- Email: [email protected]
Linomide-induced suppression of experimental autoimmune neuritis is associated with down-regulated macrophage functions
Journal of Neuroimmunology 76 Ž1997. 177–184
Linomide-induced suppression of experimental autoimmune neuritis is associated with down-regulated macrophage functions a,)
Xue-Feng Bai a
, Fu-Dong Shi a , Jie Zhu a , Bao-Guo Xiao a , Gunnar Hedlund b, Hans Link
a
DiÕision of Neurology, Karolinska Institute, Huddinge Hospital, S-141 86 Huddinge, Stockholm, Sweden b Pharmacia and Upjohn, Lund Research Centre, Lund, Sweden Received 4 September 1996; revised 3 December 1996; accepted 3 February 1997
Abstract Experimental autoimmune neuritis ŽEAN. is a T-cell mediated autoimmune disease of the peripheral nervous system, in which macrophages and T-cells feature prominently in nerve lesions. EAN represents a counterpart to Guillain-Barre´ syndrome in humans. In the present study, we investigated the in vitro and in vivo effects of Linomide ŽLS-2616, quinoline-3-carboxamide., a synthetic immunomodulatory compound, on macrophages in relation to EAN. Linomide strongly suppressed IFN-g and lipopolysaccharide ŽLPS.-induced IL-1 b , TNF-a and IL-6 mRNA expression in macrophages in vitro as demonstrated by in situ hybridisation. Linomide administered daily subcutaneously from the day of inoculation completely prevented the development of clinical symptoms of EAN. Linomide administered from day 9 post immunisation Žp.i.. significantly suppressed clinical EAN. Macrophages from Linomide-treated EAN rats showed decreased IL-1 b , TNF-a and IL-6 mRNA expression in response to IFN-g and LPS. LPS-induced nitric oxide production by macrophages was also suppressed by Linomide in vitro. Linomide, however, does not affect macrophage death and release of lactate dehydrogenase. We conclude that Linomide may exert its actions in EAN and perhaps also in other autoimmune disease models, by suppressing macrophage functions. Keywords: Experimental autoimmune neuritis; Linomide; Guillain-Barre´ syndrome; Cytokine; Macrophage
1. Introduction Linomide ŽLS-2616, quinoline-3-carboxamide., a synthetic immunomodulatory compound, is effective in inhibiting manifestations of acute and chronic-relapsing experimental autoimmune encephalomyelitis ŽEAE. ŽKarussis et al., 1993a,b. and myasthenia gravis ŽEAMG. ŽKarussis et al., 1994; Zhang et al., 1996., lupus-like disease in MRLr1 mice ŽTarkowski et al., 1986., insulin-dependent diabetes mellitus and autoimmune insulitis in NOD mice ŽGross et al., 1994. and collagen-induced arthritis in DBA mice ŽBjork and Kleinau, 1989.. The mode of action of Linomide in the modulation of autoimmune diseases is not clear. Suppressed T- and B-cell responses to specific antigens ŽKarussis et al., 1993a, 1994., augmented NK cell activity ŽKalland et al., 1985; Kalland, 1990. as well as suppressed Th1 responses ŽArad et al., 1996. have been suggested.
)
Corresponding author. Tel.: q46-8-58582279; fax: q46-8-7744822.
Experimental autoimmune neuritis ŽEAN. is a T-cell mediated autoimmune disease of the peripheral nervous system ŽPNS.. In many respects, EAN represents an experimental counterpart to acute inflammatory demyelinating polyradiculoneuropathy or Guillain-Barre´ syndrome ŽGBS. in humans. Macrophages act as professional antigen presenting cells and are of pivotal importance in the amplification and effector phase of immune-mediated demyelination. Macrophages feature prominently in the nerve lesions of EAN and GBS ŽHartung et al., 1995.. In the effector phase of EAN, macrophages exert their functions by phagocytosis of myelin components and release of proinflammatory cytokines such as IL-1, TNF-a and IL-6 and other highly active mediators that act at short range, like nitric oxide ŽNO. ŽZielasek et al., 1995.. Since macrophages are essential in the initiation and amplification of the autoimmune responses in EAN and Linomide may influence macrophage functions ŽVukanovic and Isaacs, 1995., we investigated the effects of Linomide on macrophages in relation to EAN. We found that Linomide strongly downregulated IL-1 b , TNF-a and IL-6
0165-5728r97r$17.00 Copyright q 1997 Elsevier Science B.V. All rights reserved. PII S 0 1 6 5 - 5 7 2 8 Ž 9 7 . 0 0 0 5 1 - 9
178
X.-F. Bai et al.r Journal of Neuroimmunology 76 (1997) 177–184
mRNA expression in macrophages both in vitro and in vivo, and down-regulated NO production by macrophages.
2. Materials and methods 2.1. Antigens and immunoreagents Bovine peripheral nerve myelin ŽBPM. was prepared from lumbosacral plexus ŽNorton and Poduslo, 1973.. Nicotinic acetylcholine receptor ŽAChR. was purified as described ŽLindstrom ¨ et al., 1981. from electroplax tissue of Torpedo californica ŽPacific Bio-Marine, Venice, CA.. Lipopolysaccharide ŽLPS. from Escherichia coli was provided by the Department of Microbiology, Huddinge Hospital, Karolinska Institute. Rat rIFN-g was provided by Dr. P.H. van der Meide ŽBiomedical Primate Centre, Rijswijk, The Netherlands. ŽDijkema et al., 1986.. mAbs to rat MHC class II, ICAM-1 and macrophages ŽED1. were purchased from Serotec ŽOxford, UK.. 2.2. Induction of EAN and Linomide treatment Male Lewis rats, 150–200 g, locally bred, originally obtained from Harlan CPB, Zeist, The Netherlands, were injected in both hind foot pads with 200 m l of inoculum containing 4 mg of BPM, 2 mg Mycobacterium Ž M.. tuberculosis Žstrain H37RA; Difco, Detroit, MI. in 100 m l saline and 100 m l Freund’s incomplete adjuvant ŽFIA. ŽDifco.. Symptoms of paresis were graded as follows: 0, no illness; 1, flaccid tail; 2, moderate paraparesis; 3, severe paraparesis; 4, tetraparesis. Linomide ŽPharmacia-Upjohn, Lund. was dissolved in distilled water and administered at a single dose of 160 mgrkgrday subcutaneously. 20 Lewis rats were immunised with BPM q complete Freund’s adjuvant ŽCFA.. 10 of the 20 rats were given Linomide from the day of immunisation until sacrifice or the end of experiments. The other 10 rats served as controls and were injected with PBS ŽpH 7.7. only. In both the Linomidetreated and control groups, 5 rats were used to isolate macrophages from the abdominal cavity at day 14 p.i. To evaluate Linomide effects on the efferent phase of EAN, another group of 8 rats were immunised with BPM q CFA. 4r8 rats were given Linomide at a dose of 160 mgrkg subcutaneously daily from day 9 p.i., i.e. just before onset of clinical signs, to the end of the experiment. The remaining 4 rats served as controls and were injected with PBS ŽpH 7.7. daily until the end of the experiment. 2.3. Isolation of macrophages, in Õitro culture and preparation of cells for in situ hybridisation Groups of 5 rats, including rats immunised with BPM plus CFA, rats immunised with CFA only and normal adult rats were injected intraperitoneally 4 days before sacrifice with 2 ml of 3% thioglycollate ŽSigma, St. Louis, MO. as described by Kraemer et al. Ž1985.. Four days
after injection, the peritoneal cavities were rinsed for 3 min with complete Iscove’s medium ŽFlow Lab., Irvine. supplemented with 1% Žvrv. minimum essential medium ŽFlow., 50 IUrml penicillin, 60 mgrml streptomycin ŽGibco, Paisley, UK., 2 mM glutamin ŽFlow. and 3% normal human ABq serum. After three washes in the same medium, the cell number of the suspension was adjusted to 1 = 10 6 viable cellsrml medium. The percentage of cells excluding trypan blue generally exceeded 90%. The cells were further cultured in 200 ml tissue culture flasks at 378C for 1 h. The adherent cells were collected. The purity was ) 95% as determined by flow cytometry using the PE-conjugated mAb ED1 ŽSerotec.. Purified macrophages were cultured in the medium at a concentration of 0.5 = 10 6rml in the presence of 0.5 m grml LPS or 50 IUrml of rat rIFN-g . A culture time of 24 and 48 h was selected for evaluation of cytokine mRNA expression. After washing in PBS, pH 7.4, 10 m l containing 10 4 macrophages were dried onto restricted areas of electrically charged glass slides ŽProbeOn slides; Fisher Scientific, Pittsburgh, PA. and stored in sealed boxes at y708C until hybridisation. 2.4. In situ hybridisation (ISH) ISH was performed according to Zhu et al. Ž1996. using S-labeled synthetic oligonucleotide probes ŽScandinavian Gene Synthesis AB, Koping, Sweden.. For each cytokine ¨ ŽXiao et al., 1996., a mixture of four different approximately 48-bp oligonucleotide probes were used in order to increase the sensitivity of the method. The oligonucleotide sequences were obtained from GeneBank using the MacVector system. Control slides were hybridised with the same total amount of a sense probe with nucleotide sequence for exon 4 of rat IFN-g . A constant ratio of the guaninercytosine content of approximately 60% was employed. The oligonucleotide probes were checked for absence of palindromes and long sequences of homology within the species against available GenBank data. The labelling was performed with 35 S deoxyadenosine 5Xthiotriphosphate ŽNew England Nuclear, Cambridge, MA. with terminal deoxynucleotidyl transferase ŽAmersham, Little Chalfont, UK.. After emulsion autoradiography, slides were coded and cells expressing numerous grains over their cytoplasm were counted by darkfield microscopy at 20 = magnification as previously described and validated ŽZhu et al., 1996.. The results are expressed as number of labelled cells per 10 4 macrophages. The control probe used in parallel with the cytokine probes on slides produced a uniformly weak background signal without revealing any positive cells. 35
2.5. Nitrite assay Because secreted NO quickly reacts with oxygen-yielding nitrite, the level of nitrite as a reflection of NO production in macrophage culture supernatants was measured using Griess reagent Ž1% sulfanilamide and 0.1%
X.-F. Bai et al.r Journal of Neuroimmunology 76 (1997) 177–184
N-Ž1-naphthyl. ethylenediamine dihydrochloride in 5% H 3 PO4 .. 100 m l supernatants from cultured cells were mixed with an equal volume of Griess reagent. After a 10 min reaction at room temperature, the absorbance at 540 nm was measured using an automatic plate reader. Nitrite concentration was determined by comparison with a sodium nitrite standard curve in culture medium.
179
2.7. Apoptosis assay For detection and quantification of macrophage apoptosis, an in situ cell death detection kit ŽFluorescein, Boehringer-Mannheim, Germany. was used according to the manufacturer’s instructions. Cell smear samples were used and slides were observed under fluorescence microscope.
2.6. Immunohistochemistry 2.8. LDH assay The segments of sciatic nerve close to the lumbar spinal cord from 4 control EAN rats and 4 EAN rats receiving Linomide treatment from the day of immunisation were dissected and snap-frozen in liquid nitrogen on day 14 p.i. Cryopreserved sciatic nerve tissue sections Ž10 m m. were first fixed in acetone at y208C. Non-specific binding was blocked with normal horse serum ŽVector, Burlingame, CA. for 30 min. Macrophage and T-lymphocyte infiltrates were characterised in serial sections by incubation with 5 m grml of the monoclonal antibodies ŽmAb. ED1 and W3r13 ŽSerotec, Oxford, UK., respectively. Biotinylated horse anti-mouse antibody ŽVector. was used to couple the primary mAbs with an avidin–biotin–peroxidase complex ŽVector.. After washing in PBS, substrate amino-ethyl carbazole ŽAEC. was applied to the tissue and incubated for 5 min. The slides were rinsed in tap water, counterstained with hematoxylin and mounted with Aquamount ŽLerner Lab., New Haven, CT.. Controls included the omission of the primary antibody and second antibody. Negligible background staining was observed.
To measure LDH activity in supernatants of macrophage cultures at 308C, a diagnostic kit ŽSigma. was used according to the manufacturer’s instructions. One U of LDH activity is defined as the amount of enzyme that will catalyze the formation of 1.0 mmol of reduced nicotinamide adenine dinucleotide per min. 2.9. Statistics The Student’s t-test was used. All significance tests were two-sided and rejection limit was set to a s 0.05. 3. Results 3.1. Linomide suppresses IL-1b , TNF-a and IL-6 MRNA expression in macrophages Upon culture, only low levels Ž- 10 per 10 4 . of macrophages were detected that expressed TNF-a or IL-6
Fig. 1. LPS-induced IL-1 b mRNA expressing macrophages seen at light microscopy= 400 magnification; arrows indicate examples of IL-1 b mRNA positive cells. Macrophages are from EAN rats on day 14 p.i.
180
X.-F. Bai et al.r Journal of Neuroimmunology 76 (1997) 177–184
spontaneously. Numbers of IL-1 b mRNA expressing cells were higher Ž50–100r10 4 macrophages.. After the culture of macrophages in the presence of rIFN-g or LPS, the number of cells expressing IL-1 b ŽFig. 1., TNF-a or IL-6 mRNA were upregulated. Levels of cytokine mRNA expressing macrophages did not differ upon incubation with rIFN-g during observation periods of 24 h versus 48 h ŽFig. 2.. The strongest effects of LPS on levels of cytokine mRNA expressing cells were observed at 24 h. Numbers had declined at 48 h ŽFig. 3.. In the presence of different concentrations Ž10y4 , 10y6 and 10y8 M. of Linomide to culture medium, the median and high doses of Linomide strongly suppressed IFN-g as well as LPS-induced IL-1 b and TNF-a and IFN-g-induced IL-6 mRNA expression in macrophages at both time points ŽFigs. 2 and 3.. LPS-induced IL-6 mRNA expression in macrophages was suppressed by the high dose of Linomide at 48 h. No suppressive effects were observed in cultures treated with low concentration of Linomide during the total observation period. Macrophages from EAN rats Žshown in Figs. 2 and 3., CFA immunized rats and normal rats showed similar results. 3.2. Linomide suppresses LPS-induced NO production by macrophages in Õitro Upon in vitro culture for 24 and 48 h, macrophages spontaneously produce NO. However, the concentration of
this spontaneous NO production was low. Upon in vitro culture in the presence especially of LPS but also of IFN-g , high concentrations of NO were detected in the culture supernatants. The increase of NO was time and dose dependent. Higher doses of LPS induced higher concentrations of NO. Furthermore, higher concentrations of NO were detected at 48 h than 24 h ŽFig. 4.. In the presence of median Ž10y6 M. and high Ž10y4 M. concentrations of Linomide, LPS-induced NO production by macrophages was significantly suppressed at 48 h ŽFig. 4.. Spontaneous and IFN-g induced NO production was also reduced, without reaching statistical significance. This might be due to the lower basal NO levels in the culture supernatants. Macrophages were obtained from normal Lewis rats, rats with EAN Žshown in Fig. 4. and rats immunised with CFA and the results were similar. 3.3. In ÕiÕo effects of Linomide The effects of Linomide on clinical EAN were evaluated. To evaluate Linomide effects on the afferent phase of EAN, a group of 20 rats was used and Linomide Ž160 mgrkgrday, subcutaneously. was given to 10r20 rats from day 0 p.i. until sacrifice of rats on day 14 p.i. or up to the end of experiments Žday 28 p.i... The other 10 rats served as controls and received PBS ŽpH 7.7. only. All control rats immunised with BPM q CFA developed severe EAN by day 14 p.i. In contrast, none of the rats
Fig. 2. Linomide suppresses rIFN-g induced mRNA expression in macrophages collected from EAN rats in vitro. 0.5 = 10 6 macrophages per ml were cultured in the presence of 50 IUrml of rIFN-g . Three different concentrations of Linomide were added to separate cultures for 24 h or 48 h. After washing, 10 m l containing 10 4 macrophages were dried onto ProbOn slides and hybridized with cytokine probes. 10y6 and 10y4 M of Linomide suppressed IL-1 b , TNF-a and IL-6 mRNA expression in macrophages. Bars represent mean values of separate macrophage cultures from 5 different EAN rats. ) P - 0.05; ) ) P - 0.01; ) ) ) P - 0.001.
X.-F. Bai et al.r Journal of Neuroimmunology 76 (1997) 177–184
181
Fig. 3. Linomide suppresses LPS-induced cytokine mRNA expression in macrophages in vitro. The same experimental conditions were used as described in Fig. 2, except that IFN-g was displaced by 0.5 m grml of LPS. Macrophages are purified from 5 EAN rats.
receiving Linomide injection from the day of immunisation developed EAN ŽFig. 5a.. On examination of sciatic nerve tissue sections for the presence of macrophage and T-lymphocyte infiltration on day 14 p.i., 4r4 control rats showed extensive macrophage and T-lymphocyte infiltration. In contrast, none of the 4 Linomide-treated rats showed any macrophage or T-lymphocyte infiltrations.
Fig. 4. Linomide suppresses LPS-induced NO production by macrophages. Five EAN rats were sacrificed at day 14 p.i. and macrophages were isolated as described in Section 2. 0.5=10 6 macrophages per ml were cultured in the presence of 0.5 m grml of LPS with or without the presence of 10y6 M of Linomide. Nitrite concentrations were measured in the culture supernatant. A significant suppression of nitrite production was observed at 48 h. Bars represent mean values of macrophage cultures from 5 EAN rats. ) P - 0.05.
Furthermore, Linomide significantly reduced or even eliminated inflammation at the immunisation sites Žhind footpads. in rats receiving Linomide from the day of immunisation compared to control rats. To evaluate Linomide effects on the efferent phase of EAN, another group of 8 rats were immunised with BPM q CFA, followed in 4r8 rats by injection of the same dose of Linomide at day 9 p.i., i.e. just before onset of clinical signs. The other 4 rats served as controls and were injected with PBS ŽpH 7.7. only. As shown in Fig. 5b, Linomide significantly suppressed EAN severity from day 12 to day 24 p.i. Ž P 0.01.. Macrophages were purified from both control rats and rats receiving Linomide from the day of immunisation. Cells were then challenged in vitro with IFN-g or LPS. Macrophages from Linomide-treated EAN rats showed lower numbers of IFN-g and LPS-induced IL-1 b and TNF-a mRNA expression compared to macrophages from control rats ŽFig. 6.. Borderline reduced IL-6 mRNA expression in macrophages from Linomide-treated EAN rats was also observed. 3.4. Linomide does not affect macrophages in expressing MHC class II, ICAM-1 and surÕiÕing in Õitro None of the three Linomide doses used had any effect on spontaneous, IFN-g or LPS-induced MHC class II or ICAM-1 surface expression on macrophages at 24 h and 48 h Ždata not shown.. To study induction of apoptotic
182
X.-F. Bai et al.r Journal of Neuroimmunology 76 (1997) 177–184
Fig. 5. Linomide administration in vivo suppresses clinical EAN. Ža. Linomide Ž160 mgrkgrday. administered subcutaneously from the day of immunisation to the end of experiments prevented development of EAN. Žb. Linomide Ž160 mgrkgrday. administered from day 9 p.i. to the end of experiments significantly Ž P - 0.01. suppressed severity of EAN from day 12 p.i. to day 24 p.i. Bars represent mean values from 4–5 rats at each time point.
macrophage death, different concentrations of Linomide were added to culture medium. While presence of LPS or IFN-g resulted in increased numbers of apoptotic cells,
addition of Linomide had no effect on spontaneous or IFN-g or LPS induced macrophage death Ždata not shown.. LDH release by macrophages was also investigated.
Fig. 6. In vivo administration of 160 mgrkgrday of Linomide subcutaneously suppresses IFN-g Ž50 IUrml; a–c. and LPS Ž0.5 m grml; d–f. induced mRNA expression of IL-1 b , TNF-a and IL-6 in macrophages. Five EAN rats receiving Linomide and 5 control EAN rats receiving PBS daily from the day of immunisation were sacrificed at day 14 p.i. and macrophages were isolated as described. The same experimental conditions were set as described in Figs. 2 and 3. Bars represent mean values for 5 rats. B, control EAN rats treated with PBS; I, EAN rats treated with Linomide. ) P - 0.05; )) P - 0.01; ) ) ) P - 0.001.
X.-F. Bai et al.r Journal of Neuroimmunology 76 (1997) 177–184
Fig. 7. Linomide does not affect spontaneous, LPS or IFN-g induced LDH release by macrophages. Macrophages were taken from 5 EAN rats sacrificed at day 14 p.i. and pooled. 200 m l of 0.5=10 6 macrophages per ml were cultured in the absence or presence of 0.5 m grml LPS or 50 IUrml of rIFN-g . Culture supernatants were then examined for the presence of LDH. Linomide had no pronounced effects on spontaneous, LPS or IFN-g induced LDH release. The results were confirmed by 2 repeated experiments.
Spontaneous release of LDH by macrophages was low upon in vitro culture in medium after 48 h. LPS but not IFN-g upregulated LDH release. Addition of Linomide at three different doses had no pronounced effect on spontaneous, IFN-g or LPS induced LDH release by macrophages ŽFig. 7..
4. Discussion We observed that Linomide suppressed IFN-g and LPS mediated induction of IL-1 b , TNF-a and IL-6 mRNA expression in macrophages in vitro. Linomide administered during the induction phase of EAN completely prevented the development of clinical EAN in Lewis rats. A significant suppression of EAN was also achieved when Linomide was given after day 9 p.i., i.e. the effector phase of EAN. Levels of LPS and IFN-g induced IL-1 b and TNF-a mRNA expression in macrophages from Linomide-treated EAN rats were lower than in macrophages from control EAN rats. LPS-induced NO production by macrophages was also suppressed by Linomide in vitro. In contrast, macrophage death and LDH release were not affected by Linomide. The effects observed in this study are consistent with previous observations. Vukanovic and Isaacs Ž1995. observed that monocytesrmacrophages prepared from Linomide-treated rats secreted low amounts of TNF-a when challenged with LPS in vitro. Furthermore, when rats were treated with Linomide and then challenged with LPS in vivo, the resulting elevation in serum TNF-a was inhibited by approximately 50%. Septic shock in mice induced by stimulation of macrophages with LPS is also prevented by daily treatment with 100 mgrkg of Linomide, an effect
183
that is due to inhibited TNF-a secretion ŽGonzalo et al., 1993.. The complete prevention of EAN by injection of Linomide in the induction phase indicates that systemic administration of Linomide mainly targets and interferes with trafficking of macrophages and lymphocytes into the nerve tissue. This is supported by the absence of inflammatory cell infiltrations in the target tissue. Systemic suppression of macrophage functions, reflected by the reduced capacity of macrophages in expressing IL-1 b , TNF-a and IL-6 mRNA, may interfere with further presentation of myelin antigens at the site of immunisation and ameliorate EAN by interrupting the supply of activated T-cells. How Linomide affects antigen presentation by macrophage and further T-cell activation is unclear. We postulate that, at the immunisation sites, Linomide suppresses adjuvant-induced constitutive macrophage activation. These macrophages will reduce the release of proinflammatory cytokines like TNF-a , IL-1 b and IL-6, which in turn will reduce macrophage accumulation and activation, lead to a reduced number of activated macrophages migrating to the draining lymph nodes and finally result in insufficient antigen presentation. In the present study, in Linomide-treated rats, the absence of inflammation in the immunisation sites indicates absence of macrophage activation. Furthermore, in Linomide-treated rats, macrophages migrating to draining lymph nodes lack the capacity to activate naive T-lymphocytes. Linomide reduced the capacity of macrophages to produce IL-1 b and IL-6, both of which are important for stimulation of naive T lymphocytes and synergize in the induction of T-cell activation ŽPlebanski et al., 1992; Holsti et al., 1994; Henttinen et al., 1995.. The therapeutic effects of Linomide on clinical signs of actively induced EAN suggest that Linomide may not only reduce the number of T-cells and macrophages invading the nerve but also inhibit the function of activated macrophagesrT-lymphocytes, which have already invaded peripheral nerves and impair nerve conduction or cause myelin damage through the release of toxic mediators such as TNF-a ŽSaid and Hontebeyrie-Joskowicz, 1992; Stoll et al., 1993. and NO ŽZielasek et al., 1995.. It is of note that Linomide suppressed the release of nitrogen intermediates as well as TNF-a mRNA expression in macrophages in the present study. Another possibility that deserves consideration is that macrophages are constitutively present in sciatic nerve tissues and serve as the principal accessory cells in the PNS ŽSchmidt et al., 1990.. Suppression of macrophage functions in nerves would abrogate the endoneural presentation of myelin antigens to autoreactive T-cells and consequently prevent continuous activation of autoreactive T-cells. The roles of IL-1 and IL-6 in the effector phase of EAN are unclear. We recently ŽBai et al., unpublished data. observed in sciatic nerve tissue higher numbers of IL-1 b mRNA expressing cells at the early onset Žday 7 p.i.. and
184
X.-F. Bai et al.r Journal of Neuroimmunology 76 (1997) 177–184
peak Žday 14 p.i.. of clinical EAN and declining numbers at recovery Žday 21 and 28 p.i... IL-6 mRNA expressing cells were detected at the peak of clinical EAN both in peripheral lymph node cells and in PNS tissue. Both IL-1 b and IL-6 seem thus to have a proinflammatory role in EAN. In the present study, both IFN-g and LPS induced IL-1 b and IL-6 mRNA expression were suppressed by Linomide in vitro and in vivo. These findings are also consistent with down-regulated macrophage functions induced by Linomide in EAN.
Acknowledgements This study was supported by grants from the Swedish Medical Research Council, the Swedish MS Society ŽNHR. and Karolinska Institute.
References Arad, G., Katzenellenbogen, M., Levy, R., Slavin, S., Kaempfer, R., 1996. Linomide, an immunomodulator that inhibits Th1 cytokine gene expression. Int. Immunol. 8, 1603–1607. Bjork, J., Kleinau, S., 1989. Paradoxical effects of LS-2616 ŽLinomide. treatment in the type II collagen arthritis model in mice. Agents Actions 27, 319–321. Dijkema, R., van der Meide, P.H., Dubbeld, M., Wubben, J., Schellekens, H., 1986. Cloning expression and purification of rat IFN-g . Methods Enzymol. 119, 453–463. Gonzalo, J.A., Gonzalez-Garcia, A., Kalland, T., Hedlund, G., MartinezA, C., Kroemer, G., 1993. Linomide, a novel immunomodulator that prevents death in four models of septic shock. Eur. J. Immunol. 23, 2372–2374. Gross, D.J., Sidi, H., Weiss, L., Kalland, T., Rosenmann, E., Slavin, S., 1994. Prevention of diabetes mellitus in non-obese diabetic mice by Linomide, a novel immunomodulating drug. Diabetologia 37, 1195– 1201. Hartung, H.P., Pollard, J.D., Harvey, G.K., Toyka, K.V., 1995. Immunopathogenesis and treatment of the Guillain-Barre´ syndrome part I. Muscle Nerve 18, 137–153. Henttinen, T., Levy, D.E., Silvennoinen, O., Hurme, M., 1995. Activation of signal transducer and transcription ŽSTAT. signalling pathway in a primary T-cell response. Critical role for IL-6. J. Immunol. 155, 4582–4587. Holsti, M.A., McArthur, J., Allison, J.P., Raulet, D.H., 1994. Role of IL-6, IL-1 and CD28 signalling in responses of mouse CD4q T-cells to immobilised anti-TCR monoclonal antibody. J. Immunol. 152, 1618–1628. Kalland, T., Alm, G., Stalhandske, T., 1985. Augmentation of mouse natural killer activity by LS-2616, a new immunomodulator. J. Immunol. 134, 3956–3961. Kalland, T., 1990. Regulation of natural killer progenitors. Studies with a
novel immunomodulator with distinct effects at the precursor level. J. Immunol. 144, 4472–4476. Karussis, D.M., Lehmann, D., Slavin, S., Vourka-Karussis, U., Mizrachi-Koll, R., Ovadia, H., Ben-Nun, A., Kalland, T., Abramsky, O., 1993a. Inhibition of acute, experimental autoimmune encephalomyelitis by the synthetic immunomodulator Linomide. Ann. Neurol. 34, 654–660. Karussis, D.M., Lehmann, D., Slavin, S., Vourka-Karussis, U., Mizrachi-koll, R., Ovadia, H., Kalland, T., Abramsky, O., 1993b. Treatment of chronic-relapsing experimental autoimmune encephalomyelitis with the synthetic immunomodulator Linomide Žquinoline-3-carboxamide.. Proc. Natl. Acad. Sci. USA 90, 6400– 6404. Karussis, D.M., Lehmann, D., Brenner, T., Wirguin, I., Mizrachi-Koll, R., Sicsic, C., Abramsky, O., 1994. Immunomodulation of experimental autoimmune myasthenia gravis with Linomide. J. Neuroimmunol. 55, 187–193. Kraemer, Y.I., Lopez, R.D., Reaven, G.M., 1985. Effects of noninsulindependent diabetes mellitus on the uptake of very low density lipoproteins by thioglycolate elicited mouse peritoneal macrophages. J. Clin. Endocrinol. Metab., 61–335. Lindstrom, ¨ J., Einarson, B., Tzartos, S., 1981. Production and assay of antibodies to acetylcholine receptors. Methods Enzymol. 74, 432–460. Norton, W.I., Poduslo, S.E., 1973. Myelination in rat nerve: Method of myelin isolation. J. Neurochem. 21, 749–757. Plebanski, M., Elson, C.J., Billington, W.D., 1992. Dependency on interleukin-1 of primary human in vitro T-cell responses to soluble antigens. Eur. J. Immunol. 22, 2353–2358. Said, G., Hontebeyrie-Joskowicz, M., 1992. Nerve lesions induced by macrophage activation. Res. Immunol. 143, 589–599. Schmidt, B., Stoll, G., Hartung, H.P., Heininger, K., Schafer, B., Toyka, ¨ K.V., 1990. Macrophages but not Schwann cells express Ia antigen in experimental autoimmune neuritis. Ann. Neurol. 28, 70–77. Stoll, G., Jung, S., Jander, S., van der Meide, P., Hartung, H.P., 1993. Tumor necrosis factor-a in immune-mediated demyelination and Wallerian degeneration of the rat peripheral nervous system. J. Neuroimmunol. 45, 175–182. Tarkowski, A., Gunnarsson, K., Nilsson, L.A., Lindholm, L., Stalhandske, T., 1986. Successful treatment of autoimmunity in MRLrl mice with LS-2616, a new immunomodulator. Arthritis Rheum. 29, 1405–1409. Vukanovic, J., Isaacs, J.T., 1995. Linomide inhibits angiogenesis, growth, metastasis, and macrophage infiltration within rat prostatic cancers. Cancer Res. 55, 1499–1504. Xiao, B.G., Bai, X.F., Zhang, G.X., Hojeberg, B., Link, H., 1996. Shift ¨ from anti- to proinflammatory cytokine profiles in microglia through LPS- or IFN-g-mediated pathways. Neuroreport 7, 1893–1898. Zhang, G.X., Shi, F.D., Xiao, B.G., Bjork, ¨ J., Hedlund, G., Link, H., 1996. Linomide suppresses both Th1 and Th2 cytokines in experimental autoimmune myasthenia gravis. J. Neuroimmunol., in press. Zhu, J., Mix, E., Issazadeh, S., Link, H., 1996. Dynamics of mRNA expression of interferon-g, interleukin 4 and transforming growth factor b1 in sciatic nerves and lymphoid organs in experimental allergic neuritis. Eur. J. Neurol. 3, 1–9. Zielasek, J., Jung, S., Gold, R., Liew, F.Y., Toyka, K.V., Hartung, H.P., 1995. Administration of nitric oxide synthase inhibitors in experimental autoimmune neuritis and experimental autoimmune encephalomyelitis. J. Neuroimmunol. 58, 81–88.
Linomide-induced suppression of experimental autoimmune neuritis is associated with down-regulated macrophage functions a,)
Xue-Feng Bai a
, Fu-Dong Shi a , Jie Zhu a , Bao-Guo Xiao a , Gunnar Hedlund b, Hans Link
a
DiÕision of Neurology, Karolinska Institute, Huddinge Hospital, S-141 86 Huddinge, Stockholm, Sweden b Pharmacia and Upjohn, Lund Research Centre, Lund, Sweden Received 4 September 1996; revised 3 December 1996; accepted 3 February 1997
Abstract Experimental autoimmune neuritis ŽEAN. is a T-cell mediated autoimmune disease of the peripheral nervous system, in which macrophages and T-cells feature prominently in nerve lesions. EAN represents a counterpart to Guillain-Barre´ syndrome in humans. In the present study, we investigated the in vitro and in vivo effects of Linomide ŽLS-2616, quinoline-3-carboxamide., a synthetic immunomodulatory compound, on macrophages in relation to EAN. Linomide strongly suppressed IFN-g and lipopolysaccharide ŽLPS.-induced IL-1 b , TNF-a and IL-6 mRNA expression in macrophages in vitro as demonstrated by in situ hybridisation. Linomide administered daily subcutaneously from the day of inoculation completely prevented the development of clinical symptoms of EAN. Linomide administered from day 9 post immunisation Žp.i.. significantly suppressed clinical EAN. Macrophages from Linomide-treated EAN rats showed decreased IL-1 b , TNF-a and IL-6 mRNA expression in response to IFN-g and LPS. LPS-induced nitric oxide production by macrophages was also suppressed by Linomide in vitro. Linomide, however, does not affect macrophage death and release of lactate dehydrogenase. We conclude that Linomide may exert its actions in EAN and perhaps also in other autoimmune disease models, by suppressing macrophage functions. Keywords: Experimental autoimmune neuritis; Linomide; Guillain-Barre´ syndrome; Cytokine; Macrophage
1. Introduction Linomide ŽLS-2616, quinoline-3-carboxamide., a synthetic immunomodulatory compound, is effective in inhibiting manifestations of acute and chronic-relapsing experimental autoimmune encephalomyelitis ŽEAE. ŽKarussis et al., 1993a,b. and myasthenia gravis ŽEAMG. ŽKarussis et al., 1994; Zhang et al., 1996., lupus-like disease in MRLr1 mice ŽTarkowski et al., 1986., insulin-dependent diabetes mellitus and autoimmune insulitis in NOD mice ŽGross et al., 1994. and collagen-induced arthritis in DBA mice ŽBjork and Kleinau, 1989.. The mode of action of Linomide in the modulation of autoimmune diseases is not clear. Suppressed T- and B-cell responses to specific antigens ŽKarussis et al., 1993a, 1994., augmented NK cell activity ŽKalland et al., 1985; Kalland, 1990. as well as suppressed Th1 responses ŽArad et al., 1996. have been suggested.
)
Corresponding author. Tel.: q46-8-58582279; fax: q46-8-7744822.
Experimental autoimmune neuritis ŽEAN. is a T-cell mediated autoimmune disease of the peripheral nervous system ŽPNS.. In many respects, EAN represents an experimental counterpart to acute inflammatory demyelinating polyradiculoneuropathy or Guillain-Barre´ syndrome ŽGBS. in humans. Macrophages act as professional antigen presenting cells and are of pivotal importance in the amplification and effector phase of immune-mediated demyelination. Macrophages feature prominently in the nerve lesions of EAN and GBS ŽHartung et al., 1995.. In the effector phase of EAN, macrophages exert their functions by phagocytosis of myelin components and release of proinflammatory cytokines such as IL-1, TNF-a and IL-6 and other highly active mediators that act at short range, like nitric oxide ŽNO. ŽZielasek et al., 1995.. Since macrophages are essential in the initiation and amplification of the autoimmune responses in EAN and Linomide may influence macrophage functions ŽVukanovic and Isaacs, 1995., we investigated the effects of Linomide on macrophages in relation to EAN. We found that Linomide strongly downregulated IL-1 b , TNF-a and IL-6
0165-5728r97r$17.00 Copyright q 1997 Elsevier Science B.V. All rights reserved. PII S 0 1 6 5 - 5 7 2 8 Ž 9 7 . 0 0 0 5 1 - 9
178
X.-F. Bai et al.r Journal of Neuroimmunology 76 (1997) 177–184
mRNA expression in macrophages both in vitro and in vivo, and down-regulated NO production by macrophages.
2. Materials and methods 2.1. Antigens and immunoreagents Bovine peripheral nerve myelin ŽBPM. was prepared from lumbosacral plexus ŽNorton and Poduslo, 1973.. Nicotinic acetylcholine receptor ŽAChR. was purified as described ŽLindstrom ¨ et al., 1981. from electroplax tissue of Torpedo californica ŽPacific Bio-Marine, Venice, CA.. Lipopolysaccharide ŽLPS. from Escherichia coli was provided by the Department of Microbiology, Huddinge Hospital, Karolinska Institute. Rat rIFN-g was provided by Dr. P.H. van der Meide ŽBiomedical Primate Centre, Rijswijk, The Netherlands. ŽDijkema et al., 1986.. mAbs to rat MHC class II, ICAM-1 and macrophages ŽED1. were purchased from Serotec ŽOxford, UK.. 2.2. Induction of EAN and Linomide treatment Male Lewis rats, 150–200 g, locally bred, originally obtained from Harlan CPB, Zeist, The Netherlands, were injected in both hind foot pads with 200 m l of inoculum containing 4 mg of BPM, 2 mg Mycobacterium Ž M.. tuberculosis Žstrain H37RA; Difco, Detroit, MI. in 100 m l saline and 100 m l Freund’s incomplete adjuvant ŽFIA. ŽDifco.. Symptoms of paresis were graded as follows: 0, no illness; 1, flaccid tail; 2, moderate paraparesis; 3, severe paraparesis; 4, tetraparesis. Linomide ŽPharmacia-Upjohn, Lund. was dissolved in distilled water and administered at a single dose of 160 mgrkgrday subcutaneously. 20 Lewis rats were immunised with BPM q complete Freund’s adjuvant ŽCFA.. 10 of the 20 rats were given Linomide from the day of immunisation until sacrifice or the end of experiments. The other 10 rats served as controls and were injected with PBS ŽpH 7.7. only. In both the Linomidetreated and control groups, 5 rats were used to isolate macrophages from the abdominal cavity at day 14 p.i. To evaluate Linomide effects on the efferent phase of EAN, another group of 8 rats were immunised with BPM q CFA. 4r8 rats were given Linomide at a dose of 160 mgrkg subcutaneously daily from day 9 p.i., i.e. just before onset of clinical signs, to the end of the experiment. The remaining 4 rats served as controls and were injected with PBS ŽpH 7.7. daily until the end of the experiment. 2.3. Isolation of macrophages, in Õitro culture and preparation of cells for in situ hybridisation Groups of 5 rats, including rats immunised with BPM plus CFA, rats immunised with CFA only and normal adult rats were injected intraperitoneally 4 days before sacrifice with 2 ml of 3% thioglycollate ŽSigma, St. Louis, MO. as described by Kraemer et al. Ž1985.. Four days
after injection, the peritoneal cavities were rinsed for 3 min with complete Iscove’s medium ŽFlow Lab., Irvine. supplemented with 1% Žvrv. minimum essential medium ŽFlow., 50 IUrml penicillin, 60 mgrml streptomycin ŽGibco, Paisley, UK., 2 mM glutamin ŽFlow. and 3% normal human ABq serum. After three washes in the same medium, the cell number of the suspension was adjusted to 1 = 10 6 viable cellsrml medium. The percentage of cells excluding trypan blue generally exceeded 90%. The cells were further cultured in 200 ml tissue culture flasks at 378C for 1 h. The adherent cells were collected. The purity was ) 95% as determined by flow cytometry using the PE-conjugated mAb ED1 ŽSerotec.. Purified macrophages were cultured in the medium at a concentration of 0.5 = 10 6rml in the presence of 0.5 m grml LPS or 50 IUrml of rat rIFN-g . A culture time of 24 and 48 h was selected for evaluation of cytokine mRNA expression. After washing in PBS, pH 7.4, 10 m l containing 10 4 macrophages were dried onto restricted areas of electrically charged glass slides ŽProbeOn slides; Fisher Scientific, Pittsburgh, PA. and stored in sealed boxes at y708C until hybridisation. 2.4. In situ hybridisation (ISH) ISH was performed according to Zhu et al. Ž1996. using S-labeled synthetic oligonucleotide probes ŽScandinavian Gene Synthesis AB, Koping, Sweden.. For each cytokine ¨ ŽXiao et al., 1996., a mixture of four different approximately 48-bp oligonucleotide probes were used in order to increase the sensitivity of the method. The oligonucleotide sequences were obtained from GeneBank using the MacVector system. Control slides were hybridised with the same total amount of a sense probe with nucleotide sequence for exon 4 of rat IFN-g . A constant ratio of the guaninercytosine content of approximately 60% was employed. The oligonucleotide probes were checked for absence of palindromes and long sequences of homology within the species against available GenBank data. The labelling was performed with 35 S deoxyadenosine 5Xthiotriphosphate ŽNew England Nuclear, Cambridge, MA. with terminal deoxynucleotidyl transferase ŽAmersham, Little Chalfont, UK.. After emulsion autoradiography, slides were coded and cells expressing numerous grains over their cytoplasm were counted by darkfield microscopy at 20 = magnification as previously described and validated ŽZhu et al., 1996.. The results are expressed as number of labelled cells per 10 4 macrophages. The control probe used in parallel with the cytokine probes on slides produced a uniformly weak background signal without revealing any positive cells. 35
2.5. Nitrite assay Because secreted NO quickly reacts with oxygen-yielding nitrite, the level of nitrite as a reflection of NO production in macrophage culture supernatants was measured using Griess reagent Ž1% sulfanilamide and 0.1%
X.-F. Bai et al.r Journal of Neuroimmunology 76 (1997) 177–184
N-Ž1-naphthyl. ethylenediamine dihydrochloride in 5% H 3 PO4 .. 100 m l supernatants from cultured cells were mixed with an equal volume of Griess reagent. After a 10 min reaction at room temperature, the absorbance at 540 nm was measured using an automatic plate reader. Nitrite concentration was determined by comparison with a sodium nitrite standard curve in culture medium.
179
2.7. Apoptosis assay For detection and quantification of macrophage apoptosis, an in situ cell death detection kit ŽFluorescein, Boehringer-Mannheim, Germany. was used according to the manufacturer’s instructions. Cell smear samples were used and slides were observed under fluorescence microscope.
2.6. Immunohistochemistry 2.8. LDH assay The segments of sciatic nerve close to the lumbar spinal cord from 4 control EAN rats and 4 EAN rats receiving Linomide treatment from the day of immunisation were dissected and snap-frozen in liquid nitrogen on day 14 p.i. Cryopreserved sciatic nerve tissue sections Ž10 m m. were first fixed in acetone at y208C. Non-specific binding was blocked with normal horse serum ŽVector, Burlingame, CA. for 30 min. Macrophage and T-lymphocyte infiltrates were characterised in serial sections by incubation with 5 m grml of the monoclonal antibodies ŽmAb. ED1 and W3r13 ŽSerotec, Oxford, UK., respectively. Biotinylated horse anti-mouse antibody ŽVector. was used to couple the primary mAbs with an avidin–biotin–peroxidase complex ŽVector.. After washing in PBS, substrate amino-ethyl carbazole ŽAEC. was applied to the tissue and incubated for 5 min. The slides were rinsed in tap water, counterstained with hematoxylin and mounted with Aquamount ŽLerner Lab., New Haven, CT.. Controls included the omission of the primary antibody and second antibody. Negligible background staining was observed.
To measure LDH activity in supernatants of macrophage cultures at 308C, a diagnostic kit ŽSigma. was used according to the manufacturer’s instructions. One U of LDH activity is defined as the amount of enzyme that will catalyze the formation of 1.0 mmol of reduced nicotinamide adenine dinucleotide per min. 2.9. Statistics The Student’s t-test was used. All significance tests were two-sided and rejection limit was set to a s 0.05. 3. Results 3.1. Linomide suppresses IL-1b , TNF-a and IL-6 MRNA expression in macrophages Upon culture, only low levels Ž- 10 per 10 4 . of macrophages were detected that expressed TNF-a or IL-6
Fig. 1. LPS-induced IL-1 b mRNA expressing macrophages seen at light microscopy= 400 magnification; arrows indicate examples of IL-1 b mRNA positive cells. Macrophages are from EAN rats on day 14 p.i.
180
X.-F. Bai et al.r Journal of Neuroimmunology 76 (1997) 177–184
spontaneously. Numbers of IL-1 b mRNA expressing cells were higher Ž50–100r10 4 macrophages.. After the culture of macrophages in the presence of rIFN-g or LPS, the number of cells expressing IL-1 b ŽFig. 1., TNF-a or IL-6 mRNA were upregulated. Levels of cytokine mRNA expressing macrophages did not differ upon incubation with rIFN-g during observation periods of 24 h versus 48 h ŽFig. 2.. The strongest effects of LPS on levels of cytokine mRNA expressing cells were observed at 24 h. Numbers had declined at 48 h ŽFig. 3.. In the presence of different concentrations Ž10y4 , 10y6 and 10y8 M. of Linomide to culture medium, the median and high doses of Linomide strongly suppressed IFN-g as well as LPS-induced IL-1 b and TNF-a and IFN-g-induced IL-6 mRNA expression in macrophages at both time points ŽFigs. 2 and 3.. LPS-induced IL-6 mRNA expression in macrophages was suppressed by the high dose of Linomide at 48 h. No suppressive effects were observed in cultures treated with low concentration of Linomide during the total observation period. Macrophages from EAN rats Žshown in Figs. 2 and 3., CFA immunized rats and normal rats showed similar results. 3.2. Linomide suppresses LPS-induced NO production by macrophages in Õitro Upon in vitro culture for 24 and 48 h, macrophages spontaneously produce NO. However, the concentration of
this spontaneous NO production was low. Upon in vitro culture in the presence especially of LPS but also of IFN-g , high concentrations of NO were detected in the culture supernatants. The increase of NO was time and dose dependent. Higher doses of LPS induced higher concentrations of NO. Furthermore, higher concentrations of NO were detected at 48 h than 24 h ŽFig. 4.. In the presence of median Ž10y6 M. and high Ž10y4 M. concentrations of Linomide, LPS-induced NO production by macrophages was significantly suppressed at 48 h ŽFig. 4.. Spontaneous and IFN-g induced NO production was also reduced, without reaching statistical significance. This might be due to the lower basal NO levels in the culture supernatants. Macrophages were obtained from normal Lewis rats, rats with EAN Žshown in Fig. 4. and rats immunised with CFA and the results were similar. 3.3. In ÕiÕo effects of Linomide The effects of Linomide on clinical EAN were evaluated. To evaluate Linomide effects on the afferent phase of EAN, a group of 20 rats was used and Linomide Ž160 mgrkgrday, subcutaneously. was given to 10r20 rats from day 0 p.i. until sacrifice of rats on day 14 p.i. or up to the end of experiments Žday 28 p.i... The other 10 rats served as controls and received PBS ŽpH 7.7. only. All control rats immunised with BPM q CFA developed severe EAN by day 14 p.i. In contrast, none of the rats
Fig. 2. Linomide suppresses rIFN-g induced mRNA expression in macrophages collected from EAN rats in vitro. 0.5 = 10 6 macrophages per ml were cultured in the presence of 50 IUrml of rIFN-g . Three different concentrations of Linomide were added to separate cultures for 24 h or 48 h. After washing, 10 m l containing 10 4 macrophages were dried onto ProbOn slides and hybridized with cytokine probes. 10y6 and 10y4 M of Linomide suppressed IL-1 b , TNF-a and IL-6 mRNA expression in macrophages. Bars represent mean values of separate macrophage cultures from 5 different EAN rats. ) P - 0.05; ) ) P - 0.01; ) ) ) P - 0.001.
X.-F. Bai et al.r Journal of Neuroimmunology 76 (1997) 177–184
181
Fig. 3. Linomide suppresses LPS-induced cytokine mRNA expression in macrophages in vitro. The same experimental conditions were used as described in Fig. 2, except that IFN-g was displaced by 0.5 m grml of LPS. Macrophages are purified from 5 EAN rats.
receiving Linomide injection from the day of immunisation developed EAN ŽFig. 5a.. On examination of sciatic nerve tissue sections for the presence of macrophage and T-lymphocyte infiltration on day 14 p.i., 4r4 control rats showed extensive macrophage and T-lymphocyte infiltration. In contrast, none of the 4 Linomide-treated rats showed any macrophage or T-lymphocyte infiltrations.
Fig. 4. Linomide suppresses LPS-induced NO production by macrophages. Five EAN rats were sacrificed at day 14 p.i. and macrophages were isolated as described in Section 2. 0.5=10 6 macrophages per ml were cultured in the presence of 0.5 m grml of LPS with or without the presence of 10y6 M of Linomide. Nitrite concentrations were measured in the culture supernatant. A significant suppression of nitrite production was observed at 48 h. Bars represent mean values of macrophage cultures from 5 EAN rats. ) P - 0.05.
Furthermore, Linomide significantly reduced or even eliminated inflammation at the immunisation sites Žhind footpads. in rats receiving Linomide from the day of immunisation compared to control rats. To evaluate Linomide effects on the efferent phase of EAN, another group of 8 rats were immunised with BPM q CFA, followed in 4r8 rats by injection of the same dose of Linomide at day 9 p.i., i.e. just before onset of clinical signs. The other 4 rats served as controls and were injected with PBS ŽpH 7.7. only. As shown in Fig. 5b, Linomide significantly suppressed EAN severity from day 12 to day 24 p.i. Ž P 0.01.. Macrophages were purified from both control rats and rats receiving Linomide from the day of immunisation. Cells were then challenged in vitro with IFN-g or LPS. Macrophages from Linomide-treated EAN rats showed lower numbers of IFN-g and LPS-induced IL-1 b and TNF-a mRNA expression compared to macrophages from control rats ŽFig. 6.. Borderline reduced IL-6 mRNA expression in macrophages from Linomide-treated EAN rats was also observed. 3.4. Linomide does not affect macrophages in expressing MHC class II, ICAM-1 and surÕiÕing in Õitro None of the three Linomide doses used had any effect on spontaneous, IFN-g or LPS-induced MHC class II or ICAM-1 surface expression on macrophages at 24 h and 48 h Ždata not shown.. To study induction of apoptotic
182
X.-F. Bai et al.r Journal of Neuroimmunology 76 (1997) 177–184
Fig. 5. Linomide administration in vivo suppresses clinical EAN. Ža. Linomide Ž160 mgrkgrday. administered subcutaneously from the day of immunisation to the end of experiments prevented development of EAN. Žb. Linomide Ž160 mgrkgrday. administered from day 9 p.i. to the end of experiments significantly Ž P - 0.01. suppressed severity of EAN from day 12 p.i. to day 24 p.i. Bars represent mean values from 4–5 rats at each time point.
macrophage death, different concentrations of Linomide were added to culture medium. While presence of LPS or IFN-g resulted in increased numbers of apoptotic cells,
addition of Linomide had no effect on spontaneous or IFN-g or LPS induced macrophage death Ždata not shown.. LDH release by macrophages was also investigated.
Fig. 6. In vivo administration of 160 mgrkgrday of Linomide subcutaneously suppresses IFN-g Ž50 IUrml; a–c. and LPS Ž0.5 m grml; d–f. induced mRNA expression of IL-1 b , TNF-a and IL-6 in macrophages. Five EAN rats receiving Linomide and 5 control EAN rats receiving PBS daily from the day of immunisation were sacrificed at day 14 p.i. and macrophages were isolated as described. The same experimental conditions were set as described in Figs. 2 and 3. Bars represent mean values for 5 rats. B, control EAN rats treated with PBS; I, EAN rats treated with Linomide. ) P - 0.05; )) P - 0.01; ) ) ) P - 0.001.
X.-F. Bai et al.r Journal of Neuroimmunology 76 (1997) 177–184
Fig. 7. Linomide does not affect spontaneous, LPS or IFN-g induced LDH release by macrophages. Macrophages were taken from 5 EAN rats sacrificed at day 14 p.i. and pooled. 200 m l of 0.5=10 6 macrophages per ml were cultured in the absence or presence of 0.5 m grml LPS or 50 IUrml of rIFN-g . Culture supernatants were then examined for the presence of LDH. Linomide had no pronounced effects on spontaneous, LPS or IFN-g induced LDH release. The results were confirmed by 2 repeated experiments.
Spontaneous release of LDH by macrophages was low upon in vitro culture in medium after 48 h. LPS but not IFN-g upregulated LDH release. Addition of Linomide at three different doses had no pronounced effect on spontaneous, IFN-g or LPS induced LDH release by macrophages ŽFig. 7..
4. Discussion We observed that Linomide suppressed IFN-g and LPS mediated induction of IL-1 b , TNF-a and IL-6 mRNA expression in macrophages in vitro. Linomide administered during the induction phase of EAN completely prevented the development of clinical EAN in Lewis rats. A significant suppression of EAN was also achieved when Linomide was given after day 9 p.i., i.e. the effector phase of EAN. Levels of LPS and IFN-g induced IL-1 b and TNF-a mRNA expression in macrophages from Linomide-treated EAN rats were lower than in macrophages from control EAN rats. LPS-induced NO production by macrophages was also suppressed by Linomide in vitro. In contrast, macrophage death and LDH release were not affected by Linomide. The effects observed in this study are consistent with previous observations. Vukanovic and Isaacs Ž1995. observed that monocytesrmacrophages prepared from Linomide-treated rats secreted low amounts of TNF-a when challenged with LPS in vitro. Furthermore, when rats were treated with Linomide and then challenged with LPS in vivo, the resulting elevation in serum TNF-a was inhibited by approximately 50%. Septic shock in mice induced by stimulation of macrophages with LPS is also prevented by daily treatment with 100 mgrkg of Linomide, an effect
183
that is due to inhibited TNF-a secretion ŽGonzalo et al., 1993.. The complete prevention of EAN by injection of Linomide in the induction phase indicates that systemic administration of Linomide mainly targets and interferes with trafficking of macrophages and lymphocytes into the nerve tissue. This is supported by the absence of inflammatory cell infiltrations in the target tissue. Systemic suppression of macrophage functions, reflected by the reduced capacity of macrophages in expressing IL-1 b , TNF-a and IL-6 mRNA, may interfere with further presentation of myelin antigens at the site of immunisation and ameliorate EAN by interrupting the supply of activated T-cells. How Linomide affects antigen presentation by macrophage and further T-cell activation is unclear. We postulate that, at the immunisation sites, Linomide suppresses adjuvant-induced constitutive macrophage activation. These macrophages will reduce the release of proinflammatory cytokines like TNF-a , IL-1 b and IL-6, which in turn will reduce macrophage accumulation and activation, lead to a reduced number of activated macrophages migrating to the draining lymph nodes and finally result in insufficient antigen presentation. In the present study, in Linomide-treated rats, the absence of inflammation in the immunisation sites indicates absence of macrophage activation. Furthermore, in Linomide-treated rats, macrophages migrating to draining lymph nodes lack the capacity to activate naive T-lymphocytes. Linomide reduced the capacity of macrophages to produce IL-1 b and IL-6, both of which are important for stimulation of naive T lymphocytes and synergize in the induction of T-cell activation ŽPlebanski et al., 1992; Holsti et al., 1994; Henttinen et al., 1995.. The therapeutic effects of Linomide on clinical signs of actively induced EAN suggest that Linomide may not only reduce the number of T-cells and macrophages invading the nerve but also inhibit the function of activated macrophagesrT-lymphocytes, which have already invaded peripheral nerves and impair nerve conduction or cause myelin damage through the release of toxic mediators such as TNF-a ŽSaid and Hontebeyrie-Joskowicz, 1992; Stoll et al., 1993. and NO ŽZielasek et al., 1995.. It is of note that Linomide suppressed the release of nitrogen intermediates as well as TNF-a mRNA expression in macrophages in the present study. Another possibility that deserves consideration is that macrophages are constitutively present in sciatic nerve tissues and serve as the principal accessory cells in the PNS ŽSchmidt et al., 1990.. Suppression of macrophage functions in nerves would abrogate the endoneural presentation of myelin antigens to autoreactive T-cells and consequently prevent continuous activation of autoreactive T-cells. The roles of IL-1 and IL-6 in the effector phase of EAN are unclear. We recently ŽBai et al., unpublished data. observed in sciatic nerve tissue higher numbers of IL-1 b mRNA expressing cells at the early onset Žday 7 p.i.. and
184
X.-F. Bai et al.r Journal of Neuroimmunology 76 (1997) 177–184
peak Žday 14 p.i.. of clinical EAN and declining numbers at recovery Žday 21 and 28 p.i... IL-6 mRNA expressing cells were detected at the peak of clinical EAN both in peripheral lymph node cells and in PNS tissue. Both IL-1 b and IL-6 seem thus to have a proinflammatory role in EAN. In the present study, both IFN-g and LPS induced IL-1 b and IL-6 mRNA expression were suppressed by Linomide in vitro and in vivo. These findings are also consistent with down-regulated macrophage functions induced by Linomide in EAN.
Acknowledgements This study was supported by grants from the Swedish Medical Research Council, the Swedish MS Society ŽNHR. and Karolinska Institute.
References Arad, G., Katzenellenbogen, M., Levy, R., Slavin, S., Kaempfer, R., 1996. Linomide, an immunomodulator that inhibits Th1 cytokine gene expression. Int. Immunol. 8, 1603–1607. Bjork, J., Kleinau, S., 1989. Paradoxical effects of LS-2616 ŽLinomide. treatment in the type II collagen arthritis model in mice. Agents Actions 27, 319–321. Dijkema, R., van der Meide, P.H., Dubbeld, M., Wubben, J., Schellekens, H., 1986. Cloning expression and purification of rat IFN-g . Methods Enzymol. 119, 453–463. Gonzalo, J.A., Gonzalez-Garcia, A., Kalland, T., Hedlund, G., MartinezA, C., Kroemer, G., 1993. Linomide, a novel immunomodulator that prevents death in four models of septic shock. Eur. J. Immunol. 23, 2372–2374. Gross, D.J., Sidi, H., Weiss, L., Kalland, T., Rosenmann, E., Slavin, S., 1994. Prevention of diabetes mellitus in non-obese diabetic mice by Linomide, a novel immunomodulating drug. Diabetologia 37, 1195– 1201. Hartung, H.P., Pollard, J.D., Harvey, G.K., Toyka, K.V., 1995. Immunopathogenesis and treatment of the Guillain-Barre´ syndrome part I. Muscle Nerve 18, 137–153. Henttinen, T., Levy, D.E., Silvennoinen, O., Hurme, M., 1995. Activation of signal transducer and transcription ŽSTAT. signalling pathway in a primary T-cell response. Critical role for IL-6. J. Immunol. 155, 4582–4587. Holsti, M.A., McArthur, J., Allison, J.P., Raulet, D.H., 1994. Role of IL-6, IL-1 and CD28 signalling in responses of mouse CD4q T-cells to immobilised anti-TCR monoclonal antibody. J. Immunol. 152, 1618–1628. Kalland, T., Alm, G., Stalhandske, T., 1985. Augmentation of mouse natural killer activity by LS-2616, a new immunomodulator. J. Immunol. 134, 3956–3961. Kalland, T., 1990. Regulation of natural killer progenitors. Studies with a
novel immunomodulator with distinct effects at the precursor level. J. Immunol. 144, 4472–4476. Karussis, D.M., Lehmann, D., Slavin, S., Vourka-Karussis, U., Mizrachi-Koll, R., Ovadia, H., Ben-Nun, A., Kalland, T., Abramsky, O., 1993a. Inhibition of acute, experimental autoimmune encephalomyelitis by the synthetic immunomodulator Linomide. Ann. Neurol. 34, 654–660. Karussis, D.M., Lehmann, D., Slavin, S., Vourka-Karussis, U., Mizrachi-koll, R., Ovadia, H., Kalland, T., Abramsky, O., 1993b. Treatment of chronic-relapsing experimental autoimmune encephalomyelitis with the synthetic immunomodulator Linomide Žquinoline-3-carboxamide.. Proc. Natl. Acad. Sci. USA 90, 6400– 6404. Karussis, D.M., Lehmann, D., Brenner, T., Wirguin, I., Mizrachi-Koll, R., Sicsic, C., Abramsky, O., 1994. Immunomodulation of experimental autoimmune myasthenia gravis with Linomide. J. Neuroimmunol. 55, 187–193. Kraemer, Y.I., Lopez, R.D., Reaven, G.M., 1985. Effects of noninsulindependent diabetes mellitus on the uptake of very low density lipoproteins by thioglycolate elicited mouse peritoneal macrophages. J. Clin. Endocrinol. Metab., 61–335. Lindstrom, ¨ J., Einarson, B., Tzartos, S., 1981. Production and assay of antibodies to acetylcholine receptors. Methods Enzymol. 74, 432–460. Norton, W.I., Poduslo, S.E., 1973. Myelination in rat nerve: Method of myelin isolation. J. Neurochem. 21, 749–757. Plebanski, M., Elson, C.J., Billington, W.D., 1992. Dependency on interleukin-1 of primary human in vitro T-cell responses to soluble antigens. Eur. J. Immunol. 22, 2353–2358. Said, G., Hontebeyrie-Joskowicz, M., 1992. Nerve lesions induced by macrophage activation. Res. Immunol. 143, 589–599. Schmidt, B., Stoll, G., Hartung, H.P., Heininger, K., Schafer, B., Toyka, ¨ K.V., 1990. Macrophages but not Schwann cells express Ia antigen in experimental autoimmune neuritis. Ann. Neurol. 28, 70–77. Stoll, G., Jung, S., Jander, S., van der Meide, P., Hartung, H.P., 1993. Tumor necrosis factor-a in immune-mediated demyelination and Wallerian degeneration of the rat peripheral nervous system. J. Neuroimmunol. 45, 175–182. Tarkowski, A., Gunnarsson, K., Nilsson, L.A., Lindholm, L., Stalhandske, T., 1986. Successful treatment of autoimmunity in MRLrl mice with LS-2616, a new immunomodulator. Arthritis Rheum. 29, 1405–1409. Vukanovic, J., Isaacs, J.T., 1995. Linomide inhibits angiogenesis, growth, metastasis, and macrophage infiltration within rat prostatic cancers. Cancer Res. 55, 1499–1504. Xiao, B.G., Bai, X.F., Zhang, G.X., Hojeberg, B., Link, H., 1996. Shift ¨ from anti- to proinflammatory cytokine profiles in microglia through LPS- or IFN-g-mediated pathways. Neuroreport 7, 1893–1898. Zhang, G.X., Shi, F.D., Xiao, B.G., Bjork, ¨ J., Hedlund, G., Link, H., 1996. Linomide suppresses both Th1 and Th2 cytokines in experimental autoimmune myasthenia gravis. J. Neuroimmunol., in press. Zhu, J., Mix, E., Issazadeh, S., Link, H., 1996. Dynamics of mRNA expression of interferon-g, interleukin 4 and transforming growth factor b1 in sciatic nerves and lymphoid organs in experimental allergic neuritis. Eur. J. Neurol. 3, 1–9. Zielasek, J., Jung, S., Gold, R., Liew, F.Y., Toyka, K.V., Hartung, H.P., 1995. Administration of nitric oxide synthase inhibitors in experimental autoimmune neuritis and experimental autoimmune encephalomyelitis. J. Neuroimmunol. 58, 81–88.