Transplantation of leukemic bone marrow treated with cytotoxic antileukemic antibodies and complement.

The ability of antiserum against murine L1210 leukemia to remove residual leukemia cells from murine bone marrow was investigated. Leukemic marrow was treated in vitro with antiserum and complement and used to hematologically reconstitute mice that had been irradiated with doses lethal to bone marrow. Following infusion of treated leukemic marrow, normal marrow returned without evidence of leukemia. More than 90 percent of the animals have survived for 11 months without untoward effects, suggesting that the technique may be of use in the treatment of acute leukemia in humans.

Primate model for long-term study of intraventricularly or intrathecally administered drugs and intracranial pressure.

Meaningful pharmacokinetic investigations require animal systems which approximate the human situation. This report describes a primate model in which silicone catheters are placed into the fourth ventricle and the spinal subarachnoid space and connected to subcutaneous cerebrospinal fluid without tissue damage, prod, enables spinoventricular perfusion, and permits ventricular cerebrospinal fluid sampling over extended periods in unanethetized rhesus monkeys. This animal system may provide intraventricular pressure recordings and pharmacokinetic data similar to that obtained in man.

T cell receptor alpha-, beta-, and gamma-genes in T cell and pre-B cell acute lymphoblastic leukemia.

We examined alpha-, beta-, and gamma-T cell receptor (TCR) gene activation within acute lymphoblastic leukemias (ALLs) that represent early stages of B and T cell development. We wished to determine if TCR rearrangement and expression was lineage restricted, showed any developmental hierarchy, or could identify new subsets of T cells. Rearrangement of gamma and beta TCR genes occurred early in development but in no set order, and most T-ALLs (22/26) were of sufficient maturity to have rearranged both genes. T-ALLs preferentially rearranged C gamma 2 versus the C gamma 1 complex; no preference within the beta locus was apparent. Once rearranged, the beta TCR continued to be expressed (11/13), whereas the gamma TCR was rarely expressed (3/14). The alpha TCR was expressed only in more mature T-ALLs (8/14) that usually displayed T3. The 3A-1 T cell associated antigen appeared earliest in development followed by T11 and T3. Within pre-B cell ALL a higher incidence of lineage spillover was noted for gamma TCR rearrangements (8/17) than for beta rearrangements (3/17). This also contrasts with the only occasional rearrangement of immunoglobulin (Ig) heavy chains (3/25) in T-ALL. However, in pre-B ALL the pattern of gamma TCR usage was distinct from that of T cells, with the C gamma 1 complex utilized more frequently. Almost all ALLs could be classified as pre-B or T cell in type by combining Ig and TCR genes with monoclonal antibodies recognizing surface antigens, although examples of lineage duality were noted. Unique subpopulations of cells were discovered including two genetically uncommitted ALLs that failed to rearrange either Ig or TCR loci. Moreover, two T lymphoblasts were identified that possessed the T3 molecule but failed to express alpha plus beta TCR genes. These T-ALLs may represent a fortuitous transformation of T cell subsets with alternative T3-Ti complexes.

Correlation of adenosine deaminase activity with cell surface markers in acute lymphoblastic leukemia.

Adenosine deaminase (ADA) activity has been measured in the lymphoblasts of 23 untreated patients with acute lymphoblastic leukemia and related to the presence or absence of immunologic cell surface markers. The mean ADA activity in the acute lymphoblastic leukemia population as a whole was increased fourfold over that in normal lymphocytes. 9 of the 23 patients were classified as thymus-derived (T-) cell acute lymphoblastic leukemia on the basis of erythrocyte rosette positivity; the remaining 14 patients had null-cell leukemia. The mean ADA activity (ADA U/mg protein) of T-cell lymphoblasts (102 U) was 3 times higher than the mean of null lymphoblasts (30 U). This difference is statistically significant (P less than 0.02). Measurement of ADA activity offers a biochemical method of distinguishing between immunological subtypes of lymphoblasts which may be of prognostic and therapeutic value.

Hereditary and acquired p53 gene mutations in childhood acute lymphoblastic leukemia.

The p53 gene was examined in primary lymphoblasts of 25 pediatric patients with acute lymphoblastic leukemia by the RNase protection assay and by single strand conformation polymorphism analysis in 23 of 25 cases. p53 mutations were found to occur, but at a low frequency (4 of 25). While all four mutations were identified by single strand conformation polymorphism, the comparative sensitivity of RNase protection was 50% (2 of 4). Heterozygosity was retained at mutated codons in 3 of 4 cases. One pedigree was consistent with the Li-Fraumeni syndrome, and bone marrow from both diagnosis and remission indicated a germline G to T transversion at codon 272 (valine to leucine). Although members of another family were affected with leukemia, a 2-bp deletion in exon 6 was nonhereditary. The other two nonhereditary p53 mutations included a T to G transversion at codon 270 (phenylalanine to cysteine) and a G to C transversion at codon 248 (arginine to proline). These data support the role of both hereditary and acquired p53 mutations in the pathogenesis and/or progression of some cases of childhood acute lymphoblastic leukemia.

Immunoglobulin gene rearrangement and cell surface antigen expression in acute lymphocytic leukemias of T cell and B cell precursor origins.

We have explored the relationship among immunoglobulin gene rearrangement, cytoplasmic immunoglobulin production, and cell surface antigen expression within 37 cases of acute lymphocytic leukemia. All 12 cases of the T cell type had germ-line kappa and lambda genes and 11 of 12 had germ-line heavy chain genes. In contrast, all 25 cases of the "non-T, non-B" classification, which lacked both definitive T cell markers and surface immunoglobulin, had rearranged immunoglobulin genes, indicating that they represent precursor cells already committed to the B cell lineage at the gene level. 14 had rearranged heavy chain genes, yet retained germ-line light chain genes, whereas 11 cases had both heavy and light chain gene reorganizations. All patterns of immunoglobulin gene rearrangement predicted by a model that proceeds from heavy chain gene recombination to light chain genes were observed. Despite the uniform presence of rearranged immunoglobulin genes, only five cases produced cytoplasmic mu-chain, one exceptional case produced gamma-chain, and another produced only lambda-chain. The cases of B cell precursor type that do not produce immunoglobulin may represent cells that frequently possess ineffectively rearranged immunoglobulin genes. Included in this group may be a set of cells trapped within the B cell precursor series because their ineffective rearrangements have eliminated certain gene subsegments necessary for the assemblage of an effective heavy chain gene. All seven cases of the non-T, non-B subgroup that bore HLA-DR but lacked CALLA (the common acute lymphocytic leukemia-associated antigen) represented the earliest recognizable stage of B cell precursors with rearranged heavy chain genes but germ-line light chain genes. Correlations here suggest that cells entering B cell development express HLA-DR and rearrange heavy chain genes before the expression of a B cell-associated antigen recognized by the antibody BA-1, the antigen CALLA, and any subsequent light chain gene rearrangements.

Variability in the oral bioavailability of all-trans-retinoic acid.

BACKGROUND: Orally administered all-trans-retinoic acid (all-trans-RA) can induce complete remission in a high proportion of patients with acute promyelocytic leukemia. A previous pharmacokinetic study in patients with acute promyelocytic leukemia raised the possibility that the absorption of orally administered all-trans-RA is a saturable process that would have significant clinical impact on dosing strategies. PURPOSE: This study was specifically designed to examine the saturability of all-trans-RA absorption by measuring the effect of doubling the oral dose of all-trans-RA on plasma drug concentration in patients receiving long-term oral therapy. METHODS: Six patients with solid tumors received oral doses of 10-mg gelatin capsules of all-trans-RA. Patients were studied on 2 consecutive days after they received 28 days of all-trans-RA administered as two daily 78-mg/m2 doses. The study assigned the patients to two groups. Three patients took a 156-mg/m2 dose on day 28 and a 78-mg/m2 dose on day 29; the other three patients took the lower dose on day 28 and the double dose on day 29. Blood samples for the determination of all-trans-RA plasma concentration were obtained at 30-minute intervals starting just prior to drug administration and continuing for a total of 7 hours. The plasma concentration of all-trans-RA was measured by high-performance liquid chromatography. RESULTS: Plasma concentrations following an oral dose of all-trans-RA were highly variable, with peak concentrations ranging from 0.07 to 1.2 microM for the 78-mg/m2 dose level. Doubling the dose from 78 to 156 mg/m2 increased plasma concentration in all six patients, but the increase was unpredictable and not related to dose, ranging from less than a 1.2-fold to more than a 10-fold increase. CONCLUSION: The current study does not support the hypothesis that the gastrointestinal absorption of all-trans-RA involves a saturable process but instead suggests that absorption is highly variable among patients. This wide interpatient variability suggests that pharmacokinetic drug monitoring may have an important role in the management of patients receiving all-trans-RA.

Effect of Corynebacterium parvum, methanol-extraction residue of BCG, and levamisole on macrophage random migration, chemotaxis, and pinocytosis.

Three parameters of macrophage function: random migration, chemotaxis, and pinocytosis, were studied in the guinea pig after administration of Corynebacterium parvum, methanol-extraction residue of BCG, and levamisole (LMS), a synthetic anthelmintic. Macrophage migration studies were performed with a modified Boyden chamber. Pinocytosis was assessed by the uptake of colloidal 198Au. After ip administration, each of the three immunostimulators induced an increase in macrophage chemotactic responsiveness and, to a lesser extent and duration, in random motility. Kinetic, dose-response, and time course data for the effect of each agent on macrophage movement were explored. LMS was the most effective stimulator of macrophage activation, which occurred earlier and persisted longer than it did with the other agents. Macrophages from animals receiving each of the agents showed enhanced pinocytosis. Measurement of macrophage random migration, chemotaxis, and pinocytosis appeared to provide a rapid and quantitative assessment of several parameters of macrophage function and, when studied with other immunologic parameters, may provide useful tools for the evaluation of potential immunoadjuvants.

Heterogeneity of acute lymphocytic leukemia cell surface markers as detected by monoclonal antibodies.

Monoclonal antibodies 3A1, 4F2, 5E9, OKT3, OKT6, and OKT8 were tested by flow microfluorometry for reactivity with cells from patients with T-cell or null cell acute lymphocytic leukemia (ALL). The 3A1 antibody reacted with a greater percentage of cells from 1 group of patients that could be classified as T-cell leukemia patients as compared to its reaction in the group classifiable as null cell leukemia patients. Reactivity patterns of other monoclonal antibodies with lymphocytes from patients with ALL did not clearly define groups of ALL patients. All cells from individual leukemia patients reacted differently with the seven monoclonal reagents. On the basis of patterns of reactivity, T-cell leukemias could not be classified as malignant clones of cells from the previously described stages of thymus cell maturation. The results demonstrated that the cells undergoing malignant change express a variety of combinations of antigens not found on normal lymphocytes.

Dose-dependent pharmacokinetics of all-trans-retinoic acid.

BACKGROUND: Orally administered all-trans-retinoic acid (all-trans-RA) can induce remission in a high proportion of patients with acute promyelocytic leukemia. PURPOSE: To further define the drug's pharmacokinetics, a study of intravenous all-trans-RA was performed in rhesus monkeys. METHODS: A total of nine monkeys received intravenous bolus injections of all-trans-RA. Three different doses (20, 50, and 100 mg/m2) were each tested in three monkeys. Blood samples for determination of all-trans-RA concentration were obtained prior to drug administration and at 5, 10, 15, 30, 45, 60, 75, 90, 120, 150, 180, 240, 360, and 480 minutes after drug administration. RESULTS: Plasma disappearance of all-trans-RA was characterized by three distinct phases: a brief, initial exponential decline, followed by a relative plateau in the disappearance curve (the duration of which was dose dependent), and finally a terminal exponential decay. This profile is consistent with a capacity-limited (saturable) elimination process. The first-order (terminal) half-life for all-trans-RA averaged 19 minutes, and the mean clearances were 77, 52, and 59 mL/min for the 20-, 50-, and 100-mg/m2 dose groups, respectively. The mean +/- SD Michaelis constant (Km) for the capacity-limited process was 3.2 +/- 1.9 microM. CONCLUSIONS: Peak plasma concentrations following oral administration of 45 mg/m2 all-trans-RA in humans approach the Km for the capacity-limited process; thus, the dose-dependent pharmacokinetics of all-trans-RA described here may occur within the clinically used dosage range.

Cerebrospinal fluid levels of 2'-deoxycoformycin after systemic administration in monkeys.

2'-Deoxycoformycin (DCF) is an inhibitor of the enzyme adenosine deaminase (ADA) and has shown promise as an antileukemia agent. For the assessment of the extent to which systemically administered DCF crosses into the central nervous system (CNS), rhesus monkeys were given iv boluses of DCF. Simultaneous blood and cerebrospinal fluid (CSF) samples were assayed for DCF levels at times ranging from 10 minutes to 6 hours after the drug was given. Average peak CSF drug levels of 5.5 X 10(-8) M and 3 X 10(-7) M were reached 1 1/2 - 2 hours following injections of 0.25 and 1.0 mg DCF/kg, respectively. The ratio of peak CSF to simultaneous plasma levels was 1 to 10. Data obtained from a patient who had acute lymphocytic leukemia and who was given iv DCF were comparable. Drug levels achieved within the CSF following iv administration of 0.25 mg DCF/kg are similar to those previously demonstrated to inhibit ADA. These results may be important both for understanding DCF-related CNS toxicity and for designing combination chemotherapy with DCF.

Pharmacokinetics of trimetrexate (NSC 352122) in monkeys.

The pharmacokinetics of trimetrexate was studied in Rhesus monkeys following i.v. bolus, continuous i.v. infusion, oral, and subcutaneous administration. Two methods were used to measure drug concentration in plasma, cerebrospinal fluid (CSF), and urine: the dihydrofolate reductase inhibition assay, and a reverse phase high-pressure liquid chromatography assay. The pharmacokinetic behavior of trimetrexate was characterized by triexponential plasma disappearance, elimination primarily by biotransformation, substantial plasma protein binding, poor CSF penetration, and limited oral bioavailability. Methotrexate, administered in an equimolar dose for comparison, was cleared more rapidly from plasma than was trimetrexate. Trimetrexate concentration remained above 0.1 microM 3-fold longer. In contrast to methotrexate, which is cleared almost exclusively by renal excretion, renal clearance of trimetrexate accounted for less than 5% of total clearance. A significant discrepancy was observed in plasma and urine trimetrexate concentrations measured by the two assay methods. The dihydrofolate reductase inhibition assay gave results approximately 2- to 4-fold higher in plasma. Two metabolites of trimetrexate which inhibit dihydrofolate reductase were identified in urine (one was also found in plasma) and appear to account for the different results obtained by the two assays. These metabolites would probably also interfere with the competitive protein binding assay currently being used to measure trimetrexate in ongoing phase I trials.

Modulation of thiopurine cytotoxicity in the HL-60 cell line by physiological concentrations of hypoxanthine.

Thioguanine and mercaptopurine are clinically important agents in widespread use for the treatment of acute leukemia. In this study the effect of low, physiological concentrations of hypoxanthine on thiopurine cytotoxicity was examined in the HL-60 human acute promyelocytic leukemia cell line. Initially the effect of cell concentration on medium hypoxanthine levels was investigated. When 10(5) to 10(6) cells/ml were used, medium hypoxanthine concentrations fell to undetectable (less than 0.1 microM) levels by 24 h, due to cell utilization. However, when the cell density was reduced to 10(3) cells/ml, it was possible to maintain a medium hypoxanthine level as low as 0.5 microM for 24 h without significant hypoxanthine depletion. This allowed for the investigation of the extent to which physiological concentrations of hypoxanthine (1.0 to 10 microM) could modulate thiopurine cytotoxicity. HL-60 cells were incubated in medium containing from 1.0 to 100 microM hypoxanthine concentrations and varying levels of thioguanine or mercaptopurine for 24 h. Cells were then washed and cloned in soft agar. Physiological concentrations of hypoxanthine (1.0 to 10 microM) provided significant protection to HL-60 cells from the cytotoxic effects of both thioguanine and mercaptopurine. An increase in medium hypoxanthine level from 1.0 to 3.0 microM resulted in a 3-fold increase in the thiopurine concentration required to kill 50% of cells. There was a linear relationship between the thiopurine concentration required to reduce clonogenic survival by 50% and medium hypoxanthine level over the hypoxanthine concentrations studied. Although thioguanine was 200-fold more potent than mercaptopurine, and an analogue of guanine rather than hypoxanthine, there was a similar degree of modulation of the cytotoxicity of both agents by hypoxanthine. These results indicate that low, physiological hypoxanthine concentrations can significantly modulate thiopurine cytotoxicity and suggest that endogenous hypoxanthine pools may have an important effect on the clinical activity of thioguanine and mercaptopurine.

Acceleration of myeloid recovery from cyclophosphamide-induced leukopenia by pretreatment with Bacillus Calmette-Guérin.

Treatment of C57BL/6 mice with Bacillus Calmette-Guérin i.p. 8 days prior to the induction of leukopenia by cyclophosphamide (300 mg/kg i.p.) significantly (p less than 0.002) increased peripheral granulocyte counts on each day during the recovery from leukopenia. The recovery of lymphocyte counts was unaffected by Bacillus Calmette-Guérin treatment. Further experiments indicated that the accelerated granulocyte recovery was the result of an earlier initiation of the recovery process rather than of the release of stored granulocytes. Bacillus Calmette-Guérin may have clinical value as a stimulator of myelopoiesis in patients rendered leukopenic by antineoplastic chemotherapy.

Purine pathway enzyme abnormalities in acute lymphoblastic leukemia.

The status of three purine pathway enzymes, adenosine deaminase, 5'-nucleotidase, and purine nucleoside phosphorylase, was evaluated in the leukemic cells of patients with acute lymphoblastic leukemia and correlated with routine immunological cell surface markers. A distinct pattern of enzyme activity was noted in T-lymphoblasts which have significantly higher adenosine deaminase activity (p less than 0.02) and lower 5'-nucleotidase (p less than 0.001) and purine nucleoside phosphorylase (p less than 0.01) activities than do non-T, non-B lymphoblasts. This enzyme pattern is similar to that observed in normal human thymocytes but is not shared by the mature, normal T-lymphocytes of peripheral blood, suggesting that it may reflect the differentiation status of malignant T-lymphoblasts. These findings, which confirm the biochemical heterogeneity of acute lymphoblastic leukemia, may provide an avenue for selective chemotherapy of this disease.

Bacillus Calmette-Guérin enhancement of colony-stimulating activity and myeloid colony formation following administration of cyclophosphamide.

Pretreatment of mice with Bacillus Calmette-Guérin (BCG) enhances recovery of the peripheral granulocyte count from cyclophosphamide (CTX)-induced leukopenia. In the present study, in vitro bone marrow culture was used to assess the effect of BCG pretreatment on serum levels of myeloid colony-stimulating activity and on regeneration of bone marrow myeloid colony-forming units (CFU-c) following CTX. C57BL/6 mice received either BCG or 0.9% NaCl solution i.p. 8 days prior to the administration of CTX (300 mg/kg i.p.). Following CTX, peak serum levels of myeloid colony-stimulating activity were strikingly higher in the BCG-pretreated mice [1320 +/- 30 (S.E.) units] than in the 0.9% NaCl solution-pretreated mice (764 +/- 37 units). BCG pretreatment also resulted in higher numbers of marrow CFU-c during recovery from CTX (e.g., 43.9 +/- 1.8 versus 20.0 +/- 0.9 myeloid colonies/10(5) marrow cells, Day 3 post-CTX). However, neither the rate of decline nor the absolute nadir of CFU-c, nor CFU-c cell cycle characteristics were affected by the pretreatment. The initial effect of i.p. BCG pretreatment on recovery from CTX-induced granulocytopenia is an augmentation of serum myeloid colony-stimulating activity which precedes the enhanced regeneration of CFU-c and the accelerated recovery of the peripheral granulocyte count.

Biochemical parameters of mercaptopurine activity in patients with acute lymphoblastic leukemia.

Mercaptopurine (MP) is a purine antimetabolite widely used for remission maintenance in the therapy of acute lymphoblastic leukemia. In order to study the biochemical parameters affecting MP activity, leukemic cells were obtained from ten patients with acute lymphoblastic leukemia at the time of diagnosis and from the same patients at the time of their initial marrow relapse. Hypoxanthine phosphoribosyltransferase (HPRT), the enzyme that converts MP to its active, nucleotide metabolite, thioinosine monophosphate; alkaline phosphatase, the primary catabolic enzyme of thioinosine monophosphate; and 5-phosphoribosyl-1-pyrophosphate (PRPP), the cellular ribose-phosphate donor essential for MP activation, were all measured within the patients' leukemic cells. There was marked interpatient variability in the three biochemical parameters studied with a greater than 10-fold range in alkaline phosphatase activity and an approximately 100-fold range in HPRT activity and PRPP levels. Four patients developed changes in biochemical parameters that influence MP activity at the time of relapse. In three of the four patients, alterations in more than one of these three biochemical parameters were noted. Three of four patients had a greater than 50% decrease in intracellular HPRT activity, four of four had a greater than 50% decrease in intracellular PRPP, and two of four had a greater than 9-fold increase in intracellular alkaline phosphatase activity at relapse. Two of four patients demonstrated changes in all three parameters at relapse in the directions that could have resulted in decreased MP sensitivity (i.e., decreased HPRT, decreased PRPP, and increased alkaline phosphatase). There was no correlation between pretreatment values of HPRT, PRPP, and alkaline phosphatase and remission duration. These results indicate that: (a) there is marked variation in HPRT, PRPP, and alkaline phosphatase in patients with acute lymphoblastic leukemia and b) following MP-containing maintenance chemotherapy, some patients develop biochemical changes that may result in decreased sensitivity to MP.

Pharmacokinetics of intraventricular and intravenous N,N',N''-triethylenethiophosphoramide (thiotepa) in rhesus monkeys and humans.

The cerebrospinal fluid (CSF) and plasma pharmacokinetics of N,N',N"-triethylenethiophosphoramide (thiotepa), an alkylating agent used for treatment of carcinomatous meningitis, were determined in rhesus monkeys in order to assess the relative advantage of intraventricular versus systemic administration of the drug. Following an i.v. thiotepa dose of 0.9 mg/kg (11 mg/sq m), peak plasma levels of parent drug reached approximately 1 microgram/ml. Thiotepa was rapidly equilibrated with lumbar and ventricular CSF. Systemic, lumbar, and ventricular exposure to the drug, measured as area under the curve (AUC), were similar in all cases. After a 1-mg intraventricular dose of thiotepa, peak ventricular levels were greater than 100 micrograms/ml. However, peak levels in the lumbar CSF at 1 h after intraventricular administration were less than 10 micrograms/ml. The AUC for ventricular CSF was nearly 100-fold greater for the intraventricular route than for the i.v. route; however, the AUC for lumbar CSF following intraventricular delivery was only 5% of the AUC for ventricular CSF. N,N',N''-Triethylenephosphoramide, an active metabolite of thiotepa observed in all fluids, appeared to have a much slower total body clearance than thiotepa. Comparison of the data obtained from monkey experiments with data from a patient with meningeal disease supports the use of the monkey as a model for intraventricular pharmacokinetics. The data presented indicate that there is no relative advantage to intraventricular administration of thiotepa at the doses currently used in clinical trials.

Preclinical pharmacology of arabinosyl-5-azacytidine in nonhuman primates.

The plasma and cerebrospinal fluid (CSF) pharmacokinetics of arabinosyl-5-azacytidine (AAC) were studied in rhesus monkeys following a 15-min, 1-h, or 12-h i.v. infusion of 200 mg/kg. No clinically significant toxicity was observed with these schedules. The plasma elimination of AAC is rapid and characterized by a triphasic decay with t1/2 alpha = 3.6-5.4 min, t1/2 beta = 18-24 min, and t1/2 gamma = 94-144 min for the above infusion schedules. The CSF penetration of AAC as measured by the CSF:plasma Css ratio for the 12-h infusion was 0.15. The stability of AAC in pooled plasma, phosphate buffered saline, and RPMI 1640 culture media at 37 degrees C was compared with the terminal half-life of AAC observed in vivo. The shorter in vitro AAC half-life in plasma with or without tetrahydrouridine versus that in phosphate buffered saline suggests that the terminal half-life of AAC in vivo is most likely a result of enhanced nucleophilic attack and hydrolytic degradation of the unstable triazine ring in plasma. A triexponential equation modeling the disappearance of AAC was constructed from the in vivo experimental data. Use of this equation in computer-aided simulations of current Phase I doses and schedules of AAC correctly predicts the human plasma concentrations which have been observed. The preclinical pharmacokinetic data provided here may be useful in helping to develop rational human studies with specific concentration x time goals.

Prevention of methotrexate cytotoxicity by asparaginase inhibition of methotrexate polyglutamate formation.

Escherichia coli asparaginase (Asnase) pretreatment of Asnase-sensitive L5178Y cells in vitro is thought to antagonize methotrexate (MTX) cytotoxicity through nonspecific inhibition of protein synthesis and MTX uptake. We have reexamined the mechanism of this interaction in view of recent data demonstrating the importance of MTX metabolism to polyglutamate derivatives (MTXPGs) in the cytotoxic effects of the antifolate. After a 3-hr exposure to 0.5 microM MTX, 67% of intracellular drug was in the form of MTXPGs containing a total of 2 to 5 glutamyl residues (MTX-Glu2-5), and cloning efficiency in drug-free medium was only 7% of untreated control. After a 3-hr pretreatment with E. coli Asnase (0.1 unit/ml), [3H]thymidine incorporation dropped by 29%, MTXPG formation during subsequent MTX exposure decreased by more than one-half (MTX-Glu2 unchanged; MTX-Glu3 and 4 decreased to 51.7 and 18.5% of levels achieved in cells not pretreated with Asnase; no MTX-Glu5 formed), and cloning efficiency increased to 71% of untreated control. This effect was not due to decreased MTX uptake into L5178Y cells or to decreased intracellular free L-glutamate or L-glutamine levels. A 3-hr exposure of L5178Y cells to media lacking L-isoleucine, an essential amino acid for cell growth, prior to MTX exposure inhibited [3H]thymidine incorporation by 37%, decreased subsequent MTXPG formation by 62%, and increased subsequent cloning in drug-free medium to control levels. Decreased MTXPG formation was responsible for the prevention of MTX cytotoxicity seen after both pretreatments. Unmetabolized MTX rapidly left L5178Y cells after removal of extracellular MTX. Consequently, lower levels of unbound intracellular drug, a prerequisite of drug activity, were maintained in pretreated than in control cells after passage in drug-free medium. Asnase pretreatment protects L5178Y cells from the cytotoxic effects of MTX, possibly through inhibition of cell growth which nonspecifically decreases MTXPG formation.