- Email: [email protected]
Interaction between apoptotic cells and reactive brain cells in the central nervous system of rats with autoimmune encephalomyelitis
Journal of Neuroimmunology 82 Ž1998. 168–174
Interaction between apoptotic cells and reactive brain cells in the central nervous system of rats with autoimmune encephalomyelitis Toshihiko Kohji, Naoyuki Tanuma, Yukihiko Aikawa, Yoko Kawazoe, Yoko Suzuki, Kuniko Kohyama, Yoh Matsumoto ) Department of Neuropathology, Tokyo Metropolitan Institute for Neuroscience, Musashidai 2-6, Fuchu, Tokyo 183, Japan Received 25 July 1997; received in revised form 24 September 1997; accepted 26 September 1997
Abstract To elucidate the role of brain cells in the immune regulation in the central nervous system ŽCNS., acute and chronic relapsing experimental autoimmune encephalomyelitis ŽEAE. was induced in Lewis rats and the location of apoptotic inflammatory cells and their interaction with astrocytes and microglia was investigated at various stages of the disease. Apoptotic cells detected by terminal deoxynucleotidyl transferase-mediated dUTP nick end-labeling ŽTUNEL. were few in number at day 10–12 post-immunization ŽPI., increased and peaked at day 13 PI. Then, these cells decreased gradually by day 21 PI. The most characteristic finding was that apoptotic cells were mainly distributed in the CNS parenchyma with only a few cells present in perivascular cuffs. Double staining by the TUNEL method and immunocytochemistry for astrocytes and microglia revealed that astrocytes were more closely associated with apoptotic cells than microglia. Apoptotic cell death may be one mechanism by which T cells are eliminated from the CNS. Furthermore, the present study suggests that astrocytes, rather than microglia, induce programmed cell death of infiltrating inflammatory cells. q 1998 Elsevier Science B.V. Keywords: Apoptotic cell death; Astrocytes; Experimental autoimmune encephalomyelitis ŽEAE.; T cell; Lewis rat
1. Introduction Experimental autoimmune encephalomyelitis ŽEAE. is a T cell-mediated autoimmune disease that is inducible in susceptible strains of mice, rats and other species by immunization with brain-specific antigens such as myelin basic protein ŽMBP. ŽLassmann, 1983.. Rats usually develop acute and monophasic EAE after immunization with MBP and do not show relapse of the clinical signs. We previously demonstrated that in acute EAE, T cells that enter the CNS lose their proliferating ability soon after the infiltration ŽOhmori et al., 1992.. Later, several groups showed that infiltrating T cells are eliminated from the CNS by apoptotic cell death at the peak and recovery stages of EAE ŽPender et al., 1992; Schmied et al., 1993..
) Corresponding author. Tel.: q81 423 253881r4719; fax: q81 423 218678; e-mail: [email protected]
0165-5728r98r$19.00 q 1998 Elsevier Science B.V. All rights reserved. PII S 0 1 6 5 - 5 7 2 8 Ž 9 7 . 0 0 1 9 8 - 7
In parallel with these in vivo findings, in vitro studies suggest that some brain cells such as astrocytes and microglia inhibit the proliferation of MBP-reactive T cells and induce apoptosis ŽMatsumoto et al., 1992a, 1993; Ford et al., 1996.. In general, microglia are reported to upregulate, whereas astrocytes downregulate, T cell proliferation. However, there is disagreement with regard to the antigen presenting ability of astrocytes and microglia. In the present study, we attempted to elucidate whether brain cells are involved in the process of apoptosis of infiltrating T cells and, if this is the case, to determine the type of brain cells involved. For this purpose, acute and chronic relapsing EAE was induced in Lewis rats and the location and number of apoptotic cells as well as the interaction between apoptotic cells and brain cells were investigated by terminal deoxynucleotidyl transferasemediated dUTP nick end-labeling ŽTUNEL. and immunohistochemistry. Consequently, we found that apoptotic cells were mainly located in the CNS parenchyma and that astrocytes were closely associated with apoptotic cells.
T. Kohji et al.r Journal of Neuroimmunology 82 (1998) 168–174
2. Materials and methods 2.1. Animals Lewis rats were purchased from Seiwa ŽFukuoka. and bred in our animal facility. Rats at age 8–12 weeks were used throughout the experiments. 2.2. EAE induction and tissue sampling Acute EAE was induced as described previously ŽOhmori et al., 1992.. Briefly, each rat was injected in the hind footpads bilaterally with an emulsion containing 100 m g of MBP in complete Freund’s adjuvant ŽCFA; Mycobacterium tuberculosis H37Ra, 5 mgrml.. Chronic relapsing EAE ŽCR-EAE. was induced by immunization of rats with 200 m l of guinea pig spinal cord homogenate in CFA. One gram of spinal cord tissue was homogenized in 1 ml of PBS and the suspension was emulsified with an equal volume of CFA. Starting from the day of immunization, rats were given intraperitoneal injections of cyclosporin A ŽCsA. ŽSandoz, Tokyo. at a dose of 4 mgrkg three times a week until day 21 post-immunization ŽPI.. Immunized rats were observed daily for clinical signs of EAE. The clinical stage of EAE was divided into five Žgrade 1, floppy tail; grade 2, mild paraparesis; grade 3, severe paraparesis; grade 4, tetraparesis; grade 5, death.. At various time points, rats were killed under ether anesthesia and several segments of the lumbar spinal cord were processed for paraffin-embedding. 2.3. Terminal deoxynucleotidyl transferase-mediated dUTP nick end-labeling (TUNEL) DNA fragmentation was detected by in situ nick endlabeling as described previously ŽGavrieli et al., 1992. with a few modifications. Briefly, paraffin-embedded sections were dewaxed, rehydrated through a graded alcohol series, and washed in PBS. Next, sections were treated with 0.005% protease ŽSigma, St. Louis, MO. for 20 min at 378C and subsequently incubated in a terminal deoxynucleotidyl transferase ŽTdT. buffer solution Ž140 mM sodium cacodylate, 1 mM cobalt chloride, 30 mM Tris–HCl, pH 7.2. containing 0.15 Urm l TdT ŽGibco BRL, Tokyo. and 0.004 nmolrm l digoxigenin-dUTP ŽBoehringer-Mannheim, Tokyo. for 60 min at 378C, and then in the TB buffer Ž300 mM sodium chloride, 30 mM sodium citrate. for 15 min. They were allowed to react with alkaline phosphatase ŽAP.-labeled anti-digoxigenin antibody ŽBoehringer-Mannheim. for 60 min. Positive cells were finally visualized by incubating sections with the SIGMA FAST e BCIPrNBT Buffered Substrate Tablet solution ŽSigma. containing levamisole Ž0.24 mgrml. which blocks endogenous AP.
169
2.4. Immunohistochemical staining procedures Paraffin-embedded sections were deparaffinized and rehydrated, and endogenous peroxidase was blocked by incubation in 0.3% H 2 O 2 in methanol. After incubation with normal sheep or goat serum, sections were allowed to react with the first antibody for 60 min. For the detection of microglia, biotinylated tomato lectin was used ŽAcarin et al., 1994.. Then, sections were incubated with horseradish peroxidase ŽHRP.-labeled VECTASTAIN w Elite ABC Kit ŽVector, Burlingame, CA.. HRP binding sites were detected with the SIGMA FAST e DAB Peroxidase Substrate Tablet Set solution ŽSigma.. For glial fibrillary acidic protein ŽGFAP. immunostaining, the peroxidase anti-peroxidase method was applied using Histogen PAP rabbit Universal Kit ŽBioGenex, San Ramon, ´ CA.. 2.5. Double staining for apoptotic cells and brain cells Double staining for TUNEL-positive cells and brain cells such as microglia and astrocytes was performed using paraffin-embedded sections. In the first step, apoptotic cells were detected by the TUNEL method and AP was developed in the BCIPrNBT solution as a blue color. After a thorough washing, slides were stained for microglia or astrocytes using HRP as an enzyme. HRP was developed in the DAB solution in brown. 3. Results 3.1. Clinical obserÕation The clinical course of acute and chronic relapsing EAE is shown in Fig. 1. Clinical signs of acute EAE of rats immunized with MBP developed on day 11, peaked on day
Fig. 1. Clinical course of acute ŽA. and chronic relapsing ŽB. EAE.
170
T. Kohji et al.r Journal of Neuroimmunology 82 (1998) 168–174
Fig. 2. The distribution of TUNEL-positive cells in the spinal cord during acute EAE. A few TUNEL-positive cells are located in the subarachnoid space and in the subpial region at the early stage of EAE ŽA.. TUNEL-positive cells increased in number and reached maximal levels at the peak stage ŽB and C.. At this stage, most TUNEL-positive cells were distributed in the parenchyma with very few in the perivascular cuff. There is no TUNEL-positive cells in the perivascular cuff in this field Žarrowhead in C.. TUNEL-positive cells decreased in number at the recovery stage ŽD.. Arrows in Panels A–D indicate TUNEL-positive cells. A, day 11; B and C, day 14; D, day 21. Original magnifications: =100.
13–15 and subsided by day 18 PI ŽFig. 1A.. In the present study, all the rats with acute EAE were killed for histological examination by day 21 PI. We previously observed that there was no relapse of the disease after this observation period Žour unpublished data.. In this study, the early stage of acute EAE referred to day 10 or 11 PI and the peak stage to days 13–15 PI. Rats immunized with spinal cord homogenate and treated with CsA developed clinical signs on day 11 with the peak on days 14–16 as well as acute EAE. However, all the rats that recovered from EAE showed relapse of the clinical signs between days 24 and 32 with the peak on days 25–27 ŽFig. 1B.. Based on these findings, acute EAE was examined histologically at various time points before, during and after the manifestation of clinical signs. CR-EAE was examined at the first attack, remission phase, second attack and recovery phase. 3.2. Histological and quantitatiÕe analysis of apoptotic cells in the CNS during EAE We first investigated the location of apoptotic cells in the CNS of rats with acute EAE at various time points. At
the early stage of EAE, only a small number of TUNELpositive cells were present in the subarachnoid space and the subpial region ŽFig. 2A.. At the peak stage of acute EAE, TUNEL-positive cells were distributed evenly in the parenchyma, but were rare in perivascular cuffs ŽFig. 2B
Fig. 3. The number of TUNEL-positive cells at various stages of acute EAE. The number of TUNEL-positive cellsrspinal cord section were determined using at least three rats Žtotal 9–12 segments of the spinal cord. at each time point. The values indicate means"SD.
T. Kohji et al.r Journal of Neuroimmunology 82 (1998) 168–174
Fig. 4. The number of TUNEL-positive cells at various stages of CR-EAE. The number of TUNEL-positive cellsrspinal cord section were determined using at least three rats Žtotal 9–12 segments of the spinal cord. at each time point. The values indicate means"SD. In this study, tissue sampling was performed at the 1st attack Žday 14–17 PI., remission Žday 21–23 PI., 2nd attack Žday 25–27 PI. and full recovery stage Žday 34–36 PI. of chronic relapsing EAE.
and C.. A similar distribution pattern was noticed at the first and second attack of CR-EAE Ždata not shown.. At the recovery stage of acute EAE, a few TUNEL-positive cells were present in the parenchyma ŽFig. 2D.. These findings suggest that inflammatory cells are not pro-
171
grammed to undergo apoptosis before entering the CNS, but that some factors residing in the CNS parenchyma are involved in the process. If apoptosis of infiltrating inflammatory cells is undertaken solely by the mechanism programmed in infiltrating cells, apoptotic cells would have been recognized in both the parenchyma and perivascular cuffs. We next counted TUNEL-positive cells at various stages of acute EAE ŽFig. 3.. TUNEL-positive cells were rare at the early stage, but rapidly increased and reached maximal levels on days 13 and 14 PI when the animals showed severe paraparesis. The number of TUNEL-positive cells gradually declined after the peak stage of acute EAE. Similar findings were obtained in CR-EAE ŽFig. 4.. In CR-EAE, the number of apoptotic cells in the CNS was maximal at the first attack and decreased to almost zero in the remission phase before increasing again at the second attack and declining at the recovery stage. 3.3. Relationship between apoptotic cells and brain cells Immunohistochemical study showed that the majority of apoptotic cells were located in the CNS parenchyma and
Fig. 5. Double staining for GFAP-positive Žbrown. and TUNEL-positive Ždark blue. cells ŽA and B. and for tomato lectin-positive microglia Žbrown. and TUNEL-positive cells ŽC and D.. In Panels A and B, cell processes of GFAP-positive astrocytes attach to TUNEL-positive nuclei Žarrows.. Double staining with TUNEL and tomato lectin histochemistry shows a different staining pattern. In this case, some TUNEL-positive nuclei are surrounded by lectin staining Žarrows in C and D., indicating that they are TUNEL-positive microglia. Original magnifications: =200.
172
T. Kohji et al.r Journal of Neuroimmunology 82 (1998) 168–174
Fig. 6. The number of TUNEL-positive cells contacting GFAP-positive astrocytes Žclosed square. and tomato lectin-positive microgliarmacrophages Žclosed circle.. The numbers are significantly different on days 14 Ž ) ) p- 0.01. and 15 Ž ) p- 0.05. by Student-t test. The number of contact cells was determined using at least 4 different segments of the spinal cord. The values indicate means"SD. TUNEL-positive microglia are not included in this figure.
not in the perivascular cuff ŽFig. 2.. This suggested that apoptosis occurs in association with the contact of inflammatory cells with brain cells. To examine the interaction between apoptotic cells and brain cells, we did double staining for apoptotic cells and reactive brain cells such as astrocytes ŽGFAP-positive cells. and microgliarmacrophages Žtomato lectin-positive cells.. As shown in Fig. 5A and B, cell processes of GFAP-positive astrocytes attached to TUNEL-positive nuclei. In double staining for TUNEL-positive cells and microgliarmacrophages, a similar finding was observed Ždata not shown.. In addition, some microgliarmacrophages were undergoing apoptosis ŽFig. 5C and D. as reported previously ŽNguyen et al., 1994, 1997; Tabi et al., 1995.. We then counted TUNELpositive cells contacting astrocytes or microgliarmacrophages at various stages of acute EAE ŽFig. 6.. The number of TUNEL-positive cells contacting astrocytes was low at the early stage but increased and reached maximal levels at the peak stage. Then, these cells decreased gradually at the recovery stage. TUNEL-positive cells contacting microgliarmacrophages also increased in number at the peak stage of acute EAE. However, the number of TUNEL-positive cells contacting microgliarmacrophages was significantly less than that of TUNEL-positive cells contacting astrocytes ŽFig. 6..
4. Discussion EAE is a T cell-mediated autoimmune disease characterized by the presence of inflammatory cells mainly comprising T cells and macrophages in the CNS. Although EAE is inducible in several species including mice, rats, and guinea pig by immunization with brain-specific antigen, its clinical course is different from one species to another. It is generally believed that differences in the clinical course depend on both the animal species and the encephalitogenic antigen used. Rats immunized with guinea
pig MBP show acute monophasic EAE. Infiltration of inflammatory cells into the CNS begins around day 10 PI and reaches a maximum on day 12. Between days 12 and 14, inflammation plateaus and then most lesions subside by day 21 ŽMatsumoto et al., 1993.. Although these processes are well documented, the kinetics of T cells is poorly understood. Our previous study using immunohistochemistry showed that the majority of T cells involved in EAE undergo DNA synthesis outside the CNS and then infiltrate the CNS parenchyma and that infiltrating T cells do not proliferate vigorously in the CNS ŽOhmori et al., 1992.. Later, several groups demonstrated that this phenomenon is attributable to apoptotic cell death of T cells ŽPender et al., 1991, 1992; Schmied et al., 1993; Tabi et al., 1994, 1995.. In the present study, we quantitated TUNEL-positive cells in the spinal cord throughout the course of acute EAE and found that the number of TUNEL-positive cells reached a maximal level on days 13 and 14 PI and declined gradually ŽFig. 3.. This finding is consistent with that reported by McCombe et al. Ž1996. who examined apoptotic cells in the isolated inflammatory cell population during acute EAE by flow cytometry. In our chronic relapsing model, the number of TUNEL-positive cells paralleled well with the clinical course of the disease ŽFig. 4.. Although Schmied et al. Ž1993. reported that more TUNEL-positive cells were found at the later stage, this may be due to the difference in the method of the EAE induction. They immunized SD rats with guinea pig spinal cord homogenate in CFA, which may induce chronic EAE Žthe precise clinical course was not shown in their report.. In addition to this, the difference in the detection procedures may contribute to the difference of the outcome. We used terminal deoxynucleotidyl transferase to detect DNA fragmentation in apoptosis ŽTUNEL., while they used DNA polymerase Žnick translation.. It is reported that cells undergoing apoptosis are preferentially labeled by TUNEL, whereas necrotic cells are identified by nick translation ŽGold et al., 1994.. Thus, the TUNEL method which we employed here may detect apoptotic processes at an earlier stage. One of the characteristic findings of the present study was that the majority of apoptotic cells were located in the CNS parenchyma, and not in the perivascular cuff ŽFig. 2 C . . A lth o u g h so m e to m ato lectin -p o sitiv e microgliarmacrophages underwent apoptosis, the number of such cells was far low compared to tomato lectin-negative cells Ždata not shown.. Therefore, it was judged that the majority of TUNEL-positive cells were T cells. These findings strongly suggest that brain cells are involved in the apoptotic processes of infiltrating T cells. If this is not the case, that is, if apoptotic signals are programmed in T cells before entry into the CNS, then apoptotic cells would have been found equally in both the CNS parenchyma and perivascular region. Based on these findings, we examined the relationship between apoptotic cells and brain cells by double staining. It was revealed that apoptotic cells were
T. Kohji et al.r Journal of Neuroimmunology 82 (1998) 168–174
more closely associated with astrocytes than with microglia with the most marked difference noted at the peak stage of EAE ŽFig. 6.. In our previous study ŽMatsumoto et al., 1992b., we observed that microglia react with infiltrating T cells and proliferate at the early stage whereas astrocytes react to inflammatory lesions at the late peak to recovery stages. Therefore, it is conceivable that astrocytes play some role in the induction of T cell apoptosis. Recent studies on T cell activation via T cell receptors with or without costimulatory signals revealed that T cells stimulated with an antigen presented by antigen-presenting cells ŽAPC. with costimulatory signals through CD28rB7 molecules induce T cell proliferation, whereas the same stimulation without costimulatory signals results in T cell anergy or antigen-induced cell death ŽJaneway and Bottomly, 1994.. Regarding brain cells, it is reported that microglia, but not astrocytes, express CD28rB1 molecules ŽWilliams et al., 1994; de Simone et al., 1995; Satoh et al., 1995; Iglesias et al., 1997.. In accord with this finding, in vitro studies showed that microglia expressing MHC class II antigens have an antigen-presenting ability ŽFrei et al., 1986; Matsumoto et al., 1992a; Cash et al., 1993.. However, Ford et al. Ž1996. reported that ramified microglia isolated from normal adult brain do not induce T cell proliferation. On the other hand, there is a consensus that astrocytes are poor antigen presenting cells and suppress T cell proliferation even after the induction of MHC class II molecules ŽMatsumoto et al., 1993; Weber et al., 1994.. Recently, both microglia and astrocytes have been shown to have ability to induce apoptosis of activated T cells ŽFord et al., 1996; Gold et al., 1996.. Taking these findings into consideration, the results obtained in the present study can be interpreted as follows. At the early stage of EAE, microglia proliferate in the vicinity of infiltrating T cells. However, it is unlikely that microglia at this stage induce T cell apoptosis because only a small number of apoptotic cells and of apoptotic cells contacting microglia are found in the CNS. Astrocytes do not respond to inflammatory lesions at the early stage ŽMatsumoto et al., 1992b.. At the peak stage, apoptotic cells and those contacting astrocytes reach maximal levels suggesting that astrocytes mainly prime infiltrating T cells for apoptotic cell death. This assumption does not exclude the possibility that microglia also induce apoptosis at the peak stage of acute EAE. In any case, brain cells may play a major role in apoptosis of infiltrating T cells, and thus the majority of apoptotic cells were found in the CNS parenchyma.
Acknowledgements This study was supported in part by grants from the Ministry of Education, Japan, Toyama Chemical Co. and Naito Foundation.
173
References Acarin, L., Vela, J.M., Gonzaez, ´ B., Castellano, B., 1994. Demonstration of Poly-N-acetyl lactosamine residues in ameboid and ramified microglial cells in rat brain by tomato lectin binding. J. Histochem. Cytochem. 42, 1003–1041. Cash, E., Zhang, Y., Rott, O., 1993. Microglia present myelin antigen to T cells after phagocytosis of oligodendrocytes. Cell. Immunol. 147, 129–138. de Simone, R., Giampaolo, A., Giometto, B., Gallo, P., Levi, G., Peschle, C., Aloisi, F., 1995. The costimulatory molecule B7 is expressed on human microglia in culture and in multiple sclerosis acute lesions. J. Neuropathol. Exp. Neurol. 54, 175–187. Ford, A.L., Foulcher, D., Lemckert, F.A., Sedgwick, J.D., 1996. Microglia induce CD4 T lymphocyte final effector function and death. J. Exp. Med. 184, 1737–1745. Frei, K., Bodmer, S., Schwerdel, C., Fontana, A., 1986. Astrocyte-derived interleukin 3 as a growth factor for microglia cells and peritoneal macrophages. J. Immunol. 137, 3521–3527. Gavrieli, Y., Sherman, Y., Ben-Sasson, S.A., 1992. Identification of programmed cell death in situ via specific labeling of nuclear DNA fragmentation. J. Cell Biol. 119, 493–501. Gold, R., Schmied, M., Giegerich, G., Breitschopf, H., Hartung, H.P., Toyka, K.V., Lassmann, H., 1994. Differentiation between cellular apoptosis and necrosis by the combined use of in situ tailing and nick translation techniques. Lab. Invest. 71, 219–225. Gold, R., Schmied, M., Tontsch, U., Hartung, H.P., Wekerle, H., Toyka, K.V., Lassmann, H., 1996. Antigen presentation by astrocytes primes rat T lymphocytes for apoptotic cell death. Brain 119, 651–659. Iglesias, B.M., Cerase, J., Ceracchini, C., Levi, G., Aloisi, F., 1997. Analysis of B7-1 and B7-2 costimulatory ligands in cultured mouse microglia: Upregulation by interferon-g and lipopolysaccharide and downregulation by interleukin-10, prostaglandin E 2 and cyclic AMPelevating agents. J. Neuroimmunol. 72, 83–93. Janeway, C.A. Jr., Bottomly, K., 1994. Signals and signs for lymphocyte responses. Cell 76, 275–285. Lassmann, H., 1983. Comparative Neuropathology of Chronic Experimental Allergic Encephalomyelitis and Multiple Sclerosis. SpringerVerlag, Berlin, pp. 9–94. Matsumoto, Y., Ohmori, K., Fujiwara, M., 1992a. Immune regulation by brain cells in the central nervous system: Microglia but not astrocytes present myelin basic protein to encephalitogenic T cells under in vivo-mimicking conditions. Immunology 76, 209–216. Matsumoto, Y., Ohmori, K., Fujiwara, M., 1992b. Microglial and astroglial reactions to inflammatory lesions of experimental autoimmune encephalomyelitis in the rat central nervous system. J. Neuroimmunol. 37, 23–33. Matsumoto, Y., Hanawa, H., Tsuchida, M., Abo, T., 1993. In situ inactivation of infiltrating T cells in the central nervous system with autoimmune encephalomyelitis. The role of astrocytes. Immunology 79, 381–390. McCombe, P.A., Nickson, I., Tabi, Z., Pender, M.P., 1996. Apoptosis of Vb 8.2q T lymphocytes in the spinal cord during recovery from experimental autoimmune encephalomyelitis induced in Lewis rats by inoculation with myelin basic protein. J. Neurol. Sci. 139, 1–6. Nguyen, K.B., McCombe, P.A., Pender, M.P., 1994. Macrophage apoptosis in the central nervous system in experimental autoimmune encephalomyelitis. J. Autoimmun. 7, 145–152. Nguyen, K.B., McCombe, P.A., Pender, M.P., 1997. Increased apoptosis of T lymphocytes and macrophages in the central and peripheral nervous systems of Lewis rats with experimental autoimmune encephalomyelitis treated with dexamethasone. J. Neuropathol. Exp. Neurol. 56, 58–69. Ohmori, K., Hong, Y., Fujiwara, M., Matsumoto, Y., 1992. In situ demonstration of proliferating cells in the rat central nervous system during experimental autoimmune encephalomyelitis. Evidence sug-
174
T. Kohji et al.r Journal of Neuroimmunology 82 (1998) 168–174
gesting that most infiltrating T cells do not proliferate in the target organ. Lab. Invest. 66, 54–62. Pender, M.P., Nguyen, K.B., McCombe, P.A., Kerr, J.F.R., 1991. Apoptosis in the nervous system in experimental allergic encephalomyelitis. J. Neurol. Sci. 104, 81–87. Pender, M.P., McCombe, A., Yoog, G., Nguyen, K.B., 1992. Apoptosis of ab T lymphocytes in the nervous system in experimental autoimmune encephalomyelitis: Its possible implication for recovery and acquired tolerance. J. Autoimmun. 5, 401–410. Satoh, J., Lee, Y.B., Kim, S.U., 1995. T-cell costimulatory molecules B7-1 ŽCD80. and B7-2 ŽCD86. are expressed in human microglia but not in astrocytes in culture. Brain Res. 704, 92–96. Schmied, M., Breitschopf, H., Gold, R., Zischler, H., Rothe, G., Wekerle, H., Lassmann, H., 1993. Apoptosis of T lymphocytes in experimental autoimmune encephalomyelitis. Evidence for programmed cell death as a mechanism to control inflammation in the brain. Am. J. Pathol. 143, 446–452.
Tabi, Z., McCombe, P.A., Pender, M.P., 1994. Apoptotic elimination of Vb 8.2q cells from the central nervous system during recovery from experimental autoimmune encephalomyelitis induced by the passive transfer of Vb 8.2q encephalitogenic T cells. Eur. J. Immunol. 24, 2609–2617. Tabi, Z., McCombe, P.A., Pender, M.P., 1995. Antigen-specific downregulation of myelin basic protein-reactive T cells during spontaneous recovery from experimental autoimmune encephalomyelitis: Further evidence of apoptotic deletion of autoreactive T cells in the central nervous system. Int. Immunol. 7, 967–973. Weber, F., Meinl, E., Aloisi, F., Nevinny-Stickel, C., Albert, E., Wekerle, H., Hohlfeld, R., 1994. Human astrocytes are only partially competent antigen presenting cells. Possible implication for lesion development in multiple sclerosis. Brain 117, 59–69. Williams, K., Ulvestad, E., Antel, J.P., 1994. B7rBB-1 antigen expression on adult human microglia studied in vitro and in situ. Eur. J. Immunol. 24, 3031–3037.
Interaction between apoptotic cells and reactive brain cells in the central nervous system of rats with autoimmune encephalomyelitis Toshihiko Kohji, Naoyuki Tanuma, Yukihiko Aikawa, Yoko Kawazoe, Yoko Suzuki, Kuniko Kohyama, Yoh Matsumoto ) Department of Neuropathology, Tokyo Metropolitan Institute for Neuroscience, Musashidai 2-6, Fuchu, Tokyo 183, Japan Received 25 July 1997; received in revised form 24 September 1997; accepted 26 September 1997
Abstract To elucidate the role of brain cells in the immune regulation in the central nervous system ŽCNS., acute and chronic relapsing experimental autoimmune encephalomyelitis ŽEAE. was induced in Lewis rats and the location of apoptotic inflammatory cells and their interaction with astrocytes and microglia was investigated at various stages of the disease. Apoptotic cells detected by terminal deoxynucleotidyl transferase-mediated dUTP nick end-labeling ŽTUNEL. were few in number at day 10–12 post-immunization ŽPI., increased and peaked at day 13 PI. Then, these cells decreased gradually by day 21 PI. The most characteristic finding was that apoptotic cells were mainly distributed in the CNS parenchyma with only a few cells present in perivascular cuffs. Double staining by the TUNEL method and immunocytochemistry for astrocytes and microglia revealed that astrocytes were more closely associated with apoptotic cells than microglia. Apoptotic cell death may be one mechanism by which T cells are eliminated from the CNS. Furthermore, the present study suggests that astrocytes, rather than microglia, induce programmed cell death of infiltrating inflammatory cells. q 1998 Elsevier Science B.V. Keywords: Apoptotic cell death; Astrocytes; Experimental autoimmune encephalomyelitis ŽEAE.; T cell; Lewis rat
1. Introduction Experimental autoimmune encephalomyelitis ŽEAE. is a T cell-mediated autoimmune disease that is inducible in susceptible strains of mice, rats and other species by immunization with brain-specific antigens such as myelin basic protein ŽMBP. ŽLassmann, 1983.. Rats usually develop acute and monophasic EAE after immunization with MBP and do not show relapse of the clinical signs. We previously demonstrated that in acute EAE, T cells that enter the CNS lose their proliferating ability soon after the infiltration ŽOhmori et al., 1992.. Later, several groups showed that infiltrating T cells are eliminated from the CNS by apoptotic cell death at the peak and recovery stages of EAE ŽPender et al., 1992; Schmied et al., 1993..
) Corresponding author. Tel.: q81 423 253881r4719; fax: q81 423 218678; e-mail: [email protected]
0165-5728r98r$19.00 q 1998 Elsevier Science B.V. All rights reserved. PII S 0 1 6 5 - 5 7 2 8 Ž 9 7 . 0 0 1 9 8 - 7
In parallel with these in vivo findings, in vitro studies suggest that some brain cells such as astrocytes and microglia inhibit the proliferation of MBP-reactive T cells and induce apoptosis ŽMatsumoto et al., 1992a, 1993; Ford et al., 1996.. In general, microglia are reported to upregulate, whereas astrocytes downregulate, T cell proliferation. However, there is disagreement with regard to the antigen presenting ability of astrocytes and microglia. In the present study, we attempted to elucidate whether brain cells are involved in the process of apoptosis of infiltrating T cells and, if this is the case, to determine the type of brain cells involved. For this purpose, acute and chronic relapsing EAE was induced in Lewis rats and the location and number of apoptotic cells as well as the interaction between apoptotic cells and brain cells were investigated by terminal deoxynucleotidyl transferasemediated dUTP nick end-labeling ŽTUNEL. and immunohistochemistry. Consequently, we found that apoptotic cells were mainly located in the CNS parenchyma and that astrocytes were closely associated with apoptotic cells.
T. Kohji et al.r Journal of Neuroimmunology 82 (1998) 168–174
2. Materials and methods 2.1. Animals Lewis rats were purchased from Seiwa ŽFukuoka. and bred in our animal facility. Rats at age 8–12 weeks were used throughout the experiments. 2.2. EAE induction and tissue sampling Acute EAE was induced as described previously ŽOhmori et al., 1992.. Briefly, each rat was injected in the hind footpads bilaterally with an emulsion containing 100 m g of MBP in complete Freund’s adjuvant ŽCFA; Mycobacterium tuberculosis H37Ra, 5 mgrml.. Chronic relapsing EAE ŽCR-EAE. was induced by immunization of rats with 200 m l of guinea pig spinal cord homogenate in CFA. One gram of spinal cord tissue was homogenized in 1 ml of PBS and the suspension was emulsified with an equal volume of CFA. Starting from the day of immunization, rats were given intraperitoneal injections of cyclosporin A ŽCsA. ŽSandoz, Tokyo. at a dose of 4 mgrkg three times a week until day 21 post-immunization ŽPI.. Immunized rats were observed daily for clinical signs of EAE. The clinical stage of EAE was divided into five Žgrade 1, floppy tail; grade 2, mild paraparesis; grade 3, severe paraparesis; grade 4, tetraparesis; grade 5, death.. At various time points, rats were killed under ether anesthesia and several segments of the lumbar spinal cord were processed for paraffin-embedding. 2.3. Terminal deoxynucleotidyl transferase-mediated dUTP nick end-labeling (TUNEL) DNA fragmentation was detected by in situ nick endlabeling as described previously ŽGavrieli et al., 1992. with a few modifications. Briefly, paraffin-embedded sections were dewaxed, rehydrated through a graded alcohol series, and washed in PBS. Next, sections were treated with 0.005% protease ŽSigma, St. Louis, MO. for 20 min at 378C and subsequently incubated in a terminal deoxynucleotidyl transferase ŽTdT. buffer solution Ž140 mM sodium cacodylate, 1 mM cobalt chloride, 30 mM Tris–HCl, pH 7.2. containing 0.15 Urm l TdT ŽGibco BRL, Tokyo. and 0.004 nmolrm l digoxigenin-dUTP ŽBoehringer-Mannheim, Tokyo. for 60 min at 378C, and then in the TB buffer Ž300 mM sodium chloride, 30 mM sodium citrate. for 15 min. They were allowed to react with alkaline phosphatase ŽAP.-labeled anti-digoxigenin antibody ŽBoehringer-Mannheim. for 60 min. Positive cells were finally visualized by incubating sections with the SIGMA FAST e BCIPrNBT Buffered Substrate Tablet solution ŽSigma. containing levamisole Ž0.24 mgrml. which blocks endogenous AP.
169
2.4. Immunohistochemical staining procedures Paraffin-embedded sections were deparaffinized and rehydrated, and endogenous peroxidase was blocked by incubation in 0.3% H 2 O 2 in methanol. After incubation with normal sheep or goat serum, sections were allowed to react with the first antibody for 60 min. For the detection of microglia, biotinylated tomato lectin was used ŽAcarin et al., 1994.. Then, sections were incubated with horseradish peroxidase ŽHRP.-labeled VECTASTAIN w Elite ABC Kit ŽVector, Burlingame, CA.. HRP binding sites were detected with the SIGMA FAST e DAB Peroxidase Substrate Tablet Set solution ŽSigma.. For glial fibrillary acidic protein ŽGFAP. immunostaining, the peroxidase anti-peroxidase method was applied using Histogen PAP rabbit Universal Kit ŽBioGenex, San Ramon, ´ CA.. 2.5. Double staining for apoptotic cells and brain cells Double staining for TUNEL-positive cells and brain cells such as microglia and astrocytes was performed using paraffin-embedded sections. In the first step, apoptotic cells were detected by the TUNEL method and AP was developed in the BCIPrNBT solution as a blue color. After a thorough washing, slides were stained for microglia or astrocytes using HRP as an enzyme. HRP was developed in the DAB solution in brown. 3. Results 3.1. Clinical obserÕation The clinical course of acute and chronic relapsing EAE is shown in Fig. 1. Clinical signs of acute EAE of rats immunized with MBP developed on day 11, peaked on day
Fig. 1. Clinical course of acute ŽA. and chronic relapsing ŽB. EAE.
170
T. Kohji et al.r Journal of Neuroimmunology 82 (1998) 168–174
Fig. 2. The distribution of TUNEL-positive cells in the spinal cord during acute EAE. A few TUNEL-positive cells are located in the subarachnoid space and in the subpial region at the early stage of EAE ŽA.. TUNEL-positive cells increased in number and reached maximal levels at the peak stage ŽB and C.. At this stage, most TUNEL-positive cells were distributed in the parenchyma with very few in the perivascular cuff. There is no TUNEL-positive cells in the perivascular cuff in this field Žarrowhead in C.. TUNEL-positive cells decreased in number at the recovery stage ŽD.. Arrows in Panels A–D indicate TUNEL-positive cells. A, day 11; B and C, day 14; D, day 21. Original magnifications: =100.
13–15 and subsided by day 18 PI ŽFig. 1A.. In the present study, all the rats with acute EAE were killed for histological examination by day 21 PI. We previously observed that there was no relapse of the disease after this observation period Žour unpublished data.. In this study, the early stage of acute EAE referred to day 10 or 11 PI and the peak stage to days 13–15 PI. Rats immunized with spinal cord homogenate and treated with CsA developed clinical signs on day 11 with the peak on days 14–16 as well as acute EAE. However, all the rats that recovered from EAE showed relapse of the clinical signs between days 24 and 32 with the peak on days 25–27 ŽFig. 1B.. Based on these findings, acute EAE was examined histologically at various time points before, during and after the manifestation of clinical signs. CR-EAE was examined at the first attack, remission phase, second attack and recovery phase. 3.2. Histological and quantitatiÕe analysis of apoptotic cells in the CNS during EAE We first investigated the location of apoptotic cells in the CNS of rats with acute EAE at various time points. At
the early stage of EAE, only a small number of TUNELpositive cells were present in the subarachnoid space and the subpial region ŽFig. 2A.. At the peak stage of acute EAE, TUNEL-positive cells were distributed evenly in the parenchyma, but were rare in perivascular cuffs ŽFig. 2B
Fig. 3. The number of TUNEL-positive cells at various stages of acute EAE. The number of TUNEL-positive cellsrspinal cord section were determined using at least three rats Žtotal 9–12 segments of the spinal cord. at each time point. The values indicate means"SD.
T. Kohji et al.r Journal of Neuroimmunology 82 (1998) 168–174
Fig. 4. The number of TUNEL-positive cells at various stages of CR-EAE. The number of TUNEL-positive cellsrspinal cord section were determined using at least three rats Žtotal 9–12 segments of the spinal cord. at each time point. The values indicate means"SD. In this study, tissue sampling was performed at the 1st attack Žday 14–17 PI., remission Žday 21–23 PI., 2nd attack Žday 25–27 PI. and full recovery stage Žday 34–36 PI. of chronic relapsing EAE.
and C.. A similar distribution pattern was noticed at the first and second attack of CR-EAE Ždata not shown.. At the recovery stage of acute EAE, a few TUNEL-positive cells were present in the parenchyma ŽFig. 2D.. These findings suggest that inflammatory cells are not pro-
171
grammed to undergo apoptosis before entering the CNS, but that some factors residing in the CNS parenchyma are involved in the process. If apoptosis of infiltrating inflammatory cells is undertaken solely by the mechanism programmed in infiltrating cells, apoptotic cells would have been recognized in both the parenchyma and perivascular cuffs. We next counted TUNEL-positive cells at various stages of acute EAE ŽFig. 3.. TUNEL-positive cells were rare at the early stage, but rapidly increased and reached maximal levels on days 13 and 14 PI when the animals showed severe paraparesis. The number of TUNEL-positive cells gradually declined after the peak stage of acute EAE. Similar findings were obtained in CR-EAE ŽFig. 4.. In CR-EAE, the number of apoptotic cells in the CNS was maximal at the first attack and decreased to almost zero in the remission phase before increasing again at the second attack and declining at the recovery stage. 3.3. Relationship between apoptotic cells and brain cells Immunohistochemical study showed that the majority of apoptotic cells were located in the CNS parenchyma and
Fig. 5. Double staining for GFAP-positive Žbrown. and TUNEL-positive Ždark blue. cells ŽA and B. and for tomato lectin-positive microglia Žbrown. and TUNEL-positive cells ŽC and D.. In Panels A and B, cell processes of GFAP-positive astrocytes attach to TUNEL-positive nuclei Žarrows.. Double staining with TUNEL and tomato lectin histochemistry shows a different staining pattern. In this case, some TUNEL-positive nuclei are surrounded by lectin staining Žarrows in C and D., indicating that they are TUNEL-positive microglia. Original magnifications: =200.
172
T. Kohji et al.r Journal of Neuroimmunology 82 (1998) 168–174
Fig. 6. The number of TUNEL-positive cells contacting GFAP-positive astrocytes Žclosed square. and tomato lectin-positive microgliarmacrophages Žclosed circle.. The numbers are significantly different on days 14 Ž ) ) p- 0.01. and 15 Ž ) p- 0.05. by Student-t test. The number of contact cells was determined using at least 4 different segments of the spinal cord. The values indicate means"SD. TUNEL-positive microglia are not included in this figure.
not in the perivascular cuff ŽFig. 2.. This suggested that apoptosis occurs in association with the contact of inflammatory cells with brain cells. To examine the interaction between apoptotic cells and brain cells, we did double staining for apoptotic cells and reactive brain cells such as astrocytes ŽGFAP-positive cells. and microgliarmacrophages Žtomato lectin-positive cells.. As shown in Fig. 5A and B, cell processes of GFAP-positive astrocytes attached to TUNEL-positive nuclei. In double staining for TUNEL-positive cells and microgliarmacrophages, a similar finding was observed Ždata not shown.. In addition, some microgliarmacrophages were undergoing apoptosis ŽFig. 5C and D. as reported previously ŽNguyen et al., 1994, 1997; Tabi et al., 1995.. We then counted TUNELpositive cells contacting astrocytes or microgliarmacrophages at various stages of acute EAE ŽFig. 6.. The number of TUNEL-positive cells contacting astrocytes was low at the early stage but increased and reached maximal levels at the peak stage. Then, these cells decreased gradually at the recovery stage. TUNEL-positive cells contacting microgliarmacrophages also increased in number at the peak stage of acute EAE. However, the number of TUNEL-positive cells contacting microgliarmacrophages was significantly less than that of TUNEL-positive cells contacting astrocytes ŽFig. 6..
4. Discussion EAE is a T cell-mediated autoimmune disease characterized by the presence of inflammatory cells mainly comprising T cells and macrophages in the CNS. Although EAE is inducible in several species including mice, rats, and guinea pig by immunization with brain-specific antigen, its clinical course is different from one species to another. It is generally believed that differences in the clinical course depend on both the animal species and the encephalitogenic antigen used. Rats immunized with guinea
pig MBP show acute monophasic EAE. Infiltration of inflammatory cells into the CNS begins around day 10 PI and reaches a maximum on day 12. Between days 12 and 14, inflammation plateaus and then most lesions subside by day 21 ŽMatsumoto et al., 1993.. Although these processes are well documented, the kinetics of T cells is poorly understood. Our previous study using immunohistochemistry showed that the majority of T cells involved in EAE undergo DNA synthesis outside the CNS and then infiltrate the CNS parenchyma and that infiltrating T cells do not proliferate vigorously in the CNS ŽOhmori et al., 1992.. Later, several groups demonstrated that this phenomenon is attributable to apoptotic cell death of T cells ŽPender et al., 1991, 1992; Schmied et al., 1993; Tabi et al., 1994, 1995.. In the present study, we quantitated TUNEL-positive cells in the spinal cord throughout the course of acute EAE and found that the number of TUNEL-positive cells reached a maximal level on days 13 and 14 PI and declined gradually ŽFig. 3.. This finding is consistent with that reported by McCombe et al. Ž1996. who examined apoptotic cells in the isolated inflammatory cell population during acute EAE by flow cytometry. In our chronic relapsing model, the number of TUNEL-positive cells paralleled well with the clinical course of the disease ŽFig. 4.. Although Schmied et al. Ž1993. reported that more TUNEL-positive cells were found at the later stage, this may be due to the difference in the method of the EAE induction. They immunized SD rats with guinea pig spinal cord homogenate in CFA, which may induce chronic EAE Žthe precise clinical course was not shown in their report.. In addition to this, the difference in the detection procedures may contribute to the difference of the outcome. We used terminal deoxynucleotidyl transferase to detect DNA fragmentation in apoptosis ŽTUNEL., while they used DNA polymerase Žnick translation.. It is reported that cells undergoing apoptosis are preferentially labeled by TUNEL, whereas necrotic cells are identified by nick translation ŽGold et al., 1994.. Thus, the TUNEL method which we employed here may detect apoptotic processes at an earlier stage. One of the characteristic findings of the present study was that the majority of apoptotic cells were located in the CNS parenchyma, and not in the perivascular cuff ŽFig. 2 C . . A lth o u g h so m e to m ato lectin -p o sitiv e microgliarmacrophages underwent apoptosis, the number of such cells was far low compared to tomato lectin-negative cells Ždata not shown.. Therefore, it was judged that the majority of TUNEL-positive cells were T cells. These findings strongly suggest that brain cells are involved in the apoptotic processes of infiltrating T cells. If this is not the case, that is, if apoptotic signals are programmed in T cells before entry into the CNS, then apoptotic cells would have been found equally in both the CNS parenchyma and perivascular region. Based on these findings, we examined the relationship between apoptotic cells and brain cells by double staining. It was revealed that apoptotic cells were
T. Kohji et al.r Journal of Neuroimmunology 82 (1998) 168–174
more closely associated with astrocytes than with microglia with the most marked difference noted at the peak stage of EAE ŽFig. 6.. In our previous study ŽMatsumoto et al., 1992b., we observed that microglia react with infiltrating T cells and proliferate at the early stage whereas astrocytes react to inflammatory lesions at the late peak to recovery stages. Therefore, it is conceivable that astrocytes play some role in the induction of T cell apoptosis. Recent studies on T cell activation via T cell receptors with or without costimulatory signals revealed that T cells stimulated with an antigen presented by antigen-presenting cells ŽAPC. with costimulatory signals through CD28rB7 molecules induce T cell proliferation, whereas the same stimulation without costimulatory signals results in T cell anergy or antigen-induced cell death ŽJaneway and Bottomly, 1994.. Regarding brain cells, it is reported that microglia, but not astrocytes, express CD28rB1 molecules ŽWilliams et al., 1994; de Simone et al., 1995; Satoh et al., 1995; Iglesias et al., 1997.. In accord with this finding, in vitro studies showed that microglia expressing MHC class II antigens have an antigen-presenting ability ŽFrei et al., 1986; Matsumoto et al., 1992a; Cash et al., 1993.. However, Ford et al. Ž1996. reported that ramified microglia isolated from normal adult brain do not induce T cell proliferation. On the other hand, there is a consensus that astrocytes are poor antigen presenting cells and suppress T cell proliferation even after the induction of MHC class II molecules ŽMatsumoto et al., 1993; Weber et al., 1994.. Recently, both microglia and astrocytes have been shown to have ability to induce apoptosis of activated T cells ŽFord et al., 1996; Gold et al., 1996.. Taking these findings into consideration, the results obtained in the present study can be interpreted as follows. At the early stage of EAE, microglia proliferate in the vicinity of infiltrating T cells. However, it is unlikely that microglia at this stage induce T cell apoptosis because only a small number of apoptotic cells and of apoptotic cells contacting microglia are found in the CNS. Astrocytes do not respond to inflammatory lesions at the early stage ŽMatsumoto et al., 1992b.. At the peak stage, apoptotic cells and those contacting astrocytes reach maximal levels suggesting that astrocytes mainly prime infiltrating T cells for apoptotic cell death. This assumption does not exclude the possibility that microglia also induce apoptosis at the peak stage of acute EAE. In any case, brain cells may play a major role in apoptosis of infiltrating T cells, and thus the majority of apoptotic cells were found in the CNS parenchyma.
Acknowledgements This study was supported in part by grants from the Ministry of Education, Japan, Toyama Chemical Co. and Naito Foundation.
173
References Acarin, L., Vela, J.M., Gonzaez, ´ B., Castellano, B., 1994. Demonstration of Poly-N-acetyl lactosamine residues in ameboid and ramified microglial cells in rat brain by tomato lectin binding. J. Histochem. Cytochem. 42, 1003–1041. Cash, E., Zhang, Y., Rott, O., 1993. Microglia present myelin antigen to T cells after phagocytosis of oligodendrocytes. Cell. Immunol. 147, 129–138. de Simone, R., Giampaolo, A., Giometto, B., Gallo, P., Levi, G., Peschle, C., Aloisi, F., 1995. The costimulatory molecule B7 is expressed on human microglia in culture and in multiple sclerosis acute lesions. J. Neuropathol. Exp. Neurol. 54, 175–187. Ford, A.L., Foulcher, D., Lemckert, F.A., Sedgwick, J.D., 1996. Microglia induce CD4 T lymphocyte final effector function and death. J. Exp. Med. 184, 1737–1745. Frei, K., Bodmer, S., Schwerdel, C., Fontana, A., 1986. Astrocyte-derived interleukin 3 as a growth factor for microglia cells and peritoneal macrophages. J. Immunol. 137, 3521–3527. Gavrieli, Y., Sherman, Y., Ben-Sasson, S.A., 1992. Identification of programmed cell death in situ via specific labeling of nuclear DNA fragmentation. J. Cell Biol. 119, 493–501. Gold, R., Schmied, M., Giegerich, G., Breitschopf, H., Hartung, H.P., Toyka, K.V., Lassmann, H., 1994. Differentiation between cellular apoptosis and necrosis by the combined use of in situ tailing and nick translation techniques. Lab. Invest. 71, 219–225. Gold, R., Schmied, M., Tontsch, U., Hartung, H.P., Wekerle, H., Toyka, K.V., Lassmann, H., 1996. Antigen presentation by astrocytes primes rat T lymphocytes for apoptotic cell death. Brain 119, 651–659. Iglesias, B.M., Cerase, J., Ceracchini, C., Levi, G., Aloisi, F., 1997. Analysis of B7-1 and B7-2 costimulatory ligands in cultured mouse microglia: Upregulation by interferon-g and lipopolysaccharide and downregulation by interleukin-10, prostaglandin E 2 and cyclic AMPelevating agents. J. Neuroimmunol. 72, 83–93. Janeway, C.A. Jr., Bottomly, K., 1994. Signals and signs for lymphocyte responses. Cell 76, 275–285. Lassmann, H., 1983. Comparative Neuropathology of Chronic Experimental Allergic Encephalomyelitis and Multiple Sclerosis. SpringerVerlag, Berlin, pp. 9–94. Matsumoto, Y., Ohmori, K., Fujiwara, M., 1992a. Immune regulation by brain cells in the central nervous system: Microglia but not astrocytes present myelin basic protein to encephalitogenic T cells under in vivo-mimicking conditions. Immunology 76, 209–216. Matsumoto, Y., Ohmori, K., Fujiwara, M., 1992b. Microglial and astroglial reactions to inflammatory lesions of experimental autoimmune encephalomyelitis in the rat central nervous system. J. Neuroimmunol. 37, 23–33. Matsumoto, Y., Hanawa, H., Tsuchida, M., Abo, T., 1993. In situ inactivation of infiltrating T cells in the central nervous system with autoimmune encephalomyelitis. The role of astrocytes. Immunology 79, 381–390. McCombe, P.A., Nickson, I., Tabi, Z., Pender, M.P., 1996. Apoptosis of Vb 8.2q T lymphocytes in the spinal cord during recovery from experimental autoimmune encephalomyelitis induced in Lewis rats by inoculation with myelin basic protein. J. Neurol. Sci. 139, 1–6. Nguyen, K.B., McCombe, P.A., Pender, M.P., 1994. Macrophage apoptosis in the central nervous system in experimental autoimmune encephalomyelitis. J. Autoimmun. 7, 145–152. Nguyen, K.B., McCombe, P.A., Pender, M.P., 1997. Increased apoptosis of T lymphocytes and macrophages in the central and peripheral nervous systems of Lewis rats with experimental autoimmune encephalomyelitis treated with dexamethasone. J. Neuropathol. Exp. Neurol. 56, 58–69. Ohmori, K., Hong, Y., Fujiwara, M., Matsumoto, Y., 1992. In situ demonstration of proliferating cells in the rat central nervous system during experimental autoimmune encephalomyelitis. Evidence sug-
174
T. Kohji et al.r Journal of Neuroimmunology 82 (1998) 168–174
gesting that most infiltrating T cells do not proliferate in the target organ. Lab. Invest. 66, 54–62. Pender, M.P., Nguyen, K.B., McCombe, P.A., Kerr, J.F.R., 1991. Apoptosis in the nervous system in experimental allergic encephalomyelitis. J. Neurol. Sci. 104, 81–87. Pender, M.P., McCombe, A., Yoog, G., Nguyen, K.B., 1992. Apoptosis of ab T lymphocytes in the nervous system in experimental autoimmune encephalomyelitis: Its possible implication for recovery and acquired tolerance. J. Autoimmun. 5, 401–410. Satoh, J., Lee, Y.B., Kim, S.U., 1995. T-cell costimulatory molecules B7-1 ŽCD80. and B7-2 ŽCD86. are expressed in human microglia but not in astrocytes in culture. Brain Res. 704, 92–96. Schmied, M., Breitschopf, H., Gold, R., Zischler, H., Rothe, G., Wekerle, H., Lassmann, H., 1993. Apoptosis of T lymphocytes in experimental autoimmune encephalomyelitis. Evidence for programmed cell death as a mechanism to control inflammation in the brain. Am. J. Pathol. 143, 446–452.
Tabi, Z., McCombe, P.A., Pender, M.P., 1994. Apoptotic elimination of Vb 8.2q cells from the central nervous system during recovery from experimental autoimmune encephalomyelitis induced by the passive transfer of Vb 8.2q encephalitogenic T cells. Eur. J. Immunol. 24, 2609–2617. Tabi, Z., McCombe, P.A., Pender, M.P., 1995. Antigen-specific downregulation of myelin basic protein-reactive T cells during spontaneous recovery from experimental autoimmune encephalomyelitis: Further evidence of apoptotic deletion of autoreactive T cells in the central nervous system. Int. Immunol. 7, 967–973. Weber, F., Meinl, E., Aloisi, F., Nevinny-Stickel, C., Albert, E., Wekerle, H., Hohlfeld, R., 1994. Human astrocytes are only partially competent antigen presenting cells. Possible implication for lesion development in multiple sclerosis. Brain 117, 59–69. Williams, K., Ulvestad, E., Antel, J.P., 1994. B7rBB-1 antigen expression on adult human microglia studied in vitro and in situ. Eur. J. Immunol. 24, 3031–3037.