key: cord-1021448-o1eh3gk5 authors: Mohamud, Yasir; Xue, Yuan Chao; Liu, Huitao; Ng, Chen Seng; Bahreyni, Amirhossein; Jan, Eric; Luo, Honglin title: The papain-like protease of coronaviruses cleaves ULK1 to disrupt host autophagy date: 2020-10-25 journal: bioRxiv DOI: 10.1101/2020.10.23.353219 sha: 302a286e7f66a59da6b52f33b6597a34efbee162 doc_id: 1021448 cord_uid: o1eh3gk5 The ongoing pandemic of COVID-19 alongside the outbreaks of SARS in 2003 and MERS in 2012 underscore the significance to understand betacoronaviruses as a global health challenge. SARS-CoV-2, the etiological agent for COVID-19, has infected more than 29 million individuals worldwide with nearly ~1 million fatalities. Understanding how SARS-CoV-2 initiates viral pathogenesis is of the utmost importance for development of antiviral drugs. Autophagy modulators have emerged as potential therapeutic candidates against SARS-CoV-2 but recent clinical setbacks underline the urgent need for better understanding the mechanism of viral subversion of autophagy. Using murine hepatitis virus-A59 (MHV-A59) as a model betacoronavirus, time-course infections revealed a significant loss in the protein level of ULK1, a canonical autophagy regulating serine-threonine kinase, and the concomitant appearance of a possible cleavage fragment. To investigate whether virus-encoded proteases target this protein, we conducted in vitro and cellular cleavage assays and identified ULK1 as a novel bona fide substrate of SARS-CoV-2 papain-like protease (PLpro). Mutagenesis studies discovered that ULK1 is cleaved at a conserved PLpro recognition sequence (LGGG) after G499, separating its N-terminal kinase domain from the C-terminal substrate recognition region. Consistent with this, over-expression of SARS-CoV-2 PLpro is sufficient to impair starvation-induced canonical autophagy and disrupt formation of ULK1-ATG13 complex. Finally, we demonstrated a dual role for ULK1 in MHV-A59 replication, serving a pro-viral functions during early replication that is inactivated at late stages of infection. In conclusion, our study identified a new mechanism by which PLpro of betacoronaviruses induces viral pathogenesis by targeting cellular autophagic pathway (Word count=250) IMPORTANCE The recent COVID-19 global pandemic alongside the 2003 SARS and 2012 MERS outbreaks underscore an urgent need to better understand betacoronaviruses as pathogens that pose global challenge to human health. Studying the underlying biology of how betacoronaviruses subvert innate cellular defense pathways such as autophagy will help to guide future efforts to develop anti-viral therapy. (Word count= 55) ABSTRACT 23 The ongoing pandemic of COVID-19 alongside the outbreaks of SARS in 2003 and MERS in 24 2012 underscore the significance to understand betacoronaviruses as a global health challenge. Autophagy-modulating drugs such as chloroquine/hydroxychlorine have emerged as 58 potential anti-viral agents against SARS-CoV-2; however, recent placebo-controlled trials 59 showing no clinical benefits and possible safety concerns are spearheading global research 60 efforts for better understanding (1). Autophagy is an evolutionarily conserved process that acts to 61 recycle cellular waste while also responding to invading pathogens. Cellular membranes termed 62 phagophores first enwrap cargo inside double-membraned chambers, so called autophagosomes, 63 after which cargo is degraded upon fusion of autophagosomes with digestive lysosomes. The 64 process is tightly regulated and responsive to various cellular stressors including nutrient-stress, 65 oxidative stress, as well as viral infection (2). The serine/threonine unc-51-like kinase (ULK1) is a critical upstream regulator of 67 autophagy. The role of ULK1 as a nutrient-responsive orchestrator of autophagy has been well 68 characterized (3). ULK1 is recruited to sites of autophagosome biogenesis where it 69 phosphorylates key autophagy regulatory proteins. Its central role in autophagy has implicated 70 ULK1 in diverse human diseases from cancer and neurodegeneration to inflammatory disorders 71 (4-6). Structurally, ULK1 possesses an N-terminal kinase domain and a C-terminal early 72 autophagy targeting (EAT) domain. The latter facilitates interaction of ULK1 with its various 73 substrates. Autophagy-independent functions of ULK1 have also emerged including the 74 regulation of ER-Golgi trafficking as well as innate immune signaling (7-10). For example, 75 stimulator of interferon genes (STING), a critical adaptor of the DNA sensor cyclic GMP-AMP 76 synthase (cGAS), was previously identified as a substrate of ULK1 (11). Furthermore, the master 77 innate immune kinase TANK binding kinase 1 (TBK1) was reported to be phosphorylated by 78 ULK1 and participate in metabolic signaling (12). Figure 1A & B) . Interestingly, using the polyclonal anti-ULK1 antibody that recognizes the 102 region of amino acids 351-400, we observed an additional band at ~65 kDa ( Figure 1A ). 103 However, this potential cleavage fragment was undetected with the monoclonal anti-ULK1 104 antibody that was raised against amino acids 511-750 ( Figure 1B) , indicating a cleavage event 105 may occur within this region. To investigate whether virus-mediated loss of ULK1 is through 106 transcriptional regulation, we performed RT-qPCR. Figure 1C showed that mRNA levels of 107 Ulk1 were elevated following 12h and 24h infection, suggesting that reduced protein expression 108 of ULK1 is not a result of decreased mRNA level. In contrast, elevated gene expression may 109 indicate a compensatory upregulatory mechanism following reduced protein expression. The discovery of a potential ULK1 cleavage product in MHV-infected lysates prompted us to 113 investigate whether betacoronaviral proteases are responsible for ULK1 cleavage. Betacoronaviruses encode two proteases, a papain-like cysteine protease (PL pro ) and a 3-115 chymotrypsin-like cysteine protease (3CL pro , also known as the Main protease, M pro ) that process 116 viral polyproteins into individual functional proteins (17). We generated a construct expressing 117 PL pro of SARS-CoV-2, which has 63% similarity with PL pro domain 2 of MHV-A59 ( Figure 118 2A). To determine whether PL pro targets ULK1, HEK293T cells were transiently transfected with 119 either a control vector or a plasmid expressing PL pro for 24h. Expression of PL pro alone was 120 sufficient to recapitulate the reduction in full-length ULK1 observed following MHV-A59 121 infection using the monoclonal anti-ULK1 antibody ( Figure 2B ). PL pro -mediated cleavage of 122 endogenous ULK1 was then confirmed using N-terminal 3×Flag-ULK1 construct ( Figure 2C ). HEK293T cells were co-transfected with PL pro together with 3×Flag-ULK1. Lysates probed with 124 an anti-Flag antibody revealed a significant reduction of ULK1 with the concomitant detection of 125 a lower-molecular-weight fragment of ~70kDa in size. To exclude the potential involvement of 126 3CL pro , we performed in vitro cleavage assay. Time course treatment revealed that ULK1 was 127 not targeted by the SARS-CoV-2 3CL pro , whereas coxsackievirus 3C pro , the positive control, 128 demonstrated an efficient cleavage of ULK1 ( Figure 2C ). Together, our data suggest that PL pro , 129 but not 3CL pro , targets ULK1 for cleavage. To identify the precise location within ULK1 that is targeted by PL pro , we closely investigated To understand the functional consequence of PL pro -mediated cleavage of ULK1, we first 146 examined whether the ability of ULK1 to interact with its binding partners is disrupted. ATG13 147 is a bridging molecule that directly interacts with both ULK1 and the scaffold protein FIP200 in 148 the autophagy-initiating ULK1 complex (3). Expression of PL pro resulted in reduced co-149 immunoprecipitation of ULK1 with its binding partner ATG13 ( Figure 4A) . Given that the 150 ULK1 complex is required for starvation-induced canonical autophagy, we inquired the Figure 5B ). Consistent with a potential pro-viral role for ULK1, RNA levels and viral 175 titers of MHV-A59 were significantly reduced following gene-silencing of Ulk1 ( Figure 5C ). 176 We also assessed the effects of expression of WT-or non-cleavable mutant-ULK1 on viral 177 replication. We showed that cells expressing WT-ULK1 displayed significantly enhanced viral 178 titers (Figure 5D) , supporting a pro-viral function for ULK1. However, a non-cleavable ULK1 179 mutant demonstrated diminished viral titers ( Figure 5D) . Collectively, these data support a 180 requirement of ULK1 for replication prior to its cleavage. In summary, we uncover a novel function for the PL pro of SARS-CoV-2 in cleaving the 233 autophagy-regulating kinase ULK1. These insights clarify the mechanism behind 234 betacoronaviral pathogenesis and subversion of cellular autophagy. The 3×Flag-ULK1 plasmid was generated using a multiple cloning site modified CMV10 vector 252 backbone with the corresponding cut sites for ULK1 (EcorI/BamHI). Cleavage resistant mutants, LGGG-499-QQQG and MRGG-531-MRDD, were generated using gBLOCKS DNA synthesis incubation, 50% tissue culture infective dose titer (TCID50) was calculated by the statistical 304 method of Reed and Muench (34) . Titers were expressed as plaque forming unit (PFU)/mL with 305 1 infectious unit equal to 0.7 TCID50 as described previously (35) All authors declare that they have no conflict of interest. Lysates were probed by western blotting for endogenous ULK1 using the monoclonal anti-ULK1 458 antibody and normalized to ACTB as in Figure 1A . (C) HEK293T cells were co-transfected with 3×Flag-ULK1 and either control vector or PL pro -460 expressing plasmid for 24h. Lysates were probed with anti-Flag antibody to detect exogenous 461 ULK1. Arrow denotes cleavage fragment observed at ~70 kDa. Densitometry is provided as in 462 Figure 1A . (D) Purified SARS-CoV-2 3CL wt (4 µg) or catalytically-inactive C145A 3CL mut (4 µg) of SARS- CoV-2 and CVB3 3C wt (0.1 µg) alongside HeLa lysates (30 g) were used to perform in vitro 465 cleavage assay. Western blotting was performed to probe for endogenous ULK1 and normalized 466 to ACTB as in Figure 1A . Arrows denote cleavage fragments. Lee 340 TC, Hullsiek KH. 2020. A randomized trial of hydroxychloroquine as postexposure prophylaxis 341 for Covid-19 Autophagy and the integrated stress response ULK1 induces autophagy by phosphorylating Beclin-1 and activating VPS34 lipid kinase Small molecule inhibition 349 of the autophagy kinase ULK1 and identification of ULK1 substrates Genetic variation in the autophagy gene ULK1 and risk of Crohn's disease A 354 C9ORF72/SMCR8-containing complex regulates ULK1 and plays a dual role in autophagy The noncanonical 358 role of ULK/ATG1 in ER-to-Golgi trafficking is essential for cellular homeostasis IFN-gamma-inducible antiviral responses 361 require ULK1-mediated activation of MLK3 and ERK5 Central role of ULK1 in type I interferon 365 signaling ULK1/2 restricts the 367 formation of inducible SINT-speckles, membraneless organelles controlling the threshold of 368 TBK1 activation Cyclic Dinucleotides Trigger ULK1 (ATG1) Phosphorylation 370 of STING to Prevent Sustained Innate Immune Signaling TBK1 at 372 the crossroads of inflammation and energy homeostasis in adipose tissue RNA replication of mouse 378 hepatitis virus takes place at double-membrane vesicles Subversion of the cellular autophagy pathway by viruses Coronavirus replication 382 complex formation utilizes components of cellular autophagy Discovery of Seven Novel Mammalian and Avian 385 Coronaviruses in the Genus Deltacoronavirus Supports Bat Coronaviruses as the Gene Source of 386 Alphacoronavirus and Betacoronavirus and Avian Coronaviruses as the Gene Source of 387 Gammacoronavirus and Deltacoronavirus The papain-like 389 protease of severe acute respiratory syndrome coronavirus has deubiquitinating activity Inhibits Autophagic Flux via Disruption of the SNARE Complex to Enhance Viral Replication 395 Pharmacological inhibition of ULK1 kinase blocks mammalian target of rapamycin (mTOR)-396 dependent autophagy Autophagy and viruses: adversaries or allies? Subversion of cellular autophagosomal machinery by RNA viruses The intertwined life cycles of enterovirus and autophagy Viruses and autophagy Autophagy 405 protects against Sindbis virus infection of the central nervous system Enteroviruses remodel 408 autophagic trafficking through regulation of host SNARE proteins to promote virus replication 409 and cell exit CALCOCO2/NDP52 and SQSTM1/p62 411 differentially regulate coxsackievirus B3 propagation Enteroviruses Remodel Autophagic Trafficking through Regulation of Host SNARE Proteins to Promote Virus Replication 414 and Cell Exit Involvement of autophagy in coronavirus replication Ulk1 418 governs nerve growth factor/TrkA signaling by mediating Rab5 GTPase activation in porcine 419 hemagglutinating encephalomyelitis virus-induced neurodegenerative disorders The papain-like 421 protease of severe acute respiratory syndrome coronavirus has deubiquitinating activity SARS-CoV-2 proteases cleave IRF3 and 425 critical modulators of inflammatory pathways (NLRP12 and TAB1) : Implications for disease 426 presentation across species and the search for reservoir hosts A Role for Ubiquitin in Selective Autophagy A simple method of estimating fifty percent endpoints Nature of viruses Murine 17Cl1 cells were transfected Cas9 and sg-CON or sg-ULK1 for 48h. Cells were then 501 infected with MHV-A59 (MOI=10) for an additional 24h. ULK1 knockout efficiency (left), viral 502 RNA levels in the cells (middle) and viral titers in the supernatant (right) were measured by 503 western blotting D) 17Cl1 cells were transfected with vector, 3× Flag-ULK1 WT , or 3× Flag-ULK1 MUT for 24h Cell lysates were analyzed by 506 western analysis for exogenous ULK1 with anti-Flag antibody. Viral titers in culture medium 507 were determined by TCID50 assay presented. Highlighted gray boxes denote potential consensus cleavage sites. Dashed arrows 471 denote sites of potential cleavages. (B) HEK293T cells were co-transfected with PL pro and either 3×Flag-ULK1 WT , 3×Flag-ULK1-473 G499 mut or 3×Flag-ULK1-G531 mut . Western blotting was conducted with anti-Flag and anti-GFP 474 antibody for detection of ULK1 and PL pro , respectively. were starved for 2h in the presence or absence of lysosomal inhibitor BAF (125 nM) for 2h. Lysates were probed for LC3 and normalized to ACTB. Quantification of LC3-II was conducted 490 as in Figure 1A and presented as (mean ± SD, n=3) in the right panel.