How an intrinsically disordered regulatory subunit assembles a PP1:eIF2 complex

Fatalska et al.1Fatalska A. Hodgson G. Freund S.M.V. Maslen S.L. Morgan T. Thorkelsson S.R. van Slegtenhorst M. Lorenz S. Andreeva A. Kaat L.D. Bertolotti A. Recruitment of trimeric eIF2 by phosphatase non-catalytic subunit PPP1R15B.Mol. Cell. 2024; 84: 506-521.e11https://doi.org/10.1016/j.molcel.2023.12.011Abstract Full Text Full Text PDF PubMed Scopus (1) Google Scholar use an interdisciplinary strategy to elucidate how an intrinsically disordered regulatory subunit of protein phosphatase 1 binds trimeric eIF2 and positions the phosphatase-substrate complex for dephosphorylation. As validation, they show that a disease mutation abolishes the interaction. Fatalska et al.1Fatalska A. Hodgson G. Freund S.M.V. Maslen S.L. Morgan T. Thorkelsson S.R. van Slegtenhorst M. Lorenz S. Andreeva A. Kaat L.D. Bertolotti A. Recruitment of trimeric eIF2 by phosphatase non-catalytic subunit PPP1R15B.Mol. Cell. 2024; 84: 506-521.e11https://doi.org/10.1016/j.molcel.2023.12.011Abstract Full Text Full Text PDF PubMed Scopus (1) Google Scholar use an interdisciplinary strategy to elucidate how an intrinsically disordered regulatory subunit of protein phosphatase 1 binds trimeric eIF2 and positions the phosphatase-substrate complex for dephosphorylation. As validation, they show that a disease mutation abolishes the interaction. Understanding how post-translational modifications change the function, localization, and/or lifetime of a target protein is a significant challenge. How phosphorylation sites are reversibly added/removed by specific kinases and phosphatases is still being defined. The recognition of specific substrates is complicated by the fact that many kinases and phosphatases, as well as their substrates, are multimeric or multidomain proteins. Kinases and phosphatases also rely heavily on intrinsically disordered proteins (IDPs) and intrinsically disordered regions (IDRs). Embedded in these IDPs/IDRs are short linear motifs (SLiMs) and short helical motifs (SHELMs) that contribute to the assembly and tethering of multimeric complexes2Olivieri C. Wang Y. Li G.C. V S.M. Kim J. Stultz B.R. Neibergall M. Porcelli F. Muretta J.M. Thomas D.D. et al.Multi-state recognition pathway of the intrinsically disordered protein kinase inhibitor by protein kinase A.eLife. 2020; 9e55607https://doi.org/10.7554/eLife.55607Crossref PubMed Scopus (14) Google Scholar,3Padi S.K.R. Vos M.R. Godek R.J. Fuller J.R. Kruse T. Hein J.B. Nilsson J. Kelker M.S. Page R. Peti W. Cryo-EM structures of PP2A:B55–FAM122A and PP2A:B55–ARPP19.Nature. 2024; 625: 195-203https://doi.org/10.1038/s41586-023-06870-3Crossref PubMed Scopus (4) Google Scholar or to the reorganization of multidomain proteins.4Gógl G. Kornev A.P. Reményi A. Taylor S.S. Disordered Protein Kinase Regions in Regulation of Kinase Domain Cores.Trends Biochem. Sci. 2019; 44: 300-311https://doi.org/10.1016/j.tibs.2018.12.002Abstract Full Text Full Text PDF PubMed Scopus (37) Google Scholar Recent work published in Molecular Cell by Fatalska et al.1Fatalska A. Hodgson G. Freund S.M.V. Maslen S.L. Morgan T. Thorkelsson S.R. van Slegtenhorst M. Lorenz S. Andreeva A. Kaat L.D. Bertolotti A. Recruitment of trimeric eIF2 by phosphatase non-catalytic subunit PPP1R15B.Mol. Cell. 2024; 84: 506-521.e11https://doi.org/10.1016/j.molcel.2023.12.011Abstract Full Text Full Text PDF PubMed Scopus (1) Google Scholar adds to our understanding of how phosphatases recruit their substrate proteins and the importance of IDPs by demonstrating how the intrinsically disordered regulatory subunit PPP1R15B (R15B) utilizes SHELMs to tether the protein phosphatase 1 catalytic (PP1c) subunit to its candidate substrate, eIF2α, for dephosphorylation of Ser51 (Figure 1A). Resolving the structural elements that place the active site of PP1c adjacent to Ser51 depends critically on having the full trimeric eIF2 complex, including eIF2γ, which contains the pivotal R15B binding sites. The Bertolotti team used a multidisciplinary approach that includes biochemistry, hydrogen-deuterium exchange mass spectrometry (HDX-MS), nuclear magnetic resonance (NMR), and AlphaFold predictions.5Jumper J. Evans R. Pritzel A. Green T. Figurnov M. Ronneberger O. Tunyasuvunakool K. Bates R. Žídek A. Potapenko A. et al.Highly accurate protein structure prediction with AlphaFold.Nature. 2021; 596: 583-589https://doi.org/10.1038/s41586-021-03819-2Crossref PubMed Scopus (14791) Google Scholar To validate this polyvalent recruitment strategy, the authors used a variant that drives a rare but devastating neurodevelopmental disease. Understanding phosphatase-substrate preference has remained largely underexplored. eIF2α, phosphorylated on Ser51 by protein kinase R, was one of the first substrates to be crystallized with its cognate kinase (Figure 1B). This structure demonstrated the importance of distal tethering sites that bring the P site in proximity to the kinase active site. How a phosphatase recognizes its substrates is more complex. In contrast to kinases, which all belong to a single conserved superfamily, there are several distinct phosphatase families, and most are multimeric. A plethora of regulatory subunits thus modulate the regulation/targeting mechanisms that underpin dephosphorylation. In the late 1990s, the RVxF motif, now recognized as a SLiM, was reported to recruit PP1c to its regulatory subunits.6Egloff M.-P. Johnson D.F. Moorhead G. Cohen P.T. Cohen P. Barford D. Structural basis for the recognition of regulatory subunits by the catalytic subunit of protein phosphatase 1.EMBO J. 1997; 16: 1876-1887https://doi.org/10.1093/emboj/16.8.1876Crossref PubMed Scopus (551) Google Scholar Research into other Ser/Thr phosphatases, protein phosphatase 2A (PP2A)3Padi S.K.R. Vos M.R. Godek R.J. Fuller J.R. Kruse T. Hein J.B. Nilsson J. Kelker M.S. Page R. Peti W. Cryo-EM structures of PP2A:B55–FAM122A and PP2A:B55–ARPP19.Nature. 2024; 625: 195-203https://doi.org/10.1038/s41586-023-06870-3Crossref PubMed Scopus (4) Google Scholar,7Fowle H. Zhao Z. Xu Q. Wasserman J.S. Wang X. Adeyemi M. Feiser F. Kurimchak A.N. Atar D. McEwan B.C. et al.PP2A/B55α substrate recruitment as defined by the retinoblastoma-related protein p107.eLife. 2021; 10e63181https://doi.org/10.7554/eLife.63181Crossref PubMed Scopus (14) Google Scholar and calcineurin,8Wigington C.P. Roy J. Damle N.P. Yadav V.K. Blikstad C. Resch E. Wong C.J. Mackay D.R. Wang J.T. Krystkowiak I. et al.Systematic Discovery of Short Linear Motifs Decodes Calcineurin Phosphatase Signaling.Mol. Cell. 2020; 79: 342-358.e12https://doi.org/10.1016/j.molcel.2020.06.029Abstract Full Text Full Text PDF PubMed Scopus (35) Google Scholar also shows how SLiMs and SHELMs define substrate selectivity. Understanding how specific substrates are recognized and recruited by PP1 holoenzyme complexes requires structural approaches, coupled with biochemical and cellular validation, to reveal binding sites crucial for positioning substrates for dephosphorylation. Here, Fatalska et al.1Fatalska A. Hodgson G. Freund S.M.V. Maslen S.L. Morgan T. Thorkelsson S.R. van Slegtenhorst M. Lorenz S. Andreeva A. Kaat L.D. Bertolotti A. Recruitment of trimeric eIF2 by phosphatase non-catalytic subunit PPP1R15B.Mol. Cell. 2024; 84: 506-521.e11https://doi.org/10.1016/j.molcel.2023.12.011Abstract Full Text Full Text PDF PubMed Scopus (1) Google Scholar elucidated how PP1-R15B recognizes and dephosphorylates eIF2α. R15B, the regulatory subunit of PP1, is a classic IDP, and AlphaFold predicts that there are several SHELMs embedded in R15B (Figure 1C). The authors confirmed the helicity of the SHELMs and linked them to a putative function. The authors validated these new sites by using a truncated R15B that lacks the RVxF motif responsible for binding PP1c. In vitro incubation of the full-length eIF2 trimer with this truncated R15B eluted as a complex with size-exclusion chromatography. These results established that R15B binds eIF2 independently of PP1c and forms a stable complex in vitro. To investigate putative binding regions on eIF2, the authors utilized HDX-MS in the presence or absence of R15B414–613. eIF2α peptides immediately surrounding Ser51 were not protected; instead, the C-terminal lobe of eIF2α and two regions on eIF2γ had slow exchange rates. By lowering the pH to overcome the rapid exchange rate of R15B, the authors were able to map three regions of R15B (H1, H2, and H3) that showed slower exchange in the presence of eIF2. Although R15B is largely disordered, NMR revealed three α-helical regions that correlated with those computationally predicted by AlphaFold (Figure 1C). These sites coincided with changes in relaxation properties when R15B was co-incubated with eIF2, confirming their importance for protein-protein interactions. These helices map back to the protected regions identified by HDX-MS. By independently replacing each of the three helical motifs with alanine, the authors showed that the H1 helix is required for co-immunoprecipitation of eIF2α. Perturbation of H1 also reduced dephosphorylation of Ser51 on eIF2α, consistent with a model where R15B recruits eIF2 to be dephosphorylated by PP1c. Further, the H1 mutation (N423D), associated with a neurodevelopmental disorder, reduced capture of eIF2α and dephosphorylation of Ser51. This underscores the importance of the polyvalent phosphatase-substrate recruitment mechanism and why utilizing full multimeric complexes is essential. While IDPs are challenging from a structural standpoint, Fatalska et al.1Fatalska A. Hodgson G. Freund S.M.V. Maslen S.L. Morgan T. Thorkelsson S.R. van Slegtenhorst M. Lorenz S. Andreeva A. Kaat L.D. Bertolotti A. Recruitment of trimeric eIF2 by phosphatase non-catalytic subunit PPP1R15B.Mol. Cell. 2024; 84: 506-521.e11https://doi.org/10.1016/j.molcel.2023.12.011Abstract Full Text Full Text PDF PubMed Scopus (1) Google Scholar met these challenges head on and deciphered the PP1:R15B-eIF2 binding code. The R15B-eIF2 tether relies on small, separate helical elements that poise PP1c for dephosphorylation of Ser51, a mechanism that would not have been revealed without using the complete multimeric substrate. Other substrates may also take advantage of these helices or other motifs to facilitate proper conformational alignment for PP1-mediated dephosphorylation. Global phosphoproteomics could establish the generality and cellular importance of these R15B-targeting motifs. Using AlphaFold to predict small motifs in R15B that mediate protein-protein interfaces is also striking and similar to AlphaFold predictions for PP2A-B55.9Kruse T. Garvanska D.H. Varga J. Garland W. McEwan B. Hein J.B. Weisser M.B. Puy I.B. Chan C.B. Parrila P.S. et al.Substrate recognition principles for the PP2A-B55 protein phosphatase.bioRxiv. 2024; (Preprint at)https://doi.org/10.1101/2024.02.10.579793Crossref Scopus (0) Google Scholar Although the authors showed that H1 is critical for recruitment of eIF2, a residue-specific model for H1's docking to eIF2γ is missing. Such a model, coupled with alanine scanning, would provide a more detailed mechanism for H1 binding and N423D disruption. The dynamics of complex assembly provide another future challenge. Does H1 dock first, followed by H2/3? This mechanism would be reminiscent of a PP2A-B56 substrate-recruitment strategy where high-affinity binding of an essential SLiM is followed by a secondary motif that enhances interaction.10Wang X. Garvanska D.H. Nasa I. Ueki Y. Zhang G. Kettenbach A.N. Peti W. Nilsson J. Page R. A dynamic charge-charge interaction modulates PP2A:B56 substrate recruitment.eLife. 2020; 9e55966https://doi.org/10.7554/eLife.55966Crossref Scopus (33) Google Scholar Molecular dynamics simulations and local Brownian dynamics will further define this dynamic process. Are the kinase and phosphatase part of a stable "signaling island" where the phosphorylation state of Ser51 is regulated by the dynamic properties of the respective kinase and phosphatase? Or are there separate phosphorylation/dephosphorylation islands? In any case, one of the fundamental lessons taught by this study is that understanding dynamic protein phosphorylation/dephosphorylation requires the entire complex, not just catalytic domains. Fatalska et al.1Fatalska A. Hodgson G. Freund S.M.V. Maslen S.L. Morgan T. Thorkelsson S.R. van Slegtenhorst M. Lorenz S. Andreeva A. Kaat L.D. Bertolotti A. Recruitment of trimeric eIF2 by phosphatase non-catalytic subunit PPP1R15B.Mol. Cell. 2024; 84: 506-521.e11https://doi.org/10.1016/j.molcel.2023.12.011Abstract Full Text Full Text PDF PubMed Scopus (1) Google Scholar highlight an IDP-driven tethering mechanism that facilitates phosphatase-substrate recruitment and point out the role of short motifs embedded in the IDPs that can drive the assembly of a multimeric protein complex. Signaling islands such as these reduce dimensionality and no longer describe reversible phosphorylation as a diffusion-limited process. These results exemplify the utility of biological "organized chaos," where a protein that is largely disordered can use SLiMs and SHELMs to bridge and position two highly structured proteins. S.S.T was supported by the National Institutes of Health R35 GM130389. A.C.J. was supported in part by the UCSD Graduate Training Program in Cellular and Molecular Pharmacology (T32 GM007752) and an NSF Graduate Research Fellowship (DGE-1650112). The authors declare no competing interests. Recruitment of trimeric eIF2 by phosphatase non-catalytic subunit PPP1R15BFatalska et al.Molecular CellDecember 29, 2023In BriefFatalska et al. revealed how the intrinsically disordered PP1 non-catalytic subunit PPP1R15B (R15B) captures its full trimeric eIF2 substrate by binding the far end of the complex relative to the phosphosite using three short helical elements. A homozygote variant in R15B impaired in substrate binding causes microcephaly. Full-Text PDF Open Access