Movement Within Lymph Nodes T Lymphocytes

Impaired Trafficking of Gnai2+/− and Gnai2− /− T Lymphocytes: Implications for T Cell Movement within Lymph Nodes

This information is current as Il-Young Hwang, Chung Park and John H. Kehrl of September 23, 2021. J Immunol 2007; 179:439-448; ; doi: 10.4049/jimmunol.179.1.439 http://www.jimmunol.org/content/179/1/439 Downloaded from

Supplementary http://www.jimmunol.org/content/suppl/2008/03/14/179.1.439.DC1 Material References This article cites 35 articles, 18 of which you can access for free at: http://www.jimmunol.org/content/179/1/439.full#ref-list-1 http://www.jimmunol.org/

Why The JI? Submit online.

• Rapid Reviews! 30 days* from submission to initial decision

• No Triage! Every submission reviewed by practicing scientists

by guest on September 23, 2021 • 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 © 2007 by The American Association of Immunologists All rights reserved. Print ISSN: 0022-1767 Online ISSN: 1550-6606. The Journal of Immunology

:Impaired Trafﬁcking of Gnai2؉/؊ and Gnai2؊/؊ T Lymphocytes Implications for T Cell Movement within Lymph Nodes

Il-Young Hwang, Chung Park, and John H. Kehrl1

Signals generated by the engagement of chemoattractants with their cognate receptors orchestrate lymphocyte movements into and out of lymphoid organs and sites of inflammation. Yet, the role of chemokines in organizing lymphocyte movements in lymphoid organs is controversial. Recent evidence suggests that the extensive network of fibroblastic reticular cells within the T cell areas helps guide T cells. The expression of adhesion molecules and chemokines by fibroblastic reticular cells most likely facilitates their influence on T cell movements. Consistent with this hypothesis, CD4 T cells with defective chemokine receptor signaling move very differently within lymph nodes than do normal cells. For the imaging studies, we used CD4 T ؊/؊ cells prepared from Gnai2 mice, which lack G␣i2 expression. We first demonstrate that CD4 as well as CD8 T cells from ؊/؊ these mice are markedly defective in chemokine receptor signaling. Gnai2 T cells have profound defects in chemokine- Downloaded from .induced intracellular calcium mobilization, chemotaxis, and homing, whereas Gnai2؉/؊ T cells exhibit modest defects Intravital imaging revealed that within the inguinal lymph nodes Gnai2؊/؊ CD4 T accumulate at the cortical ridge, poorly accessing the lymph node paracortex. They also lack the customary amoeboid-like cell movements and active membrane projections observed with normal CD4 T cells. These results demonstrate the importance of G␣i2 for T lymphocyte chemokine receptor signaling and argue that local chemoattractants regulate the movement of CD4 T cells in lymph nodes. The

Journal of Immunology, 2007, 179: 439–448. http://www.jimmunol.org/

roper functioning of the immune system depends upon thelial venules (HEVs), and decreased B cell motility within lymph changing intracellular contacts and cellular localization node follicles (8). P both within immune organs and within the body. Che- The impact of Gnai2 deficiency on T cell chemotaxis and T cell moattractants act as signposts to recruit and position lymphocytes trafficking has not been reported. Yet defective T cell function has and dendritic cells in lymphoid organs and inflammatory sites (1, been documented. Gnai2Ϫ/Ϫ mice develop a Th1-mediated in- 2). Most chemoattractants and chemokines signal through G pro- flammatory colitis reminiscent of human ulcerative colitis, whose tein-coupled receptors (GPCRs)2 that use the heterotrimeric G pro- penetrance depends upon the genetic background of the mice (10). Ϫ Ϫ / by guest on September 23, 2021 tein Gi to activate downstream effectors (3, 4). The binding of Gnai2 CD4 T cells exhibit augmented responses to TCR sig- ligand activates receptors triggering G␣i subunits to exchange GTP naling with enhanced intracellular calcium release and cytokine Ϫ/Ϫ for GDP, resulting in the dissociation of the G␣ subunit from its production; in contrast, Gnai3 T cells respond normally to associated G␤␥ heterodimers (5, 6). The release of Gi-associated TCR signaling (9). A relative increase in mature thymocytes in the Ϫ/Ϫ G␤␥ subunits is necessary for triggering directional migration (3, 4, thymus has also been noted in the Gnai2 mice (10). 7). Because G␣ subunits possess an intrinsic GTPase activity, GTP To characterize the impact of Gnai2 deficiency on T cell che- hydrolysis leads to the reassembly of heterotrimeric G protein, motaxis and T cell trafficking, we have examined CD4 and CD8 T causing signaling to cease (5, 6). Lymphocytes strongly express cells from wild-type, Gnai2ϩ/Ϫ, and Gnai2Ϫ/Ϫ mice. We find that Ϫ/Ϫ Ϫ/Ϫ two members of the Gi subfamily, G␣i2 and G␣i3 (8). Gnai3 Gnai2 T cells exhibit significant defects in chemokine receptor mice are reportedly without a phenotype (9); however, Gnai2Ϫ/Ϫ signaling and lymphocyte trafficking, suggesting that signals that mice exhibit defective B cell chemokine receptor signaling, as ev- modulate the level of G␣i2 present in lymphocytes directly affect idenced by depressed B cell chemotaxis, defective B cell homing the capacity of lymphocytes to respond to chemokines. The to lymph nodes, poor B cell adherence to lymph node high endo- Gnai2Ϫ/Ϫ mice have pronounced defects in chemokine receptor signaling, indicating that Gnai3 and Gnai1 poorly compensate for the loss of Gnai2 and that CXCR5 and CCR7 predominantly cou-

ple to Gi␣2 in T lymphocytes. Finally, these results argue that Laboratory of Immunoregulation, National Institute of Allergy and Infectious Dis- GPCR signaling significantly impacts the movement of T cells eases, National Institutes of Health, Bethesda, MD 20892 within the lymph node. Received for publication September 25, 2006. Accepted for publication April 16, 2007. The costs of publication of this article were defrayed in part by the payment of page charges. This article must therefore be hereby marked advertisement in accordance Materials and Methods with 18 U.S.C. Section 1734 solely to indicate this fact. Mice 1 Address correspondence and reprint requests to Dr. John H. Kehrl, Laboratory of Ϫ/Ϫ Immunoregulation, National Institute of Allergy and Infectious Diseases, National The generation of Gnai2 mice has been previously described (10). The Institutes of Health, Building 10, Room 11B08, 10 Center Drive MSC 1876, Be- mutation was backcrossed onto a C57BL/6 background six times. Mice thesda, MD 20892. E-mail address: [email protected] heterozygotic or homozygotic for the mutation were produced by crossing heterozygotic mice. Wild-type littermates were used as controls. C57BL/6 2 Abbreviations used in this paper: GPCR, G protein-coupled receptor; 3-D, three- 2ϩ 2ϩ mice were purchased from The Jackson Laboratory. All mice used in this dimensional; [Ca ]i, intracellular Ca concentration; CMFDA, 5-chloromethyl- fluorescein diacetate; CMTMR, 5-(and 6-)(((4-chloromethyl)benzoyl) amino)tetram- study were 8–14 wk of age. Mice were housed under specific pathogen- ethylrhodamine; FRC, fibroreticular cell; HEV, high endothelial venule; S1P, free conditions and used in accordance with the guidelines of the Institu- sphingosine 1-phosphate. tional Animal Care Committee at the National Institutes of Health. www.jimmunol.org 440 DEFICIENCY OF G␣i2 INHIBITS CHEMOKINE RECEPTOR SIGNALING

Reagents cytometer (BD Biosciences), and the data were analyzed using FlowJo software (Tree Star). Forward and side scatter parameters were used to gate Abs against mouse CD11a, CD11c, GR-1, CD49d, CXCR4, CXCR5, on live cells. Alternatively, the labeled cells were injected into the footpad CCR5, CD4, CD8a, B220, and CD62L were purchased from BD Pharm- of recipient mice, and 12 h later the ipsilateral popliteal and inguinal lymph ␣ ␣ ␣ ingen; CCR7 from BioLegend; and G i1,G i2, and G i3 from Santa Cruz nodes and the contralateral popliteal lymph node were removed and pro- Biotechnology. Streptavidin conjugated to PE was purchased from BD cessed, as above. In some instances, the T cells were pre incubated with Pharmingen. Pertussis toxin was purchased from Calbiochem. The 5-(and 100 ng/ml pertussis toxin for2hat37°C before transfer. 6-)(((4-chloromethyl)benzoyl) amino)tetramethylrhodamine (CMTMR) and 5-chloromethylﬂuorescein diacetate (CMFDA) were purchased from Lymph node transit assay Molecular Probes. Murine CCL19, CXCL12, and CXCL13 were purchased from R&D Systems. The assay was performed as previously described (8). Splenic CD4 T cells from wild-type or Gnai2Ϫ/Ϫ mice were labeled with either 2 ␮M CMFDA Cells or CMTMR for 15 min at 37°C, and 7–20 million cells of each population were injected i.v. to recipient mice. Two hours later, the mice were injected Splenic T cells were isolated by negative depletion using biotinylated Abs i.v. with either PBS or anti-L-selectin Ab (100 ␮g/mouse). After 12 h, to B220, GR-1, and CDllc and Dynabeads M-280 streptavidin (Dynal Bio- inguinal and popliteal lymph nodes were removed and gently dissociated tech), as previously described (11). The addition of a biotinylated Ab to into single-cell suspensions. Flow cytometric analysis was performed on a CD4 or CD8 allowed isolation of CD8 or CD4 T cells, respectively. The FACSCalibur ﬂow cytometer (BD Biosciences), and the data were ana- cell purity was greater than 95%. Cells were placed in complete RPMI lyzed using the FlowJo software (Tree Star). Forward and side scatter 1640 medium supplemented with 10% FCS, 2 mM L-glutamine, 100 IU/ml parameters were used to gate on live cells. penicillin, 100 ␮g/ml streptomycin, 1 mM sodium pyruvate, and 50 ␮M 2-ME. Chemotaxis

Intravital two-photon microscopy Chemotaxis assays were performed using a Transwell chamber, as previ- Downloaded from ously described (8, 11). In some experiments, T cells were incubated with Splenic CD4 or CD8 T cells prepared from wild-type, Gnai2ϩ/Ϫ,or Ϫ/Ϫ ϳ ␮ 100 ng/ml pertussis toxin for2hat37°C. The cells were washed twice, Gnai2 were labeled for 15 min at 37°C with 2.5 5 M CMTMR or resuspended in complete RPMI 1640 medium, and added in a volume of CMFDA (8). Seven to 10 million labeled cells of each population in 200 ␮ ␮ ␮ ϳ 100 l to the upper wells of a 24-well Transwell plate with a 5- m insert. l of PBS were adoptively transferred by tail vein injection into 6 10- Lower wells contained various doses of chemokines in 600 ␮l of complete wk-old recipient mice. Recipient mice were anesthetized with isoflurane RPMI 1640 medium. The number of cells that migrated to the lower well (Baxter). The isoflurane was used at 2.5% for induction and 1–1.5% for following a 2-h incubation was counted using a flow cytometer. maintenance vaporized in an 80:20 mixture of oxygen and air. The inguinal http://www.jimmunol.org/ lymph node was prepared microsurgically for intravital microscopy using Determination of changes in intracellular Ca2ϩ concentration a slightly modified published surgical technique (8). The mouse was placed ([Ca2ϩ] ) on a heating pad, with the skin flap raised on a glass plate over a heating i pad. The inguinal lymph node was surgically exposed on a skin flap, and Cells were seeded at 105 cells per 100 ␮l loading medium (RPMI 1640, the surrounding tissue was attached to the base of Microwell Dish with 10% FBS) into poly(D-lysine)-coated 96-well black-wall, clear-bottom mi- veterinary-grade glue (Nexaband), and 3 ml of prewarmed PBS at 37°C crotiter plates (Nalge Nunc International). An equal volume of assay load- was added. The temperature of PBS, tissue near the lymph node, and the ing buffer (FLIPR Calcium 3 assay kit; Molecular Devices) in HBSS sup- surrounding closed air were monitored and maintained at 36.5 Ϯ 0.5°C. plemented with 20 mM HEPES and 2 mM probenecid was added. Cells Inguinal lymph node was intravitally imaged from the capsule (100ϳ200 were incubated for1hat37°C before adding chemokine, and then the ␮m deep). Two-photon imaging was performed with an inverted Leica calcium flux peak was measured using a FlexStation (Molecular Devices). by guest on September 23, 2021 TCS-SP2 MP confocal microscope (Leica Microsystems) equipped with The data were analyzed with SOFT max Pro (Molecular Devices). Data are ϫ20 water-immersion objective, 0.5 W U-V-I (immersion medium used shown as fluorescent counts, and the y-axis is labeled as Lm1. Dulbecco’s PBS). Two-photon excitation was provided by a Mai Tai Ti: Sapphire laser (Spectra Physics) with a 10 W pump, tuned to 800 nm. For Immunoblotting four-dimensional analysis of cell migration, stacks of 12 section (z step ϭ 3 ␮m) were acquired every 30 s to provide an imaging volume of up to 33 Cell lysates were prepared as previously described (8). The detergent-in- ␮m in depth. Emitted fluorescence was collected using a two-channel non- soluble materials were removed, and equal amounts of protein were frac- descanned detector. Wavelength separation was through a dichroic mirror tionated by 4–20% SDS-PAGE and transferred to polyvinylidene difluo- at 560 nm, followed by 525/50 nm, and 610/75 nm emission filters. Se- ride membranes. Membranes were blocked with 5% BSA in Tween 20 plus quences of image stacks were transformed into volume-rendered four-di- PBS for 1 h and then incubated with an appropriate dilution of the primary mensional movies using Imaris software v.5.0.1 (Bitplane), and the spot Ab in 5% BSA in Tween 20 plus PBS for 2 h. The blots were incubated analysis was used for semiautomated tracking of cell motility in three di- with biotinylated Ab for 1 h, and further incubated with streptavidin con- mensions by using the parameters of autoregressive motion as an algo- jugated to HRP for 1 h. The signal was detected by ECL (Amersham rithm, 7 ␮m spot diameter, 2 ␮m background object diameter, and 15 ␮m Biosciences). maximum distance (every 30 s). The data set was corrected for tissue drift Statistics by the Imaris software. Calculations of the cell velocity, track straightness, and cell displacement were performed using the Imaris software. The cell In vivo results represent samples from three to six mice per experiment. In polarity was calculated by measuring long and short axes of individual vitro results represent mean values of sextuplet samples. All experiments cells using the point-to-point measurement feature in Imaris software. were performed more than three times. SD and p values were calculated Blood vessels within the inguinal lymph node were revealed by i.v. injec- with Student’s t test using Microsoft Excel software. tion of 0.05 mg of tetramethylrhodamine dextran (3 kDa; Molecular Probes) and 0.1 mg of FITC-dextran (150 kDa; Sigma-Aldrich) into the Results anesthetized mouse immediately before T cell imaging. For three-dimen- ϩ/Ϫ Ϫ/Ϫ sional analysis, stacks of 267 section (z step ϭ 1 ␮m) were acquired to Gnai2 and Gnai2 T lymphocytes express reduced levels provide an imaging volume of up to 266 ␮m in depth. Individual slices and of Gi␣2 compared with wild-type cells and the effects of T cell three-dimensional (3-D) reconstruction was done using the Surpass feature activation on Gi␣2 levels of the Imaris software. T cell lysates from splenic T cells prepared from wild-type, Homing assays Gnai2ϩ/Ϫ, and Gnai2Ϫ/Ϫ mice were immunoblotted for the pres- ϩ Ϫ ϩ Ϫ Ϫ Ϫ / Splenic CD4 or CD8 T cells from wild-type, Gnai2 / ,orGnai2 / mice ence of Gi␣1,Gi␣2, and Gi␣3. The T cells prepared from Gnai2 ϳ were labeled with either 1 ␮M CMFDA or 2.5 ␮M CMTMR for 15 min at mice displayed 50% less Gi␣2 than did wild-type T cells, 37°C, and 7–20 million cells of each population were injected i.v. to re- whereas no detectable Gi␣2 was found in the T cells prepared from cipient mice. After 2 h, spleen, inguinal lymph nodes, and popliteal lymph the Gnai2Ϫ/Ϫ mice. As we had previously observed with nodes were removed, and gently dissociated into single-cell suspensions. Ϫ/Ϫ Ϫ/Ϫ Peripheral blood was collected by retro-orbital eye bleeding. After remov- Gnai2 B lymphocytes (8), Gnai2 T cells express very low levels of G ␣ and an increased level of G ␣ compared with wild- ing RBC with Tris-NH4Cl, the cells were resuspended in PBS containing 1% i 1 i 3 BSA at 4°C. Flow cytometric analysis was performed on a FACSCalibur flow type T cells (Fig. 1A). To explore the impact of T cell activation on The Journal of Immunology 441

did not detect a signiﬁcant difference between the stimulated and unstimulated cells (data not shown).

Gnai2ϩ/Ϫ and Gnai2Ϫ/Ϫ T cells have altered chemokine receptor expression To explore whether haploinsufficiency or the lack of Gnai2 expression impacted chemokine receptor expression and other receptors involved in lymphocyte trafficking, we examined freshly isolated splenic CD4 and CD8 T cells from wild-type and the mutant mice for their levels of CCR5, CCR7, CXCR4, L-selectin, CD11a, and CD49d expression by flow cytometry. We found either a very modest or no change in receptor expression on both CD4 and CD8 T cells from the Gnai2ϩ/Ϫ mice, and mild to modest decreases of each of the tested receptors on CD4 and CD8 T cells prepared from the Gnai2Ϫ/Ϫ mice (Fig. 1B). In contrast, the expression levels of CD4 and CD8 on T cells from Gnai2ϩ/Ϫ and Gnai2Ϫ/Ϫ mice were similar to wild-type mice. The mechanism accounting for the re- ductions in chemokine receptor expression in the Gnia2Ϫ/Ϫ T cells Downloaded from is not known. Whether the reduction in G␣ subunit expression impacts the expression of GPCRs that couple to that G␣ subunit is not an issue that has been addressed to our knowledge. Alterations in peripheral T cells trafficking receptors may also arise from altered thymocyte development and thymocyte egress that most

likely occurs as a consequence of the absence of Gi␣2. The reduction in L-selectin levels noted in the Gnai2Ϫ/Ϫ CD4 and CD8 T http://www.jimmunol.org/ cells most likely impacts the homing of these cells to lymphoid organs because T cells from L-selectinϩ/Ϫ mice exhibit a greater than 50% reduction in homing to lymph nodes (12).

Gnai2ϩ/Ϫ T cells have diminished in vitro responses to chemokines and mildly impaired in vivo trafficking To determine whether a reduction in Gnai2 expression altered the responsiveness of T cells to chemokine stimulation, we first as- by guest on September 23, 2021 sessed responses in standard chemotaxis assays using increasing concentrations of CCL19 and CXCL12. Both CD4 and CD8 T cells from Gnai2ϩ/Ϫ mice migrated at reduced frequency compared with wild-type cells at each tested concentration of CXCL12 and CCL19 (Fig. 2A). The reduction averaged 25–30% for both chemokines. We also tested B cells from Gnai2ϩ/Ϫ mice, and they also showed a reduction in responsiveness to chemokines with decreases of 40 and 50%, when stimulated with optimal concentrations of CXCL12 and CCL19, respectively (data not shown). Pretreatment with pertussis toxin blocked the chemotaxis of both wild-type and Gnai2ϩ/Ϫ T and B cells (Fig. 2A). To assess the consequence of reducing Gnai2 expression on T cell trafficking in vivo, we examined the homing of transferred CD4 T cells and their motility in the inguinal lymph node of re- FIGURE 1. Expression of Gi␣ isoforms and chemokine receptors in T cells from Gnai2ϩ/Ϫ and Gnai2Ϫ/Ϫ mice. A, Immunoblot of cell lysates cipient mice by intravital imaging. We performed six experiments prepared from wild-type, Gnai2ϩ/Ϫ, and Gnai2Ϫ/Ϫ T cells for expression that examined CD4 T cell homing to the spleen, inguinal, and popliteal lymph nodes. In each experiment, fewer Gnai2ϩ/Ϫ CD4 of Gi␣1,Gi␣2,Gi␣3, and actin. B, Flow cytometry of splenic CD4 or CD8 T cells prepared from either wild-type, Gnai2ϩ/Ϫ,orGnai2Ϫ/Ϫ mice. Wild- T cells homed to the lymph nodes as compared with wild-type type profiles are nonshaded, Gnai2ϩ/Ϫ profiles are shaded in light gray, and cells simultaneously transferred, a result consistent with the pre- Gnai2Ϫ/Ϫ are shaded in dark gray. Isotype controls are shown with a light vious chemotaxis assays. On average 30 and 24% fewer cells en- gray line. Cell type and marker are indicated in the upper right-hand corner tered the popliteal and inguinal lymph nodes, respectively (Fig. of each profile. Representative results are from four separate experiments. 2B). In addition, we noted a modest increase in the number of Gnai2ϩ/Ϫ CD4 T in the spleen and peripheral blood of the recipient mice as compared with wild-type cells. We did not check the Gnai1, Gnai2, and Gnai3 expression, we performed RT-PCR us- efficiency at which the Gnai2ϩ/Ϫ CD4 T cells entered the splenic ing RNA samples prepared from wild-type CD4 and CD8 T cells white pulp, although we would expect a modest reduction similar stimulated for various durations with a mAb directed at CD3 and to that observed in lymph node entry. We also determined whether IL-2. We found a 2-fold enhancement in Gnai2 and Gnai3 expres- Gnai2ϩ/Ϫ CD4 T cells trafficked normally to draining lymph sion and no change in the low level of Gnai1 after 24 h of stim- nodes. We transferred labeled wild-type and Gnai2ϩ/Ϫ CD4 T into ulation; however, when we immunoblotted for Gi␣2 and Gi␣3 we the footpads of recipient mice and determined the percentage of 442 DEFICIENCY OF G␣i2 INHIBITS CHEMOKINE RECEPTOR SIGNALING Downloaded from

FIGURE 2. Gnai2ϩ/Ϫ T cells have reduced responses to chemokines and exhibit mild defects in homing. A, Chemotaxis assays. CD4 (first and second panels) and CD8 (third and fourth panels) T cells purified from wild-type (WT, u) and Gnai2ϩ/Ϫ (f) mice were subjected to a 2-h chemotaxis in response to CXCL12 or CCL19, as indicated. The percentages of cells responding to 10, 30, and 100 ng/ml CXCL12 or CCL19 are shown. Results are mean and (p Ͻ 0.01 vs wild type). In some instances, the cells were pretreated with pertussis toxin (100 ng/ml ,ء) SE of sextuplet samples from four experiments for 2 h and washed, before being subjected to a chemotaxis assay. B, Homing of i.v. transferred CD4 T cells. Ten million labeled Gnai2ϩ/Ϫ T cells and /p Ͻ 0.01 vs wild http://www.jimmunol.org ,ء ;the same number of labeled wild-type T cells were i.v. transferred to recipient mice (six experiments each with three pairs of mice type). Two hours after transfer, cells from blood, spleen, inguinal lymph nodes, and popliteal lymph nodes of recipient mice were analyzed using flow cytometer. Shown are the actual number of transferred CD4 T cells found in blood (per 10 ␮l), spleen (ϫ104 cells), inguinal lymph nodes (ϫ102 cells), and popliteal lymph nodes (ϫ10 cells). C, Homing of footpad-injected CD4 T cells. Ten million labeled Gnai2ϩ/Ϫ T cells and the same number of labeled wild-type T cells were injected into the footpad of recipient mice (four experiments each with three pairs of mice). Twelve hours after transfer, cells from ipsilateral and contralateral popliteal and ipsilateral inguinal lymph nodes of recipient mice were analyzed using flow cytometer. Shown are the percentages of recovered cells. D, Velocity profiles. Ten million labeled Gnai2ϩ/Ϫ T cells and the same number of labeled wild-type T cells were transferred i.v. to wild-type recipients. Twenty-four hours after cell transfer, inguinal lymph nodes were imaged intravitally using multiphoton microscope. Velocity profile of wild-type (Ⅺ) and Gnai2ϩ/Ϫ (f) cells. The average speeds are indicated with arrows. E, Mean velocity of wild-type and Gnai2ϩ/Ϫ T cells. Speed was

calculated from three independent movies by spot analysis using Imaris software. by guest on September 23, 2021

transferred cells that entered into the ipsilateral popliteal and in- migratory responses (percentage migrated in the presence of che- guinal lymph nodes and the contralateral popliteal lymph node. We mokine minus percentage migrated in the absence of chemokine) noted a slight reduction in the percentage of the recovered of the Gnai2Ϫ/Ϫ CD4 and CD8 T cells at the highest concentration Gnai2ϩ/Ϫ CD4 T in the ipsilateral popliteal lymph node compared of CXCL12 and CCL19 tested were reduced by 4- to 7-fold com- with wild-type cells (Fig. 2C). We also assessed Gnai2ϩ/Ϫ CD4 T pared with wild-type T cells (Fig. 3B). Similar to what we had cell motility by intravital imaging. We again performed six ex- observed with the Gnai2Ϫ/Ϫ B cells, the majority of the residual periments, each of which revealed a slight decrease in T cell responses of the Gnai2Ϫ/Ϫ CD4 and CD8 T cells to CCL19 and motility compared with wild-type cells. The velocity profile of CXCL12 were resistant to pretreatment with pertussis toxin (Fig. the Gnai2ϩ/Ϫ T cells was slightly shifted to the right compared 3B). In contrast, pertussis toxin pretreatment completely inhibited with T cells from wild-type mice; however, the difference was the chemotaxis of wild-type T cells. not statistically significant (Fig. 2D). The average velocity of Ϫ/Ϫ the Gnai2ϩ/Ϫ CD4 T cells was ϳ10% less than that observed Impaired lymph node homing of T cells from Gnai2 mice with simultaneous transferred wild-type CD4 T cells (Fig. 2E). Next, we compared the homing of transferred CD4 T cells and Ϫ/Ϫ 2ϩ CD8 T cells from wild-type and Gnai2 mice to peripheral Markedly reduced chemotaxis and [Ca ]i mobilization Ϫ/Ϫ lymph nodes, spleen, and blood. The various lymphocyte popula- following exposure of Gnai2 lymphocytes to CXCL12 or tions showed similar defects in homing with markedly reduced CCL19 homing to popliteal and inguinal lymph nodes compared with 2ϩ ϳ Examination of the increases in [Ca ]i following chemokine ex- wild-type cells. We recovered 6.5-fold more wild-type CD4 T posure of Gnai2ϩ/Ϫ and Gnai2Ϫ/Ϫ CD4 T cells revealed a signif- cells and 4-fold more wild-type CD8 T cells from the two sets of icant and more marked reduction in their responses, respectively, lymph nodes than we did the corresponding Gnai2Ϫ/Ϫ populations as compared with wild-type CD4 T cells (Fig. 3A). This indicates (Fig. 4A). By way of comparison, the recovery of wild-type B cells that the major mechanism by which CXCL12 and CCL19 increase from the inguinal and popliteal lymph nodes exceeded by 5.5-fold 2ϩ Ϫ/Ϫ [Ca ]i levels is directed by G␤␥ subunits associated with Gi␣2. the recovery of the Gnai2 B cells. These results are similar to We have previously reported markedly reduced chemotactic re- what we observed with the Gnai2ϩ/Ϫ cells, although the magni- sponses of Gnai2Ϫ/Ϫ B cells to CXCL12 and CCL19. The tude of the impairment was greater with the Gnai2Ϫ/Ϫ cells. Gnai2Ϫ/Ϫ CD4 T and CD8 T were slightly less impaired than the We also examined the ability of wild-type and Gnai2Ϫ/Ϫ T cells Gnai2Ϫ/Ϫ B cells, but all three cell types migrated poorly to to migrate from peripheral tissues into draining lymph nodes. We CXCL12 and CCL19 at each concentration tested. The specific injected each number of differentially labeled wild-type and The Journal of Immunology 443

thereby slowing Gnai2Ϫ/Ϫ T lymphocyte egress (13–15). A pre- vious study has shown that pertussis toxin treatment inhibits lym- ϩ/Ϫ phocyte egress from lymph nodes and that S1P1 cells have reduced lymph node exit efficiency (15). To compare wild-type and Gnai2Ϫ/Ϫ T cell retention in peripheral lymph nodes, we transferred differentially labeled cells to wild-type recipients, and 2 h later blocked further lymphocyte ingress by the i.v. injection of an L-selectin-specific Ab (Fig. 4D). We compared the number of wild-type and Gnai2Ϫ/Ϫ T cells in lymph nodes 24 h after Ab injection. We also plotted the data as the ratio between the recovered cells in the presence or absence of Ab. The treatment with anti-L-selectin Ab significantly reduced lymph node sizes and the recovery of lymphocytes from them. As expected, fewer Gnai2Ϫ/Ϫ T cells entered peripheral lymph nodes; however, based on the ratio indicated above, the Gnai2Ϫ/Ϫ T cells exited lymph nodes as well or better than did wild-type cells (Fig. 4E).

Intravital imaging of wild-type and Gnai2Ϫ/Ϫ CD4 T cells Finally, we wanted to compare the distribution and motility of Downloaded from Gnai2Ϫ/Ϫ CD4 T cells with that of normal cells. To determine whether wild-type and Gnai2Ϫ/Ϫ CD4 T cells localized similarly within lymph nodes, we transferred differentially labeled wild-type and Gnai2Ϫ/Ϫ CD4 T cells into a recipient mouse and directly imaged the cells in the inguinal lymph node by intravital multipho- FIGURE 3. Markedly impaired chemokine-triggered responses in ton microscopy acquiring a single stack of images at 1-␮m inter- http://www.jimmunol.org/ Ϫ/Ϫ 2ϩ Gnai2 T cells. A, Measurement of changes in [Ca ]i. Splenic T cells vals to a depth of 266 ␮m. Just before imaging, we injected a ϩ/Ϫ Ϫ/Ϫ from wild-type, Gnai2 ,orGnai2 mice were incubated for1hat mixture of green and red fluorescent dextran to delineate the mi- 37°C in the calcium assay loading buffer before adding CXCL12 or CCL19 2ϩ crocirculation of the lymph node. Five images are shown, the most (100 ng/ml). Changes in [Ca ]i were monitored over 3 min. The data were analyzed with SOFT max Pro and are shown as fluorescent counts, and the superficial image acquired (0), and images from 50, 100, 200, and ␮ 3 y-axis is labeled as Lm1. Experiment was performed twice with similar 265 m deep (Fig. 5A; see supplemental movie 1). The most results. B, Chemotaxis assays. CD4 (upper panels) and CD8 (lower panels) superficial image shows vessels near the surface of the lymph T cells purified from wild-type (WT, u) and Gnai2Ϫ/Ϫ (f) mice were node, fat pads on the right side of the image, and a lymph node subjected to a 2-h chemotaxis in response to CXCL12 or CCL19, as indi- follicle. At a depth of 50 ␮m, the lymph node follicle is well by guest on September 23, 2021 cated. Results are the mean of sextuplet samples from four experiments visualized and along the edge of the follicle wild-type (green) and p Ͻ 0.01 vs wild type). In some instances, the cells were Gnai2Ϫ/Ϫ CD4 T cells (red) are seen. Despite reduced lymph node ,ء) performed pretreated with pertussis toxin (100 ng/ml) for 2 h and washed, before homing, more Gnai2Ϫ/Ϫ CD4 T cells localized along the edge of being subjected to a chemotaxis assay. the lymph node follicle than did wild-type cells. This was even more evident at a depth of 100 ␮m, in which the ratio of Gnai2Ϫ/Ϫ

Ϫ/Ϫ CD4 T cells to wild-type T cells exceeded 2:1. In contrast, at a Gnai2 CD4 T cells into the footpad of a wild-type mouse depth of 200 and 265 ␮m, the number of wild-type cells far out- and determined the relative percentage of cells in the ipsilateral numbers Gnai2Ϫ/Ϫ CD4 T cells, although several small groups of and contralateral popliteal and inguinal lymph nodes. We did not Gnai2Ϫ/Ϫ CD4 T cells were present. Also shown is a x,z projection recover transferred cells from the inguinal lymph nodes or from of 20 ␮m compressed along the y-axis. Individual images of wild- the contralateral popliteal lymph node, but could recover cells type (green), Gnai2Ϫ/Ϫ CD4 T cells (red), and the composite im- from the ipsilateral popliteal lymph node. In the ipsilateral popli- age are shown (Fig. 5B). This again demonstrates the differing ϳ teal lymph node, we recovered 2.5-fold more transferred wild- Gnai2Ϫ/Ϫ Ϫ/Ϫ localization of the wild-type and CD4 T cells. Next, we type cells than we did Gnai2 cells (Fig. 4B). To assess the acquired a series of 3-D image stacks sequentially every 30 s at effect of pertussis toxin on T cell migration from peripheral tissues ␮ Ϫ/ϩ depths between 100 and 150 m (see supplemental movies 2 and into draining lymph nodes, we treated either wild-type, Gnai2 , 3 Ϫ/Ϫ 3). We noted a marked difference in the morphology of the wild- or Gnai2 CD4 T cells with pertussis toxin before transfer. The type and the Gnai2Ϫ/Ϫ CD4 T cells. The majority (ϳ70%) of the pretreatment completely inhibited the appearance of wild-type and Ͻ Ϫ/ϩ wild-type T cells had a polarized morphology, whereas 10% of Gnai2 CD4 T cells. In contrast, the reduced homing of the Gnai2Ϫ/Ϫ Ϫ/Ϫ the T cells did so in the imaging ﬁles. Using the ratio of Gnai2 CD4 T cells was only modestly affected by prior per- the cell length to cell width as a measure of cell polarity, the tussis toxin treatment (Fig. 4C). Ϫ/Ϫ wild-type CD4 T cells had an average polarity index of 1.8, For several reasons, the duration that Gnai2 T cells remain whereas the Gnai2Ϫ/Ϫ CD4 T cells had an index of 1.1 (Fig. 5C). within the lymph node might be different from that of wild-type T z Ϫ/Ϫ Time-lapse videos compressed along the -axis revealed that the cells. First, the reduced motility of Gnai2 T cells may impact wild-type and Gnai2Ϫ/Ϫ CD4 T cells moved within the lymph their retention rate. Second, the presence of chemokines within the node; however, the customary amoeboid-like cell movements and lymph node cortex may normally retain T lymphocytes there, or active membrane projections normally observed were less evident conversely, chemokines located at exit sites may promote T cell with the Gnai2Ϫ/Ϫ CD4 T cells (see supplemental movies 2–4).3 migration to those sites. In the absence of Gi␣2, those signals Tracking individual cells over a 20-min interval revealed that the would be greatly weakened. Third, sphingosine 1-phosphate (S1P) receptors signal lymph node egress via the activation of Gi, and the 3 absence of Gi␣2 could reduce the strength of an egress signal, The online version of this article contains supplemental material. 444 DEFICIENCY OF G␣i2 INHIBITS CHEMOKINE RECEPTOR SIGNALING Downloaded from http://www.jimmunol.org/

FIGURE 4. Gnai2Ϫ/Ϫ T cells home poorly to lymph nodes, but have no impairment in lymph node transit. A, Homing of i.v. transferred CD4 and CD8 by guest on September 23, 2021 T cells. Ten million labeled CD4 T cells, CD8 T cells, or B cells prepared from Gnai2Ϫ/Ϫ mice along with the same number of differentially labeled cells p Ͻ 0.01 vs wild type). Two ,ء ;from wild-type mice were i.v. transferred to recipient mice (results are from six experiments each with three pairs of mice hours after transfer, cells from blood, spleen, inguinal lymph nodes, and popliteal lymph nodes of recipient mice were analyzed using flow cytometer. Shown are the actual number of transferred cells found in blood (per 10 ␮l), spleen (ϫ104 cells), inguinal lymph nodes (ϫ102 cells), and popliteal lymph nodes (ϫ10 cells). B, Homing of footpad-injected CD4 T cells. Ten million labeled Gnai2Ϫ/Ϫ CD4 T cells and the same number of labeled wild-type CD4 p Ͻ 0.01 vs wild type). Twelve ,ء ;T cells were injected into the footpad of recipient mice (results from four experiments each with three pairs of mice hours after transfer, cells from ipsilateral and contralateral popliteal and ipsilateral inguinal lymph nodes of recipient mice were analyzed using flow cytometer. Shown are flow cytometry results from the ipsilateral and contralateral popliteal lymph nodes and a chart of the percentages of recovered cells. C, Pertussis toxin effects on the homing of CD4 T cells. Similar experiment as B, but the wild-type, Gnai2ϩ/Ϫ,orGnai2Ϫ/Ϫ CD4 T cells were pretreated with pertussis toxin or not. Data are shown as percentages of wild-type, nonpertussis toxin treated. D, T cell transit through lymph nodes. Ten million labeled Gnai2Ϫ/Ϫ T cells and the same number of differentially labeled wild-type cells were i.v. transferred to recipient mice (results from four experiments p Ͻ 0.01 vs wild-type). Anti-L-selectin Ab (100 ␮g/mouse, f) was injected i.v. 2 h after cell transfer. Control mice were ,ء ;each with three pairs of mice injected with PBS (u). Cells were isolated from inguinal and popliteal lymph nodes and analyzed using flow cytometer 24 h after Ab injection. Actual numbers of transferred cells recovered from the inguinal (left panel) and popliteal (right panel) nodes are shown. E, The ratios between Ab- and PBS-treated control groups are shown.

Gnai2Ϫ/Ϫ CD4 T cells exhibited oscillatory movements and rather individual cells relies on ﬁtting mean square displacements to a short tracks as compared with the longer and less erratic movement persistent random walk function. Focusing on 15 cells that re- of the wild-type CD4 T cells (Fig. 5D; supplemental movies 4 and mained in the tracking volume for 4 min, we plotted the mean 5).3 Randomly chosen tracks of CD4 T cells prepared from the squared displacement vs time. Both the wild-type and the wild-type or Gnai2Ϫ/Ϫ mice and plotted from the same origin Gnai2Ϫ/Ϫ CD4 T cells exhibited a linear relationship, although the show a striking difference in the length of the individual tracks slope, which is a measure of cell motility, differed substantially (Fig. 5E). From the tracking data, we extracted the displacement of (Fig. 5G). The slope of the wild-type cells was 275, whereas that individual cells over 20 min of imaging. These data are shown by of the Gnai2Ϫ/Ϫ CD4 T cells was 17. A measure of directional arrows beginning when individual cells are ﬁrst visualized and persistence is track straightness (displacement/track length). ending when they leave the imaging space or the imaging termi- Twelve randomly chosen tracks were sorted on the basis of their nated. Displacement of wild-type (green) and Gnai2Ϫ/Ϫ (red) CD4 track straightness and plotted (Fig. 5H). Despite their slow veloc- T cell mice is shown separately (Fig. 5F, left and right panels). The ity, there was little difference between the wild-type and Gnai2Ϫ/Ϫ displacements of 132 wild-type and Gnai2Ϫ/Ϫ CD4 T cell tracks CD4 T cells with the exception of three wild-type cells that acquired over a 10-min interval were sorted by length and plotted exhibited relatively straight tracks. When we analyzed the tracks of (Fig. 5G). One of the typical approaches to study motion paths of randomly chosen wild-type and Gnai2Ϫ/Ϫ CD4 T cells from one The Journal of Immunology 445

movie, we found that the wild-type cells moved with an average receptors may couple to nonpertussis toxin-sensitive G proteins. In speed of 8.8 ␮m/min, whereas the Gnai2Ϫ/Ϫ CD4 T cells moved fact, during T cell stimulation by APCs, T cell chemokine recep- ␮ with an average speed of 4.2 m/min (Fig. 5I, left panel). More tors coupled to Gq were shown to be recruited to the immunolog- Ϫ Ϫ Gnai2 / CD4 T cells moved with an average speed below 5 ical synapses (23). Our results contrast to studies examining the ␮ m/min as compared with wild-type cells, whereas relatively few requirement for Gi isoforms in C5a receptor-triggered chemotaxis of them moved with average speed above 10 ␮m/min. An exam- of a mouse macrophage cell line (24). Although pertussis toxin ination of three movies from separate imaging experiments re- blocks C5a receptor-triggered chemotaxis, the reduction of either Ϫ/Ϫ vealed a similar difference between wild-type and the Gnai2 Gnai2 expression (90% knockdown) or Gnai3 expression (80% CD4 T cells (Fig. 5I, right panel). We also acquired time-lapse knockdown) by lentiviral delivery of small interfering RNA did videos compressed along the z-axis from deeper in the lymph node not. This suggests that the C5a receptor couples to both G iso- ϳ ␮ Ϫ/Ϫ i ( 250 m) of the wild-type and Gnai2 CD4 T cells (supple- forms, and each is capable of triggering chemotaxis. 3 Ϫ/Ϫ mental movie 6). Relatively few Gnai2 CD4 T cells are Despite the reduced migration of Gnai2Ϫ/Ϫ CD4 T cells into the present compared with wild-type CD4 T cells (Fig. 6A). Tracking Ϫ/Ϫ deeper lymph node cortex and their reduced motility, we found individual cells showed that the motility of the Gnai2 CD4 T cells Ϫ Ϫ that the Gnai2 / CD4 T cells transited through the lymph node as was less impaired than observed more superficially in the lymph node well, if not better than did wild-type cells. Normally, entering T (Fig. 6B). The wild-type T cells moved with an average speed of 10 Ϫ Ϫ cells scan cortical ridge-dendritic cells searching for specific Ag ␮m/min, whereas the Gnai2 / CD4 T cells moved at an average (25). If they fail to find Ag, they pass through the cortical ridge, speed of 7.8 ␮m/min (data not shown). The lesser impairment may be Ϫ Ϫ enter the deeper cortical zones, and migrate toward the medulla, secondary to the selection of those Gnai2 / CD4 T best able to Downloaded from respond to chemoattractants. Nevertheless, a comparison of individual where they exit into the efferent lymphatics. The presence of high tracks from those cells that remained within the imaging space for levels of S1P within the efferent lymph facilitates lymphocyte Ϫ/Ϫ similar duration (of 6–8 min) of the 30-min imaging period revealed egress (13–15). Our data indicate that the Gnai2 T lymphocyte a significant decrease in the motility of the Gnai2Ϫ/Ϫ CD4 T cells can readily access the medullary region to exit into lymphatics. It compared with wild-type cells (Fig. 6B). is worth noting that the cortical ridge directly connects to the me- Gnai2Ϫ/Ϫ dulla, perhaps providing an avenue for T cell egress. http://www.jimmunol.org/

Discussion Our data also argue that S1P receptors can couple to Gi␣3 to trigger Naive T cells preferentially recirculate between the blood and sec- egress. ondary lymphoid organs, whereas memory and effector T cells can Two competing theories explain how GPCRs (i.e., chemokine enter into nonlymphoid tissues and inﬂammatory sites, returning to receptors) transfer information after activation (26). The ﬁrst pos- lymph nodes via the afferent lymph. Pertussis toxin has proven to tulates that freely diffusible GPCRs collide with membrane-asso-

be a valuable reagent in discerning the role of Gi proteins in the ciated heterotrimeric G proteins, which, following GDP/GTP ex- entrance of lymphocyte into lymph nodes via either route (16–18). change, localize with their primary effectors. The other and more Pretreatment of T cells with pertussis toxin before i.v. transfer to recently favored theory postulates that individual GPCRs are pre- a recipient mouse completely blocks the entrance of cells into coupled to heterotrimeric G proteins, which are in turn preas- by guest on September 23, 2021 lymph nodes. Intravital imaging of transferred lymphocytes within sembled with their effectors. A key feature of this physical scaf- HEVs has shown that pretreatment with pertussis toxin blocks firm folding is fast activation and deactivation of the signaling pathway adhesion to HEVs, thereby preventing the exit of lymphocytes as well as specificity. Because individual heterotrimeric G protein from the blood and their entrance into lymph nodes (19). Pretreat- most likely competes for precoupling, the mass ratio between the ment of mouse splenocytes with pertussis toxin also completely cell’s repertoire of GPCRs and heterotrimeric G proteins and the inhibits the migration of T cells from peripheral tissues to draining relative affinities of different GPCRs for heterotrimeric G proteins, lymph node (20). various combinations of G␣,G␤, and G␥, determine the outcome of Yet the analysis of pertussis toxin-treated cells masks any dif- ligand exposure (27). In this context, the impaired responsiveness ϩ Ϫ ferential role of the various Gi␣ isoforms, all of which are sensitive of the Gnai2 / T cells to chemokine stimulation can be explained to pertussis toxin. T lymphocytes predominantly express Gnai2 as a consequence of an altered balance between G ␣ and other and Gnai3 with little expression of Gnai1. The marked reduction i 2 heterotrimeric G proteins expressed within lymphocytes. A de- in chemokine signaling and T cell homing to peripheral lymph crease in G allows the precoupling of other heterotrimeric G nodes noted with Gnai2Ϫ/Ϫ T cells occurs despite their enhanced i␣2 proteins, thereby altering the outcome of receptor activation. These expression of Gnai3. In addition, the residual responsiveness of Ϫ Ϫ observations suggest that changes in the expression levels of G Gnai2 / T cells to chemokines in vitro and in vivo is relatively ␣ subunits, which may occur as consequence of lymphocyte devel- insensitive to pertussis toxin treatment, arguing that those Ϫ/Ϫ opment, differentiation, or activation, may significantly alter the Gnai2 T cell chemokine receptors that have coupled to Gi␣3 can no longer link to the signaling pathways needed for chemo- outcome of chemokine receptor signaling. taxis. This could result from an inability of the chemokine recep- Initial imaging studies examining lymphocyte trafficking con- cluded that the movement of T and B cells within lymph nodes is tors to trigger Gi␣3 GDP/GTP exchange or because those G␤␥ sub- best modeled by a “random walk” behavior, that is lymphocytes units paired with Gi␣3 do not activate the appropriate effectors. A lack any organized directional movement (28–31). Recently, an precedent for divergent roles for Gi␣2 and Gi␣3 in linking GPCR activation to downstream signaling pathways has been described in alternative explanation for the T cell imaging data has been put cardiac myocytes (21, 22). forth postulating that T cell movements within the lymph node are The pertussis toxin resistance of the residual chemotaxis and controlled by the presence of many small local attractors, such as lymph node homing of the Gnai2Ϫ/Ϫ T cells indicates that CXCR4 dendritic cells secreting chemokines (32). As suggested, this issue and CCR7 in the Gnai2Ϫ/Ϫ mice have most likely coupled to a is of some importance because productive dendritic-T cell inter-

nonpertussis toxin-sensitive heterotrimeric G protein such as Gq or actions are more likely to occur if there is mechanism to bias naive G12, which has some capacity to trigger chemotaxis. This raises T cells toward interacting with Ag-bearing DCs. Supporting this the possibility that under certain circumstances, T cell chemokine model in immunogen-draining lymph nodes, naive CD8ϩ T cells 446 DEFICIENCY OF G␣i2 INHIBITS CHEMOKINE RECEPTOR SIGNALING Downloaded from http://www.jimmunol.org/ by guest on September 23, 2021 The Journal of Immunology 447

Ϫ/Ϫ

FIGURE 6. Comparison of wild-type and Gnai2 CD4 T cells imaged in the T cell cortex. A, Representative image from intravital microscopy Downloaded from acquired at a depth between 252 and 267 ␮m. Twenty-four hours after cell transfer, the inguinal lymph node was imaged intravitally using multiphoton microscope (Gnai2Ϫ/Ϫ CD4 T cells, red; wild-type T cells, green) every 30 s for a total of 30 min (zoomed ϫ3). B, Tracking wild-type and Gnai2Ϫ/Ϫ CD4 T at a depth between 252 and 267 ␮m. Individual wild-type and Gnai2Ϫ/Ϫ CD4 T cell tracks from 30 min of imaging. Tracks are from those 19 cells that remained within the imaging space a minimum of 6 min (the average duration of imaging of the wild-type cells was 441 s, and the Gnai2Ϫ/Ϫ CD4 T cells 465 s). C, Direct comparison of cell tracks. Ten randomly chosen wild-type (bottom panel) and Gnai2Ϫ/Ϫ (top panel) CD4 T cell tracks from cells that persisted within the imaging space for a minimum of 7 min are shown. The initial location of each cell was set to the origin of the graphs. The average Ϫ/Ϫ duration of imaging of the wild-type cells was 429 s, and the Gnai2 CD4 T cells was 441 s. http://www.jimmunol.org/ up-regulated their expression of CCR5, which led to their attrac- CD4 T cells to chemokines may explain their improper localization by the chemokines CCL3 and CCL4 produced at sites of Ag- tion and failure to normally follow the FRC network into the T cell specific dendritic cell-CD4ϩ T cell interaction (33). A more recent zone. Second, we found that the Gnai2Ϫ/Ϫ CD4 T cells failed to study has expanded this model by showing that T lymphocytes adopt the normal polarized morphology observed with wild-type migrate along fibroreticular cells (FRC) within the T cell zone CD4 T cells. This indicates an inability to respond to environmen- (34). Because FRC not only secrete CCR7 ligands, but are also tal clues, such as chemokines, that trigger a polarized phenotype. decorated with them, solid-phased chemokines may provide hap- Third, the velocity profile of the Gnai2Ϫ/Ϫ CD4 T cells was sig- by guest on September 23, 2021 totactic guidance clues and promote chemokinesis (35). The eval- nificantly shifted to the left with the average velocity of the uation of the imaging data of the Gnai2Ϫ/Ϫ CD4 cells has some Gnai2Ϫ/Ϫ T cells reduced nearly 50% compared with wild-type bearing on these issues. First, we found that the Gnai2Ϫ/Ϫ CD4 T cells. The velocities of the Gnai2Ϫ/Ϫ T cells are most likely arti- cells clustered around the HEVs and migrated poorly into the factually elevated due to cell shape change and oscillatory move- deeper cortical region. The impaired responsiveness of Gnai2Ϫ/Ϫ ments that change the center of the cell mass affecting the spot

FIGURE 5. Intravital imaging of CD4 T cells from Gnai2Ϫ/Ϫ mice reveals major defects in T cell motility. A, Imaging CD4 T cells within the inguinal lymph node. Twenty-four hours after cell transfer, inguinal lymph nodes were imaged intravitally using multiphoton microscope immediately following injection of mixture of red and green fluorescent dextrans (Gnai2Ϫ/Ϫ CD4 T cells, red; wild-type T cells, green). A 750 ␮m ϫ 750-␮m image was collected every second at 1-␮m intervals to a depth of 266 ␮m. The first image acquired is shown in the left panel, and additional images acquired at depths of 50, 100, 200, and 265 ␮m are also shown, as indicated. The red arrow identifies a small group of Gnai2Ϫ/Ϫ CD4 T present among the much more numerous wild-type T cells. B, An XZ plane projection of the imaging data. Following a 3-D reconstruction of the imaging data, an XZ projection is shown, which was compressed along 20 ␮m y-axis at the level indicated by the white line in the first panel in A. Green (top, wild-type cells), red (middle, Gnai2Ϫ/Ϫ CD4 T cells), and composite images (bottom) are shown. Because of the lower resolution in the z-axis, the cells are elongated along the z-axis. C, Decreased polarity of Gnai2Ϫ/Ϫ CD4 T. Twenty-four hours after cell transfer, inguinal lymph nodes were imaged intravitally using a multiphoton microscope (Gnai2Ϫ/Ϫ CD4 T cells, red; wild-type T cells, green). The long and short axes of 20 randomly chosen wild-type and Gnai2Ϫ/Ϫ CD4 T cells were measured, and a polarity index was calculated by dividing the long axis by the short axis. The mean and SD of the data are shown. D, Cell tracking. Individual cells were tracked for 20 min using the Surpass feature of Imaris software within a volume of 300 ␮M ϫ 300 ␮M ϫ 30 ␮M. Individual wild-type (green) and Gnai2Ϫ/Ϫ CD4 T cell tracks (red) are shown (left panel). E, Direct comparison of cell tracks. Ten randomly chosen wild-type (middle panel) and Gnai2Ϫ/Ϫ (right panel) CD4 T cell tracks from cells that persisted within the imaging space are shown. The initial location of each cell was set to the origin of the graphs. F, CD4 T cell displacements within the lymph node. The displacement of wild-type (green) and Gnai2Ϫ/Ϫ (red) CD4 T cells tracked from when they first appeared in the imaging space to when they exited or the imaging was terminated. The arrows originate where the cell was first imaged. Wild-type and Gnai2Ϫ/Ϫ CD4 T cells are shown separately (electronically zoomed ϫ5). G, A direct comparison of wild-type and Gnai2Ϫ/Ϫ CD4 T cell displacements. The displacements of 132 randomly chosen wild-type and Gnai2Ϫ/Ϫ CD4 T cells during a 20-min imaging session were sorted by length and plotted (left panel). Fifteen wild-type and Gnai2Ϫ/Ϫ CD4 T cells that remained in the tracking volume for 4 min of the 20-min imaging session were tracked. The mean squared displacement was plotted vs time (right panel). H, Comparison of straightness indices. Twelve randomly chosen tracks were sorted on the basis of their track straightness (track displacement/track length) and plotted. I, Velocity profiles of transferred CD4 T cells. Data show the frequency histograms of 3-D velocities of wild-type (Ⅺ) and Gnai2Ϫ/Ϫ (f) CD4 T cells. Results from a representative experiment are shown. Individual cells were tracked by spot analysis using Imaris software. The average speeds of the tracked cells are indicated with arrows. The mean velocity of wild-type and Gnai2Ϫ/Ϫ CD4 T cells calculated from three independent movies from distinct experiments is shown in the right panel. 448 DEFICIENCY OF G␣i2 INHIBITS CHEMOKINE RECEPTOR SIGNALING

Ϫ Ϫ Ϫ Ϫ ϩ analysis. The reduced velocities of the Gnai2 / CD4 T cells can ness of autoimmune G␣i2 / mice is an intrinsic naive CD4 T cell disorder ␣ in part be explained by their preferential localization in regions selective for the G i2 subunit. Int. Immunol. 15: 1359–1367. 10. Rudolph, U., M. J. Finegold, S. S. Rich, G. R. Harriman, Y. Srinivasan, P. Brabet, surrounding HEVs, where T cells are known to have slower mo- G. Boulay, A. Bradley, and L. Birnbaumer. 1995. Ulcerative colitis and adeno- tility. However, those Gnai2Ϫ/Ϫ CD4 T cells that managed to enter carcinoma of the colon in G␣i2-deficient mice. Nat. Genet. 10: 143–150. the deeper cortical region also showed a reduced velocity profile 11. Moratz, C., J. R. Hayman, H. Gu, and J. H. Kehrl. 2004. Abnormal B-cell re- Ϫ Ϫ sponses to chemokines, disturbed plasma cell localization, and distorted immune compared with wild-type cells. Because the Gnai2 / T cells re- tissue architecture in Rgs1Ϫ/Ϫ mice. Mol. Cell. Biol. 24: 5767–5775. tain some responsiveness to chemokines and those cells entering 12. Tang, M. L., D. A. Steeber, X. Q. Zhang, and T. F. Tedder. 1998. Intrinsic differences in L-selectin expression levels affect T and B lymphocyte subset- the T cell zone are most likely to be those cells most responsive, specific recirculation pathways. J. Immunol. 160: 5113–5121. their movement would most likely be further impaired were that 13. Matloubian, M., C. G. Lo, G. Cinamon, M. J. Lesneski, Y. Xu, V. Brinkmann, not the case. These results indicate that within the confines of the M. L. Allende, R. L. Proia, and J. G. Cyster. 2004. Lymphocyte egress from thymus and peripheral lymphoid organs is dependent on S1P receptor 1. Nature lymph node, T cells are subject to Gi␣2-dependent signals that 427: 355–360. enhance their motility and control their localization. 14. Allende, M. L., J. L. Dreier, S. Mandala, and R. L. Proia. 2004. Expression of the sphingosine 1-phosphate receptor, S1P1, on T-cells controls thymic emigration. Many questions need answering before a comprehensive under- J. Biol. Chem. 279: 15396–15401. standing of chemokine receptor signaling and its role in lympho- 15. Lo, C. G., Y. Xu, R. L. Proia, and J. G. Cyster. 2005. Cyclical modulation of cyte trafficking and motility is fully achieved. The analysis of che- sphingosine-1-phosphate receptor 1 surface expression during lymphocyte recirculation and relationship to lymphoid organ transit. J. Exp. Med. 201: 291–301. mokine receptor signaling and lymphocyte trafficking of T cells 16. Spangrude, G. J., B. A. Braaten, and R. A. Daynes. 1984. Molecular mechanisms ϩ Ϫ Ϫ Ϫ isolated from Gnai2 / and Gnai2 / mice has illustrated that the of lymphocyte extravasation. I. Studies of two selective inhibitors of lymphocyte effect of chemokines on lymphocyte trafficking cannot be under- recirculation. J. Immunol. 132: 354–362. 17. Bargatze, R. F., and E. C. Butcher. 1993. Rapid G protein-regulated activation stood solely on the basis of the pattern of cognate receptor expres- event involved in lymphocyte binding to high endothelial venules. J. Exp. Med. Downloaded from sion. Our results provide a rationale for a comprehensive analysis 178: 367–372. 18. Cyster, J. G., and C. C. Goodnow. 1995. Pertussis toxin inhibits migration of B of the effects of other signaling pathways, such as those triggered and T lymphocytes into splenic white pulp cords. J. Exp. Med. 182: 581–586. by Ag, Toll, or cytokine receptors on heterotrimeric G protein 19. Warnock, R. A., S. Askari, E. C. Butcher, and U. H. von Andrian. 1998. Mo- expression and function as well as upon the proximal regulators of lecular mechanisms of lymphocyte homing to peripheral lymph nodes. J. Exp. Med. 187: 205–216. these proteins. Finally, the analysis of the movement of those 20. Debes, G. F., C. N. Arnold, A. J. Young, S. Krautwald, M. Lipp, J. B. Hay, and Ϫ/Ϫ Gnai2 CD4 T cells that enter into the inguinal lymph node E. C. Butcher. 2005. Chemokine receptor CCR7 required for T lymphocyte exit http://www.jimmunol.org/ supports a role for local directional clues in organizing the move- from peripheral tissues. Nat. Immunol. 6: 889–894. 21. Foerster, K., F. Groner, J. Matthes, W. J. Koch, L. Birnbaumer, and S. Herzig. ment of CD4 T cells within lymph nodes. 2003. Cardioprotection specific for the G protein Gi2 in chronic adrenergic signaling through ␤2-adrenoceptors. Proc. Natl. Acad. Sci. USA 100: 14475–14480. 22. Nagata, K., C. Ye, M. Jain, D. S. Milstone, R. Liao, and R. M. Mortensen. 2000. ␣ ␣ Acknowledgments G i2 but not G i3 is required for muscarinic inhibition of contractility and cal- We thank Mary Rust for excellent editorial assistance; Dr. Lutz Birnbaumer cium currents in adult cardiomyocytes. Circ. Res. 87: 903–909. of the National Institute of Environmental Health Sciences for the 23. Molon, B., G. Gri, M. Bettella, C. Gomez-Mouton, A. Lanzavecchia, Ϫ/Ϫ A. C. Martinez, S. Manes, and A. Viola. 2005. T cell costimulation by chemokine Gnai2 mice; Dr. Owen Schwartz, Dr. Juraj Kabat, and Dr. Mewggan receptors. Nat. Immunol. 6: 465–471. Czapiga of the Research Technology Branch of the National Institute of 24. Hwang, J. I., I. D. Fraser, S. Choi, X. F. Qin, and M. I. Simon. 2004. Analysis

Allergy and Infectious Diseases for their support in the acquisition and of C5a-mediated chemotaxis by lentiviral delivery of small interfering RNA. by guest on September 23, 2021 analysis of the multiphoton microscopy data; and Dr. Anthony Fauci Proc. Natl. Acad. Sci. USA 101: 488–493. 25. Katakai, T., T. Hara, J. H. Lee, H. Gonda, M. Sugai, and A. Shimizu. 2004. A and the National Institute of Allergy and Infectious Diseases for their novel reticular stromal structure in lymph node cortex: an immuno-platform for continued support. interactions among dendritic cells, T cells and B cells. Int. Immunol. 16: 1133–1142. 26. Nobles, M., A. Benians, and A. Tinker. 2005. Heterotrimeric G proteins pre- Disclosures couple with G protein-coupled receptors in living cells. Proc. Natl. Acad. Sci. The authors have no financial conflict of interest. USA 102: 18706–18711. 27. Huang, J. S., L. Dong, and G. C. Le Breton. 2006. Mass-dependent signaling between G protein coupled receptors. Cell. Signal. 18: 564–576. References 28. Miller, M. J., S. H. Wei, I. Parker, and M. D. Cahalan. 2002. Two-photon im- 1. Springer, T. A. 1994. Traffic signals for lymphocyte recirculation and leukocyte aging of lymphocyte motility and antigen response in intact lymph node. Science emigration: the multistep paradigm. Cell 76: 301–314. 296: 1869–1873. 2. Kunkel, E. J., and E. C. Butcher. 2002. Chemokines and the tissue-specific mi- 29. Miller, M. J., S. H. Wei, M. D. Cahalan, and I. Parker. 2003. Autonomous T cell gration of lymphocytes. Immunity 16: 1–4. trafficking examined in vivo with intravital two-photon microscopy. Proc. Natl. 3. Arai, H., C. L. Tsou, and I. F. Charo. 1997. Chemotaxis in a lymphocyte cell line Acad. Sci. USA 100: 2604–2609. transfected with C-C chemokine receptor 2B: evidence that directed migration is 30. Wei, S. H., I. Parker, M. J. Miller, and M. D. Cahalan. 2003. A stochastic view mediated by ␤␥ dimers released by activation of G␣i-coupled receptors. Proc. of lymphocyte motility and trafficking within the lymph node. Immunol. Rev. 195: Natl. Acad. Sci. USA 94: 14495–14499. 136–159. 4. Neptune, E. R., and H. R. Bourne. 1997. Receptors induce chemotaxis by re- 31. Halin, C., J. Rodrigo Mora, C. Sumen, and U. H. von Andrian. 2005. In vivo leasing the ␤␥ subunit of Gi, not by activating Gq or Gs. Proc. Natl. Acad. Sci. imaging of lymphocyte trafficking. Annu. Rev. Cell. Dev. Biol. 21: 581–603. USA 94: 14489–14494. 32. Germain, R. N., F. Castellino, M. Chieppa, J. G. Egen, A. Y. Huang, L. Y. Koo, 5. Hepler, J. R., and A. G. Gilman. 1992. G proteins. Trends Biochem. Sci. 17: and H. Qi. 2005. An extended vision for dynamic high-resolution intravital im- 383–387. mune imaging. Semin. Immunol. 17: 431–441. 6. Neer, E. J. 1995. Heterotrimeric G proteins: organizers of transmembrane signals. 33. Castellino, F., A. Y. Huang, G. Altan-Bonnet, S. Stoll, C. Scheinecker, and Cell 80: 249–257. R. N. Germain. 2006. Chemokines enhance immunity by guiding naive CD8ϩ T 7. Neptune, E. R., T. Iiri, and H. R. Bourne. 1999. G␣i is not required for chemo- cells to sites of CD4ϩ T cell-dendritic cell interaction. Nature 440: 890–895. taxis mediated by Gi-coupled receptors. J. Biol. Chem. 274: 2824–2828. 34. Bajenoff, M., J. G. Egen, L. Y. Koo, J. P. Laugier, F. Brau, N. Glaichenhaus, and 8. Han, S. B., C. Moratz, N. N. Huang, B. Kelsall, H. Cho, C. S. Shi, O. Schwartz, R. N. Germain. 2006. Stromal cell networks regulate lymphocyte entry, migra- and J. H. Kehrl. 2005. Rgs1 and Gnai2 regulate the entrance of B lymphocytes tion, and territoriality in lymph nodes. Immunity 25: 989–1001. into lymph nodes and B cell motility within lymph node follicles. Immunity 22: 35. Luther, S. A., H. L. Tang, P. L. Hyman, A. G. Farr, and J. G. Cyster. 2000. 343–354. Coexpression of the chemokines ELC and SLC by T zone stromal cells and 9. Huang, T. T., Y. Zong, H. Dalwadi, C. Chung, M. C. Miceli, K. Spicher, deletion of the ELC gene in the plt/plt mouse. Proc. Natl. Acad. Sci. USA 97: L. Birnbaumer, J. Braun, and R. Aranda. 2003. TCR-mediated hyper-responsive- 12694–12699.