Fatty acid synthesis in mitochondria elongation of short-chain fatty acids and formation of unsaturated long-chain fatty acids

Fatty acid synthesis in mitochondria elongation of short-chain fatty acids and formation of unsaturated long-chain fatty acids

BIOCHIMICA ET BIOPHYSICA ACTA BBA 55409 FATTY ACID SYNTHESIS ELONGATION IN MITOCHONDRIA OF SHORT-CHAIN UNSATURATED LONG-CHAIN FATTY FATTY A...

1MB Sizes 0 Downloads 31 Views

BIOCHIMICA ET BIOPHYSICA

ACTA

BBA 55409 FATTY

ACID SYNTHESIS

ELONGATION

IN MITOCHONDRIA

OF SHORT-CHAIN

UNSATURATED

LONG-CHAIN

FATTY

FATTY

ACIDS

AND FORMATION

OF

ACIDS

E. J. V. J. CHRIST Unilsvev Research Laboratory, (Received

Vlaardingen

(The Netherlands)

August zgth, 1967)

SUMMARY

The rate of incorporation of [14C]acetyl-CoA, [14C]malonyl-CoA and [14C]pyruvate into fatty acids by rat heart sarcosomes has been studied. Under aerobic conditions acetyl-CoA and malonyl-CoA are incorporated at about equal rates, whereas pyruvate is incorporated most rapidly. Under anaerobic conditions acetylCoA incorporation is stimulated considerably, while malonyl-CoA incorporation is stimulated only slightly, which suggests that the substrate actually incorporated is acetyl-CoA and not malonyl-CoA. The incorporation of acetyl units appears to be strictly dependent on the presence of either endogenous or added fatty acids. The labelling pattern of palmitic and myristic acids labelled with 1°C from [z-**C]pyruvate by rat heart sarcosomes does not suggest a de +zoeto synthesis. By studying the incorporation of short-chain fatty acids into fatty acids with longer chains it is found that elongation with one C, unit is the major reaction, regardless of the chain length of the fatty acid administered; butyric and hexanoic acids are very inefficient precursors for the synthesis of long-chain fatty acids. The labelling pattern of mono-, di-, and polyenoic acids has also been studied. It is shown that 2, 3, g and rr-monoenoic acids and /I-hydroxy acids are being labelled. There are indications that the g-monoenoic acids are probably formed by the action of an cc&/J,y-isomerase on the ~,~-unsaturated intermediates of B oxidation and elongation reactions, rather than by direct desaturation.

INTRODUCTION

The incorporation of acetate into fatty acids by mammalian mitochondria, first reported by DITURI eb al.‘, has been the subject of investigations by H~LsMANN~~~ and WAKIL,

&~CLAIN AND WARSHAW~

and WAKIL~.

It was shown by H~~LSMANN that

this

well as unsaturated, fatty acids was dependent on the presence of certain Krebs-cycle intermediates, an observation which, together with other experimental findings, suggested that the malonyl-CoA pathway might be operative in mitochondriaz~3~p. incorporation

B&him.

Biophys.

into

saturated,

as

Acta, 152 (1968) 50-62

FATTYACIDSYNTHESIS IN MITOCHONDRIA

51

On the other hand, according to HARLAN AND WAKIL~**,two mitochondrial systems are responsible for the incorporation of acetate into long-chain fatty acids: one for the synthesis of saturated fatty acids de novo using malonyl-CoA as intermediate, the other for the synthesis of saturated as well as unsaturated fatty acids by chain elongation of endogenous fatty acids. When avidin was added to a soluble extract of a mitochondrial acetone powder, a decrease in the synthesis of stearic acid was noted without a decrease in the production of oleic acid or other unsaturated fatty acidsa. HOLLOWAYAND WAKIL~ subsequently demonstrated that the octadecenoic acid formed in the presence of avidin is primarily cis-vaccenic acid and not oleic acid, its precursor being palmitoleic acid. The results of HARLAN AND WAKIL’~~ were confirmed by BARRON~~,who demonstrated also that optimal rates of acetate incorporation were obtained by anaerobiosis. Recently, the formation of malonyl-CoA in soluble preparations from rabbit heart sarcosomes was reported by H~~LSMANN~~. To investigate whether malonyl-CoA is used as the condensing carbon unit in mitochondrial fatty acid synthesis, the incorporation of acetyl-CoA and malonyl-CoA has been studied under aerobic as well as anaerobic conditions. To obtain information on the significance of a de nova synthesis in mitochondria, the distribution of radioactivity, derived from [z-Xlpyruvate, among the carbon atoms of the labelled product has been investigated. In addition, experiments on the elongation of short-chain fatty acids (C,-C,) are described. The formation of mono- and polyenoic acids was investigated to obtain information on the possible occurrence of fatty acid desaturation reactions in mitochondria. EXPERIMENTAL

PROCEDURES

Isolation of mitochondria Rat heart sarcosomes were isolated in 0.23 M sucrose plus 0.01 M EDTA as described by CLELANDAND SLATER12. For the initial homogenization a PotterElvehjem homogenizer with Teflon pestle was used. Fatty acid synthesis The amount of radioactivity incorporated into fatty acids from X-labelled substrates was taken as a measure of the fatty acid synthesis. Incubation medium: g5 mM KCl; I mM EDTA; 50 mM potassium phosphate buffer (pH 7.4) ; 5 mM MgCl,; 0.2 mM NADP+; IO mM ATP; 0.18 mMCoA; 1.8 mM glutathione; IO mM m.-isocitrate; 14C-labelled substrates as specified in experiments; about I mg protein; final volume r ml. Incubations were at 25” for 15 min. The experiments that were run under anaerobic conditions were carried out in Thunberg tubes. The tubes were evacuated and fluxed with N, 3 times. The reactions were stopped by addition of an equal volume of I M ethanolic KOH and the reaction mixtures saponified for I h at 75”, acidified and extracted 3 times with light petroleum. The light petroleum phase was washed once with water when 14C-labelled acetate, pyruvate, or malonate were used as substrates. If necessary, the fatty acid mixture was split into two fractions, using a column B&him.

Biophys.

Acta,

152 (1968) 50-62

52

E. J. V. J, CHRIST

of paraffin on water-repellent Hyflo 13p14.The lower fatty acids (C,, and lower) were eluted with 50% acetone in water and the eluate rendered weakly alkaline and evaporated to dryness. From the residue the fatty acids were obtained by grinding with KHSO, and pentane. These fatty acids were separated by gas chromatography as free acids. The higher fatty acids (from C,,) were eluted with 750/b acetone in water and subsequently esterified with diazomethane. The methyl esters were separated with an F & M laboratory gas chromatograph 1Model700 (IO?; polyethylenegly~ol adipate on Diatoport S at 165-200~; 3” per min). The lower fatty acids (C, up to C,,) were separated as free acids under the same conditions. In this case the single components were re-chromatographed to decrease their mutual contamination (about I%). The fatty acids were isolated from the gas chromatograph by condensation in glass cartridges filled with Ballotini beads and cooled with solid CO,. De~~adat~omoffattyacids The labelled saturated fatty acids were oxidized according to the method* of MURRAY~~. With this method, the acid is oxidized overnight in dry acetone with KMnO,. In this way the acid is oxidized only at the carboxyl end and a series of lower homologues is formed; these homologues are subsequently separated by gas chromato~aphy. In order to obtain short-chain acids (C&J also, the oxidation with KMnO, was carried out in acid aqueous medium (0.5 M H,SO,) under simultaneous steam distillation. These acids were separated on a partition column with 2 M glycine buffer on silica gel as immobile phase and chloroform-butanol mixtures as developeP.

Methyl esters of the unsaturated acids produced from the X-labelled precursors were separated by gas chromatography, enabling the isolation of the C,, group and C,,-tetraenoic acid. The C,, group was then separated into saturated, mono-, di-, and trienoic bands by thin-layer chromatography, using IS”/ w/w AgNO, in silica gel G with benzene-light petroleum (80:20, v/v) as solvent”. Methyl eicosatetraenoate was separated from accompanying impurities in the same system, but with diethyl ether as solvent. The zones were made visible under ultraviolet light by spraying with dichlorofluorescein. Appropriate zones were scraped off and extracted with ether. The different methyl esters thus obtained were then subjected to ozonolysis in methanolrs. The aldehyde fragments were either converted to dinitrophenylhydrazones (to extract the short-chain fragments, the water phase containing the dinitrophenylhydrazones being extracted exhaustively with diethyl ether) or oxidized further with performic acid to the corresponding acids. The dinitrophenylhydrazones were separated on silica gel G plates with light petroleum-ether (85 : 15, v/v) as solvent, The dicarboxylic acids were esterified with diazomethane and separated by gas chromatography on 10% polyethyleneglycol adipate on chromosorb W at 170-200~; 4’ per min. A mixture of dimethyl esters of chain length C&Z,, inclusive was usually added as carrier. Collection of the rather volatile C, ester was not attempted. The overall recovery of radioactivity after oxidation and chromatographic analysis was 60-70%. * This method was modified in 1964 for localization of ‘*C in saturated fatty acids by G. H. JOUVENAZ, D. K. NUGTEREN. H. VAN TILBORG AND R. I<. BEERTSIUI~of this laboratory.

FATTY ACID SYNTHESIS

IN MITOCHONDRIA

53

Separatiolz of mitochondrial lipids After completion of the reaction, 0.5 ml 20% trichloroacetic acid was added to the reaction mixture, which was centrifuged. The pellet was washed twice with 5% trichloroacetic acid and once with water. The pellet was then extracted with methanol-chloroform (I : I, v/v). The extract was concentrated, spotted onto silica gel plates and the lipids were separated in methanol-chloroform-water (25 : 75 : 4, v/v/v). Determination of radioactivity 14Cactivities were counted in a Packard liquid scintillation spectrometer, using dioxane scintillator fluid (IZ g s,3diphenyloxazole (PPO) ; 600 mg dimethyl-r+bis(5-phenyloxazolyl-z)-benzene (POPOP); 60 g naphthalene; IOOO ml dioxane; 200 ml cellosolve) or toluene scintillator fluid (1000 ml toluene; 4 g PPO; 0.1 g POPOP). Specific activities of fatty acids were calculated from 14Cactivities and gas chromatographic mass-records, calibrated with the use of an internal standard. Before counting, the dinitrophenylhydrazone solutions were reduced by bubbling through hydrogen gas with Pd catalyst added, thus avoiding quenching as much as possible. Decarboxylations of free fatty acids using the Schmidt reaction were performed in Thunberg tubes as described by GOLDFINE AND BLOCH~~. The determination of the enzymic decarboxylation of [r,3-14C,]malonyl-CoA was carried out in a Warburg vessel, collecting the 14C0, in Hyamine solution, placed in the centre well, after tipping 0.3 ml 4 M H,SO, into the reaction vessel, and subsequent counting. Materials [r-14C]Acetyl-CoA was prepared according to WIELAND malonyl-CoA noyl-CoA choline

and

[z-14C]malonyl-CoA

and lauryl-CoA was prepared

according

according to GOLDMAN

by the procedure

AND KBPPE~O, [1,3-*~c,]-

to TRAMS AND BRADY~~, AND VAGELOP.

and octa-

Lysophosphatidyl-

of TATTRIE 23, lysophosphatidylethanolamine

a1.24. The purity of the [I-14C]fatty acids was checked by gas chromatography or thin-layer chromatography and was found to be > 99.7% and 99.9%, respectively.

was prepared

by the procedure

of MAGEE et

Protein determination Protein was determined by the biuret method of GORNALL, DAVID~~, as modified by JACOBS et aL2e.

BARDAWILL

AND

RESULTS

Incorporation of C, units In Table I the rates of incorporation of 14C-labelled acetyl-CoA, malonyl-CoA and pyruvate are compared. During the first 15 min the incorporation of these substrates was found to be linear with time (see also Figs. I and 2). Under aerobic conditions and at equimolar concentrations, acetyl-CoA and malonyl-CoA were incorporated into fatty acids at about equal rates. (The same results were also found with sonically disintegrated sarcosomes.) Under anaerobic conditions, however, acetyl-CoA was the better substrate. Under the anaerobic conditions of Expt. I, the rate of malonyl-CoA decarboxylation is about 1.6 mpmoles/min per mg sarcosomal protein. Biochim.

Biophys.

Acta,

152 (1968) 50-62

E. J. V. J. CHRIST

54

As can be seen from Table 1, anaerobiosis stimulates malonyl-CoA incorporation only slightly. Acetyl-CoA incorporation was not influenced by addition of HCO,-, either under aerobic or anaerobic conditions. The highest rates of incorporation were obtained with jP4C]pyruvate. When octanoate in the incubation medium was replaced by octanoyl-CoA, roughly the same rates of pyruvate incorporation into fatty acids were observed (cf. Expts. 4 and 5). TABLE

I

INCORPORATION

OF ACETYL-CO&

MALONYL-CO+,

AND

PYRUVATE

INTO

FATTY

ACIDS

RY RAT

HEART

SARCOSOMES

Incubation medium: 95 mM KCl; I mM EDTA; 50 mM potassium phosphate buffer (pH 7.4); 5 mM MgCl,; 0.2 mM NADP+; ro mM ATP: o.18mM CoA; 1.8 mM glutathione; IO mM DL-iSOcitrate; [r-i*C]acetyl-CoA; [x,3-‘%,] malonyl-CoA; [2-**CJmalonyl-CoA; [z-i*C]pyruvate as indicated ; 240 ,& octanoic acid (sodium salt) ; about I mg protein: final vol. I ml. Incubations were at zs” for 15 min. In Expt. 5, 240 mpmoles octanoyl-CoA were added instead or octanoate. Ex-pt.

Substrate

Am074Td

(mpvnoles)

NO.

[I-i4C]Acetyl-CoA [2-1PC1Maionyl-CoA [r-l~C~~~alonyl-Co~~

0.20

200

0.40 0.36

200

7500 go00

0.71

0.31

3oo 200 250 300 6000

I.00

0.27

0.24

100

[I-%]

[2-*4C]Pyruvate iz-‘*CjPyruvate [z-lPC;Pyruvate

0.15

100

100

[I-%]Acetyl-CoA [r-i4C]Acetyl-CoA [r -r*CJMalonyl-Co A [I-i*C]Malonyl-CoA

[z-r*C]Pyruvate [z-X]Pyruvate [r-WZ]Pyruvatc

100

100

[r-%]Acetyl-CoA [z-i*C]Malonyl-CoA

Acetyl-CoA jr-WjAcetyl-Co.4

14C incorfiovated into ifatty acids (m~m&slmg protein per 15 min)* __ __l_____-~Aerobic conditions A~mvobic conditzons

I.15 1.16 0.41 0.39

0.48 0.45 2.31 2.81 2.01

I.35 2.30 2.32 --~ ----. * In the calculation it was assumed that one carboxyl group of the [r,3-**C,]malonyl-CoA cule was lost as i4C0, on incorporation of the molecule into fatty acids. 5

3000

7500 go00

mole-

The relationship between the incorporation of pyruvate into fatty acids by rat heart sarcosomes and the incubation time is shown in Figs. I and 2. The incorporation of pyruvate proceeded for 15 min and then stopped. The synthesis started again after addition of a further amount of the mitochondrial preparation (Fig. I). The addition of myristic acid also restored the pyruvate incorporation (Fig. z). Several other compounds were tested for their ability to stimulate pyruvate incorporation. In Table II data are summarized which demonstrate that the rate of mitochondrial pyruvate incorporation into fatty acids is determined by the availability of free fatty acids or their CoA esters. The stimulations observed on addition of lysophosphatidyl compounds were small and are not considered to be significant. In any case it is clear that lysophosphatidyl compounds, as such, are not necessary as “acceptors” for mitochondrial fatty acid synthesis. Biochim. Biophys.

Acta,

IJZ (1968) 50-62

5.5

FATTY ACID SYNTHESIS IN MITOCHONDRIA

15

.O

-Time

30

45

60

(min)

Fig. I. Effect of a second addition of sarcosomes on pyruvate incorporation by rat heart sarcosomes. Incubation medium as described under METHODS, to which 3 pmoles [z-%]pyruvate (I ymole = I PC) were added. Incubations were at 25'. The experiment was run in a series of parallel incubation flasks which were stopped after 15, 30, 45 and 60 min respectively. After 30 min, another I mg sarcosomes was added to one of the reaction flasks which was stopped after 60 min. The contents of the flasks were saponified, extracted and assayed for radioactivity as described under METHODS. Fig. 2. Effect of addition of tetradecanoic acid on pyruvate incorporation by rat heart sarcosomes. Incubation medium and incubation procedures as described for Fig. I. After 30 min, 60 pg tetradecanoic acid were added instead of sarcosomes.

TABLE

II

EFFECT OF FATTY ACIDS AND PHOSPHATIDYL COMPOUNDS ON [z-W]PYRUVATE INTO LONG-CHAIN FATTY ACIDS BY RAT HEART SARCOSOMES

INCORPORATION

Incubation medium as described under METHODS, to which 3 pmoles [2-Wlpyruvate (I pmole = I &) were added. The fatty acids were dissolved in ethanol before being added to the incubation media. Pyruvate incorporation was not influenced by the presence of ethanol (1% final concentration). Incubations were at 25’ for 30 min.

Addition None Butyric acid Hexanoic acid Octanoic acid Dodecanoic acid Tetradecanoic acid Lysophosphatidylcholine Lysophosphatidylethanolamine

(pmole)

14C incorporated into fatty acids (mpmoleslmg @otein per I5 mix)

0.6 0.6 0.6 0.6 0.6 0.1 mg 0.1 mg

0.12 0.28 0.58 0.66 0.25 0.24 0.16 0.r4

Amount

Distribution of incorporated radioactivity The distribution of the radioactivity derived from [z-Wlpyruvate among the endogenous fatty acids from rat heart sarcosomes is given in Table III. Pyruvate was incorporated into saturated, as well as unsaturated, fatty acids, as was demonstrated by H~~LSMANN~ for acetate incorporation. The mitochondrial system synthesizes chiefly phospholipid9, as can also be seen in Table III. In view of the stimulatory action of short-chain fatty acids on the [z-K]pyruvate incorporation into higher fatty acids (Table II), a more detailed analysis Biochim.

Biophys.

Acta,

152 (1968) 50-62

56

E. J. V. J. CHRIST

TABLE

III

INCORPORATION INTO

LIPIDS

OF By

RADICACTIVITY

RAT

HEART

FROM

[z-14C]~~~~~~~~

INTO

LONG-CHAIN

FATTY

ACIDS

AND

SARCOSOMES

[a-i4C]Pyruvate was incorporated into long-chain fatty acids in the incubation medium described under METHODS, to which 3 pmoles pyruvate (I pmole = I ,uC) were added. The fatty acids of several experiments were pooled (about 70000 counts/mm), and separated by gas chromatography as described under METHODS. The hydroxy acids were separated from the other tatty acids by thinlayer chromatography on silica gel G plates with light petroleum-ether (70:30. v/v) as solvent. The isolation and separation of mitochondrial lipids was carried out as described under METHODS. In this case, the data represent the average of 4 experiments. Incubations were at ~5~ for 15 min. Total radioactivit?, present in ,fraction Fatty acids C,, : o and lower Ci, : 0

(“/)

trace

17

6 8 I2

CM : o* Cl, : 0 + cm : 0 C,, : 1 Ci, : 2 Ci, : 3 C,, : II cxl : 21cm : 3


7 5 6 20

C,,:4

C,, and C,, saturated and unsaturated Hydroxy acids probably C,, and C,, Lipid fraction l’hospholipid Mono-diglyceride Free fatty acids Neutral lipid Unsaponifiable lipids

Amount (w/w) of total acid recovered I”,&)


‘4 S

I5 4 65 20

15
______-

* Not separated from C,, unsaturated.

of the distribution

of radioactivity

among the carbon atoms of the labelled products

was undertaken. Rat palmitic

heart

sarcosomes

and myristic

were incubated

with

acids isolated and subjected

[G4C]pyruvate

and the labelled

to a Murray oxidation

with KMnO,.

The resulting series of lower homologues was separated as described under METHODS. The data in Table IV show that most of the radioactivity originally present in the palmitic acid is recovered in the carboxyl carbon atom. In the first six carbon atoms of myristic acid, counted from the methyl end, no significant radioactivity was detected. From these data (see also Table II) it is clear that under these experimental conditions only chain-elongation reactions occur. In view of the results presented in Tables II and IV it was of interest to detercan also be effectively elongated to mine whether short-chain fatty acids (C,C,) long-chain fatty acids without prior cleavage to acetyl units. [r-14C]Butyric, [I-%]hexanoic, [I-Xloctanoic and [I-l%]tetradecanoic acids were incubated with rat heart sarcosomes and either acetyl-CoA or malonyl-CoA. The fatty acids, synthesized from the different precursors, were fractionated and the extent of incorporation of the precursor. into these fractions was calculated. It was found (Table V) that the greater part of the 14C of all substrates is incorporated into the corresponding fatty acid with two more carbon atoms, regardless of whether acetyl-CoA or malonyl-CoA was used as C, donor. Thus it can be expected that, most probably, not more than two or three C, units are added to endogenous fatty acids present in the preparation. Furthermore, Biochim

Biophys. Acta, 152 (1968) 50-62

57

FATTY ACID SYNTHESIS IN MITOCHONDRIA TABLE

IV

ANALYSIS 0F THE PRODUCTS FROM A K&O, OXIDATION OF HEART SARCOSOMES ON INCUBATION WITH [Z-~~~]PYRUVATE

PALMITIC

ACID

LABELLED

BY

RAT

[z-W]Pyruvate was incorporated into long-chain fatty acids in the incubation medium described under METHODS, to which 3 pmoles [G4C]pyruvate (I pmole = r &) were added. Incubations were at 25O for 30 min. The fatty acids of several experiments were pooled (70000 counts/min) and separated by gas chromatography. The isolated labelled palmitic acid was oxidized with KiYInO, and the resulting lower homologues were isolated, separated and analysed for radioactivity as described under METHODS. ______

-~~

Number of C atoms

qb Radioactivity of C,,

of fatty acid

Found

recovered

Theoretically if unifovml_y labelled

~~__

12 13 14 15 16 ~~

_

75 75 87.5 87.5 IO0

15 6 28 39 100

the proportion of 14Cin the carboxyl group of octanoic acid synthesized with [I-%]butyric acid and [r-14C]acetyl-CoA showed that successive condensation of [1-14C]acetyl units with the precursor had occurred. The same results were found with dodecanoic acid synthesized with [I-%]hexanoic acid and [r-14C]acetyl-CoA. It might be argued that the elongation systems described for mitochondria and microsomes27-28 have many characteristics in common. There are, however, also important differences (see DISCUSSION). Furthermore, it can easily be demonstrated that addition of microsomes to a mitochondrial preparation does not stimulate fatty acid synthesis when using [G4C]pyruvate as a substrate but, in fact, that it slightly inhibits the synthesis, possibly by the introduction of a microsomalATPase(TableV1). Microsomal contamination as a source of a fatty-acid elongation system, in the mito chondrial preparations used, therefore seems unimportant. Synthesis of unsaturated fatty acids

As can be seen from Table III, incubation of [G4C]pyruvate with rat heart sarcosomes leads to the appearance of radioactivity in mono- and polyunsaturated TABLE

V

ELONGATION CARBON

OF

SHORT-CHAIN

AND

LONG-CHAIN

FATTY

ACIDS

INTO

FAlTY

ACIDS

WITH

AT

LEAST

TWO

MORE

ATOMS

Incubation medium as described under METHODS, to which 2 ,~moles r-14C-labelled butyric, hexanoic, octanoic or tetradecanoic acids as sodium salts and 150 mpmoles acetyl-CoA or malonyl-CoA and ro pmoles KCN were added. The [r-i4C]tetradecanoic acid was added as serum albumin complex and sonicated before use.Incubations were at 25’ for 30 min. The labelled fatty acids from several experiments (about 30 mg protein) were pooled, isolated and separated as under METHODS. Substrates

O$ of total incorporation 6:o

[G4C]Butyric acid plus acetyl-CoA [G4C]Butyric acid plus malonyl-CoA [r-‘4C]Hexanoic acid plus malonyl-CoA [r-%]Octanoic acid $Zus malonyl-CoA [r-i4C]Tetradecanoic acid plus acetyl-CoA *

Total incorporation

73 75 -

8to-Kb 18 r9 73 -

into

12:o 5

fatty I/f:0

3

I
1: 2

: 2

expressed as mpmoles substrate incorporated

3 ??

acids presetit ilz I610

G

Total*

r8:0+2oTc,

0.50 0.46 0.54 49

0.2.5

per mg protein per 30 min.

Biochim. Biophys. Acta, 152 (1968) 50-62

58

E. J. V. J. CHRIST

fatty acids. Although the ratio of the specific activities of the monoenoic and dienoic acids (not shown) is not inconsistent with the introduction of a second double bond in the already labelled monoenoic acid, there is evidence that in mitochondria direct desaturation of saturated acids does not occur B~lo.However, more recently, the formation of 2- and g-monoenoic acids from palmityl-CoA in mitochondria was shown by DAVIDOFF ANDKoRN~~. TABLE EFFECT

VI OF ADDITION

OF MICROSOMES

ON [Z-"~]PYRUVATEINCORPORATION

BY RAT HEART

SARCO-

SOMES

Incubation medium as described under METHODS for fatty acid synthesis to which 3 /imoles [z-r*C]pyruvate (I: ,umole = I PC) and 400 m~moles dodecanoyl-CoA were added. Microsomes were isolated as the fraction from a rat heart homogenate between centrifugation at IZ 500 xg for ro min and rooooo x g for 45 min. The precipitate was washed once with sucrose-EDTA. incubations were at 25” for 15 min. Additions None 0.4 mg microsomes r.4 mg microsomes

nz,umoles 14C incorporated into fatty acids per nzg sarcasamal @rote& per 15 win I.38 I.37 I.**

per mg sarcosomes per mg sarcosomes

To obtain a better understanding of the desaturation reactions occurring in mitochondria, the unsaturated acids Iabelled with 1% from [2-14C]pyruvate by rat heart sarcosomes were isolated and subjected to ozonolysis. As can be seen from Table IV, the palmitic acid contains the bulk of its radioa~ti~ty in the carboxyl group. It seems therefore reasonable to assume that C,, and C,, acids are also labelled predominantly in the carboxyl carbon atoms. Consequently, only the first double bond in the labelled unsaturated acids can be located. To demonstrate the formation of 2- as well as 3-monoenoic acids, [IJ*C]palmitic acid was incubated with frozen and thawed sarcosomes under aerobic conditions. The monoenoic and 3-hydroxy acids were isolated with the use of carrier acids* esterified and separated by thin-layer chromatography. The conversion of palmitate to transTABLE

VII

THIN-LAYBRCHR~MATOGRAFHICANALYSIS OF~Y~~S-CL,B-UNSATURATED,~~'~~S-B,~-UNSATURATED @-HYDROXY FATTYAC~=SPR~D~CEDB~RATHEART SARCOSOMESFR~~ jr-%]PALMITATE

AND

[I-i*C]Palmitate (0.5 &mole = 0.1 PC) was incubated in an incubation medium as described under METHODS to which 0.5 mg serum albumin was added, but without added NADP+. Rat heart sarcosomes were frozen and thawed before use. The incubations were at 30’ for 30 min. The contents of several reaction flasks (about qooooo counts/mm) were pooled and, after addition of carrier acids, saponified, acidified, and extracted 3 times with ether. The fatty acids were esterified with diazomethane and separated by thin-layer chromatography as described by DAVIDOFF AND KoRN~". All acids were isolated from the plates and re-chromatographed in the same system to avoid contamination by other fractions.The hydroxy acids were separated from the other acids by thin-layer chromatography on silica gel G plates with light petroleum-ether (70 :30, V/V) as solvent. --Products (counts/min) T~caxs-B,y-zansaturated acids fl-Hydroxy acids Tvalzs-cc,P-unsatzcrated acids --. 12400 10 IO0 +?OO * Chemically prepared by Dr. D.H.

NUGTEREN.

Biochim. Biophys. Acta, 152 (1968) 50-62

FATTY ACID SYNTHESIS IN MITOCHONDRIA

59

c$-unsaturated, trans-p,y-unsaturated, and p-hydroxy acids is shown in Table VII. The isolation of cis$,y-unsaturated acids was not attempted. In Table VIII a gas-chromatographic analysis of the products of oxidative cleavage of radioactive di-, and tetraenoic acids, as identified from their gas-liquid and thin-layer chromatographic behaviour, is given. Since the rather volatile C, ester was not collected, the distribution of the radioactivity among the C,-C,, fragments is shown. The greater part of the total radioactivity eluted was found in the C,-dicarboxylic acid. The formation of the 3-monoenoic acids can be explained by the results of DAVIDOFF AND KoRN~O. The radioactivity in the Cg-dicarboxylic acid fraction could have been introduced by an exchange reaction of the carboxyl C, unit, as outlined by BARRON~~for C,,-monoenoic acid. The amount present in this fragment seems rather variable as is shown in Table VIII, Expts. I and z. TABLE VIII GAS-LIQUID

AND

SATURATED

FATTY

THIN-LAYER ACIDS

CHROMATOGRAPHIC

LABELLED

WITH

ANALYSIS

‘*c

FROM

[Z-‘*cl

OF

PRODUCTS

PYRUVATE

BY

OF RAT

OZONOLYSIS HEART

OF

UN-

SARCOSOMES

Incubation medium as described under METHODS, to which 3 pmoles [z-‘Klpyruvate (I pmole = I ,uC) were added. The incubations were at 25’ for 30 min. The labelled unsaturated acids, identified by their chromatographic behaviour, were isolated and purified as described under METHODS. After ozonolysis the resulting fragments were either converted to dinitrophenylhydrazones or oxidized to the corresponding carboxylic acids. The dinitrophenylhydrazones were separated by thin-layer chromatography; the carboxylic acids were esterified with diazomethane and separated by gas chromatography. Expt. 2 : distribution of radioactivity of dinitrophenylhydrazones. GLC = gas-liquid chromatography; TLC = thin-layer chromatography. Fragment

% Radioactivity

collected

18:2

GLC Expt.

in sample

oxidized 20:4

I

TLC zvpt.

GLC 2

C3

73

80

C, CS

IO

0

63 3 4

5

0

C6

0

0

I

C,

2

2

IO

5 =I

C8

2

0

C, C 10 C 11

5 I 2

16 0

2

-

-.__

The analysis of the products of ozonolysis of the Cl,-monoenoic acid (results not shown) revealed that most of the radioactivity was present in the C, and C,, fragments, in agreement with the results of other investigationssIlo. A smaller fraction of the radioactivity (about 15%) was found in the C, fragment. Indications that the 3-monoenoic acids are actually formed from a, p-unsaturated acids by the action of the isomerase rather than by direct desaturation, as was also demonstrated by DAVIDOFF AND KORN~O,are presented in Table IX. By comparing Expts. I and 2, it is evident that the formation of 3-monoenoic acid is substantially reduced when formation of cc&unsaturated acid by /3 oxidation is inhibited. Under these conditions, a&unsaturated C,,-acid is mainly formed during chain elongation. Controls run with the stock solutions of the r-14C-labelled acids indicated that the experimental error of the procedure is about 0.05%. Thus, it appears doubtful Biochim.

Biophys.

Acta,

152 (1968) 50-62

60

E. J. V. J. CHRIST

whether there is any desaturation at all under the conditions of Expt. 2, Table IX. The fact that, in both experiments, some radioactivity (about 0.5% of total radioactivity added) was also recovered in the fatty acid fractions with chain lengths shorter, as well as longer, than that of the precursor used, suggests that, even in the presence of IO-~ M KCN, some cleavage of labelled precursor occurred. This residual cleavage probably accounts for the radioactivity found in the 9- and IImonoenoic acids. TABLE FORMATION

IX ~F~,~-~cTA~ECEN~ATE

[I-Wjpazhlrrxc

FROM

AND

jr-'%]STEARIC

ACID

Incubation medium as described under METHODS. In both experiments I mg bovine serum albumin and ro @moles KCN were added. In Expt. r, r ymole [I-Wjpalmitic acid as sodium salt (I fxmole = 0.2 PC) pEzts roe m,umoles malonyl-CoA were used as substrate; in Expt. z, I pmoie [G4Cjstearic acid as sodium salt (I ,umole = 0.2 $.Y) was used. The fatty acid suspensions were sonicated before use. Incubations were at 25’ for 30 min. The contents of several reaction flasks were pooled. The products of ozonolysis of the C,, : 1 acids were assayed for radioactivity as described under METHODS. ____

Expt.

_-._

._ _-

--.

2

_~..._

Fractio+z Countslmin isolated

[I-iX]Palmitic acid +Eus malonyl-CoA

18:0 18:1

NO.

I

__

Precz~vsor added

190 740 I520

18:~

2224

18:3

690

rS:o

9.98‘10"

jr-‘*C:Stearic acid

18:r

--.--_--

18:3 _ ----

18:2

% Total radioactivity recovered in fractions isolated 97.7 0,s

~-MOWenoic acid present in C,, : 1

y; o- and IIMonoenoic acid present

-

-

2’,

~_____ 9

I.1 o-4 99~9 0.04

4440 760


980


in c,, : 1 78

-

-

-

-

-

--

.---~-

DISCUSSION

Recently, the formation of malonyl-CoA from acetyl-CoA by soluble preparations from sonicated rabbit heart sarcosomes was reported by H~~LSMANN’~. The indications were that propionylCob carboxylase rather than acetyl-CoA carboxylase was responsible for the carboxylation of acetyl-CoA in heart sarcosomes. At present, there seems to be no conclusive experimental evidence in support of a role for malonylCoA as the condensing carbon unit in mitochondrial fatty acid synthesis. Addition of HCO,- does not stimulate acetyl-CoA incorporation into mitochondrial systems (refs. 4, 5, IO), whereas, in microsomal elongation reactions, acetyl-CoA cannot serve as a substitute for malonyl-CoA unless HCO,- is present27p28.In addition, mitochondrial fatty acid synthesis showed more dependence on NADH8+‘, in contrast to the microsomal system where a clear preference for NADPH was notedz8. The observation that, in short-time experiments under anaerobic conditions favouring synthetic reactions, acetyl-CoA is a better substrate than malonyl-CoA, suggests that the compound actually incorporated is acetyl-CoA and not malonyl-CoA. Since in the experiments reported in Table V the incubation time was 30 min, sufficient acetyl-CoA will be formed by decarboxylation of malonyl-CoA, when used in the incubation medium. The malonyl-CoA decarboxylase %I, found in mitochondria, might possibly prevent the accumulation of malonyl-CoA formed by the action of propionyl-CoA carboxylase. Biochim. Biophys. Acta, 152

(1968)

50-62

FATTY ACID SYNTHESIS

61

IN MITOCHO~DRIA

From the results reported in this investigation it appears that de nova fatty acid synthesis in mitochondria is possible only to a very limited extent. When palmitic and myristic acids labelled with l*C from [z-‘*C]pyruvate by rat heart sarcosomes were degraded with KMnO, no significant radioactivity could be detected in the first six carbon atoms counted from the methyl end. When [I-%]butyric acid was incubated with sarcosomes and either acetyl-CoA or malonyl-CoA, only very little radioactivity could be detected in fatty acids with 14 or more carbon atoms (Table V). It is evident that the mitochondrial system in which each of the fatty acids elongated with an acetyl unit dissociates from the enzyme, can contribute to a de novo synthesis only to a very limited extent. It should be remembered that the incorporation of a butyryl unit as “primer” into palmitate by a “soluble” liver system32 is considered a de nova synthesis. Although a de novo synthesis in mitochondrial systems has been reported it is difficult to assess the degree of contamination with cytoplasmic repeatedly8t10+, de nova synthesizing enzymes in the mitochondrial preparations employed. Mammary gland mitochondria-measuring fatty acid synthesis with malonyl-CoA as substratelose already a great deal of their fatty acid synthesizing activity by simply washing the mitochondria in sucrose-EDTAa3. Addition of avidin to liver mitochondria induced a substantial reduction of mitochondrial de novo fatty acid synthesis, in spite of the fact that it seems doubtful whether avidin can penetrate into intact mitochondrialO. Total fatty acid synthesis was hardly affected. It seems possible that, by several successive additions of labelled acetyl units to endogenous fatty acids, a ratio of carboxyl carbon radioactivity to total carbon radioactivity may be obtained which is intermediate between a value for de l~ovo synthesis and a single C, addition, as observed by otherssP. The results of the investigation described above on the nature of the double bonds actually introduced indicate that the significance of mitochondrial elongation reactions is not comparable to the significance of fatty acid synthesis in microsomes, where elongation and desaturation reactions lead to the synthesis of the polyenoic acids27y28 which are necessary, e.g., as structural elements in active membranes. The fact that formation of 3-monoenoic acid is depressed when ,3 oxidation is inhibited, might be best explained by assuming that 3-monoenoic acid can only be formed by isomerization of the oc,/?isomers and not by direct desaturation of the longchain fatty acids. ACKNO’VYLEDGEMENT

The help of Mr. H. VAN TILB~RG and the technical

assistance

with

the gas-chromatographic

of Miss J. STEENHORST

are gratefully

experiments

acknowledged.

REFERENCES I 2 3 4 5 6 7

F. II)ITURI, W. N. SHAW, J. V. B. Whms AND S. GURIN, J. Bid. Chew, 226 (1957) 407. W. C. HDLSMANN, Biochim. Biophys. Acta, 45 (1960) 623. W. C. H~LSMANN, Biochim. Biophys. Acta, 58 (1962) 417. S. J. WAKIL, L.W.MCLAIN, Jr. AND J B.WARSHAW,J. Bid. Chem., 235 (1960) PC 31. S. J. WAKX, J. LlipidRes., 2 (rg6r) I. E. J. CHRIST AND W. C. HI~LSMANN, Biachim. 3~0~~~s. Acta, 60 (1962) 72. W. R. HARLAN, Jr. AND S. J. WAKIL, Biochem. Biophys. Res. common., 8 (rgtiz) 131. B&him.

Biophys. Acta,

152

(1968) 5062

E. J. V. J. CHKIST

62 8 W. K. HARLAN,

Q Ii II I2 13 I4

P. E. W. K. G. W.

Jr. AND S. J. WAKIL, J. Biol. Chem., 238 (1963)3216. H. HOLLOWAY AND S. 1. WAKIL, 1.Bid. Chem.. 239 (1964) 2489. J. BARRON, Biochim.B;ophys. A&, 116 (1966)425.. .' C. H~LSPIIANN,Biochim. Riophys. Acta, 125 (1966) 397. W. CLELAND AND E. C. SLATER, Biochern. f., 53 (1953) 547. A. HOWARD AND A. J. P. MARTIN, Biochem. J., 46 (1950) 5.~. M. L. CROMBIE, R. COMBER AND S. G. BOATMAN, Biochem. J., 59 (1955)

309. I; Ii. E. MURRAY, Australian J, Chem., 12 (1959) 657. 16 G. B. CORCORAN, Anal. Chew, 28 (1956) 168. 17 E3. DE VRIES AND G. JORRIENS, F&e, S&&q ~4~st~~~h~~~tte~,6.~ (1963) 725. 18 E. H. PRYDE. D. E. ANDERS.H.M.TEETERAND J. C.Cowa~,j. Org. Chem., 27 (1962) Ig H.GO~DFINEAND K.BI.ocH,J. Biol.Chem., 236 (1961) 2596. 20 T. WIELANDAND H. KAPPE, Ann.Chem., 581 (19.53) I. 21 E. G.TRA~XS AND R. 0. BRADY, J. Am.Cbm. So;., 82 (1960)2972. 22 I?.GOLDMAN AND P. K. VAGELOS, J.B~~~.Che~~., 236 (1961) 2620. 23 N. H. TATTRIE, J. Lipid Res., I (1959) 60. 24 W. L. MAGEE, J, GALLAI-HATCHARD, H. SANDERS AND R. H. S. THOMPSON,R~OC~~M~. 25

(1962) ‘7. A. G. CORNALL, C. J. BARDAWILL

3055.

J., 83

AND M.M. DAVID,]. Biol. Chem., 177 (x949) 751. 26 E.E.JA~oB~,M.JA~oB,D.R.SAN~~~~NI, L.B.BRADLEY,J.R~o~. Chem., 223(1956) 147. 27 W. STOFFEL, Biachem. Biophys. Res. Commute., 6 (r961) 270. 28 D. H. NUGTEREN, B&him. Biophys. Acta, 106 (1965) 280. 29 W. STOFFEL AND K. L. AcH,Z. PhysioZ.Chcm., 337 (1964) 123. 30 F. DAVIDOFP AND E. D. KoRN,~. Biol. Chem., 240(1965) 1549. 31 H. I.NAKADA, J. B. WOLFE AND A. N. WICK, J..BioE. Chem., 226 (1957) 145. 32 R, W. LONG AND J. W. PORTER, J.Bios. Chem., 234(x954) 1406. 33 W. C. H~LSMANN AND D. S. Dow, Bioclzim.Biophys. Acta, 84 (1964) 486. 34 W. BARTLEY, G. S. GETZ, B. M.NOTTON AND ‘5.RENSHAW, Riochem. J,. 82 (1962) 540. Biochim.

Biophys.

Acta,

I$! (rg68) 50-62