Expression Ratios of the Anti-Apoptotic BCL2 Family Members Dictate the Selective Addiction of KSHV-Transformed Primary Effusion Lymphoma Cell Lines to MCL1

Abstract

Kaposis sarcoma-associated herpesvirus (KSHV) causes several malignancies in people living with HIV, including primary effusion lymphoma (PEL). PEL cell lines exhibit oncogene addictions to both viral and cellular genes. Using CRISPR screens, we previously identified cellular oncogene addictions in PEL cell lines, including MCL1. MCL1 is a member of the BCL2 family, which functions to prevent intrinsic apoptosis and has been implicated in several cancers. Despite the overlapping functions of the BCL2 family members, PEL cells are only dependent on MCL1 suggesting that MCL1 may have non-redundant functions. To investigate why PEL cells exhibit selective addiction to MCL1, we inactivated the intrinsic apoptosis pathway by engineering BAX/BAK1 double knockout cells. In this context, PEL cells become resistant to MCL1 knockdown or MCL1 inactivation by the MCL1 inhibitor S63845, indicating that the main function of MCL1 in PEL cells is to prevent BAX/BAK1-mediated apoptosis. The selective requirement to MCL1 is due to MCL1 being expressed in excess over the BCL2 family. Ectopic expression of several BCL2 family proteins, as well as the KSHV BCL2 homolog, significantly decreased basal caspase 3/7 activity and buffered against staurosporine-induced apoptosis. Finally, over-expressed BCL2 family members can functionally substitute for MCL1, when it is inhibited by S63845. Together our data indicate that the expression levels of the BCL2 family likely explain why PEL tumor cells are highly addicted to MCL1. Importantly, our results suggest that caution should be taken when considering MCL1i as a monotherapy regimen for PEL, because resistance can easily develop. IMPORTANCEPrimary effusion lymphoma (PEL) is caused by Kaposis sarcoma-associated herpesvirus. We previously showed that PEL cell lines require the anti-apoptotic protein MCL1 for survival, but not the other BCL2 family proteins. This selective dependence to MCL1 is unexpected as the BCL2 family functions similarly in preventing intrinsic apoptosis. Recently, new roles for MCL1 not shared with the BCL2 family have emerged. Here, we show that non-canonical functions of MCL1 are unlikely essential. Instead, MCL1 mainly functions to prevent apoptosis. The specific requirement to MCL1 is due to MCL1 being expressed in excess over the BCL2 family. Consistent with this model, shifting these expression ratios changes the requirement away from MCL1 and towards the dominant BCL2 family gene. Together, our results indicate that although MCL1 is an attractive chemotherapeutic target to treat PEL, careful consideration must be taken as resistance to MCL1-specific inhibitors easily develops through BCL2 family overexpression.