Design and synthesis of potent non-polyglutamatable quinazoline antifolate thymidylate synthase inhibitors.

The synthesis is described of a series of analogues of the potent thymidylate synthase (TS) inhibitor, N-[4-[N-[(3,4-dihydro-2, 7-dimethyl-4-oxo-6-quinazolinyl)methyl]-N-prop-2-ynylamino]-2-f luorob enzoyl]-L-glutamic acid (4, ZM214888), in which the glutamic acid moiety is replaced by homologous amino acids and alpha-amino acids where the omega-carboxylate is replaced by acylsulfonamides and acidic heterocycles. In general these modifications when compared to 4 gave compounds with increased potency as inhibitors of isolated TS and as cytotoxic agents against murine tumor cell lines. The new compounds require transport by the reduced folate carrier for entry into cells but are not converted intracellularly into polyglutamated species. Agents with this profile are expected to show activity against tumors that are resistant to classical antifolates due to low expression of folylpolyglutamate synthetase. The analogue (S)-2-[4-[N-[(3,4-dihydro-2, 7-dimethyl-4-oxo-6-quinazolinyl)methyl]-N-prop-2-ynylamino]-2-f luorob enzamido]-4-(1H-1,2,3,4-tetrazol-5-yl)butyric acid (35, ZD9331) has been selected as a clinical development candidate and is currently undergoing phase I studies.

Quinazoline antifolates thymidylate synthase inhibitors: lipophilic analogues with modification to the C2-methyl substituent.

Modification of the potent thymidylate synthase (TS) inhibitor 1-[[N-[4-[N-[(3,4-dihydro-2-methyl-4-oxo-6-quinazolinyl)methyl]-N- prop-2-ynylamino]benzoyl]amino]methyl]-3-nitrobenzene (4a) has led to the synthesis of quinazolinone antifolates bearing functionalized alkyl substituents at C2. A general synthetic route was developed which involved coupling the appropriate 1-[[N-[4-(alkylamino)benzoyl)amino]methyl]-3-nitrobenzene 20-22 with a 6-(bromomethyl)-2-(acetoxymethyl)-3,4-dihydro-4-oxoquinazoline 9 or 10. Replacement of the 2-acetoxy group by a chlorine atom followed by the displacement of the halogen of 25a-c by various nucleophiles led to compounds 26-40. Good TS (IC50 < 1 microM) and growth inhibition (IC50 0.1-1 microM) were found with most of these new antifolates. TS inhibitors in this series do not apparently require the reduced folate carrier (RFC) for cell entry (they most likely penetrate the cell membrane by passive diffusion) and are not polyglutamated. N, O, S, Cl, and CN as well as large amino and mercapto substituents were tolerated by the enzyme. The simultaneous incorporation of 7-methyl and 2'-F substituents gave a series of highly potent agents inhibiting cell growth at concentrations < 1 microM (24, 27bc; 30-32b, 35b). The incorporation of suitable C2 substituents has overcome the decrease in aqueous solubility observed with lipophilic quinazoline antifolates. This is best illustrated by compound 31a, where up to a 54-fold increase in solubility has been achieved by the incorporation of an N-methylpiperazine nucleus into the C2-methyl group of 4a.

Quinazoline antifolate thymidylate synthase inhibitors: replacement of glutamic acid in the C2-methyl series.

The synthesis of a series of analogues of the potent thymidylate synthase (TS) inhibitor N-[4-[N-[(3,4-dihydro-2-methyl-4-oxo-6- quinazolinyl)methyl]-N-prop-2-ynylamino]benzoyl]-L-glutamic acid (ICI 198583, 1) is described in which the glutamic acid residue has been replaced by other alpha-amino acids. Most of these analogues were prepared by coupling of tert-butyl-4-(prop-2-ynylamino)benzoate (37) with 6-(bromomethyl)-3,4-dihydro-2-methyl-4-oxoquinazoline (34) followed by deprotection of the tert-butyl ester to the acid and azide-mediated coupling to the appropriate amino acid or amino acid ester. In cases where the amino acid ester was unreactive with the acid azide, a modification was used in which the quinazolinone moiety was protected as its 3-(pivaloyloxy)methyl derivative. This permitted the generation of the more reactive acid chloride of the p-aminobenzoate unit. In general these modifications result in compounds that have equivalent potency to 1 as inhibitors of isolated TS except where the amino acid lacks a lipophilic alpha-substituent. These compounds appear to require the reduced folate carrier (RFC) for transport into cells, but since they are not converted intracellularly into polyglutamated forms, they have a lower level of cytotoxicity compared to 1. The removal of the alpha-carboxylic acid has given a second set of analogues of 1 which contain simple alkyl amide, benzyl, substituted benzyl, and heterocyclic benzyl amide derivatives. These are considerably less potent than 1 as TS inhibitors but display 1-10 microM cytotoxicities due to the fact that they do not require RFC transport and can presumably readily enter cells by passive diffusion through the cell membrane. Molecular modeling and NMR studies indicated that the incorporation of, respectively, 7-methyl and 2'-fluoro substituents would favor the optimum conformation of these molecules for interaction with the TS enzyme. Accordingly, these substituents were incorporated into selected examples to give the series of analogues 47-55. These all show enhanced (approximately 10-fold) inhibition of TS compared to their unsubstituted counterparts. In the substituted benzylamides (51, 52) and heterocyclic benzyl amides (53-55) the ability to enter cells by passive diffusion results in highly potent (< 1 microM) cytotoxic agents.

The measurement of polyglutamate metabolites of the thymidylate synthase inhibitor, ICI D1694, in mouse and human cultured cells.

A method is described for the measurement of the polyglutamates of the quinazoline thymidylate synthase inhibitor, N-(5-[N-(3,4-dihydro-2-methyl-4-oxoquinazolin- 6-ylmethyl)-N-methylamino]-2-theonyl)-L-glutamic acid (ICI D1694). This involved incubation of cells with [5-3H]ICI D1694, extraction of the polyglutamates and their analysis by HPLC using an ion-pairing method. Co-chromatography with ICI D1694 and its synthetic di-hexaglutamate standards (UV detection) aided identification of the [3H]polyglutamates in the fractions recovered from the HPLC. Recovery of the polyglutamates at each stage of extraction and analysis was very good (77-84% overall recovery). Polyglutamates readily accumulated as the tri-, tetra and penta forms and occasionally a small amount of hexaglutamate was found. After mouse L1210 leukemia or human W1L2 lymphoblastoid cells were incubated for 30 min with 0.1 microM [3H]ICI D1694 there was a approximately 6-fold concentration effect intracellularly with most of the 3H associated with polyglutamate forms (approximately 75% and 96% for the L1210 and W1L2, respectively). Even some of the higher chain length tetra- and pentaglutamates could be detected at this time. After 4 hr incubation the total level of intracellular 3H had risen to 2-3 microM, greater than 96% of which was associated with polyglutamates (mainly tetra- and pentaglutamates). Four other human cell lines, two ovarian (CH1 and 41M), the MCF-7 breast and the HT-29 colon, were examined for their ability to form intracellular polyglutamates. A 4 hr incubation with 0.1 microM [3H]ICI D1694 resulted in a substantial intracellular accumulation of the drug (20-100-fold) in its polyglutamate forms with only 2-20% remaining as the parent monoglutamate, depending on the cell line. The major polyglutamate was again cell line dependent, ranging from the tri to the penta form. Prolonging the incubation time to 24 hr allowed a further accumulation of drug with a larger percentage appearing as tri- to hexaglutamates. Although cell lines differed in the total level of polyglutamates formed and the pattern of chain length observed, rapid and extensive polyglutamation of ICI D1694 occurred in all the cell types examined.

Quinazoline antifolate thymidylate synthase inhibitors: difluoro-substituted benzene ring analogues.

The synthesis of a series of new C2-methyl-N10-alkylquinazoline-based thymidylate synthase (TS) inhibitors containing difluroinated p-aminobenzoate rings is described. Derivatives of the N10-propargyl and N10-methylquinazoline antifolates were prepared with 2',3'-, 2',5'-, and 2',6'-difluoro substitution. The synthesis of the 2',5'-difluoro analogues involved oxidation of the difluoronitrotoluene to 2,5-difluoro-4-nitrobenzoic acid followed by glutamation, reduction, and alkylation (propargyl bromide or MeI) to the diethyl N-(4-(alkylamino)-2,5-difluorobenzoyl)-L-glutamates. For the synthesis of the 2',3'- and 2',6'-difluoro compounds a new route was devised starting from methyl 4-((tert-butoxycarbonyl)amino)-2,6-difluorobenzoate and its 2,3-substituted counterpart. Treatment with NaH and then an alkyl halide introduced the N10-substituent. The methyl ester was hydrolyzed and the resulting acid was condensed with diethyl L-glutamate. The secondary amine was liberated using CF3CO2H and coupled with 6-(bromo-methyl)-3,4-dihydro-2-methyl-4-oxoquinazoline to yield the antifolate diesters. Final deprotection with mild alkali completed the synthesis in each case. The target compounds were tested as inhibitors of partially purified L1210 TS and also examined for their inhibition of the growth of L1210 cells in culture. Compared to their nonfluorinated parent compounds all the difluoro analogues were poorer inhibitors of TS. The greatest loss of enzyme activity was seen in the N10-propargyl analogues which contained one of the fluorine atoms ortho to the amine substituent. This loss was less apparent in the N10-methyl derivatives. Despite this lower inhibition of TS the majority of new compounds have equivalent cytotoxicity to their nonfluorinated predecessors.

Quinazoline antifolate thymidylate synthase inhibitors: bridge modifications and conformationally restricted analogues in the C2-methyl series.

Several C2-methylquinazoline-based antifolates have been prepared in which the C9,N10 bridge has been replaced by the reversed N9,C10 unit. This series was extensively studied by incorporating further substituents at N9 and C10 as well as by modifications to the p-aminobenzoate ring. The C2-methylquinazoline analogues 29, 30, and 31 containing the methyleneoxa, methylenethia, and thia bridge units were also synthesized. In general these isosteric replacements of the bridge unit in the parent C2-methyl-N10-propargylquinazoline antifolate 2 were much less potent as inhibitors of isolated thymidylate synthase (TS) but several were at least as potent as inhibitors of L1210 cell growth in culture. The fusion of the p-aminobenzoate ring into the bicyclic systems 75 and 76 also reduced activity against TS but again gave highly cytotoxic compounds. The cytotoxicities were largely prevented by thymidine, confirming that TS is the major locus.

Quinazoline antifolate thymidylate synthase inhibitors: heterocyclic benzoyl ring modifications.

The synthesis is described of a series of C2-methyl-N10-alkylquinazoline-based antifolates in which the p-aminobenzoate ring is replaced by the heterocycles thiophene, thiazole, thiadiazole, pyridine, and pyrimidine. These were generally elaborated by the reaction of (bromomethyl)quinazoline 18 or its N3-[(pivaloyloxy)methyl]-protected derivative 36 with suitable heterocyclic amines although each heterocyclic system required its own particular synthetic approach. The compounds were tested as inhibitors of partially purified L1210 thymidylate synthase (TS). They were also examined for their inhibition of the growth of L1210 cells in culture. The thiophene system 7 and its related thiazole 8 gave analogues that were considerably more potent than the parent benzene series 2 as inhibitors of L1210 cell growth although in general these heterocycles were somewhat poorer inhibitors of the isolated TS enzyme. The enhanced cytotoxicities of the thiophene and thiazole analogues result, at least in part, from their efficient transport into the cells via the reduced folate carrier mechanism and very good substrate activity for folylpolyglutamate synthetase. The replacement of the C2-methyl group by C2-(fluoromethyl) and C2-(hydroxymethyl) substituents in the thiophene and thiazole series gave derivatives that were only slightly less potent inhibitors of the TS enzyme but which were considerably less cytotoxic.

Thymidylate synthase inhibitors: the in vitro activity of a series of heterocyclic benzoyl ring modified 2-desamino-2-methyl-N10-substituted-5,8-dideazafolates.

Heterocyclic para-aminobenzoate modifications of 2-desamino-2-methyl-5,8-dideazafolic acid and a series of its N10-substituted analogs have produced a number of interesting compounds that have enabled a deeper understanding of the biochemical events required for activity in this class of antimetabolite. There is a relationship that has become apparent between compound potency and both uptake via the reduced-folate carrier and FPGS substrate activity. Rapid cellular uptake and metabolism of polyglutamate forms that are approximately 100-fold more potent as inhibitors of TS can translate a modest TS inhibitor such as ICI D1694 into a very potent inhibitor of cell growth (approximately 500- and approximately 10-fold more potent than CB3717 or ICI 198583, respectively). Polyglutamation may therefore act as an almost essential activation step and ICI D1694 may be highly specific for tumors expressing both the reduced-folate carrier and FPGS. Polyglutamation of folate analogs also leads to drug retention which may play a major role in the pharmacodynamics of TS inhibition by ICI D1694 in vivo. Current studies with 3H-ICI D1694 are aimed at demonstrating metabolism to polyglutamates in tumor cells. The serious toxic limitations of CB3717, i.e., liver and kidney toxicities, are not seen with ICI D1694 reflecting the good water solubility of the drug compared with CB3717. The toxicities observed in mice are however to hematological tissues and are due to its TS inhibitory effects. Thus ICI D1694 may elicit toxicities in man more typical of an antimetabolite than of CB3717. The clinical evaluation of ICI D1694 may further our understanding of the role that metabolism to polyglutamates may have in therapeutic activity.

Quinazoline antifolate thymidylate synthase inhibitors: alkyl, substituted alkyl, and aryl substituents in the C2 position.

Modification of the potent thymidylate synthase (TS) inhibitor N-[4-[N-[(2-amino-3,4-dihydro-4-oxo-6-quinazolinyl)methyl]-N-prop-2- ynylamino]benzoyl]-L-glutamic acid (1a) has led to the synthesis of quinazoline antifolates bearing alkyl, substituted alkyl, and aryl substituents at C2. In general the synthetic route involved the coupling of the appropriate diethyl N-[4-(alkylamino)benzoyl]-L-glutamate with a C2-substituted 6-(bromo-methyl)-3,4-dihydro-4-oxoquinazoline followed by deprotection using mild alkali. Good enzyme inhibition and cytotoxicity were found with compounds containing small nonpolar groups in the C2 position with the 2-desamino-2-methyl analogue 3a being the most potent. Larger C2 substituents were tolerated by the enzyme, but cytotoxicity was reduced. Highly potent series were followed up by the synthesis of a number of analogues in which the N10 substituent was varied. In this manner a number of interesting TS inhibitors have been prepared. Although none of these was more potent than 1a against the isolated enzyme, over half of the compounds prepared were more potent as cytotoxic agents against L1210 cells in culture. The potential of such compounds as useful antitumor agents was further enhanced by the finding that the improved aqueous solubilities of compounds such as 3a over 1a were reflected in vivo in that 3a was at least 5 times less toxic to mice than 1a.

Quinazoline antifolate thymidylate synthase inhibitors: 2'-fluoro-N10-propargyl-5,8-dideazafolic acid and derivatives with modifications in the C2 position.

The synthesis of 2'-fluoro-10-propargyl-5,8-dideazafolic acid and its 2-desamino, 2-desamino-2-hydroxymethyl, and 2-desamino-2-methoxy analogues is described. In general the synthetic route involved the coupling of diethyl N-[2-fluoro-4-(prop-2-ynylamino)benzoyl]-L-glutamate with the appropriate 6-(bromomethyl)quinazoline followed by deprotection with mild alkali. These four compounds together with the 2-desamino-2-methyl analogue were tested for their activity against L1210 thymidylate synthase (TS). They were also examined for their inhibition of the growth of the L1210 cell line and of two mutant L1210 cell lines, the L1210:R7A that overproduces dihydrofolate reductase (DHFR) and the L1210:1565 that has impaired uptake of reduced folates. Compared with their non-fluorinated parent compounds, the 2'-fluoro analogues were all approximately 2-fold more potent as TS inhibitors. Similarly, they also showed improved inhibition of L1210 cell growth (1.5-5-fold), and this activity was prevented by co-incubation with thymidine. All had retained or improved activity against both the L1210:R7A and L1210:1565 cell lines.

Quinazoline antifolate thymidylate synthase inhibitors: benzoyl ring modifications in the C2-methyl series.

The synthesis of nine new 2-methyl-10-propargylquinazoline antifolates with substituents in the p-aminobenzoyl ring is described. In general the synthetic route involved the coupling of the appropriate ring-substituted diethyl N-[4-(prop-2-ynylamino)benzoyl]-L-glutamate with 6-(bromomethyl)-3,4-dihydro-2-methyl-4-oxoquinazoline followed by deprotection using mild alkali. The compounds were tested as inhibitors of partially purified L1210 thymidylate synthase (TS). They were also examined for their inhibition of the growth L1210 cells in culture. Compared to the parent compound 1a the 2'-fluoro analogue 2a exhibited enhanced potency in both systems whereas the 3'-fluoro analogue 3a showed enhanced growth inhibitory properties against L1210 cells despite being a poorer inhibitor of the isolated enzyme. Chloro, hydroxy, methoxy, and nitro substituents in the 2'-position were also well tolerated by the enzyme but failed to give enhanced growth inhibition. The series was extended to cover analogues of the 2'-fluoro, 3'-fluoro, 2'-chloro, 2'-methyl, 2'-amino, 2'-methoxy, and 2'-nitro derivatives with modified alkyl substituents at N10.

Quinazoline antifolate thymidylate synthase inhibitors: nitrogen, oxygen, sulfur, and chlorine substituents in the C2 position.

The synthesis of 16 new N10-propargylquinazoline antifolates with methylamino, ethylamino, (2-aminoethyl)amino, [2-(dimethylamino)ethyl]amino, (2-hydroxyethyl)amino, (carboxymethyl)amino, dimethylamino, imidazol-1-yl, methoxy, ethoxy, phenoxy, 2-methoxyethoxy, 2-hydroxyethoxy, mercapto, methylthio, and chloro substituents at C2 is described. In general, the synthetic route involved the coupling of diethyl N-[4-(prop-2-ynylamino)benzoyl]-L-glutamate (5a) with 6-(bromomethyl)-2-chloro-3,4-dihydro-4-oxoquinazoline in N,N-dimethylformamide with calcium carbonate as the base, displacement of the C2-chloro substituent with nitrogen and sulfur nucleophiles, and deprotection using mild alkali. The C2-ether analogues were most conveniently prepared by coupling 5a with 6-(bromomethyl)-2,4-diakoxy(or diphenoxy)quinazolines. In this series the final deprotection step with aqueous alkali gave simultaneous selective hydrolysis of the C4-alkoxy or C4-phenoxy substituent. The compounds were tested as inhibitors of partially purified L1210 thymidylate synthase (TS). As a measure of cytotoxicity, they were examined for their inhibition of the growth of L1210 cells in culture. The C2-methoxy analogue 11a was equivalent to the previously described tight binding TS inhibitor N10-propargyl-5,8-dideazafolic acid (CB3717, ICI 155387, 1a) against the TS enzyme and exhibited enhanced potency in culture. The C2-methoxy substituent also gave a 110-fold enhancement in aqueous solubility relative to the C2-amine. These results suggest that 11a will be an interesting compound for further study as a potential antitumor agent in vivo. A further series of 2-methoxyquinazoline antifolates with modified alkyl substituents at N10 is also described. None of these analogues equalled the activity of 11a. Thus the propargyl group appears to be the optimum N10 substituent in both 2-amino- and 2-methoxyquinazoline antifolates.