Does N-terminal Processing of Mcl-1 Occur at Mitochondrial Outer Membrane or Matrix?

Warr et al. (1) concluded in their paper that the Mcl-1ΔN isoform resides in the mitochondrial outer membrane. The major problem with this conclusion is that the controls for all three experiments attempting to address this issue (Fig. 2) were not handled properly, i.e. all controls for assigning the submitochondrial location of Mcl-1ΔN clearly were done in a separate experiment. For example, in Fig. 2A, Western blots for Tom20 and cytochrome c were not from the same samples analyzed for Mcl-1; in Fig. 2C, manganese superoxide dismutase (MnSOD), used as a control, was not analyzed together with those shown in Fig. 2B; in Fig. 2D, the Mcl-1-Bak complexes should have been analyzed in the same immunoprecipitation-Western blots, instead of two independent blots as shown in this figure. People working on mitochondria all know that the integrity of isolated mitochondria would greatly affect the experimental results and thus their interpretations. To ensure the integrity of the isolated mitochondria, one must analyze controls and experimental groups in the same blot.

Recently we demonstrated that the fast-mobility isoform of Mcl-1 (equivalent to Mcl-1ΔN) is localized to the mitochondrial matrix and is generated there via a mitochondrial processing peptidase-mediated cleavage reaction (2). Consistent with our finding, Warr et al. (1) also showed that generation of Mcl-1ΔN requires an intact electrochemical potential across the inner membrane, which actually argues against their conclusion that Mcl-1ΔN is proteolytically processed at the outer membrane. We thus feel that Warr et al. (1) should provide more solid evidence to support their conclusion.