Selective depletion of CD11c⁺CD11b⁺ dendritic cells partially abrogates tolerogenic effects of intravenous MOG in murine EAE

Abstract

Intravenous (i.v.) injection of a soluble myelin antigen can induce tolerance, which effectively ameliorates experimental autoimmune encephalomyelitis (EAE). We have previously shown that i.v. MOG induces tolerance in EAE and expands a subpopulation of tolerogenic CD11c⁺CD11b⁺ dendritic cells (DCs) with an immature phenotype having low expression of IA and co-stimulatory molecules CD40, CD86, and CD80. Here we further investigate the role of tolerogenic DCs in i.v. tolerance by injecting clodronate-loaded liposomes, which selectively deplete CD11c⁺CD11b⁺ and immature DCs, but not CD11c⁺CD8⁺ DCs and mature DCs. I.v. MOG-induced suppression of EAE was partially, yet significantly, blocked by CD11c⁺CD11b⁺ DC depletion. While i.v. MOG inhibited IA, CD40, CD80, CD86 expression and induced TGF-β, IL-27, IL-10 production in CD11c⁺CD11b⁺ DCs, these effects were abrogated after injection of clodronate-loaded liposomes. Depletion of CD11c⁺CD11b⁺ DCs also precluded i.v. autoantigen-induced T-cell tolerance, such as decreased production of IL-2, IFN-γ, IL-17 and numbers of IL-2⁺, IFN-γ⁺, and IL-17⁺ CD4⁺ T cells, as well as an increased proportion of CD4⁺CD25⁺Foxp3⁺ regulatory T cells and CD4⁺IL-10⁺Foxp3⁻ Tr1 cells. CD11c⁺CD11b⁺ DCs, through low expression of IA and co-stimulatory molecules as well as high expression of TGF-β, IL-27 and IL-10, play an important role in i.v. tolerance-induced EAE suppression.

Introduction

Multiple sclerosis (MS) is a T-cell mediated autoimmune disease of the central nervous system (CNS) that is manifested clinically as weakness and progressive paralysis [1]. Experimental autoimmune encephalomyelitis (EAE), induced by immunization of susceptible mouse strains with myelin oligodendrocyte glycoprotein peptides (MOG) or other myelin components, provides a useful animal model for MS research [2, 3]. Myelin-reactive encephalitogenic Th1 and Th17 cells are critically involved in the initiation and development of EAE [4, 5]. On the other hand, Th2, regulatory T cells (Treg cells) and Tr1 cells are considered protective [6].

Intravenous (i.v.) injection of a soluble myelin antigen that has been used for EAE induction leads to the antigen-specific tolerance, which effectively ameliorates EAE [7]. Clonal deletion and anergy of antigen-specific Th1/Th17 cells and induction of regulatory T cells are the main mechanisms involved in the induction of i.v. tolerance [8].

Antigen presenting cells (APCs), including macrophages and dendritic cells (DCs), are important for Th cell differentiation [9, 10]. APCs provide Th cells not only with antigen stimulation (Signal 1) and co-stimulatory signals (Signal 2), but also with additional polarizing signals (Signal 3), such as inflammatory cytokines IL-12 (for Th1), IL-23 (for Th17), and immunomodulatory cytokines TGF-β, IL-27 and IL-10 (for T regulatory cells) [9, 11, 12]. DCs, as the most potent professional APCs, play an essential role in Th cells differentiation and thus are involved in the induction of tolerance [13, 14]. We have shown that i.v. MOG-induced tolerance in EAE mice is associated with an increased proportion of the CD11c⁺CD11b⁺ subpopulation of DCs, while a higher proportion of the CD11c⁺CD8⁺ subpopulation was observed in non-tolerized EAE mice [15]. These CD11c⁺CD11b⁺ DCs exhibited an immature phenotype, produced high levels of IL-10 and TGF-β, and effectively suppressed EAE upon adoptive transfer, demonstrating their tolerogenic nature [15].

In the present study, we have evaluated the role of CD11c⁺CD11b⁺ DCs in i.v. tolerance induction in EAE by depleting this DC population. Clodronate-loaded liposomes selectively deplete immature DCs (iDCs), but only minimally affect the mature DC population [16]. As CD11c⁺CD11b⁺ DCs exhibited an immature phenotype, we intraperitoneally (i.p.) injected clodronate-loaded liposomes to deplete CD11c⁺CD11b⁺ DCs in MOG-i.v. treated EAE mice and then defined their outcome clinically and immunologically. Indeed, the effects of i.v. MOG-injected tolerance were significantly abrogated after CD11c⁺CD11b⁺ DC depletion, demonstrating an important role of this DC population in i.v. tolerance induction.

RESULTS

Clodronate-loaded liposomes selectively deplete CD11c⁺CD11b⁺ DCs or iDCs

We first tested the efficacy of iDC depletion. Clodronate- or PBS-loaded liposomes were i.p. injected into naïve C57BL/6 mice; splenocytes were harvested 24 hours later and analyzed by flow cytometry as described [17]. Compared with PBS-loaded liposomes, clodronate-loaded liposomes largely depleted the DCs (CD11c⁺ cells, 4.07% vs. 2.05% among total splenocytes) and F4/80⁺ macrophages (16.0% vs. 4.67%; data not shown). When CD11b and CD8 markers were analyzed in gated CD11c⁺ DCs, a reduced percentage of CD11b⁺ (51.8% vs. 40.0%, P<0.001) was observed, with virtually no change in the CD8⁺ (20.7% vs. 22.8%) and an increased percentage of CD11b⁻ CD8⁻ DCs (Fig. 1A, B). Compared to those in PBS-treated mice, the absolute numbers of total CD11⁺ DCs and CD11b⁺CD11c⁺ DCs per spleen were significantly reduced in clodronate-treated mice (1.34±0.02 × 10⁶ vs. 0.88±0.01 × 10⁶, P<0.01; 0.70±0.01 × 10⁶ vs. 0.36±0.01 ×10⁶, P<0.001, respectively). The absolute number of CD8⁺CD11c⁺ DCs was also reduced (0.27±0.01 × 10⁶ vs. 0.21±0.01 ×10⁶, P<0.01), while there was no significant difference for the numbers of CD11b⁻CD8⁻CD11c⁺ DCs between the two groups (0.34±0.01 × 10⁶ vs. 0.32±0.01 × 10⁶). Thus, while clodronate-loaded liposomes reduced the numbers of total DCs and all DC subpopulations, the major reduction was in the CD11b⁺CD11c⁺ population.

Fig. 1. Clodronate-loaded liposomes selectively deplete CD11c⁺CD11b⁺ DCs or iDCs.

Naïve C57BL/6 mice were i.p. injected with PBS-loaded or clodronate-loaded liposomes, splenocytes were harvested at 24 hours later and analyzed with flow cytometry. (A) Percentages of DCs (CD11c⁺) among splenocytes (left). Results were statistically analyzed and shown as mean ± SEM (n=3 each group) (right). (B) CD11c⁺ cells were then gated and percentages of CD11b⁺ and CD8⁺ cells were analyzed (left). Results were statistically analyzed and shown as mean ± SEM (n=3 each group) (right). (C) Expression of CD11c vs. CD83, CD80, and CD86 in splenocytes (left). Results were statistically analyzed and shown as mean ± SEM (n=3 each group) (right). (D) Expression of CD83, CD80, and CD86 in total CD11c⁺ cells and in CD11b⁺, CD8⁺, CD11b⁻CD8⁻ cells of gated CD11c⁺ cells. *p<0.05, ***p<0.001, unpaired Student’s t test. One representative of 3 experiments is shown.

Clodronate-loaded liposomes selectively reduced numbers (~3 fold) of CD83^low/−, CD80^low/− and CD86^low/− iDCs, but minimally affected the CD83^hi/+, CD80^hi/+ and CD86^hi/+ mature DCs (Fig. 1C). Mature markers of DCs, CD83, CD80, and CD86 in total CD11c⁺ cells and in CD11b⁺, CD8⁺, CD11b⁻CD8⁻ cells of gated CD11c⁺ cells were analyzed. As shown in Fig. 1D, CD11c⁺CD11b⁺ cells contained many more immature DCs and were reduced by clodronate (CD83^low/−: 91.6% vs. 39.4%, CD80^low/−: 83.3% vs. 39.4%, CD86^low/−: 91.4 vs. 38.9%), while these markers on CD8⁺, CD11b⁻CD8⁻ DCs either remained unchanged or were slightly reduced. Thus, these data confirm an immature phenotype of CD11c⁺CD11b⁺ DCs, and the efficacy of clodronate-loaded liposomes in depleting these DCs.

Clodronate-loaded liposomes partially block i.v. MOG-induced EAE suppression

We investigated the effect of clodronate-loaded liposomes in i.v. tolerance induction in EAE. Mice were immunized with MOG_35–55 in CFA to induce EAE. MOG_35–55 peptide at 200 µg/mouse was i.v. injected to induce tolerance at days 0, 3, and 6 post immunization (p.i.), and PBS-i.v. served as control. Clodronate-loaded liposomes were i.p. injected 24 hours before and 14 days p.i., while mice that received PBS-loaded liposomes served as control. The onset of disease in control EAE mice (PBS-i.v. +PBS-i.p.) was observed at day 12–14 p.i., reaching the maximum clinical score by day 21 p.i. In contrast, a mild clinical signs of EAE were observed in MOG-i.v. tolerized mice (MOG-i.v. +PBS-i.p.). Injection of clodronate-loaded liposomes to tolerized mice (MOG-i.v.+clodronate-i.p.) partially, but significantly, blocked i.v. MOG-induced EAE suppression (Fig. 2A). While injection of clodronate-loaded liposome to untreated EAE mice (PBS-i.v. +clodronate-i.p.) did not influence clinical scores (Fig. 2A).

Fig. 2. Clodronate-loaded liposome partly block i.v. MOG-induced EAE suppression.

Mice were immunized with MOG_35–55 in CFA to induce EAE. MOG_35–55 or PBS was i.v. injected at days 0, 3, and 6 p.i. Clodronate- or PBS-loaded liposome was i.p. injected 24 hours before and 14 days p.i. (A) Clinical signs were scored daily following a 0–5 scale. Data were pooled from three independent experiments and presented as mean clinical score ± SEM (n=11–15 each group). (B) On day 21 p.i., spinal cords were harvested and 5-µm sections were stained with H&E and LFB. Demyelination areas are designated by arrows. (Magnifications, × 10). (C) Inflammation and demyelination were scored using a 0–3 scale and expressed as mean scores ± SEM (n=3 each group). *p<0.05, ***p<0.001, two-way ANOVA test. One representative of 3 experiments is shown.

Consistent with clinical signs, extensive inflammatory infiltrates and demyelination in the white matter of spinal cords were seen in untreated EAE mice, while they had been significantly reduced in tolerized mice (Fig. 2B, C). On the other hand, injection of clodronate-loaded liposomes to tolerized mice significantly increased their inflammation and demyelination scores, while the same injection did not affect scores in EAE mice (Fig. 2B, C). Thus, clodronate-loaded liposomes blocked i.v. MOG-induced inhibition of EAE both in clinical severity as well as CNS inflammation and demyelination.

DCs induced by i.v. tolerance have iDC phenotype, and are depleted by clodronate-loaded liposomes

To test the depletion effect of clodronate-loaded liposomes on DC phenotype of EAE mice in vivo, splenocytes were isolated at day 21 p.i. and analyzed by flow cytometry. The percentages of macrophages (F4/80⁺ cells) and DCs (CD11c⁺ cells) in spleen cells of PBS-i.v.+clodronate-i.p. mice were lower than in untreated EAE (PBS-i.v.+PBS-i.p.) mice. Compared with untreated EAE mice, MOG-i.v. tolerized mice had decreased percentages of total macrophages (F4/80⁺; Fig. 3A) and both type 1 (M1; F4/80⁺IA⁺) and type 2 (M2; F4/80⁺CD206⁺) subtypes of these cells (Fig. 3B). Similarly, the proportion of CD11c⁺CD11b⁺ DCs was increased in tolerized mice, while this increase was significantly reduced after clodronate-loaded liposome injection (Fig. 3C, D). These results indicate that clodronate-loaded liposomes depleted both M2 macrophages and tolerogenic DCs in i.v. tolerized mice, and this effect may contribute to the observed blockade of MOG-induced EAE suppression. Interestingly, two reports showed that intracerebroventricular injection of clodronate-loaded liposomes effectively eliminated perivascular macrophages within the CNS [18, 19]. Whether i.v. injection of clodronate-loaded liposomes in our experiments had similar effect, and whether this contributed to the blockade of i.v. tolerance induction, remains to be determined in future studies.

Fig. 3. I.v. tolerance induced increase of CD11c⁺CD11b⁺ DCs is abrogated by injection of clodronate-loaded liposomes.

Mice immunized for EAE induction were i.v. injected with MOG or PBS, while PBS-loaded or clodronate-loaded liposomes were i.p. injected. Splenocytes were isolated 21 days p.i. and analyzed by flow cytometry. (A) FACS profile of macrophages (F4/80⁺ cells) in total splenocytes (left). Results were statistically analyzed and shown as mean ± SEM (n=3 each group) (right). (B) M1 (IA⁺) and M2 (CD206⁺) macrophages in gated F4/80⁺ cells (left). Results were statistically analyzed and shown as mean ± SEM (n=3 each group) (right). (C) DCs (CD11c⁺ cells) in total splenocytes (left). Results were statistically analyzed and shown as mean ± SEM (n=3 each group) (right). (D) CD11c⁺CD11b⁺ DCs (CD11b⁺ cells in gated CD11c⁺ cells) and CD11c⁺CD8⁺ DCs (CD8⁺ cells in gated CD11c⁺ cells) (left). Results were statistically analyzed and data shown as mean ± SEM (n=3 each group). *p<0.05, ** p<0.01, ***p<0.001, two-way ANOVA test. One representative of 3 experiments is shown.

To further determine DC phenotype after i.v. tolerance and iDC depletion, splenocytes and CNS-infiltrating cells were analyzed at day 21 p.i. and the expression of IA and co-stimulatory molecules CD40, CD86, CD80 on CD11c⁺CD11b⁺ DCs was determined by flow cytometry. Compared to CD11c⁺CD11b⁺ DCs in untreated EAE mice, those in MOG-i.v. tolerized mice expressed lower levels of CD40, CD86, CD80 and IA, indicating an iDC phenotype, while the percentages of DCs expressing these molecules were upregulated after injection of clodronate-loaded liposomes to MOG-i.v. tolerized mice (Fig. 4). These results, together with those shown in Figure 3, indicate that clodronate-loaded liposomes significantly depleted i.v. autoantigen-induced iDC phenotype of CD11c⁺CD11b⁺ DCs.

Fig. 4. I.v. MOG-induced tolerogenic phenotype of CD11c⁺CD11b⁺ DCs is abrogated by injection of clodronate-loaded liposomes.

Mice immunized for EAE induction were i.v. injected with MOG or PBS, while PBS-loaded or clodronate-loaded liposomes were i.p. injected. Splenocytes and CNS infiltrating cells were isolated 21 days p.i. and analyzed by flow cytometry. (A) Expression of CD40, CD80, CD86 and IA in gated CD11c⁺CD11b⁺ DCs in splenocytes (left) and CNS infiltrating cells (right). (B) Results in (A) were analyzed and shown as mean ± SEM (n=3 each group). *p<0.05, ***p<0.001, two-way ANOVA test. One representative of 3 experiments is shown.

iDC depletion reduces i.v. MOG-induced IL-27 and TGF-β production

We then determined cytokine production of DCs after CD11c⁺CD11b⁺ or iDC depletion. Compared to splenocytes of untreated EAE mice, cells of MOG-i.v. mice produced significantly higher levels of IL-27 and TGF-β in culture supernatants, while these increases were reduced after injection of clodronate-loaded liposomes (Fig. 5A). When intracellular cytokine secretion in splenic CD11c⁺CD11b⁺ DCs was analyzed by flow cytometry, increased proportions of IL-27⁺and TGF-β⁺ DCs were observed in MOG-i.v. mice compared to untreated EAE mice, while injection of clodronate-loaded liposomes reduced their numbers (Fig. 5B). These results, together, showed that the production of i.v. MOG-induced immunomodulating cytokinesIL-27 and TGF-β can be reduced by depleting CD11c⁺CD11b⁺ DCs.

Fig. 5. iDC-depletion inhibits i.v. MOG-induced IL-27 and TGF-β production.

Mice immunized for EAE induction were i.v. injected with MOG or PBS, while PBS-loaded or clodronate-loaded liposomes were i.p. injected. Splenocytes were isolated 21 days p.i. and cultured in triplicate in 24-well plates with MOG_35–55 for 24, 48 or 72 hours. (A) Supernatants were collected and analyzed for IL-27and TGF-β production by ELISA and shown as mean ± SEM (n=4 each group). (B) Intracellular secretion of TGF-β and IL-27 in gated CD11c⁺CD11b⁺ DCs was determined by flow cytometry. *p<0.05, ***p<0.001, two-way ANOVA test. One representative of 3 experiments is shown.

Effects of CD11c⁺CD11b⁺DC depletion on i.v. MOG-induced T-cell responses

To study the effect of CD11c⁺CD11b⁺ DC depletion on T cell response after i.v. MOG-induced tolerance, cytokine production of splenocytes was analyzed in vitro. Production of these cytokines by MOG-i.v. tolerized mice compared to untreated EAE mice was significantly reduced, while injection of clodronate-loaded liposomes partially, but significantly, blocked this effect (Fig. 6A). Similar results were obtained for the percentages of Th1 cells (IL-2⁺CD4⁺, IFN-γ⁺CD4⁺) and Th17 cells (IL-17⁺CD4⁺) of splenocytes and CNS-infiltrating cells, as determined by flow cytometry (Fig. 6B). In contrast, numbers of both CD4⁺CD25⁺Foxp3⁺ Treg cells and CD4⁺IL-10⁺Foxp3⁻ Tr1 cells [20, 21] in splenocytes were higher in MOG-i.v. tolerized mice than in untreated EAE mice, and the levels were significantly reduced after clodronate-loaded liposome injection (Fig. 6C). Thus, MOG-induced T cell tolerance is significantly diminished by CD11c⁺CD11b⁺ DC depletion.

Fig. 6. I.v. MOG-induced suppression of Th1/Th17 cells and increase of Treg cells and Tr1 cells is blocked by CD11c⁺CD11b⁺ DC depletion.

Mice immunized for EAE induction were i.v. injected with MOG or PBS, while PBS-loaded or clodronate-loaded liposomes were i.p. injected. Splenocytes and CNS infiltrating cells were isolated 21 days p.i. (A) Splenocytes triplicate cultured in 24-well plates restimulated with MOG_35–55 for 24, 48 or 72 hours and supernatants were collected and analyzed for IL-2, IL-17, and IFN-γ production by ELISA and shown as mean ± SEM (n=4 each group). (B) Intracellular secretion of IL-2, IFN-γ and IL-17 in gated CD4⁺ T cells in splenocytes (top) and CNS infiltrating cells (bottom) were determined by flow cytometry. (C) Treg cells and Tr1 cells in splenocytes were determinate as CD25⁺ Foxp3⁺ and IL-10⁺ Foxp3⁻ cells, respectively, in gated CD4⁺ T cells (left). Results were statistically analyzed and data shown as mean ± SEM (n=3 each group) (right). *p<0.05, ***p<0.001,***p<0.001, two-way ANOVA test. One representative of 3 experiments is shown.

Effect of iDC depletion on MOG-reactive T cells in co-cultures

To test the direct effect of iDC depletion on T cell responses, CD11c⁺ cells were purified from the spleen of MOG-i.v. + PBS-i.p. mice and MOG-i.v. + clodronate-i.p. mice, then co-cultured with purified CD4⁺ T cells from 2D2 MOG TCR transgenic mice in the presence of MOG_35–55 peptide for 3 days. As shown in Fig. 7A, reduced numbers of IL-27-and IL-10-producing CD11c⁺CD11b⁺DCs were observed in MOG-i.v. + clodronate-i.p. mice compared to MOG-i.v.+ PBS-i.p. mice. CD4⁺T cells co-cultured with DCs of MOG-i.v.+ clodronate-i.p. mice had higher proportions of IL-17⁺ and IFN-γ⁺ cells and lower proportions of IL-10⁺Foxp3⁻ cells (Fig. 7B). Supernatants of CD4⁺ T cells co-cultured with DCs of MOG-i.v.+ clodronate-i.p. mice had significantly lower levels of immunomodulatory cytokines IL-27and IL-10, but higher concentrations of proinflammatory cytokines IL-2, IFN-γ and IL-17 compared to CD4⁺T cells co-cultured with DCs of MOG-i.v.+ PBS-i.p. mice (Fig. 7C). Further, CD4⁺ T cells co-cultured with DCs of MOG-i.v. + clodronate-i.p. mice exhibited significantly higher proliferative responses to MOG_35–55 peptide stimulation (Fig. 7D). Together, these results indicate that iDC depletion significantly reduced MOG-i.v. induced tolerogenic functionality of CD11c⁺CD11b⁺ DCs.

Fig. 7. Effect of iDC depletion on MOG-reactive T cells in co-culture.

CD4⁺ T cells from naïve spleens of MOG TCR transgenic mice (2D2) were co-cultured with DCs from spleen of EAE mice treated with MOG-i.v.+PBS-i.p. (MOG-i.v.+PBS-i.p. DCs & 2D2 CD4) or MOG-i.v.+clodronate-i.p. (MOG-i.v.+clodronate-i.p. DCs & 2D2 CD4) in the presence of MOG_35–55 for 3 days. (A) CD11c⁺ cells were gated and the intracellular secretion of IL-27 and IL-10 in CD11c⁺CD11b⁺ cells were analyzed (up) and expressed as mean ± SEM (n=3 each group; bottom). (B) CD4⁺ cells were gated and their intracellular IFN-γ, IL-17, Foxp3 and IL-10 were analyzed with flow cytometry (top) and data are shown as mean ± SEM (n=3 each group; bottom). (C) Culture supernatants were harvested and production of IL-27, IL-10, IL-2, IFN-γ, IL-17 was analyzed with ELISA. Data are expressed as mean ± SEM (n=4 each group). (D) MOG-induced T-cell proliferative responses were measured by ³H-methylthymidine incorporation and the values of cpm (counts per minute) are shown as mean ±SEM (n=3 each group). *p<0.05, unpaired Student’s t test. One representative of 3 experiments is shown.

Discussion

Immunological tolerance has been defined as a mechanism by which the immune system prevents pathologic autoreactivity against self and thus prevents autoimmune diseases [22]. Administration of specific encephalitogenic antigens by various routes, including mucosal, i.v., i.p., s.c. and oral, can induce immune tolerance in EAE [7, 23, 24]. DCs, as professional APCs, can be induced into a tolerogenic phenotype, with the ability to promote the function of regulatory T cells or inhibit activation of autoreactive T cells [25–27]. Our recent study also showed that tolerogenic CD11c⁺CD11b⁺ DCs, were significantly increased in tolerized mice after i.v. injection of myelin antigen MOG_35–55 peptide [15]. In the present study we showed that this up-regulation of CD11c⁺CD11b⁺ DCs exhibited an immature phenotype and can be reduced by clodronate-loaded liposomes, which also abrogated i.v. MOG-induced suppression of EAE as well as decreased encephalitogenic IL-2-, IFN-γ-, and IL-17-producing CD4⁺ cells. Furthermore, recent studies have shown that M1 microglia/macrophages release pro-inflammatory cytokines and participate in the pathogenesis of EAE, while M2 cells produce anti-inflammatory cytokines and repair tissue [28, 29]. It has been shown that treatment with clodronate-loaded liposomes significantly reduced the numbers of CD45^high CD11b⁺ macrophages within the CNS and slightly delayed the expansion of CD45^dimCD11b⁺ microglia and infiltrating T cells [30]. In our study we found that clodronate-loaded liposomes selectively depleted M2 type, but not M1, macrophages in MOG-tolerized mice. This effect, together with reduced CD11c⁺CD11b⁺ tolerogenic DCs, might contribute to the observed blockade of MOG-induced EAE suppression by clodronate-loaded liposomes.

How DCs modulate the balance between autoimmunity and tolerance in vivo has not been fully elucidated. In general, tolerogenic DCs typically exhibit an immature phenotype characterized by low expression of IA and co-stimulatory molecules as well as low secretion of pro-inflammatory cytokines. In contrast, mature DCs facilitate activation of T cells and lead to inflammation [2, 9, 31]. In addition to the fact that co-stimulatory molecules expressed on DCs facilitate or inhibit the activity of effector cells, DCs secrete pro-or anti-inflammatory cytokines to modulate activation of CD4⁺ T cells. Further, iDCs exposed to necrotic cells or bacteria induce maturation and promote inflammation [32, 33], while binding and phagocytosis of apoptotic cells can inhibit the maturation of iDCs and render them tolerogenic [34–36].

Our group and others have shown that i.v. injection of myelin antigen into EAE mice induced by the same antigen results in apoptosis of a significant number of activated myelin-specific T cells [16, 37]. We have designed an immunotherapy using apoptotic cell-treated DCs to block EAE development in vivo [38, 39]. Our current results, together with previous studies, lead us to hypothesize that i.v. administration of the myelin antigen to EAE mice results in T cell apoptosis, and these apoptotic T cells can be ingested by iDCs, causing them to acquire a tolerogenic phenotype with lower expression of IA and co-stimulatory molecules [40, 41]. At the same time, DC deficiency in co-stimulatory molecules can induce regulatory T cells, and promote antigen-specific tolerance [42]. Indeed, Perruche et al. have shown that treatment of EAE mice with intact anti-CD3 antibody resulted in apoptosis of CD3⁺ T cells. These cells, when engulfed by phagocytes, promoted the latter to produce TGF-β, which induced CD4⁺Foxp3⁺ Treg cells in culture and contributed to immune tolerance in vivo [16]. When autoantigen-loaded DEC205⁺ DCs were injected, these cells induced TGF-beta production and Treg cells, and effectively suppressed EAE [43]. Similar results have also been observed in autoantigen-pulsed DCs [44]. In our present study, we found a higher proportion of TGF-β-producing CD11c⁺CD11b⁺ DCs in i.v. MOG tolerance mice compared to untreated EAE mice, accompanied by an increased proportion of CD4⁺CD25⁺Foxp3⁺ Treg cells, while these cells were largely reduced after depletion of CD11c⁺CD11b⁺ DCs. These results, together with the reports of other groups, indicate that induction of CD11c⁺CD11b⁺ DCs is an important step in the induction of Treg cells and, eventually, i.v. tolerance in EAE.

Another important immunomodulatory cytokine produced by APCs, including DCs and macrophages, is IL-27 [45–48]. DCs secreteIL-27 in response to Toll-like receptor (TLR) activation through a mechanism that involves the autocrine effects of interferon-β (IFN-β) [49]. The therapeutic effects of IFN-β administration on relapsing-remitting MS have been linked to the induction of IL-27 production by DCs [50]. IL-27 subunits (EBI3 and p28) and its receptor (WSX1) have been found to be upregulated in inflammatory cells in the CNS during EAE [51]. However, IL-27 administration inhibits EAE development [52]. IL-27R (WSX-1) deficient mice developed more severe EAE than WT mice, suggesting an anti-inflammatory effect of IL-27 [50, 53]. Mascanfroni et al. found that therapeutic vaccination with IL-27-conditioned DCs suppressed established relapsing-remitting EAE dependent on the expression of CD39. LPS pre-treatment with IL-27 on cDCs (F4/80⁻CD11c⁺CD11b⁺B220⁻MHC-II⁺LY6C⁻ DCs) from naive mice up-regulated IL-27 expression, suggesting a positive feedback loop. It also led to a significantly decreased expression of IA, CD40, CD80, CD86, IL-12, IL-23 and an increased production of IL-10 and TGF-β1 by these cDCs [54]. In this study, suppression of EAE by i.v. MOG was associated with a higher proportion of CD11c⁺CD11b⁺DCs, with a low expression of IA and co-stimulatory molecules and a high production of IL-27, TGF-β and IL-10 as well.

IL-27 acts directly on T cells to suppress their differentiation into effector T cells. Pre-treatment of cDCs with IL-27 led to a significant decrease in the proliferative response of naïve 2D2 T cells to MOG_35–55. Moreover, IL-27 treated cDCs had a decreased ability to induce IFN-γ and IL-17 production of CD4⁺T cells while their ability to promote IL-10⁺ and Foxp3⁺CD4⁺ T cells was increased [54]. Data from our previous studies show that IL-27 has a suppressive effect on EAE through an effect on encephalitogenic Th17 cells [50]. IL-27 also up-regulatedIL-10 in effector T cells that produce IFN-γ and T-bet, but not Foxp3, indicating a Tr1 phenotype. These IFN-γ⁺T-bet⁺Foxp3⁻ cells resemble effector T cells, which have been identified as the main source of host-protective IL-10 during inflammation [52]. In the present study, we found a higher production of IL-27 in CD11b⁺CD11c⁺ DCs of i.v. MOG mice and an increased proportion of CD4⁺IL-10⁺Foxp3⁻ Tr1 cells [20, 21]; we also found that these effects were significantly reduced after CD11b⁺CD11c⁺ DC-depletion. These results indicate that i.v. MOG-induced tolerance is achieved, at least in part, through the production of IL-27 in tolerogenic DCs, thus initializing the IL-27/Tr1 pathway [55].

In summary, our present study demonstrates that CD11b⁺CD11c⁺ DCs, upon i.v. MOG injection in EAE mice, facilitate development of both Treg cells and Tr1 cells through inhibiting co-stimulatory signals and inducing TGF-β and IL-27 signals, given that depleting these cells significantly blocked the effect of i.v. tolerance. Thus, induction of tolerogenic DCs is a key mechanism underlying i.v. autoantigen-induced tolerance.

Materials and Methods

Mice and antigen

Female C57BL/6 mice, 8–10 weeks of age, were purchased from the Jackson Laboratory. Mice were housed in the Thomas Jefferson University animal care facilities. All work was performed in accordance with the Thomas Jefferson University guidelines for animal use and care. Mouse MOG_35–55 (MEVGWYRSPFSRVVHLYRNGK) peptide was purchased from Invitrogen (Carlsbad, CA).

Depletion of DCs in naïve mice

Female naïve C57BL/6 mice were i.p. injected with the same volume of PBS- or clodronate-loaded liposomes (Encapsula Nanosciences LLC) (2 mg clodronate/mouse). Spleen cells were then harvested 24 hours later and analyzed by flow cytometry to observe the depletion of DCs.

Induction of EAE, i.v. tolerance, and iDC depletion

Mice were injected subcutaneously (s.c.) with 200 µg of MOG_35–55 in CFA containing 5 mg/ml Mycobacterium tuberculosis H37Ra (Difco Lab, Detroit, MI) over two sites on the back. All mice received 200 ng of pertussis toxin (List Biological Laboratories, Epsom, England) i.p. on days 0 and 2 post immunization (p.i.) [56]. To induce tolerance, MOG_35–55 peptide (200 µg/mouse) was i.v. injected on days 0, 3, and 6 p.i., and mice that received the same volume (100 µl) of PBS in parallel served as controls. To deplete CD11c⁺CD11b⁺ DCs, mice were intraperitoneally (i.p.) injected with clodronate-loaded liposomes (2 mg clodronate/mouse) 1 day before and 14 days p.i. and mice that received the same volume of PBS-loaded liposomes served as control.

Clinical evaluation

Clinical EAE was assessed by a 0–5 scoring system described previously [57]. Briefly, 0: normal; 1: limp tail or waddling gait with tail tonicity; 2: waddling gait with limp tail (ataxia); 2.5: ataxia with partial limb paralysis; 3: full paralysis of one limb; 3.5: full paralysis of one limb with partial paralysis of second limb; 4: full paralysis of 2 limbs; 4.5: moribund and 5: death. Mice were examined daily in a blind fashion for signs of EAE.

Histopathology

For CNS histopathology assessment, mice were perfused with 30ml PBS via the heart to eliminate peripheral blood, and spinal cords were harvested at day 21 p.i. Tissues were treated with ethanol and xylene, and paraffin-embedded 5 µm sections were stained with H&E for assessment of inflammation and with Luxol fast blue (LFB) for demyelination. Slides were assessed in a blinded fashion for inflammation and demyelination using a 0–3 scale as described [58]. For inflammation 0: none; 1: a few inflammatory cells; 2: organization of perivascular infiltrates and 3: increasing severity of perivascular cuffing with extension into the adjacent tissue. For demyelination 0: none; 1: rare foci; 2: a few areas of demyelination and 3: large (confluent) areas of demyelination.

Isolation of splenocytes and CNS-infiltrating mononuclear cells

Mice were extensively perfused at day 21 p.i. and splenocytes were harvested by forcing the tissue through a sterile 70µm nylon cell strainer to generate a single-cell suspension. Erythrocytes in the cell pellets from spleen were lysed by adding NH4Cl-Tris buffer for 5 min at room temperature followed by washing and re-suspending, and viable cells were counted in 0.4% Trypan blue.

For preparation of infiltrating mononuclear cells (MNCs) from spinal cord and brain, we mechanically dissociated spinal cords and brain through a 70 µm cell nylon strainer and enzymatically digested by incubation with Liberase TM (Roche) at 37°C for 20–30 min. Infiltrating mononuclear cells were harvested using a Percoll (Sigma-Aldrich, St. Louis, MO) gradient (70/37%), washed, and viable cells were counted in 0.4% Trypan blue.

Cytokine profiles

Splenocytes at 1–2.5 ×10⁶/ml were cultured in triplicate in RPMI1640 with 10% FBS in 24-well plates. Cells were restimulated with 20 µg/ml MOG_35–55 for 24, 48 and 72 hours. Cell-free supernatants were collected and analyzed for the production of IL-27, TGF-β, IL-2, IFN-γ, and IL-17 by ELISA, using ELISA kits (R&D Systems, Minneapolis, MN).

Flow cytometry analysis

Splenocytes and CNS-infiltrating cells were suspended in complete RPMI 1640 with 10% FBS at a density of 2 × 10⁶/ml. For surface-marker staining, cells were washed in FACS buffer (PBS containing 3% FBS and 0.02% NaN3), then incubated with fluorochrome-conjugated antibodies for 30 minutes on ice. For intracellular cytokine staining, cells were stimulated with PMA (50 ng/ml), ionomycin (500 ng/ml), and Golgi-Stop (1 mg/10⁶ cells) for 4 hours in complete RPMI 1640 medium with 10% FBS. After being washed in FACS buffer cells were stained with fluorescent antibodies to surface markers, followed by intracellular staining in accordance with the manufacturer’s instructions. In brief, cells were fixed and permeabilized using Fix/Perm cell permeabilization reagents, followed by incubation with fluorescently labeled antibodies against intracellular cytokines. All antibodies and reagents used in this section were purchased from BD Biosciences (San Jose, CA), with the exception of PMA and ionomycin (both from Sigma-Aldrich). After the last wash, data were acquired by FACS Aria (BD Biosciences, San Jose, CA). Data were analyzed using FlowJo Software (Tree Star, Ashland, OR).

Purification of DCs and co-culture with MOG-reactive CD4⁺ T cells in vitro

DCs from splenocytes were prepared as previously described [59]. In brief, single cell suspensions derived from the spleen of EAE mice treated with MOG-i.v. + PBS-i.p. or MOG-i.v. + clodronate-i.p. were incubated with mouse CD11c MicroBeads (Miltenyi Biotec) and then subjected to positive selection through MACS separation columns. Cells selected on the basis of CD11c expression routinely consisted of 90% viable DCs. Single cell suspensions derived from the spleen of naïve MOG TCR transgenic mice (2D2, 6–10 weeks of age, Jackson Laboratory) were purified by positive selection with mouse CD4 MicroBeads (Miltenyi Biotec). CD4⁺ Cells (1 × 10⁶/ml) were then incubated in 24-well culture plates with MOG (20 µg/ml) and co-cultured with DCs (1×10⁵/ml) purified from spleen cells of MOG-i.v. + PBS-i.p. and MOG-i.v.+clodronate-i.p. mice respectively. After 3 days, cell-free supernatants were collected and analyzed for the production of IL-27, IL-10, IL-2, IFN-γ, and IL-17 by ELISA, using ELISA kits. Cells were then stained with CD4, CD11c, CD11b, CD25, IFN-γ and IL-17, IL-10, IL-27 antibodies (BD Biosciences) and analyzed with flow cytometry.

For cell proliferative response, triplicate aliquots (200 µl) of these co-culture cells were applied to 96-well plates at a density of 0.5×10⁶/ml. MOG_35–55 was added to appropriate wells at final concentrations of 10, 20, and 30 µg/ml. After 54 hours, cells were incubated with 1 μCi of ³H-methylthymidine (specific activity 42 Ci/mmol) for 18 hours. Cells were harvested on fiberglass filters and thymidine incorporation was measured at a scintillation counter.

Statistical analysis

Experimental data were analyzed using Prism software (GraphPad, La Jolla, CA, USA). Comparisons of data between experimental groups were performed by two-way ANOVA test or unpaired Student’s t tests. All tests were two-sided. Differences were considered significant at a value of p<0.05.