405 Received October 15, 2015. Accepted December 28, 2015. J-STAGE Advance Publication February 19, 2016. DOI: 10.7883/yoken.JJID.2015.499 *Corresponding Author: Mailing address: Department of Bioinformatics, Graduate School of Engineering, Soka University, Tangi-cho, Hachioji, Tokyo 192-8577, Japan. Tel and Fax: ＋81-426-9465, E-mail: rihitow＠soka.ac.jp 405 Jpn. J. Infect. Dis., 69, 405–413, 2016 Original Article Contribution of Lewis X Carbohydrate Structure to Neuropathogenic Murine Coronaviral Spread Masatoshi Kakizaki1, Akira Togayachi2, Hisashi Narimatsu2, and Rihito Watanabe1* 1Department of Bioinformatics, Graduate School of Engineering, Soka University, Tokyo 192-8577; and 2Glycoscience and Glycotechnology Research Group, Biotechnology Research Institute for Drug Discovery, National Institute of Advanced Industrial Science and Technology (AIST), Ibaraki 305-8568, Japan SUMMARY: Although Lewis X (Lex), a carbohydrate structure, is involved in innate immunity through cell-to-cell and pathogen recognition, its expression has not been observed in mouse monocytes/macrophages (Mo/Mas). The Mo/Mas that infiltrate the meninges after infection with the neuropathogenic murine coronavirus strain srr7 are an initial target of infection. Furthermore, higher inflammatory responses were observed in gene-manipulated mice lacking a1,3-fucosyltransferase 9, which determines the expression of the Lex structure, than in wild type mice after infection. We inves- tigated Lex expression using CD11b-positive peritoneal exudate cells (PECs) and found that Lex is in- ducible in Mo/Mas after infection with srr7, especially in the syncytial cells during the late phase of infection. The number of syncytial cells was reduced after treatment of the infected PECs with anti-Lex antibody, during the late phase of infection. In addition, the antibody treatment induced a marked reduction in the number of the infected cells at 24 hours post inoculation, without changing the infected cell numbers during the initial phase of infection. These data indicate that the Lex structure could play a role in syncytial formation and cell-to-cell infection during the late phase of infection. INTRODUCTION cl-2 virus (cl-2), isolated from a neuropathogenic mu- rine coronavirus mouse hepatitis virus (MHV) strain, JHM virus (JHM) (1), exhibits extremely high neu- rovirulence after infection, leading infected mice to morbid states within 3 days post-inoculation (dpi), without inducing demyelination in the white matter of the brain that is observed with other JHMs (2). Instead, it produces necrotic lesions in the grey matter with in- fected neurons, which do not express the major MHV receptor carcinoembryonic antigen-related cell adhesion molecule 1 (CEACAM1) (3,4). A mutant clone isolated from cl-2, srr7, shows attenuated neurovirulence com- pared to cl-2, although its virulence is still higher than that of the other JHMs, exhibiting rapid viral spread from one organ to another non-adjacent organ within 12 hours after infection, designated as super-acute spread (5). Most of the mice infected with srr7 survive for more than 7 days and die within 10 dpi. The viral antigens in the brain are detectable in the white matter. However, the neuropathological characteristics are not demyelinating inflammation, but spongiosis, which ap- pears 48 hours post-inoculation (hpi) (6). Initial viral antigens, after inoculation with either cl-2 or srr7, are detectable in CD11b- or F4/80-positive monocytes/ macrophages (Mo/Mas) that have infiltrated the meninges. Syncytial giant cells formed of infected Mo/ Mas are often found in the meninges and ventricular cavity (5). The neuropathological lesions in the brain, including demyelination (7,8), induced by JHM infection are con- sidered to be, to some extent, the results of an indirect mechanism of infection. srr7 induces spongiotic lesions in areas where no viral antigens are detectable during the early phase of infection (6), without prominent in- flammation, but with elevated levels of proinflammato- ry cytokines in the infected brain (9). Damage of neu- rons in the hippocampus was reported to result from bystander effects of infection with Mu-3, a mutant iso- lated from srr7 (10), and another kind of virus, such as Theiler’s murine encephalomyelitis virus (TMEV) (11). Mechanisms of human brain injuries after infection with human immunodeficiency virus (12), human T-cell leukemia virus I (13), measles virus (14), and influenza virus (15) have also been discussed from the perspective of immune-triggered bystander effects. In the bystander lesions, an antigenic epitope with a Lewis X (Lex: galactose b1-4 Fucose a1-3) N-acetyl- glucosamine carbohydrate structure was recently found to be involved in infection with srr7, in mutant mice that lack the gene encoding a1,3-fucosyltransferase 9 (FUT9) (9). FUT9 determines the final structure of Lex and the gene-manipulated mice Fut9-/- reportedly do not express Lex (16). However, we failed to clearly detect in vivo Lex expression in the infected organs, including the spleen and brain, of wild type mice. The result might be a false negative due to procedures to prepare sections from tissues for immunohistochemistry. Therefore, we examined Lex expression after the infection using cul- tured Mo/Mas derived from peritoneal exudate cells (PECs). In this study, we showed for the first time that the Lex structure was inducible in Mo/Mas after infec- tion. 406 Table 1. Antibodies and reagents used in this study Target Species Clone or designation Conjugate Reference/supplier Primary antibodies for immunofluorescence staining JHMV Rabbit Polyclonal Purified 3 Gr-1 Rat RB6-8C5 Biotin BioLegend, San Diego, CA, USA CD11b Rat M1/70 Biotin BD Pharmingen, San Diego, CA, USA F4/80 Rat BM8 Purified eBioscience, San Diego, CA, USA Lex Mouse MC480 Purified Developmental Studies Hybridoma Bank, University of Iowa, Ames, IA, USA FUT9 Mouse Monoclonal Purified 22 Sialyl-Lex Mouse CSLEX1 Purified BD Biosciences, San Jose, CA, USA Primary antibodies for flow cytometry Gr-1 Rat RB6-8C5 FITC eBioscience CD11b Rat M1/70 PE eBioscience F4/80 Rat BM8 APC eBioscience Secondary antibodies for immunofluorescence staining Rabbit IgG Sheept Polyclonal FITC Abcam, Tokyo, Japan Rat IgG Donkey Polyclonal Alexa568 Abcam Mouse IgG Goat Polyclonal Alexa488 Molecular Probe, Carlsbad, CA, USA Mouse IgG Donkey Polyclonal Biotin Rockland, Gilbertsville, PA, USA Streptavidin Alexa488 Invitrogen, Carlsbad, CA, USA Streptavidin Alexa568 Invitrogen Reagent to stain nuclei Hoechst33342 Invitrogen Antibodies for inhibition assay of Lex Lex Mouse MC480 Purified Developmental Studies Hybridoma Bank Sialyl-Lex Mouse CSLEX1 Purified BD Biosciences Gr-1, granulocyte-differentiation antigen-1; FITC, Fluorescein isothiocyanate; PE, Phycoerythrin; APC, Allophycocyanin. 406 MATERIALS AND METHODS Viruses and Animals: A neuropathogenic MHV strain, srr7, was used in this study. This virus was propagated and titrated in DBT cells maintained in Dul- becco's modified Eagle's minimal essential medium (Gibco, Grand Island, NY, USA ) supplemented with 5z heat-inactivated fetal bovine serum (FBS; Sigma, Tokyo, Japan), as previously described (17). Specific pathogen-free inbred BALB/c mice purchased from Charles River (Tokyo, Japan) were housed in a specific pathogen-free animal facility and maintained according to the guidelines set by the ethical committee of our uni- versity. For the experiment, mice were transferred to a biosafety level 3 (BSL-3) laboratory, after permission from the committee of our university. Six to seven- week-old mice were used for all experiments. Preparation of peritoneal exudate cells: Thioglycol- late medium (TM) was prepared by dissolving 20 g of Brewer thioglycollate (TG) medium (BD Diagnostics, Loveton Cirvle Sparks, MD, USA), supplemented with 10.0 g of proteose peptone (BD Diagnostics), 2.38 g of sodium chloride (Wako, Osaka, Japan), and 0.945 g of dipotassium phosphate (Wako), in 1,000 ml of distilled water to obtain a TM with components similar to those of the classical product provided by Difco Laboratories (Detroit, MI, USA) (18). TM was autoclaved for 20 min at 15 lbs of pressure at 1219C, and was kept in the dark under sterile conditions at room temperature. Four days after the injection of 2 mL of TM into the peritoneal cavity, the peritoneal cavity was washed with phos- phate-buffered saline (PBS; Nissui, Tokyo, Japan), and PECs were collected. To obtain adherent PECs, the cells were seeded at 2 × 105 cells per well in 8-well glass- bottom chamber slides (Nalge Nunc International, Rochester, NY, USA), and were cultured in the RPMI 1640 medium (Gibco) supplemented with 10z heat- inactivated FBS (Gibco). After 24 hours, the adherent cells were inoculated with 1.3 × 103 or 6.5 × 103 plaque-forming units (PFU) of srr7 virus per well and incubated for a specified time. To prepare PECs for flow cytometric analysis, they were cultured in suspen- sion for 24 hours at 2 × 107 cells in 1 mL of the medium described above, in 1.5-mL microtubes, with continu- ous rotation at 20 rpm. Assay of adherent PECs: For immunostaining, ad- herent PECs were fixed in cold ethanol on chamber slides for 1 min, followed by fixation in cold acetone for 5 min; antibodies and reagents listed in Table 1 were used. Cell cultures labeled with fluorescent dyes were mounted with a gold antifade reagent (Invitrogen, Carlsbad, CA, USA). Fluorescence was visualized un- der either a confocal laser scanning microscope (Leica Microsystems, Heidelberg, Germany) or a fluorescence microscope (Keyence, Osaka, Japan). The number of infected cells was counted using the BZ analyzer (Keyence). For the inhibition test, either mouse anti- Lewis X mAb (1 : 100) or mouse anti-Sialyl-Lewis X mAb (1 : 100) (Table 1) was added to the culture medi- um. Flow cytometry: After incubating suspension culture for 24 hours, PECs were inoculated with 1.3 × 105 PFU of srr7 virus and were cultured for further 24 hours. Subsequently, the PECs were washed twice in PBS and resuspended in fluorescence-activated cell sorting (FACS) buffer, which contains 1z bovine serum 407 407 Lewis X Expression in Coronaviral Infection albumin (BSA; Sigma) and 0.1z NaN3 in PBS. For phenotypic analysis, 1 × 106 cells for each staining were initially incubated with Fc block (BD Biosciences, San Jose, CA, USA) (5 mg/mL) for 20 min at 49C. Subse- quently, the cells were incubated with the appropriate fluorochrome-conjugated antibodies (Table 1) for 30 min at 49C, followed by 3 washes with FACS buffer, and were acquired on a FACS Calibur (BD Biosciences). Appropriate isotype controls were used in all cases. Data were analyzed using FlowJo software (Tree Star, Jackson, OR, USA). RESULTS Population and susceptibility of PECs: The PECs that we used were collected after the injection of TM into the peritoneal cavity, and were expected to be mix- ed populations of leukocytes (19,20). FACS analysis re- vealed high expressions of CD11b and F4/80 in 50–70z of the PECs (Fig. 1). Around 35z of the PECs were Gr-1-positive (Fig. 1). After viral inoculation, the popu- lation appeared grossly similar. However, on double staining for CD11b and Gr-1 (CD11b:Gr-1), a small population (Fig. 1B) showed a relatively high expression of both antigens after infection. Based on F4/80:Gr-1 (Fig. 1B) and F4/80:CD11b (Fig. 1A), similar shifts of the population were observed with the upregulation of Gr-1 and CD11b, respectively, without a significant change in the F4/80 expression level after infection. The PECs attached to the culture glass slides, examined by immunofluorescence staining (IF), also exhibited high expression rates of CD11b and F4/80, in around 70z and 40z of PECs, respectively (Fig. 2A). The popula- tion of Gr-1-positive cells among PECs was around 20z. After the inoculation, the population of PECs did not show a marked difference compared to that of cul- ture without viral inoculation (Fig. 2A). Through the inoculation of 2 × 105 PECs with 1.3 × 103 PFU of srr7, the PECs were found to be susceptible to srr7 infection. The infected area in the culture plate spread rapidly between 12 and 24 hpi (Fig. 2B) forming syn- cytia (Fig. 3), all of which bore viral antigens (Fig. 4). A few syncytia had already formed before 12 hpi (Fig. 5C). The susceptibility of PECs to srr7 seemed to be comparable to that of DBT cells, which have been used to prepare virus stock with a high titer (5), because a viral inoculation with 1.3 × 103 PFU, produced and titrated using DBT cells, generated similar numbers of syncytia as in the PECs at 24 hpi (Fig. 5C). However, the level of viral production measured in the culture medium was lower in PECs compared to DBT cells, i.e., PFU at 24 hpi in PEC culture were 1–2 × 102, whereas those in DBT cell culture ranged from 5 × 104–2 × 105 (data not shown). Expression of antigens in the PECs: Interestingly, most of the syncytia (＞95z) showed the expression of Gr-1, which were also CD11b-positive (Fig. 3A and B). Solitary CD11b-positive (CD11b＋) cells, which did not form syncytia with or without viral antigens, were also observed (Fig. 3A). Some CD11b＋ and viral antigen- negative (V－) cells were detected extending their long foot processes and connecting to the V＋ cells (Fig. 3A1–3). Such a connection, using the long foot process of V－ CD11b＋ cells, was not detectable in Gr-1＋ cells, partially because most of the Gr-1＋ cells carried viral antigens after the inoculation (Fig. 3B). After the inoculation, Lex antigen appeared in the PECs cultured on the glass slides (Fig. 3C), but was not detected in the mock-infected PECs (Fig. 3D). Most of the Lex antigens were detected in the syncytial cells (Fig. 3C). Some cells with a single nucleus, bearing the Lex antigens without viral antigen expression, were found around the syncytia (Fig. 3C). Unlike CD11b expression in the foot process of PECs, the fine projection was not traceable using anti-Lex antibody (Fig. 4A). However, this does not mean that Lex was not expressed there, because the available antibodies that we used failed to detect Lex expression in monocytes, which bear CEACAM1 that has been proven to be modified by the Lex structure (21), in the normal state (Fig. 3D). Beside CD11b＋Lex＋ cells, Gr-1＋Lex＋ PECs also appeared as solitary cells with a single nucleus as well as in syncytia with multiple nuclei (Fig. 4B). FUT9, an enzyme which determines Lex expression (22), showed a similar manner of expression in PECs after infection as Lex, i.e. most of the FUT9-positive cells formed syncytia with multiple nuclei (Fig. 4D), which carried CD11b and Gr-1 antigens (Figs. 4C and D). FUT9 antigens were also detected in some solitary cells as well as Lex. A noticeable difference between the Lex and FUT9 appearance was observed after kinetic studies. At 12 hpi, Lex was not detectable even in the syncytia (Figs. 4E1 and 2). In contrast, FUT9 antigen was detected at 12 hpi (Figs. 4E3–5). In order to examine whether or not Lex is involved in syncytium formation, we added anti-Lex antibody (anti-Lex) to the PEC culture. For the control experiment, we used monoclonal antibody against sialyl-Lex (anti-sialyl-Lex), which detects human sialyl-Lex, but not that of mice (23), as shown in Fig. 5A. The number of syncytia was decreased when the cultures were treated with anti-Lex during 0–24 hpi, compared to those with anti-sialyl-Lex treatment (Figs. 5A and C). However, the treatment with antibodies did not cause reductions in the cell num- bers of the adherent cells (Figs. 5B and E). Further- more, application of the antibody before infection did not lead to significant differences between treatments with anti-Lex and anti-sialyl-Lex (Figs. 5A and C). The change in the syncytial numbers after treatment with the antibody does not mean a change in the number of infected foci, in other words initial infectivity of the viruses; instead, it means a change in the number of the syncytia that grew large enough to be detectable under a light microscope. Therefore, infected foci during the early phase of infection were counted for viral antigens at 12 hpi using IF, which proved that there were no differences in the numbers of infected cells (Fig. 5D) and the numbers of syncytia (Fig. 5C) at 12 hpi, among cultures with or without these antibodies; corresponding to our previous observation (9) that viral infectivity of PECs derived from Fut9-/- and wild type mice was the same. The treatment with anti-Lex reduced the number of viral antigen-positive cells as well, when counted at 24 hpi (Fig. 5D), which indicated that the antibody in- hibits not only syncytium formation but also cell-to-cell infection. 408 Fig. 1. (Color online) Flow cytometry of PECs in suspension culture populations of CD11b-, Gr-1-, and F4/80- positive PECs after suspension culture for 24 hours (h) with or without viral infection (Infected [24 hpi] or Unin- fected, respectively) were studied by flow cytometric analysis. A): The cells in dotted circle showed a higher expres- sion level of CD11b after infection. B): The cell populations were divided into 4 regions by forward- and side-scat- tering (FSC and SSC, respectively) profile. The antigen expression patterns of each region (R1-R3) are shown. 408 409 Fig. 2. Cell populations of PECs in adherent cell culture. A): Populations of CD11b-, Gr-1-, and F4/80-positive adherent PECs cultured for 24 h on chamber slides by immunofluorescent staining, with or without viral infection (24 hpi or Uninfected, respectively). B): The ratio of infected adherent PECs, detected by immunofluorescent staining for viral antigens, at desired intervals is shown. The adherent PECs were infected with 1.3 × 103 PFU of srr7 per well. Each infection was conducted in triplicate, and antigen-positive cells of each well were counted. Ver- tical lines indicated in the bar are the mean ± SD. 409 Lewis X Expression in Coronaviral Infection DISCUSSION Although TG-elicited PECs (TG-PECs) were in- troduced as phagocytes with high purity (24), recent stu- dies using flow cytometry revealed that these TG-PECs have heterogeneous population of leukocytes (19). Even macrophages, identified as CD11bhiF4/80hi-int, after gat- ing out B cells, neutrophils, eosinophils, and DC s from the TG-PECs through flow cytometry, appeared heter- ogeneous, based on the expression of MHC II and Ly6C antigens (20). Various populations of macrophages among TG-PECs have been reported depending upon the gating strategy, including side- or forward-scatter profiles using flow cytometry, or the efficiency of the antibodies used (25). The report of 86–95z macro- phages in the TG-PECs (26–28) was criticized as an overestimation by the research group of Misharin (20), who claimed that only 40–45z macrophages and as high as 30–40z eosinophils comprise TG-PECs. Simi- larly, reports on the population of Gr-1-positive cells in the TG-PECs range from 1z (19) to 40z (20). Our esti- mated populations of CD11b＋, F4/80＋, or Gr-1＋ cells in the TG-PECs fell into the ranges reported previously. Gr-1＋ cells among the TG-PECs were 20–35z. These PECs, composed of heterogeneous cell popula- tions, were found to be susceptible to Srr7 infection with a similar efficiency compared to DBT cells, which are used for the titration of MHV. It should be noted that most of the syncytia formed after infection of TG- PECs expressed Gr-1 antigen. The main target among leukocytes of MHV-JHM infection has been believed to be the lineages of Mo/Mas, including microglia in the brain (5,29), which express CEACAM1. CEACAM1 is also detectable as an adhesion molecule on Gr-1-positive human granulocytes (22). However, it has not been reported that Gr-1-positive cells are the main targets of MHV infection. To determine whether the colocaliza- tion of viral and Gr-1 antigens occurred through preferential viral infectivity of Gr-1＋ cells, further stu- dies need to be conducted, because Gr-1 antigen might be inducible by endogenous or exogenous factors (30). Actually, we observed that a small population, with polymorphism among PECs, (Fig. 1B) showed higher Gr-1 expression after infection compared to that detect- ed before infection. Among these Gr-1-positive cells, Lex and FUT9 anti- gen expressions were induced after infection. The anti- gens were detected predominantly in the infected Gr1- and CD11b-positive cells, which formed syncytia. The synthesis of Lex was not involved in the initial formation of syncytia, because Lex-positive syncytia appeared dur- ing the late phase of infection and the inhibition of syn- cytium formation by anti-Lex was detected during the late phase of infection, but not during the early phase. The Lex carbohydrate structure does not affect the ini- tial viral infectivity of TG-PECs, because the pre-incu- bation of TG-PECs with anti-Lex did not change the number of the infected cells, and the TG-PECs derived from Fut9-/- mice, which do not express Lex (16), showed the same infectivity as those from wild type mice (9). Therefore, it is indicated that the Lex structure decorating CEACAM1 has no contribution to the at- tachment of the virus. A possibility that this structure was involved in the increase in Lex antigen expression is unlikely, because we did not observe any increase in the CEACAM1 expression after infection, using IF (data not shown). Rather, the expression of the receptor could have been downregulated, like in other kinds of viral in- fections, including one with human coronavirus (31). In addition to the roles during development, the Lex carbohydrate structure, known as stage-specific em- bryonic antigen-1 (16), is involved in innate immunity through cell-to-cell and pathogen recognition (21,32-34). Dynamic immune regulation is induced after interaction between Lex carbohydrate structures and lectins serving as C-type lectin receptors (35), which include scavenger receptor with C-type lectin (SRCL), human dendritic cell-specific intercellular adhesion molecule-3-grabbing non-integrin (DC-SIGN), and its mouse homologue SIGN-related-1 (SIGNR1) (36,37). This CRL-mediated signaling can be turned to a pro- (37) or anti-inflammatory status (38,39) depending on 410 Fig. 3. (Color online) Antigen expression in the syncytia and solitary cells. Twenty-four hours after seeding PECs on 8-well chamber slides, the cultured cells were infected with 1.3 × 103 PFU of srr7 per well, or left uninfected (V[－]). At 24 hpi, culture slides were processed to detect JHMV (V, A–D), CD11b (A), Gr-1(B), and Lewis X (Lex, C and D) antigens. Hoechst 33342 (Hoe) was used for nuclear counter-staining, which is shown as a digital blue color (/b). Digital red and green colors are indicated as /r and /g in the pictures, respectively. Note that around syncytia are solitary Gr-1 or CD11b-positive cells not forming syncytia (A and B). Around syncytia, most of the solitary CD11b-positive cells were not infected (narrow arrows in A), whereas all solitary Gr-1-positive cells were infected (narrow arrows in B). Some of these solitary cells around the syncytia projected long foot processes connected with syncytia (bold arrows in A, shown at a higher magnification in A3). At 24 hpi, Lex antigens were detected in the syncytial cells (C). A few unifected cells with a single nucleus around syncytia expressed Lex (nar- row arrows in C). In the mock-infected PECs, Lex antigens were not detected (D). White single and double bars in- dicate 50 and 25 mm, respectively. 410 the glycan composition or agents that trigger the im- mune reaction. Our previous study indicated that Lex expression is involved in signaling that negatively in- fluences the immune responses in the course of srr7 in- fection using Fut9-/- mice. In addition, the Fut9-/- mice, which do not express Lex, showed extended inflammato- ry responses compared to wild type mice (9), which ex- hibited a poor inflammatory reaction in the infected brain accompanied by a reduced population of leuko- cytes in the spleen, after infection with srr7 (40). In spite of these important roles of Lex carbohydrate structures in immune response, the expression of Lex has not been detected by IF or flow cytometry in mouse leukocytes, even in monocytes, which express CEA- CAM1. However, it has been demonstrated, by mass spectrometry, that CEACAM1 carries rich Lex carbohy- drate structures (21). One of the reasons may be that the most sensitive anti-Lex antibody we examined was of the IgM class, which might be too large to access antigenic sites that are covered by a tertiary conformation of pro- 411 Fig. 4. (Color online) Expressions of Lex and Fut9. Adherent PECs on culture slides were stained with Lewis X (Lex; A, B, E1, and E2), CD11b (A and C), Gr-1 (B and D), JHMV (V; E), and Fut9 (D and E3-E5) antigens at 12 hpi (E) and 24 hpi (A-D). Hoechst 33342 (Hoe) was used for nuclear counter-staining, which is shown as a digital blue color (/b). Digital red and green colors are indicated as /r and /g in the pictures, respectively. At 24 hpi, most of the CD11b- or Gr-1-positive syncytia carried Lex (arrowheads in A and B) or FUT9 (arrowheads in C and D) antigens. Lex or FUT9 antigens were also detected in some solitary cells (narrow arrows in B–D). The CD11b-posi- tive fine projection (narrow arrows in A1) was not traceable using anti-Lex antibody (A2), shown at a higher mag- nification (A4 and A5). At 12 hpi, Lex was not detected even in syncytia (E1 and E2), whereas Fut9 was detected in the initial small syncytia at 12 hpi (narrow arrows in E3–5). White single and double bars indicate 50 and 25 mm, respectively. 411 Lewis X Expression in Coronaviral Infection teins or other carbohydrates. Nevertheless, we showed that highly expressed Lex was detected in the syncytia and satellite cells (SCs) around the syncytia using Mo/Mas attached to glass slides, which could have been beneficial for the antibodies to reach the antigenic sites in the Mo/Mas, with a fairly well-preserved cell ar- chitecture after fixation procedures were employed on the cells attached to the glass slides (41). Furthermore, using Mo/Mas attached to glass slides, we could detect that some SCs projected long foot processes connecting to syncytia, mimicking immune synapse, enabling cross- talk between leukocytes several hundred mm apart (42). The connection, through foot processes, between SCs and initial small syncytia might be necessary for them to grow large enough to become detectable with a light microscope, during the late phase of infection, and Lex might be involved in this process. Expression levels de- tectable by IF during extremely virulent viral infections would provide an opportunity to clarify the functional roles of Lex and its lectins, during infection and immune responses directed not only by mature Mo/Mas but also by dendritic cells, because TG-PECs contain immature CD11b＋Gr-1＋ cells that can differentiate into dendritic cells (43). Furthermore, among these CD11b＋Gr-1＋ 412 Fig. 5. (Color online) Effects of anti-Lex antibody. Anti-Lex (anti-Lex) or anti-SLex (anti-SLex) antibody were added to the PEC culture well before viral inoculation for 1 hour (＋/) and at the time of inoculation (/＋), or PECs were cultured without treatment with the antibodies (designated as -/ or /-, depending on the periods of culture, that is before or after viral inoculation, respectively). Virus was inoculated at an M.O.I of 0.02 (MOI, 0.02) through the experiments except for D-2) to count viral antigen positive cells during the early phase of infection at a higher M.O.I. (MOI, 0.1) A) and B): Immunofluorescent staining for viral antigens (V/g) and SLex (SLex/g) (A), or nuclear staining using Hoechst33324 (Hoe) (B)), with viral infection at 24 hpi (24 hpi) or without (Uninfected). White single bar indicates 50 mm. C), D), and E): Culture slides were processed for immunofluorescent staining for viral antigens (C) and D)) or nuclear staining (E), and photographed using a fluorescent microscope (Keyence). The number of viral antigen-positive cells or nuclei was counted using a BZ analyzer (Keyence). A syncytium was identified when a viral antigen-positive cell had more than 3 nuclei with a major cytoplasmic axis of more than 50 mm. Each experiment was conducted in triplicate, and viral antigen-positive cells or nuclear numbers in each well were counted. p, P-value by Student's t-test. Asterisks indicate p ＜ 0.05. Vertical lines are the mean ± SD. 412 precursor cells, there are myeloid-derived suppressor cells (MDSC) (30), which might expand their popula- tion, after extremely virulent viral infection, to contrib- ute to a rapid spread of the viruses. A small population detected by flow cytometry as a shifted population after infection, with elevated expression of CD11b＋ and Gr-1＋, may be a candidate for MDSC. The detection of Lex expression in the population is in progress in our laboratory. Ackowledgments We thank Ms Nami Suzuki of National Institute 413 413 Lewis X Expression in Coronaviral Infection of Advanced Industrial Science and Technology for technical support to flow cytometric analysis. Conflict of interest None to declare. REFERENCES 1. Taguchi F, Siddell SG, Wege H, et al. Characterization of a variant virus selected in rat brains after infection by coronavirus mouse hepatitis virus JHM. J Virol. 1985;54:429-35. 2. Bender SJ, Weiss SR. Pathogenesis of murine coronavirus in the central nervous system. J Neuroimmune Pharmacol. 2010;5: 336-54. 3. Matsuyama S, Watanabe R, Taguchi F. Neurovirulence in mice of soluble receptor-resistant (srr) mutants of mouse hepatitis virus: intensive apoptosis caused by less virulent srr mutant. Arch Virol. 2001;146:1643-54. 4. Godfraind C, Langreth SG, Cardellichio CB, et al. Tissue and cellular distribution of an adhesion molecule in the carcinoem- bryonic antigen family that serves as a receptor for mouse hepati- tis virus. Lab Invest. 1995;73:615-27. 5. Takatsuki H, Taguchi F, Nomura R, et al. Cytopathy of an in- filtrating monocyte lineage during the early phase of infection with murinecoronavirus in the brain. Neuropathology. 2010;30: 361-71. 6. Kashiwazaki H, Nomura R, Matsuyama S, et al. Spongiform de- generation induced by neuropathogenic murine coronavirus in- fection. Pathol Int. 2011;61:184-91. 7. Glass WG, Chen BP, Liu MT, et al. Mouse hepatitis virus infec- tion of the central nervous system: chemokine-mediated regula- tion of host defense and disease. Viral Immunol. 2002;15:261-72. 8. Glass WG, Hickey MJ, Hardison JL, et al. Antibody targeting of the CC chemokine ligand 5 results in diminished leukocyte in- filtration into the central nervous system and reduced neurologic disease in a viral model of multiple sclerosis. J Immunol. 2004; 172:4018-25. 9. Kashiwazaki H, Kakizaki M, Ikehara Y, et al. Mice lacking a1,3- fucosyltransferase 9 exhibit modulation of in vivo immune responses against pathogens. Pathol Int. 2014;64:199-208. 10. Kakizaki M, Kashiwazaki H, Watanabe R. Mutant murine hepa- titis virus-induced apoptosis in the hippocampus. Jpn J Infect Dis. 2014;67:9-16. 11. Buenz EJ, Sauer BM, Lafrance-Corey RG, et al. Apoptosis of hippocampal pyramidal neurons is virus independent in a mouse model of acute neurovirulent picornavirus infection. Am J Pathol. 2009;175:668-84. 12. Mattson MP, Haughey NJ, Nath A. Cell death in HIV dementia. Cell Death Differ. 2005;12:893-904. 13. Szymocha R, Akaoka H, Dutuit M, et al. Human T-cell lym- photropic virus type 1-infected T lymphocytes impair catabolism and uptake of glutamate by astrocytes via Tax-1 and tumor necro- sis factor alpha. J Virol. 2000;74:6433-41. 14. Andersson T, Schwarcz R, Love A, et al. Measles virus-induced hippocampal neurodegeneration in the mouse: a novel. Neurosci Lett. 1993;154:109-12. 15. Wang GF, Li W, Li K. Acute encephalopathy and encephalitis caused by influenza virus infection. Curr Opin Neurol. 2010;23: 305-11. 16. Kudo T, Fujii T, Ikegami S, et al. Mice lacking a1,3-fucosyltrans- ferase IX demonstrate disappearance of Lewis x structure in brain and increased anxiety-like behaviors. Glycobiology. 2007;17:1-9. 17. Taguchi F, Yamada A, Fujiwara K. Resistance to highly virulent mouse hepatitis virus acquired by mice after lowvirulence infec- tion: enhanced antiviral activity of macrophages. Infect Immun. 1980;29:42-9. 18. Li YM, Baviello G, Vlassara H, et al. Glycation products in aged thioglycollate medium enhance the elicitation of peritoneal mac- rophages. J Immunol Methods. 1997;201:183-8. 19. Cook AD, Braine EL,Hamilton JA. The phenotype of inflamma- tory macrophages is stimulus dependent: implications for the na- ture of the inflammatory response. J immunol. 2003;17:4816-23. 20. Misharin AV, Saber R, Perlman H. Eosinophil contamination of thioglycollate-elicited peritoneal mavrophage cultures skews the functional readouts of in vitro assays. J Leukoc Biol. 2012;92: 325-31. 21. Bogoevska V, Horst A, Klampe B, et al. CEACAM1, an adhesion molecule of human granulocytes, is fucosylated by fucosyltrans- ferase IX and interacts with DC-SIGN of dendritic cells via Lewis x residues. Glycobiology. 2006;16:197-209. 22. Nishihara S, Iwasaki H, Nakajima K, et a1. 3-Fucosyltransferase IX (Fut9) determines Lewis X expression in brain. Glycobiology. 2003;13:445-55. 23. Matsumura R, Hirakawa J, Sato K, et al. Novel antibodied reac- tive with sialyl Lewis X in both humans and mice define its critical role in leukocyte trafficking and contact hypersensitivity re- sponses. J Biol Chem. 2015;290:15313-26. 24. Gallily R, Warwick A, Bang FB. Effect of cortisone on genetic resistance to mouse hepatitis virus in vivo and in vitro. Proc Natl Acad Sci U S A. 1964;51:1158-64. 25. Ghosn EE, Cassado AA, Govoni GR, et al. Two physically, func- tionally, and developmentaly distinct peritoneal macrophage susets. Proc Natl Acad Sci U S A. 2010;107:2568-73. 26. Schleicher U, Hesse A, Bogdan C. Minute numbers of con- taminant CD8 ＋ T cells or CD11b ＋ CD11c ＋ NK cells are the source of IFN-g in IL-12/IL-18-stimulated mouse macrophage populations. Blood. 2005;105:1319-28. 27. Vodovotz Y, Bogdan C, Paik J, et al. Mechanisms of suppression of macrophage nitric oxide release by transform- ing growth fac- tor b. J Exp Med. 1993;178:605-13. 28. Schindler H, Lutz MB, Rollinghoff M, et al. The production of IFN-g by IL-12/IL-18-activated macrophages requires STAT4 signaling and is inhibited by IL-4. J Immunol. 2001;166:3075-82. 29. Nakagaki K, Nakagaki K, Taguchi F. Receptor-independent spread of a highly neurotropic murine coronavirus JHMV strain from initially infected microglial cells in mixed neural cultures. J Virol. 2005;79:6102-10. 30. Ribechini E, Greifenberg V, Sandwick S, et al. Subsets expansion and activation of myeloid-derives suppressor cells. Med Microbiol Immunol. 2010;199:273-81. 31. Dijkman R, Jebbink MF, Deijs M, et al. Replication-dependent downregulation of cellular angiotensin-converting enzyme 2 pro- tein expression by human coronavirus NL63. J Gen Virol. 2012; 93:1924-9. 32. Boscher C, Dennis JW, Nabi IR. Glycosylation, galectins and cel- lular signaling. Curr Opin Cell Biol. 2011;23:383-92. 33. Rabinovich GA, van Kooyk Y, Cobb BA. Glycobiology of im- mune responses. Ann N Y Acad Sci. 2012;1253:1-15. 34. van Kooyk Y, Rabinovich GA. Protein-glycan interactions in the control of innate and adaptive immune responses. Nat Immunol. 2008;9:593-601. 35. Gringhuis SI, den Dunnen J, Litjens M, et al. Carbonhydrate- specific signaling through the DC-SIGN signalosome tailors im- munity to Mycobacterium tuberculosis, HIV and Helicobacter pylori. Nat Immunol. 2009;10:1081-8. 36. Samsen A, Bogoevska V, Klampe B, et al. DC-SIGN and SRCL bind glycans of carcinoembryonic antigen (CEA) and CEA-relat- ed cell adhesion molecule 1 (CEACAM1): recombinant human glycan-binding receptors as analytical tools. Eur J Cell Biol. 2010; 89:87-94. 37. Johnson JL, Jones MB, Ryan SO, et al. The regulatory power of glycans and their binding partners in immunity. Trends Immunol. 2013;34:290-8. 38. Gracia-Vallejo JJ, Ilarregui JM, Kalay H, et al. CNS myelin in- duces regulatory functions of DC-SIGN-expressing, antigen- presenting cells via cognate interaction with MOG. J Exp Med. 2014;211:1465-83. 39. Kawauchi Y, Takagi H, Hanafusa K, et al. SIGNR1-mediated phagocytosis, but not SIGNR1-mediated endocytosis or cell adhe- sion, suppresses LPS-induced secretion of IL-6 from murine mac- rophages. Cytokine. 2015;71:45-53. 40. Kashiwazaki H, Taguchi F, Ikehara Y, et al. Characterization of splenic cells during the early phase of infection with neuropatho- genic mouse hepatitis virus. Jpn J Infect Dis. 2011;64:256-9. 41. Nomura R, Kashiwazaki H, Kakizaki M, et al. Receptor-indepen- dent infection by mutant viruses newly isolated from the neu- ropathogenic mouse hepatitis virus srr7 detected through a com- bination of spinoculation and ultraviolet radiation. Jpn J Infect Dis. 2011;64:499-505. 42. Lee SJ, Hori Y, Groves JT, et al. Correlation of a dynamic model for immunological synapse formation with effector function: two pathways to synapse formation. Trends Immunol. 2002;23:492-9. 43. Pittet MJ, Nahrendorf M, Swirski FK. The journey from stem cell to macrophage. Ann NY Acad Sci. 2014;1319:1-18.