"Sub-threshold neoplastic states" created in transgenic mice.

Several possible scenarios are already suggested by the first series of studies in which a high risk situation with respect to malignant diseases was artificially created by introducing oncogenes in the mouse germ line. Only in exceptional cases was the expression of the transgene followed in an obligatory way by cancerous proliferation. The usual situation appears now to be that somatic cells expressing the transgene have only a higher probability than normal cells of undergoing a subsequent transforming event. Apparent one-hit cases, which would constitute exceptions to this rule (insulin-SV40 and elastase-ras constructs) might correspond to combinations of host cell and cis-regulatory regions leading to very high efficiency of expression. For these two oncogenes, a dose-dependence effect was previously indicated by in vitro transformation studies, and observations on other transgenic strains also suggested that tumour development was related, possibly in a causal way, with increased levels of expression. The latter could depend on epigenetic (the host cell reaching a defined physiological or differentiation state), or genetic determinations (mutations in either cis or trans regulatory elements). More data on the accumulation of the protein product and on histological characterization of the tumours will be needed, however, in order to firmly conclude (i) that the increased expression of the transgene is immediately followed by cell proliferation, and (ii) that the observed proliferation is already a malignant one. The situations described by the "myc mice" are clearly distinct from the previous ones, because they strongly suggest the occurrence of a second, independent genetic event. Data on tumour induction by ectopic expression of the MTV transgene (Leder et al., 1986) suggest a situation analogous to the classical cooperation of oncogenes in REF cells. Identification of the second oncogene(s) by the proper transfection experiments is expected with interest. A distinct scenario of oncogenic cooperation is provided by the recent findings of Langdon et al. (1986): the first event may correspond to the block of a differentiation step resulting in the accumulation of an actively proliferating population of precursor cells, among which mutations would lead to the emergence of tumoral clones. Cooperativity between induction of self-renewal and block of differentiation was extensively studied in vitro in hematopoietic cells expressing retroviral oncogenes (Kahn et al., 1986 for review), and it may be a characteristic feature of cells whose differentiation involves complex series of proliferating precursor cells.