X-Ray crystal structures of Candida albicans dihydrofolate reductase: high resolution ternary complexes in which the dihydronicotinamide moiety of NADPH is displaced by an inhibitor

X-ray crystallographic analysis of 5-(4'-substituted phenyl)sulfanyl-2,4-diaminoquinazoline inhibitors in ternary complex with Candida albicans dihydrofolate reductase (DHFR) and NADPH revealed two distinct modes of binding. The two compounds with small 4'-substituents (H and CH3) were found to bind with the phenyl group oriented in the plane of the quinazoline ring system and positioned adjacent to the C-helix. In contrast, the more selective inhibitors with larger 4'-substituents (tert-butyl and N-morpholino) were bound to the enzyme with the phenyl group perpendicular to the quinazoline ring and positioned in the region of the active site that typically binds the dihydronicotinamide moiety of NADPH. The cofactor appeared bound to DHFR but with the disordered dihydronicotinamide swung away from the protein surface and into solution. This unusual inhibitor binding mode may play an important role in the high DHFR selectivity of these compounds and also may provide new ideas for inhibitor design.

Peptide agonists of the thrombopoietin receptor

We have screened a variety of L-amino acid peptide libraries against the extracellular domain of the human thrombopoietin (HuTPO) receptor, c-Mpl. A large number of peptide ligands were recovered and categorized into two families. Peptides from each family compete with the binding of HuTPO and with the binding of peptides from the other familiy. Representative peptides were synthesized and found to activate the full-length HuTPO receptor expressed in Ba/F3 cells to promote proliferation. These peptide families show no apparent homology to the primary sequence of TPO. We have focused our optimization efforts on one of the peptides, a linear 14-mer (IEGPTLRQWLAARA) with an IC50 of 2 nM in a competition binding assay and an EC50 of 400 nM in the proliferation assay. In order to enhance the potency of the compound, we constructed dimeric peptides by linking the carboxy-termini of the 14-mers to a lysine branch. These molecules exhibited slightly higher affinity (0.5 nM) and greatly increased potency (0.1 nM). The EC50 of the dimeric peptide was equivalent to that of the 332 aa form of baculovirus-expressed recombinant HuTPO. As previously shown for the erythropoietin-mimetic peptides, the TPO-mimetic peptides probably activate the TPO receptor by binding and inducing receptor dimerization. This supposition is supported by the observation that covalent dimerization of the peptide enhances its potency by 4,000-fold over that of the monomer. The peptide dimer is also active in stimulating in vitro proliferation of progenitors and maturation of megakaryocytes from human bone marrow, and in promoting an increase in platelet count when administered to normal mice.

Preclinical development of eniluracil: enhancing the therapeutic index and dosing convenience of 5-fluorouracil

Eniluracil (5-ethynyluracil, GW 776, 776C85) is being developed as a novel modulator of 5-fluorouracil (5-FU) for the treatment of cancer. Eniluracil is an effective mechanism-based inactivator of dihydropyrimidine dehydrogenase (DPD), the first enzyme in the catabolic pathway of 5-FU. By temporarily eliminating this prevalent enzyme, eniluracil provides predictable dosing of 5-FU and enables oral administration of 5-FU to replace intravenous bolus and continuously infused dosing. New DPD is synthesized with a half-life of 2.6 days. It also eliminates the formation of problematic 5-FU catabolites. Most importantly, in laboratory animals, eniluracil increases the therapeutic index and absolute efficacy of 5-FU. Accompanying reports in this journal indicate that eniluracil has promising clinical potential.

alpha-fluoro-beta-alanine: effects on the antitumor activity and toxicity of 5-fluorouracil

We have shown previously that (R)-5-fluoro-5,6-dihydrouracil (FUraH(2)) attenuates the antitumor activity of 5-fluorouracil (FUra) in rats bearing advanced colorectal carcinoma. Presently, we found that alpha-fluoro-beta-alanine (FBAL), the predominant catabolite of FUra that is formed rapidly via FUraH(2), also decreased the antitumor activity and potentiated the toxicity of FUra. In rats treated with Eniluracil (5-ethynyluracil, GW776), excess FBAL, in a 9:1 ratio to FUra, produced similar effects when administered 1 hr before, simultaneously with, or 2 hr after FUra. FBAL also decreased the antitumor activity of FUra in Eniluracil-treated mice bearing MOPC-315 myeloma at a 9:1 ratio with FUra, but not at a 2:1 ratio. FBAL did not affect the antitumor activity of FUra in mice bearing Colon 38 tumors. We also evaluated the effect of thymidylate synthase (TS) and thymidine kinase (TK) from tumor extracts after FUra +/- Eniluracil +/- FBAL treatment. The activity of TK was similar among the three groups at both 18 and 120 hr. There was also no difference in TS inhibition ( approximately 35%) at 18 hr. However, significantly more TS inhibition was observed in the Eniluracil/FUra group than in the FUra-alone group at 120 hr. FBAL did not alter the effect of Eniluracil/FUra in TS inhibition. Neither FUraH(2) nor FBAL affected the IC(50) of FUra in culture. Thus, the effect of FBAL did not result from direct competition with FUra uptake or immediate anabolism. Either another downstream catabolite that is not formed in cell culture is the active agent, or the effect requires the complexity of a living organism or an established tumor.

X-ray crystallographic studies of Candida albicans dihydrofolate reductase. High resolution structures of the holoenzyme and an inhibited ternary complex

The recent rise in systemic fungal infections has created a need for the development of new antifungal agents. As part of an effort to provide therapeutically effective inhibitors of fungal dihydrofolate reductase (DHFR), we have cloned, expressed, purified, crystallized, and determined the three-dimensional structure of Candida albicans DHFR. The 192-residue enzyme, which was expressed in Escherichia coli and purified by methotrexate affinity and cation exchange chromatography, was 27% identical to human DHFR. Crystals of C. albicans DHFR were grown as the holoenzyme complex and as a ternary complex containing a pyrroloquinazoline inhibitor. Both complexes crystallized with two molecules in the asymmetric unit in space group P21. The final structures had R-factors of 0.199 at 1.85-A resolution and 0.155 at 1.60-A resolution, respectively. The enzyme fold was similar to that of bacterial and vertebrate DHFR, and the binding of a nonselective diaminopyrroloquinazoline inhibitor and the interactions of NADPH with protein were typical of ligand binding to other DHFRs. However, the width of the active site cleft of C. albicans DHFR was significantly larger than that of the human enzyme, providing a basis for the design of potentially selective inhibitors.

Peptide agonist of the thrombopoietin receptor as potent as the natural cytokine

Two families of small peptides that bind to the human thrombopoietin receptor and compete with the binding of the natural ligand thrombopoietin (TPO) were identified from recombinant peptide libraries. The sequences of these peptides were not found in the primary sequence of TPO. Screening libraries of variants of one of these families under affinity-selective conditions yielded a 14-amino acid peptide (Ile-Glu-Gly-Pro-Thr-Leu-Arg-Gln-Trp-Leu-Ala-Ala-Arg-Ala) with high affinity (dissociation constant approximately 2 nanomolar) that stimulates the proliferation of a TPO-responsive Ba/F3 cell line with a median effective concentration (EC50) of 400 nanomolar. Dimerization of this peptide by a carboxyl-terminal linkage to a lysine branch produced a compound with an EC50 of 100 picomolar, which was equipotent to the 332-amino acid natural cytokine in cell-based assays. The peptide dimer also stimulated the in vitro proliferation and maturation of megakaryocytes from human bone marrow cells and promoted an increase in platelet count when administered to normal mice.

Inhibition of uridine phosphorylase. Synthesis and structure-activity relationships of aryl-substituted 1-((2-hydroxyethoxy)methyl)-5-(3-phenoxybenzyl)uracil

Structure-activity relationship studies on a series of 1-((2-hydroxyethoxy)methyl)-5-(3-(substituted-phenoxy)benzyl)uracils as inhibitors of murine liver uridine phosphorylase have led to compounds with IC50s as low as 1.4 nM. The two most potent compounds, 10j (3-cyanophenoxy) and 11f (3-chlorophenoxy) were tested in vivo for effects on steady-state concentrations of circulating uridine in mice and rats. Both compounds were substantially more efficacious than BAU (5-benzylacyclouridine) both in vitro and in vivo.

Dihydropyrimidine dehydrogenase inactivation and 5-fluorouracil pharmacokinetics: allometric scaling of animal data, pharmacokinetics and toxicodynamics of 5-fluorouracil in humans

The pharmacokinetics of 5-fluorouracil (5-FU) in different animal species treated with the dihydropyrimidine dehydrogenase (DPD) inactivator, 5-ethynyluracil (776C85) were related through allometric scaling. Estimates of 5-FU dose in combination with 776C85 were determined from pharmacokinetic and toxicodynamic analysis.

METHOD

The pharmacokinetics of 5-FU in the DPD-deficient state were obtained from mice, rats and dogs treated with 776C85 followed by 5-FU. The pharmacokinetics of 5-FU in humans were then estimated using interspecies allometric scaling. Data related to the clinical toxicity for 5-FU were obtained from the literature. The predicted pharmacokinetics of 5-FU and the clinical toxicity data were then used to estimate the appropriate dose of 5-FU in combination with 776C85 in clinical trials.

RESULTS

The allometric equation relating total body clearance (CL) of 5-FU to the body weight (B) (CL = 0.47B0.74) indicates that clearance increased disproportionately with body weight. In contrast, the apparent volume of distribution (Vc) increased proportionately with body weight (Vc = 0.58 B0.99). Based on allometric analysis, the estimated clearance of 5-FU (10.9 l/h) in humans with DPD deficiency was comparable to the observed values in humans lacking DPD activity due to genetic predisposition (10.1 l/h), or treatment with 776C85 (7.0 l/h) or (E)-5-(2-bromovinyl)-2'-deoxyuridine (BVdUrd, 6.6 l/h). The maximum tolerated dose (MTD) of 5-FU in combination with 776C85 was predicted from literature data relating toxicity and plasma 5-FU area under the concentration-time curve (AUC). Based on allometric analysis, the estimated values for the MTD in humans treated with 776C85 and receiving 5-FU as a single i.v. bolus dose, and 5-day and 12-day continuous infusions were about 110, 50 and 30 mg/m2 of 5-FU, respectively.

DISCUSSION

The pharmacokinetics of 5-FU in the DPD-deficient state in humans can be predicted from animal data. A much smaller dose of 5-FU is needed in patients treated with 776C85.

Pyrrolo[2,3-d]pyrimidines and pyrido[2,3-d]pyrimidines as conformationally restricted analogues of the antibacterial agent trimethoprim

Conformationally restricted analogues of the antibacterial agent trimethoprim (TMP) were designed to mimic the conformation of drug observed in its complex with bacterial dihydrofolate reductase (DHFR). This conformation of TMP was achieved by linking the 4-amino function to the methylene group by one- and two-carbon bridges. A pyrrolo[2,3-d]pyrimidine, a dihydro analogue, and a tetrahydropyrido[2,3-d]pyrimidine were synthesized and tested as inhibitors of DHFR. One analogue showed activity equivalent to that of TMP against DHFR from three species of bacteria. An X-ray crystal structure of this inhibitor bound to Escherichia coli DHFR was determined to evaluate the structural consequences of the conformational restriction.

High-affinity inhibitors of dihydrofolate reductase: antimicrobial and anticancer activities of 7,8-dialkyl-1,3-diaminopyrrolo[3,2-f]quinazolines with small molecular size

A series of 7,8-dialkylpyrrolo[3,2-f]quinazolines were prepared as inhibitors of dihydrofolate reductase (DHFR). On the basis of an apparent inverse relationship between compound size and antifungal activity, the compounds were designed to be relatively small and compact. Inhibitor design was aided by GRID analysis of the three-dimensional structure of Candida albicans DHFR, which suggested that relatively small, branched alkyl groups at the 7- and 8-positions of the pyrroloquinazoline ring system would provide optimal interactions with a hydrophobic region of the protein. The compounds were potent inhibitors of fungal and human DHFR, with K(i) values as low as 7.1 and 0.1 pM, respectively, and were highly active against C. albicans and an array of tumor cell lines. In contrast to known lipophilic inhibitors of DHFR such as trimetrexate and piritrexim, members of this series of pyrroloquinazolines were not susceptible to P-glycoprotein-mediated multidrug resistance and also showed significant distribution into lung and brain tissue. The compounds were active in lung and brain tumor models and displayed in vivo activity against Pneumocystis carinii and C. albicans.

5-Ethynyluracil (776C85): effects on the antitumor activity and pharmacokinetics of tegafur, a prodrug of 5-fluorouracil

We studied the effects of 5-ethynyluracil (776C85 and 776C), a potent mechanism-based inactivator of dihydropyrimidine dehydrogenase, on the antitumor efficacy and pharmacokinetics of tegafur (FT), a prodrug of 5-fluorouracil (5-FU), in rats with large s.c. colon carcinoma. Rats were dosed p.o. once daily for 7 days with either FT, FT and uracil in a 1:4 molar ratio (UFT), FT 1 h after 776C (776C/FT), or UFT 1 h after 776C (776C/UFT). 776C, which was dosed at 1 mg/kg, had neither intrinsic antitumor activity nor toxicity. The rank order in antitumor efficacy at the maximal tolerated dose of the FT (mg/kg/day) component was 776C/FT (5 mg/kg/day) > or = UFT (80 mg/kg/day) = 776C/UFT (5 mg/kg/day) >> FT (200 mg/kg/day). One-hundred % of rats treated with 776C/FT had complete and sustained tumor regression with no severe toxicity. The area under the plasma 5-FU concentration versus the time curve generated from UFT, FT, and 776C/FT at their maximum tolerated dose was 140, 50, and 27 microM.h, respectively. The area under the concentration in plasma versus time curve did not correlate with the rank order of antitumor efficacy. The vast majority of 5-FU derived from FT (alone) appeared to be rapidly catabolized. Furthermore, plasma exposure of 5-FU derived from UFT was more variable than that from 776C/FT. Each therapy also produced different levels of plasma uracil. Endogenous plasma uracil levels (1-3 microM) were not affected by FT but increased to 100 microM after dosing with 776C. Plasma uracil from UFT was 800 microM 1 h after dosing. These results suggest that moderately elevated uracil (776C/FT) may be beneficial, whereas uracil that is greatly elevated during the first 5 h (UFT) and 5-FU catabolites (FT alone) may interfere with antitumor efficacy. 776C, coadministered with FT, could provide once-a-day oral therapy for cancer patients.

Inhibition of uridine phosphorylase: synthesis and structure-activity relationships of aryl-substituted 5-benzyluracils and 1-[(2-hydroxyethoxy)methyl]-5-benzyluracils

A series of 1-[(2-hydroxyethoxy)methyl]-5-benzyluracils were synthesized and tested for inhibition of murine liver uridine phosphorylase (UrdPase). Inhibitors of UrdPase are reported to enhance the chemotherapeutic utility of 5-fluoro-2'-deoxyuridine and 5-fluorouracil and to ameliorate zidovudine-induced anemia in animal models. We prepared a series of 5-aryl-substituted analogues of 5-benzylacyclouridine (BAU), a good inhibitor of UrdPase (IC50 of 0.46 microM), to develop a compound with enhanced potency and improved pharmacokinetics. The first phase of structure-activity relationship studies on a series of 32 aryl-substituted 5-benzyluracils found several 5-(3-alkoxybenzyl) analogues of 5-benzyluracil with enhanced potency. The acyclovir side chain, the (2-hydroxyethoxy)methyl group, was substituted on the more potent aryl-substituted 5-benzyluracils. The two most potent compounds, 10y (3-propoxy) and 10dd (3-sec-butoxy), were inhibitors of UrdPase with IC50s of 0.047 and 0.027 microM, respectively. Six compounds were tested in vivo for effects on steady-state concentrations of circulating uridine in rats. Plasma uridine levels were elevated 3-9-fold by compound levels that ranged from 8 to 50 microM.

Selective inhibitors of Candida albicans dihydrofolate reductase: activity and selectivity of 5-(arylthio)-2,4-diaminoquinazolines

The recent increase in fungal infections, especially among AIDS patients, has resulted in the need for more effective antifungal agents. In our search for such agents, we focused on developing compounds which inhibit fungal dihydrofolate reductase (DHFR). A series of 25 5-(arylthio)-2,4-diaminoquinazolines were synthesized as potentially selective inhibitors of Candida albicans DHFR. The majority of the compounds were potent inhibitors of C. albicans DHFR and much less active against human DHFR. High selectivity, as defined by the ratio of the I50 values for human and C. albicans DHFR, was achieved by compounds with bulky and rigid 4-substituents in the phenylthio moiety. For example, 5-[(4-morpholinophenyl)thio]-2,4-diaminoquinazoline displayed a selectivity ratio of 540 and was the most selective inhibitor synthesized to date. Substitution in the 2- or 3-position of the 5-phenylthio group provided only marginal selectivity. 6-Substituted-5-[(4-tert-butylphenyl)thio]-2,4-diaminoquinazolines showed potent activity against the C. albicans enzyme but were equally active against human DHFR. Most of the selective compounds were also good inhibitors of C. albicans cell growth, with minimum inhibitory concentration values as low as 0.05 microgram/ mL.

5-Ethynyluracil (776C85): protection from 5-fluorouracil-induced neurotoxicity in dogs

5-Ethynyluracil (776C85) is a potent mechanism-based inactivator of dihydropyrimidine dehydrogenase (DPD), the enzyme that catalyzes the rapid catabolism of 5-fluorouracil (5-FU). Because catabolism is the major route for 5-FU clearance, we studied the effect of 5-ethynyluracil on the pharmacokinetics and toxicity of continuous i.v. 5-FU infusion in the dog. 5-FU at 40 mg/kg/24 hr resulted in a steady-state plasma 5-FU concentration of 1.3 microM and was fatal with dogs dying from apparent neurotoxicity. 5-Ethynyluracil lowered the total clearance of 5-FU from 9.9 to 0.2 L/hr/kg and enabled 1.6 mg/kg/24 hr 5-FU to achieve a steady-state plasma 5-FU concentration of 2.4 microM with no apparent toxicity. 5-FU at 4 mg/kg/24 hr achieved a steady-state plasma 5-FU concentration of 5.3 microM and produced only mild gastrointestinal disturbances in 5-ethynyluracil-treated dogs. Thus, a catabolite of 5-FU appears to be responsible for the 5-FU-induced neurotoxicity in dogs.

5-Ethynyluracil (776C85): a potent modulator of the pharmacokinetics and antitumor efficacy of 5-fluorouracil

5-Ethynyluracil (5-EU, 776C85) is a mechanism-based irreversible inhibitor of dihydropyrimidine dehydrogenase (EC 1.3.1.2), the rate-determining enzyme in 5-fluorouracil (5-FU) catabolism. In the present study, 5-EU was found to be a potent modulator of 5-FU catabolism in mice and rats. Liver extracts prepared up to 6 hr after a 5-EU dose (2 mg/kg) were > 96% inhibited in their ability to catalyze 5-FU degradation. 5-EU treatment increased the elimination t1/2 and the area under the plasma concentration-time curve of 5-FU. 5-FU oral bioavailability was approximately 100% in rats pretreated with 5-EU. Consequently, 5-EU induced a linear relationship between the area under the plasma concentration-time curve and the oral dose of 5-FU. As expected from the preservation of plasma 5-FU, 5-EU potentiated the antitumor activity and the toxicity of 5-FU in two mouse tumor models (Colon 38 and MOPC-315). However, 5-EU potentiated the antitumor activity to a greater degree and thereby increased the therapeutic index of 5-FU 2- to 4-fold.

Basis of selectivity of antibacterial diaminopyrimidines

The basis for the high affinity and selectivity of trimethoprim [2,4-diamino-5-(3',4',5'-trimethoxybenzyl)pyrimidine, TMP] and several close structural analogues is reviewed. Methoxy group substitution on the benzyl group of 2,4-diaminobenzylpyrimidine markedly affects both Escherichia coli dihydrofolate reductase (DHFR) Ki values and in vitro antibacterial activity. TMP is several hundred-fold more potent than the unsubstituted benzylpyrimidine, and the monomethoxy and dimethoxy analogues are of intermediate activity. However, equilibrium dissociation constants determined in the absence of cofactor (NADPH) show that the binding of these diaminobenzylpyrimidines in the enzyme-inhibitor binary complex is considerably weaker and does not vary among the compounds. Thus, the TMP binding affinity of E. coli DHFR is increased by NADPH in the ternary complex, and this increased affinity (cooperativity) varies with methoxy group substitution. In contrast, mouse DHFR has a weaker binding affinity for diaminobenzylpyrimidines, and none of the analogues show strong NADPH cooperative effects. The difference in the magnitude of NADPH/TMP cooperativity between bacterial and mammalian DHFR is an important factor in selectivity. The E. coli enzyme binds TMP more avidly in binary complex, and an additional selectivity factor of 30-fold arises from differences in cooperativity. Although the X-ray crystal structures of bacterial and vertebrate DHFR have been studied extensively, no single hypothesis convincingly explains the molecular basis of TMP selectivity. However, information on the three-dimensional structure of the enzyme has been used to rationally design novel, high-affinity inhibitors.

Refolding of recombinant Pneumocystis carinii dihydrofolate reductase and characterization of the enzyme

The isolation of dihydrofolate reductase (DHFR) cDNA sequences from the messenger RNA of Pneumocystis carinii using the polymerase chain reaction is described. The 206-amino acid P. carinii DHFR was expressed to high levels in Escherichia coli inclusion bodies using the T7 promoter expression system. Solubilization of the inclusion bodies in 4 M guanidine hydrochloride and refolding of the recombinant protein in the presence of 0.5% polyethylene glycol 1450 yielded correctly folded DHFR which was purified to homogeneity by methotrexate-Sepharose affinity chromatography. The refolded enzyme was readily crystallized as a ternary complex with NADPH and various inhibitors. The enzyme exhibited a sharp pH optimum with maximum activity at pH 7.0 (turnover number = 6500 min-1). Km values for dihydrofolate (DHF) and NADPH were 2.3 and 3.0 microM, respectively, in 0.1 m imidazole buffer, pH 7. Folate did not act as a substrate. Comparison of the kinetic properties of the refolded enzyme with soluble P. carinii DHFR expressed at low levels in the T7 expression system showed similar pH-activity profiles, Km values for DHF and NADPH, and IC50 values for several known antifolates which were tested as inhibitors of the enzyme.

Species-dependent differences in the biochemical effects and metabolism of 5-benzylacyclouridine

The pharmacokinetics and biochemical effects of the uridine phosphorylase (UrdPase) inhibitor 5-benzylacyclouridine (BAU) were investigated in the mouse, rat and monkey. Following i.p. administration of BAU (30 mg/kg) in the mouse and i.v. administration in the rat and monkey, initial BAU plasma half-life values were 36, 36 and 25 min, and the areas under the plasma BAU concentration versus time curves (AUC) were 127, 80 and 76 microM.hr, respectively. Rats were also dosed p.o. and i.v. with BAU at 90 mg/kg, and a comparison of the AUC values showed an oral bioavailability of 70%. Analyses of plasma samples by HPLC indicated that the metabolism of BAU differed in these species. A major BAU metabolite was observed in monkeys. Its concentration was greater than or equal to that of BAU in almost every plasma sample, and its elimination paralleled that of BAU. Urinary recovery of the metabolite was 10-fold higher than the recovery of unchanged drug. The compound was identified as the ether glucuronide of BAU by its UV absorption spectrum, its co-elution with BAU after incubation with beta-glucuronidase, and liquid chromatography/mass spectrum analysis. A different metabolite was detected in rat plasma; its maximum concentration was 15% of the BAU level, and its elution position on the HPLC chromatogram was not affected by the action of beta-glucuronidase. BAU had equivalent potency against UrdPase in liver extracts from the three species, with Ki values of about 0.17 microM. However, the in vivo effects of BAU on plasma uridine concentrations were species dependent. In mice, a 30 mg/kg i.p. dose of BAU increased the plasma uridine concentration to 11 microM from a control level of 1.8 microM. In the rat, a 30 mg/kg i.v. dose of BAU increased plasma uridine to 2.1 from 1.1 microM control levels, and a 300 mg/kg oral dose resulted in a peak plasma uridine concentration of only 6 microM. In the monkey, BAU (30 mg/kg, i.v.) had no effect on plasma uridine despite the presence of 10-100 microM BAU levels in plasma for 1.5 hr. These data show that there are significant differences in the biochemical effects and metabolism of BAU in CD-1 mice, CD rats and cynomolgus monkeys.

2,4-Diamino-5-benzylpyrimidines and analogues as antibacterial agents. 12. 1,2-Dihydroquinolylmethyl analogues with high activity and specificity for bacterial dihydrofolate reductase

Twelve 2,4-diamino-5-[(1,2-dihydro-6-quinolyl)methyl]pyrimidines containing gem-dimethyl or fluoromethyl substituents at the 2-position of the dihydroquinoline ring were prepared by condensations of dihydroquinolines with 2,4-diamino-5-(hydroxymethyl)pyrimidine. The dihydroquinolines were produced from the reaction of anilines with mesityl oxide or fluoroacetone. In some cases, 1-aryl-2,4-dimethylpyrroles were obtained as byproducts. Most of these pyrimidines were highly inhibitory to Escherichia coli dihydrofolate reductase (DHFR) and also had high specificity for the bacterial enzyme. 2,4-Diamino-5-[[1,2-dihydro-2,4-dimethyl-3-fluoro-2-(fluoromethyl)-8- methoxy-6(1H)quinolyl]methyl]pyrimidine had an apparent Ki value for E. coli DHFR 13 times lower than that of the control, trimethoprim (1), and was 1 order of magnitude more selective for the bacterial enzyme. It had outstanding activity against Gram-positive organisms in vitro, as well as broad-spectrum antibacterial activity equivalent to that of 1. The results of in vivo testing will be reported elsewhere. The gem-dimethyl substituents of the dihydroquinoline derivatives are considered to be responsible for the high selectivity, as well as contributing to potent bacterial DHFR inhibition. Molecular models are presented which suggest the probable interactions with the bacterial enzyme.

2,4-Diamino-5-benzylpyrimidines as antibacterial agents. 13. Some alkenyl derivatives with high in vitro activity against anaerobic organisms

A series of 2,4-diamino-5-(3,5-dialkenyl-4-methoxy- or -4-hydroxybenzyl)pyrimidines was prepared from [(allyloxy)benzyl]pyrimidines by Claisen rearrangements, and the resulting allyl phenols were further modified by methylation and rearrangement to 1-propenyl analogues. Analogous 3,4-dimethoxy-5-alkenyl derivatives were prepared by similar techniques. High in vitro antibacterial activity was obtained against certain anaerobic organisms, such as Bacteroides species and Fusobacterium, which was equal to or better than the control, metronidazole, in several cases. The profile was similar against Neisseria gonorrhoeae and Staphylococcus aureus. The 3,5-bis(1-propenyl)-4-methoxy derivative 8 was 1 order of magnitude more active against Escherichia coli dihydrofolate reductase than its saturated counterpart, and it was also more active than trimethoprim, 1. However, it was considerably less active in vitro against the Gram-negative organisms. The 3,4-dimethoxy-5-alkenyl, -5-alkyl, and -5-alkoxy analogues had very high broad-spectrum antibacterial activity. However, pharmacokinetic studies of four of the compounds in dogs and rats and in vivo studies with an abdominal sepsis model in rats showed no advantages over trimethoprim.

Characterization of Candida albicans dihydrofolate reductase

Dihydrofolate reductase from Candida albicans was purified 31,000-fold and characterized. In addition, the C. albicans dihydrofolate reductase gene was cloned into a plasmid vector and expressed in Escherichia coli, and the enzyme was purified from this source. Both preparations showed a single protein-staining band with a molecular weight of about 25,000 on sodium dodecyl sulfate-polyacrylamide gel electrophoresis. The enzymes were stable and had an isoelectric point of pH 7.1 on gel isoelectric focusing. Kinetic characterization showed that the enzymes from each source had similar turnover numbers (about 11,000 min-1) and Km values for NADPH and dihydrofolate of 3-4 microM. Like other eukaryotic dihydrofolate reductases, the C. albicans enzyme exhibited weak binding affinity for the antibacterial agent trimethoprim (Ki = 4 microM), but further characterization showed that the inhibitor binding profile of the yeast and mammalian enzymes differed. Methotrexate was a tight binding inhibitor of human but not C. albicans dihydrofolate reductase; the latter had a relatively high methotrexate Ki of 150 pM. The yeast and vertebrate enzymes also differed in their interactions with KCl and urea. These two agents activate vertebrate dihydrofolate reductases but inhibited the C. albicans enzyme. The sequence of the first 36 amino-terminal amino acids of the yeast enzyme was also determined. This portion of the C. albicans enzyme was more similar to human than to E. coli dihydrofolate reductases (50% and 30% identity, respectively). Some key amino acid residues in the C. albicans sequence, such as E-30 (human enzyme numbering), were "vertebrate-like" whereas others, such as I-31, were not. These results indicate that there are physical and kinetic differences between the eukaryotic mammalian and yeast enzymes.

2,4-Diamino-5-benzylpyrimidines and analogues as antibacterial agents. 9. Lipophilic trimethoprim analogues as antigonococcal agents

Lipophilic analogues of trimethoprim (1) bearing 3,5-dialkyl-4-hydroxy substituents in the benzene ring are much more active in vitro against Neisseria gonorrhoeae than is 1. The 3,5-diisopropyl-4-hydroxy derivative (2) was selected as a candidate for clinical evaluation as an antigonococcal agent, and as part of the preliminary evaluation it was submitted to extended pharmacokinetic and metabolism studies in dogs. Although the compound was not extensively conjugated by metabolic enzymes, one of the methyl groups was metabolized to produce a 3-isopropyl-4-hydroxy-5-(alpha-carboxyethyl)benzyl derivative (43), which was rapidly excreted. Related analogues were likewise extensively metabolized.

Methotrexate analogues. 30. Dihydrofolate reductase inhibition and in vitro tumor cell growth inhibition by N epsilon-(haloacetyl)-L-lysine and N delta-(haloacetyl)-L-ornithine analogues and an acivicin analogue of methotrexate

Analogues of methotrexate (MTX) with strong alkylating activity were prepared by replacing the L-glutamate side chain with N omega-haloacetyl derivatives of L-lysine and L-ornithine. Haloacetylation was accomplished in 30-40% yield by reaction of the preformed L-lysine and L-ornithine analogues of MTX with p-nitrophenyl bromoacetate or chloroacetate in aqueous sodium bicarbonate at room temperature. All four haloacetamides were potent inhibitors in spectrophotometric assays measuring noncovalent binding to purified dihydrofolate reductase (DHFR) from L1210 cells. In experiments designed to measure time-dependent inactivation of DHFR from L1210 cells and Candida albicans, the N epsilon-(bromoacetyl)-L-lysine and N delta-(bromoacetyl)-L-ornithine analogues gave results consistent with covalent binding, whereas N epsilon- and N delta-chloroacetyl analogues did not. The N delta-(bromoacetyl)-L-ornithine analogue appeared to be the more reactive one toward both enzymes. Amino acid analysis of acid hydrolysates of the L1210 enzyme following incubation with the bromoacetamides failed to demonstrate the presence of a carboxymethylated residue, suggesting that alkylation had perhaps formed an acid-labile bond. In growth inhibition assays with L1210 cultured murine leukemia cells, the four haloacetamides were all more potent than their nonacylated precursors but less potent than MTX. The greater than 40,000-fold MTX-resistant mutant cell line L1210/R81 was only partly cross-resistant to the haloacetamides. An analogue of MTX with acivicin replacing glutamate was a potent inhibitor of DHFR from chicken liver and L1210 cells but was 200 times less potent than MTX against L1210 cells in culture.

2,4-Diamino-5-benzylpyrimidines as antibacterial agents. 7. Analysis of the effect of 3,5-dialkyl substituent size and shape on binding to four different dihydrofolate reductase enzymes

A group of trimethoprim (TMP) analogues containing 3,5-dialkyl(or halo)-4-alkoxy, -hydroxy, or -amino substitution were analyzed in terms of their inhibitory activities against four dihydrofolate reductase (DHFR) isozymes. Although selectivities were lower than with TMP, the activities against vertebrate DHFR were usually at least 2 orders of magnitude less than against enzymes from microbial sources. However, the profiles of activity were remarkably similar for rat, Neisseria gonorrhoeae, and Plasmodium berghei enzymes in all three series, although somewhat different for Escherichia coli DHFR, leading to the conclusion that the hydrophobic pockets are similar for the first three isozymes. Optimal substitution was reached with 3,5-di-n-propyl or 3-ethyl-5-n-propyl groups. Branching of chains at the alpha-carbon, which resulted in increased substituent thickness, was detrimental to E. coli DHFR inhibition in particular. MR is an inadequate parameter for use in correlating such substituent effects. Conformational changes of the more bulky inhibitors can be invoked to explain some differences in inhibitory pattern. Although log P explains simple substituent effects with the vertebrate DHFRs very well, it is insufficient in the more complex cases described here, where shape is clearly involved as well. Solvent-accessible surface areas were measured for TMP in E. coli and chicken DHFRs, where the coordinates are now known. The environment is more hydrophobic in the latter case; this can also be postulated for rat DHFR, which has a very similar activity profile. As with the mammalian isozymes, N. gonorrhoeae DHFR contains an active site phenylalanine replacing Leu-28 of E. coli DHFR, thus creating a more hydrophobic pocket. A similar replacement may also occur in the P.berghei isozyme. Selectivity for bacterial DHFR is dependent on the nature of the 4-substituent, being low for polar 4-hydroxy compounds but high for polar 4-amino analogues, possibly as a result of solvation differences. With complex substituents, the environment of each atom in the active site must be taken into account to adequately explain structure-activity relationships.

Substrate-inhibitor cooperative interactions with microbial dihydrofolate reductases

Cooperativity in the binding of two substrates to an enzyme is a now well-established phenomenon. The x-ray crystallographic structure of the E. coli DHFR binary TMP complex compared with the ternary enzyme-NADPH-TMP complex suggests without too imaginative extrapolation, that the conformational changes resulting from the binding of one ligand aid in favorably positioning potential binding sites for the second ligand. Of greater importance is the fact that the extent to which inhibitor binding is enhanced by the binding of NADPH varies from species to species. To a significant extent, for example, the selectivity of TMP is enhanced by the increase in its binding to the E. coli enzyme when NADPH is present as compared with several mammalian enzymes. The reverse, negative cooperativity (a decrease in binding of a substance when moving from the binary to a ternary complex), is perhaps less common and certainly less well studied. The present paper deals with one such enzyme, the DHFR from C. albicans, and by reference to another, that from S. cerevisiae, where it is shown that the binding of substrates exhibit strong negative cooperativity. It was of interest also to determine the relationship between inhibitor/NADPH cooperativity and the relative insensitivity of N. gonorrhoeae to TMP. Equilibrium studies show that the binding of TMP in binary complex with this enzyme is exceedingly poor and that a 2,200-fold cooperative effect brings the gonococcal enzyme Ki within one order of magnitude of the E. coli enzyme Ki. Even so, it takes synergism of another sort (with sulfamethoxazole) and high doses to make co-trimoxazole therapy feasible for treating gonorrhoeae. The comparative results on the gonococcal enzyme for a family of near relatives of TMP are of interest also for the reason that the structure-activity relationships with this enzyme are quite different from those of the E. coli and other microbial enzymes. Finally, it should be pointed out that although the negative cooperativity found for the candida and saccharomyces enzymes is relatively large, it is the values of the substrate Michaelis constants that are physiologically relevant. The Km values of the yeast enzymes are within the range for other DHFR and therefore the intracellular activity of the enzymes should not be compromised.

Crystal structure of a novel trimethoprim-resistant dihydrofolate reductase specified in Escherichia coli by R-plasmid R67

Crystalline R67 dihydrofolate reductase (DHFR) is a dimeric molecule with two identical 78 amino acid subunits, each folded into a beta-barrel conformation. The outer surfaces of the three longest beta strands in each protomer together form a third beta barrel having six strands at the subunit interface. A unique feature of the enzyme structure is that while the intersubunit beta barrel is quite regular over most of its surface, an 8-A "gap" runs the full length of the barrel, disrupting potential hydrogen bonds between beta-strand D in subunit I and the adjacent corresponding strand of subunit II. It is proposed that this deep groove is the NADPH binding site and that the association between protein and cofactor is modulated by hydrogen-bonding interactions along one face of this antiparallel beta-barrel structure. A hypothetical model is proposed for the R67 DHFR-NADPH-folate ternary complex that is consistent with both the known reaction stereoselectivity and the weak binding of 2,4-diamino inhibitors to the plasmid-specified reductase. Geometrical comparison of this model with an experimentally determined structure for chicken DHFR suggests that chromosomal and type II R-plasmid specified enzymes may have independently evolved similar catalytic machinery for substrate reduction.

Receptor-based design of dihydrofolate reductase inhibitors: comparison of crystallographically determined enzyme binding with enzyme affinity in a series of carboxy-substituted trimethoprim analogues

By the use of molecular models of Escherichia coli dihydrofolate reductase (DHFR), analogues of trimethoprim (TMP) were designed which incorporated various 3'-carboxyalkoxy moieties in order to acquire ionic interactions with positively charged active-site residues. Certain of these compounds have shown exceptionally high affinity for this enzyme. For example, the 3'-(carboxypentyl)oxy analogue was found to be 55-fold more inhibitory than TMP toward E. coli DHFR (Ki = 0.024 nM vs. 1.32 nM for TMP). X-ray crystallographic studies of E. coli DHFR in binary complexes with TMP and two members of this acid-containing series of compounds defined the binding of these inhibitors and showed the carboxyl group of the latter two inhibitors to be ionically bound to Arg-57. These observations were in agreement with postulated binding modes that were based on receptor modeling.

Species-specific irreversible inhibition of Neisseria gonorrhoeae dihydrofolate reductase by a substituted 2,4-diamino-5-benzylpyrimidine

Neisseria gonorrhoeae dihydrofolate reductase undergoes a time-dependent, irreversible inactivation by 2,4-diamino-5-[3,5-dimethoxy-4-(p-bromoacetamidophenoxy)benzyl] pyrimidine. The kinetics of inactivation are consistent with the reversible formation of an enzyme-inhibitor complex followed by covalent binding to the enzyme. The reversible component is competitive with dihydrofolate and has an inhibitor binding constant of 10 nM. Irreversible inactivation proceeds as a pseudo first-order process with a minimum inactivation half-time of 20 min and a Ki of 28 nM. Using radiolabeled inhibitor, it was shown that approximately 1 mol of ligand was covalently bound to the enzyme/mol of methotrexate binding site when the enzyme was completely inhibited. Radiolabeled inhibitor remained associated with the enzyme following denaturation and sodium dodecyl sulfate-polyacrylamide gel electrophoresis. Cyanogen bromide cleavage of the 14C-labeled enzyme-inhibitor complex yielded only one radioactive polypeptide, and sequence determinations showed that His-25 was modified by covalent attachment of the inhibitor. When dihydrofolate reductases from Lactobacillus casei, Streptococcus faecium, Escherichia coli, SR-1 rodent lymphoma, and chicken liver were tested with the affinity label, only the L. casei enzyme showed a time-dependent increase in inhibition. These data, along with comparisons of known amino acid sequences and x-ray crystal structures, were used to make predictions concerning the three-dimensional conformation of the gonococcal enzyme.

Characterization and amino acid sequence of Neisseria gonorrhoeae dihydrofolate reductase

Dihydrofolate reductase has been purified from a trimethoprim-resistant strain of Neisseria gonorrhoeae. The enzyme showed a single component on sodium dodecyl sulfate-polyacrylamide gel electrophoresis (Mr = 18,000) and on isoelectric focusing in 5 M urea (pI = 6.8). Although gel electrophoresis under nondenaturing conditions resolved the preparation into two enzymatically active proteins (called form 1 and form 2), they were not genetically determined isozymes. Both had a similar dihydrofolate Km (2 microM), NADPH Km (10 microM), and trimethoprim Ki (20 nM), and form 2 (the slower migrating species) was shown to be generated from form 1 by the electrophoresis conditions. The complete covalent structure of the enzyme has also been determined. It is a single polypeptide composed of 162 residues and containing 4 cysteines. The gonococcal dihydrofolate reductase shares a 35% homology with the chicken liver enzyme and a 40% homology with the Escherichia coli enzyme. Most of these identities are residues that have been implicated in the binding of NADPH and methotrexate to the E. coli and Lactobacillus casei reductases.

Kinetics of methotrexate binding to dihydrofolate reductase from Neisseria gonorrhoeae

The kinetics of methotrexate inhibition of dihydrofolate reductase from Neisseria gonorrhoeae have been investigated. Methotrexate was shown to be a tight-binding inhibitor (Kt = 13 pM) competitive with dihydrofolate. However, "stoichiometric" or "pseudoirreversible" inhibition could not be demonstrated. Progress curves of inhibited assays quickly attained steady state regardless of the order of substrate addition, indicating that methotrexate association and dissociation processes were rapid. Kinetic techniques were used to measure the rate of methotrexate dissociation from the enzyme-NADPH-methotrexate ternary complex. At 30 degrees, the first-order off-rate constant (koff) was calculated to be 0.56 min-1. This value is approximately 40-fold greater than the dissociation rate constant of methotrexate for Escherichia coli dihydrofolate reductase. At lower temperatures, progress curves of methotrexate-inhibited gonococcal enzyme assays displayed marked increases in both curvature and the time to reach steady state. At 9 degrees, the methotrexate dissociation rate was slow enough (koff = 0.04 min-1) so that initial velocities of the reaction could be measured, and under these conditions methotrexate inhibition was shown to be "stoichiometric".

Characterization of an R-plasmid dihydrofolate reductase with a monomeric structure

A plasmid-encoded dihydrofolate reductase that originated in a clinical isolate of Salmonella typhimurium (phage type 179) moderately resistant to trimethoprim has been isolated and characterized. The dihydrofolate reductase (called type III) was purified to homogeneity using a combination of gel filtration, hydrophobic chromatography, and methotrexate affinity chromatography. Polyacrylamide gel electrophoresis under denaturing and nondenaturing conditions indicated that the enzyme is a 16,900 molecular weight monomeric protein. Kinetic analyses showed that trimethoprim is a relatively tight binding inhibitor (Ki = 19 nM) competitive with dihydrofolate. The enzyme is also extremely sensitive to methotrexate inhibition (Ki = 9 pM) and has a high affinity for dihydrofolate (Km = 0.4 microM). The sequence of the first 20 NH2-terminal residues of the protein shows 50% homology with the trimethoprim-sensitive chromosomal Escherichia coli dihydrofolate reductase and suggests that the two enzymes may be closely related. This is the first example of a plasmid encoding for a monomeric dihydrofolate reductase only moderately resistant to trimethoprim, and a resistance mechanism, dependent in part on the high dihydrofolate affinity of the type III enzyme, is proposed.

Dihydrofolate reductase hysteresis and its effect of inhibitor binding analyses

Escherichia coli dihydrofolate reductase was shown to follow slow transient kinetics (hysteresis). Nonlinear reaction velocities were detected during the enzyme assay and required 10-15 min to reach a steady-state rate. The degree of hysteresis was influenced by the enzyme concentration and the order of substrate addition. Incubation of the enzyme with NADPH before addition of dihydrofolate resulted in slow initial velocities that increased up to 2-fold during the course of the assay. Increasing the enzyme concentration from 0.2 to 1 nM resulted in diminished hysteresis. NADPH-initiated reactions were linear at all enzyme concentrations tested. Certain drugs had profound effects on hysteresis. Pyrimethamine practically eliminated the hysteresis of dihydrofolate-started reactions, whereas trimethoprime augmented the non-linearities in the sense that hysteresis was detected in both enzyme- and NADPH-started reactions. The shape of these reaction tracings makes trimethoprim is not a slow-binding inhibitor when assayed under conditions that eliminate hysteresis. Contrary to this, sulfamethoxazole did not affect hysteresis or augment inhibition of the enzyme by trimethoprim. Sulfamethoxazole alone (at 6 mM) did not inhibit the hysteresis and allow reliable determinations of Ki values of both weak and tight binding inhibitors. For example, Ki values for pyrimethamine, trimethoprim, and methotrexate were found to be 214 nM, 1.3 nM, and 0.021 nM, respectively.

Effect of a single amino acid substitution on Escherichia coli dihydrofolate reductase catalysis and ligand binding

The two isozymes of dihydrofolate reductase (Forms 1 and 2) from, a Trimethoprim-resistant strain of Escherichia coli (RT500) were separated and purified to homogeneity using a simple procedure based on differential elution from a Methotrexate affinity column. The complete amino acid sequence of the Form 2 isozyme was determined, and it differs from that of Form 1 in only one position. Residue 28 is arginine in Form 2 and leucine in Form 1. However, the isozymes differ greatly in their binding and kinetic properties. Equilibrium dialysis studies showed the Trimethoprim dissociation constants of Form 2 are about 50-fold greater than those of Form 1 in both the binary complex and the ternary complex with NADPH. Similarly, the Methotrexate dissociation constant of Form 2 is about 10-fold greater than that of Form 1. The two isozymes also differ in their turnover numbers at pH 7 (Form 1 is 10-fold more active) and inhibition by divalent cations. Form 1 is extremely sensitive to BaCl2 (50% inhibition at 0.5 mM), whereas Form 2 is much less sensitive (50% inhibition at 60 mM). In the presence of 10 mM BaCl2, Form 1 has the functional characteristics of Form 2. Its turnover number is decreased, its Trimethoprim Ki is increased, and the shape of its pH-activity profile is identical with that of Form 2. The x-ray structures and amino acid sequences of several bacterial dihydrofolate reductases indicate that Asp-27 is important in inhibitor binding and may be involved in catalysis. The present data provide kinetic evidence for this hypothesis, and it is proposed that almost all the unusual characteristics of Form 2 are the direct result of a charge interaction between Arg-28 and Asp-27. A similar interaction between Ba2+ and the Asp-27 of Form 1 can result in an enzyme complex that is kinetically similar to Form 2.

R plasmid dihydrofolate reductase with subunit structure

Dihydrofolate reductase, specified by the type II plasmid of a trimethoprim-resistant Escherichia coli, was purified 40-fold to homogeneity using a combination of gel filtration, DEAE-Sephacel chromatography, and hydrophobic chromatography. The final product shows a single protein band on polyacrylamide gel electrophoresis and has a specific activity of 1.0 unit/mg. The molecular weight of the purified enzyme is 36,000 as determined both by gel filtration and Ferguson analysis of polyacrylamide gel electrophoresis. In contrast, a single polypeptide with a molecular weight of 8,500 was observed on sodium dodecyl sulfate-gel electrophoresis. These experiments suggest that, unlike any bacteria or vertebrate dihydrofolate reductase previously examined, the type II R plasmid reductase is a tetramer composed of four identical subunits. A partial amino acid sequence determination shows no heterogeneity of the subunits and also no clear homology with any reductase sequence previously reported.

Dihydrofolate reductase from Neisseria sp

Members of the genus Neisseria are relatively nonsusceptible to trimethoprim, an inhibitor of dihydrofolate reductase. For example, the minimal inhibitory concentration (MIC) of trimethoprim for N. gonorrhoeae ranges from 2 to 70 mug/ml, whereas the MIC for Escherichia coli is 0.2 mug/ml or less. In an effort to understand this difference, dihydrofolate reductase was partially purified from five Neisseria species and compared with the enzyme from E. coli. N. gonorrhoeae dihydrofolate reductase was similar to that from E. coli in molecular weight (18,000) and affinity for the substrates reduced nicotinamide adenine dinucleotide phosphate and dihydrofolate (K(m) = 13 and 8 muM, respectively). However, the gonococcal enzyme had a decreased affinity for trimethoprim, with an apparent K(i) of 45 x 10(-9) M, some 30-fold greater than the E. coli value of 1.2 x 10(-9) M. These enzymes also differed in their isoelectric points and pH activity profiles. Within the genus Neisseria, the dihydrofolate reductase isolated from N. meningitidis and N. lactamica resembled the N. gonorrhoeae enzyme, and only small differences were detected for the N. flavescens and Branhamella catarrhalis dihydrofolate reductases. These data indicate that the relatively poor affinity of trimethoprim for the dihydrofolate reductase from these organisms may be largely responsible for the relative nonsusceptibility of Neisseria sp. to trimethoprim. The contribution of other resistance mechanisms to the overall nonsusceptibility was assessed. Strains of N. gonorrhoeae with altered cell envelope permeability had MIC values less than twofold different from those of isogenic wild-type strains. Also, a direct relationship was observed between the affinity of trimethoprim analogs for gonococcal dihydrofolate reductase and the MIC of these compounds for the gonococcus. These observations suggest that the cell envelope of N. gonorrhoeae is not impermeable to trimethoprim. Changes in the amount of dihydrofolate reductase activity could cause alterations in the susceptibility of the gonococcus to trimethoprim, as demonstrated with N. gonorrhoeae strains selected for trimethoprim resistance after chemical mutagenesis. However, the level of dihydrofolate reductase activity in wild-type N. gonorrhoeae was similar to that of E. coli, indicating that the difference in the susceptibility of these organisms is not due to greater amounts of enzyme in N. gonorrhoeae.

Escherichia coli dihydrofolate reductase: isolation and characterization of two isozymes

A combination of affinity column chromatography and preparative gel electrophoresis has been used to purify to homogeneity the two isozymes of dihydrofolate reductase from a trimethoprim-resistant strain of Escherichia coli B (RT 500). These enzyme forms are noninterconvertible and are present in crude cell lysates, but other electrophoretic species can be generated durng purification if sulfhydryl-protecting agents, such as dithiothreitol, are not present. The two isozymes, numbered form 1 and form 2 with respect to their decreasing electrophoretic mobilities, have similar molecular weights (18 500), molecular radii (21 A), and apparent Km values for reduced nico inamide adenin- dinucleotide (NADH) and NADH phosphate (NADPH). Both forms contain 2 mol of sulfhydryl/mol of enzyme which can be oxidized to intramolecular disulfide bonds. However, forms 1 and 2 differ physically in their electrophoretic mobility and isoelectric point and kinetically in their pH-activity profile, specific activity, Km for dihydrofolate, and their affinity toward a number of inhibitors.