Conformational selection and induced changes along the catalytic cycle of Escherichia coli dihydrofolate reductase

Protein function often involves changes between different conformations. Central questions are how these conformational changes are coupled to the binding or catalytic processes during which they occur, and how they affect the catalytic rates of enzymes. An important model system is the enzyme dihydrofolate reductase (DHFR) from Escherichia coli, which exhibits characteristic conformational changes of the active‐site loop during the catalytic step and during unbinding of the product. In this article, we present a general kinetic framework that can be used (1) to identify the ordering of events in the coupling of conformational changes, binding, and catalysis and (2) to determine the rates of the substeps of coupled processes from a combined analysis of nuclear magnetic resonance R₂ relaxation dispersion experiments and traditional enzyme kinetics measurements. We apply this framework to E. coli DHFR and find that the conformational change during product unbinding follows a conformational‐selection mechanism, that is, the conformational change occurs predominantly prior to unbinding. The conformational change during the catalytic step, in contrast, is an induced change, that is, the change occurs after the chemical reaction. We propose that the reason for these conformational changes, which are absent in human and other vertebrate DHFRs, is robustness of the catalytic rate against large pH variations and changes to substrate/product concentrations in E. coli. Proteins 2012;. © 2012 Wiley Periodicals, Inc.     

Conformational change of dihydrofolate reductase near the active site after thiol modification: detected by limited proteolysis

Kinetic studies of chicken liver dihydrofolate reductase (CL-DHFR) and Chinese hamster ovary DHFR (CH-DHFR) activated following p-hydroxymercuribenzoate (p-HMB) modification indicate a conformational change at the active site, suggesting a loosening of the enzyme structure upon SH modification. In the present study, limited proteolysis was applied to detect the subtle conformational changes in SH-modified DHFRs. The digested peptide fragments were separated by Tricine SDS-PAGE and sequenced by Edman auto-degradation. The thiol modifier N-iodoacetyl-N'-(5-sulfo-1-nophthyl) ethylenediamine (IAEANS), which activates these DHFRs only weakly, was used as a control. The results of sequencing showed that compared to native enzyme, there is one additional cleavage site near the active site in p-HMB-modified CL-DHFR, two additional sites in p-HMB-modified CH-DHFR, but no additional site for IAEANS-modified DHFRs. These results indicate that activation of DHFRs following thiol modification is accompanied by a conformational change at or near the active site. This subtle change in the active site conformation results in a pronounced change in enzyme activity. This provides further evidence that flexibility at the active site is essential for full expression of enzyme catalytic activity. Comparing results obtained from previous experiments on guanidine- and urea-activated CL-DHFR, this shows that a conformational change near helix(28-39) is sufficient for full activation of DHFR.     

Role of water in the catalytic cycle of E. coli dihydrofolate reductase

Dihydrofolate reductase (DHFR) catalyzes the nicotinamide adenine dinucleotide phosphate (NADPH)-dependent reduction of 7,8-dihydrofolate (H2F) to 5,6,7,8-tetrahydrofolate (H4F). Because of the absence of any ionizable group in the vicinity of N5 of dihydrofolate it has been proposed that N5 could be protonated directly by a water molecule at the active site in the ternary complex of the Escherichia coli enzyme with cofactor and substrate. However, in the X-ray structures representing the Michaelis complex of the E. coli enzyme, a water molecule has never been observed in a position that could allow protonation of N5. In fact, the side chain of Met 20 blocks access to N5. Energy minimization reported here revealed that water could be placed in hydrogen bonding distance of N5 with only minor conformational changes. The r.m.s. deviation between the conformation of the M20 loop observed in the crystal structures of the ternary complexes and the conformation adopted after energy minimization was only 0.79 A. We performed molecular dynamics simulations to determine the accessibility by water of the active site of the Michaelis complex of DHFR. Water could access N5 relatively freely after an equilibration time of approximately 300 psec during which the side chain of Met 20 blocked water access. Protonation of N5 did not increase the accessibility by water. Surprisingly the number of near-attack conformations, in which the distance between the pro-R hydrogen of NADPH and C6 of dihydrofolate was less than 3.5 A and the angle between C4 and the pro-R hydrogen of NADPH and C6 of dihydrofolate was greater than 120 degrees, did not increase after protonation. However, when the hydride was transferred from NADPH to C6 of dihydrofolate before protonation, the side chain of Met 20 moved away from N5 after approximately 100 psec thereby providing water access. The average time during which water was found in hydrogen bonding distance to N5 was significantly increased. These results suggest that hydride transfer might occur early to midway through the reaction followed by protonation. Such a mechanism is supported by the very close contact between C4 of NADP+ and C6 of folate observed in the crystal structures of the ternary enzyme complexes, when the M20 loop is in its closed conformation.     

Functional role for Tyr 31 in the catalytic cycle of chicken dihydrofolate reductase

Despite much work, many key aspects of the mechanism of the dihydrofolate reductase (DHFR) catalyzed reduction of dihydrofolate remain unresolved. In bacterial forms of DHFR both substrate and water access to the active site are controlled by the conformation of the mobile M20 loop. In vertebrate DHFRs only one conformation of the residues corresponding to the M20 loop has been observed. Access to the active site was proposed to be controlled by residue 31. MD simulations of chicken DHFR complexed with substrates and cofactor revealed a closing of the side chain of Tyr 31 over the active site on binding of dihydrofolate. This conformational change was dependent on the presence of glutamate on the para-aminobenzoylamide moiety of dihydrofolate. In its absence, the conformation remained open. Although water could enter the active site and hydrogen bond to N5 of dihydrofolate, indicating the feasibility of water as the proton donor, this was not controlled by the conformation of Tyr 31. The water accessibility of the active site was low for both conformations of Tyr 31. However, when hydride was transferred from NADPH to C6 of dihydrofolate before protonation, the average time during which water was found in hydrogen bonding distance to N5 of dihydrofolate in the active site increased almost fivefold. These results indicated that water can serve as the Broensted acid for the protonation of N5 of dihydrofolate during the DHFR catalyzed reduction.     

Importance of substrate and cofactor polarization in the active site of dihydrofolate reductase

By using a combined quantum-mechanical and molecular-mechanical potential in molecular dynamics simulations, we have investigated the effects of the enzyme electric field of dihydrofolate reductase on the electronic polarization of its 5-protonated dihydrofolate substrate at various stages of the catalyzed hydride transfer reaction. Energy decomposition of the total electrostatic interaction energy between the ligands and the enzyme shows that the polarization effect is 4% of the total electrostatic interaction energy, and, significantly, it accounts for 9kcal/mol of transition state stabilization relative to the reactant state. Therefore it is essential to take account of substrate polarization for quantitative interpretation of enzymatic function and for calculation of binding free energies of inhibitors to a protein. Atomic polarizations are calculated as the differences in the average atomic charges on the atoms in gas phase and in molecular simulations of the enzyme; this analysis shows that the glutamate tail and the pterin ring are the highly polarized regions of the substrate. Electron density difference plots of the reactant and product complexes at instantaneous configurations in the enzyme active center confirm the inferences made on the basis of partial atomic charges.     

Conformational changes during the catalytic cycle of gluconate kinase as revealed by X-ray crystallography

The crystal structure of gluconate kinase from Escherichia coli has been determined to 2.0 A resolution by X-ray crystallography. The three-dimensional structure was solved by multi-wavelength anomalous dispersion, using a crystal of selenomethionine-substituted enzyme. Gluconate kinase is an alpha/beta structure consisting of a twisted parallel beta-sheet surrounded by alpha-helices with overall topology similar to nucleoside monophosphate (NMP) kinases, such as adenylate kinase. In order to identify residues involved in substrate binding and catalysis, structures of binary complexes with ATP, the ATP analogue adenosine 5'-(beta,gamma-methylene) triphosphate and the product, gluconate-6-phosphate have been determined. Significant conformational changes are induced upon binding of ATP to the enzyme. The largest changes involve a hinge-bending motion of the NMP(bind) part and a motion of the LID with adjacent helices, which opens the cavity to the second substrate, gluconate. Opening of the active site cleft upon ATP binding is the opposite of what has been observed in the NMP kinase family so far, which usually close their active site to prevent fortuitous hydrolysis of ATP. The conformational change positions the side-chain of Arg120 to stack with the purine ring of ATP and the side-chain of Arg124 is shifted to interact with the alpha-phosphate in ATP, at the same time protecting ATP from solvent water. The beta and gamma-phosphate groups of ATP bind in the predicted P-loop. A conserved lysine side-chain interacts with the gamma-phosphate group, and might promote phosphoryl transfer. Gluconate-6-phosphate binds with its phosphate group in a similar position as the gamma-phosphate of ATP, consistent with inline phosphoryl transfer. The gluconate binding-pocket in GntK is located in a different position than the nucleoside binding-site usually found in NMP kinases.     

Modulation of cytochemically detected dihydrofolate reductase during cell cycle.

Dihydrofolate reductase (DHFR, EC 1.5.1.3) is an important enzyme involved in DNA metabolism. In this connection the cell cycle modulation of DHFR levels in HeLa S3 and HL 60 cell lines was investigated by flow cytometric analysis. A concentration of 4 micrograms/ml of aphidicolin was employed to synchronize the cell lines. DHFR was cytochemically detected by using tetrazolium salt and immunofluorescence techniques; DNA content was evaluated by means of propidium iodide staining. At 0, 2, 4, 6, 8, 10, 12 hrs. after the removal of the drug we observed a low DHFR level in G0-G1 phase, followed by an increase during late S and G2/M phases. The variations of this enzyme may represent, under well defined conditions, a marker of cycling cells.     

Conformational changes at the active site of pantetheine hydrolase during denaturation by guanidine hydrochloride

Conformational changes at the active site of pantetheine hydrolase (EC3.5.1.-) during guanidine hydrochloride (GndHCl) denaturation were investigated by UV and circular dichroism spectroscopy and by electron spin resonance spectroscopy, following the spectral behaviour of the nitroxide radicals (N- (1- oxyl - 2,2,5,5, -tetramethyl-3-pyrrolidinyl) iodacetamide) covalently linked to the two active site cysteine residues. At low denaturant concentrations (0.2 M) no conformational changes may be observed, whereas the catalytic activity, is strongly affected. The results indicate that the active site of pantetheine hydrolase is labile and unfolds under conditions in which no global tertiary struscture modifications can be observed.     

Single amino acid substitutions in flexible loops can induce large compressibility changes in dihydrofolate reductase

To address the effects of single amino acid substitutions on the flexibility of Escherichia coli dihydrofolate reductase (DHFR), the partial specific volume (v(o)) and adiabatic compressibility (beta(s)(o)) were determined for a series of mutants with amino acid replacements at Gly67 (7 mutants), Gly121 (6 mutants), and Ala145 (5 mutants) located in three flexible loops, by means of precise sound velocity and density measurements at 15 degrees C. These mutations induced large changes in v(o) (0.710-0.733 cm(3). g(-1)) and beta(s)(o) (-1.8 x 10(-6)-5.5 x 10(-6) bar(-1)) from the corresponding values for the wild-type enzyme (v(o)=0.723 cm(3). g(-1), beta(s)(o) = 1.7 x 10(-6) bar(-1)), probably due to modifications of internal cavities. The beta(s)(o) value increased with increasing v(o), but showed a decreasing tendency with the volume of the amino acid introduced. There was no significant correlation between beta(s)(o) and the overall stability of the mutants determined from urea denaturation experiments. However, a mutant with a large beta(s)(o) value showed high enzyme activity mainly due to an enhanced catalytic reaction rate (k(cat)) and in part due to increased affinity for the substrate (K(m)), despite the fact that the mutation sites are far from the catalytic site. These results demonstrate that the flexibility of the DHFR molecule is dramatically influenced by a single amino acid substitution in one of these loops and that the flexible loops of this protein play important roles in determining the enzyme function.     

Switching an active site helix in dihydrofolate reductase reveals limits to subdomain modularity

To what degree are individual structural elements within proteins modular such that similar structures from unrelated proteins can be interchanged? We study subdomain modularity by creating 20 chimeras of an enzyme, Escherichia coli dihydrofolate reductase (DHFR), in which a catalytically important, 10-residue α-helical sequence is replaced by α-helical sequences from a diverse set of proteins. The chimeras stably fold but have a range of diminished thermal stabilities and catalytic activities. Evolutionary coupling analysis indicates that the residues of this α-helix are under selection pressure to maintain catalytic activity in DHFR. Reversion to phenylalanine at key position 31 was found to partially restore catalytic activity, which could be explained by evolutionary coupling values. We performed molecular dynamics simulations using replica exchange with solute tempering. Chimeras with low catalytic activity exhibit nonhelical conformations that block the binding site and disrupt the positioning of the catalytically essential residue D27. Simulation observables and in vitro measurements of thermal stability and substrate-binding affinity are strongly correlated. Several E. coli strains with chromosomally integrated chimeric DHFRs can grow, with growth rates that follow predictions from a kinetic flux model that depends on the intracellular abundance and catalytic activity of DHFR. Our findings show that although α-helices are not universally substitutable, the molecular and fitness effects of modular segments can be predicted by the biophysical compatibility of the replacement segment.     

Probing the role of two hydrophobic active site residues in the human dihydrofolate reductase by site-directed mutagenesis

In the x-ray structure of the human dihydrofolate reductase, phenylalanine 31 and phenylalanine 34 have been shown to be involved in hydrophobic interactions with bound substrates and inhibitors. Using oligonucleotide-directed mutagenesis and a bacterial expression system producing the wild-type and mutant human dihydrofolate reductases at levels of 10% of the bacterial protein, we have constructed, expressed, and purified a serine 31 (S31) mutant and a serine 34 (S34) mutant. Fluorescence titration experiments indicated that S31 bound the substrate H2folate 10-fold tighter and the coenzyme NADPH 2-fold tighter than the wild-type human dihydrofolate reductase. The serine 31 mutation had little effect on the steady-state kinetic properties of the enzyme but produced a 100-fold increase in the dissociation constant (Kd) for the inhibitor methotrexate. The serine 34 mutant had much greater alterations in its properties than S31; specifically, S34 had a 3-fold reduction in the Km for NADPH, a 24-fold increase in the Km for H2folate, a 3-fold reduction in the overall reaction rate kcat, and an 80,000-fold increase in the Kd for methotrexate. In addition, the pH dependence of the steady-state kinetic parameters of S34 were different from that of the wild-type enzyme. These results suggest that phenylalanine 31 and phenylalanine 34 make very different contributions to ligand binding and catalysis in the human dihydrofolate reductase.     

Sequential conformation changes of chinese hamster ovary dihydrofolate reductase at its active sites

The local fluorescence probes, 2-(p-toluidino)-6-naphthalenesulfonic acid (TNS) and NADPH were employed to detect urea-induced conformation changes at each active site of dihydrofolate reductase (DHFR), respectively. The results indicate that local conformation change at DHF/TNS could be superimposed by the conformation change calculated from the enzyme activity change with a three-state model; while at NADPH site it is lagged in the first transition. This difference is further supported by the different relative changes of Michaelis constants at 0, 1 and 1.8 M urea for each substrate. Our results suggest that local conformation at DHF site is more flexible than that at NADPH site, and the urea-induced unfolding could be ascribed to a four-state transition.     

Anti-idiotypic antibodies elicited by pterin recognize active site epitopes in dihydrofolate reductases and dihydropteridine reductase

Monoclonal antibodies (mAbs) against antipterin immunoglobulin and dihydropteridine reductase (DHPR) and also polyclonal antibodies against human dihydrofolate reductase (DHFR) were obtained. The anti-idiotypic mAbs and anti-DHPR mAbs bind specifically to human DHFR, Escherichia coli DHFR, soybean seedling DHFR, and human DHPR in solid-phase immunoassays. Further, the mAbs bind to the native but not to the denatured forms of DHFRs. The monoclonal antibodies also inhibit the enzymatic activity of human DHFR but not that of human DHPR. Competitive solid-phase immunoassays show stoichiometric inhibition by methotrexate and partial inhibition by NADPH of mAb binding to human DHFR. Cyanogen bromide fragments derived from human DHFR (residues 15-52 and 53-111), containing several active site residues, bind partially to some of the monoclonal antibodies. Accordingly, polyclonal antibodies to peptide 53-111 of human DHFR cross-react to some extent with human DHPR. Data from competitive immunoassays in which the binding of the various mAbs was tested singly and in combination with other mAbs suggest that these antibodies bind to a common region on human DHFR. The results also indicate that the mAbs display some heterogeneity with respect to specific epitopes. These data suggest that despite the absence of significant amino acid sequence homologies among the various DHFRs and DHPR, they have a fundamentally similar topography at the site of binding of the pterin moiety that is recognized by the anti-idiotypic mAbs generated by pterin. In the relatively simple structure of the pterin ring system there are different substituent groups at positions C4 and C6 in methotrexate, 7,8-dihydrofolate, and 7,8-dihydrobiopterin, suggesting that these antibodies are specific for regions on various proteins that interact with the remainder of the pterin moiety. These mAbs and similar mAbs specified by substituent groups on pterin may thus be used as specific probes or inhibitors of various folate-dependent enzymes and transport proteins. They should also provide insights into some of the general features of antibody recognition of protein antigens.     

The use of triazine inhibitors in mapping the active site region of Lactobacillus casei dihydrofolate reductase.

A quantitative structure-activity relationship for the inhibition of Lactobacillus casei dihydrofolate reductase by 4,6-diamino-1,2-dihydro-2,2-dimethyl-1-(3-X-phenyl)-s-triazines has been derived: for 28 congeners. In this expression C is the molar concentration of triazine causing 50% inhibition, π₃ is the hydrophobic constant for the 3-X-phenyl substituent, MR′ is the molar refractivity of certain substituents, β is an iteratively derived coefficient, and r is the multiple least squares correlation coefficient. This correlation is quite different from those found with the same type of inhibitors acting on bovine and rat liver dihydrofolate reductase. The differences are discussed. The correlation equations for the triazines acting on purified enzymes are compared with equations correlating triazines inhibiting the growth of Staphylococcus aureus and Escherichia coli.     

R67, the other dihydrofolate reductase: rational design of an alternate active site configuration

R67 dihydrofolate reductase (DHFR) bears no sequence or structural homologies with chromosomal DHFRs. The gene for this enzyme produces subunits that are 78 amino acids long, which assemble into a homotetramer possessing 222 symmetry. More recently, a tandem array of four gene copies linked in-frame was constructed, which produces a monomer containing 312 amino acids named Quad3. Asymmetric mutations in Quad3 have also been constructed to probe the role of Q67 and K32 residues in catalysis. This present study mixes and matches mutations to determine if the Q67H mutation, which tightens binding approximately 100-fold to both dihydrofolate (DHF) and NADPH, can help rescue the K32M mutation. While the latter mutation weakens DHF binding over 60-fold, it concurrently increases kcat by a factor of 5. Two Q67H mutations were added to gene copies 1 and 4 in conjunction with the K32M mutation in gene copies 1 and 3. Addition of these Q67H mutations tightens binding 40-fold, and the catalytic efficiency (kcat/Km(DHF)) of the resulting protein is similar to that of Quad3. Since these Q67H mutations can mostly compensate for the K32M lesion, K32 must not be necessary for DHF binding. Another multimutant combines the K32M mutation in gene copies 1 and 3 with the Q67H mutation in all gene copies. This mutant is inhibited by DHF but not NADPH, indicating that NADPH binds only to the wild type half of the pore, while DHF can bind to either the wild type or mutant half of the pore. This inhibition pattern contrasts with the mutant containing only the Q67H substitution in all four gene copies, which is severely inhibited by both NADPH and substrate. Since gene duplication and divergence are evolutionary tools for gaining function, these constructs are a first step toward building preferences for NADPH and DHF in each half of the active site pore of this primitive enzyme.     

Crystallographic analysis reveals a novel second binding site for trimethoprim in active site double mutants of human dihydrofolate reductase

In order to produce a more potent replacement for trimethoprim (TMP) used as a therapy for Pneumocystis pneumonia and targets dihydrofolate reductase from Pneumocystis jirovecii (pjDHFR), it is necessary to understand the determinants of potency and selectivity against DHFR from the mammalian host and fungal pathogen cells. To this end, active site residues in human (h) DHFR were replaced with those from pjDHFR. Structural data are reported for two complexes of TMP with the double mutants Gln35Ser/Asn64Phe (Q35S/N64F) and Gln35Lys/Asn64Phe (Q35K/N64F) of hDHFR that unexpectedly show evidence for the binding of two molecules of TMP: one molecule that binds in the normal folate binding site and the second molecule that binds in a novel subpocket site such that the mutated residue Phe64 is involved in van der Waals contacts to the trimethoxyphenyl ring of the second TMP molecule. Kinetic data for the binding of TMP to hDHFR and pjDHFR reveal an 84-fold selectivity of TMP against pjDHFR (K_i 49 nM) compared to hDHFR (K_i 4093 nM). Two mutants that contain one substitution from pj- and one from the closely related Pneumocystis carinii DHFR (pcDHFR) (Q35K/N64F and Q35S/N64F) show K_i values of 593 and 617 nM, respectively; these K_i values are well above both the K_i for pjDHFR and are similar to pcDHFR (Q35K/N64F and Q35S/N64F) (305 nM). These results suggest that active site residues 35 and 64 play key roles in determining selectivity for pneumocystis DHFR, but that other residues contribute to the unique binding of inhibitors to these enzymes.