Efnb1andEfnb2ProteinsRegulateThymocyteDevelopment ... ·...

Efnb1 and Efnb2 Proteins Regulate Thymocyte Development,Peripheral T Cell Differentiation, and Antiviral ImmuneResponses and Are Essential for Interleukin-6 (IL-6)Signaling*□S

Received for publication, September 9, 2011, and in revised form, September 28, 2011 Published, JBC Papers in Press, October 5, 2011, DOI 10.1074/jbc.M111.302596

Hongyu Luo‡, Tania Charpentier§, Xuehai Wang‡, Shijie Qi‡, Bing Han‡, Tao Wu‡¶1, Rafik Terra‡, Alain Lamarre§,and Jiangping Wu‡�2

From the ‡Laboratoire Immunologie and �Service Nephrologie, Centre de Recherche du Centre Hospitalier de l’Université deMontréal, Notre-Dame Hospital, Montreal, Quebec H2L 4M1, Canada, §Institut National de la Recherche Scientifique, INRS-InstitutArmand-Frappier, Laval, Québec H7V 1B7, Canada, and the ¶Institute of Cardiology, First Affiliated Hospital, Medical College,Zhejiang University, 310003 Hangzhou, China

Background: Ephrins (Efn) are the ligands of Eph kinases. The roles of Efn in the T cell compartment are studied.Results: Efnb1 and Efnb2 double knock-out mice showed compromised thymocyte development, Th1 and Th17 function, IL-6receptor signaling, and antivirus responses.Conclusion: Efnb1 and Efnb2 are involved in the T cell development and function.Significance: This study has revealed novel biological roles of Efns.

Erythropoietin-producing hepatocellular kinases (Ephkinases) constitute the largest family of cell membrane receptortyrosine kinases, and their ligand ephrins are also cell surfacemolecules. Because of promiscuous interaction between Ephsand ephrins, there is considerable redundancy in this system,reflecting the essential roles of these molecules in the biologicalsystem through evolution. In this study, both Efnb1 and Efnb2were null-mutated in the T cell compartment of mice throughloxP-mediated gene deletion.Mice with this double conditionalmutation (double KOmice) showed reduced thymus and spleensize and cellularity. There was a significant decrease in theDN4,double positive, and single positive thymocyte subpopulationsand mature CD4 and CD8 cells in the periphery. dKO thymo-cytes and peripheral T cells failed to compete with their WTcounterparts in irradiated recipients, and the T cells showedcompromised ability of homeostatic expansion. dKO naive Tcells were inferior in differentiating intoTh1 andTh17 effectorsin vitro. The dKO mice showed diminished immune responseagainst LCMV infection. Mechanistic studies revealed that IL-6signaling in dKO T cells was compromised, in terms of abated

induction of STAT3 phosphorylation upon IL-6 stimulation.This defect likely contributed to the observed in vitro and in vivophenotype in dKO mice. This study revealed novel roles ofEfnb1 and Efnb2 in T cell development and function.

Ephs3 are the largest family of cell surface receptor tyrosinekinases, comprising about 25% of known receptor tyrosinekinases (1). There are a total of 15 Ephs that are classified into Aand B subfamilies according to their sequence homology; theformer has ninemembers and the latter six, although not all areexpressed in a given species (2, 3). The ligands of Ephs, ephrins(Efns), are also cell surface molecules (1). There are nine Efnsdivided into A and B subfamilies according to the way theyanchor to the cell surface. The Efna subfamily has six membersthat are glycosylphosphatidylinositol-anchored membraneproteins; the Efnb subfamily has three members that are trans-membrane proteins. Interactions between Ephs and Efns arepromiscuous. One Eph can interact withmultiple Efns and viceversa. In general, Epha members preferentially interact withEfna members and Ephb members with Efnb members (2–4).Such promiscuous interactions suggest that these moleculesare so vital to biological systems that heavy redundancy isessential.Ephs are receptor tyrosine kinases that can initiate signal

transduction upon ligand binding. However, it is known thatalthough Efns are ligands, they can also transduce signals intocells (2, 3) in a phenomenon called reverse signaling. As a result,the interaction between Eph and Efns results in bi-directionalsignaling. Because Ephs and Efns are cell surface molecules,

* This work was supported in part by Canadian Institutes of Health ResearchGrants MOP57697 and MOP69089 (to J. W.), IMH 79565 and MOP97829 (toH. L.), and MOP89797 (to A. L.), by grants from the Heart and Stroke Foun-dation of Quebec, the Quebec Ministry of Economic Development, Inno-vation, and Exportation Grant PSR-SIIRI-069, by The J.-Louis LevesqueFoundation (to J. W.), a group grant from the Fonds de la Recherche enSanté du Québec for Transfusional and Hemovigilance Medical Research(to J. W.), and a J.-Louis Levesque Foundation grant (to A. L.).

□S The on-line version of this article (available at http://www.jbc.org) containssupplemental Figs. S1–S3 and Table I, A and B.

1 Supported in part by Zhejiang Provincial Natural Science Foundation ofChina Grant Y2080374 and National Natural Sciences Foundation of China(Projects for Young Scientists Grant 30800999).

2 To whom correspondence should be addressed: Laboratory of Immunol-ogy, CHUM Research Centre, Notre-Dame Hospital, Pavillion DeSève, Rm.Y-5616, 1560 Sherbrooke St. East, Montreal, Quebec H2L 4M1, Canada. Tel.:514-890-8000 (Ext. 25164); Fax: 514-412-7596; E-mail: [email protected].

3 The abbreviations used are: Eph kinase, erythropoietin-producing hepato-cellular kinase; Efn, ephrin; dKO, double KO; BMTx, bone marrow transplan-tation; DP, double positive; SP, single positive; qPCR, quantitative PCR;LCMV, lymphocytic choriomeningitis virus.

THE JOURNAL OF BIOLOGICAL CHEMISTRY VOL. 286, NO. 48, pp. 41135–41152, December 2, 2011© 2011 by The American Society for Biochemistry and Molecular Biology, Inc. Printed in the U.S.A.

DECEMBER 2, 2011 • VOLUME 286 • NUMBER 48 JOURNAL OF BIOLOGICAL CHEMISTRY 41135

by guest on June 27, 2020http://w

ww

.jbc.org/D

ownloaded from

http://www.jbc.org/cgi/content/full/M111.302596/DC1

http://www.jbc.org/

they will normally be activated locally by their binding partnersexpressed on opposing cells during physical contact. As a result,the major functions of Eph and Efn are related to pattern for-mation; however, additional functions not related to patternformation have been recently observed.Most reported functions of Ephs occur in the central nervous

system where they are expressed in neurons and control axonand dendrite positioning (2, 3). They are essential in the devel-opment of neuronal connections, circuit plasticity, and repair.However, they are also involved in a variety of other processes.Some Ephs and Efns are expressed in endothelial cells and arevital in angiogenesis during normal embryonic development aswell as in tumorigenesis (4, 5). Studies have shown that intesti-nal epithelial cells express different levels of some Eph and Efnbfamilymembers thatmodulate themovement of epithelial cellsalong the crypt axis to maintain epithelium self-renewal (6). Ithas been reported that Ephb2 and Efnb2 are expressed on theendoderm during embryonic development and that their bidi-rectional interaction is essential in urorectal development (7).Pancreatic �-cells communicate with each other via Epha andEfna family members to synchronize their insulin secretion inresponse to blood glucose fluctuations (8). Several Ephb andEfnb family members are expressed on osteoclasts andosteoblasts where they regulate bone development, mainte-nance, and repair (9, 10). Multiple Eph and Efn membershave been found to be expressed in some cancer cells, andthey seem to influence cancer cell growth (11). Ephb4and Efnb2 are expressed on hematopoietic progenitor cellsand regulate red blood cell production in response tohypoxia (12). Efnb1 and Epha4 expression in platelets con-tributes to the clotting process (13). Efnb1 expression onkidney epithelial cells (podocytes) likely plays a role in glo-merular filtration (14). The interaction between Ephb2 andEfnb2 regulates the ionic homeostasis of vestibularendolymph fluid in the inner ear (15).Our group and others have reported that Ephs and Efns, par-

ticularly their B familymembers as well as someA familymem-bers, are expressed in thymocytes and T cells and are capable ofmodulating T cell responses and survival (16–24). We haveshown that Efnb1, Efnb2, and Efnb3 initiate signaling throughtheir Eph receptors and can costimulate peripheral T cells interms of enhancing cytokine production and proliferation invitro (25–27). We further demonstrated that one of these Efnbreceptors, Ephb6, although lacking kinase activity, can transmitsignals into T cells (28) and that its null mutation leads to com-promised T cells responses in vitro and in vivo (29). However,Ephb6 null mutants have normal thymus structure and thymo-cyte development (29), probably due to complementary func-tions of other Eph family members. To reveal the roles of thesehighly redundant Eph/Efn familymembers in thymocyte devel-opment and T cell immune responses in vivo and in vitro, wegenerated Efnb1/Efnb2 double gene knock-out (dKO) micewith T cell-specific deletion of these two genes. Resultsobtained from these mice showed that Efnb1 and Efnb2 areinvolved in proper thymocyte development and peripheral Tcell function.

MATERIALS AND METHODS

Generation of T Cell-specific Efnb1 and Efnb2 Gene Knock-out Mice—A PCR fragment amplified with a primer set(5�-CTGAATAAGGGCTGCGAAAG-3� and 5�-GCAAATG-GCTTAACCCAAGA-3�) based on the Efnb1 genomicsequence was used as a probe to isolate a genomic BAC DNAclone 4M20 from the 129/svmouseBACgenomic libraryRPCI-22. A genomic BAC DNA clone 85F06PCR fragment based onthe EFN2 sequence was similarly amplified with the primer set(5�-GCTGC TCTTC AGTCA GTCAG C-3� and 5�-TACCAGTCATCACATCGCAG-3�). The targeting vectors were con-structed by recombination and routine cloningmethods using a12-kb Efnb1 genomic fragment from clone 4M20 for Efnb1 anda 12-kb Efnb2 genomic fragment from clone 85F06 for Efnb2(illustrated in Fig. 1, A and E). The final targeting fragments forEfnb1 and Efnb2were excised from its cloning vector backboneby NotI and electroporated into ES cells. After G418 selection,the FRT-flanked Neo/TK cassette was eliminated by subse-quent transient transfection of the ES cells with a Flippaseexpression vector. The targeting scheme is shown in Fig. 1A.These genetic engineering steps in ES cells resulted in two netinsertions for the EFNB1 constructs as follows: a 118-bp LoxP-containing sequence (5�-AGTACGGGCC CAAGCTGGCCGCCCTAGGGG CGCGCCTGCA GATAACTTCG TATA-ATGTAT GCTATACGAA GTTATGATAT CAAGCTTATCGATACCGTCG AAGCTTGCTA GCGGTACC-3�) at posi-tion 26061 (based on the sequence of AL671478.9 of the Gen-BankTM), and a 151-bp LoxP- plus FRT-containing sequence(5�-GGCCGCCCTA GGGGCGCGCC TGCAGATAACTTCGATAATG TATGCTATAC GAAGTTATGG ATC-GAAGTTC CTATTTCTAA AAAGTATAGG AACTTCT-TAA GGCCACCGCG GCCGAACGCT AGAGCTTGTCGACGGTACCTAACTTCCTAGG-3�) at position 28713, 756bp upstream and 1285 bp downstream of the Efnb1 Exon 1,respectively. Similarly for Efnb2, the genetic engineering stepsin ES cells resulted in two net insertions as follows: a 71-bpLoxP-containing sequence (5�-TCGACCTCGA CGGTATC-GAT AAGCTTGATA TCATAACTTC GTATAGCATACATTATACGA AGTTATCGTT T-3�) at position 186685(based on the sequence of AC090007.8 in theGenBankTM), anda 102-bp LoxP plus FRT-containing sequence (5�-GCGTTA-ATTA AGAAGTTCCT ATACTTTTTA GAGAATAGGAACTTCGATCCATAACTTCGTATAGCATACATTATAC-GAAG TTATCTGCAG GCGCGCCCCT AG-3�) at position188802, 923 bp upstream and 930 bp downstream of Efnb2Exon 1, respectively.The targeted ES cell clones were injected into C57BL/6 blas-

tocysts. Chimericmalemice weremated with C57BL/6 femalesto establishmutated Efnb1 and Efnb2 allele germ line transmis-sion. Southern blotting with a probe corresponding to the 3�sequences outside the targeting region, as shown in Fig. 1A(black square), was used to screen and confirm the gene target-ing and the successful removal of the Neo-TK selectionmarkerin Efnb1 ES cells and eventually in mouse tail DNA. The tar-geted allele showed a 7.3-kb AseI/ScaI fragment, whereas theWTallele showed an 11.6-kb fragment (Fig. 1B). Southern blot-ting with a probe corresponding to the 5� sequences outside the

Efnb1 and Efnb2 Required for T Cell Development

41136 JOURNAL OF BIOLOGICAL CHEMISTRY VOLUME 286 • NUMBER 48 • DECEMBER 2, 2011


ww

.jbc.org/D

ownloaded from

http://www.jbc.org/

targeting region, as illustrated in Fig. 1E (black square), wasused to screen and confirm the gene targeting and the success-ful removal of the Neo-TK selection marker in Efnb2 ES cellsand eventually in mouse tail DNA. The targeted allele showedan 8.1-kb PacI/PacI fragment in tail DNA (Fig. 1F), although aWTallele showed a 15.5-kb fragment. PCRwas used for routinegenotyping of the targeted allele(s). The following PCR condi-tions were used for Efnb1: 4 min at 95 °C, followed by 34 cyclesof 15 s at 94 °C, 30 s at 58 °C, and 60 s at 72 °C, and a finalincubation at 72 °C for 10 min. Primers 5�-GTCTC CACTGCCCAT AGCTC-3� (forward) and 5�-TGCTC CCAGTCCAGTC ACTA-3� (reverse) amplified a 271-bp fragmentfrom the targeted allele and a 133-bp fragment from the WTallele. PCR conditions for Efnb2were similar except the anneal-ing temperature was reduced from 58 to 52 °C. Primers5�-TAGCC ATCCC TTGGA ATACG-3� (forward) and5�-TTGGC GCGCC CCTT TCGAAG-3� (reverse) amplified a456-bp fragment from the targeted allele and a 355-bp fragmentfrom the WT allele.Mice with targeted Efnb1 allele(s) were named Efnb1f/f (loxP

insertions in both alleles) orEfnb1f (loxP insertion in one allele).Mice with targeted Efnb2 allele(s) were named Efnb2f/f (loxPinsertion in both alleles). These two lines were backcrossedwith C57BL/6 for more than 10 generations and then matedwith Lck promoter-driven Cre recombinase transgenic mice inthe C57BL/6 background (strain B6.Cg-Tg (Lck-Cre) 540Jxm/J�, The Jackson Laboratory, Bar Harbor, ME) to obtain T cell-specific Efnb1 and Efnb2 gene knock-out mice (Lck-Cre-Efnb1f/f) and Efnb2 gene knock-out mice (Lck-Cre-Efnb2f/f).Because Efnb1 is an X-linked gene, Lck-Cre-Efnb1f in males isequivalent to Lck-Cre-Efnb1f/f in females in that the Efnb1 geneis completely null-mutated. For the convenience of description,Lck-Cre-Efnb1f males and Lck-Cre-Efnb1f/f females are collec-tively referred to as Lck-Cre-Efnb1f/f mice. PCR was used toconfirm the T cell-specific deletion of Efnb1 using two differentprimer pairs. Primer pair 1 (forward, 5�-GTCTC CACTGCCCATAGCTC-3�, and reverse, 5�-TGCTCCCAGTCCAGTCACTA-3�) was used to detect a 271-bp fragment derived froman allele with undeleted Exon 1. Primer pair 2 (forward,5�-GTCTC CACTG CCCAT AGCTC-3�, and reverse, 5�-AC-CCTTACATCGAAGAACTGGGCA-3�) was used to detect a492-bp fragment derived from an allele with Exon 1 deleted.Similarly, T cell-specific deletion of Efnb2was confirmed usingtwo different primer pairs. Primer pair 1 (forward, 5�-TAGCCATCCC TTGGA ATACG-3�, and reverse, 5�-TTGGCGCGCCCCTTTCGAAG-3�) was used to detect a 456-bp frag-ment derived from an allele(s) with undeleted Exon 1. Primerpair 2 (forward, 5�-CTAAG GCTCT CAGCC TCGTG-3�, andreverse, 5�-TTGGC GCGCC CCTTT CGAAG-3�) was used todetect a 291-bp fragment derived from an allele(s) with Exon 1deleted. The PCR conditions used were same as describedabove for routine genotyping.Reverse Transcription/Quantitative-PCR (RT/qPCR)—

Efnb1,Efnb2,T-bet, andROR�tmRNA levels weremeasured byRT/qPCR. Total RNA from cells was extracted using TRIzol�(Invitrogen) and then reverse-transcribed with SuperscriptIITM reverse transcriptase (Invitrogen). Primers used are listedin supplemental Table IA. qPCR conditions for the reactions

was as follows: 2 min at 50 °C, 2 min at 95 °C followed by 45cycles of 10 s at 94 °C, 20 s at 58 °C, and 20 s at 72 °C. �-Actin orGAPDHmRNA levels were used as internal controls, and datawere expressed as signal ratios of test genemRNA/control genemRNA.Flow Cytometry—Single cell suspensions from the thymus,

spleen, or lymph nodes were prepared and stained for flowcytometry as described in our previous publications (25–29).For intracellular antigen staining, cells were first stained withAbs against cell surface antigens, fixed with Cytofix/Cytoper-mTM solution (BD Biosciences), and then stained with mAbsagainst intracellular antigens. Antibodies for flow cytometryare shown in supplemental Table IB.The following synthetic peptides were purchased from

Sigma Genosys: gp(33–41), KAVYNFATC (LCMV-GP,H-2Db); np(396–404), FQPQNGQFI (LCMV-NP, H-2Db);gp(276–286), SGVENPGGYCL (LCMV-GP, H-2Db); andgp(61–80), GLNGPDIYKGVYQFKSVEFD (LCMV-GP, I-Ab).PE-gp(33–41), PE-np(396–404), and PE-gp(276–286) H-2Db

tetrameric complexes were synthesized in-house and used at1:100 dilution as described previously (30). TheseMHCtetram-ers were used to detect LCMV-specific CD8� T cells. Briefly,splenocytes were first stained with PE-gp(33–41), PE-np(396–404), or PE-gp(276–286) tetramers for 30 min at 37 °C, fol-lowed by stainingwith FITC rat anti-mouseCD8� andAPC-ratanti-mouse CD62LmAbs at 4 °C for another 20min. 7-Amino-actinomycin Dwas used for exclusion of dead cells. After wash-ing, cells were fixed in 0.5% paraformaldehyde, and sampleswere analyzed by flow cytometry.Assessment of Thymocyte and T Cell Proliferation—Mice

were injected intraperitoneally with 1 mg of 5-bromo-2-de-oxyuridine (BrdU) in 0.2 ml of PBS and sacrificed 90 min laterfor thymocyte analysis. Another group of mice was injectedwith BrdU daily for 4 days and then sacrificed for spleen T cellanalysis. Thymocytes or spleen cells were then isolated andstained with Abs against cell surface markers. They were sub-sequently permeabilized and stained with FITC-conjugatedanti-BrdUAb using a BrdU flow kit (BDBiosciences) accordingto the manufacturer’s instructions.Generation of Bone Marrow Chimeras—Eight- to 10-week-

old C57BL/6(CD45.2�)�C57B6.SJL(CD45.1�) F1 mice wereirradiated at 1100 rads. Twenty four hours later, they received4 � 106 T cell-depleted bone marrow cells from C57/B6.SJLand dKOmice in a 1:1 ratio. Efnb1f/f/Efnb2f/f mouse bonemar-row was used as a control. Thymocytes and spleen cells of therecipients were analyzed by flow cytometry 8–10weeks follow-ing the bone marrow transplantation (BMTx).In Vitro Th1, Th2, Th17, and Treg Polarization—Naive

CD4� T cells were isolated from pooled splenocytes and lymphnode cells using the naive CD4� T cells isolation kit (R&D Sys-tems). Purity of the naive CD4� cells was routinely greater than95%. Purified naive T cells (0.25 � 106 cells/well) were mixedwith T cell-depleted irradiated (3000 rads) C57BL/6 feedersplenocytes (1.25 � 106 cells/well) and cultured in 96-wellplates under Th1, Th2, Th17, or Treg polarization conditions,as described in our earlier publication (31), and then analyzedby flow cytometry.




ww

.jbc.org/D

ownloaded from



http://www.jbc.org/

LCMV Infection and Clearance—LCMV-WE was obtainedfrom Dr. R. M. Zinkernagel (Zurich, Switzerland). Viral stockwas propagated in vitro, and viral titers were determined by thefocus-forming assay as described previously (32). Mice wereinfected by the i.v. route with 200 focus-forming units ofLCMV-WE. Mice were sacrificed 8 days post-infection to col-lect spleens for primary immune response analysis or 1 monthlater for memory immune response analysis. To evaluate viralclearance, mice were bled at days 4, 5, 7, 9, 11, 14, 23, and 28post-infection, and LCMV titers in whole bloodweremeasuredby the focus-forming assay.Detection of LCMV-specific IFN-� and TNF-�-producing T

Cells—One million splenocytes from LCMV-infected micewere seeded in single wells of 96-well round-bottomed plates.Cells were maintained in 5% RPMI 1640 medium supple-mented with 100 units/mlmrIL-2, 10 �g/ml brefeldin A, 10 �M

gp(33–41) or gp(61–80) peptide. After 5 h of incubation at37 °C, cells were stained with phycoerythrin-conjugated ratanti-mouse CD8� or CD4 mAbs and 7-aminoactinomycin D.Cells were then fixed, permeabilized and stained with allophy-cocyanin-rat anti-mouse TNF-� and FITC-rat anti-mouseIFN-� mAbs. Frequency of IFN-� and TNF-�-secreting T cellswas determined by flow cytometry.

51Cr Release Assay—Eight days following LCMV infection,splenocytes were tested for cytotoxic activity in a standard 51Crrelease assay as described previously (32).Immunoblotting—dKO or control spleen T cells were puri-

fied with the EasySepTM T cell enrichment kit from StemcellTechnologies (Vancouver, British Columbia, Canada). Thesecells as well as total thymocytes were reacted with or withoutIL-6 (50 ng/ml) at 37 °C for 5 min for the detection of STAT3phosphorylation. The cells were lysed in RIPA buffer (25 mM

Tris, pH7.6, 150mMNaCl, 1%Nonidet P-40, 1% sodiumdeoxy-cholate, 0.1% SDS) supplemented with protease inhibitors andphosphatase inhibitors (CompleteTM protease inhibitor mix-ture and PhosSTOP phosphatase inhibitor mixture, RocheDiagnostics). The lysates were resolved by 8% SDS-PAGE andtransferred to nitrocellulose membranes. STAT3 was detectedby blotting with rabbit anti-mouse phospho-STAT3 (Tyr-705)monoclonal antibody (clone D3A7, Cell Signaling Technology)followed by HRP-conjugated donkey anti-rabbit IgG (GEHealthcare). The membranes were then stripped and reprobedwith rabbit anti-mouse STAT3 mAb (clone 79D7, Cell Signal-ing Technology) for total STAT3 expression. The signals weredetected with SuperSignal West Pico Chemiluminescent Sub-strate (Thermo Scientific, Rockford, IL).

RESULTS

Generation of Efnb1 and Efnb2 Conditional KOMice—Efnb1and Efnb2 null mutation caused embryonic lethality (4, 33). Tostudy the function of these genes in the T cell compartment, wegenerated conditional KOmice with LoxP sites flanking Exon 1of Efnb1 or Efnb2 (Fig. 1, A, B, E, and F). These mice (Efnb1f/f

and Efnbf/f) were crossed with transgenic mice expressing Crerecombinase driven by a proximal Lck promoter. As shown inthe upper panels of Fig. 1, C and G, in the presence of the Lck-promoter Cre transgene, Exon 1 of Efnb1 or Efnb2 was deleted

in thymocytes butwasmaintained in the genomeof tail tissue inLck-Cre-Efnb1f/f and Lck-Cre-Efnb2f/f mice.

RT-qPCR confirmed T cell-specific deletion of Efnb1 andEfnb2mRNA in thymocytes and spleenT cells but not in spleenB cells (lower panels, Fig. 1, C and G). mRNA deletion in thethymocytes and spleen T cells was not complete in this quanti-tative assay, probably because of the less than perfect effective-ness of the LoxP-Cre deletion system. Abrogated cell surfaceexpression of Efnb1 and Efnb2 proteins in thymocytes andperipheral T cells is demonstrated by flow cytometry (Fig. 1, Dand H). The transgenic Cre in transgenic mice B6.Cg-Tg (Lck-Cre) 540Jxm/J generated by Dr. J. Marth was driven by the Lckproximal promoter (34). This proximal promoter becomesmost effective in thymocytes starting from the DN3 stage (34,35). Indeed, the effective deletion of Efnb1 started from theDN3 (double negative 3) stage and onward (Fig. 1D). For Efnb2,obvious deletionwas observed from theDN4 stage and onward,although the deletion could also be seen at the DN2 and DN3stage (Fig. 1H). Because the Lck proximal promoter ceases to beactive after the SP stages (34, 35), the Efnb1 and Efnb2 deletionobserved in peripheral T cells (Fig. 1, D and H) was caused byEfnb1 and Efnb2 deletion in their progenitor cells, i.e.DN3 andDN4 cells. Although the deletion in theDN3 andDN4 stagewasnot complete, the residual undeleted DN cells did not seem todevelop into a major population in the periphery, and even theincomplete deletionwas sufficient to cause phenotypic changesin the T cell compartment, as will be described below. It isworth mentioning that deletion of Efnb1 alone did not causecompensative up-regulation of Efnb2 or vice versa (supplemen-tal Fig. 1A).Thymus and Spleen Features of dKOMice—Lck-Cre-Efnb1f/f

and Lck-Cre-Efnb2f/f single gene knock-out mice showed nodiscernable difference from the control Efnb1f/f and Efnb2f/fmice in terms of lymphoid organ size and histology (data notshown).We therefore generated Lck-Cre-Efnb1f/f/Efnb2f/f dKOmice. Efnb1f/f/Efnb2f/fmice were used as controls and are calledWT controls in the subsequent text. The dKOmice had smallerthymi and spleens (Fig. 2A). The weight and cellularity of theseorgans and the T cell number in the spleen were diminishedcompared with those of WT control mice (Fig. 2B).Efnb1 and Efnb2 Are Essential in Thymocyte Development—

Thymocyte subpopulation analysis by flow cytometry of Lck-Cre-Efnb1f/f and Lck-Cre-Efnb2f/f single gene knock-out miceshowed no significant abnormality (data not shown). The dKOthymi were then analyzed. Representative histograms areshown in Fig. 3A, and data from the 20 to 27 pairs of dKO andcontrols are summarized in Table 1. The absolute number butnot the percentage of CD4 and CD8 SP cells of dKO thymi wasdecreased compared with that of controls (Table 1). For thedouble positive (DP) cells that constitute the bulk of the thy-mus, the cell number in the dKO thymi was significantlyreduced compared with controls (a mean of 61.4 � 106 cells inthe dKO thymi versus a mean of 112.9 � 106 in the controlthymi), consistent with the reduced thymicweight and cellular-ity in the dKO mice; the percentage of these cells in the dKOthymi also showed significant decrease (Table 1). These resultsshow that Efnb1 and Efnb2 play a critical role in DP populationdevelopment, and the diminished pool of DP cells in the dKO




ww

.jbc.org/D

ownloaded from



http://www.jbc.org/




ww

.jbc.org/D

ownloaded from

http://www.jbc.org/

FIGURE 1—continued




ww

.jbc.org/D

ownloaded from

http://www.jbc.org/

thymus might contribute to smaller SP populations. There wasa significant relative increase in the percentage of double neg-ative (DN) cells in the dKO thymi, although the absolute num-ber of these cells remained unchanged compared with controls(Table 1). This suggests that the increased DN percentage ismainly a result of a decreased DP cell population.We wondered whether compromised progression from

DN cells to DP cells contributed to the significant decreasein the DN population in the dKO thymus. Because the prox-imal Lck promoter is most effective in the DN3 and DN4stages (35), we chose to analyze subpopulations from theDN2 to DN4 stages. Representative histograms are shown inFig. 3B, and a summary of data from 14 to 22 pairs of dKOmice and controls are presented in Table 2. The percentageand absolute cell number of the DN2 subpopulation in thedKO thymus were comparable with controls, whereas thoseof the DN3 subpopulation were increased, and those of theDN4 subpopulation were reduced (Table 2). This is consist-ent with the notion that the most effective Cre expressiondriven by the Lck proximal promoter starts from the DN3stage. The results suggest that deletion of both Efnb1 and

Efnb2 impairs the development from the DN3 to the DN4stage and that this is at least a contributing factor leading toa reduced DP population.The existence of different thymocyte subpopulations repre-

sents a balance between their proliferation, apoptosis, and pro-gression. We wondered whether null mutation of both Efnb1and Efnb2 affects the proliferation of thymocytes. The prolifer-ation of DN3, DN4, DP, and SP subpopulations was assessed invivo. As shown in Fig. 3C, no apparent difference in the propor-tion of BrdU-positive CD25�icTCR�� or CD25�icTCR��

DN3 cells or CD25�icTCR�� DN4 cells was observed betweendKO and control thymi nor was there any difference at the DP,CD4SP, and CD8SP stages. We used icTCR�� or icTCR�� togate DN3 and DN4 subpopulations for their BrdU uptake (Fig.3C). As dKO and control thymi did not differ significantly intheir icTCR�� cell percentage at theDN3 andDN4 stages (sup-plemental Fig. 1B), the BrdU results from Fig. 3C were notcaused by abnormal icTCR� expression in dKO cells. Theseresults indicate that the reduced cellularity of certain subpopu-lations in the dKO thymus does not seem to be the result oftheir decreased proliferation.

FIGURE 1. Generation of mice with T cell-specific Efnb1f/f and Efnb2f/f null mutation. A and E, schemes of Efnb1f/f and Efnb2f/f mouse generation. Theschemes of genetic manipulation to generate floxed Efnb1 (A) and Efnb2 (E) mice are illustrated. Bold lines represent left and right arms of genomic sequencesused in gene targeting. Empty squares represent exons (E) or thymidine kinase gene/neomycin-resistant gene cassettes (TK/Neo). LoxP and FRT sites arerepresented by large and small arrowheads, respectively. The solid box represents a genomic region from which probes were produced for Southern blotanalysis. B and F, Southern blot analysis of tail DNA of Efnb1- and Efnb2-targeted mice. For Efnb1-targeted mice (B), tail DNA was digested with AseI and ScaI andanalyzed by Southern blotting. The 11.6-kb band was derived from the wild type allele, and the 7.3-kb band was from the mutated allele. For Efnb2-targetedmice (F), the tail DNA was digested with PacI and analyzed by Southern blotting. The 15.5-kb band was derived from the wild type allele, and the 8.1-kbband was from the mutated allele. C and G, T cell-specific deletion of Efnb1 and Efnb2 in Lck-Cre-Efnb1f/f and Lck-Cre-Efnb2f/f mice according to PCR. Efnb1f/f miceand Efnb2f/f mice were crossed with transgenic mice harboring Cre recombinase driven by the proximal LCK promoter. In the upper panel, the tail andthymocyte DNA of the mice was analyzed with PCR using primer pairs 1 and 2 as described in supplemental material. For Efnb1f/f mice, primer pair 1 detecteda 271-bp fragment derived from the undeleted allele, although primer pair 2 detected a 492-bp fragment derived from the allele with Efnb1 exon 1 deleted. ForEfnb2f/f mice, primer pair 1 detected a 459-bp fragment derived from the undeleted allele, whereas primer pair 2 detected a 291-bp fragment derived from theallele with the Efnb2 exon 1 deleted. In the lower panel, total RNA was extracted from Lck-Cre-Efnb1f/f (C) or Lck-Cre-Efnb2f/f (G) thymocytes, spleen T cells, andspleen B cells. Their Efnb1 and Efnb2 mRNA was analyzed by RT/qPCR using �-actin as an internal control. Data are expressed as the means � S.D. of the ratiosof Efnb1/�-actin (C) and Efnb2/�-actin (G) signals. D and H, deletion of Efnb1 and Efnb2 expression in Lck-Cre-Efnb1f/f and Lck-Cre-Efnb2f/f thymocytes and spleenT cells according to flow cytometry. Thymocytes at different differentiation stages (DN2– 4, DP, CD4SP, and CD8SP) and spleen CD4 and CD8 cells were gatedand were analyzed for Efnb1 expression (E) or Efnb2 expression (H) by flow cytometry. The dotted line, isotypic control Ab; the dashed line, control WT cellsstained with anti-Efnb1 or anti-Efnb2 Abs; the solid line, Lck-Cre-Efnb1f/f and Lck-Cre-Efnb2f/f cells stained with anti-Efnb1 or anti-Efnb2 Abs, respectively. Thepercentage of Efnb1- or Efnb2-positive cells after the deduction of the background signal (isotypic Ab control) is shown. All the experiments in this figure wererepeated two or more times, and data from representative ones are shown.

FIGURE 2. General features of dKO lymphoid organs. A, gross morphology of the dKO thymus and spleen. B, weight and cellularity of dKO thymus and spleen.Mouse number (n) of each group is shown. Asterisks indicate a highly significant difference (p � 0.01, paired Student’s t test).




ww

.jbc.org/D

ownloaded from




http://www.jbc.org/

We then evaluated whether thymocyte over-apoptosis was afactor causing the decrease of certain thymic subpopulations indKO mice. Freshly isolated dKO thymocytes presentedenhanced annexin V expression in all stages examined (DN3,DN4, DP, CD4SP, and CD8SP) compared with the controls(Fig. 3D). The higher apoptotic rate in DN3 cells does not seemto be consistent with increased DN3 cell percentage and abso-lute number in dKO thymi; it is possible that the effect of theblockage of progress from DN3 to DN4 resulted in an accumu-

lation of DN3 cells that outweighed the detrimental effect ofapoptosis in this subpopulation. The results of this section indi-cate that in the absence of Efnb1 and Efnb2, there is increasedapoptosis in all subsets of thymocytes, and this could contributeto the reduced cellularity in the T cell compartment.We assessed the CD3, CD5, CD24, and CD69 expression in

dKO thymocyte subpopulations, particularly at the DP stage,where negative and positive selections occur. These cell surfaceproteins are critical in negative (CD3 and CD25) and positive

FIGURE 3. Flow cytometry analysis of dKO thymocytes. A, DN, DP, CD4SP, and CD8SP subpopulations in dKO thymi. Thymocytes were analyzed for DN, DP,CD4SP, and CD8SP subpopulations. Representative histograms are shown, and the percentage of each subpopulation is indicated. Data from 20 to 27 pairs ofdKO and control WT mice are summarized in Table 1. B, DN2, DN3, and DN4 subpopulations in dKO thymi. Thymocytes were first gated on Lin� cells, and theLin� cells were analyzed for DN2, DN3, and DN4 subpopulation based on CD25 and CD44 expression. Representative histograms are shown. Data from 20 to27 pairs of dKO and control WT mice are summarized in Table 2. C, DN3, DN4, DP, CD4SP, and CD8SP cell proliferation in dKO thymi according to BrdU staining.Control Efnb1f/f/Efnb2f/f and dKO mice were injected intraperitoneally with BrdU (1 mg/mouse), and their thymocytes were analyzed by flow cytometry 90 minlater. Thymocytes were stained with lineage markers as well as anti-BrdU Ab. Percentages of BrdU-positive DN3 (CD25�icTCR�� and CD25�icTCR��), DN4(CD25�icTCR��), DP, CD4SP, and CD8SP cells are shown. D, dKO thymocyte apoptosis. dKO and WT control thymocytes were stained with annexin V andlineage markers ex vivo or cultured for 24 h in plain medium and then analyzed as indicated. The percentage of annexin V-positive cells in DN3, DN4, DP, CD4SP,and CD8SP cells is shown. In C and D, the experiments were repeated more than three times, and representative histograms are shown.




ww

.jbc.org/D

ownloaded from

http://www.jbc.org/

selection or markers (CD24 and CD69) of the selection processwhen TCR activation occurs (36–39). No significant differencewas found between dKO and WT cells (supplemental Fig. 2A)with regard to the expression of thesemolecules. Therefore, thesignificantly reduced percentage and absolute number of DPcells (Table 1) in the dKO thymus is unlikely to be a result of amore aggressive negative selection or less efficient positiveselection. Rather, it is more likely due to increased general apo-ptosis in all thymocyte subpopulations and blocked progressionfrom DN3 to DN4.Efnb1 and Efnb2 Are Required for Proper Function of Periph-

eral T Cells in Vitro—The peripheral T cell subpopulation andT cell function of dKO mice were evaluated next. There was asignificant decrease of spleen weight and cellularity (Fig. 2B)and a significant reduction in T cell percentage in the spleenand lymph nodes of dKO mice compared with controls (Fig.4A). Both the CD4 and CD8 cells of the dKO spleen and lymphnodes were proportionally decreased (Fig. 4B). There was a2-fold relative increase in the percentage of CD44hiCD62Llomemory phenotype CD4 and CD8 T cells in the dKO spleencompared with the WT spleen (Fig. 4C, histograms, left panel).This increase came at the expense of CD44loCD62Lhi dKOnaive CD4 and CD8 T cells that were reduced almost 2-foldcompared with controls. Considering the approximate 2-folddecrease of the absolute T cell number in the dKO spleen (Fig.2B), the absolute memory phenotype T cell number in the dKOspleen remained unchanged, although both percentage andabsolute number of naive T cells in the dKO spleen were signif-icantly reduced (Fig. 4C, bar graph, right panel). It seems thatEfnb1 and Efnb2 affect naive CD4 and CD8 cells more thanmemory cells.Some CD44hiCD62Llo T cells in the periphery are derived

directly from CD44hiCD62Llo thymocytes (40) and displayinnate immune function by rapidly secreting IFN-� upon stim-ulation (41). We examined the SP thymocytes of dKO andWTfor their CD44 and CD62L expression and noticed a higherCD44hiCD62Llo percentage in dKO than in WT thymi (Fig.4D). We also evaluated CD44hiCD62Llo spleen T cells for theirintracellular IFN� expression 4h after stimulationwith phorbol12-myristate 13-acetate and ionomycin. IFN-�-positive WT

versus dKO CD4CD44hi cells were 1.5 versus 6.2%, and WTversus dKOCD8CD44hi cells were 5.2 versus 6.0% (supplemen-tal Fig. 2B). So, there is indeed an increase in the innate typeCD44hiCD62Llo T cell population in the periphery, and thesecells are probably directly derived from the thymus as suggestedby Rodgers (40). However, the increment is small in scale com-pared with the increase of CD44hiCD62Llo CD4 and CD8 cellsin the dKO mice as shown in Fig. 4C. Because the majority ofthe dKO peripheral CD44hiCD62Llo cells did not behave likeinnate immune cells by rapidly secreting IFN� upon phorbol12-myristate 13-acetate /ionomycin stimulation, they are prob-ably genuine memory T cells.Are there additional factors causing the skew ofmemory ver-

sus naive T cell ratio in dKOmice?We assessed the apoptosis ofthese cells from the spleen. Both naive and memory type CD4and CD8 cells from dKO mice showed an increased annexinV-positive percentage compared with those from WT mice(Fig. 4E), but naive type cells did not present a higher propor-tion compared with memory type cells. This higher apoptoticrate of all the dKO subsets likely contributes to the reducedperipheral T cell populations in dKOmice, but it is not a reasonfor a relatively higher percentage of memory type T cells inthesemice.We also demonstrated that both naive andmemorytype T cells from dKO mice manifested a similar BrdU uptakein vivo compared with cells from WT mice (supplemental Fig.3A). Taken together, these results suggest that the significantreduction of naive dKO T cells compared with memory dKO Tcells is due to reduced output of naive T cells from the dKOthymus, but Efnb1 and Efnb2 do not seem to affect significantlythe conversion from naive to memory type T cells.We next assessed different functional T cell subpopulations

either ex vivo or following in vitro differentiation. There was nochange in the relative proportion of CD25�FoxP3� Treg cellsin the dKO spleen and lymph nodes relative to the controlspleen and lymph nodes when examined ex vivo (supplementalFig. 3B). When naive CD4 cells were cultured under Th1, Th2,Th17, and Treg conditions (Fig. 4F), the dKO cells (lower row)showed compromised differentiation toward Th1 (IFN-��IL4� cells; 1st panel) and Th17 (IFN-��IL-17� cells; 3rdpanel) but not Th2 (IL-4�IFN-�� cells; 2nd panel) nor Treg

TABLE 1Percentage and absolute numbers of CD4SP, CD8SP, DP, and DN subpopulations

CD4SP CD8SP DP DN% Cells (�106) % Cells (�106) % Cells (�106) % Cells (�106)

WT 10.60 � 3.75 13.60 � 4.82 4.25 � 1.69 5.84 � 2.61 79.31 � 6.37 112.93 � 38.34 5.42. �2.90 6.84 � 2.85dKO 9.20 � 3.82 6.93 � 4.66a 4.78 � 1.82 3.91 � 2.16a 74.39 � 6.33b 61.43 � 32.22a 16.11 � 4.21a 7.98 � 3.97

n � 27 n � 20 n � 27 n � 20 n � 27 n � 20 n � 27 n � 20a p � 0.001.b p � 0.01; Student’s t test.

TABLE 2Percentage and absolute numbers of DN2, DN3, and DN4 thymocytes

DN2 DN3 DN4% Cells (�106) % Cells (�106) % Cells (�106)

WT 3.03 � 2.015 0.30 � 0.14 37.70 � 11.02 4.25 � 2.22 53.22 � 15.75 5.52 � 3.06dKO 2.97 � 1.42 0.37 � 0.23 54.16 � 12.63a 6.33 � 2.84b 40.12 � 13.76a 4.44 � 2.59b

n � 16 n � 14 n � 22 n � 14 n � 22 n � 14a p � 0.001.b p � 0.05; Student’s t test.




ww

.jbc.org/D

ownloaded from








http://www.jbc.org/

FIGURE 4. Flow cytometry analysis of dKO peripheral lymphocytes and their in vitro differentiation. A, T and B cell subpopulations in dKO spleens. Spleenand lymph node cells were stained with anti-CD3 and anti-B220 Abs for T and B cells, respectively. Percentage of CD3- and B220-positive cells is shown. B, CD4and CD8 T cell subpopulations in dKO spleens. Spleen or lymph node cells were stained with anti-CD4 and anti-CD8 Abs for CD4 and CD8 T cells, respectively.Percentage of CD4- and CD8-positive cells is shown. C, naive and memory T cell populations in dKO spleens. CD44loCD62Lhi cells and CD44hiCD62Llo cellsamong CD4� and CD8� spleen cells were analyzed, and their percentage is shown in the histogram. The bar graph at left illustrates the absolute number ofCD44loCD62Lhi cells and CD44hiCD62Llo cells in the spleen based on the calculation of the percentage and spleen cellularity of 10 pairs of dKO and WT mice.***, p � 0.001 (Student’s t tests). D, CD44loCD62Lhi cells and CD44hiCD62Llo cells in the dKO SP thymocytes. CD4SP and CD8SP thymocytes were gated, andCD44hiCD62Llo cells among these subpopulations were analyzed by flow cytometry. Their percentage is shown in the histograms. E, apoptosis of spleenCD44loCD62Lhi cells and CD44hiCD62Llo T cells of dKO mice. dKO and WT spleen CD44loCD62Lhi cells and CD44hiCD62Llo cells in the CD4 and CD8 subpopulations wereanalyzed ex vivo for their annexin V expression by flow cytometry. F, in vitro Th1, Th2, Th17, and Treg differentiation according to intracellular IFN-�, IL-4, IL-17, andFoxp3 staining. Naive CD4 cells were cultured under Th1, Th2, Th17, or Treg conditions (3 days for Th1, Th17, and Treg and 5 days for Th2). Cells were further stimulatedwith phorbol 12-myristate 13-acetate and ionomycin for 4 h before being harvested. The intracellular expression of IFN-�, IL-4, IL-17, and Foxp3 was determined byflow cytometry. Percentage of IFN-�, IL-4, IL-17, and Foxp3-positive cells is shown. G, expression of T-bet and ROR�t mRNA in Th1 and Th17 cells. T-bet and ROR�t mRNAfrom in vitro differentiated Th1 and Th17 cells, as described in F, was quantified by RT-qPCR, using GAPDH as an internal control. Data from five independentexperiments are expressed as the means � S.D. of the ratios of T-bet/GAPDH or ROR�t/GAPDH signals. *, p � 0.05 (Student’s t test). Experiments in this figure wererepeated more than three times, unless indicated otherwise, and representative data are shown.




ww

.jbc.org/D

ownloaded from

http://www.jbc.org/

(IL-17�FoxP3� cells; 4th panel), compared with controls(upper row).In testing the in vitro Th1 and Th17 differentiation, we

started with the same number of naive CD4 cells. The compro-mised Th1 and Th17 development of dKOCD4 could be due todefective expansion or differentiation, or both. To distinguishthese possibilities, we examined the expression of transcriptionfactors T-bet and ROR�t, which are essential transcription fac-tors in Th1 and Th17 cell differentiation, respectively. Asshown in Fig. 4G, under Th1 culture conditions, there was asmall but significant increase inT-bet expression in dKOTcellscompared with WT T cells, although the ROR�t expression ofthe dKO andWT cells under Th17 culture conditions was sim-ilar. The reason for this small increase is not clear, but thisresult suggests that the defective Th1 and Th17 cell develop-

ment of dKO CD4 T cells is probably mainly due to compro-mised expansion rather than differentiation.Pivotal Role of Efnb1 and Efnb2 Reverse Signaling in T Cell

Development andHomeostatic Expansion—Ephs are capable oftransducing signals in both directions (2, 3). We wanted tounderstand which direction of signaling was critical for theobserved phenotype in the dKO mice. To this end, we used amodel of whole body irradiation followed by BMTx, usingB6.SJL bone marrow cells to compete with dKO bone mar-row cells. In this model, CD45.1 single positive cells werederived from competitor B6.SJL bone marrow cells; CD45.2single positive cells were derived from the dKO bonemarrowcells; and CD45.1/CD45.2 double positive cells were derivedfrom residual recipient bone marrow cells and peripheralcells.

FIGURE 4 —continued




ww

.jbc.org/D

ownloaded from

http://www.jbc.org/

As shown in Fig. 5A, 63.9% of the thymocytes in the controlrecipients were derived from WT donor bone marrow cells(CD45.2 single positive cells; upper left panel), whereas in thetest recipients only 0.4% thymocytes were derived from dKOdonor bone marrow cells (CD45.2 single positive cells; Fig. 5A,upper right panel). Similar significant differences were alsofound in the spleen (Fig. 5A, 2nd row), lymph nodes (3rd row),and blood (last row). This clearly indicates that the dKO

bonemarrow cells were significantly inferior to theWT controlbone marrow cells in their capacity to compete with B6.SJLbone marrow cells to develop and expand in the void nichecreated by irradiation.As expected, the proportion of CD3� T cells in the spleen,

lymph nodes, and blood derived from dKO bone marrow cellswas significantly lower than the proportion derived from con-trol bone marrow cells (Fig. 5B). However, the percentage of B

FIGURE 5. Compromised development and expansion of dKO bone marrow cells and mature T cells in irradiated recipients. A, dKO bone marrow cellsfailed to compete with WT bone marrow cells in irradiated recipients. T cell-depleted dKO and WT bone marrow cells (CD45.2�) were mixed with T cell-depletedbone marrow cells from B6.SJL competitors (CD45.1�) at 1:1 ratio and transplanted to lethally irradiated C57BL/6�B6.SJL F1 recipients. After 90 days, cells fromthe thymus, spleen, lymph node, and blood were analyzed for CD45.2 and CD45.1 staining. Percentages of CD45.2� cells (derived from dKO and WT controlbone marrow cells), CD45.1� cells (derived from competing B6.SJL bone marrow cells), and CD45.1�/CD45.2� cells (derived from residual cells of the recipi-ents) are shown. B, dKO bone marrow cells in the presence of competing B6.SJL bone marrow cells had significantly reduced capability to develop into T cellsin the periphery. In the whole body irradiation-BMTx model described in A, B220� B cells and CD3� T cells among CD45.2� cells (derived from dKO or controlWT bone marrow cells) in the spleen, lymph node, and blood were determined by flow cytometry 90 days later; percentages are shown. C, dKO T cells presentedfailed homeostatic expansion in sublethally irradiated recipients. B6.SJL mice (CD45.1�) were sublethally irradiated at 600 rads and transplanted i.v. with 4 �106 CFSE-labeled spleen cells from dKO or WT control mice (CD45.2�). The histograms show profiles of carboxyfluorescein succinimidyl ester (CFSE)-positivecells gated on CD4�CD45.2� and CD8�CD45.2� cells. Experiments in this figure were repeated more than three times, and representative histograms areshown.




ww

.jbc.org/D

ownloaded from

http://www.jbc.org/

cells (B220-positive cells) derived from dKO bonemarrow cellswas greater than that derived from control bone marrow, indi-cating a T cell-specific development defect due to the Lck-Cre-caused deletion of both Enb1 and Enb2. In this system, Ephs ofdKO thymocytes and peripheral T cells should have sufficientforward Efnb1 and Efnb2 stimulation from competing WTB6.SJL bone marrow-derived cells, whose Efnb1 and Efnb2expression was normal. Therefore, we could attribute thedefective development and expansion of dKO thymocytesmainly to the absence of both Efnb1 and Efnb2 reversesignaling.To understand whether mature dKO T cells were incompe-

tent in homeostatic expansion, we transferred CFSE-labeleddKOorWT spleen cells into sublethally irradiated B6.SJLmice.As shown in Fig. 5C, dKO donor T cells showed significantly

lower homeostatic proliferation in vivo compared with WTcontrol donor T cells. This compromised homeostatic prolifer-ation of mature dKO T cells along with their increased apopto-sis and reduced thymocyte output as described above are likelyall contributing factors resulting in the significant decrease ofperipheral T cells in the dKO mice.Efnb1 and Efnb2 Deletion in the T Cell Compartment Results

in Compromised Immune Responses to LCMV Infection—Thereduced peripheral T cell population and defective T helper celldifferentiation in dKOmice strongly suggested that these micemight have incompetent immune responses in vivo. We there-fore assessed their anti-LCMV immune response. Eight daysafter themice were infected with LCMV (strainWE), there wasa significant increase in the absolute number of total spleencells in both dKO and control mice (Fis. 6A versus 2B). The

FIGURE 6. Compromised in vivo anti-LCMV immune responses of dKO mice. A, spleen cell number on day 8 following LCMV infection. Means � S.D. ofabsolute cell number of total splenocytes, CD4� cells, and CD8� cells of control WT control (n � 4) and dKO (n � 4) mice on day 8 post-LCMV infection areshown. B and C, virus-specific CD8 cells on day 8 post-LCMV infection. On day 8 post-infection, the percentage of gp33, np396, and gp276 tetramer-positivecells among CD8 cells (B) and the absolute number of gp33, np396, and gp276 tetramer-positive CD8 cells per spleen were measured by flow cytometry.Means � S.D. of data from four pairs of WT control and dKO mice are shown. D, virus-specific lymphokine-producing CD4 and CD8 cells on day 8 and day 32post-LCMV infection. Percentage of IFN-�-producing virus-specific CD4 cells (gp61-positive) and CD8 cells (gp33-positive) (left column), and percentage ofTNF-�-producing virus-specific CD4 cells (gp61-positive) and CD8 cells (gp33-positive) (right column) were measured by flow cytometry on day 8 (upper row)and day 32 (lower row) post-LCVM infection. Means � S.D. of data from 4 to 6 pairs of WT control and dKO mice are shown. E, cytotoxic T lymphocyte activityon day 8 post-LCMV infection. On day 8 post LCMV infection, virus-specific cytotoxic T lymphocyte activity was measured with 51Cr-release assays at differenteffectors versus target cell ratios. Percentage of specific lysis was calculated based on ratios of spleen cells versus target cells. Means � S.D. of data from WTcontrol (n � 4) and dKO (n � 3) mice are shown. A–E, data from 3 to 4 mice were pooled and analyzed by one-way analysis of variance followed by Bonferroni’smultiple comparison test. p values are indicated when they reached �0.05.




ww

.jbc.org/D

ownloaded from

http://www.jbc.org/

increase of CD8 cells surpassed that of CD4 cells, and theybecame the dominant cell population in the spleen (Fig. 6A),because the antiviral response is mainly a CD8 cell-mediatedevent. However, the absolute cell number of bothCD4 andCD8cell populations was significantly lower in dKOmice comparedwith that of the control Efnb1f/f/Efnb2f/f mice. Using LCMVantigen-specific tetramers, it was shown that the percentage ofgp33-, np396-, and gp276-specific CD8 cells among the totalCD8 cells was comparable in both dKO and control mice (Fig.6B), suggesting a similar rate of clonal expansion following viralantigen stimulation. The absolute number of tetramer-positiveCD8 cells per spleen tended to be lower in dKOmice (Fig. 6C),probably reflecting a smaller startingT cell population, but theydid not reach statistical significance except the gp33-positivecells at the present sample size, likely due to the level ofvariation.Wenext examined the presence of antigen-specific cytokine-

producing spleenT cells in the virus-infectedmice. As shown inFig. 6D (upper row), the relative percentage of antigen-specificIFN-�-producingCD8 cells, but notCD4 cells, was significantlylower in dKO than control spleens 8 days post-infection (1stpanel, upper row). This was also the case for antigen-specificTNF-�-producing CD8 cells (Fig. 6D, 2nd panel, upper row).Such reduction is probably causedmainly by the reduced num-ber of antigen-specific CD8 cells to start with, but their com-promised differentiation might also play a role. The defectivedKOCD4 cell help, as demonstrated in our in vitro study in Fig.4F, probably also contributed to the compromised CD8 celldevelopment. Antigen-specific IFN-� and TNF-� secretingCD4 cells in the dKOspleenwere lower in percentage andnum-ber compared with those in the control spleen, although thedifference was not statistically significant at the current samplesize (n � 4; Fig. 6D, upper row).

Interestingly, we noticed that 32 days following virus infec-tion, the dKOmice caught up with the control mice in that thatthe percentage and number of virus antigen-specific IFN-�-and TNF-�-producing CD4 and CD8 cells became similar tothe latter (Fig. 6D, lower row). On day 32 post-virus infection,the tetramer-positive cells were of CD44hi memory T cell type(data not shown), and this is consistent with our earlier findingsthat memory T cell development in dKO mice seems unim-peded (Fig. 4C). On day 8 post-infection, virus-specific cyto-toxic T cell activity among total spleen cells from the dKOmicewas significantly lower than that of controlmice (Fig. 6E). How-

ever, on a per cell basis, the cytotoxic T cell activity was similarbetween the dKO and controls (data not shown), suggestingthat the reduced cytotoxic T cell activity among spleen cells wasmainly due to reduced number of virus-specific CD8 cells (Fig.6C) in the dKO spleen. LCMVvirus clearancewasmonitored atdifferent days following virus infection. The virus was detectedin blood 7 days after infection but was cleared at 14 days in bothdKO and control mice (Table 3).Our additional results indicate that the dKO mice also had

compromised cardiac allograft rejection response (supplemen-tal Fig. 3C). Results from the virus infection along with allograftrejection show that Efnb1 and Efnb2 are relevant in T cellimmune response in vivo. It is to be noted that the heavy redun-dancy in the immune system and the remaining T cell capacity,some of which might depend on other Eph/Efn molecules andsome of which might be from T cells escaped from the incom-plete Lck-Cre-mediated deletion, are sufficient to clear theLCMV infection eventually.Implication of Efnb1 and Efnb2 in IL-6 Signaling—IL-6 plays

important roles in thymocyte development (42), Th17 develop-ment (43), and viral immunity (42, 44). The decreased in vitroTh17 cell differentiation and defective anti-LCMV immuneresponses of dKO mice prompted us to investigate whether alack of Efnb1 and Efnb2 in dKO T cells compromised IL-6 sig-naling. IL-6R� expression was first measured in T cells, but nosignificant difference was revealed between dKO and control Tcells in the absence or presence of IL-6 (data not shown). How-ever, in dKO thymocytes and T cells, IL-6-induced STAT3phosphorylation, which is a critical event in the IL-6 signalingpathway, was significantly diminished compared with that incontrol T cells (Fig. 7A). A detailed flow cytometry analysisshowed that CD4, CD8, naive T cell, and memory type T cellsfrom dKO spleens all had diminished STAT3 phosphorylationupon IL-6 stimulation (Fig. 7B). This result indicates that Efnb1and Efnb2 are necessary to maintain proper IL-6 signaling.Compromised IL-6 signaling caused by Efnb1 and Efnb2 dele-tion is likely one of the mechanisms for many of the observedphenotypes in the dKO mice.

DISCUSSION

We have shown that Efnb1 and Efnb2 have critical functionsin thymocyte development, aswell as peripheral T cell function.They are also involved in the antiviral immune responses invivo. We present evidence that Efnb1 and Efnb2 are necessary

TABLE 3LCMV virus detection in mouse blooddpi means days post-infection. The virus titers (from � to ��) in the blood were determined by the focus forming assay.

Mouse strain Mouse ID 0 dpi 2 dpi 4 dpi 7 dpi 14 dpi 23 dpi 28 dpi 32 dpi

C57BL/6 1 � � � � � � � �2 � � � � � � � �

WT 1 � � � �� 2 � � � � � � � �3 � � � � � � � �4 � � � � � � � �5 � � � � � � � �6 � � � ��

dKO 1 � � � � � � � �2 � � � � � � � �3 � � � �� 4 � � � ��




ww

.jbc.org/D

ownloaded from



http://www.jbc.org/

for proper IL-6 signaling, which is essential in the observeddKO phenotype.Because of the promiscuous interaction mostly among Ephs

and Efnbs, different Ephs and Efnbs have overlapping func-tions; as a result, deletion of only one of themmight not alwaysreveal discernable phenotypes in the immune system. Indeed,our previous study showed that although Ephb6 null mutationinmice resulted in compromised peripheral T cell function, theanimals had no obvious defect in thymus structure or thymo-cyte development (29). We also examined thymus size, struc-ture, and cellularity of Lck-Cre-Efnb1f/f and Lck-Cre-Efnb2f/fmice but found no drastic defects (data not shown).When bothEfnb1 and Efnb2 in the T cell compartment were null mutated,a significant reduction in thymus size and cellularity wasobserved. This clearly indicates that Efnb1 and Efnb2 haveoverlapping functions and could complement each other if oneismissing. It is to be noted that deletion of Efnb1 or Efnb2 alonedid not induce compensative up-regulation of the other (datanot shown).

The Cre recombinase driven by the proximal Lck promoterbecomes most active at the DN3 and DN4 stage, and the pro-moter is no longer active after the SP stage (34, 35). However,the Efnb1 and Efnb2 deletion in the DN3 and DN4 stages led totheir effective deletion in subsequent offspring T cell popula-tions in the dKO thymus and spleen. Although the deletionin the DN3 and DN4 stages was less than complete, the cellsthat escaped the deletion never became a significant populationin the subsequent stages. The functional defects detected in theperipheral T cells in terms of cytokine production, T cellhomeostatic expansion, antiviral responses, and IL-6R signal-ing confirm that most of the T cells are not escapees of unde-leted cells from the thymus, and our model is sufficient anduseful to elucidate the role of these two molecules in the T cellcompartment.In some biological systems, Cre expressionmight interfere

with function of certain cell types, but in our system this isnot an issue, as our Efnb1f/f and Efnb2f/fmice with Lck-drivenCre expression manifested no abnormalities in terms of thy-

FIGURE 7. Compromised IL-6 signaling in dKO T cells. A, compromised STAT3 phosphorylation upon IL-6 stimulation in dKO thymocytes and T cells. dKO andWT control thymocytes (left panel) and spleen T cells (right panel) were stimulated with IL-6 (50 ng/ml) for 5 min, and total and phosphorylated STAT3 (p-STAT3)in the cells were analyzed by immunoblotting. Immunoblotting images from representative experiments are shown. The bar graph at right illustrates the ratiosof p-STAT3 versus total STAT3 signals in spleen T cells according to densitometry based on results of a total of four independent experiments. The statisticallysignificant difference is indicated (Student’s t tests). B, compromised STAT3 phosphorylation upon IL-6 stimulation in naive and memory type spleen T cellsfrom dKO mice. Spleen T cells were stimulated with IL-6 (50 ng/ml) from 5 min, and phosphorylated STAT3 in the cells gated on CD4, CD8, naive(CD44loCD62Lhi), and memory (CD44hiCD62Llo) T cells was analyzed by intracellular staining followed by flow cytometry. The mean intensity of fluorescenceintensity (MIF) of each population was indicated. The experiments in A and B were repeated twice, and representative histograms are shown.




ww

.jbc.org/D

ownloaded from

http://www.jbc.org/

mocyte and T cell subpopulations and functions (data notshown).Efnb1 and Efnb2 are capable of both forward signaling by

stimulating Ephs, mostly Ephb members, and reverse signalingby accepting Ephb stimulation. In the dKO thymus, the stromacells still have normal Efnb1 and Efnb2 expression and can pro-vide forward signaling to the developing dKO thymocytesthrough their Ephbs, but obviously such forward signaling isnot sufficient. Consequently, the compromised dKO thymo-cyte development is an indication that reverse signaling isimportant to this process. The thymus is packed with thymo-cytes, and most contacts that a thymocyte receives are fromfellow thymocytes. As a result, one could argue that in the dKOthymus, thymocyte Ephbs might not receive enough forwardsignaling from neighboring thymocytes that lack Efnb1 andEfnb2. To address this concern, dKO bone marrow cells werecotransplanted with wild type B6.SJL bonemarrow cells at a 1:1ratio into irradiated recipients (C57BL/6�B6.SJL F1). In thissystem, the competitor B6.SJL bone marrow cells as well asrecipient thymic stroma cells had normal Efnb1 and Efnb2expression. The dKO thymocytes should have had sufficientforward signaling from the Efnb1 and Efnb2 of neighboringcompeting B6.SLJ-derived thymocytes as well as from C57BL/6�B6.SJL F1 stroma cells.However, they failed to competewiththe B6.SJL-derived bone marrow cells; consequently, there wasa significant decrease in the number of thymocytes and periph-eral T cells. This indicates that Efnb1 and Efnb2 reverse signal-ing but not forward signaling is largely responsible for thedefect observed in dKO mice in terms of reduction of variousthymocyte subpopulations and peripheral T cells.Such reverse signaling was also critical inmature T cell func-

tion. Naive dKO CD4 cells in the periphery were defective intheir ability to differentiate into Th1 and Th17. Under Th1 andTh17 culture conditions, Efnb1- and Efnb2-expressing WTfeeder cells were present, so the dKO naive CD4 T cells werenot lacking forward signaling, especially when considering thatthe ratio of the feeder cells to naive T cells was 5:1. Thus, thedefective Th1 and Th17 development in vitro seen in this studyis also mainly the consequence of a lack of reverse signalingthrough Efnb1 and Efnb2.Although we showed with the above-described two experi-

mental models that reverse signaling from Eph through Efnb1/Efnb2 is essential for thymocyte development and peripheral Tcell function, forward signaling from Efnb1 and Efnb2 throughEph could still play a role, maybe a minor one, in promotingthymocyte survival and T cell function. This is evidenced in ourprevious study in that forward signaling of Efnb1 could enhancethymocyte survival (45), and Efnb1 and Efnb2 could costimu-late T cells and enhance lymphokine production by forwardsignaling (25, 27).We have shown that in Efnb1 and Efnb2 dKO mice, the

peripheral naive T cell population was more affected thanmemory T cell population in terms of absolute cell number aswell as percentage. We also found that the population of anti-gen-specific IFN�- and TNF�-secreting memory type CD8cells in LCMV-infected dKO and WT mice on day 32 post-infectionwas similar (data not shown), although initially on day8 post-infection, dKO mice showed reduced antigen-specific

IFN�- and TNF�-secreting CD8 cells (Fig. 6D). This suggestsagain that Efnb1 and Efnb2 affected less the memory type Tcells in terms of function. Additional studies on secondaryimmune response in dKO mice to assess the function of bothCD4 and CD8 memory T cells are warranted to confirm suchsuggestive indications. Overall, our data are consistent with thereport by Almeida et al. (46) in that reduced thymic output of Tcells preferentially affects the naive T cell pool but exerts lesseffect on thememory T cells, probably because of a better com-pensatory mechanism than the latter.We observed that the reduction of peripheral dKO T cells in

the whole body irradiation-BMTx model was far more signifi-cant than in dKOmice. This suggests that dKO thymocytes andT cells are significantly inferior to theirwild type competitors intheir development in the thymus and subsequently theirperipheral homeostatic expansion. Such inferiority was lessobvious when dKO T cell progenitors were allowed to developwithout competitors in the thymus and periphery in the dKOmice, because they could take all the time and space to developat their own pace.In thymocytes and peripheral T cell compartment, deletion

of Efnb1 and Efnb2 led to increased cell death according to exvivo annexin V staining, but this minimally affected their pro-liferation based on BrdU uptake. However, when dKO T cellswere transplanted into irradiated B6.SJL recipients, theirhomeostatic proliferation was significantly lower than WTcontrol T cells. The T cell homeostatic expansion in irradiatedmice is much faster than inmice without irradiation. Efnb1 andEfnb2 probably play a more critical role during the fast expan-sion, either by their direct effect or indirectly via modulatingcytokine signaling pathways as will be discussed below.Efnbs have short intracellular tails, which do not possess any

enzymatic activities but have six tyrosine residues. Some of thetyrosine residues are rapidly phosphorylated upon engagementof Ephbs (47). The phosphorylated tyrosines then bind to adap-tor proteins containing Src homology 2 domains such as Grab4(48) and Disheveled (49). These proteins in turn interact withdifferent signaling pathways and transduce signals to achievebiological effects on the cells. Near the C terminus of Efnb1,there is a conserved YXXQ motif. Recently, Bong et al. (50)reported that once the tyrosine in this motif is phosphorylated,it is capable of binding the Src homology 2 domain-containingprotein STAT3. Such association leads to STAT3 phosphory-lation and activation in a JAK2-dependentmanner; JAK2 canbefound in the Efnb1-STAT3 complex. It is to be noted that theintracellular tail of Efnb2 also contains the YXXQ motif,although Efnb2 is less potent than Efnb1 in leading to STAT3phosphorylation (49). STAT3 is in the IL-6 signaling pathway,as gp130, the signaling subunit of IL-6R complex, contains theYXXQ motif that binds to STAT3. The binding of IL-6 toIL-6R� results in the formation of a hexamer containing twomolecules of IL-6, twomolecules of IL-6R�, and twomoleculesof gp130. This complex activates JAK, which in turn phosphor-ylates gp130-associated STAT3 and leads to STAT3 activation(51). Our data demonstrate that in the absence of Efnb1 andEfnb2, IL-6-induced STAT3 phosphorylation was significantlydampened. This suggests that Efnb1 and Efnb2 act in an addi-tive or synergistic fashion to enhance IL-6 signaling through




ww

.jbc.org/D

ownloaded from

http://www.jbc.org/

STAT3 phosphorylation; in other words, the full STAT3 acti-vation, and hence the full-strength IL-6R activation, requiresthe participation of both IL-6R signaling and Efnb1/Efnb2reverse signaling.Indeed,many phenotypes observed in the dKOmice could be

attributed, at least to some extent, to reduced IL-6 signaling inthe T cell compartment. We observed reduced thymocyte andperipheral T cell numbers in dKOmice, and this is also the casein IL-6 KOmice (42). The dKOCD4 cells presented diminishedTh17 but not Treg differentiation in vitro; the former but notthe latter requires the presence of exogenous IL-6 in the culturesystem (43). The dKO T cells had compromised homeostaticexpansion; in agreement with this finding, Zasragoza et al. (52)showed that IL-6 is required for the survival of naive T cellsduring such expansion. Our dKO manifested compromisedantiviral immunity against LCMV; similarly, IL-6 KO miceshow defective immune responses against vesicular stomatitisand vaccinia viruses (42); recently, Pellegrini et al. (44) havedemonstrated that IL-7 can enhance immunity against chronicLCMV infection and that an important part of this effect is viaIL-6 signaling.It is prudent to say that it is entirely possible that the mech-

anisms of action of Efnb1 and Efnb2 in the T cell compartmentare not restricted to dampening the IL-6 signaling via directacting of Efnb1 and Efnb2 on STAT3 phosphorylation. As dis-cussed above, forward signaling from Efnb1 and Efnb2 onT cells to neighboring T cells or other immune-related cellslikely contributes to the integrity of an immune response. Efnbson T cells might also affect the function of other signaling path-ways by associating with other adaptor proteins in addition toSTAT3. Efnbs are also known to bind to other cell surface mol-ecules (53), and such binding can influence their functions ofthose molecules.Although most documented functions of Ephs and Efns are

related to pattern formation, the defective T cell differentiationand function in the dKO mice obviously have little to do withsuch classical functions of Eph and Efns. This indicates thatEfnb1 and Efnb2 have previously underappreciated functionsthat go beyond their classical duty.

REFERENCES1. Eph Nomenclature Committee (1997) Cell 90, 403–4042. Wilkinson, D. G. (2000) Int. Rev. Cytol. 196, 177–2443. Flanagan, J. G., and Vanderhaeghen, P. (1998) Annu. Rev. Neurosci. 21,

309–3454. Wang, H. U., Chen, Z. F., and Anderson, D. J. (1998) Cell 93, 741–7535. Noren, N. K., and Pasquale, E. B. (2007) Cancer Res. 67, 3994–39976. Batlle, E., Henderson, J. T., Beghtel, H., van den Born, M. M., Sancho, E.,

Huls, G., Meeldijk, J., Robertson, J., van de Wetering, M., Pawson, T., andClevers, H. (2002) Cell 111, 251–263

7. Dravis, C., Yokoyama, N., Chumley, M. J., Cowan, C. A., Silvany, R. E.,Shay, J., Baker, L. A., and Henkemeyer, M. (2004) Dev. Biol. 271,272–290

8. Konstantinova, I., Nikolova, G., Ohara-Imaizumi, M., Meda, P., Kucera,T., Zarbalis, K., Wurst, W., Nagamatsu, S., and Lammert, E. (2007) Cell129, 359–370

9. Davy, A., Bush, J. O., and Soriano, P. (2006) PLoS. Biol. 4, e31510. Zhao, C., Irie, N., Takada, Y., Shimoda, K., Miyamoto, T., Nishiwaki, T.,

Suda, T., and Matsuo, K. (2006) Cell Metab. 4, 111–12111. Pasquale, E. B. (2008) Cell 133, 38–5212. Pasquale, E. B. (2005) Nat. Rev. Mol. Cell Biol. 6, 462–475

13. Arvanitis, D., and Davy, A. (2008) Genes Dev. 22, 416–42914. Hashimoto, T., Karasawa, T., Saito, A., Miyauchi, N., Han, G. D., Haya-

saka, K., Shimizu, F., and Kawachi, H. (2007) Kidney Int. 72, 954–96415. Dravis, C., Wu, T., Chumley, M. J., Yokoyama, N., Wei, S., Wu, D. K.,

Marcus, D. C., and Henkemeyer, M. (2007) Hear. Res. 223, 93–10416. Wu, J., and Luo, H. (2005) Curr. Opin. Hematol. 12, 292–29717. Wohlfahrt, J. G., Karagiannidis, C., Kunzmann, S., Epstein, M.M., Kempf,

W., Blaser, K., and Schmidt-Weber, C. B. (2004) J. Immunol.172, 843–85018. Freywald, A., Sharfe, N., Miller, C. D., Rashotte, C., and Roifman, C. M.

(2006) J. Immunol. 176, 4066–407419. Holen, H. L., Nustad, K., and Aasheim, H. C. (2010) J. Leukocyte Biol. 87,

1059–106820. Foster, K. E., Gordon, J., Cardenas, K., Veiga-Fernandes, H., Makinen, T.,

Grigorieva, E.,Wilkinson,D.G., Blackburn, C. C., Richie, E.,Manley,N. R.,Adams, R. H., Kioussis, D., and Coles, M. C. (2010) Proc. Natl. Acad. Sci.U.S.A. 107, 13414–13419

21. Muñoz, J. J., Alfaro, D., García-Ceca, J., Alonso-C, L. M., Jiménez, E., andZapata, A. (2006) J. Immunol. 177, 804–813

22. Sharfe, N., Freywald, A., Toro, A., Dadi, H., and Roifman, C. (2002) Eur.J. Immunol. 32, 3745–3755

23. Aasheim, H. C., Delabie, J., and Finne, E. F. (2005) Blood 105, 2869–287624. Smith, L. M., Walsh, P. T., Rüdiger, T., Cotter, T. G., Mc Carthy, T. V.,

Marx, A., and O’Connor, R. (2004) Exp. Cell Res. 292, 295–30325. Yu, G., Luo, H., Wu, Y., and Wu, J. (2003) J. Immunol. 171, 106–11426. Yu,G., Luo,H.,Wu, Y., andWu, J. (2003) J. Biol. Chem. 278, 47209–4721627. Yu, G., Luo, H.,Wu, Y., andWu, J. (2004) J. Biol. Chem. 279, 55531–5553928. Luo, H., Yu, G., Wu, Y., and Wu, J. (2002) J. Clin. Invest. 110,

1141–115029. Luo, H., Yu, G., Tremblay, J., and Wu, J. (2004) J. Clin. Invest. 114,

1762–177330. Altman, J. D., Moss, P. A., Goulder, P. J., Barouch, D. H., McHeyzer-

Williams, M. G., Bell, J. I., McMichael, A. J., and Davis, M. M. (1996)Science 274, 94–96

31. Zhang, J.,Wang, X., Fahmi, H.,Wojcik, S., Fikes, J., Yu, Y.,Wu, J., and Luo,H. (2009) J. Immunol. 183, 5350–5357

32. Lacasse, P., Denis., J., Lapointe, R., Leclerc, D., and Lamarre, A. (2008)J. Virol. 82, 785–794

33. Compagni, A., Logan, M., Klein., R., and Adams, R. H. (2003) Dev. Cell 5,217–230

34. Orban, P. C., Chui, D., andMarth, J. D. (1992) Proc. Natl. Acad. Sci. U.S.A.89, 6861–6865

35. Allen, J. M., Forbush, K. A., and Perlmutter, R. M. (1992) Mol. Cell. Biol.12, 2758–2768

36. DeJarnette, J. B., Sommers, C. L., Huang, K., Woodside, K. J., Emmons, R.,Katz, K., Shores, E. W., and Love, P. E. (1998) Proc. Natl. Acad. Sci. U.S.A.95, 14909–14914

37. Azzam, H. S., DeJarnette, J. B., Huang, K., Emmons, R., Park, C. S., Som-mers, C. L., El-Khoury, D., Shores, E.W., and Love, P. E. (2001) J. Immunol.166, 5464–5472

38. Kishimoto, H., and Sprent, J. (1997) J. Exp. Med. 185, 263–27139. Bendelac, A., Matzinger, P., Seder, R. A., Paul, W. E., and Schwartz, R. H.

(1992) J. Exp. Med. 175, 731–74240. Rodgers, J. R., and Cook, R. G. (2005) Nat. Rev. Immunol. 5, 459–47141. Berg, L. J. (2007) Nat. Rev. Immunol. 7, 479–48542. Kopf, M., Baumann, H., Freer, G., Freudenberg, M., Lamers, M.,

Kishimoto, T., Zinkernagel, R., Bluethmann, H., and Köhler, G. (1994)Nature 368, 339–342

43. Bettelli, E., Carrier, Y., Gao,W., Korn, T., Strom, T. B., Oukka,M.,Weiner,H. L., and Kuchroo, V. K. (2006) Nature 441, 235–238

44. Pellegrini, M., Calzascia, T., Toe, J. G., Preston, S. P., Lin, A. E., Elford,A. R., Shahinian, A., Lang, P. A., Lang, K. S.,Morre,M., Assouline, B., Lahl,K., Sparwasser, T., Tedder, T. F., Paik, J. H., DePinho, R. A., Basta, S.,Ohashi, P. S., and Mak, T. W. (2011) Cell 144, 601–613

45. Yu, G., Mao, J., Wu, Y., Luo, H., and Wu, J. (2006) J. Biol. Chem. 281,10222–10229

46. Almeida, A. R., Borghans, J. A., and Freitas, A. A. (2001) J. Exp. Med. 194,591–599

47. Kalo, M. S., Yu, H. H., and Pasquale, E. B. (2001) J. Biol. Chem. 276,




ww

.jbc.org/D

ownloaded from

http://www.jbc.org/

38940–3894848. Cowan, C. A., and Henkemeyer, M. (2001) Nature 413, 174–17949. Lee, H. S., Bong, Y. S., Moore, K. B., Soria, K., Moody, S. A., and Daar, I. O.

(2006) Nat. Cell Biol. 8, 55–6350. Bong, Y. S., Lee, H. S., Carim-Todd, L., Mood, K., Nishanian, T. G., Tes-

sarollo, L., and Daar, I. O. (2007) Proc. Natl. Acad. Sci. U.S.A. 104,

17305–1731051. Taga, T., and Kishimoto, T. (1997) Annu. Rev. Immunol. 15, 797–81952. Zaragoza, B., Evaristo, C., Kissenpfennig, A., Libri, V., Malissen, B., Rocha,

B., Freitas, A. A., and Almeida, A. R. (2011) PLoS One 6, e1742353. Freywald, A., Sharfe, N., and Roifman C. M. (2002) J. Biol. Chem. 277,

3823–3828




ww

.jbc.org/D

ownloaded from

http://www.jbc.org/

Terra, Alain Lamarre and Jiangping WuHongyu Luo, Tania Charpentier, Xuehai Wang, Shijie Qi, Bing Han, Tao Wu, Rafik

Interleukin-6 (IL-6) SignalingDifferentiation, and Antiviral Immune Responses and Are Essential for

Efnb1 and Efnb2 Proteins Regulate Thymocyte Development, Peripheral T Cell

doi: 10.1074/jbc.M111.302596 originally published online October 5, 20112011, 286:41135-41152.J. Biol. Chem.

10.1074/jbc.M111.302596Access the most updated version of this article at doi:

Alerts:

When a correction for this article is posted•

When this article is cited•

to choose from all of JBC's e-mail alertsClick here

Supplemental material:

http://www.jbc.org/content/suppl/2011/10/05/M111.302596.DC1

http://www.jbc.org/content/286/48/41135.full.html#ref-list-1

This article cites 53 references, 24 of which can be accessed free at


ww

.jbc.org/D

ownloaded from

http://www.jbc.org/lookup/doi/10.1074/jbc.M111.302596

http://www.jbc.org/cgi/alerts?alertType=citedby&addAlert=cited_by&cited_by_criteria_resid=jbc;286/48/41135&saveAlert=no&return-type=article&return_url=http://www.jbc.org/content/286/48/41135

http://www.jbc.org/cgi/alerts?alertType=correction&addAlert=correction&correction_criteria_value=286/48/41135&saveAlert=no&return-type=article&return_url=http://www.jbc.org/content/286/48/41135

http://www.jbc.org/cgi/alerts/etoc

http://www.jbc.org/content/suppl/2011/10/05/M111.302596.DC1

http://www.jbc.org/content/286/48/41135.full.html#ref-list-1

http://www.jbc.org/

Efnb1andEfnb2ProteinsRegulateThymocyteDevelopment ... ·...

Documents

Transcript of Efnb1andEfnb2ProteinsRegulateThymocyteDevelopment ... ·...