Studies on dendritic cells and cytokines in inflammatory diseases of the central nervous system

Abstract

Multiple sclerosis (MS) is characterised by chronic inflammation in the central nervous system (CNS), which results in primary demyelination, axonal loss and neurological deficit. MS is believed to be mediated by myelin antigen-reactive T cells, which after priming in the periphery enter the CNS and initiate demyelination. The cause of the amoaggressive T cell response in MS remains unclear. Consequently, there is no ewe in this disease. In most neuroinfections, on the other hand, the CNS inflammation resolves either spontaneously, or after appropriate antimicrobial therapy.

Dendritic cells (DCs) are key players in induction and regulation of T cell responses. DCs capture antigens in non-lymphoid tissues and, subsequently, migrate to regional lymphoid nodes, in order to activate naive T cells. Up to recently, this paradigm appeared not to be applicable to the CNS, due to presumed absence of DCs. However, recent animal studies have shown that DCs normally occur in the meninges and choroid plexus, and infiltrate inflamed brain parenchyma. DC migration from the CNS to deep cervical lymph nodes has also been demonstrated. These findings have made it likely that DCs are directly involved in T cell activation in MS as well as in neuroinfections.

The aim of this thesis was to explore the presence of DCs in human cerebrospinal fluid (CSF) and possible involvement of CSF DCs in regulation of immune responses in MS and bacterial neuroinfections (studies 3-6). Expression of pro-inflammatory cytokines (interferon (IFN)-gamma interleukin (IL)-12, IL-15) by blood DCs and mononuclear cells (MNCs) was also studied in MS patients and controls (studies 1-2).

Numbers of IL-15-expressing blood MNCs, presumably monocytes or DCs, were elevated in MS compared to healthy donors, as detected by immunocytochemistry. This elevation was confined to patients with primary and secondary progressive MS, and was not observed in the subgroup of relapsing-remitting MS. Thus, EL-15 might be of importance in clinically more advanced forms of MS.

Spontaneous production of cytokines by enriched blood DCs was analysed by flow cytometry. Cells expressing IFN-gamma, IL- 12 and IL- 15 but not IL- 10 were more frequent among enriched DC than among corresponding MNC samples. Human blood contains two DC subsets, namely myeloid (CD1 1c+) DC, which are potent antigen-presenting cells, and plasmacytoid (CD 11 c) DC, which can produce large amounts of type I IFNs upon infection with certain bacteria and enveloped viruses. Using flow cytometry, we found the same two DC subsets in the CSF DCs were present in minute numbers in CSF from patients with non-inflammatory neurological diseases. In MS, numbers of myeloid and plasmacytoid CSF DCs were increased, in particular during the early stage of the disease, inversely correlated with MS duration. The highest levels were observed in acute monosymptomatic optic neuritis (ON) with oligoclonal IgG in CSF, which usually represents the first bout of MS. In bacterial meningitis (BM), only numbers of myeloid DCs were consistently increased in CSF In Lyme meningoencephalitis (LM), there was a striking accumulation of myeloid and especially of plasmacytoid DCs in CSF Myeloid DCs in CSF expressed higher levels of HLA-DR, CD40, CD80 and CD86 than in blood and were thus wore mature, while plasmacytoid DCs in CSF and in blood were equally immature.

In BM and LM, DCs appeared to be recruited to the CSF compartment by chemokines present in CSF, like SW-1alpha, MCP-1 and C5a. MS patients had increased expression of CCR5 by myeloid blood DCs, which could facilitate their recruitment in response to CCR5 ligands expressed in MS brain (MIP-1alpha/beta, RANTES). CSF supernatants from BM patients inhibited the T cell-stimulatory and Th1-deviating capacities of myeloid DCs in vitro, likely due to the presence of IL-10, whereas CSF from LM and MS patients augmented these capacities of DCs.

Based on the present data and those of others, we suggest that CSF DCs participate in immune surveillance in the CNS and can activate T cells locally andlor in regional lymph nodes. The character of the T cell response induced by CSF DCs may depend on DC numbers, cytokine milieu and, in MS, on inherited properties of DCs. Further studies are warranted to investigate, whether repeated presentation of myelin antigens by CSF DC triggers the chronic autoimmune response in MS.