It is also possible that calpain-mediated cleavage of proteins is elevated by bortezomib; this has been proposed to explain the increased degradation of IkBa caused by bortezomib high content screening treatment of various human cell lines. However, a previous peptidomics study did not detect major changes in levels of intracellular peptides when SH-SY5Y cells were treated with a calcium ionophore known to activate calpains. Another possibility is activation of autophagy by bortezomib, which is known to induce autophagy in several systems. This idea is attractive because of the large number of mitochondrial protein fragments found to be elevated by bortezomib. However, the standard marker for autophagy, LC3, showed no evidence of autophagy upon treatment of HEK293T or SH-SY5Y cells with high concentrations of bortezomib for 1 hour. Furthermore, treatment of cells with rapamycin for 1 hour to produce autophagy had little effect on the cellular peptides. Taken together, these results suggest that autophagy does not contribute to the altered peptidome observed upon treatment of cells with bortezomib for short time periods. A third possibility to explain 
the increase in many peptides is that protein levels are induced by bortezomib. A previous RNA microarray analysis found that thousands of mRNAs were either up- or down-regulated by treatment with bortezomib for 14, 24, or 48 hours. A cross-comparison of the RNA microarray study and our results found no correlation between those proteins corresponding to up-regulated peptides and mRNA changes at 14 and 24 hour time points. Furthermore, our finding that most of the peptides were up-regulated as early as 30 minutes after the start of the exposure to bortezomib argues against a general effect on protein synthesis; while the synthesis of some proteins may be stimulated within 30 minutes of the start of bortezomib treatment, it is unlikely that all of the affected proteins will show such a rapid increase. Furthermore, a change in protein levels would not explain why all peptides derived from a particular protein are not similarly affected. For example, although many fragments of heat shock 10 kDa protein 1are elevated, one fragment is significantly decreased by 50 and 500 nM bortezomib. Similar variability in the changes of peptides derived from other proteins is observed. Therefore, an increase in the synthesis of these proteins would not explain why some of the peptides derived from these proteins were decreased by the bortezomib treatment. A fourth possibility is that bortezomib BIBW2992 interferes with the further degradation of the peptides produced by the proteasome. Although bortezomib is usually described in the literature as being highly specific for the proteasome with no off-target effects, a recent study has shown that bortezomib inhibits serine proteases such as cathepsins A and G, chymase, dipeptidyl peptidase II, and HtrA2/Omi. Although none of these enzymes are thought to function in the degradation of peptides produced by the proteasome, it is possible that bortezomib has additional off target effects and inhibits a cytosolic peptidase. This possibility would be consistent with the increase in peptides derived from cytosolic proteins as well as mitochondrial proteins; this organelle contains transporters that export peptides generated by proteases within the mitochondria, and these peptides are subsequently degraded by cytosolic peptidases. The mechanism of action of bortezomib as an antitumor agent is thought to involve a reduction in protein turnover and/or protein activation, especially for proteins such as NFkB and cyclins. To generate the active NFkB transcriptional dimeric complexes, two precursors, NFkB1and NFkB2, have to undergo limited proteolytic processing by the proteasome to yield.