Clearance Across Lymphatic Endothelium Endothelium And

Home , CXCR5

Prostaglandin D2 Regulates CD4+ Memory T Cell Trafficking across Blood Vascular Endothelium and Primes These Cells for Clearance across Lymphatic Endothelium This information is current as of October 4, 2021. S. Rumel Ahmed, Helen M. McGettrick, Clara M. Yates, Christopher D. Buckley, Marianne J. Ratcliffe, Gerard B. Nash and G. Ed Rainger J Immunol 2011; 187:1432-1439; Prepublished online 29

June 2011; Downloaded from doi: 10.4049/jimmunol.1100299 http://www.jimmunol.org/content/187/3/1432

Supplementary http://www.jimmunol.org/content/suppl/2011/06/29/jimmunol.110029 http://www.jimmunol.org/ Material 9.DC1 References This article cites 34 articles, 11 of which you can access for free at: http://www.jimmunol.org/content/187/3/1432.full#ref-list-1

Why The JI? Submit online.

by guest on October 4, 2021 • Rapid Reviews! 30 days* from submission to initial decision

• 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 © 2011 by The American Association of Immunologists, Inc. All rights reserved. Print ISSN: 0022-1767 Online ISSN: 1550-6606. The Journal of Immunology

+ Prostaglandin D2 Regulates CD4 Memory T Cell Trafﬁcking across Blood Vascular Endothelium and Primes These Cells for Clearance across Lymphatic Endothelium

S. Rumel Ahmed,*,1 Helen M. McGettrick,*,†,1 Clara M. Yates,* Christopher D. Buckley,† Marianne J. Ratcliffe,‡ Gerard B. Nash,* and G. Ed Rainger*

Memory lymphocytes support inflammatory and immune responses. To do this, they enter tissue via blood vascular endothelial cells (BVEC) and leave tissue via lymphatic vascular endothelial cells (LVEC). In this study, we describe a hierarchy of signals, including novel regulatory steps, which direct the sequential migration of human T cells across the blood and the lymphatic EC. Cytokine- stimulated (TNF and IFN) human BVEC preferentially recruited memory T cells from purified PBL. Lymphocyte recruitment from flow could be blocked using a function-neutralizing Ab against CXCR3. However, a receptor antagonist directed against the Downloaded from PGD2 receptor DP2 (formerly chemoattractant receptor-homologous molecule expressed on Th2 cells) inhibited transendothelial migration, demonstrating that the sequential delivery of the chemokine and prostanoid signals was required for efficient lymphocyte recruitment. CD4+ T cells recruited by BVEC migrated with significantly greater efficiency across a second barrier of human LVEC, an effect reproduced by the addition of exogenous PGD2 to nonmigrated cells. Migration across BVEC or exogenous PGD2 modified the function, but not the expression, of CCR7, so that chemotaxis toward CCL21 was significantly enhanced. Thus, chemokines may not regulate all stages of lymphocyte migration during inflammation, and paradigms describing http://www.jimmunol.org/ their trafficking may need to account for the role of PGD2. The Journal of Immunology, 2011, 187: 1432–1439.

ollowing infection, foreign Ag is presented to naive T cells thesized and presented to the flowing lymphocytes in response to constitutively recirculating through secondary lymphoid endothelial cell (EC) activation by cytokines such as TNF-a (ΤNF), F organs. There they become activated and acquire the ef- IL-1b (IL-1), or IFN-g (IFN) (3). Initial capture is followed by fector functions and phenotypic markers associated with Ag- activation of the lymphocytes by surface-presented chemokine(s). specific memory T cells (reviewed in Refs. 1, 2). A defining as- Chemokines induce the rapid activation of the b1 and b2 inte- pect of immune memory is the persistence of the response. This grin receptors constitutively expressed on the lymphocyte surface. is achieved when a fraction of tissue-resident cells leave the High-affinity interactions between a4b1 integrin (VLA-4) and its by guest on October 4, 2021 inflamed tissue via the lymphatic vasculature and re-enter the EC counterligand, VCAM-1, and between aLb2 integrin and peripheral blood during the resolution phase of the inflammatory ICAM-1 (4–6) stabilize lymphocyte adhesion to the EC mono- response. There, in the absence of antigenic stimulation, they layer. There is also a strong presumption that chemokine signals patrol the body as long-lived memory T cells, conducting immune are sufficient to promote lymphocyte migration over and through surveillance for their cognate Ag (reviewed in Refs. 1, 2). the endothelial monolayer and into inflamed tissue. In the case of During inflammation, memory T cells are recruited to tissue after EC treatment with TNF plus IFN, chemokines acting through capture by specialized, fast-acting adhesion receptors (i.e., VCAM- the lymphocyte-borne receptor CXCR3 have been shown to sta- 1 and E- or P-selectin) on blood vascular endothelial cells (BVEC) bilize attachment, but the signals inducing transendothelial mi- of the postcapillary venules. These adhesion molecules are syn- gration have not been defined (7, 8). Transendothelial migration of T cells has been observed within minutes of adhesion to EC activated with TNF (where blockade of b2 integrins was inhibitory) *School of Clinical and Experimental Medicine, Medical School, University of Bir- (4), with TNF plus IFN (7), and for EC stimulated with TNF and † mingham, Birmingham B15 2TT, United Kingdom; School of Immunity and In- to which stromal-derived factor-1a (CXCL12) or CCL19 (ELC) fection, Medical School, University of Birmingham, Birmingham B15 2TT, United Kingdom; and ‡AstraZeneca Research and Development Charnwood, Leicestershire had been added to the EC surface (9). In the last example, the LE11 5RH, United Kingdom homeostatic chemokines (CXCL12 and CCL19) were not en- 1S.R.A. and H.M.M. contributed equally to this work. dogenous products of blood vascular endothelium, and it is Received for publication February 1, 2011. Accepted for publication May 23, 2011. doubtful whether they contribute to lymphocyte activation in the This work was supported by Biotechnology and Biological Sciences Research Board inflamed vasculature. Collaborative Awards in Science and Engineering Studentship, which included fund- In addition to signals that permit admission to tissue, T cell ing in part from AstraZeneca. egress from the site of inflammation is necessary for immune Address correspondence and reprint requests to Dr. G. Ed Rainger, School of Clinical surveillance and the efficient regulation and resolution of the and Experimental Medicine, Medical School, University of Birmingham, Birming- ham B15 2TT, United Kingdom. E-mail address: [email protected] immune response. For example, memory T cells with Ag reactivity The online version of this article contains supplemental material. that is not relevant to the ongoing response must achieve efficient Abbreviations used in this article: BVEC, blood vascular endothelial cell; CRTh2, release from the inflamed tissue or the immune system risks de- chemoattractant receptor-homologous molecule expressed on Th2 cells; DC, den- letion of their memory repertoire when the immune infiltrate is dritic cell; EC, endothelial cell; GPCR, G protein-coupled receptor; LVEC, lymphatic cleared during inflammatory resolution (10). In addition, T cell vascular endothelial cell; PBSA, PBS with albumin; S1P, sphingosine 1-phosphate. helper functions that support the humoral immune response or- Copyright Ó 2011 by The American Association of Immunologists, Inc. 0022-1767/11/$16.00 dinarily occur in secondary lymphoid tissue, as it is here that www.jimmunol.org/cgi/doi/10.4049/jimmunol.1100299 The Journal of Immunology 1433 professional APC [i.e., dendritic cells (DC), B lymphocytes, and Analysis of lymphocyte migration Th cells] are brought into functional association (11). Movement Migration through BVEC on Transwell filters under static conditions. out of tissue toward the draining lymph node occurs in the afferent Lymphocyte migration was assessed using 24-well format Transwell filters lymphatic vasculature. This process is less well characterized than as described (17). BVEC were washed to remove residual cytokines, fresh recruitment across the blood vasculature; however, it also appears M199+BSA was placed in the lower chamber, and PBL were added to the upper chamber. In some experiments, lymphocytes were pretreated with to be tightly regulated, permitting the migration of specific pop- Abs at 10 mg/ml (unless otherwise stated) against the integrin subunits + ulations of leukocytes. Studies to date indicate that CD4 memory CD29 (anti-b1; clone Mab13 a gift from M. Humphries, University of T cell migration out of tissue and into the lymphatic vasculature is Manchester, Manchester, U.K.), CD18 (anti-b2; clone R6.5E a gift from dependent upon interactions between the lymphocyte chemokine Dr. Tony Shock; Cell Tech, Slough, U.K.), CD61 (anti-b3 integrin; clone SZ21 from Immunotech, Marseilles, France), or CD18 activation epitope receptor CCR7 and the chemokines CCL19 and/or CCL21 (sec- (anti-b2; clone KIM127 a gift from Martyn Robinson, Cell Tech) (18) ondary lymphoid tissue chemokine), both of which are expressed against the chemokine receptors CXCR3 (clone 1C6; R&D Systems) or on lymphatic endothelium (12, 13). CCR7 (clone 150503; R&D Systems), with antagonists of the PGD2 In a recent study, we reported that although activation by receptors DP1 (BW868c; Cayman Chemicals) or DP2 (BAY-u3405 chemokines was a first and essential signal for integrin activation [Ramatroban]; Cayman Chemicals), or with NBD-phallacidin (Molecular and immobilization of neutrophils on TNF-stimulated EC, onward Probes) to measure F-actin. The lymphocytes were allowed to settle, adhere, and migrate through EC and the filter at 37˚C in a CO2 incubator for migration across the endothelium was mediated by PGD2 (14). In 24 h. From previous experience, little transmigration was detected within this study, we show that the same paradigm is relevant to the a few hours using filters, and 24 h were needed to obtain a proportion recruitment of CD4+ and CD8+ memory T cells. When recruited of lymphocytes under the filter comparable to that seen in minutes during by EC stimulated by TNF and IFN, an initial chemokine signal, direct observations of transendothelial migration (17). Transmigrated cells operating through the lymphocyte-borne CXCR3 receptor, was in the lower chamber were removed and counted using a Coulter Multi- Downloaded from sizer II (Coulter Electronics, Essex, U.K.). From the known number of essential for immobilization. However, efficient penetration of added lymphocytes, the percentage of lymphocytes that transmigrated was the endothelial monolayer required a PGD2-mediated signal. Im- calculated. In some experiments, the surface phenotype of transmigrated portantly, passage of the BVEC monolayer and, specifically, lymphocytes was assessed by flow cytometry. Lymphocytes were labeled + exposure to PGD2 also primed CD4 T cells for transmigration with anti–CD4-PE, anti–CD8-FITC (BD Biosciences, Oxford, U.K.), or across lymphatic endothelium. Efficient translymphatic migra- anti-CD45RA-CY5 (Serotech, Oxford, U.K.) for 30 min on ice. Fixed- tion was CCR7 dependent, and migration across BVEC, or the volume counts for positively labeled cells were made using a Coulter http://www.jimmunol.org/ XL flow cytometer (Coulter Electronics) and analyzed using WinMDI. In provision of exogenous synthetic PGD2, greatly increased the this way, we calculated the percentage of transmigration for CD4+ and number of lymphocytes migrating across lymphatic vascular EC CD8+ and memory T cell subsets. Apoptotic cells were assessed by DiOC6 (LVEC) as well as promoting chemotaxis of lymphocytes to- and 7-aminoactinomycin D and excluded from counts; we observed little if ward CCL21. any apoptosis over the course of the assay. Migration through LVEC on Transwell filters under static conditions. Lymphocyte migration across LVEC was assessed as described above for Materials and Methods the assay of lymphocytes across BVEC with the exception that the LVEC + Isolation and culture of human PBL and CD4 memory T cells were cultured on the underside of the filter so that lymphocytes migrating

across this barrier would contact the monolayer from a physiologically by guest on October 4, 2021 Venous blood from healthy individuals was collected in EDTA at a final relevant direction. The lymphocyte populations used were: 1) PBL or concentration of 1.6 mg/ml (Sarstedt, Leicester, U.K.). PBMC were isolated + 2 purified CD4 CD45RA T cells collected from the lower chamber of by centrifugation of blood on Histopaque 1077 (Sigma-Aldrich, Poole, + 2 BVEC migration assays; and 2) control PBL or purified CD4 CD45RA U.K.), and PBL were prepared by panning of PBMC on culture plastic to 2 T cells that were maintained in culture flasks in BVEC medium for 24 h. remove monocytes (15). In some experiments, CD4+ memory (CD45RA ) The latter represented lymphocyte populations of equivalent age and prior T cells were purified from PBMC using a positive selection immunobead manipulation to cells migrated across BVEC, but had not migrated through kit (Miltenyi Biotec, Surrey, U.K.). Isolated cells were washed, counted, BVEC. and adjusted to a final concentration of 1 3 106/ml in Medium 199 (Life Technologies Invitrogen Compounds, Paisley, U.K.) supplemented with Microscopic observation of migration through EC under static conditions. 1% BSA (Sigma-Aldrich) or PBS with albumin (0.15%; PBSA). Adhesion and transmigration were assessed by direct microscopic observation as previously described (19). BVEC in six-well plates were washed In some experiments, lymphocytes were pretreated with 10 nM PGD2 or agonists against the PGD receptors, DP1 (BW 245C) or DP2 (15[R]- with PBSA to remove residual cytokines, and purified PBL were added for 2 5 min. Nonadherent cells were removed from the EC by gentle washing PGD2; all from Cayman Chemicals), or with the G protein-coupled receptor (GPCR) inhibitor pertussis toxin (100 ng/ml; Sigma-Aldrich) prior with PBSA, and video recordings of five fields of view of the endothelial to stimulation with PGD . Alternatively, to examine the mechanism of surface were made using phase-contrast videomicroscopy so that the 2 number of adherent cells and their position above or below the EC action of PGD2, PBL were pretreated with 10 nM DP2 agonist prior to staining with anti–CD31-PE (R&D Systems, Oxford, U.K.). monolayer could be assessed. Manipulations and microscopy were carried out at 37˚C. The video recordings were digitized and analyzed offline using Image-Pro Plus software (DataCell, Finchampstead, U.K.). The numbers Isolation and culture of EC of adherent cells were counted in the video fields, averaged, converted 2 HUVEC were isolated from umbilical cords as previously described (16) to cells per mm using the calibrated microscope field dimensions, and and cultured in M199 supplemented with 20% FCS, 10 ng/ml epidermal multiplied by the known surface area of the EC monolayer to calculate the growth factor, 35 mg/ml gentamicin, 1 mg/ml hydrocortisone (all from total number adherent. This number was divided by the known total Sigma-Aldrich), and 2.5 mg/ml amphotericin B (Life Technologies Invi- number of lymphocytes added to obtain the percentage of the lymphocytes trogen Compounds). Primary HUVEC were dissociated using trypsin/ that had adhered. Each lymphocyte was classified as either: 1) phase bright EDTA (Sigma-Aldrich) and seeded on six-well tissue culture plates (Fal- and adherent to the surface of the EC; or 2) phase dark and spread and con; Becton Dickinson Labware), uncoated low-density 3.0-mm pore migrating below the EC. The percentage of adherent lymphocytes that had polycarbonate Transwell filters (which were placed in matching plates; BD transmigrated was calculated. Pharmingen, Oxford, U.K.), or iBidi chamber slides (iBidi, Martinsried, Microscopic observation of migration through EC under conditions of flow. Germany) for flow assays. Seeding density was chosen to yield confluent iBidi chamber slides (iBidi) containing a confluent EC monolayer were monolayers within 24 h. TNF-a (TNF; 100 U/ml; Sigma-Aldrich) and attached to a perfusion system mounted on the stage of a phase-contrast IFN-g (IFN; 10ng/ml; PeproTech, London, U.K.) were added to confluent videomicroscope enclosed in a Perspex chamber at 37˚C, as described monolayers for 24 h before adhesion assay with lymphocytes. (19, 20). At one end, they were connected to a Harvard withdrawal syringe Primary human dermal LVEC were purchased from Promocell and pump that delivered flow at a rate equivalent to a wall shear stress of 0.1 cultured in the manufacturer’s recommended medium (Promocell, Hei- Pa. At the other end, they were connected to an electronic switching valve delberg, Germany). Lymphatic EC were subcultured on to Transwell filters (Lee Products, Gerards Cross, U.K.), which selected flow from two res- and treated with inflammatory cytokines as described above. ervoirs containing PBL in PBSA or cell-free PBSA. A 4-min bolus of PBL 1434 PGD2 AND LYMPHOCYTE TRAFFIC DURING INFLAMMATION was perfused over the HUVEC followed by cell-free wash buffer. Video recordings were made of a series of microscope fields along the centerline of the flow channel after 11 min of washout. Video recordings were analyzed as above, except that lymphocytes adherent to the HUVEC could be classified as rolling adherent (spherical cells moving over the surface much slower than free-flowing cells), stationary adherent (typically with dis- torted shape and actually migrating slowly on the surface), or phase-dark transmigrated cells. The sum of the numbers adherent in all categories was divided by the number perfused during the bolus to obtain total PBL adhesion as percent of cells perfused. Lymphocyte chemotaxis toward CCL21 Lymphocyte chemotaxis was assayed in 96-well MultiScreen plates with 5-mm pores (Millipore). Freshly isolated PBL, PBL that had migrated through BVEC or PBL pretreated with exogenous PGD2, were allowed to migrate across uncoated filters toward CCL21 (80 ng/ml added to lower well; R&D Systems) for 4 h to allow sufficient time for lymphocytes to cross the filter. Transmigrated cells in the lower chamber were removed and counted using a Coulter Multisizer II (Coulter Electronics). From the known number of added lymphocytes, the percentage of lymphocytes that migrated across the filter was calculated. mRNA quantification by PCR Downloaded from Trypsin/EDTA was used to detach HUVEC from the inside of filters or LVEC from the underside of the filters, and mRNA was isolated from the cells using the RNeasy Mini Kit (Qiagen, Crawley, U.K.). CCL19 FAM- labeled and 18S VIC-labeled primers were bought as Assay on Demand kits from Applied Biosystems (Warrington, U.K.) to analyze mRNA. mRNA were analyzed by quantitative real-time PCR using the Quanti-Tect probe RT-PCR kit according to the manufacturer’s instructions (Qiagen). Samples were amplified using the 7900HT Real-Time PCR machine http://www.jimmunol.org/ (Applied Biosystems) and analyzed using the software package SDS 2.2 (Applied Biosystems). Data were expressed as relative expression units relative to 18S. FIGURE 1. Memory lymphocytes preferentially migrate across cyto- Statistical analysis kine-stimulated BVEC. A, Cytokine stimulation of BVEC had no consis- Effects of multiple treatments were tested using ANOVA, followed by post tent effect on the levels of PBL adhesion. B, Stimulation of BVEC with hoc comparison with control by Dunnett test. Single treatments were a combination of TNF and IFN significantly increased the efficiency of compared with controls by paired t test. lymphocyte migration into the lower Transwell chamber. **p , 0.01 compared with untreated BVEC by paired t test. C, The percentage of Results CD4+ and CD8+ memory T cells in isolated PBL before and after mi- by guest on October 4, 2021 Vascular EC stimulated with TNF and IFN preferentially gration across BVEC shows enrichment for memory T cells. All data are 6 , , recruit memory T cells mean SEM of five independent experiments. *p 0.05, **p 0.01 compared with PBL by paired t test. When unfractionated PBL were added to unstimulated BVEC cultured in Transwell with 3-mm pores, ∼40% became adherent, and 10% of all lymphocytes added migrated across the endothelial necessary to promote passage across the EC barrier. Previously, monolayer into the bottom chamber of the assay (Fig. 1A,1B). we found that chemokine-activated neutrophils required a signal Activation of the EC with TNF, IFN, or a combination of these delivered through the PGD2 receptor DP-1 to efficiently migrate agents did not lead to a significant increase in lymphocyte adhe- across EC (14). In this study, inclusion of antagonists of the PGD2 sion, although the combination of TNF and IFN did increase the receptors DP-1 or DP-2 in a Transwell migration assay demon- efficiency of migration significantly (Fig. 1A,1B). CD4 and CD8 strated that the latter but not the former could effectively ablate memory (CD45RA2) T cells preferentially migrated through EC the migration of lymphocytes (Fig. 2B). We also tested the effi- stimulated with TNF and IFN compared with naive cells of the cacy of the DP-2 receptor antagonist in a flow-based assay. On same types (Fig. 1C), in agreement with previous data from static cytokine-stimulated EC, lymphocytes were recruited, activated, (17) and flow assays (9). Memory cells constituted ∼50% and 20% and migrated on the surface and underneath the monolayer after of the CD4 and CD8 populations in peripheral blood, respectively. diapedesis (Supplemental Video 1). DP-2 receptor antagonist did Both CD4 and CD8 memory populations were enriched after not reduce the number adhering (data not shown) but was able to migration so that they represented ∼90% and 50% of the migrated inhibit .50% of lymphocyte migration across the monolayer (Fig. CD4 and CD8 populations, respectively (Fig. 1C). 2C–E, Supplemental Video 2). Of those cells that migrated, movement across the endothelium occurred in a paracellular and A CXCR3-mediated signal immobilizes lymphocytes on not transcellular manner. The adherent lymphocytes recruited cytokine-stimulated vascular EC, but signaling through the from flow did not roll, implying that they received the CXCR3- PGD receptor DP-2 (chemoattractant receptor-homologous 2 mediated chemokine signal necessary for immobilization, but molecule expressed on Th2 cells) is essential for efficient lacked the prostanoid signal required for transit of the monolayer. transmigration of recruited lymphocytes In agreement with previous observations (7, 8), we found that Transit of blood vascular EC primes lymphocytes for migration a function-blocking Ab against CXCR3 nearly abolished the ad- across lymphatic vascular EC hesion of flowing PBL to TNF- and IFN-stimulated EC (Fig. 2A). PBL were allowed to migrate across TNF- and IFN-stimulated These data implied that a CXCR3-mediated signal was required BVEC for 24 h and collected from the lower chamber of the as- for stable adhesion to EC, but could not reveal whether it was say. These cells (predominantly CD4 and CD8 memory T cells) The Journal of Immunology 1435 Downloaded from

FIGURE 2. IFN-g–inducible chemokines and PGD2 are both essential for the efficient migration of lymphocytes across BVEC. A, The effect of blocking the chemokine receptor CXCR3 on the recruitment of flowing lymphocytes to BVEC stimulated with TNF and IFN. B, The effect of blocking the PGD2 receptors DP-1 or DP-2 on lymphocyte migration across BVEC stimulated with TNF and IFN under static assay conditions. ANOVA showed a significant effect of inhibitor treatment on lymphocyte migration (p , 0.01). C and D, Micrographs from flow-based adhesion assays of lymphocytes adherent to TNF- and IFN-stimulated BVEC in the absence or presence of 1 mM DP-2 antagonist (original magnification 3200). Arrows refer to surface-adherent phase- bright lymphocytes (SA) and transmigrated phase-dark lymphocytes (TM). E, The effect of DP-2 antagonist on lymphocyte behavior following recruitment http://www.jimmunol.org/ from flow. Cells were classified as rolling (white bars), surface adherent (black bars), or transmigrated (gray bars). All data are mean 6 SEM of at least three independent experiments. *p , 0.05 compared with untreated lymphocytes by paired t test, **p , 0.01 by Dunnett post hoc test. were then placed on LVEC that were unstimulated or activated with layer was significantly inhibited (Fig. 4A). As controls, we used TNF and IFN. As a control, freshly isolated PBL were also cultured a function-neutralizing Ab against CXCR3 and an isotype- in dishes for the same period prior to adhesion assay on LVEC. The matched nonspecific Ab, neither of which had any consistent ef- cultured lymphocytes could adhere to LVEC, and a small pro- fect on migration across LVEC (Fig. 4A). We confirmed the ex-

portion migrated across the monolayer (Fig. 3A). Stimulation of pression of CCL21 mRNA in LVEC but not HUVEC by by guest on October 4, 2021 the LVEC with cytokines did not increase the efficiency of mi- quantitative PCR (data not shown). In addition, pretreatment with gration of these cells (Fig. 3A). However, when lymphocytes had pertussis toxin inhibited PGD2-induced lymphocyte migration previously migrated across BVEC, the efficiency of migration across LVEC by 41 6 11% (mean 6 SEM, n =4;p , 0.05 by across LVEC was significantly increased (Fig. 3A). Importantly, paired t test), demonstrating a role for GPCR in the movement passage of lymphocytes across cytokine-stimulated BVEC did of lymphocytes across these EC. not increase the efficiency of migration across a second mono- The lymphocyte adhesion receptors that support transit of LVEC layer of BVEC (Fig. 3B), implying that the priming process have not been described. Because integrins are crucial for transit modified lymphocyte responses that were specific to the transit of through BVEC, we tested the effect of function-neutralizing Abs LVEC. against b1, b2, and b3 integrins. Blockade of b1 and b2 integrins As PBL that migrate across cytokine stimulated BVEC are had a significant effect on the efficiency of transmigration (40% highly enriched for CD4 memory T cells, it was possible that the and 80% inhibition, respectively; Fig. 4B). A b3 integrin-blocking enhanced capacity to migrate across LVEC was an intrinsic pro- reagent had no more effect than an isotype-matched control Ab perty of this subset of cells that was not reliant on a priming signal. (Fig. 4B). In addition, the priming effect of BVEC migration did To investigate this, we repeated the above experiment using CD4 not alter the activation status of b2 integrins or F-actin on memory T cells isolated from the mixed lymphocyte populations migrated lymphocytes compared with nonmigrated controls (data found in peripheral blood. Fig. 3C shows that cultured CD4 not shown). memory T cells were poor at migrating across LVEC. Importantly, Migration across BVEC or delivery of exogenous PGD CD4 memory T cells that had experienced migration across BVEC 2 increases the efficiency of CD4 memory T cell migration across again displayed significantly increased migration across LVEC, LVEC and chemotaxis toward the CCR7 ligand CCL21 demonstrating that a priming signal was indeed necessary for efficient T cell migration across LVEC. As lymphocyte transit of BVEC was a PGD2-dependent process, we speculated that this prostanoid signal might also be an Efficient migration of CD4 memory T cells across lymphatic important signal for increasing the efficiency of migration ac- b endothelium is dependent on CCR7 and supported by 1 and ross LVEC. In a separate series of experiments, passage of CD4 b 2 integrins memory T cells across BVEC again delivered a considerable in- The migration of CD4+ T cells from skin into draining lymphatics crease in migration efficiency across LVEC compared with CD4 has been shown to depend on the chemokine receptor CCR7 memory T cells cultured on plastic (Fig. 5A). However, if isolated (12). In this study, when we took CD4 memory T cells that had CD4 memory T cells were cultured on plastic in the presence of migrated through BVEC and treated them with a function- exogenous PGD2 for 24 h, then an increase in efficiency of mi- neutralizing Ab against CCR7, migration through a LVEC mono- gration across LVEC was also obtained (Fig. 5A). Some increase 1436 PGD2 AND LYMPHOCYTE TRAFFIC DURING INFLAMMATION Downloaded from FIGURE 4. CCR7 and b1 and b2 integrins are required for lymphocyte migration across LVEC. A, The effect of Ab treatment against CXCR3 or CCR7 on the migration of CD4+CD45RO+ T cells across cytokine-stimulated (TNF and IFN) LVEC. **p , 0.01 compared with isotype control

by paired t test. B, The effect of Ab treatment against CD29 (b1 integrin), + CD18 (b2 integrin), or CD61 (b3 integrin) on the migration of CD4 CD45RO+ T cells across cytokine-stimulated (TNF and IFN) LVEC. ANOVA showed a significant effect of Ab treatment on T cell migration http://www.jimmunol.org/ FIGURE 3. Migration across BVEC primes lymphocytes for efficient (p , 0.01). Data are mean 6 SEM of three independent experiments. migration across LVEC. A, The effect of stimulation of LVEC with **p , 0.01 compared with isotype control by Dunnett post hoc test. cytokines (TNF and IFN) and/or transmigration of PBL across BVEC on the percentage of added lymphocytes migrating across LVEC. B, The effect of stimulation of a second BVEC monolayer with cytokines (TNF and As CD4 memory T cell migration across LVEC was driven by IFN) and/or transmigration of PBL across BVEC on the percentage of added lymphocytes migrating across a second monolayer of BVEC. C, The CCR7, we examined the effect of transit of these cells across BVEC effect of stimulation of LVEC with cytokines (with TNF and IFN) and/or on the levels of expression of this receptor. We also analyzed other transmigration of CD4+CD45RO+ T cells across BVEC on the percentage chemokine receptors reported to support migration of different of added lymphocytes migrating across LVEC. Data are mean 6 SEM of lymphocyte subsets into lymph nodes via the high endothelial by guest on October 4, 2021 three to six independent experiments. *p , 0.05 compared with control by venule or into the afferent lymphatic vasculature via the tissue. paired t test, **p , 0.01 compared with control by paired t test. Although lymphocytes in peripheral blood expressed CCR7, CXCR5, and CXCR3, expression levels were not altered by the process of migration across BVEC (Table I). Importantly, how- in migration efficiency was evident within 30 min of pretreatment ever, we could show that in a Transwell chemotaxis assay, CD4 memory T cells that had previously been recruited across BVEC with PGD2 (19 6 1.7% compared with untreated cells 14 6 5%; mean 6 SEM, n = 3). Moreover, stimulation of the DP2 receptor or had been stimulated with exogenous PGD2 migrated with sig- with an agonist for 1 h significantly enhanced the migratory ef- nificantly increased efficiency to the CCR7 ligand CCL21 (Fig. ficiency of lymphocytes through LVEC compared with the un- 5C). Thus, the migration of CD4 T cells across BVEC appears to treated control (60 6 2% versus 52 6 2% migration, respectively; enhance the function of CCR7 without increasing its surface ex- mean 6 SEM, n =3;p , 0.05 by paired t test). Importantly, pression on primed cells (Fig. 6). Interestingly, using LVEC cul- activating the DP1 receptor using a specific agonist had no con- tured on plastic to track the migration velocity of lymphocytes sistent or significant effect on the level of lymphocyte migration during migration, we could show that stimulation of DP2 with (data not shown). Once again, cells migrated paracellularly rather a specific agonist significantly reduced the velocity of migration 6 than transcellularly. In addition, we examined whether DP2 stim- underneath LVEC (6.4 0.2 mm/min for untreated compared 6 6 ulation altered the surface expression of CD31 (PECAM-1). Pre- with 4.2 0.2 mm/min for DP2 agonist pretreated PBL; mean , treatment with DP2 agonist had no effect on CD31 surface ex- SEM, n =3;p 0.001 by paired t test). These data show that the pression (data not shown). Thus, PGD2 signaling through the DP2 process of CCR7-mediated diapedesis across LVEC is more effi- receptor primes lymphocytes to move with greater efficiency cient after PGD2 stimulation, but this is not due to increased through LVEC. chemokinesis. We have demonstrated that passage of CD4 memory T cells across the LVEC monolayer is a process dependent upon b1 and b2 Discussion integrins. To determine whether upregulation of these receptors In the current model of lymphocyte trafficking, we have identified provided a mechanism for increased efficiency of LVEC transit, new steps that regulate the migration of memory T cells across the we measured the integrin expression on CD4 memory T cells in EC of both the blood and lymphatic vasculature. Thus, PGD2 is isolated PBL after their migration across BVEC and after culture required for the efficient transendothelial migration of memory on plastic. However, the levels of b1 and b2 integrins were in- T cells across BVEC, and this PGD2-driven step of recruitment sensitive to BVEC migration and not altered by 24-h culture on also primes CD4 memory T cells for efficient migration across plastic (Fig. 5B). LVEC (Fig. 6). During migration across BVEC, the PGD2 signal is The Journal of Immunology 1437

mediated priming for efficient migration across LVEC did not cause upregulation of integrin adhesion receptors or chemokine receptors on CD4 memory T cells, but did modify the function of CCR7 so that the efficiency of chemotaxis toward CCL21 was significantly increased. Current paradigms of T cell trafficking during an inflammatory response assume that the delivery of a chemokine signal is sufficient to promote integrin-mediated stable adhesion, followed by reorganization of the actin cytoskeleton during spreading on the EC surface, eventually leading to transendothelial migration. However, blockade of the PGD2 receptor DP-2 [formally known as chemoattractant receptor-homologous molecule expressed on Th2 cells (CRTh2)] greatly reduces the transmigration of memory T cells, leaving the great majority attached to the apical surface of the EC monolayer. These observations strongly imply the presence of a hierarchy of activation signals that are required for efficient recruitment. Thus, we propose an initial chemokine signal ac- tivates lymphocyte integrins allowing immobilization on the endothelium. In turn, PGD2 permits downstream functions such as migration and diapedesis. Such a model for the recruitment of Downloaded from lymphocytes during inflammation resembles our recently published observations of the recruitment of neutrophils (14). In that study, we showed that migration across EC of the blood vasculature required PGD2-mediated signaling through the DP-1 receptor. This prostanoid signal was again subordinate to an initial chemokine stimulus delivered to the neutrophil by CXCR2. Taken http://www.jimmunol.org/ together, these observations indicate that leukocyte access to inflamed tissue may require both chemokine and prostanoid signals, although the source of PGD2 in these assays is still to be de- finitively identified. Previously, we showed that PGD2 production by cultured EC was at the limits of detection by mass spectrom- + FIGURE 5. Exogenous PGD2 enhances migration of CD4 CD45RO etry (14). However, the use of cyclooxygenase inhibitors in EC T cells across LVEC and their chemotaxis toward CCL21. A, The effect of cultures was very effective at reducing PGD2-mediated leukocyte + + exogenous PGD2 on the migration of CD4 CD45RO T cells across LVEC migration. The implication of these observations is that formation by guest on October 4, 2021 stimulated with cytokines (TNF and IFN). *p , 0.05 compared with of small amounts of this agent by EC may be highly localized and control by paired t test. B, The expression of CD29 (b1 integrin) and CD18 occur in response to leukocyte adhesion to the endothelial cell (b integrin) on CD4+CD45RO+ T cells freshly isolated from peripheral 2 surface rather than being an endogenously synthesized agent re- blood (None), T cells migrated across BVEC (Migrated), or cultured on plastic (Plastic). ANOVA showed no significant effect of lymphocyte leased constitutively. This aspect of prostanoid production requires further elucidation in future studies. treatment on integrin expression. C, The effect of exogenous PGD2 on the migration of CD4+CD45RO+ T cells toward CCL21 in Transwell che- The process of migrating across BVEC primed CD4 memory motaxis assay. Lymphocytes were cultured on plastic for 24 h (None), T cells for efficient migration across LVEC. This phenomenon migrated across BVEC (Migrated), or treated with exogenous PGD2. One- could be recapitulated by treating PBL with PGD2, strongly im- way ANOVA showed a significant effect of lymphocyte treatment on plying that this eicosanoid promoted passage across both EC chemotaxis (p , 0.05). Data mean are 6 SEM of three (B) or four in- monolayers. The mechanism by which T cell migration across dependent experiments. *p , 0.05 compared with untreated by Dunnett LVEC was expedited was also identified. Thus, either migration post hoc test. across BVEC or exogenous PGD2 could promote the function of the CCR7 without increasing its surface expression. Evidence for subordinate to a primary activating stimulus delivered by che- this came through the use of a function-neutralizing Ab against mokines of the IFN inducible family (CXCL9–11), which promote CCR7 and also blockade of GPCR signaling, both of which sig- stable integrin-mediated adhesion via the CXCR3 chemokine re- nificantly retarded CD4 memory T cell migration across LVEC. In ceptor. However, without receipt of the PGD2 signal, memory addition, CD4 memory T cells, which had migrated across BVEC T cells do not traffic efficiently across BVEC or LVEC. PGD2- or received an exogenous PGD2 stimulus, were capable of migrating more efficiently toward the CCR7 ligand CCL21 in a Transwell chemotaxis assay. The use of CCR7 by CD4 T cells for Table I. Mean fluorescent intensities for flow cytometry of chemokine migration across LVEC is in accordance with previous observa- receptor expression on peripheral blood lymphocytes before and after migration across BVEC tions made in murine models, in which T cells lacking the receptor were unable to efficiently traffic into the lymphatic vasculature PBL Migrated Across from the skin (12) or the lung (13). A role for prostanoids in Chemokine Receptor Nonmigrated PBL BVEC regulating the function of CCR7 in T cells has not previously been demonstrated, although the phenomenon has close parallels with CCR7 46.12 6 10.14 36.23 6 6.271 CXCR3 22.79 6 2.481 16.77 6 1.227 the regulated migration of DCs out of peripheral tissue and into CXCR5 33.34 6 6.286 22.56 6 9.035 the afferent lymphatic vasculature. In this case, the differentiation Data are mean 6 SEM of five experiments. ANOVA showed no significant effect and maturation of DCs within peripheral tissue is driven by PGE2 of migration on the expression of chemokine receptors. (21–24). This prostanoid signal promotes the function (but not the 1438 PGD2 AND LYMPHOCYTE TRAFFIC DURING INFLAMMATION

FIGURE 6. New steps in the regulation of CD4+CD45RO+ T cell trafficking that are dependent upon PGD2. 1, BVEC stimulated with cytokines (TNF and IFN) selectively recruit CD4+CD45RO+ T cells from flowing blood. 2, The T cells receive an activating stimulus from IFN-g–inducible chemokines (CXCL9– 11), which operate exclusively through the CXCR3 receptor. 3, T cell integrins are tran- siently activated and immobilize the cell on the Downloaded from surface of BVEC. 4, PGD2 stimulates the DP-2 receptor (formally CRTh2), a signal that is essential for efficient passage across the BVEC monolayer. 5, Receipt of a PGD2 signal during passage of the BVEC or stimulation with exogenous PGD2 promotes the function, but not the expression, of the lymphocyte chemokine http://www.jimmunol.org/ receptor CCR7. 6, CD4+CD45RO+ T cells migrate across LVEC in response to the chemokines CCL19 and CCL21 with significantly increased efficiency. by guest on October 4, 2021

expression) of the chemokine receptor CCR7, so that DCs also E-selectin, VCAM-1, and ICAM-1 on cytokine-stimulated LVEC respond more efficiently to the chemokines CCL19 and CCL21. and demonstrated that they function to support migration of DCs The striking similarity of the pathways used by DCs and CD4 across LVEC (26–28). In this study, we show for the first time, memory T cells for exit from peripheral tissue may indicate that as to our knowledge, that passage of CD4 memory T cells across the adaptive immune system has evolved, the different populations LVEC requires both b1 and b2 integrins. Interestingly, the PGD2- of effector cells have adopted a similar mechanism for localization dependent priming of these cells for efficient transit of the LVEC in secondary lymphoid tissue, except that their emigration from did not require upregulation of integrin receptors that were highly tissue is regulated by different prostanoids. expressed on this subset of T cells. However, taken together, these Although CCR7 regulates CD4 memory T cell trafficking across studies indicate that migration of leukocytes across LVEC uses LVEC, other aspects of adhesion and migration of these cells across adhesion receptors that are remarkably similar to those used for LVEC are less well characterized. However, it does appear that transit of the BVEC barrier. sphingosine 1-phosphate (S1P) is a potent retention signal for Although the importance of PGD2 for successful navigation of tissue-resident T cells (25). In vivo activation via the S1P receptor, both blood vascular and lymphatic vascular EC by memory T cells S1P1, retarded migration into the lymphatic vasculature, whereas is a novel observation, an important role for PGD2 and its receptor in vitro, activation of polarized T cells through the S1P1 receptor DP-2 in memory T cell biology has been previously reported. caused firm adhesion of these cells to the basal surface of lym- Indeed, Nagata and colleagues (29) demonstrated the presence of phatic but not vascular EC (25). Such retention signals do not a G protein-coupled orphan receptor CRTh2 on CD4+ and CD8+ preclude a role for PGD2 in encouraging migration away from memory T cells before they identified its cognate ligand, PGD2 tissue. Indeed, they might represent an additional step in the relay (30). Interestingly, others claimed that this receptor was the most of directional cues that control entry, residence, and removal of reliable marker for some memory T cell subsets (31). In these leukocytes from tissue during the evolution of an inflammatory studies, signaling by PGD2 via the DP-2 receptor could upregulate response. Other studies have identified adhesion receptors such the b2 integrin CD11b and the TNF superfamily member CD154 The Journal of Immunology 1439

(CD40L) and could induce chemotactic responses in some mem- 15. Rainger, G. E., P. Stone, C. M. Morland, and G. B. Nash. 2001. A novel system for investigating the ability of smooth muscle cells and fibroblasts to regulate ory T cells (30, 32). Moreover, PGD2 signaling through DP-2 was adhesion of flowing leukocytes to endothelial cells. J. Immunol. Methods 255: shown to exacerbate cutaneous (33) and pulmonary (34) allergic 73–82. inflammation in rodent models. Thus, it is now becoming apparent 16. Cooke, B. M., S. Usami, I. Perry, and G. B. Nash. 1993. A simplified method for culture of endothelial cells and analysis of adhesion of blood cells under con- that prostanoid signals operating through the DP-2 receptor are ditions of flow. Microvasc. Res. 45: 33–45. important regulators of both memory T cell trafficking and im- 17. McGettrick, H. M., K. Hunter, P. A. Moss, C. D. Buckley, G. E. Rainger, and mune function. G. B. Nash. 2009. Direct observations of the kinetics of migrating T cells suggest active retention by endothelial cells with continual bidirectional migration. J. In conclusion, the current paradigms that postulate an exclusive Leukoc. Biol. 85: 98–107. role for chemokines in the recruitment of memory T cells from the 18. Stanley, P., A. Smith, A. McDowall, A. Nicol, D. Zicha, and N. Hogg. 2008. blood require updating (see Fig. 6 for details). Although an initial Intermediate-affinity LFA-1 binds alpha-actinin-1 to control migration at the leading edge of the T cell. EMBO J. 27: 62–75. chemokine signal is essential for the activation of T cells on 19. Butler, L. M., G. E. Rainger, M. Rahman, and G. B. Nash. 2005. Prolonged BVEC, a downstream prostanoid signal, delivered by PGD2 via culture of endothelial cells and deposition of basement membrane modify the the DP-2 receptor, is required to drive the process of diapedesis. In recruitment of neutrophils. Exp. Cell Res. 310: 22–32. 20. Chakravorty, S. J., H. M. McGettrick, L. M. Butler, C. D. Buckley, addition, passage of the blood vasculature and receipt of a PGD2 G. E. Rainger, and G. B. Nash. 2006. An in vitro model for analysing neutrophil signal significantly increases the efficiency of passage across EC migration into and away from the sub-endothelial space: Roles of flow and CD31. Biorheology 43: 71–82. of the lymphatic vasculature by modulating the function of CCR7. 21. Scandella, E., Y. Men, S. Gillessen, R. Fo¨rster, and M. Groettrup. 2002. Pros- taglandin E2 is a key factor for CCR7 surface expression and migration of Disclosures monocyte-derived dendritic cells. Blood 100: 1354–1361. 22. Luft, T., M. Jefford, P. Luetjens, T. Toy, H. Hochrein, K. A. Masterman, The authors have no financial conflicts of interest, with the exception of

C. Maliszewski, K. Shortman, J. Cebon, and E. Maraskovsky. 2002. Functionally Downloaded from M.R., who is employed by AstraZeneca but will not benefit financially distinct dendritic cell (DC) populations induced by physiologic stimuli: prosta- from this work. glandin E(2) regulates the migratory capacity of specific DC subsets. Blood 100: 1362–1372. 23. Scandella, E., Y. Men, D. F. Legler, S. Gillessen, L. Prikler, B. Ludewig, and References M. Groettrup. 2004. CCL19/CCL21-triggered signal transduction and migration of dendritic cells requires prostaglandin E2. Blood 103: 1595–1601. 1. Mora, J. R., and U. H. von Andrian. 2006. T-cell homing specificity and plas- 24. Legler, D. F., P. Krause, E. Scandella, E. Singer, and M. Groettrup. 2006. ticity: new concepts and future challenges. Trends Immunol. 27: 235–243. Prostaglandin E2 is generally required for human dendritic cell migration and

2. Marelli-Berg, F. M., L. Cannella, F. Dazzi, and V. Mirenda. 2008. The highway http://www.jimmunol.org/ exerts its effect via EP2 and EP4 receptors. J. Immunol. 176: 966–973. code of T cell trafficking. J. Pathol. 214: 179–189. 25. Ledgerwood, L. G., G. Lal, N. Zhang, A. Garin, S. J. Esses, F. Ginhoux, 3. Springer, T. A. 1995. Traffic signals on endothelium for lymphocyte recircula- M. Merad, H. Peche, S. A. Lira, Y. Ding, et al. 2008. The sphingosine 1- tion and leukocyte emigration. Annu. Rev. Physiol. 57: 827–872. phosphate receptor 1 causes tissue retention by inhibiting the entry of pe- 4. Luscinskas, F. W., H. Ding, and A. H. Lichtman. 1995. P-selectin and vascular ripheral tissue T lymphocytes into afferent lymphatics. Nat. Immunol. 9: 42– cell adhesion molecule 1 mediate rolling and arrest, respectively, of CD4+ T lymphocytes on tumor necrosis factor alpha-activated vascular endothelium 53. under flow. J. Exp. Med. 181: 1179–1186. 26. Sawa, Y., and E. Tsuruga. 2008. The expression of E-selectin and chemokines in 5. Jones, D. A., L. V. McIntire, C. W. Smith, and L. J. Picker. 1994. A two-step the cultured human lymphatic endothelium with lipopolysaccharides. J. Anat. adhesion cascade for T cell/endothelial cell interactions under flow conditions. J. 212: 654–663. Clin. Invest. 94: 2443–2450. 27. Sawa, Y., E. Tsuruga, K. Iwasawa, H. Ishikawa, and S. Yoshida. 2008. Leukocyte 6. Yago, T., M. Tsukuda, H. Yamazaki, T. Nishi, T. Amano, and M. Minami. 1995. adhesion molecule and chemokine production through lipoteichoic acid recog- nition by toll-like receptor 2 in cultured human lymphatic endothelium. Cell Analysis of an initial step of T cell adhesion to endothelial monolayers under by guest on October 4, 2021 flow conditions. J. Immunol. 154: 1216–1222. Tissue Res. 333: 237–252. 7. Piali, L., C. Weber, G. LaRosa, C. R. Mackay, T. A. Springer, I. Clark-Lewis, and 28. Johnson, L. A., S. Clasper, A. P. Holt, P. F. Lalor, D. Baban, and D. G. Jackson. B. Moser. 1998. The chemokine receptor CXCR3 mediates rapid and shear- 2006. An inflammation-induced mechanism for leukocyte transmigration across resistant adhesion-induction of effector T lymphocytes by the chemokines lymphatic vessel endothelium. J. Exp. Med. 203: 2763–2777. IP10 and Mig. Eur. J. Immunol. 28: 961–972. 29. Nagata, K., K. Tanaka, K. Ogawa, K. Kemmotsu, T. Imai, O. Yoshie, H. Abe, 8. McGettrick, H. M., E. Smith, A. Filer, S. Kissane, M. Salmon, C. D. Buckley, K. Tada, M. Nakamura, K. Sugamura, and S. Takano. 1999. Selective expression G. E. Rainger, and G. B. Nash. 2009. Fibroblasts from different sites may pro- of a novel surface molecule by human Th2 cells in vivo. J. Immunol. 162: 1278– mote or inhibit recruitment of flowing lymphocytes by endothelial cells. Eur. J. 1286. Immunol. 39: 113–125. 30. Hirai, H., K. Tanaka, O. Yoshie, K. Ogawa, K. Kenmotsu, Y. Takamori, 9. Cinamon, G., V. Shinder, and R. Alon. 2001. Shear forces promote lymphocyte M. Ichimasa, K. Sugamura, M. Nakamura, S. Takano, and K. Nagata. 2001. migration across vascular endothelium bearing apical chemokines. Nat. Immu- Prostaglandin D2 selectively induces chemotaxis in T helper type 2 cells, nol. 2: 515–522. eosinophils, and basophils via seven-transmembrane receptor CRTH2. J. Exp. 10. Marelli-Berg, F. M., H. Fu, F. Vianello, K. Tokoyoda, and A. Hamann. 2010. Med. 193: 255–261. Memory T-cell trafficking: new directions for busy commuters. Immunology 130: 31. Cosmi, L., F. Annunziato, M. I. G. Galli, R. M. E. Maggi, K. Nagata, and 158–165. S. Romagnani. 2000. CRTH2 is the most reliable marker for the detection of 11. Pereira, J. P., L. M. Kelly, and J. G. Cyster. 2010. Finding the right niche: B-cell circulating human type 2 Th and type 2 T cytotoxic cells in health and disease. migration in the early phases of T-dependent antibody responses. Int. Immunol. Eur. J. Immunol. 30: 2972–2979. 22: 413–419. 32. Tanaka, K., H. Hirai, S. Takano, M. Nakamura, and K. Nagata. 2004. Effects of 12. Debes, G. F., C. N. Arnold, A. J. Young, S. Krautwald, M. Lipp, J. B. Hay, and prostaglandin D2 on helper T cell functions. Biochem. Biophys. Res. Commun. E. C. Butcher. 2005. Chemokine receptor CCR7 required for T lymphocyte exit 316: 1009–1014. from peripheral tissues. Nat. Immunol. 6: 889–894. 33. Satoh, T., R. Moroi, K. Aritake, Y. Urade, Y. Kanai, K. Sumi, H. Yokozeki, 13. Bromley, S. K., S. Y. Thomas, and A. D. Luster. 2005. Chemokine receptor H. Hirai, K. Nagata, T. Hara, et al. 2006. Prostaglandin D2 plays an essential role CCR7 guides T cell exit from peripheral tissues and entry into afferent lym- in chronic allergic inflammation of the skin via CRTH2 receptor. J. Immunol. phatics. Nat. Immunol. 6: 895–901. 177: 2621–2629. 14. Tull, S. P., C. M. Yates, B. H. Maskrey, V. B. O’Donnell, J. Madden, 34. Shiraishi, Y., K. Asano, K. Niimi, K. Fukunaga, M. Wakaki, J. Kagyo, R. F. Grimble, P. C. Calder, G. B. Nash, and G. E. Rainger. 2009. Omega-3 Fatty T. Takihara, S. Ueda, T. Nakajima, T. Oguma, et al. 2008. Cyclooxygenase-2/ acids and inflammation: novel interactions reveal a new step in neutrophil re- prostaglandin D2/CRTH2 pathway mediates double-stranded RNA-induced en- cruitment. PLoS Biol. 7: e1000177. hancement of allergic airway inflammation. J. Immunol. 180: 541–549.