Search

The enzymatic activity of ammodytoxins (Atxs), secreted phospholipases A(2) (sPLA(2)s) in snake venom, is essential for expression of their presynaptic neurotoxicity, but its exact role in the process is unknown. We have analyzed in detail the enzymatic properties of Atxs, their mutants, and homologues. The apparent rates of phospholipid hydrolysis by the sPLA(2)s tested vary by up to 4 orders of magnitude, and all enzymes display a strong preference for vesicles containing anionic phospholipids, phosphatidylglycerol or phosphatidylserine (PS), over those containing zwitterionic phosphatidylcholine (PC). Nevertheless, Atxs are quite efficient in hydrolyzing pure PC vesicles as well as PC-rich plasma membranes of intact HEK293 cells. The presence of anionic phospholipids in PC vesicles dramatically increases the interfacial binding affinity and catalytic activity of Atxs, but not of their nontoxic homologue ammodytin I(2), that displays unusually low binding affinity and enzymatic activity on PS-containing vesicles and HEK293 plasma membranes. Aromatic and hydrophobic residues on the interfacial binding surface of Atxs are important for productive binding to both zwitterionic and anionic vesicles, while basic and polar residues have a negative impact on binding to zwitterionic vesicles. When tightly bound to the membrane interface, Atxs can reach full enzymatic activity at low micromolar concentrations of Ca(2+). Although Atxs have evolved to function as potent neurotoxins that specifically target presynaptic nerve terminals, they display a high degree of phospholipolytic efficiency on various phospholipid membranes.

An increasing number of mammalian secreted phospholipases A(2) (sPLA(2)s) has been identified over the past few years. Here, we report the identification and recombinant expression of a novel sPLA(2)-like protein in mouse and human species that has been called group XIIB (GXIIB). The mature protein has a molecular mass of 19.7 kDa and structural features similar to those of the previously identified GXII sPLA(2), now called GXIIA. Strikingly, the GXIIB sPLA(2) has a mutation in the active site, replacing the canonical histidine by a leucine, suggesting that this sPLA(2) is catalytically inactive. Recombinant expression of human (hGXIIB) and mouse (mGXIIB) sPLA(2)s in Escherichia coli indicates that GXIIB sPLA(2)s display no measurable lipolytic activity on various types of phospholipid substrates. Furthermore, these sPLA(2)-like proteins display relatively weak affinity to phospholipid vesicles. Binding experiments indicate that these proteins are also unable to bind to the well-known M-type sPLA(2) receptor. The RNA tissue distribution of GXIIB sPLA(2)s is distinct from that of other sPLA(2)s including the homologous GXIIA. Strong expression was observed in liver, small intestine, and kidney in both human and mouse species. Interestingly, the expression of the novel sPLA(2) is dramatically decreased in human tumors from the same tissues. The absence of enzymatic activity suggests that the GXIIB sPLA(2)-like proteins probably exert their biological roles by acting as ligands for as yet unidentified receptors.

An important characteristic of the human group IIA secreted phospholipase A(2) (IIA PLA(2)) is the extremely low activity of this enzyme with phosphatidylcholine (PC) vesicles, mammalian cell membranes, and serum lipoproteins. This characteristic is reflected in the lack of ability of this enzyme to bind productively to zwitterionic interfaces. Part of the molecular basis for this lack of activity is an absence of tryptophan, a residue with a known preference for residing in the interfacial region of zwitterionic phospholipid bilayers. In this paper we have replaced the eight residues that make up the hydrophobic collar on the interfacial binding surface of the enzyme with tryptophan. The catalytic and interfacial binding properties of these mutants have been investigated, particularly those properties associated with binding to and hydrolysis of zwitterionic interfaces. Only the insertion of a tryptophan at position 3 or 31 produces mutants that significantly enhance the activity of the human IIA enzyme against zwitterionic interfaces and intact cell membranes. Importantly, the ability of the enzyme mutants to hydrolyze PC-rich interfaces such as the outer plasma membrane of mammalian cells was paralleled by enhanced interfacial binding to zwitterionic interfaces. The corresponding double tryptophan mutant (V3,31W) displays a specific activity on PC vesicles comparable to that of the human group V sPLA2. This enhanced activity includes the ability to interact with human embryonic kidney HEK293 cells, previously reported for the group V enzyme [Kim, Y. J., Kim, K. P., Rhee, H. J., Das, S., Rafter, J. D., Oh, Y. S., and Cho, W. (2002) J. Biol. Chem. 277, 9358-9365].

Human plasma platelet activating factor acetylhydrolase (pPAF-AH) is a phospholipase A(2) that specifically hydrolyzes the sn-2 ester of platelet activating factor (PAF) and of phospholipids with oxidatively truncated sn-2 fatty acyl chains. pPAF-AH is bound to lipoproteins in vivo, and it binds essentially irreversibly to anionic and zwitterionic phospholipid vesicles in vitro and hydrolyzes PAF and PAF analogues. Substrate hydrolysis also occurs in the absence of vesicles, with a maximum rate reached at the critical micelle concentration. A novel pre-steady-state kinetic analysis with enzyme tightly bound to vesicles and with a substrate that undergoes slow intervesicle exchange establishes that pPAF-AH accesses its substrate from the aqueous phase and thus is not an interfacial enzyme. Such a mechanism readily explains why this enzyme displays dramatic specificity for phospholipids with short sn-2 chains or with medium-length, oxidatively truncated sn-2 chains since a common feature of these lipids is their relatively high water solubility. It also explains why the enzymatic rate drops as the length of the sn-1 chain is increased. pPAF-AH shows broad specificity toward phospholipids with different polar headgroups. Additional results are that PAF undergoes intervesicle exchange on the subminute time scale and it does not undergo transbilayer movement over tens of minutes.

Cytosolic group IV phospholipase A2 (cPLA2) plays a role in liberating arachidonic acid from the sn-2 position of mammalian cellular phospholipids. The enzyme consists of a catalytic domain joined to an N-terminal calcium-dependent, membrane binding domain (C2 domain). The interfacial binding properties of the full-length, nonphosphorylated enzyme and its C2 domain to phospholipid vesicles were studied as a function of vesicle phospholipid composition and calcium concentration. The binding of cPLA2 to phosphatidylcholine vesicles is mostly governed by its C2 domain; binding is relatively weak, and calcium enhances binding and interfacial catalysis by about 10-fold. Catalytically productive interfacial binding was measured by monitoring the increase in the rate of cPLA2-catalyzed hydrolysis of a fluorimetric substrate present in vesicles as a function of bulk vesicle concentration. Enzyme-vesicle binding was also measured by fluorescence as was enzyme-calcium binding. Compared to zwitterionic vesicles, cPLA2 binding to anionic phosphatidylmethanol vesicles is of higher affinity and calcium-independent, although calcium is required for the binding of the C2 domain to these anionic vesicles. cPLA2 is fully catalytically active on phosphatidylmethanol vesicles in the absence of calcium. Phosphatidylserine is not a good replacement for phosphatidylmethanol for inducing high-affinity, calcium-independent binding of cPLA2. These results reveal two modes of catalytically productive interfacial binding of cPLA2: calcium-dependent anchoring via the C2 domain and a calcium-independent component involving a phosphatidylmethanol recognition element in the catalytic domain. They also show that membrane binding of cPLA2 is not, in general, predicted by the interfacial binding properties of its C2 domain.

Human secretory group IIa phospholipase A2 (hIIa-PLA2) contains a large number of prominent cationic patches on its molecular surface and has exceptionally high affinity for anionic surfaces, including anionic membranes. To identify the cationic amino acid residues that support binding of hIIa-PLA2 to anionic membranes, we have performed extensive site-directed mutagenesis of this protein and measured vesicle binding and interfacial kinetic properties of the mutants using polymerized liposomes and nonpolymerized anionic vesicles. Unlike other secretory PLA2s, which have a few cationic residues that support binding of enzyme to anionic membranes, interfacial binding of hIIa-PLA2 is driven in part by electrostatic interactions involving a number of cationic residues forming patches on the putative interfacial binding surface. Among these residues, the amino-terminal patch composed of Arg-7, Lys-10, and Lys-16 makes the most significant contribution to interfacial adsorption, and this is supplemented by contributions from other patches, most notably Lys-74/Lys-87/Arg-92 and Lys-124/Arg-127. For these mutants, complete vesicle binding occurs in the presence of high vesicle concentrations, and under these conditions the mutants display specific activities comparable to that of wild-type enzyme. These studies indicate that electrostatic interactions between surface lysine and arginine residues and the interface contribute to interfacial binding of hIIa-PLA2 to anionic vesicles and that cationic residues closest to the opening of the active-site slot make the most important interactions with the membrane. However, because the wild type binds extremely tightly to anionic vesicles, it was not possible to exactly determine what fraction of the total interfacial binding energy is due to electrostatics.

Analysis of phospholipases A2 on model phospholipid bilayers in which enzyme is essentially irreversibly bound at the lipid-water interface, termed "scooting mode", is a useful tool for studying the kinetic properties of interfacial enzymes. It is shown that human cytosolic 85 kDa phospholipase A2 (cPLA2) hydrolyzes sn-2-arachidonyl-containing phospholipids or the gamma-linolenoyl ester of 7-hydroxycoumarin (GLU) dispersed in vesicles of 1,2-dioleoyl-sn-glycero-3-phosphomethanol (L-DOPM) in the scooting mode. Trapping of cPLA2 on L-DOPM vesicles is rapid and independent of product formation. Slowing of cPLA2-catalyzed hydrolysis of substrates present in phosphatidylmethanol and phosphatidylcholine vesicles is primarily due to apparent inactivation rather than to substrate depletion. cPLA2 phosphorylated on serine 505 by mitogen-activated protein kinase displays a 30% increase in the rate of sn-2-arachidonylphosphatidylcholine hydrolysis in the scooting mode compared to that of the nonphosphorylated enzyme. Kinetic parameters of cPLA2 acting on a variety of different phosphatidylmethanol vesicles were evaluated, and the results are discussed in terms of active site affinities for substrates and of lateral organization of substrates in the bilayer. A key result is that the sigmoidal kinetics reported previously using 1,2-dimyristoyl-sn-glycero-3-phosphomethanol (DMPM) vesicles are most prominent near the phase transition temperature of DMPM. No sigmoidal kinetics was observed using L-DOPM vesicles. The results of kinetic experiments and the behavior of a fluorescent substrate analog are consistent with nonideal mixing of substrate in DMPM vesicles, but not in L-DOPM vesicles, suggesting that apparent saturation and sigmoidal kinetics are more a result of nonideal mixing of substrate in DMPM vesicles than of active site binding of substrate. The fluorescence assay described using L-DOPM/GLU vesicles is useful for evaluating the interfacial behavior of cPLA2, including its substrate preferences and the effect of active site-directed inhibitors.

The small GTP-binding protein G25K and the protein K-Ras 4B contain prenyl groups (geranylgeranyl and farnesyl, respectively) that are thioether linked to a C-terminal cysteine which is methylated on its alpha-carboxyl group. These proteins, like many other prenyl proteins, also have a string of basic residues near their C-termini. A series of prenylated peptides based on the C-terminal sequences of human brain G25K and human K-Ras 4B were synthesized and analyzed for their membrane binding affinities. G25K peptides containing an N-terminal N-acetyltryptophan group were studied because their binding to membranes containing a trace of dansylated phospholipid could be detected by fluorescence resonance energy transfer. The G25K peptide lacking a prenyl group and a C-terminal methyl ester did not detectably bind to vesicles, and binding was enhanced by more than 500-fold if the peptide was geranylgeranylated. For the farnesylated peptide, methylation of the C-terminus increased membrane affinity by at least 60-fold if the vesicles contained phosphatidylserine and by 3-fold if they lacked this acidic lipid. The geranylgeranylated and methylated G25K peptide remains irreversibly attached to vesicles over several minutes only if the vesicles contain phosphatidylserine, whereas the corresponding nonmethylated or farnesylated and methylated peptides dissociate rapidly (less than a few seconds) from neutral or anionic vesicles. Farnesylation of the nonmethylated K-Ras 4B peptide enhances its affinity to vesicles containing acidic phospholipids (phosphatidylglycerol or phosphatidylserine) by 70-fold, and methylation leads to an additional dramatic (150-fold) increase in membrane affinity.(ABSTRACT TRUNCATED AT 250 WORDS)

A glycerophospholipid:cholesterol acyltransferase (GCAT) that also has lipase activity is secreted by the bacterium Aeromonas hydrophila. Hydrolysis of the sn-2-ester bond of 1,2-dimyristoyl-sn-glycero-3-phosphomethanol (DMPM) vesicles by this enzyme is shown to occur in a highly processive scooting mode in which the enzyme, substrate, and the products of hydrolysis remain bound to the vesicle interface. This conclusion is based on the following observations. (a) When there is an excess of vesicles over enzyme, the hydrolysis of the sn-2-acyl group ceases after only a fraction of the total available substrate is hydrolyzed. Addition of more enzyme, but not of more substrate, leads to a new round of hydrolysis. (b) The extent of hydrolysis of vesicles per enzyme increases with the size of the vesicles, and it corresponds to the total hydrolysis of the outer monolayer of one vesicle by one enzyme. (c) The enzyme bound to vesicles composed of reaction products or of the non-hydrolyzable phospholipid 1,2-ditetradecyl-sn-glycero-3-phosphomethanol (DTPM) is not able to undergo intervesicle exchange. Instead, intervesicle transfer of the substrate or the bound enzyme due to vesicle fusion promotes hydrolysis of all of the vesicles present in the reaction mixture. (d) Addition of DTPM vesicles to a reaction mixture containing DMPM substrate vesicles and the enzyme has no noticeable effect on the course of hydrolysis. Substrate specificity studies in the scooting mode on DMPM vesicles reveal that GCAT displays essentially no selectivity in the hydrolysis of phospholipids with different polar head groups. Treatment of GCAT with trypsin, which removes a small peptide, results in an enzyme that displays comparable catalytic activity but increased affinity for the interface. Alkyltrifluoromethyl ketones are shown to be tight-binding competitive inhibitors of GCAT. The scooting mode analysis, which has previously been shown to provide a simplified approach for analyzing the steady-state kinetics of interfacial catalysis by secreted phospholipase A2, is also useful for analyzing the interfacial kinetic behavior of lipases.

A trifluoromethyl ketone analogue of arachidonic acid in which the COOH group is replaced with COCF3 (AACOCF3) was prepared and found to be a tight- and slow-binding inhibitor of the 85-kDa cytosolic human phospholipase A2 (cPLA2). Enzyme inhibition was observed when AACOCF3 was tested in assays using either phospholipid vesicles or phospholipid/Triton X-100 mixed micelles. The fact that the inhibition developed over several minutes in both assays establishes that AACOCF3 inhibits by direct binding to the enzyme rather than by decreasing the fraction of enzyme bound to the substrate interface. From the measured values of the inhibitor association and dissociation rate constants, an upper limit of the equilibrium dissociation constant for the Ca(2+).AACOCF3.PLA2 complex of 5 x 10(-5) mole fraction was obtained. Thus, detectable inhibition of cPLA2 by AACOCF3 occurs when this compound is present in the assay at a level of one inhibitor per several thousand substrates. Arachidonic acid analogues in which the COOH group is replaced by COCH3, CH(OH)CF3, CHO, or CONH2 did not detectably inhibit the cPLA2. The arachidonyl ketones AACOCF2CF3 and AACOCF2Cl were found by 19F NMR to be less hydrated than AACOCF3 in phospholipid/Triton X-100 mixed micelles, and compared to AACOCF3 these compounds are also weaker inhibitors of cPLA2. In keeping with the fact that cPLA2 displays substrate specificity for arachidonyl-containing phospholipids, the arachidic acid analogue C19H39COCF3 is a considerably less potent inhibitor compared to AACOCF3.(ABSTRACT TRUNCATED AT 250 WORDS)

The kinetics of the phospholipase A2-catalyzed hydrolysis of bilayer vesicles and mixed micelles of several oxyglycero and thioglycero analogues of phospholipids have been studied. The results with vesicles show that, depending on the source of the enzyme, the rates of hydrolysis of the oxy-containing long-chain phosphatidylmethanols are 2.5- to 28-fold higher compared to the rates of hydrolysis of the analogous thio substrates. The oxygen to sulfur substitution does not significantly alter the affinities of the enzymes for the reaction products or calcium. Since it is unlikely that sulfur substitution changes the rate constants for the formation and dissociation of the enzyme-product complex by the same factor, the element effects seen in the rates of hydrolysis of the oxy- and thioester phospholipids in vesicles are primarily due to a change in the rate constant for the chemical step of the catalytic turnover cycle. For bovine pancreatic phospholipase A2, various mutants with lower catalytic activity were used to show that the value of the element effect does not increase in the mutants. These results establish that, for the pancreatic phospholipase A2, the element effect is fully expressed, and the chemical step is fully rate-limiting for both oxyglycero and thioglycero phospholipids in vesicles. It was found that the element effect decreases from 7 to 1 when long-chain phosphatidylmethanols are present in micelles of a neutral diluent. This result suggests that the chemical step is not rate-limiting during the hydrolysis of these mixed micelle substrates.

Difluorostatine- and difluorostatone-containing peptides have been evaluated as potent inhibitors of penicillopepsin, a member of the aspartic proteinase family of enzymes. Isovaleryl-Val-Val-StaF2NHCH3 [StaF2 = (S)-4-amino-2,2-difluoro-(R)-3-hydroxy-6-methylheptanoic acid] and isovaleryl-Val-Val-StoF2NHCH3 [StoF2 = (S)-4-amino-2,2-difluoro-3-oxo-6-methylheptanoic acid] have measured Ki's of 10 x 10(-9) and 1 x 10(-9) M, respectively, with this fungal proteinase. The StoF2-containing peptide binds 32-fold more tightly to the enzyme than the analogous peptide containing the non-fluorinated statine ethyl ester. Each compound was cocrystallized with penicillopepsin, intensity data were collected to 1.8-A resolution, and the atomic coordinates were refined to an R factor [formula: see text] of 0.131 for both complexes. The inhibitors bind in the active site of penicillopepsin in much the same fashion as do other statine-containing inhibitors of penicillopepsin analyzed earlier [James, M. N. G., Sielecki, A. Salituro, F., Rich, D. H., & Hofmann, T. (1982) Proc. Natl. Acad. Sci. U.S.A. 79, 6137-6141; James, M.N.G., Sielecki, A., & Hofmann, T. (1985) in Aspartic Proteinases and their Inhibitors (Kosta, V., Ed.) pp 163-177, Walter deGruyter, Berlin]. The (R)-3-hydroxyl group in StaF2 binds between the active site carboxyl groups of Asp33 and Asp213, making hydrogen-bonding contacts to each one. The ketone functional group of the StoF2 inhibitor is bound as a hydrated species, with the gem-diol situated between the two aspartic acid carboxyl groups in a manner similar to that predicted for the tetrahedral intermediate expected during the catalytic hydrolysis of a peptide bond [James, M. N. G., & Sielecki, A. (1985) Biochemistry 24, 3701-3713]. One hydrogen-bonding interaction from the "outer" hydroxyl group is made to O delta 1 of Asp33, and the "inner" hydroxyl group forms two hydrogen-bonding contacts, one to each of the carboxyl groups of Asp33 (O delta 2) and Asp213 (O delta 2). The only structural difference between the StaF2 and StoF2 inhibitors that accounts for the factor of 10 in their Ki's is the additional (R)-3-OH group on the tetrahedral sp3 carbon atom of the hydrated StoF2 inhibitor. The intermolecular interactions involving the fluorine atoms of each inhibitor are normal van der Waals contacts to one of the carboxyl oxygen atoms of Asp213 (F2-O delta 2 Asp213, 2.9 A). The observed stereochemistry of the bound StoF2 group in the active site of penicillopepsin has stimulated our reappraisal of the catalytic pathway for the aspartic proteinases.(ABSTRACT TRUNCATED AT 400 WORDS)

The kinetics of hydrolysis of 1-palmitoyl-2-arachidonyl-sn-glycero-3-phosphocholine vesicles catalyzed by the high molecular weight phospholipase A2 from rat kidney show an anomalous behavior. The reaction progress lasts for several minutes and then stops after only 5-10% of the available substrate has been hydrolyzed. Addition of more enzyme but not more substrate leads to a new round of hydrolysis. Although this initially suggested that the enzyme becomes inactivated during the turnover, such a conclusion could not be substantiated. Addition of buffer containing 0.15 M NaCl and bovine serum albumin to the reaction after the progress ceased leads to the re-initiation of the lipolysis. The enzyme is not strongly inhibited by the reaction products. Although the enzyme does not bind irreversibly to vesicles composed of pure 1-palmitoyl-2-arachidonyl-sn-glycero-3-phosphocholine, it does become irreversibly trapped on vesicles that contain a critical mole percentage of reaction products. This trapping is the most likely explanation for the cessation of the reaction progress. Both the binding of enzyme to 1,2-dipalmitoyl-sn-glycero-3-phosphocholine vesicles and the hydrolysis of 1-stearoyl-2-[3H]arachidonyl-sn-glycerophosphocholine contained in these vesicles require the presence of products. Furthermore, the trapping of enzyme is independent of catalytic turnover. The trapping is sensitive to the structure of the fatty acid present in the vesicles and requires the presence of divalent metals (either Ca2+, Sr2+, Ba2+, or Mg2+). Since the concentrations of the metals needed for the enzymatic activity correlate with the amounts needed to promote the trapping, it is suggested that the role of the metal is only to promote the interfacial binding of the enzyme.

Manoalogue, a synthetic analogue of the sea sponge-derived manoalide, has been previously shown to partially inactivate the phospholipase A2 from cobra venom (Reynolds, L. J., Morgan, B. P., Hite, E. D., Mihelich, E. D., & Dennis, E. A. (1988) J. Am. Chem. Soc. 110, 5172) by reacting with enzyme lysine residues. In the present study, the inactivation of the phospholipases A2 from pig pancreas, bee venom, and cobra (Naja naja naja) venom by manoalogue was studied in detail. Manoalogue-treated enzymes were examined in the scooting mode on vesicles of 1,2-dimyristoyl-sn-glycero-3-phosphomethanol. Here the native enzymes bound irreversibly to the vesicles and hydrolyzed all of the phospholipids in the outer monolayer without leaving the surface of the interface. All three manoalogue-treated enzymes showed reduced catalytic turnover for substrate hydrolysis in the scooting mode, and the modified enzymes did not hop from one vesicle to another. Thus, inactivation by manoalogue is not due to the decrease in the fraction of enzyme bound to the substrate interface. This result was also confirmed by fluorescence studies that directly monitored the binding of phospholipase A2 to vesicles. A chemically modified form of the pig pancreatic phospholipase A2 in which all of the lysine epsilon-amino groups have been amidinated was not inactivated by manoalogue, indicating that the modification of lysine residues and not the amino-terminus is required for the inactivation. Several studies indicated that the manoalogue-modified enzymes contain a functional active site. For example, studies that monitored the protection by ligands of the active site from attack by a alkylating agent showed that manoalogue-modified pig phospholipase A2 was capable of binding calcium, a substrate analogue, lipolysis products, and a competitive inhibitor. Furthermore, relative to native enzymes, manoalogue-modified enzymes retained significantly higher catalytic activities when acting on water-soluble substrates than when acting on vesicles in the scooting mode. Intact manoalogue had no affinity for the catalytic site on the enzyme as it did not inhibit the enzyme in the scooting mode and it did not protect the active site from alkylation. Pig pancreatic phospholipase A2 bound to micelles of 2-hexadecyl-sn-glycero-3-phosphocholine was resistant to inactivation by manoalogue, suggesting that the modification of lysine residues on the interfacial recognition surface of the enzyme was required for inactivation.(ABSTRACT TRUNCATED AT 400 WORDS)

The kinetics of hydrolysis of phospholipid vesicles by phospholipase A2 (PLA2) in the scooting mode can be described by the Michaelis-Menten formalism for the action of the enzyme in the interface (E*). E* + S in equilibrium E*S in equilibrium E*P in equilibrium E* + Products The values of the interfacial rate constants cannot be obtained by classical methods because the concentration of the substrate within the lipid bilayer is not easily manipulated. In the present study, carbonyl-carbon heavy atom isotope effects for the hydrolysis of phospholipids have been measured in both vesicles and in mixed micelles in which the phospholipid was present in the nonionic detergent Triton X-100. A large [14C]carbonyl carbon isotope effect of 1.12 +/- 0.02 was measured for the cobra venom PLA2-catalyzed hydrolysis of dipalmitoylphosphatidylcholine in Triton X-100. In contrast, no isotope effect (1.01 +/- 0.01) was measured for the action of the porcine pancreatic and cobra venom enzymes on vesicles of dimyristoylphosphatidylmethanol in the scooting mode. In a second experiment, the hydrolysis of vesicles was carried out in oxygen-18 enriched water. Analysis of the released fatty acid product by mass spectrometry showed that it contained only a single oxygen-18. All of these results were used to estimate both the forward and reverse commitments to catalysis. The lack of doubly labeled fatty acid demonstrated that the product is released from the E*P complex faster than the reverse of the esterolysis step. The small isotope effect in vesicles demonstrated that the E*S complex goes on to products faster than substrate is released from the enzyme. The relevance of these results to an understanding of substrate specificity and inhibition of PLA2 is discussed. In addition, the conditions placed on the values of the rate constants obtained in the present study together with results obtained in the other studies described in this series of papers have led to the evaluation of most of the interfacial rate constants for the hydrolysis of phospholipid vesicles by PLA2.

The binding equilibrium of phospholipase A2 (PLA2) to the substrate interface influences many aspects of the overall kinetics of interfacial catalysis by this enzyme. For example, the interpretation of kinetic data on substrate specificity was difficult when there was a significant kinetic contribution from the interfacial binding step to the steady-state catalytic turnover. This problem was commonly encountered with vesicles of zwitterionic phospholipids, where the binding of PLA2 to the interface was relatively poor. The action of PLA2 on phosphatidylcholine (PC) vesicles containing a small amount of anionic phospholipid, such as phosphatidic acid (PA), was studied. It was shown that the hydrolysis of these mixed lipid vesicles occurs in the scooting mode in which the enzyme remains tightly bound to the interface and only the substrate molecules present on the outer monolayer of the target vesicle became hydrolyzed Thus the phenomenon of scooting mode hydrolysis was not restricted to the action of PLA2 on vesicles of pure anionic phospholipids, but it was also observed with vesicles of zwitterionic lipids as long as a critical amount of anionic compound was present. Under such conditions, the initial rate of hydrolysis of PC in the mixed PC/PA vesicles was enhanced more than 50-fold. Binding studies of PLA2 to vesicles and kinetic studies in the scooting mode demonstrated that the enhancement of PC hydrolysis in the PC/PA covesicles was due to the much higher affinity of the enzyme toward covesicles compared to vesicles of pure PC phospholipids. A novel and technically simple protocol for accurate determination of the substrate specificity of PLA2 at the interface was also developed by using a double-radiolabel approach. Here, the action of PLA2 in the scooting mode was studied on vesicles of the anionic phospholipid 1,2-dimyristoyl-sn-glycero-3-phosphomethanol that contained small amounts of 3H- and 14C-labeled phospholipids. From an analysis of the 3H and 14C radioactivity in the released fatty acid products, the ratio of substrate specificity constants (kcat/KMS) was obtained for any pair of radiolabeled substrates. These studies showed that the PLA2s from pig pancreas and Naja naja naja venom did not discriminate between phosphatidylcholine and phosphatidylethanolamine phospholipids or between phospholipids with saturated versus unsaturated acyl chains and that the pig enzyme had a slight preference for anionic phospholipids (2-3-fold). The described protocol provided an accurate measure of the substrate specificity of PLA2 without complications arising from the differences in binding affinities of the enzyme to vesicles composed of pure phospholipids.

To date, 12 secreted phospholipases A2 (sPLA2s) have been identified in the mouse species and divided into three structural collections (I/II/V/X, III, and XII). On the basis of their different molecular properties and tissue distributions, each sPLA2 is likely to exert distinct functions by acting as an enzyme or ligand for specific soluble proteins or receptors, among which the M-type receptor is the best-characterized target. Here, we present the properties of binding of the full set of mouse sPLA2s to the mouse M-type receptor. All enzymes have been produced in Escherichia coli or insect cells, and their properties of binding to the cloned and native M-type receptor have been determined. sPLA2s IB, IIA, IIE, IIF, and X are high-affinity ligands (K0.5 = 0.3-3 nM); sPLA2s IIC and V are low-affinity ligands (K0.5 = 30-75 nM), and sPLA2s IID, III, XIIA, and XIIB bind only very weakly or do not bind to the M-type receptor (K0.5 > 100 nM). Three exogenous parvoviral group XIII PLA2s and two fungal group XIV sPLA2s do not bind to the receptor. Together, these results indicate that the mouse M-type receptor is selective for only a subset of mouse sPLA2s from the group I/II/V/X structural collection. Binding of mouse sPLA2s to a recombinant soluble mouse M-type receptor leads in all cases to inhibition of enzymatic activity, and the extent of deglycosylation of the receptor decreases yet does not abolish sPLA2 binding. The physiological meaning of binding of sPLA2 to the M-type receptor is discussed on the basis of our current knowledge of sPLA2 functions.

Mammalian secreted phospholipases A(2) (sPLA(2)s) constitute a family of structurally related enzymes that are likely to play numerous biological roles because of their phospholipid hydrolyzing activity and binding to soluble and membrane-bound proteins, including the M-type receptor. Over the past decade, a number of competitive inhibitors have been developed against the inflammatory-type human group IIA (hGIIA) sPLA(2) with the aim of specifically blocking its catalytic activity and pathophysiological functions. The fact that many of these inhibitors, including the indole analogue Me-Indoxam, inhibit several other sPLA(2)s that bind to the M-type receptor prompted us to investigate the impact of Me-Indoxam and other inhibitors on the sPLA(2)-receptor interaction. By using a Ca(2+) loop mutant derived from a venom sPLA(2) which is insensitive to hGIIA inhibitors but still binds to the M-type receptor, we demonstrate that Me-Indoxam dramatically decreases the affinity of various sPLA(2)s for the receptor, yet an sPLA(2)-Me-Indoxam-receptor complex can form at very high sPLA(2) concentrations. Me-Indoxam inhibits the binding of iodinated mouse sPLA(2)s to the mouse M-type receptor expressed on live cells but also enhances binding of sPLA(2) to phospholipids. Because Me-Indoxam and other competitive inhibitors protrude out of the sPLA(2) catalytic groove, it is likely that the inhibitors interfere with the sPLA(2)-receptor interaction by steric hindrance and to different extents that depend on the type of sPLA(2) and inhibitor. Our finding suggests that the various anti-inflammatory therapeutic effects of sPLA(2) inhibitors may be due not only to inhibition of enzymatic activity but also to modulation of binding of sPLA(2) to the M-type receptor or other as yet unknown protein targets.

A prominent secondary four-bond hydrogen/deuterium isotope effect was observed from proton NMR at the active site histidine imidazole ring of bovine pancreatic sPLA(2) in the presence of a phosphonate transition state analogue. The cross-modulation of H(epsilon2)/H48 and H(delta1)/H48 resonances was confirmed by line shape simulation that follows the McConnell equation with fractionation factors incorporated to account for the change in the signal magnitude as well as the resonance line shape at various H(2)O/D(2)O solvent mixtures. While the downfield shift of each individual proton upon deuteration on the opposite site can be attributed to the proton-relay system of the H48-D99 catalytic dyad in sPLA(2), the observation that H(delta1)/H48 induces a 3-fold larger H/D secondary isotope effect ( approximately 0.15 ppm) on H(epsilon2)/H48 than vice versa ( approximately 0.05 ppm) is interpreted as additional spectroscopic evidence for the previously proposed short-strong hydrogen bond formed between the donor N(delta1)/H48 and a nonbridging phosphonate oxygen atom of the transition state analogue. These results provide additional details for the catalytic mechanism of sPLA(2) and demonstrate that the intrinsic H/D secondary isotope effect is a useful tool to probe hydrogen bond strength.

The secreted phospholipase A(2) from bee venom (bvPLA(2)) contains a membrane binding surface composed mainly of hydrophobic residues and two basic residues that come in close contact with the membrane. Previous studies have shown that the mutant in which these two basic residues (K14 and R23) as well as three other nearby basic residues were collectively changed to glutamate (charge reversal), like wild-type enzyme, binds with high affinity to anionic phospholipid vesicles. In the present study, we have measured the equilibrium constants for the interaction of wild-type bvPLA(2), the charge-reversal mutant (bvPLA(2)-E5), and the mutant in which the five basic residues were changed to neutral glutamine (bvPLA(2)-Q5) with phosphatidylcholine (PC) vesicles containing various amounts of the anionic phosphatidylserine (PS). Remarkably, bvPLA(2)-E5 with an anionic membrane binding surface binds more tightly to vesicles as the mole percent of PS is increased. Computational studies predict that this is due to a significant upward shift in the pK(a) of E14 (and to some extent E23) when the enzyme binds to PC/PS vesicles such that the carboxylate of the glutamate side chain near the membrane surface undergoes protonation. The experimental pH dependence of vesicle binding supports this prediction. bvPLA(2)-E5 binds more weakly to PS/PC vesicles than does wild-type enzyme due to electrostatic protein-vesicle repulsion coupled with the similar energetics of desolvation of basic residues and glutamates that accompanies enzyme-vesicle contact. Studies with bvPLA(2)-Q5 show that only a small fraction of the total bvPLA(2) interfacial binding energy ( approximately 10%) is due to electrostatics.

Platelet-activating factor acetylhydrolases (PAF-AHs) are a group of enzymes that hydrolyze the sn-2 acetyl ester of PAF (phospholipase A(2) activity) but not phospholipids with two long fatty acyl groups. Our previous studies showed that membrane-bound human plasma PAF-AH (pPAF-AH) accesses its substrate only from the aqueous phase, which raises the possibility that this enzyme can hydrolyze a variety of lipid esters that are partially soluble in the aqueous phase. Here we show that pPAF-AH has broad substrate specificity in that it hydrolyzes short-chain diacylglycerols, triacylglycerols, and acetylated alkanols, and displays phospholipase A(1) activity. On the basis of all of the substrate specificity results, it appears that the minimal structural requirement for a good pPAF-AH substrate is the portion of a glyceride derivative that includes an sn-2 ester and a reasonably hydrophobic chain in the position occupied by the sn-1 chain. In vivo, pPAF-AH is bound to high and low density lipoproteins, and we show that the apparent maximal velocity for this enzyme is not influenced by lipoprotein binding and that the enzyme hydrolyzes tributyroylglycerol as well as the recombinant pPAF-AH does. Broad substrate specificity is also observed for the structurally homologous PAF-AH which occurs intracellularly [PAF-AH(II)] as well as for the PAF-AH from the lower eukaryote Physarum polycephalum although pPAF-AH and PAF-AH(II) tolerate the removal of the sn-3 headgroup better than the PAF-AH from P. polycephalum does. In contrast, the intracellular PAF-AH found in mammalian brain [PAF-AH(Ib) alpha 1/alpha 1 and alpha 2/alpha 2 homodimers] is more selectively operative on compounds with a short acetyl chain although this enzyme also displays significant phospholipase A(1) activity.