A point mutation in the nucleotide exchange factor eIF2B constitutively activates the integrated stress response by allosteric modulation

  1. Morgane Boone
  2. Lan Wang
  3. Rosalie E Lawrence
  4. Adam Frost
  5. Peter Walter
  6. Michael Schoof

  • Curated by eLife

    eLife logo

    Evaluation Summary:

    This manuscript addresses a significant and timely topic in translational control and will be of interest to researchers studying molecular biology or diseases impacted by the Integrated Stress Response (ISR). The combination of biochemical, structural, and in-cell experiments constitutes a comprehensive study that supports the proposed model for allosteric regulation of the active/inactive states of the eIF2B complex. The findings are relevant to neuropathologies, infectious and inflammatory diseases, diabetes, and metabolic disorders.

    (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.)

Reviewed by

Read the full article See related articles

Listed in

Log in to save this article

Abstract

In eukaryotic cells, stressors reprogram the cellular proteome by activating the integrated stress response (ISR). In its canonical form, stress-sensing kinases phosphorylate the eukaryotic translation initiation factor eIF2 (eIF2-P), which ultimately leads to reduced levels of ternary complex required for initiation of mRNA translation. Previously we showed that translational control is primarily exerted through a conformational switch in eIF2’s nucleotide exchange factor, eIF2B, which shifts from its active A-State conformation to its inhibited I-State conformation upon eIF2-P binding, resulting in reduced nucleotide exchange on eIF2 (Schoof et al. 2021). Here, we show functionally and structurally how a single histidine to aspartate point mutation in eIF2B’s β subunit (H160D) mimics the effects of eIF2-P binding by promoting an I-State like conformation, resulting in eIF2-P independent activation of the ISR. These findings corroborate our previously proposed A/I-State model of allosteric ISR regulation.

Article activity feed

  1. eLife

    Author Response:

    Reviewer #1 (Public Review):

    The integrated stress response (ISR) controls cellular protein synthesis in response to diverse stimuli. A set of related protein kinases, with distinct regulatory domains that respond to different stress conditions, share a common kinase domain that specifically phosphorylates the translation factor eIF2 on its alpha subunit. Phosphorylation of eIF2 inhibits translation by inactivating eIF2B, the guanine nucleotide exchange factor (GEF) for eIF2. The decameric eIF2B, a dimer of heteropentamers, is the key control hub of the ISR. Previously, a small molecule inhibitor of the ISR called ISRIB was found to bind to eIF2B and was proposed to reverse the impacts of eIF2 phosphorylation by increasing stabilizing the association of eIF2B heteropentamers into the functional decameric complex. However, more recently, an alternative model ISRIB action has been proposed. eIF2B is proposed to toggle between inactivate and active states. Binding of phosphorylated eIF2 to a regulatory site is proposed to trigger the inactive state by allosterically weakening binding of eIF2 at the active site. In the new model, ISRIB has been proposed to favor the active state conformation of eIF2B and thereby overcome the effects of eIF2 phosphorylation.

    In this paper, the authors further study a previously described H160D mutation in the eIF2Bbeta subunit. This mutation at one of the dimer interfaces in eIF2B was previously proposed to inhibit eIF2B by weakening dimerization. Consistent with this hypothesis, the H160D mutation impaired dimerization of eIF2B(beta, gamma, delta, epsilon) tetramers. However, in this study, the authors show that the H160D mutation does not impair dimerization when eIF2Balpha is included; thus, the mutation impairs eIF2B activity without impairing dimerization. Using biochemical assays, the authors show that the H160D mutation impairs nucleotide exchange by eIF2B decamers and weakens the binding eIF2 to eIF2B. However, the binding of phosphorylated eIF2 to eIF2B is not weakened.

    Cryo-EM structural analysis of the mutant eIF2B complex reveals a partial rocking of the decameric structure that resembles the structure of the eIF2B complex when bound to its inhibitor phosphorylated eIF2. In this partially rocked structure, both the ISRIB binding site at the dimer interface and the functional eIF2alpha binding sites are widened, providing a structural solution to why the mutation weakens eIF2 binding. Interestingly, the inhibitory binding site for phosphorylated eIF2 is not affected the H160D mutation. The authors propose that the H160D mutation in eIF2Bbeta induces an allosteric conformational change that mimics the effects of phosphorylated eIF2 binding to eIF2B.

    Finally, the authors generated cell lines that exclusively express the mutant eIF2Bbeta subunit. The mutation impairs total protein synthesis and cell growth rate and leads to elevated expression of the ISR marker ATF4.

    This is a high-quality study, the results are convincing and the authors conclusions are supported by the data. As the ISR has been implicated in a variety of diseases, further elucidation of the mechanism of action of eIF2B and ISRIB will be critical in the development of therapeutic interventions.

    A weakness of the paper (that hopefully can be easily remedied) would be to show the quality control data to verify the mutant cell lines used in Figure 6. It would be good to see that the mutant allele is present in the cells and that no WT alleles remain. In addition, examination of eIF2alpha Ser51 phosphorylation in Figure 6A would strengthen the conclusion that the eIF2Bbeta mutation is activating ATF4 expression independent of changes in eIF2 phosphorylation. Also, use of ATF4 reporters in Figure 6A, in addition to the presented Western data, would provide a nice quantitative read-out for the impact of the H160D mutation on ATF4 mRNA translation. Finally, as the biochemical and structural data indicate that the H160D mutation impairs ISRIB activity, it would be worthwhile testing whether ISRIB will rescue the slow-growth of the H160D cell lines in Figure 6D (the anticipation is that this slow-growth phenotype will not be rescued by ISRIB).

    • The genotype of our cell lines at the EIF2B2 target locus was screened for by PCR + restriction enzyme digest, and later sequence verified by deep sequencing. We used the CRISPResso2 pipeline to calculate allele frequencies and HDR editing efficiencies from the sequencing data, and now also include those results in a supplementary figure (Figure 6 – supplementary figure 1).

    • The levels of baseline eIF2 phosphorylation are indeed the same in WT and both H160D clones, both when assessed using a phospho-specific antibody (for eIF2alpha Ser51-P) or through band shift using phospho-retention gels (Phos-tag). We now include a new supplementary figure with this data (Figure 6 – figure supplement 3A-B).

    • It is well-established in the field that ATF4 is regulated at the translational level during acute ISR activation, and indeed, reporters with the ATF4 5’ UTR have been instrumental in studying and quantifying this, allowing scientists to forego time-intensive western blots and perform high throughput analyses. Stable integration, however, can notoriously affect genomic integrity and otherwise introduce clonal variation, even when the construct is targeted to a specific locus (for example when using the FlpIn system). We have observed heterogeneity in baseline ATF4 reporter signal even when comparing polyclonal cell lines generated by lentiviral integration. As it is best practice to avoid comparing between reporter cell lines generated in different backgrounds (WT vs H160D), particularly when investigating basal conditions, we consider it more appropriate to directly measure the levels of proteins of interest by western blot, as is also commonly done in the field. By showing that ATF4 protein levels increase (Figure 6A) but its transcript levels do not (Figure 6B), while those of its target genes do (Figure 6B), we equally confirm that ATF4 is translationally upregulated in the eIF2B H160D mutant. Moreover, our Western blot conditions provide enough sensitivity to differentiate no stress (lane 1) from mild stress (lanes 5 and 9) and high stress (lanes 7 and 11). We have added notation of these specific relevant lanes to the text to make the point more accessible to the reader. We therefore consider the generation of reporter cell lines in different genetic backgrounds to be a redundant abstraction of a phenotype that we already directly show.

    • Indeed, as predicted from both our in vitro and cellular work, ISRIB did not alter growth half-life of H160D cells. We included these new data in Figure 6 – supplementary figure 3C.

  2. eLife

    Evaluation Summary:

    This manuscript addresses a significant and timely topic in translational control and will be of interest to researchers studying molecular biology or diseases impacted by the Integrated Stress Response (ISR). The combination of biochemical, structural, and in-cell experiments constitutes a comprehensive study that supports the proposed model for allosteric regulation of the active/inactive states of the eIF2B complex. The findings are relevant to neuropathologies, infectious and inflammatory diseases, diabetes, and metabolic disorders.

    (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. eLife

    Reviewer #1 (Public Review):

    The integrated stress response (ISR) controls cellular protein synthesis in response to diverse stimuli. A set of related protein kinases, with distinct regulatory domains that respond to different stress conditions, share a common kinase domain that specifically phosphorylates the translation factor eIF2 on its alpha subunit. Phosphorylation of eIF2 inhibits translation by inactivating eIF2B, the guanine nucleotide exchange factor (GEF) for eIF2. The decameric eIF2B, a dimer of heteropentamers, is the key control hub of the ISR. Previously, a small molecule inhibitor of the ISR called ISRIB was found to bind to eIF2B and was proposed to reverse the impacts of eIF2 phosphorylation by increasing stabilizing the association of eIF2B heteropentamers into the functional decameric complex. However, more recently, an alternative model ISRIB action has been proposed. eIF2B is proposed to toggle between inactivate and active states. Binding of phosphorylated eIF2 to a regulatory site is proposed to trigger the inactive state by allosterically weakening binding of eIF2 at the active site. In the new model, ISRIB has been proposed to favor the active state conformation of eIF2B and thereby overcome the effects of eIF2 phosphorylation.

    In this paper, the authors further study a previously described H160D mutation in the eIF2Bbeta subunit. This mutation at one of the dimer interfaces in eIF2B was previously proposed to inhibit eIF2B by weakening dimerization. Consistent with this hypothesis, the H160D mutation impaired dimerization of eIF2B(beta, gamma, delta, epsilon) tetramers. However, in this study, the authors show that the H160D mutation does not impair dimerization when eIF2Balpha is included; thus, the mutation impairs eIF2B activity without impairing dimerization. Using biochemical assays, the authors show that the H160D mutation impairs nucleotide exchange by eIF2B decamers and weakens the binding eIF2 to eIF2B. However, the binding of phosphorylated eIF2 to eIF2B is not weakened.

    Cryo-EM structural analysis of the mutant eIF2B complex reveals a partial rocking of the decameric structure that resembles the structure of the eIF2B complex when bound to its inhibitor phosphorylated eIF2. In this partially rocked structure, both the ISRIB binding site at the dimer interface and the functional eIF2alpha binding sites are widened, providing a structural solution to why the mutation weakens eIF2 binding. Interestingly, the inhibitory binding site for phosphorylated eIF2 is not affected the H160D mutation. The authors propose that the H160D mutation in eIF2Bbeta induces an allosteric conformational change that mimics the effects of phosphorylated eIF2 binding to eIF2B.

    Finally, the authors generated cell lines that exclusively express the mutant eIF2Bbeta subunit. The mutation impairs total protein synthesis and cell growth rate and leads to elevated expression of the ISR marker ATF4.

    This is a high-quality study, the results are convincing and the authors conclusions are supported by the data. As the ISR has been implicated in a variety of diseases, further elucidation of the mechanism of action of eIF2B and ISRIB will be critical in the development of therapeutic interventions.

    A weakness of the paper (that hopefully can be easily remedied) would be to show the quality control data to verify the mutant cell lines used in Figure 6. It would be good to see that the mutant allele is present in the cells and that no WT alleles remain. In addition, examination of eIF2alpha Ser51 phosphorylation in Figure 6A would strengthen the conclusion that the eIF2Bbeta mutation is activating ATF4 expression independent of changes in eIF2 phosphorylation. Also, use of ATF4 reporters in Figure 6A, in addition to the presented Western data, would provide a nice quantitative read-out for the impact of the H160D mutation on ATF4 mRNA translation. Finally, as the biochemical and structural data indicate that the H160D mutation impairs ISRIB activity, it would be worthwhile testing whether ISRIB will rescue the slow-growth of the H160D cell lines in Figure 6D (the anticipation is that this slow-growth phenotype will not be rescued by ISRIB).

  4. eLife

    Reviewer #2 (Public Review):

    The integrated stress response (ISR) features eIF2 phosphorylation that invokes translational control in response to diverse stress conditions. The eIF2 combined with GTP is central for escorting the initiator tRNA to ribosomes and phosphorylation of eIF2 serves to repress its dedicated nucleotide exchange factor (eIF2B), those impairing the ability to recycle to the active eIF2-GTP that is required for global protein synthesis. There is much interest in the multisubunit eIF2B functions (five subunits in a decameric complex) including the mechanisms for its repression by phosphorylated eIF2 and small molecules render eIF2B insensitive to eIF2 phosphorylation or those that can repress eIF2B. This manuscript addresses the consequences of a missense mutation in eIF2B (beta subunit), with an eye towards delineating mechanisms regulating eIF2B in the ISR. Given that the important roles of eIF2B and the ISR in stress responses and different diseases, including neuropathologies, infectious and inflammatory diseases, and diabetes and metabolic disorders, this line of research is significant and is of broad interest.

    Critique:

    The beta subunit of eIF2B (H160D) was previously shown by this laboratory to block eIF2B tetramer assembly in vitro upon treatment with ISRIB. This study addressed how the residue substitution affects decamer formation and regulation of the ISR. Using a number of elegant biochemical and biophysical assays, along with cryo-EM imaging, the manuscript concludes that the H160D mutant retains the ability to form the decameric holoenzyme, but stabilizes eIF2B in an inhibited state, which would be typically induced by phosphorylated eIF2. This inhibited state is akin to the genetically defined gcd mutations in yeast by Hinnebusch and colleagues. Overall, the manuscript does support the activated/inhibited state eIF2B model for ISR regulation, although there are some concerns about the cryo-EM interpretations that should be considered and addressed. The manuscript will be of broad interest to the field.

  5. Version published to 10.7554/elife.76171 on eLife
  6. Version published to 10.1101/2021.12.03.471181v1 on bioRxiv