Dendritic cells are potent antigen-presenting cells endowed with the unique ability to initiate adaptive immune responses upon inflammation. Inflammatory processes are often associated with an increased production of serotonin, which operates by activating specific receptors. However, the functional role of serotonin receptors in regulation of dendritic cell functions is poorly understood. Here, we demonstrate that expression of serotonin receptor 5-HT7 (5-HT7R) as well as its downstream effector Cdc42 is upregulated in dendritic cells upon maturation. Although dendritic cell maturation was independent of 5-HT7R, receptor stimulation affected dendritic cell morphology through Cdc42-mediated signaling. In addition, basal activity of 5-HT7R was required for the proper expression of the chemokine receptor CCR7, which is a key factor that controls dendritic cell migration. Consistent with this, we observed that 5-HT7R enhances chemotactic motility of dendritic cells in vitro by modulating their directionality and migration velocity. Accordingly, migration of dendritic cells in murine colon explants was abolished after pharmacological receptor inhibition. Our results indicate that there is a crucial role for 5-HT7R–Cdc42-mediated signaling in the regulation of dendritic cell morphology and motility, suggesting that 5-HT7R could be a new target for treatment of a variety of inflammatory and immune disorders.
Serotonin (5-hydroxytryptamine, 5-HT) is one of the most extensively studied neurotransmitters in the central nervous system (CNS) regulating multiple physiological functions including the control of anger, aggression, body temperature, appetite, sleep, mood and pain (Mössner and Lesch, 1998). Serotonin operates by activating a family of specific 5-HT receptors, which comprise seven distinct classes based on their structural and functional characteristics. Serotonin receptors belong to the G-protein-coupled receptor family with exception of the 5-HT3 receptor, which is an ion channel (Barnes and Sharp, 1999). Despite the major role of 5-HT in the CNS, ∼90% of 5-HT is produced in the gastro-intestinal tract (Racké et al., 1996), where it is involved in the regulation of multiple physiological processes such as stimulation of cytokine and chemokine production (Dürk et al., 2005; Idzko et al., 2004; Müller et al., 2009), cell proliferation (Pakala and Benedict, 1998; Pakala et al., 1997), migration (Kushnir-Sukhov et al., 2006; Müller et al., 2009; Tamura et al., 1997) and the regulation of the immune system (Ahern, 2011). Expression of 5-HT receptors has been identified on a broad range of immune cells, including T cells (O'Connell et al., 2006), macrophages (Mikulski et al., 2010) and dendritic cells (Idzko et al., 2004).
Dendritic cells are important innate immune cells, which are known as the most effective antigen-presenting cells (APCs). They play a key role in the induction of immune responses by capturing and transferring antigens to the cells of the adaptive immune system (Banchereau et al., 2000; Banchereau and Steinman, 1998; Figdor et al., 2004; Steinman and Witmer, 1978). In the steady state, immature dendritic cells continuously capture and take up self-antigens as well as injurious environmental proteins (Steinman et al., 2003). Following tissue damage, increased levels of inflammatory chemokines, cytokines, and foreign antigens (e.g. lipopolysaccharide, LPS) induce the maturation of dendritic cells (Ricart et al., 2011). During this process, dendritic cells lose the ability to further uptake and process antigens, while gaining the capacity to activate T cells (Banchereau et al., 2000; Langenkamp et al., 2000; Roses et al., 2008). For that, mature dendritic cells have to migrate to the secondary lymphoid organs, where they directly interact with naïve T cells to initiate immune responses (Banchereau and Steinman, 1998; Harada et al., 2012; Müller et al., 2009; Randolph et al., 2005; Steinman, 1991). In addition, maturating dendritic cells undergo fundamental morphological changes by cytoskeleton reorganizations leading to the development of cellular extensions (Hubo et al., 2013; Verdijk et al., 2004). Maturation is an essential factor for dendritic cell functions, because this process includes the upregulated surface expression of the crucial chemokine receptor CCR7, which drastically enhances migratory capacity of cells (Ricart et al., 2011; Sallusto et al., 1999; Yanagihara et al., 1998). Beside the chemokine receptor CCR7, 5-HT receptors expression profiles are also altered during maturation of dendritic cells. It has been demonstrated that immature dendritic cells highly express 5-HT1BR, 5-HT1ER and 5-HT2BR, whereas mature dendritic cells show expression of 5-HT4R and 5-HT7R (Idzko et al., 2004). The 5-HT7R is one of the most recently identified members of the 5-HT receptor family. Its activation induces cAMP elevation through the stimulatory Gs protein (Albayrak et al., 2013; Hedlund, 2009) and upregulates the release of pro-inflammatory cytokines and co-stimulatory molecules for T cell activation in mature dendritic cells (Dienz and Rincon, 2009; Müller et al., 2009). It has also been shown that in neurons 5-HT7R is involved in regulation of neuronal morphology through the G12-protein-mediated activation of the small GTPase Cdc42 (Kvachnina et al., 2005; Ponimaskin et al., 2007).
Although Cdc42 involvement in dendritic cell morphology and motility has been intensively studied during the last decade, the regulatory upstream mechanisms controlling Cdc42 functions have not been elucidated so far. In the present study, we demonstrate that expression of the 5-HT7R and Cdc42 is significantly increased in dendritic cells upon maturation. Although maturation of dendritic cells was not influenced by 5-HT7R, stimulation of the 5-HT7R–Cdc42 signaling pathway influenced formation and elongation of cellular processes in mature bone-marrow-derived dendritic cells (BMDCs). Moreover, our data demonstrate that, in dendritic cells, 5-HT7R-mediated signaling is involved in regulation of the expression level of the chemokine receptor CCR7. In addition, our data show that 5-HT7R plays an important role in the regulation of chemotactic motility of BMDCs under in vitro conditions by modulating directionality and migration velocity. Accordingly, the dendritic cell migration in colon preparations from mice was significantly decreased after pharmacological receptor inhibition. Taken together, our results suggest that 5-HT7R–Cdc42-mediated signaling regulates dendritic cell morphology and migration, and thus might represent an attractive target for the treatment of a variety of inflammatory and immune diseases.
Expression of 5-HT7R is upregulated upon maturation of BMDCs
It has been suggested that different effects of serotonin on immune cells might be mediated by the variable expression of the corresponding 5-HT receptors (Idzko et al., 2004). Therefore, we analyzed whether the expression of 5-HT7R changes upon dendritic cell maturation. For that, we determined receptor expression profiles in immature and mature BMDCs using quantitative real-time PCR (qRT-PCR). Maturation of BMDCs was induced by LPS treatment for 48 h. As shown in Fig. 1A, BMDC maturation caused a significant upregulation of the 5-HT7R mRNA level (9.5-fold) as compared to immature BMDCs (Fig. 1A). Given that the level of mRNA transcripts do not necessarily correlate with the protein expression level, we analyzed protein expression of 5-HT7R in mature and immature BMDCs by western blot analysis. In line with transcript levels, protein expression of 5-HT7R in membrane fractions significantly increased upon BMDC maturation (Fig. 1B). Immunocytochemical analysis further confirmed a higher expression of 5-HT7R in mature BMDCs (Fig. 1C). Taken together, these results demonstrate upregulation of 5-HT7R during maturation of BMDCs.
5-HT7R-mediated signaling does not affect differentiation and maturation of BMDCs
To investigate the functional impact of 5-HT7R on dendritic cell development, we compared the maturation profile of BMDCs isolated from wild-type (WT) and 5-HT7R-deficient mice. Flow cytometry analysis revealed a similar expression profile of several well-established dendritic cell surface markers, such as major histocompatibility complex class II (MHC II) molecules, CD11c (also known as ITGAX) and CD86 in LPS-stimulated mature 5-HT7R-deficient and WT BMDCs (Fig. 2A,B). Similar results were obtained in immature BMDCs (data not shown). In addition, the number and ratio of mature and immature BMDCs in WT and 5-HT7R-deficient BMDC cell cultures was quite similar (Fig. 2A,B). Thus, 5-HT7R does not affect differentiation of BMDCs under basal conditions.
Next, we studied the effect of receptor stimulation on BMDC maturation. Immature and mature BMDCs were treated with the 5-HT7R agonist 5-carboxamidotryptamine maleate (5-CT; 10 µM) either alone or in combination with 10 µM of the highly selective 5-HT7R antagonist SB 269970 for 48 h. These concentrations were selected based on previously published data (Amireault and Dube, 2005; Crider et al., 2003; Kvachnina et al., 2005; Saitow et al., 2009). Quantitative analysis of flow cytometry data showed no difference in the number of immature and mature BMDCs after 5-HT7R stimulation (Fig. 2C,D) demonstrating that 5-HT7R-mediated signaling has no influence on the maturation of BMDCs.
Stimulation of 5-HT7R leads to process formation and elongation in mature BMDCs through Cdc42-mediated signaling
We have previously shown that activation of the 5-HT7R–Cdc42 signaling pathway leads to an increased neurite outgrowth and to the formation of dendritic spines in hippocampal neurons (Kvachnina et al., 2005; Ponimaskin et al., 2007). Therefore, we hypothesized that the morphology of dendritic cells, including the events of process formation and elongation, might also be associated with 5-HT7R–Cdc42-mediated signaling. To test this hypothesis, BMDCs isolated from transgenic mice expressing GFP under the control of the β-actin promoter (Berberich et al., 2008; Okabe et al., 1997), were seeded as 95×103 cells/well on 12-mm coverslips and stimulated with 1 µg/ml LPS. Cells were incubated with vehicle (H2O), 10 µM 5-CT alone or 5-CT in combination with 10 µM SB 269970 for 24 h, followed by morphometric analysis. BMDCs were identified by incubation of GFP-expressing cells with antibodies against the specific dendritic cells markers CD11c and CD86 typically expressed by mature dendritic cells (Fig. 3A). The impact of Cdc42 activation on dendritic cell morphology was investigated by treatment of GFP-expressing mature BMDCs with 50 µM of the selective Cdc42 inhibitor ZCL 278 (Friesland et al., 2013) alone or in combination with 5-CT for 24 h.
Morphometric analysis of long-term experiments (24 h) revealed a significant increase in the length of processes in mature BMDCs treated with the receptor agonist in comparison with H2O-treated control cells (Fig. 3A,B), whereas no significant changes in the number of processes were observed (Fig. 3C). This effect was receptor specific, because it was completely blocked by the treatment with receptor antagonist SB 269970. Interestingly, the inhibition of Cdc42 resulted in a significant reduction in both the length and number of processes, as compared to the vehicle-treated cells (Fig. 3A–C), suggesting the importance of basal Cdc42 activity for BMDC morphology. This effect could not be rescued by receptor stimulation with agonist, further emphasizing that Cdc42-mediated signaling is downstream of 5-HT7R.
Long-term experiments were extended by short-term analysis, where stimulation of cells occurred for 30 min. Similar to results obtained after prolonged treatment, short-term stimulation with 5-CT also led to an increased length of processes, although the effect was less pronounced (supplementary material Fig. S1). Consistent with the results of long-term stimulation experiments, the pharmacological inhibition of Cdc42 alone or along with 5-HT7R stimulation by 5-CT also induced a significant decrease in process length (supplementary material Fig. S1).
The impact of Cdc42 in process formation was further confirmed by the observation that the expression level of Cdc42 was significantly increased in both mature WT and 5-HT7R-deficient BMDCs in comparison to immature cells (Fig. 3D,E). It is noteworthy that 5-HT7R deficiency does not lead to an altered expression of Cdc42 in mature BMDCs, because we obtained a comparable level of Cdc42 in WT and 5-HT7R-deficient BMDCs (supplementary material Fig. S1D). Taken together, these results demonstrate a crucial role of 5-HT7R–Cdc42-mediated signaling in regulation of dendritic cell morphology.
5-HT7R–Cdc42-mediated signaling enhances the chemotactic migration of mature BMDCs in Transwell migration assays
The capacity of dendritic cells to initiate immune responses is dependent on their specific chemotactic migration allowing dendritic cells to reach lymph nodes and to interact with naïve T cells. To analyze the ability of 5-HT7R to induce chemotactic migration of dendritic cells in vitro, we subjected BMDCs isolated from WT or 5-HT7R-deficient mice to Transwell migration assays. Cells were transferred on the porous membrane in the upper compartment of Transwell migration chambers, accompanied with 5-HT7R agonist treatment either in the upper or lower compartment chamber. The number of cells migrated to the lower compartment was determined after 24 h. These experiments demonstrate that mature BMDCs do not possess a chemotactic response to 5-HT7R activation with 5-CT (Fig. 4A). In addition, the CXCL12-induced BMDC migration was not influenced by the 5-HT7R stimulation with 5-CT (supplementary material Fig. S2A). CXCL12 is a ligand of CXCR4 receptor known to induce strong chemotactic migration of immature dendritic cells.
To analyze whether 5-HT7R stimulation can modulate the chemotactic behavior of mature dendritic cells, we transferred mature WT and 5-HT7R-deficient BMDCs to the upper compartment of the Transwell chamber and added 0.1 µg/ml C-C motif chemokine 19 (CCL19) to the lower chamber compartment. CCL19 is a ligand of the chemokine receptor CCR7 (Ricart et al., 2011; Sallusto et al., 1999; Vecchi et al., 1999; Yanagihara et al., 1998) and a strong activator of mature dendritic cell migration (Sozzani et al., 1998). The concentration of 0.1 µg/ml CCL19 was selected in accordance with previous data indicating a maximal chemotactic response at this concentration (Ricart et al., 2011; Yoshida et al., 1998). In parallel, 10 µM of the 5-HT7R agonist 5-CT, either alone or in combination with 10 µM of the 5-HT7R antagonist SB 269970, was added to the upper Transwell compartment. The role of Cdc42 activity was elucidated by the pre-treatment of cells with 50 µM ZCL 278, a selective Cdc42 inhibitor, added either alone or in combination with 5-CT. The number of cells migrated to the lower compartment was determined in long-term (analysis after 24 h) and in short-term (analysis after 5 h) experiments.
Quantitative analysis of long-term experiments revealed that CCL19 induces a strong chemotactic movement of mature WT BMDCs with a chemotactic index of 7.9±0.2 (mean±s.e.m., n=3), whereas the chemotactic index of mature 5-HT7R-knockout (KO) BMDCs was reduced to 5±0.5 (Fig. 4B). Stimulation of 5-HT7R with 5-CT results in a further increase of the chemotactic index to 10.4±0.4 in WT BMDCs, whereas no significant chemotactic index changes appeared in 5-HT7R-deficient BMDCs upon 5-CT treatment (chemotactic index of 5.6±0.5; Fig. 4B). Application of the 5-HT7R-specific antagonist SB 269970 in combination with receptor stimulation leads to a strong and significant decrease of CCL19-induced chemotaxis in mature WT BMDCs (chemotactic index of 6.5±0.1 versus 10.4±0.4), demonstrating that the 5-CT-induced increase in chemotactic index was mediated by 5-HT7R. This was further confirmed by the observation that combined application of 5-CT and SB 269970 had no significant influence on the chemotactic movement of 5-HT7R-deficient BMDCs (chemotactic index of 4.7±0.9).
Treatment of mature WT BMDCs with Cdc42 inhibitor caused no significant changes in chemotactic index (9±0.5) in comparison to WT cells treated only with CCL19 (Fig. 4B). In contrast, the increase in chemotactic motility obtained upon 5-HT7R stimulation with 5-CT was completely abolished upon simultaneous Cdc42 inhibition with ZCL 278 (chemotactic index of 10.4±0.4 versus 5.5±0.2), suggesting that 5-HT7R-mediated Cdc42 activation is important for the chemotactic dendritic cell motility. Treatment of 5-HT7R-deficient BMDCs with ZCL 278 either alone or in combination with 5-CT did not demonstrate any significant changes of the chemotactic index, when compared with 5-HT7R-deficient control cells (chemotactic index of 4.8±1.4 and 3.3±1.3 versus 5±0.5; Fig. 4B).
Similar to results obtained in long-term stimulation experiments, quantitative analysis of short-term experiments revealed that CCL19 induced a strong chemotactic movement of mature WT BMDCs with a chemotactic index of 12.2±1.7, whereas the chemotactic index of mature 5-HT7R-KO BMDCs was reduced to 6.9±0.3 (supplementary material Fig. S3). In addition, the treatment of mature WT and 5-HT7R-deficient BMDCs with Cdc42 inhibitor ZCL 278 resulted in pronounced inhibition of chemotactic motility (chemotactic index of WT 1.5±0.4 and 1.6±0.7; chemotactic index of KO 0.8±0.1 and 1.4±0.3). It is noteworthy that, in short-term experiments, 5-CT treatment was not sufficient to increase the chemotactic index of WT BMDCs, as it was obtained after prolonged stimulation (chemotactic index 12.05±2.1), suggesting that the nature of 5-HT7R-mediated increase of chemotactic movement of BMDCs is complicated.
To analyze the molecular mechanisms underlying 5-HT7R-mediated facilitation of chemotactic dendritic cell motility in more details, we compared the expression profile of CCR7 in mature WT and 5-HT7R-deficient BMDCs treated with 10 µM 5-CT or H2O for 24 h. Analysis of flow cytometry data revealed a significantly reduced expression of the chemokine receptor in 5-HT7R-deficient BMDCs when compared with WT cells under basal conditions (Fig. 4C,D). In contrast, the CCR7 expression profile was not affected upon 5-HT7R activation with 5-CT (supplementary material Fig. S4A,B), indicating that basal 5-HT7R activation is sufficient for proper CCR7 expression.
In addition to CCR7, mature dendritic cells also express CXCR4 receptor and thus could respond to CXCL12. However, it has been shown that CCL19 is a ∼100-fold more potent chemoattractant for mature dendritic cells than CXCL12. Furthermore, mature dendritic cells show a strong decrease in the chemotactic response to CXCR4 ligands, demonstrating that the CXCL12–CXCR4 axis is much less important for migration of mature dendritic cells, as compared to CCR7 (Humrich et al., 2006; Ricart et al., 2011). Based on these findings, together with our observation that CXCL12-mediated migration of immature dendritic cells was not modulated by the 5-HT7R (supplementary material Fig. S2B), we did not further study the role of 5-HT7R in modulating CXCR4-mediated chemotactic responses in mature dendritic cells.
Taken together, we conclude that 5-HT7R per se does not possess chemotactic properties. However, the presence of 5-HT7R plays an important role for CCL19-induced chemotaxis of mature BMDCs, most probably by regulating the expression of CCR7. Our data also demonstrate that 5-HT7R-induced Cdc42 activation is involved in the control of chemotactic motility of mature BMDCs.
5-HT7R-mediated signaling enhances directionality of migrating BMDCs in a 3D collagen gel
Although a Transwell assay is commonly used for analysis of chemotaxis (Honig et al., 2003; Ledgerwood et al., 2008; Yopp et al., 2005), it does not allow the measurement of migration velocities and directionality (Tayalia et al., 2011). In contrast, the collagen-based 3D system for analysis of chemotactic cell migration is widely accepted as a system allowing detailed analysis of migration and is a better system to recapitulate the physiological cell environment (Lämmermann et al., 2008, 2009). Therefore, we extended our analysis to 3D collagen gels, which mimic a complex fibrillar scaffold obtained in intact tissues. In this experiment, collagen gels containing mature WT or 5-HT7R-KO BMDCs were cast in custom-made migration chambers and overlaid with medium containing 0.1 µg/ml CCL19. The role of 5-HT7R in chemotactic migration was validated by treatment of cells either with 10 µM 5-CT or 5-CT together with 10 µM SB 269970. Cell migration was analyzed by time-laps microscopy and single-cell tracking for 5 h (Fig. 5A,B). In accordance with results obtained in the Transwell system, both cell types (WT and 5-HT7R-KO) migrated along the chemokine gradient. However, the directionality (i.e. the ratio between the euclidean distance and accumulated distance) of WT BMDCs was significantly higher compared to that of 5-HT7R-deficient cells (Fig. 5A–C; supplementary material Movies 1 and 2), suggesting the impact of basal receptor activity for defined migration directionality. This was further confirmed by the observation that treatment of WT BMDCs with receptor antagonist resulted in a decrease of directionality to the values obtained for 5-HT7R-deficient BMDCs (Fig. 5C). Activation of 5-HT7R by 5-CT led to a significantly increased directionality in WT BMDCs when compared with control cells (0.60±0.02 versus 0.53±0.02, mean±s.e.m., n=4, Fig. 5C). Neither application of 5-HT7R agonist nor antagonist influenced 5-HT7R-KO BMDC directionality when compared with control H2O-treated 5-HT7R-KO cells (Fig. 5C).
Additionally, we investigated the role of 5-HT7R activity in regulating migration velocity of BMDCs. As shown in Fig. 5D, the migration velocity of cells was significantly reduced in both antagonist-treated WT BMDCs and untreated 5-HT7R-deficient BMDCs, when compared with control WT BMDCs (1.68±0.05 µm/min and 1.75±0.05 µm/min versus 1.91±0.05 µm/min). It is noteworthy, that no significant changes in mean velocity were obtained between differently treated 5-HT7R-deficient BMDCs (Fig. 5D). Taken together, these results suggest that 5-HT7R plays an important role in the regulation of chemotactic motility of BMDCs in 3D collagen environment by modulating directionality and migration velocity.
5-HT7R-mediated signaling increases velocity of migratory active intestinal dendritic cells in colon
To extend our observations to the in vivo situation, we analyzed the migration behavior of dendritic cells in colon explants from CD11c–EYFP transgenic mice. In this transgenic mouse line, the enhanced yellow fluorescent protein (EYFP) is placed under the control of the CD11c promoter, allowing an unaltered live-cell imaging without any further labeling procedures. Moreover, in our previous study we have demonstrated that a large proportion of the EYFP-expressing cells possess markers described to be characteristic for dendritic cells (Lindquist et al., 2004; Voedisch et al., 2012). Dendritic cell migration was analyzed by two-photon time-lapse microscopy in longitudinally opened colon sections with the lamina propria placed upside and the muscle layer downside after treatment with 1 µg/ml LPS. The impact of 5-HT7R was investigated by treatment of preparations with 10 µM of the 5-HT7R antagonist SB 269970 or H2O, for 1 h. To ensure that only dendritic cells within the colon epithelium were captured for the analysis, additional second harmonic generation (SHG) signals of collagen fibers were collected. The collagen fibers closely underlie the colon epithelium and therefore represent suitable markers for in-tissue orientation. For motility analysis, only EYFP-positive cells colocalizing with collagen fibers were evaluated (Fig. 6A). Trajectories analysis revealed that after pharmacological blockade of 5-HT7R with antagonist, dendritic cells show a reduced and diffuse migration path, when compared with straight and long-distance moving control cells (Fig. 6B). In accordance with these observations, the mean migration distance of dendritic cells in preparations treated with 5-HT7R antagonist was significantly decreased in comparison to control dendritic cells [Fig. 6C; 112±5.9 µm for LPS plus H2O (N=7; n=343 and 61.4±3.4 µm for LPS SB 269970 (N=6; n=231); N is the number of performed experiments and n the number of analyzed cells]. These results demonstrate that 5-HT7R-mediated signaling can modulate the motility of dendritic cells in vivo by regulating their migration velocity.
In the present study, we analyzed the impact of 5-HT7R-mediated signaling in the regulation of dendritic cell functions with a particular focus on dendritic cell morphology and motility. The ability of dendritic cells to prime naïve T cells within the draining lymph nodes and to initiate immune responses is dependent on their chemotactic migration. Leukocyte migration is known to be mainly driven by cell shape changes mediated by rearrangements of the cytoskeletal actin network (Fenteany and Glogauer, 2004; Lämmermann et al., 2008). In this context, members of the Rho family of small GTPases, whose most prominent feature is the regulation of the actin cytoskeleton (Hall, 1994; Machesky and Hall, 1997; Norman et al., 1994) represent key players that are crucially involved in the regulation of dendritic cell migratory properties (Benvenuti et al., 2004; Harada et al., 2012; Lammermann et al., 2009; Ocana-Morgner et al., 2011). Our previous data have indicated that, in neurons, actin cytoskeleton rearrangements can be regulated by 5-HT7R through the G12 protein-mediated activation of the small GTPase Cdc42 (Kobe et al., 2012; Kvachnina et al., 2005; Nobes and Hall, 1995a,b). We have also demonstrated a high basal activity of 5-HT7R towards Cdc42 signaling in neuronal cells (Kvachnina et al., 2005), suggesting that both basal activity and agonist-mediated stimulation of the 5-HT7R–Cdc42 signaling pathway might contribute to morphological rearrangements of cells. Referring to dendritic cell morphology, Cdc42 has previously been shown to modulate cell shape changes by inducing alterations in the length and number of membrane protrusions, and it is required for the induction and maintenance of the typical stellate cytoskeletal architecture of mature dendritic cells (Jaksits et al., 2004; Nikolic et al., 2011; Swetman et al., 2002). Furthermore, it has been shown that Cdc42 deficiency induces a more rounded morphology in cultured cells (Chen et al., 2000; Wu et al., 2007). Similar results have also been obtained under in vivo conditions (Luckashenak et al., 2013), where authors observed a reduced number of protrusions and more rounded morphology in Cdc42-deficient Langerhans cells. Although the significance of small GTPases for morphogenesis of dendritic cells is well established, the upstream signaling components regulating Cdc42-mediated pathways in dendritic cells, as well as its role in the regulation of dendritic cell motility have not been fully characterized.
In the present study, we found that expression of 5-HT7R at both the mRNA and the protein level was upregulated in mature dendritic cells, suggesting an elevated functional mode and the enhancement of receptor-mediated downstream signaling that might be involved in the regulation of dendritic cell functions. Our data also demonstrate that 5-HT7R-mediated signaling modulates morphology as well as the migratory capacity of dendritic cells and that Cdc42 is crucially involved in these processes. Accordingly, we obtained increased protein expression of Cdc42 in mature BMDCs. Reorganization of actin filaments results not only in morphological changes, but is also highly involved in cell motility and migration events that are essential for proper responses to tissue damage as well as infection and for the control of immune surveillance (Kaibuchi et al., 1999; Raftopoulou and Hall, 2004; Ridley, 2001). Cdc42 has also been previously revealed as an important regulator of cell polarity by stabilizing spatial and temporal asymmetries, thereby enabling a clear distinction between cell front and rear to navigate the straight movement of cells (Cau and Hall, 2005; Etienne-Manneville, 2004; Lauffenburger and Horwitz, 1996; Szczur et al., 2006). Furthermore, we have previously shown that Cdc42 deficiency affects the spatial and temporal coordination of protrusions, thereby entailing the development of multiple competing leading edges in the presence of chemoattractant CCL19, which leads to decreased directionality of cells. This finding suggests that Cdc42 is not only involved in formation and/or elongation of protrusions, but also important for the reorganization of these protrusions in leading edges thus enabling proper migration towards chemoattractant (Lammermann et al., 2009). Results of the current study identify 5-HT7R as an important upstream regulator of Cdc42-mediated signaling involved in dendritic cell migration. Receptor stimulation induces an enhanced chemotactic capacity, which was accompanied and therefore attributed to an increased directionality and velocity of dendritic cells. Beside the involvement in the regulation of dendritic cell motility in vitro, 5-HT7R-mediated signaling was here additionally demonstrated to be essentially involved in proper dendritic cell migration in colon.
Results of in vitro Transwell analysis suggest that 5-HT7R-mediated signaling might modulate dendritic cell migration not only through activation of Cdc42, but also by induction of the other small GTPases. In our previous works, we have shown that RhoA can be activated by the 5-HT7R, although to a lesser extent than Cdc42 (Kvachnina et al., 2005; Kobe et al., 2012). In contrast to Cdc42, which propels cell movement by promoting cellular extension and spreading, RhoA mediates detachment and cell rounding through regulation of actin filament rearrangements (Graness et al., 2006; Hall, 1994, 1998; Sepp and Auld, 2003; Swetman et al., 2002). In dendritic cells, constitutively active mutants of RhoA have been shown to induce cell rounding, unpolarized cell morphology, decreased migration velocity and reduced chemotaxis (Shurin et al., 2005). Thus, the mechanism regulating dendritic cell motility through 5-HT7R might involve a cross-talk between Cdc42 and RhoA (Kvachnina et al., 2005; Li et al., 2002). In this context, a reduced chemotactic capacity of dendritic cells obtained under 5-HT7R stimulation with Cdc42 inhibition in comparison to effect obtained in BMDCs treated only with Cdc42 inhibitor might be attributed to 5-HT7R-mediated activation of RhoA, which results in reduced chemotaxis.
Our data also revealed that 5-HT7R deficiency impairs the expression of the chemokine receptor CCR7 on mature BMDCs. This suggests that beside its role in the regulation of dendritic cell motility through induction of cytoskeletal rearrangements, 5-HT7R-mediated signaling might also affect motility of dendritic cells through regulation of CCR7 expression (Förster et al., 1999; Jang et al., 2006). In accordance with this hypothesis, a previous study reported a reduced ability of maturing Cdc42-deficient dendritic cells to upregulate CCR7 expression (Luckashenak et al., 2013). Furthermore, it has been demonstrated that 5-HT7R-mediated signaling regulates the activation of NF-κB in colon-derived dendritic cells (Kim et al., 2013). By contrast, inhibition of NF-κB has been shown to significantly reduce the induction of CCR7 expression in monocyte-derived dendritic cells (van de Laar et al., 2010). These findings indicate that 5-HT7R–Cdc42-mediated signaling might regulate dendritic cell motility through induction of chemokine receptor CCR7 expression by activation of the NF-κB signaling pathway. Furthermore, 5-HT7R–Cdc42-mediated signaling has been reported to mediate activation of the transcription factor serum response factor (SRF) (Kvachnina et al., 2005). Downregulated expression of SRF target genes paralleled by impaired CCR7 signaling has previously been shown in lesion macrophages of uremic mice (Ponda et al., 2010). In addition, a decreased expression of other SRF-driven genes involved in dendritic cell motility, for example MMP-9 (Yen et al., 2010), could be involved in decreased 3D migration of 5-HT7R-deficient BMDCs. Thus, it would be interesting to analyze in future studies whether the expression of corresponding genes are changed in 5-HT7R-deficient dendritic cells.
Our finding that 5-HT7R can modulate the migratory properties of mature dendritic cells suppose that impaired 5-HT7R-mediated signaling in dendritic cells might be involved in the pathogenesis of inflammatory diseases, which are often accompanied by reduced dendritic cell motility (Angeli and Randolph, 2006; Eriksson and Singh, 2008; Hermida et al., 2014). In our recent work, we have demonstrated that 5-HT7R-mediated signaling is crucially involved in the progression of chemically induced acute and chronic colitis in mice (Guseva et al., 2014). In particular, we observed that pharmacological inhibition as well as genetic ablation of 5-HT7R result in increased severity of intestinal inflammation in both acute and chronic colitis. In addition, in mice with dextran sulfate sodium (DSS)-induced colitis as well as in mucosa biopsies from patients with Crohn's disease, we obtained a highly increased expression level of 5-HT7R and this increase was attributed to 5-HT7R expression on CD11c/CD86 double-positive immune cells. The results of the present study might provide the molecular mechanism explaining the increased inflammatory response obtained upon inhibition of 5-HT7R. Inactivation of 5-HT7R-mediated signaling will result in impaired cytoskeletal rearrangements and reduced expression of CCR7 on dendritic cells leading to decreased directionality and migration velocity of cells as well as to their decelerated motility. Thus, dendritic-cell-mediated activation of specific T cell subsets to induce immune responses against foreign antigens as well as to mediate tolerance to self-antigens will be delayed. Furthermore, impaired 5-HT7R-mediated signaling increases the production of pro-inflammatory cytokines (Guseva et al., 2014). These combined events might exacerbate the severity of inflammatory bowel diseases such as Crohn's disease.
In conclusion, results of the current study reveal that serotonin can modulate motility of dendritic cells by regulating cell morphology, velocity and directionality of migration as well as expression level of the chemokine receptor CCR7 through 5-HT7R–Cdc42-mediated signaling. Our findings represent a new serotonin-mediated mechanism in the regulation of immune cell polarization and pathfinding of dendritic cells and indicate the 5-HT7R could be a new target for the therapy of inflammatory diseases.
MATERIALS AND METHODS
Mice were bred at the central animal facility of Hannover Medical School (Hannover, NI, Germany) or at Charles River Laboratories (Sulzfeld, BW, Germany). The following mice strains were used: C57BL/6J (wild type, WT), C57BL/6J-Htr7tm1Sut (5-HT7R knockout, 5-HT7R-deficient, 5-HT7R-KO) (Hedlund et al., 2003), C57BL/6-Tg(ACTbEGFP) (designated here as GFP-transgenic mice; these mice constitutively express EGFP in all cells under the control of the β-actin promoter) and CD11c-EYFP transgenic mice (these mice express EYFP under the control of the CD11c promoter) (Berberich et al., 2008; Lindquist et al., 2004; Okabe et al., 1997; Voedisch et al., 2012). This study was conducted in accordance with German law for animal protection and with the European Communities Council Directive 86/609/EEC for the protection of animals used for experimental purposes. All experiments were approved by the Local Institutional Animal Care and Research Advisory committee and permitted by the local government (number 33.14-42502-04-12/0753).
Generation of dendritic cells in vitro
Bone marrow-derived dendritic cells (BMDCs) were generated from a flushed bone marrow suspension as previously described (Czeloth et al., 2005). In brief, 5×106 bone marrow cells from tibia and femur of WT, 5-HT7R-KO and GFP-transgenic mice were cultured in 10 ml dendritic cell medium containing RPMI 1640 (Invitrogen, Dun Laoghaire, Ireland) with 10% FCS (Invitrogen, Dun Laoghaire, Ireland), 0.0035% β-mercaptoethanol (Roth, Karlsruhe, BW, Germany), 1% L-glutamine (Invitrogen, Dun Laoghaire, Ireland) and 1% penicillin-streptomycin (Invitrogen, Dun Laoghaire, Ireland) supplemented with 100–200 ng/ml granulocyte-macrophage colony-stimulating factor (GM-CSF) produced by a NIH-3T3 cell line infected with Psi2-pM5DGM#6 (Ohl et al., 2004). After 3 days in vitro (3 div), another 10 ml cell medium supplemented with GM-CSF was added to the cells. At 6 div, half of the culture supernatant was collected, centrifuged at 1200 g for 10 min at room temperature and the cell pellet was resuspended in 10 ml fresh medium with GM-CSF. The cell suspension was added back to the original plate and cells were cultured. At 8 div, non-adherent cells were collected and centrifuged at 1200 g for 10 min at room temperature. Cell pellet was resuspended in dendritic cell medium and cells were counted using a hemocytometer. Cells were seeded on petri dishes (TPP, Trasadingen, Switzerland) at 5×106 cells/dish in 10 ml dendritic cell medium. To generate mature BMDCs, 1 µg/ml lipopolysaccharide (LPS) (Sigma-Aldrich, Buchs SG, Switzerland) was added to the cell suspensions. To get immature BMDCs, H2O was added instead. Further treatments of cells for analysis of cell morphology, cell surface marker expression profiles and cell migration behavior were as described in the following sections. At 9 or 10 div, mature and immature WT BMDCs were harvested, respectively.
Flow cytometry was performed according to standard procedures. BMDCs were blocked with 5% rat serum (AbD serotec, Hercules, CA) in 3% FCS (Invitrogen, Dun Laoghaire, Ireland) in PBS (140 mM NaCl, 6 mM Na2HPO4, 3 mM KCl, 2 mM KH2PO4; Roth, Karlsruhe, BW, Germany) for 15 min at 4°C and surface markers were stained for 15 min at 4°C with anti-CD11c (1:200, anti-mouse CD11c, APC conjugated, eBioscience, Vienna, Austria), anti-MHC II (1:200, anti-mouse I-Ab antibody, FITC conjugated, Biolegend, Fell, RP, Germany), anti-CD86 (1:200, anti-mouse CD86, phycoerythrin-conjugated, Biolegend, Fell, RP, Germany) antibodies and DAPI (1:10,000, Sigma-Aldrich, Buchs SG, Switzerland). For CCR7 expression analysis, BMDCs were preliminarily stained with anti-CCR7 [1:40, anti-mouse CD197 (CCR7), phycoerythrin-conjugated, eBioscience, Vienna, Austria] antibody or the isotype-specific control antibody anti-IgG2a (1:40, anti-mouse IgG2a, κ Isotype Ctrl, phycoerythrin-conjugated, Biolegend, Fell, RP, Germany) for 20 min at 37°C. Flow cytometry was performed on LSRII and FACSAria II (both BD Biosciences, Franklin Lakes, NJ). Data were analyzed with Summit 5.1 software (Beckman Coulter Inc., Krefeld, NM, Germany).
To analyze the physiological function of 5-HT7R on the development of BMDCs, LPS-stimulated or H2O-treated BMDCs were additionally stimulated with 10 µM of the 5-HT7R agonist 5-carboxamidotryptamine maleate (5-CT) (Tocris, Bristol, UK) alone or in combination with 10 µM of the 5-HT7R antagonist SB 269970 hydrochloride (SB 269970) (Tocris, Bristol, UK) at 8 div. To analyze 5-HT7R functions in the regulation of CCR7 expression on mature BMDCs, LPS-stimulated cells were additionally treated with 10 µM of the 5-HT7R agonist 5-CT or H2O for 24 h. The application of SB 269970 was performed 30 min before 5-CT treatment. H2O was used as a control. Cells were incubated at 37°C with 5% CO2. At 10 div, WT and 5-HT7R-KO mature and immature BMDCs were harvested, blocked and incubated with antibodies against MHC II, CD11c and CD86 followed by DAPI solution. Thereafter, cell suspensions were sorted and analyzed for the amount of immature (CD11cHIGH, MHC IILOW and CD86LOW) and mature (CD11cHIGH, MHC IIHIGH and CD86HIGH) BMDCs by using the FACSAria IIU or the XDP (both BD Biosciences, Franklin Lakes, NJ) with 405 nm, 488 nm and 640 nm excitation lasers, fitted with a 100 μm nozzle. Data were collected and the total amount of BMDCs as well as the maturation state of each condition was analyzed using Summit 5.1 software (Beckman Coulter Inc., Krefeld, NM, Germany).
Morphological analysis of GFP-transgenic BMDCs
To estimate the role of 5-HT7R activation on dendritic cells morphology, at 8 div, BMDCs were seeded as 95×103 cells/well on 12-mm coverslips and treated with 1 µg/ml LPS to stimulate cell maturation. In addition, BMDCs were treated with 10 µM 5-CT alone or in combination with 10 µM SB 269970. To estimate the role of Cdc42 signaling on dendritic cell morphology, LPS-treated cells were incubated with 50 µM of the selective Cdc42 inhibitor ZCL 278 (Tocris, Bristol, UK) alone or in combination with 10 µM 5-CT. Application of SB 269970 and ZCL 278 occurred 30 min before 5-CT administration. In both cases, H2O was applied for control measurements. Cells were incubated at 37°C with 5% CO2 for 24 h (long-term analysis). Dendritic cell surface markers were stained with anti-CD11c (1:50, anti-mouse CD11c, APC conjugated, eBioscience, Vienna, Austria) and anti-CD86 (1:50, anti-mouse CD86, phycoerythrin-conjugated, BioLegend, Fell, RP, Germany) antibodies for 30 min at room temperature. For short-term analysis of dendritic cell morphology, stimulation and staining of cells occurred simultaneously for 30 min. Afterwards, dendritic cells were immediately fixed with 4% paraformaldehyde (PFA).
The morphology of BMDCs was analyzed using a Carl Zeiss LSM 780 microscope (Zeiss, Oberkochen, BW, Germany) with a C-Apochromat 40×1.20 NA water-immersion objective. A series of z-stacks (with a width of 0.5 µm/stack) were acquired using a 488 nm laser, a 561 nm laser and a 633 nm laser at a pixel count of 1500×1500 with a 43 µm pinhole. The morphology of BMDCs in short-term experiments was analyzed using a Carl Zeiss LSM 780 (Zeiss, Oberkochen, BW, Germany) with a Plan-Apochromat 63×1.40 NA oil-immersion objective. A series of z-stacks (with a width of 0.5 µm/stack) were acquired using a 488 nm laser, a 561 nm laser and a 633 nm laser at a pixel count of 800×800 with a 55 µm pinhole. The mean number and length of dendritic cell processes was analyzed manually with Motic images Plus 2.0 (Motic China group Co., Ltd., Xiamen, PO, China). Represented images shown in figures are maximum intensity projections of z-stacks.
Immunofluorescence analysis of 5-HT7R expression on BMDCs
To analyze cellular distribution of 5-HT7R, on 8 div, BMDCs were seeded as 95×103 cells/well on 12-mm coverslips and got additionally stimulated with or without 1 µg/ml LPS to generate immature and mature cells. At 9 div, immature and mature BMDCs were treated with 10% normal goat serum (NGS, Jackson Immuno Research Laboratories, West Grove, PA) in cell medium for 15 min at room temperature to block unspecific binding of antibodies. Afterwards, dendritic cells were incubated with primary antibody against 5-HT7R (1:10, Novus biological, Littleton, CO) for 30 min at room temperature. After washing, cells were treated with secondary antibody (1:400, goat anti-rabbit-IgG conjugated to Alexa Fluor 546, Life Technologies, Carlsbad, CA) for 5-HT7R detection in combination with cell surface marker staining with anti-CD11c [1:50, anti-mouse CD11c, allophycocyanin (APC) conjugated, eBioscience, Vienna, Austria] and anti-MHC II (1:50, anti-mouse I-Ab antibody, FITC conjugated, Biolegend, Fell, RP, Germany) antibodies for 30 min at room temperature. After staining, cells were fixed with 4% PFA (Roth, Karlsruhe, BW, Germany) for 10 min at room temperature. The cellular distribution of 5-HT7R was analyzed using a Carl Zeiss LSM 780 microscope (Zeiss, Oberkochen, BW, Germany) with a Plan-Apochromat 63×1.40 NA oil-immersion objective. A series of z-stacks (with a width of 0.4 µm/stack) were acquired from the middle part of the cell using a 488 nm laser, a 561 nm laser and a 633 nm laser at a pixel count of 512×512 and a 50 µm pinhole. Represented images are maximum intensity projections of specific z-stack subsets.
Quantitative real-time PCR analysis
To quantify 5-HT7R mRNA level in immature and mature BMDCs, at 10 div, cells were harvested, blocked with rat serum and stained with antibodies against CD11c, MHC II and CD86 as described above. By using FACS analysis, mature (CD11cHIGH, MHCIIHIGH and CD86HIGH) and immature dendritic cells (CD11cHIGH, MHCIILOW and CD86LOW) BMDCs were sorted. After sorting, total RNA was isolated and purified by using TRIzol Reagent (Invitrogen, Dun Laoghaire, Ireland), RNeasy mini kit (Qiagen, Hilden, NW, Germany) and an RNase-Free DNase set (Qiagen, Hilden, NW, Germany) according to manufacturer's instructions. RNA was reverse transcribed using the SuperScriptIII First-Strand Synthesis System (Invitrogen, Dun Laoghaire, Ireland). Real-time PCR was performed in triplicates on a StepOnePlus™ real-time PCR system (Applied Biosystems, Darmstadt, HE, Germany) using a TaqMan Universal PCR Master Mix (Applied Biosystems, Darmstadt, HE, Germany). For detection of 5-HT7R mRNA, corresponding Gene Expression Assays containing gene-specific primers were used [5-hydroxytryptamine (serotonin) receptor 7 (Dye:FAM), Mm00434133_m1*, Applied Biosystems, Darmstadt, HE, Germany]. GAPDH was used as a reference gene. Data were collected and analyzed using Step one plus 2.1 software (Applied Biosystems, Darmstadt, HE, Germany) and the analysis was performed by using the comparative CT Method (ΔΔ CT Method) according to the procedure described at: http://www3.appliedbiosystems.com/cms/groups/ mcb_support/documents/generaldocuments/cms_042380.pdf.
Preparation of membrane protein extracts and whole-cell lysates
For analysis of 5-HT7R protein expression in BMDCs, on 10 div, mature and immature WT BMDCs were harvested and centrifuged at 1200 g for 10 min at 4°C. Cells were counted as described above. For the preparation of membrane protein extracts, 107 cells were always used. All steps were performed at 4°C. Cell suspensions were centrifuged at 2900 g for 4 min and cells were homogenized in 150 µl homogenization buffer. Homogenates were centrifuged at 2300 g for 5 min. Supernatants were collected and centrifuged again at 13,600 g for 20 min. Cell pellets, containing membrane proteins were resuspended in 50 µM HEPES (Roth, Karlsruhe, BW, Germany) and stored until usage at −80°C.
For analysis of Cdc42 protein expression in BMDCs, on 10 div, mature and immature WT and 5-HT7R-deficient BMDCs were harvested and centrifuged at 1200 g for 10 min at 4°C. Cells were counted as described above. For the preparation of whole-cell lysates, 107 cells were always used. Cell suspensions were centrifuged at 1200 g for 4 min at 4°C and cell pellets were resuspended in 100 µl MAPK buffer (150 mM NaCl, 50 mM Tris, 0.5% C24H39NaO4, 1% Triton, 0.1% SDS, 1 mM Na3VO4, 50 mM NaF,10 mM Na2H2P2O7; Roth, Karlsruhe, BW, Germany) containing 1% Clap and 1% PMSF (both Roth, Karlsruhe, BW, Germany). Cells were incubated with buffer for 15 min on a shaker at 4°C. Next, cells were centrifuged at 14,000 g for 15 min at 4°C and cell pellets were removed. Cell lysates were stored until usage at −20°C.
Protein samples were analyzed on 12% SDS-polacrylamide gels. Proteins were transferred on nitrocellulose membranes (GE Healthcare, Buckingham, UK) (24 h, 40 mA, room temperature), which were then blocked [in 5% dried milk powder in TBS-T (0.2 M Tris, 5 M NaCl, 0.1% Tween 20; Roth, Karlsruhe, BW, Germany) or 5% bovine serum albumin (BSA; Roth, Karlsruhe, BW, Germany) in TBS-T, 1 h, room temperature] and probed with rabbit monoclonal antibody directed against 5-HT7R (Abcam, Milton, UK) or mouse antibody directed against Cdc42 (BD Bioscience, Franklin Lakes, NJ). Membranes were developed with peroxidase-conjugated goat anti-rabbit-IgG antibody (Thermo Fisher Scientific Inc., Waltham, MA) or peroxidase-conjugated goat anti-mouse-IgG antibody (Calbiochem, Bad Soden, HE, Germany). Membranes were developed by chemiluminescence staining using ‘super signal western femto maximum sensitifity substrate’ (Thermo Fisher Scientific Inc., Waltham, MA) according to the manufacturer's instructions. The chemiluminescence was detected on a Fusion SL advanced MP with a 16-bit CCD camera (Peqlab, Erlangen, BY, Germany). Expression levels of 5-HT7R and Cdc42 proteins were normalized to GAPDH expression and evaluated using the ImageQuant TL software (GE Healthcare, Buckingham, UK).
12-month-old CD11c-EYFP-transgenic mice were killed by cervical dislocation. Colons were isolated, washed with PBS containing 3% FCS (Invitrogen, Dun Laoghaire, Ireland) and opened longitudinally.
Colon sections of 5 to 10 cm length were covered by 3 ml cell medium (DMEM with F-12, Life Technologies, Carlsbad, CA) and kept in place by a slice anchor within an imaging chamber. Sections were treated with 1 µg/ml LPS for dendritic cells activation alone or in combination with 10 µM of the 5-HT7R antagonist SB 269970. Imaging was performed for 1 h at 37°C with a TriMScope two-photon microscope (La Vision BioTec, Bielefeld, NRW, Germany) attached to a Titanium Sapphire MaiTai HP ultrafast laser (Spectra-Physics, Mountain view, CA) using a 20× water immersion objective. Control and SB 269970-treated (10 µM) sections were imaged alternately on the same day. To make sure that only dendritic cells of the colon epithelium were captured for analysis, the second harmonic generation (SHG) signal of collagen fibers was collected. For motility measurements, only cells colocalizing with these collagen fibers and being defined as EYFP-expressing colon dendritic cells were included. SHG signals of collagen fibers and EYFP fluorescence of CD11c–EYFP transgenic cells were excited with 920 nm (10.25% of 2.65 W output power) and detected by 490/50 or 525/50 band pass filters, respectively. To generate four-dimensional time-lapse images of moving CD11c–EYFP cells each 5 min of 1 h z-stacks were acquired with a 1 mm z-resolution resulting in image volumes of 500 mm in the xy- and 30 to 60 mm in the z-direction. Image stacks were combined to volumes using Imaris version 6.5.3 (Bitplane AG, Zürich, Switzerland) and the location of cells was determined using the ‘surpass mode spot tracking’ function. Cells with a diameter of 10 µm (mature dendritic cells) were marked in their center with a spot-to-track cell movement. To subtract peristaltic movement of the colon containing dendritic cells chosen for analysis, the ‘translational drift correction’ function was applied on resting single cells, lying within the collagen fibers with SHG signal.
Transwell migration assay
At 9 div, lower chambers of Transwell plates (6.5 mm Transwell, 5.0 µm Pore Polycarbonate Membrane Insert, Corning, NY) were filled with medium containing RPMI 1640 (Invitrogen, Dun Laoghaire, Ireland), 0.0035% β-mercaptoethanol (Roth, Karlsruhe, BW, Germany), 1% L-glutamine (Invitrogen, Dun Laoghaire, Ireland) and 1% penicillin/streptomycin (Invitrogen, Dun Laoghaire, Ireland). The chemokines CCL19 for mature BMDCs and CXCL12 for immature BMDCs (both Peprotech, Rocky Hill, NJ, diluted in RPMI 1640) were added as 0.1 µg/ml. H2O was used as control. 1×105 mature or immature WT or 5-HT7R-KO BMDCs were resuspended in 100 µl cell medium were transferred into the upper chamber. To verify the role of Cdc42 signaling on dendritic cell migration behavior, mature BMDCs were treated in the upper chamber with 50 µM ZCL 278 alone or in combination with 10 µM 5-CT. To estimate the role of 5-HT7R on dendritic cell migration behavior, BMDCs were treated in the upper chamber with 10 µM 5-CT alone or in combination with 10 µM SB 269970 hydrochloride. Application of SB 269970 and ZCL 278 occurred 30 min before 5-CT administration. H2O was applied as control. For long-term experiments, cells were incubated at 37°C with 5% CO2 for 24 h. For analysis, cells from lower chambers were collected and cell numbers were evaluated with the LSRII and FACSAria II (BD Biosciences, Franklin Lakes, NJ) and Summit 5.1 software (Beckman Coulter Inc., Krefeld, NM, Germany). For short-term experiments, cells were incubated at 37°C with 5% CO2 for 5 h. For analysis, cells from lower chambers were collected and cell numbers were evaluated using a hemocytometer. All conditions were performed in duplicates. The chemotactic index was calculated as the ratio between numbers of dendritic cells migrated in the presence and in the absence of stimuli (i.e. chemotatic index=dendritic cells migrated with stimuli/dendritic cells migrated without stimuli).
3D collagen gel migration assay
The 3D collagen gel chemotaxis assay was performed as described previously (Lammermann et al., 2009). Briefly, at 9 div, mature WT and 5-HT7R-KO BMDCs were washed once with PBS. Cells were resuspended in RPMI medium containing RPMI 1640 (Invitrogen, Dun Laoghaire, Ireland), 0.0035% β-mercaptoethanol (Roth, Karlsruhe, BW, Germany), 1% L-glutamine (Invitrogen, Dun Laoghaire, Ireland) and 1% penicillin-streptomycin (Invitrogen, Dun Laoghaire, Ireland). Cells were added in a 1:2 ratio to a collagen mix containing PureCol (Advanced biomatrix, San Diego, CA) in MEM (Invitrogen, Dun Laoghaire, Ireland) and 0.4% sodium bicarbonate (Sigma-Aldrich, Buchs SG, Switzerland). The end concentration of collagen gels was 1.6 mg/ml and final cell number for one assay was 4×105 dendritic cells/ml. 5-HT7R stimulation and inhibition was performed by adding 10 µM 5-CT alone or in combination with 10 µM SB 269970. For control measurements, H2O was applied. Collagen–cell mixtures were transferred in a custom-made migration chamber with a thickness of 0.5 to 1 mm. After 30 min, as assembly of the collagen fibers occurred at 37°C, the gels were overlaid with 0.1 µg/ml CCL19 (Peprotech, Rocky Hill, NJ) diluted in RPMI 1640. Low-magnification bright-field movies of 3D collagen chemotaxis assays were recorded for 5 h at 1-min intervals in custom-built climate chambers (5% CO2, 37°C, humidified) and PAL camera (Prosilica, Vancouver, Canada) triggered by custom-made software (SVS Vistek, Seefeld, BY, Germany). Dendritic cells were tracked over 2 h with ImageJ (National Institutes of Health, Bethesda, MD) and the manual tracking plugin Velocity (µm/min) and directionality (ratio between euclidean distance and accumulated distance, 0≤values≤1) parameters were calculated and visualized as plots by analyzing the acquired data with the chemotaxis and migration tool Plugin.
Statistical analyses were performed with Sigma Plot 11.0 (Systat software Inc., San Jose, CA) and GraphPad prism 5 (GraphPad software Inc., San Diego, CA) software using one-way and two-way ANOVA with Bonferroni and Fisher post-hoc corrections, respectively. All experiments were performed three times at least. All data shown as bar graphs represent mean±s.e.m. Statistical differences for the mean values are indicated as follows: *P<0.05; **P<0.01; ***P<0.001.
The authors declare no competing or financial interests.
K.H., E.P., D.G., O.P., A.B. and M.S. conceived the study. K.H., D.G., S.S., M.S., A.B., H.S. and O.P. were responsible for experimental design. K.H., D.G., S.S., M.S. and O.P. performed experiments. K.H., S.S., D.G. and E.P. analyzed the data. K.H., D.G., E.P., S.S., A.B. and M.S. provided reagents, materials and/or analysis tools. K.H., D.G. and E.P. wrote the paper.
These studies were supported by the Deutsche Forschungsgemeinschaft (DFG) [grant number PO732 and REBIRTH project to E.P.].
Supplementary material available online at http://jcs.biologists.org/lookup/suppl/doi:10.1242/jcs.167999/-/DC1
- Received December 23, 2014.
- Accepted June 15, 2015.
- © 2015. Published by The Company of Biologists Ltd