The integrated stress response remodels the microtubule-organizing center to clear unfolded proteins following proteotoxic stress

Curation statements for this article:
  • Curated by eLife

    eLife logo

    Evaluation Summary:

    This interesting study identifies why or how the integrated stress response pathway regulates cell recovery upon proteotoxic stress, which is especially interesting in cancer cells resistant to proteasome inhibitors. The authors conclude that translation initiation of mRNAs encoding microtubule cytoskeleton, centrosome and ATF5 proteins is necessary to recover from proteotoxic stress. This paper will make a strong contribution to the literature.

    (This preprint has been reviewed by eLife. We include the public reviews from the reviewers here; the authors also receive private feedback with suggested changes to the manuscript. The reviewers remained anonymous to the authors.)

This article has been Reviewed by the following groups

Read the full article See related articles

Abstract

Cells encountering stressful situations activate the integrated stress response (ISR) pathway to limit protein synthesis and redirect translation to better cope. The ISR has also been implicated in cancers, but redundancies in the stress-sensing kinases that trigger the ISR have posed hurdles to dissecting physiological relevance. To overcome this challenge, we targeted the regulatory node of these kinases, namely, the S51 phosphorylation site of eukaryotic translation initiation factor eIF2α and genetically replaced eIF2α with eIF2α-S51A in mouse squamous cell carcinoma (SCC) stem cells of skin. While inconsequential under normal growth conditions, the vulnerability of this ISR-null state was unveiled when SCC stem cells experienced proteotoxic stress. Seeking mechanistic insights into the protective roles of the ISR, we combined ribosome profiling and functional approaches to identify and probe the functional importance of translational differences between ISR-competent and ISR-null SCC stem cells when exposed to proteotoxic stress. In doing so, we learned that the ISR redirects translation to centrosomal proteins that orchestrate the microtubule dynamics needed to efficiently concentrate unfolded proteins at the microtubule-organizing center so that they can be cleared by the perinuclear degradation machinery. Thus, rather than merely maintaining survival during proteotoxic stress, the ISR also functions in promoting cellular recovery once the stress has subsided. Remarkably, this molecular program is unique to transformed skin stem cells, hence exposing a vulnerability in cancer that could be exploited therapeutically.

Article activity feed

  1. Author Response

    Reviewer #1 (Public Review):

    This study by Hurwitz et al. defines a functional relationship between the ISR and microtubule dynamics. This is mediated through the mRNA-specific translation of genes including ATF5 in the context of proteotoxic stress. They further show that this relationship is particularly important in the context of recovery from bortezomib treatment which in the WT setting leads to efficient clearance of protein aggregates. However, this process is significantly less efficient when the ISR is impaired through phosphorylation defective eIF2alpha. The authors use whole transcriptome ribosome profiling to identify mRNAs that are differentially translated upon treatment with bortezomib and uncover a subset of mRNAs which exhibit an increase in translation. This list of 24 mRNAs includes ATF5 which they functionally show is important for cell survival in the context of proteotoxic stress. Overall, this work is solid and provides new insights into SCC management proteotoxic stress. This work reveals a new arm of cellular regulation dynamics that is controlled by the ISR and helps grow our understanding of how the ISR enables survival in the context of distinct stress.

    We thank this Reviewer for his/her comments. We are pleased to notice that he/she finds our work to be solid and innovative and that it will help to advance our understanding of how the ISR promotes recovery in the context of distinct stresses. We also appreciate this Reviewer’s enthusiasm for our work, which he/she defines as interesting and a strong contribution to the literature.

    Reviewer #2 (Public Review):

    The manuscript from Hurwitz et al. documents a connection between the integral stress response (ISR) and a centrosome-mediated protein clearance mechanism, using skin carcinoma cells as a model system. Using ribosomal profiling and molecular approaches, the authors identify that upon stress, the IRS promotes a shift in the translation of centrosomal proteins required for the clearance of unfolded protein-enriched aggregates in the pericentrosomal area. Abrogating the ISR response sensitizes cancer cells, promoting cell death. The authors generated useful cellular tools for the community and information about the translational changes of specific proteins involved in the IRS response, paving the way for future studies.

    There are no major significant weaknesses, and the authors achieve their aim of dissecting the relevance of the ISR in skin carcinoma cells.

    We are grateful to this Reviewer for his/her comments. We are happy to note that this Reviewer acknowledges the generation of useful research tools for the scientific community and that our work will likely pave the way to future research endeavors on this topic. We are also pleased that he/she found our work to be convincing and statistically robust without any significant weaknesses.

    Reviewer #3 (Public Review):

    This is an interesting study that identifies why or how the ISR pathway regulates cell recovery upon proteotoxic stress, which is especially interesting in cancer cells resistant to proteasome inhibitors. The study concludes that only by favouring canonical translation initiation of mRNAs encoding microtubule cytoskeleton, centrosome and ATF5 proteins are necessary to recover from proteotoxic stress. The study is robust and uses advanced pre-clinical models and sequencing techniques to explore the translatome of stressed cancer cells.

    The authors claim that they find a proteotoxic mechanism exclusive to SCC stem cells. However the authors do not use stem cells, they work with primary SCC cells. They would need to actually show in stem cells that this is the case and that normal keratinocytes or epidermal stem cells do not use this exclusive mechanism. In addition, it would be very interesting to translate these findings into the clinic. It would be interesting to know how relevant this mechanism is for human tumour cells.

    We thank this Reviewer for his/her useful comments. We are pleased to notice that he/she found our study to be interesting and robust. We share this Reviewer’s excitement for our pre-clinical model and agree that future study should aim at dissecting the clinical relevance of the combination of proteasome inhibitors together with ISR inactivation.

    In our study, we used primary keratinocytes, which have been transformed by expression of mutant HRasG12V and deletion of TgfbrII. These cells have been previously characterized and used as a model of SCC stem cells (Yang et al., 2015). To address whether transformed keratinocytes are differentially sensitive to proteotoxic stress, we compared responses to bortezomib in primary keratinocytes isolated from littermates: either WT control (TGFbrIIfl/fl), HRasG12V expressing, pre-transformed keratinocytes (HRasG12V; TGFbrIIfl/fl) and fully transformed keratinocytes (HRasG12V; TGFbrIID/D), which were used to generate S51 cells.

    We find that WT keratinocytes are significantly more sensitive to bortezomib, underscoring a unique way in which cancer cells are able to protect themselves against proteotoxic stress (Figure 1G and Figure1figure supplement 3A). We show that a major difference the unique ability of cancer cells to translationally upregulate ATF5 during stress (Figure 7H). The ability of the cancer cells to do so relies upon the ISR, as when eIF2a phosphorylation is prevented by S51A, the cancer cells are no longer protected. In agreement with a role for ATF5 in centrosome dynamics, WT cells fail to increase MTOC size (Figure 4-figure supplement 3). Altogether these data suggest that SCC cells have acquired a resistance mechanism to proteotoxic stress, which allows rapid recovery and preservation of microtubule function.

  2. Evaluation Summary:

    This interesting study identifies why or how the integrated stress response pathway regulates cell recovery upon proteotoxic stress, which is especially interesting in cancer cells resistant to proteasome inhibitors. The authors conclude that translation initiation of mRNAs encoding microtubule cytoskeleton, centrosome and ATF5 proteins is necessary to recover from proteotoxic stress. This paper will make a strong contribution to the literature.

    (This preprint has been reviewed by eLife. We include the public reviews from the reviewers here; the authors also receive private feedback with suggested changes to the manuscript. The reviewers remained anonymous to the authors.)

  3. Reviewer #1 (Public Review):

    This study by Hurwitz et al. defines a functional relationship between the ISR and microtubule dynamics. This is mediated through the mRNA-specific translation of genes including ATF5 in the context of proteotoxic stress. They further show that this relationship is particularly important in the context of recovery from bortezomib treatment which in the WT setting leads to efficient clearance of protein aggregates. However, this process is significantly less efficient when the ISR is impaired through phosphorylation defective eIF2alpha. The authors use whole transcriptome ribosome profiling to identify mRNAs that are differentially translated upon treatment with bortezomib and uncover a subset of mRNAs which exhibit an increase in translation. This list of 24 mRNAs includes ATF5 which they functionally show is important for cell survival in the context of proteotoxic stress. Overall, this work is solid and provides new insights into SCC management proteotoxic stress. This work reveals a new arm of cellular regulation dynamics that is controlled by the ISR and helps grow our understanding of how the ISR enables survival in the context of distinct stress.

  4. Reviewer #2 (Public Review):

    The manuscript from Hurwitz et al. documents a connection between the integral stress response (ISR) and a centrosome-mediated protein clearance mechanism, using skin carcinoma cells as a model system.

    Using ribosomal profiling and molecular approaches, the authors identify that upon stress, the IRS promotes a shift in the translation of centrosomal proteins required for the clearance of unfolded protein-enriched aggregates in the pericentrosomal area. Abrogating the ISR response sensitizes cancer cells, promoting cell death.

    The authors generated useful cellular tools for the community and information about the translational changes of specific proteins involved in the IRS response, paving the way for future studies.

    There are no major significant weaknesses, and the authors achieve their aim of dissecting the relevance of the ISR in skin carcinoma cells.

  5. Reviewer #3 (Public Review):

    This is an interesting study that identifies why or how the ISR pathway regulates cell recovery upon proteotoxic stress, which is especially interesting in cancer cells resistant to proteasome inhibitors. The study concludes that only by favouring canonical translation initiation of mRNAs encoding microtubule cytoskeleton, centrosome and ATF5 proteins are necessary to recover from proteotoxic stress. The study is robust and uses advanced pre-clinical models and sequencing techniques to explore the translatome of stressed cancer cells.

    The authors claim that they find a proteotoxic mechanism exclusive to SCC stem cells. However the authors do not use stem cells, they work with primary SCC cells. They would need to actually show in stem cells that this is the case and that normal keratinocytes or epidermal stem cells do not use this exclusive mechanism. In addition, it would be very interesting to translate these findings into the clinic. It would be interesting to know how relevant this mechanism is for human tumour cells.