Search

Phosphatidylinositol-specific phospholipase C (PI-PLC) enzymes simultaneously interact with the substrate, PI, and with nonsubstrate lipids such as phosphatidylcholine (PC). For Bacillus thuringiensis PI-PLC these interactions are synergistic with maximal catalytic activity observed at low to moderate mole fractions of PC (X(PC)) and maximal binding occurring at low mole fractions of anionic lipids. It has been proposed that residues in alpha-helix B help to modulate membrane binding and that dimerization on the membrane surface both increases affinity for PC and activates PI-PLC, yielding the observed PI/PC synergy. Vesicle binding and activity measurements using a variety of PI-PLC mutants support many aspects of this model and reveal that while single mutations can disrupt anionic lipid binding and the anionic lipid/PC synergy, the residues important for PC binding are less localized. Interestingly, at high X(PC) mutations can both decrease membrane affinity and increase activity, supporting a model where reductions in wild-type activity at X(PC) > 0.6 result from both dilution of the substrate and tight membrane binding of PI-PLC, limiting enzyme hopping or scooting to the next substrate molecule. These results provide a direct analysis of vesicle binding and catalytic activity and shed light on how occupation of the activator site enhances enzymatic activity.

Phospholipase Cbeta (PLCbeta) enzymes are activated by Galpha q and Gbetagamma subunits and catalyze the hydrolysis of the minor membrane lipid phosphatidylinositol 4,5-bisphosphate [PI(4,5)P2]. Activation of PLCbeta2 by Gbetagamma subunits has been shown to be conferred through its N-terminal pleckstrin homology (PH) domain, although the underlying mechanism is unclear. Also unclear are observations that the extent of Gbetagamma activation differs on different membrane surfaces. In this study, we have identified a unique region of the PH domain of the PLCbeta2 domain (residues 71-88) which, when added to the enzyme as a peptide, causes enzyme activation similar to that with Gbetagamma subunits. This PH domain segment interacts strongly with membranes composed of lipid mixtures but not those containing lipids with electrically neutral zwitterionic headgroups. Also, addition of this segment perturbs interaction of the catalytic domain, but not the PH domain, with membrane surfaces. We monitored the orientation of the PH and catalytic domains of PLC by intermolecular fluorescence resonance energy transfer (FRET) using the Gbetagamma activatable mutant, PLCbeta2/delta1(C193S). We find an increase in the level of FRET with binding to membranes with mixed lipids but not to those containing only lipids with electrically neutral headgroups. These results suggest that enzymatic activation can be conferred through optimal association of the PHbeta71-88 region to specific membrane surfaces. These studies allow us to understand the basis of variations of Gbetagamma activation on different membrane surfaces.

We report the combined effects of phospholipase C (PLC), a pronucleating factor, and apolipoprotein A-I (apo A-I), an antinucleating factor, in solutions of model bile. Results indicate that apo A-I inhibits cholesterol nucleation from unilamellar lecithin vesicles by two mechanisms. Initially, inhibition is achieved by apo A-I shielding of hydrophobic diacylglycerol (DAG) moieties so as to prevent vesicle aggregation. Protection via shielding is temporary. It is lost when the DAG/apo A-I molar ratio exceeds a critical value. Subsequently, apo A-I forms small ( approximately 5-15 nm) complexes with lecithin and cholesterol that coexist with lipid-stabilized (400-800 nm) DAG oil droplets. This microstructural transition from vesicles to complexes avoids nucleation of cholesterol crystals and is a newly discovered mechanism by which apo A-I serves as an antinucleating agent in bile. The critical value at which a microstructural transition occurs depends on binding of apo A-I and so varies with the cholesterol mole fraction of vesicles. Aggregation of small, unilamellar, egg lecithin vesicles (SUVs) with varying cholesterol composition (0-60 mol %) was monitored for a range of apo A-I concentrations (2 to 89 microg/mL). Suppression of aggregation persists so long as the DAG-to-bound-apo A-I molar ratio is less than 100. A fluorescence assay involving dansylated lecithin shows that the suppression is an indirect effect of apo A-I rather than a direct inhibition of PLC enzyme activity. The DAG-to-total apo A-I molar ratio at which suppression is lost increases with cholesterol because of differences in apo A-I binding. Above this value, a microstructural transition to DAG droplets and lecithin/cholesterol A-I complexes occurs, as evidenced by sudden increases in turbidity and size and enhancement of Forster resonance energy transfer; structures are confirmed by cryo TEM.

The neural cell adhesion molecule (NCAM) is implicated in important functions during development and maintenance of the nervous system. Two of the three major isoforms, NCAM 140 and NCAM 180, are transmembrane glycoproteins with large cytoplasmic domains of different length. The purpose of this study was to identify novel intracellular binding partners of NCAM 140 and NCAM 180. We expressed both cytoplasmic domains, as well as cytoplasmic fragments of NCAM, as fusion proteins in Escherichia coli and used them for ligand affinity chromatography or glutathione S-transferase (GST) pull-down assays. By peptide mass fingerprinting Western blot analysis, or both, we identified PLCgamma, LANP, syndapin, PP1, and PP2A as binding partners for both NCAM 140 and NCAM 180, whereas TOAD-64 was identified as a NCAM 180-specific interacting protein. Furthermore, we were able to show that binding of these novel binding proteins, as well as the previously described interaction partners ROK alpha (rho A binding kinase alpha) and alpha- and beta-tubulin, bind to specific cytosolic sequences of NCAM. For this purpose, we performed GST pull-down experiments using cytosolic fragments of NCAM as GST-fusion proteins and cytosolic- or cytoskeleton-enriched protein fractions of rat brain.

Phosphatidylinositol-specific phospholipase C (PLC) enzymes catalyze the hydrolysis of phosphatidylinositol 4,5 bisphosphate in a two step reaction that involves a cyclic intermediate. The PLCbetafamily are activated by both the alpha and betagamma subunits of heterotrimeric G proteins. To determine which catalytic step is affected by Gbetagamma subunits, we compared the change in PLCbeta(2) activity catalysis toward monomeric short-chain phosphatidylinositol (PI) substrates and monomeric water-soluble cyclic inositol phosphates as well as long-chain PI in bilayer and micellar interfaces in the absence and presence of Gbetagammasubunits. Unlike other PLC enzymes, no cyclic products were detected for either wild-type PLCbeta(2) or a chimeric protein composed of the PH domain of PLCbeta(2) and the catalytic domain of PLCdelta(1). Using cIP as a substrate to examine the second step of the reaction, we found that the presence of Gbetagamma subunits stimulated this step by a higher level than that for the overall reaction (k(cat) 1.5-fold (cIP) as opposed to 1.20-fold for soluble diC(4)PI). Detergents above their CMC can generate the same kinetic activation of PLCbeta(2) as Gbetagamma, suggesting that hydrophobic compounds stabilize the activated state of the enzyme. The most pronounced effect of Gbetagamma is that it relieves competitive product inhibition. Taken together, our results show that activation of PLCbeta(2) occurs through enhancement in the catalytic rate of hydrolysis of the cyclic intermediate and increased product release, and that hydrophobic interactions play a key role.

Alpha-synuclein plays a key role in the pathogenesis of many neurodegenerative diseases. To date, its cellular role has yet to be determined, although it has been proposed to be connected to calcium and G protein-mediated dopamine signaling. Alpha-synuclein is known to bind strongly to model membrane surfaces where it may interact with other membrane-associated proteins. Here, we find that the membrane association of alpha-synuclein is enhanced by the presence of phosphatidylinositol 4,5-bisphosphate [PI(4,5)P(2)] and Ca(2+). We also find that alpha-synuclein interacts with high affinity with the G protein-regulated enzyme phospholipase Cbeta(2) (PLCbeta(2)), which catalyzes the hydrolysis of PI(4,5)P(2). Binding of alpha-synuclein to PLCbeta(2) reduces its catalytic activity by 50%, but causes its level of activation by Gbetagamma subunits to increase from 4- to 24-fold. This effect is greatly reduced for A53T alpha-synuclein, which is a mutant associated with familial Parkinson's disease. PI(4,5)P(2) hydrolysis by PLCbeta(2) results in an increase in the intracellular Ca(2+) concentration, and we find that in cultured cells the presence of alpha-synuclein results in a 6-fold enhancement in the release of Ca(2+) from intracellular stores in response to agents that release Gbetagamma subunits relative to controls. Alpha-synuclein also enhances the increase in the level of inositol phosphates seen upon G protein stimulation, suggesting that it also may interact with PLCbeta(2) in cells. Given that Ca(2+) and dopamine regulation are mediated through PLCbeta and G protein signals, our results suggest that alpha-synuclein may play a role in inositol phospholipid signaling.

The signal transduction protein phospholipase C-gamma1 (PLC-gamma1) is activated when its C-terminal SH2 domain (PLCC) binds the phosphorylated Tyr-1021 site (pTyr-1021) in the beta-platelet-derived growth factor receptor (PDGFR). To better understand the contributions that dynamics make to binding, we have used NMR relaxation experiments to investigate the motional properties of backbone amide and side chain methyl groups in a peptide derived from the pTyr-1021 site of PDGFR, both free and in complex with the PLCC SH2 domain. The free peptide has relaxation properties that are typical for a small, unstructured polymer, while the backbone of the bound peptide is least flexible for residues in the central portion of the binding site with the amplitude of pico- to nanosecond time scale motions increasing toward the C-terminus of the peptide. The increase in large amplitude motion toward the end of the pY1021 peptide is consistent with the bound peptide existing as an ensemble of states with C-terminal residues having the broadest distribution of backbone conformations, while residues in the central binding site are the most restricted. Deuterium spin relaxation experiments establish that the protein-peptide interface is highly dynamic, and this mobility may play an important role in modulating the affinity of the interaction.

The generation of multiprotein complexes at receptors and adapter proteins is crucial for the activation of intracellular signaling pathways. In this study, we used multiple biochemical and biophysical methods to examine the binding properties of several SH2 and SH3 domain-containing signaling proteins as they interact with the adapter protein linker for activation of T-cells (LAT) to form multiprotein complexes. We observed that the binding specificity of these proteins for various LAT tyrosines appears to be constrained both by the affinity of binding and by cooperative protein-protein interactions. These studies provide quantitative information on how different binding parameters can determine in vivo binding site specificity observed for multiprotein signaling complexes.

Listeria monocytogenes is a bacterium responsible for localized and generalized infections in humans and animals. It has the ability to spread from the cytoplasm of an infected cell to neighboring cells without becoming exposed to the extracellular space. The bacterium secretes a phospholipase C (PLC(LM)) that is active on glycerophospholipids, e.g., phosphatidylcholine, and on sphingomyelin; thus, PLC(LM) should be described more appropriately as a phospholipase C/sphingomyelinase. We have obtained PLC(LM) free from a frequent contaminant, listeriolysin O, using an improved purification procedure. PLC(LM) has been assayed on large unilamellar liposomes of defined lipid composition. The enzyme is activated by K(+) and Mg(2+), and readily degrades phospholipids in bilayer form, in the absence of detergents. Enzyme activity is accompanied by important changes in the structure of the phospholipid vesicles, namely, vesicle aggregation, intervesicular mixing of lipids, and mixing of aqueous contents, with very low leakage of vesicular contents. The data are interpreted as indicative of PLC(LM)-induced vesicle fusion. This is confirmed by the demonstration of intervesicular mixing of inner monolayer lipids, using a novel procedure. The observation of PLC(LM)-induced membrane fusion suggests a mechanism for the cell-to-cell propagation of the bacterium, which requires disruption of a double-membrane vacuole.

An assay detecting and quantifying cholesterol nucleation from low-density lipoproteins has been established. Förster resonance energy transfer between dehydroergosterol and dansylated lecithin becomes significantly alleviated as a consequence of conucleation of dehydroergosterol and cholesterol. The assay, in combination with dynamic light scattering, absorbance spectroscopy, and fluorescence microscopy, can be used to study aggregation and nucleation in model blood systems. Human plasma LDL was labeled with dehydroergosterol and dansylated lecithin by incubation with donor multilamellar liposomes and isolated by centrifugation. Exposure of labeled LDL (0.5 mg/mL of total lipids) to sphingomyelinase (0.0-0.2 unit/mL) led to modest particle aggregation but produced no changes in energy transfer and no crystallization. However, addition of sphingomyelinase produced significant particle aggregation, nucleation, and crystallization, in a dose-dependent fashion, in samples that were previously treated with the enzyme, cholesterol esterase (0.2 unit/mL). The combination of cholesterol esterase and sphingomyelinase led to a significant alleviation of energy transfer, which preceded by 24 h the appearance of fluorescent, microscopic sterol crystals. These results point to a synergistic effect between cholesterol esterase and sphingomyelinase, suggesting that mere aggregation of LDL is insufficient to promote nucleation, and crystal formation likely proceeds in the intracellular space after LDL uptake by macrophages.

Eukaryotic phosphatidylinositol-specific phospholipase Cs (PI-PLCs) utilize calcium as a cofactor during catalysis, whereas prokaryotic PI-PLCs use a spatially conserved guanidinium group from Arg69. In this study, we aimed to construct a metal-dependent mutant of a bacterial PI-PLC and characterize the catalytic role of the metal ion, seeking an enhanced understanding of the functional differences between these two positively charged moieties. The following results indicate that a bona fide catalytic metal binding site was created by the single arginine-to-aspartate mutation at position 69: (1) The R69D mutant was activated by Ca(2+), and the activation was specific for R69D, not for other mutants at this position. (2) Titration of R69D with Ca(2+), monitored by (15)N-(1)H HSQC (heteronuclear single quantum coherence) NMR, showed that addition of Ca(2+) to R69D restores the environment of the catalytic site analogous to that attained by the WT enzyme. (3) Upon Ca(2+) activation, the thio effect of the S(P)-isomer of the phosphorothioate analogue (k(O)/k(Sp) = 4.4 x 10(5)) approached a value similar to that of the WT enzyme, suggesting a structural and functional similarity between the two positively charged moieties (Arg69 and Asp69-Ca(2+)). The R(P)-thio effect (k(O)/k(Rp) = 9.4) is smaller than that of the WT enzyme by a factor of 5. Consequently, R69D-Ca(2+) displays higher stereoselectivity (k(Rp)/k(Sp) = 47,000) than WT (k(Rp)/k(Sp) = 7600). (4) Results from additional mutagenesis analyses suggest that the Ca(2+) binding site is comprised of side chains from Asp33, Asp67, Asp69, and Glu117. Our studies provide new insight into the mechanism of metal-dependent and metal-independent PI-PLCs.

PLC(Bc) is a 28.5 kDa monomeric enzyme that catalyzes the hydrolysis of the phosphodiester bond of phosphatidylcholine, phosphatidylethanolamine, and phosphatidylserine to provide a diacylglycerol and the corresponding phosphorylated headgroup. Because single replacements of Glu4, Tyr56, and Phe66 in the headgroup binding pocket led to changes in substrate specificity [Martin et al. (2000) Biochemistry 39, 3410-3415], a combinatorial library of approximately 6000 maltose binding protein-PLC(Bc) fusion protein mutants containing random permutations of these three residues was generated to identify PLC(Bc) mutants with altered specificity profiles and high catalytic activities. Members of this library were screened for hydrolytic activity toward the water soluble substrates C6PC, C6PE, and C6PS using a novel protocol that was conducted in a 96-well format and featured the in situ cleavage of the fusion protein to release the mutant PLC(Bc)s. Ten mutant enzymes that exhibited significant preferences toward C6PE or C6PS were selected and analyzed by steady-state kinetics to determine their specificity constants, k(cat)/K(M). The C6PS selective clones E4G, E4Q/Y56T/F66Y, and E4K/Y56V exhibited higher specificity constants toward C6PS than wt, whereas Y56T, F66Y, and Y56T/F66Y were C6PE selective and had comparable or higher specificity constants than wt for C6PE. The corresponding wt residues were singly reinserted back into the E4Q/Y56T/F66Y and E4K/Y56V mutants via site-directed mutagenesis, and the E4Q/F66Y mutant thus obtained exhibited a 10-fold higher specificity constant toward C6PS than wt, a value significantly higher than other PLC(Bc) mutants. On the basis of available data, an aromatic residue at position 66 appears important for significant catalytic activity toward all three substrates, especially C6PC and C6PE. The charge of residue 4 also appears to be a determinant of enzyme specificity as a negatively charged residue at this position endows the enzyme with C6PC and C6PE preference, whereas a polar neutral or positively charged residue results in C6PS selectivity. Replacing Tyr56 with Val, Ala, Thr, or Ser greatly reduces activity toward C6PC. Thus, the substrate specificity of PLC(Bc) can be modulated by varying three of the amino acid residues that constitute the headgroup binding pocket, and it is now apparent that this enzyme is not evolutionarily optimized to hydrolyze phospholipids with ethanolamine or serine headgroups.

Phosphatidic acid (PA) stimulates phospholipase C-beta(1) (PLC-beta(1)) activity and promotes G protein stimulation of PLC-beta(1) activity. The isoform dependence for PA regulation of PLC-beta activity as well as the role of PA in modulating regulation of PLC-beta activity by protein kinase C (PKC) and G protein subunits was determined. As compared to PLC-beta(1), the phospholipase C-beta(3) (PLC-beta(3)) isoform was less sensitive to PA, requiring greater than 15 mol % PA for stimulation. PLC-beta(3) bound weakly to PA. PKC had little effect on PA stimulation of PLC-beta(3) activity. PKC, however, inhibited PA stimulation of PLC-beta(1) activity through a mechanism dependent on the mol % PA. Stimulation by 7.5 mol % PA was completely inhibited by PKC. Increasing the PA and Ca(2+) concentration attenuated PKC inhibition. The binding of PLC-beta(1) to PA containing phospholipid vesicles was also reduced by PKC, in a manner dependent on the mol % PA. PA increased the stimulation of PLC-beta(1) activity by G alpha q but had little effect on the stimulation by beta gamma subunits. These results demonstrate that PA stimulation of PLC-beta activity is tightly regulated, suggesting the existence of a distinct PA binding region in PLC-beta(1). PA may be an important component of a receptor mediated signaling mechanism that determines PLC-beta(1) activation.

Clostridium perfringens alpha-toxin is a 370-residue, zinc-dependent, phospholipase C that is the key virulence determinant in gas gangrene. It is also implicated in the pathogenesis of sudden death syndrome in young animals and necrotic enteritis in chickens. Previously characterized alpha-toxins from different strains of C. perfringens are almost identical in sequence and biochemical properties. We describe the cloning, nucleotide sequencing, expression, characterization, and crystal structure of alpha-toxin from an avian strain, SWan C. perfringens (SWCP), which has a large degree of sequence variation and altered substrate specificity compared to these strains. The structure of alpha-toxin from strain CER89L43 has been previously reported in open (active site accessible to substrate) and closed (active site obscured by loop movements) conformations. The SWCP structure is in an open-form conformation, with three zinc ions in the active site. This is the first example of an open form of alpha-toxin crystallizing without the addition of divalent cations to the crystallization buffer, indicating that the protein can retain three zinc ions bound in the active site. The topology of the calcium binding site formed by residues 269, 271, 336, and 337, which is essential for membrane binding, is significantly altered in comparison with both the open and closed alpha-toxin structures. We are able to relate these structural changes to the different substrate specificity and membrane binding properties of this divergent alpha-toxin. This will provide essential information when developing an effective vaccine that will protect against C. perfringens infection in a wide range of domestic livestock.

Platelet endothelial cell adhesion molecule-1 (PECAM-1/CD31) is a 130K transmembrane glycoprotein that belongs to the immunoglobulin gene superfamily and is expressed on the surface of hematological or vascular cells, including platelets and endothelial cells. Although the importance of this adhesion molecule in various cell-cell interactions is established, its function in platelets remains ill-defined. In the process of clarifying the mechanism by which the lectin wheat germ agglutinin (WGA) activates platelets, we unexpectedly discovered that PECAM-1 is involved in signal transduction pathways elicited by this N-acetyl-D-glucosamine (NAGlu)-reactive lectin. WGA, which is a very potent platelet stimulator, elicited a rapid surge in Syk and phospholipase C (PLC)-gamma 2 tyrosine phosphorylation and the resultant intracellular Ca(2+) mobilization; collagen, as reported, induced these responses, but in a much slower and weaker manner. WGA strongly induced tyrosine phosphorylation of a 130-140K protein, which was confirmed to be PECAM-1 by immunoprecipitation and immunodepletion studies. WGA-induced PECAM-1 tyrosine phosphorylation occurred rapidly, strongly and in a manner independent of platelet aggregation or cell-cell contact; these characteristics of PECAM-1 phosphorylation were not mimicked at all by receptor-mediated platelet agonists. In addition, WGA was found to associate with PECAM-1 itself, and anti-PECAM-1 antibody, as well as NAGlu, specifically inhibited WGA-induced platelet aggregation. In PECAM-1 immunoprecipitates, Src family tyrosine kinases existed, and a kinase activity was detected, which increased upon WGA stimulation. Furthermore, the Src family kinase inhibitor PP2 inhibited WGA-induced platelet aggregation, Ca(2+) mobilization, and PLC-gamma 2 tyrosine phosphorylation. Finally, WGA induced PECAM-1 tyrosine phosphorylation and cytoskeletal reorganization in vascular endothelial cells. Our results suggest that (i) PECAM-1 is involved in WGA-induced platelet activation, (ii) PECAM-1 clustering by WGA activates unique and strong platelet signaling pathways, leading to a rapid PLC activation via Src family kinases, and (iii) WGA is a useful tool for elucidating PECAM-1-mediated signaling with wide implications not confined to platelets.

Phosphatidylinositol-specific phospholipase Cs (PI-PLCs, EC 3.1.4.10) are ubiquitous enzymes that cleave phosphatidylinositol or phosphorylated derivatives, generating second messengers in eukaryotic cells. A catalytic diad at the active site of Bacillus cereus PI-PLC composed of aspartate-274 and histidine-32 was postulated from the crystal structure to form a catalytic triad with the 2-OH group of the substrate [Heinz, D. W., et al. (1995) EMBO J. 14, 3855-3863]. This catalytic diad has been observed directly by proton NMR. The single low-field line in the 1H NMR spectrum is assigned by site-directed mutagenesis: The peak is present in the wild type but absent in the mutants H32A and D274A, and arises from the histidine Hdelta1 forming the Asp274-His32 hydrogen bond. This hydrogen is solvent-accessible, and exchanges slowly with H2O on the NMR time scale. The position of the low-field peak shifts from 16.3 to 13.8 ppm as the pH is varied from 4 to 9, reflecting a pKa of 8.0 at 6 degrees C, which is identified with the pKa of His32. The Hdelta1 signal is modulated by rapid exchange of the Hepsilon2 with the solvent. Estimates of the exchange rate as a function of pH and protection factors are derived from a line shape analysis. The NMR behavior is remarkably similar to that of the serine proteases. The postulated function of the Asp274-His32 diad is to hydrogen-bond with the 2-OH of phosphatidylinositol (PI) substrate to form a catalytic triad analogous to Asp-His-Ser of serine proteases. This is an example of substrate-assisted catalysis where the substrate provides the catalytic nucleophile of the triad. This hydrogen bond becomes shorter as the imidazole is protonated, suggesting it is stronger in the transition state, contributing further to the catalytic efficiency. The hydrogen bond fits the NMR criteria for a short, strong hydrogen bond, i.e., a highly deshielded proton resonance, bond length of 2.64 +/- 0.04 A at pH 6 measured by NMR, a D/H fractionation factor significantly lower than 1.0, and a protection factor > or = 100.

Phosphatidylinositol-specific phospholipase C (PI-PLC) has been proposed previously to employ a catalytic mechanism highly reminiscent of that of ribonuclease A (RNase A). Both catalytic sites are comprised of two histidine side chains acting as a general base-general acid pair and a phosphate-activating residue: an arginine in the case of PI-PLC and a lysine in RNase A. Despite these structural similarities, the PI-PLC reaction is slowed 10(5)-fold upon substitution of one of the phosphate nonbridging oxygen atoms with sulfur, whereas a much smaller effect is observed in the analogous RNase A reaction. Here, we report a systematic study of this property in PI-PLC, conducted by means of site-directed chemical modification of a cysteine residue replacing the arginine at position 69. The results show that mutant enzymes featuring bidentate side chains at this position display significantly higher activity, higher thio effects, and greater stereoselectivity than do those with monodentate side chains. The results suggest that the bidentate nature of Arg69 is the origin of the large thio effects and stereoselectivity in PI-PLC. We propose that in addition to binding the phosphate, the function of arginine 69 is to bring the phosphate group and the 2-OH group of inositol into proximity and to induce proper alignment for nucleophilic attack, and possibly to lower the pK(a) of the 2-OH. The results presented here could be important to mechanisms of phosphoryl transfer enzymes in general, suggesting that a major part of thio effects observed in enzymatic phosphoryl transfer reactions can originate from factors other than direct interaction between a side chain and a phosphate group, and caution the use of the absolute magnitude of the thio effect as an indicator of the strength of such interactions.

Phosphatidylinositol-specific phospholipase C (PI-PLC) catalyzes the cleavage of the P-O bond in phosphatidylinositol via intramolecular nucleophilic attack of the 2-hydroxyl group of inositol on the phosphorus atom. Our earlier stereochemical and site-directed mutagenesis studies indicated that this reaction proceeds by a mechanism similar to that of RNase A, and that the catalytic site of PI-PLC consists of three major components analogous to those observed in RNase A, the His32 general base, the His82 general acid, and Arg69 acting as a phosphate-activating residue. In addition, His32 is associated with Asp274 in forming a catalytic triad with inositol 2-hydroxyl, and His82 is associated with Asp33 in forming a catalytic diad. The focus of this work is to provide a global view of the mechanism, assess cooperation between various catalytic residues, and determine the origin of enzyme activation by the hydrophobic leaving group. To this end, we have investigated kinetic properties of Arg69, Asp33, and His82 mutants with phosphorothioate substrate analogues which feature leaving groups of varying hydrophobicity and pK(a). Our results indicate that interaction of the nonbridging pro-S oxygen atom of the phosphate group with Arg69 is strongly affected by Asp33, and to a smaller extent by His82. This result in conjunction with those obtained earlier can be rationalized in terms of a novel, dual-function triad comprised of Arg69, Asp33, and His82 residues. The function of this triad is to both activate the phosphate group toward the nucleophilic attack and to protonate the leaving group. In addition, Asp33 and His82 mutants displayed much smaller degrees of activation by the fatty acid-containing leaving group as compared to the wild-type (WT) enzyme, and the level of activation was significantly reduced for substrates featuring the leaving group with low pK(a) values. These results strongly suggest that the assembly of the above three residues into the fully catalytically competent triad is controlled by the hydrophobic interactions of the enzyme with the substrate leaving group.

Apolipophorin III (apoLp-III) from the greater wax moth Galleria mellonella is an exchangeable insect apolipoprotein that consists of five amphipathic alpha-helices, sharing high sequence identity with apoLp-III from the sphinx moth Manduca sexta whose structure is available. To define the minimal requirement for apoLp-III structural stability and function, a C-terminal truncated apoLp-III encompassing residues 1-91 of this 163 amino acid protein was designed. Far-UV circular dichroism spectroscopy revealed apoLp-III(1-91) has 50% alpha-helix secondary structure content in buffer (wild-type apoLp-III 86%), increasing to essentially 100% upon interactions with dimyristoylphosphatidylcholine (DMPC). Guanidine hydrochloride denaturation studies revealed similar stability properties for wild-type apoLp-III and apoLp-III(1-91). Resistance to denaturation for both proteins increased substantially upon association with phospholipid. In the absence of lipid, wild-type apoLp-III was monomeric whereas apoLp-III(1-91) partly formed dimers and trimers. Discoidal apoLp-III(1-91)-DMPC complexes were smaller in diameter (13.5 nm) compared to wild-type apoLp-III (17.7 nm), and more molecules of apoLp-III(1-91) associated with the complexes. Lipid interaction revealed that apoLp-III(1-91) binds to modified spherical lipoprotein surfaces and efficiently transforms phospholipid vesicles into discoidal complexes. Thus, the first three helices of G. mellonella apoLp-III contain the basic features required for maintenance of the structural integrity of the entire protein.

Cell signaling proteins may form functional complexes that are capable of rapid signal turnover. These contacts may be stabilized by either scaffolding proteins or multiple interactions between members of the complex. In this study, we have determined the affinities between a regulator of G protein signaling protein, RGS4, and three members of the G protein-phospholipase Cbeta (PLC-beta) signaling cascade which may allow for rapid deactivation of intracellular Ca(2+) release and activation of protein kinase C. Specifically, using fluorescence methods, we have determined the interaction energies between the RGS4, PLC-beta, G-betagamma, and both deactivated (GDP-bound) and activated (GTPgammaS-bound) Galpha(q). We find that RGS4 not only binds to activated Galpha(q), as predicted, but also to Gbetagamma and PLCbeta(1). These interactions occur through protein-protein contacts since the intrinsic membrane affinity of RGS4 was found to be very weak in the absence of the protein partner PLCbeta(1) or a lipid regulator, phosphatidylinositol-3,4,5 trisphosphate. Ternary complexes between Galpha(q), Gbetagamma and phospholipase Cbeta(1) will form, but only at relatively high protein concentrations. We propose that these interactions allow RGS4 to remain anchored to the signaling complex even in the quiescent state and allow rapid transfer to activated Galpha(q) to shut down the signal. Comparison of the relative affinities between these interacting proteins will ultimately allow us to determine whether certain complexes can form and where signals will be directed.

Large unilamellar vesicles containing phosphatidylinositol (PI), neutral phospholipids, and cholesterol are induced to fuse by the catalytic activity of phosphatidylinositol-specific phospholipase C (PI-PLC). PI cleavage by PI-PLC is followed by vesicle aggregation, intervesicular lipid mixing, and mixing of vesicular aqueous contents. An average of 2-3 vesicles merge into a large one in the fusion process. Vesicle fusion is accompanied by leakage of vesicular contents. A novel method has been developed to monitor mixing of lipids located in the inner monolayers of the vesicles involved in fusion. Using this method, the mixing of inner monolayer lipids and that of vesicular aqueous contents are seen to occur simultaneously, thus giving rise to the fusion pore. Kinetic studies show, for fusing vesicles, second-order dependence of lipid mixing on diacylglycerol concentration in the bilayer. Varying proportions of PI in the liposomal formulation lead to different physical effects of PI-PLC. Specifically, 30-40 mol % PI lead to vesicle fusion, while with 5-10 mol % PI only hemifusion is detected, i.e., mixing of outer monolayer lipids without mixing of aqueous contents. However, when diacylglycerol is included in the bilayers containing 5 mol % PI, PI-PLC activity leads to complete fusion.

A recent report that microinjection of the SH3 domain of PLC-gamma1 could induce DNA synthesis raised the functional importance of the SH3 domain of PLC-gamma1 in mitogenic signaling. In this report, we provide evidence that SOS1, a p21Ras-specific guanine nucleotide exchange factor, directly binds to the SH3 domain of PLC-gamma1, and that the SH3 domain of PLC-gamma1 is involved in SOS1-mediated p21Ras activation. SOS1 was coprecipitated with the GST-fused SH3 domain of PLC-gamma1 in vitro. The interaction between SOS1 and the PLC-gamma1 SH3 domain is mediated by direct physical interaction. The carboxyl-terminal proline-rich domain of SOS1 is involved in the interaction with the PLC-gamma1 SH3 domain. Moreover, PLC-gamma1 could be co-immunoprecipitated with SOS1 antibody in cell lysates. From transient expression studies, we could demonstrate that the SH3 domain of PLC-gamma1 is necessary for the association with SOS1 in vivo. Intriguingly, overexpression of the SH3 domain of PLC-gamma1, lipase-inactive PLC-gamma1, or wild-type PLC-gamma1 elevated p21Ras activity and ERK activity when compared with vector transfected cells. The PLC-gamma1 mutant lacking the SH3 domain could not activate p21Ras. p21Ras activities in cell lines overexpressing either PLC-gamma1 or the SH2-SH2-SH3 domain of PLC-gamma1 were elevated about 2-fold compared to vector transfected cells. This study is the first to demonstrate that the PLC-gamma1 SH3 domain enhances p21Ras activity, and that the SH3 domain of PLC-gamma1 may be involved in the SOS1-mediated signaling pathway.

The phosphatidylcholine-preferring phospholipase C from Bacillus cereus (PLC(Bc)) is a 28.5 kDa enzyme with three zinc ions in its active site. The roles that a number of amino acid residues play as zinc ligands and in binding and catalysis have been elucidated. Recent mechanistic studies indicate that the rate of the reaction is limited by a proton-transfer step during chemical hydrolysis and not substrate binding or product release. An X-ray structure of PLC(Bc) complexed with a phosphonate inhibitor related to phosphatidylcholine revealed that the three amino acid residues Glu4, Tyr56, and Phe66 comprise the choline binding pocket. However, because the contributions that these three residues make to substrate recognition and specificity were unknown, a series of site-specific mutants for Glu4, Tyr56, and Phe66 were constructed by PCR mutagenesis. On the basis of a comparison of their respective CD spectra and melting temperatures, it appears that the mutants adopt folded structures in solution that are virtually identical to that of wild-type PLC(Bc). The kinetic parameters k(cat) and K(m) for the hydrolysis of the three soluble substrates 1, 2-dihexanoyl-sn-glycero-3-phosphocholine (C6PC), 1, 2-dihexanoyl-sn-glycero-3-phosphoethanolamine (C6PE), and 1, 2-dihexanoyl-sn-glycero-3-phospho-L-serine (C6PS) at concentrations below their corresponding critical micelle concentration (cmc) values were determined for each mutant. Replacement of Phe66 with a nonaromatic residue dramatically decreased k(cat) (approximately 200-fold) and reduced PLC(Bc) activity toward C6PC, C6PE, and C6PS, whereas changes to Glu4 and Tyr56 typically led to much more modest losses in catalytic efficiencies. Mutations of Glu4 had relatively little effect upon k(cat) and K(m) for C6PS, but they significantly influenced K(m) for C6PC and C6PE. Replacing Tyr56 with nonaromatic residues also affects catalytic efficiency, albeit to a much lesser degree than the corresponding changes at position 66. However, the presence of an aromatic residue at position 56 seems to confer some substrate selectivity for C6PC and C6PE, which bear a positive charge on the headgroup, relative to C6PS, which has no net charge on the headgroup; this increase in specificity arises largely from a reduced k(cat) for C6PS.

The phosphatidylcholine-preferring phospholipase C from Bacillus cereus (PLCBc) is a 28.5 kDa enzyme with three zinc ions in its active site. Although much is known about the roles that various PLCBc active site amino acids play in binding and catalysis, there is little information about the rate-determining step of the PLCBc-catalyzed hydrolysis of phospholipids and the catalytic cycle of the enzyme. To gain insight into these aspects of the hydrolysis, solvent viscosity variation experiments were conducted to determine whether an external step (substrate binding or product release) or an internal step (hydrolysis) is rate-limiting. The data indicate that the PLCBc-catalyzed reaction is unaffected by changes in solvent viscosity. This observation is inconsistent with the notion of substrate binding or product release being rate-determining and supports the hypothesis that a chemical step is rate-limiting. Furthermore, a deuterium isotope effect of 1.9 and a linear proton inventory plot indicate one proton is transferred in the rate-determining step. These data may be used to formulate a comprehensive catalytic cycle that is for the first time based on experimental evidence. In this mechanism, Asp55 of PLCBc activates an active site water molecule for attack on the phosphodiester bond, the hydrolysis of which is rate-limiting. The phosphorylcholine product is the first to leave the active site, followed by diacylglycerol.

Phosphatidylinositide-specific phospholipase C-betas play a key role in Ca2+ signaling and are specifically activated by the alphaq family of heterotrimeric G proteins and as well as betagamma subunits. We have determined the affinity between Gbetagamma subunits and GTPgammaS and GDP-liganded Galphaq subunits on membrane surfaces, and their respective affinities to PLC-beta1, -beta2 and -beta3 effectors by fluorescence spectroscopy. We find that activation of Galphaq by GTPgammaS decreases its affinity for Gbetagamma subunits at least 36-fold compared to the GDP-liganded form, but increases its affinity for PLC-betas at least 40-200-fold depending on the PLC-beta isoform. The affinity of Galphaq(GTPgammaS) is similar for PLC-beta1 and -beta3 and 10-fold stronger for PLC-beta2, which corresponds to the reported relationship between the concentration of Galphaq(GTPgammaS) and PLC-beta activation on lipid bilayers. We find that a large portion of the PLC-beta-Galphaq association energy lies within the 400 residue C-terminal region of PLC-beta1 since truncating this region reduces its Galphaq affinity. In contrast, the isolated N-terminal region does not interact with Galphaq. Gbetagamma subunits interact with all three PLC-beta isotypes, but only showed strong binding to PLC-beta2, and activation of the three PLC-betas by Gbetagamma subunits parallels this behavior. We also tested the possibility that both Galphaq and Gbetagamma can simultaneously bind PLC-beta2. Our data argue against simultaneous binding and show that Galphaq and Gbetagamma independently regulate this effector.

The activity of mammalian phosphoinositide-specific phospholipase C beta 2 (PLC-beta 2) is regulated by the alpha q family of G proteins and by beta gamma subunits. We measured the affinity between the laterally associating PLC-beta 2 and G beta gamma on membrane surfaces by fluorescence resonance energy transfer. Using a simple model, we translated this apparent affinity to a bulk or three-dimensional equilibrium constant (Kd) and obtained a value of 3.2 microM. We confirmed this Kd by separately measuring the on and off (kf and kr) rate constants. The kf was slower than a diffusion-limited value, suggesting that conformational changes occur when the two proteins interact. The off rate shows that the PLC-beta 2.G beta gamma complexes are long-lived ( approximately 123 s) and that activation of PLC-beta 2 by G beta gamma would be sustained without a deactivating factor. The addition of alpha i1(GDP) subunits failed to physically dissociate the complex as determined by fluorescence. However, enzyme activity studies performed under similar conditions show that the addition of G alpha i1(GDP) results in reversal of PLC-beta 2 activation by G beta gamma during the time of the assay (30 s). From these results, we propose that G alpha(GDP) subunits can bind to the PLC-beta 2.G beta gamma complex to allow for rapid deactivation without complex dissociation. In support of this model, we show by fluorescence that G alpha i1(GDP).G beta gamma.PLC-beta 2 can form.

The hydrolysis of phosphatidylinositol 4,5-bisphosphate (PIP2) by phosphoinositide-specific phospholipase C (PLC) is absolutely dependent on Ca2+. The PH domain truncated catalytic core of rat phospholipase C delta1 (PLC-delta1) has Ca2+ binding sites in its catalytic and C2 domains, and potential Ca2+ binding sites in two EF-hands. A catalytically inactive PLC-delta1 catalytic core bound with low affinity to PIP2-containing vesicles in the presence of Ca2+. A mutant PLC-delta1 has been engineered which lacks the C2 domain Ca2+ binding site and the surrounding loops known as the jaws. Isothermal calorimetric titration showed four Ca2+ ions bind to the wild-type PLC-delta1 catalytic core in solution but only one binds to the C2 domain jaws deletion mutant. The activity and Ca2+ dependence of wild-type and mutant phospholipase Cs were determined using substrate incorporated in detergent micelles and in large unilamellar vesicles. The activities of wild-type and mutant were identical to each other in both assay systems. Wild-type and the C2 jaws deletion mutant of PLC have Hill coefficients of 1.12-1.16 with respect to [Ca2+]. We conclude that a single Ca2+ bound to the catalytic domain is entirely responsible for the Ca2+ dependence of the basal activity of PLC-delta1.

The mechanism of phosphatidylinositol-specific phospholipase C (PI-PLC) has been suggested to resemble that of ribonuclease A. The goal of this work is to rigorously evaluate the mechanism of PI-PLC from Bacillus thuringiensis by examining the functional and structural roles of His-32 and His-82, along with the two nearby residues Asp-274 and Asp-33 (which form a hydrogen bond with His-32 and His-82, respectively), using site-directed mutagenesis. In all, twelve mutants were constructed, which, except D274E, showed little structural perturbation on the basis of 1D NMR and 2D NOESY analyses. The H32A, H32N, H32Q, H82A, H82N, H82Q, H82D, and D274A mutants showed a 10(4)-10(5)-fold decrease in specific activity toward phosphatidylinositol; the D274N, D33A, and D33N mutants retained 0. 1-1% activity, whereas the D274E mutant retained 13% activity. Steady-state kinetic analysis of mutants using (2R)-1, 2-dipalmitoyloxypropane-3-(thiophospho-1d-myo-inositol) (DPsPI) as a substrate generally agreed well with the specific activity toward phosphatidylinositol. The results suggest a mechanism in which His-32 functions as a general base to abstract the proton from 2-OH and facilitates the attack of the deprotonated 2-oxygen on the phosphorus atom. This general base function is augmented by the carboxylate group of Asp-274 which forms a diad with His-32. The H82A and D33A mutants showed an unusually high activity with substrates featuring low pKa leaving groups, such as DPsPI and p-nitrophenyl inositol phosphate (NPIPs). These results suggest that His-82 functions as the general acid with assistance from Asp-33, facilitating the departure of the leaving group by protonation of the glycerol O3 oxygen. The Bronsted coefficients obtained for the WT and the D33N mutant indicate a high degree of proton transfer to the leaving group and further underscore the "helper" function of Asp-33. The complete mechanism also includes activation of the phosphate group toward nucleophilic attack by a hydrogen bond between Arg-69 and a nonbridging oxygen atom. The overall mechanism can be described as "complex" general acid-general base since three elements are required for efficient catalysis.

The role of amino acid residues located in the active site pocket of phosphatidylinositol-specific phospholipase C (PI-PLC) from Bacillus cereus[Heinz, D. W., Ryan, M., Bullock, T., & Griffith, O. H. (1995) EMBO J. 14, 3855-3863] was investigated by site-directed mutagenesis, kinetics, and crystal structure analysis. Twelve residues involved in catalysis and substrate binding (His32, Arg69, His82, Gly83, Lys115, Glu117, Arg163, Trp178, Asp180, Asp198, Tyr200, and Asp274) were individually replaced by 1-3 other amino acids, resulting in a total number of 21 mutants. Replacements in the mutants H32A, H32L, R69A, R69E, R69K, H82A, H82L, E117K, R163I, D198A, D198E, D198S, Y200S, and D274S caused essentially complete inactivation of the enzyme. The remaining mutants (G83S, K115E, R163K, W178Y, D180S, Y200F, and D274N) exhibited reduced activities up to 57% when compared with wild-type PI-PLC. Crystal structures determined at a resolution ranging from 2.0 to 2.7 A for six mutants (H32A, H32L, R163K, D198E, D274N, and D274S) showed that significant changes were confined to the site of the respective mutation without perturbation of the rest of the structure. Only in mutant D198E do the side chains of two neighboring arginine residues move across the inositol binding pocket toward the newly introduced glutamic acid. An analysis of these structure-function relationships provides new insight into the catalytic mechanism, and suggests a molecular explanation of some of the substrate stereospecificity and inhibitor binding data available for this enzyme.

Posttranslational phosphorylation of proteins is an important event in many cellular processes. Phosphorylated tyrosine residues can serve as association sites for other proteins in signal transduction cascades of tyrosine kinase receptors. Formation of phosphohistidine residues in proteins has been found in eukaryotic and prokaryotic organisms. Furthermore, it has been suggested that phosphohistidine might substitute for phosphotyrosine in conferring high-affinity binding to proteins involved in signal transduction. We have analyzed the ability of 3-phosphohistidine to associate with the known phosphotyrosine-specific phosphotyrosine binding and src homology 2 protein domains. From our binding studies using synthetic peptides, we conclude that 3-phosphohistidine cannot replace phosphotyrosine in conferring high-affinity binding to the phosphotyrosine binding domain of shc or the src homology 2 domain of phospholipase C-gamma1.

Large enhancements (maximum of 82-fold in terms of enzyme efficiency, Vmax/Km) of bacterial PI-PLC cyclic phosphodiesterase activity were observed in the presence of organic solvents miscible in water (dimethyl sulfoxide, dimethylformamide, and 2-propanol). In general, organic solvents lowered the Km for myo-inositol 1,2-cyclic phosphate (cIP) and increased Vmax substantially. This kinetic effect was similar to that obtained with phosphatidylcholine micelles and bilayers in an aqueous assay system for cyclic inositol phosphate hydrolysis [Zhou, C., et al. (1997) Biochemistry 36, 347-355]. Solvent properties were examined to determine which ones correlated with the activation of PI-PLC toward cIP in each solvent. Activation correlated best with the solvent polarity as measured by ET(30); no significant correlation was observed with solution surface tension, the bulk dielectric constant (epsilon), 1/epsilon (a measure of the strength of charge interactions), or the Hildebrand solubility parameter. The sigmoidal curve of the enzyme activity versus solvent polarity was consistent with the solvent promoting a transition in the enzyme from a low-activity to a high-activity form. Possible candidates for this change, including enzyme dimerization, helix B/loop stabilization, and dehydration of the active site, are discussed.

Proteins recognize ligands by forming specific intermolecular interactions that often involve solvent exposed residues. Changes in the motional properties of these residues upon binding can affect the conformational entropy of the system and thus are related to the energetics of binding. The role that dynamics plays in ligand recognition can be investigated by comparing the motional properties of a free and ligated protein. NMR relaxation studies are well suited for examining changes in dynamics, especially for motions on a nanosecond to picosecond time scale. Recently, we determined the solution structure of the phosphotyrosine binding (PTB) domain of the insulin receptor substrate (IRS-1) complexed to a tyrosine-phosphorylated peptide derived from the interleukin 4 (IL-4) receptor [Zhou et al., (1996) Nat. Struct. Biol. 3, 388-393]. The peptide binds tightly to the protein in a surface exposed pocket, resulting in the partial burial of many protein residues. Using NMR relaxation studies, the dynamics of the backbone nitrogens of IRS-1 PTB domain were studied in both the free protein and the protein when complexed to the IL-4 receptor phosphopeptide. The backbone nitrogens of many residues that make important contacts to the ligand are motionally restricted in the free and complexed protein. Additional residues become motionally restricted only after ligand binding, including several residues that do not make any direct contacts with the ligand. These observed changes in the dynamics are compared to structural features of the complex.

We have determined the crystal structures of complexes of phosphoinositide-specific phospholipase C-delta1 from rat with calcium, barium, and lanthanum at 2.5-2.6 A resolution. Binding of these metal ions is observed in the active site of the catalytic TIM barrel and in the calcium binding region (CBR) of the C2 domain. The C2 domain of PLC-delta1 is a circularly permuted topological variant (P-variant) of the synaptotagmin I C2A domain (S-variant). On the basis of sequence analysis, we propose that both the S-variant and P-variant topologies are present among other C2 domains. Multiple adjacent binding sites in the C2 domain were observed for calcium and the other metal/enzyme complexes. The maximum number of binding sites observed was for the calcium analogue lanthanum. This complex shows an array-like binding of three lanthanum ions (sites I-III) in a crevice on one end of the C2 beta-sandwich. Residues involved in metal binding are contained in three loops, CBR1, CBR2, and CBR3. Sites I and II are maintained in the calcium and barium complexes, whereas sites II and III coincide with a binary calcium binding site in the C2A domain of synaptotagmin I. Several conformers for CBR1 are observed. The conformation of CBR1 does not appear to be strictly dependent on metal binding; however, metal binding may stabilize certain conformers. No significant structural changes are observed for CBR2 or CBR3. The surface of this ternary binding site provides a cluster of freely accessible liganding positions for putative phospholipid ligands of the C2 domain. It may be that the ternary metal binding site is also a feature of calcium-dependent phospholipid binding in solution. A ternary metal binding site might be a conserved feature among C2 domains that contain the critical calcium ligands in their CBR's. The high cooperativity of calcium-mediated lipid binding by C2 domains described previously is explained by this novel type of calcium binding site.

The crystal structures of various ternary complexes of phosphoinositide-specific phospholipase C-delta 1 from rat with calcium and inositol phosphates have been determined at 2.30-2.95 A resolution. The inositol phosphates used in this study mimic the binding of substrates and the reaction intermediate and include D-myo-inositol-1,4,5-trisphosphate, D-myo-inositol-2,4, 5-trisphosphate. D-myo-inositol-4,5-bisphosphate, and D,1-myo-inositol-2-methylene-1,2-cyclićmonophosphonate. The complexes exhibit an almost invariant mode of binding in the active site, each fitting edge-on into the active site and interacting with both the enzyme and the catalytic calcium at the bottom of the active site. Most of the active site residues do not undergo conformational changes upon binding either calcium or inositol phosphates. The structures are consistent with bidentate liganding of the catalytic calcium to the inositol phosphate intermediate and transition state. The complexes suggest explanations for substrate preference, pH optima, and ratio of cyclic to acyclic reaction products. A reaction mechanism is derived that supports general acid/base catalysis in a sequential mechanism involving a cyclic phosphate intermediate and rules out a parallel mechanism where acyclic and cyclic products are simultaneously generated.

We have measured the membrane binding affinities of purified phosphatidylinositol-specific phospholipases C-beta 1 and C-beta 2 to membranes of varying lipid composition using fluorescence methods. Our studies show that these proteins bind with affinities of 10(-5)-10(-4) M, with a small dependence on lipid type. Binding was relatively insensitive to the presence of phosphatidylinositol-specific phospholipases C-beta s' major physiological substrate, phosphatidylinositiol 4,5-bisphosphate, as well as the presence of Ca2+, which is required for activity. The presence of purified GTP gamma S-activated alpha 11 subunits of heterotrimeric guanine nucleotide binding proteins (G proteins) did not alter the membrane binding affinity of phosphatidylinositol-specific phospholipases C-beta 1, even though alpha 11 is a potent activator of this protein. Similarly, the presence of purified beta gamma subunits of G proteins did not alter the membrane association of phosphatidylinositol-specific phospholipases C-beta 2 even though these subunits strongly activate this isoform. These results argue against a recruitment model for PLC-beta activation by G proteins, negatively charged lipids, Ca2+, or substrate, and suggest that activation occurs through association of the membrane-bound species.

A series of site-specific mutants of the phosphatidylcholine-preferring phospholipase C from Bacillus cereus (PLCBc) was prepared in which the glutamic acid residue at position 146 was replaced with glutamine, aspartic acid, histidine, and leucine to elucidate what role Glu146 might play in catalysis. An expression system for the native enzyme in Escherichia coli was first developed to provide PLCBc that was fused via an intervening factor Xa protease recognition sequence at its N-terminus to maltose binding protein (MBP). This MBP-PLCBc fusion protein was isolated at levels of 50-70 mg/L of culture; selective trypsin digestion of the MBP-PLCBc fusion protein followed by chromatographic purification yielded recombinant PLCBc at levels of ca. 10 mg/L. Polymerase chain reaction (PCR) mutagenesis on the PLCBc gene (plc) was then used to replace the Glu146 codon with those for glutamine (E146Q), aspartic acid (E146D), histidine (E146H), and leucine (E146L). The catalytic efficiency of the E146Q mutant was 1.6% that of native PLCBc, while the other mutants each possessed activities of 0.2-0.3% of the wild type. The kcat/Km vs pH profiles for both E146Q and native PLCBc have ascending acidic limbs, suggesting that Glu146 does not serve as the general base in the hydrolysis reaction. As measured by circular dichroism, all of the mutant proteins contained less helical structure and underwent denaturation at lower temperatures than the wild type in the order: wild type > E146Q > E146D approximately E146H approximately E146L. Atomic absorption analyses indicated that the mutant proteins also exhibited lower Zn2+ content than the wild type. Thus, the Glu146 residue in PLCBc stabilizes the secondary and tertiary structure of the enzyme and serves as a critical ligand for Zn2, but it does not appear to have any specific catalytic role.

Numerous proteins on the external surface of the plasma membrane are anchored by glycosylated derivatives of phosphatidylinositol (GPI), rather than by hydrophobic amino acids embedded in the phospholipid bilayer. These GPI anchors are cleaved by phosphatidylinositol-specific phospholipases C (PI-PLCs) to release a water-soluble protein with an exposed glycosylinositol moiety and diacylglycerol, which remains in the membrane. We have previously determined the crystal structure of Bacillus cereus PI-PLC, the enzyme which is widely used to release GPI-anchored proteins from membranes, as free enzyme and also in complex with myo-inositol [Heinz, D.W., Ryan, M. Bullock, T.L., & Griffith, O. H. (1995) EMBO J. 14, 3855-3863]. Here we report the refined 2.2 A crystal structure of this enzyme complexed with a segment of the core of all GPI anchors, glucosaminyl(alpha 1-->6)-D-myo-inositol [GlcN-(alpha 1-->6)Ins ]. The myo-inositol moiety of GlcN(alpha 1-->6)Ins is well-defined and occupies essentially the same position in the active site as does free myo-inositol, which provides convincing evidence that the enzyme utilizes the same catalytic mechanism for cleavage of PI and GPI anchors. The myo-inositol moiety makes several specific hydrogen bonding interactions with active site residues. In contrast, the glucosamine moiety lies exposed to solvent at the entrance of the active site with minimal specific protein contacts. The glucosamine moiety is also less well-defined, suggesting enhanced conformational flexibility. On the basis of the positioning of GlcN(alpha 1-->6)Ins in the active site, it is predicted that the remainder of the GPI-glycan makes little or no specific interactions with B. cereus PI-PLC. This explains why B. cereus PI-PLC can cleave GPI anchors having variable glycan structures.

Protein-protein interfaces can consist of interactions between large numbers of residues of each molecule; some of these interactions are critical in determining binding affinity and conferring specificity, while others appear to play only a marginal role. Src-homology-2 (SH2) domains bind to proteins containing phosphorylated tyrosines, with additional specificity provided by interactions with residues C-terminal to the phosphotyrosine (pTyr) residue. While the C-terminal SH2 domain of phospholipase C-gamma 1 (PLCC SH2) interacts with eight residues of a pTyr-containing peptide from its high affinity binding site on the beta-platelet-derived growth factor receptor, it can still bind tightly to a phosphopeptide containing only three residues. Novel deuterium (2H) based nuclear magnetic resonance (NMR) spin relaxation experiments which probe the nanosecond-picosecond time scale dynamics of methyl containing side chain residues have established that certain regions of the PLCC SH2 domain contacting the residues C-terminal to the pTyr have a high degree of mobility in both the free and peptide complexed states. In contrast, there is significant restriction of motion in the pTyr binding site. These results suggest a correlation between the dynamic behavior of certain groups in the PLCC SH2 complex and their contribution to high affinity binding and binding specificity.

The pleckstrin homology (PH) domain of phospholipase C-delta 1 (PLC-delta 1) binds to phosphatidylinositol 4,5-bisphosphate (PI(4,5)P2) in phospholipid membranes with an affinity (Ka approximately 10(6) M-1) and specificity comparable to those of the native enzyme. PLC-delta 1 and its PH domain also bind inositol 1,4,5-trisphosphate, the polar head group of PI(4,5)P2, with comparable affinity and approximately 1:1 stoichiometry. A peptide corresponding to amino acids 30-43 of the PLC-delta 1 PH domain contains several basic residues predicted to bind PI(4,5)P2, but binds weakly and with little specificity for PI(4,5)P2; hence the tertiary structure of the isolated PH domain is required for high affinity PI(4,5)P2 binding. Our PI-(4,5)P2 binding results support the hypothesis that the intact PH domain, serving as a specific tether, directs PLC-delta 1 to membranes enriched in PI(4,5)P2 and permits the active site, located elsewhere in the protein, to hydrolyze multiple substrate molecules before this enzyme dissociates from the membrane surface.

Arginine side chains are often involved in protein--protein and protein--nucleic acid interactions. Due to a number of factors, resonance assignment and detection of NOEs involving the arginine side chains via standard NMR techniques can be difficult. We present here an approach to characterization of the interaction between a phosphopeptide (pY1021) and four arginine residues that line the phosphotyrosine-binding pocket of the C-terminal SH2 domain of phospholipase C-gamma 1 (PLCC SH2). Previously published [Pascal, S. M., et al. (1994) Cell 77, 461] NOE data provide a partial description of this interaction, including contacts between the aliphatic region of Arg 59 and the phosphotyrosine (pTyr) aromatic ring. Further characterization has now been accomplished by using 15N and 13C NMR relaxation studies of arginine N episilon and C zeta spins, respectively, and proton exchange rates of arginine H episilon nuclei. Differences between the chemical shifts of the arginine guanidino groups of the free SH2 domain in imidazole and phosphate buffers or in complex with pY1021 have provided insight into specific interactions with the phosphate and the aromatic ring of the pTyr. The resulting data are consistent with the most stable hydrogen bonds to phosphate donated by the Arg 39 epsilon-NH and the two Arg 37 eta-NH2 groups and with pTyr aromatic ring interactions involving the Arg 39 and possibly the Arg 18 guanidino groups.

Several phosphatidylinositol 4,5-bisphosphate (PtdInsP2)-regulated actin-binding proteins and most phosphoinositide-specific phospholipases C (PI-PLCs) comprise a basic amino acid motif (KxxxKxKK, where x denotes any amino acid), which was previously suggested to represent a PtdInsP2-binding site commonly present in these proteins. We have shown earlier that a peptide corresponding to amino acids 448-464 of human PLC beta 2 (LPSPEDLRGKILIKNKK, peptide P1) markedly and specifically stimulated the activity of this enzyme [Simões et al. (1993) FEBS Lett. 331, 248]. Here, we present a detailed analysis of the effects of various peptides related to peptide P1 aimed at understanding the mechanisms of peptide-mediated PLC beta 2 stimulation. Peptide KILIKNKK (P2), which comprises only the basic amino acid consensus motif, also stimulated PLC beta 2, although higher concentrations were required to observe this stimulatory effect. The effects of P1 and P2 were not additive, indicating that the two peptides affect PLC beta 2 activity via the same mechanism. Peptide LPSPEDLRG (P3), composed of the amino-terminal half of P1, did not affect the activity of PLC beta 2. Peptide KILIKNKKQFSGPTSS (P4), which includes the nine amino acids flanking the carboxy-terminus of the KILIKNKK motif within the sequence of PLC beta 2, stimulated the enzyme but was indistinguishable in potency from P2. Circular dichroism analysis revealed that peptide P1 changes its conformation in the presence of PtdInsP2 but not in the presence of other phospholipids including phosphatidylinositol 4-phosphate. The results suggest that the basic amino acid sequence physically interacts with PtdInsP2.(ABSTRACT TRUNCATED AT 250 WORDS)

Photoaffinity labeling and site-directed mutagenesis have been used to identify amino acid residues of the phospholipase C gamma 1 (PLC gamma 1) N-terminal SH2 domain involved in recognition of the activated epidermal growth factor receptor (EGFR). The photoactive amino acid p-benzoylphenylalanine (Bpa) was incorporated into phosphotyrosine-containing peptides derived from EGFR autophosphorylation sites Tyr992 and Tyr1068. Irradiation of these labels in the presence of SH2 domains showed cross-linking which was time-dependent and specific; labeling was inhibited with non-Bpa-containing peptides from EGFR in molar excess. The phosphotyrosine residue on the peptides was important for SH2 recognition, as dephosphorylated peptides did not cross-link. Radiolabeled peptides were used to identify sites of cross-linking to the N-terminal SH2 of PLC gamma 1. Bpa peptide-SH2 complexes were digested with trypsin, and radioactive fragments were purified by HPLC and analyzed by Edman sequencing. These experiments showed Arg562 and an additional site in the alpha A-beta B region of the SH2 domain, most likely Glu587, to be labeled by the Tyr992-derived peptide. Similar analysis of the reaction with the Tyr1068-derived photoaffinity label identified Leu653 as the cross-linked site. Mutation of the neighboring residues of Glu587 decreased photo-cross-linking, emphasizing the importance of this region of the molecule for recognition. These results are consistent with evidence from the v-Src crystal structure and implicate the loop spanning residues Gln640-Ser654 of PLC gamma 1 in specific recognition of phosphopeptides.

The backbone dynamics of the C-terminal SH2 domain of phospholipase C gamma 1 have been investigated. Two forms of the domain were studied, one in complex with a high-affinity binding peptide derived from the platelet-derived growth factor receptor and the other in the absence of this peptide. 2-D 1H-15N NMR methods, employing pulsed field gradients, were used to determine steady-state 1H-15N NOE values and T1 and T2 15N relaxation times. Backbone dynamics were characterized by the overall correlation time (tau m), order parameters (S2), effective correlation times for internal motions (tau e), and, if required, terms to account for motions on a microsecond-to-millisecond-time scale. An extended two-time-scale formalism was used for residues having relaxation data and that could not be fit adequately using a single-time-scale formalism. The overall correlation times of the uncomplexed and complexed forms of SH2 were found to be 9.2 and 6.5 ns, respectively, suggesting that the uncomplexed form is in a monomer-dimer equilibrium. This was subsequently confirmed by hydrodynamic measurements. Analysis of order parameters reveals that residues in the so-called phosphotyrosine-binding loop exhibited higher than average disorder in both forms of SH2. Although localized differences in order parameters were observed between the uncomplexed and complexed forms of SH2, overall, higher order parameters were not found in the peptide-bound form, indicating that on average, picosecond-time-scale disorder is not reduced upon binding peptide. The relaxation data of the SH2-phosphopeptide complex were fit with fewer exchange terms than the uncomplexed form. This may reflect the monomer-dimer equilibrium that exists in the uncomplexed form or may indicate that the complexed form has lower conformational flexibility on a microsecond-to-millisecond-time scale.

We added phospholipase C-delta 1 (PLC-delta) to the aqueous subphase beneath monolayers formed from mixtures of phosphatidylinositol 4,5-bisphosphate (2% PIP2), phosphatidylserine (33% PS), and phosphatidylcholine (65% PC) and then measured the initial rate of hydrolysis of PIP2 after addition of 10 microM free calcium. Increasing the surface pressure of the monolayer, pi, from 20 to 40 mN/m decreased the rate of hydrolysis 200-fold. The rate of hydrolysis depends exponentially on the surface pressure: rate alpha exp(-pi Ap/kT) where k is the Boltzmann constant, T is the temperature, and Ap congruent to 1 nm2. Similar results were obtained with different (1 and 100 microM) free [Ca2+] and with different mole fractions of PIP2. The results are consistent with a model in which PLC-delta binds to PIP2 with high affinity (Ka = 10(6) M-1) in the absence of calcium ions [Rebecchi, M.J., Peterson, A., & McLaughlin, S. (1993) Biochemistry (preceding paper in this issue)], and a portion of PLC-delta of area Ap inserts into the monolayer doing work = pi Ap prior to hydrolysis of PIP2. Removing the monovalent acidic lipid PS from the monolayer decreases the activity of PLC-delta 4-fold, this effect of PS on activity is similar to the effect of monovalent acidic lipids on the binding of PLC-delta to PIP2 in bilayer vesicles.

We studied the binding of phosphoinositide-specific phospholipase C-delta 1 (PLC-delta) to vesicles containing the negatively charged phospholipids phosphatidylinositol 4,5-bisphosphate (PIP2) and phosphatidylserine (PS). PLC-delta did not bind significantly to large unilamellar vesicles formed from the zwitterionic lipid phosphatidylcholine (PC) but bound strongly to vesicles formed from mixtures of PC and PIP2. The apparent association constant for the putative 1:1 complex formed between PLC-delta and PIP2 was Ka congruent to 10(5) M-1. The binding strength increased further (Ka congruent to 10(6) M-1) when the vesicles also contained 30% PS. High-affinity binding of PLC-delta to PIP2 did not require Ca2+. PLC-delta bound only weakly to vesicles formed from mixtures of PC and either PS or phosphatidylinositol (PI); binding increased as the mole fraction of acidic lipid in the vesicles increased. We also studied the membrane binding of a small basic peptide that corresponds to a conserved region of PLC. Like PLC-delta, the peptide bound weakly to vesicles containing monovalent negatively charged lipids; unlike PLC-delta, it did not bind strongly to vesicles containing PIP2. Our data suggest that a significant fraction of the PLC-delta in a cell could be bound to PIP2 on the cytoplasmic surface of the plasma membrane.