Choice of PP1 catalytic subunit affects neither the requirement for G-actin nor the insensitivity to Sephin1 of PPP1R15A-regulated eIF2αP dephosphorylation

The integrated stress response is regulated by kinases that phosphorylate translation initiation factor 2α and phosphatases that dephosphorylate it. Genetic and biochemical data indicate that the eIF2α-directed holophosphatase - a therapeutic target in diseases of protein misfolding - is comprised of a regulatory, PPP1R15, and a catalytic, Protein Phosphatase 1 (PP1), subunit. However, differing reports have appeared regarding the requirement for an additional co-factor, G-actin, in enabling substrate-specific de-phosphorylation1-3. An additional concern relates to the sensitivity of this PP1 holoenzyme to the [(o-chlorobenzylidene)amino]guanidines (Sephin1 or Guanabenz) small molecule proteostasis modulators3,4. We find that in the absence of G-actin, PPP1R15A regulatory subunit fragments were unable to accelerate eIF2α dephosphorylation beyond that affected by a catalytic subunit alone, whether PP1 was purified from rabbit muscle or from bacteria. Furthermore, we did not observe Sephin1 or Guanabenz inhibition of eIF2α dephosphorylation by any PPP1R15A-containing holophosphatase.


Introduction
The integrated stress response (ISR) is a signal transduction pathway that couples diverse stressful conditions to the activation of a rectifying translational and transcriptional program that is implicated in biological processes ranging from memory formation to immunity and metabolism (reviewed in Ref. 5). The mammalian ISR and its yeast counterpart (the general control response) are initiated by the phosphorylation of the α subunit of translation initiation factor 2 (eIF2α) on serine 51 6,7 and its activity is terminated by eIF2α P dephosphorylation.
Despite genetic evidence pointing to the sufficiency of the conserved C-terminal portion of PPP1R15 in reversing the eIF2α P -dependent ISR in vivo 2,8,9 , complexes formed in vitro between PPP1R15 regulatory subunit fragments and PP1 have been observed to unexpectedly lack specificity towards eIF2α P 2 . Dephosphorylation of eIF2α P is no faster by a complex of PPP1R15A-PP1 (or PPP1R15B-PP1) than by PP1 alone, showing that PP1R15A/B do not influence k cat or K m of PP1 towards the specific substrate eIF2α P 2 . However, addition of G-actin to the binary complex selectively accelerates eIF2α P dephosphorylation. G-actin binds directly to the Cterminus of PPP1R15 to form a ternary complex, whose affinity (K d~1 0 -8 M) matches V1.17F the EC 50 of G-actin's stimulatory effect 2,4 . The in vivo relevance of G-actin to eIF2α P dephosphorylation is attested to by the finding that actin sequestration in fibres (as Factin) enfeebles eIF2α P dephosphorylation, implying a role for factors that affect the actin cytoskeleton in ISR regulation 1 .
The ability to dephosphorylate eIF2α P is an essential function 16 . Nonetheless, inactivation of the PPP1R15A gene, which decelerates eIF2α P dephosphorylation and prolongs the ISR and has proven protective in certain cellular and animal models of diseases associated with enhanced unfolded protein stress [17][18][19][20] . This has generated interest in targeting the PPP1R15A-containing holophosphatase for inhibition by small molecules (reviewed in Ref. 21), an endeavour that requires detailed knowledge of the enzymatic mode of action.
A recent report published in Nature Structure and Molecular Biology 3 challenged the need for G-actin as a co-factor in PPP1R15A-mediated eIF2α P dephosphorylation.
Instead, it was suggested that a binary complex assembled from PP1α and a fragment of PPP1R15A (PPP1R15A 325-636 ), encompassing both the C-terminal PP1binding region and an N-terminal extension, dephosphorylates eIF2α P faster than PP1 alone 3 . Importantly, dephosphorylation of eIF2α P by this active binary complex was reported to be selectively inhibited by Guanabenz and Sephin1, two structurallyrelated small molecule proteostasis modifiers 22,23 . These findings contradict our observation that neither the non-selective PPP1R15A-PP1 binary complex, nor the eIF2α P -selective PPP1R15A-PP1-G-actin ternary complex were susceptible to these inhibitors 4,13 .
To establish if these discrepant findings reflected differences in enzyme subunit preparations or experimental regimes, we set out to reproduce the experiments in ref. 3

Results & Discussion
Native PP1 or bacterially-expressed PP1α require the presence of G-actin to promote PPP1R15A-regulated eIF2α P dephosphorylation PP1 produced in E. coli may differ in its enzymatic activity from PP1 purified from animal tissues, both in its substrate specificity and sensitivity to regulatory subunits (reviewed in ref. 24). To determine if the G-actin-dependence of PP1-PPP1R15Amediated eIF2α P dephosphorylation observed in our laboratory was a peculiarity of the bacterially-expressed PP1γ used 2,4 , we purified the native catalytic subunit of PP1 from rabbit skeletal muscle (PP1 N ), following an established protocol 25 , and compared the two PP1 preparations. Native PP1 (PP1 N ) is a mixture of PP1α, PP1β and PP1γ isoforms and gave rise to two prominent bands on SDS-PAGE ( Supplementary Fig.1a, left panel). The mass spectra of tryptic peptides derived from the PP1 N sample was analysed by Maxquant with iBAQ (intensity based absolute quant) to estimate the relative contribution of PP1 and the major contaminating species, tropomyosin, and enable a comparison of the catalytic subunit content of PP1 N preparation with the bacterially-expressed PP1γ, which served as a reference.  Supplementary Fig.1a). However, in this system, aimed to closely reproduce the experiments reported in ref. 3, eIF2α P dephosphorylation also exhibited a stringent requirement for both PPP1R15A and G-actin (Fig.1c). G-actin also exerted a saturable concentration-dependent stimulatory effect on the activity of a three-component holophosphatase constituted with native PP1 N (Fig.2c).
The EC 50 for G-actin with PP1 N (24 nM) was similar to that previously observed using To further promote comparability of the experimental conditions used here to those of ref. 3, we used PhosTag gels from the same commercial source (Alpha laboratories), and confirmed that the proteins used in our experiments exhibited the expected mobility for these gels (Supplementary Fig. 2c). Despite our best efforts we have V1.17F been unable to reproduce the stimulatory effect of MBP-PPP1R15A 325-636 on eIF2α P dephosphorylation.
Substrate recruitment by the PPP1R15A 325-512 region plays a secondary role in the kinetics of eIF2α P dephosphorylation and its reported disruption is unlikely to account for Sephin1's activity.
PPP1R15A interacts directly with eIF2α, both in cells 13  which exceeds by two-fold that required for a proteostatic effect in cultured cells 4,23 .

Conclusions:
The new experiments presented here cover a range of conditions with realistic

Protein expression and purification
The plasmids used to express protein in E. coli are presented in Supplementary  The stability test of PP1α (Supplementary Fig.2) was performed by preparing a fresh 240 nM solution of PP1α in the assay buffer described above. Separate aliquots were pre-incubated either at 30˚C or on ice for the specified times (30 minutes to 7 hours, see schema in Supplementary Fig.2a). At termination of the preincubation, 5 µL of these pre-incubated solutions were added into 20 µL dephosphorylation reactions as described above.
Reactions performed in Fig.5