Methotrexate and gamma-tert-butyl methotrexate transport in CEM and CEM/MTX human leukemic lymphoblasts.

In a continuing investigation of determinants of their 200-fold methotrexate resistance and their collateral sensitivity to gamma-tert-butyl methotrexate, the ability of CEM/MTX cells to transport the two drugs was analyzed and compared with that of CEM cells. The Km and Vmax values for the influx of methotrexate into CEM cells did not differ significantly from those of CEM/MTX cells, and this was the case for gamma-tert-butyl methotrexate as well. Surface binding and influx rates were proportional to cell surface area, but differences in efflux rates and methotrexate uptake were too large to be explained on this basis. Neither methotrexate nor trimetrexate competed with gamma-tert-butyl methotrexate influx in CEM cells. However, both drugs perturbed the gamma-tert-butyl methotrexate steady state in CEM cells, resulting in slightly less uptake than with gamma-tert-butyl methotrexate alone. However, the major difference between the two cell types was in the methotrexate uptake plateau, which was much greater in the case of the parental cell line. A related observation was the more rapid efflux of methotrexate from CEM/MTX cells than from CEM cells. The poor uptake, the associated meager capacity to polyglutamylate methotrexate and the enhanced methotrexate efflux appear to be responsible for its decreased activity against CEM/MTX cells. Half-lives for gamma-tert-butyl methotrexate efflux were the same in both cell lines, allowing the drug to accumulate to cytotoxic levels despite its inability to form polyglutamates.

Methotrexate analogues. 33. N delta-acyl-N alpha-(4-amino-4-deoxypteroyl)-L-ornithine derivatives: synthesis and in vitro antitumor activity.

N delta-Acyl derivatives of the potent folylpolyglutamate synthetase (FPGS) inhibitor N alpha-(4-amino-4-deoxypteroyl)-L-ornithine (APA-L-Orn) were synthesized from N alpha-(4-amino-4-deoxy-N10-formylpteroyl)-L-ornithine by reaction with an N-(acyloxy)succinimide or acyl anhydride, followed by deformylation with base. The N delta-hemiphthaloyl derivative was also prepared from 4-amino-4-deoxy-N10-formylpteroic acid by reaction with persilylated N delta-phthaloyl-L-ornithine, followed by simultaneous deformylation and ring opening of the N delta-phthaloyl moiety with base. The products were potent inhibitors of purified dihydrofolate reductase (DHFR) from L1210 murine leukemia cells, with IC50's ranging from 0.027 and 0.052 microM as compared with 0.072 microM for APA-L-Orn. Several of the N delta-acyl-N10-formyl intermediates also proved to be good DHFR inhibitors. One of them, N alpha-(4-amino-4-deoxy-N10-formylpteroyl)-N delta-(4-chlorobenzoyl)-L- ornithine, had a 2-fold lower IC50 than its deformylated product, confirming that the N10-formyl group is well tolerated for DHFR binding. While N delta-acylation of APA-L-Orn did not significantly alter anti-DHFR activity, inhibition of FPGS was dramatically diminished, supporting the view that the basic NH2 on the end of the APA-L-Orn side chain is essential for the activity of this compound against FPGS. N delta-Acylation of APA-L-Orn markedly enhanced toxicity to cultured tumor cells. However, N delta-acyl derivatives also containing an N10-formyl substituent were less cytotoxic than the corresponding N10-unsubstituted analogues even though their anti-DHFR activity was the same, suggesting that N10-formylation may be unfavorable for transport. Two compounds, the N delta-benzoyl and N delta-hemiphthaloyl derivatives of APA-L-Orn, with IC50's against L1210 cells of 0.89 and 0.75 nM, respectively, were more potent than either methotrexate (MTX) or aminopterin (AMT) in this system. These compounds were also more potent than MTX against CEM human lymphoblasts and two human head and neck squamous cell carcinoma cell lines (SCC15, SCC25) in culture. Moreover, in assays against SCC15/R1 and SCC25/R1 sublines with 10-20-fold MTX resistance, the N delta-hemiphthaloyl derivative of APA-L-Orn showed potency exceeding that of MTX itself against the parental cells.(ABSTRACT TRUNCATED AT 400 WORDS)

Combination of N,N',N"-triethylenethiophosphoramide and cyclophosphamide in vitro and in vivo.

In vitro and in vivo studies with the drug combination thioTEPA and cyclophosphamide (CPA) were carried out using the MCF-7 human breast carcinoma cell line and the EMT6 mouse mammary carcinoma cell line. In vitro, survival curves were essentially linear. The EMT6 cell line was less sensitive to thioTEPA than the MCF-7 cell line, with concentrations which reduce cell survival to 10% of 440 and 140 microM, respectively. The response of both cell lines to 4-hydroperoxycyclophosphamide was similar. Simultaneous and immediate sequential treatments with these drugs produced supraadditive cell killing of both cell lines, although the magnitude of the supraadditivity was greater in the MCF-7 cell line than in the EMT6 cell line. Both of these drugs appeared to be as effective as thiol-depleting agents as is diethyl maleate. By DNA alkaline elution, there was a pattern of increasing DNA cross-linking similar to the increasing levels of cytotoxicity of this drug combination with increasing thioTEPA concentrations. In the EMT6 tumor in vivo, the maximally tolerated combination therapy (5 mg/kg x 6 thioTEPA and 100 mg/kg x 3 CPA) produced about 25 days of tumor growth delay which was not significantly different than expected for additivity of the individual drugs. The survival of EMT6 tumor cells after treatment of the animals with various single doses of thioTEPA and CPA was assayed. Tumor cell killing by thioTEPA produced a very steep, linear survival curve through 5 logs. The tumor cell survival curve for CPA out to 500 mg/kg gave linear tumor cell kill through almost 4 logs. In all cases, the combination treatment tumor cell survivals fell well within the envelope of additivity. Both of these drugs are somewhat less toxic toward bone marrow cells by the granulocyte-macrophage colony-forming unit in vitro assay method than to tumor cells. The combination treatments were subadditive or additive in bone marrow granulocyte-macrophage colony-forming unit killing. When bone marrow is the dose-limiting tissue, there is a therapeutic advantage to the use of this drug combination.

Collateral methotrexate resistance in cultured human head and neck carcinoma cells selected for resistance to cis-diamminedichloroplatinum(II).

A human head and neck squamous cell carcinoma line (SCC25) derived from a patient with no prior history of radiotherapy or chemotherapy was made resistant to cis-diamminedichloroplatinum(II) (CDDP) by continuous escalation of weekly 30-min pulses of the CDDP from 0 to 0.2 mM over 20 months and then cloned and pulsed weekly with 0.2 mM CDDP for another 20 months. This afforded a resistant subline, SCC25/CP[1], with an IC50 for CDDP 12-fold higher than that of the parental cells. The SCC25/CP[1] cells unexpectedly proved to be cross-resistant to methotrexate (MTX) (24-fold for 30-min treatment and 8-fold for continuous treatment). Resistance was associated with a modest (about 2-fold) increase in the dihydrofolate reductase (DHFR) content according to radioligand-binding assay, and in the rate of cell division. In addition there was a 4-fold decrease in the fraction of long-chain MTX polyglutamates MTX(G4-6) in the cell after 24 h exposure to either 0.2 or 2.0 microM MTX. When the SCC25/CP[1] cells were kept out of CDDP for 8-9 months and 12 months to give the sublines SCC25/CP[2] and SCC25/CP[3], respectively, MTX sensitivity to continuous exposure returned to normal. The SCC25/CP[3] cells still exhibited a slightly elevated DHFR level, but their generation time became shorter than that of the parental SCC25 line. In addition the SCC25/CP[3] cells had an initial uptake velocity (V0) for MTX that was 9-fold greater than the V0 of the SCC25 or SCC25/CP[1] cells, while its ability to form MTX(G4-6) was comparable to that of the SCC25 cells. When SCC25/CP[2] cells were rechallenged with weekly 0.2 mM CDDP pulses for 4-6 months, a MTX-resistant line, SCC25/CP[4], was produced. The SCC25/CP[4] cells retained a slightly elevated DHFR content and a high proliferation rate, but the V0 for MTX influx was intermediate between SCC25 and SCC25/CP[3] cells. The ability to form the longer-chain polyglutamates MTX(G4-6) was again impaired. Thus, MTX cross-resistance can develop in cultured head and neck carcinoma cells when CDDP is used as the selecting agent for primary resistance. MTX resistance is multifactorial, as it is when MTX itself is used as the selecting agent, and appears to involve various combinations of altered growth rate, DHFR content, MTX uptake, and ability to form noneffluxing long-chain MTX polyglutamate species. These results are potentially of clinical relevance, since CDDP and MTX are often used in combination with other drugs or with radiation to treat patients with squamous cell carcinoma of the head and neck.

Metabolism of methotrexate and gamma-tert-butyl methotrexate by human leukemic cells in culture and by hepatic aldehyde oxidase in vitro.

The cellular uptake and metabolism of methotrexate (MTX) and gamma-tert-butyl methotrexate (TBM) were compared in CEM human leukemic lymphoblasts and a highly MTX-resistant subline (CEM/MTX) in which MTX uptake is defective. The CEM/MTX cells were found previously to be as sensitive as the parent line to TBM. While MTX was polyglutamylated extensively in the CEM cells, giving abundant levels of non-effluxing conjugates, polyglutamylation in CEM/MTX cells was reduced severely, even after exposure to a high MTX concentration (100 microM) in the medium. This treatment provided free intracellular MTX in greater than 100-fold excess over the dihydrofolate reductase level. In contrast to MTX, the ester TBM was unmetabolized in either cell line. Uptake levels after incubation of CEM and CEM/MTX cells with 2 microM TBM for 24 hr were 17 and 15 pmol/mg protein respectively. Thus, TBM accumulated equally in both cells and was well retained despite the lack of polyglutamylation. These results, together with the previously observed affinity of the drug for dihydrofolate reductase, provide a plausible rationale for the comparable sensitivity of CEM and CEM/MTX cells to TBM. Experiments were also performed to determine the susceptibility of TBM to metabolic detoxification by hepatic aldehyde oxidase. Km values were 8-fold lower for TBM than for MTX in assays using an enzyme preparation from rabbit liver, and Vmax values were 8-fold higher. Neither MTX nor TBM was oxidized to its 7-hydroxy derivative in intact CEM or CEM/MTX cells. Because TBM is capable of overcoming at least one of the modalities of MTX resistance, defective polyglutamylation, and may be more efficiently detoxified than MTX by the action of hepatic aldehyde oxidase, it has the potential to be a useful agent for the treatment of MTX-resistant tumors.

Characterization of a human squamous carcinoma cell line resistant to cis-diamminedichloroplatinum(II).

We have developed a human head and neck squamous cell carcinoma cell line (SCC-25/CP) which is relatively stably resistant to cis-diamminedichloroplatinum(II) (CDDP) after repeated exposure to escalating doses of the drug. The studies reported elucidate the mechanism(s) by which the SCC-25/CP cell line is resistant to CDDP. The SCC-25/CP cell line is approximately 30-fold resistant to CDDP, approximately 10-fold resistant to carboplatin, and about 9-fold resistant to iproplatin. Using [195mPt]CDDP, we examined the levels of platinum in whole cells and cellular fractions of both the SCC-25 and SCC-25/CP cells after 1 h exposure to 100 microM drug. The SCC-25 cells took up 30 pmol of platinum/10(6) cells in 1 h; 64% of the drug was in the nucleus and 21% in the cytosol. The SCC-25/CP cells took up 7 pmol of platinum/10(6) cells; of this, 41% was in the nucleus and 33% in the cytosol. The SCC-25 cell nuclei contained 331 pmol of platinum/mg protein and the cytosol 21 pmol of platinum/mg protein, whereas the SCC-25/CP cell nuclei contained 47 pmol of platinum/mg protein and the cytosol 8.1 pmol/mg protein. The release of drug from both cell lines followed a very similar course and was most rapid over the first 6 h. There was no difference in the non-protein sulfhydryl content of the cell lines. The protein sulfhydryl content, as measured by Ellman's procedure, indicated that the SCC-25/CP cell line has approximately a 2-fold increase in protein sulfhydryl content compared to the SCC-25 cell line. The SCC-25/CP cell line is about 2-fold resistant to cadmium chloride at 50% cell kill and about 2.5-fold resistant at 1 log kill compared to the SCC-25 cell line. Glutathione transferase activity in crude cytoplasmic extracts was measured and found to be approximately 2- to 3-fold higher in the CDDP resistant cells. The isoelectric point of the glutathione transferase isozyme was 4.8 in both the sensitive and resistant cell lines, suggesting induction of the predominant isozyme present in the parent cell line. By alkaline elution there was greater cross-link formation by CDDP in the SCC-25 cell line than in the SCC-25/CP cell line at the same drug concentrations. In conclusion, the mechanism of resistance of the SCC-25/CP cell line to CDDP is multifactorial, involving plasma membrane changes, increased cytosolic binding, and decreased DNA cross-linking.

Methotrexate analogues. 28. Synthesis and biological evaluation of new gamma-monoamides of aminopterin and methotrexate.

Lipophilic gamma-monoamide derivatives of aminopterin (AMT) were synthesized in high overall yield from 4-amino-4-deoxy-N10-formylpteroic acid and gamma-N-tert-alkyl-, gamma-N-aralkyl-, or gamma-N-arylamides of alpha-benzyl L-glutamate via a modification of the mixed carboxylic-carbonic anhydride coupling method. Coupling was also accomplished with p-nitrophenyl 4-amino-4-deoxy-N10-formylpteroate. Compounds obtained in this manner included the gamma-tert-butylamide, gamma-(1-adamantylamide), gamma-benzylamide, gamma-(3,4-dichlorobenzylamide), gamma-(2,6-dichlorobenzylamide), gamma-anilide, gamma-(3,4-methylenedioxyanilide), and gamma-(3,4-dihydroxanilide) derivatives of AMT. Also prepared, from 4-amino-4-deoxy-N10-methylpteroic acid via diethyl phosphorocyanidate coupling, was the gamma-(3,4-methylenedioxyanilide) of MTX. The methylenedioxyanilides were cleaved smoothly to dihydroxyanilides with boron tris(trifluoroacetate) in trifluoroacetic acid. All the gamma-monoamides were tested as inhibitors of purified dihydrofolate reductase (DHFR) from murine L1210 leukemia cells and as inhibitors of the growth of wild-type L1210 cells and a subline (L1210/R81) with high-level resistance to MTX and AMT based mainly on a defect in drug uptake via active transport. Several compounds were also tested against human leukemic lymphoblasts (CEM cells) and a resistant subline (CEM/MTX) whose resistance is likewise based on uptake. The IC50 of the gamma-monoamides against DHFR was 1.5- to 5-fold higher than that of the parent acids, but the IC50 against cultured cells varied over a much broader range, suggesting that uptake and/or metabolism rather than DHFR binding are principal determinants of in vitro growth inhibitory activity for these compounds. gamma-N-Aryl and gamma-N-aralkyl derivatives appeared to be more potent than gamma-N-tert-alkyl derivatives. Where comparison could be made, AMT gamma-monoamides were more potent than MTX gamma-monoamides. Several of the gamma-monoamides showed potency comparable to that of the parent acid against wild-type L1210 and CEM cells; all of them were more potent than MTX against the L1210/R81 subline; and some of the AMT gamma-monoamides were also more potent than the parent acid against resistant CEM/MTX cells. As a group, however, the gamma-monoamides were considerably more active against the murine cells than against the human cells, suggesting that the former may take up the amides better or may be able to metabolize them more efficiently than the parent acids. All the gamma-monoamides were tested in vivo against L1210 leukemia in mice.(ABSTRACT TRUNCATED AT 400 WORDS)

Alkylating agents: in vitro studies of cross-resistance patterns in human cell lines.

The alkylating agents represent one of the most important classes of antitumor agents and play a major role in combination with other agents in the curative chemotherapy of selected human cancers. By repeatedly exposing cells to escalating doses of an alkylating agent, we have developed four human tumor cell lines which are relatively stably resistant to the drug with which the culture was treated. The response of these cell lines to a variety of alkylating agents was compared to the response of the parent cell lines to the same drug. The Raji/HN2 line was 7-fold resistant to nitrogen mustard and about 3-fold resistant to 4-hydroxyperoxycyclophosphamide, but it was not resistant to N,N'-bis(2-chloroethyl)-N-nitrosourea (BCNU), melphalan (MEL), busulfan, trimethyleneiminethiophosphoramide, 4-hydroperoxyifosfamide, or cisplatin [cis-diamminedichloroplatinum(II)] (CDDP). The Raji/BCNU line was 5.3-fold resistant to BCNU and 4-fold resistant to both MEL and CDDP. The Raji/CP line was 7-fold resistant to CDDP and 3-fold resistant to both nitrogen mustard and BCNU, but it was not resistant to busulfan, trimethyleneiminethiophosphoramide, or 4-hydroperoxyifosfamide. The SCC-25/CP line, which was 12-fold resistant to CDDP, was 5-fold resistant to MEL and 3-fold resistant to 4-hydroxyperoxycyclophosphamide. The SCC-25/CP line was almost 24-fold resistant to methotrexate after 30-min treatment and about 7-fold resistant to methotrexate after continuous treatment. None of the other cell lines was resistant to methotrexate. The survival of SCC-25 and SCC-25/CP cells exposed to several antineoplastic agents was examined over several logs of survival. The SCC-25/CP cells are highly resistant to CDDP; the ratio of the slopes of the survival curves (SCC-25/CP to SCC-25) of the two lines was 43. At survivals of 1%, resistance to MEL and BCNU became evident in the SCC-25/CP line. At survivals of 0.1%, resistance to mitomycin C and, to a lesser degree, to Adriamycin and vincristine was evident. It is more difficult to produce resistance to alkylating agents, even with extended selection pressure, than to other antineoplastic drugs such as antimetabolites and natural products. We found no evidence of pleiotropic resistance in any alkylating agent-resistant cell line. Our results suggest that a judicious choice of alkylating agents given in sequential or concurrent combination may be a rational treatment strategy with potential applications in the clinic.

Phenotypic heterogeneity in cultured human head and neck squamous cell carcinoma lines with low-level methotrexate resistance.

Low-level methotrexate (MTX) resistance (less than 20-fold) was induced by gradual selection pressure in four human head and neck squamous cell carcinoma lines established in culture from biopsies of patients not previously treated with MTX. Each parental and resistant line was characterized with respect to MTX uptake and polyglutamylation, dihydrofolate reductase (DHFR) content, and growth rate. Relative DHFR gene copy numbers and amounts of DHFR-related cytoplasmic messenger RNA were analyzed by plasmid complementary DNA hybridization in a dot blot assay and were correlated with the amount of gene product. The resistant lines were not cloned in order to simulate in vitro the conditions which might exist in an in vivo setting, where multiple resistant subpopulations of cells may be present in a tumor. The study was restricted to cells with low-level resistance since these are likely to be the clinically most relevant type. Of the four resistant lines characterized, one showed a severe defect in MTX uptake and polyglutamylation, another was a DHFR overproducer with only small changes in uptake and polyglutamylation, a third was likewise a DHFR overproducer but also showed lower MTX uptake, and the fourth was minimally altered except for growth rate. The diversity in resistance phenotype among these cells in vitro suggests that in vivo resistance in patients with head and neck carcinoma who are treated with MTX may similarly involve multiple mechanisms and that further therapeutic intervention using MTX or other antifolates should take this into account.

Methotrexate analogues. 25. Chemical and biological studies on the gamma-tert-butyl esters of methotrexate and aminopterin.

gamma-tert-Butylaminopterin (gamma-tBAMT), the first example of an aminopterin (AMT) gamma-monoester, was synthesized, and new routes to the known N10-methyl analogue gamma-tert-butyl methotrexate (gamma-tBMTX) were developed. The inhibitory effects of gamma-tBAMT on the activity of purified dihydrofolate reductase (DHFR) from L1210 murine leukemia cells, the growth of L1210 cells and CEM human leukemic lymphoblasts in suspension culture, and the growth of several lines of human squamous cell carcinoma of the head and neck in monolayer culture were compared with the effects of gamma-tBMTX and the parent acids AMT and methotrexate (MTX). Patterns of cross-resistance to gamma-tBAMT, gamma-tBMTX, and AMT among several MTX-resistant cell lines were examined. In vivo antitumor activities of gamma-tBAMT and gamma-tBMTX were compared in mice with L1210 leukemia. While the activity of gamma-tBAMT was very close to that of gamma-tBMTX in the DHFR inhibition assay, the AMT ester was more potent than the MTX ester against cells in culture and against L1210 leukemia in vivo. Only partial cross-resistance was shown against gamma-tBMTX and gamma-tBAMT in cultured cells that were resistant to MTX by virtue of a transport defect or a combination of defective transport and elevated DHFR activity.

Alkylating agent resistance: in vitro studies with human cell lines.

Development of in vitro resistance to HN2 (also called mustargen or mechlorethamine hydrochloride), N,N'-bis(2-chloroethyl)-N-nitrosourea (BCNU), and cisplatin [cis-diamminedichloroplatinum(II)] was achieved in two human cell lines, the Raji/Burkitt lymphoma and a squamous cell carcinoma of the tongue. A 10- to 20-fold increase in resistance relative to the parental line was achieved in 3-4 months of continuous selection pressure. At this time, further increase in selection pressure resulted in cell death, while removal of drug led to rapid loss of resistance. However, by holding selection pressure constant over 8-12 months, semistable clones ranging in resistance up to 8- to 12-fold were obtained. The half-life for resistance loss upon removal of drug was 2-3 months. In the presence of intermittent low concentrations of the alkylating agent, resistance has been maintained in excess of 9 months. With one exception, the growth kinetics of the resistant clones were slightly slower than those of the parental lines. Cross-resistance studies were performed against HN2, BCNU, cisplatin, phenylalanine mustard, and hydroperoxycyclophosphamide. There was, in general, a lack of cross-resistance. We conclude that stable resistance to alkylating agents is produced with difficulty. We propose that these semistable cloned human tumor lines represent clinically relevant models for the study of alkylating agent resistance and that the cross-resistance patterns among these cells have important therapeutic and mechanistic implications.

Development of methotrexate resistance in a human squamous cell carcinoma of the head and neck in culture.

Four methotrexate (MTX)-resistant sublines of a human squamous cell carcinoma (SCC15) were established in culture by progressive dose escalation. The biochemical basis of resistance was studied. The line with the lowest resistance (R1) had a normal dihydrofolate reductase (DHFR) content but showed decreased MTX transport and polyglutamation. Lines of intermediate resistance (R2 and R3) showed an increased DHFR content and DHFR gene copy number and a defect in MTX transport. The line with the greatest resistance (R4) showed increased DHFR content and gene copy number but nearly normal MTX transport. These results demonstrate that multiple mechanisms of MTX resistance occur in human epithelial cells in culture. We also find evidence of alterations in DHFR gene expression. The MTX-resistant cells were either not cross-resistant or only partly cross-resistant to two lipophilic MTX ester derivatives. These compounds are of potential therapeutic interest for the treatment of MTX-resistant tumors.