CCR7 Signaling Inhibits T Cell Proliferation¹

Effective T cell priming depends on distinct interactions between T cells and APCs in secondary lymphoid organs. The contact between both cell types is the result of a highly coordinated migration and functional maturation program in which the chemokine receptor CCR7 and its homeostatic ligands, CCL19 and CCL21, play pivotal roles (1). The essential role of CCR7 and its ligands in the migration of mature dendritic cells (DCs)³ and naive T cells to lymph nodes was largely discovered by studying plt/plt mice, which lack the CCR7 ligands (2), and CCR7 gene-deficient (CCR7^−/−) mice (3). Among other mediators, activated DCs produce proinflammatory cytokines and subsequently up-regulate CCR7, which makes them responsive to CCR7 ligands. This in turn promotes DC migration from peripheral tissues to secondary lymphoid organs, such as lymph nodes and spleen. Furthermore, CCR7 is also important in mediating colocalization and, thus, Ag-specific T cell activation (4, 5).

An efficient way to induce immunosuppression is to inhibit the entry and exit of immune cells in secondary lymphoid organs. Proof of the principle for this mechanism of immunosuppression was shown by the efficacy of FTY720. This drug induces a migratory block that prevents lymphocytes from exiting the lymph nodes. A notable immunosuppressive effect could be demonstrated in human solid organ transplantation (6, 7).

There is emerging evidence that CCR7 signaling can also influence other key steps in the immune response. For example, CCL19 costimulates LFA-1 activation (8) by accelerating the movement of T cells scanning for their Ag (9, 10). Furthermore, subsets of T cells require CCR7 for efficient priming (11), and DC maturation and endocytic capacity are facilitated by CCL19 (12, 13, 14).

Although these effects are still being investigated, blocking these coactivation capacities and impairing cellular migration are promising targets for suppressing cellular immune responses.

In our previous work (15), systemic application of pharmacological doses of CCL19-IgG in mice was used to disturb this particular chemokine/chemokine-receptor system. We were able to show that recirculation and colocalization of T cells and DCs in secondary lymphoid organs were impaired, which prompted us to test CCL19-IgG in different models of solid organ transplantation. Using murine models for allogeneic kidney as well as allogeneic heart transplantation, we were able to show that treatment with CCL19-IgG significantly delayed allograft rejection in vivo. In particular, we could further demonstrate that treatment with CCL19-IgG had a significant impact on Ag-induced T cell proliferation in vivo. However, because these in vivo experiments cannot differentiate between effects mediated by impaired migration and/or potential direct cellular events, we set up this study to test the hypothesis that the CCR7 ligands CCL19 and CCL21 are responsible for reduced T cell proliferation by a direct cellular effect.

In the current study, we present evidence that CCL19 and CCL21 are also able to directly inhibit T cell proliferation after TCR activation, and we show that CCR7 signaling is linked to the regulation of cell cycle progression.

These data add a new aspect to the multiple capabilities of this chemokine receptor. We also show that chemokines and their receptors are closely linked not only to cell migration but also to the homeostasis of the immune response by controlling other cell functions such as proliferation.

The CCL19-IgG fusion protein was generated as described previously (15). The CCL21-IgG construct was generated by exchanging the CCL19 domain with the CCL21 domain, which was obtained by PCR amplification of murine lymph node cDNA. The primers were designed with restriction sites appropriate for ligation into the IgG1-coding vector pCDNA3.1 (sense: 5′-CTGCCTAAGCTTTCTGCCATGGCTCAGATGATGACTCT GAG-3′; antisense: 5′-CACATCTCGGATCCCATCCTCTTGAGGGCT GTGTCTG-3′).

The chimeric molecules were expressed in suspension Chinese hamster ovary (CHO) cells that were adapted to grow in serum-free ProCHO4-CDM medium (Cambrex). The culture supernatant containing either secreted CCL19-IgG1 or CCL21-IgG was purified using protein A-Sepharose Fast Flow (Amersham Biosciences) as described previously (16). Protein concentrations were determined with the Bio-Rad protein assay; serum concentrations in mice were determined by an ELISA capture assay as previously described (17). The biologic activity of the recombinant fusion proteins was routinely checked in a chemotaxis assay.

To exclude an Fc-mediated effect by FcγR binding or complement activation, we created a CCL19-control protein (CCL19-ΔIgG). Three amino acids within the hinge region critical for FcγR binding (18) at positions 233 to 235 were mutated (ELL→PVA) using the QuikChange site-directed mutagenesis kit from Stratagene.

Female inbred C57BL/6 (H2^b), female BALB/c (H2^d), and OVA TCR-transgenic DO11.10 mice were supplied by Charles River Laboratories and housed at the central animal facilities of the University of Kiel (Kiel, Germany) under conventional conditions. Generation of CCR7^−/− mice has been described earlier (19). These animals were maintained at the central animal facility of Hannover Medical School (Hannover, Germany). The CCR7-deficient mice have been backcrossed to the BALB/c genetic background for at least seven generations. All mice were maintained under specific pathogen-free conditions and were used at 8–12 wk of age in all experiments.

HUT78, a CCR7-expressing human T cell lymphoma cell line, and CHO cells were obtained from American Type Culture Collection. Cell lines were grown in RPMI 1640 supplemented with 10% FCS, 2 mM l-glutamine, streptomycin (100 μg/ml), and penicillin (100 U/ml). For large-scale protein production and purification, the adherent transfected CHO cells were successively adapted to serum-free conditions and maintained in ProCHO4-CDM medium.

Preparation and culture of bone marrow cells to generate DCs has been described previously (20). For DC maturation, 1 μg/ml LPS and 1 μg/ml PGE₂ (Sigma-Aldrich) were added on day 10 for 24 h. For Ag-specific T cell priming experiments, bone marrow DCs were additionally pulsed with 5 μg/ml OVA peptide 323–339 (NeoMPS).

Murine lymphocytes were isolated from spleen and lymph node suspensions of untreated mice. T cells were purified using a MACS system (Miltenyi Biotec) after magnetic labeling and depletion of non-T cells (Pan T cell Ab mixture; Miltenyi Biotec). Human T cells were extracted from healthy donor blood. Purity levels of >90% of T cells were confirmed by anti-CD3 staining (anti-CD3-PC5, clone UCHT1; Beckman Coulter) and FACS analysis. FACS analysis was performed on a Coulter Epics XL flow cytometer (Beckman Coulter). Analyses of flow cytometry listmode data files were performed using WinMDI 2.8 software.

To perform a MLR, splenocytes from C57BL/6 mice were treated with 50 μg/ml mitomycin C (Sigma-Aldrich) and mixed (as stimulator cells) with allogeneic BALB/c spleen cells (as responders) at a 1:1 ratio and a final density of 2 × 10⁵ cells/well. For in vitro T cell proliferation assays, primary T cells were cultured with either 0.1–3 μg/ml plate-bound anti-human CD3 (clone OKT3; Janssen-Cilag) or anti-mouse CD3 (clone 145-2C11; BD Pharmingen) and 1 μg/ml anti-human CD28 (clone CD28.2, BD Pharmingen) or anti-mouse CD28 (clone 37.51, BD Pharmingen), respectively. For Ag-specific T cell priming experiments, primary T cells were cocultured with activated DCs at a DC/T cell ratio of 1:40. Cells were seeded (in quadruplicate) in 96-well round-bottom microtiter plates in the presence of rCCL19 (R&D Systems), CCL19-IgG, CCL19-ΔIgG, CCL21-IgG, or equimolar ChromePure human IgG (hIgG; Dianova) at 37°C for a total of 4 days. As an additional specificity control 2.5 μM eotaxin (CCL11) (R&D Systems) was added in a subset of assays. After 3 days, 10 μM BrdU was added to each well. T cell proliferation was determined after fluorescence labeling with PKH26 (Sigma-Aldrich) and FACS analysis or by using a chemiluminescence immunoassay based on the measurement of BrdU incorporation during DNA synthesis according to the manufacturer’s instruction (Roche Diagnostics). Signals were quantified using a Victor Light 1420 luminometer (PerkinElmer).

Plastic plates were preincubated with CCL19-IgG or CCL21-IgG (2–200 μg/ml) in combination with an anti-CD3 Ab (2 μg/ml) to analyze the effect of immobilized chemokine ligands. Up to a final concentration of 200 μg/ml total protein hIgG or BSA was supplemented to ensure equal amounts of plate-bound anti-CD3 while using different concentrations of chemokine ligands. As a negative control, anti-CD3 was preincubated with either hIgG or BSA alone (each 200 μg/ml) without chemokine ligands. After 2 h at 37 °C, plates were washed twice and cell suspensions containing anti-CD28 (1 μg/ml) were incubated for 3 days on coated plates before analyzing proliferation using BrdU incorporation.

To measure IL-2 production, cells were stimulated as described and supernatants were obtained after 24 h by centrifugation. IL-2 concentrations were quantified using an ELISA kit (IL-2 Quantikine; R&D Systems).

Surface marker expression was analyzed by FACS analysis using fluorescence-labeled mAbs (clones anti-CD3 UCHT1, anti-CD4 13B8.2, and anti-CD8 B9.11, Beckman-Coulter; clones anti-CD69 FN50 and anti-CD25 M-A251, BD Pharmingen).

Cultured cells (5 × 10⁵) were stimulated as described above for 4 days and washed with PBS buffer before resuspension in propidium iodide/sodium citrate/Triton X-100 solution (50 μg/ml, 0.1%, and 0.1%, respectively). Cells were incubated at 4°C in the dark for 20 min. Stained cells were analyzed by flow cytometry. The percentages of cells in different stages of the cell cycle were determined.

Protein levels of p27^Kip1, CDK1 (BD Biosciences), and β-actin (Cell Signaling Technology) were determined by Western blotting (17).

Data are shown as mean ± SD. The difference between groups was tested by using Student’s t test for statistical significance with SPSS 12.01 software.

We have previously shown that the CCL19-IgG fusion protein is able to suppress an alloimmune response in vivo (15). This effect can be partially explained by an impairment of T cell and APC accumulation in secondary lymphoid organs, reducing T cell Ag priming.

To determine whether this immunosuppressive effect is also mediated by a direct cellular effect of CCL19 or its second ligand, CCL21, independently of cell trafficking, we used the MLR with BALB/c splenocytes as effector cells and mitomycin C-treated C57/B6 splenocytes as stimulator cells in an in vitro alloimmune response model. Fig. 1 illustrates that the presence of CCL19 significantly reduced the proliferative response of splenocytes in a concentration-dependent manner. Recombinant CCL19 and equimolar concentrations of CCL19-IgG or CCL21-IgG showed a comparable effect, whereas control hIgG had no influence on proliferation compared with vehicle treated cells. To exclude the possibility that the inhibitory effect on MLR was influenced by the IgG part of the fusion protein, we repeated the experiment with a control fusion protein (CCL19-ΔIgG) carrying a mutation in the Fc receptor binding domain (18). The effects mediated by the CCL19-Fc-mutated IgG fusion protein and the nonmutated CCL19-IgG were identical. The presence of either CCL19-IgG or CCL21-IgG reduced the proliferation of effector cells. Thus, both CCR7 ligands specifically inhibit Ag-induced splenocyte proliferation.

We next wanted to determine whether CCL19 and CCL21 exert antiproliferative effects directly on T cells. Furthermore, we aimed to clarify whether CCL19 acts as an overall inhibitor of T cell proliferation or specifically inhibits T cell activation due to TCR engagement.

We initially chose the human T cell line HUT78 because these cells constitutively express the chemokine receptor CCR7 at high levels and respond to CCL19 chemotactically (17, 21). Fig. 2,A shows that CCL19 strongly inhibited the spontaneous proliferation of this T cell line when used at higher concentrations than those necessary for optimal chemotactic responses in Transwell assays (50–200 nM) (22). At concentrations >1 μM, highly significant inhibition of HUT78 cell proliferation was observed in a dose-dependent manner. This effect was not attributable to increased cell death or apoptosis, which were measured by propidium iodide and annexin V staining, respectively (data not shown). This effect could also be demonstrated when CCL21-IgG was present. This experiment, using a human T cell line, also shows that CCL19 and CCL21 can inhibit proliferation independently of TCR engagement. The concentrations of CCL19, CCL19-IgG, CCL21-IgG, and hIgG from Fig. 1 were used.

In addition, it was of interest to determine whether the antiproliferative effect mediated by both CCR7 ligands is also operative in T cells activated by TCR engagement. For this experiment, T cells were purified from PBMCs or splenocytes and stimulated with plate-bound anti-CD3 and anti-CD28 Abs. In the presence of CCL19, human T cell proliferation was significantly reduced compared with vehicle-treated controls (Fig. 2 B). Equimolar concentrations of either CCL19-IgG or CCL21-IgG caused the same inhibitory effect without significant differences between each other.

Next, we studied the effect of the fusion protein in purified T lymphocytes stimulated with optimal and suboptimal doses of anti-CD3 and 1 μg/ml anti-CD28, respectively (Fig. 2 C). CCL19-mediated inhibition of T cell proliferation was most pronounced under conditions of TCR stimulation with suboptimal doses of anti-CD3 (≤0.3 μg/ml). To exclude effects mediated by unspecific binding of the chemokine ligands to surface receptors and, thus, unspecific signaling, we analyzed the effects of high concentrations of another CC chemokine, eotaxin (CCL11). Eotaxin did not influence the proliferation of T cells. Because this chemokine lacks a specific receptor on T cells, the lack of inhibitory effect supports the specificity of the effects mediated by CCL19 and CCL21.

In addition, we analyzed the fluorescence fading of labeled lymphocytes in purified human CD3⁺ T cells after stimulation with immobilized anti-CD3/anti-CD28 mAbs over time. Fig. 2 D shows that activated T cells underwent a reduced in vitro cellular expansion of 35.7% in the presence of CCL19-IgG in comparison to 82.9% in the control group. Examination of T cell subsets indicates that the inhibition of cell division is evident in both CD4⁺ and CD8⁺ T cells. The proliferative response of CD8 T cells to anti-CD3/anti-CD28 was greater than that of CD4 T cells. In the CD8⁺ T cell subset, the inhibition of cell division seems to be more pronounced.

To further substantiate the observation that the inhibitory effects are solely mediated by the chemokine receptor CCR7, we analyzed T cells from CCR7 knockout mice. Fig. 3,A shows the anti-CD3-/anti-CD28-induced proliferation of CCR7^−/− T cells compared with wild-type T cells. The presence of CCL19 had no influence on CCR7 knockout T cell proliferation. In addition, the proliferative response was significantly reduced in both control and CCL19-treated cells compared with wild-type T cells. Fig. 3 B confirms this finding: CCR7^−/− T cells exhibit less proliferation than wild-type T cells in DC-mediated stimulation. Again, CCR7^−/− T cells show markedly reduced proliferation after stimulation independent of treatment that was more profoundly reduced than that mediated by CCL19-IgG in wild-type T cells, indicating a robust influence of CCR7 on proliferation. Compared with TCR cross-linking with anti-CD3/anti-CD28, DC-mediated proliferation was reduced to a lesser extent, albeit still significantly.

To verify the above results in a more physiological system, we used OVA transgenic CD4 T cells stimulated by OVA-pulsed DCs. Fig. 3 C shows that Ag-specific DC-induced T cell proliferation using OVA TCR transgenic T cells is also significantly reduced by CCL19-IgG. Moreover, we could not detect a difference in proliferative response when CCR7^−/− were used as stimulator cells instead of wild-type DCs. This demonstrates that the inhibitory effect of CCL19 on the proliferative response is mediated by T cell CCR7 and that CCR7 signaling in DCs does not contribute to this effect.

To identify the underlying molecular mechanisms for our findings, we next analyzed the key steps influencing T cell proliferation, cytokine secretion, and cell cycle progression.

Because of the pivotal role of IL-2 in T cell activation and proliferation, we analyzed IL-2 production and the expression of IL-2Rα (CD25) in activated T cells. Fig. 4,A indicates that primary human T cells secreted significant amounts of IL-2 after anti-CD3/anti-CD28 stimulation. The presence of CCL19 led to drastically reduced IL-2 release, demonstrating a down-regulation of this important autocrine activation loop by CCL19, which may contribute to a reduction in T cell proliferation. However, Fig. 4,B demonstrates that exogenously adding IL-2 up to 100 ng/ml did not abolish the antiproliferative effects of CCL19. In contrast to IL-2 secretion, we detected no relevant influence of CCL19 on other activation markers, such as expression of CD25 or CD69, after TCR stimulation. FACS analysis shows that expression of both receptors increased after a 24-h stimulation period with anti-CD3/anti-CD28, independently of the presence of CCL19 (Fig. 4 C). In addition, phosphorylated target proteins of the early TCR-induced signaling cascade, like STATs, protein kinase C, and IκB (23, 24), were not significantly affected by CCL19 (data not shown).

Recent studies demonstrated that the function of a variety of chemokines is partly dependent on surface presentation and immobilization (25, 26). With regard to this physiological condition, we asked whether immobilization of the CCR7 ligands CCL19 and CCL21 would influence the demonstrated inhibitory effects. Therefore, we precoated plastic plates with different amounts of chemokines in combination with anti-CD3 Ab to imitate a surface-bound presentation. Subsequently, T cells were stimulated by incubation on such a surface. Fig. 5 shows that both CCR7 ligands are also effective in inhibiting the proliferative response to anti-CD3/anti-CD28 stimulation when immobilized. Neither immobilized hIgG nor BSA was able to inhibit proliferation. The inhibitory effect was concentration dependent.

When high concentrations of immobilized CCL19 or CCL21 were presented on the surface, no further inhibitory effect could be elicited by adding soluble CCR7 ligands. Furthermore, soluble CCL19 could not increase the effect when immobilized CC21 induced the maximal inhibition of T cell proliferation (nor vice versa; Fig. 5 B). This suggests that both CCR7 ligands use identical signaling pathways to inhibit proliferation.

TCR activation forces lymphocytes to enter the cell cycle (27). Because of the reduced cell division and the lack of increased cell death after CCL19 addition, it was intriguing that the cell cycling of T cells may be influenced by CCL19. Thus, we analyzed the cell cycle distribution of antiCD3-/anti-CD28-activated T cells in the presence or absence of CCL19. We performed time course experiments (12, 24, 26, 48, 60, and 72 h of activation) to determine which phases of the cell cycle were influenced by CCL19. As shown in Fig. 6,A, resting unstimulated T cells sustained for 4 days were almost exclusively in the G₀/G₁ phase of the cell cycle in the absence of mitogenic stimulation or CCL19-IgG. Treatment with anti-CD3/anti-CD28 for 4 days induced the proliferation of primary T cells as indicated by a shift of lymphocytes into the S and G₂/M phases. The proportion of T cells within the S and G₂/M phases was significantly increased by CCL19, from 6.54 to 19.9%, consistent with a cell cycle delay or arrest in activated T cells induced by CCL19. We next determined whether the influence of CCL19 on the cell cycle involves previously identified regulatory proteins. We observed that cyclin-dependent kinase (CDK) inhibitor p27^Kip1 was down-regulated in TCR-stimulated cells (Fig. 6,B). This effect was abolished in the presence of CCL19, which is consistent with the increased number of activated T cells in the S and G₂/M phases (Fig. 6 A). Furthermore, CDK1, a serine-threonine kinase required for entry into mitosis, was down-regulated by CCL19 treatment of activated T cells. TCR stimulation alone results in strong up-regulation of CDK1 in proliferating T cells. CCL19-IgG-treatment during TCR activation was associated with reduced levels of CDK1 and high levels of p27^Kip1.

CCR7 is a prerequisite for immune cells to home into secondary lymphoid organs. Moreover, evidence has accumulated over recent years indicating that CCR7-mediated signaling influences more cell functions than migration. For example, recently it has been shown in DCs that CCR7 ligands induce alterations in cytoarchitecture, increase the ability of endocytosis, and promote survival (14, 28, 29). In T cells, CCR7 is responsible for a variety of proactivation events such as migratory speed or activation of LFA-1 or as a costimulatory factor in the priming of T cell subtypes (8, 11, 30).

In a previous study, we showed that targeting the CCL19/CCR7 system can also result in a profound immunosuppression that cannot be solely explained by the impairment of cell migration (15).

In the current study, we present data showing that the CCR7 ligands CCL19 and CCL21 are not only capable of impairing immune responses by influencing the migratory behavior of T cells and DC in vivo but that these ligands can also directly induce a thus far undescribed inhibitory program. As demonstrated by MLR, CCL19 and CCL21 can reduce alloimmune responses. We could attribute this effect to reduced T cell proliferation. This effect is valid for CCR7-expressing T lymphoma cells and both murine and human T cells. It applies to CD4⁺ as well as CD8⁺ T cells, and we demonstrate that it is independent of the mode of stimulation, namely TCR cross-linking or DC-mediated activation.

The proliferation stimulus in T cells was induced by either Ab- or DC-mediated TCR activation. However, even though the immortalized transformed lymphoma cell line HUT78 does not express CD3 (31), the antiproliferative properties of CCL19 could still be observed. A possible explanation of these observations is a direct effect of CCR7 signaling on cell proliferation that is independent of TCR signals. In agreement with this hypothesis, CCR7^−/− T cells show reduced proliferation upon TCR stimulation. To our knowledge, there are no published reports that demonstrate an altered proliferation pattern of CCR7^−/− T cells. In this study we could show for the first time that the genetic knockout of CCR7, independent of CCL19 signaling, is sufficient for a reduced proliferative response of T cells, both after TCR cross-linking as well as after DC-mediated stimulation. As expected, exogenous CCL19 did not have any influence on the proliferation rate of these cells because the specific receptor is missing. Furthermore, the CC chemokine eotaxin (CCL11), the receptor of which is not expressed on lymphocytes, did not have any influence on proliferation. These results are convincing evidence that CCR7 ligands specifically exert the demonstrated effects via CCR7 and that expression and function of CCR7 are linked to proliferation control.

Because CCR7-mediated effects are independent of TCR activation, the downstream signaling events of CCR7 are likely to be distinct from TCR signaling pathways. This is substantiated by the fact that early TCR-mediated activation signals seem to be unaffected by CCR7 engagement. For example, the expression profiles of the very early activation markers CD25 or CD69 were not reduced by CCL19 treatment. In addition, there were no differences in key signals downstream of the TCR, such as STAT or protein kinase C activation. Moreover, the effect of CCL19 on T cell division was cumulative over a 1-wk observation period with increasing differences in cell division after longer incubation. Thus, we concluded that early T cell activation is not responsible for the demonstrated inhibitory effect of CCL19 and that the sustained presence of CCL19 is able to continuously decelerate the division of CCR7⁺ T cells, which would be consistent with an impaired cell cycle progression.

The analysis of the distribution of cells within the cell cycle confirmed this. The cell cycle distribution of T cells cocultured with CCL19 showed a significant increase in G₂/S as compared with untreated cells. Consistent with this finding, CCL19 induced an up-regulation of p27^Kip1 and a down-regulation of CDK1, both leading to a transitional block at the G₂ phase. To our knowledge, this is the first time that a cell cycle-regulating protein has been associated with the CCR7 signaling pathway.

Nevertheless, there are a few examples of a link between chemokine receptors and cell cycle regulation in the literature. The CXC chemokine CXCL4 (PF4) has been identified as a cell proliferation blocker in endothelial cells. PF4-induced signaling resulted in the down-regulation of cell cycle regulators like p21^Cip1/WAF1 (32). Recently, it was shown that CXCR4 influences cell cycle-regulating proteins in medulloblastoma cells (33). Moreover, the chemokine CCL2 is thought to be a direct inhibitor of the proliferation of pancreas carcinoma cells (34). CXCR3 was recently be described as a metastasis-promoting receptor (35), but when used in pharmacological concentrations (>0.5 μM) its ligands negatively influence the growth of myeloma cells (36). Interestingly, the connection between chemokine receptors and cell cycle control might already be used therapeutically, as Fujino and colleagues suggest (37). They ascribe one of the pleiotropic effects of statins to an influence on the cell cycle-regulating proteins p21 and cyclin D1, which were associated with the down-regulation of CCR2 and CCR5 (37).

These examples show that there is expanding evidence that chemokine receptors and their ligands, known to mediate trafficking events, are also implemented in pivotal cell function control, such as proliferation.

In conclusion, we identified the modulation of cell proliferation as a new function of the chemokine receptor CCR7. CCR7^−/− cells as well as continuous receptor engagement by high concentrations of the agonistic ligand CCL19 show reduced proliferation caused by cell cycle arrest.

To date, it has been reported that CCR7 ligands have proactivation capabilities. For example, CCL19 facilitates the maturation of DCs and thus enhances T cell priming (12). A direct effect on T cell proliferation was not observed using concentrations up to 0.1 μM CCL19. In this study, we observe inhibitory effects that are not DC-mediated because there is no difference when stimulating with wild-type or CCR7^−/− DC (Fig. 3 B). This indicates that the inhibition of proliferation is due to an influence on proliferation itself and not as a consequence of altered priming. Furthermore, we used concentrations higher than those used for optimal chemotaxis response. Consistent with our data, 0.1 μM CCL19 did not influence T cell proliferation directly (12).

The physiologic concentrations of a chemokine in tissue are hard to determine exactly. Concentrations in certain microenvironments might well be higher than what is suggested by an analysis of whole tissue. Local concentrations, e.g., in the center of a lymph node, may well be high enough to result in immune-modulating effects. Recently, published work has confirmed this thesis. Concentrations of 2 to 10 μg/ml CCR7 ligands were measured in lymph node tissue (9), which is in the range used in our study. For example, on endothelial surfaces even locally higher concentrations are discussed, leading to growing evidence that in vivo chemokines are likely to be presented bound to surface structures like glycosaminoglycans (25). Moreover, the functional relevance of this surface binding has been demonstrated for a couple of CC chemokine receptors (26, 38). For example, the CCR2 ligand RANTES could not promote migration in an in vivo peritonitis model if its glycosaminoglycan binding domain was mutated, although this mutated chemokine could induce chemotaxis in vitro as well as nonmutated RANTES (39). Therefore, to test whether the inhibitory effects are influenced by the mode of chemokine presentation, we tested the inhibitory potency of surface-immobilized CCL19 and CCL21. Again, a high density on the surface of CCL19 and CCL21 was equally effective in inhibiting the proliferative response of T cells.

Although these in vitro models cannot assess quantitatively whether bound CCR7 ligands are more potent than soluble ones in terms of inhibition of proliferation, the concentration dependence in both settings shows that a sufficient number of CCR7 receptors have to be engaged to elicit this negative signaling. Although there is data that show differing functions of CCL19 and CCL21 (40), both chemokines induced the same effects in all of our experiments. Furthermore, we could not demonstrate an additional effect if one of them was used in an assay with the other present at the maximal concentration. This strongly suggests that both ligands induce the same signaling pathway, leading to inhibition of proliferation.

These results could be valid for the physiological environment within lymph nodes in which high densities of surface-bound chemokines, high concentrations of soluble chemokines, and T cells with various degrees of activation occur. In such a microenvironment, bystander T cells that do not receive optimal antigenic stimulation could be inhibited from proliferating by high concentrations of CCL19 or CCL21, whereas cells with strong TCR stimulation would proliferate despite CCR7 engagement. This model is consistent with the finding that higher concentrations of anti-CD3 or stimulation by DCs were less affected by CCL19-mediated inhibition of proliferation than by suboptimal stimulation (Fig. 2 C).

As another example, environments rich in chemokine ligands could also occur during lymphoid tissue development and could control organ structure by influencing proliferation. The altered lymphoid structures of CCR7^−/− mice are indeed a result of both impaired homing (19) and lymphoid organogenesis (41).

These examples illustrate that CCR7 ligands may induce antiproliferative effects in physiological situations. In addition, the finding that CCR7 is linked to proliferation control offers interesting options for therapeutic interventions independently of the physiological range of in vivo concentrations and the mode of presentation.

In vivo use of CCL19-IgG has already proven to mediate immunosuppression at least by impairment of migration and colocalization. The antiproliferative capacity of CCL19 described here could functionally contribute to this impairment of allograft responses in mice. For the in vitro proliferation assays, we chose concentrations of CCR7 ligands comparable to those that were used in the in vivo transplantation studies (15). The results allow differentiating intracellular signaling events independently of the influence on recirculation or colocalization and show a direct cellular effect by CCR7-mediated signaling.

Recently, different reports have already demonstrated that the chemokine receptor CCR7 and its ligands can modulate the immune response by participating in signaling events apart from chemotaxis and maturation (42, 43), but it has not been previously demonstrated that CCL19 or CCL21 itself directs CCR7⁺ cells to an antiproliferative status. However, there are a few reports that have already indicated inhibitory effects. One example is that CCL19 is able to inhibit colony forming of human CCR7⁺CD34⁺ chronic myelogenous leukemia progenitor cells (44).

From a clinical aspect, this newly described capacity of modulation might render CCR7 an even more valuable target for therapy, not only for the purpose of immunosuppression but also for tumor therapy, because CCR7-expressing metastatic tumors have a poor prognosis (45, 46, 47, 48). CCR7 targeting might work not only by interfering with tumor cell homing, and thereby metastases, but also by reducing the growth of these tumors. Alfonso-Perez and colleagues (49) have demonstrated a largely specific lysis of CCR7-positive lymphoma cells in vitro by using Ab-mediated complement cytotoxicity. Although we used another approach, this work shows that the high expression of CCR7 in tumor cells makes them an easy target for CCR7-directed therapy. One possible therapeutical approach is to use, as in this study, agonistic ligands in high concentrations. A combination of traffic impairment, inhibition of proliferation by agonistic signaling, and targeting the receptor to lyse the cell could indeed be a powerful, clinically usable tool.

In summary, our results indicate that the CCR7 ligands CCL19 and CCL21 are more than just potent chemoattractants for T cells but are sufficient to mediate antiproliferative effects via CCR7 by influencing cell cycle progression. Thus, we show a novel association between chemotaxis and proliferation of T cells and thereby extend the view on the functions of the receptor CCR7. To date, CCR7 is known to mediate LFA-1 up-regulation, providing T cells with costimulatory signals (8), whereas higher concentrations of CCL19 or CCL21 suppress the production of the costimulatory cytokine IL-2. Furthermore, CCR7 signals support the phosphorylation of activating molecules and induce calcium flux (40, 50), while CCR7 ligands also induce cell cycle arrest and an inhibit T cells from responding to Ags by proliferating. Although the circumstances under which these different properties are effective in vivo need clarification, the inhibitory function of CCR7 might be of clinical relevance. Indeed, under certain conditions the net effect in vivo is a suppression of the allograft immune response. Thus, CCR7 features at least two outstanding properties in terms of immunosuppression: orchestrating cellular traffic into secondary lymphoid organs and modulating T cell proliferation. The potency of CCL19-IgG to both provide antiproliferative effects and interfere with the trafficking behavior of CCR7⁺ cells might be used to extend current immunosuppression therapies.

We thank Mareike Newsky for technical assistance.

The authors have no financial conflict of interest.