Identification of a novel dimeric phosphoprotein (PP29/30) associated with signaling receptors in human T lymphocytes and natural killer cells

Two-dimensional gel electrophoresis of in vitro phosphorylated proteins coprecipitated by CD2 monoclonal antibody (mAb) from Brij58 lysates of resting human T lymphocytes and natural killer (NK) cells resulted in the identification of a novel 29/30-kD disulfide-linked dimer (pp29/30). Comparative two-dimensional analysis of CD2, CD3, CD4, CD5, and CD8 immunoprecipitates revealed that pp29/30 associates with these signaling receptor complexes but not with CD18, CD27, and CD29 in human T lymphocytes. Analysis of CD2 immunoprecipitates prepared from T cell antigen receptor/CD3-modulated T lymphocytes indicated that pp29/30 preferentially associates and comodulates with the human T cell antigen receptor (TCR). Since tyrosine phosphorylated pp29/30 selectively interacts with the Src homology type 2 domains (SHZ) of the protein tyrosine kinases p56lck and p59fyn but not ZAP70 the present data suggest that pp29/30 represents a novel signaling receptor associated phosphoprotein likely involved in the activation of human T lymphocytes and NK cells.

5,10-Methylenetetrahydro-5-deazafolic acid and analogues: synthesis and biological activities

The synthesis of 5,10-methylene-5-deazatetrahydrofolic acid (2), a stable, rigid analogue of 5,10-methylenetetrahydrofolate (1), is reported as a potential inhibitor of thymidylate synthase. The target compound was obtained by a Fisher-indole type cyclization of the hydrazone 16 from 2-amino-6-hydrazino-4-oxopyrimidine (10) and diethyl N-[4-(3-formyl-1-pyrrolyl)benzoyl]-L-glutamate (15) followed by catalytic reduction of the product 17. Similarly, modification of the Fisher-indole type cyclization of the appropriate hydrazone precursors 11 and 12 afforded the nonclassical analogues 3-amino-7,8,9-trimethyl-2H-pyrrolo[3',4':4,5]pyrido[2,3-d]pyrimidin-1- one (4) and 3-amino-8-benzyl-7,9-dimethyl-2H-pyrrolo[3',4':4,5]pyrido [2,3-d]pyrimidin-1-one (5), respectively. The target compound 2, its aromatic precursor 18, and the nonclassical analogue 4 were evaluated as inhibitors of the growth of Manca human lymphoma cells and also as inhibitors of human dihydrofolate reductase, human thymidylate synthase, glycinamide ribonucleotide formyltransferase, and aminoimidazole carboxamide ribonucleotide formyltransferase. Compound 18 showed weak inhibition of lymphoma cell growth (IC50 = 42 microM) and of AICAR formylTF (IC50 = 17 microM). Compounds 2 and 4 did not inhibit lymphoma cell growth or thymidylate synthase. The inactivity of 2 was attributed to its lack of flexibility leading to its inability to bind to thymidylate synthase.

Quantitation of folic acid enhancement of antifolate synergism

Trimetrexate (TMTX), 5,10-dideazatetrahydrofolate (DDATHF), and 10-propargyl-5,8-dideazafolate (PDDF, CB3717) are antifolates whose primary intracellular targets are dihydrofolate reductase, glycinamide ribonucleotide formyltransferase, and thymidylate synthase, respectively. Varying the medium folic acid (PteGlu) concentration over the range of 0.5 to 100 microM increasingly blocks the growth inhibitory effects of the individual antifolates in Manca human lymphoma cells, but increasingly enhances the synergistic interaction of both TMTX + DDATHF and TMTX+ PDDF combinations. Drug interactions were quantitated using the universal response surface approach, which consists of fitting a concentration-effect surface to experimental data with weighted nonlinear regression, enabling the estimation of the synergism parameter, alpha. Estimates for alpha are larger (more intense synergism) for the TMTX + DDATHF combination (alpha = 4.68 +/- 0.66 at 2 microM PteGlu; alpha = 53.6 +/- 3.7 at 40 microM PteGlu) than for the TMTX + PDDF combination (alpha = 0.690 +/- 0.25 at 2 microM PteGlu; alpha = 7.20 +/- 1.8 at 40 microM PteGlu). However, the relative increase brought about by increasing the PteGlu concentration from 2 microM to 40 microM is similar in each instance, 11- and 10-fold, respectively. These experiments suggest that the enhanced cytotoxic interaction brought about by increased PteGlu concentration with the TMTX + DDATHF combination and the TMTX + PDDF combination may share a common mechanism. The dramatic intensity of the synergism between DDATHF and TMTX at 40 microM PteGlu, along with the concept of modulation of antifolate synergism by folates, suggests future in vivo and/or clinical applications of combinations of these (or similar) compounds.

Folate analogues. 35. Synthesis and biological evaluation of 1-deaza, 3-deaza, and bridge-elongated analogues of N10-propargyl-5,8-dideazafolic acid

Structural modifications at the pyrimidine ring and at the C9,N10-bridge region of the thymidylate synthase (TS) inhibitors N10-propargyl-5,8-dideazafolate (1; PDDF; CB 3717), 2-desamino-N10-propargyl-5,8-dideazafolate (2, DPDDF), and 2-desamino-2-methyl-N10-propargyl-5,8-dideazafolate (3, DMPDDF) have been carried out. Methods for the synthesis of 2-desamino-N10-propargyl-1,5,8-trideazafolate (4), 2-desamino-2-methyl-N10-propargyl-3,5,8-trideazafolate (5a), and 2-desamino-2-methyl-N10-propargyl-5,8-dideaza-1,2-dihydrofolate (6) have been developed. The bridge-extended analogues isohomo-PDDF (7) and isohomo-DMPDDF (8) contain an additional methylene group interposed between N10 and the phenyl ring of 1 and 3, respectively. All new compounds were evaluated as inhibitors of TS and the growth of tumor cells in culture. Selected analogues were tested as substrates of folylpolyglutamate synthetase (FPGS) and striking differences in substrate activity were observed among these compounds, indicating that structural modifications at the pyrimidine ring of classical antifolates profoundly influence their polyglutamylation. Enzyme inhibition data established that both N1 and N3-H of the pyrimidine ring are essential for efficient binding of quinazoline-type antifolates to human TS.

Synthesis and biological activity of 5,11-methylenetetrahydro-5- deazahomofolic acid

The synthesis of 5,11-methylene-5-deazatetrahydrohomofolate (5), a stable, semirigid mimic of 5,10-methylenetetrahydrofolate (4) is reported as a potential inhibitor of thymidylate synthases (TS). The key intermediate 3-amino-1-oxo-tetrahydropyrimido[4,5-c] [2,6]naphthyridine (6) was obtained by the regiospecific cyclocondensation of 2,4,6-triaminopyrimidine with ethyl 1-benzyl-3-oxo-4-piperidinecarboxylate followed by halogenation (of the resulting lactam 9) and catalytic hydrogenolysis. Selective reduction of 6 followed by arylation with tert-butyl p-fluorobenzoate, saponification, and coupling with diethyl L-glutamate followed by saponification afforded the target compound 5. The title compound was tested as an inhibitor of the growth of Manca human lymphoma cells and also as an inhibitor of TS from Manca cells and Lactobacillus casei and was found to be inactive. In addition, compound 5 also failed to inhibit glycinamide ribonucleotide formyltransferase from L. casei and from Manca cells.

Folate analogues. 34. Synthesis and antitumor activity of non-polyglutamylatable inhibitors of dihydrofolate reductase

Five analogues of methotrextate (MTX), 10-deazaaminopterin (10-DAM), and 10-ethyl-10-deazaaminopterin (10-EDAM) in which the glutamate moiety was replaced by either a gamma-methyleneglutamate or beta-hydroxyglutamate were synthesized and evaluated for their antifolate activity. These analogous are 4-amino-4-deoxy-N10-methylpteroyl-beta-hydroxyglutamic acid (1), 4-amino-4-deoxy-10-deazapteroyl-beta-hydroxyglutamic acid (2), 4-amino-4-deoxy-N10-methylpteroyl-gamma-methyleneglutamic acid (3, MMTX), 4-amino-4-deoxy-10-deazapteroyl-gamma-methyleneglutamic acid (4, MDAM), and 4-amino-4-deoxy-10-ethyl-10-deazapteroyl-gamma-methyleneglutamic acid (5, MEDAM). None of these compounds were metabolized to the respective polyglutamate derivative as judged by their inability to serve as substrates for CCRF-CEM human leukemia cell folylpolyglutamate synthetase (FPGS) in vitro. All compounds inhibited recombinant human-dihydrofolate reductase (DHFR) at nearly equivalent magnitude as MTX. Growth-inhibition studies with H35 hepatoma, Manca human lymphoma, and CCRF-CEM human leukemia cells established greater cytotoxic effects with compounds 3-5 than with compounds 1 and 2. gamma-Methyleneglutamate derivatives 3-5 were transported to H35 hepatoma cells better than MTX or beta-hydroxyglutamate derivatives 1 and 2. Compound 3 was 2.5 times better than MTX in competing with folinic acid transport in H35 hepatoma cells. Compound 1 did not have a significant inhibitory effect on folinic acid transport even at 50 microM under identical conditions. The IC50 for compound 1 against H35-hepatoma cell growth was 8.5-fold higher than MTX. Compounds with the gamma-methyleneglutamate moiety (3-5) exhibited almost equal or lower IC50 values than MTX against the growth of CCRF-CEM human leukemia cells. These studies show that on continuous exposure, the non-polyglutamylatable inhibitors DHFR (3-5) can exhibit superior antifolate activity compared to the polyglutamylatable methotrexate, presumably due to their enhanced transport to these cell lines. Compounds 3-5 appear to be excellent models to study the role of polyglutamylation of antifolates in antitumor activity and host toxicity.

Tetrahydrohomofolate polyglutamates as inhibitors of thymidylate synthase and glycinamide ribonucleotide formyltransferase in Lactobacillus casei

In order to determine the mechanism for the effects of homofolates on growth of Lactobacillus casei, polyglutamated derivatives of homofolate (HPteGlu), dihydrohomofolate and tetrahydrohomofolate (H4HPteGlu) were synthesized and tested as inhibitors of folate-requiring enzymes. The following L. casei enzymes were examined: thymidylate synthase (TS), glycinamide ribonucleotide formyltransferase (GARFT), aminoimidazolecarboxamide ribonucleotide formyltransferase, serine hydroxymethyltransferase and dihydrofolate reductase. Polyglutamates of (6R,S)-H4HPteGlu are potent inhibitors of TS and GARFT. For example, the IC50 values of (6R,S)-H4HPteGlu6 are 0.7 microM for TS and 0.3 microM for GARFT. By contrast, the value for HPteGlu6 is greater than 10 microM for both TS and GARFT. Inhibition of TS and GARFT by (6R,S)-H4HPteGlu derivatives increases with polyglutamate chain length. For TS, the Glu5 and Glu6 derivatives of (6R,S)-H4HPteGlu are 20 and 30 times more potent than the monoglutamate, respectively. For GARFT, the Glu2-6 derivatives are 2-3 times more potent than Glu1. Inhibition of TS and GARFT by (6R,S)-H4HPteGlu polyglutamates is almost entirely due to the unnatural (6R) diastereomer at C-6. Homofolate derivatives are only weak inhibitors of aminoimidazolecarboxamide ribonucleotide formyltransferase, serine hydroxymethyltransferase, and dihydrofolate reductase. We conclude that both TS and GARFT are potential targets of (6R)-H4HPteGlu polyglutamates.

Folate analogues. 33. Synthesis of folate and antifolate poly-gamma-glutamates by [(9-fluorenylmethoxy)oxy]carbonyl chemistry and biological evaluation of certain methotrexate polyglutamate polylysine conjugates as inhibitors of the growth of H35 hepatoma cells

Representative examples of folate and antifolate poly-gamma-glutamyl metabolites were synthesized via the [(9-fluorenylmethoxy)oxy]carbonyl (Fmoc) chemistry using the KH polyamide resin. Polyglutamate yields were consistently better in all cases compared to the previous Merrifield method, and the crude products were obtained in greater than 85% purity. The symmetrical anhydride (7) derived from alpha-tert-butyl N-Fmoc-L-glutamate (6) was used for the initial coupling of the first glutamate residue to the KH resin and also for subsequent chain elongation. The alpha-tert-butyl protective groups were not labile under the conditions used for the cleavage of the finished peptide from the resin. A series of poly-gamma-glutamyl metabolites of methotrexate (MTX) with a chain length ranging from two to five glutamyl residues were synthesized and coupled with poly(L-lysine) having an average molecular weight of 27,000 and 52,000. Each conjugate was tested for its ability to inhibit the growth of wild type (H35) and MTX transport resistant (H35R) strains of hepatoma cells in culture, the latter having a 100-fold reduced sensitivity to MTX. 4-Amino-4-deoxy-N10-methylpteroylglutamyl-gamma-glutamylpoly (L-lysine) conjugate [MTX(G2)-poly-L-Lys-52000] and MTX(G4)-poly-L-Lys-52000 were among the most active (I50 = 8.0 and 10 nM against H35 cells) MTX-polylysines synthesized to date, and they were somewhat more inhibitory to the transport resistant cells. MTX(G5)-poly-L-Lys-52000 was approximately 1000 times more effective than MTX(G5)-poly-D-Lys-52000 in inhibiting the growth of H35R hepatoma cells in culture, indicating that internal cleavage of the gamma-glutamate chain of the conjugate with subsequent release of MTX or shorter chain polyglutamates of MTX is unlikely to be an important determinant of MTX-polyglutamate polylysine cytotoxicity. The results indicate that MTX-polyglutamate poly(L-lysine) conjugates are taken up by the cells independently of MTX and probably via endocytosis.

Synthesis and antifolate properties of 10-alkyl-5,10-dideaza analogues of methotrexate and tetrahydrofolic acid

Synthesis of the 10-methyl and 10-ethyl analogues of 5,10-dideazatetrahydrofolic acid (DDTHF), a potent inhibitor of glycinamide ribotide (GAR) formyltransferase, is reported. Key intermediates in the process were 10-methyl- and 10-ethyl-4-amino-4-deoxy-5,10-dideazapteroic acid. Condensation of the piperidine enamines of branched 4-(p-carbomethoxyphenyl)butyraldehydes with (acetoxymethylene)malononitrile afforded 1,1-dicyano-4-piperidinobutadiene 5a,b. Subsequent reaction with alcoholic ammonium hydroxide yielded the appropriately substituted 2-amino-3-cyanopyridines 6a,b. Ring closure with guanidine gave 10-methyl- and 10-ethyl-4-amino-4-deoxy-5,10-dideazapteroic acids (7a,b). Coupling with diethyl glutamate followed by ester hydrolysis afforded 10-alkyl-5,10-dideazaminopterin analogues 9a,b. Hydrolysis of the 4-amino group of 7a,b yielded the 10-alkylpteroic acids, which were coupled with diethyl glutamate, hydrogenated over PtO2, and saponified to afford 10-alkyl-5,10-dideazatetrahydrofolic acids 13a,b. Aminopterin analogues 9a,b were effective inhibitors of DHFR derived from L1210, but were less potent than methotrexate for inhibition of growth of L1210 in culture. The 10-ethyl (13b) analogue of 5,10-DDTHF was about twice as potent an inhibitor of L1210 cell growth as 5,10-DDTHF, but was only 1/7 as potent for inhibition of GAR formyltransferase. 10-Methyl analogue 13a was similar in potency to 5,10-DDTHF. All of the compounds showed moderately improved transport into L1210 cells relative to methotrexate.

Synthesis and antifolate properties of 9-alkyl-10-deazaminopterins

Reformatski condensation of benzyl 2-bromopropionate with 4-carbomethoxybenzaldehyde, followed by dehydration afforded benzyl 2-methyl-p-carbomethoxycinnamate (4a). Hydrogenation over a Pd catalyst gave the hydrocinnamic acid 5a. Conversion to the chloromethyl (6a) and azidomethyl ketone (7a) was followed by hydrogenation to the aminomethyl ketone (8a). Direct N-alkylation by 2,4-diamino-5-nitro-6-chloropyrimidine followed by reductive ring closure in Zn-HOAc and subsequent saponification of the benzoate ester yielded 4-amino-4-deoxy-9-methyl-10-deazapteroic acid (11a). Coupling with diethyl L-glutamate and saponification afforded 9-methyl-10-deazaminopterin (13a). The 9-ethyl analogue (13b) was similarly prepared from benzyl 2-bromobutyrate. The 9-methyl analogue (13a) was 21 times more potent than MTX as an inhibitor of cell growth in L1210 cells. The reason for this enhanced cytotoxicity in L1210 is unclear, since enzyme inhibition and transport parameters were similar to those of MTX. In human Manca leukemia cells growth inhibition was not dramatic and paralleled MTX.

Folate analogues. 32. Synthesis and biological evaluation of 2-desamino-2-methyl-N10-propargyl-5,8-dideazafolic acid and related compounds

The chemical synthesis of three close analogues (2-4) of N10-propargyl-5,8-dideazafolate (PDDF) is described. The quinazoline ring of 2 and 4 was constructed from the pivotal intermediate 9 in a novel and unambiguous manner during the final step of the synthesis under very mild conditions. 2-Desamino-2-methyl-N10-propargyl-5,8-dideazafolate (DMPDDF) (2) was a strong inhibitor of human and Lactobacillus casei thymidylate synthases, whereas 2-desamino-2-(trifluoromethyl)-N10-propargyl-5,8-didezafolate (3) and 2-desamino-2,3-dimethyl-N10-propargyl-5,8-dideazafolate (4) were only weak inhibitors of this enzyme. DMPDDF exhibited excellent growth inhibition of Manca human lymphoid leukemia and H35 hepatoma cells in culture. The inhibitor activities of 2 were 43- and 65-fold greater than that of PDDF, respectively, in these cell lines. H35R cells that are resistant to methotrexate by virtue of a transport defect were cross resistant to DMPDDF but not to PDDF. H35FF cells which have 70-fold greater amounts of thymidylate synthase compared to H35N cells were 130-fold resistant to DMPDDF. Furthermore, the toxicity of DMPDDF to H35 hepatoma cells could be completely reversed by thymidine, establishing its locus of action as thymidylate synthase. Transport studies in vitro established that DMPDDF effectively inhibits MTX influx into H35 hepatoma cells, whereas PDDF has no effect on MTX transport in this cell line. These data suggest that the greater activity of DMPDDF relative to PDDF is partly due to the ability of the former compound to enter cells via the MTX/reduced folate transport system. Enzyme inhibition data of 4 suggest that the presence of N3H in DMPDDF is essential for binding to thymidylate synthase.

Folate analogues. 31. Synthesis of the reduced derivatives of 11-deazahomofolic acid, 10-methyl-11-deazahomofolic acid, and their evaluation as inhibitors of glycinamide ribonucleotide formyltransferase

The Boon-Leigh procedure, involving condensation of a 6-chloro-5-nitropyrimidine (22) with an alpha-amino ketone (20 or 21) followed by reduction of the nitro group, cyclization, and L-glutamylation, led to the formation of 11-deazahomofolate (29) and its 10-methyl derivative (30). The corresponding (6R,S)-5,6,7,8-tetrahydro (4, 5) and 7,8-dihydro (31, 32) derivatives were prepared by catalytic hydrogenation. (6S)-11-Deazatetrahydrohomofolate was prepared from 29 by enzymatic reduction. Compounds 29 and 30 had little effect (IC50 greater than 2 x 10(-5) M) on Lactobacillus casei glycinamide ribonucleotide (GAR) formyltransferase but (6R,S)-11-deazatetrahydrohomofolate (4) is a potent inhibitor of this enzyme (IC50 = 5 x 10(-8) M). It is at least 100 times more inhibitory than 33, the 6S compound, indicating that the 6R component of the mixture having the unnatural configuration at C6 (34) is responsible for the potent inhibition. Compound 4 is a much weaker inhibitor of murine (L1210) and human (MOLT-4) leukemia cell GAR formyltransferases (IC50 greater than 1 x 10(-5) M). (6R,S)-11-Deaza-10-methyltetrahydrohomofolate (5) (IC50 = 1.1 x 10(-5) is 200 times weaker than 4 against L. casei GAR formyltransferase. However, 11-deaza-10-methyldihydrohomofolate (32) is more inhibitory (IC50 = 5.5 x 10(-7) M) than 5 or 30. None of the compounds showed inhibition of L. casei aminoimidazolecarboxamide ribonucleotide (AICAR) formyltransferase, dihydrofolate reductase, or thymidylate synthase. The dihydro derivatives 31 and 32 are 5% as active as dihydrofolate as substrates for L. casei dihydrofolate reductase. Compound 4 showed moderate inhibition of the growth of L. casei, Streptococcus faecium, MOLT-4 cells, and MCF-7 human breast adenocarcinoma cells.

Inhibition of glycinamide ribonucleotide formyltransferase and other folate enzymes by homofolate polyglutamates in human lymphoma and murine leukemia cell extracts

In order to determine the biochemical basis for the cytotoxicity of homofolates, poly-gamma-glutamyl derivatives of homofolate (HPteGlu) and tetrahydrohomofolate (H4HPteGlu) were synthesized and tested as inhibitors of glycinamide ribonucleotide formyltransferase (GARFT), aminoimidazolecarboxamide ribonucleotide formyltransferase (AICARFT), thymidylate synthase, and serine hydroxymethyltransferase (SHMT) in extracts of Manca human lymphoma and L1210 murine leukemia cells. The most striking inhibitions are that of GARFT by (6R,S)-H4HPteGlu4-6 with IC50 values from 1.3 to 0.3 microM. Both diastereomers, (6R)-H4HPteGlu6 and (6S)-H4HPteGlu6, inhibit GARFT activity. In Manca cell extracts, the (6S) form is more potent than the (6R) form whereas in the murine system the reverse is true. The (6R,S)-H4HPteGlu polyglutamates are weak inhibitors of human AICARFT (IC50, 6-10 microM). Polyglutamates of HPteGlu, however, are more inhibitory to AICARFT, with HPteGlu4-6 having IC50 values close to 2 microM. Polyglutamates of HPteGlu and of H4HPteGlu are weaker inhibitors of thymidylate synthase (IC50, 8 microM for HPteGlu5-6 and greater than 20 microM for H4HPteGlu1-5). Polyglutamates of HPteGlu and of H4HPteGlu are poor inhibitors of SHMT (IC50, greater than 20 microM). Manca cell growth is inhibited 50% by HPteGlu and (6R,S)-5-methyl-H4HPteGlu at 6 and 8 microM, respectively. Both of these effects are reversed by 0.1 mM inosine. Trimetrexate at a subinhibitory concentration, 10 nM, antagonizes growth inhibition by HPteGlu, raising the IC50 from 6 to 64 microM, but enhances inhibition by (6R,S)-5-methyl-H4HPteGlu, lowering the IC50 from 8 to 5 microM. Our results support the view that homofolates become toxic after conversion to H4HPteGlu polyglutamates which block GARFT, a step in purine biosynthesis.

Synthesis and antifolate activity of 5-methyl-5,10-dideaza analogues of aminopterin and folic acid and an alternative synthesis of 5,10-dideazatetrahydrofolic acid, a potent inhibitor of glycinamide ribonucleotide formyltransferase

The title compounds were prepared in extensions of a general synthetic approach used earlier to prepare 5-alkyl-5-deaza analogues of classical antifolates. Wittig condensation of 2,4-diaminopyrido[2,3-d]pyrimidine-6-carboxaldehyde (2a) and its 5-methyl analogue 2b with [4-(methoxycarbonyl)benzylidene] triphenylphosphorane gave 9,10-ethenyl precursors 3a and 3b. Hydrogenation (DMF, ambient, 5% Pd/C) of the 9,10-ethenyl group of 3b followed by ester hydrolysis led to 4-[2-(2,4-diamino-5-methylpyrido[2,3-d]pyrimidin-6-yl)ethyl]ben zoi c acid (5), which was converted to 5-methyl-5,10-dideazaaminopterin (6) via coupling with dimethyl L-glutamate (mixed-anhydride method using i-BuOCOCl) followed by ester hydrolysis. Standard hydrolytic deamination of 6 gave 5-methyl-5,10-dideazafolic acid (7). Intermediates 3a and 3b were converted through concomitant deamination and ester hydrolysis to 8a and 8b. Peptide coupling of 8a,b (using (EtO)2POCN) with diesters of L-glutamic acid gave intermediate esters 9a and 9b. Hydrogenation of both the 9,10 double bond and the pyrido ring of 9a and 9b (MeOH-0.1 N HCl, 3.5 atm, Pt) was followed by ester hydrolysis to give 5,10-dideaza-5,6,7,8-tetrahydrofolic acid (11a) and the 5-methyl analogue 11b. Biological evaluation of 6, 7, 11a, and 11b for inhibition of dihydrofolate reductase (DHFR) isolated from L1210 cells and for growth inhibition and transport characteristics toward L1210 cells revealed 6 to be less potent than methotrexate in the inhibition of DHFR and cell growth. Compounds 6, 11a, and 11b were transported into cells more efficiently than methotrexate. Growth inhibition IC50 values for 11a and 11b were 57 and 490 nM, respectively; the value for 11a is in good agreement with that previously reported (20-50 nM). In tests against other folate-utilizing enzymes, 11a and 11b were found to be inhibitors of glycinamide ribonucleotide formyltransferase (GAR formyltransferase) from one bacterial (Lactobacillus casei) and two mammalian (Manca and L1210) sources with 11a being decidedly more inhibitory than 11b. Neither 11a nor 11b inhibited aminoimidazolecarboxamide ribonucleotide formyltransferase. These results support reported evidence that 11a owes its observed antitumor activity to interference with the purine de novo pathway with the site of action being GAR formyltransferase.

Synthesis and biological evaluation of 8-deazahomofolic acid and its tetrahydro derivative

The syntheses of 8-deazahomofolic acid and its tetrahydro derivative, potential inhibitors of thymidylate synthase (TS) and other folate related enzymes, are described. Wittig condensation of 2-acetamido-6-formyl-4-pyrimidinol with the triphenylphosphine ylide 3 derived from N-acetyl-4-(p-carbethoxyanilino)-1-chloro-2-butanone, hydrogenation of the enone intermediate 5, introduction of a 5-amino group via diazonium coupling, and reductive ring closure yielded ethyl N11-acetyl-8-deazahomopteroate (8). Alkaline hydrolysis gave 8-deazahomopteroic acid, which was blocked as the 11-trifluoroacetyl derivative, coupled with diethyl L-glutamate, and the blocking groups saponified to afford 8-deazahomofolic acid (12). Hydrogenation of the glutamate diester intermediate and subsequent saponification yielded the tetrahydro-8-deazahomofolate (14). Growth inhibition of Streptococcus faecium, Lactobacillus casei, and L1210 cells in culture by the target compounds was modest. They were also weak inhibitors of thymidylate synthase, dihydrofolate reductase, glycinamide-ribonucleotide transformylase, and aminoimidazolecarboxamide ribonucleotide transformylase. In contrast, 8-deazafolate showed moderate inhibition of aminoimidazolecarboxamide ribonucleotide transformylase, suggesting that inhibition of this enzyme may be related to its cytotoxic action. Tetrahydro-8-deazahomofolate showed low substrate activity with thymidylate synthase.

Synthesis and biological evaluation of poly-gamma-glutamyl metabolites of 10-deazaaminopterin and 10-ethyl-10-deazaaminopterin

The chemical synthesis of a series of poly-gamma-glutamyl metabolites of the experimental anticancer drugs 10-deazaaminopterin (10-DAAM) and 10-ethyl-10-deazaaminopterin (10-EDAAM) has been carried out by the solid-phase procedure. The synthetic products were identical with the poly-gamma-glutamyl metabolites of radiolabeled 10-DAAM and 10-EDAAM produced by normal mouse tissues with regard to elution volume from [(diethylamino)ethyl]cellulose columns and susceptibility to hydrolysis by human plasma folylpolyglutamate hydrolase. Poly-gamma-glutamyl metabolites with a glutamate chain length of up to four glutamate residues were detected in the tissues. The antifolate activity was evaluated with methotrexate (MTX) sensitive and MTX-resistant strains of Lactobacillus casei and Streptococcus faecium. In general, inhibitory potency decreases with increasing Glu chain length. However there are two exceptions. Addition of one Glu residue to 10-DAAM enhances its potency for MTX-resistant L. casei and addition of one Glu residue to 10-EDAAM enhances its potency for the MTX-sensitive L. casei. As shown earlier for MTX polyglutamates, polyglutamylation greatly enhances the inhibitory potency of 10-DAAM and 10-EDAAM for L. casei thymidylate synthase. MTX polyglutamates are 15-30 times more inhibitory than the corresponding 10-DAAM derivatives and 30-60 times more inhibitory than the corresponding 10-EDAAM derivatives. Polyglutamylation of 10-DAAM had little influence on its ability to inhibit L. casei dihydrofolate reductase; however, with 10-EDAAM, addition of one or two Glu residues enhanced its inhibitory potency 2.3-fold.

Folate analogues. 30. Synthesis and biological evaluation of N10-propargyl-5,8-dideaza-5,6,7,8-tetrahydrofolic acid and related compounds

The 5,6,7,8-tetrahydro derivative (1) of the powerful thymidylate synthase inhibitor N10-propargyl-5,8-dideazafolic acid (PDDF) has been synthesized and evaluated for its antifolate activity. A convenient method for the preparation of the key intermediate 2-amino-6-(bromomethyl)-4-hydroxy-5,6,7,8-tetrahydroquinazoline (18) is described. Two closely related analogues of 1 were also synthesized and evaluated for their antifolate activity and thymidylate synthase inhibition. N10-Propargyl-5,8-dideaza-5,6,7,8-tetrahydrofolate (1) and N10-methyl and N10-hydrogen analogues 2 and 3 were weaker inhibitors of Lactobacillus casei thymidylate synthase compared to PDDF. N10-Methyl-5,8-dideaza-5,6,7,8-tetrahydrofolate (2) exhibited the most potent antifolate activity against L. casei (IC50 = 2.8 nM) and Streptococcus faecium (IC50 = 0.57 nM). In intact and permeabilized murine leukemia L1210 cells, the replacement of the quinazoline moiety with its tetrahydro derivative resulted in a marked decrease in potency and a loss of the contribution of the propargyl substituent to enzyme inhibition, indicating an altered binding mode to thymidylate synthase.

Carrier-mediated transport of folate compounds in L1210 cells. Initial rate kinetics and extent of duality of entry routes for folic acid and diastereomers of 5-methyltetrahydrohomofolate in the presence of physiological anions

Comparison of the kinetic parameters for influx of highly purified [3H]folic acid versus [3H]methotrexate in L1210 cells under anionic buffer conditions showed a marked discordancy. In addition, the kinetics for influx of [3H]folic acid were unchanged in variant L1210 cells defective in [3H]methotrexate transport. In these variant cells, the Vmax for methotrexate was reduced 17-fold and the Km was increased 3-fold. The results show that [3H]folic acid influx is mediated by a system which has a low affinity, but a 20-fold higher capacity, for folate compounds than the classical high-affinity system mediating [3H]methotrexate influx. Since the latter system also exhibits very low affinity for [3H]folic acid, it would not be expected to contribute significantly to the total influx of [3H]folic acid. The high-capacity system for [3H]folic acid influx is different from that believed to mediate pterin influx in L1210 cells since it was not inhibited by adenine, a potent inhibitor of pterin influx. However, exposure of cells to [3H]folic acid in a nonanionic buffer resulted in marked stimulation of initial influx, and a fraction of influx under these conditions was inhibited by methotrexate. These results suggest that anions modulate the extent of multiplicity of [3H]folic acid influx by their known effects on the high-affinity, reduced folate/methotrexate system. The diastereomers, at carbon 6, of [14C]5-methyltetrahydrohomofolate shared both transport systems. The influx Km for the natural diastereomer was one-half that of the unnatural form for both transport systems. Both diastereomers showed a much greater differential in affinity between the two transport systems than did [3H]folic acid. Our results suggest that an analog which could be effectively transported by the low-affinity/high-capacity route may be useful in the treatment of tumors resistant to methotrexate due to a defective high-affinity/low capacity influx system. We also found that incubation of L1210 cells with [3H]folic acid or the natural diastereomer [14C]5-methyltetrahydrohomofolate for 10 min resulted in the formation of a nonexchangeable fraction of radioactivity amounting to 20-40% of the total accumulation. This non-exchangeable fraction may be explained by the accumulation of metabolites other than polyglutamates. Preloading of cells with methotrexate prior to incubation with [3H]folic acid prevented the accumulation of radioactivity as a nonexchangeable fraction.

Interaction of polyglutamyl derivatives of methotrexate, 10-deazaaminopterin, and dihydrofolate with dihydrofolate reductase

Polyglutamyl derivatives of methotrexate (MTX) and 10-deazaaminopterin (10-DAM) containing a total of one through six glutamate residues (Glu residues) were tested as inhibitors of dihydrofolate reductase (DHFR) derived from sheep, chicken, and beef liver. The ability of dihydropteroylpentaglutamate to antagonize the inhibitory activity of these analogues was also studied. The most striking effects were seen with sheep liver DHFR, where polyglutamylation of MTX causes stepwise decreases in the concentration required for 50% inhibition (IC50) with each additional Glu residue until MTX with a total of six Glu residues has an IC50 value 1/3 that of MTX. With 10-DAM the pattern is more complex. The IC50 values increase with addition of Glu residues until a maximum is reached with 10-DAM having a total of three Glu residues which has a value twice that of 10-DAM. 10-DAM with a total of four Glu residues and 10-DAM with a total of five Glu residues have progressively lower IC50 values, the latter being equipotent with 10-DAM. With dihydropteroylpentaglutamate as substrate instead of dihydrofolate, the IC50 values are increased 2- to 5-fold for both MTX and 10-DAM derivatives. The results obtained with chicken liver and beef liver DHFR are generally similar to those described for the sheep liver enzyme, but the effects of polyglutamylation are less pronounced. The addition of 0.2 M KCl to the assay system reduces the differences in inhibitory potency of the polyglutamyl derivatives with all three enzymes tested. We conclude that polyglutamylation can alter the interaction of folate analogues and dihydrofolate with DHFR.

Synthesis and antifolate properties of 5,10-methylenetetrahydro-8,10-dideazaminopterin

The synthesis of the 5,10-methylene analogue of 5,6,7,8-tetrahydro-8,10-dideazaminopterin, a potential dual inhibitor of dihydrofolate reductase (DHFR) and thymidylate synthase (TS) enzymes, is described. The dimethyl ester of 10-carboxy-4-amino-4-deoxy-8,10-dideazapteroic acid was converted to the tetrahydro derivative by hydrogenation. Thermally induced cyclization of the 10-carbomethoxy and the 5-NH groups afforded the 5,10-carbonyl analogue. Reduction of the lactam with borane readily yielded the key 5,10-methylene-4-amino-4-deoxy-8,10-dideazatetrahydropteroic acid methyl ester. Saponification of the benzoate ester and coupling with L-glutamate concluded the synthesis. The title compound was a modest inhibitor of growth in folate-dependent bacteria. Streptococcus faecium and Lactobacillus casei, but inhibition of DHFR or TS derived from L. casei was poor. The compound was also a weak inhibitor of DHFR derived from L1210 murine leukemia and was a weak inhibitor of L1210 growth in culture.

Folate analogues. 26. Syntheses and antifolate activity of 10-substituted derivatives of 5,8-dideazafolic acid and of the poly-gamma-glutamyl metabolites of N10-propargyl-5,8-dideazafolic acid (PDDF)

The poly-gamma-glutamyl derivatives of n10-propargyl-5,8-dideazafolic acid (PDDF) with a chain length of up to five glutamate residues were synthesized from N10-propargyl-5,8-dideazapteroic acid by the solid-phase procedure. These compounds were evaluated for their antifolate activity using folate-requiring microorganisms and intact and permeabilized L1210 cells and as inhibitors of dihydrofolate reductase and thymidylate synthase derived from L. casei. The polyglutamylated derivatives of PDDF (1) were more active than the parent compound in inhibiting the growth of L. casei, thymidylate synthesis in permeabilized L1210 cells, and L. casei thymidylate synthase. Two analogues of 5,8-dideazafolic acid (2 and 3), one with a 2-butyne and another with a cyclopropylmethyl substituent at N10, were also synthesized and evaluated for their antifolate activities using the above-mentioned test systems. They were considerably less active than PDDF or its polyglutamylated derivatives. N10-Propargyl-5,8-dideazapteroyl tri-, tetra-, and pentaglutamates were equipotent with 5-fluorodeoxyuridylate as inhibitors of thymidylate synthesis in permeabilized L1210 cells. The polyglutamyl metabolites of PDDF were shown to be the most potent antifolate inhibitors of L. casei and L1210 thymidylate synthases yet described.

Folate analogues. 25. Synthesis and biological evaluation of N10-propargylfolic acid and its reduced derivatives

N10-Propargylfolic acid (2), which is the closest pteridine analogue of the thymidylate synthase inhibitor N10-propargyl-5,8-dideazafolic acid (PDDF), was synthesized starting from diethyl [p-(N-propargylamino)benzoyl]-L-glutamate (5) and N-(3-bromo-2-oxopropyl)phthalimide (8). The 7,8-dihydro derivative of propargylfolic acid served as a synthetic substrate of Lactobacillus casei dihydrofolate reductase. Propargylfolic acid and its reduced derivatives were weak inhibitors of L. casei thymidylate synthase compared to PDDF. All derivatives of propargylfolate were active against the growth of Streptococcus faecium, but with the exception of 7,8-dihydropropargylfolic acid, all were inactive against L. casei. Although less potent than PDDF, marked inhibition of thymidylate synthase by 2 was observed in permeabilized L1210 cells.

Synthesis and biological activity of resolved C-10 diastereomers of 10-methyl- and 10-ethyl-10-deazaminopterin

Synthesis and evaluation of the antitumor drugs 10-methyl- and 10-ethyl-10-deazaminopterin (15a,b) were previously reported for the diastereomeric mixtures, lacking resolution at the C-10 position. In order to assess biological properties of the individual diastereomers, the C-10 isomers of 4-amino-4-deoxy-10-methyl- and 10-ethyl-10-deazapteroic acids (13a,b) were prepared by total synthesis. Coupling with L-glutamate afforded the appropriate diastereomers of the title compounds. Biochemical, transport, and cell growth inhibitory properties in L1210 cells and folate-dependent bacteria were measured. Differences were generally less than 2-fold between diastereomeric pairs, but a factor of 3 was noted for d,L-15b vs. l,L-15b in inhibition of DHFR from L1210 cells and in cytotoxicity toward L1210 cells. An in vivo comparison of the isomers of 15b with racemic compound against L1210 in mice did not show a significant efficacy difference (ILS) among the compounds. However, d,L-15b showed an acute toxicity about 2.5 times that of l,L-15b.

Syntheses and antifolate activity of 5-methyl-5-deaza analogues of aminopterin, methotrexate, folic acid, and N10-methylfolic acid

Evidence indicating that modifications at the 5- and 10-positions of classical folic acid antimetabolites lead to compounds with favorable differential membrane transport in tumor vs. normal proliferative tissue prompted an investigation of 5-alkyl-5-deaza analogues. 2-Amino-4-methyl-3,5-pyridinedicarbonitrile, prepared by hydrogenolysis of its known 6-chloro precursor, was treated with guanidine to give 2,4-diamino-5-methylpyrido[2,3-d]pyrimidine-6-carbonitrile which was converted via the corresponding aldehyde and hydroxymethyl compound to 6-(bromomethyl)-2,4-diamino-5-methylpyrido[2,3-d]pyrimidine. Reductive condensation of the nitrile 8 with diethyl N-(4-amino-benzoyl)-L-glutamate followed by ester hydrolysis gave 5-methyl-5-deazaaminopterin. Treatment of 12 with formaldehyde and Na(CN)BH3 afforded 5-methyl-5-deazamethotrexate, which was also prepared from 15 and dimethyl N-[(4-methylamino)benzoyl]-L-glutamate followed by ester hydrolysis. 5-Methyl-10-ethyl-5-deazaaminopterin was similarly prepared from 15. Biological evaluation of the 5-methyl-5-deaza analogues together with previously reported 5-deazaaminopterin and 5-deazamethotrexate for inhibition of dihydrofolate reductase (DHFR) isolated from L1210 cells and for their effect on cell growth inhibition, transport characteristics, and net accumulation of polyglutamate forms in L1210 cells revealed the analogues to have essentially the same properties as the appropriate parent compound, aminopterin or methotrexate (MTX), except that 20 and 21 were approximately 10 times more growth inhibitory than MTX. In in vivo tests against P388/0 and P388/MTX leukemia in mice, the analogues showed activity comparable to that of MTX, with the more potent 20 producing the same response in the P388/0 test as MTX but at one-fourth the dose; none showed activity against P388/MTX. Hydrolytic deamination of 12 and 20 produced 5-methyl-5-deazafolic acid and 5,10-dimethyl-5-deazafolic acid, respectively. In bacterial studies on the 2-amino-4-oxo analogues, 5-deazafolic acid proved to be a potent inhibitor of Lactobacillus casei DHFR and also the growth of both L. casei ATCC 7469 and Streptococcus faecium ATCC 8043. Its 5-methyl congener 22 is also inhibitory toward L. casei, but its IC50 for growth inhibition is much lower than its IC50 values for inhibition of DHFR or thymidylate synthase from L. casei, suggesting an alternate site of action.

Synthesis and antifolate properties of 10-alkyl-8,10-dideazaminopterins

The synthesis of 10-alkyl analogues of the potent antitumor agent 8,10-dideazaminopterin is described. Alkylation of appropriate alpha-alkyl homoterephthalate esters with 2,4-diamino-6-(bromomethyl)-8-deazapteridine afforded 10-alkyl-10-carboxy-4-amino-4-deoxy-8,10-dideazapteroic acid diesters. Ester cleavage and decarboxylation at C-10 were accomplished by heating with sodium cyanide in Me2SO at 170-180 degrees C to afford the 2,4-diamino-10-alkyl-8,10-dideazapteroic acids. The acids were coupled with diethyl glutamate, followed by saponification, to give the 10-alkyl-8,10-dideazaminopterins. The compounds were potent inhibitors of growth in folate-dependent bacteria, Streptococcus faecium and Lactobacillus casei. The 10-methyl and 10-ethyl analogues gave the highest percent increases in life span for mice infected with L1210 leukemia with ILS values of +203 and +235%, respectively.

Folate analogues. 22. Synthesis and biological evaluation of two analogues of dihydrofolic acid possessing a 7,8-dihydro-8-oxapterin ring system

Two analogues of dihydrofolic acid possessing a 7,8-dihydro-8-oxapterin ring system have been synthesized and evaluated for their antifolate activities. These compounds, N-[(2-amino-4-hydroxy-7,8-dihydro-8-oxa-6-pteridinyl)benzoyl]-L-glutamic acid (3) and N-[[(2-amino-4-hydroxy-7,8-dihydro-8-oxa-6-pteridinyl) methyl]benzoyl]-L-glutamic acid (4), were synthesized by reacting the appropriately substituted alpha-halo ketones with 2,5-diamino-4,6-dihydroxypyrimidine (2). Elaboration of p-carbomethoxybenzaldehyde (5) to p-carbomethoxyphenacyl bromide (7) was accomplished by its oxidation with Jones reagent and the successive treatment of the oxidation product with SOCl2, CH2N2, and HBr. Commercially available p-vinylbenzoic acid (11) was converted to its glutamate conjugate 12 and was further converted to the bromo ketone, diethyl N-[p-(1-bromo-2-oxopropyl)benzoyl]-L-glutamate (17), by a series of reactions involving epoxidation, oxirane ring opening with HBr, Jones oxidation, Zn/HOAc reduction, and successive treatment of the reduction product 16 with SOCl2, CH2N2, and HBr. These bromo ketones, 7 and 17, upon reaction with pyrimidine 2, gave the diethyl esters of the target compounds, which were hydrolyzed to 3 and 4 with NaOH. Compound 4 underwent an interesting acid-catalyzed isomerization where the double bond of 4 was shifted from the 5,6-position to the 6,9-position to give the isomer 19. Both compounds 3 and 4 were inactive against Lactobacillus casei (ATCC 7469) and did not serve as synthetic substrates of L. casei dihydrofolate reductase. Compound 4 showed activity against Streptococcus faecium (ATCC 8043), but 3 was inactive against this organism.

Folate analogues. 21. Synthesis and antifolate and antitumor activities of N10-(cyanomethyl)-5,8-dideazafolic acid

A close analogue of the antileukemic agent 5,8-dideaza-N10 propargylfolic acid (2) was synthesized by replacing the propargyl moiety of 2 with a cyanomethyl group. This compound, N10-(cyanomethyl)-5,8-dideazafolic acid (3), was evaluated for its antifolate and antitumor activities in several biological test systems. Alkylation of diethyl N-(4-aminobenzoyl)-L-glutamate with bromoacetonitrile gave diethyl N-[4-[(cyanomethyl)amino]benzoyl]-L-glutamate (7). Reaction of 7 with 2 amino-6-(bromomethyl)-4-hydroxyquinazoline (9) in dimethylacetamide gave the corresponding diethyl ester 11, which was hydrolyzed to the target compound 3. The known antileukemic agent 2 was also synthesized for comparative studies by employing a modified procedure, which resulted in a better yield of this product. Both compounds 2 and 3 were evaluated for their antifolate activities by using two folate-requiring microorganisms, Streptococcus faecium and Lactobacillus casei. They were further evaluated as inhibitors of thymidylate synthase and dihydrofolate reductase derived from the above organisms, as well as for their antitumor activity by using selected tumor cells in culture. Compound 2 was found to be as equally potent as methotrexate (MTX) against S. faecium, and it was an excellent inhibitor of L. casei thymidylate synthase. The cyanomethyl analogue 3 was less active than 2 in all the test systems, except the inhibition of dihydrofolate reductase.

Folate analogues. 20. Synthesis and antifolate activity of 1',2',3',4',5',6'-hexahydrohomofolic acid

The synthesis of 1',2',3',4',5',6'-hexahydrohomofolic acid (3), a close analogue of homofolic acid (2), has been carried out by replacement of the benzene ring of 2 with a cyclohexane ring. The synthetic methods employed here were based on the Boon-Leigh strategy to obtain products with unambiguous structures. Based on a number of chemical and spectral observations, a tentative cis stereochemistry was assigned to the 1,4-substituents of the cyclohexane ring of both the homopteroate analogue 13 and the target compound 3. We investigated hexahydrohomopteroic acid (13), hexahydrohomofolic acid (3), and their 7,8-dihydro and d,l-5,6,7,8-tetrahydro derivatives for antifolate activity employing several biological test systems. The dihydro and tetrahydro derivatives of both 13 and 3 were active against Streptococcus faecium, whereas they were inactive against Lactobacillus casei. These compounds were neither substrates nor inhibitors of L. casei dihydrofolate reductase or thymidylate synthase.

An hypothesis on the role of pteroylpolyglutamate derivatives as coenzymes

Evidence is assembled supporting the view that the poly-gamma-Glu chain of pteroyl poly-gamma-Glu derivatives plays a role in coordinating the activities of enzymes involved in the sequential metabolism of single carbon units.

Synthesis and antitumor activity of 10-alkyl-10-deazaminopterins. A convenient synthesis of 10-deazaminopterin

Requirements for large-scale synthesis of the potent antitumor drug 10-deazaminopterin have led to development of a facile synthesis of this compound and its 10-alkyl analogues. The lithium diisopropyl amide generated dianions of appropriate p-alkylbenzoic acids were alkylated with 3-methoxyallyl chloride. The resulting 4-(p-carboxyphenyl)-1-methoxy-1-butenes were brominated at pH 7-8 to afford the 2-bromo-4-(p-carboxyphenyl)butyraldehydes. Condensation with 2,4,5,6-tetraminopyrimidine and subsequent in situ oxidation of the resulting dihydropteridines yielded crystalline 10-alkyl-10-deaza-4-amino-4-deoxypteroic acids. The pteroic acids were coupled with diethyl glutamate via the mixed anhydride method, followed by saponification at room temperature, to give the target 10-deazaminopterins. The 10-alkyl compounds were approximately equipotent to 10-deazaminopterin as growth inhibitors of folate-dependent bacteria. Their abilities to inhibit Lactobacillus casei and L1210 derived dihydrofolate reductases were also similar. Transport properties in vitro were suggestive of an improved therapeutic index for the 10-alkyl analogues. Against L1210 in mice, the percent increase in life span at the LD10 dosage was +151% (methotrexate), +178% (10-deazaminopterin), +235% (10-methyl analogue), and +211% (10-ethyl analogue). 10,10-Dimethyl-10-deazaminopterin was less effective at an equimolar dosage, but the ILS at the maximum dose tested (72 mg/kg) was +135%. It was far less toxic than the other analogues possibly because of enhanced clearance.

10-Propargylaminopterin and alkyl homologues of methotrexate as inhibitors of folate metabolism

Reported antifolate activity against leukemia L1210 by N-[14-[[(2-amino-4-hydroxy-6-quinazolinyl)methyl]-propargylamino]benzoyl]]-L-glu tamic acid through potent inhibition of thymidylate synthase (EC 2.1.1.45) prompted us to include the propargyl group in a study of the effect on folate metabolism and membrane transport of replacing the 10-methyl group of methotrexate with other groups. Along with the propyl (8a) and octyl (8b) homologues of methotrexate, the propargyl compound 8c was prepared for evaluation. Syntheses of 8a,b were achieved by a standard multistep sequence involving preparation of the side-chain precursors via tosylated intermediates and then their alkylation with 6-(bromomethyl)-2,4-pteridinediamine hydrobromide. The side-chain precursor to 8c was prepared by direct alkylation of diethyl N-(4-aminobenzoyl)-L-glutamate with propargyl bromide and was separated from unchanged amine and dipropargyl coproduct by a combination of methods, including dry-column chromatography and recrystallization. Subsequent steps leading to 8c were like those used to prepare 8a,b. Biological evaluations of the three compounds consisted of studies of their effects on enzyme inhibition [(dihydrofolate reductase (EC 1.5.1.3) and thymidylate synthase)], L1210 cell growth inhibition, cellular membrane transport with various murine cell types (L210, S180, Ehrlich, and epithelial), in vivo (mice) activity vs. L1210 leukemia and S180 ascites, and plasma clearance in mice. The in vivo results vs. S180 ascites offered evidence that 8c might have a better therapeutic index against this tumor than methotrexate, but no other result from either of these compounds suggested significant superiority over methotrexate.

Further studies stereospecificity at carbon 6 for membrane transport of tetrahydrofolates. Diastereoisomers of 5-methyltetrahydrofolates as competitive inhibitors of transport of methotrexate in L1210 cells

The unnatural d diastereoisomer at carbon 6 of 5-methyltetrahydrofolate was only slightly less effective than the natural 1 diastereoisomer as a competitive inhibitor of the carrier-mediated membrane transport of [3H]methotrexate into L1210 murine leukemia cells. The apparent Ki for a mixture containing equal amounts of both natural and unnatural diastereoisomers was not significantly different from that found for the unnatural form. These results show that the reduced folate carrier system in these cells has a strong affinity for the unnatural stereoisomer, a finding in contrast to that obtained with the corresponding diastereoisomer of 5-formyltetrahydrofolate.

Folate analogues altered in the C9-N10 bridge region. 18. Synthesis and antitumor evaluation of 11-oxahomoaminopterin and related compounds

The chemical synthesis of 11-oxahomoaminopterin (1) has been carried out using procedures which were also found to be applicable to the synthesis of 11-oxahomofolic acid (2). Reaction of 1-bromo-4-[p-(caarbomethoxy)phenoxy]-2-butanone (10) with sodium azide gave 1-azido-4-[p-(carbomethoxy)phenoxy]-2-butanone (11). Protection of the carbonyl group of 11 as the ethylene ketal and subsequent base hydrolysis of the product gave 1-azido-4-(p-carboxyphenoxy)-2-butanone ketal (13). The glutamate conjugate 14 was prepared from 13 by the isobutyl chloroformate method and was hydrogenated to diethyl N-[(alpha-amino-2-oxo-4-butanoyl)-p-anisoyl]-L-glutamate ketal (15). Reaction of 15 with 6-chloro-2,4-diamino-5-nitropyrimidine (16) and 2-amino-6-chloro-4-hydroxy-5-nitropyrimidine (17) and deprotection of the corresponding products gave the intermediates 18 and 19, which were elaborated to 1 and 2 using a series of steps involving deprotection, dithionite reduction, cyclization, oxidation, and hydrolysis. Although 11-oxahomoaminopterin showed antifolate activity against two folate-requiring microorganisms and inhibited Lactobacillus casei DHFR, it was inactive against L-1210 leukemia in mice at a maximum dose of 48 mg/kg. Compound Lactobacillus casei DHFR, it was inactive against L-1210 leukemia in mice at a maximum dose of 48 mg/kg. Compound 1 was also tested for its ability to be transported via the methotrexate transport system using the L-1210 and Ehrlich tumor cell lines, and these results are compared with those of related analogues. The growth inhibitory activity of 1 in the L-1210 cell lines in culture was found to be 15 times weaker than that of methotrexate.

Polyglutamyl derivatives of tetrahydrofolate as substrates for Lactobacillus casei thymidylate synthase

Tetrahydropteroylpolyglutamates containing up to seven Glu residues were tested as substrates for Lactobacillus casei thymidylate synthase. The Km values decreased from 24 microM for the monoglutamate to 1.8 microM for the triglutamate. Addition of residues 4, 5, 6, and 7 did not decrease the Km further. When monoglutamate and polyglutamate substrates were simultaneously incubated with the enzyme, the rate observed was characteristic of the polyglutamate even when the monoglutamate concentration was 44 times that of the polyglutamate. Iodoacetamide treatment inhibited the enzyme to the same extent with monoglutamate and polyglutamate substrates. Addition of 0.3 M NaCl doubled the rate obtained with the polyglutamate substrate whereas the rate with the monoglutamate was inhibited 25%. MgCl2 stimulated the reaction only 10% with the polyglutamate substrate compared with 80% stimulation obtained with the monoglutamate. Inhibition by fluorodeoxyuridylate was similar with both mono- and polyglutamate substrates; however, with the phosphonate derivative of fluorodeoxyuridine, the polyglutamate substrate enhanced inhibition 5- to 8-fold.

Folate analogues altered in the C9-N10 bridge region. 16. Synthesis and antifolate activity of 11-thiohomoaminopterin

The synthesis of 11-thiohomoaminopterin (1), which is a close analogue of 11-thiohomofolic acid (2), has been carried out by modification of the Boon-Leigh procedure. Treatment of 1-chloro-4-[p-(carbomethoxy)thiopenoxy]-2-butanone (5) with sodium azide gave 1-azido-4-[p-(carbomethoxy)thiophenoxy]-2-butanone (6). After protection of the carbonyl group of 6, the product 7 was catalytically hydrogenated to 1-amino-4-[p-(carbomethoxy)thiophenoxy]-2-butanone ketal (3). Reaction of 32 with 6-chloro-2,4-diaminmo-5-nitropyrimidine gave the desired pyrimidine intermediate, which was elaborated to 4-amino-4-deoxy-11-thiohomopteroic acid (20) by standard procedures. Alternately, 1-azido-4-[p-(carbomethoxy)thiophenoxy]-2-butanone ketal (7) was hydrolyzed to the corresponding acid (8) and coupled with diethyl L-glutamate to obtain diethyl N-[p-(1-azido-2-oxo-4-thiobutanoyl)benzoyl]-L-glutamate ketal (10), which was used for the large-scale preparation of 11-thiohomoaminopterin (1). Although 11-thiohomoaminopterin showed antifolate activity against two folate-requiring microorganisms and inhibited Lactobacillus casei dihydrogolate reductase, it did not exhibit any antitumor activity against L-1210 lymphoid leukemia in mice at a maximum dose of 48 mg/kg.

Folate analogues altered in the C9-N10 bridge region: 11-thiohomofolic acid

The synthesis of 11-thiohomofolic acid (2) has been accomplished by an unambiguous procedure. Reaction of 1-chloro-4-[p-(carbomethoxy)thiophenoxy]-2-butanone (10) with hydroxylamine under carefully controlled conditions gave the corresponding oxime 33. Conversion of this oxime to 1-phthalimido-4-[p-(carbomethoxy)thiophenoxy]-2-butanone oxime (4) was carried out by its reaction with potassium phthalimide using crown 18 ether as a catalyst. Hydrazinolysis of compound 4 gave 1-amino-4-[p-(carbomethoxy)thiophenoxy]-2-butanone oxime (5), which was used for the construction of the title compound 2 by modification of the Boon and Leigh procedure. An alternate synthesis utilizing 1-hydroxy-4-[p-(carbomethoxy)thiophenoxy]-2-butanone (11) and 4-hydroxy-2,5,6-triaminopyrimidine has also been carried out. Compound 2 did not exhibit any antifolate activity against Lactobacillus casei or Streptococcus faecium. The dithionite reduction product, 7,8-dihydro-11-thiohomofolic acid, was able to function as a substrate of L. casei dihydrofolate reductase. The catalytic reduction product of 2, consisting of a mixture of diastereomers, exhibited powerful antifolate activity against both these organisms.

Methotrexate analogues. 12. Synthesis and biological properties of some aza homologues

Methotrexate analogues, in which an additional nitrogen atom is inserted between the phenyl ring and the carbonyl group of the side chain, were prepared by photochemical methods. The compounds were less inhibitory toward dihydrofolate reductase and thymidylate synthetase derived from Lactobacillus casei than was methotrexate. They were also less cytotoxic against human lymphoblastic leukemia cells (CCRF-CEM). In vivo against L-1210 leukemia in mice, the aza homologue of methotrexate showed significant antitumor activity (%ILS = 55%) compared to methotrexate (%ILS = 88%).

Synthesis of aza homologues of folic acid

Folic acid analogues containing an additional nitrogen atom between the phenyl ring and the carbonyl group of the side chain were synthesized. None of the compounds showed significant inhibitory activity against human lymphoblastic leukemia cells (CCRF-CEM) in culture or against Lactobacillus casei (ATCC 7469) growth. Against L1210 leukemia in mice, the aza homologue of folic acid, 4, and the aspartic acid analogue, 14, showed no increase in life span over control animals. These compounds were more toxic in vivo than the corresponding methotrexate analogues. Compound 4 supported the growth of Streptococcus faecium (ATCC 8043), and its tetrahydro derivative supported the growth of Pediococcus cerevisiae (ATCC 8081). These results strongly suggest that 4 can substitute for folate derivatives as cofactors for serine transhydroxymethylase, thymidylate synthetase, and dihydrofolate reductase.

Phosphonate analogue of 2'-deoxy-5-fluorouridylic acid

The phosphonate analogue (6) of 2'-deoxy-5-fluorouridylic acid has been prepared via a Pfitzner--Moffatt oxidation and Witting reaction. This compound was found to inhibit thymidylate synthetase from three sources and to be cytotoxic to H.Ep.-2 cells in culture.

Nalpha-(pteroyltetra (gamma-glutamyl))-lysine as a ligand for the purification of thymidylate synthetase by affinity chromatography

Nalpha-(pteroyltetra (gamma-glutamyl))-lysine Sepharose was synthesized and shown to be a stable high capacity affinity matrix capable of bringing about the purification of Lactobacillus casei thymidylate synthetase to maximum specific activity from crude extracts in high yield. Under conditions optimal for binding of thymidylate synthetase, dihydrofolate reductase was not bound.

Diastereoisomers of 5,10-methylene-5,6,7,8-tetrahydropteroyl-D-glutamic acid

The diastereoisomers of 5,10-methylene 5,6,7,8-tetrahydropteroyl-D-glutamate were resolved and tested as substrates and inhibitors of Lactobacillus casei thymidylate synthetase. No activity was observed. The compounds were neither growth factors nor inhibitors for Lactobacillus casei, Streptococcus faecium, or Pediococcus cerevisiae. 7,8-Dihydropteroyl-D-glutamate is 50% as active as 7,8-dihydropteroyl-L-glutamate (dihydrofolate) as a substrate for L. casei dihydrofolate reductase.

Aspects of the reversal of methotrexate toxicity in rodents

An ip injection of a Lactobacillus casei dihydrofolate reductase preparation into rats and mice given a single lethal dose of methotrexate (MTX) caused a marked lowering of free MTX in the blood. Alternatives to citrovorum factor as agents for reversing MTX toxicity were explored in mice. dl, L-5-Methyltetrahydrofolate, dl,L-5,10-methylenetetrahydrofolate, l,L-5,10-methylenetetrahydrofolate, and dihydrofolate were also effective MTX antagonists; d,L-5,10-methylenetetrahydrofolate was inert.

Synthesis and biological activity of 10-thia-10-deaza analogs of folic acid, pteroic acid, and related compounds

The 10-thia analogs of pteroic acid, folic acid, their esters, and their 4-amino analogs were synthesized through a reaction sequence involving, as a key step, the condensation of 2-amino-3-cyano-5-chloromethylpyrazine with appropriately substituted thiols. The abilities of the products to inhibit the growth of methotrexate (MTX)-sensitive and MTX-resistant microorganisms were investigated as were their abilities to inhibit dihydrofolic acid reductase and thymidylic acid synthetase. Several compounds had high activity.