The chemokine receptor CCR7, together with its ligands CCL19 and CCL21, is responsible for the correct homing and trafficking of dendritic cells and lymphocytes to secondary lymphoid tissues. Moreover, cancer cells can utilize CCR7 to metastasize to draining lymph nodes. However, information on CCR7 signaling leading to cell migration or receptor trafficking is sparse. Using novel CCR7 deletion mutants with successive truncations of the intracellular C-terminus and a mutant with impaired G-protein coupling, we identified distinct motifs responsible for various aspects of CCR7 signal transduction. Deleting a Ser/Thr motif at the tip of the intracellular tail of CCR7 resulted in an impaired chemokine-mediated activation of Erk1/2 kinases. Interestingly, deleting an additional adjacent motif restored the ability of CCL19-mediated Erk1/2 phosphorylation, suggesting the presence of a regulatory motif. Both the Ser/Thr and the regulatory motif are dispensable for signaling events leading to cell migration and receptor trafficking. A CCR7 mutant lacking virtually the complete C-terminus readily bound CCL19 and was internalized, but was unable to activate the G protein and to transmit signals required for cell migration, mobilization of [Ca2+]i and Erk1/2 activation. Finally, G-protein coupling was critical for [Ca2+]i mobilization, Erk1/2 phosphorylation and chemotaxis, but not for CCR7 trafficking.
Chemokine receptors are seven-transmembrane-domain G-protein-coupled receptors (GPCRs) that are widely expressed on many cell types of the body and responsible for guiding migrating cells along a chemokine gradient. A highly regulated network of chemokine receptors and chemokines is responsible for directing inflammatory and homeostatic cell trafficking (Moser et al., 2004). Homing of antigen-bearing dendritic cells to secondary lymphoid organs for instance, depends on the chemokine receptor CCR7. In the T-cell zone of lymphoid organs, where the two ligands of CCR7, CCL19/ELC and CCL21/SLC are expressed (Luther et al., 2000), incoming dendritic cells present their peptide antigens to naive T cells, thereby inducing an immune response. In addition to its expression on mature dendritic cells, CCR7 is also expressed on subpopulations of T cells, B cells, NK cells and thymocytes (Campbell et al., 2003; Muller et al., 2003). Homing of dendritic cells and naive T cells is strongly affected by the lack of either CCR7 or its ligands, and consequently results in a defect in initiating primary immune responses (Forster et al., 1999; Gunn et al., 1999; Luther et al., 2000; Ohl et al., 2004). Moreover, CCR7 is also expressed in breast, gastric, non-small-cell lung, and oesophageal squamous cancer, as well as in chronic lymphocytic leukemia (reviewed by Balkwill, 2004; Zlotnik, 2006). CCR7-expressing cancer cells respond to CCL19 and CCL21 and metastasize to the draining lymph nodes and the lung (Muller et al., 2001). Consequently, understanding how chemokine receptors transmit signals resulting in cell migration and identifying ways of modulating chemotaxis or receptor silencing is of interest for future therapeutic strategies.
Chemokine receptor signaling is initiated by activation of the Bordetella pertussis toxin-sensitive Gi family of heterotrimeric G proteins (Thelen, 2001). Chemokine-bound receptors trigger the Gα subunits to exchange GTP for GDP, leading to the dissociation of the Gα subunit from the βγ heterodimer. Whereas the Gα subunit itself appears not to be required for chemotaxis, the released Gβγ subunit transduces signals leading to cell locomotion (Neptune and Bourne, 1997). Gβγ activates phospholipase Cβ (PLCβ) isoforms leading to the formation of inositol (1,4,5)-trisphosphate and a transient rise in the concentration of intracellular free calcium. Gβγ also activates phosphoinositide 3-kinases (PI3Ks) and downstream pleckstrin-homology-domain-containing proteins, such as protein kinase B (PKB)/Akt or guanine-nucleotide exchange factors and mitogen-activated protein kinases (MAPKs) (Thelen, 2001; Wymann et al., 2003). Despite its well-characterized biological function, CCR7 signaling has not yet been characterized in depth. So far, CCR7 triggering has been shown to induce receptor phosphorylation, β-arrestin (official symbol ARRB1) recruitment, phosphorylation of PKB and extracellular regulated kinases 1 and 2 (Erk1 and Erk2; official symbols MAPK3 and MAPK1) and to mobilize intracellular calcium (Bardi et al., 2001; Kohout et al., 2004; Otero et al., 2006; Riol-Blanco et al., 2005; Sanchez-Sanchez et al., 2004; Scandella et al., 2004; Willimann et al., 1998). Of note, despite similar CCR7 binding affinities, G-protein activation and chemotactic properties of CCL19 and CCL21 (Willimann et al., 1998), CCL19 induces stronger receptor phosphorylation and β-arrestin recruitment than CCL21 (Kohout et al., 2004). Moreover, CCL19, unlike CCL21, induces rapid endocytosis of CCR7 (Bardi et al., 2001; Otero et al., 2006). We have shown recently, that CCR7 is internalized through clathrin-coated pits together with CCL19 and transported to early endosomes (Otero et al., 2006). Subsequently, CCR7 recycles back to the plasma membrane to participate again in chemokine gradient sensing, whereas CCL19 is sorted to lysosomes for degradation (Otero et al., 2006). The molecular mechanism underlying this observation has not yet been elucidated.
Crucial motifs of chemokine receptors involved in migration and endocytosis have been mapped to the intracellular loops and the C-terminus. For instance, the Asp-Arg-Tyr (DRY) motif in the second intracellular loop is generally believed to be responsible for G-protein activation. Whereas G-protein activation is essential for chemotaxis, it is still a matter of debate whether and how G-protein activation is involved in GPCR endocytosis (Vilardaga et al., 2001). For GPCR internalization, two main consensus signals within the C-terminus have been described (Bonifacino and Traub, 2003; Neel et al., 2005). A tyrosine-based (Yxxϕ or NPxY) and a di-Leu/Ile-Leu/Leu-Ile motif have been identified which interact with the clathrin adaptor protein AP2. The di-Leu motif is essential for internalization and trafficking of many chemokine receptors, including CXCR2, CXCR3, CXCR4 and CCR5 (reviewed by Neel et al., 2005). However, CCR7 does not contain such known endocytosis motifs and domains responsible for delivering different signals remain to be identified.
In the present study, we searched for distinct motifs within the chemokine receptor CCR7 that regulate signal transduction, receptor trafficking, and cell migration. To this end, we generated three C-terminal truncation mutants of CCR7, where the intracellular tail was gradually removed. In addition, we also mutated the conserved DRY motif in the second intracellular loop of the receptor. Using this strategy, we identified segregated motifs in CCR7 that are responsible for distinct aspects of CCR7 signal transduction, such as ligand induced mobilization of intracellular calcium, Erk phosphorylation, cell migration and receptor trafficking.
CCR7 and its mutants are expressed to a similar degree at the plasma membrane
In order to identify distinct motifs within the chemokine receptor CCR7 that regulate signal transduction, receptor trafficking, and cell migration, we generated three truncation mutants of CCR7 lacking parts or the entire C-terminal tail and a mutant with a modified DRY motif in the second intracellular loop of the receptor (Fig. 1A). The cytoplasmic C-terminal part of CCR7 was defined in accordance with the TOPO_DOM prediction by the UniProtKB/Swiss-Prot (www.expasy.org/uniprot) entry for human CCR7 (accession number P32248). In the mutant MT1 (residues 1-334), all except the three most proximal amino acids of the predicted C-terminal tail were removed, and Lys was replaced by Arg to prevent a putative modification by ubiquitin. The last 34 amino acids were removed in the MT2 (residues 1-345) mutant, whereas in MT3 (residues 1-355) only the last 24 amino acids were deleted (Fig. 1A). Within these last 24 amino acids of CCR7 (residues 365-378), threonine and serine clusters were reported to be phosphorylated after ligand binding (Kohout et al., 2004), indicating a possible role in receptor signaling and/or trafficking for these amino acids. These three constructs were used to establish stable cloned cell lines in HEK293 cells and in pre-B 300-19 cells that both do not express endogenous CCR7. Moreover, to analyze the role of the highly conserved DRY motif of CCR7 (residues 153-155) with respect to receptor trafficking and signaling we created a mutated CCR7 receptor where we replaced the DRY sequence by DNY. This single mutation has been reported to disrupt the G-protein coupling of the chemokine receptors CCR5 and CXCR4 (Berchiche et al., 2007; Lagane et al., 2005). The DNY construct was transfected into 300-19 cells and stable cell clones were established. Cell surface staining of CCR7 with a specific antibody was assessed by flow cytometry (Fig. 1B and Fig. 2) and confocal microscopy (data not shown), which revealed that all CCR7 mutants are expressed at the plasma membrane of stably transfected cell lines. In addition, we also quantified the ratio of intracellular compared with extracellular protein by flow cytometry and found no difference between wild-type CCR7 and its mutants (data not shown), indicating that all mutants were not trapped within the ER or heavily misfolded. These results provide clear evidence that the C-terminus of CCR7 is not required for proper insertion of the receptor into the plasma membrane.
CCR7 and its mutants equally bind its ligand CCL19
To investigate whether all four CCR7 mutants are able to interact with their ligand, we performed a binding assay with CCL19. Therefore, 300-19 cells stably transfected with wild-type receptor or the CCR7 mutants were incubated with biotinylated CCL19 and FITC-labeled streptavidin. In parallel, CCR7 surface expression was measured by flow cytometry using a CCR7-specific antibody. As shown in Fig. 2, all CCR7 mutants were able to bind CCL19, demonstrating that neither the C-terminal tail of the receptor nor the point mutation of the DRY motif is required for chemokine binding.
The membrane proximal part of the C-terminal tail of CCR7 and the DRY motif are crucial for cell migration
Next, we investigated whether motifs in the C-terminal tail of CCR7 or the DRY motif are required for signaling leading to cell migration. To this end, we tested the chemotactic abilities of 300-19 cells stably expressing CCR7 or mutants thereof in Transwell™ chemotaxis assays. CCR7-expressing cells readily migrated in a dose-dependent manner in response to both CCL19 and CCL21 as expected (Fig. 3). Cells expressing truncated CCR7 that lacks parts of the intracellular tail (MT2 and MT3), migrated towards CCL19 and CCL21, comparable with 300-19 cells that express wild-type CCR7. This somewhat unexpected finding indicates that the Ser/Thr containing motifs (Ser356-Ser357, Ser364-Ser365 and Thr372-Thr373-Thr374-Thr375) at the end of the C-terminus, which were postulated as putative targets for GRK-mediated phosphorylation (Kohout et al., 2004), are not critical for eliciting signals leading to cell migration. In marked contrast, MT1, which lacks the entire intracellular tail, was unable to induce a chemotactic response to CCL19 and CCL21 (Fig. 3). This points to an important role of the proximal part of the cytoplasmic CCR7 C-terminus (residues 335-344) in signal transduction of chemotactic processes. As expected, 300-19 cells stably transfected with the DNY mutant were unable to migrate, confirming the key role of G-protein coupling in the migration process (Fig. 3).
Chemokine-mediated mobilization of intracellular calcium depends on the membrane proximal part of the C terminus and the DRY motif of CCR7
The mobilization of intracellular calcium is an early event after chemokine binding to its receptors. To examine whether CCR7 mutants are able to release Ca2+ from intracellular stores upon ligand binding, cells were loaded with Fluo-3/AM and chemokine-mediated [Ca2+]i changes were monitored by flow cytometry. MT2 and MT3 both mobilized Ca2+ in response to CCL19 and CCL21 at levels comparable with wild-type CCR7 (Fig. 4). By contrast, CCL19 and CCL21 were unable to induce changes in [Ca2+]i in cells expressing the CCR7 mutants MT1 and DNY (Fig. 4). All cell lines responded to ionomycin, indicating comparable dye loading of the cells (Fig. 4). Similarly to their role in cell migration, the DRY motif and residues 335-345 of CCR7 are indispensable for calcium signaling.
Chemokine-mediated triggering of the CCR7 mutants MT1 and DNY does not lead to Erk1/2 activation
Signal transduction by chemokine receptors also results in a transient activation of the MAPKs Erk1 and Erk2. Thus, we stimulated HEK293-CCR7 transfectants with CCL19 for 5 minutes and analyzed the phosphorylation of Erk1/2 by western blot analysis using an antibody recognizing phosphorylation at positions T202 and Y204. As shown in Fig. 5A, activation of Erk1/2 after CCL19 stimulation was readily detected in wild-type CCR7 and MT2 transfectants. Surprisingly, triggering of MT3 led only to a weak, but reproducible, activation of Erk1/2 (Fig. 5A). The CCR7 mutant lacking the entire intracellular tail (MT1), however, was unable to induce Erk1/2 phosphorylation upon ligand binding.
To address whether Erk1/2 phosphorylation after CCR7 activation is dependent on G-protein signaling, we incubated HEK293 cells stably expressing wild-type CCR7 with Bordetella pertussis toxin (PTx), which ADP-ribosylates the α subunit of Gi proteins and thus blocks their inhibitory function on adenylyl cyclase. As shown in Fig. 5B, CCL19-induced Erk1/2 activation was inhibited by PTx, indicating that G-protein coupling is required for this process. This was confirmed in the DNY mutant, which also failed to induce Erk1/2 activation (Fig. 5C).
The CCR7 C-terminus and the DRY motif are required for G-protein coupling
To determine whether the CCR7 mutants couple to G proteins, membrane preparations of 300-19 cell lines stably expressing wild-type CCR7 or mutants thereof were stimulated with CCL19 and the GDP-GTP exchange determined by a GTPγS assay. As expected, the DNY mutant was unable to activate G proteins whereas wild-type CCR7 led to a significant CCL19-dependent GTPγS binding (Fig. 6). Similarly, both MT2 and MT3 also stimulated GTPγS binding in a ligand-dependent manner. Again, CCL19 binding to MT1 did not lead to G-protein activation. Comparable results were obtained with HEK293 transfectants (data not shown). Interestingly, in both cell lines, G-protein binding of MT2 was slightly more efficient than of MT3, which might suggest that the amino acids situated between positions 345-355 could act as a putative repression signal. This result fits nicely with the more efficient chemokine-mediated Erk1/2 activation by MT2 compared with MT3.
The intracellular tail and G-protein activation are dispensable for CCR7 trafficking
Finally, we investigated whether G-protein activation and the membrane proximal part of the C-terminus of CCR7 are also required for receptor trafficking. To this end, 300-19 cells stably expressing CCR7 deletion mutants were incubated with CCL19 for 30 minutes at 37°C followed by determining cell surface expression of CCR7 with a specific antibody. Surprisingly, triggering of CCL19 provoked receptor endocytosis of all CCR7 C-terminal deletion mutants (Fig. 7A). Next, we measured recycling of internalized CCR7 back to the plasma membrane. For this, CCR7 transfectants were stimulated for 30 minutes with CCL19, washed extensively to remove unbound CCL19 followed by incubation for 1 hour in the absence of chemokines to permit receptor recycling. CCR7 recycling was comparable between wild-type CCR7 and the C-terminal deletion mutants (Fig. 7A), providing clear evidence that motifs within the intracellular tail of CCR7 are dispensable for CCR7 trafficking. Although CCL21 induces only marginal CCR7 endocytosis (Bardi et al., 2001; Otero et al., 2006), we found no differences in endocytosis between CCR7 and its deletion mutants upon CCL21 triggering (data not shown). Moreover, similar results were obtained when CCR7 mutants were expressed in 300-19 and HEK293 cells (data not shown). To address whether endocytosis and recycling of CCR7 depend on G-protein activation, we incubated 300-19 cells expressing wild-type CCR7 with PTx. Gαi inhibition influenced neither CCL19-induced CCR7 internalization nor recycling (Fig. 7B). Studies on the DNY mutant confirmed that CCR7 internalization and recycling were independent of G-protein activation (Fig. 7C).
Previous studies showed that phorbol esters, such as PMA, can downmodulate chemokine receptors such as CXCR4 and CCR5 in the absence of ligands by receptor phosphorylation through second messenger kinases (Oppermann et al., 1999; Signoret et al., 1997). Kohout and co-workers demonstrated that the second messenger kinase PKC, but not PKA, induces CCR7 phosphorylation (Kohout et al., 2004). To investigate the role of PKC in CCR7 internalization, we stimulated 300-19 cells expressing CCR7 with PMA for 30 minutes and determined cell surface expression of CCR7 by flow cytometry. As depicted in Fig. 8 PMA stimulation slightly mediated CCR7 downmodulation, but was much less efficient than CCL19. Moreover, CCL19-mediated CCR7 endocytosis was only marginally affected in cells pretreated with the PKC inhibitor bisindolylmaleimide (Bim) (Fig. 8). Treating cells with Bim alone or Bim together with PMA did not diminish CCR7 expression. These data indicate that PKC is involved in CCR7 trafficking but seems to play a minor role. Moreover, these results are in line with our finding that deleting the C-terminus of CCR7 comprising putative PKC and PKA phosphorylation sites (SS356/7/SS364/5) slightly affected CCL19-mediated internalization (Fig. 7).
In conclusion, we identified distinct motifs within the intracellular domains of CCR7 that are critical for at least two different signaling pathways. Receptor internalization and recycling do not require G-protein coupling and are independent of motifs located within the C-terminus of CCR7 (summarized in Table 1). By contrast, signal transduction triggered by CCR7 leading to cell migration, changes in [Ca2+]i and Erk1/2 activation require both G-protein activation and a motif within the intracellular tail of CCR7. In fact, we identified a novel motif within the C-terminus juxtaposed to the seventh transmembrane domain of CCR7 that is critical for G-protein activation, and hence cell migration. Moreover, we discovered that the tail of CCR7 might include a regulation motif located within residues 346-355, because CCR7 lacking the tip of the tail (residues 356-378; MT3) has an impaired capacity to activate Erk1/2, which is restored if the adjacent amino acids (residues 346-355; MT2) are also deleted. Moreover, our data indicate that Erk1/2 phosphorylation is not mandatory for cell migration, because MT3 mediated normal cell migration in response to CCL19 but had little effect on Erk1/2 phosphorylation.
The pivotal function of CCR7 in the migration of T lymphocytes and mature dendritic cells to the T-cell zones of secondary lymphoid organs has been extensively characterized throughout recent years, but the motifs required for migration and trafficking have barely been investigated. In particular, the C-termini of chemokine receptors are poorly conserved and may account for major differences in trafficking and the intracellular fates of chemokine receptors (Neel et al., 2005). For some chemokine receptors, including CCR3 (Sabroe et al., 2005), CCR5 (Kraft et al., 2001), CXCR1, CXCR2 (Richardson et al., 2003; Sai et al., 2004), CXCR3 (Dagan-Berger et al., 2006) and CXCR4 (Doranz et al., 1999; Haribabu et al., 1997; Roland et al., 2003), successive deletions of the C-terminal cytoplasmic tails have been analyzed, but comparable information on CCR7 is lacking. Considering that these receptors possess different internalization motifs and behave differently upon deletional analysis, it is pertinent to identify motifs in CCR7 required for endocytosis, signaling and migration through systematic truncation analysis. Kohout et al. replaced five serine and four threonine residues with alanine in the distal C-terminal region of CCR7 and showed that these point mutations strongly reduced CCL19-induced receptor phosphorylation and Erk1/2 activation (Kohout et al., 2004). However, neither migration nor receptor trafficking was analyzed in this study.
Here, we demonstrate that truncation of the C-terminal tail has no adverse effect on CCR7 cell surface expression, ligand binding or receptor trafficking. In fact, the CCR7 mutant MT1, which lacks the whole C-terminal region, could easily reach the plasma membrane and underwent endocytosis and recycling almost as efficiently as wild-type CCR7 (Fig. 7). This finding is remarkable because most chemokine receptors show impaired internalization when their C-terminus is removed, as demonstrated for CCR3 (Sabroe et al., 2005), CXCR4 (Roland et al., 2003), CXCR3 (Dagan-Berger et al., 2006), CXCR1 and CXCR2 (Richardson et al., 2003). Interestingly, the C-terminus of CXCR3, particularly the serine clusters therein, was shown to be involved in CXCL9- and CXCL10-mediated receptor internalization, whereas CXCL11-mediated CXCR3 endocytosis depends on a motif within the third intracellular loop of the receptor (Colvin et al., 2004). Serial C-terminal truncation of CCR5 resulted in progressive loss of cell surface expression (Venkatesan et al., 2001) or impaired ligand-induced internalization (Kraft et al., 2001). Despite the normal trafficking of MT1 in HEK293 and 300-19 cell transfectants, CCR7 lacking its whole C-terminus was not able to transmit signals leading to cell migration (Fig. 3). A previous study showed that a cluster of Ser/Thr at the tip of the tail of CCR7 is phosphorylated upon ligand binding, suggesting that this region may be important for CCR7 internalization (Kohout et al., 2004). Our MT2 mutant lacks all possible phosphorylation sites of the C-terminus, whereas in MT3 the above-mentioned Ser/Thr motifs were deleted, leaving one putative phosphorylation site (Ser348). CCL19-induced receptor endocytosis and recycling occurred in both MT2 and MT3 expressing cells, and even in cells expressing the tailless mutant MT1, providing clear evidence that the intracellular tail of CCR7 does not include major signaling motifs for receptor trafficking. Moreover, CCR7 trafficking does not require G-protein coupling, as inhibiting G-protein activation by either PTx treatment or by the DRY to DNY mutation, did not abrogate receptor internalization or recycling (Fig. 7). In this respect, CCR7 behaves like other chemokine receptors as CXCR3, CXCR4 and CCR5 internalization for instance does not depend on G-protein activation (Amara et al., 1997; Colvin et al., 2004; Lagane et al., 2005) Noteworthy, deletion of the Ser/Thr motifs (MT3) at the C-terminus significantly impaired chemokine-mediated Erk1/2 activation (Fig. 5). This finding is in agreement with the previous observation that replacing Ser356-Ser357, Ser364-Ser365, Ser378 and Thr373-Thr374-Thr375-Thr376 with Ala resulted in diminished Erk1/2 phosphorylation (Kohout et al., 2004). Interestingly, this phenotype is completely recovered if another ten amino acids are deleted. CCL19-induced Erk1/2 phosphorylation is completely restored in the MT2 mutant, indicating that residues 346-355 may contain a novel regulatory motif. Moreover, CCR7 signaling mediated by MT2 and MT3 also suggests that Erk1/2 activation is not required for migration or for inducing changes in [Ca2+]i. The removal of potential phosphorylation sites within the C-terminus of CCR7 in the mutants MT2 and MT3 did not impair chemotaxis (Fig. 3), indicating that the Ser356-Ser357, Ser364-Ser365 and Thr372-Thr373-Thr374-Thr375 putative targets for GRK phosphorylation (Kohout et al., 2004) are not required for chemotactic signal transduction.
Truncation of the C-terminus of CCR7 led to the discovery of other exciting findings. Deletion of the entire C-terminus of CCR7 resulted in the loss of chemotactic activity (Fig. 3). Chemotaxis depends on the activation of G proteins, and chemokine-mediated G-protein activation is believed to exclusively depend on the highly conserved DRY motif within the second intracellular loop. Indeed, modifying the DRY motif completely abolished CCL19-mediated G-protein activation (Fig. 6), cell migration (Fig. 3), [Ca2+]i mobilization (Fig. 4) and Erk1/2 phosphorylation (Fig. 5), but did not affect receptor trafficking (Fig. 7). As the impaired migration of MT1 is due to the inability to activate the G protein (Fig. 6), it is tempting to speculate that the C-terminus contains an additional unidentified second motif critical for G-protein activation. This G-protein activation motif of CCR7 must be located within residues 335-345 (sequence: NDIFKLFKDLG), because MT2 readily activated the G protein upon chemokine binding. To our knowledge, such a G-protein activation motif located in the C-terminus of a chemokine receptor has not been identified. However, removing the C-terminus of CXCR4 resulted in an even higher G-protein activation in response to CXCL12 compared with cells expressing wild-type CXCR4 (Haribabu et al., 1997).
So far, we have been unable to identify a motif responsible for CCR7 trafficking. We demonstrate that neither G-protein coupling nor the intracellular tail contribute to CCR7 trafficking. Most commonly, tyrosine- or leucine-based internalization motifs can target membrane receptors to clathrin-coated pits. The C-terminal tail of CXCR2, CXCR4 and CCR5 for instance contains di-Leu motifs (LLKIL, LKIL and LL, respectively) which are determinants for receptor endocytosis upon ligand binding (Fan et al., 2001; Kraft et al., 2001; Orsini et al., 1999). The C-terminal tail of CCR7 contains five leucine residues, but they are not arranged in pairs and hence do not represent a bona fide internalization motif. Moreover, MT1 was efficiently endocytosed upon ligand binding although none of these leucines was present (Fig. 7). Ubiquitylation of the C-terminus also seems not to be the main driving force for CCR7 endocytosis, because in MT1 the remaining lysine was replaced by arginine. Thus, further studies on the intracellular loops are required to identify the motif involved in CCR7 trafficking.
In summary, the present study reveals that endocytosis and recycling signals are not contained within the C-terminus of CCR7, whereas signal transduction leading to cell migration relys on a ten residue motif adjacent to the seventh transmembrane domain and on G-protein activation. Deleting a Ser/Thr motif at the distal part of the tail of CCR7 resulted in an impaired chemokine-mediated activation of Erk1/2 kinases. Interestingly, deleting an additional adjacent motif restored the ability of CCL19-mediated Erk1/2 phosphorylation, indicating the presence of a regulatory motif. Both the Ser/Thr motif and the regulatory motif are dispensable for chemotactic signals. Moreover, we identified a motif in the C-terminus of CCR7 juxtaposed to the plasma membrane that is critical for G-protein activation and cell migration.
Materials and Methods
Antibodies and reagents
Antibodies were obtained from the following sources: FITC-coupled mouse anti-human CCR7 (clone 150503; R&D Systems Inc., Minneapolis, MN), mouse anti-HA, (Sigma, Buchs, Switzerland), streptavidin-FITC (Jackson ImmunoResearch Laboratories), phospho-Erk-1/2, total Erk-2 (Santa Cruz Biotechnology Inc, Santa Cruz, CA), and HRP-conjugated secondary anti-mouse and anti-rabbit antibodies (DAKO, Hamburg, Germany). Similar results were obtained using rat anti-human CCR7 (clone 3D12, BD Biosciences PharMingen, San Diego, CA) (data not shown). Recombinant human chemokines CCL19 and CCL21 were purchased from PromoCell (Heidelberg, Germany). Bordetella pertussis toxin was purchased from Calbiochem (La Jolla, CA). Ionomycin, bisindolylmaleimide (Bim) and PMA were from Sigma. [35S]GTPγS was from Amersham Biosciences (Little Chalfont, Buckinghamshire, UK). Monobiotinylated CCL19 was from RMF Dictagene SA (Epalinges, Switzerland).
Construction of expression plasmids
The cloning of pcDNA3-CCR7-HA has been described (Otero et al., 2006). C-terminal deletion mutants of CCR7, MT1 (aa 1-334), MT2 (aa 1-345) and MT3 (1-355) were generated by PCR using full-length CCR7 (aa 1-378) as template. PCR primers are as follows: CCR7-MTse 5′-ATA AAG CTT CGT CAT GGA CCT GGG G-3′ (HindIII restriction site underlined), CCR7-MT1as 5′-ATA GAA TTC CGC GGA ACC TGA CGC C-3′ (EcoRI restriction site underlined), CCR7-MT2as (5′-ATA GAA TTC CGC CCA GGT CCT TGA AG-3′) and CCR7-MT3as (5′-ATA GAA TTC CCC CAC TGC CGG AGC TG-3′). PCR fragments were cloned into pcDNA3-HA, revealing C-terminal-tagged CCR7 mutants. In CCR7-MT1, a K332R substitution was introduced. The CCR7-DNY mutant was generated by site-directed mutagenesis using pcDNA3-CCR7-HA as template. Briefly, a first PCR was performed with the primer CCR7se2 (5′-ATA GAA TTC CGT CAT GGA CCT GGG GAA AC-3′; EcoRI site underlined) and the primer DRYas (5′-GGC CAC GTA GTT GTC AAT GCT GAT-3′; modification underlined). A second PCR was performed with the primers DRYse (5′-ATC AGC ATT GAC AAC TAC GTG GCC-3′; modification underlined) and CCR7as (5′-TAT GCG GCC GCT GGG GAG AAG GTG GTG-3′; NotI site underlined). Then, both PCR products were mixed and a third PCR was performed with the primers CCR7as and CCR7se2 and the amplification product was cloned into pcDNA3-HA. HA-tagged CCR7 behaves like wild-type CCR7 in terms of chemotaxis, Erk activation, Calcium mobilization and receptor trafficking (Otero et al., 2006).
Cell lines and transfection
The human embryonic kidney cell line HEK293 was grown in DMEM (Invitrogen, Basel, Switzerland) supplemented with 10% (v/v) fetal bovine serum. HEK293 cells were stably transfected with FuGENE6 (Roche, Basel, Switzerland). Cell clones were established by limiting dilution in the presence of 0.8 mg/ml of G-418 (Invitrogen). The murine pre-B cell line 300-19 (Legler et al., 1998; Willimann et al., 1998) was grown in RPMI1640 (Invitrogen) with 10% (v/v) fetal bovine serum, 50 μM β-mercaptoethanol and 2 mM nonessential amino acids. Transfection was performed by electroporation and stable cell lines were generated by limiting dilution as described (Legler et al., 1998; Willimann et al., 1998).
CCL19-bio binding assay
CCR7-transfected 300-19 cells were incubated with monobiotinylated CCL19 (1 μg/ml) for 20 minutes at 4°C. Cells were extensively washed with PBS, 2% FBS, 5 mM EDTA and incubated with streptavidin-FITC (1 μg/ml) for 20 minutes, washed and fluorescence was monitored by flow cytometry (FACScan II™ or LSR II™) using CellQuest™ or FACSDiva™ software (BD Biosciences). Data were analyzed with the FlowJo™ software (Tree Star, San Carlos, CA).
Cell migration (1×105 cells in 100 μl) in response to graded concentrations of CCL19 and CCL21 was measured in 24-well Transwell™ chambers (Corning Costar, Cambridge, MA) through a polycarbonate filter of 5 μm pore size. After 3 hours of incubation at 37°C, a 600 μl aliquot of cells that migrated to the bottom chamber was counted by flow cytometry acquiring events for a fixed time period of 60 seconds and the number of migrated cells were expressed as percent of input cells.
Cells (2×106) were lysed with 0.5 ml of 1% Triton X-100 in 150 mM NaCl, 50 mM HEPES, 0.1 M EGTA, 2 mM MgCl2 and 10% glycerol. Proteins from total cell lysates were resolved by SDS-PAGE and transferred to Protran nitrocellulose membranes (Schleicher & Schuell, Dassel, Germany). Membranes were blocked with PBS containing 5% low-fat dry milk and incubated with Erk-specific antibodies overnight at 4°C (anti-phospho-Erk-1/2) or for 1 hour at room temperature (anti-total Erk-2) on a rocking plate. After washing, HRP-conjugated secondary antibodies were detected using enhanced chemiluminescence (Pierce/Socochim, Lausanne, Switzerland).
Analysis of chemokine-mediated changes in intracellular free Ca2+ concentrations
Cells (106/ml in 145 mM NaCl, 5 mM KCl, 1 mM Na2HPO4, 1 mM MgCl2, 5 mM glucose, 1 mM CaCl2 and 10 mM HEPES, pH 7.5) were loaded with 1.5 μl/ml Fluo-3/AM (4 mM in DMSO) for 30 minutes at 37°C. Samples were divided into 500 μl aliquots and changes in intracellular calcium concentrations in response to 1 μg/ml of chemokines or 0.02 μg/ml of ionomycin were acquired by FACS analysis. Changes in the fluorescence of Fluo-3 were recorded by flow cytometry for a period of 150 seconds.
To investigate receptor internalization, 300-19 transfectants in suspension (106/ml) or HEK293 cells derived from a 90% confluent six-well plates, respectively, were stimulated with 2 μg/ml chemokine or 1 μM PMA for 30 minutes at 37°C. Where indicated, cells were pre-treated with 10 μM Bim for 30 minutes at 37°C as described (Oppermann et al., 1999). Cell surface expression of CCR7 after ligand-mediated endocytosis was measured by flow cytometry using a CCR7-specific antibody. For recycling experiments, internalized receptor (2 μg/ml of chemokine for 30 minutes at 37°C) was allowed to recycle back to the plasma membrane by washing off unbound chemokines twice with warm PBS following incubation in chemokine-free medium for 1 hour at 37°C. Surface expression of CCR7 was thereafter assessed by flow cytometry.
Pertussis toxin treatment
300-19 cells (106/ml) or HEK293 cells (one well of a six-well plate) expressing CCR7 were preincubated with 100 ng/ml of PTx for 2 hours at 37°C prior to the respective assay. PTx was maintained in the medium throughout the experiments to ensure complete inhibition of Gαi.
GTPγS binding assay
Cell transfectants (2×106/assay point) were washed twice with PBS and serum-starved for 3 hours. Membrane fractions were prepared from cells resuspended in 1 ml membrane buffer (20 mM HEPES, 6 mM MgCl2, 1 mM EGTA, pH 7.2) supplemented with protease inhibitors (Roche), by incubation on ice for 10 minutes, and subsequent squeezing the cells six times through a syringe (G25). The supernatant of a first, low-speed centrifugation step (500 g, 10 minutes, 4°C) was again spun down at a higher speed (20,000 g, 30 minutes, 4°C) and the resulting membrane pellet was resuspended in GTPγS assay buffer (50 mM HEPES, 100 mM NaCl, 10 mM MgCl2, 1 mM EGTA, 0.1% BSA, pH 7.2) supplemented with protease inhibitors. GTPγS binding was assessed in the presence and absence of CCL19 (5 μg/ml) by adding [35S]GTPγS (0.5 nM) and GDP (10 μM, Sigma) to 200 μg of membrane preparations. After 30 minutes at 37°C, each sample was sucked onto a GF/C filter (Whatman, Maidstone, UK). Filters were washed four times with washing buffer (50 mM HEPES, 5 mM MgCl2) and dried at 60°C. Scintillation fluid (rotiszinteco; Roth, Karlsruhe, Germany) was added and scintillography performed with a Beckman LS 6000IC counter (Fullerton, CA).
This work was supported by the German Research Foundation (DFG, TR-SFB 11 TP-C9 and GR 1517/8-1), the Thurgauische Stiftung für Wissenschaft und Forschung, the State Secretariat for Education and Research and the Thurgauische Krebsliga. D.F.L. is a recipient of a career development award from the Prof. Dr Max Cloëtta Foundation. The authors declare no competing financial interests.
↵* These authors contributed equally to this work
- Accepted April 30, 2008.
- © The Company of Biologists Limited 2008