Search

Mammalians are devoid of autonomous biosynthesis of folates and hence must obtain them from the diet. Reduced folate cofactors are B9-vitamins which play a key role as donors of one-carbon units in the biosynthesis of purine nucleotides, thymidylate and amino acids as well as in a multitude of methylation reactions including DNA, RNA, histone and non-histone proteins, phospholipids, as well as intermediate metabolites. The products of these S-adenosylmethionine (SAM)-dependent methylations are involved in the regulation of key biological processes including transcription, translation and intracellular signaling. Folate-dependent one-carbon metabolism occurs in several subcellular compartments including the cytoplasm, mitochondria, and nucleus. Since folates are essential for DNA replication, intracellular folate cofactors play a central role in cancer biology and inflammatory autoimmune disorders. In this respect, various folate-dependent enzymes catalyzing nucleotide biosynthesis have been targeted by specific folate antagonists known as antifolates. Currently, antifolates are used in drug treatment of multiple human cancers, non-malignant chronic inflammatory disorders as well as bacterial and parasitic infections. An obligatory key component of intracellular folate retention and intracellular homeostasis is (anti)folate polyglutamylation, mediated by the unique enzyme folylpoly-γ-glutamate synthetase (FPGS), which resides in both the cytoplasm and mitochondria. Consistently, knockout of the FPGS gene in mice results in embryonic lethality. FPGS catalyzes the addition of a long polyglutamate chain to folates and antifolates, hence rendering them polyanions which are efficiently retained in the cell and are now bound with enhanced affinity by various folate-dependent enzymes. The current review highlights the crucial role that FPGS plays in maintenance of folate homeostasis under physiological conditions and delineates the plethora of the molecular mechanisms underlying loss of FPGS function and consequent antifolate resistance in cancer., (Copyright © 2016 Elsevier Ltd. All rights reserved.)

Folylpoly-gamma-glutamate synthetase (FPGS) is the enzyme responsible for metabolic trapping of reduced folate cofactors in cells for use in nucleotide and amino acid biosynthesis. There are two isoforms of FPGS expressed in mouse tissues, one is expressed in differentiated tissue, principally liver and kidney, and the other in all rapidly proliferating cell types. The present study sought the functional difference that would explain the evolution of two mouse FPGS species. Recombinant cytosolic mouse isozymes were compared with respect to steady state kinetics, chain length of polyglutamate derivatives formed, and end-product inhibition by the major reduced folylpentaglutamate cofactors. Both isoforms were equally effective in catalyzing the addition of a mole of glutamic acid to reduced folate monoglutamate substrates. Each isoform was also capable of forming long chain polyglutamate derivatives of the model folate, 5,10-dideazatetrahydrofolate. In contrast, the FPGS isoform derived from rapidly proliferating tissue was much more sensitive to inhibition by (6R)-5,10-CH(2)-H(4)PteGlu(5) and (6S)-H(4)PteGlu(5) than the isoform expressed in differentiated tissues, as demonstrated by 13- and 6-fold lower inhibition constants (K(i)), respectively. Interestingly, each isozyme was equally sensitive to inhibition by (6R)-10-CHO-H(4)PteGlu(5). We drew the conclusion that the decreased sensitivity of the FPGS expressed in mouse liver and kidney to feedback inhibition by 5,10-CH(2)-H(4)PteGlu(5-6) and H(4)PteGlu(5-6) may have evolved to permit accumulation of a larger folate cofactor pool than that found within rapidly proliferating tissue.

Antifolate drugs are molecules directed to interfere with the folate metabolic pathway at some level. They can be recognized among the first rationally designed compounds applying the principle of structural analogy with the substrate developing the antimetabolite strategy. This strategy has taken advantage of the basic different features of the microbial and human folate metabolism and therefore allows targeting the pathway at different steps generating a specificity tools for Medicinal Chemists. Two main problems are giving renewed importance to such targets and therefore improving the efforts to discover new targets in the folate metabolism area. The first one is the increasing resistance to the present drugs due to different mechanisms such as the enzyme modification and the increased production of enzymes with not well recognized importance. The second one is the development of techniques directed to highlight the interference at genetic level of molecular probes as antifolate drug to develop new enzymes previously unknown. This approach is defined as genetic approach to drug discovery, from gene to drugs. The present article describes the importance in drug design and discovery of some antifolate targets among the best known at the present status of research such as thymidylate synthase (TS), dhydrofolate reductases, (DHFR) serine hydroxymethyltransferase (SHMT), folyilpolyglutamyl synthetase (FPGS), gamma-glutamyl hydrolase (gamma-GH), glycinamide-ribonucleotide transformylase (GARTfase), amino-imidazole-carboxamide-ribonucleotide transformylase (AICARTfase) and Folate transporters. Discovery, known functions, structure/function studies and inhibition will be described.

Rotationally restricted analogues of 5-deazapteroyl-L-glutamate and (6R,6S)-5-deaza-5,6,7,8-tetrahydropteroyl-L-glutamate with a one-carbon bridge between the amide nitrogen and the 6'-position of the p-aminobenzoyl moiety were synthesized and tested as substrates for folylpolyglutamate synthetase (FPGS), a key enzyme in folate metabolism and an important determinant of the therapeutic potency and selectivity of classical antifolates. The corresponding bridged analogues of 5-deazapteroyl-L-ornithine and (6R,6S)-5-deaza-5,6,7, 8-tetrahydropteroyl-L-ornithine were also synthesized as potential inhibitors. Condensation of diethyl L-glutamate with methyl 2-bromomethyl-4-nitrobenzoate followed by catalytic reduction of the nitro group, reductive coupling with 2-acetamido-6-formylpyrido[2, 3-d]pyrimidin-4(3H)-one in the presence of dimethylaminoborane, and acidolysis with HBr/AcOH yielded 2-L-[5-[N-(2-acetamido-4(3H)-oxopyrido[2, 3-d]pyrimidin-6-yl)methylamino]-2, 3-dihydro-1-oxo-2(1H)-isoindolyl]glutaric acid (1). When acidolysis was preceded by catalytic hydrogenation, the final product was the corresponding (6R,6S)-tetrahydro derivative 2. A similar sequence starting from methyl N(delta)-benzyloxycarbonyl-L-ornithine led to 2-L-[5-[N-(2-amino-4(3H)-oxopyrido[2, 3-d]pyrimidin-6-yl)methylamino]-2, 3-dihydro-1-oxo-2(1H)-isoindolyl]-5-aminopentanoic acid (3) and the (6R,6S)-tetrahydro derivative 4. Compounds 3 and 4 were powerful inhibitors of recombinant human FPGS, whereas 1 and 2 were exceptionally efficient FPGS substrates, with the reduced compound 2 giving a K(m) (0.018 microM) lower than that of any other substrate identified to date. (6R,6S)-5-Deazatetrahydrofolate, in which the side chain is free to rotate, was rapidly converted to long-chain polyglutamates. In contrast, the reaction of 1 and 2 was limited to the addition of a single molecule of glutamic acid. Hence rotational restriction of the side chain did not interfere with the initial FPGS-catalyzed reaction and indeed seemed to facilitate it, but the ensuing gamma-glutamyl adduct was no longer an efficient substrate for the enzyme.

Folypolyglutamate synthetase (FPGS) is responsible for the conversion of naturally occurring folates and antifolates to their poly-gama-glutamyl derivatives, which are the forms required for intracellular retention of folates and are also the preferred substrates (cofactors) for most folate-dependent enzymes. Folate and methotrexate analogues 6 and 4, with L-histidine in place of L-glutamate, were designed and synthesized as potential FPGS inhibitors. Target compound 5, the N tau-(carboxymethyl)-L-histidine derivative of 4, was also prepared. Compounds 4 and 5 inhibited the growth of L1210 cells (IC50 values: 0.091 and 0.15 microM, respectively) and were potent inhibitors of L1210 dihydrofolate reductase. No significant inhibition of FPGS by 4, 5, or 6 was observed at the high pH of the standard enzyme assay. This could be the consequence of a lack of protonation of the basic side chains, which is likely to be required for FPGS inhibitory activity. The observed cytotoxicity indicates that partial protonation of the imidazole ring permits cellular uptake of the analogues.

Many quinazoline thymidylate synthase (TS) inhibitors undergo intracellular metabolism to polyglutamate forms which can significantly alter their activity and pharmacodynamics through improved TS inhibition and drug retention. When a series of quinazolines was tested for inhibitory activity towards TS (IC50 0.001-2 microM) and the growth of L1210 cells (IC50 0.005-10 microM), no direct correlation was observed. However, a very good correlation was apparent if a L1210 variant cell line (L1210: RD1694) was used. This line is deficient in its ability to form antifolate polyglutamates. A number of other intact cell methods have also been developed which estimate the contribution that intracellular polyglutamation makes to a compound's activity. These assays were validated using a series of quinazoline-based TS inhibitors with well-defined activity for TS, folypolyglutamate synthetase (FPGS) and the reduced-folate cell membrane carrier (RFC). Short-exposure growth-inhibition assays or the measurement of TS activity in situ after various incubation times, followed by different lengths of time in drug-free medium, can indicate both the speed and extent of appearance of retentive forms (usually polyglutamates). Continuous-exposure growth-inhibition assays, in the presence of leucovorin (LV), are also useful, since only the growth-inhibitory potency of polyglutamated analogues is significantly decreased by LV. Highly polyglutamated compounds, e.g. ZD1694, are virtually inactive in the presence of a high concentration of LV. It is proposed that these methods, when considered together, provide a greater degree of information concerning the rate and extent of polyglutamation of a particular compound than isolated FPGS assays alone.

Variation of the bridge linking the heterocyclic ring and p-aminobenzoyl-L-glutamate portions of our previously described classical 2,4-diaminofuro[2,3-d]pyrimidines 1 and 2 are reported as inhibitors of dihydrofolate reductase (DHFR) and thymidylate synthase (TS) and as antitumor agents. Specifically -CH2CH2- and -CH2NHCH2- bridged analogues, N-[4-[2-(2,4-diaminofuro[2,3-d]pyrimidin-5-yl) ethyl]benzoyl]-L-glutamic acid (3) and N-[4-[[N-[(2,4-diaminofuro[2,3-d]pyrimidin-5-yl) methyl]amino]methyl]benzoyl]-L-glutamic acid (4), respectively, were synthesized. Compound 3 was obtained via a Wittig reaction of the tributylphosphonium salt of 2,4-diamino-5-(chloromethyl)furo[2,3-d]pyrimidine (5) and methyl 4-formylbenzoate (6) followed by reduction and coupling with the diethyl ester of L-glutamic acid. Compound 4 was synthesized by the nucleophilic displacement of 5 with diethyl N-[4-(aminomethyl)benzoyl]-L-glutamate (15) and saponification. Both analogues were evaluated in vitro as inhibitors of DHFRs from (recombinant) human, human CCRF-CEM cells, and Lactobacillus casei. Compound 3 showed moderate activity (IC50 10(-6)-10(-7) M). Compound 4 was essentially inactive (IC50 10(-5) M, CCRF-CEM). The compounds were also evaluated against TS from (recombinant) human and L. casei and were of low activity (IC50 10(-5) M). The three-atom-bridged analogue 4 was somewhat more inhibitory to human TS than methotrexate (MTX). Compound 3 inhibited the growth of tumor cells in culture (IC50 10(-7) M) while 4 showed a low level of growth inhibitory activity. The inhibition of the growth of leukemia CCRF-CEM cells by both compounds parallels their inhibition of CCRF-CEM DHFR. Analogue 3 was a good substrate for human folylpolyglutamate synthetase (FPGS) derived from CCRF-CEM cells (Km 8.5 microM). Further evaluation of the growth inhibitory activity of 3 against the MTX-resistant subline of CCRF-CEM cells (R30dm) with decreased FPGS indicated that poly-gamma-glutamylation was important for its action. Protection studies with 3 in the FaDu squamous cell carcinoma cell line indicated that inhibition was completely reversed by leucovorin [(6R,S-5-formyltetrahydrofolate] or by a combination of thymidine and hypoxanthine, suggesting an antifolate effect directed at DHFR.

Previous attempts to design inhibitors of mammalian folylpolyglutamate synthetase (FPGS) have resulted in three classes of active compounds, all of which have charged moieties in the side chain, but structural alteration of the rest of the folate molecule has not seemed to be an avenue for drug discovery. However, groups in the side chain of folate analogs that bear charge distributions different from that of glutamic acid appear to prevent efficient transport into mammalian cells on the reduced folate carrier system. We now report that substituents at the 7-, 2'-, or 3'-position of 2-desamino-2-methyl-4-hydroxyquinazoline antifolates decrease or prevent the catalysis of diglutamate formation by FPGS but are compatible with efficient binding to the reduced folate carrier system. Thus, 5,8-dideazafolates with a 3'-alkyl group had a lower Vmax for FPGS than did the corresponding unsubstituted quinazolines, by a factor of 4-12, but these compounds inhibited the reaction of control FPGS substrates, indicating that the 3'-groups had much larger effects on catalytic activity than on binding to enzyme. A 7-methyl substituent affected the Vmax of a series of 5,8-dideazafolate compounds by a factor of 2-8, but this decrease in the catalytic rate was also accompanied by an increase in the Km of the substituted compounds by a factor of 10-100. The extent of the effect of a 7-methyl substituent on Vmax appeared to be dependent on the size of the substituent at N10. Different substituents at the 2'-position affected the kinetics of the FPGS reaction with one of three patterns, i.e., 1) a 2'-fluoro substituent both increased Vmax and decreased Km slightly, 2) either -OH or -NH2 decreased the Vmax without affecting the Km, and 3) 2'-Cl-, -CH3, -CF3, or -OCH3 substituents were found to both decrease Vmax and increase Km. Substitutions at the 7-, 2'-, or 3'-position had only minor effects on the ability of 2-desamino-2-methyl-4-oxoquinazolines to interfere with the transport of [3H]methotrexate into L1210 cells. Hence, these classes of compounds are likely to be efficiently transported by the reduced folate carrier system. We conclude that the region of the folate molecule bounded by the 7-, 6-, 9-, 10-, 3'-, and 2'-positions, the "bay region," is of major importance both for the binding of folates and folate analogs to FPGS and for the assumption of a conformation of the enzyme-substrate complex compatible with catalysis.(ABSTRACT TRUNCATED AT 400 WORDS)

Five new N alpha-(5,8-dideazapteroyl)-L-ornithines have been prepared using multistep synthetic sequences. These include N alpha-[5-(trifluoromethyl)-5,8-dideazapteroyl]-L-ornithine, 3, as well as N alpha-[5-(trifluoromethyl)-5,8-dideazaisopteroyl]-L-ornithine, 4, and its 5-fluoro and 5-chloro analogues. Both of the compounds containing a 5-(trifluoromethyl) group (3 and 4) were found to be excellent inhibitors of homogeneous hog liver folylpolyglutamate synthetase, having Ki values in the same range as N alpha-(5-chloro-5,8-dideazapteroyl)-L-ornithine, 2, (approximately 10 nM). However, the bridge-reversed isomer of 2 was 60-fold less inhibitory than 2.

A novel folic acid analogue, N alpha-(5-deaza-5,6,7,8-tetrahydropteroyl)-L-ornithine, 3, was prepared via a multistep synthetic sequence. The key steps involved the conversion of 5-deazapteroic acid to its N10-formyl derivative followed by catalytic hydrogenation of the pyridine ring and subsequent heating in dilute sodium hydroxide to afford the new 5-deaza-5,6,7,8-tetrahydropteroic acid. After trifluoroacetylation, this compound was coupled to N delta-(tert-butyl-oxycarbonyl)-L-ornithine using conventional peptide bond forming conditions. Deprotection first in base and then in acid gave the title compound. Compound 3 was an effective inhibitor of hog liver folylpolyglutamate synthetase (Kis, estimated = 64 nM), and was shown to retard the formation of polyglutamates of a structurally related folic acid analogue in HCT-8 cells in vitro.

5-Deazafolate and 5-deazatetrahydrofolate (DATHF) analogues with the glutamic acid side chain replaced by homocysteic acid (HCysA), 2-amino-4-phosphonobutanoic acid (APBA), and ornithine (Orn) were synthesized as part of a larger program directed toward inhibitors of folylpolyglutamate synthetase (FPGS) as probes of the FPGS active site and as potential therapeutic agents. The tetrahydro compounds were also of interest as non-polyglutamatable inhibitors of the purine biosynthetic enzyme glycinamide ribonucleotide formyltransferase (GARFT). Reductive coupling of N2-acetamido-6-formylpyrido[2,3-d]pyrimidin-4(3H)-one with 4-aminobenzoic acid, followed by N10-formylation, mixed anhydride condensation of the resultant N2-acetyl-N10-formyl-5- deazapteroic acid with L-homocysteic acid, and removal of the N2-acetyl and N10-formyl groups with NaOH, afforded N-(5-deazapteroyl)-L-homocysteic acid (5-dPteHCysA). Mixed anhydride condensation of N2-acetyl-N10-formyl- 5-deazapteroic acid with methyl D,L-2-amino-4-(diethoxyphosphinyl)butanoic acid, followed by consecutive treatment with Me3SiBr and NaOH, yielded D,L-2-[(5-deazapteroyl)amino]-4-phosphonobutanoic acid (5-dPteAPBA). Treatment with NaOH alone led to retention of one ethyl ester group on the phosphonate moiety. Catalytic hydrogenation of N2-acetyl-N10-formyl-5-deazapteroic acid followed by mixed anhydride condensation with methyl L-homocysteate and deprotection with NaOH afforded N-(5,6,7,8-tetrahydro-5-deazapteroyl)-L-homocysteic acid (5-dH4PteHCysA). Similar chemistry starting from methyl D,L-2-amino-4-(diethoxyphosphinyl)butanoic acid and methyl N delta-(benzyloxycarbonyl)-L-ornithinate yielded D,L-2-[(5-deaza-5,6,7,8-tetrahydropteroyl)amino]-4-phosphonobut ano ic acid (5-dH4Pte-APBA) and N alpha-(5-deaza-5,6,7,8-tetrahydropteroyl)-L-ornithine (5-dH4PteOrn), respectively. The 5-deazafolate analogues were inhibitors of mouse liver FPGS, and the DATHF analogues inhibited both mouse FPGS and mouse leukemic cell GARFT. Analogues with HCysA and monoethyl APBA side chains were less active as FPGS inhibitors than those containing an unesterified gamma-PO(OH)2 group, and their interaction with the enzyme was noncompetitive against variable folyl substrate. In contrast, Orn and APBA analogues obeyed competitive inhibition kinetics and were more potent, with Ki values as low as 30 nM. Comparison of the DATHF analogues as GARFT inhibitors indicated that the Orn side chain diminished activity relative to DATHF, but that the compounds with gamma-sulfonate or gamma-phosphonate substitution retained activity, with Ki values in the submicromolar range. The best GARFT inhibitor was the 5-dH4PteAPBA diastereomer mixture, with a Ki of 47 nM versus 65 nM for DATHF. None of the compounds showed activity against cultured WI-L2 or CEM human leukemic lymphoblasts at concentrations of up to 100 microM.(ABSTRACT TRUNCATED AT 400 WORDS)

A variety of folate analogues were synthesized to explore the specificity of the folate binding site of hog liver folylpolyglutamate synthetase and the requirements for catalysis. Modifications of the internal and terminal glutamate moieties of folate cause large drops in on rates and/or affinity for the protein. The only exceptions are glutamine, homocysteate, and ornithine analogues, indicating a less stringent specificity around the delta-carbon of glutamate. It is proposed that initial folate binding to the enzyme involves low-affinity interactions at a pterin and a glutamate site and that the first glutamate bound is the internal residue adjacent to the benzoyl group. Processive movement of the polyglutamate chain through the glutamate site and a possible conformational change in the protein when the terminal residue is bound would result in tight binding and would position the gamma-carboxyl of the terminal glutamate in the correct position for catalysis. Steric limitations imposed on the internal glutamate residues that loop out and additional steric constraints imposed by binding of different pterin moieties would be expected to effect slight conformational changes in the protein and/or the terminal glutamate and would explain the decrease in on rate and catalytic rate with increased polyglutamate chain length, and the differential effect of one-carbon substitution on the catalytic rate with polyglutamate derivatives. The 4-amino substitution of folate increases the on rate for monoglutamate derivatives but severely impairs catalysis with diglutamate derivatives. Pteroylornithine derivatives are the first potent and specific inhibitors of folylpolyglutamate synthetase to be identified and may act as analogues of reaction intermediates.(ABSTRACT TRUNCATED AT 250 WORDS)

Six new 5,8-dideaza analogues of folic acid and aminopterin containing a terminal L-ornithine residue were prepared by using multistep synthetic sequences. Each was evaluated as an inhibitor of hog liver folylpolyglutamate synthetase and human dihydrofolate reductase. Structural modifications at positions 2, 4, 5, and 10 were included to help define structure-activity relationships for compounds of this type. The compound N alpha-(4-amino-4-deoxy-5-chloro-5,8-dideazapteroyl)-L-ornithine (3f) was identified as the most potent inhibitor of mammalian folylpolyglutamate synthetase reported thus far (Ki congruent to 2 nM). Its 4-oxy counterpart, N alpha-(5-chloro-5,8-dideazapteroyl)-L-ornithine, was only 5-fold less inhibitory than 3f toward folylpolyglutamate synthetase but was found to be a much weaker inhibitor of dihydrofolate reductase than 3f.

Dihydrofolate (H2-folate) synthetase (EC 6.3.2.12) was isolated from Escherichia coli B. A radiochemical assay was developed to determine the activity of H2-folate synthetase in order to study the effects of folate metabolites and antimetabolites which would interfere with the microbiological assay method previously used. The effects of folate and pteroate derivatives on the activity of this enzyme were investigated to determine if inhibition of this enzyme could constitute a site of action for these compounds as chemotherapeutic agents or a site of metabolic regulation. H2-folate synthetase was inhibited by its product, H2-folate, and by the antimetabolite dihydrohomopteroate, with apparent Ki values of 23.4 and 9.2 muM, respectively.

A series of folate analogs containing ornithine instead of glutamate was synthesized and tested for inhibition of folylpolyglutamate synthetase (FPGS) and other folate-dependent enzymes of human leukemia cell lines. Reduced derivatives of 2-amino-4-oxo-10-methyl-pteroyl-ornithine had dramatically increased inhibitory potency against FPGS compared to the oxidized parent. The amino-pterin analog (2,4-diamino-pteroylornithine) was a potent inhibitor of both dihydrofolate reductase and FPGS. It was a much more potent linear competitive inhibitor of human FPGS than the corresponding methotrexate derivative previously described (Ki = 0.15-0.26 and 3 microM respectively). A quinazoline folate analog, 2-amino-4-oxo-5,8-dideazapteroyl-ornithine, was a relatively poor inhibitor of isolated dihydrofolate reductase and thymidylate synthase; however, it is the most potent human FPGS inhibitor identified to date (Ki = 100-150 nM). Because of the lack of appreciable interaction with other folate-dependent enzymes, structures incorporating the 2-amino-4-oxo-5,8-dideazapteroate nucleus may thus lead to selective inhibition of FPGS. Substitution of ornithine for glutamate caused a profound decrease in cytotoxic potency for these analogs; this was apparently the result of poor transport. Together with earlier studies, these data indicate that the potency of FPGS inhibition by an analog containing ornithine closely parallels the relative substrate activity of its glutamate-containing counterpart. The substitution of ornithine apparently does not perturb the pterin specificity of FPGS. The close parallel between substrate and inhibitor specificity may thus allow the use of currently available structure-activity studies on FPGS to design more potent and more selective inhibitors of FPGS.

Derivatives of folic acid occur in nature predominantly as poly (gamma-glutamyl) derivatives containing 2-8 glutamate residues. The data regarding the function of these derivatives, and their biosynthesis by eucaryotic and procaryotic folylpolyglutamate synthetases, is reviewed. The most universal functions of folylpolyglutamates appear to be (a) as the actual cofactors in vivo for folate dependent enzymes, (b) as inhibitors of folate dependent enzymes for which they are not substrates, and (c) to increase retention of folates after they are transported into cells as monoglutamates. Folylpolyglutamates also have numerous specialized functions in specific organisms, e.g. as structural components of some coliphage, and as allosteric regulators in Neurospora crassa. A single enzyme appears responsible for synthesis of all polyglutamate derivatives, regardless of length. With the recent introduction of sensitive assays this folylpolyglutamate synthetase has begun to be characterized. Although procaryotic and eucaryotic synthetases have many dissimilar properties, both types catalyze the ATP-dependent addition of L-glutamate to the gamma-carboxyl of the glutamate present in the folate. Both types also require a monovalent cation and relatively high pH. The most significant differences between the two types are in their folate substrate specificity and the product lengths derived from various folates.