Tangential migration of corridor guidepost neurons contributes to anxiety circuits.

In mammals, thalamic axons are guided internally toward their neocortical target by corridor (Co) neurons that act as axonal guideposts. The existence of Co-like neurons in non-mammalian species, in which thalamic axons do not grow internally, raised the possibility that Co cells might have an ancestral role. Here, we investigated the contribution of corridor (Co) cells to mature brain circuits using a combination of genetic fate-mapping and assays in mice. We unexpectedly found that Co neurons contribute to striatal-like projection neurons in the central extended amygdala. In particular, Co-like neurons participate in specific nuclei of the bed nucleus of the stria terminalis, which plays essential roles in anxiety circuits. Our study shows that Co neurons possess an evolutionary conserved role in anxiety circuits independently from an acquired guidepost function. It furthermore highlights that neurons can have multiple sequential functions during brain wiring and supports a general role of tangential migration in the building of subpallial circuits.

Apple ethanol extract promotes proliferation of human adult stem cells, which involves the regenerative potential of stem cells.

Tissue regeneration using adult stem cells (ASCs) has significant potential as a novel treatment for many degenerative diseases. Previous studies have established that age negatively affects the proliferation status and differentiation potential of ASCs, suggesting a possible limitation in their potential therapeutic use. Therefore, we hypothesized that apple extract might exert beneficial effects on ASCs. The specific objectives were to investigate the proliferative effect of apple ethanol extract on human adipose tissue-derived mesenchymal stem cells (ADSCs) and human cord blood-derived mesenchymal stem cells (CB-MSCs), and identify the possible molecular mechanisms. Apple extract promoted proliferation of ADSCs and CB-MSCs as determined by 3-(4,5-dimethylthiazol-2-yl)-2,5-diphenyltetrazolium bromide and Click-iT 5-ethynyl-2'-deoxyuridine flow cytometry assays. In addition, phosphorylation of p44/42 MAPK (ERK), mammalian target of rapamycin (mTOR), p70 S6 kinase (p70S6K), S6 ribosomal protein (S6RP), eukaryotic initiation factor (eIF) 4B and eIF4E was induced stepwise in ADSCs. Furthermore, apple extract significantly induced the production of vascular endothelial growth factor and interleukin-6 in both ADSCs and CB-MSCs. Similarly, apple extract-induced phosphorylation of the mTOR/p70S6K/S6RP/eIF4B/eIF4E pathway was blocked by pretreatment with PD98059, a specific ERK inhibitor. These results indicate that apple extract-induced proliferation of ADSCs under serum-free conditions is mediated by ERK-dependent cytokine production. Moreover, the beneficial effect of apple extract on proliferation of ASCs may overcome the limitation in therapeutic use of stem cells in tissue regeneration and maintenance of stem cell homeostasis.

Icariin promotes self-renewal of neural stem cells: an involvement of extracellular regulated kinase signaling pathway.

OBJECTIVE: To investigate the effects and underlying molecular mechanisms of icariin (ICA) on self-renewal and differentiation of neural stem cells (NSCs). METHODS: NSCs were derived from forebrains of mice embryos by mechanical dissociation into single cell suspension. The self-renewal of NSCs was measured by neurosphere formation assay. The proliferation of NSCs was detected by water-soluble tetrazolium (WST) and 5-ethynyl-2'-deoxyuridine (EdU) incorporation assay. Protein expression of neuron-specific marker tubulin-ßIII(TuJ1) and astrocyte-specific marker glial fibrillary acidic protein (GFAP) were measured by immunofluorescence and Western blotting. Using microarray, the differentially expressed genes (DEGs) were screened between NSCs with or without ICA treatment. The signaling pathways enriched by these DEGs and their role in mediating effects of ICA were analyzed. RESULTS: ICA significantly promoted neurosphere formation of NSCs cultured in growth protocol in a dose-dependent manner and achieved the maximum effects at 100 nmol/L. ICA also increased optical absorbance value and EdU incorporation into nuclei of NSCs. ICA had no significant effects on the percentage of TuJ1 or GFAP-positive cells, and TuJ1 or GFAP protein expression in NSCs cultured in differentiation protocol. A total of 478 genes were found to be differentially regulated. Among signaling pathways significantly enriched by DEGs, mitogen activated protein kinase (MAPK) pathway was of interest. Blockade of extracellular signal-regulated kinase (ERK)/MAPK, other than p38/MAPK subfamily pathway partially abolished effects of ICA on neurosphere formation and EdU incorporation of NSCs. CONCLUSION: ICA can promote the selfrenewal of NSCs at least partially through ERK/MAPK signaling pathway.

Trypanosoma brucei (UMP synthase null mutants) are avirulent in mice, but recover virulence upon prolonged culture in vitro while retaining pyrimidine auxotrophy.

African trypanosomes are capable of both de novo synthesis and salvage of pyrimidines. The last two steps in de novo synthesis are catalysed by UMP synthase (UMPS) - a bifunctional enzyme comprising orotate phosphoribosyl transferase (OPRT) and orotidine monophosphate decarboxylase (OMPDC). To investigate the essentiality of pyrimidine biosynthesis in Trypanosoma brucei, we generated a umps double knockout (DKO) line by gene replacement. The DKO was unable to grow in pyrimidine-depleted medium in vitro, unless supplemented with uracil, uridine, deoxyuridine or UMP. DKO parasites were completely resistant to 5-fluoroorotate and hypersensitive to 5-fluorouracil, consistent with loss of UMPS, but remained sensitive to pyrazofurin indicating that, unlike mammalian cells, the primary target of pyrazofurin is not OMPDC. The null mutant was unable to infect mice indicating that salvage of host pyrimidines is insufficient to support growth. However, following prolonged culture in vitro, parasites regained virulence in mice despite retaining pyrimidine auxotrophy. Unlike the wild-type, both pyrimidine auxotrophs secreted substantial quantities of orotate, significantly higher in the virulent DKO line. We propose that this may be responsible for the recovery of virulence in mice, due to host metabolism converting orotate to uridine, thereby bypassing the loss of UMPS in the parasite.

The additivity of substrate fragments in enzyme-ligand binding.

BACKGROUND: Enzymes have evolved to recognise their target substrates with exquisite selectivity and specificity. Whether fragments of the substrate--perhaps never available to the evolving enzyme--are bound in the same manner as the parent substrate addresses the fundamental basis of specificity. An understanding of the relative contributions of individual portions of ligand molecules to the enzyme-binding interaction may offer considerable insight into the principles of substrate recognition. RESULTS: We report 12 crystal structures of Escherichia coli thymidylate synthase in complexes with available fragments of the substrate (dUMP), both with and without the presence of a cofactor analogue. The structures display considerable fidelity of binding mode and interactions. These complexes reveal several interesting features: the cofactor analogue enhances the localisation of substrate and substrate fragments near the reactive thiol; the ribose moiety reduces local disorder through additional specific enzyme-ligand interactions; the pyrimidine has multiple roles, ranging from stereospecificity to mechanistic competence; and the glycosidic linkage has an important role in the formation of a covalent attachment between substrate and enzyme. CONCLUSIONS: The requirements of ligand-protein binding can be understood in terms of the binding of separate fragments of the ligand. Fragments which are subsystems of the natural substrate for the enzyme confer specific contributions to the binding affinity, orientation or electrostatics of the enzymatic mechanism. This ligand-binding analysis provides a complementary method to the more prevalent approaches utilising site-directed mutagenesis. In addition, these observations suggest a modular approach for rational drug design utilising chemical fragments.

[Effects of modulators and cytostatics on catabolism of I-125 desoxyuridine in tumors (express-method of antineoplastic modulator screening)]. / Deistvie immunomoduliatorov i tsitostatikov na katabolizm 125I-dezoksiuridina v opukholi (ékspress-metod skrininga protivoopukholevykh immunomoduliatorov).

El-4 and P-815 murine tumor cells labelled by 125I-deoxyuridine or 51Cr were administered in 7-day subcutaneous syngeneic tumors or subcutaneously. At the same time different groups of mice were treated by LPS plus MDP, beta-C7H15-MDP, dexal-MDP, polyacrylamide-MDP-phosphatidylethanolamine, adriblastin or cyclophosphamide. It was shown that cytostatics and immunomodulators significantly delayed catabolism and withdrawing of 125I-deoxyuridine (that has not been incorporated in DNA) from tumor cells. This delay was correlated with the inhibition of tumor nodes growth rate. It is concluded that influence of cytostatics and immunomodulators on catabolism and withdrawing rate of 125I-deoxyuridine from tumor cells relates to their cytostatic effect and may be used at the earliest screening step of immunomodulator analysis.

Synergy of inhibition of DNA synthesis in human bone marrow by azidothymidine plus deficiency of folate and/or vitamin B12?

The effect of azidothymidine (Zidovudine, AZT) on pyrimidine (thymidine, deoxyuridine, and thymidine triphosphate) incorporation into DNA in folate- and/or vitamin B12-deficient and normal human bone marrow cells was studied to investigate whether such vitamin deficiency affects susceptibility to AZT-induced hematologic toxicity. Bone marrow cells from 12 patients were studied: 5 had folate and/or vitamin B12 deficiency; 7 controls included 5 with anemia related to chronic disease and 2 with iron deficiency. At 0.2 microM AZT (3 hr, 37 degrees C), the approximate pharmacologic serum trough level, pyrimidine incorporation into DNA was suppressed by 12 to 19% in folate- and/or vitamin B12-deficient cells and by 16 to 23% in normal cells. At 2.0 microM AZT (3 hr, 37 degrees C), the approximate pharmacologic serum peak level, this was suppressed by 15 to 40% in folate- and/or vitamin B12-deficient cells and by 32 to 47% in controls. Deoxyuridine incorporation into DNA was inhibited significantly greater than thymidine at 2.0 microM AZT (3 hr, 37 degrees C) in both groups. Inhibition of deoxyuridine incorporation was not reversed with methyltetrahydrofolate or vitamin B12. There tended to be less striking suppression by AZT of deoxyuridine incorporation into DNA in bone marrow cells from vitamin B12-deficient patients, which was made more striking by adding vitamin B12. This suggests that some of what passes for "AZT damage" to bone marrow cells may in fact be coincident deficiency of vitamin B12. AZT inhibition of DNA synthesis in 3 hr bone marrow cultures is relatively consistent in a variety of hematologic disorders. As approximately two-thirds of AIDS patients appear to be in negative balance with respect to folate and/or vitamin B12, the fact that AZT-induced inhibition of pyrimidine incorporation into DNA is occurring in cells which may be megaloblastic, i.e., in a state of impaired DNA synthesis, suggests that these cells may be more susceptible to AZT toxicity. The data also support the notion that AZT inhibition results predominantly from termination of DNA chain elongation. Whether folate or vitamin B12 supplementation may partially overcome apparent "AZT inhibition" of DNA synthesis (hematologic toxicity) and whether the benefit of such therapy exceeds the risk will require further study.

Diet-induced DNA damage and altered nucleotide metabolism in lymphocytes from methyl-donor-deficient rats.

Tumor induction with chronic feeding of methyl-donor-deficient diets has been well established; however, the biochemical and molecular mechanisms which predipose to tumorigenesis in this model are still not well understood. The purpose of the present investigation was to assess DNA damage and altered nucleotide metabolism in lymphocytes from Fischer 344 rats fed one of four semi-purified diets: (i) deficient in methionine and choline; (ii) deficient in folic acid; (iii) deficient in methionine, choline and folic acid; or (iv) a supplemented control diet. The accumulation of DNA-strand breaks, as assessed by DNA unwinding in alkali, was increased in lymphocytes from both the methionine/choline-deficient and folate-deficient groups, but was most severe in the group deficient in all three methyl donors. Lymphocyte DNA damage was consistently associated with alterations in folate-dependent thymidylate synthesis, and a decrease in intracellular levels of the DNA-repair-associated pyridine nucleotide, nicotinamide adenine dinucleotide. In the liver, a synergistic lipotropic interaction between folate deficiency and methionine/choline deficiency was observed, confirming the metabolic inter-relationship between these nutrients. Taken together, the results suggest that folate deficiency interacts with methionine/choline deficiency to potentiate symptoms of methyl-donor deficiency and that alterations in folate-dependent thymidylate synthesis are related to DNA damage in lymphocytes. These metabolic aberrations may contribute to immune dysfunction with chronic feeding of methyl-donor-deficient diets.

Further studies on the pharmacologic effects of the 7-hydroxy catabolite of methotrexate in the L1210 murine leukemia cell.

This paper describes studies that further explore the pharmacologic activity of the 7-hydroxy catabolite of methotrexate (7-OH-MTX). A 3-hr exposure of L1210 leukemia cells to 100 microM 7-OH-MTX produced negligible suppression of cell growth despite the build-up of intracellular polyglutamyl congeners to levels 2.7 times greater than the dihydrofolate reductase (DHFR) binding capacity. There was no evidence for direct inhibition of DHFR under these conditions based upon measurements of cellular tetrahydrofolate cofactor and dihydrofolate levels, nor was there suppression of [3H]deoxyuridine incorporation into DNA or [14C]formate incorporation into purines. When the interval of exposure to 100 microM 7-OH-MTX was increased to 6 hr, cell growth was inhibited by 60% and there was mild (approximately 50%) inhibition of purine and thymidylate biosynthesis associated with a small increase in cellular dihydrofolate and a small decline in cellular tetrahydrofolates. Consistent with weak inhibition of DHFR was the absence of significant binding of 7-OH-MTX polyglutamates to DHFR as assessed by gel filtration of cell extracts. Mild direct inhibition of purine biosynthetics by 7-OH-MTX- or MTX-polyglutamyl congeners was demonstrated based upon inhibition of [14C]formate incorporation into purines in cells pretreated with fluorodeoxyuridine so as to prevent tetrahydrofolate cofactor depletion or dihydrofolate polyglutamate build-up. Effects of a 6-hr exposure of cells to 100 microM 7-OH MTX on cell growth were reversed completely by 10 microM leucovorin; effects on cells containing comparable levels of MTX polyglutamyl congeners were unaffected by leucovorin. These studies demonstrate very weak inhibition of L1210 leukemia cell growth and purine, pyrimidine and tetrahydrofolate synthesis by the polyglutamyl congeners of 7-OH-MTX. The data suggest that effects of 7-OH-MTX polyglutamates on folate-requiring enzymes are not likely to play an important role in moderate-dose MTX regimens. However, pharmacologic activity may be expressed in high-dose MTX protocols when high blood levels of 7-OH-MTX are sustained over long intervals to the extent to which polyglutamate congeners accumulate in tumor cells and add to the much more potent inhibitory effects of MTX polyglutamates already present. Pharmacologic activity, however, would be diminished, if not completely reversed, by the concurrent administration of leucovorin.

Folate requirement and metabolism in nonpregnant women.

Folate metabolism and requirements were studied in 10 adult nonpregnant women maintained for 92 d in a metabolic unit. After a folate depletion period of 28 d, the subjects received increasing supplements of folate from food items or as pteroylmonoglutamic acid (PGA). Plasma folate levels fell 60% during the depletion period and continued to fall until 200 micrograms/d of naturally occurring food folates were provided. Supplements of 300 micrograms/d of naturally occurring folates produced a small rise in plasma folate levels although erythrocyte folate levels continued to fall. Lymphocyte deoxyuridine suppression, neutrophil hypersegmentation, and other measurements related to folate metabolism were performed. When compared with PGA, dietary folates appeared to be no more than 50% available. A daily intake of 200-250 micrograms of dietary folates appears to meet the folate requirements of nonpregnant adult women whereas an intake of 300 micrograms/d provides an allowance for storage.

Growth and DNA synthesis of folate- and methionine-depleted L1210 mouse leukemia cells in culture.

The growth and DNA synthesis of L1210 mouse leukemia cells were examined under folate- and methionine-deficient conditions. Cell proliferation was dependent on methionine supplementation rather than on folate concentration. The UdR suppression value was abnormally high in the folate-deficient condition. However, it was also high when the methionine was low, despite folate supplementation. In accordance with this, UdR incorporation was significantly improved with various folates by cells grown in low-methionine conditions. Methionine depletion resulted in marked impairment of UdR incorporation regardless of folate concentration. These findings indicate close metabolic interrelations between folate and methionine, which may be relevant to the pathological biochemistry of human megaloblastic anemia.

High-dose methotrexate therapy with leucovorin rescue: in vitro investigations on human osteosarcoma cell lines.

High-dose methotrexate (MTX) therapy with subsequent leucovorin (LV) rescue (HDMTX-LV) in the treatment of osteosarcoma is based on the assumption that this tumor has a deficient uptake system for MTX and reduced folates. To simulate features of HDMTX-LV therapy protocols in vitro, sensitive and MTX-resistant human osteosarcoma cell lines and a lymphoblastoid cell line were exposed to MTX and/or LV at various dosages and time schedules and the effects on DNA metabolism and on cell growth were evaluated. The data show that in osteosarcoma cells and in lymphoblasts the cytotoxic effects of 10(-6) M to 10(-7) M MTX can be substantially reversed by LV if the antidote is applied within the first 12 h of MTX exposure. The results are not consistent with the assumption mentioned above and should be taken into consideration when designing new therapeutic regimens. An alternative hypothesis for the efficacy of HDMTX-LV is discussed. It is concluded that HDMTX-LV therapy may be effective in the treatment of osteosarcoma, even when subpopulations of the tumor cells exhibit different mechanisms of resistance to MTX, such as elevated levels of dihydrofolate reductase or a deficient transport system for MTX, if high doses of MTX are applied long enough to ensure lethal intracellular MTX levels and low-dose LV schedules instituted after a long delay are used.

Abnormal deoxyuridine suppression as a rapid predictor for antimetabolite chemosensitivity: correlation with in vitro tests of growth inhibition.

The deoxyuridine (dU) suppression test, which estimates the activity of the de novo pathway to DNA synthesis from dU, was evaluated as a predictor of antimetabolite growth inhibition. Observations of growth inhibition were made using flask cell culture and soft agar clonogenic assay and correlated with results of the rapidly performed dU suppression test in human (SK-L7) leukemia cells, in methotrexate-sensitive and -resistant murine (L1210) leukemia cells, and in human tumor explants. The concentration of methotrexate resulting in a positive dU suppression test was closely correlated with the methotrexate concentrations required for growth inhibition in flask and soft agar culture systems. The fact that the dU suppression test can be rapidly interpreted in 4 hours compared to the longer period required for clonogenic assay suggests that further evaluation of this procedure as a rapid predictor for clinical antimetabolite response is warranted.

Biochemical pharmacology of the lipophilic antifolate, trimetrexate.

Trimetrexate is a novel lipophilic folate antagonist that causes growth inhibition, inhibition of nucleic acid biosynthesis, and cytotoxicity at nanomolar concentrations in tissue cultures. The potency of trimetrexate cytotoxicity against most cell lines is greater than that of methotrexate. Trimetrexate has antitumor activity in vivo in several murine leukemia and solid tumor systems, including tumors in which methotrexate is inactive. Antitumor activity was seen following oral, intravenous, or intraperitoneal administration. Trimetrexate causes a pronounced and early depression in incorporation of deoxyuridine into DNA. In tumor cell lines resistant to methotrexate because of a drug transport defect, trimetrexate retains activity. In many such cases the methotrexate-resistant tumors show collateral sensitivity to trimetrexate. In methotrexate-resistant cells with impaired drug transport, trimetrexate sensitivity was even more pronounced when cells were grown in folate-free medium supplemented with physiological levels of tetrahydrofolate cofactor. In the human tumor stem cell colony assay, trimetrexate, at concentrations achievable in vivo, gave activity against many human tumors, including samples that were unresponsive to methotrexate. Trimetrexate crosses the blood-brain barrier, and at very high doses may cause neurotoxicity. At conventional doses the primary toxic effects in mice are gastrointestinal. This toxicity is reversible at therapeutic doses. Unlike earlier lipophilic antifolates, trimetrexate has rapid plasma clearance (t1/2 in mice of 45 minutes). Trimetrexate is a tight-binding competitive inhibitor of dihydrofolate reductase. The Ki,slope for inhibition of the human enzyme was 4 X 10(-11) M. A dose-dependent decrease in cellular purine ribonucleotide pools is given by trimetrexate. Pyrimidine ribonucleotide pools tend to increase in treated cells. Trimetrexate caused a marked depression of cellular pools of dTTP and dGTP, and a lesser depression in dATP. Cytotoxicity of trimetrexate in vitro was prevented by leucovorin. Leucovorin also protected mice from trimetrexate toxicity. Thymidine protected cells from lethal effects of low concentrations of trimetrexate, but not from high concentrations. The combination of thymidine and hypoxanthine completely protected cells from low and high concentrations of trimetrexate. A new, stable and highly water-soluble formulation of trimetrexate has been developed. Because of the interesting biochemical and pharmacological properties of trimetrexate, and its experimental antitumor activity, clinical trials are planned.

Effects of iron deficiency anemia on delayed cutaneous hypersensitivity in mice.

In the present study, the effect of iron-deficiency anemia on delayed cutaneous hypersensitivity was measured using weanling C57BL/6 female mice which were fed either an ad libitum control diet supplemented with 25 to 30 mg Fe/kg diet (FePO4), an iron-deficient test diet (5 to 6 mg Fe/kg diet), or pairfed control diet (25 to 30 mg Fe/kg diet). When skin sensitizing agent (dinitrofluorobenzene) was applied to these animals and skin responses were measured 3 to 5 days later, anemic mice showed a significantly decreased inflammatory skin response than either control or pairfed mice. Five days after sensitization, the animals were challenged with dinitrofluorobenzene painted on the right ear and an equal dose of only the solvent on the left ear followed by 125I-deoxyuridine injected intraperitoneally. The ratio of either total or DNA associated radioactivity incorporated into the right over the left ears was significantly lower in anemic mice than either control or pairfed mice. A single dose of Imferon injected 24 h before the recall dose of dinitrofluorobenzene restored the ratio of 125I-dUR incorporated in anemic mice without having any significant effect on either the control or pairfed groups. The results suggest that iron is not required for sensitization but is required for an effective inflammatory response.

[Radiational basis and practice of the citrovorum factor (Leucovorin) protection after high-dose methotrexate therapy. High-dose methotrexate/leucovorin]. / Rationale Grundlagen und Praxis des Citrovorumfaktor (Leucovorin)-Schutzes nach hochdosierter Methotrexat-Therapie.

The effects of methotrexate on the DNA metabolism of human lymphoblast cultures and bone marrow cells were estimated. The ratio of the 3H-deoxyuridine and 3H-thymidine incorporation rates ia a good parameter for the methotrexate effect. Citrovorum factor (5-formyltetrahydrofolic acid) reverses methotrexate toxicity only when its concentration exceeds that of methotrexate at least ten times. Guidelines for the practice of high-dose methotrexate therapy and the following citrovorum factor rescue, especially for patients with retarded methotrexate elimination are presented. A new simple formula enables the exact calculation of the citrovorum factor dose for individual patients according to their methotrexate serum levels. This formula, derived from the experimental data of the bone marrow and lymphoblast culture cells, gives a rational basis for the high-dose rescue with citrovorum factor in cases with expected or manifest toxicity, and means, moreover, a safety factor for the practice of high-dose methotrexate therapy.

Biochemical control of high-dose methotrexate/Leucovorin rescue therapy.

High-dose methotrexate/Leucovorin rescue therapy is based on the assumption of differences in the transport system for folate compounds between normal and malignant proliferating cells. Thus, under normal conditions, methotrexate (MTX) and Leucovorin (citrovorum factor, CF) in low doses can enter the cells by an active transport system, whereas in some malignancies - such as osteosarcoma - these substances only penetrate through the cell membrane by passive diffusion if they are given in very high doses. Therefore, after high-dose MTX treatment, the cytotoxic effect of the folate antagonist is compensated for by rescue with Leucovorin in low doses only in the normal cell system. The consequence of this kind of treatment is a selective antitumor effect. To avoid cytotoxic side effects, this therapeutic regimen must be monitored carefully. The decrease of the ratio of 3H-deoxyuridine (dUR) beta H-thymidine (dTR) incorporation into the DNA of the cells is a good biochemical parameter for estimating the MTX effect on rapidly proliferating cell systems. Using this indicator, it was shown that the usually administered dose of Leucovorin is not sufficient for an effective rescue of the bone marrow cells as long as the MTX serum concentration is equal or higher than 10(-6) M. If in critical cases the MTX elimination is retarded, a rescue can only be achieved by Leucovorin at doses tenfold higher than the actual amount of MTX in the whole body system. The Leucovorin rescue does under such circumstances can be calculated according to the formula Leucovorin (mg) = 10 x MTX (mg/l) x 0.76 x body weight (kg).

New dose-time relationships of folate antagonists to sustain inhibition of human lymphoblasts and leukemic cells in vitro.

In this study, three methods are utilized to analyze toxicity produced by methotrexate and the lipid-soluble antifolate, 2,4-diamino-5-(3',4'-dichlorophenyl)-6-methylpyrimidine, in human lymphoblasts (WIL-2) and leukemic cells. These methods detect increasingly severe metabolic damage; inhibition of deoxyuridine incorporation into DNA, the reversibility of inhibition of deoxyuridine incorporation by supplementation with formyltetrahydrofolate as Ca2+ leucovorin, and the ability of cells to form clones in soft agarose. The critical dose and exposure time for establishing and maintaining the metabolic toxicity of methotrexate is examined in detail. It is shown that, if an initial loading dose of methotrexate is of high enough concentration or is maintained for a sufficient period to achieve greater than 98% inhibition of deoxyuridine incorporation, this inhibition can be sustained by low concentrations of 2,4-diamino-5-(3',4'-dichlorophenyl)-6-methylpyrimidine or methotrexate. Concentrations of methotrexate or 2,4-diamino-5-(3',4'-dichlorophenyl)-6-methylpyrimidine that equal or exceed 0.1 microM are sufficient for maintenance of inhibition by an initial loading dose of methotrexate but escape from inhibition that occurs if lower levels of drug are used. The possible implications of these observations for in vivo protocols are discussed.