Epithelial Cells and Provides a Regulatory Link to Lung Ephrin-A1

Home , EPHB6, Ephrin, Ephrin A1, Ephrin A4, Ephrin receptor, Receptor tyrosine kinase

Ephrin-A1 Suppresses Th2 Cell Activation and Provides a Regulatory Link to Lung Epithelial Cells

This information is current as Jan G. Wohlfahrt, Christian Karagiannidis, Steffen of September 27, 2021. Kunzmann, Michelle M. Epstein, Werner Kempf, Kurt Blaser and Carsten B. Schmidt-Weber J Immunol 2004; 172:843-850; ; doi: 10.4049/jimmunol.172.2.843

http://www.jimmunol.org/content/172/2/843 Downloaded from

References This article cites 42 articles, 15 of which you can access for free at: http://www.jimmunol.org/content/172/2/843.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

• Fast Publication! 4 weeks from acceptance to publication by guest on September 27, 2021

*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 © 2004 by The American Association of Immunologists All rights reserved. Print ISSN: 0022-1767 Online ISSN: 1550-6606. The Journal of Immunology

Ephrin-A1 Suppresses Th2 Cell Activation and Provides a Regulatory Link to Lung Epithelial Cells1

Jan G. Wohlfahrt,* Christian Karagiannidis,* Steffen Kunzmann,* Michelle M. Epstein,† Werner Kempf,‡ Kurt Blaser,* and Carsten B. Schmidt-Weber2*

Gene expression screening showed decreased ephrin-A1 expression in CD4؉ T cells of asthma patients. Ephrin-A1 is the ligand of the Eph receptor family of tyrosine kinases, forming the largest family of receptor tyrosine kinases. Their immune regulatory properties are largely unknown. This study demonstrates significantly reduced ephrin-A1 expression in T cells of asthma patients using real time-PCR. Immunohistological analyses revealed strong ephrin-A1 expression in lung tissue and low expression in cortical areas of lymph nodes. It is absent in T cell/B cell areas of the spleen. Colocalization of ephrin-A1 and its receptors was found only in the lung, but not in lymphoid tissues. In vitro activation of T cells reduced ephrin-A1 at mRNA and protein levels. T cell proliferation, activation-induced, and IL-2-dependent cell death were inhibited by cross-linking ephrin-A1, and not by Downloaded from engagement of Eph receptors. However, anti-EphA1 receptor slightly enhances Ag-specific and polyclonal proliferation of PBMC cultures. Furthermore, activation-induced CD25 up-regulation was diminished by ephrin-A1 engagement. Ephrin-A1 engagement reduced IL-2 expression by 82% and IL-4 reduced it by 69%; the IFN-␥ expression remained unaffected. These results demonstrate that ephrin-A1 suppresses T cell activation and Th2 cytokine expression, while preventing activation-induced cell death. The reduced ephrin-A1 expression in asthma patients may reflect the increased frequency of activated T cells in peripheral blood. That http://www.jimmunol.org/ the natural ligands of ephrin-A1 are most abundantly expressed in the lung may be relevant for Th2 cell regulation in asthma and Th2 cell generation by mucosal allergens. The Journal of Immunology, 2004, 172: 843–850.

NA array-based screening of T cell cDNA of asthmatic group can bind several ephrins beside EphA1, which can bind only patients revealed reduced ephrin-A1 expression com- ephrin-A1 (2, 7). The ligands of the Eph receptor family have to be D pared with those of healthy donors (1). This ﬁnding is membrane bound to be active. Therefore, cell-cell interaction is interesting because ephrin-A1 expression was previously not re- necessary for Eph receptor-ephrin signaling, and soluble ephrins ϩ ported in peripheral CD4 T cells and in addition Eph receptors are active only if they are artiﬁcially clustered (9Ð11).

are expressed on epithelial cells, which are important cells for the Characteristic for the ephrin biology is that not only is a forward by guest on September 27, 2021 lung function. Therefore, ephrins may represent a regulatory in- signaling pathway by the Eph receptors induced upon ligand bind- terface of epithelial surfaces in the lung and the immune system. ing but also the ligand itself transmits a reverse signals into the cell The ephrins are the ligands of the Eph receptors, which form (4). Interestingly, even the GPI-anchored ephrin-A ligands can in- with at least 14 members the largest group of tyrosine-kinase re- duce signals due to lipid raft microdomain-associated uncharac- ceptors (2Ð4). Ephrins are divided into two subclasses. The A terized receptors, which are specifically phosphorylated after eph- subclass (ephrin-A1ÐA6) consists of proteins tethered to the cell rin engagement (12Ð15). Involvement of Eph receptors and ephrins membrane via a GPI anchor, whereas the B subclass (ephrin-B1Ð is currently discussed in developmental processes (16Ð19), as well B3) consist of a transmembrane domain and a short cytoplasmic as their role in carcinogenesis (20). region (5, 6). Similarly, the Eph receptors are divided in an A Human ephrin-A1 is a 25-kDa glycoprotein with high homology (EphA1Ð9) and a B group (EphB1Ð6), dependent on the group of to murine and rat orthologs (21) and is abundantly expressed in ephrins they are binding. All Eph receptors of the A subclass can lung, but is also found in kidney, salivary gland, skin, and intestine bind ephrin-A1 (7), but the receptor EphA2 binds ephrin-A1 with (21, 22). It was found as an immediate-early response gene of ϭ the highest affinity (Kd 25 nM (8)). All receptors of the EphA endothelium induced by TNF-␣ (23) that possesses angiogenic and endothelial cell chemoattractant activity (24). EphA1 receptor is the least conserved member of the EphA receptor family and is * Swiss Institute of Allergy and Asthma Research, Davos, Switzerland; † Division of expressed in high levels in the liver, kidney, thymus, and lung but Immunology, Allergy and Infectious Disease, Department of Dermatology, Univer- sity of Vienna, Vienna, Austria; and ‡ Department of Dermatology, University Hos- not in the brain (25, 26). Eph receptors are also transcribed at high pital Zurich, Zurich, Switzerland. levels in tumor cells of epithelial origin (27, 28) and overproduc- Received for publication May 28, 2003. Accepted for publication November 4, 2003. tion of EphA1 induces tumorigenic ability of normal fibroblasts or The costs of publication of this article were defrayed in part by the payment of page melanoma progression (27, 29). This could support a role for charges. This article must therefore be hereby marked advertisement in accordance ephrins in the growth of the cells, but it could also indicate that the with 18 U.S.C. Section 1734 solely to indicate this fact. immune system recognizes the ephrin-overexpressing cells to a 1 This work was supported by Swiss National Foundation Grants 31.52986.97 and 31-65436.01, Roche Research Foundation Grant Mkl/stm 115-2001, the Gebert Ru¬f reduced degree. In fact, recent studies showed that immuno-com- Foundation Grant G-074/99, EMDO Foundation, Saurer Foundation, Zurich and OPO petent cells express Eph receptors and ephrin ligands. A soluble Foundation Zurich, Deutsche Forschungsgemeinschaft Grant KU 1403/1-1 (to splice variant of ephrin-A4 is secreted by activated human B lym- Z.S.K.). phocytes, and EphA2 receptors are present on dendritic cells/mac- 2 Address correspondence and reprint requests to Carsten B. Schmidt-Weber, Swiss Institute of Allergy and Asthma Research, Obere Strasse 22, CH-7270 Davos, Swit- rophages in tonsils (30) and peripheral blood (31). TCR engage- zerland. E-mail address: [email protected] ment along with EphB6 receptor cross-linking in the Jurkat cell

line results in FAS-mediated cell death (32). EphB6 receptor over- manufacturer. After hybridization, membranes were washed for 20 min with expression in this cell line suppresses TCR-mediated IL2 secretion 30 ml of a 0.1ϫ standard saline-citrate-phosphate/EDTA, 1% (w/v) SDS so- and CD25 expression by inhibition of the c-jun kinase pathway lution. Subsequently, moistened arrays were sealed in plastic bags and exposed for 24 h on phosphoimager screens (Fuji Film, Tokyo, Japan). The screens (33). Furthermore, plate-bound ephrin-A1 inhibits strong chemo- were analyzed using the FLA 3000 phosphoimager system (Fuji) and evalu- taxis in thymocytes and Jurkat cell line (34). These studies clearly ated with the AIDA program (Raytest, Urdorf, Switzerland). demonstrate that ephrins participate in immune regulation and possibly convey an off signal to the immune system. T cell proliferation assay In this study, we describe ephrin-A1 expression by T cells and Triplicates of 1 ϫ 105 CD4ϩ T cells or PBMC were set up in 200 ␮lof the impact of ephrin-A1 cross-linking on T cell function. The RPMI 1640 in 96-well flat-bottom microtiter plates (Costar, Corning, NY). strong localized expression of Eph receptors on lung epithelia sug- Cells were activated by plate-bound mouse anti-human CD3 mAb (1 ␮g/ ml; clone CRL 8001; ATCC) and mouse anti-human CD28 mAb (2 ␮g/ml; gests that ephrins create microenvironmental conditions, which are clone 15E8; Netherlands Red Cross blood transfusion service). The bot- known to contribute to the pathogenesis of asthma such as hyper- toms of the wells were as well covered with rabbit Ig Fraction (DAKO- reactivity of the lung and airway remodeling. Cytomation, Zug, Switzerland), rabbit anti-human ephrin-A1 (Santa Cruz Biotechnology, Santa Cruz, CA), rabbit anti-human EphA1 receptor (R&D Materials and Methods Systems), or rabbit anti-human EphA2 receptor Abs (Sigma-Aldrich, Buchs, Switzerland). The concentration of the plate-bound Abs was 4 or 1 Patients ␮g/ml for soluble addition of Abs to the cultures. One triplicate of cells was All asthmatic patients met the American Thoracic Society definitions of not activated to calculate the stimulation index for the activated cells. After asthma (35). At the time of the study, allergic asthma (AA)3 patients had an incubation time of 1, 2, 3, or 6 days, cells were pulsed for 16 h with 1 ␮ 3 stable, severe bronchial asthma; were not in acute respiratory distress; had Ci [ H]thymidine (Hartmann) and harvested on glass fiber filters using an Downloaded from no evidence of chest infection; and had never had atopic dermatitus (AD); automated multisample harvester (Tomtec, Hamden, CT). Filters were but were all characterized by an atopic predisposition. transferred in sample bags with liquid scintillation fluid and analyzed using a ␤-scintillation counter (Pharmacia-Wallac, Turku, Finland). Isolation of T cells Quantitative real time PCR PBMC were isolated from blood of nonsmoking healthy donors by Ficoll ϩ (Biochrom, Berlin, Germany) density gradient centrifugation. The inter- Total RNA was isolated from 3 ϫ 106 CD4 T lymphocytes using the ϩ phase cells were washed three times, and CD4 T cells were purified using RNEasy minikit (Qiagen) according to the manufacturer’s protocol. The http://www.jimmunol.org/ anti-CD4-Dynal magnetic beads and Detach-a-Bead Abs (both Dynal, RNA was eluted in 50 ␮l of water. Approximately 4 ␮g of total RNA (10 Hamburg, Germany) as previously described (36). The purity of CD4ϩ T ␮l) were reverse transcribed using the TaqMan Reverse Transcription Re- cells was initially tested by flow cytometry and was Ͼ95%. CD19ϩ B agents kit (Roche/Applied Biosystems, Rotkreuz, Switzerland) with ran- lymphocytes were isolated with CD19 MACS MicroBeads (Miltenyi Bio- dom hexamers as primer following the recommendation of the supplier in tec, Bergisch Gladbach, Germany) following the manufacturer’s protocol. a total volume of 30 ␮l. The PCR primers and probes detecting ephrin-A1, The viability of the T cells was verified after isolation and was Ն99%. EphA1 receptor, and IL-2 were designed based on the sequences reported in GenBank with the Primer Express software version 1.2 (Perkin Elmer/ Cell cultures Applied Biosystems, Foster City, CA) as follows: ephrin-A1 forward ϩ primer 5Ј-CCA TAC ATG TGC AGC TGA ATG AC-3Ј; ephrin-A1 re- All cell cultures with human CD4 T cells were grown in RPMI 1640 verse primer 5Ј-CAG CGT CTG CCA CAG AGT GA-3Ј; ephrin-A1 probe

supplemented with 2 mM L by guest on September 27, 2021 -glutamine, 1 mM sodium pyruvate, 1% MEM FAM 5Ј-ACG TGG ACA TCA TCT GTC CGC ACT ATG AA-3Ј nonessential amino acids and vitamins, 100 U/ml penicillin, 100 ␮g/ml Ј Ј ␮ TAMRA; ephA1 forward primer 5 -TGC TGG ATC CCC CAA AAG-3 ; streptomycin, 50 M 2-ME (all from Life Technologies, Basel, Switzer- EphA1 reverse primer 5Ј-TTG GGC AGT CCT GGT ACA TG-3Ј; EphA1 land) and 10% FBS (Sera-Lab, Loughborough, U.K.) in a humidified at- probe FAM 5Ј-TGG GTG GAG TGA ACA GCA ACA GAT AC-3Ј mosphere containing 5% CO at 37¡C. 2 TAMRA; IL-2 forward primer 5Ј-TCA CCA GGA TGC TCA CAT TTA Ј Ј DNA-array hybridization AGT-3 ; IL-2 reverse primer 5 -GAG GTT TGA GTT CTT CTT CTA GAC ACT GA-3Ј; IL-2 probe FAM 5Ј-ATG CCC AAG AAG GCC ACA CD4ϩ T cells were isolated out of 80 ml of peripheral blood from three GAA CTG AAA C-3Ј TAMRA. For EphA2, IL-4, ␤-glucuronidase (GUS), patients with allergic asthma and three healthy persons. For RNA isolation, and 18S rRNA, commercially available primers and probes were purchased 5 ϫ 106 purified CD4ϩ T cells were lysed with RNEasy lysis buffer (Qia- (Applied Biosystems). The TaqMan probes of the target genes were labeled gen, Hamburg, Germany) after overnight stimulation using plate-bound with FAM as the reporter dye and TAMRA as the fluorescent quencher. anti-CD3 (OKT3, 1 ␮g/ml) and anti-CD28 (15E8; Netherlands Red Cross The probes for the housekeeping genes were labeled with VIC (Applied blood transfusion service, Amsterdam, The Netherlands; 2 ␮g/ml). The Biosystems) as the reporter dye and TAMRA as fluorescent quencher. OKT3 hybridoma was purchased from American Type Culture Collection When possible, primers were spanning exon-intron borders to eliminate (ATCC, Manassas, VA), and supernatants were purified using affinity chro- amplification of contaminations of genomic DNA. Accumulation of the matography. RNA was isolated using the RNEasy Mini Kit (Qiagen) and PCR products was detected in real time by monitoring the probe cleavage- ␮ eluted in 100 lofH2O. The RNA was precipitated with sodium acetate induced mobilization of the reporter dye. The prepared cDNAs were am- ␮ and ethanol, washed with 70% ethanol, and resuspended in 11 lofH2O. plified using an UNG-containing PCR mastermix (Perkin Elmer/Applied The RNA was mixed with 4 ␮l of human array-specific primers (R&D Biosystems) according to the recommendations of the manufacturer in a Systems, Abingdon, U.K.) and incubated in a thermal cycler at 90¡C for 2 total volume of 25 ␮l in an ABI PRISM 7000 Sequence Detection System min. The temperature was ramped to 42¡C in 20 min. For reverse tran- (Applied Biosystems). To determine a correct housekeeping gene to com- scription, the following components were added: 6 ␮lof5ϫ reverse first- pare differently activated CD4ϩ T lymphocytes, a TaqMan Human Endog- strand buffer (Life Technologies, Basel, Switzerland); 2 ␮lof0.1MDTT enous Control Plate (Applied Biosystems) was used. Both GUS and 18S (Life Technologies); 20 U RNase inhibitor (Roche, Mannheim, Germany); rRNA did not show significant differences between cDNA content of rest- 1 ␮l of dCTP, dGTP, dTTP (each 10 mM; PerkinElmer), 5 ␮lof ing and activated CD4ϩ T cells. Amplification with serial dilutions of [␣-32P]dATP (0.4 MBq/␮l; Hartmann, Braunschweig, Germany); and 1 ␮l cDNA revealed that the used assay reflects the cDNA concentration in a of Superscript II RNase H-reverse transcriptase (Life Technologies). The linear manner (Fig. 3C) and that these efficiencies of the target and the ␮ mixture was incubated at 42¡C for 3 h and filled up to 100 l with H2O. housekeeping genes are in the same range. The cDNA was purified with the QIAquick Nucleotide Removal Kit (Qiagen) Relative quantification and calculation of the confidential range was ␮ ␮ ⌬⌬ and eluted in 100 lofH2O. An aliquot of 1 l was measured in a scintillation performed using the comparative Ct method as described (37, 38). All counter to estimate the percentage incorporation of labeled nucleotides in the amplifications were conducted at least in duplicates. cDNA. CD4ϩ T cell cDNA probes prepared from patients’ blood were hy- bridized in parallel with cDNA probes from healthy donors on human cytokine Immunoblotting expression array membrane (R&D Systems) following the instructions of the CD4ϩ T cells or CD19ϩ B cells 2 ϫ 106 were lysed in Laemmli buffer (125 mM Tris HCl (pH 6.8), 4% SDS, 20% glycerol, 2.5% 2-ME, bromphenol 3 Abbreviations used in this paper: AA, allergic asthma; GUS, ␤-glucuronidase; TBS, blue) and homogenized with a Qiashredder (Qiagen). Lysates were heated Tris-buffered saline; AICD, activation-induced cell death. for 10 min to 75¡C and loaded next to a Benchmark protein-mass ladder The Journal of Immunology 845

(Invitrogen, Basel, Switzerland) on a NuPAGE 4Ð12% bis-tris gel (In- vitrogen) and run until proteins were sufficiently separated. The proteins were electroblotted onto a nitrocellulose membrane (Amersham Life Sci- ence, Du¬bendorf, Switzerland). The membrane was blocked overnight with TBS containing 5% BSA fraction V (Sigma-Aldrich), washed three times for 5 min, and incubated with 1 ␮g/ml rabbit anti-human ephrin-A1 Ab (Santa Cruz Biotechnology) in TBS buffer with 5% BSA fraction V for 1 h. The blot was washed three times with TBS after incubation with HRP- labeled donkey anti-rabbit Ig (Amersham Life Science) in a dilution of 1/50,000 in TBS containing 5% BSA fraction V for 1 h. After three wash- ings, the blot was developed using the ECL Plus System (Amersham Life Science) following the recommendation of the supplier. Flow cytometry After stimulation with plate-bound anti-CD3 Ab and anti-CD28 Ab for 1 and 16 h, 1 ϫ 106 unstimulated and stimulated CD4ϩ T cells were incubated for 45 min with 1 ␮g of a rabbit anti-human ephrin-A1 Ab (Santa FIGURE 1. Ephrin-A1 mRNA and protein expression of CD4ϩ T cells. Cruz Biotechnology). After a washing, the first Ab was stained with 5 ␮g A, Hybridization of radioactive cDNA generated from CD4ϩ T cells from Ј of a FITC-conjugated F(ab )2 fragment of a swine anti-rabbit Ab (DAKO). healthy donors and asthma patients in duplicates. L19 was used as house- Cells were fixed with 2% paraformaldehyde and subjected to FACS (EP- keeping gene; the experiment represents the results of three independent ICS XL-MCL, Beckman Coulter, Miami, FL). CD25 expression was mea- experiments. B, Quantitative real time PCR analysis of ephrin-A1 expres-

sured accordingly, using an RD1-labeled anti-CD25 Ab (Beckman Downloaded from ϩ ■ Coulter). Viabilities of cell populations were measured by FACS using 25 sion of cDNA generated from CD4 T cells from healthy donors ( ) and Œ ⌬⌬ ␮M ethidium bromide. asthma patients ( ). Units were calculated by the Ct method and nor- (Statistical significance ( p Յ 0.05 ,ء .malized to a value of a healthy donor Allergic asthma induction in mice based on the nonparametric Mann-Whitney U test. C, Western blot analysis ϩ ϩ BALB/c mice were immunized i.p. with 10 ␮g of OVA (Grade V; Sigma- of whole cell lysates of purified CD4 T cells and CD19 B cells kD, Aldrich, St. Louis, MO) in 200 ␮l of PBS or with PBS alone and on days kilodaltons. The image is representative of three independent experiments. 0 and 21, and 1 wk later on days 28 and 29 they were nebulized with 1% http://www.jimmunol.org/ OVA in PBS using an ultrasonic nebulizer for 60 min twice daily, as previously described (39). Mice were then rechallenged a second time with 2 ϫ 3 min of incubation in 100, 95, 90, 80, and 70% ethanol. Slides were aerosolized 1% OVA or PBS for 60 min twice daily on 2 consecutive days heated for 5 min to 90¡C in 10 mM sodium citrate buffer (pH 6.0). Cold on days 89 and 90. After 48 h, lungs were perfused with PBS and fixed slides were incubated in peroxidase block buffer of the DAKO EnVision ϩ with 4% formalin. Memory, recovered, and control groups are referred to System Peroxidase (3-amino-9-ethylcarbazole) for mice (DAKOCytoma- as OVA-OVA-OVA (OVA -OVA -OVA ), OVA-OVA-PBS i.p. aerosol aerosol tion) for 5 min and washed with TBST (0.05 M Tris, 0.3 M NaCl, 0.1 M (OVA -OVA -PBS ), OVA-PBS-OVA (OVA -PBS - i.p. aerosol aerosol i.p. aerosol Tween 20, pH 7.4). Slides were pre-incubated with 25% FCS, 0.1% BSA OVA ), and PBS-PBS-PBS (PBS -PBS -PBS ). aerosol i.p. aerosol aerosol in TBST for 20 min and washed; anti-ephrin-A1, anti-EphA1, and anti- Immunochemical staining EphA2 (Santa Cruz Biotechnology) or rabbit Ig fraction (DAKOCytoma- tion) as isotype control were added at a concentration of 1 ␮g/ml in block- by guest on September 27, 2021 Paraformaldehyde-fixed sections were incubated overnight at 55¡C. Sec- ing buffer for an incubation period of 16 h at 4¡C. Biotinylated rabbit tions were deparaffinized by 2 ϫ 5 min of incubation in xylene followed by anti-mouse IgG and HRP-conjugated streptavidin (DAKOCytomation) was

FIGURE 2. Immunohistological staining of ephrin-A1 and EphA1 receptor. A, Anti-ephrin-A1 staining of a murine lung. The isotype control for B–D is shown in A. Black ar- rows, Ephrin-A1 expression (B)on epithelial cells. Lung sections of OVA-sensitized mice that had been treated with aerosolized PBS (C)or OVA in PBS (D). Arrow in C, EphA1 receptor expression by epithelial cells in bronchi. The black arrow in D highlights epithelial cell thickening, and the blue arrow perivascular, peribronchial inﬁl- trates. Values are representative of at least three independent experiments. 846 EPHRIN-A1 INHIBITS Th2 CELLS

FIGURE 3. Real time RT-PCR analysis of ephrin-A1 and GUS. The ﬂuorescent (passive dye)- normalized reporter signal (⌬Rn) is shown against the number of PCR cycles (A). Bars, SEM of triplicate samples. B, Decrease of ephrin-A1 expression relative to housekeeping gene expression Downloaded from (GUS). The percentage was calculated on the basis ⌬⌬ of the Ct method. Bars, 95% conﬁdence interval calculated on the basis of deviation of GUS and ephrin-A1 expression. The experiment is representative of three independent experiments. C, Surface expression of ephrin-A1 on human CD4ϩ T cells stimulated for 0, 1, and 16 h on plate-immobilized http://www.jimmunol.org/ anti-CD3 and anti-CD28. Gray histogram, ephrin-A1 staining; clear histogram, isotype control. The experiment is representative of three independent experiments. by guest on September 27, 2021

used for detection, and AEC--substrate (DAKOCytomation) was used as the ephrin-A1 level in the cells of the AA patients was 2.0-fold (Ϯ substrate as described by the manufacturer. All sections were counter- 0.65; p Յ 0.05) decreased (Fig. 1B). To show the expression of stained with hematoxylin and analyzed using an Axiovert microscope and ephrin-A1 in CD4ϩ T cells on the protein level, lysates of CD4ϩ an AxioCam digital camera (both Zeiss, Go¬ttingen, Germany). T cells and CD19ϩ B cells were immunoblotted. In both cases, a Statistical analysis single band at 21 kDa was visible. Ephrin-A1 is expressed in ϩ ϩ The data were tested for significance using the Mann-Whitney U test. Dif- higher amounts in CD4 T lymphocytes than in CD19 B lym- ferences of p Յ 0.05 were considered to be significant. phocytes (Fig. 1C). The EphA1 and EphA2 receptors were expressed only in background levels as mRNA and as protein (data Results not shown). Expression of ephrin-A1 in T cells Screening for gene expression changes of 475 different genes in Ephrin-A1 and Eph-A1 expression in vivo cDNA libraries of activated CD4ϩ T cells of three AA patients Ephrin-A1 expression was analyzed in lymphoid organs of NMRI compared with three healthy persons using the cDNA array tech- mice. Ephrin-A1 expression was low in spleen (Fig. 2A) and was nology showed a decreased expression of the mRNA of ephrin-A1. predominantly observed in the marginal zone areas around the The intensity of the array spots were 1.9-fold lower (Ϯ0.28) in the lymphoid follicles. In lymph nodes, ephrin-A1 was detected pre- AA patients than in the healthy counterparts (Fig. 1A). We con- dominantly in cortical areas. The EphA2 receptor was expressed at firmed this result using the quantitative real time RT-PCR method. low levels in the red pulp of the spleen (data not shown) but was Comparing the mRNA levels of peripheral CD4ϩ T cells of five totally absent in the T or B cell-rich white pulp of the spleen. other patients with severe AA and of five healthy volunteers, again Neither EphA2 nor EphA1 receptors were visible in inguinal and The Journal of Immunology 847

FIGURE 4. A,[3H]Thymidine incorporation of CD4ϩ T cells after 3-day culture on plate-immobilized anti-CD3 and CD28 along with Abs as indicated. The stimulation index was calculated by the division with the respective medium control. Bars, SEM of six statistical signiﬁcance ,ء ;independent experiments ( p Յ 0.001). B, T cells were stimulated on plate-im- Downloaded from mobilized anti-CD3 and CD28 along with Abs as indicated and rested for 10 days. The restimulation was performed as with anti-CD3/CD28 without any anti- ephrin Abs; ■, mean of three independent experiments (n ϭ 3). C, CD25 (IL-2R␣) surface expression of anti-ephrin-A1-treated cells. Cells were treated as

described for A. Gray histogram, CD25 staining; clear http://www.jimmunol.org/ histogram, isotype control. The histograms are representative of three independent experiments. by guest on September 27, 2021

mesenteric lymph nodes, whereas ephrin-A1 expression was low isolated mRNA reveals that ephrin-A1 expression of CD4ϩ T cells and predominantly in cortical areas of the lymph nodes (data not stimulated for 1 h was 2.0-fold (Ϯ 0.6) and 3.7-fold (Ϯ 0.8) re- shown). In contrast, ephrin-A1 expression was frequently observed duced after 16 h of stimulation with plate-immobilized anti-CD3 in bronchi (Fig. 2B). Furthermore, strong EphA1 receptor expres- and anti-CD28 mAb (Fig. 3, A and B). The real time-based quan- sion was observed on any epithelial cell of small and large bronchi. tifications were performed in the linear range of ephrin-A1, and all This expression pattern was also observed in OVA-sensitized mice other genes were quantified in the present study (data not shown). 90 days after OVA aerosol challenge. Inflammatory infiltrates can Flow cytometric analysis clearly demonstrates ephrin-A1 expres- be observed, containing scattered and weak EphA1 receptor-ex- sion on 9.52 Ϯ 0.12% CD4ϩ T cells (Fig. 3C). However, after pressing cells. Epithelial cells appear thickened and slightly lower stimulation of T cells, ephrin-A1 disappears from the cell surface in EphA1 receptor expression (Fig. 2D). Similarly, EphA1 recep- within 1 h and remained absent (16 h; Fig. 3C). tor and ephrin-A1 expression was found on epithelial cells of human asthmatic lung sections. Ephrin-A1 was also found on scat- Ephrin-mediated inhibition of T cell proliferation tered subepithelial cells. The expression results demonstrated that ephrin-A1 is activation dependently expressed and thus may also be itself functionally Regulation of ephrin-A1 in T cells involved in the regulation of the cellular activation process. Pro- Because DNA array comparison showed a decreased ephrin-A1 liferation assays of activated CD4ϩ T cells demonstrate the role of expression in cell populations of CD4ϩ T cells of the peripheral the expression of ephrin ligands in the regulation of lymphocytes. blood of AA patients compared with those of healthy persons, we Plate-bound anti-ephrin-A1 Ab provoked a significant decrease of asked whether observed changes are due to a disease-mediated anti-CD3/28-induced T cell proliferation 3 days after T cell acti- activation of T cells and whether the level of ephrin-A1 expression vation. The proliferation index was significantly lower ( p Յ 0.001; is changed after engagement of the TCR. The expression of eph- 33.5 Ϯ 9.6-fold) compared with T cells stimulated in the presence rin-A1 is reduced in CD4ϩ T cells after activation on both the of the isotype control. In contrast, anti-EphA1 receptor Ab and mRNA and the protein level. Quantitative real time RT-PCR of anti-EphA2 receptor Ab (Fig. 4A) induced only a minor decrease. 848 EPHRIN-A1 INHIBITS Th2 CELLS Downloaded from

FIGURE 5. [3H]Thymidine incorporation of T cells treated with autologous APC (irradiated CD4Ϫ fraction of PBMC) (A) or unfractionated PBMC (B). The T cells and/or the APC were pretreated with anti-EphA receptor Abs or matched Ig control and subsequently cultured in the presence of recombinant DerPI (100 nM, except the first two columns) for 5 days. Alternatively (B), anti-EphA1 receptor Abs were added in a soluble form to PBMC, cultured in the absence or presence of plate-immobilized anti-CD3 and anti-CD28 mab (1 and 2 ␮g/ml respectively). Bars, SEM of three independent experiments. http://www.jimmunol.org/ The reduced proliferation was visible at days 5 and 7 after stim- containing APC. To separate effects of anti-EphA1 on T cells vs ulation (data not shown). Anti-ephrin-A1 inhibited similarly effec- APC, we pretreated irradiated CD4Ϫ PBMC or CD4ϩ T cells with tive the PHA-stimulated proliferation of T cells as TCR-specific anti-EphA1 receptor Abs. Although the intensity and/or frequency stimulation (data not shown). Proliferation assays performed with of EphA1 is very low (31), we observed a slightly, but consistently CD4ϩ T cells isolated from asthmatic donors were not different from enhanced proliferation in Ag-specific (Fig. 5A) and polyclonal T those of healthy donors. The inhibition of proliferation was dose de- cell responses (Fig. 5B). The effect on the Ag-specific response pendent in a range of 0.04 to 40 ␮g (data not shown). Restimulated, was observed only among anti-EphA1 receptor APC, whereas anti- anti-ephrin-A1-treated cells did not show anergy (Fig. 4B). Anti-eph- EphA1-pretreated T cells proliferated normally. rin-A1 treatment of the cells did not affect the viability of the unac- by guest on September 27, 2021 Ͼ Ϯ Incubation with anti-ephrin-A1 Abs reduces activation-induced tivated T cells and was always 87.2 0.3% after 3 days (Fig. 6A) ϩ and 84.8 Ϯ 0.9% after 5 days (data not shown), as tested by ethidium cell death (AICD) of activated CD4 T cells bromide exclusion. The dramatic effect of anti-ephrin-A1 also effec- The pronounced effect on IL-2 cytokine and receptor expression is tively reduced (7.3 Ϯ 2.3-fold) the surface expression of the early likely to affect AICD, which is known to be IL-2 dependent (40). activation marker CD25 (Fig. 4C). We therefore induced AICD with plate immobilized anti-CD3 mAb in the presence and absence of anti-ephrin-A1 Abs. In fact, Incubation with anti-EphA1 receptor Abs anti-ephrin-A1 prohibited AICD to large extent, whereas it did not To verify that not only Ab-mediated cross-linking of ephrin-A1 induce cell death in resting T cells (Fig. 6A). However, addition of affects T cell proliferation, but also the natural ligand-receptor in- exogenous IL-2 rendered ephrin-A1 engagement ineffective to interaction, we blocked the EphA1 receptor with Abs in cultures duce AICD (Fig. 6B). Ephrin-A1 engagement reduces the expression of IL-2 and IL-4 but not IFN-␥ in activated T cells Because an increased production of IL-2 is typical for activated T cells, we analyzed the effect of an incubation of anti-ephrin-A1 Ab on the expression of IL-2 mRNA of activated CD4ϩ T cells using the quantitative real time RT-PCR method. After 3 days, the incubation of activated CD4ϩ T cells with anti-ephrin-A1 Ab led to a 21.3 Ϯ 2.4-fold decrease of IL-2 mRNA (Fig. 7, A and B) and a 12.9 Ϯ 3.2-fold decrease of IL-4 (Fig. 7, C and D) compared with activated T cells without this Ab. In contrast, IFN-␥ expression was not affected (Ϯ 0.2; data not shown). The results were con- firmed by ELISA; however, total amounts of IL-4 were very low. Discussion Activated CD4ϩ T cells from asthma patients have decreased ex- FIGURE 6. Viability of CD4ϩ T cells quantified by flow cytometric pression of ephrin cDNA compared with healthy controls (41, 42). analysis of ethidium bromide exclusion. Cell culture conditions are iden- The present findings confirm reduced ephrin-A1 expression in pa- tical with those described in Fig. 4. The experiment is representative of at tients suffering from severe asthma and verifies the presence of ϩ least three independent experiments. ephrin-A1 protein in T cells and CD19 B cells (30). Most studies The Journal of Immunology 849

FIGURE 7. Real time RT-PCR analysis of IL-2 (A and B) and GUS or IL-4 (C and D) following engagement of anti-CD3-stimulated T cells with anti-ephrin A1 or isotype control (IC). The ﬂuorescent (passive dye)-normalized reporter signal (␦Rn) is shown against the number of PCR cycles. Bars, SEM of triplicate samples. The decrease of IL-2 (B) or IL-4 (D) expression is shown relative to housekeeping gene expression (GUS). The percentage was calculated on the ⌬⌬ basis of the Ct method; the mean of three independent experiments is shown. Bars, 95% conﬁdence in- Downloaded from terval (IC) calculated on the basis of deviation of GUS and ephrin-A1 expression. http://www.jimmunol.org/

to date support the notion that ephrin-mediated signals are sup- duced proliferation, whereas anti-ephrin-A1 Abs coimmobilized pressive. Thus, we speculate that diminished ephrin-A1 expression with anti-CD3/28 mAb significantly diminished clonal expansion. by guest on September 27, 2021 in allergic disease and an apparent suppressive effect of ephrins Consistently, anti-EphA1 receptor Abs enhanced slightly Ag spe- may be associated with a lack of suppression and, in turn, a loss of cifically and polyclonally induced T cell proliferation, despite the peripheral tolerance, as is often observed in allergic diseases. fact that EphA1 receptor is expressed only at very low levels on We observed that ephrin-A1 was marginally expressed in lymph APC of peripheral blood (31). These data show that T cells can be nodes and spleen and that expression of the receptors was not in T inhibited by ephrin A1 and EphA1 receptor interactions. cell-rich areas (e.g., white pulp), confirming previous studies demon- The fact that T cells of healthy and asthmatic patients were strating that EphA2 receptor is expressed only in dendritic cells in the equally suppressed, was expected, because low ephrin-A1-ex- crypts of human tonsils (30). In addition to peripheral lymphoid ex- pressing T cells of asthmatics are preactivated in vivo and thus pression, overlapping expression of ephrin-A1 and EphA1 receptors cannot by further activated in vitro and in turn show proliferative has been shown in rat thymus (43). Remarkably, we observed that responses comparable with those of healthy donors. The decrease EphA1 receptors were strongly and homogeneously expressed by in T cell proliferation correlated with anti-ephrin-A1-mediated large and small bronchi in mouse and human biopsy samples. In a suppression of the IL-2R␣ chain (CD25). These results support mouse model of OVA-induced allergic lung inflammation, we ob- previous studies showing that EphB6 cross-linking prevents anti- served epithelial thickening and a slight decrease in EphA1 receptor CD3-induced CD25 up-regulation in Jurkat cells (33) and suggests expression in areas in close proximity to inflammatory infiltrates. that ephrins may influence cytokine-mediated immune regulation. Taken together, these data demonstrate high expression in thymus and The dramatic effect of ligand cross-linking also confirms the bidi- bronchial-associated lymphoid tissue, which are both epithelial-rich rectional nature of ephrin signaling, in which GPI-linked ephrins tissues. High expression of EphA1 receptor within the lungs and the can specifically induce signal transduction by interacting with sig- changes in ephrin-A1 expression during allergic disease may not naling competent receptors in lipid rafts (4). be merely a consequence of asthma but may relate to disease The reduction in T cell proliferation was not due to T cell an- pathogenesis. ergy, because anti-ephrin-A1-treated cells could be restimulated, Activation of T cells by plate-immobilized anti-CD3 and anti- and was independent of the induction cell death. In contrast, CD28 mAb induced a strong, immediate, and long lasting (50Ð ephrin-A1 engagement prevented IL-2-dependent AICD, which 80%) decrease in ephrin-A1 expression compared with resting T depends on IL-2-dependent expression of FAS ligand (40). cells both at the RNA and protein level, suggesting that reduced Ephrin-A1-mediated AICD prevention was reversed by addition ephrin-A1 expression in asthma patients is likely due to recent T of exogenous IL-2, which strongly suggests that ephrin-A1 cell activation. This is supported by the observations that periph- directly affects the IL-2 pathway and is likely to underlie the eral T cells of asthma patients have a high frequency of cells bear- observed difference in proliferation assays. Moreover, the ab- ing the activation marker CD25 (44, 45), CD69 (46), or HLA-DR sence of IL-2 receptor (CD25) expression upon anti-ephrin (44). Anti-EphA1 receptor or anti-EphA2 receptor marginally re- treatment implies an ephrin-mediated effect on IL-2, further 850 EPHRIN-A1 INHIBITS Th2 CELLS supported by a strong reduction of IL-2 mRNA expression 14. Knoll, B., and U. Drescher. 2002. Ephrin-As as receptors in topographic projec- (82%) when T cells were stimulated with anti-ephrin-A1 Abs, tions. Trends Neurosci. 25:145. 15. Huai, J., and U. Drescher. 2001. An ephrin-A-dependent signaling pathway con- coimmobilized with anti-CD3/CD28 mAb. Additionally, incuba- trols integrin function and is linked to the tyrosine phosphorylation of a 120-kDa tion with anti-ephrin-A1 reduced IL-4 mRNA (69%) but had no protein. J. Biol. Chem. 276:6689. ␥ 16. Mellitzer, G., Q. Xu, and D. G. Wilkinson. 2000. Control of cell behaviour by effect on IFN- production. These findings are particularly impor- signalling through Eph receptors and ephrins. Curr. Opin. Neurobiol. 10:400. tant for allergic diseases, which are dominated by a Th2-like 17. Holder, N., and R. Klein. 1999. Eph receptors and ephrins: effectors of morpho- cytokine pattern. Taken together, these data suggest ephrin- genesis. Development 126:2033. 18. Cheng, N., D. M. Brantley, and J. Chen. 2002. The ephrins and Eph receptors in mediated cytokine regulation, and it is tempting to speculate that in angiogenesis. Cytokine Growth Factor Rev. 13:75. allergic disease, ephrin receptor-ligand interactions perturb periph- 19. O’Leary, D. D., and D. G. Wilkinson. 1999. Eph receptors and ephrins in neural eral suppression, thus leading to enhanced Th2 cell expansion. development. Curr. Opin. Neurobiol. 9:65. 20. Nakamoto, M., and A. D. Bergemann. 2002. Diverse roles for the Eph family of The present study demonstrates for the first time ephrin-A1 expres- receptor tyrosine kinases in carcinogenesis. Microsc. Res. Tech. 59:58. sion in T cells and lung. The abundant presence of ephrin-A1 and its 21. Takahashi, H., and T. Ikeda. 1995. Molecular cloning and expression of rat and mouse B61 gene: implications on organogenesis. Oncogene 11:879. main receptors EphA1 and EphA2 in the lung suggests a regulatory 22. Shao, H., A. Pandey, K. S. O’Shea, M. Seldin, and V. M. Dixit. 1995. Charac- interface of the immune system and the physiologically important terization of B61, the ligand for the Eck receptor protein-tyrosine kinase. J. Biol. epithelial surface in the lung. The impacts of ephrin-A1 and EphA Chem. 270:5636. 23. Holzman, L. B., R. M. Marks, and V. M. Dixit. 1990. A novel immediate-early receptors are of a suppressive consequence for the recipient T cell. It response gene of endothelium is induced by cytokines and encodes a secreted appears that T cells, which infiltrate bronchial tissues are hindered in protein. Mol. Cell. Biol. 10:5830. IL-2 and IL-4 expression, whereas AICD is supposed to be decreased 24. Pandey, A., H. Shao, R. M. Marks, P. J. Polverini, and V. M. Dixit. 1995. Role of B61, the ligand for the Eck receptor tyrosine kinase, in TNF-␣-induced an- in a ephrin/Eph receptor-dominated microenvironment, provided that giogenesis. Science 268:567. Downloaded from experimental, Ab-mediated ephrin-A1 cross-linking reflects the phys- 25. Fox, G. M., P. L. Holst, H. T. Chute, R. A. Lindberg, A. M. Janssen, R. Basu, and A. A. Welcher. 1995. cDNA cloning and tissue distribution of five human EPH- iological receptor ligand interaction. The physiological relevance of like receptor protein-tyrosine kinases. Oncogene 10:897. such an immunoprivilege may be seen in the important physiological 26. Lickliter, J. D., F. M. Smith, J. E. Olsson, K. L. Mackwell, and A. W. Boyd. function of the lung. Because ephrin signaling is bidirectional, it can 1996. Embryonic stem cells express multiple Eph-subfamily receptor tyrosine kinases. Proc. Natl. Acad. Sci. USA 93:145. be expected that T cell-mediated ephrin contact also affects the func- 27. Maru, Y., H. Hirai, and F. Takaku. 1990. Overexpression confers an oncogenic tion of the bronchial epithelial cells. These effects are the subject of potential upon the eph gene. Oncogene 5:445. http://www.jimmunol.org/ current investigations in our laboratory. The surface-exposed expres- 28. Lindberg, R. A., and T. Hunter. 1990. cDNA cloning and characterization of eck, an epithelial cell receptor protein-tyrosine kinase in the eph/elk family of protein sion of the ephrin system along with its negative regulatory function kinases. Mol. Cell. Biol. 10:6316. for the immune system, and possibly also for the epithelial cells, iden- 29. Easty, D. J., S. P. Hill, M. Y. Hsu, M. E. Fallowfield, V. A. Florenes, M. Herlyn, and D. C. Bennett. 1999. Up-regulation of ephrin-A1 during melanoma progres- tifies these genes as pharmacological targets and important molecules sion. Int. J. Cancer 84:494. for the pathogenesis of asthma. 30. Aasheim, H. C., E. Munthe, S. Funderud, E. B. Smeland, K. Beiske, and T. Logtenberg. 2000. A splice variant of human ephrin-A4 encodes a soluble molecule that is secreted by activated human B lymphocytes. Blood 95:221. References 31. De Saint-Vis, B., C. Bouchet, G. Gautier, J. Valladeau, C. Caux, and P. Garrone. 1. Wohlfahrt, J. G., S. Kunzmann, G. Menz, W. Kneist, C. A. Akdis, K. Blaser, and 2003. Human dendritic cells express neuronal Eph receptor tyrosine kinases: role

C. B. Schmidt-Weber. 2003. T cell phenotype in allergic asthma and atopic der- of EphA2 in regulating adhesion to fibronectin. Blood. In press. by guest on September 27, 2021 matitis. Int. Arch. Allergy Immunol. 131:272. 32. Luo, H., X. Wan, Y. Wu, and J. Wu. 2001. Cross-linking of EphB6 resulting in 2. Zhou, R. 1998. The Eph family receptors and ligands. Pharmacol. Ther. 77:151. signal transduction and apoptosis in Jurkat cells. J. Immunol. 167:1362. 3. van der Geer, P., T. Hunter, and R. A. Lindberg. 1994. Receptor protein-tyrosine 33. Freywald, A., N. Sharfe, C. Rashotte, T. Grunberger, and C. M. Roifman. The kinases and their signal transduction pathways. Annu. Rev. Cell Biol. 10:251. EphB6 receptor inhibits JNK activation in T lymphocytes and modulates T cell 4. Kullander, K., and R. Klein. 2002. Mechanisms and functions of Eph and ephrin receptor-mediated responses. J. Biol. Chem. 278:10150. signalling. Nat. Rev. Mol. Cell. Biol. 3:475. 34. Sharfe, N., A. Freywald, A. Toro, H. Dadi, and C. Roifman. 2002. Ephrin stim- 5. Pandey, A., R. A. Lindberg, and V. M. Dixit. 1995. Cell signalling: receptor ulation modulates T cell chemotaxis. Eur. J. Immunol. 32:3745. orphans find a family. Curr. Biol. 5:986. 35. 1992. International Consensus Report on Diagnosis and Treatment of Asthma. 6. Eph Nomenclature Committee. 1997. Unified nomenclature for Eph family re- National Institutes of Health Publication, Bethesda. ceptors and their ligands, the ephrins. Cell 90:403. 36. Schmidt-Weber, C. B., S. I. Alexander, L. E. Henault, L. James, and 7. Gale, N. W., S. J. Holland, D. M. Valenzuela, A. Flenniken, L. Pan, T. E. Ryan, A. H. Lichtman. 1999. IL-4 enhances IL-10 gene expression in murine Th2 cells M. Henkemeyer, K. Strebhardt, H. Hirai, D. G. Wilkinson, T. Pawson, S. Davis, in the absence of TCR engagement. J. Immunol. 162:238. and G. D. Yancopoulos. 1996. Eph receptors and ligands comprise two major 37. Kunzmann, S., J. G. Wohlfahrt, S. Itoh, H. Asao, M. Komada, C. A. Akdis, specificity subclasses and are reciprocally compartmentalized during embryogen- K. Blaser, and C. B. Schmidt-Weber. 2003. SARA and Hgs attenuate suscepti- esis. Neuron 17:9. bility to TGF-␤1-mediated T cell suppression. FASEB J. 17:194. 8. Bartley, T. D., R. W. Hunt, A. A. Welcher, W. J. Boyle, V. P. Parker, R. A. Lindberg, 38. Biosystems, P. A. 1997. ABI Prism 7700 Sequence detection system. User Bul- H. S. Lu, A. M. Colombero, R. L. Elliott, B. A. Guthrie, et al. 1994. B61 is a ligand letin No. 2. Company Newslr. for the ECK receptor protein-tyrosine kinase. Nature 368:558. 39. Mojtabavi, N., G. Dekan, G. Stingl, and M. M. Epstein. 2002. Long-lived Th2 9. Winslow, J. W., P. Moran, J. Valverde, A. Shih, J. Q. Yuan, S. C. Wong, memory in experimental allergic asthma. J. Immunol. 169:4788. S. P. Tsai, A. Goddard, W. J. Henzel, F. Hefti, et al. 1995. Cloning of AL-1, a 40. Van Parijs, L., A. Ibraghimov, and A. K. Abbas. 1996. The roles of costimulation ligand for an Eph-related tyrosine kinase receptor involved in axon bundle for- and Fas in T cell apoptosis and peripheral tolerance. Immunity 4:321. mation. Neuron 14:973. 41. Wohlfahrt, J. G., S. Kunzmann, G. Menz, W. Kneist, C. A. Akdis, K. Blaser, and 10. Davis, S., N. W. Gale, T. H. Aldrich, P. C. Maisonpierre, V. Lhotak, T. Pawson, C. B. Schmidt-Weber. 2003. T cell phenotype in allergic asthma and atopic der- M. Goldfarb, and G. D. Yancopoulos. 1994. Ligands for EPH-related receptor matitis. Int. Arch. Allergy Immunol. 131:272. tyrosine kinases that require membrane attachment or clustering for activity. Sci- 42. Schmidt-Weber, C. B., J. G. Wohlfahrt, and K. Blaser. 2001. DNA arrays in ence 266:816. allergy and immunology. Int. Arch. Allergy Immunol. 126:1. 11. Sakano, S., R. Serizawa, T. Inada, A. Iwama, A. Itoh, C. Kato, Y. Shimizu, 43. Munoz, J. J., C. L. Alonso, R. Sacedon, T. Crompton, A. Vicente, E. Jimenez, F. Shinkai, R. Shimizu, S. Kondo, M. Ohno, and T. Suda. 1996. Characterization A. Varas, and A. G. Zapata. 2002. Expression and function of the Eph A recep- of a ligand for receptor protein-tyrosine kinase HTK expressed in immature he- tors and their ligands ephrins A in the rat thymus. J. Immunol. 169:177. matopoietic cells. Oncogene 13:813. 44. Walker, C., J. C. Virchow, Jr., P. L. Bruijnzeel, and K. Blaser. 1991. T cell 12. Wang, X., P. J. Roy, S. J. Holland, L. W. Zhang, J. G. Culotti, and T. Pawson. subsets and their soluble products regulate eosinophilia in allergic and nonallergic 1999. Multiple ephrins control cell organization in C. elegans using kinase-de- asthma. J. Immunol. 146:1829. pendent and -independent functions of the VAB-1 Eph receptor. Mol. Cell 4:903. 45. Durham, S. R., S. J. Till, and C. J. Corrigan. 2000. T lymphocytes in asthma: 13. Davy, A., N. W. Gale, E. W. Murray, R. A. Klinghoffer, P. Soriano, bronchial versus peripheral responses. J. Allergy Clin. Immunol. 106:S221. C. Feuerstein, and S. M. Robbins. 1999. Compartmentalized signaling by GPI- 46. Huang, J. L., L. S. Ou, C. H. Tsao, L. C. Chen, and M. L. Kuo. 2002. Reduced anchored ephrin-A5 requires the Fyn tyrosine kinase to regulate cellular adhe- expression of CD69 and adhesion molecules of T lymphocytes in asthmatic chil- sion. Genes Dev. 13:3125. dren receiving immunotherapy. Pediatr. Allergy Immunol. 13:426.