Evaluating the stability of disulfide bridges in proteins: a torsional potential energy surface for diethyl disulfide.

Disulfide bonds formed by the oxidation of cysteine residues in proteins are the major form of intra- and inter-molecular covalent linkages in the polypeptide chain. To better understand the conformational energetics of this linkage, we have used the MP2(full)/6-31G(d) method to generate a full potential energy surface (PES) for the torsion of the model compound diethyl disulfide (DEDS) around its three critical dihedral angles (χ2, χ3, χ2'). The use of ten degree increments for each of the parameters resulted in a continuous, fine-grained surface. This allowed us to accurately predict the relative stabilities of disulfide bonds in high resolution structures from the Protein Data Bank. The MP2(full) surface showed significant qualitative differences from the PES calculated using the Amber force field. In particular, a different ordering was seen for the relative energies of the local minima. Thus, Amber energies are not reliable for comparison of the relative stabilities of disulfide bonds. Surprisingly, the surface did not show a minimum associated with χ2 ∼ - 60°, χ3 ∼ 90, χ2' ∼ - 60°. This is due to steric interference between Hα atoms. Despite this, significant populations of disulfides were found to adopt this conformation. In most cases this conformation is associated with an unusual secondary structure motif, the cross-strand disulfide. The relative instability of cross-strand disulfides is of great interest, as they have the potential to act as functional switches in redox processes.

31P MAS-NMR of human erythrocytes: independence of cell volume from angular velocity.

31P magic angle spinning NMR (MAS-NMR) spectra were obtained from suspensions of human red blood cells (RBCs) that contained the cell-volume-sensitive probe molecule, dimethyl methylphosphonate (DMMP). A mathematical representation of the spectral-peak shape, including the separation and width-at-half-height in the 31P NMR spectra, as a function of rotor speed, enabled us to explore the extent to which a change in cell volume would be reflected in the spectra if it occurred. We concluded that a fractional volume change in excess of 3% would have been detected by our experiments. Thus, the experiments indicated that the mean cell volume did not change by this amount even at the highest spinning rate of 7 kHz. The mean cell volume and intracellular 31P line-width were independent of the packing density of the cells and of the initial cell volume. The relationship of these conclusions to other non-NMR studies of pressure effects on cells is noted.

CO(2) fixation by Rubisco: computational dissection of the key steps of carboxylation, hydration, and C-C bond cleavage.

Despite intensive experimental and computational studies, some important features of the mechanism of the photosynthetic CO(2)-fixing enzyme, Rubisco, are still not understood. To complement our previous investigation of the first catalytic step, the enolization of D-ribulose-1,5-bisphosphate (King et al., Biochemistry 1998, 44, 15414-15422), we present the first complete computational dissection of subsequent steps of the carboxylation reaction that includes the roles of the central magnesium ion and modeled residues of the active site. We investigated carboxylation, hydration, and C-C bond cleavage using the density functional method and the B3LYP/6-31G(d) level to perform geometry optimizations. The energies were determined by B3LYP/6-311+G(2d,p) single-point calculations. We modeled a fragment of the active site and substrate, taking into account experimental findings that the residues coordinated to the Mg ion, especially the carbamylated Lys-201, play critical roles in this reaction sequence. The carbamate appears to act as a general base, not only for enolization but also for hydration of the beta ketoacid formed by addition of CO(2) and, as well, cleavage of the C2-C3 bond of the hydrate. We show that CO(2) is added directly, without assistance of a Michaelis complex, and that hydration of the resultant beta ketoacid occurs in a separate subsequent step with a discrete transition state. We suggest that two conformations of the hydrate (gem-diol), with different metal coordination, are possible. The step with the highest activation energy during the carboxylation cycle is the C-C bond cleavage. Depending on the conformations of the gem-diol, different pathways are possible for this step. In either case, special arrangements of the metal coordination result in bond breaking occurring at remarkably low activation energies (between 28 and 37 kcal mol(-1)) which might be reduced further in the enzyme environment.

Energetically most likely substrate and active-site protonation sites and pathways in the catalytic mechanism of dihydrofolate reductase.

Despite much experimental and computational study, key aspects of the mechanism of reduction of dihydrofolate (DHF) by dihydrofolate reductase (DHFR) remain unresolved, while the secondary DHFR-catalyzed reduction of folate has been little studied. Major differences between proposed DHF mechanisms are whether the carboxylate group of the conserved active-site Asp or Glu residue is protonated or ionized during the reaction, and whether there is direct protonation of N5 or a proton shuttle from an initially protonated carboxylate group via O4. We have addressed these questions for both reduction steps with a comprehensive set of ab initio quantum chemical calculations on active-site fragment complexes, including the carboxyl side chain and, progressively, all other polar active-site residue groups including conserved water molecules. Addition of two protons in two steps was considered. The polarization effects of the remainder of the enzyme system were approximated by a dielectric continuum self-consistent reaction field (SCRF) model using an effective dielectric constant (epsilon) of 2. Optimized geometries were calculated using the density functional (B3LYP) method and Onsager SCRF model with the 6-31G basis. Single-point energy calculations were then carried out at the B3LYP/6-311+G level with either the Onsager or dielectric polarizable continuum model. Additional checking calculations at MP2 and HF levels, or with other basis sets or values of epsilon, were also done. From the results, the conserved water molecule, corresponding to W206 in the E. coli DHFR complexes, that is H-bonded to both the OD2 oxygen atom of the carboxyl (Asp) side chain and O4 of the pterin/dihydropterin ring, appears critically important and may determine the protonation site for the enzyme-bound substrates. In the absence of W206, the most stable monoprotonated species are the neutral-pair 4-enol forms of substrates with the carboxyl group OD2 oxygen protonated and H-bonded to N3. If W206 is included, then the most stable forms are still the neutral-pair complexes but now for the N3-H keto forms with the protonated OD2 atom H-bonding with W206. A second proton addition to these complexes gives protonations at N8 (folate) or N5 (DHF). Calculated H-bond distances correlate well with those for the conserved W206 observed in many X-ray structures. For all structures with occluded M20 loop conformations (closed active site), OD2-N3 distances are less than OD2-NA2 distances, which is consistent with those calculated for protonated OD2 complexes. Thus, the results (B3LYP; epsilon = 2 calculations) support a mechanism for both folate and DHF reduction in which the OD2 carboxyl oxygen is first protonated, followed by a direct protonation at N8 (folate) and N5 (DHF) to obtain the active cation complexes, i.e., doubly protonated. The results do not support a proposed protonated carboxyl with DHF in the enol form for the Michaelis complex, nor an ionized carboxyl with protonated enol-DHF as a catalytic intermediate. However, as additional calculations for the monoprotonated complete complexes show a reduction in the energy differences between the neutral-pair keto and ion-pair keto (N8- or N5-protonated) forms, we are extending the treatment using combined quantum mechanics and molecular mechanics (QM/MM) and molecular dynamics simulation methods to refine the description of the protein/solvent environment and prediction of the relative stabilization free energies of the various (OD2, O4, N5, and N8) protonation sites.

Identification of active-site residues of the pro-metastatic endoglycosidase heparanase.

Heparanase is a beta-D-endoglucuronidase that cleaves heparan sulfate (HS) and has been implicated in many important physiological and pathological processes, including tumor cell metastasis, angiogenesis, and leukocyte migration. We report herein the identification of active-site residues of human heparanase. Using PSI-BLAST and PHI-BLAST searches of sequence databases, similarities were identified between heparanase and members of several of the glycosyl hydrolase families (10, 39, and 51) from glycosyl hydrolase clan A (GH-A), including strong local identities to regions containing the critical active-site catalytic proton donor and nucleophile residues that are conserved in this clan of enzymes. Furthermore, secondary structure predictions suggested that heparanase is likely to contain an (alpha/beta)(8) TIM-barrel fold, which is common to the GH-A families. On the basis of sequence alignments with a number of glycosyl hydrolases from GH-A, Glu(225) and Glu(343) of human heparanase were identified as the likely proton donor and nucleophile residues, respectively. The substitution of these residues with alanine and the subsequent expression of the mutant heparanases in COS-7 cells demonstrated that the HS-degrading capacity of both was abolished. In contrast, the alanine substitution of two other glutamic acid residues (Glu(378) and Glu(396)), both predicted to be outside the active site, did not affect heparanase activity. These data suggest that heparanase is a member of the clan A glycosyl hydrolases and has a common catalytic mechanism that involves two conserved acidic residues, a putative proton donor at Glu(225) and a nucleophile at Glu(343).

Molecular dynamics simulation of human prion protein including both N-linked oligosaccharides and the GPI anchor.

Although glycosylation appears to protect prion protein (PrP(C)) from the conformational transition to the disease-associated scrapie form (PrP(Sc)), available NMR structures are for non-glycosylated PrP(C), only. To investigate the influence of both the two N-linked glycans, Asn181 and Asn197, and of the GPI anchor attached to Ser230, on the structural, dynamical and electrostatic behavior of PrP, we have undertaken molecular dynamics simulations on the C-terminal region of human prion protein HU:PrP(90-230), with and without the three glycans. The simulations used the AMBER94 force field in a periodic box model with explicit water molecules, considering all long-range electrostatic interactions. The results suggest the structured part of the protein, HU:PrP(127-227) is stabilized overall from addition of the glycans, specifically by extensions of Helix-B and Helix-C and reduced flexibility of the linking turn containing Asn197, although some regions such as residues in the turn (165-170) between Strand-B and Helix-B have increased flexibility. The stabilization appears indirect, by reducing the mobility of the surrounding water molecules, and not from specific interactions such as H bonds or ion pairs. The results are consistent with glycosylation at Asn197 having a stabilizing role, while that at Asn181, in a region with already stable secondary structure, having a more functional role, in agreement with literature suggestions. Due to three negatively charged SiaLe(x) groups per N-glycan, the surface electrostatic properties change to a negative electrostatic field covering most of the C-terminal part, including the surface of Helix-B and Helix-C, while the positively charged N-terminal part PrP(90-126) of undefined structure creates a positive potential. The unusual hydrophilic Helix-A (144-152) is not covered by either of these dominant electrostatic fields, and modeling shows it could readily dimerize in anti parallel fashion. In combination with separate simulations of the GPI anchor in a membrane model, the results show the GPI anchor is highly flexible and would maintain the protein at a distance between 9 and 13 A from the membrane surface, with little influence on its structure or orientational freedom.

QM/MM and SCRF studies of the ionization state of 8-methylpterin substrate bound to dihydrofolate reductase: existence of a low-barrier hydrogen bond.

Using combined semiempirical quantum mechanics and molecular mechanics (QM/MM) and ab initio self-consistent reaction field (SCRF) calculations, we determined that a low-barrier hydrogen bond (LBHB) is formed when the mechanism-based substrate 8-methylpterin binds to dihydrofolate reductase (DHFR). The substrate initially was assumed bound either in the ion-pair form corresponding to N3-protonated substrate hydrogen (H) bonded to the unprotonated (carboxylate) of the conserved Glu30 residue in the active site, or in the neutral-pair form corresponding to unprotonated substrate H bonded to the neutral (carboxylic acid) from of Glu30. The free energy of interaction of these H-bonded systems with the protein/solvent surroundings was computed using a coordinate-coupled free energy perturbation (FEP) method implemented within the molecular dynamics (MD) simulation scheme and using a semiempirical (PM3) QM/MM force field. The free energy obtained from the QM/MM force-field simulations corresponds most closely with the corresponding free energy component obtained from HF/6-31G* SCRF calculations using a value of 2 for the dielectric constant (epsilon) for the solvated protein. Calculations were performed at levels ranging from HF/6-31G to MP2/6-31G* to B3LYP/6-31 + G**, with varying dielectric constants. The energy-minimized path for motion of the proton in the H bond along a one-dimensional reaction coordinate was calculated at HF/6-31G, HF/6-31G* (epsilon = 1) and B3LYP/6-31G* (epsilon = 2) levels. These calculations identified a second neutral-pair complex, involving the 2-amino group of substrate, which also interacts with Glu30, which is lower in energy than the ion-pair form. A harmonic vibrational analysis shows that the first vibrational state appears to lie near or above the TS connecting potential energy minima corresponding to the two neutral-pair configurations, thus indicating an LBHB. Consequently, the H-bonded system will have a significant probability of being found in the ion-pair form, in agreement with experimental spectral studies indicating an enzyme-bound cation and suggesting that the LBHB would activate substrate towards hydride-ion transfer from NADPH.

Acute oxygen supplementation restores markers of hepatocyte energy status and hypoxia in cirrhotic rats.

The oxygen limitation hypothesis states that hepatocyte hypoxia is the mechanism determining metabolic restriction in the cirrhotic liver. Therefore we studied markers of hepatocyte energy state and cellular hypoxia in livers of normal and cirrhotic rats before and after oxygen supplementation. Rats with carbon tetrachloride-induced cirrhosis and procedural control rats were exposed to either room air or a hyperoxic gas mixture for 1 h immediately before freeze clamping and perchloric acid extraction of liver tissue. Extracts were assessed by (31)P NMR and enzymatic assays. Livers from cirrhotic rats breathing room air showed a reduced ratio of ATP/ADP, an increased ratio of inorganic phosphate/ATP, and a trend toward an increased ratio of lactate/pyruvate compared with procedural control livers (ATP/ADP 1.73 +/- 0.35 versus 2.68 +/- 0.61, P <.05; P(i)/ATP 2.74 +/- 0.48 versus 1.56 +/- 0.26, P <.05; lactate/pyruvate 29.3 +/- 6.4 versus 22.5 +/- 7.4, P =.18). After supplementation with oxygen for 1 h, these ratios in cirrhotic livers approached control values. A variety of other metabolic markers affected by cirrhosis showed variable trends toward normal in response to oxygen supplementation, whereas minor trends toward an increase in ATP levels in control animals suggest the possibility of marginal oxygen limitation in normal livers. The data are consistent with the hypothesis that hepatocytes in cirrhotic livers have normal metabolic capacity but are constrained by a deficit in oxygen supply. Interventions aimed at increasing oxygen supply to the liver may have both short- and long-term therapeutic value in the management of cirrhosis.

Enzyme polarization of substrates of dihydrofolate reductase by different theoretical methods.

We have investigated the importance of polarization by the enzyme dihydrofolate reductase (DHFR) on its substrates, folate and dihydrofolate, using a series of quantum mechanical (QM) techniques (Hartree-Fock (HF), Møller-Plesset second-order perturbation theory (MP2), local density approximation (LDA) and generalized gradient approximation (GGA) density functional theory (DFT) calculations) in which the bulk enzyme is included in the calculations as point charges. Polarization, in terms of both charges on components (residues) of the folate and dihydrofolate molecules and changes in the electron density, particularly of the pterin ring of the substrates, and the implications for the catalytic reduction are discussed. Significant differences in polarization behavior are observed for the different theoretical methods employed. The consequences of this, particularly for choosing an appropriate model for quantum mechanical/molecular mechanical (QM/MM) calculations, are pointed out. The HF and MP2 QM methods show small polarizations (approximately 0.04 electrons) of the pterin ring but quite large polarizations with both LDA and GGA DFT methods (0.3-0.5 electrons). This large difference in polarization for both folate and dihydrofolate arises as a result of substantial differences between the charge distributions for the gasphase DFT and HF calculations, specifically the charges on the dianionic glutamate side chain. Some recent literature reports of incorrect representation of anionic systems by DFT methods are noted. The DFT results are similar to the previously reported LDA DFT results of Bajorath et al. predicting a large polarization of the pterin ring of folate (Proteins 9:217-224, 1991) and dihydrofolate (PNAS 88:6423-6426, 1991) of approximately 0.5-0.6 electrons.

31P and 1H NMR spectroscopic studies of liver extracts of carbon tetrachloride-treated rats.

NMR spectroscopy was used to examine hepatic metabolism in cirrhosis with a particular focus on markers of functional cellular hypoxia. (31)P and (1)H NMR spectra were obtained from liver extracts from control rats and from rats with carbon tetrachloride-induced cirrhosis. A decrease of 34% in total phosphorus content was observed in cirrhotic rats, parallelling a reduction of 40% in hepatocyte mass as determined by morphometric analysis. Hypoxia appeared to be present in cirrhotic rats, as evidenced by increased inorganic phosphate levels, decreased ATP levels, decreased ATP:ADP ratios (1.72 +/- 0.40 vs 2.48 +/- 0.50, p < 0.01), and increased inorganic phosphate:ATP ratios (2.77 +/- 0.48 vs 1.62 +/- 0.24, p < 0.00001). When expressed as a percentage of the total phosphorus content, higher levels of phosphoethanolamine and lower levels of NAD and glycerophosphoethanolamine were detected in cirrhotic rats. Cirrhotic rats also had increased phosphomonoester:phosphodiester ratios (5.73 +/- 2.88 vs 2.53 +/- 0.52, p < 0.01). These findings are indicative of extensive changes in cellular metabolism in the cirrhotic liver, with many findings attributable to the presence of intracellular hypoxia.

Molecular dynamics simulations of human prion protein: importance of correct treatment of electrostatic interactions.

Molecular dynamics simulations have been used to investigate the dynamical and structural behavior of a homology model of human prion protein HuPrP(90-230) generated from the NMR structure of the Syrian hamster prion protein ShPrP(90-231) and of ShPrP(<90-231) itself. These PrPs have a large number of charged residues on the protein surface. At the simulation pH 7, HuPrP(90-230) has a net charge of -1 eu from 15 positively and 14 negatively charged residues. Simulations for both PrPs, using the AMBER94 force field in a periodic box model with explicit water molecules, showed high sensitivity to the correct treatment of the electrostatic interactions. Highly unstable behavior of the structured region of the PrPs (127-230) was found using the truncation method, and stable trajectories could be achieved only by including all the long-range electrostatic interactions using the particle mesh Ewald (PME) method. The instability using the truncation method could not be reduced by adding sodium and chloride ions nor by replacing some of the sodium ions with calcium ions. The PME simulations showed, in accordance with NMR experiments with ShPrP and mouse PrP, a flexibly disordered N-terminal part, PrP(90-126), and a structured C-terminal part, PrP(127-230), which includes three alpha-helices and a short antiparallel beta-strand. The simulations showed some tendency for the highly conserved hydrophobic segment PrP(112-131) to adopt an alpha-helical conformation and for helix C to split at residues 212-213, a known disease-associated mutation site (Q212P). Three highly occupied salt bridges could be identified (E146/D144<-->R208, R164<-->D178, and R156<-->E196) which appear to be important for the stability of PrP by linking the stable main structured core (helices B and C) with the more flexible structured part (helix A and strands A and B). Two of these salt bridges involve disease-associated mutations (R208H and D178N). Decreased PrP stability shown by protein unfolding experiments on mutants of these residues and guanidinium chloride or temperature-induced unfolding studies indicating reduced stability at low pH are consistent with stabilization by salt bridges. The fact that electrostatic interactions, in general, and salt bridges, in particular, appear to play an important role in PrP stability has implications for PrP structure and stability at different pHs it may encounter physiologically during normal or abnormal recycling from the pH neutral membrane surface into endosomes or lysomes (acidic pHs) or in NMR experiments (5.2 for ShPrP and 4.5 for mouse PrP).

Quantum chemical analysis of the enolization of ribulose bisphosphate: the first hurdle in the fixation of CO2 by Rubisco.

A study, using ab initio quantum chemical methods, of the first step in the reaction mechanism of Rubisco, the enolization of the substrate, ribulose bisphosphate, is reported. This is the first such study that takes into account the likely roles of critical features within the active site. On the basis of molecular dynamics relaxation of the complex between activated enzyme and substrate using X-ray crystallographic structures as starting coordinates, a 29-atom fragment that mimicked the active site was constructed. States along a proposed reaction pathway were calculated using density functional theory and Moller-Plesset second-order perturbation theory. The results are consistent with the postulate that the base that abstracts the C3 proton of ribulose bisphosphate is the metal-stabilized carbamate of Lys-201 formed during the activation process. The calculations suggest that the active-site residue, Lys-175, is charged before enolization commences and they indicate a possible means by which the enzyme directs the incoming CO2 to attack the C2 carbon atom of the enediol, rather than the chemically very similar C3 atom.

Identification and energetic ranking of possible docking sites for pterin on dihydrofolate reductase.

The reliability of new methodology for detecting sites for ligand binding on the surfaces of proteins has been tested using a range of dihydrofolate reductase (DHFR) crystal structures. Docking of the pterin molecule to ten such DHFR structures has been examined. Initial docking sites were selected using the VDW-FFT method we have developed recently. This procedure was followed by rigid geometry optimization and solvation energy calculations using our parametrized reaction field multipoles (PRFM) method and the finite difference solution of the Poisson equation (FDPB) method. Two different sets of MM parameters, from the OPLS and Amber94 force fields, have been used. In eight cases the energy of the complexes with pterin bound at the active site was the lowest with the recent Amber94 parameters. In one case the spurious first-ranked site was only 1.8 kcal/mol lower in energy compared with the active site. The other 'failure' of the method may, in fact, represent a valid initial binding site. The calculations with the old OPLS parameters gave slightly worse results.

Predicted structure of the extracellular region of ligand-gated ion-channel receptors shows SH2-like and SH3-like domains forming the ligand-binding site.

Fast synaptic neurotransmission is mediated by ligand-gated ion-channel (LGIC) receptors, which include receptors for acetylcholine, serotonin, GABA, glycine, and glutamate. LGICs are pentamers with extracellular ligand-binding domains and form integral membrane ion channels that are selective for cations (acetylcholine and serotonin 5HT3 receptors) or anions (GABAA and glycine receptors and the invertebrate glutamate-binding chloride channel). They form a protein superfamily with no sequence similarity to any protein of known structure. Using a 1D-3D structure mapping approach, we have modeled the extracellular ligand-binding domain based on a significant match with the SH2 and SH3 domains of the biotin repressor structure. Refinement of the model based on knowledge of the large family of SH2 and SH3 structures, sequence alignments, and use of structure templates for loop building, allows the prediction of both monomer and pentamer models. These are consistent with medium-resolution electron microscopy structures and with experimental structure/function data from ligand-binding, antibody-binding, mutagenesis, protein-labeling and subunit-linking studies, and glycosylation sites. Also, the predicted polarity of the channel pore calculated from electrostatic potential maps of pentamer models of superfamily members is consistent with known ion selectivities. Using the glycine receptor alpha 1 subunit, which forms homopentamers, the monomeric and pentameric models define the agonist and antagonist (strychnine) binding sites to a deep crevice formed by an extended loop, which includes the invariant disulfide bridge, between the SH2 and SH3 domains. A detailed binding site for strychnine is reported that is in strong agreement with known structure/function data. A site for interaction of the extracellular ligand-binding domain with the activation of the M2 transmembrane helix is also suggested.

Molecular dynamics simulations of the docking of substituted N5-deazapterins to dihydrofolate reductase.

Orientations of the deazapterin ring and the conformational preferences of groups appended to the deazapterin ring in a set of 8-substituted deazapterin cations docked into the dihydrofolate reductase (DHFR) binding site have been investigated using a methodology based on the simulated annealing technique within molecular dynamics (MD) simulations. Of five possible binding pockets for the 8-substituents, identified from a preliminary manual docking study, one has been definitively eliminated after an analysis of MD trajectories, while another remains uncertain. Using a new method based on standard thermodynamic cycles and a linear approximation of polar and non-polar free energy contributions from MD averages, binding affinities of the different ligands in each binding site have been correlated with experimental dissociation constants. The study has provided insights into structure-activity relationships for use in the design of modified inhibitors of DHFR.

Structure-activity relationships for the 8-alkylpterins: a new class of mechanism-based substrates for dihydrofolate reductase (DHFR).

The substrate activity with both human and chicken dihydrofolate reductases (DHFR) has been examined for a series of 8-alkylpterins, 6-methyl-8-alkylpterins, and 7-methyl-8-propylpterin. All the 8-alkylpterins exhibited substrate activity with Vmax/[E]o values ranging from 1.0 to 5.4 and 2.6 to 14.8 s-1 for chicken and human DHFRs, respectively, with activity varying in the order 8-methyl > 8-allyl > 8-isopropyl > or = 8-ethyl > or = 8-propyl for both enzymes. Km values were found to range from 6.2 to 47 and 14 to 261 microM for chicken and human DHFRs, respectively, with the strength of binding varying in the order 8-propyl > 8-isopropyl > 8-ally > 8-methyl > 8-ethyl for both enzymes. Addition of a 6-methyl substituent affected the activity of the 8-alkylpterins significantly. While 6,8-dimethylpterin was a much better substrate than 8-methylpterin, the 6-methyl-8-propyl, 6-methyl-8-allyl, and 6-methyl-8-isopropyl compounds showed no substrate activity and 6-methyl-8-ethylpterin showed very weak activity with chicken enzyme only. 7-Methyl-8-propylpterin showed no substrate activity. Thermodynamic dissociation constants (Kd) for the compounds in binary complex with both human and chicken DHFRs ranged from 23 to 351 and 15 to 127 microM, respectively. Trends for the KdS were consistent with the kinetic data in suggesting stronger binding for 8-alkylpterins with larger 8-substituents. Comparison of Kd values with corresponding Km values suggested both strong cooperativity (6,8-dimethylpterin) and antagonism (6-methyl-8-isopropylpterin) with NADPH in binding to DHFR. Kd values of 20 and 10 microM for the ternary complexes of 7-methyl-8-propylpterin with human and chicken enzyme, respectively, suggest modest inhibitory activity. Application of molecular graphics modeling of ligands in the DHFR binding site has provided insight in interpreting the structure-activity relationships. The finding that different binding orientations are possible for ligands with small (8-methyl) or larger (e.g., 8-propyl) 8-substituents helps to explain the 6-methyl substituent effect and the transition from weak binding and high activity to tight binding and low activity as a function of ring-substituent pattern.

pH dependence of enzyme reaction rates and deuterium isotope effects on the reduction of a new mechanism-based substrate by dihydrofolate reductase DHFR.

The enzyme kinetics of the reduction of the substrate 6,8-dimethylpterin by chicken and recombinant human dihydrofolate reductases (DHFRs) have been examined over the pH range 5.0-8.0 in the presence of NADPH or (4R)-[2H]NADPH (NADPD). The pH profiles of the catalytic constant (Vmax/[E]o or kcat) showed pH independence for chicken DHFR and little pH dependence for human DHFR. For both DHFRs, the pH profiles of the Michaelis constant (Km(substrate)) and the apparent second-order rate constant (Vmax/Km(substrate)[E]o or kcat/Km(substrate)) indicated that two ionizable groups, deduced to be the substrate and the conserved Glu carboxy group in the enzyme active site, should be ionized in their cationic and anionic forms, respectively, for formation of the enzyme-substrate complex and for catalysis. The pKa values of about 5.3 and 6.5 which were obtained from the pH profiles of Km(substrate) and kcat/Km(substrate) were assigned to the ionizations of the substrate and the enzyme carboxy group, respectively. Deuterium isotope effects on DV and d(V/K) were significant for both enzymes, approximately 3 for chicken DHFR and approximately 4 for recombinant human DHFR, and were pH independent. Thus, the rate-limiting step in the enzymic reduction of 6,8-dimethylpterin is hydride-ion transfer at acidic pHs as well as neutral pHs. The results demonstrate that, compared with dihydrofolate, 6,8-dimethylpterin is a superior substrate for mechanistic investigations as it allows direct study of the effects of both enzyme and substrate ionizations involved in the catalysis and also avoids the obscuration of the catalytic rate by product release.

Structure-activity relationships and pH dependence of binding of 8-alkyl-N5-deazapterins to dihydrofolate reductase.

Thermodynamic dissociation constants (Kd) have been determined for two series of 8-alkyl-N5-deazapterins in binary complexes with human and chicken dihydrofolate reductases (DHFRs) and ternary complexes with the enzyme.NADPH complex. For an initial series of 12 compounds with variable 8-alkyl substitutents and pyrazine ring-methyl substitution patterns, Kd values at pH 6.6 were found to range from > 100 to 0.5 microM, with consistent trends depending on the enzyme source, the size of the 8-substituent, and the presence and position of the pyrazine ring-methyl substituent. For most compounds in this first series, Kd values were significantly lower for the ternary complex than for the binary complex with ratios of Kd(binary)/Kd(ternary) ranging from 0.6 to 62, suggesting a degree of cooperativity in binding to the enzyme between ligand and cofactor. This effect was more pronounced for the human enzyme. The structure-activity relationships developed in the first series suggested a number of strategies for developing ligands with greater affinity for DHFR. These were tested with a second series of four compounds. The Kd of 80 nM at pH 6.6 of one of these compounds [5-methyl- 8-isobutyl-N5-deazapterin (15)] in ternary complex with human DHFR is more than 200 times lower than that for the lead compound (8-methyl-N5-deazapterin (1); Kd 21 microM). Studies of binding stoichiometry indicated two binding sites in binary complexes with DHFR for 8-alkyl-N5-deazapterins with smaller 8-substituents. The second site was not found in ternary complexes or for ligands with larger 8-substituents, suggesting that the second ligand molecule in binary complexes is probably binding in the cofactor site and that the larger 8-substituents also bind in this area. A detailed study of the inhibition kinetics for one compound, 6,8-dimethyl-N5-deazapterin (5), showed it to be a competitive inhibitor of the chicken DHFR-catalyzed reduction of 6,8-dimethylpterin suggesting that the 8-alkyl-N5-deazapterins bind in the substrate site of DHFR. The pH dependence of the binding of several ligands in binary and ternary complexes with DHFR was examined by determining their Kd values at a range of pH's. This suggested that binding was predominantly between protonated ligand and deprotonated enzyme, but with variable contributions to binding observed between deprotonated enzyme and neutral ligand, and protonated enzyme and protonated ligand, depending on compound and complex type.

A method of preparation and purification of (4R)-deuterated-reduced nicotinamide adenine dinucleotide phosphate.

(4R)-Deuterated-reduced nicotinamide adenine dinucleotide phosphate, (4R)-[2H]NADPH, was prepared by reduction of NADP+ using an NADP(+)-dependent alcohol dehydrogenase (EC 1.1.1.2) from Thermoanaerobium brockii and isopropanol-d8 as substrate at 43 degrees C, pH 9. More than 80% of the product was identified as reduced cofactor by reverse-phase (ODS) HPLC, and a 1H NMR study showed that all of the reduced cofactor was (4R)-deuterated. Less than 10% of the product was oxidized cofactor, the remainder being impurities from the breakdown of the dinucleotide compound. Subsequent purification carried out by semipreparative reverse-phase HPLC with 0.1 M NaCl at pH 8.5 gave a compound of more than 96% purity. Separated (4R)-[2H]NADPH fractions were freeze-dried and the white solid was stored at 5 degrees C with desiccant.

Ionization state and pKa of pterin-analogue ligands bound to dihydrofolate reductase.

The ionization state and pKa of the inhibitor 6,8-dimethyl-N5-deazapterin bound to the recombinant human dihydrofolate reductase (rhH2folate reductase) complex with NADPH was determined by a spectrofluorimetric method. The excitation spectra for bound ligand as a function of pH from 6.1 to 9.7 indicated it was the same cationic form as for unbound ligand, which is protonated on N3. However, the lower limit for the pKa of the bound form was determined to be 9.1, a value about pH 2.5 higher than that for free ligand, indicating that ligand bound to the enzyme is protonated at neutral pH. The excitation spectra for bound ligand as a function of pH were generated by computer simulation by employing corrections for the pH dependence of the proportion of bound ligand (variable Kd; ligand-dissociation constant) and taking account of the different pKa values for bound and unbound forms. A plot of Kd values against pH showed a bell-shaped curve indicating that 6,8-dimethyl-N5-deazapterin bound to rhH2folate reductase.NADPH to form a ternary complex of ionised enzyme with protonated ligand and/or protonated enzyme with unprotonated ligand; the spectrofluorimetric results are consistent with the first alternative.