At least three categories of genes are envisaged to be involved in the natural history of B-CLL. First, the genes that are responsible for the transforming event(s) in the (presently unknown) target cell; second, the gene(s) that help the progressive accumulation of malignant cells and finally the gene(s) that cause the progression toward a more aggressive lymphoma. The possibility that the clonal expansion of B-CLL is due to a prolonged life-span of monoclonal B cells rather than to an acceleration of their proliferative activity may now be reinterpreted by taking into account some recent findings on the expression of Bcl-2 gene in B-CLL cells. The Bcl-2 gene product regulates programmed cell death and a number of experiments suggest that Bcl-2 is involved in the selection and maintenance of long-lived memory B cells rescuing them from apoptotic death and leading to their accumulation in the GO phase of the cell cycle. Variant chromosomal translocations have been detected in a small fraction (5-10%) of B-CLL, involving Bcl-2 and the Ig light chain gene. Despite the low percentage of Bcl-2 rearrangements the expression of mRNA and protein is appreciable in most samples of fresh B-CLL cells in an amount comparable to that observed in Karpas 422 cells, which contain a t(14;lS). Southern blot analysis of 16 B-CLL cases with high expression of Bcl-2 failed to reveal both gene amplification and rearrangement, thereby suggesting that the level of Bcl-2 expression in B-CLL does not reflect a rearrangement of Bcl-2 gene, which appears to be a rare event in CLL cells. In addition, the half life of Bcl-2 mRNA is not prolonged in CLL cells. The expression of Bcl-2 mRNA can be modulated in B-CLL cells by using B cell activators and cytokines. In the transition from a resting to a cycling phase, Bcl-2 expression is downregulated indicating that proliferation and Bcl-2 expression are inversely related in malignant B-CLL cells from peripheral blood as observed for normal B lymphocytes derived from secondary follicles. By culturing B-CLL cells in the presence of an antisense phosphorothioate oligodeoxynucleotide complementary to the translation-initiation site of Bcl-2 gene, DNA fragmentation, characteristic of apoptosis, can be seen in a proportion of cases. These data lead to the plausible hypothesis that the high Bcl-2 expression in B-CLL cells might confer on them a survival advantage favouring their relentless accumulation, perhaps by inhibition of apoptosis.