Peritoneal Exudate Cells Treated with Calcitonin Gene-Related Peptide Suppress Murine Experimental Autoimmune Uveoretinitis via IL-10¹

Experimental autoimmune uveitis (EAU)³ is an animal model of intraocular inflammatory disease, a condition caused by idiopathic disease or systemic diseases such as Behçet syndrome and sarcoidosis, which may lead to blindness. In animals, EAU can be induced by immunization with interphotoreceptor retinoid-binding protein (IRBP), an eye-specific retinal Ag, or by transfer of the Ag-specific T cells (1, 2). Mouse and rat EAU is a dominant Th1 response to the uveitogenic retinal Ag, and uveitogenic effector T cells display a Th1-like cytokine profile. Up-regulation of Th2 cytokine responses has been known to protect EAU (3, 4).

Inoculation of Ags, ranging from viral-encoded molecules to soluble heterologous proteins, into the anterior chamber of the eye typically leads to the induction of a stereotypic systemic immune response selectively deficient in Ag-specific effectors that evoke immunogenic inflammation, including T cells that mediate delayed hypersensitivity (DH) and IgG Abs that fix complement. This response, termed anterior chamber-associated immune deviation (ACAID), arises in part from unique properties of the anterior chamber in the eye. Aqueous humor that fills the anterior chamber is secreted by cells of the ciliary body and contains numerous immunomodulatory cytokines and neuropeptides (5, 6, 7). Among the factors in the aqueous humor, TGF-β2, α-melanocyte stimulating hormone, vasoactive intestinal peptide, and calcitonin gene-related peptide (CGRP) have been indicated to play particularly important roles as immunomodulatory factors (8, 9, 10, 11).

CGRP, a 37-aa peptide, is a neuropeptide widely distributed in the CNS and peripheral nervous system (12), and is constitutively found in the sensory neurons of the iris and ciliary body (13). Specific receptors for CGRP localize in most tissues, including the CNS, heart, liver, and spleen (14).

tk;2Signals via CGRP receptors have been explored in murine macrophage cell line, and one of the signals leads to accumulation of intracellular cAMP (15). CGRP has been shown to modulate inflammatory and immune responses, such as inhibition of proliferative responses of PBMC, IL-2 production by lymphocytes, and the induction of DH (16, 17, 18). Torii et al. (19) have demonstrated in vitro that CGRP augments IL-10 production and suppresses IL-1β production by LPS-stimulated peritoneal exudated macrophages. The immunomodulatory effects of CGRP have been demonstrated in in vivo studies. CGRP injected intracutaneously impairs the induction of contact hypersensitivity in mice, and contributes to the pathogenesis of failed contact hypersensitivity induction after acute, low-dose UVB radiation (20). In addition, CGRP contributes to immunological tolerance in normal and mast cell-deficient mice induced by acute, low-dose UVB radiation (20). These immunoregulatory mechanisms of CGRP should enhance IL-10 production by macrophages, because IL-10 impairs IL-12 production and inhibits B7 expression on APC (21).

Wilbanks et al. (22, 23, 24) reported that F4/80⁺ macrophages harvested from peritoneal exudate cells (PEC) cultured with a soluble protein Ag in the presence of TGF-β2 similarly induced ACAID when injected i.v. into naive mice, indicating that by treatment with TGF-β2, which is found abundantly in aqueous humor, F4/80⁺ macrophages are able to acquire ACAID-inducing properties similar to F4/80⁺ bone marrow-derived cells in iris/ciliary body. CGRP is an immunosuppressive factor present in high concentration in aqueous humor, like TGF-β2. In this study, we examined whether CGRP confers immunosuppressive abilities on PEC, and whether CGRP-treated PEC pulsed with the human IRBP peptide spanning aa residues 1–20 (hIRBP 1–20) are able to prevent development of EAU in mice induced by immunization with hIRBP 1–20. We also investigated whether CGRP-induced IL-10 production by PEC contributes to the immunomodulatory mechanisms. Our results indicated that by injection of PEC treated with CGRP (100 ng/ml) and pulsed with hIRBP 1–20, immune tolerance to hIRBP 1–20 was established and EAU was suppressed. However, no tolerance and EAU suppression were observed by CGRP and hIRBP 1–20-treated PEC when recipient mice were given anti-IL-10 Abs i.p., or when the PEC were derived from IL-10 knockout mice. These results demonstrate that macrophages treated with CGRP suppress murine EAU by inducing immunosuppression to the pathogenic Ag, and IL-10 secreted from the macrophages may play an important role in the CGRP-mediated suppression of murine EAU.

Six- to 8-wk-old female C57BL/6 mice were obtained from Japan CLEA (Shizuoka, Japan). C57BL/6 IL-10 knockout mice (C57BL/6-IL-10^tmlCgn) were purchased from The Jackson Laboratory (Bar Harbor, ME). All animals were treated according to the Association for Research in Vision and Ophthalmology resolution on the use of animals in research. All procedures were performed under sodium pentobarbital anesthesia or ketamine and xylazine mixtures as anesthetics.

hIRBP peptide sequence 1–20 (GPTHLFQPSLVLDMAKVLLD) (hIRBP 1–20) was synthesized by conventional solid-phase techniques, as described elsewhere (25). Purified Bordetella pertussis toxin (PTX) was from Sigma-Aldrich (St. Louis, MO), and CFA and Mycobacterium tuberculosis strain H37Ra were from Difco (Detroit, MI). Rat CGRP, human CGRP fragment (8–37) (antagonist of CGRP), and BSA were purchased from Sigma-Aldrich. As mitogen, Con A was purchased from Vector Laboratories (Burlingame, CA). Monoclonal goat anti-mouse IL-10 Ab and goat IgG (control for anti-IL-10 Ab) were purchased from R&D Systems (Minneapolis, MN).

Serum-free medium was used for all PEC cultures (26). This medium was composed of RPMI 1640 (Sigma-Aldrich), 10 mM HEPES (Invitrogen Life Technologies, Carlsbad, CA), 0.1 mM nonessential amino acids (Invitrogen Life Technologies), 1 mM sodium pyruvate (Invitrogen Life Technologies), 100 U/ml penicillin (Invitrogen Life Technologies), 100 μg/ml streptomycin (Invitrogen Life Technologies), and 1 × 10⁻⁵ M 2-ME (Sigma-Aldrich), supplemented with 0.1% BSA (Sigma-Aldrich) and ITS⁺ culture supplement (1 μg/ml iron-free transferrin, 10 ng/ml linoleic acid, 0.3 ng/ml Na₂Se, and 0.2 μg/ml Fe(NO₃)₃) (Collaborative Biochemical Products, Bedford, MA). In some studies, RPMI 1640 containing 10% FBS instead of 0.1% BSA was used.

PEC were collected by washing the peritoneal cavity of normal mice that were injected i.p. with 2 ml of 3% thioglycolate medium (Sigma-Aldrich) 3 days earlier.

The PEC were washed and resuspended, seeded in 24-well culture plates (1 × 10⁶/well), and treated with or without 2 or 100 ng/ml rat CGRP or 100 ng/ml CGRP fragment peptide (8–37) in the presence of 4 μM hIRBP 1–20 in serum-free medium at 37°C in an atmosphere of 5% CO₂. In some experiments, BSA was used as a control of hIRBP 1–20. After overnight culture, the plates were washed three times with culture medium to remove CGRP, Ag, and nonadherent cells. Adherent cells were used in all subsequent experiments. More than 95% of the adherent cells were F4/80⁺ macrophages. Intravenous injection of cells was performed according to standard methods. Briefly, cells were washed and counted twice by using trypan blue and diluted to the desired cell concentration, and then injected into recipient mice i.v. (3 × 10⁵ cells in 150 μl) via the tail vein.

hIRBP 1–20 (200 μg) was emulsified in CFA (1:1 w/v) containing 5 mg/ml Mycobacterium tuberculosis H37Ra. A total of 200 μl of the emulsion was injected s.c. in the thigh and neck. Concurrent with immunization, 0.1 μg of PTX was injected i.p. For histological assessment of EAU, eyes were collected and assessed on day 21 after immunization. Eyes were fixed in Bouin’s solution. Sections of samples were embedded in paraffin and stained with H&E for pathological study. The severity of EAU in each eye was scored on a scale of 0–4 in half-point increments, according to a semiquantitative system described previously (27). Briefly, focal nongranulomatous, monocytic infiltrations in the choroids, ciliary body, and retina were scored as 0.5. Retinal perivascular infiltration and monocytic infiltration in the vitreous were scored as 1. Granuloma formation in the uvea and retina, the presence of occluded retinal vasculitis, along with photoreceptor folds, serous detachment, and loss of photoreceptor were scored as 2. In addition, the formation of Dalen-Fuchs nodules (granuloma at the level of the retinal pigmented epithelium) and the development of subretinal neovascularization were scored as 3 and 4 according to the number and the size of the lesions.

Splenectomy was conducted by a modification of the procedure described elsewhere (28). Briefly, a midline incision was made in the abdomen and the spleen was exposed. The splenic arteries were ligated using chronic gut sutures, and the spleen was then excised. The incision was sutured with silk. In sham-operated animals, the same procedure was done, but the spleen was left intact. On day 14 after splenectomy, the mice were immunized with hIRBP 1–20 and injected with the CGRP-treated PEC. On day 21 after immunization, the mice were sacrificed for histological assessment of EAU.

For induction of EAU by adoptive transfer of primed cells, donor C57BL/6 mice were immunized with the hIRBP 1–20 (200 μg), as above. The splenic cells were collected on day 21 after immunization, and cultured with 3 μg/ml Con A (vector) for 48 h. The cultured spleen cells were collected and injected i.p. (15–20 × 10⁷ cells/mouse) into naive C57BL/6 mice, and then CGRP-treated PEC were injected into these mice. EAU was assessed histopathologically 14 days after the adoptive transfer. In another experiment, CGRP-treated PEC were injected into the mice following immunization with hIRBP 1–20. On day 21 after immunization, the mice were sacrificed and spleen cells were collected. The collected spleen cells were cultured with 3 μg/ml Con A (vector) for 48 h in RPMI 1640 containing 10% FBS, and then splenic T cells were purified using mouse pan T cell isolation kit and autoMACS purchased from Miltenyi Biotec (Gladbach, Germany). More than 99% of cells were CD3⁺ T cells (data not shown). The purified T cells were injected (4 × 10⁷ cells/mouse) into mice following immunization with hIRBP 1–20. On day 21 after immunization, the mice were sacrificed for histological assessment of EAU.

In some experiments, C57BL/6 mice were injected i.p. with 0.2 mg/0.1 ml/mouse goat anti-mouse IL-10 Ab (R&D Systems) on day 4 after immunization. Control was treated with an equivalent dose of goat Ig (R&D Systems). Eyes were collected on day 21 after immunization with hIRBP peptide.

On day 19 after immunization, mice were injected intradermally with 20 μg/10 μl of hIRBP 1–20 suspended in PBS into the pinna of one ear. Ear swelling was measured after 24 and 48 h using a micrometer (Mitutoyo, Tokyo, Japan). Ag-specific DH was measured as the difference in ear thickness before and after challenge. Results were expressed as: specific ear swelling = (24-h measurement – 0-h measurement) for test ear – (24-h measurement – 0-h measurement) for control ear.

Results of some experiments were analyzed by ANOVA and Scheffe’s test. Means were considered to be significantly different when p < 0.05. Mann-Whitney U test was used in the analysis of EAU. Data are presented as mean ± SEM.

We first investigated whether DH response against immunized Ag was suppressed by CGRP-treated PEC. PEC were cultured with 2 or 100 ng/ml rat CGRP or human CGRP fragment (8–37) as control in the presence of hIRBP 1–20, and then injected into recipient mice i.v. (3 × 10⁵ cells in 150 μl) via the tail vein. In this experiment, human CGRP fragment peptide (CGRP-(8–37)) was used as a control for CGRP. CGRP-(8–37) binds to the CGRP receptor with high affinity, but functions as an antagonist of CGRP by interfering with signaling via the CGRP receptor. Injected mice were simultaneously immunized with hIRBP 1–20 to induce EAU. On day 19 after immunization, hIRBP 1–20 challenge was conducted by intradermal injection into the ear. After 24 and 48 h, ear swelling was measured for DH assay. Fig. 1,A shows the 24-h data of a representative experiment. DH response against hIRBP 1–20 was markedly inhibited in mice that received PEC treated with CGRP in a dose-dependent manner. Subsequently, we examined whether the development of EAU was also suppressed in mice that received CGRP-treated PEC. Histopathological evaluation of EAU was performed at 21 days after immunization (see Materials and Methods). Fig. 1,B shows the representative results of the experiment. Compared with mice injected with CGRP-(8–37)-treated PEC, both frequency and severity of EAU were markedly suppressed in mice injected with CGRP-treated PEC, especially when PEC were treated with 100 ng/ml CGRP. Fig. 1, C–F, shows the representative histopathology of the retina in mice that received PEC cultured with CGRP-(8–37) (Fig. 1, C and D), 2 ng/ml CGRP (Fig. 1,E), or 100 ng/ml CGRP (Fig. 1,F). In mice injected with CGRP-(8–37)-treated PEC, inflammatory cell infiltration into the entire retina was observed, and anatomical retinal layers were partially destroyed (Fig. 1,C, score 1; Fig. 1,D, lower magnification). However, cell infiltration into the retina was markedly reduced in mice injected with CGRP-treated PEC, and retinal layer structures were completely preserved (Fig. 1,E, score 0.5; Fig. 1 F, score 0).

Although the development of EAU in mice was suppressed by injection of CGRP-treated PEC pulsed with hIRBP 1–20, it is possible that the suppression might be Ag nonspecific. Therefore, we also examined whether CGRP-treated PEC pulsed with an Ag other than hIRBP 1–20 possess the ability to suppress EAU induced by immunization with hIRBP 1–20. In this experiment, PEC treated with BSA in the presence of 100 ng/ml CGRP were injected i.v. into mice, then the recipients were immunized with hIRBP 1–20 to induce EAU. Representative results of the experiments are shown in Fig. 2. Although 3 of 10 mice that received CGRP-treated PEC pulsed with hIRBP 1–20 developed EAU (30%) with a mean pathological EAU score of 0.12, 7 of 10 mice with BSA-pulsed, CGRP-treated PEC developed EAU (70%) with a mean EAU score of 0.79. Both frequency and severity of EAU for BSA-pulsed, CGRP-treated PEC were comparable to those for hIRBP 1–20-pulsed, CGRP-untreated PEC (70%, 0.81). These results indicated that CGRP-treated PEC suppressed the development of EAU in an Ag-specific manner.

Next, we examined whether the spleen is required to express the suppressive effects of the CGRP-treated PEC. We investigated the effects of splenectomy on the immune suppression induced by CGRP-treated PEC. Sham-operated mice served as controls. On day 14 after splenectomy, the mice were immunized with hIRBP 1–20 and injected with the CGRP-treated PEC. On day 21 after immunization, the mice were sacrificed for histological assessment of EAU. The representative results are shown in Fig. 3,A. Although 7 of 9 sham-operated mice that received CGRP-untreated PEC pulsed with hIRBP 1–20 developed EAU (78%) with a mean pathological EAU score of 1.56, 3 of 10 sham-operated mice injected with CGRP-treated PEC developed EAU (30%) with a mean EAU score of 0.40 (p = 0.026). In contrast, in the splenectomized mice, both frequency and severity of EAU in mice that received CGRP-treated PEC (70%, 1.30) were comparable to those injected with CGRP-untreated PEC (80%, 1.20). Fig. 3 B shows the results of 24-h evaluation of a representative DH assay. DH response against hIRBP 1–20 was markedly inhibited in sham-operated mice that received PEC treated with CGRP. In contrast, DH response against hIRBP 1–20 was not inhibited in splenectomized mice that received PEC treated with CGRP. These data indicated that the spleen plays an important role in the suppression of EAU by CGRP-treated PEC.

To determine whether CGRP-treated PEC impaired cellular pathogenicity in the efferent phase, we performed adoptive transfer experiments. To induce EAU by adoptive transfer of primed cells, donor C57BL/6 mice were immunized with the hIRBP 1–20 (200 μg), as in Materials and Methods. The splenic cells were collected on day 21 after immunization and cultured with 3 μg/ml Con A for 48 h. Forty-eight hours later, cultured spleen cells were collected and injected i.p. (15–20 × 10⁷ cells/mouse) into naive C57BL/6 mice. Simultaneously, PEC treated with hIRBP 1–20 and rat CGRP or human CGRP-(8–37) were injected into adoptively transferred mice. EAU was assessed histopathologically 14 days after the adoptive transfer. The results of representative experiments are shown in Fig. 4. Five of eight (63%) adoptively transferred mice injected with hIRBP 1–20-pulsed, CGRP-(8–37)-treated PEC developed EAU, whereas none of eight adoptively transferred mice injected with hIRBP 1–20-pulsed, CGRP-treated PEC had EAU. These data indicated that: 1) Th1-like effector cells responsible for the pathogenesis of EAU were not induced, and 2) hIRBP 1–20-pulsed, CGRP-treated PEC suppressed EAU in the efferent phase.

We investigated whether EAU is suppressed by the suppressor T cells generated from hIRBP 1–20-immunized mice injected with CGRP-treated PEC. CGRP-treated PEC were injected into mice following immunization with hIRBP 1–20. On day 21 after immunization, the mice were sacrificed, and spleen cells were collected. The collected spleen cells were cultured with Con A for 48 h, and then T cells were purified from the spleen cells. The purified T cells were injected into the hIRBP 1–20-immunized mice. On day 21 after immunization, the mice were sacrificed for histological assessment of EAU. The results of a representative experiment are shown in Fig. 5. Six of 10 (60%) mice adoptively transferred T cells obtained from mice injected with hIRBP 1–20-pulsed, CGRP-(8–37)-treated PEC developed EAU with a mean pathological EAU score of 0.65, whereas none of the eight mice adoptively transferred T cells obtained from mice injected with hIRBP 1–20-pulsed, CGRP-treated PEC had EAU (p = 0.023). DH response against hIRBP 1–20 was markedly inhibited in immunized mice adoptively transferred T cells obtained from mice injected with hIRBP 1–20-pulsed, CGRP-treated PEC (data not shown). These data indicated that the recipients of CGRP-treated PEC were able to induce suppressor T cells.

To elucidate whether IL-10 is pivotal for the immunosuppressive effects of CGRP-treated PEC, mice injected with hIRBP 1–20-pulsed, CGRP-treated PEC were given i.p. anti-IL-10 Ab or goat IgG as control on day 4 after immunization with hIRBP 1–20. On day 19 after immunization, hIRBP 1–20 was injected intradermally into the ear. After 24 h of hIRBP 1–20 challenge, ear swelling was measured for DH assay. In addition, eyes were collected for histopathological evaluation 21 days after immunization. Representative results are shown in Fig. 6,A. DH response against hIRBP 1–20 was completely restored by administration of anti-IL-10 Ab in mice injected with hIRBP 1–20-pulsed, CGRP-treated PEC (Fig. 6 B). In addition, anti-IL-10 Ab dramatically impaired the ability of hIRBP 1–20-pulsed, CGRP-treated PEC to suppress EAU in both frequency and severity.

CGRP has been shown to promote IL-10 production by LPS-stimulated macrophages in vitro (17). In addition, because the above in vivo study showed that CGRP-treated PEC failed to suppress EAU by inducing immune tolerance to the immunized Ag in the absence of IL-10, it is possible that IL-10 production by PEC promoted by treatment with CGRP would be critical for the immunomodulatory effects of CGRP-treated PEC. To investigate this hypothesis, the experiments were planned using PEC derived from IL-10 knockout mice. In brief, PEC derived from IL-10 knockout mice or wild-type mice were treated with hIRBP 1–20 in the presence of 100 ng/ml CGRP or CGRP-(8–37), and were then injected i.v. into syngeneic mice. The recipient mice were immunized with hIRBP 1–20, then DH responses were measured on day 19, and histological evaluation of EAU was done on day 21 after immunization. Representative results of the experiments are shown in Fig. 7. Mice injected with CGRP-treated PEC from IL-10 knockout mice developed EAU at a high frequency, similar to mice injected with CGRP-(8–37)-treated PEC from wild-type or IL-10 knockout mice (60 vs 70%). The average EAU score for CGRP-treated PEC from IL-10-deficient mice was slightly lower than that for CGRP-(8–37)-treated PEC, but significantly higher than that for CGRP-treated PEC from wild type (Fig. 7,A). In addition, suppression of DH responses to the immunized Ag was observed with wild-type PEC treated with CGRP, but was not found with PEC derived from IL-10 knockout mice treated with CGRP (Fig. 7 B). These data suggested that IL-10 derived from CGRP-treated PEC is important for their abilities to suppress development of EAU and induce immune suppression to the immunogenic regimen.

This study demonstrates that injection of macrophages pulsed with hIRBP 1–20 in the presence of CGRP induces Ag-specific immune suppression via IL-10-dependent mechanisms, which suppresses DH responses and development of EAU even in the efferent phase.

The eye is an immune-privileged site that prevents induction of immunogenic inflammation within its microenvironment. The immunosuppression is considered to be an evolutionary adaptation of several immunosuppressive mechanisms that protect the eye’s delicate structure from damages associated with immunogenic inflammation. ACAID is induced when soluble Ag is injected into the anterior chamber of the eye (29). Bone marrow-derived macrophages in the iris/ciliary body, which are bathed in aqueous humor, capture the Ag and migrate out of the eye across the trabecular meshwork directly into the blood (30). Upon the arrival of these cells in the spleen, they act as ACAID-inducing cells and selectively activate CD8⁺ T cells (31), B cells that secrete noncomplement-fixing Abs (32), and regulatory T cells that suppress DH and complement-fixing Ab formation.

In addition, PEC that are pulsed with Ag in the presence of aqueous humor in vitro are able to induce Ag-specific immune deviation like ACAID, when injected i.v. into naive mice (33). These results led to the notion that factors that alter the characteristics of APCs exist in aqueous humor. The first molecule identified in this regard was TGF-β (9). PEC pulsed in vitro with Ag in the presence of TGF-β2 succeeded in inducing Ag-specific immune deviation like ACAID (22, 23, 24). However, we considered that bone marrow-derived macrophages in the eye also would be influenced by the other factors in aqueous humor. Taylor et al. (10, 11, 34, 35) identified several immunosuppressive propertie: α-melanocyte-stimulating hormone, vasoactive intestinal peptide, CGRP, and smatostatin in aqueous humor. They have recently examined whether the ocular microenvironment quenches NO production by activated macrophages. They observed that NO production was virtually abolished by aqueous humor and by CGRP, and neutralization of CGRP in aqueous humor totally abolished the fluid’s NO-inhibiting activity (34). Thus, soluble factors that promote adaptive immune privilege are present in aqueous humor. Recent studies have indicated that CGRP plays an immunomodulatory role in APCs (19, 21). ACAID induced by intracameral injection of IRBP alters the subsequent immune responses of mice, rendering them incapable of developing IRBP-specific DH or expressing significant autoimmune uveitis following a uveitogenic regimen (36). Moreover, when splenic cells from mice with ACAID to IRBP were harvested and injected into mice with established uveitis, the intraocular inflammation was arrested abruptly (36). PEC pulsed with hIRBP 1–20 in the presence of CGRP suppressed both DH responses and development of EAU in an Ag-specific manner. In addition, injection of hIRBP 1-20-pulsed, CGRP-treated PEC ameliorated EAU already developed by transfer of hIRBP 1-20-sensitized spleen cells, indicating that treatment with CGRP enables PEC to induce Ag-specific immune deviation like ACAID, as in the case of TGF-β2. In our experiments using splenectomized mice, both the frequency and severity of EAU in mice injected with CGRP-treated PEC were comparable to those that received CGRP-untreated PEC. DH response against hIRBP 1–20 was not inhibited in splenectomized mice that received PEC treated with CGRP. The system of EAU suppression by CGRP-treated PEC closely resembles the observation of ACAID system, in which it was shown that the spleen is essential for the induction of Ag-specific immunosuppression indicated by the prolonged acceptance of DBA/2 mouse skin allografts after intracameral injections of P815 mastocytoma cells (37).

Torii et al. (19) demonstrated that CGRP moderates Ag-presenting functions of Langerhans cells and macrophages, augmenting LPS-induced IL-10 production and suppressing IL-1β production. Moreover, B7-2 expression up-regulated by stimulation with LPS and GM-CSF was suppressed by CGRP in macrophages (19). They demonstrated that CGRP-induced modulation of the Ag-presenting functions of Langerhans cells and macrophages was related to cytokines they produce that favor less robust Ag presentation for cell-mediated immunity. Niizeki et al. (20, 38) have reported that CGRP participates in the immune aberrations in skin caused by UV radiation, and that IL-10 is an important mediator of tolerance induced by painting hapten on the skin of animals exposed to acute, low-dose UV radiation in the afferent phase only. In a subsequent study, Kitazawa and Streilein (21) reported that CGRP has the capacity to promote cutaneous tolerance through an IL-10-dependent mechanism. In the present study, CGRP-treated PEC obtained from IL-10 knockout mice failed to suppress EAU, and the failure was also observed when the recipient mice were treated with anti-IL-10 Ab. These results suggest that not only IL-10 production by CGRP-treated PEC, but also IL-10 produced by some T cells stimulated by CGRP-treated PEC are required for inducing the immune suppression. However, consistent with previous reports, the immunosuppression induced by CGRP-treated macrophages with altered Ag-presenting functions was also mediated by IL-10-dependent mechanisms.

It is also possible that IL-10 secretion from CGRP-treated PEC is directly associated with remission of EAU and suppression of DH responses in part. Takeuchi et al. (39) reported that IRBP-reactive Th2-type cells, which produce IL-4 and IL-10, were generated in the spleen in the remission phase of EAU, whereas IRBP-reactive Th1-type cells were induced in both the draining lymph nodes and the spleen. Rizzo et al. (40) described that administration of exogenous IL-10 ameliorated EAU and inhibited proliferation and IFN-γ production by mature uveitogenic effector T cells. In the present study, IL-10-deficient mice developed milder EAU compared with the control, although the difference in EAU score was not statistically significant. In previous studies, IL-10-deficient mice developed inflammatory bowel disease and severe inflammatory responses (41, 42). For this reason, we cannot exclude the possibility that alternative functions of macrophages in IL-10-deficient mice are responsible for the failure of CGRP-treated macrophages from IL-10-deficient mice to inhibit development of EAU in the present study. We consider that although IL-10 production induced by CGRP may in part account for the suppression of EAU and DH responses by CGRP-treated PEC, the preferential mechanism would be suppressor T cells induced by altered Ag-presenting functions of CGRP-treated PEC. To support this hypothesis, we demonstrated that the suppressor T cells could be generated in mice by adoptively transferred CGRP-treated PEC (Fig. 5).

In this study, it remains unclear whether APCs affected by CGRP directly establish tolerance by inducing suppressor T cells via IL-10, or whether other mechanisms are involved.

A recent study shows that besides direct inhibition of T cell stimulation, blockade of B7/CD28 facilitates induction of T cell unresponsiveness by generating alternatively activated macrophages via IL-10 (43). Another study shows that B7-H1 can be induced to express on macrophages and to enhance T cell proliferation and secretion of IL-10, IFN-γ, and GM-CSF, but not IL-2 or IL-4 (44). Furthermore, B7-H1 preferentially costimulates CD4⁺ T cells independent of CD28, and enhances mixed lymphocyte responses to allogeneic Ags (44). Based on these facts, we are investigating functions of costimulatory molecules that play a role in CGRP-associated regulatory responses.

In conclusion, Ag-specific DH and development of EAU were suppressed in an Ag-specific manner by injection of PEC cultured with CGRP and human IRBP peptide. However, the suppression of both DH responses and development of EAU was abolished when mice treated with anti-IL-10 Abs were used as recipients of CGRP-treated PEC, or when the PEC were derived from IL-10-deficient mice, indicating that the immunosuppression induced by CGRP-treated macrophages with altered Ag-presenting functions was mediated by IL-10-dependent mechanisms.