Homing to Cardiac Allografts for Double-Negative Regulatory T Cell ...

The Journal of Immunology

CXCR5/CXCL13 Interaction Is Important for Double-Negative Regulatory T Cell Homing to Cardiac Allografts1 Boris P.-L. Lee,*† Wenhao Chen,*† Hui Shi,* Sandy D. Der,† Reinhold Fo¨rster,‡ and Li Zhang2*†§ Accumulating evidence indicates that regulatory T (Treg) cells control development of various diseases both systemically and locally. However, molecular mechanisms involved in Treg cell homing remain elusive. We have shown previously that ␣␤TCRⴙCD3ⴙCD4ⴚCD8ⴚ double-negative (DN) Treg cells selectively accumulate in tolerant allografts to maintain localized immune regulation. However, the molecular mechanism leading to the accumulation of DN Treg cells in tolerant grafts was not known. Our cDNA microarray analysis revealed significant up-regulation of chemokine receptor CXCR5 mRNA in DN Treg clones compared with nonregulatory clones. In this study, we examined the importance of CXCR5 in mediating DN Treg migration. Compared with CD4 and CD8 T cells, both primary DN Treg cells and clones constitutively express high levels of CXCR5 protein, enabling them to migrate toward increasing CXCL13 gradients in vitro. After infusion into recipient mice, CXCR5ⴙ DN Treg clones, but not their CXCR5ⴚ mutants, preferentially accumulated in cardiac allografts and could prevent graft rejection. Furthermore, we found that allogeneic cardiac allografts express high levels of CXCL13 mRNA compared with either recipient native hearts or nontransplanted donor hearts. Ab neutralization of CXCL13 abrogated DN Treg cell migration in vitro and prevented in vivo homing of DN Treg clones into allografts. These data demonstrate that DN Treg cells preferentially express CXCR5, and interaction of this chemokine receptor with its ligand CXCL13 plays an important role in DN Treg cell migration both in vitro and in vivo. The Journal of Immunology, 2006, 176: 5276 –5283.

A

ccumulating evidence indicates that regulatory T (Treg)3 cells are critical for the maintenance of immune tolerance by limiting the responses of effector CD4 and CD8 T cells. Many studies have shown the importance of Treg cells using models of autoimmunity (1, 2), allergy (3), infections (4), cancer (5), and transplantation (6). Various subsets of Treg cells have been identified in both animal models and humans. These include the most commonly studied CD4⫹CD25⫹ Tregs (7) and IL-10-producing type I regulatory T cells (8) and CD4⫹CD103⫹ T cells (9), CD4⫹CD25⫺ Tregs (10), CD8⫹CD28⫺ T cells (11), ␥␦ T cells (12), NK T cells (13), and ␣␤TCR⫹CD3⫹NK1.1⫺CD4⫺CD8⫺ double-negative (DN) T cells (14 –18). DN Treg cells comprise 1–3% of peripheral T lymphocytes in rodents (15) and 1% of total peripheral blood CD3⫹ T cells in healthy volunteers (18). Recently, it has been demonstrated that both murine and human DN Treg cells can down-regulate immune responses mediated by syngeneic CD4 and CD8 T cells in vitro and in vivo in an Ag-specific manner (16 –19).

*Toronto Medical Discovery Towers, University Health Network, Toronto, Ontario, Canada; †Department of Laboratory Medicine and Pathobiology, University of Toronto, Toronto, Ontario, Canada; ‡Institute of Immunology, Hannover Medical School, Hannover, Germany; and §Department of Immunology, University of Toronto, Toronto, Ontario, Canada Received for publication December 28, 2005. Accepted for publication February 21, 2006. The costs of publication of this article were defrayed in part by the payment of page charges. This article must therefore be hereby marked advertisement in accordance with 18 U.S.C. Section 1734 solely to indicate this fact. 1 This work is supported by Canadian Institutes of Health Research (MOP 14431) and National Cancer Institute of Canada (Grant 15067) (to L.Z.). Additional funding is provided by Wyeth-Ayerst Canada. 2 Address correspondence and reprint requests to Dr. Li Zhang, Toronto General Research Institute, University Health Network, 101 College Street, TMDT 2-807, Toronto, Ontario, Canada, M5G 1L7. E-mail address: [email protected] 3

Abbreviations used in this paper: Treg, regulatory T; DLI, donor lymphocyte infusion; DN, double negative; lpr, lymphoproliferative.

Copyright © 2006 by The American Association of Immunologists, Inc.

Although it has been demonstrated repeatedly that Treg cells can be isolated from spleens of tolerant mice, recent work using transplantation models has suggested that T cells infiltrating tolerated kidney (20) and skin grafts (19, 21) are enriched for Treg cells when compared with splenic T cells. Similarly, it was found recently in various tumor models that CD4⫹CD25⫹ Treg cells accumulated within tumors (22–24). Depletion of intratumor CD4⫹CD25⫹ Treg cells led to eradication of well-established highly aggressive tumors (22, 23). Collectively, these findings suggest that Treg cells may suppress immune responses locally. Although these studies demonstrate Treg cells can home into transplanted allografts, autoimmune diseased organs, and tumor sites, the molecular mechanism that is used by Treg cells to migrate is currently unknown. We have demonstrated that pretransplant donor lymphocyte infusion (DLI) can activate recipient DN Treg cells and induce longterm skin graft survival in single class I or class II locus-mismatched mouse models (15, 17, 19, 25). Interestingly, it was found that DLI-activated DN Treg cells can home into allogeneic skin grafts and form the majority of graft-infiltrating T cells in accepted donor-specific, but not the third-party rejecting allografts (19). Furthermore, these graft-infiltrating DN Treg cells can suppress syngeneic CD8 effector T cell activity through a cytotoxic mechanism (19). We have established DN Treg cell clones from DLItreated tolerant mice. These DN Treg clones can specifically suppress syngeneic CD8 effector T cells in vitro (15, 26, 27), and prevent the rejection of donor-specific skin and cardiac allografts in the absence of other immunosuppressants (15, 26, 27). From these DN Treg clones arose spontaneous mutants that lost suppressive capacity in vitro (26) and failed to enhance allograft survival when infused into syngeneic mice (26, 27). To identify molecules that are critical for DN Treg cell function, we compared differences in gene expression between DN Tregs and their mutants using cDNA microarray and quantitative real-time PCR analysis. 0022-1767/06/$02.00

The Journal of Immunology The chemokine receptor CXCR5 mRNA was highly expressed in DN Treg clones compared with their natural mutants (26). Chemokine receptors and their ligands are crucial for lymphocyte trafficking under both homeostatic and inflammatory conditions. The receptor CXCR5 was originally identified from Burkitt’s lymphoma as a G protein-coupled receptor (28), and subsequently found in B cells and a small subset of CD4⫹ T cells (29 –31). To date, CXCL13 has been the exclusive ligand for CXCR5, which initiates chemotaxis toward an increasing gradient in vitro and to B cell zones in lymph nodes that express CXCL13 (32). More recently, CXCR5 expression has also been shown in DN T cells in mice (26, 30), a subset of CD4 follicular Th cells (33, 34), and some tonsillar CD4⫹CD25⫹ Tregs in humans (29). In CD4 follicular Th cells, it was demonstrated recently that expression of CXCR5 is crucial for transient homing into B cell zones for efficient Ab production (35). In general, however, the role of CXCR5 in Treg cell migration is poorly understood. The goal of this study was to determine whether CXCR5 expression on DN Treg cells plays an important role in their migration. In this study, we demonstrate that both DN Treg clones and primary DN Treg cells express a high level of CXCR5 protein on their surface, and that DN Treg cells, but not CD4 or CD8 T cells, can use this receptor to migrate toward CXCL13 in vitro. Furthermore, CXCR5⫹ Treg clones, but not their mutants, are able to migrate to and accumulate in cardiac allografts, and prevent graft rejection after adoptive transfer. Upon neutralization of CXCL13, DN Treg cell migration was abrogated in vitro and in vivo. These results demonstrate that CXCR5-CXCL13 interaction plays an important role in DN Treg cell migration, and suggest that modulation of this interaction may help control DN Treg cell migration and their regulatory function locally, and thus modulate the development of various diseases.

Materials and Methods Mice C57BL/6 ⫻ BALB/c (CB6F1) and BALB/c-H-2-dm2 (dm2) mice were purchased from The Jackson Laboratory. A breeding stock of 2C transgenic mice (C57BL/6 background) was provided by D. Loh (Nippon Research Center, Kanagawa, Japan). The 2C mice were bred with dm2 mice to obtain anti-Ld transgenic TCR⫹ (2C ⫻ dm2)F1 (H-2b/d, Ld⫺) mice. Anti-Ld transgeneic TCR is detectable by specific mAb 1B2 (36). All animals were kept in the animal facility at the University Health Network, and protocols were approved by the University Health Network Animal Care Committee, which meets the guidelines of the Canadian Council on Animal Care.

Generating and maintaining DN Treg and mutant clones Generation of DN Treg clones was performed using previously described methods (15). In brief, splenocytes were collected from (2C ⫻ dm2)F1 mice tolerant to CB6F1 skin allografts following DLI and stimulated with irradiated Ld⫹ CB6F1 splenocytes with 30 U/ml human rIL-2 and 30 U/ml rIL-4 for 10 days. Activated 1B2⫹ T cells were cultured in 96-well tissue culture plates at limiting dilutions of 0.5 cells/well with resulting colonies subcloned to ensure clonality. To maintain the T cell clones, 5 ⫻ 104 cells were cultured in 24-well plates containing 5 ⫻ 105 irradiated Ld⫹ CB6F1 splenocytes in ␣MEM supplemented with 10% FBS, 0.1% 2-ME, and 30 U/ml human rIL-2 and 30 U/ml rIL-4 (defined as complete medium). Cells were incubated at 37°C with 5% CO2 and restimulated this way every 3– 4 days.

5277 bated with CD3-FITC, CD4-PE, and CD8-PECy5. Data were acquired and analyzed on an EPICS XL-MCL flow cytometer.

Migration assays Regulatory or mutant clones (5 ⫻ 105) were placed in each upper Transwell chamber with 8-␮m pores (Costar). Lower wells contained complete medium with or without 2 ␮g/ml CXCL13 (R&D Systems). Plates were incubated at 37°C with 5% CO2 for 6 h, after which upper chambers were removed and migrated cells were stained with trypan blue, followed by visual hemocytometer quantification. Assays with primary lymphocytes used Transwells with 5-␮m pores due to smaller cells. Briefly, lymphocytes from lymph nodes of (2C ⫻ dm2)F1 or CB6F1 mice were harvested and processed through 40-␮m cell strainers. Aliquots were examined using FACS, as described above, to determine percentages of CD4, CD8, and DN Treg cells. In total, 106 lymphocytes were placed in each upper chamber. Total numbers of each cell type were calculated as percentage of cell type as determined by FACS ⫻ 106. Lower chambers contained complete medium with varying doses (0, 0.1, 1.25, 2.5, 5, or 10 ␮g/ml) of CXCL13. Transwell plates were incubated at 37°C with 5% CO2 for 6 h, and upper chambers were removed. Total migrated cells were quantified, as described above. Migrated cells were then analyzed using FACS to determine percentages of CD4, CD8, and DN Treg cells. Total percentage of migrated cells of a specific population was determined by: (total migrated cells/total input cells) ⫻ 100. In vitro CXCL13-neutralizing studies were performed by addition of varying concentrations (0, 5, 10, 25, or 50 ␮g/ml) of CXCL13-neutralizing Ab (R&D Systems) added to lower wells of the migration assay in which CB6F1 lymphocytes and 2 ␮g/ml CXCL13 were used.

Cardiac transplantation and organ harvesting Naive 8- to 10-wk-old (B6 ⫻ dm2)F1 mice were preinfused with 107/ mouse of either regulatory clone TN12 or mutant clone TN12.8, or left untreated 24 h pregrafting. In mice receiving Ab, 100 ␮g of CXCL13 mAb was injected i.p. 6 h postinfusion of regulatory clones. Allogeneic heart grafts from 6- to 8-wk-old CB6F1 mice were heterotopically transplanted into the abdomen of mice, as previously described (27). Mice were sacrificed 24 h posttransplantation; spleen, lymph nodes, cardiac allografts, and native hearts were harvested where applicable. Cardiac allografts were monitored daily by palpation, as previously described (16).

Immunohistological analysis of tissue Native hearts and cardiac allografts were harvested 24 h after transplantation, embedded in freezing medium, and immediately placed on dry ice. After sectioning, samples were coincubated with either mAb 1B2-FITC to specifically recognize injected clones or a rat Ab directed to murine CXCR5 (37), followed by incubation with PE-labeled goat anti-rat Ab. Samples were visualized using a Nikon TE200 fluorescent microscope at ⫻400 magnification (⫻40 objective and ⫻10 ocular). Images were captured using a Hamamatsu ORCA-285 digital camera (Hamamatsu) and SIMPLEPCI v 4.0.0 imaging software (Compix). Images were cropped using Adobe Photoshop 7.0.

Detection of graft-infiltrating lymphocytes Cardiac allografts harvested 24 h posttransplantation were mechanically processed and incubated in a solution of 0.2 U/ml collagenase (SigmaAldrich) in complete medium at 37°C for 30 min. Following digestion, tissue was forced through a 40-␮m cell strainer to isolate graft-infiltrating cells. Following RBC lysis, cells were washed three times with PBS containing 0.2% BSA. Cells were coincubated with 1B2-FITC and analyzed by FACS.

Quantitative real-time PCR Cell surface marker staining For detection of the chemokine receptor CXCR5, regulatory and mutant clones were collected at day 3 poststimulation and incubated with rat antimouse CXCR5 mAb (37), followed by incubation with PE-conjugated secondary Ab specific for rat IgG (Cedarlane Laboratories). Primary lymphocytes were stained in the same manner, followed by coincubation with CD3-FITC, CD4-PECy5, and CD8-PECy5. Graft-infiltrating cells were incubated with 1B2-FITC. For migration assays, lymphocytes were coincu-

Quantitative real-time PCR analysis was performed, as previously described (26). In brief, RNA was isolated from native hearts, transplanted hearts, and untransplanted donor hearts. cDNA was generated using random hexamer. Detection of CXCL13 was conducted using mouse specific primers (sense, 5⬘-TGGCTGCCCCAAAACTGA-3⬘; antisense, 5⬘-TGGCACGAGGATTCACACAT-3⬘). The expression of ␤-actin was detected, as previously described, and used to normalize starting cDNA concentrations (26).

5278

Results DN Treg clones, but not mutants, express functional surface CXCR5 We have demonstrated recently that pretransplant infusion of Ldspecific DN Treg clones, such as TN12, but not their natural mutants, such as TN12.8, into Ld-negative syngeneic mice can prevent rejection of Ld⫹ skin and cardiac allografts (19, 26). Furthermore, cDNA microarray analysis revealed a significantly higher expression of CXCR5 mRNA in DN Treg clones compared with their mutant progeny (26). To determine whether DN Treg clones express functional CXCR5 protein, DN Treg clone TN12 and its natural mutant TN12.8 were stained with anti-CXCR5 mAb and analyzed by flow cytometry. A significantly higher level of CXCR5 protein expression was observed on the regulatory TN12 clone cells when compared with nonregulatory TN12.8 clone cells (Fig. 1A). To study whether expression of CXCR5 on TN12 clone cells enables them to migrate toward the specific ligand CXCL13, TN12 Treg or mutant TN12.8 clones were seeded in upper chambers of Costar Transwell migration plates. CXCL13 was added to the lower chambers, creating a concentration gradient. After 6 h, cells that migrated to the bottom chambers were quantified. The regulatory TN12 clone cells had lower basal migration (2.6 ⫾ 0.9%) compared with their CXCR5⫺ mutants (8.3 ⫾ 0.8%). However, in the presence of CXCL13, TN12 DN Treg clone migration was significantly increased (43.8 ⫾ 3.0%), whereas no significant

FIGURE 1. Expression of functional CXCR5 on DN Treg clones. A, CXCR5 expression in clones. The DN Treg clone TN12 (dotted line) and mutant clone TN12.8 (dashed line) were stained with purified rat mAb directed to murine CXCR5, followed by PE-conjugated goat anti-rat IgG secondary Ab. Regulatory TN12 clones were coincubated with isotype control rat IgG, followed by PE-conjugated goat anti-rat IgG mAb as the negative control (solid line). Data are representative of at least five independent experiments. B, DN Treg cell migration in vitro. A total of 5 ⫻ 105 CXCR5⫹ regulatory TN12 and CXCR5⫺ TN12.8 mutant clones was added to the top wells, and 2 ␮g/ml CXCL13 was placed in the lower wells of Costar Transwell plates. The number of cells that migrated to the lower wells was determined after 6 h by staining with trypan blue, followed by quantification using a hemocytometer. Percentage of migrated cells was calculated using the equation: (total migrated cells/total input cells) ⫻ 100. Data are representative of four independent experiments.

ROLE OF CXCR5 IN DN Treg CELL MIGRATION change in migration of mutant TN12.8 clone cells (9.2 ⫾ 0.8%) was observed (Fig. 1B). These data indicate that DN Treg clones, which can prevent graft rejection, express a functional chemokine receptor CXCR5 and can migrate toward CXCR13, whereas their natural mutants that have lost suppressive function do not express CXCR5 or migrate toward CXCL13. Primary DN, but not CD4 or CD8 T cells express CXCR5 and preferentially migrate toward increasing CXCL13 gradients It was reported previously that DN T cells from MRL-lymphoproliferative (lpr) mice express cell surface CXCR5 (30). Our recent studies demonstrated that DN T cells obtained from C57BL/6, (2C ⫻ dm2)F1, and B6-lpr mice have potent regulatory function both in vitro and in vivo (15, 17, 25, 27). To determine whether CXCR5 is also preferentially expressed on primary DN Treg cells compared with CD4 or CD8 cells, lymphocytes were harvested from naive CB6F1 mice and incubated with fluorescently labeled mAbs directed to CD3 (FITC), CD4 (PE-Cy5), CD8 (PE-Cy5), NK1.1 (PE-Cy5), and CXCR5 (PE). As controls, the same population of lymphocytes was coincubated with mAb to CXCR5 (PE) along with either CD4 (PE-Cy5) or CD8 (PE-Cy5). CD3⫹CD4⫺CD8⫺NK1.1⫺ cells (DN Treg cells), CD3⫹CD4⫹, or CD3⫹CD8⫹ cells were gated, and the expression of CXCR5 was analyzed using flow cytometry. As shown in Fig. 2A, primary DN Treg cells express high levels of CXCR5 on their surface. Little or no expression of CXCR5 on primary naive CD4 or CD8 T cells was detected. This constitutive expression on DN Tregs was similarly observed in (2C ⫻ dm2)F1, C57BL/6, BALB/c, and B6-lpr mice (data not shown). Next, we investigated whether primary DN Tregs preferentially use CXCR5 for migration compared with other subsets of T cells (Fig. 2B). Lymph node cells from naive CB6F1 mice were seeded in the upper chambers of Transwell plates. An increasing gradient of CXCL13 (0 –10 ␮g/ml) was added to the lower chambers. After 6 h, the total number of migrated cells in the lower chambers was quantified. To determine the percentages of each subpopulation of T cells that migrated toward CXCL13, aliquots of lymphocytes input into upper chambers of the migration assay and migrated cells from the lower chambers following 6-h incubation were stained using CD3-FITC, CD4-PE, and CD8-PE-Cy5 mAbs, and analyzed by flow cytometry. Total percentages of migration for CD4⫹, CD8⫹, and DN Treg cells were calculated. DN Treg cells showed a dose-dependent migration toward increasing concentrations of CXCL13. At 2.5 ␮g/ml CXCL13, 57% of DN Treg cells migrated to the lower chambers. In contrast, neither CD4 nor CD8 showed noticeable migration toward any concentration of CXCL13 (Fig. 2B). To ensure that the preferential expression of functional CXCR5 on DN Treg cells is not strain specific, these studies were also performed by using primary anti-Ld TCR transgenic (2C ⫻ dm2)F1 mice and C57BL/6 mice; the same results were obtained (data not shown). Our data demonstrate that, unlike CD4 or CD8 T cells, both DN Treg clones and primary DN Treg cells express high levels of CXCR5, and can migrate toward CXCL13. Blocking DN Treg cell migration with CXCL13-neutralizing mAb in vitro To further confirm that CXCR5 and CXCL13 interaction is important for DN Treg cell migration, blocking studies were performed. Lymphocytes from CB6F1 mice were seeded at 5 ⫻ 105/ well in upper chambers and 2 ␮g/ml CXCL13 in lower chambers of Transwell migration plates. Increasing concentrations of CXCL13-neutralizing mAb (0, 5, 10, 25, 50 ␮g/ml) were added to

The Journal of Immunology

FIGURE 2. Expression of functional CXCR5 on primary DN Treg cells, but not CD4 or CD8 T cells. A, Expression of CXCR5 on naive primary T cells. Lymph node cells from CB6F1 mice were coincubated with mAbs to mouse CD3 (FITC), CD4 and CD8 (PECy5), and CXCR5 (purified). Following CXCR5 staining, lymphocytes were coincubated with PE-conjugated goat anti-rat IgG secondary Ab. Cells were gated on CD3⫹CD4⫺CD8⫺, CD3⫹CD4⫹CD8⫺, or CD3⫹CD8⫹CD4⫺ populations, and the expression of CXCR5 on the DN Treg cells (dotted line) CD4 (dashed line) and CD8 (alternating dots and dashes) is shown. Data are representative of at least seven independent experiments. Lymphocytes were coincubated with isotype control rat IgG, followed by PE-conjugated goat anti-rat IgG secondary Ab as the negative control (solid line). B, Naive lymph node T cell migration in vitro. Primary lymphocytes from the lymph nodes of CB6F1 mice were collected, and stained using CD3-FITC, CD4PE, and CD8-PECy5 mAbs to determine the percentages and total numbers of each subpopulation of the cells before migration assays. In total, 5 ⫻ 105 lymphocytes were then added to the upper chambers. Varying concentrations (0 –10 ␮g/ml) of CXCL13 were added to the lower chambers. After 6 h of culture, migrated cells were similarly stained and analyzed using flow cytometry. Total percentage of migrated cells in each subpopulation was calculated based on the equation: (total migrated cells of a given population)/(total input cells of a given population) ⫻ 100. Data are representative of 12 independent experiments.

the lower chambers, and cell migration was examined, as described above. As observed in Fig. 3, neither CD4 nor CD8 T cell populations showed significant increases in migration after 6 h. However, DN Treg cells again showed increased migration to the bottom wells. This migration was reduced with increasing concentrations of CXCL13-neutralizing mAb, with complete abrogation at 25 ␮g/ml. These data clearly demonstrate that interaction of CXCR5 with its ligand CXCL13 is critical for DN Treg cell migration in vitro. CXCR5⫹ DN Treg clones home to cardiac allografts We have shown previously that DN Treg cells preferentially accumulate in long-term accepted allogeneic skin grafts (19). The molecular mechanism by which DN Treg cells home into tolerant allografts is not clear. Because DN Treg cells express high levels of CXCR5, we thus studied whether CXCR5-CXCL13 interaction plays an important role in DN Treg cell homing into allografts. Naive (B6 ⫻ dm2)F1 mice were infused with either CXCR5⫹ DN Treg clone TN12 or CXCR5⫺ mutant clone TN12.8, followed by transplantation of an Ld⫹ cardiac allograft the following day.

5279

FIGURE 3. Abrogation of migration using neutralizing CXCL13 Ab. In total, 5 ⫻ 105 lymph node cells from CB6F1 mice were placed in the upper chambers of Transwell migration plate. A total of 2 ␮g/ml CXCL13 and increasing concentrations (0 –50 ␮g/ml) of CXCL13-neutralizing mAb were added to the lower chambers. Migrated cells were stained for CD3, CD4, and CD8 and analyzed by FACS, and total percentages were determined, as described in Fig. 2B. DN Treg cells (u) show significantly increased migration relative to both CD4⫹ (䡺) and CD8⫹ (f) T cells. Migration is reduced at 10 ␮g/ml and completely abrogated by 25 ␮g/ml neutralizing Ab. Data are representative of five independent experiments.

These clones are syngeneic to recipients and express an anti-Ld transgenic TCR detectable by mAb 1B2. One day posttransplantation, heart allografts along with recipient native hearts were harvested, and the presence of the infused T cell clones was examined by immunohistology using 1B2-FITC mAb. Native hearts of the transplanted animals show no infiltration of infused cells (Fig. 4A). Although a significant number of infused DN Treg clones accumulated in the cardiac allografts (Fig. 4B), no infiltration was observed in allografts from mice preinfused with CXCR5⫺ mutant clones (Fig. 4C). To determine whether DN Treg cells that migrate to allografts still express CXCR5, allografts were harvested from (B6 ⫻ dm2)F1 mice that were infused with the syngeneic DN Treg clone TN12 1 day posttransplantation. Expression of CXCR5 on the infused DN Treg clones was assessed by dual staining of heart tissue with 1B2-FITC (Fig. 4D) and anti-mouse CXCR5 mAb (PE) (Fig. 4E). Analysis using immunofluorescent microscopy revealed that most 1B2-expressing graft-infiltrating DN Treg clones also expressed CXCR5 (Fig. 4F). These data indicate that the CXCR5⫹ DN Treg clones, but not CXCR5⫺ mutants, can home into cardiac allografts, and suggest that the expression of CXCR5 is important for DN Treg cell migration in vivo. DN Treg clones preferentially accumulate in cardiac allografts and can prevent rejection following adoptive transfer To further investigate whether DN Treg cells can preferentially migrate to and persist in allografts and prevent rejection, naive (B6 ⫻ dm2)F1 mice were infused with either CXCR5⫹ DN Treg clone TN12 or CXCR5⫺ mutant clone TN12.8 1 day before cardiac transplantation. Groups of mice were sacrificed on days 1, 5, and 20 posttransplantation. The cardiac allograft-infiltrating cells and lymphocyte suspensions of the spleen and lymph nodes of recipient mice were stained with 1B2-FITC, followed by FACS analysis. The total numbers of infused DN Treg and mutant clone cells in each organ were determined. Interestingly, while a similar number of DN Treg and mutant clones was detected in the spleens and lymph nodes, the infused DN Treg clones selectively accumulated in cardiac allografts, which composed ⬎20% of total graft-infiltrating cells at all three time points (Fig. 5A). In addition, we examined the ability of infused DN Treg clones to prolong

5280

ROLE OF CXCR5 IN DN Treg CELL MIGRATION

FIGURE 4. CXCR5-expressing DN Treg clones preferentially home into transplanted cardiac allografts. (B6 ⫻ dm2)F1 mice were i.v. infused with 1 ⫻ 107 cells/mouse of either DN Treg TN12 clones (A, B, and D–F) or mutant clone TN12.8 (C). One day later, each mouse was transplanted with a CB6F1 cardiac allograft. Both native (A) and transplanted (B–F) hearts were harvested at 24 h posttransplantation. To monitor the infused cells and CXCR5 expression, heart tissue samples were stained with either 1B2-FITC alone (shown in green fluorescence) (A–C) or 1B2-FITC and CXCR5-PE (shown in red) (D–F). No infused cells were detected in native hearts or in the allografts from the recipients that were infused with mutant clones (C) at 24 h after transplantation. A significant number of infused cells (shown in green fluorescence; B and D) and CXCR5⫹ cells (shown in red; E) was found in cardiac allografts from mice infused with DN Treg clones (B and D). Further examination of cardiac allografts showed that 1B2 (D) and CXCR5 (E) expression were colocalized (F), indicating that graft-infiltrating DN Treg clones also expressed CXCR5. Data shown here are representative of experiments using four to five mice.

survival of cardiac allografts. In all mice infused with CXCR5⫹ TN12 Treg clones, graft survival showed significant prolongation (⬎100 days), whereas mice receiving CXCR5⫺ mutant TN12.8 clones showed no increased graft survival compared with untreated controls

FIGURE 5. DN Treg clones accumulate in cardiac allografts and can prolong graft survival. The (2C ⫻ dm2)F1 mice were i.v. infused with 1 ⫻ 107/ mouse either DN Treg TN12 or mutant TN12.8 clones. All mice were transplanted with CB6F1 cardiac allografts. A, Spleen, lymph nodes, and transplanted hearts were harvested at days 1, 5, and 20. Graft-infiltrating cells (GIC) and lymphocytes in the spleen and lymph nodes were isolated and stained with 1B2-FITC and analyzed using FACS. Percentages of infused cells (detected by 1B2 mAb) in the lymphocytes of each organ are shown. B, Survival of cardiac allografts in mice receiving infusion of regulatory TN12 clones (n ⫽ 4, dashed line) or TN12.8 mutant clones (n ⫽ 4, dotted line), or left untreated (n ⫽ 5, solid line) was monitored by palpation. A, Represents data from four different experiments.

(Fig. 5B). These data indicate that while CXCR5⫺ mutant clones are unable to home to cardiac allografts, CXCR5-expressing DN Treg clones can preferentially migrate, and persist in cardiac allografts,

FIGURE 6. Detection and neutralization of CXCL13 expression in transplanted cardiac allografts. A, (B6 ⫻ dm2)F1 mice were injected with 107 DN Treg clone cells 1 day before transplantation. Native hearts from transplant recipients and cardiac allografts were harvested from mice 6 h posttransplantation along with untransplanted donor hearts. Tissues were processed for RNA isolation and examined using real-time quantitative PCR. Primers specific for murine CXCL13 were used to detect gene expression. Results are expressed as values relative to ␤-actin expression. Data are representative of at least five independent experiments. B, (B6 ⫻ dm2)F1 mice were infused with 107 DN Treg clone cells. These mice were injected i.p. with either 100 ␮g of CXCL13-neutralizing mAb (n ⫽ 4) or medium or PBS (n ⫽ 5) at 6 h after infusion of cells. All mice were transplanted with CB6F1 allogeneic heart grafts 18 h later. Allografts were harvested 1 day postgrafting, and graft-infiltrating cells were stained with specific mAb 1B2 and analyzed by FACS.

The Journal of Immunology which may generate a localized immune regulation and contribute to preventing graft rejection. Expression of CXCL13 in cardiac allografts is important for DN Treg cell homing Because CXCL13 is the only known ligand for CXCR5, it should be expressed by cardiac allografts for DN Treg cell homing to occur through CXCR5. To test this hypothesis, CB6F1 cardiac allografts and native recipient hearts were harvested from (B6 ⫻ dm2)F1 recipients 6 h posttransplantation. As a baseline control, hearts were also harvested from donors without transplantation. Tissues were then processed for RNA, and CXCL13 expression was determined using real-time PCR. All results were normalized to ␤-actin expression. It was found that the cardiac allografts from DN Treg cell-treated mice showed a 45-fold increase in CXCL13 mRNA expression compared with that of the native hearts, and a 14-fold greater expression compared with untransplanted hearts from the same donor origin (Fig. 6A). To further confirm that the interaction between CXCR5 and CXCL13 is important for DN Treg cell homing to allografts, an in vivo blocking study was performed. Recipient (B6 ⫻ dm2)F1 mice were preinfused with 107 DN Treg clone TN12 cells 24 h pretransplantation. These mice were further subjected to i.p. injections with 100 ␮g of CXCL13-neutralizing mAb 6 h following clone injection (18 h pretransplantation). Mice were sacrificed 24 h posttransplantation, and the spleen, lymph nodes, and cardiac allografts were harvested and examined for the presence of infused DN Treg clone cells using flow cytometry. Homing of DN Treg cells into the spleen and lymph nodes appeared to be unaffected by treatment with neutralizing mAb. However, CXCL13 neutralization resulted in a 9-fold reduction of graft infiltration by DN Treg clones when compared with the controls (Fig. 6B). Taken together, these data demonstrate that CXCR5-CXCL13 interaction plays an important role in DN Treg cell homing into cardiac allografts, suggesting that modulation of this interaction may be one of the important approaches to induce transplantation tolerance by increasing localized accumulation of DN Treg cells.

Discussion It is well known that different chemokine receptors are expressed and used preferentially by different lymphocyte subpopulations for migration (32). For instance, CXCR4 and CCR7 are preferentially expressed on naive T cells and allow for homing to local lymph nodes (38, 39). The receptors CXCR3 and CCR5 are preferentially expressed on Th1 cells, which are important in homing of Th1 cells to regions of inflammatory response (40). Expression of CCR3 and CCR4 is important for Th2 cell homing (41, 42). CXCR5 has been shown to be a major chemokine receptor used by peripheral B lymphocytes to migrate to the B cell zones in lymph nodes (37). Studies using other mouse models have observed low levels of CXCR5 expression in the naive CD4 population (35), and on a subpopulation of memory CD4⫹ T cells (33, 41– 43). Treg cells, as a functionally distinct subset of lymphocytes, play a critical role in the development of various diseases. Accumulating data indicate that Treg cells can home to the draining lymph nodes and diseased organs to regulate immune responses locally (44 – 46). However, little is known about the in vivo homing of Treg cells and the chemokine receptors that are preferentially expressed and used by these cells. Recently, we and others have demonstrated, in mice and humans, that DN Tregs are able to suppress immune responses in an Ag-specific and cell contact-dependent manner in vitro (15, 17, 18, 25). Furthermore, injection of DN Treg cells or clones can induce long-term skin and cardiac allo- and xenograft survival (15–17, 19,

5281 26, 27, 47). Interestingly, at 3 days posttransplantation, a low number of Ld-specific DN Treg cells could be detected in both Ld⫹ donor-specific as well as in third-party skin grafts at similar levels. However, at 7 days after transplantation, the number of Ld-specific DN Treg cells decreased in third-party rejecting grafts, but accumulated in large numbers in Ld⫹ tolerant grafts (19). These data suggest that the migration of DN Treg cells at early time points may be due to inflammation resulting from the transplantation procedure, while the accumulation and persistence of DN Treg cells at later stages may be Ag specific. The mechanism leading to the migration and accumulation of DN Treg cells into tolerant allografts is not clear. In this study, we focused on the identification of chemokine receptors that are preferentially expressed by DN Treg cells. Our data indicate that DN Treg clones express a high level of surface CXCR5 protein, whereas their natural mutants that have lost the capacity to prolong graft survival express a very low level of CXCR5 mRNA (26) and protein (Fig. 1A). Furthermore, we demonstrate that primary DN Treg cells from CB6F1 mice constitutively express high levels of CXCR5 protein on their surface. In contrast, neither naive CD4⫹ nor CD8⫹ T cells showed detectable levels of CXCR5 expression (Fig. 2A). Ansel et al. (30) have shown that DN T cells from lpr mice also express CXCR5. We have demonstrated previously that DN Treg cells obtained from B6-lpr mice also possess potent regulatory function both in vitro and in vivo (17). Collectively, these data clearly demonstrate that unlike CD8 or most CD4 T cells, DN Treg cells preferentially express CXCR5. It has been shown recently that CD4⫹CD25⫹ Treg cells in human tonsils up-regulate CXCR5 (29). These CD4⫹CD25⫹ Treg cells can home to B cell zones and prevent Ab production by B lymphocytes through inhibition of B helper CD4⫹ T cells (29). Whether CXCR5 is also expressed in other subsets of Treg cells and serves as a common pathway for Treg cell homing to sites of immune response requires further study. The only known CXCR5 ligand observed to date is CXCL13, which in turn is exclusively detected by CXCR5 (48). In examining migration of lymphocytes toward an increasing gradient of CXCL13, we found that both DN Treg clones (Fig. 1B) and primary DN Treg cells (Fig. 2B) showed strong chemotactic responses in vitro, which can be blocked by CXCL13-neutralizing Ab (Fig. 3). In contrast, neither nonregulatory mutant clones nor primary CD4⫹ or CD8⫹ T cells could migrate toward CXCL13 (Figs. 1b and 2b). Furthermore, CXCR5⫹ DN Treg clones can preferentially migrate to and persist in cardiac allografts (Figs. 4 and 5), and administration of CXCL13-neutralizing Ab to the recipients significantly reduced homing of DN Treg clones to cardiac allografts (Fig. 6). Taken together, these results demonstrate that expression of CXCR5 on DN Treg cells and the interaction of this receptor with its ligand CXCL13 are important for DN Treg cell migration both in vitro and in vivo. Interactions between CXCR5 and CXCL13 are known to be involved in both physiological and pathophysiological conditions. Under normal circumstances, CXCL13 is produced predominantly by follicular dendritic cells in lymph nodes (49). After encountering Ags, a small subset of CD4⫹ T cells up-regulates CXCR5 and homes into lymph nodes. These CD4⫹ T cells are juxtaposed to CXCR5-expressing B cells, and play a role in humoral immune responses of B cells within the lymphoid follicle (35, 50). In various disease models, ectopic expression of CXCL13 has been reported in vascular endothelium of the CNS (51, 52) as well as in germinal centers of ectopic lymphoid follicles within the synovium of chronic arthritis patients (53–57). Furthermore, it has been observed recently that CXCL13 expression can be colocalized to B

5282 cell cluster formations in rejecting human renal grafts (58). In general, these studies suggest that CXCL13 expression may be associated with development of diseases. Whether CXCL13 is also involved in recruiting Treg cells to the sites of immune responses is not clear. An earlier report showed that mice challenged with Ag intratracheally had increased numbers of DN Treg cell accumulation in the lung (59). Furthermore, DN T cell clones generated from the lung-infiltrating DN T cells were able to suppress Agstimulated proliferative responses of syngeneic Th1 clones (59). This study indicates that DN Treg cells can infiltrate to the lung. However, the mechanism involved in DN Treg cell homing is not known. We have demonstrated previously that DN Treg cells accumulate in tolerant allografts (19), and that infusion of DN Treg clones leads to long-term cardiac allograft survival (26, 27). By using real-time PCR, we found a 45-fold increase in CXCL13 mRNA expression within the transplanted allografts compared with native hearts (Fig. 6A). Although the source of CXCL13 in the allografts is not known, the fact that CXCL13-neutralizing mAb can inhibit DN Treg cell homing into cardiac allografts indicates that expression of CXCL13 is important for DN Treg cells to selectively migrate into heart allografts after transplantation (Figs. 4 and 5). These results suggest, at least with DN Treg cells, that CXCR5-CXCL13 interaction is central to homing into transplanted grafts. Interestingly, injection of CXCL13 neutralization mAb at 6 h after DN Treg cell infusion did not inhibit DN Treg clone homing into secondary lymphoid organs that express high CXCL13 concentration. Moreover, similar numbers of DN Treg and mutant clones were found in the spleen and lymph nodes. These findings suggest that DN Treg clones may use an alternate chemokine receptor for homing into the lymphoid organs (37, 41– 43). Although CXCR5 allows for B cell and Th cell homing into B cell zones in lymphoid follicles, mature dendritic cells, naive, and central memory T cells use CCR7 for homing into T cell zones of the lymphatic organs (32, 60). We have observed that both DN Treg and mutant clones express CCR7 mRNA (data not shown). Similarly, ⬎50% of DN Treg cells in human peripheral blood express CCR7 (18). It is therefore possible that DN Treg clones may use CCR7 for homing into secondary lymphoid organs. The role of CCR7 and possibly other chemokine receptors and their ligands in DN Treg cell migration is currently under investigation. In summary, we demonstrate in this work that DN Treg cells express high levels of CXCR5 and can migrate toward CXCL13. Furthermore, cardiac allografts up-regulate CXCL13 expression following transplantation, and infused CXCR5⫹ DN Treg clones preferentially migrate into and persist in cardiac allografts that can prevent graft rejection. Blocking CXCL13 by using neutralizing Ab prevents DN Treg cell migration in vitro and homing into cardiac allografts. Taken together, these data demonstrate that interaction between CXCR5 and CXCL13 is important for graft infiltration by DN Treg cells. This preferential homing of DN Treg cells toward graft-derived CXCL13 could be exploited to increase graft-infiltrating DN Tregs as a possible means of maintaining localized immune suppression.

Disclosures The authors have no financial conflict of interest.

References 1. Sakaguchi, S. 2004. Naturally arising CD4⫹ regulatory T cells for immunologic self-tolerance and negative control of immune responses. Annu. Rev. Immunol. 22: 531–562. 2. O’Garra, A., P. L. Vieira, P. Vieira, and A. E. Goldfeld. 2004. IL-10-producing and naturally occurring CD4⫹ Tregs: limiting collateral damage. J. Clin. Invest. 114: 1372–1378.

ROLE OF CXCR5 IN DN Treg CELL MIGRATION 3. Robinson, D. S., M. Larche, and S. R. Durham. 2004. Tregs and allergic disease. J. Clin. Invest. 114: 1389 –1397. 4. Mills, K. H. 2004. Regulatory T cells: friend or foe in immunity to infection? Nat. Rev. Immunol. 4: 841– 855. 5. Sutmuller, R. P., L. M. van Duivenvoorde, A. van Elsas, T. N. Schumacher, M. E. Wildenberg, J. P. Allison, R. E. Toes, R. Offringa, and C. J. Melief. 2001. Synergism of cytotoxic T lymphocyte-associated antigen 4 blockade and depletion of CD25⫹ regulatory T cells in antitumor therapy reveals alternative pathways for suppression of autoreactive cytotoxic T lymphocyte responses. J. Exp. Med. 194: 823– 832. 6. Lechler, R. I., O. A. Garden, and L. A. Turka. 2003. The complementary roles of deletion and regulation in transplantation tolerance. Nat. Rev. Immunol. 3: 147–158. 7. Sakaguchi, S., N. Sakaguchi, M. Asano, M. Itoh, and M. Toda. 1995. Immunologic self-tolerance maintained by activated T cells expressing IL-2 receptor ␣-chains (CD25): breakdown of a single mechanism of self-tolerance causes various autoimmune diseases. J. Immunol. 155: 1151–1164. 8. Groux, H., A. O’Garra, M. Bigler, M. Rouleau, S. Antonenko, J. E. de Vries, and M. G. Roncarolo. 1997. A CD4⫹ T-cell subset inhibits antigen-specific T-cell responses and prevents colitis. Nature 389: 737–742. 9. Huehn, J., K. Siegmund, J. C. U. Lehmann, C. Siewert, U. Haubold, M. Feuerer, G. F. Debes, J. Lauber, O. Frey, G. K. Przybylski, et al. 2004. Developmental stage, phenotype, and migration distinguish naive- and effector/memory-like CD4⫹ regulatory T cells. J. Exp. Med. 199: 303–313. 10. Graca, L., S. Thompson, C. Y. Lin, E. Adams, S. P. Cobbold, and H. Waldmann. 2002. Both CD4⫹CD25⫹ and CD4⫹CD25⫺ regulatory cells mediate dominant transplantation tolerance. J. Immunol. 168: 5558 –5565. 11. Manavalan, J. S., S. Kim-Schulze, L. Scotto, A. J. Naiyer, G. Vlad, P. C. Colombo, C. Marboe, D. Mancini, R. Cortesini, and N. Suciu-Foca. 2004. Alloantigen specific CD8⫹CD28⫺ FOXP3⫹ T suppressor cells induce ILT3⫹ ILT4⫹ tolerogenic endothelial cells, inhibiting alloreactivity. Int. Immunol. 16: 1055–1068. 12. Skelsey, M. E., J. Mellon, and J. Y. Niederkorn. 2001. ␥␦ T cells are needed for ocular immune privilege and corneal graft survival. J. Immunol. 166: 4327– 4333. 13. Godfrey, D. I., and M. Kronenberg. 2004. Going both ways: immune regulation via CD1d-dependent NKT cells. J. Clin. Invest. 114: 1379 –1388. 14. Young, K. J., B. DuTemple, M. J. Phillips, and L. Zhang. 2003. Inhibition of graft-versus-host disease by double-negative regulatory T cells. J. Immunol. 171: 134 –141. 15. Zhang, Z. X., L. Yang, K. J. Young, B. DuTemple, and L. Zhang. 2000. Identification of a previously unknown antigen-specific regulatory T cell and its mechanism of suppression. Nat. Med. 6: 782–789. 16. Chen, W., M. S. Ford, K. J. Young, M. I. Cybulsky, and L. Zhang. 2003. Role of double-negative regulatory T cells in long-term cardiac xenograft survival. J. Immunol. 170: 1846 –1853. 17. Ford, M. S., K. J. Young, Z. Zhang, P. S. Ohashi, and L. Zhang. 2002. The immune regulatory function of lymphoproliferative double negative T cells in vitro and in vivo. J. Exp. Med. 196: 261–267. 18. Fischer, K., S. Voelkl, J. Heymann, G. K. Przybylski, K. Mondal, M. Laumer, L. Kunz-Schughart, C. A. Schmidt, R. Andreesen, and A. Mackensen. 2005. Isolation and characterization of human antigen-specific TCR␣␤⫹ CD4⫺CD8⫺ double-negative regulatory T cells. Blood 105: 2828 –2835. 19. Young, K. J., L. Yang, M. J. Phillips, and L. Zhang. 2002. Donor-lymphocyte infusion induces transplantation tolerance by activating systemic and graft-infiltrating double-negative regulatory T cells. Blood 100: 3408 –3414. 20. Sawitzki, B., M. Lehmann, T. Ritter, E. Graser, J. W. Kupiec-Weglinski, and H. D. Volk. 2001. Regulatory tolerance-mediating T cells in transplantation tolerance. Transplant. Proc. 33: 2092–2093. 21. Graca, L., S. P. Cobbold, and H. Waldmann. 2002. Identification of regulatory T cells in tolerated allografts. J. Exp. Med. 195: 1641–1646. 22. Yu, P., Y. Lee, W. Liu, T. Krausz, A. Chong, H. Schreiber, and Y. X. Fu. 2005. Intratumor depletion of CD4⫹ cells unmasks tumor immunogenicity leading to the rejection of late-stage tumors. J. Exp. Med. 201: 779 –791. 23. Antony, P. A., C. A. Piccirillo, A. Akpinarli, S. E. Finkelstein, P. J. Speiss, D. R. Surman, D. C. Palmer, C. C. Chan, C. A. Klebanoff, W. W. Overwijk, et al. 2005. CD8⫹ T cell immunity against a tumor/self-antigen is augmented by CD4⫹ T helper cells and hindered by naturally occurring T regulatory cells. J. Immunol. 174: 2591–2601. 24. Unitt, E., S. M. Rushbrook, A. Marshall, S. Davies, P. Gibbs, L. S. Morris, N. Coleman, and G. J. Alexander. 2005. Compromised lymphocytes infiltrate hepatocellular carcinoma: the role of T-regulatory cells. Hepatology 41: 722–730. 25. Zhang, Z. X., W. L. Stanford, and L. Zhang. 2002. Ly-6A is critical for the function of double negative regulatory T cells. Eur. J. Immunol. 32: 1584 –1592. 26. Lee, B. P., E. Mansfield, S. C. Hsieh, T. Hernandez-Boussard, W. Chen, C. W. Thomson, M. S. Ford, S. E. Bosinger, S. Der, Z. X. Zhang, et al. 2005. Expression profiling of murine double-negative regulatory T cells suggests mechanisms for prolonged cardiac allograft survival. J. Immunol. 174: 4535– 4544. 27. Chen, W., M. S. Ford, K. J. Young, and L. Zhang. 2003. Infusion of in vitrogenerated DN T regulatory cells induces permanent cardiac allograft survival in mice. Transplant. Proc. 35: 2479 –2480. 28. Dobner, T., I. Wolf, T. Emrich, and M. Lipp. 1992. Differentiation-specific expression of a novel G protein-coupled receptor from Burkitt’s lymphoma. Eur. J. Immunol. 22: 2795–2799. 29. Lim, H. W., P. Hillsamer, and C. H. Kim. 2004. Regulatory T cells can migrate to follicles upon T cell activation and suppress GC-Th cells and GC-Th celldriven B cell responses. J. Clin. Invest. 114: 1640 –1649.

The Journal of Immunology 30. Ansel, K. M., L. J. McHeyzer-Williams, V. N. Ngo, M. G. McHeyzer-Williams, and J. G. Cyster. 1999. In vivo-activated CD4 T cells up-regulate CXC chemokine receptor 5 and reprogram their response to lymphoid chemokines. J. Exp. Med. 190: 1123–1134. 31. Forster, R., T. Emrich, E. Kremmer, and M. Lipp. 1994. Expression of the Gprotein-coupled receptor BLR1 defines mature, recirculating B cells and a subset of T-helper memory cells. Blood 84: 830 – 840. 32. Ebert, L. M., P. Schaerli, and B. Moser. 2005. Chemokine-mediated control of T cell traffic in lymphoid and peripheral tissues. Mol. Immunol. 42: 799 – 809. 33. Breitfeld, D., L. Ohl, E. Kremmer, J. Ellwart, F. Sallusto, M. Lipp, and R. Forster. 2000. Follicular B helper T cells express CXC chemokine receptor 5, localize to B cell follicles, and support immunoglobulin production. J. Exp. Med. 192: 1545–1552. 34. Chtanova, T., S. G. Tangye, R. Newton, N. Frank, M. R. Hodge, M. S. Rolph, and C. R. Mackay. 2004. T follicular helper cells express a distinctive transcriptional profile, reflecting their role as non-Th1/Th2 effector cells that provide help for B cells. J. Immunol. 173: 68 –78. 35. Hardtke, S., L. Ohl, and R. Forster. 2005. Balanced expression of CXCR5 and CCR7 on follicular T helper cells determines their transient positioning to lymph node follicles and is essential for efficient B-cell help. Blood 106: 1924 –1931. 36. Sha, W. C., C. A. Nelson, R. D. Newberry, D. M. Kranz, J. H. Russell, and D. Y. Loh. 1988. Positive and negative selection of an antigen receptor on T cells in transgenic mice. Nature 336: 73–76. 37. Forster, R., A. E. Mattis, E. Kremmer, E. Wolf, G. Brem, and M. Lipp. 1996. A putative chemokine receptor, BLR1, directs B cell migration to defined lymphoid organs and specific anatomic compartments of the spleen. Cell 87: 1037–1047. 38. Sallusto, F., P. Schaerli, P. Loetscher, C. Schaniel, D. Lenig, C. R. Mackay, S. Qin, and A. Lanzavecchia. 1998. Rapid and coordinated switch in chemokine receptor expression during dendritic cell maturation. Eur. J. Immunol. 28: 2760 –2769. 39. Potsch, C., D. Vohringer, and H. Pircher. 1999. Distinct migration patterns of naive and effector CD8 T cells in the spleen: correlation with CCR7 receptor expression and chemokine reactivity. Eur. J. Immunol. 29: 3562–3570. 40. Burns, W. R., Y. Wang, P. C. Tang, H. Ranjbaran, A. Iakimov, J. Kim, M. Cuffy, Y. Bai, J. S. Pober, and G. Tellides. 2005. Recruitment of CXCR3⫹ and CCR5⫹ T cells and production of interferon-␥-inducible chemokines in rejecting human arteries. Am. J. Transplant. 5: 1226 –1236. 41. Yamamoto, J., Y. Adachi, Y. Onoue, Y. S. Adachi, Y. Okabe, T. Itazawa, M. Toyoda, T. Seki, M. Morohashi, K. Matsushima, and T. Miyawaki. 2000. Differential expression of the chemokine receptors by the Th1- and Th2-type effector populations within circulating CD4⫹ T cells. J. Leukocyte Biol. 68: 568 –574. 42. Bonecchi, R., G. Bianchi, P. P. Bordignon, D. D’Ambrosio, R. Lang, A. Borsatti, S. Sozzani, P. Allavena, P. A. Gray, A. Mantovani, and F. Sinigaglia. 1998. Differential expression of chemokine receptors and chemotactic responsiveness of type 1 T helper cells (Th1s) and Th2s. J. Exp. Med. 187: 129 –134. 43. Rivino, L., M. Messi, D. Jarrossay, A. Lanzavecchia, F. Sallusto, and J. Geginat. 2004. Chemokine receptor expression identifies pre-T helper (Th)1, pre-Th2, and nonpolarized cells among human CD4⫹ central memory T cells. J. Exp. Med. 200: 725–735. 44. Chen, M. L., M. J. Pittet, L. Gorelik, R. A. Flavell, R. Weissleder, H. von Boehmer, and K. Khazaie. 2005. Regulatory T cells suppress tumorspecific CD8 T cell cytotoxicity through TGF-␤ signals in vivo. Proc. Natl. Acad. Sci. USA 102: 419 – 424.

5283 45. Ochando, J. C., A. C. Yopp, Y. Yang, A. Garin, Y. Li, P. Boros, J. Llodra, Y. Ding, S. A. Lira, N. R. Krieger, and J. S. Bromberg. 2005. Lymph node occupancy is required for the peripheral development of alloantigen-specific Foxp3⫹ regulatory T cells. J. Immunol. 174: 6993–7005. 46. Siegmund, K., M. Feuerer, C. Siewert, S. Ghani, U. Haubold, A. Dankof, V. Krenn, M. P. Schon, A. Scheffold, J. B. Lowe, et al. 2005. Migration matters: regulatory T cell compartmentalization determines suppressive activity in vivo. Blood 106: 3097–3104. 47. Chen, W., D. Zhou, J. R. Torrealba, T. K. Waddell, D. Grant, and L. Zhang. 2005. Donor lymphocyte infusion induces long-term donor-specific cardiac xenograft survival through activation of recipient double-negative regulatory T cells. J. Immunol. 175: 3409 –3416. 48. Moser, B., P. Schaerli, and P. Loetscher. 2002. CXCR5⫹ T cells: follicular homing takes center stage in T-helper-cell responses. Trends Immunol. 23: 250 –254. 49. Cyster, J. G., K. M. Ansel, K. Reif, E. H. Ekland, P. L. Hyman, H. L. Tang, S. A. Luther, and V. N. Ngo. 2000. Follicular stromal cells and lymphocyte homing to follicles. Immunol. Rev. 176: 181–193. 50. Moser, B., and L. Ebert. 2003. Lymphocyte traffic control by chemokines: follicular B helper T cells. Immunol. Lett. 85: 105–112. 51. Smith, J. R., R. M. Braziel, S. Paoletti, M. Lipp, M. Uguccioni, and J. T. Rosenbaum. 2003. Expression of B-cell-attracting chemokine 1 (CXCL13) by malignant lymphocytes and vascular endothelium in primary central nervous system lymphoma. Blood 101: 815– 821. 52. Barone, F., M. Bombardieri, A. Manzo, M. C. Blades, P. R. Morgan, S. J. Challacombe, G. Valesini, and C. Pitzalis. 2005. Association of CXCL13 and CCL21 expression with the progressive organization of lymphoid-like structures in Sjogren’s syndrome. Arthritis Rheum. 52: 1773–1784. 53. Corcione, A., S. Casazza, E. Ferretti, D. Giunti, E. Zappia, A. Pistorio, C. Gambini, G. L. Mancardi, A. Uccelli, and V. Pistoia. 2004. Recapitulation of B cell differentiation in the central nervous system of patients with multiple sclerosis. Proc. Natl. Acad. Sci. USA 101: 11064 –11069. 54. Magliozzi, R., S. Columba-Cabezas, B. Serafini, and F. Aloisi. 2004. Intracerebral expression of CXCL13 and BAFF is accompanied by formation of lymphoid follicle-like structures in the meninges of mice with relapsing experimental autoimmune encephalomyelitis. J. Neuroimmunol. 148: 11–23. 55. Zheng, B., Z. Ozen, X. Zhang, S. De Silva, E. Marinova, L. Guo, D. Wansley, D. P. Huston, M. R. West, and S. Han. 2005. CXCL13 neutralization reduces the severity of collagen-induced arthritis. Arthritis Rheum. 52: 620 – 626. 56. Shi, K., K. Hayashida, M. Kaneko, J. Hashimoto, T. Tomita, P. E. Lipsky, H. Yoshikawa, and T. Ochi. 2001. Lymphoid chemokine B cell-attracting chemokine-1 (CXCL13) is expressed in germinal center of ectopic lymphoid follicles within the synovium of chronic arthritis patients. J. Immunol. 166: 650 – 655. 57. Sato, T., S. Ishikawa, K. Akadegawa, T. Ito, H. Yurino, M. Kitabatake, H. Yoneyama, and K. Matsushima. 2004. Aberrant B1 cell migration into the thymus results in activation of CD4 T cells through its potent antigen-presenting activity in the development of murine lupus. Eur. J. Immunol. 34: 3346 –3358. 58. Steinmetz, O. M., U. Panzer, U. Kneissler, S. Harendza, M. Lipp, U. Helmchen, and R. A. Stahl. 2005. BCA-1/CXCL13 expression is associated with CXCR5positive B-cell cluster formation in acute renal transplant rejection. Kidney Int. 67: 1616 –1621. 59. Chang, J. C., L. Zhang, S. G. Distler, G. Ziang, and A. M. Kaplan. 1992. Characterization and function of CD3⫹CD4⫺CD8⫺ TCR-␣␤ bearing cells infiltrating the lung during the immune response. Reg. Immunol. 4: 25–33. 60. Sprent, J., and C. D. Surh. 2002. T cell memory. Annu. Rev. Immunol. 20: 551–579.