The Multiple Personalities of the Chemokine Receptor CCR7 in Dendritic Cells¹

Dendritic cells (DCs)⁴ are potent professional APCs that derive from bone marrow progenitors (1, 2, 3). Progenitors may differentiate into circulating precursors that later home to peripheral tissues as immature myeloid DCs (1, 2, 3). Alternatively, progenitors may differentiate into blood plasmacytoid DCs (PDCs), another DC subtype not normally found in peripheral tissues, that enter directly in the lymph nodes through the high endothelial veins (HEVs) (4, 5). Unless indicated otherwise, the results discussed below refer mainly to myeloid DCs on which most information is available. Myeloid DCs (hereafter called simply DCs) are found in the immature state in epithelia and in the interstitial space of most solid organs where they sample the environment by Ag uptake through receptor-mediated endocytosis, phagocytosis, and macropinocytosis but display a low T cell stimulatory ability (1, 2, 3). Upon sensing of “danger signals” (tissue damage, inflammatory cytokines, or pathogens), immature DCs start a differentiation process called maturation (1). The phenotypical changes that take place during maturation include the blunting of the endocytic ability and the up-regulation of MHCs and costimulatory molecules that eventually transform these DCs in potent APCs. Maturation takes place concomitantly with the migration of the DCs from their niches in the peripheral tissues to the lymph nodes where they arrive through the lymphatics. In this regard, during maturation is up-regulated the expression of the chemokine receptor CCR7 that guides the migratory DCs to the nodes (6, 7, 8).

Apart from the maturing DCs, CCR7-expressing subsets of immature DCs (often called “semimature” DCs because of their intermediate phenotype) use this receptor to migrate continuously, in the absence of danger signals, to the lymph nodes, where they contribute to the peripheral immune tolerance against self-Ags (9, 10). Finally, CCR7 also directs the PDCs to the lymph nodes, although, as mentioned above, these cells use HEVs to penetrate in the nodes (5, 8).

CCR7 has two ligands, CCL19 and CCL21, that are highly expressed by stromal cells in the T cell-rich lymph node areas (7, 8). In addition, CCL19 is expressed by mature DCs and is presented on the luminal side of the HEV cells (7, 8). CCL21 is expressed by afferent lymphatic endothelial cells and by HEV cells in mice (in humans CCL21 is only exposed on the lumen, but not expressed, by the HEVs) (7, 8, 11). The importance of CCR7 for the immune response is shown clearly in mice deficient in this receptor, which lack contact sensitivity and delayed-type hypersensitivity reactions and show severely delayed kinetics regarding the Ab response (6).

Chemotaxis has been the function of chemokine receptors most extensively studied in DCs (12). In this regard, in the immature DCs, it has been shown that the chemotaxis was mediated by inflammatory chemokine receptors CCR1, CCR2, CCR5, CCR6, CXCR1, and CXCR2, which guide these cells to inflammatory sites (12). In both immature and mature DCs, the chemotaxis can also be mediated by chemokine receptor CXCR4 (12). Apart from chemotaxis, less is known on other possible functions of chemokine receptors in DCs. CCR7 was described initially as a potent chemotactic receptor expressed by migratory leukocytes, including DCs (13, 14). However, recently, several reports have shown that, apart from chemotaxis, CCR7 can control additional functions in these cells (15, 16, 17, 18, 19).

It was observed that the stimulation of CCR7 in DCs induces the apparition of dendritic protrusions, indicating that CCR7 can regulate the cytoarchitecture of these cells (15). CX3CR1 also induces transepithelial dendrites in intestinal lamina propia DCs, which enables these cells to sample luminal pathogens (20). In the case of CCR7, its ability to control DC cytoarchitecture is probably related to the capability of this receptor to regulate the organization of the actin cytoskeleton (Table I) (21, 22). Based on in vitro analysis and in vivo studies that directly observe the DCs in the lymph nodes, it has been suggested that the dendrites, by enhancing the net surface of the DCs, confer these cells a high capability to “capture” T cells in the lymph nodes (23, 24). This suggestion is supported by experimental data that show that reduction of the dendritic protrusions of the DCs results in the dampening of their Ag-presenting ability (25). Therefore, the CCR7-mediated induction of dendritic cytoplasmic extensions may contribute positively to the immune response by regulating the APC ability of the DCs.

It also has been reported that the stimulation of CCR7 with ligands CCL19 and CCL21 positively regulates the rate of endocytosis of the mature DCs (16). Stimulation of immature DCs with CCL3, the ligand of inflammatory chemokine receptors CCR1 and CCR5, also enhances the endocytic activity of these cells (16). As immature DCs are endocytic cells that sample the environment by Ag uptake (1, 2, 3), this finding implies that stimulation with CCL3, by increasing the endocytic activity, may enhance still further the sampling ability of the immature DCs. The stimulation of the endocytosis by CCR7 in the mature DCs is paradoxical because, as indicated above, DCs are known to reduce their endocytic ability when they differentiate into mature DCs, i.e., at the differentiation stage when they express CCR7 (1, 2, 3). However, it has also been shown that engagement of B7-DCs or CD40 in mature DCs also enhances the endocytosis in these cells (26). This finding suggest that B7-DCs, CD40 and CCR7, may belong to a group of surface molecules that can restore in the mature DCs the ability to uptake exogenous Ags, a feature typical of immature DCs. It remains to be investigated if CCR7 may up-regulate further the endocytosis in the CCR7-expressing semimature DCs.

Stimulation of CCR7 in mature DCs protects these cells from apoptosis caused by serum deprivation, suggesting that CCR7 may regulate survival (17). Before, it was also shown that the stimulation of the chemokine receptor CXCR4 may induce protection from IL-10-induced apoptosis in PDCs (27). In the case of CCR7, the protection that this receptor confers to the DCs is highly relevant for the immune response when considering the life cycle of these cells. DCs become apoptotic and die in the lymph nodes (28), implying that from the moment in which Ag-loaded DCs reach the lymph node until their final demise in these regions, the DCs have a limited span of time to encounter and activate a quiescent naive or memory T cell that displays a TCR that recognize its antigenic load. In this regard, the number of T cells that display a suitable TCR for a certain Ag peptide presented on a DC is very low (1 cell in 10⁵–10⁶ T cells). Therefore, predictably the mechanisms that delay the apoptosis and enhance the longevity of the DCs also increase the probability of the encounters of Ag-loaded DCs with cognate T cells, which would have a beneficial effect in the immune response. Confirming these predictions, it has been observed that factors that moderately enhance DC longevity (the increase cannot be excessive because it can lead to immunopathology) concomitantly increase the magnitude of the cellular immune response (29). In contrast, premature induction of DC death leads to the prevention of T cell activation and expansion (30). Therefore, the prosurvival effects that CCR7 may induce in the DCs that are either in route to or inside the lymph nodes may potentially contribute to the extended longevity of these cells and, therefore, to an improved immune response.

Stimulation of CCR7 also increases the migratory speed of the DCs, indicating that this receptor can regulate DC locomotion (18). This function is not exclusive of CCR7 because a variety of publications have shown previously, using checkerboard assays, that chemokines may often enhance the migratory speed of the DCs (12). In the case of CCR7, as between certain limits, there is a direct and positive relationship between adhesion and motility (31); a possibility is that CCR7 may regulate the speed of the DCs by inducing an increase in the adhesion of these cells to the substrate. However, and although CCR7 has been shown to up-regulate integrin activity and, consequently, adhesion in other cells types (see references in Table I), we have not been able to detect CCR7-dependent changes in the adhesion in the DCs in static adhesion experiments (18). It is still possible that the effects of CCR7 on migratory speed of DCs may be due to dynamic changes in the integrin avidity that could only be detected in adhesion assays under flow conditions (32). Another possibility is that CCR7 may affect the DC motility by regulating targets that are downstream of the integrins (18). It has also been suggested that the characteristic “dendritic” morphology of the DCs may also contribute to high motility of these cells (8). If this were the case, CCR7 may also contribute to regulate indirectly the motile phenotype of the DCs by inducing dendrite formation (8, 15). Predictably, the CCR7-dependent regulation of the migratory speed may operate in vivo in concert with chemotaxis to rapidly direct the DCs to the lymph nodes. In this regard, it has been reported that the number of DCs that reach the lymph nodes may positively affect the immune response (33). Therefore, the increase in DC motility (and survival (17)) controlled by CCR7 may increase the magnitude of the immune response by increasing the number of DCs that reach the nodes. The CCR7-induced stimulation of the migratory speed may also contribute to the rapid movement of the DCs inside the nodes and, consequently, to the high number of encounters between the DCs and the T cells in these regions. It has been calculated that every hour in the range of 500-5000 encounters can take place among DCs and T cells in the lymph nodes (23, 34). As the number of contacts rise, it also increases concomitantly the probability of productive interactions between Ag-loaded DCs and cognate T cells (24). In sum, the enhanced motility induced by stimulation of CCR7 may increase the number of DCs that enter the lymph node per unit of time and the number of encounters of DCs with the T cells in these regions, resulting in a positive effect on the immune response (24).

Finally, recently, it has been shown that stimulation of CCR7 enhances the mature phenotype of DCs, leading to the secretion of inflammatory cytokines, an increase in the levels of MHC and costimulatory molecules, and the potentiation of the ability of the DCs to activate naive T cells (19). Induction of maturation of the DCs was also earlier observed upon stimulation of CCR1, CCR3, and CCR5 receptors (35, 36). In the case of CCR7, it is important to note that the stimulation of this receptor caused the apparition of a mature phenotype only in the DCs that were licensed by a previous exposure to microbial stimuli because induction of maturation was not observed in immature DCs that express CCR7 following the stimulation of this receptor (19). This suggests that, following the exposure of the DCs to danger signals and the subsequent onset of maturation, CCR7 can further enhance this process (19). In the described study, where authors used the mouse as a model to analyze in vivo the effect of CCR7 on maturation (19), it was also found that, compared with wild-type animals, the maturation of DCs was reduced in plt mice (19). As plt mice are deficient in CCL19 and only express CCL21 in the lymphatic vessels, which allows stimulation of the DCs by this chemokine and some small migration of the DCs to the lymph nodes, the results indicate that both CCL19 and CCL21 are required to induce full maturation of the DCs (19, 37).

Some of the functions described for CCR7 in DCs have also been observed in other cells types (Table I). Concerning the regulation of the cytoarchitecture, it has been shown that stimulation of CCR7 induces filopodia formation in mesangial cells (38) and morphological polarization in T cells (39, 40). Stimulation of CCR7 has also been shown to protect mesangial cells and CD8 T cells from CD95- and etoposide-induced apoptosis, respectively (41, 42, 43). Moreover, CCR7 regulates the migratory speed of naive T cells (40) and human squamous cell carcinoma (44). Finally, CCR7 also modulates the differentiation of naive T cell to a Th1 phenotype (45).

With respect to functions that have not been described so far for DCs, CCR7 has been reported to regulate adhesion in human mesangial cells (38) and T cells (46, 47). At least in T cells, it has been shown that CCR7 may control adhesion by regulating the activity of different integrins (32, 48). CCR7 has also been described to regulate the proliferation in mesangial cells (42), T cells (45, 49), and hemopoietic precursor cells (21). Finally, recently, it has been shown that CCR7 regulates the invasive ability of carcinoma cells (50).

At least in the case of the chemotaxis, the expression of CCR7 is not sufficient to guarantee its functionality. This has been clearly observed in PDCs where CCR7 is not functional unless the cells receive an activation stimulus (51). During the last years, some of the signaling components of the molecular machinery that confers functionally to CCR7 in DCs have been identified. It has been shown that stimulation of PGE₂ receptors EP2 and EP4 is necessary for the function of CCR7 (52, 53). Moreover, it also has been shown that LPS-matured DCs only migrate toward the ligand CCL19 in the presence of the free radical gas NO (see below) (54). It is likely that both PGE₂ and NO may regulate the function of CCR7 in the inflammation sites where they are produced (55). Finally, CCR7 requires the expression of the ADP-ribosyl cyclase CD38 on the membrane to be functional in DCs (56). CD38 is a glycoprotein that catalyzes the conversion of NAD⁺ in cADP-ribose, an intracellular messenger that regulates the liberation of calcium from ryanodine receptor-regulated intracellular stores (56).

The reason(s) why PGE₂ and CD38 are required to make CCR7 functional is unclear; however, in the case of the NO, it has been indicated that its stimulatory effect on migration was mediated by the enzyme cGMP protein kinase (cGK) and its target, the vasodilator-stimulated protein (VASP) (54). In the absence of NO, cGK phosphorylates VASP, and this causes the disruption of the focal adhesions and the consequent inhibition of the migration of the DCs (54). However, in the presence of NO, cGK is inhibited, implying that the VASP-mediated disruption of the focal adhesions does not take place, and therefore, the DCs can migrate. Finally, it should be mentioned that the requirement for PGE₂ receptors and CD38 ectoenzyme has been observed only for CCR7-regulated chemotaxis and is not known whether other CCR7-dependent functions require also these surface molecules.

The chemokines CCL19 and CCL21 were found to be indistinguishable when comparing their binding affinities for CCR7 (57), their ability to regulate the survival (17), the chemotaxis and the migratory speed (18), and the maturation (19). However, at least under the conditions used in the experiments, in the case of the regulation of the cytoarchitecture (15), this function was observed only when the DCs were stimulated with CCL19 but not with CCL21 (Table I). In T cells, it has also been observed that CCL19, but not CCL21, induces desensitization of the receptor, i.e., inability of the receptor to respond to a second stimulus of the chemokine (58, 59). These differences in response of CCR7 to its ligands can be relevant in the immune system because, at least in the case of T cells, it has been observed that the generation of CTLs in mice was blunted by specific CCL19, but not CCL21, peptide antagonists (60). The results that indicate different responses in the DCs for CCL19 or CCL21 are in line with other reports that show that the knocking down of the leukotriene C (4) transporter multidrug-resistance protein 1 affects the signaling from CCR7 only when the DCs are stimulated with CCL19 but not with CCL21 (55).

The reason(s) that cause these differences in response to the stimulation of CCL19 and CCL21 in DCs and other cells has not been clarified. It is possible that these chemokines may induce a different conformation of CCR7 that can be reflected in a different signaling capability and consequent functional outcome (see below) (59). In this regard, and despite the mentioned equivalence concerning their affinity for the receptor and capability to induce chemotaxis in several cell types, it should be mentioned that CCL19 and CCL21 share only 32% aa identity. Furthermore, compared with CCL19, CCL21 includes some 30 additional aa with two extra cysteines in the carboxyl terminus (65). Also suggesting a different signaling capability for both chemokines, a recent report (59) has shown that stimulation of the CCR7-transfected HEK293 cells with CCL19 or CCL21 induces different phosphorylation of intracellular residues of the receptor and different signaling outcomes.

Little information is available on the transcriptional regulation of CCR7 (55). It has been described that CCR7 expression can be regulated positively by NF-κB and AP-1 transcriptional factors in Hodgkin’s disease-derived cell lines (66, 67). Moreover, the CCR7 promoter includes potential binding sites for NF-κB and AP-1, indicating that these transcriptional factors may bind directly to the promoter (66, 67). CCR7 expression is regulated negatively by the peroxisome proliferator-activated receptor (PPAR-γ) in monocyte-derived DCs, by the IFN consensus sequence binding protein, and by the Runx3 in murine bone marrow-derived DCs and by STAT4 in murine CD4 Th1 lymphocytes (68, 69, 70).

Like all G protein-coupled receptors (GPCRs), CCR7 may use G protein-dependent and -independent mechanisms to convey intracellular signals (see below) (71). G protein-dependent mechanisms involve the use of the α subunit and the βγ dimer of the heterotrimeric G proteins (72, 73, 74). Further versatility is introduced by the use of different α protein families and βγ components to relay the signals. In this regard, there are four families of α subunits, which include several members, namely families Gα_s, Gα_q, Gα_i, and Gα₁₂ (72, 73, 74). There are also 5 β and 12 γ members that have not been classified in families (74). The specific type of the α or βγ G proteins that CCR7 uses in DCs is limited by the selective expression of these molecules in these cells (A. L. Corbí and M. Vega, unpublished observation) (75). Finally, CCR7 may also use adapters β-arrestins (β-arrestin-1 and -2) to relay G protein-independent signals from CCR7 (59, 71).

We have found that three functions controlled by CCR7 in DCs, namely, survival, migratory speed, and chemotaxis, are regulated by three independent signaling modules (Fig. 1) (17, 18). We observed that chemotaxis was regulated by the pertussis toxin-sensitive Gi proteins and also by the three MAPK family members Erk1/2, p38, and JNK. Migratory speed was regulated by Rho/Pyk2/cofilin, a signaling axis that was independent of G_i and probably dependent on the G₁₂ family of α G proteins (data not shown) (Fig. 1) (18, 76). Finally, we found that CCR7 uses G_i, βγ subunits of the heterotrimeric G-proteins, the pair PI3K/Akt, and the transcription factor NF-κB to regulate CCR7-dependent survival in the DCs (17) (Fig. 1). We found that interference with the molecular components, different from G_i, of any single one of the three modules identified, by using selective inhibitors or dominant-negative constructs, did not perturb in the other two modules neither the activity of the corresponding signaling components nor the functions under the control of these modules (Fig. 1 and Refs.17 and 18). These results point out the independence of the signaling/functional modules identified. Concerning G_i, this is a signaling component shared by the signaling/functional modules that control survival and chemotaxis but not by the module that control migratory speed (Fig. 1). In line with our data, other authors have also indicated that CCR7-regulated endocytosis is controlled by a set of molecules that are independent of those that regulate chemotaxis (62). Although it remains to be investigated if signaling modules are also used to regulate other CCR7-dependent tasks in DCs, such as endocytosis, cytoarchitecture, and maturation, however, our results suggest that CCR7 may potentially use independent modules to control different functions (Fig. 1) (17, 18). These data are also in keeping with recent reports that indicate that the constitution of independent signaling modules may be a common strategy to control different cell functions (77).

The results reviewed in this article indicate that CCR7 may contribute to modulate the immune response through the control of the functions of the DCs. In addition, it is suggested that the formation of independent signaling modules may be a strategy that may allow CCR7 to regulate its several functions in the DCs. Several issues will be relevant in future studies on CCR7 in these cells. Briefly, among other goals, it would be important to analyze the factors that regulate CCR7 expression and function. Furthermore, the identification of the whole gamut of functions under the control of CCR7 in DCs and other cells will be important. In this regard, it would be interesting to investigate if some of the functions described for CCR7 in mature DCs are also observed in other cell types that express this receptor, including immature DCs subtypes, PDCs, and other leukocytes. With the exception of the induction of maturation (19), the new functions described for CCR7 have been described in in vitro studies (Table I). Although some of these functions are difficult to investigate in vivo, further analysis will be required to confirm their relevance for the immune response.

Scanty information is available on the signaling capabilities of CCR7 in different systems (Table I). Therefore, an important goal will be the identification of the signaling components that regulate different CCR7-dependent functions in DCs and the contribution of different GPCR-protein molecular machinery in the process (72, 73, 74). The existence of independent signaling modules to control different CCR7-dependent functions offers the possibility of intervening selectively in these functions. Another important question is to explain how these signaling/functional modules may maintain their independence. Although at the moment it can only be speculated on potential mechanisms, the formation of scaffolds is an attractive possibility. Scaffolding formation has emerged as a common strategy that allows diversion of signals into different pathways (78). In the case of CCR7, the formation of scaffolding complexes near the receptor may potentially contribute to segregate the signals to regulate specific functions. Signaling molecules associated to GPCRs, including β-arrestins, may be involved in the formation of these scaffolds (71). Proteomic technology will be invaluable to investigate the possible formation of complexes associated to CCR7. In sum, the knowledge on the functional capabilities of CCR7 and mechanisms whereby this receptor regulates these functions in DCs will be invaluable for a better understanding of the immune response under normal and pathological conditions. As with the Sufi story at the beginning of this review, at the present there are more questions than answers regarding CCR7. More answers will have to be provided in the next years to understand the role of CCR7 in DCs and other cells of the immune system.

We sincerely appreciate the support of Prof. Angel L. Corbí. We acknowledge two anonymous referees for their most useful comments.