The triose-phosphate site of homogeneous reconstituted sn-glycerol-3-phosphate acyltransferase of Escherichia coli

The sn-glycerol-3-phosphate (glycerol-phosphate) acyltransferase of Escherichia coli was purified to near homogeneity and its activity reconstituted with phospholipids (Green, P.R., Merrill, A.M., Jr. and Bell, R.M. (1981) J. Biol. Chem. 256, 11151-11159). The competency of glycerol-P analogues to serve as inhibitors and as substrates was investigated. Dihydroxyacetone-P, ethyleneglycol-P, 1,3-propanediol-P, 3,4-dihydroxybutylphosphonate and DL-glyceraldehyde-3-P were inhibitors of the reconstituted purified glycerol-phosphate acyltransferase. The kinetics of inhibition, while formally of the mixed type, most closely resembled that of a simple competitive inhibition with respect to glycerol-3-P. Inorganic phosphate was also found to be a competitive inhibitor. All of the glycerol-3-P analogues except DL-glyceraldehyde-3-P were substrates. Of these, dihydroxyacetone-P proved to be the best substrate. The secondary hydroxyl was not necessary for activity. Glycerol-phosphate acyltransferase catalyzed the hydrolysis of palmitoyl-CoA in the presence of DL-, but not D-glyceraldehyde-3-P. This suggests that the gem diol of L-glyceraldehyde-3-P may be a substrate, and that the acylated adduct may be unstable. The enzyme was inactivated by phenylglyoxal and butanedione, suggesting that arginine may be at or near the active site.     

Asymmetric reconstitution of homogeneous Escherichia coli sn-glycerol-3-phosphate acyltransferase into phospholipid vesicles

The sn-glycerol-3-phosphate (glycerol-P) acyltransferase of Escherichia coli cytoplasmic membrane was purified in Triton X-100 (Green, P. R., Merrill, A. H., Jr., and Bell, R. M. (1981) J. Biol. Chem. 256, 11151-11159) and incorporated into mixed micelles containing Triton X-100, phosphatidylethanolamine, phosphatidylglycerol, cardiolipin, and beta-octyl glucoside. Enzyme activity was quantitatively reconstituted from the mixed micelle into single-walled phospholipid vesicles by chromatography over Sephadex G-50. Activity coeluted with vesicles of 90-nm average diameter on columns of Sepharose CL-4B and Sephacryl S-1000. These vesicles contained less than 2 Triton X-100 and 5 beta-octyl glucoside molecules/100 phospholipid molecules. Calculations suggested that up to eight 91,260-dalton glycerol-P acyltransferase polypeptides were incorporated per 90-nm vesicle. The pH dependence and apparent Km values for glycerol-P and palmitoyl-CoA of the glycerol-P acyltransferase reconstituted into vesicles were similar to those observed upon reconstitution by mixing of the enzyme in Triton X-100 with a 20-fold molar excess of sonicated phosphatidylethanolamine:phosphatidylglycerol:cardiolipin, 6:1:1. The integrity of vesicles containing glycerol-P acyltransferase was established by trapping 5,5'-dithiobis-(2-nitrobenzoic acid). Chymotrypsin inactivated greater than 95% of the glycerol-P acyltransferase in intact vesicles and cleaved the 91,260-dalton polypeptide into several vesicle-bound and several released peptides, indicating that critical domains of the enzyme are accessible in intact vesicles. Trinitrobenzene sulfonate and 4,4'-diisothiocyano-2,2'-disulfonic acid stilbene caused greater than 90% loss of glycerol-P acyltransferase in vesicles. Disruption of vesicles with Triton X-100 did not reveal significant latent activity. These data strongly suggest that the glycerol-P acyltransferase was reconstituted asymmetrically into the vesicles with its active site facing outward.     

Phospholipid dependence of homogeneous, reconstituted sn-glycerol-3-phosphate acyltransferase of Escherichia coli

A novel mixed micelle assay for the sn-glycerol-3-phosphate acyltransferase of Escherichia coli was developed using the nonionic detergent octaethylenegly-coldodecyl ether. The assay permitted investigation of the phospholipid dependence of enzyme activity at phospholipid/detergent ratios of 5:1 (w/w) to 2:1 depending on the phospholipid employed. The higher ratio yielded maximal activity when E. coli phospholipids were used; the lower ratio was observed with cardiolipin(E. coli). Phosphatidylglycerol(E. coli) and phosphatidylethanolamine(E. coli) also restored enzyme activity. Activation by phosphatidylethanolamine(E. coli) was pH-dependent and relatively inefficient. The synthetic, disaturated (1,2-palmitoyl)phosphatidylglycerol reconstituted only 25% of the total enzyme activity as that observed with the monounsaturated (1-palmitoyl, 2-oleoyl) species. Full activation of enzyme was achieved with (1,2-dioleoyl)phosphatidylglycerol. Phosphatidylcholine and phosphatidic acid were unable to reconstitute enzyme activity. Chromatographic sizing of the sn-glycerol-3-phosphate acyltransferase, following reconstitution in cardiolipin(E. coli)/octaethyleneglycoldodecyl ether mixed micelles, suggested that the monomeric form of the enzyme was active.     

Role of spermidine in the activity of sn-glycerol-3-phosphate acyltransferase from Escherichia coli

The integral membrane protein, sn-glycerol-3-phosphate acyltransferase, catalyzes the first committed step in phospholipid synthesis, and both acyl-CoA and acyl-acyl carrier protein can be used as acyl donors in this reaction. We found that spermidine increased the specific activity of the acyltransferase when either substrate was used as the acyl donor. Magnesium, as well as other cations, also increased acyltransferase activity but were not nearly as effective as spermidine. Two roles for spermidine in this reaction were deduced from our data. First, spermidine dramatically lowered the K_m for glycerol 3-phosphate resulting in an overall rate enhancement when either substrate was used as the acyl donor. This effect was attributed to the modification of the acyl-transferase environment due to the binding of spermidine to membrane phospholipids. A second effect of spermidine was evident only when acyl-acyl carrier protein was used as substrate. Using this acyl donor, a pH optimum of 7.5 was found in the absence of spermidine, but in its presence, the pH optimum was shifted to 8.5. Between pH 7.5 and 8.5, palmitoyl-acyl carrier protein undergoes a conformational change to a more expanded, denatured state and its activity in the acyltransferase assay decreases dramatically. Spermidine restored the native conformation of palmitoyl-acyl carrier protein at pH 8.5, thus accounting for the majority of rate enhancement observed at elevated pH.     

Purification and positional specificity of sn-glycerol-3-phosphate acyltransferase from Escherichia coli membranes.

SN-Glycerol-3-phosphate acyltransferase was solubilized from membranes of Escherichia coli B and K-12 and purified on an affinity column of Sepharose 4B coupled with 6-phosphogluconic acid. Phosphatidylglycerol was required for activation and stabilization of the purified enzyme. The acyl residues were exclusively transferred to the position 1 of sn-glycerol 3-phosphate by the enzyme, regardless of whether the acyl-CoA was saturated or unsaturated.     

Efficiency of reconstitution of the membrane-associated sn-glycerol 3-phosphate acyltransferase of Escherichia coli

Membrane-associated enzymes are often solubilized with detergents, purified, and then reconstituted with phospholipid cofactors to regain function. Insofar as most purification and reconstitution procedures are not quantitative, the final reconstituted preparations could reflect a population of molecules ranging from fully functional to completely inactive. Quantitative studies on the efficiency of reconstitution of the Triton X-100-solubilized sn-glycerol 3-phosphate (glycerol-P) acyltransferase of Escherichia coli cytoplasmic membrane were undertaken at each step of purification. Physical recovery of the 83,000 Mr polypeptide was quantitated in polyacrylamide gels using membranes from cells labeled with [3H]leucine. The 83,000 Mr polypeptide in such gels was demonstrated to consist exclusively of the glycerol-P acyltransferase peptide by V8 peptide mapping. Comparison between physical recovery of 83,000 Mr polypeptide and reconstituted activity allowed the efficiency of reconstitution to be determined. Unexpectedly, disproportionalities occurred during the purification. However, the final purification of reconstituted enzyme activity matched that of the 83,000 Mr polypeptide. This method also allowed measurement of the specific activities of the glycerol-P acyltransferase in membranes from a wild type E. coli strain and from plasmid-containing strains which express the plsB gene product to different extents. The physical amounts of the 83,000 Mr polypeptide and glycerol-P acyltransferase activity measured in membranes were not strictly proportional. In strains where the amount of 83,000 Mr polypeptide was enhanced, a larger proportion of latent activity was observed following solubilization and reconstitution. The results establish the suitability of the reconstituted preparations of glycerol-P acyltransferase for detailed kinetic analysis and permit inferences pertaining to regulation.     

Structural characterization of ordered arrays of sn-glycerol-3-phosphate acyltransferase from Escherichia coli.

Overproduction of the sn-glycerol-3-phosphate acyltransferase in Escherichia coli leads to incorporation of this integral membrane protein into ordered tubular arrays within the cell. Freeze-fracture-etch shadowing was performed on suspensions of partially purified tubules and whole bacteria. This procedure revealed the presence of ridges and grooves defining a set of long-pitch left-handed helical ridges. The long-pitch helices represented chains of acyltransferase dimers. Tubules observed within the cell were often closely packed, with an apparent alignment of grooves and ridges in adjacent tubules. Fracture planes passing through the tubules indicated the presence of a bilayer structure, with some portion of the enzyme being associated with the membrane. The major portion of the enzyme extended from the hydrophilic surface, forming a large globular structure that, in favorable views, displayed a central cavity facing the cytoplasm. Computer analysis of shadowed tubules revealed that the left-handed helices were six stranded, with a pitch of 1,050 A (105.0 nm) and a spacing of 75 A (7.5 nm) between acyltransferase dimers along the chains. Analysis of the predicted secondary structure failed to reveal obvious transmembrane segments, suggesting that very little of the protein was inserted into the bilayer.     

Function of phospholipids on the regulatory properties of solubilized and membrane-bound sn-glycerol-3-phosphate acyltransferase of Escherichia coli.

Glycerophosphate acyltransferase (acyl-CoA:sn-glycerol-3-phosphate O-acyltransferase, EC 2.3.1.15) solubilized from Escherichia coli membranes was highly activated by phosphatidylglycerol. Phosphatidylethanolamine, cardiolipin and 1,2-diacyl-sn-glycerol 3-phosphate showed no effect. The Km of the enzyme for sn-glycerol 3-phosphate was increased 20-fold by solubilization. The value could not be restored by the addition of phospholipids. Temperature-sensitive regulation of the synthesis of either 1-palmitoyl- or cis-vaccenoyl-sn-glycerol 3-phosphate by the solubilized enzyme was identical with that by the membrane-bound enzyme in vivo and in vitro. The proportion of the molecular species of 1-acyl-sn-glycerol 3-phosphate varied when the ratios of palmitoyl-CoA and cis-vaccenoyl-CoA were changed, but changes in the sn-glycerol 3-phosphate concentration had no effect on selective acylation by both the solubilized and membrane-bound enzymes.     

Partial NH2- and COOH-terminal sequence and cyanogen bromide peptide analysis of Escherichia coli sn-glycerol-3-phosphate acyltransferase

The sn-glycerol-3-phosphate acyltransferase from Escherichia coli, an integral membrane protein whose activity is dependent on phospholipids, was purified to near homogeneity (Green, P. R., Merrill, A. H., Jr., and Bell, R. M., (1981) J. Biol. Chem. 256, 11151-11159). Determination of a partial NH2-terminal sequence and the COOH terminus permitted alignment of the polypeptide on the sequenced sn-glycerol-3-phosphate acyltransferase structural gene (Lightner, V. A., Bell, R. M., and Modrich, P. (1983) J. Biol. Chem. 258, 10856-10861). Processing of the sn-glycerol-3-phosphate acyltransferase is apparently limited to the removal of the NH2-terminal formylmethionine. Thirteen of 27 possible cyanogen bromide peptides predicted from the DNA sequence were purified, characterized, and assigned to their location in the primary structure. Three peptides located at positions throughout the sequence were partially sequenced by automated Edman degradation. The partial sequence analysis of the homogeneous sn-glycerol-3-phosphate acyltransferase is fully in accord with the primary structure inferred from the DNA sequence.     

Mutants of Escherichia coli defective in membrane phospholipid synthesis. Phenotypic suppression of sn-glycerol-3-phosphate acyltransferase Km mutants by loss of feedback inhibition of the biosynthetic sn-glycerol-3-phosphate dehydrogenase.

Revertants of Escherichia coli mutants defective in the first enzyme of membrane phospholipid synthesis, sn-glycerol-3-phosphate (glycerol-P) acyltransferase, were investigated. These glycerol-P acyltransferase mutants, selected as glycerol-P auxotrophs, contained membranous glycerol-P acyltransferase activity with an apparent Km for glycerol-P 10 times higher than the parental activity. The glycerol-P acyltransferase activity was also more thermolabile in vitro than the parental activity. Most revertants no longer requiring glycerol-P for growth regained glycerol-P acyltransferase activity of normal thermolability and apparent Km for glycerol-P. However, two novel revertants were isolated which retained an abnormal glycerol-P acyltransferase activity. The glycerol-P dehydrogenase activities of these novel revertants were about 20-fold less sensitive to feedback inhibition by glycerol-P. The feedback-resistant glycerol-P dehydrogenase co-transduced with gpsA, the structural gene for the glycerol-P dehydrogenase. Further transduction experiments demonstrated that the feedback resistant glycerol-P dehydrogenase phenotypically suppressed the glycerol-P acyltransferase Km lesion. The existence of the class of glycerol-P auxotrophs which owe their phenotype to the glycerol-P acyltransferase Km lesion therefore depends on the feedback regulation of glycerol-P synthesis in E. coli.     

Localization and characterization of the sn-glycerol-3-phosphate acyltransferase in Rhodopseudomonas sphaeroides

The membrane localization and properties of the Rhodopseudomonas sphaeroides sn-glycerol-3-phosphate acyltransferase have been examined utilizing enzymatically prepared acyl-acyl carrier protein (acyl-ACP) substrates as acyl donors for sn-glycerol-3-phosphate acylation. Studies conducted with membranes prepared from chemotrophically and phototrophically grown cells show that sn-glycerol-3-phosphate acyltransferase activity is predominantly (greater than 80%) associated with the cell's cytoplasmic membrane. Enzyme activity associated with the intracytoplasmic membranes present in phototrophically grown R. sphaeroides was within the range attributable to cytoplasmic membrane contamination of this membrane fraction. Enzyme activity was optimal at 40 degrees C and pH 7.0 to 7.5, and required the presence of magnesium. No enzyme activity was observed with any of the long-chain acyl-CoA substrates examined. Vaccenoyl-ACP was the preferred acyl-ACP substrate and vaccenoyl-ACP and palmitoyl-ACP were independently utilized to produce lysophosphatidic and phosphatidic acids. With either vaccenoyl-ACP or palmitoyl-ACP as sole acyl donor substrate, the lysophosphatidic acid formed was primarily 1-acylglycerol-3-phosphate and the Km(app) for sn-glycerol-3-phosphate utilization was 96 microM. The implications of these results to the mode and regulation of phospholipid synthesis in R. sphaeroides are discussed.     

A second transport system for sn-glycerol-3-phosphate in Escherichia coli.

Strains containing phage Mucts inserted into glpT were isolated as fosfomycin-resistant clones. These mutants did not transport sn-glycerol-3-phosphate, and they lacked GLPT, a protein previously shown to be a product of the glpT operon. By plating these mutants on sn-glycerol-3-phosphate at 43 degrees C, we isolated revertants that regained the capacity to grow on G3P. Most of these revertants did not map in glpT and did not regain GLPT. These revertants exhibited a highly efficient uptake system for sn-glycerol-3-phosphate within an apparent Km of 5 micron. In addition, three new proteins (GP 1, 2, and 3) appeared in the periplasm of these revertants. None of these proteins were antigentically related to GLPT. However, like GLPT, GP1 exhibits abnormal behavior on sodium dodecyl sulfate-polyacrylamide gels. GP 2 is an efficient binding protein. The new uptake system showed different characteristics than the system that is coded for by the glpT operon. It was inhibited neither by phosphate nor fosfomycin. So far, none of the systems that transport organic acids in Escherichia coli could be implicated in the new sn-glycerol-3-phosphate uptake activity. The mutation ugp+, which was responsible for the appearance of the new transport system and the appearance of GP 1, 2, and 3 in the periplasm was cotransducible with araD by phage P1 transduction and was recessive in merodiploids.     

Periplasmic protein related to the sn-glycerol-3-phosphate transport system of Escherichia coli.

Two-dimensional gel electrophoresis of shock fluids of Escherichia coli K-12 revealed the presence of a periplasmic protein related to sn-glycerol-3-phosphate transport (GLPT) that is under the regulation of glpR, the regulatory gene of the glp regulon. Mutants selected for their resistance to phosphonomycin and found to be defective in sn-glycerol-3-phosphate transport either did not produce GLPT or produced it in reduced amounts. Other mutations exhibited no apparent effect of GLPT. Transductions of glpT+ nalA phage P1 into these mutants and selection for growth on sn-glycerol-3-phosphate revealed a 50% cotransduction frequency to nalA. Reversion of mutants taht did not produce GLPT to growth on sn-glycerol-3-phosphate resulted in strains that produce GLPT. This suggests a close relationship of GLPT to the glpT gene and to sn-glycerol-3-phosphate transport. Attempts to demonstrate binding activity of GLPT in crude shock fluid towards sn-glycerol-3-phosphate have failed so far. However, all shock fluids, independent of their GLPT content, exhibited an enzymatic activity that hydrolyzes under the conditions of the binding assay, 30 to 60% of the sn-glycerol-3-phosphate to glycerol and inorganic orthophosphate.     

Characterization of the ugp region containing the genes for the phoB dependent sn-glycerol-3-phosphate transport system of Escherichia coli

Summary The ugp structural genes, coding for the pho regulon dependent sn-glycerol-3-phosphate transport system, were cloned in pBR322 and characterized. The expression of the cloned ugp system was phoB dependent. Cells containing the ugp plasmid overproduced the G3P binding protein upon phosphate starvation. Tn5 mutagenesis of the cloned DNA revealed that the ugp genes are organized in two separate operons which comprise at least four genes: ugpB and ugpD constitute one operon, ugpA and ugpC constitute the other. The structural gene for the G3P binding protein (G3PBP) is ugpB.The ugpC gene product was also synthesized in minicells as a polypeptide, with an apparent molecular weight of 40,000. No gene products could be assigned to the ugpA and ugpD genes. Hybridization experiments allowed the physical characterization of 20 kb of DNA adjacent to the ugp genes on the E. coli chromosome including the liv genes.     

Mutants of Escherichia coli defective in membrane phospholipid synthesis. Properties of wild type and Km defective sn-glycerol-3-phosphate acyltransferase activities.

The sn-glycerol-3-phosphate (glycerol-P) acyltransferase, the first enzyme of membrane phospholipid synthesis in Escherichia coli, was investigated in a wild type and a mutant strain defective in this activity. The mutant strain, selected as a glycerol-P auxotroph, was previously shown to contain a glycerol-P acyltransferase activity with an apparent Km for glycerol-P 10 times higher than that of its parent or revertants. The membranous mutant glycerol-P acyltransferase but did not appear to be thermolabile in vivo. Revertants no longer requiring glycerol-P for growth, showed glycerol-P acyltransferase activity with thermolability properties similar to the wild type. The second phospholipid biosynthetic enzyme, 1-acylglycerol-P acyltransferase, was not thermolabile in membranes containing a thermolabile glycerol-P acyltransferase activity. The pH optimum for the mutant acyltransferase was over 1 pH unit higher than that of the parental activity. Further, the mutant and wild type glycerol-P acyltransferase differed in their response to magnesium chloride and potassium chloride. The palmitoyl-CoA dependence of the wild type and mutant glycerol-P acyltransferase activities were different. The mutant glycerol-P acyltransferase activity was inhibited greater than 90% by Triton X-100 under conditions where the wild type activity was not affected. These experiments provide novel information about the wild type glycerol-P acyltransferase activity of E. coli and provide six additional lines of evidence for the mutant character of the glycerol-P acyltransferase in the mutant strains.     

Analysis of amino acid motifs diagnostic for the sn-glycerol-3-phosphate acyltransferase reaction

Alignment of amino acid sequences from various acyltransferases [sn-glycerol-3-phosphate acyltransferase (GPAT), lysophosphatidic acid acyltransferase (LPAAT), acyl-CoA:dihydroxyacetone-phosphate acyltransferase (DHAPAT), 2-acylglycerophosphatidylethanolamine acyltransferase (LPEAT)] reveals four regions of strong homology, which we have labeled blocks I-IV. The consensus sequence for each conserved region is as follows: block I, [NX]-H-[RQ]-S-X-[LYIM]-D; block II, G-X-[IF]-F-I-[RD]-R; block III, F-[PLI]-E-G-[TG]-R-[SX]-[RX]; and block IV, [VI]-[PX]-[IVL]-[IV]-P-[VI]. We hypothesize that blocks I-IV and, in particular, the invariant amino acids contained within these regions form a catalytically important site in this family of acyltransferases. Using Escherichia coli GPAT (PlsB) as a model acyltransferase, we examined the role of the highly conserved amino acid residues in blocks I-IV in GPAT activity through chemical modification and site-directed mutagenesis experiments. We found that the histidine and aspartate in block I, the glycine in block III, and the proline in block IV all play a role in E. coli GPAT catalysis. The phenylalanine and arginine in block II and the glutamate and serine in block III appear to be important in binding the glycerol 3-phosphate substrate. Since blocks I-IV are also found in LPAAT, DHAPAT, and LPEAT, we believe that these conserved amino acid motifs are diagnostic for the acyltransferase reaction involving glycerol 3-phosphate, 1-acylglycerol 3-phosphate, and dihydroxyacetone phosphate substrates.