Characterization of the ubiquitin-proteasome system in bortezomib-adapted cells.

Resistance towards the proteasome inhibitor bortezomib is poorly understood. We adapted the HL-60, ARH-77 and AMO-1 cell lines (myeloid leukemia, plasmocytoid lymphoma, myeloma) to bortezomib exceeding therapeutic plasma levels, and compared characteristics of the ubiquitin-proteasome system, alternative proteases and the unfolded protein response (UPR) between adapted cells and parental lines. Adapted cells showed increased transcription rates, activities and polypeptide levels of the bortezomib-sensitive beta5, but also of the beta2 proteasome subunit and consistently retained elevated levels of active beta1/beta5-type proteasome subunits in the presence of therapeutic levels of bortezomib. Bortezomib-adapted HL-60 cells showed increased expression and proteasome association of the 11S proteasome activator, and did not accumulate poly-ubiquitinated protein, activate the UPR or UPR-mediated apoptosis in response to bortezomib. The rate of protein biosynthesis was reduced, and the transcription of chaperone genes downmodulated. We did not observe major changes in the activities of TPPII, cathepsins or deubiquitinating proteases. We conclude that different types of bortezomib-adapted cell lines, including myeloma, show similar patterns of changes in the proteasomal machinery which result in residual proteasome activity in the presence of bortezomib and a quantitative balance between protein biosynthesis and destruction.

Activity patterns of proteasome subunits reflect bortezomib sensitivity of hematologic malignancies and are variable in primary human leukemia cells.

Proteasomal proteolysis relies on the activity of six catalytically active proteasomal subunits (beta1, beta2, beta5, beta1i, beta2i and beta5i). Applying a functional proteomics approach, we used a recently developed activity-based, cell-permeable proteasome-specific probe that for the first time allows differential visualization of individual active proteasomal subunits in intact primary cells. In primary leukemia samples, we observed remarkable variability in the amounts of active beta1/1i-, beta2/2i- and beta5/5i-type of subunits, contrasting with their constant protein expression. Bortezomib inhibited beta5- and beta1-type, but to a lesser extend beta2-type of subunits in live primary cells in vitro and in vivo. When we adapted the bortezomib-sensitive human acute myeloid leukemia cell line HL-60 to bortezomib 40 nM (HL-60a), proteasomal activity profiling revealed an upregulation of active subunits, and residual beta1/beta5-type of activity could be visualized in the presence of bortezomib 20 nM, in contrast to control cells. In a panel of cell lines from hematologic malignancies, the ratio between beta2-type and (beta1 + beta5)-type of active proteasomal polypeptides mirrored different degrees of bortezomib sensitivity. We thus conclude that the proteasomal activity profile varies in primary leukemia cells, and that the pattern of proteasomal subunit activity influences the sensitivity of hematologic malignancies toward bortezomib.