Carcinogen-specific induction of genetic instability

Abstract

It has been proposed recently that the type of genetic instability in cancer cells reflects the selection pressures exerted by specific carcinogens. We have tested this hypothesis by treating immortal, genetically stable human cells with representative carcinogens. We found that cells resistant to the bulky-adduct-forming agent 2-amino-1-methyl-6-phenylimidazo[4,5-b]pyridine (PhIP) exhibited a chromosomal instability (CIN), whereas cells resistant to the methylating agent N-methyl-N'-nitro-N-nitrosoguanidine (MNNG) exhibited a microsatellite instability (MIN) associated with mismatch repair defects. Conversely, we found that cells purposely made into CIN cells are resistant to PhIP, whereas MIN cells are resistant to MNNG. These data demonstrate that exposure to specific carcinogens can indeed select for tumor cells with distinct forms of genetic instability and vice versa.

Figures

Figure 1

Resistance to MNNG and PhIP. (a) Approximately 105 cells of control H3 clones (not previously exposed to carcinogens) or clones derived after exposure to carcinogens (see Table 1 for enumeration) were exposed to either 50 μM PhIP or 5 μM MNNG as described in Materials and Methods, and cells were stained with crystal violet 14 days later. Untreated cells served as a plating control (Untreated). (b) BUB-DLD1 cells, inducibly expressing a dominant mutant hBUB1 gene were exposed to PhIP or MNNG and stained as in a. As DLD1 cells are MMR deficient, they are resistant to MNNG, irrespective of induction.

Figure 2

FISH analysis of chromosomal instability in clones surviving carcinogen exposure. A chromosome 12-specific centromeric probe was labeled with FITC (yellow), and a contig of three bacterial artificial chromosome clones mapping to the distal part of chromosome 12q was labeled with tetramethylrhodamine B isothiocyanate (red). Cells were counterstained with 4′,6-diamidino-2-phenylindole (blue). Nuclei of control cells (a) and MNNG-resistant clones (b and c) exhibited two yellow and two red signals in virtually every nucleus (a and c) and metaphase spread (b). In contrast, cells of PhIP-resistant clones (d and e) often exhibited more or less than two copies of chromosome 12.

Figure 3

Expression of hMLH1 and hMSH2 proteins in clones surviving carcinogen exposure. Western blotting was performed with anti-hMLH1 and anti-hMSH2 antibodies. H3 cells expressed full-length forms of both hMLH1 and hMSH2, as did every PhIP-resistant clone (a). In contrast, each MNNG-resistant clone expressed either a full-length hMSH2 or a full-length hMLH1 protein, but not both, demonstrating a highly specific mechanism for loss of MMR gene activity in each clone (b). The asterisk (*) indicates a nonspecific band crossreacting with the anti-hMLH1 antibody.

Figure 4

Inducible expression of a mutant hBUB1 allele confers CIN. (a) Sequence of hBUB1 transcripts assessed by reverse transcription–PCR analysis of RNA from BUB-DLD1 cells. The exogenous mutant hBUB1 gene contained a C-to-A transversion at codon 492 (marked by *), resulting in a substitution of tyrosine for serine. Before induction, there was no mutant hBUB1 transcript detectable, whereas after induction, the level of mutant hBUB1 expression was similar to that of the endogenous wild-type hBUB1 gene. (b) Fluorescence microscopy showed no detectable expression of the coexpressed GFP gene before induction, but (c) uniform expression of GFP after induction. (e) Cells expressing mutant hBUB1 after induction were chromosomally unstable, as indicated by an abnormal number of FISH signals in a high fraction of cells, whereas uninduced cells (d) were stable. The red and yellow dots represent centromeric probes specific for chromosomes 7 and 12, respectively.