【病毒外文文献】2013 MERS-coronavirus replication induces severe in vitro cytopathology and is strongly inhibited by cyclosporin A or in

1 VIR 2013 052910 REVISED VERSION1 2 MERS coronavirus replication induces severe in vitro cytopathology and is strongly 3 inhibited by cyclosporin A or interferon alpha treatment4 5 6 Adriaan H de Wilde 1 V Stalin Raj 2 Diede Oudshoorn 1 Theo M Bestebroer 2 Stefan van 7 Nieuwkoop 2 Ronald W A L Limpens 3 Clara C Posthuma 1 Yvonne van der Meer 1 Montserrat 8 B rcena 3 Bart L Haagmans 2 Eric J Snijder 1 and Bernadette G van den Hoogen 2 9 10 1 Molecular Virology Laboratory Department of Medical Microbiology Leiden University Medical 11 Center Leiden The Netherlands12 2 Viroscience Lab Erasmus MC Rotterdam The Netherlands13 3 Section Electron Microscopy Department of Molecular Cell Biology Leiden University Medical 14 Center Leiden The Netherlands15 16 Corresponding authors Eric J Snijder E J Snijder lumc nl and Bernadette G van den Hoogen17 b vandenhoogen erasmusmc nl 18 19 Abstract word count 23520 Text word count 549921 Number of figures 722 23 24 Running title MERS CoV replication and sensitivity to CsA and IFN g302g325 26 Contents Category Animal Viruses Positive strand RNA27 28 29 JGV Papers in Press Published June 19 2013 as doi 10 1099 vir 0 052910 0 2 SUMMARY30 31 Coronavirus CoV infections are commonly associated with respiratory and enteric disease in 32 humans and animals The 2003 outbreak of severe acute respiratory syndrome SARS highlighted the 33 potentially lethal consequences of CoV induced disease in humans In 2012 a novel CoV Middle 34 East Respiratory Syndrome coronavirus MERS CoV emerged causing 49 human cases thus far of 35 which 23 had a fatal outcome In this study we characterized MERS CoV replication and cytotoxicity 36 in human and monkey cell lines Electron microscopy of infected Vero cells revealed extensive 37 membrane rearrangements including the formation of double membrane vesicles and convoluted 38 membranes which were previously implicated in the RNA synthesis of SARS CoV and other CoVs 39 Following infection we observed rapidly increasing viral RNA synthesis and release of high titres of 40 infectious progeny followed by pronounced cytopathology These characteristics were used to 41 develop an assay for antiviral compound screening in 96 well format which was used to identify42 cyclosporin A as an inhibitor of MERS CoV replication in cell culture Furthermore MERS CoV was 43 found to be 50 100 times more sensitive to interferon alpha IFN g302g12g3g87g85g72g68g87g80g72g81g87g3g87g75g68g81g3g54g36g53g54 CoV an 44 observation that may have important implications for the treatment of MERS CoV infected patients 45 MERS CoV infection did not prevent the IFN induced nuclear translocation of phosphorylated 46 STAT1 in contrast to infection with SARS CoV where this block inhibits the expression of antiviral 47 genes These findings highlight relevant differences between these distantly related zoonotic CoVs in 48 terms of their interaction with and evasion of the cellular innate immune response 49 50 3 INTRODUCTION51 52 In June 2012 a previously unknown coronavirus was isolated from a 60 year old Saudi 53 Arabian patient who died from acute respiratory distress syndrome and multiple organ failure Zaki et 54 al 2012 Subsequently the novel virus was isolated from several additional residents and visitors of 55 the Arabian Peninsula suffering from similar respiratory symptoms In retrospect also a cluster of 56 respiratory infections in Jordan April 2012 was linked to the same agent although no convincing57 evidence for human to human transmission was obtained This was clearly different for a cluster of 58 three U K cases in early 2013 consisting of a patient who had travelled to Saudi Arabia and two 59 family members without recent travel history outside the U K In the past year various names have 60 been used to refer to this newly identified CoV including novel beta coronavirus nCoV and human 61 coronavirus EMC HCoV EMC but following a recent recommendation by the coronavirus study 62 group of ICTV and other experts de Groot et al 2013 we will use Middle East Respiratory 63 Syndrome coronavirus MERS CoV throughout this paper Up to May 2013 49 confirmed MERS64 cases including 23 fatalities have been recorded65 http www who int csr don archive disease coronavirus infections en 66 Coronavirus CoV infections are associated with respiratory and enteric disease in humans 67 and animals Since the 1960s two human CoVs HCoVs OC43 and 229E were known to cause mild 68 respiratory disease Hamre McIntosh et al 1967 but it was the 2003 outbreak of 69 severe acute respiratory syndrome SARS fatality rate 10 that revealed the potentially lethal 70 consequences of CoV induced disease in humans Drosten et al 2003 Ksiazek et al 2003 Two 71 years later bats were identified as the most likely animal reservoir for this zoonotic CoV Lau et al 72 2005 Li et al 2005 Subsequently a wide variety of bat associated CoVs was discovered 73 Vijaykrishna et al 2007 Woo et al 2007 and also two additional human CoVs NL63 and HKU1 74 Fouchier et al 2004 van der Hoek et al 2004 Woo et al 2005 were identified Although the 75 general capacity of bat CoVs to switch hosts appears to be rather restricted Muller et al 2012 the 76 4 possibility of SARS CoV re emergence or zoonotic transfer of other animal CoVs has remained a 77 public health concern over the past 10 years 78 Coronaviruses are classified in four genera alpha beta gamma and deltacoronaviruses de 79 Groot et al 2012 and our previous analysis of the MERS CoV genome van Boheemen et al 2012 80 identified the newly emerging agent as a member of lineage C of the genus Betacoronavirus 81 Strikingly as in the case of SARS CoV the closest known relatives of MERS CoV are bat 82 coronaviruses like HKU 4 and HKU 5 van Boheemen et al 2012 Woo et al 2007 The 83 evolutionary distance to SARS CoV lineage B is considerable a notion further supported by recent 84 comparative studies revealing important differences in receptor usage Muller et al 2012 Raj et al 85 2013 86 Mammalian viruses have to cope with the host cell s innate responses including those87 triggered by activation of the type I interferon IFN pathway reviewed by Randall Zhong et al 2012 90 For example it was postulated that the sensing of double stranded ds RNA replication intermediates91 by the innate immune system is inhibited by the elaborate virus induced membrane structures with 92 which CoV RNA synthesis is associated Knoops et al 2008 Versteeg et al 2007 Other evasion 93 mechanisms were attributed to protein functions that can be either conserved across CoVs or specific 94 for certain CoV lineages Proteins such as the nsp3 proteinase Ratia et al 2006 the nsp16 2 O 95 methyltransferase Zust et al 2011 and the products of SARS CoV ORFs 3b 6 and 7a Frieman et 96 al 2007 Hussain et al 2008 Kopecky Bromberg et al 2006 Zhou et al 2012 have all been97 described to prevent IFN induction signalling In particular the SARS CoV ORF6 protein is known to 98 inhibit IFN induced JAK STAT signalling by blocking the nuclear translocation of phosphorylated 99 STAT1 p STAT1 which contributes to the pathogenic potential of the virus in a mouse model Sims100 et al 2013 In spite of these immune evasion strategies treatment with type I IFNs can inhibit CoV 101 replication in vitro Garlinghouse et al 1984 Haagmans et al 2004 Paragas et al 2005 Taguchi Zheng et al 2004 and for example protected type I pneumocytes against SARS CoV 103 infection in macaques Haagmans et al 2004 104 5 Clearly well characterized systems for MERS CoV replication in cell culture will be 105 invaluable for future studies into basic virus properties and interactions with the host including innate 106 immune responses Therefore we set out to characterize the replication of MERS CoV in different 107 cell lines Using this information an assay to screen for antiviral compounds was developed which 108 identified cyclosporin A CsA as an inhibitor of MERS CoV replication Our first screening 109 experiments also established that compared to SARS CoV MERS CoV replication is more sensitive 110 to type I interferon treatment 111 112 RESULTS113 114 Kinetics of MERS CoV replication in Vero and Huh7 cells Only a few laboratory studies on 115 MERS CoV replication have been reported thus far Cells from a variety of mammalian hosts were 116 found to be susceptible and infection can induce pronounced cytopathology and cell death Muller et 117 al 2012 Zaki et al 2012 Following entry the CoV replicative cycle starts with the translation of 118 the positive stranded RNA genome into replicase polyproteins that are cleaved into 16 nsps 119 Gorbalenya et al 2006 van Boheemen et al 2012 These direct both genome replication and the 120 synthesis of the subgenomic sg mRNAs required to express the structural and accessory proteins To 121 investigate MERS CoV replication in more detail we used Vero and Huh7 cells to analyse viral RNA 122 synthesis and progeny release in single cycle infection experiments 123 Hybridisation analysis of the accumulation of viral RNA revealed the presence of genome 124 RNA and seven sg transcripts with sizes closely matching those previously predicted from the 125 positions of conserved transcription regulatory sequences TRS in the viral genome van Boheemen126 et al 2012 Fig 1a The relative abundance of the various sg mRNAs is similar to what has been 127 observed for other CoVs with the smallest species encoding the N protein being by far the most 128 abundant transcript Fig 1b In both cell lines viral mRNAs could be readily detected at 7 h p i and 129 reached maximum levels around 13 h p i Fig 1a Viral RNA levels remained more or less constant 130 until 24 h p i in Vero cells whereas the amount isolated from Huh7 cells declined due to the more 131 rapid development of cytopathology in this cell line between 13 and 24 h p i see below After the 132 6 peak of viral RNA accumulation had been reached the titre of virus released from MERS CoV 133 infected Vero cells steadily increased from 5 x 10 5 to 5 x 10 7 p f u per ml Fig 1c Interestingly 134 the bulk of the viral progeny was released significantly earlier from Huh7 cells although the final titre135 at 24 h p i was comparable to that obtained from Vero cells 136 137 Antisera raised against non structural proteins of other betacoronaviruses cross react with 138 MERS CoV proteins Despite the relatively large evolutionary distance to better characterized 139 CoVs we tested a panel of antisera from our laboratory for cross reactivity with MERS CoV infected 140 cells In contrast to a polyclonal serum recognizing the SARS CoV nucleocapsid N protein data not 141 shown antisera against various SARS CoV nsps nsp3 nsp5 nsp8 Snijder et al 2006 raised using 142 purified recombinant proteins as antigen were found to strongly cross react Fig 2a In addition 143 rabbit antisera raised against synthetic peptides 23 mers representing a small but conserved C 144 terminal part of SARS CoV and MHV nsp4 strongly cross reacted with MERS CoV Only small but 145 apparently immunogenic parts of these peptides e g LYQPP are absolutely conserved between 146 MHV and MERS CoV nsp4 Fig 2b Conservation in other betacoronaviruses data not shown 147 suggests that antisera recognizing this nsp4 region may be used for immunodetection of additional 148 newly emerging CoVs 149 150 MERS CoV replication structures Subsequently we employed a monoclonal antibody recognizing 151 dsRNA to localize intermediates in viral RNA synthesis Knoops et al 2008 Weber et al 2006 In 152 various cell types the immunolabelling signals for both replicase and dsRNA localized to the 153 perinuclear region Fig 2c where the replication structures induced by other CoVs are known to 154 accumulate Brockway et al 2003 Gosert et al 2002 Knoops et al 2008 Snijder et al 2006 155 Stertz et al 2007 Ulasli et al 2010 156 We next used electron microscopy EM to investigate the ultra structural and potentially 157 cytopathic changes that MERS CoV induces in infected cells and focused on the membranous 158 replication structures that support MERS CoV RNA synthesis The preservation of such structures 159 typically double membrane vesicles DMVs and convoluted membranes CMs was previously 160 7 found to be significantly improved by using protocols that include cryo fixation and freeze 161 substitution Knoops et al 2008 Snijder et al 2006 We now applied these advanced preservation 162 techniques including newly developed protocols for high pressure freezing HPF to MERS CoV 163 infected Vero cells Images of similarly prepared SARS CoV infected Vero E6 cells are included for 164 comparison Fig 3f 165 Compared to mock infected control cells Fig 3e different degrees of distinct alterations166 were observed at 8 h p i Some cells contained relatively small DMV clusters Fig 3a b black 167 arrowheads inset whereas in others large numbers of DMVs occupied extensive areas of the 168 perinuclear region Fig 3c d differences that likely reflect different stages in infection progression 169 The diameter of MERS CoV induced DMVs ranged from 150 to 320 nm comparable to what was170 previously measured for SARS CoV induced structures Knoops et al 2008 An interesting171 morphological difference with our previous studies of SARS CoV infected cells was the presence of a 172 dense inner DMV core which can be attributed to technical differences in sample preparation In 173 terms of ultrastructural preservation HPF is widely considered superior to the previously used 174 plunge freezing protocols Also in the case of SARS CoV Fig 3f and the distantly related equine 175 arteritis virus Knoops et al 2012 a similar dense DMV core became apparent when HPF was 176 employed Although DMV cores are known to contain dsRNA the implications of these 177 ultrastructural observations remain unclear Interestingly CMs were always surrounded by DMV 178 clusters and were only observed in cells that appeared to be more advanced in infection Fig 3c d 179 white arrows inset This observation strengthens the notion that DMV formation precedes the180 development of CMs as previously postulated for SARS CoV Knoops et al 2008 181 182 MERS CoV induced cytophatology and cell death In cell culture many CoVs induce severe 183 cytopathic effect CPE and cell death Infection with a number of CoVs can also induce extensive 184 syncytium formation due to fusion activity of the viral spike protein at neutral pH reviewed in 185 Belouzard et al 2012 MERS CoV induced cytopathology was monitored by light microscopy 186 following low m o i inoculation of monkey and human cells Fig 4 In line with previous 187 observations Zaki et al 2012 Vero cells developed clear CPE at 2 days post infection d p i and 188 8 detached at 3 d p i Fig 4a Similar observations were made for Calu3 2B4 cells Fig 4b In 189 contrast MERS CoV infected Vero E6 cells displayed only mild CPE starting at 3 d p i and cell death 190 was not complete after six days Fig 4c The development of CPE in Huh7 cells was strikingly faster 191 compared to the three other cell lines and following extensive syncytium formation cells detached 192 already around 17 h Fig 4d Given the low m o i used and the viral replication kinetics Fig 1 the 193 syncytium formation in these only partially infected Huh7 cultures appeared to be a major factor in 194 CPE development DPP4 expression on Vero and Huh7 cells Raj et al 2013 and expression levels 195 of DPP4 on Calu3 2B4 and Vero E6 cells correlated with susceptibility to MERS CoV data not 196 shown 197 198 Development of an assay to screen for compounds inhibiting MERS CoV replication The virus 199 induced CPE in Vero and Huh7 cells was used to develop a first assay to screen for compounds that 200 inhibit MERS CoV replication in cell culture Vero cells were seeded in 96 well plates and infected at 201 an m o i of 0 005 or 0 05 Fig 5a After two and three days CPE formation was monitored 202 microscopically and cytotoxicity was measured using a commercial cell viability assay Moderate 203 CPE was observed on day 2 and by day 3 cell viability had dropped below 10 with both virus doses 204 used Fig 5a indicating near complete cell death In MERS CoV infected Huh7 cells Fig 5b 205 already after day 1 cell viability had dropped to 79 or 24 after m o i 0 005 or 0 05 infection 206 respectively which was in line with our observations on rapid syncytium formation and CPE in this 207 particular cell line Fig 4d One day later CPE was complete for both virus doses used and cells had 208 detached Fig 5b Based on this comparison further experiments were done using an m o i of 0 005209 and Huh7 and Vero cells were incubated for two or three days respectively before measuring cell 210 viability 211 Previously it was shown that replication of various CoVs including SARS CoV can be 212 inhibited by the immunosuppressive drug CsA de Wilde et al 2011 Pfefferle et al 2011 213 Therefore while testing whether the CPE based assay described above could be used as an antiviral 214 screening method we used CsA treatment to obtain a first proof of principle Infected Vero cells were 215 given g22g3g82g85g3g28g3g541g48g3of CsA and were analysed at 3 d p i At the concentrations used CsA did not 216 9 adversely affect the viability of mock infected cells Fig 5c g55g85g72g68g87g80g72g81g87g3g90g76g87g75g3g28g3g541g48g3g70g82g80g83g79g72g87g72ly217 inhibited CPE and left cell viability unchanged compared to mock infected control cells The 218 inhibitory effect of CsA was confirmed in Huh7 cells Fig 5d which displayed reduced and lack of 219 g38g51g40g3g88g83g82g81g3g87g85g72g68g87g80g72g81g87g3g90g76g87g75g3g26g17g24g3g541g48g3g68g81g71g3g20g24g3g541g48g3g38g86g36g15g3g85g72g86g83g72g70g87g76g89g72g79g92g17g3g55g75g72g86g72g3g85g72g86g88g79g87g86g3g90g72g85g72g3g70g82g85g85g82g69g82g85g68g87g72g71g3g69g92g3220 immunofluorescence microscopy analysis of CsA treated and high m o i infected Vero and Huh7 221 cells and by determining virus titres released into the medium Both assays confirmed an almost 222 complete block of MERS CoV infection data not shown However as previously reported for other 223 CoVs de Wilde et al 2011 a small fraction of MERS CoV infected cells appeared to be refractive 224 to CsA treatment and supported a low level of MERS CoV replication even at high CsA 225 concentrations data not shown 226 227 Enhanced sensitivity of MERS CoV to pegylated IFN g302g3g87g85g72g68g87g80g72g81g87g3g76g81g3g70g82g80g83g68g85g76g86g82g81g3g87g82g3g54g36g53g54 CoV 228 Type I IFNs inhibit CoV replication and can protect against infection in animal models Haagmans et 229 al 2004 Taguchi van Boheemen249 et al 2012 In particular MERS CoV does not encode a homolog of the SARS CoV ORF6 protein 250 which was reported to block the IFN induced nuclear translocation of phosphorylated transcription 251 factor STAT1 As nuclear translocation of p STAT1 is essential for transcriptional activation of 252 downstream antiviral genes the ORF6 protein makes SARS CoV less sensitive to treatment with type 253 I IFN Frieman et al 2007 Sims et al 2013 IFN induced translocation of p STAT1 was readily 254 observed in IFN treated mock infected Vero cells Fig 7a d but not in IFN treated SARS CoV 255 infected cells Fig 7e f In contrast in MERS CoV infected and IFN treated cultures the 256 translocation of p STAT1 was detected Fig 7g h Together with the data on IFN sensitivity Fig 5 257 these observations highlight important differences between SARS CoV and MERS CoV in terms of 258 their interaction with the IFN signalling pathways 259 260 11 261 DISCUSSION262 263 Following the 2003 SARS epidemic global CoV hunting efforts identified a wealth of previously 264 unknown family members in particular in bat species from several continents de Groot et al 2012 265 Moreover at least three of the four current established human CoVs NL63 229E and OC43 were 266 postulated to have originated from zoonotic reservoirs Huynh et al 2012 Pfeff