Prevention of Experimental Autoimmune Uveoretinitis by Vasoactive IntestinalPeptide

Background  Vasoactive intestinal peptide (VIP), a neuropeptide that is known tobe present in lymphoid tissue microenvironments, shows prominent anti-inflammatoryactions.

Objective  To examine the potential effect of VIP on the development of experimentalautoimmune uveoretinitis (EAU).

Design  We immunized C57BL/6 mice with human interphotoreceptor retinoid-bindingprotein peptide 1-20 (h-IRBP peptide). Vasoactive intestinal peptide was administeredintraperitoneally on alternate days until day 21 after immunization (entiregroup). In some cases, VIP was injected at different time points after theinduction of immunity with h-IRBP peptide (efferent group). In each experiment,a control group of mice was injected with phosphate-buffered saline insteadof VIP. Development of EAU was evaluated by means of histological examinationon day 21 after immunization. Furthermore, we determined whether intravenousinjection of peritoneal exudate cells cultured with VIP overnight in vitroabrogated EAU. We analyzed delayed hypersensitivity for h-IRBP peptide andthe occurrence and severity of EAU using evaluation of histopathological sectionsfor inflammatory ocular disease.

Results  Treatment with VIP suppressed the expression of delayed hypersensitivityresponses to h-IRBP peptide significantly (positive control vs entire group, P = .02; positive control vs efferent group, P<.001). Mice treated with VIP (n = 10) showed a lower occurrence(40%) and decreased severity of EAU (entire group mean score, 0.3; medianscore, 0) compared with untreated mice (occurrence, 80%; mean score, 0.85;median score, 0.75), as assessed by histopathological analyses (P = .049). Suppressive effects of VIP on EAU were also observed, evenwhen VIP was administered on days 8 through 20 after immunization (efferentgroup [n = 9] occurrence, 11%; mean score, 0.1; median score, 0) (P = .003). Moreover, expression of EAU was significantly suppressedwhen the animals were pretreated with peritoneal exudate cells pulsed withh-IRBP in the presence of VIP (control mean score, 1.2; median score, 1.0;occurrence, 80% [n = 10]) compared with the VIP-treatment group (mean score,0.3; median score, 0; occurrence, 30% [n = 10]) (P =.004). In addition, VIP-treated peritoneal exudate cells generated regulatorT cells in the spleens of recipient mice that were able to interfere withthe development of EAU (control group mean score, 0.5; median score, 0.5;occurrence, 63% [n = 8]) compared with the VIP-treatment group (mean score,0.08; median score, 0; occurrence, 17% [n = 6]) (P =.08).

Conclusion  Treatment with VIP is a highly effective therapy to suppress EAU.

Clinical Relevance  As a result of its efficacy in preventing EAU, VIP might be consideredas a novel therapeutic modality for human uveitis.

The eye is an immune-privileged site that inhibits the induction ofconventional immunological responses within its microenvironment.¹ Studies of the immunosuppressive activity of aqueoushumor, the fluid that fills the anterior chamber, have revealed the presenceof soluble factors that actively suppress immune-mediated inflammation.² Aqueous humor contains numerous potential immunomodulatingfactors such as transforming growth factor β₂,³ theneuropeptide α-melanocyte–stimulating hormone,⁴ andcalcitonin gene–related peptide.⁵ Anotherneuropeptide constitutively detected in aqueous humor is vasoactive intestinalpeptide (VIP).⁶

Vasoactive intestinal peptide, a neuropeptide that can be detected inlymphoid microenvironments, displays a broad spectrum of biological functions,including modulation of innate and adaptive immunity and suppression of inflammatoryreactions.⁷ Recent reports have indicated thatVIP promotes helper T cell type 2 (T_H2) differentiation and inhibitshelper T cell type 1 (T_H1) responses by regulating macrophage costimulatorysignals and probably interleukin 12/interferon γ production.⁸ All of these activities are likely to have beneficialeffects in autoimmune diseases such as Behçet syndrome. From thesereasons, we elected to evaluate the therapeutic effects and immunomodulatoryproperties of VIP in experimental autoimmune uveoretinitis (EAU), an animalmodel for human uveitis.

Experimental autoimmune uveoretinitis is a T-cell–mediated autoimmunedisease induced by immunizing susceptible strains of mice with interphotoreceptorretinoid-binding protein (IRBP) or its peptide.⁹ Theinflammatory CD4⁺ T cells that mediate EAU are known to secreteT_H1-type cytokines, including interleukin 2, interferon γ,and tumor necrosis factor α, as well as cytokines responsible for expressionof delayed-type hypersensitivity (DH) responses.¹⁰ BecauseEAU shares a number of clinical, histological, and immunological featureswith human uveoretinitis syndromes, including Behçet syndrome, Vogt-Koyanagi-Haradasyndrome, sympathetic ophthalmia, and sarcoidosis,¹¹ itis a suitable model to study the potential effects of VIP on the pathogenesisof uveitis. Herein we report that VIP prevents the development of and leadsto an improvement in the pathological score and occurrence of EAU inducedby human IRBP peptide 1-20 (h-IRBP peptide). In addition, peritoneal exudatecells (PECs) cultured in vitro with h-IRBP peptide in the presence of VIP-inducedh-IRBP peptide–specific regulatory T cells were injected into syngeneicmice and suppressed the h-IRBP peptide–specific DH response and developmentof EAU. These results suggest that VIP might be an effective therapy for humanuveoretinitis.

Specific pathogen-free female C57BL/6 mice were obtained from ShizuokaLaboratory Service, Shizuoka, Japan, and maintained in our animal facilities.Mice aged 8 to 12 weeks were used in this study. Mice were maintained in accordancewith the statement of the Association for Research in Vision and Ophthalmologyregarding the use of animals in research. A mixture of ketamine hydrochloride(30 mg/kg) and xylazine hydrochloride (125 mg/kg) was used for anesthesiaand administered by means of intraperitoneal injection.

Human IRBP peptide 1-20 (GPTHLFQPSLVLDMAKVLLD) was synthesized by conventionalsolid-phase techniques⁹ on a peptide synthesizer(TakaraBio Inc, Shiga, Japan). Complete Freund adjuvant and Mycobacterium tuberculosis H37Ra were purchased from Difco Labs, Detroit,Mich. Purified Bordetella pertussis toxin was purchasedfrom Sigma-Aldrich Corp, St Louis, Mo.

Mice were immunized subcutaneously in the thigh and neck with a peptide(200 µg) derived from h-IRBP peptide in a 0.2-mL emulsion in completeFreund adjuvant that had been supplemented with M tuberculosis strain H37Ra to 5 mg/mL and were given 100 ng of purified Bordetella pertussis toxin intraperitoneally as an additional adjuvant.Eyes were collected and assessed on day 21 after immunization. In anotherexperiment, spleen cells were collected from immunized mice on day 21 afterPEC transfer, and T cells were enriched using a magnetic, activated cell-sortinghigh-gradient separation column purchased from Miltenyi Biotec (Auburn, Calif)that yields 99% CD3⁺ cells, as confirmed by flow cytometry. PurifiedT cells were injected intraperitoneally (40 × 10⁶ to 50 ×10⁶ cells/mouse) into naive C57BL/6 mice. The presence of EAU wasassessed by histopathological examination 14 days after the adoptive transfer.Eyes were fixed in Bouin solution. Sections of samples were embedded in paraffinand stained with hematoxylin-eosin for histopathological study. The EAU scorewas evaluated in a masked fashion. Occurrence and severity of EAU were examinedfor each eye and scored on a scale of 0 to 4 in half-point increments, accordingto a semiquantitative system described previously.¹¹ The1.0 grade of EAU means that inflammatory cell infiltration into the entireretina is observed, and anatomical retinal layers are partially destroyed.¹¹ The 0.5 grade of EAU means that retinal layer structuresare completely preserved, but slight cell infiltration into the retina andciliary body is observed.¹¹

Vasoactive intestinal peptide (5 nmol/mouse) was administered intraperitoneallyon alternate days until day 21 after h-IRBP peptide immunization. In anotherexperiment, VIP was administered at different times during the efferent phaseof EAU. In each experiment, a control group of mice was injected with phosphate-bufferedsaline (PBS) alone.

Mice received an intradermal injection of 20 µg/10 µg ofh-IRBP peptide suspended in PBS into the right ear pinnae. After 24 and 48hours, the ear swelling was measured by a micrometer (Mitutoyo, Tokyo, Japan).

Peritoneal exudate cells were obtained from naive B6 mice that had received2 mL of intraperitoneal 3% thioglycolate solution 3 days earlier. The cellswere washed and suspended, then placed in 24-well culture plates (8 ×10⁵/well). Based on the report by Taylor et al⁶ thatthe concentration of VIP in rabbit aqueous humor is 12nM, we treated PECswith 12nM VIP or PBS in serum-free medium at 37°C in an atmosphere of5% carbon dioxide. The h-IRBP peptide was added at concentrations of 10 mg/mL.After overnight culture, plates were washed 3 times with culture medium toremove VIP and nonadherent cells. Adherent cells were cooled for 30 minutesat 4°C, dislodged from the culture plate by vigorous pipetting, washedtwice, and resuspended in PBS. When analyzed by means of flow cytometry forexpression of F4/80, greater than 99% of these cells were positive (data notshown), and we used the cells as macrophages. A total of 100 µL of PBScontaining 2 × 10⁵ PECs was injected into the tail vein ofnaive syngeneic mice. Thirty minutes later, the recipients received a subcutaneousinjection of a uveitogenic regimen of h-IRBP peptide in complete Freund adjuvant.

Analysis of ear swelling was performed by using analysis of varianceand the Scheffé test. We used analysis of variance and the Scheffétest to determine any significant difference that existed among all 4 experimentalgroups. We chose these tests to correct for multiple comparisons. We comparedthe EAU scores using the Mann-Whitney test. Means were considered to be significantlydifferent when P<.05.

To determine whether VIP treatment suppresses h-IRBP peptide–specificimmune responses in vivo, as revealed by suppression of antigen-specific DH,C57BL/6 mice were immunized with h-IRBP peptide as described. The mice thenreceived intraperitoneal VIP on alternate days for 3 weeks after immunization.Twenty days after immunization, the ears of the animals underwent challengewith h-IRBP peptide. As the results in Figure1A show, VIP impaired the capacity of mice immunized subsequentlyby means of h-IRBP peptide to display DH response (VIP entire group) (P = .02). Furthermore, to determine the potential effectof VIP on the DH response after the immune system had already been exposedto an uveitogenic regimen, panels of mice that received VIP on days 8 through20 after immunization also underwent testing for DH. The results, depictedin Figure 1A, show that mice treatedwith VIP in this manner displayed feeble DH (VIP efferent group) (P<.001). Thus, pretreatment of B6 mice with VIP impairs the capacityof these mice to acquire DH responses. Furthermore, delayed VIP treatmentsuppresses the expression of DH responses.

To estimate the potential effect of VIP on the development of EAU, weadministered VIP intraperitoneally on alternate days for 3 weeks after immunizationwith h-IRBP peptide. The development of EAU was assessed by means of histopathologicalexamination of the eye tissue sections. As shown in Figure 1B, EAU developed in 8 of 10 control mice (mean score, 0.85;median score, 0.75; range, 0-2.0). In contrast, EAU developed in only 4 of10 VIP-recipient mice (mean score, 0.3; median score, 0; range, 0-1.0) (Figure 1B) (P =.049). As shown in Figure 2A, inflammatorycell infiltration was present in the vitreous and the retina, and damagedphotoreceptor layer and retinal vasculitis were observed in control mice.In contrast, pathological findings in VIP-treated mice were distinctly milder(Figure 2B). These results indicatea potent preventive effect of VIP on the development of EAU. Furthermore,the result shown in Figure 1A encouragedus to attempt to suppress DH after the induction of immunity with h-IRBP peptide(on the efferent limb of immune response to h-IRBP peptide). Therefore, todetermine the potential effect of VIP on EAU after the immune system had beenexposed to an uveitogenic regimen, VIP was administered on days 8 through20 after immunization. As shown in Figure1C, VIP blocked disease development (mean score, 0.1; median score,0; range, 0-0.5) compared with control group results (mean score, 0.85; medianscore, 0.75; range, 0-2.0) (P = .003) and led toan improvement in the pathological score and occurrence of EAU (control group,8/10 mice; VIP-treatment group, 1/9 mice). These results suggest that VIPhas a suppressive effect on established EAU.

We next wished to determine whether intravenous injection of in vitro–treatedPECs with VIP had the capacity to abrogate EAU. The PECs were harvested fromC57BL/6 mice, pulsed with h-IRBP peptide, and incubated overnight with VIP.The cells were then harvested, washed free of h-IRBP peptide and VIP, andinjected intravenously into naive B6 mice. Thirty minutes later, the recipientsreceived a uveitogenic regimen, as described previously. The results are shownin Figure 3. Uveitis was detectedin the eyes of only 3 of 10 mice (mean score, 0.3; median score, 0; range,0-0.5) that received intravenous injection of h-IRBP peptide–pulsedPECs cultured overnight in VIP. In contrast, uveitis was detected in the eyesof 8 of 10 mice (mean score, 1.2; median score, 1.0; range, 0-2.0) that receivedintravenous injection of h-IRBP peptide–pulsed PECs cultured overnightin PBS only (P = .004). Thus, mice that receivedPECs treated with VIP and that carried an h-IRBP peptide–specific signaldisplayed significantly reduced evidence of EAU.

To test the postulate that VIP-treated PECs generate regulatory T cellsin the spleen that interfere with development of EAU, we examined whetherthe splenic T cells primed by VIP-treated PECs could transfer tolerance tonaive recipient mice. Forty million to 50 million T cells obtained from spleensof mice that first received VIP-treated PECs were transferred to syngeneicrecipients. Within 30 minutes, these recipients were immunized with h-IRBPpeptide and complete Freund adjuvant. Mice that received T cells derived frommice injected with VIP-treated PECs displayed lower scores of uveoretinitisand a lower occurrence of disease (Figure4). Experimental autoimmune uveoretinitis developed in 5 of 8 miceinjected with PBS-treated PECs (control group), with a mean EAU score of 0.5(median score, 0.5; range, 0-1.0), whereas EAU developed in only 1 of 6 miceinjected with VIP treated PECs, with a mean EAU score of 0.08 (median score,0; range, 0-0.5) (P = .08). These data demonstratethat VIP-treated PECs induce the generation of CD3⁺ regulatoryT cells that are capable of suppressing the h-IRBP peptide–specificpathogenic T-cell responses of uveitis.

The data presented herein show that the neuropeptide VIP is effectivein suppressing EAU, a murine experimental model for human uveitis. In thisstudy, we demonstrated that (1) VIP had a significant prophylactic effecton both the occurrence and severity of EAU; (2) VIP was effective in amelioratingestablished EAU; and (3) PECs pulsed with h-IRBP peptide in the presence ofVIP possessed the ability to ameliorate EAU via inducing h-IRBP peptide–specificregulatory T cells.

It is important to gain insight into the mechanisms by which VIP down-regulatesDH to h-IRBP peptide. It is possible that VIP directly down-regulates immuneeffector T cells. Taylor et al⁶ reported thatVIP suppresses antigen-specific lymph node cell proliferation and interferon γproduction in vitro. Furthermore, it has been shown recently that mice deficientin VIP receptors (VPAC₂R), which are highly up-regulated on T_H1 cells, display enhanced DH.¹² Ourresults showed that injection of VIP during the induction of immunity withh-IRBP peptide suppressed DH expression and EAU occurrence and severity. Inaddition, VIP even inhibited DH responses to h-IRBP peptide after the immunesystem had been exposed to a uveitogenic regimen, indicating that VIP impairsthe capacity of mice immunized with h-IRBP peptide to mount DH responses duringthe afferent and efferent limbs of the immune response. These findings arein agreement with the view that VIP has the capacity to suppress inductionand expression of T_H1 immune responses.

Hara et al¹³ showed that PECs incubatedwith IRBP in the presence of aqueous humor or supernatants from cultured iris/ciliarybody induced anterior chamber–associated immune deviation in vivo (impairedDH and IgG2a antibody production) when injected into naive, syngeneic mice,and in vitro–treated PECs had the capacity to abrogate EAU. Similarly,the present study showed that PECs pulsed with h-IRBP peptide in the presenceof VIP conferred on recipient mice impaired DH responses to h-IRBP peptide(data not shown) and protected those mice against EAU. Moreover, VIP-treatedPECs induced regulatory T cells that were able to suppress the T_H1response after immunization, which may explain the cure of uveitis after onset.However, how and where regulatory T cells are generated and alter the immuneresponse in vivo is still unclear and needs further investigation.

This is not the first report to suggest a role of VIP in the developmentof experimental intraocular inflammation. Although our data indicate thattreatment with VIP is effective at inhibiting EAU, Zhang and colleagues¹⁴ reported that treatment with VIP exacerbated theinflammatory process of endotoxin-induced uveitis and that the treatment decreasedthe production of tumor necrosis factor α and elevated the serum levelof interleukin 10. This difference in the effectiveness of VIP in endotoxin-induceduveitis and EAU may be due to the different features of ocular inflammationin these models. Moreover, Katayama¹⁵ previouslyshowed that high-dose intravitreous injection of VIP resulted in miosis andthe breakdown of the blood-aqueous barrier, whereas low-dose injection ofVIP induced mydriasis. These findings suggest that VIP plays a role as a proinflammatoryfactor at high concentrations and has the capacity for different immune responsesat low concentrations. However, in the present study, we demonstrated thatthe transfer of macrophage cultured with VIP at the intraocular concentration(12nM) significantly suppressed the induction of EAU. Also, systemic injectionof VIP (5 nmol/mouse) was effective for amelioration of uveitis. These findingsare supported by a recent report showing that VIP treatment (5-nmol dose)reduced the incidence and severity of collagen-induced arthritis.¹⁶ The in vitro study also showed that the VIP inhibitionof messenger RNA expression of proinflammatory cytokines (such as tumor necrosisfactor α and interleukin 1β) induced by collagen-induced arthritiswas found at concentrations of 10 ^{− 8}M to 10⁻⁶M.¹⁶ Therefore, although VIP might exertdual effects on induction of ocular inflammation, our results indicate thatVIP provides a highly suppressive effect on EAU at the intraocular concentrationof VIP.

We have demonstrated that VIP has the ability to ameliorate EAU andto down-regulate established pathogenic T_H1 responses to h-IRBPpeptide. Moreover, h-IRBP peptide–pulsed PECs, cultured in the presenceof VIP, acquired the capacity to suppress DH responses to antigen and preventthe development of EAU by inducing regulatory T cells. Vasoactive intestinalpeptide might offer a new approach for the management of clinical uveitis.

Correspondence: Takeshi Kezuka, MD, PhD, Department of Ophthalmology,Tokyo Medical University, 6-7-1 Nishishinjuku, Shinjuku-ku, Tokyo, 160-0023Japan (tkezuka@tokyo-med.ac.jp).

Submitted for publication July 14, 2003; final revision received November6, 2003; accepted December 3, 2003.

This study was supported in part by grant 13771047 from Grant-in-Aidfor Encouragement of Young Scientists and grant 15791009 from Grant-in-Aidfor Young Scientists (B) from the Japan Society for the Promotion of Science,Tokyo.

We thank J. Wayne Streilein, MD, Schepens Eye Research Institute, Boston,Mass, for critical reading of the manuscript, and J. Patrick Barron of theInternational Medical Communications Center of Tokyo Medical University, Tokyo,for his revision of the manuscript.