Structures of three inhibitor complexes provide insight into the reaction mechanism of the human methylenetetrahydrofolate dehydrogenase/cyclohydrolase.

Enzymes involved in tetrahydrofolate metabolism are of particular pharmaceutical interest, as their function is crucial for amino acid and DNA biosynthesis. The crystal structure of the human cytosolic methylenetetrahydrofolate dehydrogenase/cyclohydrolase (DC301) domain of a trifunctional enzyme has been determined previously with a bound NADP cofactor. While the substrate binding site was identified to be localized in a deep and rather hydrophobic cleft at the interface between two protein domains, the unambiguous assignment of catalytic residues was not possible. We succeeded in determining the crystal structures of three ternary DC301/NADP/inhibitor complexes. Investigation of these structures followed by site-directed mutagenesis studies allowed identification of the amino acids involved in catalysis by both enzyme activities. The inhibitors bind close to Lys56 and Tyr52, residues of a strictly conserved motif for active sites in dehydrogenases. While Lys56 is in a good position for chemical interaction with the substrate analogue, Tyr52 was found stacking against the inhibitors' aromatic rings and hence seems to be more important for proper positioning of the ligand than for catalysis. Also, Ser49 and/or Cys147 were found to possibly act as an activator for water in the cyclohydrolase step. These and the other residues (Gln100 and Asp125), with which contacts are made, are strictly conserved in THF dehydrogenases. On the basis of structural and mutagenesis data, we propose a reaction mechanism for both activities, the dehydrogenase and the cyclohydrolase.

Cellular ATP depletion by LY309887 as a predictor of growth inhibition in human tumor cell lines.

The antifolate LY309887 is a specific glycinamide ribonucleotide formyltransferase inhibitor that blocks de novo purine synthesis and produces a depletion of purine nucleotides. The activity of LY309887 in six human tumor cell lines has been examined by growth inhibition and clonogenic assay after continuous exposure for three cell doubling times and by ATP depletion at 24 h. Three cell lines (CCRF-CEM, MCF7, and GC3) were sensitive to LY309887-induced growth inhibition (IC50: 5.6-8.1 nM), whereas the other cell lines (COR-L23, T-47D, and A549) were comparatively resistant (IC50: 36-55 nM). Sensitivity to LY309887 cytotoxicity was consistent with sensitivity to growth inhibition in four of five cell lines tested (MCF7/GC3: 0.01% survival and COR-L23/T-47D: 1-5% survival at 100 nM LY309887). LY309887-induced ATP depletion was measured by luciferase-based ATP assay and confirmed by high performance liquid chromatography measurements. There was a linear relationship between ATP depletion and growth inhibition when data were analyzed for all six cell lines (r2 = 0.93; P < 0.0001). Depletion of 24-h cellular ATP concentrations to < 1 mM was associated with both cell growth inhibition and cytotoxicity in all cell lines studied. In conclusion, cellular ATP depletion induced by LY309887 can be used to predict growth inhibition and cytotoxicity in human tumor cells.

Enzyme inhibition, polyglutamation, and the effect of LY231514 (MTA) on purine biosynthesis.

The pyrrolopyrimidine-based antifolate, N-¿4-[2-(2-amino-3,4-dihydro-4-oxo-7H-pyrrolo[2,3-d]pyrimidin-5-yl)ethyl ]benzoyl¿glutamic acid, LY231514 (MTA) has demonstrated antitumor activity in a broad array of human tumors, including breast cancer, colon cancer, non-small cell lung cancer, head and neck cancer, pancreatic cancer, and other solid tumors. The biochemical basis of this activity was explored by measuring activation of MTA by polyglutamation and the activity of MTA to inhibit several folate-dependent enzymes: thymidylate synthase, dihydrofolate reductase, and glycinamide ribonucleotide formyltransferase (GARFT). The enzyme folylpolyglutamate synthase (FPGS) activated MTA very efficiently. Using FPGS from two different species, Km values below 2 micromol/L and high relative first order rate constants, k' (Vmax/Km) of 6.4 and 13.7 compared with another substrate, lometrexol, were obtained. The formation of polyglutamates of several antifolates were compared in vitro at high and low substrate concentrations. At low MTA concentrations, tetraglutamated and pentaglutamated MTA were the predominant forms identified after a 24-hour incubation period. In contrast, only diglutamyl methotrexate and a mixture triglutamylated, tetraglutamylated, and pentaglutamylated forms of the GARFT inhibitor lometrexol were formed under the same conditions. At higher substrate concentrations (20 micromol/L, 24 hours), greater amounts of each product were formed. The major metabolites, however, were triglutamated MTA or triglutamated lometrexol, while only diglutamyl methotrexate was recovered. Thus, MTA was an excellent substrate for FPGS and it was efficiently metabolized to highly polyglutamated species by this enzyme. The activity of MTA and its polyglutamated metabolites to inhibit several folate-dependent enzymes was measured. In vitro, MTA and its polyglutamates were potent, tight-binding inhibitors of several folate-dependent enzymes, including thymidylate synthase, dihydrofolate reductase, and GARFT. Preliminary cell-based assays (CCRF-CEM) demonstrated inhibition of the purine de novo pathway by MTA, consistent with its multitargeted mechanism of action against tumor cells. The combined effects of activation of MTA to highly polyglutamated metabolites and the potency of these polyglutamates to inhibit multiple folate-dependent enzymes provide a mechanistic basis for understanding the broad antitumor activity of this compound against many human tumor types.

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.

Biological activity of the multitargeted antifolate, MTA (LY231514), in human cell lines with different resistance mechanisms to antifolate drugs.

Prior studies have indicated that MTA requires intracellular polyglutamation for optimal cytotoxic effect and that these polyglutamates potently inhibit several key enzymes of folate metabolism, including thymidylate synthase (TS), dihydrofolate reductase, and glycinamide ribonucleotide formyltransferase (GARFT). In the present studies, we have investigated the mechanistic basis for resistance to MTA in several human tumor cell lines. The cell lines were developed for resistance by the gradual exposure to stepwise (fivefold) increases in the concentration of MTA over a 5-month period. The degree of resistance was 140-fold for GC3 colon carcinoma, 117-fold for HCT-8 ileocecal carcinoma, and 729-fold for CCRF-CEM leukemia cells adapted to 2 micromol/L MTA. The lines had strong cross-resistance (>3,200-fold) to raltitrexed. Only modest resistance was noted for methotrexate and the GARFT inhibitor, LY309887. The cytotoxicity of MTA in wild-type cells was only partially alleviated by thymidine addition (5 micromol/L) and complete protection required the addition of both hypoxanthine (100 micromol/L) and thymidine. In contrast, thymidine alone totally lacked protective activity in the MTA-resistant lines. The cells either demonstrated a GARFT-like reversal pattern (complete protection by hypoxanthine) for GC3MTA or a dihydrofolate reductase-like reversal pattern (complete protection by the combination of hypoxanthine and thymidine) for HCT-8MTA and CCRF-CEM(MTA) cells. Cellular resistance was multifactorial and stable on removal of selective pressure. Only GC3MTA cells showed increased TS activity (approximately 40-fold). Accumulations of 3H-MTA at 24 hours in CCRF-CEM(MTA), HCT-8MTA, and GC3MTA cells were 2%, 6%, and 46% of wild-type values, respectively. We also evaluated the cytotoxic activity of MTA in MCF-7 breast carcinoma and H630 colon carcinoma cells selected for resistance to raltitrexed and 5-fluorouracil, respectively, via TS amplification (provided by Dr P.G. Johnston, Belfast, Ireland). These cells demonstrated more than 200-fold less resistance to MTA compared with raltitrexed and MTA-induced cytotoxicity was prevented by hypoxanthine. These studies suggest that in addition to TS modulation, secondary targets emerge during the development of MTA resistance.

Multiple folate enzyme inhibition: mechanism of a novel pyrrolopyrimidine-based antifolate LY231514 (MTA).

Extensive biochemical and pharmacological evidence indicates that LY231514 is a novel antifolate antimetabolite. LY231514 is transported into cells mainly through the reduced folate carrier system and extensively metabolized to polyglutamated forms. The polyglutamates of LY231514 inhibit at least three key folate enzymes: TS, DHFR, and GARFT, and to a lesser extent AICARFT and C1-tetrahydrofolate synthase. The combined effects of the inhibition exerted by LY231514 at each target give rise to an unusual end-product reversal pattern at the cellular level that is distinct from those of other inhibitors such as methotrexate and the quinazoline antifolates. The metabolic effects exerted by LY231514 on the folate and nucleotide pools are also quite distinct from those of MTX and LY309887. The efficient polyglutamation, longer cellular retention and the multiple folate enzyme inhibition mechanism may all have contributed directly to the exciting antitumor responses now observed in Phase I and II studies. The multitargeted inhibition mechanism of LY231514 is particularly intriguing. This new level of mechanistic insight, which has evolved from the study of LY231514, challenges the traditional concept and paradigm of antifolate drug discovery and development which focused on developing very potent and selective inhibitors of single folate enzyme targets, such as DHFR, TS or one of the enzymes along the de novo purine biosynthetic pathway. Given the complex nature of folate metabolism and the critical role of folates in maintaining the physiological functions of living systems, it is completely reasonable to suspect that agents which can interfere at multiple sites in the folate pathway may trigger and cause more biochemical imbalance in the cellular DNA and RNA synthesis of malignant cells than agents that act on a single point (Fig. 5). In conclusion, LY231514 (MTA) is a new generation antifolate antimetabolite demonstrating inhibitory activity against multiple folate enzymes including TS, DHFR and GARFT. In current phase II studies, MTA is broadly active as a single agent and is showing very encouraging antitumor activity in multiple solid tumors including colorectal, breast and non-small cell lung cancers (38-43). The every three week dosing schedule has proven to be convenient and easy to administer and the clinical toxicities of LY231514 seem to be well tolerated. More advanced and extensive clinical trials of LY231514 are currently in progress.

The antiproliferative and cell cycle effects of 5,6,7, 8-tetrahydro-N5,N10-carbonylfolic acid, an inhibitor of methylenetetrahydrofolate dehydrogenase, are potentiated by hypoxanthine.

5,6,7,8-Tetrahydro-N5,N10-carbonylfolic acid (LY354899) has been demonstrated to inhibit the dehydrogenase activity of C1-tetrahydrofolate synthase. This compound was only moderately antiproliferative toward CCRF-CEM lymphocytic leukemia cells in culture, but induced apoptosis after long incubation times. Slightly greater potency was observed in CEM cells adapted to grow in low folate media. Cell cycle alterations induced by LY354899 were unique relative to antifolates that inhibit either the purine or thymidine de novo biosynthetic pathways. Based on the observed changes in DNA content, we hypothesized that inhibition of the dehydrogenase resulted in two temporally distinct events: the first was a purineless-like effect and the second was a thymineless-like effect that resulted in apoptosis. To test this hypothesis, we combined LY354899 with the purine salvage metabolite, hypoxanthine. This combination resulted in an earlier and more dramatic apoptotic response, indicating that the thymineless effect had been potentiated. Biochemical analysis of ribo- and deoxyribonucleoside triphosphates confirmed that inhibition of the dehydrogenase activity initially resulted in decreased pools of deoxypurines and deoxypyrimidines, followed 16 hr later by an increase in deoxyadenosine triphosphate (dATP) and a further decrease in deoxythymidine triphosphate (dTTP). These studies demonstrate that the inhibition of the dehydrogenase activity of C1-tetrahydrofolate synthase may represent a viable target for the development of novel antifolates. The results are discussed in terms of deoxypurine and deoxypyrimidine biosynthesis.

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.

Test of a convergence model for color transparency perception.

Models of color transparency suggest that a region in which colors of surfaces converge in color space will appear transparent. The convergence is described by a transparency parameter alpha and a target of convergence. To test such models psychophysically, observers were presented a display with four colored areas. The colors of three of the areas were chosen in advance by the experimenter. The task of the observer was to choose the color of the fourth area to make a central region appear transparent. Settings for the fourth color were collected for a total of twenty-four color combinations chosen from three planes in color space. Observers' settings agreed well with the model, which predicts that choices for the fourth color lie along a line segment in color space that is parameterized by alpha. The results suggest further that color discriminability and color opponency also influence transparency judgment.