the conversion of long chain saturated fatty acids to ... · pdf filethe conversion of long...

THE JOURNAL OF BIOIIOGICA~ CHEMISTRY Vol. 239, No. 8, August 1964

Prinled in U.S.A.

The Conversion of Long Chain Saturated Fatty Acids to Their

@Unsaturated, ,&y-Unsaturated, and @-Hydroxy Derivatives by Enzymes from the Cellular Slime Mold,

Dictyostelium discoideum

FRANK DAVIDOFF AND EDWARD D. KORN

From the Laboratory of Biochemistry, Section on Cellular Physiology, National Heart Institute, National Institutes of Health, Bethesda 14, Maryland

(Received for publication, January 29, 1964)

A major problem in the study of fatty acid metabolism at the subcellular level has been the demonstration of the individual steps of both oxidation and synthesis (1, 2). Oxidation of long and short chain fatty acids by intact mitochondria is considered to proceed directly to acetate, without the accumulation of intermediates (3). Synthesis of fatty acids from acetyl coenzyme A and malonyl-CoA by the soluble “synthetase complex” like- wise occurs as a concerted process, and significant amounts of intermediates have not been detected (4). Only in the incom- pletely defined synthetic system from mammary gland have such intermediates been directly demonstrated, and these are limited to compounds of chain length Cq to Ca (5).

Each of the enzymes of /3 oxidation has been highly purified and studied in detail as the catalyst of a single step (6). Most of these studies have been carried out with fatty acids of chain length Cl2 or less, and a number of them have relied on spectrophotometric and enzymatic assays for characterization of the products (7-Q). The structures of the long chain fatty acid intermediates have generally been inferred from the identification of the Cq, Ce, and Cs intermediates.

In the biosynthetic system, a number of separate intermediate steps have also been studied, using chemical, spectrophotometric, and exchange tracer techniques (2, 10). As in /3 oxidation, only the short chain fatty acid intermediates have been isolated and characterized. Using substrates of Ce chain length, Brady and his coworkers (11, 12) have shown the participation of synthetic c~,@unsaturated, P-keto, and D( -)-@-hydroxyacyl-CoA derivatives in the biosynthesis of fatty acid chains (11, 12).

The inability to detect long chain fatty acid intermediates in these intact systems may have several causes. Until the advent of gas-liquid chromatography and thin layer chromatography, these compounds were difficult to isolate and characterize. Also, there is evidence that, at least in some systems, the intermediates may exist only bound to enzyme sulfhydryls as thiol esters (2), and, therefore, never would accumulate unless the enzymes were used in substrate concentrations. Such a phenomenon could also explain the difficulty of demonstrating the entry of synthetic intermediates into the biosynthesis of fatty acids (4).

Whatever the explanation, the long chain fatty acid derivatives presumed to be the intermediates in fatty acid oxidation a,nd synthesis appear never to have been isolated and charac-

terized. It is therefore of interest that enzymes from the cellular slime mold Dictyostelium discoideum have been found which transform saturated long chain fatty acyl-CoA derivatives to a series of long chain intermediates, each of which accumulates in considerable quantity. These compounds include long chain trans-a,P-unsaturated fatty acids which are the postulated intermediates in fatty acid oxidation and synthesis, long chain D( -)-/3-hydroxy acids which are the presumed intermediates in fatty acid synthesis, but not in p oxidation, and long chain cis- and trans-p,y-unsaturated fatty acids for which a metabolic role is not clear. The same compounds are formed during the incubation of palmityl-CoA with the mitochondrial fraction of guinea pig liver,1 and, therefore, the reactions described in this paper are not unique to the cellular slime mold.

EXPERIMENTAL PROCEDURE

Materials-Radioactive fatty acids were obtained commer- cially. The l-14C-palmitic acid used in several experiments was contaminated with 1 .S% of fatty acids having the same retention time as palmitoleate on gas-liquid chromatography, but giving rise on oxidative cleavage (13) to approximately equal quantities of radioactive dicarboxylic acids of chain lengths Cs to Clz. Appropriate corrections were made for the presence of these contaminants. Subsequent batches of 1-14C-palmitate were hydrogenated before use. 16-14C-Palmitic acid was radiochemi- tally pure, as determined by gas-liquid chromatography. 2- Pentadecanone and lauric, myristic, and palmitic aldehydes were obtained from K and K Laboratories. CoA, NAD, and ATP were obtained from Pabst Laboratories.

Synthesis of @-Hydroxy Acids@Hydroxymyristate, /?-hydroxypalmitate and ,8-hydroxystearate were synthesized by the Reformatsky procedure (14). Each product was purified either by crystallization of the free acid from n-heptane, or chromatography of the methyl ester on silicic acid. Analysis of each compound by gas-liquid chromatography revealed only one peak. Synthetic @-hydroxypalmitate was optically inactive; the optical rotations of the other hydroxy acids were not determined. The ,&hydroxy acids were further characterized by their infrared spectra, and by identification of the unsaturated

1 F. Davidoff and E. D. Korn, unpublished results.

2496

by guest on May 8, 2018

http://ww

w.jbc.org/

Dow

nloaded from

http://www.jbc.org/

August 1964 F. Davidof and E. D. Korn 2497

acids formed by dehydration. Methyl esters of the P-hydroxy acids were saponified in 0.5 N NaOH in 50% methanol for 2 hours at 85”. The P-hydroxy acids were stable under these conditions.

Synthesis of Unsaturated Acids-A mixture of o( , p- and p, y- hexadecenoic acids was prepared from @-hydroxypalmitic acid by reaction with p-toluene sulfonyl chloride in pyridine, followed by treatment with potassium t-butoxide. The products were methylated, and separated by preparative gas-liquid chromatography. Alternatively, more efficient separation of larger quantities of a!, p- and pi y-unsaturated fatty acids was achieved by chromatography, as methyl esters, on a column of silicic acid impregnated with silver nitrate, according to the technique of de Vries (15). The cis and trans isomers of the /3, y-hexadecenoate fraction were separated by rechromatography on a smaller silicic acid-silver nitrate column. The final preparation of trans-0, y-hexadecenoic acid contained 1.5 y0 of the cis isomer, and the cis preparation contained 0.6% of the trans isomer. The proportions of each isomer were determined by thin layer chromatography of the purified compounds, as methyl esters, on silica gel G impregnated with silver nitrate, a technique that separates cis and trans isomers (16). The spots corresponding to each isomer were eluted with diethyl ether, and the fatty acid determined quantitatively by analytical gas-liquid chromato-

graphy. Dehydration of P-hydroxypalmitate produced about equal

amounts of cis- and bans-p, y-hexadecenoate, but only the trans-a,/3-isomer could be detected among the products. The synthetic CY, fl- and ,Q, y-unsaturated compounds were identified by their infrared spectra, specifically the absorption of the trans- P, y-hexadecenoic acid at 965 cm-l, and the absorption of trans- m,/%hexadecenolc acid at 985 cm+ (17). Further confirmation of structure was achieved by oxidative cleavage of the unsaturated fatty acids by the Mn04--104- procedure of von Rudloff (13), and identification of the resultant aliphatic monocarboxylic acids by gas-liquid chromatography. Myristic acid was formed from cr ,/3-hexadecenoic acid, and tridecanoic acid from ,@,r- hexadecenoic acid. Hydrogenation of the synthetic unsaturated fatty acids in methanol, with PtOs as catalyst, gave rise only to palmitic acid.

Free c&s- and trans-/3, y-unsaturated acids were formed from their methyl esters by saponification in 0.5 N NaOH in 50% methanol for 2 hours at 85”. However, under these conditions, the oc,&unsaturated fatty acid underwent addition to form the P-methyl ether, and was partially isomerized to the P,y-unsaturated acid. Hence, methyl-a,/3-hexadecenoate was saponified in 0.5 N NaOH in a mixture of tert-butyl alcohol and water.

Synthesis of Acyl-CoA Derivatives-Before synthesizing the CoA derivatives, the purity of each fatty acid was checked by gas-liquid chromatography. Fatty acyl-CoA derivatives were prepared according to the method of Goldman and Vagelos (18). The reaction between CoA and mixed anhydride was continued until the nitroprusside test for free -SH groups was negative (19). Unreacted fatty acid was extracted with ether from the acidified aqueous reaction mixture. Yields were calculated, in the case of the labeled fatty acids, from the radioactivity not extractable with ether. For unlabeled fatty acids, the yield of acyl-CoA was quantitated by hydrolyzing an aliquot, converting it to its methyl ester, and quantitating by gas-liquid chromatography. Yields of acyl-CoA varied from 18% of theoretical for a,@hexadecenoyl-CoA to 80 to 90yG for saturated, and

,8,y-unsaturated, fatty acyl-CoA’s. Aliquots of each acyl-CoA preparation were analyzed for purity by gas-liquid chromatography after hydrolysis and conversion to methyl esters. The (Y , /?-hexadecenoyl-CoA contained 3 y0 /3, y-hexadecenoyl-CoA and 1.9% ,&hydroxypalmityl-CoA; ,&hydroxypalmityl-CoA contained 2.6 ‘% a!, P-hexadecenoyl-CoA and no ,0, y-hexadecenoyl-CoA; trans- ,and cis-p, y-hexadecenoyl-CoA contained 1.1 y0 and 1.9%, respectively, of cr,&hexadecenoyl-CoA and no /3-hydroxypalmityl-CoA. Corrections were made for these components in calculating the results.

Preparation of EnzymesStock cultures of D. discoideum, ag- gregateless mutant Agg 204, were maintained on agar. Sub- mersion cultures were grown on lipid-extracted Escherichia coli as described previously, except that streptomycin, 75 mg per liter, was used instead of penicillin (20). Amebas were grown in submersion culture until nearly all bacteria were consumed; the yield was 7 to 10 g of amebas per liter of culture. The cells were washed in 0.04 M phosphate buffer, pH 6.0, by centrifugation until free of residual bacteria. They were then suspended in 0.25 M sucrose at O”, and packed by centrifuging at 3000 r.p.m. for 10 minutes. The amebas were ground with 2 volumes of Superbrite glass beads for 5 minutes with a chilled mortar and pestle. Breakage of cells was virtually quantitative, as determined by phase contrast microscopy. The homogenate was washed from the beads with a total of 4 volumes of ice-cold 0.25 M sucrose, and the remaining beads and intact cells were removed by centrifugation at low speed.

The heavier subcellular particles were then separated by centrifugation at 12,000 X g for 15 minutes. The supernatant was centrifuged at 105,000 x g for 45 minutes to sediment lighter particles. The resultant pellets of heavy and light particles were resuspended in 0.25 M sucrose by gentle homogenization with a ground glass homogenizer. These two fractions are re- ferred to as “heavy particle” and “light particle” fractions since the mitochondria of D. discoideum are of unusual shape (21), and their sedimentation properties are not known. Particle suspensions were diluted to a known optical density at 600 mp before use.

Subcellular particulate fractions prepared in either 0.1 M

phosphate buffer, pH 7.25, or 0.25 M sucrose, appeared to have the same enzymatic activities toward long chain fatty acyl-CoA substrates. Activity was unchanged for at least several weeks when particles were kept frozen at -20” in sucrose or phosphate buffer.

For preparation of acetone powders, the pellets of light and heavy particles were resuspended in a small volume of 0.25 M sucrose or 0.1 M phosphate buffer, pH 7.25, and then extracted with 10 volumes of acetone, according to the technique of Drys- dale and Lardy (22). The powder was washed once with cold acetone, then with ether, and dried at 0”. Activity of the acetone powder appeared to be stable for at least 5 months.

Soluble enzymes were extracted from acetone powders by suspending 50 mg of acetone powder in 3 ml of 0.1 M phosphate buffer, pH 7.25, and stirring for 15 minutes at 0”. The suspen- sion was then centrifuged at 25,000 X g, and the supernatant removed. The soluble enzymes were stable for several weeks when kept frozen. They were used at a known concentration of protein as determined by the technique of Lowry et al. (23).

Incubations and Analysis of Products--Incubations were carried out aerobically in test tubes. The reaction was stopped by addition of an equal volume of 1 N NaOH. In experiments


http://ww

w.jbc.org/

Dow

nloaded from

http://www.jbc.org/

2498 Biosynthesis of Fatty Acid Derivatives Vol. 239, No. 8

with intact subcellular particles, the samples were then saponified for 1 hour at 70”. Following incubations with soluble enzymes, hydrolysis was performed at room temperature for 15 minutes (7). The mixture was then acidified with 10 N HzS04, and extracted three times with diethyl ether. Ether was evapo- rated under an air stream, and the free acids converted to the methyl esters using BFs-methanol reagent (24), with heating to 100” for 2 to 4 minutes. Neither isomerization of unsaturated acids nor dehydration of hydroxy acids occurred under these conditions.

Analysis of the reaction products from incubations with nonradioactive fatty acyl-CoA derivatives was carried out directly by gas-liquid chromatography on ethylene glycol succinate. The area under each peak was quantitated by triangulation. Detector response was calibrated by analysis of known quantities of fatty acid methyl esters. In each analysis, the concentration of each component was calculated as a percentage of the total fatty acids present in the chromatogram. The absolute quantity of each product was then calculated from these per- centages and the known amount of substrate added to the original incubation mixture. The soluble enzyme preparations did not contribute significant amounts of fatty acids.

When radioactive saturated acyl-CoA derivatives were used as substrates, the products of reaction were first separated as mercuric acetate adducts by chromatography on silicic acid (25). Carrier methyl esters were added when appropriate. Methyl esters of saturated fatty acids were eluted with 10 ml of benzene; adducts of the unsaturated fatty acids, and the hydroxy fatty acids were eluted together with 2 ml of benzene-5% acetic acid in methanol, 1: 1, followed by 8 ml of 5% acetic acid in methanol. Adducts were then decomposed by addition of 0.5 ml of concentrated HCl directly to the acetic acid-methanol solution. After 1 hour, water was added, and the methyl esters of the unsaturated and hydroxy acids were extracted with petroleum ether.

Direct analysis and quantitation of the radioactive products was accomplished with either the continuous combustion-continuous counting gas-liquid chromatographic technique of Kar- men, McCaffrey, and Bowman (26), or fractional collection of samples as they emerged from a preparative gas-liquid chromatogram onto anthracene coated with silicone oil, followed by direct counting in a scintillation spectrometer (27). For most analyses, ethylene glycol succinate polyester served as the liquid phase for chromatography, but analysis of the radioactive products was also carried out with SE 30 silicone rubber as the liquid phase.

In some experiments, after removal of the methyl esters of the saturated fatty acids by the adducts technique, the radioactive products were quantitated after separation of the components, as methyl esters, by thin layer chromatography on silica gel G impregnated with silver nitrate. This technique provides separation of tram-cu, /?-unsaturated, cis- and bans-0, y-unsaturated, A”unsaturated, and /3-hydroxy long chain fatty acid methyl esters (16). The developing solvent was 5y0 ether in petroleum ether. Spots were detected by spraying with dichlorofluorescein and viewing under ultraviolet light. The silica gel was then either scraped into counting vials containing 0.4% diphenyloxazole in toluene and the radioactivity determined, or the esters were eluted with diethyl ether and an aliquot counted. In

control experiments, recovery of radioactivity from the plates was 80%.

Characterization of ProductsIn order to obtain sufficient ma- terial for complete identification, the fatty acid methyl esters were separated and isolated by preparative gas-liquid chromatography (20) or thin layer chromatography. Unsaturated fatty acid methyl esters were subjected to oxidative cleavage by a modification of the technique of von Rudloff (13). Oxidation products were then identified by gas-liquid chromatography. When it was desired to collect radioactive CO, produced by this reaction, samples of radioactive fatty acids were incubated with the oxidative reagents in flasks fitted with center wells and sealed with serum caps. After completion of the reaction, a methanolic solution of Hyamine was injected into the center well, and the reaction mixture was acidified with 0.2 ml of 10 N H&O,; in this case sodium metabisulfite was not added. The flasks were then incubated for 1 hour at 37”, the Hyamine was transferred to a

scintillation vial, 0.4y0 diphenyloxazole in toluene was added, and the radioactivity determined in a scintillation spectrometer. In control experiments, no radioactive CO* was produced from 1-W-linoleic acid.

/3-Hydroxy acids were degraded by the procedure of van Slyke (28). Fatty acids were heated to 100” in capped, center- well flasks with 1 ml of 1.4 N H&O4 and 0.3 ml of tert-butyl alcohol. Then 0.05 ml of 5% K&r207 was injected into the reaction mixture, and the flasks were shaken at 100” for 1 hour. The flasks were then chilled, methanolic Hyamine was added to the center well, and the CO2 was collected and counted as above. In control experiments with lJ4C-linoleic acid, no radioactive COZ was formed. Oxidations and degradations were performed at least in duplicate.

Hydroxy fatty acids were acetylated with isopropenyl acetate- sulfuric acid, according to the micro technique of Kishimoto and Radin (29). Infrared spectra of the synthetic methyl esters were determined in carbon tetrachloride in a Beckman IR 7 or Per- kin-Elmer 21 spectrophotometer. Spectra of the enzymatically produced fatty acids were obtained with a micro-KBr pellet in the Beckman IR 7. Optical rotations were obtained in ethanol, in a Rudolph polarimeter, with a 0.75-ml cell.

RESULTS

Identification of Products of Palmityl-CoA Metabolism

A gas-liquid chromatogram of the products of the incubation of I-14C-palmityl-CoA with the light particle fraction of the homogenate of D. discoideum is shown in Fig. 1. Essentially identical results were obtained with 16J4C-palmityl-CoA. In both cases, the light particle fraction converted approximately 30 to 45% of the substrate to the three products A, B, and C.2 In a control experiment, only 3 y0 of sodium 1-14C-palmitate was metabolized.

Compound A-Compound A was identified by the following procedures as a mixture of methyl trans-/3, y-hexadecenoate (75%) and methyl cis-fi, y-hexadecenoate (25%). Its retention time in gas-liquid chromatography was very close, but not identical, to that of methyl palmitoleate (Table I). Hydrogenation

2 If, as found by Goldman and Vagelos (18) for acetyl-CoA, the substrates were not 100% enzymatically active palmityl-CoA, then a greater percentage of the available substrate was converted to products.


http://ww

w.jbc.org/

Dow

nloaded from

http://www.jbc.org/

August 1964 F. Davidoff and E. D. Korn 2499

TIME

FIG. 1. Gas-liquid chromatographic analysis of the methyl esters of the nroducts of incubation of 1-r4C-nalmitvl-CoA with the light part‘icle fraction of D. discoideum homogenate. Satu- rated fatty acids were removed prior to analysis as the adducts with mercuric acetate (25). Radioactivity was determined by continuous combustion-continuous counting, as described in “Experimental Procedure;” full scale represents 1000 c.p.m. Mass was analyzed with a hydrogen flame detector. Nonradio-

of Compound A changed its retention time to that of methyl palmitate. Position of the double bond was established by analysis of the products of oxidation with Mn04-104. When Compound A derived from 1-r4C-palmityl-CoA was oxidized, 60% of the radioactivity was recovered as COS. This is consistent with the production of malonic acid by cleavage of the original olefin, since under these conditions malonate is degraded to CO2 (30). When radioactive Compound A produced from 16-14C-palmityl-CoA was isolated and degraded in similar fashion, the only radioactive compound produced was tridecanoic acid. Compound A, therefore, is a 16 carbon fatty acid with a double bond at the 3,4-position.

Synthetic methyl cis- and trans-/3, y-hexadecenoate and Com- pound A all had identical retention times on gas-liquid chromatography (Table I) and, therefore, this technique could not be used to determine stereochemistry. The proportions of ci.s and trans isomers in the enzymatic product were established in several ways. The products formed from incubation of light particles with l-14C-palmityl-CoA were separated by thin layer chromatography on silicic acid impregnated with AgN03 (Table II). The spots corresponding to &an@, y-hexadecenoate and cis-p, y-hexadecenoate were found to contain radioactivity in the ratio of approximately 3 to 1. Also, methyl /3,y-hexadecenoate was isolated by preparative gas-liquid chromatography of the products of incubation of soluble enzyme with nonradioactive nn-P-hydroxypalmityl-CoA. Details of this enzymatic reaction will be described later. The infrared spectrum of the methyl /3,y-hexadecenoate indicated a preponderance of the trans isomer. The two isomers were then separated by thin layer chromatography on AgNO,-silicic acid. The compounds

active fatty acids from D. discoideum were added as carrier. The radioactive peaks were identified (see text) as Palmilate, methyl palmitate; A, methyl cis- and trans-fi,r-hexadecenoate; B, methyl trans.a,@-hexadecenoate; and C, methyl P-hydroxypalmitate. Nonradioactive peaks are the methyl esters of 1, palmitoleate; d, 5,9-hexadecadienoate; 9, oleate and vaccenate; and 4, 5,9-octadecadienoate and 5,11-octadecadienoate.

TABLE I

Gas-liquid chromatographic retention times of substituted fatty acid methyl esters

All of the hydroxy acids and the unsaturated 16-carbon acids were identified both as the synthetic and enaymatically produced compounds; the remaining unsaturated acids were identified only as enzymatic products. Chromatography was done on ethylene glycol succinate.

Compound Retention time relative to methyl stearate

Methyl @,r-tetradecenoate ............... 0.41* Methyl a,p-tetradecenoate ............... 0.57* Methyl palmitoleate. .................... 0.67* Methyl cis- and trans-p, y-hexadecenoate 0.71 Methyl trans-a,b-hexadecenoate .......... 0.93 Methyl @,r-octadecenoate. ............... 1.22 Methyl ol,&octadecenoate. ............... 1.62 Methyl &hydroxymyristate .............. 2.12 Methyl @-hydroxypalmitate .............. 3.61 Methyl fi-hydroxystearate ............... 5.95

* The retention times marked with an asterisk were obtained at 183” and 15 p.s.i.; the remainder were obtained at 193” and 15 p.s.i.

were eluted from the silicic acid, and the amounts of each quantitated by analytical gas-liquid chromatography. Again the ratio of trans to cis was 3 to 1. The products of incubation of soluble enzyme with a!, P-hexadecenoyl-CoA were also separated by thin layer chromatography and trans-P, y-hexadecenoate and c&-/3, y-hexadecenoate were both demonstrated to be present.


http://ww

w.jbc.org/

Dow

nloaded from

http://www.jbc.org/


TABLE II

Comparison of the conversion of myristyl-CoA, palmityl-CoA, and stearyl-CoA to a,&unsaturated, 0, y-unsaturated, and

p-hydroxy fatty acids

Light particle fraction of D. discoideum homogenate at a final dilution of approximately 1.2 mg of protein per ml (optical density, 0.48 at 600 ma) was incubated with each radioactive long chain fatty acyl-CoA, in 0.5 ml of 0.1 M phosphate buffer, pH 7.25. Substrates were lJ4C-stearyl-CoA, 3.5 X lo5 c.p.m. (162 mpmoles); 1-*4C-palmityl-CoA, 4.1 X lo6 c.p.m. (75 msmoles); and 1-‘Gmyristyl-CoA, 4.6 X lo5 c.p.m. (101 mpmoles). After 3 hours of incubation at 31”, fatty acid esters were saponified, and the fatty acids were extracted and esterified, and carrier methyl esters were added. The mercuric acetate adducts were made and separated as described in “Experimental Procedure.” Saturated fatty acids were separated from unsaturated and hydroxy acids on small columns of silicic acid, and an aliquot of each eluate counted. Recovery of radioactivity from the incubation mixture was quantitative. Regenerated methyl esters of the products were then analyzed by thin layer chromatography on silver nitrate-impregnated silica gel. Individual spots were scraped into counting vials, and radioactivity was determined in diphenyloxazole-toluene in a scintillation counter. Recovery of radioactivity from the plates was 65 to 80%.

Substrate

1-IKl-Myristyl-CoA l-W-Palmityl-CoA 1-W-Stearyl-CoA I -

* The remainder of the radioactivity was in the saturated fatty acid (unreacted substrate).

Compound B-Compound B was identified as methyl trans- a!,@-hexadecenoate. Hydrogenation of this compound converted it to methyl palmitate. Upon Mn04--104 oxidation of Compound B enzymatically produced from 1-14C-palmityl-CoA, 78% of the radioactivity was recovered as COZ. This is consistent with the production of oxalic acid from the original unsaturated fatty acid, since oxalate (like malonate) is degraded to COZ under these conditions (31). Oxidative cleavage of Com- pound B produced enzymatically from 16-l%-palmityl-CoA gave rise only to radioactive myristic acid. Compound B is thus a 16 carbon fatty acid with a double bond at the 2,3-position.

The specific rotation of methyl D( -)-/3-hydroxybutyrate is -21.09” (33). I f one assumes the same rotation on a molar

basis for the two compounds, the expected specific rotation for methyl D( -)-@-hydroxypalmitate would be -8.71”, a value in good agreement with that found.

Metabolism of Saturated Long Chain Fatty Acyl-CoA Derivatives

The infrared spectrum of methyl oc,&hexadecenoate isolated as the enzymatic product derived from nonradioactive ,&hydroxypalmityl-CoA had the adsorption band at 985 cm-r characteristic of trans olefins conjugated with carbonyl groups (17). No evidence was found for the presence of the cis isomer. The retention time in gas-liquid chromatography of the enzymatic product from palmityl-CoA, &hydroxypalmityl-CoA, and ,l3, y-hexadecenoyl-Cob was identical with the retention time of synthetic methyl trans-a, /3-hexadecenoate. A sample of synthetic methyl cis-a,@-hexadecenoate was not available, but cis and tram isomers of cr,fl-unsaturated fatty acids are reported to have very different retention times (32) and the cis isomer should have been detected if it were present. The cy,P-hexadecenoate produced enzymatically had the same mobility in thin layer chromatography as synthetic trans-a, P-hexadecenoate.

With a given homogenate preparation, there were consistent differences between subcellular fractions in total conversion and ratios of products formed with l-14C-palmityl-CoA as substrate (Table III). With light particles (the most active fraction), total conversion of radioactive palmityl-CoA averaged about 39%, with maximum conversion of about 45%. Analysis of the products by gas-liquid chromatography and thin layer chromatography gave comparable ratios of products (Tables II and III). Recombination of the enzyme fractions caused no increase in the total conversion, and no significant shifts in the ratios of products.

The light particle fraction was also assayed for activity against myristyl-Coil and stearyl-CoA. Somewhat more myristyl-CoA was converted to products than either stearyl-CoA or palmityl- CoA. However, the ratio of products formed from each substrate was quite similar (Table II).

To investigate the relationship of the formation of the long chain fatty acyl derivatives to fatty acid degradation, particulate and supernatant fractions of homogenates of D. discoideum were incubated with l-r4C-palmityl-CoA, with and without cofactors (Table III). The radioactivity of each of the fatty acid products and of COz was determined.

Compound C-Compound C was characterized as methyl Although the oxidation of l-14C-palmityl-CoA to COZ was not

D( -)-/3-hydroxypalmitate. Its retention time in gas-liquid chromatography (Table I) and its mobility in thin layer chromatography were identical with those of synthetic methyl nn-@-hydroxypalmitate. The chromatographic behavior of Compound C did not change upon hydrogenation. However, acetylation of Compound C changed its retention time in a manner analogous to the effect of acetylation on the retention time of synthetic methyl 9(10)-hydroxypalmitate.

Upon van Slyke oxidation of radioactive Compound C enzymatically produced from lJ4C-palmityl-CoA, 720/, of the radioactivity was recovered as COZ. To characterize the other fragment of the molecule resulting from such oxidation, radioactive Compound C from 16J4C-palmityl-CoA was used. The oxidation mixture was extracted with petroleum ether, unlabeled carrier 2-pentadecanone was added, and the mixture was analyzed for mass and radioactivity by gas-liquid chromatography. Radioactivity cochromatographed exactly with the carrier 2-pentadecanone.

In order to obtain sufficient @-hydroxypalmitate to determine its optical activity, 36.5 pmoles of synthetic trans-or,Bhexa- decenoyl-Coil were incubated for 2 hours at 31” with 24 mg of soluble enzyme from the acetone powder of light particles. After hydrolysis of the thiol esters for 15 minutes at room temperature in 0.5 N NaOH, the solution was acidified and extracted with ether. The products were methylated with BFa-methanol. Gas-liquid chromatographic analysis of an aliquot of this preparation revealed 45% conversion of the substrate to P-hydroxypalmitate. Methyl ,&hydroxypalmitate was then isolated by chromatography on a column of silicic acid. Its specific rotation in ethanol was [cy]:’ -10.33”. The rotation was increasingly negative with decreasing wave length.


http://ww

w.jbc.org/

Dow

nloaded from

http://www.jbc.org/


TABLE III Radioactive products from incubation of D. discoideum homogenate

fractions with 1 -W-palmityl-CoA

Particulate fractions diluted to a final optical density of 0.32 at 600 rnp (approximately 0.8 mg of protein per ml) or supernatant fraction (approximately 1 mg of protein per ml) were incubated with 8.4 X lo5 c.p.m. (73 mMmoles) of 1-W-palmityl-CoA. All flasks contained 100 rmoles of phosphate buffer, pH 7.25; flasks with cofactors contained 1 pmole of CoA, 10 pmoles of ATP, 1 pmole of NAD, and 5 pmoles of Mg2+ in a final volume of 1.0 ml. Incubations were carried out in sealed flasks equipped with center wells containing 0.3 ml of Hyamine in methanol to trap COS. At the end of 2 hours of incubation at 30”, 0.2 ml of 10 N HzS04 was added to the reaction compartment, and the flasks incubated for 1 hour more. The Hyamine solution was then counted in diphenyloxazole-toluene in a scintillation spectrometer. Un- saturated and hydroxy fatty acids were extracted and separated from the substrate by the mercuric acetate adduct technique as described under “Experimental Procedure,” and determined quantitatively by continuous combustion-counting gas-liquid chromatography, as in Fig. 1. Total recovery of radioactivity was from 75 to 95%.

Enzyme and additions

Heavy particles. . Heavy particles + cofactors Light particles. Light particles + cofactors. Supernatant . Supernatant + cofactors.

Product

6 10 0.04 4 5 3.1

16 20 0.03 8 16 0.7 4 17 0.04 3 9 0.10

* The remainder of the recovered radioactivity was in palmitate.

great in any fraction, the heavy particle fraction was by far the most active. Oxidation to CO a was completely dependent on the presence of cofactors. The presence of oxaloacetate did not affect the results. It is also apparent that the accumulation of long chain acyl intermediates was only moderately affected by the presence of cofactors, and that in this regard, cofactors had similar effects whether or not the enzyme fraction was capable of degrading fatty acid to COZ. Thus, the light particle and supernatant fractions, although very active in transforming palmityl- CoA to the several intermediate fatty acids, were virtually unable to oxidize 1-r4C-palmityl-CoA to CO*.

Interconversions of Four Intermediates

To delineate more clearly their metabolic relationships, truns- OL , /3-hexadecenoyl-CoA, n@-hydroxypalmityl-CoA, trans-p, y-

hexadecenoyl-Coil, and cis-P,y-hexadecenoyl-CoA were each incubated separately with soluble enzymes extracted from the acetone powder of light particles. The soluble enzymes catalyzed the conversion of each substrate to all of the other compounds with one exception: there was only slight conversion of trans-0, y-hexadecenoyl-CoA to cis-/3, y-hexadecenoyl-Cob and cis-/3,y-hexadecenoyl-CoA was not converted to the trans isomer. Details of each incubation are presented below.

Metabolism of Bans-cu,P-Hexadecenoyl-CoA-Only the trans

isomer of this compound was available for study as substrate, but as discussed before, this appears to be the isomer produced enzymatically from the other substrates. truns-a,&Hexa- decenoyl-CoA was converted to &hydroxypalmitate much more rapidly than it was converted to /3,-r-hexadecenoate (Fig. 2). Control incubation with boiled enzyme showed no nonenzymatic conversion to either compound.

When a, P-hexadecenoyl-CoA was incubated with excess enzyme (Table IV), the relative amounts of products were similar to those formed when intact particulate enzyme was incubated with radioactive palmityl-CoA (Tables II and III). Qualitative analysis of the /3, y-hexadecenoic acid produced from a,&hexadecenoyl-CoA showed both cis and trans isomers to be

250 -

1

=, Y 200-

P

I Substrate :a,,&tfexadecenoyf CoA

150-

L 3

E

Et IOO-

.& Y - ffexadecenoate

0 30 60 90 120 150 180 TIME IN MINUTES

FIG. 2. Conversion of trans-cu,p-hexadecenoyl-CoA to P-hydroxypalmitate and p,-r-hexadecenoate. The incubation mixture contained 160 pg of soluble enzyme protein and 560 mpmoles of substrate per ml of 0.1 M phosphate buffer, pH 7.25. The reaction was started by addition of enzyme; incubation was at 31”. Ali- quots of 1.0 ml were removed at increasing time intervals, and analyzed for substrate and products by gas-liquid chromatography as described in “Experimental Procedure.”

TABLE IV Proportions of residual substrate and products after incubation of

excess soluble enzyme from D. discoideum homogenate with synthetic acyl-CoA substrates

Incubation mixtures contained either 110 mpmoles of a,p-hexadecenoyl-CoA, 165 mfimoles of cis-ply-hexadecenoyl-CoA, 195 m&moles of trans.p,r-hexadecenoyl-CoA, or 192 mpmoles of p- hydroxypalmityl-CoA, respectively, and 400 pg of soluble enzyme protein in 0.5 ml of 0.1 M phosphate, pH 7.25. After incubation for 1 hour at 31”, each sample was analyzed for substrate and products by gas-liquid chromatography.

Substrate

Final proportions of compounds

%

trans-Ly,p-Hexadecenoyl-CoA ...... 8.7 42.3 49.0 nr&Hydroxypalmityl-CoA ....... 7.4 10.5 82.1 trans$,r-Hexadecenoyl-CoA. ..... 71.3 8.3 19.4 cis-p,r-Hexadecenoyl-CoA ........ 68.5 8.6 20.8


http://ww

w.jbc.org/

Dow

nloaded from

http://www.jbc.org/


present. As noted before, the ,&hydroxypalmitate formed was the D( -) isomer.

Metabolism of DL-/SHydroxypalmityl-CoA-When nn-@-hydroxypalmityl-CoA was incubated with soluble enzyme under conditions identical with those employed for the incubation of a!, P-hexadecenoyl-CoA, the products were trans-cr , /3-hexadecenoate, and /3,y-hexadecenoate, of which 75% was the trans isomer and 25 y0 the cis isomer (Table IV). OL , P-Hexadecenoate was formed more rapidly than was 0, y-hexadecenoate (Fig. 3). The large proportion of @hydroxypalmitate remaining when the substrate was incubated with excess enzyme (Table IV) is probably because the L( +) isomer was not metabolized.

Metabolism of cis- and tran.@, y-Hexadecenoyl-CoA-Since both stereoisomers of P,y-hexadecenoate were formed enzymati- tally from all the other substrates, it was of interest to study the behavior of both isomers as substrate. Both cis- and trans-0, y- hexadecenoyl-CoA were converted to @-hydroxypalmitate and to oc,P-hexadecenoate (Fig. 4). The rate of conversion of cis-p, y- hexadecenoyl-Coil to P-hydroxypalmitate was about twice the rate of conversion of the trans isomer. Despite the differences in the rates of reaction, the equilibrium concentrations of products and substrates appeared to be about the same for both isomers (Table IV).

The ar,@hexadecenoate produced from both cis- and trans- ,L3, y-hexadecenoyl-Coil had the same mobility on thin layer chromatography and gas-liquid chromatography as the a:, ,6- hexadecenoate formed from P-hydroxypalmityl-CoA, and is presumed to be the trans isomer.

Isomerization of cis- and trans-/3 , y-Hexadecenoyl-CoA-The foregoing studies demonstrated the reversible interconversion of all of the compounds except for cis- and trans-p, y-hexadecenoate. To test for this, an aliquot of each of the samples analyzed in Fig. 4 was chromatographed on thin layer plates of AgNOa-silica gel. The areas corresponding to cis- and trans-methyl /3, y-hexadecenoate were eluted and quantitatively analyzed by gas-liquid

- Substrate : p-Hydroxypolmifyl CoA

0 Q, /-Hexadecenoafe

0 30 60 90 120 150 180 TIME IN MINUTES

FIG. 3. Conversion of nn-fl-hydroxypalmityl-CoA to a&hexadecenoate and@,r-hexadecenoate. The conditions of incubation were identical with those in Fig. 2, except that the substrate was 395 mpmoles of nn-p-hydroxylpalmityl-CoA per ml.

go- .8- Hydroxypolmif of e

ao-

o Substrate : m P, y - o Substrate : m P,r- -

Hexodecenoyl CoA

0 Substrate: Ls &y - 0 Substrate: Ls &y -

Hexodecenoyl CoA

08 1 I 1 I I I 0 30 60 90 120 150 180

TIME IN MINUTES

FIG. 4. Comparison of the rates of conversion of trans-p,r- hexadecenoyl-CoA and cis-fi,r-hexadecenoyl-Coil to P-hydroxypalmitate and cu,p-hexadecenoate. Incubation conditions were the same as in Fig. 2, except that the substrate was 310 mpmoles of trans.p,r-hexadecenoyl-CoA or 305 mpmoles of cis+,r-hexadecenoyl-Coil, and the reaction mixture contained 135 rg of soluble enzyme protein per ml.

chromatography. No production of the trans isomer from the cis substrate was observed. However, the cis isomer comprised about 5 to 6% of the total @,y-unsaturated fatty acid remaining at the end of the experiment in which the trans isomer was substrate.

The reasons for the absence or very low level of “cis-trans isomerase” activity are not clear. Thiol esterase activity has been detected in these enzyme extracts and may partially account for the results. For example, interconversion of the cis and trans isomers might occur only through the intermediate formation of ,&hydroxy- or cr , ,&unsaturated acyl thiolesters. If these were hydrolyzed as rapidly as they were formed, “isomerase” activity would not be observed.

Alternatively, both stereoisomers may bind to the same catalytic site on the enzyme, in competitive fashion. Again assuming that interconversion of the 0, y isomers can occur only through the intermediate formation of P-hydroxy- or a!,@-unsaturated compounds, the presence of a very large amount of either isomer of p,y-unsaturated acyl-CoA as substrate might prevent significant back reaction to the other isomer. By extension of this hypothesis, the observation that some cis-@,y- hexadecenoate is produced from the tram isomer, without the reverse occurrence, could be interpreted to mean tighter binding of the cis isomer than of the trans to the enzyme. The more rapid over-all rate of metabolism of the cis isomer than the trans isomer is consistent with this interpretation.

Evidence for Pathway of /3,y-Hexadecenoate Production-Both dehydration of a /I-hydroxy acyl-CoA (34, 35) and direct isomerization of the double bond of an cr,P-unsaturated acyl-CoA (36) have been postulated as mechanisms for the enzymatic formation of P,y-unsaturated fatty acyl-CoA in other systems. To investigate this problem, the rates of production of the P,-r-


http://ww

w.jbc.org/

Dow

nloaded from

http://www.jbc.org/

August 1964 F. Davidof and E. D. Kern

30 - 0, ,l? - Hexodecenoyl CoA

r’ \ z

TIME IN MINUTES

FIG. 5. Comparison of the rates of production of P,r-hexadecenoate from a,@-hexadecenoyl-CoA and &hydroxypalmityl-CoA. This figure is partially a composite of the data from Figs. 2 and 3. In addition, data are shown from an experiment using as substrate 870 mpmoles of P-hydroxypalmityl-CoA per ml. This was done to adjust the concentration of the D isomer of fi-hydroxypalmityl- CoA to a level more comparable to that of a,@hexadecenoyl-CoA.

hexadecenoic acids from each of the other synthetic substrates were compared. The data clearly indicate that 0, y-hexadecenoate was formed more rapidly from o(, P-hexadecenoyl-CoA than from /3-hydroxypalmityl-CoA (Fig. 5). These data are

inconsistent with a pathway in which /3-liydroxypalmityl-CoA

is an intermediate in the production of /3, y-hexadecenoate from cy , P-hexadecenoyl-CoA, and suggest the direct conversion of a!, ,&hexadecenoyl-CoA to ,8, y-hexadecenoate.

Further data consistent with this hypothesis were obtained from a kinetic study of the reverse reactions (i.e. the rates of conversion of cis- and tram-p, y-hexadecenoyl-CoA to LY , /?- hexadecenoate and P-hydroxypalmitate). From Fig. 4 it may be seen that a,&hexadecenoate was formed very quickly, and then remained relatively constant in concentration from the time of the earliest point at 15 minutes, whereas P-hydroxypalmitate was synthesized at a constant rate for 1 hour from tram-@, y- hexadecenoyl-Coil, and for 30 minutes from cis-P,y-hexadecenoyl-CoA.

A more detailed study of the initial rates of formation of 01, /3-hexadecenoate and /%hydroxypalmitate (Fig. 6) from trans- and cis- /3, y-hexadecenoyl-CoA showed more clearly the rapid appearance of c~,@hexadecenoate, most of which was formed in the first 5 minutes, It should be emphasized that in the experiment with trans-/3,y-hexadecenoyl-CoA (Fig. 6), the rate of formation of P-hydroxypalmitate increased during the first 10 minutes.

These data are compatible with the hypothesis that P,y- hexadecenoyl-Coil is converted directly to 01, ,B-hexadecenoyl- CoA, which is then converted to P-hydroxypalmityl-CoA. After the first few minutes of incubation, the rates of formation and conversion of a!, ,B-hexadecenoyl-CoA must be approximately equal, resulting in a relatively constant concentration of the oc,P-unsaturated acid and a continuing rise in concentration of @-hydroxypalmitate.

Many points remain to be clarified, however; for instance, it is not known if both stereoisomers of /z?, y-hexadecenoate are made by the same reaction, or that the reactions occur at the CoA level. There remains the real possibility that the acyl moiety must be transferred from CoA to a sulfhydryl group on the enzyme and that this is the rate-determining step. Purification of the enzymes should be helpful in elucidating the mechanisms involved.

Chain Length Specificity towards Hydroxy Substrates-Incuba- tions with P-hydroxymyristyl-CoA, P-hydroxypalmityl-CoA, and P-hydroxystearyl-CoA were carried out to determine the relative rates of reaction with each substrate. /3, y-Unsaturated fatty acids were produced at comparable rates from all three substrates (Fig. 7), although p, y-tetradecenoate was formed somewhat more rapidly. The ol,&unsaturated acid was formed more

I I I I I I 1

Substrot e : @ans /,r - Subslfale : & 1,~ -

50 - tiexodecenoyl Co A Hexadecenoyl CoA ’

/-

0 IO 20 30 0 IO 20 30

TIME IN MINUTES

FIG. 6. Comparison of the rates of conversion of trans-fi,r- hexadecenoyl-CoA and cis-p,r-hexadecenoyl-CoA to P-hydroxypalmitate anda&hexadecenoate. Conditions of incubation were the same as in Fig. 2 except that the substrate was 364 mpmoles of trans-p,r-hexadecenoyl-CoA per ml or 315 mpmoles of cis-fl,r- hexadecenoyl-Coil per ml.

x a

I I 1 I I I

p- Hydroxymyrisslyf Co A

OF I I 1 I I I 0 30 60 90 120 150 180

TIME IN MINUTES

FIG. 7. Conversion of P-hydroxymyristyl-CoA, fl-hydroxypalmityl-CoA, and p-hydroxystearyl-CoA to the corresponding p,r-unsaturated fatty acids. Conditions of incubation were the same as in Fig. 2, except that substrates were 183 mpmoles of &hydroxymyristyl-CoA, 117 mpmoles of p-hydroxypalmityl-CoA, or 163 mpmoles of P-hydroxystearyl-CoA per ml. Enzyme concentration was 320 pg of protein per ml.


http://ww

w.jbc.org/

Dow

nloaded from

http://www.jbc.org/


rapidly than the corresponding P,y-unsaturated acid in each case, and a ,,f3-tetradecenoate was formed more rapidly than were cr,&hexadecenoate and a,/3-octadecenoate.

DISCUSSION

The interconversions demonstrated in this paper are most readily explained by the following reactions:

Saturated fatty acid -+ trans-or,@

unsaturated or cis- or trans.p,y-unsaturated fatty acid (1)

trans-ol,p-Unsaturated fatty acid s

D(-)-fl-hydroxy fatty acid (2)

trans.a,@-Unsaturated fatty acid s

cis- and trans.@,r-unsaturated fatty acid (3)

These reactions occur either as the CoA derivatives or possibly as thiol esters bound directly to sulfhydryl groups on the enzymes; there are no data to exclude either possibility. Although both Reactions 2 and 3 have been shown to be reversible, meas- urement of equilibrium constants has been impossible in the crude system because of the presence of thiolesterase activity, and contamination of the substrates by inactive acyl thiolesters (18). Each of the above reactions will now be discussed in some detail.

The evidence for Reaction 1 is indirect, depending entirely upon the demonstration of conversion of saturated fatty acyl- CoA to all the intermediates. Direct dehydrogenation of the saturated fatty acid is the simplest explanation. This reaction is presumably analogous either to that catalyzed by the acyl dehydrogenase of fatty acid oxidation (37), or the reversal of the reaction catalyzed by the ethylene reductase of fatty acid synthesis (2).

The hydration of truns-a! , P-unsaturated fatty acid (Reaction 2) is similar to the hydration step in fatty acid synthesis in that the D(-)-fi-hydroxy acid is formed (2). The product of the analogous step in fatty acid oxidation is the L( +) isomer when truns-oc,P-unsaturated fatty acid is the substrate (38). The cellular slime mold enzyme appears to form only trans-a,@-

unsaturated fatty acids from @-hydroxy acids. The enoyl hydrase of &oxidation, although not specific for the tram-a,& unsaturated isomer as substrate (38), synthesizes only the tram isomer from L(+)-@hydroxy fatty acids, for which substrate the enzyme is specific (34, 35, 39). The stereochemistry of the a,&unsaturated fatty acid produced by the dehydrase of fatty acid synthesis is apparently also tram (2). The stereochemistry of the intermediates in fatty acid synthesis and oxidation have not been directly determined for compounds of chain length greater than C~Z.

The evidence for Reaction 3 is the kinetics of the interconversions of (Y, /3-hexadecenoate, 0, y-hexadecenoate, and ,&hydroxypalmitate when each is added as the CoA derivative. The data strongly suggest that cr , &hexadecenoate is the intermediate between /3,y-hexadecenoate and &hydroxypalmitate. The al- ternative pathway with /?-hydroxypalmitate as the intermediate between the two unsaturated fatty acids cannot be definitely excluded by these data, however, especially if the interconversions occur as enzyme-bound substrates rather than as the free CoA derivatives. There are, in fact, some possible precedents

for the direct int.erconversion of /3-hydroxy acids and /3, y-unsaturated acids. Enoyl hydrase appears to catalyze the hydration of /3, y-unsaturated acids although at a much slower rate than the hydration of a,/%unsaturated acids (34, 35). However, the possibility exists that this hydration actually follows isomerization of the /3,r-isomer to the o(,/%isomer (36). Bloch et al. (32,40) have studied a bacterial enzyme that converts P-hydroxydecanoyl-Cob to p, y-decenoate. (The stereochemistry of the product has not been reported.) These investigators have suggested that the reaction is a direct dehydration, but the possible presence of an cr,@-P,r isomerase in their preparations has not been eliminated. The reaction is specific for D( -)-p-

hydroxydecanoyl-CoA. There are also precedents, however, for the direct isomerization

of cr,/3- and /3,y-unsaturated fatty acids as postulated in Reac- tion 3. Such reactions have been described previously in bacterial (41,42) and mammalian systems (36), but only for short chain fatty acids. In these cases, vinylacetate has been shown to be converted to crot0nat.e without the intermediate formation of @-hydroxybutyrate. With increasing chain length much less activity was found. The data reported in this paper indicate that this type of isomerization also occurs with long chain fatty acids, and are also the first demonstration of the formation and conversion of both cis and trans isomers of p, y-unsaturated fatty acids. The apparent lack of stereospecificity is unexplained.

It is perhaps surprising that the cellular slime mold enzymes form OL ,&hexadecenoate and p, y-hexadecenoate in the ratio 3 : 1. The reaction catalyzed by vinylacetyl isomerase, for example, leads to 98% crotonate at equilibrium (41). The differences may be attributable to the difference in chain length of the substrates. The proportions of crotonate and vinylacetate produced by alkali-catalyzed isomerization (43) are identical with those produced enzymatically. Elongation of the fatty acid chain beyond 4 carbons, however, appears to change the ratio of products formed by alkali isomerization, so that 30% of the P,r isomer is found at equilibrium (43). The influence of the thiol ester bond on the enzyme-catalyzed reaction cannot be quantitatively evaluated, but by analogy to ketones (44, 45), one might expect the equilibrium mixture to contain relatively more of the cy , ,&unsaturated compound than is found with the free acids.

The metabolic role of these fatty acid intermediates is not obvious; they may function in oxidation, synthesis, or elongation of fatty acids. The data cannot be consistently interpreted to favor any one of these possibilities.

The least likely role for these reactions would seem to be in oxidation of fatty acids, for the following reasons. (a) Growing cells of D. discoideum do not degrade palmitate or stearate to any significant extent (46), although laurate and myristate are partially degraded. (b) There was no correlation (Table III) between the ability of subcellular fractions to oxidize l-14C- palmityl-CoA to COZ, and to convert it to any of the intermediates under consideration. (c) The D(-)-hydroxy acids are formed, rather than the L( +) isomer thought to be the intermediate in fatty acid oxidation. (d) It is difficult to envision a role for @,y-unsaturated fatty acids in /3 oxidation unless they be intermediates in the oxidation of unsaturated fatty acids. Dur- ing /I oxidation of oleate, for example, P, y-dodecenoate would presumably be formed. The naturally occurring unsaturated fatty acids of D. discoideum have double bonds in positions that similarly would lead to the formation of /3, y-unsaturated fatty


http://ww

w.jbc.org/

Dow

nloaded from

http://www.jbc.org/


acids upon fi oxidation (21). The lack of st.ereospecificity with regard to synthesis of the p, y-unsaturated acids in vitro suggests that Reaction 3 might serve to consume, rather than to produce, /3, y-unsaturated fatty acids in Go.

In this connection it is pertinent to note that although there is indirect evidence that the acyl dehydrogenases of mitochondrial fatty acid oxidation dehydrogenate short chain fatty acids to o( ,&unsaturated fatty acids, the possible formation of a p, y-unsaturated product has not been eliminated, especially for long chain fatty acids. In fact, /3,y-unsaturated acyl-CoA binds to acyl dehydrogenase and produces the same spectral change in the flavoprotein as does oc,&unsaturated acyl-CoA (47).

Bloch (48) has proposed that the formation of ,l3, y-unsaturated fatty acids in certain bacteria is a necessary step in the synthesis of unsaturated fatty acids. A similar mechanism does not seem likely in D. discoideum despite the fact that the diunsatu- rated fatty acids of the cellular slime mold, 5,9-hexadecadienoate, 5,9-octadecadienoate, and 5,l I-octadecadienoate, all have a double bond at C-5. One might suppose that these acids arise by /3,y-unsaturation of 7tetradecenoate, 7-hexadecenoate, and 9-hexadecenoate, respectively, followed by addition of a 2 carbon unit. However, the naturally occurring unsaturated acids are all of the cis configuration, while the trans isomer is the predominant product of the enzymatic reactions described in this paper. The major objection, however, is that experiments in the intact organism have demonstrated that desaturation at the 5-position occurs after the carbon chain is completed (46).

The predominant fate of long chain fatty acids (C,, and C,,) in growing cells of D. discoideum is elongation (and desaturation). This suggests that the reactions described in this paper may be biosynthetic. The configuration of the hydroxy acid is consistent with this role. It is still difficult, however, to define a role for the /?, y-unsaturated fatty acids. It is conceivable that the pathway of fatty acid elongation involves condensation of long chain fatty acyl-CoA with acetyl-CoA (49) to form the @-keto acid, reduction to the P-hydroxy acid, and dehydration to the cr,S-unsaturated acid which is then isomerized to the /?,y-unsaturated acid. Reduction of a double bond at the @,y-position would be more exothermic than reduction at the cu,&position, which might provide some thermodynamic advan- tage in the reduction step. This might explain the function of the truns-/?, y-unsaturated fatty acids. The reactions described in these studies might thus represent reversal of those steps following condensation in a chain elongation pat’hway.

Formation of &s-/3, y-unsaturated acids might be incidental to the formation of the saturated fatty acid chain, but serve, in some systems, for formation of unsaturated fatty acids. Al- though it is believed that in the usual fatty acid synthetase the intermediate which is reduced is the a,P-unsaturated acid (a), there is no direct evidence on this point for long chain fatty acids, and the intermediates in the chain elongation reactions of mitochondria are unknown as yet (49). We have no data on which unsaturated fatty acid is reduced in D. discoideum.

SUMMARY

Subcellular fractions from homogenates of the cellular slime mold, Dictyostelium discoideum, convert saturated C1+ C& and Cl8 fatty acyl coenzyme A to the trans-ac,p-unsaturated-, D( -)-

&hydroxy, and cis- and trans-P,y-unsaturated fatty acids of the same chain length. Soluble enzymes, derived from acetone

powders of such subcellular fractions, catalyze the interconversion of all of these products. Evidence is presented that the reactions involved include the reversible hydration of (Y,P-unsaturated acids, and the direct isomerization of a!, /3- and p, y-unsaturated acids.

REFERENCES

1. GREEN, D. E., Biol. Revs. Cambridge Phil. Sot., 29, 330 (1954). 2. LYNEN. F.. Federation Proc.. 20, 941 (1961). 3. KENNI&Y,‘E. P., AND LENN~NGER, A. L.,’ in W. D. MCELROY

AND B. GLASS (Editors), Phosphorus.metabolism, Vol. 2, Johns Hopkins Press, Baltimore, 1952. n. 253.

4. WAKIL, S. >., Proceedings of the’$fth ‘international congress of biochemistry, Vol. ‘Y, The Macmillan Company, New York, 1963, p. 3.

5.

6.

7. 8.

9.

10.

11.

12.

13. 14.

15. 16. 17.

18.

19.

HELE, P., POPJAK, G., AND LIURYSSENS, M., Biochem. J., 66, 348 (1957).

GREEN, D. E., AND WAKIL, S. J., in K. BLOCH (Editor), Lipid metabolism, John Wiley and Sons, Inc., New York, 1960, p. 1.

BEINERT, H., J. Biol. Chem., 206, 575 (1953). WAKIL, S. J.. AND MAHLER, H. R., J. Biol. Chem.. 207. 125

(1954). ,

CRANE, F. L., MII, S., HAUGE, J. G., GREEN, D. E., AND BEINERT. H., J. Biol. Chem., 218, 701 (1956).

GOLDMAN, P., ALBERTS, A. W.; AND VAG~LOS; P. R., J. Biol. Chem., 238, 3579 (1963).

ROBINSON, J. D., BRADLEY, R. M., AND BRADY, R. O., Bio- chemistry, 2, 191 (1963).

ROBINSON, J. D., BRADLEY, R. M., AND BRADY, R. O., J. Biol. Chem.. 238.528 (1963).

VON RUDLOF;, E.,. Can.‘J. Chem., 34, 1413 (1956). SHRINER, R. L., in R. ADAMS (Editor), Organic reactions, Vol.

1. John Wilev and Sons. Inc., New York, 1942, D. 1. DE VRIES, B., j. Am. Oil 3hemists’ Sot., 4d, 184 if963). MORRIS, L. J., Chem. and Ind. (Londok), 1238 (1962): SINCLAIR. R. G.. MCKAY. A. F.. MYERS. G. S.. AND JONES.

R. N., >. Am. ‘Chem. So;., 74, i578 (195i). ’ GOLDMAN, P., AND VAGELOS, P. R., J. Biol. Chem., 236, 2620

(1961).

20.

21. 22.

23.

24.

25. 26.

27.

28. 29.

30.

31. 32.

33.

34.

STADTMAN, E. R., in S. P. COLOWICK AND N. 0. KAPLAN (Edi- tors), Methods in enzymology, Vol. 5, Academic Press, Inc., New York, 1951, p. 931.

DAVIDOFF, F., AND KORN, E. D., J. Biol. Chem., 238, 3199 (1963).

GEZELIUS, K., Exptl. Cell Research, 23, 300 (1961). DRYSDALE, G. R., AND LARDY, H. A., J. Biol. Chem., 202, 119

(1953). LOWRY, 0. H., R,OSEBROUGH, N. J., FARR, A. L., AND RANDALL,

R. J., J. Biol. Chem., 193, 265 (1951). METCALFE, L. D., AND SCHMITZ, A. A., Anal. Chem., 33, 363

(1961). GOLDFINE, H., AND BLOCH, K., J. Biol. Chem., 236,2596 (1961). KARMEN, A., MCCAFFREY, I., AND BOWMAN, R. L., J. Lipid

Research, 3,281 (1962). KARMEN, A., GIUFFRIDA, L., AND BOWMAN, R. L., J. Lipid

Research, 3,44 (1962). VAN SLYKE, D. D., J. Biol. Chem., 32,455 (1917). KISHIMOTO, Y., AND RADIN, N. S., J. Lipid Research, 4, 130

(1963). HUEBNER, C. F., AMES, S. R., AND BUBL, E. C., J. Am. Chem.

Sot., 68, 1621 (1946). LAUN~R, H. F., j. A&. Chem. Sot., 64, 2597 (1932). LENNARZ, W. J., LIGHT, R. J., AND BLOCH. K.. Proc. Natl.

Acad. Sci. U. S., 48, 840 (1962). Beilsteins Handbuch der Organ&hen Chemie, Vol. 5, Ed. 4,

Berlin. &ringer-Verlag. 1921. II. 308. STERN, J., in PrD. BOYE;, H. L~RDY, AND K. MYRB&K (Edi-

tors), The enzymes, Vol. 6, Academic Press, Inc., New York, 1961, p. 511.

35. STERN, J. R., AND DEL CAMPILLO, A., J. Biol. Chem., 218, 985 (1956).


http://ww

w.jbc.org/

Dow

nloaded from

http://www.jbc.org/

Biosynthesis of Fatty Acid Derivatives Vol. 239, No. 8

36. RILLING, H. C., AND COON, M. J., J. Biol. Chem., 236, 3087 (1960).

37. HAUGE, J. G., CRANE, F. L., AND BEINERT, H., J. Biol. Chem., 219, 727 (1956).

38. WAHIL, S. J., Biochim. et Biophys. Acta, 19,497 (1956). 39. MAHLER, H. R., Federation Proc., 12, 694 (1953). 40. NORRIS, A. T., AND BLOCH, K., J. Biol. Chem., 238, PC3133

(1963). 41. BARTSCH. R. G., AND BARKER, H. A.. Arch. Biochem. Biovhus.,

92, 122’ (1961): _ 1

42. HARDMAN, J. K., AND STADTMAN, T. C., J. Biol. Chem., 238, 2088 (1963).

43. LINSTEAD, R. P., AND NOBLE, E. G., J. Chem. Sot., 614 (1934).

44. KON, G. A. R., LINSTEAD, R. P., AND WRIGHT, J. M., J. Chem. Sot., 599, (1934).

45. ABBOTT, A. E., KON, G. A. R., AND SATCHELL, R. D., J. Chem. Sot., 2514 (1928).

46. DAVIDOFF, F., AND KORN, E. D., J. Biol. Chem., 238, 3210 (1963).

47. BEINERT, H., AND CRANE, F. L., in W. D. MCELROY AND B. GLASS (Editors), Inorganic nitrogen metabolism, The Johns Hopkins Press, Baltimore, 1956, p. 601.

48. BLOCH, Ii., BARONOWSKY, P. E., GOLDFINE, H., LENNARZ, W. J., L&IT, R., NORRIS, A. T;, AND SCHE~ER~RANDT, G.; Federation Proc.. 20, 921 (1961).

49. HARLAN, W. R., JR., AND %‘AKI;, S. J., J. Biol. Chem., 238, 3216 (1963).


http://ww

w.jbc.org/

Dow

nloaded from

http://www.jbc.org/

Frank Davidoff and Edward D. KornDictyostelium discoideumthe Cellular Slime Mold,

-Hydroxy Derivatives by Enzymes fromβ-Unsaturated, and γ,β-Unsaturated, β,αThe Conversion of Long Chain Saturated Fatty Acids to Their

1964, 239:2496-2506.J. Biol. Chem.

http://www.jbc.org/content/239/8/2496.citation

Access the most updated version of this article at

Alerts:

When a correction for this article is posted•

When this article is cited•

to choose from all of JBC's e-mail alertsClick here

http://www.jbc.org/content/239/8/2496.citation.full.html#ref-list-1

This article cites 0 references, 0 of which can be accessed free at


http://ww

w.jbc.org/

Dow

nloaded from

http://www.jbc.org/content/239/8/2496.citation

http://www.jbc.org/cgi/alerts?alertType=citedby&addAlert=cited_by&cited_by_criteria_resid=jbc;239/8/2496&saveAlert=no&return-type=article&return_url=http://www.jbc.org/content/239/8/2496.citation

http://www.jbc.org/cgi/alerts?alertType=correction&addAlert=correction&correction_criteria_value=239/8/2496&saveAlert=no&return-type=article&return_url=http://www.jbc.org/content/239/8/2496.citation

http://www.jbc.org/cgi/alerts/etoc

http://www.jbc.org/content/239/8/2496.citation.full.html#ref-list-1

http://www.jbc.org/

the conversion of long chain saturated fatty acids to ... · pdf filethe conversion of long...

Documents