Hsf1 and the molecular chaperone Hsp90 support a “rewiring stress response” leading to an adaptive cell size increase in chronic stress

Reviewer #1 (Public Review):

The manuscript describes that cultured mammalian cells adapt to chronic stress by increasing their size and protein translation through Hsp90. The authors extensively use Hsp90 knockout cells and mass spectrometry to provide solid evidence that chronic heat shock response is accompanied by cell size changes and stress resistance in large cells. The major strength of the work is the authors ability to document the heat shock response in detail, while the main weakness is that the cell size changes appear not to be quantitative making it difficult to assess how much the cell density is changed in chronic stress. Nevertheless, the increased stress resistance of large cells is conceptually important and provides one potential explanation why cells need to control their size. This work adds to our understanding of how cellular stress is managed, and while stress responses have been observed previously in relation to cell size, this work provides evidence for increased stress resistance in larger cells.

Reviewer #2 (Public Review):

The paper by Maiti et al. reporting a highly interesting, previously un-noticed, phenomenon of cell size increase as part of the response to chronic proteotoxic stresses, such as heat shock, which the authors term "rewiring stress response". Furthermore, they establish that it is mediated via HSF1, and, strikingly, necessitates a certain threshold levels of HSP90. Dwelling deeper into the underlying mechanisms, they find that HSP90 help scale protein synthesis with the increased cell sizes, and when diminished, this scaling is impaired, and also cell viability in chronic stress is also compromised. These findings correspond with a previous study by this group on the lethality of HSP90 deficient mice, and moreover, have implications to our understanding of cellular adaptation to stress, and generate interesting hypotheses about the possible links of this mechanism to the impairments of the ability to cope with stress during aging and senescence.

This is an excellent study, with highly novel and important findings, which illuminate a new phenomenon related to cellular adaptation to chronic stress. I have only one major concern, about some technical aspects, specifically over-crowding effects, which could confound the results, which should be answered by the authors. Other than that, further details which I think are pertinent to the study most likely already exist in the experiments performed, and most could be answered with additional simple experiments and by further analyses of the proteomics data which has already been performed, but which results are not sufficiently shown in detail.

Reviewer #3 (Public Review):

This study focuses on defining the specific importance of HSP90 isoforms in stress-resistance. Specifically, addressing the importance of the two HSP90 isoforms alpha and beta in adapting cells to chronic stress. Noting that chronic stresses of different types can induce increases of cellular size, the authors investigated the role of HSP90a/b in this process. Intriguingly, they found that KO of either of these isoforms did not influence chronic stress-dependent increases of cell size. However, they did find that HSF1 plays an important role in this process through undefined mechanisms. The authors go on to show that this increase in cell size appears to be correlated with enhanced protein synthesis during conditions of stress, which allow cells to maintain protein density in the enlarged cell. Intriguingly, this correlation is disrupted in HSP90a/b KO cells, where cell size increases, but there is a deficiency in recovery of protein synthesis following the initial insult. This appears to involve sustained ISR signaling that does not resolve in HSP90-deficient cells. Using a number of different compounds that increase (e.g., CDKi) or inhibit (e.g., rapamycin) cell size changes, the authors demonstrate that protection against chronic stress correlates with cell size and protein density, linking cell expansion to stress resistance.

Overall, this is an observational study that heavily relies on correlation to define a proposed stress responsive signaling mechanism termed the 'rewiring stress response' to explain the coordinated increase in cell size and protein translation in protection against chronic stress.

Due to the reliance on correlation, there remains many questions unanswered related to this work. For example. What is the specific role for HSP90a/b in regulating protein translation during chronic stress through the ISR or related pathways? The authors indicate that the induction of the eIF2a phosphatase GADD34 is not impacted in HSP90-deficient cells, so what role does HSP90 have in this process. Is HSP90 required for proper folding of GADD34? Would you see similar effects in protein translation recovery if other ISR activators are used in HSP90-deficient cells? Addressing this central unanswered question that would significantly enhance the current study. While the authors are undoubtedly pursuing this in subsequent studies, it is difficult to fully gauge the impact of this work without more clarity on that point specifically.

Along the same lines, another critical unanswered question is 'Are similar effects observed in non-dividing cells?' Does chronic stress lead to increases of size and regulation of protein translation in primary cell models that are not undergoing division.

Ultimately, this is an interesting study that does a good job of establishing correlations between increases in cell size and protein translation, but does not get to the really intriguing questions related to this coordination. As this study is extended through either revisions to this manuscript or subsequent papers, the importance of this rewiring stress response in the context of cellular stress and pathologic conditions (e.g., age-associated disease) will become increasing apparent.

Author Response:

Regarding the two main points emphasized by the eLife assessment:

• Potentially confounding effects of overcrowding: This is indeed an important point, which we avoided, unfortunately without explicitly mentioning it in the manuscript (assuming that it went without saying.) We will point out that our proliferation assays, already part of the original manuscript, indicated that cells were not overcroweded. Nevertheless, we will include additional evidence indicating that our cells were not overcrowded and remained subconfluent.

• Mechanisms: We will mention even more explicitly than we already did that this is beyond the scope of this story and why that is. As we did say, there are lots of factors directly or indirectly involved in translation that depend on Hsp90. Figuring out which one or which ones it might be is a whole new and totally open-ended project.

Regarding some of the other public comments:

• While we did provide quantitative (!) data on changes in cytoplasmic density (e.g. diffusion coefficients, total amount of protein relative to cell size), we will emphasize in the revised manuscript that the changes in cell size, as measured by both flow cytometry and image analyses, are a relative and approximate measure of the 3D changes in cell volume. Although our data on the diffusion coefficients, which report on cytoplasmic density, are directly comparable, our measurements of the amounts of protein relative to cell size (if this is what the comment meant with "cell density") have at least relative value.

• Results of proteomic data not shown in sufficient detail: We recognize that it is not trivial to "read" the data as presented in the paper (volcano plots, full datasets as an Excel file and through ProteomeXchange). We will add subsets of the proteomic data to the Excel file and include some Gene Ontology analyses.

• We did demonstrate that Hsf1 most likely acts transcriptionally to promote the observed cell size increase.

• We acknowledge that a large fraction of our data is "observational", but some experiments clearly go beyond providing correlations. When we manipulate some of the players genetically (KO, knockdown, overexpression) or pharmacologically, we get results that support our conclusions about underlying mechanistic connections.

• GADD34: This protein is not known to be an Hsp90 client (or interactor), which is also supported by our mass spec data since its steady-state levels don't change in Hsp90α or β KO cells compared to wild-type cells.

• Non-dividing cells: it would indeed be exciting to determine whether the same phenomena and mechanisms apply to non-dividing cells. However, there are likely to be substantial technical challenges. We would need primary human (or alternatively murine) cells such as B-cells or hepatocytes, and it is difficult to predict whether they would tolerate mild heat stress for several days. It might also be possible to explore this with a mouse model, but clearly, this must be left to future studies.