Академический Документы
Профессиональный Документы
Культура Документы
and Bilins
Methods and Protocols
EDITED BY
Alison G. Smith
Michael Witty
HUMANA PRESS
Heme, Chlorophyll, and Bilins
Heme, Chlorophyll,
and Bilins
Methods and Protocols
Edited by
Alison G. Smith
Department of Plant Sciences, University of Cambridge, Cambridge, UK
and
Michael Witty
Department of Biochemistry, University of Cambridge, Cambridge, UK
www.humanapress.com
The content and opinions expressed in this book are the sole work of the authors and editors, who have
warranted due diligence in the creation and issuance of their work. The publisher, editors, and authors are
not responsible for errors or omissions or for any consequences arising from the information or opinions
presented in this book and make no warranty, express or implied, with respect to its contents.
Authorization to photocopy items for internal or personal use, or the internal or personal use of specific
clients, is granted by Humana Press Inc., provided that the base fee of US $10.00 per copy, plus US $00.25
per page, is paid directly to the Copyright Clearance Center at 222 Rosewood Drive, Danvers, MA 01923. For
those organizations that have been granted a photocopy license from the CCC, a separate system of
payment has been arranged and is acceptable to Humana Press Inc. The fee code for users of the Transac-
tional Reporting Service is: [1-58829-111-1/02 (hardcover) $10.00 + $00.25].
Heme, chlorophyll, and bilins: methods and protocols / edited by Alison G. Smith and Michael Witty
p. cm.
Includes bibliographical references (p.)
ISBN 1-58829-111-1 (alk. paper)
1. Chlorophyll. 2. Heme. 3. Tetrapyrroles. 4. Plant pigments. I. Smith, Alison G. II. Witty, Michael.
QK898.C5 H46 2001
572'.46–dc21 2001039604
Preface
The men of experiment are like the ant, they only collect and use; the
reasoners resemble spiders, who make cobwebs out of their own
substance. But the bee takes the middle course: it gathers its material
from the flowers of the garden and field, but transforms and digests it by
a power of its own. Not unlike this is the true business of philosophy
[science]; for it neither relies solely or chiefly on the powers of the mind,
nor does it take the matter which it gathers from natural history and
mechanical experiments and lay up in the memory whole, as it finds it,
but lays it up in the understanding altered and digested. Therefore, from
a closer and purer league between these two faculties, the experimental
and the rational (such as has never been made), much may be hoped.
Francis Bacon, Novum Organum, 1620
(Republished in 1960 by Liberal Arts Press, New York, p. 93)
Each time a new researcher joins a laboratory, there is a passing on of methods and
technical know-how from existing members, so that expertise is maintained and refined.
As long as the procedures are current, then the information remains easily accessible,
and can be transferred to other research groups by exchange visits, or when a researcher
moves labs. But it is seldom that the methods are published in anything other than an
abbreviated form, or with the inclusion of technical tips that can make the difference
between a method working or failing. With the handling and manipulation of
tetrapyrroles, a discipline that has been carried out over the last hundred years or so,
there have been a number of excellent handbooks published over the years that detail
the characteristics of these important compounds, and provide methods for their
preparation, analysis, and use. However, these books are now mostly out-of-print, and
in many cases had a theoretical rather than practical orientation. In the experience of
one of us (MW), as someone new moving into the area of tetrapyrrole research, despite
collecting all the methods from publications and colleagues, the knowledge was
disjointed and hard to put into practice. Furthermore, it seemed that although many
modern and state-of-the-art procedures were practiced, the simpler, more traditional
methods had been forgotten about, or lost with the retirement of older scientists.
Our goal in producing this book, therefore, was to ask scientists who routinely carry
out the experiments, to describe their basic protocols and technology for the study of
chlorophyll, heme, and related molecules, including technical tips and ways to avoid
common pitfalls. In the editing process, we have worked hard to ensure that the
contributions from each author provided a coherent and accessible introduction to their
topic, be it chemical, biophysical, or molecular biological, and that the protocols were
comprehensible to novices (us!). We are extremely grateful to all the contributors for
v
vi Preface
their willingness to modify their chapters as we requested, and for their forbearance in
the length of time it has taken to complete the project. We would also like to thank Tom
Lanigan at Humana Press Inc., for being prepared to take the project on, and Christine
McAndrew for all her help at a difficult time.
Alison G. Smith
Michael Witty
Contents
Preface ................................................................................................................... v
Contributors ....................................................................................................... ix
1 Laboratory Methods for the Study of Tetrapyrroles
Alison G. Smith and Michael Witty ................................................................ 1
2 Syntheses of Tetrapyrroles
Kevin M. Smith ................................................................................................ 13
3 General Laboratory Methods for Tetrapyrroles
Jerry C. Bommer and Peter Hambright ........................................................ 39
4 Enzymatic Preparation of Tetrapyrrole Intermediates
Martin J. Warren and Peter M. Shoolingin-Jordan .................................. 69
5 Analysis of Biosynthetic Intermediates, 5-Aminolevulinic Acid
to Heme
Chang Kee Lim ................................................................................................... 95
6 Analysis of Intermediates and End Products of the Chlorophyll
Biosynthetic Pathway
Constantin A. Rebeiz ......................................................................................111
7 Analysis of Heme and Hemoproteins
Angela Wilks ..................................................................................................157
8 Hemoproteins Purification and Characterization by Using Aqueous
Two-Phase Systems
Daniel Forciniti ............................................................................................. 185
9 Structural Study of Heme Proteins by Electron Microscopy
of 2-Dimensional Crystals
Terrence G. Frey ............................................................................................. 209
10 Analysis and Reconstitution of Chlorophyll–Proteins
Harald Paulsen and Volkmar H. R. Schmid ............................................. 235
11 Two-Dimensional Crystallization of Chlorophyll Proteins
Georgios Tsiotis ............................................................................................255
12 Biosynthesis and Analysis of Bilins
Matthew J. Terry ........................................................................................... 273
vii
viii Contents
ix
Laboratory Methods for the
1 Study of Tetrapyrroles
1
A.G. Smith and M. Witty
vulinic acid (ALA) has produced pyrroles although most deposits contain only a few
(33), while electrical discharge in the pres- parts per million, some contain significant
ence of pyrrole and formaldehyde has pro- amounts of free or complexed porphyrins,
duced porphyrins (14), and they have been for example the Gibellina sedimentary
detected in sterile meteorites (14,15). deposits, which contain 24 mg/g copper
Porphyrins are chemically stable (30) and nickel porphyrins (29).
and can persist in the environment for Although they are found in abiotic sys-
many millions of years. Porphyrins are tems, most tetrapyrroles are biological, and
found in large fossils such as mollusk shells indeed they are the most conspicuous liv-
(17) and also in molecular fossil forms in ing molecule on earth. Chlorophylls can be
geological strata. The best examples of seen from satellites in space, where vegeta-
these are coal and oil deposits, where they tion types can be identified and used to
are found as mostly nickel(II) and vanadyl predict underlying geology (31). Even
complexes (9). Mineral porphyrins have when viewed from outside, the Earth looks
been detected in sedimentary deposits with enticing because of tetrapyrroles. If there
high organic content laid down as early as are Men from Mars, they would pick on
precambrian times (8). They may precipi- Earth for special interest, and they would
tate to form distinct bedding planes and, be right to do so (25).
Figure 1. The structure of chlorophyll a. Chlorophylls are present in protein complexes in the membrane of photosynthetic bac-
teria and the thylakoid membrane of chloroplasts, where they harvest and trap light energy during photosynthesis (Chapters 10
and 11).
2
Laboratory Methods for the Study of Tetrapyrroles
3
A.G. Smith and M. Witty
Figure 3. The structure of phytochromobilin. This is the chromophore of phytochromobilin, which is the red-light receptor of
higher plants (Chapter 13). Linear tetrapyrroles are also found as accessory light-harvesting pigments in cyanobacteria and many
algae (Chapter 14).
4
Laboratory Methods for the Study of Tetrapyrroles
Figure 4. The tetrapyrrole biosynthetic pathway, showing the different endproducts and the major intermediates (Chapters 4,
5, 6, and 12). Enzymes are shown in italics.
5
A.G. Smith and M. Witty
reduction of vinyl group, and formation of a only provides the means for photosynthet-
fifth ring to produce protochlorophyllide. In ic organisms to live, but also indirectly sup-
the presence of light, protochlorophyllide is ports almost all life on earth with carbohy-
reduced to form chlorophyllides, which drates and oxygen.
undergo esterification by phytyl diphos-
phate or geranylgeranyldiphosphate to pro- 3.2. Oxidation of Carbohydrates to
duce chlorophyll (see Chapter 6). Produce Usable Energy
Nonphotosynthetic cells obtain their
3. ROLES OF TETRAPYRROLES
energy by the oxidation of carbohydrates,
3.1. Light Harvesting which in aerobic organisms results in the
formation of CO2. This process involves a
Photosynthetic organisms contain a series of reduction-oxidation (redox) reac-
sophisticated system of several hundred tions whereby the large gap in oxidation
chlorophylls (or bacteriochlorophylls) and state between carbohydrate and carbon
other accessory pigments, which act as dioxide is released in a series of gentle and
antennae to absorb light and pass the ener- efficient steps, with oxygen as the final
gy to special chlorophylls in reaction cen- electron acceptor. Transition metals, such
ters. Here the light energy is trapped as as the iron found in heme (Figure 2), are
excited electrons, which are then trans- well suited to catalyze these reactions
ferred through an electron transfer chain to because they contain d-electron orbital sys-
generate ATP. In higher plants, algae, and tems with small differences in energy lev-
cyanobacteria, this process results in the els, thus allowing a range of oxidation
oxidation of water to evolve molecular oxy- states so that energy can be released in a
gen and the production of reduced nicoti- controlled and useful way (cf section 3.4).
namide adenine dinucleotide phosphate In the bacterial membrane, and the
(NADPH) (see Chapters 10 and 11 for mitochondria of eukaryotes, a series of pro-
more detail). The ATP and NADPH gen- tein complexes containing cofactors, which
erated by the light-dependent reactions are include heme (see Chapter 7), undergo a
used to fix CO2 into organic combination series of reversible redox reactions that gen-
via the Calvin cycle. Photosynthesis not erates ATP. In this respect, the process of
6
Laboratory Methods for the Study of Tetrapyrroles
Figure 6. The formation of a tetrapyrrole. The reaction catalyzed by PBG deaminase. The holoenzyme (E) contains an active site
dipyrromethane cofactor. This is used to accept PBG monomers and form enzyme substrate complexes. A, acetate; P, propionate
moiety. Pyrrole rings are labeled A, B, C, and D.
7
A.G. Smith and M. Witty
oxidative phosphorylation is similar to that ply from one beat to the next. Intracellular
in photosynthesis (section 3.1), reflecting myoglobin is also found in bacteria, proto-
common evolutionary ancestors of some of zoans, plants, and invertebrates (32).
the components involved. Plants also contain leghemoglobin used
for the homeostasis of oxygen. The best
3.3. Transport and Homeostasis of studied example is in the legume root nod-
Oxygen ule, where symbiotic bacteria consume
large amounts of energy during the reduc-
Metabolism requires the consumption tion of atmospheric nitrogen to ammonia,
of large amounts of oxygen, and therefore which is then available to the host plant.
transport from the atmosphere to cells Leghemoglobin facilitates the maintenance
buried deep in animal bodies is necessary of high levels of oxygen needed for bacteri-
to allow a rapid rate of metabolism. While al respiration, while preventing poisoning
several kinds of oxygen transport proteins of the nitrogen fixing machinery by molec-
exist, including hemerythrins (non-heme ular oxygen (2). Unlike in animals, the
iron proteins) and hemocyanins (copper protein is not circulated, but rather simple
proteins), globins are the most abundant diffusion down an oxygen gradient created
and widespread oxygen transport protein by bacterial respiration promotes transport
(12). They use reversible oxygen coordina- of oxygen from the outside.
tion to protoheme IX iron (FeII) for trans-
port of oxygen from respiratory organs 3.4. Protection of Cellular Processes
throughout the animal body. from Reactive Oxygen Species
Hemoglobins transport oxygen in many
animal groups, and most have similar As well as the desired biological reaction
structures and functions (35). In circulat- of oxygen as a terminal acceptor of electrons
ing red blood cells, hemoglobins consist of in the controlled oxidation of carbohydrates
17 kDa polypeptides, each containing a in respiration, oxygen will also react with
heme group that can bind one oxygen mol- electrons encountered at random, to pro-
ecule. Hemoglobins are typically tetramers duce reactive oxygen species (ROS) such as
that allow for cooperative binding and oxygen radicals and superoxide. These are
release of oxygen, depending on the PO of very harmful to living systems, causing lipid
2
the surrounding tissue. peroxidation, membrane damage, and
In addition to oxygen transport, hemo- genetic mutation. Cells contain a number
globins are associated with other important of enzymes that act to remove these inter-
functions. For example, circulating hemo- mediates quickly. Many of these oxidases
globin I of the swamp clam Lucina pectina- are heme-containing proteins, including
ta is involved in sulfide transport. Sulfides peroxidases and catalase (7).
are bound with high affinity within a cage
of heme and three phenylalanine residues 3.5. Taking Advantage of Reactive
and are released to symbiotic bacteria upon Oxygen Species
reduction (3).
Myoglobin is a monomeric protein that Although ROS can be harmful, organ-
contributes to the transport of oxygen by isms have evolved systems in which they
diffusion, and large amounts are found in are useful. One important example of this
skeletal and cardiac muscles of mammals. is the biosynthesis and degradation of
In cardiac muscle, myoglobin acts as a lignin. Lignocellulose is a composite of
short-term oxygen buffer, smoothing sup- lignin and cellulose and probably the most
8
Laboratory Methods for the Study of Tetrapyrroles
abundant organic molecule in the bios- prosthetic group, tetrapyrroles also func-
phere, functioning as material for mechan- tion in key regulatory processes. These
ical strength in wood. Rather than a regular include control of gene expression, cellular
polymer such as cellulose, lignin is synthe- signaling pathways, and control of protein
sized by peroxidases, which oxidize phenol- transport within the cell.
propane units (such as coniferyl alcohol) to An example is heme-mediated feedback
form reactive radical species. These poly- control of its own synthesis, which appears
merize in an irregular fashion to form to occur in all groups of organisms, most
lignin, which thereby contains a wide array importantly at the production of the first
of chemical linkages (38). Because of this committed precursor ALA. In plants, there
unsymmetrical arrangement, lignin is is evidence that heme inhibits the enzyme
unusually resistant to enzymatic break- glutamyl-tRNA reductase (36). In animals,
down, and only a few microbes, such as the although heme inhibits the activity of
white rot fungi, have enzymes capable of ALAS, control is exerted at several other
doing so (20). Lignin is also an important points as well. In mammals, there are two
byproduct of the paper industry, which forms of the enzyme, constitutive and ery-
generates about 30 million tons of unused throid-specific. Liver ALAS is inhibited by
lignin per year, in a process involving harsh heme in a negative feedback loop (11) to
chemical treatments (13). Genetic modifi- maintain levels of heme production for the
cation of tree species to reduce lignin con- maintenance of cellular processes. This feed-
tent is being explored as a means of avoid- back regulation is achieved by a combina-
ing this costly and polluting process (34). tion of effects including inhibition of ALAS
A more general, and essential, role of gene expression, increased ALAS mRNA
these reactive species is found in both degradation, and inhibition of pre-ALAS
plants and animals, where ROS have been protein transport to the mitochondrion
shown to modulate a wide range of cellular (19), with only a minor contribution by
and physiological processes, acting as part inhibition of ALAS catalytic activity (28).
of the signal transduction pathway. For In contrast to the liver, in erythroid cells,
example, one of the earliest responses to transcription of the ALAS gene, together
pathogen attack is a marked increase in with genes for later enzymes in the pathway
ROS in the infected tissue, produced in and for globins, is stimulated by heme to
part by the activity of plasma membrane- produce the large amounts of heme needed
bound NADPH oxidase, a hemoprotein for hemoglobin in red blood cells.
complex. The ROS, in turn, act as a trigger In yeast, expression of the ALAS gene is
for defense responses, such as modification controlled by heme and mediated through
of membrane permeability and ion fluxes, the transcription factor HAP1, which
and systemic acquired resistance in plants binds heme for activity. The binding
(1). Similarly in animals, ROS influence domain contains multiple copies of a short
signaling cascades and transcriptional– motif, which is also found in the mito-
posttranscriptional control of gene expres- chondrial transit peptide of mammalian
sion, thereby playing an essential role in ALAS. This motif, involved in transient
processes such as apoptosis (23). binding of heme, is quite different to the
more stable heme-binding domain of
3.6. Control of Metabolic and Cellular cytochromes and globins (39).
Processes by Signaling There is accumulating evidence that
tetrapyrrole intermediates play a role in sig-
As well as functioning as an enzymic naling. In plants, there is coordination
9
A.G. Smith and M. Witty
between the chloroplast and the nucleus, rigid structures that they are able to form
such that nuclear-encoded genes for chlo- and the fact that they can coordinate a
roplast-targeted proteins are transcribed number of metal ions which are involved
only in cells with functional chloroplasts. in the catalysis. For example, using por-
Although the exact identity of the so-called phyrin molecular boxes and zinc coordina-
“plastid-factor” remains elusive, plant tion, it has been possible to influence the
mutants with defects in certain steps of the stereospecificity of reactions by the geo-
tetrapyrrole biosynthetic pathway have metrical constraints of a host cavity (26).
altered plastid-nuclear signaling (24). Other novel uses of tetrapyrroles have
been established in clinical medicine, in
3.7. Subtle Pigmentation particular for the treatment of cancer cells,
in a technique called photodynamic thera-
While plants are green because of the py (PDT). The rationale behind the meth-
presence of chlorophyll and animal tissues od is to load the cancerous cells with pho-
are largely red due to heme, some of the tosensitizing porphyrin mixtures, which,
more subtle animal colors are also conferred upon irradiation with visible light, cause
by tetrapyrroles. The cuticle of birds’ eggs the production of singlet oxygen, thereby
with colored shells contain tetrapyrroles leading to the destruction of the cells as
which contribute to their camouflage. Most described in section 3.4. Porphyrins are
markings and pigmentation are due to pro- ideal compounds for this technique, not
toporphyrin IX, which is associated with only because of their light absorption prop-
brown and black coloring. Blue eggshells erties, but also because there is some pref-
are associated with biliverdin IXα, and erential uptake of these molecules by
green eggshells are associated with zinc- tumor cells. Initially, in the 1960s and
biliverdin IXα with traces of copropor- 1970s, the major photosensitizers used
phryin III (18). The feathers of some birds were hematoporphyrins and related prepa-
also contain tetrapyrrole pigments. The rations derived from acid extraction of
feathers of Turocos contain red turacin (cop- blood (or hemoglobin), and therefore, are
per-uroporphyrin III) and green tura- not chemically defined compounds. How-
coverdin (22). Uroporphryin I is found in ever, since 1980, new sensitizers have been
many calcified mollusk tissues such as shells developed, including chlorins and phthalo-
(17) and pearls (16), though the function of cyanines, which have been chemically syn-
the tetrapyrrole is unknown. thesized (5).
our own laboratory experience, there are due to exocyclic ring size. Inorg. Chem. 35:199-209.
certain “tricks-of-the-trade” which are nec- 10.Dailey, K.K. and T.B. Rauchfuss. 1997. π-Complex-
es of metalloporphyrins as model intermediates in
essary to use in order to carry out success- hydrodemetallation (HDM) catalysis. Polyhedron
ful experiments. In this volume, we have 16:3129-3136.
selected articles written by people who 11.Granick, S. 1966. The induction in vitro of the syn-
thesis of δ-aminolevulinic acid synthase in chemical
actually carry out these procedures on a porphyria: a response to certain drugs, sex hormones,
daily basis in their own laboratories. Each and foreign chemicals. J. Biol. Chem. 241:1359-
chapter provides an overview of the topic 1375.
12.Hardison, R. 1998. Hemoglobins from bacteria to
with general information on the experi- man: evolution of different patterns of gene expression.
mental approach, as well as a number of J. Exp. Biol. 201:1099-1117.
13.Hartley, B.S., P.M.A. Broda, and P.J. Senior. 1987.
step-by-step procedures, which should pro- Technology in the 1990s: Utilization of Lignocellulosic
vide the basis for any novice tetrapyrrolo- Wastes. The Royal Society, London.
gist taking their first steps into this field. 14.Hodgson, G.W. and B.L. Baker. 1964. Evidence for
porphyrins in the orgueil meteorite. Nature 202:125-
131.
15.Hodgson, G.W. and B.L. Baker. 1967. Porphyrin
ABBREVIATIONS abiogenesis from pyrrole and formaldehyde under sim-
ulated geochemical conditions. Nature 216:29-32.
16.Iwahashi, Y. and S. Akamatsu. 1994. Porphyrin pig-
ALA, 5-aminolevulinic acid; ALAS, ment in black-lip pearls and its application to pearl
ALA synthase; PBG, porphobilinogen; identification. Fisheries Sci. 60:69-71.
17.Kennedy, G.Y. 1975. Porphyrins in invertebrates. Ann.
PDT, photodynamic therapy; ROS, reac- NY Acad. Sci. 244:662-673.
tive oxygen species. 18.Kennedy, G.Y. and H.G. Vevers. 1976. A survey of
avian eggshell pigments. Comp. Biochem. Physiol. B
55:117-123.
19.Lathrop, J.T. and M.P. Timko. 1993. Regulation by
REFERENCES heme of mitochondrial protein-transport through a
conserved amino-acid motif. Science 259:522-525.
1.Alvarez, M.E., R.I. Pennell, P.J. Meijer, A. Ishikawa, 20.Leonowicz, A., A. Matuszewska, J. Luterek, D.
R.A. Dixon, and C. Lamb. 1998. Reactive oxygen Ziegenhagen, M. Wojtas-Wasilewska, N.S. Cho, M.
intermediates mediate a systemic signal network in the Hofrichte, and J. Rogalski. 1999. Biodegradation of
establishment of plant immunity. Cell 98:773-784. lignin by white rot fungi. Fungal Genet. Biol. 27:175-
2.Appleby, C.A. 1984. Leghemoglobin and rhizobium 185.
respiration. Annu. Rev. Plant Physiol. 35:443-478. 21.McDonagh, A.F. 1979. Bile pigments: bilatrienes and
3.Bolognesi, M., C. Rosano, R. Losso, A. Borassi, M. 5,15-biladienes, p. 293-491. In D. Dolphin (Ed.), The
Rizzi, J.B. Wittenberg, A. Boffi, and P. Ascenzi. 1999. Porphyrins, Vol. 1. Academic Press, London.
Cyanide binding to Lucina pectinata hemoglobin I and 22.Nicholas, R.E.H. and C. Rimington. 1952. Isolation
to sperm whale myoglobin: an X-ray crystallographic of unequivocal uroporphyrin III, a further study of
study. Biophys. J. 77:1093-1099. turacin. Biochem. J. 50:194-201.
4.Bonarlaw, R.P., L.G. Mackay, C.J. Walter, V. Mar- 23.Nose K. 2000. Role of reactive oxygen species in the
vaud, and J.K.M. Sanders. 1994. Towards synthetic regulation of physiological functions. Biol. Pharmacol.
enzymes based on porphyrins and steroids. Pure Appl. Bull. 23:897-903.
Chem. 66:803-810. 24.Oster, U., H. Brunner, and W. Rudiger. 1996. The
5.Bonnett, R. 1999. Photodynamic therapy in historical greening process in cress seedlings. 5. Possible interfer-
perspective. Rev. Contemp. Pharmacother. 10:1-17. ence of chlorophyll precursors, accumulated after thu-
6.Buchler, J.W. 1975. Static coordination chemistry of japlicin treatment, with light-regulated expression of
metalloporphyrins, p. 157-231. In K.M. Smith (Ed.), Lhc genes. J. Photochem. Photobiol. B 36:255-261.
Porphyrins and Metalloporphyrins. Elsevier, Amster- 25.Sagan, C., W.R. Thompson, R. Carlson, D. Gurnett,
dam. and C. Hord. 1993. A search for life on earth from the
7.Cadenas, E. 1989. Biochemistry of oxygen toxicity. Galileo spacecraft. Nature 365:715-721.
Annu. Rev. Biochem. 58:79-100. 26.Sanders, J.K.M. 1998. Supramolecular catalysis in
8.Callot, H.J. 1991. Geochemistry of chlorophylls, p. transition. Chem. Eur. J. 4:1378-1383.
339-364. In H. Scheer (Ed.) Chlorophylls. CRC Press, 27.Sanders, J.K.M., N. Bampos, Z. Clyde-Watson, S.L.
Boca Raton. Darling, J.C. Hawley, H.J. Kim, C.C. Mak, and S.J.
9.Czernuszewicz, R.S., J.G. Rankin, and T.D. Lash. Webb. 2000. Axial coordination chemistry of metallo-
1996. Fingerprinting petroporphyrin structures with porphyrins, p. 349-390. In K.M. Kadish, K.M. Smith,
vibrational spectroscopy. 4. Resonance raman spectra and R. Guilard (Eds.), The Porphyrin Handbook. Aca-
of nickel(II) cycloalkanoporphyrins: structural effects demic Press, London.
11
A.G. Smith and M. Witty
28.Sassa, S. and T. Nagai. 1996. The role of heme in gene tors from Antirrhinum regulate phenylpropanoid and
expression. Int. J. Hematol. 63:167-178. lignin biosynthesis in transgenic tobacco. Plant Cell
29.Schaeffer, P., R. Ocampo, H.J. Callot, and P. 10:135-154.
Albrecht. 1993. Extraction of bound porphyrins from 35.Terwilliger, N.B. 1998. Functional adaptations of
suphur-rich sediments and their use for reconstruction oxygen-transport proteins. J. Exp. Biol. 201:1085-
of palaeoenvironments. Nature 364:133-136. 1098.
30.Smith, K.M. 1975. Porphyrins and Metallopor- 36.Vothknecht, U.C., C.G. Kannangara, and D. Wett-
phyrins, p. 829-836. Elsevier, Amsterdam. stein. 1998. Barley glutamyl tRNA(Glu) reductase:
31.Smith, M.O., S. Jacquemond, M. Verstraete, and Y. mutations affecting haem inhibition and enzyme activ-
Govaerts. 1999. Geobotany: vegetation mapping for ity. Phytochemistry 47:513-519.
earth sciences, p. 189-248. In Remote Sensing for the 37.Warren, M.J. and A.I. Scott. 1990. Tetrapyrrole
Earth Sciences, Manual of Remote Sensing, Vol. 3. assembly and modification into the ligands of biologi-
John Wiley & Sons, New York. cally functional cofactors. Trends Biochem. Sci.
32.Suzuki, T. and K. Imai. 1998. Evolution of myoglo- 15:486-491.
bin. Cell. Mol. Life Sci. 54:979-1004. 38.Whetten, R.W., J.J. MacKay, and R.R. Sedoroff.
33.Szutka, A. 1966. Formation of pyrrolic compounds by 1998. Recent advances in understanding lignin biosyn-
ultra-violet irradiation of δ-aminolevulinic acid. thesis. Annu. Rev. Plant Physiol. Plant Mol. Biol.
Nature 212:401-402. 49:585-609.
34.Tamagnone, L., A. Merida, A. Parr, S. Mackay, F.A. 39.Zhang, L. and L. Guarente. 1995. Heme binds to a
Culianez-Macia, K. Roberts, and C. Martin. 1998. short sequence that serves a regulatory function in
The AmMYB308 and AmMYB330 transcription fac- diverse proteins. EMBO J. 14:313-320.
12
Syntheses of Tetrapyrroles
2
Kevin M. Smith
Department of Chemistry, University of California–Davis,
Davis, CA, USA
13
K.M. Smith
14
Syntheses of Tetrapyrroles
chloride. After cooling and removal of coag- (16,44). Hemes [iron(II) porphyrins] can be
ulated protein (usually with a wooden obtained from hemins [iron(III) chloride
stick), the hemin separates and can be col- porphyrins] most commonly by reduction
lected by filtration. Alternatively, the messy with sodium dithionite under nitrogen or
protein can be precipitated by addition of a argon. Since autoxidation of iron(II) to
solution of strontium chloride, followed by iron(III) porphyrins is very facile in air, use
concentration to give hemin as above of nitrogen or (preferably the heavier) argon
16
Syntheses of Tetrapyrroles
18
Syntheses of Tetrapyrroles
tion, or ether formation, at the 3,8-(1- [11] (29). Deuterohemin [22] can be
hydroxyethyl) groups is a problem; it is obtained from “proto” hemin by brief heat-
best to use diazomethane in methanol to ing of [11] in a resorcinol melt (60), via the
obtain the dimethyl ester [20] (CAUTION). so-called Schumm reaction in which the
vinyl groups are replaced by hydrogen
3.1.5. Deuteroporphyrin IX [23] atoms (10,12,17,42). Demetalation, as
reported above for the preparation of pro-
Deuteroporphyrin IX [23] is of signifi- toporphyrin IX from hemin, then affords
cant historical importance because it was deuteroporphyrin IX [23].
the first porphyrin isolated in Fischer’s Numerous 3,8-disubstitution products
Nobel Prize winning synthesis of hemin (and 3- or 8-monosubstitution analogues)
19
K.M. Smith
of deuteroporphyrin IX and its esters can 3.2 Porphyrins and Chlorins from Plants
be prepared, usually by way of aromatic and Algae
electrophilic substitution on the hemin
or its copper(II) complex. A typical In this section, some simple degradation
example is 3,8-diacetyldeuteroporphyrin reactions, which furnish porphyrins and
IX [21] (see above), which was also an chlorins in useful quantities from plants
intermediate in the Fischer’s hemin total and algae, will be described.
synthesis. The traditional source for chlorophylls a
[12] and b [13], usually present in a ratio
3.1.6. Coproporphyrin III [24] of about 3:1, was leaf tissue, usually
spinach (25,68). A very useful chemical
Coproporphyrin III [24] is a biologically adjunct for simplification of the mandato-
significant porphyrin because its hexahydro- ry chromatographic separation of the
derivative, coproporphyrinogen III [25], is a chlorophyll a and b pigments has been
colorless intermediate on the pathway reported (41); it employs the Girard
between uroporphyrinogen III [26], proto- reagent T as a means of dramatically
porphyrinogen IX [27], and protoporphyrin increasing the polarity of the series b com-
IX [1] in normal porphyrin metabolism. ponent in the mixture. For example, reac-
Under normal circumstances, the amount of tion of methyl pheophorbide a [28] and b
[25] present at steady state is small. Howev- [29] mixture (see above) with Girard’s
er, biological oxidation of coproporphyrino- reagent T gives a mixture consisting of
gen III yields the colored coproporphyrin unreacted a series compound, i.e., methyl
III, which takes it out of the normal meta- pheophorbide a [28], and the salt [30]
bolic sequence. Hence, certain diseases of from the b series. Column chromatography
porphyrin metabolism can result in a then achieves a very simple separation in
buildup of excess photochemically active which [30] remains adsorbed to the top of
porphyrins in tissues; such diseases are the column, whereas the relatively nonpo-
known collectively as porphyrias. Chemical- lar a series compound [28] is eluted quick-
ly, porphyrinogens can be oxidized very effi- ly. Use of a polar solvent then elutes the b
ciently to porphyrins by use of 2,3-dichloro- series salt, which can be hydrolyzed to give
5,6-dicyanobenzoquinone (DDQ). If pure methyl pheophorbide b [29].
biosynthetic work using porphyrinogens is Investigators wishing only to deal with
to be carried out, the corresponding por- chlorophyll derivatives in the a series were
phyrin can usually be reduced to por- advantaged when it was shown that Spir-
phyrinogen using sodium amalgam or cat- ulina maxima (from Mexico) or S. pacifica
alytic hydrogenation (15). When vinyl (from Hawaii) contain only the chloro-
groups are present on the porphyrin macro- phyll a series of pigments. On account of
cycle, of course, only the sodium amalgam the fairly drastic extraction conditions,
route is recommended—catalytic hydro- chlorophyll a itself is usually not obtained
genation will probably reduce the vinyls to directly from the alga, but large quantities
ethyls. It must be kept in mind when han- of pheophytin a [31] and methyl pheo-
dling porphyrinogens, that oxygen and light phorbide a [28] (up to 0.4% measured by
can efficiently oxidize the hexahydro mater- dry weight) can be obtained (67).
ial to the porphyrin level, which will make it Treatment of the plant chlorophylls
inactive in biosynthetic investigations—the (either separately or as a mixture) with
first true porphyrin in porphyrin biosynthe- mild acid gives the metal-free pheophytins
sis is protoporphyrin IX itself. a [31] and b [32]; this, as a dried paste, is
20
Syntheses of Tetrapyrroles
usually the form in which the pigments are pheophorbides still contain one ester, and
stored prior to further degradation to use- that hydrolysis of this ester will cause con-
ful materials. Hydrolysis of the pheo- comitant decarboxylation on ring E).
phytins gives the corresponding pheophor- Alternatively, and preferably (for ease of
bides a [33] and b [34]; (note that the handling), methanolysis of pheophytin a
21
K.M. Smith
22
Syntheses of Tetrapyrroles
or b provides the corresponding methyl 18 [38], which bears a very useful anhy-
pheophorbides a or b [28 or 29, respective- dride ring [45]. On the other hand, dia-
ly]—these contain two methyl esters. zomethane esterification (CAUTION) yields
Transesterification of the phytyl ester for purpurin 7 trimethyl ester [39] (26–
methyl, without removal of the magnesium 28,45). Heating of [39] in collidine gives a
atom, can be accomplished to afford the diversely substituted porphyrin, 3-vinyl-
methyl chlorophyllides [35] and [36] (26). rhodoporphyrin XV dimethyl ester [40]
A number of simple to perform but (28). If the so-called “meso” (i.e., 3-ethyl
mechanistically complex reactions can be instead of 3-vinyl) series of pigments is
carried out on chlorophyll derivatives. For used, another porphyrin, rhodoporphyrin
example, oxidation of pheophytin a [31] XV dimethyl ester [41], is obtained.
under highly alkaline conditions accom- The isocyclic ring (E) in chlorophylls
plishes cleavage the 131-132 bond in the β- and their derivatives contains a β-ketoester
ketoester ring E, with hydrolysis of the of function which imparts a high degree of
phytyl ester, to give Fischer’s “unstable chemical reactivity upon the compounds
chlorin” [37] (28). Simple evaporation of containing it. Such lability is often a disad-
the solution affords the so-called purpurin vantage in the use of chlorophyll derivatives
23
K.M. Smith
for specific purposes; the spectrum of are much less reactive than are conjugated
chemical reactivity in the ring E portion of ketoesters. Thus, heating of methyl pheo-
the pigments can be dramatically decreased phorbide a [28] (or b [29]) in collidine (30)
by removal of the 132-CO2Me group. gives methyl pyropheophorbide a [42] (or
When the 132-CO2Me group is removed, b [43]) in virtually quantitative yield; use of
the so-called “pyro” series of chlorophyll collidine is a yield-enhancing improvement
derivatives are obtained. Basically, ketones upon the classical method (28) which uti-
24
Syntheses of Tetrapyrroles
lized pyridine. Identical demethoxycar- infra] can be treated with alkali to give,
bonylation reactions take place with the so- after esterification with diazomethane,
called meso- (3-ethyl) series of compounds. chlorin e6 trimethyl ester [45] and chloro-
The 5-membered isocyclic ring in the porphyrin-e6 trimethyl ester [46], respec-
pyro-series of chlorophylls cannot be tively (24). Methanolysis of pheophorbide
cleaved, but the ring E in its β-ketoester a also affords [45]. This reaction can be
form can be readily opened since the high- reversed, and ring E is reformed either by
ly reactive conjugated functionality pro- treatment with methoxide (24), with tert-
vides a handle for chemical elaboration of butoxide (65), or best of all using tri-
ring E. For example, pheophorbide a [33] phenylphosphine and bis(trimethylsilyl)
and 3-vinylpheoporphyrin a5 [44, vide amide (31).
25
K.M. Smith
Although the chlorophyll b series of pig- 4. The green filtrate is evaporated and
ments is less accessible than those from then purified by flash chromatography
chlorophyll a (and indeed, as mentioned on Grade V neutral alumina, eluting
above, Spirulina algae contains no chloro- first with n-hexane to remove a fast run-
phyll b) a series of reactions parallel to ning yellow band, with dichlormethane
those described above also occurs in the b to remove the major blue-gray pheo-
series; the analogue of chlorin e6 trimethyl phytin a band, and finally with 97:3
ester in the b series is called rhodin g7 dichloromethane:tetrahydrofuran to
trimethyl ester [47] and of chloropor- remove some bright green magnesium-
phyrin e6 trimethyl ester is rhodinpor- containing pigments.
phyrin g7 trimethyl ester [48]. 5. The evaporated pheophytin a fraction is
Chlorins can be converted into por- treated with 500 mL of 5% sulfuric acid
phyrins by using DDQ as a dehydrogena- in methanol (degassed by bubbling with
tion agent. The β-ketoester functionality nitrogen gas) for 12.5 hours at room
does not take kindly to oxidative stress, so temperature in the dark (aluminum
methyl pheophorbide a [28] gives only a foil) under nitrogen, followed by dilu-
low yield of 3-vinylpheoporphyrin a5 di- tion with dichloromethane, and rinsing
methyl ester [44]. Using a “sledgehammer” with water.
approach to preparation of porphyrins 6. The mixture is rinsed with 10% saturat-
from chlorophyll derivatives, chlorophyll a ed aqueous sodium bicarbonate, the
under very vigorous basic conditions fol- organic layer is dried over anhydrous
lowed by esterification (methanolysis), sodium sulfate, followed by evapora-
affords phylloporphyrin XV methyl ester tion and crystallization of the residue
[49] and pyrroporphyrin XV methyl ester from dichloromethane:methanol. This
[50] (23). gives methyl pheophorbide a [28]
(average yield 1.8 g).
❖ Procedure 1. Isolation of Methyl Pheo-
phorbide a [28] from S. maxima (67)
4. CHEMICAL SYNTHESES OF
1. About 500 g of dried S. maxima alga is PORPHYRINS
slurried in 2 L of acetone, and then liq-
uid nitrogen is added to form a frozen Porphyrin chemical synthesis will be dis-
slush. cussed here in connection with two series
2. After transferring to a 5-L 3-necked of compounds: (i) those porphyrins which
round-bottom flask, the mixture is have been most often used in connection
heated at reflux with mechanical stir- with model studies, e.g., 5,10,15,20-tetra-
ring for 2 hours. The supernatant is fil- phenylporphyrin (TPP) [51] and 2,3,7,8,
tered through a Whatman filter paper 12,13,17,18-octaethylporphyrin (OEP)
(Whatman, Clifton, NJ, USA) using a [52]; and (ii) those related to protopor-
Buchner funnel, and more acetone is phyrin IX [1]. Simply based on the sym-
added to the solid debris. metry in the substituent arrays of [51] and
3. The extraction process, with refluxing, [52] and the lack of symmetry in [1], it is
is repeated twice more—note that the obvious that it would be a waste of time to
debris retains its deep green color, but approach the synthesis of both series of
amounts of additional chlorophyll compounds using the same strategy. To
obtained are marginal. attempt the synthesis of OEP [52] by labo-
26
Syntheses of Tetrapyrroles
rious multistep construction of an open- give pyrrole [54]. Simply pouring the
chain tetrapyrrolic intermediate would be cooled reaction mixture into iced water
plainly unwise—such symmetrically sub- causes precipitation of the product pyrrole.
stituted compounds are most efficiently The reaction works with a variety of sub-
obtained by tetramerization of a suitable stituents on the central (i.e., 2-) carbon of
monopyrrole (see below). On the other the 1,3-dione and with a variety of esters
hand, there is no way (in the absence of on the acetoacetate. The reaction described
enzymes) that protoporphyrin IX [1] can above (using acetoacetates) is the Johnson
be synthesized chemically by monopyrrole version, while the Kleinspehn modification
chemical self-condensation, so a more employs oximinomalonic esters in place of
sophisticated chemical approach is essen- the acetoacetates.
tial. As it happens, porphyrins [51] and Compared with the above synthesis of
[52] can be synthesized by self-condensa- pyrroles, methodology for preparing
tion of monopyrroles, while protopor- pyrroles such as [58] is relatively new. A
phyrin IX [1] can be accessed by a number major advance in the field was made when
of routes, the most simple (and the one to the Barton–Zard pyrrole synthesis was
be used as an example in this chapter) reported (8); the importance of this route
being from dipyrroles. was related to the substituent patterns
which could be accessed using it. Thus,
4.1. Syntheses of Pyrroles treatment of a nitroalkene [59a] or its syn-
thetic precursor, an acetoxynitroalkane
For both series of compounds men- [e.g., 59b], with an isocyanoacetate [e.g.,
tioned above, it is first essential to 60] affords excellent yields of pyrroles such
synthesize monopyrroles. Pyrrole itself as [58].
[53] is commercially available. Syntheses
of two common examples of useful
pyrroles (from the many dozens in the ❖ Procedure 2. Synthesis of Ethyl 3,4-
literature) (5,20,32,37,38) will be illus- Diethylpyrrole-2-Carboxylate [58]
trated here. (55)
Pyrroles bearing peripheral substituents
1. A mixture of 3-acetoxy-4-nitrohexane
are those which are most useful for appli-
[59b] (8)(16.3 g), ethyl isocyanoacetate
cation to dipyrrole and porphyrin synthe-
[60] (9.8 g; Sigma, St. Louis, MO,
sis. The Johnson–Kleinspehn synthesis
USA), and 1,8-diazabicyclo[5.4.0]
(11,43) is perhaps the most useful for tetra-
undec-7-ene (26.4 g; Sigma) in tetrahy-
substituted pyrroles. For example, pyrrole
drofuran (100 mL) is stirred at 20°C
[54], bearing very useful methyl and pro-
for 12 hours.
pionate groups, is prepared by the reaction
of dione [55] with benzyl oximinoacetoac- 2. The mixture is poured into water con-
etate [56]—compound [56] is in turn taining 1 M HCl and then extracted
obtained by the reaction of benzyl acetoac- with ethyl acetate.
etate [57] with sodium nitrite in the pres- 3. The extracts are washed with water and
ence of acetic acid. Slow admixture of dried over anhydrous magnesium sul-
equimolar amounts of [55] and [56] and fate.
excess zinc powder and sodium acetate in 4. After evaporation of the solvents, the
hot acetic acid results in reduction of the residue is chromatographed on a col-
oximinoderivative [56] to the amine, fol- umn of silica gel eluted with hexane:
lowed by in situ condensation with [55] to dichloromethane mixtures.
27
K.M. Smith
28
Syntheses of Tetrapyrroles
5. Evaporation of the eluates containing advantage of using the clay is that it can be
the red band will give the required pyr- removed simply by filtration after the reac-
role [58], with an average yield of 14.2 g. tion is complete. Compound [62] is
obtained from the corresponding methyl-
4.2. Syntheses of Dipyrromethanes pyrrole [64] simply by treatment with lead
tetra-acetate. Note that pyrrole [64] pos-
Unsymmetrically substituted dipyrro- sesses the same “symmetry pattern” as does
methanes, [e.g., 61], can be prepared by pyrrole [54]; its synthesis is relatively
condensation of 2-acetoxymethylpyrroles straightforward (21). Pyrrole [63] can be
[62] with 2-unsubstituted pyrroles [63] in obtained from [64] by following a
acetic acid containing a catalytic amount sequence of reactions as shown in Scheme
(<0.1 equiv.) of toluene p-sulfonic acid 1. The dipyrromethane [61] will form
(13). Montorillonite K-10 clay has also rings A and B of protoporphyrin IX
been shown to be a very useful acid catalyst dimethyl ester [2] in the synthesis, which
in dipyrromethane syntheses (30,36); the will be described later. The 1- and 9-car-
29
K.M. Smith
30
Syntheses of Tetrapyrroles
31
K.M. Smith
32
Syntheses of Tetrapyrroles
from the Rothemund and Adler–Longo 6. The alumina is washed with dichloro-
(propionic acid) approaches is somewhat methane (200 mL), and the combined
impure (though highly crystalline) and filtrates are concentrated to approxi-
contains (4) about 5% or less of meso- mately 200 mL before addition of 200
tetraphenylchlorin [70]. Brief treatment (6) mL of methanol.
of the crude product with DDQ accom- 7. Filtration results in the chlorin-free
plishes transformation of [70] into [51]; product as glistening purple crystals,
earlier methodology involved the separa- with an average yield of 19.2 g.
tion of these two components on a chro-
matography column, but transformation of Approaches to so-called octaalkylpor-
[70] into [51] instead of separation of [51] phyrins (such as OEP [52]) are a little more
from [70] is much more sensible. Using complicated, but only with regard to the
this kind of methodology, literally kilo- difficulty of preparing the pyrrole starting
grams of TPP can be prepared. TPP can, in materials. In this case, the future meso-
any case, be purchased either “chlorin-free” (i.e., interpyrrolic) carbons can either be
or crude. Additionally, with only relatively present already on the pyrrole, or as in the
few exceptions, the reaction tolerates sub- case with TPP (wherein the meso-carbons
stitution of other arylaldehydes for ben- were provided by the formyl carbon in
zaldehyde, and good yields of a variety of benzaldehyde), the meso-carbons can be
tetra-arylporphyrins can be obtained (47). added separately from the pyrrole. A prima-
ry stricture, as mentioned above, is that a
❖ Procedure 3. Synthesis of Chlorin-Free monopyrrole tetramerization approach can
TPP [51] (6) only be used to give structurally unique
product if the substituents at positions 3-
1. Benzaldehyde (66.5 mL) and pyrrole and 4- in the monopyrrole are identical.
(46.5 mL) are simultaneously added to Thus, fully symmetrical porphyrins such
refluxing propionic acid (2.5 L), and as octaethylporphyrin [52] can be prepared
the mixture is refluxed for a further 30 easily using two major routes. The first
minutes before being allowed to cool approach, which chronologically was devel-
overnight to room temperature. oped first, involves the tetramerization of
2. The crude TPP is filtered off, washed pyrroles [71] bearing 2-CH2R substituents;
with hot water, and then washed with the “R” group must be a good leaving
methanol until the filtrate is colorless, group, and the methylene carbon of the 2-
to give 20.4 g (20% yield) of purple substituent will eventually be the source of
glistening crystals. the 5,10,15, and 20-carbons of the product
porphyrin. After the condensation reaction,
3. Concentration of the propionic acid
an oxidation step is necessary to afford good
filtrate affords a second crop of crystals.
yields of symmetrical porphyrin. Useful
4. The crude TPP (20 g) is dissolved in examples for attachment of CH2R groups
refluxing ethanol-free chloroform (2.5 to pyrroles are (i) the Mannich reaction of
L) before addition of DDQ (5 g) in pyrrole [72] with formaldehyde and
dry toluene (150 mL). dimethylamine [or better, with commercial-
5. The mixture is refluxed for 3 hours ly available (N,N-dimethylmethylene)am-
before filtration of the yellowish solu- monium iodide, Eschenmoser’s reagent
tion, under suction, through a sintered (56,59)] to give the 2-(N,N-dimethy-
glass funnel containing Grade I alumina laminomethyl)pyrrole [73]; heating of this
(300 g). in acetic acid gives a 52% yield of [52]
33
K.M. Smith
(18,70); and (ii) hydrolysis of the pyrrole 6. Evaporation gives a residue which is
[74] to give pyrrole [75] which is tetramer- chromatographed on silica gel, eluted
ized to give [52] in 44% yield by heating in with dichloromethane to give OEP
acetic acid containing potassium ferri- [52], with an average 55% yield (240
cyanide (35,63). Most recently, the Bar- mg).
ton–Zard pyrrole synthesis (see above) (8) Alternatively, tetramerization of 2,5-di-
has greatly simplified preparative approach- unsubstituted pyrroles [e.g., 72] in the
es to monopyrroles of the type [58]; lithium presence of reagents that can provide the
aluminum hydride (CAUTION: reacts vio- four meso-methine carbons of the product
lently with moisture) reduction, followed by can be used. Cyclization of 3,4-diethylpyr-
tetramerization of the resulting pyrrole-2- role [72] with formaldehyde affords
carbinol [76] under acidic conditions, gives 55%–75% yields of OEP [52] (61).
[52] in 55% yield (2,55). If the 3- and 4-substituents on the
monopyrrole component are not identical,
❖ Procedure 4. Synthesis of OEP [52] mixtures will usually result due to acid cat-
from Pyrrole [58] (54) alyzed pyrrole ring scrambling reactions
(49). Thus, acid catalyzed tetramerization
1. Ethyl 3,4-diethylpyrrole-2-carboxylate of pyrrole [77] will result in production of
[58] (see above, 657 mg) is added drop- a mixture of the four etioporphyrin type
wise at 0°–5°C to a stirred solution of isomers [6–9]. However, a method has
lithium aluminum hydride (320 mg; been devised which does produce only
Sigma) in dry tetrahydrofuran (15 mL). etioporphyrin I [6] from tetramerization of
The mixture is stirred for 2 hours at a pyrrole related to [77]; treatment of 2-
0°–5°C before destroying the excess (N,N-dimethylaminomethyl)pyrroles [e.g.,
lithium aluminum hydride by addition 78] with methyl iodide gives [79], which
of ethyl acetate. has a leaving group that is labile even under
2. It is then poured into saturated aqueous neutral conditions (i.e., no acid), which
ammonium chloride, extracted with would cause pyrrole ring scrambling, is
ethyl acetate (3 times 10 mL), washed present (53,54). This, quaternized in meth-
with aqueous sodium chloride, and dried anol containing potassium ferricyanide (to
over anhydrous magnesium sulfate. accomplish rapid in situ oxidation of the
labile porphyrinogen intermediate), gives a
3. The solution is evaporated to dryness
good yield of pure etioporphyrin I [6].
under vacuum before addition of
dichloromethane (15 mL).
❖ Procedure 5. Synthesis of
4. To this solution is added dimethoxy- Etioporphyrin I [6] from Pyrrole [78]
methane (0.7 mL; Sigma) and toluene
p-sulfonic acid (110 mg), and the mix- 1. Benzyl 4-ethyl-5-(N,N-dimethylamino-
ture is stirred for 12 hours at room tem- methyl)-3-methylpyrrole-2-carboxylate
perature. Aerial oxidation occurs under (53,54) (1.88 g) is dissolved in tetrahy-
these conditions, but chloranil (Sigma) drofuran (600 mL), and 10% palladi-
can also be used, although without any um on carbon (500 mg; Sigma) is
improvement in yield. added.
5. The mixture is washed with aqueous 2. The resulting mixture is stirred under
sodium bicarbonate, and the organic hydrogen at room temperature for 12
layer is dried over anhydrous magne- hours before the catalyst is removed,
sium sulfate. and the solvent is evaporated to dryness.
34
Syntheses of Tetrapyrroles
35
K.M. Smith
developed within my own research group. I of haemoproteins. J. Chem. Soc. [Perkin 1] 1771-
would like to thank all of my collaborators, 1781.
15.Cavaleiro, J.A.S., G.W. Kenner, and K.M. Smith.
over the years, for their important contri- 1974. Biosynthesis of protoporphyrin-IX from copro-
butions to our group effort; their names porphyrinogen-III. J. Chem. Soc. [Perkin 1] 1188-
can be found in the various literature cita- 1194.
16.Chu, T.C. and E.J.-H. Chu. 1955. Paper chromatog-
tions. I would also like to thank the raphy of iron complexes of porphyrins. J. Biol. Chem.
National Institutes of Health (HL 22252) 212:1-7.
and the National Science Foundation 17.DiNello, R.K. and D.H. Dolphin. 1981. Evidence
for a fast (major) and slow (minor) pathway in the
(CHE 99-04076) for uninterrupted sup- Schumm devinylation reaction of vinylporphyrins. J.
port of our efforts over the past 24 years. Org. Chem. 46:3498-3502.
18.cf. Eisner, U., A. Lichtarowicz, and R.P. Linstead.
1957. Chlorophylls and related compounds. Part VI.
REFERENCES The synthesis of octaethylchlorin. J. Chem. Soc. 733-
739.
1.Adler, A.D., F.R. Longo, J.D. Finarelli, J. Goldmach- 19.Fischer, H. 1955. Preparation of hemin. Org. Synth.
er, J. Assour, and L. Korsakoff. 1967. A simplified 3:442.
synthesis for meso-tetraphenylporphyrin. J. Org. 20.Fischer, H. and H. Orth. 1934. Die Chemie des
Chem. 32:476. Pyrrols. Vol. 1. Akademische Verlagsgesellschaft,
2.Aoyagi, K., T. Haga, H. Toi, Y. Aoyama, T. Mizutani, Leipzig.
and H. Ogoshi. 1997. Electron deficient porphyrins. 21.Fischer, H. and H. Orth. 1934. Die Chemie des
III. Facile syntheses of perfluoroalkylporphyrins Pyrrols. Vol. 1, p. 333. Akademische Verlagsge-
including water soluble porphyrin. Bull. Chem. Soc. sellschaft, Leipzig.
Jpn. 70:937-943. 22.Fischer, H. and H. Orth. 1937. Die Chemie des
3.Arsenault, G.P., E. Bullock, and S.F. MacDonald. Pyrrols. Vol. 2, part 1. Akademische Verlagsge-
1960. Pyrromethanes and porphyrins therefrom. J. sellschaft, Leipzig.
Am. Chem. Soc. 82:4384-4389. 23.Fischer, H. and H. Orth. 1937. Die Chemie des
4.Badger, G.M., R.A. Jones, and R.L. Laslett. 1964. Pyrrols. Vol. 2, part 1, p 358. Akademische Verlagsge-
Porphyrins. VII. The synthesis of porphyrins by the sellschaft, Leipzig.
Rothemund reaction. Aust. J. Chem. 17:1028-1035. 24.Fischer, H. and A. Stern. 1940. Die Chemie des
5.Baltazzi, E. and L.I. Krimen. 1963. Recent advances Pyrrols. Vol. 2, part 2. Akademische Verlagsge-
in the chemistry of pyrrole. Chem. Rev. 63:511. sellschaft, Leipzig.
6.Barnett, G.H., M.F. Hudson, and K.M. Smith. 1975. 25.Fischer, H. and A. Stern. 1940. Die Chemie des
Concerning meso-tetraphenylporphyrin purification. J. Pyrrols. Vol. 2, part 2, p. 48. Akademische Verlagsge-
Chem. Soc. [Perkin 1] 1401-1403. sellschaft, Leipzig.
7.Barrett, J. 1959. Detection of hydroxyl groups in por- 26.Fischer, H. and A. Stern. 1940. Die Chemie des
phyrins and chlorins. Nature 183:1185-1186. Pyrrols. Vol. 2, part 2, p. 52. Akademische Verlagsge-
8.Barton, D.H.R., J. Kervagoret, and S.Z. Zard. 1990. sellschaft, Leipzig.
A useful synthesis of pyrroles from nitroolefins. Tetra- 27.Fischer, H. and A. Stern. 1940. Die Chemie des
hedron 46:7587-7598. Pyrrols. Vol. 2, part 2, p. 73. Akademische Verlagsge-
9.Battersby, A.R. and E. McDonald. 1975. Biosynthe- sellschaft, Leipzig.
sis of porphyrins, chlorins, and corrins. In K.M. Smith 28.Fischer, H. and A. Stern. 1940. Die Chemie des
(Ed.), Porphyrins and Metalloporphyrins. Elsevier, Pyrrols. Vol. 2, part 2, p. 108. Akademische Verlagsge-
Amsterdam. sellschaft, Leipzig.
10.Bonnett, R., I.H. Campion-Smith, and A.J. Page. 29.Fischer, H. and K. Zeile. 1929. Synthese des haemato-
1977. Protiodevinylation: the Schumm reaction of porphyrins, protoporphyrins und haemins. Liebigs
vinylporphyrins. J. Chem. Soc. [Perkin 1] 68-71. Ann. Chem. 468:98-116.
11.Bullock, E., A.W. Johnson, E. Markham, and K.B. 30.Freeman, B.A. and K.M. Smith. 1995. Novel uses of a
Shaw. 1958. A synthesis of coproporphyrin III. J. naturally occurring heterogeneous catalyst to improve
Chem. Soc. 1430-1440. organic syntheses. Proc. NOBCChE 22:227-242.
12.Burbidge, P.A., G.L. Collier, A.H. Jackson, and G.W. 31.Gerlach, B., S.E. Brantley, and K.M. Smith. 1998.
Kenner. 1967. Syntheses and spectra of some meso- Novel synthetic routes to 8-vinyl chlorophyll deriva-
methylated porphyrins. J. Chem. Soc. B 930-937. tives. J. Org. Chem. 63:2314-2320.
13.Cavaleiro, J.A.S., A.M.d’A.R. Gonsalves, G.W. Ken- 32.Gossauer, A. 1974. Die Chemie der Pyrrole. Springer,
ner, and K.M. Smith. 1973. Synthesis of Berlin.
pyrromethanes and a tripyrrane. J. Chem. Soc. [Perkin 33.Grinstein, M. 1947. Studies of protoporphyrin. J. Biol.
1] 2471-2478. Chem. 167:515-519.
14.Cavaleiro, J.A.S., A.M.d’A.R. Gonsalves, G.W. Ken- 34.Iakovides, P. and K.M. Smith. 1996. Syntheses of oxy-
ner, and K.M. Smith. 1974. Total synthesis of deuteri- gen analogues of sulfhemes-A and -C. Tetrahedron
ated derivatives of protoporphyrin-IX for NMR studies 52:1123-1148.
37
K.M. Smith
35.Inhoffen, H.H., J.-H. Fuhrhop, H. Voigt, and H. 52.Morell, D.B. and M. Stewart. 1956. Removal of iron
Brockmann, Jr. 1966. Formylierung der meso- from hemins. Aust. J. Exp. Biol. Med. Sci. 34:211-
Kohlenstoffeatome von alkylsubstituierten porphyri- 218.
nen. Liebigs Ann. Chem. 695:133. 53.Nguyen, L.T., M.O. Senge, and K.M. Smith. 1996.
36.Jackson, A.H., R.K. Pandey, K.R.N. Rao, and E. Simple methodology for syntheses of porphyrins pos-
Roberts. 1985. Reactions on solid supports, part II: a sessing multiple peripheral substituents with an ele-
convenient method for synthesis of pyrromethanes ment of symmetry. J. Org. Chem. 61:998-1003.
using a Montmorillonite clay as catalyst. Tetrahedron 54.Nguyen, L.T. and K.M. Smith. 1996. Syntheses of
Lett. 26:793-796. type-I porphyrins via monopyrrole tetramerization.
37.Jackson, A.H. and K.M. Smith. 1973. In J.W. ApSi- Tetrahedron Lett. 37:7177-7180.
mon (Ed.), Total Synthesis of Natural Products, Vol. 1, 55.Ono, N., H. Kawamura, M. Bougauchi, and K.
p. 143-278. John Wiley & Sons, New York. Maruyama. 1990. Porphyrin synthesis from nitrocom-
38.Jackson, A.H. and K.M. Smith. 1984. In J.W. ApSi- pounds. Tetrahedron 46:7483-7496.
mon (Ed.)Total Synthesis of Natural Products Vol. 6, 56.Ono, M., R. Lattmann, K. Imomota, C. Lehmann, T.
p. 237-280. John Wiley & Sons, New York. Früh, and, A. Eschenmoser. 1985. Monopyrrole pre-
39.Kadish, K., K.M. Smith, and R. Guilard. (Eds.). cursors for the synthesis of uroporphyrinogenoctani-
1999. The Porphyrin Handbook. Academic Press, trile. Croat. Chem. Acta 58:627.
Boston. 57.Rothemund, P. 1935. Formation of porphyrins from
40.Kenner, G.W., S.W. McCombie, and K.M. Smith. pyrrole and aldehydes. J. Am. Chem. Soc. 57:2010-
1973. Protection of porphyrin vinyl groups. A synthe- 2011.
sis of coproporphyrin-III from protoporphyrin-IX. 58.Rothemund, P. and A.R. Menotti. 1941. Porphyrin
Liebigs Ann. Chem. 1329-1338. studies. IV. The synthesis of α,β,γ,δ-tetraphenylpor-
41.Kenner, G.W., S.W. McCombie, and K.M. Smith. phyrin. J. Am. Chem. Soc. 63:267-270.
1973. Separation and oxidative degradation of chloro- 59.Schreiber, J., H. Maag, N. Hashimoto, and A.
phyll derivatives. J. Chem. Soc. [Perkin 1]. 2517-2523. Eschenmoser. 1971. Dimethyl(methylene)ammonium
42.Kenner, G.W., J.M.E. Quirke, and K.M. Smith. 1976. iodide. Angew. Chem. Int. Ed. Engl. 10:330-331.
Transformations of protoporphyrin-IX into harderopro- 60.Schumm, O.Z. 1928. Preparation of hemin derivatives
phyrin, pemptoporphyrin, chlorocruoroporphyrin and by pyro reactions. Z. Physiol. Chem. 178:1-18.
their isomers. Tetrahedron 32:2753-2756. 61.Sessler, J.L., A. Mozaffari, and M.R. Johnson. 1991.
43.Kleinspehn, G.G. 1955. Pyrrole series XXVII. Novel 3,4-Diethylpyrrole and 2,3,7,8,12,13,17,18-octaethyl-
route to certain 2-pyrrolecarboxylic esters and nitriles. porphyrin. Org. Synth. 70:68-78.
J. Am. Chem. Soc. 77:1546-1548. 62.Shiau, F.-Y., R.K. Pandey, S. Ramaprasad, T.J.
44.Labbe, R.F. and G. Nishida. 1957. A new method of Dougherty, and K.M. Smith. 1990. The isomeric
hemin isolation. Biochim. Biophys. Acta 26:437. mono-acetyl-mono-(1-hydroxyethyl)-deuteroporphyrins:
45.Lee, S.-J.H., N. Jagerovic, and K.M. Smith. 1993. Use synthesis, characterization and use for syntheses of
of the chlorophyll derivative, purpurin-18, for synthe- regioselectively methyl- and vinyl-deuterated hemins.
ses of sensitizers for photodynamic therapy. J. Chem. J. Org. Chem. 55:2190-2195.
Soc. [Perkin 1] 2369-2377. 63.cf. Siedel, W. and F. Winkler. 1943. Oxydation von
46.Lemberg, R., B. Bloomfield, P. Caiger, and W. Lock- pyrrolderivaten mit bleitetraacetat. Liebigs Ann.
wood. 1955. Porphyrins with formyl groups. V. Isola- Chem. 554:162-201.
tion of porphyrin a from heart muscle and determina- 64.Smith, K.M. (Ed.). 1975. Porphyrins and Metallopor-
tion of its heme a content. Aust. J. Exp. Biol. Med. Sci. phyrins. Elsevier, Amsterdam.
33:435-450. 65.Smith, K.M., G.M.F. Bisset, and M.J. Bushell. 1980.
47.Lindsey, J.S. 1999. Synthesis of meso-substituted por- Partial syntheses of optically pure methyl bacterio-
phyrins. In K. Kadish, K.M. Smith, and R. Guilard pheophorbides c and d from methyl pheophorbide a. J.
(Eds.), The Porphyrin Handbook, Chapter 2. Acade- Org. Chem. 45:2218-2224.
mic Press, Boston. 66.Smith, K.M., E.M. Fujinari, K.C. Langry, D.W.
48.Lindsey, J.S., I.C. Schreiman, H.C. Hsu, P.C. Kear- Parish, and H.D. Tabba. 1983. Manipulation of vinyl
ney, and A.M. Marguerettaz. 1987. Rothemund and groups in protoporphyrin-IX: introduction of deuteri-
Adler-Longo reactions revisited: synthesis of tetraphenyl- um and carbon-13 labels for spectroscopic studies. J.
porphyrins under equilibrium conditions. J. Org. Am. Chem. Soc. 105:6638-6646.
Chem. 52:827-836. 67.Smith, K.M., D.A. Goff, and D.J.Simpson. 1985.
49.Mauzerall, D. 1960. The thermodynamic stability of Meso-substitution of chlorophyll derivatives: direct
porphyrinogens. J. Am. Chem. Soc. 82:2601-2605. route for transformation of bacteriopheophorbides-d
50.Mironov, A.F., T.R. Ovsepyan, R.P. Evstigneeva, and into bacteriopheophorbides-c. J. Am. Chem. Soc.
N.A. Preobrazenskii. 1965. Synthetic studies in the 107:4946-4954.
dipyrrylmethane series. Synthesis of 4,4′-dimethyl- 68.Strain, H.H. and W.A. Svec. 1966. In L.P. Vernon and
3,3′-bis(β-carbomethoxyethyl)dipyrrylmethane. Zh. G.R. Seely (Eds.), p. 21. The Chlorophylls, Academic
Obshch. Khim. 35:324-328. Press, New York.
51.Morell, D.B., J. Barrett, and P.S. Clezy. 1961. The 69.Thudichum, J.L.W. 1867. Rpt. Med. Off. Privy
prosthetic groups of cytochrome oxidase. 1. Purifica- Council, App. 7. 10:152.
tion as porphyrin a and conversion into haemin a. 70.Whitlock, H.W. and R. Hanauer. 1968. Octaethyl-
Biochem. J. 78:793-797. porphyrin. J. Org. Chem. 33:2169-2171.
38
General Laboratory Methods
3 for Tetrapyrroles
39
J.C. Bommer and P. Hambright
Compounds by Milgrom in 1998 (84) is range and is valuable for the separation of
suggested reading both for beginners and compounds having similar retention factor
experienced workers desiring a broader (Rf ) values. High-pressure liquid chroma-
view of the field. The Porphyrin Handbook, tography (HPLC) can separate and quanti-
a ten volume series edited by Kadish, tate samples on the microgram scale. Many
Smith, and Guilard was published in 2000 porphyrins and diamagnetic metallopor-
(61a). Dupré maintains “Porphynet, the phyrins fluoresce under UV irradiation,
porphyrin site” (www. porphyrin.net), and a hand-held long wavelength 366 nm
which lists 40 currently available mono- UV lamp is an aid in monitoring the sepa-
graphs on porphyrins. This site is a source rations. Porphyrin solution should always
of general information to those working in be filtered before application to columns or
all aspects of the porphyrin field, and lists plates, and one should avoid “overloading”
vendors, e-mail addresses of porphyrin the media with too high a concentration of
chemists, future conferences, etc. The Jour- pigment. A small amount of the column
nal of Porphyrins and Phthalocyanines began material often accompanies the purified
publication in 1997. The Chemical Ab- porphyrin and tends to complicate subse-
stracts Service “CA Selects: Porphyrins” quent CHN determinations. Filtering the
gives biweekly coverage of books, disserta- solution through a 47 mm 0.2- or 0.45-µm
tions, reports, reviews, patents, and journal nylon filter (Millipore, Bedford, MA, USA)
abstracts dealing with most areas of por- aids in the removal of small particles. Such
phyrin chemistry. The free Medline filters are also useful for the collection of
(www.ncbi.nih.gov/PubMed) has over small quantities of a valuable precipitate.
23 000 listings under “porphyrins”. Our seventy years experience at the bench
The chapter by Smith in this volume indicates that accidents do happen, and we
describes the nomenclature of the por- have tried to indicate throughout this chap-
phyrin nucleus, and the preparations of ter where accidental loss of a valuable por-
numerous 3,8-disubstituted deuteropor- phyrin can be avoided. A clean heavy-wall
phyrins, chlorophylls, and recent advances empty flask should be between the water
in porphyrin synthesis. We intend to cover aspirator and the filtration flask, and the fil-
laboratory techniques useful for the syn- ter flask should always be securely clamped
thesis, purification, and analysis of water- to a ring stand. It is worthwhile to date all
soluble and insoluble porphyrins and met- sample vials containing solid porphyrins.
alloporphyrins, and general aspects of the This is an aid in rapidly locating in your
solution chemistry of such macrocyclic laboratory book at a later time the synthet-
species. While only a few examples are ic or isolation method used for the particu-
mentioned, such procedures can usually be lar preparation, and also because certain
applied to whole classes of compounds. compounds have unexpectedly short shelf-
lives. Columns and TLC work should be
done under the hood to minimize exposure
2. CHROMATOGRAPHY to the often hazardous solvents, and it is
worthwhile to wear disposable gloves in all
Most porphyrins are purified by some porphyrin manipulations. Lim has edited a
form of chromatography. Column chro- volume on porphyrin chromatography to
matography is used for isolation of 100 mg appear in 2000 (75a). White, Bachman,
to multigram quantities of material. Thin and Burnham (116) have an excellent chap-
layer chromatography (TLC) on coated ter on chromatography of porphyrins and
glass plates is employed in the 1 to 100 mg metalloporphyrins, and Varaldi, Longo and
40
General Laboratory Methods for Tetrapyrroles
Adler (114) have summarized nonchro- methylene chloride, acetone, ethyl acetate,
matographic methods (electrophoresis, tetrahydrofuran, acetonitrile, pyridine, n-
countercurrent distribution, sublimation, propanol, ethanol, methanol, acetic acid,
extraction, precipitation, and crystalliza- and water. Numerous examples of useful
tion) for the purification of porphyrins. solvent mixtures and sorbents are men-
tioned in later sections of this chapter for
2.1. Column Chromatography particular compounds. For silica gel or alu-
mina chromatography, we find that ethyl
Most column chromatography involves acetate in various concentrations (1%–
glass columns of varying lengths, although 15%) mixed with CHCl3 or CH2Cl2, in
some prefer plastic columns that can be cut combination with increasing amounts of
to isolate the porphyrin bands. Others methanol, is effective for the separation of
claim that better separations occur in pear- a wide range of synthetic porphyrins and
shaped separating funnels. In contrast to for porphyrin esters in general.
gravity techniques, some groups prefer Many groups prefer to evaporate a con-
“flash chromatography”, in which a low centrated organic solution of the impure
pressure (5–15 psi) of an inert gas is porphyrin(s) in the presence of the packing
applied to force the sample and solvent material. This solid is then placed on top of
rapidly through the sorbent. A glass-wool the column, and elution is begun with a
plug and sand are placed on the bottom of relatively nonpolar solvent. One rationale
open columns, or commercial columns for this procedure is that many porphyrins
with glass frits overlaid with a layer of sand are relatively insoluble in solvents of low
are employed. The “dry column” technique polarity. For example, the room temperature
(116) in which the column is packed with solubility of nickel(II) etioporphyrin I
dry absorbent is sometimes useful, but in (µg/mL) in various solvents is: chloroform
general, better separations are achieved (2.2 × 103), methylene chloride (1.6 × 103),
when the column is loaded with a slurry of ethyl acetate (2.2 × 102), cyclohexane
the packing material, which is prepared by (approximately 7), followed by acetonitrile
stirring the solid into a large volume of the cyclohexane, ethanol, methanol, and
liquid. A layer of sand is added to the top hexane at approximately 2 µg/mL. Solu-
of the column so that the bed is not dis- bility depends on both the type of por-
turbed by solvent addition. It is best to dis- phyrin ring and the identity of the coordi-
solve the impure porphyrin mixture in a nated metal; for example, the solubilities
relatively nonpolar solvent, so that the in chloroform are Ni-Etio I (2.2 × 103),
majority of the porphyrin and impurities are Ni-octaethylporphyrin (5.1 × 103), VO-
initially adsorbed near the top of the col- Etio I (6.8 × 103), and VO- octaethylpor-
umn. This is followed by selective elution phyrin (OEP) (8.2 × 103). OEP has eight
with a solvent, solvent mixture, or successive ethyl groups in the β- pyrrole positions,
solvent mixtures of higher polarity, which while etioporphyrin I has four methyl and
move the desired porphyrin away from the four ethyl substituents.
adsorbed impurities, or separate a mixture Commercial CHCl3 is often stabilized
of porphyrins one from another. One with approximately 0.75% ethanol, and
elutropic series of solvents useful in por- the polar ethanol affects the separations. If
phyrin work are n-hexane, petroleum necessary, the ethanol can be removed by
ether, cyclohexane, carbon tetrachloride, shaking the CHCl3 with an equal volume
toluene, benzene, diethyl ether, trichloroeth- of water, drying with CaCl2, and distilling
ylene, 1,2-dichloroethane, chloroform, over P2O5, collecting only the middle frac-
41
J.C. Bommer and P. Hambright
tions. Some groups directly use CHCl3 sta- water, and grade V contains 15% water.
bilized with amylenes. CHCl3 and Other adsorbents include Florisil (60–
CH2Cl2 (both carcinogenic to laboratory 100 mesh), calcium carbonate (for proto-
animals) often contain HCl, which can porphyrin IX dimethyl ester [DME]),
protonate the porphyrin free base H2-P sugar (for chlorophylls), talc (to remove
into the charged diacid H4-P2+, leading to reduced porphyrin impurities), 80% mag-
poor separations. To test for HCl, the inex- nesol-20% cellulose (for magnesium por-
pensive and readily available tetraphenyl- phyrin esters) microcrystalline cellulose,
porphyrin (H2-TPP) is dissolved in ben- Kieselguhr, Fullers earth, and diatoma-
zene, producing a red solution of the free ceous earth (for sulfonated deuteropor-
base. Several milliliters of CHCl3 or phyrins). Bachmann and Burnham (8) sep-
CH2Cl2 are added to a similar volume of arated, on the macro or micro scale, the
the red reagent, and a green solution of the methyl esters of meso, proto, copro I, and
porphyrin diacid forms if HCl is present. uro I by gel filtration on Sephadex LH-20
The acid can be removed by shaking these (Amersham Pharmacia Biotech, Piscat-
solvents with an aqueous solution of 0.1 M away, NJ, USA), with 1/1 CHCl3-MeOH
NaOH or 0.1 M ammonia until the test (containing 1 g Tris base/L) as the eluant.
solution remains red. The solvents are then Sephadex has been used to remove unde-
dried over solid anhydrous CaCl2, sired salts from porphyrin solutions.
Na2SO4, or K2CO3 before use. For the Polyamide columns are useful for iron por-
separation of a metalloporphyrin that is phyrin carboxylic acids and acid-labile
stable in acid from small amounts of the metalloporphyrins (29). Reverse phase
unmetalated free base, the mixture is often octadecyl (C18) columns using methanol-
dissolved in chloroform containing 2% tri- buffered water mixtures can separate water-
fluoroacetic acid to purposely protonate soluble anionic porphyrins by total por-
the undesired H2-P into H4-P2+. The phyrin charge (110). Cationic porphyrins
green diacid will tend to stick near the top of differing charge can be resolved on
of the column, resulting in a better separa- silica gel plates developed with saturat-
tion than can be obtained simply with the ed KNO 3:water:acetonitrile in a 1:1:8
metalloporphyrin and its free base. mixture (10). Cation exchange columns
A variety of packing materials have been (-SO3-H+ or -SO 3-Na+) are used to separ-
used in column techniques. Silica gel ate divalent metal cations from aqueous
(100–200 mesh) and alumina (80–200 solutions of anionic water-soluble por-
mesh) are perhaps the most common phyrins. For cationic porphyrins, anion
-
choices. Activated alumina can be pur- exchange columns [-N(CH3)3+Cl ] replace
chased in acidic, neutral, or basic forms, tosylates, or the precipitating agents iodide,
and these activated (Grade I) materials or perchlorate in solution with chloride.
often adsorb the porphyrin so strongly that TLC or absorption spectrophotometry
it becomes difficult to remove from the (350–800 nm) are used to monitor the
column. The alumina can be deactivated fractions finally coming off of columns and
by adding water to the adsorbent (wt/wt) also to determine what impurities are pre-
in a stoppered Erlenmeyer flask. The mix- sent prior to running a column. In many
ture is shaken to remove all lumps and cases, a given fraction must be rechro-
allowed to cool to room temperature matographed many times, filtered, and
before use. Activity grade I has no added then recrystallized to obtain pure material.
water, grade II contains 3% water, grade III It is advantageous to use a long spatula or
contains 6% water, grade IV contains 10% aspirator to remove the obvious impurity
42
General Laboratory Methods for Tetrapyrroles
band(s) from the top of the column once and tetra-aryl porphyrins on alumina. With
development has begun. Sometimes, it is sterically hindered compounds not allowing
best to carefully empty the column and dimerization, such as tetrakis(2,4,6-tri-
slice out the bands of interest for further methyl-phenyl)porphyrin, the monohyd-
treatment. It is prudent to empty the col- roxy Fe(III) porphyrin would be obtained.
umn of packing material after the separa- To isolate the solid OEP dimer, the
tion, as some sorbents swell and will break solution was filtered through a 0.45-µm
the column upon prolonged standing. filter, the solvent removed on a rotary evap-
Chromatography is an art, and often trial orator at 60°C, and the purple solid dried
and error is the only means of discovering overnight at 60°C in a vacuum oven. To
an efficient separation technique for new transform the dimer into the Fe(OEP)Cl
and unusual compounds. The following monomer, the CH2Cl2 solution was fil-
procedure illustrates a typical column chro- tered, saturated with HCl(g) followed by
matographic purification of a water insolu- N2(g) to remove excess HCl(g), and the
ble iron porphyrin, crude iron(III) OEP solvent was then removed under the hood
chloride (section 10.2, Procedure 4). on a steam bath. The monomer was dried
TLC of this iron porphyrin in HCl-free overnight at 70°C in a vacuum oven. The
CH2Cl2 on an alumina plate showed one HCl(g) could come from a lecture bottle
major band with an Rf value of 0.6 and a with a trap between the cylinder and solu-
moderate amount of substance that did not tion, or blown into the CH2Cl2 from the
move from the origin. The absorption spec- head-space of a reagent bottle of HCl.
tra of the impure compound in the same
solvent gave bands (and relative peak ❖ Procedure 1. Column Chromato-
heights) at 633.5 nm (1.0), 535.5 nm (1.9), graphy of Iron-Octaethylporphyrin
and 506 nm (1.8), indicating that the
major form in solution was the Fe(OEP)Cl 1. Wearing gloves and under the hood, a
monomer. If present, the metal-free H2- 3- × 50-cm glass column is fitted with
OEP would be the first species to elute, glass wool, tamped down with a meter
having bands in CH2Cl2 at 619.5 nm stick, and overlaid with approximately
(1.0), 566.0 nm (1.5), 531.5 nm (2.7), and 2 cm of acid-washed sea sand.
497.5 nm (2.9). In this particular example,
no H2-OEP was found, and approximately 2. The bottom of the column is placed in
500 mL of the red metalloporphyrin solu- a small diameter circular metallic
tion was collected. An approximately 2-cm clamp (to prevent the heavy column
thick impurity band was present at the ori- from sliding into the collection
gin, which did not move in CH2Cl2. The beaker), and the upper portion is held
addition of ethanol-stabilized chloroform vertical with a vinyl covered extension
eluted an unidentified minor band from the clamp (not overly tightened such that
origin material, and thus chromatography the column cracks).
in CHCl3 would have been undesirable. 3. A beaker is placed under the column,
The spectra of the eluted metalloporphyrin which is then loaded with 250 g of
in CH2Cl2 showed peaks at 588.0 nm (1.0) Grade IV 80 to 200 mesh alumina (A-
and 560.0 nm (1.7) indicating the 540; Fisher Scientific, Pittsburgh, PA,
Fe(OEP)Cl monomer had been trans- USA) that has been slurried in 400 mL
formed into the u-oxo dimer (OEP)FeIII- CH2Cl2. The slurry is poured into the
O-FeIII(OEP), which is typical behavior for column with the Teflon stopcock
most iron(III) protoporphyrin type esters open, taking care that solvent is always
43
J.C. Bommer and P. Hambright
present above the solid phase. If the the zinc invariably incorporates into a frac-
upper layer runs dry, more solvent tion or all of the H2-P on the plate forming
should be added and the upper layer the unwanted Zn-P complex. For fairly
stirred with a glass rod until no more basic porphyrins, shaking the organic sol-
bubbles percolate up through the slurry. vent containing the Zn-P with an approxi-
4. After settling, sand is added to a depth mately 3 M solution of HCl, followed by a
of approximately 2 cm. The alumina water wash, is often sufficient to remove
fills 64% of the column, leaving room the zinc from Zn-P. Another technique for
such that a reasonable amount of solu- less basic porphyrins involves adding triflu-
tion or solvent can be added at one oroacetic acid to a chloroform solution
time. containing the Zn-P and stirring until the
5. Two grams of impure iron OEP is dis- absorption spectrum of neutralized samples
solved in 350 mL of CH2Cl2, filtered, indicate that the two bands between 500
and slowly poured onto the column and 700 nm, due to Zn-P, have been trans-
with the stopcock in the open position. formed into the four-banded spectra char-
acteristic of many H2-P derivatives. The
6. Once all of the metalloporphyrin acidic mixture should be extracted repeat-
reaches the sand layer, small portions of edly with deionized water before shaking
solvent are added until the entire com- with a base. Immediate neutralization with a
pound has been absorbed, and then the dilute ammonia or NaOH often forms reac-
upper space is filled with methylene tive Zn(NH3)×2+ or Zn(OH)+/Zn(OH)3-
chloride. complexes, which then reintroduce zinc
7. The void volume of the solid phase is back into the porphyrin.
approximately 200 mL, and since the All TLC manipulations should be done
metalloporphyrin runs near the solvent under the hood and in dim light, as certain
front, about 200 mL of pure solvent is porphyrins and metalloporphyrins have a
collected before the porphyrin enters tendency to photodecompose on the plates.
the collection beaker. A line of closely spaced spots resolves better
than material simply streaked onto the
2.2. Thin Layer Chromatography plates, and the walls of the developing
chamber can be lined with absorbent paper
To purify small amounts of porphyrins towels to saturate the atmosphere in the
in the 1 to 100 mg range, or larger enclosure. A single- edged razor blade is
amounts if necessary, TLC plates usually useful for scraping off the various fractions,
give better resolutions than do columns, which should then be ground into a fine
with the resolution inversely proportional powder. Methanol often must be added to
to the thickness of the plate. Commercial- deactivate the packing material before the
ly, 200-, 250-, 500-, and 1000-µm silica porphyrin can be eluted from the powder
gel plates are available, and 2000-µm or with an organic solvent. With low Rf mate-
thicker preparative plates can be prepared rial, pretreatment with acetic acid is some-
in house. Alumina, microcrystalline cellu- times necessary.
lose, RP-C18-silica gel, Kieselguhr, An excellent TLC method for determin-
polyamide and Florisil (Mg silicate) plates ing the purity of methyl esters of natural
can be purchased. Fluorescent TLC plates porphyrins is one originally developed by
(and fluorescent column material) are to be Chu and Chu (25) and modified by Elder
avoided. These plates often contain Zn2+ (33). The system consists of chloroform,
salts as a component of the phosphor, and kerosene, and methanol in a volume ratio
44
General Laboratory Methods for Tetrapyrroles
45
J.C. Bommer and P. Hambright
work are chosen for their sensitivity it is injected, will often change the relative
towards the compounds of interest, e.g., size of the eluted peaks to give a good indi-
wavelengths near the Soret maxima for cation that artifacts are indeed a problem.
UV/VIS absorbance detection systems. Generally, we have found that normal
This can give a good idea of purity relative phase HPLC of porphyrins tends to be less
to contamination by other tetrapyrroles, reproducible than reverse phase chro-
but tells one nothing about what other matography. Very precise conditioning of
types of dissimilar organic or inorganic the column seems to be required, and
compounds may reside in the sample. The therefore, we have moved to reverse phase
same can be said for fluorescence detection, systems exclusively. Systems that seem to
except in this case, another class of conta- work well for monocarboxylic through
minant, nonfluorescent or weakly fluores- tetracarboxylic tetrapyrroles use C-18
cent metallo-tetrapyrroles, may be missed columns with eluant consisting of tetrabut-
or under-represented along with other ylammonium phosphate (0.002–0.01 M)
nonfluorescent compounds. Also, only at about pH 2.7 and methanol or acetoni-
substances that are actually eluted from the trile as organic modifiers. Some systems for
column are detected. In the case of more polar tetrapyrroles rely on sodium or
tetrapyrroles, there are a number of conta- potassium phosphate buffers at near neu-
minants that will stick more or less irre- tral pH with concentrations from 0.01 to
versibly on the column, such as partially 0.1 M containing methanol or acetonitrile.
esterified porphyrins in the instance of sili- The higher salt concentrations are most
ca chromatography of porphyrin car- useful for the most polar tetrapyrroles,
boxylic acid esters and black polypyrroles, such as highly sulfonated species or uro-
etc., from synthesis of meso-substituted porphyrins. It is best to use some type of
porphyrins. Thus, initially, TLC is useful gradient system unless analyzing very pure
in determining whether such contaminants samples, so that less polar contaminants are
are present in the HPLC samples. eluted from the column in a reasonable
HPLC chromatography of tetrapyrroles period of time.
is particularly prone to artifacts. This prob- As an example, the system we use to
ably arises from the relative insolubility of evaluate a mixture of porphyrin carboxylic
most of these compounds and their ten- acids, which contain from 2 to 8 carboxyl
dency to aggregate. This is especially pro- groups is as follows: Hamilton PRP-1 poly-
nounced in reverse phase chromatography meric supported C-18 column with a gra-
of some of the less aqueous soluble por- dient from 0% acetonitrile in 0.01 M sodi-
phyrins such as mono- or dicarboxylic acid um phosphate buffer at pH 6.85 and flow
porphyrins or chlorins derived from natur- 2 mL/minute to 25% acetonitrile over 14
al sources and some of the derivatives of minutes, then a further gradient to 80%
meso-tetraphenylporphine. Systems that acetonitrile over 2 minutes.
tend to minimize this problem usually
involve ion-pairing reagents such as tetra-
butylammonium ion and a relatively low 3. RECRYSTALLIZATION
pH. Sometimes, multiple peaks appear on
the chromatogram, which seem to be due As with many organic compounds,
to dimers and higher aggregates that are recrystallization can be a powerful tool for
relatively stable under the chromatograph- the purification of tetrapyrroles. It is gen-
ic conditions. Changing the concentration erally only practical however when more
of injected sample, or the solvent in which than a few milligrams of compound are
46
General Laboratory Methods for Tetrapyrroles
available. It has been our experience that As noted in Section 4, tetrapyrroles are
for most porphyrins, a fairly high level of notorious for retaining solvents, especially
purity, perhaps 80% or better for tetrapyr- if crystallized from a high boiling solvent
role impurities and somewhat less if impu- like N,N-dimethylformamide (DMF) and
rities are not tetrapyrroles, is necessary a certain amount of DMF decomposes
before satisfactory results can be obtained upon heating into dimethylamine, which
by recrystallization. It has also been our binds to certain metals in metallopor-
experience that certain types of porphyrins phyrins. Sometimes these molecules can
do not crystallize well, and recrystallization only be removed by heating the solid near
as a purification step is not productive. the boiling point of the solvent, if the sta-
Type III isomers of the biologically impor- bility of the tetrapyrrole allows. The re-
tant porphyrins for instance are very diffi- moval can be followed by monitoring the
cult to crystallize, but type I isomers crys- increase in absorbance at the Soret peak for
tallize nicely with improvement in isomeric a standard concentration by weight over a
purity as well as removal of other por- period of hours or days. Lower tempera-
phyrin contaminants. Some porphyrins tures can be employed if the substance is
can be recrystallized as the dihydrochlo- placed under vacuum in a vacuum oven or
rides by dissolution in concentrated HCl drying pistol.
and refluxing until constant boiling HCl is
achieved then adding water slowly to the ❖ Procedure 3. Recrystallization of
refluxing solution until a lower normality Protoporphyrin IX Dimethyl Ester
of HCl is achieved in which the porphyrins
have reduced solubility. In general, dissolu- 1. Crude protoporphyrin IX DME (ap-
tion in a minimum amount of low boiling proximately 7 g) is obtained by remov-
solvent in which they have high solubility, ing the iron from 10 g of hemin with
then adding an equal or greater amount of concurrent esterification [Grinstein
a higher boiling solvent in which they have method (29)] and partially purified by
only slight solubility can recrystallize small flash chromatography over silica or alu-
quantities of porphyrins or other tetra- mina eluting with 10% ethyl acetate in
pyrroles. The solution in an open contain- dichloromethane.
er is then left to evaporate at room temper-
ature or slightly warmed on a trivet until 2. The solution containing the porphyrin
the desired degree of crystal formation is (about 300–500 mL) is placed in a
complete. Filtering and washing with the L-rotary evaporating flask, and the
higher boiling solvent completes the opera- removal of solvent under vacuum is
tion. Larger amounts of tetrapyrroles can commenced using a 50° to 60°C bath
be handled more efficiently and rapidly by with continuous addition of 500 mL of
recrystallization on a rotary evaporator. In ethyl acetate. The rate of addition is
this case, the tetrapyrrole is dissolved in a controlled such that the volume in the
low boiling solvent, placed in a round bot- flask approaches about 300 mL con-
tom flask on the rotary evaporator fitted currently with complete addition of the
with a continuous feed tube, and the sol- ethyl acetate.
vent removed under reduced pressure while 3. Completion of the crystallization occurs
adding the higher boiling solvent in which upon standing in the refrigerator for sev-
the compound is less soluble, until the eral hours, at which time the crystals are
desired degree of crystallization is reached. collected by filtration on a Buchner fun-
A typical procedure is described below. nel and washed with ethyl acetate.
47
J.C. Bommer and P. Hambright
4. The product is analyzed by the Elder monomer and 15% of the µ-oxo-bridged
kerosene TLC technique (33), and the P-Fe-O-FeP dimer (91). The CHN analy-
procedure can be repeated as desired if ses and corresponding ratios are always
further purification is necessary by dis- helpful, but the moles of water are usually
solving in minimum dichloromethane determined by calculation based on the
and again exchanging with ethyl acetate. observed values. Crystal structures of
5. The yield is about 5 g, and an overall water-insoluble porphyrins usually contain
purity of about 97% to 98% can be several molecules of the crystallizing sol-
expected after one or two recrystalliza- vent, and tetra-arylporphyrin crystals have
tions. channels throughout the solid that occlude
solvent molecules and act as “porphyrin
sponges” (19). On new compounds, most
4. ANALYSIS OF PORPHYRINS AND authors report CHN and metal analyses,
METALLOPORPHYRINS the absorption spectra and extinction coef-
ficients, infrared (IR) band positions, the
It is often important to know the con- analyzed 1H nuclear magnetic resonance
centration in solution of a particular por- (NMR) data, and the major peaks obtained
phyrin or metalloporphyrin. Since this by mass spectrometry.
class of molecules often contains various It is better to determine solution con-
kinds and amounts of nonporphyrin mate- centrations based on known extinction
rial that are isolated along with the por- coefficients, than to rely on calculations
phyrin itself, simply weighing out a given based on the weight of the sample. Due to
amount of compound may give inaccurate aggregation effects, the spectra should be
results. With water-soluble porphyrins, measured at concentration levels similar to
water molecules are usually indicated by the value quoted in the literature. DiNello
chemical analysis of the isolated solids. The and Chang (29) give band positions and
anionic water-soluble H2-TPPS4 (Figure 1) extinction coefficients in CHCl3 for the
analyzes as Na4H2-TPPS4·10 H2O when free bases of modified natural porphyrin
oven dried at 110°C (83), the cationic pyri- derivatives, and for the corresponding
dinium compound as H2-TMPyP(4)Cl4·4 Fe(II) pyridine hemochromes, measured in
H2O, various phenyl/(4-sulfonatophenyl), 4 M pyridine-0.2 M KOH with the addi-
and phenyl/(N-methyl-4-pyridyl) por- tion of small amounts of the reducing
phyrins contain from 2 to 11 moles of agent sodium dithionite. Caughey and
water, often in nonstoichiometric amounts, coworkers (23) have spectra on other free
as Zn-TPPS4 analyzes for 16.6 H2O. base 3,8-disubstituted deuteroporphyrin
Thermo-gravimetric work (53) indicated DMEs and metallo derivatives (22).
that the latter compound gradually lost Caughey et al. have compiled data on iso-
weight up to 200°C, a constant-weight mers of uroporphyrins and copropor-
region was found in the 220° to 400°C phyrins and other biologically important
range, while rapid decomposition was molecules (21). Fuhrhop and Smith (44)
noted above 400°C. Compounds prepared have five tables of extinction coefficients of
at different times sometimes contain differ- various porphyrins and metalloporphyrins
ing amounts of water. The Fe(III)- in organic solvents and in aqueous sodium
TMPyP(4) is a monomer in acid, but dodecyl sulfate solutions. The spectra of
when precipitated from acid solution by centrally N-alkylated porphyrins and met-
the addition of 1 M HClO4, Mossbauer alloporphyrins are found in Lavallee’s
spectra on the solid indicated 85% monograph (71), and James and coworkers
48
General Laboratory Methods for Tetrapyrroles
(83) give data on well-characterized water- Finally, there are a number of laboratory
soluble porphyrins and their precursors. procedures to determine porphyrin con-
Methods for the determination of por- centrations. In Drabkin’s method, iron
phyrins in biological material have been protoporphyrin IX is digested with hydro-
reviewed (103). gen peroxide in basic solution, the liberat-
Figure 1. The structures of some water-soluble porphyrins and several of their precursors. Each compound is the porphyrin
with the indicated substituents on the four meso (5, 10, 15 and 20) positions. H2-TPP: meso-tetraphenylporphyrin; H2-
TPyP(4), meso-tetrakis(4-pyridyl)porphyrin; H2-TMPyP(X): the ortho (2), meta (3) and para (4) isomers of meso-tetrakis(N-
methyl-X-pyridyl)porphyrin; H2-TPPC4, meso-tetrakis(4-carboxyphenyl)porphyrin; H2-TPPS4: meso-tetrakis(4-sul-
fonatophenyl)porphyrin; H2-TAPP, meso-tetrakis(4-N,N,N-trimethylanilinium)porphyrin. A fuller explanation of the
nomenclature of porphyrins can be found in Chapter 2.
49
J.C. Bommer and P. Hambright
ed ferric iron reduced with ascorbate, and in the delivery of oligonucleotides into the
the Fe(II) spectrophotometrically analyzed nucleus (43). Four coordinate Cu(II),
with o-phenanthroline (32). With metallo- Ni(II), and Au(III) TMPyP(4) complexes
TPPS4 complexes, approximately 10-mg intercalate into certain DNAs at low ionic
samples were digested with a 3:1:1 mixture strengths, and show external groove bind-
of H2SO4-HNO3-HClO4, and the metal ing at higher salt levels (89). The N-methyl
then determined by titration with EDTA groups in the ortho position of metallo
(53). Brisbin and coworkers (13) spectro- TMPyP(2) complexes provide a steric bar-
photometrically determined the concentra- rier to intercalation, and the bulky tetraca-
tion of protoporphyrin IX DME to ± 2%. tionic tetrakis(4-N,N,N-trimethylanilini-
To a constant amount of porphyrin um)porphyrin, H2-TAPP4+ also does not
(approximately 10-5 M) in acetic acid, high- intercalate (81). Fe(III) and Mn(III)-
er and lower concentrations of standard- TMPyP(X) compounds are excellent super-
ized metal acetate solutions (divalent Zn, oxide dismutase mimics (10–12), and they
Co, Ni, Fe, and Cu) were added. After also transform the cytotoxic peroxynitrite
heating to completely form the 1:1 com- (ONOO-) into NO2 and the O = Mn(IV)
plex, appropriate plots of absorbance ver- porphyrin (36,73). Mn(III)-tetra(N-ethyl-
sus metal ion concentration indicate the 2-pyridyl)porphyrin is superior to oxyhe-
point at which the concentration of the moglobin for the determination of NO
metal is equal to the concentration of the concentrations. Water-soluble Fe(III),
porphyrin. This method can also be Mn(III), and Gd(III) porphyrins have been
applied to determine the concentration of explored as liver and tumor imaging agents
the metals (2–5 × 10-5 M) by titration with using magnetic resonance imaging (MRI)
a known concentration of porphyrin. Petro techniques (79), and radioactive sulfonated
and Marzilli (93) determined the molar metalloporphyrins have potential as tumor
extinction coefficients of a series of cation- localizing agents in nuclear medicine. Nat-
ic porphyrins by adding excess standard- ural and synthetic water-soluble porphyrins
ized Zn(II) to aqueous solutions of the por- act against the human immunodeficiency
phyrins at pH 12.0. After the reaction was virus (30), and H2-TMPyP(2) and H2-
complete, the pH was brought to 9.0, and TMPyP(4) both inhibit human telomerase
excess zinc was determined with the colori- (51,58). Porphyrins that produce singlet
metric zinc reagent, zincon. The precision dioxygen within tumors are valuable in
of this method was approximately 3%. photodynamic therapy (108). The struc-
tures and common abbreviations used for
certain water-soluble compounds are
5. WATER-SOLUBLE PORPHYRINS shown in Figure 1.
a 1:1 ratio (each approximately 10-2 M) are reaction is complete. To ascertain the
mixed in CH2Cl2 or CHCl3 containing degree of N-alkylation, a sample from the
0.75% EtOH and approximately 10-3 M pot is spotted on a silica gel TLC plate,
BF3-OEt2 (as the acid catalyst) and stirred and the plate is developed with a 1:1:8
for several hours at 25°C. The cyclic por- (vol/vol/vol) mixture of saturated aqueous
phyrinogen formed is then oxidized in the KNO3-water-acetonitrile (10). During
same pot to the porphyrin with DDQ or the course of the reaction, six bands are
p-chloranil at reflux, and the impurities are observed, with the slowest moving and
removed by chromatography. last remaining the tetra (N-alkylated)-por-
phyrin. Other workers use 3:3:1:2:1 iso-
propanol-H2O-acetone-acetic acid-con-
7. N-ALKYLATIONS TO PREPARE centrated NH3 for the separation of
CATIONIC PORPHYRINS differently charged cationic porphyrins
and metalloporphyrins. The sterically hin-
It should be noted that all alkylating dered H2-TPyP(2) was also tetra-N-
agents are hazardous, and extreme caution methylated in neat dimethyl sulfate at
should be taken when working with these 110°C. The same N-methylation tech-
substances. Many workers methylate the niques in DMF are used to prepare the
H2-TPyP(X) compounds in hot or reflux- tetrakis[N-methyl-4 (or 3) quinolyl]por-
ing chloroform in the presence of excess phyrins (1), and the popular tetra
CH3I, and the solid iodide salts of H2- (4-N,N,N-trimethylanilinium)porphyrin,
TMPyP(X) precipitate from solution. Since H2-TAPP from tetra(4-N,N-dimethyl-
the iodides are not very soluble in water, the anilinium)porphyrin (65). Evaporating
product is stirred with the chloride form of the water from an aqueous solution of
an ion-exchange resin either in water, or in M-TAPP in the oven leads to loss of the
water–methanol, and warmed until the N-methyl groups.
solid dissolves. After filtration, the solution Several examples of water-soluble “pick-
is slowly passed through a long column of et-fence” porphyrins have been prepared.
chloride resin, and the water is removed by The starting material, tetra(2-nitro-
lyophilization (90). In some cases, both the phenyl)porphyrin, is synthesized by the
tri- and tetra-N-methylated iodides precipi- Adler-Longo technique in acetic acid (26).
tate from chloroform, as indicated by elec- This compound is dissolved in concentrat-
trophoresis studies on the products (16), ed HCl and reduced to the tetra(2-
and thus full tetra-N-methylation is not aminophenyl)porphyrin with SnCl2 at
always achieved in chloroform with CH3I. 70°C. The H2-T(2-NH2P)P is a mixture
The N-methylation is perhaps best done in of four atropisomers, with the amino
DMF with methyl-p-toluenesulfonate groups above and below the porphyrin
(MTS). In a typical procedure, 0.5 g of por- plane. A TLC method to separate these iso-
phyrin is added to 50 mL of DMF in a mers is given in section 2.2.1. An 8- × 30-
100-mL flask (83). The solution is warmed, cm column filled with a silica gel-chloro-
and before boiling, 2 mL of MTS is added. form slurry was used on the preparative
The solution is refluxed for 4 hours, and the scale. The column was loaded with a chlo-
tosylate salt of the porphyrin is removed roform solution of the atropisomers, and
from the cooled solution by filtration and the three undesired and less polar com-
ion-exchanged into the chloride form. In pounds removed with 1:1 benzene:diethyl
some instances, the N-alkylated porphyrins ether. The target and most polar cis-
decompose if not isolated soon after the α,α,α,α isomer was then eluted with 1:1
52
General Laboratory Methods for Tetrapyrroles
acetone:diethyl ether. The other three isomers per with concentrated H2SO4. While most
were re-equilibrated at 100°C in CHCl3- water-soluble manganese porphyrins are pro-
toluene, forming more of the desired duced in the 3+ oxidation state, the Mn(II)-
α,α,α,α species. More efficiently, the isomer β-Br8-TMPyP(4) is the stable form of this
mixture is refluxed overnight in benzene in electron deficient porphyrin having a
the presence of silica gel. As it forms, the deformed nuclear structure (11). One to
α,α,α,α is preferentially adsorbed on the four chlorine atoms can be added to the
solid and can be removed with 1:1 acetone: β-pyrroles of H2-TPyP(2) by refluxing the
ether (76). The reaction of nicotinic anhy- compound in CHCl3 with N-chlorosuccin-
dride at room temperature in CH2Cl2-pyri- imide (60). The products are separated by
dine with the amino compound forms the chromatography, and the H2-β-ClxTEt-
α,α,α,α-tetrakis(o-nicotinamidophenyl)por- PyP(4) are then formed in DMF by the
phyrin (85). This species can be gently N- addition of ethyl-p-toluenesulfonate. The
methylated in dry trimethyl phosphate by the sterically hindered 2,6-dichloro-TMPyP(4)
addition of methyl trifluoromethylsulfonate, has been prepared (57), as well as an octaca-
with added 2,6-lutidine to scavenge protons. tionic derivative (54). A series of compound
The ortho-isonicotinamido compound has containing (N-methyl-4-pyridyl) groups on
also been prepared (50,113). The four atro- the β-pyrrole positions have also been synthe-
pisomers of the water-soluble Cu(II)- sized (34). Tetraphenyl type porphyrins with
TMPyP(2) could be separated on silica gel -CH2X substituents in the para positions,
TLC plates developed with 2-butanone-con- with X = N+Et3, N+Ph3, NH2, and PO32- are
centrated NH3-NH4PF6-n-butylamine. The known, and porphyrins have been made
Zn(II) and Ni(II) isomers, but not those of water-soluble by the addition of sugar
the metal-free H2-TMPyP(2) or its Mn(III) residues (47). Other compounds contain
complex, could also be separated under such three (N-methyl-4-pyridyl) groups for water
conditions (64). solubility, and the fourth phenyl or pyridyl is
Refluxing the cobalt(II) complex of the derivatized with substituents that can interact
meso-tetrakis(pentafluorophenyl) porphy- with nucleic acids (75). Four moles of ethyl-
rin overnight in DMF (61) leads to the enediamine (and related diamines) have been
production of meso-tetrakis-[2,3,5,6,tetra- added to protoporphyrin IX DME to form
fluoro-4-(dimethy lamino)phenyl]porphy- acid-soluble compounds (117), and similar
rinato cobalt(II). This complex can be con- species containing two moles of ethylenedi-
verted into the water-soluble triflate salt amine can be prepared from meso or
(68,69) using methyl trifluoromethanesul- deuteroporphyrin IX DME. These protopor-
fonate in trimethyl phosphate overnight at phyrin derivatives are soluble over a wider pH
60°C under N2. The metallo triflate salts range if the -NH-(CH2)2-N+Me3 forms are
are stable at room temperature, while the prepared, using techniques similar to those
solid chlorides decompose within days. described above. Then an octacationic
The electron withdrawing tetrafluo- tetrakis[2,4,6-trimethyl-3,5-bis(-CH2N+Me3)
rophenyl groups reduce the electron densi- phenyl]porphyrin is known (5).
ty at the central nitrogen atoms, and a larg-
er effect can be achieved by halogenations
at the β-pyrrole positions (31). Thus, 8. NEGATIVELY CHARGED
Cu(II)-TMPyP(4) dissolved in DMF can PORPHYRINS
be β-octabromonated (96) by addition of 8.1. Synthetic Derivatives
Br2(l), and the metal-free H2-β-Br8TM-
PyP(4) is prepared by removal of the cop- The synthetic tetranegatively charged
53
J.C. Bommer and P. Hambright
acid forms can be evaluated on C-18 plates tion, then twice more with the same vol-
with 70% to 80% methanol–sodium phos- ume of deionized water. The volume is
phate buffer system for porphyrins with reduced on a rotary evaporator, and the
four or fewer carboxy groups and 50% to porphyrin mixture is applied to a silica or
60% methanol/1 or 2 M ammonium alumina column to effect separation and
acetate for porphyrins with four or more purification of the components.
carboxyl groups. If porphyrin esters are not desired, the
porphyrins may be collected from the
8.3. Isolation of Natural Porphyrins decanted and filtered growth medium
from Bacterial Cultures directly onto the bulk C-18 silica reverse
phase packing such as that available in 55
Many bacteria, especially photosynthet- to 105 µm size from Waters or Millipore
ic bacteria, produce substantial amounts of (Bedford, MA, USA) activating first with
porphyrins, porphyrinogens, and bacteri- methanol then washing with water. The
ochlorophyll, or can be made to do so packing is then washed with water, and the
under certain conditions of stress. In gen- porphyrins eluted with 90% methanol and
eral, the porphyrins or porphyrinogens are water vol/vol. The solvent is removed by
mostly excreted into the growth media and rotary evaporation, and the porphyrins
can be treated separately from the bacteri- taken up in water, filtered, and either col-
ochlorophyll in the case of photosynthetic lected by flocculation at pH 4.0, or applied
bacteria that remain within the cellular to a reverse phase column for further
structure of the bacterium. purification. These procedures are applica-
The cells are separated from the medium ble to most tetrapyrroles found in any
by centrifugation at a minimum 2000× g. aqueous-based solution whether of mam-
The medium is decanted and stirred or malian origin, such as urine and extracted
shaken while adding 5 mL of 5% iodine in feces, or aqueous extracts of plant material.
ethanol per liter of the medium. The solu- One must be careful of course of treating
tion is allowed to stand for 1 hour protect- some tetrapyrroles of biological origin with
ed from light to complete oxidation of any strong acids such as in the esterification
porphyrinogens to the corresponding por- steps described. Such porphyrin may
phyrins. If porphyrin esters are desired, the require slightly different handling tech-
medium is passed through a layer of diethy- niques.
laminoethyl (DEAE) cellulose (about 100
mL of aqueous gravity packed adsorbent
per liter of medium) on a Buchner funnel, 9. PORPHYRINS AND METALLO-
which binds the anionic porphyrin tightly. PORPHYRINS IN SOLUTION
The packing is washed with water, dried in
the oven, or preferably air-dried, or dried 9.1. Behavior in Solution at the
under vacuum. The porphyrins are eluted Molecular Level
from the cellulose with 5% wt/vol sulfuric
acid in methanol or methanol saturated Under certain conditions, porphyrins
with HCl until color ceases and allowed to and metalloporphyrins undergo intermole-
stand protected from light for 24 hours at cular association in solution. In water at
room temperature. The esterifying mixture pH 7.5 in 0.01 M Tris buffer, plots of
is diluted with an equal volume of absorbance versus concentration for H2-
dichloromethane, and washed first with an TPPS3 follow Beers law from approximate-
equal volume of 1 M sodium acetate solu- ly 5 × 10-7 M to 1.4 × 10-5 M, and the
56
General Laboratory Methods for Tetrapyrroles
compound is considered monomeric (90). acid converts the absorbed free base into
In the presence of 0.1 M KNO3 at this the more weakly adsorbed green diacid H4-
pH, however, increasingly negative devia- TMPyP6+. For low porphyrin concentra-
tions from Beers law are observed as the tion work, many workers prefer to use a
concentration of the porphyrin increases, new plastic cuvette for each measurement
consistent with a monomer–dimer equilib- on freshly prepared solutions. The ampho-
rium, where the absorbance of the dimer is teric compound tetrakis[N-(2-sulfoethyl)-
less than that of the monomer. Equations 4-pyridyl]porphyrin is monomeric in the
have been developed to determine the pH range 2.0 to 12.0, and does not adsorb
dimerization constant, KD, from such on glass surfaces (55).
absorbance–concentration data. The position of substituents on the por-
phyrin periphery influence the extent of
2H2-TPPS3 [H2-TPPS3]2 KD [Eq. 1] aggregation. While H2-TPPS4, sulfonated
Since dimerization increases with ionic para-substituted TPP species and H2-TPPC
strength, overlay spectra of a given concen- associate, the ortho-and di-ortho-substituted
tration of porphyrin measured at differing sulfonated TPP derivatives, as well as tetra(2-
salt concentrations also provides evidence carboxyphenyl)porphyrin are monomeric
for the extent of porphyrin aggregation. (110,111). The electron-rich natural por-
Another method is to obtain the spectra of phyrins such as meso and protoporphyrin
the porphyrin at a given salt concentration IX examined by fluorescence techniques
in a 0.10-cm cell. The solution is diluted have high KD values of 2.7 × 106 M-1 and
1/10 and the spectra then taken in a 1.0 1.9 × 107 M-1, respectively (80), while the
cm cell, followed by another 1/10 dilution, octanegative uroporphyrin I shows no evi-
run in a 10.0-cm cell. If the compound is dence of dimerization at moderate ionic
monomeric, the overlay spectra should be strengths above pH 7.0. Dimerization also
superimposable. If dimers form, the most depends on the nature of the coordinated
dilute solution produces the highest ab- metal ion. The five or six coordinate
sorbance. Isosbestic points are often noted. Zn(II), VO(IV), Cr(III), Mn(III), and
Temperature–jump relaxation methods Co(III) complexes of TPPS4 are monomer-
allow the determination of the rate con- ic under conditions in which the four coor-
stants for dimer formation (kf) and dissoci- dinate Cu(II), Pd(II), and Ag(II)-TPPS4
ation (kd), and for H2-TPPS3, KD = 4.8× complexes are dimers (67). In nonaqueous
104 M-1, kf = 2.2 × 108 M-1 s-1, and kd = solutions, electron spin resonance studies
4.6 × 103 s-1. Under the same conditions, on paramagnetic metalloporphyrins and
both with and without added electrolyte, concentration-dependent NMR work on
H2-TMPyP(4) follows Beers law and metal-free compounds are used to access
shows no kinetic relaxation behavior, and monomer–dimer behavior (115).
thus behaves as a monomer. The electron An example of the practical conse-
withdrawing pyridinium groups remove quences of dimerization and larger aggre-
electron density from the porphyrin ring, gate formation is the sometimes anomalous
disfavoring the van der Waals interactions behavior of porphyrins during HPLC
leading to dimerization. Many water-solu- analysis. We have noted, for instance, the
ble porphyrins, such as H2-TMPyP(4), are reverse phase HPLC of H2-TPPS4 in phos-
adsorbed strongly on glassware. A flask that phate buffer systems can show a series of
once contained this compound when peaks eluting at rather regular intervals in
washed with water appears clean, but 0.1 what has been shown to be an essentially
M HCl added to the flask turns green, as pure sample by other chromatographic
57
J.C. Bommer and P. Hambright
means. This can be explained as a separa- with KD values in the 105-106 M-1 range
tion of the monomeric, dimeric, and high- (88). Addition of salts such as NaClO4 lead
er aggregated species, which are not rapidly to dimer dissociation. Job’s law mole-
dissociated under these HPLC conditions. ratio spectrophotometric studies indicated
Changing the concentration of the injected that the double decker porphyrin CeIII-
sample or the solvent composition of the [TMPyP(4)]27+ reacted with two moles of
injection media changes the relative size of Ni-TPPS4- or VO-TPPS4-, presumably with
the eluting peaks. one porphyrin on either face of the cerium
In aqueous solution, certain porphyrin dimer (18). Also, two moles of CeIV-[TAPP]28+
systems exhibit supramolecular behavior. complexed with one mole of either Ni(II) or
The diacid H4-TPPS2- has a Soret band at VO-TPPS4-.
433 nm. The new narrow peak at approxi- Only 1:1 molecular complexes were
mately 489 nm, which appears below pH formed between indigo di-, tri-, and tetrasul-
2.0 at high ionic strengths, is attributed to fonates (18) with Cu(II) and Zn(II)-TAPP4+
the presence of J-aggregates, edge-to-edge and Zn-TMPyP(4)4+. The magnitudes of the
ribbon-like zwitterionic electronically cou- association constants for molecular complexes
pled species in which the central diacid involving H2-TPPC44-, H2-TAPP4+, and
protons of one porphyrin interact with the tetrakis(N-propyl-4-pyridyl)porphyrin with
sulfonic acid groups of another (87). At various ligands that bind in a face-to-face fash-
higher porphyrin concentrations, another ion can be predicted (102). Molecular com-
peak at 422 nm appears, due to even larg- plexes between uroporphyrin-I and a variety
er H-aggregates, which are face-to-face of large organic cations and neutral hetero-
associations of the J-species, and involve cyclic molecules such as caffeine, o-phenan-
hundreds of thousands of interacting throline, methyl viologen, nicotinamide, and
porphyrin units (95). Resonance light adenine have been investigated (82).
scattering experiments (92) indicate that Porphyrins show acid base behavior, and
the trans diphenyl/di(4-sulfonatophenyl) the first two acid dissociation constants are
porphyrin, as the free base at pH 6.0 and defined as follows, where the charges of the
as the diacid at pH 3.0 show supramolecu- peripheral groups are neglected:
lar behavior, as does the diacid H4-(β-Br8-
TPPS)42- at pH approximately 1.0 and the H4-P2+ H3-P+ + H+ K4 [Eq. 2]
free base and Cu(II) complex of trans H3-P +
H2-P + H +
K3 [Eq. 3]
diphenyl/di(N-methyl-4-pyridyl)porphyrin.
Supramolecular chiral H- and J-aggregates The equilibria are measured spectrophoto-
of H4-TPPS42- are formed on poly-L- metrically, and it is important to make sure
glutamate at pH 2.9, but only in the pres- that the porphyrin is monomeric under the
ence of the cationic Zn(II), Mn(III), and pH titration conditions, and that the
Au(III)-TMPyP(4) species. The positive buffers used do not complex with the por-
porphyrins act as spacers, allowing the anion- phyrins (56). For example, H2-
ic porphyrins to approach the polypeptide TMPyP(4)4+ shows pK4 = 0.8, pK3 = 1.4,
and gain chirality (94). while for the more basic H2-TAPP4+, pK4
Heteronuclear dimers are formed = 3.95 and pK3 = 4.11. The pK2 and pK1
between oppositely charged porphyrins and values for most porphyrins are above 12,
metalloporphyrins in solution. Thus, 1:1 although the H2-[β-Br8-TMPyP(4)]4+
complexes in acetone–water were found with eight electron-withdrawing bromines
between H2-TAPP4+ (and Zn-TAPP4+) with on the β-pyrrole positions (96) gives pK2 =
H2-TPPS44-, Cu-TPPS44-, and Zn-TPPS44- 6.5 and pK1 = 10.2. The magnitudes of
58
General Laboratory Methods for Tetrapyrroles
these acid dissociation constants depend on DME. Partial potential values could be
the ionic strength and temperature. The assigned to substituent groups, such that
pK3 values are used to rank the relative when added to the potential of the refer-
basicities of water-soluble porphyrins (47), ence compound porphin, allow the calcu-
as in the series H2-TMPyP(2) [-0.9], H2- lation of E1/2(1) for new derivatives. For a
TMPyP(4) [1.4], H2-TMPyP(3) [1.8], series of porphyrins, a variety of spectro-
H2-TAPP [4.11], H2-TPPS4 [4.76], H2- scopic, kinetic, and equilibrium data can
TPPC4 [5.5], uroporphyrin I [6.0], and be correlated with either pK3 or E1/2 (1).
hematoporphyrin IX [6.1]. The cationic
porphyrins are usually less basic than the 9.2. Practical Aspects of Porphyrin
anionic compounds. Dissolution
Detergents have been used to bring 3,8-
disubstituted deuteroporphyrin IX DME As noted in the previous section, most
compounds into aqueous solution (46). porphyrins exist in an aggregated state in
With cationic detergents such as cetyltri- aqueous solution. This can translate into
methylammonium bromide, the monoca- the more fundamental problem of how to
tion H3-P+ is destabilized, and only the successfully dissolve some of the less hydro-
diacid–free base equilibria (K3K4) can be philic porphyrins and stabilize solutions
observed. In 2.5% sodium laurel sulfate, long enough to do meaningful biological
both pK3 and pK4 can be obtained, and experiments. This problem is generally
typical pK3 values for these esters are 5.8 for associated with mono- and dicarboxylic
mesoporphyrin, 5.5 for deuteroporphyrin, porphyrins and chlorins and their divalent
4.8 for protoporphyrin, and 3.3 for the 3,8 metallo derivatives, which lack hydrophilic
diacetyldeuteroporphyrin (23). Electron substituents other than the carboxylic acid
withdrawing groups decrease the proton groups. Protoporphyrin IX is an example of
affinity of the central nitrogen atoms. this type of porphyrin in that it is colloidal
The reduction potential of the free base in aqueous alkali (21). As a general rule,
porphyrin into its radical anion, E1/2 (1), these porphyrins can be dissolved by treat-
have been measured under the same condi- ing them with dilute base (0.01 to 0.1 M
tions in DMF for over a hundred water- NaOH, KOH, or better yet, ammoniacal
insoluble porphyrins (119,120), and such bases such as Tris or NH4OH if usage will
constants are also a measure of relative allow), then diluting to about 50% with
basicities that allows comparisons between aqueous soluble organic solvents such as
meso and β-pyrrole substituted com- ethanol, DMF, dimethylsulfoxide, etc. Rea-
pounds. sonably concentrated stock solutions on the
order of several millimolar can usually be
H2-P + e- H2-P- E1/2(1) [Eq. 4]
prepared in this way. The stock solution can
The more basic the porphyrin, the less ten- then be diluted into a large volume of
dency it has to add an electron, and the buffered aqueous solution or media to
more negative its reduction potential. A adjust the pH and minimize contribution
very basic porphyrin is OEP with E1/2(1) = from the organic solvent and base. Both
-1.85 V, followed by -1.82 for meso DME, solutions are likely to be unstable with time
-1.77 for deutero DME, -1.70 for proto resulting in increasing aggregation and pre-
DME, -1.66 for the unsubstituted por- cipitation, and should be prepared freshly
phyrin porphin, -1.56 for H2-TPP, -1.48 and used immediately for each experiment.
for the 3,8-diacetyl deutero DME and, Although the porphyrins likely exist as very
-1.32 for the less basic 3,8-dicyano deutero large aggregates in the final buffered solu-
59
J.C. Bommer and P. Hambright
60
General Laboratory Methods for Tetrapyrroles
Typical values for KCd are 7.9 × 10-7 M for Ce(IV)-[TPPC4]2 was produced by basic
TMPyP(2) and 4.2 × 10-11 M for TPPS4. hydrolysis of the tetramethyl ester. As
The deformed Cd-P reacts with Cu2+ (and noted, the oxidation state of the coordinat-
Fe2+, Zn2+, Mn2+ ions) approximately 102 ed metal may or may not change during
to 103 times faster than Cu2+ incorporates the reactions. A list of the metal ions that
into H2-P itself. Such room temperature have been incorporated into water-soluble
metal-catalyzed electrophilic substitution porphyrins has been compiled (47).
reactions have been used to insert metal
ions into picket-fence type porphyrins, 10.2. Water-Insoluble Porphyrins
where refluxing the solution would lead to
atropisomerization (85,113). Mercury(II) Adler’s DMF method is often employed
in acid forms Hg2-P2+ complexes, and sim- for insertion of various metal ions into
ilar displacement reactions occur after ini- water-insoluble porphyrins (4). The free
tial loss of a mercury ion (99). Lithium base porphyrin and a metal salt (acetate,
ions are in equilibrium with Li-P- com- chloride) are refluxed in DMF until the
plexes of TMPyP(X) (56) and β-Br8- absorption spectra indicates that metala-
TMPyP(4) (96) in base. Deformed central- tion is complete. The addition of water to
ly mono-N-alkylated porphyrins react with the cooled solution precipitates the metal-
metal ions several orders of magnitude loporphyrin, which can then be purified by
faster than do the parent compounds (71). chromatography. An example of this proce-
This fact has been used for the rapid prepa- dure is given below. One or two molecules
ration of short half-life radiolabeled por- of dimethylamine are often found bound
phyrins of divalent Cu, Co, and Pd, where to trivalent complexes. Buchler has devel-
the central N-benzyl group is lost upon oped techniques of incorporation of high
metalation (72). oxidation state metal ions in which the
In cases where high temperatures in reactions are run in imidazole or phenol
nonaqueous solvents are necessary for melts, and he has reviewed other useful
metalation with water-insoluble or metalation systems (15). These include
organometallic reagents, it is often best to reactions in acetic acid–sodium acetate, in
first metalate the water-insoluble precursor, pyridine and benzonitrile for acid labile
which can usually be purified by chro- complexes, and the uses of metallo acety-
matography. The water-soluble metallo- lacetonates, phenoxides, and organometal-
porphyrin is then formed in a subsequent lic reagents as metal carriers. Buchler’s “sta-
step. For example, H2-TPyP(4) in bility index” Si (the product of the Pauling
trichlorobenzene was reacted with n-BuLi electronegativity and cation charge divided
at room temperature to form Li-TPyP(4)-, by the ionic radius in picometers) is a guide
and after addition of Ce(acac)3.H2O, the to the tendency of a metalloporphyrin to
solution was refluxed until metalation was be demetalated by acids of various concen-
complete (17). The Ce(IV)-[TPyP(4)]2 trations (14), and relationships between
sandwich complex was purified by chro- the acid-catalyzed demetalation rate con-
matography on alumina, and after reaction stants for a series of M-TAPP complexes
in DMF with MTS (100°C for 5 days), the and Si have been explored (2). The loss of
water-soluble Ce(III)-[TMPyP(4)]2 was the metal ion by acid solvolysis reactions is
formed. The Ce(IV)-[TAPP]2 was made usually first-order in metalloporphyrin and
from the cerium(IV)-N.N-dimethylanilin- second-order in (H+).
ium precursor by N-methylation in The incorporation of many metals
CHCl3/EtOH with CH3I, and the requires high temperatures, which can be
61
J.C. Bommer and P. Hambright
problematic for most anionic porphyrins sis provided the ester has some solubility in
derived from natural sources. These por- this mixture (44).
phyrins and chlorins often have peripher- Porphyrins having acetic acid side chains
al groups that are labile or reactive with are prone to decarboxylate or undergo
the solvents at high temperature. In the other types of degradation if attempts are
case of vinyl or other unsaturated groups, made to metalate even the ester forms at
this can be as low as 80°C depending on high temperature. Thus, porphyrins such
the solvent, but in most cases, tempera- as uroporphyrin are usually not successful-
tures in excess of 150°C tend to cause the ly metalated in refluxing solvents such as
most difficulty. Synthetic procedures phenol, benzonitrile, dichlorobenzene, and
involving protection and regeneration of imidazole.
vinyl groups on porphyrins have been Insertion of such metals as Al, the lan-
described by Smith et al. (107). Metala- thanides, Pt, Sc, VO, TiO, and Zr into
tion of hematoporphyrin even at room these porphyrins is generally not successful.
temperature generally results in some Cobalt incorporation into porphyrins con-
dehydration of the hydroxyethyl groups to taining free carboxyl groups, even at room
vinyl groups, and if not during the meta- temperature, usually results in predomi-
lation, then certainly during the isolation nantly insoluble black polymer-like prod-
and drying process. ucts. This can sometimes be avoided by the
In general, it is best to do metal incor- addition of large amounts of pyridine to
porations on the ester form of porphyrins the metalating solution or in some cases by
with carboxyl groups. This tends to protect starting with the porphyrin ester and
these groups from decarboxylation, anhy- hydrolyzing the purified product. Use of
dride formation, and unwanted interac- pyridine may result in a product with one
tions with solvents or metalating agents. or two pyridines coordinated to the central
Purification of metalloporphyrin esters is metal ion. These ligands can usually be
generally easier than the free acid forms removed by washing with strong acid, but
using chromatographic and crystallization often this results in the formation of insol-
techniques. The resulting products can be uble polymer-like materials, or in certain
hydrolyzed with strong base, e.g., a stirred cases, loss of the coordinated metal
mixture of 2 to 4 M NaOH or KOH (24) through acid hydrolysis reactions. The pro-
with the metalloporphyrin ester dissolved cedure below is an example of the incorpo-
in an equal volume of tetrahydrofuran. ration of iron into OEP, using the DMF
Complete hydrolysis is usually accom- method of Adler (4). With its eight ethyl
plished in 12 to 24 hours at room temper- groups on the β-pyrrole positions, OEP is
ature and can be ascertained by reverse the most widely used model compound for
phase TLC. Hydrolysis is usually marked the natural protoporphyrins, which have
by precipitation of the product as the Na+ eight β-pyrrole substituents.
or K+ salt or the observed transfer of the
compound from the tetrahydrofuran ❖ Procedure 4. Incorporation of Iron
(THF) into the aqueous part of the two into Octaethylporphyrin
phase system. Removal of the THF, which
is dissolved in the aqueous layer by rotary 1. Under a well ventilated hood and wear-
evaporation, allows collection of the free ing gloves, pour 1.2 L of DMF and
acid metalloporphyrin by flocculation at approximately 10 mL of acetic acid
pH 4.0. Methanol and 1% KOH with a into a 4-L beaker containing 9.0 g
trace of water can also be used for hydroly- (16.8 mM) of OEP and stirring bar.
62
General Laboratory Methods for Tetrapyrroles
63
J.C. Bommer and P. Hambright
5.Almarsson, O., H. Adalsteinsson, and T.C. Bruice. 19.Byrn, M.P., C.J. Curtis, I. Goldberg, Y. Hsiou, S.J.
1995. Synthesis and characterization of an octacation- Khan, P.A. Sawin, S.K. Tendick, and C.E. Strouse.
ic iron tetraphenylporphyrin, which is soluble in water 1991. Porphyrin sponges: structural systematics of the
and non-m-oxo dimer forming. J. Am. Chem. Soc. host lattice. J. Am. Chem. Soc. 113:3617-3621.
117:4524-4532. 20.Cardinal, R.A., I. Bossenmaier, Z.J. Petryka, L. John-
6.Anton, J.A. and P.A. Loach. 1975. Synthesis of some son and C.J. Watson. 1968. Isolation of porphyrins
mono, di and trisubstituted tetraarylporphyrins. J. from porphyria urine by preparative thin-layer chro-
Heterocyclic. Chem. 12:573-576. matography. J. Chromatogr. 38:100-105.
7.Artaud, I., H. Grennberg, and D. Mansuy. 1992. 21.Caughey, W. S., A. Adler, B.F. Burnham, D. Dolphin,
Selective oxidative cleavage of 1,2-dimethoxyarenes to S.F. MacDonald, D.C. Mauzerall, and Z.J. Petyka.
muconic diesters catalyzed by an iron -sulfonated- 1972. Porphyrins, p. 185-216. In Specification and
tetrakis(pentafluorophenyl)porphyrin. J. Chem. Soc., Criteria for Biochemical Compounds, 3rd ed. Nation-
Chem. Commun. 1036-1038. al Academy of Sciences, Washington, D.C.
8.Bachmann, R.C. and B.F. Burnham. 1969. Purifica- 22.Caughey, W.S., R.M. Deal, C. Weiss, and M. Gouter-
tion of porphyrin esters by gel-filtration. J. Chromatogr. man. 1965. Electronic spectra of substituted metal
41:394-399. deuteroporphyrins. J. Mol. Spectrosc. 16:451-463.
9.Barnett, G.H., M.F. Hudson, and K.M. Smith. 23.Caughey, W.S., W.Y. Fujimoto, and B. Johnson. 1966.
1975. Concerning meso-substituted tetraphenylpor- Substituted deuteroporphyrins. II. Substituent effects
phyrin purification. J. Chem. Soc. [Perkins 1] 1401- on electronic spectra, nitrogen basicities, and ligand
1403. affinities. Biochemistry 5:3830-3843.
10.Batinic-Haberle, I., L. Benov, I. Spasojevic, and I. 24.Cavaleiro, J.A.S., G.W. Kenner, and K.M. Smith.
Fridovich. 1998. The ortho effect makes 1974. Pyrroles and related compounds. Part XXXII.
manganese(III) meso-tetrakis(N-methylpyridinium-2- Biosynthesis of protoporphyrin IX from copropor-
yl)porphyrin a powerful and potentially useful super- phyrininogen III. J. Chem. Soc. [Perkins I]. 1188-
oxide dismutase mimic. J. Biol. Chem. 273:24521- 1194.
24528. 25.Chu, T.C. and E.J.-H. Chu. 1966. Thin-layer chro-
11.Batinic-Haberle, I., S.I. Liochev, I. Spasojevic, and matography of methyl esters of porphyrins, chlorins
I. Fridovich. 1997. A potent superoxide dismutase and related compounds with Eastman “chromagram”.
mimic: manganese-β-octabromo-meso-tetrakis-(N- J. Chromatogr. 21:46-51.
methylpyridinium-4-yl)porphyrin. Arch. Biochem. 26.Collman, J.P., R.R. Gagne, C.A. Reed, T.P. Halbert,
Biophys. 343:225-233. G. Lang, and W.T. Robinson. 1975. “Picket fence por-
12.Batinic-Haberle, I., I. Spasojevic, P. Hambright, L. phyrins.” Synthetic models for oxygen binding hemo-
Benov, A.L. Crumbliss, and I. Fridovich. 1999. The proteins. J. Am. Chem. Soc. 97:1427-1439.
relationship between redox potentials, proton dissocia- 27.Conant, J.B. and W.W. Moyer. 1930. Studies in the
tion constants of pyrroic nitrogens and in vivo super- chlorophyll series. III. Products of the phase test. J.
oxide dismutase activities of manganese(III) and Am. Chem. Soc. 52:3013-3023.
iron(III) cationic and anionic porphyrins. Inorg. 28.Datta-Gupta, N. and T.J. Bardos. 1966. Synthetic
Chem. 38:4011-4022. porphyrins. I. Synthesis and spectra of some para-sub-
13.Brisbin, D.A. and J.O. Asgill. 1974. The micro-deter- stituted meso-tetraphenylporphines. J. Heterocyclic.
mination of porphyrins. An intergrated laboratory Chem. 3:495-502.
experiment. J. Chem. Ed. 51:211-213. 29.DiNello, R.K. and C.K. Chang. 1978. Isolation and
14.Buchler, J.W. 1975. Static coordination chemistry of modification of natural porphyrins, p. 289-339. In D.
metalloporphyrins, p. 157-231. In K.M. Smith (Ed.), Dolphin (Ed.), The Porphyrins, Vol. I, Part A. Acade-
Porphyrins and Metalloporphyrins, Elsevier, Amster- mic Press, New York.
dam. 30.Dixon, D.W., R. Schinazi, and L.G. Marzilli. 1990.
15.Buchler, J.W. 1979. Synthesis and properties of metal- Porphyrins as agents against the human immunodefi-
loporphyrins, p. 389-483. In D. Dolphin (Ed.), The ciency virus. Ann. NY Acad. Sci. 616:511-513.
Porphyrins, Vol. I, Part A. Academic Press, New York. 30a. Dolphin, D., (Ed.). 1978. The Porphyrins, Vol. I-VII.
16.Buchler, J.W., F.M. Kunzel, U. Mayer, and M. Nawra. Academic Press, New York.
1994. Metal complexes with tetrapyrrole ligands. Part 31.Dolphin, D., T.G. Traylor, and L.Y. Xie. 1997. Poly-
64. Electrophoresis of water-soluble porphyrins and haloporphyrins: unusual ligands for metals and metal
their metal complexes. Fresenius J. Anal. Chem. catalyzed oxidations. Acc. Chem. Res. 30:251-259.
348:371-376. 32.Drabkin, D.L. 1941. Spectrophotometric studies VIII.
17.Buchler, J.W. and M. Nawra. 1994. Metal complex- The microdetermination of iron in cytochrome c and
es with tetrapyrrole ligands. 67. Synthesis and hemin preparations. J. Biol. Chem. 140:387-396.
spectroscopic properties of water-soluble cerium 33.Elder, G. 1971. Separation of porphyrin methyl esters
bisporphyrinate double-decker ions. Inorg. Chem. by two-dimensional thin layer chromatography. J.
33:2830-2837. Chromatogr. 59:234-236.
18.Buchler, R.A. and M. Nawra. 1996. Metal complexes 34.Endisch, C., J.-H. Fuhrhop, J. Buschmann, P. Luger,
with tetrapyrrole ligands. 71. Heteroaggregation of and U. Siggel. 1996. β-Tetraethyl-β′-tetrapyridin-4-yl
cationic metal mono- and bisporphyrinates with anion porphyrins, their N-methylated tetracations, and het-
metal porphyrinates or indigosulfonates. Inorg. Chim. erodimers with ms-tetraphenylsulfonato porphyrins. J.
Acta. 251:227-237. Am. Chem. Soc. 118:6671-6680.
64
General Laboratory Methods for Tetrapyrroles
35.Falk, J.E. 1964. Porphyrins and Metalloporphyrins. 52.Haye, S.E. and P. Hambright. 1984. Equilibrium con-
Elsevier, Amsterdam. stants and electrophylic exchange reactions of lead(II)
36.Ferrer-Sueta, G., I. Batinic-Haberle, I. Spasojevic, I. porphyrins. Inorg. Chem. 23:4777-4779.
Fridovich, and R. Radi. 1999. Catalytic scavengening 53.Herrmann, O., S.H. Mehdi, and A. Corsini. 1978.
of peroxynitrite by isomeric Mn(III) N-methylpyridyl- Heterogeneous metal-insertion: a novel reaction with
porphyrins in the presence of reductants. Chem. Res. porphyrins. Can. J. Chem. 56:1084-1089.
Toxicol. 12:442-449. 54.Hoffmann, P., G. Labat, A. Robert, and B. Meunier.
37.Fischer, H. and R. Baumler. 1929. Uber phao und 1990. Highly selective bromination of tetramesitylpor-
phyllerythro-porphyrine. Ann. Chem. 474:65-120. phyrin: an easy access to robust metallo-porphyrins,
38.Fischer, H. and H. Orth. 1934. Die Chemie des M-Br8-TMP and M-Br8-TMPS. Tetrahedron Lett.
Pyrroles. Vol 1. Akademische Verlagsgellschaft, Liepzig. 31:1991-1994.
39.Fischer, H. and H. Orth. 1937. Die Chemie des 55.Igarashi, S. and T. Yotsuyanagi. 1984. New “ampho-
Pyrroles. Vol 2, part 1. Akademische Verlagsgellschaft, teric ion type” water-soluble porphines as a spec-
Liepzig. trophotometric agent. Chem. Lett. 1871-1874.
40.Fischer, H. and H. Siebel. 1932. Uber phaeophorbide 56.Islam, M.Q. and P. Hambright. 1998. Lithium com-
a, chorin e and chlorophyll a. Ann. Chem. 499:84- plexes and the kinetics of interactions of zinc ions with
108. tetra(N-methyl-4-pyridyl)porphyrins in basic solution.
41.Fischer, H. and A. Stern. 1940. Die Chemie des Trans. Met. Chem. 23:727-733.
Pyrroles. Vol 2, part 2. Akademische Verlagsgellschaft, 57.Ito, Y., K. Kunimoto, S. Miyachi, and T. Kato. 1991.
Liepzig. Photocatalytic cleavage of 1,2-diols by a cofacial hin-
42.Fleischer, E.B., R. Foust, D. Jeter, and R. Near. 1973. dered water-soluble iron(III) porphyrin. Tetrahedron
The solubilities of tetra(monofluorophenyl)porphines. Lett. 32:4007-4010.
Inorg. Nucl. Chem. Lett. 9:1219-1220. 58.Izbicka, E., R.T. Wheelhouse, E. Raymond, K.K.
43.Flynn, S.M., S.T. George, L. White, W. Devonish, Davidson, R.A. Lawrence, D. Sun, B.E. Windle, L.H.
and G.B. Takle. 1999. Water-soluble, meso-substitut- Hurley, and D.D. Von Hoff. 1999. Effect of cationic
ed cationic porphyrins—a family of compounds for porphyrins as G-quadruplex interactive agents in
cellular delivery of oligonucleotides. BioTechniques human tumor cells. Cancer Res. 59:639-644.
26:736-740. 59.Jimenez, H.R., M. Julve, J.M. Moratal and J. Fuas.
44.Fuhrhop, J.-H. and K.M. Smith. 1975. In K.M. 1987. Stability constants of metalloporphyrins. a study
Smith (Ed.), Porphyrins and Metalloporphyrins, of the protonation, deprotonation and formation of
Chapter 19. Elsevier, Amsterdam. copper(II) and zinc(II) complexes of meso-tetra(p-
45.Gonsalves, A.M.d’A.R., R.W.A. Johnstone, M.M. sufonatophenyl)porphyrin. J. Chem. Soc. Chem.
Pereira, A.M.P. de SantAna, A.C. Serra, A.J.F.N. Sor- Commun. 910-911.
bal, and P.A. Stocks. 1996. New procedures for the 60.Kachadourian, R., I. Batinic-Haberle, and I.
synthesis and analysis of 5,10,15,20-tetrakis(sul- Fridovich. 1999. Synthesis and superoxide dismuting
fophenyl)-porphyrins and derivatives through chloro- activities of partially (1-4) β-chloronated derivatives of
sulfonation. Heterocycles 43:829-838. manganese(III) meso-tetrakis(N-ethylpyridinium-3-
46.Hambright, P. 1975. Dynamic coordination chemistry yl)porphyrin. Inorg. Chem. 38:391-396.
of metalloporphyrins, p. 233-278. In K.M. Smith 61.Kadish, K.M., C. Araullo-McAdams, B.C. Han, and
(Ed.), Porphyrins and Metalloporphyrins, Elsevier, M.M. Franzen. 1990. Synthesis and spectroscopic
Amsterdam. characterization of (T(p-Me2N)F4PP)H2 and (T(p-
47.Hambright, P. 2000. The chemistry of water-soluble Me2N)F4PP)M where T(p-Me2N)F4PP is the dianion
porphyrins, Ch. 18. In K. Kadish, K.M. Smith and R. of meso-tetrakis(o,o,m,m,-tetrafluoro-p-(dimethy-
Guilard (Eds.), The Porphyrin Handbook. Academic lamino)phenyl)porphyrin and M = Co(II), Cu(II) or
Press, New York. Ni(II). Structures of (meso-tetrakis(pentafluo-
48.Hambright, P., A. Adeyemo, A. Shamim, and S. rophenyl)porphyrinato)cobalt(II), (TF5PP)Co. J. Am.
Lemelle. 1985. [[4,4′,4′′,4′′′-porphyrin-5,10,15,20- Chem. Soc. 112:8364-8368.
tetrayltetrakis(1-methylpyridiniumato](2-)]-indium(III) 61a. Kadish, K., K.M. Smith, and R. Guilard (Eds.).
pentaperchlorate. Inorg. Synth. 23:55-59. 2000. The Porphyrin Handbook. Academic Press,
49.Hambright, P. and E.B. Fleischer. 1970. The acid-base New York.
equilibria, kinetics of copper ion incorporation and 62.Kalyanasundaram, K. 1984. Photochemistry of water-
acid-catalyzed zinc ion displacement from the water- soluble porphyrins: comparative study of isomeric
soluble α,β,γ,δ-tetra(4-N-methylpyridyl)porphyrin. tetrapyridyl and tetraksi(N-methylpyridiniumyl)-por-
Inorg. Chem. 9:1757-1761. phyrins. Inorg. Chem. 23:2453-2459.
50.Hambright, P., A. Turner, J.S. Cohen, R.C. Lyon, A. 63.Kalyansundaram, K. 1991. Photochemistry of Poly-
Katz, P. Neta, and A. Adeyemo. 1987. An iron(III) pyridine and Porphyrin Complexes. Academic Press,
porphyrin that exhibits minimal dimerization in aque- New York.
ous solution. Inorg. Chim. Acta 128:L11-L14. 64.Kaufmann, T., B. Shamsai, R.S. Lu, R. Bau, and
51.Han, F.X., R.T. Wheelhouse, and L.H. Hurley. 1999. G.M. Miskelly. 1995. Separation of the rotational iso-
Interactions of TMPyP4 and TMPyP2 with quadru- mers of tetrakis(N-methyl-2-pyridiniumyl)-porphyrin
plex DNA. Structural basis for differential effects on and the crystal structure of α,α,α,β-(tetrakis(N-
telomerase inhibition. J. Am. Chem. Soc. 121:3561- methyl-2-pyridiniumyl)porphyrin copper hexacyano-
3570. ferrate. Inorg. Chem. 34:5073-5079.
65
J.C. Bommer and P. Hambright
65.Krishnamurthy, M. 1977. Synthesis and characteriza- 83.Meng, G.G., B.R. James, K.A. Skov, and M. Korbelik.
tion of a new water-soluble porphyrin. Indian J. 1994. Porphyrin chemistry pertaining to the design of
Chem. 15B:964-966. anti-cancer drugs; part 2, the synthesis and in vitro test
66.Krishnamurthy, M. 1977. Kinetics of anation reac- of water-soluble porphyrins containing, in the meso
tions of a water-soluble rhodium(III) porphyrin. Inorg. positions, the functional groups: 4-methylpyridinium,
Chim. Acta. 25:215-218. or 4-sulfonatophenyl, in combination with phenyl, 4-
67.Krishnamurthy, M., J.R. Sutter, and P. Hambright. pyridyl, 4-nitrophenyl, or 4-aminophenyl. Can. J.
1975. Monomer-dimer equilibration of water-soluble Chem. 72:2447-2457.
porphyrins as a function of the coordinated metal ion. 84.Milgrom, L.R. 1997. The Colours of Life: An Intro-
J. Chem. Soc. Chem. Commun. 13-14. duction to the Chemistry of Porphyrins and Related
68.La, T., R.A. Richards, R.S. Lu, R. Bau, and G.M. Compounds. Oxford Press, New York.
Miskelly. 1995. Solution chemistry and crystal struc- 85.Miskelly, G.M., W.S. Webley, C.R. Clark, and D.A.
ture of nickel tetrakis(2,3,5,6-tetrafluofo-N,N,N- Buckingham. 1988. Acidity and dimerization of three
trimethyl-4-aniliniumyl)porphyrin trifluoromethane- water-soluble iron(III) porphyrin cations: (meso-
sulfonate (NiTF4TMAP(CF3SO3)4). Inorg. Chem. α,β,α,β-tetrakis(o-(Nmethylnicotinamido) phenyl)
34:5632-5640. porphyrinato)iron(III), (meso α,α,α,α-tetrakis(o-N-
69.La, T., R.A.Richards, and G.M. Miskelly. 1994. Syn- methylnicotinamido) phenyl)porphyrinato)iron(III),
thesis and characterization of the cationic porphyrin and (meso-tetrakis(1-methylpyridinium-4-yl)porphyri-
meso-tetrakis(2,3,5,6-tetrafluoro-N,N,N-trimethyl-4- nato)iron(III). Inorg. Chem. 27:3773-3781.
aniliniumyl)porphyrin. Inorg. Chem. 33:3159-3163. 86.Neilands, J.B. and J.A. Garibaldi. 1960. Copropor-
70.Lascelles, J. 1964. Tetrapyrrole biosynthesis and its reg- phyrin III tetramethyl ester. Biochem. Prep. 7:36-38.
ulation. Benjamin, New York. 87.Ohno, O., Y. Kaizu, and H. Kobayashi. 1993. J-aggre-
71.Lavallee, D.K. 1987. The Chemistry and Biochemistry of gate formation of a water-soluble porphyrin in acidic
N-Substituted Porphyrins. VCH Publishers, New York. aqueous media. J. Chem. Phys. 99:4128-4439.
72.Lavallee, D.K., A. White, A. Diaz, J.-P. Battioni, and 88.Ojadi, E., R. Selzer, and H. Linschitz. 1985. Properties
D. Mansuy. 1986. Efficient metalloporphyrin synthe- of porphyrin dimers, formed by pairing cationic and
sis under mild conditions using N-benzylporphyrins. anionic porphyrins. J. Am. Chem. Soc. 107:7783-7784.
Tetrahedron Lett. 27:3521-3524. 89.Pasternack, R.F. and E.J. Gibbs. 1996. Porphyrin and
73.Lee, J., J.A. Hunt, and J.A. Groves. 1998. Manganese metalloporphyrin interactions with nucleic acids. In A.
porphyrins as redox-coupled peroxynitrite reductases. Sigel and H. Sigel (Eds.), Metal Ions In Biological Sys-
J. Am. Chem. Soc. 120:6053-6061. tems 33. Marcel Decker, New York.
74.Lemberg, R. and J.W. Legge. 1949. Haematin Com- 90.Pasternack, R.F., P.R. Huber, P. Boyd, G. Engasser, L.
pounds and Bile Pigments. Interscience, New York. Francesconi, E. Gibbs, P. Fasella, G. Cerio Venturo,
75.Li, H. and L. Czuchajowski. 1994. Ribofuransoides and L. deC. Hinds. 1972. On the aggregation of
N-substituted with meso-porphyrin as nucleoside-like meso-substituted water-soluble porphyrins. J. Am.
compounds. Tetrahedron Lett. 35:1629-1630. Chem. Soc. 94:4511-4517.
75a.Lim, C.K. (Ed.). 2001. Porphyrins. In Methods of 91.Pasternack, R.F., H. Lee, P. Malek, and C. Spenser.
Chromatography, Vol. 2. World Scientific Publishers, 1977. Solution properties of tetrakis-(4-N-methyl)
New York. pyridylporphineiron(III). J. Inorg. Nucl. Chem. 39:
76.Lindsey, J. 1980. Increased yield of a desired isomer by 1865-1870.
equilibria displacement of binding to silica gel, applied 92.Pasternack, R.A., K.F. Schaefer, and P. Hambright.
to meso-tetrakis(o-aminophenyl)-porphyrin. J. Org. 1994. Resonance light scattering studies of porphyrin
Chem. 45:5215. diacid aggregates. Inorg. Chem. 33:2062-2065.
77.Lindsey, J.S. and R.W. Wagner. 1989. Investigation of 93.Petho, G. and L.G. Marzilli. 1994. A new spectropho-
the synthesis of ortho-substituted tetraphenylpor- tometric method for determination of cationic por-
phyrins. J. Org. Chem. 54:828-836. phyrinis. Microchem. J. 50:178-183.
78.Longo, F. R., M. G. Finarelli, and J. Kim. The synthe- 94.Purrello, R., L. Monsu’ Scolaro, E. Bellacchio, S. Gur-
sis and some physical properties of ms-tetra(pentafluo- rieri, and A. Romer. 1998. Chiral H- and J-type aggre-
rophenyl)porphin and ms-tetra(pentachloro-phenyl) gates of meso-tetrakis(4-sulfonatophenyl)porphine on
porphin. 1969. J. Heterocycl. Chem. 6:927-931. α-helical polyglutamic acid induced by cationic por-
79.Lyon, R.C., P.J. Faustino, J.S. Cohen, A. Katz, F. phyrins. Inorg. Chem.37:3647-3648.
Mornex, D. Colcher, C. Baglin, S.H. Koenig and P. 95.Ribo, J.M., J. Crusats, J.-A. Farrera, and M.L. Valero.
Hambright. 1987. Tissue distribution and stability of 1994. Aggregation in water solutions of tetrasodium
metalloporphyrin MRI contrast agents. Magn. Reson. diprotonated meso-tetrakis(4-sulfonato-phenyl)por-
Med. 4:24-33. phyrin. J. Chem. Soc. Chem. Commun. 681-682.
80.Margalit, R. and M. Rotenberg. 1984. Thermody- 96.Richards, R.A., K. Hammons, M. Joe, and G.M.
namics of porphyrin dimerization in aqueous solution. Miskelly. 1996. Observation of a stable water-soluble
Biochem. J. 219:445-450 lithium porphyrin. Inorg. Chem 35:1940-1944.
81.Marzilli, L.G. 1990. Medical aspects of DNA-por- 97.Rimington, C. 1952. Haems and porphyrins in health
phyrin interactions. New J. Chem. 14:409-420. and disease. II. Acta Med. Scand. 143:177-196.
82.Mauzerall, D. 1965. Spectra of molecular complexes of 98.Rimington, C. and C.A. Miles. 1951. A study of the
porphyrins in aqueous solution. Biochemistry. 4:1801- porphyrins excreted in urine by a case of congenital
1810. porphyria. Biochem. J. 50:202-206.
66
General Laboratory Methods for Tetrapyrroles
99.Robinson, L.R. and P. Hambright. 1992. Mer- olytic reduction of porphyrins and metalloporphyrins
cury(II) reactions with water-soluble porphyrins. to chlorins or phlorins. J. Chem. Soc. Faraday Trans.
Inorg. Chem. 31:652-656. 89: 495-502.
100.Rousseau, K. and D. Dolphin. 1974. A purification 111.Turay, J., P. Hambright and N. Datta-Gupta. 1978.
of meso-tetraphenyl-porphyrin. Tetrahedron Lett. 48: Intermolecular association of natural and synthetic
4251-4254. water-soluble porphyrins, J. Inorg. Nucl. Chem.
101.Sari, M.A., J.P. Battioni, D. Dupre, D. Mansuy, 40:1687-1688.
and J.B. Le Pecq. 1990. Interaction of cationic por- 112.Turk, H. and W.T. Ford. 1991. Epoxidation of
phyrins with DNA: importance of the number and styrene with aqueous hypochlorite catalyzed by a
position of the charges and minimum structural manganese(III) porphyrin bound to collodial anion-
requirements for interaction. Biochemistry 29:4205- exchange particles. J. Org. Chem. 56:1253-1260.
4215. 113.Valiotti, A., A. Adeyemo, R.F.X. Williams, L.
102.Schneider, H.-J. and M. Wang. 1994. Ligand-por- Ricks, J. North and P. Hambright. 1981. A water-
phyrin complexes: quantitative evaluation of stacking soluble “picket fence” porphyrin and its isomers. J.
and ionic contributions. J. Org. Chem. 59:7464- Inorg. Nucl. Chem. 43:2653-2658.
7472. 114.Varadi, V., F.R. Longo, and A.D. Adler. Nonchro-
103.Schwartz, S., M.H. Berg, I. Bossenmaier, and H. matographic methods of purification of porphyrins.
Dinsmore. 1960. Determination of porphyrins in 1978, p. 581-588. In D. Dolphin (Ed.), The Por-
biological material, p. 221-293. In D. Glick (Ed.), phyrins, Vol. I, Part A. Academic Press, New York.
Methods of Biochemical Analysis, Vol. VIII. Wiley & 115.White, W.I. 1978. Aggregation of porphyrins and
Sons, New York. metalloporphyrins, p. 303-339. In D. Dolphin (Ed.),
104.Shamim, A. and P. Hambright. 1980. An equilibri- The Porphyrins, Vol. V. Chapter 7. Academic Press,
um and kinetic study of water-soluble cadmium por- New York.
phyrins. Inorg. Chem. 19:564-566. 116.White, W.I., R.C. Bachman, and B.F. Burnham.
105.Shamim, A., P. Worthington, and P. Hambright. 1978. Chromatography of porphyrins and metallo-
1981. Synthesis and characterization of phenyl/4- porphyrins, p 553-580. In D. Dolphin (Ed.), The
pyridyl meso substituted porphyrins. J. Chem. Soc. Porphyrins, Vol. I, Part A. Academic Press, New York.
Pakistan 3:1-3. 117.White, W.I. and R.A. Plane. 1974. A homologous
106.Smith, K.M. (Ed.) 1975. Porphyrins and Metallo- series of water-soluble porphyrins and metallopor-
porphyrins. Elsevier, Amsterdam. phyrins: synthesis, dimerization, protonation and self-
107.Smith, K.M., E.M. Fujinara, K.C. Langrey, D.W. complexation. Bioinorg. Chem. 4:21-35.
Parish, and H.D. Tabba. 1983. Manipulation of 118.With, T.K. 1958. Preparation of crystalline porphyrin
vinyl groups in protoporphyrin IX: introduction of esters from bovine porphyria urine. Biochem. J.
deuterium and carbon-13 labels for spectroscopic 68:717-720.
studies. J. Am. Chem. Soc. 105:6638-6646. 119.Worthington, P., P. Hambright, R.F.X. Williams,
108.Sternberg, E.D., D. Dolphin, and C. Bruckner. M.R. Feldman, K.M. Smith, and K.C. Langry.
1998. Porphyrin based photosensitizers for use in 1980. Reduction potentials of beta-substituted free
photodynamic therapy. Tetrahedron 54:4151-4202. base porphyrins. Inorg. Nucl. Chem. Lett.16:441-
109.Sugata, S., S. Yamanouchi, and Y. Matsushima. 447.
1977. Meso-tetrapyridyl-porpyrins and their metal 120.Worthington, P., P. Hambright, R.F.X. Williams, J.
complexes. Synthesis and physico-chemical proper- Reid, C, Burnham, A. Shamim, D.M. Bell, R.
ties. Chem. Pharm. Bull. 25:884-889. Kirkland, R.G. Little, N. Datta-Gupta, and U. Eis-
110.Sutter, T.P.G., R. Rahimi, P. Hambright, J. Bom- ner. 1980. Reduction potentials of seventy-five free
mer, M. Kumar, and P. Neta. 1993. Steric and base porphyrin molecules: reactivity correlations and
inductive effects on the basicity of porphyrins and on the production of potentials. J. Inorg. Biochem.
the site of protonation of porphyrin dianions: radi- 12:281-291.
67
Enzymatic Preparation of Tetrapyrrole
4 Intermediates
69
M.J. Warren and P.M. Shoolingin-Jordan
SD2 = BL21(DE3)pLysE/ E. coli hemD cloned into pET14b. Raux, Davlin, and
pET14b-HemD Inducible expression of His-tagged UROS. Warren, unpublished.
Figure 1. Biosynthesis of heme from ALA. The figure also highlights uroporphyrinogen III as the branchpoint for siroheme and
cobalamin synthesis. Abbreviations used: A, acetate side chain; p, propionate side chain.
71
M.J. Warren and P.M. Shoolingin-Jordan
Figure 2. The biosynthesis of ALA. (a) ALA can be synthesized from glutamate by the C-5 pathway or (b) from glycine and suc-
cinyl-CoA by the Shemin route. In the case of the latter, it is known that the proR-hydrogen of glycine is removed in the overall
transformation into ALA.
72
Enzymatic Preparation of Tetrapyrrole Intermediates
ALAS has been greatly enhanced by the um phosphate buffer, pH 7.2, contain-
availability of the recombinant enzyme ing 0.5 mM pyridoxal 5′ phosphate, 2
from R. spheroides arising from the cloned mM EDTA, 10% glycerol, 5 mM 2-
and overexpressed hemA gene (4). mercaptoethanol, and 100 µM phenyl-
methanesulfonyl fluoride (PMSF). All
3.1. Enzyme Purification of subsequent stages are carried out at 4°C.
R. spheroides ALAS 4. The suspension is sonicated by placing
Expressed in Escherichia coli a large sonicator probe (e.g., a SANYO
Soniprep 150 Ultrasonic Disintegrator,
ALAS can be purified from wild-type R. Integrated Services TCP, Palisades Park,
spheroides (NCIB) according to published NJ, USA) about one third of the way
methods (27). However, preparing the into the bacterial suspension and soni-
media for the growth of this organism is cating the solution at medium ampli-
tedious, and the yield of purified enzyme is tude (10–12 µm) for 4 1-minute bursts
low. To overcome these problems, we have with 2 minutes cooling in between.
produced a recombinant strain of Escher- Cooling is achieved by placing the ves-
ichia coli (JMA19) that overexpresses the R. sel containing the bacterial solution in
spheroides ALAS (HemA) (Table 1), an ice-water slurry.
derived from strain JM109 that had been
5. After sonication, the extract is cen-
transformed with the plasmid pA19. The
trifuged at 15 000× g for 20 minutes to
plasmid (pA19) was constructed from a
remove cell debris. The clarified straw-
HindIII/EcoRI fragment containing the
colored supernatant contains the active
hemA gene from R. spheroides, which had
soluble enzyme.
been modified at the 5′ end by polymerase
chain reaction (PCR) and cloned into To those unfamiliar with the procedures
pUC19 (Table 1) (4). of protein purification, they are encour-
aged to read an excellent account of the
common procedures employed in protein
❖ Procedure 1. Preparation of E. coli
isolation (26).
Lysate Containing Recombinant
R. spheroides ALAS
❖ Procedure 2. Purification of
1. Bacterial growth: From an agar plate of Homogeneous Recombinant
recombinant E. coli harboring the R. R. spheroides ALAS
spheroides hemA gene (JMA18), a single
bacterial colony is removed and used to 1. Ammonium sulfate fractionation: Frac-
inoculate a starter culture (5 mL) of tionation with solid ammonium sulfate
Luria-Bertani (LB) medium containing is the first step of the purification
50 µg/mL ampicillin. process. This procedure is sometimes
2. The culture is grown for between 5 to referred to as salting out and is depen-
10 hours at 37°C and then used to dent upon the concentration of the
inoculate a larger (1 L) culture, which protein solution and the amount of salt
is grown overnight at 37°C with rotary that is added. In the case of ALAS, the
shaking (160–180 rpm) for 18 hours. enzyme is known to precipitate from
3. Harvesting and cell lysis: The cells are solution when the solution is saturated
harvested by centrifugation at 3000× g with 60% ammonium sulfate. To the
for 20 minutes, and the cell pellet is clarified bacterial extract, solid ammo-
resuspended in 10 mL of 20 mM sodi- nium sulfate is added to a saturation of
73
M.J. Warren and P.M. Shoolingin-Jordan
Figure 3. The biosynthesis of porphobilinogen from 2 molecules of ALA. It has been established that the proR hydrogen of the
ALA molecule occupying the P (propionate) site is stereoselectively removed during the reaction.
76
Enzymatic Preparation of Tetrapyrrole Intermediates
potassium phosphate buffer, pH 7.0, and to precipitate the thiol and protein.
containing 100 µM ZnSO4, and 20 After centrifugation, an aliquot of the
mM 2-mercaptoethanol. supernatant is mixed with an equal volume
5. High resolution anion exchange chro- of modified Ehrlich’s reagent (21), and the
matography: Final purification may be absorbance is measured at 555 nm (E555 =
achieved by chromatography using a 6.02 × 104 M-1 cm-1). Most ALADs are
Mono Q 5HR FPLC column susceptible to end-product inhibition, a
(Amersham Pharmacia Biotech) equili- factor that tends to limit the yields of PBG.
brated with the same buffer. The Typically a yield of 80% is achieved.
enzyme is eluted in buffer with a linear
gradient from 0 to 1 M KCl, and active 4.2.2. Large-Scale Preparation of PBG
fractions are collected and pooled.
ALAD (500 U) is incubated in a stop-
6. Storage: The pooled active fractions are
pered conical flask at 37°C in 1.9 L of 5
concentrated to about 2 mg/mL, and
mM potassium phosphate buffer, pH 6.8,
the purified enzyme is filter-sterilized
containing 5 µM ZnSO4, 5 mM 2-mercap-
for storage at 4°C in 50 mM potassium
toethanol, and ALA. The 5-aminolevulinic
phosphate buffer, pH 6.0, containing
acid hydrochloride (1 g; Sigma) is dissolved
100 µM ZnSO4, and 20 mM 2-mer-
in about 95 mL of the same buffer, adjust-
captoethanol. Activity is maintained
ed carefully to pH 6.8 with 0.1 M NaOH,
for 2 weeks. From a 0.5-L culture,
and made up to 100 mL before adding to
between 10 to 20 mg of purified
the above flask. Incubation is carried out
ALAD is obtained.
under nitrogen, typically, for 10 hours or
until the rate of PBG production has
4.2. Enzyme Assay and Incubation ceased. The reaction is followed by remov-
Protocol ing 10 µL of the incubation mixture at
Different protocols may be employed intervals and adding to 490 µL of 10%
for the enzymatic synthesis of PBG, trichloroacetic acid containing 0.1 M
depending on whether a small- or large- HgCl2 to precipitate the thiol. After cen-
scale preparation is required. Indeed, the trifugation, 0.4 mL of the supernatant is
small-scale synthesis is identical to that mixed with an equal volume of modified
used for the assay of the enzyme. Ehrlich’s reagent, and the absorbance is
measured as above. The PBG is purified
4.2.1. Assay and Small-Scale Enzymatic from the reaction mixture by adjusting the
Synthesis of PBG pH to 7.5 and passing the incubation mix-
ture slowly through a column (2 × 12 cm)
Purified E. coli ALAD (1–10 µg) is of Dowex 1 × 8 acetate (200–400 mesh).
preincubated in a total volume of 500 µL The column is first washed with 1 L of dis-
of 50 mM potassium phosphate buffer, pH tilled water, and the PBG is eluted with 1
8.0, containing 50 µM ZnSO4, and 10 M acetic acid and collected by freeze-dry-
mM 2-mercaptoethanol. The reaction is ing the solution or rapid-flash evaporation
initiated by the addition of ALA to give a below 30°C. The PBG is recrystallized by
final concentration of 5 mM. Incubation is dissolving it in a minimum volume of 1 M
carried out at 37°C for 3 minutes, after ammonia and adding 1 M acetic acid to
which time an equal volume (500 µL) of bring the pH to the isoelectric point of 5.5.
10% trichloroacetic acid containing 0.1 M After allowing the crystallization to proceed
HgCl2 is added to terminate the reaction for 1 hour, the crystals are filtered off and
78
Enzymatic Preparation of Tetrapyrrole Intermediates
PBGD is assayed using a stopped assay. The enzyme uroporphyrinogen III syn-
To 750 µL of 0.1 M Tris-HCl buffer, pH thase (UROS) (also known as uropor-
8.0, is added 40 µL of enzyme containing phyrinogen III cosynthase) catalyzes a
between 0.1 to 1 µg of purified enzyme. remarkable reaction in which preuropor-
81
M.J. Warren and P.M. Shoolingin-Jordan
Figure 4. The synthesis of uroporphyrinogen I and III from preuroporphyrinogen. Note the action of UROS, which is able to
invert the orientation of ring d during the macrocyclic ring closure process.
82
Enzymatic Preparation of Tetrapyrrole Intermediates
porphyrinogens can be generated either final concentration of 0.4 mM, and the
non-enzymically, by the reduction of uro- cells are grown for another 2 hours.
porphyrin or, in situ, using a coupled 2. Harvesting and cell lysis: The bacteria
enzyme system. The non-enzymic reduc- are collected by centrifugation (10 000×
tion of uroporphyrin is achieved by the use g at 4°C). The bacterial pellet is resus-
of sodium amalgam. Although the reaction pended in 10 mL of binding buffer
is easy to perform, problems can arise from (5 mM imidazole, 0.5 M NaCl, 20 mM
the pH of the solution, which becomes Tris-HCl, pH 7.9). The bacterial sus-
very high by the end of the reductive pension is sonicated as described in
process. For these reasons, we have favored Procedure 1, and the solution is cen-
the generation of uroporphyrinogen by trifuged (10 000× g at 4°C) to remove
enzymatic transformation of PBG, the cellular debris.
employing the actions of the enzymes 3. His-bind column: The His-tag sequence
PBGD and UROS. of the fusion protein can bind to diva-
lent metal cations such as Co2+ and
6.1. Purification of UROS Ni2+ immobilized on to His-bind resin
(Novagen, Madison, WI, USA; howev-
UROS can be purified from a recombi- er, many suppliers make different forms
nant version of the E. coli hemD that has of metal chelate resin and readers are
been modified to incorporate a 6-histidine encouraged to browse the multitude of
(His) tag at the N terminus of the protein. catalogues available). After unbound
The E. coli hemD was amplified by PCR
proteins are washed away, the His-
with appropriately designed primers such
tagged protein is eluted with imidazole.
that the gene was cloned into the NdeI and
The resin (poured into a small column,
BamHI sites of pET14b, giving the plas-
1 × 2.5 cm) is initially prepared by rins-
mid pET14b-HemD (Table 1). When
ing with 15 mL of water, charged with
transformed into E. coli BL21(DE3)pLysE,
25 mL of a 50 mM divalent cation solu-
the strain was found to overproduce the
tion (normally Ni2+) (charge buffer),
His-tagged version of the protein, which
and equilibrated with 15 mL binding
has a molecular mass of 29 kDa. The strain
buffer. The supernatant is loaded onto
harboring the plasmid is comparatively
the charged His-bind column. The col-
unstable, and fresh transformants are
required when cultures are to be grown. umn is washed with 10 column vol-
The following protocol can be adapted for umes of binding buffer, 6 column vol-
purification of all the His-tagged enzymes umes of wash buffer (100 mM
described in this chapter. imidazole, 0.5 M NaCl, 20 mM Tris-
HCl, pH 7.9), and finally the protein is
eluted in 6 column volumes of elution
❖ Procedure 5. Purification of His- buffer (400 mM imidazole, 0.5 M
Tagged UROS from E. coli NaCl, 20 mM Tris-HCl, pH 7.9). The
1. Bacterial growth: The bacteria are grown protein eluting from the column can be
from a starter culture in 2-L baffled detected by the use of the Bio-Rad pro-
flasks containing 1 L of LB media (with tein assay and SDS-PAGE.
appropriate antibiotics) at 37°C with 4. Storage: Fractions containing the mod-
vigorous shaking until an A600 = 0.6 is ified UROS are pooled and desalted by
reached, at which point isopropyl-β-D- passing through a PD-10 column, pre-
thiogalactoside (IPTG) is added to a viously equilibrated in 50 mM Tris-
83
M.J. Warren and P.M. Shoolingin-Jordan
HCl, pH 7.8. The protein is lyo- tion mixture) the solution can be filtrated
philized and is stable in this form for in an ultrafiltration unit fitted with a PM-
up to 1 year. In comparison to some of 10 membrane. The filtrate should be kept
the other enzymes described in this under argon to help prevent any oxidation.
chapter, UROS is poorly expressed, The yield of uroporphyrinogen III from
and a yield of about 2 mg/L of culture PBG is normally in excess of 95%.
is normally achieved. The uroporphyrinogen I isomer can also
be synthesized by this method simply by
6.2. Enzymatic Preparation of omitting UROS from the incubation.
Uroporphyrinogen III
Uroporphyrinogen III can be synthe- 7. SYNTHESIS OF PRECORRIN-2
sized in vitro using PBG and purified (DIHYDROSIROHYDRO-
PBGD and UROS. The reaction can be CHLORIN), SIROHYDRO-
undertaken in a range of buffers between CHLORIN, AND SIROHEME
pH 7.5 and 9.0, although the uropor-
phyrinogen III is generally more stable at Enzymatic transformations of uropor-
the higher pH values. To prevent oxidation phyrinogen III into precorrin-2 are depen-
of the product, the buffers are normally dent upon the presence of the enzyme uro-
thoroughly degassed by freeze–thawing porphyrinogen III methyltransferase
under a vacuum of less than 1 mbar. For (Figure 5), which requires S-adenosyl-L-
efficient transformation of PBG into uro- methionine (SAM) as a methyl donor (3).
porphyrinogen III, the reaction mixture There are a number of sources of this
should contain PBGD at 10 µg/mL, enzyme including Pseudomonas denitrifi-
UROS at 2 µg/mL, and PBG at 100 µM. cans, Bacillus megaterium, and Bacillus
The reaction is effectively quantitative, stearothermophilus. The CysG enzyme from
thus producing uroporphyrinogen III at a both E. coli and Salmonella typhimurium
concentration approaching 25 µM. This can also be used, although CysG is, in fact,
can be verified by taking 50 µL of the incu- a multifunctional enzyme responsible for
bation, mixing with 950 µL of 1 N HCl, the conversion of precorrin-2 into siroheme
and leaving under a bright light for 20 (30). However, in the presence of only
minutes. After centrifugation in an Eppen- SAM and uroporphyrinogen III, the en-
dorf model microfuge at 13 000 rpm for 5 zyme will effectively transform uropor-
minutes, the absorbance of the solution at phyrinogen III into precorrin-2. The uro-
405 nm can be measured, and the concen- porphyrinogen methyltransferases are
tration of porphyrin can be determined normally homodimers with a subunit
using the extinction coefficient of 5.48 × molecular mass of about 30 kDa, while the
105 M-1 L. CysG proteins, which are also homodimers,
So long as the enzymatic incubation is have a subunit molecular mass of 50 kDa.
kept in an anaerobic environment under
reduced light, the uroporphyrinogen III is 7.1. Purification of Uroporphyrinogen
stable for several hours. The solution Methyltransferases
should appear colorless, but if it starts to
turn pink then this is diagnostic of the Although the uroporphyrinogen methyl-
solution starting to oxidize. To isolate the transferases can be purified from recombi-
uroporphyrinogen III from the incubation nant sources, the preparations are often
(i.e., to remove the enzymes from the reac- laborious and in low yields. We have favored
84
Enzymatic Preparation of Tetrapyrrole Intermediates
Figure 5. The biosynthesis of siroheme from uroporphyrinogen. Uroporphyrinogen III is methylated at positions 2 and 7 to give precorrin-2 by the enzyme uroporphyrinogen methyltrans-
the B. stearothermophilus CobA and the E.
coli CysG, since these can be purified easily
in a couple of hours by metal chelate chro-
matography. For instance, the B. stearother-
mophilus CobA can be purified from strain
ER262 (Table 1), which is BL21(DE3)
pLysS transformed with pER259 (cobA
cloned into pET14b). As for all His-tagged
enzymes, the isolation procedure is very
similar to that described in Procedure 5
(Section 6.1). About 15 mg of purified
enzyme can be obtained per liter of culture.
tion of SAM helps to overcome inhibition ferred to a small plastic column. The resin is
with S-adenosyl-L-homocysteine. The reac- washed with buffer, and buffer containing
tion should be undertaken at pH 8.0 in 50 250 mM NaCl, to remove the more loosely
mM Tris-HCl buffer at either 30° or 37°C. bound proteins, and the precorrin-2 is elut-
As precorrin-2 is so unstable, we recom- ed in buffer containing 2 M NaCl. Precor-
mend that a high concentration of the uro- rin-2 is highly unstable with a tendency to
porphyrinogen methyltransferase be used in form mono- and dilactones. The com-
the reaction at a concentration of about 50 pound is difficult to store and should be
µg/mL. This ensures a rapid synthesis of used immediately. The uroporphyrinogen
precorrin-2, which can be monitored visu- methyltransferases are very susceptible to
ally since the solution turns a bright yellow feedback inhibition by S-adenosyl-L-homo-
color. In fact, precorrin-2 has a broad cysteine, and therefore, to achieve high
absorption maximum around 350 to 400 yields of precorrin-2 (in excess of 90%), a
nm. The newly synthesized precorrin-2 can high concentration of enzyme and SAM are
be separated from the other components of required in the incubation mixture.
the incubation mixture by ion exchange Sirohydrochlorin can be synthesized
chromatography. After mixing in a few mil- from precorrin-2 by the inclusion of either
liliters of ion exchange resin such as DEAE CysG or Met8p together with NAD+ to
Sephacel, the solution is slowly stirred for the above incubation (25). These enzymes
about 1 minute. Once the resin has settled, are purified in the same manner as
the majority of the supernatant can be described for the CobA (above) from the
decanted, and the resin slurry can be trans- appropriate strains shown in Table 1. The
Figure 6. Spectra of precorrin-2, sirohydrochlorin, and cobalt-sirohydrochlorin. The spectrum of precorrin-2 (large dashed line)
has a broad absorption maximum around 350 to 400 nm. The spectrum of sirohydrochlorin (filled line) has a more defined
absorption maximum at 378 nm, while cobalt sirohydrochlorin (dashed line) has defined maxima at 410 and 595 nm.
86
Enzymatic Preparation of Tetrapyrrole Intermediates
87
M.J. Warren and P.M. Shoolingin-Jordan
8.3. Synthesis of Coproporphyrinogen a requirement for metal ions for activity, the
human enzyme appears functional in the
Coproporphyrinogen III can be effi- absence of any metal or cofactors. Indeed,
ciently generated by the following proto- the simplest source of the enzyme is a His-
col. An incubation mixture containing 50 tagged version of the human enzyme,
mM Tris-HCl buffer, pH 8.0, 2 mM which is easily overproduced in E. coli,
dithiothreitol (DTT), and 5 µM uropor- yielding in excess of 10 mg/L.
phyrinogen III is prepared. The uropor-
phyrinogen III is made as described in Sec- 9.1. Purification of CPO
tion 4. The buffer should be thoroughly
degassed by freeze–thawing under reduced Although the human CPO is thought to
pressure. The reaction is started by the be associated with the outer surface of the
addition of purified UROD (5 µg/mL), inner membrane of the mitochondrion,
and the incubation is performed at 37°C when expressed in E. coli it is easily solubi-
under dim light. The coproporphyrinogen lized in the presence of 0.5% n-octyl-β-D-
III can be removed from the enzyme mix- glucopyranoside (22). Recombinant
ture by ultrafiltration through a PM-10 expression of the human CPO was
membrane in an ultrafiltration unit. The achieved by cloning the cDNA into the
solution should appear colorless, and any expression vector pBTac such that the
appearance of reddish coloration should be cDNA was cloned in-frame with a 6-histi-
taken as a sign of oxidation. The copropor- dine tag at the 5′ end (Table 1). The result-
phyrinogen should be used immediately, ing plasmid, termed pHHCPO, was trans-
although it may be possible to freeze the formed into E. coli JM109. Purification of
solution so long as it is kept under argon. the enzyme is essentially as described in
The yield of coproporphyrinogen from Procedure 5 (Section 6.1), except that the
uroporphyrinogen is in excess of 95%. resuspension buffer for the cell pellet (step
2) is 50 mM NaH2PO4, 300 mM NaCl,
0.5% n-octyl-β-D-glucopyranoside, and
9. SYNTHESIS OF 100 mM Tris-HCl, pH 8.0, containing 1
PROTOPORPHYRINOGEN mM PMSF. The Ni-column is washed with
resuspension buffer containing 20 mM
The synthesis of protoporphyrinogen imidazole, and the CPO is eluted from the
requires the decarboxylation of the two pro- column in resuspension buffer plus 250
pionate side chains on rings a and b of the mM imidazole. After dialysis against resus-
coproporphyrinogen III isomer by the pension buffer to remove the imidazole, the
enzyme coproporphyrinogen oxidase enzyme can be stored frozen at -20°C for
(CPO). There are two independent enzyme several months. The yield of purified
systems that achieve this transformation, enzyme is in excess of 10 mg/L of culture.
representing aerobic (encoded by hemF)
and anaerobic processes (encoded by 9.2. Assay of CPO and Synthesis of
hemN). However, the aerobic enzyme is Protoporphyrinogen
much better characterized, where purified
recombinant hemF-encoded CPO has been The incubation for the synthesis of pro-
shown to require two molecules of oxygen toporphyrinogen is undertaken in 50 mM
during the reaction with the release of two Tris-HCl, pH 8.0, containing 0.2%
molecules of carbon dioxide (22). Although Tween 20 and 2.5 mM glutathione.
some reports have suggested the enzyme has Coproporphyrinogen III is generated by
88
Enzymatic Preparation of Tetrapyrrole Intermediates
the reduction of coproporphyrin III dihy- introducing three new double bonds. In
drochloride (Porphyrin Products, Logan, aerobic organisms, the enzyme would
UT, USA) with 3% sodium amalgam. The appear to require the services of a flavin
reduction itself should be undertaken in cofactor and passes the electrons onto mol-
100 Tris-HCl, pH 8.0, and once the solu- ecular oxygen. The corresponding anaero-
tion turns colorless, or nearly colorless, the bic oxidation of protoporphyrinogen
solution is passed through a small 10-mL remains poorly understood, but in E. coli
column of glass wool. This not only serves it would appear to be a multiprotein com-
to remove the amalgam and mercury, but plex that is coupled to the respiratory chain
the glass wool also appears to bind the por- of the cell. From a commercial standpoint,
phyrin while allowing the porphyrinogen the enzymatic oxidation of protopor-
to pass through (Dailey, personal commu- phyrinogen represents an important target
nication). The pH of the solution is then for a number of herbicides, diphenyl ether
adjusted back to around 7.0 to 8.0 by addi- derivatives, which selectively inhibit the
tion of 2 M morpholinepropanesulfonic enzyme. Defects in the human enzyme are
acid (MOPS), pH 7.0. The copropor- associated with variegate porphyria, the
phyrinogen is added to the incubation form of porphyria that is particularly com-
mixture at a final concentration of 5 µM mon in South Africa (19).
and incubated with CPO at a concentra-
tion of 10 µg/mL. The synthesis of proto- 10.1. Purification of PPO
porphyrin can be followed by coupling a
small portion of the incubation with pro- Perhaps the simplest recombinant
toporphyrinogen oxidase (PPO) (see Sec- source of this enzyme is the PPO from
tion 8). Alternatively, the conversion of Myxococcus xanthus, as described by Dailey
coproporphyrinogen to protoporphyrino- and Dailey (6). This is a PPO that uses
gen can be determined by analysis of the molecular oxygen as the terminal electron
oxidized methyl esters and quantified using acceptor and is a single subunit enzyme. In
an HPLC system (see Chapter 5). The eukaryotes, the enzyme is found on the
incubation should be performed at 37°C cytosolic side of the inner mitochondrial
under dim light. The enzyme can be membrane or associated with chloroplast
removed from the incubation mixture by membranes, while in bacteria, it is a
ultrafiltration through a PM-10 membrane peripheral membrane protein. The gene
attached to an ultrafiltration cell. As with corresponding to the M. xanthus PPO was
the other porphyrinogens, protopor- amplified and modified such that the N
phyrinogen should be used immediately. terminus encodes for a 6-histidine tag. The
The yield of protoporphyrinogen from construct was subsequently cloned into a
coproporphyrinogen is in excess of 95%. Tac-driven derivative of pBTac-1, yielding
the plasmid pMx-PPO (Table 1).
are centrifuged at 100 000× g, and the fluorescence at 635 nm after excitation at
supernatant discarded, then this mem- 405 nm (29). Coproporphyrin emits at
brane fraction is resuspended in 60 mL 610 nm, so it is important to make sure
of NaH2PO4, pH 7.4, 300 mM NaCl, that the auto-oxidation of copropor-
and 0.5% n-octyl-β-D-glucopyrano- phyrinogen is not being observed. The
side. The suspension is centrifuged increase in protoporphyrin fluorescence is
again at 100 000× g to separate the sol- measured over a 10-minute period, and the
ubilized enzyme from the remaining enzyme activity can be deduced with refer-
membranes. ence to a calibration curve for the fluores-
3. His-bind column: This is carried out as cence of a standard solution of protopor-
for CPO, except that PPO is eluted in phyrin. The yield of protoporphyrin from
buffer containing 150 mM imidazole. protoporphyrinogen is in excess of 95%.
The recombinant protein can be
detected by SDS-PAGE, migrating
with a molecular mass of about 50 000 11. SYNTHESIS OF PROTOHEME
Da. The purified protein is yellow in
color due to the presence of the flavin The final step in the synthesis of proto-
cofactor and has a characteristic flavo- heme is the insertion of ferrous iron in a
protein UV/VIS spectrum. reaction that is catalyzed by ferrochelatase.
4. Storage: The protein can be stored In eukaryotes, this enzyme is normally
frozen at -20°C for several months. peripherally associated with the inner mem-
Purified PPO is obtained in excess of brane of the mitochondrion. Quite surpris-
10 mg/L of culture. ingly, the human enzyme contains an iron
sulphur center, although no immediate role
10.2. Synthesis of Protoporphyrin and has been forwarded for its presence. Defects
Assay in the human enzyme are associated with
erythropoietic protoporphyria, a relatively
The synthesis of protoporphyrin can be severe form of porphyria that can cause
achieved either by use of a coupled enzyme severe liver damage (19). In B. subtilis, fer-
system from coproporphyrinogen III or by rochelatase exists as a soluble protein and
chemical reduction of protoporphyrin by represents one of the simplest sources of the
sodium amalgam. The use of a coupled enzyme (11). The increased solubility of the
enzyme system is perhaps more attractive Bacillus enzyme was a major expedient in
and will be discussed here. The incubation the crystallization of the enzyme (2).
is set up as described for coproporphyrino-
gen synthesis above. The incubation for the 11.1. Enzyme Purification
synthesis of protoporphyrinogen is under-
taken in 50 mM Tris-HCl, pH 8.0, con- The B. subtilis hemH is cloned into
taining 0.2% Tween 20, and 2.5 mM glu- pUC18 under control of the lac promoter
tathione. Coproporphyrinogen is added to to give plasmid pLUG18e2. When trans-
the incubation mixture at a final concen- formed into E. coli JM109 cells, the plas-
tration of 5 µM and incubated with CPO mid causes the bacteria to constitutively
at a concentration of 10 µg/mL and PPO overproduce the enzyme to a level of about
at 20 µg/mL. The synthesis of protopor- 10 mg/L of culture (Table 1). The strain
phyrin can be followed fluorometrically by harboring pLUG18e2 is somewhat unsta-
making a 1:10 dilution of the incubation ble, and fresh transformants need to be
mixture with buffer and determining the used for new cultures.
90
Enzymatic Preparation of Tetrapyrrole Intermediates
protoporphyrin IX at 2 µM, 0.3 mg/mL 6.Dailey, H.A. and T.A. Dailey. 1996. Protoporphyrino-
Tween 80, 20 µM Fe2+, 6 mM DTT, 5 gen oxidase of Myxococcus xanthus. J. Biol. Chem.
271:8714-8718.
mM sodium dithionite, and 2 µg/mL fer- 7.Erskine, P.T., N. Senior, S. Awan, R. Lambert, G.
rochelatase. Fe2+ is prepared daily as a stock Lewis, I.J. Tickle, M. Sarwar, P. Spencer, P. Thomas,
solution of 50 mM (NH4)2Fe(SO4)2 in 0.3 M.J. Warren et al. 1997. X-ray structure of 5-amino-
laevulinic acid dehydratase, a hybrid aldolase. Nat.
M DTT. Protoporphyrin IX is prepared as Struct. Biol. 4:1025-1031.
a stock of 100 µM disodium protopor- 8.Erskine, P.T., E. Norton, J.B. Cooper, R. Lambert, A.
phyrin dissolved in water containing 15 Coker, G. Lewis, P. Spencer, M. Sarwar, S.P. Wood,
M.J. Warren, and P.M. Shoolingin-Jordan. 1999. X-
mg/mL Tween 80. The insertion of ferrous Ray structure of 5-aminolevulinic acid dehydratase
iron can also be followed spectrofluoromet- from Escherichia coli complexed with the inhibitor lev-
ulinic acid at 2.0 A resolution. Biochemistry 38:4266-
rically by measuring the rate of protopor- 4276.
phyrin disappearance, as described above. 9.Ferreira, G.C. and J. Gong. 1995. 5-Aminolaevuli-
The yield of protoheme is in excess of 90%. nate synthase and the first step of heme biosynthesis. J.
Bioenerg. Biomembr. 27:151-159.
10.Guo, G.G., M. Gu, and J.D. Etlinger. 1994. 240-kDa
proteasome inhibitor CF-2. is identical to delta-
ABBREVIATIONS aminolevulinic acid dehydratase. J. Biol. Chem.
269:12399-12402.
11.Hansson, M. and L. Hederstedt. 1994. Purification
ALA, 5-aminolevulinic acid; ALAS, 5- and characterisation of a water-soluble ferrochelatase
aminolevulinic acid synthase; PBG, por- from Bacillus subtilis. Eur. J. Biochem. 220:201-208.
12.Hunter, G.A. and G.C. Ferreira. 1995. A continuous
phobilinogen; PBGD, porphobilinogen spectrophotometric assay for 5-aminolevulinate syn-
deaminase; CPO, coproporphyrinogen thase that utilizes substrate cycling. Anal. Biochem.
oxidase; Da, Dalton molecular mass unit; 226:221-224.
13.Jaffe, E.K. 1995. Porphobilinogen synthase, the first
LB medium, Luria-Bertani medium; source of heme's asymmetry. J. Bioenerg. Biomembr.
PMSF, phenylmethanesulfonyl fluoride; 27:169-179.
PPO, protoporphyrinogen oxidase; SAM, 14.Jaffe, E.K. 2000. The porphobilinogen synthase family
of metalloenzymes. Acta Crystallogr. D 56:115-128.
S-adenosyl-L-methionine; SDS-PAGE, 15.Jordan, P.M., G. Burton, H. Nordlöv, M.M. Schnei-
sodium dodecyl sulfate polyacrylamide gel der, L. Pryde, and A.I. Scott. 1979. J. Chem. Soc.,
electrophoresis. Chem. Commun. 204-205.
16.Jordan, P.M. and M.J. Warren. 1987. Evidence for a
dipyrromethane cofactor at the catalytic site of E. coli
porphobilinogen deaminase. FEBS Lett. 225:87-92.
REFERENCES 17.Jordan, PM. 1991. The biosynthesis of 5-aminolae-
vulinic acid and its transformation into uropor-
1.Akhtar, M. and C. Jones. 1986. Preparation of stere- phyrinogen III, p. 1-66. In A. Neuberger and L.L.M.
ospecifically labelled porphobilinogens. Methods van Deenen (Eds.), and P.M. Jordan (Vol. Ed.), New
Enzymol. 123:375-383. Comprehensive Biochemistry, Vol. 19, Biosynthesis of
2.Al-Karadaghi, S., M. Hansson, S. Nikonov, B. Jons- Tetrapyrroles. Elsevier, Amsterdam.
son, and L. Hederstedt. 1997. Crystal structure of fer- 18.Kannangara, C.G., R.V. Andersen, B. Pontoppidan,
rochelatase: the terminal enzyme in heme biosynthesis. R. Willows, and D. von Wettstein. 1994. Enzymic
Structure 5:1501-1510. and mechanistic studies on the conversion of gluta-
3.Blanche, F., L. Debussche, D. Thibaut, J. Crouzet, mate to 5-aminolaevulinate, p. 3-25. In D.J. Chad-
and B. Cameron. 1989. Purification and characteriza- wick, and K. Ackrill (Eds.), The Biosynthesis of
tion of S-adenosyl-L-methionine:uroporphyrinogen Tetrapyrrole Pigments, Ciba Foundation Symposium
methyltransferase from Pseudomonas denitrificans. J. 180. John Wiley & Sons, New York.
Bacteriol. 171:4222-4231. 19.Kappas, A., S. Sassa, R.A. Galbraith, and Y. Nord-
4.Bolt, E.L., L. Kryszak, J. Zeilstra-Ryalls, P.M. mann. 1995. The porphyrias, p. 2103-2160. In C.R.
Shoolingin-Jordan, and M.J. Warren. 1999. Char- Scriver, A.L. Beaudet, W.S. Sly, and D. Valle (Eds.),
acterisation of the R. sphaeroides 5-aminolevulinic The Metabolic and Molecular Basis of Inherited Dis-
acid synthase isoenzymes, HemA and HemT, isolated ease, 7th ed. McGraw Hill, New York.
from recombinant Escherichia coli. Eur. J. Biochem. 20.Li, J.M., C.S. Russell, and S.D. Cosloy. 1989. The
265:1-11. structure of the E. coli hemB gene. Gene 75:177-184.
5.Dailey, H.A. 1977. Purification and characterisation of 21.Mauzerall, D. and S. Granick. 1956. The occurrence
the membrane bound ferrochelatase from Spirillum and determination of δ-aminolevulinic acid and por-
itersonii. J. Bacteriol. 132:302-307. phobilinogen in urine. J. Biol. Chem. 219:435-446.
92
Enzymatic Preparation of Tetrapyrrole Intermediates
22.Medlock, A.E. and H.A. Dailey. 1996. Human proto- Nakanishi, and O. Meth-Cohn (Eds.), and J.W. Kelly
porphyrinogen oxidase is not a metalloprotein. J. Biol. (Vol. Ed.), Comprehensive Natural Products Chem-
Chem. 271:32507-32510. istry, Vol. 4, Amino Acids, Peptides, Porphyrins and
23.Neidle, E.L. and S. Kaplan. 1993. Expression of Alkaloids. Elsevier, Amsterdam.
Rhodobacter sphaeroides hemA and hemT genes encod- 29.Smith, A.G. and W.T. Griffiths. 1993. Enzymes of
ing two 5-aminolaevulinic acid synthase isoenzymes. J. chlorophyll and heme biosynthesis. Methods Plant
Bacteriol. 175:2292-2303. Biochem. 9:299-343.
24.Phillips, J., F.G. Whitby, J.P. Kushner, and C.P. Hill. 30.Spencer, J.B., N.J. Stolowich, C.A. Roessner, and A.I.
1997. Characterisation and crystallization of human Scott. 1993. The Escherichia coli cysG gene encodes the
uroporphyrinogen decarboxylase. Prot. Sci. 6:1343- multifunctional protein, siroheme synthase. FEBS
1346. Lett. 335:57-60.
25.Raux, E., T. McVeigh, S.E. Peters, T. Leustek, and 31.Spencer, P. and P.M. Jordan. 1993. Purification and
M.J. Warren. 1999. The role of Saccharomyces cerevisi- characterisation of 5-aminolaevulinic acid dehydratase
ae Met1p and Met8p in siroheme and cobalamin from E. coli and a study of reactive thiols at the metal
biosynthesis. Biochem. J. 338:701-708. binding domain. Biochem. J. 290:279-287.
26.Scopes, R.K. 1987. Protein Purification, Principles and 32.Thomas, S.D. and P.M. Jordan. 1986. Nucleotide
Practice, 2nd ed. Springer Verlag, Basel. sequence of the hemC locus encoding porphobilinogen
27.Shoolingin-Jordan, P.M., J.E. LeLean, and A.J. Lloyd. deaminase of Escherichia coli K12. Nucleic Acids Res.
1997. Continuous coupled assay for 5-aminolevulinate 14:6215-6226.
synthase. Methods Enzymol. 281:309-316. 33.Whitby, F.G., J.D. Phillips, J.P. Kushner, and C.P.
28.Shoolingin-Jordan, P.M. and K.-M. Cheung. 1999. Hill. 1998. Crystal structure of human uropor-
Biosynthesis of heme, p. 61-107. In D.H.R. Barton, K. phyrinogen decarboxylase. EMBO J. 17:2463-2471.
93
Analysis of Biosynthetic Intermediates,
5 5-Aminolevulinic Acid to Heme
95
C.K. Lim
The above CE method has been modi- applications where high sensitivity and
fied by inclusion of 10% (vol/vol) acetoni- specificity are required.
trile in the running buffer (50 mM ammo-
nium acetate, pH 5.20) and coupled on-line
to electrospray ionization mass spectrome- 3. ANALYSIS OF PORPHYRINS
try (ESIMS) to provide an extremely sensi-
tive and specific analytical method for ALA The naturally occurring porphyrins exist
and PBG (12). The detection limits were in complex mixtures including isomeric
estimated to be 100 and 10 amol of ALA forms. Effective analysis, therefore, requires
and PBG on column, respectively. The sen- high resolution coupled with sensitive
sitivity could be further improved by the use detection. To date, the best technique for
of selected ion recording (SIR) scans or the separation of porphyrins and their iso-
nanospray ionization, or both. mers is HPLC. The resolution achieved by
Figure 1 shows the separation and detec- HPLC has not been reproduced by other
tion of ALA and PBG by CE-ESIMS. The separation methods.
protonated ion of ALA is at m/z 132 and
that of PBG is at m/z 227. However, the 3.1. Extraction of Porphyrins from
protonated PBG was found to lose ammo- Biological Materials for HPLC
nia (NH3) easily in the electrospray source Analysis
to give an intense ion at m/z 210, corre-
sponding to a methylenepyrrolenine ion. Sample preparation is an important and
PBG was, therefore, detected at m/z 210 integrated part of the successful application
for the methylenepyrrolenine ion and mul- of HPLC to the analysis of porphyrins in
tiple reaction monitoring (MRM) acquisi- biological materials. A good sample prepara-
tions could be used for PBG by monitor- tion procedure minimizes quantitative errors
ing the transition from m/z 227 to m/z and places less demand on the chromatogra-
210. This method is recommended for phy, allowing faster and better analysis.
Figure 1. CE-ESIMS of ALA and PBG. Capillary, 70 cm × 75 µm i.d.; running buffer, 50 mM ammonium acetate, pH 5.2:ace-
tonitrile (90:10, vol/vol); running voltage, 20 kV; ESI voltage, 3.5 kV.
96
Analysis of Heme and Its Precursors
Figure 2. Separation of uroporphyrin I and III isomers. (a) On Hypersil-BDS C18 with acetonitrile:1 M ammonium acetate, pH
5.16 (9:91, vol/vol), as eluent; (b) on Hypersil-ODS with acetontrile:1 M ammonium acetate, pH 5.16 (13:87, vol/vol), as eluent;
and (c) on Hypersil-BDS C18 with acetonitrile:1 M ammonium acetate, pH 5.55 (9:91, vol/vol), as eluent. Flow-rate, 1 mL/minute.
98
Analysis of Heme and Its Precursors
fewer residual silanol groups through boxylic acid porphyrin could not be com-
exhaustive end-capping or made by different pletely separated by RP-HPLC, although
bonding procedures to those normally used the type I isomer easily resolved from the
for conventional ODS columns. Residual type III isomers either with 15% (vol/vol)
silanol groups on silica-based C18 columns acetonitrile in 1 M ammonium acetate
interact adversely with basic compounds, buffer, pH 5.16, as eluent on a convention-
causing peak tailing or broadening. In gen- al C18 column or with 28% acetonitrile:
eral, BDS C18 columns give better resolu- methanol (10:90) in 1 M ammonium
tion and faster separation for porphyrins acetate buffer, pH 5.55, as eluent on a
than conventional C18 columns. BDS C18 column. The four type III iso-
Methanol should not be used as the mers were resolved into three peaks in the
organic modifier for the separation of uro- elution order of 7c, 7d, and 7a + 7b (7),
porphyrins, especially when isocratic elution with 28% acetonitrile:methanol (10:90) in
is used. It causes severe peak tailing and 1 M ammonium acetate buffer, pH 5.16,
excessive retention with loss of resolution. as eluent. The letters a, b, c, and d are used
Methanol is a hydrogen-bonding organic throughout this chapter to denote the posi-
modifier. A layer of methanol adsorbed onto tions of methyl groups on rings A, B, C,
the hydrophobic hydrocarbonaceous C18 and D, respectively.
stationary phase surface can form extensive
hydrogen bonds with the 8 carboxylic acid 3.2.3. Type I and Type III
groups of uroporphyrin. The result is long Hexacarboxylic Acid Porphyrins
retention and peak tailing. This phenome-
non is not observed for porphyrins with one There are two type I and six type III
or more methyl groups, since the interaction hexacarboxylic acid porphyrin isomers.
is dominated by hydrophobic interaction The two type I isomers (6Iab and 6Iac)
between the hydrophobic methyl group(s) have been separated from the most abun-
and the stationary phase surface. dant type III isomer (6IIIad) by isocratic
A small proportion (e.g., 10%) of ace- RP-HPLC with 16% (vol/vol) acetonitrile
tonitrile can be added to methanol, and the in 1 M ammonium acetate buffer, pH
mixture (10% acetonitrile and 90% 5.16, as eluent on a Hypersil-ODS column
methanol) can be used as the organic mod- (ThermoQuest, Bellafonte, PA, USA). The
ifier. The more hydrophobic acetonitrile, complete separation of all 8 isomers has
which is also a nonhydrogen bonding not been achieved. Using the above system,
organic modifier, will be preferentially 6IIIac coeluted with 6IIIbd, and 6IIIab
adsorbed onto the stationary phase surface, coeluted with 6IIIbc (7).
thus preventing hydrogen bond formation.
The complete separation of uroporphyrin 3.2.4. Type I and Type III
I, II, III, and IV isomers has not been Pentacarboxylic Acid Porphyrins
achieved. They were resolved into three
peaks in the elution order of I, III + IV, and There are four type III and one type I
II (7,18). The resolution was not improved pentacarboxylic acid porphyrin isomers.
by employing a BDS C18 column. These 5 isomers have been separated by RP-
HPLC on a Hypersil-ODS column with
3.2.2. Type I and Type III 45% (vol/vol) acetonitrile:methanol (10:90)
Heptacarboxylic Acid Porphyrins in 1 M ammonium acetate buffer, pH 5.16,
as eluent (Figure 3a). The elution order was
The four type III isomers of heptacar- 5I, 5bcd, 5abc, 5acd, and 5abd. A reversal
99
C.K. Lim
of elution order between 5I and 5abd was 3.2.6. Protoporphyrin, Heme, and
observed when 19% acetonitrile in 1 M Related Compounds
ammonium acetate, pH 5.16, was used as
the mobile phase (Figure 3b). The presence Dicarboxylic acid porphyrins, heme, and
of methanol in the mobile phase resulted in related compounds are much more hydro-
an overall improvement in resolution. phobic than the other porphyrins described
above. They require a much higher propor-
3.2.5. Coproporphyrin I, II, III, and IV tion of organic modifier for elution. Since
Isomers acetonitrile is immiscible with 1 M ammo-
nium acetate above the proportion of about
Coproporphyrin isomers are easily sepa- 35%, it cannot be used as the sole organic
rated by RP-HPLC (7). The separations of modifier for the separation of this group of
the 4 isomers on a Hypersil-ODS and a compounds. Either a mixture of acetoni-
Hypersil-BDS C18 column are shown in trile:methanol (10:90) or methanol alone
Figure 4 (a and b). Better resolution with can be used instead. Methanol is completely
faster elution times was achieved on the miscible with 1 M ammonium acetate.
Hypersil-BDS C18 column. The Hypersil- The separation of dicarboxylic por-
BDS C18 column also required less ace- phyrins and metalloporphyrins by RP-
tonitrile (23%) for elution than the Hyper- HPLC has been described (10). A typical
sil-ODS column (30%), which is an separation of protoporphyrin and heme-
obvious advantage. related compounds on a Hypersil-ODS
Figure 3. Separation of type I and type III pentacarboxylic acid porphyrin isomers. Column, Hypersil-ODS; eluent (a), 45%
acetonitrile:methanol (10:90, vol/vol) in 1 M ammonium acetate, pH 5.16; eluent (b), 19% (vol/vol) acetonitrile in 1 M ammo-
nium acetate, pH 5.16. The letters a, b, c, and d denote the positions of methyl groups on rings A, B, C, and D, respectively.
100
Analysis of Heme and Its Precursors
column with 86% (vol/vol) methanol in 1 ammonium acetate, pH 5.16, has been
M ammonium acetate buffer, pH 5.16, as described for the separation of type I and
eluent is shown in Figure 5. type III isomers of 8-, 7-, 6-, 5-, and 4-car-
boxylated porphyrins (7). The system is
3.2.7. Separation of Porphyrin Mixtures applicable to analysis where the separation
from Uroporphyrin to of dicarboxylated porphyrins is not
Protoporphyrin required, e.g., urinary porphyrins. The elu-
tion of dicarboxylated porphyrins requires
From uroporphyrin to protoporphyrin, an acetonitrile content higher than its mis-
the compounds differ widely in hydropho- cibility with 1 M ammonium acetate. It
bicity. Gradient elution is therefore essen- should be emphasized that using this gra-
tial for the simultaneous separation of these dient system the acetonitrile content
porphyrins, including their isomers. should not be allowed to exceed 35%, and
A 15-minute linear gradient elution the column must not be washed with ace-
from 13% to 30% acetonitrile in 1 M tonitrile at the end of the separation due to
101
C.K. Lim
the immiscibility problem. It is also impor- um acetate buffer, pH 5.16; solvent B, 10%
tant to remember that whenever acetoni- (vol/vol) acetonitrile in methanol. Various
trile and 1 M ammonium acetate is used elution programs can be used, depending
for elution, as in the separation of the indi- on the applications, with these two solvent
vidual group of porphyrin isomers by iso- mixtures for porphyrin separation (15). The
cratic elution, the column should not be pH of the buffer, 5.16, is optimal for the
washed with acetonitrile before removal of separation of porphyrin mixtures.
ammonium acetate with a solvent in which Figure 7 shows the separation of por-
it is completely miscible. The column may phyrins in the feces and urine of a patient
be washed with 90% methanol or acetoni- with porphyria cutanea tarda (PCT) on a
trile:methanol in water. C18-bonded RP column (Hypersil-ODS).
Porphyrin mixtures including protopor- It clearly demonstrates the flexibility and
phyrin are best separated by gradient elu- applicability of the system.
tion RP-HPLC with (1,7,17) or without
ion-pairing agents (7). Columns of silica
gel chemically bonded with different
hydrocarbon chain lengths, from C1 to
C18, have all been successfully used for the
RP-HPLC separation of porphyrin mix-
tures in biological materials (7,8). With the
increasing use of on-line HPLC-MS in
analysis, including the tetrapyrroles, gradi-
ent elution solvent mixtures that are fully
compatible with MS are necessary. This
rules out systems that use involatile inor-
ganic phosphate in separation.
A simple RP-HPLC system with 0.1%
trifluroroacetic acid (solvent A) and ace-
tonitrile (solvent B) as the gradient elution
solvent mixture has been used for the sepa-
ration of porphyrins. The system, fully
compatible with MS, is able to resolve the
type I and III isomers of 6-, 5-, and 4-car-
boxylated porphyrins. Separation of uro-
and heptacarboxylic acid porphyrin iso-
mers, however, was not achieved (Figure
6). The system is best used for the separa-
tion of porphyrins with fewer numbers of
carboxylic acid groups, including the dicar-
boxylic acid porphyrins.
The ammonium acetate buffer system
that is fully compatible with MS and pro-
vides high efficiency separation of por-
phyrins is the buffer of choice. It is recom- Figure 5. Separation of protoporphyrin and metallopor-
mended that the following gradient phyrins. Column, Hypersil-ODS; eluent, methanol:1 M
ammonium acetate, pH 5.16 (86:14, vol/vol); flow rate, 1
mixtures are used for elution: solvent A, mL/minute. Peaks: 1 = Zn-deuteroporphyrin; 2 = heme; 3 =
10% (vol/vol) acetonitrile in 1 M ammoni- Zn-protoporphyrin; 4 = protoporphyrin.
102
Analysis of Heme and Its Precursors
Figure 6. Separation of a
standard mixture of por-
phyrins. Column, Hyper-
sil-ODS (25 cm × 4.6 mm
i.d., 5 µm particle size); sol-
vents, 0.1% trifluoroacetic
acid (solvent A) and ace-
tonitrile (solvent B); elu-
tion, linear gradient from
25% solvent B (75% sol-
vent A) to 50% solvent B in
30 minutes; flow rate, 1
mL/minute; detection, ab-
sorbance 404 nm. Peaks: 4,
5, 6, 7, and 8 refer, respec-
tively, to tetra(copro)-,
penta-, hexa-, hepta-, and
octa(uro)carboxylic acid
porphyrin; I and III denote
type I and type III isomers.
103
C.K. Lim
ing the relative hydrophobicity. In copro- The type III heptacarboxylic acid por-
porphyrin III, these are 5 (from position 1 phyrin isomers each have only a single
to 3) and 6 (from position 8 to 5) bond CH3 group that dominated the hydropho-
lengths apart, respectively. In copropor- bic interaction. These isomers are, there-
phyrin IV, each of the adjacent CH3 groups fore, very similar in hydrophobicity and
is 5 bond lengths away (from positions 2 to difficult to separate.
8 and 3 to 5) from their nearest nonadja- In metalloporphyrins, hydrophobic
cent CH3 group. The slightly shorter dis- interaction between the side-chain sub-
tance (one bond length) between one of the stituents and stationary phase surface is
adjacent pair CH3 groups and its nearest also an important retention mechanism.
nonadjacent CH3 group (5 instead of 6 However, the insertion of a metal ion,
bonds apart) is sufficient to make the type which completely occupies the center of
IV isomer slightly more hydrophobic and, the porphyrin hole, significantly altered the
therefore, be retained longer than the type electronic environment around the central
III isomers. Similar arguments apply to the N atoms. The retention of metallopor-
separation of the penta- and hexacarboxylic phyrins is also dependent on the ability,
acid porphyrin isomers. and therefore the species, of the inserted
104
Analysis of Heme and Its Precursors
metal ion to accept axial ligands from the phyrin methyl esters on an ODS column
mobile phase. Addition of polar axial lig- by gradient elution from 70% (vol/vol)
ands leads to a decrease in the overall acetonitrile in water to 100% acetonitrile
hydrophobicity of a molecule and, conse- in 30 minutes is shown in Figure 8. The
quently, its retention (10). system is also fully compatible with MS.
In some applications, it is more conve- The porphyrinogens are the true inter-
nient to separate porphyrins as their methyl mediates in the biosynthesis of heme. Por-
ester derivatives. The majority of methods phyrinogens are unstable to oxidation by
reported for the separation of porphyrin air, especially under acidic conditions and
methyl esters are by normal phase (adsorp- in the presence of light, to the correspond-
tion) chromatography on silica gel columns ing porphyrins, and are therefore rarely
(7). RP-HPLC, however, provides better analyzed. However, studies have shown
resolution and, although unable to separate that for isomer separation, the porphyrino-
the type I and III isomers of uro- and hep- gens are usually better resolved than the
tacarboxylic acid porphyrin methyl esters, is porphyrins (3–6). Isomerically pure por-
able to simultaneously separate the type I phyrinogens are needed as substrates for
and III isomers of hexacarboxylic acid, pen- enzymic experiments, and their separation
tacarboxylic acid, and coproporphyrin may also have application in situations
methyl esters in a single gradient elution where the complete resolution of isomers is
run. Furthermore, the polar hydroxylated essential. For example, investigation of the
porphyrins, e.g., hydroxyuroporphyrin, preferred order of uroporphyrinogen decar-
which are difficult to elute from a silica gel boxylation requires the complete separa-
column, are easily eluted from a RP column. tion of all four type III heptacarboxylic
The separation of a mixture of por- porphyrinogen isomers (14).
105
C.K. Lim
Porphyrinogens are easily prepared by re- ammonium acetate buffer, pH 5.16 (8:12:
duction of porphyrins, usually with 3% (wt/ 80, by vol) as mobile phase (4,7). The
wt) sodium amalgam, as described below. superior resolution of porphyrinogen iso-
mers is again demonstrated.
❖ Procedure 2. Preparation of The complete separation of the five type
Porphyrinogens by Reduction I and III pentacarboxylic acid porphyrino-
of Porphyrins gen isomers was achieved on a Hypersil-
ODS column with methanol:1 M ammo-
1. Dissolve the porphyrin in 0.01 M nium acetate buffer, pH 5.16 (40:60,
KOH in a tube. Cover the tube with vol/vol), as eluent (3,7).
foil to prevent exposure to light. The type I, II, III, and IV isomers of
2. Flush the tube and solution with nitro- coproporphyrinogen could be rapidly sepa-
gen and add excess 3% sodium amal- rated on an ODS column with acetoni-
gam. trile:1 M ammonium acetate, pH 5.16
3. Shake or vortex mix the stoppered tube (25:75, vol/vol) as mobile phase. With this
vigorously until no red fluorescence is system, the four coproporphyrinogen iso-
detected under UV light. mers could also be simultaneously separated
4. Transfer the porphyrinogen solution from the four coproporphyrin isomers (5).
immediately into a clean tube contain- Protoporphyrinogen was easily separat-
ing a small volume of 0.1 M EDTA, ed from protoporphyrin by RP-HPLC
flush with nitrogen, and keep the stop- with methanol:1 M ammonium acetate
pered tube on ice in the dark until buffer, pH 5.16 (90:10, vol/vol), as eluent.
analysis. The porphyrinogens are stable The separation of these two compounds
for at least 1 hour under these condi- may be useful for studying their intercon-
tions. vertion either chemically or enzymatically.
Uroporphyrinogen I and III isomers
have been separated on an ODS column
with 6% (vol/vol) acetonitrile in 1 M 7. RETENTION BEHAVIOR OF
ammonium acetate buffer, pH 5.16, as elu- PORPHYRINOGENS IN RP-HPLC
ent. The elution order was III before I,
which is the reversed of that observed for Although hydrophobic interaction is
the corresponding porphyrins. The por- undoubtedly still the mechanism govern-
phyrinogens are also more hydrophilic ing the retention of porphyrinogens in RP-
than the porphyrins, requiring less organic HPLC, the reversal in the elution order
modifier for elution. observed for many of the porphyrinogen
The four type III heptacarboxylic acid isomers in comparison to porphyrins
porphyrin isomers, which could not be requires explanation. For example, while
separated by RP-HPLC, were completely coproporphyrin isomers were eluted in the
resolved as the porphyrinogens with ace- order I, III, IV, and II based on the
tonitrile:methanol:1 M ammonium acetate hydrophobic interaction hypothesis, the
buffer, pH 5.16 (7:3:90, by vol), as eluent corresponding coproporphyrinogen iso-
(Figure 9). The elution order was 7a, 7c, mers were eluted in the order of I, II, III,
7b, and 7d, which is also different from and IV, an apparent contradiction to the
that of the porphyrins (6,14). hydrophobic interaction hypothesis. How-
The six type III hexacarboxylic acid por- ever, reduction of the methine bridges of
phyrinogen isomers have been separated by the rigid porphyrin macrocycle resulted in
RP-HPLC with acetonitrile:methanol:1 M a relatively flexible porphyrinogen mole-
106
Analysis of Heme and Its Precursors
cule, which is able to adopt various confor- to 415 nm and 600 to 620 nm, respective-
mations. In the flexible coproporphyrino- ly, provides a highly sensitive and specific
gen molecules, the small CH3 group in method of detection. Heme is a nonfluo-
each isomer may be subjected to varying rescence compound and cannot be detect-
degrees of steric hindrance or shielding by ed fluorimetrically.
the larger propionic acid groups, depend- The porphyrinogens do not fluoresce
ing on the adopted conformation. This and have only relatively weak UV absorp-
alters the expected hydrophobic surface tion at the 220 to 240 nm region. They
areas available for interaction, which results can be detected with a UV detector set at
in a change in elution order. Since the con- 220 nm, but are best detected electrochem-
formations of the various groups of por- ically with an amperometric or coulomet-
phyrinogens under RP-HPLC conditions ric detector because of their ease of oxida-
are unknown, it makes prediction of the tion. Porphyrins and metalloporphyrins are
elution order difficult. also electro-active and can be detected with
an electrochemical detector.
In terms of sensitivity, specificity, and
8. HPLC DETECTORS FOR ability to positively identify and characterize
PORPHYRINS, METALLO- compounds, the best “detector” currently
PORPHYRINS, AND available is the mass spectrometer. On-line
PORPHYRINOGENS HPLC-ESIMS and tandem mass spectrom-
etry (MS/MS) provide unsurpassed speci-
Porphyrins and metalloporphyrins have ficity for the analysis of tetrapyrroles.
an intense absorption band at the 400 nm The methyl esters of porphyrins are usu-
region (Soret band). Detection at the Soret ally used in MS analysis because they ionize
band region with an UV-visible detector better in the ion source than free acid por-
set at 400 to 405 nm allows the simultane- phyrins. Porphyrins with a higher number
ous detection of all porphyrins and metal- of carboxylic acid groups, e.g., uro- and hep-
loporphyrins. tacarboxylic acids porphyrins, are more dif-
Porphyrins also have intense red fluores- ficult to ionize than those with a lesser num-
cence, and a fluorescence detector set at ber of carboxylic acid groups, such as copro-
excitation and emission wavelengths of 400 and protoporphyrins. With the recent devel-
107
C.K. Lim
108
Analysis of Heme and Its Precursors
friendly instruments of very high sensitivi- and coproporphyrinogen isomers with amperometric
ty and resolution. This will allow the por- detection. Biochem. J. 234:629-633.
6.Lim, C.K., F. Li, and T.J. Peters. 1987. High-perfor-
phyrins to be analyzed with even greater mance liquid chromatography of type-III heptacar-
sensitivity and specificity, especially in boxylic porphyrinogen isomers. Biochem. J. 247:229-
HPLC-MS/MS analysis. The use of micro- 232.
7.Lim, C.K., F. Li, and T.J. Peters. 1988. High-perfor-
column HPLC-nanospray MS will further mance liquid chromatography of porphyrins. A review.
enhance the sensitivity of detection and J. Chromatogr. Biomed. Appl. 429:123-153.
should be explored for possible application 8.Lim, C.K. and T.J. Peters. 1984. Urine and faecal por-
phyrin profiles by reversed-phase high-performance liq-
in the analysis of the biosynthetic interme- uid chromatography in the porphyrias. Clin. Chim.
diates from ALA to heme. Acta 139:55-63.
9.Lim, C.K., M.A. Razzaque, J. Luo, and P.B. Farmer.
2000. Isolation and characterization of protoporphyrin
glycoconjugates from rat Harderian gland by HPLC,
ABBREVIATIONS capillary electrophoresis and HPLC/electrospray ioniza-
tion MS. Biochem. J. 347:757-761.
10.Lim, C.K., J.M. Rideout, and T.J. Peters. 1984. High-
ALA, 5-aminolevulinic acid; CAPS, 3- performance liquid chromatography of dicarboxylic
(cyclohexylamino)-1-propane sulfonic acid; porphyrins and metalloporphyrins: retention behaviour
and biomedical applications. J. Chromatogr. 317:333-
CE, capillary electrophoresis; CEP, con- 341.
genital erythropoietic porphyria; DMSO, 11.Lim, C.K., J.M. Rideout, and D.M. Samson. 1979.
Determination of 5-aminolaevulinic acid and porpho-
dimethyl sulfoxide; ESIMS, electrospray bilinogen by high-performance liquid chromatography.
ionization mass spectrometry; HPLC, high- J. Chromatogr. 185:605-611.
performance liquid chromatography; LC, 12.Lord, G.A., J.L. Luo, and C.K. Lim. 1999. Capillary
zone electrophoresis/mass spectrometry of 5-aminolae-
liquid chromatography; MECC, micellar vulinic acid and porphobilinogen. Rapid. Comm. Mass
electrokinetic capillary chromatography; Spec. 14:314-316.
MRM, multiple reaction monitoring; MS, 13.Luo, J.L., J. Deka, and C.K. Lim. 1996. Determina-
tion of 5-aminolaevulinic acid dehydratase activity in
mass spectrometry; MS/MS, tandem mass erythrocytes and porphobilinogen in urine by micellar
spectrometry; ODS, octadecylsilyl; PBG, electrokinetic capillary chromatography. J. Chromatogr.
porphobilinogen; RP, reversed-phase; SDS, 722:353-357.
14.Luo, J. and C.K. Lim. 1993. Order of uropor-
sodium dodecylsulfate; SIR, selected ion phyrinogen III decarboxylation on incubation of por-
recording. phobilinogen and uroporphyrinogen III with ery-
throcyte uroporphyrinogen decarboxylase. Biochem.
J. 289:529-532.
15.Luo, J. and C.K. Lim. 1995. Isolation and characteri-
REFERENCES zation of new porphyrin metabolites in human porphyria
cutanea tarda and in rats treated with hexachloroben-
1.Bonnett, R., A.A. Charalambides, K. Jones, I.A. Mag- zene by HPTLC, HPLC and liquid secondary ion mass
nus, and R.J. Ridge. 1978. The direct determination of spectrometry. Biomed. Chromatogr. 9:113-122.
porphyrin carboxylic acids. High-pressure liquid chro- 16.Mauzerall, D. and S. Granick. 1956. The occurrence
matography with solvent systems containing phase- and determination of δ-aminolevulinic acid and por-
transfer agents. Biochem. J. 173:693-695. phobilinogen in urine. J. Biol. Chem. 219:435-446.
2.Chiang, S.C.C. and S.F.Y. Li. 1997. Separation of por- 17.Meyer, H.D., W. Vogt, and K. Jacob. 1984. Improved
phyrins by capillary electrophoresis in fused-silica and eth- separation and detection of free porphyrins by high-
ylene vinyl acetate copolymer capillaries with visible performance liquid chromatography. J. Chromatogr.
absorbance detection. Biomed. Chromatogr. 11:366-370. 290:207-213.
3.Li, F., C.K. Lim, and T.J. Peters. 1987. Separation and 18.Rideout, J.M., D.J. Wright, and C.K. Lim. 1983. High
characterization of pentacarboxylic porphyrinogen iso- performance liquid chromatography of uroporphyrin
mers by high-performance liquid chromatography with isomers. J. Liq. Chromatogr. 6:383-394.
electrochemical detection. Biochem. J. 243:421-423. 19.Rossi, E. and D.H. Curnow. 1986. Porphyrins, Ch.
4.Li, F., C.K. Lim, and T.J. Peters. 1989. Preparation, 10, p. 261-303. In C.K. Lim (Ed.), HPLC of Small
high-performance liquid chromatographic separation Molecules. IRL Press, Oxford.
and characterization of hexacarboxylic porphyrinogens. 20.Weinberger, R., E. Sapp, and S. Moring. 1990. Capil-
J. Chromatogr. 461:353-359. lary electrophoresis of urinary porphyrins with
5.Lim, C.K., F. Li, and T.J. Peters. 1986. High-perfor- absorbance and fluorescence detection. J. Chromatogr.
mance liquid chromatography of uroporphyrinogen 516:217-285.
109
Analysis of Intermediates and End
6 Products of the Chlorophyll
Biosynthetic Pathway
Constantin A. Rebeiz
University of Illinois, Urbana, IL, USA
111
C.A. Rebeiz
biosynthetic pathways that constitute the nm, a bandwidth of 40 nm, and a photon
multibranched Chl a/b biosynthetic path- density of about 0.01 µmol/m2s.
way. This chapter will describe various ana-
lytical techniques that permit the qualitative ❖ Procedure 1. Preparation of an
and quantitative analysis of the intermedi- Ammoniacal Extract
ates of this complex pathway. Chapter 10 in
this volume describes in more general terms 1. Since homogenization of plant tissues
the isolation, quantification, and character- disrupts the plant vacuoles and releases
ization of total Chl a/b. organic acids, NH4OH is added to the
acetone to neutralize the medium and
prevent loss of the acid-labile central
2. MATERIALS AND METHODS Mg atom from the Mg-tetrapyrroles
(19). For many of the compounds
All tetrapyrrole intermediates and end which are labile, such as protopor-
products of the Chl biosynthetic pathway phyrin IX (Proto), the extraction
are loosely or tightly bound to plastid mem- should be carried out under subdued
brane lipoproteins (4,31,32,38). Before lighting and at ice bucket temperatures.
qualitative and quantitative analysis, these 2. Cut the plant tissue into small pieces
tetrapyrroles need to be extracted from their about 1 cm in length.
native environment with organic solvents.
On the basis of their polarity, plant 3. Homogenized in acetone: 0.1 N
tetrapyrroles fall into two broad categories: NH4OH (9:1 vol/vol) at 0° to 4°C for
(i) very apolar fully esterified tetrapyrroles 60 seconds, using a homogenizer (PT
which are soluble in apolar solvents such as 10/35 probe; Brinkman Instruments,
petroleum ethers and hexane; and (ii) less Westbury, NY, USA), at a ratio of 7 mL
apolar tetrapyrroles soluble in more polar of solvent per gram of tissue.
solvents such as acetone. Once tetrapyrroles 4. Centrifugation at 39 000× g for 12 min-
have been partitioned between polar and utes at 1°C.
apolar solvents, they can be subjected to a 5. Decant the ammoniacal acetone extract
variety of purification and analytical tech- and store at -80°C until required.
niques. Before discussing details of purifica- Isolated plastids and subplastidic prepa-
tion and analysis, tetrapyrrole extraction rations can be used, for example, to study
and partitioning into apolar and polar frac- precursor product relationships in organel-
tions will be described. lo or in vitro. Plastids or subplastidic
preparations (see Chapter 12), which are
2.1. Tetrapyrrole Extraction usually suspended in buffers with a pH
range of about 7.0 to 8.0, are extracted
Tetrapyrrole extraction from green tis-
with acetone: 0.1 N NH4OH (9:1 vol/vol)
sues can be carried out under laboratory
at a rate of 10 mL per 1 mL of preparation
light (about 5 µmol/m2s). However, pig-
and then treated as above.
ment extraction from etiolated tissues
should be performed under a green safe- 2.2. Tetrapyrrole Partitioning Between
light that does not photoconvert pro- Apolar and Polar Solvents
tochlorophyllide a (Pchlide a) and Pchlide
a ester (Pchlide a E) to chlorophyllide a Once an ammoniacle extract has been
(Chlide a) and Chl a (66). In our laborato- made, partitioning tetrapyrroles between
ry, we routinely use a low irradiance green apolar and polar solvents is an easy way of
light with a transmission maximum at 503 achieving partial purification prior to
112
Analysis of Intermediate and End Products of Chlorophyll
113
Figure 1. Integrated Chl a/b biosynthetic pathway adapted from Reference 44. To facilitate understanding of the text, various biosynthetic pathways are designated by the numbers 1 to 14.
C.A. Rebeiz
source, which in most cases is a Xenon arc, mixtures. In the following, a brief descrip-
has a limited life span. As the arc lamp tion of common chromatographic proce-
ages, so does its output. As a consequence, dures used in tetrapyrrole separation and
the sensitivity of the instrument decreases purification are described.
as a function of the age of the excitation
lamp. This can be corrected for by adjust- 2.4.1. Paper Chromatography
ing the sensitivity of the instrument on a
daily basis against a stable fluorescent stan- Paper chromatography is one of the old-
dard, such as an acrylic block of rhodamine est techniques used for the separation of
b, by adjustment of the photomultiplier various tetrapyrroles including Chls.
voltage. The stability of the excitation Although more powerful techniques have
source also varies continuously as the volt- since been developed, it is still a viable and
age in the laboratory fluctuates as a func- cheap alternative for the separation of
tion of power demands. This caveat can Proto by ascending chromatography, from
simply be corrected for by recording spec- more polar tetrapyrroles such as uropor-
tra in the ratio mode. In that mode, the phyrins (Uro) and coproporphyrins
excitation beam is split into two beams. (Copro) (37).
One beam passes through the sample (sam-
ple beam), and the other beam, the refer- 2.4.2. Thin Layer Chromatography
ence beam, passes through a quantum
counter such as a cuvette filled with a con- For the past 35 years, thin layer chro-
centrated rhodamine b solution (about 10 matography (TLC) has been the workhorse
mg/mL polyethylene glycol 600). Since of tetrapyrrole purification. However, a
rhodamine b exhibits a constant quantum comprehensive treatment of TLC is
yield between 380 and 600 nm by moni- beyond the scope of this chapter. In our
toring the ratio of the sample and reference laboratory, we prepare our own TLC plates
beams instead of monitoring only the sam- using a variety of adsorbents including cel-
ple beam, the effect of any fluctuation in lulose, silica gel H, and polyethylene.
the electrical current on signal intensity is Silica gel H plates are prepared by mix-
eliminated. The recording of fluorescence ing 42 g of silica gel H (EM Science, Gibb-
spectra in the ratio mode is routine in most stown, NJ, USA) and 135 mL of water.
commercial scanning spectrofluorometers. Cellulose (Cellulosepulver, MN 300;
Finally, the photomultiplier response of Macherey-Nagel, Duren, Germany) plates
the instrument as a function of wavelength are prepared by mixing 22 g of cellulose
is not constant and usually decreases dra- and 135 mL of water. Polyethylene plates
matically in the red region of the spectrum. are prepared by mixing 40 g of chromato-
In computer-interfaced instruments, this is graphic grade polyethylene powder (Poly-
corrected for on-line by multiplying the sciences, Warrington, PA, USA) and 120
signal ratios at every wavelength by a cor- mL of pure or 90% aqueous acetone. Mix-
rection factor. This feature is standard on ing is carried out for 60 seconds in a War-
most sophisticated scanning spectrofluo- ing blender. The plates (0.25 or 0.5 mm in
rometers. thickness) are prepared by using a Desaga
spreader. After air-drying for about 30
2.4. Purification of Tetrapyrroles minutes, the plates are activated by heating
at 105°C overnight. The activated plates
Prior to some analyses, it is desirable to are stored dry in a dessicator until further
purify individual tetrapyrroles from crude use. Unless otherwise indicated, we favor
115
C.A. Rebeiz
nm, Chlide a at 432 and 675 nm, pheo- Chlide a and b emissions.
phytin and Pheobide a [Pheo(bide) a] at F (F622 E400) = undeconvoluted total
413 and 674 nm and Pheo(bide) b at 439 fluorescence emission amplitude observed
and 660 nm. By excitation of the HEAR at 622 nm upon excitation of the tetrapyr-
fraction at 400 nm, i.e., close to the Soret role mixture at 400 nm. This factor cor-
excitation maximum of Proto, and by rects for Copro fluorescence emission.
recording an emission spectrum between F (F638 E440) = undeconvoluted total
580 and 700 nm, it is possible to maximize fluorescence emission amplitude observed
the fluorescence emission of Proto at 633 at 638 nm upon excitation of the
nm. The fluorescence emission of Mg- tetrapyrrole mixture at 440 nm. This fac-
Proto and Mpe are far removed from the tor corrects for Pchlide a fluorescence
emission maximum of Proto and do not emission.
interfere with its determination. Only fluo- F (F675 E440) = undeconvoluted total
rescence emissions from Copro at 622 nm, fluorescence emission amplitude observed
from Pchlide a at 638 nm and to a minor at 675 nm upon excitation of the tetrapyr-
extent from Chlide a and b, are likely to role mixture at 440 nm. This factor corrects
interfere. The possible contribution of for Chlide a and b fluorescence emission.
these fluorescences to Proto fluorescence is See Table 2 for further equations.
eliminated by spectral deconvolution. This The various fluorescence amplitudes to
is achieved by exciting the tetrapyrrole be substituted into Equation 1 are read
mixture at 440 nm, i.e., close to the Soret from the recorded emission spectra. Next,
excitation maximum of Pchlide a and the net fluorescence emission amplitude of
Chlide a and b, and by recording a second Proto, as calculated from Equation 1, is
emission spectrum of the mixture between converted to Proto concentration by refer-
580 to 700 nm. The net fluorescence ence to a standard calibration curve. The
amplitude due only to Proto emission is latter is prepared from standard Proto solu-
deconvoluted from the Copro, Pchlide a, tions of known concentrations and from
and Chlide a and b emissions with the use their fluorescence emission amplitudes.
of Equation 1 (45). The latter are recorded under the same
instrumental conditions by excitation at
Equation 1. Deconvolution of 400 nm. In this manner, Proto can be
Fluorescence Data determined with a precision of about 4%
(45). Minimum detection levels are about
Proto (F633 E400) = 1.055 F (F633 E400) 0.2 pmol/mL. Equation 1 can be used with
- 0.1190 F (F622 E400) - 0.442 F (F638 any spectrofluorometer capable of record-
E440) - 0.0141 F (F675 E440) ing corrected fluorescence emission and
Where: excitation spectra. The user has to con-
Proto (F633 E400) = deconvoluted net struct, however, the required calibration
fluorescence emission amplitude of Proto curve that relates net fluorescence emission
at 633 nm upon excitation of the tetrapyr- amplitudes to Proto concentrations.
role mixture at 400 nm. In the Laboratory of Plant Biochemistry
F (F633 E400) = undeconvoluted total and Photobiology, Proto calculations as
fluorescence emission amplitude observed well as all other spectrofluorometric calcu-
at 633 nm upon excitation of the tetrapyr- lations are automatically carried out by the
role mixture at 400 nm. This fluorescence PC interfaced with the spectrofluorometer,
amplitude is usually contaminated by con- using Porphyrin Analytical Tools® (PAT)
tributions from Copro, Pchlide a, and software (48).
117
C.A. Rebeiz
118
Analysis of Intermediate and End Products of Chlorophyll
119
C.A. Rebeiz
Figure 4. Protochlorophyllide group. Tetrapyrroles which incorporate the fifth isocyclic ring, ring E. LCFA = long-chain fatty acid.
Compound Abbreviation R1 R2 R3
Divinyl-protochlorophyllide a DV Pchlide a methyl vinyl H
Monovinyl-protochlorophyllide a MV Pchlide a methyl ethyl H
Divinyl-protochlorophyllide a ester DV Pchlide a E methyl vinyl LCFA
Monovinyl-protochlorophyllide a ester MV Pchlide a E methyl ethyl LCFA
Divinyl-protochlorophyllide b DV Pchlide b formyl vinyl H
Monovinyl-protochlorophyllide b MV Pchlide b formyl ethyl H
Divinyl-protochlorophyllide b ester DV Pchlide b E formyl vinyl LCFA
Monovinyl-protochlorophyllide b ester MV Pchlide b E formyl ethyl LCFA
120
Analysis of Intermediate and End Products of Chlorophyll
light. The separated tetrapyrroles are can be efficiently used for the quantitative
detected by their red fluorescence. In this separation of Proto from other
system, Proto migrates with a retention tetrapyrroles. Using large columns, it can
factor (Rf) of about 0.8, while Copro and also be conveniently used for preparative
Uro migrate with Rfs of about 0.6 and 0.3, purposes.
respectively (37). Although paper chro- We have successfully used the solvent
matography is useful for separating Proto system of Ho (23) for separating Proto
form other tetrapyrroles, it is not conve- from Uro, Copro, and other multi-
nient for preparative purposes because of caboxylic tetrapyrroles in HEAR fractions
the difficulty of eluting the separated prepared from cancer cell cultures (50).
tetrapyrroles. The same system can be used with HEAR
preparation of plant sources. With this sol-
3.3.2. Chromatographic Separation and vent system, there is no need to convert
Determination by HPLC tetrapyrroles to their methyl esters prior to
chromatography. If feasible, it is always
If HPLC instrumentation is available, it advantageous to avoid derivatization prior
Figure 5. Chlorin group. Tetrapyrroles reduced to the oxidation state of chlorophylls, i.e., a single bond between carbons 7 and 8.
Compound Abbreviation R1 R2 R3
121
C.A. Rebeiz
molar extinction coefficient is very low, In the following sections, each group of
and such determinations require the use of compounds are considered in turn, and
more concentrated solutions. similarities and differences in analytical
The molar extinction coefficient is relat- procedures are highlighted as appropriate.
ed to Proto absorbance and concentration
by Equation 2 (15).
5. ANALYSIS OF MG-
Equation 2. Beer's Law PROTOPORPHYRIN IX
C.A. Rebeiz
Table 1. Molar Extinction Coefficients (Ext Coefficient) of Metabolic Intermediates of the Chl a Biosynthetic Pathway in Different Solvents at
Room Temperature
Tetrapyrrole Solvent Abs ( nm) Ext Coefficient Reference
Molar extinction coefficients of Pchlide a were calculated from the specific absorption coefficients reported in Reference 30 and a molecular weight of
613 for Pchlide a. In Reference 30 the cited specific absorption coefficients were mistakenly assigned to Pchlide a E by the authors.
125
C.A. Rebeiz
with a maximum at 594 to 596 nm. In this tudes vary with the freezing pattern of the
spectral region, interference by other nat- sample. For a given spectrum, however, the
ural tetrapyrroles is not encountered, and ratio of fluorescence amplitudes at any two
correction for fluorescence band overlap is or more wavelengths is independent of the
unnecessary. The fluorescence amplitude at freezing pattern.
595 to 596 nm or the integrated area As a consequence, determination of the
under the fluorescence emission band is amounts of DV and MV Mg-Proto is a 2-
next converted to Mg-Proto concentration step process. First, the total amount of DV
by reference to calibration curves prepared plus MV Mg-Proto in the ether-extracted
under the same instrumental and analytical HEAR fraction is determined at room
conditions. The calibration curves should temperature, exactly as described in section
relate authentic Mg-Proto concentrations 3.2. Next, the DV/MV ratio of the Mg-
to fluorescence emission amplitude at 595 Proto pool is determined from 77 K spec-
to 596 nm or to integrated fluorescence tra recorded in diethyl ether. The amount
areas under the Mg-Proto emission band. of DV and MV Mg-Proto is then calculat-
In this manner, Mg-Proto can be deter- ed from the total amount of Mg-Proto, as
mined with a precision of about 7% to determined from the ether-extracted
8%. Minimum detection levels are about HEAR fraction at room temperature and
0.2 pmol/mL. from the DV/MV Mg-Proto ratio as deter-
mined at 77 K in diethyl ether (63).
5.2. Quantitative Determination of DV To calculate the DV/MV Mg-Proto
and MV Mg-Proto in Crude ratio, two sharp 77 K excitation spectra in
Extracts by Spectrofluorometry diethyl ether need to be recorded. For the
recording of sharp DV and MV Mp-Proto
Before structural studies or determina- and all other 77 K DV and MV Mg-
tion of DV and MV Mg-Proto, it is neces- tetrapyrroles spectra, it is essential to thor-
sary to extract Mg-Proto from the Mpe- oughly wash the diethyl ether fraction with
free ether-extracted HEAR fraction into a 0.5 M solution of KH2PO4 adjusted to
less polar solvents such as diethyl ether. pH 4.8 (or to pH 7.0 for monocarboxylic
Extraction of DV and MV Mg-Proto from Mg-tetrapyrroles) as described in section
the ether-extracted Mpe-free HEAR frac- 3.2. At a pH lower than 4.8, Mg loss
tion is achieved exactly as described in sec- becomes a problem, and at a pH higher
tion 3.2 for Proto. than 4.8, the dicarboxylic Mg-Proto pool
The use of low temperature (77 K) is will be lost by passing into the aqueous
essential to differentiate between DV and phase.
MV Mg-Proto and all other DV and MV The first 77 K excitation spectrum is
Mg-tetrapyrroles. At 77 K, due to consid- recorded from 380 to 500 nm by position-
erable fluorescence excitation and emission ing the emission monochromator at the
band narrowing, the DV and MV Mg- emission maximum of DV Mg-Proto at
Proto pools exhibit respective excitation 591 nm. In this spectrum, DV Mg-Proto
and emission maxima at 424 and 591 nm will exhibit a sharp and narrow excitation
(DV Mg-Proto) and 417 and 589 nm (MV band with a maximum at 424 nm (63).
Mg-Proto) (12). It is not possible, however, The second 77 K excitation spectrum is
to directly determine the amounts of DV recorded from 380 to 500 nm by position-
and MV Mg-Proto from 77 K spectra, ing the emission monochromator at 587
since for a given amount of DV and MV nm, just below the 589 nm emission max-
Mg-Proto, the overall fluorescence ampli- imum of MV Mg-Proto. In this spectrum,
126
Analysis of Intermediate and End Products of Chlorophyll
MV Mg-Proto will exhibit a sharp and nar- Essentially, the tetrapyrrole mixture, as in
row excitation band with a maximum at an aliquot of HEAR or acetone extract, is
417 nm (63). In most cases, if both DV spotted at the origin of a 5 × 20 cm TLC
and MV Mg-Proto are present, two sharp plate of silica gel H. While the plate is still
maxima are observed, one at 424 nm for wet, it is inserted into a cut 1000-mL glass
DV Mg-Proto and one at 417 nm for MV cylinder containing about 10 mL of
Mg-Proto. The net fluorescence excitation toluene:ethyl acetate:ethanol (8:2:2 vol/
amplitudes of DV and MV Mg-Proto are vol/vol). The cylinder is capped with a piece
then deconvoluted and calculated with of aluminum foil. Development is carried
Equations 3 and 4 (Table 2) (63). out at 4°C in darkness. After the solvent
The various undeconvoluted fluores- front has migrated about 10 cm, the plate
cence excitation amplitudes to be substitut- is viewed under 366 nm UV. The separated
ed into Equation 3 and 4 are read from the tetrapyrroles are detected by their red fluo-
recorded excitation spectra. Next, the ratio rescence. In this system, Proto remains at
of the deconvoluted DV Mg-Proto excita- the origin, and Mg-Proto moves with an Rf
tion amplitude to the deconvoluted MV of about 0.09. Monocarboxylic Mg-por-
Mg-Proto excitation amplitude is calculat- phyrins such as Mpe, Pchlides, and Chlides
ed. This ratio is referred to as the apparent move toward the center of the plate. Mg-
DV/MV Mg-Proto excitation ratio. That Proto is eluted in methanol:acetone (4:1
ratio is converted to an authentic DV/MV vol/vol). Elution is achieved by scraping the
Mg-Proto concentration ratio by reference wet silica gel H band into a small beaker
to a calibration curve. The latter relates containing a few milliliters of the metha-
apparent ratios of DV/MV Mg-Proto exci- nol:acetone mixture. This is followed by
tation amplitudes to known ratios of transfer of the slurry to a conical centrifuge
DV/MV Mg-Proto concentrations. The tube with a Pasteur pipet and centrifuga-
calibration curve is constructed as follows: tion at 1800× g for 1 minute on a table top
first, known concentrations of DV and centrifuge. The supernatant containing
MV Mg-Proto dissolved in diethyl ether Mg-Proto, with or without Proto, is
are mixed in different proportions. Second, removed using a Pasteur pipet. The propor-
fluorescence excitation spectra are recorded tions of DV/MV Mg-Proto components
on every mixture at 77 K as described can be determined by spectrofluorometry
above. Third, the apparent fluorescence in diethyl ether at 77 K as described in sec-
excitation ratios at 424/417 nm are plotted tion 5.2. The Mg-Proto pool is transferred
against the authentic concentration ratio to ether, as described in section 3.2, after
for every mixture. In this manner, DV and dilution with a small volume of HEAR.
MV Mg-Proto can be determined with a The ether extract should be washed with a
precision of about 5% to 6% (63). 0.5 M solution of KH2PO4 adjusted to pH
4.8 prior to 77 K spectroscopy.
5.3. Chromatographic Separation of
Mg-Proto 5.3.2. Chromatographic Separation of
DV Mg-Proto from MV Mg-Proto
5.3.1. Chromatographic Separation on on Thin Layers of Polyethylene
Thin Layers of Silica Gel H
Separation of the purified Mg-Proto
The mixed DV-MV Mg-Proto pool can pool into DV and MV components can be
be separated from monocarboxylic Mg-por- achieved on thin layers of polyethylene
phyrins on thin layers of silica gel H (12). developed in 90% aqueous acetone (12).
127
128
C.A. Rebeiz
Table 2. Spectrofluorometric and Spectrophotometric Equations Used to Deconvolute Spectral Band Overlaps
DV Mg-Proto (E424 F591) Diethyl ether 0.1 3 1.014 F (E424 F591) - 0.106. F (E417 F587)
MV Mg-Proto (E417 F587) Diethyl ether 0.1 4 1.013 F (E417 F587) - 0.131. F (E424 F591)
Pchlide a (F638 E440) HEAR 0.1 5 1.0080 F (F638 E440) - 0.0141 F (F675 E440)
- 0.0197 F (F633 E400) + 0.0028 F (F622 E440)
DV Pchlide a (E453 F625) Diethyl ether 0.1 6 1.061 F (E451 F625) - 0.068 F (E437 F625)
1.3307
MV Pchlide a (E437 F625) Diethyl ether 0.1 7 1.060 F (E437 F625) - 0.964 F (E451 F625)
0.8056
Pchlide a (623) Diethyl ether – 8 41.10 (Abs 623) - 4.93 (Abs 663) - 4.93 (Abs 644)
Pchlide a (626) 80% Acetone – 9 33.20 (Abs 626) - 4.48 (Abs 663) - 7.58 (Abs 645)
MV Pchlide b (E463 F643) Diethyl ether 0.1 10 1.0022 F (E463 F643) - 0.0507 F (E455 F660)
MV Chlide b (E455 F660) Diethyl ether 0.1 11 1.0022 F (E455 F660) - 0.0438 F (E463 F643)
MV Pchlide b (F643 E443) Diethyl ether 0.1 12 1.04 F (F643 E463) - 0.54 F (F635 E440)
Pchlide a (F635 E440) Diethyl ether 0.1 13 1.04 F (F635 E440) - 0.08 F (F643 E463)
MV Chlide a (E433 F674) HEAR 0.2 14 [1.2600 F (E433 F674) - 0.1661 F (E412 F674)
- 0.0894 F (E460 F660) - 0.2796 F (E438 F 660)]
- 0.1 Chlide a (E433 F674)
DV Chlide a (E458 F674) Diethyl ether 0.1 15 1.0205 F (E458 F674) - 0.0557 F (E447 F674)
MV Chlide a (E447 F674) Diethyl ether 0.1 16 1.0205 F (E447F674) - 0.3749 F (E458 F674)
Chlide a (663) Diethyl ether – 17 12.162 (Abs 663) - 1.08 (Abs 644) - 0.32
(Abs 624)
Chlide a (663) 80% Acetone – 18 14.1803 (Abs 663) - 2.9099 (Abs 645) - 0.2238
(Abs 626)
MV Chlide b (E460 F660 ) HEAR 0.2 19 [1.0520 F (E460 F660) - 0.1450 F (E438 F660)
- 0.0551 F (E433 F674) - 0.0030 F (E412 F674)]
+ 0.1810 Chlide b (E460 F660)
Table 2, continued.
DV Chlide b (E498 F666) Diethyl ether 0.1 20 1.0184 F (E498 F666) - 0.0425 F (E475 F660)
MV Chlide b (E475 F660) Diethyl ether 0.1 21 1.0185 F (E475 F660) - 0.4421 F (E498F666)
Chlide b (644) Diethyl ether – 22 20.96 (Abs 644) - 3.31 (Abs 663) - 0.59 (Abs 624)
Chlide b (645) 80% Acetone – 23 26.0064 (Abs 645) - 4.6613 (Abs 6663) - 0.3636
(Abs 626)
MV Pheo(bide) a (E412 F674) HEAR 0.2 24 [1.132 F (E412 F674) - 0.9713 F (E433 F674)
MV Pheo(bide) b (E438 F660) HEAR 0.2 25 [1.179 F (E438 F660) - 0.4033 F (E460 F660)
- 0.541 F (E433 F674) + 0.04873 F (E412 F 674)]
+ 0.095 Pheo(bide) b (E438 F660)
E = excitation wavelength; F = emission wavelength; Abs = absorbance wavelength. The (E…F…) terminology is used to refer to equations that use exci-
tation spectra, the (F….E…) terminology is used to refer to equations that use emission spectra, while the Abs terminology is used in equations that use
absorbance spectra. For example: (E424 F591) = Soret excitation amplitude at 424 nm recorded at an emission wavelength of 591 nm; (F638E440) =
fluorescence emission amplitude at 638 nm elicited by excitation at 440 nm; Abs 663 = absorbance at 663 nm. DL = detection limit (pmol/mL). Eq =
equation number. See section 3.1, Equation 1 for a fully described example using figures for protoporphyrin IX.
129
C.A. Rebeiz
tetrapyrrole, while Mg-Proto is a more Separation of the crude (i.e., prior to sil-
polar dicarboxylic tetrapyrrole. ica gel H purification) or silica gel H-puri-
fied Mpe pool into DV and MV compo-
6.1. Quantitative Determination of Mpe nents can be achieved on thin layers of
in Crude Extracts by polyethylene developed in 90% aqueous
Spectrofluorometry acetone in darkness as described in section
5.3.2. In this solvent, DV and MV Mpe
Upon tetrapyrrole extraction from plant migrate with approximate Rfs of 0.54 and
tissues and partitioning of the pigments 0.68, respectively.
between hexane and acetone (sections 2.1
and 2.2), the mixed DV-MV Mpe pool 6.3. Quantitative Determination of
passes into the HEAR along with other Purified DV and MV Mpe
dicarboxylic and monocarboxylic tetra-
pyrroles. Since Mg-Proto and Mpe exhibit As in the case of DV and MV Mg-
similar spectrophotometric and fluores- Proto, the concentration of purified DV
cence properties, it is necessary to correct and MV Mpe can be determined in a vari-
for the presence of Mg-Proto. In a first ety of organic solvents either by fluores-
step, the total Mg-Proto plus Mpe content cence or absorbance spectroscopy, using
is determined on an aliquot of HEAR as the same Equations and molar extinction
described in section 5.1 for Mg-Proto. coefficients.
Next, the Mpe pool is extracted into
diethyl ether as described in section 3.2 for
monocarboxylic tetrapyrroles. The Mg- 7. ANALYSIS OF MG-
Proto content is then determined on the PROTOPRPHYRIN IX DIESTER
ether-extracted HEAR fraction as
described in section 5.2. The amount of The Mg-Protoporphyrin IX diester
Mpe is calculated by subtracting the (Mpde) pool (35) is a mixed DV-MV fully
amount of Mg-Proto from the total esterified tetrapyrrole pool (Figure 3). The
amount of Mg-Proto plus Mpe. proportion of DV and MV Mpde appears
Similarly, DV and MV Mpe are deter- to depend upon the plant species (12). The
mined, as for Mg-Proto (section 5.2), after same equations used for the determination
transfer of the Mpe pool to diethyl ether. of Mg-Proto and Mpe are also used for
Mpde determination. What differs are the
6.2. Chromatographic Separation of extraction procedures, since Mpde is an
Mpe from other Tetrapyrroles apolar fully esterified tetrapyrrole.
131
C.A. Rebeiz
tion of the vinyl group at position 4 to mixed DV-MV Pchlide a in the presence of
ethyl, a reaction catalyzed by [4-vinyl]- other di- and monocarboxylic tetrapyrroles
Pchlide a reductase (65). The other is can be determined without further purifi-
formed by conversion of MV Mpe to MV cation, after correction for spectral band
Pchlide a (64), a reaction catalyzed by a overlap.
putative MV cyclopentanone ring syn- This is achieved by exciting the
thetase (Figure 1, pathway 9). MV Pchlide tetrapyrroles in the HEAR fraction at 440
a formation from DV Pchlide a via path- nm, i.e., close to the Soret excitation maxi-
way 2 takes place in both dark divinyl mum of Pchlide a and Chlides a, and by
(DDV)-light-dark divinyl (LDDV), i.e., recording an emission spectrum between
DDV-LDDV, plant species such as cucum- 580 to 700 nm. A second emission spec-
ber and in dark monovinyl (DMV)-light- trum is elicited by excitation at 400 nm,
dark monovinyl (LDMV), i.e., DMV- i.e., close to the Soret excitation maxima of
LDMV, plant species such as corn wheat Proto and Copro. The net fluorescence
and barley. Conversely, MV Pchlide a for- amplitude due only to Pchlide a emission
mation from MV Mpe via pathway 9 is deconvoluted from the Copro, Proto,
appears to predominate in DMV-LDMV and Chlide a emissions with the use of
plant species (1). Equation 5 (Table 2) (45). The Pchlide a
DV Pchlide a occupies a central position net fluorescence amplitude is converted to
in the DV carboxylic Chl a biosynthetic Pchlide a concentration by reference to a
pathway as a precursor of MV Pchlide a standard calibration curve. The latter is
and DV Chlide a (Figure 1, pathways 1 prepared as described in section 3.1 from
and 2). In contrast to MV Pchlide a, only standard Pchlide a solutions of known con-
one pool of DV Pchlide a, which is part of centrations and from their fluorescence
biosynthetic pathway 1, exists in etiolated emission amplitudes. Pchlide a can be
and green plants (44). determined with a precision of about 6%
(45). Minimum detection levels are about
8.1. Quantitative Determination of 0.1 pmol/mL.
Pchlide a in Crude Extracts by
Spectrofluorometry 8.2. Quantitative Determination of MV
and DV Pchlide a in Crude Extracts
Upon tetrapyrrole extraction from plant by Spectrofluorometry
tissues and partitioning of the pigments
between hexane and acetone (sections 2.1 It is necessary to extract the Pchlide a
and 2.2), the mixed DV-MV Pchlide a pool from the HEAR fraction into less
pool passes into the HEAR fraction along polar solvents such as diethyl ether prior to
with other dicarboxylic and monocar- structural studies or determination of MV
boxylic tetrapyrroles. In HEAR at room and DV Pchlide a (section 3.2).
temperature, DV and MV Pchlide a exhib- As in the case of other MV and DV Mg-
it similar emission maxima at 637 to 639 tetrapyrroles, the use of low temperature
nm. In this region of the spectrum, the (77 K) is essential to differentiate between
main interfering tetrapyrrole is DV Proto, MV and DV Pchlide a. At this tempera-
which emits at 632 to 633 nm. Lesser ture, due to considerable fluorescence exci-
interference by the long wavelength emis- tation and emission band narrowing, MV
sion tail of Copro, and the short wave- Pchlide a exhibits a split Soret excitation
length emission tail of Chlide a may also band with maxima at 437 and 443 nm and
be encountered. The concentration of the an emission maximum at 625 nm. Like-
133
C.A. Rebeiz
wise, DV Pchlide a exhibits split Soret exci- next calculated. That ratio is converted to
tation maxima at 443 and 451 nm and an an authentic DV/MV Pchlide a concentra-
emission maximum at 625 nm (63). In this tion ratio by reference to a calibration
case too, determination of the amounts of curve (see section 5.2). The latter relates
DV and MV Pchlide a is a 2-step process. apparent ratios of DV/MV Pchlide a exci-
First, the total amount of DV plus MV tation amplitudes to known ratios of
Pchlide a in the HEAR fraction is deter- DV/MV Pchlide a concentrations. In this
mined at room temperature exactly as manner, DV and MV Pchlide a can be
described in section 8.4. Next, the determined with a precision of 4% to 7%
DV/MV ratio of the Pchlide a pool is (63).
determined at 77 K in diethyl ether. The
amount of DV and MV Pchlide a is then 8.3. Chromatographic Separation of
calculated from the total amount of Pch- Pchlide a
lide a and from the DV/MV Pchlide a
ratio (63). 8.3.1. Chromatographic Separation on
To calculate the DV/MV Pchlide a Thin Layers of Silica Gel H
ratio, one sharp excitation spectrum needs
to be recorded at 77 K in diethyl ether (see The mixed DV-MV Pchlide a pool can
section 5.2). The 77 K excitation spectrum be separated from other monocarboxylic
is recorded from 380 to 500 nm by posi- Mg-porphyrins on thin layers of silica gel
tioning the emission monochromator at H as described in section 5.3.1 (12). Pch-
the emission maximum of MV and DV lide a moves with an Rf of about 0.2 just
Pchlide a at 625 nm. In this spectrum, the behind Mpe (Rf of 0.31) (4). Pchlide a is
short wavelength MV Pchlide a Soret exci- eluted in methanol:acetone (4:1 vol/vol) as
tation maximum at 437 nm will appear as described for Mg-Proto. The proportions
a sharp distinct peak. Likewise, the long of DV/MV Pchlide a components are
wavelength DV Pchlide a Soret excitation determined by spectrofluorometry in
maximum at 451 nm will also appear as a washed diethyl ether extracts at 77 K as
sharp distinct peak (63). The MV Pchlide described in section 8.4.
a long wavelength excitation maximum
and the DV Pchlide a short wavelength 8.3.2. Chromatographic Separation of
excitation maximum will overlap com- DV Pchlide a from MV Pchlide a
pletely and appear as a single excitation
maximum at 443 nm. For calculation of Separation of the of the crude (i.e., prior
the DV/MV Pchlide a ratio, only the Soret to silica gel H purification) or silica gel H-
excitation maxima at 437 and 451 nm are purified Pchlide a pool into DV and MV
used. Because of complete overlap of the components can be achieved on thin layers
DV and MV Soret excitation maxima at of polyethylene developed in 90% aqueous
443 nm, this wavelength cannot be used in acetone in darkness as described in section
the calculations. The net fluorescence exci- 5.3.2 (11). In this solvent, DV and MV
tation amplitudes of DV and MV Pchlide Pchlide a migrate with approximate Rfs of
a are deconvoluted and calculated with the 0.55 and 0.75, respectively. The separated
use of Equations 6 and 7 (63). The ratio of bands are eluted in diethyl ether.
the deconvoluted DV Pchlide a excitation Partial separation of DV and MV Pch-
amplitude to the deconvoluted MV Pch- lide a by HPLC on a polyvinyl alcohol
lide a excitation amplitude, referred to as polymer column has been described by
the apparent excitation amplitude ratio, is Shioi and coworkers (61). The separation
134
Analysis of Intermediate and End Products of Chlorophyll
maximum at 643 nm. The detailed spectral by positioning the emission monochroma-
properties of MV Pchlide b have been pre- tor at 660 nm, the emission maximum of
viously described (56). MV Chlide b in diethyl ether at 77 K. In
this spectrum, the MV Chlide b Soret exci-
9.1.1. Determination of Pchlide b in Crude tation maximum at 475 nm will also
Extracts Containing Chl(ide) a and b appear as a distinct peak. For deconvolu-
tion and calculation purposes, the best dis-
Under natural conditions, MV Pchlide crimination between MV Pchlide b and
b can only be observed in green tissues or MV Chlide b is accomplished by deconvo-
during the advanced stages of greening of luting the net Soret excitation fluorescence
etiolated tissues. Spectrofluorometric amplitudes of the diethyl ether extract at
determination of MV Pchlide b in the 455 and 463 nm, respectively. The net flu-
presence of Pchlide a and Chlide a and b is orescence excitation amplitudes of MV
best achieved in diethyl ether at 77 K. Pchlide b and MV Chlide b at 463 and
However, as mentioned for other MV and 455 nm are deconvoluted and calculated
DV Mg-porphyrins, 77 K spectroscopic with the use of Equations 10 and 11 (25).
determinations cannot be used alone for The apparent MV Pchlide b/MV Chlide b
quantitative measurements. A 2-step tech- excitation ratio is converted to an authentic
nique using room temperature and 77 K MV Pchlide b/MV Chlide b concentration
spectrofluorometry has therefore been ratio by reference to a calibration curve
developed. In a first step, the amount of that relates apparent ratios of MV Pchlide
MV Chlide b in HEAR is determined by b/MV Chlide b excitation amplitudes to
room temperature spectrofluorometry as known ratios of MV Pchlide b/MV Chlide
described in section 13.1. In a second step, b concentrations (see section 5.2). Using
the MV Chlide b/MV Pchlide b ratio is this method MV Pchlide b can be deter-
determined at 77 K in diethyl ether. The mined with a precision of about 10% (25).
amount of MV Pchlide b is then calculated
from the total amount of Chlide b as deter- 9.1.2. Determination of Pchlide b in
mined at room temperature and from the Crude Extracts Containing Pchlide a
MV Chlide b/MV Pchlide b ratio as deter-
mined at 77 K in diethyl ether (25). Experimental conditions may arise
To calculate the MV Chlide b/MV Pch- where MV Pchlide b needs to be deter-
lide ratio, two sharp 77 K excitation spec- mined in the presence of MV Pchlide a
tra in diethyl ether need to be recorded. and in the absence of MV Chlide b. This
First, the HEAR fraction is extracted with situation may be encountered in model
diethyl ether (section 3.2). To optimize the systems prepared from etiolated tissues and
detection of MV Pchlide b, the first 77 K synthetic exogenous MV Pchlide b
excitation spectrum is recorded from 380 (51,54). Under such conditions, quantita-
to 500 nm in the ether extract by position- tive determination of MV Pchlide by spec-
ing the emission monochromator at 643 trofluorometry can be achieved on an
nm, the emission maximum of MV Pch- aliquot of HEAR and on a diethyl ether
lide b in diethyl ether at 77 K. In this spec- extract of HEAR. In this manner, losses
trum, the MV Pchlide b Soret excitation incurred during purification of small
maximum at 463 nm will appear as a sharp amounts of MV Pchlide b are avoided.
distinct peak. To optimize the detection of Again, a 2-step technique using room tem-
MV Chlide b, a second 77 K excitation perature and 77 K spectrofluorometric
spectrum is recorded from 380 to 500 nm determinations are used. In a first step, the
137
C.A. Rebeiz
amount of Pchlide a in the HEAR fraction known ratios of Pchlide a/MV Pchlide b
is determined by room temperature spec- concentrations (see section 5.2). In this
trofluorometry as described in section 8.4. manner, MV Pchlide b can be determined
In a second, step the Pchlide a/MV Pchlide with a precision of about 7% (25).
b ratio is determined at 77 K in diethyl
ether. The amount of MV Pchlide b is then 9.2. Chromatographic Separation of
calculated from the total amount of Pch- Pchlide b from other Tetrapyrroles
lide a as determined at room temperature
and from the Pchlide a/MV Pchlide b ratio A promising HPLC technique has been
as determined at 77 K in diethyl ether described by Scheumann et al. for synthet-
(25). ic MV Pchlide b (54). Separation of MV
To calculate the Pchlide a/MV Pchlide Pchlide b from MV Pchlide a is achieved
b ratio, two sharp 77 K emission spectra in on a C-18 reverse phase silica gel column
diethyl ether need to be recorded. First, the (Hypersil ODS, 5 µm particle size; Shan-
Pchlide a and b pools are extracted from don Lipshaw, Pittsburgh, PA, USA). Elu-
the HEAR fraction with diethyl ether as tion is at a flow rate of 1.0 mL/minute,
described in section 3.2 for monocar- with a step gradient starting with 34% 25
boxylic tetrapyrroles. To optimize the mM aqueous NH4OAC, 15% acetone,
detection of MV Pchlide b, the first 77 K and 51% methanol, and increasing to 16%
emission spectrum is recorded from 580 to water, 60% acetone, and 24% methanol
700 nm by excitation at the Soret excita- within 20 minutes. Finally, the gradient is
tion maximum of MV Pchlide b at 463 held at 100% acetone for 14 minutes. In
nm. In this spectrum, MV Pchlide b emis- this system, MV Pchlide b and a are eluted
sion would appear as a sharp distinct peak with respective retention times of about 9
at 643 nm. To optimize the detection of and 15 minutes.
MV Pchlide a, a second 77 K emission
spectrum is recorded from 580 to 700 nm 9.3. Quantitative Determination of
by excitation at 440 nm. In this spectrum, Purified Pchlide b
MV Pchlide a would exhibit a sharp emis-
sion maximum at 625 nm and a broader Purified MV Pchlide b can be deter-
emission band between 632 and 650 nm. mined in any appropriate solvent from
The best discrimination between Pchlide a room temperature emission spectra elicited
and MV Pchlide b is accomplished by by excitation close to its Soret absorbance
deconvoluting the net emission fluores- maximum. In each case, a calibration curve
cence amplitudes of the crude diethyl ether using known amounts of MV Pchlide b
extract at 635 and 643 nm (25). should be constructed in order to relate
The net fluorescence emission ampli- MV Pchlide b concentration to fluores-
tudes of MV Pchlide b and Pchlide a at cence emission amplitudes of the pure MV
643 and 625 nm, respectively, are decon- Pchlide b solutions at its emission maxi-
voluted and calculated with the use of mum. A similar procedure using Soret
Equations 12 and 13 (25). Next, the calcu- excitation spectra instead of emission spec-
lated apparent Pchlide a/MV Pchlide b tra can also be used.
emission ratio is converted to an authentic Alternatively, absorbance spectroscopy
Pchlide a/MV Pchlide b concentration may be used. At room temperature, MV
ratio by reference to a calibration curve Pchlide b exhibits a 6-banded absorption
that relates apparent ratios of Pchlide spectrum in various organic solvents. The
a/MV Pchlide b emission amplitudes to wavelengths of absorption maxima depend
138
Analysis of Intermediate and End Products of Chlorophyll
on the solvent. In diethyl ether, a strong has been demonstrated, however, that the
Soret absorbance band with an absorption bulk of the Pchlide a E pool of cucumber
maximum at 442 nm and a less intense red cotyledons, which consists mainly of MV
absorbance band with a maximum at 630 Pchlide a E, is not formed from MV Pch-
nm are observed (Table 1). Table 1 reports lide a by esterification. Instead, MV Pchlide
absorption maxima and molar extinction a and MV Pchlide a E appear to be formed
coefficient values for MV Pchlide b phytyl in parallel from a common precursor at the
ester in various solvents. MV Pchlide b and level of Proto and/or Mg-Proto or Mpe
its esterified analog exhibit identical spec- (36). The MV Pchlide a GG, dihydroGG,
tral and molar extinction properties. The and tetrahydroGG may be considered
Soret and red absorption maxima can both intermediates on the way to the formation
be used for quantitative determinations. If of MV Pchlide a P.
the Soret absorbance is used, a highly puri- The DV Pchlide a pool is considered to
fied sample free of carotenoids and other originate from DV Mpde (Figure 1, path-
tetrapyrroles is required. Carotenoids do way 15) via a series of reactions similar to
not interfere with quantitative determina- those of MV Pchlide a E. Because of exper-
tions using the red absorbance maximum. imental difficulties, a precursor product
The molar extinction coefficients are relat- relationship between DV Mpde and DV
ed to MV Pchlide b absorbance and con- Pchlide a ester has not yet been demon-
centration by Beer's law (Equation 2). strated. The DV Pchlide a GG, DHGG,
and THGG may be considered intermedi-
ates on the way to the formation of DV
10. ANALYSIS OF PCHLIDE ESTER a Pchlide a E.
The protochlorophyllide ester (Pch-
lide E) a pool is a highly heterogeneous 10.1. Quantitative Determination of
fully esterified tetrapyrrole pool. It is pre- Pchlide a E in Crude Extracts by
sent in all etiolated and green tissues so Spectrofluorometry
far surveyed (Rebeiz, unpublished). It
consists of MV and DV Pchlide a E (Fig- Upon tetrapyrrole extraction from plant
ure 4) (10). tissues and partitioning of the pigments
The long chain fatty alcohols esterified between hexane and acetone (sections 2.1
at position 7 of the macrocycle are variable and 2.2), the mixed Pchlide a E pool pass-
and have been reported to consist of ger- es into hexane along with other fully ester-
anylgeraniol (GG), dihydroGG (DHGG), ified tetrapyrroles. Pchlide a and its ester
tetrahydroGG (THGG), and hexahy- exhibit identical spectrofluorometric and
droGG (i.e., phytol) in Scenedesmus obliqu- spectrophotometric properties, so quanti-
us (27), in etiolated and greening cucum- tative determination of Pchlide a E is
ber cotyledons (58), and in etiolated leaves exactly as described in section 8.1. In most
of kidney bean (59). cases, Pchlide a E in the hexane extract will
In the integrated Chl a/b pathway, the be contaminated by trace amounts of Pch-
MV Pchlide a E pool is considered to orig- lide a carried over from acetone to hexane.
inate from MV Mpde via biosynthetic This minor Pchlide a contamination can
pathway 13 (Figure 1). Because of experi- be eliminated by washing the hexane with
mental difficulties, a precursor product rela- ammoniacal acetone or by purification of
tionship between MV Mpde and MV Pch- Pchlide a E on thin layers of silica gel H as
lide a E has not yet been demonstrated. It described below.
139
C.A. Rebeiz
Chlide a is a 2-step process. First, the total Pchlide a move with an Rf of about 0.2 just
amount of DV plus MV Chlide a in behind Mpe (8). The mixed Chlide a-Pch-
HEAR is determined at room temperature lide a band is eluted in methanol:acetone
exactly as described in section 12.1. Next, (4:1 vol/vol). Further separation of Chlide a
the DV/MV ratio of the Chlide a pool is from Pchlide a can be achieved after methy-
determined at 77 K in diethyl ether. The lation with freshly prepared diazomethane.
amount of DV and MV Chlide a is then The methylated pigments are chro-
calculated from the total amount of Chlide matographed on thin layers of silica gel H in
a as determined at room temperature and toluene:ethyl acetate:ethanol exactly as
from the DV/MV Chlide a ratio as deter- described above. In this solvent, Pchlide a E
mined at 77 K in diethyl ether (68). and Chl run with respective Rf values of
To calculate the DV/MV Chlide a ratio, about 0.90 and 0.86. Methyl Chlide a and
one sharp 77 K excitation spectrum in methyl Pchlide a run with respective values
diethyl ether need to be recorded (see sec- of about 0.68 and 0.75. The methylated
tion 5.2). The 77 K excitation spectrum is Chlide a is eluted in diethyl ether. It is usu-
recorded from 380 to 500 nm by position- ally contaminated with small amounts of
ing the emission monochromator at the methylated Pchlide a. The recovery of a
emission maximum of MV and DV Chlide highly pure methylated Chlide a prepara-
a at 674 nm. In this spectrum, the DV tion requires a second purification on silica
Chlide a Soret excitation maximum at 458 gel H as described above (8).
nm will appear as a sharp distinct peak.
The MV Chlide a Soret excitation maxi- 12.5. Separation and Analysis of DV and
mum at 447 nm will also appear as a sharp MV Chlide a
distinct peak (67). The net fluorescence
excitation amplitudes of DV and MV Separation of silica gel H-purified
Chlide a are deconvoluted and calculated methyl Chlide a into DV and MV compo-
with the use of Equations 15 and 16 (Table nents can be achieved on thin layers of
2) (68). The ratio of the deconvoluted DV polyethylene developed in 2-propanol:ace-
Chlide a excitation amplitude to the tone (1:1 vol/vol) as described for DV and
deconvoluted MV Chlide a excitation MV Pchlide a E (section 10.2) (10).
amplitude is converted to an authentic An efficient separation of cucumber DV
DV/MV Chlide a concentration ratio by and MV Chlide a from DV and MV Pch-
reference to a calibration curve that relates lide a has been described (21). Separation is
the apparent ratios of DV/MV Chlide a achieved on a 201 TP 250 × 4.6 mm I.D.,
excitation amplitudes to known ratios of polymeric octadecylsilica, 5 µm particle
DV/MV Chlide a concentrations (see sec- size, and 300 Å pore size column
tion 5.2). In this manner, DV and MV (Vydac/The Separation Group, Hesperia,
Chlide a can be determined with a preci- CA, USA). Elution at a rate of 1.2
sion of 2% to 6% (68). mL/minute is with a linear acetone gradient
from 30% aqueous acetone to 55% aque-
12.4. Chromatographic Separation of ous acetone in 8 minutes, then to 100%
Chlide a from other Tetrapyrroles acetone in 4 minutes. The mobile phase is
kept at 100% acetone for an additional 4
The mixed DV-MV Chlide a pool can be minutes. In this system, MV and DV
separated from other monocarboxylic Mg- Chlide a eluted separately between 3 and 4
porphyrins on thin layers of silica gel H as minutes, while MV and DV Chlide a elut-
described in section 5.3.1 (8). Chlide a and ed separately between 6 and 8 minutes.
143
C.A. Rebeiz
partitioning between hexane and acetone below the DV Chlide b excitation maxi-
(sections 2.1 and 2.2), the MV Chlide b mum. The net fluorescence excitation
pool passes into HEAR along with other amplitude of the Chlide b pool at 460 nm
dicarboxylic and monocarboxylic tetra- is next converted to DV Chlide b concen-
pyrroles such as MV Chlide a (673–674 tration by reference to a standard calibra-
nm emission), MV pheophorbide a (673 tion curve that relates standard DV Chlide
nm emission), and smaller amounts of MV b excitation amplitudes at 460 nm to con-
Pheophorbide b (660 nm emission). It is centrations.
also accompanied by Pchlide a (638 nm
emission) and in some cases by small 13.2. Quantitative Determination of MV
amounts of Mg-Proto and Mpe (595 nm and DV Chlide b in Crude Extracts
emission). by Spectrofluorometry
It is possible to determine the amounts
of MV Chlide b without prior purification Before structural studies or determina-
by room temperature excitation spectroflu- tion of MV and DV Chlide b, it is neces-
orometry, since the emission maxima of sary to extract Chlide b from HEAR into
MV Chlide a and b and MV pheophorbide diethyl ether (section 3.2).
a and b fall in the red region of the spec- As in the case of other MV and DV Mg-
trum, far away from the emission of other tetrapyrroles, the use of low temperature
Mg-tetrapyrroles. The concentration of (77 K) is essential to differentiate between
Chlide b can be determined from the exci- MV and DV Chlide b. At 77 K, MV
tation spectra after correction for excitation Chlide b exhibits a Soret excitation band
band overlap. This is achieved by recording with an excitation maximum at 475 nm,
2 Soret excitation spectra in HEAR at an excitation shoulder at 485 nm, and an
room temperature. One spectrum is emission maximum at 659 to 660 nm. DV
recorded close to the emission maximum Chlide b exhibits a red-shifted Soret excita-
of MV Chlide a at 674 nm. The other tion maximum at 490 nm, a Soret excita-
spectrum is recorded close to the emission tion shoulder at 498 nm, and an emission
maximum of MV Chlide b at 660 nm. The maximum at 666 nm (67). In this case too,
net fluorescence excitation amplitude of determination of the amounts of DV and
MV Chlide b, as calculated from Equation MV Chlide b is a 2-step process. First, the
19 (7), is converted to MV Chlide b con- total amount of DV plus MV Chlide b in
centration by reference to a standard cali- HEAR is determined at room temperature
bration curve that relates standard MV (section 13.1). Next, the DV/MV ratio of
Chlide b solutions of known concentra- the Chlide b pool is determined at 77 K in
tions to their fluorescence excitation ampli- diethyl ether. The amount of DV and MV
tudes. In this manner, MV Chlide b can be Chlide b is then calculated from the total
determined with a precision of about 1% amount of Chlide b as determined from
for pure MV Chlide b to 39% for extracts the HEAR fraction at room temperature
containing only 6% MV Chlide b. and from the DV/MV Chlide b ratio as
In those cases where the Chlide b pool determined at 77 K in diethyl ether (67).
consists mainly or exclusively of DV To calculate the DV/MV Chlide b ratio,
Chlide b with or without smaller amounts 2 sharp 77 K excitation spectra in diethyl
of DV Chlide a (as described above), total ether need to be recorded. The first 77 K
Chlide b can be evaluated using Equation excitation spectrum is recorded from 380
19 to calculate the net Soret fluorescence to 500 nm by positioning the emission
excitation amplitude at 460 nm, i.e., 8 nm monochromator at the emission maximum
145
C.A. Rebeiz
pool vary widely depending on the green- formed in etiolated plants via pathway 13
ing group affiliation of the plant species by photoconversion of MV Pchlide a E (9).
(24), the phase of the photoperiod (9), and The third subpool appears to be formed via
light pretreatment of the plant tissue (2). pathway 14 by dark reduction of MV Pch-
The chemical structure of MV Chl a phy- lide a E during seed germination in total
tol (Chl a P) (Figure 5) was determined by darkness (Rebeiz, unpublished).
Fischer and Stern (20). In the integrated In the integrated Chl a/b pathway, DV
Chl a/b biosynthetic pathway, the MV Chl Chl a is depicted as a precursor of one of
a P pool is considered to consist of 4 dif- the MV Chl a subpools (Figure 1, path-
ferent subpools of MV Chl a P formed via ways 1 and 8). Its specific formation from
different Chl a biosynthetic pathways (44). DV Chlide a and its conversion to MV
Two subpools are formed from MV Pch- Chl a were recently documented (2). The
lide a and MV Chlide a via pathways 2 and nature of the fatty alcohol(s) at position 7
9 (Figure 1). The third subpool is formed has not been investigated. A second DV
from DV Chlide a and MV Chlide a via Chl a pool, with a long chain esterifying
pathways 4 and 5. The fourth subpool is fatty alcohol at position 7 of the macrocy-
formed from DV Chl a by direct conver- cle other than phytol, may also occur in
sion of the vinyl group at position 4 to some plant tissues enriched in DV Pchlide
ethyl via pathway 8 (2). It is conjectured a E such as cucurbit seed coats. Such a pool
that the 4 MV Chl a P subpools are part of may be formed by (photo)conversion of
different Chl-protein complexes having DV Pchlide a E via pathway 8 (Figure 1).
different roles in photosynthesis (44).
During the early phases of greening of 14.1. Quantitative Determination of Chl
etiolated tissues, in addition to MV Chl a a in Crude Extracts
P, each MV Chl a subpool (Figure 5) may
also contain Chl a GG, Chl a DHGG, and Upon tetrapyrrole extraction from
Chl a THGG. These Chls may be consid- green(ing) plant tissues and partitioning of
ered as transient intermediates on the way the pigments between hexane and acetone
to the formation of Chl a P (52,53). (sections 2.1 and 2.2), Chl a passes into
In the integrated Chl a/b biosynthetic hexane along with Pchlide a E. In the
pathway, 3 additional subpools of MV Chl process, Chl a may become contaminated
a, with long chain fatty alcohols at position by very small amounts of Chlide a, which
7 of the macrocycle other than phytol, are may be removed by washing the hexane
depicted. These Chls occur in trace extract with an equal volume of acetone:
amounts and may be either intermediates, water:0.1 N NH4OH (8:1:1 vol/vol/vol).
on the way to the formation of MV Chl a P, In most green higher and lower plants,
or end products of the Chl biosynthetic the Chl a pool in the hexane fraction con-
pathway. Because of experimental difficul- sists mainly of MV Chl a (673 nm emis-
ties, the esterifying long chain fatty alco- sion). During the early stages of greening
hol(s) at position 7 of the macrocycle have of DDV-LDDV plant species such as
not yet been characterized. The 3 subpools cucumber, the Chl a pool may also contain
have been detected in DDV-LDMV plant small amounts of DV Chl a (673 nm emis-
species such as corn and barley during etio- sion). In green tissues, the Chl a pool is
lation and/or during the early phases of usually accompanied by MV Chl b (657
greening. One subpool is formed from MV nm emission), MV pheophytin a, i.e.,
Chlide a via pathway 12 during the early demetalated Chl a (673 nm emission), and
phases of greening. The second subpool is smaller amounts of MV Pheophytin b (660
147
C.A. Rebeiz
viewed under 366 nm UV light. The sepa- Hypersil MOS2, end-capped, C-8
rated Chls are detected by their red fluores- (6.2%–6.8% carbon), 120 Å pore size, 100
cence. MV Chl a and b migrate with × 4.6-mm column (Shandon Lipshaw) at
respective Rfs of about 0.5 and 0.21, while 30°C. Pigments are separated at a flow rate
DV Chl a and b migrate with respective Rfs of 1 mL/minute by a linear gradient
of about 0.56 and 0.16 (68). The Chl expressed by minute; % solvent A; % sol-
bands are eluted in diethyl ether. vent B and programmed as follows: (0; 75;
25), (1; 50; 50), (20; 30; 70); (25; 0; 100),
14.4. Chromatographic Separation of Chl and (32; 0; 100). The column is recondi-
a by HPLC tioned to original conditions over a 7-minute
period. In this system, solvent A consists of
HPLC has become a very popular ana- methanol:1 M ammonium acetate (70:30
lytical tool for the quantitative separation vol/vol). Solvent B consists of 100%
of complex plant extracts containing Chl a methanol. DV and MV Chl a, elute much
and b. A few HPLC applications will be slower (26–27 min) than DV and MV Chl
described below. b (20–21 min).
ing. As evidenced by on-line emission spec- scribed for Chlide a (sections 12.6 and
tra of eluting peaks, the formation of 4 dif- 12.7).
ferent Chlide a E can be detected during
the first 500 milliseconds, following a 2.5-
millisecond light treatment of etiolated 15. ANALYSIS OF CHL b
barley and corn (Figure 1, pathway 13).
The various Chlide a Es exhibit retention In most higher plants, the Chl b pool
times of 4.5, 5.2, 6.3, and 7.7 minutes. consists exclusively of MV Chl b (Figure
The chemical nature of the esterifying 5). In the Nec1 corn mutant, it consists
LCFAs at position 7 of the macrocycle is exclusively of DV Chl b (Figure 5) (6,68).
still undetermined. It is not clear either In primitive prochlorophyte picoplank-
whether the various Chlide a E are formed tons, the Chlide b pool is a mixed DV-MV
by photoconversion of Pchlide a Es or by tetrapyrrole pool.
rapid esterification of newly formed Chlide In the integrated Chl a/b pathway, the
a (Rebeiz, unpublished). MV Chl b pool consists of 8 different sub-
pools formed via different Chl b biosyn-
14.4.4. Determination of Chl a Esterified thetic pathways (Figure 1). Four subpools
with Geranylgeraniol Derivatives are formed from DV Pchlide a via path-
ways 2, 5, 6, and 8. Two subpools are
During the early stages of greening of formed from MV Pchlide b via pathways 3
etiolated tissues, newly formed Chlide a is and 10. The last 2 subpools are formed
converted to Chl a P via Chl a GG, Chl a from MV Chlide a via pathways 9 and 11.
DHGG, and Chl a THGG (53). Separa- MV Chl b formation via pathways 5 and 8
tion of these various Chl a has been is confined to DDV-LDDV plant species
achieved after pheophytinization (i.e., such as cucumber, while its formation via
demetalation) on a LiChrosorb RP-8, 5 to pathways 9, 10, and 11 is confined to
10 µm particle size column (Knauer, DDV-LDMV plant species such as corn
Oberusel, Germany). Isocratic elution is at barley and wheat and to DMV-LDMV
a flow rate of 1.5 mL/minute, in plant species such as Johnson grass (Figure
methanol:water (95:5 vol/vol) at room 1). Biosynthetic pathways 2, 3, and 6 occur
temperature (55). After extraction with in all three greening groups.
ethyl acetate, pheophytinization is achieved As far as we know, the esterifying alco-
by treatment with HCL (14). In this sys- hol at position 7 of the MV Chl b macro-
tem, the pheophytinized GG Chl a deriva- cycle is phytol. Because of the extreme
tives exhibit the following retention times: biosynthetic heterogeneity of the Chl b
pheophytin a GG (Pheo a GG) = 10 to 11 pool, it is possible that a more thorough
minutes, Pheo a DHGG = 11.8 to 12 min- investigation of the esterifying alcohol at
utes, Pheo a THGG = 14 to 14.2 minutes, position 7 of the Chl b macrocycle may
and Pheo a P = 16 to 16.4 minutes (55). reveal the presence of minor amounts of
LCFAs other than phytol.
14.5. Quantitative Determination of In the integrated Chl a/b pathway, the
Purified DV and MV Chl a DV Chl b pool consists of 2 different sub-
pools formed via 2 different Chl b biosyn-
Since DV and MV Chl a exhibit identi- thetic pathways (Figure 1). One subpool is
cal absorbance properties to DV and MV formed from DV Chlide b via pathway 7.
Chlide a, the concentration of purified Chl The second subpool is formed from DV
a solutions is determined exactly as de- Chl a via pathway 1. DV Chl b formation
150
Analysis of Intermediate and End Products of Chlorophyll
from Chl a and b by chromatography on nations when the red absorbance maximum
thin layers of cellulose. In ligroin (b.p: is used. The molar extinction coefficients are
60°–80°C):acetone:n-propanol (99:10:0.45 related to MV and DV Pheobide
vol/vol/vol), pheophytin a migrates close to absorbance and concentration by Beer's law
the solvent front with an Rf of about 0.9, (Equation 2). Table 1 reports absorption
while Chl a and b move with respective Rfs maxima and molar extinction coefficient
of about 0.46 and 0.23 (7). The pheophytin values for MV and DV Pheo(bide) a in
a band is eluted in diethyl ether. diethyl ether and 80% aqueous acetone.
Separation of pheophytin a and its
derivatives by HPLC has been described in
section 14.4. 17. ANALYSIS OF PHEO(BIDE) b
153
C.A. Rebeiz
19.Falk, J.E. 1964. Porphyrins and Metalloporphyrins, p. phyll(ide) via the acidic and fully esterified biosynthet-
250-253. Elsevier, Amsterdam. ic branches in higher plants. Biochemistry 21:242-247.
20.Fischer, H. and A. Stern. 1940. Die Chimie des 37.Nicolas, R.E.H. and C. Rimington. 1949. Qualitative
Pyrroles, p. 321-324. Kademische, Verlagsgesellschaft. analysis of the porphyrins by partition chromatography.
21.Garrido, J.L. and M. Zapata. 1997. Reversed-phase Scand. J. Cli. Lab. Invest. 1:12-18.
high performance liquid chromatographic separation of 38.Parham, R. and C.A. Rebeiz. 1995. Chloroplast bio-
mono- and divinyl chlorophyll forms using pyridine- genesis 72: a (4-vinyl) chlorophyllide a reductase assay
containing mobile phases and a polymeric octadecylsil- using divinyl chlorophyllide a as an exogenous sub-
ica column. Chromatographia 44:43-49. strate. Anal. Biochem. 231:164-169.
22.Granick, S. 1948. Protoporphyrin 9 as a precursor of 39.Rebeiz, C.A. 1978. The separation of chlorophyll and
chlorophyll. J. Biol. Chem. 172:717-727. pheophytins by reversed phase HPLC. Chromatogra-
23.Houssier, C. and K. Sauer 1969. Optical properties of phy Rev. 4:8-9.
the protochlorophyll pigments II. Electronic absorp- 40.Rebeiz, C.A. and M.B. Bazzaz. 1979. Cell-free agri-
tion, fluorescence and circular dichroism spectra. culture: the concept and its initial implementation, p.
Biochim. Biophys. Acta 172:261-266. 453-471. In C.D. Scott (Ed.), Biotechnology in Energy
24.Ioannides, I.M., D.M. Fasoula, R.K. Robertson, and Production and Conservation. John Wiley & Sons,
C.A. Rebeiz. 1994. An evolutionary study of chloro- New York.
phyll biosynthetic heterogeneity in green plants. 41.Rebeiz, C.A., F.C. Belanger, S.A. McCarty, G.
Biochem. Sys. Ecol. 22:211-220. Freyssinet, J.X. Duggan, S.M. Wu, and J.R. Mattheis.
25.Ioannides, I.M., V.P. Shedbalkar, and C.A. Rebeiz. 1981. Biosynthesis and accumulation of novel chloro-
1997. Quantitative determination of 2-monovinyl pro- phyll a and b chromophoric species in green plants, p.
tochlorophyll(ide) b by spectrofluorometry. Anal. 197-212. In G. Akoyunoglou (Ed.), Photosynthesis V.
Biochem. 249:241-244. Chloroplast Development. Balaban International Ser-
26.Kim, J.S. and C.A. Rebeiz. 1996. Origin of the chloro- vices, Philadelphia.
phyll a biosynthetic heterogeneity in higher plants. J. 42.Rebeiz, C.A., and P. Castelfranco. 1971. Chlorophyll
Biochem. Mol. Biol. 29:327-334. biosynthesis in a cell-free system from higher plants.
27.Knaust, R. and H. Senger. 1994. Monovinyl and Plant Physiol. 47:33-37.
divinyl protochlorophyll in different stages of esterifica- 43.Rebeiz, C.A. and P. Castelfranco. 1971. Protochloro-
tion isolated from mutant C-2A' of the unicellular green phyll biosynthesis in a cell-free system from higher
alga Scenedesmu obliquus. Physiol. Plant. 90:490-496. plants. Plant Physiol. 47:24-32.
28.Kolossov, V., I.M. Ioannides, S. Kulur, and C.A. 44.Rebeiz, C.A., I.M. Ioannides, V. Kolossov, and K.J.
Rebeiz. 1999. Chloroplast biogenesis 82: development Kopetz. 1999. Chloroplast biogenesis 80. Proposal of a
of a cell-free system capable of the net synthesis of unified multibranched chlorophyll a/b biosynthetic
chlorophyll(ide) b. Photosynthetica 36:253-258. pathway. Photosynthetica 36:117-128.
29.Koski, V.M. 1950. Chlorophyll formation in seedlings 45.Rebeiz, C.A., J.R. Mattheis, B.B. Smith, C.C. Rebeiz,
of Zea mays L. Arch. Biochem. 29:339-343. and D.F. Dayton. 1975. Chloroplast biogenesis.
30.Koski, V.M. and J.H.C. Smith. 1948. The isolation Biosynthesis and accumulation of protochlorophyll by
and spectral absorption properties of protochlorophyll isolated etioplasts and developing chloroplasts. Arch.
from etiolated barley seedlings. J. Am. Chem. Soc. Biochem. Biophys. 171:549-567.
70:3558-3562. 46.Rebeiz, C.A., A. Montazer-Zouhoor, and H. Daniell.
31.Lee, H.J., M. Ball, and C.A. Rebeiz. 1991. Intraplas- 1984. Chloroplast culture X: thylakoid assembly in
tidic localization of the enzymes that convert δ- vitro. Isr. J. Bot. 33:225-235.
aminolevulinic acid to protoporphyrin IX in etiolated 47.Rebeiz, C.A., R. Parham, D.A. Fasoula, and I.M.
cucumber cotyledons. Plant Physiol. 96:910-915. Ioannides. 1994. Chlorophyll biosynthetic heterogene-
32.Lee, H.J., M.D. Ball, R. Parham, and C.A. Rebeiz. ity, p. 177-193. In D.J. Chadwick and K. Ackrill (Eds.),
1992. Chloroplast biogenesis 65: enzymic conversion of The Biosynthesis of the Tetrapyrrole Pigments. John
protoporphyrin IX to Mg-protoporphyrin IX in a sub- Wiley & Sons, New York.
plastidic membrane fraction of cucumber etiochloro- 48.Rebeiz, C.A. and D.G. Saab. 1995. Porphyrin Analyt-
plasts. Plant Physiol. 99:1134-1140. ical Tools. Software protected by copyright.
33.Lenning, K.V., J.L. Garrido, J. Aristegui, and M. Zap- 49.Rebeiz, C.A., S.M. Wu, M. Kuhadje, H. Daniell, and
ata. 1995. Temperature-programmed high performance E.J. Perkins. 1983. Chlorophyll a biosynthetic routes
liquid chromatography of mono- and divinyl chloro- and chlorophyll a chemical heterogeneity. Mol. Cell.
phyll forms from marine phytoplankton. Chro- Biochem. 58:97-125.
matographia 41:539-543. 50.Rebeiz, N., S. Arkins, K.W. Kelley, and C.A. Rebeiz.
34.Mackinney, G. 1941. Absorption of light by chloro- 1996. Enhancement of coproporphyrinogen III trans-
phyll solutions. J. Biol. Chem. 140:315-322. port into isolated leucocyte mitochondria by ATP.
35.McCarthy, S.A., F.C. Belanger, and C.A. Rebeiz. Arch. Biochem. Biophys. 333:475-481.
1981. Chloroplast biogenesis: detection of a magne- 51.Reinbothe, C., N. Lebedev, and S. Reinbothe. 1999. A
sium protoporphyrin diester pool in plants. Biochem- protochlorophyllide light-harvesting complex involved
istry 20:5080-5087. in de-etiolation of higher plants. Nature 397:80-84.
36.McCarthy, S.A., J.R. Mattheis, and C.A. Rebeiz. 1982. 52.Rudiger, W. 1993. Esterification of chlorophyllide and
Chloroplast biogenesis: biosynthesis of protochloro- its implication for thylakoids development, p. 219-240.
154
Analysis of Intermediate and End Products of Chlorophyll
In C. Sundqvist and M. Ryberg (Eds.), Pigment-Pro- 62.Smith, J.H.C. and C.S. French. 1963. The major
tein Complexes in Plastids: Synthesis and Assembly. accessory pigment in photosynthesis. Ann. Rev. Plant
Academic Press, New York. Physiol. 14:181-224.
53.Rudiger, W. and S. Schoch. 1991. The last steps of 63.Tripathy, B.C. and C.A. Rebeiz. 1985. Chloroplast
chlorophyll biosynthesis, p. 451-464. In H. Scheer biogenesis. Quantitative determination of monovinyl
(Ed.), Chlorophylls. Academic Press, New York. and divinyl Mg-protoporphyrins and protochloro-
54.Scheumann, V., H. Klement, M. Helfrish, U. Oster, phyll(ides) by spectrofluorometry. Anal. Biochem.
S. Schoch, and W. Rudiger. 1999. Protochlorophyl- 149:43-36.
lide b does not occur in barley etioplasts. FEBS Lett. 64.Tripathy, B.C. and C.A. Rebeiz. 1986. Chloroplast
445:445-448. biogenesis. Demonstration of the monovinyl and
55.Schoch, S. 1978. The esterification of chlorophyllide a divinyl monocarboxylic routes of chlorophyll biosyn-
in greening bean leaves. Z. Naturforsch 33c:712-714. thesis in higher plants. J. Biol. Chem. 261:13556-
56.Shedbalkar, V.P., I.M. Ioannides, and C.A. Rebeiz. 13564.
1991. Chloroplast biogenesis. Detection of monovinyl 65.Tripathy, B.C. and C.A. Rebeiz. 1988. Chloroplast
protochlorophyll(ide) b in plants. J. Biol. Chem. biogenesis 60. Conversion of divinyl protochlorophyl-
266:17151-17157. lide to monovinyl protochlorophyllide in green(ing)
57.Shedbalkar, V.P. and C.A. Rebeiz. 1992. Chloroplast barley, a dark monovinyl/light divinyl plant species.
biogenesis: determination of the molar extinction coef- Plant Physiol. 87:89-94.
ficients of divinyl chlorophyll a and b and their pheo- 66.Vernon, L.P. 1960. Spectrophotometric determination
phytins. Anal. Biochem. 207:261-266. of chlorophylls and pheophytins in plant extracts. Anal.
58.Shioi, Y. and T. Sasa. 1883. Formation and degradation Biochem. 32:1144-1150.
of protochlorophylls in etiolated and greening cotyle- 67.Withrow, R.B. and L. Price. 1957. A darkroom safe-
dons of cucumber. Plant Cell Physiol. 24:835-840. light for research in plant physiology. Plant Physiol.
59.Shioi, Y. and T. Sasa. 1983. Compositional hetero- 32:244-248.
geneity of protochlorophyllide ester in etiolated leaves 68.Wu, S.M., J.M. Mayasich, and C.A. Rebeiz. 1989.
of higher plants. Arch. Biochem. Biophys. 220:286- Chloroplast biogenesis: quantitative determination
292. of monovinyl and divinyl chlorophyll(ide) a and b
60.Shioi, Y. and T. Sasa. 1982. Separation of protochloro- by spectrofluorometry. Anal. Biochem. 178:294-
phylls esterified with different alcohols from inner seed 300.
coat of three cucurbitaceae. Plant Cell Physiol. 69.Wu, S.M. and C.A. Rebeiz 1985. Chloroplast biogene-
23:1315-1321. sis. Molecular structure of chlorophyll b (E489 F666).
61.Shioi, Y., K. Watanabe, K.-i. Takmiya, J.L. Garrido, J. Biol. Chem. 260:3632-3634.
and M. Zapata. 1995. Separation of mono-and divinyl 70.Zscheile, F.P. and C.P. Comar. 1941. Influence of
chlorophyll species by high-performance liquid chro- preparative procedure on the purity of chlorophyll
matography using an octadecyl polyvinyl alcohol poly- components as shown by absorption spectra. Bot.
mer column. Anal. Biochem. 231:225-229. Gazette 102:463-481.
155
7 Analysis of Heme and Hemoproteins
Angela Wilks
University of Maryland, Baltimore, MD, USA
157
A. Wilks
158
Analysis of Heme and Hemoproteins
large quantities of protein for biochemical cells when grown on minimal media alone.
and biophysical characterization. The The individual globin chains of hemo-
advance in molecular biological techniques globin can be expressed as fusion proteins,
has not only furthered our understanding of in recombinant E. coli systems, or as is
hemoprotein function but also contributed more common, with the β-globin chain,
to the rapidly expanding role of hemopro- the apoprotein itself (27,41,72). It is also
teins in areas such as cell signaling and regu- possible in E. coli cells to co-express the α
lation of gene expression and function. and β-globins together and obtain active
holoprotein (41,42,44,47,100). In all of
2.1. Hemoprotein Expression Systems these systems, however, some heterogeneity
is observed. A functionally homogenous
2.1.1. Oxygen-Binding Proteins protein is best obtained by removal of the
heme, separation of the α and β-globin
The successful expression and purifica- chains, and subsequent reassembly in the
tion of recombinant myoglobins from a presence of cyanohemin to reform the
number of species have been reported active tetrameric protein (41,42,47).
(103,110,122,127). Initial studies on the More recently, a new class of ligand
expression in Escherichia coli of human binding hemoproteins that act as biological
myoglobin utilized a fusion protein con- sensors have been identified in a number of
sisting of the first 31 amino acids of the organisms. This class of proteins includes
phage lambda cII gene and the tetrapeptide the eukaryotic soluble guanylate cyclase,
Ile-Glu-Gly-Arg, followed by the myoglo- FixL of Rhizobia, and CooA of Rhodospiril-
bin gene sequence (122). The fusion prod- lum rubrum, which sense NO, O2, and
uct was then isolated, reconstituted with CO, respectively (91). While mammalian
heme, cleaved with trypsin, and purified to soluble guanylate cyclases have been
generate the active protein in gram quanti- intractable to expression in E. coli, some
ties. Subsequent studies focused on the progress has been made in insect cell lines
generation of synthetic genes for sperm (135). The oxygen sensor FixL is a modular
whale myoglobin, which allowed for opti- protein consisting of an N-terminal heme
mization of the preferred E. coli codon domain and a C-terminal kinase transmit-
usage and the incorporation of unique ter domain. Expression of both the full-
restriction sites throughout the gene length protein and the heme domain alone
(110,127). The introduction of multiple have been successfully carried out in E. coli
restriction sites enabled the design of cas- TG1 under control of the T7 promoter in
sette based primers for site-directed muta- a pUC8 derived vector (32). The CooA
genesis (110,127). The synthetic gene transcription factor has been subcloned
when cloned into pUC18 (Life Technolo- into pKK223-3 (Amersham Pharmacia
gies, Rockville, MD, USA) and expressed Biotech, Piscataway, NJ, USA) under con-
in E. coli DH5α (Life Technologies), trol of the T7 promoter and expressed in E.
produced active holoprotein indistinguish- coli JM109 cells (2,99).
able from that of the native protein. High
expression yields of myoglobin apoprotein 2.1.2. Oxidases
as inclusion bodies has been obtained in
the T7 promoter based vector pET17b It is only recently that the membrane-
(Novagen, Madison, WI, USA) (49). The bound eukaryotic cytochrome P-450
authors were able to refold and reconstitute enzymes have been routinely expressed in
the protein in quantities of 200 mg/L of recombinant E. coli systems. There are a
159
A. Wilks
number of factors critical for successful enzymes have also been expressed in yeast,
recombinant expression of this family of baculovirus, and mammalian expression
enzymes. Many of the specific require- systems (33). The yeast and mammalian
ments for the successful expression of the expression systems in general produce
cytochrome P-450 enzymes involve the much less total protein, but have been
ability of the cell to adjust their synthesis of extremely valuable in the area of metabo-
heme and lipids to a level that allows for lite and drug research.
correct folding and membrane association. The recent emergence of nitric oxide in
If the rate of expression exceeds the ability physiological functions, such as signal
of the cell to synthesize these factors, then transduction in the cardiovascular and ner-
the protein irreversibly accumulates as vous systems and cytostatic functions of
inclusion bodies. The requirements for the immune system, has led to an extensive
functional cytochrome P-450 expression in body of work on nitric oxide synthase
E. coli were determined to be factors con- (NOS) (66,77). A number of isoforms of
tained in the expression vector, i.e., the pro- nitric oxide synthase have been expressed
moter and lac repressor gene, the structure in baculovirus, including the human endo-
and/or sequence of the mRNA around the thelial, inducible and neuronal enzymes, as
initiation codon, as well as variables such as well as the rat neuronal nitric oxide syn-
the E. coli strain and culture conditions (3). thase (11,12,73,90). Large-scale expression
The powerful T7 phage promoter is capa- of nitric oxide synthase in baculovirus has
ble of synthesizing large amounts of the been achieved, but a critical factor in
recombinant protein, but the majority was obtaining soluble active protein was the
found to accumulate in inclusion bodies. addition of exogenous heme required for
The most successful vectors for the expres- correct folding and binding of the cofactor
sion of cytochrome P-450 are based on the tetrahydrobiopterin (62,64,98).
lac promoter and its derivatives. The two E. The successful expression of nitric oxide
coli expression vectors that have been suc- synthase isoforms in E. coli has developed
cessfully used for cytochrome P-450 more recently and was critical to the rapid
expression are pCW Ori+, a derivative of progress in elucidating many of the struc-
pHSe5 (28,71), and pSP19g10L, a deriva- tural and mechanistic features of NOS (66,
tive of pSPORT-1 (Life Technologies). 77). As with the other members of the cyto-
The fusion of a promoter and ribosomal chrome P-450 family, the NOS enzymes
binding site of a prokaryotic plasmid and were expressed successfully in the pCW
the nucleotide sequence from a eukaryotic Ori+ vectors (30,31,93,131). Factors
gene can have an inhibitory effect on trans- reported to be critical for maximizing the
lation due to the formation of mRNA sec- expression of the rat neuronal nitric oxide
ondary structure. In cytochrome P-450, synthase were the co-expression of groEL
17α-hydroxylase (CYP17) changes in the and groES chaperonins in the protease defi-
amino terminal codons promoted high cient BL21 (DE3) pLysS E. coli strain (93).
level expression, whereas the unmodified However, Gerber and Ortiz de Montellano
DNA failed to produce any detectable pro- have reported successful expression of the
tein (3). Utilization of the pCW Ori+ plas- rat neuronal nitric oxide synthase in the
mid vector together with the first 8 codons absence of chaperonins (31). The inducible
of the modified CYP17 has resulted in the NOS, unlike the neuronal isoform, requires
high-level expression of a number of calmodulin, and co-expression of calmod-
cytochrome P-450 enzymes (35,124). ulin with iNOS is essential for active holo-
The eukaryotic cytochrome P-450 protein (30,131). All of the isoforms
160
Analysis of Heme and Hemoproteins
ber of reasons, including the simpler proto- vectors (Novagen) to high levels utilizing
col and cost effectiveness. Recently, the sol- E. coli BL-21 (pLysS) strain (95). The
uble bacterial heme oxygenases of Syne- yields of protein from this vector ranged
chocystis sp. PCC 6803 and the pathogen from 100 mg/L in rich medium to 40
Corynebacterium diphtheriae have been mg/L in minimal media.
expressed in E. coli as catalytically active It is only recently that S. cerevisiae
proteins (14,129). The C. diphtheriae heme cytochrome c has been expressed heterolo-
oxygenase (HmuO) was expressed under gously in E. coli (81). The successful
the control of the T7 promoter with and expression of the holoprotein required the
without a 6-histidine tag at the C terminus, co-expression of cytochrome c heme–lyase
and no significant differences in protein for covalent attachment of heme a to the
yields or activity were noted (129). apoprotein. The successful expression of
the proteins was achieved by cloning the
2.1.3. Electron Transfer Proteins genes encoding the cytochrome (CYC1)
and the heme–lyase (CYC3) in parallel
As previously described, two of the crit- under the control of the Lac and Trc pro-
ical factors in the expression of many moters in the vector pUC18. The expres-
mammalian hemoproteins has been the sion system yielded 15 mg/L of active iso-
removal of a hydrophobic membrane 1-cytochrome c holoprotein.
anchor domain or redesigning the codons The expression systems described above
encoding the first 5 amino acids of the N encompass only a small fraction of hemo-
terminus. Sligar and coworkers (5) in early proteins, and while it is hard to generalize
studies completely synthesized the genes on the expression of a given class or a partic-
for both the complete rat hepatic ular hemoprotein, some simple generaliza-
cytochrome b5 with the membrane anchor tions can be made. First, hemoproteins with
and the protease-treated soluble form. In covalently attached hemes (cytochrome c)
addition, they incorporated an optimal and/or posttranslational modifications, such
ribosomal binding site and spacer region as glycosylation (myeloperoxidase, lactoper-
for expression of the proteins in E. coli (5). oxidase), have been more successful in
The soluble protein, when expressed in eukaryotic expression systems. Second, a
pUC, accounted for 8% of the total pro- critical factor in recombinant E. coli expres-
tein, with the membrane-bound protein sion of membrane bound hemoproteins,
being somewhat lower and fractionating such as the cytochrome P-450 enzymes, is
with the cell membrane. In later studies, the ability of the cells to synthesize heme
the use of the high expression T7 promoter and other cofactors required for correct fold-
was utilized in the expression of both the ing and activity of the holoprotein.
rat and human cytochrome b5. In these
later studies, polymerase chain reaction 2.2. Protein Purification Methods
(PCR) was used to engineer a sequence
encoding a 4-histidine tag at the N termi- 2.2.1. Oxygen-Binding Proteins
nus allowing rapid purification by nick-
el–chelate affinity chromatography (46). Purification of many hemoproteins has
Optimization of expression for structural been simplified with the development of
nuclear magnetic resonance (NMR) stud- recombinant expression systems and the
ies was carried out by Guiles and coworkers recent advances in affinity chromatogra-
in which they expressed the rat liver phy. Recombinant sperm whale myoglobin
cytochrome b5 in the T7 derived pET3C has been purified to homogeneity primari-
162
Analysis of Heme and Hemoproteins
mg/mL in cold deionized water equili- tion and solubilization of the membranes.
brated with carbon monoxide (CO) by The selection of a suitable detergent and its
bubbling with CO, and the pH was concentration in which to solubilize a
adjusted to 8.0 with a few crystals of given cytochrome P-450 enzyme is largely
Tris. The solution was left undisturbed a matter of trial and error. Solubilization of
overnight at 5°C and was then concen- the protein from bacterial cell pellets is
trated through PM30 membranes. The usually carried out in 100 mM Tris buffer
concentrated reassembled hemoglobin in the pH range 7.5 to 8.0, containing
was then reduced anaerobically with either 500 mM sucrose or 20% glycerol
sodium dithionite in the presence of (18,36). The outer membrane is then solu-
CO and passed through a Sephadex G- bilized in the presence of lysozyme and
25 column equilibrated with 15 mM protease inhibitors (1 mM PMSF, 2 µM
Tris-HCl (pH 8.4) to remove the leupeptin, 10 µM bestatin, and 0.04 U/mL
excess hemin and dithionite. aprotinin) with gentle stirring for 1 hour at
4. Separation of the excess α and β-globin 4°C. The spheroplasts are collected by cen-
chains was accomplished on a DE52 trifugation at 100 000× g for 1 hour. The
column (4 × 5 cm; Whatman) in 15 pellets can then be resuspended in the start
mM Tris-HCl (pH 8.4). The α-globin buffer for solubilization. Solubilization of
and intact hemoglobin were then elut- the cytochrome P-450 is carried out at 4°C
ed from the column with a gradient of for 60 minutes, and the solubilized enzyme
750 mL each of start buffer and 80 is centrifuged at 100 000× g for 60 min-
mM Tris-HCl (pH 8.0). The α-globin utes. In the purification of CYP4All, 1%
was eluted first followed by the hemo- Emulgen 911 appeared the most effective
globin as the major band with the in solubilizing the protein from bacterial
excess β-globin remaining bound to membranes (18). In the case of CYP1A1,
the column. CYP2C10, and CYP3A4, a combination
of 0.625% cholate and 0.62% Triton
2.2.2. Oxidases N101 was found to yield the greatest level
of soluble active holoprotein (36).
The purification of the cytochrome P- Although there is no overall scheme for
450 enzymes has been greatly simplified the purification of recombinant cytochrome
with the advent of recombinant E. coli P-450 enzymes, two general approaches
expression systems and the use of metal- have been taken. First, the purification of
chelate affinity chromatography tech- the native protein has largely been carried
niques. For the purpose of this review, we out by a series of ion exchange chromatogra-
will focus on the purification of cyto- phy steps on DEAE Sephacel (Amersham
chrome P-450 enzymes from bacterial sys- Pharmacia Biotech) and CM Sepharose
tems in light of the considerable advan- (36). More recently, a number of cyto-
tages in high level of expression, ease of chrome P-450 enzymes have been purified
manipulation, and the relatively low cost. utilizing nickel chelate chromatography
A number of laboratories have utilized the (18,69). In these proteins, a 6-histidine tail
pCW Ori+ vector (see Section 2.1.2) in the was engineered at the C terminus allowing
development of expression and purification binding to a nickel–nitriloacetic acid (Ni-
systems for many cytochrome P-450 en- NTA) agarose column. The protein can
zymes (30,31,33,35,38,65,131). then be eluted with imidazole in a 1-step
The critical step in the purification of purification. The nickel–chelate affinity
cytochrome P-450 enzymes is the prepara- method has also been extensively utilized in
165
A. Wilks
the purification of the nitric oxide synthase cytochrome c from E. coli cells has recently
enzymes (30,92,131). been reported (81). The preparation of the
soluble fraction is essentially the same as
2.2.3. Electron Transfer Proteins previously described for the other soluble
hemoproteins such as cytochrome b5 and
The purification of the engineered solu- myoglobin. Purification of the protein was
ble recombinant cytochrome b5 (5) has carried out by successive cation exchange
been the basis for a number of subsequent chromatography steps on CM Sepharose
purification schemes. CL6B and on Mono S HR 10/10 by FPLC.
pathogenic bacteria. Pathogenic bacteria be carried out using the pyridine hemo-
require iron for their survival, and this chrome assay (24,25,114).
requirement is in part linked to their viru-
lence (8). The major source of iron within ❖ Procedure 5. Pyridine Hemochrome
the host is found in heme and hemopro- Method
teins, and many pathogens have developed
1. The isolated hemoprotein in a 4.0-mL
sophisticated uptake systems for the acqui-
sample volume is converted to a pyri-
sition of heme from their environment. A
dine hemochrome by the addition of
number of heme receptors and periplasmic
0.5 mL of pyridine and 0.5 mL 0.5 N
transport proteins have been isolated utiliz-
NaOH solution.
ing hemin–agarose, including outer mem-
brane lipoprotein heme receptors from 2. The reduced spectrum (+ dithionite)
Haemophilus influenzae (60) and Haemo- minus the oxidized spectrum (no
philus ducreyi (22), as well as a secreted dithionite) is calculated at 550 (proto-
hemoglobin–protease in pathogenic E.coli heme) or 557 nm (heme c). The
strain EB1 (80). millimolar extinction coefficients
(∆εox-red) at each of the wavelengths
are 30.0 and 19.1 at 557 and 550 nm,
3. DETECTION AND QUANTITA- respectively (87).
TION OF HEMOPROTEINS The method has also been applied to
more complicated samples such as mito-
A unique and useful property of hemo- chondria, yeast, and photosynthetic bacte-
proteins is the ability to determine the ria (1,4,54,56,87,101). In such cases, it
nature of the heme as well as quantitate was necessary to remove lipids and conta-
both the heme and protein based on the minating photosynthetic pigments by sol-
absorption characteristics of the chromo- vent extraction. Protoheme and heme c can
phore. In addition, the absorption proper- then be separated by differential extraction
ties provide valuable information as to the in acidic organic solvent (25). However,
redox and spin state of individual hemopro- the pretreatment of heme samples must be
teins in multicomponent systems. These approached with some caution. Degrada-
techniques are invaluable in the study of tion of hematins in alkali solution has been
complex systems where individual spectral observed, and the pyridine ferrohemo-
characteristics can be accurately determined. chrome of protoheme is especially labile.
Differential extraction of heme in HCl/
3.1. Detection of Heme in Microsomes acetone results in the partial extraction of
and Whole Cells heme c from tissue samples. Improved
A number of methods for heme detec- methods of detection and quantitation of
tion and quantitation in tissue extracts heme have been developed (as described
have been described. All of the methods above). These methods are based on the
rely on monitoring the distinctive Soret redox difference absorbance measured at 2
absorbance either directly or following wavelengths of the pyridine hemochrome.
extraction of the chromophore. These methods can be utilized without
prior separation of the hemes and organic
3.1.1. Pyridine Hemochrome solvent treatment (52). The method
described below has been successfully used
Assignment of heme type in isolated to quantitate the hemes in the photosyn-
hemoproteins as well as quantitation can thetic bacteria R. sphaeroides (52).
167
A. Wilks
between the reduced and oxidized proteins. tain the heme-associated peroxidase activi-
Subsequent development of a difference ty. The gels (12 cm in length and 1.5 mm
spectral analysis program (DSAP) has fur- thickness) are run overnight at 4°C with a
ther aided the separation and identification constant current (8 mA/gel) (21).
of individual cytochromes of organelles The following procedure has been mod-
such as mitochondria (74). The combina- ified slightly by Dutta and Henry (21) for
tion of HSAP and DSAP provide a method use in electroblotting experiments on
for interpreting complex spectral informa- polyvinylene difluoride (PVDF) mem-
tion of biological samples and may prove brane. Following transfer of the protein to
useful with other measurements of tissue PVDF membranes, the gels are incubated
oxidation in assessing clinical problems in the TMBZ solution for 3 hours at 4°C
associated with free radical reactions. (in the dark). H2O2 is added to a final con-
centration of 30 mM, and the staining is
3.1.5. Detection of Hemoprotein visible after 1 hour (storage in the staining
Peroxidase Activity on SDS-PAGE solution for periods greater than 3 hours
results in irreversible background staining).
Electrophoresis of proteins is commonly The membranes can be quickly destained
used in determination of protein purity, by washing twice in 100% methanol, fol-
molecular weight, and in the case of iso- lowed by destaining in 70 mM sodium sul-
electrofocusing (IEF), the pI of a given fite for 5 minutes. The membranes can
protein. This technique can be useful in then be stained with Coomassie Brilliant
identifying hemoproteins of unknown Blue R-250, 50% methanol, and 10%
molecular weight in an impure protein acetic acid and destained in 90% methanol.
fraction or isolate by taking advantage of
the heme-associated peroxidase activity as a ❖ Procedure 7. TMBZ Staining of
visualization tool (96,116). In this proce- Polyacrylamide Gels
dure, a substrate specific for heme, 3,3′,5,
5′-tetramethylbenzidine (TMBZ) is incu- 1. A solution of 6.3 mM TMBZ in meth-
bated with the SDS-PAGE gel and visual- anol is prepared freshly prior to use.
ized by the addition of hydrogen peroxide Immediately before use, 3 volumes of
(H2O2). One major disadvantage of the the TMBZ methanol solution is mixed
use of this staining technique in SDS- with 7 parts of 0.25 M sodium acetate,
PAGE analysis is that the denaturing con- pH 5.0. The gels are then immersed in
ditions often lead to loss of noncovalently this solution for a period of 1 to 2
bound heme from the protein during the hours (in the dark) and mixed every 15
electrophoresis run. This problem has been minutes.
circumvented with the use of lithium 2. Following incubation, the gels are
dodecyl sulfate (LDS) in place of the SDS. stained with the addition of H2O2 to a
The gels are run as previously described final concentration of 30 mM. The
(59), except SDS is replaced by LDS in all staining should be visible within 3 to 5
buffers. The sample buffer is prepared with minutes with increasing intensity over
a lower concentration of DTT (5 mM), as a 30-minute period.
this can interfere with the TMBZ staining 3. The gels are then placed in isopropa-
of the gel following electrophoresis. The nol:0.25 M sodium acetate (pH 5.0)
samples are immediately loaded onto the (3:7). The solution is changed 3 times
gel following addition of LDS-PAGE sam- to remove any precipitated TMBZ and
ple buffer and are not boiled so as to main- clear the gel background.
170
Analysis of Heme and Hemoproteins
utilized for the detection of blood in urine ture for the membrane-bound P-450 iso-
(hematuria). The detection of blood in zymes has led to the use of indirect meth-
urine showed a linear relationship over a ods for determining active-site structures,
wide range of concentrations with the detec- such as homology modeling (69,83),
tion level as low as 0.1 to 1 pg/µL (19). mechanism-based inactivation (57,58),
The use of chemiluminescence in the and photoaffinity labeling (67,75,134). Al-
detection of hemoproteins has been modi- though such probes have not yet been rou-
fied to specifically detect oxidative metabo- tinely used with hemoproteins, the future
lites of hemoglobin and myoglobin (125). development of photoaffinity probes with
Hemoglobin is a major target for reactive such a broad specificity opens up the possi-
metabolites of drugs as well as environmen- bility of determining topological informa-
tal toxins, and the relatively long half-life of tion on the active sites of hemoproteins
the red blood cell has led to its use as a whose structures are unknown.
marker of exposure to such toxic metabo-
lites (79). Hydrogen peroxide at low con- 4.1. Photoaffinity Labeling with
centrations reacts with the heme group of Synthesizing Trifluoromethyl-
myoglobin and hemoglobin to form a cova- diazirinylphenyl Diazenes
lent heme–protein adduct. The modified
hemoprotein is then capable of generating Recently, Tschirret-Guth and Ortiz de
hydrogen peroxide in an essentially autocat- Montellano (119) have advanced the
alytic process. Thus, the covalent modifica- active-site labeling methodology developed
tion of the protein is an important marker in their laboratory. In previous studies, it
for oxidative damage and, more specifically, was noted that reaction of arylhydrazines
in the case of myoglobin, a sensitive marker and aryldiazenes with hemoproteins gave
for ischemic reperfusion injury. an iron–aryl complex (76). Early work uti-
Utilizing the chemiluminescence proce- lized the fact that, upon oxidation, the aryl
dure outlined above, Vuletich and Osawa complex shifted to form a mixture of N-
have modified the reaction conditions to arylprotoporphyrin-IX regioisomers. The
specifically select for protein bound heme ratio of the 4 regioisomers represented a
adducts on SDS-PAGE gels (125). The use low-resolution topology of the active-site,
of DTT is largely excluded as a reducing but gave no information on the specific
agent as it degrades the heme and, hence, amino acid residues within the active-site.
reduces the peroxidase activity of the heme The recent synthesis of azidoaryldiazenes as
adduct. Replacement of DTT with tris(2- photoaffinity probes for the active-site of
carboxyethyl) phosphine (TCEP), which hemoproteins has addressed this question
does not degrade the heme, allows detec- (119). The formation of the iron–aryl
tion of the peroxidase activity of the heme complex with the heme and subsequent
adduct with no contamination from non- formation of the activated nitrene by oxi-
covalent heme-containing proteins, as the dation limits the interaction to amino acid
noncovalent heme is lost on denaturation residues within the active site of the hemo-
and electrophoresis. protein (Figure 1a). The reaction of azido-
phenyldiazene with myoglobin was specific
for the distal histidine (His-64), suggesting
4. HEMOPROTEIN ACTIVE-SITE the probe preferentially selected for nucle-
PROBES ophilic residues (117). To develop pho-
toaffinity probes that were capable of label-
The lack of any available crystal struc- ing multiple residues, the authors replaced
172
Analysis of Heme and Hemoproteins
the azido function with the more reactive electronic nature of the heme. In addition
trifluoromethyl group by synthesizing 3- to providing valuable information on the
and 4-(trifluoromethyldiazirinyl)phenyl ligation and coordination state of the
diazene (118) (Figure 1b). Irradiation of heme, these techniques have aided greatly
the sample following formation of the in elucidating and confirming the mecha-
iron–diazirinylphenyl complex at 350 nm nism of action of the enzyme.
generates a highly reactive carbene capable HO-1 catalyzes the rate limiting step in
of inserting into unactivated C-H bonds. the conversion of heme to biliverdin (Fig-
The reaction of sperm whale myoglobin ure 2), and although not a hemoprotein on
with the activated carbene and subsequent reconstitution with heme, the enzyme
analysis of the tryptic peptides by mass forms a 1:1 ferric heme:HO-1 (heme:HO-
spectrophotometric techniques identified 1) complex with electronic spectra similar
at least 4 residues within the heme-binding to that of myoglobin (128). At pH 6.0, the
site of the protein, Leu-29, Val-68, His-64, heme:HO-1 complex has a Soret at 404
and Ile-107. With the exception of His-64, nm and a charge transfer band at 632 nm,
which was previously identified in the ear- characteristic of a high spin species (Figure
lier azidophenyldiazene work (117), the 3 3b). On increasing the pH from 6.0 to
remaining residues were nonpolar and less 10.0, the heme Soret shifts from 404 to
than 3 to 4 Å from the heme. 409 nm, and the charge transfer band at
630 nm decreases with the appearance of
α/β bands at 539 and 574 nm, a process
5. SPECTROPHOTOMETRIC attributed to a pH-dependent high-spin to
ANALYSIS OF HEMOPROTEINS low-spin conversion of a water ligand to a
hydroxide (Figure 3b) (40,111). The
The unique chromophore of hemopro- UV/visible spectra of the reduced complex
teins allows the use of a number of spec- in the presence of CO has a Soret at 420
trophotometric techniques to yield detailed nm and α/β bands at 567 and 537 nm,
information on the nature of the heme lig- consistent with a histidine proximal ligand.
and. Although individual spectroscopy Although the UV/visible spectrum can
techniques can yield valuable information, provide some limited information on the
it is the combined use of techniques such proximal ligand of a hemoprotein of
as UV/visible, magnetic circular dichroism unknown structure, it is not conclusive.
(MCD), resonance Raman, and NMR that MCD is a difference technique that will
can provide powerful insight as to the have both positive and negative compo-
nature of the coordination, ligation, and nents yielding a more detailed spectrum
oxidation state of the heme. than the corresponding UV/visible spec-
It will not be the focus of this review to trum. The increased sensitivity of MCD
give a detailed account of each technique, has been shown to be an effective “finger-
but present a brief synopsis of the informa- printing tool” for the determination of the
tion that can be gained as it relates to ligand coordination, oxidation state, and
hemoprotein structure and function. On spin state of hemoproteins (15,123).
studies of the human heme oxygenase Assignment of the proximal ligand in a
(HO-1), we have previously used a combi- hemoprotein of unknown structure by
nation of spectrophotometric techniques MCD, like UV/visible spectroscopy and
such as UV/visible, resonance Raman, resonance Raman (see below), is dependent
MCD, and NMR spectrophotometry to on comparison to a protein whose proxi-
determine the ligation, coordination, and mal ligand is known, such as myoglobin
173
A. Wilks
(histidine), catalase (tyrosine), and cyto- dine ligation in the heme:HO-1 complex
chrome P-450 (cysteine). The sample to be (Figure 3a) (40). The derivative bands of
analyzed is prepared with a known sixth the heme:HO-1 complex are blue-shifted
ligand such as cyanide, which together when compared to myoglobin, but are still
with the 4 pyrroles can then be compared very similar in structure (Figure 3b) (40).
to the ferric cyano complex of hemopro- Raman spectroscopy is a vibrational
teins, whose proximal ligands are known. spectroscopy technique that can detect
The electronic absorption and MCD spec- transitions between different ligation states
tra of the cyanoferric heme:HO-1 complex of porphyrins as the spin state is changed
is very similar to that of cyanoferric myo- (107,109). The change in spin state alters
globin, thus confirming the proximal histi- the size of the iron and its displacement
174
Analysis of Heme and Hemoproteins
from the heme plane which directly effects corresponding to the ferric 6-coordinate
the vibrational frequencies of the por- low-spin species (6CLS) at 1503 (υ3),
phyrin ring bond stretches. This technique 1582 (υ2), and 1638 (υ10) increasing (Fig-
can be used to characterize the spin state ure 4). This again supports the ionization
and coordination state in both the ferric of a coordinating water molecule to a
and ferrous oxidation states of a given hydroxide above pH 8.0.
hemoprotein (108). As in the MCD exper- Resonance Raman studies of the heme:
iments, the resonance Raman spectra of HO-1 complex provided direct confirma-
the heme:HO-1 complex show a marked tion of histidine as the proximal ligand
change in the heme axial coordination and (111). The iron–imidazole or iron–thiolate
spin state on changing pH (111). At pH ligand is not generally detectable due to the
6.0, the spectrum shows the heme to be weak coupling of the π electronic system.
primarily 6-coordinate high-spin (6CHS) However, in the case of the iron–imidazole
similar to myoglobin, in which the heme is (Fe-His) ligation, the reduced ferrous 5-
coordinated through a proximal histidine coordinate species shows significant en-
with water bound as the sixth ligand. hancement due to the iron being out of the
Increasing the pH to 8.0 significantly alters plane of the porphyrin. This transition
the Raman spectrum, with the bands at provides a characteristic marker for the Fe-
1483 (υ3) and 1565 cm-1 (υ2) of the ferric His stretching mode in the low frequency
6CHS species decreasing, and the bands ferrous spectra. Interestingly, the position
Figure 2. Conversion of
heme to biliverdin by
HO-1. The abbreviations
are as follows; Me,
methyl; V, vinyl; Pr, pro-
pionic acid.
175
A. Wilks
176
Analysis of Heme and Hemoproteins
pound II in peroxidase and shown that this trophilic mechanism is consistent with the
species is not an intermediate in the nor- presence of a proximal ligand that is nei-
mal physiological reaction of heme oxyge- ther ionized or hydrogen-bonded.
nase (128). Subsequent studies with ethyl- Isotopic labeling and 2-dimensional
hydroperoxide showed that the reaction NMR studies on the rat HO-1 have
proceeded by the normal pathway with the revealed an unusual heme electronic struc-
formation of α-meso-ethoxyheme, analo- ture (43). The proton contact shift patterns
gous to the formation of α-meso-hydroxy- of the heme reveal large differences in the
heme in the enzymatic conversion of heme delocalized spin density in which the spin
to biliverdin (130). The ethylhydroperox- density is highest at positions 3 (methyl)
ide reaction, therefore, favors an electro- and 2 (vinyl) and much smaller at posi-
philic mechanism with the addition of the tions 1 (methyl) and 4 (vinyl) (Figure 7).
terminal ethoxy or hydroxy (Fe-O-OH) The asymmetric spin density within a pyr-
moiety to the heme (Figure 6). The addi- role is unprecedented in any low-spin ferric
tion of the terminal oxygen of the ferric hemoprotein, where it is more common to
peroxo anion (Fe-O-O-) to the α-meso- find asymmetry between different pyrroles.
edge of the heme is ruled out as the termi- Based on studies with model hemes, the
nal oxygen of the ethylhydroperoxide distribution of the spin density suggests a
species is blocked. The proposed elec- direct electronic perturbation of the heme
177
A. Wilks
178
Analysis of Heme and Hemoproteins
is with great anticipation that we look for- tral analysis program; IEF, isoelectric focus-
ward to the discovery of new and unique ing; H2O2, hydrogen peroxide; LDS, lithi-
hemoproteins and the critical role they may um dodecyl sulfate; MCD, magnetic circu-
play in molecular and cellular processes. lar dichroism; NMR, nuclear magnetic
resonance; PAGE, polyacrylamide gel elec-
trophoresis; PCA, principal component
ABBREVIATIONS analysis; PMSF, phenylmethylsulfonyl flu-
oride; PVDF, polyvinylene difluoride
DTT, dithiothreitol; EDTA, ethylene membrane; SDS, sodium dodecyl sulfate;
diamine tetra acetic acid; HO-1, human SVD; singular value decomposition analy-
heme oxygenase isozyme 1; HRP, horserad- sis; TCEP, tris(2-carboxyethyl) phosphine;
ish peroxidase; HSAP, hemoprotein spec- TMBZ, 3,3′,5,5′-tetramethylbenzidine.
Figure 6. Electrophilic oxidation of the porphyrin macrocycle by the ferric peroxide complexes.
179
A. Wilks
ACKNOWLEDGMENTS 14.Cornejo, J., R.D. Willows, and S.I. Beale. 1998. Phy-
tobilin biosynthesis: cloning and expression of a gene
encoding soluble ferredoxin-dependent heme oxyge-
I would like to thank Paul Ortiz de nase from Synechocystis sp. PCC 6803. Plant J. 15:99-
Montellano for his continuous support and 107.
encouragement over the years. I also thank 15.Dawson, J.H. and D.M. Dooley. 1989. Iron Por-
phyrins. Part 3, p. 1-135. In A.P.B. Lever and H.B.
the National Institutes of Health for their Gray (Eds.). VCH publishers, New York.
financial support. 16.Dawson, J.H., S. Kadkhodayan, C. Zhuang, and M.
Sono. 1992. On the use of iron octa-alkylporphyrins
as models for protoporphyrin IX-containing heme sys-
tems in studies employing magnetic circular dichroism
REFERENCES spectroscopy. J. Inorg. Biochem. 45:179-192.
17.DePillis, G.D., S. Ozaki, J.M. Kuo, D.A. Maltby, and
1.Agalidis, I., E. Jauneau, and F. Reiss-Husson. 1974. P.R. Ortiz de Montellano. 1997. Autocatalytic pro-
Influence of iron deficiency on the cytochromes of cessing of heme by lactoperoxidase produces the native
Rhodopseudomonas spheroides. Eur. J. Biochem. 47:573- protein-bound prosthetic group. J. Biol. Chem.
580. 272:8857-8860.
2.Aono, S., H. Nakajima, K. Saito, and M. Okada. 18.Dierks, E.A., Z. Zhang, E.F. Johnson, and P.R. de
1996. A novel heme protein that acts as a carbon Montellano. 1998. The catalytic site of cytochrome
monoxide-dependent transcriptional activator in Rho- P4504A11 (CYP4A11) and its L131F mutant. J. Biol.
dospirillum rubrum. Biochem. Biophys. Res. Commun. Chem. 273:23055-23061.
228:752-756. 19.Dorward, D.W. 1993. Detection and quantitation of
3.Barnes, H.J., M.P. Arlotto, and M.R. Waterman. heme-containing proteins by chemiluminescence.
1991. Expression and enzymatic activity of recombi- Anal. Biochem. 209:219-223.
nant cytochrome P450 17 alpha-hydroxylase in 20.Durante, W., M.H. Kroll, N. Christodoulides, K.J.
Escherichia coli. Proc. Natl. Acad. Sci. USA 88:5597- Peyton, and A.I. Schafer. 1997. Nitric oxide induces
5601. heme oxygenase-1 gene expression and carbon monox-
4.Bartsch, R.G. 1971. Cytochromes: bacterial. Methods ide production in vascular smooth muscle cells. Circ.
Enzymol. 23:344-363. Res. 80:557-564.
5.Beck von Bodman, S., M.A. Schuler, D.R. Jollie, and 21.Dutta, C. and H.L. Henry. 1990. Detection of hemo-
S.G. Sligar. 1986. Synthesis, bacterial expression, and protein peroxidase activity on polyvinylidene difluo-
mutagenesis of the gene coding for mammalian cyto- ride membrane. Anal. Biochem. 184:96-99.
chrome b5. Proc. Natl. Acad. Sci. USA 83:9443-9447. 22.Elkins, C., C.J. Chen, and C.E. Thomas. 1995. Char-
6.Black, S.M. and P.R. Ortiz de Montellano. 1995. acterization of the hgbA locus encoding a hemoglobin
Characterization of rat neuronal nitric oxide synthase receptor from Haemophilus ducreyi. Infect. Immun.
expressed in Saccharomyces cerevisiae. DNA Cell Biol. 63:2194-2200.
14:789-794. 23.Estabrook, R.W., J. Peterson, J. Barn, and A. Hilde-
7.Boyle, R.C., A.L. Tappel, A.A. Tappel, H. Chen, and brant. 1972. The spectrophotometric measurement of
H.J. Andersen. 1994. Quantitation of heme proteins turbid suspensions of cytochromes associated with
from mixtures. J. Agric. Food Chem. 42:100-104. drug metabolism, p. 303-350. In C.F. Chignell (Ed.),
8.Braun, V., K. Hantke, and W. Koster. 1998. Bacterial Methods in Pharmacology. Vol. 2. Appleton-Century-
iron transport: mechanisms, genetics, and regulation. Crofts, New York.
Met. Ions Biol. Syst. 35:67-145. 24.Falk, J.E. 1963. Part A. Pyrrole pigments: chemistry
9.Champiat, D., A. Roux, O. Lhomme, and G. Nosen- and biochemistry of porphyrins and metallopor-
zo. 1994. Biochemiluminescence and biomedical phyrins, p. 3-33. In M. Florkin and E.H. Stotz (Eds.),
applications. Cell Biol. Toxicol. 10:345-351. Comprehensive Biochemistry, Vol. 9. Elsevier, Amster-
10.Chance, B. 1954. Spectrophotometry of intracellular dam.
respiratory pigments. Science 120:767. 25.Falk, J.E. 1964. Porphyrins and Metalloporphyrins.
11.Charles, I.G., A. Chubb, R. Gill, J. Clare, P.N. Lowe, Elsevier, New York.
L.S. Holmes, M. Page, J.G. Keeling, S. Moncada, and 26.Fishel, L.A., J.E. Villafranca, J.M. Mauro, and J.
V. Riveros-Moreno. 1993. Cloning and expression of a Kraut. 1987. Yeast cytochrome c peroxidase: mutagen-
rat neuronal nitric oxide synthase coding sequence in a esis and expression in Escherichia coli show tryptophan-
baculovirus/insect cell system. Biochem. Biophys. Res. 51 is not the radical site in compound I. Biochemistry
Commun. 196:1481-1489. 26:351-360.
12.Charles, I.G., C.A. Scorer, M.A. Moro, C. Fernandez, 27.Fronticelli, C., J.K. O’Donnell, and W.S. Brinigar.
A. Chubb, J. Dawson, N. Foxwell, R.G. Knowles, and 1991. Recombinant human hemoglobin: expression
S.A. Baylis. 1996. Expression of human nitric oxide and refolding of beta-globin from Escherichia coli. J.
synthase isozymes. Methods Enzymol. 268:449-460. Prot. Chem. 10:495-501.
13.Chen, H., A.L. Tappel, and R.C. Boyle. 1993. Oxida- 28.Gegner, J.A. and F.W. Dahlquist. 1991. Signal trans-
tion of heme proteins as a measure of oxidative damage duction in bacteria: CheW forms a reversible complex
to liver tissue slices. Free Radic. Biol. Med. 14:509- with the protein kinase CheA. Proc. Natl. Acad Sci.
517. USA 88:750-754.
180
Analysis of Heme and Hemoproteins
29.George, H.J., D.E. Van Dyk, R.A. Straney, J.M. Trza- 43.Hernandez, G., A. Wilks, R. Paolesse, K.M. Smith,
skos, and R.A. Copeland. 1996. Expression purifica- P.R. Ortiz de Montellano, and G.N. La Mar. 1994.
tion and characterization of recombinant human Proton NMR investigation of substrate-bound heme
inducible prostaglandin G/H synthase from bac- oxygenase: evidence for electronic and steric contribu-
ulovirus-infected insect cells. Protein Expr. Purif. 7:19- tions to stereoselective heme cleavage. Biochemistry
26. 33:6631-6641.
30.Gerber, N.C., C.R. Nishida, and P.R. Ortiz de Mon- 44.Hoffman, S.J., D.L. Looker, J.M. Roehrich, P.E.
tellano. 1997. Characterization of human liver indu- Cozart, S.L. Durfee, J.L. Tedesco, and G.L. Stetler.
cible nitric oxide synthase expressed in Escherichia coli. 1990. Expression of fully functional tetrameric human
Arch. Biochem. Biophys. 343:249-253. hemoglobin in Escherichia coli. Proc. Natl. Acad. Sci.
31.Gerber, N.C. and P.R. Ortiz de Montellano. 1995. USA 87:8521-8525.
Neuronal nitric oxide synthase. Expression in Escheri- 45.Hofrichter, J., J.H. Sommer, E.R. Herry, and W.A.
chia coli, irreversible inhibition by phenyldiazene, and Eaton. 1983. Nanosecond absorption spectroscopy of
active site topology. J. Biol. Chem. 270:17791-17796. hemoglobin. Proc. Natl. Acad. Sci. USA 80:2235-
32.Gilles-Gonzalez, M.A., G. Gonzalez, M.F. Perutz, L. 2239.
Kiger, M.C. Marden, and C. Poyart. 1994. Heme- 46.Holmans, P.L., M.S. Shet, C.A. Martin-Wixtrom,
based sensors, exemplified by the kinase FixL, are a C.W. Fisher, and R.W. Estabrook. 1994. The high-
new class of heme protein with distinctive ligand bind- level expression in Escherichia coli of the membrane-
ing and autoxidation. Biochemistry 33:8067-8073. bound form of human and rat cytochrome b5 and
33.Gonzalez, F.J. and K.R. Korzekwa. 1995. Cyto- studies on their mechanism of function. Arch. Bio-
chromes P450 expression systems. Annu. Rev. Phar- chem. Biophys. 312:554-565.
macol. Toxicol. 35:369-390. 47.Hui, H.L., J.S. Kavanaugh, M.L. Doyle, A. Wierz-
34.Guengerich, F.P. 1983. Oxidation-reduction properties ba, P.H. Rogers, A. Arnone, J.M. Holt, G.K. Ackers,
of rat liver cytochromes P-450 and NADPH-cyto- and R.W. Noble. 1999. Structural and functional
chrome p-450 reductase related to catalysis in reconsti- properties of human hemoglobins reassembled after
tuted systems. Biochemistry 22:2811-2820. synthesis in Escherichia coli. Biochemistry 38:1040-
35.Guengerich, F.P., N.A. Hosea, A. Parikh, L.C. Bell- 1049.
Parikh, W.W. Johnson, E.M. Gillam, and T. Shimada. 48.Ishikawa, K., M. Sato, and T. Yoshida. 1991. Expres-
1998. Twenty years of biochemistry of human P450s: sion of rat heme oxygenase in Escherichia coli as a cat-
purification, expression, mechanism, and relevance to alytically active, full-length form that binds to bacteri-
drugs. Drug Metab. Dispos. 26:1175-1178. al membranes. Eur. J. Biochem. 202:161-165.
36.Guengerich, F.P., M.V. Martin, Z. Guo, and Y.J. 49.Jennings, P.A., M.J. Stone, and P.E. Wright. 1995.
Chun. 1996. Purification of functional recombinant Overexpression of myoglobin and assignment of its
P450s from bacteria. Methods Enzymol. 272:35-44. amide, C alpha and C beta resonances. J. Biomol.
37.Halaka, F.G., G.T. Babcock, and J.L. Dye. 1985. The NMR. 6:271-276.
use of principal component analysis to resolve the spec- 50.Jones, S., M. Tasab, J.E. Ogden, D.J. Ballance, and
tra and kinetics of cytochrome c oxidase reduction by M.J. Powell. 1996. Expression of rat neuronal nitric
5, 10-dihydro-5-methyl phenazine. Biophys. J. oxide synthase in Saccharomyces cerevisiae. J. Biotech-
48:209-219. nol. 48:37-41.
38.Halpert, J.R., T.L. Domanski, O. Adali, C.P. Biagini, 51.Jung, C., O. Ristau, and H. Rein. 1991. The high-
J. Cosme, E.A. Dierks, E.F. Johnson, J.P. Jones et al. spin/low-spin equilibrium in cytochrome P-450—a
1998. Structure-function of cytochromes P450 and new method for determination of the high-spin con-
flavin-containing monooxygenases: implications for tent. Biochim. Biophys. Acta 1076:130-136.
drug metabolism. Drug Metab. Dispos. 26:1223-1231. 52.Kakuno, T., R.G. Bartsch, K. Nishikawa, and T.
39.Hartmann, C. and P.R. Ortiz de Montellano. 1992. Horio. 1981. Redox componenets associated with
Baculovirus expression and characterization of catalyt- chromatophores from Rhodospirillum rubrum. J.
ically active horseradish peroxidase. Arch. Biochem. Biochem. 70:79-94.
Biophys. 297:61-72. 53.Kincaid, J., P. Stein, and T.G. Spiro. 1979. Absence of
40.Hawkins, B.K., A. Wilks, L.S. Powers, P.R. Ortiz de heme-localized strain in T state hemoglobin: insensi-
Montellano, and J.H. Dawson. 1996. Ligation of the tivity of heme-imidazole resonance Raman frequencies
iron in the heme-heme oxygenase complex: X-ray to quaternary structure. Proc. Natl. Acad. Sci. USA
absorption, electronic absorption and magnetic circu- 76:549-552.
lar dichroism studies. Biochim. Biophys. Acta 54.King, M.T. and G. Drews. 1975. The respiratory elec-
1295:165-173. tron transport system of heterotrophically-grown Rho-
41.Hernan, R.A., H.L. Hui, M.E. Andracki, R.W. Noble, dopseudomonas palustris. Arch. Microbiol. 102:219-
S.G. Sligar, J.A. Walder, and R.Y. Walder. 1992. 231.
Human hemoglobin expression in Escherichia coli: 55.Kitagawa, T., K. Nagai, and M. Tsubaki. 1979.
importance of optimal codon usage. Biochemistry Assignment of the Fe-Nepsilon (His F8) stretching
31:8619-8628. band in the resonance Raman spectra of deoxy myo-
42.Hernan, R.A. and S.G. Sligar. 1995. Tetrameric hemo- globin. FEBS Lett. 104:376-378.
globin expressed in Escherichia coli. Evidence of hetero- 56.Knaff, D.B. and B.B. Buchanan. 1975. Cytochrome b
geneous subunit assembly. J. Biol. Chem. 270:26257- and photosynthetic sulfur bacteria. Biochim. Biophys.
26264. Acta. 376:549-560.
181
A. Wilks
57.Koenigs, L.L. and W.F. Trager. 1998. Mechanism- 73.Nakane, M., J.S. Pollock, V. Klinghofer, F. Basha, P.A.
based inactivation of cytochrome P450 2B1 by 8- Marsden, A. Hokari, T. Ogura, H. Esumi, and G.W.
methoxypsoralen and several other furanocoumarins. Carter. 1995. Functional expression of three isoforms
Biochemistry 37:13184-13193. of human nitric oxide synthase in baculovirus-infected
58.Koenigs, L.L. and W.F. Trager. 1998. Mechanism- insect cells. Biochem. Biophys. Res. Commun.
based inactivation of P450 2A6 by furanocoumarins. 206:511-517.
Biochemistry 37:10047-10061. 74.North, J.A., R. Dietrich, and A.L. Tappel. 1996. Mul-
59.Laemmli, U.K. 1970. Cleavage of structural proteins ticomponent analysis of heme protein spectra in bio-
during the assembly of the head of bacteriophage T4. logical materials. Anal. Biochem. 233:115-123.
Nature 227:680-685. 75.Ohnishi, T., S. Miura, and Y. Ichikawa. 1993. Pho-
60.Lee, B.C. 1992. Isolation of an outer membrane he- toaffinity labeling of cytochrome P-45011 beta with
min-binding protein of Haemophilus influenzae type b. methyltrienolone as a probe for the substrate binding
Infect. Immun. 60:810-816. region. Biochim. Biophys. Acta 1161:257-264.
61.Lindenmayer, A. and R.W. Estabrook. 1958. Low 76.Ortiz de Montellano, P.R. 1995. Arylhydrazines as
temperature spectral studies on the biosynthesis of probes of hemoprotein structure and function. Bio-
cytochromes in bakers yeast. Arch. Biochem. Biophy. chimie 77:581-593.
78:66-82. 77.Ortiz de Montellano, P.R., C. Nishida, I. Rodriguez-
62.List, B.M., P. Klatt, E.R. Werner, K. Schmidt, and B. Crespo, and N. Gerber. 1998. Nitric oxide synthase
Mayer. 1996. Overexpression of neuronal nitric oxide structure and electron transfer. Drug Metab. Dispos.
synthase in insect cells reveals requirement of haem for 26:1185-1189.
tetrahydrobiopterin binding. Biochem. J. 315:57-63. 78.Ortiz de Montellano, P.R. and A. Wilks. 2000. Struc-
63.Maines, M.D. 1997. The heme oxygenase system: a ture and mechanism of heme oxygenase. Adv. Inorg.
regulator of second messenger gases. Annu. Rev. Phar- Chem. 51:359-402.
macol. Toxicol. 35:517-554. 79.Osawa, Y., K. Nakatsuka, M.S. Williams, J.T. Kindt,
64.Mayer, B., P. Klatt, B.M. List, C. Harteneck, and K. and M. Nakatsuka. 1996. Reactions of reactive me-
Schmidt. 1996. Large-scale purification of rat brain tabolites with hemoproteins—toxicological implica-
nitric oxide synthase from baculovirus overexpression tions: covalent alteration of hemoproteins. Adv. Exp.
system. Methods Enzymol. 268:420-427. Med. Biol. 387:37-45.
65.McMillan, K., D.S. Bredt, D.J. Hirsch, S.H. Snyder, 80.Otto, B.R., S.J. van Dooren, J.H. Nuijens, J. Luirink,
J.E. Clark, and B.S. Masters. 1992. Cloned, expressed and B. Oudega. 1998. Characterization of a hemoglo-
rat cerebellar nitric oxide synthase contains stoichio- bin protease secreted by the pathogenic Escherichia coli
metric amounts of heme, which binds carbon monox- strain EB1. J. Exp. Med. 188:1091-1103.
ide. Proc. Natl. Acad. Sci. USA 89:11141-11145. 81.Pollock, W.B., F.I. Rosell, M.B. Twitchett, M.E.
66.McMillan, K., J.C. Salerno, and B.S. Masters. 1996. Dumont, and A.G. Mauk. 1998. Bacterial expression
Nitric oxide synthases: analogies to cytochrome P450 of a mitochondrial cytochrome c. Trimethylation of
monooxygenases and characterization of recombinant lys72 in yeast iso-1-cytochrome c and the alkaline con-
rat neuronal nitric oxide synthase hemoprotein. Meth- formational transition. Biochemistry 37:6124-6131.
ods Enzymol. 268:460-472. 82.Poole, R.K., R.I. Scott, and B. Chance. 1980. Low-
67.Miller, J.P. and R.E. White. 1994. Photoaffinity label- temperature spectral and kinetic properties of
ing of cytochrome P450 2B4: capture of active site cytochromes in Escherichia coli K-12 grown at lowered
heme ligands by a photocarbene. Biochemistry oxygen tension. Biochim. Biophys. Acta 591:471-482.
33:807-817. 83.Poulos, T.L. 1991. Modeling of mammalian P450s on
68.Miyake, Y. and N. Takayama. 1976. Spectral interme- basis of P450cam X-ray structure. Methods Enzymol.
diates during the reduction of hepatic microsomal 206:11-30.
cytochrome P-450. J. Biochem. (Tokyo) 79:1077- 84.Poulos, T.L. 1993. Peroxidases. Curr. Opin. Biotech-
1087. nol. 4:484-489.
69.Modi, S., M.J. Paine, M.J. Sutcliffe, L.Y. Lian, W.U. 85.Ramos, L.S., K.R. Beebe, W.P. Karey, M.E. Sanches,
Primrose, C.R. Wolf, and G.C. Roberts. 1996. A B.C. Erickson, B.E. Wilson, B.E. Wangen, and B.R.
model for human cytochrome P450 2D6 based on Kowalski. 1986. Chemometrics. Anal. Chem. 58:
homology modeling and NMR studies of substrate 294R-515R.
binding. Biochemistry 35:4540-4550. 86.Renaud, J.P., D.R. Davydov, K.P. Heirwegh, D. Man-
70.Mok, T.C., P.A. Rickard, and F.J. Moss. 1969. The suy, and G.H. Hui Bon Hoa. 1996. Thermodynamic
carbon monoxide-reactive haemoproteins of yeast. studies of substrate binding and spin transitions in
Biochim. Biophys. Acta 172:438-449. human cytochrome P-450 3A4 expressed in yeast
71.Muchmore, D.C., L.P. McIntosh, C.B. Russell, D.E. microsomes. Biochem. J. 319:675-681.
Anderson, and F.W. Dahlquist. 1989. Expression and 87.Rieske, J.S. 1967. The quantitative determination of
nitrogen-15 labeling of proteins for proton and nitro- mitochondrial hemoproteins. Methods Enzymol.
gen-15 nuclear magnetic resonance. Methods Enzy- 10:488-493.
mol. 177:44-73. 88.Ristau, O., H. Rein, S. Greschner, G.R. Janig, and K.
72.Nagai, K. and H.C. Thogersen. 1984. Generation of Ruckpaul. 1979. Quantitative analysis of the spin
beta-globin by sequence-specific proteolysis of a hybrid equilibrium of cytochrome P-450 LM2 fraction from
protein produced in Escherichia coli. Nature 309:810- rabbit liver microsomes. Acta Biol. Med. Ger. 38:177-
812. 185.
182
Analysis of Heme and Hemoproteins
89.Rivera, M., C. Barillas-Mury, K.A. Christensen, J.W. 103.Shu, F., V. Ramakrishnan, and B.P. Schoenborn.
Little, M.A. Wells, and F.A. Walker. 1992. Gene syn- 1996. High-level expression and deuteration of sperm
thesis, bacterial expression, and 1H NMR spectroscop- whale myoglobin. A study of its solvent structure by
ic studies of the rat outer mitochondrial membrane X-ray and neutron diffraction methods. Basic Life
cytochrome b5. Biochemistry 31:12233-12240. Sci. 64:309-323.
90.Riveros-Moreno, V., B. Heffernan, B. Torres, A. 104.Smith, A.T., N. Santama, S. Dacey, M. Edwards,
Chubb, I. Charles, and S. Moncada. 1995. Purifica- R.C. Bray, R.N. Thorneley, and J.F. Burke. 1990.
tion to homogeneity and characterisation of rat brain Expression of a synthetic gene for horseradish peroxi-
recombinant nitric oxide synthase. Eur. J. Biochem. dase C in Escherichia coli and folding and activation of
230:52-57. the recombinant enzyme with Ca2+ and heme. J. Biol.
91.Rodgers, K.R. 1999. Heme-based sensors in biological Chem. 265:13335-13343.
systems [In Process Citation]. Curr. Opin. Chem. Biol. 105.Smulevich, G., J.M. Mauro, L.A. Fishel, A.M. Eng-
3:158-167. lish, J. Kraut, and T.G. Spiro. 1988. Heme pocket
92.Rodriguez-Crespo, I., N.C. Gerber, and P.R. Ortiz de interactions in cytochrome c peroxidase studied by
Montellano. 1996. Endothelial nitric-oxide synthase. site-directed mutagenesis and resonance Raman spec-
Expression in Escherichia coli, spectroscopic characteri- troscopy. Biochemistry 27:5477-5485.
zation, and role of tetrahydrobiopterin in dimer for- 106.Smulevich, G., M.A. Miller, J. Kraut, and T.G.
mation. J. Biol. Chem. 271:11462-11467. Spiro. 1991. Conformational change and histidine
93.Roman, L.J., E.A. Sheta, P. Martasek, S.S. Gross, Q. control of heme chemistry in cytochrome c peroxidase:
Liu, and B.S. Masters. 1995. High-level expression of resonance Raman evidence from Leu-52 and Gly-181
functional rat neuronal nitric oxide synthase in Escherichia mutants of cytochrome c peroxidase. Biochemistry
coli. Proc. Natl. Acad. Sci. USA 92:8428-8432. 30:9546-9558.
94.Sari, M.A., S. Booker, M. Jaouen, S. Vadon, J.I. 107.Spiro, T.G. 1978. Resonance Raman spectra of
Boucher, D. Pompon, and D. Mansuy. 1996. Expres- hemoproteins. Methods Enzymol. 54:233-249.
sion in yeast and purification of functional macrophage 108.Spiro, T.G. 1985. Resonance Raman spectroscopy as
nitric oxide synthase. Evidence for cysteine-194 as iron a probe of heme protein structure and dynamics. Adv.
proximal ligand. Biochemistry 35:7204-7213. Prot. Chem. 37:111-159.
95.Sarma, S., R.J. DiGate, D.L. Banville, and R.D. 109.Spiro, T.G. and R.S. Czernuszewicz. 1995. Reso-
Guiles. 1996. 1H, 13C and 15N NMR assignments nance Raman spectroscopy of metalloproteins. Meth-
and secondary structure of the paramagnetic form of ods Enzymol. 246:416-460.
rat cytochrome b5. J. Biomol. NMR 8:171-183. 110.Springer, B.A. and S.G. Sligar. 1987. High-level
96.Schmidt, M.L. and J.Q. Trojanowski. 1986. Enzy- expression of sperm whale myoglobin in Escherichia
matic detection of native and derivatized horseradish coli. Proc. Natl. Acad. Sci. USA 84:8961-8965.
peroxidase in sodium dodecyl sulfate polyacrylamide 111.Sun, J., A. Wilks, P.R. Ortiz de Montellano, and
gels. Anal. Biochem. 155:371-375. T.M. Loehr. 1993. Resonance Raman and EPR spec-
97.Schuller, D.J., A. Wilks, P.R. Ortiz de Montellano, troscopic studies on heme-heme oxygenase complexes.
and T.L. Poulos. 1999. Crystal structure of human Biochemistry 32:14151-14157.
heme oxygenase-1. Nat. Struct. Biol. 6:860-867. 112.Tamburini, P.P., G.G. Gibson, W.L. Backes, S.G.
98.Seo, H.G., J. Fujii, H. Soejima, N. Niikawa, and N. Sligar, and J.B. Schenkman. 1984. Reduction kinetics
Taniguchi. 1995. Heme requirement for production of of purified rat liver cytochrome P-450. Evidence for a
active endothelial nitric oxide synthase in baculovirus- sequential reaction mechanism dependent on the
infected insect cells. Biochem. Biophys. Res. Com- hemoprotein spin state. Biochemistry 23:4526-4533.
mun. 208:10-18. 113.Taylor, K.L., D.J. Uhlinger, and J.M. Kinkade, Jr.
99.Shelver, D., M.V. Thorsteinsson, R.L. Kerby, S.Y. 1992. Expression of recombinant myeloperoxidase
Chung, G.P. Roberts, M.F. Reynolds, R.B. Parks, and using a baculovirus expression system. Biochem. Bio-
J.N. Burstyn. 1999. Identification of two important phys. Res. Commun. 187:1572-1578.
heme site residues (cysteine 75 and histidine 77) in 114.Teale, F.W.J. 1959. Cleavage of the haem-protein link
CooA, the CO-sensing transcription factor of Rho- by acid methyl ethyl ketone. Biochim. Biophy. Acta.
dospirillum rubrum. Biochemistry 38:2669-2678. 35:543.
100.Shen, T.J., N.T. Ho, V. Simplaceanu, M. Zou, B.N. 115.Teraoka, J. and T. Kitagawa. 1981. Structural impli-
Green, M.F. Tam, and C. Ho. 1993. Production of cation of the heme-linked ionization of horseradish
unmodified human adult hemoglobin in Escherichia peroxidase probed by the Fe-histidine stretching
coli. Proc. Natl. Acad. Sci. USA 90:8108-8112. Raman line. J. Biol. Chem. 256:3969-3977.
101.Shioi, Y., K. Takamiya, and M. Nishimura. 1972. 116.Thomas, P.E., D. Ryan, and W. Levin. 1976. An
Studies on energy and electron transfer systems in improved staining procedure for the detection of the
green photosynthetic bacterium Chloropseudomonas peroxidase activity of cytochrome P-450 on sodium
ethylica strain 2K. I. Isolation and characterization of dodecyl sulfate polyacrylamide gels. Anal. Biochem.
cytochromes from Chloropseudomonas ethylica strain 75:168-176.
2K. J. Biochem. (Tokyo) 71:285-294. 117.Tschirret-Guth, R.A., K.F. Medzihradszky, and P.R.
102.Shrager, R.I. and R.W. Hendler. 1982. Titration of Ortiz de Montellano. 1998. Specific azidophenyl-
individual components in a mixture with resolution diazene hemoprotein active site probes. Cross-linking
of difference spectra, pK’s and redox transitions. Anal. of the heme to His-64 in myoglobin. J. Am. Chem.
Chem. 54:1147-1152. Soc. 120:7404-7410.
183
A. Wilks
118.Tschirret-Guth, R.A., K.F. Medzihradszky, and 128.Wilks, A. and P.R. Ortiz de Montellano. 1993. Rat
P.R. Ortiz de Montellano. 1999. Trifluoromethyl- liver heme oxygenase. High level expression of a trun-
diazirinylphenyldiazenes: new hemeprotein active-site cated soluble form and nature of the meso-hydroxy-
probes. J. Amer. Chem. Soc. 121:4731-4737. lating species. J. Biol. Chem. 268:22357-22362.
119.Tschirret-Guth, R.A. and P.R. Ortiz de Motellano. 129.Wilks, A. and M.P. Schmitt. 1998. Expression and
1998. Synthesis of photoaffinity probes for heme- characterization of a heme oxygenase (Hmu O) from
containing proteins. J. Org. Chem. 63:9711-9715. Corynebacterium diphtheriae. Iron acquisition requires
120.Tsutsui, K. 1986. Affinity chromatography of heme- oxidative cleavage of the heme macrocycle. J. Biol.
binding proteins: synthesis of hemin-agarose. Meth- Chem. 273:837-841.
ods Enzymol. 123:331-338. 130.Wilks, A., J. Torpey, and P.R. Ortiz de Montellano.
121.Tsutsui, K. and G.C. Mueller. 1982. A protein with 1994. Heme oxygenase (HO-1). Evidence for elec-
multiple heme-binding sites from rabbit serum. J. trophilic oxygen addition to the porphyrin ring in the
Biol. Chem. 257:3925-3931. formation of alpha-meso-hydroxyheme. J. Biol.
122.Varadarajan, R., A. Szabo, and S.G. Boxer. 1985. Chem. 269:29553-29556.
Cloning, expression in Escherichia coli, and reconsti- 131.Wu, C., J. Zhang, H. Abu-Soud, D.K. Ghosh, and
tution of human myoglobin. Proc. Natl. Acad. Sci. D.J. Stuehr. 1996. High-level expression of mouse
USA 82:5681-5684. inducible nitric oxide synthase in Escherichia coli
123.Vickery, L.E. 1978. Spin states of heme proteins by requires coexpression with calmodulin. Biochem.
magnetic circular dichroism. Methods Enzymol. 54: Biophys. Res. Commun. 222:439-444.
284-302. 132.Yoshida, T., K. Ishikawa, and M. Sato. 1991.
124.von Wachenfeldt, C., T.H. Richardson, J. Cosme, Degradation of heme by a soluble peptide of heme
and E.F. Johnson. 1997. Microsomal P450 2C3 is oxygenase obtained from rat liver microsomes by mild
expressed as a soluble dimer in Escherichia coli follow- trypsinization. Eur. J. Biochem. 199:729-733.
ing modification of its N-terminus. Arch. Biochem. 133.Yoshida, T. and G. Kikuchi. 1978. Purification and
Biophys. 339:107-114. properties of heme oxygenase from pig spleen micro-
125.Vuletich, J.L. and Y. Osawa. 1998. Chemilumines- somes. J. Biol. Chem. 253:4224-4229.
cence assay for oxidatively modified myoglobin. Anal. 134.Yun, C.H., G.J. Hammons, G. Jones, M.V. Mar-
Biochem. 265:375-380. tin, N.E. Hopkins, W.L. Alworth, and F.P.
126.Wilks, A., S.M. Black, W.L. Miller, and P.R. Ortiz Guengerich. 1992. Modification of cytochrome
de Montellano. 1995. Expression and characterization P450 1A2 enzymes by the mechanism-based inacti-
of truncated human heme oxygenase (hHO- 1) and a vator 2-ethynylnaphthalene and the photoaffinity
fusion protein of hHO-1 with human cytochrome label 4-azidobiphenyl. Biochemistry 31:10556-
P450 reductase. Biochemistry 34:4421-4427. 10563.
127.Wilks, A. and P.R. Ortiz de Montellano. 1992. Intra- 135.Zabel, U., M. Weeger, M. La, and H.H. Schmidt.
molecular translocation of the protein radical formed 1998. Human soluble guanylate cyclase: functional
in the reaction of recombinant sperm whale myoglobin expression and revised isoenzyme family. Biochem. J.
with H2O2. J. Biol. Chem. 267:8827-8833. 335:51-57.
184
Hemoproteins Purification and
8 Characterization by Using Aqueous
Two-Phase Systems
Daniel Forciniti
Chemical Engineering Department, University of Missouri-Rolla,
Rolla, MO, USA
185
Table 1. Some Examples of the Use of Aqueous Two-Phase Systems for the Protein Purification
186
D. Forciniti
Concentration
Substance Source Yield Factor/Purity Observations Reference
Thaumatin Cell homogenate of 92% 65% Two extraction steps using PEG/phosphate 5
E. coli systems.
Lactate dehydrogenase Porcine muscle extract 54% 11.8-fold PEG/Dx followed by affinity precipitation. 19
Ecdysone (hormone) Spinach 88% ND UCON 50-hb-5111/Reppal PES-200a. 33
3-phosphoglycerate kinase Yeast homogenate 57% 5.2-fold UCON 50-HB-5100/Dx T-500. 47
NAD-kinase Lactobacillus 100% 3-fold PEG/salt. 23
cellobiosus
Leucine dehydrogenase Bacillus sphaericus 98% 9.5-fold PEG/crude Dx (unfractionated Dx of high, 23
>1 000 000 molecular weight).
Beta interferon Human fibroblasts 350-fold 31
Beta interferon Recombinant E. coli 76% ND PEG/salt/PEG phosphate ester followed by 18
back extraction.
Fractionation of proteins Human serum ND ND PEG/Dx and PEG/Aquaphase PPTb plus addition 7
of various affinity ligands. Separation done by
counter-current distribution.
Lactate dehydrogenases Recombinant B. 80% 8-fold Cu(II)poly-VI-VCL and Dx-70. 30
stearothermophilus
Chymosin Recombinant A. ND ND PEG-8000/NA2SO4 Industrial scale production 27
awamori (Genencor Intrl.).
Recombinant Protein Mammalian cell 90% 100-fold PEG/Pluronic F68/Ammonium sulfate. Industrial 4
culture scale production (MedImmune, Gaithersburg,
MD, USA)
Cutinase Recombinant ND 26-fold PEG-1500/sodium phosphate (plus Triton 9
Saccharomyces X-100, or Tween 20, or butyrate).
cerevisiae
Vitreoscilla hemoglobin E. coli lysate 47% 95% Sequential extraction: (i) PEG/Dx, 21
(ii) PEG/Na2SO4, (iii) PEG/MgSO4
aReppal PES-200 is a hydroxypropyl starch produced by CarbamyLab (Kristianstad, Sweden).
bAquaphase PPT is a modified starch product produced by Perstop (Sweden).
ND: not determined by the authors.
Hemoproteins Purification and Characterization
PEG Dx 2,16
PEG Ficoll 2
Dx Ficoll 2
Methyl Cellulose Dx 2
PEG Benzoil Dextran 28
P(EO-PO) Water (NaCl) 47
P(EO-PO) Reppal (Hydroxypropyl starch) 20
PEG Sodium citrate/citric acid 48
PEG NaH2PO4/K2HPO4 2,43
PEG Ammonium Sulfate 2
Sodium Dextran Sulfate PEG (NaCl) 2
PEG NaCl 44
PEG SCNNa/NaCl 10
PEG GuHCl/NaCl 46
DX PVP 56
Dx PVA 56
PEG Polyvinyl methyl ether 56
Methyl Cellulose Hydroxypropyl Dextran 2
All systems contain a large amount of water (or buffer), up to 90% (wt/wt).
or affinity ligands is also included. The the top and bottom phases (VT/VB). The
derivatized species often constitute a minor working tie-line and the volume ratio
constituent (1%–5%) of the total system. between the top and the bottom phase can
The equilibrium data for aqueous 2- be chosen by using published equilibrium
phase systems are usually presented by plot- data (56). Since there are variations in the
ting the amount of one phase forming molecular weight of the polymers and,
species (A) versus the amount of the other sometimes, the experimental conditions
(B) in both top and bottom phases. The used to determine a binodal are not com-
locus of equilibrium compositions of top pletely specified, it may be appropriate to
and bottom phase in the A-B plane is called create specific a phase diagram if a particu-
the binodal (See Figure 1). Figure 1 shows lar system is going to be used routinely.
the phase diagram for a polyethylene The equilibrium curve or binodal is
propylene random copolymer P(EO- affected by:
PO)/salt system at 25°C. A few tie-lines are 1. Polymer molecular weight. The higher
also indicated. Simple geometrical argu- the molecular weight of the polymer
ments show that the length of the tie-line is used, the lower the polymer concentra-
equal to TLL = (∆A2 + ∆B2)1/2, where ∆A = tion (on a weight by weight basis)
[A]top - [A]bottom and ∆B = (B)bottom - needed to obtain phase separation. The
(B)top. The ratio between segments BO and binodal curve becomes more asymmet-
AO is equal to the volume ratio between rical as the molecular weight difference
188
Hemoproteins Purification and Characterization
[PEG] and [Dx] are the overall concentration of PEG and Dx (before phase separation). The
densities (ρ in units of g/cm3) of each phase were measured with U-shaped oscillator
densitometer, and the interfacial tensions were measured with a spinning drop tensiometer
(Krüss Site 04, Hamburg, Germany). All systems are at 25°C.
Figure 1. Equilibrium curve for a P(EO-PO)/NaCl system at 25°C. Points to the left the equilibrium curve correspond to one
phase, and points to the right correspond to 2 liquid phases at equilibrium Point B representing the composition of the top phase,
point A the composition of the bottom phase, and points O and P two possible overall concentrations of P(EO-PO) and NaCl.
An overall concentration equal to O yields VT/VB = 1, and an overall concentration equal to P yields VT/VB = 1/3.
189
D. Forciniti
4. Polydispersity of the polymers. Com- ent pHs, we must consider any change in
mercial grade polymers have a wide the physical properties of the proteins due
distribution of molecular weights. to changes in the pH and how these
Therefore, phase separation occurs in a changes may affect the protein partition
small range of polymer concentrations coefficient. For example, whereas lysozyme
changing the binodal curve to a bin- does not change its conformation over the
odal zone. If the polymer molecular pH range 1.2 to 11.3 in dilute salt solu-
weight distribution is too wide, more tions at moderate temperatures, it polymer-
than two phases can be formed. izes reversibly at pH above 5 (45). This
5. Ionic Strength. The length of the tie- change in the molecular weight of the pro-
line increases when salts are added to tein will change the partition coefficient.
PEG-salt systems. In PEG/Dx/water The molecular weight and concentration
systems the ionic strength does not of the phase-forming species also affect the
have an appreciable influence on the partition coefficient strongly (13). For
position of the binodal. example, in a PEG/Dx system, low PEG
molecular weight favors the partitioning of
1.1. Distribution of Proteins in Aqueous proteins into the PEG-rich phase. The
Two-Phase Systems effect of polymer concentration on the par-
tition coefficient is also well known. If K p
The separation power of a given aque- is smaller than 1, an increase in either one
ous 2-phase system is given by the partition of the phase-forming species decreases Kp.
coefficient, Kp, of a protein. Kp is defined Similarly, if Kp is larger than 1, an increase
as the ratio of the protein concentration in in either of the phase-forming species
the top and bottom phases. This partition increases Kp. The amount of the phase-
coefficient depends on the difference in forming species also affects the volume ratio
chemical potential of the protein between between the phases. For example, in Figure
top and bottom phases, and therefore, it is 1, a system of total composition O has a
a function of the chemical nature of the top:bottom volume ratio of 1, whereas a
polymers, the protein, added electrolytes, system of total composition P has a
and temperature. It is usually manipulated top:bottom volume ratio of 1/3. Therefore,
by changing the pH (13,14) or the temper- the targeted protein can be not only puri-
ature, by adding salts or affinity ligands, or fied, but also concentrated in a single step.
by changing the molecular weight of the Generally, the partition coefficient in-
polymers or their concentration. creases with increasing temperature in Dx/
Manipulation of the pH is of primary PEG systems between 4° and 40°C (12).
importance in partitioning studies of pro- The partition coefficient of small and
teins. It is convenient to split the partition hydrophilic proteins is only slightly affect-
coefficient into 2 contributions, one that is ed by changes in temperature, whereas the
independent of pH, and the other that is partition coefficient of bigger and more
proportional to the net charge of the pro- hydrophobic proteins is strongly affected
tein. In so doing, electrostatic and nonelec- by temperature changes. High tempera-
trostatic contributions to the partition coef- tures (around 40°C) may be used to mini-
ficient can be conveniently (but artificially) mize protein association, whereas low tem-
separated. This approach predicts that the peratures may be desirable to maintain
logarithm of the partition coefficient is a protein stability.
linear function of the charge of the protein. Salts have unequal affinities for the top
In the analysis of partitioning data at differ- and bottom phases of aqueous 2-phase sys-
190
Hemoproteins Purification and Characterization
tems (25). This uneven partition of salts dent on protein’s charge, Z is the charge of
between the 2 phases affects the chemical the protein, and γ depends on the types
potential of the protein in each phase and and concentration of polymers, tempera-
thus its partition coefficient. The following ture, and types of electrolytes. Values of γ
mental exercise helps to understand this are plotted in Figure 2 for various salts (val-
phenomenon. Consider 2 phases at equi- ues are for Dx-500/PEG-8000 at 4°C).
librium in which a salt has been previously The net effect of a salt can be calculated by
partitioned. The different affinity of the γ = γcation - γanion. For example, the addi-
salt for the bottom and top phases creates 2 tion of tetrabutyl ammonium phosphate
distinct ionic atmospheres in both phases. yields a γ = -79, whereas the addition of
Picture a charged protein molecule at infi- potassium perchlorate yields a γ = 30. This
nite distance from the phases. Bring the figure can be used to select the appropriate
protein and try to insert it in each of the 2 electrolyte to move the desired protein to
phases. Since the protein is going to see dif- either the top or bottom phases. A pH
ferent ionic atmospheres in both phases, away from the isoelectric point of the pro-
the work needed to insert it in one or the tein must be selected, since the partition
other phase will be different. This differ- coefficient of proteins at their isoelectric
ence in the work of insertion of a charged point is quite insensitive to salt type and
protein in each of the 2 phases at equilibri- concentration (see Equation 1).
um is proportional to the electrochemical The partition coefficient of proteins can
potential difference of the protein between be manipulated by adding an affinity lig-
them. Consequently, the partition coeffi- and or by using liquid–liquid chromatog-
cient, which is a function of that potential raphy, counter–current extraction, or by a
difference, is strongly affected by the type combination of the first two. Affinity lig-
and concentration of salts and by the ands, such as PEG-palmitate (42) or PEG-
charge on the protein. red (24), have been routinely used to
In general, the partition coefficient of improve the selectivity of aqueous 2-phase
proteins away from their isoelectric point extraction. In affinity partitioning (26), it
depends on both the type and concentra- is customary to define ∆ln Kp = ln KL - ln
tion of cation and anion (25). For example, K0 (KL and K0 are the partition coeffi-
for positively charged proteins the partition cients of the protein with and without
coefficient in PEG/Dx systems is higher in added affinity ligand, respectively) to
potassium chloride than in potassium quantify the enhancement in partitioning
phosphate; the reverse is true for negatively as a result of the addition of the affinity lig-
charged proteins. The effect of the cation and. Liquid–liquid chromatography, with
on the partition coefficient of positively or without the addition of an affinity lig-
charged proteins is K approximately equal and, can be used to improve the selectivity
to Cs greater than Na greater than Li, of aqueous 2-phase systems. For example,
whereas the reverse order holds for nega- we have used liquid–liquid chromatogra-
tively charged proteins. It is possible to cor- phy with an immobilized Dx-rich phase to
relate the partition coefficient of a protein purify formate dehydrogenase (FDH) from
with its charge and with the type of salt. Candida boinidi with and without the
Johansson (25) found that, addition of PEG-red as an affinity ligand
(50). Another attractive alternative tech-
log Kp = log K0 + γZ [Eq. 1]
nique is the use of metal affinity aqueous
where, K0 depends on the particular pro- 2-phase extraction (1,40,55). In this tech-
tein and phase-system, but it is indepen- nique, a small portion (less than 1%) of the
191
D. Forciniti
total PEG is replaced by PEG-iminodi- forward. Two sets of 2-phase systems (usu-
acetic acid (PEG-IDA), which chelates di- ally Dx/PEG) covering a wide pH range
valent cations like copper or zinc. Histidine are prepared. One set contains NaCl,
groups on the protein surface recognize whereas the other set contains Na2SO4.
these metals, and the strength of the bind- The pH dependence of the partition coef-
ing is proportional to the number of histi- ficient of proteins in the presence of NaCl
dine groups on the protein surface. is usually different from that in the pres-
One of the most common analytical ence of Na2SO4. So, the partition coeffi-
uses of aqueous 2-phase systems is the cient versus pH curves cross each other at a
determination of isoelectric points by pH (cross-point) that usually agrees with
determining cross-partitioning points (3, the isoelectric point of the proteins. Dis-
41,51,53). The method uses the different crepancies between the isoelectric points
sensitivity of the partition coefficient on determined by cross-partitioning and by
the kind of salt used and determines the isoelectric focusing can be due to confor-
pH at which the partition coefficient is mational changes of the protein in the
independent of salt type. If the protein of phases system or by interactions between
interest does not interact with contaminat- the polymers and the proteins.
ing materials in the solution, its isoelectric The knowledge of the hydrophobicity of
point can be determined without prior a protein is useful for the design of reverse
protein purification (other than desalting). phase chromatography units, for the
The determination of isoelectric points understanding of protein–ligand interac-
of proteins by cross-partitioning is straight- tions, and for the understanding of protein
Figure 2. Effect of cation and anion on the partition coefficient of proteins. The values of γ obtained from this figure must be
used with Equation 1.
192
Hemoproteins Purification and Characterization
folding and refolding. The usual way of partition coefficient with the molecular size
measuring hydrophobicity of a solute is to of the solutes. For example, Petersen
measure the free energy of transfer of that (38,39) found that the partition coefficient
solute from water into an organic solvent. in a Dx/PEG system of cytochrome c oxi-
Because of the need to use an organic sol- dase was 20, whereas that of cytochrome c
vent, these methods are not well suited to was 0.275 and used these differences to
measure hydrophobicities of polar and flex- show that both oxidized and reduced
ible biological molecules. Aqueous 2-phase cytochrome c formed a 1:1 complex with
systems have been used to determine the the oxidase. In another study, Middaugh
relative hydrophobicity of various proteins and Lawson (32) determined the associa-
(56). The overall idea is to “calibrate” a tion constant of hemoglobin by using
series of aqueous 2-phase systems (Dx/ aqueous 2-phase systems. They found that
Ficoll and Dx/PEG have been used) by the partition coefficients of oxyhemoglobin
partitioning a series of small solutes (amino and methemoglobin produced a sigmoidal
acids) of increasing hydrophobicity. It has curve when they were plotted against pro-
been found that the partition coefficient of tein concentration. From these plots, they
amino acid i, correlates with the partition determined the dimer–tetramer association
coefficient of glycine by (56): constant for these proteins.
The potential uses of this technique are
1n Ki = 1n KGly + (HF)(RHi) [Eq. 2]
vividly pictured in the following examples.
where HF is known as the hydrophobicity PEG-coated liposomes are a current alter-
factor (a constant for a given 2-phase sys- native to increase the stability of liposomes.
tem) and RHi is the hydrophobicity of The behavior of the modified liposomes
solute i relative to glycine. For a protein, will depend on their surface properties.
the surface hydrophobicity of the i protein Because the partition behavior of a particle
relative to molecule j, HSFi is given by: is a signature of its surface properties, par-
titioning of PEG-coated liposomes in
HSFi = 1n Kij /HFj [Eq. 3]
aqueous 2-phase systems can be used to
The technique has been used to deter- anticipate their behavior in the blood
mine the hydrophobicity of several hemo- stream. For example, Moribe et al. (34)
proteins including hemoglobin, apomyo- used aqueous 2-phase systems to detect
globin, and cytochrome c (56). The change surface differences of PEG-coated lipo-
of surface properties caused by pH- somes. They partitioned the coated and
induced denaturation of some of these pro- uncoated liposomes (after exposing them
teins has been investigated by this tech- to plasma) in two kinds of systems, charge
nique (56). For example, it has been found sensitive (5% PEG-8000, 5% Dx T-500,
that the difference in hydrophobic index and 0.11 M NaPO4, pH 7.0) and charge
(expressed in equivalent CH2 groups) for insensitive (0.01 NaPO4, 5% PEG-8000,
denatured and native cytochrome c is 146, 5% Dx T-500, and 0.15 M NaCl). Here
which indicates the exposure, as the pro- charge sensitive or insensitive refers to sys-
tein denatures, of hydrophobic residues tems in which the charge of the substance
otherwise buried in the interior of the to be partitioned either affects its partition
polypeptide. behavior or not. They concluded that in
It also is possible to follow protein–pro- spite of PEG being a steric barrier for the
tein and protein–ligand association by 2- interaction between plasma proteins and
phase partitioning (29). The basic princi- the liposomes, a weak interaction remains
ple here is the strong dependence of the between the PEG-coated liposomes and
193
D. Forciniti
matography using the appropriate stan- Biotech, whereas narrow fractions of PEG
dards. Some companies, like Wyatt Tech- can be purchased from PolySciences.
nology (Santa Barbara, CA, USA), provide
determination of molecular weights for a 2.2. Buffers
fee. The molecular weights of PEG and Dx
can be determined by using a Superose A variety of buffers have been used to
12 column (Amersham Pharmacia regulate the pH in aqueous 2-phase sys-
Biotech) eluted with a 3% NaCl solution tems. The two most commonly used are
at room temperature (17). Figure 3 shows a phosphate and Tris buffers. The buffers
typical molecular weight distribution curve must be kept in a refrigerator and used
for Dx. Molecular weight standards for within 30 days.
PEG can be bought from Polysciences.
Since molecular weight standards for Dx 2.3. Additives
are difficult to obtain, narrow fractions of
pullulan (Polysciences) can be used. Poly- A series of additives are normally used in
disperse polymers of good quality can be aqueous 2-phase systems. Bacteriocides
purchased from speciality chemical compa- (either sodium azide or chloroacetamide) are
nies like Sigma. Polymer batches of narrow conveniently added to the polymer stock
molecular weight distributions can also be solutions or in solid form to the 2-phase sys-
purchased but at a much higher price. For tem. A series of salts (shown in Figure 2) are
example, less polydisperse Dx can be commonly used to drive the protein of
bought from Amersham Pharmacia interest into one or another phase.
Figure 3. Molecular weight distribution of Dx T-500. This was obtained by running a Dx solution through a Superose 12 col-
umn eluted with NaCl (3%). The column was calibrated with pullulan standards.
195
D. Forciniti
by ultrafiltration prepare 50 g of a 1 g/L phase systems for the first time should
protein solution and ultrafiltrate with 5 choose equal volumes of top and bottom
volumes of a 50 mmol Tris buffer, pH 7.0. phases to facilitate sampling and protein
Stock solutions of affinity ligands, e.g., partition coefficient determination. For ana-
PEG-bound ligand, can be prepared at a lytical purposes, 5- or 10-g systems are very
concentration of hundredfold or more, as convenient. Although any buffer can be
the final concentration of these species in used, for acidic and neutral pHs, phosphate
the 2-phase system is quite low. The cost of buffers are recommended, whereas for basic
the PEG derivatives limits the amount of pHs, Tris buffers can be used. Buffer con-
stock solution to be prepared. centration should be kept between 20 to 50
mM. It is less cumbersome to work at room
3.2. Selection and Preparation of temperature since the mixing, equilibration,
Aqueous Two-Phase Systems and sampling has to be done at the same
temperature. Because protein partitioning is
The selection of the appropriate phase only marginally affected by temperature
system depends on the final application. changes, low temperatures may be desirable
PEG/Dx is by far the most common pair of to maintain protein stability. The final
polymers used. Other incompatible poly- preparation of an aqueous 2-phase system is
mers were already mentioned and summa- quite straightforward, and a detailed recipe
rized in Table 2. High molecular weight Dx can be found in Reference 11. An example
(Dx-500 000) is highly recommended, since is given in Procedure 2.
it can be used with low molecular weight
PEGs reducing the viscosity of the phases. ❖ Procedure 2. Preparation of a
Also popular are PEG/K2HPO4-KH2PO4 Dx-500 000/PEG-4000 System
systems. After the phase forming species at Room Temperature
have been selected, the next step is to select a
particular tie-line. Good sources of tie-lines 1. Shake the stock solutions well so that
are the monograph by P.Å. Albertsson (2), there are no density gradients.
the book by Zaslavsky (56), which contains 2. Place a graduated centrifuge tube of 15
about 150 phase diagrams for Dx/PEG, Dx/ mL total volume on a weighing balance.
Polyvinylpyrrolidine (PVP), Dx/Polyvinyl 3. Weigh out the stock solutions into the
alcohol (PVA), Dx/Ficoll, PEG/Polyvinyl tube in order of their increasing densi-
methyl ether, and PEG-salt, and several arti- ties, and layer them carefully over each
cles (16,43,44). Some convenient 2-phase other. This facilitates the removal of
systems are shown in Table 4. As a rule of portions of one stock solution in case
thumb, those who are using aqueous 2- of error during weighing. Because of
198
Hemoproteins Purification and Characterization
the problem of accurately pipetting the the ratio between top and bottom phase
polymer stock solutions due to their volumes of both published and prepared
high viscosity, they are best measured systems may not be the same. The appear-
by weight and are easily transferred ance of only one phase after following a
using a Pasteur pipet with a broken tip. published recipe step-by-step is equally
4. Mix the contents of the test tube thor- frustrating. These apparent inconsistencies
oughly, first by hand, and then in a cause people to believe that the lack of
rotary shaker (20 min is enough) at the reproducibility is an inherent property of
equilibration temperature. aqueous 2-phase systems. Fortunately, this
is not true. The most common reasons for
5. Let the systems settle for a period of 30 these inconsistencies are:
minutes to 24 hours depending on the
system composition, or centrifuge • The selected tie-line is too close to the
them for 2 to 15 minutes at 1500× g. critical point. So, small differences in
Poor temperature control in centrifuges the molecular weight or molecular
makes it more convenient to sediment weight distribution of the polymers,
the systems in a water bath or in a presence of additives, or differences in
chromatographic chamber when work- temperature moved the system into
ing at a temperature other than ambi- the one phase region. Addition of
ent. In general, the time of phase sepa- small amounts of one of the two poly-
ration depends on the distance of the mers will move the system into the 2-
working tie-line from the critical point. phase region.
Close to the critical point, the phase • The selected tie-line is too far from the
separation time is long. At intermedi- critical point. When working at longer
ate tie-lines the phase separation time is tie-lines poor mixing is normally the
shorter. If the more viscous phase vol- cause for the lack of production of 2
ume is larger than the volume of the phases. Since the denser stock solution
less viscous phase, the phase separation is added to the centrifuge tube first, it
time increases. If the system is to be is quite difficult to mix the residue of
used in a liquid–liquid partition chro- stock solution that is trapped in the tip
matography system, one must chose a of the tube with the rest of the solu-
total concentration of polymers such tion. So, the 2-phase system is actually
that the PEG-rich phase constitutes prepared using a considerable smaller
most of the volume. This phase system amount of one of the two polymers.
must be allowed to settle for 24 to 48 To assure good mixing, mix the con-
hours before the PEG-rich phase is tent of the tube in a vortex mixer and
used as the mobile phase. inspect the tip and walls of the tubes
If instructions are followed carefully, the for stock solution residues. Continue
preparation of aqueous 2-phase systems mixing after no deposits are present,
should be routine laboratory work. Still, a and place the tube in a rotary shaker.
common source of frustration for those
using aqueous 2-phase systems for first 3.3. Preparation of Liquid–Liquid
time is their apparent lack of reproducibil- Partitioning Chromatography
ity. As indicated before, systems are nor- Systems
mally prepared according to some pub-
lished binodal. Often, the prepared system Specific guidelines for the use of this
differs from the one published. Specifically, method can be found in the various articles
199
D. Forciniti
by Muller (35–37). The main steps are attempting to run a liquid–liquid par-
outlined in Procedure 3. tition chromatography (LLPC) col-
umn. The partition coefficient of the
❖ Procedure 3. Preparation of target protein must be adjusted to be
Partitioning Chromatography Systems between 0.3 and 0.1 (the data shown
in Figure 2 can be used for this pur-
1. Measure the partition coefficient of the pose), and the salt concentration
raw materials in batch systems before should be high enough to shield elec-
Figure 4. Purification of hemoglobin from E. coli. Cell homogenate is mixed with Dx and PEG solutions. The Dx-rich phase is
discarded, and Na2SO4 is added to the PEG-rich phase. The Na2SO4-rich phase is discarded, and MgSO4 is added to the PEG-
rich phase. The protein is recovered from the salt-rich phase.
200
Hemoproteins Purification and Characterization
trostatic interactions between the pro- relate quite well with the partition coeffi-
teins and the gel (ionic strength of 0.05 cients of the proteins obtained in batch
or higher). experiments, so scale-up is straightforward;
2. After the optimum conditions to (iii) very low amounts of Dx are needed,
obtain an appropriate partition coeffi- which is of direct benefit as far as costs go.
cient have been identified, prepare In addition, if the losses are proportional to
enough top phase at the right pH and the total Dx, then the losses are expected to
at the right ionic strength to elute the be low as well; (iv) PEG precipitates large
column. The top (PEG-rich) phase is proteins (above 200 000 Da) at the station-
allowed to equilibrate for several days ary phase–mobile phase interface. This is
in the presence of small amount of bot- avoided at all costs, as the precipitate clogs
tom phase. the column; and (v) the eluant contains
significant amounts (around 10% wt/wt)
3. The column is packed according to
PEG. Depending on the final application,
Muller (36) and equilibrated until the
this can be removed as described in the fol-
UV noise of the effluent has dropped
lowing section.
below 0.005 OD units at 280 nm. This
ensures that all the Dx-rich phase that
is not bound to the beads has been ❖ Procedure 4. Determination of
washed out. Partition Coefficients
4. The sample is dissolved in the mobile
phase (it should not exceed 2%–3% of 1. Approximately 1 mL of the protein
the bed volume for analytical runs and solution to be purified is added to the
as twice as much for preparative runs) phase-forming species mixture replac-
and injected into the column. The elu- ing an equal amount of buffer. Mixing
tion is started immediately. and phase separation are done as
described above for systems that do not
As in any chromatographic separation, contain any protein.
sample preparation is quite important. If
the starting material is a cell homogenate, 2. Mixing has to be done carefully. It has
solids must be sedimented out by centrifu- to be vigorous enough to allow distrib-
gation for about 15 minutes at 2000× g. ution of the proteins between the 2
The clear supernatant is mixed with the phases but gentle enough to prevent
appropriate amount of PEG that is going protein denaturation (a rotary shaker is
to be used as a mobile phase. If aggregates highly recommended, whereas the use
are observed, they must be eliminated by of a vortex mixer is discouraged).
centrifugation. If no precipitation is 3. The phase systems are centrifuged at
observed, more PEG is added (to reach 1500× g for 20 minutes to speed phase
30%), the liquid is cooled in an ice bath settling.
for 10 minutes, and the protein precipitate 4. Sampling is done by pipetting carefully
is removed and resuspended in buffer. 1 mL of top phase and 1 mL of bottom
Some features of these type of systems phase from each partitioning tube (the
are: (i) the partition coefficients must be amount pipetted should be controlled
sufficiently different from 1 to make an by weighing for more precise sam-
impact on the retention times. That is, pling). Impurities may accumulate at
they must be in the right range to make a the liquid–liquid or liquid–air inter-
multicomponent chromatographic separa- faces. They do not constitute a prob-
tion possible; (ii) the elution volumes cor- lem unless they are pipetted during
201
D. Forciniti
points should be obtained in the neighbor- generally due to poor sampling, and a pro-
hood of the cross-point. tein mass balance should be done to assure
that sampling has been done correctly.
❖ Procedure 5. Cross-Point The sensitivity of cross-partitioning
Determination depends upon the angle at which the 2
lines intersect. If the lines are perpendicu-
1. Set A. Add to a 10-mL centrifuge tube lar, the sensitivity is at a maximum, while
(37): (i) 2.5 g of Dx stock solution; (ii) parallel or nearly parallel lines yield no
1.0 g of PEG stock solution; (iii) 3 g of cross-point or a “cross-point range”. The
sodium chloride; (iv) 2.5 g of buffer; slope of the lines depends upon the type of
and (v) 1 g of the protein stock solution. salt, the change in net charge of the protein
2. Set B. Add to a 10-mL centrifuge tube: with pH, the specific interactions between
(i) 2.5 g of Dx stock solution; (ii) 1.0 g the ions and the proteins, and the salt-
of PEG stock solution; (iii) 3 g of or induced changes in the interactions
sodium sulfate solution; (iv) 2.5 g of between polymer and protein. The sensi-
buffer; and (v) 1 g of the protein stock tivity can be manipulated by varying the
solution. molecular weight of the polymers, the tem-
3. The final phase system composition is perature, the concentration of the poly-
7.5% (wt/wt) Dx, 5.0 (wt/wt) PEG, mers, and the type of salt. So, cross-parti-
0.1 M alkali chloride or 0.05 M alkali tioning should be done using the lowest
sulfate, and 0.04 M glycine or sodium possible PEG molecular weight to mini-
phosphate buffer. mize problems associated with the high vis-
4. Prepare blanks of the phases without cosity of the phases. If the sensitivity is not
added protein. good enough, the experimentalist needs to
5. Mix, equilibrate, and sample the phases explore different conditions until a good
sensitivity is found.
as explained in Section 3.4.6.
The pH and the partition coefficient at
6. The pH in each phase is measured with the cross-point are only marginally depen-
a microelectrode directly on the undi- dent on the combination of salts used and
luted phases. Because of the high vis- on their concentration. For example, NaCl
cosity of the phases, the pH measure- can be replaced by potassium chloride
ments must be done over a relatively and/or sodium sulfate by lithium sulfate
long period of time. without affecting the results. Still, some
7. The partition coefficients of Sets A and small differences in pH values at the cross-
B are plotted versus the pH. The pH point with different salts have been ob-
and the partition coefficient values at served. These differences are similar to
which one Kp versus pH line (Set A) those encountered in the electrophoretic
crosses the other one (Set B) and are determination of isoelectric points, which
read from the axes. can also be slightly affected by the salt used.
The lines of Kp versus pH may not cross This independence of cross-partitioning
each other because of errors in pH or in the on the type and concentration of salt
values of the partition coefficients. One makes cross-partitioning a viable option for
must be sure that the pH has been mea- determining the isoelectric point of pro-
sured long enough to reach equilibrium and teins that are stable only at high salt con-
that the pH of both phases agrees within the centrations. In contrast, the type and con-
experimental uncertainty (approximately centration of salts have a strong influence
0.05 pH units). Erroneous values of Kp are on the shape of the ln Kp versus pH curves.
203
D. Forciniti
point: 9.8). The cross-partitioning of he- The tune-up of the separation is quite
moproteins from cytochrome c (MW straight forward, and it includes optimiza-
12 000) to catalase (MW 240 000) in Dx/ tion of the partition coefficient of the target
PEG systems (51) yields partition coeffi- protein as compared to the contaminant
cients at the cross-point that do not show proteins, optimization of the amount of
the clear dependence on protein molecular affinity ligand, and inhibition of other
weight found with nonhemoproteins. Even enzymes that compete with the target pro-
though the molecular weights of human tein for the affinity ligand. For example, in
hemoglobin variants (A, F, S, and C) and the purification of FDH by using affinity
hemoglobins from different species are liquid–liquid partition chromatography, we
essentially the same, the partition coeffi- found that the optimum conditions were
cient at the cross-point of hemoglobins A achieved when the stationary phase was Dx-
and F and those of hemoglobins from dif- 500, the mobile phase PEG-20 000 (2.7%
ferent mammalian species show measurable PEG and 4.5 Dx), 75 mM potassium bro-
differences. Although the 4 human hemo- mide, 12 mM phosphate buffer at pH 7.5,
globin variants differ in charge, adult and a concentration of PEG-red in the
hemoglobins A, S, and C have the same the mobile phase of 5 × 10-5 M. The sample
partition coefficient, while the fetal hemo- was 500 µL of crude extract of C. boidinii
globin (F) has a lower partition coefficient. heated for 10 minutes at 55°C. We also
showed that the separation of the ligand
4.4. Liquid–Liquid Partitioning from the enzyme is quite straightforward.
Chromatography The eluate containing the FDH–ligand
complex was treated with potassium phos-
There are numerous examples of the use phate forming a PEG-salt system. Using a
of this technique for the purification of concentration of potassium phosphate of
nucleic acids and proteins. In a few of 10% (wt/wt) the K value of FDH was
them, the addition of affinity ligands to 0.003, and the volume ratio was 1:4.22 (top
improve the separation has been explored. to bottom). The yield of the enzyme in the
Because the elution volume is extremely lower phase was 99%. The column was sta-
sensitive to the partition coefficient of the ble for more than 1 year, and scale-up was
protein (if it is smaller than 3), even in the straightforward.
absence of affinity ligands considerable res-
olution can be obtained. For example,
Muller (37) separated a synthetic mixture 5. CONCLUDING REMARKS
of lysozyme (retention volume, V = 17
mL), peroxidase (V = 19 mL), cytochrome Aqueous 2-phase extraction is a very
c (V = 25 mL), myoglobin (V = 30 mL), β- well-established technique that has been
lactoglobulin (V = 39 mL), ovotransferrin used in biochemical laboratories for the last
(V = 50 mL), ovalbumin (V = 60), and 40 years. It is a versatile, easy to use, and low
human serum albumin (V = 100) (0.6–1 cost technique. Although the primary use of
mg each) in a 300 × 10 mm column the method is the purification of proteins
packed with Lipargel coated with a solu- and other biological materials, it can be also
tion of Dx-40 and eluted at a flow rate of used for protein characterization studies.
0.3 mL/minute with a PEG-6000 solution. Uncountable proteins have been purified
We have demonstrated that affinity lig- using this technique either from cell
ands can be used in LLPC to increase the homogenates or from previously fractionat-
purification factor of a given protein (49). ed mixtures. Although it has been used to
205
D. Forciniti
separate cell debris from proteins providing 4.Alred, A. 1999. Purification of a recombinant protein
a first cheap purification step, it has been from mammalian cell culture: an industrial applica-
tion. 11th International Conference on Partitioning in
also used to purify proteins in a single step Aqueous Two-Phase Systems, Gulf Shore, Alabama.
from cell debris by making use of a variety 5.Asenjo, J.A., R.E. Turner, S.L. Mistry, and A. Kaul.
of affinity ligands. The possibility of using 1994. Separation and purification of recombinant pro-
teins from Escherichia coli with aqueous two-phase sys-
the technique in continuous mode by tems. J. Chromatogr. A 668:129-138.
immobilizing 1 phase opens even more pos- 6.Bergreen, K., G. Johansson, and F. Tjerneld.
sibilities. One of its main advantages is its 1995. Effects of salts and the surface hydrophobici-
ty of proteins on partitioning in aqueous two-phase
low cost and easy use. The cost of chemicals systems containing thermoseparating ethylene
is minimum (except if affinity ligands has to oxide-propylene oxide copolymers. J. Chromatogr.
be used), and the hardware needed to A 718:67-79.
7.Birkenmeir, G., G. Kopperschlager, P.A. Albertsson,
implement it is available in any biochem- G. Johansson, F. Tjerneld, H.E. Akerlund, S. Berner,
istry laboratory. Members of the aqueous 2- and H. Wickstroem. 1987. Fractionation of proteins
from human serum by counter-current distribution. J.
phase systems community can be reached at Biotechnol. 5:115-129.
our Web page for helping those who wish to 8.Bradford, M.M. 1976. A rapid and sensitive method
use this technique. I hope that this chapter for the quantitation of microgram quantities of protein
utilizing the principle of protein dye binding. Anal.
will encourage researchers in the heme and Biochem. 72:248-254.
related compounds research field to consid- 9.Fernandez, S., G. Johansson, and R. Hatti-Kaul.
er aqueous 2-phase systems for their isola- 1999. Two-phase partitioning of recombinant Cuti-
nase. 11th International Conference on Partitioning in
tion procedures. Aqueous Two-Phase Systems, Gulf Shore, Alabama.
10.Forciniti, D. 1994. Protein refolding using aqueous
two-phase systems. J. Chromatogr. A 668:95-100.
ABBREVIATIONS 11.Forciniti, D. 1999. Preparation of aqueous two-phase
systems. In R. Hatti-Kaul (Ed.), Aqueous Two-Phase
Systems–Methods and Protocols. Humana Press, New
Dx, dextran; HF, hydrophobicity factor; Jersey.
12.Forciniti, D., C.K. Hall, and M.-R. Kula. 1991. Effect
HSFi, surface hydrophobicity of protein I; of temperature on protein partitioning. Bioseparations
Kp, partition coefficient (= Ct/Cb); K0, 2:115-128.
partition coefficient at zero charge; K0, 13.Forciniti, D., C.K. Hall, M.-R. Kula. 1991. Protein
partitioning at the isoelectric point: effect of polymer
partition coefficient in the absence of affin- concentration and polymer molecular weight. Biotech-
ity ligands; KL, partition coefficient in the nol. Bioeng. 38:986-994.
presence of affinity ligands; PEG, poly- 14.Forciniti, D., C.K. Hall, and M.-R. Kula. 1992. Pro-
tein partitioning. Effect of pH and polymer molecular
ethyleneglycol; P(EO-PO), polyethylene weight. Chem. Eng. Sci. 47:165-175.
propylene random copolymer (UCON); 15.Forciniti, D., M.-R. Kula, and C.K. Hall. 1990. Inter-
PVA, polyvinyl alcohol; PVP, polyvinyl- facial tension in aqueous two-phase systems. Influence
of temperature and polymer molecular weight. J.
pyrrolidone; RHi, hydrophobicity of solute Biotechnol. 16:279-290.
i relative to glycerin; TLL, tie-line length. 16.Forciniti, D., M.-R. Kula, and C.K. Hall. 1991. Influ-
ence of polymer molecular weight and temperature on
phase composition in aqueous two-phase systems.
Fluid Phase Equilib. 61:243-262.
REFERENCES 17.Forciniti, D., M.-R. Kula, and C.K. Hall. 1991. Mol-
ecular weight distribution and aqueous two-phase sys-
1.Aguinada-Diaz, P.A. and R.Z. Guzman. 1996. Affin- tems. Biotechnology 20:151-162.
ity partitioning of metal ions in aqueous polyethylene 18.Guan, Y., T.H. Lilley, T.E. Treffry, C.-L. Zhou, and
glycol/salt two-phase systems with PEG-modified P.B. Wilkinson. 1996. Use of aqueous two-phase sys-
chelators. Sep. Sci. Technol. 31:1483-1499. tems in the purification of human interferon-alpha1
2.Albertsson, P.Å. 1986. Partition of Cell Particles and from recombinant E. Coli. Enzyme Microb. Technol.
Macromolecules, 3rd ed. Wiley (Interscience), New 19:446-455.
York. 19.Guoqiang, D., R. Kaul, and B. Mattiasson. 1994.
3.Albertsson, P.Å., S. Sasakawa, and H. Walter. 1970. Integration of aqueous two-phase extraction and affin-
Cross partition and isoelectric points of proteins. ity precipitation for the purification of lactate dehy-
Nature 228:1329-1330. drogenase. J. Chromatogr. A 668:145-154.
206
Hemoproteins Purification and Characterization
20.Harris, P.A., G. Karlström, and F. Tjerneld. 1991. 36.Muller, W. 1994. Columns using aqueous two-phase
Enzyme purification using temperature- induced phase systems. In H. Walter and G. Johansson (Eds.), Meth-
formation. Bioseparation 2:237-246. ods in Enzymology. Academic Press, NY.
21.Hart, R.A. and J.E. Bailey. 1991. Purification and 37.Muller, W. 1994. Separation of proteins and nucleic
aqueous two-phase partitioning properties of recombi- acids. In H. Walter and G. Johansson (Eds.), Methods
nant Vitreoscilla hemoglobin. Enzyme Microb. Tech- in Enzymology. Academic Press, New York.
nol. 13:788-795. 38.Petersen, L.C. 1978. Cytochrome c-cytochrome aa3
22.Hatti-Kaul, R. (Ed.). 1999. Aqueous Two-Phase Sys- complex formation a low ionic strength studied by
tems–Methods and Protocols, Chs. 2 and 3. Humana aqueous two-phase partition. FEBS Lett. 94:105-108.
Press, New Jersey. 39.Petersen, L.C. 1978. Measurements of cytochrome c-
23.Hustedt, H., K.H. Kroner, and M.-R. Kula. 1984. cytochrome aa3 complex formation by aqueous two-
Applications of phase partitioning in biotechnology, p. phase partition. Biochem. Soc. Trans. 6:1274-1275.
529-589. In H. Walter, D.E. Brooks, and D. Fisher 40.Plunkett, S.D. and F.H. Arnold. 1990. Metal affinity
(Eds.), Partitioning in Aqueous Two-Phase Systems. extraction of human hemoglobin in an aqueous poly-
Academic Press, New York. ethylene glycol-sodium sulfate two-phase system.
24.Johansson, G. 1984. Partitioning of proteins, p. 161- Biotechnol. Tech. 4:45-48.
226. In H. Walter, D.E. Brooks, and D. Fisher, (Eds.), 41.Sasakawa, S. and H. Walter. 1972. Partition behavior
Partitioning in Aqueous Two-Phase Systems. Academ- of native proteins in aqueous Dx-poly(ethylene glycol)
ic Press, New York. phase systems. Biochemistry 11:2760-2765.
25.Johansson, G. 1994. Charge determination by parti- 42.Shanbhag, V.P. and P.E.H. Jensen. 1999. Affinity par-
tioning. In H. Walter and G. Johansson (Eds.), Meth- titioning using poly(ethylene glycol) with covalently
ods in Enzymology. Academic Press, New York. coupled hydrophobic groups. In R. Hatti-Kaul (Ed.),
26.Kopperschlager, G. 1994. Affinity extraction with dye Aqueous Two-Phase Systems—Methods and Protocols.
ligands. In H. Walter and G. Johansson (Eds.), Meth- Humana Press, New Jersey.
ods in Enzymology. Academic Press, New York. 43.Silva, L.H.M., J.S.R. Coimbra, and A.J.A. Meirelles.
27.Lorch, J. 1999. Two-phase aqueous extraction as a 1997. Equilibrium phase behavior of poly(ethylene
process development tool. 11th International Confer- glycol) + potassium phosphate + water two-phase sys-
ence on Partitioning in Aqueous Two-Phase Systems, tems at various pH and temperatures. J. Chem. Eng.
Gulf Shore, Alabama. Data 42:398-401.
28.Lu, M., F. Tjernald, G. Johansson, and P.Å. 44.Snyder, S.M., K.D. Cole, and C.C. Szlag. 1992. Phase
Albertsson. 1991. Preparation of benzoyl-Dx and its composition viscosities, and densities for aqueous two-
use in aqueous two-phase systems. Bioseparation 2: phase systems composed of polyethylene glycol and
247-255. various salts at 25 C. J. Chem. Eng. Data 37:268-274.
29.Lundeberg, S. and L. Backman. 1994. Protein-protein 45.Sophianopulos, A.J. and K.E. Van Holde. 1964. Phys-
and protein-ligand interactions, p. 241-254. In H. ical studies of muramidase (lysozyme). J. Biol. Chem.
Walter and G. Johansson (Eds.), Methods in Enzymol- 239:2516-2524.
ogy. Academic Press, New York. 46.Spears, T. and D. Forciniti. 1994. Protein refolding
30.Mattiasson, B. and I.Y. Galaev. 1999. Affinity parti- using chaotropic aqueous two-phase systems. In R.D.
tioning using aqueous two phase systems formed by Rogers (Ed.), Aqueous Two-Phase Systems: From
thermosensitive polymers. 11th International Confer- Metal Ions to Biomolecules. ACS Books, Washington.
ence on Partitioning in Aqueous Two-Phase Systems, 47.Tjerneld, F., P.A. Alred, R.F. Modlin, A. Kozlowski,
Gulf Shore, Alabama. and J.M. Harris. 1995. Purification of biomolecules
31.Menge, U., M. Morr, U. Mayr, and M.-R. Kula. using temperature-induced phase separation. In R.D.
1983. Purification of human fibroblast interferon by Rogers and M.A. Eiteman (Eds.), Aqueous Biphasic
extraction in aqueous two-phase systems. J. Appl. Separations: Biomolecules to Metal Ions. Plenum
Biochem. 5:75-90. Press, New York.
32.Middaugh, C.R. and E.Q. Lawson. 1980. Analysis of 48.Vernau, J. and M.-R. Kula. 1990. Extraction of pro-
protein association by partitioning in aqueous two- teins from biological raw materials using aqueous
phase polymer systems: applications to the tetramer- PEG/citrate phase systems. Biotechnol. Appl. Bio-
dimer association of hemoglobin. Anal. Biochem. chem. 12:397-404.
105:364-368. 49.Walsdorf, A., D. Forciniti, and M.-R. Kula. 1990.
33.Modlin, R.F., P.A. Alred, and F. Tjerneld. 1994. Uti- Investigation of affinity partition chromatography
lization of temperature-induced phase separation for using formate dehydrogenase as a model. J. Chro-
the purification of ecdysone and 20-hydroxyecdysone matogr. 523:103-117.
from spinach. J. Chromatogr. A 668:229-236. 50.Walter, H., D.E. Brooks, and D. Fisher (Eds.). 1984.
34.Moribe, K., K. Maruyama, and M. Iwatsuru. 1997. Partitioning in Aqueous Two-Phase Systems, p. 498-
Estimation of surface state of poly(ethylene glycol)- 528. Academic Press, New York.
coated liposomes using an aqueous two-phase parti- 51.Walter, H. and D. Forciniti. 1994. Cross-partition-
tioning technique. Chem. Pharm. Bull. 45:1683- ing: determination of isoelectric point by partitioning,
1687. p. 223-233. In H. Walter and G. Johansson (Eds.),
35.Muller, W. 1989. Aqueous two-phase polymer systems Methods in Enzymology. Academic Press, New York.
for liquid/liquid partition-chromatography of biopoly- 52.Walter, H. and G. Johansson (Eds.). 1994. Methods
mers. Ber. Bunsen-Ges. Phys. Chem. 93:956-961. in Enzymology, Vol. 228. Academic Press, New York.
207
D. Forciniti
53.Walter, H., S. Sasakawa, and P.Å Albertsson. 1972. tioning in Aqueous Two-Phase Systems, Gulf Shore,
Cross partition of proteins. Effect of ionic composition Alabama.
and concentration. Biochemistry 11:3880-3883. 55.Wuenschell, G.E., E. Naranjo, and F.H. Arnold.
54.Winter, C., D. Ansaldi, J. Clifford, P. Lester, and B. 1990. Aqueous two-phase metal affinity extraction of
Wolk. 1999. Initial isolation of recombinant proteins heme proteins. Bioprocess Eng. 5:199-202.
from whole fermentation lysates using aqueous two 56.Zaslavsky, B.Y. 1995. Aqueous Two-Phase Partition-
phase systems. 11th International Conference on Parti- ing. Marcel Dekker, New York.
208
Structural Study of Heme Proteins by
9 Electron Microscopy of 2-Dimensional
Crystals
Terrence G. Frey
San Diego State University San Diego, CA, USA
209
T.G. Frey
not only excess lipid but also other conta- gent used, two different crystal forms have
minating proteins. For example, cytochrome been obtained. Multiple extractions with
c oxidase constitutes approximately 10% of nonionic Triton® X-114 and X-100 pro-
the protein of the mitochondrial inner mem- duces a vesicular preparation of nearly pure
brane, and one can purify a membrane frac- cytochrome oxidase and 25% by weight of
tion of nearly pure cytochrome oxidase by residual phospholipid. The molecules of
treating beef heart mitochondria with appro- cytochrome oxidase are arranged as dimers
priate concentrations of detergents followed related by a crystallographic 2-fold axis.
by centrifugation to separate detergent solu- The crystal is formed when a large vesicle
bilized components from the membrane collapses causing molecules from two sides
residue (30,66,80). This requires 2 to 3 of the vesicle to interdigitate in the center
detergent treatments, and in each case, the of the vesicle, as shown in Figure 1b, form-
pellet becomes enriched in cytochrome ing a 2D crystal in the space group p22121
oxidase while other membrane proteins with unit cell dimensions of a = 100 Å, b =
and excess phospholipid are removed by 125 Å, and a thickness of 210 Å (42). In
decanting the supernatants. Depending this case, each unit cell of the crystal con-
upon the type and concentration of deter- tains one molecule from each layer of the
211
T.G. Frey
collapsed vesicle, i.e., each unit cell con- with purified detergent solubilized protein
tains two dimers. The space group and detergent solubilized lipids, either
describes a primitive orthorhombic lattice “native” lipids purified from the same type
with a 2-fold axis of rotation perpendicular of membrane as the protein are derived or
to the membrane. The symmetry operators synthetic lipids using Procedure 1
21 indicate 2-fold screw axes (rotation by (55,60,82,84). This is described in more
360°/2 followed by translation by half of a detail in Chapter 11.
unit cell) parallel to the a and b crystal axes
in the plane of the crystal. ❖ Procedure 1. Growth of 2D Crystals
Treatment with sodium deoxycholate
removes a large proportion of phospholipid 1. Dissolve lipids in an organic solvent
and dissociates cytochrome oxidase dimers (e.g., chloroform) and dry a measured
into monomers. The resulting 2D crystal is amount on the surface of a suitable ves-
not vesicular but consists of sheets of sel under a stream of nitrogen.
cytochrome oxidase monomers arranged on 2. Suspend the dried lipid in an appropri-
a primitive 2D lattice in the space group ate volume of buffered detergent by
p121, denoting no symmetry perpendicular sonication, producing mixed deter-
to the membrane and a 21 screw axis of sym- gent–lipid micelles.
metry along one of the crystal axes (37).
These crystals are of the class shown in Figure 3. Mix the protein and lipid solutions
1c. The unit cell dimensions are a = 69 Å and approximately 1–10 mg/mL) to give a
b = 140 to 170 Å depending upon the prepa- relatively high protein to lipid ratio
ration; the thickness is approximately 110 Å (approximately 1:1 by weight).
(35,37). In both cases, the structures of the 4. Remove the detergent by dialysis or by
preparations are more complex than adsorption.
described here. The vesicular p22121 dimer As the detergent concentration decreas-
crystals are predominantly multilayered vesi- es, the protein–detergent micelles and the
cles with many small 3D crystals containing protein–lipid micelles merge and eventual-
layers of the 2D crystal motif stacked in reg- ly form bilayer membranes containing a
ister (42,51,80). Single 2D crystals are more high concentration of protein. Under the
rare but can readily be found. The p121 crys- proper conditions, the protein molecules
tal form contains single layer crystals as well within the bilayers arrange themselves onto
as stacks of the simple 2D crystal motif in a 2D lattice forming 2D crystals (see Fig-
which successive layers are offset from one ure 2) (50,55,81). There are two common
another, and electron micrographs show sev- methods to remove detergent. One is to
eral differently appearing projections (35). adsorb excess detergent by adding com-
Multilayered crystals of membrane proteins mercially available resin beads, e.g., Bio-
are commonly observed and often severely Beads® (Bio-Rad Laboratories, Hercules,
inhibit structural study because of the het- CA, USA) (81). A slower more controlled
erogeneity in the structures of different crys- method is to dialyze the detergent–pro-
tals (35,54,69). tein–lipid solution against detergent-free
buffer. The speed of this process depends
2.3. Reconstitution of Purified Protein on the characteristics of the detergent
with Purified Lipids employed, principally the critical micelle
concentration (cmc)(50,55). In aqueous
The most general method for growing solution, detergent molecules aggregate to
2D crystals of membrane proteins begins form micelles that are in equilibrium with
212
Structural Study of Heme Proteins by Electron Microscopy
individual detergent molecules; the cmc is ple which should be pure and monodis-
essentially the concentration of individual perse. The latter property can be difficult
molecules in the presence of micelles. Since to achieve in the case of membrane pro-
detergent micelles are too large to pass teins which, owing to their hydrophobic
through the pores of common dialysis tub- surfaces, can adopt multiple aggregation
ing, dialysis of detergent solutions proceeds states even in the presence of detergents.
by the movement of individual detergent Thus, the choice of detergent can be criti-
molecules through the dialysis tubing. cal and affects both the aggregation of the
Thus, the rate of dialysis depends on the protein and the methods employed in
cmc; the higher the cmc, the faster dialysis removing it during crystallization trials.
occurs. Of course dialysis also proceeds For example, Suarez et al. found that beef
more rapidly at higher temperature. Triton heart mitochondrial cytochrome c oxidase
detergents have low cmc’s, so dialysis even is polydisperse in many common deter-
at room temperature takes a relatively long gents, but became monodisperse when
period of time. For this reason, Weiss et al. transferred to dodecyl-maltoside (also
used the Bio-Bead adsorption method to known as lauryl maltoside), a detergent
crystallize Complex III (cytochrome c oxi- consisting of a maltose polar head and a 12
doreductase) purified in Triton X-100 from carbon saturated hydrocarbon tail that they
Neurospora mitochondria (81,84). Kim et synthesized for this purpose (now commer-
al. grew crystals of purified cytochrome c cially available) (70). Choice of detergent
oxidase essentially identical to the dimer also affects the crystallization process,
crystals described above by reconstitution since, as mentioned above, the cmc deter-
with purified phospholipids (53). mines the dialysis rate, and other properties
Many factors can affect the prospects of affect the efficacy of adsorption to beads.
success in crystallizing a membrane pro- Frequently, two detergents within the same
tein. As with most crystallization experi- generic class can have significantly different
ments, a critical factor is the protein sam- cmcs. For example, dodecyl-maltoside has
Figure 2. Reconstitution of purified phospholipids and purified membrane protein into a protein–lipid bilayer by mixing
detergent solubilized lipid (left) and detergent solubilized protein (center) and then removing the detergent molecules.
213
T.G. Frey
214
Structural Study of Heme Proteins by Electron Microscopy
specimen preparation is negative staining Freshly evaporated carbon films are gen-
first developed in the 1960s. Negative erally hydrophilic and readily adsorb pro-
staining works well with 2D membrane teins and lipid vesicles. As they are stored,
protein crystals, but the useful resolution however, they quickly become more
that can be achieved is limited to about 10 hydrophobic and then do not adsorb the
Å, and 20 Å is more common. The speci- specimen as well. This change is readily
men in an aqueous suspension of approxi- observed as one draws a liquid drop off of
mately 1 mg protein/mL is adsorbed onto the grid with a piece of filter paper. If the
a thin carbon film. Carbon films are gener- liquid draws off evenly leaving a thin film,
ated by evaporation of carbon onto a suit- the grid is hydrophilic, but if it is all
able surface, for example freshly cleaved rapidly drawn off leaving none behind,
mica, from which it can be floated off onto the grid is hydrophobic and will probably
a water surface and lowered onto the sur- not adsorb the specimen. One can over-
face of electron microscope grids (com- come this change to some extent by
monly 400 mesh copper grids) previously increasing the concentration of the speci-
placed on a support beneath the film. The men, but it is often necessary to render
carbon film can be lowered gently onto the the surface hydrophilic again by a process
grid surfaces by draining water out of the called glow discharge, in which the films
vessel. Alternatively, one can coat grids are exposed to the ions formed by an elec-
with a plastic film, generally Collodion, tric discharge in the residual gases in a
but Formvar is also used, after which a car- vacuum apparatus pumped down to sev-
bon film is evaporated onto the plastic eral tens of microns of pressure. A glow
film. These carbon–Collodion films can be discharge accessory is common in com-
used directly, but best results are usually mercial vacuum evaporators found in
obtained if the plastic is dissolved away electron microscopy laboratories, or an
with a solvent, since pure carbon films are inexpensive glow discharge apparatus can
thinner and more conductive giving less be constructed from a plastic vacuum
movement when irradiated with an elec- dessicator, a Tesla coil vacuum tester, and
tron beam. Ideally, one should use the an inexpensive rotary pump (3). Proce-
thinnest carbon film possible in order to dure 2 describes a typical method to pre-
minimize its contribution to the image, pare negatively stained specimens.
but very thin carbon films are fragile, and
400 mesh grids are too coarse to support ❖ Procedure 2. Preparation of
them adequately. In this case, one can first Negatively Stained Specimens
coat the grid with a holey film, a plastic
film containing numerous circular or 1. Allow a drop of the specimen to adsorb
ellipsoidal holes of varying sizes. There to a carbon-coated grid for 1 to 5 min-
are several methods by which holey plas- utes.
tic films can be formed (36,57,63), and 2. Wash the grid with several drops of
they are generally stabilized to electron water or buffer followed by several
irradiation by evaporation of carbon onto drops of a suitable negative stain, com-
them. Holey films make an excellent sup- monly 1% to 2% uranyl acetate, but
port for very thin carbon films and can uranyl formate and phosphotungstic
even be used to support thin layers of acid are also used. Uranyl stains gener-
stain over the holes without a carbon sup- ally provide higher resolution, but
port film. must be used at a pH of about 4.5,
215
T.G. Frey
which may disrupt the structures of 2. Rinse with water or volatile buffer to
some specimens. remove sample that is not adsorbed to
3. Draw off excess stain with a piece of fil- the surface.
ter paper leaving a very thin layer that 3. Freeze by plunging into a cryogen; liq-
dries down forming a glass-like elec- uid nitrogen is acceptable, but ethane
tron dense replica surrounding the or propane cooled in liquid nitrogen
specimen molecules supporting and freezes much more rapidly.
contrasting them. 4. Place the sample in a freeze frac-
In the electron microscope, the actual ture–etch instrument and remove ice by
protein structure appears light against a sublimation at approximately 70°C.
dark background formed by the stain, 5. Contrast the surface by evaporating a
hence the term “negative” stain. It is heavy atom, generally platinum evapo-
important to realize that negative staining rated with carbon, but tungsten–tanta-
contrasts the 3D surface of the molecule lum may produce smaller metal grains
with atoms that are approximately 7 Å in yielding higher resolution (12). This
diameter. Thus, the resolution in negative- creates a shadow effect that highlights
stained specimens is limited to approxi- the surface topography and can also be
mately 10 Å at best and more commonly used to measure the thickness of the
20 Å. Furthermore, the internal structure specimen.
of protein domains are not contrasted. This technique has been applied to con-
trast selectively the surface of vesicle crystals
3.2. Other Methods of cytochrome oxidase dimers indicating
that the enzyme protrudes 20 to 30 Å
Images of negatively stained specimens beyond the bilayer surface on the exterior
are a projection of the electron density of surface (corresponding to the matrix side of
the 3D stain replica onto the 2D image the inner mitochondrial membrane) (33)
plane. Ultimately, one would wish to cal- and to measure the thickness of a number
culate a 3D structure from multiple images of 2D crystals (37). Berriman, Leonard,
of tilted specimens (see below), but infor- and coworkers used shadowing to demon-
mation on the 3D configuration of the strate that crystals of the mitochondrial
molecule can often be quickly obtained by cytochrome bc1 complex thinned markedly
other common specimen preparation tech- during electron irradiation (5), and Smith
niques. Indeed, this information is often and Ivanov have published a procedure to
essential to confirm the molecular packing compute the surface relief structure from
of a new crystal form, information that is images of shadowed specimens (67). A
required in calculating the 3D structure. related technique, freeze fracture–replica-
One of the most useful techniques is heavy tion, can also be used to study the structure
metal shadowing of freeze-dried specimens of membrane protein crystals in order to
(Procedure 3). help determine the molecular packing (13).
Heuser has adapted the technique of rapid
❖ Procedure 3. Heavy Metal slam freezing and freeze fracture–etch to
Shadowing look at molecules adsorbed to a slurry of
small mica chips (44). Conventional plastic
1. Adsorb the specimen to a flat embedding and thin sectioning can also be
hydrophilic surface, such as a carbon used to evaluate the gross structure of a new
film or freshly cleaved mica. crystal preparation of a membrane protein
216
Structural Study of Heme Proteins by Electron Microscopy
and to help confirm the molecular packing since the resulting images would be of the
model (31,51,80). actual biological molecules rather than the
distribution of heavy atom stains around
3.3. Low Dose Electron Microscopy them. There are several problems in achiev-
ing this goal, however, beginning with the
The challenges of specimen preparation low contrast afforded by biological speci-
for electron microscopy are compounded mens and by their sensitivity to exposure to
by the sensitivity of biological specimens to high energy electrons. Through the use of
electron irradiation. The exposure required low dose techniques, one can record elec-
to record a single image at moderate reso- tron micrographs of unstained specimens
lution, approximately 1 nm, can be as high at electron doses low enough to minimize
as 100 to 300 electrons/Å2. However, mea- radiation damage. The disadvantage is that
surements of electron damage at much although these images may technically be
lower exposures paint a gloomy picture for high resolution, the signal-to-noise ratio
prospects of achieving even this modest
(S/N) is very low, often less than 1.0, and
resolution. An exposure of even 20 to 30
cannot be interpreted. The S/N can be
electrons/Å2 results in loss of 20% to 30%
increased by averaging many images of
of the mass of a typical biological specimen
identical structures such as the unit cells of
(20), exposure of less than 10 electrons/Å2
a crystal; statistically the S/N is increased
disrupts the higher order features of a pro-
tein crystal (38), and an electron dose of by a factor equal to the square root of the
approximately 0.5 electrons/Å2 is sufficient number of structures averaged. This can be
to inactivate enzymes (39). The primary a very powerful tool in the case of 2D crys-
function of stain is to increase the contrast tals, where a very small area might contain
of biological specimens so lower electron 100 unit cells giving an increase in the S/N
doses can be used to achieve useful images. of tenfold. A more typical situation would
Furthermore, heavy atom stains are more be a crystal containing several thousand
resistant to damage by electron irradiation. unit cells giving and an increase in S/N of
Nevertheless, studies in the 1970s demon- thirty- to fiftyfold. This was first demon-
strated the efficacy of recording electron strated by Unwin and Henderson with
micrographs using minimal electron expo- their images of unstained purple mem-
sure even for negatively stained specimens brane containing thousands of bacteri-
(77,83). Now, most electron microscopes orhodopsin molecules; these electron
allow one to focus and correct astigmatism micrographs appear featureless, but the
on an area of the specimen grid adjacent to averaged image was clearly interpreted at 7
the specimen and then record an image of Å resolution as resulting from the presence
the specimen, exposing it to only the elec- of transmembrane α-helices (78).
trons required to expose the photographic The remaining problem is how to pre-
film. This procedure is absolutely essential pare unstained specimens for the high vacu-
when recording images of unstained speci- um conditions in an electron microscope.
mens that are much more sensitive to elec- Unwin and Henderson dried their purple
tron irradiation than are stained specimens. membrane specimens in a thin layer of 1%
glucose in order to surround them with a
3.4. Unstained Specimens hydrophilic substance that could also sup-
port them structurally. This approach has
Ideally, one would like to record elec- worked very well with purple membrane
tron micrographs of unstained specimens, and with a number of other examples, but
217
T.G. Frey
has the disadvantage that glucose has a (Pleasanton, CA, USA) and by Oxford
density very similar to protein and thus Instruments (Concord, MA, USA) for the
actually reduces the already low contrast popular side entry electron microscopes.
rather than increasing it. This was accept- The techniques of cryoelectron microscopy
able in the case of bacteriorhodopsin mole- have been described in an excellent review
cules as nearly all of the protein lies within by Dubochet et al. (21), and I will only
the lipid bilayer surrounded by the lower summarize the important points here. The
density hydrocarbon tails of the lipid mol- key to this technique is to freeze the speci-
ecules. Cytochrome oxidase crystals, how- men very rapidly in a thin layer of water.
ever, project much of their structure With freezing velocities above approxi-
beyond the lipid bilayer surface, and these mately 10 000 degrees/second, the water is
portions of the structure are virtually invisi- transformed to vitreous ice, a noncrys-
ble above 10 Å resolution if the crystals are talline ice form that has a density and
embedded in glucose (17,18,42). Better structure similar to liquid water. It is very
results have been obtained with auroth- difficult to freeze a thick specimen this
ioglucose, a glucose derivative containing rapidly, but a thin layer of water clinging to
gold atoms, or with glucose mixed with an electron microscope grid can be readily
uranyl acetate (79). These mixtures provide frozen to vitreous ice by plunging it into a
low resolution contrast of the hydrophilic suitable cryogen such as liquid propane or
domains of membrane proteins while liquid ethane cooled by liquid nitrogen.
embedding the structure in a hydrophilic The specimen grid is held in a pair of fine
substance. Kuhlbrandt and others have tweezers clamped into a simple device for
obtained excellent results using tannic acid plunging, and Procedure 4 is followed (see
rather than glucose (54). Figure 43 in Reference 21).
ment at a very low temperature where from the image. This is analogous to
movement is inhibited. The specimen may absorption contrast in the light microscope
be adsorbed to a thin carbon film as for and contrasts relatively low resolution
negative staining, or it may be suspended details; and (ii) phase contrast is generated
over the holes of a holey film. The latter when the phases of electrons are retarded as
method has the advantage that the speci- they pass through the specimen. The phase
men is not in contact with a solid support of these electrons are further modified by
prior to freezing, but the holey film has a the objective lens, and the extent of this
significantly smaller area suitable to record phase shift depends upon the:
images. The specimen grid should be • Angle of diffraction.
frozen immediately after blotting, but the
thin film of water supporting the specimen • Spherical aberration of the objective
may still dry significantly if the relative lens.
humidity of the environment is low. To • Focus of the objective lens.
minimize drying, one can maintain higher Thus, it is possible to control the
humidity around the specimen by: amount of phase shift of the diffracted
• Blowing humidified air across it (21). electrons by changing the focus of the
objective lens, and with an appropriate
• Freeze in a cold room where humidity
level of underfocus, some of the diffracted
is high.
electrons can be further phase-shifted by
• Use a specially constructed freezing approximately 90°, generating appropriate
device that incorporates a humidity phase contrast when combined with undif-
chamber (64). fracted electrons at the image plane. But
Adrian et al. have adapted this proce- for each choice of underfocus, only elec-
dure to incorporate heavy atom salts in trons diffracted at particular angles are
the vitrified water layer in order to phase-shifted by 90°, generating proper
increase the contrast of the frozen speci- contrast. Electrons diffracted at other
mens (1). A very important benefit of cry- angles are phase-shifted by smaller amounts,
oelectron microscopy is the reduction of generating less contrast, or are phase-shift-
electron beam damage at low tempera- ed in the wrong direction, generating
ture. In measurements of loss of higher inverted contrast.
resolution information as a function of Diffraction angle correlates with resolu-
electron irradiation, specimens at -170°C tion, and electrons diffracted at higher
can be exposed to 5 to 10 times the num- angles contain higher resolution informa-
ber of electrons as those at room tempera- tion. The changes in the phase of electrons
ture (38). as a function of their diffraction angle is
The low contrast provided by the rela- described by the contrast transfer function
tively small density differences between vit- (CTF) (23,26). In order to visualize indi-
reous ice and protein can be enhanced by vidual macromolecules, one must adjust
appropriate choice of focus of the objective the objective lens to a relatively large
lens. In the brightfield mode of a transmis- underfocus, one micron or more. This gen-
sion electron microscope, contrast is gener- erates contrast of lower resolution, features
ated by two mechanisms: (i) amplitude making the molecules visible, but may
contrast is generated when electrons are introduce contrast reversals at high resolu-
scattered by the specimen at a wide enough tion. Figure 3c is an optical diffraction pat-
angle to cause them to be intercepted by tern (equivalent to a plot of Fourier trans-
the objective aperture subtracting them form intensities) of the cytochrome oxidase
219
T.G. Frey
crystal in Figure 3a, and the effects of the reversed contrast, are shown in the plot of
CTF can be seen in the concentric rings of the CTF displayed as an insert in Figure 3c
high background noise. The regions of the on the same scale as the as the diffraction
Fourier transform, where phases have been pattern. By definition, CTF values greater
shifted in the wrong direction generating than zero represent incorrect phase shifts,
Figure 3. (a) An electron micrograph of a frozen hydrated crystal of cytochrome oxidase dimers; one unit cell is outlined. (b) A
Fourier-filtered image with dramatically increased S/N calculated from 5 electron micrographs similar to panel a. One unit cell is
outlined with unit cell axes of a = 100 Å and b = 125 Å. (c) Optical diffraction pattern of the crystal in panel a; the optical dif-
fraction pattern is equivalent to a plot of the intensities of the Fourier transform. The reciprocal lattice vectors, a* and b*, are indi-
cated. The inset is a plot of the phase CTF, χ(α), for this defocus shown on the same scale, and the zeros in the CTF are indicat-
ed by horizontal lines showing regions of minimal contrast in the diffraction pattern.
220
Structural Study of Heme Proteins by Electron Microscopy
and the circled diffraction spots lie within stained specimens, one normally records
rings of the diffraction pattern that have several images at different tilt angles for
been phase-shifted in the wrong direction. each specimen, since these specimens are
In order to calculate a high resolution more resistant to radiation damage.
image of a biological specimen, one must Unstained specimens are much more sensi-
correct for the effects of the CTF. Since dif- tive to electron irradiation, and generally,
ferent values of underfocus optimally con- only one high resolution image is recorded
trast features at different levels of resolu- from each. Specimens are more stable to
tion (different diffraction angles), it is irradiation at low temperature (10,38), so
sometimes advantageous to record more it is possible to record more than one
than one image of the same specimen, the micrograph from a single specimen at low
first at lower defocus for high resolution temperature, particularly if the highest res-
information, and the second at greater olution is not required.
defocus for lower resolution information
(11,28,73). 3.7. Specific Labeling
Figure 4. Lattice lines of the 3D Fourier transform of a 2D crystal. The positions of the reflections in the diffraction pattern
of an untilted crystal are shown as black ellipses. In the 3D Fourier transform, these reflections extend perpendicular to the plane
of the crystal as shown by the lines of periodically varying intensity. The Fourier transform of an electron micrograph of a crystal
that has been tilted is a central section of the 3D Fourier transform that intercepts the lattice lines at the points shown by the X’s.
222
Structural Study of Heme Proteins by Electron Microscopy
5a), consistent with the cytochrome c bind- plex is to compare structures of the intact
ing site in the atomic structure, determined complex with structures of subcomplexes.
by X-ray diffraction and from biochemical This approach was used in the study of the
studies (see section 4.1 and Figure 5b). mitochondrial cytochrome bc1 complex.
Other labeling studies have taken a dif- Weiss, Leonard, and coworkers crystallized
ferent approach, using heavy atom cluster Neurospora mitochondrial cytochrome c
molecules that have been modified to react reductase (cytochrome bc1or Complex III)
selectively with certain functional groups by reconstituting it with purified lipids and
of proteins, generally reactive sulfhydryl adsorbing excess detergent with Bio-Beads.
groups of cysteine residues (29). A special This produced crystals of the type shown in
issue of The Journal of Structural Biology is Figure 1c, although these were generally
devoted to results from this approach using formed in the two layers of a collapsed vesi-
gold cluster compounds (1999, volume cle giving two overlapping crystalline layers
127, issue 2). Crum et al. used a mono- (56,84). Their low resolution 3D recon-
maleimide derivative of an undecagold struction of the intact complex is shown in
cluster compound to label specifically Cys- Figure 6b. Hovmoller et al. subsequently
115 of cytochrome oxidase subunit III in formed crystals of the purified subcomplex
crystals of cytochrome oxidase dimers. lacking two large “core” proteins, and com-
They then identified the binding site by parison of the 3D structure with that of the
low dose cryoelectron microscopy of speci- intact complex allowed them to identify the
mens embedded in glucose and uranyl functional components as shown in Figure
acetate (16). 6 (46,47). The core subunits, which proba-
A different approach to identify the vari- bly function in facilitating the assembly of
ous components of a macromolecular com- the complex, were later purified, and their
Figure 5. A comparison of the structure of cytochrome oxidase monomers determined by electron crystallography (a) and by
X-ray crystallography (b). (a) A low resolution structure in projection calculated from electron micrographs of frozen hydrated
crystals of cytochrome oxidase monomers. The dark peak outlined in white contour lines is the position of cytochrome c bind-
ing calculated from difference images. (b) A ribbon diagram produced from the atomic coordinants calculated from the high res-
olution X-ray structure and displayed by the program RasMol. The cytochrome c binding site is placed between Cys-115 of sub-
unit III and the acidic residues of subunit II as determined biochemically.
223
T.G. Frey
low resolution structure was determined lution information. The simplest method
from helical aggregates confirming the to screen micrographs for quality is optical
assignment in Figure 6 (48). The cyto- diffraction, since the diffraction pattern is
chrome b6f complex found in the thylakoid the Fourier transform of the object and is a
membrane of chloroplasts has a function in diagnostic of several important image char-
photosynthesis very similar to that of acteristics. Formation of an optical diffrac-
cytochrome bc1 in mitochondria, but lacks tion pattern is readily accomplished with
the core subunits. The low resolution pro- an optical diffractometer consisting of a
jection structure of the cytochrome b6f laser (generally a 1–5 mW He/Ne laser),
complex purified from Chlamydomonas beam expansion–spatial filter, and a single
reinhardtii was determined by electron crys- diffraction lens to collect the diverging
tallography of 2D crystals grown by recon- beam and focus it to a point at some dis-
stitution and found to be very similar to the tance (24,65). When an electron micro-
cytochrome bc1 subcomplex lacking the graph is placed in the optical path just after
core subunits (8). the diffraction lens, the focused diffraction
pattern of the illuminated portion of the
4. DATA PROCESSING micrograph can be observed at the focus of
the lens. The undiffracted beam appears as
4.1. Selecting Micrographs — Optical a bright spot in the center of the diffraction
Diffraction pattern and is surrounded by the diffrac-
tion pattern of the object image. In the
Although data collection by low dose case of crystalline objects, the diffraction
electron microscopy is a critical element in pattern consists of peaks of light at the
determining the structure of a membrane points of a 2D lattice whose spacings are
protein, data processing is equally impor- the reciprocal of the crystal lattice; thus,
tant. The first step is to identify which of the lattice in the diffraction image is called
the many micrographs recorded are suit- the “reciprocal lattice”. The structural
able for further processing; generally, only a information common to all unit cells with-
minority of micrographs contain high reso- in the illuminated area of the micrograph
Figure 6. A comparison of the structure of cytochrome c reductase (cytochrome bc1 complex) determined by electron crystal-
lography and by X-ray crystallography. (a) A drawing interpreting the positions of 5 subunits in the dimeric complex with respect
to the lipid bilayer in the center. Core subunits I and II face the matrix space. (b) Balsa wood models of the low resolution struc-
tures determined by electron microscopy–crystallography of: (i) a subcomplex lacking core subunits I and II on the left, and (ii) the
intact complex. (iii) A ribbon diagram based upon the atomic coordinants of all subunits determined by X-ray crystallography.
224
Structural Study of Heme Proteins by Electron Microscopy
is contained at the points of the diffraction circles in the diffraction pattern (Figure
pattern’s reciprocal lattice, while nonperi- 3c), and the radii of the circles can be
odic noise is distributed throughout the used to calculate accurately the focus of
diffraction pattern. Although one can the objective lens. If the objective lens
obtain equivalent information by digitizing has residual astigmatism, the focus is
the electron micrograph and calculating its different in orthogonal directions, and
diffraction pattern, an optical diffractome- the zeroes produce concentric ellipses
ter accomplishes this instantaneously, rather than circles.
allowing one to move the micrograph
around in the beam to select the best area 4.2. Digitizing
quickly. One generally looks for two crite-
ria revealed by the optical diffraction In order to process electron micrographs
pattern: by computer, they must first be converted
to digital form. Although it is now possible
1. Does the image contain high resolution to purchase sensitive high resolution digi-
information about the crystal structure? tal cameras for transmission electron
The diffraction pattern is the Fourier microscopes, film is still the best media on
transform of the object illuminated, which to record low dose high resolution
representing its structure in frequency images. The most accurate scanners have
space, with points furthest from the been mechanical, based upon wrapping the
center representing higher frequency micrograph around a rotating drum or on
components contributing higher reso- precise movement in two dimensions, and
lution information. One therefore looks these are generally quite expensive. More
for micrographs whose diffraction pat- recently, high resolution digital cameras
terns display diffraction spots on the based upon charge-coupled device (CCD)
reciprocal lattice that extend relatively technology have become available, and
far from the origin of the diffraction these represent a suitable lower cost alter-
pattern. Once the diffraction constant native to mechanical scanners for many
of a particular optical diffractometer is applications. Whatever the device used,
calibrated, usually with a diffraction there are several factors to consider in digi-
object or grating of known spacing, the tizing an image, and these have been cov-
resolution of the information from each ered in earlier publications (19,26). The
micrograph can be calculated based first is the resolution one requires or
upon the distance from the origin of expects in the digitized image. In digital
the furthest diffraction spot. sampling of a continuous function (analog
2. Is the micrograph properly focused signal), one must sample the function at an
with astigmatism corrected? Character- interval that is one half the interval or reso-
istics of the transfer of information lution one wants to obtain in the digitized
from the object to the image are image; this is called the Nyquist sampling
described by the CTF as described rate. Thus, if one requires 10 Å informa-
below. A plot of the CTF in diffraction tion in a digital image, it must be sampled
space shows that it periodically passes (digitized) at 5 Å or smaller intervals. In
through zero at points determined by order to obtain very high resolution infor-
the wavelength of the electron wave, mation from a low dose image, one must
the spherical aberration of the objective digitize the image of a large 2D crystal, at
lens, and by the focus of the objective very small intervals, producing a very large
lens. The zeroes appear as concentric data file. Electron micrographs of 2D crys-
225
T.G. Frey
228
Structural Study of Heme Proteins by Electron Microscopy
have been shown to bind to opposite sur- and the maleimide group bound to the
faces of cytochrome c in the active site sulfhydryl of Cys-115.
(9,25). Although the position of Cys-115
of subunit III in the X-ray structure 5.2. Cytochrome c Reductase
appears to be quite distant from the peak
identified for the undecagold cluster com- The low resolution structure of the
pound bound to Cys-115 in the low reso- Neurospora cytochrome c reductase (cyto-
lution structure determined by electron chrome bc1 complex) determined by elec-
crystallography (16), one must remember tron microscopy is very similar the atomic
that the dimer observed in 2D crystals by structure of the beef heart mitochondrial
electron microscopy is much more com- enzyme determined by X-ray diffraction
pact than the dimer found in 3D crystals (49,85,88). As shown in Figure 6, the
by X-ray diffraction. In order to compare structure calculated from electron micro-
these 2 structures, the monomers in the X- graphs of 2D crystals displays an asymmet-
ray structure must each be moved approxi- ric distribution of mass protruding 30 Å on
mately 25 Å towards one another placing one side of the bilayer and 70 Å on the
Cys-115 of subunit III within the 15 Å other with a 50 Å domain within the lipid
length of the link between the undecagold bilayer (56,81). The smaller domain pro-
cluster observed by electron microscopy truding 30 Å was identified as the
Figure 7. A comparison of the structure of a cytochrome oxidase dimer determined by (a) electron crystallography at 15 Å res-
olution and (b) a ribbon diagram based upon the atomic structure determined by X-ray crystallography. The position of sub-
unit IV in the electron microscopy structure was determined by labeling with anti-IV Fabs. The structures are on the same scale.
230
Structural Study of Heme Proteins by Electron Microscopy
ABBREVIATIONS REFERENCES
231
T.G. Frey
6.Bhairi, S.M. 1997. Detergents: A Guide to the Proper- 23.Erickson, H.P. and A. Klug. 1971. Measurement and
ties and Uses of Detergents in Biological Systems. Cal- compensation of defocusing and aberrations by Fouri-
biochem-Novabiochem Corporation, San Diego. er processing of electron micrographs. Phil. Trans. R.
7.Bozzola, J.J. and L.D. Russell. 1999. Electron Soc. Lond. B 261:105-118.
Microscopy. Principles and Techniques for Biologists. 24.Erickson, H.P., W.A. Voter, and K. Leonard. 1978.
Jones and Bartlett, Boston. Image reconstruction in electron microscopy: enhance-
8.Bron, P., J.J. Lacapere, C. Breyton, and G. Mosser. ment of periodic structure by optical filtering. Meth-
1999. The 9 Å projection structure of cytochrome b6f ods Enzymol. 49:39-63.
complex determined by electron crystallography. J. 25.Ferguson-Miller, S., D.L. Brautigan, and E. Margo-
Mol. Biol. 287:117-126. liash. 1978. Definition of cytochrome c binding
9.Capaldi, R.A., V. Darley-Usmar, S. Fuller, and F. Mil- domains by chemical modification. III. Kinetics of
lett. 1982. Structural and functional features of the reaction of carboxydinitrophenyl cytochromes c with
interaction of cytochrome c with complex III and cytochrome c oxidase. J. Biol. Chem. 253:149-159.
cytochrome c oxidase. FEBS Lett. 138:1-7. 26.Frank, J. 1973. Computer processing of electron
10.Chiu, W. and T.W. Jeng. 1982. Electron radiation sen- micrographs, p. 215-274. In J.K. Koehler (Ed.),
sitivity of protein crystals. Ultramicroscopy 10:63-69. Advanced Techniques in Biological Electron Microscopy.
11.Conway, J.F. and A.C. Steven. 1999. Methods for Springer-Verlag, Berlin.
reconstructing density maps of “single” particles from 27.Frank, J., W. Chiu, and L. Degn. 1988. The charac-
cryoelectron micrographs to subnanometer resolution. terization of structural variations within a crystal field.
J. Struct. Biol. 128:106-118. Ultramicroscopy 26:345-360.
12.Costello, M.J., J. Escaig, K. Matsushita, P.V. Viitanen, 28.Frank, J. and P. Penczek. 1995. On the correction of
D.R. Menick, and H.R. Kaback. 1987. Purified lac the contrast function in biological electron microscopy.
permease and cytochrome o oxidase are functional as Optik 98:125-129.
monomers. J. Biol. Chem. 262:17072-17082. 29.Frey, P.A. and T.G. Frey. 1999. Synthesis of
13.Costello, M.J. and T.G. Frey. 1982. Membranous undecagold labeling compounds and their applications
cytochrome c oxidase. A freeze-fracture electron micro- in electron microscopic analysis of multiprotein com-
scopic analysis. J. Mol. Biol. 162:131-156. plexes. J. Struct. Biol. 127:94-100.
14.Crowther, R.A., D.J. De Rosier, and A. Klug. 1970. 30.Frey, T.G., S.H. Chan, and G. Schatz. 1978. Structure
The reconstruction of a three-dimensional structure and orientation of cytochrome c oxidase in crystalline
from projections and its application to electron membranes. Studies by electron microscopy and by
microscopy. Proc. R. Soc. A 317:319. labeling with subunit-specific antibodies. J. Biol.
15.Crowther, R.A. and U.B. Sleytr. 1977. An analysis of Chem. 253:4389-4395.
the fine structure of the surface layers from two strains 31.Frey, T.G. and T. Chang. 1986. The structure of mem-
of Clostridia, including correction for distorted images. brane bound cytochrome c oxidase. Ann. NY Acad.
J. Ultrastruct. Res. 58:41-49. Sci. 483:120-130.
16.Crum, J., K.J. Gruys, and T.G. Frey. 1994. Electron 32.Frey, T.G., M.J. Costello, and S.H. Chan. 1984.
microscopy of cytochrome c oxidase crystals: labeling Selective contrast in electron microscopy of crystalline
of subunit III with a monomaleimide undecagold clus- cytochrome oxidase. Ultramicroscopy 13:85-91.
ter compound. Biochemistry 33:13719-13726. 33.Frey, T.G., M.J. Costello, B. Karlsson, J.C. Hasel-
17.Deatherage, J.F., R. Henderson, and R.A. Capaldi. grove, and J.S. Leigh. 1982. Structure of the
1982. Relationship between membrane and cytoplas- cytochrome c oxidase dimer. Electron microscopy of
mic domains in cytochrome c oxidase by electron two-dimensional crystals. J. Mol. Biol. 162:113-
microscopy in media of different density. J. Mol. Biol. 130.
158:501-514. 34.Frey, T.G., L.A. Kuhn, J.S. Leigh, Jr., M.J. Costello,
18.Deatherage, J.F., R. Henderson and R.A. Capaldi. and S.H. Chan. 1985. Cytochrome oxidase: structural
1982. Three-dimensional structures of cytochrome c insights from electron microscopy and from secondary
oxidase vesicle crystals in negative stain. J. Mol. Biol. structure prediction. J. Inorg. Biochem. 23:155-162.
158:487-499. 35.Frey, T.G. and J.M. Murray. 1994. Electron
19.De Rosier, D.J. and P.B. Moore. 1970. Reconstruction microscopy of cytochrome c oxidase crystals.
of three-dimensional images from electron micro- Monomer-dimer relationship and cytochrome c bind-
graphs of structures with helical symmetry. J. Mol. ing site. J. Mol. Biol. 237:275-297.
Biol. 52:355-369. 36.Fukami, A. and K. Adachi. 1965. A new method of
20.Dubochet, J. 1975. Carbon loss during irradiation of preparation of a self-perforated micro plastic grid and
T4 bacteriophages and E. coli bacteria in electron its application. J. Electron Microsc. 14:112-118.
microscopes. J. Ultrastruct. Res. 52:276-288. 37.Fuller, S.D., R.A. Capaldi, and R. Henderson. 1979.
21.Dubochet, J., M. Adrian, J.J. Chang, J.C. Homo, J. Structure of cytochrome c oxidase in deoxycholate-
Lepault, A.W. McDowall, and P. Schultz. 1988. Cryo- derived two-dimensional crystals. J. Mol. Biol.
electron microscopy of vitrified specimens. Q. Rev. 134:305-327.
Biophys. 21:129-228. 38.Glaeser, R.M. and K.A. Taylor. 1978. Radiation dam-
22.Dykstra, M.J. 1992. Biological Electron Microscopy. age relative to transmission electron microscopy of bio-
Theory, Techniques, and Troubleshooting. Plenum logical specimens at low temperature: a review. J.
Press, New York. Microsc. 112:127-138.
232
Structural Study of Heme Proteins by Electron Microscopy
39.Hahn, M., J. Seredynski, and W. Baumeister. 1976. 55.Kühlbrandt, W. 1992. Two-dimensional crystalliza-
Inactivation of catalase monolayers by irradiation with tion of membrane proteins. Q. Rev. Biophys. 25:1-49.
100 keV electrons. Proc. Natl. Acad. Sci. USA 73:823- 56.Leonard, K., P. Wingfield, T. Arad, and H. Weiss. 1981.
827. Three-dimensional structure of ubiquinol:cytochrome c
40.Hayat, M.A. 1986. Transmission Electron Microscopy. reductase from Neurospora mitochondria determined by
Academic Press, Orlando. electron microscopy of membrane crystals. J. Mol. Biol.
41.Henderson, R., J.M. Baldwin, K.H. Downing, J. Lep- 149:259-274.
ault, and F. Zemlin. 1986. Structure of purple mem- 57.Lunsdorf, H. and E. Spiess. 1986. A rapid method of
brane from halobacterium halobium: recording, mea- preparing perforated supporting foils for the thin car-
surement and evaluation of electron micrographs at 3.5 bon films used in high resolution transmission electron
A resolution. Ultramicroscopy 19:147-178. microscopy. J. Microsc. 144:211-213.
42.Henderson, R., R.A. Capaldi, and J.S. Leigh. 1977. 58.Malhotra, A., P. Penczek, R.K. Agrawal, I.S.
Arrangement of cytochrome oxidase molecules in two- Gabashvili, R.A. Grassucci, R. Junemann, N.
dimensional vesicle crystals. J. Mol. Biol. 112:631-648. Burkhardt, K.H. Nierhaus, and J. Frank. 1998.
43.Henderson, R. and P.N. Unwin. 1975. Three-dimen- Escherichia coli 70 S ribosome at 15 A resolution by
sional model of purple membrane obtained by electron cryo-electron microscopy: localization of fMet-
microscopy. Nature 257:28-32. tRNAfMet and fitting of L1 protein. J. Mol. Biol.
44.Heuser, J. 1989. Protocol for 3-D visualization of mol- 280:103-116.
ecules on mica via the quick- freeze, deep-etch tech- 59.Mannella, C.A. 1984. Phospholipase-induced crystal-
nique. J. Electron Microsc. Tech. 13:244-263. lization of channels in mitochondrial outer mem-
45.Holser, W.T.Z. 1958. Point groups and plane groups branes. Science 224:165-166.
in a two-sided plane and their subgroups. Z. Kristallo- 60.Michel, H. 1991. Crystallization of Membrane Pro-
gr. 110:266-281. teins. CRC Press, Boca Raton.
46.Hovmoller, S., K. Leonard, and H. Weiss. 1981. 61.Miyazawa, A., Y. Fujiyoshi, M. Stowell, and N.
Membrane crystals of a subunit complex of mitochon- Unwin. 1999. Nicotinic acetylcholine receptor at 4.6
drial cytochrome reductase containing the cyto- A resolution: transverse tunnels in the channel wall. J.
chromes b and c1. FEBS Lett. 123:118-122. Mol. Biol. 288:765-786.
47.Hovmoller, S., M. Slaughter, J. Berriman, B. Karls- 62.Mohraz, M. and P.R. Smith. 1984. Structure of
son, H. Weiss, and K. Leonard. 1983. Structural (Na+,K+)-ATPase as revealed by electron microscopy
studies of cytochrome reductase. Improved membrane and image processing. J. Cell Biol. 98:1836-1841.
crystals of the enzyme complex and crystallization of a 63.Murray, J.M. and R. Ward. 1987. Preparation of holey
subcomplex. J. Mol. Biol. 165:401-406. carbon films suitable for cryo-electron microscopy. J.
48.Hutchinson, E.G., W. Tichelaar, G. Hofhaus, H. Elec. Microsc. Tech. 5:285-290.
Weiss, and K.R. Leonard. 1989. Identification and 64.Murray, J.M. and R. Ward. 1987. Principles for the
electron microscopic analysis of a chaperonin oligomer construction and operation of a device for rapidly
from Neurospora crassa mitochondria. EMBO J. freezing suspensions for cryo-electron microscopy. J.
8:1485-1490. Electron Microsc. Tech. 5:279-284.
49.Iwata, S., J.W. Lee, K. Okada, J.K. Lee, M. Iwata, B. 65.Salmon, E.D. and D. DeRosier. 1981. A surveying
Rasmussen, T.A. Link, S. Ramaswamy, and B.K. Jap. optical diffractometer. J. Microsc. 123:239-247.
1998. Complete structure of the 11-subunit bovine 66.Seki, S., H. Hayashi, and T. Oda. 1970. Studies on
mitochondrial cytochrome bc1 complex (see com- cytochrome oxidase. I. Fine structure of cytochrome
ments]. Science 281:64-71. oxidase-rich submitochondrial membrane. Arch.
50.Jap, B.K., M. Zulauf, T. Scheybani, A. Hefti, W. Biochem. Biophys. 138:110-121.
Baumeister, and U. Aebi. 1992. 2D crystallization: 67.Smith, P.R. and I.E. Ivanov. 1980. Surface reliefs
from art to science. Ultramicroscopy 46:45-84. computed from micrographs of isolated heavy metal
51.Jayaraman, U., T. Chang, T.G. Frey, and J.K. Blasie. shadowed particles. J. Ultrastruct. Res. 71:25-36.
1987. Electron density profile of two-dimensionally 68.Sosinsky, G.E. and G.A. Perkins. 2000. Electron
crystalline membranous cytochrome c oxidase at low crystallographic methods for investigating gap junc-
resolution. Biophys. J. 51:475-486. tion structure [In Process Citation]. Methods 20:140-
52.Karlsson, B., S. Hovmoller, H. Weiss, and K. 155.
Leonard. 1983. Structural studies of cytochrome 69.Stokes, D.L. and N.M. Green. 1990. Three-dimen-
reductase. Subunit topography determined by electron sional crystals of CaATPase from sarcoplasmic reticu-
microscopy of membrane crystals of a subcomplex. J. lum. Symmetry and molecular packing. Biophys. J.
Mol. Biol. 165:287-302. 57:1-14.
53.Kim, C.H., M. Radermacher, M. Kessel, J. Frank, 70.Suarez, M.D., A. Revzin, R. Narlock, E.S. Kempner,
and T.E. King. 1985. Three-dimensional reconstruc- D.A. Thompson, and S. Ferguson-Miller. 1984. The
tion of cytochrome oxidase vesicle crystals prepared by functional and physical form of mammalian
cholate solubilization. J. Inorg. Biochem. 23:163-169. cytochrome c oxidase determined by gel filtration,
54.Kühlbrandt, W. 1988. Structure of light-harvesting radiation inactivation, and sedimentation equilibrium
chlorophyll a/b protein complex from plant photosyn- analysis. J. Biol. Chem. 259:13791-13799.
thetic membranes at 7 A resolution in projection. J. 71.Taylor, K.A., L. Dux, S. Varga, H.P. Ting-Beall, and
Mol. Biol. 202:849-864. A. Martonosi. 1988. Analysis of two-dimensional crys-
233
T.G. Frey
tals of Ca2+-ATPase in sarcoplasmic reticulum. Meth- cytochrome-c reductase from Neurospora mitochondria
ods Enzymol. 157:271-289. and structure analysis by electron microscopy. Methods
72.Taylor, K.A. and R.M. Glaeser. 1976. Electron Enzymol. 126:191-201.
microscopy of frozen hydrated biological specimens. J. 82.Wikoff, W.R., G. Wang, C.R. Parrish, R.H. Cheng,
Ultrastruct. Res. 55:448-456. M.L. Strassheim, T.S. Baker, and M.G. Rossmann.
73.Toyoshima, C. and N. Unwin. 1988. Contrast trans- 1994. The structure of a neutralized virus: canine par-
fer for frozen-hydrated specimens: determination from vovirus complexed with neutralizing antibody frag-
pairs of defocused images. Ultramicroscopy 25:279- ment. Structure 2:595-607.
291. 83.Williams, R.C. and H.W. Fisher. 1970. Electron
74.Tsukihara, T., H. Aoyama, E. Yamashita, T. Tomiza- microscopy of Tobacco Mosaic virus under condi-
ki, H. Yamaguchi, K. Shinzawa-Itoh, R. Nakashima, tions of minimal beam exposure. J. Mol. Biol. 52:
R. Yaono, and S. Yoshikawa. 1995. Structures of metal 121-123.
sites of oxidized bovine heart cytochrome c oxidase at 84.Wingfield, P., T. Arad, K. Leonard, and H. Weiss.
2.8 A [see comments]. Science 269:1069-1074. 1979. Membrane crystals of ubiquinone:cytochrome c
75.Tsukihara, T., H. Aoyama, E. Yamashita, T. Tomiza- reductase from Neurospora mitochondria. Nature
ki, H. Yamaguchi, K. Shinzawa-Itoh, R. Nakashima, 280:696-697.
R. Yaono, and S. Yoshikawa. 1996. The whole struc- 85.Xia, D., C.A. Yu, H. Kim, J.Z. Xia, A.M. Kachurin,
ture of the 13-subunit oxidized cytochrome c oxidase L. Zhang, L. Yu, and J. Deisenhofer. 1997. Crystal
at 2.8 A [see comments]. Science 272:1136-1144. structure of the cytochrome bc1 complex from bovine
76.Unger, V.M., N.M. Kumar, N.B. Gilula, and M. Yea- heart mitochondria [published erratum appears in Sci-
ger. 1999. Three-dimensional structure of a recombi- ence 1997 Dec 19;278(5346):2037]. Science 277:60-
nant gap junction membrane channel. Science 283: 66.
1176-1180. 86.Yeager, M., V.M. Unger, and A.K. Mitra. 1999. Three-
77.Unwin, P.N.T. 1974. Electron microscopy of the dimensional structure of membrane proteins deter-
stacked disk aggregate of Tobacco Mosaic Virus pro- mined by two-dimensional crystallization, electron cry-
tein. II. The influence of electron irradiation on the omicroscopy, and image analysis. Methods Enzymol.
stain distribution. J. Mol. Biol. 87:657-670. 294:135-180.
78.Unwin, P.N.T. and R. Henderson. 1975. Molecular 87.Zeppenzauer, M. 1971. Formation of large crystals, p.
structure determination by electron microscopy of 253. In W.B. Jacoby (Ed.), Methods of Enzymology.
unstained crystalline specimens. J. Mol. Biol. 94:425- Academic Press, New York.
440. 88.Zhang, Z., L. Huang, V.M. Shulmeister, Y.I. Chi,
79.Valpuesta, J.M., R. Henderson, and T.G. Frey. 1990. K.K. Kim, L.W. Hung, A.R. Crofts, E.A. Berry, and
Electron cryo-microscopic analysis of crystalline S.H. Kim. 1998. Electron transfer by domain move-
cytochrome oxidase. J. Mol. Biol. 214:237-251. ment in cytochrome bc1. Nature 392:677-684.
80.Vanderkooi, G., A.E. Senior, R.A. Capaldi, and H. 89.Zhu, J., P.A. Penczek, R. Schroder, and J. Frank.
Hayashi. 1972. Biological membrane structure. 3. The 1997. Three-dimensional reconstruction with contrast
lattice structure of membranous cytochrome oxidase. transfer function correction from energy-filtered cryo-
Biochim. Biophys. Acta 274:38-48. electron micrographs: procedure and application to the
81.Weiss, H., S. Hovmoller, and K. Leonard. 1986. 70S Escherichia coli ribosome. J. Struct. Biol. 118:197-
Preparation of membrane crystals of ubiquinol- 219.
234
Analysis and Reconstitution of
10 Chlorophyll–Proteins
235
H. Paulsen and V.H.R. Schmid
mild nondenaturing conditions (Table 1). native and recombinant Chl-protein com-
However, other nonionic detergents like plexes usually is the analysis of the pigment
Triton® X-100 or ionic detergents such as and protein components. Very often, the
lauryldimethylamine oxide (LDAO) and first step consists of an extraction of the
SDS (or lithiumdodecylsulfate, LDS, complexes by adding 80% acetone, which
where SDS would precipitate at low tem- denatures and precipitates the protein moi-
peratures) are also in use. SDS or LDS, in ety, whereas the noncovalently bound pig-
combination with nonionic detergents, ments end up in the supernatant and can
allows the control of the denaturing strin- subsequently be separated and analyzed. If
gency in partially denaturing polyacry- Chl-protein complexes have been separat-
lamide gels and, thus, is useful in separat- ed by polyacrylamide gel electrophoresis,
ing more stable complexes from less stable acetone does not extract pigments from gel
ones (70). slices. This can be done quantitatively by
using 2-butanol (52).
2.1.2. Purification The pigments in the 80% acetone
extract can be analyzed and quantified
The overview of separation procedures either spectrophotometrically or by high-
given in Table 2 is far from being complete, performance liquid chromatography
but rather gives a few examples of where (HPLC). For the quantification of mix-
the procedures have been used. tures of Chl a and Chl b in 80% acetone
The majority of the procedures used for (buffered to pH 7.8), the algorithm by
the isolation of native Chl-protein com- Porra et al. (73) is frequently used:
plexes includes ultracentrifugation steps in concentrationChl a (µg/mL) = 12.25 A663.6 - 2.55 A646.6
a density gradient or various chromato- concentrationChl b (µg/mL) = 20.31 A646.6 - 4.91 A663.6
graphic techniques or both; these tech- where A646.6 and A663.6 are the
niques are also useful to isolate recombi- absorbances at the given wavelengths at 1
nant Chl-protein complexes. Some of the cm pathlength, minus the absorbance at
gel electrophoretic techniques tend to have 750 nm. Some alternative algorithms are
a more strongly denaturing effect on pig- discussed in the same reference. For the
ment–protein complexes and, therefore, determination of bacteriochlorophyll a in
may lead to the loss of pigments. However, acetone:methanol (7:2, vol/vol), a molar
gel electrophoresis provides a rapid means absorption coefficient of ε770 = 76
to separate in vitro reconstituted Chl-pro- 000/Mcm (19) has been measured. For the
tein complexes from unbound pigments rough quantification of α- and β-
and, thus, is useful for analyzing reconsti- carotenoids and their derivatives, the spe-
tution products. Therefore, gel elec- cific absorption coefficient of Davies (22)
trophoresis along with sucrose gradient is very useful: ε440 = 240 L/gcm.
centrifugation and ion exchange tech- Numerous HPLC protocols for separat-
niques will be described in some more ing Chls and carotenoids have been devel-
detail in section 3.4. oped (29,34,65,93). One of the simplest
ones is a gradient from 80% to 100% ace-
2.2. Analysis of Chlorophyll–Protein tone. Usually, the conversion of peak areas
Complexes to pigment quantities is calibrated on the
basis of absorption coefficients such as
2.2.1. Biochemical Characterization those given above.
Quantification of the precipitated and
The first biochemical characterization of redissolved protein can be performed by
237
Table 1. Detergents Used for Solubilizing Chl-Proteins
238
Sucrose-gradient PSI 57
ultracentrifugation
PSII core and antenna subunits 20
PSII core complex and PSII-LHCII 37
super complex
LH1 and LH2 from R. molischianum 92
LHCI 63
LHCII 13,48
69,84
2,45 containing SDS or LDS, co-pre-
36
46
54
47
31
21
5
3
cipitation of the dodecylsulfate
salt can be reduced by acidifying
the solution to pH 4.0–5.0 with
acetic acid. Note that acidification
is incompatible with subsequent
Chl analysis in the supernatant, as
Chl-protein complexes of thylakoids
His-tagged bacterial photosynthetic
LHCII subunits
PSII LHCs
interaction chromatography on
Preparative flat-bed isoelectric focussing
ments, the supernatant is treated in differ- 6. The aliquots are dried in a nitrogen
ent ways. stream and can be stored for months to
years at -20°C under nitrogen or argon.
❖ Procedure 1. Total Pigment Extract For the isolation of individual Chls and
carotenoids, the following procedure is
1. The pigment solution in acetone is useful:
mixed with 0.25 volumes diethyl ether
in a separating funnel. ❖ Procedure 2. Isolation of Individual
2. To improve phase separation, solid Pigments
NaCl (e.g., 35 g to 600-mL solution) is
added and dissolved by gently moving 1. The acetonic pigment solution is
the funnel. If the lower acetone phase cooled to 0°C.
remains colored, the ether extraction 2. Dioxane is added to give a final con-
should be repeated, adding more NaCl centration of about 15% (vol/vol) (42).
if phase separation is poor. 3. To the homogenous solution 0.16 vol-
3. Combine and dry the ether phases, umes of water is added drop-wise
either by the addition of solid NaCl or under constant stirring, and the resul-
by placing the ether solution in a tant solution is kept on ice for 1 hour
-20°C freezer for at least 1 hour. Then without further stirring.
ice or NaCl can be removed by filtra- 4. Aggregated Chls (as well as pheophytin
tion through a sintered glass funnel and β-carotene) are collected by cen-
(precooled to -20°C if ice crystals are to trifugation (15 000× g, for 10 min),
be removed). and the pelleted pigments are used for
4. Evaporate the ether in a rotary evapora- column chromatographic separation of
tor or nitrogen stream. the Chls and β-carotene. If the super-
5. Pigments are dissolved in acetone, natant still contains a substantial
quantified on the basis of their Chl con- amount of the Chl originally present,
tent (see section 2.2.1), and aliquoted. more water may be added drop-wise,
and the additional precipitate collect-
ed. If the addition of water is too exces-
sive or too fast, xanthophylls will also
aggregate and contaminate the crude
Chl preparation. The supernatant is
kept for xanthophyll isolation.
5. For the separation of individual Chls
(and xanthophylls), a reversed phase
C18 material [e.g., 55–105 µ-Bonda-
pak (Waters, Milford, MA, USA)] with
acetone–water mixtures as the mobile
phase is suitable. Chromatography can
be done either at low pressure or by
Figure 1. Partially denaturing gel electrophoresis of recon- using an HPLC apparatus. A column
stitution mixtures with LHCI- (Lhca4; lanes a and b) or volume of at least 4 mL (low pressure
LHCII- (Lhcb1; lanes c and d) apoprotein. The mixtures
were subjected to either the freeze-thaw (lanes a and c) or the
development) or 0.8 mL (high pressure
detergent exchange method (lanes b and d). The resolution processing) is recommended for each
of bands with monomeric complexes (M) and free pigments milligram of raw pigments applied.
(FP) is visible.
243
H. Paulsen and V.H.R. Schmid
Table 3. Comparison of the Experimental Steps of the Two Most Commonly Used Reconstitution
Techniques
Detergent Exchange
Freeze-Thaw Method Method
Protein denaturation by + +
LDS and heating
Addition of OG - +
Application to nondenaturing + +
polyacrylamide gel electrophoresis or
sucrose density gradient centrifugation
6. Preequilibrate the column with 86% 12. For isolation of individual xantho-
acetone. phylls, the pigments in the supernatant
7. Dissolve the Chl pellet in a small vol- of the dioxane precipitation (step 4) are
ume of 86% acetone. ether-extracted and dried as described
8. To remove any particulate material the for total pigment extracts (steps 1–4).
solution is centrifuged (15 000× g for 13. The residue is dissolved in ethanol
10 min). and made up to 8% KOH by addi-
9. The supernatant is loaded on the tion of 0.1 volumes from an 80%
column. (wt/vol) stock solution in water (22).
10. Elution of Chl b (green) and Chl a The mixture is overlaid with nitrogen
(blue-green) is done with acetone of and kept overnight in a tightly
that concentration. For the elution of capped bottle at 55°C in the dark.
pheophytin (brown) and β-carotene The saponification with KOH con-
(orange), the acetone concentration has verts residual Chl and lipids into
to be raised to 90% and 95%, respec- more hydrophilic products.
tively. To obtain pure pigments, only 14. Xanthophylls are extracted by ether as
the central part of the individual described for total pigment extract
pigment fractions should be collected. (steps 1–2).
11. The eluted pigments are transferred to 15. The ether fraction is washed twice with
ether, dried, quantitated, and stored as 3 volumes of water.
described for total pigment extracts 16. The xanthophylls are then obtained
(steps 1–6). from the ether solution as described
244
Analysis and Reconstitution of Chlorophyll–Proteins
above (total pigment extracts, steps described by Nagai and Thøgersen (58), is
3–4). They can be either used as “total described here in detail.
xanthophylls” (see section 2.2.1 for
absorption coefficient for carotenoids) ❖ Procedure 3. Isolation of
for reconstitution or, alternatively, sub- Recombinant LHC-Apoproteins
jected to a column chromatography
procedure as outlined for the Chls 1. Start with an 5 mL overnight culture
(step 5) to isolate the individual xan- [Luria-Bertani medium supplemented
thophylls. with 100 µg ampicillin/mL (LB-Amp)]
17. Proceed as in steps 6 to 9, but use of Eschericia coli that harbors the
74% acetone for preequilibration and respective expression plasmid.
as solvent. 2. Inoculate a 250-mL Erlenmeyer flask
18. Isocratic elution is started with 74% containing 100 mL LB-Amp with 1 mL
acetone. Following elution of neoxan- of the overnight culture and grow the
thin and violaxanthin the acetone con- cells on a rotary shaker at 170
centration is raised to 80% acetone for rounds/minute and 37°C to mid-log
the elution of lutein. phase (OD550 of around 0.5), which
takes about 2 hours.
The purity of the individual pigments is
3. Induce overexpression by addition of a
most conveniently tested by analytical
1 M isopropyl-β-D-thiogalactoside
HPLC or by thin-layer chromatography
solution to 1 mM final concentration.
(TLC) with, e.g., RP18-plates (Merck,
The cells are cultivated for another 4 to
Darmstadt, Germany) and methanol as
5 hours under the same conditions.
solvent. For the quantification of Chls see
section 2.2.1; carotenoids can be quanti- 4. Harvest the cells by centrifugation
fied by means of the absorption coefficients (5 min at 5 000× g). The cell pellets are
of Davies (23). The specific absorption either stored at -20°C or processed fur-
coefficients, for example for an ethanolic ther immediately.
solution, are ε439 = 224.3 L/gcm (neoxan- 5. Suspend the cell pellet in 500 µL lysis
thin), ε443 = 255 L/gcm (violaxanthin), buffer [50 mM Tris-HCl (pH 8.0),
ε445 = 255 L/gcm (lutein), and ε453 = 262 25% (wt/vol) sucrose, and 1 mM
L/gcm (β-carotene). EDTA) and bring it to 800 µL with
lysis buffer.
3.3.2. Protein Isolation 6. Add 200 µL lysozyme from a 1%
(wt/vol, in lysis buffer) solution which
Originally, authentic proteins isolated is freshly prepared each time. The solu-
from the membranes of the corresponding tion is mixed well and incubated at
species were extracted and used for recon- room temperature for 30 minutes.
stitution experiments (72). Meanwhile, 7. Add 10 µL DNase I solution [0.1%
however, cDNAs derived from the genes (wt/vol) solution in 20 mM Tris-HCl
are cloned into expression vectors, e.g., (pH 8.0), 50 mM NaCl, 1 mM dithio-
pDS-vectors (12), in order to obtain large threitol (DTT), and 50% (vol/vol)
quantities of the desired protein. All bacte- glycerol; this solution can be stored at
rially expressed LHC-apoproteins are accu- -20°C], together with 10 µL of 0.1 M
mulated in the form of insoluble inclusion MnCl2 and 1 M MgCl2. Incubation
bodies. Therefore, inclusion body isolation, for another 30 minutes at room tem-
which follows mainly the procedure perature follows.
245
H. Paulsen and V.H.R. Schmid
247
H. Paulsen and V.H.R. Schmid
Laemmli system (68), and for LHCI, we For both gel types, the same running
prefer a modified Neville system (85). For buffer with 25 mM Tris, 196 mM glycine,
most applications, analytical slab minigels and 0.1% (wt/vol) LDS (only required in
are a good choice. The 30% (wt/vol) the cathode buffer) is used. The buffer is
acrylamide stock solution we use has an best prepared as a tenfold stock solution
acrylamide: N,N′-methylenebisacrylamide and diluted before use.
ratio of 30. Prior to electrophoresis, the gel and the
Laemmli Gel buffer should be cooled to 4°C. After
Prepare the required volume of the removal of the comb, the gel pockets are
resolving gel solution with 12% (wt/vol) rinsed with running buffer. Samples equiv-
acrylamide, 400 mM Tris-HCl (pH 8.8) alent to 10 µg Chl are applied to 4-mm-
and 10% (vol/vol) glycerol. While the solu- wide wells of a 1-mm-thick gel. Elec-
tion is stirred, ammonium persulfate (APS) trophoresis is either conducted with a
[10% (wt/vol) stock solution] and constant voltage of 60 V (Laemmli) or
N,N,N′,N′-tetramethylethylenediamine with a constant current of 0.1 mA/ mm2 of
(Temed) are added to final concentrations gel cross-section (Neville gel) for about 2.5
of 0.07% (wt/vol) and 0.05% (vol/vol), hours. Afterwards, the gel sandwich is dis-
respectively. The gel solution is poured assembled, the gel documented, and indi-
between the assembled glass plates up to vidual bands can be excised and the protein
about 7 mm below that point where the eluted for further characterization (see sec-
bottom of the comb will be located. The tion 2.2).
gel surface is overlaid with a thin layer of
water to obtain homogenous polymeriza- 3.4.2. Sucrose Density Gradients
tion and a plane gel surface. After 1 hour,
the gel should be polymerized, the water is Compared to nondenaturing gel elec-
poured off, and, if necessary, the gel surface trophoresis, ultracentrifugation through
is dabbed with filter paper. The stacking sucrose density gradients is a more gentle
gel solution with 4.5% (wt/vol) acry- method for isolating labile LHC. More-
lamide, 130 mM Tris-HCl (pH 6.8), 10% over, this method allows the isolation of
(vol/vol) glycerol, 0.05% (wt/vol) APS, sufficient material for further analyses and
and 0.05% (vol/vol) Temed is poured on has the advantage that the green band col-
top of the resolving gel, and the comb is lected from the centrifuge tube can be used
inserted by gently pushing it down starting immediately without the need to extract it
from one side. from a gel.
Neville Gel Sucrose gradients can be formed either
The resolving gel is composed of 12% by means of a gradient mixer in combina-
(wt/vol) acrylamide, 424 mM Tris-HCl tion with a peristaltic pump or, more con-
(pH 9.1), and 10% (wt/vol) sucrose. Poly- veniently, by the freeze-thaw method
merization is initiated by the addition of described by Bassi and Simpson (6). For
10% APS solution and Temed to final con- the latter method, a solution with 0.5 M
centrations of 0.03% (wt/vol) and 0.075% sucrose, 5 mM Tricine-NaOH (pH 7.8),
(vol/vol), respectively. The stacking gel is and 0.1% (wt/vol) LM is filled in the cen-
composed of 4% (wt/vol) acrylamide, 54 trifuge tubes. The tubes are placed in a
mM Tris-H2SO4 (pH 6.1), 10% (wt/vol) -20°C freezer. Three hours before sample
sucrose, and polymerization is initiated by application, the tubes are transferred to a
final concentrations of 0.06% (wt/vol) refrigerator and kept there until completely
APS and 0.075% (vol/vol) Temed. thawed. Subsequently, the upper tenth of
248
Analysis and Reconstitution of Chlorophyll–Proteins
4. CONCLUDING REMARKS
Figure 2. Sucrose gradient fractionation of reconstitution Several aspects have been mentioned in
mixtures containing the two apoproteins of LHCI-730. FP, the previous paragraphs of how reconstitu-
free pigments; m-LHCI, monomeric LHCI; d-LHCI, dimer-
ic LHCI (LHCI-730). tion of Chl a/b-protein complexes can be
249
H. Paulsen and V.H.R. Schmid
250
Analysis and Reconstitution of Chlorophyll–Proteins
11.Büchel, C. and C. Wilhelm. 1993. Isolation and char- 25.Delepelaire, P. and N.H. Chua. 1981. Electrophoretic
acterization of a photosystem I-associated antenna purification of chlorophyll a/b protein complexes from
(LHC I) and a photosystem I-core complex from the Chlamydomonas reinhardtii and spinach and analysis of
chlorophyll c-containing alga Pleurochloris meiringensis their polypeptides. J. Biol. Chem. 256:9300-9307.
(Xanthophyceae). J. Photochem. Photobiol. B Biol. 26.Douady, D., B. Rousseau, and L. Caron. 1994. Fucox-
20:87-93. anthin chlorophyll a/c light-harvesting complexes of
12.Bujard, H., R. Gentz, M. Lanzer, D. Stueber, M. Laminaria saccharina — partial amino acid sequences
Mueller, I. Ibrahimi, M.T. Haeuptle, and B. Dob- and arrangement in thylakoid membranes. Biochem-
berstein. 1987. A T5-promoter-based transcription- istry 33:3165-3170.
translation system for the analysis of protein expres- 27.Durnford, D.G. and B.R. Green. 1994. Characteriza-
sion in vivo and in vitro. Methods Enzymol. tion of the light harvesting proteins of the chromo-
155:416-433. phytic alga, Olisthodiscus luteus (Heterosigma carterae).
13.Burke, J.J., C.L. Ditto, and C.J. Arntzen. 1978. Biochim. Biophys. Acta 1184:118-126.
Involvement of the light-harvesting complex in cation 28.Flachmann, R. and W. Kühlbrandt. 1996. Crystalliza-
regulation of excitation energy distribution in chloro- tion and identification of an assembly defect of recom-
plasts. Arch. Biochem. Biophys. 187:252-263. binant antenna complexes produced in transgenic
14.Butler, P.J.G. and W. Kühlbrandt. 1988. Determina- tobacco plants. Proc. Natl. Acad. Sci. USA 93:14966-
tion of the aggregate size in detergent solution of the 14971.
light-harvesting chlorophyll a/b-protein complex from 29.Francis, G.W., B. Huseby, and O.M. Andersen. 1993.
chloroplast membranes. Proc. Natl. Acad. Sci. USA An improved HPLC system for the analysis of photo-
85:3797-3801. synthetic pigments. Chromatographia 35:189-192.
15.Camm, E.L. and B.R. Green. 1980. Fractionation of 30.Fromme, P., H.T. Witt, W.D. Schubert, O. Klukas,
thylakoid membranes with the nonionic detergent W. Saenger, and N. Krauss. 1996. Structure of photo-
octyl-β-D-glucopyranoside. Plant Physiol. 66:428- system I at 4.5 angstrom resolution: a short review
432. including evolutionary aspects. Biochim. Biophys. Acta
16.Camm, E.L. and B.R. Green. 1982. The effects of 1275:76-83.
cations and trypsin on extraction of chlorophyll-pro- 31.Funk, C., W.P. Schröder, A. Napiwotzki, S.E. Tjus,
tein complexes by octyl glucoside. Arch. Biochem. Bio- G. Renger, and B. Andersson. 1995. The PSII-S pro-
phys. 214:563-572. tein of higher plants: a new type of pigment-binding
17.Cammarata, K.V., F. Plumley, and G.W. Schmidt. protein. Biochemistry 34:11133-11141.
1990. Reconstitution of light-harvesting complexes: a 32.Gantt, E., F.X. Cunningham, B. Grabowski, and S.
single apoprotein binds Chla, Chlb and xanthophylls, Tan. 1998. Relatedness of caroteno-chlorophyll anten-
p. 341-344. In M. Baltscheffsky (Ed.), Current na complexes in algae and plants, p. 239-247. In G.
Research in Photosynthesis, Vol. 2. Kluwer Academic Garab (Ed.), Photosynthesis: Mechanism and Effects,
Publishers, Dordrecht. Vol. I. Kluwer Academic Publishers, Dordrecht.
18.Cammarata, K.V. and G.W. Schmidt. 1992. In-vitro 33.Garab, G., J. Kieleczawa, J.C. Sutherland, C. Busta-
reconstitution of a light-harvesting gene product–dele- mante, and G. Hind. 1991. Organization of pig-
tion mutagenesis and analyses of pigment binding. ment–protein complexes into macrodomains in the
Biochemistry 31:2779-2789. thylakoid membranes of wild-type and chlorophyll b-
19.Clayton, R.K. and B.J. Clayton. 1981. B 850 pig- less mutant of barley as revealed by circular dichroism.
ment–protein complex of Rhodopseudomonas Photochem. Photobiol. 54:273-281.
sphaeroides. Extinction coefficients, circular dichroism, 34.Gilmore, A.M. and H.Y. Yamamoto. 1991. Resolution
and the reversible binding of bacteriochlorophyll. Proc. of lutein and zeaxanthin using a non-endcapped, lightly
Natl. Acad. Sci. USA 78:5583-5587. carbon-loaded C18 high-performance liquid chro-
20.Dainese, P. and R. Bassi. 1991. Subunit stoichiom- matographic column. J. Chromatogr. 543:137-145.
etry of the chloroplast photosystem II antenna 35.Giuffra, E., D. Cugini, R. Croce, and R. Bassi. 1996.
system and aggregation state of the component Reconstitution and pigment-binding properties of
chlorophyll a/b binding proteins. J. Biol. Chem. recombinant CP29. Eur. J. Biochem. 238:112-120.
266: 8136-8142. 36.Goldsmith, J.O. and S.G. Boxer. 1996. Rapid isola-
21.Dainese, P., G. Hoyer-Hansen, and R. Bassi. 1990. tion of bacterial photosynthetic reaction centers with
The resolution of chlorophyll a/b-binding proteins by a an engineered poly-histidine tag. Biochim. Biophys.
preparative method based on flat bed isoelectric focus- Acta 1276:171-175.
ing. Photochem. Photobiol. 51:693-703. 37.Hankamer, B., J. Nield, D. Zheleva, E. Boekema, S.
22.Davies, B.H. 1965. Analysis of carotenoid pigments, p. Jansson, and J. Barber. 1997. Isolation and biochemi-
489-532. In T.W. Goodwin (Ed.), Chemistry and Bio- cal characterisation of monomeric and dimeric photo-
chemistry of Plant Pigments. Academic Press, New York. system II complexes from spinach and their relevance
23.Davies, B.H. 1976. Carotenoids, p. 38-165. In T.W. to the organisation of photosystem II in vivo. Eur. J.
Goodwin (Ed.), Chemistry and Biochemistry of Plant Biochem. 243:422-429.
Pigments, Vol. 2. Academic Press, London. 38.Haworth, P., J.L. Watson, and C.J. Arntzen. 1983.
24.Deisenhofer, J., O. Epp, K. Miki, R. Huber, and H. The detection, isolation, and characterisation of a
Michel. 1985. Structure of the protein subunits in the light-harvesting complex which is specifically associat-
photosynthetic reaction center of Rhodopseudomonas ed with photosystem I. Biochim. Biophys. Acta
viridis at 3 Å resolution. Nature 318:618-624. 724:151-158.
251
H. Paulsen and V.H.R. Schmid
39.Hiller, R.G., C.D. Scaramuzzi, and J. Breton. 1992. reconstituted algal chlorophyll a/b-binding light-har-
The organisation of photosynthetic pigments in a Crypto- vesting complexes of Chlorella fusca with different pig-
phyte alga — a linear dichroism study. Biochim. Bio- ment compositions and pigment–protein stoichiome-
phys. Acta 1102:360-364. tries. Photosynth. Res. 49:71-81.
40.Hobe, S., R. Förster, J. Klingler, and H. Paulsen. 56.Mullet, J.E. and C.J. Arntzen. 1980. Simulation of
1995. N-proximal sequence motif in light-harvesting grana stacking in a model membrane system. Media-
chlorophyll a/b-binding protein is essential for the tion by a purified light-harvesting pigment–protein
trimerization of light-harvesting chlorophyll a/b com- complex from chloroplasts. Biochim. Biophys. Acta
plex. Biochemistry 34:10224-10228. 589:100-117.
41.Hobe, S., S. Prytulla, W. Kühlbrandt, and H. 57.Mullet, J.E., J.J. Burke, and C.J. Arntzen. 1980.
Paulsen. 1994. Trimerization and crystallization of Chlorophyll proteins of photosystem I. Plant Physiol.
reconstituted light-harvesting chlorophyll a/b complex. 65:814-822.
EMBO J. 13:3423-3429. 58.Nagai, K. and H.C. Thøgersen. 1987. Synthesis and
42.Iriyama, K., N. Ogura, and A. Takamiya. 1974. A sequence-specific proteolysis of hybrid proteins pro-
simple method for extraction and partial purification duced in E. coli. Methods Enzymol. 153:461-481.
of chlorophyll from plant material, using dioxane. J. 59.Neville, D.M. 1971. Molecular weight determination
Biochem. 76:901-904. of protein-dodecyl sulfate complexes by gel elec-
43.Kamimura, Y., T. Mori, T. Yamasaki, and S. Katoh. trophoresis in a discontinous buffer system. J. Biol.
1997. Isolation, properties and a possible function of a Chem. 246:6328-6334.
water soluble chlorophyll a/b protein from Brussels 60.Nishio, N. and H. Satoh. 1997. A water-soluble
sprouts. Plant Cell Physiol. 38:133-138. chlorophyll protein in cauliflower may be identical to
44.Kan, K.S. and J.P. Thornber. 1976. The light-harvest- Bnd22, a drought-induced, 22-kilodalton protein in
ing chlorophyll a/b-protein of Chlamydomonas rein- rapeseed. Plant. Physiol. 115:841-846.
hardtii. Plant Physiol. 57:47-53. 61.Nußberger, S., K. Dörr, D.N. Wang, and W.
45.Knötzel, J., I. Svendsen, and D.J. Simpson. 1992. Kühlbrandt. 1993. Lipid-protein interactions in crys-
Identification of the photosystem-I antenna polypep- tals of plant light-harvesting complex. J. Mol. Biol.
tides in barley — isolation of 3 pigment-binding 234:347-356.
antenna complexes. Eur. J. Biochem. 206:209-215. 62.Pagano, A., G. Cinque, and R. Bassi. 1998. In vitro
46.Krupa, Z., N.P.A. Huner, J.P. Williams, E. Maissan, reconstitution of the recombinant photosystem II
and D.R. James. 1987. Development at cold-harden- light-harvesting complex CP24 and its spectroscopic
ing temperatures. The structure and composition of characterization. J. Biol. Chem. 273:17154-17165.
purified rye light harvesting complex II. Plant Physiol. 63.Palsson, L.O., S.E. Tjus, B. Andersson, and T. Gill-
84:19-24. bro. 1995. Ultrafast energy transfer dynamics resolved
47.Kügler, M., L. Jansch, V. Kruft, U.K. Schmitz, and in isolated spinach light-harvesting complex I and the
H.P. Braun. 1997. Analysis of the chloroplast protein LHC I-730 subpopulation. Biochim. Biophys. Acta
complexes by blue-native polyacrylamide gel elec- 1230:1-9.
trophoresis (BN-PAGE). Photosynth. Res. 53:35-44. 64.Parkes-Loach, P.S., J.R. Sprinkle, and P.A. Loach.
48.Kühlbrandt, W. 1984. Three-dimensional structure of 1988. Reconstitution of the B873 light-harvesting
the light-harvesting chlorophyll a/b-protein complex. complex of Rhodospirillum rubrum from the separately
Nature 307:478-480. isolated alpha and beta-polypeptides and bacteri-
49.Kühlbrandt, W., D.N. Wang, and Y. Fujiyoshi. 1994. ochlorophyll a. Biochemistry 27:2718-2727.
Atomic model of plant light-harvesting complex by 65.Patzlaff, J.S. and B.A. Barry. 1996. Pigment quantita-
electron crystallography. Nature 367:614-621. tion and analysis by HPLC reverse phase chromatogra-
50.Laemmli, U.K. 1970. Cleavage of structural proteins phy: a characterization of antenna size in oxygen-evolv-
during the assembly of the head of bacteriophage T4. ing photosystem II preparations from cyanobacteria
Nature 227:680-685. and plants. Biochemistry 35:7802-7811.
51.Lam, E., W. Ortiz, and R. Malkin. 1984. Chlorophyll 66.Paulsen, H., B. Finkenzeller, and N. Kühlein. 1993.
a/b proteins of photosystem I. FEBS Lett. 168:10-14. Pigments induce folding of light-harvesting chlorophyll
52.Martinson, T.A. and F.G. Plumley. 1995. One-step a/b-binding protein. Eur. J. Biochem. 215:809-816.
extraction and concentration of pigments and acyl 67.Paulsen, H. and S. Hobe. 1992. Pigment-binding
lipids by sec-butanol from in vitro and in vivo samples. properties of mutant light-harvesting chlorophyll a/b-
Anal. Biochem. 228:123-130. binding protein. Eur. J. Biochem. 205:71-76.
53.McDermott, G., S.M. Prince, A.A. Freer, A.M. 68.Paulsen, H., U. Rümler, and W. Rüdiger. 1990.
Hawthornthwaite-Lawless, M.Z. Papiz, R.J. Cogdell, Reconstitution of pigment-containing complexes from
and N.W. Isaacs. 1995. Crystal structure of an integral light-harvesting chlorophyll a/b-binding protein over-
membrane light-harvesting complex from photosyn- expressed in E. coli. Planta 181:204-211.
thetic bacteria. Nature 374:517-521. 69.Peter, G.F. and J.P. Thornber. 1991. Biochemical
54.Meyer, M. and C. Wilhelm. 1993. Reconstitution of composition and organization of higher plant photo-
light-harvesting complexes from Chlorella fusca system II light-harvesting pigment–proteins. J. Biol.
(Chlorophyceae) and Mantoniella squamata (Prasino- Chem. 266:16745-16754.
phyceae). Z. Naturforsch. C 48:461-473. 70.Peter, G.F. and J.P. Thornber. 1991. Electrophoretic
55.Meyer, M., C. Wilhelm and G. Garab. 1996. Pig- procedures for fractionation of photosystems I and II
ment-pigment interactions and secondary structure of pigment proteins of higher plants and for determina-
252
Analysis and Reconstitution of Chlorophyll–Proteins
tion of their subunit composition, p. 195-210. In L.G. 84.Sarvari, E. and P. Nyitrai. 1994. Separation of chloro-
Rogers (Ed.), Methods in Plant Biochemistry, Vol. 5. phyll-protein complexes by Deriphat polyacrylamide
Academic Press, New York. gradient gel electrophoresis. Electrophoresis 15:1068-
71.Peterson, G.L. 1977. A simplification of the protein 1071.
assay method of Lowry et al. which is more generally 85.Schmid, V. and C. Schäfer. 1994. Alterations of the
applicable. Anal. Biochem. 83:346-356. chlorophyll-protein pattern in chronically photo-
72.Plumley, F.G. and G.W. Schmidt. 1987. Reconstitu- inhibited Chenopodium rubrum cells. Planta 192:473-
tion of chlorophyll a/b light-harvesting complexes: 479.
xanthophyll-dependent assembly and energy transfer. 86.Schmid, V.H.R., K.V. Cammarata, B.U. Bruns, and
Proc. Natl. Acad. Sci. USA 84:146-150. G.W. Schmidt. 1997. In vitro reconstitution of the
73.Porra, R.J., W.A. Thompson, and P.E. Kriedemann. photosystem I light-harvesting complex LHCI-730:
1989. Determination of accurate extinction coeffi- heterodimerization is required for antenna pigment
cients and simultaneous equations for assaying chloro- organization. Proc. Nat. Acad. Sci. USA 94:7667-
phylls a and b extracted with four different solvents: 7672.
verification of the concentration of chlorophyll stan- 87.Setlikova, E., S. Ritter, R. Hienerwadel, J. Kopecky, J.
dards by atomic absorption spectroscopy. Biochim. Komenda, W. Welte, and I. Setlik. 1995. Purification
Biophys. Acta 975:384-394. of a photosystem II reaction center from a thermophilic
74.Rhee, K.H., E.P. Morris, J. Barber, and W. cyanobacterium using immobilized metal affinity chro-
Kühlbrandt. 1998. Three-dimensional structure of the matography. Photosynth. Res. 43:201-211.
plant photosystem II reaction centre at 8 Å resolution. 88.Smith, P.K., R.I. Krohn, G.T. Hermanson, A.K.
Nature 396:283-286. Mallia, F.H. Gartner, M.D. Provenzano, E.K. Fuji-
75.Rhee, K.H., E.P. Morris, D. Zheleva, B. Hankamer, moto, N.M. Goeke, B.J. Olson, and D.C. Klenk.
W. Kühlbrandt (Reprint Author), and J. Barber. 1985. Measurement of protein using bicinchoninic
1997. Two dimensional structure of plant photosystem acid. Anal. Biochem. 79:76-85.
II at 8 Å resolution. Nature 389:522-526. 89.Tan, S., G.R. Wolfe, F.X. Cunningham, and E.
76.Rhiel, E., W. Lange, and E. Mörschel. 1993. The Gantt. 1995. Decrease of polypeptides in the PS I
unusual light-harvesting complex of Mantoniella squa- antenna complex with increasing growth irradiance in
mata — supramolecular composition and assembly. the red alga Porphyridium cruentum. Photosynth. Res.
Biochim. Biophys. Acta 1143:163-172. 45:1-10.
77.Ritter, S., J. Komenda, E. Setlikova, I. Setlik, and W. 90.Thornber, J.P. 1995. Thirty years of fun with antenna
Welte. 1992. Immobilized metal affinity chromatogra- pigment–proteins and photochemical reaction centers:
phy for the separation of photosystem I and photosys- a tribute to the people who have influenced my career.
tem II from the thermophilic cyanobacterium Syne- Photosynth. Res. 44:3-22.
chococcus elongatus. J. Chromatogr. 625:21-31. 91.Tjus, S.E., M. Roobol-Boza, L.O. Palsson, and B.
78.Robinson, N.C., D. Wiginton, and L. Talbert. 1984. Andersson. 1995. Rapid isolation of Photosystem I
Phenyl sepharose-mediated detergent-exchange chro- chlorophyll-binding proteins by anion exchange perfu-
matography: its application to exchange of detergents sion chromatography. Photosynth. Res. 45:41-49.
bound to membrane proteins. Biochemistry 23:6121- 92.Todd, J.B., P.S. Parkes-Loach, J.F. Leykam, and P.A.
6126. Loach. 1998. In vitro reconstitution of the core and
79.Rogl, H., K. Kosemund, W. Kühlbrandt, and I. peripheral light-harvesting complexes of Rhodospiril-
Collinson. 1998. Refolding of Escherichia coli pro- lum molischianum from separately isolated compo-
duced membrane protein inclusion bodies immo- nents. Biochemistry 37:17458-17468.
bilised by nickel chelating chromatography. FEBS Lett. 93.Val, J., E. Monge, and N.R. Baker. 1994. An
432:21-26. improved HPLC method for rapid analysis of the xan-
80.Rögner, M., U. Mühlenhoff, E.J. Boekema, and H.T. thophyll cycle pigments. J. Chromatogr. Sci. 32:286-
Witt. 1990. Mono-, di- and trimeric PS I reaction cen- 289.
ter complexes isolated from the thermophilic 94.van Amerongen, H. and W.S. Struve. 1995. Polarized
cyanobacterium Synechococcus sp.* — size, shape and optical spectroscopy of chromoproteins. Biochem.
activity. Biochim. Biophys. Acta 1015:415-424. Spectroscopy 246:259-283.
81.Roobol-Boza, M. and B. Andersson. 1996. Isolation 95.van der Staay, G.W.M., A. Brouwer, R.L. Baard, F.
of hydrophobic membrane proteins by perfusion chro- Vanmourik, and H.C.P. Matthijs. 1992. Separation of
matography — purification of photosystem II reaction photosystem I and photosystem II from the oxy-
centers from spinach chloroplasts. Anal. Biochem. chlorobacterium (prochlorophyte) Prochlorothrix hol-
235:127-133. landica and association of chlorophyll-b binding anten-
82.Ros, F., R. Bassi, and H. Paulsen. 1998. Pigment- nae with photosystem II. Biochim. Biophys. Acta
binding properties of the recombinant photosystem II 1102:220-228.
subunit CP26 reconstituted in vitro. Eur. J. Biochem. 96.van Holde, K.E., C.E. Johnson, and P.S. Ho. 1998.
253:653-658. Physical Biochemistry. Prentice Hall, London.
83.Sandoná, D., R. Croce, A. Pagano, M. Crimi, and R. 97.Welte, C., R. Nickel, and A. Wild. 1995. Three-
Bassi. 1998. Higher plants light harvesting proteins. dimensional crystallization of the light-harvesting
Structure and function as revealed by mutation analysis complex from Mantoniella squamata (Prasinophyceae)
of either protein or chromophore moieties. Biochim. requires an adequate purification procedure. Biochim.
Biophys. Acta 1365:207-214. Biophys. Acta 1231:265-274.
253
Two-Dimensional Crystallization of
11 Chlorophyll Proteins
Georgios Tsiotis
Department of Chemistry, University of Crete, Heraklion, Greece
255
G. Tsiotis
Successful imaging of biological mole- gent is critical. Ideally, detergent and isola-
cules requires the adoption of experimental tion protocols should be selected not only
protocols that take into account the fragili- to yield a homogeneous structural state but
ty of such molecules when exposed to the also to preserve the protein in a unique
hostile environment encountered in an functional state (11,22). The pure protein
electron microscope. Initially, due to the is obtained in a detergent solution, often
high vacuum, it is essential to provide pro- with residual lipids. In fact, the latter often
tection against dehydration and stabiliza- contribute to the stability of the membrane
tion against molecular collapse. Methods protein and may be essential for successful
such as freezing in vitreous ice (8), sugar 2D crystallization.
embedding (16), and negative staining
(14) have been developed for this purpose. 2.1.1. Detergent Choice and
Frozen-hydrated and sugar-embedded Concentration
techniques yield information on internal
protein density, whereas negative staining Many membrane proteins are destabi-
provides information only on the molecu- lized on extraction from their native mem-
lar environment which is stain accessible. branes, especially when short-chain [high
critical micellar concentration (CMC)]
detergents are used. Although the proteins
2. TWO-DIMENSIONAL can be effectively solubilized with deter-
CRYSTALLIZATION gents that replace the lipid and keep the
hydrophobic surfaces of the protein shield-
2D crystals of membrane proteins have ed from water, delipidation may destabilize
been obtained in three different ways: (i) the protein. The choice of detergent is crit-
reconstitution of the purified membrane ical; there is a fine balance between disrup-
protein into a lipid bilayer at high protein tion of the membrane to solubilize a mem-
density; (ii) improvement of the packing of brane protein and preserving its structural
a highly abundant protein into regular integrity. As discussed above, reconstitu-
arrays in its native membrane; and (iii) tion is closely linked to the properties of
addition of precipitants to promote pro- the detergents used both during purifica-
tein–protein interactions, analogous to 3D tion and the reconstitution itself. There-
crystallization, but yielding 2D crystals. fore, the detergents used for purification
can be exchanged for a different detergent
2.1. Crystallization of Purified used for reconstitution. This makes possi-
Membrane Protein by ble the use of mild detergents for isolation
Reconstitution into Lipid Bilayers (most with long alkyl chain and low
CMC) and then to change to detergents
The crystallization of membrane pro- with high CMC for the crystallization.
teins in 2D arrays requires membrane pro- Among the detergents with an alkyl chain,
teins which are solubilized and purified to those with a charged head group (e.g.,
homogeneity. A unique oligomeric species dodecyl sulfate and hexatrimethylammoni-
possessing a high intrinsic molecular sym- um) are generally not suited for isolation of
metry is likely to favor the growth of “native” membrane proteins. Zwitterionic
crystalline arrays. Indeed, heterogeneous alkyl detergents (e.g., sulfobetains) and
protein solutions and the presence of dena- those with an N-oxide as polar head group
tured proteins greatly hinder the formation can be used with many proteins, but they
of crystalline patches. The choice of deter- are still too harsh for most of the mem-
257
G. Tsiotis
brane proteins. Among the detergents with of Chl a and b at 664 and 647 nm,
an alkyl chain, those with polyoxyethylene respectively.
(e.g., C12E8, Lubrol-series, Triton® X- Chl a (µM) = 13.19 × A664 - 2.57 × A647
series, and Brij-series) or a disaccharide Chl b (µM) = 21.10 × A647 - 5.26 × A664
head group are the mildest (33). Different Total Chl (µM) = 7.93 × A664 + 19.53 × A647
protocols have been developed for the For cyanobacteria, which contain only
exchange of detergent using ultrafiltration Chl a, the concentration of an 80% ace-
(e.g., with Centricon-100 concentrator tone extract can be calculated from A663
devices [Amicon Division, W.R. Grace, using the formula:
MA, USA]), gel filtration, anion or cation
Chl a (µg/mL) = 12.2 × A663 (ε663 =
exchange chromatography, sucrose density
82 mg-1.Chl a.L-1)
gradient, isoelectric focusing, and selective
absorption onto hydrophobic resin beads.
❖ Procedure 2. Bio-Rad Protein Assay
2.1.2. Determination of Chlorophyll This assay is based on the observation
and Protein Content that the absorbance maximum for an acidic
solution of Coomassie® Brilliant Blue G-
To set up optimal 2D crystallization trials,
250 changes from 465 to 595 nm when
accurate knowledge of both lipid and protein
bound to protein (5).
is required, and the following procedures
have been used with success in my laboratory. 1. Prepare a set of protein standards by
An advantage of Chl proteins is that the chro- diluting a stock of 2 mg/mL bovine
mophores that are present (see Procedure 1 serum albumin (BSA) standard in the
and Chapter 10 in this book by Paulsen and buffer used for the experimental sam-
Schmid) correlate with the protein content ples. Include a blank with no BSA.
(see Procedures 2 or 3 in this chapter). 2. Place 0.1 mL of each standard concen-
tration of BSA and appropriately dilut-
❖ Procedure 1. Determination of ed experimental samples in test tubes.
Chlorophyll Concentration 3. Add 5.0 mL of the Bio-Rad protein assay
dye reagent concentrate (Bio-Rad Labo-
Absorption spectroscopy is a quick and ratories, Hercules, CA, USA) (diluted 1:4
easy method for the estimation of chro- with water) to each test tube and vortex mix.
mophore content of Chl binding proteins. 4. After a standard period from 5 minutes
It is potentially nondestructive, although to 1 hour, measure the absorbance at
Chls are usually extracted from the protein 595 nm.
using organic solvents. 5. Plot the absorbance at 595 nm versus
1. Dilute 0.1 mL of the sample to 20 mL the protein concentration of standards.
(200-fold) with 80% acetone. 6. Estimate the protein concentration of
2. Mix well and centrifuge at 3000× g for experimental samples from the stan-
5 minutes. dard curve.
3. Measure the absorbance (A) of the ace-
tone extract at 664 and 647 nm in a 1- ❖ Procedure 3. Bicinchoninic Acid Assay
cm path length cuvette. to Determine Protein Concentration
4. Use the following equation to calculate
the Chl a, Chl b, and total Chl derived The water-soluble salt of bicinchoninic
from the known extinction coefficients acid (BCA) is a specific reagent for Cu+.
258
Two-Dimensional Crystallization of Chlorophyll Proteins
Peptide bonds and 4 amino acids (cysteine, and transition temperatures, and they also
cystine, tryptophan, and tyrosine) reduce provide mixtures of head group charges and
cupric (Cu2+) ions in alkaline medium to molecular geometries similar to membranes
Cu+ (Biuret reaction). The reaction prod- from which the protein originated. Howev-
uct of 2 molecules BCA and 1 Cu+ exhibits er, the complexity of such preparations may
a strong absorbance at 562 nm. Using preclude crystal formation. Nevertheless,
BCA protein assay reagent available from since synthetic lipids, Escherichia coli lipids,
Pierce Chemical (Rockford, IL, USA), the soybean lecithin, and egg lecithin have all
method below should be followed. been successfully used for 2D crystalliza-
1. Prepare a set of protein standards of tion, no general recommendations can be
known concentration by diluting a made on which lipid or lipid mixture is
stock 2 mg/mL BSA standard in the most suitable for any one particular mem-
buffer used for the experimental sam- brane protein. Commercially available
ples. Include a blank with no BSA. lipids (from Sigma [St. Louis, MO, USA]
2. Prepare a working reagent by mixing or Lipid Products [South Nutheld Redhill,
50 parts Reagent A with 1 part Reagent B. Surrey RH1 5 PG, UK]) are often stored as
chloroform solutions or a lyophilized pow-
3. Dilute 0.1 mL of each standard con-
der at -20°C where they are stable for long
centration of BSA and each experimen-
periods. Prior to use for 2D crystallization
tal sample to 2 mL in a working re-
trials, lipids have to be transferred to deter-
agent and incubate for 30 minutes at
gent-containing buffer solutions. The lipid
37°C or room temperature for 2 hours.
content of the reconstitution mixture is, in
4. Allow the samples to cool and measure general, a well-controlled parameter.
the absorbance at 562 nm. Lyophilized lipids can easily be weighed
5. Prepare a standard curve by plotting and redissolved in buffer at 1 to 10 mg/mL
the absorbance at 562 nm versus the containing a high concentration of deter-
protein concentration. Using the stan- gent. Lipids from chloroform can be han-
dard curve, determine the protein con- dled in the same way after removal of the
centration for each unknown protein organic solvent. The amount of detergent
sample. to solubilize completely a lipid stock can be
Since color development will continue calculated by use the following equation:
slowly, it is necessary that all absorbance Concentration of detergent (mol/L) =
readings be done immediately. Care must CMC (mol/L) + 3 × concentration of the
be taken to avoid the presence of reducing lipid (mol/L).
agents such as thiols [dithiothreitol (DTT) All organic solvent should be removed
or mercaptoethanol] or large amounts of before solubilization in detergent buffer, as
sugars, which will interfere with the assay. it interferes with the crystallization.
Figure 1. Hydrodynamic radii on dilution of a mixed micellar suspension containing a detergent (8-POE) and either lipid
(EggPC) or lipid and a membrane protein (porin OmpF). Above 15 mM 8-POE, uniform micelles are seen. Below 15 mM 8-
POE, each treatment produces a range of structure radii with large crystals seen in the presence of OmpF. The dilution is repre-
sented as a function of detergent concentration to illustrate the 3-stage model. Large bilayer structures (crystals/vesicles) at low
detergent concentration (Stage I). Small micelles at high detergent concentration (Stage III). Mixture of structures at intermedi-
ate detergent concentration (Stage II). Black arrows indicate the saturation points, and white arrows indicate the solubilization
points. Data from Reference 7. Figure was modified from Reference 15.
260
Two-Dimensional Crystallization of Chlorophyll Proteins
At the start of a typical reconstitution all components by equal factors, until the
experiment, an excess of detergent ensures a free detergent concentration drops below
homogenous distribution of protein and saturation. Crystallization by the dilution
lipid in micelles. As detergent concentration method requires a significant dilution of
is decreased, lipids and proteins interact due the protein, and therefore, rather high ini-
to the exposure of their hydrophobic sur- tial concentrations are required. On the
faces. With an excess of lipid over protein, other hand, the dilution method allows the
the protein is mainly incorporated into lipid process to be arrested when the saturation
bilayers, similar to its native state. In an point is reached, extending the time in
excess of protein over lipid, the protein which an ordered assembly of the compo-
mostly ends up in amorphous aggregates, nents can take place. As an illustration of
perhaps denatured. An important parameter this, the following 2 procedures describe
is, therefore, the lipid:protein ratio (LPR), the preparation of 2D crystals of PSI com-
which should be low enough to promote plexes from the thermophilic cyanobacteri-
crystal contacts between protein molecules, um Synechococcus sp. OD 24 (Procedure 4)
but not so low that the protein is lost to and tubular crystals of spinach PSII (Pro-
aggregation. When the membrane protein is cedure 5).
reconstituted from a mixture of solubilized
components, crystal ordering of proteins ❖ Procedure 4. 2D Crystals of the PSI
may occur during reconstitution. For crystal Complex from Synechococcus sp. OD
packing during reconstitution, the LPR of 24 (7)
the reconstitution experiment must be as
low as possible to ensure close packing with- PSI complexes are isolated from Syne-
out leading to excessive aggregation. While chococcus sp. OD 24 according to Refer-
the lipid content of the reconstitution mix- ence 24.
ture is, in general, a well-controlled parame-
1. Make 600 µL of a starting mixture
ter, the content of monodisperse protein is
containing 1 mg/mL PSI, 1 mg/mL
sometimes unknown, because protein assays
lecithin, and 7.5 mg/mL octyl-β-thio-
do not indicate the amount of aggregates.
glucoside (OTG).
2. Dilute this mixture by the addition of
2.2. Crystallization Methods sequential 25-µL aliquots of 10 mM 2-
[N-morpholino]ethanesulfonic acid
The manner in which the detergent (MES), pH 6.0, 100 mM NaCl, 10
concentration is decreased for reconstitu- mM MgCl2 to achieve a slow dilution
tion and subsequent 2D crystallization is of PSI.
an important consideration. The common- 3. After the addition of 4 aliquots (i.e.,
ly used techniques for detergent removal 100 µL, thus reducing the concentra-
are dilution (7,47), dialysis (11,22), and tion of OTG to about 6 mM), large
selective adsorption of the detergent on interconnected protein–lipid aggre-
solid supports such as the hydrophobic gates can be observed.
resin beads (43). 4. Upon further dilution of detergent (ad-
dition of 12 aliquots, i.e., 300 µL), dis-
2.2.1. Dilution Method tinct vesicles can be seen. Crystalline
packing of PSI complexes should be
Diluting a solution of protein, lipid, and obtained after the addition of 20
detergent decreases the concentrations of aliquots (i.e., 500 µL).
261
G. Tsiotis
4. Add DMPC solution to achieve a LPR β-glucoside (OG), pH 7.8. After dialysis
of 1. against buffer containing 0.8% OG, 5 mM
5. Dialyze the mixture against a deter- MgCl2, and 50 mM NaCl in the dark at
gent-free buffer containing 25 mM 4°C for 5 days, 2D crystals were obtained,
ammonium ferric citrate. which had a hexagonal pattern with a lat-
Dialysis cell temperature: 26°C for tice constant of 12.3 nm. Modification of
24 hours; increase to 37°C over 12 hours; this procedure, by the use of sonicated vesi-
37°C for 24 hours; and decrease to 26°C cles of dioleoyl-9-10 phosphatidylcholine
over 10 hours. Total dialysis time is 70 lipid solubilized in OG in a ratio 1:1 to
hours. proteins, allowed the formation of crystals
Digital image processing of negatively which diffracted at 8.5 Å (23). Digital
stained and frozen-hydrated specimens pre- image processing of frozen-hydrated speci-
pared using Procedure 7 revealed ortho- mens revealed crystals with a p22121 sym-
rhombic crystals with unit cell dimensions a metry and unit cell with dimensions of a =
= 13.8 nm, b = 14.5 nm, and p121 symme- 12.8 nm, b = 19.4 nm.
try. The same procedure has also been used As an alternative to the dialysis appara-
with trimeric PSI isolated from mesophilic tus shown in Figure 2, an inexpensive mi-
cyanobacterium Synechococcus PCC 7002 by crodialysis arrangement with Eppendorf®
isoelectric focusing (46) to provide 2D crys- tubes or dialysis buttons may be used (Fig-
tals (Tsiotis, unpublished results). ure 3). This method enables the dialysis of
Purple sulfur and nonsulfur bacteria small volumes (<50 µL) of protein–lipid–
possess membrane-bound LH complexes, detergent. Another interesting microdialy-
which serve to transfer energy to the RC, sis device using a bent glass capillary tube,
where charge separation occurs. LH com- shown in Figure 4 has been described by
plex (170 µg), isolated from a carotenoid- Kuhlbrandt (26). A glass tube of 2.5 mm
less mutant of the purple nonsulfur bac- inner diameter and about 6.5 mm outer
terium Rhodospirillum rubrum G9 as diameter is bent by 90° near to one end.
described previously (13), was dissolved in The end is melted into a smooth surface,
100 µL of 50 mM NH4HCO3, 1% octyl- which forms a tight seal with the dialysis
263
G. Tsiotis
membrane. The dialysis membrane is fixed and the device is placed in a glass beaker
with a ring of silicon tubing, and 20 to 50 with dialysis buffer. One interesting advan-
mL dialysate is fed into the capillary from tage of this device is that a sample for elec-
the open end using a syringe (Figure 4). tron microscopy can be taken at anytime
The dialysate is shaken down against the with a finer glass capillary without disturb-
dialysis membrane to remove air bubbles, ing the progress of the experiment.
264
Two-Dimensional Crystallization of Chlorophyll Proteins
50% slurry in water supplemented with vantage is that these spontaneously formed
0.2% NaN3. Procedure 9 describes the use 2D crystals are rarely highly ordered
of this technique to crystallize the Chlamy- because other components can be trapped
domonas reinhardii cytochrome b6f com- and hinder the crystal growth. Another dis-
plex, purified as in Reference 41. advantage is that these methods are limited
to cases where the membrane protein
❖ Procedure 9. 2D Crystals of occurs at high density. However, the quali-
Cytochrome b6f Complex from ty of naturally occurring 2D crystals can be
C. reinhardii (37) improved by either detergent extraction,
fusion of crystalline patches, or incubation
1. Resuspend the purified cytochrome b6f with additives. Examples of naturally
complex in buffer containing: 6.8 mM occurring crystalline membranes are the
Tricine, 245 mM ammonium phos- photosynthetic membranes from purple
phate, 0.3 mM NaN3, 2 mM CaCl2, bacteria (35). PSII from higher plants is
1.1 mM benzamidine, 5.6 mM Â-ami- found in the thylakoid membrane at high
nocaproic acid, 0.2 mM phenylmethyl- density and sometimes in a regular lattice
sulfonyl fluoride (PMSF), 0.3% glyc- form (45). Detergent extraction of the thy-
erol, 20 mM Hecameg, pH 8.0. lakoid membrane under nonsolubilizing
2. Add a mixture of EggPC and di-C18:1- conditions promotes or improves the crys-
phosphatidylglycerol (1:1 ratio). The tallinity (3,19,31).
final protein concentration in reconsti- For example, crystals of PSII, which
tution mixture should be 0.5 mg/mL provided evidence for the presence of a
and the LPR 0.2. PSII dimer (3) were obtained as follows.
3. Preincubate the sample overnight in PSII membranes, isolated as in Reference
the cold room under gentle stirring. 4, were resuspended in 150 mM NaCl, 5
4. Treat with 200 mg/mL SM2 Bio-Beads. mM MgCl2, and 20 mM Tricine, pH 7.5,
at a final Chl concentration of 2 mg/mL.
5. Incubate for 12 hours. Triton X-100 was added to a concentration
6. Pipet off the reconstituted material and of 4% (wt/vol), and the mixture was incu-
keep for 24 hours at 4°C. bated at 20°C for 20 minutes in the dark.
7. Freeze in liquid nitrogen, then thaw It was then centrifuged for 30 minutes
the sample at 37°C 3 times. (45 000× g), and the pellet was washed
The crystals have a unit cell (a = 17.5 once with buffer containing 10 mM NaCl,
nm, b = 6.8 nm, and γ= 90°) and a p22121 5 mM MgCl2, and 10 mM HEPES, pH
symmetry. 7.6. The crystals obtained had a rectangu-
lar unit cell (a = 17.8 nm and b = 26.7
2.3. 2D Crystallization in Native nm). In contrast, similar crystals obtained
Membranes using dodecyl maltoside (0.06%) instead
of Triton X-100, and a subsequent sucrose
Some membrane proteins have a natural gradient (0–2 M) (18) had unit cell dimen-
propensity to form regular arrays within sions of a = 16.8 nm, b = 18.9 nm, with γ =
the native membrane. Since the membrane 91°, and a p1 symmetry group, and led to
protein does not dissociate from the lipid the proposal of the presence of a PSII
bilayer, its native orientation is maintained. monomer (19). Lastly, using a similar
Additionally, the proteins are not exposed approach but with altered conditions,
to harsh detergents and, therefore, are kept tubular PSII crystals have been obtained
under stabilizing conditions. One disad- from spinach, which indicate the presence
266
Two-Dimensional Crystallization of Chlorophyll Proteins
of 2 monomeric PSII complexes (31) (see the chloroplast, was resolved to atomic reso-
Procedure 10). lution after formation of 2D crystals in a
“batch method” (49) (see Procedure 11).
❖ Procedure 10. Tubular Crystals of The temperature profile proved to be critical
PSII from Spinach (31) for the crystallization of LHC-II. In addi-
tion, 2D crystals of some proteins with large
Thylakoid membranes were isolated as extramembrane domains can be obtained at
in Reference 35. a pH close to the isoelectric point (35) or by
1. Resuspend thylakoid membranes in 15 increasing the ionic strength (6,12). These
mM NaCl, 50 mM sucrose, 5 mM methods are best interpreted as variations of
MgCl2, and 20 mM HEPES, pH 7.5. 3D crystallization and not as proper recon-
2. Adjust Chl concentration to 0.8 stitution of membrane proteins into a
mg/mL. native-like environment.
3. Add Triton X-100 to give a detergent
to Chl ratio of 4.5:1. ❖ Procedure 11. 2D Crystals of LHC-II
4. Incubate for 20 minutes in ice with Complex from Pea Chloroplasts
stirring.
LHC-II complexes from pea chloroplas-
5. Centrifuge for 20 minutes (20 000× g) ts are isolated as described in Reference 27.
and resuspend the pellet in 15 mM
NaCl, 5 mM MgCl2, 20 mM HEPES, 1. An aliquot of LHC-II complex (with a
pH 7.5 Chl concentration of 4 mg/mL in
6. Determine the Chl concentration. 0.4% Triton X-100) is diluted 50 times
7. Add Triton X-100 to give a ratio of 3:1 with distilled water and KCl added to a
of detergent to Chl. final concentration of 300 mM.
8. Incubate for 10 minutes on ice. 2. Centrifuge and collect the pellet. Pro-
ceed with either Method A or B.
9. Centrifuge for 15 minutes (15 000× g)
and wash the pellet with the same Method A
buffer.
(i) Dissolve the pellet to a Chl concen-
Additional washing steps can be includ- tration of 0.7 mg/mL in Triton X-100
ed to separate the crystals from the remain- and glycerol at final concentrations of
ing material. 0.23% and 40%, respectively.
The tubular crystals have a length of 1
to 2 µm and a diameter of 0.2 µm, and the (ii) Incubate at 35° to 40°C for 2 hours.
unit cell has dimensions of a = 11.2 nm Method B
and b = 16.1 nm.
(i) Dissolve the pellet with 0.11% Triton
X-100 and 0.24% n-nonyl-β-gluco-
2.4. Crystallization of Purified Membrane
pyranoside.
Protein by Precipitation
(ii) Add glycine buffer (100 mM, pH
In a few cases, a solubilized membrane 7.0) and glycerol to a final concentra-
protein has been crystallized into 2D sheets tion of 10 mM and 40%, respective-
without detergent removal under condi- ly. The Chl concentration should be
tions similar to those used in 3D crystalliza- 0.75 mg/mL.
tion experiments. Pea thylakoid LHC-II, (iii) Incubate for 2 days at 25°C and then
the most abundant membrane protein in for 2 hours at 40°C.
267
G. Tsiotis
The crystals have a unit cell of a = 12.7 than biological atoms (C, H, O, N, P, and
nm, b = 12.7 nm, with γ = 60°. The plane S), the contrast is drastically increased but
group has a p321 symmetry (28). also inverted (hence the term “negative”
staining). In addition, the heavy atom salts
partially substitute the water in the native
3. ANALYSIS OF 2D CRYSTALLIZA- environment of the molecules, thus em-
TION REACTIONS bedding the specimen and protecting it
from collapse upon drying in air. The most
Even if 2D crystals of large membrane commonly used heavy atom salts are uranyl
proteins can sometimes be observed by acetate or formate and sodium or potassi-
light microscopy, the shape of the crystals um phosphotungstate. Uranyl salts are
and their degree of order can best be more suitable for more protein samples,
assessed by electron microscopy. Thus, whereas tungstate is useful for lipid struc-
screening of 2D crystallization experiments tures. The pH of tungstate solutions can be
requires access to an electron microscope. modified, whereas uranyl salts precipitate
The quickest specimen preparation meth- at pH greater than 4.5.
od for screening for 2D crystals is negative After negative staining, examination of
staining. This method is rapid, needs a the samples is carried out by transmission
small amount of sample (less than 5 µL), is electron microscopy. The presence of large
remarkably simple, and can provide struc- vesicles or sheets can be checked at low
tural information to a resolution of about 2 magnification (2000×–5000×) because of
nm. A negatively stained 2D crystal is the high contrast obtained when operating
embedded in a dry microcrystalline heavy at large underfocus even for very thin
atom replica. As heavy atoms used for neg- objects (Figure 5A). Further observations
ative staining scatter electrons much more are done at higher magnification (Figure
268
Two-Dimensional Crystallization of Chlorophyll Proteins
5B). If no crystals are found, the presence ic acid]; LHC, light-harvesting complex;
of remaining detergent, single protein par- LPR, lipid:protein ratio; MES, 2-[N-mor-
ticles incorporated in lipid vesicles, solubi- pholino]ethanesulfonic acid; OG, octyl-β-
lized proteins, or protein aggregates can all glucoside; OTG, octyl-β-thioglucoside; 8-
be identified. This information may be POE, octyl-polyoxyethylene; PSI and PSII,
valuable for designing further crystalliza- photosystem I and II; RC, reaction center.
tion trials.
REFERENCES
269
G. Tsiotis
13.Ghosh, R., A. Hoenger, A. Hardmeyer, D. Mihailes- 30.Lichtenberg, D., R.J. Robson, and E.A. Dennis.
cu, R. Bachofen, A. Engel, and J.P. Rosenbusch. 1993. 1983. Solubilization of phospholipids by detergents.
Two-dimensional crystallization of the light-harvesting Structural and kinetic aspects. Biochim. Biophys. Acta
complex from Rhodospirillum rubrum. J. Mol. Biol. 737:285-304.
231:501-504. 31.Lyon, M.K., K.M. Marr, and P.S. Furcinitti. 1993.
14.Haschenmeyer, R.H. and R.J. Meyers. 1972. Nega- Formation and characterization of 2-dimensional
tive staining, p. 101-107. In M.A. Hayat (Ed.), Princi- crystals of photosystem-II. J. Struct. Biol. 110:133-
ples and Techniques of Electron Microscopy. Vol. 2, 140.
Van Nostrand, New York. 32.McDermott, G., S.M. Prince, A.A. Freer, A.M. Haw-
15.Hasler, L., J.B. Heymann, T. Walz, J. Kistler, and A. thornthwaite-Lawless, M.Z. Papiz, R.J. Cogdell, and
Engel. 1998. 2D crystallization of membrane proteins: N.W. Isaacs. 1995. Crystal structure of an integral
rationales and examples. J. Struct. Biol. 121:162-172. membrane light-harvesting complex from photosyn-
16.Henderson, R. and P.N.T. Unwin. 1975. Three- thetic bacteria. Nature 374:517-521.
dimensional model of purple membrane obtained by 33.Michel, H. 1991. General and practical aspects of
electron microscopy. Nature 257:28-32. membrane protein crystallization, pp. 74-88. In
17.Henderson, R., J.M. Baldwin, K.H. Downing, J. Lep- H. Michel (Ed.), Crystallization of Membrane Pro-
ault, and F. Zemlin. 1986. Structure of purple mem- teins, CRC Press, Boca Raton.
brane from Halobacterium halobium: recording, mea- 34.Michel, H., K.A. Weyer, H. Gruenberg, and F.
surement and evaluation of electron micrographs at 3.5 Lottspeich. 1985. The ‘heavy’ subunit of the photo-
Å resolution. Ultramicroscopy 19:147-178. synthetic reaction centre from Rhodopseudomonas
18.Holzenburg, A., F.H. Wilson, M.E. Finbow, and R.C. vidris: isolation of the gene, nucleotide and amino acid
Ford. 1992. Structural investigations of membrane sequence. The EMBO Journal 4:1667-1672.
proteins: the versality of electron microscopy. Bio- 35.Miller, K.R. and J.S. Jacob. 1983. Two-dimensional
chem. Soc. Trans. 20:591-597. crystals formed from the photosynthetic reaction cen-
19.Holzenburg, A., M.C. Bewley, F.H. Wilson, W.V. ters. J. Cell Biol. 97:1266-1270.
Nicholson, and R.C. Ford. 1993. 3-dimensional struc- 36.Mishra, R.K. and D.F. Ghanotakis. 1994. Selective
ture of photosystem-II. Nature 363:470-472. extraction of CP 26 and CP 29 proteins without affect-
20.Iwata, S., C. Ostermeier, B. Ludwig, and H. Michel. ing the binding of the extrinsic proteins (33, 23 and 17
1995. Structure at 2.8 Å resolution of cytochrome c kDa) and the DCMU sensitivity of a photosystem II
oxidase from Paracoccus denitrificans. Nature 376:660- core complex. Photosynth. Res. 42:37-42.
669. 37.Mosser, G., C. Breyton, A. Olofsson, J.L. Popot,
21.Jansson, S. 1994. The light-harvesting chlorophyll a/b- and J.L. Rigaud. 1997. Projection map of cyto-
binding proteins. Biochim. Biophys. Acta 1184:1-19. chrome b6f complex at 8 Å resolution. J. Biol. Chem.
22.Jap, B.K., M. Zulauf, T. Scheybani, A. Hefti, W. 272:20263-8.
Baumeister, U. Aebi, and A. Engel. 1992. 2D crystal- 38.Nakazato, K., C. Toyoshima, I. Enami, and Y. Inoue.
lization: from art to science. Ultramicroscopy 46:45-84. 1996. Two-dimensional crystallization and cryo-elec-
23.Karrasch, S., P.A. Bullough, and R. Ghosh. 1995. The tron microscopy of photosystem II. J. Mol. Biol. 257:
8.5 Å projection map of the light-harvesting complex I 225-232.
from Rhodospirillum rubrum reveals a ring composed 39.Ostermeier, C., S. Iwata, B. Ludwig, and H. Michel.
of 16 subunits. EMBO J. 14:631-638. 1995. Fv fragment-mediated crystallization of the
24.Karrasch, S., D. Typke, T. Walz, M. Miller, G. Tsio- membrane protein bacterial cytochrome c oxidase.
tis, and A. Engel. 1996. Highly ordered two-dimen- Nat. Struct. Biol. 2:842-6.
sional crystals of photosystem I reaction center from 40.Pebay-Peyroula, E., G. Rummel, J.P. Rosenbusch, and
Synechococcus sp.: functional and structural analyses. E.M. Landau. 1997. X-ray structure of bacteri-
J. Mol. Biol. 262:336-348. orhodopsin at 2.5 angstroms from microcrystals grown
25.Kreusch, A., A. Neubuser, E. Schiltz, J. Weckesser, in lipidic cubic phases [see comments]. Science
and G. Schulz. 1994. Structure of the membrane 277:1676-81.
channel porin from Rhodopseudomonas blastica at 2.0 Å 41.Pierre, Y., C. Breyton, D. Kramer, and J.-L. Popot.
resolution. Prot. Sci. 3:58-63. 1995. Purification and characterization of the
26.Kühlbrandt, W. 1992. Two-dimensional crystalliza- cytochrome b6f complex from Chlamydomonas rein-
tion of membrane proteins. Q. Rev. Biophys. 25:1-49. hardtii. J. Biol. Chem. 270:29342-29349.
27.Kühlbrandt, W., T. Thaler, and E. Wehrli. 1983. The 42.Rhee, K.H., E.P. Morris, D. Zheleva, B. Hankamer,
structure of membrane crystals of the light-harvesting W. Kuhlbrandt, and J. Barber. 1997. Two-dimension-
chlorophyll a/b protein complex. J. Cell Biol. 96:1414- al structure of plant photosystem II at 8-angstrom res-
1424. olution. Nature 389:522-526.
28.Kühlbrandt, W. and D.N. Wang. 1991. Three- 43.Rigaud, J.L., G. Mosser, J.J. Lacapere, A. Olofsson,
dimensional structure of plant light-harvesting com- D. Levy, and J.L. Ranck. 1997. Bio-Beads: an effi-
plex determined by electron crystallography. Nature cient strategy for two-dimensional crystallization of
350:130-134. membrane proteins. J. Struct. Biol. 118:226-235.
29.Landau, E.M. and J.P. Rosenbusch. 1996. Lipidic 44.Schirmer, T., T. Keller, Y. Wang, and J. Rosenbusch.
cubic phases: a novel concept for the crystallization of 1995. Structural basis for sugar translocation through
membrane proteins. Proc. Natl. Acad. Sci. USA 93: maltoporin channels at 3.1 Å resolution [see com-
14532-14535. ments]. Science 267:512-514.
270
Two-Dimensional Crystallization of Chlorophyll Proteins
45.Staehelin, L.A. 1975. Chloroplast membrane struc- 48.Tsukihara, T., H. Aoyama, E. Yamashita, T. Tomiza-
ture. Intramembrenous particles of different sizes make ki, H. Yamaguchi, K. Shinzawa-Itoh, R. Nakashima,
contact in stacked membrane regions. Biochim. Bio- R. Yaono, and S. Yoshikawa. 1995. Structure of metal
phys. Acta 408:1-11. sites of oxidized bovine heart cytochrome c oxidase at
46.Tsiotis, G., W. Nitschke, W. Haase, and H. Michel. 2.8 Å. Science 269:1069-1074.
1993. Purification and crystallization of photosystem I 49.Wang, D.N. and W. Kühlbrandt. 1991. High-resolu-
complex from a phycobilisome-less mutant of the tion electron crystallography of light-harvesting
cyanobacterium Synechococcus PCC 7002. Photosynth. chlorophyll a/b-protein complex in three different
Res. 35:285-297. media. J. Mol. Biol. 217:691-699.
47.Tsiotis, G., T. Walz, A. Spyridaki, A. Lustig, A. 50.Weiss, M.S., A. Kreusch, E. Schiltz, U. Nestel, W.
Engel, and D. Ghanotakis. 1996. Tubular crystals of Welte, J. Weckesser, and G.E. Schulz. 1991. The
a photosystem II core complex. J. Mol. Biol. 259: structure of porin from Rhodobacter capsulatus at 1.8 Å
241-248. resolution. FEBS Lett. 280:379-382.
271
Biosynthesis and Analysis of Bilins
12
Matthew J. Terry
University of Southampton, Southampton, England, UK
273
M.J. Terry
PEB, which it obtains from a diet of red bilin 2,3-reductase, catalyses the formation
seaweed, as a defensive ink pigment (43). of the A-ring ethylidene group required for
The biosynthetic relationship between covalent assembly to apophytochrome (see
the bilins discussed in this chapter is shown Reference 55 and Chapter 13 in this text
in Figure 2. The first committed step in the by McDowell and Lagarias). The product
pathway for the synthesis of all bilins is the of this reaction, 3(Z)-PΦB (57), is subse-
oxidation of protoheme to BV IXα by the quently converted to 3(E)-PΦB, the imme-
enzyme heme oxygenase (5,34,42,53). Fol- diate precursor of the bound phytochrome
lowing this universal reaction, BV IXα can chromophore, by an, as yet unidentified,
be reduced at one of three different posi- bilin 31,32 cis-trans isomerase. In red algae,
tions in a somewhat phylogenetic-depen- the pathways for the synthesis of bilins are
dent manner. In animals, BV IXα is quite complex, but have been recently been
reduced at the C-10 position (see Figure 1) reviewed in some depth (5). Concentrating
by the enzyme BV reductase to give BR just on the central pathway, BV IXα is first
IXα (32,50). In higher plants, the plastid- reduced by a bilin 15,16-reductase to give
localized enzyme PΦB synthase, which is a 15,16-DHBV IXα (10) and subsequently
274
Biosynthesis and Analysis of Bilins
by a bilin 2,3-reductase to 3(Z)-PEB (9). ther increased by recent evidence that the
This later enzyme is presumably related to reduction of BV IXα is not simply related
PΦB synthase, as both enzymes accom- to broad phylogenetic group. The green
plish the same reaction. 3(Z)-PEB then alga Mesotaenium caldariorum is able to
undergoes a series of isomerizations to pro- synthesize 3(Z)-PCB directly from 3(Z)-
duce the 3(E)-isomers of both PEB and PΦB (67). Such a pathway suggests a rea-
PCB (5,9). sonable biosynthetic alternative to the red
The complexity of bilin synthesis is fur- algal pathway for PCB synthesis, and
Figure 2. The biosynthetic relationship of the bilins discussed in the text. The full pathway shown has been constructed from
biosynthetic pathways identified in a number of different organisms with no single organism containing all pathways. Heme oxy-
genase is present in animals, plants, algae, and cyanobacteria. Of the four reductases, BV reductase has been detected in animals
and cyanobacteria, PΦB synthase (bilin 2,3-reductase) in plants and green algae, bilin 15,16-reductase in red algae and possibly
cyanobacteria, and bilin 181,182-reductase in green algae alone. There are two isomerase activities. 3(Z),3(E) cis-trans-isomerase
(31,32-isomerase) appears to be present in plants and algae, while (15,16),(181,182)-isomerase activity has only been detected in
red algae. Broken arrows indicate possible biosynthetic routes that have yet to be shown experimentally. Abbreviations: BR, biliru-
bin; BV, biliverdin; DHBV, dihydrobiliverdin; PCB, phycocyanobilin; PEB, phycoerythrobilin; PΦB, phytochromobilin.
275
M.J. Terry
indeed the authors report the unpublished more cost-effective method of obtaining
observation that Cyanidium caldarium has BV is to purify it from a mixture of bile
PΦB synthase activity. Since the yeast acids, which can be purchased quite cheap-
Pichia pastoris also has this activity (66), ly from both Sigma and ICN Pharmaceuti-
bilin 2,3-reductases may be considerably cals. An alternative method of obtaining
more widespread than first thought. To BV IXα is by oxidation of the BR IXα
complicate matters further, it now appears with 2,3-dichloro-5,6-dicyano-1,4-benzo-
that cyanobacteria also possess a BV (bilin quinone (22,36). This method is also cost-
10,11-)reductase (49), while two more effective if large quantities of BV IXα are
novel enzymes (a bilin 4,5-reductase and a required and is the most suitable for
bilin 121,122-dehydrogenase) have been obtaining (14C)-labeled BV IXα. In addi-
proposed to account for the synthesis of tion, it can be used for the synthesis of the
the five additional phycobilins identified to BV IIIα and XIIIα isomers from the corre-
date (5). sponding BR isomers or for MBV IXα
While the diversity of bilins and their from mesobiliverdin (MBR) IXα. Exten-
biosynthesis is a fascinating topic, this sive methods for the synthesis of a wide
chapter is primarily concerned with the range of BV dimethyl esters, including
methods required to study bilin biosynthe- DHBV dimethyl ester, have also been pub-
sis in general. In particular, I will focus on lished (52).
a few specific steps in bilin synthesis in
photosynthetic organisms, only addressing 2.2. Phycobilins
the biochemistry of these pathways in ani-
mals for comparative purposes. It is intend- The phycobilins can most easily be
ed that the methods described here could obtained by cleavage from the appropriate
be applied equally well to the biosynthetic phycobiliprotein using a HgCl2-assisted
steps that have yet to be characterized. methanolysis protocol. A method for the
preparation of PCB from lyophilized,
mixed Spirulina species is shown in Proce-
2. ISOLATION AND PREPARATION dure 1. This method has been described in
OF BILINS Terry et al. (56) and is based on the earlier
methods of Beale and Cornejo (9) for the
2.1. Biliverdins isolation of phycobilins from Porphyridium
cruentum and Arciero et al. (1) who used
BV can be purchased as a dihydrochlo- Synechococcus.
ride of about 80% purity from ICN Phar-
maceuticals (Costa Mesa, CA, USA) and, ❖ Procedure 1. Isolation and
until very recently, from Sigma (St. Louis, Purification of PCB from
MO, USA). One problem with these sam- Cyanobacteria
ples is that they contain a mixture of iso-
mers making them unsuitable for many Reagents
purposes. An alternative is to buy a more
pure sample of the IXα isomer (the isomer • Lyophilized mixed Spirulina species
that predominates in biological systems) (Sigma)
from Porphyrin Products (Logan, UT, • C18 cartridge e.g., Sep-Pak (Waters-
USA). The IXα isomer can then be further Millipore, Milford, MA, USA) or
purified by high-performance liquid chro- Bond Elut (Varian Instruments, Santa
matography (HPLC) (see section 3.1). A Clarita, CA, USA).
276
Biosynthesis and Analysis of Bilins
of the observation that the acetone treat- fonic acid (HEPES)-NaOH, pH 7.7,
ment required for extraction of P. cruentum 1% (wt/vol) polyvinylpyrrolidone (solu-
cells prior to phycoerythrin methanolysis ble PVP), 2 mM MgCl2, 2 mM
leads to oxidation of the bound PEB to EDTA (free acid), 2 mM EDTA (di-
give PΦB, which is then cleaved by the Na salt)].
normal methanolysis procedure (17). Phy- • Bovine serum albumin (BSA) and
coerythrin-containing cells extracted with cysteine.
methanol do not produce PΦB. The • Assay buffer [500 mM sorbitol, 20
methanolysis and purification procedures mM TES, 10 mM HEPES-NaOH,
are exactly as described above, and the yield pH 7.7, with 1 mM phenylmethyl-
of, predominantly, 3(E)-PΦB was estimat- sulphonyl fluoride (PMSF), 2 M leu-
ed to be approximately one third the yield peptin, 0.5 mM dithiothreitol (DTT)
of PEB (17). added fresh] 300 000 U/mL catalase in
A reasonable alternative is to synthesize 5 mM citrate buffer, pH 7.5.
PΦB enzymatically by incubating the BV
• NADPH regenerating system (12 mM
IXa precursor with isolated plastids (57,
NADP+, 100 mM glucose-6-phos-
62). In principle, plastid preparations from
phate, and 15 U/mL glucose-6-phos-
a wide variety of tissues or species could be
phate dehydrogenase).
used, but in my experience, there is consid-
erable variation in the yields of PΦB that • 1 mM BV IXα stock solution in
can be obtained. Of the plants that have DMSO.
been investigated, the best starting materi-
al appears to be pea and oat seedlings, Method
while tomato, cucumber, and Arabidopsis
are much less suitable. It is also generally 1. Grow pea seedlings in moist vermi-
better to use etioplasts than chloroplasts, as culite in the dark for 8 to 10 days at
bilin recovery is greater from these prepa- 25°C.
rations. The procedure is relatively straight- 2. Prepare homogenization buffer from
forward and only requires crude etioplast stock solution by diluting 2-fold with
preparations. Indeed, attempts to purify water. Add BSA and cysteine to final
plastids further with Percoll gradients led concentrations of 0.2% (wt/vol) and 5
to decreased bilin yields, possibly because mM, respectively, and adjust back to
of interactions between the bilins and the pH 7.7.
Percoll itself. Procedure 2 shows a simple 3. Under green safelight, harvest approxi-
procedure for obtaining PΦB from BV mately 30 g apical tissue (top 3–4 cm
using isolated pea (Pisum sativum L.) etio- of about 200 seedlings) and homoge-
plasts (55,62). nize in a prechilled mortar and pestle
in 2 mL/g fresh weight ice-cold
homogenization buffer. The sample
❖ Procedure 2. Synthesis of PΦB from
should be kept chilled (4°C) through-
BV Using Isolated Pea Etioplasts
out the isolation procedure.
4. Filter homogenate through 4 layers of
Reagents
muslin.
• Homogenization stock solution [1 M 5. Centrifuge for 1 minute at 8000× g.
sorbitol, 40 mM TES, 20 mM 4-(2- 6. Wipe away any starch from the side of
hydroxyethyl)-1-piperazineethanesul- the tube with a tissue and gently sus-
279
M.J. Terry
280
Biosynthesis and Analysis of Bilins
unknown bilin(s), then a wider range of with this system BV IXα and 3(Z)-PΦB
techniques may be appropriate. Although still elute close to each other, and further
HPLC and absorption spectroscopy will improvements in separation have been
again prove useful, it may be necessary to achieved using a mobile phase of ace-
determine the chemical structure in order tone:20 mM formic acid (50:50 vol/vol).
to confirm or elucidate the identity of the Interestingly, this fairly small change in sol-
unknown bilin. In this case, 1H nuclear vent system leads to a complete reversal in
magnetic resonance (NMR) spectroscopy the order that the bilins elute (67).
is the most appropriate technique. Another problem associated with most
of these solvent systems is that retention
3.1. HPLC Analysis and Purification times vary considerably from day to day
and even during a series of runs. This prob-
Bilins are readily separated by C18 ably reflects a pronounced temperature
reverse phase HPLC using isocratic solvent sensitivity of these systems, but may also be
systems, and their distinctive absorption related in part to the condition of the col-
spectra make them easy to detect in the vis- umn itself. This variation highlights the
ible wavelength range. The most common- need to identify peaks by co-injection of
ly used solvent system for analyzing most known standards rather than by retention
bilins is that of Beale and Cornejo (8), who time. Perhaps the single biggest difficulty
separated phycobilins on an Altex Ultras- of using HPLC for bilin analysis is that
phere octadecylsilane (ODS) column (250 there is tremendous variation between
mm long × 46 mm diameter, 5 µm particle columns, with new columns, apparently
size). Separation was achieved using a identical to columns already in use, being
mobile phase of acetone:ethanol:water: completely unsuitable for bilin separation.
acetic acid (38:50:11:1 vol/vol/vol/vol) This normally manifests itself as an inabil-
with a flow rate of 4 mL/minute at 30°C ity to retain the bilins on the column, and
(8,17). An almost identical system (ace- this difference in performance is even
tone:ethanol:water:acetic acid, 34:48:17:1 apparent between columns from the same
vol/vol/vol/vol) has also proved suitable for company. However, ODS columns from a
separating phytochromobilins, in this case number of companies (some of which have
using an Ultrasphere ODS column (250 × been named above) have been used success-
10 mm, 5 µm particle size; Beckman Coul- fully, and researchers new to the field
ter, Fullerton, CA, USA) at room tempera- should persevere in finding a suitable one.
ture (57). However, one potentially serious The solvent systems described above
problem with this solvent system is that have not only been used for separating
there is almost no resolution of BV IXα phycobilins, but are also suitable for ana-
and 3(Z)-PΦB. To solve this problem, lyzing both BV IXα and MBV IXα syn-
some improvements on this system have thesized from heme (54,62) and mesoheme
been developed in the Lagarias laboratory. (unpublished results), respectively. Howev-
The first of these, changing the mobile er, a simpler system comprising 95%
phase to acetone:ethanol:100 mM formic (vol/vol) aqueous ethanol:acetic acid:water
acid (25:65:10 vol/vol/vol), has been used (92:1:7 vol/vol/vol) has also commonly
successfully for the analysis of PΦB synthe- been used (16,45). Both systems are also
sis using a Supercosil LC-18 ODS column capable of separating the 4 BV IX isomers.
(250 × 4.6 mm, 5 µm particle size; Supel- For the analysis of BR IXα formation from
co, Bellefonte, PA, USA) with a flow rate BV IXα by cyanobacterial BV reductase,
of 1.5 mL/minute (54,62). However, even Schluchter and Glazer (49) used a linear
281
M.J. Terry
3(Z)-phycocyanobilin 369, 686 (9) ε368 = 46, 774 (63)e 63e,f, 9e,f
3(E)-phycocyanobilin 375, 692 (9) ε374 = 47, 900 (13)e 19, 2, 30, 28e
282
Biosynthesis and Analysis of Bilins
lished the pathway for bilin biosynthesis in partially purified using differential ammo-
photosynthetic organisms has utilized the nium sulfate precipitation and DEAE-cel-
classical biochemical approach of feeding lulose and blue-Sepharose affinity chro-
putative bilin precursors. Using these tech- matography steps (16). This procedure was
niques, and in particular taking advantage subsequently extended to include a ferre-
of radiolabeled precursors, it was estab- doxin-Sepharose affinity chromatography
lished that the phycobilins were derived step to give a 200-fold purification of heme
from heme synthesized from ALA by the oxygenase activity (46).
main tetrapyrrole pathway. These experi- The employment of a ferredoxin affinity
ments are discussed by Beale (5). Similarly, purification step highlights one of the dis-
ALA and BV IXα were also shown to be tinguishing features of heme oxygenases
precursors of phytochromobilin (reviewed from photosynthetic organisms: the
in Reference 58), though only recently was requirement for reduced ferredoxin, pro-
the intermediacy of heme confirmed (62). duced by ferredoxin-NADP+ oxidoreduc-
While this type of approach will continue tase and NADPH, for enzyme activity. In
to be important in bilin research, this sec- contrast, mammalian heme oxygenase (first
tion will concentrate on assaying the described in Reference 53), found associat-
enzymes committed to bilin synthesis. ed with microsomal membranes, utilizes an
NADPH-cytochrome P450 reductase (35,
4.1. Heme Oxygenase 69). A second distinguishing feature is that
the algal enzyme requires a second reduc-
Heme oxygenase catalyses the synthesis tant in addition to ferredoxin (16): ascor-
of BV IXα from protoheme in a reaction bate appears to be the most effective with
that requires O2 and reducing power, and Trolox (6-hydroxy-2,5,7,8-tetramethyl-
liberates CO and Fe2+ (Figure 3) (34,42). chroman-2-carboxylic acid) also proving
In photosynthetic organisms, the most suitable (46). The reason that the reaction
extensively characterized heme oxygenase requires a second reductant is currently
system is that of C. caldarium (5). Heme unknown.
oxygenase activity was first measured in The heme oxygenase assay in extracts or
Cyanidium extracts by Beale and Cornejo partially purified fractions of Cyanidium is
(7). The enzyme is soluble and has been an HPLC-based assay that takes advantage
Figure 3. The reaction catalyzed by the enzyme heme oxygenase. Substrates and products are not shown stoichiometrically.
284
Biosynthesis and Analysis of Bilins
of the observation that MBV IXα, in con- (18). However, in most other respects, the
trast to BV IXα, is not further metabolized cyanobacterial heme oxygenase appears to
in these preparations. A typical reaction be similar to its algal counterpart (15,18).
buffer contains 10 µM mesohemin IX in One important consideration when
20 mM Tris-HCl, pH 7.6, with 5 mM L- assaying heme oxygenase is product identi-
ascorbate, 0.1 mg/mL catalase, 0.5 mg/mL fication. Heme readily undergoes coupled
BSA, 10% (wt/vol) glycerol, and reducing oxidation to a mixture of BV IXα, β, γ,
power provided by an NADPH-regenerat- and δ (particularly in the presence of ascor-
ing system (10 mM glucose 6-phosphate, bate), and it is necessary to establish that
0.5 mM NADP+, and 0.3 U/mL glucose- the reaction product is exclusively the IXα
6-phosphate dehydrogenase), together with isomer in order to demonstrate that BV
5 µg/mL Porphyra umbilicalis ferredoxin, synthesis is enzymatic. This can be done by
and 0.01 U/mL spinach ferredoxin- HPLC as described in section 3.1. To pre-
NADP+ reductase (46). O2 is also required vent the formation of additional BV IX
for the reaction to proceed. Mesohemin isomers from H2O2 generated by the reac-
can be obtained from Porphyrin Products, tion of ascorbate and O2, catalase is rou-
while all the other components are readily tinely included in the reaction mixture (7).
available from Sigma. In the Beale labora- Heme oxygenase can be stimulated by the
tory, assays are performed at 42°C for 1 addition of strong iron chelators such as
hour in the dark, and the bilin products are desferrioxamine or Tiron (4,5-dihydroxy-
extracted into dichloromethane:1-butanol 1,3-benzene disulphonic acid). These are
(2:1 vol/vol) and concentrated on DEAE- thought to aid the dissociation of an
Sepharose prior to HPLC analysis (46). Fe(III)-BV IXα complex that may be the
The electrons required for the reaction primary product in vitro (46). A number
are transferred directly from the ferredox- of inhibitors are also available. Sn-proto-
in, as evidenced by the observation that porphyrin IX is a potent inhibitor of the
light-driven ferredoxin reduction via a algal enzyme (16), while the animal
spinach photosystem I fraction can support enzyme is inhibited by a wide range of
heme oxygenase activity (45). As there is metal porphyrins (60) that have yet to be
likely to be a direct interaction between tested on heme oxygenases from photosyn-
heme oxygenase and the ferredoxin, it is thetic organisms. Heme oxygenase is also
unsurprising that the source of ferredoxin inhibited by diethylpyrocarbonate, proba-
is important for maximum activity. Por- bly through binding to active-site his-
phyra ferredoxin is almost as good as a tidines, but is insensitive to the sulfhydryl
ferredoxin-containing fraction from Cyani- reagent p-hydroxymercuribenzoate (16).
dium (86%), but spinach ferredoxin only In preparations from higher plants,
resulted in 20% of activity (16). There is heme oxygenase has only been assayed in
also considerable variation in the effective- isolated plastids (62). The method is essen-
ness of different second reductants. L- and tially the same as that described in Proce-
D-ascorbate were equally effective, while dure 2, except that 10 µM heme replaces
dehydroascorbate could not support any the BV, and the O2 depletion step is not
activity. Trolox (29% of the activity with required. Under these assay conditions, BV
ascorbate), dihydroquinone (16%), and IXα is completely metabolized to 3(Z)-
DTT (15%) were all partially active (46). PΦB. However, as for the algal system,
In the cyanobacterium Synechocystis sp. mesoheme is also a substrate for the plant
PCC 6701, Trolox was actually more effec- heme oxygenase, while the product MBV
tive as a second reductant than ascorbate IXα is not further metabolized (unpub-
285
M.J. Terry
287
M.J. Terry
is not entirely clear, and it is unknown activity has been measured in isolated plas-
whether algae or higher plants also have tids from oat (57), pea (62), and tomato
this enzyme. (54). The conditions for measuring this
Algae appear to have at least four differ- enzyme in isolated plastids have been
ent bilin-reducing enzymes for which the described in Procedure 2. Recently,
substrate specificity has yet to be clearly progress has been made in purifying PΦB
defined (Figure 2). The best characterized synthase (38). This work has revealed that
of these are the enzymes from the red alga the enzyme is soluble, has an apparent
C. caldarium. Beale and Cornejo (6) molecular mass of 29 kDa, and functions
showed that cell-free extracts of Cyanidium in the same ferredoxin-dependent manner
reduced BV IXα to give the final products as the algal (45) and cyanobacterial (15)
of 3(Z)- and 3(E)-PCB. Subsequently, reductases.
these authors identified two soluble activi-
ties, one of which reduced BV IXα to 4.3. Bilin Isomerases
15,16 DHBV, while a second reduced
15,16 DHBV to 3(Z)-PEB (9,10). 3(Z)- Little is known about the bilin isomeras-
PEB is then further isomerized to PCB (see es. Two types of isomerase activity have
section 4.3.). In contrast to BV reductase, been reported. A glutathione-dependent
the two bilin reductases required reduced 3(Z) to 3(E) cis-trans isomerase was identi-
ferredoxin for activity (8,45). The assay fied in Cyanidium extracts (9), and a simi-
conditions used were as follows: 25 mM lar activity has been proposed to exist in
HEPES buffer, pH 7.3, 10 µM BV IXα, plants as incubation of BV IXα with plas-
10% (vol/vol) glycerol, 1 mM MgCl2, tids results in the synthesis, first of 3(Z)-
approximately 1 mg/mL BSA, 3000 U/mL PΦB and then of 3(E)-PΦB (57). It is still
catalase, and reducing power in the form of not entirely clear whether this is an
an NADPH-regenerating system together enzyme-catalyzed reaction, as it may sim-
with ferredoxin and ferredoxin NADP+ ply reflect the fact that the (Z)-isomer is
reductase (see conditions for the heme oxy- chemically less stable than the (E)-isomer.
genase assay in section 4.1). Following However, cyanobacterial extracts synthe-
incubation at 40°C for 30 minutes, the sized 3(Z)-PCB, but not 3(E)-PCB, from
bilin products were extracted and analyzed BV IXα supporting a role for a specific
by HPLC as described in section 3.1. 3(Ζ) to 3(E) cis-trans isomerase in other
Incubation of BV IXα with soluble pro- organisms (15). The second isomerase
tein extracts from plastids of the green alga activity converts PEB to PCB with both
M. caldariorum under similar assay condi- the (Z)- and (E)-isomers serving as sub-
tions resulted in completely different prod- strates (9). In this case, the reaction was
ucts. In this case, 3(Z)-PΦB and 3(Z)-PCB catalyzed by a high molecular weight
were synthesized sequentially (67). This (>60 kDa) protein fraction, but did not
result suggests the presence of a homolog require NADPH or reduced ferredoxin for
of the plant enzyme, PΦB synthase, and a activity (9).
previously undescribed 181,182-reductase
(see Figures 1 and 2). PΦB synthase activi-
ty was first detected in cucumber using a ABBREVIATIONS
coupled assay to measure PΦB synthesis by
assembly with apophytochrome (55). This ALA, 5-aminolevulinic acid; BV,
assay has now been superceded by an biliverdin; BR, bilirubin; BSA, bovine
HPLC-based assay and PΦB synthase serum albumin; DHBV, dihydrobiliverdin;
288
Biosynthesis and Analysis of Bilins
DMSO, dimethyl sulfoxide; MBV, meso- 10.Beale, S.I. and J. Cornejo. 1991. Biosynthesis of phy-
biliverdin; PCB, phycocyanobilin; PEB, cobilins. 15,16-dihydrobiliverdin IXα is a partially
reduced intermediate in the formation of phycobilins
phycoerythrobilin; PΦB, phytochromo- from biliverdin IXα. J. Biol. Chem. 266:22341-
bilin; TFA, trifluoroacetic acid. 22345.
11.Brockmann, H., Jr. and G. Knobloch. 1973. Die
absolute Konfiguration des 2E-Äthyliden-3-methyl-
succinimids. Ein Beitrag zur Bestimmung der
ACKNOWLEDGMENTS absoluten Konfiguration von Phycobilinen und Phy-
tochrom. Chem. Ber. 106:803-811.
12.Chapman, D.J., W.J. Cole, and H.W. Siegelman.
I would like to thank the Royal Society 1967. The structure of phycoerythrobilin. J. Am.
for their support through a Royal Society Chem. Soc. 89:5976-5977.
13.Cole, W.J., D.J. Chapman, and H.W. Siegelman.
University Research Fellowship, Professors 1967. The structure of phycocyanobilin. J. Am. Chem.
Sam Beale and Peter Shoolingin-Jordan for Soc. 89:3643-3645.
reading this manuscript prior to publica- 14.Cole, W.J., D.J. Chapman, and H.W. Siegelman.
1968. The structure and properties of phyco-
tion, Mark Milford for checking the Proce- cyanobilin and related bilatrienes. Biochem. 7:2929-
dures, and Professor Clark Lagarias for the 2935.
opportunity to learn about the wonderful 15.Cornejo, J. and S.I. Beale. 1997. Phycobilin biosyn-
thetic reactions in extracts of cyanobacteria. Photosyn.
world of bilins. Res. 51:223-230.
16.Cornejo, J. and S.I. Beale. 1988. Algal heme oxyge-
nase from Cyanidium caldarium. Partial purification
REFERENCES and fractionation into three required protein compo-
nents. J. Biol. Chem. 263:11915-11921.
1.Arciero, D.M., D.A. Bryant, and A.N. Glazer. 1988. 17.Cornejo, J., S.I. Beale, M.J. Terry, and J.C. Lagarias.
In vitro attachment of bilins to apophycocyanin. I. 1992. Phytochrome assembly. The structure and bio-
Specific covalent adduct formation at cysteinyl residues logical activity of 2(R),3(E)-phytochromobilin derived
involved in phycocyanobilin binding in C-phyco- from phycobiliproteins. J. Biol. Chem. 267:14790-
cyanin. J. Biol. Chem. 263:18343-18349. 14798.
2.Arciero, D.M., J.L. Dallas, and A.N. Glazer. 1988. 18.Cornejo, J., R.D. Willows, and S.I. Beale. 1998.
In vitro attachment of bilins to apophycocyanin. II. Phytobilin biosynthesis: cloning and expression of a
Determination of the structures of tryptic bilin pep- gene encoding soluble ferredoxin-dependent heme
tides derived from the phycocyanobilin adduct. J. Biol. oxygenase from Synechocystis sp. PCC 6803. Plant J.
Chem. 263:18350-18357. 15:99-107.
3.Arciero, D.M., J.L. Dallas, and A.N. Glazer. 1988. 19.Crespi, H.L., L.J. Boucher, G.D. Norman, J.J. Katz,
In vitro attachment of bilins to apophycocyanin. III. and R.C. Dougherty. 1967. Structure of phyco-
Properties of the phycoerythrobilin adduct. J. Biol. cyanobilin. J. Am. Chem. Soc. 89:3642-3643.
Chem. 263:18358-18363. 20.Crespi, H.L. and J.J. Katz. 1969. Exchangeable hydro-
4.Austin, C.C. and K.W. Jessing. 1994. Green-blood gen in phycoerythrobilin. Phytochem. 8:759-761.
pigmentation in lizards. Comp. Biochem. Physiol. 21.Davis, S.J., J. Kurepa, and R.D. Vierstra. 1999. The
109A:619-626. Arabidopsis thaliana HY1 locus, required for phy-
5.Beale, S.I. 1993. Biosynthesis of phycobilins. Chem. tochrome-chromophore biosynthesis, encodes a pro-
Rev. 93:785-802. tein related to heme oxygenases. Proc. Natl. Acad. Sci.
6.Beale, S.I. and J. Cornejo. 1984. Enzymic Transfor- USA 96:6541-6546.
mation of biliverdin to phycocyanobilin by extracts of 22.Elich, T.D., A.F. McDonagh, L.A. Palma, and J.C.
the unicellular red alga Cyanidium caldarium. Plant Lagarias. 1989. Phytochrome chromophore biosynthe-
Physiol. 76:7-15. sis. Treatment of tetrapyrrole-deficient Avena explants
7.Beale, S.I. and J. Cornejo. 1984. Enzymatic heme with natural and non-natural bilatrienes leads to for-
oxygenase activity in soluble extracts of the unicellular mation of spectrally active holoproteins. J. Biol. Chem.
red alga, Cyanidium caldarium. Arch. Biochem. Bio- 264:183-189.
phys. 235:371-384. 23.Fang, L.-S. and J.L. Bada. 1990. The blue-green
8.Beale, S.I. and J. Cornejo. 1991. Biosynthesis of phy- blood plasma of marine fish. Comp. Biochem. Physiol.
cobilins. Ferredoxin-mediated reduction of biliverdin 97B:37-45.
catalyzed by extracts of Cyanidium caldarium. J. Biol. 24.Fu, E., L. Friedman, and H.W. Siegelman. 1979.
Chem. 266:22328-22332. Mass-spectral identification and purification of phy-
9.Beale, S.I. and J. Cornejo. 1991. Biosynthesis of phy- coerythrobilin and phycocyanobilin. Biochem. J. 179:
cobilins. 3(Z)-phycoerythrobilin and 3(Z)-phyco- 1-6.
cyanobilin are intermediates in the formation of 3(E)- 25.Glazer, A.N. 1989. Light guides. Directional energy
phycocyanobilin from biliverdin IXα. J. Biol. Chem. transfer in a photosynthetic antenna. J. Biol. Chem.
266:22333-22340. 264:1-4.
289
M.J. Terry
26.Glazer, A.N. and G.J. Wedemayer. 1995. Cryptomon- californica: concentration and storage of phycoerythro-
ad biliproteins—an evolutionary perspective. Photosyn. bilin in the ink gland. J. Exp. Biol. 201:1595-1613.
Res. 46:93-105. 44.Provasoli, L., J.J.A. McLaughlin, and M.R. Droop.
27.Goodman, W.G., B. Adams, and J.T. Trost. 1985. 1957. The development of artificial media for marine
Purification and characterization of a biliverdin-associ- algae. Archiv. Mikrobiol. 25:392-428.
ated protein from the hemolymph of Manduca Sexta. 45.Rhie, G. and S.I. Beale. 1992. Biosynthesis of Phyco-
Biochem. 24:1168-1175. bilins. Ferredoxin-supported NADPH-independent
28.Gossauer, A. and W. Hirsch. 1974. Totalsynthese des heme oxygenase and phycobilin-forming activities
racemischen Phycocyanobilins (Phycobiliverdins) from Cyanidium caldarium. J. Biol. Chem. 267:16088-
sowie eines “Homophycobiliverdins”. Liebigs Ann. 16093.
Chem. 1974:1496-1513. 46.Rhie, G. and S.I. Beale. 1995. Phycobilin biosynthesis:
29.Gossauer, A. and J.-P. Weller. 1978. Chemical total reductant requirements and product identification for
synthesis of (+)-(2R,16R)- and (+)-(2S, 16R)-phycoery- heme oxygenase from Cyanidium caldarium. Arch.
throbilin dimethyl ester. J. Am. Chem. Soc. 100:5928- Biochem. Biophys. 320:182-194.
5933. 47.Rippka, R., J. Deruelles, J.B. Waterbury, M. Herd-
30.Kakiuchi, T., H. Kato, K.P. Jayasundera, T. Higashi, man, and R.Y. Stanier. 1979. Generic assignments,
K. Watabe, D. Sawamoto, H. Kinoshita, and K. Ino- strain histories and properties of pure cultures of
mata. 1998. Total syntheses of (±)-phycocyanobilin cyanobacteria. J. Gen. Microbiol. 111:1-61.
and its derivatives bearing a photoreactive group at D- 48.Ryter, S., E. Kvam, and R.M. Tyrell. 1999. Heme
ring. Chem. Lett. 1998:1001-1002. oxygenase activity determination by high-performance
31.Kennedy, G.Y. and H.G. Vevers. 1976. A survey of liquid chromatography. Methods Enzymol. 300:322-
avian eggshell pigments. Comp. Biochem. Physiol. 336.
55B:117-123. 49.Schluchter, W.M. and A.N. Glazer. 1997. Character-
32.Kutty, R.K. and M.D. Maines. 1981. Purification and ization of cyanobacterial biliverdin reductase. J. Biol.
characterization of biliverdin reductase from rat liver. Chem. 272:13562-13569.
J. Biol. Chem. 256:3956-3962. 50.Singleton, J.W. and L. Laster. 1965. Biliverdin reduc-
33.Lagarias, D.M., M.W. Crepeau, M.D. Maines, and tase of guinea pig liver. J. Biol. Chem. 240:4780-4789.
J.C. Lagarias. 1997. Regulation of photomorphogen- 51.Stocker, R., Y. Yamamoto, A.F. McDonagh, A.N.
esis by expression of mammalian biliverdin reductase in Glazer, and B.N. Ames. 1987. Science 235:1043-
transgenic Arabidopsis plants. Plant Cell 9:675-688. 1046.
34.Maines, M.D. 1988. Heme oxygenase: function, mul- 52.Stoll, M.S. and C.H. Gray. 1977. The preparation
tiplicity, regulatory mechanisms, and clinical applica- and characterization of bile pigments. Biochem. J.
tions. FASEB J. 2:2557-2568. 163:59-101.
35.Maines, M.D., N.G. Ibrahim, and A. Kappas. 1977. 53.Tenhunen, R., H.S. Marver, and R. Schmid. 1968.
Solubilization and partial purification of heme oxyge- The enzymatic conversion of heme to bilirubin by
nase from rat liver. J. Biol. Chem. 252:5900-5903. microsomal heme oxygenase. Proc. Natl. Acad. Sci.
36.McDonagh, A.F. and F. Assisi. 1971. Commercial USA 61:748-755.
bilirubin: a trinity of isomers. FEBS Lett. 18:315-317. 54.Terry M.J. and R.E. Kendrick. 1996. The aurea and
37.McDonagh, A.F. and L.A. Palma. 1980. Preparation yellow-green-2 mutants of tomato are deficient in phy-
and properties of crystalline biliverdin IXα. Simple tochrome chromophore synthesis. J. Biol. Chem.
methods for preparing isomerically homogenous 271:21681-21686.
biliverdin and (14C) biliverdin by using 2,3-dichloro- 55.Terry, M.J. and J.C. Lagarias. 1991. Holophy-
5,6-dicyanobenzoquinone. Biochem. J. 189:193-208. tochrome assembly. Coupled assay for phytochromo-
38.McDowell, M.D. and J.C. Lagarias. 1997. Partial bilin synthesis in organello. J. Biol. Chem. 266:22215-
purification, photoaffinity labeling, and characteriza- 22221.
tion of phytochromobilin synthase. Plant Physiol. 56.Terry, M.J., M.D. Maines, and J.C. Lagarias. 1993.
114:S739. Inactivation of phytochrome- and phycobiliprotein-
39.Muramoto, T., T. Kohchi, A. Yokota, I. Hwang, and chromophore precursors by rat liver biliverdin reduc-
H.M. Goodman. 1999. The Arabidopsis photomor- tase. J. Biol. Chem. 268:26099-26106.
phogenic mutant hy1 is deficient in phytochrome 57.Terry, M.J., M.D. McDowell, and J.C. Lagarias.
chromophore biosynthesis as a result of a mutation in a 1995. (3Z)- and (3E)-phytochromobilin are interme-
plastid heme oxygenase. Plant Cell 11:335-347. diates in the biosynthesis of the phytochrome chro-
40.Murphy, J.T. and J.C. Lagarias. 1997. The phyto- mophore. J. Biol. Chem. 270:11111-11119.
fluors: a new class of fluorescent protein probes. Curr. 58.Terry M.J., J.A. Wahleithner, and J.C. Lagarias.
Biol. 7:870-876. 1993. Biosynthesis of the plant photoreceptor phy-
41.Oren, D.A. 1997. Bilirubin, rem sleep, and photo- tochrome. Arch. Biochem. Biophys. 306:1-15.
transduction of environmental time cues. A Hypothe- 59.Turner, L., J.D. Houghton, and S.B. Brown. 1992.
sis. Chronobiol. Int. 14:319-329. Isolation and partial purification of phycocyanin
42.Ortiz de Montellano, P.R. 1998. Heme oxygenase apoprotein and its role in studies of bilin-apoprotein
mechanism: evidence for an electrophilic, ferric perox- attachment. Plant Physiol. Biochem. 30:309-314.
ide species. Acc. Chem. Res. 31:543-549. 60.Vreman, H.J., D.A. Cipkala, and D.K. Stevenson.
43.Prince, J., T.G. Nolen, and L. Coelho. 1998. Defen- 1996. Characterization of porphyrin heme oxygenase
sive ink pigment processing and secretion in Aplysia inhibitors. Can. J. Physiol. Pharmacol. 74:278-285.
290
Biosynthesis and Analysis of Bilins
61.Vreman, H.J. and D.K. Stevenson. 1988. Heme oxy- ly active chromophore precursor of the plant photore-
genase activity as measured by carbon monoxide pro- ceptor phytochrome. Proc. Natl. Acad. Sci. USA
duction. Anal. Biochem. 168:31-38. 93:8989-8994.
62.Weller, J.L., M.J. Terry, C. Rameau, J.B. Reid, and 67.Wu, S.-H., M.T. McDowell, and J.C. Lagarias.
R.E. Kendrick. 1996. The phytochrome-deficient pcd1 1997. Phycocyanobilin is the natural precursor of
mutant of pea is unable to convert heme to biliverdin the phytochrome chromophore in the green alga
IXa. Plant Cell 8:55-67. Mesotaenium caldariorum. J. Biol. Chem. 272:25700-
63.Weller, J.-P. and A. Gossauer. 1980. Synthese und 25705.
photoisomerisierung des racem. Phytochromobilin- 68.Yamaguchi, T., Y. Komoda, and H. Nakajima.
dimethylesters. Chem. Ber. 113:1603-1611. 1994. Biliverdin-IXα and biliverdin IXβ reductase
64.Wilks, A. and P.R. Ortiz de Montellano. 1993. Rat from human liver. J. Biol. Chem. 269:24343-
liver heme oxygenase. High level expression of a trun- 24348.
cated soluble form and nature of the meso-hydroxylat- 69.Yoshida, T. and G. Kikuchi. 1978. Features of the
ing species. J. Biol. Chem. 268:22357-22362. reaction of heme degradation catalyzed by the recon-
65.Wilks, A. and M.P. Schmitt. 1998. Expression and stituted microsomal heme oxygenase system. J. Biol.
characterization of a heme oxygenase (Hmu O) from Chem. 253:4230-4236.
Corynebacterium diphtheriae. Iron acquisition requires 70.Yoshida, T., M. Noguchi, and G. Kikuchi. 1982. The
oxidative cleavage of the heme macrocycle. J. Biol. step of carbon monoxide liberation in the sequence of
Chem. 273:837-841. heme degradation catalyzed by the reconstituted
66.Wu, S.-H. and J.C. Lagarias. 1996. The methy- microsomal heme oxygenase system. J. Biol. Chem.
lotrophic yeast Pichia pastoris synthesizes a functional- 257:9345-9348.
291
Analysis and Reconstitution of
13 Phytochromes
293
M.T. McDowell and J.C. Lagarias
Figure 1. Domain structure of eukaryotic and cyanobacterial phytochromes and phytofluors. (A) Phytochromes exist in pho-
tointerconvertible red light absorbing (Pr) and far-red light absorbing (Pfr) forms. Phytochromes possess either phytochromobilin
or phycocyanobilin prosthetic groups bound to a conserved cysteine residue indicated with an asterisk. (B) Phytofluors are orange
fluorescent biliproteins consisting of phycoerythrobilin thioether-linked to the conserved cysteine residue on an apophytochrome.
Regulatory domains include the PAS-related domain (PRD), which contain 2 direct repeats shown as dark boxes, and the histi-
dine kinase domain (HKD) and the histidine kinase-related domain (HKRD), which are depicted with cross hatching.
294
Analysis and Reconstitution of Phytochromes
ences (19,46). Two major caveats about be performed in a laboratory and kept
phytochrome measurements need to be under dim green safelight (54). Second, the
made at this point. First, to avoid continu- validity of the phytochrome difference
ous photocycling, all measurements should assay depends upon the lack of other pig-
Figure 2. Absorption and difference spectra of purified recombinant oat phytochrome. (A) Absorption spectra of Pr and Pfr
forms of PΦB and PCB adducts of recombinant oat apophytochrome A (AsphyA). (B) Difference spectra of PΦB and PCB
adducts of recombinant AsphyA. Adapted from Murphy and Lagarias (44).
296
Analysis and Reconstitution of Phytochromes
ments that absorb in the 500 to 800 nm absorption maximum) no longer chan-
wavelength range. For this reason, chloro- ges with further illumination. For puri-
phyll must be removed from extracts, from fied phytochrome preparations, absor-
light-grown plant extracts which can be bances between 250 and 800 nm are
accomplished by precipitation with poly- recorded; for crude samples, a 500 to
ethyleneimine (41). The latter is not a con- 800 nm range is used. The wavelength
cern with recombinant phytochromes pre- of red light needed is dependent on the
pared from bacterial, yeast, or insect cell species of phytochrome under exami-
extracts. Our standard method for phy- nation. For phytochromes containing
tochrome photoassay is outlined in Proce- the PΦB chromophore, the red actinic
dure 1. Because actinic illumination and light should be 660 nm (± 5 nm), and
sample measurement can be performed for PCB-containing phytochrome, the
simultaneously, we recommend the use of actinic light should be less than 650
a diode array spectrophotometer, such as nm (e.g., we use 636 ± 5 nm). Actinic
the Model 8453 from Hewlett Packard light is typically provided by a 250 W
(Wilmington, DE, USA). The length of quartz halogen lamp filtered through
irradiation will also depend on the light an appropriate interference filter (Cori-
source. Typically, for a focused 250 W on, Franklin, MA, USA).
quartz halogen lamp, a 200-second irradia- 3. The sample is then irradiated with sat-
tion is sufficient. The limit of sensitivity for urating far-red light, and another ab-
this assay is approximately 10 nM phy- sorption spectrum is taken. Far-red
tochrome (i.e., 1 µg/mL protein). light is obtained using a 250 W quartz
halogen lamp filtered through an FRS
❖ Procedure 1. Phytochrome Photoassay 700 black plexiglas filter (Rohm and
Hass, Philadelphia, PA, USA).
1. The protein sample, either from plant 4. The difference spectrum and quantita-
tissue or reconstituted recombinant tion of photoactive phytochrome is
phytochrome (see later section), is obtained by mathematically subtract-
diluted or reconstituted in the desired ing the spectrum of the red light-irra-
amount of TEGE buffer [25 mM Tris- diated sample from the spectrum of the
HCl, pH 7.8, containing 25% (vol/ far-red light-irradiated sample (see Fig-
vol) ethylene glycol, 1 mM EDTA, 2 ure 2B). The quantitation is obtained
mM phenylmethylsulfonyl fluoride by measuring the ∆∆A from the Amax
(PMSF) and 1 mM dithiothreitol and the Amin of the difference spec-
(DTT)], typically 500 µL. During irra- trum. To correct for the incomplete
diation and the subsequent analysis, photoconversion of Pr to Pfr and back,
the sample is kept at 10° to 16°C. the ∆∆A is multiplied by 0.86. To cal-
2. The sample is transferred to a 1-cm culate the molar concentration of
pathlength spectrophotometric cu- phytochrome, the corrected ∆∆A value
vette, placed in the spectrophotometer, is entered into the Lambert-Beers Law
irradiated with saturating red light, and equation of Acorr = εcl, where ε665 =
an absorption spectrum is taken. Satu- 132 000 M-1cm-1 for purified oat phy-
rating illumination is defined as the tochrome A (31).
amount of illumination needed to pro- 5. To determine the relative purity for a
duce a photostationary state and is particular phytochrome preparation,
determined empirically when the ab- the specific absorbance ratio (SAR) is
sorbance at 650 to 665 nm (i.e., Pr’s determined. To do this, the full spec-
297
M.T. McDowell and J.C. Lagarias
spectrum, a small aliquot of concen- examined to date can catalyze thioether for-
trated recombinant Cph1 is added to mation with PΦB, as well as the phyco-
the PΦB synthase assay mixture. bilins, PCB, and phycoerythrobilin (PEB),
5. The assay was incubated an additional both precursors of the phycobiliproteins
20 to 30 minutes at room temperature (37). This work is in striking contrast with
to facilitate assembly, and then a differ- phycobiliprotein assembly, whose bilin liga-
ence spectrum is obtained as outlined tion requires separate lyases for proper
above. assembly (see Chapter 14 in this text by
Schluchter and Bryant). Analyses of bilin
assembly to apophytochrome have relied
3. ASSAYS FOR HOLOPHYTO- on 3 major tools: (i) difference spectro-
CHROME ASSEMBLY scopy (described above); (ii) zinc blot ana-
lysis; and (iii) fluorescence spectroscopy. In
3.1. History and Development the sections that follow, the latter 2 tools
will be highlighted.
Based on the pioneering work of Gard-
ner and collaborators (22), it is well estab-
lished that the chromophore and apophy- 3.2. Zinc Blot Assay for
tochrome biosynthetic pathways are Bilin Attachment
essentially independent. Plants treated with
inhibitors of 4-aminolevulinic acid (ALA) Early examination of bile pigments re-
biosynthesis, such as gabaculine or amino- vealed that bilatrienes form intensely fluo-
5-hexynoic acid, possess reduced amount rescent complexes with zinc ions in the
of photoactive holophytochrome, but accu- presence of iodine (1). This and related
mulate near normal amounts of apophy- observations have led to the examination of
tochrome (12,14,22). Using these in- fluorescent zinc complexes of bilin-linked
hibitors, it was shown that co-incubating polypeptides. The initial work was per-
the plants with ALA and BV IXα, as well formed using standard sodium dodecyl sul-
as the unnatural isomer BV XIIIα, could fate polyacrylamide gel electrophoresis
restore the levels of spectrally active phy- (SDS-PAGE) gels with the fluorescent
tochrome (15). The phycobiliprotein chro- products being observed following UV
mophore precursor PCB was also found to illumination (2). The purpose of SDS-
restore the levels of spectrally active phy- PAGE is to remove the unbound bilin
tochrome, albeit with a blue-shifted differ- from the protein-bound bilin. This
ence spectrum (see Figure 2B). Follow-up methodology, which can be used for
investigations utilizing apophytochrome phycobiliproteins as well (48), has been
isolated from inhibitor-treated plants indi- further improved by the extension to
cated that either the lyase enzyme copuri- electroblotted protein samples or the
fies with apophytochrome, or that apophy- “zinc blot” (37). The zinc blot is used as a
tochrome is itself a bilin lyase (13). The general diagnostic technique to assess the
development of recombinant apophyto- ligation competency of a particular
chrome expression systems by many labo- apophytochrome. The limit of sensitivity of
ratories (7,10,13,18,20,21,29,32,33,35,55, this technique is approximately 50 ng/cm
62,65) have corroborated the latter hypo- of phytochrome per lane for gels and
thesis that apophytochrome is a bilin lyase 12 ng/cm of phytochrome per lane for
that catalyzes thioether linkage formation blots. Procedure 2, for the zinc blot assay,
with bilins. All bonafide apophytochromes is based on 2 references (2,37).
299
M.T. McDowell and J.C. Lagarias
1. Protein samples to be analyzed are elec- The natural biological function of phy-
trophoresed in a SDS-polyacrylamide tochrome is to act as a light switch due to
gel using any standard procedure. After its ability to photointerconvert between the
electrophoresis, the gel is transblotted Pr and Pfr light-absorbing forms. Based on
to a polyvinylidene difluoride (PVDF) NMR and resonance Raman spectroscopic
membrane using any standard proce- analysis, this photoconversion has been
dure for electroblotting. proposed to involve the Z to E isomeriza-
NOTE: Nylon and nitrocellulose are con- tion of the C15 methine bridge double
siderably less effective owing to their bond (16,52). While the 2 known phy-
greater autofluorescence background tochrome chromophore precursors, PΦB
and large UV absorption, respectively. and PCB, possess this double bond, PEB
does not. Indeed, PEB adducts of apophy-
2. After electroblotting, the membrane is tochromes are inactive photochemically, a
washed briefly with deionized water. result that led to the hypothesis that such
3. After the water wash, the blot is trans- adducts might be fluorescent. That PEB
ferred to 1.3 M zinc acetate in deion- incubation with apophytochromes produce
ized water and incubated at room tem- intensely fluorescent adducts was first docu-
perature for roughly 30 minutes under mented in 1995 (39). This finding led to
reduced light. the development of a real-time kinetic assay
4. Just prior to visualization, the blot is for the study of phytochrome assembly that
rinsed with deionized water to remove is described below. This assay has not only
excess Zn2+ ion. enabled the determination of both bilin
5. The blot is visualized by placing on a binding and catalytic rate constants for the
long wavelength UV transilluminator reconstitution of holophytochrome with its
and photographing with Technical Pan natural chromophore (i.e., Kd ∼ 1 µM, kcat
film (Type 4415; Eastman Kodak, ∼ 0.25–0.3 s-1), but it has also facilitated
Rochester, NY, USA) using an RG-630 the analysis of potential inhibitors of this
cutoff filter (Schott, Laurel, NJ, USA) process, such as the PΦB precursor BV (i.e.,
(2-min exposure). Alternatively, the Ki ∼ 1 mM). The procedures for these
blot can be imaged using a Storm assays, summarized below, are based on the
860 PhosphorImager® (Molecular work of Li, Murphy, and Lagarias (39). Two
Dynamics, Sunnyvale, CA, USA) with major fluorescent assay methodologies are
the red laser in fluorescence mode. The described below, the standard and the com-
image obtained using the Storm can be petitive assays. The only difference between
analyzed using the ImageQuant soft- the kinetic assays is the data analysis. The
ware or converted to tagged image for- data analysis for the standard assay is out-
mat files (TIFF) and analyzed using a lined in Scheme 1. The analyses of the 2
types of competitive fluorescence assays are
program such as National Institutes of
outlined in Schemes 2 and 3, respectively.
Health (NIH) Image. For quantitative
analyses, a dilution series of known ❖ Procedure 4. Fluorescence Assay for
quantities of holophytochrome (or Holophytochrome Assembly
phycobiliprotein) should be included
on each blot. 1. The formation of the fluorescent PEB-
300
Analysis and Reconstitution of Phytochromes
Scheme 1. Kinetic analysis of holophytochrome assembly (adapted from Reference 39). The enzyme reaction of a bilin, typical-
ly PEB, with apoPC to produce a fluorescent product is shown in the diagram. Equation 1 is the integrated rate equation describ-
ing this reaction. Equation 2 defines the kinetics of bilin–apophytochrome adduct formation. If the bilin precursor concentration
is kept essentially constant, by providing a large excess, semilog plots of the fraction of apophytochrome remaining are expected to
be linear as described in Equation 3. Equation 4 and its reciprocal, Equation 5, provide a means for determining the affinity of
apophytochrome for a particular bilin, Kbilin, and the rate constant, or turnover, of apophytochrome for a particular bilin, k2.
301
M.T. McDowell and J.C. Lagarias
Scheme 2. Reversible competitive inhibition of PEB phytochrome assembly. A kinetic model for PEB adduct formation in the
presence of a reversible competitive inhibitor such as BV. The kinetics of PEB adduct formation should be pseudo-first-order as
predicted by Equation 7. The raw fluorescence data is transformed and replotted as described by Equation 8. The slopes of these
semilog replots yield kapp values. These values are used to construct the plot described by Equation 10. The x-intercept of Equa-
tion 10 yields an estimate of the equilibrium dissociation constant for the reversible competitive inhibitor.
303
M.T. McDowell and J.C. Lagarias
Scheme 3. Irreversible inhibition of PEB phytochrome assembly. A kinetic model for PEB adduct formation in the presence of
an irreversible competitive inhibitor such as PΦB. The formation of both PEB-phytochrome and competitor bilin–phytochrome
are described by Equations 11 and 12. In the presence of large molar excesses of all bilins, these equations are first-order expres-
sions. The kappi values are calculated using Equation 13, then plotted as a function of the competitive inhibitor according to
Equations 14 and 15. A plot of Equation 15 yields the dissociation constant (Ki) and the catalytic rate constant (k4) for the com-
petitive inhibitor of fluorescent adduct formation.
304
Analysis and Reconstitution of Phytochromes
switching of the C17 and C18 methyl and required for assembly, as demonstrated by
vinyl moieties and elaboration of the C18 the binding of PEB to apophytochome.
side chain reveal that the bilin binding BVs and bilirubins, which lack the A-ring
pocket of apophytochrome is not very dis- ethylidene moiety, also do not form cova-
cerning with regard to the C18 substituent lent adducts with phytochrome, although
(15,37,40). The C15 methine bridge is not the former are capable of noncovalent
Figure 3. Fluorescence assay for holophytochrome assembly. Representative data for standard fluorescence analysis of PEB
attachment to recombinant oat phytochrome A, after Li et al. (39). The upper panel shows raw fluorescence kinetic data as a func-
tion of increasing PEB concentration. The middle panel is a replot of the same data according to Equation 3 (Scheme 1), from
which kapp values were estimated. The bottom panel depicts a replot of 1/kapp versus 1/(PEB) according to Equation 5 (Scheme
1). See text and Reference 39 for details.
305
M.T. McDowell and J.C. Lagarias
interaction with phytochrome and, there- presently unresolved. However, the observed
fore, can act as reversible competitive bilin lyase activity recombinant pea apophy-
inhibitors as discussed above. The require- tochrome mutants, in which this histidine
ment for both propionic acid moieties has residue was changed to a glutamine or argi-
been established by the inability of bilin– nine residue, suggest otherwise (3,58).
esters to bind to apophytochrome (3). Ongoing studies to identify catalytically
important residues for bilin lyase activity
4.2. Apophytochrome Specificity for will take advantage of the growing family of
Thioether Linkage Formation phytochrome-related proteins in cyanobac-
teria, in which deletions, insertions, and
While bilin specificity has been actively amino acid substitutions, which influence
addressed, less is known about the regions of bilin ligation, can be assessed.
the apophytochrome required for bilin
attachment. Thus far, the only unequivocal 4.3. Assembly of Recombinant
requirement is the conserved cysteine, Phytochromes In Vivo
through which the bilin forms its thioether
linkage (cys-321 in the case of oat PHYA3) Recently, recombinant expression of
(3,4). Much effort has been directed at try- phytochrome led to a novel application
ing to determine other amino acid residues (11,28,38,67). Phytochrome expressed in
or regions of the protein that are involved a heterologous system such as Saccharo-
either directly or indirectly with the lyase myces cerevisiae could be assembled in vivo
activity. Deletion analysis of phytochrome if the chromophore was supplied exoge-
and expression of the truncated proteins in nously. The key stumbling block was get-
either E. coli or yeast have established that ting the chromophore into living cells.
neither the first 68 amino acids nor the This could be accomplished by dissolving
entire C-terminal domain are required for the chromophore precursor in DMSO,
the autocatalytic assemble of recombinant which was added to a minimal buffer
phytochromes (10,20). With recombinant medium at a final concentration of 50 µM
rice apophytochrome however, the deletion (38). The dissolved chromophore was then
of the first 80 residues abolished bilin bind- diluted in the appropriate buffer to no
ing (62). Site-directed mutagenesis of the more than 10% (vol/vol). The cells were
region surrounding the conserved cysteine able to take up the bilin that assembled
attachment site has been undertaken by sev- with the recombinant phytochrome, while
eral groups (3,9,49,58). These experiments the cells remained viable. The ability to
have so far failed to identify other residues reconstitute holophytochromes in living
essential for bilin assembly, although an cells provides a powerful tool for struc-
important role for the histidine residue adja- ture–function analysis of this photorecep-
cent to the conserved cysteine (i.e., H324 in tor family in nonplant cell systems and has
pea PHYA) has been proposed based of the also led to the development of a new fami-
loss-of-function of site-directed mutants of ly of apophytochrome-based fluorescent
this histidine residue (3,9,49). Interestingly, probes called phytofluors (43).
for the Cph1-related bacteriophytochrome
BphP from Deinococcus radiodurans, which 4.4. Phytofluors: A New Class of
lacks the conserved cysteine residue, this Fluorescent Protein Probe
adjacent histidine appears to be the site of
bilin binding (8). Whether this histidine Phytofluors are intensely orange fluores-
represents a catalytically important residue is cent adducts that are formed spontaneous-
306
Analysis and Reconstitution of Phytochromes
307
M.T. McDowell and J.C. Lagarias
13.Elich, T.D. and J.C. Lagarias. 1989. Formation of a 29.Kunkel, T., K. Tomizawa, R. Kern, M. Furuya, N.H.
photoreversible phycocyanobilin-apophytochrome Chua, and E. Schafer. 1993. In vitro formation of a
adduct in vitro . J. Biol. Chem. 264:12902-12908. photoreversible adduct of phycocyanobilin and tobac-
14.Elich, T.D. and J.C. Lagarias. 1987. Phytochrome co apophytochrome B. Eur. J. Biochem. 215:587-594.
chromophore biosynthesis. Both 5-aminolevulinic acid 30.Lagarias, D.M., S.H. Wu, and J.C. Lagarias. 1995.
and biliverdin overcome inhibition by gabaculine in Atypical phytochrome gene structure in the green alga
etiolated Avena sativa L. seedlings. Plant Physiol. mesotaenium caldariorum. Plant Mol. Biol. 29:1127-
84:304-310. 1142.
15.Elich, T.D., A.F. McDonagh, L.A. Palma, and J.C. 31.Lagarias, J.C., J.M. Kelly, K.L. Cyr, and W.O. Smith,
Lagarias. 1989. Phytochrome chromophore biosyn- Jr. 1987. Comparative photochemical analysis of high-
thesis. Treatment of tetrapyrrole-deficient Avena explants ly purified 124 kilodalton oat and rye phytochromes in
with natural and non-natural bilatrienes leads to for- vitro. Photochem. Photobiol. 46:5-13.
mation of spectrally active holoproteins. J. Biol. Chem. 32.Lagarias, J.C. and D.M. Lagarias. 1989. Self assembly
264:183-189. of synthetic phytochrome holoprotein in vitro. Proc.
16.Fodor, S.P.A., J.C. Lagarias, and R.A. Mathies. 1990. Natl. Acad. Sci. USA 86:5778-5780.
Resonance Raman analysis of the Pr and Pfr forms of 33.Lagarias, J.C. and F.M. Mercurio. 1985. Structure
phytochrome. Biochemistry 29:11141-11146. function studies on phytochrome. Identification of
17.Furuya, M. 1993. Phytochromes—their molecular light-induced conformational changes in 124-kDa
species, gene families, and functions. Ann. Rev. Plant Avena phytochrome in vitro. J. Biol. Chem. 260:2415-
Physiol. Plant Mol. Biol. 44:617-645. 2423.
18.Grimm, R., G.K. Donaldson, SM Vandervies, E. 34.Lagarias, J.C. and H. Rapoport. 1980. Chromopep-
Schafer, and A.A. Gatenby. 1993. Chaperonin-medi- tides from phytochrome. The structure and linkage of
ated reconstitution of the phytochrome photoreceptor. the Pr form of the phytochrome chromophore. J. Am.
J. Biol. Chem. 268:5220-5226. Chem. Soc. 102:4821-4828.
19.Gross, J., M. Seyfried, L. Fukshansky, and E. Schaefer. 35.Lamparter, T., F Mittmann, W. Gartner, T. Borner,
1984. In vivo spectrophotometry, p. 131-158. In H. E. Hartmann, and J. Hughes. 1997. Characterization
Smith and M.G. Holmes (Eds.), Techniques in Photo- of recombinant phytochrome from the cyanobacteri-
morphogenesis. Academic Press, New York. um Synechocystis. Proc. Natl. Acad. Sci. USA
20.Hill, C., W. Gartner, P. Towner, S.E. Braslavsky, and 94:11792-11797.
K. Schaffner. 1994. Expression of phytochrome 36.Lapko, V.N. and P.S. Song. 1995. A simple and
apoprotein from Avena sativa in Escherichia coli and improved method of isolation and purification for
formation of photoactive chromoproteins by assembly native oat phytochrome. Photochem. Photobiol.
with phycocyanobilin. Eur. J. Biochem. 223:69-77. 62:194-198.
21.Hughes, J., T. Lamparter, F. Mittmann, E. Hart- 37.Li, L. and JC Lagarias. 1992. Phytochrome assem-
mann, W. Gartner, A. Wilde, and T. Borner. 1997. A bly—defining chromophore structural requirements
prokaryotic phytochrome. Nature 386:663-663. for covalent attachment and photoreversibility. J. Biol.
22.Jones, A.M., C.D. Allen, G. Gardner, and P.H. Quail. Chem. 267:19204-19210.
1986. Synthesis of phytochrome apoprotein and chro- 38.Li, L. and J.C. Lagarias. 1994. Phytochrome assembly
mophore are not coupled obligatorily. Plant Physiol. in living cells of the yeast Saccharomyces cerevisiae. Proc.
81:1014-1016. Natl. Acad. Sci. USA 91:12535-12539.
23.Jones, A.M., and M.D. Edgerton. 1994. The anatomy 39.Li, L., J.T. Murphy, and J.C. Lagarias. 1995. Contin-
of phytochrome, a unique photoreceptor in plants. uous fluorescence assay of phytochrome assembly in
Sem. Cell Biol. 5:295-302. vitro. Biochem. 34:7923-7930.
24.Jones, A.M., R.D. Vierstra, S.M. Daniels, and P.H. 40.Lindner, I., B. Knipp, S.E. Braslavsky, W. Gartner,
Quail. 1985. The role of separate domains in the struc- and K. Schaffner. 1998. A novel chromophore selec-
ture of phytochrome from etiolated Avena sativa L. tively modifies the spectral properties of one of the two
Planta. 164:505-506. stable states of the plant photoreceptor phytochrome.
25.Kehoe, D.M. and A.R. Grossman. 1996. Similarity of Angew. Chem., Int. Ed. 37:1843-1846.
a chromatic adaptation sensor to phytochrome and 41.Litts, J.C., J.M. Kelly, and J.C. Lagarias. 1983. Struc-
ethylene receptors. Science 273:1409-1412. ture-function studies on phytochrome. Preliminary
26.Kendrick, R.E. and G.H.M. Kronenberg (Eds.). characterization of highly purified phytochrome from
1994. Photomorphogenesis in Plants, 2nd ed. Marti- Avena sativa enriched in the 124-kilodalton species. J.
nus Nijhoff Publishers, Dordrecht, The Netherlands. Biol. Chem. 258:11025-11031.
27.Kolukisaoglu, H.U., S. Marx, C. Wiegmann, S. 42.McDowell, M.T. and J.C. Lagarias. Purification and
Hanelt, and H.A.W. Schneider-Poetsch. 1995. Diver- properties of phytochromobilin synthase from etiolat-
gence of the phytochrome gene family predates ed oat seedlings. Plant Physiol. (In press).
angiosperm evolution and suggests that Selaginella and 43.Murphy, J.T. and J.C. Lagarias. 1997. The phyto-
Equisetum arose prior to Psilotum. J. Mol. Evol. fluors: a new class of fluorescent protein probes. Curr.
41:329-337. Biol. 7:870-876.
28.Kunkel, T., V. Speth, C. Buche, and E. Schafer. 44.Murphy, J.T. and J.C. Lagarias. 1997. Purification and
1995. In vivo characterization of phytochrome-phyco- characterization of recombinant affinity peptide-tagged
cyanobilin adducts in yeast. J. Biol. Chem. 270:20193- oat phytochrome A. Photochem. Photobiol. 65:750-
20200. 758.
308
Analysis and Reconstitution of Phytochromes
45.Pratt, L.H., M.M. Cordonnier-Pratt, P.M. Kelmen- 58.Song, P.S., M.H. Park, and M. Furuya. 1997. Chro-
son, G.I. Lazarova, T. Kubota, and R.M. Alba. 1997. mophore: apoprotein interactions in phytochrome A.
The phytochrome gene family in tomato (Solanum Plant Cell Environ. 20:707-712.
lycopersicum L). Plant Cell Environ. 20:672-677. 59.Taylor, B.L. and I.B. Zhulin. 1999. PAS domains:
46.Pratt, L.H., J.E. Wampler, and E.S. Rich, Jr. 1984. internal sensors of oxygen, redox potential, and light.
An automated dual-wavelength spectrophotometer Microbiol. Mol. Biol. Rev. 63:479-506.
optimized for phytochrome assay. Anal. Instrum. 60.Terry, M.J. and J.C. Lagarias. 1991. Holophyto-
13:269-287. chrome assembly—coupled assay for phytochromobi-
47.Quail, P.H., M.T. Boylan, B.M. Parks, T.W. Short, Y. lin synthase in organello. J. Biol. Chem. 266:22215-
Xu, and D. Wagner. 1995. Phytochromes: photosen- 22221.
sory perception and signal transduction. Science 61.Terry, M.J., M.D. Maines, and J.C. Lagarias.
268:675-680. 1993. Inactivation of phytochrome-chromophore
48.Raps, S. 1990. Differentiation between phycobilipro- and phycobiliprotein-chromophore precursors by rat
tein and colorless linker polypeptides by fluorescence liver biliverdin reductase. J. Biol. Chem. 268:26099-
in the presence of ZnSO4. Plant Physiol. 92:358-362. 26106.
49.Remberg, A., P. Schmidt, S.E. Braslavsky, W. Gartner, 62.Tomizawa, K., J. Stockhaus, N.H. Chua, and M.
and K. Schaffner. 1999. Differential effects of muta- Furuya. 1995. Spectrophotometric and molecular
tions in the chromophore pocket of recombinant phy- properties of mutated rice phytochrome A. Plant Cell
tochrome on chromoprotein assembly and Pr-to-Pfr Physiol. 36:511-516.
photoconversion. Eur. J. Biochem. 266:201-208. 63.Vierstra, R.D. 1993. Illuminating phytochrome func-
50.Rudiger, W., T. Brandlmeier, I. Blos, A. Gossauer, and tions. Plant Physiol. 103:679-684.
J.P. Weller. 1980. Isolation of the phytochrome chro- 64.Wada, M., T. Kanegae, K. Nozue, and S. Fukuda.
mophore. The cleavage reaction with hydrogen bro- 1997. Cryptogam phytochromes. Plant Cell Environ.
mide. Z. Naturforsch. 35:763-769. 20:685-690.
51.Rudiger, W. and F. Lopez-Figueroa. 1992. Photore- 65.Wahleithner, J.A., L. Li, and J.C. Lagarias. 1991.
ceptors in algae. Photochem. Photobiol. 55:949-954. Expression and assembly of spectrally active recombi-
52.Rudiger, W., F. Thummler, E. Cmiel, and S. Schnei- nant holophytochrome. Proc. Natl. Acad. Sci. USA
der. 1983. Chromophore structure of the physiologi- 88:10387-10391.
cally active form (Pfr) of phytochrome. Proc. Natl. 66.Wilde, A., Y. Churin, H. Schubert, and T. Borner.
Acad. Sci. USA 80:6244-6248. 1997. Disruption of a Synechocystis sp. PCC 6803 gene
53.Sage, L.C. 1992. Pigment of the Imagination: A History with partial similarity to phytochrome genes alters
of Phytochrome Research. Academic Press, San Diego. growth under changing light qualities. FEBS Lett.
54.Schiff, J.A. 1972. A green safelight for the study of 406:89-92.
chloroplast development and other photomorpho- 67.Wu, S.H. and J.C. Lagarias. 1996. The methy-
genetic phenomena. Methods Enzymol. 24B:321-322. lotrophic yeast synthesizes a functionally active
55.Schmidt, P., U.H. Westphal, K. Worm, S. Braslavsky, chromophore precursor of the plant photoreceptor
W. Gartner, and K. Schaffner. 1996. Chromophore- phytochrome. Proc. Natl. Acad. Sci. USA 93:8989-
protein interaction controls the complexity of the phy- 8994.
tochrome photocycle. J. Photochem. Photobiol. B. 68.Wu, S.H., M.T. McDowell, and J.C. Lagarias. 1997.
Biol. 34:73-77. Phycocyanobilin is the natural chromophore precursor
56.Schneider-Poetsch, H.A.W., B. Braun, S. Marx, and of phytochrome from the green alga Mesotaenium cal-
A. Schaumburg. 1991. Phytochromes and bacterial dariorum. J. Biol. Chem. 272:25700-25705.
sensor proteins are related by structural and functional 69.Yeh, K.C. and J.C. Lagarias. 1998. Eukaryotic phy-
homologies—hypothesis on phytochrome-mediated tochromes: light-regulated serine/threonine protein
signal-transduction. FEBS Lett. 281:245-249. kinases with histidine kinase ancestry. Proc. Natl. Acad.
57.Schneider-Poetsch, H.A.W., S. Marx, H.U. Koluk- Sci. USA 95:13976-13981.
isaoglu, S. Hanelt, and B. Braun. 1994. Phytochrome 70.Yeh, K.C., S.H. Wu, J.T. Murphy, and J.C. Lagarias.
evolution—phytochrome genes in ferns and mosses. 1997. A cyanobacterial phytochrome two-component
Physiol. Plant. 91:241-250. light sensory system. Science 277:1505-1508.
309
Analysis and Reconstitution of
14 Phycobiliproteins: Methods for
the Characterization of Bilin
Attachment Reactions
311
W.M. Schluchter and D.A. Bryant
have been shown to contain proteins simi- biliproteins (26,35,63,65). Recently, the X-
lar to eukaryotic phytochrome (44,47, ray structure of the AP trimers carrying the
64,77,78), definitive evidence for the core linker polypeptide was solved (65).
occurrence of phytochromobilin (PφB) in This structure shows that this AP linker
cyanobacteria has not yet been obtained. (and probably the related rod linker that
The phycobilisome is composed of an interacts with phycocyanin) modifies the
AP core that is surrounded by rods con- spectroscopic properties of the phyco-
taining PC that radiate outwards from this biliprotein with which it is associated by
core. In some organisms, PE is also a com- causing slight shifts in bilin conformation
ponent of these peripheral rods and is as well as by bringing two bilins closer
found distal to the PC (28). In other together within the trimer. The linker pro-
cyanobacteria, phycoerythrocyanin (PEC) tein is located between two of the three
is present in small to moderate amounts β-AP subunits in the trimer and directly
under low light-intensity conditions and is interacts with the PCBs of these two sub-
likewise found distal to phycocyanin in the units (65). Approximately half of the sur-
peripheral rods (9,12). face of the linker protein is located within
The α and β subunits that compose the cavity of the trimer (65).
each phycobiliprotein share amino acid In some strains of marine cyanobacteria,
sequence similarity to each other and to three different bilins may occur on their
the subunits of other phycobiliproteins, phycobiliproteins (55), whereas in other
and this observation supports the hypothe- strains, such as Synechococcus sp. PCC
sis that this family of proteins evolved 7002 and Synechocystis sp. PCC 6803, only
through gene duplication (63). Indeed, it is PCB is present. Even in those strains which
the β subunit of phycoerythrin that is only contain PCB, two different stereoiso-
thought to be the ancestral phycobilipro- mers occur on the C-phycocyanin β sub-
tein from which all others evolved (36, 71). unit at the C3′ of the bilin: the R configu-
The three-dimensional structures of at least ration is found for the PCB attached at
one member of each of the major spectro- cysteine β-82 and the S configuration is
scopic classes of phycobiliproteins have found for the PCB attached at cysteine
been determined (8,13,18,19,23,24, β-153 (62). The biochemical basis for how
59–62,65), and these structures show that the biosynthesis of the phycobiliproteins is
the amino acid similarity translates into controlled, such that the correct bilin is
remarkable structural conservation (5). attached to the proper cysteine residue
The subunit structure for this family of with the appropriate stereochemistry, is a
proteins resembles that of members of the fascinating but incompletely understood
globin family with a predominance of α process. Since autocatalytic reactions with
helices and the complete absence of apophycobiliproteins and free bilins have
β-pleated sheets (61). The unique spectro- yielded nonnatural products (2,4,20), all
scopic properties of each phycobiliprotein evidence currently indicates that bilins are
are believed to be due to the type(s) of enzymatically attached to the appropriate
bilin(s) attached, to the immediate electri- apophycobiliprotein.
cal charge and polarity of the environment Phycobiliproteins have been studied for
of the chromophore, and to the way by more than a century and a half now, and
which the phycobiliprotein subunits hold they have captured the imaginations of
the bilins in a stretched conformation many scientists because of their brilliant
(26,35,63). Linker proteins also affect the colors. These proteins are relatively easy to
spectroscopic properties of the phyco- isolate and purify because they comprise
312
Analysis and Reconstitution of Phycobiliproteins
such a large proportion of the total protein tion, and quantitation of phycobiliproteins
in many cells. Much is known about their utilizing reverse-phase high-performance
structure and function, but much less is liquid chromatography (HPLC) was devel-
known about the biosynthesis of the indi- oped (66). This method has been exten-
vidual proteins and the assembly of the sively used in the characterization of phy-
macromolecular phycobilisome. Most cobiliproteins from newly discovered
approaches toward understanding how organisms (29) and from mutants defective
these proteins are synthesized have been in phycobiliprotein biosynthesis (42,68).
made in the attempts to reconstitute them. The methods necessary for the reconstitu-
Most of these reconstitution studies have tion of phycobiliproteins are summarized
taken place in the last 10 years when below.
recombinant DNA technology has allowed
one to overproduce the apoproteins for Nomenclature
such studies and to generate mutants. The
majority of work done on phycobiliprotein The nomenclature for phycobiliproteins
reconstitution has been performed using can be somewhat confusing and reflects, in
cyanobacterial proteins. Therefore, this part, historical developments in the study
chapter will summarize some of the many of phycobiliproteins. PCs and PEs were all
methods that have been developed for ana- originally given the prefix C- or R- to des-
lyzing and reconstituting phycobiliproteins ignate whether they were purified from
from these organisms. However, it is hoped cyanophytes (cyanobacteria) or rho-
that this information will serve as a good dophytes (red algae). The designation B-
starting point for researchers who are inter- was later introduced for a distinct type of
ested in studying the reconstitution and PE from the red alga Smithora naiadum,
biosynthesis of phycobiliproteins from red which is a member of the order Bangiales
algae, cyanobacteria, or cryptomonads. (1,37). The three major types of Class I PE
Also, since the last review on phycobilipro- (those that contain five bilins per αβ
tein purification was written (27), a new monomer; see below) differ in their
method for the separation, characteriza- absorbance properties due to the types of
Figure 1. Structures of the four singly-linked peptide-linked bilins present in the phycobiliproteins of cyanobacteria. The num-
bering scheme for the carbon atoms is indicated.
313
W.M. Schluchter and D.A. Bryant
bilins present on their αβ subunits. These entum, contains PEB at cysteine β-155 and
proteins exhibit one, two, or three distinct PCB at the other two positions (33). R-PC
peaks in the visible region of the spectrum II, isolated from several unicellular marine
and are called C-phycoerythrin (C-PE; cyanobacterial strains, contains PEB at cys-
containing only PEB), B-phycoerythrin teines α-84 and β-155 and PCB at cysteine
(B-PE), and R-phycoerythrin (R-PE), β-82 (55,56). R-PC-III was isolated from
respectively, regardless of the group from Synechococcus sp. WH7805 and has a
which they have been isolated (References PCB:PEB ratio of 2:1, but differences in
33, 37, and references therein). These des- the absorption properties of this PC sug-
ignations seemed sufficient until marine gest that the chromophores are distributed
unicellular cyanobacteria were shown to differently than in R-PC-I (57). The fourth
contain two forms of PE, PE I, and PE II, form of PC, R-PC-IV, was isolated from
in the rods of their phycobilisomes (55, Synechococcus sp. WH8501 and was found
67). PE I was less abundant than PE II. PE to contain PUB attached at α-84 and PCB
II has an extra bilin (PUB) on the α sub- at the other two positions on the β subunit
unit (α-75) and contains both PEB and (67). Finally, PEC is structurally more sim-
PUB. Thus far, only two Synechococcus ilar to PCs than to PEs, but is found distal
strains (WH8020 and WH8103) have to PC in the phycobilisome rods of some
been shown to contain a PE with six chro- cyanobacterial strains. PEC carries PXB at
mophores per αβ monomer (55). There- α-84 and PCB at both positions on the β
fore, these two PEs are members of a new subunit (9,12). Spectroscopic variants of
class of PE, dubbed Class II PE. However, AP (which contains one PCB on each sub-
PE I is more like other PEs that have been unit) have not yet been identified. Thus,
characterized, in that it contains only five the nomenclature for biliproteins devised
bilins per αβ monomer and contains either previously has been rather haphazard and
only PEB or both PEB and PUB (55). confusing. A new form of nomenclature
Thus far, no red algal PE has been shown has been suggested (51), but has not been
to be a member of Class II PE. B-PE, R- widely used thus far.
PE, and PE II complexes carry bilins on
their associated linker protein, called γ (34,
45). A recently discovered red algal PE 2. HOLO-PHYCOBILIPROTEINS
(from Audouinella macrospora) contains a
PE with PCB, PEB, and PUB chro- 2.1. Phycobiliprotein Purification
mophores and is more like B- and R-PEs in
that it contains 5 bilins per αβ monomer Phycobiliproteins may exist in different
and has bilins present on its γ subunit (29). aggregation states depending upon the
This PE is a Class I member, but is not by individual type of biliprotein, the organism
definition an R-PE, which have been from which it was isolated, the composi-
shown to contain PUB and PEB that con- tion of the solution containing it (pH,
tribute to the three absorption peaks in the ionic strength), and such factors as temper-
visible region. ature and protein concentration. The
Four major types of PC have been char- purification of individual phycobiliproteins
acterized (15,63), and all PC types contain has been summarized previously (27). A
PCB as the terminal acceptor bilin at cys- few minor improvements have been intro-
teine β-82. C-PC contains PCB at all three duced using fast protein, peptide, and
cysteines (25,46,60). R-PC-I, present in polynucleotide liquid chromatography
some red algae including Porphyridium cru- (FPLC) (Mono Q; Amersham Pharmacia
314
Analysis and Reconstitution of Phycobiliproteins
Biotech, Piscataway, NJ, USA) (66), but tributions of each bilin type at various
for the most part, the conventional chro- wavelengths must be considered. The con-
matographic methods are still widely used tributions of the various chromophores at
today. Therefore, this chapter will primari- different wavelengths are listed in Table 1.
ly summarize the methods for separation In some cases, the concentration of the
and purification of individual α and β sub- phycobiliprotein sample may be limiting;
units. for example, this is often the case when
isolating phycoerythrins from field sam-
2.2. Storage and Recovery of Purified ples of red algae. The spectra for PEs in
Phycobiliproteins 20% acetic acid (vol/vol) have been deter-
mined to be identical to those for the
Phycobiliproteins are very stable when same protein dissolved in 8 M urea, pH
stored in phosphate buffer at pH 7.0 in the 3.0 (37). This was also found to be true
presence of a reducing agent [1–5 mM for AP and PC (A.N. Glazer, personal
β-mercaptoethanol or dithiothreitol communication).
(DTT)] and sodium azide (1 mM) in the
dark at 4°C. For long-term storage, ammo- 2.4. Purification of Individual Subunits
nium sulfate may be added to 65% satura- by Conventional Chromatographic
tion at 4°C. When sealed at 4°C in the Methods
dark, such slurries–precipitates can be
stored indefinitely. The phycobiliproteins The first method for the separation and
can be recovered by centrifugation at purification of phycobiliprotein subunits
27 000× g for 15 minutes or by centrifuga- was developed by Glazer and Fang and was
tion in a microcentrifuge at 13 000× g for based upon methods used for the separa-
15 minutes. The phycobiliprotein pellet tion of the subunits of hemoglobin
should be resuspended in 5 mM phosphate (30,31). All methods described thus far are
buffer, pH 7.0, 1 mM β-mercaptoethanol, performed under denaturing and acidic
and dialyzed against the same buffer at 4°C conditions which limit oxidation and other
prior to use. side reactions that can modify the bilins.
Each procedure described will include the
2.3. Concentration Determination cyanobacterial source for the phycobilipro-
tein. Most of these conditions have been
Because the absorption properties of the shown to work successfully for the separa-
phycobiliproteins are highly dependent on tion of phycobiliproteins from a wide vari-
the aggregation state, pH, ionic strength, ety of sources, but some optimization of
and protein concentration (26), the most the method may be required if the user is
reliable method to determine the concen- attempting to adapt the method to the
tration of phycobiliprotein solutions is to purification of subunits from a different
measure the absorption spectrum of the phycobiliprotein (see Procedure 1). Each
peptide-bound bilins by dissolving an subunit, when renatured without its part-
aliquot of the protein in 8 M urea, pH 1.9, ner, is much less stable and tends to aggre-
or in 10 mM TFA (trifluoroacetic acid); its gate over time when in solution. Proce-
concentration can then be determined by dures 2 and 3 describe renaturation of
using the extinction coefficients given in phycocyanin subunits. Both renaturation
Table 1 (Reference 27 and references there- procedures, followed by the last diethyl-
in). For phycobiliproteins that contain 2 or aminoethyl (DEAE) chromatography step,
more different bilins per subunit, the con- have a recovery rate for renatured protein
315
W.M. Schluchter and D.A. Bryant
316
Analysis and Reconstitution of Phycobiliproteins
chococcus sp. PCC 6301 and Synechocystis The Bio-Rex 70 column (1.5 × 15 cm)
sp.; (22663; ATCC, Manassas, VA, USA) with the PE subunits adsorbed was washed
also called Microcystis aeruginosa (14) (Pro- with 15 mL of 2.0 M urea, 30 mL of 4.0
cedure 5). M urea, and 50 mL of 6.0 M urea before
development with a linear gradient of 6.0
❖ Procedure 5. Separation of to 10.0 M urea, pH 3.0 (20 mL total vol-
Allophycocyanin Subunits from ume). The α subunit eluted first followed
Synechococcus sp. PCC 6301 and by the β subunit. Subunits were renatured
Synechocystis sp. by exhaustive dialysis against 50 mM K-
phosphate buffer at pH 7.0 at room tem-
1. Dissolve 325 mg purified and lyo- perature.
philized AP in 50 mL of 10 mM K- Separation of the subunits of Anabaena
phosphate, 8 M urea, 10 mM β-mer- variabilis PEC was first demonstrated by
captoethanol, pH 8.0, and equilibrate Bryant et al. using a modification of the
for 1 hour at room temperature. See method developed for the separation of PC
section 2.1. subunits described above (12). The Bio-
2. Load the material onto a DEAE Rex 70 column (3.9 × 51 cm) was subject-
Sephadex A-50 column (3.5 × 12 cm) ed to incremental step gradients of acidic
and wash with equilibration buffer. urea as described previously, followed by
3. Use 200 mL of equilibration buffer elution of the α subunit by addition of
plus 50 mM KCl to elute the elute the 7.4 M urea, pH 3.0. Once the elution of the
β subunit of AP. α subunit was complete, elution of the
β subunit was accomplished by addition
4. Residual β subunit is eluted by repeat-
of 9.0 M urea, pH 3.0. Subunits were dia-
ed washes with 150 mL equilibration
lyzed against 50 mM ammonium acetate,
buffer plus 80 mM KCl.
pH 6.8.
5. Use equilibration buffer plus 180 mM
KCl to elute the α subunit of AP.
2.5. Purification of Phycobiliproteins by
6. Pool fractions of each subunit from HPLC
steps 3 and 5 for dialysis against 25
mM ammonium acetate, pH 6.8, and In 1987, HPLC was used to verify the
concentration by ultrafiltration using an purity of PC and AP preparations from M.
Amicon cell with a 10 000 MWCO aeruginosa (58); however, the method also
membrane (Millipore, Bedford, MA, showed that the AP and PC subunits could
USA). be separated on a C18 reverse-phase col-
A method similar to the one developed umn. In 1990, Swanson and Glazer intro-
for the separation of the PC subunits was duced a method for separation of phyco-
successfully used in the separation of the α, biliprotein subunits using C4 reverse-phase
β, and γ subunits of phycoerythrin from P. HPLC (66). These HPLC methods have
cruentum (34). The only significant differ- several advantages over the conventional
ence was in the development of the col- chromatographic methods. They are more
umn. The γ subunit was eluted with 7.4 M rapid and require much less starting mater-
urea, the α subunit with 8.0 M urea, and ial. When used in conjunction with a pho-
the β subunit with 9.0 M urea. Similar todiode array detector, these methods also
conditions were used to separate the sub- give immediate spectroscopic information
units of phycoerythrin II (PE II) from the about bilin content and subunit stoichiom-
cyanobacterium Gloeobacter violaceus (10). etry. The method of Swanson and Glazer
318
Analysis and Reconstitution of Phycobiliproteins
has also successfully been used to separate The method of Swanson and Glazer
phycobiliproteins obtained directly from uses a C4 reverse-phase analytical column
purified phycobilisomes, giving quantita- (250 × 10 mm) and a solvent system con-
tive information regarding phycobilipro- sisting of 0.1% TFA in water (Buffer A)
tein stoichiometry and content in these and a 2:1 acetonitrile: isopropanol mixture
mixtures (39). When both HPLC methods (Buffer B). This purification procedure has
were compared, the method of Swanson been very successful in the separation and
and Glazer gave better resolution of phyco- resolution of diverse types and mixtures of
biliproteins isolated from Arthrospira max- phycobiliproteins. The purified phyco-
ima (38,39). The use of reverse-phase biliprotein or phycobiliprotein mixture,
HPLC is clearly a better choice than con- typically 100 to 1500 µg in 200 to 500 µL
ventional chromatographic procedures for in 5 mM Na-phosphate, pH 7.0, 1 mM β-
determining stoichiometric information mercaptoethanol is combined with an
when the amount of starting material and equal volume of 9.0 M urea, pH 2.0 (fresh-
speed are primary concerns. ly prepared), and subjected to centrifuga-
Figure 3. HPLC separation of cyanobacterial C-PC subunits. Purified PC from Synechococcus sp. PCC 6301 (top panel) or
Anabaena sp. PCC 7120 (bottom panel) was separated on a C4 reverse-phase HPLC column as described in the text. Elution of
subunits was monitored at 660 nm in order to follow the absorbance of peptide-linked PCB. In each case, the α subunit elutes
prior to the β subunit. This figure was modified with permission from Reference 66.
319
W.M. Schluchter and D.A. Bryant
tion in a microcentrifuge for 5 minutes very important to wash the column exten-
prior to injection on the column. A Hi- sively between injections using a linear gra-
Pore RP304 column (Bio-Rad Laborato- dient to 100% Buffer B over 5 minutes fol-
ries) equilibrated in 65% Buffer A and lowed by at least 5 to 10 minutes of
35% Buffer B (1.5 mL/min) has typically washing the column with 100% Buffer B.
been used. After injection of the sample, The β subunits of phycobiliproteins are
proteins are eluted with a linear gradient to sometimes retained on the column, and
30% Buffer A and 70% Buffer B over 35 these will usually be eluted by this treat-
to 40 minutes depending upon the source ment. If careful quantitation of a sample is
of the phycobiliprotein (see Figure 3). required, it is wise to perform a blank
With a few alterations of the elution gradi- injection between each run with samples in
ent profile, this method has been success- order to insure that the column is entirely
fully employed in the separation of a wide free of residual phycobiliprotein subunits.
variety of phycobiliproteins from cyano- Preparative separation of phycobilipro-
bacteria, red algae, and cryptomonads tein subunits can be accomplished using
(17,29,39,66,74). In fact, researchers have this method in conjunction with a semi-
had success in the separation of phyco- preparative C4 reverse-phase column (or by
biliproteins from phycobilisome samples employing multiple runs on an analytical
taken directly from sucrose gradients (after column). Subunits can be collected as they
dialysis against 5 mM Na-phosphate, pH elute from the column, and the solvents
7.0 followed by combination with an equal can be removed by rotary evaporation. The
volume of 9.0 M urea, pH 1.9, prior to aqueous subunits can then be diluted 2:9
injection) (see Figure 4). Toole et al. com- with 9.0 M urea, pH 2.0, 10 mM β-mer-
bined the phycobilisomes taken directly captoethanol, followed by dialysis against
from sucrose gradients (in sucrose–phos- 50 mM Na-phosphate, pH 7.0 (22).
phate) with an equal volume of 8.4 M
guanidine hydrochloride, pH 6.4 (followed 2.6. Methods for Analyzing the Quality
by centrifugation), prior to loading on the of the Renatured Subunits
C4 column (Vydac/The Separations
Group, Hesperia, CA, USA) using the gra- The best method to analyze the quality
dient conditions described above (72). This of renatured subunits is to compare the
method has also been successfully used to absorption spectrum of a dilute solution
characterize the linker polypeptide and containing the subunit with the fluores-
phycobiliprotein stoichiometry in phyco- cence excitation spectrum of the same solu-
bilisomes from A. maxima (38,39). tion. In order to obtain an accurate excita-
Some technical considerations to keep tion spectrum, the absorbance at the
in mind for each separation include the long-wavelength maximum should be less
need to use higher concentrations of urea than 0.05 OD so that reabsorption of
to solubilize phycobiliprotein mixtures that emitted light will be minimized. If the two
may contain any given apophycobilipro- spectra differ significantly, then it is likely
tein. It has been observed that apophyco- that the renatured subunit is not folded
biliprotein subunits often do not bind as properly or that the chromophore(s) may
well as holo-subunits under these condi- have been chemically modified during
tions, but that addition of urea to at least 6 purification and renaturation. If the major-
M final concentration in the solution to be ity of the protein has been oxidized, it is
injected greatly increases the yield of unlikely that the sample will be a good
apophycobiliprotein material (22). It is also source of bilin in bilin transfer assays.
320
Analysis and Reconstitution of Phycobiliproteins
material was removed by ultracentrifuga- it is much easier to store frozen E. coli cells
tion at 100 000× g for 30 minutes. The containing the overproduced apophyco-
supernatant containing both subunits was biliprotein fusion and purify small batches
applied to a Bio-Gel P100 gel filtration of protein when one needs it. This insures
column (5 × 75 cm; Bio-Rad Laboratories) that the substrate for in vitro addition reac-
with 3 M urea-HCl, 10 mM β-mercap- tions is properly folded and contains fully
toethanol, pH 2.5, as the buffer at room reduced cysteines.
temperature. The β subunit eluted first,
followed by fractions containing both α 3.1.4. Attaching Apophycobiliproteins to
and β subunits, and finally followed by Agarose Beads
fractions containing only the α subunit.
Attempts to renature the β subunit were The covalent attachment of apo-α-PC
unsuccessful. However, the α subunit to agarose beads greatly facilitated reconsti-
could be renatured as long as the protein tution studies because it was possible to
concentration remained below 0.1 mg/mL. perform addition reactions in a small
Dialysis against 50 mM Na-phosphate, 1.0 microcentrifuge tube, to wash away excess
mM DTT, pH 7.0, and 0.1 mM NaN3 bilin after the reaction was terminated if
resulted in renaturation of some apo-α-PE. necessary, and then to measure the fluores-
cence of the sample after this process
3.1.3. Producing Apophycobiliproteins as (21,22). The apoprotein in 50 mM Na-
Fusions phosphate, pH 7.0, 5 mM EDTA was
mixed with Affi-Gel 15 (Bio-Rad Labo-
Several different phycobiliprotein struc- ratories) at 1 mg of protein per mL of
tural genes have been successfully fused beads (22). The covalent attachment of the
with the genes encoding other proteins, protein to the beads continued for 30 min-
and this has allowed the purification pro- utes at 4°C until the reaction was stopped
cedure to be simplified to a single affinity by the addition of 0.05 volumes of 1 M
chromatography step (Y.A. Cai, W.M. glycylglycine, pH 7.0 (incubated for 1
Schluchter, and A.N. Glazer, unpublished hour at 4°C). To remove excess unbound
results). The maltose binding protein has protein, the beads were washed with 50
been employed in such fusions, as well as a mM Na-phosphate, pH 7.0, 5 mM
domain of 24 amino acids containing 6 EDTA, 0.5 M NaCl, followed by 50 mM
contiguous histidine residues that has usu- Na-phosphate, pH 7.0, 5 mM EDTA. The
ally been fused to the N termini of several air was evacuated out of the flask contain-
phycobiliprotein subunits (including α- ing the beads, and the beads were stored at
PC, β-PC, α-AP, and β-AP) from several 4°C in the same buffer with the addition
different cyanobacteria. Following the of 5 mM DTT.
manufacturer’s procedures for purification
of the fusion proteins, high yields of prod- 4. RECONSTITUTION OF
ucts were generally obtained. An important HOLOPHYCOBILIPROTEINS
factor to remember is to add reductant
throughout the purification procedure in 4.1. Nonenzymatic Assays
order to keep the cysteines reduced. It is
also best to purify only as much protein as The first evidence that enzymes might
is needed in the next week. Within 2 weeks be required for bilin addition to phyco-
at 4°C, these proteins tend to oxidize and biliproteins was revealed through the
begin denaturing. As a matter of practice, experiments of Arciero et al. with apo-PC
322
Analysis and Reconstitution of Phycobiliproteins
(2–4). When either PCB or PEB was must take place in vivo in organisms which
added to apo-αβ-PC, covalent addition contain more than one bilin attached to
took place at the α-84 and β-82 sites, but phycobiliproteins.
not at the β-153 site. The primary prod-
ucts of those nonenzymatic additions were 4.1.1. Assay Conditions
bilins at a higher oxidation state, with an
extra double bond between C2-C3 of ring The single most important factor in
A (see Figure 1 for numbering scheme). these assays is that the apoprotein be fully
Mesobiliverdin (MBV) was the product reduced prior to addition of the bilin sub-
when PCB was added, and 15,16 dihydro- strate. This is accomplished by using fresh-
biliverdin was the product when PEB was ly purified apoprotein, adding DTT to 10
added. mM, and incubating this mixture for 30
Nonenzymatic addition reactions have minutes at room temperature (22). The
also been performed with apo-α-PC (20) DTT should be removed by gel filtration
and with apoallophycocyanin subunits prior to addition of bilin. It has been
(W.M. Schluchter and A.N. Glazer, observed that bilins will react with DTT
unpublished results). In all cases, a phyco- when this compound is present at high
biliprotein adduct is formed, and there is concentrations (W.M. Schluchter and
no discrimination between the bilin iso- A.N. Glazer, unpublished results) (20).
mers observed in such in vitro addition Generally, apophycobiliproteins are very
experiments. Such discrimination clearly stable when present in 5 to 50 mM Na-
uct of which shows limited sequence simi- 8.0. Washing entails full resuspension,
larity to the family of putative lyases preferably using a tissue homogenizer,
including CpcE and which is located followed by centrifugation at 8000× g;
downstream of the structural genes encod- the inclusion bodies containing
ing PE, produced markedly lower levels of CpcE/CpcF will be in the pellet frac-
PE. These observations suggest that CpeY tion.
is a lyase subunit as well (43). 4. The inclusion body proteins are solubi-
The activities of only a few lyases have lized with 9.0 M urea-HCl, pH 1.9, 1
been tested in vitro, and to date, the most mM DTT. The concentrations of each
extensive analyses have been performed protein should be determined spec-
using Synechococcus sp. PCC 7002 CpcE trophotometrically using the ε280 nm
and CpcF. So far, no lyase that can specifi- for each protein (calculated from the
cally attach bilins to the β subunit of any Trp [ε = 5540 M-1cm-1] and Tyr [ε =
phycobiliprotein has been identified. 1480 M-1cm-1] content of the pro-
Methods for assaying these enzymes will be teins) (54).
summarized here in the hopes that this will 5. This estimate should be compared with
encourage additional research in this area. the staining intensities of diluted
aliquots of each urea-solubilized pro-
4.2.1. CpcE CpcF Expression tein on SDS-PAGE. The ε280 nm for
Synechococcus sp. PCC 7002 CpcE and
Recombinant CpcE and CpcF were pro-
CpcF under denaturing conditions are
duced in both soluble form and in the
35 640 M-1cm-1 and 20 220 M-1cm-1,
form of inclusion bodies in E. coli. Howev-
respectively (22).
er, Fairchild et al. showed that corenatura-
tion of these two proteins in a 1:1 ratio led 6. These proteins should be mixed in a
to the most activity (22). 1:1 molar ratio at a concentration of
0.15 to 0.3 mg/mL prior to renatura-
❖ Procedure 7. Purification of tion.
Recombinant CpcE and CpcF Several methods have been used success-
fully to renature these proteins. A concen-
1. The cpcE and cpcF genes overexpressed trated mixture can be diluted approximate-
using a T7/pET vector system and the ly 1:10 with 50 mM Tris-HCl, 75 mM
majority of the recombinant proteins NaCl, pH 8.0; the dilution is followed by
are found in inclusion bodies. extensive dialysis against the same buffer at
2. The inclusion bodies are collected by 4°C. This procedure yielded renatured het-
low-speed centrifugation after the cells erodimeric CpcECpcF, but direct dialysis
have been lysed by passage through a of the diluted proteins in 9.0 M urea
French pressure cell. The inclusion against the same Tris-NaCl buffer pro-
bodies appear as a chalky white pellet duced similar results. In both cases, the
and are easily differentiated from yield of renatured CpcECpcF was approxi-
unbroken cells which usually appear mately 50%. The extinction coefficients
more tan or brownish in color. for native CpcE and CpcF were calculated
3. The inclusion bodies should be washed (from the Trp and Tyr content of each pro-
extensively using the following solu- tein) to be 38 440 M-1cm-1 and 21 060
tions: 50 mM Tris-HCl, 5 mM EDTA, M-1cm-1, respectively. After filter steriliza-
pH 8.0; 50 mM Tris-HCl, pH 8.0, 1% tion through a 0.2 µm membrane to pre-
Triton X-100; 50 mM Tris-HCl, pH vent microbial growth, these proteins were
325
W.M. Schluchter and D.A. Bryant
stable for weeks at 4°C. Although other elsewhere in this volume (see Chapter 8)
purification procedures have been utilized and in References 2 and 22. There is
for preparations of proteins for more rigor- presently no reported method for the
ous kinetic analyses (21), the procedure purification of the precursor of peptide-
described above yielded a preparation of linked PUB or of the doubly-linked forms
enzyme with high activity. of PEB and PUB. However, it has been
shown that CpcECpcF from Synechococcus
4.2.2. Bilin Donors sp. PCC 7002 (13) and Anabaena sp. PCC
7120 (C.F. Chan, W.M. Schluchter, and
PEB and PCB can be cleaved from holo- A.N. Glazer, unpublished results) will
phycobiliproteins and purified as described transfer the bilin from a holo-α-PC sub-
Figure 5. Bilin addition assays with Anabaena sp. PCC 7120 apo-α-PC resin. Assay conditions were as follows. Approximately
300 µL of settled resin (containing Anabaena sp. PCC 7120 apo-α-PC subunit covalently attached as described in Reference
22 was in a 1.5-mL microcentrifuge tube containing 0.8 mL of reaction assay buffer (50 mM Tris-HCl, pH 8.0, 75 mM NaCl,
1 mM MgCl2, 1 mM Na pyrophosphate, 1 mM thioglycollate). The enzyme to be tested was Anabaena sp. PCC 7120
CpcECpcF (overproduced and purified as described in this chapter; W.M. Schluchter, C. Chan, and A.N. Glazer, manuscript in
preparation). In assays where the enzyme was added (+CpcEF), Anabaena sp. PCC 7120 CpcECpcF were present at 0.25 µM. In
control assays, the same volume of reaction assay buffer was added in place of CpcECpcF (-CpcEF). The reaction was initiated by
the addition of the bilin donor. After incubation at 37°C in the dark for 1 hour, the resin was washed extensively as described in
the text to remove any remaining donor bilin. The fluorescence emission of the resin present in each assay was measured at 640
nm because this is the peak of fluorescence emission for the native holo-α-PC. The donor bilin was purified PCB (11.6 µM;
labeled as PCB), purified holophycocyanin from Anabaena sp. PCC 7120 (0.92 µM; labeled as 7120 PC), or purified holophy-
cocyanin from Synechococcus sp. PCC 7002 (1.0 µM; labeled as 7002 PC). The Anabaena sp. PCC 7120 CpcECpcF lyase cat-
alyzed the addition of free PCB to Anabaena sp. PCC 7120 apo-α-PC. However, this enzyme also catalyzed the reverse reaction
by transferring bilin from the α-PC subunit (purified either from Anabaena sp. PCC 7120 or from Synechococcus sp. PCC 7002;
W.M. Schluchter, C. Chan, and A.N. Glazer, unpublished results).
326
Analysis and Reconstitution of Phycobiliproteins
The first assays performed to test the 4.3. Methods for Analysis: Detection of
activity of Synechococcus sp. PCC 7002 Covalent Products
CpcECpcF used apo-α-PC bound to resin
as the substrate (22). This greatly facilitated Enzymatic bilin addition reactions
the removal of unreacted bilins or the holo- should always be compared with control
α-PC substrate and the enzyme prior to nonenzymatic reactions using one or more
the measurement of the incorporation of of the following methods. If holophyco-
PCB onto the α-PC resin. However, affin- biliproteins were the source of bilin for the
ity-tagged apophycobiliproteins have been addition reaction, care must be taken to
successfully used as substrates in these same insure that all residual holophycobilipro-
reactions (W.M. Schluchter and A.N. tein has been removed. This is most easily
Glazer, unpublished results). accomplished when the apophycobilipro-
tein is attached to a solid support. The
❖ Procedure 8. Enzymatic Assay of beads are washed extensively with 9.0 M
Adduct Formation urea-HCl, pH 2.5, followed by 50 mM
Na-phosphate, pH 7.0. When the apo-sub-
1. The fully prereduced apophycobilipro- unit has been affinity tagged, it is very like-
tein is added to a microcentrifuge tube. ly that any holo-phycobiliprotein added as
If the subunit is affinity tagged, a source of bilin can dimerize with either
approximately 0.3 to 0.6 mg is used. affinity-tagged apo-subunits or affinity-
However, if the subunit is attached to a tagged enzyme-mediated bilin adducts and
solid support, an aliquot corresponding will copurify with the affinity-tagged sub-
to approximately 300 to 500 µL of set- unit. Therefore, purification of the affini-
tled beads is added. ty-tagged protein must be performed
2. The buffer conditions (as determined according the manufacturer’s procedure
for optimal activity of the Synechococ- under denaturing conditions whenever
cus sp. PCC 7002 CpcECpcF) are 50 possible. If this is not possible, then anoth-
mM Tris-HCl, pH 8.0, 75 mM NaCl, er method for the detection and separation
1 mM MgCl2, 1 mM disodium pyro- of these two subunits should be used (see
phosphate, and 1 mM thioglycolate. HPLC separation below).
3. The enzyme to be tested should be
added to a final concentration of 0.1 to 4.3.1. Absorbance
0.4 µM. Absorbance is the easiest and most
4. The reaction is usually initiated by the straightforward method to detect an addi-
addition of the bilin substrate. Free tion product. Unfortunately, this is the
bilin should be dissolved in DMSO at method that gives one the least amount of
327
W.M. Schluchter and D.A. Bryant
information about the product. Although absorbance and fluorescence are red-shift-
it is a good starting point, this method ed relative to the PCB product. The MBV
should never be used as the only indicator adduct, with a fluorescence emission maxi-
of which product(s) is present. When PCB mum at 668 nm, is much less fluorescent
is added to apo-α-PC in the absence of than the PCB adduct, which has a fluores-
CpcECpcF, the unnatural MBV adduct cence emission maximum at 643 nm
predominates and can be easily distin- (3,22). Additionally, the extinction coeffi-
guished from the PCB product. The cients for the long wavelength absorbing
absorbance maximum of MBV attached to species of MBV peptides in 10 mM TFA
the native PC subunit occurs at 647 nm, were determined to be 40% lower than
whereas the absorbance maximum for the those of the naturally occurring PCB-bear-
proper PCB adduct occurs at 622 nm (2). ing peptides (2).
However, in cases in which multiple prod- Much less is known about the fluores-
ucts may be attached at the same site, the cence properties of the unnatural DBV
absorbance spectrum of the addition prod- adduct formed when PEB is added to apo-
uct will usually be difficult to interpret PC or apo-α-PE (4,20). The use of fluores-
(20). The absorbance of the peptide-bound cence to monitor product accumulation
bilins present can be determined by denat- with putative lyases that attach PEB may
uration of the addition product using one be complicated by the fact that multiple
of the methods described above. For PEB products accumulate in nonenzymatic
addition experiments, the nonenzymatical- reactions. Therefore, absorbance and fluo-
ly favored product, DBV, was found to rescence spectroscopy may not work as well
accumulate (4,20). DBV exhibits charac- as one of the following methods for the
teristic absorbance maxima at 606 and 330 characterization of enzymatic bilin addi-
nm in native proteins (74); for denatured tion to apo-PE.
subunits in acidic urea solutions, a 330 nm
absorbance peak is diagnostic of peptide- 4.3.3. HPLC Separation and Detection
linked DBV, whereas a 308 nm peak is
characteristic of peptide-linked PEB If the holo- and apo-subunits, which
(33,74). might be produced or used as substrates in
an enzymatic reaction, can be separated by
4.3.2. Fluorescence C4 reverse-phase chromatography as
described above, then this method provides
Free bilins exhibit little fluorescence in an excellent way to detect the transfer of
solution but become highly fluorescent bilin from a holophycobiliprotein to an
once they have been covalently attached to apo-subunit. Such separations are usually
phycobiliproteins, because they are rigidly best achieved if the source of the holo-sub-
held in a stretched conformation that does unit is from another organism. The trans-
not facilitate nonradiative decay of the fer reaction of PCB from Anabaena sp.
excited state (22). Therefore, the fluores- PCC 7120 holo-α-PC to Synechococcus sp.
cence emission spectrum of both control PCC 7002 apo-α-PC mediated by Syne-
and enzymatic reactions can be measured chococcus sp. PCC 7002 CpcECpcF was
as a way of monitoring the products of the detected using this method (22). Syne-
reaction (see Figure 5). The MBV product chococcus sp. PCC 7002 CpcECpcF pro-
of nonenzymatic PCB addition to apophy- teins can also transfer a bilin from Syne-
cocyanin is easily distinguished from the chococcus sp. PCC 7002 holo-PC to
natural PCB product, because both the Anabaena sp. PCC 7120 apo-α-PC sub-
328
Analysis and Reconstitution of Phycobiliproteins
unit (see Figure 6). aliquot of trypsin is added, and the incuba-
tion is continued for another 2 hours
4.3.4. Characterization of the Product by under the same conditions. The reaction is
Tryptic Digestion stopped by the addition of glacial acetic
acid to 30% (vol/vol). If a large amount of
This is the most quantitative method of protein is being digested, then fractiona-
characterization of the bilin product tion on Sephadex G-50 in 30% acetic acid
(2,3,20). The addition product is cleaved (vol/vol) is a good method to separate
using trypsin, and tryptic peptides are sep- undigested material from tryptic peptides.
arated on a C18 reverse-phase column (45). If the amount of material is scaled for ana-
Tryptic peptides can be collected, their lytical purposes, then the colored material
absorption spectra in 10 mM TFA deter- can be collected and loaded directly onto a
mined, and their composition evaluated by SepPak C18 cartridge. The cartridge can be
amino acid analysis or sequencing to show washed with 0.1% TFA followed by elu-
rigorously which bilin was added to a par- tion by 60% acetonitrile, 40% 0.1% TFA.
ticular site(s) on the apophycobiliprotein The eluate should be collected, dried under
subunit. If multiple products are present, N2, and redissolved in 10 mM TFA prior
this is the best method to determine how to HPLC separation. However, if the
many products have been formed and to amount of material is scaled for preparative
quantitate their relative amounts. Keep in purposes, the colored material in the eluate
mind that for each phycobiliprotein, diges- from the gel exclusion chromatography in
tion by more than one protease may be 30% acetic acid should then be concentrat-
required to obtain a fragment sufficiently ed under N2 before dilution with 50 mM
small to allow its isolation and characteri- Na-phosphate, pH 2.5. The mixture
zation. Digestion procedures for each type should then be fractionated on an ion-
of phycobiliprotein have been published exchange column (SP-Sephadex G-25, 2 ×
(7,49,50,55,67,74–76), and it is recom- 6.5 cm) and eluted with a linear gradient
mended that the user refer to the opti- of 0 to 0.6 M NaCl in 50 mM Na-phos-
mized procedure for the particular phyco- phate, pH 2.5. Fractions containing col-
biliprotein with which he/she is working. ored material should be collected, desalted
The procedure described below was used on the SepPak C18 cartridge as described
successfully on C-PC and R-PE (2,45). above, before separation by HPLC.
The addition product should be separat- The conditions used for separating the
ed from unreacted bilin by chromatogra- tryptic peptides of phycocyanin follow.
phy on Sephadex G-25. The phycobilipro- However, for each phycobiliprotein, differ-
tein should then be fully denatured by ent gradient conditions may be required,
acidification with 1 N HCl to pH 2.0 and and optimization of these conditions
stored under N2 for 45 minutes. Trypsin should be pursued prior to preparative-
(TCPK-treated; Worthington Biochemical, scale analyses. For the phycocyanin of
Lakewood, NJ, USA), dissolved in 1 mM Synechococcus sp. PCC 7002, a C18 reverse-
HCl at 5 mg/mL concentration, is added phase analytical column (5 µm, 4.6 × 250
to 2% (wt/wt) to the denatured phyco- mm) should be used for separation of tryp-
biliprotein in HCl. This mixture is titrated tic peptides (see Figure 7). The solvent sys-
to pH 7.5 with 1 N NaOH after the addi- tem is 0.1 M Na-phosphate, pH 2.1
tion of ammonium bicarbonate to 100 (Buffer A) and acetonitrile (Buffer B) with
mM. After incubation of this mixture for 2 flow rates of 1.5/mL min. Peptides are
hours at 30°C in the dark, an additional loaded at 20% Buffer B (80% Buffer A)
329
W.M. Schluchter and D.A. Bryant
Figure 6. Monitoring the transfer of bilin from Synechococcus sp. PCC 7002 PC to Anabaena sp. PCC 7120 apo-α-PC by
C4 reverse-phase HPLC. Each assay contained 100 µg of Anabaena sp. PCC 7120 apo-α-PC, 75 µg of Synechococcus sp. PCC 7002
PC, 0.2 µM Synechococcus sp. PCC 7002 CpcECpcF (if present) in a volume of 400 µL (reaction assay buffer conditions are as
described in Figure 5). Reactions were allowed to proceed for 16 hours at room temperature in the dark. Each reaction was com-
bined with 800 µL of 9 M urea, pH 1.9, mixed, and centrifuged prior to injection on the C4 column (as described in this chap-
ter). After injection, buffer conditions (buffers are those from Swanson and Glazer; Reference 66) are as follows: 2 minutes at 35%
Buffer B (65% Buffer A), a 1-minute linear gradient to 53% Buffer B (47% Buffer A), followed by a linear gradient to 63% Buffer
B over 20 minutes (22). Each assay was monitored at 280 nm (reflecting protein content) and 680 nm (reflecting bilin content).
Retention times for various components are as follows: Anabaena sp. PCC 7120 apo-α-PC, 9.5 minutes; Anabaena sp. PCC
7120 holo-α-PC, 10 minutes; Synechococcus sp. PCC 7002 apo-α-PC, 11.7 minutes; Synechococcus sp. PCC 7002 holo-α-PC,
12.2 minutes; Synechococcus sp. PCC 7002 holoβ-PC, 15.8 minutes. Synechococcus sp. PCC 7002 CpcECpcF is capable of trans-
ferring bilin from 7002 holo-α-PC to Anabaena sp. PCC 7120 apo-α-PC (W.M. Schluchter and A.N. Glazer, unpublished
results).
330
Analysis and Reconstitution of Phycobiliproteins
332
Analysis and Reconstitution of Phycobiliproteins
32.Glazer, A.N., S. Fang, and D.M. Brown. 1973. Spec- 48.Lokstein, H., C. Steglich, and W.R. Hess. 1999.
troscopic properties of C-phycocyanin and of its α and Light-harvesting antenna function of phycoerythrin in
β subunits. J. Biol. Chem. 248:5679-5685. Prochlorococcus marinus. Biochim. Biophys. Acta.
33.Glazer, A.N. and C.S. Hixson. 1975. Characterization 1410:97-98.
of R-phycocyanin: chromophore content of R-phyco- 49.Lundell, D.J. and A.N. Glazer. 1981. Allophyco-
cyanin and C-phycoerythrin. J. Biol. Chem. 250: cyanin B: α common β subunit in Synechococcus allophy-
5487-5495. cocyanin B (λmax 670 nm) and allophycocyanin (λmax
34.Glazer, A.N. and C.S. Hixson. 1977. Subunit struc- 650 nm). J. Biol. Chem. 246:12600-12606.
ture and chromophore composition of rhodophytan 50.Lundell, D.J., A.N. Glazer, R.J. DeLange, and D.M.
phycoerythrins: Porphyridium cruentum B-phyco- Brown. 1984. Bilin attachment sites in the α and β
erythrin and b-phycoerythrin. J. Biol. Chem. 252: subunits of B-phycoerythrin: amino acid sequence
32-42. studies. J. Biol. Chem. 259:5472-5480.
35.Glazer, A.N., D.J. Lundell, G. Yamanaka, and R.C. 51.MacColl, R. 1998. Cyanobacterial phycobilisomes. J.
Williams. 1983. The structure of a “simple” phycobil- Struct. Biol. 124:311-334.
isome. Ann. Microbiol. (Inst. Pasteur). 134B:159-180. 52.MacColl, R., L.E. Eisele, M. Dhar, J.-P. Ecuyer, S.
36.Glazer, A.N. and G.J. Wedemayer. 1995. Cryptomon- Hopkins, J. Marrone, R. Barnard, H. Malak, and A.J.
ad biliproteins—an evolutionary perspective. Photo- Lewitus. 1999. Bilin organization in cryptomonad
syn. Res. 46:93-105. biliproteins. Biochemistry 38:4097-4105.
37.Glazer, A.N., J.A. West, and C. Chan. 1982. Phyco- 53.MacColl, R. and D. Guard-Friar. 1987. Phycobilipro-
erythrins as chemotaxonomic markers in red algae: a teins. CRC Press, Boca Raton.
survey. Biochem. Sys. Ecol. 10:203-215. 54.Mach, H., C.R. Middaugh, and R.V. Lewis. 1992.
38.Gómez-Lojero, C., B. Pérez-Gómez, D.W. Krog- Statistical determination of the average values of the
mann, and A. Peña-Diaz. 1997. The tricylindrical core extinction coefficients of tryptophan and tyrosine in
of the phycobilisome of the cyanobacterium Arthrospi- native proteins. Anal. Biochem. 200:74-80.
ra (Spirulina) maxima. Int. J. Biochem. Cell Biol. 55.Ong, L.J. and A.N. Glazer. 1991. Phycoerythrins of
29:959-970. marine unicellular cyanobacteria I: bilin types and
39.Gómez-Lojero, C., B. Pérez-Gómez, G. Prado-Flores, locations and energy transfer pathways in Synechococ-
D.W. Krogmann, A. Cárabez-Rejo, and A. Peña- cus sp. phycoerythrins. J. Biol. Chem. 266:9515-9527.
Diaz. 1997. The phycobilisomes of the cyanobacteri- 56.Ong, L. J. and A.N. Glazer. 1987. R-phycocyanin II, a
um Arthrospira (Spirulina) maxima. Int. J. Biochem. new phycocyanin occurring in marine Synechococcus
Cell Biol. 29:1191-1205. species. Identification of the terminal energy acceptor
40.Gysi, J. and H. Zuber. 1974. Isolation and character- bilin in phycocyanins. J. Biol. Chem. 262:6323-6327.
ization of allophycocyanin from the thermophilic blue- 57.Ong, L.J. and A.N. Glazer. 1988. Structural studies of
green alga Mastigocladus laminosus cohn. FEBS Lett. phycobiliproteins in unicellular marine cyanobacteria,
48:209-213. p. 102-121. In S.E. Stevens, Jr. and D.A. Bryant
41.Hess, W.R., F. Partensky, G.W.M. van der Staay, (Eds.), Light-Energy Transduction in Photosynthesis:
J.M. Garcia-Fernandez, T. Börner, and D. Vaulot. Higher Plant and Bacterial Models. American Society
1996. Coexistence of phycoerythrin and a chlorophyll of Plant Physiologists, Rockville, MD.
α/β antenna in a marine prokaryote. Proc. Natl. Acad. 58.Padgett, M.P. and D.W. Krogmann. 1987. Large scale
Sci. USA 93:11126-11130. preparation of pure phycobiliproteins. Photosyn. Res.
42.Jung, L.J., C.F. Chan, and A.N. Glazer. 1995. Candi- 11:225-235.
date genes for the phycoerythrocyanin α subunit lyase: 59.Reuter, W., G. Wiegand, R. Huber, and M.E. Than.
biochemical analysis of pecE and pecF interposon 1999. Structural analysis at 2.2 Å of orthorhombic
mutants. J. Biol. Chem. 270:12877-12884. crystals presents the asymmetry of the allophyco-
43.Kahn, K., D. Mazel, J. Houmard, N. Tandeau de cyanin-linker complex, AP-LC7.8, from phycobilisomes
Marsac, and M.R. Schaefer. 1997. A role for cpeYZ in of Mastigocladus laminosus. Proc. Natl. Acad. Sci. USA
cyanobacterial phycoerythrin biosynthesis. J. Bacteriol. 96:1363-1368.
179: 998-1006. 60.Schirmer, T.,W. Bode, and R. Huber. 1987. Refined
44.Kehoe, D. and A. Grossman. 1996. Similarity of a three-dimensional structures of two cyanobacterial C-
chromatic adaptation sensor to phytochrome and eth- phycocyanins at 2.1 and 2.5 Å resolution. J. Mol. Biol.
ylene receptors. Science 273:1409-1412. 196:677-695.
45.Klotz, A.V. and A.N. Glazer. 1985. Characterization 61.Schirmer, T., W. Bode, R. Huber, W. Sidler, and H.
of the bilin attachment sites in R-phycoerythrin. J. Zuber. 1985. X-ray crystallographyic structure of the
Biol. Chem. 260:4856-4863. light-harvesting biliprotein C-phycocyanin from the
46.Lagarias, J.C., A.V. Klotz, J.L. Dallas, A.N. Glazer, thermophilic cyanobacterium Mastigocladus laminosus
J.E. Bishop, J.F. O’Connell, and H. Rapoport. 1988. and its resemblance to globin structures. J. Mol. Biol.
Exclusive A-ring linkage for singly attached phyco- 184:257-277.
cyanobilins and phycoerythrobilins in phycobilipro- 62.Schirmer, T., R. Huber, M. Schneider, W. Bode, M.
teins. J. Biol. Chem. 263:12977-12985. Miller, and M.L. Hackert. 1986. Crystal structure
47.Lamparter, T., F. Mittmann, W. Gartner, T. Börner, E. analysis and refinement at 2.5 Å of hexameric C-phy-
Hartmann, and J. Hughes. 1997. Characterization of cocyanin from the cyanobacterium Agmenellum qua-
recombinant phytochrome from the cyanobacterium druplicatum: the molecular model and its implications
Synechocystis. Proc. Natl. Acad. Sci. USA 94:11792-7. from light-harvesting. J. Mol. Biol. 188:651-676.
333
W.M. Schluchter and D.A. Bryant
63.Sidler, W.A. 1994. Phycobilisome and phycobilipro- in cyanobacterial light-harvesting proteins. Mol.
tein structure, p. 139-216. In D.A. Bryant (Ed.), The Microbiol. 30:475-486.
Molecular Biology of Cyanobacteria, Vol. 1. Kluwer 73.Wedemayer, G.J., D.G. Kidd, and A.N. Glazer. 1996.
Academic, Dordrecht, The Netherlands. Cryptomonad biliproteins: bilin types and locations.
64.Sineshchekov, V., J. Hughes, E. Hartmann, and T. Photosyn. Res. 48:163-170.
Lamparter. 1998. Fluorescence and photochemistry of 74.Wedemayer, G.J., D.G. Kidd, D.E. Wemmer, and
recombinant phytochrome from the cyanobacterium A.N. Glazer. 1992. Phycobilins of cryptophycean
Synechocystis. Photochem. Photobiol. 67:263-267. algae: occurrence of dihydrobiliverdin and meso-
65.Stec, B., R.F. Troxler, and M.M. Teeter. 1999. Crystal biliverdin in cryptomonad biliproteins. J. Biol. Chem.
structure of C-phyocyanin from Cyanidium caldarium 267:7315-7331.
provides a new perspective on phycobilisome assembly. 75.Wemmer, D.E., G.J. Wedemayer, and A.N. Glazer.
Biophys. J. 76:2912-2921. 1993. Phycobilins of cryptophycean algae: novel link-
66.Swanson, R.V. and A.N. Glazer. 1990. Separation of age of dihydrobiliverdin in a phycoerythrin 555 and a
phycobiliprotein subunits by reverse-phase high pressure phycocyanin 645. J. Biol. Chem. 268:1658-1669.
liquid chromatography. Anal. Biochem. 188:295-299. 76.Wilbanks, S.M. and A.N. Glazer. 1993. Rod struc-
67.Swanson, R.V., L.J. Ong, S.M. Wilbanks, and A.N. ture of a phycoerythrin II-containing phycobilisome
Glazer. 1991. Phycoerythrins of marine unicellular II: complete sequence and bilin attachment site of a
cyanobacteria II: characterization of phycobiliproteins phycoerythrin g subunit. J. Biol. Chem. 268:1236-
with unusually high phycourobilin content. J. Biol. 1241.
Chem. 266:9528-9534. 77.Wilde, A., Y. Churin, H. Schubert, and T. Börner.
68.Swanson, R.V., J. Zhou, J.A. Leary, T. Williams, R. de 1997. Disruption of a Synechocystis sp. PCC 6803 gene
Lorimier, D.A. Bryant, and A.N. Glazer. 1992. Char- with partial similarity to phytochrome genes alters
acterization of phycocyanin produced by cpcE and cpcF growth under changing light qualities. FEBS Lett. 406:
mutants and identification of an intergenic suppressor 89-92.
of the defect in bilin attachment. J. Biol. Chem. 78.Yeh, K. C., S.-H. Wu, J.T. Murphy, and J.C.
267:16146-16154. Lagarias. 1997. A cyanobacterial phytochrome two-
69.Tandeau de Marsac, N. and G. Cohen-Bazire. 1977. component light sensory system. Science 277:1505-
Molecular composition of cyanobacterial phycobili- 1508.
somes. Proc. Natl. Acad. Sci. USA 74:1635-1639. 79.Zhao, K.H., M.G. Deng, M. Zheng, M. Zhou, A.
70.Tandeau de Marsac, N., D. Mazel, V. Capuano, T. Parbel, M. Storf, M. Meyer, B. Strohmann, and H.
Damerval, and J. Houmard. 1990. Genetic analysis Scheer. 2000. Novel activity of a phycobiliprotein
of the cyanobacterial light-harvesting antenna complex, lyase: both the attachment of phycocyanobilin and the
p. 143-153. In D. Drews and E.A. Dawes (Eds.), Mol- isomerization to phycobiliviolin are catalyzed by the
ecular Biology of Membrane-Bound Complexes in proteins PecE and PecF encoded by the phycoerythro-
Phototropic Bacteria. Plenum, New York. cyanin operon. FEBS Lett. 469:9-13.
71.Thomas, J.-C. and C. Passquet. 1999. Characteriza- 80.Zhou, J., G.E. Gasparich, V.L. Stirewalt, R. de Lorim-
tion of a phycoerythrin without α-subunits from a uni- ier, and D.A. Bryant. 1992. The cpcE and cpcF genes
cellular red alga. J. Biol. Chem. 274:2472-2482. of Synechococcus sp. PCC 7002: construction and phe-
72.Toole, C.M., T.L. Plank, A.R. Grossman, and L.K. notypic characterization of interposon mutants. J. Biol.
Anderson. 1998. Bilin deletions and subunit stability Chem. 267:16138-16145.
334
Index 335
Index
A
ALA (see Aminolevulinic acid)
ALA synthase, 9, 70, 72, 73–76, 79
ALAD preparation (procedure), 77
ALAS (see ALA synthase)
ALAS preparation (procedure), 73
Allophycocyanin, 311, 317–318, 321, 323
Aminolevulinic acid, 4, 71, 72–76, 95–96, 278, 299
Ammoniacal extraction (procedure), 112
AP (see Allophycocyanin)
Aqueous two phase partitioning procedure, 198–200
B
Bacteriochlorophyll, 6, 237, 255
Bacteriocide, 195, 196
Bacteriophytochrome, 5, 306
Bacteriorhodopsin, 210, 218, 256
BChl (see Bacteriochlorophyll)
Bilatrienes, 299
Bile, 273, 276, 299
Bilin adduct assay (procedure), 324, 327
Bilirubin, 273–275
Biliverdin reductase assay (procedure), 287
Biliverdin, 10, 161, 173, 175, 177, 273-276, 282, 298, 323
Binodal curve, 188–189, 190, 199
Blood, 8, 10, 15, 17, 171, 172, 193
BR (see Bilirubin)
BV (see Biliverdin)
C
Capillary electrophoresis, 95, 108
Carotenoid, 97, 122, 237, 241–245, 263
CD (see Circular dichroism)
CE (see Capillary electrophoresis)
Chl (see Chlorophyll)
Chlide (see Chlorophyllide)
Chlorophyll determination (procedure), 258
335
336 Index
Hematuria, 172
Heme chemiluminescence procedure, 171
Heme detection (procedure), 168
Heme, 8, 9, 17, 100, 102, 158, 175, 179, 209, 273, 284
Hemin, 47, 55, 165, 166–167, 285, 286
Hemoglobin preparation (procedure), 163, 164
Hemoglobin, 3, 9, 10, 15, 172, 193, 200, 204, 315
Hemoprotein spectral analysis, 169, 170, 178
Hexane extracted acetone residue, 114, 116–119, 122, 123,
126–129, 131, 133, 134, 136-138, 141-143, 145, 151, 152
High performance liquid chromatography, 17, 45–46, 57,
58, 87, 89, 95, 96, 100, 102, 103, 105–108, 116, 121–123,
132, 134, 138, 140, 149, 152, 164, 237, 243, 245, 277, 278,
280–282, 284–286, 288, 298, 318-319, 323, 327–330
HO-1 (see human heme oxygenase isozyme 1)
Horse radish peroxidase, 161, 171
H-PHEN+, 63
HPLC (see High pressure/performance liquid chromatography)
HRP (see Horse radish peroxidase)
HSAP (see Hemoprotein spectral analysis)
Human heme oxygenase isozyme 1, 173–178
Hydroxymethylbilane (also called Preuroporphyrinogen),
4, 71, 80-82
I
Insecticyanin, 273
Iron octaethylporphyrin chromatography (procedure), 43
I
LCFA (see Long chain fatty alcohol)
Leghemoglobin, 8
LHC (see Light harvesting complex)
LHC preparation (procedure), 245–246
Light harvesting complex, 111, 235, 236, 238–243, 245–249,
256, 267
Long chain fatty alcohol, 119-121, 149, 150
Lutein, 242, 245
M
Magnetic circular dichroism, 173–174
MBV (see Mesobiliverdin)
MCD (see Magnetic circular dichroism)
Mesobiliverdin, 273, 276, 281, 285, 323, 328
Methine group, 1, 34, 106, 300, 302, 305
Methyl para-toluenesulfonate, 52, 61, 63
Methyl pheophorbide isolation (procedure), 26
Mg-protoporphyrin IX monomethyl ester, 114, 116, 117,
123, 126, 127, 130-132, 134, 139, 142, 143, 145
338 Index
S
Shemin pathway, 4, 72
Siroheme, 4, 87,
Sucrose density gradient, 248–249
T
TAPP, 49, 50, 52, 63
Tetrahydrogeranylgeraniol, 139, 140, 147, 150
Tetrakis(2-amino-phenyl)porphyrin TLC (procedure), 45
Tetramethylbenzidine, 170
Tetraphenylporphyrin, 30, 31, 33, 57, 63
Thaumatin, 186
THGG (see Tetrahydrogeranylgeraniol)
Thylakoid, 111, 224, 238, 242, 266, 267
TMBZ (see Tetramethylbenzidine)
TMBZ PAGE staining procedure, 170
TMPyP(X), 48, 63
TPP (see Tetraphenylporphyrin)
TPP synthesis (procedure), 33
TPP, 26, 30, 31, 33, 42, 51, 54, 57
TPPC4, 49, 54, 63
TPPS1, 54, 63
TPPS2, 54, 63
TPPS3, 54, 56, 57, 63
TPPS4, 48, 49, 50, 54, 57, 63
TPyP(X), 49, 50, 63
Turacin, 10
Turacoverdin, 10
Two-dimensional crystal growth procedure, 212, 261–263,
265, 266–268
U
Urine, 97
Uro (see Uroporphyrin)
Uroporphyrin, 10, 98
Uroporphyrinogen III, 4, 20, 80, 81–85
UROS preparation (procedure), 83
V
Violaxanthin, 245
X
Xanthophyll, 148, 243, 244–45, 246–247
Z
Zeaxanthin, 242
Zinc blot assay (procedure), 300
Heme, Chlorophyll, and Bilins
Methods and Protocols
Edited by
Alison G. Smith
Department of Plant Sciences, University of Cambridge, UK
Michael Witty
Department of Biochemistry, University of Cambridge, UK
Although researchers can profitably investigate heme, chlorophyll, and related tetrapyrroles in a wide
range of academic and medical research programs, the handling and manipulation of these delicate com-
pounds requires considerable skill and cross-boundary knowledge. In Heme, Chlorophyll, and Bilins: Methods
and Protocols, an interdisciplinary panel of hands-on investigators overcomes these limitations by describing in
detail how to work successfully with chlorophyll, heme, and bilins in biological, medical, chemical, and bio-
chemical research. Each method is presented by a researcher who actually uses it on a daily basis and includes
step-by-step instructions and pertinent tricks-of-the-trade that often make the difference between laboratory
success and failure. Topics range from methods for the analysis of tetrapyrroles, heme, and hemoproteins, to
the biosynthesis and the analysis of chlorophyll and bilins.
Timely and highly practical, Heme, Chlorophyll, and Bilins: Methods and Protocols is a gold-standard
collection of readily reproducible techniques suitable for a wide range of researchers, whether it be a clinician
studying photodynamic therapy, an ecologist studying the chlorophyll composition of leaves in a tropical
forest, or a cell biologist investigating the function of specific hemoproteins.
Features
• Detailed step-by-step protocols that have been • Time-saving techniques that even a highly
optimized for robust results skilled researcher will find helpful
• Numerous tricks-of-the-trade that often make • Troubleshooting tips, alternative ways of doing
the difference between success and failure things, and informative explanations
Contents
Laboratory Methods for the Study of Tetrapyrroles. Syn- Phase Systems. Structural Study of Heme Proteins by Elec-
theses of Tetrapyrroles. General Laboratory Methods for tron Microscopy of 2-Dimensional Crystals. Analysis and
Tetrapyrroles. Enzymatic Preparation of Tetrapyrrole In- Reconstitution of Chlorophyll–Proteins. Two-Dimensional
termediates. Analysis of Biosynthetic Intermediates, 5- Crystallization of Chlorophyll Proteins. Biosynthesis and
Aminolevulinic Acid to Heme. Analysis of Intermediates Analysis of Bilins. Analysis and Reconstitution of Phyto-
and End Products of the Chlorophyll Biosynthetic Path- chromes. Analysis and Reconstitution of Phycobiliproteins:
way. Analysis of Heme and Hemoproteins. Hemoproteins Methods for the Characterization of Bilin Attachment Reac-
Purification and Characterization by Using Aqueous Two- tions. Index.
90000