Abstract
Multicellular organisms remove damaged or superfluous cells through a highly regulated cellular process known as programmed cell death. There are two main forms of programmed cell death, apoptosis and necrosis. Necrosis (necroptosis) previously thought to be an unregulated death pathway was recently found to be highly regulated. The manner by which a cell dies has important implications. In apoptotic cell death, caspases digest the cell to cause implosion in an immunologically silent process. In necroptotic cell death, increased Rip kinase signaling effects rupture of the plasma membrane, cellular explosion, and the activation of an inflammatory response.
Death receptors, such as the TNFα receptor, can activate either apoptotic or necroptotic death. The upstream activators and transducers including Caspase-8, Rip1, and Fadd, are common to both forms of cell death. Interestingly, Caspase-8 and c-FlipL, a caspase homolog, were recently shown to inhibit the necrotic pathway during embryonic development through the formation of a catalytically active complex. The BH3-only Bcl-2 family member, Bid is one of the strongest substrates of Caspase-8, placing it at the interface of the apoptotic and necroptotic pathways, and in position to mediate cell death fate.
The role of apoptosis in hematopoietic homeostasis has been well characterized. We developed a mouse model of unrestrained necroptosis in order to determine how unrestrained necroptosis impacts hematopoietic homeostasis and bone marrow function. To do this we generated a mouse model in which apoptosis is prevented by the deletion of the pro-apoptotic effectors Bax and Bak. We further deleted the upstream activator Bid (VavBaxBakBid TKO mice). Surprisingly, these mice die of bone marrow failure due to unrestrained necroptotic cell death. TKO bone marrow displays necroptotic cells by electron microscopy, and markedly increased Rip1 expression by immunofluorescence. TKO mice die of bone marrow failure with marked myeloid dysplasia between the age of 3 and 12 months, and a small number develop leukemia, a phenotype that closely resembles MDS. Further analysis revealed expansion and increased BrdU incorporation of the SLAM-HSC population, consistent with increased HSC proliferation in response to death of more mature cells. To assess function of these HSCs, we performed competitive reconstitution assays. TKO bone marrow initially outcompetes WT bone marrow, but the mice eventually succumb to bone marrow failure beginning at 5 months post–transplantation, despite the presence of ~10-15% wild type bone marrow. These results demonstrate that increased necroptotic signaling results in a cell autonomous stem cell defect. In addition, the presence of necroptotic bone marrow also kills normal HSCs in a non cell-autonomous manner, due to a feed-forward inflammatory process.
To further characterize how necroptotic cell death is regulated, we developed myeloid progenitor cell lines (MPCs) from the bone marrow of WT, Bid KO, BaxBak DKO, and BaxBakBid TKO mice to facilitate biochemical and mechanistic studies. Our studies demonstrated increased activation (phosphorylation) and markedly increased levels of Rip1 in the pronecrotic complex (Complex II) with Rip3, Caspase-8, and Fadd in our TKO MPCs following LPS treatment. This association of Rip1 with Complex II is abrogated by reintroduction of Bid by retrovirus into TKO MPCs, demonstrating that Bid inhibits Rip1 association with complex II, suggesting that Bid is a key factor that determines cell death fate.
Increased bone marrow cell death is well documented in MDS. To determine if necroptosis plays a role in this bone marrow cell death, we evaluated RIP1 and Caspase 3 expression in 17 human MDS samples. Remarkably, we found increased RIP1 expression, but not activated caspase 3 in bone marrow samples from patients with the RCMD, RAEB-1, and RAEB-2 subtype of MDS, but not in 4 control bone marrow samples (normal lymphoma staging marrows). Our study thus demonstrates that increased necroptosis signaling can result in bone marrow failure with dysplasia, and that necroptotic cell death signaling is increased in bone marrow from MDS patients, highlighting the potential importance of this targetable signaling pathway in bone marrow failure disorders such as MDS.
Disclosures
No relevant conflicts of interest to declare.
Autophagy and cell death: The jury's still out