Graefe's Archive for Clinical and Experimental Ophthalmology

, Volume 245, Issue 10, pp 1549–1557

CD4+CD25+Tregs express an increased LAG-3 and CTLA-4 in anterior chamber-associated immune deviation

Authors

  • Xuefei Zhu
    • State Key Laboratory of Ophthalmology, Zhongshan Ophthalmic Center, Uveitis Study CenterSun Yat-sen University
    • State Key Laboratory of Ophthalmology, Zhongshan Ophthalmic Center, Uveitis Study CenterSun Yat-sen University
  • Hongyan Zhou
    • State Key Laboratory of Ophthalmology, Zhongshan Ophthalmic Center, Uveitis Study CenterSun Yat-sen University
  • Bing Li
    • State Key Laboratory of Ophthalmology, Zhongshan Ophthalmic Center, Uveitis Study CenterSun Yat-sen University
  • Xiangkun Huang
    • State Key Laboratory of Ophthalmology, Zhongshan Ophthalmic Center, Uveitis Study CenterSun Yat-sen University
  • Qianli Meng
    • State Key Laboratory of Ophthalmology, Zhongshan Ophthalmic Center, Uveitis Study CenterSun Yat-sen University
  • Li Wang
    • State Key Laboratory of Ophthalmology, Zhongshan Ophthalmic Center, Uveitis Study CenterSun Yat-sen University
  • Aize Kijlstra
    • Animal Sciences Group van Wageningen UR Divisie Veehouderij
    • Department Ophthalmology, Eye Research Institute MaastrichtUniversity of Maastricht
Laboratory Investigation

DOI: 10.1007/s00417-007-0591-8

Cite this article as:
Zhu, X., Yang, P., Zhou, H. et al. Graefes Arch Clin Exp Ophthalmol (2007) 245: 1549. doi:10.1007/s00417-007-0591-8

Abstract

Background

Regulatory CD4+CD25+ T cells have been proven to be essential for maintenance of peripheral tolerance and autoimmune diseases. ACAID is a model of immune privilege in the eye. Relatively little is known about the role and phenotype of these regulatory CD4+CD25+ T cells in ACAID.

Methods

Injection of OVA into the anterior chamber of BALB/C mice was performed to induce ACAID. The frequencies of splenic CD4+CD25+ Tregs and the expression of CTLA-4 and LAG-3 on these cells were determined by flow cytometry. Magnetic cell sorting was used to isolate CD4+CD25+ and CD4+CD25T cells. The function of CD4+CD25+ T cells was detected by in vitro immunosuppression assays and in vivo adoptive transfer.

Results

ACAID was successfully induced following an i.c. injection of OVA. Frequencies of CD4+CD25+ and Tregs were significantly increased in ACAID mice as compared to those in controls. The CD4+CD25+ T cells stimulated with OVA in ACAID mice showed a stronger suppressive ability in vitro than those seen in non-ACAID mice. CD4+CD25+ T cells from ACAID mice, but not from non-ACAID mice, were able to suppress DTH responses in an antigen-specific manner following adoptive transfer. The frequencies of CTLA-4 or LAG-3 on Tregs in ACAID mice were higher as compared with those in naive mice.

Conclusion

Splenic CD4+CD25+Foxp3+T cells expressing CTLA4 and LAG3 play an important role in the induction of ACAID.

Keywords

Regulatory T cellsImmune regulationFoxp3ACAID

Introduction

Anterior chamber-associated immune deviation (ACAID), a model of systemic immune tolerance, is characterized by an antigen-specific suppression of delayed type hypersensitivity (DTH) and a selective deficiency of complement-fixing antibodies (Abs) [16, 17]. DTH is inhibited by two types of suppressor cells: one is a CD4+ T cell population that inhibits the induction and the other is a CD8+T cell population that suppresses DTH expression [20, 24].

CD4+T suppressor cells from ACAID mice are able to transfer immune tolerance to naïve mice [21]. Further study proved that CD4+T cells expressing CD25 and producing IL-10 are required for the development of ACAID [19]. CD4+CD25+ regulatory T cells include two populations: one naturally exists in thymus [10], the other is induced in the periphery [13]. A recent report suggests that development of ACAID is independent of the naturally occurring CD4+CD25+ regulatory T cell population [11]. This suggests a possible role for antigen-specific CD4+CD25+ regulatory T cells during ACAID. In the study presented here, we focused on a possible role of these regulatory T cells from the spleen in the development of ACAID. This was performed by quantifying CD4+CD25+ T cells during the development of ACAID using a set of novel T cell markers such as Foxp3, LAG-3 and CTLA-4.

Foxp3 is considered as an important marker for regulatory T cells, and CD4+CD25+Foxp3+ T cells are preferentially regarded as regulatory T cells (Tregs) [3, 4, 8]. LAG-3 selectively marks the induced regulatory CD4+CD25+T cells [9], and CTLA-4 modulates the suppressive activity of CD4+CD25+ T cells by providing co-stimulatory signals [22].

Our results showed that Tregs were increased in ACAID mice. Over 90% CD4+CD25+ T cells from ACAID mice, positive controls, negative controls or naïve mice were Foxp3 positive. Splenic Tregs expressed high levels of CTLA-4 and LAG-3 in ACAID mice. Furthermore, we showed that CD4+CD25+ T cells from the spleens of ACAID mice could transfer tolerance when adoptively transferred into naïve mice.

Materials and methods

Mice

Female 6- to 8-week old BALB/c mice were used in the experiments. Mice were obtained from the Animal Center of the Sun Yat-sen University (Guangzhou, China). Mice were treated humanely and in accordance with the ARVO statement for the use of animals in vision and ophthalmology research.

Induction of ACAID and DTH assay

ACAID was induced as previously described [16]. In brief, ACAID mice received an injection of 2-μl ovalbumin (OVA, 20 mg/ml, Sigma, St. Louis, MO) dissolved in PBS into the anterior chamber (AC) and were immunized on day 7 by subcutaneous (s.c.) injection 250 μg/200 μl of OVA or bovine serum albumin (BSA, Amresco, USA) emulsified in Complete Freudn’s adjuvant (CFA, Sigma, St. Louis, MO). Mice only receiving a s.c. injection of OVA in CFA alone served as positive controls. Mice receiving an intracameral injection of 2 μl sterile PBS alone acted as negative controls. Untreated mice were used as naïve controls.

DTH to specific alloantigen was measured using a conventional ear swelling assay [16]. Ag (200 μg of OVA/10 μl of PBS) and an equal volume of sterile PBS were respectively injected into the right and left ear pinnae. Ear pinnae of experimental and control animals were measured with a Mitutoyo engineer’s micrometer (Mitutoyo, Japan) immediately before challenge and 24 h later.

Phenotypes of Tregs determined by flow cytometry

Splenocytes were suspended in PBS and 0.1% BSA at a final concentration of 1 × 106 cells/100 μl. These cells were incubated with FITC-anti-CD3 mAb, FITC or APC-anti-CD4 mAb, PE-cy7-anti-CD25 mAb and PE-anti-LAG-3 mAb for 30 min in the dark at 4°C. Cells were washed, fixed and permeabilized and stained with APC or PE-anti-Foxp3 mAb, PE-anti-CTLA-4 mAb or matched isotype according to the manufacturer’s instructions and analyzed by flow cytometry (FCM). All conjugated antibodies were purchased from eBioscience and BD Pharmingen (San Diego, CA).

CD4+CD25+ and CD4+CD25T-cell selection

CD4+CD25+ T cells were isolated from mouse splenocytes with a CD4+CD25+ Regulatory T Cells Isolation kit (Miltenyi Biotec, Auburn, CA). Briefly, spleen cell suspensions were incubated with a cocktail of biotin-conjugated antibodies and anti-biotin microbeads for negatively sorting CD4+T cells. The enriched CD4+T cells were incubated with PE-labeled CD25+ and anti-PE microbeads for positively sorting CD4+CD25+T cells. The purities of isolated CD4+CD25+ or CD4+CD25T cells identified by flow cytometric analysis both were shown to be higher than 90%.

In vitro suppression assay of CD4+CD25+T cells

Splenocytes isolated from naïve BALB/c mice were used as antigen presenting cells (APCs) and were treated with 50 μg/ml mitomycin-C (MMC, Sigma, St. Louis, MO) for 45 min at 37°C.These cells were washed three times with RPMI-1640 (Gibco, USA) before incubation in proliferation assays. Purified CD4+CD25T cells from positive controls served as effector cells, and purified CD4+CD25+T cells acted as regulator cells. Effector cells (2 × 104 cells/well) cultured with or without OVA (200 μg/ml), MMC-treated splenocytes (8 × 104 cells/well) and regulator cells (2 × 104 cells/well) from each group were incubated in RPMI-1640 medium, supplemented with 10% heat-inactivated FCS, 50 μM 2-ME, 2 mM L-glutamine, 0.1 mM nonessential amino acids, 1 mM sodium pyruvate and 1% streptomycin/penicillin at 37°C for 72 h and pulsed with 1 μCi/well [3H]thymidine (Shanghai Institute of Applied Physics, Chinese Academy of Sciences, China) during the last 16 h of incubation. T cell proliferation was determined by the incorporation of [3H]thymidine. Proliferation data were the mean value of triplicate wells and represented three independent experiments.

In vivo suppression assay of CD4+CD25+T cells

The isolated CD4+CD25+T cells or CD4+CD25T cells were washed three times with PBS, resuspended in PBS and subsequently injected intravenously into the tail vein of naive BALB/c mice at a concentration of 5 × 105 or 5 × 106/200 μl. Twenty-four hours later, recipients were s.c. immunized with OVA in CFA or BSA in CFA. On day 7, DTH responses were assayed according to the aforementioned method.

Statistics

The results were expressed as the mean±SEM. Data were analyzed by ANOVA using SPSS11.0. A value of P < 0.05 was considered significantly different.

Results

Frequency of splenic CD4+Tregs is increased in ACAID mice

CD4+CD25+T cells have been shown to play a critical role in immune tolerance and immune responses in various diseases and experimental models of disease in rodents and humans [5, 7, 12]. Our study aimed to examine whether CD4+CD25+T cells were increased in ACAID, an eye-derived immune tolerance model. The results showed that ACAID was successfully induced by injection of OVA into the anterior chamber of the mice as evidenced by a markedly decreased delayed type hypersensitivity response (Fig. 1). The frequency of splenic CD4+CD25+ T cells in CD4+T cells from ACAID mice was significantly higher than that from non-ACAID groups (P < 0.01) (Fig. 2a).
https://static-content.springer.com/image/art%3A10.1007%2Fs00417-007-0591-8/MediaObjects/417_2007_591_Fig1_HTML.gif
Fig. 1

ACAID was successfully induced. ACAID mice received an intracameral injection of OVA followed by s.c. immunization with OVA in CFA after 7 days. Positive controls were immunized s.c. with OVA in CFA. Negative controls received an intracameral injection of sterile PBS. Naïve mice did not receive any stimulation. Mice were challenged 14 days later by injection of antigen into the right ear pinnae and an equal volume of sterile PBS into the left ear pinnae. Delayed-type hypersensitivity responses were measured 24 h later. The results shown are the means±SEM of three experiments performed. In the ACAID group (n = 4), DTH responses were significantly suppressed compared with those in the positive control group (n = 4). *P < 0.001

https://static-content.springer.com/image/art%3A10.1007%2Fs00417-007-0591-8/MediaObjects/417_2007_591_Fig2a_HTML.gifhttps://static-content.springer.com/image/art%3A10.1007%2Fs00417-007-0591-8/MediaObjects/417_2007_591_Fig2b_HTML.gifhttps://static-content.springer.com/image/art%3A10.1007%2Fs00417-007-0591-8/MediaObjects/417_2007_591_Fig2c_HTML.gif
Fig. 2

The frequencies of CD4+CD25+ and CD4+CD25+Foxp3+T cells among CD4+T cells. All the spleens from each group were harvested on day 7 after the last treatment. Splenocytes were prepared as described in the methods section and stained with anti-CD3, anti-CD4, anti-CD25 mAb and anti-Foxp3 mAb. Data in the quadrant indicate the percentage of positive cells, and derive from gating on CD3+CD4+ cells. a: The frequency of CD4+CD25+ among CD4+T cells; b: the frequency of CD4+CD25+Foxp3+ among CD4+T cells; c: the percentages of CD4+CD25+Foxp3+ on CD4+CD25+ T cells or CD4+CD25Foxp3+ on CD4+CD25T cells. The frequencies of CD4+CD25+ and CD4+CD25+Foxp3+T cells among CD4+T cells in the ACAID group (n = 4) are significantly increased compared with those in the positive control (n = 4). There is no significant difference in the percentages of CD4+CD25+Foxp3+ on CD4+CD25+ T cells or CD4+CD25Foxp3+ on CD4+CD25T cells between the ACAID group and the positive control group. One representative dot blot of four independent experiments is shown. The results shown in the histogram are the means±SEM of four independent experiments. *P < 0.05

As CD25 is not a definitive marker for regulatory T cells, Foxp3 was used to further identify these cells [3, 4, 8]. The results showed that the frequency of splenic CD4+CD25+Foxp3+ in CD4+T cells in ACAID mice were slightly, but significantly increased as compared with that observed in the control groups (P < 0.05) (Fig. 2b), although the frequencies of Foxp3 in CD4+CD25+T cells were similar (all being more than 90%) (Fig. 2c). Foxp3 was also found to be expressed by approximately 7% of CD4+CD25T cells. There was no significant difference among these four groups concerning the frequency of CD4+CD25Foxp3+T cells in the spleen.

Splenic CD4+CD25+T cells from ACAID mice have a stronger suppressive ability in vitro

To characterize the function of CD4+CD25+T cells and CD4+CD25 T cells in vitro, these two populations were purified from naïve mice, the positive control and the ACAID group, and proved to be over 90% pure (Fig. 3a). Purified CD4+CD25+T cells from three groups were unresponsive to stimulation with OVA and APCs in vitro. A slight proliferation of CD4+CD25T cells was observed upon exposure to APC alone, whereas these CD4+CD25T cells showed vigorous proliferation after stimulation with OVA and APCs. We subsequently examined whether the splenic CD4+CD25+T cells from ACAID mice had a stronger regulatory activity than those from other groups. The results revealed that CD4+CD25+T cells from each group could inhibit the proliferation of effector cells (CD4+CD25T cells) in the presence or absence of OVA. There was no significant difference concerning the suppressive ability among these three groups without OVA. However, CD4+CD25+T cells from ACAID mice had a much stronger suppressive ability than those obtained from naïve mice (P < 0.01) or positive controls (P < 0.01) in the presence of OVA. There was no difference concerning the suppressive ability between negative controls and positive controls (Fig. 3b).
https://static-content.springer.com/image/art%3A10.1007%2Fs00417-007-0591-8/MediaObjects/417_2007_591_Fig3_HTML.gif
Fig. 3

Suppressive effect of CD4+CD25+T cells on CD4+CD25 effector T cells in vitro. a: The purity of isolated of CD4+CD25+ (left) and CD4+CD25 T (right) cells by MACS; b: the suppressive effect of CD4+CD25+T cells on the proliferation of CD4+CD25 T cells. Splenocytes from naïve mice were treated with mitomycin-C as APC. CD4+CD25 effector T cells (2 × 104 cells/per well) from positive control were cultured with or without OVA (200 μg/ml), APC (8 × 104 cells/per well) and equal numbers of CD4+CD25+T cells (regulator cells, 2 × 104 cells/per well), respectively, from naïve mice, positive controls and ACAID mice. The cells were cultured for 72 h and pulsed with [3H]thymidine for the last 16 h of culture. Splenic CD4+CD25+T cells stimulated with OVA have a stronger suppressive ability when obtained from ACAID mice than from the other two groups.Data are the means±SEM of triplicate samples and are representative of three independent experiments. *P < 0.01

Adoptively transferred CD4+CD25+T cells from ACAID mice significantly suppress DTH responses

The isolated CD4+CD25+T cells (5 × 105 cells per mouse) or CD4+CD25T (5 × 105 or 5 × 106 cells per mouse) from naïve mice, positive control or ACAID mice were adoptively transferred into the tail vein of syngeneic mice. These recipients were s.c. immunized with OVA in CFA or BSA in CFA 24 h later. On day 7 DTH responses were assayed. The results, as shown in Fig. 4, revealed that the CD4+CD25+T cells from ACAID mice, but not from non-ACAID mice, could adoptively transfer DTH inhibition. When a large number of CD4+CD25 T cells (5 × 106 cells per mouse) from ACAID mice were transferred, DTH responses could also be suppressed, but not as strongly as when CD4+CD25+T cells had been transferred. Moreover, transferred CD4+CD25+T cells from the ACAID mice induced by OVA did not show any effect on the ear swelling of those mice receiving a s.c. injection of BSA in CFA and challenged with BSA, showing the antigen specificity of the transferred cells (P < 0.01).
https://static-content.springer.com/image/art%3A10.1007%2Fs00417-007-0591-8/MediaObjects/417_2007_591_Fig4_HTML.gif
Fig. 4

DTH responses to adoptive transferred cells. Splenic CD4+CD25+ and CD4+CD25 T cells from naïve mice, positive controls and ACAID mice were sorted by MACS. Isolated CD4+CD25+T cells (5 × 105 cells/per mouse), CD4+CD25T cells (5 × 106 cells/per mouse) or PBS were injected intravenously into the tail vein of syngeneic mice. At 24 h, the recipients received a s.c. injection of OVA or BSA in CFA. On day 7, the mice were challenged by injection of OVA (10 μl) into the right ear pinnae and an equal volume of sterile PBS into the left. DTH responses were measured 24 h later. Splenic CD4+CD25+ T cells or large numbers of CD4+CD25 T cells from ACAID mice transferred suppression of DTH responses. Each group contained four animals. The experiments were repeated two times. All results are expressed as mean±SEM. * P < 0.01

Expression of CTLA-4 and LAG-3 on CD4+Tregs in ACAID mice is increased

As CTLA-4 and LAG-3 have been shown to be involved in the function of regulatory CD4+CD25+T cell, we went on to study the expression of CTLA-4 and LAG-3 on Tregs in ACAID [9, 22]. The results showed that the frequency of Foxp3+CTLA-4+ in CD4+CD25+T cells (Fig. 5a) was significantly increased in ACAID mice and positive controls as compared with that in naïve and negative controls (P < 0.01). There was no significant difference concerning the frequency of Foxp3+CTLA-4+CD4+CD25+T cells between ACAID mice and positive controls, although a tendency for a light increase was observed in the former. The frequency of Foxp3+LAG-3+ in CD4+CD25+T cells (Fig. 5b) in ACAID mice was significantly higher than that in naïve, negative controls and positive controls (P < 0.01).
https://static-content.springer.com/image/art%3A10.1007%2Fs00417-007-0591-8/MediaObjects/417_2007_591_Fig5_HTML.gif
Fig. 5

The frequencies of CTLA-4+Foxp3+ or LAG-3+Foxp3+cells among CD4+CD25+T cells. All the spleens from each group were harvested on day 7 after the last treatment. Splenocytes were prepared as described in the methods section and CD4+T cells were required by negative sorting. CD4+T cells were stained with anti-CD4, anti-CD25, anti-Foxp3 and anti-CTLA-4 or anti-LAG-3 mAb. Data in the quadrants indicated the percentage of positive cells that derived from gating on CD4+CD25+T cells. a: The frequencies of CTLA-4+Foxp3+ among CD4+CD25+T cells; b: the frequencies of LAG-3+Foxp3+cells among CD4+CD25+T cells. The frequency of CTLA-4+Foxp3+ or LAG-3+Foxp3+cells among CD4+CD25+T cells in ACAID group (n = 4) was markedly increased compared with the negative control group (n = 4). One representative dot blot of four independent experiments is shown. The results shown in the histogram are the means±SEM of four independent experiments. *P < 0.01

Discussion

The results presented in this paper provide strong evidence for an important role for CD4+CD25+Foxp3+T cells (Tregs) expressing a high level of CTLA-4 and LAG-3 in the development of ACAID. Our earlier work has shown that CD4+PD-1+T cells played a potent regulatory role in ACAID. PD-1, a negative regulatory molecule in the CD28/B7 family, may be a marker for CD4+ regulatory cells [14]. In the present paper, our experiments showed that the frequency of Tregs was significantly increased in ACAID mice compared to the non-ACAID groups. A vast majority of CD4+CD25+T cells (>90%) in all groups were Foxp3 positive. CD4+CD25+T cells from ACAID mice displayed a stronger inhibitory ability to CD4+CD25T cells in vitro as compared with that observed in naïve mice and positive controls. CD4+CD25+T cells from ACAID mice, but not from non-ACAID mice, were able to adoptively transfer tolerance to naïve mice. Our further phenotypic analysis of Tregs in ACAID mice showed that these cells expressed higher levels of both CTLA-4 and LAG-3. Taken together, these results suggest that both CD4+PD-1+T cells and CD4+CD25+Foxp3+T cells may be the subpopulations of the CD4+ suppressor cell in ACAID.

It has been shown that ACAID is a phenomenon mediated or maintained by generating an active regulatory T cell population to suppress DTH [15]. Regulatory CD4+T cells are one of the suppressive populations responsible for the development of ACAID [2]. These CD4+ T cells are highly enriched with a subpopulation expressing the IL-2Rα chain (CD25) [13]. In our study, the frequencies of CD4+CD25+ and CD4+CD25+Foxp3+T cells were significantly higher in ACAID mice as compared with those in non-ACAID groups. In an attempt to exclude the influence of intracameral injection on the frequency of CD4+CD25+T cells as well as CD4+CD25+Foxp3+T cells, an experiment with PBS injection into the anterior chamber was performed. This experiment showed that this manipulation did not influence the frequency of these cells. All these results suggest that an upregulation of Tregs occurs during ACAID.

Tregs are clearly shown to be increased following an intracameral injection of OVA in our study. It is important to test if these cells have a suppressive property. In vitro experiments revealed that this population without stimulation had a suppressive ability on CD4+CD25 effector T cells, no matter if they were obtained from ACAID mice, positive mice or naïve mice. Upon stimulation with OVA, these CD4+CD25+T cells from ACAID mice showed a stronger suppressive activity than those from the other two groups. These results suggest that OVA-specific CD4+CD25+Tregs are induced during ACAID. Our in vivo experiment showed that DTH responses were impaired by adoptively transferring splenic CD4+CD25+T cells from ACAID mice, but not from non-ACAID mice. These results indicate that splenic CD4+CD25+T cells in ACAID mice are not identical to natural Tregs in naïve mice, and seem to have a stronger immunosuppressive capacity. As ACAID is an immune tolerance induced by foreign antigens, it is presumed that these CD4+CD25+T cells with an antigen-specific inhibitory function (Tregs) play a crucial role in the development of this immune deviation. Future studies using Foxp3−/− mice may further clarify their role. Additionally, time response studies should be performed in the future to determine the half-life of the suppressive activity of CD4+CD25+T cells generated during ACAID. It is interesting to note that a ten-fold higher number of CD4+CD25T cells from ACAID mice, when adoptively transferred to naïve mice, could also suppress DTH responses, although the suppression was weaker than when transferring CD4+CD25+T cells. This result suggests that a subset of regulatory T cells with a weak suppressive property or in a very small frequency also exists in the CD4+CD25T cells subpopulation. It has been found that a small proportion of CD4+CD25T cells express Foxp3 [4]. In our experiment we also found that about 7% of these T cells express Foxp3. It is possible that the observed suppressive activity could be ascribed to these Foxp3-expressing CD4+CD25T cells This is consistent with a recent report that both CD4+CD25+ and CD4+CD25 regulatory cells mediate dominant transplantation tolerance [6]. We cannot rule out the possibility that contamination with CD4+CD25+T cells may also be involved in the inhibition of the DTH response observed when using high numbers of CD4+CD25T cells. Activation of CD25 cells into CD25+ cells following injection can also explain the observed findings. Further studies are needed to clarify this issue.

As CD4+CD25+Foxp3+T cells in ACAID mice showed an increased frequency and suppressive ability, we further explored the phenotype of this regulatory population. CTLA-4 and LAG-3, two important molecules for Tregs, were studied for their expression in Tregs from ACAID mice as well as other groups. Our study revealed an increased expression of LAG-3 in Tregs in ACAID mice as compared with that seen in the various control groups used, suggesting that LAG-3 may be involved in the specific regulation in this immune tolerance. The results also showed an increased expression of CTLA-4 in Tregs in ACAID mice as well as in the positive group, although a little higher expression was observed in the former. As CTLA-4 is a marker of activated Tregs, it is likely that these CTLA-4-expressing Tregs in the positive control are those primed by complete Freund’s adjuvant used in this study. As the number of LAG-3+ T cells are very low and CD4+CD25+Foxp3+CTLA-4+ T cells are not isolated by MACS, further studies with CTLA-4 or LAG-3 knockout mice or using anti-CTLA-4 or anti-LAG-3 antibodies to block the corresponding pathway will greatly help us to clarify the exact role of CTLA-4 or LAG-3 in the induction of ACAID. A number of regulatory molecules, such as GITR, PD-1 and BTLA, have been recently documented [1, 2, 18, 23]. It is also necessary to investigate their role in Tregs in the development of ACAID.

In conclusion, our results suggest that CD4+CD25+Foxp3+T cells expressing a high level of CTLA-4 and LAG-3 represent an important subset of CD4+Tregs in ACAID. However, further studies on the phenotype and function of regulatory cells will provide more profound insight into the development of ACAID.

Acknowledgments

The authors thank Prof. Dr. Changyou Wu for his constructive suggestions and Dr. Haining Zhang for her technical assistance.

Copyright information

© Springer-Verlag 2007