Cerebrospinal fluid affects phenotype and functions of myeloid dendritic cells

Abstract

Myeloid (CD11c+) dendritic cells (DC) are present in cerebrospinal fluid (CSF), as well as in the meninges and choroid plexus. Functional studies of these DC are hindered or impossible. To obviate this problem, we investigated the effects of CSF supernatants from patients with non-inflammatory neurological diseases (NIND), multiple sclerosis (MS), bacterial meningitis (BM) and Lyme meningoencephalitis (LM) on immature monocyte-derived DC (moDC) from healthy donors. CSF supernatants caused maturation of moDC (MS > LM > NIND > BM), as reflected by a decrease in CD1a, and an increase in HLA-DR, CD80 and CD86 expression. The maturation effect of MS CSF and LM CSF could be blocked by anti-TNF-alpha MoAb or recombinant human IL-10. moDC cultured with BM CSF either remained immature or turned into CD14+ macrophage-like cells and were relatively inefficient at inducing T cell responses in vitro. In contrast, moDC cultured with LM CSF induced strong Th1 responses. Both BM CSF and LM CSF contained IFN-gamma, a cytokine that augments IL-12 production by moDC and hence should confer an ability to induce a Th1 response. However, BM CSF also contained high levels of IL-10, which could antagonize the effects of IFN-gamma on moDC. moDC cultured with MS CSF induced a higher production of IFN-gamma from T cells compared to moDC cultured with NIND CSF or BM CSF. In summary, soluble factors present in the CSF may influence the phenotype and functions of meningeal, choroid plexus and CSF DC which, in turn, may have an impact on the character of intrathecal T cell responses.

Fig. 1

Influence of CSF supernatants from patients with MS (n = 34), ON (n = 10), BM (n = 5), LM (n = 5) and NIND (n = 9) on surface phenotype of donor moDC. MoDC obtained in four independent experiments were cultured for 24 h with either CSF supernatants (1/2, v/v), medium as a negative control, or rhTNF-α (100 ng/ml) as a positive control. Expression of surface markers was analysed by flow cytometry. Data are expressed as percentage MFI after culture with CSF or rhTNF-α in relation to MFI after culture with medium alone in the same experiment (background MFI). For CD1a, closed circles in the BM group indicate two cases where simultaneous up-regulation of CD14 was observed. Bars indicate medians. *P < 0·05, **P < 0·01, ***P < 0·001

Fig. 2

Influence of CSF on the ability of donor moDC to induce proliferation of allogeneic naive T cells (MLR). MoDC were cultured for 24h with either medium, or CSF supernatants, or rhTNF-α (100 ng/ml), washed, irradiated (30 Gy) and co-cultured in graded numbers with 5 × 10⁴ naïve allogeneic T cells/well for 5 days. [³H]-thymidine was added for an additional 18 h. (a) MLR induced by moDC cultured with medium, rhTNF-α, and CSFs from an MS, an NIND and a BM patient (representative experiment). Shown are means of triplicates ± s.e.m. ▪, Medium; ▴, TNF-α; ▾, MS CSF; ♦, NIND CSF; •, BM CSF. (b, c) Allogeneic MLR induced by moDC cultured with CSF from patients with MS (n = 16), ON (n = 8), BM (n = 5), LM (n = 5) and NIND (n = 7). DC:T ratio is 1:10 in (b) and 1:100 in (c). Data are pooled from three experiments and are expressed as percentage cpm induced by CSF- or TNF-α-treated moDC in relation to cpm induced by moDC cultured with medium in the same experiment (background cpm). Bars refer to medians. ***P < 0·001.

Fig. 3

Influence of CSF on the ability of moDC to induce differentiation of naïve T cells. Donor moDC were cultured for 24 h with either medium, or CSF supernatants (same as in Fig. 2), or TNF-α (100 ng/ml), washed and co-cultured for 5 days with naive allogeneic T cells (2 × 10⁴ DC per 2 × 10⁵ T cells in 96-well plates). Proliferating T cells were expanded with IL-2 for 6 days and restimulated with PMA/ionomycin. Intracellular cytokine expression (IFN-γ/IL-4 and IL-10) was analysed by flow cytometry. Results were obtained in three experiments and were expressed as percentages of cytokine-positive T cells. Bars refer to medians. *P < 0·05, **P < 0·01.

Fig. 4

Influence of CSF from different patient groups on the ability of donor moDC to produce IL-12p40 in allogeneic MLR. Donor moDC were cultured for 24 h with either medium, or CSF supernatants (same as in Figs 2 and 3), or rhTNF-α (100 ng/ml), washed and co-cultured for 60 h with naive allogeneic T cells (1:10). Supernatants were collected, and levels of IL-12p40 were analysed by ELISA. Bars refer to medians. *P < 0·05, **P < 0·01.

Fig. 5

Levels of IFN-γ and IL-10 (ELISA) in CSF from patients with BM (n = 5), LM (n = 5), MS (n = 16) and NIND (n = 7). Mean ± s.d. ▪, IFN-γ; □, IL-10.

Fig. 6

Effect of anti-TNF-α MoAb and rIL-10 on the ability of MS CSF and LM CSF to induce maturation of moDC. CSF from three MS and three LM patients was preincubated for 1 h either with no MoAb or with anti-TNF-α MoAb (10 μg/ml), and added to moDC (1/2, v/v). rIL-10 (1 ng/ml) was added directly to CSF/DC cultures. After 24 h, moDC were harvested, and expression of surface markers was analysed by flow cytometry. Data are expressed as percentage MFI after culture with CSF in relation to MFI after culture with medium alone (background MFI). ▪, MS; □, LM.