Recent observations that Notch1 mutations are present in more than 50% of human T-ALL patient samples indicate that Notch signals have a central role in the pathogenesis of T-ALL (Weng et al., 2004, Science, 306, 269–271). Notch1 mutations are of two types: i) mis-sense mutations in an extracellular heterodimerization (HD) domain; and ii) frameshift or stop codon mutations that result in deletion of an intracellular PEST destruction box. The likely effect(s) of these mutations is to i) cause heightened proteolytic activation of Notch1, an event that requires the enzyme gamma-secretase; and to ii) increase the duration of Notch1 signaling, respectively. To investigate the contribution of the HD and PEST domain mutations to the pathogenesis of human T-ALL, Notch1 alleles with common HD mutations were introduced retrovirally into hematopoietic progenitor cells, which were then transplanted into lethally irradiated mice. These mice developed transiently circulating abnormal CD4+CD8+ T cells, but failed to develop leukemia after 9 months of monitoring. Similar findings occurred when Notch1 alleles bearing only PEST mutations were introduced into mice, as previously noted by others (

Feldman et al.,
2000
,
Blood
,
96
,
1906
–1913
). However, when HD mutations were combined with a PEST mutation in cis, the mice developed persistent circulating CD4+CD8+ T cells, leukocytosis, and eventually T-ALL (median time, 18 weeks post-transplantation). These tumors infiltrated lymphoid organs, marrow, and viscera, were monoclonal or oligoclonal, and had intact proviral integrations. Cell lines derived from these tumors are susceptible to gamma-secretase inhibitor-mediated cell death, which indicates a continued reliance on Notch signaling for growth and survival. These results show the utility of the murine BMT assay to score Notch1 mutants found in human T-ALL with respect to leukemogenicity, and suggest that the level of signal that is necessary to induce and sustain Notch1-associated T-ALL is relatively high. It will be of interest to determine if HD domain mutations generally require PEST deletions in cis to induce T-ALL, or if weak and strong mutations of varying leukemogenic potential exist.

Author notes

Corresponding author

Sign in via your Institution