Methotrexate, aminopterin, and folate have been synthesized with 90% enrichment of C-2 with 13C. 13C nuclear magnetic resonance has been used to examine the state of protonation of the pteridine ring of these compounds under various conditions and gives much more clear-cut results than most other methods. For the free compounds the following pK values were obtained: methotrexate, 5.73 +/- 0.02 (N-1); aminopterin, 5.70 +/- 0.03 (N-1); folic acid, 2.40 (N-1) and 8.25 +/- 0.05 (N-3, O-4 amide group). The state of protonation of these compounds when complexed to dihydrofolate reductase (isoenzyme 2 from Streptococcus faecium) was also studied over the pH range 6--10. The resonance from bound methotrexate showed a constant chemical shift over the whole pH range studied, and it is inferred that in the complex the pteridine ring remains protonated to at least pH 10. The same result was obtained for the binary complex of aminopterin with the reductase and for either methotrexate or aminopterin in ternary complex with reductase and NADPH4. The latter is an inhibitor of the reductase competitive with NADPH. However, folate bound to the reductase in either the binary or the ternary complex shows the same protonation behavior as in the free state. The data indicate that the association constant for binding of methotrexate is increased enough when protonation of N-1 occurs to account for the enhanced binding of methotrexate as compared with folate.

No abstract

Methionine has been incorporated with high efficiency by Streptococcus faecium var. Durans strain A into dihydrofolate reductase isoenzyme 2. In the l3C NMR spectrum of the purified enzyme the resonances corresponding to the seven methionine residues are partially resolved into three composite peaks. Denaturation with urea collapses these into a single peak centered at 15.32 ppm, whereas the resonance of free methionine is at 15.04 ppm. Spectra of the free enzyme, its complex with methotrexate, and its complex with methotrexate and reduced nicotinamide adenine dinucleotide phosphate (NADPH) have been simulated, permitting more accurate estimates of line widths and nuclear Overhauser enhancement (NOE) values. These, together with the T\ values, cannot be explained solely by the effects of macromolecular tumbling and very rapid rotation of the methionine methyl group about its axis. A model assuming, in addition, the occurrence of free rotation about the methionine CHi-S bond is also unsatisfactory, and it is concluded that internal rotation about the CH2-S bond is highly restricted so that the methyl group oscillates through a relatively narrowangular range. Complex formation with NADPH produced rather small changes in the spectrum of the native enzyme, probably 4 he structure of dihydrofolate reductase is currently under intensive study in a number of laboratories. Interest in this enzyme derives in part from its clinical relevance. The reductase is the target of methotrexate, a drug used in the treatment of various types of cancer, and bacterial reductase is strongly and specifically inhibited by trimethoprim which is now used extensively in combination with sulfa drugs for treating a variety of infectious diseases.Although studies of the amino acid sequence, the effects of chemical modification, and x-ray diffraction of the crystalline protein have made important contributions to understanding the structure of the reductase, NMR studies are able to make a unique contribution. The method does not perturb the structure of the enzyme and is potentially capable of giving information about the internal motions of amino acid residues and about the interactions of residues at the active site with a great variety of substrates and inhibitors in binary or ternary f From the

Dihydrofolate reductase isozyme 2 of Streptococcus faecium has been labeled with 13C in the C gamma position of tryptophan residues by growing the organism on a defined medium containing L-[gamma-13C]tryptophan (90% 13C). The 13C nuclear magnetic resonance (NMR) spectrum of the enzyme shows four well-resolved resonances which have nuclear Overhauser enhancements of 1.1-1.3. Values of T1 (spin-lattice relaxation time) and T2 (spin-spin relaxation time) are significantly less than predicted for an isotropically rotating, rigid sphere with no intermolecular dipole-dipole interactions. Three of the resonances have chemical shifts downfield from the 13C resonance of urea-denatured enzyme by amounts up to 1.43 ppm. The chemical shift of resonance 4 in the spectrum is 4.0 ppm upfield from Trp C gamma of urea-denatured enzyme. This large upfield shift is attributed to electric field effects generated by polar side chains. The two more upfield peaks both provide evidence that the corresponding tryptophan residues, WC and WD, each undergo chemical exchange between alternative microenvironments. In the case of WC, which gives a resonance with two components, exchange is slow (ve, exchange rate much less than 55 s-1), and the relative populations of the two stable states are in the ratio 2:3. WD is apparently in intermediate to fast exchange on the NMR time scale. With a two-state model, ve increases from approximately 90 to 150 s-1 as the temperature is increased from 5 to 25 degrees C. This increases in temperature is also accompanied by an increase in the fractional population of the minor downfield state(s), from about 0.062 at 5 degrees C to 0.24 at 25 degrees C. However, the data may also be interpreted as a temperature-dependent equilibrium between a continuum of many states. WD is tentatively identified with Trp-22 since comparison of the sequences of Lactobacillus casei dihydrofolate reductase and S. faecium dihydrofolate reductase and inspection of the crystal structure of the L. casei enzyme indicate that Trp-6, Trp-115, and Trp-160 are probably all involved in regions of beta sheet whereas Trp-22 is in a loop joining beta A to alpha B. Earlier crystallographic evidence for the Escherichia coli reductase suggests that in the methotrexate complex with this enzyme the corresponding loop has a good deal of flexibility. It is probable that in the uncomplexed S. faecium reductase the motion of this loop is the major mechanism for the exchange process involving Trp-22. The upfield chemical shift of resonance 4 is attributed to electric field effects on C gamma of Trp-22 produced by the carboxylate groups of Asp-27 and Asp-9. On the basis of the small difference between the chemical shift of resonance 3 and that of tryptophan C gamma in urea-denatured reductase, it is suggested that WC may be identified with Trp-6.

No abstract