【病毒外文文献】2018 Attenuation of replication by a 29 nucleotide deletion in SARS-coronavirus acquired during the early stages of huma

1SCIENTIFIC REPORTS 2018 8 15177 w v w v y z w v w y y z Attenuation of replication by a x Doreen w x y w x y y y an z Gottula w x y Gloza Balboni Battilani Danijela w v Pfeifer w w z w x y M ller w x y Drosten w x y x x w Rhinolophus x x x y x Emerging zoonotic viruses such as Severe Acute Respiratory Syndrome coronavirusuni00A0 SARS CoV are a consider able concern for public health 1 Bats of the genus Rhinolophus are the bona f ide animal reservoir for SARS and SARS related CoV SARSr CoV 2 Epidemics may emerge from wild animal reservoirs in which a plethora of viral variants exists Af ter initial cross host infection the occurrence of positive selection with adaptive changes w Charit Universit tsmedizin Berlin corporate member of Freie Universit t Berlin Humboldt Universit t zu Berlin w w v w w x y Institute of Virology University of Bonn Medical Centre Sigmund x y w x z x y y w z Department of Infectious w x x y v w x x x y Dipartimento di Scienze Mediche Veterinarie Facolt di Medicina Veterinaria Alma Mater Studiorum Universit v z v v z Virology Unit Institute of Microbiology and v w v v v Institute of w w v w v x x w y x v w w Institute for x y w x Received 16 July 2018 Accepted 27 September 2018 Published xx xx xxxx 2SCIENTIFIC REPORTS 2018 8 15177 w v w v y z w v w y y z is thought to be essential for viral emergence 3 4 T he conf irmation of phenotypic changes acquired in the process of adaptation requires virus isolation which is not always possible In the case of SARS CoV isolation of virus strains from the bat reservoir has widely failed 5 except in singular instances involving bat borne viruses whose spike protein showed an exceptionally high amino acid identity with the epidemic strain 6 8 Additional reasons for low isolation success include the lack of appropriate cell culture systems as well as low virus concentrations and of ten inappropriate storage conditions for samples from f ield investigations Studies of phenotypic properties of reservoir borne viruses can be facilitated by the reconstruction of genetic traits through reverse genetics In the present study we reconstructed variant SARS CoVs carrying dif ferent forms of open reading frame ORF 8 an accessory gene in the SARS CoV genome that is among the most variable genes in bat associated SARSr CoV ORF8 is also one of the most relevant genes when studying potential viral adaptation to humans as the ORF8 coding sequence has undergone gradual deletion during the human epidemic Early epidemic SARS CoVs contained a full ORF8 that is also present in the genomes of almost all bat and carnivore associated precursor viruses 9 12 A 29 nucleotide nt deletion within ORF8 occurred in all strains involved in the middle and late phase of the human epidemic 9 10 Subtotal or total deletions of up to 415 nts occurred within and around the ORF8 region in SARS CoV strains circulating in the very late phase of the epidemic We have previously found that a European SARS related CoV carried by rhinolophid bats in Bulgaria did not contain any ORF8 or similar gene 13 while virtually all SARS related CoV from Asia possess a single continuous ORF8 8 14 15 T he 29 nt deletion in ORF8 was the most obvious genetic change during human to human transmission of SARS causing the expression of truncated gene products termed ORF8a and ORF8b It has been widely hypoth esized that the truncated products led to a modulation of pathogenicity or replication that favored adaptation of SARS CoV to humans 9 11 12 For instance it was found that replication of SARS CoV is increased in cells that overexpress the protein encoded by ORF8a 16 In contrast the same protein led to an attenuation of replication when engineered into recombinant infectious bronchitis virus IBV 17 T he 8b protein when overexpressed was found to induce apoptosis and to be involved in cellular degradation of the viral envelope protein during SARS CoV infection 16 18 19 When engineered into an IBV infectious clone protein 8b inhibited the induction of interferonuni00A0 IFN as triggered by poly I C stimulation and thus enhanced IBV replication T his ef fect was found to involve binding of ubiquitin and interference with induction of the IFN type 1 promoter via IRF3 17 Another study found that the full 122 amino acid protein encoded by ORF8 induces ATF6 dependent transcription which triggers the expression of chaperones and leads to a general attenuation of the protein translation level thus mod ifying the unfolded protein response 20 Loss of ORF8 s original cellular localization in the endoplasmatic retic ulum would have ablated this function in human viruses of the middle and late epidemic phases 20 21 However the ef fect that this change had on SARS CoV replication level remains unclear Studies of SARS CoV replication in experimental mice have not provided indications as to essential functions of ORF8a or 8b 22 However these studies involved high inoculated virus doses and utilized a mouse model that may not ref lect relative changes of replication in transition from bats to humans In sum the available data leave it unclear whether the changes that occurred in ORF8 led to a modif ication of viral replication when comparing bat versus human hosts and specif ically whether the deletion of 29 nt involved an increase of replication level in human cells as of ten impli cated Genome deletions have also been observed in MERS CoV and also here it was speculated that deletion may ref lect adaptation to the human host or the release of selective pressure exerted exclusively in the zoonotic reservoir 23 24 In the present study on SARS CoV we f irst followed up on our initial f inding of the absence of ORF8 in genomes of SARSr CoV from European bats and sequenced rhinolophid bat associated SARSr CoV from four countries across Europe Spain Bulgaria Italy Slovenia We then studied the inf luence of ORF8 on replication in the context of a full replicating SARS CoV genome based on a host transition scenario represented by cell cul tures We constructed recombinant viruses with full ORF8 truncated ORF8 29 nt deletion as well as completely deleted ORF8 Replication was compared in primate cell lines VeroFM MA104 a novel cell line generated from the lung of a rhinolophid bat three additional non chiropteran cell lines as well as human airway epithelial cul tures We f ind that the 29 nt deletion conferred an attenuation of replication level irrespective of the host system studied We have previously described the detection of SARS related CoV in Rhinolophus in Bulgaria 13 T he full length genome of the virus sequenced in that study con tained no ORF8 To further investigate the presence of ORF8 in SARS related CoV in Europe we analyzed fecal samples from rhinolophid bats in Spain Bulgaria Italy as well as Slovenia All f ive Rhinolophus species commonly encountered in Europe were tested positive for coronavirus RNA Tableuni00A01 However none of 92 geographically and phylogenetically representative viruses provided evidence for the presence of an ORF8 gene 19 of these viruses were fully sequenced conf irming absence of ORF8 at orthotopic or heterotopic genome positions To understand the consequences of the absence of ORF8 recombinant SARS CoV rSCV were generated as shown in Fig uni00A01a T he variant termed rSCV 8full contained a complete ORF8 as encountered in bats civet cats as well as in early human cases of SARS thus resembling the pre epidemic and starting phase of the SARS outbreak T he variant termed rSCV epi corre sponds to isolates from the main phase also referred to as middle and late phases of the epidemic containing a 29 nt deletion that introduces a ribosomal frameshif t separating ORF8 into ORF8a and ORF8b Two alternative virus variants without ORF8 were constructed Variant 1 rSCV del8 1 had ORF8 removed without any replace ment Variant 2 rSCV del8 2 had ORF8 replaced by a 5 nt spacer sequence AATAA which occurs instead of ORF8 between ORF7b and the nucleocapsid gene in a natural SARS related CoV variant from European rhinolophid bats 13 3SCIENTIFIC REPORTS 2018 8 15177 w v w v y z w v w y y z Infectious viruses were rescued from cDNA as described 25 24 h af ter co electroporation of in vitro transcribed RNA and nucleocapsid subgenomic RNA parts of the cell culture supernatant were transferred to VeroFM cells and virus replication was monitored by real time RT PCR 72 h post infection hpi rSCV 8full rSCV epi and rSCV del8 2 gave evidence of replication by real time RT PCR while rSCV del8 1 RNA increased more slowly reach ing a plateau level only by 96 hpi Country N individuals tested Rhinolophid species and number of individuals n yielding positive RT PCR Number of RT PCR results indicating absence of ORF8 Bulgaria 506 R euryale n 44 R blasii n 11 R ferrumequinum n 1 R mehelyi n 4 44 11 1 4 Italy 45 R ferrumequinum n 6 6 Slovenia 53 36 R hipposideros n 5 5 Spain 285 R hipposideros n 21 21 Total 872 All 5 species known in Europe 92 100 Table 1 Bat specimens included in this study and conf irmation of absence of ORF8 Absence of ORF8 was determined by RT PCR using reverse primers F29260R TTTGTATGCGTCAATGTGCTTG for reverse transcription and F28182R GGGTCCACCAAATGTAATGCGG and forward primer F27626F GAGAAAGACAGAATGAATGAGC for PCR Figure 1 Generation and evaluation of ORF8 variant recombinant SARS CoV Variants of the open reading frame 8 ORF8 were designed in accordance to their appearance in nature a rSCV 8full represents a single ORF8 as found in reservoir bats and amplif ication hosts in China as well as in the early phase of the SARS epidemic T he middle phase of the epidemic was dominated by a virus variant carrying a 29 nt deletion resulting in the disruption of ORF8 in 2 reading frames 8a and 8b as seen in rSCV epi T he absence of ORF8 is a genomic feature of reservoir bats in Europe and the epidemic virus in the late phase Two deletion variants were constructed Variant 1 rSCV del8 1 perfectly misses ORF8 In variant 2 rSCV del8 2 ORF8 is replaced by the short substitutional sequence AATAA in accordance to the upstream region of the nucleocapsid gene of the European SARS related bat CoV BtCoV BM48 31 Rhi bla Bulgaria 2008 NC 014470 b Plaque morphology of rSCV 8full and rSCV epi were very similar while rSCV del8 1 produced only dif fuse and rSCV del8 2 reduced plaques Western Blot analysis revealed that only rSCV 8full rSCV epi and rSCV del8 2 infected cells expressed the nucleocapsid protein while none could be detected in cells infected with rSCV del8 1 Detection of actin served as loading control Virus replication of the three ORF8 variants rSCV del8 rSCV del8 2 was monitored by plaque titration af ter infection of VeroFM cells at two dif ferent multiplicities of infection 1 c and 0 001 d Virus growth was determined in at least 3 independent experiments in triplicates Shown is one representative experiment Error bars represent standard deviation of the mean 4SCIENTIFIC REPORTS 2018 8 15177 w v w v y z w v w y y z All variants were plaque quantif ied During this process dif ferences in plaque morphology were noted While rSCV 8full and rSCV epi yielded plaques of similar sizes rSCV del8 2 showed smaller plaque size and rSCV del8 1 gener ated no distinct plaques Fig uni00A01b T he absence of plaques in rSCV del8 1 could point to a replication defect caused by the loss of essential gene func tions including the nucleocapsid gene whose expression is indispensable for virus replication 26 Modif ications of ORF8 af fect the upstream context of the transcription regulatory sequence TRS of the nucleocapsid gene 27 Western blot analyses showed that rSCV 8full rSCV epi as well as rSCV del8 2 expressed similar amounts of the nucle ocapsid protein while no expression was seen in cells infected with rSCV del8 1 Fig uni00A01b Subsequent experiments were therefore carried out using variant 2 which is hereaf ter referred to as rSCV del8 Replication of rSCV 8full rSCV epi and rSCV del8 were tested in VeroFM cells at high and low multiplicity of infection MOI 1 and 0 001 plaque forming unit PFU cell Supernatants were harvested at 8 24 48 and 72 hpi and titered in VeroFM cells Fig uni00A01c uni00A0d At MOI 1 30 40 of the cells were already lysed by 24 hpi precluding meaningful comparisons of viruses Fig uni00A01c At MOI 0 001 replication was in the exponential phase by 24 hpi Fig uni00A01d rSCV 8full grew signif icantly more ef f iciently than both other variants at 24 hpi p 0 05 and p 0 001 comparing to rSCV epi and rSCV del8 respectively using t Test T he fully deleted variant rSCV del8 replicated signif icantly less ef f iciently than rSCV epi at 24 hpi p 0 002 t Test By 48 hpi replication of all variants reached plateau levels Due to the essential role of the type I IFN system in the restriction of virus infection it was tested whether the ORF8 dependent replication phenotype might be linked to the type I IFN response MA104 monkey kidney cells were used because they are competent for IFN induction and signaling 28 VeroFM cells would only ref lect dif ferences in IFN signaling due to their defect in IFN beta gene expression 29 According to previous results virus infections with all three variants were carried out at MOI 0 001 and virus progeny was measured at 24 hpi by plaque assay All viruses generally replicated more ef f iciently in VeroFM cells than MA104 cells However relative dif fer ences between strains were very similar in both cell lines suggesting that IFN beta gene induction that is only functional in MA104 cells may not play a signif icant role for the observed phenotypes Fig uni00A02a To test the potential inf luence of the ORF8 gene product on IFN signaling virus growth af ter external addition of IFN was compared To this end VeroFM cells were incubated with indicated concentrations of universal type I IFN alpha prior to infection with rSCV 8full and rSCV 8del At 24 h p i rSCV 8full grew to titers 30 times higher than rSCV del8 Fig uni00A02b If cells were treated with IFN prior to infection the overall replication of both viruses decreased in a dose dependent manner However the replication dif ference between both viruses remained constant T his suggests that ORF8 has a SARS CoV specif ic replication enhancing ef fect that is independent of IFN antagonism In previous studies it has been suggested that the function of ORF8 may be more relevant for replication in cells from the actual animal reservoir than for replication in human or primate cell cultures 30 Because bats of the genus Rhinolophus are the bona f ide animal reservoir for SARS related CoV 2 an epithelial cell line was generated from the lung of Rhinolophus alcyone Fig uni00A03a One pregnant female individual was euthanized and embryonal tissues were prepared using techniques described previously Fig uni00A03b 31 33 T he resulting immortalized Rhinolophus lung cell Figure 2 Replication of ORF8 variants in primate cell cultures with and without IFN pre treatment a VeroFM and MA104 cells were infected with three ORF8 variant viruses at MOI 0 001 Supernatants were harvested at 24 hpi and virus titers determined by plaque titration T he experiment was done in triplicates Signif icance of replication dif ferences between virus variants was determined by t Test p 0 001 p rSCV epi rSCV del8 T hese results did not point to a specif ic ef fect of the gene product of ORF8 on replication in bat cells but rather suggested a general promoting ef fect on virus replica tion conferred by ORF8 To determine whether the ef fect was more general and independent of host taxa a low passage air way epithelial cell line from Sigmodon hispidus cotton rat 34 and primary lung cell lines from Capra hircus goat and Ovis aries sheep were transduced using vesicular stromatitis virus G protein pseudotyped lentiviruses to transiently express hACE2 as described above Fig uni00A04a and subsequently infected with the ORF8 virus vari ants at MOI 0 001 Again the same hierarchy of viral replication levels namely rSCV 8full rSCV epi rSCV del8 was observed at 24 hpi Fig uni00A04b Virus rSCV 8full replicated signif icantly less ef f iciently than rSCV epi in all three non host cell lines t Test p 0 001 p 0 05 p 0 05 7SCIENTIFIC REPORTS 2018 8 15177 w v w v y z w v w y y z T he gradual deletion of ORF8 constituted the most obvious change in SARS CoV af ter emergence Because few changes occurred in other parts of the genome we have focused our study on ORF8 and kept the rest of the genome constant representing the sequence of SARS CoV strain Frankfurt 1 T his prototype virus was isolated from the late phase of the SARS CoV epidemic when transmission chains already occurred in countries outside China 36 Even in this late epidemic strain a reconstitution of the full ORF8 reading frame led to an increased replicative capability suggesting that the viral genome was still compatible with its primordial ORF8 element in spite of possible onward evolution in humans 9 Compared to earlier observations with ORF8 deleted SARS CoV our results show a clear phenotypic dif fer ence in replication in relevant models of human respiratory tract infection T he phenotypic dif ference depends on the presence of full ORF8 with enhanced replication as opposed to the deletion variant Earlier experiments in which no clear dif ferences in replication have been noted were aimed at the identif ication of strong attenuation markers as necessary for live vaccine development and hence worked at very high MOIs at or above 1 22 We have also included such high MOIs in our experiments for comparison and our results were highly concordant with those earlier studies For instance like in the study by Yount et al 22 dif ferences between viruses with full ORF8 and 29nt deletions were only about three fold at MOI 1 Natural infection however does not involve high virus doses 37 When a virus is passed from human to human the stochastic nature of infection success implicates that inocula just below or above one unit of human infectious virus are transmitted For instance during inf luenza A transmission in ferrets as few as 2 virus units were transmitted between animals 38 Under conditions of inter host transmission the observed phenotypic dif ferences as observed in our study may have signif icant ef fects on viral f itness 37 In endemic viral infections loss of f itness should cause viral lineage extinction in competition with more reproductively capable lineages that co circulate within the host population 39 Af ter a single time zoonotic intro duction however competing viral lineages are unlikely to exist Variants with slightly deleterious mutations ran domly selected through transmission bottlenecks can continue to reproduce in spite of reduced f itness an ef fect known as founder ef fect Interestingly it has been shown by in vitro studies that virus populations with reduced initial f itness suf fer less from slightly deleterious mutations than populations that replicate on peak f itness level 40 Reduced initial f itness is a condition that can be expected in early stage zoonotic epidemics when the virus is not yet adapted to the new host environment 41 Considering the present results we therefore suggest that the 29 nt deletion in SARS CoV is the result of a founder ef fect that has permitted survival in spite of reduction of f itness T his interpretation contrasts with the earlier notion that the 29 nt deletion ref lects adaptation to humans T he conclusion of adaptation is partly based on results from expression of ORF8 or its truncation products in overexpression systems or expression in heterologous virus genomes suggesting various inf luences on virus cell interaction 16 21 For instance one study provided evidence for IFN evasion mediated by ORF8b 17 T his is not conf irmed by our experiments studying ORF8 in full virus context Other authors have proposed that the 29 nt deletion may have been neutral for f it ness af ter viral host transition to hum