5'-Haloacetamido-5'-deoxythymidines: novel inhibitors of thymidylate synthase

5'-Bromoacetamido-5'-deoxythymidine (BAT), 5'-iodoacetamido-5'-deoxythymidine (IAT), 5'-chloroacetamido-5'-deoxythymidine (CAT) and [14C]BAT were synthesized and their interactions with thymidylate synthase purified from L1210 cells were investigated. The inhibitory effects of these compounds on thymidylate synthase were in the order BAT greater than IAT greater than CAT, which is in agreement with their cytotoxic effects in L1210 cells. In the presence of substrate during preincubation, the concentration required for 50% inhibition of the enzyme activity by these inhibitors was 4-8-fold higher than it was in the absence of dUMP. The I50 values for BAT were 1 X 10(-5) M and 1.2 X 10(-6) M in the presence and absence, respectively, of dUMP during preincubation. These results were in agreement with the observed inhibition of thymidylate synthase by BAT in intact L1210 cells. A Lineweaver-Burk plot revealed that BAT behaved as a competitive inhibitor. The Km for the enzyme was 9.2 microM, and the Ki determined for competitive inhibition by BAT was 5.4 microM. Formation of a tight, irreversible complex is inferred from the finding that BAT-inactivation of thymidylate synthase was not reversible on prolonged dialysis and that the enzyme-BAT complex was nondissociable by gel filtration through a Sephadex G-25 column or by TSK-125 column chromatography. Incubation of thymidylate synthase with BAT resulted in time-dependent, irreversible loss of enzyme activity by first-order kinetics. The rate constant for inactivation was 0.4 min-1, and the steady-state constant of inactivation, Ki, was estimated to be 6.6 microM. The 5'-haloacetamido-5'-deoxythymidines provide specific inhibitors of thymidylate synthase that may also serve as reagents for studying the enzyme mechanism.     

Interaction of 5-fluoro-4-thio-2'-deoxyuridine 5'-phosphate with mammalian tumour thymidylate synthase: role of the pyrimidine N(3)-H dissociation

The role of the pyrimidine N(3)-H in binding of dUMP derivatives to thymidylate synthase was evaluated with the aid of a new dUMP analogue, 5-fluoro-4-thio-dUMP, synthesized by an improved thiation and enzymatic phosphorylation. The interaction of this analogue, and of 5-FdUMP, with the enzyme, and the pH-dependence of these interactions, were compared. Both were slow-binding competitive inhibitors of the enzyme from Ehrlich carcinoma, L1210 and CCRF-CEM cells, with Ki an order of magnitude higher for 5-fluoro-4-thio-dUMP than for 5-FdUMP. With both nucleotides, as well as the parent nucleosides, enzyme inactivation increased as the pH was lowered from 8 to 6. Maximum inactivation with 5-FdUrd was at pH 7.0, and with 5-fluoro-4-thio-dUrd at pH 6.0, in agreement with the higher pKa for the N(3)-H dissociation of the former, and pointing to participation of the N(3)-H as a hydrogen donor in binding to the enzyme.     

Strategies for the measurement of the inhibitory effects of thymidine analogs on the activity of thymidylate synthase in intact murine leukemia L1210 cells

Two strategies have been pursued to monitor the inhibition of thymidylate (dTMP) synthase (5,10-methylenetetrahydrofolate:dUMP C-methyltransferase, EC 2.1.1.45) by thymidine (dThd) analogs in intact murine leukemia L1210 cells. The first method was based on the determination of tritium release from 2'-deoxy[5-3H]uridine [( 5-3H]dUrd) or 2'-deoxy[5-3H]cytidine [( 5-3H]dCyd); the second method was based on an estimation of the amount of dCyd incorporated into DNA as dTMP. The validity of these procedures was assessed by evaluating the inhibition of thymidylate synthase in murine leukemia L1210 cells by a series of 18 dThd analogs. There was a strong correlation between the inhibitory effects of the dThd analogs on the proliferation of L1210 cells on the one hand, and (i) their inhibitory effects on tritium release from [5-3H]dCyd (r = 0.926) and (ii) their inhibitory effects on the incorporation of dCyd into DNA dTMP (r = 0.921), on the other hand. Evaluation of tritium release from [5-3H]dCyd proved to be the most convenient method that has been described so far to measure thymidylate synthase activity and to follow the inhibitory effects of thymidylate synthase inhibitors in intact L1210 cells, since this method is rapid and very sensitive, and since it proved superior to the evaluation of tritium release from [5-3H]dUrd because it circumvents possible interactions of the inhibitors with thymidine kinase activity.     

Purification and properties of mouse thymus thymidylate synthase. Comparison of the enzyme from mammalian normal and tumour tissues

Mouse thymus thymidylate synthase has been purified to apparent electrophoretic homogeneity and compared with the enzyme from mouse tumour L1210 and Ehrlich ascites carcinoma cells. The enzyme is a dimer composed of 35,000 mol. wt monomers. Mouse thymus and tumour enzymes exhibit allosteric properties reflected by cooperative binding of both dUMP and 5-fluoro-dUMP. Activation energy for the reaction, catalyzed by thymidylate synthase from mouse tumour but not from mouse thymus, lowers at temperatures above 34 degrees C, reflecting a change of rate-limiting step in dTMP formation. MgATP at millimolar concentrations inhibits mouse thymus enzyme.     

Synthesis and biological activity of N(alpha)-[4-[N-[(3,4-dihydro-2-methyl-4-oxo-6-quinazolinyl)methyl]-N-propargylamino]phenylacetyl]-L-glutamic acid

2-Deamino-2-methyl-N10-propargyl-5,8-dideazafolic acid (ICI 198583) is a potent inhibitor of thymidylate synthase. Its analogue, N(alpha)-[4-[N-[(3,4-dihydro-2-methyl-4-oxo-6-quinazolinyl)methyl]-N-propargylamino]phenylacetyl]-L-glutamic acid, containing p-aminophenylacetic acid residue substituting p-aminobenzoic acid residue, was synthesized. The new analogue exhibited a moderately potent thymidylate synthase inhibition, of linear mixed type vs. the cofactor, N(5,10)-methylenetetrahydrofolate. The Ki value of 0.34 microM, determined with a purified recombinant rat hepatoma enzyme, was about 30-fold higher than that reported for inhibition of thymidylate synthase from mouse leukemia L1210 cells by ICI 198583 (Hughes et al., 1990, J. Med. Chem. 33, 3060). Growth of mouse leukemia L5178Y cells was inhibited by the analogue (IC50 = 1.26 mM) 180-fold weaker than by ICI 198583 (IC50 = 6.9 microM).     

A convenient procedure for purification of thymidylate synthase from L1210 cells

Several laboratories have described procedures for purification of thymidylate synthase (TMP synthase) that utilize folates or folate analogs covalently attached to a matrix. The principle of separation is the formation of a ternary complex between dUMP, TMP synthase, and the bound ligand and the subsequent elution of the enzyme with buffers that do not contain dUMP. We have successfully used 10-formylfolic acid as the bound ligand for the purification of TMP synthase. As compared to other ligands that have been used, 10-formylfolic acid has the advantages that it can be easily synthesized, it is stable, and the enzyme is eluted as a sharp peak. Application of this procedure to L1210 leukemia cells gave 1765-fold purification of TMP synthase with a recovery of 39%. The native enzyme had a molecular weight of 78,000, which is about the same as that reported.     

5′-Haloacetamido-5′-deoxythymidines: novel inhibitors of thymidylate synthase

5′-Bromoacetamido-5′-deoxythymidine (BAT), 5′-iodoacetamido-5′-deoxythymidine (IAT), 5′-chloroacetamido-5′-deoxythymidine (CAT) and [¹⁴C]BAT were synthesized and their interactions with thymidylate synthase purified from L1210 cells were invesatigated. The inhibitory effects of these compounds on thymidylate synthase were in the order BAT > IAT > CAT, which is in agreement with their cytotoxic effects in L1210 cells. In the presence of substrate during preincubation, the concentration required for 50% inhibition of the enzyme activity by these inhibitors was 4–8 fold higher than it was in the absence of dUMP. The I₅₀ values for BAT were 1·10⁻⁵ M and 1.2·10⁻⁶ M in the presence and absence, respectively, of dUMP during preincubation. These results were in agreement with the observed inhibition of thynmidylate synthase by BAT in intact L1210 cells. A Lineweaver-Burk plot revealed that BAT behaved as a competitive inhibitor. The K_m for the enzyme was 9.2 μM, and the K_i determined for competitive inhibition by BAT was 5.4 μM. Formation of a tight, irreversible compledx is referred from the finding that BAT-inactivation of thymidylate synthase was not reversible on prolonged dialysis and that the enzyme-BAT complex was nondissociable by gel filtration through a Sephadex G-25 column or by TSK-125 column chromatography. Incubation of thymidylate synthase with BAT resulted in time-dependent, irreversible loss of enzyme activity by first-order kinetics. The rate constant for inactivation was 0.4 min⁻¹, and the steady-state constant of inactivation, K_i, was estimated to be 6.6 μM. The 5′-haloacetamido-5′-deoxythymidines provide specific inhibitors of thymidylate synthase that may also serve as reagents for studying the enzyme mechanism.     

Mechanism of inhibition of mammalian tumor and other thymidylate synthases by N4-hydroxy-dCMP, N4-hydroxy-5-fluoro-dCMP, and related analogues

N4-Hydroxy-dCMP (N4-OH-dCMP), N4-methoxy-dCMP (N4-OMe-dCMP), and their 5-fluoro congeners (syntheses of which are described) were all slow-binding inhibitors of Ehrlich carcinoma thymidylate synthase (TS), competitive with respect to dUMP, and had differing kinetic constants describing interactions with the two TS binding sites. N4-OH-dCMP was not a substrate (no dihydrofolate produced; no tritium released with 5-3H-labeled molecule), and its inactivation of TS was methylenetetrahydrofolate-dependent, hence mechanism-based, with arrest of a step posterior to addition of cofactor and blocking abstraction of the C(5) hydrogen. Ki values for N4-OH-dCMP and its 5-fluoro analogue were in the range 10(-7) - 10(-8) M, 2-3 orders of magnitude higher for the corresponding N4-OMe analogues. The 5-methyl analogue of N4-OH-dCMP was 10(4)-fold less potent, pointing to the anti rotamer of the imino form of exocyclic N4-OH, relative to the ring N(3), as the active species. This is consistent with weaker slow-binding inhibition of the altered enzyme from 5-FdUrd-resistant, relative to parent, L1210 cells by both FdUMP and N4-OH-dCMP, suggesting interaction of both N4-OH and C(5)-F groups with the same region of the active center. Kinetic studies with purified enzyme from five sources, viz., Ehrlich carcinoma, L1210 parental, and 5-FdUrd-resistant cells, regenerating rat liver, and the tapeworm Hymenolepis diminuta, demonstrated that addition of a 5-fluoro substituent to N4-OH-dCMP increased its affinity from 2- to 20-fold for the enzyme from different sources.(ABSTRACT TRUNCATED AT 250 WORDS)     

Thymidylate synthetase from Escherichia coli K12. Purification, and dependence of kinetic properties on sugar conformation and size of the 2' substituent.

Thymidylate synthetase from Escherichia coli K12 has been purified 3600-fold by a series of chromatographic procedures. The final preparation had a specific activity of 1.47 units/mg protein and was approximately 80% pure. The enzyme is a dimer of relative molecular mass, Mr, 64000 composed of two subunits of Mr 32,000 each. Its isoelectric point is 4.7 and it is stimulated by Mg2+. Michaelis constants for (+)5,10-methylene-5,6,7,8-tetrahydrofolate [(+)CH2H4folate] were 0.014 mM in the case of methylation of 2'-deoxyuridine-5'-phosphate (dUMP) and 0.55 mM when it served as methyl-group donor for 2'-fluoro-2'-deoxyuridine-5'-phosphate (dUflMP); the corresponding Km values for dUMP and dUflMP were 0.01 mM and 0.11 mM, respectively. The activation energies for the two reactions were found to be 72.8 kJ/mol (methylation of dUMP) and 66.1 kJ/mol (methylation of dUflMP). The data support a recognition mechanism between thymidylate synthetase and that fraction of the nucleotide the sugar moiety of which is in the 2'-endo-3'-exo conformation.     

Characteristics of thymidylate synthase purified from a human colon adenocarcinoma

Thymidylate synthase has been purified greater than 4000-fold from a human colon adenocarcinoma maintained as a xenograft in immune-deprived mice. In this disease, the enzyme is an important target for the cytotoxic action of 5-fluorouracil, which is influenced by the reduced folate substrate CH2-H4PteGlu. Due to the importance of this interaction, and the existence in cells of folate species as polyglutamyl forms, the interaction of folylpolyglutamates with thymidylate synthase was examined. Polyglutamates of PteGlu were used as inhibitors, and the interaction of CH2-H4PteGlu polyglutamates as substrates or in an inhibitory ternary complex were also examined. Using PteGlu1-7, Ki values were determined. A maximal 125-fold decrease in Ki was observed between PteGlu1 and PteGlu4; further addition of up to three glutamyl residues did not result in an additional decrease in Ki. Despite the increased binding affinity of folypolyglutamates for this enzyme, no change in the Km values for either dUMP (3.6 microM) or CH2-H4PteGlu (4.3 microM) were detected when polyglutamates of [6R]CH2-H4PteGlu were used as substrates. Product inhibition studies demonstrated competitive inhibition between dTMP and dUMP in the presence of CH2-H4PteGlu5. In addition, CH2-H4PteGlu4 stabilized an inhibitory ternary complex formed between FdUMP, thymidylate synthase, and CH2-H4PteGlu4. Thus the data do not support a change in the order of substrate binding and product release upon polyglutamylation of CH2-H4PteGlu reported for non-human mammalian enzyme. This is the first study to characterize kinetically thymidylate synthase from a human colon adenocarcinoma.     

Site-directed mutagenesis of arginine 179 of thymidylate synthase. A nonessential substrate-binding residue

X-ray structural studies have shown that Arg-179 of thymidylate synthase is complexed to bound inorganic phosphate or to the 5'-phosphate of the bound substrate dUMP. The importance of Arg-179 to the structure/function of thymidylate synthase is also indicated by its complete conservation among the 17 thymidylate synthases thus far sequenced. In the present work, Arg-179 has been replaced by Thr, Ala, Lys, and Glu using site-directed mutagenesis with a mixture of four synthetic oligonucleotides as primers. The mutant proteins complement thymidylate synthase-deficient Escherichia coli and show high enzyme activity. Each of these mutants has been purified to homogeneity, partially sequenced to verify the mutation, and has had its steady state kinetic parameters determined. The most significant effect of all mutations is localized to a decrease in the net rate of association of thymidylate synthase with dUMP; the Lys mutant also shows an apparent increase in the dissociation constant of the folate cofactor of the reaction. The high activity in the mutant enzymes is explained by "plasticity" of the enzyme and compensatory actions of the other Arg residues. Why the Arg-179 residue has been conserved during evolution remains an open question.     

Adducts of 3'-azido-2,3'-dideoxythymidine 5'-phosphate or 5'-phosphonate as inhibitors of cytopathic effect and transformation of cells under the influence of retroviruses in cell culture]

Inhibition of HIV-1- or HIV-2-induced cytopathicity and (Moloney) murine sarcoma virus (MSV)-induced cell transformation by amino acid and amino alcohol adducts of either 3'-azido-2',3'-dideoxythymidine 5'-monophosphate (AZTMP) or 5'-hydrogenphosphonate (AZTHP) were investigated. Both types of nucleotide adducts inhibited replication of HIV-1 and HIV-2 in MT-4 cells at a 1.5- to 3-fold higher EC50 (50% effective concentration) than AZT; and, also, selectivity indexes of these adducts were approximately 1.5 to 3-fold lower than that of AZT. The activity of the AZTMP and AZTHP adducts against MSV-induced transformation of C3H/3T3 cells was equal to or only slightly inferior than that of AZT, but their toxicity was 10-fold lower, so that their selectivity indexes were 2- to 7-fold higher. The nature of the aminoacyl component of the adducts significantly influence the antiretroviral activity of the test compounds.     

Suppression of the human immunodeficiency virus in cultured cells by 5'-phosphites of 3'-azido-2',3'-dideoxynucleosides]

5'-Phosphites (5'-hydrogenphosphonates) of 3'-azido-2'-, 3'-dideoxynucleosides are shown to be effective inhibitors of the human immunodeficiency virus (HIV-1) in MT4 cell culture. 5'-Phosphite of 3'-azido-2', 3'-dideoxythymidine was the most active among these compounds and even a little more active as compared to the well-known anti-AIDS drug 3'-azido-2',3'-dideoxythymidine; at the same time 5'-phosphites of 3'-azido-2',3' -dideoxynucleosides with adenine, guanine and cytosine bases were more active than the corresponding nucleosides. The toxicity of all four phosphites was comparatively low and the equimolar mixture of all four phosphites was 2-3 fold less toxic than each of them separately. Data on the decreased toxicity of the phosphite mixture are explained from the viewpoint of a decreased pool disbalance of natural 2'-deoxynucleoside 5'-triphosphates in cells; a significant pool disbalance is developed in the case of 3'-azido-2',3'-dideoxythymidine action.     

Database searching for thymidine and thymidylate kinase inhibitors using three-dimensional structure-based methods

Structure-based drug design methods were used to search for novel inhibitors of herpes simplex virus type 1 (HSV-1) thymidine kinase and Mycobacterium tuberculosis thymidylate kinase. The method involved the use of crystal structure complexes to guide database searching for potential inhibitors. A number of weak inhibitors of HSV-2 were identified, one of which was found to inhibit HSV-1 TK and HSV-1 TK-deficient viral strains. Each compound tested against M. tuberculosis thymidylate kinase was found to have some activity. The best of these compounds was only 4.6-fold less potent than 3'-azido-3'-deoxythymidine-5'-monophosphate (AZTMP). This study demonstrates the utility of structure-based drug design methods in the search for novel enzyme inhibitors.     

Kinetic properties of human thymidylate synthase,an anticancer drug target

We have determined the kinetic parameters of human recombinant thymidylate synthase (hrTS) with its natural substrate, dUMP, and E-5-(2-bromovinyl)-2(')-deoxyuridine monophosphate (BVdUMP), a nucleotide derivative believed to be the active species of the novel anticancer drug NB1011. NB1011 is activated by hrTS and is selectively toxic to high thymidylate synthase expressing tumor cells. BVdUMP undergoes hrTS-catalyzed thiol-dependent transformation. dUMP and BVdUMP act as competitive hrTS substrates. The natural folate cofactor, CH(2)-THF, inhibits the TS-catalyzed reaction with BVdUMP. We suggest that lower folate levels found in tumor cells favor TS-catalyzed BVdUMP transformation, which, in addition to higher levels of TS expression in tumor cells, contributes to the favorable therapeutic index of the drug NB1011.     

Synthesis and anti-HIV activity of beta-D-3'-azido-2',3'-unsaturated nucleosides and beta-D-3'-azido-3'-deoxyribofuranosylnucleosides

Since the discovery of 3'-azido-3'-deoxythymidine (AZT) and 2',3'-didehydro-2',3'-dideoxythymidine (d4T) as potent and selective inhibitors of the replication of human immunodeficiency virus (HIV), there has been a growing interest for the synthesis of 2',3'-didehydro-2',3'dideoxynucleosides with electron withdrawing groups on the sugar moiety. Here we described an efficient method for the synthesis of such nucleoside analogs bearing structural features of both AZT and d4T The key intermediate, 3-azido-1,2-bis-O-acetyl-5-O-benzoyl-3-deoxy-D-ribofuranose, 5 was synthesized from commercially available D-xylose in five steps, from which a series of pyrimidine and purine nucleosides were synthesized in high yields. The resultant protected nucleosides were converted to target nucleosides using appropriate chemical modifications. The final nucleosides were evaluated as potential anti-HIV agents.