Mutational spectra vary with exposure conditions: benzo[a]pyrene in human cells

The AHH-1 human lymphoblastoid line was exposed to benzo[a]pyrene under markedly different conditions: a single toxic exposure of 30 microM for 28 h, a nontoxic exposure of 0.5 microM for 6 days and an exposure approximating estimates of BP concentration in the human lung of 20 nM for 20 days. Duplicate cultures containing 2 x 10(9) cells each were used to assure the statistical quality of the mutational spectra. Point mutational hotspots were observed in bp 215 to 318 of the third exon of the hprt gene after mutants were selected en masse with 6-thioguanine, using a combination of denaturing gradient gel electrophoresis and high fidelity polymerase chain reaction. The spectra were highly reproducible in replicate experiments but varied dramatically among treatment conditions. These data demonstrate that mutational spectra were critically dependent upon conditions of exposure. The results significantly extend prior reports on this subject and clarify an important issue for the use of mutational spectra obtained in vitro to create hypotheses about what spectra may be expected in humans in vivo. We conclude that commonly used protocols of short-term exposure to mutagenic chemicals at high concentrations should not be used to define such expectations. Rather, the more difficult protocols of long-term and low-concentration mutation studies are justified as conditions necessary, although perhaps not sufficient, to approximate human in vivo mutational pathways.