Dihydrofolate reductase from a methotrexate-resistant strain of Escherichia coli: dihydrofolate monooxygenase activity

Dihydrofolate reductase from strain MB 1428 of Escherichia coli was shown to catalyze the oxidative cleavage of dihydrofolate at the C(9)N(10) bond. One of the products of the reaction was identified as 7,8-dihydropterin-6-carboxaldehyde through its proton magnetic resonance spectrum. The maximal enzymatic rate was 0.05 moles dihydrofolate cleaved per minute per mole enzyme at 25° and pH 7.2, and the K_M for dihydrofolate was 17.5 ± 2.5 μM. The enzymatic reaction was fully inhibitable with methotrexate. The mechanism of enzyme action was proposed to be an apparent “acidification” of dihydrofolate upon binding to the enzyme. Folate underwent an analogous oxidative cleavage by enzyme with a turnover number of 0.0014, which produced pterin-6-carboxaldehyde. Methotrexate was also slowly degraded by the enzyme.     

Dihydrofolate reductase: the amino acid sequence of the enzyme from a methotrexate-resistant mutant of Escherichia coli.

The determination of the amino acid sequence of the enzyme dihydrofolate reductase (5,6,7,8-tetrahydrofolate:NADP+ oxidoreductase, EC 1.5.1.3) from a mutant of Escherichia coli B is described. The 159 residues were positioned by automatic Edman degradation of the whole protein, of the reduced and alkylated cyanogen bromide fragments, and of selected tryptic, chymotryptic, and thermolytic digestion products. An N-bromosuccinimide produced fragment of the largest cyanogen bromide peptide was also used in the sequence determination.     

Circular dichroism studies of dihydrofolate reductase from a methotrexate-resistant strain of Escherichia coli B, MB 1428: ternary complexes.

Circular dichroism has been used to monitor the binding of pyridine nucleotide cofactors to enzyme-folate analog complexes of dihydrofolate reductase from Escherichia coli B (MB 1428). The enzyme binds one molar equivalent of many folate analogs and two molar equivalents of several pyridine nucleotide cofactors. The apo-enzyme has very low optical activity. The binding of folate analogs including folate, dihydrofolate, methotrexate, trimethoprim and pyrimethamine induce large Cotton effects. Pyridine nucleotides when bound to the enzyme-folate analog complexes also induce new optically active bands; all the effects being due to the first molar equivalent of cofactor bound. NADPH and NADP+ induce very similar bands when bound to the enzyme-methotrexate complex suggesting that the geometry of the complexes formed are very similar. The oxidized and reduced cofactor likewise have similar effects on the enzyme-folate complex. However, NADPH and NADP+ addition to both the enzyme-trimethoprim and enzyme-pyrimethamine complexes have significantly different effects on the circular dichroism spectra, suggesting that the inhibitors which are less homologous to the natural dihydrofolate substrate allow more conformational freedom in the enzyme-inhibitor-cofactor complex. In most cases the prior binding of the folate analog greatly increases the binding of the first molar equivalent of cofactor so that at concentrations of approx. 5-20 muM the binding appears stoichiometric. Pyrimethamine is an exception in that it apparently has no effect on the binding of NADPH to the enzyme.     

31P n.m.r. studies of complexes of NADPH and NADP+ with Escherichia coli dihydrofolate reductase

We have measured the ³¹P n.m.r. spectra of NADP⁺ and NADPH in their binary complexes with Escherichia coli dihydrofolate reductase and in ternary complexes with the enzyme and folate or methotrexate. The ³¹P chemical shift of the 2′ phosphate group is the same in all complexes; its value indicates that it is binding in the dianionic state and its pH independence suggests that it is interacting strongly with cationic residue(s) on the enzyme. Similar behaviour has been noted previously for the complexes with the Lactobacillus casei enzyme although the ³¹P shift is somewhat different in this complex, possibly due to an interaction between the 2′ phosphate group and His 64 which is not conserved in the E. coli enzyme. For the coenzyme complexes with both enzymes ³¹POC₂¹H_2′ spin-spin interactions were detected (7.5–7.8 Hz) on the 2′ phosphate resonances, indicating a POC₂H_2′ dihedral angle of 30 or 330 : this is in good agreement with the value of 330° measured in crystallographic studies¹ (Matthews et al., 1978) on the L. casei enzyme. NADPH-MTX complex. The pyrophosphate resonances are shifted to different extents in the various complexes and there is evidence that there is more OPO bond angle distortion in the E. coli enzyme complexes than in those with the L. casei enzyme. The effects of ³¹POC₅¹H_5′ spin coupling were detected on one pyrophosphate resonance and indicate that the POC₅H_5′ torsion angle has changed by at least ～30° on binding to the E. coli enzyme: this is considerably less than the distortion (～50°) observed previously in the L. casei enzyme complex.     

Interaction of 13C-enriched folate with dihydrofolate reductase studied by carbon magnetic resonance spectroscopy

In the presence of dihydrofolate reductase the carbon magnetic resonance spectrum of folate labeled at the benzoylcarbonyl carbon with ¹³C contains two broadened peaks arising from free and enzyme-bound folate, the latter appearing over 2 ppm upfield from free folate. Addition of TPN⁺ causes sharpening of both peaks indicating formation of a single folate-TPN⁺-enzyme ternary complex. Methotrexate specifically displaces folate from the ternary complex regenerating a single sharp resonance at 170.4 ppm characteristic of free folate. Line width changes show that folate is bound more tightly in the ternary than in the binary complex. Increased shielding of this carbonyl upon binding is inconsistent with its participation in a H-bond.     

Proton magnetic resonance studies on Escherichia coli dihydrofolate reductase. Assignment of histidine C-2 protons in binary complexes with folates on the basis of the crystal structure with methotrexate and on chemical modifications.

The effects of pH upon the C-2 resonances of the 5 histidine residues of Escherichia coli MB 1428 dihydrofolate reductase in binary complexes with methotrexate, aminopterin, folate, methopterin, and trimethoprim were studied by 300-MHz 1H nmr spectroscopy. Three of the five histidine residues, labeled 1, 2, and 3, exhibited similar pK' values and chemical shifts for their C-2 protons in the five binary complexes. One histidine, 4, was quite different in the folate complex and the last histidine, 5 was quite different in the trimethoprim complex. For all five binary complexes, each histidine had a pK' which was significantly different from the other 4 histidines of that complex. Titration of the binary methotrexate complex of a 5,5'-dithiobis(2-nitrobenzoate)-modified enzyme showed that 2 histidines were not perturbed by this modification of Cys 152, and that the alkaline form of histidine 2, the acid form of histidine 4, and, to a lesser extent, the acid form of histidine 3 were slightly perturbed. Titration of the binary methotrexate complex of a N-bromosuccinimide-modified enzyme demonstrated that this modification slightly affected all of the histidines and drastically affected histidine 5. Histidines 3 and 5 of the binary methotrexate complex reacted rapidly with the histidine-specific reagent, ethoxyformic anhydride, while histidines 2 and 4 reacted at a moderate rate and histidine 1 reacted slowly if at all. The local electrostatic environments of the 5 histidine residues as deduced from the crystal structure of the binary complex of the enzyme with methotrexate (Matthews, D.A., Alden, R.A., Bolin, J.T., Freer, S.T., Hamlin, R., Xuong, N., Kraut, J., Poe, M., Williams, M.N., and Hoogsteen, K. (1977) Science 197, 594-597) were used as the basis for proposed assignments of the five histidine C-2 nmr resonances. The assignments were: 1, pK' 7.9 to 8.2, His 124; 2, pK' 7.2 to 7.4, His 141; 3, pK' 6.5 to 6.7, His 149; 4, pK' 5.7 to 6.3, His 114; and 5, pK' 5.2 to 5.9, His 45. The effect of the chemical modifications upon the enzyme's histidine residues were consistent with the assignments, but no direct chemical evidence in support of the assignments was obtained. It was proposed that, since the crystallographic data provided consistent assignments of the histidine nmr data for both native and chemically modified enzyme, the local environment of each of the 5 histidine residues was similar in the crystal and in solution.     

The purification of methionine sulfoxide reductase from Escherichia coli

A sensitive assay, based on the acylation of tRNA^Met, has been developed to measure the enzymatic reduction of methionine sulfoxide to methionine. Using this assay, methionine sulfoxide reductase has been purified to near homogeneity from extracts of Escherichia coli.     

On the measurement of pH in Escherichia coli by 31P nuclear magnetic resonance

The ³¹P high resolution NMR spectra of concentrated suspensions of Escherichia coli cells have been measured at 145.8 MHz. The position of the orthophosphate resonance is used as a measure of internal and external pH. In accord with Paddan, Zilberstein and Rottenberg ((1976) Eur. J. Biochem. 63, 533–541) it is shown that when properly energized the internal pH is 7.5 ± 0.1. By synchronizing the NMR data acquisition with 3-s bursts of O₂ it is possible to measure the internal pH with a time resolution of about 1 s. It is shown that at 20°C the pH remains constant for times longer than 15 s after the oxygen is discontinued and it decays in several minutes.     

Proton magnetic resonance studies of peroxidases from turnip and horseradish.

Proton NMR spectra at 270 MHz have been measured for horseradish peroxidase and turnip peroxidase isoenzymes (P1, P2, P3 and P7) in both their high spin ferric native states and as the low spin ferric cyanide complexes. Resonances of amino acids near the heme have been identified and used to investigate variations in the structure of the heme crevice amongst the enzymes. Ligand proton resonances have been resolved in spectra of the cyanide complexes of the peroxidases and these provide information on the heme electronic structure. The electronic structure of the heme and the tertiary structure of the heme crevice are essentially the same in the acidic turnip isoenzymes, P1, P2 and, to a lesser extent, P3 but differ in the basic turnip enzyme, P7. The heme electronic structure and nature of the iron ligands in peroxidases are discussed. Further evidence is presented for histidine as the proximal ligand. A heme-linked ionizable group with a pK of 6.5 has been detected by NMR in the cyanide complex of horseradish peroxidase.     

13C NMR studies of complexes of Escherichia coli dihydrofolate reductase formed with methotrexate and with folic acid.

13C NMR studies of 13C-labelled ligands bound to dihydrofolate reductase provide (DHFR) a powerful means of detecting and characterizing multiple bound conformations. Such studies of complexes of Escherichia coli DHFR with [4,7,8a,9-13C]- and [2,4a,6-13C]methotrexate (MTX) and [4,6,8a-13C]- and [2,4a,7,9-13C]folic acid confirm that in the binary complexes, MTX binds in two conformational forms and folate binds as a single conformation. Earlier studies on the corresponding complexes with Lactobacillus casei DHFR indicated that, in this case, MTX binds as a single conformation whereas folate binds in multiple conformational forms (both in its binary complex and ternary complex with NADP+); two of the bound conformational states for the folate complexes are very different from each other in that there is a 180 degrees difference in their pteridine ring orientation. In contrast, the two different conformational states observed for MTX bound to E. coli DHFR do not show such a major difference in ring orientation and bind with N1 protonated in both forms. The major difference appears to involve the manner in which the 4-NH2 group of MTX binds to the enzyme (although the same protein residues are probably involved in both interactions). Addition of either NADP+ or NADPH to the E. coli DHFR-MTX complex results in a single set of 13C signals for bound methotrexate consistent with only one conformational form in the ternary complexes.     

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.