Human Naive CD8 T Cells Down-Regulate Expression of the WNT Pathway Transcription Factors Lymphoid Enhancer Binding Factor 1 and 7 This information is current as (T Cell Factor-1) following Antigen of September 29, 2021. Encounter In Vitro and In Vivo Tim Willinger, Tom Freeman, Mark Herbert, Hitoshi Hasegawa, Andrew J. McMichael and Margaret F. C. Callan Downloaded from J Immunol 2006; 176:1439-1446; ; doi: 10.4049/jimmunol.176.3.1439 http://www.jimmunol.org/content/176/3/1439 http://www.jimmunol.org/ References This article cites 32 articles, 12 of which you can access for free at: http://www.jimmunol.org/content/176/3/1439.full#ref-list-1

Why The JI? Submit online.

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

by guest on September 29, 2021 • No Triage! Every submission reviewed by practicing scientists

• Fast Publication! 4 weeks from acceptance to publication

*average

Subscription Information about subscribing to The Journal of Immunology is online at: http://jimmunol.org/subscription Permissions Submit copyright permission requests at: http://www.aai.org/About/Publications/JI/copyright.html Email Alerts Receive free email-alerts when new articles cite this article. Sign up at: http://jimmunol.org/alerts

The Journal of Immunology is published twice each month by The American Association of Immunologists, Inc., 1451 Rockville Pike, Suite 650, Rockville, MD 20852 Copyright © 2006 by The American Association of Immunologists All rights reserved. Print ISSN: 0022-1767 Online ISSN: 1550-6606. The Journal of Immunology

Human Naive CD8 T Cells Down-Regulate Expression of the WNT Pathway Transcription Factors Lymphoid Enhancer Binding Factor 1 and Transcription Factor 7 (T Cell Factor-1) following Antigen Encounter In Vitro and In Vivo1

Tim Willinger,2* Tom Freeman,† Mark Herbert,‡ Hitoshi Hasegawa,§ Andrew J. McMichael,* and Margaret F. C. Callan¶

The transcription factors lymphoid enhancer binding factor 1 (LEF1) and transcription factor 7 (TCF7) (T cell factor-1 (TCF-1)) are downstream effectors of the , which is a critical regulator of T cell development in the thymus. In this

study, we show that LEF1 and TCF7 (TCF-1) are not only expressed in thymocytes, but also in mature T cells. Our data Downloaded from demonstrate that Ag encounter in vivo and engagement of the TCR or IL-15 in vitro leads to the down-regulation of LEF1 and TCF7 (TCF-1) expression in human naive CD8 T cells. We further show that resting T cells preferentially express inhibitory LEF1 and TCF7 (TCF-1) isoforms and that T cell activation changes the isoform balance in favor of stimulatory TCF7 (TCF-1) isoforms. Altogether, our study suggests that involved in the WNT signaling pathway not only regulate T cell develop- ment, but also peripheral T cell differentiation. The Journal of Immunology, 2006, 176: 1439–1446. http://www.jimmunol.org/ he differentiation of T lymphocytes is mainly controlled currently unknown whether the WNT pathway has any role in by signals through their Ag-specific TCR. In the thymus, mature T cells. T TCR signals regulate the development of a mature T cell The canonical WNT signaling pathway is a critical regulator of repertoire via positive and negative selection. In the periphery, stem cell function, e.g., it controls the maintenance and self-re- 3 naive T cells (TN) further differentiate into effector and memory newal of hemopoietic stem cells (8–11). Furthermore, dysregula- T cells when recognizing foreign Ags derived from infectious tion of the WNT pathway commonly occurs in human cancers pathogens via their TCR (1). Apart from TCR signals and cyto- (11). In the absence of a WNT signal, cytoplasmic ␤-catenin is kines, other factors contributing to peripheral T cell differentiation phosphorylated and targeted for degradation by the proteasome are not well characterized at present. (3, 11). WNT signaling allows ␤-catenin to escape proteasomal by guest on September 29, 2021 In addition to TCR signals, major developmental pathways such degradation and to translocate to the nucleus. In the nucleus, as the WNT, Notch, and Hedgehog signaling pathways influence T ␤-catenin interacts with members of the lymphoid enhancer cell development (2–4). Interestingly, there is recent evidence that binding factor (LEF)/T cell factor (TCF) family of transcription these major developmental pathways also control the differentia- factors (LEF1, TCF7 (TCF-1), TCF7L1 (TCF-3), and TCF7L2 tion of peripheral T cells. Thus, Notch regulates the decision of (TCF-4)) to activate the transcription of WNT target such CD4 T cells between the Th1 vs the Th2 fate (5), while the Hedge- as c- and cyclin D1 (11). In the absence of WNT signaling, hog pathway can influence the proliferation and cytokine produc- i.e., when not interacting with ␤-catenin, LEF/TCF family tion of human peripheral CD4 T cells (6, 7). In contrast, it is members act as transcriptional repressors by recruiting Groucho repressor proteins (12). LEF1 and TCF7 (TCF-1) share common motifs, in par- *Medical Research Council Human Immunology Unit, Weatherall Institute of Mo- lecular Medicine, Oxford, United Kingdom; †Medical Research Council Rosalind ticular, the C-terminal HMG domain of both proteins is responsi- Franklin Centre for Genomics Research, Wellcome Trust Genome Campus, Hinxton, ble for DNA binding, while a ␤-catenin-binding domain at the N Cambridge, United Kingdom; ‡The Neonatal Unit, Women’s Centre, John Radcliffe ␤ Hospital, Oxford, United Kingdom; §University School of Medicine, Shigenobu, terminus mediates the interaction with -catenin (12). Interest- Ehime, Japan; and ¶Department of Rheumatology, Division of Medicine, Imperial ingly, there are multiple LEF1 and TCF7 (TCF-1) protein isoforms College London, Chelsea and Westminster Hospital, London, United Kingdom with distinct functional properties (13–17). In addition to the stim- Received for publication August 31, 2005. Accepted for publication November ulatory full-length isoforms, there are N-terminally truncated iso- 14, 2005. forms that are without the ␤-catenin-binding domain (referred to as The costs of publication of this article were defrayed in part by the payment of page ⌬ charges. This article must therefore be hereby marked advertisement in accordance CTNNB) but retain the ability to interact with Groucho repres- with 18 U.S.C. Section 1734 solely to indicate this fact. sors (12). Importantly, these truncated isoforms can function in a 1 This work was supported by Grants from the Medical Research Council (U.K.) and dominant-negative manner in the WNT signaling pathway as has Leukaemia Research Fund. T.W. is a Medical Research Council Clinical Research been demonstrated for ⌬CTNNB isoforms of LEF1 (17, 18), TCF7 Fellow, and M.F.C.C. is a Medical Research Council Senior Clinical Fellow. (TCF-1) (19, 20), and the Xenopus homolog of LEF1/TCF7 2 Address correspondence and reprint requests to Dr. Tim Willinger, Medical Re- search Council Human Immunology Unit, Weatherall Institute of Molecular Medi- (TCF-1) (21). Finally, there are also LEF1 and TCF7 (TCF-1) cine, John Radcliffe, Oxford, U.K. E-mail address: [email protected] isoforms with alternative C-termini (termed tails) known as N- or 3 Abbreviations used in this paper: TN, naive T cell; CB, cord blood; LEF1, lymphoid B-tailed isoforms. Currently, very little is known about the differ- enhancer binding factor 1; qRT-PCR, quantitative RT-PCR; TCF7 (TCF-1), tran- ent functional properties of these multiple LEF1 and TCF7 scription factor 7 (T cell factor 1); TCM, central memory T cell; TEM, effector memory T cell; TEMRA, effector memory RA T cell; CD62L, CD62 ligand. (TCF-1) isoforms in T cells.

Copyright © 2006 by The American Association of Immunologists, Inc. 0022-1767/06/$02.00 1440 LEF1 AND TCF7 (TCF-1) EXPRESSION IN PERIPHERAL T CELLS

Knockout and transgenic studies in mice have clearly shown a TGGTCAGG; reverse, AGTCTGGCTTATATCCAACACTTCG; probe: redundant and ␤-catenin-dependent role of LEF1 and TCF7 FAM-TTTCACCAGCAAGCTTGCGACCTTGAC-TAMRA. All probes (TCF-1) in T cell development (3). TCF7 (TCF-1)Ϫ/Ϫ mice have span exon-intron junctions. We applied the comparative threshold cycle method for relative quantification of mRNA expression according to the impaired T cell development with a partial block at the interme- manufacturer’s recommendations. Validation experiments demonstrated diate single-positive to double-positive transition due to reduced that the amplification efficiencies of LEF1 and TCF7 (TCF-1) were equal thymocyte proliferation and survival (22–24). Although T cell de- to that of the endogenous control hypoxanthine phosphoribosyltransferase. velopment is normal in LEF1 Ϫ/Ϫ mice, B cell development is impaired (25). Thymocytes from TCF7 (TCF-1)Ϫ/Ϫ LEF1 Ϫ/Ϫ Intracellular FACS staining mice show a profound block at the intermediate single-positive We analyzed intracellular expression of LEF1 and TCF7 (TCF-1) in CD8 stage with neither double-positive nor single-positive thymocytes T cell subsets using the methanol permeabilization protocol as previously present, and consequently no mature T cells in the periphery (26). described (27). LEF1 and TCF7 (TCF-1) were detected by indirect staining ␤ using pretitrated mAbs REMB6 (Oncogene Research Products) and 7H3 Taken together, the WNT- -catenin-LEF1/TCF7 (TCF-1) axis (Upstate Biotechnology), respectively. Briefly, 2 ␮g of primary mAb was plays a pivotal role in T cell development. However, it is unknown added to 2 ϫ 106 cells and incubated for1hatroom temperature. After whether LEF1, TCF7 (TCF-1), and the WNT pathway have a spe- washing, this was followed by staining with PE-conjugated rabbit anti- cific function in peripheral T cells. mouse Ab (DakoCytomation) for1hatroom temperature. Cells were subsequently washed in blocking buffer (PBS containing 2% mouse serum) Therefore, we undertook a detailed analysis of the expression of before surface staining with directly conjugated mAbs specific for CD62 LEF1 and TCF7 (TCF-1) in human peripheral T cells. We found ligand (CD62L), CD45RA, and CD8 (all BD Biosciences). CD62L was ϩ that both LEF1 and TCF7 (TCF-1) are expressed in mature CD8 used instead of CCR7 as a surface marker because CCR7 staining was

compromised following methanol permeabilization (27). Downloaded from TN and that their expression is down-regulated following TCR or IL-15R engagement in vitro and Ag encounter in vivo. Further- Stimulation of cord blood (CB) CD8 T cells in vitro more, T cell activation changed the balance of stimulatory vs in- hibitory LEF1 and TCF7 (TCF-1) isoforms. Our results suggest We obtained CB samples from the John Radcliffe Hospital maternity unit, that the WNT pathway, in addition to its well-known role in T cell upon written consent and approval by the local Medical Ethics Committee. development, is likely to be involved in regulating peripheral T CB CD8 T cells were isolated by immunomagnetic selection as described above. Phenotyping was conducted using mAbs specific for CCR7 (R&D cell differentiation. Systems), CD45RA (BD Biosciences), CD8 (BD Biosciences), CD3, http://www.jimmunol.org/ CD25, and HLA class II (all DakoCytomation). We stimulated CB CD8 T Materials and Methods cells (1–2 ϫ 106/ml) with either plate-bound anti-CD3 mAb OKT3 (1 Isolation of CD8 T cell subsets ␮g/well), IL-15 (50 ng/ml), or TGF␤1 (3 ng/ml) in 24-well plates for the indicated time points. All cytokines were from obtained R&D Systems. For microarray and quantitative RT-PCR (qRT-PCR) experiments, CD8 T Cells were cultured in complete medium (RPMI 1640 supplemented with cell subsets were isolated from healthy donors in accordance with institu- 10% FCS, 1% sodium pyruvate, 2 mM L-glutamine, 100 U/ml penicillin, tional ethics approval as previously described (27). CD8 T cells were and 100 ␮g/ml streptomycin) at 37°C in 5% CO . Alternatively, CD8 T ϩ ϩ 2 6 sorted into either TN (CCR7 CD45RA ) and effector memory T cells/ cells (1 ϫ 10 ) were cocultured with irradiated allogeneic EBV-trans- Ϫ ϩ/Ϫ ϩ effector memory RA T cells (TEM/EMRA; CCR7 CD45RA ) (microarray formed B cells at a 1:1 ratio. From day 3 onward, CD3 cells represented ϩ ϩ data set 1), or TN (CCR7 CD45RA ), central memory T cells (TCM; Ͼ95% of live cells in these cocultures, i.e., the stimulator B cells had ϩ Ϫ Ϫ Ϫ by guest on September 29, 2021 CCR7 CD45RA ), TEM (CCR7 CD45RA ), and TEMRA practically disappeared. (CCR7ϪCD45RAϩ) populations (microarray data set 2). Microarray expression analysis Cloning of LEF1 mRNA isoforms from primary CD8 T cells RNA extraction and labeling was performed as previously described (27, Total RNA was extracted from primary CD8 T cells using TRI Reagent 28). For microarray data set 1, total RNA was pooled from several donors (Sigma-Aldrich) and cDNA was prepared as described above for the qRT- (replicate pool 1: n ϭ 10; replicate pool 2: n ϭ 6) and hybridized to PCR experiments. RT-PCR was performed with the following primers: Affymetrix HG-U95Av2 arrays. Two independent microarray experiments forward primers, CAGCGGAGCTCAGATTACAGAG (full-length iso- ϩ forms) and ACTCGAGCTCTTCCGGGTACATAATG (⌬CTNNB iso- were performed with RNA from CD8 TN and TEM/EMRA for data set 1. For microarray data set 2 (described in Ref. 27), total RNA from individual forms); reverse primers, CTTCGAATTCCACCATGTTTCAGATG (N- donors was used and hybridized to Affymetrix HG-U133 plus 2.0 arrays tail isoforms) and GTCAGAATTCCTTTGGCGTCGACTG (B-tail (Affymetrix). Four independent microarray experiments were performed isoforms). PCR products were cloned into the vector pIRES2-EGFP (BD ϩ Clontech) followed by DNA sequencing. with RNA from CD8 TN,TCM,TEM, and TEMRA for data set 2. We used GCOS software (Affymetrix) and the software package BRB-ArrayTools for data analysis and the identification of differentially expressed genes Western blot analysis between CD8 T cells subsets (27). We prepared protein lysates from ϳ3to5ϫ 106 CD8 T cells by washing qRT-PCR analysis the cells in PBS and resuspending them in an equal volume of 2ϫ sample buffer (100 mM Tris-HCl (pH 6.8), 4% SDS, 8% 2-ME, 0.2% bromphenol qRT-PCR was conducted on cDNA from the indicated CD8 T cell popu- blue, and 20% glycerol). After sonication for 2 ϫ 20 s and boiling for 5–10 lations with the 5Ј nuclease/TaqMan assay. Briefly, we prepared cDNA min, protein samples were separated by SDS-PAGE and gels blotted to from ϳ1 ␮g of DNase-treated total RNA using the SuperScript First- nitrocellulose membranes. We performed immunodetection of LEF1 and Strand Synthesis System (Invitrogen Life Technologies). We then per- TCF7 (TCF-1) with the REMB6 mAb (Exalpha) at a 1/500 dilution and formed quantitative PCR in a final volume of 25 ␮l with 300 nM of the with the 7H3 mAb (Upstate Biotechnology) at a 1/1000 dilution, respec- forward and reverse primers and 100–250 nM of the fluorogenic TaqMan tively. This was followed by incubation with secondary HRP-conjugated probes (Eurogentec) using 2ϫ quantitative PCR Mastermix Plus (Euro- anti-mouse Ig (DakoCytomation) and signal detection with ECL reagent gentec). Reactions were run on an ABI Prism 7700 Sequence Detection (BD Amersham). Blots were stripped by incubating the membrane at 50°C System machine (Applied Biosciences) in triplicate (initial steps: 50°C/2 for 30 min in stripping buffer (62.5 mM Tris-HCl (pH 6.7), 2% SDS, and min and 95°C/10 min, followed by 40 cycles: 95°C/15 s and 60°C/1 min) 100 mM 2-MEl) and reprobed with anti-␤-actin mAb AC-15 at a 1/5000 The following primers and probes were used: 1) LEF1: forward, TGA dilution (Sigma-Aldrich). CAGCTGCCTACATCTGAAAC; reverse, GCTGCCTTGGCTTTGCAC; probe: FAM-TGGTGGAAAACGAAGCTCATTCCCAA-TAMRA. LEF1 Statistical analysis primers and probes target exon 11/exon 12 and are specific for all LEF1 N-tail isoforms. 2) TCF7 (TCF-1): forward, TGCAGCTATACCCAG A two-sample, two-tailed t test assuming unequal variances was used to GCTGG; reverse, CCTCGACCGCCTCTTCTTC; probe: FAM-TCCCG determine the significance of differences in mRNA expression between two TAGTTGTCCCGCGCTG-TAMRA. TCF7 (TCF-1) primers and probes groups (␣ ϭ 0.05). For multigroup comparisons, we applied one-way target exon 7/exon 8 and are specific for all TCF7 (TCF-1) isoforms. 3) ANOVA with post hoc testing using Tukey’s significant difference test hypoxanthine phosphoribosyltransferase: forward, GACTTTGCTTTCCT (␣ ϭ 0.05). The Journal of Immunology 1441

Results cytometry. Importantly, as shown in Fig. 2, TN (and TCM)ex- CD8ϩ T down-regulate expression of LEF1 and TCF7 (TCF-1) pressed significantly higher levels of both LEF1 and TCF7 N Ͻ upon Ag encounter in vivo (TCF-1) protein than TEM and TEMRA ( p 0.05, one-way ANOVA/Tukey’s post hoc test). In summary, we found that the We used microarray technology to screen for genes that are differen- WNT pathway effectors LEF1 and TCF7 (TCF-1) are expressed in tially expressed between human T and Ag-primed CD8 T cells. Our N peripheral T cells and that LEF1 is the most differentially ex- first, exploratory, microarray data set compared in pressed transcription factor between T and Ag-primed CD8 T cell purified CD8ϩ T (CCR7ϩCD45RAϩ) and CD8ϩ T N N EM/EMRA subsets. (CCR7ϪCD45RAϩ/Ϫ) populations using RNA pooled from several donors. We also generated a more detailed second data set that ana- ϩ ϩ ϩ ϩ Expression of LEF1 and TCF7 (TCF-1) in CD8 TN is down- lyzed the gene expression profiles of CD8 TN (CCR7 CD45RA ) ϩ ϩ Ϫ regulated by TCR signals and IL-15 in vitro in relation to that of CD8 TCM (CCR7 CD45RA ), TEM Ϫ Ϫ Ϫ ϩ (CCR7 CD45RA ), and TEMRA (CCR7 CD45RA ) subsets. Tech- We studied the regulation of LEF1 mRNA expression in peripheral nical advances allowed us to use RNA from individual donors instead CD8 T cells in vitro by qRT-PCR. We observed down-regulation of pooled RNA and to perform a greater number of replicate exper- of LEF1 mRNA expression in CD8 T cells in response to TCR iments (four instead of two) for the second data set. Furthermore, gene triggering (Fig. 3A). This down-regulation was rapid (within 12 h) chips (HG-U133 Plus 2.0 instead of HG-U95Av2) with a greater and persisted for Ͼ48 h. Stimulation with homeostatic cytokines, number of probes and greater coverage of the were such as IL-15, also lead to a persistent decrease in LEF1 mRNA available for this data set. Using this data set, we have previously levels (Fig. 3B), with IL-2 having a similar effect (data not shown). ␤ Downloaded from investigated the molecular relationships between TN and memory In contrast, stimulation with TGF 1 increased LEF1 expression in (Ag-primed) CD8 T cell subsets and the molecular basis for their CD8 T cells, compared with the medium control (Fig. 3C). different functional properties (27). We now aimed to examine se- To investigate the expression of LEF1 and TCF7 (TCF-1) in ϩ lected differentially expressed genes in more detail to gain further CD8 TN, we used CD8 T cells isolated from CB. CB CD8 T cells insight into the molecular basis of CD8 T cell behavior. have a predominant CCR7ϩCD45RAϩ phenotype similar to adult Ϫ ϩ We identified LEF1 as a gene highly expressed in TN, compared TN (Fig. 4A) with some CCR7 CD45RA cells also present. Im- with Ag-primed CD8 T cell subsets. In both microarray data sets, portantly, it has been shown that CB T cells (including the http://www.jimmunol.org/ LEF1 was found to be in the top three most differentially expressed CCR7ϪCD45RAϩ subset) are functionally naive (29). CB CD8 T genes when comparing the different mature CD8 T cell populations cells showed strong up-regulation of activation markers such as (Table I). Indeed, within data set 2, probe identifications for LEF1 CD25 (IL-2R␣; Fig. 4A) and HLA class II (data not shown) in accounted for two of the top three differentially expressed genes. response to TCR (allogeneic) stimulation. At later time points,

LEF1 mRNA levels were 5- to 10-fold higher in TN, compared poststimulation CD8 T cells converted back to a resting state as with the three Ag-experienced subsets, with TCM expressing LEF1 shown by the absence of activation marker expression (Fig. 4A). more strongly than TEM and TEMRA (Fig. 1A). Interestingly, we Similar to bulk peripheral CD8 T cells, we observed down-regu- ϩ found that TCF7 (TCF-1) was also expressed at higher levels in lation of LEF1 mRNA expression in CD8 TN (from CB) follow- by guest on September 29, 2021 TN, compared with TCM,TEM, and TEMRA (Fig. 1B). ing TCR stimulation as measured by qRT-PCR (Fig. 4B). Impor- qRT-PCR experiments confirmed the differential expression of tantly, at the time points examined (more than day 3) allogeneic

LEF1 and TCF7 (TCF-1) in TN and Ag-primed CD8 T cell subsets stimulator B cells had practically disappeared from the T cell-B (Fig. 1). We also examined LEF1 and TCF7 (TCF-1) protein ex- cell cocultures. Interestingly, after the initial down-regulation, pression in peripheral CD8 T cell subsets by intracellular flow there was a progressive increase in LEF1 mRNA expression at

a Table I. LEF1 is among the 10 most differentially expressed genes between TN and Ag-primed CD8 T cell subsets

No. GenBank Identification Gene Symbol Description TN TEM/EMRA Microarray data set 1 1 AB018295 NY-REN-7 NY-REN-7 Ag 93.0 6.6 2 M95178 ACTN1 Actinin ␣1 376.7 29.4 3 AL049409 LEF1 Lymphoid enhancer-binding factor-1 853.2 65.3 4 D90144 CCL3 MIP-1 ␣ 66.4 491.0 5 U11276 KLRB1 CD161 232.5 1155.1 6 L31584 CCR7 CCR7 1385.7 238.6 7 X76220 MAL T lymphocyte maturation-associated protein 1242.9 106.9 8 AJ223603 C4.4A GPI-anchored metastasis-associated protein homolog 367.0 41.2 9 AA478904 KLF7 Kruppel-like factor-7 189.2 38.0 10 X02910 TNF TNF-␣ 93.9 704.7

No. GenBank Identification Gene Symbol Description TN TCM TEM TEMRA Microarray data set 2 1 AA992805 LEF1 Lymphoid enhancer-binding factor-1 421.5 52.4 29.8 28.8 2 AI082078 ACTN1 Actinin, ␣1 928.9 47.4 25.5 41.4 3 AF294627 LEF1 Lymphoid enhancer-binding factor-1 1666.5 208.2 80.8 111.1 4 NM_014795 ZFHX1B Zinc finger 1b 20.8 127.0 405.6 708.2 5 NM_006144 GZMA Granzyme A 98.3 503.9 2240.4 2429.5 6 AI636759 SLC15A4 Solute carrier family 15, member 4 108.3 191.6 531.6 1020.8 7 AI356412 LYN v-yes-1 Yamaguchi sarcoma viral-related oncogene homolog 22.0 23.6 85.0 313.1 8 BC004344 MGC29816 Hypothetical protein MGC29816 1181.2 240.1 170.6 169.8 9 AI246687 CTSC Cathepsin C 286.7 529.8 1330.6 1541.0 10 NM_002835 PTPN12 Protein tyrosine phosphatase, non-receptor type 12 65.3 87.5 281.6 718.5

a Listed are the 10 most differentially expressed genes between CD8 T cell subsets. Mean expression values (data set 1: n ϭ 2; data set 2: n ϭ 4) are shown. 1442 LEF1 AND TCF7 (TCF-1) EXPRESSION IN PERIPHERAL T CELLS

FIGURE 2. Intracellular LEF1 and TCF7 (TCF-1) protein expression in

TN and Ag-primed CD8 T cell subsets. LEF1 (A) and TCF7 (TCF-1) (B) protein expression was detected in CD8 T cell subsets by intracellular FACS staining as described in Materials and Methods. Thin line histo- grams indicate isotype control staining, thick line histograms LEF1 (A)or

TCF7 (TCF-1) (B) staining. A representative example from one donor is Downloaded from shown. Similar results were obtained for CD8 T cell subsets from two other donors.

ϩ

decreased LEF1 and TCF7 (TCF-1) mRNA expression in CD8 http://www.jimmunol.org/

TN (from CB) (Fig. 4C). The level of LEF1 down-regulation (5- to 10-fold) was similar to that observed when comparing TN and CD8 T cells primed with Ag in vivo (Fig. 1A). Thus, the signals that control mature CD8 T cell differentiation, i.e., TCR triggering and homeostatic cytokines, also regulate the expression of LEF1 and ϩ TCF7 (TCF-1) in CD8 TN. TCR and IL-15 stimulation of peripheral CD8 T cells lead to a

relative decrease in inhibitory LEF1 and TCF7 (TCF-1) isoform by guest on September 29, 2021 expression Several LEF1 and TCF7 (TCF-1) isoforms with different func- tional properties have been described (13–21), but which of these different isoforms are expressed in CD8 T cells is unknown. There- fore, we analyzed LEF1 and TCF7 (TCF-1) isoform expression in mature T cells. First, cloning of LEF1 mRNA isoforms by RT- PCR and DNA sequencing showed that peripheral CD8 T cells express both stimulatory full-length and inhibitory ⌬CTNNB LEF1 mRNA isoforms (Fig. 5A). Furthermore, we found that CD8 T cells preferentially express N-tail LEF1 mRNA isoforms, al- though isoforms carrying a B-tail could also be detected (Fig. 5B). These mRNA isoforms were distinguished by the presence (N-tail isoforms) or absence (B-tail isoforms) of exon 11 (16). The pre- of N-tail isoforms applied to both full-length and ⌬CT- ϩ NNB LEF1 mRNA isoforms (Fig. 5B). FIGURE 1. CD8 TN down-regulate expression of LEF1 and TCF7 (TCF-1) mRNA upon Ag encounter in vivo. Upper panels, Expression of Second, we investigated LEF1 and TCF7 (TCF-1) protein iso-

LEF1 (A) and TCF7 (TCF-1) (B) mRNA as examined in TN form expression by Western blot. Jurkat cells, a transformed im- ϩ ϩ Ϫ ϩ/Ϫ (CCR7 CD45RA ) and TEM/TEMRA (CCR7 CD45RA ) CD8 T cell sub- mature T cell line with a high proliferative capacity, predominantly p Ͻ expressed stimulatory full-length LEF1 isoforms of 48–55 kDa ,ء .(sets by microarray analysis (data set 1, n ϭ 2) and qRT-PCR (n ϭ 4 0.05 (two-sample t test with unequal variances), compared with TN.Inthe (Fig. 6A) as previously reported (17). In contrast, primary resting second microarray data set (lower panels), LEF1 (A) and TCF7 (TCF-1) (B) ϩ ϩ CD8 T cells showed predominant expression of a 38-kDa LEF1 mRNA expression was determined in TN (CCR7 CD45RA ), TCM ϩ Ϫ Ϫ Ϫ protein isoform (Fig. 6A). It has previously been demonstrated that (CCR7 CD45RA ), TEM (CCR7 CD45RA ), and TEMRA Ϫ ϩ this 38-kDa LEF1 band corresponds to the ⌬CTNNB LEF1 iso- (CCR7 CD45RA ) subsets from four individual donors. LEF1 and TCF7 p Ͻ 0.05 (one-way form that has a dominant-negative function in the WNT signaling ,ء .TCF-1) mRNA expression relative to T is displayed) N pathway (17, 18). The predominance of inhibitory LEF1 protein ANOVA, Tukey’s significant difference post hoc test), compared with TN. ϩ isoforms also applied to resting CD8 TN (from CB; Fig. 6B, lane 1). In line with our mRNA expression results (Fig. 4), there was an later time points poststimulation, although it varied between do- overall down-regulation of LEF1 protein expression following nors. A further set of experiments corroborated these results: in TCR triggering or stimulation with IL-15 in vitro (Fig. 6B, lanes 2 addition to TCR triggering, homeostatic signals such as IL-15 also and 3). The Journal of Immunology 1443 Downloaded from http://www.jimmunol.org/

FIGURE 4. TCR and homeostatic signals inhibit expression of LEF1 ϩ and TCF7 (TCF-1) in CD8 TN. A, Characterization of CB CD8 T cells. CD8 T cells were isolated from CB and their CCR7/CD45RA phenotype examined by FACS. A representative example from one donor is shown (left panel). The right panel shows expression of the activation marker CD25 (IL-2R␣) in resting CD8 T cells (day (d), d0), thin line histogram) and in cells poststimulation with alloantigen (d6, thick line histogram; d14,

thin line histogram). B, LEF1 mRNA expression in response to allogeneic FIGURE 3. TCR signals and IL-15 down-regulate, while TGF␤1 up- ϩ by guest on September 29, 2021 stimulation of CD8 TN. CB CD8 T cells (A) were stimulated with irra- regulates LEF1 mRNA expression in CD8 T cells. Human CD8 T cells diated allogeneic EBV-transformed B cells and LEF1 mRNA expression were isolated by immunomagnetic selection and stimulated with either was determined by qRT-PCR at the indicated time points. Results from ,p Ͻ 0.05 (one-way ANOVA ,ء .plate-bound anti-CD3 mAb/OKT3 at 1 ␮g/well (A), IL-15 at 50 ng/ml (B), three independent experiments are shown or TGF␤1 at 3 ng/ml (C). At the indicated time points, LEF1 mRNA Tukey’s significant difference post hoc test), compared with day 0. C, Ex- expression was examined by qRT-PCR relative to the resting state (ϭ 0 h). pression of LEF1 and TCF7 (TCF-1) mRNA in response to alloantigen and p Ͻ 0.05 (two-sample t test with unequal variances), compared with ϩ ,ء IL-15. CD8 TN (from CB) were stimulated with either IL-15 at 50 ng/ml medium control. Two independent experiments with CD8 T cells from or with irradiated allogeneic EBV-transformed B cells for 5 days. Relative different donors were performed. expression of LEF1 and TCF7 (TCF-1) as determined by qRT-PCR is p Ͻ 0.05 (two-sample ,ء .shown for one of three independent experiments t test with unequal variances), compared with day 0. Similar to LEF1, we observed differences in the stimulatory vs inhibitory TCF7 (TCF-1) isoform balance between primary T cells and transformed Jurkat T cells. Jurkat cells showed predominant Discussion expression of stimulatory full-length TCF7 (TCF-1) isoforms of Ag-experienced T cells are better able than TN to respond to Ag. 42–48 kDa (Fig. 7A). In contrast, the ratio of stimulatory full- This reflects both an increase in their frequency and a change in length (42–48 kDa)/inhibitory ⌬CTNNB (26–32 kDa) TCF7 their state of differentiation. The molecular mechanisms that un- (TCF-1) isoforms (14, 15) was about equal in resting (Fig. 7A) and derpin the changes in cellular state and hence the development of ϩ CD8 TN (from CB; Fig. 7B, lane 1). Clevers and colleagues (15) T cell memory remain poorly understood. The capacity of tran- have previously demonstrated that the 26- to 32-kDa bands corre- scription factors to modulate many different aspects of T cell func- spond to ⌬CTNNB TCF7 (TCF-1) isoforms that have a dominant- tion makes them an attractive candidate for study in this respect. In negative function (19, 20). Interestingly, TCR and IL-15 stimula- two separate microarray studies, we observed that the WNT path- ϩ tion of CD8 TN (from CB) lead to preferential down-regulation way effector, LEF1, was expressed at higher levels in TN, com- of the inhibitory TCF7 (TCF-1) ⌬CTNNB isoforms (Fig. 7B, lanes pared with Ag-experienced CD8 T cells. Indeed, in both studies, it 2 and 3). Thus, the TCF7 (TCF-1) protein isoform balance in CD8 proved to be one of the most differentially expressed genes and the T cells changed in favor of the stimulatory isoforms following most differentially expressed transcription factor between these activation. In conclusion, our data suggest that the negative effects two populations of cells. A second effector of the same pathway, of inhibitory LEF1 and TCF7 (TCF-1) isoforms prevail in resting TCF7 (TCF-1), was also differentially expressed. CD8 T cells. In activated cells this negative effect seems to be The differential expression of these molecules in the different relieved by down-regulation of overall LEF1 protein expression CD8 T cell subsets proved to be consistent and robust; it was and by specific down-regulation of inhibitory TCF7 (TCF-1) iso- confirmed at the level of mRNA using qRT-PCR and at the level form expression. of protein using intracellular staining in many different donors. 1444 LEF1 AND TCF7 (TCF-1) EXPRESSION IN PERIPHERAL T CELLS

ϩ Downloaded from FIGURE 6. CD8 TN preferentially express inhibitory LEF1 isoforms and down-regulate LEF1 protein expression upon activation. A, Primary resting T cells show preferential expression of inhibitory LEF1 protein isoforms. Expression of full-length and truncated (⌬CTNNB) LEF1 protein isoforms in the Jurkat T cell line and in primary human CD8 T cells was examined by Western blotting. B, Down-regulation of LEF1 protein ex- pression in activated CD8 T cells. CD8ϩ T (from CB) were either left N http://www.jimmunol.org/ unstimulated (lane 1) or stimulated with IL-15 at 50 ng/ml (lane 2)or alloantigen (lane 3). LEF1 protein expression was analyzed by Western blotting. Blotting with ␤-actin Ab (bottom panel) was used to assess pro- tein loading. Two independent experiments are shown.

FIGURE 5. Expression of different LEF1 isoforms in primary CD8 T cells. A, Cloning of LEF1 isoforms from CD8 T cells. LEF1 mRNA iso- lation induce very similar changes in gene expression in human forms were amplified from primary human CD8 T cells by RT-PCR using CD8 T cells (30). This would suggest that IL-15 and TCR stimu- ⌬ primers specific for either full-length or CTNNB isoforms. Left panel, lation probably activate common signaling pathways, and this

Gel electrophoresis of LEF1 PCR products. First lane on the left, DNA by guest on September 29, 2021 could also apply to the regulation of LEF1/TCF7 (TCF-1) expres- marker; and second lane, no template control. LEF1 PCR products were ligated into the vector pIRES2-EGFP and analyzed by restriction digestion sion in CD8 T cells. Furthermore, one recent study reported that (right panel). First lane on left, DNA marker; second lane, undigested another IL-2 family member, IL-7, can also inhibit LEF1 and parental vector (pIRES2-EGFP); and third lane, digested parental vector; TCF7 (TCF-1) expression (31). In contrast, we found that TGF␤1, lanes 1–4, digested pIRES2-LEF1. B, CD8 T cells predominantly express which is known to inhibit T cell differentiation and maintain T cell N-tail LEF1 mRNA isoforms. DNA sequencing results for cloned LEF1 quiescence (32), increased LEF1 expression. Interestingly, we mRNA isoforms from A are displayed. The number of clones identical with noted a partial recovery of LEF1 expression in CD8 T cells in vitro a specific isoform among all of the clones sequenced is given in when the cells converted back to a resting state following Ag stim- parentheses. ulation. Similarly, one murine microarray study demonstrated ini- tial down-regulation/partial recovery of LEF1 expression upon naive 3 effector 3 memory CD8 T cell differentiation in Furthermore, in a longitudinal study in an in vitro system, we were vivo (33). able to show that stimulation of TN, by alloantigen, led to a down- The observed correlations between expression of LEF1 and regulation of LEF1 and TCF7 (TCF-1). TCF7 (TCF-1) and T cell naivety and quiescence were not con- The importance of these two transcription factors in regulating sistent with published data showing that LEF1 and TCF7 (TCF-1) thymocyte development is accepted. Their possible role in regu- are able to drive cellular proliferation. However, this paradox was lating peripheral T cell function has not been considered previ- resolved by additional experiments that analyzed the expression of ously. TN can be regarded as peripheral stem cells, while TEM and the different LEF1 and TCF7 (TCF-1) isoforms. Our work shows TEMRA are differentiated cells with effector function. The idea that that, compared with Jurkat cells, resting CD8 T cells express - expression of LEF1 and TCF7 (TCF-1) may be relevant to main- atively more of the inhibitory LEF1 and TCF7 (TCF-1) protein taining the TN stem cell population is in line with the known role isoforms. T cell stimulation results in down-regulation of this in- of the WNT-␤-catenin-LEF1/TCF7 (TCF-1) pathway in the main- hibitory isoform. Importantly, it has previously been shown that tenance of hemopoietic stem cells (8–11). The observation that although stimulatory full-length LEF1 and TCF7 (TCF-1) iso- peripheral T cells from TCF7 (TCF-1)Ϫ/Ϫ mice have a spontane- forms drive the proliferation of Jurkat T cells, dominant-negative ously activated phenotype (CD44highCD62Llow) that is character- ⌬CTNNB isoforms inhibit proliferation (20). Furthermore, colon istic of Ag-primed cells is also in line with this notion (23). cancer cells predominantly express full-length LEF1 isoforms Our results are most consistent with the idea that LEF1 and while down-regulating the expression of ⌬CTNNB isoforms (17). TCF7 (TCF-1) might control T cell quiescence. Similar to TCR Finally, inhibitory ⌬CTNNB isoforms are the most abundant signals, we observed that other pro-proliferative signals such as the TCF7 (TCF-1) isoforms in the intestine and TCF7 (TCF-1)Ϫ/Ϫ cytokines IL-2 and IL-15 inhibit LEF1 and TCF7 (TCF-1) expres- mice develop intestinal and mammary adenomas (19). Thus, it is sion. Interestingly, it has been shown that IL-15 and TCR stimu- been suggested that the balance between stimulatory and inhibitory The Journal of Immunology 1445

ϩ FIGURE 7. Activation of CD8 TN changes the TCF7 (TCF-1) isoform balance in favor of stimulatory isoforms. A, Expression of TCF7 (TCF-1) protein isoforms in resting CD8 T cells. Expression of full-length and truncated (⌬CTNNB) TCF7 (TCF-1) protein isoforms in the Jurkat T cell line and in primary human CD8 T cells was examined by Western blotting. B, Preferential down-regulation of inhibitory TCF7 (TCF-1) isoform expression in activated CD8 ϩ T cells. CD8 TN (from CB) were either left unstimulated (lane 1) or stimulated with IL-15 at 50 ng/ml (lane 2) or alloantigen (lane 3). TCF7 (TCF-1) isoform expression was analyzed by Western blotting. Two (of four) independent experiments are shown. Downloaded from

LEF1 and TCF7 (TCF-1) is a checkpoint for cellular proliferation 2. Outram, S. V., A. Varas, C. V. Pepicelli, and T. Crompton. 2000. Hedgehog in the context of malignancy. signaling regulates differentiation from double-negative to double-positive thy- mocytes. Immunity 13: 187–197. In a conceptually similar way, the experiments we describe lead 3. Van de Wetering, M., W. de Lau, and H. Clevers. 2002. WNT signalling and us to formulate the hypothesis that the balance between stimula- lymphocyte development. Cell 109: S13–S19. tory and inhibitory LEF1 and TCF7 (TCF-1) isoforms represent a 4. Maillard, I., S. H. Adler, and W. S. Pear. 2003. Notch and the immune system. http://www.jimmunol.org/ Immunity 19: 781–791. checkpoint for the quiescence of peripheral T cells. Direct evi- 5. Amsen, D., J. M. Blander, G. R. Lee, K. Tanigaki, T. Honjo, and R. A. Flavell. dence for this will require future studies, in which expression of 2004. Instruction of distinct CD4 T helper cell fates by different Notch ligands on individual LEF1 and TCF7 (TCF-1) isoforms or combinations of antigen-presenting cells. Cell 117: 515–526. 6. Lowrey, J. A., G. A. Stewart, S. Lindey, G. F. Hoyne, M. J. Dallman, individual isoforms is manipulated in primary mature T cells. The S. E. Howie, and J. R. Lamb. 2002. Sonic hedgehog promotes cell cycle pro- functional redundancy between LEF1 and TCF7 (TCF-1) and the gression in activated peripheral CD4ϩ T lymphocytes. J. Immunol. 169: presence of numerous isoforms will make such experiments diffi- 1869–1875. 7. Stewart, G. A., J. A. Lowrey, S. J. Wakelin, P. M. Fitch, S. Lindey, cult to design and perform. Consistent with this, in preliminary M. J. Dallman, J. R. Lamb, and S. E. Howie. 2002. Sonic hedgehog signaling modulates activation of and cytokine production by human peripheral CD4ϩ T experiments, we did not find a clear phenotype when knocking by guest on September 29, 2021 down total LEF1 (i.e., all isoforms) by RNA interference in human cells. J. Immunol. 169: 5451–5457. 8. Willert, K., J. D. Brown, E. Danenberg, A. W. Duncan, I. L. Weissman, T. Reya, peripheral T cells. Studies in knockout and transgenic mice are J. R. Yates III, and R. Nusse. 2003. Wnt proteins are lipid-modified and can act probably more suited to address the role of the WNT pathway in as stem cell growth factors. Nature 423: 448–452. 9. Reya, T., A. W. Duncan, L. Ailles, J. Domen, D. C. Scherer, K. Willert, L. Hintz, mature T cells, although the conditional knockout of individual R. Nusse, and I. L. Weissman. 2003. A role for Wnt signalling in self-renewal of LEF1 and TCF7 (TCF-1) isoforms will be challenging. haematopoietic stem cells. Nature 423: 409–414. In conclusion, our study identifies LEF1 as the most differen- 10. Duncan, A. W., F. M. Rattis, L. N. DiMascio, K. L. Congdon, G. Pazianos, C., Zhao, K. Yoon, J. M. Cook, K. Willert, N. Gaiano, and T. Reya. 2005. Integration tially expressed transcription factor between TN and Ag-experi- of Notch and Wnt signaling in hematopoietic stem cell maintenance. Nat. Im- enced CD8 T cells. It shows that, compared with a Jurkat cell line, munol. 6: 314–322. ϩ 11. Reya, T., and H. Clevers. 2005. Wnt signaling in stem cells and cancer. Nature CD8 TN express more of the inhibitory isoform of this and an- 434: 843–850. other (TCF7 (TCF-1)) member of the WNT signaling pathway. We 12. Hurlstone, A., and H. Clevers. 2002. T-cell factors: turn-ons and turn-offs. EMBO provide evidence that down-regulation of these inhibitory isoforms J. 21: 2303–2311. is associated with T cell stimulation. Our results suggest that the 13. Van de Wetering, M., M. Oosterwegel, F. Holstege, D. Dooyes, R. Suijkerbuijk, A. G. van Kessel, and H. Clevers. 1992. The human T cell transcription factor-1 WNT pathway may have a specific function not only in immature, gene. J. Biol. Chem. 267: 8530–8536. but also in mature T cells and provide a strong rationale for further 14. Castrop, J., D. van Wichen, M. Koomans-Bitter, M. van de Wetering, molecular studies aimed at directly investigating the functional R. de Weger, J. van Dongen, and H. Clevers. 1995. The human TCF-1 gene encodes a nuclear DNA-binding protein uniquely expressed in normal and neo- importance of individual isoforms of members of the WNT-LEF1/ plastic T-lineage lymphocytes. Blood 86: 3050–3059. TCF7 (TCF-1) signaling pathway in peripheral T cell 15. Van de Wetering, M., J. Castrop, V. Korinek, and H. Clevers. 1996. Extensive differentiation. alternative splicing and dual promoter usage generate TCF-1 protein isoforms with differential transcription control properties. Mol. Cell. Biol. 16: 745–752. 16. Hovanes, K., T. W. H. Li, and M. L. Waterman. 2000. The human LEF-1 gene Acknowledgments contains a promoter preferentially active in lymphocytes and encodes multiple We thank Reg Boone, Manuela Herber, and Ann Atzberger for expert cell isoforms derived from alternative splicing. Nucleic Acids Res. 28: 1994–2003. 17. Hovanes, K., T. W. H. Li, J. E. Munguia, T. Truong, T. Milovanovic, J. L. Marsh, sorting, as well as Laurie Scott and Hayley Woffendin of the Leukaemia R. F. Holcombe, and M. L. Waterman. 2001. ␤-catenin-sensitive isoforms of Research Fund Microarray Facility for assistance in performing the lymphoid enhancer factor-1 are selectively expressed in colon cancer. Nat. Genet. GeneChip analysis. 28: 53–57. 18. Behrens, J., J. P. von Kries, M. Kuehl, L. Bruhn, D. Wedlich, R. Grosschedl, and W. Birchmeier. 1996. Functional interaction of ␤-catenin with the transcription Disclosures factor LEF-1. Nature 382: 638–642. The authors have no financial conflict of interest. 19. Roose, J., G. Huls, M. van Beest, P. Moerer, K. van der Horn, R. Goldschmeding, T. Logtenberg, and H. Clevers. 1999. Synergy between tumor suppressor APC and the ␤-catenin-Tcf4 target Tcf1. Science 285: 1923–1926. References 20. Chung, E. J., S.-G. Hwang, P. M. Nguyen, S. Lee, J.-S. Kim, J. W. Kim, 1. Lanzavecchia, A., and F. Sallusto. 2002. Progressive differentiation and selection P. A. Henkart, D. P. Bottaro, L. Soon, P. Bonvini, et al. Regulation of leukemic of the fittest in the immune response. Nat. Rev. Immunol. 2: 982–987. cell adhesion, proliferation, and survival by ␤-catenin. Blood 100: 982–990. 1446 LEF1 AND TCF7 (TCF-1) EXPRESSION IN PERIPHERAL T CELLS

21. Molenaar, M., M. van de Wetering, M. Oosterwegel, J. Peterson-Maduro, 27. Willinger, T., T. Freeman, H. Hasegawa, A. J. McMichael, and M. F. C. Callan. S. Godsave, V. Korinek, J. Roose, O. Destree, and H. Clevers. 1996. XTcf-3 2005. Molecular signatures distinguish human central memory from effector transcription factor mediates ␤-catenin-induced axis formation in Xenopus em- memory CD8 T cell subsets. J. Immunol. 175: 5895–5903. bryos. Cell 86: 391–399. 28. Petalidis, L., S. Bhattacharyya, G. A. Morris, V. P. Collins, T. C. Freeman, and 22. Verbeek, S., D. Izon, F. Hofhuls, E. Robanus-Maandag, H. te Riele, P. A. Lyons. 2003. Global amplification of mRNA by template-switching PCR: M. van de Wetering, M. Oosterwegel, A. Wilson, H. R. MacDonald, and linearity and application to microarray analysis. Nucleic Acids Res. 31: e142. H. Clevers. 1995. An HMG-box-containing T-cell factor required for thymocyte 29. Zippelius, A., G. Bioley, F.-A. Le Gal, Rufer, N., M. Brandes, P. Batard, differentiation. Nature 374: 70–74. M. de Smedt, J. Plum, D. E. Speiser, J.-C. Cerottini, et al. 2004. Human thymus 23. Schilham, M. W., A. Wilson, P. Moerer, B. J. Benaissa-Trouw, A. Cumano, and exports naı¨ve CD8 T cells that can home to nonlymphoid tissues. J. Immunol. H. Clevers. 1998. Critical involvement of Tcf-1 in expansion of thymocytes. 172: 2773–2777. J. Immunol. 161: 3984–3991. 30. Liu, K., M. Catalfamo, Y. Li, P. A. Henkart, and N. P. Weng. 2002. IL-15 mimics T cell receptor cross-linking in the induction of cellular proliferation, gene ex- 24. Ioannidis, V., F. Beermann, H. Clevers, and W. Held. 2001. The ␤-catenin-TCF-1 ϩ ϩ pression, and cytotoxicity in CD8ϩ memory T cells. Proc. Natl. Acad. Sci. USA pathway ensures CD4 CD8 thymocyte survival. Nat. Immunol. 2: 691–697. 99: 6192–6197. 25. Reya, T., M. O’Riordan, R. Okamura, E. Devaney, K. Willert, R. Nusse, and 31. Yu, Q., B. Erman, J.-H. Park, L. Feigenbaum, and A. Singer. 2004. IL-7 receptor R. Grosschedl. 2000. Wnt signaling regulates B lymphocyte proliferation through signals inhibit expression of transcription factors TCF-1, LEF-1, and ROR␥t: a LEF-1 dependent mechanism. Immunity 13: 15–24. impact on thymocyte development. J. Exp. Med. 200: 797–803. 26. Okamura, R. M., M. Sigvardsson, J. Galceran, S. Verbeek, H. Clevers, and 32. Gorelik, I., and R. A. Flavell. 2002. Transforming growth factor-␤ in T-cell R. Grosschedl. 1998. Redundant regulation of T cell differentiation and TCR␣ biology. Nat. Rev. Immunol. 2: 46–53. gene expression by the transcription factors LEF-1 and TCF-1. Immunity 8: 33. Kaech, S. M., S. Hemby, E. Kersh, and R. Ahmed. 2002. Molecular and func- 11–20. tional profiling of memory CD8 T cell differentiation. Cell 111: 837–851. Downloaded from http://www.jimmunol.org/ by guest on September 29, 2021