Over the last decade there has been considerable debate about how Bcl-2 pro-survival proteins regulate apoptosis. Several different models have been proposed, suggesting that the pro-survival proteins inhibit apoptosis through sequestration of “activator” BH3-only proteins (Mode 1 / Direct activation model), and through interactions with Bax/Bak that prevent them from oligomerising on the outer mitochondrial membrane (Mode 2 / Indirect activation model). The evidence for both scenarios, based on in vitro studies, is very strong, however there are very few studies that have examined precisely how pro-survival proteins inhibit apoptosis in vivo.

Here we describe a novel mouse model that provides a complete physiological readout of the interaction between Bcl-x_L and Bak. These mice carry a germline point mutation in the BH3 domain of Bak which biochemical and X-ray crystallographic studies (using both mouse and human proteins) demonstrate impaired binding to Bcl-x_L, but not other pro-survival Bcl-2 family members. Mice carrying this mutation exhibit severe thrombocytopenia and significantly reduced thymocyte counts compared to wild-type animals. Other cell types such as fibroblasts from the mutant mice are also more sensitive to apoptotic stimuli and chemotherapeutics compared to those from wild-type animals. Hence, this model provides a powerful tool to study the mechanisms of cell death regulation in platelets and T-cells that are still not fully understood. Moreover, these mice extend previous genetic studies on apoptosis regulation as they provide precise mechanistic information on how the phenotypes are generated – such information cannot readily be discerned from more conventional genetic approaches such as pro-survival or pro-apoptotic protein knock-outs. The data we have to date demonstrates that inhibition of Bak by Bcl-x_L (i.e Mode 2 / Indirect activation) is critical for survival of certain cell types in vivo.