Structure, function and interfacial allosterism in phospholipase A2: insight from the anion-assisted dimer

Enzymes that function on membrane surfaces offer many challenges to understanding structural and functional details due to the difficulties of obtaining relevant information of the protein in a physiological environment. Focusing on this aspect of structural biology, it is important to develop conditions that mimic the interaction of membrane proteins with their binding surface and ultimately the mechanisms of action. This approach has been used to characterize the allosteric nature of secreted phospholipase A2 (PLA2) to its substrate interface. The breakthrough here was to crystallize the pancreatic group-IB PLA2 in an anion-assisted dimer with five coplanar phosphate anions bound. In the anion-assisted dimer structure one molecule of a tetrahedral mimic inhibitor and five anions are shared between the two subunits of the dimer. The sn-2-phosphate of the inhibitor, which mimics the tetrahedral intermediate of the esterolysis reaction, is bound in the active site of one subunit, and the alkyl chain extends into the active site slot of the second subunit across the subunit-subunit interface. This interface-bound structural mimic provided insight into the active site environment and specific anionic interactions to the i-face of the protein. The presence or absence of a single critical active site water, corresponds to the difference between the activated or inactivated form of the enzyme. The anion-assisted dimer structure supports a calcium coordinated nucleophilic water mechanism, with its pK(a) modulated by this assisting water. This working model has been further strengthened with an enzyme-product complex structure solved with the hydrolysis products of the substrate PAF also bound to the anion-assisted dimer form of PLA2. Additional confirmation of the assisting-water mechanism comes from a structure of the inactive zymogen proPLA2 also crystallized in an anion-assisted dimer. Remarkably, the assisting water present in the activated complex is absent in this proPLA2 structure.     

Desolvation map of the i-face of phospholipase A2

The changes in the microenvironment of the Trp-3 on the i-face of pig pancreatic IB phospholipase A₂ (PLA2) provide a measure of the tight contact (Ramirez and Jain, Protein Sci. 9, 229-239, 1991) with the substrate interface during the processive interfacial turnover. Spectral changes from the single Trp-substituent at position 1, 2, 6, 10, 19, 20, 31, 53, 56 or 87 on the surface of W3F PLA2 are used to probe the Trp-environment. Based on our current understanding only the residue 87 is away from i-face, therefore all other mutants are well suited to report modest differences along the i-face. All Trp-mutants bind tightly to anionic vesicles. Only those with Trp at 1, 2 or 3 near the rim of the active site on the i-face cause significant perturbation of the catalytic functions. Most other Trp-mutants showed < 3-fold change in the interfacial processive turnover rate and the competitive inhibition by MJ33. Binding of calcium to the enzyme in the aqueous phase had modest effect on the Trp-emission intensity. However, on the binding of the enzyme to the interface the fluorescence change is large, and the rate of oxidation of the Trp-substituent with N-bromosuccinimide depends on the location of the Trp-substituent. These results show that the solvation environment of the Trp-substituents on the i-face is shielded in the enzyme bound to the interface. Additional changes are noticeable if the active site of the bound enzyme is also occupied, however, the catalytically inert zymogen of PLA2 (proPLA2) does not show such changes. Significance of these results in relation to the changes in the solvent accessibility and desolvation of the i-face of PLA2 at the interface is discussed.     

Desolvation map of the i-face of phospholipase A2

The changes in the microenvironment of the Trp-3 on the i-face of pig pancreatic IB phospholipase A2 (PLA2) provide a measure of the tight contact (Ramirez and Jain, Protein Sci. 9, 229-239, 1991) with the substrate interface during the processive interfacial turnover. Spectral changes from the single Trp-substituent at position 1, 2, 6, 10, 19, 20, 31, 53, 56 or 87 on the surface of W3F PLA2 are used to probe the Trp-environment. Based on our current understanding only the residue 87 is away from i-face, therefore all other mutants are well suited to report modest differences along the i-face. All Trp-mutants bind tightly to anionic vesicles. Only those with Trp at 1, 2 or 3 near the rim of the active site on the i-face cause significant perturbation of the catalytic functions. Most other Trp-mutants showed < 3-fold change in the interfacial processive turnover rate and the competitive inhibition by MJ33. Binding of calcium to the enzyme in the aqueous phase had modest effect on the Trp-emission intensity. However, on the binding of the enzyme to the interface the fluorescence change is large, and the rate of oxidation of the Trp-substituent with N-bromosuccinimide depends on the location of the Trp-substituent. These results show that the solvation environment of the Trp-substituents on the i-face is shielded in the enzyme bound to the interface. Additional changes are noticeable if the active site of the bound enzyme is also occupied, however, the catalytically inert zymogen of PLA2 (proPLA2) does not show such changes. Significance of these results in relation to the changes in the solvent accessibility and desolvation of the i-face of PLA2 at the interface is discussed.     

Crystal structure of phospholipase A2 complex with the hydrolysis products of platelet activating factor: equilibrium binding of fatty acid and lysophospholipid-ether at the active site may be mutually exclusive

We have solved the 1.55 A crystal structure of the anion-assisted dimer of porcine pancreatic group IB phospholipase A2 (PLA2), complexed with the products of hydrolysis of the substrate platelet activating factor. The dimer contains five coplanar phosphate anions bound at the contact surface between the two PLA2 subunits. This structure parallels a previously reported anion-assisted dimer that mimics the tetrahedral intermediate of PLA2 bound to a substrate interface [Pan, Y. H., et al. (2001) Biochemistry 40, 609-617]. The dimer structure has a molecule of the product acetate bound in subunit A and the other product 1-octadecyl-sn-glycero-3-phosphocholine (LPC-ether) to subunit B. Therefore, this structure is of the two individual product binary complexes and not of a ternary complex with both products in one active site of PLA2. Protein crystals with bound products were only obtained by cocrystallization starting from the initial substrate. In contrast, an alternate crystal form was obtained when PLA2 was cocrystallized with LPC-ether and succinate, and this crystal form did not contain bound products. The product bound structure has acetate positioned in the catalytic site of subunit A such that one of its oxygen atoms is located 3.5 A from the catalytic calcium. Likewise, a longer than typical Ca-to-Gly(32) carbonyl distance of 3.4 A results in a final Ca coordination that is four-coordinate and has distorted geometry. The other oxygen of acetate makes hydrogen bonds with N(delta)(1)-His(48), O(delta)(1)-Asp(49), and the catalytic assisting water (w7). In contrast, the glycerophosphocholine headgroup of LPC-ether in subunit B makes no contacts with calcium or with the catalytic residues His(48) or Asp(49). The tail of the LPC-ether is located near the active site pocket with the last nine carbons of the sn-1- acyl chain refined in two alternate conformations. The remaining atoms of the LPC-ether product have been modeled into the solvent channel but have their occupancies set to zero in the refined model due to disorder. Together, the crystallographic and equilibrium binding results with the two products show that the simultaneous binding of both the products in a single active site is not favored.     

The basis for k(cat) impairment in prophospholipase A(2) from the anion-assisted dimer structure

Kinetic results in this paper show that, contrary to earlier reports, pig pancreatic prophospholipase A(2) (proPLA2) does not hydrolyze monodisperse short chain phosphatidylcholine below the critical micelle concentration. ProPLA2 is active on an anionic interface, but at a rate that is decreased by more than 100-fold compared to that of PLA2, the active form. Solution studies show that both proPLA2 and PLA2 bind to an anionic interface and also bind a tetrahedral intermediate mimic at the active site. The 1.5 A resolution crystal structure of the anion-assisted dimer of proPLA2 reported in this paper is compared with the corresponding structure for PLA2 [Pan, Y. H., et al. (2001) Biochemistry 40, 609-617]. As a mimic for the forms bound to the anionic interface, these structures provide insights into the possible structural basis for the impaired chemical step of the zymogen. The proPLA2 dimer contained within one crystallographic asymmetric unit has one molecule of the inhibitor 1-hexadecyl-3-(trifluoroethyl)-sn-glycero-2-phosphomethanol and is bridged by four coplanar sulfate anions. Relative to the structure of PLA2, the subunit contact surface in proPLA2 displays a tilted orientation, an altered mode of inhibitor binding, displacement of a mechanistically significant loop that includes Tyr69, and a critical active site water seen in PLA2 that is not seen in proPLA2. These differences are interpreted to suggest possible origins of the functional differences between the pro and active enzyme at an anionic interface. A structural origin of this difference is discussed in terms of the calcium-coordinated activated water mechanism of the esterolysis reaction. Together, a comparison of the structures of the anion-assisted dimers of PLA2 and proPLA2 not only offers an explanation of why the zymogen form is k(cat)-impaired and binds poorly even to the anionic interface but also supports a mechanism for the activated enzyme that includes a critical second-sphere assisting water bridging His48 and the calcium-coordinated catalytic water.     

Control of the chemical step by leucine-31 of pancreatic phospholipase A2

A well-defined region of pancreatic and other secreted phospholipase A2 (PLA2), which we call the i-face, makes a molecular contact with the interface to facilitate and control the events and processivity of the interfacial catalytic turnover cycles. The structural features of the i-face and its allosteric relationship to the active site remain to be identified. As a part of the calcium binding (26-34) loop, Leu-31 is located on the surface near the substrate binding slot of PLA2. Analysis of the primary rate and equilibrium parameters of the Leu-31 substitution mutants of the pig pancreatic PLA2 shows that the only significant effect of the substitution is to impair the chemical step at the zwitterionic interface in the presence of added NaCl, and only a modest effect is seen on kcat at the anionic interface. Leu-31 substitutions have little effect on the binding of the enzyme to the interface; the affinity for certain substrate mimics is modestly influenced in W3F, L31W double mutant. The fluorescence emission results with the double mutant show that the microenvironment of Trp-31 is qualitatively different at the zwitterionic versus anionic interfaces. At both of the interfaces Trp-31 is not shielded from the bulk aqueous environment as it remains readily accessible to acrylamide and water. The NaCl-induced change in the Trp-31 emission spectrum of the double mutant on the zwitterionic interface is similar to that seen on the binding to the anionic interface. Together, the kinetic and spectroscopic results show that the form of PLA2 at the zwitterionic interface (Ez) is distinguishably different from the catalytically more efficient form at the anionic interface (Ea). This finding provides a structural basis for the two-state model for kcat activation by the anionic interface. In conjunction with earlier results we suggest that neutralization of certain cationic residues of PLA2 exerts a control on the calcium loop through residue 31.     

Contributions of residues of pancreatic phospholipase A2 to interfacial binding, catalysis, and activation

Primary rate and equilibrium parameters for 60 site-directed mutants of bovine pancreatic phospholipase A2 (PLA2) are analyzed so incremental contributions of the substitution of specific residues can be evaluated. The magnitude of the change is evaluated so a functional role in the context of the N- and C-domains of PLA2 can be assigned, and their relationship to the catalytic residues and to the i-face that makes contact with the interface. The effect of substitutions and interfacial charge is characterized by the equilibrium dissociation constant for dissociation of the bound enzyme from the interface (Kd), the dissociation constant for dissociation of a substrate mimic from the active site of the bound enzyme (KL), and the interfacial Michaelis constants, KM and kcat. Activity is lost (>99.9%) on the substitution of H48 and D49, the catalytic residues. A more than 95% decrease in kcat is seen with the substitution of F5, I9, D99, A102, or F106, which form the substrate binding pocket. Certain residues, which are not part of the catalytic site or the substrate binding pocket, also modulate kcat. Interfacial anionic charge lowers Kd, and induces kcat activation through K56, K53, K119, or K120. Significant changes in KL are seen by the substitution of N6, I9, F22, Y52, K53, N71, Y73, A102, or A103. Changes in KM [=(k2+k-1)/k1] are attributed to kcat (=k2) and KL (=k-1/k1). Some substitutions change more than one parameter, implying an allosteric effect of the binding to the interface on KS, and the effect of the interfacial anionic charge on kcat. Interpreted in the context of the overall structure, results provide insights into the role of segments and domains in the microscopic events of catalytic turnover and processivity, and their allosteric regulation. We suggest that the interfacial recognition region (i-face) of PLA2, due to the plasticity of certain segments and domains, exercises an allosteric control on the substrate binding and chemical step.     

Contiguous binding of decylsulfate on the interface-binding surface of pancreatic phospholipase A2

Pig pancreatic IB phospholipase A 2 (PLA2) forms three distinguishable premicellar E i (#) ( i = 1, 2, and 3) complexes at successively higher decylsulfate concentrations. The Hill coefficient for E 1 (#) is n 1 = 1.6, and n 2 and n 3 for E 2 (#) and E 3 (#) are about 8 each. Saturation-transfer difference nuclear magnetic resonance (NMR) and other complementary results with PLA2 show that decylsulfate molecules in E 2 (#) and E 3 (#) are contiguously and cooperatively clustered on the interface-binding surface or i-face that makes contact with the substrate interface. In these complexes, the saturation-transfer difference NMR signatures of (1)H in decylsulfate are different. The decylsulfate epitope for the successive E i (#) complexes increasingly resembles the micellar complex formed by the binding of PLA2 to preformed micelles. Contiguous cooperative amphiphile binding is predominantly driven by the hydrophobic effect with a modest electrostatic shielding of the sulfate head group in contact with PLA2. The formation of the complexes is also associated with structural change in the enzyme. Calcium affinity of E 2 (#) appears to be modestly lower than that of the free enzyme and E 1 (#). Binding of decylsulfate to the i-face does not require the catalytic calcium required for the substrate binding to the active site and for the chemical step. These results show that E i (#) complexes are useful to structurally characterize the cooperative sequential and contiguous binding of amphiphiles on the i-face. We suggest that the allosteric changes associated with the formation of discrete E i (#) complexes are surrogates for the catalytic and allosteric states of the interface activated PLA2.     

Structural basis of the anionic interface preference and kcat* activation of pancreatic phospholipase A2

Pancreatic phospholipase A(2) (PLA2) shows a strong preference for the binding to the anionic interface and a consequent allosteric activation. In this paper, we show that virtually all the preference is mediated through 3 (Lys-53, -56, and -120) of the 12 cationic residues of bovine pancreatic PLA2. The lysine-to-methionine substitution enhances the binding of the enzyme to the zwitterionic interface, and for the K53,56,120M triple mutant at the zwitterionic interface is comparable to that for the wild type (WT) at the anionic interface. In the isomorphous crystal structure, the backbone folding of K53,56M K120,121A and WT are virtually identical, yet a significant change in the side chains of certain residues, away from the site of substitution, mostly at the putative contact site with the interface (i-face), is discernible. Such reciprocity, also supported by the spectroscopic results for the free and bound forms of the enzyme, is expected because a distal structural change that perturbs the interfacial binding could also affect the i-face. The results show that lysine-to-methionine substitution induces a structural change that promotes the binding of PLA2 to the interface as well as the substrate binding to the enzyme at the interface. The kinetic results are consistent with a model in which the interfacial Michaelis complex exists in two forms, and the complex that undergoes the chemical step is formed by the charge compensation of Lys-53 and -56. Analysis of the incremental changes in the kinetic parameters shows that the charge compensation of Lys-53 and -56 contributes to the activation and that of Lys-120 contributes only to the structural change that promotes the stability of the Michaelis complex at the interface. The charge compensation effects on these three residues also account for the differences in the anionic interface preference of the evolutionarily divergent secreted PLA2.     

Cooperative binding of monodisperse anionic amphiphiles to the i-face: phospholipase A2-paradigm for interfacial binding

Equilibrium parameters for the binding of monodisperse alkyl sulfate along the i-face (the interface binding surface) of pig pancreatic IB phospholipase A(2) (PLA2) to form the premicellar complexes (E(i)(#)) are characterized to discern the short-range specific interactions. Typically, E(i)(#) complexes are reversible on dilution. The triphasic binding isotherm, monitored as the fluorescence emission from the single tryptophan of PLA2, is interpreted as a cooperative equilibrium for the sequential formation of three premicellar complexes (E(i)(#), i = 1, 2, 3). In the presence of calcium, the dissociation constant K(1) for the E(1)(#) complex of PLA2 with decyl sulfate (CMC = 4500 microM) is 70 microM with a Hill coefficient n(1) = 2.1 +/- 0.2; K(2) for E(2)(#) is 750 microM with n(2) = 8 +/- 1, and K(3) for E(3)(#) is 4000 microM with an n(3) value of about 12. Controls show that (a) self-aggregation of decyl sulfate alone is not significant below the CMC; (b) occupancy of the active site is not necessary for the formation of E(i)(#); (c) K(i) and n(i) do not change significantly due to the absence of calcium, possibly because alkyl sulfate does not bind to the active site of PLA2; (d) the E(i)(#) complexes show a significant propensity for aggregation; and (e) PLA2 is not denatured in E(i)(#). The results are interpreted to elaborate the model for atomic level interactions along the i-face: The chain length dependence of the fit parameters suggests that short-range specific anion binding of the headgroup is accompanied by desolvation of the i-face of E(i)(#). We suggest that allosteric activation of PLA2 results from such specific interactions of the amphiplies and the desolvation of the i-face. The significance of these primary interfacial binding events and the coexistence of the E and E(i)(#) aggregates is discussed.     

The interfacial binding surface of phospholipase A2s

For membrane-associated enzymes, which access substrate from either a monolayer or bilayer of the aggregate substrate, the partitioning from the aqueous phase to this phospholipid interface is critical for catalysis. Despite a large and expanding body of knowledge regarding interfacial enzymes, the biophysical steps involved in interfacial recognition and adsorption remain relatively poorly understood. The surface of the enzyme that contacts the phospholipid surface is referred to as its interfacial binding surface, or more simply, its i-face. The interaction of a protein's i-face with the aggregate substrate may simply control access to substrate. However, it can be more complex, and this interaction often serves to allosterically activate the enzyme on this surface. First we briefly review what is currently known about i-face structure and function for a prototypical interfacial enzyme, the secreted Phospholipase A2 (PLA2). Then we develop, characterize, compare, and discuss models of the PLA2 i-face across a subset of five homologous PLA2 family members, groups IA, IB, IIA, V, and X. A homology model of human group-V is included in this comparison, suggesting that a similar approach could be used to explore interfacial function of any of the PLA2 family members. Despite moderate sequence identity, structural homology and sequence similarity are well conserved. We find that the residues predicted to be interfacial, while conserved structurally, are not highly conserved in sequence. Implications for this divergence on interfacial selectivity are discussed.     

The crystal structure of the catalytic domain of the site-specific recombination enzyme gamma delta resolvase at 2.7 A resolution

The crystal structure of the catalytic domain of the site-specific recombination enzyme gamma delta resolvase has been determined at 2.7 A resolution. Its first 120 amino acids form a central five-stranded, beta-pleated sheet surrounded by five alpha helices. In one of the four dyad-related dimers, the two active site Ser-10 residues are 19 A apart, perhaps close enough to contact and become covalently linked to the DNA at the recombination site. This dimer also forms the only closely packed tetramer found in the crystal. The subunit interface at a second dyad-related dimer is more extensive and more highly conserved among the homologous recombinases; however, its active site Ser-10 residues are more than 30 A apart. Side chains, identified by mutations that eliminate catalysis but not DNA binding, are located on the subunit surface near the active site serine and at the interface between a third dyad-related pair of subunits of the tetramer.     

Dimers generated from tetrameric phosphorylating glyceraldehyde-3-phosphate dehydrogenase from Bacillus stearothermophilus are inactive but exhibit cooperativity in NAD binding

Tetrameric phosphorylating glyceraldehyde-3-phosphate dehydrogenase (GAPDH) from Bacillus stearothermophilus has been described as a "dimer of dimers" with three nonequivalent interfaces, P-axis (between subunits O and P and between subunits Q and R), Q-axis (between subunits O and Q and between subunits P and R), and R-axis interface (between subunits O and R and between subunits P and Q). O-P dimers, the most stable and the easiest to generate, have been created by selective disruption of hydrogen bonds across the R- and Q-axis interfaces by site-directed mutagenesis. Asp-186 and Ser-48, and Glu-276 and Tyr-46, which are hydrogen bond partners across the R- and Q-axis interfaces, respectively, have been replaced with glycine residues. All mutated residues are highly conserved among GAPDHs from different species and are located in loops. Both double mutants D186G/E276G and Y46G/S48G were dimeric, while all single mutants remained tetrameric. As previously described [Clermont, S., Corbier, C., Mely, Y., Gerard, D., Wonacott, A., and Branlant, G. (1993) Biochemistry 32, 10178-10184], NAD binding to wild type GAPDH (wtGAPDH) was interpreted according to the induced-fit model and exhibited negative cooperativity. However, NAD binding to wtGAPDH can be adequately described in terms of two independent dimers with two interacting binding sites in each dimer. Single mutants D186G, E276G, and Y46G exhibited behavior in NAD binding similar to that of the wild type, while both dimeric mutants D186G/E276G and Y46G/S48G exhibited positive cooperativity in binding the coenzyme NAD. The fact that O-P dimer mutants retained cooperative behavior shows that (1) the P-axis interface is important in transmitting the information induced upon NAD binding inside the O-P dimer from one subunit to the other and (2) the S-loop of the R-axis-related subunit is not directly involved in cooperative binding of NAD in the O-P dimer. In both O-P dimer mutants, the absorption band of the binary enzyme-NAD complex had a highly decreased intensity compared to that of the wild type and, in addition, totally disappeared in the presence of G3P or 1,3-dPG. However, no enzymatic activity was detected, indicating that the formed ternary enzyme-NAD-G3P or -1, 3-dPG complex was not catalytically efficient. In the O-P dimers, the interaction with the S-loop of the R-axis-related subunit is disrupted, and therefore, the S-loop should be less structured. This resulted in increased accessibility of the active site to the solvent, particularly for the adenosine-binding site of NAD. Thus, together, this is likely to explain both the lowered affinity of the dimeric enzyme for NAD and the absence of activity.     

N-terminal domain swapping and metal ion binding in nitric oxide synthase dimerization

Nitric oxide synthase oxygenase domains (NOS(ox)) must bind tetrahydrobiopterin and dimerize to be active. New crystallographic structures of inducible NOS(ox) reveal that conformational changes in a switch region (residues 103-111) preceding a pterin-binding segment exchange N-terminal beta-hairpin hooks between subunits of the dimer. N-terminal hooks interact primarily with their own subunits in the 'unswapped' structure, and two switch region cysteines (104 and 109) from each subunit ligate a single zinc ion at the dimer interface. N-terminal hooks rearrange from intra- to intersubunit interactions in the 'swapped structure', and Cys109 forms a self-symmetric disulfide bond across the dimer interface. Subunit association and activity are adversely affected by mutations in the N-terminal hook that disrupt interactions across the dimer interface only in the swapped structure. Residue conservation and electrostatic potential at the NOS(ox) molecular surface suggest likely interfaces outside the switch region for electron transfer from the NOS reductase domain. The correlation between three-dimensional domain swapping of the N-terminal hook and metal ion release with disulfide formation may impact inducible nitric oxide synthase (i)NOS stability and regulation in vivo.     

Interactions between the catalytic centers and subunit interface of triosephosphate isomerase probed by refolding, active site modification, and subunit exchange.

The effects of unfolding, refolding, and hybridization of triosephosphate isomerase (TPI) subunits from different species and subunits which have been specifically modified at the active site have been examined. These effects have been evaluated in terms of changes in catalytic parameters, CD spectra, and susceptibility to denaturation. Dissociation followed by reassociation yields an active dimer but with increased Km, reduced kcat, and increased susceptibility to inactivation and unfolding in denaturants. These data suggest that while the general structure of the refolded dimer is similar to the native enzyme, its complete original structure is not restored. Covalent reaction of the active site Glu165 with the substrate analogue 3-chloroacetol phosphate (CAP) results in dimers with increased susceptibility to unfolding and inactivation by denaturants (i.e. the rates of inactivation and unfolding are (TPICAP)2 greater than (TPI-TPICAP) greater than (TPI)2). These data point to the interactions between the catalytic center and the subunit interface. Subunits of TPI from different species, in spite of structural differences at the subunit interface, hybridized to active heterodimers. Subunit hybridization was random among monomers from different mammals, preferential between yeast and mammalian or avian monomers. Hybridization did not occur between avian and mammalian monomers under these conditions. These data provide information on the elements in the interface of the dimer and the relationship of the catalytic center with the subunit interface.     

The structure of bovine F1-ATPase in complex with its regulatory protein IF1

In mitochondria, the hydrolytic activity of ATP synthase is prevented by an inhibitor protein, IF1. The active bovine protein (84 amino acids) is an alpha-helical dimer with monomers associated via an antiparallel alpha-helical coiled coil composed of residues 49-81. The N-terminal inhibitory sequences in the active dimer bind to two F1-ATPases in the presence of ATP. In the crystal structure of the F1-IF1 complex at 2.8 A resolution, residues 1-37 of IF1 bind in the alpha(DP)-beta(DP) interface of F1-ATPase, and also contact the central gamma subunit. The inhibitor opens the catalytic interface between the alpha(DP) and beta(DP) subunits relative to previous structures. The presence of ATP in the catalytic site of the beta(DP) subunit implies that the inhibited state represents a pre-hydrolysis step on the catalytic pathway of the enzyme.     

Membrane binding modulates the quaternary structure of CTP:phosphocholine cytidylyltransferase

CTP:phosphocholine cytidylyltransferase (CCT), a key enzyme that controls phosphatidylcholine synthesis, is regulated by reversible interactions with membranes containing anionic lipids. Previous work demonstrated that CCT is a homodimer. In this work we show that the structure of the dimer interface is altered upon encountering membranes that activate CCT. Chemical cross-linking reactions were established which captured intradimeric interactions but not random CCT dimer collisions. The efficiency of capturing covalent cross-links with four different reagents was diminished markedly upon presentation of activating anionic lipid vesicles but not zwitterionic vesicles. Experiments were conducted to show that the anionic vesicles did not interfere with the chemistry of the cross-linking reactions and did not sequester available cysteine sites on CCT for reaction with the cysteine-directed cross-linking reagent. Thus, the loss of cross-linking efficiency suggested that contact sites at the dimer interface had increased distance or reduced flexibility upon binding of CCT to membranes. The regions of the enzyme involved in dimerization were mapped using three approaches: 1) limited proteolysis followed by cross-linking of fragments, 2) yeast two-hybrid analysis of interactions between select domains, and 3) disulfide bonding potential of CCTs with individual cysteine to serine substitutions for the seven native cysteines. We found that the N-terminal domain (amino acids 1-72) is an important participant in forming the dimer interface, in addition to the catalytic domain (amino acids 73-236). We mapped the intersubunit disulfide bond to the cystine 37 pair in domain N and showed that this disulfide is sensitive to anionic vesicles, implicating this specific region in the membrane-sensitive dimer interface.     

Evidence for two conformers of the beta subunit of tryptophan synthase in solution.

To explain our finding that the dimeric beta subunit of tryptophan synthase is only 50% inactivated by beta-chloro-L-alanine (Ahmed, S. A., Ruvinov, S. B., Kayastha, A. M., and Miles, E. W. (1991) J. Biol. Chem. 266, 21548-21557), we have extended our investigation using spectroscopic, steady-state kinetic, and electrophoretic methods. The spectroscopic properties of the half-active beta 2 dimer and the reactivation after alkali treatment show that the inactivation proceeds by an "enamine" mechanism. Although the fully active beta 2 dimer associates with the tryptophan synthase alpha subunit to form alpha 2 beta 2 complex, the inactive beta subunits in the half-active enzyme associate weakly or not at all with the alpha subunit. Our results provide evidence for two conformers of the beta subunit in solution: one is rapidly inactivated by beta-chloro-L-alanine and the other is not inactivated. Thermal inactivation studies and non-denaturing polyacrylamide gel electrophoresis of the half-active enzyme show that the beta 2 dimer exists in both homologous and heterologous combinations of these two forms. After removal of the reaction products and unreacted beta-chloro-L-alanine from the half-active beta 2 dimer by gel filtration, further incubation with beta-chloro-L-alanine results in the loss of 50% of the remaining activity. This result suggests that the subunits undergo rearrangement via an intermediate monomer form to regenerate the two conformers of the active beta subunit. This mechanism of rearrangement is supported by our finding that the extent of inactivation increases at lower concentrations of the beta 2 dimer.     

Dihydroorotase from Escherichia coli: loop movement and cooperativity between subunits

Escherichia coli dihydroorotase has been crystallized in the presence of the product, L-dihydroorotate (L-DHO), and the structure refined at 1.9A resolution. The structure confirms that previously reported (PDB entry 1J79), crystallized in the presence of the substrate N-carbamyl-D,L-aspartate (D, L-CA-asp), which had a dimer in the asymmetric unit, with one subunit having the substrate, L-CA-asp bound at the active site and the other having L-DHO. Importantly, no explanation for the unusual structure was given. Our results now show that a loop comprised of residues 105-115 has different conformations in the two subunits. In the case of the L-CA-asp-bound subunit, this loop reaches in toward the active site and makes hydrogen-bonding contact with the bound substrate molecule. For the L-DHO-bound subunit, the loop faces in the opposite direction and forms part of the surface of the protein. Analysis of the kinetics for conversion of L-DHO to L-CA-asp at low concentrations of L-DHO shows positive cooperativity with a Hill coefficient n=1.57(+/-0.13). Communication between subunits in the dimer may occur via cooperative conformational changes of the side-chains of a tripeptide from each subunit: Arg256-His257-Arg258, near the subunit interface.