Proc. Nat. Acad. Sci. USAVol. 72, No. 5, pp. 1940-1944, May 1975

Presence of Two Polypeptide Chains Comprising Fatty Acid Synthetase(fatty acid synthetases of chicken and rat livers and yeast/molecular weight/size of subunit in denaturing solutions)

JAMES K. STOOPS, MICHAEL J. ARSLANIAN, YANG H. OH, KIRK C. AUNE, THOMAS C. VANAMAN*,AND SALIH J. WAKIL

Baylor College of Medicine, Marrs McLean Department of Biochemistry, Texas Medical Center, Houston, Texas 77025; and* Department of Microbiology, Duke University, Durham, North Carolina 27710

Communicated by Charles Tanford, February 19, 1975

ABSTRACT Highly purified fatty acid synthetases ofchicken and rat livers have molecular weights of 500,000and dissociate in solutions of low ionic strength into sub-units of molecular weight 250,000 with loss of synthetaseactivity. The subunits can be reassociated in phosphatebuffer with full restoration of the activity. In the presenceof sodium dodecyl sulfate or guanidine HCI, the syn-thetases dissociate into polypeptide chains of molecularweight 220,000 as determined by sodium dodecyl sulfate-gel electrophoresis and sedimentation equilibrium. Thepolypeptide contains the 4'-phosphopantetheine groupand the [14C]acetyl and [14C]malonyl groups if the syn-thetases were prelabeled with [14C]acetyl-CoA and [14C]-malonyl-CoA. Similar results were obtained with thesynthetase from yeast, except the subunit has a molecularweight of 200,000. These observations indicate that themulti-catalytic activities of the synthetases and the acylcarrier protein are associated only with the two poly-peptide chains. The findings suggest a novel structuralorganization for multienzyme complexes.

The palmitic acid synthesizing system isolated from Esche-richia coli consists of at least six separable enzymes and a pro-tein containing 4'-phosphopantetheine (acyl carrier protein)(1, 2). The fatty acid synthetases (FAS) of animal tissues andyeast, however, are multienzyme complexes having molecularweights of about 500,000 and 2,300,000, respectively (3-8).The synthetases of animal tissues and yeast have been disso-ciated into inactive subunits of 250,000 daltons (4-7, 9). Theenzymatic composition and the number of proteins comprisingthese subunits are not known. Lynen (10, 11) has proposedthat the yeast complex consists of seven individual enzymes.This conclusion was based on the finding of equal amounts ofseven different N-terminal amino acids and on the presence ofat least six different proteins in starch gels of the urea treatedenzyme. Consistent with the multiprotein complex hypothesiswas the isolation of a 16,000-molecular-weight peptide con-taining 4'-phosphopantetheine from guanidine- HCl (Gdn.HCl)-treated synthetase (12, 13). Similarly, the finding of fivedifferent N-terminal amino acids and eight protein bands inphenol/acetic acid/urea gels led Porter (14) to propose thatthe pigeon liver FAS consists of eight proteins. The multi-protein concept was further supported by the isolation of aprotein of 10,000 molecular weight containing the 4'-phospho-pantetheine from pigeon and dog synthetases (15-17).

In contrast to the model of an E. coli-like system in whichthe individual enzymes and the acyl carrier protein are tightlyheld together by noncovalent interactions, Schweizer et al.

(18) have proposed that the yeast complex is an aggregate oftwo polypeptide chains of 179,000 and 185,000 daltons.Contrary to the results of Willecke et al. (12), Gdn-HC1 didnot release a small peptide from the complex. Instead, sodiumdodecyl sulfate (NaDodSO4)-gel electrophoresis of panto-thenate-labeled protein indicated that the prosthetic groupwas associated with the 185,000 subunit.

In our laboratory we were able to separate several activitiesfrom the complex and to isolate a highly purified phospho-pantetheine-containing protein only after exposure of thechicken FAS to trypsin (19). Treatment of the complex withdenaturing agents (NaDodSO4 or Gdn- HCl) yielded manypeptide fragments, which led us to suspect that proteolysis ofthe synthetase had occurred during this treatment. Inactiva-tion of the protease made it possible to dissociate the complexin the presence of denaturing agents into two polypeptides ofequal molecular weight. In this paper we report the results ofstudies of the subunit size of fatty acid synthetase of animaland yeast origin using NaDodSO4-gel electrophoresis andsedimentation equilibrium analyses in Gdn* HCl.

MATERIALS AND METHODS

Enzyme Preparations. Chicken liver synthetase was pre-pared according to the procedure of Arslanian and Wakil (20)and had a specific activity of 1100-1500 nmol of NADPHoxidized per min/mg of protein at 250. Some preparationswere purified further by a DEAE Bio-Gel A step which wascarried out as follows: FAS (400 mg) which had been dialyzedovernight against a buffer containing 0.01 M potassium phos-phate (pH 7.4), 10 mM 2-mercaptoethanol (HS-EtOH), 1 mMEDTA, and 10% (v/v) glycerol was applied to a DEAE Bio-Gel A column (3.5 X 15 cm) equilibrated with the same buffer.The enzyme was eluted with a linear gradient consisting of 250ml each of the starting buffer and a buffer containing thesame components except 0.15 M phosphate was used.

[14C]Pantothenate-labeled FAS was prepared from chickensthat were injected intraperitoneally with 0.5 ml of [1-14C]-pantothenate (2.0 uCi, 10.3 Ci/mol). Injections were carriedout daily during the periods of starvation and refeeding (20).The FAS from rat liver was prepared by essentially the same

procedure and had a specific activity of 1000-1200 nmol/minper mg. The yeast FAS was prepared by a modification of theprocedure of Lynen (18, 21) and had a specific activity of1000-1200. A more detailed account of these procedures will bepublished elsewhere.NaDodSO4-Gel Electrophoresis. The preparations of the gels

and the electrophoresis were performed in a similar manner tothe procedure of Weber and Osborn (22). The 5% gel buffersolution contained 0.5% 3-dimethylaminopropionitrile and

Abbreviations: FAS, fatty acid synthetase; Gdn* HCl, guani-dine HCl; NaDodSO4, sodium dodecyl sulfate; HS-EtOH, 2-mercaptoethanol; Tos-LysCH2Cl, N-a-p-tosyl--lysine chloro-methyl ketone HCl.

Proc. Nat. Acad. Sci. USA 72 (1975)

0.2% ammonium persulfate. Protein samples of 50-100 Al in asolution of 0.1 M sodium phosphate (pH 7.0), 1 mM EDTA,1% HS-EtOH, 1-1.5% NaDodSO4, 10% glycerol, and con-taining bromophenol blue were layered on the gels. The elec-trophoresis was conducted for 6-7 hr at a constant current of 9mA per gel. Under these conditions the tracking dye traveled7-9 cm. The NaDodSO4-urea-gel method described bySchweizer et al. (18) was used for the electrophoresis of thesamples for a 24-hr period. Proteins were stained withCoomassie blue.NaDodSO4-Treatment of the Proteins. Before NaDodSO4

treatment, the enzyme was dialyzed overnight at 30 against asolution containing 0.1 M sodium phosphate (pH 7.0), 1.0mM EDTA, and 0.1 mM HS-EtOH (buffer A), or for 3 hragainst the same buffer at room temperature. For estimationof molecular weight, marker proteins and FAS were dissolvedin 0.1 M sodium phosphate (pH 7.0), 1 mM EDTA, and 1%HS-EtOH at a relative NaDodSO4:protein ratio of >2 andNaDodSO4 concentration of >1%. In order to achieve rapidheat inactivation of the protease in the synthetase prepara-tions, the enzyme was placed in a boiling-water bath for 1min within 5 sec after the addition of NaDodSO4. The volumeof this solution did not exceed 0.5 ml. When larger quantitiesof synthetase were required, the enzyme solution and theNaDodSO4 buffer were mixed in a funnel attached to an open-ended glass tube (4 mm inside diameter) having a coiled por-tion submerged in a boiling-water bath.

Treatment of the Synthetase with Alkylating Agents. DialyzedFAS (4.5 mg/ml) was treated with iodoacetate (10 mM),iodoacetamide (10 mM), or N-2-p-tosyl-ilysine chloromethylketone HCl (Tos-LysCH2CI) (1.3 mM) within 15 sec afterthe addition of NaDodSO4 (1.9%) to the protein. After 48 hrat 220, HS-EtOH was added to a final concentration of 1%and after an additional 48 hr of incubation, NaDodSO4-gelelectrophoresis was performed.

Preparation of the Synthetases for Sedimentation EquilibriumStudies in Gdn- HCl. The chicken synthetase was subjected toNaDodSO4 and heat as described above for large quantitypreparation. The protein was then dialyzed overnight againstbuffer A containing 1% NaDodSO4 and further dialyzedagainst 10 mM iodoacetamide in the same buffer for 8 hr. Theiodoacetamide was removed by dialysis against 0.1 M sodiumphosphate (pH 7.0), 1% NaDodSO4. The NaDodSO4 wasremoved by making this solution 8 M in urea and passing theresulting mixture through Dowex-1 Cl column (23). Finally,the protein solution was made 6 M with Gdn HCl by exten-sive dialysis. The synthetase from yeast was subjected directlyto 6M Gdn HCl containing 0.1 M HS-EtOH. The rat enzymewas alkylated with acrylonitrile (24) in 6 M Gdn-HCl andexhaustively dialyzed against 6 M Gdn HCl.

Ultracentrifugation Studies. The sedimentation experimentswere performed and the computation of molecular weight wascarried out as described previously (25, 26). The expecteduncertainty of molecular weights is approximately 6% of thedetermined values. The absorptivity of FAS was determinedby using the analytical ultracentrifuge (25). A value of 1.15ml/mg- cm at 280 nm was obtained at pH 8.3 with lysozymeused as a reference. The partial specific volume of FAS wastaken to be 0.74 cm3/g, as computed from the amino-acidcomposition (27).

Acetate and Malonate Binding to FAS. Aliquots (30 ,) ofacetyl-CoA or malonyl-CoA([1-14C]acetyl-CoA, 0.39 mM, 51Ci/mol, [1,3-_4C]malonyl-CoA, 0.61 mM, 33 Ci/mol) were

ab6c de f ghii

FIG. 1. NaDodSO4-gel (5%) pattern of FAS from rat andchicken livers and yeast. Electrophoreses of gels a to f wererun for 6-7 hr and the arrow indicates the approximate positionof the dye front; gels g to j were run for 24 hr. (a) and (b) 23 and46 uig of rat FAS (specific activity 1046 nmol/min per mg); (c)and (d) 20 and 50 jig of chicken FAS (specific activity 1180);(e) 5 jig of yeast FAS (specific activity 1500); (f) 32 jig of yeastFA8 (specific activity 1260); (g) 20 Aig of myosin; (h) 25 tg ofyeast FAS (specific activity 1260); (i) 25 jig of rat FAS (specificactivity 1000); (j) 25 jig of chicken FAS (specific activity 1180).

added to 30 MlI of dialyzed chicken synthetase (9.7 mg/ml) atroom temperature. After 1 min, 200 Ml1 of 1.7% NaDodSO4 inbuffer A and 30 jIl of iodoacetamide (0.1 M) were added.Labeling of the enzyme with acetate was also performed with-out subsequent alkylation by substituting 30 jil of H20 foriodoacetamide. The FAS of rat liver and yeast were labeled ina similar manner except that 60 and 100 Ml of the acyl CoAderivatives were added to 40 and 100 M~lof the rat (5.0 mg/ml)and yeast (1.3 mg/ml) enzymes, respectively. Aliquots of 100and 200,Ml of 2% NaDodSO4, 20 mM iodoacetamide in bufferA were added to the rat and yeast samples. After incubatingthe samples for 24 hr at 220, the chicken FAS was made 1% inHS-EtOH and after an additional 24 hr, NaDodSO4-gel elec-trophoresis was performed on 50 M~lof these samples. The gelswere stained, destained, and cut in 2-mm sections. The sec-tions were heated overnight at 7Q0 with 1 ml of 0.2 N NaGH insealed scintillation vials. After the solutions were acidifiedwith 0.2 ml of 2 M glacial acetic acid, 10 ml of Triton-toluenecounting solution (33% toluene, 67% Triton X-100 containing100 mg of 1,4-bis [2-(5-phenyloxazolyl) ]benzene and 5.5 g of2,5-diphenyloxazole per liter) were added. Scintillation count-ing was carried out using a Nuclear-Chicago Mark II LiquidScintillation System. Counting efficiency was determined bythe use of ['14C]toluene as internal standard.

RESULTS

Criteria of Purity of the Synthetases. The following resultsindicated that the FAS preparations were essentially ho-mogeneous and contained one protein component. The chickenand rat preparations had the highest specific activity reported-3300 and 2100 nmol of NADPH per mmn/mg at 380,Yrespectively. NaDodSO4-gels loaded with 5-50 jig of the yeast,chicken, and rat liver preparations showed the presence ofonly one major protein staining band (Fig. 1). Electrophoresisof these preparations for 24 hr caused the protein to migratehalfway through the gels without further resolution (Fig. 1).-

Proteolysis of the Chicken Synthetase in NaDodSO4. In our

studies of the dissociation of the FAS of chicken liver in

Fatty Acid Synthetase 1941

1942 Biochemistry: Stoops et al.

3 1.4

I4 4I.8

E0C-1

cs

1.2

1.0

0. .8

ab c d e f g hi a b c d e f g

FIG. 2. Effect of protein concentration and time of incubationin NaDodSO4 on NaDodSO4-gel (10%) pattern of chicken FAS.Synthetase (specific activity 1375) incubated at the indicatedconcentrations for 24 hr (a-f) and 96 hr (g-i) in a solution con-taining 2.2% NaDodSO4, 1% HS-EtOH and 0.1 M sodiumphosphate, pH 7.0. Each gel was loaded with 50 jAg of FAS oralbumin or both. (a), (b), and (c) contained FAS at 11.5, 2.3and 1.1 mg/ml, respectively; (d) albumin (5.0 mg/ml), (e) FAS(5.8 mg/ml), and albumin (5.0 mg/ml); (f) FAS (5.8 mg/ml);(g) albumin (5.0 mg/ml); (h) albumin (5 mg/ml) and FAS (5.8mg/ml); (i) FAS (5.8 mg/ml). The arrow indicates the approxi-mate position of the dye front.

FIG. 3. NaDodSO4-gel pattern of chicken FAS after treat-ment with heat or alkylating agents. Synthetase (specific activity1250) was heated or alkylated as described in Materials andMethods and 50 /Ag was loaded on 5% gels (b-e and g) or 10%gels (a, f). The synthetase was treated as follows: (a) no treat-ment; (b) iodoacetate; (c) iodoacetamide; (d) Tos-LysCH2Cl;(e) heat; (f) Tos-LysCH2Cl plus albumin (5 mg/ml); (g) FASsubjected to DEAE Bio-Gel A chromatography and dissociatedin Tris- glycine, pH 8.3, prior to treatment with heat. Thearrow indicates the approximate position of dye front.

NaDodSO4 we noted the occurrence of NaDodSO4-activatedproteolysis. The extent of proteolysis was dependent uponprotein concentration and time of incubation in NaDodSO4, asshown in Fig. 2. Within 24 hr of addition of NaDodSO4 to thesynthetase at a concentration of 11.5 mg/ml, the protein bandsmigrated near the dye front; however, at lower concentrationsof synthetase peptides of various sizes were obtained.Proteolysis was not limited to synthetase proteins, since albu-min underwent extensive degradation when added to syn-thetase treated with NaDodSO4 (Fig. 2, gels d through i). Inaddition, the NaDodSO4-activated proteolysis could bedetected by the method described by Mycek (28) whenalbumin was used as the substrate. Highly purified synthetasedoes not appear to undergo degradation in the native state.The FAS (11 mg/ml) that had been incubated for 24 hr in theabsence of NaDodSO4 did not show any detectable degrada-tion when analyzed by NaDodSO4-gel electrophoresis. Incontrast to the chicken enzyme, the NaDodSO4-gel patterns ofthe highly purified rat and yeast synthetases indicated thatthese two preparations contain little or no protease. Theabsence of NaDodSO4-activated proteolysis in the yeastFAS is consistent with the results of Schweizer et al. (18).

Partial removal of the protease from FAS was achieved bychromatographic procedures, but complete elimination of thisactivity could not be accomplished. However, treatment of thechicken liver FAS in NaDodSO4 with heat and the alkylatingagents iodoacetic acid, iodoacetamide, and the trypsin in-hibitor Tos-LysCH2Cl inactivated the protease (Fig. 3).Preparations treated, with these agents showed little or no

I II

0 t 0. ) C 0hicken ( X )

00

0.1 0. 2 0.3 0.4 0. 5 0. 6

RELATIVE MOBILITY

FIG. 4. Determination of molecular weights (mwt) of FASfrom chicken and rat livers and yeast. Protein samples (10 ,ug)were loaded on 5% NaDodSO4-gels. A myosin (canine heart),molecular weight 210,000; C albumin (dimer), molecular weight132,000; ® f,-galactosidase (E. coli), molecular weight 130,000;o albumin, molecular weight 66,000.

increase in the low-molecular-weight peptides resulting fromproteolysis (Fig. 3, compare gels a through e). Furthermore,inhibition of the protease prevented the degradation of al-bumin (Fig. 3, gel f). Iodoacetamide-treated FAS did notshow significant degradation upon storage at room tempera-ture for 7 days, whereas heat treatment of the enzyme did notprevent proteolysis during this incubation period.

Since the relative mobility of the major protein band was notaltered by these agents, it was concluded that these agentsdid not cause the protein to behave abnormally upon NaDod-S04-gel electrophoresis. This conclusion was further supportedby the results obtained with the rat liver synthetase, whichupon NaDodSO4-gel electrophoresis gave the same mobilitywhether or not it was subjected to these procedures.

Determination of the M1olecular Weight of FAS. Sedimenta-tion equilibrium studies of the chicken enzyme (0.2 mg/ml) inphosphate buffer (0.10 M, pH 7.4) containing 1 mM EDTAand 1 mM HS-EtOH gave a molecular weight value of480,000. Incubation of the chicken liver FAS in 5 mM Tris.glycine (pH 8.3), 1 mM EDTA, and 5 mM HS-EtOH for4-6 hr at 50 inactivated the enzyme. Sedimentation velocityanalysis showed that a peak sedimenting with a S20,uw of 13.5was transformed to a species with a 520,,of 8.5. Sedimentationequilibrium studies confirmed the apparent dissociation, sincethe molecular weight under the above condition was found tobe 240,000. The subunits reassociate into the dimer in 0.05 Mphosphate buffer with the regeneration of the original activity.It is of interest that the dissociated species gave the patternshown in Fig. 3, gel g, which is essentially the same as thatobtained for the native FAS. In contrast, the rat FAS (0.2mg/ml) in 0.1 M phosphate, pH 7.4, 1 mM EDTA andHS-EtOH yielded a molecular weight range of 250,000-400,000. Increasing the phosphate concentration to 0.5 Mresulted in a molecular weight range of 350,000-420,000,indicating that the monomer of molecular weight 250,000 is inequilibrium with the dimer. These results are essentially thesame as those reported by other workers for the avian and rat

enzymes (5-7). The native yeast FAS has been reported to

dissociate into smaller units of 500,000, which could be dis-sociated into subunits of 250,000 (9). The latter was the small-est subunit obtained that reassociated into the active enzyme.

Subunits of similar size were also obtained from the syn-

thetases after treatment with denaturing agents. As can be

Proc. Nat. Acad. Sci. USA 72 (1975)

Proc. Nat. Acad. Sci. USA 72 (1975)

1200

1000

E800

"3 600-0

0

< 400

200

38 57 79

ELUTION VOLUME (ml)

400F

E

a0

cJ

w

zcr

0coco

FIG. 5. Chromatography of ['4C] pantothenate-labeled chickenFAS on Sepharose 6B. Radio-labeled synthetase (7.5 mg, specificactivity 1330, radioactivity 1000 dpm/mg) was dissolved in 1.5ml of a solution containing 0.1 M sodium phosphate (pH 7.0),1% NaDodSO4, and 1 mM EDTA, heated, and alkylated withiodoacetamide as described in Materials and Methods. The samplewas then applied to the column and fractions were analyzed forabsorbance (0) and radioactivity (X). Another sample was pre-

pared in the same manner except that the heat and alkylationsteps were omitted. The sample was chromatographed on thesame column after incubation for 72 hr at 220. Absorbance (0),radioactivity (A). Vo is void volume.

seen in Fig. 1, NaDodSO4-gel electrophoresis of FAS indicatedthe presence of one protein component comprising over 95%of the protein. The values of the relative mobilities of thiscomponent obtained from NaDodSO4-gel electrophoresis ofthe synthetases are shown in Fig. 4. The molecular weightsestimated from this figure are 220,000, 220,000, and 200,000 forthe FAS proteins of chicken, rat, and yeast, respectively.When albumin and its aggregates corresponding to a molecu-lar weight range of 65,000-260,000 were used as markers (29),the estimated molecular weights for the synthetases were

10% higher than the values given above.In addition, molecular weight determinations of the FAS

by sedimentation equilibrium in Gdn HCl were carried out.Values of 240,000, 247,000, and 184,000 were obtained forthe chicken, rat, and yeast synthetases, respectively. Thesevalues are in good agreement with those obtained fromNaDodSO4-gel electrophoresis. No correction was made forpreferential solvation that may be occurring in 6 M Gdn * HCl.This correction has not been found to exceed 15% (30).

The Association of 4'-Phosphopantetheine with and theBinding of Acetyl and ilIalonyl Groups to FAS. Since the FASdissociates to a subunit of approximately 200,000 molecularweight, it was of interest to determine whether this subunitcarries the 4'-phosphopantetheine prosthetic group and thecatalytic sites that bind acetyl and malonyl groups. A prepara-

tion of chicken liver FAS labeled with [14C]pantothenate was

heat-treated and alkylated with iodoacetamide in NaDodSO4as described in Materials and Methods and chromatographedon Sepharose 6B (1.2 X 90 cm) equilibrated with 0.1 Msodium phosphate (pH 7.0), 0.1% NaDodSO4, 0.1 mMI HS-EtOH, and 10 mMI NaN3. As can be seen in Fig. 5, all theradioactivity was associated with the protein componenteluting near the void volume. The protein had a molecular

300

200

-n 100

E lo

CC

0

0

0

5)

800 L

600

400

200

2 4 6 8 10DISTANCE FROM ORIGIN (cm)

FIG. 6. Distribution of radioactivity and NaDodSO4-gel pat-terns of [14C] acetyl-FAS (- - -) and [14C]malonyl-FAS (0 *).In the upper graph the [14Clacetyl-FAS was not alkylated. Thearrows indicate the position of the dye front.

weight of 220,000 as determined by NaDodSO4-gel electro-phoresis. This result indicated that the prosthetic group 4'-phosphopantetheine is bound to the subunit, presumably bycovalent linkage to a serine residue (31, 32). Association of the[I4C]pantothenate with lower molecular weight proteins was

found only in NaDodSO4-treated synthetase when the pro-

tease was not inactivated (Fig. 5). The estimated molecularweights of these proteins were less than 65,000 and the Na-DodSO4-gel pattern was similar to that shown in Fig. 2, gel a.

The binding of acetyl and malonyl groups to the synthe-tases was also investigated. ['4C]acteyl- and ['4C malonyl-labeled chicken FAS were prepared and subjected to NaDod-S04-gel electrophoresis. Fig. 6 shows that the radioactivitywas associated only with protein of molecular weight of220,000. When the protease was not inactivated, the FAS was

degraded and numerous radiolabeled peptides were detected.The binding of ['4C]acetyl and ['4C]malonyl groups to therat and yeast FAS was also found to be associated onlywith the 220,000 and 200,000 molecular weight proteins.Amino Terminal Determination. By means of the dansyl

procedure with NaDodSO4-disrupted protein (33) only traceamounts of a-dansyl amino acids were detected in samplesequivalent to 0.5 mg (1 nmol) of chicken or rat fatty acidsynthetases. Quantitative amino terminal determinations,performed on chicken fatty acid synthetase by the cyanateprocedure (34) gave less than 0.5 moles of each amino acid per

460,000 g of protein after correction for destructive losses.These results indicate that chicken and rat fatty acid synthe-tases possess blocked amino termini.

DISCUSSION

Studies of the native complexes in the ultracentrifuge gave an

estimated molecular weight of 460,000-500,000 for the chickenand rat liver enzymes. Dissociation of the FAS in Tris - glycinebuffer produced a subunit of half the molecular weight whichcan be reassociated into the active native enzyme in phosphatebuffer. The molecular weight studies in NaDodSO4 and Gdn

S.--

I"1.1.1 KtTIKI.. \J

K

w

Fatty Acid Synthetase 1943

1944 Biochemistry: Stoops et al.

HCl indicate that this subunit consists of a single polypeptidechain of 220,000 daltons. Consistent with this observation isthe finding that the 4'-phosphopantetheine is covalentlybound to the 220,000 subunit and that no small-molecular-weight protein containing this prosthetic group could beisolated without prior proteolysis (compare Fig. 5). Moreover,the binding of [14C]acetyl and ['4C]malonyl groups to thesynthetases, presumably by acylating the transacylases, thecondensing enzyme, and the pantetheine components of FAS(3, 13, 35), followed by NaDodSO4-gel electrophoresis gaveonly one peak of radioactivity, corresponding to the 220,000molecular weight species. It can be concluded from theseresults that the synthetase consists of two polypeptide chainsof similar size. Whether these chains are identical or notremains to be determined.

Similarly, the yeast FAS has been reported to reversiblydissociate into subunits of 250,000 (9). Our studies of thisenzyme in NaDodSO4 and Gdn- HCl indicated that the sub-unit consists of a single polypeptide chain of 185,000-200,000molecular weight. This polypeptide binds both acetyl andmalonyl groups, and according to Schweizer et al. (18) itcontains the 4'-phosphopantetheine. It was further reported(18) that the yeast synthetase consists of two nonidenticalsubunits of molecular weights 179,000 and 185,000 based ongenetic and NaDodSO4-gel electrophoresis studies. Eventhough our preparation of the yeast synthetase gave only oneband on NaDodSO4-gel electrophoresis (Fig. 1), our conclusionregarding the size (about 200,000 daltons) of the subunitcomprising the yeast complex is essentially the same.Our findings and those of Schweizer (18) are in variance

with the concept that the FAS of animal and yeast origin areaggregates of enzymes and acyl carrier protein held togetherby noncovalent interactions. The exposure of the FAS toproteolysis during the various manipulations may explain whyothers have been able to detect the presence of many proteinsand to isolate a pantetheine-containing polypeptide of molec-ular weight of 10,000-16,000 (10-14, 16, 17).

Since the synthetase catalyzes seven different reactions andfive or six of the activities are present in the subunit afterdissociation in low ionic strength buffers, it was concludedthat the polypeptide having molecular weight of 220,000contains multiple catalytic sites. This type of structurewould provide the cell with a highly integrated and efficientsystem for multistep metabolic processes. Also, the synthesisand assembly of this multienzyme system is highly proficient.In the case of fatty acid synthetase, this assembly may assurethe formation of palmitate and eliminate the production ofshort chain fatty acids. This type of organization may notbe confined to fatty acid synthetase but may involve othermultienzyme systems (36, 37).

The authors wish to thank Dr. Y. Reddy for his generous giftof canine heart myosin. This work was supported in part byGrant GM 19091 from the National Institutes of Health andGrant Q-587 from the Welch Foundation.

1. Wakil, S. J. (1970) in Lipid Metabolism, ed. Wakil, S. J.(Academic Press, New York), pp. 1-48.

2. Prescott, D. J. & Vagelos, P. R. (1972) Advances in Enzy-mology, ed. Meister, A. (Interscience, New York), Vol. 36,pp. 269-311.

3. Joshi, V. C., Plate, C. A. & Wakil, S. J. (1970) J. Biol.Chem. 245, 2857-2867.

4. Kumar, S., Muesing, R. A. & Porter, J. W. (1972) J. Biol.Chem. 247, 4749-4762.

5. Yun, S. & Hsu, R. Y. (1972) J. Biol. Chem. 247, 2689-2698.6. Smith, S. & Abraham, S. (1970) J. Biol. Chem. 245, 3209-

3217.7. Burton, D. N., Haavik, A. G. & Porter, J. W. (1968) Arch.

Biochem. Biophys. 126, 141-154.8. Lynen, F. (1961) Fed. Proc. 20, 941-951.9. Sumper, M. & Lynen, F. (1972) Colloq. Ges. Biol. Chem. 23,

365-394.10. Lynen, F. (1967) in Progress in Biochemical Pharmacology,

eds. Kritchevsky, D., Paoletti, R. & Steinberg, M. D.(Karger, Basel, New York), Vol. 3, pp. 1-31.

11. Lynen, F. (1964) in New Perspectives in Biology, ed. Sela,M. (Elsevier, Amsterdam), Vol. 4, pp. 132-146.

12. Willecke, K., Ritter, E. & Lynen, F. (1969) Eur. J. Bio-chem. 8, 503-509.

13. Schweizer, E., Willecke, K., Winnewisser, W. & Lynen, F.(1970) in Vitamins and Hormones, eds. Harris, R. S.,Munson, P. L. & Diczfalusy, E. (Academic Press, NewYork), Vol. 28, pp. 329-343.

14. Yang, C. P., Butterworth, P. H. W., Bock, R. M. & Porter,J. W. (1967) J. Biol. Chem. 242, 3501-3507.

15. Lornitzo, F. A., Qureshi, A. A. & Porter, J. W. (1974)J. Biol. Chem. 249, 1654-1656.

16. Qureshi, A. A., Lornitzo, F. A. & Porter, J. W. (1974)Biochem. Biophys. Res. Commun. 60, 158-165.

17. Roncari, D. A. K. (1974) J. Biol. Chem. 249, 7035-7037.18. Schweizer, E., Kniep, B., Castorph, H. & Holzner, U.

(1973) Eur. J. Biochem. 39, 353-362.19. Arslanian, M., Stoops, J., McClure, J., Barakat, H. &

Oh, Y. (1973) Fed. Proc. 32, 633.20. Arslanian, M. J. & Wakil, S. J. (1975) in Methods of Enzy-

mology, ed. Lowenstein, J. M. (Academic Press, New York),Vol. 35, pp. 59-65.

21. Lynen, F. (1969) in Methods in Enzymology, ed. Lowenstein,J. M. (Academic Press, New York), Vol. 14, pp. 17-33.

22. Weber, K. & Osborn, M. (1969) J. Biol. Chem. 244, 4406-4412.

23. Weber, K. & Kuter, D. J. (1971) J. Biol. Chem. 246, 4504-4509.

24. Seibles, T. S. & Weil, L. (1967) in Methods in Enzymology,ed. Hirs, C. H. W. (Academic Press, New York), Vol. 11,pp. 204-206.

25. Aune, K. C. & Timasheff, S. N. (1971) Biochemistry 10,1609-1617.

26. Kar, E. G. & Aune, K. C. (1974) Anal. Biochem. 62, 1-18.27. Cohn, E. J. & Edsall, J. T. (1943) Proteins, Amino Acids,

and Peptides (Reinhold Publishing Corp., New York).28. Mycek, M. J. (1970) in Methods in Enzymology, eds.

Perlmann, G. E. & Lorand, L. (Academic Press, NewYork), Vol. 19, pp. 285-315.

29. Dunker, A. K. & Rueckert, R. R. (1969) J. Biol. Chem.244, 5074-5080.

30. Hade, E. P. K. & Tanford, C. (1967) J. Amer. Chem. Soc.89, 5034-5040.

31. Vanaman, T. C., Wakil, S. J. & Hill, R. L. (1968) J. Biol.Chem. 243, 6420-6431.

32. Roncari, D. A. K., Bradshaw, R. A. & Vagelos, P. R.(1972) J. Biol. Chem. 247, 6234-6242.

33. Stark, G. R. (1967) in Methods in Enzymology, ed. Hirs,C. H. W. (Academic Press, New York), Vol. 11, pp. 125-138.

34. Weiner, A. M., Platt, T. & Weber, K. (1972) J. Biol.Chem. 247, 3242-3251.

35. Phillips, G. T., Nixon, J. E., Abramovitz, A. S. & Porter,J. W. (1970) Arch. Biochem. Biophys. 138, 357-371.

36. Vernon, M., Falcoz-Kelly, F. & Cohen, G. N. (1972) Eur.J. Biochem. 28, 520-527.

37. White, R. C. & Nelson, T. E. (1974) Biochim. Biophys.Acta 365, 274-280.

Proc. Nat. Acad. Sci. USA 72 (1975)