Reply to Hu et al.: Mcl-1 Reduction Due to Caspase-dependent Cleavage during Endoplasmic Reticulum Stress-induced Apoptosis

This is a response to a letter by Hu et al. (1)

As Hu et al. pointed out (1), caspase-3 could process Mcl-1 during apoptosis, depending on the type of apoptotic stimuli or cell (2, 3). However, we did not detect any Mcl-1 fragments by immunoblotting during apoptosis in myoblast cells after induction by ATF6, WBP1, or endoplasmic reticulum (ER) stressors. We thus examined whether active ATF6 could reduce Mcl-1 levels in the absence of caspase activation. These conditions were established by overexpressing Bcl-x_L. Again, Mcl-1 was reduced by active ATF6. Our conclusion is consistent with the current consensus that the level of the Mcl-1 protein is principally regulated by the ubiquitin/deubiquitin system. The cleavage of Mcl-1 by caspases during apoptosis may contribute to the feed-forward amplification of apoptotic signals after caspase activation (4). It is likely that, in myoblast cells, the completion of apoptosis does not depend on this feed-forward mechanism. We did not detect up-regulation of Mcl-1, Noxa, or Bim by DNA microarray analysis or quantitative PCR.

A significant point of the study is that the unfolded protein response is linked to the Bcl-2 family during ER stress-induced apoptosis. Because expression of Bcl-2 family members can differ greatly among cell types (e.g. normal versus cancer cells), and complex interactions occur within the Bcl-2 family, Bcl-2 family proteins may respond to ER stress differently in different cell types. This would generate apparent discrepancies among studies. Nevertheless, these discrepancies may provide valuable clues for clinical applications (5).