Biosynthesis in Escherichia coli fo sn-glycerol 3-phosphate, a precursor of phospholipid.

Homogeneous biosynthetic sn-glycerol-3-phosphate dehydrogenase (EC 1.1.1.8) of Escherichia coli was potently inhibited by palmitoyl-CoA and other long chain acyl-CoA thioesters. The concentration dependence of this inhibition was not cooperative. Enzyme activity was inhibited 50% at 1 PM palmitoyl-CoA; thus, this inhibition occurred at concentrations below the critical micellar concentration of palmitoyl-CoA. PalmitoylCoA was a reversible, noncompetitive inhibitor with respect to both NADPH and dihydroxyacetone phosphate. Palmitoyl-CoA did not affect the quaternary structure of the enzyme. This inhibition could be prevented or reversed by the addition of phospholipid vesicles prepared from E. coli phospholipids. PalmitoylCoA did not alter the kinetics of inhibition by sn-glycerol 3-phosphate, which is a proven physiological regulator of this enzyme. Decanoyl-CoA, dodecanoyl-CoA, myristoyl-CoA, palmitoyl-(la’-etheno)CoA, stearoylCoA, and oleoyl-CoA inhibited sn-glycerol-3-phosphate dehydrogenase at concentrations below their critical micellar concentrations. Palmitate inhibited sn-glycerol-3-phosphate dehydrogenase activity 50% at 200 PM. Palmitoyl-carnitine, deoxycholate, taurocholate, and dodecyl sulfate were more potent inhibitors than Triton X-100, Tween-20, or Tween-80. Palmitoyl-acyl carrier protein at concentrations up to 50 PM had no effect on sn-glycerol-3-phosphate dehydrogenase activity. The possible physiological role of long chain fatty acyl-CoA thioesters in the regulation of sn-glycerol 3phosphate and phospholipid biosynthesis in E. coli is discussed.

Palmitoyl-carnitine, deoxycholate, taurocholate, and dodecyl sulfate were more potent inhibitors than Triton X-100, Tween-20, or Tween-80. Palmitoyl-acyl carrier protein at concentrations up to 50 PM had no effect on sn-glycerol-3-phosphate dehydrogenase activity. The possible physiological role of long chain fatty acyl-CoA thioesters in the regulation of sn-glycerol 3phosphate and phospholipid biosynthesis in E. coli is discussed.
The biosynthesis of sn-glycerol 3-phosphate (glycerol-P), which is required for phospholipid biosynthesis in Escherichia coli, occurs in a single step from the glycolytic intermediate, dihydroxyacetone phosphate (dihydroxyacetone-P). The enzyme which catalyzes the NAD(P)H-dependent reduction of dihydroxyacetone-P to produce glycerol-P is the biosynthetic glycerol-P dehydrogenase (EC 1.1.1.8). The biosynthetic role for the enzyme was established by the isolation of glycerol (1) and glycerol-P (2) auxotrophs deficient in glycerol-P dehydrogenase activity. Since glycerol-P was a potent inhibitor of the * The costs of publication of this article were defrayed in part by the payment of page charges. This article must therefore be hereby marked "nduertisement" in accordance with 18  it was suggested that glycerol-P might be a feedback inhibitor of glycerol-P biosynthesis in vivo (3). The physiological significance of the inhibition by glycerol-P was established by the isolation of strains bearing a mutation within the structural gene for the glycerol-P dehydrogenase, gpsA (4). Crude extracts of the mutant strains contained glycerol-P dehydrogenase activity about lo-fold less sensitive to glycerol-P inhibition than the parental activity (4). Glycerol-P dehydrogenase has been purified to homogeneity from the parental and feedback-resistant strains (5). Comparison of the kinetic properties of the homogeneous enzymes demonstrated that the property of feedback resistance was inherent to the mutant dehydrogenase and indicated that regulation of the enzyme by glycerol-P occurred at an allosteric site (6).
Although the regulation of glycerol-P dehydrogenase by glycerol-P has been extensively investigated in vitro (3,5,6) and its physiological significance has been established (4), little information exists regarding other mechanisms which may regulate glycerol-P biosynthesis. Kito and Pizer reported that glycerol-P dehydrogenase activity present in crude extracts and in partially purified preparations (3,7) was potently inhibited by palmitoyl-coenzyme A. Numerous mammalian (8)(9)(10)(11)(12) and microbial enzymes (3,13) are inhibited in vitro by palmitoyl-CoA, but the physiological significance of t,his inhibition has been questioned because of the inherent detergent properties of this compound (14,15) and its irreversible inhibition of some enzymes (8). These criticisms are especially relevant for enzymes which exhibit inhibition at concentrations of palmitoyl-CoA in excess of its critical micellar concentration. Another criticism has stemmed from the absence of teleonomic rationales for the regulation of some enzymes which are strongly inhibited in vitro by this compound (8). The physiological significance of inhibition by palmitoyl-CoA remains to be established.
In this paper, we report investigations on the inhibition by palmitoyl-CoA of homogeneous, biosynthetic glycerol-P dehydrogenase from E. cob. In an attempt to evaluate the physiological significance of such inhibition with respect to glycerol-P synthesis, we have investigated the concentration dependencies, mode of inhibition, and reversibility of inhibition by palmitoyl-CoA.
The effects of palmitoyl-CoA on the sedimentation coefficient of glycerol-P dehydrogenase and on the inhibition by glycerol-P were investigated. was fixed in these studies; v-' was plotted against S-' at several fixed levels of inhibitor.
The data were reduced by least squares analysis, and in all cases, correlation coefficients were 0.99 or greater. Either the slopes or l/u intercepts derived from least squares analysis varied with Z, depending on whether inhibition was competitive, noncompetitive, or uncompetitive. Replots of these slopes and intercepts against Z were used to determine K,, and K,,. CoA-The inhibition of homogeneous glycerol-P dehydrogenase by palmitoyl-CoA is shown in Fig. 1 3. Effect of palmitoyl-CoA on SK,,,,, of glycerol-P dehydrogenase. Centrifugation in sucrose density gradients was performed as described under "Experimental Procedures" using 10 pg of enzyme. A, gradient without palmitoyl-CoA. B, gradient containing 20 pM palmitoyl-CoA. Gradient fractions were assayed for glycerol-P dehydrogenase activity after IOO-fold dilution into assay mixtures. Glycerol-P dehydrogenase activity is plotted along with the positions of markers: catalase, A; yeast alcohol dehydrogenase, B; and ovalbumin, c.

Inhibition
immediate. In the absence and presence of 1.0 JI~M palmitoyl-CoA, constant activity was observed for at least 8 min, and the initial rate was proportional to the amount of enzyme employed over the range 0 to 100 ng. The amount of inhibition observed was independent of whether the reaction was initiated by addition of either substrate or with the enzyme. When glycerol-P dehydrogenase was incubated at 4°C for 1 min in the presence of 20 PM palmitoyl-CoA, a loo-fold dilution into the assay mix resulted in activity equal to that observed when an equivalent amount of enzyme was assayed with 0.2 ,uM palmitoyl-CoA.
Other preincubations of enzyme at 0 or 23°C for 5 min and with 2, 4, or 5 PM palmitoyl-CoA established that inhibition was fully reversible by loo-fold dilution. The inhibition of glycerol-P dehydrogenase by several concentrations of palmitoyl-CoA was investigated by varying the concentration of one substrate and keeping the concentration of the other fixed. At saturating NADPH (0.1 mM), palmitoyl-CoA was a noncompetitive inhibitor with respect to dihydroxyacetone-P (Fig. 24). The replots of the slopes or intercepts derived from these data were linear functions of palmitoyl-CoA concentration (Fig. 2B). The inhibitor constants, K,, and K,,, derived from these replots were 1.1 and 0.7 IJM, respectively. At 0.4 mM dihydroxyacetone-P, palmitoyl-CoA was a noncompetitive inhibitor with respect to NADPH (Fig. 2C). The replots of the slopes or intercepts derived from these data were linear functions of palmitoyl-CoA concentration (Fig.  20). The K,, and K,, derived from these replots were 1.5 PM and 0.2 FM, respectively. Since palmitoyl-CoA was a noncompetitive inhibitor with respect to either substrate, the inhibitor appears to be able to bind all enzyme-substrate intermediates and free enzyme. Previous kinetic data indicated the kinetic mechanism required NADPH to bind prior to dihydroxyacetone-P (6). Since the K,, with respect to NADPH was considerably lower than the other K,, and the two KcL values, the free enzyme appears to have higher affinity for palmitoyl-CoA than the enzyme-substrate intermediates. Since several enzymes subject to inhibition by palmitoyl-CoA have been demonstrated to dissociate in its presence (8,  3-4 (9) 4.5 (14) 2 (14) zreater than 60 PM (14) 200 (24) 2000 (25) 10,ooo (25) 200 (25) 1000 (25) 0.06 mg/ml (25) 0.085 mg/ml (25) a References are provided for the reported critical micellar concentrations.
* The modification of assay conditions by addition of 0.2 M NaCl or 10% methanol inhibited glycerol-P dehydrogenase by 30 and 60%, respectively. 12, 13), the effect of palmitoyl-Cot\ upon the ~20,~ of glycerol-P dehydrogenase was investigated by centrifugation in sucrose gradients. The szo,w in the absence of palmitoyl-CoA was 4.2 S (Fig. 3), in agreement with the previously reported value (5). The presence of 20 pM palmitoyl-CoA did not alter the szo,w (Fig. 3). Hence, the addition of palmitoyl-CoA did not cause dissociation or aggregation of the enzyme. The recovery of glycerol-P dehydrogenase activity after 20 h at 4'C was approximately 75% from the gradient containing 20 PM palmitoyl-CoA and from the control. Irreversible inhibition of the enzyme by palmitoyl-CoA did not occur. The similar recoveries of activity indicate that palmitoyl-CoA did not produce an inactive species with an altered sedimentation coefficient.  Fig. 4. C&oA, C&oA, C&CoA, and Cls:I-CoA caused substantial inhibition at levels less than 5 pM. Glycerol-P dehydrogenase activity was quantitatively inhibited at 20 pM concentrations of each of these acyl-CoA derivatives except C&oA.
In all cases, the concentration dependence of the inhibition was noncooperative and appeared rectilinear.
The potency of the long chain acyl-CoA thioesters employed appeared to be inversely related to their critical micellar concentrations (Table I). This generalization was also valid for palmitoyl-&oA2 (Fig. 1). When the concentration dependency of the inhibition of palmitoyl-CoA inhibition was investigated in assays containing 10% methanol or 0.2 M NaCl to alter the critical micellar concentration of the palmitoyl-CoA, the concentration dependencies were essentially unchanged ( Fig. 1; Table I). When assays were conducted in buffers of lower ionic strength, 10 mM Tris-HCl (pH 7.4) and 10 mM potassium phosphate (pH 7.4), the concentration dependence of pahnitoyl-CoA inhibition was unaltered. CoA, C2-CoA, C&oA, C&oA, and Cs-CoA did not inhibit the glycerol-P dehydrogenase at concentrations up to 20 PM, while 20 pM C&oA inhibited 30%. The combination of 20 pM CoA and 20 pM palmitate was not inhibitory.
Other amphiphilic molecules were tested to define the specificity of the inhibition by acyl-CoA thioesters. The effect of palmitoyl-ACP was investigated, because it is a principal product of de novo fatty acid biosynthesis in E. coli (26) and serves as an acyl donor in vitro for the biosynthesis of phospholipids (20,26). Neither palmitoyl-ACP nor ACP at concentrations up to 50 pM (Table I) had any effect on glycerol-P dehydrogenase activity. Palmitate produced detectable inhibition at levels greater than 50 pM (Fig. 5); 200 pM palmitate caused 50% inhibition.
The anionic detergents, deoxycholate and dodecyl sulfate, were potent inhibitors of glycerol-P dehydrogenase activity (Fig. 5). Fifty per cent inhibition occurred at 20 pM dodecyl sulfate and at 600 pM deoxycholate while nearly total inhibition occurred at 45 pM and 4 mM, respectively. When enzyme was preincubated at 4 or 23°C for 2 min in the presence of 45 pM dodecyl sulfate, greater than 95% of the glycerol-P activity was recovered after a loo-fold dilution. Palmitoyl-carnitine inhibited glycerol-P dehydrogenase activity 50% at 30 pM (Fig. 5). Substantial inhibition by deoxycholate and dodecyl sulfate occurred below the reported * The ECOA derivative of palmitoyl-CoA was employed as a test of the specificity of inhibition by palmitoyl-CoA (9). This derivative would be expected to have a slightly lower critical micellar concentration than palmitoyl-CoA. The similarity of the inhibition by palmitoyl-CoA and palmitoyl-eCoA demonstrated that the mechanism of inhibition is not sensitive to modification of the adenine ring of CoA.  (Table I). This was not the case for nonionic detergents like Triton X-100 (Fig. 5), Tween-20, or Tween-80 (Table I).
Effect of Phospholipid Vesicles on Palmitoyl-CoA Inhibition of Glycerol-P Dehydrogenase-Since long chain acyl-CoA compounds associate spontaneously with membranes or phospholipid vesicles in vitro (27,28), significant amounts of these compounds would be expected to be associated with membranes in vivo. In order to simulate the conditions likely to exist in vivo, the effect of phospholipid palmitoyl-CoA. The concentration dependency of the inhibition by palmitoyl-CoA was determined in the presence and absence of 20 pM phospholipid vesicles; the addition of vesicles diminished by approximately 60% the inhibition at every concentration of palmitoyl-CoA tested (Fig. 6B). This result was independent of the order of addition of vesicles and palmitoyl-CoA.
At palmitoyl-Cob levels of 10 and 20 PM, the activity was quantitatively inhibited (Fig. l), but the addition of 50 and 100 pM phospholipid vesicles, respectively, restored 50% of the enzyme activity (data not shown). Phospholipid vesicles reduced the inhibition by palmitoyl-CoA present at levels below or above the critical micellar concentration.
Thus, vesicle-associated palmitoyl-Cob was a less potent inhibitor than the free compound.
Experiments employing bovine serum albumin or cyclodextrin to sequester palmitoyl-CoA and, thereby, to lessen its inhibitory effect were undertaken. Unfortunately, both of these materials inhibited glycerol-P dehydrogenase activity at the levels required to bind palmitoyl-CoA.
Effect of Palmitoyl-CoA on Glycerol-P Inhibition of Glycerol-P Dehydrogenase-The available kinetic, physiological, and genetic data indicate that glycerol-P synthesis is regulated in vivo by glycerol-P inhibition of the glycerol-P dehydrogen-ase (4)(5)(6). The effect of 2 pM palmitoyl-CoA on the kinetics of glycerol-P inhibition was investigated. Glycerol-P was a competitive inhibitor with respect to dihydroxyacetone-P in the absence (Fig. 7A) and presence of 2 pM palmitoyl-CoA 7B). Since the Ki values for glycerol-P in the absence and presence of palmitoyl-CoA were 5 and 6 PM, respectively (Fig.  7C), the two inhibitors, glycerol-P and palmitoyl-CoA, appeared to act independently. DISCUSSION While the mechanisms which regulate the biosynthesis of membrane phospholipids in Escherichia coli remain incompletely understood, significant progress has been made recently by combined biochemical, genetic, and physiological approaches (29). Such investigations have established the biosynthetic role of the glycerol-P dehydrogenase (2) and the physiological significance of feedback inhibition by glycerol-P (4-6). The objective of the present investigations was to evaluate by kinetic studies of homogeneous glycerol-P dehydrogenase whether palmitoyl-CoA, another biosynthetic precursor of phospholipid, may function as a modulator of glycerol-P synthesis. Recent investigations have suggested that palmitoyl-Cob may regulate the activities of several lipogenic enzymes in both microbial (13) and mammalian systems (9)(10)(11). At present, the physiological significance of palmitoyl-CoA inhibition has not been established for any of the enzymes investigated.
Homogeneous glycerol-P dehydrogenase is potently inhibited by palmitoyl-CoA and other long chain acyl-CoA thioesters (Figs. 1 and 4). This demonstrates that the inhibition, noted by previous workers for activity in crude extracts (7) and in partially purified preparations (3), is an inherent property of the enzyme. Long chain fatty acyl-CoAs by virtue of their inherent detergent properties have the potential to perturb enzyme structure nonspecifically.
The general mechanism of inhibition by detergents occurs through cooperative binding at concentrations equal to or greater than the critical micellar concentration.
Since the concentrations of acyl-CoA thioesters which cause significant inhibition of glycerol-P dehydrogenase activity are less than their respective critical micellar concentrations (Table I)," inhibition may be mediated by specific binding. 4 The concentration dependencies of inhibition by all acyl-CoA thioesters tested were not cooperative ( Figs. 1 and 4), when both NADPH and dihydroxyacetone-P were saturating. The conclusion that palmitoyl-CoA inhibition was noncooperative can be extended to include various concentrations of NADPH and dihydroxyacetone-P (Fig. 2). The findings that the inhibition by palmitoyl-CoA is fully reversible is consistent with physiological regulation by long chain acyl-CoA thioesters. These data indicate that inhibition by palmitoyl-CoA and other acyl-CoA thioesters may, in princi- ' The critical micellar concentrations of the compounds in Table I  were measured at ionic strengths lower than that routinely employed in the assays of glycerol-P dehydrogenase activity. The maximum decrease in-the critical micellar concentration which might be expected due to this increased ionic strength (0.1 M Tris-HCl) is 2-to 3fold (14). Since the inhibition by acyl-CoA thioesters is detectable at concentrations 20-fold below the reported critical micellar concentrations (Table I) The binding of-a few moles of palmitoyl-CoA per mol of enzyme in the range of concentrations at which inhibition occurs might suggest interaction with a specific, regulatory site, but the binding of many moles per mol of enzyme could not rule out the presence of such a site (9). ple, function ' Acyl-CoA and acyl-ACP thioesters are acyl donors for phospholipid synthesis in vitro (20), but the contributions of acyl-ACP or acyl-CoA as acyl donors in viva remains to be established (26,29). Acyl-ACP thioesters have long been recognized to be the products of de nouo fatty acid synthesis (26). Only recently, has an enzvme capable of activating fatty acids as ACP thioesters been described (33).