Crystal structure of CTP:glycerol-3-phosphate cytidylyltransferase from Staphylococcus aureus: examination of structural basis for kinetic mechanism

Integrity of the cell wall is essential for bacterial survival, and as a consequence components involved in its biosynthesis can potentially be exploited as targets for antibiotics. One such potential target is CTP:glycerol-3-phosphate cytidylyltransferase. This enzyme (TarD(Sa) in Staphylococcus aureus and TagD(Bs) in Bacillus subtilis) catalyzes the formation of CDP-glycerol, which is used for the assembly of linkages between peptidoglycan and teichoic acid polymer in Gram-positive bacteria. Intriguingly, despite the high sequence identity between TarD(Sa) and TagD(Bs) (69% identity), kinetic studies show that these two enzymes differ markedly in their kinetic mechanism and activity. To examine the basis for the disparate enzymological properties, we have determined the crystal structure of TarD(Sa) in the apo state to 3 A resolution, and performed equilibrium sedimentation analysis. Comparison of the structure with that of CTP- and CDP-glycerol-bound TagD(Bs) crystal structures reveals that the overall structure of TarD(Sa) is essentially the same as that of TagD(Bs), except in the C-terminus, where it forms a helix in TagD(Bs) but is disordered in the apo TarD(Sa) structure. In addition, TarD(Sa) can exist both as a tetramer and as a dimer, unlike TagD(Bs), which is a dimer. These observations shed light on the structural basis for the differing kinetic characteristics between TarD(Sa) and TagD(Bs).     

Negative cooperativity of substrate binding but not enzyme activity in wild-type and mutant forms of CTP:glycerol-3-phosphate cytidylyltransferase

CTP:glycerol-3-phosphate cytidylyltransferase (GCT) catalyzes the synthesis of CDP-glycerol for teichoic acid biosynthesis in certain Gram-positive bacteria. This enzyme is a model for a cytidylyltransferase family that includes the enzymes that synthesize CDP-choline and CDP-ethanolamine for phosphatidylcholine and phosphatidylethanolamine biosynthesis. We have used quenching of intrinsic tryptophan fluorescence to measure binding affinities of substrates to the GCT from Bacillus subtilis. Binding of either CTP or glycerol-3-phosphate to GCT was biphasic, with two binding constants of about 0.1-0.3 and 20-40 microm for each substrate. The stoichiometry of binding was 2 molecules of substrate/enzyme dimer, so the two binding constants represented distinctly different affinities of the enzyme for the first and second molecule of each substrate. The biphasic nature of binding was observed with the wild-type GCT as well as with several mutants with altered Km or kcat values. This negative cooperativity of binding was also seen when a catalytically defective mutant was saturated with two molecules of CTP and then titrated with glycerol-3-phosphate. Despite the pronounced negative cooperativity of substrate binding, negative cooperativity of enzyme activity was not observed. These data support a mechanism in which catalysis occurs only when the enzyme is fully loaded with 2 molecules of each substrate/enzyme dimer.     

Inhibition of glycerol-3-phosphate acyltransferase by analogs of glycerol-3-phosphate.

Analogs of glycerol-3-phosphate were tested as substrates or inhibitors of the glycerol-3-phosphate acyltransferases of mitochondria and microsomes. (rac)-3,4-Dihydroxybutyl-1-phosphonate, (rac)-glyceraldehyde 3-phosphate, (rac)-3-hydroxy-4-oxobutyl-1-phosphonate, (1S,3S)-1,3,4-trihydroxybutyl-1-phosphonate, and (1R,3S)-1,3,4 trihydroxybutyl-1-phosphonate were competitive inhibitors of both mitochondrial and microsomal sn-glycerol-3-phosphate acyltransferase activity. An isosteric analog of dihydroxyacetone phosphate, 4-hydroxy-3-oxobutyl-1-phosphonate, was a much stronger competitive inhibitor of the microsomal than the mitochondrial enzyme. Phenethyl alcohol was a noncompetitive inhibitor of both the microsomal and the mitochondrial acyltransferases. The product of the mitochondrial acyltransferase reaction with (rac)-3,4-dihydroxybutyl-1- phosphonate was almost exclusively (rac)-4-palmitoyloxy-3-hydroxybutyl-1-phosphonate. The microsomal acylation reaction generated both the monoacyl product and (S)-3,4-dipalmitoyloxybutyl-1-phosphonate. The apparent Km for (S)-3,4-dihydroxybutyl-1-phosphonate was 2.50 and 1.38 mM for the mitochondrial and microsomal enzymes, respectively.     

Substrate specificity of CTP:phosphocholine cytidylyltransferase.

The specificity of CTP:phosphocholine cytidylyltransferase from rat liver for phosphorylated bases has been investigated. The apparent Km for phosphocholine was 0.17 mM. As the number of methyl substituents on the phospho-base decreased, the apparent Km increased: 4.0 mM for phosphodimethylethanolamine, 6.9 for phosphomonomethylethanolamine and 68.4 for phosphoethanolamine. The Vmax for the reaction was similar for phosphocholine (12.6 mumol/min per mg protein), phosphomonomethylethanolamine (13.5 mumol/min per mg protein) and phosphoethanolamine (9.2 mumol/min per mg protein). When phosphodimethylethanolamine was the substrate, the Vmax was 3-fold higher (40.3 mumol/min per mg protein). Phosphoethanolamine, phosphomonomethylethanolamine and phosphodimethylethanolamine were competitive inhibitors of the cytidylyltransferase when phosphocholine was used as substrate with Ki values of 18.5 mM, 9.3 mM and 1.5 mM, respectively. The results show that the cytidylyltransferase is highly specific for phosphocholine.     

Identification of a mitochondrial glycerol-3-phosphate dehydrogenase from Arabidopsis thaliana: evidence for a mitochondrial glycerol-3-phosphate shuttle in plants

We report molecular characterization of an Arabidopsis gene encoding a mitochondrial FAD-dependent glycerol-3-phosphate dehydrogenase (FAD-GPDH) that oxidizes glycerol-3-phosphate (G-3-P) to dihydroxyacetone phosphate. We demonstrate through in vitro targeting assays that the encoded gene product can be imported into mitochondrial membrane systems. Enzyme activity of the protein was confirmed through heterologous expression in Escherichia coli. The Arabidopsis gene is expressed throughout plant development, but at the highest level during seed germination. We also show that expression of the Arabidopsis FAD-GPDH gene is coupled to oxygen consumption and affected by ABA and stress conditions. Together with an NAD(+)-dependent GPDH, this enzyme could form a G-3-P shuttle, as previously established in other eukaryotic organisms, and links cytosolic G-3-P metabolism to carbon source utilization and energy metabolism in plants.     

Isoenzymic forms of NAD-linked glycerol-3-phosphate dehydrogenase from rabbit brain.

Two major enzyme forms of cytosolic NAD-linked glycerol-3-phosphate dehydrogenase in rabbit brain have been purified to apparent homogeneity. One major enzyme form designated I6.5 exhibits an iso-electric point at pH 6.5, and is indistinguishable from the major form I6.5 found in other tissues. The other major form, designated I5.9, has an isolectric point at pH 5.9, and by amino acid analysis is shown to be a true isoenzyme distinct from form I6.5. Form I5.9 appears to be closely related to or identical with the major enzyme characteristic of heart. Neither the brain enzyme form I5.9 nor the major heart isoenzyme are inhibited by antiserum to the muscle enzyme. Because of the high apparent Km for NADH, it is postulated that the brain isoenzyme I5.9 serves to maintain glycolysis when NADH levels rise under relatively anaerobic conditions especially during fetal and neonatal development.     

CTP:phosphocholine cytidylyltransferase in intestinal mucosa

CTP:phosphocholine cytidylyltransferase is thought to be a rate-limiting enzyme in phosphatidylcholine synthesis. This enzyme has not been well studied in intestine. We found that activity was greater in the non-lipid stimulated state (cytosolic form of the enzyme) than any previous tissue investigated (2.7 nM/min per mg protein). On addition of lysophosphatidylethanolamine, the enzyme only increased in activity 2.4-fold which is less than any previously reported tissue on lipid stimulation. As compared to liver, the enzyme was resistant to inhibition by chlorpromazine (gut, 100% activity remaining at 80 microM; 14% in liver). Tetracaine and propranolol were found to be impotent as inhibitors of the intestinal enzyme. Octanol-water partitioning showed that both chlorpromazine and tetracaine were hydrophobic, propranolol was not. pKa studies demonstrated that at the reaction pH, chlorpromazine would be uncharged. Physiologic experiments in which de novo phosphatidylcholine synthesis was either stimulated by bile duct fistulization and triacylglycerol infusion or suppressed by including phosphatidylcholine in a lipid infusion demonstrated that the enzyme (cytosolic enzyme) responded by decreasing Vmax but that the Km remained the same. In sum, these studies suggest that CTP:phosphocholine cytidylyltransferase in intestine is unique as compared to other tissues and that its response to a physiological stimulus is counter to that which would be adaptive.     

Structural analysis of glycerol-3-phosphate dehydrogenase from severalDrosophila species

This report describes preliminary protein structural studies of glycerol-3-phosphate dehydrogenase (alpha-GPDH) from Drosophila spp. and an important innovative feature of our enzyme purification protocol. The scheme involves the coupling of substrate (alpha-glycerophosphate) elution from CM-Sephadex and cofactor (NADH) elution from Affi-Gel blue resin. Using this method a 32.7% yield and a 111-fold purification were obtained from a D. melanogaster line carrying the alpha-GpdhS allele at the alpha-Gpdh locus. The product obtained from 0 to 3-day-old adult flies was electrophoretically homogeneous and consisted mainly of the adult alpha-GPDH-1 isozyme. The method was used to obtain alpha-GPDH protein from D. melanogaster (two lines), D. hydei, D. immigrans, and D. mercatorum. Peptide mapping revealed structural differences among the enzymes from the different species, and amino acid sequencing showed many similarities between D. melanogaster alpha-GPDH and the rabbit muscle enzyme.     

c-Jun N-terminal kinase regulates CTP:phosphocholine cytidylyltransferase

CTP:phosphocholine cytidylyltransferase (CCTalpha) is a rate-regulatory enzyme required for phosphatidylcholine (PtdCho) synthesis. CCTalpha is also a phosphoenzyme, but the physiologic role of kinases on enzyme function remains unclear. We report high-level expression of two major isoforms of the c-Jun N-terminal kinase family (JNK1 and JNK2) in murine lung epithelia. Further, JNK1 and JNK2 phosphorylated purified CCTalpha in vitro, and this was associated with a dose-dependent decrease (approximately 40%) in CCT activity. To evaluate JNK in vivo, lung epithelial cells were infected with a replication defective adenoviral vector encoding murine JNK2 (Adv-JNK2) or an empty vector. Adv-JNK2 infection, unlike the empty vector, markedly increased JNK2 expression concomitant with increased incorporation of [32P]orthophosphate into endogenous CCTalpha. Although Adv-JNK2 infection only modestly reduced CCT activity, it reduced PtdCho synthesis by approximately 30% in cells. These observations suggest a role for JNK kinases as negative regulators of phospholipid synthesis in murine lung epithelia.     

Colony-stimulating factor 1 regulates CTP: phosphocholine cytidylyltransferase mRNA levels.

Growth factor regulation of phosphatidylcholine (PtdCho) metabolism during the G1 stage of the cell cycle was investigated in the colony-stimulating factor 1 (CSF-1)-dependent murine macrophage cell-line BAC1.2F5. The transient removal of CSF-1 arrested the cells in G1. Incorporation of [3H]choline into PtdCho was stimulated significantly 1 h after growth factor addition to quiescent cells. Metabolic labeling experiments pointed to CTP:phosphocholine cytidylyltransferase (CT) as the rate-controlling enzyme for PtdCho biosynthesis in BAC1.2F5 cells. The amount of CT mRNA increased 4-fold within 15 min of CSF-1 addition and remained elevated for 2 h. The rise in CT mRNA levels was accompanied by a 50% increase in total CT specific activity in cell extracts within 4 h after the addition of CSF-1. CSF-1-dependent elevation of CT mRNA content was neither attenuated nor superinduced by the inhibition of protein synthesis with cycloheximide. The rate of CT mRNA turnover decreased in the presence of CSF-1 indicating that message stabilization was a key factor in determining the levels of CT mRNA. These data point to increased CT mRNA abundance as a component in growth factor-stimulated PtdCho synthesis.     

The purification and characterization of CTP:phosphorylcholine cytidylyltransferase from rat liver

We have purified CTP:phosphorylcholine cytidylyltransferase from rat liver cytosol 2180-fold to a specific activity of 12,250 nmol/min/mg of protein. The purified enzyme was stable at -70 degrees C in the presence of Triton X-100 and 0.2 M phosphate. The purified enzyme gave a single protein and activity band on nondenaturing polyacrylamide electrophoresis. Separation by sodium dodecyl sulfate-polyacrylamide electrophoresis indicated that the purified enzyme contained subunits with Mr of 39,000 and 48,000. Gel filtration analysis indicated that the native enzyme was a tetramer containing two 39,000 and two 48,000 subunits. The purified enzyme appeared to bind to Triton X-100 micelles, one molecule of tetramer/micelle. Maximal activity was obtained with 100 microM phosphatidylcholine-oleic acid vesicles (8-10-fold stimulation). Phosphatidylglycerol produced a 4-5-fold increase in activity at 10 microM. The pH optimum and true Km values for CTP and phosphorylcholine were similar to those reported previously for crude preparations of cytidylyltransferase. The overall behavior of cytidylyltransferase during purification and subsequent analysis suggested that it has hydrophobic properties similar to those exhibited by membrane proteins.     

Isolation and properties of glycerol-3-phosphate oxidoreductase from human placenta

Glycerol-3-phosphate oxidoreductase (sn-glycerol 3-phosphate: NAD+ 2-oxidoreductase, EC 1.1.1.8) from human placenta has been purified by chromatography on 2,4,6-trinitrobenzenehexamethylenediamine-Sepharose, DEAE-Sephadex A-50 and 5'-AMP-Sepharose 4B approximately 15800-fold with an overall yield of about 19%. The final purified material displayed a specific activity of about 88 mumol NADH min-1 mg protein-1 and a single protein band on polyacrylamide gel electrophoresis in the presence of sodium dodecyl sulphate. The native molecular mass, determined by Ultrogel AcA 44 filtration, was 62000 +/- 2000 whereas the subunit molecular mass, established on polyacrylamide gel in the presence of 0.1% sodium dodecyl sulphate, was 38000 +/- 500. The isoelectric point of the enzyme protein, determined by column isoelectric focusing, was found to be 5.29 +/- 0.09. The pH optimum of the placental enzyme was in the range 7.4-8.1 for dihydroxyacetone phosphate reduction and 8.7-9.2 for sn-glycerol 3-phosphate oxidation. The apparent Michaelis constants (Km) for dihydroxyacetone phosphate, NADH, sn-glycerol 3-phosphate and NAD+ were 26 microM, 5 microM, 143 microM and 36 microM respectively. The activity ratio of cytoplasmic glycerol-3-phosphate oxidoreductase to mitochondrial glycerol-3-phosphate dehydrogenase in human placental tissue was 1:2. The consumption of oxygen by human placental mitochondria incubated with the purified glycerol-3-phosphate oxidoreductase, NADH and dihydroxyacetone phosphate was similar to that observed in the presence of sn-glycerol 3-phosphate. The possible physiological role of glycerol-3-phosphate oxidoreductase in placental metabolism is discussed.     

Purification and kinetic characterization of CTP:phosphocholine cytidylyltransferase from Saccharomyces cerevisiae

CTP:phosphocholine cytidylyltransferase (CCT) regulates the biosynthesis of phosphatidylcholine in mammalian cells. In order to understand the mechanism by which this enzyme controls phosphatidylcholine synthesis, we have initiated studies of CCT from the model genetic system, the yeast Saccharomyces cerevisiae. The yeast CCT gene was isolated from genomic DNA using the polymerase chain reaction and was found to encode tyrosine at position 192 instead of histidine, as originally reported. Levels of expression of yeast CCT activity in Escherichia coli or in the yeast, Pichia pastoris, were somewhat low. Expression of yeast CCT in a baculovirus system as a 6x-His-tag fusion protein was higher and was used to purify yeast CCT by a procedure that included delipidation. Kinetic characterization revealed that yeast CCT was activated approximately 20-fold by 20 microM phosphatidylcholine:oleate vesicles, a level 5-fold lower than that necessary for maximal activation of rat CCT. The k(cat) value was 31.3 s(-1) in the presence of lipid and 1.5 s(-1) in the absence of lipid. The K(m) values for the substrates CTP and phosphocholine did not change significantly upon activation by lipids; K(m) values in the presence of lipid were 0.80 mM for phosphocholine and 1.4 mM for CTP while K(m) values in the absence of lipid were 1.2 mM for phosphocholine and 0.8 mM for CTP. Activation of yeast CCT, therefore, appears to be due to an increase in the k(cat) value upon lipid binding.     

Characterization of a lipid activated CTP:phosphocholine cytidylyltransferase from Drosophila melanogaster

CTP:phosphocholine cytidylyltransferase (CCT) is an enzyme critical for cellular phosphatidylcholine (PC) synthesis, converting phosphocholine and cytidine 5'-triphosphate (CTP) to CDP-choline. We have isolated a cDNA encoding an isoform of CCT from Drosophila melanogaster and expressed the recombinant native and 6 x -His-tagged forms using a baculovirus expression system in Spodoptera frugiperda (Sf9) insect cells. Immunoblot using anti-phospho amino acid antibodies reveals the enzyme is phosphorylated on serine and threonine residues, but not tyrosine. The purified native enzyme exhibits a V(max) value of 1352+/-159 nmol CDP-choline/min/mg, a K(m) value of 0.50+/-0.09 mM for phosphocholine, and a K' (Hill constant) value of 0.72+/-0.10 mM for CTP. The 6 x -His-tagged enzyme has similar properties with a V(max) value of 2254+/-253 nmol CDP-choline/min/mg, a K(m) value of 0.63+/-0.13 mM for phosphocholine and a K' for CTP equal to 0.81+/-0.20 mM. Each form of the enzyme was activated to a similar extent by synthetic PC vesicles containing 50 mol% oleate. The efficiency of lipid activation was greatest using PC vesicles containing diphosphatidylglycerol (DPG), significantly less efficient activation was seen when phosphatidylserine (PS) and phosphatidylinositol (PI) were incorporated into vesicles, and PC alone or PC vesicles containing phosphatidylethanolamine were the least efficient enzyme activators.     

CTP:phosphorylcholine cytidylyltransferase from rat liver. Isolation and characterization of the catalytic subunit

We reported previously the purification of CTP:phosphorylcholine cytidylyltransferase from rat liver (Weinhold, P. A., Rounsifer, M. E., and Feldman, D. A. (1986) J. Biol. Chem. 261, 5104-5110). The purified enzyme appeared to contain equal amounts of two nonidentical proteins, with Mr of about 38,000 and 45,000. We have now separated and purified these proteins. Polyacrylamide electrophoresis in the presence of sodium dodecyl sulfate indicated that each protein was homogeneous. The 45,000 protein contained the catalytic activity. Analysis by gel filtration chromatography and glycerol gradient centrifugation indicated that the 38,000 and 45,000 proteins in the purified cytidylyltransferase were independently associated with Triton X-100 micelles. The apparent Mr of the complexes suggested that a tetramer of each protein was bound to one Triton X-100 micelle. The isolated 45,000 catalytic protein had the same lipid requirement and kinetic properties as the purified cytidylyltransferase containing both proteins. Enzyme activity was stimulated to maximal values by phosphatidylcholine vesicles containing 9 mol % of either oleic acid, phosphatidylinositol, or phosphatidylglycerol. The amino acid compositions of the isolated 38,000 and 45,000 proteins were distinctly different. Overall, the results suggested that a tetramer of the 45,000 protein possessed nearly optimal catalytic activity. A functional role of the 38,000 protein as part of a cytidylyltransferase enzyme complex could not be documented. However, the need for stabilizing concentrations of Triton X-100 in the purified enzyme preparation may have prevented the association of the two proteins.     

Structure and mechanism of CTP:phosphocholine cytidylyltransferase (LicC) from Streptococcus pneumoniae.

Pneumococcal LicC is a member of the nucleoside triphosphate transferase superfamily and catalyzes the transfer of a cytidine monophosphate from CTP to phosphocholine to form CDP-choline. The structures of apo-LicC and the LicC-CDP-choline-Mg(2+) ternary complex were determined, and the comparison of these structures reveals a significant conformational change driven by the multivalent coordination of Mg(2+). The key event is breaking the Glu(216)-Arg(129) salt bridge, which triggers the coalescence of four individual beta-strands into two extended beta-sheets. These movements reorient the side chains of Trp(136) and Tyr(190) for the optimal binding and alignment of the phosphocholine moiety. Consistent with these conformational changes, LicC operates via a compulsory ordered kinetic mechanism. The structures explain the substrate specificity of LicC for CTP and phosphocholine and implicate a direct role for Mg(2+) in aligning phosphocholine for in-line nucleophilic attack and stabilizing the negative charge that develops in the pentacoordinate transition state. These results provide a structural basis for assigning a specific role for magnesium in the catalytic mechanism of pneumococcal LicC.