1,25-Dihydroxyvitamin D3 inhibits the expression of inducible nitric oxide synthase in rat central nervous system during experimental allergic encephalomyelitis

Molecular Brain Research 45 Ž1997. 255–267

Research report

1,25-Dihydroxyvitamin D 3 inhibits the expression of inducible nitric oxide synthase in rat central nervous system during experimental allergic encephalomyelitis E. Garcion a , S. Nataf a , A. Berod b, F. Darcy a , P. Brachet a

a,)

Institut National de la Sante´ et de la Recherche Medicale, Unite´ 298, Centre Hospitalier UniÕersitaire, Angers, France ´ b Unite´ 339, Hopital St. Antoine, Paris, France ˆ Accepted 17 September 1996

Abstract The inducible form of nitric oxide synthase ŽiNOS. generates nitric oxide of which the excessive production is associated with central nervous system ŽCNS. inflammatory diseases. The investigation of iNOS expression during experimental allergic encephalomyelitis ŽEAE. of the Lewis rat demonstrated iNOS immunoreactivity and mRNA both during inflammatory bursts Ždays 12 and 23 post-immunization. and during the remission phase Žday 18.. iNOS expression was region-specific and expanded with time along a caudo-rostral axis, thus, correlating with the development of inflammatory infiltrates. Whereas cells of the monocytermacrophage lineage continuously contributed to iNOS expression, astrocytes only expressed iNOS immunoreactivity or mRNA during the relapse Žday 23.. In order to investigate possible regulatory effects of 1,25-dihydroxyvitamin D 3 Ž1,25-D 3 . on iNOS expression, rats were treated with the hormone after the beginning of clinical signs Ždays 11, 13, 19, 21 and 23 post-immunization., and areas of the CNS were examined at day 23. 1,25-D 3 exerted a drastic inhibitory effect on iNOS expression, both at the protein and the mRNA levels. However, this effect was region-specific, and was most pronounced in the cerebellum and brainstem, but non-existent in cerebral cortex. iNOS down-regulation occurred in macrophages, activated microglia and astrocytes. The inhibition of iNOS expression in some CNS structures could account for the improvement of clinical signs observed in EAE-rats treated with 1,25-D 3. Since 1,25-D 3 can be synthesized by activated macrophages or microglia, our results support the hypothesis that this hormone might be implicated in the control of the CNS-specific immune responses. 1,25-D 3 or its analogues could, thus, be of therapeutic value in the management of iNOS-associated diseases of the CNS. Keywords: Nitric oxide; Monocyte; Macrophage; Microglia; Astrocyte; Vitamin D; Calcitriol; Brain; Glial cell; Free radical; Inflammation; Cytokine

1. Introduction Initially called EDRF Žendothelium-derived relaxing factor., nitric oxide ŽNO. is a free gaseous radical which mediates a variety of biological functions, including vasodilation, neurotransmission and cytotoxicity w9x. NO is produced from L-arginine by the enzyme NO-synthase ŽNOS. for which different isoforms have been described in the central nervous system ŽCNS.. Constitutive forms of NOS ŽcNOS. are Ca2q- and calmodulin-dependent, while inducible forms ŽiNOS. are Ca2q and calmodulin-independent and continuously produce NO w16,26,28x. During brain inflammation, such as in viral infections

)

Corresponding author. Fax: q33 Ž2. 4173-1630.

or experimental allergic encephalomyelitis ŽEAE., the iNOS gene is up-regulated w21x. Increased NO production may be responsible, to some extent, for the alteration of brain functions or CNS tissue destruction in EAE-rats w21,25,29,46,58x. In vitro studies suggest that glial cell-derived NO may cause the death of oligodendrocytes and, thus, could participate in the formation of lesions during multiple sclerosis ŽMS. w3,37x. This hypothesis is supported by the fact that human iNOS mRNA is markedly elevated in demyelinated regions of the brains of MS patients when compared to control brains w6x. However, conflicting results have been reported concerning the effects of pharmacological inhibitors of iNOS on the clinical course of EAE w8,59x, probably because NO can exert both beneficial and harmful effects. In view of the potential role of NO in the pathogenesis

0169-328Xr97r$17.00 Copyright q 1997 Elsevier Science B.V. All rights reserved. PII S 0 1 6 9 - 3 2 8 X Ž 9 6 . 0 0 2 6 0 - 4

256

E. Garcion et al.r Molecular Brain Research 45 (1997) 255–267

of the CNS, the regulation of iNOS expression has become a matter of intense investigation. Lipopolysaccharide ŽLPS. and IFNg , alone or in combination with TNFa or IL1 b , are potent inducers of iNOS gene expression in macrophages, astrocytes or microglia w57x. In contrast, anti-inflammatory cytokines, such as interleukin-4 ŽIL-4., IL-10, IL-13 and TGFb , have been shown to prevent iNOS induction andror counteract NO cytotoxicity w11x. Likewise, glucocorticoids and the steroid-like hormone, retinoic acid, have been shown in vitro to antagonize iNOS synthesis in different cell types, including murine macrophages w10,18,35,49,53x. Like glucocorticoids and retinoic acid, the active hormonal form of vitamin D, 1,25-dihydroxyvitamin D 3 Ž1,25-D 3. , exerts profound immunosuppressive effects. However, little is known about its ability to regulate iNOS, especially in CNS cells. 1,25-D 3 binds to an intracellular receptor, the vitamin D receptor ŽVDR., which belongs to the steroid-thyroid hormone receptor superfamily and acts as a ligand-dependent transcription factor w17x. In addition to its role in the control of calcemia and bone metabolism, 1,25-D 3 displays anti-inflammatory and immunomodulatory properties w30x. This hormone inhibits B and T lymphocyte proliferation w22,30x, down-regulates the production of IL-2 and IFNg by T lymphocytes w4x, and prevents immunoglobulin secretion by B lymphocytes w24x. In vivo, the immunosuppressive properties of 1,25-D 3 have been demonstrated in various animal models of autoimmunity, such as experimental autoimmune thyroiditis or diabetes w15,31x. Moreover, 1,25-D 3 administration at the time of immunization with myelin basic protein ŽMBP., also prevents the development of EAE in SJL mice w7,23x. There are presently several lines of evidence which indicate that the CNS constitutes a target of 1,25-D 3 . Binding sites of the radiolabelled hormone have been demonstrated in certain neurons and in non-neuronal structures, such as perivascular organs w5,55x. Moreover, the VDR mRNA was detected in the human hippocampus w56x and in newborn rat astrocyte cultures w42x in which 1,25-D 3 was shown to enhance the expression of nerve growth factor ŽNGF. and neurotrophin-3 ŽNT3. transcripts w41,42x. In addition, 1,25-D 3 can be synthesized from its precursor, 25-hydroxyvitamin D 3 Ž25-D 3. , by activated microglial cells w43x, a finding which suggests the existence of local regulatory loops controlling the action of 1,25-D 3 on brain homeostasis. A recent study from our laboratory has shown that 1,25-D 3 can exert immunomodulatory effects inside the CNS during an ongoing EAE immune process in the Lewis rat w39x. Treatment of animals after the appearance of clinical signs of EAE resulted in a significant clinical improvement, associated with a down-regulation of CD4 antigen expression by infiltrating macrophages and activated microglial cells in the brainstem and cerebellum. Since EAE pathology is associated with an excessive production of NO, the aim of the present study was to investigate the regional and cellular expression of the

iNOS gene and iNOS immunoreactivity during the time course of chronic relapsing EAE in rats, and to explore the modulatory effects of 1,25-D 3 on the iNOS synthesis during the progression of EAE.

2. Materials and methods 2.1. Animals and EAE induction 8–9-week-old Lewis female rats were obtained from Charles River France ŽCleon ´ . and bred in our animal facilities. Chronic relapsing EAE was induced in rats as previously described w14,39x. Briefly, guinea-pig spinal cords ŽGPSC. were obtained from female Dunkin-Hartley animals. 1 g GPSC was homogenized in 1 ml saline, then emulsified with 2 ml of complete Freund’s adjuvant ŽCFA. supplemented with 40 mg of Mycobacterium tuberculosis H37RA ŽDifco Laboratories, Detroit, MI, USA.. Rats were injected intradermally into each footpad with 0.1 ml of this emulsion. Immunized rats underwent a first crisis which began at day 10 post-immunization, characterized by a severe ataxia with a peak at day 12 followed by a remission phase at days 14–18 post-immunization. A relapse was subsequently observed in most of the animals between days 19 and 23 post-immunization. The time course of EAE progression was very reproducible w39x. 2.2. Treatment A total of 37 age-matched Lewis female rats were used for this study. The 5 experimental groups comprised: Ž1. control rats which were not immunized for EAE Žcontrol rats, n s 3.; Ž2. control rats not immunized, treated with 1,25-D 3 Žcontrol 1,25-D 3-treated rats, n s 4.; Ž3. rats immunized for EAE ŽEAE-control rats, n s 15.; Ž4. rats immunized for EAE and treated with vehicle alone ŽEAE vehicle-treated rats, n s 7.; and Ž5. rats immunized for EAE and treated with 1,25-D 3 ŽEAE-1,25-D 3-treated rats, n s 7.. Groups 4 and 5 were further pooled since no significant differences were observed either at the level of clinical manifestation of EAE or histochemical data. In the present study, they were, thus, referred to as EAE-control rats. 1,25-D 3 was a kind gift of Dr. L. Binderup ŽLeo Pharmaceutical Products, Ballerup, Denmark.. 1,25-D 3 was dissolved at the appropriate concentration in 80% propylene glycol and 20% 0.05 M disodium phosphate at pH 7.4, and administered by i.p. injection. For EAE 1,25-D 3treated rats and control 1,25-D 3-treated rats, a 2-step protocol was applied. It consisted of 2 injections of 1,25-D 3 at 5 m grkg on days 11 and 13, followed by 3 injections at 1 m grkg on days 19, 21 and 23. The treatment was interrupted during the remission phase Ždays 14–18.. In parallel, EAE vehicle-treated rats received an i.p. injection of propylene glycol and sodium diphosphate without 1,25-D 3 . This therapeutic protocol allowed us to study the effects of

E. Garcion et al.r Molecular Brain Research 45 (1997) 255–267

1,25-D 3 after the appearance of the clinical signs, when neuroimmunological interactions had begun. 2.3. Immunohistochemical techniques Animals were killed on days 12, 18 or 23 post-immunization for the EAE-control group. EAE-1,25-D 3-treated rats were all killed at day 23. Brains and lumbothoracic spinal cords were surgically removed, snap frozen in isopentane cooled with liquid nitrogen and stored at y808C until immunostaining was performed. 10-m m transverse sections of anterior brain, midbrain, brainstem and cerebellum, and 20-m m transverse sections of lower thoracic or upper lumbar spinal cord were cut with a cryostat. The sections were fixed in cold absolute ethanol for 10 min, then washed in phosphate-buffered saline ŽPBS. and incubated for 30 min in a solution of 10% normal goat serum in PBS. The sections were then incubated overnight at 48C with the following primary antibodies: OX42 ŽSerotec, Kidlington, UK., a monoclonal antibody ŽMab. directed against the complement receptor type 3 ŽCR3. which is expressed on monocytesrmacrophages and microglia; a rabbit polyclonal antibody directed against GFAP ŽDakopatts, Glostrup, Denmark., a specific marker of astrocytes. An affinity-purified rabbit polyclonal antibody directed against a 21-kDa fragment of the mouse macrophage iNOS was obtained from Transduction Laboratories ŽLexington, KN, USA.. The specificity of this antibody was assessed by western blots of CNS tissues from EAE-rats, which showed a 130-kDa band corresponding to the molecular weight of iNOS w26x Ždata not shown.. Control experiments were performed with a monoclonal antibody ŽTransduction Laboratories., whose specificity was previously established w33x. Staining with this monoclonal antibody was fainter than that obtained with the polyclonal antibody. However, both antibodies decorated the same cells, so that the staining pattern was strictly coincident. Because of its higher signal to noise ratio, experiments presented below were performed with the polyclonal antibody. For this purpose, sections were incubated with the 1st antibody in the presence of 0.1% Triton X-100. After rinsing in PBS, sections were incubated for 40 min with the corresponding secondary antibodies: a rat-adsorbed biotinylated horse anti-mouse antibody ŽVector, Burlingame, CA, USA. or a rat-adsorbed biotinylated donkey anti-rabbit antibody ŽAmersham, Little Chalfont, UK.. The sections were then washed again, exposed to an avidin-biotin-peroxidase complex for 60 min at room temperature ŽVectastain ABC kit; Vector. and reacted with 0.04% diaminobenzidine in PBS with 0.01% H 2 O 2 for 6–10 min. For GFAP-staining, the antibody fixation was revealed by a streptavidin-fluorescein complex ŽAmersham, Little Chalfont, UK. instead of the ABC kit. In all cases, sections processed without primary antibodies were used as controls. Double-staining experiments carried out to co-localize iNOS- and CR3-positive cells or iNOS- and GFAP-posi-

257

tive astrocytes were achieved by first incubating the sections with a mixture of the 2 primary antibodies produced in different species Žrabbit anti-iNOS, mouse anti-CR3 ŽOX42. and mouse anti-GFAP ŽTED1, kind gift of Dr. E. Geisert, University of Tennessee Health Science Center, Memphis, TN... After washing in PBS, the sections were incubated with the 2 corresponding secondary antibodies Žgoat anti-rabbit TRITC-conjugated antibody ŽBiosys, Compiegne, France. and biotinylated anti-mouse antibody ` ŽAmersham.. for 60 min. The antibody fixation was finally revealed by the streptavidin-fluorescein complex. 2.4. In situ hybridization Two oligodeoxynucleotide probes were used: a 39-mer Žsequence 5X-TCC CCT CTG ATG GTG CCA TCG GGC ATC TGG TAG CCA GCG-3X . complementary to iNOS mRNA of both monocytesrmacrophages and astrocytes; and a 48-mer Žsequence: 5X-CTC CTG CAT TTC TTC CTG ATA GAG GTG GTC CTC CTC TGG GTG CCT GCA-3X . complementary to iNOS mRNA of astrocytes only. The latter sequence differs from that of monocytermacrophage iNOS by 5 bases w18,30,32x. The 2 probes were 3X end-labelled using a- 35 S-deoxyadenosine triphosphate Žspecific activity ) 37 TBqrmmol, Amersham. and terminal deoxynucleotidyltransferase ŽBoehringer Mannheim, Germany.. The specific activity was in the range of 37–74 TBqrmmol. Frozen sections were thawed at room temperature for 10 min, fixed with 4% formaldehyde in phosphate-buffered saline ŽPBS. pH 7.4 for 10 min, acetylated in a solution of 4 = standard sodium citrate ŽSSC. pH 8.0, containing 0.25% acetic anhydride and 0.1 M triethanolamine for 10 min at room temperature, dehydrated in ethanol and airdried. The labelled oligonucleotides Ž0.5 pmolrml. were dissolved in the hybridization solution Ž50% vrv formamide, 4 = SSC, 1 = Denhardt’s solution, 1% sarcosyl, 10 mM dithiothreitol, 0.1 M potassium phosphate pH 7.4, 250 mgrml yeast tRNA and 250 mgrml herring sperm DNA., and applied onto each slide. Hybridization was allowed to proceed for 18 h at q428C. Coverslips were then carefully removed in 4 = SSC and slides were rinsed at 458C in 4 = SSC, 0.02% sodium dodecyl sulfate ŽSDS. for 30 min, then twice in 1 = SSC for 30 min and twice in 0.1 = SSC for 30 min before being dehydrated in absolute ethanol and air-dried at room temperature. Autoradiograms were generated by apposition of 35 Slabelled sections to b max Hyperfilm ŽAmersham. for 10 days at room temperature, while other sections were dipped in K5 nuclear emulsion ŽIlford, Cheshire, UK. and exposed for 1 month in the dark for cellular localization. Sections were then slightly counterstained by toluidine blue. The specificity of the labelling was assessed by means of several criteria. In situ hybridizations were performed with 2 probes which are complementary to different regions of the iNOS mRNA. In addition, for hybridiza-

258

E. Garcion et al.r Molecular Brain Research 45 (1997) 255–267

tion-positive control, we used a heterologous probe Žcomplementary to tyrosine hydroxylase mRNA.. Competition between labeled iNOS probes and excess of iNOS cold probes Ž50 = . and hybridization of samples of other tissues and structures not known to contain iNOS mRNA Žcontrol brains. allowed us to assess the labelling specificity. 2.5. Semi-quantitatiÕe analysis CR3- and iNOS-positive cells were counted under medium magnification Ž=20. by 2 independent investigators. For each region, 2 adjacent slides and 10 fieldsrslide were randomly examined. Results are given as the mean cell number obtained from the analysis of 20 fieldsrregion. Semi-quantitative results were recorded using the following arbitrary scale: 0, no positive cells; ", 1–2 cells; q, 2–20 cells; qq, 20–100 cells; qqq, ) 100 cells; qqqq, intense and widespread staining. 2.6. In situ hybridization analysis Radioautographs obtained from sections of brainstem and cerebellum were quantified by image computer analysis ŽBiocom, Les Ullis, France.. For each slide, the percentage of total cerebral area corresponding to foci which express iNOS mRNA were determined. Results are given as the mean value obtained from the analysis of 2 brain sections from 3 different rats minimum.

3. Results 3.1. Characterization of iNOS-immunoreactiÕe cells in CNS during EAE 3.1.1. Cells of monocyter macrophage lineage In control rats, no iNOS-immunoreactive cells were detected in CNS sections ŽFig. 1C.. OX42-staining revealed resting microglia which appeared as cells with a small body and thin elongated processes, either with a bipolar morphology in the white matter or with a stellate morphology in the grey matter ŽFig. 1A.. A few OX42positive monocytes, identified on the basis of their small size and rounded morphology, were also detected in the ventricular compartments, but not in the cerebral parenchyma Žnot shown.. The kinetics of development of iNOS immunoreactivity was next investigated during the course of EAE induction. On day 12 ŽTable 1., OX42-positive macrophages and reactive microglia were predominantly localized to the spinal cord and were less abundant in brainstem and cerebellum sections. The OX42-positive cells observed in the cerebellum were predominantly located inside the granule cell layer and the adjacent white matter. Few foci of stained cells were observed in anterior brain and midbrain

sections. Staining of sections in the same area with OX42ŽFig. 1B. and iNOS-antibodies ŽFig. 1D. and double-stainings Žnot shown. showed that a large proportion of OX42positive cells also displayed iNOS immunoreactivity. This observation was consistent throughout all areas examined. Interestingly, the iNOS-immunoreactive cells mainly displayed the rounded morphology of monocytesrmacrophages ŽFig. 1D,E.. Hence, these cells could be monocytesrmacrophages or reactive microglial cells, with a rounded body and few short and thick processes. Indeed, it is well-recognized that these 2 cell populations are very difficult to distinguish on the sole basis of morphological criteria w38x. In addition, a few iNOS-immunoreactive cells consisted of activated microglia, characterized by a hypertrophic cell body with numerous thick processes ŽFig. 1F.. Resting microglia was not stained with the iNOS antibody. It is worth mentioning that, at this early time point Žday 12. in EAE development, we never observed iNOS-positive immunoreactive astrocytes, such as those illustrated in Fig. 1G, which were only detected at later times. On day 18, during the remission phase, the pattern of iNOS immunoreactivity was very similar to that observed at day 12 in the spinal cord ŽFig. 2C., but the staining appeared more pronounced in the cerebellum ŽFig. 2A. and the brainstem ŽTable 1.. This reflected a parallel increase in the number of OX42-positive cells in the same areas ŽTable 1.. Likewise, while almost absent on day 12, immunostaining with iNOS- and OX42-antibodies became detectable on day 18 in the anterior brain as well as in the midbrain, being especially abundant in the striatum and corpus callosum. As observed at day 12, double-staining experiments confirmed that more than half of the OX42positive cells were iNOS-positive ŽFig. 3A,B.. Moreover, most of the iNOS-positive cells were of round morphology, consistent with the phenotype of monocytesrmacrophages or reactive microglial cells. A few iNOS-positive cells had the typical morphology of activated microglial cells. On day 23, during the 1st relapse of EAE, the iNOS immunoreactivity observed in monocytesrmacrophages appeared similar to that observed on day 18 ŽFig. 2B., with the exception of the spinal cord in which the number of iNOS-positive cells appeared to decrease ŽFig. 2D.. Nevertheless, the distribution pattern of OX42-positive infiltrating cells was quite similar to that observed on day 18 ŽTable 1.. 3.1.2. Astrocytes During the early stages of EAE Žday 12., GFAP-positive activated astrocytes were not found, as observed in the normal CNS ŽTable 2.. This observation is in agreement with previous studies which showed that astroglial activation during EAE occurs later than microglial activation w32x. Moreover, none of these cells were labelled with iNOS antibody at day 12. During the recovery stage, at day 18, only a few astrocytes expressed iNOS immuno-

E. Garcion et al.r Molecular Brain Research 45 (1997) 255–267

259

Fig. 1. OX42- ŽA. and iNOS-labelling ŽC. in control rat brain Žbrainstem.. OX42- ŽB. and iNOS-staining ŽD. in EAE-control rat brain at day 12 Žbrainstem.. Illustration of the iNOS-immunoreactive cell types observed in the rat CNS during the course of EAE: monocytesrmacrophages or reactive microglia ŽE., activated microglia ŽF. and astrocytes ŽG.. Bar ŽA–D. s 40 m m; bar ŽE–G. s 20 m m.

260

E. Garcion et al.r Molecular Brain Research 45 (1997) 255–267

Fig. 2. iNOS-labelling of monocytesrmacrophages and reactive microglia at day 18 in EAE-control rats in the cerebellum ŽA. and spinal cord ŽC.. iNOS immunoreactivity at day 23 in EAE-control rats in cerebellum ŽB., spinal cord ŽD., fronto-parietal cortex ŽE. and brainstem ŽG.. The iNOS-labelling is observed both on inflammatory cells and astrocytes ŽB and D, arrows.. iNOS immunoreactivity at day 23 in EAE-1,25-D 3 -treated rats in fronto-parietal cortex ŽF. and brainstem ŽH.. Bar s 40 m m.

E. Garcion et al.r Molecular Brain Research 45 (1997) 255–267

261

Fig. 3. Identification of the cell types expressing iNOS by double-staining experiments. In EAE-control rats Žday 18., double-staining of cerebellum sections with OX42- ŽA. and iNOS-antibodies ŽB. demonstrated that numerous cells of the monocytermacrophage lineage express iNOS. At day 23 in the cerebellum, almost all the GFAP-positive reactive astrocytes ŽC. were labelled by the iNOS-antibody ŽD.. In the cerebellum of EAE rats treated with 1,25-D 3 , GFAP-positive reactive astrocytes ŽE. did not express iNOS immunoreactivity ŽF.. Bar s 20 m m.

E. Garcion et al.r Molecular Brain Research 45 (1997) 255–267

262

Table 1 Immunohistochemical analysis of iNOS expression and OX42-labelling in cells of the monocytermacrophage lineage

iNOS-D12 OX42-D12 iNOS-D18 OX42-D18 iNOS-D23 OX42-D23 iNOS-D23rD3 OX42-D23rD3

Anterior brain

Midbrain

Brainstem

Cerebellum

Spinal cord

q q qq qq qq qqq q q

q q qq qq qq qqq qq qq

qq qq qqq qqq qqq qqqq " qqqq

qq qq qqq qqq qqq qqq " qqq

qqqq qqqq qqqq qqqq qq qqqq " qqqq

Semi-quantitative counts were carried out in EAE-rats at days 12 and 18. At day 23, data are compared with those from EAE-rats treated with 1,25-D 3 ŽD 3 ..

reactivity in the cerebellum and in the spinal cord. In contrast, at day 23, the majority of the inflammatory infiltrates were surrounded by dense processes of astrocytes which were iNOS-positive ŽFig. 1G, Fig. 2B,D.. Double-staining experiments with GFAP ŽFig. 3C. and anti-iNOS ŽFig. 3D. antibodies confirmed that iNOS immunoreactivity was expressed by almost all the activated astrocytes during the second thrust. 3.1.3. Neurons, ependymal and endothelial cells iNOS immunoreactivity could not be detected in periventricular or ependymal cells. With regard to iNOS immunoreactivity in endothelial cells, we were unable to reach a conclusion because infiltrating cells Žmonocytes, peripheral macrophages., perivascular microglia or migrating parenchymal microglia, were superposed or proximal to cerebral microvessels. We could not identify any iNOS-positive cells that would clearly correspond to neurons. 3.2. Study of iNOS mRNA expression by in situ hybridization iNOS mRNA was not detected by in situ hybridization in control rat CNS ŽTable 3, Fig. 4A.. In contrast, these transcripts were found in those brain regions of EAE-rats where iNOS immunoreactivity was observed ŽFig. 4B..

In most cases, positive cells formed clusters and corresponded to inflammatory foci ŽFig. 4C,D.. The 2 oligonucleotide probes did not discriminate between the macrophagermicroglia iNOS-mRNA and the astrocyte iNOS-mRNA, due to their high sequence homology. However, analysis of adjacent sections in different brain structures established that clusters of cells expressing iNOS mRNA were consistently iNOS-immunoreactive. More isolated positive cells were also observed ŽFig. 4D.. They corresponded to astrocytes as identified with GFAP-staining in adjacent sections. This radiolabelling strongly differed from that of the large positive areas ŽFig. 4C., which evidence the iNOS mRNA of infiltrating cells of the monocytermacrophage lineage. 3.3. Effects of 1,25-D3 on iNOS-immunolabelling and -mRNA expression during EAE In agreement with a previous study w39x, the pattern of immunostaining with the OX42 Mab was not modified in the different areas of the CNS in EAE-animals treated with 1,25-D 3 , with the notable exception of the anterior brain ŽTable 1., especially in the striatum. In this structure, OX42-labelling appeared much less pronounced that in the corresponding structure of untreated EAE-control rats. In contrast, iNOS immunoreactivity was markedly reduced both in cells of the monocytermacrophage lineage

Table 2 Immunohistochemical analysis of iNOS immunoreactivity and GFAP-staining of activated astrocytes

iNOS-D12 GFAP-D12 iNOS-D18 GFAP-D18 iNOS-D23 GFAP-D23 iNOS-D23rD3 GFAP-D23rD3

Anterior brain

Midbrain

Brainstem

Cerebellum

Spinal cord

– – – – q q q q

– – – – q q q q

– – – – qq qq " qq

– – qq qq qq qq " qq

– – " " q q " q

Semi-quantitative counts were carried out in EAE-rats at days 12 and 18. At day 23, data are compared with those from EAE-rats treated with 1,25-D 3 ŽD 3 ..

E. Garcion et al.r Molecular Brain Research 45 (1997) 255–267

263

Fig. 4. Study of iNOS-mRNA expression by in situ hybridization experiments. No iNOS-mRNA was detected in the cerebellum of control rats ŽA.. In the cerebellum of EAE-control rats ŽB., numerous foci expressing iNOS-mRNA were present in the adjacent white matter Žwm. and granule cell layer Žgl., but not in the molecular layer Žml.. During EAE, 2 types of labelling were observed: ŽC. positive dense areas observed during all stages of EAE, which evidence iNOS-mRNA expression by infiltrating inflammatory cells, and ŽD. small clusters Žarrows. of silver grains present only in the later stages of EAE and corresponding to astrocytes identified in the same areas. iNOS-mRNA expression was observed at day 23 in EAE-control rats, in the brainstem ŽE. and in the cerebellum ŽG. in a region proximal to a vessel Žv.. iNOS-mRNA expression had almost disappeared at day 23 in EAE-1,25-D 3-treated rats, in the brainstem ŽF. and in the cerebellum ŽH. around a blood vessel Žv., despite the presence of infiltrating cells which were faintly visible Žarrows.. Bar ŽA–D. s 40 m m; bar ŽE–H. s 20 m m.

E. Garcion et al.r Molecular Brain Research 45 (1997) 255–267

264

Table 3 In situ hybridization analysis of iNOS-mRNA expression in control rats or EAE-rats at days 12 and 18 In situ hybridization computer analysis Ž=10y2 . Control D12 D18 D23 D23rD3

0.6"1.2 151.7"50.0 66.6"31.3 88.3"56.7 9.1"3.0

At day 23, data are compared with those from EAE-rats treated with 1,25-D 3 ŽD 3 .. Data are obtained from autoradiographs corresponding to sections of the brainstem and cerebellum. They are expressed as the mean"S.D. of the percentage of total cerebral area expressing iNOSmRNA.

and in astrocytes in the cerebellum, brainstem and in the spinal cord of 1,25-D 3-treated rats when compared to EAE-control rats ŽTables 1 and 2, Fig. 2G,H.. This effect was not observed in the anterior brain where 1,25-D 3 treatment only slightly reduced or even did not modify the weak intensity of iNOS-immunostaining ŽFig. 2E,F.. Double-staining with anti-iNOS and OX42 antibodies Žnot shown., or with anti-iNOS and GFAP antibodies ŽFig. 3C–F., showed that iNOS immunoreactivity was downregulated by 1,25-D 3 both in cells of the macrophage lineage and in astrocytes. In parallel, in situ hybridization experiments ŽTable 3. revealed that the inhibition of iNOS immunoreactivity was correlated with a striking decrease of iNOS mRNA, both in macrophagesrmicroglia and in astrocytes. At day 23, iNOS mRNA could only very faintly be detected in the brainstem ŽFig. 4F. and in the cerebellum ŽFig. 4H. of treated rats whereas the transcripts remained broadly expressed in the same regions in EAE-control rats ŽFig. 4E,G.. There was also a decrease of iNOS transcripts in the spinal cord of EAE-1,25-D 3-treated animals when compared to EAE-control rats, but the difference was less pronounced than in the brainstem or cerebellum Žnot shown..

4. Discussion The analysis of iNOS immunoreactivity and mRNA in the rat model of chronic relapsing EAE demonstrated a time- and cell-specific expression of this enzyme. Macrophages and activated microglia appear to express iNOS during all stages of EAE, while astrocytes only express iNOS during the 2nd inductive phase of EAE. The fact that iNOS remains expressed in inflammatory cells during the remission phase of EAE suggests that its role is more complex than expected from other studies, such as in the EAE mouse model w46x. A first explanation of simulta-

neous NO synthesis and clinical remission is that the deleterious effect exerted on oligodendrocytes or neurons by the persistent production of NO must be counteracted by other factors that specifically protect these cells. It has been reported that reactive oxygen species, including NO, induce the synthesis of NGF w40x, whose neuroprotective effects are well-established. A second explanation, which is not exclusive, is that NO itself limits the extent of the immune process inside the CNS. This hypothesis is supported by observations that NO can induce apoptosis in both lymphocytes w13x and cells of the monocyte lineage w1,44x, while extensive cell death has been observed during the remission phase of EAE w47x. Alternately, it is possible that NO produced during the remission phase exerts adverse effects which contribute to the induction of the 2nd clinical phase of EAE. It also appears that the synthesis of iNOS during EAE is region-specific and progresses in time in a caudo-rostral direction. Thus, iNOS expression is strongly correlated with the development of inflammatory infiltrates except in the spinal cord at day 23 where iNOS expression appears to decrease in OX42-positive cells, relative to that observed at days 12 and 18. This observation supports the hypothesis that regions of the CNS display different immunological status w39x, notably at the level of the regulation of cytokine expression. It is intriguing that, besides macrophages and activated microglia, astrocytes become able to express iNOS during the 2nd inductive phase of EAE. This observation confirms that astrocytes are probably not the main antigen-presenting cell in the CNS w32x since our data do not support any major role for these cells during the 1st phase of EAE. However, some astrocytes participate afterward in the inflammatory and immune responses. Interestingly, all the hypertrophied astrocytes which were strongly immunoreactive for GFAP were also iNOS-positive. Hence, in addition to GFAP being a phenotypic marker of astrocytes w2,27x, iNOS appears to constitute a functional marker of activated astrocytes during the later stages of EAE as is the case during transient global ischemia w12x. These data provide support to the concept that the gliotic reaction which occurs during rat EAE is not a passive phenomenon. Indeed, since activated astrocytes are a potential source of NO, they may mediate paracrine andror autocrine interactions with the neighbouring cells. Nevertheless, it remains to be determined whether NO is deleterious or beneficial for CNS integrity, since it has been demonstrated that astrocytes can exert immunomodulatory effects on T cells during EAE w32x or MS w36x. The inhibition of iNOS expression by 1,25-D 3 , observed at day 23 post-immunization, is of major interest since it coincides with the clinical improvement observed in EAE-1,25-D 3-treated rats during the 2nd relapse phase of EAE w39x. Both immunohistochemical and in situ hybridization data confirm that a down-regulation of iNOS expression takes place in the 2 major cell types expressing

E. Garcion et al.r Molecular Brain Research 45 (1997) 255–267

iNOS, viz. macrophages and astrocytes. As VDR expression has been demonstrated in both astrocytes w42x and macrophages w4x, it is tempting to speculate that 1,25-D 3 regulation of iNOS is a receptor-mediated process rather than a non-genomic effect. It should be stressed, however, that 1,25-D 3 actions on the iNOS gene may not be direct, if the hormone, for instance, decreases the synthesis of pro-inflammatory cytokines involved in iNOS gene activation w57x. The fact that 1,25-D 3 also reduces the clinical manifestation of EAE suggests that the hormone could attenuate the deleterious effects of cell-derived NO. iNOS regulation by 1,25-D 3 appears reminiscent of that mediated by another ligand of a nuclear hormone receptor, all-trans retinoic acid, which can down-regulate in vitro cytokine-induced iNOS mRNA expression in several cell types, including rat glial cells. Furthermore, retinoic acid can improve the disease course during EAE w48x, like 1,25-D 3 . There is increasing evidence that retinoic acid receptor ŽRAR.rVDR heterodimers mediate most of the effects initially thought to depend on VDR homodimers w52x. Hence, it would be of interest to determine if both RA and 1,25-D 3 might cooperate in the regulation of iNOS gene. Some of the acute and chronic actions of 1,25-D 3 that influence the immune process w30x can now be ascribed to the inhibition of iNOS synthesis. However, regulation of iNOS gene in the CNS contrasts with that observed in osteoblasts, in which 1,25-D 3 appeared to potentiate cytokine-induced NO release w50x. This supports the concept of a tissue-specific action of 1,25-D 3 . In this respect, it is intriguing that 1,25-D 3 appears to exert more pronounced effects in the brainstem and especially in the cerebellum. Such a phenomenon, already observed using other immunological parameters w39x, could result from a higher penetration of 1,25-D 3 through the fenestrate capillaries which are proximal to the 4th ventricle. Alternately, it is possible that glial cells present a region-dependent functional heterogeneity, so that their response to 1,25-D 3 varies from one CNS structure to another. The fact that 1,25-D 3 can be synthesized in vitro by activated microglial cells w43x and macrophages w51x raises the possibility that, in vivo, an endogenous production of the hormone might contribute to maintain brain homeostasis, and confer to some CNS structures their special immunological status. The finding that 1,25-D 3 down-regulates iNOS expression by different cell types, including macrophages, microglia and astrocytes, could have broad clinical implications, since NO and iNOS have been involved in the physiopathology of MS and of a wide variety of acute and chronic neurologic disorders w19,20,34,54x. In addition, it was recently reported that a majority of female patients suffering from MS have a deficit in 25-D 3 w45x, the precursor of 1,25-D 3 . On the basis of the present work, these clinical observations suggest that this deficit may have deleterious consequences on the evolution of the disease. It appears possible, therefore, that 1,25-D 3 or less

265

hypercalcemiant analogues could be valuable tools in the management of MS and other iNOS-associated diseases of the CNS.

Acknowledgements We would like to thank Dr. C. Montero-Menei for fruitful discussion and Dr. D. Djakiew for critical reading of the manuscript. We thank also Professor N. Passutti from the Laboratory of Experimental Surgery, Faculte´ de Medecine de Nantes; L. Sindji, A. Leboterff and P. Trelo´ han for technical assistance; and Dr. S. Carvajal-Gonzalez for computer analysis. This work was supported by INSERM and by a grant from the ‘Ligue Franc¸aise contre la Sclerose en Plaques’. ´

References w1x Albina, J.E., Cui, S., Mateo, R.B. and Reichner, J.S., Nitric oxidemediated apoptosis in murine peritoneal macrophages, J. Immunol., 150 Ž1993. 5080–5085. w2x Aquino, D.A., Chiu, F-C., Brosnan, C.F. and Norton, W.T., Glial fibrillary acidic protein increases in the spinal cord of Lewis rats with acute experimental autoimmune encephalomyelitis, J. Neurochem., 51 Ž1988. 1085–1096. w3x Bagasra, O., Michaels, F.H., Zheng, Y.M., Bobroski, L.E., Spitsin, S.V., Fu, Z.F., Tawadros, R. and Koprowski, H., Activation of the inducible form of nitric oxide synthase in the brains of patients with multiple sclerosis, Proc. Natl. Acad. Sci. USA, 92 Ž1995. 12041– 12045. w4x Bhalla, A.K., Amento, E.P. and Krane, S.M., Differential effects of 1,25-dihydroxyvitamin D 3 on human lymphocytes and monocytermacrophages: inhibition of interleukin-2 and augmentation of interleukin-1 production, Cell Immunol., 98 Ž1986. 311–322. w5x Bidmon, H.J. and Stumpf, W.E., Choroid plexus, ependyma and arachnoidea express receptor for vitamin D: differences between ‘seasonal’ and ‘non seasonal’ breeders, Prog. Brain Res., 91 Ž1992. 279–283. w6x Bo, ¨ L., Dawson, T.M., Wesselingh, S., Mork, S., Choi, S., Kong, P.A., Hanley, D. and Trapp, B.D., Induction of nitric oxide synthase in demyelinating regions of multiple sclerosis brains, Ann. Neurol., 36 Ž1994. 778–786. w7x Branisteanu, D.D., Waer, M., Sobis, H., Marcelis, S., Vandeputte, M. and Bouillon, R., Prevention of murine experimental allergic encephalomyelitis: cooperative effects of cyclosporine and 1 a ,25ŽOH. 2 D 3 , J. Neuroimmunol., 61 Ž1995. 151–160. w8x Cross, A.H., O’Mara, T. and Raine, C.S., Chronologic localization of myelin-reactive cells in the lesions of relapsing EAE: implication for the study of multiple sclerosis, Neurology, 43 Ž1993. 1028–1033. w9x Dawson, T.M. and Snyder, S.H., Gases as biological messengers: nitric oxide and carbon monoxide in the brain, J. Neurosci., 14 Ž1994. 5147–5159. w10x Di Rosa, M., Radomski, M., Carnuccio, R. and Moncada, S., Glucocorticoids inhibit the induction of nitric oxide synthase in macrophages, Biochem. Biophys. Res. Commun., 172 Ž1990. 1246– 1252. w11x Ding, A., Nathan, C.F., Graycar, J., Derynck, R., Stuehr, D. and Srimal, S., Macrophage deactivating factor and transforming growth factors-b 1, -b 2 and -b 3 inhibit induction of macrophage nitrogen oxide synthesis by INF-g , J. Immunol., 145 Ž1990. 940–944. w12x Endoh, M., Maiese, K. and Wagner, J., Expression of the inducible

266

w13x

w14x

w15x

w16x

w17x

w18x

w19x

w20x

w21x

w22x

w23x

w24x

w25x

w26x

w27x

w28x

w29x

w30x

E. Garcion et al.r Molecular Brain Research 45 (1997) 255–267 form of nitric oxide synthase by reactive astrocytes after transient global ischemia, Brain Res., 651 Ž1994. 92–100. Fehsel, K., Kroncke, K.-D., Meyer, K.L., Huber, H., Wahn, V. and ¨ Kolb-Bochofen, V., Nitric oxide induces apoptosis in mouse thymocytes, J. Immunol., 155 Ž1995. 2858–2865. Feurer, C., Prentice, D.E. and Cammisuli, S., Chronic relapsing encephalomyelitis in the Lewis rat, J. Neuroimmunol., 10 Ž1985. 159–165. Fournier, C., Gepner, P., Sadouk, M.B. and Charriere, ` J., In vivo beneficial effects of cyclosporin A and 1,25-dihydroxyvitamin D 3 on the induction of experimental autoimmune thyroitidis, Clin. Immunol. Immunopathol., 54 Ž1990. 53–63. Galea, E., Reis, D.J. and Feinstein, D.L., Cloning and expression of nitric oxide synthase from rat astrocytes, J. Neurosci. Res., 37 Ž1994. 406–414. Green, S. and Chambon, P., Nuclear receptor enhance our understanding of transcription regulation, Trends Genet., 4 Ž1988. 309– 313. Hirokawa, K., O’Shaughnessy, K.M., Ramrakha, P. and Wilkins, M.R., Inhibition of nitric oxide synthesis in vascular smooth muscle by retinoids, Br. J. Pharmacol., 113 Ž1994. 1448–1454. Iadecola, C., Zhang, F. and Xu, X., Inhibition of inducible nitric oxide synthase ameliorates cerebral ischemic damage, Am. J. Physiol., 268 Ž1995. R286–R292. Koka, P., He, K., Zack, J.A., Kitchen, S., Peacock, W., Fried, I., Tran, T., Yashar, S.S. and Merrill, J.E., Human immunodeficiency virus 1 envelop proteins induce interleukin 1, tumor necrosis factor a , and nitric oxide in glial cultures derived from fetal, neonatal, and adult human brain, J. Exp. Med., 182 Ž1995. 941–952. Koprowski, H., Zheng, Y.M., Heber-Katz, E., Fraser, N., Rorke, L., Fu, Z.F., Hanlon, C. and Dietzschold, B., In vivo expression of inducible nitric oxide synthase in experimentally induced neurologic diseases, Proc. Natl. Acad. Sci. USA, 90 Ž1993. 3024–3027. Koren, R., Liberman, U.A., Maron, L., Novogrodsky, A. and Ravid, A., 1,25-Dihydroxyvitamin D 3 acts directly on human lymphocytes and interferes with the cellular response to interleukin-2, Immunopharmacology, 18 Ž1989. 187–194. Lemire, J.M. and Archer, C.D., 1,25-Dihydroxyvitamin D 3 prevents the in vivo induction of murine experimental allergic encephalomyelitis, J. Clin. InÕest., 87 Ž1990. 1103–1107. Lemire, J.M., Adams, J.S., Sakai, R. and Jordan, S.C., 1 a ,25-Dihydroxyvitamin D3 suppresses proliferation and immunoglobulin production by normal human peripheral blood mononuclear cells, J. Clin. InÕest., 74 Ž1984. 657–661. Lin, R.F., Lin, T.S., Tilton, R.G. and Cross, A.H., Nitric oxide localized to spinal cords of mice with experimental allergic encephalomyelitis: an electron paramagnetic resonance study, J. Exp. Med., 178 Ž1993. 643–648. Lowenstein, C.J., Glatt, C.S., Bredt, D.S. and Snyder, S.H., Cloned and expressed macrophage nitric oxide synthase contrasts with the brain enzyme, Proc. Natl. Acad. Sci. USA, 89 Ž1992. 6711–6715. Ludowyk, P.A., Hughes, W., Hugh, A., Willenborg, D.O., Rockett, K.A. and Parish, C.R., Astrocytic hypertrophy: an important pathological feature of chronic experimental autoimmune encephalitis in aged rats, J. Neuroimmunol., 48 Ž1993. 121–134. Lyons, C.R., Orloff, G.J. and Cunningham, J.M., Molecular cloning and functional expression of an inducible nitric oxide synthase from a murine macrophage cell line, J. Biol. Chem., 267 Ž1992. 6370– 6374. MacMicking, J.D., Willenborg, D.O., Weidemann, M.J., Rockett, K.A. and Cowden, W.B., Elevated secretion of reactive nitrogen and oxygen intermediates by inflammatory leukocytes in hyperacute experimental autoimmune encephalomyelitis: enhancement by the soluble products of encephalitogenic T cells, J. Exp. Med., 176 Ž1992. 303–307. Manolagas, S.C., Hustmyer, F.G. and Yu, X.-P., 1,25-Dihydroxy-

w31x

w32x

w33x

w34x

w35x

w36x

w37x

w38x

w39x

w40x

w41x

w42x

w43x

w44x

w45x

w46x

w47x

vitamin D 3 and the immune system, Proc. Soc. Exp. Biol. Med., 191 Ž1989. 238–245. Mathieu, C., Waer, M., Casteels, K., Laureys, J. and Bouillon, R., Prevention of type 1 diabetes in NOD mice by non hypercalcemic doses of a new structural analog of 1,25-dihydroxyvitamin D 3 , KH1060, Endocrinology, 136 Ž1995. 866–872. Matsumoto, Y., Ohmori, K. and Fujiwara, M., Microglial and astroglial reactions to inflammatory lesions of experimental autoimmune encephalomyelitis in the rat central nervous system, J. Neuroimmunol., 37 Ž1992. 23–33. McKee, M., Scavone, C. and Nathanson, J.A., Nitric oxide, cGMP, and hormone regulation of active sodium transport, Proc. Natl. Acad. Sci. USA, 91 Ž1994. 12056–12060. Meda, L., Cassatella, M.A., Szendrei, G.I., Otvos Jr., L., Baron, P., Villalba, M., Ferrari, D. and Rossi, F., Activation of microglial cells by b-amyloid protein and interferon-g , Nature (London), 374 Ž1995. 647–650. Mehta, K., McQueen, T., Tucker, S., Pandita, R. and Aggarwal, B.B., Inhibition by all-trans-retinoic acid of tumor necrosis factor and nitric oxide production by peritoneal macrophages, J. Leukocyte Biol., 55 Ž1994. 336–342. Meinl, E., Aloisi, F., Ertl, B., Weber, F., De Waal Malefyt, R., Wekerle, H. and Hohfeld, R., Multiple sclerosis: immunomodulatory effects of human astrocytes on T cells, Brain, 117 Ž1994. 1323–1332. Merrill, J.E., Ignarro, L.J., Sherman, P., Melinek, J. and Lane, T.H., Microglial cell cytotoxicity of oligodendrocytes is mediated through nitric oxide, J. Immunol., 151 Ž1993. 2132–2141. Montero-Menei, C., Sindji, L., Garcion, E., Mege, M., Couez, D., Gamelin, E. and Darcy, F., Early events of the inflammatory reaction induced in rat brain by lipopolysaccharide intracerebral injection: relative contribution of peripheral monocytes and activated microglia, Brain Res., 724 Ž1996. 55–66. Nataf, S., Garcion, E., Darcy, F., Chabannes, D., Muller, J.-Y. and Brachet, P., 1,25-Dihydroxyvitamin D 3 exerts regional effect in the central nervous system during experimental allergic encephalomyelitis, J. Neuropathol. Exp. Neurol., 55 Ž1996. 904–914. Naveilhan, P., Neveu, I., Jehan, F., Baudet, C., Wion, D. and Brachet, P., Reactive oxygen species influence nerve growth factor synthesis in primary rat astrocytes, J. Neurochem., 62 Ž1994. 2178– 2186. Neveu, I., Naveilhan, P., Baudet, C., Brachet, P. and Metsis, M., 1,25–Dihydroxyvitamin D 3 regulates NT-3, NT-4 but not BDNF mRNA in astrocytes, NeuroReport, 6 Ž1994. 124–126. Neveu, I., Naveilhan, P., Jehan, F., Baudet, C., Wion, D., DeLuca, H.F. and Brachet, P., 1,25-Dihydroxyvitamin D 3 regulates the synthesis of nerve growth factor in primary culture of glial cells, Mol. Brain Res., 24 Ž1994. 70–76. Neveu, I., Naveilhan, P., Menei, C., Wion, D., Brachet, P. and Garabedian, M., Synthesis of 1,25-dihydroxyvitamin D 3 by rat brain ´ macrophages in vitro, J. Neurosci. Res., 38 Ž1994. 214–220. Nguyen, K.B., McCombe, P.A. and Pender, M.P., Macrophage apoptosis in the central nervous system in experimental autoimmune encephalomyelitis, J. Autimmun., 7 Ž1994. 145–152. Nieves, J., Cosman, F., Herbert, J., Shen, V. and Lindsay, R., High prevalence of vitamin D deficiency and reduced bone mass in multiple sclerosis, Neurology, 44 Ž1994. 1687–1692. Okuda, Y., Nakatsuji, Y., Fujimura, H., Esumi, H., Ogura, T., Yanagihara, T. and Sakoda, S., Expression of the inducible isoform of nitric oxide synthase in the central nervous system of mice correlates with the severity of actively induced experimental allergic encephalomyelitis, J. Neuroimmunol., 62 Ž1995. 103–112. Pender, M.P., McCombe, P.A., Yoong, G. and Nguyen, K.B., Apoptosis of alpha-beta T lymphocytes in the nervous system in experimental autoimmune encephalomyelitis: its possible implications for recovery and acquired tolerance, J. Autoimmun., 5 Ž1992. 401–410.

E. Garcion et al.r Molecular Brain Research 45 (1997) 255–267 w48x Racke, M.K., Burnett, D., Pak, S.-H., Albert, P.S., Cannella, B., Raine, C.S., McFarlin, D.E. and Scott, D.E., Retinoid treatment of experimental allergic encephalomyelitis: IL-4 production correlates with improved disease course, J. Immunol., 154 Ž1995. 450–458. w49x Radomski, M.W., Palmer, R.M.J. and Moncada, S., Glucocorticoids inhibits the expression of an inducible, but not the constitutive, nitric oxide synthase in vascular endothelial cells, Proc. Natl. Acad. Sci. USA, 87 Ž1990. 10043–10047. w50x Riancho, J.A., Zarrabeitia, M.T., Fernandez-Luna, J.L. and Gonzalez-Macias, J., Mechanisms controlling nitric oxide synthesis in osteoblasts, Mol. Cell. Endocrinol., 107 Ž1995. 87–92. w51x Reichel, H., Koeffler, H.P. and Norman, A.W., Synthesis in vitro of 1,25-dihydroxyvitamin D 3 and 24,25-dihydroxyvitamin D 3 by interferon-g-stimulated normal human bone marrow and alveolar macrophages, J. Biol. Chem., 262 Ž1987. 10931–10937. w52x Schrader, M., Muller, K.M. and Carlberg, C., Specificity and flexi¨ ¨ bility of vitamin D signalling, J. Biol. Chem., 269 Ž1994. 5501–5504. w53x Simmons, M.L. and Murphy, S., Cytokines regulate L-arginine-dependent cyclic GMP production in rat glial cells, Eur. J. Neurosci., 5 Ž1993. 825–831. w54x Sparrow, J.R., Inducible nitric oxide synthase in the central nervous system, J. Mol. Neurosci., 5 Ž1995. 219–229.

267

w55x Stumpf, W.E. and O’Brien, L.P., 1,25ŽOH. 2 vitamin D 3 sites of action in the brain: an autoradiographic study, Histochemistry, 87 Ž1987. 393–406. w56x Sutherland, M.K., Somerville, M.J., Yoong, L.K.K., Bergeron, C., Haussler, M.R. and Crapper McLachlan, D.R., Reduction of vitamin D hormone receptor mRNA levels in Alzheimer as compared to Huntington hippocampus: correlation with calbindin-28k mRNA levels, Mol. Brain Res., 13 Ž1992. 239–250. w57x Szabo, C. and Thiemermann, C., Regulation of the expression of the inducible isoform of nitric oxide synthase, AdÕ. Pharmacol., 34 Ž1995. 113–153. w58x Van Dam, A.-M., Bauer, J., Man-A-King, W.K.H., Marquette, C., Tilders, F.J.H. and Berkenbosch, F., Appearance of inducible nitric oxide synthase in the rat central nervous system after rabies virus infection and during experimental allergic encephalomyelitis but not after peripheral administration of endotoxin, J. Neurosci. Res., 40 Ž1995. 251–260. w59x Zielasek, J., Jung, S., Gold, R., Liew, F.Y., Toyka, K.V. and Hartung, H.P., Administration of nitric oxide synthase inhibitors in autoimmune neuritis and experimental autoimmune encephalomyelitis, J. Neuroimmunol., 58 Ž1995. 81–88.