dsRNA-induced condensation of antiviral proteins promotes PKR activation

Mammalian cells respond to dsRNA in multiple manners. One key response to dsRNA is the activation of PKR, an eIF2α kinase, which triggers translational arrest and the formation of stress granules. However, the process of PKR activation in cells is not fully understood. In response to increased endogenous or exogenous dsRNA, we observed that PKR forms novel cytosolic condensates, referred to as dsRNA-induced foci (dRIFs). dRIFs contain dsRNA, form in proportion to dsRNA, and are enhanced by longer dsRNAs. dRIFs also enrich several other dsRNA-binding proteins including ADAR1, Stau1, NLRP1, and PACT. Strikingly, dRIFs correlate with and form prior to translation repression by PKR and localize to regions of cells where PKR activation is initiated. We suggest that dRIF formation is a mechanism cells utilize to enhance the sensitivity of PKR activation in response to low levels of dsRNA, or to overcome viral inhibitors of PKR activation.

Heme-Regulated eIF2α Kinase (HRI) Inhibition Decreases PKR Activation in HUDEP2 Cells

Introduction: Sickle Cell Disease (SCD) is a group of inherited disorders caused by mutations in the β-globin gene which encodes the hemoglobin subunit β in erythrocytes [1]. Hemoglobin containing the mutant β-globin polymerizes and causes sickling of erythrocytes, which subsequently leads to vaso-occlusion, hemolysis, and activation of the immune system by release of free heme [2]. Heme-regulated eIF2α kinase, also known as heme-regulated inhibitor (HRI/EIF2AK1), the dsRNA-activated kinase Protein Kinase R (PKR/EIF2AK2), PKR-like endoplasmic reticulum kinase (PERK/EIF2AK3) and General Control Non-Depressible 2 (GCN2/EIF2AK4) are four kinases belonging to the eIF2α kinases family and play key functions in the Integrated Stress Response-ATF4 pathway, which is critical for translational control in response to various stress conditions [3]. These kinases are activated respectively by heme depletion, viral infection, endoplasmic reticulum stress, and amino acid starvation, and they phosphorylate eukaryotic initiation factor-2α (eIF2α). Recently, it has been shown that HRI inhibition induces fetal hemoglobin in HUDEP-2 cells and CD34+ hematopoietic progenitor stem cells and prevents sickling, suggesting HRI as a potential therapeutic target for SCD [4, 5]. Moreover, it has been well documented that kinase inhibition can activate compensatory loops (bypass signaling) to circumvent the inhibited target, in particular by overexpression and activation of kinases having the same substrate [6, 7]. Thus, in this study, we investigated if the inhibition of HRI in HUDEP-2 cells lead to compensatory mechanisms by modulation of the expression and activation of the other eIF2α kinases. Methods: To inhibit HRI, we generated HUDEP-2 [8] HRI Knock-Out cells (HRI-KO) and HUDEP-2 HRI Kinase Dead cells (K196R) (clones HRI-KD1 and HRI-KD2) by using CRISPR-Cas9 gene editing technology [9]. We confirmed the induction of fetal hemoglobin for each clone by flow cytometry. To evaluate a potential compensatory mechanism, we measured the effects of HRI inhibition on the expression and activation of the other eIF2α kinases by western blot (WB) and the regulation at the transcriptomic level by qPCR. Based on the results of preliminary studies, we generated HUDEP-2 PKR Knock-Out cells (PKR-KO) by CRISPR-Cas9. We differentiated them for 7 days and we quantified the level of fetal hemoglobin by flow cytometry, AlphaLISA® and WB. Results: HRI-KO, HRI-KD1 and HRI-KD2 clones expressed fetal hemoglobin after 7-day of differentiation consistent with published data [4]. HRI inhibition did not result in any modulation of PKR protein expression, but the activation of PKR, measured by phosphorylation at its residue Threonine 446, was decreased in HRI-KD1, HRI-KD2, and in HRI-KO cells at day 0 and day 7 of differentiation. As PKR and HRI have the same downstream target eIF2α and HRI inhibition induces fetal hemoglobin through eIF2α-ATF4-Bcl11a axis, we verified if fetal hemoglobin induction is due to HRI inhibition exclusively and is not a consequence of the decrease in PKR activation when HRI is inhibited. We measured the protein expression level of fetal hemoglobin in PKR-KO cells and results obtained by flow cytometry, western blot and AlphaLISA® did not show any regulation in fetal hemoglobin in PKR-KO cells after 7 days of differentiation. Finally, HRI inhibition did not result in any regulation of kinases PERK and GCN2 activation and expression, at the RNA and protein level. The expression of these two eIF2α kinases was low compared to HRI and PKR. Conclusion: HRI inhibition does not cause any modulation in the expression and activation of GCN2 and PERK in HUDEP-2 cells but results in a decrease in PKR activation. This outcome could be explained by a possible increase in the expression of proteins that inhibit PKR, such as TRBP or Hsp40, and are induced by HRI silencing (Hsp40) [10-12]. Nonetheless PKR inhibition does not induce fetal hemoglobin in HUDEP-2 cells after 7 days of differentiation. Overall, this study provides evidence that fetal hemoglobin induction by HRI inhibition in HUDEP-2 cells is independent from the other eIF2α kinases and supports HRI as a potential therapeutic target in SCD. However, the biological implications of a potential compensatory effect on PKR signaling in HRI-expressing tissues warrant further investigation. Disclosures Krishnamoorthy: Cellarity, Inc.: Current Employment, Current holder of stock options in a privately-held company.

Higher order phosphatase-substrate contacts terminate the Integrated Stress Response

Many regulatory PPP1R subunits join few catalytic PP1c subunits to mediate phosphoserine and phosphothreonine dephosphorylation in metazoans. Regulatory subunits are known to engage PP1c's surface, locally affecting flexible phosphopeptides access to the active site. However, catalytic efficiency of holophosphatases towards their natively-folded phosphoprotein substrates is largely unexplained. Here we present a Cryo-EM structure of the tripartite PP1c/PPP1R15A/G-actin holophosphatase that terminates signalling in the Integrated Stress Response (ISR) in pre-dephosphorylation complex with its substrate, translation initiation factor 2α (eIF2α). G-actin's role in eIF2α dephosphorylation is supported crystallographically by the structure of the binary PPP1R15A-G-actin complex, and by biochemical and genetic confirmation of the essential role of PPP1R15A-G-actin contacts to eIF2αP dephosphorylation. In the pre-dephosphorylation CryoEM complex, G-actin aligns the catalytic and regulatory subunits, creating a composite surface that engages eIF2α's N-terminal domain to position the distant phosphoserine-51 at the active site. eIF2α residues specifying affinity for the holophosphatase are confirmed here to make critical contacts with the eIF2α kinase PERK. Thus, a convergent process of higher-order substrate recognition specifies functionally-antagonistic phosphorylation and dephosphorylation in the ISR.

eIF2α integrates proteotoxic signals both from ER and cytoplasm: Hsp70-Bag3 module regulates HRI-dependent phosphorylation of eIF2α

The major heat shock protein Hsp70 has been implicated in many stages of cancer development. These effects are mediated by a scaffold protein Bag3 that binds to Hsp70 and links it to components of multiple cancer-related signaling pathways. Accordingly, the Hsp70-Bag3 complex has been targeted by small molecules, which showed strong anti-cancer effects. Here, our initial question was how JG-98, an allosteric inhibitor of Hsp70 that blocks its interaction with Bag3, causes cell death. Breast epithelial cells MCF10A transformed with a single oncogene Her2 showed higher sensitivity to JG-98 then parental MCF10A cells. RNA expression analysis showed that this enhanced sensitivity correlated with higher induction of the UPR genes. Indeed, depletion of the pro-apoptotic UPR responsive transcription factor CHOP significantly protected cells from JG-98. Surprisingly, only the eIF2α-associated branch of the UPR was activated by JG-98, suggesting that the response was not related to the ER proteotoxicity. Indeed, it was dependent on activation of a distinct cytoplasmic eIF2α kinase HRI. HRI-dependent phosphorylation of eIF2α was also activated by the cytoplasmic proteotoxicity via Hsp70-Bag3 complex, which directly associates with HRI. Dissociation of Hsp70-Bag3 complex led to Bag3-dependent degradation of HRI via autophagy. Therefore, eIF2α integrates proteotoxicity signals from both ER and cytoplasm, and the cytoplasmic response mediates cytotoxicity of the Hsp70-Bag3 inhibitors.

A KDM6 inhibitor potently induces ATF4 and its target gene expression through HRI activation and by UTX inhibition

UTX/KDM6A encodes a major histone H3 lysine 27 (H3K27) demethylase, and is frequently mutated in various types of human cancers. Although UTX appears to play a crucial role in oncogenesis, the mechanisms involved are still largely unknown. Here we show that a specific pharmacological inhibitor of H3K27 demethylases, GSK-J4, induces the expression of transcription activating factor 4 (ATF4) protein as well as the ATF4 target genes (e.g. PCK2, CHOP, REDD1, CHAC1 and TRIB3). ATF4 induction by GSK-J4 was due to neither transcriptional nor post-translational regulation. In support of this view, the ATF4 induction was almost exclusively dependent on the heme-regulated eIF2α kinase (HRI) in mouse embryonic fibroblasts (MEFs). Gene expression profiles with UTX disruption by CRISPR-Cas9 editing and the following stable re-expression of UTX showed that UTX specifically suppresses the expression of the ATF4 target genes, suggesting that UTX inhibition is at least partially responsible for the ATF4 induction. Apoptosis induction by GSK-J4 was partially and cell-type specifically correlated with the activation of ATF4-CHOP. These findings highlight that the anti-cancer drug candidate GSK-J4 strongly induces ATF4 and its target genes via HRI activation and raise a possibility that UTX might modulate cancer formation by regulating the HRI-ATF4 axis.

Unfolded protein response in colorectal cancer

Colorectal cancer (CRC) is a gastrointestinal malignancy originating from either the colon or the rectum. A growing number of researches prove that the unfolded protein response (UPR) is closely related to the occurrence and progression of colorectal cancer. The UPR has three canonical endoplasmic reticulum (ER) transmembrane protein sensors: inositol requiring kinase 1 (IRE1), pancreatic ER eIF2α kinase (PERK), and activating transcription factor 6 (ATF6). Each of the three pathways is closely associated with CRC development. The three pathways are relatively independent as well as interrelated. Under ER stress, the activated UPR boosts the protein folding capacity to maximize cell adaptation and survival, whereas sustained or excessive ER triggers cell apoptosis conversely. The UPR involves different stages of CRC pathogenesis, promotes or hinders the progression of CRC, and will pave the way for novel therapeutic and diagnostic approaches. Meanwhile, the correlation between different signal branches in UPR and the switch between the adaptation and apoptosis pathways still need to be further investigated in the future.

PKR and the Integrated Stress Response drive immunopathology caused by ADAR1 mutation

SummaryMutations in ADAR, the gene that encodes the ADAR1 RNA deaminase, cause numerous human diseases, including Aicardi-Goutières Syndrome (AGS). ADAR1 is an essential negative regulator of the RNA sensor MDA5, and loss of ADAR1 function triggers inappropriate activation of MDA5 by self-RNAs. However, the mechanisms of MDA5-dependent disease pathogenesis in vivo remain unknown. Here, we introduce a knockin mouse that models the most common ADAR AGS mutation in humans. These Adar-mutant mice develop lethal disease that requires MDA5, the RIG-I-like receptor LGP2, type I interferons, and the eIF2α kinase PKR. We show that a small molecule inhibitor of the integrated stress response (ISR) that acts downstream of eIF2α phosphorylation prevents immunopathology and rescues the mice from mortality. These findings place PKR and the ISR as central components of immunopathology in vivo and identify new therapeutic targets for treatment of human diseases associated with the ADAR1-MDA5 axis.

The Erythropoietin Receptor Stimulates Rapid Cycling and Formation of Larger Red Cells During Mouse and Human Erythropoiesis

Erythroid terminal differentiation entails cell divisions that are coupled to progressive decreases in cell size. EpoR signaling is essential for the survival of erythroid precursors, but it is unclear whether it has other functions in these cells. Here we endowed mouse precursors that lack the EpoR with survival signaling, finding that this was sufficient to support their differentiation into enucleated red cells, but that the process was abnormal. Precursors underwent fewer and slower cell cycles and yet differentiated into smaller red cells. Surprisingly, EpoR further accelerated cycling of early erythroblasts, the fastest cycling cells in the bone marrow, while simultaneously increasing their cell size. EpoR-mediated formation of larger red cells was independent of the established pathway regulating red cell size by iron through Heme-regulated eIF2α kinase (HRI). We confirmed the effect of Epo on red cell size in human volunteers, whose mean corpuscular volume (MCV) increased following Epo administration. This increase persisted after Epo declined and was not the result of increased reticulocytes. Our work reveals a unique effect of EpoR signaling on the interaction between the cell cycle and cell growth. Further, it suggests new diagnostic interpretations for increased red cell volume, as reflecting high Epo and erythropoietic stress.

The eIF2α kinase HRI triggers the autophagic clearance of cytosolic protein aggregates

Large cytosolic protein aggregates are removed by two main cellular processes, autophagy and the ubiquitin-proteasome system (UPS), and defective clearance of these protein aggregates results in proteotoxicity and cell death. Recently, we found that the eIF2α kinase heme-regulated inhibitory (HRI) induced a cytosolic unfolded protein response (cUPR) to prevent aggregation of innate immune signalosomes, but whether HRI acts as a general sensor of proteotoxicity in the cytosol remains unclear. Here we show that HRI controls autophagy to clear cytosolic protein aggregates when the UPS is inhibited. We further report that silencing HRI expression resulted in decreased levels of BAG3 and HSPB8, two proteins involved in chaperone-assisted selective autophagy (CASA), suggesting that HRI controls proteostasis in the cytosol at least in part through CASA. Moreover, knocking down the expression of HRI resulted in cytotoxic accumulation of over-expressed α-synuclein, a protein known to aggregate in Parkinson’s disease, dementia with Lewy bodies, and multiple system atrophy. In agreement with these data, protein aggregate accumulation and microglia activation were observed in the spinal cord white matter of 7-month old Hri-/- mice as compared to Hri+/+ littermates. Moreover, aged Hri-/- mice showed accumulation of misfolded α-synuclein, indicative of misfolded proteins, in the lateral collateral pathway, a region of the sacral spinal cord horn that receives visceral sensory afferents from the bladder and distal colon, a pathological feature common to α-synucleinopathies in humans. Together, these results suggest that HRI contributes to a general cUPR that could be leveraged to bolster the clearance of cytotoxic protein aggregates.