T Cells + Mature CD8 Positive Selection by the Pre-TCR Yields

Positive Selection by the Pre-TCR Yields Mature CD8 + T Cells Yuriko Ito, Satoko Arai, Nicolai S. C. van Oers, Iannis Aifantis, Harald von Boehmer and Toru Miyazaki This information is current as of September 29, 2021. J Immunol 2002; 169:4913-4919; ; doi: 10.4049/jimmunol.169.9.4913 http://www.jimmunol.org/content/169/9/4913 Downloaded from

References This article cites 39 articles, 17 of which you can access for free at: http://www.jimmunol.org/content/169/9/4913.full#ref-list-1

Why The JI? Submit online. http://www.jimmunol.org/

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

• No Triage! Every submission reviewed by practicing scientists

• Fast Publication! 4 weeks from acceptance to publication

*average by guest on September 29, 2021

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 © 2002 by The American Association of Immunologists All rights reserved. Print ISSN: 0022-1767 Online ISSN: 1550-6606. The Journal of Immunology

Positive Selection by the Pre-TCR Yields Mature CD8؉ T Cells1

Yuriko Ito,2* Satoko Arai,2*† Nicolai S. C. van Oers,* Iannis Aifantis,‡ Harald von Boehmer,‡ and Toru Miyazaki3*

It has been of much interest whether there is functional redundancy between the constitutively signaling pre-T␣/TCR␤ (pre-TCR) and ligated TCR␣␤ complexes, which independently operate the two distinct checkpoints during thymocyte development, i.e., the pre-TCR involved in ␤-selection at the CD4؊CD8؊ double-negative stage and the TCR␣␤ being crucial for positive/negative -␣selection at the CD4؉CD8؉ double-positive stage. We found that the pre-TCR expressed on double-positive cells in TCR ,deﬁcient (TCR␣؊/؊) mice produced a small number of mature CD8؉ T cells. Surprisingly, when pre-T␣ was overexpressed resulting in augmentation of pre-TCR expression, there was a striking increase of the CD8؉ T cells. In addition, even in the ؉ ؊ ؊ absence of up-regulation of pre-TCR expression, a similar increase of CD8 T cells was also observed in TCR␣ / mice over- Downloaded from expressing Egr-1, which lowers the threshold of signal strength required for positive selection. In sharp contrast, the CD8؉ T cells drastically decreased in the absence of pre-T␣ on a TCR␣؊/؊ background. Thus, the pre-TCR appears to functionally promote positive selection of CD8؉ T cells. The biased production of CD8؉ T cells via the pre-TCR might also support the potential involvement of signal strength in CD4/CD8 lineage commitment. The Journal of Immunology, 2002, 169: 4913–4919.

he development of T cells in the thymus is controlled at phosphorylation of ZAP-70 and syk protein kinases, activation of http://www.jimmunol.org/ two consecutive checkpoints (1, 2). At the first check- the ras/mitogen-activated protein kinase pathway, and rapid in- T point, termed ␤-selection, CD4ϪCD8Ϫ double-negative crease in intracellular calcium levels (5Ð9). Because of these sim- thymocytes with a productive TCR␤ rearrangement are rescued ilarities, we asked the question whether in certain situations the from programmed cell death by the pre-TCR consisting of the pre-TCR can partially or even entirely substitute for the ligated varied TCR␤ and the invariant pre-TCR␣ (pT␣)4 chain. As a re- TCR␣␤, resulting in the generation of “␣␤-less” mature T cells. In sult, the cells proliferate, and begin to coexpress CD4 and CD8 and this study, we approached this question by qualifying different to rearrange the TCR␣ locus. Some evidence suggests that the strains of mice expressing the pre-TCR on the DP precursor cells. pre-TCR is also involved in channeling cells into the ␣␤-lineage because it reduces the number of ␥␦ T cells with productive TCR␤ by guest on September 29, 2021 rearrangements (3, 4). At the second checkpoint, the engagement Materials and Methods of the TCR␣␤ by appropriate MHC ligands rescues CD4ϩCD8ϩ Generation of transgenic mice double-positive (DP) thymocytes from apoptotic death, resulting in A 1.3-kb fragment of pT␣ cDNA (kindly provided by H. J. Fehling, Ulm, the generation of mature CD4ϩCD8Ϫ or CD4ϪCD8ϩ single-pos- Germany) was subcloned into the blunt-ended BamHI site of the expres- itive (SP) thymocytes. Very interestingly, there exists an obvious sion vector controlled by the lck-proximal promoter (10). The complete plasmid (plck-pT␣) was digested with NotI, and the transgene fragment no similarity of proximal signaling events between the constitutively longer containing vector sequence was purified by Geneclean II kit (Bio signaling pre-TCR and the ligated TCR␣␤, resulting in tyrosine 101, Vista, CA). DNA was microinjected into fertilized eggs of C57BL/6 phosphorylation of CD3␨ and CD3⑀ subunits, recruitment and mice. Resulting founders were screened for transgene by PCR and South- ern blotting. Five independent founders carried the transgene. Two of the five mice that expressed transgene at relatively higher levels, as assessed by Northern blotting using thymus RNA, were backcrossed within TCR␣Ϫ/Ϫ *Center for Immunology, University of Texas Southwestern Medical Center, Dallas, background. All animals used in the experiments were cared in accordance TX 75390; †Department of Applied Biological Chemistry, University of Tokyo, To- with institutional guidelines. kyo, Japan; and ‡Dana-Farber Cancer Institute, Harvard Medical School, Boston, MA 02115 Received for publication June 11, 2002. Accepted for publication August 26, 2002. RNA analysis The costs of publication of this article were defrayed in part by the payment of page Total RNA was isolated from thymus and lymph nodes (LNs) by using charges. This article must therefore be hereby marked advertisement in accordance RNAzolB (Tel-Test, Friendswood, TX) from various mice. A quantity with 18 U.S.C. Section 1734 solely to indicate this fact. amounting to 15 ␮g total RNA was denatured, electrophoresed in a 1% 32 1 This work was supported by research grant of Juvenile Diabetes Foundation (1- agarose gel, blotted on a nylon membrane, and then hybridized with P- 2001-581), American Liver Foundation Herman Lopata Memorial Liver Scholar labeled either pT␣ or ␤-actin cDNA fragment. Award, and Howard Hughes Medical Institute research support (to T.M.); National Institutes of Health Grant RO1AI42953 (to N.S.C.v.O.); National Institutes of Health Grants RO1A14728 and RO1A145846 (to H.v.B.); and the “Eugenia Spanopolou” Immunoprecipitation and Western blotting (to I.A.). Thymocytes from various mice (6.0 ϫ 107) were lysed at a concentration 2 Y.I. and S.A. contributed equally to this study. of 1 ϫ 108 cells/ml in a Triton X-100-containing lysis buffer (20 mM 3 Address correspondence and reprint requests to Dr. Toru Miyazaki, Center for Im- Tris-Cl, pH 7.6, 150 mM NaCl, 2.0 mM EDTA, and protease and phos- munology, University of Texas Southwestern Medical Center, 6000 Harry Hines Boulevard phatase inhibitors). The TCR␨ subunit was specifically immunoprecipi- NA 7200, Dallas, TX 75390-9093. E-mail address: [email protected] tated with an anti-␨ mAb (6B10.2). The immunoprecipitates were separated 4 ␣ ␣ ␤ ␤ on SDS-PAGE and transferred to Immobilon-P membranes (Millipore, Bed- Abbreviations used in this paper: pT , pre-TCR ; DP, double positive; 2m, 2- microglobulin; HSA, heat-stable Ag; int␤, intracellular TCR␤; LN, lymph node; SP, ford, MA). Membranes were immunoblotted with anti-phosphotyrosine or single positive; TG, transgenic; WT, wild type. anti-␨ mAb, as previously described (7).

FIGURE 1. Positive selection of CD8ϩ T cells by the pre-TCR. A, The pT␣ expression was markedly enhanced in the thymus (Thy) of pT␣-TG/ TCR␣Ϫ/Ϫ mice, but not in the LN, as assessed by Northern blotting. B, Increased CD3␨ chain phosphorylation under the overexpression of pT␣ in TCR␣Ϫ/Ϫ mouse thymocytes. Lysates of thymocytes from WT, TCR␣Ϫ/Ϫ, and pT␣-TG/TCR␣Ϫ/Ϫ mouse thymi were immunoprecipitated with an anti- CD3␨ mAb and probed sequentially on Western blots with anti-phosphotyrosine mAb 4G10 (top panel) and anti-CD3␨ mAb 6B10.2 (bottom panel). C, The Journal of Immunology 4915

Antibodies to achieve efficiently increased signals through the pre-TCR in ␣ transgenic thymi. Abs used in this study: anti-mouse CD4 (GK1.5), anti-mouse CD8 (53- Ϫ/Ϫ 6.7), anti-mouse TCR␤ chain (H57-597), anti-mouse TCR␥␦ (GL-3), anti- Surprisingly, peripheral LNs of pT␣-TG/TCR␣ mice con- mouse CD8␤ (H35-17.2), anti-pan NK cells (DX-5), anti-Thy-1.2 tained a large number (2.7 ϫ 105 cells in all of collected mesen- (30-H12), anti-mouse CD25 (3C7), anti-mouse CD69 (H1.2F3), anti- teric and inguinal LNs: 4.5% of total LN cells) of CD8ϩ cells that mouse heat-stable Ag (HSA) (30-F ), anti-phosphotyrosine (4G10), and 1 harbored int␤ (Fig. 1C, right). A small, but clearly detectable num- anti-CD3␨ (6B10.2) were purchased from BD PharMingen (San Diego, ␤ϩ ϩ ␣Ϫ/Ϫ CA). The anti-CD3⑀ mAb (145-2C11) was purchased from eBioscience ber of the int CD8 cells also did exist in LNs from TCR 4 (San Diego, CA). Intracellular staining was performed by using the Cyto- mice (1.2 ϫ 10 cells in LNs, 0.2% of total LN cells; Fig. 1C, left). fix/Cytoperm kits (BD PharMingen). These int␤ϩCD8ϩ cells were TCR␥␦Ϫ (Fig. 1D), Thy-1.2ϩ, and CD8␣ϩ␤ϩ, and lacked expression of DX-5, a pan-NK cell marker T cell activation (Fig. 1E), thus satisfying the criteria for ␣␤-lineage T lymphocytes For IL-2 production, LN cells (1.0 ϫ 106 cells/well) either from wild-type despite the absence of TCR␣␤ complexes. These findings strongly (WT) or pT␣-transgenic (TG)/TCR␣Ϫ/Ϫ mice were cultured in the pres- Ϫ implicate that the pre-TCR harbors some potential of mediating ence or absence of 10 ng/ml of PMA ϩ 5 ϫ 10 7 M of ionomycin (Sigma- signals required for positive selection of CD8ϩ SP cells. The pres- Aldrich, St. Louis, MO) in a 96-well plate for 16 h. The cells were then ϩ ϩ stained for intracellular IL-2 and cell surface CD8. CD8ϩ cells were ana- ence of a small number of int␤ CD8 cells even in LNs from lyzed for IL-2 levels by FACSCalibur cytometer (BD Biosciences, San nontransgenic TCR␣Ϫ/Ϫ mice is consistent with the very low, but Jose, CA). For expression of activation markers, LN cells (1.0 ϫ 106 cells/ ␨ Ϫ/Ϫ significant amount of CD3 chain phosphorylation detectable in well) either from WT or pT␣-TG/TCR␣ mice were cultured in the Ϫ/Ϫ ␮ ⑀ TCR␣ thymocytes (Fig. 1B). Consistently, a far increased presence or absence of 1 g/well of immobilized anti-CD3 mAb or 1 ϩ ␮g/ml Con A in a 96-well plate for 16 h. The anti-CD3⑀ mAb was immo- number of mature CD8 SP thymocytes, displaying a low level of Downloaded from bilized by preincubation of wells at 4¡C overnight. The cells were then the HSA (CD24) expression, were detected in the thymus of pT␣- stained for CD8, ␥␦, intracellular TCR␤ (int␤), and either for CD25 or TG/TCR␣Ϫ/Ϫ mice than in the thymus of TCR␣Ϫ/Ϫ mice (Table ϩ␥␦Ϫ ␤ϩ CD69, and then analyzed by cytometer. CD8 int cells were ana- I). In LNs from both TCR␣Ϫ/Ϫ and pT␣-TG/TCR␣Ϫ/Ϫ mice, a lyzed for expression of either CD25 or CD69. small number of int␤ϩCD4ϩ cells was also detected (Fig. 1C). Such int␤ϩCD4ϩ cells had been previously reported as Results ϩ lowϩ

␤ http://www.jimmunol.org/ Positive selection of mature CD8ϩ T cells by the pre-TCR CD4 TCR cells, and it was not clear whether they were derived from the thymus (16Ð18). In contrast to the int␤ϩCD8ϩ First, we analyzed a set of mice expressing either low or high cells, the number of these CD4ϩ cells was comparable in both levels of the pre-TCR in the absence of TCR␣ chain, i.e., TCR␣- ϫ 4 ϫ 5 Ϫ Ϫ Ϫ Ϫ types of mice (1.8 10 in LNs, 1.3 10 in the spleen in ␣ / ␣ / Ϫ Ϫ deﬁcient (TCR ) mice (11) or TCR mice expressing a TCR␣ / ,vs1.6ϫ 104 in LNs, 1.1 ϫ 105 in the spleen in pT␣- ␣ Ϫ Ϫ transgenic pT chain under the regulation of lck-proximal pro- TG/TCR␣ / ), suggesting that the level of the pre-TCR com- ␣Ϫ/Ϫ moter. In thymocytes of TCR mice, the pre-TCR should be plexes expressed in the thymus does not inﬂuence the development expressed at some level in DP thymocytes because most of the of the cells. Both strains also had int␤ϩCD4ϪCD8Ϫ cells, which ␤ ␤ ϩ cells express the TCR chain due to -selection, and still express consisted of ␥␦ cells (3) and a proportion of DX-5 -NK T-like

␣ by guest on September 29, 2021 the pT chain that in this situation cannot be outcompeted by the cells (data not shown). TCR␣ chain (12). In the newly developed pT␣-transgenic mice on ␣Ϫ/Ϫ ␣ ␣Ϫ/Ϫ ␣ a TCR background (pT -TG/TCR mice), the pT ex- ϩ pression was much higher than in TCR␣Ϫ/Ϫ mice in the thymus, Egr-1 enhances the positive selection of CD8 T cells by the although it was at undetectable level in the peripheral lymphoid pre-TCR organs by Northern blotting, which is compatible with the char- We further analyzed whether lowering the threshold for signaling acteristic of the transgene prompter (10) (Fig. 1A). To assess required for positive selection by overexpressing of the Egr-1, a whether the amount of signals mediated via the pre-TCR in each zinc-ﬁnger transcription factor, would increase the generation of cell at the DP stage was actually strengthened by pT␣ overexpres- int␤ϩCD8ϩ T cells in TCR␣Ϫ/Ϫ mice (19, 20). We previously sion in transgenic mice, we analyzed the tyrosine phosphorylation reported that overexpression of Egr-1 in thymocytes markedly of CD3␨ chains in thymocytes, which are known to be functionally lowered the threshold of signal strength through the TCR␣␤ (or coupled to both the pre-TCR as well as the TCR␣␤ (7, 13Ð15). To the avidity of TCR/MHC interaction) required for positive selec- this end, CD3␨ chains were immunoprecipitated from lysates of tion of mature SP thymocytes (19), and thus one might expect that thymocytes from either pT␣-TG/TCR␣Ϫ/Ϫ, TCR␣Ϫ/Ϫ,orWT Egr-1 could also lower the signal strength of the pre-TCR required mice, and their phosphorylation status was examined. In pT␣-TG/ for positive selection of int␤ϩCD8ϩ T cells. The result illustrated TCR␣Ϫ/Ϫ and TCR␣Ϫ/Ϫ mice, CD3␨ chain should associate only in Fig. 2A shows that this is in fact true: the Egr-1 transgene ex- with the pre-TCR, because of the lack of TCR␣␤ in the absence of pression on the TCR␣Ϫ/Ϫ background (Egr-TG/TCR␣Ϫ/Ϫ)re- ␣-chain. As illustrated in Fig. 1B, there was much more CD3␨ sulted in a phenotype very similar to that of pT␣-TG/TCR␣Ϫ/Ϫ chain phosphorylation in thymocytes from pT␣-TG/TCR␣Ϫ/Ϫ as mice, in which a large (even larger than in pT␣-TG/TCR␣Ϫ/Ϫ) compared with those in TCR␣Ϫ/Ϫ (highlighted as ␨-PO4), whereas proportion (5.5 ϫ 105 in LNs, 7.5% of total LN cells) of total amounts of precipitable ␨-chains were almost equal in both int␤ϩCD8ϩ cells with a ␥␦Ϫ, Thy-1.2ϩ, CD8␣ϩ␤ϩ, DX-5Ϫ phe- strains (bottom panel, CD3-␨). Thus, pT␣ overexpression appeared notype was present in peripheral LNs. As shown in Fig. 2B, their

LN (mesenteric and inguinal) cells from either TCR␣Ϫ/Ϫ or pT␣-TG/TCR␣Ϫ/Ϫ mice were stained with anti-CD4, anti-CD8␣, and anti-TCR␤ chain (for intracellular staining) and analyzed by FACSCalibur cytometer (BD Biosciences). CD4/CD8␣ proﬁles of the cells pregated on int␤ϩ population are displayed. The int␤ϩCD8ϩ cells were remarkably increased in pT␣-TG/TCR␣Ϫ/Ϫ mice (right panel) as compared with those in TCR␣Ϫ/Ϫ mice (left panel). These int␤ϩCD8ϩ cells were analyzed for TCR␥␦ (D), and for CD8␤, Thy-1.2, and DX-5 (E) by using speciﬁc mAbs. For expression of CD8␤, Thy-1.2, or DX-5 presented in E, the int␤ϩCD8ϩ cells gated on ␥␦-negative population were analyzed. Dashed lines present positive or negative controls: ␥␦ϩ populations within CD4ϪCD8Ϫ cell population (for D), B220 (B cell marker)-positive cells within LN cells stained for CD8␤ or Thy-1.2, and DX-5-positive NK cell population within CD4ϪCD8Ϫ cell population (for E). 4916 PRE-TCR-MEDIATED POSITIVE SELECTION

FIGURE 2. Peripheral CD8ϩ T cells were also increased under the Egr-1 overexpression in TCR␣Ϫ/Ϫ mice. A, LN (mesenteric and inguinal) cells from Egr-1 transgenic mice on a TCR␣Ϫ/Ϫ background (Egr-TG/ Downloaded from TCR␣Ϫ/Ϫ mice) were analyzed for the presence of int␤ϩCD8ϩ T cells by staining, as described in the leg- end for Fig. 1, C–E. B, Thymocytes from WT or Egr- TG/TCR␣Ϫ/Ϫ mice were stained for CD4, CD8, and HSA, then analyzed by cytometer. Demonstrated are the CD4/CD8 proﬁles of mature thymocytes, which low

were pregated on HSA populations. http://www.jimmunol.org/ by guest on September 29, 2021

thymi also harbored a proportion of the HSAlow CD8ϩ mature SP int␤ϩCD4ϪCD8Ϫ cells in E␣Ϫ/Ϫ mice was much smaller than in thymocytes. Likewise, the absolute number of the HSAlow CD8ϩ TCR␣Ϫ/Ϫ mice (compare Figs. 1C and 3). The precise reason for mature SP cells in Egr-TG/TCR␣Ϫ/Ϫ thymus was larger (ϳ2.5 times) this difference is unclear. E␣ might also influence the development than that in pT␣-TG/TCR␣Ϫ/Ϫ thymus (see Table I and Fig. 2B). of the int␤ϩCD4ϪCD8Ϫ cells, although further study will be required. All facts considered, it is very likely that signals mediated ␣ Direct evidence for the involvement of the pT chain by the pre-TCR and the ligated TCR␣␤ are functionally equivalent ϩ Further evidence for the involvement of the pre-TCR in positive with regard to positive selection of CD8 cells from DP selection of int␤ϩCD8ϩ T cells was observed in TCR␣ enhancer- precursors. deficient (E␣Ϫ/Ϫ) mice (21): They likewise possessed a compara- ␤ϩ ϩ ble (or even a slightly larger) proportion of int CD8 cells in the ϩ LNs as TCR␣Ϫ/Ϫ mice (1.9 ϫ 104 in LNs, 0.4% of total LN cells; CD8 T cells selected by the pre-TCR are functional Fig. 3, left panel). Moreover, in mice doubly deficient for E␣ and Having int␤ϩCD8ϩ cells undergo positive selection by the pre- pT␣ (22) (pT␣Ϫ/Ϫ/E␣Ϫ/Ϫ mice), the number of the CD8ϩ cells TCR, we then addressed whether these cells are functionally drastically decreased (Ͻ5 ϫ 103 in LNs; Fig. 3, right panel), con- equivalent to the T cells from WT mice. First, we addressed the firming the indispensable requirement of the pT␣ chain for the IL-2 production by the CD8ϩ cells in response to ionomycin ϩ selection event. Again, the presence of a similar proportion of the PMA. LN cells from either pT␣-TG/TCR␣Ϫ/Ϫ or WT mice were int␤ϩCD4ϩ cells in E␣Ϫ/Ϫ and pT␣Ϫ/Ϫ/E␣Ϫ/Ϫ mice (3.6 ϫ 104 in stimulated in the presence or absence of ionomycin ϩ PMA for LNs of E␣Ϫ/Ϫ,vs3.4ϫ 104 in LNs of pT␣Ϫ/Ϫ/E␣Ϫ/Ϫ) supports 16 h, then IL-2 production in the CD8ϩ cells were determined by the notion that the development of the int␤ϩCD4ϩ cells in the intracellular staining analysis for IL-2. As shown in Fig. 4A, CD8ϩ absence of TCR␣ is pre-TCR independent. Interestingly, the num- cells from both pT␣-TG/TCR␣Ϫ/Ϫ and WT mice efficiently pro- ber of the int␤ϩCD4ϩ cells was larger in E␣Ϫ/Ϫ mice than in duced IL-2 upon stimulation. Second, the activation status of the TCR␣Ϫ/Ϫ mice (see Fig. 1C). In addition, the proportion of the CD8ϩ cells in response to either CD3⑀ cross-linking or Con A The Journal of Immunology 4917

Table I. Absolute numbers of mature CD8ϩ thymocytes in TCR␣Ϫ/Ϫ and pT␣-TG/TCR␣Ϫ/Ϫ micea

TCR␣Ϫ/Ϫ pT␣-TG/TCR␣Ϫ/Ϫ

Total thymocytes 1.1 ϫ 108 0.76 ϫ 108 CD8ϩ SP cells Ͻ1.9 ϫ 104 1.8 ϫ 105 HSAlow CD8ϩ SP cells Ͻ1.6 ϫ 104 1.4 ϫ 105

a Thymocytes of TCR␣Ϫ/Ϫ or pT␣-TG/TCR␣Ϫ/Ϫ mice were stained for CD4, CD8, and HSA and analyzed by a cytometer. The absolute numbers of total thymocytes, whole CD8ϩ cells, and HSAlow mature CD8ϩ cells are presented. Numbers are the averages of five to seven mice analyzed. In TCR␣Ϫ/Ϫ thymi, CD8ϩ cells were ϩ low ϩ Ϫ/Ϫ often undetectable. Numbers for CD8 SP and HSA CD8 SP in the TCR␣ ϩ represent the maximum number observed for each population. FIGURE 3. Peripheral CD8 T cells were remarkably decreased in the absence of pT␣. The presence of the CD8ϩ T cells was comparably analyzed in the LNs from TCR␣ enhancer-deficient (E␣Ϫ/Ϫ) mice and E␣, Ϫ/Ϫ Ϫ/Ϫ stimulation was also studied. LN cells were stimulated by immo- pT␣ doubly deficient (pT␣ /E␣ ) mice by staining, as in the previous ␤ϩ ϩ bilized anti-CD3⑀ Ab or Con A, then the expression levels of T cell experiments (see legends for Figs. 1 and 2). The int CD8 cells were ␣Ϫ/Ϫ ␣Ϫ/Ϫ activation markers, CD25 and CD69, on the CD8ϩ cells were an- drastically decreased in pT /E mice. alyzed. As illustrated in Fig. 4B, CD8ϩ cells from pT␣-TG/ TCR␣Ϫ/Ϫ mice functionally responded to either stimulation, dem- Downloaded from onstrating up-regulation of these markers. Probably due to the less peptide complex. We wondered whether positive selection of amount of CD3⑀ on the cell surface, as demonstrated in Fig. 4C, CD8ϩ cells operated by the pre-TCR is also dependent on MHC CD69 and CD25 expression levels after stimulation were lower in molecules. We investigated this question by analyzing the propor- Ϫ Ϫ ϩ pT␣-TG/TCR␣ / than in WT. Together, the CD8 cells selected tion of the int␤ϩCD8ϩ T cells in mice doubly deficient for class I by the pre-TCR were functionally competent. MHC and TCR␣, or for class II MHC and TCR␣, which were generated by cross-breeding of TCR␣Ϫ/Ϫ mice with either class I http://www.jimmunol.org/ MHC molecules are not required for the positive selection MHCϪ/Ϫ (KϪ/ϪDϪ/Ϫ) (23, 24) or class II MHCϪ/Ϫ (A␤Ϫ/Ϫ) (25) events by the pre-TCR mice. No obvious difference in the size of the int␤ϩCD8ϩ T cell Positive selection mediated by TCR␣␤ strictly requires interaction population was detected between the MHC/TCR␣ doubly deficient between the TCR and appropriate class I or class II MHC/Ag- vs TCR␣Ϫ/Ϫ mice (Table II), suggesting that the recognition of MHC

FIGURE 4. CD8ϩ T cells selected by

the pre-TCR are functionally competent. A, by guest on September 29, 2021 LN cells either from WT or pT␣-TG/ TCR␣Ϫ/Ϫ mice were cultured in the presence or absence of 10 ng/ml of PMA ϩ 5 ϫ 10Ϫ7 M of ionomycin. The cells were then stained for intracellular IL-2 and CD8. Intracellular IL-2 levels in CD8ϩ cells are present. The solid lines represent the level of intracellular IL-2 when incubated with stimulation, whereas the dashed lines represent those on cells incubated without stimulation. B, LN cells either from WT or pT␣-TG/TCR␣Ϫ/Ϫ mice were cultured in the presence or absence of 1 ␮g/well of immobilized anti-CD3⑀ mAb or 1 ␮g/ml of Con A. The cells were then stained for CD8, ␥␦, int␤, and either for CD25 or CD69, and then analyzed by cytometer. CD25 and CD69 expression on CD8ϩ␥␦Ϫint␤ϩ cells are presented. The solid lines demonstrate the expression of CD25 or CD69 when incubated with stimulation, whereas the dashed lines represent those on cells incubated without stimulation. C, Lower expression of CD3⑀ on CD8ϩ cells from pT␣-TG/ TCR␣Ϫ/Ϫ mice. LN cells either from WT or pT␣-TG/TCR␣Ϫ/Ϫ mice were stained for CD3⑀, CD8, ␥␦, and int␤ϩ, and analyzed by cytometer. CD3⑀ expression on CD8ϩ␥␦Ϫint␤ϩ cells is presented. Bold line, WT; solid line, pT␣-TG/TCR␣Ϫ/Ϫ; dashed line, negative control (CD3⑀ expression on B220ϩ LN B cells). 4918 PRE-TCR-MEDIATED POSITIVE SELECTION

Table II. Absolute numbers of int␤ϩ CD8ϩ cells in the LNs of various micea

␣Ϫ/Ϫ Ϫ/Ϫ Ϫ/Ϫ ␣Ϫ/Ϫ ␤Ϫ/Ϫ ␣Ϫ/Ϫ ␤ Ϫ/Ϫ ␣Ϫ/Ϫ Ϫ/Ϫ ␣Ϫ/Ϫ ␤ Ϫ/Ϫ ␤Ϫ/Ϫ ␣Ϫ/Ϫ TCR K D TCR A TCR 2m TCR TAP TCR 2m A /TCR int␤ϩCD8ϩ (ϫ104) 1.2 Ϯ 0.2 1.5 Ϯ 0.2 1.2 Ϯ 0.3 1.6 Ϯ 0.4 1.6 Ϯ 0.3 1.7 Ϯ 0.5

a LNs of the six different strains of mice were analyzed for the number of int␤ϩ CD8ϩ cells. Three to four mice for each strain were analyzed. All of the strains of mice harbored comparable number of the int␤ϩ CD8ϩ cells.

Ags by the pre-TCR is not required for the selection event. We also compared in DP cells. However, the potential difference of the pT␣ ␤ ␤ Ϫ/Ϫ ␣Ϫ/Ϫ Ϫ/Ϫ ␣Ϫ/Ϫ ϩ lowϩ studied 2-microgloblin ( 2m) TCR and TAP TCR transgene expression in DP and CD4 CD8 intermediate cells mice (26) to test possible involvement of nonclassical MHC mole- will need to be determined to extend this explanation to the priority ␤ ϩ ␣ ␣Ϫ/Ϫ cules (most of which are dependent on 2m and/or TAP for their cell of CD8 cell production in pT -TG/TCR mice as well. The surface expression). Again, a comparable fraction of the int␤ϩCD8ϩ alternative view of course is that pre-TCR signals never have the ␤ Ϫ/Ϫ ␣Ϫ/Ϫ Ϫ/Ϫ ␣Ϫ/Ϫ ϩ T cells was observed in 2m TCR and TAP TCR strength required for CD4 T cell production. when compared with TCR␣Ϫ/Ϫ mice (Table II). In addition, we also In summary, our observations provide new insights on thymo- ␤ Ϫ/Ϫ ␤Ϫ/Ϫ ␣Ϫ/Ϫ analyzed 2m A TCR mice, which lack both class I and cyte development, two of which may be worth re-emphasizing. class II MHC as well as most of nonclassical MHC molecules. Like- First, the fact that the pre-TCR can promote positive selection of ϩ ϩ ϩ wise, the number of int␤ CD8 T cells in these mice was equivalent CD8 T cells further suggests the functional resemblance of the Downloaded from to that in TCR␣Ϫ/Ϫ mice (Table II). Thus, recognition of MHC mol- signaling pre-TCR and the ligated TCR␣␤. Second, the biased ecules (either classical or nonclassical) is not necessary for the pre- production of CD8ϩ T cells by the pre-TCR signals may support TCR to mediate positive selection. At the transition stage from the the propriety of hypothesis for the involvement of signal strength double-negative to the DP (␤-selection), it has been shown that pre- in CD4/CD8 lineage commitment. TCR localizes cell autonomously to membrane rafts, where it appears to signal in a constitutive and ligand-independent manner (4, 27, 28). Acknowledgments http://www.jimmunol.org/ Likewise, the recognition of an extracellular ligand might not be re- We thank Drs. H. J. Fehling (Ulm, Germany) for providing pT␣ cDNA; quired for positive selection by the pre-TCR, while we cannot mutu- F. Lemonnier (Paris, France) for KϪ/ϪDϪ/Ϫ mice; F. W. Alt (Boston, MA) ally exclude an involvement of non-MHC cell surface ligand(s). for E␣Ϫ/Ϫ mice; C. Minjares and S. Hall for creating transgenic mice and help with care of the mice; and Drs. S. Ward and J. D. Farrar (Dallas, TX) Discussion for critical reading of the manuscript. In this study, we demonstrated that the pre-TCR can promote positive selection of CD8ϩ T cells in the thymus, by comparatively References 1. Rodewald, H. R., and H. J. Fehling. 1998. Molecular and cellular events in early analyzing various strains of mice. Our observation strongly implies thymocyte development. Adv. Immunol. 69:1. by guest on September 29, 2021 that the pre-TCR signals bear functional resemblance to those 2. Malissen, B., and M. Malissen. 1996. Functions of TCR and pre-TCR subunits: transduced by the TCR␣␤ following ligand engagement. lessons from gene ablation. Curr. Opin. Immunol. 8:383. 3. Aifantis, I., O. Azogui, J. Feinberg, C. Saint-Ruf, J. Buer, and H. von Boehmer. One may wonder why the pre-TCR appeared to select predom- ␣␤ ␥␦ ϩ 1998. On the role of the pre-T cell receptor in versus T lineage commit- inantly CD8 cells. One possible explanation might be insufficient ment. Immunity 9:649. ϩ ␣Ϫ/Ϫ 4. Saint-Ruf, C., M. Panigada, O. Azogui, P. Debey, H. von Boehmer, and F. Grassi. signal strength to induce CD4 cells in TCR mice, and even ␥␦ ␣ ␣Ϫ/Ϫ ␣Ϫ/Ϫ 2000. Different initiation of pre-TCR and TCR signalling. Nature 406:524. in pT -TG/TCR or Egr-TG/TCR mice. Accumulating 5. Punt, J. A., R. T. Kubo, T. Saito, T. H. Finkel, S. Kathiresan, K. J. Blank, and evidence suggests that the strength and/or duration of TCR signal- Y. Hashimoto. 1991. Surface expression of a T cell receptor ␤ (TCR-␤) chain in ␣ ␦ ␥ ing appear to influence CD4/CD8 lineage commitment (29Ð33). the absence of TCR- ,- , and - proteins. J. Exp. Med. 174:775. ϩ 6. Groettrup, M., A. Baron, G. Griffiths., R. Palacios, and H. von Boehmer. 1992. T Indeed, positive selection of CD8 cells by weak TCR signals has cell receptor (TCR) ␤ chain homodimers on the surface of immature but not been demonstrated in various experiments (34, 35). The lower mature ␣, ␥, ␦ chain deficient T cell lines. EMBO J. 11:2735. ␨ ␣ 7. Van Oers, N. S., H. von Boehmer, and A. Weiss. 1995. The pre-T cell receptor level of CD3 chain phosphorylation in thymocytes from pT -TG/ (TCR) complex is functionally coupled to the TCR-␨ subunit. J. Exp. Med. 182: Ϫ/Ϫ ϩ TCR␣ (selecting predominantly CD8 cells) as compared with 1585. WT animals (selecting both CD4ϩ and CD8ϩ cells) as shown in 8. Cheng, A. M., I. Negishi, S. J. Anderson, A. C. Chan, J. Bolen, D. Y. Loh, and ϩ T. Pawson. 1997. The Syk and ZAP-70 SH2-containing tyrosine kinases are Fig. 1B might be sufficient for positive selection of CD8 cells, but implicated in pre-T cell receptor signaling. Proc. Natl. Acad. Sci. USA 94:9797. ϩ not for CD4 cells. Recently, a new perspective, referred to as 9. Hogquist, K. A. 2001. Signal strength in thymic selection and lineage commit- kinetic signaling model, was proposed based on a precise in vitro ment. Curr. Opin. Immunol. 13:225. 10. Mombaerts, P., A. R. Clarke, M. A. Rudnicki, J. Iacomini, S. Itohara, analysis of DP cells undergoing TCR signals (36Ð38). This model J. J. Lafaille, L. Wang, Y. Ichikawa, R. Jaenisch, M. L. Hooper, et al. 1992. postulates that: DP cells terminate CD8 transcription and convert Mutations in T-cell antigen receptor genes ␣ and ␤ block thymocyte development ϩ lowϩ ϩ at different stages. Nature 360:225. into CD4 CD8 intermediate cells in response to TCR ( co- 11. Trop, S., M. Rhodes, D. L. Wiest, P. Hugo, and J. C. Zuniga-Pflucker. 2000. ϩ lowϩ receptor) signals; these CD4 CD8 intermediate cells are not Competitive displacement of pT␣ by TCR-␣ during TCR assembly prevents sur- yet lineage committed and retain the potential to differentiate into face coexpression of pre-TCR and ␣␤ TCR. J. Immunol. 165:5566. 12. Allen, J. M., K. A. Forbush, and R. M. Perlmutter. 1992. Functional dissection of either CD4 or CD8 mature SP cells; sustained signaling in these the lck proximal promoter. Mol. Cell. Biol. 12:2758. cells results in their differentiation into CD4 SP cells, whereas 13. Cenciarelli, C., D. Hou, K. C. Hsu, B. L. Rellahan, D. L. Wiest, H. T. Smith, cessation of signaling results in coreceptor reversal and differen- V. A. Fried, and A. M. Weissman. 1992. Activation-induced ubiquitination of the Ϫ/Ϫ T cell antigen receptor. Science 257:795. tiation into CD8 SP cells. In TCR␣ as well as Egr-TG/ 14. Van Oers, N. S., N. Killeen, and A. Weiss. 1994. ZAP-70 is constitutively as- TCR␣Ϫ/Ϫ mice, the biased CD8ϩ cell production might fit with sociated with tyrosine-phosphorylated TCR ␨ in murine thymocytes and lymph ␣ node T cells. Immunity 1:675. the model, because endogenous pT expression is rapidly down- 15. Van Oers, N. S., W. Tao, J. D. Watts, P. Johnson, R. Aebersold, and H. S. Teh. regulated during the transition from the DP to the SP stage (lower 1994. Constitutive tyrosine phosphorylation of the T-cell receptor (TCR) ␨ sub- in late DP cells than in early DP cells, then almost undetectable in unit: regulation of TCR-associated protein tyrosine kinase activity by TCR ␨. Mol. Cell. Biol. 13:5771. mature SP thymocytes) (39), perhaps resulting in a decreased ex- ϩ ϩ 16. Kersh, E. N., A. S. Shaw, and P. M. Allen. 1998. Fidelity of T cell activation pression level of the pre-TCR in CD4 CD8low intermediate as through multistep T cell receptor ␨ phosphorylation. Science 281:572. The Journal of Immunology 4919

17. Eichelberger, M., A. McMickle, M. Blackman, P. Mombaerts, S. Tonegawa, and 28. Aifantis, I., C. Borowski, F. Gounari, H. D. Lacorazza, J. Nikolich-Zugich, and P. C. Doherty. 1995. Functional analysis of the TCR ␣-␤ϩ cells that accumulate H. von Boehmer. 2002. A critical role for the cytoplasmic tail of pT␣ in T in the pneumonic lung of influenza virus-infected TCR-␣Ϫ/Ϫ mice. J. Immunol. lymphocyte development. Nat. Immun. 3:483. 154:1569. 29. Anderson, A. C., E. A. Robey, and Y. H. Huang. 2001. Notch signaling in lym- 18. Mombaerts, P., E. Mizoguchi, H. G. Ljunggren, J. Iacomini, H. Ishikawa, phocyte development. Curr. Opin. Genet. Dev. 11:554. L. Wang, M. J. Grusby, L. H. Glimcher, H. J. Winn, A. K. Bhan, et al. 1994. 30. Izon, D. J., J. A. Punt, L. Xu, F. G. Karnell, D. Allman, P. S. Myung, N. J. Boerth, Peripheral lymphoid development and function in TCR mutant mice. Int. Immu- J. C. Pui, G. A. Koretzky, and W. S. Pear. 2001. Notch1 regulates maturation of nol. 6:1061. CD4ϩ and CD8ϩ thymocytes by modulating TCR signal strength. Immunity 14: 19. Miyazaki, T., and F. A. Lemonnier. 1998. Modulation of thymic selection by 253. expression of an immediate-early gene, early growth response 1 (Egr-1). J. Exp. 31. Hernandez-Hoyos, G., S. J. Sohn, E. V. Rothenberg, and E. Alberola-Ila. 2000. Med. 188:715. Lck activity controls CD4/CD8 T cell lineage commitment. Immunity 12:313. 20. Miyazaki, T. 1997. Two distinct steps during thymocyte maturation from 32. Lin, K., N. S. Longo, X. Wang, J. A. Hewitt, and K. Abraham. 2000. Lck do- CD4ϪCD8Ϫ to CD4ϩCD8ϩ distinguished in the early growth response (Egr)-1 mains differentially contribute to pre-T cell receptor (TCR)- and TCR-␣/␤-reg- transgenic mice with a recombinase-activating gene-deficient background. J. Exp. ulated developmental transitions. J. Exp. Med. 191:703. Med. 186:877. 33. Singer, A., R. Bosselut, and A. Bhandoola. 1999. Signals involved in CD4/CD8 21. Sleckman, B. P., C. G. Bardon, R. Ferrini, L. Davidson, and F. W. Alt. 1997. lineage commitment: current concepts and potential mechanisms. Semin. Immu- Function of the TCR ␣ enhancer in ␣␤ and ␥␦ T cells. Immunity 7:505. nol. 11:273. 22. Fehling, H. J., A. Krotkova, C. Saint-Ruf, and H. von Boehmer. 1995. Crucial 34. Slifka, M. K., and J. L. Whitton. 2001. Functional avidity maturation of CD8ϩ T role of the pre-T-cell receptor ␣ gene in development of ␣␤ but not ␥␦ T cells. cells without selection of higher affinity TCR. Nat. Immun. 2:711. Nature 375:795. 35. Stefanski, H. E., S. C. Jameson, and K. A. Hogquist. 2000. Positive selection is 23. Tanchot, C., F. A. Lemonnier, B. Perarnau, A. A. Freitas, and B. Rocha. 1997. limited by available peptide-dependent MHC conformations. J. Immunol. 164: Differential requirements for survival and proliferation of CD8 naive or memory 3519. T cells. Science 276:2057. 36. Brugnera, E., A. Bhandoola, R. Cibotti, Q. Yu, T. I. Guinter, Y. Yamashita, 24. Pascolo, S., N. Bervas, J. M. Ure, A. G. Smith, F. A. Lemonnier, and B. Perarnau. S. O. Sharrow, and A. Singer. 2000. Coreceptor reversal in the thymus: signaled 1997. HLA-A2.1-restricted education and cytolytic activity of CD8ϩ T lympho- CD4ϩ8ϩ thymocytes initially terminate CD8 transcription even when differen- ␤ ␤ ϩ Downloaded from cytes from 2 microglobulin ( 2m) HLA-A2.1 monochain transgenic H-2Db tiating into CD8 T cells. Immunity 13:59. ␤ 2m double knockout mice. J. Exp. Med. 185:2043. 37. Bosselut, R., L. Feigenbaum, S. O. Sharrow, and A. Singer. 2001. Strength of 25. Cosgrove, D., D. Gray, A. Dierich, J. Kaufman, M. Lemeur, C. Benoist, and signaling by CD4 and CD8 coreceptor tails determines the number but not the D. Mathis. 1991. Mice lacking MHC class II molecules. Cell 66:1051. lineage direction of positively selected thymocytes. Immunity 14:483. 26. Van Kaer, L., P. G. Ashton-Rickardt, H. L. Ploegh, and S. Tonegawa. 1992. 38. Singer, A. 2002. New perspectives on a developmental dilemma: the kinetic TAP1 mutant mice are deficient in antigen presentation, surface class I molecules, signaling model and the importance of signal duration for the CD4/CD8 lineage and CD4Ϫ8ϩ T cells. Cell 71:1205. decision. Curr. Opin. Immunol. 14:207. 27. Irving, B. A., F. W. Alt, and N. Killeen. 1998. Thymocyte development in the 39. Saint-Ruf, C., K. Ungewiss, M. Groettrup, L. Bruno, H. J. Fehling, and

absence of pre-T cell receptor extracellular immunoglobulin domains. Science H. von Boehmer. 1994. Analysis and expression of a cloned pre-T cell receptor http://www.jimmunol.org/ 280:905. gene. Science 266:1208. by guest on September 29, 2021