Glycerophosphocholine Cholinephosphodiesterase Research Articles

Purification, characterization, molecular cloning, and extracellular production of a novel bacterial glycerophosphocholine cholinephosphodiesterase from Streptomyces sanglieri

A novel metal ion-independent glycerophosphocholine cholinephosphodiesterase (GPC-CP) of Streptomyces sanglieri was purified 53-fold from culture supernatant with 1.1% recovery (583 U/mg-protein). The enzyme functions as a monomer with a molecular mass of 66 kDa. The gene encoding the enzyme consists of a 1941-bp ORF that produces a signal peptide of 38 amino acids for secretion and a 646 amino acid mature protein with a calculated molecular mass of 70,447 Da. The maximum activity was found at pH 7.2 and 40°C. The enzyme hydrolyzed glycerol-3-phosphocholine (GPC) over a broad temperature range (37-60°C) and within a narrow pH range near pH 7. The enzyme was stable at 50°C for 30 min and between pH 5-10.5. The enzyme exhibited specificity toward GPC and glycerol-3-phosphoethanolamine and hydrolyzed glycerol-3-phosphate and lysophosphatidylcholine. However, the enzyme showed no activity toward any diacylglycerophospholipids and little activity toward other glycerol-3-phosphodiesters and lysophospholipids. The enzyme was not inhibited in the presence of 2 mM SDS and Mg(2+); however, Cu(2+), Zn(2+), and Co(2+) remarkably inhibited activity. Enzyme activity was also slightly enhanced by Ca(2+), Na(+), EDTA, DTT, and 2-mercaptoethanol. During the hydrolysis of GPC at 37°C and pH 7.2, apparent Vmax and turnover number (kcat) were determined to be 24.7 μmol min(-1) mg-protein(-1) and 29.0 s(-1), respectively. The apparent Km and kcat/Km values were 1.41 mM and 20.6 mM(-1) s(-1), respectively. GPC hydrolysis by GPC-CP might represent a new metabolic pathway for acquisition of a phosphorus source in actinomycetes.

Release of GPI-anchored Zn2+-glycerophosphocholine cholinephosphodiesterase as an amphiphilic form from bovine brain membranes by bee venom phospholipase A2.

Enzymatic release of Zn(2+)-glycerophosphocholine (GPC)cholinephosphodiesterase, as an amphiphilic form, from bovine brain membranes was examined. Of various membrane hydrolases, bee PLA2 was the most effective in the release of the GPC cholinephosphodiesterase (amphiphilic form, 63-70%) from membrane. Compared to pancreatic PLA2, bee PLA2 was more efficient in the release of GPC cholinephosphodiesterase. In pH-dependent release of GP1-anchored phosphodiesterase, there was a similar pH-release profile between PLA2-mediated release and spontaneous one, implying the involvement of membrane disruption in the PLA2 action. The PLA2-mediated release showed a limited time-dependence (until 45 min) and a limited dose dependence (up to 3 units/ml), characteristic of a receptor-type binding. An ionic binding of PLA2 to membrane may be alluded from the interfering effect of anionic phospholipids on the PLA2 action. In support of an interaction between PLA2 and membrane glycoproteins, the PLA2 action was found to be blocked by lectins, wheat germ agglutinin or concanavalin A. In combination with detergent, the PLA2-mediated release was found to be enhanced synergistically by saponin, a cholesterol-complexing agent. Meanwhile, an additive interaction between PLA2 and lysolecithin suggests that PLA2 action is independent of lysolecithin. It is suggested that the binding of PLA2 to specific sites of membranes, probably rich in GPI-anchored glycoproteins, may be related to the facilitated release of GPI-anchored proteins as amphiphilic form.

Regulation of brain glycosylphosphatidylinositol-specific phospholipase D by natural amphiphiles.

Brain glycosylphosphatidylinositol-specific phospholipase D (GPI-PLD)-catalyzed conversion of amphiphilic form of Zn2+ -glycerophosphocholine cholinephosphodiesterase (Amp-GPC PDE) into hydrophilic form was investigated in the presence of natural amphiphiles. Monoacylglycerols enhanced considerably the conversion by GPI-PLD of Amp-GPC PDE to hydrophilic form, with the enhancing effect of monoacylglycerols being dependent on the size of acyl group (C8-C18). Whereas the maximal enhancement of GPI-PLD action was the greatest with monodecanoylglycerol, the concentration (EC50) required to achieve 50% maximal effect was the smallest for monomyristoyl- or monopalmitoylglycerol. In addition, monolaurylglycerol or its alkyl analogue, monododecylglycerol, showed a remarkable decrease in enhancing effect at high concentrations (>1 mM). Presence of double bond in acyl chain, as exemplified by monooleoylglycerol or mono-11-eicosenoin, further enhanced the conversion by GPI-PLD. Meanwhile, lysophosphatidylcholine (IC50, 25 microM) and phosphatidic acid (IC50, >100 microM), ionic amphiphiles, inhibited the GPI-PLD activity, which was determined in the presence of monooleoylglycerol as a detergent. From these results, it is suggested that the activity of GPI-PLD in vivo system may be regulated by natural amphiphiles.

Ascorbate-induced oxidative inactivation of Zn2+-glycerophosphocholine cholinephosphodiesterase.

Zn2+-glycerophosphocholine cholinephosphodiesterase, responsible for the conversion of glycerophosphocholine into glycerol and phosphocholine, was inactivated during incubation with ascorbic acid at 38 degrees C. The inclusion of copper ions or Fe2+ accelerated the ascorbate-induced inactivation, with Cu2+ or Cu+ being much more effective than Fe2+, suggestive of ascorbate-mediated oxidation. Dehydroascorbic acid had no effect on the phosphodiesterase, but H2O2 inactivated the enzyme in a concentration-dependent manner. Also, the enzyme was inactivated partially by a superoxide anion-generating system but not an HOCl generator. In support of involvement of H2O2 in the ascorbate action, catalase and superoxide dismutase expressed a complete and a partial protection, respectively. However, hydroxy radical scavengers such as mannitol, benzoate, or dimethyl sulfoxide were incapable of preventing the ascorbate action, excluding the participation of extraneous .OH. Although p-nitrophenylphosphocholine exhibited a modest protection against the ascorbate action, a remarkable protection was expressed by amino acids, especially by histidine. In addition, imidazole, an electron donor, showed a partial protection. Separately, when Cu2+-induced inactivation of the phosphodiesterase was compared with the ascorbate-mediated one, the protection and pH studies indicate that the mechanism for the ascorbate action is different from that for the Cu2+ action. Here, it is proposed that Zn2+-glycerophosphocholine cholinephosphodiesterase is one of brain membrane proteins susceptible to oxidative inactivation.

Active site of brain Zn2+-glycerophosphocholine cholinephosphodiesterase and regulation of enzyme activity.

Properties of active site of Zn2+-glycerophosphocholine cholinephosphodiesterase from ox brain were examined using substrates and inhibitors of the phosphodiesterase. The anionic binding site expressed a selectivity for a positively-charged group. Meanwhile, the glyceryl moiety-binding site appeared to be a narrow crevice of a limited size, excluding the entry of acylglycerophospholipids containing long acyl chains. While endogenous quaternary ammonium compounds such as phosphocholine, choline or carnitine inhibited the enzyme, divalent metal ions such as Co2+, Mn2+ or Zn2+ enhanced the activity by 1.5 to 2-folds. The pH dependence for the inhibition by phosphocholine or the hydrolysis of substrate implies the involvement of a basic amino acid residue with a pK value of 9.6-9.7, probably lysine, in the binding of phosphoryl group. In further support, the lysine modifiers such as trinitrobenzene sulfonic acid or diethylpyrocarbonate expressed some inactivation. The pH-rate profile indicates that an amino acid residue with a pK value of 10.2, presumably tyrosine, may participate as a nucleophile in the catalysis. This might be further supported by the inactivation of the enzyme by tyrosine modifiers such as tetranitromethane or HOI-generating system. Separately, the phosphodiesterase was observed to be susceptible to the action of hydrogen peroxide or peroxynitrite-generating system. From these results, it is implied that the phosphodiesterase may be affected by endogenous sources.

Interaction of divalent metal ions with Zn(2+)-glycerophosphocholine cholinephosphodiesterase from ox brain.

The effect of divalent metal ions on the activity of glycerophosphocholine cholinephosphodiesterse from ox brain was examined. Zn(2+)- and Co(2+)-glycerophosphocholine cholinephosphodiesterases were prepared from the exposure of apoenzyme to Zn2+ and Co2+, respectively, and the properties of two metallo-phosphodiesterases were compared to those of native phosphodiesterase. Although two metallo-enzymes were similar in expressing Km value, optimum pH or sensitivity to Cu2+, they differed in the susceptibility to the inhibition by thiocholine or tellurite; while Co(2+)-phosphodiesterase was more sensitive to tellurites, Zn(2+)-phosphodiesterase was more susceptible to inhibition by thiocholine. In addition, Zn(2+)-phosphodiesterase was more thermo-stable than Co2+ enzyme. Separately, when properties of native phosphodiesterase were compared to those of each metallo-phosphodiesterase, native phosphodiesterase was found to be quite similar to Zn(2+)-phosphodiesterase in many respects. Even in thermo-stability, native enzyme resembled Zn(2+)-phosphodiesterase rather than Co(2+)-enzyme. Consistent with this, the stability of native phosphodiesterase was maintained in the presence of Zn2+, but not Co2+, Mn2+ was also as effective as Zn2+ in the stabilization of the enzyme. Noteworthy, the native enzyme was found to be inhibited competitively by Cu2+ with a Ki value of 20 microM, and its inhibitory action was antagonized effectively by Zn2+ or Co2+. Also, choline, another competitive inhibitor of the enzyme, appeared to antagonize the inhibitory action of Cu2+. Taken together, it is suggested that there may be multiple binding sites for divalent metal ions in the molecule of glycerophosphocholine cholinephosphodiesterase.

Properties of a Zn(2+)-glycerophosphocholine cholinephosphodiesterase from bovine brain membranes.

A Zn(2+)-glycerophosphocholine cholinephosphodiesterase was purified with a specific activity of 4.6 mumole/min.mg protein from bovine brain membranes by procedures involving PI-PLC solubilization, concanavalin A affinity chromatography, CM-sephadex chromatography and Sephadex G-150 chromatography. Based on molecular weight determination gel chromatography and SDS polyacrylamide gel electrophoresis, the phosphodiesterase activity appears to be a dimeric protein (110 kDa) composed of two subunits with a molecular weight of approximately 54 kDa. The K(m) value for p-nitrophenylphosphocholine and the optimum pH were found to be 16 microM and pH 10.5, respectively. The phosphodiesterase was inhibited by Cu2+, but not the other divalent metal ions. The activity of the apoenzyme was remarkably activated by Co2+ or Zn2+, but not Mn2+ or Mg2+. In addition, the inactivation of the enzyme in glycine buffer was prevented by Mn2+ or Zn2+, but not Co2+ or Mg2. In a separate experiment, comparing properties of the purified and membrane-bound phosphodiesterases, the forms of two enzymes were quite similar except in stability. Both enzymes were more stable at pH 7.4 than pH 5 or 10. However, the membrane-bound enzyme was more stable than the soluble enzyme at all three pHs. These data suggest that the activity of the phosphodiesterase may be stabilized in-vivo.

Involvement of both a Zn2+ site and an anionic binding site in the selective inhibition of a Zn2+-glycerophosphocholine cholinephosphodiesterase by thiols and tellurites

Inhibition of a Zn(2+)-glycerophosphocholine cholinephosphodiesterase by thiols or tellurites were examined mechanistically. Inactivation of the phosphodiesterase by thio-carboxylates, which was due to the removal of Zn2+ in the catalytic site, was enhanced by introduction of an amino group in the structure of thiols, suggesting the presence of an anionic site adjacent to a Zn2+ site. In support of the suggestion, it was found that thiols, associable with both a Zn2+ site and an anionic site, were more potent reversible inhibitors; dimethylaminoethanethiol (Ki, 17 microM), diethylaminoethanethiol (Ki, 1.2 microM) and thiocholine (Ki, 2.6 microM). Meanwhile, the inhibition of the phosphodiesterase by tellurites is ascribed to the binding of tellurite anions to a Zn2+ site, based on the protective action of tellurite anions against the inactivation of the enzyme by EDTA. Moreover, the inhibition of the phosphodiesterase by tellurites was prevented by phosphate ions, which expressed the protective effect against EDTA inactivation. In further support, it was observed that tellurite and thiocholine appeared to interact with active site in an additive manner, in contrast to a synergistic action between tellurites and quaternary ammonium compounds such as acetylcholine or choline.

Brain myelin-bound Zn(2+)-glycerophosphocholine cholinephosphodiesterase is a glycosylphosphatidylinositol-anchored enzyme of two different molecular forms.

An Zn(2+)-GPC cholinephosphodiesterase activity, which is present more predominantly in myelin than in microsome or cytosol, has been examined using rho-nitrophenylphosphocholine as a substrate. In the solubilization of enzyme activity from myelin membranes, lysolecithin was found to be more effective than Triton X-100 or deoxycholate. Especially, the myelin-bound phosphodiesterase was suggested to be a glycosylphosphatidyl-inositol-anchored protein, based on solubilization by B. cereus phospholipase C and Triton X-114 phase separation. Interestingly, it was found that while phospholipase C-solubilized enzyme, a hydrophilic protein, was associable with Concanavalin A column, detergent-solubilized amphiphilic form of enzyme was not. Either detergent extract or cytosol was observed to contain both amphiphilic form and hydrophilic one. In CM-sephadex chromatography, the soluble hydrophilic phosphodiesterase was observed to be separatable into two forms of enzyme. In comparative studies, both forms of phosphodiesterase showed much similarity in substrate specificity, optimum pH, Km value and Zn2+ requirement, although they differed in charge property and molecular weight.

Characterization of a Zn(2+)-requiring glycerophosphocholine cholinephosphodiesterase possessing p-nitrophenylphosphocholine phosphodiesterase activity.

p-Nitrophenylphosphocholine phosphodiesterase activity was purified 5000-fold from mouse brain by treatment of membranes with Bacillus cereus phospholipase C preparation and sequential chromatographies on concanavalin A-Sepharose and CM-Sephadex columns. The phosphodiesterase (Zn(2+)-requiring) showed Km and Vmax. values of 5.5 microM and 4.2 mumol/min per mg respectively in the hydrolysis of p-nitrophenylphosphocholine, and possessed an optimum pH of 10.5 and a molecular mass of approx. 74 kDa. The purified enzyme was found to convert glycerophosphocholine into glycerol and phosphocholine, with Km and Vmax. of 48 microM and 5 mumol/min per mg respectively. In the hydrolysis of glycerophosphocholine the enzyme also exhibited a Zn2+ requirement and optimal pH at 10.5. Additionally, the p-nitrophenylphosphocholine phosphodiesterase activity was competitively inhibited by glycerophosphocholine, with a Ki value of 50 microM. These observations, together with chromatographic behaviour and heat-denaturation analyses, indicate that both p-nitrophenylphosphocholine phosphodiesterase and glycerophosphocholine cholinephosphodiesterase activities reside in the same protein.

Selective inhibition of Zn(2+)-glycerophosphocholine cholinephosphodiesterase by tellurium tetrachloride.

A Zn(2+)-glycerophosphocholine cholinephosphodiesterase (EC 3.1.4.38) purified from mouse brain was found to be reversibly inhibited by tellurium tetrachloride. This effect was characterized by a competitive pattern of inhibition, with apparent Ki values of 0.7 microM and 1.5 microM for the hydrolysis of p-nitrophenylphosphocholine and glycerophosphocholine respectively. Interestingly, the inhibitory effect of tellurium tetrachloride was found to be greatly potentiated by tetramethylammonium salt, indicative of a synergistic interaction between the two compounds. Additionally, it was observed that the effect of tellurium tetrachloride was not affected by a number of other metal ions, and was more pronounced at neutral pH, suggesting that the inhibitory role of the tellurium tetrachloride may be of importance under physiological conditions. Thus Zn(2+)-glycerophosphocholine cholinephosphodiesterase is proposed to be one of the target enzymes which is susceptible to the inhibitory effect of tellurium tetrachloride.

Evidence for a membrane defect in Alzheimer disease brain.

To determine whether neurodegeneration in Alzheimer disease brain is associated with degradation of structural cell membrane molecules, we measured tissue levels of the major membrane phospholipids and their metabolites in three cortical areas from postmortem brains of Alzheimer disease patients and matched controls. Among phospholipids, there was a significant (P less than 0.05) decrease in phosphatidylcholine and phosphatidylethanolamine. There were significant (P less than 0.05) decreases in the initial phospholipid precursors choline and ethanolamine and increases in the phospholipid deacylation product glycerophosphocholine. The ratios of glycerophosphocholine to choline and glycerophosphoethanolamine to ethanolamine were significantly increased in all examined Alzheimer disease brain regions. The activity of the glycerophosphocholine-degrading enzyme glycerophosphocholine choline-phosphodiesterase was normal in Alzheimer disease brain. There was a near stoichiometric relationship between the decrease in phospholipids and the increase of phospholipid catabolites. These data are consistent with increased membrane phospholipid degradation in Alzheimer disease brain. Similar phospholipid abnormalities were not detected in brains of patients with Huntington disease, Parkinson disease, or Down syndrome. We conclude that the phospholipid abnormalities described here are not an epiphenomenon of neurodegeneration and that they may be specific for the pathomechanism of Alzheimer disease.

Alterations of phospholipid metabolites in postmortem brain from patients with Alzheimer's disease.

To test the hypothesis that brain cell membranes degenerate in Alzheimer's disease (AD), we measured the levels of phospholipids, their water-soluble metabolites, and glycerophosphocholine (GPC) cholinephosphodiesterase activity in postmortem brain tissue from patients with AD and age-matched controls. We found significantly higher levels of the phospholipid catabolite GPC in AD brain. In contrast, choline and ethanolamine levels were significantly lower in AD, and phospholipid levels were slightly decreased. Furthermore, in AD the activity of the GPC-degrading enzyme GPC cholinephosphodiesterase was unaltered. Our results indicate that membrane phospholipid catabolism is increased in AD brain. Inasmuch as the tissue levels of initial phospholipid precursors were decreased, we suggest that phospholipid turnover is elevated in this neurodegenerative disease.

The hydrolysis of glycerophosphocholine by rat brain microsomes: activation and inhibition.

Experiments with glycerophosphocholine phosphodiesterase (GPC diesterase, EC 3.1.4.2.) in rat brain microsomes suggest that, although its activity is inhibited by low concentrations of calmidazolium, its dependence on Ca2+ ions is not modulated by calmodulin. The activity of glycerophosphocholine choline phosphodiesterase (choline phosphohydrolase, EC 3.1.4.38) was much lower than that of the GPC diesterase. A relatively inexpensive method for the preparation of sn-glycero-3-phospho [Me-14C]choline is described.