Antiestrogen resistance is a major limitation to improving breast cancer survival rates and elucidating its mechanisms remains an important challenge 
. In breast tumors, BCL2 expression measured prior to therapy correlates with ER expression and an improved response to antiestrogens 
. However, BCL2 levels decrease after TAM therapy, but only in those women who obtain clinical benefit 
. In breast tumors, apoptosis increases after the first 24 hr of TAM treatment but markedly decreases 3-months later. Moreover, BCL2 expression is elevated in residual (resistant) tumors 
. We hypothesized that increased expression of BCL2 and/or BCL-W may play a role in antiestrogen resistance by allowing resistant cells to evade apoptosis. We show that in the absence of estrogen there is an increase in basal and ICI-regulated BCL2 mRNA, protein, and promoter activity in resistant cells, observations consistent with data showing elevated activity of two upstream regulators of BCL2: NFκB and XBP1 
. However, co-inhibition of BCL-W and BCL2 is required to restore ICI sensitivity, a process that is driven by increased autophagy and necrosis, but not apoptosis. We also show that increased autophagy may activate necrotic cell death in resistant cells.
Little is known about BCL-W expression and function in breast cancer. Since BCL-W is overexpressed in some human colon cancer cells 
and its expression is regulated by estrogen in cerebrocortical neuron cultures 
, we hypothesized that BCL-W could play a role in antiestrogen resistance. BCL-W expression is increased by ICI in both sensitive and resistant cells, suggesting that an increased co-expression of both BCL2 and BCL-W is required for antiestrogen resistance. Hence, the ICI-induced increase in the expression of BCL-W alone in antiestrogen-sensitive cells has little effect on responsiveness unless accompanied by a concurrent increase in BCL2, as is seen in resistant cells.
Small-molecule inhibitors of proapoptotic BCL2 family members can restore sensitivity to some therapeutic agents that induce apoptosis 
. Some of these compounds inhibit the proliferation of cells that express high levels of BCL2 
. However, several antiapoptotic BCL2 family members also regulate autophagy through their interactions with BECN1 
. In resistant MCF-7/LCC9 and LY2 cells, only the levels of autophagy and necrosis increase after YC137+ICI treatment; there is no increase in either MMP or apoptosis.
We used BCL-W and/or BCL2 siRNA to confirm the results with YC137. As expected, BCL-W+BCL2 co-inhibition has no effect on apoptosis in ICI treated MCF-7/LCC9 cells, whereas both autophagy and necrosis increase. Inhibition of BCL2 and BCL-xL decreases cellular ATP and increases necrosis (but not apoptosis) in acinar cells hyperstimulated with CCK-8 
. Autophagy can also activate necrosis in apoptosis-deficient mouse embryonic fibroblasts 
. Increased autophagosome formation is induced early during necrotic cell death and contributes to the cellular destruction that occurs during necrosis in Caenorhabditis elegans 
. These results suggest that BCL-W+BCL2 coinhibition can increase antiestrogen sensitivity in resistant breast cancer cells by preferentially activating necrosis, apparently in association with the induction of autophagy. In contrast, inhibiting autophagy in some TAM-resistant breast cancer cells can increase apoptosis 
No change occurs in the proportion of cells undergoing S-phase after 3MA+YC137+ICI treatment. Thus, it is unlikely that autophagy plays a major role in the cell cycle arrest effects of antiestrogens. We also show that the inhibition of autophagy, in combination with BCL-W+BCL2 co-inhibition in ICI treated resistant cells, does not further reduce total cell number but shifts programmed cell death such that apoptosis increases and necrosis decreases. Our results strongly suggest that functional autophagy is a central component of the cell fate decision machinery in ICI-resistant breast cancer cells, although we cannot exclude the possibility that autophagy also alters the kinetics of cell death. Nonetheless, in addition to being a cell death effector mechanism, autophagy appears to be a central component in influencing how breast cancer cells die in response to antiestrogens.
In summary, our results show that BCL-W+BCL2 co-inhibition restores ICI sensitivity in antiestrogen-resistant cells and increases ICI sensitivity in antiestrogen-sensitive cells. We show that the overexpression of BCL-W and BCL2 is linked to determining cell fate through autophagy in ICI resistant breast cancer models (Figure S5A
). We have shown that BCL-W+BCL2 coinhibition increases autophagy and necrosis with no effect on the extent of apoptotic cell death (Figure S5B
). These data suggest that BCL-W and BCL2 activate apoptosis and necrosis by initially regulating autophagy (Figure S5C
). We conclude that the co-inhibition of BCL-W and BCL2 restores sensitivity in antiestrogen-resistant breast cancer cells by promoting an autophagy-associated increase in necrosis. Antiestrogen sensitive cells undergo autophagy and/or apoptosis, whereas resistant cells undergo autophagy and necrosis when resensitized. These different cell death outcomes in sensitive and resistant cells show the notable plasticity of cell fate mechanisms in breast cancer. In resistant cells, resensitization to antiestrogens can also occur without the cell cycle arrest that accompanies cell death in de novo
sensitive cells. Thus, antiestrogen-regulated signaling that modifies cell cycling occurs through mechanisms independent of mitochondrial function and cell death.
From a therapeutic perspective, these data also suggest that broad rather than specific BCL2 family member inhibitors will have greater clinical value and may explain the apparent lack of activity of targeted BCL2 antisense monotherapy in clinical trials 
. Combination therapy with endocrine agents and broadly active small molecule inhibitors of BCL2 family members may delay, prevent, or reverse the acquisition of antiestrogen resistance in breast cancer patients and lead to significant improvements in survival.