An Em-BCL-2 transgene facilitates leukaemogenesis by ionizing radiation Deena L Gibbons 1 , Denise MacDonald 2 , Keith P McCarthy 3 , Helen J Cleary 2 , Mark Plumb 2 , Eric G Wright 2 and Mel F Greaves* ,1 1 Leukaemia Research Fund Centre, Institute of Cancer Research, Chester Beatty Laboratories, 237 Fulham Road, London SW3 6JB, UK; 2 Radiation and Genome Stability Unit, Medical Research Council, Harwell, Oxfordshire OX11 0RD, UK; 3 Department of Histology, Cheltenham General Hospital, Cheltenham, Gloucester GL53 7AN, UK Clonogenic murine B cell precursors are normally ultra- sensitive to apoptosis following genotoxic exposure in vitro but can be protected by expression of an Em-BCL-2 transgene. Such exposures are likely to be mutagenic. This in turn suggests that a level of in vivo genotoxic exposure that usually has minimal pathological con- sequences might become leukaemogenic when damaged cells fail to abort by apoptosis. If this were to be the case, then the cell type that becomes leukaemic and the chromosomal/molecular changes that occur would also be of considerable interest. We tested this possibility by exposing Em-BCL-2 and wild-type mice of diering ages to a single dose of X-irradiation of 1 ± 4 Gy. Young (*4 ± 6 weeks) transgenic mice developed leukaemia at a high rate following exposure to 2 Gy but adult mice (4 ± 6 months) did not. Exposure to 4 Gy produced leukaemia in both young and adult transgenic mice but at a higher frequency in the former. Leukaemic cell populations showed clonal rearrangements of the IGH gene but in most cases analysed had immunophenotypic features of an early B lympho-myeloid progenitor population which has not previously been recorded in radiation leukaemogenesis. Molecular cytogenetic analy- sis of leukaemic cells by banded karyotype and FISH revealed a consistent double abnormality: trisomy 15 plus an interstitial deletion of chromosome 4 that was con®rmed by LOH analysis. Keywords: apoptosis; leukaemogenesis; BCL-2; irradia- tion Introduction Genetic alterations that interfere with the execution of the apoptotic cell death pathway are important components of both the clonal evolution of cancer cells and the response of these cells in patients to therapy (Fisher, 1994; Yang and Korsmeyer, 1996). One mechanism involved is the response of cells to DNA damage. In normal cells, detection of DNA damage, involving p53 and other proteins, results in the instigation of cell cycle arrest and DNA repair or cell death (Kuerbitz et al., 1992; Paulovich et al., 1997). These alternative pathways vary in dominance and in hemopoietic cells, the cell death option is frequently adopted. This outcome can however be compromised if key regulatory proteins are dysfunctional. Lack of p53 (Clarke et al., 1993; Griths et al., 1997) or over- expression of BCL-2 protein (Griths et al., 1994a; Merino et al., 1994; Vaux et al., 1988) can, for example, rescue lymphoid cells otherwise destined for destruction following low level exposure to genotoxic chemicals, ionizing irradiation or corticosteroids. The acquisition of such blocks to cell death can have important consequences to both leukaemogenesis, response to genotoxic therapy and further malignant progression. Animals that are homozygous null for p53 are at high risk of developing leukaemia/lymphoma (Donehower et al., 1992; Purdie et al., 1994) and exposure of p53 null lymphocytes in vitro to X- irradiation results in survival of clonogenic mutants that would have died if in a wild-type, p53 competent background (Griths et al., 1997). These observations suggest that the availability of apoptotic pathways in cells can make a very signi®cant impact on the balance of transformation and cell death outcomes for genotoxic exposures. If this is correct, then mice transgenic for BCL-2 should be considerably more vulnerable to cancer induction by genotoxic exposure than wild-type controls. For example, in the fraction of clonogenic Em-BCL-2 expressing B lineage cells surviving diering doses of X-irradiation there could be cells that had incurred chromosomal/DNA damage that was compatible with continued growth but that included molecular alterations that could contribute to leukaemogenesis. In a normal genetic background, such cells would be eliminated by apoptosis. This idea has not hitherto been assessed. A positive outcome would be instructive for our understanding of mechanisms of leukaemogenesis and perhaps relevant to the susceptibility to leukaemia/lymphoma of normal adult human individuals who have clones of B cells with an endogenous rearranged and activated BCL-2 gene (Ji et al., 1995; Liu et al., 1994). Em-BCL-2 transgenic mice constitutively express human BCL-2 protein at high levels in the B cell lineage (McDonnell et al., 1989). These animals may spontaneously develop B lineage malignancies at a relatively modest rate (5 ± 15%) and with a protracted latency (12 ± 18 months) (McDonnell and Korsmeyer, 1991; Strasser et al., 1993) although a cross between Em-BCL-2 and Em-MYC transgenic mice results in a high rate of leukaemias with a brief latency (Strasser et al., 1990), indicating that single oncogenes can have highly penetrant eects in the context of expression of an anti-apoptotic gene. We challenged such mice with a single dose of X-irradiation of between 1 and 4 Gray (Gy). We also compared very young mice (4 ± 6 weeks) with adult mice. The rationale here relates to an *Correspondence: MF Greaves Received 3 November 1998; revised 18 January 1999; accepted 19 January 1999 Oncogene (1999) 18, 3870 ± 3877 ã 1999 Stockton Press All rights reserved 0950 ± 9232/99 $12.00 http://www.stockton-press.co.uk/onc