IMMUNOBIOLOGY

CCL21 mediates CD4! T-cell costimulation via a DOCK2/Rac-dependentpathwayKathrin Gollmer,1 Francois Asperti-Boursin,2,3 Yoshihiko Tanaka,4,5 Klaus Okkenhaug,6 Bart Vanhaesebroeck,7

Jeffrey R. Peterson,8 Yoshinori Fukui,4,5 Emmanuel Donnadieu,2,3 and Jens V. Stein1

1Theodor Kocher Institute, University of Bern, Bern, Switzerland; 2Institut Cochin, Universite Paris Descartes, Centre National de la Recherche Scientifique(UMR 8104), Paris, France; 3Inserm (U567), Paris, France; 4Division of Immunogenetics, Department of Immunobiology and Neuroscience, Medical Institute ofBioregulation, Kyushu University, Fukuoka, Japan; 5Japan Science and Technology, CREST, Tokyo, Japan; 6Laboratory of Lymphocyte Signaling andDevelopment, Babraham Institute, Babraham Research Campus, Cambridge, United Kingdom; 7Center for Cell Signaling, Institute of Cancer, Queen MaryUniversity of London, London, United Kingdom; and 8Department of Basic Sciences, Fox Chase Cancer Center, Philadelphia, PA

CD4! T cells use the chemokine receptorCCR7 to home to and migrate withinlymphoid tissue, where T-cell activationtakes place. Using primary T-cell receptor(TCR)–transgenic (tg) CD4! T cells, weexplored the effect of CCR7 ligands, inparticular CCL21, on T-cell activation. Wefound that the presence of CCL21 duringearly time points strongly increased invitro T-cell proliferation after TCR stimula-tion, correlating with increased expres-sion of early activation markers. CCL21

costimulation resulted in increased Ras-and Rac-GTP formation and enhancedphosphorylation of Akt, MEK, and ERKbut not p38 or JNK. Kinase-deadPI3K"D910A/D910A or PI3K#-deficient TCR-tgCD4! T cells showed similar responsive-ness to CCL21 costimulation as controlCD4! T cells. Conversely, deficiency inthe Rac guanine exchange factor DOCK2significantly impaired CCL21-mediatedcostimulation in TCR-tg CD4! T cells, con-comitant with impaired Rac- but not Ras-

GTP formation. Using lymph node slicesfor live monitoring of T-cell behavior andactivation, we found that G protein-coupled receptor signaling was requiredfor early CD69 expression but not forCa2! signaling. Our data suggest that thepresence of CCL21 during early TCR sig-naling lowers the activation thresholdthrough Ras- and Rac-dependent path-ways leading to increased ERK phosphor-ylation. (Blood. 2009;114:580-588)

Introduction

Naive T cells continuously traffic to secondary lymphoid organs,including peripheral lymph nodes (PLNs), where they screenantigen-presenting cells (APCs), in particular dendritic cells (DCs),for the presence of specific peptide Ag presented on MHC (pMHC)complexes.1 In proinflammatory conditions, DCs expressing cog-nate pMHC and costimulatory signals, such as B7 molecules,induce efficient T-cell activation through the T-cell receptor (TCR)and CD28.2 This leads to an early signaling response characterizedby activation of multiple signaling pathways, including tyrosinekinase cascades, sustained increase in intracellular Ca2!, andactivation of phosphoinositide-3-kinase (PI3K), small GTPases ofthe Ras and Rho family, mitogen-activated protein kinase, nuclearfactor of activated T-cell, and nuclear factor-"B. Activated T cellssubsequently increase surface expression of early activation mark-ers, such as CD69 and CD25, produce interleukin-2 (IL-2), expandclonally, and differentiate into effector cells.

Direct observations of lymphocytes and DCs presenting cog-nate pMHC complexes in explanted PLNs or live mice using2-photon microscopy have uncovered a dynamic range of cellularinteractions within lymphoid tissue. In some settings, T cellsalmost immediately arrest on encountering DCs.3 Alternatively,T cells were observed to continue to migrate during the first severalhours after entry into lymphoid tissue along the stromal networkformed by fibroblastic reticular cells (FRCs), where they under-went brief serial contacts with DCs.4 This first phase of high

motility is reminiscent of naive T-cell migration in the absence ofpMHC-loaded DCs.4,5 Of note, even during continuous motilityduring initial DC encounters, T cells integrate TCR-derived sig-nals, as they gradually increase CD44 and CD69 surface levels.6

Differences in T-cell deceleration in distinct models are probablythe result of variations in total Ag load and TCR-MHC affinity,which influence the time T cells require to reach a threshold forefficient arrest and formation of long-lasting contacts with DCs.6-8

FRCs in the T-cell area of PLN express the homeostaticchemokines CCL19 and CCL21.9,10 Their G-protein coupledreceptor (GPCR) CCR7 is highly expressed on naive T cells andcontributes to random motility during DC scanning in vitro and invivo.11-16 TCR signaling events are thus spatiotemporally tightlyconnected to chemokine receptor signaling, suggesting a potentialcrosstalk between both pathways during early T-cell activation.Indeed, chemokines have been shown to contribute to T-cellactivation in at least 2 ways. First, in vitro assays showed thatCCL19 and CCL21 indirectly contribute to lymphocyte activationby allowing efficient screening of rare Ag-bearing DCs.11,12 DC-bound CCL21 was also found to sensitize CD4! T cells to pMHCcomplexes on neighboring DCs, correlating with efficient DCscanning of the leading edge of polarized T cells.17 These observa-tions support a role for CCR7 ligands in promoting efficientencounters with other cell types present in lymph nodes, such asDCs or B cells, through increased motility and scanning.

Submitted January 27, 2009; accepted April 22, 2009. Prepublished online asBlood First Edition paper, May 18, 2009; DOI 10.1182/blood-2009-01-200923.

The online version of this article contains a data supplement.

The publication costs of this article were defrayed in part by page chargepayment. Therefore, and solely to indicate this fact, this article is herebymarked ‘‘advertisement’’ in accordance with 18 USC section 1734.

© 2009 by The American Society of Hematology

580 BLOOD, 16 JULY 2009 ! VOLUME 114, NUMBER 3

Independent of their chemoattractant and guiding activities,chemokines also act directly as costimulatory factors and conse-quently modulate the outcome of an immune response.18-20 CCL5costimulates Jurkat T cells by recruiting its receptor CCR5 andGq/11 to the immunologic synapse without inducing migration.21

Similarly, CXCL12 and CCL21 increase anti-CD3-induced T-cellactivation, suggesting that migration-independent chemokine recep-tor- and TCR-triggered signals are combined for optimal T-cellactivation.19,22 Thus far, the underlying molecular mechanisms ofchemokine-mediated costimulation and intracellular integratorsacting both downstream TCR and chemokine receptors remainincompletely described. In addition, it is not known whetherchemokines costimulate T cells in a lymphoid environment.

Chemokine-induced cell migration, scanning of Ag-bearingDCs, and subsequent formation of a stable immunologic synapserequire cytoskeletal reorganizations in T cells.23,24 The small GT-Pase Rac is a key modulator of F-actin polymerization in cells andis important for cell migration.25 The hematopoietic lineage-restricted guanine exchange factor (GEF) DOCK2 catalyzes GDP-GTP exchange of Rac downstream chemokine receptors. Accord-ingly, DOCK2-deficient lymphocytes migrate poorly to homeostaticchemokines in chemotaxis assays, 2-dimensional surfaces, andwithin lymph nodes.26-28 DOCK2 also acts downstream the TCR topromote Rac-GTP formation. In the absence of DOCK2, clusteringof TCR and lipid rafts, and to a lesser extent ERK phosphorylation,are impaired, resulting in decreased DC-induced T-cell prolifera-tion.29 Furthermore, PI3K isoforms are involved in DOCK2-independent chemokine-induced lymphocyte migration.30,31 PI3K#has also been reported to contribute to T-cell proliferation insome32,33 but not all studies.31 PI3K$ is a major class IA isoformexpressed in immune cells and is activated by tyrosine kinasesdownstream T- and B-cell antigen receptors.34,35 In vitro, B cellsexpressing a kinase-dead PI3K$ mutant, p110D910A/D910A, showdecreased migration toward CXCL13, whereas T-cell migration toCCR7 ligands appears not to be affected.36 Thus, similar toDOCK2, PI3K isoforms act downstream TCR and chemokinereceptors, making them potential transmitters of chemokine-mediated costimulation.

Here, we report a detailed analysis of the costimulatory effect ofchemokines expressed in the T-cell area of secondary lymphoidorgans. Using in vitro proliferation assays of various TCR tg mouselymphocytes, we identify CCL21 as having the greatest costimula-tory activity during early TCR signaling, in particular at suboptimalstimulation. We performed a biochemical and genetic analysis ofsignaling molecules underlying the costimulatory activity, focusingon pathways known to be triggered by both receptor classes. Ourdata suggest a model in which CCL21-mediated interstitial cellmotility and TCR signaling are integrated, in part via DOCK2,during the early, promigratory phase of T-cell–DC encounters, thuslowering the activation threshold in lymphoid tissue.

Methods

Mice

Four- to 12-week-old control or p110#-deficient and p110$D910A/D910A–knock-in mice backcrossed to the OT-II TCR tg C57BL/6 background and2B4 TCR tg DOCK2-deficient mice on a B10.BR background weredescribed before.29,37 Female Marylin (anti–H-Y peptide called DBY)TCR-transgenic Rag2%/% CD45.1!/! mice were obtained from the CDTA.C57BL/6, BALB/c, and DO11.10 mice were bred at the Theodor KocherInstitute or purchased from Harla. All experiments were performed inaccordance with approval from the Swiss Kanton of Bern Veterinary Officeand French veterinary animal experimentation regulations.

Reagents

All antibodies and other reagents are listed in supplemental data (availableon the Blood website; see the Supplemental Materials link at the top of theonline article).

Proliferation assays

Primary T cells were purified from spleen and PLN single-cell suspensionsvia negative selection using antibody-coated magnetic beads (& 90%purity; Dynal). Isolated T cells (5 ' 105 cells/mL) were either plated on96-well plates coated with anti-CD3( monoclonal antibodies (mAbs) orstimulated with soluble anti-CD3( mAb and cultured in RPMI 1640/10%fetal calf serum/standard supplements, in the presence or absence ofanti-CD28 mAb or chemokines (100 nM final concentration). [3H]Thymi-dine was added for the last 16 hours of a 48- or 72-hour culture, followed byquantification in a scintillation counter. Where indicated, T cells werepretreated with 400 ng/mL pertussis toxin (PTX; 30 minutes at 37°C, 7%CO2), washed twice, and cultured as described. For activation of TCR-transgenic CD4! T cells, isolated T cells (4 ' 106 cells/mL) were culturedfor 48 or 72 hours with irradiated congenic spleen cells (5 ' 106 cells/mL)in the presence or absence of 100 nM chemokine and indicated concentra-tions of agonist peptide. [3H]Thymidine incorporation was determined asdetailed earlier in this paragraph.

In some experiments, T cells were loaded with carboxyfluoresceinsuccinimidyl ester (CFSE) (1 )M, 30 minutes at 37°C, 7% CO2) beforeactivation and analyzed for dye dilution by flow cytometry (FACScalibur,BD Biosciences). From obtained histograms, a proliferation index wascalculated as follows. The percentage of divided cells in the presence ofanti-CD3( mAb alone (without chemokines), or, in case of APC-triggeredproliferation, at a peptide concentration of 0.1 )g/mL without chemokines,was arbitrarily normalized to “1.” Other percentages were adjustedaccordingly, with their ratio to “1” being the proliferation index. For Figure2A, the cpm value of 0.1 )g/mL without chemokines was normalized to “1”and the other cpm values expressed as proliferation index accordingly.

Flow cytometry

TCR Tg DO11.10 T cells were activated by chicken or turkey OVA323-339-pulsed irradiated splenocytes in the presence or absence of CCL21(100 nM) for indicated times and stained for CD25 and CD69. Formeasurement of intracellular IL-2, DO11.10 T cells were restimulated withionomycin/phorbol myristate acetate in the presence of brefeldin A for3 hours and labeled with mAbs against Thy1.2, KJI-26, and IL-2 followingthe manufacturer’s instructions (BD Biosciences PharMingen).

Immunoblotting

Isolated T cells (5 ' 107 cells/mL) were incubated with anti-CD3( mAb(5 )g/mL; 4°C, 20 minutes) after overnight incubation in RPMI 1640/0.5%fatty acid free bovine serum albumin. Cells were stimulated by crosslinkingof primary antibody with goat anti–hamster-IgG Ab (G94-56, 20 )g/mL) at37°C for indicated times, in the presence or absence of CCL21 (100 nM).For quantification, all blots were normalized to the loading control. Foldincrease over background was calculated using ImageJ software oraccording to the manufacturer’s instructions (LI-COR Biosciences). Addi-tional information is available in the supplemental data.

Time-lapse imaging

Splenocytes were washed in Hanks Balanced Salt Solution and 4 ' 105

cells were left to settle to glass coverslips for 15 minutes at 37°C. Inparallel, 5 ' 105 T cells prepared from peripheral and mesenteric lymphnodes were incubated for 5 minutes at 37°C with 0.5 )M 5-chloromethyl-fluorescein diacetate (CMFDA). When indicated, T cells were treated with1 )g/mL CCL21 for 5 minutes at 37°C before adding to the splenocyteslayer. Imaging was performed as described16 on an inverted microscope in aheating chamber (37°C). Images were acquired every 10 seconds during10 to 15 minutes using MetaFluor software (Molecular Devices). Celldisplacements were analyzed with Imaris software (Bitplane).

CHEMOKINE COSTIMULATION OF CD4! T CELLS 581BLOOD, 16 JULY 2009 ! VOLUME 114, NUMBER 3

PLN slice preparation and video imaging

PLN slice preparation was performed as previously described16 with somemodifications (supplemental data). The preparation was perfused at a rate of1 mL/minute with RPMI without phenol red medium bubbled with amixture of 95% O2 and 5% CO2. A single section located 20 to 30 )m fromthe surface of slice was acquired every 30 seconds. Five minutes afterbeginning of image acquisition, slices were perfused with oxygenatedRPMI medium containing DBY peptide at indicated concentrations.Fura-2-loaded T cells were alternatively excited at 350 and 380 nm andemission at 510 nm was used for analysis of Ca2! responses usingMetaFluor software. Ca2! values were represented as the fluorescenceintensity ratio at 340/380 nm. T cells were considered responsive when theamplitude of their responses reached at least twice that of the background.When several Ca2! traces were averaged, the rising phases of the traceswere synchronized. T-cell motility was analyzed with Imaris software.

Measurement of CD69 expression on T cells activated withinPLN slices

PLN slices overlaid with CMFDA-loaded T cells were treated withindicated concentrations of DBY peptide at 37°C, 6% CO2. After 2 hours,slices were washed and mechanically dissociated using 30-G needles toobtain single-cell suspensions. Cells were stained with phycoerythrin(PE)–labeled anti-CD69 and PE-Cy5-labeled anti-CD45.1 for analysis on aFACScan (BD Biosciences).

Statistical analysis

Data were analyzed using Prism software (GraphPad Software). Studentt test was used for statistical analysis, unless indicated otherwise. Signifi-cance was set at P less than .05.

Results

Homeostatic chemokines act as costimulatory factors duringmAb- or APC-induced T-cell activation

The lymphoid tissue-expressed chemokines CCL21 and CXCL12have been shown to act as costimulatory factors during T-cellactivation.17,19,22 In CFSE dilution experiments, we confirmed thatCCL21 enhanced proliferation of T cells activated with plate-

bound anti-CD3( mAb alone or in combination with anti-CD28mAb (Figure 1A-B). CXCL12 and CCL19 were not effective,although CCL19 showed a nonsignificant tendency to increaseproliferation on plates coated with anti-CD3( mAb alone (data notshown). Similar results were obtained when stimulating withsoluble anti-CD3( mAb (data not shown). No increased prolifera-tion was observed in the presence of human CCL2 or heat-inactivated CCL21 (data not shown). Thus, CCR7 ligands, espe-cially CCL21, possess higher costimulatory potency than theCXCR4 ligand CXCL12, reflecting receptor expression levels onmurine naive T cells. In the conditions used here, chemokinesalone did not induce T-cell proliferation (Figure 1B), whereas bothCCL19 and CCL21 slightly increased the percentage of live(7-amino-actinomycin D-negative) T cells after 3 days of culturefrom 18.4% plus or minus 1.1% to 27% plus or minus 1% and27.4% plus or minus 0.7%, respectively. For subsequent experi-ments, we focused on CCL21, which is expressed by FRCs at100-fold higher levels than CCL19 and therefore probably physio-logically more relevant.38

To evaluate the costimulatory potential of CCL21 duringphysiologic TCR activation, we activated DO11.10 TCR tg CD4!

T cells with either chicken or turkey OVA323-339 peptide-pulsedirradiated splenocytes as APCs. Proliferation induced by bothpeptides was increased in the presence of CCL21 (Figure 1C). Therelative increase in the presence of CCL21 was particularlynoticeable at low peptide concentrations and with the low affinityturkey OVA323-339 peptide, representing suboptimal T-cell activa-tion conditions (Figure 1C). In line with previous observations,17

inhibition of G*i signaling by preincubation of T cells with PTXcompletely blocked the costimulatory effect of CCL21, indicatingthat CCL21 acted directly on T cells (Figure 1D).

CCL19 and CCL21 have been reported to induce cell motilityunder certain experimental conditions, thus increasing the likeli-hood of T-cell–DC encounters.11,12 We examined whether motility,rather than direct signaling, was involved in CCL21-mediatedcostimulation. Using time-lapse videomicroscopy, we followedindividual CD4! T cells coincubated with splenocytes on glasscoverslips during 10-minute observation periods in the presence or

Figure 1. Homeostatic chemokines function as costimulatory factors during Ab- and peptide-induced T-cell activation. (A) Flow cytometry histogram of CFSE-labeledT-cell division after 48 and 72 hours of activation with anti-CD3( mAb in the presence (bold line) or absence (gray fill) of 100 nM CCL21. Data are representative from 1 of6 independent experiments. (B) Proliferation in the presence and absence of homeostatic chemokines. CFSE-labeled T cells were stimulated with plate-bound anti-CD3( mAbwith or without anti-CD28 mAb in the presence or absence of CCL21, CCL19, or CXCL12 (100 nM). Proliferation was determined after 48 hours by fluorescence-activated cellsorter and normalized to fold increase compared with cells cultured with anti-CD3( mAb (proliferation index). The presence of chemokines increases the percentage of cellshaving undergone cell divisions within 48 hours. Data are pooled from 4 to 6 independent experiments. Statistical significance was determined using analysis of variancecomparison of anti-CD3 or anti-CD3/CD28–stimulated T-cell proliferation with or without chemokines. *P + .05 compared with “no chemokine.” (C) DO11.10 T cells werecocultured with chicken or turkey OVA323-339-loaded congenic splenocytes in the presence or absence of CCL21 (100 nM). T-cell activation was determined by3H-thymidine-incorporation after 48 hours. Numbers indicate fold increase of proliferation in the presence of CCL21. *P + .05 compared with “no chemokine.” (D) Prioractivation as in panel C; DO11.10 T cells were incubated with PTX. Data are pooled from 2 independent experiments.

582 GOLLMER et al BLOOD, 16 JULY 2009 ! VOLUME 114, NUMBER 3

absence of CCL21. In the absence of CCL21, very few T cellsdisplayed a motile behavior. CCL21 did not increase the percentageof motile T cells under these conditions, although cells were moreelongated (supplemental Figure 1; supplemental Videos 1-2).Taken together, our data suggest that, in the conditions usedhere, CCL21 acts mainly as a costimulatory factor independentof motility.

CCL21-mediated costimulation is required during an early timewindow

In lymphoid tissue containing few pMHC complexes, T-cellactivation starts with an early phase of continuous motility mostprobably codependent on CCL21, followed by stable T-cell–DCinteractions.4,5,8,39 We thus examined the effect of CCL21 on T-cellactivation triggered by low and high pMHC levels during early orlate time points in vitro. Addition of CCL21 to DO11.10 TCR tgCD4! T cells at 8 or 24 hours after activation led to a gradualdecrease in pMHC-stimulated proliferation, indicating that forefficient costimulation, presence of CCL21 is required within anearly time frame of T-cell activation (Figure 2A). Similarly,addition of neutralizing anti-CCL21 Abs to the T-cell culture onlyblocked chemokine-induced costimulation when added from thestart but not when added 8 hours after activation (data not shown).

In line with a priming effect of CCL21, the percentage of T cellsexpressing the early activation marker CD69 was doubled as soonas 8 hours after activation in the presence of CCL21, with cellsexpressing higher mean CD69 levels (Figure 2B). We also ob-served an increase in the percentage of CD25! DO11.10 CD4!

T cells 8 hours after activation (Figure 2B), paralleled by enhancedIL-2 production 24 hours after antigen-specific stimulation (Figure2C). Taken together, T-cell costimulation through CCL21 iseffective during an early time window and directly increases thepercentage and expression levels of early activation markers.

To examine whether CCR7-mediated costimulation was ageneral phenomenon, we stimulated peripheral blood humanT cells with anti-CD3 mAbs in the presence or absence of CCL19or CCL21. Addition of CCL19 accelerated the initiation ofTCR-triggered Ca2! increase but did not affect the amplitude of theresponse (supplemental Figure 2A). Furthermore, simultaneousactivation of TCR and CCR7 resulted in a marked increase in cells

expressing CD69, with CCL19 being more potent than CCL21, incontrast to mouse lymphocytes (supplemental Figure 2B). Alto-gether, these results support a conserved role for CCR7 duringT-cell activation.

Signal transduction analysis in primary T cells after TCR andCCR7 costimulation

To explore the biochemical basis of CCL21-induced costimulation,we investigated signaling events activated downstream of bothTCR and chemokine receptors. The TCR on primary mouse T cellswas cross-linked for various times in the presence or absence ofCCL21 and lysates analyzed by quantitative Western blotting. Asshown in Figure 3A (left panel), simultaneous activation of TCRand CCR7 resulted in additive Rac-GTP formation up to 2 minutes(supplemental Table 1). Unexpectedly, early Ras-GTP formationwas mainly mediated by CCR7 at early time points, whereas at5 minutes, Ras-GTP was only observed in CCR7 and TCR-costimulated cells (2.2- , 0.7-fold increase over added values ofsingle stimulation; mean , SEM). This was paralleled by in-creased phosphorylation of MEK1/2 when both anti-CD3 andCCL21 were combined and by synergistic phosphorylation of itsdownstream target ERK1/2 (2.4- , 1.3-fold and 2- , 0.1-foldincrease over added values of single stimulation at 5 and 10 min-utes, respectively). Flow cytometric analysis uncovered that bothpercentage of pERK! cells as well as cellular pERK levels werestrongly increased when CCL21 and TCR signaling coincided(Figure 3B).

Conversely, the phosphorylation of JNK (Figure 3A), p38, andRaf (supplemental Table 1) did not increase during chemokine-mediated costimulation. We also analyzed PI3K activity by quanti-fying phosphorylation of its downstream target Akt. CCL21triggered robust Akt phosphorylation, which was less pronouncedin CD3-crosslinked T cells at the early time points measured here.Simultaneous TCR and CCR7 activation by CCL21 had an additiveeffect on Akt phosphorylation (Figure 3A). In summary, our datasuggest a rapid and pronounced activation of Rac, Ras, andMEK1/2 on simultaneous TCR and CCR7 triggering, resulting inenhanced ERK1/2 phosphorylation.

Figure 2. CCL21 costimulation is effective during early T-cell activa-tion. DO11.10 T cells were cocultured with OVA323-339-loaded congenicsplenocytes in the presence or absence of CCL21 (100 nM). (A) CCL21was added at indicated times, and proliferation of T cells was analyzedafter 48 hours. Statistical significance was determined using analysis ofvariance comparison of peptide-stimulated T-cell proliferation without orwith CCL21 added at indicated times. *P + .05 compared with “nochemokine.” (B) Up-regulation of early activation markers CD69 and CD25on CD4! KJ1-26! DO11.10 T cells 8 and 24 hours after Ag-specific T-cellactivation in the presence or absence of CCL21 (100 nM). One represen-tative experiment of 3 is shown. (C) IL-2 production of DO11.10 T cellsafter 24 hours as determined by intracellular staining. One representativeexperiment of 2 is shown. In panels B and C, numbers indicate percentageof positive cells and mean fluorescence intensity (MFI), respectively.

CHEMOKINE COSTIMULATION OF CD4! T CELLS 583BLOOD, 16 JULY 2009 ! VOLUME 114, NUMBER 3

PI3K activity is not required for CCL21-mediated costimulation

We wanted to examine whether PI3K activity had an effect onbaseline and CCL21-enhanced T-cell proliferation, focusing on thelymphocyte-enriched PI3K# and PI3K$ isoforms. To this end, wecompared the proliferation of control, PI3K$D910A/D910A, or PI3K#%/%

OT-II TCR-tg CD4! T cells in response to increasing concentra-tions of chicken OVA323-339. Similar to our previous results withDO11.10 T cells, OT-II TCR tg T cells proliferated more in thepresence of CCL21, in particular at low peptide concentrations(Figure 4). Similarly, PI3K$D910A/D910A and PI3K#-deficient OT-IITCR tg T cells OT-II cells showed comparable antigen-induced andCCL21-costimulated proliferation (Figure 4). Furthermore, pretreat-ment of human lymphocytes with the pan-PI3K inhibitor Wortman-nin had no effect on chemokine-induced enhancement of T-cellactivation as measured by Ca2! and CD69 responses (supplementalFigure 2A; and data not shown). Similar results were obtained withmurine T lymphocytes (data not shown). In summary, these datasuggest that PI3K activity downstream TCR or CKR is notinvolved in CCL21-costimulated proliferation, although we did notassess effector differentiation in these assays.

Decreased costimulatory effect of CCL21 in the absence ofDOCK2

T cells lacking the RacGEF DOCK2 show reduced Rac-GTPformation and ERK phosphorylation after TCR stimulation, whereasmost other signaling pathways remain intact.29 Given the strongsynergistic effect of simultaneous TCR and CCR7 stimulation onERK activation and the involvement of DOCK2 downstream both

receptors, we examined the costimulatory effect of CCL21 incontrol and DOCK2%/% 2B4 TCR-tg CD4! T cells activated byAg-pulsed splenocytes. Control 2B4 TCR tg cells responded withincreased proliferation to the presence of CCL21. In agreementwith previous results,29 proliferation of DOCK2%/% 2B4 TCR-tgT cells was strongly reduced at all peptide concentrations examined(Figure 5A). Importantly, the costimulatory effect of CCL21 wassignificantly reduced in the absence of DOCK2. This was particu-larly noticeable comparing the costimulatory effect of CCL21 atsimilar baseline proliferation (eg, 0.03 )g/mL MCC peptide incontrol cells vs 0.1 )g/mL MCC peptide in DOCK2%/% 2B4 TCRtg CD4! T cells). In addition, CCL21-stimulated CD69 and CD25expression was reduced in DOCK2-deficient cells (supplementalFigure 3A). Similarly, pharmacologic inhibition of the Rac effectorPak1 using a novel, highly specific compound40 strongly reducedCCL21-induced proliferation and CD69 expression (supplementalFigure 4; and data not shown).

Next, we determined whether the decreased costimulatoryeffect in DOCK2%/% T cells correlated with impaired intracellularsignaling. We performed anti-CD3( crosslinking experiments inthe presence and absence of CCL21. As reported,29 DOCK2%/%

2B4 TCR-tg T cells did not show any detectable Rac-GTP evenwhen TCR and CCR7 were simultaneously activated (Figure 5B).In contrast, both wild-type and DOCK2%/% T cells showed normalRas-GTP formation after activation of TCR and CCR7, or CCR7alone (Figure 5C). Simultaneous activation of TCR and CCR7 inboth control and DOCK2%/% 2B4 TCR-tg T cells synergisticallyincreased ERK phosphorylation (Figure 5C; supplemental Table 1).However, absolute ERK phosphorylation levels at 2 minutes were

Figure 3. Biochemical analysis of CCL21 costimulation. Primarymouse T cells were stimulated with anti-CD3 crosslinking, in the presenceor absence of CCL21 (100 nM), for indicated times, and the activation ofearly signaling molecules was analyzed by immunoblotting. (A) Immuno-blots of Rac2-GTP, Ras-GTP, and phosphorylated MEK1/2, ERK1/2, JNK,and Akt after costimulation of TCR and/or CCL21. For loading controls,blots were stripped and probed for total protein, or alternatively, a separategel with lysates was analyzed. (B) Flow cytometric analysis of phosphory-lated ERK1/2 formation in nonstimulated primary mouse T cells (dark grayfill) or after 5 minutes of stimulation with CCL21 (100 nM; dashed line),TCR-crosslinking (light gray fill), and TCR-crosslinking in the presence of100 nM CCL21 (bold line). One representative experiment of 2 is shown.

Figure 4. CCL21-mediated costimulation in the absence of PI3K#- orPI3K"-activity. OT-II TCR-tg CD4! T cells were cocultured with chickenOVA323-339-pulsed irradiated congenic splenocytes in the presence orabsence of CCL21 (100 nM). T-cell activation was determined as de-scribed in “Proliferation assays.” Proliferation of control, PI3K#-deficient,and PI3K$D910A/D910A OT-II TCR-tg CD4! T cells after 72 hours is shown.Numbers indicate fold increase of proliferation in the presence of CCL21.Data are pooled from 4 independent experiments. *P + .05 compared with“no chemokine.”

584 GOLLMER et al BLOOD, 16 JULY 2009 ! VOLUME 114, NUMBER 3

reduced in DOCK2%/% T cells by 72% plus or minus 5%(mean , SD), with a recovery 5 minutes after stimulation (Figure5C). CCL21 elicited similar Akt phosphorylation in both controland DOCK2%/% T cells, indicating normal PI3K function (supple-mental Figure 3B). Taken together, these data suggest that DOCK2/Rac contributes to efficient ERK activation downstream both TCRand CCR7 at early time points.

Optimal CD4! T-cell activation inside lymphoid tissue requiresGPCR signaling

Our in vitro results indicated that CCR7 ligands act as costimula-tory factors during CD4! T-cell activation. We next performedexperiments to study the influence of the chemokine-rich lymphoidmicroenvironment on T-cell responses, using PLN slices. One ofthe advantages of this system is the possibility to interfere acutelywith the molecular composition of the tissue. PLN slices (320 )mthickness), containing previously overlaid Marylin TCR tg CD4!

T cells, were perfused with the cognate DBY peptide during theimaging experiment. This experimental setting enabled us tomeasure the initiation of Ca2! responses of T cells after theirantigen encounter. Large concentrations of the antigenic peptide(100-1000 nM) triggered a strong increase in the intracellular Ca2!

concentration, which was associated with a reduction in the T-cellvelocity (supplemental Figure 5; supplemental Video 3). The

perfused peptide probably binds rapidly to MHC moleculesexpressed by resident DCs that form a dense network within theT-cell zone.41 One to 2 hours after the first Ca2! response, asignificant proportion of Marylin CD4! T cells showed increasedCD69 levels (Figure 6, supplemental Figure 4).

Next, we investigated whether GPCR ligands are involved inT-cell activation leading to increased CD69 surface levels. MarylinTCR-tg CD4! T cells pretreated or not with PTX for 15 minuteswere overlaid on PLN slices. Our previous experiments revealedthat the blocking effect of PTX is only complete after 2 hours,allowing efficient migration of both control and inhibitor-treatedlymphocytes into the slice. After 2 hours of PTX treatment,interstitial T-cell motility was significantly impaired compared withthat of cells treated with the B subunit (suB) of the toxin that doesnot possess a catalytic activity and was used as a control.16

Addition of 100 nM specific peptide to slices induced rapidinduction of Ca2! flux in both control and PTX-treated populations(supplemental Video 3). On average, the mean delay between theperfusion of the peptide and the initiation of the Ca2! response was12.3 plus or minus 4.1 minutes for suB-treated cells and 12.0 plusor minus 1.9 minutes (n - 4) for PTX-treated cells. Moreover, theamplitude of Ca2!-flux was not affected by PTX, indicating thatearly TCR signaling is intact in PTX-treated lymphocytes, althoughthe percentage of responding cells was significantly decreased

Figure 5. CCL21-mediated costimulation and signaling pathways inDOCK2-deficient T cells. (A) 2B4 TCR-tg CD4! T cells were coculturedwith MCC88-103-pulsed irradiated splenocytes in the presence or absenceof CCL21 (100 nM). T-cell activation was determined as described in“Proliferation assays.” Proliferation of TCR-tg DOCK2!/% and DOCK2%/%

2B4 TCR-tg T cells after 48 hours is shown. Numbers indicate foldincrease of proliferation in the presence of CCL21. Data are pooled from3 independent experiments. *P + .05 compared with “no chemokine.”(B) Control and DOCK2%/% 2B4 TCR-tg T cells were stimulated as inFigure 3 and analyzed for Rac-GTP formation. One representativeexperiment of 2 is shown. (C) Control and DOCK2%/% 2B4 TCR-tg T cellswere stimulated as in Figure 3 and analyzed for phosphorylation ofERK1/2 and Ras-GTP formation. One representative experiment of3 is shown.

Figure 6. Antigen-induced T-cell responses measured in PLN slicesare decreased by PTX. (A) Average Ca2! responses of Marylin TCR-tgCD4! T cells incubated with suB or PTX and stimulated with DBY peptide.Fura-2-loaded T cells were incubated for 10 minutes with 100 ng/mL suB(black line) or PTX (gray line), washed, and overlaid on PLN slices.Time-lapse imaging was started 2 hours after suB or PTX treatment. Afterseveral minutes of imaging, the preparation was perfused with a solutioncontaining 100 nM DBY peptide. Ca2!-signals of responding T cells weresynchronized at the rising phases of the response. (B) Percentage ofCa2!-responding CD4! T cells induced by DBY peptide (100 nM). Meanplus or minus SD of 3 independent experiments in which more than50 cells were analyzed per experiment. *P + .05. (C) CD69 expressionmeasured by flow cytometry on T cells activated within PLN slices.Marylin CD4! TCR-tg T cells were incubated with 100 ng/mL of suB orPTX, labeled with CMFDA and overlaid on PLN slices. Two hours afterDBY peptide treatment, slices were mechanically dissociated and recov-ered cells stained with anti–CD69-PE and anti–CD45.1-PE-Cy5 Abs.Results in the left panel are representative of 3 independent experimentsand show percentage and MFI of CD69! cells. The right panel showscombined MFI of CD69 expression after normalization to the value ofsuB-treated T cells. *P + .05.

CHEMOKINE COSTIMULATION OF CD4! T CELLS 585BLOOD, 16 JULY 2009 ! VOLUME 114, NUMBER 3

(Figure 6A-B). Accordingly, the percentage of CD69! cells afterpeptide perfusion was reduced in PTX-treated CD4! T cellscompared with suB-treated lymphocytes. Notably, the mean expres-sion levels of CD69 in activated T cells were reduced by approxi-mately 60% after PTX treatment at 10 and 100 nM DBY peptide(Figure 6C). These data support the notion that GPCR ligandscontribute to efficient CD4! T-cell activation inside the lymphoidmicroenvironment containing chemokines and other promigra-tory factors.

Discussion

The aim of this study was to examine whether CCL21 and otherhomeostatic chemokines present in the lymph node paracortexhave a direct effect on TCR-induced intracellular signaling leadingto T-cell activation. Our data suggest that the presence of CCL21results in higher T-cell proliferation, with the costimulatory actionof CCL21 being more pronounced at suboptimal activation. Weprovide evidence that the presence of CCL21 during the earlystages of CD4! T-cell activation leads to a selective and synergisticincrease in ERK but not JNK or p38 phosphorylation, concomitantwith increased expression of the early activation markers CD69and CD25. Furthermore, we provide evidence that CCL21-triggered costimulation correlates with increased and prolongedRas- and Rac-GTP levels. The latter was mediated by the RacGEFDOCK2 activated downstream of CCR7 and TCR, whereas PI3Kactivity was not required for costimulation in our system. Finally,observation of T-cell activation in PLN slices supports a costimula-tory function of GPCR ligands in the paracortex.

Chemokines have previously been shown to participate in T-cellactivation. For example, CCL5 induces recruitment of its ligandCCR5 and coupling with Gq/11 proteins at the interface betweenAPC and T cells, enhancing conjugate stability and proliferation.21

This mechanism probably does not underlie the CCL21-stimulatedcostimulation described here, as the increased proliferation weobserved was pertussis toxin sensitive, and hence G*i-dependent,and because CCR7 is not thought to accumulate at the immuno-logic synapse.21 Furthermore, CCL21 also increased Ab-elicitedproliferation, in the absence of adhesive ligands. It is nonethelessconceivable that CCR7-triggered integrin avidity contributes tomore stable T-cell–DC interactions and therefore influences theoutcome of an immune response.42 In addition, T-cell adhesionleads to potentiation of TCR signaling through ERK activation.43

The participation of adhesion in chemokine-induced T-cell costimu-lation deserves further investigations.

Similar to the Ras-GTP formation downstream of CCR7reported here, CXCL12- and PI3K-dependent signals trigger aphysical interaction between CXCR4 and the TCR, which pro-motes Ras-GTP formation and ERK phosphorylation.44-46 The factthat CCR7 and TCR do not colocalize at the T-cell–APC interface21

and the refractiveness of CCL21 costimulation to PI3K inhibitionargue against a physical interaction of CCR7 and TCR. Althoughthe precise mechanisms by which CCR7- and TCR-derived signalsare integrated for synergistic Ras activation remain unknown, weidentified a role for the RacGEF DOCK2 in integrating TCR- andCCR7-triggered Rac activation. DOCK2-mediated Rac-GTP forma-tion correlated with increased and prolonged ERK phosphoryla-tion, possibly via the Rac-effector Pak, which can phosphorylateRaf and MEK.47 Further support for a downstream role for Pak1 isprovided by pharmacologic inhibition, which largely phenocopiesthe DOCK2 deficiency. Although we did not detect increased Rafphosphorylation after costimulation with CCL21, levels of phos-

phorylated MEK1/2 were increased in TCR- and CCR7-stimulatedCD4! T cells. Thus, both Ras- and Rac-GTP are efficiently formedafter costimulation with homeostatic chemokines and, via activa-tion of MEK1/2, mediate ERK phosphorylation (Figure 7).A digital all-or-nothing, highly amplified ERK phosphorylation is acentral feature of T-cell activation in vitro48,49 and may help explainthe narrow peptide concentration range between no T-cell prolifera-tion and a full proliferative response observed in vivo.8

Although DOCK2 deficiency leads to delayed ERK phosphory-lation downstream of TCR and CCR7, we still observed asignificant chemokine-dependent costimulatory effect on ERKphosphorylation in these cells, potentially mediated by Ras-GTP. Incontrast, the proliferation of DOCK2-deficient T cells was stronglyimpaired. These observations suggest additional roles for DOCK2and Rac during TCR signal transduction, including TCR and lipidraft polarization.29,50 Lymphocytes also express the RacGEF Vav1,which acts downstream both TCR and chemokine receptors.51,52

Vav1-deficiency leads to decreased Rac and ERK activation,51,53-55

among other defects. As most Rac-GTP formation downstream theTCR is mediated by DOCK2,29 defective activation of Rac inVav-1–deficient T cells may be linked because of its adapterfunction, rather than its GEF activity. Alternatively, Vav-inducedRac activation may depend on preceding DOCK2-mediated F-actinformation.

Deficiency in Rac2 inhibits Th1 differentiation,56 and DOCK2%/%

CD4! T cells are skewed toward a Th2 phenotype because ofinefficient down-regulation of the IL-4R* chain.57 Conversely,exposure to CCL21 increases Th1 differentiation in vitro throughincreased IFN-# production,22 and CCR7 ligands induce IL-12production in DCs.58 Taken together, these observations indicatethat the CCL21-rich paracortex favors Th1 differentiation throughDOCK2-Rac activity. During infections, expression of homeostaticchemokines is strongly decreased,59,60 implying a change not onlyin streptolysin O architecture but also in lymphocyte differentiationpathways. Secondary challenges in mice with low lymphoidchemokine levels may thus show a preferential Th2 differentiation,although other cytokines are probably involved in decision-making.

We were unable to detect any defects in activation of PI3K#-deficient T cells or an impairment of CCL21-mediated costimula-tion in our experimental system. The discrepancy to other pub-lished reports32,33 is unclear at the moment but could be the result ofdifferent transgenic models. Although PI3K$ activity is requiredfor full ERK phosphorylation downstream TCR signaling,61,62

expression of inactive PI3K$ did not reduce CCL21-mediatedcostimulation in the experimental model used here. PI3K activitythus appears dispensable for CCL21-mediated costimulation invitro. It will be important to investigate the function of PI3K duringT-cell activation under more physiologic conditions, in particularfor the generation of cytokine-producing effector T cells. This is

Figure 7. Proposed model for integration of TCR and CCR7-triggered signals.During an early promigratory phase of an ongoing antigenic response, CD4! T cellsundergo sequential encounters with pMHC-presenting DCs (left panel). During thisperiod, T cells are exposed to TCR- and CCR7 (in addition to other GPCR)-derivedsignals, both of which activate PI3K, DOCK2-Rac, and Ras. Active Rac- andRas-GTP contribute to enhanced MEK-ERK activation leading to up-regulation ofearly activation markers, such as CD69, and rapid production of IL-2 (right panel).

586 GOLLMER et al BLOOD, 16 JULY 2009 ! VOLUME 114, NUMBER 3

especially relevant as directional migration of PI3K#-deficientT cells is affected in the paracortex,28 and PI3K $D910A/ D910A T cellsshow defective raft recruitment when stimulated with surrogateAPCs in vitro.62

As CCL21 contributes to some extent to the exploratorybehavior of T cells in lymphoid tissue,13-16 it is experimentallydifficult to dissect a costimulatory effect of chemokines from theirindirect facilitation of efficient scanning of rare DCs. Besides adirect signaling effect, the promigratory functions of CCR7 ligandsand other G*i-dependent factors probably influence T-cell re-sponses by increasing the chances for T cells to encounter andinteract with APCs. Using the PLN slice system, we were able tosynchronize T-cell activation in a physiologic environment whilefollowing motility, Ca2! responses, and CD69 up-regulation asfunctional readout. Although CCL21 is abundant inside lymphoidtissue, other GPCR ligands contribute to DOCK2-dependentinterstitial T-cell migration13,15,28 and may thus add to T-cellactivation in vivo. We therefore used the general G*i inhibitorPTX, which reduced the percentage of T cells showing Ca2! fluxand CD69 up-regulation. This observation may reflect the de-creased cellular motility and ability to engage in productiveencounters with DCs, although the responding PTX-treated T cellsincrease their Ca2! with a delay similar to that of control T cells,indicating that DCs are forming a tight network inside lymphoidtissue. Notably, PTX-treated T cells showed normal Ca2!-fluxresponses, indicating that G*i signaling is not generally requiredfor T-cell activation. Nonetheless, PTX exerted an inhibitory effecton the level of CD69 expressed by antigen-stimulated T cells at theearly time points measured here, suggesting that GPCR ligandscontribute to optimal T-cell activation within lymphoid tissue.Because CD69 levels continuously increase over the first 24 to48 hours of an immune response, it is possible that PTX-treatedcells eventually reach similar activation levels as untreated cells.

Our data support a scenario where, within lymphoid tissue,migrating CD4! T cells measure and integrate TCR- and GPCR-derived signals, resulting in more efficient CD69 up-regulation andIL-2 production. The physiologic situation is probably morecomplex, as CCL21 and other promigratory factors may influenceT-cell activation both in positive and negative manners. It has beensuggested that promigratory and antimigratory signals mediated byCCR7 and TCR, respectively, compete with each other, onlyallowing efficient T-cell activation to take place when the TCR-triggered stop signal overrides promigratory signals from chemo-kine gradients.23 How may these seemingly contradictory proposi-tions on the role of CCL21 be reconciled? One potential answermay lie in the ability of T cells to integrate activation signals overtime. At low doses of antigen, T cells do not immediately form tightstable contacts with DCs but undergo brief serial interaction withAPCs for several hours.8 On repeated encounters with low amountsof pMHC complexes, intracellular promigratory signaling mol-

ecules may contribute to reach a lower activation threshold, thusactively participating in T-cell activation. In case T cells do notrepeatedly encounter peptide-MHC complexes with a thresholdfrequency, no signal integration takes place. Homeostatic chemo-kines may thus serve as tissue “rheostat” for recirculating lympho-cytes, informing them on their presence inside lymphoid tissue byincreasing their sensitivity to repetitive TCR-derived signals. Thismay also help prevent unwanted lymphocyte activation in thenonlymphoid tissue to which naive T cells occasionally migrate.63

Simultaneously, promigratory signals may act to “silence noise” byavoiding T-cell adhesion at individual DCs displaying low affinityantigenic peptide. At later time points, full effector T-cell differen-tiation requires prolonged interactions with DCs, in which celldisplacement is suppressed by continuous TCR signaling.23

In conclusion, our data support a role for homeostatic chemo-kines, in particular CCL21, during CD4! T-cell costimulation invitro and a role for GPCRs for T-cell activation in situ. Wehypothesize that CCL21 and other GPCR ligands lower thethreshold for T-cell activation in the early promigratory phase ofT-cell activation by sustaining increased ERK phosphorylationlevels downstream Rac and Ras.

Acknowledgments

The authors thank Prof Britta Engelhardt (University of Bern) andDr Alain Trautmann (CNRS and INSERM, Paris) for continuoussupport, Drs Emilio Hirsch (University of Turin) and MatthiasWymann (University of Basel) for providing PI3K#-deficient mice,and Drs Marcus and Silvia Thelen (IRB, Bellinzona) and MichaelHuber (MPI, Freiburg) who generously shared reagents andprotocols.

This work was supported by the Swiss National Foundation(grants SNF3100A0-107510 and EU-MEXT 25405; J.V.S.). J.R.P.was supported by the Department of Defense Research Program(W81XWH-05-1-0200) and the National Institutes of Health(GM083025).

Authorship

Contribution: K.G., F.A.-B., and Y.T. performed experiments andanalyzed results; K.O., B.V., J.R.P., and Y.F. provided geneticallymodified mouse strains and synthesized compounds; and K.G.,E.D., and J.V.S. designed the research and wrote the paper.

Conflict-of-interest disclosure: B.V. is an advisor to Intellikine(San Diego, CA). The remaining authors declare no competingfinancial interests.

Correspondence: Jens V. Stein, University of Bern, TheodorKocher Institute, Freiestrasse 1, 3012 Bern, Switzerland; e-mail:jstein@tki.unibe.ch.

References1. von Andrian UH, Mempel TR. Homing and cellu-

lar traffic in lymph nodes. Nat Rev Immunol.2003;3:867-878.

2. Lanzavecchia A, Sallusto F. Dynamics of T lym-phocyte responses: intermediates, effectors, andmemory cells. Science. 2000;290:92-97.

3. Stoll S, Delon J, Brotz TM, Germain RN. Dynamicimaging of T cell-dendritic cell interactions inlymph nodes. Science. 2002;296:1873-1876.

4. Mempel TR, Henrickson SE, Von Andrian UH. T-cellpriming by dendritic cells in lymph nodes occurs inthree distinct phases. Nature. 2004;427:154-159.

5. Miller MJ, Wei SH, Parker I, Cahalan MD. Two-

photon imaging of lymphocyte motility and anti-gen response in intact lymph node. Science.2002;296:1869-1873.

6. Henrickson SE, von Andrian UH. Single-cell dy-namics of T-cell priming. Curr Opin Immunol.2007;19:249-258.

7. Skokos D, Shakhar G, Varma R, et al. Peptide-MHC potency governs dynamic interactions be-tween T cells and dendritic cells in lymph nodes.Nat Immunol. 2007;8:835-844.

8. Henrickson SE, Mempel TR, Mazo IB, et al. T cellsensing of antigen dose governs interactive be-

havior with dendritic cells and sets a threshold forT-cell activation. Nat Immunol. 2008;9:282-291.

9. Link A, Vogt TK, Favre S, et al. Fibroblastic reticu-lar cells in lymph nodes regulate the homeostasisof naive T cells. Nat Immunol. 2007;8:1255-1265.

10. Lammermann T, Sixt M. The microanatomy ofT-cell responses. Immunol Rev. 2008;221:26-43.

11. Kaiser A, Donnadieu E, Abastado JP, TrautmannA, Nardin A. CC chemokine ligand 19 secreted bymature dendritic cells increases naive T cell scan-ning behavior and their response to rare cognateantigen. J Immunol. 2005;175:2349-2356.

CHEMOKINE COSTIMULATION OF CD4! T CELLS 587BLOOD, 16 JULY 2009 ! VOLUME 114, NUMBER 3

12. Stachowiak AN, Wang Y, Huang YC, Irvine DJ. Ho-meostatic lymphoid chemokines synergize with ad-hesion ligands to trigger T and B lymphocyte chemo-kinesis. J Immunol. 2006;177:2340-2348.

13. Worbs T, Mempel TR, Bolter J, von Andrian UH,Forster R. CCR7 ligands stimulate the intranodalmotility of T lymphocytes in vivo. J Exp Med.2007;204:489-495.

14. Woolf E, Grigorova I, Sagiv A, et al. Lymph nodechemokines promote sustained T lymphocytemotility without triggering stable integrin adhe-siveness in the absence of shear forces. Nat Im-munol. 2007;8:1076-1085.

15. Okada T, Cyster JG. CC chemokine receptor 7contributes to Gi-dependent T cell motility in thelymph node. J Immunol. 2007;178:2973-2978.

16. Asperti-Boursin F, Real E, Bismuth G, TrautmannA, Donnadieu E. CCR7 ligands control basal Tcell motility within lymph node slices in a phos-phoinositide 3-kinase-independent manner. J ExpMed. 2007;204:1167-1179.

17. Friedman RS, Jacobelli J, Krummel MF. Surface-bound chemokines capture and prime T cells for syn-apse formation. Nat Immunol. 2006;7:1101-1108.

18. Taub DD, Turcovski-Corrales SM, Key ML, Longo DL,Murphy WJ. Chemokines and T lymphocyte activa-tion: I. Beta chemokines costimulate human T lympho-cyte activation in vitro. J Immunol. 1996;156:2095-2103.

19. Nanki T, Lipsky PE. Cutting edge: stromal cell-derived factor-1 is a costimulator for CD4! T-cellactivation. J Immunol. 2000;164:5010-5014.

20. Trautmann A. Chemokines as immunotransmit-ters? Nat Immunol. 2005;6:427-428.

21. Molon B, Gri G, Bettella M, et al. T cell costimula-tion by chemokine receptors. Nat Immunol. 2005;6:465-471.

22. Flanagan K, Moroziewicz D, Kwak H, Horig H,Kaufman HL. The lymphoid chemokine CCL21costimulates naive T cell expansion and Th1 po-larization of non-regulatory CD4! T cells. CellImmunol. 2004;231:75-84.

23. Dustin ML. Stop and go traffic to tune T cell re-sponses. Immunity. 2004;21:305-314.

24. Friedman RS, Jacobelli J, Krummel MF. Mecha-nisms of T cell motility and arrest: deciphering therelationship between intra- and extracellular de-terminants. Semin Immunol. 2005;17:387-399.

25. Etienne-Manneville S, Hall A. Rho GTPases incell biology. Nature. 2002;420:629-635.

26. Fukui Y, Hashimoto O, Sanui T, et al. Haemato-poietic cell-specific CDM family protein DOCK2 isessential for lymphocyte migration. Nature. 2001;412:826-831.

27. Shulman Z, Pasvolsky R, Woolf E, et al. DOCK2regulates chemokine-triggered lateral lymphocytemotility but not transendothelial migration. Blood.2006;108:2150-2158.

28. Nombela-Arrieta C, Mempel TR, Soriano SF, etal. A central role for DOCK2 during interstitiallymphocyte motility and sphingosine-1-phosphate-mediated egress. J Exp Med. 2007;204:497-510.

29. Sanui T, Inayoshi A, Noda M, et al. DOCK2 is es-sential for antigen-induced translocation of TCRand lipid rafts, but not PKC-theta and LFA-1, inT cells. Immunity. 2003;19:119-129.

30. Nombela-Arrieta C, Lacalle RA, Montoya MC, et al.

Differential requirements for DOCK2 andphosphoinositide-3-kinase gamma during T and Blymphocyte homing. Immunity. 2004;21:429-441.

31. Martin AL, Schwartz MD, Jameson SC, ShimizuY. Selective regulation of CD8 effector T cell mi-gration by the p110# isoform of phosphatidylinosi-tol 3-kinase. J Immunol. 2008;180:2081-2088.

32. Alcazar I, Marques M, Kumar A, et al. Phospho-inositide 3-kinase gamma participates in T cellreceptor-induced T-cell activation. J Exp Med.2007;204:2977-2987.

33. Sasaki T, Irie-Sasaki J, Jones RG, et al. Functionof PI3Kgamma in thymocyte development, T-cellactivation, and neutrophil migration. Science.2000;287:1040-1046.

34. Okkenhaug K, Vanhaesebroeck B. PI3K in lym-phocyte development, differentiation and activa-tion. Nat Rev Immunol. 2003;3:317-330.

35. Okkenhaug K, Bilancio A, Emery JL,Vanhaesebroeck B. Phosphoinositide 3-kinase inT-cell activation and survival. Biochem Soc Trans.2004;32:332-335.

36. Reif K, Okkenhaug K, Sasaki T, Penninger JM,Vanhaesebroeck B, Cyster JG. Cutting edge: differ-ential roles for phosphoinositide 3-kinases,p110gamma and p110delta, in lymphocyte chemo-taxis and homing. J Immunol. 2004;173:2236-2240.

37. Garcon F, Patton DT, Emery JL, et al. CD28 pro-vides T-cell costimulation and enhances PI3Kactivity at the immune synapse independently ofits capacity to interact with the p85/p110 het-erodimer. Blood. 2008;111:1464-1471.

38. Luther SA, Bidgol A, Hargreaves DC, et al. Differ-ing activities of homeostatic chemokines CCL19,CCL21, and CXCL12 in lymphocyte and dendriticcell recruitment and lymphoid neogenesis. J Im-munol. 2002;169:424-433.

39. Mempel TR, Scimone ML, Mora JR, von AndrianUH. In vivo imaging of leukocyte trafficking inblood vessels and tissues. Curr Opin Immunol.2004;16:406-417.

40. Deacon SW, Beeser A, Fukui JA, et al. Anisoform-selective, small-molecule inhibitor targetsthe autoregulatory mechanism of p21-activatedkinase. Chem Biol. 2008;15:322-331.

41. Lindquist RL, Shakhar G, Dudziak D, et al. Visu-alizing dendritic cell networks in vivo. Nat Immu-nol. 2004;5:1243-1250.

42. Scholer A, Hugues S, Boissonnas A, Fetler L,Amigorena S. Intercellular adhesion molecule-1-dependent stable interactions between T cellsand dendritic cells determine CD8! T cellmemory. Immunity. 2008;28:258-270.

43. Conche C, Boulla G, Trautmann A,Randriamampita C. T cell adhesion primes anti-gen receptor-induced calcium responses througha transient rise in adenosine 3.,5.-cyclic mono-phosphate. Immunity. 2009;30:33-43.

44. Weber KS, Ostermann G, Zernecke A, SchroderA, Klickstein LB, Weber C. Dual role of H-Ras inregulation of lymphocyte function antigen-1 activ-ity by stromal cell-derived factor-1alpha: implica-tions for leukocyte transmigration. Mol Biol Cell.2001;12:3074-3086.

45. Kumar A, Humphreys TD, Kremer KN, et al.CXCR4 physically associates with the T cell re-ceptor to signal in T cells. Immunity. 2006;25:213-224.

46. Patrussi L, Ulivieri C, Lucherini OM, et al. p52Shc is

required for CXCR4-dependent signaling and che-motaxis in T cells. Blood. 2007;110:1730-1738.

47. Bagrodia S, Cerione RA. Pak to the future.Trends Cell Biol. 1999;9:350-355.

48. Stefanova I, Hemmer B, Vergelli M, Martin R,Biddison WE, Germain RN. TCR ligand discrimi-nation is enforced by competing ERK positiveand SHP-1 negative feedback pathways. Nat Im-munol. 2003;4:248-254.

49. Altan-Bonnet G, Germain RN. Modeling T cellantigen discrimination based on feedback controlof digital ERK responses. PLoS Biol. 2005;3:e356.

50. Yu H, Leitenberg D, Li B, Flavell RA. Deficiencyof small GTPase Rac2 affects T-cell activation.J Exp Med. 2001;194:915-926.

51. Fischer KD, Kong YY, Nishina H, et al. Vav is aregulator of cytoskeletal reorganization mediatedby the T-cell receptor. Curr Biol. 1998;8:554-562.

52. Takesono A, Horai R, Mandai M, Dombroski D,Schwartzberg PL. Requirement for Tec kinases inchemokine-induced migration and activation ofCdc42 and Rac. Curr Biol. 2004;14:917-922.

53. Holsinger LJ, Graef IA, Swat W, et al. Defects inactin-cap formation in Vav-deficient mice impli-cate an actin requirement for lymphocyte signaltransduction. Curr Biol. 1998;8:563-572.

54. Costello PS, Walters AE, Mee PJ, et al. The Rho-family GTP exchange factor Vav is a critical trans-ducer of T cell receptor signals to the calcium,ERK, and NF-kappaB pathways. Proc Natl AcadSci U S A. 1999;96:3035-3040.

55. Reynolds LF, Smyth LA, Norton T, et al. Vav1transduces T cell receptor signals to the activa-tion of phospholipase C-gamma1 via phospho-inositide 3-kinase-dependent and -independentpathways. J Exp Med. 2002;195:1103-1114.

56. Li B, Yu H, Zheng W, et al. Role of the guanosinetriphosphatase Rac2 in T helper 1 cell differentia-tion. Science. 2000;288:2219-2222.

57. Tanaka Y, Hamano S, Gotoh K, et al. T helpertype 2 differentiation and intracellular trafficking ofthe interleukin 4 receptor-alpha subunit controlledby the Rac activator Dock2. Nat Immunol. 2007;8:1067-1075.

58. Marsland BJ, Battig P, Bauer M, et al. CCL19 andCCL21 induce a potent proinflammatory differen-tiation program in licensed dendritic cells. Immu-nity. 2005;22:493-505.

59. Mueller SN, Hosiawa-Meagher KA, KoniecznyBT, et al. Regulation of homeostatic chemokineexpression and cell trafficking during immune re-sponses. Science. 2007;317:670-674.

60. Scandella E, Bolinger B, Lattmann E, et al. Restora-tion of lymphoid organ integrity through the interac-tion of lymphoid tissue-inducer cells with stroma ofthe T cell zone. Nat Immunol. 2008;9:667-675.

61. Matheu MP, Deane JA, Parker I, Fruman DA,Cahalan MD. Class IA phosphoinositide 3-kinasemodulates basal lymphocyte motility in the lymphnode. J Immunol. 2007;179:2261-2269.

62. Okkenhaug K, Bilancio A, Farjot G, et al. ImpairedB and T cell antigen receptor signaling inp110delta PI 3-kinase mutant mice. Science.2002;297:1031-1034.

63. Staton TL, Habtezion A, Winslow MM, Sato T, LovePE, Butcher EC. CD8! recent thymic emigrantshome to and efficiently repopulate the small intestineepithelium. Nat Immunol. 2006;7:482-488.

588 GOLLMER et al BLOOD, 16 JULY 2009 ! VOLUME 114, NUMBER 3

Supp  Data  Gollmer  et  al.  

Table  1  

Table  S1.  Quan-fica-on  of  western  blot  analysis  in  primary  T  cells  The  signal  average  of  the  “0”  <me  point  was  arbitrarily  put  to  “1,”  and  the  signal  intensi<es  at  various  <me  points,   and   treatments   was   adjusted   accordingly.   Signals   in   cos<mulated   samples   (CCL21   +   TCR)  represen<ng  more  than  double  the  sum  of  single-­‐s<mulated  (CCL21  or  TCR)  are  shown  in  bold.  All  values  are  mean   ±   SEM.   The   occasionally   high   SEM  was   due   to   varia<ons   in   absolute   fold   induc<on   between  individual  experiments,  while  the  paPerns  of  signal  molecule  ac<va<on  as  shown  in  Figs.  3  and  5  remained  consistent  between  experiments.  To  normalize  for  this  variability,  we  also  calculated  the  “fold  increase”  of  cos<mulated  over  single-­‐s<mulated  cells  as   shown   in   the   lower  panel  of   the   table   for  Rac-­‐  and  Ras-­‐GTP  forma<on  and  ERK  phosphoryla<on.  

Supp  1  

Figure  S1  (A)   Soluble   CCL21   does   not   increases   the   migra<on   of   T   cells   in   contact   with   splenocytes.  CMFDA-­‐loaded  T  cells  were  cocultured  with  splenocytes  and  imaged  during  10  min.  Cells  were  treated  or  not  with  1  µg/ml  CCL21  added  5  min  before  imaging.  T-­‐cell  displacements  measured  in   PLN   slices   are   also   presented   for   comparison.   The   red   line   indicates   average   velocity.   (B)  CCL21-­‐induced   elonga<on   of   CMFDA-­‐labeled   T   cells.   The   le[   panel   shows   a   representa<ve  micrograph  of  T  cells  prior  to  adding  CCL21  (top)  and  5  min  a[er  CCL21  addi<on  (boPom).  The  right   panel   shows   the   shape   index   calculated   using   the   Morphometric   Analysis   Func<on   of  Metamorph.  

Supp  2  

Figure  S2    (A)   CCL19  mediates   PI3K-­‐independent   cos<mula<on   of   human   T   cells.   Fura-­‐2   loaded   human   T  cells  preincubated  with  100  nM  WMN  for  30  min  or  medium  alone  were  s<mulated  at  the  arrow  with  5  µg/ml  an<-­‐CD3  in  the  presence  or  absence  of  0.5  µg/ml  CCL19.  Ca2+  signal  was  measured  on  cell  popula<on  with  a  spectrofluorimeter.   (B)  Percentage  of  CD69+  human  T  cells  s<mulated  with  5  µg/ml  an<-­‐CD3  in  the  presence  or  absence  of  0.5  µg/ml  CCL19  or  CCL21.  CD69  expression  was  measured  by  flow  cytometry  12  h  later.  The  red  number  indicates  fold  increase  in  presence  of  chemokine.  Data  are  representa<ve  of  two  to  three  experiments.  

Supp  3  

 Figure  S3    (A)  Expression  of  CD69  and  CD25   in  absence  of  DOCK2.  2B4  control  or  DOCK2-­‐deficient  2B4  cells   were   s<mulated   using   irradiated   splenocytes   loaded   with   indicated   concentra<ons   of  MCC  pep<de,   in  presence   (black  bars)  or  absence   (white  bars)  of  100  nM  CCL21.  CD69  and  CD25   surface   expression   was   measured   8   and   16   h   later,   respec<vely.   (B)   Normal   Akt  phosphoryla<on   in   absence   of   DOCK2.   2B4   control   and   DOCK2-­‐deficient   T   cells   were  s<mulated  and  analyzed  for  phosphoryla<on  of  Akt  as  in  Figs.  3  and  5.  One  representa<ve  blot  of  five  is  shown.  

Supp  4  

Figure  S4.  Reduced  CCL21-­‐mediated  cos-mula-on  in  Pak1-­‐inhibited  CD4+  T  cells    OT-­‐II   TCR   tg   CD4+   T   cells   were   s<mulated   using   irradiated   splenocytes   loaded   with   indicated  amounts  of  OVA  pep<de,  in  presence  of  increasing  amounts  of  the  Pak1  inhibitor  IPA-­‐3.  A[er  72  h,  prolifera<on   was   measured   using   3H   thymidine   incorpora<on.   The   numbers   indicate   the   fold  increase  over  control  (“ctrl”  =  no  chemokine).  Shown  are  quadruple  values  from  one  of  three  similar  experiments.  

Supp  5  

Figure  S5.  An-gen-­‐induced  T-­‐cell  responses  measured  in  PLN  slices  (A)  Ca2+  responses  (black)  and  mo<lity  (red)  in  one  representa<ve  Marylin  TCR-­‐tg  CD4+  T  cell  within  a  PLN  slice.  T  cells  were  loaded  with  fura-­‐2  and  added  to  a  PLN  slice  30  min  before   the   recording.   A[er   several  minutes   of   imaging,   the   prepara<on  was   perfused  with  a  solu<on  containing  1  µM  DBY  pep<de.  The  doPed  line  represents  the  curve-­‐fiPed  velocity.  A  <me-­‐lapse  anima<on  of  Ca2+  responses  is  shown  in  Video  3.  (B)  Percentage  of  CD69+  cells  as  a  func<on  of  the  DBY  pep<de  concentra<on.  CMFDA-­‐labeled  Marylin  TCR-­‐tg  T   cells  were  overlaid  on  PLN   slices.   Two  h  a[er  DBY  pep<de   s<mula<on,  CD69  expression  was  determined  on  CMFDA+  CD45.1+  cells.  (C)  Kine<cs  of  CD69  expression  in  PLN   slices   treated   with   100   nM   DBY   pep<de.   Results   are   representa<ve   of   three  experiments.