A novel class of monoglutamated antifolates exhibits tight-binding inhibition of human glycinamide ribonucleotide formyltransferase and potent activity against solid tumors.

Tight-binding inhibition of recombinant human monofunctional glycinamide ribonucleotide formyltransferase by Lometrexol (6R-5,10-dideazatetrahydrofolate) requires polyglutamation. LY254155 and LY222306 differ from 5,10-dideazatetrahydrofolate in the replacement of the 1',4'- phenylene moiety by a 2',5'-thiophene and a 2',5'-furan, respectively. Compared to Lometrexol, the thiophene and furan analogues had 25- and 75-fold greater inhibitory potencies against human monofunctional glycinamides ribonucleotide formyltransferase (Ki = 2.1 and 0.77 nM, respectively). The binding affinities of the thiophene and furan analogues for membrane folate-binding protein from human KB cells were 6- and 350-fold weaker than Lometrexol, respectively. Both the thiophene analogue and 5,10-dideazatetrahydrofolate inhibited the in vivo growth of murine 6C3HED lymphosarcoma, murine C3H mammary carcinoma, and human xenograft HXGC3, HC1, and VRC5 colon carcinomas by 95-100%. The thiophene analogue was efficacious against human xenograft PANC-1, a pancreatic carcinoma which was completely resistant to 5,10- dideazatetrahydrofolate. These novel antifolates represent the first monoglutamated tight-binding inhibitors of glycinamide ribonucleotide formyltransferase. By eliminating the need for polyglutamation, this class of antifolates may have clinical activity in the treatment of solid tumors expressing low levels of folylpolyglutamate synthetase or tumors resistant to antifolate therapy due to increased gamma-glutamyl hydrolase activity.

LY231514, a pyrrolo[2,3-d]pyrimidine-based antifolate that inhibits multiple folate-requiring enzymes.

N-[4-[2-(2-amino-3,4-dihydro-4-oxo-7H-pyrrolo[2,3-d]pyrimidin-5-yl)ethyl ]-benzoyl]-L-glutamic acid (LY231514) is a novel pyrrolo[2,3-d]pyrimidine-based antifolate currently undergoing extensive Phase II clinical trials. Previous studies have established that LY231514 and its synthetic gamma-polyglutamates (glu3 and glu5) exert potent inhibition against thymidylate synthase (TS). We now report that LY231514 and its polyglutamates also markedly inhibit other key folate-requiring enzymes, including dihydrofolate reductase (DHFR) and glycinamide ribonucleotide formyltransferase (GARFT). For example, the Ki values of the pentaglutamate of LY231514 are 1.3, 7.2, and 65 nM for inhibition against TS, DHFR, and GARFT, respectively. In contrast, although a similar high level of inhibitory potency was observed for the parent monoglutamate against DHFR (7.0 nM), the inhibition constants (Ki) for the parent monoglutamate are significantly weaker for TS (109 nM) and GARFT (9,300 nM). The effects of LY231514 and its polyglutamates on aminoimidazole carboxamide ribonucleotide formyltransferase, 5,10-methylenetetrahydrofolate dehydrogenase, and 10-formyltetrahydrofolate synthetase were also evaluated. The end product reversal studies conducted in human cell lines further support the concept that multiple enzyme-inhibitory mechanisms are involved in cytotoxicity. The reversal pattern of LY231514 suggests that although TS may be a major site of action for LY231514 at concentrations near the IC50, higher concentrations can lead to inhibition of DHFR and/or other enzymes along the purine de novo pathway. Studies with mutant cell lines demonstrated that LY231514 requires polyglutamation and transport via the reduced folate carrier for cytotoxic potency. Therefore, our data suggest that LY231514 is a novel classical antifolate, the antitumor activity of which may result from simultaneous and multiple inhibition of several key folate-requiring enzymes via its polyglutamated metabolites.

Medicinal chemistry of difluoropurines.

A series of over 70 difluoropurine analogs was synthesized by varying the C-2, 6 and 8 substituents about the purine ring system. After initial in vitro and in vivo screening, testing concentrated on the 2,6-diaminopurine analog (dFdAP) and the guanosine analog (dFdG). dFDAP appears to be a prodrug for dFdG. Both compounds significantly inhibited mammary tumor growth in mice, caused a moderate inhibition in ovarian and lymphosarcoma models, and demonstrated no activity in lung and melanoma models. This is a narrower spectrum of activity than that of gemcitabine (dFdC). The antitumor activity of dFdAP in human xenografts that are refractory to standard clinical agents was comparable or superior to that of gemcitabine. However, during the preliminary toxicology testing, dFdG was associated with several deaths caused by cardiac toxicity. Therefore, although dFdG is a potentially useful oncolytic, further investigation is required.

Cellular pharmacology of MTA: a correlation of MTA-induced cellular toxicity and in vitro enzyme inhibition with its effect on intracellular folate and nucleoside triphosphate pools in CCRF-CEM cells.

The mechanism of action of an antifolate may be investigated using a variety of experimental methods. These include experiments in a cell culture setting to observe possible protection against drug effects afforded by the end products of metabolic pathways, assessing the activity of purified target enzymes in the presence of the antifolate, and, finally, the measurement of drug effects on intracellular folate and nucleoside triphosphate pools. The current discussion is focused on studies using CCRF-CEM leukemia cells that were designed to compare and contrast mechanisms of action of the antifolates methotrexate, which is primarily a dihydrofolate reductase inhibitor, raltitrexed, a thymidylate synthase inhibitor, LY309887, a glycinamide ribonucleotide formyltransferase inhibitor, and MTA (multitargeted antifolate), which is a novel antifolate antimetabolite. The results of these studies support the hypothesis that MTA affects multiple enzymatic targets and has a distinct mechanism of action from methotrexate, raltitrexed, and LY309887.

Novel cryptophycin antitumor agents: synthesis and cytotoxicity of fragment "B" analogues.

A general synthetic approach to novel cryptophycin analogues 6 is described. N-Hydroxysuccinimide active ester 15, a key common intermediate, was converted to beta-epoxide 6 in three steps, via initial coupling with unprotected amino acid 9, followed by deprotection/macrolactamization of acyclic precursor 16, and final oxidation of styrene 7 to install the C7-C8 beta-epoxide. Cryptophycin styrenes 7 and beta-epoxides 6, bearing diverse side chains in fragment "B", were evaluated for cytotoxic activity. beta-Epoxides 6, in general, were significantly more potent than the corresponding alpha-epoxides 17 and styrenes 7. A benzyl side chain was required for potent activity, with beta-epoxide 6u, possessing a 3-Cl,4-(dimethylamino)benzyl moiety, as the most potent cytotoxic agent prepared, with an IC(50) = 54 pM, only 2-fold less than that of Cryptophycin-52 (3).

Synthesis, pharmacological characterization, and molecular modeling of heterobicyclic amino acids related to (+)-2-aminobicyclo[3.1.0] hexane-2,6-dicarboxylic acid (LY354740): identification of two new potent, selective, and systemically active agonists for group II metabotropic glutamate receptors.

As part of our ongoing research program aimed at the identification of highly potent, selective, and systemically active agonists for group II metabotropic glutamate (mGlu) receptors, we have prepared novel heterobicyclic amino acids (-)-2-oxa-4-aminobicyclo[3.1. 0]hexane-4,6-dicarboxylate (LY379268, (-)-9) and (-)-2-thia-4-aminobicyclo[3.1.0]hexane-4,6-dicarboxylate (LY389795, (-)-10). Compounds (-)-9 and (-)-10 are structurally related to our previously described nanomolar potency group II mGlu receptor agonist, (+)-2-aminobicyclo[3.1.0]hexane-2,6-dicarboxylate monohydrate (LY354740 monohydrate, 5), with the C4-methylene unit of 5 being replaced with either an oxygen atom (as in (-)-9) or a sulfur atom (as in (-)-10). Compounds (-)-9 and (-)-10 potently and stereospecifically displaced specific binding of the mGlu2/3 receptor antagonist ([3H]LY341495) in rat cerebral cortical homogenates, displaying IC50 values of 15 +/- 4 and 8.4 +/- 0.8 nM, respectively, while having no effect up to 100 000 nM on radioligand binding to the glutamate recognition site on NMDA, AMPA, or kainate receptors. Compounds (-)-9 and (-)-10 also potently displaced [3H]LY341495 binding from membranes expressing recombinant human group II mGlu receptor subtypes: (-)-9, Ki = 14.1 +/- 1.4 nM at mGlu2 and 5.8 +/- 0.64 nM at mGlu3; (-)-10, Ki = 40.6 +/- 3.7 nM at mGlu2 and 4.7 +/- 1.2 nM at mGlu3. Evaluation of the functional effects of (-)-9 and (-)-10 on second-messenger responses in nonneuronal cells expressing human mGlu receptor subtypes demonstrated each to be a highly potent agonist for group II mGlu receptors: (-)-9, EC50 = 2.69 +/- 0.26 nM at mGlu2 and 4.58 +/- 0.04 nM at mGlu3; (-)-10, EC50 = 3.91 +/- 0.81 nM at mGlu2 and 7.63 +/- 2. 08 nM at mGlu3. In contrast, neither compound (up to 10 000 nM) displayed either agonist or antagonist activity in cells expressing recombinant human mGlu1a, mGlu5a, mGlu4a, or mGlu7a receptors. The agonist effects of (-)-9 and (-)-10 at group II mGlu receptors were not totally specific, however, as mGlu6 agonist activity was observed at high nanomolar concentrations for (-)-9 (EC50 = 401 +/- 46 nM) and at micromolar concentrations (EC50 = 2 430 +/- 600 nM) for (-)-10; furthermore, each activated mGlu8 receptors at micromolar concentrations (EC50 = 1 690 +/- 130 and 7 340 +/- 2 720 nM, respectively). Intraperitoneal administration of either (-)-9 or (-)-10 in the mouse resulted in a dose-related blockade of limbic seizure activity produced by the nonselective group I/group II mGluR agonist (1S,3R)-ACPD ((-)-9 ED50 = 19 mg/kg, (-)-10 ED50 = 14 mg/kg), indicating that these molecules effectively cross the blood-brain barrier following systemic administration and suppress group I mGluR-mediated limbic excitation. Thus, heterobicyclic amino acids (-)-9 and (-)-10 are novel pharmacological tools useful for exploring the functions of mGlu receptors in vitro and in vivo.

Cell cycle effects of antifolate antimetabolites: implications for cytotoxicity and cytostasis.

PURPOSE: Cell cycle-related events in CCRF-CEM lymphocytic leukemia cells were examined subsequent to inhibition of thymidylate synthase (TS) or GAR formyltransferase (GARFT) and prior to cell death or stasis. METHODS: Cell populations were treated with the GARFT inhibitors 6R-5, 10-dideazatetrahydrofolate (Lometrexol) or LY309887, the TS inhibitor ZD1694, or the multitargeted antifolate LY231514. DNA content, nucleoside precursor incorporation and proliferating cell nuclear antigen (PCNA) expression as functions of drug treatment were assessed by multiparameter flow cytometry. Cellular respiration was measured by MTT analysis and apoptosis was detected by extraction of DNA fragments. RESULTS: Cell populations treated for up to 96h with lometrexol or LY309887 did not replicate and maintained a cell cycle distribution with distinct G1, S and G2/M regions. The number of S phase cells in treated populations was slightly elevated relative to control as measured by DNA content and PCNA. However, these cells were unable to incorporate 5-bromodeoxyuridine (BrdU). Throughout treatment, cells incubated with GARFT inhibitors maintained intact membranes and respired at a level comparable to untreated cells. In contrast, ZD1694 as well as LY231514, induced synchronization of the treatment population at the G1/S interface within 12h of drug addition. This was followed by synchronous entry of the population into S phase. After 24 h of treatment, more than 90% of the cells were capable of incorporating BrdU and stained positive for PCNA. DNA fragmentation occurred in cells treated with ZD1694 or LY231514 but not in those treated with GARFT inhibitors. In addition, the viable cells remaining after 24-48 h of treatment with ZD1694 or LY231514 were respiring at twice the level of untreated cells. CONCLUSION: These results demonstrate that the distinct endpoints of GARFT and TS inhibition are preceded by distinct cell cycle and metabolic alterations.

Role of membrane-associated folate binding protein in the cytotoxicity of antifolates in KB, IGROV1, and L1210A cells.

Based on differential levels of membrane-associated folate binding protein (mFBP) expression, murine L1210 leukemia, human KB epidermoid carcinoma, and human IGROV1 ovarian carcinoma cells maintained under low (physiological) folate conditions (2 nM folinic acid) were used as model systems to investigate the potential role of mFBP in antifolate transport. In addition, L1210 parental cells were compared to a subline, L1210A, expressing high levels of mFBP and defective reduced folate carrier. Antifolates for which KB-derived mFBP has high affinity (5, 10-dideazatetrahydrofolic acid [DDATHF] and homo-DDATHF [0.24 and 0.78 respectively relative to folic acid]) and low affinity (methotrexate [0.002]) were chosen for this study. Protection against DDATHF/homo-DDATHF induced cytotoxicity was achieved preferentially by folic acid compared to folinic acid in IGROV1 and L1210A cells. In IGROV1 cells, cytotoxicity IC50s were increased 18- and 5.5-fold for DDATHF and homo-DDATHF respectively by 20 nM folic acid. Moreover, greater protection was observed in L1210A cells, where IC50s were increased 354- and 80-fold for these same compounds by 20 nM folic acid. Similar protection was not observed in KB cells, suggesting that KB mFBP was not functional in DDATHF transport. Although mFBP expression may be an important determinant in the cytotoxicity of antifolates for certain tumor cells, our data demonstrate a lack of correlation between levels of mFBP and function of mFBP for DDATHF transport in the models studied.

Evaluation of new anticancer agents against the MIA PaCa-2 and PANC-1 human pancreatic carcinoma xenografts.

Human pancreatic carcinoma xenograft models were developed from established MIA PaCa-2 and PANC-1 cell lines (ATCC, Rockville, MD). Tumors were maintained by serial trocar implantation in CD1 nu/nu mice, and attempts were made to test all drugs under optimal schedules at maximum tolerated doses. In both models, adriamycin, cisplatin, and 5-fluorouracil were inactive (< 60% inhibition of tumor weight), whereas gemcitabine (LY188011] produced modest activity (69% inhibition in MIA PaCa-2 and 76% inhibition in PANC-1. Major differences in tumor sensitivity were noted with diarylsulfonylureas (DSU) and taxol. The DSU (Sulofenur [LY186641] and LY295501) produced complete inhibition in the MIA PaCa-2 xenograft, but were inactive in PANC-1. Conversely, taxol produced 80% inhibition of PANC-1 tumor growth, but was inactive against MIA PaCa-2. In general, in vivo antitumor activity roughly correlated with in vitro tumor cytotoxicity with the exception of DSU. We have previously shown that DSU are extensively bound to albumin and that in vitro cytotoxic activity in serum-containing medium is not predictive of in vivo antitumor activity. The MIA PaCa-2 and PANC-1 xenograft models may be useful for selecting potential candidates for therapy of human pancreatic cancer.

Effect of albumin on antitumor activity of diarylsulfonylureas.

Several diarylsulfonylureas (DSU), including Sulofenur (LY186641) and LY181984, have been described that exhibit wide spectrum and high therapeutic activity against murine solid tumors and human tumor xenografts. The mechanism for antitumor activity is poorly understood. Moreover, in vitro cytotoxic activity in serum-containing medium is not predictive of in vivo antitumor activity for DSU. Since DSU are extensively bound to serum albumin (> 99%), we sought to determine the effect of albumin on tumor cytotoxicity. We adapted human CCRF-CEM leukemia and GC3 colon carcinoma cells for growth in UltraCHO serum- and albumin-free medium. In comparisons between normal growth medium (RPMI-1640 with 10% fetal bovine serum) and UltraCHO medium, the unbound fraction of drug correlated better with cytotoxic activity than did the total drug. Tumor cytotoxicity by DSU required > 24 h and was markedly enhanced in UltraCHO medium. For example, LY181984 and Sulofenur had IC50 values of 7.4 and 12.1 micrograms/ml against CCRF-CEM in normal growth medium and 0.6 and 0.2 microgram/ml in UltraCHO. Moreover, DSU with the lowest IC-50s in albumin-free medium displayed the most potent in vivo antitumor activity in the 6C3HED lymphosarcoma. A Sulofenur-resistant CCRF-CEM cell line was developed by culturing the cells for > 20 passages in UltraCHO medium containing LY186641 at 2 micrograms/ml (10X IC-50). This line showed approximately 18-fold resistance to LY186641, but did not show cross-resistance to vinblastine, actinomycin D, or doxorubicin. The albumin-free conditions may be useful for further mechanistic studies on the antitumor action by DSU. Further studies are underway to determine whether DSU structural requirements for cytotoxicity an albumin binding are intrinsically linked.

In vitro and in vivo antitumor activity of the phosphatidylinositol-3-kinase inhibitor, wortmannin.

The microbial product wortmannin has previously been shown to be a potent inhibitor of phosphatidylinositol-3-kinase. In view of the potential role of this enzyme in transduction of mitogenic signals, we determined the cytotoxic activity of wortmannin against several human tumor cell lines in vitro. The most sensitive lines included GC3 colon carcinoma, IGROV1 ovarian carcinoma, and CCRF-CEM leukemia (IC-50s ranging from 0.7-2.1 microM). The cytotoxicity of wortmannin was decreased approximately 10-fold by serum-free conditions. Wortmannin was generally less active in low passage human breast cancer cell lines that overexpress either epidermal growth factor receptor or Her2/neu. Wortmannin was also tested for in vivo antitumor activity against seven murine tumor and ten human tumor xenograft models. Activity (> 60% inhibition of tumor growth) was observed in only the C3H mammary carcinoma and the human BxPC-3 pancreatic carcinoma xenograft. In vivo antitumor activity did not correlate with in vitro sensitivity to wortmannin cytotoxicity.

Substrate specificity of mammalian folylpolyglutamate synthetase for 5,10-dideazatetrahydrofolate analogs.

The metabolism of 5,10-dideazatetrahydrofolate (DDATHF [lometrexol]) to polyglutamate derivatives by folylpoly-gamma-glutamate synthetase (FPGS) plays a central role in the activity of this compound as an antineoplastic agent. The availability of a series of DDATHF derivatives differing in structure throughout the molecule has allowed a study of the structural requirements for substrate activity with mouse liver and hog liver FPGS. Kinetics of the polyglutamation reaction in vitro have been related to the potency of these compounds as inhibitors of the growth of human CEM leukemic cells. The structure-activity relationships for enzyme from both sources were nearly identical. FPGS from both species showed a broad acceptance for structural changes in the pyridopyrimidine ring, in the phenyl group, and in the intermediate bridge region, with structural changes in these regions being reflected in changes in Km for FPGS but much more modest alterations in Vmax. The data suggested that the phenyl ring was not contributing to any pi-pi hydrophobic interactions. It appeared to function primarily in maintaining a favorable distance between the pyridopyrimidine ring and the glutamate side chain. The lowest Km values were found for DDATHF analogs in which there were small alterations at the 10 position, e.g., 5-deazatetrahydrofolate, 10-methyl-DDATHF, and 10-formyl-5-deazatetrahydrofolate; the first-order rate constants for these substrates were the highest in this series, an indication of the efficiency of polyglutamation at low substrate concentrations. After correction for the intrinsic inhibitory activity of the parent DDATHF analog as an inhibitor of the target enzyme, the first-order rate constants for FPGS were found to be predictive of the potency of tumor cell growth inhibition for most of the compounds in this structural series.

Synthesis and biological evaluation of cryptophycin analogs with substitution at C-6 (fragment C region).

Analogs of the antitumor agents cryptophycins 1 and 8 with dialkyl substitution at C-6 (fragment C) were synthesized and evaluated for in vitro cytotoxicity against human leukemia cells (CCRF-CEM). The activity of these analogs decreased as the size of the substituents at C-6 increased. The C-6 spirocylopropyl compound (2g) was highly potent in vitro and showed excellent antitumor activity in animal models.

The synthesis and biological activity of a series of 2,4-diaminopyrido[2,3-d]pyrimidine based antifolates as antineoplastic and antiarthritic agents.

A new series of 2,4-diaminopyrido[2,3-d]pyrimidine based antifolates 1-3 were synthesized through an efficient conversion of 2-pivaloyl-4-oxo-6-ethynylpyrido[2,3-d]pyrimidine 5 to the corresponding 4-amino analog 7 via the activated 1,2,4-triazole intermediate 6. Compound 7 was used as the key intermediate for the preparation of the final products. The detailed biological evaluation of these compounds both as antineoplastic and antiarthritic agents will be discussed.

Synthesis and biological evaluation of novel cryptophycin analogs with modification in the beta-alanine region.

Structure modification of the beta-alanine region (fragment C) of the potent antimitotic agent cryptophycin was investigated. This includes: (1) introduction of substituents at the previously unsubstituted C7 position of the macrolide ring and (2) replacement of the (2R)-3-amino-2-methyl-propanoic acid (beta-alanine) with various (1)-amino acids to give the corresponding 15-membered unnatural cryptophycin analogs.

Preclinical cellular pharmacology of LY231514 (MTA): a comparison with methotrexate, LY309887 and raltitrexed for their effects on intracellular folate and nucleoside triphosphate pools in CCRF-CEM cells.

LY231514 (N-[4-[2-(2-amino-3,4-dihydro-4-oxo-7H-pyrrolo[2,3-d]pyrimidin-5-yl)ethy l]-benzoyl]-L-glutamic acid) is a new folate-based antimetabolite currently in broad phase II clinical evaluation. Previous in vitro studies (C. Shih et al, CancerRes 57: 1116-1123, 1997) have suggested that LY231514 could be a multitargeted antifolate (MTA) capable of inhibiting thymidylate synthase (TS), dihydrofolate reductase (DHFR) and glycinamide ribonucleotide formyltransferase (GARFT). The present study compared LY231514 with methotrexate, raltitrexed and a glycinamide ribonucleotide formyltransferase inhibitor, LY309887, at 300, 100, 30 and 100 nM, respectively, for their effects on intracellular folate and at 100, 66, 20 and 30 nM respectively, for their effects on nucleoside triphosphate pools in CCRF-CEM cells. Methotrexate induced an accumulation of dihydrofolate species, together with a rapid depletion of ATP, GTP and all of the deoxynucleoside triphosphates. LY309887 caused an accumulation of 10-formyltetrahydrofolate, a rapid loss of ATP, GTP and dATP, but a slower loss in dCTP, dTTP and dGTP. Both LY231514 and raltitrexed had minimal effects on folate pools. In contrast, they caused rapid depletion of dTTP, dCTP and dGTP, but induced an accumulation of dATP at different rates, with raltitrexed doing so about 2.5 times faster. Most of the observed metabolic changes could be understood on the basis of current knowledge of folate and nucleotide metabolism. We concluded that LY231514 was distinct from methotrexate, LY309887 and raltitrexed based on their metabolic effects in CCRF-CEM cells, and that in this cell line the inhibitory effects of LY231514 were exerted primarily against the thymidylate cycle and secondarily against de novo purine biosynthesis.