Hsp90?mediated regulation of DYRK3 couples stress granule disassembly and growth via mTORC1 signaling

Article19 March 2021Open Access Source DataTransparent process Hsp90-mediated regulation of DYRK3 couples stress granule disassembly and growth via mTORC1 signaling Laura Mediani Department Biomedical, Metabolic Neural Sciences, Centre for Neuroscience Nanotechnology, University Modena Reggio Emilia, Modena, Italy Search more papers by this author Francesco Antoniani ItalyThese authors are contributed equally to work as second, third Veronica Galli Jonathan Vinet Genomic Post-Genomic Center, IRCCS Mondino Foundation, Pavia, Arianna Dorotea Carrà Ilaria Bigi Vadreenath Tripathy Center Regenerative Therapies TU Dresden, Technische Universität Germany Tatiana Tiago Marco Cimino Giuseppina Leo Triana Amen orcid.org/0000-0003-4808-7806 Experimental Neurodegeneration, Medical Göttingen, Daniel Kaganovich orcid.org/0000-0003-2398-1596 Cristina Cereda orcid.org/0000-0001-9571-0862 Orietta Pansarasa Jessica Mandrioli Neuroscience, St. Agostino Estense Hospital, Azienda Ospedaliero Universitaria di Priyanka Tripathi Institute Neuropathology, RWTH Aachen Aachen, Dirk Troost (Neuro)Pathology, Amsterdam UMC, Amsterdam, The Netherlands Eleonora Aronica Johannes Buchner orcid.org/0000-0003-1282-7737 Integrated Protein Science Munich at the Chemie, München, Garching, Anand Goswami Jared Sterneckert Simon Alberti Corresponding Author [email protected] orcid.org/0000-0003-4017-6505 Biotechnology (BIOTEC), Molecular Cellular Bioengineering (CMCB), Serena Carra orcid.org/0000-0003-0939-0140 Information Mediani1, Antoniani1, Galli1, Vinet1,2, Carrà1, Bigi1, Tripathy3, Tiago1, Cimino1, Leo1, Amen4, Kaganovich4, Cereda2, Pansarasa2, Mandrioli5, Tripathi6, Troost7, Aronica7, Buchner8, Goswami6, Sterneckert3, *,9 *,1 1Department 2Genomic 3Center 4Department 5Department 6Institute 7Department 8Center 9Biotechnology *Corresponding author. Tel: +49 351 46340243; E-mail: (lead contact). +39 059 2055265; EMBO Reports (2021)22:e51740https://doi.org/10.15252/embr.202051740 PDFDownload PDF article text main figures. Peer ReviewDownload a summary editorial decision including letters, reviewer comments responses feedback. ToolsAdd favoritesDownload CitationsTrack CitationsPermissions ShareFacebookTwitterLinked InMendeleyWechatReddit Figures & Info Abstract Stress granules (SGs) dynamic condensates associated with protein misfolding diseases. They sequester stalled mRNAs factors, such subunit raptor, suggesting that SGs coordinate cell during after stress. However, molecular mechanisms linking SG dynamics remain undefined. We report chaperone Hsp90 is required dissolution. binds stabilizes dual-specificity tyrosine-phosphorylation-regulated kinase 3 (DYRK3) in cytosol. Upon inhibition, dissociates from becomes inactive. Inactive subjected two different fates: it either partitions into SGs, where protected irreversible aggregation, or degraded. In presence Hsp90, active promotes disassembly, restoring translation. Thus, links adaptation regulating activity key involved condensate translation restoration. SYNOPSIS other part binding stabilizing (DYRK3). Inhibition loss results destabilization, persistence failure restore regulates stability activity. Partitioning protects aggregation. mTORC1-dependent DYRK3. ALS fibroblasts motor neurons show reduced expression Introduction ribonucleoprotein assemblies form response conditions transiently translationally along 40S ribosomal subunits, enzymes, molecules, p90 S6 raptor (Kedersha et al, 2013). Raptor-mediated sequestration inside cells apoptosis (Thedieck relief, dissolution, subsequent release reactivation complex (Wippich This suggests hubs metabolism changes environmental conditions, thus enabling stressors details how cellular assembly dissolution have so far eluded us. Recent impaired dysfunction death. dysfunctionality emerging an important pathomechanism number age-related neurodegenerative diseases amyotrophic lateral sclerosis (ALS), frontotemporal lobar degeneration (FTD), Alzheimer disease (AD). these diseases, disassemble has been linked disturbances RNA development aggregates cause neuronal death (Zhang 2019). Hence, identifying characterizing factors govern will provide targets therapeutic potential treat uncurable Surprisingly, although we gained detailed knowledge on promote 1999; Matsuki 2013; Van Treeck 2018), still poorly understood. RNA-binding proteins (RBPs) high concentrations free crucial drive (Bounedjah 2012; Ciryam 2015; Kroschwald 2018; Guillen-Boixet 2020; Sanders Yang 2020). many RBPs assemble propensity misfold aggregate, which can affect material properties their ability dissolve (Molliex Patel 2015). Such aberrant also arise accumulation misfolding-prone proteins, defective products (DRiPs) ALS-linked C9orf72 dipeptide repeat (Ganassi 2016; Lee Mateju 2017; As result, under close surveillance quality control (PQC) system. PQC system includes chaperones, recognize aberrantly folded refold misfolded well ubiquitin–proteasome autophagy degradation systems, clear (Balchin 2016). Indeed, chaperones Hsp70 valosin-containing (VCP) able prevent maintaining 2017). addition, SGs. When machinery fails, persisting targeted proteasomal involving p62/SQSTM1 ZFAND1/VCP (Buchan Chitiprolu Turakhiya 2018). our understanding its infancy do not know whether maintain healthy state. One essential ubiquitous folding maturation large variety components, (Jain Markmiller suggested regulate processing bodies (PBs) (Matsumoto 2011), cytosolic RNA-protein mRNA storage (Standart Weil, Moreover, promotes, through yet unknown mechanism, recruitment argonaute 2, eIF4E, partner eIF4E transporter (4E-T) (Pare 2009; Suzuki 2009). agreement this, recent proteomic analysis fungal pathogen identified novel role PB components (O'Meara despite well-known function tolerance, unclear functionality. Here, use genetics fluorescence microscopy fixed live study functional effects mammalian cells. Results delays Hsp70-independent manner cells, majority recovery phase, only minor fraction (< 5–10%) clearance Using time-lapse HeLa-Kyoto expressing marker G3BP2-GFP, confirmed most sodium arsenite induced when removed; dissolved regardless lysosomal proteases were inactivated ammonium chloride (Fig EV1A Movie EV1), does play major role. By contrast, inhibition ATPase VER-155008 resulted (VER; Figs 1A EV1B, literature was absolute: could inhibitor, slower kinetics persisted ca. 20–30% 4 h removal. Combined data suggest regulated additional besides Hsp70. Click here expand figure. Figure EV1. independently Related Fig 1 A. Kinetics living absence chloride. G3BP2-GFP treated (50 µM) 45 min. Then, allowed recover drug-free medium (recovery control) (NH4Cl, 20 mM). Images taken over time period every 10 Dashed lines = 95% confidence intervals. Number counted: 193 (Recovery Control); 104 B. left untreated (control) GA (5 µM), 17AAG VER (40 h. Cells fixed, percentage counted. n independent experiments, ± s.e.m.; 126–185 counted/sample. n.s.: non-significant (one-way ANOVA). C, D. HEK293 express SG-resident PABP endogenously tagged fluorescent probe Dendra2 (PABPC1-Dendra2). PABPC1-Dendra2 HEK293K (100 mM), µM). 299 control); 160 mM); 361 (GA 5 µM); (17AAG 259 (VER 40 E. Confocal showing colocalize DRiPs. HeLa OP-puro (25 DRiPs visualized click chemistry, while Hsc70, Hsp70, ?, ? immunostaining. Scale bar µm. F. Quantitation DRiP enrichment Automated imaging segmentation based G3BP signal. Data presented histogram. HS 43.5°C alone segmented: 2,198 (control); 3,056 (VER); 1,484 (GA); 3,220 (17AAG); P < 10?10 G, H. stably V5-tagged inducible tetracycline (Flp-In) cultured (V5-HSP70 OFF V5-HSP70 ON, respectively). (G) extracts prepared 24 treatment, levels analyzed immunoblotting. TUBA4A used loading control. (H) (500 min, followed 90 then 115-378 counted/sample; I. h, respectively. stained endogenous DCP1A, P-bodies. Nucleic acid DAPI. P-bodies treatment shown. h: 104–124 counted/sample, 0.02; 111–178 10?7 J. lipofected cDNA encoding mRFP-DCP1A. post-transfection, 2 Representative images GFP-G3BP2 mRFP-DCP1A 0, 30, 60, 90, 120 min Download figure PowerPoint 1. types A, inhibitors. Control) µM 0.5 Control, A); 340 187 209 363 B); 125 353 1.5 G3BP1-mCherry 72 non-targeting siRNA siRNAs specific ? ?. (C) Efficacy knockdown HSPA1A verified total extracts. (D) medium. reported. 159 (siRNA 166 + ?). 37°C TIA-1, 605 510 VER); 1,334 GA); 468 17AAG). 3–4 s.e.m. 0.01 (One-way MG132 (20 439 649 637 0.0001 information: EV1 Movies EV1–EV5. (Taipale 2010; Biebl Buchner, 2019; Moran Luengo 2019) interacts (Markmiller Yet, currently dynamics. To test idea, measured well-established inhibitors activity, geldanamycin (GA) 17-Allylamino-17-demethoxygeldanamycin (17AAG) (Schulte Neckers, 1998). experiment, determined induces spontaneous assembly. up 3% EV1B). added phase both delayed B EV2), dose-dependent effect 1B EV2). Delayed upon be reproduced line EV1C D, EV3 EV4, HEK293T cells) observed siRNA-mediated 1C D EV5). Of note, stress, heat shock proteasome inhibitor 1E F). regulator type line. (or persistence) accumulate example, dysfunctional asked affects Although assisting subset newly synthesized (Schopf 2017), did observe colocalization alpha beta EV1E). contrast HSPA8 EV1E), targeting (Hartl Hayer-Hartl, 2002; Ganassi VER, 17AAG, GA, quantified microscopy. caused strong EV1F; 2016), led milder EV1F). Next, if upregulation assists prevents rescue found induction EV1G) EV1H). Together, demonstrate than Short-term It shown long-term decreases (Suzuki PBs cytoplasmic mRNP present growing increase size 2005). intimately connected SGs: equilibrium, share several interact, allowing transfer mRNPs between may perturb equilibrium PBs, indirectly affecting hypothesis, first using antibody PB-resident DCP1A. previous findings, disappearance EV1I), whereas had no EV1I). Tre