Clinical scores were analysed using the non-parametric Mann–Whitney U-test. The level of significance was set at P < 0·05. EAE was induced in C57BL/6 mice by immunization with the MOG35–55 peptide in CFA followed by i.v. injection of PT. EAE mice exhibited three disease phases: preclinical, peak and remission phases. AG-014699 nmr Clinical signs (partial limp tail) presented at 7 dpi. Disease

then progressed to limp tail, waddling gait and paralysis during the peak phases (at 16 dpi). Finally, mice recovered but still presented with clinical signs during the remission phases (at 28 dpi). CFA mice showed no clinical signs at all (Fig. 1a). Lymph node MNCs were isolated from 7 dpi EAE and CFA mice and then co-cultured with astrocytes at lymphocyte : astrocyte ratios of 10:1, 1:1, and 1:5. At the lymphocyte : astrocyte ratios tested there were no differences in proliferation among cells isolated from the CFA group, with the exception of CD3/CD28 and concanavalin A (ConA)-stimulated cells (Fig. 1b). Conversely, lymphocytes isolated

from EAE mice proliferated significantly in response to stimulation with MOG35–55 peptide (P < 0·001). In the EAE lymphocyte : astrocyte co-cultured group, lymphocyte proliferation was inhibited by half at a ratio of 10:1 (P < 0·01) and inhibited completely at ratios of 1:1 and 1:5 (P < 0·001) compared to proliferation observed for MOG35–55 peptide-stimulated EAE lymphocytes alone. These data indicate that the inhibitory effect of astrocytes on MOG35–55-specific lymphocytes is correlated with lymphocyte : astrocyte ratios. Lymphocytes were then co-cultured with astrocytes PF2341066 at a lymphocyte : astrocyte Isoconazole ratio of 10 : 1. Supernatants were obtained 72 h later and cytokine levels were detected by ELISA. In the supernatants collected from EAE lymphocyte : astrocyte cultures, IFN-γ (P < 0·001) and IL-17 (P < 0·001) levels were decreased significantly; IL-4 and TGF-β levels were also decreased compared to levels observed for EAE lymphocytes. There were no significant differences in cytokine production by cells harvested from mice

in the CFA groups. Levels of the above cytokines were lower in the supernatants of astrocytes cultured alone (Fig. 1c). The suppressing effect of astrocyte on MOG35–55-specific lymphocytes might be mediated by soluble factors as well as cell contact. We cultured astrocyte and MOG35–55-specific lymphocytes without contact between both cells using Transwell plates. Supernatants were taken out to test cytokine levels after 72 h. Results are shown in Fig. 1d. Significant reductions of IFN-γ (P < 0·001) and IL-17 (P < 0·001) levels were also observed at the co-culture group without contact between both cells. These results suggest that cell contact is not required in astrocyte-mediated suppression of lymphocyte secreting, and might be mediated by soluble factors. Astrocytes were incubated in the presence or absence of IFN-γ and then co-cultured with lymphocytes for 72 h.