The ability of phorbol 12-myristate 13-acetate (PMA) at 0.02 jpg/ml to induce neoplastic transformation in NIH 3T3 cells is highly dependent on the culture conditions. Optimal transformation, indicated by the saturation density and extent of focus formation in transferred cultures raised under standard conditions, was observed when the ornal cells were grown in 2% calf serum (CS) and exposed continually to PMA for at least 4 weeks before transfer into the assay. Transormation of stationary cultures in 10% CS occurred later and to a lesser degree than in 2% CS. The same cells subjected to thrice-weekly transfer in 2% or 10% CS at low cell density so that they were in a constant state of exponential growth exhibited no evidence of transformation in response to PMA. This strong condition-dependence of PMAenhanced transformation is indicative of a selection process similar to that described for spontaneous transformation. In both cases, transformation is promoted by inhibiting multiplication and prevented by zing multiplication. Therefore, it has the earmarks of an epigenetic rather than a mutational process and requires phenotypic rather than genotypic variation to supply the states for selection. The concept of "progressive state selection," originally proposed to account for spontaneous transformation, can also account for PMAenhnced transformation.Phorbol esters induce a wide spectrum of short-term responses in various cell culture systems. Such responses include changes in morphology, enzyme activities, ion transport, nutrient transport, and effects on saturation density, among others (for a review of the effects of promoters in cell culture, see ref. 1). Strict correlations between these shortterm responses and the classic long-term response, tumor promotion in the skin ofmice, have been difficult to establish, particularly in view of the extremely divergent types of responses observed in cell culture. Both suppression and enhancement by phorbol 12-myristate 13-acetate (PMA) are observed in several systems depending, at least to some degree, on the amount of time elapsed between treatments with the initiating carcinogen and PMA (2, 3). Changes in the nature of the cellular response to PMA invite speculation about the role of the metabolic state in mediating these responses.NIH 3T3 cells first came to wide attention as transformable recipients of ras oncogene sequences (4, 5). Studies conducted over the last several years have revealed, however, that these cells undergo spontaneous transformation in the absence of exogenous oncogenes in a manner that is dependent on environmental conditions (6,7). For example, exposure to growth constraints, such as those that occur at confluence or in low concentrations of serum, will induce transformation (8). High concentrations of glutamine, an important energy source for cultured cells, are required to obtain maximal transformation (9). Maintenance of transformed cells at low densities in high serum concentrations results in the gradual reversal of transformation a...