The Role of CD4 + T Cells in Biphasic Hind Limb Paralysis Induced by the D Variant of Encephalomyocarditis Virus (EMC-D) in DBA/2 Mice

Size: px

Start display at page:

Download "The Role of CD4 + T Cells in Biphasic Hind Limb Paralysis Induced by the D Variant of Encephalomyocarditis Virus (EMC-D) in DBA/2 Mice"

Erik Rogers
5 years ago
Views:

1 Note Exp. Anim. 53(1), 31 35, 2004 The Role of CD4 + T Cells in Biphasic Hind Limb Paralysis Induced by the D Variant of Encephalomyocarditis Virus (EMC-D) in DBA/2 Mice Makio TAKEDA 1), Ryoichi OHTSUKA 1, 2), Yumi NAKAYAMA 2), and Kunio DOI 2) 1) Institute of Environmental Toxicology, Uchimoriya 4321, Mitsukaido, Ibaraki , and 2) Department of Veterinary Pathology, Faculty of Agriculture, The University of Tokyo, Yayoi, Bunkyo-ku, Tokyo , Japan Abstract: DBA/2 CrSlc mice infected with the D variant of encephalomyocarditis virus (EMC-D) (10 PFU/head) developed biphasic hind limb paralysis due to spinal cord lesion. The early phase lesion was characterized by demyelination with infiltration of macrophages in the funiculus lateraris and the late phase lesion by degeneration of motor neurons with infiltration of CD4 + T cells in the cornu ventrale (Takeda et al., Int. J. Exp. Pathol., 1993, 1995). In the present study, treatment with anti-mac1 monoclonal antibody (MAb) or anti- CD4 MAb prior to virus infection (-3 to -1 days) reduced the early phase lesion and the incidence of the first paralysis. Signals of viral RNAs were observed only in a few oligodendrocytes in the funiculus lateraris. Treatment with anti-cd4 MAb from 31 to 33 days post infection when mice showed recovery from the first paralysis reduced the late phase lesion and prevented the second paralysis. Signals of viral RNAs were still detected in a few degenerated neurons in the cornu ventrale. These results indicate that while macrophages and CD4 + T cells participate in the early phase lesion and paralysis and only CD4 + T cells in the late phase lesion and paralysis. Key words: CD4 + T cells, EMC-D, mouse spinal cord lesion We first reported that DBA/2 CrSlc mice inoculated with a low dose of the D variant of encephalomyocarditis virus (EMC-D) (10 PFU/head) developed biphasic hind limb paralysis [14]. Briefly, about 60% of the infected mice developed hind limb paralysis by 12 days post infection (DPI), twothirds of them showed recovery by 33 DPI, and 30% of the mice which had shown recovery developed paralysis again by 56 DPI. The character of this disease is biphasic spinal cord lesions. The degree of lesion is most prominent in the lumbar spinal cord. The lesion in the early phase is characterized by demyelination associated with infiltration of macrophages in the funiculus lateraris and that in the late phase by degeneration of motor neurons associated with infiltration of CD4 + T cells in the cornu ventrale. The virus titer of the spinal cord peaked at 7 DPI, decreased thereafter, and could no longer be detected even in paralysed mice at 28 DPI. We clarified using an in situ (Received 14 April 2003 / Accepted 6 August 2003) Address corresponding: K. Doi, Department of Veterinary Pathology, Graduate School of Agricultural and Life Sciences, The University of Tokyo, Yayoi, Bunkyo-ku, Tokyo , Japan

2 32 M. TAKEDA, ET AL. hybridization method that viral RNAs were observed in oligodendrocytes around demyelinated lesions with infiltration of macrophages in the early phase while they were observed in degenerated motor neurons with infiltration of CD4 + T cells in the late phase [15]. The aim of this study was to clarify the role of macrophages and especially CD4 + T cells in the development of spinal cord lesions and hind limb paralysis in DBA/2 mice infected with EMC-D using anti-mac1 and anti-cd4 monoclonal antibodies (MAb). The protocol of this study was approved by the Animal Care and Use Committee of Graduate School of Agricultural and Life Sciences, The University of Tokyo. One hundred and ten 8-week-old DBA/2 male mice were obtained from Charles River Japan Inc. (Kanagawa). The mice were housed in an animal room at a temperature of 23 ± 2 C with a relative humidity of 55 ± 5%, and fed MF pellets (Oriental Yeast Co., Ltd., Tokyo) and water ad libitum. EMC-D was kindly provided by Dr. J. W. Yoon [17], and anti-mac1 and anti-cd4 MAbs were prepared using hybridoma cell lines obtained from the American Type Culture Collection (Rockville, MO, USA). The mice were randomly divided into two groups, group A (35 mice) for the study of early lesion, and group B (75 mice) for the study of late lesion. Group A was further divided into 4 groups. Namely, 10 mice were treated intraperitoneally (i.p.) with 0.5 mg of anti- CD4 MAb three times (on days -3, -2 and -1) prior to virus infection (10 PFU/head) (A-1), and 10 mice were similarly treated with 0.5 mg of anti-mac1 MAb (A-2). Ten mice were inoculated i.p. with virus (10 PFU/head) alone as positive controls (A-3), and 5 mice were inoculated i.p. with 0.1 ml of PBS as negative controls (A-4). At 7 DPI, surviving mice were sacrificed by exsanguination under ether anesthesia. In group B, 70 mice were inoculated i.p. with EMC- D (10 PFU/head), and were checked for the sequence of clinical signs. Another 5 mice were inoculated i.p. with 0.1 ml of PBS as negative controls (B-4). Fifty surviving mice, showing recovery from clinical signs at 31 DPI were further divided into 3 groups. Fifteen mice were treated i.p. with 0.5 mg of anti-cd4 MAb three times (on 31, 32 and 33 DPI) (B-1), and 15 mice were treated with anti-mac1 MAb (B-2) in the same way. The remaining 20 mice were not treated with MAb and checked for further sequence of clinical signs as positive controls (B-3). At 51 DPI, all mice were sacrificed by exanguination under ether anesthesia. At autopsy, spleen cells of 5 randomly selected infected mice and 5 positive control mice were assayed with a fluorescence-activated cell sorter. More than 95% of MAb target cells were depleted in comparison with positive controls. During the experimental period, clinical signs and mortality were checked daily. In the spinal cord of EMC-D-infected mice, the lumbar spinal cord is most prominently affected as mentioned above [13]. Therefore, at autopsy, the lumbar spinal cords were fixed in 4% paraformaldehyde, and coronal paraffin sections of 4 µm were made. For histopathological observations, some of these sections were stained with hematoxylin and eosin (HE). In addition, 10-µm cryostat sections of small pieces of the fresh lumbar spinal cord were stained by the avidin-biotin-peroxidase complex (ABC) method using Vectastain Elite ABC kit (Vector Lab., Inc., USA). The primary antibodies used were as follows:monoclonal anti-mac1 (macrophage) rabbit IgG (Boehringer Mannheim Yamanouchi, Tokyo, Japan), monoclonal anti-l3t4 (CD4 + T cell) rabbit IgG (Biosys, Compiegne, France) and monoclonal anti-lyt2 (CD8 + T cell) rabbit IgG (Biosys). To detect viral RNAs, in situ hybridization was performed on paraffin sections of the lumbar spinal cord as described in a previous paper [14]. The crna probe for genome of EMC-D using in this experiment was 800 bases long and complementary to the structural proteins VP3-VP1, and was labeled by digoxigenin using a DIG-RNA labeling kit (Roche Diagnostics, Tokyo). The specificity of this probe was confirmed by northern blot analysis and sequencing [15]. As shown in Table 1, two mice died and the remaining 5 out of 8 mice (62.5%) developed hind limb paralysis by 7 DPI in group A-3 (0+EMC-D). On the other hand, in both groups A-1 (anti-cd4 + EMC-D) and A-2 (anti-mac1 + EMC-D), 1 of 10 mice developed hind limb paralysis and no mice died by 7 DPI. There were no clinical signs in group A-4. In group B, hind limb paralysis was first seen in some mice at 3 DPI and its incidence peaked at 12 DPI (60%). Twenty mice had died by 21 DPI, and the 50 surviving mice recovered from clinical signs and appeared normal at 30 DPI. Thereafter, as shown in Table 2, hind limb paralysis recurred in 6 of 20 mice of group

3 CD4 + T CELLS AND EMC-D INFECTION 33 Table 1. Effects of monoclonal antibodies on the incidence of EMC-D-induced firstphase paralysis at 7 DPI Table 2. Effects of monoclonal antibodies on the incidence of EMC-D-induced secondphase paralysis at 51 DPI Group Incidence of paralysed mice (%) A-1 (anti-cd4 + EMC-D) 1/10 (10) A-2 (anti-mac 1 + EMC-D) 1/10 (10) A-3 (EMC-D) 5/8 (62.5) A-4 (PBS) 0/4 (0) Group Incidence of paralysed mice (%) B-1 (EMC-D + anti-cd4) 0/15 (0) B-2 (EMC-D + anti-mac 1) 3/15 (20) B-3 (EMC-D) 4/15 (26.7) B-4 (PBS) 0/6 (0) B-3 (EMC-D +0) and 3 of 15 mice of group B-2 (EMC- D + anti-mac1) from 42 to 51 DPI. On the other hand, in group B-1 (EMC-D + anti-cd4), no mice developed the second hind limb paralysis. There were no clinical signs in group B-4 until 51 DPI. In group A, demyelination was seen in the funiculus lateralis in the spinal cords of mice showing the first hind limb paralysis. In group A-3, the lesions were severe and associated with infiltration of many anti- Mac1-positive macrophages (Fig. 1a) and a few anti-l3t4-positive CD4 + T cells. Signals of viral RNAs were observed in many oligodendrocytes in the lesion (Fig. 1b). On the other hand, in both group A-1 and group A-2, the lesions were minimal and not associated with mononuclear cell infiltration, and signals of viral RNAs were observed in only a few oligodendrocytes (Fig. 1c). There were no signals of viral RNAs in the spinal cords of mice of group A-4. In group B, in the spinal cord of 6 mice of group B-3 which exhibited the second hind limb paralysis, prominent infiltration of CD4 + T cells was found around degenerated motor neurons in the cornu ventrale at 51 DPI (Fig. 2a). In the funiculus lateralis, remyelination occurred in the previous demyelinated lesion as previously reported [14]. Lesions with similar characteristics and severity were also observed in the spinal cords of the 3 mice of group B-2 which exhibited the second hind limb paralysis. In the spinal cords of mice of group B-2 and group B-3 which showed the second hind limb paralysis, signals of viral RNAs were observed in a small number of degenerated motor neurons (Fig. 2b). On the other hand, in group B-1, a few degenerated neurons accompanying no mononuclear cell infiltration were observed, and a small number of signals of viral RNAs was observed in only a few motor neurons (Fig. 2c). There were neither spinal cord lesions nor signals of viral RNAs in group B-4. The mortality and the sequence of hind limb paraly- sis in EMC-D-infected mice observed in the present study were similar to those in our previous report [14]. In the early phase of the EMC-D-induced biphasic central nervous disease in DBA/2 mice, the treatment of anti-mac1 MAb or anti-cd4 MAb clearly reduced both demyelination in the funiculus lateralis and the incidence of hind limb paralysis. Macrophages and signals of viral RNAs in oligodendrocytes were also reduced. These results indicate that macrophages and CD4 + T cells play an important role in the development of demyelination in the funiculus lateralis and subsequent hind limb paralysis in addition to the direct effect of the virus on oligodendrocytes. In this regard, Sriram et al. [12] reported that treatment of anti-cd4 MAb prevented demyelination induced by EMC-M, and Topham et al. [16] reported that treatment of anti-cd4 MAb prevented encephalitis induced by EMC-M. On the other hand, Baek and Yoon [2, 3] and Hirasawa et al. [5 7] reported that macrophages, not T cells, played a crucial role in pancreatic β cell destruction in the early phase of EMC-D infection. The reason for this discrepancy is still obscure. Matsuzaki et al. [9] reported that EMC-D replicated first in the pancreas and then spread to the other organs. Furthermore, Craighead et al. [4] reported that phagocytosis of viral particles by macrophages led to picornavirus-induced immunity. Taking these reports into account, it seems reasonable to consider that, the mechanism/route of macrophage infiltration in the pancreas being different from that in the spinal cord, macrophage infiltration in the spinal cord might be concerned with CD4 + T cells. As reported in our previous papers [14, 15], in the present study, many macrophages and a few CD4 + T cells infiltrated the demyelinated lesions in the funiculus lateralis of mice infected with 10 PFU of EMC-D at 7 DPI, and mononuclear cell infiltration was reduced not only in mice treated with anti-mac1 MAb but also in mice treated with anti-cd4

34 M. TAKEDA, ET AL. Fig. 1. Funiculus lateraris of lumbar spinal cord at the early phase after EMC-D infection. (a) Many anti- Mac1-positive cells are seen (0+EMC-D group). Immunostaining, 120.

(c) Signals of viral RNAs are seen in a few oligodendrocytes and there are no infiltrating cells (anti-mac1+e MC-D group). In situ hybridization, 120. Fig. 2.

4 34 M. TAKEDA, ET AL. Fig. 1. Funiculus lateraris of lumbar spinal cord at the early phase after EMC-D infection. (a) Many anti- Mac1-positive cells are seen (0+EMC-D group). Immunostaining, 120. (b) Signals of viral RNA are seen in many oligodendrocytes (0+EMC-D group). In situ hybridization, 120. (c) Signals of viral RNAs are seen in a few oligodendrocytes and there are no infiltrating cells (anti-mac1+e MC-D group). In situ hybridization, 120. Fig. 2. Cornu ventrale of lumber spinal cord at the late phase after EMC-D infection. (a) Many anti-cd4 + T cells are seen around degenerated motor neurons (0+EMC-Dgroup). Immunostaining, 150. (b) Signals of viral RNAs are seen in a small number of degenerated mortor neurons (arrowheads) surrounded by many mononuclear cells (0 +EMC-D group). In situ hybridization, 150. (c) Signals of viral RNAs are seen in a few motor neurons (arrowhead) and there are no infiltrating cells. In situ hybridization, 150. MAb. These findings suggest the possibility that the CD4 + T cells might be memory T cells and rule the activity of macrophages. Only treatment with anti-cd4 MAb prevented the second hind limb paralysis, although a small number of signals of viral RNAs was still detected in a few degenerated neurons in the cornu ventrale. In addition, as described in our previous report [14], there were many CD4 + T cells around the degenerated neurons bearing viral RNAs in the cornu ventrale of the spinal cord of mice developing the second phase hind limb paralysis in the present study. Taken together, these suggest that CD4 + T cells may recognize some signals from the surface of degenerated neurons in which viral RNAs remain, resulting in destruction of these neu- rons. Huber [8] reported that 70 kda heat shock protein (hsp 70) was expressed on cardiocytes infected with EMC-M and coxsackie virus B3 (CVB3) in vitro, and cytolytic T lymphocytes which belong to the CD4 population were detected in the mice infected with CVB3. Furthermore, it was reported that CD4 + T cells reacted cytotoxically to myelin by releasing tumor necrosis factor alpha (TNF-α) in experimental autoimmune encephalomyelitis mice [10, 11]. Taking these reports into account, we hypothesize that CD4 + T cells might act as memory T cells in the early phase lesion and as cytotoxic T cells in the late phase lesion. As mentioned above, EMC-D-induced biphasic central nervous disease is thought to be a virus-induced autoimmune disease in which CD4 + T cells play an

5 CD4 + T CELLS AND EMC-D INFECTION 35 important role in the pathogenesis. Therefore, we consider that this model offers a good experimental tool for investigating viral immunity. Recently it was reported that Lewis rats, on recovery from monophasic clinical experimental allergic encephalitis, were induced to develop repeated paralytic relapses following intraperitoneal administration of IL-12. However the characteristics and mechanisms of the spinal cord lesion seem to be different from those of the present model. References 1. Ahmed, Z., Gveric, D., Pryce, G., Baker, D., Leonard, J.P., and Cuzner, M.L Am. J. Pathol. 158: Baek, H.S. and Yoon, J.W J. Virol. 64: Baek, H.S. and Yoon, J.W Diabetes 40: Craighead, J.E., Huber, S.A., and Sriam, S Lab. Invest. 63: Hirasawa, K., Jun, H.S., Han, H.S., Zhang, M.D., Hollenberg, M.D., and Yoon, J.W J. Virol. 73: Hirasawa, K., Jun H.S., Maeda, K., Kawaguchi, Y., Itagaki, S., Mikami, T., Baek, H.S., Doi, K., and Yoon, J.W J. Virol. 71: Hirasawa, K., Tsutsui, S., Takeda, M., Mizutani, M., Itagaki, S., and Doi, K J. Gen. Virol. 77: Huber, S.A Lab. Invest. 67: Matsuzaki, H., Doi, K., Doi, C., Onodera, T., and Mitsuoka, T Exp. Anim. 38: Powell, M.B., Miychell, D., and Lederman, J Int. Immunol. 2: Ruddle, N.H., Bergman, C.M., and McGrath, K.M J. Exp. Med. 172: Sriram, S., Topham, D.J., Huang, S.K., and Rodriguez, M J. Virol. 63: Takeda, M., Hirasawa, K., and Doi, K J. Vet. Med. Sci. 53: Takeda, M., Itagaki, S., and Doi, K Int. J. Exp. Path. 74: Takeda, M., Miura, R., Shiota, K., Hirasawa, K., Lee, M. J., Itagaki, S., and Doi, K Int. J. Exp. Path. 76: Topham, D.J., Adesina, A., Shenoy, M., Craighead, J.E., and Sriram, S J. Virol. 65: Yoon, J.W., McClintock, P.R., Onodera, T., and Notokins, A.L J. Exp. Med. 152:

The pathogenesis of nervous distemper

Veterinary Sciences Tomorrow - 2004 The pathogenesis of nervous distemper Marc Vandevelde Canine distemper is a highly contagious viral disease of dogs and of all animals in the Canidae, Mustellidae and