|
|
||||||||


* Department of Microbiology and Immunology,
Interdepartmental Graduate Program in Neuroscience, and
Department of Neurology, University of Rochester School of Medicine and Dentistry, Rochester, NY 14642
| Abstract |
|---|
|
|
|---|
| Introduction |
|---|
|
|
|---|
Despite their importance in initiating and perpetuating adaptive immune responses in the periphery, CD11c+ dendritic cells (DCs) were previously assumed not to be important in neuroinflammation because they are absent from the CNS parenchyma during homeostasis (4). The CNS is one of the few tissues bereft of a lymphatic vasculature capable of supporting DC trafficking. Nevertheless, several recent studies have demonstrated that CD11c+ leukocytes with morphological characteristics of DCs are prominent components of CNS infiltrates in neuroinflammatory states including EAE and MS (5, 6). Some studies suggest that these cells might be particularly efficient in supporting the activation and effector functions of encephalitogenic T cells, while others indicate that they are incompetent in priming naive as well as in expanding memory myelin-specific T cells (5, 6, 7). In light of this controversy, we assessed the cell surface and functional characteristics of CNS CD11c+ cells at EAE onset and peak and compared them to those of CD11cCD11b+ macrophages/microglia (the other major myeloid subset in EAE lesions) as well as splenic and bone marrow-derived DCs (BMDCs).
| Materials and Methods |
|---|
|
|
|---|
Wild-type and CD45.1 congenic C57BL/6 mice were purchased from National Cancer Institute-Frederick (Frederick, MD). C57BL/6-Tg (Tcra2D2, Tcrb2D2)1Kuch/J mice (transgenic for the TCR V
3.2 and V
11 chains reactive to myelin oligodendrocyte glycoprotein (MOG)3555), commonly known as 2D2 mice, were a gift from Dr. V. Kuchroo (Harvard Medical School, Boston, MA) (8). Experiments were performed using University Committee on Animal Resources-approved protocols.
Reagents
MOG3555 (MEVGWYRSPFSRVVHLYRNGK) was synthesized by Macromolecular Resources and purified by HPLC. Oligodeoxynucleotides (ODN) were phosphorothioate-modified to increase their resistance to nuclease degradation (Operon Technologies). The sequences were 5'-ATAATCGACGTTCAAGCAAG-3' for CpG ODN (CpG1760) and 5'-ATAATAGAGCTTCAAGCAAG-3' for control ODN (CpG1908).
Abs/fusion proteins used for FACS and whole-mount immunofluorescence were purchased from BD Biosciences except for IgG 2a,
-APC (eBioscience).
Generation of bone marrow chimeras
C57BL/6 bone marrow cells (10 x 106) were injected into the tail veins of lethally irradiated (1000 rad) CD45.1 congenic hosts. In all cases, the reconstitution efficiency of CD11b+ cells exceeded 95% as assessed by flow cytometry at 5 wk after irradiation.
Induction and evaluation of EAE
Mice were immunized with MOG3555 (150 µg) in CFA (5 mg/ml heat-killed Mycobacterium tuberculosis H37Ra; v/v) by s.c. injection at four sites over the flanks. Bordetella pertussis toxin (List Laboratories) was injected i.p. (2 ng/mouse) on days 0 and 2. Mice were observed for signs of EAE on a daily basis and graded on a standard scale as previously described (9).
CD4+ T cell and myeloid subset purification/preparation
CD4+ T cells were purified from lymph node cells with CD4 T cell enrichment columns (R & D Systems). CD11c+ cells were purified from splenocytes with anti-CD11c microbeads (Miltenyi Biotec). CD11b+CD11c and CD11c+ cells were FACS sorted from CNS mononuclear cell (MNC) fractions, as were CD4+CD62Lhigh cells from 2D2 splenocytes. All isolated cells were 9398% pure. BMDCs were induced by culture with GM-CSF and IL-4 (10).
Cell cultures
Myeloid cells were cultured in 24-well plates (2.5 x 105 cells/ml) with either anti-CD40 mAb (FGK45; (11)), isotype matched rat IgG (5 µg/ml), CpG1760, or control ODN (100 nM). Naive 2D2 cells or MOG-primed CD4+ T cells (2 x 105 cells/well) were cultured with APCs (105 cells/well) in the presence or absence of MOG3555 (50 µg/ml).
Cytokine ELISA
All cytokines were quantified using a sandwich ELISA technique based on noncompeting pairs of Abs. Capture and detection mAbs were obtained from BD Biosciences except for the anti-IL-23 p19 mAb (gift of J. Benson, Centocor).
Whole-mount immunofluorescence
Spinal cord sections (3 x 2 mm) were stained with labeled mAbs and viewed under a fluorescent microscope (Olympus BX40F) as previously described (9). Background staining with isotype-matched control mAbs was minimal.
| Results |
|---|
|
|
|---|
In the initial experiments, mice were sacrificed shortly after the onset of clinical EAE to analyze CNS CD11c+ cells and other APC subsets at a time point when local T cell activation is driving the development of inflammatory infiltrates. In repeated experiments CD11c+CD11b+ cells, consistent with myeloid DCs, comprised between 16 and 25% of live CNS mononuclear cells as determined by flow cytometric analysis (Fig. 1A, upper left panel). By contrast, CD11c+CD11b cells were scarce (28%); those detected expressed either CD8
or Gr-1, indicative of alternative DC subsets (P. Deshpande and B. Segal, unpublished observations). The majority of the remaining cells were CD11cCD11b+ macrophages/microglia (2050%) and CD4+ T cells (2030%).
|
Next, we analyzed intact EAE lesions by whole-mount immunofluorescence to assess the spatial relationship between DC-like cells and T cells. Consistent with the flow cytometry data, virtually all CD11c+ cells coexpressed the myeloid marker CD11b (Fig. 1A, lower panels). CD11c+ cells formed clusters with CD4+ T cells adjacent to CD31+ blood vessels (Fig. 1A, upper right panel). Hence, CNS myeloid DC-like cells are well positioned to activate myelin-specific T cells in situ.
CNS DCs originate from the peripheral pool of circulating leukocytes
Microglia can exhibit DC-like morphology, phenotype, and T cell-activating properties following the activation of CD40 or TLRs in vitro (5). To distinguish whether CD11c+ cells in EAE lesions arise from CNS resident microglia or circulating hemopoietic cells, we constructed bone marrow chimeras with CD45.1 congenic donors and CD45.2 hosts. As expected, the vast majority of CD11b+ cells in spinal cord mononuclear fractions from unimmunized chimeras, representative of microglia, were of host origin (Fig. 1C). In contrast, >95% of CD11c+ and CD11b+ cells in spinal cord mononuclear fractions from MOG-immunized mice sacrificed at peak EAE were of donor origin, suggesting that these cells or their precursors infiltrated the CNS from the circulation (Fig. 1D).
CNS CD11c+ cells from early EAE infiltrates stimulate naive MOG-specific T cells to divide
Mature myeloid DCs in bone marrow, lymphoid organs, and other tissues external to the CNS are distinguished by their ability to prime naive T cells. However, there is conflicting data regarding the outcome of interactions between CNS DCs and CD4+ T cells (2, 5, 6, 7). To investigate the biological properties of CNS DCs in our experimental system, we purified CD11c+ cells from spinal cords within several days of EAE onset and cocultured them with CFSE-labeled, naive (CD62L+CD25CD44) MOG-specific CD4+ (2D2) T cells. For comparative purposes, 2D2 cells were cocultured with CNS CD11cCD11b+ and splenic CD11c+ cells from the same donors in parallel. As shown in Fig. 2, 2D2 cells underwent multiple rounds of division upon culture with MOG-pulsed CNS CD11c+ cells, a response approximating that elicited by CD11c+ splenocytes and exceeding that elicited by T cell-depleted splenocytes. By contrast, relatively few 2D2 cells proliferated upon coculture with CNS CD11cCD11b+ macrophages and Ag, even at high APC to T cell ratios (Fig. 2, lower panels). A similar pattern was observed when OVA-specific TCR transgenic cells were combined with the same APC populations (not shown).
|
In addition to activating naive CD4+ T cells, peripheral myeloid DCs direct Th lineage commitment via the secretion of polarizing cytokines. Encephalitogenic CD4+ T cells that accumulate in EAE lesions predominantly fall within the Th17 and Th1 subsets (8, 9). The IL-12 p40 monokines IL-12 and IL-23, promote/stabilize Th1 and Th17 differentiation, respectively. Therefore we measured IL-12 and IL-23 production by CNS CD11c+ cells and other APCs in response to various inflammatory stimuli and during cognate interactions with T cells.
Approximately 50% of CD11c+CD11b+ cells and 17% of CD11cCD11b+ cells expressed intracellular IL-12 p40 when analyzed during the early phases of clinical EAE (Fig. 3A). Both subsets secreted IL-12 and IL-23 heterodimers following short-term stimulation with a CpG-containing ODN, although CD11c+ cells consistently secreted significantly larger quantities (Fig. 3B). CNS CD11c+, but not CD11cCD11b+, cells secreted IL-12 and IL-23 in response to CD40 ligation. Similarly, CNS CD11c+ cells produced significantly more IL-12 and IL-23 than CNS CD11cCD11b+ cells during coculture with either naive 2D2 cells or primed polyclonal MOG-specific CD4+ T cells in the presence of Ag (Fig. 3, C and D).
|
To directly compare the relative capacities of CNS APC subsets to direct Th differentiation of autoreactive CD4+ T cells, we measured IL-2, IL-17, IFN-
, and GM-CSF levels in supernatants from the same cocultures as those illustrated in Fig. 3C. CNS CD11c+ cells were essentially as effective as splenic DC in inducing MOG-specific IFN-
and GM-CSF production during culture with naive 2D2 cells (Fig. 4A). They also stimulated the naive T cells to secrete substantial quantities of IL-17. In contrast, CNS CD11cCD11b+ cells induced little if any cytokine production beyond baseline levels. Consistent with the proliferation data shown in Fig. 2, only CD11c+ CNS cells stimulated significant IL-2 secretion.
|
induce Th17 differentiation of uncommitted T cells (8). We detected TGF-
at comparable concentrations across all cocultures, irrespective of APC phenotype or the presence or absence of Ag (Fig. 4B). By contrast, IL-6 was preferentially up-regulated in cultures that contained CNS CD11c+ cells or BMDCs. Hence, IL-6 production may be the pivotal factor that predisposes CNS CD11c+ cells to induce Th17 differentiation.
We also assessed CNS APCs with regard to their ability to reactivate primed myelin-specific T cells. Both CD11c+ and CD11c myeloid subsets stimulated CD4+ T cells from MOG3555-immunized mice to secrete IFN-
, IL-17, and GM-CSF in an Ag-specific manner; the levels of IL-2 and GM-CSF were consistently higher in supernatants from cocultures with CD11c+ cells (data not shown).
CD11c+ cells harvested from the CNS immediately before the remission phase of EAE are relatively inefficient APCs
MOG-induced EAE characteristically follows a monophasic course in C57BL/6 mice, reminiscent of idiopathic transverse myelitis, optic neuritis, and some forms of acute disseminated encephalomyelitis in humans. Little is known about the cellular and molecular events underlying the transition between the active inflammatory and remitting stages of autoimmune demyelination. We questioned whether the CNS CD11c+ subpopulation evolves as acute disease advances, losing T cell stimulatory potency at the peak of clinical disability and thereby curbing the autoimmune response preparatory to the remission phase. To test this hypothesis we analyzed CNS CD11c+ cells for the expression of maturation markers as well as for the ability to activate CD62LhighCD44CD25 2D2 T cells when harvested shortly following clinical presentation vs during peak disease. Splenic CD11c+ cells were used as a control APC population. CNS CD11c+ cells harvested on day 18 postimmunization, immediately before the expected onset of remission, expressed relatively low levels of CCR7, CD86, and MHC class II, suggestive of an immature phenotype (Fig. 5B). Furthermore, these cells were clearly and consistently less efficient than the CNS CD11c+ cells harvested during early EAE in stimulating naive myelin-specific T cells to expand (Fig. 5A) or secrete proinflammatory cytokines (data not shown) in vitro.
|
In addition to losing optimal T cell stimulatory capacity during disease progression, we considered the possibility that CNS DCs actively engage immunoregulatory pathways earlier in the course to limit the extent of CNS inflammation and to set the stage for subsequent remission. CD25+CD4+ FoxP3+ T cells (Treg) begin accumulating in the CNS before peak disease; depletion of these cells before EAE induction exacerbates severity and delays recovery (12). We questioned whether mature CNS CD11c+ cells could support CD25+CD4+ T cell (Treg)-mediated suppression. It was previously reported that LPS-activated splenic DCs overcome CD25+ Treg-mediated suppression of conventional CD4+ T cells (10, 13). In contrast, we found that Tregs retained their suppressive properties in the presence of CNS CD11c+ cells (Fig. 5C).
| Discussion |
|---|
|
|
|---|
CNS DCs express lower levels of MHC class II and costimulatory molecules at peak disease than at disease onset, providing one possible explanation for their reduced APC potency over time (Fig. 5B). As yet, we have no evidence that they secrete immunosuppressive cytokines (Fig. 4B; data not shown). We have not established whether CNS DCs down-regulate stimulatory functions on an individual basis or whether mature DCs that dominate the infiltrating myeloid population at EAE onset are replaced by immature DCs at peak disease. We found that CNS CD11c+ cells are derived from hemopoietic cells that infiltrate the CNS from the circulation (Fig. 1D). Based on that observation and taking into account the relatively short lifespan of DCs following migration into peripheral tissues, we believe that the latter scenario is most likely. The factors that govern DC differentiation and maturation status within the CNS microenvironment is a topic of continuing interest and research in our laboratory.
There is growing evidence that DCs accumulate in the CNS during autoimmune demyelinating diseases in humans such as MS (6). Therefore, our findings may have relevance regarding the immunopathogenesis and treatment of those disorders. Interventions that accelerate the conversion of CNS DCs from an activated to an immature state might be therapeutically useful in MS.
| Disclosures |
|---|
|
|
|---|
| Footnotes |
|---|
1 This work was supported by National Multiple Sclerosis Society Grant RG 3866-A-3 and National Institutes of Health Grant NS047687-01A1. ![]()
2 Address correspondence and reprint requests to Dr. Benjamin M. Segal, Department of Neurology/Neuroimmunology, University of Rochester School of Medicine, 601 Elmwood Avenue, Box 605, Rochester, NY 14642. E-mail address: Benjamin_Segal{at}urmc.rochester.edu ![]()
3 Abbreviations used in this paper: EAE, experimental autoimmune encephalomyelitis; DC, dendritic cell; BMDC, bone marrow DC; MS, multiple sclerosis; MNC, mononuclear cell; MOG, myelin oligodendrocyte glycoprotein; ODN, oligodeoxynucleotide; Treg, CD4+CD25+ regulatory T cell. ![]()
Received for publication February 20, 2007. Accepted for publication March 27, 2007.
| References |
|---|
|
|
|---|
heavy chain class switching. Immunity 5: 319-330. [Medline]This article has been cited by other articles:
![]() |
I. L. King, T. L. Dickendesher, and B. M. Segal Circulating Ly-6C+ myeloid precursors migrate to the CNS and play a pathogenic role during autoimmune demyelinating disease Blood, April 2, 2009; 113(14): 3190 - 3197. [Abstract] [Full Text] [PDF] |
||||
![]() |
A. L. Zozulya, S. Ortler, J. Lee, C. Weidenfeller, M. Sandor, H. Wiendl, and Z. Fabry Intracerebral Dendritic Cells Critically Modulate Encephalitogenic versus Regulatory Immune Responses in the CNS J. Neurosci., January 7, 2009; 29(1): 140 - 152. [Abstract] [Full Text] [PDF] |
||||
| |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |