Substrate binding and catalytic mechanism in phospholipase C from Bacillus Cereus: A molecular mechanics and molecular dynamics study

For the first time a consistent catalytic mechanism of phospholipase C from Bacillus cereus is reported based on molecular mechanics calculations. We have identified the position of the nucleophilic water molecule, which is directly involved in the hydrolysis of the natural substrate, phosphatidylcholine, in phospholipase C. This catalytically essential water molecule, after being activated by an acidic residue (Asp55), performs the nucleophilic attack on the phosphorus atom in the substrate, leading to a trigonal bipyramidal pentacoordinated intermediate (and structurally similar transition state). The subsequent collapse of the intermediate, regeneration of the enzyme, and release of the products has to involve a not yet identified second water molecule. The catalytic mechanism reported here is based on a series of molecular mechanics calculations. First, the x-ray structure of phospholipase C from B. cereus including a docked substrate molecule was subjected to a stepwise molecular mechanics energy minimization. Second, the location of the nucleophilic water molecule in the active site of the fully relaxed enzyme–substrate complex was determined by evaluation of nonbonded interaction energies between the complex and a water molecule. The nucleophilic water molecule is positioned at a distance (3.8 Å) from the phosphorus atom in the substrate, which is in good agreement with experimentally observed distances. Finally, the stability of the complex between phospholipase C, the substrate, and the nucleophilic water molecule was verified during a 100 ps molecular dynamics simulation. During the simulation the substrate undergoes a conformational change, but retains its localization in the active site. The contacts between the enzyme, the substrate, and the nucleophilic water molecule display some fluctuations, but remain within reasonable limits, thereby confirming the stability of the enzyme–substrate–water complex. The protocol developed for energy minimization of phospholipase C containing three zinc ions located closely together at the bottom of the active site cleft is reported in detail. In order to handle the strong electrostatic interactions in the active site realistically during energy minimization, delocalization of the charges from the three zinc ions was considered. Therefore, quantum mechanics calculations on the zinc ions and the zinc-coordinating residues were carried out prior to the molecular mechanics calculations, and two different sets of partial atomic charges (MNDO-Mulliken and AM1-ESP) were applied. After careful assignment of partial atomic charges, a complete energy minimization of the protein was carried out by a stepwise procedure without explicit solvent molecules. Energy minimization with either set of charges yielded structures, which were very similar both to the x-ray structure and to each other, although using AM1-ESP partial atomic charges and a dielectric constant of 4, yielded the best protein structure. © 1997 John Wiley and Sons, Inc. Biopoly 42: 319–336, 1997     

Structural mechanisms of constitutive activation in the C5a receptors with mutations in the extracellular loops: molecular modeling study

Previously we demonstrated by random saturation mutagenesis a set of mutations in the extracellular (EC) loops that constitutively activate the C5a receptor (C5aR) (Klco et al., Nat Struct Mol Biol 2005;12:320-326; Klco et al., J Biol Chem 2006;281:12010-12019). In this study, molecular modeling revealed possible conformations for the extracellular loops of the C5a receptors with mutations in the EC2 loop or in the EC3 loop. Comparison of low-energy conformations of the EC loops defined two distinct clusters of conformations typical either for strongly constitutively active mutants of C5aR (the CAM cluster) or for nonconstitutively active mutants (the non-CAM cluster). In the CAM cluster, the EC3 loop was turned towards the transmembrane (TM) helical bundle and more closely interacted with EC2 than in the non-CAM cluster. This suggested a structural mechanism of constitutive activity where EC3 contacts EC2 leading to EC2 interactions with helix TM3, thus triggering movement of TM7 towards TM2 and TM3. The movement initiates rearrangement of the system of hydrogen bonds between TM2, TM3 and TM7 including formation of the hydrogen bond between the side chains of D82(2.50) in TM2 and N296(7.49) in TM7, which is crucial for formation of the activated states of the C5a receptors (Nikiforovich et al., Proteins: Struct Funct Gene 2011;79:787-802). Since the relative large length of EC3 in C5aR (13 residues) is comparable with those in many other members of rhodopsin family of GPCRs (13-19 residues), our findings might reflect general mechanisms of receptor constitutive activation. The very recent X-ray structure of the agonist-induced constitutively active mutant of rhodopsin (Standfuss et al., Nature 2011;471:656-660) is discussed in view of our modeling results.     

The effect of different molecular species of sphingomyelin on phospholipase C δ1 activity

Bovine brain sphingomyelin was separated into different molecular species using a reverse phase column. PLC δ1 was inhibited by all molecular species of sphingomyelin. The extent of this inhibition was dependent on the hydrophobicity. Based on fatty acid analysis, we conclude that the inhibition of PLC δ1 depends on the chain length and degree of unsaturation of the fatty acid moiety of SM. N-palmitoyl-D-sphingomyelin and N-stearoyl-D-sphingomyelin inhibited PLC δ1 less then N-oleoyl-D-sphingomyelin. In the absence of Ca²⁺ (1 mM EGTA) all tested molecular species of SM inhibited weakly the enzyme. The sensitivity of PLC δ1 to inhibition by SM increased with increasing Ca²⁺ concentration. The shape of calcium curve differed for molecular species with saturated and unsaturated fatty acids. Inhibition of PLC δ1 by N-palmitoyl-D-sphingomyclin and N-stearoyl-D-sphingomyclin reached a maximum at 0.2 μM Ca²⁺, while inhibition by N-oleoyl-D-sphingomyclin reached maximum at 2 μM Ca²⁺. PLC δ1 is more sensitive to inhibition by SM when it is maximally activated by spermine and calcium and the extent of this inhibition depends on the length and degree of fatty acid unsaturation of the molecular species.     

Investigation of the molecular interactions of triclocarban with human serum albumin using multispectroscopies and molecular modeling

Triclocarban (TCC), as a broad spectrum antibacterial agent widely used in personal care products, has recently been recognized as environmental pollutant with the potential of adversely affecting wildlife and human health. However, the behavior of TCC in blood circulatory system and the potential toxicity of TCC at the molecular level have been poorly investigated. In this study, the effect of TCC on human serum albumin (HSA) and the binding mechanism of TCC to HSA were examined using spectroscopic techniques and molecular modeling methods. The fluorescence results suggested that the fluorescence of HSA was quenched by TCC through a static quenching mechanism and nonradiation energy transfer, and TCC was bound to HSA with moderately strong binding affinity via hydrophobic interaction based on the analysis of the thermodynamic parameters. The site marker competitive experiments revealed that TCC bound into subdomain IIA (site I) of HSA. In addition, the results obtained from the circular dichroism, Fourier transform infrared (FT-IR), 8-anilino-1-naphthalenesulfonic acid fluorescence, synchronous fluorescence, three-dimensional fluorescence spectra and dynamic light scattering suggested the change in the microenvironment and conformation of HSA during the binding reaction. Finally, the best binding mode of TCC and specific interaction of TCC with amino acid residues were determined using molecular docking and molecular dynamics simulations. In a word, the present studies can provide a way to help us well understand the transport, distribution and toxicity effect of TCC when it diffused in the human body.

Communicated by Ramaswamy H. Sarma     

Structural insights into the interactions of phorbol ester and bryostatin complexed with protein kinase C: a comparative molecular dynamics simulation study

Protein kinase C (PKC) isozymes are important regulatory enzymes that have been implicated in many diseases, including cancer, Alzheimer’s disease, and in the eradication of HIV/AIDS. Given their potential clinical ramifications, PKC modulators, e.g. phorbol esters and bryostatin, are also of great interest in the drug development. However, structural details on the binding between PKC and its modulators, especially bryostatin – the highly potent and non-tumor promoting activator for PKCs, are still lacking. Here, we report the first comparative molecular dynamics study aimed at gaining structural insight into the mechanisms by which the PKC delta cys2 activator domain is used in its binding to phorbol ester and bryostatin-1. As anticipated in the phorbol ester binding, hydrogen bonds are formed through the backbone atoms of Thr242, Leu251, and Gly253 of PKC. However, the opposition of H-bond formation between Thr242 and Gly253 may cause the phorbol ester complex to become less stable when compared with the bryostatin binding. For the PKC delta-bryostatin complex, hydrogen bonds are formed between the Gly253 backbone carbonyl and the C30 carbomethoxy substituent of the ligand. Additionally, the indole Nε1 of the highly homologous Trp252 also forms an H-bond to the C20 ester group on bryostatin. Backbone fluctuations also suggest that this latter H-bond formation may abrogate the transient interaction between Trp252 and His269, thus dampening the fluctuations observed on the nearby Zn²⁺-coordinating residues. This new dynamic fluctuation dampening model can potentially benefit future design of new PKC modulators.     

Myrosinase: insights on structural,catalytic, regulatory,and environmental interactions

Glucosinolate–myrosinase is a substrate-enzyme defense mechanism present in Brassica crops. This binary system provides the plant with an efficient system against herbivores and pathogens. For humans, it is well known for its anti-carcinogenic, anti-inflammatory, immunomodulatory, anti-bacterial, cardio-protective, and central nervous system protective activities. Glucosinolate and myrosinase are spatially present in different cells that upon tissue disruption come together and result in the formation of a variety of hydrolysis products with diverse physicochemical and biological properties. The myrosinase-catalyzed reaction starts with cleavage of the thioglucosidic linkage resulting in release of a D-glucose and an unstable thiohydroximate-O-sulfate. The outcome of this thiohydroximate-O-sulfate has been shown to depend on the structure of the glucosinolate side chain, the presence of supplementary proteins known as specifier proteins and/or on the physiochemical condition. Myrosinase was first reported in mustard seed during 1939 as a protein responsible for release of essential oil. Until this date, myrosinases have been characterized from more than 20 species of Brassica, cabbage aphid, and many bacteria residing in the human intestine. All the plant myrosinases are reported to be activated by ascorbic acid while aphid and bacterial myrosinases are found to be either neutral or inhibited. Myrosinase catalyzes hydrolysis of the S-glycosyl bond, O-β glycosyl bond, and O-glycosyl bond. This review summarizes information on myrosinase, an essential component of this binary system, including its structural and molecular properties, mechanism of action, and its regulation and will be beneficial for the research going on the understanding and betterment of the glucosinolate–myrosinase system from an ecological and nutraceutical perspective.     

Nucleotide-CF1 interactions and current views on the catalytic mechanism

The binding of nucleotides on isolated subunits as well as on reconstituted CF1 core complex is reviewed. Nucleotide interaction with CF1 and consequent ATPase activity are always associated with the presence of Mg2+. The metal binding site studies using Electron Paramagnetic Resonance (EPR) and pulsed EPR conclude that the metal binding occurs prior to any nucleotide addition. The addition of nucleotide does not modify the enzyme's metal binding site but brings on additional ligands with the phosphates of the nucleotides. The ATPase and nucleotide binding experiments with CF1 are also better interpreted by the hypothesis that Mg2+ is an activator rather than an inhibitor of the enzyme and that the actual substrate of CF1-ATPase is ATP rather than MgATP. The dual role of tentoxin as an inhibitor at low concentration (10^-8-10^-7 M) and activator at higher concentrations (10^-6 M) of the enzymatic activity of CF1, is due to the presence of two different binding sites on CF1. The synthesis of a new cyclic analogue of tentoxin with alanine changed for a serine has shown that it was possible to dissociate the two roles. The serine tentoxin analogue has the same inhibition effect on CF1 but is no longer an activator. The binding of nucleotides may influence the stability, produce structural changes and, over long distance, cause movements of CF1. All these effects of nucleotide or metal binding and activation or inhibition of CF1 may help also to elucidate the role played by the catalytic and non catalytic sites. These questions are reviewed and analyzed with respect to the current views on the catalytic mechanism.     

Insights into the catalytic mechanism of cofactor-independent phosphoglycerate mutase from X-ray crystallography,simulated dynamics and molecular modeling

Phosphoglycerate mutases catalyze the isomerization of 2 and 3-phosphoglycerates, and are essential for glucose metabolism in most organisms. Here, we further characterize the 2,3-bisphosphoglycerate-independent phosphoglycerate mutase (iPGM) from Bacillus stearothermophilus by determination of a high-resolution (1.4A) crystal structure of the wild-type enzyme and the crystal structure of its S62A mutant. The mutant structure surprisingly showed the replacement of one of the two catalytically essential manganese ions with a water molecule, offering an additional possible explanation for its lack of catalytic activity. Crystal structures invariably show substrate phosphoglycerate to be entirely buried in a deep cleft between the two iPGM domains. Flexibility analyses were therefore employed to reveal the likely route of substrate access to the catalytic site through an aperture created in the enzyme's surface during certain stages of the catalytic process. Several conserved residues lining this aperture may contribute to orientation of the substrate as it enters. Factors responsible for the retention of glycerate within the phosphoenzyme structure in the proposed mechanism are identified by molecular modeling of the glycerate complex of the phosphoenzyme. Taken together, these results allow for a better understanding of the mechanism of action of iPGMs. Many of the results are relevant to a series of evolutionarily related enzymes. These studies will facilitate the development of iPGM inhibitors which, due to the demonstrated importance of this enzyme in many bacteria, would be of great potential clinical significance.     

Docking of 4‐oxalocrotonate tautomerase substrates: Implications for the catalytic mechanism

The enzyme 4‐oxalocrotonate tautomerase catalyzes the ketonization of dienols, which after further processing become intermediates in the Krebs cycle. The enzyme uses a general acid–base mechanism for proton transfer: the amino‐terminal proline has been shown to function as the catalytic base and Arg39 has been implicated as the catalytic acid. We report the results of molecular docking simulations of 4‐oxalocrotonate tautomerase with two substrates, 2‐hydroxymuconate and 5‐carboxymethyl‐2‐hydroxymuconate. pK_a calculations are also performed for the free enzyme. The predicted binding mode of 2‐hydroxymuconate is in agreement with experimental data. A model for the binding mode of 5‐carboxymethyl‐2‐hydroxymuconate is proposed which explains the lower catalytic efficiency of the enzyme toward this substrate. The pK_a predictions and docking simulations support residue Arg39 as the general acid for the enzyme catalysis. © 1999 John Wiley & Sons, Inc. Biopoly 50: 319–328, 1999     

Revisiting the structural flexibility of the complex p21(ras)-GTP: the catalytic conformation of the molecular switch II.

The hydrolysis of GTP in p21(ras) triggers conformational changes that regulate the ras/ERK signaling pathway. An important active site residue is Gln61, which has been found to be mutated in 30% of human tumors. The dynamics of the active site conformation is studied by using molecular dynamics simulation of two independent structures of the GTP-bound uncomplexed enzyme. Two distinct conformations of the enzyme are observed, in which the side-chain residue Gln61 is in different orientations. Essential dynamics analysis is used to describe the essential motions in the transition between the two conformations. Results are compared with earlier simulations of p21(ras) and its complex with GTPase activating protein p21-GAP.     

NMR spectroscopic and molecular modeling studies of protein-carbohydrate and protein-peptide interactions

Investigations of the conformations of carbohydrates, their analogues and their molecular mimics are described, with emphasis on structural and functional information that can be gained by NMR spectroscopic techniques in combination with molecular modeling. The transferred nuclear Overhauser effect (trNOE) has been employed to determine the bound conformations of carbohydrates and other bioactive molecules in complex with protein receptors. The corresponding experiments in the rotating frame (trROE) and selective editing experiments (e.g., QUIET-NOESY) are used to eliminate indirect cross-relaxation pathways (spin diffusion), thereby minimizing errors in the data used for calculation of conformations. Saturation transfer difference NMR experiments reveal detailed information about intermolecular contacts between ligand and protein. Computational techniques are integrated with NMR-derived information to construct structural models of these bioactive molecules and of their complexes with proteins. Recent investigations into the nature of molecular mimicry with regard to protein-ligand interactions are described, along with applications in determining the mode of action of enzyme inhibitors. The results are relevant for the design of the next generation of drug and vaccine candidates.     

Hepatitis C virus RNA-dependent RNA polymerase: Study on the inhibition mechanism by pyrogallol derivatives

Pyrogallol reversibly and noncompetitively inhibits the activity of the hepatitis C RNA-dependent RNA polymerase. Based on molecular modeling of the inhibitor binding in the active site of the enzyme, the inhibition was suggested to be realized via chelation of two magnesium cations involved in the catalysis at the stage of the phosphoryl residue transfer. The proposed model allowed us to purposefully synthesize new derivatives with higher inhibitory capacity.     

The catalytic mechanism of Drosophila alcohol dehydrogenase: evidence for a proton relay modulated by the coupled ionization of the active site Lysine/Tyrosine pair and a NAD+ ribose OH switch

The ionization properties of the active site residues in Drosophila lebanonensis alcohol dehydrogenase (DADH) were investigated theoretically by using an approach developed to account for multiple locations of the hydrogen atoms of the titratable and polar groups. The electrostatic calculations show that (a) the protonation/deprotonation transition of the binary complex of DADH is related to the coupled ionization of Tyr151 and Lys155 in the active site and (b) the pH dependence of the proton abstraction is correlated with a reorganization of the hydrogen bond network in the active site. On this basis, a proton relay mechanism for substrate dehydrogenation is proposed in which the O2' ribose hydroxyl group from the coenzyme has a key role and acts as a switch. The proton relay chain includes the active site catalytic residues, as well as a chain of eight water molecules that connects the active site with the bulk solvent.     

The molecular mechanism of acylation stimulating protein regulation of adipophilin and perilipin expression: involvement of phosphoinositide 3-kinase and phospholipase C

The novel adipokine acylation stimulating protein (ASP) is involved in lipid metabolism and obesity‐related disorders. Adipophilin and perilipin, two members of the lipid droplet protein family, participate not only in fat storage within adipocytes, but also in ectopic lipid deposition in the form of cytoplasmic triglyceride (TG) droplets within many types of mammalian cells. During differentiation to mature adipocytes, mechanisms controlling the synthesis and turnover of these lipid droplet proteins are only partially understood, the mechanisms regulating gene/protein expression as yet unidentified. In our previous study, ASP has been shown to regulate adipophilin and perilipin expression to facilitate TG synthesis during 3T3‐L1 cell differentiation. Our aim in this study was to provide insight into the physiological importance of phosphoinositide 3‐kinase (PI3K) and phospholipase C (PLC) in ASP‐triggered alteration of adipophilin and perilipin expression. We found that acute (2.5 h) inhibition of PLC or PI3K results in a decrease in mRNA and protein of perilipin and adipophilin at any time during differentiation. The fact that there is such a rapid change even with mRNA levels suggests a rapid turnover of both mRNA and protein independent of a direct ASP effect. Also, the presence of these inhibitors blocked the ASP stimulatory effects with a maximal decrease in gene and protein expression of adipophilin (?45% and ?60%, respectively, P < 0.01) and perilipin (?96% and ?63%, respectively, P < 0.01 and P < 0.05). These findings provide further understanding of the adipogenic properties of ASP in adipocytes. J. Cell. Biochem. 112: 1622–1629, 2011. © 2011 Wiley‐Liss, Inc.     

Specific molecular alterations in the norpA-encoded phospholipase C of Drosophila and their effects on electrophysiological responses in vivo

A large number of mutants in the norpA gene, which encodes the phospholipase C (PLC) involved in Drosophila phototransduction, is available for the investigation of the effects of specific amino acid substitutions in PLC on biochemical and electrophysiological properties of these mutants. Of the 47 norpA mutants screened for PLC protein content, all but one (H43) displayed drastically decreased amounts of the protein suggesting that almost any mutational alteration has a deleterious effect on the integrity of the protein. Three new amino acids were identified in the catalytic domains X and Y that are important for PLC catalytic activity and the generation of photoreceptor responses (ERG). One of them was found substituted in H43, which showed a low specific PLC activity, a pronounced decrease in ERG sensitivity, and a wild-type-like response termination time. The response termination times obtained from three mutants was found to be approximately inversely proportional to the amount of PLC. In addition, we show that (i) the specific PLC activity is a key factor determining the photoreceptor sensitivity; (ii) the catalytic activity and response termination are separable functions of PLC; and (iii) a mutation in the putative G alpha-interacting C2 domain causes a preferentially strong defect in latency.     

Conformational analysis of the nonapeptide leuprorelin using NMR and molecular modeling

Summary The nonapeptide Leuprorelin, one of the LHRH agonists, was studied by means of 2D nuclear magnetic resonance spectroscopy and molecular modeling. NOESY spectra in aqueous/deuterated methanol solution (50% H₂O/CD₃OD) at low temperature (268 K) revealed short-range nOe connectivities (i, i+1), characteristic of flexibility of the molecule. The H ^N -H ^N sequential connectivities observed provide evidence that the sequence has the propensity to form a bend involving residues 5 and 6 and the N-terminal segment. The α-proton chemical shifts compared to random coil and additional data from the amide proton temperature coefficients support this assumption. One long-range nOe cross peak between H ₂ ^α -H ^NEth is indicative of proximity between C- and N-termini.     

Conformational analysis of the nonapeptide leuprorelin using NMR and molecular modeling

The nonapeptide Leuprorelin, one of the LHRH agonists, was studied by means of 2D nuclear magnetic resonance spectroscopy and molecular modeling. NOESY spectra in aqueous/deuterated methanol solution (50%H₂O/CD₃OD) at low temperature (268 K) revealed short-range nOe connectivities (i, i+1), characteristic of flexibility of the molecule. The H^N–H^N sequential connectivities observed provide evidence that the sequence has the propensity to form a bend involving residues 5 and 6 and the N-terminal segment. The -proton chemical shifts compared to random coil and additional data from the amide proton temperature coefficients support this assumption. One long-range nOe cross peak between H₂ –H^NEth is indicative of proximity between C- and N-termini.     

Gating and regulation of the calcium release-activated calcium channel: Recent progress from experiments and molecular modeling

Calcium release-activated calcium (CRAC) channels are highly calcium ion (Ca²⁺)-selective channels in the plasma membrane. The transient drop of endoplasmic reticulum Ca²⁺ level activates its calcium sensor stromal interaction molecule (STIM) and then triggers the gating of the CRAC channel pore unit Orai. This process involves a variety of activities of the immune system. Therefore, understanding how the activation and regulation of the CRAC channel can be accomplished is essential. Here we briefly summarize the recent progress on Orai gating and its regulation by 2-aminoethoxydiphenylborate (2-APB) obtained from structural biology studies, biochemical and electrophysiological measurements, as well as molecular modeling. Indeed, integration between experiments and computations has further deepened our understanding of the channel gating and regulation.     

Homology modeling and molecular dynamics study on Schwanniomyces occidentalis alpha-amylase

With consumers growing increasingly aware of environmental issues, industries find enzymes as a reasonable alternative over physical conditions and chemical catalysts. Amylases are important hydrolase enzymes, which have been widely used in variety of industrial process such as pharmaceutical, food, and fermentation industries. Among amylases α-Amylase is in maximum demand due to its wide range of applications. The homology modeling study on Schwanniomyces occidentalis amylase (AMY1, UniProt identifier number: P19269) was performed by Modeller using Aspergillus oryzae (6TAA) as the template. The resulting structure was analyzed for validity and subjected to 14 ns of molecular dynamics (MD) simulation trough GROMACS. The validity of obtained model may represent that utilized OPLS force field is suitable for calcium-containing enzymes. DSSP secondary structure and contact map analysis represent the conservation of domain A TIM barrel feature together with calcium ion coordination sphere. Investigating the covariance matrix followed by principle component analyses for the first five eigenvectors of both trajectories indicate a little more flexibility for AMY1 structure. The electrostatic calculation for the final structures shows similar isoelectric point and superimposed buffering zone in the 5–8 pH range.     

Pharmacophore-directed Homology Modeling and Molecular Dynamics Simulation of G Protein-coupled Receptor： Study of Possible Binding Modes of 5-HT2c Receptor Agonists

A new pharmacophore-based modeling procedure, including homology modeling, pharmacophore study, flexible molecular docking, and long-time molecular dynamics (MD) simulations, was employed to construct the structure of the human 5-HT_(2C) receptor and determine the characteristics of binding modes of 5-HT_(2C) receptor agonists. An agonist-receptor complex has been constructed based on homology modeling and a pharmacophore hypothesis model based on some high active compounds. Then MD simulations of the ligand-receptor complex in an explicit membrane environment were carried out. The conformation of the 5- HT_(2C) receptor during MD simulation was explored, and the stable binding modes of the studied agonist were determined. Flexible molecular docking of several structurally diverse agonists of the human 5-HT_(2C) receptor was carried out, and the general binding modes of these agonists were investigated. According to the models presented in this work and the results of Flexi-Dock, the involvement of the amino acid residues Asp134, Ser138, Ash210, Asn331, Tyr358, Ile131, Ser132, Val135, Thr139, Ile189, Val202, Val208, Leu209, Phe214, Val215, Gly218, Ser219, Phe223, Trp324, Phe327, and Phe328 in agonist recognition was studied. The obtained binding modes of the human 5-HT_(2C) receptor agonists have good agreement with the site-directed mutagenesis data and other studies.