The effect of aromatic CoA esters on fatty acid synthetase: Biosynthesis of ω-phenyl fatty acids

The effect of aromatic CoA esters on fatty acid synthetase: Biosynthesis of ω-phenyl fatty acids

ARCHIVES OF BIOCHEMISTRY AND BIOPHYSICS Vol. 222, No. 1, April 1, pp. 259-265, (1983) The Effect of Aromatic CoA Esters on Fatty Acid Synthetase: Bio...

565KB Sizes 20 Downloads 83 Views

ARCHIVES OF BIOCHEMISTRY AND BIOPHYSICS Vol. 222, No. 1, April 1, pp. 259-265, (1983)

The Effect of Aromatic CoA Esters on Fatty Acid Synthetase: Biosynthesis of w-Phenyl Fatty Acids’ STUART Bruce

Lyon

Received

SMITH’

AND

ALAN

STERN

Memorial Research Laboratory, Children’s Hospital Medical Northern Calgbrnia, Oakland, Cal$brnia 94609 September

20, 1982, and in revised

form

November

Center

of

22, 1982

Aromatic carboxylic acids ingested by, or formed in, the body can be converted to the CoA derivatives but the possible metabolic fate of these thioesters has not been investigated extensively. We have examined the effects of two such thioesters, benzoylCoA and phenylacetyl-CoA, on the mammalian fatty acid synthetase. Benzoyl-CoA inhibited the enzyme, apparently by competing with acetyl-CoA and malonyl-CoA for substrate binding sites. Phenylacetyl-CoA, on the other hand, could replace acetyl-CoA as a primer for the fatty acid synthetase reaction; the product was almost exclusively o-phenyldodecanoic acid. The Km of the synthetase for phenylacetyl-CoA was considerably higher than that for acetyl-CoA and the rate of synthesis of w-phenyldodecanoic acid was only 16% of that of palmitic acid. Experiments in which the rate of synthesis and release of o-phenyl moieties from the synthetase was compared with that of naliphatic moieties indicated that the rate limiting step was the initiation of chain growth from phenylacetyl-CoA; release of the synthesized acyl moieties by chain terminating thioesterases was equally rapid in the case of w-phenyl and n-aliphatic acids. Possible metabolic consequences of the effects of these aromatic CoA esters on lipogenesis are discussed.

Aromatic carboxylic acids are formed in the body by a variety of reactions. For example, toluene .is detoxified by oxidation to benzoic acid (1) and phenylalanine is metabolized to phenylpyruvic, phenyllactic, and phenylacetic acid (2); these phenylalanine metabolites are present in elevated levels in the phenylketonuric condition (3). Benzoic acid finds its way into the human diet as a natural constituent of certain fruits and also as a food preservative. In addition, a number of hypolipidemic drugs have been described which are carboxylic acids with aromatic rings. One such compound, truns-6-{2-[4-(l,l-dimethylethyl)-phenyllethyl} tetrahydro-4hydroxy-2H-pyran-2-one (SC-33459), is

believed to be capable of entering the poxidation pathway, giving rise to p-tertbutylbenzoyl-CoA (4). Apparently aromatic carboxylic acids, such as benzoic acid (5), can indeed be activated to the CoA thioester. It has been suggested that the hypolipidemic effect of certain of these drugs might be attributable, at least in part, to the inhibitory effect of the CoA esters on lipogenesis (4). In view of the lack of experimental data on the effects of CoA esters of aromatic carboxylic acids on lipogenic enzymes, we conducted a study on the effect of phenylacetyl-Cod and benzoyl-CoA on the fatty acid synthetase enzyme. MATERIALS

i This work was supported RR 05467 from the National ’ To whom correspondence

by Grants AM16073 and Institutes of Health. should be addressed.

AND

METHODS

Enzymes. The procedure for the purification assay of fatty acid synthetase (6), trypsinized 259

0003-9861/83 Copyright All rights

and fatty

$3.00

0 1983 by Academic Press, Inc. of reproduction in any form reserved.

260

SMITH

AND

acid synthetase core (7), thioesterase I (6), and thioesterase II (7) have been described in earlier publications from this laboratory. Substrates. All CoA esters were purchased from PL Biochemicals, Wisconsin. Radioactive malonyl-CoA was purchased from New England Nuclear Corporation, Massachusetts, and NADPH from Sigma Chemical Company, Missouri. Identijication of acyl-enzyme thioesters. Acyl-S-enzyme species were synthesized by elongation of either acetyl-S-enzyme or phenylacetyl-S-enzyme with [2i4C]malonyl-CoA, as described in the text. Reactions were stopped by addition of CoA to give a final concentration of 0.1 mM; CoA causes almost instantaneous unloading of acetyl and malonyl moieties from the fatty acid synthetase (8,9), thus preventing further chain initiation or elongation. Acyl-S-enzyme species were precipitated by addition of an equal volume of 95% (v/v) ice-cold ethanol. Carrier enzyme, 0.8 mg, was added and the mixture was incubated at 0°C for 1 h. The precipitated protein was collected by centrifugation, washed twice with 1 ml of 50% ethanol, and dissolved in 0.5 ml of 0.5 M NaOH. The solution was left at room temperature overnight to allow complete hydrolysis of the acyl-S-enzyme thioesters to take place. One hundred microliters of concentrated HCl and a few milligrams of a carrier fatty acid mixture were added and the fatty acids were extracted by partitioning twice with 2.5 ml of petroleum ether. A portion of the extract was taken for determination of radioactivity and the remainder was used to prepare methyl esters (10) for gas-liquid radiochromatography. Fatty acid identi)ication Radioactive fatty acids, both n-aliphatic and o-phenyl, were identified by gasliquid radiochromatography of the methyl esters on two stationary phases. One column, 8 ft X I/s in., packed with 15% DEGS on Gas Chrom Q 800/100, was run isothermally at 200°C with a flow rate of 25 ml/min. The second column, 6 ft X i/s in., packed with 3% SE on Supelcoport 80/100, was run isothermally at 160°C using a flow rate of 26 ml/min. Preliminary experiments, performed using nonradioactive, authentic standards confirmed that plots of log retention time against aliphatic carbon number were linear for both n-aliphatic and o-phenyl fatty acids on both DEGS and SE 30 columns. In all experiments, the elution times of the radioactive zones corresponded precisely with those predicted for either an n-aliphatic or an w-phenyl fatty acid. RESULTS

Benzoyl-CoA inhibited the fatty acid synthetase reaction. Inhibition appeared to be competitive for both acetyl-CoA and malonyl-CoA (Fig. 1). Benzoyl-CoA could not replace acetyl-CoA as a primer for the

STERN

fatty acid synthetase reaction. In contrast, phenylacetyl-CoA could act as a substrate for the enzyme (Fig. 2). The K, for phenylacetyl-CoA was higher than that for acetyl-CoA and the V,,, considerably lower. The major product of fatty acid synthesis from malonyl-CoA and phenylacetyl-CoA was o-phenyldodecanoic acid (Table I). Some long chain n-fatty acids, mainly palmitic, were also synthesized, presumably as a result of decarboxylation of malonyl moieties to form acetyl moieties (12). Several experiments were performed in an attempt to determine whether the lower rate of synthesis of w-phenyl fatty acid, compared to n-fatty acid, was attributable to a lower rate of chain initiation, chain elongation, or chain termination. In the first experiment, the trypsinized fatty acid synthetase core was used to compare the rate of assembly of fatty acyl moieties starting with acetyl-CoA and phenylacetyl-CoA (Fig. 3). The rate of long chain acyl-S-enzyme formation was much slower with phenylacetyl-Cod as primer (reaction 2) than with acetyl-CoA (reaction 1). To distinguish between the possibilities that either chain initiation or chain elongation might be retarded when phenylacetyl-CoA is used as primer, a third incubation was performed which was initially identical to reaction 2; however, in reaction 3 acetyl-CoA was added to the incubation after the reaction had been allowed to proceed for a short time in the presence of phenylacetyl-CoA. Acetyl moieties participate only in the chain initiation process and not in the chain elongation process. We reasoned therefore that if chain initiation with phenylacetyl-CoA occurred rapidly and elongation was rate limiting then the addition of acetyl-CoA would have no effect on the rate of longchain acyl-S-enzyme synthesis. If on the other hand, chain initiation was rate limiting then the addition of acetyl-CoA should cause a marked increase in the rate of long-chain acyl-S-enzyme synthesis. The experimental results shown in Fig. 3 favored this latter alternative and an experiment performed subsequently using radiolabeled substrate confirmed this interpretation. In this experiment (Table II),

EFFECT

OF I

AROMATIC

CoA I

ESTERS

ON

FATTY

16

500pM

I

-4

-2

261

SYNTHETASE

.

A

I

ACID

6--

/

l

I

I

2

4

6

0

-6

0

6

l/S

FIG. 1. The effect of benzoyl-CoA on fatty acid synthetase activity. Activity was assayed spectrophotometrically at 37°C. Incubation systems contained, in a final volume of 0.5 ml: 0.1 M potassium phosphate buffer (pH 6.6), 250 PM NADPH, 20 pg fatty acid synthetase, and various concentrations of acetyl-CoA, malonyl-CoA, and benzoyl-CoA. Lineweaver-Burk plots are shown, at the indicated concentrations of benzoyl-CoA, for malonyl-CoA (A) and acetyl-CoA (B). Acetyl-CoA concentration was fixed at 50 PM in A and malonyl-CoA at 56 PM in B. Units of V are nanomoles of NADPH oxidized per minute per microgram of FAS. Units of S are nanomoles per milliter.

it was revealed that after incubation for 1.5 min with phenylacetyl-CoA and [2“C]malonyl-CoA that only 23% of the enzyme molecules had initiated acyl chain growth. Most of the products were wphenyl acyl moieties but some n-aliphatic acyl moieties had been formed, presumably as a result of the enzyme’s ability to generate acetyl moieties by decarboxylation of malonyl moieties. No acyl moieties shorter than n-C& or w-phenyl Cl0 were detected. Ater 12 min of incubation, many more enzyme molecules (68%) had initiated assembly of acyl moieties and growth had advanced to a longer chain length. In the third incubation, where acetyl-CoA was added after an initial exposure of 1.5 min to phenylacetyl-CoA, there

was a marked increase in the number of enzyme molecules which had initiated acyl chain growth (from 23 to 60%) and most of the increase was attributable to the formation of n-aliphatic acyl moieties. For comparative purposes, an incubation system containing acetyl-CoA as the sole primer was included; after only 1.5 min 86% of the enzyme molecules had completed formation of a long-chain acyl thioester. These experiments showed conclusively that when phenylacetyl-CoA, rather than acetyl-CoA, is offered as a primer for the fatty acid synthetase there is a marked reduction in the rate of initiation of chain growth by the enzyme. A second series of experiments was performed to determine whether the chain

262

SMITH

AND

Acyl-CoA,

STERN

pM

FIG. 2. Comparison of the kinetics of elongation of aeetyl-CoA and phenylacetyl-CoA by the fatty acid synthetase. Incubation systems contained 0.1 M potassium phosphate buffer (pH 6.6), 125 pM NADPH, 50 pM malonyl-CoA, and either acetyl-CoA or phenylacetyl-CoA, as indicated; reactions were started with 5 pg of enzyme to give a final volume of 0.5 ml. The malonyl-CoA used in this experiment had been purified by high-performance liquid chromatography (11) in order to remove all traces of acetyl-CoA. Note the different magnitude of the units on the vertical axis for acetylCoA and phenyl acetyl-CoA experiments. Lineweaver-Burk plots, shown as an inset, gave a K,,, of 3 pM and V,, of 2500 with acetyl-CoA and a K,,, of 20 pM and a V,,,,. of 400 with phenylacetyl-CoA.

termination of fatty acid synthesis might be affected by replacing acetyl-CoA with phenylacetyl-CoA. Again the trypsinized fatty acid synthetase core was utilized, this time to generate acyl-S-enzyme species TABLE

I

PRODUCTS OF FATTY ACID SYNTHESIS USING ACET~LCoA AND PHENYLACETYL-COA AS PRIMERS Products

n-Fatty Substrate

(25 PM)

acetyl-CoA phenylacetyl-CoA

(mol

%)

w-Phenyl fatty acids

acids

Cl1

Cl6

Cl8

Cl0

Clz

Cl4

4 1

87 26

9 1

0 0

0 72

0 0

Note. Assay conditions were essentially the same as described in Fig. 2 except that [2-“C]malonyl-CoA (0.1 pCi) was used. Incubations were conducted at 37°C for either 1 min (acetyl-CoA) or 5 min (phenylacetyl-CoA).

which could be presented as potential substrates for two chain-terminating enzymes, thioesterases I and II. Thioesterase I, a covalently linked domain of the fatty acid synthetase which can be removed by limited trypsinization (13), is the enzyme responsible for chain termination on the native synthetase (14); thioesterase II is a separate enzyme, not part of the synthetase complex (15), which is responsible for early termination of acyl chain growth in some mammary glands (14). Under the conditions selected for chain elongation, all of the available malonyl-CoA was utilized in the formation of the acyl-s-enzyme (Table III). Thus, the release of acyl moieties from the enzyme, catalyzed by the thioesterases, could be studied in the absence of complications arising from resynthesis of new acyl-S-enzyme species. With acetyl-CoA as primer, the major species synthesized was palmityl-S-enzyme; using phenylacetyl-CoA, it was w-phenyl-tetradecanoyl-S-enzyme. Neither thioesterase

EFFECT

OF

AROMATIC

0.61

1 0

CoA

I 2

ESTERS

I 4

ON

I 6

FATTY

ACID

I 8

I 10

263

SYNTHETASE

I 12

I 14

Minutes FIG. 3. Rate of acyl chain elongation by the trypsinized fatty acid synthetase core using acetylCoA and phenylacetyl-CoA as primer. Incubation systems contained initially: 0.2 M potassium phosphate buffer (pH 6.6), 125 pM malonyl-CoA, 250 pM NADPH, and either 50 pM acetyl-CoA (reaction 1) or 50 FM phenylacetyl-CoA (reactions 2 and 3). In reaction 3, acetyl-CoA (50 pM, final concentration) was added 1.5 min after initiating the reaction which contained phenylacetyl-CoA. Reactions were started by the addition of 1 mg of trypsinized fatty acid synthetase (indicated by the arrow). Incubation temperature was 10°C rather than 37°C since the elongation rate is too fast to measure spectrophotometrically at the higher temperature.

showed any preference for palmityl-S-enzyme over w-phenyl acyl-S-enzyme as substrate (Table III). However, thioesterase II was about 30 times more effective than thioesterase I in catalyzing acyl-S-fatty acid synthetase thioester hydrolysis. Finally we performed an experiment to determine whether thioesterase II exhibited the same broad acyl specificity with o-phenyl thioesters as had been shown previously with n-aliphatic thioesters (12). Indeed, when thioesterase II was present

during the assembly of acyl moieties on the trypsinized fatty acid synthetase core, a wide spectrum of product chain lengths was synthesized and released irrespective of whether acetyl-CoA or phenyl acetylCoA was used as primer (Table IV). DISCUSSION

The observation that phenylacetyl-CoA can be elongated by the fatty acid synthetase was surprising to us. Although for-

TABLE

II

FORMATIONOFACYL-ENZYMETHIOESTERSFROMACETYL-COA AND~HENYLACETYL-COA BYTRYPSINIZEDFATTYACIDSYNTHETASE Products

Incubation

conditions

Phenylacetyl-CoA, 1.5 min Phenylacetyl-CoA, 12 min Phenylacetyl-CoA, 1.5 min then acetyl-CoA, 2 min Acetyl-CoA, 1.5 min

Total n-Aliphatic + w-phenyl

(pmol

acyl-S-enzyme

n-Aliphatic

w-Phenyl

Cl6

Cl8

Czo

Total

Cl0

Cl2

Cl4

Cl6

Total

113 339

3 0

22 13

4 25

29 38

3 12

19 25

49 104

13 160

84 301

295 431

10 43

135 328

33 60

178 431

0 0

10 0

‘71 0

36 0

117 0

Note. Reaction conditions were identical to those in Fig. 3 except that [2-%]malonyl-CoA (43 Ci/mol) was used in place of unlabeled malonyl-CoA and the reaction volume was 0.1 rather than 0.5 ml. Reaction mixtures contained 500 pmol trypsinized FAS. Acyl-S-enzyme species were identified as described under Experimental Procedures.

264

SMITH

TABLE

AND

III

RELEASEOFW-PHENYLANDR-FMTYACIDSFROM CoVALENTLINKAGETOTHETRYPSINIZEDFATTY ACID SYNTHETASE COREBYCHAIN-TERMINATING THIOESTERASES w-phenyltetradecanoyl-S-FAS hydrolysis (nmol/min/mg protein)

Enzyme Thioesterase Thioesterase

I II

11.3 333

PalmitylS-FAS hydrolysis 11.8 364

Note. Radioactively labeled w-phenyltetradecanoyl-S-FAS and palmityl-S-FAS were synthesized by incubating 4.6 FM trypsinized fatty acid synthetase core in 0.1 M potassium phosphate buffer (pH 8) containing 31.4 pM [2-“Clmalonyl-CoA (1.5 Ci/mol), 0.25 mM NADPH, and 50 PM phenylacetyl-CoA or acetylCoA, for 2 min at 37°C. Thioesterase was then added, over a range of enzyme concentrations, and the incubation was continued for an additional 5 min when the reaction was terminated by the addition of 100 ~1 of 20% perchloric acid. Radioactive fatty acids released from thioester linkage were extracted into hexane and assayed by liquid scintillation spectrometry.

STERN

iments of Knoop (16). He showed, in 1905, that rabbits fed an o-phenyl even-carbon fatty acid excreted phenylacetic acid in their urine, whereas those fed an o-phenyl odd-carbon fatty acid excreted benzoic acid. Those observations led Knoop to propose that fatty acids were oxidized by removal of successive two-carbon fragments from the carboxyl end. We can now say that the fatty acid biosynthetic pathway is able to catalyze the reverse of only one of these two P-oxidation sequences; whereas the CoA ester of phenylacetic acid is a viable precursor for synthesis of w-phenyl even carbon acids, the CoA ester of benzoic acid cannot be used to form w-phenyl odd-carbon acids. McCune et aZ. (4) have proposed that the active form of certain hypolipidemic drugs, which can be metabolized to aromatic carboxylic acids, may be the CoA ester and TABLE

IV

CHAINLENGTHOFFAW~ACIDSSYNTHESIZEDAND RELEASEDFROMFATTYACIDSYNTHETASEINTHE PRESENCEOFTHIOESTERASE II mol

mation of the w-phenyl aliphatic moiety on the synthetase is slower than is the formation of the n-aliphatic moiety, the thioesterase enzymes operative in fatty acid synthesis apparently do not discriminate against w-phenyl aliphatic esters. Thus both thioesterases “see” the w-phenyl group as equivalent to a 4-carbon aliphatic chain; thioesterase I terminates acyl chain growth at either n-Cls or w-phenyl& and thioesterase II exhibits an equally broad specificity for n-aliphatic (C,&,-,) and wphenyl (w-phenyl-Cd-w-phenyl-C& moieties. The rate-limiting step in the synthesis of w-phenyl fatty acids appears to be at the level of chain initiation, either the transfer of the phenylacetyl moiety from CoA to the enzyme or the initial condensation of the phenylacetyl moiety with a malonyl moiety. Since no short-chain wphenyl acyl moieties accumulate on the enzyme, it is evident that subsequent elongation reactions, at least up to the wphenyl-Cl0 stage, occur relatively rapidly. Our experiments provide an interesting, yet belated, sequel to the classical exper-

Aliphatic chain length

n-fatty

acids

G

0

G

0 2 16

G GO Cl2

%

o-phenyl

fatty 8 24 12 13

15

25

Cl4

18

C 4:

25

7 11

GO

6 18

acids

0 0

Note. Incubation systems contained, in a volume of 0.5 ml: 0.1 M potassium phosphate buffer (pH 8.0), 250 rnrd NADPH, 96 PM [2-“C]malonyl-CoA (1.65 Ci/mol), 50 PM acetyl-CoA or phenylacetyl-CoA, 2.27 pM trypsinized fatty acid synthetase, and 102 pg thioesterase II. Reactions, at 37”C, were terminated after 36 s (acetyl-CoA primer) or 188 s (phenylacetyl-CoA primer). A previous experiment, in which the reaction rate was monitored spectrophotomerically, indicated that these incubation conditions were sufficient to permit the oxidation of 40 nmol of NADPH in both acetyl-CoAand phenylacetyl-CoA-primed reactions. Acyl-S-enzyme thioesters were removed by ethanol precipitation, as described under Experimental Procedures, and free fatty acids present in the supernatant extracted and identified.

EFFECT

OF

AROMATIC

CoA

ESTERS

that the CoA esters may inhibit specific lipogenic enzymes. Our results indicate that there may be some merit to this idea since benzoyl-CoA proved to be a good inhibitor of the fatty acid synthetase. However, the fact that some aromatic carboxylic acids such as phenylacetic acid may actually serve as substrates for fatty acid synthesis reveals that the situation is more complex than might have been anticipated. For example, it is possible that certain aromatic carboxylic acids could potentially give rise to unusual lipids with abnormal properties. Indeed it has already been shown that rat tissues will utilize the hypolipidemic compound ethyl 4-benzyloxybenzoate (BRL 10894) and synthesize a triglyceride containing the 4-benzyloxybenzoyl moiety (17). Whether any of these aromatic carboxylic acids can exert an effect on lipogenesis in viva may depend largely on how their metabolism is compartmentalized within the cell. Although the formation of hippurate from benzoate and glycine has been shown to take place in rat liver mitochondria (5) it remains unclear whether the formation of CoA esters of aromatic acids is confined to the mitochondrial compartment. For aromatic carboxylic acids to influence directly the de novo fatty acid synthesis pathway would necessitate access of the CoA thioesters of these compounds to the cytosolic compartment. ACKNOWLEDGMENTS We thank malonyl-CoA

Dr. Zafar Randhawa for purifying used in some of our experiments.

the

ON

FATTY

ACID

SYNTHETASE

265

REFERENCES 1. GLEASON, M. N., GOSSELIN, R. E., HODGE, H. C., AND SMITH, R. P. (1969) in Clinical Toxicology of Commercial Products, p. 228, Williams & Wilkins, Baltimore. 2. LEHNINGER, A. I. (1970) in Biochemistry, p. 442, Worth, New York. 3. JERVIS, G. A. (1947) J. Biol. Chem 169, 651-656. 4. McC~NE, S. A., DURANT, P. J., FLANDERS, L. E., AND HARRIS, R. A. (1982) Arch B&hem. Bie phys. 214. 124-133. 5. GATLEY, S. J., AND SHERRAIT, H. S. (19’77) Bioehem J. 166,39-47. 6. SMITH, S. (1981) in Methods in Enzymology (Lowenstein, J. M., ed.), Vol. 71, pp. 181-188, Academic Press, New York. in Enzymology (Low7. SMITH, S. (1981) in Methods enstein, J. M., ed.), Vol. 71, pp. 188-200, Academic Press, New York. 8. STERN, A., SEDGWICK, B., AND SMITH, S. (1982) J.

Biol. Chem 257, 799-803. 9. SMITH, S. (1982) Arch. B&hem Biophys. 218.249253. 10. METCALFE, L. D., AND SCHMITZ, A. A. (1961) AnaL Chem. 33, 363-364. 11. CORKEY, B. E., BRANDT, M., WILLIAMS, R. J., AND WILLIAMSON, J. R. (1981) Analytical B&hem 118,30-41. 12. KATIYAR, S. S., AND PORTER, J. W. (1975) Arch. B&hem Biophys. 170,220-227. 13. SMITH, S., AGRADI, E., LIBERTINI, L. J., AND DILEEPAN, K. N. (1976) Proc. Nat. Acad Sci USA 73.11841188. 14. SMITH, S. (1980) J. Dairy Sci. 63,337-352. 15. LIBERTINI, L. J., AND SMITH, S. (1978) J. Biol Chem. 253,1393-1401. 16. KNOOP, F. (1905) Beitr. Chem Physiol. Pathol. 6. 150-162. 17. FEARS, R., BAGGALEY, K. H., ALEXANDER, R., MORGAN, B., AND HINDLEY, R. M. (1978) J. Lipid Res. 19, 3-11.