【病毒外文文献】2002 Murine Coronavirus Spike Glycoprotein Mediates Degree of Viral Spread, Inflammation, and Virus-Induced Immunopathol

Virology 301 109 120 2002 acute virus induced neurological disease We have pre viously demonstrated that the spike S gene is a major determinant ofMHV neurovirulence Recombinant vi ruses containing the spike gene ofthe highly neuroviru lent MHV 4 strain S 4 R exhibit a dramatically more neu rovirulent phenotype 3 log 10 decrease in intracranial LD 50 than isogenic recombinants containing the spike gene ofthe mildly neurovirulent MHV A59 strain S A59 R Phillips et al 1999 S A59 R and S 4 R have all other genes derived from MHV A59 By defining the contribution of direct virus mediated pathology and immune mediated pathology to the differential neurovirulence of S 4 R and S A59 R we can better elucidate the mechanisms by which the MHV 4 spike confers high neurovirulence The MHV spike glycoprotein is required for viral entry and spread Boyle et al 1987 Collins et al 1982 Stur viral spread or the immune response to infection Although MHV infects many cell types in the brain including neurons astrocytes and oligodendrocytes Knobler etal 1981 Lavi etal 1984 1988 Weiner 1973 it has been suggested that neuronal tropism is a major determinant ofMHV neurovirulence Dubois Dalcq et al 1982 Fleming et al 1986 Knobler et al 1981 The high neurovirulence ofS 4 R is associated with extensive spread ofvirus in the brain Phillips et al 1999 In creased viral spread could either reflect increased infec tion ofa particular cell type such as neurons or alter natively increased infection of multiple cell types in the CNS In the CNS clearance ofinfectious MHV requires multiple components ofthe immune response including CD8 H11001 and CD4 H11001 T cells and B cells Adoptive transfer experiments in combination with depletion experiments have demonstrated that both CD8 H11001 and CD4 H11001 T cells are 1 Murine Coronavirus Spike Glycoprotein Mediates and Virus Induced Immunopathology Joanna J Phillips Ming Ming Chua Department of Microbiology University of Pennsylvania School Basic Science The Fox Chase Cancer Center 770 Received February 25 2002 returned to author The mouse hepatitis virus MHV spike glycoprotein is alterations in spike affect neurovirulence using two isogenic containing the MHV 4 spike gene is dramatically more neurovirulent Phillips M M Chua E Lavi and S R Weiss 1999 J Virol cellular tropism viral spread and the immune response to infection spike mediated neurovirulence was associated with extensive Infection of primary hippocampal neuron cultures demonstrated spread may occur between cells in close physical contact into the brain a higher percentage of CD8 H11001 T cells and infection Despite this robust and viral specific immune response immune mediated pathology also contributes to the high neurovirulence INTRODUCTION The severity ofviral infections depends on the extent oftissue destruction and cellular dysfunction mediated by a combination ofdirect virus infection and immune mediated destruction The relative contribution ofthese two components differs depending on a number of virus and host factors including viral tropism rate of viral spread and specificity of the immune response Infection of mice with the murine coronavirus mouse Degree of Viral Spread Inflammation in the Central Nervous System F Rall and Susan R Weiss 1 Philadelphia Pennsylvania 19104 6076 and Division of Avenue Philadelphia Pennsylvania 19111 April 26 2002 accepted May 5 2002 determinant of neurovirulence We investigated how recombinant viruses differing exclusively in spike S 4 R than S A59 R containing the MHV A59 spike gene J J 60 We examined the contribution of differences in to the differential neurovirulence of S 4 R and S A59 R MHV 4 spread in the brain in both neurons and astrocytes S 4 R spread more rapidly than S A59 R and suggested that S 4 R infection induced a massive influx of lymphocytes frequency of MHV specific CD8 H11001 T cells relative S A59 R S 4 R infection infection of RAG1H11002 H11002mice suggested that of S 4 R 2002 Elsevier Science USA man and Holmes 1981 Expressed on the virion surface spike is responsible for binding to the viral receptor and mediating virus cell fusion and subsequent to infection spike expressed on the host cell membrane can medi ate cell cell fusion The spike is also vital for the immune response to infection as it is able to induce both a cell mediated and a humoral mediated immune re sponse Bergmann et al 1996 Castro and Perlman 1995 Collins et al 1982 Thus the high neurovirulence hepatitis virus MHV provides a model for studying To whom correspondence and reprint requests should be ad dressed Fax 215 573 4858 E mail weisssr mail med upenn edu doi 10 1006 viro 2002 1551 conferred by the MHV 4 spike may be due to alterations in a number ofaspects ofinfection including viral entry critical for normal viral clearance Korner et al 1991 0042 6822 02 35 00 2002 Elsevier Science USA 109 Glenn of Medicine 1 Burholme for revision a major 73 7752 77 viral that In addition a higher to All rights reserved Stohlman et al 1986 1995 1998 Sussman et al 1989 Williamson and Stohlman 1990 Yamaguchi et al 1991 Furthermore the peak ofT lymphocyte infiltration into the CNS is coincident with falling titers of infectious virus in the CNS Williamson et al 1991 Relative to infection with S A59 R infection with S 4 R results in a dramatic infil tration ofinflammatory cells into the CNS Phillips et al 1999 Differences in either the composition of these infiltrates or their virus specific activity could have a profound influence on neurovirulence The MHV spike is a major determinant ofMHV neu rovirulence To identify the critical parameters of infec tion mediating the differential neurovirulence of S 4 R and S A59 R we characterized the cellular tropism and compo sition and function of the lymphocytic response to infec tion with S 4 R and S A59 R Our results suggest that the high neurovirulence conferred by alterations in the MHV spike are mediated by both increased viral spread in multiple cell types in the brain and immune mediated pathology RESULTS S 4 R and S A59 R exhibit similar cellular tropism We previously observed that at the peak ofvirus rep lication day 5 postinfection the more neurovirulent S 4 R infects a significantly greater number of cells in the basal FIG 1 Identification of S 4 R and S A59 R infected neurons in the brain 5 days after intracranial inoculation Representative pictures are shown of double immunofluorescent labeling of sagittal brain sections taken from S 4 R inoculated A and B and S A59 R inoculated C and D mice A and C show immunofluorescent labeling of viral antigen in the basal forebrain red B and D show the corresponding labeling of MAP2b in neurons green Note MAP2b positive cells exhibit the characteristic morphology of neurons Arrows identify cells double positive for viral antigen and MAP2b in both S 4 R and S A59 R inoculated brains Note the greater total number ofviral antigen positive cells following inoculation with S 4 R No viral antigen staining was observed in mock infected mice when sections were incubated in preimmune sera or when the primary antibody was omitted data not shown Magnification H11003380 FIG 2 Quantification of viral antigen positive cells in the brain following inoculation with S 4 R and S A59 R The total number ofviral antigen positive cells per brain section following infection with S 4 R open squares and S A59 R open diamonds was determined on days 3 and 5 postinfection On day 5 postinfection two brain sections per mouse were examined and the mean number ofviral antigen positive cells per sagittal section was determined nH11005 5 For the quantification ofviral antigen positive cells on da y 3 a single sagittal section was examined for each infected mouse n H11005 4 The mean number ofviral antigen positive cells per brain section horizontal line following infec tion with S 4 R was significantly greater than with S A59 R on days 3 and 5 postinfection two tailed t test P H11021 0 05 and P H11021 0 001 respectively 110 PHILLIPS ET AL forebrain hippocampus and cingulate gyrus than S A59 R Phillips et al 1999 Increased infection could reflect differences in viral spread in a single cell type or in multiple cell types To quantitatively compare the cellular tropism ofS 4 R and S A59 R we performed double label immunofluorescence on sagittal brain sections for viral TABLE 1 Identification of CNS Cells Infected by S 4 R and S A59 R on Day 5 Postinfection a Virus Animal Section 1 Section 2 No ofviral antigen positive cells No ofcells double positive with AP14 b No ofviral antigen positive cells No ofcells double positive for GFAP b S 4 R 1 599 328 54 8 478 132 27 6 2 1016 493 48 5 1096 111 10 1 3 883 469 53 1 765 148 19 3 4 1192 597 50 1 713 202 28 3 5 249 143 57 4 738 99 13 4 Group mean 52 8 Group mean 19 7 S A59 R 1 529 291 55 0 288 61 21 2 2 97 52 53 6 330 44 13 4 3 239 134 56 1 259 37 14 3 4 403 208 51 6 616 170 27 6 5 326 163 50 0 282 80 28 4 Group mean 53 3 Group mean 21 0 a Mice were inoculated intracranially with 10 PFU ofeither S 4 RorS A59 R Animals were sacrificed on day 5 postinfection and perfused and the brain was fixed in formalin embedded in paraffin and sectioned sagittally Section 1 and 2 are adjacent b The slides were double immunostained with anti MHV A59 serum and AP14 a monoclonal antibody to MAP2b or a mouse anti GFAP antibody Sagittal sections were coded and systematically examined in a blinded fashion magnification H11003190 and the total number ofviral antigen positive cells and double positive cells per section was determined The is the percentage ofthe total number ofviral antigen positive cells counted that were also positive for the indicated marker FIG 3 Viral spread in primary hippocampal neuron cultures Neurons were cultured for 4 days and then infected with S 4 R A and B or S A59 R C and D Representative pictures on day 3 postinfection are shown Colocalization of viral antigen A and C and neuronal cell markers B and D were demonstrated using double immunofluorescence for viral antigen green and MAP2 red The arrows indicate viral antigen positive neurons while the arrowheads indicate viral antigen positive neurites Note the large focus of viral antigen positive cells following infection with S 4 R A Magnification H11003190 111MHV SPIKE MEDIATES VIRAL SPREAD AND INFLAMMATION antigen and MAP2b a neuron specific marker De Cam illi et al 1984 Riederer et al 1995 or glial fibrillary acidic protein GFAP an astrocyte specific marker Mice were inoculated intracranially with 10 PFU ofS 4 RorS A59 R and animals were sacrificed at 3 and 5 days postinfec tion The sections were systematically scanned magnifica tion H11003190 in a blinded fashion and the total number of viral antigen positive and double positive cells was de termined Representative pictures ofdouble immunoflu orescence for viral antigen and MAP2b are shown Fig 1 On days 3 and 5 postinfection the mean number of viral antigen positive cells per brain section was signif icantly greater following infection with S 4 R than S A59 R two tailed t test day 3 P H11021 0 05 and day 5 P H11021 0 001 Fig 2 The pattern ofviral antigen expression in the brain was similar following infection with S 4 RorS A59 R Despite the difference in the absolute number of in fected cells S 4 R had a similar cellular tropism as S A59 R As shown in Table 1 we found that the percentage of viral antigen positive cells that were also positive for neuronal or astrocytic markers was remarkably similar for S 4 R and S A59 R On day 5 postinfection neurons ap peared to account for approximately 53 of either S 4 R or S A59 R infected cells while astrocytes accounted for ap proximately 20 or 21 ofS 4 R or S A59 R infected cells respectively Thus the increased neurovirulence ofS 4 R correlated with increased infection of multiple CNS cell types including both neurons and astrocytes Viral spread of S 4 R and S A59 R differs in primary neuronal cultures The extent ofneuronal infection is thought to be a determinant ofMHV neurovirulence Dubois Dalcq et al 1982 Fleming et al 1986 Knobler et al 1981 As an in vitro model ofCNS infection and to examine the ability of S 4 R and S A59 R to spread in neurons we infected primary hippocampal neuron cultures obtained from B6 mice with the two viruses Immunostaining for MAP2 and GFAP demonstrated that the cultures consisted primarily ofneurons with less than 8 ofthe cells exhibiting GFAP immunoreactivity data not shown On days 0 1 and 3 after infection supernatants were titered for infectious virus and the number ofviral antigen positive cells was determined by immunofluorescence Neurons did not exhibit cytopathic effect for the duration of the experi ment as assessed by bright field microscopy To compare the number ofcells infected with each virus we used immunofluorescence to detect viral anti gen expression S 4 R and S A59 R infection of neurons was confirmed by cell morphology and by double immunos taining for viral antigen and MAP2 Fig 3 Viral antigen positive cells were often located in discrete foci More over viral antigen positive neurites could often be ob served between infected cells or between foci of infected cells suggesting that perhaps one means ofviral spread in these cultures was along neurites To examine the pattern ofviral spread in these cultures we determined the total number ofviral antigen positive cells and the total number offoci ofinfected cells per coverslip at different time points Table 2 On day 1 postinfection the average number offoci and the total number ofantigen positive cells was similar following infection with S 4 R and S A59 R On day 3 postinfection although the average number offoci was similar the average number ofviral antigen positive cells was significantly greater following infection with S 4 R than S A59 R two tailed t test PH11021 0 01 Thus on day 3 the number ofviral antigen positive cells per focus was greater for S 4 R 5 79 H11006 1 15 than S A59 R 3 17 H11006 1 59 mean H11006 SD The difference in the number of infected cells despite a similar number of foci between S 4 R and S A59 R suggested that S 4 R spread faster via cell cell spread than S A59 R Thus the results in primary neuronal cultures were consistent with the results invivo and suggested that the MHV 4 spike conferred an inher ent ability to spread rapidly from cell to cell in the CNS Consistent with previous studies in which primary neu ronal cultures were infected with MHV Dubois Dalcq etal 1982 Pasick et al 1994 the amount ofinfectious virus released into the media was minimal Following infection with S 4 RorS A59 R the titers ofinfectious virus in the super natants initially dropped day 0 to day 1 postinfection and then remained the same from day 1 to day 3 postinfection Table 2 The inactivation ofinfectious virus in the residual inoculum most likely accounted for the precipitous drop in TABLE 2 S 4 R and S A59 R Infection of Primary Neuronal Cultures a Virus Day Viral titers log PFU mL mean H11006 SD b No offoci mean H11006 SD c No ofviral antigen positive cells mean H11006 SD d S4R 0 3 79H110060 39 0 0 1 2 13H110060 34 60H1100617 367H11006108 3 2 28H110060 22 127H1100652 842H11006 31 SA59R 0 4 21H110060 12 0 0 1 3 59H110060 15 122H1100679 302H11006165 3 3 38H110060 19 162H1100634 471H11006109 a Primary hippocampal neuron cultures were infected on day 4 post explant at an m o i of5 Coverslips containing infected neurons were fixed in 2 paraformaldehyde and immunostained for viral antigen mouse antinucleocapsid b Supernatants from infected cultures were collected at the indicated times and viral titers were determined on L2 cells The titers are the mean titers from two independent experiments n H11005 6 c A group ofviral antigen positive cells that appeared to exhibit cell cell contact was defined as a focus The data shown represent the results from two independent experiments n H11005 4 d The total number ofviral antigen positive cells per coverslip was determined n H11005 4 except on day 3 following infection with S 4 R n H11005 3 On day 3 the mean number ofviral antigen positive cells was greater following infection with S 4 R than S A59 R two tailed t test P H11021 0 01 112 PHILLIPS ET AL viral titers from day 0 to 1 postinfection On days 1 and 3 the titers ofinfectious virus were lower in S 4 R infected cultures than S A59 R infected cultures Differences in the composition of the lymphocytic response to infection with S 4 R and S A59 R The MHV 4 spike appears to mediate more efficient spread in cells ofthe CNS than the MHV A59 spike Previously we observed a difference in the immune re sponse to infection with S 4 R and S A59 R by H therefore the total frequency of MHV specific CD8 H11001 T cells in the CNS following S 4 R infection was approximately 22 9 The MHV A59 spike lacks the immunodominant S510 epitope Luytjes et al 1987 Thus following infection with S A59 R the MHV specific CD8 H11001 T cell response was S598 spe cific and accounted for approximately 11 8 of the CD8 H11001 T cell response Within the population ofcells isolated from the CNS a proportion ofCD4 H11001 cells exhibited intracellular IFN H9253 in the absence ofpeptide Fig 5B The reason for this background level ofIFN H9253 expression is not known how ever it has been suggested that infected microglia and macrophages in the CNS cell preparation may stimulate CD4 H11001 T cells Haring et al 2001 The frequency of M133 specific CD4 H11001 T cells following infection with S 4 R or S A59 R was similar accounting for approximately 16 0 and 15 4 ofthe CD4 H11001 T cell response respectively As this may overestimate the frequency of MHV specific cells in the CNS we also subtracted the percentage of IFN H9253 expressing cells in the absence ofpeptide from the total percentage ofIFN H9253 expressing cells after pep tide stimulation The calculated frequency of M133 spe cific CD4 H11001 T cells following infection with S 4 R 10 6 and S A59 R 8 9 was also similar Thus infection with either S 4 RorS A59 R resulted in a robust CD8 H11001 and CD4 H11001 T cell 113MHV SPIKE MEDIATES VIRAL SPREAD AND INFLAMMATION response that was both epitope specific and functional The major difference between the frequency of virus specific T cells induced by S 4 R and S A59 R was the in creased frequency of functionally activated CD8 H11001 T cells following infection with the more neurovirulent S 4 R Note that due to the increased numbers ofinflammatory cells in the brain following S 4 R infection relative to S A59 Rin fection the total number of virus specific CD8 H11001 and CD4 H11001 T cells in the brain was greater following infection with S 4 R than S A59 R Contribution of lymphocytes to differential neurovirulence S 4 R infection resulted in a greater number of lympho cytes and a higher frequency of virus specific immune FIG 4 Lymphocyte populations in the brain following inoculation with S 4 R and S A59 R Mice were infected with S 4 RorS A59 R and animals were sacrificed at the peak of inflammation day 7 postinfection A The upper row depicts the forward scatter and side scatter characteristics of representative mononuclear cell preparations from the brains of S 4 R and S A59 R infected mice Surface expression of CD45 was used to assist in the gating ofmononuclear cells and representative flow cytometry density plots showing expression ofCD45 x axis and CD8 y axis in S 4 R infected mice left lower row and S A59 R infected mice right lower row Numbers in the upper right quadrants represent the percentage of CD8 H11001 CD45 H11001 monocytes and the numbers in the right lower quadrant represent the percentage ofCD8 H11002 CD45 H11001 monocytes B Relative percentages ofCD8 H11001 CD4 H11001 and B220 H11001 cells in brain derived monocytes C Absolute numbers oftotal cells CD45 H11001 CD8 H11001 CD4 H11001 and B220 H11001 cells in cell preparations from the brain Mononuclear cell preparations from six mice were pooled and the results