Activity Profiling of Proteasome Subunits and Deubiquitinating Enzymes in Human Hematopoetic Malignancies.

Proteasomal proteolysis is based on the activity of six different catalytic proteasomal subunits (β1, β2, β5, β1i, β2i, β5i) as well as ubiquitin-mediated recruitment of cytosolic proteins to the proteasomal complex. The latter is regulated by a balance between ubiquitinating enzymes and a variety of ubiquitin-specific proteases (USP). Although proteasomal proteolysis has become a major therapeutic target, the individual activity profiles of proteasomal subunits and USP in primary human malignant cells are unknown. Recently, an activity-based, cell-permeable affinity probe has been developed, which allows to visualize the individual proteasomal subunit activity in living cells in a semiquantitative fashion (Berkers et al., Nature Methods, May 2005). Similarly, a ubiquitin-based synthetic probe (HAUbVS) allows to visualize a broad range of USP based on their activity in cellular lysates. We have used both types of probes to assess i) the differential activity of the proteasomal catalytic subunits in intact cells and ii) the activity profiles of a panel of USP in primary cells of human hematologic malignancies.

Individual subunits of both the constitutive and the immuoproteasome were tagged by the probe in live cells in situ, resolved by 1D or 2D SDS-PAGE, and identified by mass spectrometry. Comparison of different cases of ALL and CLL revealed a remarkable variability in the relative activities of the β1/1i–, β2/2i–, and β5/5i–type of subunits, respectively, which contrasted with their stochiometrically constant relation of the respective polypeptides in the architecture of the proteasome. Of note, live primary myeloma cells lacked active β1/1i–type subunits and contained only extremely low amounts of β5/5i–type active subunits, while active β2/2i subunits were present in amounts comparable with primary leukaemia cells or monocytes. The variability in active proteasomal subunits contrasted with a relatively homogeneous distribution of the activity profiles of seven different USP that were targeted by the USP-specific probe. In particular, and in contrast to control cell lines, we failed to confirm the substantial upregulation of UCH-L1 in primary cells, which had been suggested to be a characteristic of the proteasomal machinery in malignant cells, based on studies using human cell lines.

In addition, we here for the first time directly visualize the effect of bortezomib on individual proteasome subunits in primary human cells: although Bortezomib is reported to selectively target the β5 subunit, we here show that both the β1 and β1i subunits are also inhibited by the substance, the latter with even higher affinity (IC50 approx. 2.5 nM). In one case of ALL that lacked detectable active β5/5i subunits in intact cells, Bortezomib induced cytotoxic death at the concentration where β1/β1i activity was eliminated.

In summary, our study for the first time visualizes the differential activity of proteasomal subunits in living primary human cells, demonstrates significant variability in their activity profiles in leukaemia and myeloma cells and suggests that differential proteasomal activity might contribute to the variable sensitivity of cancer cells towards proteasome inhibitors such as Bortezomib.