NMR studies of ligand carboxylate group interactions with arginine residues in complexes of Lactobacillus casei dihydrofolate reductase with substrates and substrate analogues

In a series of complexes of Lactobacillus casei dihydrofolate reductase (DHFR) formed with substrates and substrate analogues, the (1)H/(15)N NMR chemical shifts for the guanidino group of the conserved Arg 57 residue were found to be sensitive to the mode of binding of their H(eta) protons to the charged oxygen atoms in ligand carboxylate groups. In all cases, Arg 57 showed four nonequivalent H(eta) signals indicating hindered rotation about the N(epsilon)-C(zeta) and C(zeta)-N(eta) bonds. The H(eta)(12) and H(eta)(22) protons have large downfield shifts as expected for a symmetrical end-on interaction with the ligand carboxylate group. The chemical shifts are essentially the same in the complexes with folate and p-aminobenzoyl-L-glutamate (PABG) and similar to those found previously for the methotrexate complex reflecting the strong and similar hydrogen bonds formed with the carboxylate oxygens. Interestingly, the rates of rotation about the N(epsilon)-C(zeta) bond for the complexes containing the weakly binding PABG fragment are almost identical to those measured in the complex with methotrexate, which binds 10(7) times more tightly. In the methotrexate complex, this rotation depends on correlated rotations about the N(epsilon)-C(zeta) bond of Arg 57 and the C(alpha)-C' bond of the ligand glutamate alpha-carboxylate group. Thus, even in a fragment such as PABG, which has a much faster off-rate, the carboxylate group binds to the enzyme in a similar way to that in a parent molecule such as folate and methotrexate with the rotation about the N(epsilon)-C(zeta) bond of Arg 57 being essentially the same in all the different complexes.     

Structural comparisons of complexes of methotrexate analogues with Lactobacillus casei dihydrofolate reductase by two-dimensional 1H NMR at 500 MHz

We have used two-dimensional (2D) NMR methods to examine complexes of Lactobacillus casei dihydrofolate reductase and methotrexate (MTX) analogues having structural modifications of the benzoyl ring [the 3',5'-difluoro and 3',5'-dichloro analogues (II and III)] and also the glutamic acid moiety [the alpha- and gamma-monoamides (IV and V)]. Assignments of the 1H signals in the spectra of the various complexes were made by comparison of their 2D spectra with those of complexes containing methotrexate where we have previously assigned resonances from 32 of the 162 amino acid residues. In the complexes formed with the dihalomethotrexate analogues, the glutamic acid and pteridine ring moieties were shown to bind to the enzyme in a manner similar to that found in the methotrexate-enzyme complex. Perturbations in 1H chemical shifts of protons in Phe-49, Leu-54, and Leu-27 and the methotrexate H7 and NMe protons were observed in the different complexes and were accounted for by changes in orientation of the benzoyl ring in the various complexes (15 degrees and 25 degrees in the difluoro- and dichloromethotrexate complexes, respectively). Binding of oxidized or reduced coenzyme (NADP+ or NADPH) to the binary complexes did not result in different shifts for Leu-27, Leu-54, or Leu-19 protons, and thus, the orientation of the benzoyl ring of the methotrexate analogues is not perturbed greatly by the presence of either oxidized or reduced coenzyme.(ABSTRACT TRUNCATED AT 250 WORDS)     

1H/15N HSQC NMR studies of ligand carboxylate group interactions with arginine residues in complexes of brodimoprim analogues and Lactobacillus casei dihydrofolate reductase

1H and 15N NMR studies have been undertaken on complexes of Lactobacillus casei dihydrofolate reductase (DHFR) formed with analogues of the antibacterial drug brodimoprim (2,4-diamino-5-(3', 5'-dimethoxy-4'-bromobenzyl)pyrimidine) in order to monitor interactions between carboxylate groups on the ligands and basic residues in the protein. These analogues had been designed by computer modeling with carboxylated alkyl chains introduced at the 3'-O position in order to improve their binding properties by making additional interactions with basic groups in the protein. Specific interactions between ligand carboxylate groups and the conserved Arg57 residue have been detected in studies of 1H/15N HSQC spectra of complexes of DHFR with both the 4-carboxylate and the 4, 6-dicarboxylate brodimoprim analogues. The spectra from both complexes showed four resolved signals for the four NHeta protons of the guanidino group of Arg57, and this is consistent with hindered rotation in the guanidino group resulting from interactions with the 4-carboxylate group in each analogue. In the spectra of each complex, one of the protons from each of the two NH2 groups and both nitrogens are considerably deshielded compared to the shielding values normally observed for such nuclei. This pattern of deshielding is that expected for a symmetrical end-on interaction of the carboxylate oxygens with the NHeta12 and NHeta22 guanidino protons. The differences in the degree of deshielding between the complexes of the two structurally similar brodimoprim analogues and the methotrexate indicates that the shielding is very sensitive to geometry, most probably to hydrogen bond lengths. The 1H/15N HSQC spectrum of the DHFR complex with the brodimoprim-6-carboxylate analogue does not feature any deshielded Arg NHeta protons and this argues against a similar interaction with the Arg57 in this case. It has not proved possible to determine whether the 6-carboxylate in this analogue is interacting directly with any residue in the protein. 1H/15N HSQC spectra have been fully assigned for the complexes with the three brodimoprim analogues and chemical shift mapping used to explore interactions in the binding site. The 1H signals of the bound ligands for all three brodimoprim analogues have been assigned. Their 1H chemical shifts were found to be fairly similar in the different complexes indicating that the 2, 4-diaminopyrimidine and the benzyl ring are binding in essentially the same binding sites and with the same overall conformation in the different complexes. The rotation rate about the NepsilonCzeta bond in the brodimoprim-4,6-dicarboxylate complex with DHFR has been determined from a zz-HSQC exchange experiment, and its value is quite similar to that observed in the DHFR.methotrexate complex (24 +/- 10 s-1 at 8 degrees C and 50 +/- 10 s-1 at 15 degrees C, respectively). The 1H and 15N chemical shift differences of selected amide and guanidino NH groups, measured between the DHFR complexes, provided further evidence about the interactions involving Arg57 with the 4-carboxylate and 4,6-dicarboxylate brodimoprim analogues.     

Characterization of rates of ring-flipping in trimethoprim in its ternary complexes with Lactobacillus casei dihydrofolate reductase and coenzyme analogues

NMR measurements have been used to investigate rates of ring-flipping and the activation parameters for the trimethoxybenzyl ring of the antibacterial drug trimethoprim (TMP) bound to Lactobacillus casei dihydrofolate reductase (DHFR) for a series of ternary complexes formed with analogues of the coenzyme NADPH. Rates were obtained at several temperatures from line shape analyses ((13)C-edited HSQC (1)H spectra) and transfer of magnetization measurements (zz-HSQC) on complexes containing 3'-O-[(13)C]trimethoprim. Examination of the structures of the complexes indicates that ring-flipping can only be achieved following major conformational changes and transient fluctuations of the protein and coenzyme structure around the trimethoxybenzyl ring. There is no simple correlation between rates of ring-flipping and binding constants. The presence of the coenzyme nicotinamide ring (in either its reduced or its oxidized forms) in the binding site close to the trimethoxybenzyl ring moiety is the major factor reducing the ring-flipping on coenzyme binding. Thus, the ternary complex with NADPH shows the largest reduction in the rate of ring-flipping (11 +/- 3 s(-)(1) at 298 K) as compared with the binary complex (793 +/- 80 s(-)(1) at 298 K). Complexes with NADPH analogues that either have no nicotinamide ring or are known to have their nicotinamide rings removed from the binding site show the smallest reductions. For the DHFR.TMP.NADP(+) complex where there are two conformations present, very different rates of ring-flipping were observed for the two forms. The activation parameters (DeltaH() and DeltaS()) for the ring-flipping in all the complexes are discussed in terms of the protein-ligand interactions and the possible constraints on the pathway through the transition state.     

DNA binding by dihydrofolate reductase from Lactobacillus casei

The protein-dependent retention of double-stranded DNA molecules on nitrocellulose filters has been used to show that pure dihydrofolate reductase from Lactobacillus casei has affinity for DNA. Dihydrofolate reductase will bind to end-labeled linear double-stranded DNA and to DNA in supercoiled form. Coenzymes and certain inhibitors do not affect the affinity of the protein to DNA, indicating that the DNA-binding region of the protein is distinct from the binding sites for these molecules. Comparison of the retention on filters by dihydrofolate reductase of two plasmid DNAs, differing only in a 3000-base pair insert containing the L. casei gene for dihydrofolate reductase, showed that in the presence of this DNA region lower concentrations of the protein were required to give significant retention; it is possible that a specific DNA-protein interaction underlies this effect. This presents the possibility of studying the interaction with DNA of a protein for which a crystal structure and considerable nuclear magnetic resonance data are already available.     

Interaction of the carboxamide of NADPH with Lactobacillus casei dihydrofolate reductase

Dihydrofolate reductase from Lactobacillus casei and its complexes with NADPH and methotrexate yield well-resolved Raman spectra. The 1685-cm^?1 Raman band assigned to the carboxamide of NADPH persists in the NADPH-enzyme binary complex but is absent from the NADPH-methotrexate-enzyme ternary complex. This is ascribed to stabilization of the polarized form of the carboxamide by H bonding to the NH and CO groups of Ala 6 and Ile 13 of the peptide backbone.     

Ultraviolet difference-spectroscopic studies of substrate and inhibitor binding to Lactobacillus casei dihydrofolate reductase.

The u.v. difference spectra generated when methotrexate, trimethoprim or folate bind to Lactobacillus casei dihydrofolate reductase were analysed. The difference spectrum producted by methotrexate binding is shown to consist of three components: (a) one closely resembling that observed on protonation of methotrexate, reflecting an increased degree of protonation on binding; (b) a pH-independent contribution corresponding to a 40 nm shift to longer wavelengths of a single absorption band of methotrexate: (c) a component arising from perturbation of tryptophan residue(s) of the enzyme. Quantitative analysis of the pH-dependence of component (a) shows that pK of methotrexate is increased from 5.35 to 8.55 (+/-0.10) on binding. In contrast, folate is not protonated when bound to the enzyme at neutral pH. At pH7.5, where methotrexate is bound 2000 times more tightly than folate, one-third of the difference in binding energy between the two compounds arises from the difference in chaarge stage. A similar analysis of the difference spectra generated on trimethoprim binding demonstrates that this compound, too, shows an increase in pK on binding but only from 7.22 to 7.90 (+/-0.10), suggesting that its 2,4-diaminopyrimidine ring does not bind to the enzyme in precisely the same way as the corresponding moiety of methotrexate.     

Proton nuclear magnetic resonance saturation transfer studies of coenzyme binding to Lactobacillus casei dihydrofolate reductase

The chemical shifts of all the aromatic proton and anomeric proton resonances of NADP+, NADPH, and several structural analogues have been determined in their complexes with Lactobacillus casei dihydrofolate reductase by double-resonance (saturation transfer) experiments. The binding of NADP+ to the enzyme leads to large (0.9-1.6 ppm) downfield shifts of all the nicotinamide proton resonances and somewhat smaller upfield shifts of the adenine proton resonance. The latter signals show very similar chemical shifts in the binary and ternary complexes of NADP+ and the binary complexes of several other coenzymes, suggesting that the environment of the adenine ring is similar in all cases. In contrast, the nicotinamide proton resonances show much greater variability in position from one complex to another. The data show that the environments of the nicotinamide rings of NADP+, NADPH, and the thionicotinamide and acetylpyridine analogues of NADP+ in their binary complexes with the enzyme are quite markedly different from one another. Addition of folate or methotrexate to the binary complex has only modest effects on the nicotinamide ring of NADP+, but trimethoprim produces a substantial change in its environment. The dissociation rate constant of NADP+ from a number of complexes was also determined by saturation transfer.     

Effects of coenzyme analogues on the binding of p-aminobenzoyl-L-glutamate and 2,4-diaminopyrimidine to Lactobacillus casei dihydrofolate reductase

The binding of p-aminobenzoyl-L-glutamate and 2,4-diaminopyrimidine to dehydrofolate reductase from Lactobacillus casei MTX/R in the presence of a series of co-enzymes and coenzyme analogues has been measured fluorometrically. These two ligands, which can be regarded as "fragments" of the powerful inhibitor methotrexate, have been shown to bind cooperatively in the absence of coenzyme [Birdsall, B., Burgen, A. S. V., Rodrigues de Miranda, J., & Roberts, G. C. K. (1978) Biochemistry 17, 2102], p-amino-benzoyl-L-glutamate binding 58 times more tightly in the presence of 2,4-diaminopyrimidine than in its absence. In the presence of coenzymes, this cooperativity ranges from 1.8- to 428-fold. The effects of coenzymes on individual binding steps range from an 8-fold decrease in binding constant to a 23-fold increase. The structural specificity of these effects are discussed in terms of a model involving ligand-induced conformational changes and compared with the effects on trimethoprim and methotrexate binding described in the preceding paper [Birdsall, B., Burgen, A. S. V., & Roberts, G. C. K. (1980) Biochemistry (first paper of four in this issue)].     

Circular-dichroism studies of ligand binding to dihydrofolate reductase from Lactobacillus casei MTX/R.

Circular-dichroism spectra (200--450 nm) were recorded for Lactobacillus casei MTX/R dihydrofolate reductase and its complexes with substrates, inhibitors and coenzymes. These spectra are compared with those reported by others for dihydrofolate reductase from other sources. The binding of NADP+ or NADPH is associated with the perturbation of one or more aromatic amino acid residues, and there is marked enhancement of the negative c.d. band at 340 nm arising from the dihydronicotinamide chromophore of NADPH. The substrates folate and dihydrofolate give rise to substantial extrinsic c.d. bands on binding, which show a number of specific differences between enzymes from different sources. The binary complexes between the enzyme and the inhibitors methotrexate or trimethoprim also show strong c.d. bands, and these are qualitatively very similar for all dihydrofolate reductases studied so far. The ternary complexes between enzyme, NADPH and trimethoprim or methotrexate are very different from the sum of the spectra of the binary complexes. Trimethoprim leads to the disappearance of the 340 nm c.d. band of bound NADPH, whereas in the methotrexate--NADPH--enzyme ternary complex a "couplet" c.d. spectrum is observed at long wavelengths. Analysis of this latter feature suggests that it arises from a direct interaction between the dihydronicotinamide and pteridine rings in the ternary complex.     

Nucleotide sequence of the dihydrofolate reductase gene of methotrexate-resistant Lactobacillus casei

The nucleotide sequence of the dihydrofolate reductase (DHFR) gene of a methotrexate-resistant strain of Lactobacillus casei, which is the source of DHFR for nuclear magnetic resonance (NMR) studies, has been determined. The derived amino acid sequence differs from that obtained by protein sequencing by the presence of aspartic acid instead of asparagine at position 8 and proline instead of leucine at position 90. The nucleotide sequences of 320-bp 5' and 335-bp 3' flanking regions of this gene have also been determined.     

Molecular cloning of the gene for dihydrofolate reductase from Lactobacillus casei

The Lactobacillus casei gene for dihydrofolate reductase has been cloned in Escherichia coli using the multicopy vector pBR322. A restriction map of the cloned DNA has been prepared. The cloned DNA directs the synthesis of L. casei dihydrofolate reductase in E. coli and confers trimethoprim and methotrexate resistance.     

Reduction of oxidised folates by dihydrofolate reductase from methotrexate-resistant Lactobacillus casei.

The use of alternative substrates by dihydrofolate reductase (5,6,7,8-tetrahydrofolate: NADP+ oxidoreductase, EC 1.5.1.3) was investigated as a possible mechanism for the resistance of Lactobacillus casei to the cytotoxic drug methotrexate. The reduction of folic acid and 10-formylfolic acid by homogeneous enzyme was compared to that of the normal substrate, dihydrofolic acid. The three substrates have different pH optima and Km values. In addition, it was found that the reduction of 10-formylfolic acid was markedly stimulated by the presence of ions. Although the reduction was sensitive to methotrexate in all cases, the ion activation may be of importance in partially inhibited systems.     

NMR studies of differences in the conformations and dynamics of ligand complexes formed with mutant dihydrofolate reductases

Two mutants of Lactobacillus casei dihydrofolate reductase, Trp 21----Leu and Asp 26----Glu, have been prepared by using site-directed mutagenesis methods, and their ligand binding and structural properties have been compared with those of the wild-type enzyme. 1H, 13C, and 31P NMR studies have been carried out to characterize the structural changes in the complexes of the mutant and wild-type enzymes. Replacement of the conserved Trp 21 by a Leu residue causes a decrease in activity of the enzyme and reduces the NADPH binding constant by a factor of 400. The binding of substrates and substrate analogues is only slightly affected. 1H NMR studies of the Trp 21----Leu enzyme complexes have confirmed the original resonance assignments for Trp 21. In complexes formed with methotrexate and the mutant enzyme, the results indicate some small changes in conformation occurring as much as 14 A away from the site of substitution. For the enzyme-NADPH complexes, the chemical shifts of nuclei in the bound coenzyme indicate that the nicotinamide ring binds differently in complexes with the mutant and the wild-type enzyme. There are complexes where the wild-type enzyme has been shown to exist in solution as a mixture of conformations, and studies on the corresponding complexes with the Trp 21----Leu mutant indicate that the delicately poised equilibria can be perturbed. For example, in the case of the ternary complex formed between enzyme, trimethoprim, and NADP+, two almost equally populated conformations (forms I and II) are seen with the wild-type enzyme but only form II (the one in which the nicotinamide ring of the coenzyme is extended away from the enzyme structure and into the solvent) is observed for the mutant enzyme complex. It appears that the Trp 21----Leu substitution has a major effect on the binding of the nicotinamide ring of the coenzyme. For the Asp 26----Glu enzyme there is a change in the bound conformation of the substrate folate. Further indications that some conformational adjustments are required to allow the carboxylate of Glu 26 to bind effectively to the N1 proton of inhibitors such as methotrexate and trimethoprim come from the observation of a change in the dynamics of the bound trimethoprim molecule as seen from the increased rate of the flipping of the 13C-labeled benzyl ring and the increased rate of the N1-H bond breaking.     

Trimethoprim binding to Lactobacillus casei dihydrofolate reductase: a 13C NMR study using selectively 13C-enriched trimethoprim

We have measured the 13C chemical shifts for trimethoprim molecules selectively enriched with 13C at the 2-, 4-, 5-, 6-, and 7-positions and the p-OCH3 position in their complexes with Lactobacillus casei dihydrofolate reductase in the presence and absence of coenzyme analogues. The C2 carbon shifts indicate that the pyrimidine ring is protonated at N1 in all the complexes of trimethoprim with the enzyme and coenzymes and in each case the pyrimidine ring is binding in a similar way to that of the corresponding part of methotrexate in the enzyme-methotrexate complex. The C6 carbon of trimethoprim shows a large upfield shift in all complexes (3.51 to 4.70 ppm) but no shift in the complex of 2,4-diaminopyrimidine with the enzyme: these shifts probably arise from steric interactions between the C1' and C2' carbons and the H6 proton, which approach van der Waals contact in the folded conformation adopted by trimethoprim when bound to the enzyme. The large shift observed for C6 in all complexes indicates that the basic folded conformation is present in all of them. A comparison of the 13C shifts in the enzyme-trimethoprim-NADPH complex with those in the enzyme-trimethoprim binary complex shows substantial changes even for carbons such as C6 and p-OCH3 (0.46 and -0.36 ppm, respectively), which are remote from the coenzyme: these are caused by ligand-induced conformational changes that may involve displacement of the helix containing residues 42-49.(ABSTRACT TRUNCATED AT 250 WORDS)