【病毒外文文献】2016 Middle East respiratory syndrome coronavirus shows poor replication but significant induction of antiviral response

Middle East respiratory syndrome coronavirus shows poor replication but significant induction of antiviral responses in human monocyte derived macrophages and dendritic cells Janne Tynell 1 Veera Westenius 1 Esa Ro nkko 1 Vincent J Munster 2 Krister Mele n 1 Pamela O sterlund 1 and Ilkka Julkunen 1 3 Correspondence Janne Tynell janne tynell thl fi 1 National Institute for Health and Welfare THL Helsinki Finland 2 Laboratory of Virology Division of Intramural Research National Institute of Allergy and Infectious Diseases National Institutes of Health Rocky Mountain Laboratories Hamilton MT USA 3 Department of Virology University of Turku Turku Finland In this study we assessed the ability of Middle East respiratory syndrome coronavirus MERS CoV to replicate and induce innate immunity in human monocyte derived macrophages and dendritic cells MDDCs and compared it with severe acute respiratory syndrome coronavirus SARS CoV Assessments of viral protein and RNA levels in infected cells showed that both viruses were impaired in their ability to replicate in these cells Some induction of IFN l1 CXCL10 and MxA mRNAs in both macrophages and MDDCs was seen in response to MERS CoV infection but almost no such induction was observed in response to SARS CoV infection ELISA and Western blot assays showed clear production of CXCL10 and MxA in MERS CoV infected macrophages and MDDCs Our data suggest that SARS CoV and MERS CoV replicate poorly in human macrophages and MDDCs but MERS CoV is nonetheless capable of inducing a readily detectable host innate immune response Our results highlight a clear difference between the viruses in activating host innate immune responses in macrophages and MDDCs which may contribute to the pathogenesis of infection Received 1 July 2015 Accepted 19 November 2015 INTRODUCTION The Middle East respiratory syndrome coronavirus MERS CoV was first discovered in 2012 in Saudi Arabia from a man suffering from an acute respiratory distress syndrome Zaki et al 2012 de Groot et al 2013 Since then 1611 confirmed cases with 575 fatalities have been reported as of October 2015 http www who int csr don 29 october 2015 mers saudi arabia en The high morbidity and novel nature of the virus have drawn comparisons with the severe acute respiratory syndrome coronavirus SARS CoV which infected w8000 people causing almost 800 fatalities in 2002 2003 Cheng et al 2007 Whilst sharing many similarities in terms of the clinical picture Hui et al 2014 studies focusing on pathogenesis have identified several notable differences between the viruses including different receptor usage Li et al 2003 Raj et al 2013 differences in cell tropism Chan et al 2013 Zielecki et al 2013 different suscepti bility to type I IFN Zielecki et al 2013 and differences in host response Josset et al 2013 Dromedary camels are considered to be the direct source of MERS CoV human infections as evidenced by isolation of MERS CoV from camels Azhar et al 2014 Raj et al 2014 and widespread seropositivity of camels in Africa and the Arabian Penin sula Raj et al 2013 Reusken et al 2013 2014 Haagmans et al 2014 Meyer et al 2014 Viruses very similar to MERS CoV have also been isolated from bats Ithete et al 2013 Memish et al 2013 Corman et al 2014 and a recent study showed binding of bat CoV HKU4 spike protein to the MERS CoV receptor human dipep tidyl peptidase 4 DPP4 Wang et al 2014 supporting a bat origin of MERS CoV MERS CoV causes a lower respiratory tract infection pre senting as pneumonia and common symptoms include fever cough sore throat and myalgia Hui et al 2014 A large portion of patients especially those with severe illness had some underlying condition such as diabetes or chronic renal disease Gastrointestinal symptoms and renal failure were also frequently observed in patients and MERS CoV RNA was detected in blood urine and rectal swabs of patients suggesting systemic dissemination One supplementary figure is available with the online Supplementary Material Journal of General Virology 2016 97 344 355 DOI 10 1099 jgv 0 000351 000351 G 2015 The Authors Printed in Great Britain344 and infection of the kidneys and gastrointestinal system Poissy et al 2014 This view is also supported by in vitro studies on MERS CoV host cell tropism confirm ing efficient replication in human renal and intestinal cell lines Chan et al 2013 Macrophages and dendritic cells DCs are abundantly present in infected lungs and they play an important role in infection control as producers of inflammatory cytokines and as antigen presenting cells Kopf et al 2015 They act as the first responders to invading pathogens and their interaction with the patho gens can be a strong determinant for the outcome of an infection Some respiratory viruses are known to infect and replicate in macrophages and DCs including human CoVs OC43 and 229E Collins 1998 Funk et al 2012 In addition to providing a platform for propagation infec tion of DCs has also been speculated to contribute to sys temic spread of influenza A virus through the lymphatic system Moltedo et al 2011 SARS CoV has been shown to infect macrophages and DCs but the infection is abortive and does not result in detectable viral protein synthesis or production of progeny viral particles Ziegler et al 2005 In the present study we investigated the ability of MERS CoV to replicate and induce innate antiviral genes in human lung epithelial cells and primary human monocyte derived macrophages and DCs MDDCs and compared the responses with those induced by SARS CoV RESULTS Monocyte derived macrophages and MDDCs are non permissive for MERS CoV infection In order to determine whether primary human macrophages and MDDCs are permissive for MERS CoV infection we assayed supernatant and cell lysate samples from infected cells by end point dilution and quantitative reverse tran scription qRT PCR to search for evidence of viral replica tion SARS CoV was included as a comparison and MDDCs were also infected with influenza A virus strain A Beijing 89 H3N2 as a positive control As shown in Fig 1 a no evi dence of efficient replication of MERS CoV was observed in the qRT PCR analysis in either cell type with viral RNA levels remaining the same decreasing or increasing only 0 001 0 01 0 1 1 10 100 1000 10 000 100 000 SARS CoV MERS CoV Relative copy number Macrophages 1 h 6 h 24 h 48 h 0 1 1 10 100 1000 10 000 100 000 SARS CoV MERS CoV Relative copy number DCs 1 h 6 h 24 h 48 h 0 1 2 3 4 5 6 7 8 1 6 Time h post infection Time h post infection 24 48 Viral titre log TCID 50 ml 1 Viral titre log TCID 50 ml 1 Macrophages MERS CoV SARS CoV 0 1 2 3 4 5 6 7 8 1 6 24 48 DCs MERS CoV SARS CoV b a Fig 1 Replication of MERS CoV and SARS CoV in human monocyte derived macrophages and MDDCs Cells were infected at m o i 1 and cell lysate and cell culture supernatant samples were collected at 1 6 24 and 48 h post infection a Total cellular RNA was isolated from cell lysates and qRT PCR analysis carried out to quantify MERS CoV and SARS CoV viral RNA in infected macrophages and MDDCs Relative RNA amounts were compared with the 1 h sample Results mean SD are representative of two independent experiments carried out in macrophages and MDDCs obtained from three different blood donors P 0 05 b Viral titres were measured from macrophage and MDDC supernatants using the end point dilution assay Results were calculated using the Spearman Karber method MERS CoV infection in macrophages and MDDCs http jgv microbiologyresearch org 345 weakly throughout the infection Similar results were obtained with SARS CoV Fig 1a End point dilution assays performed on the supernatants confirmed the obser vations showing a clear decrease in viral titre during the infection Fig 1b As a control to confirm the functionality of our experimental setting A549 Calu 3 and Vero E6 cells were infected with the viruses in a similar fashion as the leu kocytes A549 cells were found to be non permissive for infection with either virus as evidenced by a lack of increase in viral RNA amounts and decreasing supernatant titre levels throughout the infection Fig 2 Both viruses showed effi cient replication on Calu 3 and Vero E6 cells Fig 2 To further verify our observations we analysed cell lysates of infected macrophages MDDCs Calu 3 cells and Vero E6 cells for any increase in viral N protein amounts by West ern blotting Fig 3 As expected no increase was seen in macrophages or MDDCs whereas strong expression of N protein was observed in Calu 3 and Vero E6 cells Fig 3 MERS CoV infection results in induction of IFN b IFN l1 CXCL10 and MxA innate immune response genes Next we compared the ability of MERS CoV and SARS CoV to induce innate immune responses by analysing the expression levels of IFN b IFN l1 CXCL10 TNF a and MxA mRNAs by qRT PCR from cellular RNAs isolated from virus infected cells In macrophages MERS CoV infection resulted in increased levels of IFN l1 and CXCL10 mRNAs at the 6 h time point and some induction of IFN inducible MxA as well as a high induction of CXCL10 at the 24 h time point Fig 4 In SARS CoV infected macrophages no induction of cytokine or MxA expression was observed Fig 4 In MERS CoV infected MDDCs induction of CXCL10 and MxA was observed at the 6 and 24 h time points coupled with some induction of IFN b and IFN l1 at the 6 h time point Fig 5 SARS CoV infection in MDDCs led to a modest increase in the IFN l1 mRNA level at the 24 h time point Fig 5 In comparison with A Beijing 89 the induction of innate immune response genes by MERS CoV was weak with the exception of the MxA gene which was induced to a similar level as seen in influenza A virus infected cells at the 24 h time point In Calu 3 cells no sig nificant difference between MERS CoV and SARS CoV was detected with both viruses causing a clear induction of all the tested cytokine and MxA genes Fig 6 In A549 cells no induction of IFN b IFN l1 CXCL10 TNF a or MxA mRNAs was seen in response to infection with either one of the CoVs data not shown In support of the qRT PCR data we performed ELISAs on supernatants from the macrophage MDDC and Calu 3 infection experiments to determine whether increased 0 1 1 10 100 1000 10 000 100 000 SARS CoV MERS CoV Relative copy number A549 1 h 6 h 24 h 48 h 0 1 1 10 100 1000 10 000 100 000 SARS CoV MERS CoV Relative copy number Calu 3 1 h 24 h 48 h 72 h 96 h 1 h 24 h 48 h 72 h 96 h 0 1 1 10 100 1000 10 000 100 000 SARS CoV MERS CoV Relative copy number Vero E6 a b ND ND NDND ND 0 1 2 3 4 5 6 7 8 1 Viral titre log TCID 50 ml 1 Viral titre log TCID 50 ml 1 Viral titre log TCID 50 ml 1 Vero E6 MERS CoV SARS CoV 0 1 2 3 4 5 6 7 8 1 Calu 3 MERS CoV SARS CoV 0 1 2 3 4 5 6 7 8 16 A549 MERS CoV SARS CoV 24 48 72 96 24 48 72 9624 Time h post infection Time h post infection Time h post infection 48 Fig 2 Replication of MERS CoV and SARS CoV in Vero E6 cells and in human A549 and Calu 3 cells Vero E6 cells were infected at m o i 0 1 and A549 and Calu 3 cells at m o i 1 Cell lysate and cell culture supernatant samples were collected at the indicated time points a Total cellular RNA was isolated from cell lysates and qRT PCR analysis was carried out to quantify MERS CoV and SARS CoV viral RNA Relative RNA amounts were compared with the 1 h sample Data are presented as mean SD of triplicate measurements P 0 05 P 0 01 ND Not determined time points where sufficient amounts of high quality RNA could not be recovered due to extensive cytotoxicity caused by the viruses b Viral titres were measured from supernatants using the end point dilution assay Results were calculated using the Spearman Karber method J Tynell and others 346 Journal of General Virology 97 cytokine gene expression could be seen at the protein level Despite some expression at the mRNA level analysis of IFN l1 and TNF a showed no production of either cytokine in any cell type in response to MERS CoV or SARS CoV infection data not shown However strong induction of CXCL10 mRNA by MERS CoV in macro phages and MDDCs did correlate with high amounts of CXCL10 produced in some of the donors Fig 7 There was a large variation between different donors as three of the six tested donors produced high levels of CXCL10 in both macrophages and MDDCs whilst the other three showed no detectable CXCL10 production Fig 7 In Calu 3 cells only modest CXCL10 production at the 48 h time point was observed in response to MERS CoV Fig 7 SARS CoV induced strong CXCL10 production in Calu 3 cells but no production was detected in either macrophages or MDDCs Fig 7 Western blot analysis of MxA from cell lysates revealed high expression in MERS CoV infected macrophages and a more modest expression in MDDCs Fig 3 SARS CoV infection resulted in MxA expression in macrophages at the 24 h time point but no expression in MDDCs was detected Fig 3 In order to estimate the biological significance of the observed cytokine production UV inactivated super natants from the infection experiments were used to prime A549 cells for 24 h followed by a 6 h infection with A Beijing 89 virus Fig S1 available in the online Supplementary Material Cells primed with macrophage and MDDC supernatants from MERS CoV and A Beijing 89 infection experiments were the only cells to show a statistically significant reduction in influenza A virus M1 RNA expression levels Fig S1 DPP4 is expressed in macrophages and MDDCs Finally a quantitative assay for MERS CoV receptor DPP4 mRNA expression was set up to investigate whether differ ences in DPP4 expression would explain the observed differences in CoV replication DPP4 was expressed at similar levels in macrophages and MDDCs whereas approximately six to eightfold higher expression levels copy numbers were seen in Calu 3 cells compared with macrophages and MDDCs Fig 8a This observation is an unlikely explanation for differences in replication though as the DPP4 copy number was two to three times lower in Vero E6 cells than in macrophages or MDDCs Fig 8a Interestingly almost no DPP4 mRNA expression was detected in A549 cells Fig 8a To study the DDP4 protein expression in different cell types we car ried out Western blot analysis for DPP4 protein expression DCs Macrophages Calu 3 MERS CoV N protein Actin Mock 1 6 24 48 72 Mock 1 6 24 48 72 1 48 h post infection SARS CoV N protein Actin Calu 3 Vero E6 Mock Mock 1 24 48 1 24 48 72 MERS CoV N protein Actin h post infection M ock Mock Calu 3 Vero E6 12448124487296 SARS CoV N protein Actin h post infection MxA MxA Fig 3 Quantification of MERS CoV and SARS CoV N protein expression during infection Cell lysates from MDDC macro phage Calu 3 and Vero E6 infection experiments collected at the indicated time points were analysed by Western blotting for the expression of actin MxA or MERS CoV or SARS CoV N protein The 1 h sample and 1 5 diluted 48 h sample from the Calu 3 experiment were included as positive controls in the MDDC macrophage Western blots MERS CoV infection in macrophages and MDDCs http jgv microbiologyresearch org 347 using highly specific antibodies Fig 8b As shown in Fig 8 b DPP4 was detected in all tested cell types with A549 cells showing the lowest DPP4 expression levels DISCUSSION In the present study we investigated the characteristics of MERS CoV infection in human monocyte derived macro phages and MDDCs Our results show that like SARS CoV MERS CoV is unable to establish a productive infection in human macrophages and MDDCs in vitro as evidenced by no significant increase in viral titres RNA levels or expression of viral N protein during the infection We did however observe an increase in IFN b IFN l1 CXCL10 and MxA mRNA expression and in CXCL10 pro tein expression in response to MERS CoV infection This increase was not seen with SARS CoV revealing a clear difference between the viruses Calu 3 cells supported effi cient growth of both viruses and exhibited a strong expression of IFN b IFN l1 CXCL10 and MxA mRNAs but only SARS CoV induced strong CXCL10 protein expression An obvious explanation for the lack of MERS CoV replica tion in macrophages and MDDCs would be the lack of 0 1 1 10 100 1000 10 000 a b c e d Relative fold induction Time h post infection SARS CoV MERS CoV 0 1 1 10 100 1000 10 000 Relative fold induction Time h post infection SARS CoV MERS CoV 0 1 1 10 100 1000 10 000 Relative fold induction Time h post infection SARS CoV MERS CoV 0 1 1 10 100 1000 10 000 Relative fold induction Time h post infection SARS CoV MERS CoV 0 1 1 10 100 1000 10 000 Relative fold induction Time h post infection SARS CoV MERS CoV 24 48 24 48 24 48 24 48 24 48 16 16 16 16 16 Fig 4 MERS CoV and SARS CoV induced cytokine and MxA gene expression in human macrophages Human monocyte derived macrophages were left uninfected or infected with MERS CoV and SARS CoV at m o i 1 Cells were collected at 1 6 24 and 48 h after infection Total cellular RNA was isolated and the levels of a IFN b b IFN l1 c CXCL10 d TNF a and e MxA mRNAs were determined by qRT PCR Values were normalized to 18S rRNA levels Gene expression data are presented as relative fold induction of gene expression in relation to uninfected samples P 0 05 P 0 01 Results mean SD are representative of two independent experiments carried out in cells obtained from three different blood donors J Tynell and others 348 Journal of General Virology 97 MERS CoV receptor DPP4 on the cell surface DPP4 is ubiquitously expressed in many different tissues Lambeir et al 2003 and has been reported to be expressed on the cell surface of most monocyte derived macrophages and MDDCs Zhong et al 2013 In order to verify the expression of DPP4 in our cell systems we quantified DPP4 mRNA copy numbers in our samples by qRT PCR and also analysed DPP4 protein levels in cell lysates Fig 8 We found that both macrophages and MDDCs readily expressed DPP4 with macrophages showing even higher DPP4 protein levels than Calu 3 cells Fig 8 However low DPP4 expression correlated with the lack of MERS CoV replication on A549 cells Even though receptor avail ability is evidently not an issue a block at a very early stage of macrophage MDDC infection is suggested as no pro duction of MERS CoV or SARS CoV virus RNA or pro teins is seen Another possible explanation for impaired internalization besides receptor availability is the deficiency in spike protein cleavage CoV S protein is classified as a class I viral fusion protein and it needs to be cleaved for efficient internalization into a host cell Bosch et al 2003 This cleavage can occur either during the infection 0 1 1 10 100 1000 10 000 100 000 Relativ e fold induction Time h post infection 0 1 1 10 100 1000 10 000 a b c e d Relativ e fold induction Time h post infection SARS CoV MERS CoV A Beijing 89 SARS CoV MERS CoV A Beijing 89 0 1 1 10 100 1000 10 000 100 000 Relativ e fold induction Time h post infection A Beijing 89 SARS CoV MERS CoV 0 1 1 10 100 1000 10 000 Relativ e fold induction Time h post infection A Beijing 89 SARS CoV MERS CoV 0 1 1 10 100 1000 10 000 1616 16 16 16 24 4824 48 24 48 24 48 24 48 Relativ e fold induction Time h post infection A Beijing 89 SARS CoV MERS CoV ND ND NDND ND Fig 5 MERS CoV SARS CoV and influenza A virus induced cytokine and MxA gene expression in human MDDCs Human MDDCs were left uninfected or infected with MERS CoV and SARS CoV or influenza A virus strain A Beijing 89 H3N2 at m o i 1 Cells were collected at 1 6 24 and 48 h after infection Total cellular RNA was isolated and the levels of a IFN b b IFN l1 c CXCL10 d TNF a and e MxA mRNAs were determined by qRT PCR Values were normalized to 18S rRNA levels Gene expression data are presented as relative fold induction of gene expression in relation to uninfected samples P 0 05 P 0 01 ND Not determined time points where sufficient amounts of high quality RNA could not be recovered due to extensive cytotoxicity caused by the viruses Results mean SD are representative of two independent experiments carried out in cells obtained from three different blood donors MERS C