Potentiation of carcinogen-induced methotrexate resistance and dihydrofolate reductase gene amplification by inhibitors of poly(adenosine diphosphate-ribose) polymerase

Poly(ADP-ribosyl)ation of nuclear proteins is an immediate response of most eukaryotic cells to DNA strand breaks, as induced by carcinogen treatment. DNA amplification, on the other hand, can be induced in cell culture systems by chemical or physical carcinogens, too, reaching peak levels a few days after induction treatment. We have previously shown that 3-aminobenzamide, an inhibitor of poly(ADP-ribosyl)ation, potentiates carcinogen-induced simian virus 40 DNA amplification in hamster cells which served as a short-term model system (Bürkle et al., Cancer Res., 47: 3632-3636, 1987). Here we report that those results can be extended to the development of methotrexate (MTX) resistance associated with dihydrofolate reductase (DHFR) gene amplification in a different hamster cell line. (a) Treatment with the alkylating carcinogen N-methyl-N'-nitro-N-nitrosoguanidine (MNNG) 3 days before selection with 350 nM MTX induced the MTX resistance frequency by 17- to 100-fold, as expected. Addition of 3-aminobenzamide (0.1 to 1 mM) before MNNG treatment further potentiated the frequency of MTX resistance by up to 5-fold in a dose-dependent manner, parallel to a potentiation of cytotoxicity. MTX resistance frequency was potentiated not only relative to the decrease in cell survival but also in absolute terms. The same potentiation occurred after cotreatment with benzamide (1 mM), another poly(ADP-ribosyl)ation inhibitor, under conditions which precluded direct drug interactions. Benzoic acid, a noninhibitory analogue, had no effect on the MNNG-induced MTX resistance frequency. (b) Neither 3-aminobenzamide, nor benzamide, nor benzoic acid at 1 mM, respectively, had any effect on the spontaneous frequency of MTX resistance. (c) Individual MTX-resistant colonies were expanded to determine their DHFR gene copy number. The relative frequency of DHFR gene amplification was similar (14% versus 22%) whether clones were derived from cultures induced with MNNG alone or MNNG in the presence of 1 mM 3-aminobenzamide. We conclude that poly(ADP-ribosyl)ation should act as a negative regulatory factor in the induction of DNA amplification, since inhibition of poly(ADP-ribose) polymerase potentiates both MNNG-induced simian virus 40 DNA amplification, as shown previously, and MNNG-induced MTX resistance associated with DHFR gene amplification, as shown in this paper.