Phenotypes of T Cells Infiltrating the Eyes in Autoimmune Anterior Uveitis Associated with EAE

Abstract

purpose. Lewis rats immunized with myelin basic protein (MBP) develop experimental autoimmune encephalomyelitis (EAE) and associated anterior uveitis (AU), which recurs. The goal was to analyze cellular activation markers and adhesion molecules of T cells that infiltrate the eyes and spinal cord during acute and recurrent AU in actively and passively induced diseases simultaneously in the same animals.

methods. EAE-AU was induced in Lewis rats by immunization with MBP in CFA, or by adoptive transfer of MBP-specific T-cell lines, and the signs of clinical EAE and AU was scored. Cells isolated from the iris-ciliary body were tested by flow cytometry for expression of CD4, CD8, CD45RC, T-cell receptor (TCR) Vβ8.2, α4 integrin, l-selectin, CD44, and CD134.

results. Ocular T cells showed a significantly higher expression of CD62L (l-selectin) than did T cells in the spinal cord. In addition, a much lower percentage of infiltrating CD8⁺ T cells was found in the eyes during AU. In passive transfer experiments, T-cell lines derived from acute and recurrent uveitis showing similar phenotypes differing in specificities but possessed the capacity of inducing both AU and EAE. Pretreatment of rats with effector CD4⁺ T cell before MBP immunization did not induce suppression of EAE or AU. However, pretreatment with regulatory CD8⁺ T cells significantly reduced the severity and duration of both EAE and AU.

conclusions. T cells recruited into the inflamed eyes or central nervous system (CNS) are mainly activated/memory T cells expressing different levels of l-selectin. Regulatory CD8⁺ T cells may contribute to the susceptibility of the eye to recurrent AU. The differences in phenotypes of T cells recruited simultaneously to two different organs suggest that microenvironment also plays a role in determining lymphocyte homing.

Experimental autoimmune encephalomyelitis (EAE) is a CD4⁺ T-helper cell type 1 (Th1)-mediated inflammation of the central nervous system (CNS). EAE can be induced in Lewis rats by immunization with myelin basic protein (MBP) in CFA or can be adoptively transferred by MBP-specific activated T cells. At the same time, the rats exhibit an inflammation of the anterior segment of the eye called anterior uveitis (AU). Possible target antigens are myelinated neurons situated within the CNS and iris. EAE is generally regarded as a model for studying the etiology and pathogenicity of multiple sclerosis (MS), and most studies have focused on disease in the spinal cord and brain. However, there is little information on ocular disease during EAE. Uveitis also can be transferred to naïve animals by activated T-cell lines specific for MBP or its peptides. ¹ The dominant encephalitogenic MBP epitope, residues 69 to 89, is capable of inducing AU, suggesting a common effector mechanism for EAE and AU. ^{2 3} Moreover, other encephalitogenic epitopes were found to be uveitogenic. ^{1 3} Rats recovered from EAE are resistant to subsequent reinduction of EAE with MBP but develop AU. ⁴  

Lymphocytes play a significant role in autoimmune ocular inflammation. Cell adhesion molecules are involved in the recruitment of those specific leukocytes into the target tissue. However, their role in uveitis is not well defined. Cells infiltrating the eye express surface molecules that direct other leukocyte migration during ocular inflammation. Therefore, surface molecules of T lymphocytes are actively involved in the recruitment of a specific leukocyte subset into the eye over the course of disease. Moreover, the activation status and ability to express adhesion molecules are vital for T cells to migrate to the spinal cord (SC) and the eye because of the existence of the blood-brain barrier and the blood-ocular barrier in those organs. The inflammatory cells infiltrating the iris-ciliary body in AU in the Lewis rat are composed of T cells, monocytes, and granulocytes. ^{3 5} In those earlier experiments, we have shown that CD4⁺ T cells that infiltrate the iris-ciliary body express activation markers, such as T-cell receptor (TCR) Vβ8.2 and OX40 antigen in acute AU. ³  

The phenotype of T cells that infiltrate the spinal cord has been studied in EAE. It has been shown that in EAE in mice, T cells are mainly activated/memory (CD44^high/LFA-1^high/ICAM-1^high/CD45RB^low). These CNS-seeking T lymphocytes are phenotypically distinct from the T cells found in the inflamed lung, subcutaneous tissue, and gut. ⁶ T cells that infiltrate the CNS express α₄β₁ integrin, but do not express α₆, α_E, α₇, and l-selectin. T cells with a high expression of l-selectin have been found in the gut and subcutaneous tissue, and α_Eβ₇ is expressed in inflamed lungs. ⁷ These studies suggest that the existence of phenotypically distinct T cell populations recruited to different tissues and the role of the local environment in defining specificity of inflammatory cells. Susceptibility of the eye, but not the CNS, in Lewis rats to recurrent inflammation after reimmunization with MBP also suggests the difference between the microenvironments of the organs. Based on those observations, we hypothesized that a unique cell population may be recruited to the eye. 

In the current study, we investigated the phenotypes of T cells recruited into inflamed eyes during acute and recurrent AU. Primary immunization with MBP induced both EAE and AU in Lewis rats, but reimmunized rats were resistant to EAE, although susceptible to recurring AU. Results of a previous study showed that recurrent AU (RAU) is caused by a subset of CD4⁺ T cells that recognize new MBP epitopes and use a diverse repertoire of TCRs. ⁴ MBP-specific T cells, derived from the actively immunized rats, are capable of transferring EAE and AU into naïve rats, suggesting the encephalitogenic and uveitogenic potential of those cells. However, the capacity of CD4⁺ T cells derived from RAU to transfer AU or EAE has not been studied. Moreover, the phenotypes of the pathogenic T cells that infiltrate target organs have not been well studied in AU or RAU. In addition, there is limited information on infiltrating T cell phenotypes during eye inflammation; however, results thus far have shown that in human idiopathic uveitis, there is a significant increase in CD4⁺ CD25⁺ and CD69⁺ CD4⁺ T cells in T-cell infiltrate within the aqueous. ^{8 9} In the current study, we examined the capacity of T cells to transfer the disease and analyzed the phenotype of T cells that infiltrate the eyes during acute AU and RAU in actively and passively induced disease. We characterized the expression of cellular activation markers and adhesion molecules simultaneously in the same animals, not only in the eyes but also in spinal cords and circulating blood. 

Materials and Methods

Induction and Assessment of AU and EAE

Female LEW rats (8 to 12 weeks old; Harlan Sprague-Dawley, Indianapolis, IN) were used in this study. The rats were kept in germ-free conditions, according to institutional and federal guidelines, at the Animal Care Facility, Neurologic Sciences Institute, Oregon Health and Science University. All animal experimentation procedures adhered to the ARVO Statement for the Use of Animals in Ophthalmic and Vision Research and have been approved by the institutional animal experimentation committee. 

EAE was induced by subcutaneous injection of 25 μg guinea pig MBP in CFA supplemented with 150 μg Mycobacterium tuberculosis strain H37Ra (Difco, Detroit, MI). The rats were assessed daily for changes in clinical signs as follows: 0, no signs; 1, limp tail; 2, hind leg weakness, ataxia; 3, paraplegia; and 4, paraplegia with forelimb weakness, a moribund condition. Clinical signs of ocular inflammation were also scored by biomicroscopy, according to the following scale: 0, normal; 1, slight iris vessel dilation and thickened iris stroma, a few scattered inflammatory cells, or both; 2, engorged blood vessels in iris; abnormal pupil contraction, occasional cells in vitreous; 3, hazy anterior chamber, decreased red reflex; and 4, marked cells in vitreous. Rats were reinjected after they had recovered from uveitis with the same dose of 25 μg MBP in CFA usually 40 days after primary immunization. We defined recovery as having a score of 0 for two consecutive days and a relapse as a score higher than 0 for at least two consecutive days after a recovery. 

MBP-Reactive Cell Lines and Adoptive Transfer

Lymphocytes harvested from draining lymph nodes of the cervical, inguinal, and popliteal areas and the spleen of immunized animals were used to develop MBP-specific cell lines. Cells were cultured at a density of 1 × 10⁶/mL in RPMI 1640 containing 10% fetal bovine serum, gentamicin (50 μg/mL), amphotericin B (2.5 μg/mL), 1% sodium pyruvate, and 50 μM 2-mercaptoethanol. MBP was used at 20 μg/mL to stimulate lymphocytes for 3 days. At the end of stimulation, cells were rested in culture medium containing 10 U/mL human recombinant (r)IL-2 (Sigma, St. Louis, MO) for 7 to 10 days. Subsequently, the cycles of stimulation and rest were repeated to expand the specific T-cell population. Syngeneic irradiated thymocytes were used as antigen-presenting cells (APCs) at a ratio of 1:20 (T cells to APCs) and then were rested in culture medium containing 10 U/mL human rIL-2 for 7 to 10 days. For adoptive transfer, 1 × 10⁷ MBP-activated T cells were administrated intraperitoneally per naïve rat. The clinical signs were monitored daily and scored as described. 

A regulatory cell line was prepared from normal Lewis rat spleen cells. ¹⁰ Splenic T cells were stimulated with irradiated (2000 rad) resting MBP-reactive T cells for 3 days, and the cycles of stimulation and rest were repeated to expand the specific T-cell population. For adoptive transfer, we used 1 × 10⁷ T cells after two rounds of stimulation. 

Isolation of Infiltrating Cells

Cells infiltrating the iris-ciliary body were isolated as described previously. ³ Briefly, the iris-ciliary body tissue was obtained from the anterior portion of the eye by microdissection. The tissue was incubated in RPMI medium containing 10% FBS and 1 mg/mL collagenase for 2 hours at 37°C, with occasional agitation by pipetting. At the end of the incubation, a single-cell suspension was prepared by filtering through a nylon filter. Cells were washed twice before use. 

Spinal cords were removed from rats by insufflation and dissociated by gently grinding the tissue into a single-cell suspension inside a nylon filter with the plunger of a syringe. Lymphocytes were isolated by Percoll gradient centrifugation (Amersham Pharmacia Biotech, Piscataway, NJ). The dissociated tissue pellet from two spinal cords was mixed with 7 mL 80% Percoll (Amersham Pharmacia Biotech) and placed on top of 2 mL 100% Percoll (90% Percoll and 10% 10× PBS). An additional 6 mL of 40% Percoll was overlayered onto the preparation. The cells were recovered from the interface and washed twice before use. 

Cell Surface Maker Labeling and Flow Cytometric Analysis

Cells in aliquots of 1 × 10⁶ cells per tube were labeled with monoclonal antibodies (mAbs) against rat cell surface markers, purchased from PharMingen (San Diego, CA). After a wash with flow cytometry washing buffer (PBS containing 2% FBS and 0.01 M sodium azide), the cells were pelleted by centrifugation and resuspended in 50 μL washing buffer. An anti-CD4 mAb conjugated with phycoerythrin (PE; 1 μL) and one of the following antibodies conjugated to FITC were added to each tube: anti-CD8, anti-CD45RC, anti-TCR Vβ8.2, anti-α4 integrin, anti-l-selectin, anti-CD44, and anti-CD134 (1 μL). A PE-conjugated mouse IgG1 or FITC-conjugated mouse IgG1 was used as an isotype-matched control to establish background staining and to set the quadrants before calculating the percentage of positively stained cells. After incubation for 20 minutes on ice, the cells were washed two times by centrifugation and resuspended in 0.5 mL serum-free PBS for immediate data acquisition or in 0.5 mL flow cytometry fixative (PBS with 10% formalin, 2% glucose, and 0.01 M sodium azide) for overnight storage at 4°C. 

Lymphocyte Proliferation Assay

The lymphocyte proliferation assay (LPA) was performed in 96-well plates in triplicate with RPMI medium containing 10% FBS, 5 × 10⁻⁵ M 2-mercaptoethanol, and 50 μg/mL gentamicin. Cells were seeded at a density of 2 × 10⁵ per well and incubated with RPMI medium only, 1 μg ConA, or 10 μg MBP at 37°C in 5% CO₂ for 72 hours and then pulsed with 1 μCi tritiated thymidine per well for an additional 18 hours. The cells were harvested onto a glass fiber filter, and the thymidine uptake was assessed by liquid scintillation counting (model 1250 Betaplate counter; Wallac, PE Life Sciences, Boston, MA). The data were expressed as a stimulation index (SI), which was calculated by dividing the proliferation (counts per minute incorporated) measured in the presence of antigen by the proliferation measured with medium alone. Stimulation was considered positive if the SI of immunized rats was equal to or greater than twice the background (SI = 2). 

Rat Cytokine ELISA

Cytokine levels of IFN-γ and IL-10 in cell culture supernatants were determined by specific sandwich ELISA kits from BioSource International, (Camarillo, CA), according to the manufacturer’s procedure. A 50-μL sample of each supernatant taken from cultures at 48 hours was used for the ELISA. Cytokine concentration was determined from standard curves, using recombinant standards supplied by the manufacturer. 

Statistical Analysis

Student’s t-test was used to calculate the probability in comparison of the data between experiment groups. P < 0.05 was considered significant. 

Results

T-Cell Subsets in Actively Induced AU and RAU

Because Lewis rats had a recurrence of AU which developed after immunization with the second dose of MBP, we asked whether the T-cell population found in the eye during acute AU differs from the cell population in the recurrent phase. To perform a qualitative comparison of T cells, we isolated cells from inflamed organs by collecting the infiltrating cells from the same animals on the same day and then subjecting them to a flow cytometric analysis. The type of T cells that infiltrated the target organs is presented in Figure 1 , Table 1 . Based on the absolute number of infiltrating cells, approximately five times more CD4⁺ than CD8⁺ T cells were found in the eyes at both the AU and RAU stages (Table 1) . The ratio of CD4⁺ to CD8⁺ in the T cells that infiltrated the eye was almost the same in both phases of uveitis (Fig. 1 , bottom). In the spinal cord, the percentage of infiltrating CD8⁺ T cells was approximately four times higher in RAU than in AU (P < 0.01), which produced a markedly increased in the CD4⁺-to-CD8⁺ ratio in AU, when compared with that in RAU (P < 0.001) or blood (P < 0.01). In RAU, the spinal cord CD4⁺-to-CD8⁺ ratio decreased because the organ was mostly infiltrated with CD8⁺ T cells. In conclusion, the main difference between the two organs was in the number of CD8⁺ T cells present, as indicated by the CD4⁺-to-CD8⁺ ratio, which was high in AU, but low in RAU in spinal cord cells, whereas it remained very similar in the eye T cells. In the peripheral blood, draining lymph nodes and spleen, there was no evident difference in the ratios of CD4⁺ to CD8⁺ T cells among the naïve, AU, and RAU groups (Fig. 1) . 

Activation Markers in Actively Induced AU

Our previous studies have shown that T cells that infiltrate the iris express primarily TCR Vβ8.2, which has been associated with the pathogenicity of MBP-specific T cells. ^{11 12} In the current studies we found that CD4⁺ T cells from the iris at the peak of clinical signs of RAU also expressed TCR Vβ8.2 but in a lower number of cells (49% and 34%, respectively; Fig. 2 , Table 2 ). Moreover, the percentage of CD4⁺ T cells that expressed TCR Vβ8.2 was higher in eyes than in spinal cords. The expression of CD45RC was downregulated, and CD44 was upregulated in infiltrating CD4⁺ T cells of both AU and RAU, which was shown by fluorescence intensity (Fig. 3) . Therefore, these infiltrating T cells were mainly activated/memory CD44^high/CD45RC^low CD4⁺ T cells. Compared with T lymphocytes in peripheral blood, a high level of CD134 and CD25 expression was detected on T cells in the eyes and spinal cords during AU and RAU (up to 6% in blood vs. 30% to 50% in target organs at both stages). Our findings are summarized in Table 2 . 

Adhesion Molecules

To determine whether a specific adhesion phenotype is associated with the localization of T cells in the eye, we examined the expression of VLA-4 and CD62L on T cells recruited to the eye and for comparison to the spinal cord. The expression of α4 integrin (VLA-4) was increased in infiltrating T cells when compared with T cells in the blood, on average approximately 23% vs. 85% to 90% in the organs. However, there was a difference in the level of l-selectin expressed (CD62L) on CD4⁺ T cells between the eye and the spinal cord as well as between AU and RAU (Fig. 3 , Table 2 ). A higher percentage of T cells that infiltrated the eye expressed CD62L in RAU (73%) than in AU (55%). In recurrent disease, most of the T cells in the eye were positive for l-selectin in contrast to CD4⁺ T cells in the spinal cord (73% vs. 25%). 

Adoptive Transfer of AU and EAE with T Cells Derived from Active AU and RAU

Active immunization of Lewis rats with MBP induces EAE and AU. These recovered rats develop a resistance to further attempts to induce active EAE, although they are still susceptible to AU (Fig. 4A) . This suggests that spontaneous remission of EAE coincides with the termination of the pathogenic immune response and development of a resistance to reinduction of EAE but not to recurrence of AU. To determine whether T cells from different stages of disease have the ability to induce AU and EAE, we developed T-cell lines derived from acute EAE-AU and recurrent AU. T-cell lines, regardless of the postimmunization time of cell collection or the source of cells (draining lymph nodes or spleen), were capable of transferring AU and EAE after activation with MBP (Fig. 4B) . The clinical signs of EAE and AU caused by passive transfer were similar in severity and duration. To address the question of antigenic specificity, we performed a T-cell proliferation assay with MBP and its peptides. T-cell lines derived from AU were mostly specific to the major immunopathogenic peptide 69-89, whereas T-cell lines derived from RAU showed a broader specificity toward different MBP epitopes, including peptide 69-89 (Fig. 4C) . This was similar to our earlier findings indicating the shift in epitope recognition after reimmunization with MBP. ⁴ Phenotypically, both MBP-stimulated T cell lines were similar in the percentage of cells that expressed CD4, CD25, CD44, CD62L, CD134, and VLA-4 (Fig. 4D) . However, T-cell lines derived from AU differed from the T-cell lines derived from RAU significantly in the number of cells that expresses TCR Vβ8.2 (78% vs. 21%) and CD45RC (30% vs. 8%). Our findings suggest that phenotypical similarities and the recognition of encephalitogenic and uveitogenic epitopes enabled those T cells to induce both EAE and AU. 

Phenotypes of Cells that Infiltrate Eyes after Adoptive Transfer

To examine the phenotypes of CD4⁺ T cells that infiltrate the iris-ciliary body and spinal cord after adoptive transfer of MBP-specific T cells derived from the acute and recurrent phase of AU, we performed immunofluorescence staining on the isolated cells from the inflammation sites, iris-ciliary body, and spinal cord. In peripheral blood, lymph nodes, and spleen, the ratio of CD4⁺ to CD8⁺ T cells was higher in recipients of cells derived from AU, compared with recipients of T cells derived from RAU. The ratio was lower in the target organs, except for spinal cords of recipients of cells derived from AU (Fig. 5) . We observed a significantly higher number of CD4⁺ cells in the eyes when T cells derived from acute AU were used for transfer than when T cells from the recurrent phase were used (P < 0.01). In contrast to the CD8⁺ T cell population found in the spinal cord after actively induced disease, no significant increase was found in CD8⁺ T cells after passive transfer (Fig. 5) . The composition of T cells in the spinal cord after transfer of both cell lines was compared with that of the first episode of EAE after initial immunization. 

The phenotypes of CD4⁺ T cells found in both target organs are summarized in Table 3 . The CD4⁺ T cells that infiltrated the iris-ciliary body showed considerably higher expression of CD62L and CD134 than did CD4⁺ T cells found in the spinal cord. The expression of l-selectin was again higher in the eyes of recipients of T-cell lines derived from both AU and RAU, but was downregulated in the spinal cords, which was similar to the results from the actively immunized animals. It suggests that different mechanisms may be involved in CD4⁺ T-cell migration to the eye than to the spinal cord. Based on the fluorescence intensity in the flow cytometric analysis, T cells were characterized by high intensity of CD44 and low CD45RC (not shown), indicating that T cells mostly consisted of activated/memory (CD44^high/CD45RC^low) cells. 

Effect of Passively Transferred Effector CD4⁺ T Cells before Immunization with MBP

To investigate whether the administration of effector CD4⁺ T cells provides an environment that inhibits induction of AU and EAE by subsequent immunization with MBP, naïve rats first received 10⁷ MBP-activated CD4⁺ T cells by adoptive transfer and then were immunized with MBP. Those rats showed signs of clinical AU and EAE that lasted for several days (Fig. 6A) , after which animals recovered, but they did not have total resistance to further attempts to cause disease. There were noticeable differences in severity and duration. EAE developed with much lower severity. In the spinal cord, CD4⁺ T-cell phenotypes were comparable to cells found in acute EAE. However, the CD4⁺ to CD8⁺ ratio was close to 1:1 (not shown). In contrast, AU resulted in accelerated onset and stronger severity when compared with the first episode. The cell population found in the iris-ciliary body was similar to that of RAU (not shown). 

Effect of Passively Transferred CD8⁺ T Cells before Immunization with MBP

In the next experiment, we asked whether regulatory T cells have the capacity to inhibit the development of AU. We developed a suppresser cell line that was composed of 66% CD8⁺ and 34% CD4⁺ cells. After stimulation, more than 85% of CD8⁺ cells expressed CD25, as determined by flow cytometry. Rats pretreated with these cells (10⁷ T cells/animal) did not show any symptoms of EAE or AU. Ten days after transfer, the rats were immunized with MBP in CFA. Figure 6B shows that the rats had both diseases, but the severity and duration were significantly reduced (maximum AU score of 1 compared with 3 in control animals, P < 0.01). This CD8⁺ T cell line inhibited proliferative responses of the effector CD4⁺ T cell line in vitro (Fig. 6C) , and in contrast to the CD4⁺ T cell line, it produced high levels of IL-10 and low levels of IFN-γ (Fig. 6D) . 

Discussion

AU is associated with EAE after immunization of Lewis rats with MBP. These animals were resistant to further induction of EAE; however, RAU developed after they were reimmunized. In this study, we have shown that MBP-specific T-cell lines derived from donors with either AU or RAU were capable of transferring both EAE and AU to naïve rats. It suggests that memory CD4⁺ T cells at from both stages of disease, and in particular from RAU, had the capacity to cause inflammation in both eye and spinal cord, although initially they did not cause EAE in actively reimmunized animals. Some unknown in vivo factors at the RAU stage abrogated the potential of MBP-specific T cells to induce inflammatory responses in the CNS, but still sustained their capacity to induce inflammation of the eye’s anterior segment. Two possible mechanisms can be proposed. One is that the active suppression of inflammatory processes may occur in the spinal cords of MBP reimmunized rats but not in passively immunized rats. Another is the existence of a subset of CD4⁺ T cells with a unique phenotype, which may influence the migration of circulating T cells into the target organs. 

In this model, rats that recovered from active EAE were protected from the second induction of disease by MBP reimmunization and from recurrences but not from recurrence of AU. One of the mechanisms for controlling the outgrowth and differentiation of activated CD4⁺ T cells and their downregulation resides in regulatory T cells, including both the CD4⁺ and CD8⁺ phenotypes. It has been shown that the regulatory T cell subset in the peripheral lymphoid organs consisting of CD8⁺ T cells may be responsible for resistance in rats to EAE. ^{10 13} These regulatory T cells that are present in naïve animals respond to MBP-reactive T cells and expand substantially in actively immunized animals. Earlier studies have shown that resistance to active induction of EAE could be induced by pretreating animals with nonpathogenic inoculation of autoantigen or effector cells. ^{14 15 16} However, in contrast to those studies, our pathogenic CD4⁺ T effector cell line did not provide resistance to disease. Moreover, those cells provided a favorable environment to AU, suggesting that MBP-reactive T cells were not effective in stimulation of regulatory cells and in affording protection from disease. There is evidence, however, that CD8⁺ T cells can regulate immune responses by controlling the Th phenotype of MBP-reactive CD4⁺ T cells. ¹⁷ One possibility is that CD4⁺ and CD8⁺ T cells stimulate or inhibit activated CD4⁺ T cells by secreting cytokines and, in effect, determine whether cells become IFN-γ secreting Th1 cells or IL-4 secreting Th2 cells. Thus, the secretion of particular cytokines acts directly on the target CD4 cells as a suppressive factor. During acute AU, cytokines such as IL-4 and IL-10 were not detected in the eye but were detected the spinal cord. ¹⁸ Such an environment could facilitate recurrence. 

In addition, passive transfer of regulatory T cells, characterized by expression of IL-10 and suppression of the proliferation of CD4 effector cells, reduced clinical AU and EAE. The fact that we observed a significantly higher number of CD8⁺ T cells in the spinal cord than in the eye supports the notion that CD8⁺ T cells contribute to the resistance of EAE. In contrast, the CD4⁺-to-CD8⁺ ratio in the eyes remained almost the same in AU and RAU, suggesting that the absence of suppression due to the low number of CD8⁺ T cells may in part be responsible for susceptibility to RAU. Uchio et al. ¹⁹ showed that the development of experimental autoimmune uveoretinitis (EAU) could be prevented by adoptive transfer of CD4⁺ T cells, whereas CD8⁺ T cells do not prevent onset. However, postrecovery CD4⁺ T cells fail to inhibit EAU induced by passive immunization with uveitogenic T cells. Based on these findings the investigators suggest that suppressor CD4⁺ T cells may play an important role in the remission of EAU. ¹⁹ Furthermore, mice without CD8⁺ T cells have milder but more chronic EAE, suggesting that CD8⁺ T lymphocytes may act as both effectors and regulators. ²⁰ Depletion of CD8⁺ T cells in mice predisposes them to the second induction of EAE. ²¹ These findings, along with our data, suggest that CD8⁺ T cells may be important players in resistance to EAE at the stage of restimulation of Lewis rats with MBP, but other factors also should be considered, such as CD4⁺ T cell phenotype and local environment. 

T cells that infiltrate the target organs consisted of the activated/memory CD4⁺ T cells expressing CD44^high/CD45RC^low. The T cells in the eye closely resembled those from the spinal cord in their pattern of expression of activation markers. CD4⁺ T cells expressed TCR Vβ8.2 associated with pathogenicity of EAE and AU. ^{11 22} However, there were differences in the number of CD4⁺ T cells that expressed l-selectin (CD62L), between acute and recurrent AU. Selectins are essential in the primary step of leukocyte migration, but the role of CD62L on CD4⁺ T cells in EAE is controversial. It has been reported that T cells from inflamed CNS in mice are also mainly CD4⁺ and express a typical activated/memory phenotype: CD44^high/LFA-1^high/ICAM-1^high/CD45RB^low but did not express l-selectin. ⁶ The investigators reported that the CNS T cells express α4- and β1-integrin, proteins involved in primary and secondary adhesion. They propose the existence of a phenotypically distinct CNS-seeking T-lymphocyte population. Another study has shown that treatment with an mAb to l-selectin effectively suppresses rat EAE induced by active immunization, but has only a mild inhibitory effect on EAE induced by adoptive transfer. ²³ Moreover, activated T-cell lines and clones express CD44 and α4-integrin, but not l-selectin and enter the CNS independent of their antigenic specificity. mAbs directed against CD44 and α4-integrin prevent the transfer of EAE by MBP-specific T cells; however, anti-CD62L have no effect on homing of encephalitogenic T cells to the inflamed CNS in mice. ²⁴ In our studies, the expression of l-selectin in infiltrating T cells found in spinal cords of EAE rats was low, which may be due to the difference in types of lymphocytes recruited into the inflamed sites. It is also possible that the blood-brain barrier and blood-ocular barrier play an active role in inflammatory cell recruitment, and endothelial cells adhesion molecules may allow access of different cell population. ^{24 25 26 27}  

Our study showed that T-cell lines derived from the acute or recurrent phase of AU were capable of transferring uveitis and encephalomyelitis. The similarity of infiltrating T cells in target tissues between actively and passively immunized animals showed that the cell phenotype is important in the pathogenic process. This may agree with the recent studies on the fate of MBP-specific T cells after adoptive transfer. Flugel et al. ²⁸ have demonstrated that before onset of EAE, migratory T effector cells downregulate the activation markers (CD25, CD134) but upregulate several chemokine receptors. ²⁸ After entering the CNS, these T cells are reactivated by local APC-bearing autoantigen. Therefore, at this stage, the phenotypes of infiltrating T cells in the CNS of actively or passively immunized rats would be similar. 

In conclusion, we provide new insight into a possible contribution of regulatory CD8⁺ T cells to the susceptibility of the eye to recurrent AU, although determining their role in pathogenicity needs to be further investigated. It is still not clear whether relapse of AU is regulated by specific T-cell responses or bystander mechanisms. However, based on our earlier studies and these findings, we conclude that the microenvironment of the eye, lacking regulatory cytokines (IL-4 and IL-10) and CD8⁺ T suppressor cells, determines in part the recruitment of specific T cells. Our findings also imply that the nature of the target site, rather than the antigenic specificity of the T cell is important in facilitating homing of the cells. A combination of those effects may contribute to the susceptibility of the eye to RAU. 

 Submitted for publication September 28, 2001; revised January 2, 2002; accepted January 14, 2002.

 Supported by Grant EY12477 from the National Eye Institute (GA).

 Commercial relationships policy: N.

 The publication costs of this article were defrayed in part by page charge payment. This article must therefore be marked “advertisement” in accordance with 18 U.S.C. §1734 solely to indicate this fact.

 

Figure 1.

 

CD4⁺ and CD8⁺ T cells in peripheral blood, lymphoid, and target organs. Top: Bars represent the percentage of CD4⁺ and CD8⁺ T cells infiltrating the iris-ciliary body (I/CB) and spinal cord (SC) determined by flow cytometric analysis. Bottom: Bars represent the ratio of CD4⁺ to CD8⁺ T-cells in peripheral blood, lymph nodes, spleen, I/CB, and SC. Note that in SC the CD4⁺-CD8⁺ ratio is high in AU but low in RAU, whereas in the I/CB the CD4⁺-CD8⁺ ratio is similar. Data are expressed as the mean ± SD of three to five separate experiments. Results were considered significantly different when **P < 0.01, ***P < 0.001.

Table 1.

 

Number of Organ-Infiltrating Cells at the Peak of AU and RAU in the Lewis Rat

Organ	Disease Stage	Recovered Cells (n)*	CD4+ Cells (n)	CD8+ Cells (n)
Iris/ciliary body	AU	14 × 104	7 × 104	1.5 × 104
	RAU	7.2 × 104	2.6 × 104	0.5 × 104
Spinal cord	AU	90 × 104	27 × 104	3.6 × 104
	RAU	45 × 104	2.3 × 104	9 × 104

 Cell numbers were calculated in each eye or spinal cord.

*  Cells recovered from the iris/ciliary body or spinal cord were counted after trypan blue staining. The number of CD4⁺ and CD8⁺ cells was calculated based on the percentage of T cells determined by flow cytometry.

Figure 2.

 

Expression profiles of TCR Vβ8.2 and adhesion molecules of infiltrating CD4⁺ T cells analyzed by flow cytometry. The cells were surface stained with PE-conjugated anti-CD4 mAb and FITC-conjugated anti-Vβ8.2, anti-CD62L (l-selectin) and anti-VLA4 (α4 integrin). Cells harvested from eyes at (top) the peak of AU and (bottom) the peak of RAU.

Table 2.

 

Distribution of Surface Membrane Molecules in Blood and Target Organs’ CD4⁺ T Cells after Active Immunization

Cell Surface Markers	AU					RAU
		Blood	Iris/Ciliary	Spinal Cord	Blood	Iris/Ciliary	Spinal Cord
TCR Vβ8.2	6 ± 1	49 ± 6	31 ± 3	7 ± 1	34 ± 4	21 ± 1
CD44	83 ± 9	95 ± 5	97 ± 1	61 ± 24	81 ± 7	93 ± 4
CD45RC	62 ± 16	27 ± 6	31 ± 1	68 ± 3	23.5 ± 4	22 ± 2
CD25	5 ± 1	36 ± 4	41 ± 5	4 ± 1	33 ± 3	36 ± 3
CD134	3 ± 1	34 ± 9	48 ± 6	2 ± 1	13 ± 3	23 ± 3
l-selectin	65 ± 19	55 ± 3	45 ± 3	92 ± 2	73 ± 2	25 ± 3
α4 integrin	28 ± 5	77 ± 7	93 ± 3	19 ± 5	86 ± 1	91 ± 2

 Data represent the mean ± SD percentage of CD4⁺ cells of three separate experiments using three to five rats.

Figure 3.

 

Expression profiles of activation markers of infiltrating CD4⁺ T cells, analyzed by flow cytometry. The cells were surface stained with PE-conjugated anti-CD4 mAb and FITC-conjugated anti-CD44, anti-CD45RC, anti-CD134, and anti-CD25. Cells harvested from eyes at (top) the peak of AU and (bottom) the peak of RAU.

Figure 4.

 

Kinetics of actively and passively induced EAE and AU in Lewis rats and MBP-specific T-cell line characteristics. (A) Kinetics of EAE and AU after active immunization with MBP-CFA. Recurrent AU develops after reimmunization with MBP. Data represent clinical scores ± SD of EAE and AU recorded daily. Ten days after primary immunization with MBP-CFA, rats showed signs of development of EAE-AU. (B) Kinetics of EAE and AU after adoptive transfer of 1 × 10⁷ T cell lines derived from an episode of AU or RAU. Data represent clinical scores ± SD of EAE and AU recorded daily. Note that clinical features were similar in severity and duration, regardless of the origin of cells used for passive transfer (AU- or RAU-derived). (C) Specificities of responses of cell lines developed from cells derived from acute and recurrent phase of disease measured by lymphocyte proliferation assay. The proliferative responses were measured to whole MBP and its peptides. Note that T-cell line specificities were similar to epitope recognition during the appropriate phases of uveitis, with dominance of 69-89 epitopes in AU and diversity of recognition in RAU. (D) Phenotypes of CD4⁺ T cells in MBP-specific cell lines derived from AU and RAU, determined by flow cytometry. The expressions of TCR Vβ8.2, CD25, CD44, CD134, and l-selectin was significantly upregulated in both T cell lines. The expression of CD45RC decreased significantly (P < 0.01).

Figure 5.

 

Presence of CD4⁺ T cells and CD8⁺ T cells in the target organs after adoptive transfer of MBP-specific T-cell lines derived from AU and RAU. (A) Percentage of CD4⁺ and CD8⁺ T cells infiltrating the target organs after adoptive transfer of 1 × 10⁷ T cell lines derived from acute and recurrent AU, determined by flow cytometric analysis. (B) Ratios of CD4⁺ to CD8⁺ T cell in peripheral blood, lymph nodes, spleen, iris-ciliary body (I/CB), and spinal cord (SC). (□) Naïve blood cells; (▪) cells after passive transfer of AU-derived cells (AT-AU); (╡) cells after passive transfer of RAU-derived cells (AT-RAU). The results were pooled from three individual experiments, five rats per experiment. ***P < 0.001.

Table 3.

 

Distribution of Surface Membrane Molecules in Blood and Target Organs’ CD4⁺ T Cells after Adoptive Transfer of MBP-Specific T-Cell Lines

Cell Surface Markers	AU					RAU
		Blood	Iris/Ciliary	Spinal Cord	Blood	Iris/Ciliary	Spinal Cord
TCR Vβ8.2	6 ± 1	46 ± 1	23 ± 6	4 ± 1	41 ± 7	15 ± 5
CD44	87 ± 5	85 ± 3	95 ± 5	53 ± 3	82 ± 3	94 ± 5
CD45RC	58 ± 6	64 ± 7	24 ± 10	59 ± 6	45 ± 10	16 ± 6
CD25	4 ± 1	29 ± 1	26 ± 2	6 ± 1	32 ± 2	30 ± 2
CD134	2 ± 1	43 ± 7	33 ± 2	2 ± 1	47 ± 2	21 ± 2
l-selectin	86 ± 1	85 ± 1	44 ± 12	48 ± 7	80 ± 2	26 ± 8
α4 integrin	30 ± 8	74 ± 10	93 ± 7	20 ± 7	86 ± 2	93 ± 3

 Data represent the mean percentage ± SD of CD4⁺ cells of three separate experiments using three to five rats.

Figure 6.

 

Effect of MBP-specific effector and regulatory T cells on development of AU and EAE. (A) Kinetics of EAE and AU after adoptive transfer of 10⁷ MBP-specific CD4⁺ T cells followed by immunization with MBP-CFA. Data are the mean ± SD clinical score of EAE and AU recorded daily in five rats. (B) Kinetics of EAE and AU after adoptive transfer of 1 × 10⁷ regulatory CD8⁺ T-cell line followed by immunization with MBP-CFA. Data are the clinical mean ± SD score of EAE and AU recorded daily. (C) Proliferative responses of the effector cell line to MBP and its immunodominant peptide 69-89 measured by lymphocyte proliferation assay. Regulatory cells were added to effector cells at a 1:1 ratio. (D) Production of IFN-γ and IL-10 by effector and regulatory cell lines stimulated for 48 hours in vitro determined by ELISA.