Inheritance of acquired changes in growth capacity of spontaneously transformed BALB/3T3 cells propagated in mice and in culture.

Five subclones were derived from a spontaneously transformed BALB/3T3 clone soon after its isolation. Despite their common clonal origin, the subclones were different from each other in appearance, colony-forming efficiency in agar (CFEag), and rate of tumor formation in mice. A comparison of growth properties during repeated passages in culture was made between the cells derived from the tumors and the cells used to initiate the tumors. In most cases, the tumor-derived cells had a much lower CFEag than did their parental in vitro-propagated cells, and the CFEag was restored slowly to the original level or remained at a reduced level during the period of study. In a few cases, the tumor-derived cells had almost as high a CFEag as their parental cells or were quickly restored to this level during cultivation. It was shown by karyotypic and clonal analysis that the reduced CFEag of the tumor-derived cells arose from a change in the transformed cells; i.e., it was not due to the presence of normal host cells in the explanted tumor. Clones of the tumor-derived cells tended to show the same patterns of change in CFEag as the uncloned tumor cell populations, but there were cases of individual variation in pattern among the tumor cell clones. Tumor cells with greatly reduced CFEag also grew a little more slowly on plastic than did the parental nontumor cells, and their growth rate tended to increase along with CFEag in long-term culture. Clonal analysis of one of the five original subclones failed to reveal cells which had the CFEag properties of its progeny tumor cells. This suggests that the reduction of CFEag during tumor formation arose by adaptation rather than selection of preexisting variants. A similar conclusion was drawn about the restoration of CFEag during cultivation of the tumor-derived cells. Although the decrease in CFEag which accompanied tumor formation varied in magnitude and stability, some tumor cell populations retained their reduced capacity through months of weekly passaging in culture, involving up to 100 cell divisions. The results are therefore consistent with the heretical notion of inheritance of acquired characteristics. In addition, the wide variation of in vitro growth capacity among tumors initiated by different subclones, and even among tumors initiated by the same subclone, raises the possibility that the complete chain of causality underlying the variation is intrinsically indeterminate.