Molecular modeling and site-directed mutagenesis define the catalytic motif in human gamma -glutamyl hydrolase.

Human gamma-glutamyl hydrolase (hGH) is a central enzyme in folyl and antifolylpoly-gamma-glutamate metabolism, which functions by catalyzing the cleavage of the gamma-glutamyl chain of substrates. We previously reported that Cys-110 is essential for activity. Using the sequence of hGH as a query, alignment searches of protein data bases were made using the SSearch and TPROBE programs. Significant similarity was found between hGH and the glutamine amidotransferase type I domain of Escherichia coli carbamoyl phosphate synthetase. The resulting hypothesis is that the catalytic fold of hGH is similar to the folding of this domain in carbamoyl phosphate synthetase. This model predicts that Cys-110 of hGH is the active site nucleophile and forms a catalytic triad with residues His-220 and Glu-222. The hGH mutants C110A, H220A, and E222A were prepared. Consistent with the model, mutants C110A and H220A were inactive. However, the V(max) of the E222A hGH mutant was reduced only 6-fold relative to the wild-type enzyme. The model also predicted that His-171 in hGH may be involved in substrate binding. The H171N hGH mutant was found to have a 250-fold reduced V(max). These studies to determine the catalytic mechanism begin to define the three dimensional interactions of hGH with poly-gamma-glutamate substrates.

Glutamyl hydrolase. pharmacological role and enzymatic characterization.

gamma-Glutamyl hydrolase (GH, EC 3.4.19.9) is a lysosomal and secreted glycoprotein that hydrolyzes the gamma-glutamyl tail of antifolate and folate polyglutamates. Tumor cells that have high levels of GH are inherently resistant to classical antifolates, and further resistance can be acquired by elevations in GH following exposure to this class of antitumor agents. The highest level of expression in normal tissues occurs in the liver and kidney in humans. When panels of tumors are compared with normal tissues, GH expression is elevated in cancerous hepatic and breast tissue. A second poly-gamma-glutamate hydrolyzing enzyme, glutamate carboxypeptidase II, is a transmembrane protein whose active site is on the outside of the cell, occurring in the prostate gland, small intestine, brain, kidney, and tumor neovasculature. It is a high-affinity (nanomolar), low-turnover, zinc co-catalytic enzyme. In contrast, GH is a low-affinity (micromolar), high-turnover enzyme that has a cysteine at the active site. Data are presented suggesting that Cys110 is the nucleophile that attacks the gamma-amide linkage and causes hydrolysis. GH is being evaluated as an intracellular target for inhibition in order to enhance the therapeutic activity of antifolates and fluorouracil.

Glutamyl hydrolase: properties and pharmacologic impact.

Glutamyl hydrolase cleaves the poly-gamma-glutamate chain folate and antifolate poly-gamma-glutamates. Its cellular location is lysosomal with large amounts of the enzyme constitutively secreted. The highest levels of glutamyl hydrolase mRNA in humans is found in the liver and kidney. Baculovirus-expressed human enzyme has been used to evaluate the method of hydrolysis of methotrexate-gamma-glu4 and MTA-gamma-glu4. In both cases, the substrates are hydrolyzed by removal of the outer two gamma-glutamate linkages, yielding glu and gamma-glu2 as the glutamate products. Cell lines resistant to 5,10-dideazatetrahydrofolate (lometrexol) have sevenfold higher activities of glutamyl hydrolase. These cultures have a 60% to 90% reduced amount of antifolate polygamma-glutamates and 30% reduced folyl poly-gamma-glutamates. These results suggest the possibility of using glutamyl hydrolase to favorably modulate the activity of antifolate therapy.

Structural organization of the human gamma-glutamyl hydrolase gene.

Gamma-glutamyl hydrolase (GH) plays an important role in the metabolism of folic acid and the pharmacology of antifolates such as methotrexate. We have previously cloned and characterized the human GH cDNA. In this report, the complete organization and structure of the human GH gene was determined. The human GH gene spans 24 kb in the human genome, with nine exons sized from 51 to 371 bp. All of exon-intron splice junctions follow the GT-AG rule. The sequence upstream of exon 1 consists of a promoter-like, GC-rich region and a number of putative cis active elements including Sp1, AP1, and MZF1 sites. A TATA sequence in the 5' region of human GH gene was not observed, similar to housekeeping genes known to be tissue-specific and differentially expressed. S1 nuclease protection analysis with human liver, prostate, brain, and mammary gland revealed a major transcription start point at nucleotide -125 relative to the ATG start codon and several minor transcription start points. Analysis of GH cDNA isolated from human liver indicated a nucleotide change, T-->C, in the leader sequence of GH, which suggested a polymorphism. Studies of cDNA from different human tissue sources provided evidence that there is a single spliced cDNA species in human.

Site-directed mutagenesis establishes cysteine-110 as essential for enzyme activity in human gamma-glutamyl hydrolase.

Gamma-glutamyl hydrolase (GH), which hydrolyses the gamma-glutamyl conjugates of folic acid, is a key enzyme in the maintenance of cellular folylpolyglutamate concentrations. The catalytic mechanism of GH is not known. Consistent with earlier reports that GH is sulphydryl-sensitive, we found that recombinant human GH is inhibited by iodoacetic acid, suggesting that at least one cysteine is important for activity [Rhee, Lindau-Shepard, Chave, Galivan and Ryan (1998) Mol. Pharmacol. 53, 1040-1046]. Using site-directed mutagenesis, the cDNA for human GH was altered to encode four different proteins each with one of four cysteine residues changed to alanine. Three of the mutant proteins had activities similar to wild-type GH and were inhibited by iodoacetic acid, whereas the C110A mutant had no activity. Cys-110 is conserved among the human, rat and mouse GH amino acid sequences. The wild-type protein and all four mutants had similar intrinsic fluorescence spectra, indicating no major structural changes had been introduced. These results indicate that Cys-110 is essential for enzyme activity and suggest that GH is a cysteine peptidase. These studies represent the first identification of the essential Cys residue in this enzyme and provide the beginning of a framework to determine the catalytic mechanism, important in defining GH as a therapeutic target.

Glutamyl hydrolase and the multitargeted antifolate LY231514.

PURPOSE: To examine the activity of glutamyl hydrolase (GH) on the poly-gamma-glutamates of multitargeted antifolate (MTA) (LY231514) and the effect of enhanced GH on the pharmacological activity of MTA. METHODS: Expressed and purified GH were used to study the enzymatic cleavage of MTA poly-gamma-glutamates and wild-type and GH-enhanced H35 hepatoma cell lines to evaluate growth inhibition. RESULTS: MTA tri- and penta-gamma-glutamates were good substrates for human GH, having higher rates than MTX tri- and penta-gamma-glutamates. Preferential hydrolysis with human enzyme occurred at the two gamma-glutamyl bonds at the carboxyl end of the molecule, whereas the rat enzyme preferred the innermost gamma-linkage. Incubation of rat H35 hepatoma cell lines with MTA resulted in the intracellular accumulation of primarily tetra-, penta-, and hexa-gamma-glutamate. The formation of these were markedly reduced in H35D cells, which is a line resistant to antifolates chiefly through enhanced cellular levels of GH activity. CONCLUSIONS: MTA poly-gamma-glutamates are effective substrates for GH and their pharmacological effectiveness bears an inverse relationship to cellular GH activity. This observation, along with enhanced resistance to MTA of thymidylate synthase-amplified cells, substantiates the importance of the poly-gamma-glutamates of MTA inhibiting TS as the primary target. Further evidence for the inverse relationship of GH to classical antifolate pharmacological activity is established.

Cloning of mouse gamma-glutamyl hydrolase in the form of two cDNA variants with different 5' ends and encoding alternate leader peptide sequences.

Mouse-liver gamma-glutamyl hydrolase (GH) is a lysosomal endopeptidase with an acid pH optimum that is activated by sulfhydryl compounds and preferentially hydrolyzes the most proximal gamma-glutamyl linkage of longer chain polyglutamates of folates and their analogues. We describe the cloning of this mouse lysosomal cDNA enzyme from liver GH mRNA in the form of two cDNA variants (1.295 and 1.268 kb in length) differing 14-fold (Variant I versus Variant II) in relative frequency that exhibited 5'-end heterogeneity and encoded alternate leader peptides. The 5' UTR in these variants also differs in length by 27 nucleotides. Otherwise, the ORF and 3' UTR in each case are the same. These cDNAs encode a protein in which the deduced amino acid sequence shares 78.9 and 69. 1% identity to rat and human GH sequences, respectively. Amino acid sequence comparisons among the three species identified three conserved Asn sites and two conserved Cys residues that may be sites of glycosylation and sulfhydryl compound activation, respectively. Variant I GH mRNA was more abundant than Variant II GH mRNA in all mouse tissues examined. Variant I GH mRNA levels were extremely high in salivary gland, moderately high in kidney, liver, lung, stomach and uterus, low in small intestine, brain and fetal liver and relatively rare in thymus, spleen and skeletal muscle. Abundance of GH mRNA among tumors varied from low to high, with no discernible correlation with their tissue of origin.

Characterization of human cellular gamma-glutamyl hydrolase.

A previously identified cDNA encoding a human gamma-glutamyl hydrolase was expressed in a baculovirus system. The expressed protein had molecular mass of 37 kDa. Treatment of the protein with PNGase F produced a protein of molecular mass of 30 kDa, indicating that the protein contained asparagine-linked glycosylation. Sequence analysis of the expressed protein indicated that a 24-amino-acid signal peptide had been removed. A polyclonal antibody to the expressed enzyme was used in Western blot analysis of partially purified lysates of HL-60 promyeloid leukemia cells and MCF-7 breast cancer cells. The HL-60 and MCF-7 enzymes appeared as two closely spaced bands with a molecular mass of 37 kDa. Treatment of the HL-60 enzyme with PNGase F produced a protein with a molecular mass of 30 kDa. The activities of the expressed enzyme and the enzyme from HL-60 cells were similar on methotrexate polyglutamates. Methotrexate-gamma-Glu is a poor substrate for the human enzyme relative to methotrexate gamma-Glu2-5. During hydrolysis of methotrexate-gamma-Glu4, all possible pterin-containing cleavage products (methotrexate and methotrexate-gamma-Glu1-3) appear. The results demonstrated that the human enzyme cleaves both the ultimate and penultimate gamma-linkages of methotrexate polyglutamates. Glutamate was released as either glutamic acid or gamma-Glu2. Longer chain species of gamma-Glun>2 were not observed. Inhibition by iodoacetic acid suggested that both the expressed enzyme and the HL-60 enzyme may contain a catalytically essential cysteine. These results indicate that the identified cDNA encodes the intracellular gamma-glutamyl hydrolase found in a variety of human tumor cells and that the baculovirus-expressed enzyme is a suitable model for further structural and enzymatic studies.

Intracellular location of thymidylate synthase and its state of phosphorylation.

Thymidylate synthase (TS), an enzyme that is essential for DNA synthesis, was found to be associated mainly with the nucleolar region of H35 rat hepatoma cells, as determined both by immunogold electron microscopy and by autoradiography. In the latter case, the location of TS was established through the use of [6-3H]5-fluorodeoxyuridine, which forms a tight ternary complex of TS with 5-fluorodeoxyuridylate (FdUMP) and 5, 10-methylenetetrahydrofolylpolyglutamate within the cell. However, with H35 cells containing 50-100-fold greater amounts of TS than unmodified H35 cells, the enzyme, although still in the nucleus, was located primarily in the cytoplasm as shown by autoradiography and immunohistochemistry. In addition, TS was also present in mitochondrial extracts of both cell lines, as determined by enzyme activity measurements and by ternary complex formation with [32P]FdUMP and 5,10-methylenetetrahydrofolate. Another unique observation is that the enzyme appears to be a phosphoprotein, similar to that found for other proteins associated with cell division and signal transduction. The significance of these findings relative to the role of TS in cell division remains to be determined, but suggest that this enzyme's contribution to the cell cycle may be more complex than believed previously.

Biological properties of fluoroglutamate-containing analogs of folates and methotrexate with altered capacities to form poly (gamma-glutamate) metabolites.

Fluoroglutamate-containing analogs of folates and methotrexate (MTX) with altered capacities for poly (gamma-glutamate) metabolism were synthesized to probe the biological roles of polyglutamates. Compared to folic acid, DL-e,t-gamma-fluorofolic acid, a compound that is a poor substrate for polyglutamylation, was approximately 25-fold less potent in promoting growth of folate-depleted H35 rat hepatoma cells. DL-beta,beta-Difluorofolic acid, a compound that forms diglutamates more readily than does folic acid, was at least equivalent to folic acid in potency. Leucovorin (LV), a reduced folate, was 30-fold more potent than folic acid in promoting growth, whereas the analogous form of DL-e,t-gamma-fluorofolate, DL-e,t-gamma-fluoroleucovorin (DL-e,t-gamma-FLV) was only 4-fold more potent than folic acid. Both LV and DL-e,t-gamma-FLV protected or "rescued" cells from the growth inhibitory effects of MTX; however a 37- to 46-fold higher concentration of the fluoro analog was required. Folic acid, DL-e,t-gamma-fluorofolic acid, LV, and DL-e,t-gamma-FLV each potentiated the growth inhibitory effect of 5-fluoro-2'-deoxyuridine on CCRF-CEM human leukemia cells; higher concentrations of fluorinated analogs again were required. Stereochemically pure L-t-gamma-fluoromethotrexate (L-t-gamma-FMTX), a poor substrate for polyglutamylation, was evaluated as a cell growth inhibitor. In continuous exposure, L-t-gamma-FMTX), was 7-fold less potent than MTX as an inhibitor of CCRF-CEM growth. Results with these fluorinated folate and MTX analogs offer insight into the importance of polyglutamate metabolism to these biological and pharmacological effects.

Human gamma-glutamyl hydrolase: cloning and characterization of the enzyme expressed in vitro.

A cDNA encoding human gamma-glutamyl hydrolase has been identified by searching an expressed sequence tag data base and using rat gamma-glutamyl hydrolase cDNA as the query sequence. The cDNA encodes a 318-amino acid protein of Mr 35,960. The deduced amino acid sequence of human gamma-glutamyl hydrolase shows 67% identity to that of rat gamma-glutamyl hydrolase. In both rat and human the 24 amino acids preceding the N terminus constitute a structural motif that is analogous to a leader or signal sequence. There are four consensus asparagine glycosylation sites in the human sequence, with three of them conserved in the rat enzyme. Expression of both the human and rat cDNA in Escherichia coli produced antigenically related proteins with enzyme activities characteristic of the native human and rat enzymes, respectively, when methotrexate di- or pentaglutamate were used as substrates. With the latter substrate the rat enzyme cleaved the innermost gamma-glutamyl linkage resulting in the sole production of methotrexate as the pteroyl containing product. The human enzyme differed in that it produced methotrexate tetraglutamate initially, followed by the triglutamate, and then the diglutamate and methotrexate. Hence the rat enzyme is an endopeptidase with methotrexate pentaglutamate as substrate, whereas the human enzyme exhibits exopeptidase activity. Another difference is that the expressed rat enzyme is equally active on methotrexate di- and pentaglutamate whereas the human enzyme has severalfold greater activity on methotrexate pentaglutamate compared with the diglutamate. These properties are consistent with the enzymes derived from human and rat sources.

Identification, cloning, and sequencing of a cDNA coding for rat gamma-glutamyl hydrolase.

Purified gamma-glutamyl hydrolase secreted from rat H35 hepatoma cells has been characterized as a diffuse band of 55 kDa on SDS-polyacrylamide gel electrophoresis that is converted to bands of 35 and 33 kDa after enzymatic removal of N-linked carbohydrate. Polyclonal antibodies against 55-kDa gamma-glutamyl hydrolase captured the enzyme activity and recognized the glycosylated and both deglycosylated forms of gamma-glutamyl hydrolase. A complete cDNA sequence of gamma-glutamyl hydrolase was obtained using degenerate oligonucleotides derived from peptide sequences, screening of a rat hepatoma cDNA library, and reverse transcription polymerase chain reaction. Based upon the deduced amino acid sequence the peptide component of gamma-glutamyl hydrolase had a molecular weight of 33,400. The results of amino acid analysis of the purified protein agreed with the deduced amino acid sequence in which there are seven potential asparagine-containing glycosylation sites.

Polyglutamylation of the dihydrofolate reductase inhibitor gamma-methylene-10-deazaaminopterin is not essential for antitumor activity.

As part of a continuing program aimed at developing nonpolyglutamylatable inhibitors of dihydrofolate reductase that are less toxic and more specific in their action, we herein report the therapeutic efficacy and toxicity of gamma-methylene-10-deazaaminopterin (MDAM) in athymic nude mice bearing advanced human HCT-8 ileocecal xenografts and its antitumor activity in C57BL/6 x DBA/2 F1 (hereafter called B6D2F1) mice bearing P388 murine leukemia. For the xenograft study, MDAM was administered at the maximum tolerated dose by the following dose schedules: (a) 5-day continuous i.v. infusion at 1.0 mg/kg/day (schedule I); and (b) i.v. push, daily for 5 days at 50 mg/kg/day (schedule II). The maximum tolerated dose values for methotrexate (MTX) under these conditions were 0.2 and 1.0 mg/kg/day for schedule I and schedule II, respectively. MTX did not exhibit any significant antitumor activity in this model system by both schedules; however, MDAM induced complete responses of 13 and 25% and partial responses of 25 and 50% by schedules I and II, respectively. MDAM also exhibited antitumor activity significantly superior to that of MTX in the P388 tumor model. One of the enantiomers of MDAM, which possesses the natural configuration at the gamma-methyleneglutamate moiety (l-MDAM), has been shown to be a better inhibitor of human recombinant dihydrofolate reductase and H35 hepatoma cell growth than D,L-MDAM. L-MDAM inhibited the uptake of radiolabeled folinic acid to H35 hepatoma cells eight times more efficiently than MTX. The results indicate that the superior activity of MDAM relative to MTX may be partially due to a combination of enhanced transport to tumor cells and slower deactivation by aldehyde oxidase.

Exploitation of folate and antifolate polyglutamylation to achieve selective anticancer chemotherapy.

Synthesis of poly(gamma-glutamate) metabolites of natural folates and antifolates is a critical process. Folypolyglutamates are essential for cell proliferation. Polyglutamates of glutamate (Glu)-containing antifolates are often critical for their cytotoxic action and are relevant to antifolate resistance. However, the role of polyglutamate synthesis in selectivity is less clear. We have undertaken a research program to further define the significance of polyglutamate metabolism and to devise ways to exploit this metabolism to achieve greater therapeutic selectivity in cancer chemotherapy. This article briefly reviews several approaches tested thus far. Inhibition of folypolyglutamate synthesis should lead to cell death. Current ornithine (Orn)-containing folate-based inhibitors of the enzyme responsible for their synthesis, folypolyglutamate synthetase (FPGS), are poorly transported, apparently because of interference by the protonated delta-amine. Replacement of Orn with 4, 4-difluoroOrn, the delta-amine of which has a much lower pKa and is thus less protonated at physiological pH, was explored. Since it is unclear how polyglutamylation contributes to selectivity, we explored generic means either to eliminate or to enhance polyglutamylation. The data indicate that substitution for Glu in an antifolate by some Glu analogs in which the gamma-COOH is either altered or replaced (e.g., gamma-tetrazole-Glu) leads to loss of both FPGS substrate activity and binding; antifolate target specificity is unchanged, while uptake is actually enhanced. Substitution of 3,3-difluoroGlu for Glu leads to enhanced polyglutamylation (although probably only to the diglutamate), retention of target specificity, and at least equal uptake. Comparative studies of the same antifolate containing different replacements for Glu, such as gamma-tetrazole-Glu (no polyglutamylation) or 3,3-difluoroGlu (enhanced polyglutamylation), will be useful in exploring the role and significance of polyglutamylation.

Insulin-dependent suppression in glutamyl hydrolase activity and elevated cellular methotrexate polyglutamates.

Folates and antifolates are converted to polyglutamates, which are better retained in cells and may also bind more tightly to cellular proteins than the parent compounds. The regulation of the process of polyglutamate formation and breakdown is not fully clarified yet and is being studied by a number of approaches. An early observation concerning the potential regulation of polyglutamate formation was that insulin caused a marked increase in the rate and accumulation of polyglutamates of methotrexate (MTX) in rat hepatoma cells. The present study demonstrated that insulin caused a decrease in the activity of gamma-glutamyl hydrolase (GH), the enzyme that degrades polyglutamates, that was inversely commensurate with the increase in the synthesis of MTX polyglutamates. The effects of insulin on GH activity with regard to concentration, time of onset, and the effect of N6, O2' dibutyryl cAMP and theophylline were consistent with the reduction in GH being responsible for the increase in cellular MTX polyglutamate accumulation. Insulin addition also led to an increase in folate polyglutamates. The insulin effects were also seen with H35D cells, a subline with enhanced glutamyl hydrolase activity as a result of having been made resistant to 5, 10-dideazatetrahydrofolic acid. When H35 cells with insulin were compared with H35D cells lacking insulin, there was an 8-fold increase in GH and a 44-fold decrease in the number of gamma-glutamate residues added to MTX.

Effects of gamma-glutamyl hydrolase on folyl and antifolylpolyglutamates in cultured H35 hepatoma cells.

A subline of H35 hepatoma cells (H35D cells) that have been made resistant to 5,10-dideazatetrahydrofolate exhibits an increase in gamma-glutamyl hydrolase (GH) activity. GH is a lysosomal enzyme in H35 and H35D cells on the basis of comparison of the distribution of enzyme activity with other known lysosomal enzymes. The hydrolysis rate of methotrexate polyglutamate with isolated, intact lysosomes is 4-5-fold greater in H35D cells than in H35 cells. GH activity in isolated lysosomes is in part dependent on the presence of a reducing agent such as mercaptoethanol. Permeabilization of lysosomal preparations from both cell types by Triton X-100 causes a 10-fold enhancement in GH activity. The result of the enhanced activity of GH in H35D cells is a marked reduction in antifolylpolyglutamate concentration, with the parent antifolate being the predominant intracellular species found under all conditions tested. Unlike antifolates, the total intracellular folate concentration is nearly identical in both cells under standard culture conditions up to 10 microns folic acid. However, the chain length of folylpolyglutamates consists of predominantly triglutamates and tetraglutamates in H35D cells with increased GH, whereas it consists of pentaglutamates and hexaglutamates in the parental cells. At 50 and 100 microns folic acid, the folate accumulation in H35D cells is less than half that of H35 cells, and the predominant polyglutamate species in the H35D cells are the diglutamates through the tetraglutamates. The results demonstrate that the two H35 cell lines having equal folylpolyglutamate synthetase but that one with enhanced lysosomal GH activity exhibits a marked reduction in the amount and gamma-glutamyl chain length of folylpolyglutamates and antifolylpolyglutamates.

DL-beta,beta-difluoroglutamic acid mediates position-dependent enhancement or termination of pteroylpoly(gamma-glutamate) synthesis catalyzed by folylpolyglutamate synthetase.

Poly(gamma-glutamylation) of glutamate (L-Glu)-containing antifolates and natural folates is important in pharmacological mechanisms and in physiological processes. Based on previous work from our laboratories, we hypothesized that replacement of the L-Glu moiety in parent molecules with DL-beta,beta-difluoroglutamic acid (DL-beta,beta-F2Glu) might be a generic means of increasing polyglutamylation by both increasing the synthesis rate and decreasing the degradation rate (J. J. McGuire et al., J. Biol. Chem. 265, 14073-14079 (1990)); thus biological potency might be increased without other biochemical properties being altered. DL-beta,beta-F2Glu, synthesized by an improved route (B. P. Hart and J. K. Coward, Tetrahedron Lett. 34, 4917-4920 (1993)), has been incorporated into a methotrexate (MTX) homolog, beta,beta-difluoromethotrexate (beta,beta-F2MTX), and a folic acid (PteGlu) homolog, beta,beta-difluorofolic acid (beta,beta-PteF2Glu). Biochemical properties of beta,beta-F2MTX (e.g., inhibition of isolated dihydrofolate reductase, transport in whole cells) are similar to those of MTX except that, in accord with our hypothesis, apparent substrate efficiency for rat and human folylpolyglutamate synthetase (FPGS) is 4- to 7.5-fold higher, respectively, for beta,beta-F2MTX than for MTX. Analysis of the products synthesized by purified FPGS, however, suggests that while addition of the first gamma-Glu to beta,beta-F2MTX is highly efficient, subsequent additions occur at a negligible rate; this premise was confirmed by directly comparing the in vitro FPGS substrate activity of MTX-gamma-Glu and beta,beta-F2MTX-gamma-Glu. Furthermore, the dramatically diminished in vitro growth inhibitory potency of beta,beta-F2MTX as compared to MTX when exposure time to drug is decreased (despite otherwise similar biochemical properties) suggests that polyglutamylation is also impaired in intact cells. Similar results with FPGS have been obtained with oxidized and reduced forms of beta,beta-PteF2Glu. These data suggest that the effect of beta,beta-F2Glu on polyglutamylation by FPGS is dependent on its position relative to the point of L-Glu ligation. When beta,beta-F2Glu is the acceptor amino acid (i.e., point of attachment), ligation of Glu is enhanced; however, if beta,beta-F2Glu is one residue distal to the acceptor amino acid, further elongation is blocked.

Multifactorial resistance to 5,10-dideazatetrahydrofolic acid in cell lines derived from human lymphoblastic leukemia CCRF-CEM.

5,10-dideaza-5,6,7,8-terrahydrofolic acid (DDATHF) is a potent antiproliferative agent in cell culture systems and in vivo in a number of murine and human xenograft tumors. In contrast to classical antifolates, which are dihydrofolate reductase inhibitors, DDATHF primarily inhibits GAR transformylase, the first folate-dependent enzyme along the pathway of de novo purine biosynthesis. The (6R) diastereomer of DDATHF (Lometrexol), currently undergoing clinical investigation, was used to develop CCRF-CEM human leukemia sublines resistant to increasing concentrations of the drug. Three cell lines were selected for ability to grow in medium containing 0.1 microM, 1.0 microM, and 10 microM of (6R)DDATHF, respectively. Impaired polyglutamylation was identified as a common mechanism of resistance in all three cell lines. A progressive decrease in the level of polyglutamylation was associated with diminished folylpolyglutamate synthetase activity and paralleled increasing levels of resistance to the drug. However, the expression of folylpolyglutamate synthetase RNA was not altered in the resistant cell lines compared to the parent cells. The most resistant cell subline also displayed an increased activity of gamma-glutamyl hydrolase. The sublines were scrutinized for other possible mechanisms of resistance. No alterations in drug transport or in purine economy were found. Modest increases were found in the activity of methylene tetrahydrofolate dehydrogenase but no alterations of other folate-dependent enzymes were observed. Increases in accumulation and conversion of folic acid to reduced forms, particularly 10-formyltetrahydrofolate, was also seen. The resistant cell lines were sensitive to dihydrofolate reductase inhibitors, methotrexate and trimetrexate, for a 72-h exposure period but showed cross-resistance to methotrexate for 4 and 24 h exposures. Cross-resistance was also shown toward other deazafolate analogues for both short- and long-term exposures.

Biochemical studies on PT523, a potent nonpolyglutamatable antifolate, in cultured cells.

Studies on the mode of action of PT523 [N alpha-(4-amino-4-deoxypteroyl)-N delta-hemiphthaloyl-L-ornithine], a potent nonpolyglutamatable antifolate, were carried out in sensitive and resistant H35 rat hepatoma cell lines in culture, to compare it with other antifolates, including three dihydrofolate reductase (DHFR) inhibitors, i.e., methotrexate (MTX), gamma-fluoro-MTX, and trimetrexate (TMQ), two thymidylate synthase inhibitors, i.e., N10-propargyl-5,8- dideazafolate (PDDF) and 2-desamino-2-methyl-N10-propargyl-5,8-dideazafolate (dmPDDF), and the glycinamide ribonucleotide formyltransferase inhibitor 5,10-dideaza-5,6,7,8-tetrahydrofolate. PT523 was the most active compound in this group against the parental H35 cells, with an IC50 ranging from 2.5 nM for 72 hr of treatment to 0.21 microM for 2 hr of treatment. Sublines resistant to MTX by virtue of a transport defect or a combination of defective transport and increased DHFR activity were resistant to PT523 and MTX but not to PDDF, whereas sublines resistant to fluoropyrimidines by virtue of increased thymidylate synthase activity were resistant to PDDF but not to PT523, TMQ, or MTX. Inhibition of H35 cell growth by PT523 was associated with a concentration- and time-related decrease in de novo dTMP and purine biosynthesis. Growth inhibition by PT523, MTX, and TMQ was prevented by leucovorin or a combination of thymidine (dThd) and hypoxanthine but not by dThd or hypoxanthine alone; in contrast, growth inhibition by dmPDDF was prevented by dThd alone. Intracellular reduced folate polyglutamate pools were markedly altered by PT523 treatment, with the most pronounced effect being an increase in 7,8-dihydrofolate mono- and polyglutamates and a decrease in 5,10-methylene-5,6,7,8-tetrahydrofolate mono- and polyglutamates, 5,6,7,8-tetrahydrofolate mono- and polyglutamates, and 10-formyl-5,6,7,8-tetrahydrofolate mono- and polyglutamates. This pattern was qualitatively similar to that observed with MTX and TMQ but different from that observed with dmPDDF or 5,10-dideaza-5,6,7,8-tetrahydrofolate, which resulted in little or no change in the folate species. Uptake of [3H]MTX and [3H]folinic acid, but not [3H]folic acid, by H35 cells was inhibited in a dose-related manner by PT523, suggesting that penetration of the cell probably involves, at least in part, active transport by the MTX/reduced folate carrier. To determine whether the potent cellular effects of PT523 might be due to chemical or enzymic clevage to N'-(4-amino-4-deoxypteroyl)-L-ornithine, a potent inhibitor of folylpolyglutamate synthetase, the formation of [3H]MTX polyglutamates in CCRF-CEM lymphoblasts pulsed with [3H]MTX after preincubation with PT523 was examined.(ABSTRACT TRUNCATED AT 400 WORDS)

Biological activity of a novel rationally designed lipophilic thymidylate synthase inhibitor.

AG-331 (N6[4-(N-morpholinosulfonyl)benzyl]-N6-methyl-2,6-diamino- benz[cd]indole glucuronate) is a novel lipophilic thymidylate synthase (TS) inhibitor. The properties of this compound were investigated in H35 rat hepatoma cells and in three variant cell lines resistant to antifolates by differing mechanisms. There was no evidence for any intracellular effect of AG-331 on dihydrofolate reductase (DHFR); however, the low degree of cross-resistance found for the H35FF line, which has elevated TS levels, suggested that TS may not be the sole locus of action of AG-331 in hepatoma cells. TS-directed effects of AG-331 were suggested by the pattern of its inhibition of deoxyuridine incorporation into DNA and the lesser effects of purine incorporation. In addition, H35 cells treated with 10 microM AG-331 were shown to accumulate in the S phase of the cell cycle, and this effect could be reversed by coadministration of thymidine. However, when treatments were conducted at a 5-fold higher concentration of AG-331, no S-phase block was apparent, suggesting the loss of a TS-directed effect at high inhibitor concentrations. Thymidine and folinic acid also failed to protect cells against AG-331 cytotoxicity, suggesting an alternate mode of action. Similar results were also obtained in protection experiments with a human hepatoma cell line, HEPG2, although previous results obtained in colon- and breast-cancer cell lines have suggested TS specific effects for AG-331. The possibility that biotransformation of AG-331 to other toxic species may occur in liver-derived cell lines has yet to be investigated.