Purification and properties of the fatty acid synthetase from Mycobacterium phlei.

Abstract The fatty acid synthetase from Mycobacterium phlei has been purified 340-fold to homogeneity. The enzyme has a molecular weight of 1.39 x 106. At low concentrations of phosphate buffer (0.005 m), the synthetase dissociates into an enzymatically inactive species (7.65 S) which can be partially reaggregated and reactivated by dialysis against 0.5 m potassium phosphate buffer. The mycobacterial polysaccharides, 3-O-methylmannose-containing polysaccharide (MMP) and 6-O-methylglucose-containing polysaccharide (MGLP), stimulate the fatty acid synthetase markedly. Their presence lowers the Km values for acetyl-CoA and malonyl-CoA 9-fold and 4-fold, respectively. The polysaccharides also appear to function by altering the rate-limiting step of fatty acid synthesis. MMP stimulates fatty acid synthesis more effectively than MGLP. Various chemical modifications of the polysaccharides do not markedly alter their stimulating activity. Acetyl-CoA is the most effective primer and its concentration affects the degree to which MMP and MGLP stimulate fatty acid synthesis. It is proposed that the polysaccharides function primarily by binding long chain acyl-CoA and thereby relieve product inhibition of the fatty acid synthetase.


From the James Bryant Conant Chemical
Laboratories, Harvard University, Cambridge, Massachusetts OWlS8 SUMMARY The fatty acid synthetase from Mycobacterium phlei has been purified 340-fold to homogeneity.
The enzyme has a molecular weight of 1.39 x 106. At low concentrations of phosphate buffer (0.005 M), the synthetase dissociates into an enzymatically inactive species (7.65 S) which can be partially reaggregated and reactivated by dialysis against 0.5 M potassium phosphate buffer.
Their presence lowers the K, values for acetyl-CoA and malonyl-CoA g-fold and 4-fold, respectively.
The polysaccharides also appear to function by altering the rate-limiting step of fatty acid synthesis. MMP stimulates fatty acid synthesis more effectively than MGLP. Various chemical modifications of the polysaccharides do not markkdly alter their stimulating activity. Acetyl-CoA is the most effective primer and its concentration affects the degree to which MMP and MGLP stimulate fatty acid synthesis.
It is proposed that the polysaccharides function primarily by binding long chain acyl-CoA and thereby relieve product inhibition of the fatty acid synthetase.
1\Iultienzyme complexes catalyzing the synthesis of long chain fatty acids from acetyl-CoA and malony-Cob have been isolated from various eucaryotic sources and their properties are known in great detail (1). That such multienzyme complexes can occur also in procaryotic organisms was demonst)rated by the presence of a fatty acid synthetase of high molecular weight in Jlycobacterium phlei (2). In most bacteria, fatty acid synthesis is catalyzed by individual, nonaggregating enzymes (3). Apart from comparative and phylogenet,ic aspects, the X. phlei fatty acid synthetase is of considerable interest because of its exceptional cofactor and substrate requirements, its relative instability, and the bimodal fatty acid pattern that this enzyme system produces (2,4,5). We now describe a procedure for purifying the J/r. phlei synthetase to homogeneity, conditions for dissociating and * This research was supported by grants-in-aid from the United States Public Health Service, the National Science Foundation, the Life Insurance Medical Research Fund, and the Eugene P. Higgins Trust Fund of Harvard University. regenerating the complex, and the influence of substrate and cofactor concentrations on enzyme activity.

EXPERIMENTAL PROCEDURE
Materials-Acetyl-CoA was synthesized by the method of Simon and Shemin (6). [2-14C]Malonyl-CoA was purchased from New England Nuclear.
TPNH, DPNH, and FMN were obtained from Calbiochem. All other acyl-CoA derivatives and DTTI were purchased from P. L. Hiochemicals, Inc., DEAEcellulose from the urown Company, a-amylase from Worthington Riochemicals, and IXo-Gel products from Bio-Rad Laboratories. Oyster glycogen, yeast mannan, and fat-free bovine serum albumin were purchased from Sigma, and DEAE-Sephades was obtained from Pharmacia. J1. phlei ATCC-356 cells were grown to stationary phase (48 hours) 011 a medium containing glucose and Tween-80 (2).

Isolation
of Polysaccharides-M. phlei cells (80 g) were suspended in 400 ml of distilled water and heated at 90" for 15 min. This extract was fractionated as previously described up to and including 13io-Gel P-10 chromatography (4). The misturc of polysaccharides obtained from the ljio-Gel P-10 column was chromatographcd on a DEAE-Sephadex column as described by Keller and Ballou (7). The MMP eluted with distilled water and the 6-O-methylglucose-containing polysaccharides QIGLP-I, -11, -III) were eluted with a 0 to 0.15 M NH4HC03 gradient. The separated polysaccharides were lyophilizcd and further purified by passage through a llio-Gel P-6 column (3 x 75 cm) with distilled water as eluting solvent.
After acid hydrolysis, carbohydrates were characterized and qualitatively identified by gas chromatography of the trimethylsilyl derivatives of the methyl glycosides as previously described (4). Polysaccharides wcrc quantitatively estimated with cr-naphthol reagent with n-glucose as a standard (8 TPNH, and 1 PM FNIN, and the final volume was 0.5 ml. Unless otherwise indicated, reaction mixtures contained 20 PM malonyl-CoA, 300 FM acetyl-CoA, 200 pg of crude polysaccharide (fraction after the Bio-Gel P-10 step), and 1 pg of enzyme. At low concentrations of enzyme (1 to 4 pg/O.5 ml), fatty acid synthesis was linear with time for at least 30 min, but all reactions were terminated after I5 min. One unit of activity is defined as the amount of enzyme required to incorporate 1 nmole of malonate per min into the fatty acids.
In some instances the results obtained by radioactive assay were checked and confirmed by the spectrophotometric assay described by Lynen (9) with the above concentrations of substrates and cofactors.
Chew&al Mod$cation of Polysaccharides-For deacylation, a sample of MGLP-II (5 mg) was dissolved in 1 ml of 0.5 N NaOH and allowed to stand at room temperature for 1 hr. The product (MGP), after desalting by passage over a Bio-Gel P-6 column, contained less than 0.4 mole of acyl residue per mole of MGP as judged by a hydroxamate assay (10). MGLP-II contains 7 moles of acyl residues per mole of polysaccharide (7, 11). MMP or MGLP was partially methylated according to the method of Falconer and Adams (12). Polysaccharide (1 mg) was dissolved in 0.5 ml of distilled water and 0.3 ml of dimethyl sulfate was added dropwise at 0", followed by 0.6 ml of 30% NaOH.
The mixture was stirred overnight at room temperature and the additions of dimethylsulfate and NaOH were repeated at this time and again after 8 hours. The reaction mixture was neutralized with 6 N HCl and the partially methylated polysaccharide was extracted 3 times with 4-ml portions of chloroform. Evaporation of the solvent yielded 0.8 mg of product. An SO-pg aliquot of the methylated polysaccharide was subjected to methanolysis and gas chromatography of the trimethylsilyl derivatives as previously described (4). MGP methylated by this procedure contained no glucose or A-O-CHB-glucose and methylated MMP contained neither mannose nor 3-O-CHsmannose, indicating that at least one additional methyl group had been introduced into each saccharide residue in the polysaccharide.
MGP was digested with cll-amylase to remove 3 sugar residues from the nonreducing end of the polysaccharide as described by Saier and Ballou (11). A 4.6-mg sample of MGP was dissolved in 1 ml of 0.09 M potassium phosphate, pH 7.0, containing 0.005 M NaCI.
cr-Amylase (1.75 mg) was added and, after 48 hours at room temperature, digestion was terminated by boiling the reaction mixture for 5 min; the precipitate was removed by centrifugation and discarded. The oc-amylasetreated MGP was desalted by percolation through a Bio-Gel P-6 column (34 x 3 cm), lyophilized, and dissolved in 1 ml of distilled water (yield 2.42 mg). A 125.pg sample was subjected to mcthanolysis and the methyl glycosides were analyzed by gas chromatography of the trimethylsilyl derivatives (4). By comparison of the areas of the 3-0-CHS-glucose and the glucose peaks to that of 6-0-CHB-glucose peak obtained from MGP and a!amylase-digested MGP, it was calculated that a-amylase digestion had removed the 3-0-CHs-glucose residue and 2 glucose residues from the polysaccharide.
PuriJication of Fatty Acid Synthetase-All steps were performed at O-4". Protein was determined by the method of Warburg and Christian (260 and 280 nm) (13). ,411 buffers contained 1 mM DTT and 1 mu EDTA.
Eighty grams of frozen M. phlei cells were thawed in 300 ml of 0.1 M potassium phosphate buffer, pH 7.0. The cells were broken by passage through a French pressure cell operated at 8,000 p.s.i., and the disrupted cells were centrifuged at 17,000 x g for 20 min. The resulting supernatant was centrifuged at 105,000 x g for 90 min and subsequently brought to 35% saturation with ammonium sulfate. After stirring for 15 min, the precipitate was removed by centrifugation at 37,000 x g for 20 min and discarded.
The supernatant was slowly brought to 557" saturation, stirred for 30 min, and centrifuged at 37,000 x g. The 35 to 5570 ammonium sulfate precipitate was dissolved in 0.1 M potassium phosphate buffer (pH 7.0), and 32.8 ml of a 1 y0 protamine sulfate solution (0.12 mg of protamine sulfate per mg of protein) were added dropwise while stirring.
After 30 min the solution was centrifuged at 37,000 x g and the precipitate discarded.
The supernatant was percolated onto a DEAEcellulose column (3 X 20 cm) that had been previously equilibrated with 0.1 ?rz potassium phosphate buffer, pH 7.0. The column was washed with 1 liter of 0.25 1\1 potassium phosphate buffer, pH 7.0. The enzyme was eluted by a linear gradient of 650 ml of 0.25 i\r potassium phosphate (pH 7.0) and 0.70 M potassium phosphate buffer (pH 7.0). The major fractions containing activity were concentrated to 50 ml on a IIiaflo apparatus with an XM-50 membrane.
The concentrated solution was slowly brought to 60yc saturation with ammonium sulfate and centrifuged at 37,000 x g for 20 min. The precipitate was dissolved in 2 ml of 0.5 M potassium phosphate, pH 7.0, and applied to a Bio-Gel A-5m column (70 X 2.8 cm) that had been equilibrated with the same buffer. The enzymatic activity eluted with the first protein peak from the column.
The collected peak fractions were combined and concentrated to 2.5 ml on a Diaflo apparatus with an X111-50 membrane.
The results of this purification are summarized in Table I.

RESULTS
Physical Properties of Fatty Acid Synfhefase-After 340.fold purification the enzyme was homogeneous as judged by centrifugation in a Beckman model E analytical ultracentrifuge (top of Fig. 1).
Sedimentation velocity studies of the purified enzyme in 0.5 M phosphate buffer gave an S value of 23.61. From this number, and using a Stokes radius of 108 A (a), a molecular weight of 1.39 x IO6 was calculated (14). A partial specific volume of 0.725 cm3 per g was assumed. During dialysis for 24 hours at 4" against 0.005 M potassium phosphate buffer, pH 7.0, containing 1 mM DTT and 1 mM EDTA, the enzyme complex dissociated into a smaller species with an S value of 7.65 (bottom of Fig. 1) corresponding to a molecular weight of 0.25 x 106. As shown in Fig. 2, the enzyme rapidly loses fatty acid synthetase activity on dialysis in 0.005 M phosphate with a half-time of less than 1 hour.
All activity had disappeared after 10 hours  Ten milligrams of enxyme in 1 ml of 0.5 M potassium phosphate buffer (pH 7.0) were dialyzed against 200 ml of 0.005 M potassium phosphate buffer, pH 7.0, for 26 hours at 4". At this time (indicated by the arrow) the dialysis buffer was changed to 0.5 M potassium phosphate and dialysis was continued for 12 hours. Aliquots of enzyme were removed at the times indicated and assayed for fatty acid synthetase activity as described under "Experimental Procedure." All buffers contained 1 mM DTT and 1 mM EDTA. The enzymatic assay was performed as described under "Experimental Procedure." The incubations without PS contained 140 pM malonyl-CoA and those with PS contained 80 PM malonyl-CoA. of dialysis.
Partial reversal of this inactivation was demonstrated as follows.
After 26 hours of dialysis, the buffer was changed to 0.5 M potassium phosphate, pH 7.0, containing 1 mM DTT and 1 mM EDTA.
Two hours later the enzyme had regained approximately 35% of the fatty acid synthesis activity; but continued dialysis did not result in a further increase.
When the reactivated enzyme was subjected to sedimentation velocity analysis in the ultracentrifuge, two species were detected corresponding to the active enzyme (23.1 S) and to the subunit (7.3 S).
EJects of Polysaccharides on Fatty Acid Synthesis-In an earlier report (4), we demonstrated that the polysaccharides lowered the apparent IL, for acetyl-CoA.
At that time, TPNH (600 PM) was the only source of reductant used in the assay mixture. Later it was found that DPNH as well as TPNH is required for optimal fatty acid synthesis (5). Under these conditions, the rate of malonyl-CoA incorporation was about 2.5 greater than with 600 PM TPXH alone (5). For this reason, the kinetic effects of polysaccharide on fatty acid synthesis as a function of acetyl-CoA concentration were reinvestigated with assay mixtures containing 30 PM each of DPNH and TPNH.
In the presence of polysaccharide, acetyl-CoA saturates the synthetase at concentrations between 400 and 600 PM (Fig. 3); further increases in acetyl-CoA concentration cause inhibition. Without polysaccharide, acetyl-CoA does not reach saturating levels until about 2000 PM. There is less than a X-fold difference in maximum velocities achieved with or without polysaccharide. However, the apparent K, for acetyl-CoA is lowered by the polysaccharide from approximately 800 PM to 90 PM.* These 2 The following difficulty arises in estimating Km values for a two-substrate reaction in the presence or absence of a modifier as in the present instance.
First of all, the concentration of one substrate (malonyl-CoA) affects the Km value for the other (acetyl-CoA, (4)). Second, the optimal malonyl-CoA concentrations are 140 PM and 80 PM, respectively, in the absence and presence of polysaccharide, the former (140 pM) being slightly inhibitory when the assay is done in the presence of polysaccharide.
Hence, we had two options : (a) to use the same malonyl-CoA concentrations K, values for acetyl-CoA in the Al. phlei system are unusually high compared to other fatty acid synthetase systems which generally operate with K, values for acetyl-CoA of less than 50 /.a1 (15). Earlier we reported (4) that polysaccharide lowers the K, for acetyl-CoA 50.fold, from 200 to 4 PM. These differences c5tn be attributed to modification in assay conditions, i.e. substitution of 600 PM TPNH by 30 PM each of DPNH and TPNH.
The effect of polysaccharide on the apparent I<, for malonyl-CoA is less striking than for acetyl-CoA (Fig. 4). The value is reduced from about 40 PM to 9.6 PM.~ Relative E$ectiveness of Polysaccharides-The apparent K a (concentration of oolvsaccharide at which one-half maximum for comparing K,,, values for acetyl-CoA in the presence and absence of polysaccharide, in which event the malonyl-CoA concentration was optimal in one situation and slightly inhibitory in the other, or (b) to use malonyl-CoA concentrations which are different but optimal for each of the two situations (plus or minus polysaccharide).
We chose the latter on the grounds that it is more meaningful to compare polysaccharide effects on Km values at optimal substrate concentrations.
The same arguments apply to the choice of conditions for the experiments of Fig. 4. is observed) for each of the polysaccharides was determined from a Lineweaver-Burk plot of the data shown in Fig. 5. In order to compare the effects of the various polysaccharides, fatty acid synthetase activity was plotted as a function of polysaccharide concentrations at saturating concentrations of acetyl-CoA and malonyl-CoA.
The apparent K, for MMP was estimated to be 8 /*RI (16.6 ,ug l)cr ml); for MGLPI, 23 PM (86 pg per ml) ; for MGLP-II, 34 PM (130 pg per ml) ; and for RIGLP-III, 30 FM (116 kg per ml). I)espite the marked structural diffcrences between MNP (16) and MGLP (II)., the magnitude of their effects on the rate of fatty acid synthesis under these conditions is not strikingly different.
The various sljccies of MGLP which differ only in their acyl content have closely similar activities but they arc all inferior to MW'.
At suboptimal concentrations of acetyl-CoA, the differences in stimulatory effects of 1\IMP and MGLP become greatly magnified (Fig. 6). Whereas at 300 PM acctyl-CoA stimulation b) &IGLP and M>lP is nearly the same, at 20 PM acctyl-CoA MGLP is less cffectivc than nII\Il' by a factor of 8. In addition, acetyl-CoA appears to influence the apparent K, for MMP.
From a Lineweavcr-Rurk plot of the data in Fig. 6, it is calculated that at 300 PM acetyl-CoA the K, for i\'IMl' is 20 pc11\1 and 1250 phi at 20 PM acetgl-Cod.
Acetyl-CoA appears to modify the complex, altering its affinity for polysaccharide.
EJect of MGLP on JlNP Stimulation-Fatty acid synthetase activity is roughly proportional to polysaccharide at low concentrations and the effects of MGLP and MMP are then additive (Fig. 7). Hut at saturating concentrations of MMP, stimulatiqn is not potentiated by MGLP.
If the two polysaccharidcs acted at different sites, a synergistic effect would be expected.
EQects ~j Polysaccharide Structure- Table  11 compares the effects of various natural and chemically modified polysaccharides on fatty acid synthesis.
Partial methylation of MMP with incorporation of at least one additional methoxy group per hexose residue did not alter its stimulating activity.
When MGLP was deacylated to MGP, stimulating activity declined only slightly.
Furthermore, removal of the 3 carbohydrate residues from the nonreducing terminal of MGP by cY-amylase did not diminish the effect on fatty acid synthesis.
On the other hand, partial methylation of MGP did sharply lower the capacity to stimulate the synthetase.
Glycogen, mannan, and While the structure of the native polysaccharides from J1. phlei can be substantially modified without elimination of stimulating activity, the results with glycogen and mannan demonstrate that certain structural features, including perhaps the hydrophobic methoxy group, are essential.
Primer Specijicity of Fatty Acid Synthesis-In view of the report by Lin and Kumar that butyryl-CoA is the preferred substrate for fatty acid synthesis in the mammary gland system (17)) various short chain acyl-CoA derivatives were tested as primers for fatty acid synthesis.
The results in Table III clearly demonstrate that acetyl-CoA is the most effective primer for the M. phl ei fatty acid synthetase, yet the stimulation by the polysaccharide is greater for the higher homologues.
In the range of 25 PM to 400 PM, butyryl-CoA and hexanoyl-CoA were essentially inactive unless polysaccharide was present.
Since acetyl-CoA is the only primer that is significantly active in the absence of polysaccharide, a special activating effect of acetyl-CoA on the synthetase is again indicated. Effect of Polysaccharide on Rate-determining Step in Fatty Acid Synthesis-In a study on an acetyl transferase, Riddle and Jencks (18) have demonstrated that "an observed Michaelis constant may reflect either a binding of substrate or a change in rate- determining step." A plot of fatty acid synthetase activity against malonyl-CoA concentration in the presence or absence of MMP at 900 ~11 acetyl-CoA (Fig. 8) suggests that polysaccharide may function similarly, i.e. by changing the step in fatty acid synthesis which is rate-limiting.
At concentrations of 4 PM or Less, malonyl-CoA activity is virtually the same with or without MMP.
It seems likely that at such low concentrations of malonyl-CoA, malonyl transacylase activity is rate-determining.
As the malonyl-CoA concentration is raised, the rate of fatty acid synthesis without MMP levels off, indicating that some other reaction has become limiting.
With MMP present in the incubation, the velocity increases almost linearly up to about 10 PM malonyl-CoA and above that concentration continues to rise with a much steeper slope than in the absence of MMP.
Following the arguments of Riddle and Jencks, we may interpret these data to show that the apparent K, for malonyl-CoA without MMP (5.3 PM) reflects a change in the rate-determining step in fatty acid synthesis from malonyl transacylase to another reaction (18). When MMP is present, the apparent K, (28.5 pM) may be equal to or closer to K,, the true dissociation constant for malonyl-CoA. Curves of similar type were obtained at 300 pM and 600 PM and state of aggregation, but whatever their nature, it is clear that the physiological milieus in which the three enzyme systems operate in the cell must differ vastly.
acetyl-CoA, and in these instances also the K, values for malonyl-Co-4 differed in the presence and absence of MMP.

DISCUSSIOx
The Jf. phlei fatty acid synthetase shares a number of properties with functionally analogous multienzyme complexes from other sources but displays certain features that are unique.
The molecular weight of this enzyme (1.39 x 106) falls within the range observed for other fatty acid synthetase complexes (0.5 to 2.5 x 106) (2). We had earlier reported a molecular weight of 1.7 y lo6 for the -II. phlei synthetase, a value based on gel filtration and sucrose density gradient centrifugation of the enzyme in 0.1 M phosphate buffer (2). We consider the present value (1.39 x 10") determined by sedimentatiorl-velocity analysis of the synthetase in 0.5 M phosphate to be the more reliable.
One of the distinctive properties of the M. phki synthetase is the unusually high K, for acetyl-CoA, a value about 20 times greater than reported for fatty acid synthetases from any other source. For comparative purposes we show corresponding values for the enzyme systems from other sources (Table IV).
The very dramatic reduction of the E;', values for acetyl-Coil in M. phlei by MMP and MGLP, with the consequence that they allow fatty acid synthesis to occur at more physiological substrate levels, has been reported previously (4). Polysaccharide also lowers the K, value for malonyl-CoA, but the reduction is of lesser magnitude.
While we still 1 k ac a complete explanation for these effects, we wish to record here several observations and some speculations which bear on the mode of action of these polysaccharides.
Dissociation of the M. phlei enzyme on exposure to media of low ionic strength affords a subunit of the same molecular weight (0.25 x 106) that has been observed for the fatty acid synthetases from animal tissues (19) and yeast (20). Dissociation and the concomitant inactivation of the JI. phlei synthetase are relatively rapid.
The enzyme partially reaggregates to the enzymatically active form but only in buffers of relatively high ionic strength (0.5 ~1 phosphate).
It is of interest to note that dissociation and reassociation do not alter the characteristic specificity of the enzyme, i.e. the bimodal fatty acid product pattern or the relative rates of total synthesis from acetyl-CoA and palmityl-CoA elongation to C22 and C&4 acids.3 Comparison of the conditions affording optimal activity of fatty acid synthetases from various sources shows striking differences of which susceptibility to ionic strength is one. Fig. 9 shows the response to ionic strength of three fatty acid synthetase complexes under investigation in this laboratory.
As noted, the synthetase from M. phlei is unstable and inactive in phosphate buffer below 0.1 M. A high molecular weight synthetase from Euglena gracilis is fully active in 0.1 M phosphate and rapidly inactivated as ionic strength is raised (21). Another bacterial multienzyme complex, isolated by H. Knoche from Corynebacterium diphtheriae (22), shows only slight activity in 0.1 M phosphate and is optimally active at 0.5 M phosphate.
What is being observed is probably a combination of effects on catalytic activity Examining several of the partial reactions catalyzed by the fatty acid synthetasc in order to localize the polysaccharidc effect, we found none that was significantly stimulated by MMP or LMGLP.~ It was also observed that these partial reactions are strongly inhibited by palmityl-CoA.
'The inhibitions are relieved by inclusion of BSA or of MMP or MGLP in t,he reaction mistures. l'hc H&4-like effects of the polysaccharides suggested that they might bind or complex palmityl-CoA and thercbg lower the effective concentration of this potent enzyme inhibitor. Such interactions between the mycobacterial polgsaccharides and palmityl-CoA have, in fact, been demonstrated by chromatography on Sephadex (23). If palmityl-CoA, which is one of the products of fatty acid synthesis, inhibited the system either by a generalized detergent effect or by negative feedback, then its removal due to sequestering with polysaccharides could result in an acceleration of over-all synthesis. Yet, as previously reported (a), BSA will not replace polysaccharide as a stimulatory agent for over-all fatty acid synthesis by the M. phlei system. This observation suggests that binding of palmityl-CoA or other long chain end products is not the only mechanism by which polysaccharides influence fatty acid synthesis in the N. phlti system.
An accompanying paper describes the effects of MMP and MGLP on the fatty acid synthetases from yeast and C. diphtherise (23). In these systems polysaccharide stimulation is significant but much less marked than in X. phlei.
Moreover, in these two systems, in contrast to JF. phlei, the polysaccharide effects are fully reproducible by BSA.
It would, therefore, appear that the mycobacterial polysaccharides play a species-specific role in regulating fatty acid synthesis in X. phlei.
If the polysaccharides function in 114. phlei also by binding long chain acyl-CoA and, therefore, relieving inhibition of fatty acid synthesis from acetyl-Coh, the question arises as to why HSA will replace the polysaccharides in the C. diphfheriae system and in yeast but not in M. phlei (23). The answer may lie in the unique ability of the X. phlei etlzyme to synthesize C2?-to C&oA derivatives as well as the Cl6 and Cl8 thioesters and to do so in a bimodal fashion.
As the acyl-CoA chain becomes longer, the molecule will become more hydrophobic.
Hence C&COB is less likely to diffuse from a lipophilic enzyme region to an aqueous environment than C16-Co-1. If the site for &CoA transacylation were not fully exposed on the enzyme surface, a large molecule such as bovine serum albumin (mol wt 69,000) might be unable to approach it, whereas this site might be more accessible to a smaller molecule (MMP, mol wt 2,100). As a result, polysaccharide and long chain acyl-CoA could interact to form a relatively hydrophilic complex which would more readily diffuse into the aqueous environment..
On the other hand, to explain the equivalence of MA and polysaccharide in their effects on the yeast and C. diphfheriae system (23), one need only assume that in these instances the end product (palmityl-CoA) diffuses from the enzyme at an appreciable rate. The complesing agents can then bind free palmityl-CoA and thereby 1)rotect against end product inhibition. The above schcmc for the mechanism of polysaccharide stimulation of the -l/. phlei fatty acid synthetase accounts for many of the experimental findings reported in this paper. For example, if the size of the binding molecule (polysaccharide or MA) were important for access to the enzyme site and hence for interaction with long chain acyl-CcA, and if the two polysaccharides were to bind the acyl-CoA derivatives with the same affinity, then one would esprct MMP (mol wt, 2100) to be more effective than MGLP (tiol wt 3700) for stimulating fatty acid synthesis. MMP does, indeed, have a h', value 3 to 4 times lower than NGLP.
It should be pointed out that the comparative effectiveness of MMP and MGLP is a function of acetyl-Coil concentration. The two polysaccharides stimulate fatty acid synthesis equally at high acetyl-CoA concentrations (300 PM), whereas at 20 PM acetyl-CoA, MMP is superior to MGLP by a factor of 8. One can rationalize these results by postulating that high acetyl-CoA concentrations induce a conformational change which makes the enzyme complex equally accessible to the two polysaccharidcs of different bize.
To account for thelowering of the K, for acetyl-CoA by polysaccharide, the following explanation is offered. If acetyl-Cob and &CoA compete for the same enzyme site and if this site has a greater affinity for C&CoA, then the apparent K, for acetyl-CoX will be a function of the C&d&oh concentration.
Polysaccharides, by binding t,he C&ZoA, will, therefore, lower the apparent K, for acetyl-CoA.
The biphasic chain length pattern of the end products, apart from posing special problems of regulation, makes it highly probable that the X. phlei synthetase serves multiple purposes. For example, it might ljrovide fatty acids not only for membrane phospholipids but also for the complex lipids of the mycobacterial cell envelope.
The membrane phospholipids of 11s. phlei contain mainly Cl6 and CM acids while the C& and CZ4 acids are probably b&ding blocks for the class of wall components known as mycolic a ids (24). lIevices are, therefore, needed for controlling the supply of shorter and longer chain synthetase products for the two separate biosynthetic pathways and it is attractive to implicate the mycobacterial polysaccharides in this regulatory role. Whether or not this speculation is correct, one further aspect of fatty acid synthesis in M. phlei is puzzling.
M. phlei contains, apart from the multienzyme complex under discussion (type I), a second fatty acid synthetase (type II) which is acyl carrier protein-dependent and functions strictly as an elongating system, converting palmityl-CoA or stearyl-CoA to C&2 and CU products (2, 25). Since the type I synthetase also produces these long chain acids, either by total synthesis or by elongating palmityl-CoA, the type II synthetase would appear to be redundant.
Another possibility is that the acyl carrier proteindependent elongating system arises artefactually from the multienzyme complex during the processing of the bacterial extracts. However, the experimental evidence so far does not favor this explanation.
Throughout purification of Synthetase I, the relative activities for total synthesis and palmityl-CoA elongation remain constant and no conditions have so far been found for either inactivating the complex differentially or dissociating it into functionally separate entities.