Talanta 153 (2016) 301305

Contents lists available at ScienceDirect

Talanta
journal homepage: www.elsevier.com/locate/talanta

A method for determining arsenolipids in seawater by HPLC-high

resolution mass spectrometry
Muslim Khan, Kenneth B. Jensen, Kevin A. Francesconi n
Institute of Chemistry-Analytical Chemistry, NAWI Graz, University of Graz, Universitaetsplatz 1, 8010 Graz, Austria

art ic l e i nf o a b s t r a c t

Article history: Arsenic-containing lipids (arsenolipids), naturally occurring arsenicals in algae, have never been detected
Received 18 December 2015 in seawater even though they might be introduced to the water column on senescence of marine algae or
Received in revised form by active excretion. The complex nature of seawater presents an analytical challenge to detect these
4 March 2016
compounds and to monitor their environmental fate. We developed a simple sample preparation method
Accepted 7 March 2016
Available online 11 March 2016
using liquid-liquid extraction combined with HPLC-high resolution mass spectrometry (HRMS) capable
of measuring six standard arsenolipids in seawater at the ng As/L level (o 1% of the total arsenic in
Keywords: seawater). The method is suitable for studies on the biotransformation and pathways of arsenolipids in
Arsenolipids the marine environment. When we applied the method to four samples of natural seawater, however, we
Seawater
did not nd any of the six standard arsenolipids.
Mass spectrometry
& 2016 Elsevier B.V. All rights reserved.

1. Introduction separation and detection by atomic spectroscopy [19]. Subsequent

Arsenic is present in the world's oceans at low and fairly uni-
form concentrations of 12 g/L [1], mostly as the oxyanion ar-
senate. In marine organisms, however, the concentration of arsenic
can be very high (50 mg/kg or more wet mass) in both marine
animals and algae [2], and most of this arsenic is present as or-
ganoarsenic compounds.
The early research on these naturally occurring organoarsenic
compounds dealt almost exclusively with water-soluble arsenicals,
and led to the discovery of major arsenic natural products such as
arsenobetaine [3] and the arsenosugars [4]. Over the last eight
years, however, the focus has shifted to the study of lipid-soluble
arsenicals, so-called arsenolipids, whereupon several types have
been identied including arsenic-containing fatty acids [511],
hydrocarbons [1214,10,11], phospholipids [15,13,16] and fatty
alcohols [14]. The arsenolipids are likely biosynthesised by algae
by a process of oxidative methylation to produce dimethylarsinic
acid [17], which is then further transformed into arsenolipids.
Algal biotransformation processes are central to the biogeo-
chemical cycle of arsenic in the sea. The classic studies on seawater
proles of arsenic species by Andreae [1,18] in the late 1970s de-
monstrated that primary production controlled arsenic speciation
in terms of the simple mono- and dimethylated arsenic acids.
Andreae used an analytical method based on a vapour generation
derivatisation step to form volatile arsines followed by their
n
Corresponding author.
E-mail address: kevin.francesconi@uni-graz.at (K.A. Francesconi).
arsenic speciation studies in seawater have similarly used methods
based on vapour generation [20]. Those methods, however, would
not detect refractory arsenicals, such as arsenobetaine, or arseni-
cals with bulky substituents such as arsenosugars or arsenolipids
(these species were unknown at the time of Andreae's studies).
There have been several reports over the years of refractory ar-
senic in natural waters [2123], its presence being inferred from
speciation data showing that the sum of arsenic species generated
by vapour generation was less than the total arsenic value.
In view of the analytical limitations imposed by vapour gen-
eration techniques, the possibility remains that other arsenic
species dominant in marine biota are also present to some extent
in seawater, but have remained undetected. These compounds
could be released to the seawater by simple dissolution on se-
nescence of an organism or by active biological efux. These
processes could be similar to those involved in the cycling of
phospholipids in marine ecosystems [24]. Thus, arsenolipids might
represent a convenient way to remove accumulated arsenic be-
cause the lipid-soluble compounds would likely diffuse easily
through algal cell membranes, and in that event, arsenolipids
would be associated with primary productivity in the oceans. To
test this hypothesis, a sensitive analytical procedure is needed that
can measure arsenolipids in seawater at the low ng/L level, i.e.
o1% of the total arsenic. We report the development of a method
for detecting traces of arsenolipids in seawater based on solvent
extraction and HPLC/high resolution mass spectrometry.

http://dx.doi.org/10.1016/j.talanta.2016.03.030
0039-9140/& 2016 Elsevier B.V. All rights reserved.
302 M. Khan et al. / Talanta 153 (2016) 301305
2. Experimental
2.1. Chemicals and standards
Water used for preparing reagents and standards was obtained
from a Milli-Q system (18.2 M cm, Millipore GmbH, Vienna,
Austria). Methanol ( Z99.9%, MeOH), ethanol (100%, EtOH), hy-
drochloric acid (ROTIPURAN HCl 25%), dichloromethane (Z99.9%,
DCM), chloroform ( Z99.9%), methyl-tertiary-butyl ether
(Z99.5%, MTBE), acetonitrile (Z99.9%, ACN) and formic acid
(Z98%) were obtained from Carl Roth GmbH (Karlsruhe, Ger-
many); hexane (Z 95%), ethyl acetate (Z99.5%, EtOAc), diethyl
ether (Z 99.5%, Et2O), sodium chloride (Bioultra used in molecular
biology) were purchased from Sigma-Aldrich (Vienna, Austria).
Filters of cellulose acetate membrane (0.20 mm, 47 mm) were
purchased from ALBET lab science (Austria). NASS 6 and CASS 5
(both reference sea water) were purchased from National Research
Council of Canada (Ottawa, Ontario, Canada). Six naturally occur-
ring arsenolipids (AsFA362, AsFA388, AsFA418, AsHC332, AsHC360,
and AsHC444; see Fig. 1) were synthesized in-house [25] and their
purity ( 499%) assessed by NMR and HPLC/MS. Stock solutions of
these compounds (3.33 mM, 250 mg As/L) were prepared in etha-
nol; further solutions were prepared by dilution in ethanol. Ex-
ternal calibrations were performed in the concentration range of
0.2 mg As/L to 4 mg As/L for each of the standards.
3. Instrumentation and operating conditions
3.1. Conditions for graphite furnace- atomic absorption spectrometry
(GF-AAS)
In preliminary experiments performed with high arsenic
spikes, the arsenic content of fractions was determined by using a
GF-AAS (AA 240Z) from Varian Australia Pty Ltd (Mulgrave, Aus-
tralia) under the following conditions: Light source arsenic hollow
cathode lamp: 10 mA; wavelength: 193.7 nm; drying conditions,
150 C, 20 s (ramp mode); ashing conditions, 1200 C, 40 s (ramp
mode); atomization, 2600 C, 3 s (temperature control mode);
purge gas argon: 3.0 L min
1. Quantication was made from peak
area; each standard and sample was measured in triplicate.
3.2. HPLC/HRMS conditions
HPLC separations were performed with a Dionex Ultimate 3000
series instrument (Thermo Fischer Scientic, Erlangen, Germany).
Compounds were measured with a high resolution mass spectro-
meter (Q-Exactive Hybrid Quadrupole-Orbitrap MS from Thermo
Fischer) under electrospray ionization (ESI) conditions. The con-
ditions were optimised for positive mode while infusing an
Fig. 1. Standard arsenolipids, three arsenic fatty acids (AsFA) and three arsenic hydrocarbons (AsHC), used in this work. The compounds are referred to by the afore-
mentioned abbreviation followed by the nominal molecular mass of the compound.
ethanolic solution of AsFA-362 (50 mg As/L) at 5 ml/min combined
by a T-piece with a 1 ml/min ow of a solution of 60% ACN (0.1%
HCOOH) and 40% water (0.1% HCOOH) and resulted in settings:
Spray voltage3.3 kV, Capillary temperature300 C, Gas (N2)
temperature450 C, Gas ow75 au arbitrary units, Aux gas
ow20 au. Resolution was set at 70,000 FWHM, with automatic
gain control set to 106, maximum injection time 100 ms. Mea-
surements were carried out in full scan mode between m/z
320450.
3.3. Samples
To serve as a test sample in preliminary experiments, 35 g of
high purity NaCl was dissolved in 1000 ml of water to give a 3.5%
salt solution (the typical salinity of seawater); this solution had pH
5.7. Seawater samples used for testing the method were collected
in acid-washed polypropylene bottles (2.5 L) from Piran in Slove-
nia and from Venice in Italy. In addition, two samples of certied
reference seawater (CASS 5 and NASS 6) were also analysed. All
water samples were ltered through cellulose acetate membrane
lters (47 mm diameter, ca 0.20 mm pore size) before the solvent
extraction experiments.
3.4. Solvent selection and extraction conditions
Six solvents, namely DCM, EtOAc, chloroform, Et2O, MTBE and
hexane, were tested for their efciency at extracting arsenolipids
spiked to the NaCl solution (salt water). In these preliminary ex-
periments, we spiked the salt solution (100 mL, in a separatory
funnel) with AsFA 362, the most polar of the six arsenolipids
tested, at a concentration of 10 mg As/L and extracted it with sol-
vent (2  10 mL); the two extracts were combined and evaporated
on a centrifugal lyophilisator (Christ RVC 2-33 CD plus, Martin
Christ GmbH, Hartz Osterode am Germany) to dryness and the
residue was weighed and later analysed for arsenic by GF-AAS. For
DCM, sequential extractions were also performed by extracting
100 ml of salt water (spiked with 0.5 mg As/L each of AsHC-332 and
AsFA-362) three times with DCM (each 10 mL) and separately re-
cording the dry mass and arsenolipid content of the DCM extracts.
Recoveries for the six arsenolipids by using DCM were checked by
spiking a mixture of the compounds (each 0.01 mg As/L) to the salt
water, and then extracting the aqueous layer before measuring the
arsenolipids in the DCM layer by HPLC/HRMS. All experiments
were performed in triplicate.
Following the preliminary experiments, the method was ap-
plied to the seawater samples. Thus, a seawater sample (100 ml,
from Piran at natural pH 8.0, or from Venice at natural pH 8.0 or at
pH 5.70, adjusted by adding a few drops of 1 M HCl) were spiked
with six arsenolipids (each 0.01 mg As/L), transferred to a separa-
tory funnel and extracted with DCM (2  10 ml). The DCM
 M. Khan et al. / Talanta 153 (2016) 301305 303

fractions were combined and solvent was evaporated by cen- Table 2

trifugal lyophilisation. The residue was dissolved in 250 ml of Extraction efciency of six solvents of AsFA-362 from salt water (spiked at 10 mg As/
L) measured by GF-AAS, and their solubility in water.
ethanol (EtOH) and analysed for arsenolipids by using HPLC/
HRMS. Solvents % AsFA 362 Ex- Total solids extracted, Solubility in water
tracted Mean (SD), mg, Mean (SD), n 3 g/100 g [29]
n 3
3.5. Identication and quantication of arsenolipids by HPLC/HRMS
Hexane 44 (2) 0.9 (0.09) 0.001 g
Arsenolipids were determined in the seawater extracts by MTBE 49 (2) 1.7 (0.2) 4.8 g
HPLC/HRMS under the conditions summarized in Table 1. Data Et2O 52 (4) 1.9 (0.1) 6.8 g
evaluation was performed with the standard instrument software DCM 96 (2) 1.3 (0.08) 1.6 g
Chloroform 80 (2) 2.4 (0.2) 0.8 g
Xcalibur Version 3.0.63 (Thermo Scientic, San Jose, USA). Quan- EtOAc 83 (4) 1.3 (0.1) 8.7 g
tication was performed on the basis of peak area against the six
arsenolipid standards.

4.2. Sequential extraction by dichloromethane

4. Results and discussion
Sequential DCM extractions (3  10 mL) were performed on
100 ml of salt water spiked with 0.5 mg As/L of AsHC-332 and
We wished to develop a method able to detect traces of ar-
AsFA-362, and the extracts were individually measured. Recovery
senolipids in seawater for future investigations of their stability
results are shown in the Supplementary material (Fig. S1). For both
and fate in seawater. We chose a method of solvent extraction and
arsenolipids, the rst extraction removed 4 80%, the second ex-
HPLC/HRMS measurement, and preliminary experiments were
traction ca 10%, and only a trace (ca o 1%) was removed in the
performed with a 3.5% salt solution prepared from high purity
third extraction. Hence in further experiments, 2  10 ml of DCM
NaCl, and then applied to two seawater samples from the Adriatic
was used, and the two extracts were combined.
Sea and two CRM seawater samples from Canada. We had ar-
senolipids standards of three arsenic fatty acids and three arsenic
4.3. Recovery of six arsenolipids spiked to salt water
hydrocarbons to test and guide our sample preparation steps as
we investigated the effect of the extracting solvent and the num-
The preliminary experiments described above were performed
ber of extractions.
to determine a practical sample preparation procedure for ex-
tracting arsenolipids from water. To facilitate the detection of the
4.1. Selection of solvent
compounds, a salt water solution was spiked with standard ar-
senolipids at high levels (10 mg As/L or 0.5 mg As/L). Having es-
Chloroform has traditionally been used for the extraction of
tablished a practical method in terms of handling, experiments
lipids [26], including lipids from seawater [27], while recent work
were then performed with spiking at a more realistic arsenolipid
has also demonstrated MTBE as an efcient extraction solvent for
concentration of 0.01 mg As/L, a level representing ca 1% of typical
lipids [28]. In our work, we tested chloroform and MTBE, as well as
total arsenic seawater values (ca 12 mg/L). The residue obtained
four additional solvents with diverse polarity indices, for their
following evaporation of the solvent was dissolved in 250 ml of
efciency at extracting the most polar of the arsenolipids, namely
ethanol, so theoretically a 400-fold increase in arsenolipids con-
AsFA-362, from a salt solution (Table 2).
centration was achievable. The HPLC chromatograms of a standard
Although all solvents showed reasonable efciency at extract-
solution of arsenolipids and of arsenolipids recovered from salt
ing AsFA 362 from the salt solution, the best results were obtained
water are shown in the Fig 2. recoveries were acceptable for all
with DCM. This solvent also had the advantage of lower water
compounds, ranging from 67% to 94% (Fig. 3).
solubility compared with EtOAc and the two ethers tested. Total
extractable solid (NaCl) was negligible for all solvents in all cases
4.4. Application of method to seawater samples
more than 99% of the initial dissolved solids stayed in the aqueous
layer. On the basis of these results, DCM was chosen as the ex-
The method developed using the spiked salt solution was then
traction solvent in the subsequent experiments.
applied to the seawater samples collected from Piran in Slovenia
and Venice in Italy. However, none of the standard arsenolipids
Table 1
Optimised conditions for HPLC/HRMS determination of arsenolipids.

HPLC and HRMS

Column Asahipak reversed phase C-8 column

(4.6  150 mm, particle size 5 mm)
Column temperature 30 C
Injection volume 10 mL
Flow rate 1 mL/min
Mobile phase A: 0.1% formic acid in water
B: 0.1% formic acid in ACN
Gradient 01 min, 80% A and 20% B; 1.18.00 min, 20
100% B; and 8.110.00 min, 80% A and 20% B
ESMS Thermo Fisher Q Exactive
Mode Positive
Spray voltage 3.2 kV Fig. 2. HPLC/HRMS chromatograms of (a) standard solution of six arsenolipids
Capillary temperature 350 C (each at 4 mg As/L). (b) Extract (nal volume 250 mL) of salt solution (100 mL) that
Resolution (FWHM) R 70,000 had been spiked with arsenolipids (each at 0.01 mg As/L, i.e. 1 ng As). Extracted m/z
Scan range and SIM of char- Full range scan: m/z 320450 for each arsenolipid species. Asahipak C-8 (4.6  150 mm, 5 mm); mobile phase
acteristic ions water/ACN gradient (20100% ACN, inc. 0.1% formic acid); ow rate 1 mL min
1;
column temperature 30 C; injection volume 10 ml.
304 M. Khan et al. / Talanta 153 (2016) 301305
Fig. 3. Recoveries of the six standard arsenolipids after their liquid-liquid extrac-
tion from a spiked salt solution (1 ng As of each compound added to 100 mL of salt
solution) and measurement by HPLC/HRMS (n 3).
Fig. 4. HPLC/HRMS chromatograms of (a) sea water extract (Piran, pH 8.0) offset of
2  106 counts; (b) seawater extract (Piran, pH 8.0) spiked with six arsenolipids
(each at 0.01 mg As/L) offset of 4  106 counts; (c) standard 4 mg As/L of six ar-
senolipids offset of 6  106 counts. Extracted m/z for each arsenolipid species
Asahipak C-8 (4.6  150 mm, 5 mm); mobile phase water/ACN gradient (20100%
ACN, containing 0.1% formic acid); ow rate 1 mL/min; column temperature: 30 C;
injection volume 10 ml.
was detected in these two sea water samples (Fig. 4a). When the
Piran seawater sample was spiked with the six standard arseno-
lipids each at 0.01 mg As/L, recoveries for ve of the compounds
were similar to those obtained for the earlier experiments with
salt water (Fig. S2). The exception was AsFA 362, which returned
only ca 20% (cf 65%) of the spiked amount; this low recovery was
conrmed in a repeat triplicate experiment.
A likely reason for the observed differences is that at the lower
pH of the salt solution (5.7 versus 8.0 for Piran seawater) proto-
nation of the carboxylate group of the AsFA is favoured and the
resultant species might be expected to be more extractable into
organic solvents. However, lower pH would also favour the pro-
tonation of the dimethylarsinoyl group giving the more polar
Me2As (OH)- species. Possibly these competing effects on lipid-
solubility were more or less balanced for the AsFAs except for the
AsFA 362, the most polar of the compounds. Because sea water is
often acidied before extracting lipids from it [27], we compared
the extraction efciencies with natural seawater (pH 8.0) and
seawater acidied to pH 5.7. For seawater at the lower pH, the
recovery for AsFA 362 increased more than two-fold while the
recoveries for the other ve arsenolipids showed no signicant
differences (Figs. S3 and S4). In view of the varying response of the
different types of arsenolipids to pH, and in the interests of
minimizing sample preparation steps, initial testing of seawater
for arsenolipids is probably best performed at unadjusted sea-
water pH.
We also applied the method, under natural, unadjusted pH
conditions, to two additional seawater samples the seawater
CRMs CASS 5 and NASS 6 with certied total arsenic contents of
1.24 mg/L and 1.43 mg/L, respectively. We did not, however, detect
any of the six standard arsenolipids in these samples.
The detection limit of the method, estimated from the mean
3*SD of ten blanks of seawater (Piran, Slovenia) was between
0.050.2 ng As/L for the six arsenolipids. These levels represent
o0.1% of the total arsenic content of seawater. The fact that these
compounds were not detected in samples of seawater from four
regions suggests that arsenolipids do not make any signicant
contribution to the total arsenic content of seawater, at least not in
the seawater samples measured in this study. Calculations based
on total seawater volume (1.35  1021 L, http://www.ngdc.noaa.
gov/mgg/global/etopo1_ocean_volumes.html) and estimated total
biomass (5  1012 kg, [30,31]) containing 1 mg As/kg as arsenoli-
pids [32] would suggest an average concentration of arsenolipids
in the oceans in the order of 0.004 ng As/L. We would expect this
value to be much higher in coastal areas and waters with high
primary productivity, in which case arsenolipids should be de-
tectable by the analytical procedure reported here. It is possible,
however, that arsenolipids released to the water column, by algae
for example, have a short residence time, and are quickly degraded
to dimethylarsinic acid, a known marker of marine primary pro-
ductivity [1]. We are currently investigating these processes.
In summary, we report a method able to measure six arseno-
lipids in seawater at the ng As/L level. We developed the method
to test the hypothesis that arsenolipids could be transported to
seawater either through active biological processes or after se-
nescence of the organisms, and that previous reports of unknown
arsenic species in seawater might be explained by arsenolipids.
Arsenolipids were not found, however, in four samples of seawater
tested in this study, possibly because of the arsenolipids short
residence times. Future work will look at experimental systems
simulating marine waters to gain insight into arsenolipid de-
gradation processes that might take place in the environment.
Such experiments will depend on the method for determining
trace levels of arsenolipids in seawater described here.
Acknowledgement
This research has been supported by the Austrian Science Fund
(FWF) project number 23761-N17. We thank NAWI Graz and the
Styrian Government for supporting the Graz Central Lab En-
vironmental Metabolomics.
Appendix A. Supplementary material
Supplementary data associated with this article can be found in
the online version at http://dx.doi.org/10.1016/j.talanta.2016.03.
030.
References
[1] M.O. Andreae, Distribution and speciation of arsenic in natural waters and
some marine algae, Deep-Sea Res. 25 (1978) 391402.
[2] K.A. Francesconi, J.S. Edmonds, Arsenic species in marine samples, Croat.
Chem. Acta 71 (1998) 343359.
[3] J.S. Edmonds, K.A. Francesconi, Isolation, crystal structures and synthesis of
arsenobetaine, the arsenical constituent of the western rock lobster Panulirus
longgipes cygnus George, Tetrahedron. Lett. 18 (1977) 15431546.
[4] J.S. Edmonds, K.A. Francesconi, Arseno-sugar from brown kelp (Ecklonia ra-
diata) as intermediates in cycling of arsenic in marine ecosystem, Nature 289
(1981) 602604.
[5] A. Rumpler, J.S. Edmonds, M. Katsu, K.B. Jensen, W. Goessler, G. Raber,
H. Gunnlaugsdottir, K.A. Francesconi, Arsenic-containing long-chain fatty acids
in cod-liver oil: a result of biosynthetic indelity? Angew. Chem. Int. Ed. 47
(2008) 26652667.
[6] K.O. Amayo, A. Petursdottir, C. Newcombe, H. Gunnlaugsdottir, A. Raab, E.
M. Krupp, J. Feldmann, Identication and quantication of arsenolipids using
reversed-phase HPLC coupled simultaneously to high-resolution ICPMS and
high-resolution electrospray MS without species-specic standards, Anal.
Chem. 83 (2011) 35893595.
 M. Khan et al. / Talanta 153 (2016) 301305
[7] S. Lischka, U. Arroyo-Abad, J. Mattusch, A. Kuehn, C. Piechotta, The high di-
versity of arsenolipids in herring llet (Clupea harengus), Talanta 110 (2013)
144152.
[8] U. Arroyo-Abad, S. Lischka, C. Piechotta, J. Mattusch, T. Reemtsma, Determi-
nation and identication of hydrophilic and hydrophobic arsenic species in
methanol extract of fresh cod liver by RP-HPLC with simultaneous ICP-MS and
ESI-Q-TOF-MS detection, Food Chem. 141 (2013) 30933102.
[9] K.O. Amayo, A. Raab, E.M. Krupp, J. Feldmann, Identication of arsenolipids
and their degradation products in cod-liver oil, Talanta 118 (2014) 217223.
[10] M.S. Taleshi, G. Raber, J.S. Edmonds, K.B. Jensen, K.A. Francesconi, Arsenolipids
in oil from blue whiting Micromesistius poutassou - evidence for arsenic-
containing esters, Sci. Rep. 4 (2014) 17.
[11] U. Arroyo-Abad, Z. Hu, M. Findeisen, D. Pfeifer, J. Mattusch, T. Reemtsma,
C. Piechotta, Synthesis of two new arsenolipids and their identication in sh,
Eur. J. Lipid Sci. Technol. 117 (2015) 18.
[12] M.S. Taleshi, K.B. Jensen, G. Raber, J.S. Edmonds, H. Gunnlaugsdottir, K.
A. Francesconi, Arsenic-containing hydrocarbons: natural compounds in oil
from the sh capelin, Mallotus villosus, Chem. Commun. 39 (2008) 47064707.
[13] A. Raab, C. Newcombe, D. Pitton, R. Ebel, J. Feldmann, Comprehensive analysis
of lipophilic arsenic species in a brown alga (Saccharina latissima), Anal. Chem.
85 (2013) 28172824.
[14] K.O. Amayo, A. Raab, E.M. Krupp, H. Gunnlaugsdottir, J. Feldmann, Novel
identication of arsenolipids using chemical derivatizations in conjunction
with RP-HPLC-ICPMS/ESMS, Anal. Chem. 85 (2013) 93219327.
[15] S. Garca-Salgado, G. Raber, R. Raml, C. Magnes, K.A. Francesconi, Arsenosugar
phospholipids and arsenic hydrocarbons in two species of brown macroalgae,
Environ. Chem. 9 (2012) 6366.
[16] Sandra A. Viczek, Kenneth B. Jensen, Kevin A. Francesconi, Arsenic-containing
phosphatidylcholines: a new group of arsenolipids discovered in herring ca-
viar, Angew. Chem. Int. Ed. Engl. (2016), http://dx.doi.org/10.1002/
anie.201512031.
[17] R.V. Cooney, R.O. Mumma, A.A. Benson, Arsoniumphospholipid in algae, Proc.
Natl. Acad. Sci. USA 75 (1978) 42624264.
[18] M.O. Andreae, D. Klumpp, Biosynthesis and release of organoarsenic com-
pounds by marine algae, Environ. Sci. Technol. 13 (1979) 738741.
[19] M.O. Andreae, Determination of arsenic species in natural waters, Anal. Chem.
49 (1977) 820823.
305
[20] H. Hasegawa, Y. Sohrln, M. Matsul, M. Hojo, M. Kawashlmal, Speciation of
arsenic in natural waters by solvent extraction and hydride generation atomic
absorption spectrometry, Anal. Chem. 66 (1994) 32473252.
[21] A.G. Howard, S.D.W. Comber, The discovery of hidden arsenic species in
coastal waters, Appl. Organomet. Chem. 3 (1989) 509514.
[22] A.M. de Bettencourt, M.0 Andreaes, Refractory arsenic species in estuarine
waters, Appl. Organomet. Chem. 5 (1991) 111116.
[23] A.M. de Bettencourt, M.H. Florencio, M.F.N. Duarte, M.L.R. Gomes, L.F.C. Vilas
Boa, Refractory methylated arsenic compounds in estuarine waters: Tracing
back elusive species, Appl. Organomet. Chem. 8 (1994) 4356.
[24] M. Suzumura, Phospholipids in marine environments, Talanta 66 (2005)
422434.
[25] M.S. Taleshi, R.K. Seidler-Egdal, K.B. Jensen, T. Schwerdtle, K.A. Francesconi,
Synthesis and characterization of arsenolipids: naturally occurring arsenic
compounds in sh and algae, Organometallics 33 (2014) 13971403.
[26] J. Folch, M. Lees, G.H. Sloane Stanley, A simple method for the isolation and
purication of total lipids from animal tissues, J. Biol. Chem. 226 (1957)
497509.
[27] M. Goutx, A. Saliot, Relationship between dissolved and particulate fatty acids
and hydrocarbons, chlorophyll a and zooplankton biomass in Villefranche Bay,
Mediterranean Sea, Mar. Chem. 8 (1980) 299318.
[28] V. Matyash, G. Liebisch, T.V. Kurzchalia, A. Shevchenko, D. Schwudke, Lipid
extraction by methyl-tert-butyl ether for high-throughput lipidomics, J. Lipid
Res. 49 (2008) 11371146.
[29] C. Reichardt, Solvents and Solvent Effects in Organic Chemistry, 3rd ed., Wiley-
VCH, Weinheim, 2003.
[30] F. Garcia-Pichel, J. Belnap, S. Neuer, F. Schanz, Estimates of global cyano-
bacterial biomass and its distribution, Algol. Stud. 109 (2003) 213227.
[31] X. Irigoien, T.A. Klevjer, A. Rstad, U. Martinez, G. Boyra, J.L. Acua, A. Bode,
F. Echevarria, J.I. Gonzalez-Gordillo, S. Hernandez-Leon, S. Agusti, D.L. Aksnes,
C.M. Duarte, S. Kaartvedt, Large mesopelagic shes biomass and trophic ef-
ciency in the open ocean, Nat. Commun. 5 (2014).
[32] R.A. Glabonjat, G. Raber, K.B. Jensen, J. Ehgartner, K.A. Francesconi, Quanti-
cation of arsenolipids in the Certied Reference Material NMIJ 7405-a (Hijiki)
using HPLC/Mass Spectrometry after Chemical Derivatization, Anal. Chem. 86
(2014) 1028210287.