Compromising the 19S proteasome complex protects cells from reduced flux through the proteasome

Thank you for submitting your work entitled “Compromising the 19S proteasome complex protects cells from reduced flux through the proteasome” for peer review at eLife. Your submission has been favorably evaluated by Randy Schekman (Senior Editor) and two reviewers, one of whom, Jeffery W Kelly, is a member of our Board of Reviewing Editors.

The reviewers have discussed the reviews with one another, and the Reviewing editor has drafted this decision to help you prepare a revised submission.

The Lindquist Laboratory exposed cells to pharmacologic proteasome inhibitors and performed genome-wide screens for mutations that allowed the cells to remain viable. Survival was increased by reduced expression of subunits comprising the 19S regulatory subunit of the proteasome, thereby increasing the ratio of 20S to 26S proteasomes. The central question both reviewers would like discussed in the paper is: how is it that a reduction of 19S would render the cells less sensitive to inhibition by bortezomib?

The main concern of the expert reviewer has to do with the hypothesis put forward by the authors to explain why cells with diminished 19S cap are more resistant to bortezomib and/or MG132. The chief argument appears to be that a subset of the 20S is refractory to inhibition in these cells. However, there are some problems with the experiments that show this. In the first experiment (Figure 2 panel F), there is indeed less 20S activity in WT cells treated with bortezomib compared with 19S-deficient cells. However, there is also much less proteasome. Where is the proteasome going in the WT control? Is this simply a consequence of cell death? But compare with panel 2B – cell number actually seems to be increasing over the first 24h.

This experiment would benefit from being repeated as a time-course, so that it is possible to evaluate proteasome activity at a time before the proteasome has largely disappeared from the WT cells. It would also be much better to do this experiment with epoxomicin or carfilzomib because BTZ can potentially dissociate during the native gel electrophoresis, given that the t1/2 has been reported to be in the range of 100 minutes. Finally, it would be desirable to see this effect quantified (e.g. with an LLVY-AMC assay or similar). Please address as many of these suggestions as possible when revisiting your paper for serious consideration for publication.

Reviewer #1:

This very well written paper entitled “Compromising the 19S proteasome complex protects cells from reduced flux through the proteasome” is competitive for space in eLife from this reviewer's perspective.

The Lindquist Laboratory exposed cells to pharmacologic proteasome inhibitors and performed genome-wide screens for mutations that allowed the cells to remain viable. Survival was increased by reduced expression of subunits comprising the 19S regulatory subunit of the proteasome, thereby increasing the ratio of 20S to 26S proteasomes. While this cellular alteration has fitness costs in the absence of stress (stable cell lines could not be isolated under basal conditions), this means of counteracting reduced flux through the proteasome under stressful conditions is conserved from yeast to humans. All insertions that reached a high level of statistical enrichment occur in genes encoding subunits of the proteasome 19S regulatory complex. These included both subunits that hydrolyze ATP (PSMC2, PSMC3, PSMC4, PSMC5, and PSMC6) subunits that are not ATPases (PSMD2, PSMD6, PSMD7, and PSMD12). No insertions were recovered in genes encoding subunits of the 20S catalytic component of the proteasome. The Bortezomib screen uncovered less loss-of-function hits, consistent with its superior selectivity relative to MG-132 that hits many targets in addition to the 20S proteasome. ShRNA targeting of the PSMD2 subunit of the 19S regulatory subunit rendered cells much more resistant to Bortezomib without activating the heat shock response or the NRF2 transcription factor-this treatment did not ameliorate ER stress or Hsp90 inhibition or non-proteasome flux stresses. In cells with reduced levels of PSMD2, bortezomib treatment strongly repressed genes involved in DNA replication. This heightened anti-proliferation response suggests that cells with reduced 19S subunits are primed to enter a protected, quiescent-like state when flux through the proteasome is reduced. Moreover, reducing PSMD2 levels maintained proteasome degradation flux upon bortezomib treatment and maintained the normal rate of protein synthesis. These phenotypes in mammalian cells are preserved in yeast, identifying a mechanism preserved over a billion years of evolution. In cancer cell lines, proteasome resistance is often coupled to reduction in the expression of one or more genes encoding the 19S subunits of the proteasome, speaking to significance. The central questions I have are:

1) How is it that a reduction of 19S regulatory subunit of the proteasome would render the cells less sensitive to inhibition by proteasome inhibitors?

2) How the 20S subunit increases relative to the 26S subunit is not clear to me mechanistically; please address.

Reviewer #2:

This is a very interesting paper that makes the surprising observation that knockdown of subunits of the proteasome 19S cap renders cells moderately resistant to CFZ. This is a very counter-intuitive observation that could be of considerable importance as it sheds new light on regulatory mechanisms that enable cells to deal with proteotoxic stress.

Overall, I find this to be worthy of serious consideration for publication in eLife. My main concern has to do with the hypothesis put forward by the authors to explain why cells with diminished 19S cap are more resistant to bortezomib and/or MG132. The chief argument appears to be that a subset of the 20S is refractory to inhibition in these cells. However, there are some problems with the experiments that show this. In the first experiment (Figure 2 panel F), there is indeed less 20S activity in WT cells treated with bortezomib compared with 19S-deficient cells, However, there is also much less proteasome. Where is the proteasome going in the WT control? Is this simply a consequence of cell death? But compare with panel 2B – cell number actually seems to be increasing over the first 24h. This experiment would benefit from being repeated as a time-course, so that it is possible to evaluate proteasome activity at a time before the proteasome has largely disappeared from the WT cells. It would also be much better to do this experiment with epoxomicin or carfilzomib because BTZ can potentially dissociate during the native gel electrophoresis, given that the t1/2 has been reported to be in the range of 100 minutes. Finally, it would be desirable to see this effect quantified (e.g. with an LLVY-AMC assay or similar). In Figure 4 panel C, a similar experiment is done with transiently depleted cells. Unlike Fig. 2B, there is not dramatic loss of proteasome in 19S-depleted cells. However, there is essentially undetectable level of 20S activity in the cells treated with BTZ. A time-course and/or dose-titration would be informative. Together, Figures 2B and 4C are key results for the paper, but the existing data are just not all that compelling.

The second major point is that I would have liked to see more of a discussion regarding possible mechanisms. How is it that a reduction of 19S would render the cells less sensitive to inhibition by bortezomib? There are several possible explanations, some of which are testable, but the authors are relatively silent on this salient point.