A degradatory fate for CCR4 suggests a primary role in Th2 inflammation

Abstract CCR4 is the sole receptor for the chemokines CCL22 and CCL17. Clinical studies of asthmatic airways have shown levels of both ligands and CCR4+ Th2 cells to be elevated, suggestive of a role in disease. Consequently, CCR4 has aroused much interest as a potential therapeutic target and an understanding of how its cell surface expression is regulated is highly desirable. To this end, receptor expression, receptor endocytosis, and chemotaxis were assessed using transfectants expressing CCR4, CCR4+ human T cell lines, and human Th2 cells polarized in vitro. CCL17 and CCL22 drove rapid endocytosis of CCR4 in a dose‐dependent manner. Replenishment at the cell surface was slow and sensitive to cycloheximide, suggestive of de novo synthesis of CCR4. Constitutive CCR4 endocytosis was also observed, with the internalized CCR4 found to be significantly degraded over a 6‐h incubation. Truncation of the CCR4 C‐terminus by 40 amino acids had no effect on cell surface expression, but resulted in significant impairment of ligand‐induced endocytosis. Consequently, migration to both CCL17 and CCL22 was significantly enhanced. In contrast, truncation of CCR4 did not impair constitutive endocytosis or degradation, suggesting the use of alternative receptor motifs in these processes. We conclude that CCR4 cell surface levels are tightly regulated, with a degradative fate for endocytosed receptor. We postulate that this strict control is desirable, given that Th2 cells recruited by CCR4 can induce the further expression of CCR4 ligands in a positive feedback loop, thereby enhancing allergic inflammation.

Asthma is a chronic condition of the airways that causes significant morbidity and mortality worldwide, despite the availability of effective symptom-relieving treatments. 4 CCR4 has been widely implicated in the pathogenesis of inflammatory diseases such as asthma and atopic dermatitis due to its expression on Th2 cells. [5][6][7] Following allergen provocation of the airways, the endogenous chemokine ligands of CCR4, CCL17/Thymus and activation regulated chemokine (TARC), and CCL22/M -derived chemokine (MDC), are secreted by activated APCs, including dendritic cells, monocytes, and M s. 8 Gradients of these 2 chemokines lead to the recruitment of CCR4-expressing, CD4 + Th2 cells, which infiltrate the lung as part of the late allergic response.
Once there, Th2 cells contribute to the development of inflammation through the production of IL-4, IL-5, and IL- 13. 9 CCL17 and CCL22 purportedly arose from a gene duplication event, yet share only 32% amino acid homology. [10][11][12] Several studies have reported differences in the ability of these 2 chemokines to activate CCR4, providing evidence that they act as biased agonists of this receptor. Mariani et al initially reported that CCL22 but not CCL17 was able to induce CCR4 internalization in Th2 cells and CCR4 transfected cells. 13 Conversely, Ajram et al reported that CCL22 but not CCL17 coupled CCR4 to -arrestin2 recruitment in a transfectant system. 14 Studies of bronchial epithelial cells that also express CCR4 found that CCL17 induces significant transcription of the vasodilatory proteincalcitonin gene-related peptide (CGRP) while CCL22 is ineffective. 15 -CGRP is implicated in asthma pathogenesis, 16 which is significant given that both CCL17 and CCL22 are increased in the bronchoalveolar lavage of asthmatics compared to healthy subjects. 7 In this study, we examined the regulation of CCR4 expression, elucidating structural features of CCR4 that are important for this process.
We found that CCR4 expression is controlled at multiple levels, suggestive of an important role for this receptor in tissue homeostasis and the control of Th2 inflammation, underscoring the potential for CCR4 targeting in the treatment of allergic disease.

METHODS
Unless otherwise stated, reagents were purchased from Thermo Fisher Scientific (Paisley, UK) or Sigma-Aldrich (Poole, UK). Recombinant human CCL17 and CCL22 were purchased from PeproTech EC (London, UK). The anti-CCR4 Ab 10E4 was generated by Millennium Pharmaceuticals and has been previously described. 17 The goat anti-mouse FITC-labeled F (ab') 2 secondary Ab was purchased from Dako Cytomation (Cambridge, UK). The rabbit anti-glyceraldehyde 3-phosphate dehydrogenase (GAPDH) Ab was purchased from Millipore (MA, USA). pcDNA3.1 plasmids encoding either a wildtype (WT) CCR4 construct or a C-terminal truncation mutant of CCR4, CCR4-Δ40, were generated by site-directed mutagenesis as previously described. 18

Human Th2 cells
PBMCs were isolated from the whole blood of healthy donors, as previously described. 19    To obtain a loading control, membranes were stripped and re-probed with GAPDH primary Ab overnight in blocking buffer.

Western blotting
In experiments comparing the constitutive degradation of WT CCR4 and CCR4-Δ40 constructs it was observed that staining of the loading control GAPDH also degraded markedly over time in the presence of cycloheximide. In these experiments, the intensity of CCR4 staining was therefore normalized to the 0-hour time-point for each construct rather than to the GAPDH loading controls, which nonetheless demonstrated consistent protein loading for the first 3 h of the time-course.

Flow cytometry
Cell surface CCR4 expression was measured using a FACS Calibur flow cytometer (BD), as previously described using the 10E4 anti-CCR4 Ab. 21

Chemotaxis
Chemotaxis was assessed as previously described 21 using ChemoTx plates (Neuroprobe, Rockville MD, USA). Briefly, 0.1% BSA in RPMI 1640 was used to generate several chemokine concentrations, which were pipetted into previously blocked wells of the plate. The membrane of the apparatus was subsequently placed on top, and 20 µl of a cell suspension containing 2 × 10 5 cells were added on top of the membrane above each well. After a 5-h incubation at 37 • C and 5% CO 2 , the membrane was removed and cell migration into the well assessed by staining with the live cell dye CellTiter Glo (Promega, Southampton, UK). Data are reported as Chemotactic Index, which is defined as the ratio of chemotactic responses to chemokine to those of media alone.

Statistical analysis
Data are presented as the mean ± SEM of at least 3 independent experiments. All statistical tests were performed using GraphPad Prism software V6 (San Diego, CA) with the appropriate test detailed in the fig.

Cell surface CCR4 and total CCR4 can be detected by the 10E4 Ab with N-linked glycosylation of CCR4 apparent in transfectant systems
Initially, we validated expression of CCR4 on a CHO-K1 derived cell line engineered to express high levels of CCR4 on the cell surface. 14 As anticipated, the anti-human CCR4 mAb 10E4 17,21 was readily able to detect CCR4 on the cell surface (Fig. 1A). The same mAb was also able to detect CCR4 in CHO-CCR4 cell lysates, with a major band running at around 41 kDa and a minor band running at an apparently lower molecular weight (Fig. 1B). We hypothesized that the lower molecu- To strengthen these findings, we also assessed CCR4 expression in L1.2 transfectants transiently expressing CCR4. 10E4 was able to detect CCR4 at the cell surface, albeit at lower levels than in the stable CHO-CCR4 cell line (Fig. 1C). Tunicamycin was seen to reduce the molecular weight of the receptor, again suggestive of N-linked glycosylation ( Fig. 1D). 10E4 was also able to detect CCR4 on the surface of human Th2 cells (Fig. 1E) and in cell lysates from Th2 cells by Western blot (Fig. 1F). Notably, although the band ran at the expected molecular weight, it appeared much less diffuse than in the transfectant systems.
Moreover, the apparent molecular weight of CCR4 was unchanged by tunicamycin treatment, suggesting that N-linked glycosylation of CCR4 is restricted to expression in transfectant systems. Alternatively, given the much lower levels of CCR4 expression in Th2 cells, the 10E4 Ab may be insufficiently sensitive to detect changes in CCR4 via Western blotting.
A key pathway by which GPCR responses are regulated is by means of receptor endocytosis. 3 We therefore set out to characterize the downregulation of CCR4 in response to its endogenous ligands, CCL17 and CCL22. We used the 10E4 mAb to probe downregulation of CCR4 cell surface levels in response to CCL17 and CCL22 on the CHO-CCR4 cell line (Fig. 2). When endocytosis of CCR4 on CHO-CCR4 cells  The addition of cycloheximide had a marked inhibitory effect on receptor replenishment at the cell surface 6 h after the removal of both CCL22 and CCL17 ( Fig. 3B and D). These results were also reproduced in human Th2 cells (Fig. 3C and E) suggesting that CCR4 cell surface replenishment is dependent on de novo protein synthesis.

CCR4 undergoes constitutive internalization in Hut78 and CHO-CCR4 cells
Constitutive receptor internalization has been reported for several chemokine receptors notably CXCR3 which is associated with Th1

Truncation of the C-terminus of CCR4 does not affect cell surface receptor expression but significantly impairs receptor endocytosis and enhances chemotaxis
The intracellular C-terminal region of chemokine receptors is a key region involved in regulating receptor turnover, by virtue of numerous phosphorylation sites which are the target of G protein-coupled receptor kinases (GRKs). 22 To examine the role of this motif in the  (Fig. 5A).
Plasmids encoding either WT CCR4 or CCR4-Δ40 were transiently transfected into L1.2 cells. As shown in Fig. 5B, transfectants expressed both constructs at identical levels on the cell surface, as determined by flow cytometry. The 2 receptors responded differently, however, in receptor endocytosis assays. WT CCR4 underwent significant endocytosis in response to stimulation with both 100 nM CCL17 and 100 nM CCL22 with a 60% reduction in cell surface levels (Fig. 5C). In contrast, no significant endocytosis of the CCR4-Δ40 construct was observed in response to treatment with either chemokine for 30 min (Fig. 5C).
C-terminal CCR4 truncation also had consequences for the chemotactic responses of L1.2 transfectants. Cells expressing the CCR4-Δ40 construct had significantly more efficacious chemotactic responses toward both CCL17 (Fig. 5D) and CCL22 (Fig. 5E) although the responses were still bell-shaped.

C-terminal CCR4 truncation has no effect on constitutive degradation
Given the differences in the internalization responses of wild-type and truncated CCR4, it was of interest to determine whether C-terminal truncation also affects other processes that may influence CCR4 turnover, namely constitutive endocytosis and degradation. L1.  (Fig. 5B). When CCR4 protein was examined over the 6 h time course both constructs exhibited similar patterns of CCR4 degradation over time (Fig. 6B) with the rate of receptor degradation over time directly comparable for both constructs (Fig. 6C). This suggests that the 40-most C-terminal residues of CCR4 do not play a major role in the constitutive endocytosis or degradation of the receptor.

DISCUSSION
Despite the licensing of the CCR5 antagonist maraviroc and the CXCR4 antagonist plerixafor for the inhibition of HIV-1 infection 23 and stem cell mobilization respectively, 24 no chemokine receptor antagonists to date have shown efficacy for the treatment of inflammatory conditions. This is despite intense efforts by many groups In this study, we have shown that CCR4 was endocytosed following exposure to both CCL17 and CCL22 with CCL22 appearing to be the most efficacious ligand when assays were performed in CHO-CCR4 cells. This was not the case in human Th2 cells where although both ligands showed efficacy, there were no significant differences in their ability to downregulate CCR4. This could be an apparent example of We have also shown that CCR4 cell surface levels are strictly regulated. CCR4 is endocytosed in the presence or absence of ligand and once internalized, it is targeted for degradation, being slowly replenished at the cell surface by de novo protein synthesis rather than undergoing receptor recycling. This is in contrast to an earlier study by Mariani and colleagues that found cycloheximide to have no effect on CCR4 cell surface replenishment and concluded that CCR4 is recycled following internalization by ligand. 13 These findings appear to be incompatible with the extremely slow rate of cycloheximide-sensitive replenishment we observed here, using both CCR4 transfected cells and Th2 cells. It is notable that although the authors of the earlier study used a 5-fold higher concentration of cycloheximide that was shown to be effective here, their study lacked a positive control for the efficacy of cycloheximide. Thus, 1 potential explanation for the discrepancies is that the cycloheximide used in the earlier study was ineffective, or inadvertently used at a lower concentration than that reported. Other differences in the 2 protocols used include the source of recombinant The chemokines serve to recruit CCR4 + Th2 cells that secrete the pro-inflammatory cytokines IL-4, IL-5, and IL-13. These cytokines act on eosinophils, airway smooth muscle cells and mucus-producing goblet cells in the airways and result in many of the hallmark features of allergic inflammation. IL-4 also induces the expression of CCL17 and CCL22, thereby perpetuating Th2 cell recruitment by a positive feedback loop. Both chemokines also recruit CCR4 + Treg cells that produce IL-10 and transforming growth factor (TGF)-. These 2 cytokines act to dampen down Th2-induced inflammation. Loss of Treg functions leads to dysregulated Th2 activation, uncontrolled inflammatory responses and the development of autoimmunity. Fine-control of CCR4 signaling by manipulation of cell-surface levels is therefore necessary to effectively maintain homeostasis and control inflammation.
did not rule out the potential for autologous cell production of CCL17 or CCL22 to drive constitutive endocytosis, this would appear to be unlikely given the identical patterns of endocytosis seen in 3 quite different cell lines, the human T cell line Hut78 and the 2 non-human transfectant systems L1.2 and CHO.
Curiously, a consistent finding was the apparent increase in cell surface levels of the Δ40 CCR4 construct at the 1-h time point following cycloheximide treatment (Fig. 6A). This was not supported when total Δ40 protein expression was examined by Western blot (Fig. 6B) suggesting The effects of truncation on ligand-induced endocytosis also closely mirror the effects of CCR4-truncating mutations observed in a study of ATLL in which ligand-induced CCR4 internalization in ATLL cell lines was impaired by C-terminal truncation at glutamine 330 (CCR4-Q330), with a concomitantly enhanced chemotactic response toward CCL17 and CCL22. Importantly, the gain-of-function CCR4 phenotype in ATLL cells gave the cells a competitive advantage over cells expressing wild-type CCR4, leading to increased growth and survival. 40 The parallels between CCR4-Q330 and CCR4-Δ40 could perhaps be predicted given the close proximity of the 2 truncations at positions 320 (CCR4Δ40) and 330 (CCR4-Q330). What is remarkable, however, is that both truncation mutants are still able to signal via G-proteins to drive chemotaxis. We have previously shown that mutation of helix VIII within the C-terminus of CCR4 at lysine 310 (K310N) completely ablated chemotaxis in response to CCL17, while robust chemotaxis was observed in response to CCL22, albeit with reduced potency. 21 These results suggest that, in conjunction with the conserved DRY-LAIV motif within ECL2, the residues within helix VIII (Fig. 5) are also likely to be involved in G-protein coupling and signaling, with an absolute requirement for residue K310 for activation of CCR4 by CCL17.
Why might CCR4 expression be under such strict levels of control? The regulation of CCR4 we observed is reminiscent of that of the Th1-associated receptor CXCR3, 41,42 which our group has previously shown undergoes constitutive endocytosis, degradation, and replenishment at the cell surface by de novo synthesis. 20 Th2 cells are a notable source of IL-4 and IL-13, which induce the expression of CCL17 and CCL22 by monocytes and M s, allowing for a positive-feedback loop in both Th2 and Treg recruitment (Fig. 7), as previously put forward by Bonecchi et al. 41 As such, it may be important to conclusively "switch off" CCR4 signaling to effectively maintain both homeostasis and the control of inflammatory responses by Tregs.
In summary, we have shown that CCR4 expression is subject to multiple levels of regulation, explained by a need to prevent inappropriate activation of a key chemokine receptor that is expressed by both Th2 and Treg cells. CCR4 expression and signaling was tightly regulated by the biased, endogenous ligands CCL17 and CCL22, which induced receptor internalization and degradation, followed by slow cell surface receptor replenishment. Finally, we identified the final 40 amino acids of the C-terminus as a key region involved in receptor endocytosis and intracellular receptor trafficking and signaling. The specific targeting of this region of CCR4 is plausible by the use of previously identified small molecules that recognize this region. 43 It will be interesting to see if such compounds interfere with regulation of CCR4 as might be predicted from this study.

DISCLOSURE
The authors declare no conflict of interest related to the content of this work.