CXCL1 Inhibits Airway Smooth Muscle Cell Migration through the Decoy Receptor Duffy Antigen Receptor for

This information is current as Laila A. Al-Alwan, Ying Chang, Simon Rousseau, James G. of September 23, 2021. Martin, David H. Eidelman and Qutayba Hamid J Immunol 2014; 193:1416-1426; Prepublished online 30 June 2014; doi: 10.4049/jimmunol.1302860

http://www.jimmunol.org/content/193/3/1416 Downloaded from

Supplementary http://www.jimmunol.org/content/suppl/2014/06/30/jimmunol.130286 Material 0.DCSupplemental http://www.jimmunol.org/ References This article cites 62 articles, 13 of which you can access for free at: http://www.jimmunol.org/content/193/3/1416.full#ref-list-1

Why The JI? Submit online.

• Rapid Reviews! 30 days* from submission to initial decision by guest on September 23, 2021 • No Triage! Every submission reviewed by practicing scientists

• Fast Publication! 4 weeks from acceptance to publication

*average

Subscription Information about subscribing to The Journal of Immunology is online at: http://jimmunol.org/subscription Permissions Submit copyright permission requests at: http://www.aai.org/About/Publications/JI/copyright.html Email Alerts Receive free email-alerts when new articles cite this article. Sign up at: http://jimmunol.org/alerts

The Journal of Immunology is published twice each month by The American Association of Immunologists, Inc., 1451 Rockville Pike, Suite 650, Rockville, MD 20852 Copyright © 2014 by The American Association of Immunologists, Inc. All rights reserved. Print ISSN: 0022-1767 Online ISSN: 1550-6606. The Journal of Immunology

CXCL1 Inhibits Airway Smooth Muscle Cell Migration through the Decoy Receptor Duffy Antigen Receptor for Chemokines

Laila A. Al-Alwan,1 Ying Chang,1 Simon Rousseau, James G. Martin, David H. Eidelman, and Qutayba Hamid

Airway smooth muscle cell (ASMC) migration is an important mechanism postulated to play a role in airway remodeling in asthma. CXCL1 has been linked to tissue growth and metastasis. In this study, we present a detailed examination of the inhibitory effect of CXCL1 on human primary ASMC migration and the role of the decoy receptor, Duffy AgR for chemokines (DARC), in this inhibition. Western blots and pathway inhibitors showed that this phenomenon was mediated by activation of the ERK-1/2 MAPK

pathway, but not p38 MAPK or PI3K, suggesting a biased selection in the signaling mechanism. Despite being known as a non- Downloaded from signaling receptor, small interference RNA knockdown of DARC showed that ERK-1/2 MAPK activation was significantly depen- dent on DARC functionality, which, in turn, was dependent on the presence of heat shock protein 90 subunit a. Interestingly, DARC- or heat shock protein 90 subunit a–deficient ASMCs responded to CXCL1 stimulation by enhancing p38 MAPK activation and ASMC migration through the CXCR2 receptor. In conclusion, we demonstrated DARC’s ability to facilitate CXCL1 inhibition of ASMC migration through modulation of the ERK-1/2 MAPK–signaling pathway. The Journal of Immunology, 2014, 193:

1416–1426. http://www.jimmunol.org/

tructural cell migration is a central mechanism involved in viability, would disrupt the active recruitment of cells, potentially a range of physiological conditions, such as wound healing providing an avenue for therapeutic intervention. S (1), neurogenesis (2), and embryogenesis (3). This very Recently, we demonstrated that IL-17–induced CXC chemo- fundamental mechanism may have detrimental consequences for kines, also known as growth-related oncogenes (GROs), CXCL1 tissue structure and function when migration leads to an exces- (GRO-a), CXCL2 (GRO-b), and CXCL3 (GRO-g), had the po- sive accumulation of cells. Tissue fibrosis (4), inappropriate angio- tential to collectively induce ASMC migration (15). Although

genesis (5), cancer cell metastasis (6), and increased airway smooth primarily known as a potent inducer of neutrophil by guest on September 23, 2021 muscle mass in asthma (7) are a few examples of progressively (16), CXCL1 was shown to induce cancer cell invasion (17) and disproportionate cell migration. Attempts have been made to endothelial (18) and epithelial cell migration (19). CXCL1 facil- understand the source of migrating cells (8–12) with the hope of itates its signaling by binding to its putative receptor, the che- better understanding the role of migration in diseases such as mokine receptor CXCR2, although it also was found to bind to asthma, where the origins of migrating cells are not known in vivo CXCR1, albeit with lower affinity (20). Chemokine receptors (13) and, hence, cannot be specifically targeted. However, regard- belong to class A rhodopsin-like, seven-transmembrane, or hep- less of the origins of migrating cells, active recruitment of cells tahelical, G protein–coupled receptors (GPCRs), which in general, from different sites, near or far, is central for disease chronicity and favor coupling to the Gai protein subunit (21). severity (8, 14). Therefore, it is possible to speculate that interfer- Consisting of more than 1000 receptors capable of mediating ence with the machinery of cell migration, without affecting cellular multiple biological functions, GPCRs are amenable targets for drug development (22). The classical view of GPCR signaling requires their association with a heterotrimeric complex of small G pro- Meakins-Christie Laboratories, Respiratory Division, Department of Medicine, McGill University, Montreal, Quebec H2X 2P2, Canada teins (Gabg) through a highly conserved DRYLAIV motif in 1L.A.A.-A. and Y.C. contributed equally to this work. the second loop near the C terminus of the GPCR (23). Atypical Received for publication October 23, 2013. Accepted for publication May 26, 2014. or decoy receptors are a group of seven-transmembrane receptors This work was supported by the Canadian Institutes of Health Research, the Amer- that are unable to recruit small G proteins and, hence, cannot ican Thoracic Society, the Costello Memorial Fund, and the McGill University activate classical GPCR signaling. There are four known atypical Health Centre-Research Institute of the Fonds de la Recherche en Sante´ Que´bec. receptors: D6, CCX-CKR1, CXCR7, and Duffy AgR for chemo- Address correspondence and reprint requests to Dr. Qutayba Hamid, Meakins-Christie kines (DARC) (23). The lack of signaling by these receptors is Laboratories, Respiratory Division, Department of Medicine, McGill University, 3626 Rue St. Urbain, Montreal, QC H2X 2P2, Canada. E-mail address: Qutayba.Hamid@ ascribed either to the complete absence of the DRYLAIV motif, as McGill.ca is the case for DARC, or to the presence of presumably non- The online version of this article contains supplemental material. functional modified forms of this motif (DKYLEIV in D6, DRY- Abbreviations used in this article: 17-AAG, 17-N-allylamino-17-demethoxygeldana- VAVT in CCX-CKR1, and DRYLSIT in CXCR7) (23). DARC mycin; ASMC, airway smooth muscle cell; DARC, Duffy AgR for chemokines; Gai- was first discovered on RBCs as the Duffy blood group Ag (24) PCR; Gai–protein-coupled receptor; GPCR, G protein–coupled receptor; GRO, growth-related oncogene; HSP, heat shock protein; PDGF, platelet-derived growth and was subsequently shown to serve as a receptor for the malarial factor; PI, propidium iodide; Ptx, pertussis toxin; rh, recombinant human; RT, room parasites Plasmodium vivax and Plasmodium knowlesi (25, 26). temperature; siRNA, small interference RNA; siScram, scrambled siRNA. DARC has been established as a promiscuous Copyright Ó 2014 by The American Association of Immunologists, Inc. 0022-1767/14/$16.00 that can bind several chemokines, including CXCL1, to which it www.jimmunol.org/cgi/doi/10.4049/jimmunol.1302860 The Journal of Immunology 1417 binds with the highest affinity (27). The expression of KC (the Small interference RNA transfection mouse ortholog of human CXCL1) and DARC was found to ASMCs were grown to 60–80% confluence in medium containing 10% follow parallel temporal patterns and tissue localization during FBS. The cells were transfected with small interference RNA (siRNA) mouse development (28), suggesting that a relationship may exist targeting CXCR1 and/or CXCR2, DARC, or HSP90a using the siRNA between the two. In general, DARC was found to maintain che- Reagent System (all from Santa Cruz Biotechnology, Santa Cruz, CA). mokine homeostasis in the blood (29) and to facilitate chemokine Scrambled siRNAs (Santa Cruz Biotechnology) were used as a negative control. After transfection, ASMCs were either used directly for migration transcytosis and presentation across endothelium (30, 31). experiments or stimulated with CXCL1 and prepared for Western blotting. In this study, we investigated the effect of CXCL1 on ASMC Protein lysates from transfected, nonstimulated ASMCs were saved to migration and the role of CXCR1, CXCR2, and DARC in medi- evaluate the efficiency of siRNA knockdown by Western blotting using anti- ating this effect. Our results establish that CXCL1 inhibits ASMC CXCR2 (Abcam, Cambridge, MA), anti-CXCR1 (R&D Systems, Minne- apolis, MN), anti-DARC (Abcam), or anti-HSP90a (Abcam) Abs. GAPDH migration through activation of the ERK-1/2 MAPK pathway (Millipore, Billerica, MA) was used as a loading control. and that this effect is mediated via DARC but not the CXCR1 or CXCR2 receptors. In addition, we observed a regulatory effect of Immunofluorescence staining the heat shock protein (HSP)90a on DARC protein expression in ASMCs were seeded on a Lab-Tek Chamber Slide System (Thermo Fisher ASMCs. Finally, we found that, in the absence of DARC, CXCL1 Scientific, Rochester, NY) and left to adhere overnight at 37˚C, after engaged CXCR2 and enhanced ASMC migration through p38 which they were treated with 4 ng/ml rhCXCL1 for 30 min, washed with MAPK/HSP27 pathway activation. 13 PBS, and fixed with 4% paraformaldehyde. To eliminate nonspecific binding, cells were incubated with Protein Block (Dako Canada, Bur- lington, ON, Canada) for 1 h at room temperature (RT). Subsequently, cells Materials and Methods were treated with 1:250 of rabbit monoclonal anti-DARC (Abcam) and Downloaded from Culture of human ASMCs mouse monoclonal anti-HSP90a (Abcam) Abs diluted in Ab Diluent (Dako Canada) at 4˚C overnight. Cells were washed with 0.1% Tween 20/ Human ASMCs were isolated from lung transplant donors and prepared as PBS and incubated with 1:300 Alexa Fluor 555–conjugated donkey anti- described previously (32). The protocol was approved by institutional re- rabbit IgG (highly cross-adsorbed) and Alexa Fluor 488–conjugated goat search ethics boards of the McGill University Health Centre and the anti-mouse IgG (Invitrogen, Camarillo, CA) for 1 h at RT. After washing, Universite´ de Montre´al, Montreal, QC, Canada and informed consent for cells were stained with the nuclear staining Hoechst 33342 (AnaSpec, ASMC harvesting was obtained at surgery or at the time of lung trans- Fremont, CA) for 10 min. Using laser scanning confocal microscopy http://www.jimmunol.org/ plantation. Cultured cells were identified as smooth muscle cells by im- (LSM780; Carl Zeiss Microscopy, Jena, Germany), stained ASMCs were munohistochemical staining for smooth muscle specific a-actin, myosin H viewed by a 633/1.40 oil DIC “Plan Apochromat” objective, and images chain, and calponin by Western blotting and flow cytometry. ASMCs were were analyzed using ImageJ (36) (National Institutes of Health, Bethesda, cultured in DMEM/F12 (1:1; Life Technologies/Invitrogen, Grand Island, MD). Negative controls were performed using 1:250 of species-specific NY) supplemented with 10% FBS and 100 U/ml penicillin/streptomycin isotype Ab. (Clonetics, Walkersville, MD) at 37˚C with 5% CO2 and were used be- tween passages 4 and 8. Prior to each experiment, ASMCs were starved for Western blot for signaling pathway analysis 24 h in DMEM/F12 culture medium supplemented with 0.1% FBS and 100 U/ml penicillin/streptomycin. The primary ASMCs from asthmatic subjects Normal ASMCs, or transfected ASMCs where applicable, were treated with 3 used in our protocols were purchased from Lonza (Basel, Switzerland). 4 ng/ml rhCXCL1 for 0, 5, 15, 30, 45, or 60 min, washed twice with cold 1 PBS, and lysed. Whole-cell lysates (10–15 mg) were loaded on 10% ac- by guest on September 23, 2021 Migration assay rylamide SDS-PAGE NEXT GEL (AMRESCO, Solon, OH), followed by transfer to nitrocellulose membranes (Bio-Rad, Hercules, CA). The blots ASMC migration was assessed as previously described (33) using a 48-well were blocked for 1 h at RT and incubated for 24 h at 4˚C with Abs specific microchemotaxis Boyden chamber (Neuro Probe, Cabin John, MD). to total ERK1/2 MAPK and phosphorylated ERK1/2 MAPK (42 and 44 Briefly, different concentrations of recombinant human (rh)CXCL1, alone kDa; both from ), phosphorylated p38 MAPK (38 kDa; or in combination with CXCL2 and/or CXCL3, were added to the lower Millipore), GAPDH (38 kDa; Millipore), phosphorylated HSP27 (27 kDa; 6 chamber, and 1 3 10 cells/ml normal (or asthmatic, where applicable) Cell Signaling), phospho-Ser473 Akt (∼60 kDa; Cell Signaling) for PI3K ASMCs were added to the upper chamber. A 10-mm-pore polycarbonate pathway activation, and b-arrestin 2 (50 kDa; Santa Cruz Biotechnology). membrane (Neuro Probe) treated with 0.01% collagen type I (rat tail) After washing (0.1% Tween 20/PBS), the membranes were incubated with (Nalgene Culture 3D Matrix: Nalgene, Rochester, NY) was placed be- a 1:15,000 dilution of IRDye 680 goat anti-mouse IgG and IRDye 680 goat tween the two chambers. After 4 h, the membrane was removed, the upper anti-rabbit IgG (Rockland, Philadelphia, PA) in blocking buffer and ana- face (containing nonmigrated cells) was wiped, and the cells that had lyzed with an Odyssey IR scanner using Odyssey imaging software 3.0 migrated to the lower side of the membrane were fixed with 4% parafor- (LI-COR Biosciences). maldehyde and stained using a Hema 3 Kit (Fisher, Kalamazoo, MI). The number of migrated cells was counted in five random fields at an original Apoptosis assays magnification 3200. Viability of ASMCs after treatment with pathway inhibitors or different The role of CXCR1 and/or CXCR2 receptors or other concentrations of CXCL1 (0.25–4 ng/ml) or after siRNA transfection was G(ai)-coupled receptors in CXCL1 inhibition of ASMC measured using flow cytometry (BD FACSCalibur; BD Biosciences) migration for annexin V and propidium iodide (PI) staining. Briefly, ASMCs from healthy controls, and asthmatic subjects where applicable, were treated ASMCs were treated for 1 h at 37˚C with 10 mg/ml anti-CXCR1 and/or with 1) DMSO or PD184352 for 1hr at 37˚C or 2) different concentrations anti-CXCR2 neutralizing Abs (R&D Systems, Minneapolis, MN) or with of CXCL1 (0.25–4 ng/ml) for 4 h at 37˚C or 3) transfected with different 10, 50, or 100 ng/ml pertussis toxin (Ptx; TOCRIS Bioscience, Bristol, siRNAs (scrambled siRNA [siScram] controls, siCXCR1, siCXCR2, U.K.) prior to the addition of CXCL1 to the Boyden chamber. Species- siDARC or siHSP90a). Subsequently, cells were trypsinized, washed, and specific isotype controls were used as controls for the neutralizing Abs. Ptx stained using an annexin V and PI staining kit (Annexin V:FITC Apoptosis was reconstituted in sterile distilled water. Detection Kit; BD Biosciences) to quantify the percentage of cells un- dergoing apoptosis, following the manufacturer’s suggested protocol. The role of ERK-1/2 MAPK–signaling pathway and HSP90 in There was no evidence of cytotoxicity in ASMCs treated with the inhib- CXCL1 inhibition of ASMC migration itors or the siRNAs at the concentrations used (data not shown). ASMCs were treated for 1 h at 37˚C with either ERK-1/2 MAPK inhibitor Assessment of HSP90a and HSP90b mRNA by quantitative PD184352 (2 mM; US Biological, Swampscott, MA) (34) or with different real-time PCR concentrations (1, 10, or 100 nM) of 17-N-allylamino-17-demethoxygel- danamycin (17-AAG), a low-toxicity profile pan inhibitor of HSP90 (35) Serum-starved ASMCs were treated with 4 ng/ml rhCXCL1 for 0.5, 1, or 3 (AG Scientific, San Diego, CA) prior to the addition of CXCL1 to the h, and total RNA was extracted using the RNeasy Mini kit (QIAGEN; Boyden chamber. Controls were exposed to the same concentration of Mississauga, ON, Canada). Quality control and RNA concentrations were DMSO. determined using a spectrophotometer (NanoDrop Products, Wilmington, 1418 CXCL1 INHIBITS ASMC MIGRATION VIA DARC

DE). A total of 40 ml reaction mix containing 250 ng total RNA, 1 ml ASMC migration was not sensitive to the effect of neutralizing Abs reverse transcriptase Superscript II (Invitrogen), 0.5 ml RNase inhibitor, 8 ml (Fig. 2A) or siRNA knockdown (Fig. 2B) of CXCR1 and CXCR2 3 5 first-strand buffer (Invitrogen), 4 ml0.1MDTT,0.5mM each deoxy- receptors, or to the effect of Ptx, which inhibits Gai signaling ribonucleotide triphosphate, and 0.5 mg oligo dT12–18 primers (Invitrogen) was prepared. The reaction program was 37˚C for 10 min, 42˚C for 50 min, (40) (Fig. 2C), suggesting that it is unlikely for CXCR1, CXCR2, 45˚C for 10 min, and then cooling at 4˚C. cDNA was amplified using the and other Gai–protein-coupled receptors (Gai-PCRs) to be in- 7500 Fast Real-Time PCR system thermal cycler, using SYBR Green (both volved in mediating CXCL1 inhibition of ASMC migration. from Applied Biosystems, Foster City, CA) and the following primers: DARC, a decoy receptor that binds to CXCL1 with high affinity HSP90a (forward 59-GGC AGA GGC TGA TAA GAA CG-39 and reverse 59-CGT GAT GTG TCG TCATCT CC-39), HSP90b (forward 59-GAA CCA (27), was shown to be expressed in RBCs (24), endothelial cells TTG CCA AGT CTG GT-39 and reverse 59-ACG CAC AGT GAA GGA (30), brain (41), and epithelial cells (42), but it has never been ACC TC-39), and GAPDH (forward 59-AGT CAA CGG ATT TGG TCG described in ASMCs. After establishing DARC expression in TAT T-39 and reverse 59-ATG GGT GGA ATC ATA TTG GAA C-39). The ASMCs by immunofluorescence staining of cultured normal ASMCs HSP90 mRNA expression was normalized to GAPDH and compared using (Fig. 2D), we investigated its involvement in CXCL1-mediated the DDCt method. All results were expressed as relative quantity compared with nonstimulated controls. inhibition of ASMC migration using siRNA knockdown (siDARC). Transfection of ASMCs with siDARC eliminated the inhibitory Statistical analysis effect of CXCL1 on ASMC migration and actually led to en- All data are presented as mean 6 SEM. Statistical analysis was performed hanced ASMC migration compared with the siSram control using ANOVA, followed by the post hoc Dunnett test. The Bonferroni (siScram: 0.64-fold 6 0.06, siDARC: 1.4-fold 6 0.08; p , 0.001, multiple-comparison post hoc test was performed for statistical analysis Fig. 2E), indicating that DARC facilitates CXCL1-mediated in- , within and between stimulated conditions. The p values 0.05 were hibition of ASMC migration. Moreover, in the absence of DARC, Downloaded from regarded as statistically significant. Statistical analysis was performed using GraphPad Prism version 5.00 for Windows (GraphPad Software, San CXCL1 was able to enhance ASMC migration over that of control Diego, CA; http://www.graphpad.com). levels, suggesting the activation of alternate signaling mecha- nisms, possibly through other receptor(s). The siRNA knockdown efficiency reached 45% for CXCR1, Results 49% for CXCR2, and 56% for DARC, as measured by Western

CXCL1 is a negative regulator of ASMC migration blotting (Supplemental Fig. 2A, 2B, 2C, respectively). http://www.jimmunol.org/ Recently, we showed that CXCL2 and CXCL3 induced significant CXCL1 inhibits ASMC migration through selective activation migration of ASMCs over a range of concentrations (37). Using of ERK1/2 MAPK, but not p38 or PI3K, pathways similar approaches, we sought to investigate the effect of CXCL1 on ASMC migration and found that, over a range of 10-fold in- DARC, which lacks the ability to recruit small G proteins, has never creasing concentrations, CXCL1 did not appear to affect normal been shown to participate in signaling cascades (23). However, in or asthmatic ASMC migration (Supplemental Fig. 1A, 1B, re- light of our observation that DARC mediates the inhibitory effect of spectively). However, by exploring a narrower range of concen- CXCL1 on ASMC migration, we sought to test whether treatment trations (0.25, 0.5, 1, 2, and 4 ng/ml), it was possible to demonstrate with CXCL1 activated any of the signaling pathway(s) (ERK-1/2, that CXCL1 could significantly inhibit both normal and asthmatic p38, or PI3K) in ASMCs. We found that, although CXCL1 induced by guest on September 23, 2021 ASMC migration at 2 ng/ml (normal: 0.52-fold 6 0.07, p , 0.01; significant activation of ERK1/2 MAPK starting at 5 min (10.22- asthmatic: 0.52-fold 6 0.11, p , 0.01) and 4 ng/ml (normal: 0.48- fold 6 1.7; p , 0.001) and continuing for up to 60 min (4.34-fold 6 fold 6 0.11, p , 0.001; asthmatic: 0.52-fold 6 0.10, p , 0.01) 0.89; p , 0.05) after stimulation (Fig. 3A), the p38 MAPK and (Fig. 1A, 1B, respectively). The inhibitory effect of CXCL1 was PI3K pathways were not activated (Fig. 3B, 3C, respectively). We not attributable to toxicity (Supplemental Fig. 1C, 1D). These used the pharmacological inhibitor PD184352 to further establish results indicate that CXCL1 inhibits ASMC migration over a very a possible role for the ERK1/2 MAPK pathway in CXCL1- narrow range of concentrations, without affecting cell viability. mediated inhibition of ASMC migration. Similar to the effect of We previously established that the effect of CXCL2 and CXCL3 siDARC, inhibition of the ERK1/2 MAPK pathway reversed the on ASMC migration was not significantly different from platelet- effect of CXCL1 on ASMC migration from inhibitory to stimula- derived (PDGF)-BB (37), one of the known potent tory (DMSO: 0.74-fold 6 0.05, PD184352: 1.334-fold 6 0.08, p , mediators of ASMC migration (38). Therefore, we sought to in- 0.001, Fig. 3D). These results indicate that CXCL1 inhibits ASMC vestigate the extent of the inhibitory effect of CXCL1 on ASMC migration through activation of the ERK1/2 MAPK pathway, but migration by adding CXCL1 to CXCL2 and/or CXCL3 in equal not the p38 MAPK or PI3K pathway, suggesting a biased selection concentrations, using normal ASMCs. We found that CXCL1 in the activation of signaling cascades. inhibits CXCL2-induced ASMC migration at all concentrations siDARC inhibits ERK-1/2 MAPK while enhancing p38 MAPK tested, with the exception of 0.25 ng/ml (Fig. 1C). It also inhibits and HSP27 activation in ASMCs in response to CXCL1 CXCL3- and CXCL2/CXCL3-induced ASMC migration at final concentrations of 2 and 4 ng/ml (Fig. 1D, 1E, respectively). In- Administration of siDARC led to a significant reduction in ERK1/2 terestingly, however, the addition of 4 ng/ml of CXCL1 to 5 ng/ml MAPK activation compared with siScram, beginning at 5 min of PDGF-BB significantly enhanced PDGF-BB–induced ASMC (siScram: 9.36-fold 6 0.53, siDARC: 5.33-fold 6 0.36; p , migration (Fig. 1F). These results indicate that CXCL1 may act as 0.001) and persisting for up to 60 min (siScram: 6.06-fold 6 0.66, an inhibitor of CXCL2- and/or CXCL3-enhanced ASMC migra- siDARC: 3.37-fold 6 0.36; p , 0.01) after stimulation (Fig. 4A). tion but not PDGF-induced ASMC migration. We used siCXCR1 and siCXCR2 to exclude the possibility that CXCR1 and CXCR2 play roles in ERK1/2 MAPK pathway acti- CXCL1 inhibition of ASMC migration is mediated through vation and found no significant difference compared with siScram a DARC but not CXCR1, CXCR2, or other G i–protein-coupled (Supplemental Fig. 3). These findings suggest that DARC, but not receptors CXCR1 and/or CXCR2, mediates CXCL1-induced ERK1/2 MAPK Because CXCL1 was shown to bind to both CXCR1 and CXCR2 activation. (39), we investigated the role of both receptors in CXCL1-mediated To investigate the signaling pathway(s) responsible for CXCL1- inhibition of ASMC migration. We found that CXCL1 inhibition of enhanced ASMC migration in the absence of DARC, we sought to The Journal of Immunology 1419 Downloaded from http://www.jimmunol.org/ by guest on September 23, 2021

FIGURE 1. CXCL1 is a negative regulator of ASMC migration. A Boyden chamber was used to investigate the effect of low doses (0.25–4 ng/ml) of CXCL1 alone on normal (A) or asthmatic (B) ASMC migration or the effect of equal concentrations of CXCL1/CXCL2 (C), CXCL1/CXCL3 (D), and CXCL1/CXCL2/CXCL3 (E) or the addition of 4 ng/ml of CXCL1 to 5 ng/ml of PDGF-BB (F) on normal ASMC migration. CXCL1 was able to inhibit ASMC migration alone or when combined with CXCL2 and/or CXCL3 (n = 5–9 subjects). Data are mean 6 SEM. *p , 0.05, **p , 0.01, ***p , 0.001 versus nonstimulated control (A and B) and between similar conditions (C–E). examine the activation of the PI3K and p38 MAPK pathways in by Western blotting. Administration of CXCL1 to siDARC-ASMCs CXCL1-stimulated ASMCs after siDARC. We found a significant significantly enhanced the protein expression of CXCR2 (Fig. 5B). In increase in p38 MAPK activation in ASMCs at 5 min (siScram: 1.32- contrast, CXCR1 protein expression was not affected (Fig. 5A). We fold 6 0.17, siDARC: 2.32-fold 6 0.40; p , 0.01) and 15 min next examined the effect of CXCL1 on ASMC migration and p38 (siScram: 0.98-fold 6 0.18, siDARC: 2-fold 6 0.28; p , 0.01) after MAPK activation after simultaneous double siRNA knockdown of stimulation (Fig. 4B). In contrast, the PI3K pathway was not acti- DARC and CXCR2 (siDARC/CXCR2) and found that it signifi- vated significantly (Fig. 4C). Activation of HSP27 has been estab- cantly blunted both the inhibitory and the enhancing effects of lished as a regulator of cell migration downstream of the p38 CXCL1 on ASMC migration compared with siScram and to MAPK pathway in both endothelial cells (43) and ASMCs (44). siDARC, respectively, (siScram: 0.66 6 0.03, siDARC: 1.45 6 0.09, Therefore, we examined HSP27 in response to CXCL1 after siDARC/CXCR2: 0.9 6 0.04; p , 0.001, Fig. 5C). In addition, it siDARC and found that it also was activated at 5 min (siScram: reduced p38 MAPK enhanced activation to control levels (Fig. 5D). 0.9-fold 6 0.35, siDARC: 1.79-fold 6 0.21; p , 0.05) and 15 min These results indicate that, in the absence of DARC, CXCL1 (siScram: 0.81-fold 6 0.25, siDARC: 1.73-fold 6 0.27; p , 0.05) enhances ASMC migration through CXCR2/p38 MAPK pathway after stimulation (Fig. 4D). These results suggest that, in the ab- activation. sence of DARC, CXCL1-enhanced ASMC migration is likely to siHSP90a mediates DARC protein expression and CXCL1 be mediated through the p38 MAPK/HSP27 pathway. inhibition of ASMC migration In the absence of DARC, CXCR2 mediates CXCL1-enhanced HSP90 is a chaperone protein involved in many cellular processes ASMC migration and p38 MAPK pathway activation and is known to act as an essential molecular chaperone of GPCR- As a first step in investigating which receptor(s) mediates CXCL1- and non–GPCR-mediated ERK1/2 phosphorylation (45, 46). enhanced ASMC migration and p38 MAPK activation after Therefore, we sought to investigate the possible role of HSP90 siDARC, we examined the protein expression of CXCR1 and CXCR2 in CXCL1/DARC-mediated ERK1/2 MAPK pathway activation 1420 CXCL1 INHIBITS ASMC MIGRATION VIA DARC Downloaded from http://www.jimmunol.org/

FIGURE 2. CXCL1 inhibition of ASMC migration is mediated through DARC but not CXCR1, CXCR2, or other Gai-PCRs. ASMCs were incubated with Abs or Ptx (1 h) or transfected with siRNA prior to the designated experiments. Blockade of CXCR1, CXCR2, or Gai-PCR signaling with neutralizing Abs (A), siRNA transfection (B), or Ptx (C) did not reverse CXCL1-mediated inhibition of ASMC migration. (D) DARC is expressed in ASMCs (one

representative subject of four is shown). Original magnification 363. (E) Blockade of DARC function using siDARC reversed the effect of CXCL1 on by guest on September 23, 2021 ASMC migration from inhibitory to stimulatory. Results are expressed as mean 6 SEM (n = 3–5 subjects). **p , 0.01, ***p , 0.001 versus nonstimulated control; $$$p , 0.001 versus scrambled siRNA control. and ASMC migration inhibition. Using 17-AAG, we found that The siRNA knockdown efficiency reached 68% for HSP90a, CXCL1 inhibition of ASMC migration was reversed to control as measured by Western blotting (Supplemental Fig. 2D). levels at 10 and 100 nM (Fig. 6A), supporting a possible role for HSP90 in CXCL1-mediated ASMC migration inhibition. How- Discussion ever, there are two isoforms of HSP90, HSP90a and HSP90b (47, Although cell migration is a fundamental mechanism involved in 48), which are known to differentially regulate cellular processes many biological processes, when dysregulated or inappropriate in (49). When we examined the mRNA expression of HSP90a and magnitude it could lead to extremely damaging outcomes, such HSP90b in ASMCs in response to 4 ng/ml of CXCL1, we found as airway remodeling in asthma (7). Directed cell migration, also a significant increase in the mRNA expression of HSP90a, but not known as chemotaxis, is a complex process initiated by a con- HSP90b, after 30 and 60 min of stimulation (Supplemental Fig. centration gradient of stimulus sensed by specific receptors, fol- 4). Similar to siDARC, administration of siHSP90a changed the lowed by activation of signaling pathways, cellular polarization, effect of CXCL1 on ASMC migration from inhibitory to stimu- recycling of actin filaments and actin binding proteins, and re- latory (Fig. 6B), and it significantly reduced ERK1/2 MAPK ac- cruitment of cell–cell and cell–matrix adhesion proteins (50, 51). tivation (Fig. 6C) while increasing p38 MAPK activation at 5 and There has been little progress in understanding the underlying 15 min (Fig. 6D) after treatment. These results suggest that the mechanisms driving cell migration, and there are few examples of HSP90a isoform plays a role in CXCL1-induced ERK1/2 MAPK molecules capable of its inhibition (52, 53). In this study, we pathway activation and ASMC migration inhibition. demonstrate that CXCL1 is a physiologically significant inhibitor Given that siHSP90a produced similar results to siDARC, we of both basal and stimulated ASMC migration. We also show that postulated that some relationship may exist between the two proteins. the DARC receptor is able to modulate this inhibition through its Using confocal microscopy, we found that, in the presence of effect on the activity of the ERK-1/2 MAPK–signaling pathway. HSP90a, treatment of ASMCs with CXCL1 increased DARC re- Previously, we investigated the effect of chemokines on ASMC cruitment at the cell periphery, near the cell surface (Fig. 6E). Con- migration and found evidence that the range of concentrations in versely, siHSP90a caused near-complete disappearance of DARC conventional use (0.01–100 ng/ml) may not accurately reflect the expression, which was not rescuedbyCXCL1treatment(Fig.6F). capacity of chemokines to induce ASMC migration (37). More- These results indicate that HSP90a may be important for the regu- over, depending on disease status, the concentrations of CXCL1 lation and/or the stability of DARC protein expression in ASMCs. measured in biological fluids of humans or animal models do not The Journal of Immunology 1421 Downloaded from http://www.jimmunol.org/

FIGURE 3. CXCL1 inhibits ASMC migration through selective activation of the ERK1/2 MAPK pathway but not the p38 or PI3K pathways. Whole-cell lysates from ASMCs stimulated with 4 ng/ml of CXCL1 for 0, 5, 15, 30, 45, or 60 min were used for Western blot. (A) Activation of the ERK-1/2 MAPK pathway was strongly detected in ASMCs after 5 min and for up to 60 min after CXCL1 stimulation. In contrast, the p38 MAPK (B) and PI3K (C) pathways by guest on September 23, 2021 were not significantly activated. (D) Inhibition of the ERK-1/2 MAPK pathway using the pharmacological inhibitor PD184352 significantly reversed the effect of CXCL1 on ASMC migration from inhibitory to stimulatory. Data are mean 6 SEM (n = 4–7 subjects). *p , 0.05, **p , 0.01, ***p , 0.001 versus nonstimulated control; $$$p , 0.001 versus vehicle control (DMSO). –, CXCL1 only. exceed 10 ng/ml (18, 54, 55). Therefore, using a narrower range of hanced the latter’s effect on ASMC migration (data not shown). concentrations of CXCL1 to examine its effect on ASMC mi- The discrepancy in the combinatorial effect of CXCL1 on ASMC gration may better reflect its biological significance in asthmatic migration also was shown with primary oligodendrocyte progen- airways remodeling. In this article, we show that, over a narrower itor cells (58), albeit using different chemotactic meditators, range of concentrations, CXCL1 exhibited an inhibitory effect on namely PDGF-A and fibroblast growth factor-2. Using these normal and asthmatic ASMC migration, an outcome that was not mediators, Vora et al. (58) showed that CXCL1 was able to inhibit observed using the conventional concentrations. In support of our PDGF-induced, but not fibroblast growth factor-2–induced, oligo- finding that CXCL1 is capable of inhibiting ASMC migration in dendrocyte progenitor cell migration, and that the effect of CXCL1 vitro, a very recent study on a murine model of prostate cancer was concentration dependent. Taken together, these findings indi- demonstrated that in vivo engraftment of cancer cells capable of cate that CXCL1 is capable of inhibiting cell migration alone or in producing CXCL1 (TRAMP-G2LG) inhibited tumor growth and the presence of other chemotactic mediators and that its inhibitory invasion, in contrast with cell lines that do not produce CXCL1 effect may not be restricted to ASMCs. Moreover, our results and (TRAMP-G2) (56). Interestingly, these investigators found that the those of other investigators demonstrate the complexity of the in- production of CXCL1 from TRAMP-G2LG cancer cells reached teraction between CXCL1 and other chemotactic mediators in dif- ∼4 ng/ml, similar to the concentration of CXCL1 that inhibited ferent cell types in which the inhibitory effect of CXCL1 is likely to ASMC migration in our study. These results suggest that, although be governed by many factors, including, but not limited to, cell known as a mediator of cell chemotaxis (16, 17), at specific type, the chemotactic mediators used, and the concentrations of concentrations CXCL1 may act as a negative regulator of cell CXCL1 that were used in these combinations. migration/invasion both in vitro and in vivo. Although CXCL1 is known to specifically signal through We also found that combining CXCL1 with CXCL2 and/or CXCR2 and, in some instances, CXCR1 (20), we were able to CXCL3 in equal concentrations effectively inhibited CXCL2- demonstrate, using neutralizing Abs, siRNA transfection, and in- and/or CXCL3-induced ASMC migration. These results are sup- hibition by Ptx, that the inhibitory effect of CXCL1 on ASMC ported by a recent report demonstrating that both endogenously migration was not mediated through CXCR1, CXCR2, or any produced and the exogenous addition of CXCL1 equally inhibited other G(ai)-PCR. Instead, our findings indicate that CXCL1 CXCL8- and CXCL10-induced mast cell migration (57). Con- inhibited ASMC migration via DARC. This finding is of particular versely, combining CXCL1 with PDGF-BB significantly en- interest given that a recent study of three populations of African 1422 CXCL1 INHIBITS ASMC MIGRATION VIA DARC Downloaded from http://www.jimmunol.org/ by guest on September 23, 2021

FIGURE 4. siDARC inhibits ERK-1/2 MAPK while enhancing p38 MAPK and HSP27 activation in ASMCs in response to CXCL1. Whole-cell lysates from siDARC ASMCs treated with 4 ng/ml of CXCL1 for 0, 5, 15, 30, 45, or 60 min were used for detection of pathway activation by Western blotting. siDARC significantly reduced CXCL1-induced ERK-1/2 MAPK pathway activation (A) while enhancing p38 MAPK pathway (B) and downstream HSP27 (D) activation. (C) In contrast, there was no change in PI3K pathway activation before or after siDARC. Representative blots of ERK-1/2 MAPK, p38 MAPK, and HSP27 activation in response to CXCL1 after siScram (E) or siDARC (F). Results are expressed as mean 6 SEM (n = 4–5 subjects). *p , 0.05, **p , 0.01, ***p , 0.001 versus nonstimulated controls; $p , 0.05, $$p , 0.01, $$$p , 0.001 versus siScram. descent found a relationship between a functional polymorphism with CXCL1 significantly induced ERK-1/2 MAPK pathway ac- in the DARC gene and an increased susceptibility to asthma (59). tivation, which mediated CXCL1 inhibition of ASMC migration. Moreover, in several types of cancer, increased DARC expression However, administering CXCL1 to ASMCs after siDARC resulted was associated with decreased tumor metastasis and tumor size, as in a significant reduction in ERK-1/2 MAPK activity; concomi- well as higher survival rates (60, 61). Our results indicate that the tantly, we observed enhanced activity of p38 MAPK and down- inhibitory effect of CXCL1 is mediated through DARC and sug- stream HSP27, a signaling pathway that has long been established gest that DARC may play a protective role in migration-dependent as a modulator of cell migration (44). Interestingly, we found this diseases, such as asthma and cancer. enhancing effect to be mediated through CXCR2 in DARC- It is widely accepted that, because of its lack of the DRYLAIV deficient ASMCs. Of interest, binding studies using a mutant motif, DARC does not induce canonical GPCR signaling (23). form of CXCL1, named E6A, demonstrated that it was able to Nevertheless, our finding that CXCL1’s inhibition of ASMC mi- bind to DARC on RBCs and inhibit malaria parasite invasion as gration was mediated through DARC suggested the possibility that efficiently as the wild-type CXCL1. However, it was not able to DARC contributes to signaling initiation. Indeed, treating ASMCs bind to CXCR2 on neutrophils (62), suggesting the possibility that The Journal of Immunology 1423 Downloaded from http://www.jimmunol.org/ by guest on September 23, 2021

FIGURE 5. In the absence of DARC, CXCR2 mediates CXCL1-enhanced ASMC migration and p38 MAPK pathway activation. ASMCs were incubated with Abs (1 h) or transfected with siRNA to knockdown either DARC (siDARC) or DARC and CXCR2 (siDARC/CXCR2) expression and, where applicable, were stimulated with 4 ng/ml of CXCL1 prior to the designated experiments. Administering siDARC significantly increased the protein expression of CXCR2 (B), but not CXCR1 (A), in response to CXCL1, whereas administration of siDARC/CXCR2 significantly reduced ASMC migration (C)andp38MAPKactivation(D)in ASMCs in response to CXCL1. (E) Representative blots of p38 MAPK activation after siScram, siDARC, or siDARC/CXCR2. Results are expressed as mean 6 SEM (n = 4–5 subjects). *p , 0.05, **p , 0.01, ***p , 0.001 versus nonstimulated controls; $p , 0.05, $$p , 0.01, $$$p , 0.001 versus siScram or siDARC.

DARC may compete with CXCR2 for the binding of CXCL1. In function, either through competition or by influencing its expres- addition, CCX-CKR, another decoy receptor, was reported to in- sion, conformation, or binding capacity to CXCL1. hibit CXCR3-mediated chemotaxis of cotransfected HEK293 There has been recent interest in the ability of GPCRs to signal cells by forming heterodimeric complexes with CXCR3 and sub- through both G protein and arrestin pathways in a ligand-specific sequently affecting its binding properties (63). Accordingly, it is manner. The latter mechanism, termed “bias signaling,” primarily reasonable to speculate that DARC may also form complexes with activates the ERK1/2 MAPK pathway and is specifically regulated the CXCR2 receptor and, consequently, prevents its binding to by b-arrestin 2, independent of G protein coupling (64–66). Using CXCL1. However, in the aforementioned study, the investigators Western blotting, we could not detect b-arrestin 2 protein ex- did not find any difference in CXCR3 expression in HEK293 pression in ASMCs with or without CXCL1 stimulation (data not cells, whether or not they were cotransfected with CCX-CKR (63). shown), suggesting that b-arrestin 2 is unlikely to mediate In contrast, we observed an increase in CXCR2 protein expression CXCL1/DARC-induced ERK-1/2 MAPK pathway activation. in DARC-deficient ASMCs upon treatment with CXCL1, sug- Moreover, neither inhibition of src nor blockade of epidermal gesting that the presence of DARC may have a regulatory influ- growth factor receptor, two pathways known to mediate ERK-1/2 ence on CXCR2 expression. Moreover, CCX-CKR was only able MAPK pathway activation (67) using the pharmacological in- to inhibit CXCR3/CXCL9- and CXCR3/CXCL10-enhanced che- hibitor PP2 (68) and PD153035 (69), was able to inhibit CXCL1/ motaxis. In contrast, DARC/CXCL1 inhibited both basal and en- DARC-mediated ERK-1/2 MAPK pathway activation (data not hanced ASMC migration, suggesting that this inhibition is probably shown), suggesting that they may not be involved in this activa- regulated through a different mechanism. Taken together, our results tion. Instead, we chose to examine the effect of HSP90, a chap- establish that ERK1/2 MAPK is the sole pathway mediating erone protein that was found to be involved in the activation of CXCL1 inhibition of ASMC migration and that, in the absence of ERK1/2 MAPK pathway in vascular smooth muscle cells through DARC, CXCL1 enhances ASMC migration through the CXCR2/ direct association with phospho-ERK1/2 (46). We found that in- p38MAPK/HSP27-signaling pathway. Moreover, they suggest that hibition of HSP90, specifically HSP90a, gave similar results to the presence of DARC in ASMCs may negatively regulate CXCR2 those obtained with siDARC. Surprisingly, however, the observed 1424 CXCL1 INHIBITS ASMC MIGRATION VIA DARC Downloaded from http://www.jimmunol.org/ by guest on September 23, 2021

FIGURE 6. siHSP90a mediates DARC protein expression and CXCL1 inhibition of ASMC migration. Serum-starved ASMCs were incubated with 17- AAG (1 h) or Abs (1 h) or were transfected with siHSP90a and, where indicated, were stimulated with 4 ng/ml of CXCL1 prior to the designated experiments. Administration of 10 or 100 nM of 17-AAG reversed ASMC migration to control levels (A), whereas siHSP90a enhanced ASMC migration (B), inhibited ERK-1/2 MAPK activation (C), and enhanced p38 MAPK activation (D) in response to CXCL1. (E) Using immunofluorescence confocal microscopy we observed, in the presence of HSP90a, increased DARC expression at the cell periphery after CXCL1 stimulation (yellow arrows). (F)In contrast, siHSP90a decreased DARC expression, regardless of CXCL1 treatment. Photomicrographs of stained ASMCs were acquired from one subject (representative of three subjects). Original magnification 363 (E and F). Results are expressed as mean 6 SEM (n = 5 subjects). *p , 0.05, **p , 0.01, ***p , 0.001 versus nonstimulated controls; $p , 0.05, $$p , 0.01, $$$p , 0.001 versus siScram or DMSO. effect of siHSP90a was not directly related to ERK-1/2 activation some of the materials used in this study, and Dr. Pasquale Ferraro (Univer- per se; rather, it appears to be due to the effect that HSP90a has on site´ de Montre´al) for providing surgical specimens for ASMC culture. We DARC protein expression. Taken together, our results suggest that also thank Dr. Min Fu for assistance with the confocal microscopy imaging HSP90a may have a regulatory effect on DARC protein expres- and Severine Audusseau for technical contributions. sion and function. In conclusion, we found that, over a specific range of concentrations, Disclosures CXCL1 acts as a negative regulator of ASMC migration and that this The authors have no financial conflicts of interest. effect is mediated by activation of the ERK-1/2 MAPK path- way. In addition, our results provide a demonstration of the abil- References ity of the decoy receptor DARC to mediate CXCL1 signaling and 1. Majesky, M. W., and S. M. Schwartz. 1990. Smooth muscle diversity in arterial function. They also establish the CXCL1/DARC/ASMC interaction wound repair. Toxicol. Pathol. 18: 554–559. as a pathway that may play a role in the pathology of asthma re- 2. Hatten, M. E. 1993. The role of migration in central nervous system neuronal development. Curr. Opin. Neurobiol. 3: 38–44. modeling, meriting further study to establish its role in ASMC mi- 3. Ikawa, M., N. Inoue, A. M. Benham, and M. Okabe. 2010. Fertilization: gration and as a therapeutic objective in airway remodeling in asthma. a sperm’s journey to and interaction with the oocyte. J. Clin. Invest. 120: 984–994. 4. Fan, D., A. Takawale, J. Lee, and Z. Kassiri. 2012. Cardiac fibroblasts, fibrosis and extracellular matrix remodeling in heart disease. Fibrogenesis Tissue Repair Acknowledgments 5: 15. 5. Pickens, S. R., N. D. Chamberlain, M. V. Volin, R. M. Pope, N. E. Talarico, We thank Dr. Carolyn Baglole (McGill University) for valuable advice, A. M. Mandelin, II, and S. Shahrara. 2012. Role of the CCL21 and CCR7 Dr. Andrew Halayko (University of Winnipeg, MB, Canada) for providing pathways in rheumatoid arthritis angiogenesis. Arthritis Rheum. 64: 2471–2481. The Journal of Immunology 1425

6. Zhang, Y., S. Takahashi, A. Tasaka, T. Yoshima, H. Ochi, and K. Chayama. 2013. 32. Govindaraju, V., M. C. Michoud, M. Al-Chalabi, P. Ferraro, W. S. Powell, and Involvement of microRNA-224 in cell proliferation, migration, invasion and J. G. Martin. 2006. -8: novel roles in human airway smooth muscle anti-apoptosis in hepatocellular carcinoma. J. Gastroenterol. Hepatol. 28: 565– cell contraction and migration. Am. J. Physiol. Cell Physiol. 291: C957–C965. 575. 33. Goncharova, E. A., D. A. Goncharov, and V. P. Krymskaya. 2006. Assays for 7. Johnson, P. R., and J. K. Burgess. 2004. Airway smooth muscle and fibroblasts in in vitro monitoring of human airway smooth muscle (ASM) and human pul- the pathogenesis of asthma. Curr. Allergy Asthma Rep. 4: 102–108. monary arterial vascular smooth muscle (VSM) cell migration. Nat. Protoc. 1: 8. Schmidt, M., G. Sun, M. A. Stacey, L. Mori, and S. Mattoli. 2003. Identification 2933–2939. of circulating fibrocytes as precursors of bronchial myofibroblasts in asthma. J. 34. Sebolt-Leopold, J. S., D. T. Dudley, R. Herrera, K. Van Becelaere, A. Wiland, Immunol. 171: 380–389. R. C. Gowan, H. Tecle, S. D. Barrett, A. Bridges, S. Przybranowski, et al. 1999. 9. Yasukawa, A., K. Hosoki, M. Toda, Y. Miyake, Y. Matsushima, T. Matsumoto, Blockade of the MAP kinase pathway suppresses growth of colon tumors in vivo. D. Boveda-Ruiz, P. Gil-Bernabe, M. Nagao, M. Sugimoto, et al. 2013. Eosino- Nat. Med. 5: 810–816. phils promote epithelial to mesenchymal transition of bronchial epithelial cells. 35. Giubellino, A., C. Sourbier, M. J. Lee, B. Scroggins, P. Bullova, M. Landau, PLoS ONE 8: e64281. W. Ying, L. Neckers, J. B. Trepel, and K. Pacak. 2013. Targeting heat shock 10. Saunders, R., S. Siddiqui, D. Kaur, C. Doe, A. Sutcliffe, F. Hollins, P. Bradding, protein 90 for the treatment of malignant pheochromocytoma. PLoS ONE 8: A. Wardlaw, and C. E. Brightling. 2009. Fibrocyte localization to the airway e56083. smooth muscle is a feature of asthma. J. Allergy Clin. Immunol. 123: 376–384. 36. Schneider, C. A., W. S. Rasband, and K. W. Eliceiri. 2012. NIH Image to 11. Wang, C. H., C. D. Huang, H. C. Lin, K. Y. Lee, S. M. Lin, C. Y. Liu, ImageJ: 25 years of image analysis. Nat. Methods 9: 671–675. K. H. Huang, Y. S. Ko, K. F. Chung, and H. P. Kuo. 2008. Increased circulating 37. Al-Alwan, L. A., Y. Chang, A. Mogas, A. J. Halayko, C. J. Baglole, J. G. Martin, fibrocytes in asthma with chronic airflow obstruction. Am. J. Respir. Crit. Care S. Rousseau, D. H. Eidelman, and Q. Hamid. 2013. Differential roles of CXCL2 Med. 178: 583–591. and CXCL3 and their receptors in regulating normal and asthmatic airway 12. Walker, N., L. Badri, S. Wettlaufer, A. Flint, U. Sajjan, P. H. Krebsbach, smooth muscle cell migration. J. Immunol. 191: 2731–2741. V. G. Keshamouni, M. Peters-Golden, and V. N. Lama. 2011. Resident tissue- 38. Suganuma, N., S. Ito, H. Aso, M. Kondo, M. Sato, M. Sokabe, and Y. Hasegawa. specific mesenchymal progenitor cells contribute to fibrogenesis in human lung 2012. STIM1 regulates platelet-derived growth factor-induced migration and allografts. Am. J. Pathol. 178: 2461–2469. Ca2+ influx in human airway smooth muscle cells. PLoS ONE 7: e45056. 13. Berair, R., R. Saunders, and C. E. Brightling. 2013. Origins of increased airway 39. Damaj, B. B., S. R. McColl, K. Neote, C. A. He´bert, and P. H. Naccache. 1996. smooth muscle mass in asthma. BMC Med. 11: 145. Diverging signal transduction pathways activated by (IL-8) and Downloaded from 14. Nihlberg, K., K. Larsen, A. Hultga˚rdh-Nilsson, A. Malmstro¨m, L. Bjermer, and related chemokines in human neutrophils. IL-8 and Gro-alpha differentially G. Westergren-Thorsson. 2006. Tissue fibrocytes in patients with mild asthma: stimulate calcium influx through IL-8 receptors A and B. J. Biol. Chem. 271: a possible link to thickness of reticular basement membrane? Respir. Res. 7: 50. 20540–20544. 15. Al-Alwan, L. A., Y. Chang, C. J. Baglole, P. A. Risse, A. J. Halayko, J. G. Martin, 40. Milligan, G., and E. Kostenis. 2006. Heterotrimeric G-proteins: a short history. D. H. Eidelman, and Q. Hamid. 2012. Autocrine-regulated airway smooth muscle Br. J. Pharmacol. 147(Suppl. 1): S46–S55. cell migration is dependent on IL-17-induced growth-related oncogenes. J. Allergy 41. Horuk, R., A. Martin, J. Hesselgesser, T. Hadley, Z. H. Lu, Z. X. Wang, and Clin. Immunol. 130: 977–985.e6 S. C. Peiper. 1996. The Duffy antigen receptor for chemokines: structural

16. De Filippo, K., A. Dudeck, M. Hasenberg, E. Nye, N. van Rooijen, K. Hartmann, analysis and expression in the brain. J. Leukoc. Biol. 59: 29–38. http://www.jimmunol.org/ M. Gunzer, A. Roers, and N. Hogg. 2013. Mast cell and chemokines 42. Liu, X. H., T. J. Hadley, L. Xu, S. C. Peiper, and P. E. Ray. 1999. Up-regulation CXCL1/CXCL2 control the early stage of neutrophil recruitment during tissue of Duffy antigen receptor expression in children with renal disease. Kidney Int. inflammation. Blood 121: 4930–4937. 55: 1491–1500. 17. Kuo, P. L., K. H. Shen, S. H. Hung, and Y. L. Hsu. 2012. CXCL1/GROa 43. Rousseau, S., F. Houle, J. Landry, and J. Huot. 1997. p38 MAP kinase activation increases cell migration and invasion of prostate cancer by decreasing fibulin-1 by vascular endothelial growth factor mediates actin reorganization and cell expression through NF-kB/HDAC1 epigenetic regulation. Carcinogenesis 33: migration in human endothelial cells. Oncogene 15: 2169–2177. 2477–2487. 44. Hedges, J. C., M. A. Dechert, I. A. Yamboliev, J. L. Martin, E. Hickey, 18. Jones, C. P., S. C. Pitchford, C. M. Lloyd, and S. M. Rankin. 2009. CXCR2 L. A. Weber, and W. T. Gerthoffer. 1999. A role for p38(MAPK)/HSP27 pathway mediates the recruitment of endothelial progenitor cells during allergic airways in smooth muscle cell migration. J. Biol. Chem. 274: 24211–24219. remodeling. Stem Cells 27: 3074–3081. 45. He, F., A. Kumar, and Z. H. Song. 2012. Heat shock protein 90 is an essential 19. Kroeze, K. L., M. A. Boink, S. C. Sampat-Sardjoepersad, T. Waaijman, molecular chaperone for CB2 cannabinoid receptor-mediated signaling in tra-

R. J. Scheper, and S. Gibbs. 2012. Autocrine regulation of re-epithelialization becular meshwork cells. Mol. Vis. 18: 2839–2846. by guest on September 23, 2021 after wounding by chemokine receptors CCR1, CCR10, CXCR1, CXCR2, and 46. Liu, D. H., H. Y. Yuan, C. Y. Cao, Z. P. Gao, B. Y. Zhu, H. L. Huang, and CXCR3. J. Invest. Dermatol. 132: 216–225. D. F. Liao. 2007. Heat shock protein 90 acts as a molecular chaperone in late- 20. Ahuja, S. K., and P. M. Murphy. 1996. The CXC chemokines growth-regulated phase activation of extracellular signal-regulated kinase 1/2 stimulated by oxi- oncogene (GRO) alpha, GRObeta, GROgamma, neutrophil-activating peptide-2, dative stress in vascular smooth muscle cells. Acta Pharmacol. Sin. 28: 1907– and epithelial cell-derived neutrophil-activating peptide-78 are potent agonists 1913. for the type B, but not the type A, human interleukin-8 receptor. J. Biol. Chem. 47. Kobayakawa, T., S. Yamada, A. Mizuno, and T. K. Nemoto. 2008. Substitution 271: 20545–20550. of only two residues of human Hsp90alpha causes impeded dimerization of 21. Zlotnik, A., and O. Yoshie. 2012. The chemokine superfamily revisited. Im- Hsp90beta. Cell Stress Chaperones 13: 97–104. munity 36: 705–716. 48. Gao, Y., S. Yechikov, A. E. Vazquez, D. Chen, and L. Nie. 2013. Distinct roles of 22. Luttrell, L. M. 2008. Reviews in molecular biology and biotechnology: trans- molecular chaperones HSP90a and HSP90b in the biogenesis of KCNQ4 membrane signaling by G protein-coupled receptors. Mol. Biotechnol. 39: 239–264. channels. PLoS ONE 8: e57282. 23. Graham, G. J., M. Locati, A. Mantovani, A. Rot, and M. Thelen. 2012. The 49. Cortes-Gonzalez, C., J. Barrera-Chimal, M. Ibarra-Sanchez, M. Gilbert, G. Gamba, biochemistry and biology of the atypical chemokine receptors. Immunol. Lett. A. Zentella, M. E. Flores, and N. A. Bobadilla. 2010. Opposite effect of Hsp90alpha 145: 30–38. and Hsp90beta on eNOS ability to produce nitric oxide or superoxide anion in 24. Cutbush, M., and P. L. Mollison. 1950. The Duffy blood group system. Heredity human embryonic kidney cells. Cell. Physiol. Biochem. 26: 657–668. (Edinb) 4: 383–389. 50. Ridley, A. J., M. A. Schwartz, K. Burridge, R. A. Firtel, M. H. Ginsberg, 25. Miller, L. H., S. J. Mason, J. A. Dvorak, M. H. McGinniss, and I. K. Rothman. G. Borisy, J. T. Parsons, and A. R. Horwitz. 2003. Cell migration: integrating 1975. Erythrocyte receptors for (Plasmodium knowlesi) malaria: Duffy blood signals from front to back. Science 302: 1704–1709. group determinants. Science 189: 561–563. 51. Wehrle-Haller, B. 2012. Structure and function of focal adhesions. Curr. Opin. 26. Horuk, R., C. E. Chitnis, W. C. Darbonne, T. J. Colby, A. Rybicki, T. J. Hadley, Cell Biol. 24: 116–124. and L. H. Miller. 1993. A receptor for the malarial parasite Plasmodium vivax: 52. Aso, H., S. Ito, A. Mori, N. Suganuma, M. Morioka, N. Takahara, M. Kondo, and the erythrocyte chemokine receptor. Science 261: 1182–1184. Y. Hasegawa. 2013. Differential regulation of airway smooth muscle cell mi- 27. Chaudhuri, A., V. Zbrzezna, J. Polyakova, A. O. Pogo, J. Hesselgesser, and gration by E-prostanoid receptor subtypes. Am. J. Respir. Cell Mol. Biol. 48: R. Horuk. 1994. Expression of the Duffy antigen in K562 cells. Evidence that it 322–329. is the human erythrocyte chemokine receptor. J. Biol. Chem. 269: 7835–7838. 53. Ning, Y., Q. Sun, Y. Dong, W. Xu, W. Zhang, H. Huang, and Q. Li. 2011. Slit2-N 28. Luan, J., Y. Furuta, J. Du, and A. Richmond. 2001. Developmental expression of inhibits PDGF-induced migration in rat airway smooth muscle cells: WASP and two CXC chemokines, MIP-2 and KC, and their receptors. 14: 253– Arp2/3 involved. Toxicology 283: 32–40. 263. 54. Luan, J., R. Shattuck-Brandt, H. Haghnegahdar, J. D. Owen, R. Strieter, 29. Fukuma, N., N. Akimitsu, H. Hamamoto, H. Kusuhara, Y. Sugiyama, and M. Burdick, C. Nirodi, D. Beauchamp, K. N. Johnson, and A. Richmond. 1997. K. Sekimizu. 2003. A role of the Duffy antigen for the maintenance of plasma Mechanism and biological significance of constitutive expression of MGSA/ chemokine concentrations. Biochem. Biophys. Res. Commun. 303: 137–139. GRO chemokines in malignant melanoma tumor progression. J. Leukoc. Biol. 30. Lee, J. S., C. W. Frevert, M. M. Wurfel, S. C. Peiper, V. A. Wong, K. K. Ballman, 62: 588–597. J. T. Ruzinski, J. S. Rhim, T. R. Martin, and R. B. Goodman. 2003. Duffy antigen 55. Antoniou, K. M., A. Tzouvelekis, M. G. Alexandrakis, K. Sfiridaki, facilitates movement of chemokine across the endothelium in vitro and promotes I. Tsiligianni, G. Rachiotis, N. Tzanakis, D. Bouros, J. Milic-Emili, and neutrophil transmigration in vitro and in vivo. J. Immunol. 170: 5244–5251. N. M. Siafakas. 2006. Different angiogenic activity in pulmonary sarcoidosis and 31. Pruenster, M., L. Mudde, P. Bombosi, S. Dimitrova, M. Zsak, J. Middleton, idiopathic pulmonary fibrosis. Chest 130: 982–988. A. Richmond, G. J. Graham, S. Segerer, R. J. Nibbs, and A. Rot. 2009. The 56. Benelli, R., S. Stigliani, S. Minghelli, S. Carlone, and N. Ferrari. 2013. Impact of Duffy antigen receptor for chemokines transports chemokines and supports their CXCL1 overexpression on growth and invasion of prostate cancer cell. Prostate promigratory activity. Nat. Immunol. 10: 101–108. 73: 941–951. 1426 CXCL1 INHIBITS ASMC MIGRATION VIA DARC

57. Alkhouri, H., L. M. Moir, C. L. Armour, and J. M. Hughes. 2014. CXCL1 is bition of CXCR3-mediated chemotaxis by the human chemokine receptor-like a negative regulator of mast cell chemotaxis to airway smooth muscle cell protein CCX-CKR. Br. J. Pharmacol. 168: 1375–1387. products in vitro. Clin. Exp. Allergy 44: 381–392. 64. Goupil, E., S. A. Laporte, and T. E. He´bert. 2012. Functional selectivity in GPCR 58. Vora, P., P. Pillai, J. Mustapha, C. Kowal, S. Shaffer, R. Bose, M. Namaka, and signaling: understanding the full spectrum of receptor conformations. Mini Rev. E. E. Frost. 2012. CXCL1 regulation of oligodendrocyte progenitor cell mi- Med. Chem. 12: 817–830. gration is independent of calcium signaling. Exp. Neurol. 236: 259–267. 65. Maudsley, S., S. A. Patel, S. S. Park, L. M. Luttrell, and B. Martin. 2012. 59. Vergara, C., Y. J. Tsai, A. V. Grant, N. Rafaels, L. Gao, T. Hand, M. Stockton, Functional signaling biases in G protein-coupled receptors: Game Theory and M. Campbell, D. Mercado, M. Faruque, et al. 2008. Gene encoding Duffy receptor dynamics. Mini Rev. Med. Chem. 12: 831–840. antigen/receptor for chemokines is associated with asthma and IgE in three 66. Luttrell, L. M., F. L. Roudabush, E. W. Choy, W. E. Miller, M. E. Field, populations. Am. J. Respir. Crit. Care Med. 178: 1017–1022. K. L. Pierce, and R. J. Lefkowitz. 2001. Activation and targeting of extracellular 60. Sun, G., Y. Wang, Y. Zhu, C. Huang, and Q. Ji. 2011. Duffy antigen receptor for signal-regulated kinases by beta-arrestin scaffolds. Proc. Natl. Acad. Sci. USA chemokines in laryngeal squamous cell carcinoma as a negative regulator. Acta 98: 2449–2454. Otolaryngol. 131: 197–203. 67. Eishingdrelo, H., and S. Kongsamut. 2013. Minireview: Targeting GPCR Activated 61. Latini, F. R., A. U. Bastos, C. P. Arnoni, J. G. Muniz, R. M. Person, W. Baleotti, ERK Pathways for Drug Discovery. Curr. Chem. Genomics Transl. Med. 7: 9–15. Jr., J. A. Barreto, L. Castilho, and J. M. Cerutti. 2013. DARC (Duffy) and BCAM 68. Wang, Y., Y. Zhan, R. Xu, R. Shao, J. Jiang, and Z. Wang. 2014. Src mediates (Lutheran) reduced expression in thyroid cancer. Blood Cells Mol. Dis. 50: 161–165. extracellular signal-regulated kinase 1/2 activation and autophagic cell death 62. Hesselgesser, J., C. E. Chitnis, L. H. Miller, D. G. Yansura, L. C. Simmons, induced by cardiac glycosides in human non-small cell lung cancer cell lines. W. J. Fairbrother, C. Kotts, C. Wirth, B. L. Gillece-Castro, and R. Horuk. 1995. Mol. Carcinog. DOI: 10.1002/mc.22147. A mutant of melanoma growth stimulating activity does not activate neutrophils 69. Martel, G., J. Be´rube´, and S. Rousseau. 2013. The protein kinase TPL2 is es- but blocks erythrocyte invasion by malaria. J. Biol. Chem. 270: 11472–11476. sential for ERK1/ERK2 activation and cytokine gene expression in airway ep- 63. Vinet, J., M. van Zwam, I. M. Dijkstra, N. Brouwer, H. R. van Weering, ithelial cells exposed to pathogen-associated molecular patterns (PAMPs). PLoS A. Watts, M. Meijer, M. R. Fokkens, V. Kannan, D. Verzijl, et al. 2013. Inhi- ONE 8: e59116. Downloaded from http://www.jimmunol.org/ by guest on September 23, 2021