HPRT MUTANTS IN A TRANSPLANTABLE MURINE TUMOR ARISE MORE FREQUENTLY IN-VIVO THAN IN-VITRO

A model system was developed to allow investigation of the frequency a t which clastogenic and/or mutagenic events occur in situ in a transpl antable murine fibrosarcoma tumour (MC1A-C1) compared with in vitro cu lture. The marker selected for detecting these events was the X-linked hprt (hypoxanthine-guanine phosphoribosyltransferase) gene. We found that the hprt gene in MC1A-C1 was not suitable for this purpose, most likely because multiple active copies were present. To circumvent the problem, HPRT(-)[6-thioguanine (6-TG)-resistant] clones were isolated by inactivating all hprt genes with methylnitrosourea. Spontaneous rev ertants to hypoxanthine/aminopterin/thymidine resistance (HAT(R)) were isolated and found to be approximately 1000 times more sensitive than the parental tumour to induction of 6-TG(R) mutants by cobalt-60 gamm a-rays. This sensitivity is expected for a heterozygous marker; these revertants may therefore possess only one functional hprt locus but tw o or more active X chromosomes. A clone with a stable hprt gene was id entified and a neo gene was introduced. The resulting cell line (MN-11 ) could be grown as a subcutaneous tumour in syngeneic C57BL/6 animals . The frequency of mutations arising in vivo in the marker hprt gene c ould be estimated by culturing explanted tumour cells in the presence of 6-TG, using G418 selection to distinguish tumour from host cells. T he frequency of mutants in MN-II cells grown as tumours was found to b e 3.4-fold higher than in tissue culture for an equivalent period of t ime. These data provide the first direct evidence for the existence of mutagenic factors in a tumour environment that might contribute to tu mour progression.