Kinetics of anaphylatoxin C5a receptor expression during experimental allergic encephalomyelitis

Kinetics of anaphylatoxin C5a receptor expression during experimental allergic encephalomyelitis

Journal of Neuroimmunology 91 Ž1998. 147–155 Kinetics of anaphylatoxin C5a receptor expression during experimental allergic encephalomyelitis Serge N...

2MB Sizes 0 Downloads 13 Views

Journal of Neuroimmunology 91 Ž1998. 147–155

Kinetics of anaphylatoxin C5a receptor expression during experimental allergic encephalomyelitis Serge Nataf, Nathalie Davoust, Scott R. Barnum

)

Department of Microbiology, DiÕision of Clinical Immunology and Rheumatology, UniÕersity of Alabama at Birmingham, BHSB 306, 1918 UniÕersity BouleÕard, Birmingham, AL 35294, USA Received 1 May 1998; revised 24 June 1998; accepted 24 June 1998

Abstract In this study, we investigated the expression of the C5aR in spinal cords of Lewis rats with experimental allergic encephalomyelitis ŽEAE.. Using in situ hybridization ŽISH. we analyzed the kinetics of C5aR at different time points of EAE Žpreclinical stage, clinical peak, remission phase.. We observed that C5aR mRNA was readily detected in the CNS of EAE rats at all the stages of the disease. Using a combination of ISH and immunohistochemistry, we formally demonstrated that C5aR is strongly expressed on microglial cells and hypertrophic astrocytes during EAE. The potential involvement of C5a receptor in EAE physiopathology is discussed. q 1998 Elsevier Science B.V. All rights reserved. Keywords: Complement; C5a receptor; Experimental allergic encephalomyelitis; Multiple sclerosis; Central nervous system

1. Introduction Experimental allergic encephalomyelitis ŽEAE. is an autoimmune disease of the central nervous system ŽCNS. considered a relevant animal model for multiple sclerosis ŽMS.. In this model, T cells sensitized with neuroantigens proliferate at the periphery and migrate across the blood– brain barrier ŽBBB. into the central nervous system ŽCNS.. It is assumed that a second presentation of the antigen inside the CNS induces a localized activation of the encephalitogenic T cells leading to a massive recruitment of inflammatory cells, a breakdown of the BBB and extensive demyelination ŽLassman, 1983; Hart and Fabry, 1995.. Although EAE in its classical form is a T cell-mediated disease, studies have demonstrated a role for complement ŽC. in the development of both clinical signs and demyelination ŽVanguri et al., 1982; Liu et al., 1983; Linington et al., 1989; Piddlesden et al., 1993.. Support for the role of complement in the pathogenesis of EAE also comes from studies by Piddlesden et al. Ž1994. who demonstrated that the soluble complement receptor type 1 ŽsCR1., which blocks C activation, inhibited antibody-mediated EAE. )

Corresponding author: Tel.: q1 205 9344972; fax: q1 205 9344985

Activation of complement generates proteolytic fragments, with potent inflammatory properties. Among these fragments is the anaphylatoxin C5a whose functions in inflammation during EAE have not yet been investigated. C5a is a potent chemoattractant for neutrophils and monocytermacrophages ŽShin et al., 1968; Fernandez et al., 1978.. In addition, microglia, astrocytes and O2-A progenitors derived from the rat brain, have also been shown to be chemotactic for C5a ŽArmstrong et al., 1990; Yao et al., 1990.. Taken together these results suggest that C5a may play a role in recruitment of both infiltrating and resident cells to sites of CNS inflammation. C5a binds to a specific cell-surface receptor, the C5aR, which belongs to a large family of seven-transmembrane-domain receptors that are G-protein-coupled ŽGerard and Gerard, 1994; Wetsel, 1995.. Definitive evidence of the expression of the C5aR on cultured human astrocytes and microglia at the mRNA and protein level has been obtained only recently ŽLacy et al., 1995; Gasque et al., 1995.. Neuronal expression of the C5aR has also been documented in a mouse model of Listeria meningitis in which infected animals had high levels of C5aR expression on neurons throughout the brain ŽStahel et al., 1997a.. Similarly, C5aR expression was elevated on neurons in a brain-injury model ŽStahel et al.,

0165-5728r98r$ - see front matter q 1998 Elsevier Science B.V. All rights reserved. PII: S 0 1 6 5 - 5 7 2 8 Ž 9 8 . 0 0 1 6 9 - 6

148

S. Nataf et al.r Journal of Neuroimmunology 91 (1998) 147–155

1997b.. Finally, additional evidence supporting C5aR involvement in CNS inflammatory diseases is derived from recent studies showing that C5aR expression is markedly elevated in acute and chronic MS lesions ŽMuller-Ladner ¨ et al., 1996.. Both blood-derived infiltrating cells Žfoamy macrophages. and glial-like cells identified as hypertrophic fibrous astrocytes, as well as endothelial cells, expressed the C5aR at elevated levels in the inflammatory stages of the disease, but not in chronic silent disease tissue ŽMuller-Ladner et al., 1996.. Other investigators have re¨

cently confirmed increased C5aR expression in active MS lesions ŽGasque et al., 1997.. To overcome some of the limitations of using postmortem tissue and to be able to follow changes in C5aR expression during the course of disease, we analyzed C5aR expression in the CNS during chronic relapsing EAE of the Lewis rat at different disease stages. We found that C5aR mRNA expression, as assessed by in situ hybridization ŽISH., is elevated in cells infiltrating the CNS and blood vessels before the occurrence of clinical symptoms,

Fig. 1. Basal expression of the C5aR in Lewis rat spinal cord in healthy control and CFA-treated rats. Spinal cords from healthy control and CFA-treated rats, sacrificed 9 days after immunization, were removed, snap frozen and then transverse sections Ž8 mm. were prepared for in situ hybridization using C5aR antisense and sense riboprobes, as described in Section 2. ŽA. In situ hybridization using a C5aR antisense riboprobe on spinal cord meninges and white matter from a control rat. Note the large cytoplasmic cells Žblack arrows. and the blood vessel-like structure Žwhite arrow.. ŽB. Same as in panel A, except the section shows more parenchyma and some gray matter. Neurons in the ventral horn Žwhite arrow. and the dorsal horn Žblack arrow. are weakly C5aR-positive. ŽC. In situ hybridization with a C5aR antisense riboprobe on CFA-treated rat spinal cord. Motor neurons in the ventral horn are strongly positive Žwhite arrow. as well as blood vessels in the white matter. ŽD. Same as C, but showing the white matter at higher magnification. ŽE. Same as C and D, showing a C5aR-positive branching vessel at high magnification. ŽF. In situ hybridization using a C5aR sense probe on a spinal cord from an healthy control rat. Original magnifications A, 50 = ; B and C, 20 = ; D, 100 = ; E, 500 = ; F, 100 = .

S. Nataf et al.r Journal of Neuroimmunology 91 (1998) 147–155

and then increases as disease progresses. Moreover, combining ISH and immunohistochemistry allowed us to identify C5aR expressing cells as monocytermacrophages, activated microglia, as well as astrocytes. Surprisingly, we also found increased expression of C5aR on blood vessels and neurons of control animals immunized with CFA alone.

2. Materials and methods 2.1. EAE induction and clinical assessment One set of experiments were performed in which a total of 14 female Lewis rats Ž8–9 weeks old. obtained from Charles River France ŽCleon ´ ., were immunized for EAE ŽEAE rats. as previously described ŽFeurer et al., 1985.. Briefly, 1 g of guinea pig spinal cord tissue was homogenized with 1 ml saline and then emulsified with 2 ml of Difco bacto complete Freund’s adjuvant ŽCFA. supplemented with 40 mg of Mycobacterium tuberculosis H37RA ŽDifco Laboratories, Detroit, MI.. Rats were injected intradermally into each footpad with 0.1 ml of this emulsion, then examined daily for grading of clinical signs as follows: 1 s loss of tail tonicity, 2 s weakness of one or both hind legs or mild ataxia, 3 s severe ataxia or paralysis, 4 s severe hind leg paralysis accompanied by urinary incontinence. Clinical symptoms appeared on day 9 Žgrade 1., and all animals developed a severe paralysis of grade 4 on day 12 and completely recovered on days 17–18. There was no variation in clinical expression of the disease from rat to rat at these time points, as previously reported ŽFeurer et al., 1985.. As a control, three Lewis female rats were immunized with CFA alone ŽCFA control rats. and three did not receive any injections Žhealthy control rats.. 2.2. Riboprobes C5aR riboprobes were prepared by linearizing a murine C5aR cDNA clone prepared in our laboratory ŽStahel et al., 1997a., using XbaI and ClaI endonucleases ŽPromega, Madison, WI. for generation of antisense and sense riboprobes, respectively. In vitro transcription was performed using a RNA transcription kit ŽPromega. and probes were labeled with digoxigenin-UTP ŽBoehringer Mannheim.. 2.3. In situ hybridization EAE rats were randomly chosen and killed by CO 2 inhalation on days 7 Ž n s 3., 9 Ž n s 3., 12 Ž n s 4. or on day 18 Ž n s 4., a time at which all animals had completely recovered. CFA control rats Ž n s 3. were killed 9 days after immunization as were age-matched healthy control rats Ž n s 3.. Lumbothoracic spinal cords were surgically removed, snap frozen in liquid nitrogen-chilled isopentane and stored at y808C until use. Prior to sectioning, the

149

frozen tissue was embedded in Tissue-Tek. In situ hybridization was performed on 8-mm-thick transversal sections using digoxigenin-labeled cRNA and detection by alkaline phosphatase color reaction, as previously described ŽStahel et al., 1997a.. Briefly, sections were fixed for 1 h in 3% paraformaldehyde ŽPFA; pH 7.4. and hybridized for 16 h at 508C with riboprobes diluted 1r10 in hybridization buffer ŽFisher Scientific, Fairlawn, NJ.. After hybridization, three consecutive washes at increasing stringency Žfinal wash in 0.1 = SSC containing 0.1% SDS for 5 min. were performed at 508C. Immunological detection was performed by incubating slides with anti-digoxigenin alkaline phosphatase-conjugated Fab ŽBoehringer Mannheim. followed by 5-bromo-4-chloro-3-indolyl-phosphate ŽBCPI.r4-nitro blue tetrazolium chloride ŽNBT.color substrate solution ŽSigma.. Sections were examined under medium magnification and the number of C5aRpositive cells was graded semi-quantitatively as Ž0. absence of positive cells, Ž". very few positive cells, Žq. few positive cells, Žqq . numerous positive cells. 2.4. Immunohistochemistry To identify the cells expressing C5aR mRNA, ISH was combined with immunohistochemistry using the following primary antibodies: ED1 Mab ŽSerotec, Kidlington, UK. specific for monocytermacrophages and activated microglia, and anti-glial fibrillary acidic protein ŽGFAP. polyclonal antibody ŽAccurate Chemical and Scientific, Westbury, NY. specific for astrocytes. In control experiments, immunohistochemistry without ISH was performed according to an immunoenzymatic staining method supplied by the manufacturer ŽVectastain ABC kit, Vector, Burlingame, CA.. Briefly, sections were fixed in cold absolute ethanol for 10 min then sequentially incubated for 60 min with ED1 Mab or anti-GFAP antibody, followed by biotin-conjugated goat anti-rabbit IgG ŽSigma. for GFAP staining or biotin-conjugated donkey anti-mouse

Table 1 C5aR mRNA expression during EAE as assessed by ISH Infiltrating cells

Control Ž ns 3. CFA day 9 Ž ns 3. EAE day 7 Ž ns 3. EAE day 9 Ž ns 3. EAE day 12 Ž ns 4. EAE day 18 Ž ns 4.

Blood vessels

Meninges

Parenchyma

0 0 q qq qq qq

0 0 q qq qq: Mc,T qq: As

" qq " qq qq qq

The number of C5aR mRNA expressing cells or blood vessels was semi-quantitatively evaluated at medium magnification. The following arbitrary scale was used: 0: no signal, ": very few cells or blood vessels, q: few, qq: numerous. Mc: macrophagermicroglia. T: T lymphocytes. As: astrocytes.

150

S. Nataf et al.r Journal of Neuroimmunology 91 (1998) 147–155

IgG ŽJackson Immunoreaserch, Baltimore, USA. for ED1 staining. Sections were then rinsed in PBS and incubated in a solution of 1% hydrogen peroxide for 15 min. After several rinses, the sections were treated with an avidin-biotinylated–peroxidase complex for 50 min at room temperature ŽVectastain ABC kit, Vector, Burlingame, CA. and followed by 0.04% diaminobenzidine ŽSigma. in PBS with

0.01% H 2 O 2 for 10 min. For the combination of anti-GFAP or ED1 staining with ISH, immunohistochemistry was performed after ISH using a procedure similar to the one described above except for the fixation step in ethanol which in this case was omitted. As a control and to verify that ISH procedures did not interfere with immunohistochemistry, sections were also immunostained with either

Fig. 2. Kinetics of C5aR expression in EAE spinal cord meninges and white matter. Spinal cords from EAE rats sacrificed at days 7, 9 or 12 after immunization, were analyzed by in situ hybridization for C5aR mRNA expression. ŽA. In situ hybridization on spinal cord from an EAE rat on day 7 after immunization. Note the few C5aR-positive cells infiltrating the parenchyma. ŽB. In situ hybridization on spinal cord from an EAE rat on day 9 after immunization. ŽC. Same as B at higher magnification. Note the marked increase in C5aR expression on blood vessels Žblack arrows. and the C5aR-positive infiltrating cells Žwhite arrows.. ŽD,E. In situ hybridization on spinal cord from two different EAE rats on day 12 after immunization. Note the marked meningeal infiltration in panel E. ŽF. Same as panel E, using a sense probe on an adjacent section. Original magnifications A, B, F, 100 = ; C, E, 200 = ; D, 50 = .

S. Nataf et al.r Journal of Neuroimmunology 91 (1998) 147–155

an isotype-matched mouse monoclonal antibody or without primary antibody. In all these procedures, ISH was performed using both a sense or an antisense C5aR probe.

3. Results 3.1. Expression of C5aR in control rats In healthy control rats, a faint signal obtained with a C5aR antisense probe was observed in only few vessel-like structures as well as some parenchymal cells with a large cytoplasm and short processes ŽFig. 1A.. By combining ISH and immunostaining with an anti-GFAP Ab, these cells were shown to belong to the astrocytic lineage ŽFig. 3F.. In the gray matter, C5aR mRNA could be detected in the cell body of motor neurons and in some neurons in the

151

dorsal horn ŽFig. 1B.. In CFA-treated rats killed on day 9 post-immunization, numerous blood vessels located in the white matter expressed C5aR mRNA ŽFig. 1C–E, Table 1.. However, these vessels and their vicinity were totally free of any blood-derived infiltrating cells. In 2 of 3 animals, the level of specific C5aR mRNA detected in motor neurons appeared slightly increased compared to the basal level observed in healthy control rats ŽFig. 1C.. As a control, hybridization with a C5aR sense probe was performed and no signal was observed either in healthy control rats ŽFig. 1F. or CFA-treated rats Ždata not shown.. 3.2. Kinetic of C5aR expression in EAE rats At day 7 post-immunization, the animals examined Ž n s 3. did not have any clinical signs of disease, and only few round cells expressed a strong specific signal for a

Fig. 3. Expression of the C5aR by astrocytes in EAE. Spinal cords from EAE rats sacrificed on day 18 after immunization were analyzed by a combination of in situ hybridization for the C5aR and immunohistochemistry for GFAP to identify astrocytes, as described in Section 2. ŽA. In situ hybridization using a C5aR antisense riboprobe in spinal cord parenchyma of an EAE rat at day 18. ŽB. In situ hybridization using a C5aR antisense riboprobe Žpurple color. combined with immunohistochemistry staining for GFAP to identify astrocytes Žbrown color.. ŽC. Higher magnification of a double-positive cell shown in panel B. ŽD. In situ hybridization and immunohistochemistry as in panel B in EAE rat spinal cord gray matter. ŽE. In situ hybridization using a C5aR sense riboprobe combined with immunohistochemistry staining for GFAP in EAE rat spinal cord parenchyma. ŽF. In situ hybridization and immunohistochemistry as in panel B in healthy control rat spinal cord gray matter. Original magnifications A and D, 200= ; B, 100 = ; C, E and F, 500 = .

152

S. Nataf et al.r Journal of Neuroimmunology 91 (1998) 147–155

C5aR mRNA. These cells were located either in the parenchyma of the white matter or in the meninges ŽFig. 2A, Table 1.. However, the meninges were largely clear of C5aR-positive infiltrating cells at this time point and blood vessels and neurons showed the same pattern of C5aR mRNA expression as control healthy rats Ždata not shown.. At day 9 post-immunization, all the animals examined Ž n s 3. presented with mild clinical signs at the time of sacrifice Žaverage score: 1.. Numerous round infiltrating cells in the meninges were strongly positive for the C5aR. These cells were also seen in the parenchyma of the white matter, as well as, inside and around the blood vessels ŽFig. 2B and C.. A large number of blood vessels also expressed specific C5aR mRNA even when not surrounded by any infiltrate ŽFig. 2B and C.. On day 12 post-immunization, at the peak of clinical disease Žaverage score: 4., we did not observe any striking qualitative or quantitative modifications of C5aR mRNA expression when compared to day 9 ŽFig. 2D.. However, in 1 animal of 4, some round cells were also observed in the gray matter Ždata not shown. along with a diffuse and intense meningeal infiltration ŽFig. 2E.. The possibility of artifact linked to an edge effect was ruled out using a sense probe on adjacent sections ŽFig. 2F.. On day 18 post-immunization, during the remission phase Žaverage score: 0.5., strongly C5aR-positive cells were seen, presenting as either round or of glial-like morphology ŽFig. 3A.. Moreover, most of these positive cells were located in the vicinity of infiltrated blood vessels and only few round

cells could be detected in the parenchyma. At any of the time points examined, the background signal on hybridization with a sense probe was essentially identical to that shown in Fig. 2F. 3.3. Identification of the cells expressing the C5aR mRNA To determine the identity of C5aR-expressing cells, we examined sections of spinal cord from animals at various times post-immunization, using a combination of immunohistochemistry Žfor cellular identification. and ISH Žfor the C5aR.. On day 18 post-immunization, during the remission phase of EAE, some of the cells expressing C5aR mRNA clearly appeared as glial-like cells and were located in the parenchyma surrounding infiltrated vessels ŽFig. 3A.. Double staining experiments using an anti-GFAP antibody and an antisense C5aR probe showed that these cells were indeed hypertrophic reactive astrocytes ŽFig. 3B and C.. Using the same technique, we also observed a population of GFAP-positive astrocytes in the gray matter, near to or in contact with motor neurons. These astrocytes did not express or only very weakly expressed C5aR mRNA ŽFig. 3D.. As a control, ISH was performed using a C5aR sense probe which showed no background signal, while GFAP staining was readily detectable ŽFig. 3E.. In addition, omission of the GFAP antibody resulted a very low background for the immunohistochemistry procedure Ždata not shown.. Double staining of astrocytes from

Fig. 4. Expression of the C5aR by microglia and infiltrating mononuclear cells in EAE. Spinal cords from EAE rats sacrificed on day 12 after immunization were analyzed by a combination of in situ hybridization for the C5aR and immunohistochemistry using the ED-1 antibody to identify activated mononuclear cells, as described in Section 2. ŽA. In situ hybridization using a C5aR antisense riboprobe Žpurple color. combined with immunohistochemistry staining using the ED-1 antibody to identify monocytic cells Žbrown color.. ŽB. High magnification of a double-positive cell bearing short processes. ŽC. In situ hybridization using a C5aR sense riboprobe combined with immunohistochemistry staining with the ED-1 antibody in EAE rat spinal cord parenchyma. Original magnifications, A, 50 = ; B and C, 500 = ; D, 100 = .

S. Nataf et al.r Journal of Neuroimmunology 91 (1998) 147–155

healthy control rats showed a markedly lower level of expression of the C5aR compared to EAE rats ŽFig. 3F.. Infiltrating macrophages or activated microglia were identified using the ED1 Mab which binds to a specific moiety expressed on these cells ŽFlaris et al., 1993.. We examined spinal cord sections from animals at 12 days post-immunization and found strong staining in a large number of infiltrating cells in the meninges, in and around blood vessels and in the parenchyma as would be expected at this stage of disease Ždata not shown.. By a combination of immunohistochemistry and ISH, we observed ED1-positive cells that were either positive or negative for the expression of the C5aR in the parenchyma of these animals ŽFig. 4A.. Some ED1-positive, C5aR-positive cells had short cellular processes and can be considered as putative activated microglia ŽBauer et al., 1994. ŽFig. 4B.. In control experiments, we did not observe colocalization of the signals when combining ED1 staining and ISH using a C5aR sense probe ŽFig. 4C..

4. Discussion Though different studies have suggested a role for complement in EAE pathophysiology ŽShin and Koski, 1992; Spiegel et al., 1997., the function of the receptor for the anaphylatoxin C5a has not been investigated. Our report is the first demonstration that C5aR mRNA is expressed by both infiltrating cells and CNS resident cells during the course of EAE. Moreover, C5aR-expressing cells were seen in the meninges and CNS parenchyma during all the stages of the disease and as early as 2 days before the onset of clinical signs. These data extend previous reports demonstrating increased expression of the C5aR at the protein level on hypertrophic astrocytes and macrophagermicroglial cells in multiple sclerosis ŽMS. ŽMuller-Ladner et al., 1996; Gasque et al., 1997.. Al¨ though the kinetics of C5aR expression have not been examined in MS, we have previously shown that C5aR expression, along with receptors for IL-8 and N-formylMet-Leu-Phe, appears highest in the acute and chronic active stages of the disease ŽMuller-Ladner et al., 1996.. ¨ These data suggest that C5a, the C5aR ligand, could exert specific functions depending on the target-cell and the stage of disease. During EAE, the early events which lead to a massive transmigration of inflammatory cells into the CNS include an upregulation of adhesion molecules on both endothelial cells and leukocytes, a phenomenon which has been assumed to be predominantly TNF-a and IFN-g-mediated ŽBaron et al., 1993; Barten and Ruddle, 1994.. Interestingly, we found that C5aR mRNA expression is strongly upregulated in CNS blood vessels in EAE rats, as well as CFA-treated rats. Mycobacterium tuberculosis, the main immunogenic agent contained in CFA, is able to induce C3 cleavage and the subsequent activation of the alternative

153

complement pathway ŽSchorey et al., 1997.. Therefore, it is conceivable that C5aR expression on blood vessels of CFA-treated rats results, in part, from the inflammatory cascade induced by alternative pathway activation leading to C3a and C5a generation. It should be noted that, at day 9, EAE rats differ from CFA-treated rats by the presence of numerous, round infiltrating cells expressing the C5aR mRNA. Thus, in EAE, CNS infiltration could require both a CFA-dependent endothelial activation and a neuroantigen-specific T cell stimulation. Since C5a has previously been shown to induce adhesion molecule expression on endothelial cells in vitro ŽForeman et al., 1994., our data suggest that one possible function of the C5aR, in EAE or MS, would be activation of the endothelium, along with cytokines, to increase adhesion molecule expression, thereby facilitating attachment of activated leukocytes to CNS vasculature. During EAE relapses, the transendothelial migration of leukocytes Žchemotaxis. into the surrounding CNS parenchyma, is due, in part, to locally synthesized a- and b-chemokines ŽGlabinski et al., 1997.. Similarly, the C5aR expression we observed on monocytermacrophages in the spinal cord meninges ŽFig. 2. may have contributed to recruitment of these cells to sites of EAE-induced inflammation. Support for this hypothesis is derived from studies demonstrating C5a-induced meningitis in animal models of bacterial meningitis and, in sterile models of meningitis, in which injection of C5a alone into CSF induced a massive infiltration of leukocytes into the CSF and brain parenchyma ŽErnst et al., 1984; Kadurugamuwa et al., 1989; Faustmann et al., 1995.. Further, other studies have shown that depletion of complement delays the intrathecal appearance of leukocytes in meningitis models ŽTuomanen et al., 1986.. Along with blood-derived infiltrating cells, C5aR mRNA was detected in other CNS parenchymal cells including activated microglia, hypertrophic astrocytes as well as neurons. Interestingly, we found that, in healthy rats, a subset of astrocytes located in the white matter, expressed constitutively a low level of C5aR mRNA ŽFig. 1.. Similarly, C5aR expression has been recently demonstrated in primary cultures of rat astrocytes ŽSayah et al., 1997.. It is thus possible that the C5aR-positive astrocytes we observed in healthy rats belong to the subset of reactive astrocytes which had a high level of C5aR mRNA expression during EAE remission Žday 18. and were located exclusively in the white matter ŽFig. 2.. This suggests that, under inflammatory conditions, these ‘sentinel’ astrocytes could upregulate C5aR expression at the same time they become activated and then be recruited to sites of inflammation. In fact, recruitment more than proliferation accounts for astrocyte accumulation during reactive gliosis in EAE ŽMatsumoto et al., 1992.. Since C5a is a potent chemoattractant for rat astrocytes ŽArmstrong et al., 1990., we would hypothesize that C5aR-expressing astrocytes first migrate toward sites of inflammation then differentiate to hypertrophic reactive astrocytes and participate in

154

S. Nataf et al.r Journal of Neuroimmunology 91 (1998) 147–155

glial scar formation. Based on this hypothesis, we suggest that a heterogeneous distribution of chemotactic receptors on different astrocyte subpopulations would allow a fine regulation of astrocyte activation and trafficking through the effects exerted by multiple chemotactic factors including C5a, chemokines Žsuch as IL-8. or f-Met-Leu-Phe ŽFMLP., all of which have receptors on astrocytes ŽGlabinski et al., 1995; Lacy et al., 1995.. In contrast to astrocytes, resting microglial cells did not appear to express C5aR mRNA in healthy rats and, markedly elevated levels of C5aR mRNA were detected in activated ED1-positive microglia at the peak of disease ŽFig. 4.. These differences suggest that the regulation of C5aR expression on CNS cells depends on numerous parameters including cell-type, state of activation andror proliferation and spatial localization. Little is known about regulation of C5aR expression and, until recently, no inflammatory mediator has clearly been shown to upregulate C5aR expression in glial cells. Studies from our laboratory have shown that TNF-a mediates a striking upregulation of neuronal C5aR expression in a mouse model of experimental Listeria-induced meningoencephalomyelitis ŽStahel et al., 1997a.. The fact that we did not observe such a phenomenon in EAE rats, a model in which TNF-a is readily synthesized locally ŽVillaroya et al., 1996., indicates that neuronal regulation of C5aR expression may vary depending on the species andror the inflammatory conditions. In another model system, using CNS-specific expression of IL-3 ŽChiang et al., 1996., we have also observed elevated expression of the C5aR on microglia and neurons, suggesting that IL-3 can regulate C5aR expression in the CNS ŽParadisis et al., 1998.. It is worth noting the differences in C5aR expression on neurons in EAE vs. the bacterial meningitis model we have recently examined. In EAE, we observed a variable, but low constitutive expression of the C5aR on motor neurons in the spinal cord ŽFig. 1.. This expression was elevated in some of the CFA-treated animals and in some of the EAE rats. Receptor expression was variable even within a given section and no clear pattern of induction of C5aR expression on neurons was seen. In contrast, C5aR expression was markedly elevated in the cerebrum and cerebellum of mice with Listeria-induced meningitis ŽStahel et al., 1997a.. Further, constitutive expression of the C5aR was almost undetectable in control mice in the meningitis model. The differences in C5aR expression on neurons in these two model systems may reflect species-specific differences, responses to different subsets of cytokines and neurotrophic factors andror differential expression on neuronal subsets Žmotor neurons vs. cortical neurons and Purkinje cells.. We are pursuing studies to address this issue. In the present study, we clearly demonstrate that C5aR mRNA is expressed in the CNS during the entire course of EAE, beginning before the onset of clinical signs. However, the number of C5aR-expressing cells did not appear

to correlate with the clinical expression of EAE. C5aR mRNA assessed by ISH appeared quantitatively similar from day 9 Žbefore the peak of the disease. to day 18 Žduring the remission phase.. Indeed, the regulation of C5aR expression at the protein level may involve posttranscriptional mechanisms and analyzing C5aR protein expression may reveal differences between stages of the disease. Moreover, as shown in Table 1, qualitative differences in C5aR expression were demonstrated between day 9 or 12, and day 18. Thus, at day 18, C5aR-expressing cells surrounding blood vessels were mostly astrocytes while, at the peak of the disease, a majority of these cells were either macrophagermicroglia or T cells. During chronic relapsing EAE, a second wave of inflammatory cells infiltrate the CNS around day 23 post-immunization ŽNataf et al., 1996.. Therefore, the fact that astrocytes strongly express C5aR during the remission phase, may indeed reflect an ongoing immune process more than a cicatricial event. We speculate that an intravascular generation of C5a participates in the chemoattraction and activation of astrocytes which in turn will stimulate the extravasation of leukocytes. In summary, since C5a is a potent inflammatory mediator, our data suggest that C5a- andror C5aR-blocking strategies could be of interest in both treating or preventing autoimmune disorders of the CNS. However, in vitro studies are needed to elucidate the exact significance of C5aR mRNA expression we observed on neurons and T lymphocytes. The recent cloning of the rat C5aR ŽAkatsu et al., 1997., will allow the development of rat-specific anti-C5aR antibodies which will be useful tools for characterizing C5aR expression on these cell types and evaluating C5aR-blocking treatments in EAE.

Acknowledgements This study was supported by NIH grant NS29719 Žto S.R.B... The authors thank Mike Strawn ŽDepartment of Photography and Instructional Graphics, UAB. for technical help with the illustrations.

References Akatsu, H., Miwa, T., Sakurada, C., Fukuoka, Y., Ember, J.A., Yamamoto, T., Hugli, T.E., Okada, H., 1997. cDNA cloning and characterization of rat C5a anaphylatoxin receptor. Microbiol. Immunol. 41, 575–580. Armstrong, R.C., Harvath, L., Dubois-Dalcq, M.E., 1990. Type 1 astrocytes and oligodendrocyte-type 2 astrocyte glial progenitors migrate toward distinct molecules. J. Neurosci. Res. 27, 400–407. Baron, J.L., Madri, J.A., Ruddle, N.H., Hashim, G., Janeway, C.A. Jr., 1993. Surface expression of a4 integrin by CD4 T cells is required for their entry into brain parenchyma. J. Exp. Med. 177, 57–68. Barten, D.M., Ruddle, N.H., 1994. Vascular cell adhesion molecule-1 modulation by tumor necrosis factor in experimental allergic encephalomyelitis. J. Neuroimmunol. 51, 123–133. Bauer, J., Sminia, T., Wouterlood, F.G., Dijkstra, C.D., 1994. Phagocytic activity of macrophages and microglial cells during the course of

S. Nataf et al.r Journal of Neuroimmunology 91 (1998) 147–155 acute and chronic relapsing experimental autoimmune encephalomyelitis. J. Neurosci. Res. 38, 365–375. Chiang, C.S., Powell, H.C., Gold, L.H., Samimi, A., Campbell, I.L., 1996. Macrophagermicroglial-mediated primary demyelination and motor disease induced by the central nervous system production of interleukin-3 in transgenic mice. J. Clin. Invest. 97, 1512–1524. Ernst, J.D., Hartiala, K.T., Goldstein, I.M., Sande, A.M., 1984. Complement ŽC5.-derived chemotactic activity accounts for accumulation of polymorphonuclear leukocytes in cerebrospinal fluid of rabbits with pneumococcal meningitis. Infect. Immun. 46, 81–86. Faustmann, P.M., Krause, D., Dux, R., Dermietzel, R., 1995. Morphological study in the early stages of complement C5a fragment-induced experimental meningitis: activation of macrophages and astrocytes. Acta Neuropathol. 89, 239–247. Fernandez, H.N., Henson, P.M., Otani, A., Hugli, T.E., 1978. Chemotactic response to human C3a and C5a anaphylatoxins: I. Evaluation of C3a and C5a leukotaxis in vitro and under stimulated in vivo conditions. J. Immunol. 120, 109–115. Feurer, C., Prentice, D.E., Cammisuli, S., 1985. Chronic relapsing encephalomyelitis in the Lewis rat. J. Neuroimmunol. 10, 159–165. Flaris, N.A., Densmore, T.L., Molleston, M.C., Hickey, W.F., 1993. Characterization of microglia and macrophages in the central nervous system of rats: definition of the differential expression of molecules using standard and novel monoclonal antibodies in normal CNS and in four models of parenchymal reaction. Glia 7, 34–40. Foreman, K.E., Vaporciyan, A.A., Bonish, B.K., Jones, M.L., Johnson, K.J., Glovsky, M.M., Eddy, S.M., Ward, P.A., 1994. C5a-induced expression of P-selectin in endothelial cells. J. Clin. Invest. 94, 1147–1155. Gasque, P., Chan, P., Fontaine, M., Ishenko, A., Lamacz, M., Gotze, O., ¨ Morgan, B.P., 1995. Identification and characterization of the complement C5a anaphylatoxin receptor on human astrocytes. J. Immunol. 155, 4882–4889. Gasque, P., Singhrao, S.K., Neal, J.W., Gotze, O., Morgan, B.P., 1997. Expression of the receptor for complement C5a ŽCD88. is up-regulated on reactive astrocytes, microglia, and endothelial cells in the inflamed human central nervous system. Am. J. Pathol. 150, 31–41. Gerard, C., Gerard, N.P., 1994. C5a anaphylatoxin and its seven transmembrane-segment receptor. Annu. Rev. Immunol. 12, 775–808. Glabinski, A., Tani, M., Aras, S., Stoller, M., Tuohy, V., Ransohoff, R.M., 1995. Regulation and function of central nervous system chemokines. Int. J. Dev. Neurosci. 13, 153–165. Glabinski, A.R., Tani, M., Strieter, R.M., Tuohy, V.K., Ransohoff, R.M., 1997. Synchronous synthesis of a- and b-chemokines by cells of diverse lineage in the central nervous system of mice with relapses of chronic experimental autoimmune encephalomyelitis. Am. J. Pathol. 150, 617–630. Hart, M.N., Fabry, Z., 1995. CNS antigen presentation. Trends Neurosci. 18, 475–481. Kadurugamuwa, J.L., Hengstler, B., Bray, M.A., Zak, O., 1989. Inhibition of complement-factor-5a-induced inflammatory reactions by prostaglandin E2 in experimental meningitis. J. Infect. Dis. 160, 715–719. Lacy, M., Jones, J., Whittemore, S.R., Haviland, D.L., Wetsel, R.A., Barnum, S.R., 1995. Expression of the receptors for the C5a anaphylatoxin, interleukin-8 and fMLP by human astrocytes and microglia. J. Neuroimmunol. 61, 71–78. Lassman, H., 1983. Comparative neuropathology of chronic experimental allergic encephalomyelitis and multiple sclerosis. Springer-Verlag, Berlin. Linington, C., Morgan, B.P., Scolding, N.J., Wilkins, P., Piddlesden, S., Compston, D.A.S., 1989. The role of complement in the pathogenesis of experimental allergic encephalomyelitis. Brain 112, 895–911. Liu, W.T., Vanguri, P., Shin, M.L., 1983. Studies on demyelination in vitro: the requirement of membrane attack components of the complement system. Ann. Neurol. 131, 778–782.

155

Matsumoto, Y., Ohmori, K., Fujiwara, M., 1992. Microglial and astroglial reactions to inflammatory lesions of experimental autoimmune encephalomyelitis in the rat central nervous system. J. Neuroimmunol. 37, 23–33. Muller-Ladner, U., Jones, J., Gay, S., Wetsel, R.A., Raine, C., Barnum, ¨ S.R., 1996. Enhanced expression of chemotactic receptors in multiple sclerosis lesions. J. Neurol. Sci. 144, 135–144. Nataf, S., Garcion, E., Darcy, F., Chabannes, D., Muller, J.Y., Brachet, P., 1996. 1,25-dihydroxyvitamin D3 exerts regional effects in the central nervous system during experimental allergic encephalomyelitis. J. Neuropathol. Exp. Neurol. 55, 904–914. Paradisis, P.M., Campbell, I.L., Barnum, S.R., 1998. Elevated complement C5a receptor expression on glial cells and neurons induced by the central nervous system production of interleukin-3 in transgenic mice. Glia, in press. Piddlesden, S.J., Lassman, H., Zimprich, F., Morgan, B.P., Linington, C., 1993. The demyelinating potential of antibodies to myelin oligodendrocyte glycoprotein is related to their ability to fix complement. Am. J. Pathol. 143, 555–564. Piddlesden, S.J., Storch, M.K., Hibbs, M., Freeman, A.M., Lassman, H., Morgan, B.P., 1994. Soluble recombinant complement receptor 1 inhibits inflammation and demyelination in antibody-mediated demyelinating experimental allergic encephalomyelitis. J. Immunol. 152, 5477–5484. Sayah, S., Patte, C., Gasque, P., Chan, P., Ishenko, A., Vaudry, A., Fontaine, M., 1997. Characterization of rat C5a anaphylatoxin receptor ŽC5aR.: cloning of rat C5aR cDNA and study of C5aR expression by rat astrocytes. Mol. Brain Res. 48, 215–222. Schorey, J.S., Carroll, M.C., Brown, E.J., 1997. A macrophage invasion mechanism of pathogenic mycobacteria. Nature 277, 1091–1093. Shin, M.L., Koski, C.L., 1992. The complement system in demyelination. In: Martenson, R.E. ŽEd.., Myelin: Biology and Chemistry. Boca Raton, CRC Press, pp. 801–831. Shin, H.S., Snyderman, R., Friedman, E., Mellors, A., Mayer, M.M., 1968. Chemotactic and anaphylatoxic fragment cleaved from the fifth component of guinea pig complement. Science 162, 361–363. Spiegel, K., Emmerling, M.R., Barnum, S.R., 1997. Acute phase proteins: Strategies for inhibition of complement activation in the treatment of neurodegenerative disorders. In: Totowa, W.P. ŽEd.., Inflammatory Mechanisms and Management of Neurodegeneration. Humana Press, NJ, pp. 129–176. Stahel, P.F., Frei, K., Eugster, H.P., Fontana, A., Hummel, K.M., Wetsel, R.A., Ames, R., Barnum, S.R., 1997a. TNF-a-mediated expression of the receptor for anaphylatoxin C5a ŽC5aR. on neurons in experimental Listeria meningocephalitis. J. Immunol. 159, 861–869. Stahel, P.F., Kossmann, T., Hans, V., Morganti-Kossmann, M.C., Barnum, S.R., 1997b. Experimental diffuse axonal injury induces enhanced neuronal C5a receptor expression in rats. Mol. Brain Res. 50, 205–212. Tuomanen, E., Hengstler, B., Zak, O., Tomasz, A., 1986. The role of complement in inflammation during experimental pneumococcal meningitis. Microb. Pathog. 1, 15–32. Vanguri, P., Koski, C.L., Silverman, B., Shin, M.L., 1982. Complement activation by isolated myelin: activation of the classical pathway in the absence of myelin-specific antibodies. P.N.A.S. USA 79, 3290– 3294. Villaroya, H., Violleau, K., Younes-Chenoufi, A.B., Baumann, N., 1996. Myelin-induced experimental allergic in Lewis rats: tumor necrosis factor a levels in serum and cerebrospinal fluid. Immunohistochemical expression in glial cells and macrophages of optic nerve and spinal cord. J. Neuroimmunol. 64, 105–118. Wetsel, R.A., 1995. Structure, function and cellular expression of complement anaphylatoxin receptors. Curr. Opin. Immunol. 1, 48–53. Yao, J., Harvath, L., Gilbert, D.L., Colton, C.A., 1990. Chemotaxis by a CNS macrophage, the microglia. J. Neurosci. Res. 27, 36–42.