Вы находитесь на странице: 1из 219

Biochemistry Involving Carbon-Fluorine Bonds

In Biochemistry Involving Carbon-Fluorine Bonds; Filler, R.; ACS Symposium Series; American Chemical Society: Washington, DC, 1976.

In Biochemistry Involving Carbon-Fluorine Bonds; Filler, R.; ACS Symposium Series; American Chemical Society: Washington, DC, 1976.

Biochemistry Involving Carbon-Fluorine Bonds

Rober t

Filler ,

EDITOR

Illinois Institute of Technology

A symposium sponsored by

the Divisions o f Fluorine

and Biological Chemistry

at the 170th Meeting of the

American Chemical Society,

Chicago, Ill., Aug. 26, 1975.

ACS SYMPOSIUM SERIES

28

AMERICA N

CHEMICAL

SOCIETY

WASHINGTON ,

D.

C .

1976

In Biochemistry Involving Carbon-Fluorine Bonds; Filler, R.; ACS Symposium Series; American Chemical Society: Washington, DC, 1976.

Library of

Congress

Data

Biochemistry

involving carbon-fluorine fonds.

(ACS symposium series; 28, ISSN 0097-6156)

Includes bibliographical references and index.

1. Organofluorine compounds—Physiological effect—

Congresses. I. Filler, Robert, 1923- . II. American Chemical Society. Division of Fluorine. III. American

Chemical

Series: American Chemical Society. ACS symposium series;

Society. Division of Biological Chemistry. IV .

28.

QP981.F55B56

612'.0157

 

76-13037

ISBN 0-8412-0335-0

ACSMC8

28

1-215

(1976)

Copyright © 1976

American Chemical Society

Al l Rights Reserved. N o part of this book may

be reproduced or transmitted in any form or by

any means—graphic, electronic, including photo-

copying, recording, taping, or information storage

and retrieval systems—without written permission

from the American Chemical Society.

PRINTED IN TH E UNITED STATES OF AMERICA

In Biochemistry Involving Carbon-Fluorine Bonds; Filler, R.; ACS Symposium Series; American Chemical Society: Washington, DC, 1976.

ACS Symposium Series

Rober t

F.

Gould , Series

Editor

In Biochemistry Involving Carbon-Fluorine Bonds; Filler, R.; ACS Symposium Series; American Chemical Society: Washington, DC, 1976.

FOREWORD

The

ACS SYMPOSIU

a medium for publishing symposia quickly in book form. The format of the SERIES parallels that of the continuing ADVANCES IN CHEMISTRY SERIES except that in order to save time the papers are not typeset but are reproduced as they are sub- mitted by the authors in camera-ready form. As a further means of saving time, the papers are not edited or reviewed except by the symposium chairman, who becomes editor of the book. Papers published in the AC S SYMPOSIUM SERIES are original contributions not published elsewhere in whole or major part and include reports of research as well as reviews since symposia may embrace both types of presentation.

In Biochemistry Involving Carbon-Fluorine Bonds; Filler, R.; ACS Symposium Series; American Chemical Society: Washington, DC, 1976.

PREFACE

Τη recent years there has been increasing interest and research activity in the biochemistry and medicinal applications of compounds con­ taining the carbon-fluorine bond. The medicinal chemistry of fluorinated organic compounds was last discussed at a symposium held at the national ACS meeting in Washington, D.C. in September 1971. Simultaneously a Ciba Foundation symposium on the biochemistry and biological activities of carbon—fluorine compound and incisive discussions a symposium, promi nent contributors to the field participated, were published in 1972 by Elsevier. A survey of fluorinated compounds of medicinal interest was presented in the December 1974 issue of Chemical Technology. The biochemical aspects of C- F chemistry have developed rapidly since the pioneer studies in the early 1940s by Sir Rudolph Peters, who elucidated the mechanism of the toxic action offluoroacetateby invoking the concept of 'lethal synthesis." Special mention should also be made

of the elegant studies since the late 1950s by Charles Heidelberger and his colleagues o n th e tumor-inhibitory effects o f nucleotides o f fluorinted pyrimidines. Important advances in the biochemistry of organofluorine compounds continue unabated and a symposium on this subject, co- sponsored by the Divisions of Fluorine and Biological Chemistry, was held at the Chicago ACS meeting in August 1975. This volume includes all ten presentations and discussions at that symposium. Th e subjects are of broad interest, ranging from fluorocarboxylic acids as enzymatic

and

We hope that these reports will further stimulate those already active in the field and create a sense of excitement and enlightenment to the curious. To the newcomer we say "Welcome—don't be afraid of fluorine compounds—they're lots of fun." Finally, I wish to express my appreciation to all of the contributors for their patience and unstinting cooperation in making this book possible.

Illinois Institute of Technology Chicago, Ill. January 1976

metabolic probes to the use of perfluorocarbons as "artificial blood."

ROBERT FILLER

ix

In Biochemistry Involving Carbon-Fluorine Bonds; Filler, R.; ACS Symposium Series; American Chemical Society: Washington, DC, 1976.

1

Fluorocarboxylic Acids as Enzymatic and Metabolic Probes

ERNES T

KUN

University of California, San Francisco, Department of Pharmacology, and the

Cardiovascular Research Institute,

San Francisco, Calif. 94143

The development

of

site specific reagents

necessarily

depends on the knowledge of specific biochemical reactions, which should include the chemical structure of catalysts, sub- strates, modifiers, and products of enzymatic processes. Very littl e is known about catalytic sites of most enzymes, therefore comparison of substrates with analogues may be one of the experi- mental approaches to study this problem. It follows also that extensive information related to analytical enzymology (i.e., catalytic properties of isolated enzymes, enzymatic composition of cells) has to preceed the meaningful application of specific probes in complex systems. This endeavor is conceptually a synthetic one and intends to elucidate biological functions. It is only in relatively rare cases that inhibitors of cellular systems act in a manner predicted from in vitro enzymology. For example, a linea r multienzymatic process can be indeed regulated by a rate limiting enzymatic component, susceptible to a specific inhibitor. In this case the biological significance of the linea r multienzyme system can be studied with success. If acute inhibition does not cause rapid irreversible changes in cellula r economy, sustained inhibition may trigger a variety of compen- satory cellular processes, encompassing both intermediary and macromolecular metabolism. Experimental pathology and toxicology may benefit from these studies, since pathophysiology of environ- mental toxic effects is likely to be traced to specific initia- ting reactions of inhibitors with specific cellular sites. This experimental subject is at present only in its beginning stages. In contrast to linear systems-distributive, branching,and cycli c multienzymatic processes (1) respond i n a fa r more complex manner to perturbations by a specific inhibitor. Alterations of steady state concentrations of metabolites, and subcellular changes of topographic distribution of intermediates of metabolic path- ways will occur. Most of these consequences are unpredictable from idealized metabolic charts (2). Experimental results obtained with relatively uncomplicated fluorocarboxylic acids indicate that major metabolic pathways are apparently not

1

In Biochemistry Involving Carbon-Fluorine Bonds; Filler, R.; ACS Symposium Series; American Chemical Society: Washington, DC, 1976.

2

BIOCHEMISTRY

INVOLVING

CARBON-FLUORINE

BONDS

regulated

importantly

regulatory enzymes in cellula r systems, problems of molar ratio s

of substrate to enzymes in biologica l systems > have been reviewed

by

properties

by

of

enzymatic

of

components

The

alone

rol e

but

of

more

primary

availabilit y

substrates .

(3)

but

the

majority

of

biochemical

texts

have

not

incorporated

these

advances

at

present

time .

The

choice

of

F

substitutio n

of

H or

OH groups

in

well

known

carboxylic

acid

substrates

to

obtai n specifi c inhibitor s was based on straigh t forward chemi-

cal

consideration s

(4) .

The

present

paper

i s

restricte d

to

experimental

work

related

to

the

mode of

action

of

fluorocarbox-

yli c

acids,

developed

in

our

laboratory.

F1uoro-dicarboxylic

acids

 
 

A

serie s

of

chemical

and

enzymological

experiments

(5,6,7 ,

8,9,10,11,12,13,14,15,16,17,18,19) with mono and

acetates, fluoroglutamates mal ate , fluorolactat e and mal at e dehydrogenases , transaminases ,

glutamate and lactat e

difluor o

oxal -

dehydrogenases suggested a

reasonably uni -

form pattern

of interactio n between fluorodicarboxylic acid sub-

strates with

respective enzymatic sites . It is of interest that

introductio n

of

one

or

two

F

atoms

di d

not

change

the

conforma-

tion of the parent molecule to the extent that i t could not be recognized by enzymatic sites . In general^enzyme-fluoro-sub- strate Michaelis-Menten complexes are readily formed, but in some instances the rates of conversion to products, i.e. , the process of enzymatic catalysis itsel f is drasticall y reduced by F-substitution . The molecular reasons fo r thi s inhibitio n by F- substitutio n are as yet unexplored, and constitut e a significan t problem of mechanistically oriented enzymology, worthy of more detailed investigations. As would be expected F-dicarboxylic acids behave as relativel y uncomplicated linearl y competitive inhibitors with respect to the non-fluorinated substrate mole- cule. Kinetic analyses of the effect of fluoro-oxalacetate in bisubstrat e systems o f mal at e dehydrogenases (19) and lactat e dehydrogenase (14) support thi s conclusion. A summary of ex- perimental results i s shown in Table I. Whereas monofluoro oxalacetate i s a slowly reacting substrate of MDH, difluor o sub-

stitutio n converts thi s substrate homolog to an equally good sub- strate of MDH to oxalacetate itsel f with respect to Vmax,except difluoro-oxalacetat e has a Km of 4.0 mM (about 4.10^ higher than Km of oxalacetate). Further exploration of this significan t effec t of difluor o substitutio n may shed ligh t on the as yet un- known molecular mechanism of catalysi s of MDH. Since the pri - mary purpose of our investigation s was to obtai n biologica l probes, the stabilit y of F-dicarboxylic acids were also investi - gated in biological systems. Despite the promising in vitr o properties of e-monofluoro oxalacetatic acid as an inïïïbito r of

MDH, instabilit y prevented

As would be expected.monofluoro oxalacetate is susceptible to

it s

applicatio n

in

complex

systems.

In Biochemistry Involving Carbon-Fluorine Bonds; Filler, R.; ACS Symposium Series; American Chemical Society: Washington, DC, 1976.

fluoro-analogue

1.0 defluorination

15.0 amino­

MâXF-S)

13.0

101

ACIDS

Ma

Ma =

oxalacetate

F-CARBOXYLIC

;

dehydrogenase

= with

0.5 μΜ

transaminative

45. μΜ

μΜ

μΜ

μΜ

η *

710 μΜ

330 μΜ

300 μΜ

330

300

13

Κ

= glutamate

lactat e (F-S)

OF

4.0 mM

μΜ

1800 μΜ

640 μΜ

PROPERTIES

0.5

- -

substrate;

1^

GOT

=

LDH

(-)-erythrofluoromalate

dehydrogenase;

INHIBITORY

physiological

a-F-glutamate(NADP + )

3-fluorolactat e

a-F-glutamate(NAD + )

;

dehydrogenase

33'-difluoromalate

33* -F2-oxalacetic

33'-F2-oxalacetic

acid

F-oxalacetic

3-F-oxalacetic

F-glutaric

F-carboxylic

AND

MDH = malate

= with

SUBSTRATE

e

L(+)

GDH = ; glutamat

3
3

(S)

enzyme

(decarboxylating)

MDH(mito)

MDH(mito)

GOT(mito)

GOT(mito)

transferase activity

I:

malic

TABLE

Kidney MDH

LDH

GDH

GDH

GDH

;

Enzyme

Kidney

Kidney

Muscle

TO

molecular

Liver

Liver

Liver

Live r

Liver

r

Live

LEGEND

No.

7
8
9 10

5
6

3
4

1
2

In Biochemistry Involving Carbon-Fluorine Bonds; Filler, R.; ACS Symposium Series; American Chemical Society: Washington, DC, 1976.

4

BIOCHEMISTRY INVOLVING CARBON-FLUORINE BONDS

COOH

COOH

I

I

FCH

ÇH 2

+ Ε

I

c=o

HCNH ?

I

I

COOH

COOH

Figure

I

1.

COOH

HC

II

C-NH 2

COOH

Transaminative

COOH

COOH

I I

CH 2

C=0

I

COOH

FCH

I

HCNH 2

COOH

COOH

COOH

^

I

HCH

H 2 0

CH 2

I

I

+ NH,

C=NH

c=o

I

I

COOH

 

COOH

degradation

acid

of

monofluoro-oxahcetic

In Biochemistry Involving Carbon-Fluorine Bonds; Filler, R.; ACS Symposium Series; American Chemical Society: Washington, DC, 1976.

1.

KU N

Fluorocarboxylic

Acids

as

Probes

5

bivalent metal io n catalyzed decarboxylation to fluoro pyruvate.

The kinetic s o f thi s reactio n has been studie d i n detai l (20) . Interaction of monofluoro oxalacetate with glutamate-oxalacetate aminotransferase yields rapid elimination of F~ and ΝΗΛ + from a system containing both aspartate and the fluoro-acid (b) , repre­ senting a rare case of elimination reactions characteristic of pyridoxal-phosphate catalysis , as well known from the work o f Snell and hi s school. This i s shown i n Figure 1. From a com­ bined use of fluoro-glutarate, a relatively specific inhibito r of GDH (11) and of difluoro-oxalacetate, an inhibito r of GOT (10) the regulation o f both transaminative and oxidative pathways o f mitochondrial glutamate metabolism was furthe r elucidated (12) . Enzymatic reduction of monofluoro-oxalacetate by MDH on a prepar­ ative scale yielded (-)erythrofluoromalic acid (21) which i s an excellent inhibito r of malate dehydrogenases i n vitro . In isolated hepatocytes this inhibito r acts only on cytosolic MDH isoenzyme because i t doe mitochondrial membrane (cf . 21) , therefore i t i s useful as a cellula r probe o f this MDH isoenzyme. Extensive trial s with mitochondria and isolate d hepatocytes - as models fo r the study of complex systems - indicated that only difluoro-oxalaceti c and (-)erythrofluoromalic acidsproved useful. Difluoro-oxalace-

tate , by inhibitin g GOT,proved to be stable

in it s action . Mono fluoro-malate, besides inhibitin g cytosoli c MDH f is also an effective activator of the citrat e carrie r system of the inner mitochondrial membrane, as shown i n Figure 2 (cf . 21). When mitochondria are incubated with citrat e and the exit

and highly

specifi c

.

Cit

F-Mal

I

i

1

2

4

6

8

10

Figure

2.

Min.

Molecular Pharmacology

Spectrophotometric assay of citrate entry into mitochondria

In Biochemistry Involving Carbon-Fluorine Bonds; Filler, R.; ACS Symposium Series; American Chemical Society: Washington, DC, 1976.

6

BIOCHEMISTRY

INVOLVING

CARBON-FLUORINE

BONDS

of isocitrat e i s monitored by an externally added NADP + + Mg 2 + + isocitrat e dehydrogenase isocitrat e detector system in the dual wave length spectrophotometer, rapid isocitrat e efflu x i s observed under a wide variety of experimental conditions. This process is greatly stimulated by (-)erythrofluoromalate (see Figure 2) as indicated by the large increase of the rate of extramitochondrial NADPH formation. Fluoromalate appears to act simila r to malate (cf. 22) except it s effec t i s not complicated by penetration and subsequent metabolism in the matrix - as i t is the case with malate. Consequently (-)erythrofluoromalate is

a

translocating membrane system than oxidizable dicarboxylic acids

more

specifi c

and

useful

activato r

of

the

citrate-isocitrat e

(compare wit h 22) .

I t

i s

a

puzzl e

why

citrate-isocitrat e

exchange in energized mitochondria proceeds with the observed high efficienc y (about 80% o f added citrat e appears as isocitrat e in the extramitochondrial compartment). We présumeras discussed more extensively later through the inner mitochondrial membrane i s a more complex process than a carrie r mediated carboxylic acid exchange (22)

and may reflec t an energy coupled membrane unknown physiological significance .

process

of

as

ye t

In

vitr o

chemical

and

enzymatic

studies

thus

fa r

provided

fluor o malate and difluoro-oxalacetat e as candidates useful as probes i n cellula r systems. The discovery of a valuable tech -

nique, that of isolatio n of hepatocytes by Berry and Friend

gave significan t impetus to further studies. Gluconeogenesis

from lactat e was characteristicall y inhibite d by difluoro-oxal -

acetate

glucose formatio n from pyruvate was more sensitiv e to fluoro -

malate

and Figure 3. Combination of both fluoro-dicarboxyli c acids

(23),

and

than

to

to

a

lesse r

extent

by

(-)erythrofluoromalate,whereas

(24)

as

shown

in

Table

II

difluoro-oxalacetate

Tabl e

II

Effects of fluoro-dicarboxylic acids on rates of gluco- neogenesis in intact isolated liver cells Cells (125—225 mg wet weight), obtained from the livers of animals faeted 18 h, were incubated for 40 min at 37 °C. The initial substrate concentration was 10 mM and inhibitor concentration 2.5 mM. The values given are means ± S.E . with the number of observations in parenthesis. Rates of glucose formation have been calculated after subtraction of the average rate observed in the absence of added substrate. Tlli 8 wa s °· 0 8 ± 0 -°° 3 nmolxg^Xmin" 1 (32 observations)

Substrate

.

.

.

Apparent

added

Inhibitor added

Glucose formed

inhlbltlo n

μπιοί x g" 1 x miu - 1

·/·

L-Lactate -

0.45 ±

0.02

(20)

-

L-Lactate

Fluoromalate

0.34 ±

0.06

(4)

24

L -Lactate

Difluorooxaloacetate 0.24

±

0.04

(4)

56

L-Lactate Fluoromalate,

 

Pyruvate

Difluorooxaloacetate 0.11 ±0.0 5

0.02

0.51 ±

(3)

(20)

76

Pyruvat e

Fluoromalate

0.21 ±

0.04

(5)

59

Pyruvat e

Difluorooxaloacetate 0.38 ±

0.08

(4)

25

Pyruvat e

Fluoromalate ,

0.22 ±0.0 7

(4)

57

Difluorooxaloacetate

Fructose

-

2.82 ±

0.11

(14)

-

Fructos e

Fluoromalat e

2.81

(2)

Fructose

Difluorooxaloacetate 2.S i - -

(2)

European Journal of Biochemistry

In Biochemistry Involving Carbon-Fluorine Bonds; Filler, R.; ACS Symposium Series; American Chemical Society: Washington, DC, 1976.

1.

KUN

Fluorocarboxylic

Acids

as

Probes

7

 

J

I

.

I

0

1

2

3

4

5

 

Inhibitor

concentration

(mM )

 

European Journal of Biochemistry

Figure 3. on rate of cells from pyruvate (·) or foctate (O). Each point is the average of two experiments. Con- ditions were the same as given in the legend of Table 1. , with difluorooxaloacetate; , with fluoromalate.

inhibi t gluconeogenesis from both precursors by 76-80% without

significantly alterin g cellula r processes other than those invol - ved in the transfe r of reducing equivalents through the inner mito-

chondrial

membrane

(25,

26).

In

preliminary

studies

we

found

no

noticable

injecte d intravenously

Marked changes i n glutamate and aspartate concentrations i n the live r indicated that both fluoro carboxylic acids actually pene-

trated

undertake more extensive in vivo studies with both fluoro carbox-

toxic

effects

of

eithe r

fluoro

body

would

carboxylic

weight)

appear

acids when

mice.

to

(20 mg/100 g

cells .

It

into

live r

parenchyma

reasonable

yli c

possible

t i c approach to metabolic disorders lik e diabetes, characterized

by

chemotherapeu-

aci-ds

as

a model

large

fo r

a

experimental

abnormally

gluconeogenesis.

Studies

with

fluorocitrate

a.)Structure

of

the

inhibitory

isomer.

In

contrast

to

fluorodicarboxyli c acids of apparently low or undetectable acute

toxicity,th e

most

important

fluorotricarboxyli c

acid :

monofluoro-

citri c acid,

because

of

it s

remarkable

toxicity ,

 

plays

a

histor -

ical

significance

in

the

fiel d

of

biochemical

lesions,

an

area

established

by

Si r

Rudolph

Peters

(cf .

27).

In

a

series

of

classica l experiments Peters and his school establishe d that the

is

site

of

action

of

fluorocitrate

localized

at

an

initia l

step

of

citrat e

metabolism

(28) .

The

enzymatic

sit e

of

actio n

of

fluorocitrat e was proposed

t o

be

mitochondria l

aconitas e

(27) .

In Biochemistry Involving Carbon-Fluorine Bonds; Filler, R.; ACS Symposium Series; American Chemical Society: Washington, DC, 1976.

8

BIOCHEMISTRY

INVOLVING

CARBON-FLUORINE

BONDS

This mechanism appears to be incompatible with the results of

Guarrierra-Bobylev a and Buff a

administration of toxic doses of fluorocitrate,only the metabo-

lis m

tate - which is also a substrate of aconitase - i s oxidized nor­ mally. Because of the apparent uncertainties surrounding both the chemistry of the toxi c isomer of fluorocitri c acid and it s subcellula r mode o f actio n thi s problem was reinvestigated .

i n mitochondria whereas cis-aconi -

(29) , who found tha t afte r i n viv o

o f

citrat e

was

inhibite d

As

illustrate d

in

Figure4

(cf.

30)

the

four

possible

isomers

of monofluorocitric acid are formed eithe r from fluoroacetyl-CoA

and oxalaceti c acid

CoA. I t was shown (31,32) tha t the toxi c isomer i s formed onl y

oxalacetat e

by enzymatic condensatio n

fluor o

or

from

o f

monofluoro-oxalacetic

acety l

acid

and

acetyl

CoA wit h

whereas al l other isomers had no significan t inhibitor y effec t on

aconitase. Because

fluorocitrate (Κ η · * 50-80 μΜ) the distinctio n between "inhibi ­ tory" and "non-inhibitory

would

as well as others,observed a time dependent, anomalous inhibitio n

of

coupled test system, containing either Mg 2 + or Mn?+ (see late r fo r

of the relativel y weak inhibitor y effects of

initia l

velocity

kinetic

analyses,

we,

wish

for.

Besides

aconitase

activity ,

but only in the isocitrat e d hydrogenase

more

detail s

).

Elucidatio n

of

the

chemical

structure

of

the

toxi c fluorocitrat e isomer was successfu l despit e the difficultie s encountered i n the enzymology of aconitase . Synthesis and resol ­ utio n o f isomers was accomplished i n 1969 (33) and i t was shown that the electrophoretically separated erythro isomers (Figure 5 and Figure 6) containe d the toxi c species , which was furthe r re ­ solved and identifie d as (-)erythrofluorocitri c acid , correctl y defined as : 1R: 2Rl-fluoro-2-hydroxy-l,2,3-propan e tricarboxyli c acid. Crystallographic analysis of rubidium ammonium hydrogen fluorocitrate dihydrate (34) lent further support to our deduction, based originall y on NMR and pk analyses of electrophoreticall y resolved diastereoisomers (31,32).

Since

the

biosynthesis

of

citrat e

condensing

enzyme

plays

a

from

(-)erythrofluorocitri c

acid

key

fluoroacetic

role

in

the

acid

some

kineti c

characteristic s of

thi s

enzyme were

also

determined

wit h

F-acetv l

CoA

as

substrate .

Result s

are

shown

i n

Table

III ,

 

TABL E

3

 

Summary of Kinetic Properties of Citrate Synthase from

Pig

Heart"

 

Con­

stant

Substrate or inhibitor

Kinetic constant

Km

Acetyl-CoA Fluoroacetyl-CoA Fl uo roace ty 1-Co A Acetyl-CoA

v Fluoroacetyl-CoA

Kx

Km

25 μλί

23

2.2

2.77

μλί (/xmolcs DPNH/mg/min )

0.0084.5 (pinoles DPNH/mg/min )

'

mil

malate and

reaction was followed

DPN ,

β

malate dehydrogenase served as asourceof oxalacetate, and the by appearance of DP N H .

'The

Citric Acid Cycle"

In Biochemistry Involving Carbon-Fluorine Bonds; Filler, R.; ACS Symposium Series; American Chemical Society: Washington, DC, 1976.

KUN

Fluorocarboxylic

Acids

as

Probes

Fofmulot

(•)-Mixture

 

f

 

_

\

COOH

Corbons derived from

|

oxoloocetote or<

H-C- H

fluoro-oxaloacetote

1

|HOOC-C-O H

Carbons derived Γ from ocetote or <

H-C- F

1

fluoroocetote

L

C00H

1

2

~\

COOH1

|

H-C- F

1

2

HOOC-C-O H

H-C- H

1

CCO

H

(•)-

COOH

1

H-C- H

HOOC-i-O H

F-C- H

2

COOH1

Newman projections olong C - 2 . C - 3 axis (ionized forms)

COO"

"OOC ^ L TA J F-V^

OH

coo " HO^ L COO" JXj

coo "

Mixture

COOH 1

1

F-C- H

2

HOOC-fj-OH

H-Ç- H

COOH

COO"

F^C>

fi/S

System

Racemates

E «tensions of carbo- hydrote-omino acid rules (COOH group Ignored)

Racemates

Racemates

Configurational Prefix *

(2S.3R)-

(2R.3S)-

(2RS.3SR)-

L

«

L

9 "

^s^gLg "

(2R.3R)-

(2S.3S h

0$Lg-

(2RS.3RS)-

U t Og -

0 $ L t .LgC^ -

erythro-L $ -

erythro-0 $ -

erythro-D $ L $ -

threo-Oj-

threo-L,-

fhreo-O e L f -

Weaker acids (Component

H

I

Provisional Assignments

A )

Stronger octds (Component 8 )

Stronger octds (Component 8 )

From fluoroocetyl-CoA

«-

From fluoro-/

oxaloacetate

è

Not formed enzymaticolly

Not formed enzymaticolly

enanttomers

Figure 4.

"The

"The

Citric Acid Cycle"

Citric Acid Cycle"

The isomers of monofluorocitric acid *

*Th e

terminology

o f L. D* an d erythro

D„ originates fro m

.

B .

Vickcry [ M A

Suggested

Ne w

Nomenclature

fo r the Isomers

o f Isocitric

Acid, "

/ . Biol.

Chan.

237,

1739 (1962)1.

Fo r clarification o f RS

nomenclature,

the reader

is referred

to:

Cahn ,

R .

S.,

Ingold,

C , an d Prelog,

V. : Angew.

Client.

Intern.

Ed.

Engl.

3S5(5) ,

511

(1966) ,

an d Cahn ,

R .

S.,

7.

Chem.

Educ.

116(41) ,

50 8

(1964) .

In Biochemistry Involving Carbon-Fluorine Bonds; Filler, R.; ACS Symposium Series; American Chemical Society: Washington, DC, 1976.

BIOCHEMISTRY INVOLVING CARBON-FLUORINE 0 0 0 ο C Y ο ο ο ο ο 0
BIOCHEMISTRY
INVOLVING
CARBON-FLUORINE
0
0
0
ο
C
Y
ο
ο
ο
ο
ο
0
0 0
0 0
0000
ο
n
ο
ν
U
χ
CL
Ο
·
— ·
LJL
I
2
3
4
5
6
Journal of Biological Chemistry
Figure 5. Electrophoresis of fluorocitric acid from enzymatic and
chemical syntheses. CS: chemically synthesized fluorocitric acid; P:
picric acid marker; fluorocitric acids I and II (see text) (32).
6 £
5 Ό
4 0
3.0
2.0
Ι
Ό
0
ppm ( δ
)

I

1

l

I

CNF

1

I

I

I

t

Figure

I

I

I

6.

CH F CH 2

I

I

i

1

I

»

1

1

1

1

I

I

CH 2

t

I

1

1

1

1

W

i

1

1

1

I

I

1

CH ,

I

I

1

t

1

I

I

I

1

I

1

I

1

1

I

t

1

I

I

Journal of Biological Chemistry

Nuclear magnetic resonance spectrum of triethyl fluorocitrate (32)

I

1

1

1

BONDS

In Biochemistry Involving Carbon-Fluorine Bonds; Filler, R.; ACS Symposium Series; American Chemical Society: Washington, DC, 1976.

1.

KUN

Fluorocarboxylic

Acids

as

Probes

11

(cf .

Vmax

30) .

i n

th e

Whereas

Km fo r o f

acety l

F-acety l

presence

CoA

CoA

and

i s

it s

about

F-analo g

l/300th

i s

o f

th e

V mx

same,

measured

in

the

presence

of

the

physiological

substrate.

Fluoro-

acety l

CoA

act s

als o

as

an

inhibito r

o f

th e

condensatio n

o f

acety l

CoA

wit h

oxalacetate .

 

b.)Mode of action of (-)erythrofluorocitrate . A molecular

mechanism of fluorocitrat e toxicit y has to account fo r the irre - versible cessation of cell function in certain specific cells of the nervous system, known to be the probable anatomical organ sit e of thi s poison (28). Whereas the question of organ specificit y in the manifestation of pharmacological responses belongs to the do- main of physiology, universalit y of metabolic enzymes in most or- gans stil l provides acceptable in vitr o models fo r the study of biochemical toxicology. Consequently i f inhibitio n of aconitase is the mode of action of fluorocitrate , irreversibl e inhibitio n of thi s enzyme should b tions of the inhibito r becaus most probably present in vivo (compare with slow rates of fluoro - citrat e biosynthesis cf . 30) . Even i f these requirements are met there i s some uncertaint y why aconitas e inhibitio n should prove

fatal ,

relativel y small toxicity : eg. malonate i s not a powerful poison (cf. 1) , or as stated earlier , inhibitio n of malate dehydrogenase in vivo elicit s no detectable toxic effects . This question is of

particula r significance , since the existence of cytoplasmic and mitochondria l isoenzymes o f aconitas e (35) make i t uncertai n why the operation of the tricarboxylate cycle could not proceed un- disturbed i f only one, i.e. , the mitochondrial isoenzyme^were

when inhibitor s of other citri c acid cycle enzymes causes

inhibite d

(compare with 29). Because of the historicall y develop-

ed notion

(36) identifyin g fluorocitrat e toxicit y with aconitase

inhibition , the present evidence related to this question wil l

be

at advanced stages of purificatio n are well known to those con- cerned with the enzymology of aconitase . In al l but one instance

millimola r concentrations of ascorbate, cysteine , and Fe 2 + are required to demonstrate ful l aconitase activit y in vitro . More recent studies concerned with the mechanism of aconitase action were performed almost entirely in this highly artificia l in vitr o environment (37,38). This enzyme had a molecular weight of 68,000. Gawron et al . (39,40) while unable to reproduce the iso - latio n procedure of Villafranc a and Mildvan (37,38) obtained an enzyme (mol. weight = 66,000) which had higher specifi c activit y

examined in detail . Notorious problems of enzyme instabilit y

than

that previous in vitr o studies (37,38) are entirely valid even within the framework of cuvette enzymology. Non-competitive in -

reported by earlie r workers. It appears therefore doubtful

hibitio n of heart aconitase (37) by transaconitate with citrat e

or

therefore the explanatory mechanism of Villafranc a (41) postula- tin g two isomeric forms of aconitase appears to be without fir m

isocitrat e

as

substrates

(41)

could

not

be

reproduced

(42)»

In Biochemistry Involving Carbon-Fluorine Bonds; Filler, R.; ACS Symposium Series; American Chemical Society: Washington, DC, 1976.

12

BIOCHEMISTRY

INVOLVING CARBON-FLUORINE BONDS

experimental foundation . One o f th e - unti l recentl y - unrecog­ nized complications o f i n vitr o enzymology o f aconitase lie s i n the frequently used NADP^ dependent isocitrat e dehydrogenase assay system itself . As demonstrated i n 1974 (43)^various pre­ parations o f aconitase are susceptible to time dependent inacti -

vation by bivalent cations eg. , Mg 2 + or Mn^ + , which are necessary constituents o f the coupled assay system. Unawareness o f thi s complication can result in double reciprocal plots of reaction

rates ,

mixed

types of inhibition s by fluorocitrate

which

can simulate

competitive

o r non-competitive

with

highly

o r

variable

apparent Κ·,· values calculated from primary plots . We have also been misled by these types o f experimental observations and pos­

tulated earlie r a time dependent inactivatio n fluorocitrate , presumably by fluoro-aconitate

ted product (cf . 30) . The very low K i values fo r fluorocitrat e reported by Brand e t al . (44), assaying aconitase activitie s o f crude mitochondrial extract are apparently due to the inactivatio n o f aconitase by Mg2 , which i s superimposed to the competitive inhibitory effec t o f fluorocitrat e as reported more recently (43). In view o f the artificialit y o f the enzymology o f aconitase, we pursued this problem by isolatin g an electrophoretically homogeneous enzyme

which is , fo r a finit e period of time, full y active without any artificia l activato r whatsoever (43). Purificatio n i s shown i n

or it s defluorina -

o f

aconitase by

Table IV.

The molecular weight o f thi s cytoplasmic isoenzyme i s

 

Table

IV

 

Purification

of cytoplasmic aconitase from pig liwr

 

Purification

Total

Total acon­

Specific

activitv

at 25°'

 

slq>

protein

itase activ­

 

ity at 25°

μηιοΐα/ηηη

 

min/mg

 

Step 1 (liver

 

protein

extract)

12,300

1,045

0.134

Step 2

 

5(i0

730

1.32

Ste p

:>

180

500

2.70

Step I

 

8

85

10.0

 

Molecular Pharmacology

 

111,000.

considerably

higher

than

purifie d

preparations

requir­

ing

Fe^+ cysteine

fo r activation .

A much

less

purifie d

mito­

chondrial

isoenzyme

has also

been

isolate d

under

conditions

which

maintained maximal

activit y

without

added

activators .

From

Figure 7

i t

i s

evident

that

fluorocitrat e

acted

as

a

linearl y

competitive

reversible

inhibitor .

From

suitable

compu­

ter

summarized

models % substrate

i n Table

and inhibitory

constants

V.

I t

i s

apparent

that

were

(-)erythrofluoro -

calculated as

In Biochemistry Involving Carbon-Fluorine Bonds; Filler, R.; ACS Symposium Series; American Chemical Society: Washington, DC, 1976.

1.

KU N

Fluorocarboxylic

Acids

as

Probes

Ε

40θ|

//·>

u

COO

Jj

2ΌΟΟ

l/S(M )

I

I

OOO

I

L_

2000

Molecular Pharmacology

Figure 7. Competitive inhibition of cy­ toplasmic (A) and mitochondrial (B) aconitase of pig liver by (—)-erythro-

fluorocitrate. Rates of cis-aconitate for­ mation 240 nm at 25°. plasmic aconitase, and in B, 32 μ% of mitochondrial aconitase, were used per

test system (5-cm light path;

ume) at varied concentrations of citrate (abscissa). Curve 1, 500 μΜ. fluorocitrate; 2,10 0 μΜ fluorocitrate; 3, no fluorocitrate.

3-ml vol­

Table V

13

Siihslrutt:

constants

fitr

citrate

ami fl-ixoritralc

and

inhihilor

constants

f<»r

(—

)-r.rf/thro-Jliiornrilratc

determined

at μ/Ι 7.5 (10 w.\t Tris-ΙΚΊ)

and 25°

Isoenzyme

K m

K,

 

Citrate

</-Iso

Citrate

d Iso-

 

citrate

citrate

 

Α·

;

A

(A)

tnU

μΜ

μΜ

μΜ

Cytoplasmic

220

700

17

18

27

20

Mitochon­

drial

420

250

17

00

57

35

" A , determined by the cis-aconitatc assay; B ,

determined by the isocitrate dehydrogenase assay

(Μβ-'"

concentration

varied

between 0.1

an d 5.3

nm;.

citrat e

behaved

Molecular Pharmacology

as an uncomplicated competitive

and reversible

inhibitor ,

therefore

this

kinetic

property

did not satisfy the

mechanistic

requirements

se t forth

by

the well

known

irreversibl e

and

highly

potent

toxicologica l

effec t

o f

fluorocitrate .

Although

i t

i s

apparent

that

the enzyme

requiring

no arificia l

activato r

i s

probably

close r

to

the enzyme

functioning

i n the

cellula r

environ-

In Biochemistry Involving Carbon-Fluorine Bonds; Filler, R.; ACS Symposium Series; American Chemical Society: Washington, DC, 1976.

14

BIOCHEMISTRY

INVOLVING

CARBON-FLUORINE

BONDS

mentjits properties are stil l different from aconitase located

i n isolate d mitochondria . I t was shown (43) tha t Mg2 + is a

potent inactivato r of the

isolate d

enzyme.

In

sharp

contrast,

lysosome fre e mitochondria , afte r accumulating

over

200 mM

Mg^ +

in the inner membrane and matrix (46) maintain a full y activ e intramitochondrial aconitase for several hours at 37°, clearly indicating that in vitr o sensitivit y to Mg^ + of isolated aconi­ tase has no relationshi p to the intramitochondrial catalyti c environment of thi s enzyme. This fac t further stresses the need fo r extreme caution in projecting cuvette enzymology to biologi ­ cal systems.

Defluorination

of

fluorocitrate

by

isolated

aconitase

(37),

has been recently reported by Villafranc a and Platu s (45)

presence of an about 100 fol d molar excess of fluorocitrat e over

aconitase. This system als o contained 10 mM cysteine and 5 mM

Fe^ + . It i s noteworthy that enzyme preparations requiring no activators faile d to defluorinat lar results were also obtaine

Gawron

again in the presence of 5 mM Fe* + , 10 mM cysteine, 30 mM ascor-

bate, 0.126 μ moles of aconitase and 1 mM fluorocitrate . Only 3% o f 1.0 mM fluorocitrat e was defluorinate d and thi s reversibl e

reaction stopped in 1 to 1.5 minutes. Since

in

the

(48)

did

observe

in

vitr o

defluorinatio n

of

fluorocitrate ,

inhibitio n

of

the

enzyme which occurs during thi s process i s reversible (cf. 45),

whereas, as shown later , mitochondrial citrat e utilizatio n i s inhibite d in an irreversibl e manner by very low concentrations of fluorocitrate , i t seems unlikely that this phenomenon bears on the molecular mechanism of fluorocitrat e poisoning.

In

contrast

to

kinetic

studies

with

isolated

aconitase

pre­

parations, isolated mitochondria respond to fluorocitrate in a

manner

which

seems

to

bear

more

directl y

on

toxicology.

When

the citrat e

influ x

and

isocitrat e

efflu x

(21)

of

isolate d intac t

mitochondria

are

measured

following

preincubation

of

mitochondria

with less than micromolar concentrations of (-)erythrofluoro - citrate , marked and irreversibl e inhibitio n of thi s process i s

obtained. Isocitrate efflu x i s inhibite d when eithe r citrat e or

cis-aconitate are added externally (49).

Results

are

shown

in

Figure 8 (a & b) and Table VI.

structur e was disrupte d by the nonioni c detergen t Trito n X-100

When the

mitochondrial

membrane

ful l aconitas e activit y o f mitochondria was obtaine d even i n

the

presence of low concentrations of fluorocitrate > whic h in the

same

preparation completely inhibited the flux of tricarboxylic acids

into

intac t

mitochondria.

It

i s

also

apparent

(Figure

8b)

that

activation of isocitrate efflu x by fluoromalate is also inhibited by preincubatio n with 2 μΜ fluorocitrate . The absence of inhibi ­

tion by 10" 8 to ΙΟ" 6 M fluorocitrate of aconitase of lysed mito­

chondria agreed with

trate in intact mitochondria therefore irreversibly inhibited a membrane associated process essentia l fo r energy dependent tri- ' carboxylic acid translocation. Similar experimental results

in

vitr o

enzyme

kinetic s

(43).

Fluoroci ­

In Biochemistry Involving Carbon-Fluorine Bonds; Filler, R.; ACS Symposium Series; American Chemical Society: Washington, DC, 1976.

1.

KUN

Fluorocarboxylic

Acids

as

Probes

Mia

Biochemical and Biophysical Research Communications

15

Figure 8a. Rates of isocitrate efflux from cis-aconitate

as substrate (1

cis-aconitate

C = cis-aconitate added after 4 min preincubation

without F-citrate; D = mitochondria preincubated with 5 μΜ F-citrate for 4 miny then cis-aconitate addi­ tion. At arrows, Triton X-100 is added (see Results).

Abbreviations used in Figures 8a and b: C is = cis- aconitate; F-Cit = fluorocitrate; F-Mai = fluoro-

malate; Cit =

citrate;

=

Triton X-100.

2~

4

6

8

K)

12

Min

Biochemical and Biophysical Research Communications

Figure 8b. Rates of isocitrate efflux from citrate (1 mM) as substrate. Upper curve = at 4 min, 1 mM F-malate added. Lower curve = at 2 min, 2 μΜ F-

F-

citrate;

malate, being added in succession.

at 4 min,

1 mM

citrate;

at 7 min,

1 mM

were

citrat e i n preloaded mitochondria was not inhibite d by added fluorocitrat e (44),hence i t was concluded tha t the anion carrie r

obtained

by

Brand

et

al .

(44).

Exchange

diffusio n

of

itsel f

must

was

be

not

noted

the

targe t

of

however,

that

inhibitio n

by

the

essentia l

fluorocitrate .

prerequisite

for

I t

In Biochemistry Involving Carbon-Fluorine Bonds; Filler, R.; ACS Symposium Series; American Chemical Society: Washington, DC, 1976.

16

•Added

(see

BIOCHEMISTRY

INVOLVING CARBON-FLUORINE BONDS

 

Table

VI

INHIBITIO N

OF

ISOCITRAT E

EFFLU X

FROM

ADDED

CITRAT E

 

AND

CIS-ACONITATE

BY

FLUOROCITRATE

 
 

%

Inhibitio n

o f

Isocitrat e

Efflu x

 

M

M

F-citrat e

F-citrat e

citrate *

 

cis-aconitate *

1.0

X

10" 8

7

7

1.2 5

10" 8

 

11

13

2 . 5

10"

8

31

30

5.

0

10~ 8

61

51

1.0

X

10" 7

5.

0

10" 7

100

78

a s

substrate s

(1

mM)

afte r

1

minut e

preincubatio n

wit h

Colum n

1) .

F-citrat e

Biochemical and Biophysical Research Communications

inhibitio n of mitochondrial citrat e transfer i s preincubation of mitochondria fo r 2 to 8 minutes with very low concentrations of

fluorocitrat e i n the absence

taneously

inhibition . In preloaded mitochondria citrat e concentration i s

3-4 mM (cf . 44). Itwould be therefore

o f

citrat e

which,if

added

simul­

with

fluorocitrate

would

at 5 mM concentration > prevents

expected

that

under

these

1(T 6

to

10" 8

We have reinvestigated

Eva

a

extraordinary stabilit y o f lysosome free mitochondria

have followed the time course o f ATP synthetase activit y o f these organells fo r 40-60 minutes. Since ATP synthetase activit y under these conditions i s much faste r than substrate permeation into

the

M

fluorocitrat e

Kirste n

differen t

not be effective

this

problem

scientis t

from

i n

technique.

conditions.

with

Dr.

collaboration

o f

advantage

(visitin g

the University

Taking

Berlin)b y o f the

(50)

ATP,

we

experimental

matrix,

the time

course

o f

32p incorporation

into

induced by external substrate t i s a sensitive measure o f the rate

of

substrate

to

i s

translocation.

in

This

mitochondria

ADP

illustrate d

fo r

10 minutes

endogenous

deplete

Figure

After

i n the presence

9.

preincubation

of

unlabel led

of P i + (either

substrates,

32 P + substrates

10

mM citrat e

-

Tris ,

0.5 mM malate-Tris

separately

or

simul­

taneously)

were

added

and th e rate

o f

ATP synthesis

assayed (51)

by

th e radiochemical

method.

The rate o f ATP synthesis

reached

a

plateau

i n

15 minutes

when eithe r

substrates

were

added

sep­

arately ,

but proceeded

a t a high

rate

fo r 60 minutes

when

In Biochemistry Involving Carbon-Fluorine Bonds; Filler, R.; ACS Symposium Series; American Chemical Society: Washington, DC, 1976.

1.

KUN

Fluorocarboxylic

Acids

as

Probes

MINUTE S

17

Figure 9. Time course of ATP synthesis by lysosome- free mitochondria (50). After preincubation for 10 min in the presence of 5 mM ADP + 5 m M P<, to deplete endogenous substrates, 32 Ρ (0.5 to 1.0 X JO 5 dpm) and substrates (see text) were added Τ =30° .

were added simultaneously. This time course i s

characteristic of the kinetics of malate or fluoromalate activa ­ ted citrat e translocation (Figure 2) . When^during preincubation of mitochondrials nanomoles of (-)erythrofluorocitrate per mg mitochondrial protein and 40 mM Mg2+ were present and the reac ­

tio n was starte d by eithe r glutamate (2. 5 mM) plu s malate

mM)

synthesi s was completely inhibite d when

substrat e but was unaffecte d when glutamate + malate were the

substrates (FigurelO). It i s interesting that Mg2 + accumulation

citrat e was the permeant

citrat e + malate

(2. 5

o r

citrat e

(10 mM) plu s

malate

(0. 5 mM)^the

rat e

o f

ATP

(^230

mM into

the

matrix,

cf .

50)

caused

a

3-fol d

augmentation

o f ATP synthetas e activit y (compare Figure s 9 wit h 10^ thus Mg 2 +

has

no

inhibitor y

effec t

on

aconitase

i n

situ .

The

critica l

role

of

the

time

of

preincubation

with

50 picomoles

fluorocitrate

In Biochemistry Involving Carbon-Fluorine Bonds; Filler, R.; ACS Symposium Series; American Chemical Society: Washington, DC, 1976.

18

C

'to

BIOCHEMISTRY

INVOLVING CARBON-FLUORINE

MINUTE S

Figure 10. Conditions were the same as described

for Figure 9,

C = citrate + malate control; A = glutamate +

malate control; Ό = glutamate + malate + preincu­ bation with 50 ρ moles of F-citrate per mg mitochon­ drial protein; Β = citrate + malate + preincubation with 50 ρ moles of F-citrate per mg mitochondrial

except 40 mM Mg 2 * was also present.

protein;

=

30° .

BONDS

per mg mitochondrial protein is illustrate d by the following results . Citrat e dependent ATP synthetas e activit y was inhibite d by 52% afte r 3 minutes preincubation , 63% afte r 6 minutes pre - incubation^and 90-100% afte r 12 minutes preincubation. This time course has direc t relationshi p to the known kinetic s of fluorocitrat e poisoning , which has a slow onset but i s irreversi ­ ble. Once citrat e metabolism of mitochondria i s inhibite d by traces of fluorocitrat e no subsequent manipulation can reactivate this specific process. Penetration of other mitochondrial sub­ strate s and thei r subsequent metabolism i s unaffected under con­ dition s when citrat e permeation i s completely inhibited . No precise molecular interpretation of these results is as yet

In Biochemistry Involving Carbon-Fluorine Bonds; Filler, R.; ACS Symposium Series; American Chemical Society: Washington, DC, 1976.

1.

KUN

Fluorocarboxylic

Acids

as

Probes

19

possible , however i t seems clea r that predictions based on cu - vette enzymology of isolate d aconitase does not full y account fo r these phenomena. Technology of enzyme isolatio n in al l but

one

pendent on artificia l activators . The preparatio n which i s full y active without Fe 2 + + cysteine is inactivated in vitr o by Mg 2 +

case modifies aconitase and makes it s catalyti c functio n de-

(or

Mn 2 + )

yet

in

the

intact

mitochondria

this

inactivation

by

Mg 2 + does not take place , therefore the micro environment of the mitochondrial enzyme has not yet been reproduced i n vitro . Inhibitio n of the energy dependent citrat e transport by traces of fluorocitrat e in isolate d mitochondria exhibits many charac- teristic s which appear to predict the in vivo toxicit y of fluoro - citrate . Further analysis of thi s system and isolatio n of inner membrane components of the transport apparatus are necessary fo r real progress in this field . It is significan t that Peters (cf. 27) intuitivel y predicted that the biochemical lesio n in fluoro - citrate toxicit y is probabl chemistry. Our recent ciall y because no biochemically meaningful argument can be pro- posed for the critica l role of citrat e transfer in mitochondrial and especiall y i n cellula r metabolism unless an as yet unknown function of this process in the maintenance of energy transduc- tion of the inner mitochondrial membrane i s postulated.

Acknowledgement

Most

of

the

experimenta l

work

was

except

9

and

the

was

Awardee

experiments

sponsored

of

the

10)

Career

concerned

with

by

USPH.

GM-20552.

Literature

1. Webb,

Cited

J.L .

"Enzyme and Metabolic

supported

ATP-synthetase

by

HD-01239,

(Figures

E.

Kun

i s

a

researc h

Inhibitors, "

Acad.

Press ,

New

York 1963.

2. Dagley, S.

and

Nicholson ,

D.E.

"An

Introduction

to

Metabolic

Pathways,"

John

Wiley

& Sons,

Inc.,

New

York

1970.

 

3. Sols , A.

and

Gancedo,

C.

"Primary