β-Catenin Expression Enhances IL-7 Receptor Signaling in Thymocytes during Positive Selection

Abstract

Differentiation of CD4⁺CD8⁺ double-positive thymocytes into CD8⁺ single-positive (SP) thymocytes is regulated by TCR and cytokine receptor signals. Previously, we have shown that expression of stabilized β-catenin, the major transcriptional cofactor of T cell factor, results in increase in both CD4SP and CD8SP thymocytes with a preferential effect on CD8SP thymocytes. In this report, using mice expressing stabilized β-catenin and mice with T cell specific deletion of β-catenin, we show that β-catenin expression augments IL-7Rα-chain expression and down-regulates suppressor of cytokine signaling-1 expression in thymocytes undergoing positive selection. Consequently, β-catenin expression augments IL-7R signaling in thymocytes during positive selection and promotes the development of CD8SP thymocytes.

Development of mature T cells in the thymus is mediated by signals transduced through surface αβ TCR and CD4 and CD8 coreceptors upon interaction with MHC-self-Ag complexes on the thymic epithelial cells. These signals ensure the development of immature CD4⁺CD8⁺ double-positive (DP)³ thymocytes into CD4⁻CD8⁺ single-positive (SP) and CD4⁺CD8⁻ SP thymocytes with coordinated expression of CD8 with MHC class I-restricted TCR and CD4 with MHC class II-restricted TCR (1-5). Additional survival and differentiation signals are provided by IL-7-IL-7R signals. A role of IL-7R signaling in positive selection has been difficult to assess as mice deficient in IL-7, IL-7Rα, and the common cytokine receptor γ chain (γ_c) are profoundly lymphopenic and have severely hypoplastic thymuses due to defects in survival and differentiation of the most immature thymocytes (6-8). However, a role of IL-7R in inducing Bcl-2 expression during positive selection was shown using a blocking anti-IL-7R Ab in vivo. Accordingly, enforced expression of Bcl-2 partially rescued generation of mature thymocytes in IL-7R-deficient mice, suggesting that IL-7 provides survival signals during positive selection (9).

Blocking IL-7R signals with Abs in fetal thymic organ culture (FTOC) system selectively blocked development of CD8SP thymocytes but not CD4SP thymocytes, indicating preferential requirement for IL-7R signals for generation of CD8SP thymocytes (10). Accordingly, suppressor of cytokine signaling (SOCS)-1-deficient (SOCS-1^−/−) mice (11) and mice expressing constitutively active STAT5 ((Stat5CA)-Tg mice) (12) both have enhanced cytokine signaling and show an increase in the number of cells in the CD8SP lineage. Additionally, positive selecting TCR signals have been shown to up-regulate expression of IL-7Rα and down-regulate expression of SOCS-1, thereby confer IL-7 responsiveness to signaled thymocytes, which facilitates subsequent differentiation of CD8SP thymocytes (13). Thus to date, TCR coreceptors CD4 and CD8 as well as IL-7R are believed to be the major cell surface receptors that play a role in positive selection of thymocytes. However, molecular aspects of signals transmitted by these receptors during positive selection remain to be fully defined.

β-Catenin is a major cofactor for the T cell factor family of transcription factors (14, 15). β-catenin expression is regulated at the posttranslational level by glycogen synthase kinase-3β-mediated phosphorylation followed by ubiquitination and degradation. Deletion of the glycogen synthase kinase-3β phosphorylation sites results in a stabilized form of β-catenin, which is fully functional and has been used to study the role of β-catenin in vivo (16-18). We have previously shown that expression of stabilized β-catenin in thymocytes using the proximal Lck promoter (CAT-Tg) results in an increase in the number of mature CD4SP and CD8SP thymocytes at the cost of DP thymocytes, with a greater effect on CD8SP thymocytes (19). Conversely, T cell-specific deficiency of β-catenin (CAT-KO) results in a decrease in the number of mature thymocytes with a statistically significant decrease in CD8SP thymocytes (20). Together these studies suggest a role for β-catenin during positive selection of thymocytes.

The present study was undertaken to investigate the mechanisms by which β-catenin expression regulates the number of mature CD8SP thymocytes. In CAT-Tg mice, targets of positive selection such as IL-7Rα up-regulation and SOCS-1 down-regulation are enhanced in thymocytes undergoing positive selection. Conversely, in CAT-KO mice, developing thymocytes fail to fully down-regulate expression of SOCS-1. As a result, CAT-Tg mice have a higher number and CAT-KO have a lower number of mature thymocytes with a preferential effect on CD8SP thymocytes. Furthermore, we demonstrate that in vitro and ex vivo CAT-Tg thymocytes exhibit a more robust response to IL-7 treatment. Taken together with the observation that CAT-Tg, SOCS-1-deficient, and Stat5CA-Tg mice are highly similar with respect to the generation of CD8SP thymocytes, these data demonstrate that β-catenin expression enhances IL-7R signaling in developing thymocytes.

Materials and Methods

Mice

Generation of CAT-Tg and CAT-KO mice were previously described (19, 20). SOCS-1⁺/⁻IFN-γ^−/− mice were provided by Dr. J. Ihle (St. Jude Children’s Research Hospital, Memphis, TN) (21). Thymuses from Stat5CA-Tg mice were provided by Dr. M. Farrar (University of Minnesota, Minneapolis, MN) (12). Age-matched littermate controls were used in all experiments.

Flow cytometry and cell sorting

Cells were harvested, stained, and analyzed on a FACSCalibur (BD Biosciences). Dead cells were excluded by forward light scatter and propidium iodide gating. All the data were acquired and presented on log scale. Abs with the following specificities were used for staining: allophycocyanin CD4 (GK1.5), PE-CD8α or PerCP-Cy5.5 CD8α (53-6.7), FITC-TCRβ (H57–597), PE-γ_c (4G3), Alexa Fluor 488-phosphorylated STAT5 (pSTAT5, clone 47), and PE-Bcl-2 (3F11) (all from BD Pharmingen). PE IL-7Rα (A7R34) was purchased from eBioscience. Stained thymocytes were sorted into double negative, DP, CD4⁺CD8^low, CD4SP, and TCR^high CD8SP subpopulations using a DakoCytomation MoFlo.

Intracellular staining

Freshly isolated thymocytes or spleen cells were fixed and permeabilized first with 4% paraformaldehyde, and then with methanol/acetone mixture (1:1, v/v). Cells were then stained with anti-Bcl-2 Ab or anti-pSTAT5 Ab. After intracellular staining, cells were further stained for CD4 and CD8 (22). Thymocytes were cultured in vitro at 5 × 10⁶/ml in medium or 6 ng/ml IL-7 (R&D Systems) for 1 h to assess pSTAT5 expression or cultured for 18 h to assess Bcl-2 expression.

Retroviral infection

For retroviral infection, purified CD4 T cells from C57BL/6 were activated by plate-bound anti-CD3 and anti-CD28 Abs with IL-2 for 2 days, and then incubated with supernatant containing MSCV-CAT-huCD8 or MSCV-huCD8 for a total of 3 h. The virus supernatant was then removed and the cells were further cultured in medium with IL-2 for 2 days before being examined for IL-7Rα-chain mRNA expression.

FTOC system

Embryonic day 17.5 fetal thymic lobes were placed in FTOCs in medium or with 50 μg/ml anti-IL-7Rα (eBioscience) plus anti-γ_c Abs (BD Pharmingen). On day 3 of culture, thymocytes were harvested, made into suspension culture and treated with 0.01% pronase (to remove surface CD4 and CD8) for 10 min. The cells were then cultured overnight in medium to allow re-expression of CD4 or CD8 molecules that were being actively synthesized in the cell. Thymocytes were then analyzed by flow cytometry, and the number of CD4SP and TCRβ^high CD8SP thymocytes per thymic lobe was determined.

Quantitative real-time RT-PCR

Total RNA from sorted thymocyte subpopulations was reverse transcribed using poly(dT) and Superscript III reverse transcriptase (Invitrogen Life Technologies). The cDNA was subjected to real-time PCR amplification (Applied Biosystems) for 40 cycles with annealing and extension temperature at 60°C.

Results

CAT-Tg, SOCS-1-KO, and Stat5CA-Tg mice have increased mature CD8SP thymocytes

In CAT-Tg mice expressing stabilized β-catenin under proximal Lck promoter (19), the frequency of both CD4SP and CD8SP thymocyte populations relative to DP thymocytes was increased, with a significantly greater increase in CD8SP population (Fig. 1A, top left). Conversely, in CAT-KO mice (20) with a T cell-specific deletion of β-catenin gene, there was a modest but statistically significant decrease in the frequency of CD8SP thymocytes (Fig. 1A, top right). The slightness in the degree of changes in CAT-KO mice may be attributed to the presence of γ-catenin that has been shown to play a redundant role in other biological systems (23, 24). Interestingly, in SOCS-1^−/−IFN-γ^−/− mice (21) (IFN-γ deficiency prevents disease and perinatal death) and in mice expressing an active form of STAT5b (Stat5CA-Tg mice) (12), a preferential increase in CD8SP thymocytes was observed (Fig. 1A, bottom panels) (11, 12). The absolute number of CD8SP thymocytes, but not CD4SP thymocytes, in CAT-Tg, SOCS-1^−/−IFN-γ^−/−, and Stat5CA-Tg mice showed a significant increase compared with control mice (Fig. 1B). In CAT-KO mice, a decrease in the number of both CD4SP and CD8SP thymocytes was observed (Fig. 1B). However, it mainly reflects the decrease in total thymic cellularity (20). Thus expression of stabilized β-catenin preferentially increases the number of CD8SP thymocytes, and the similarity in thymic phenotypes among CAT-Tg, SOCS-1-KO, and Stat5CA-Tg mice suggests the possibility that β-catenin expression may regulate the number of CD8SP thymocytes by affecting cytokine signaling during positive selection.

FIGURE 1.

CAT-Tg, SOCS-1-KO, and Stat5CA mice have increased mature thymocytes. A, Percentage of CD4SP and TCR^high CD8SP thymocytes (relative to DP) in CAT-Tg, CAT-KO, SOCS-1^−/−IFN-γ^−/−, and Stat5CA-Tg mice compared with their littermate controls (grey bar). CAT-Tg (n = 8), CAT-KO (n = 7), SOCS-1^−/−IFN-γ^−/− (n = 4), and Stat5CA-Tg (n = 4) mice are represented. B, Absolute number of CD4SP and CD8SP thymocytes in the mice groups described in A are shown.

β-Catenin expression enhances IL-7R signals in thymocytes undergoing positive selection

To determine whether β-catenin expression enhanced IL-7 response, we directly examined IL-7-induced phosphorylation of STAT5 (pSTAT5) in thymocytes. Mature CD4SP and CD8SP thymocytes from CAT-Tg mice showed significantly higher IL-7 responsiveness compared with cells from control mice, as noted by higher pSTAT5 (shown by mean fluorescence intensity; MFI) induced in response to in vitro IL-7 treatment (Fig. 2A, top left). We also examined IL-7 responsiveness in SOCS-1^−/−IFN-γ^−/− CD4SP and CD8SP thymocytes to determine whether the increase in IL-7 responsiveness in CAT-Tg thymocytes was functionally significant. We found that the increase in IL-7 responsiveness in CAT-Tg thymocytes was similar or even greater than that seen in SOCS-1^−/−IFN-γ^−/− thymocytes (Fig. 2A, compare top left and bottom panels). These data suggest that the enhanced IL-7 responsiveness caused by β-catenin expression is biologically significant. To determine whether enhanced IL-7R signaling results in enhanced downstream target gene expression, we examined IL-7 induced up-regulation of Bcl-2 by intracellular staining. Both CD4SP and CD8SP thymocytes from CAT-Tg mice showed higher level of Bcl-2 induced by IL-7 treatment compared with control counterparts, with CD8SP thymocytes showing a more dramatic increase that is consistent with a higher level pSTAT5 induction by IL-7 in these cells (Fig. 2A, top right). Thymocytes from CAT-KO mice did not show decreased responsiveness to IL-7 (data not shown), suggesting that γ-catenin may play a redundant role in this function. During positive selection and maturation to the SP stage, DP thymocytes that have received the initial positive selection signal first differentiate into a transitional stage described as CD4⁺CD8^low cells. This developmental intermediate, along with CD69⁺ DP thymocytes, is believed to be precursor for mature thymocytes (25). In particular, enhanced IL-7R signaling in CD4⁺CD8^low thymocytes is believed to promote the generation of CD8SP thymocytes from these positive selection intermediates. CAT-Tg CD4⁺CD8^low thymocytes showed higher levels of pSTAT5 compared with control CD4⁺CD8^low cells in response to in vitro IL-7 treatment (Fig. 2B). CAT-Tg CD4⁺CD8^low cells also showed higher levels of Bcl-2 expression upon IL-7 treatment compared with control CD4⁺CD8^low thymocytes (Fig. 2B). Statistical analysis of several experiments showed that the increased responses, though modest, were highly significant (Fig. 2B). Again, the increase in IL-7 responsiveness in CAT-Tg CD4³ CD8^low thymocytes was comparable to or greater than that seen in SOCS-1^−/−IFN-γ^−/− CD4⁺CD8^low thymocytes (Fig. 2B, compare top and bottom panels), suggesting the enhanced IL-7 responsiveness caused by β-catenin expression, although seemingly small, is biologically significant. Thus transgenic expression of β-catenin significantly enhanced cytokine responsiveness of positive selection intermediates to IL-7 in vitro.

FIGURE 2.

β-Catenin expression enhances IL-7R signals of thymocytes. A, Enhanced responsiveness of CAT-Tg mature thymocytes to IL-7 treatment in vitro. Thymocytes from control or CAT-Tg mice were stimulated with IL-7 in vitro. Induction of pSTAT5 after 1 h of treatment and Bcl-2 after 18 h of treatment of thymocytes was analyzed by intracellular staining and flow cytometry. Level of pSTAT5 induction (left) or Bcl-2 induction (right) is determined by mean fluorescence intensity (MFI) and is calculated by subtracting MFI before IL-7 treatment from after IL-7 treatment. Enhanced IL-7 response (upper panels) in CAT-Tg mature thymocytes is shown for control mice (n = 2) and CAT-Tg mice (n = 4). The results are representative of three independent experiments. Enhanced IL-7 response (lower panels) in mature thymocytes in SOCS-1^−/−IFN-γ^−/− mice (n = 4). B, Enhanced response of CAT-Tg CD4⁺CD8^low thymocytes to IL-7 in vitro. IL-7 responsiveness in CD4⁺CD8^low thymocytes was assessed similarly as described in A. Enhanced IL-7 responses in CAT-Tg CD4⁺CD8^low thymocytes (top left panels) and histograms of pSTAT5 and Bcl-2 staining in CAT-Tg thymocytes (top right panels) are represented. Control mice (n = 2) and CAT-Tg mice (n = 4) are shown. The results are representative of three independent experiments. Bottom panels represent enhanced IL-7 responses in CD4⁺CD8^low thymocytes in SOCS-1^−/−IFN-γ^−/− mice (n = 4). C, Increased level of pSTAT5 in immediately ex vivo thymocytes from CAT-Tg mice. Freshly isolated thymocytes were fixed and permeabilized, and then subjected to intracellular staining for pSTAT5. MFI of pSTAT5 (left) in CD4⁺CD8^low cells and histograms of intracellular pSTAT5 (right panels) in CD4⁺CD8^low, CD4SP, and CD8SP thymocytes from control and CAT-Tg mice are shown. Data are representative of eight independent analyses.

To determine whether CAT-Tg CD4⁺CD8^low thymocytes had received higher IL-7R signaling in vivo, we examined pSTAT5 in ex vivo cells. Intracellular staining for pSTAT5 of ex vivo CAT-Tg CD4⁺CD8^low thymocytes showed markedly higher level of pSTAT5 compared with control CD4⁺CD8^low thymocytes (Fig. 2C, right top). Ex vivo CAT-Tg CD4SP and CD8SP mature thymocytes also showed higher level of pSTAT5 (Fig. 2C, right top). Quantification of pSTAT5 level, based on intracellular staining, showed that CAT-Tg CD4⁺CD8^low cells had significantly higher pSTAT5 level compared with control CD4⁺CD8^low cells (Fig. 2C, left). The increase in pSTAT5 was not due to an increase in total STAT5 level, as level of total STAT5 is the same in CAT-Tg thymocytes or control thymocytes (Fig. 2C, right bottom). The degree of increase in pSTAT5 in ex vivo CAT-Tg CD4⁺CD8^low thymocytes was greater than that seen in SOCS-1^−/−IFN-γ^−/− CD4⁺CD8^low thymocytes (data not shown), suggesting the enhanced IL-7 signaling caused by β-catenin expression is biologically relevant. These data indicate that CAT-Tg CD4⁺CD8^low cells and CAT-Tg SP thymocytes experience higher level of cytokine signals in vivo. Together these data demonstrate that β-catenin expression enhances IL-7R signaling both in vitro and in vivo. This result suggests that enhanced IL-7R signaling in CD4⁺CD8^low intermediates promotes increased generation of CD8SP thymocytes.

β-Catenin expression promotes up-regulation of IL-7R and down-regulation of SOCS-1 expression in thymocytes during positive selection

Positive selecting TCR signals up-regulate expression of IL-7Rα and down-regulate expression of SOCS-1 to confer IL-7R responsiveness to CD4⁺CD8^low thymocytes (9, 11, 13). To determine molecular basis for β-catenin-mediated enhancement of IL-7R signaling, we assayed the expression of IL-7Rα and SOCS-1 molecules. CAT-Tg CD4⁺CD8^low thymocytes expressed a higher level of surface IL-7Rα protein compared with control CD4⁺CD8^low thymocytes (Fig. 3A). CAT-Tg mature thymocytes as well as peripheral CD4 and CD8 T cells also expressed a higher level of IL-7Rα compared with their control counterparts (Fig. 3A). Cell surface expression of IL-7Rα was accompanied by up-regulation of IL-7Rα mRNA in CAT-Tg CD4⁺CD8^low cells (Fig. 3A, far right). To further investigate whether β-catenin expression induces IL-7Rα expression, we expressed a stabilized form of mouse β-catenin in peripheral CD4 T cells via retroviral infection. Two days after infection, CD4 T cells expressing stabilized β-catenin expressed 4-fold higher level of IL-7Rα mRNA compared with cells infected with control vector (Fig. 3B). Thus, expression of β-catenin induces IL-7Rα expression. In addition to IL-7Rα, the level of surface γ_c was also modestly increased on CAT-Tg CD4⁺CD8^low thymocytes, but not DP thymocytes, compared with control thymocytes (Fig. 3C). This result shows that a signaling-competent IL-7R was assembled on CAT-Tg CD4⁺CD8^low thymocytes and its level was enhanced by β-catenin expression.

FIGURE 3.

β-Catenin expression increases IL-7R expression and decreases SOCS-1 expression. A, Expression of IL-7Rα-chain. Surface expression of IL-7Rα-chain on thymocyte subpopulations and spleen CD4 or CD8 T cells from control or CAT-Tg mice is shown (left panels). Data are representative of six to eight independent analyses for each type of mice. RNA from sorted CD4⁺CD8^low thymocytes was analyzed by real-time RT-PCR. Relative expression level of IL-7Rα to β-actin is shown (right) from three independent experiments. B, Induction of IL-7Rα by expression of β-catenin. Purified CD4 T cells from C57BL/6 mice were activated and infected with MSCV-CAT-huCD8 or control vector MSCV-huCD8. Two days after retroviral infection, the cells were stained for huCD8 to confirm efficiency of infection and analyzed for IL-7Rα mRNA by real-time RT-PCR. Relative expression level of IL-7Rα to β-actin is shown. C, Surface expression of γ_c. Surface expression of γ_c on control or CAT-Tg thymocytes was analyzed by flow cytometry, and MFI of γ_c is shown from eight independent experiments. D, SOCS-1 gene expression in CD4⁺CD8^low thymocytes. Purified CD4⁺CD8^low thymocytes from CAT-Tg or CAT-KO mice were analyzed for SOCS-1 gene expression by real-time RT-PCR. Relative expression of SOCS-1 to β-actin in control thymocytes is set as 1, and the fold change relative to control thymocytes is shown from three independent experiments.

We next assessed SOCS-1 expression in CD4⁺CD8^low thymocytes from CAT-Tg and control mice. Expression of SOCS-1 mRNA was significantly lower in CD4⁺CD8^low subset from CAT-Tg mice compared with CD4⁺CD8^low control thymocytes (Fig. 3D). Conversely, higher level expression of SOCS-1 mRNA was observed in CD4⁺CD8^low cells from CAT-KO mice (Fig. 3D). Statistical analysis showed that regulation of SOCS-1 expression in the presence and absence of β-catenin was significant. In this function γ-catenin does not appear to play a redundant role. The ratio of SOCS-1 mRNA in DP to SOCS-1 mRNA in CD4⁺CD8^low was 2.8 in control mice and 5.9 in CAT-Tg mice. This result shows a 2-fold further down-regulation of SOCS-1 during DP to CD4⁺CD8^low differentiation in CAT-Tg mice. We conclude that β-catenin expression down-regulates SOCS-1 expression in TCR-signaled CD4⁺CD8^low thymocytes in addition to up-regulating IL-7R components. Thus, β-catenin enhances IL-7R signaling by positively regulating IL-7R expression and negatively regulating SOCS-1 expression. The observation that a modest increase in IL-7R expression and decrease in SOCS-1 expression in CD4⁺CD8^low cells results in a more dramatic increase in CD8SP thymocytes underscores the importance of the level of IL-7R signaling in CD4⁺CD8^low intermediate cells for positive selection.

β-Catenin expression enhances but does not replace IL-7R signal during positive selection of CD8SP thymocytes

To verify that β-catenin-induced enhanced IL-7R signaling was essential for increased CD8 cell generation, we studied the generation of CD8SP thymocytes in FTOC in the presence of Abs that block signal transduction through the IL-7R. Treatment of FTOC with blocking IL-7R Abs has been shown to selectively block development of CD8SP thymocytes but not CD4SP thymocytes (10). Thymic lobes from embryonic day 17.5 mice were placed in FTOC with or without Abs against IL-7Rα and γ_c. Three days later, cells were harvested from the FTOC and treated with pronase to strip cell surface CD4 and CD8 proteins. Treated cells were then cultured in medium to allow re-expression of CD4 or CD8 molecules that were actively synthesized in the cells. CD8SP thymocytes in both control and CAT-Tg FTOC were dramatically reduced when IL-7R signal was blocked by the addition of IL-7R Abs (Fig. 4, left). These data show that IL-7R signal was required for the generation of CD8SP thymocytes in CAT-Tg mice. A small number of CD8SP thymocytes were consistently generated in the presence of blocking Abs to IL-7R (Fig. 4, left) (10). Interestingly the number of CD8SP thymocytes generated in the presence of blocking Abs to IL-7R was the same between control and CAT-Tg mice (Fig. 4, left). These data suggest that a minority of CD8SP thymocytes may not use IL-7R signals and the generation of these cells is not affected by β-catenin expression. As expected, addition of IL-7R and γ_C Abs did not affect the number of CD4SP thymocytes whether in control or CAT-Tg thymic lobes, consistent with the notion that generation of majority of CD4SP thymocytes does not require IL-7R signal (Fig. 4, right). Taken together, the effect of CAT-Tg on promoting positive selection of CD8SP thymocytes required IL-7R signal. We conclude that β-catenin expression enhances but does not replace IL-7R signals during positive selection of thymocytes.

FIGURE 4.

β-Catenin expression enhances but does not replace IL-7R signal during positive selection of CD8SP thymocytes Embryonic day 17.5 fetal thymic lobes from control or CAT-Tg mice were placed in FTOC in medium alone or medium containing anti-IL-7Rα plus anti- γ_c Abs. After 3 days of culture, thymocytes were harvested, treated with pronase, and cultured overnight in medium. Thymocytes were then stained with Abs to CD4, CD8, and TCRβ and analyzed by flow cytometry. The absolute number of CD4SP and TCRβ^high CD8SP thymocytes per thymic lobe is shown from three independent experiments.

Discussion

In this report, we demonstrate that β-catenin expression enhances IL-7R signals during positive selection, resulting in increased numbers of CD8SP thymocytes. The enhancement of IL-7R signaling in CAT-Tg thymocytes was modest but significant and similar in degree to the enhancement of IL-7R signaling in SOCS-1^−/−IFN-γ^−/− thymocytes. Both SOCS-1^−/− and CAT-Tg mice have a significantly higher number of CD8SP thymocytes with no increase in CD4SP thymocytes. This finding shows that modestly enhanced cytokine signaling in CD4⁺CD8^low thymocytes is sufficient to result in the significant increase in CD8SP thymocytes and suggests that a modest effect in the “correct” developmental intermediate (CD4⁺CD8^low) can result in biologically significant increase in the product population (CD8SP thymocytes).

In CAT-Tg mice, transgenic stabilized β-catenin is expressed at equal level in both CD4SP and CD8SP thymocytes (data not shown). However, only CD8SP thymocytes show an increase in number, whereas CD4SP thymocytes are largely unaffected. This phenotype is strikingly similar to SOCS-1^−/− and Stat5CA-Tg mice that have enhanced cytokine signaling in both CD4SP and CD8SP thymocytes but a preferential increase in CD8SP thymocytes. We thus speculated that expression of β-catenin might regulate cytokine receptor signaling, such as IL-7R signaling, during positive selection and enhance maturation of CD8SP thymocytes.

Analysis of proximal IL-7 signaling, indicated by phosphorylation of STAT5, shows that both CAT-Tg CD4SP and CAT-Tg CD8SP thymocytes have enhanced IL-7 signaling either in vitro or in vivo. Furthermore, analysis of IL-7 downstream target Bcl-2 expression shows that IL-7-induced effect is elevated in both CAT-Tg CD4SP and CAT-Tg CD8SP thymocytes. Importantly, β-catenin induced enhancement of IL-7 signaling and effect is observed in positive selection intermediates CD4⁺CD8^low thymocytes. At this stage, the enhanced IL-7 signaling has been shown to preferentially promote the further differentiation of CD4⁺CD8^low cells into CD8SP thymocytes (10, 11, 25). The effect of β-catenin on IL-7R signaling is further supported by the observation that SOCS-1^−/− mice and Stat5CA-Tg mice, both of which have enhanced IL-7R signaling in thymocytes, show increased CD8SP thymocytes similar to that seen in CAT-Tg mice, as described in this study and other studies (11, 12). The enhancement of IL-7R signaling in CAT-Tg thymocytes, though modest, is similar or even greater than that caused by SOCS-1 deficiency. This response suggests the enhanced cytokine signaling in CAT-Tg thymocytes is sufficient to result in the significant increase in CD8SP thymocytes.

IL-7 signaling is regulated by both induced expression of the IL-7α and γ_c chains as well as expression of negative regulators such as SOCS-1 in thymocytes. Indeed, CAT-Tg CD4⁺CD8^low thymocytes show increased level of both IL-7Rα-chain and γ_c and decreased level of SOCS-1, which provides a direct molecular explanation for enhanced IL-7 signaling in these cells. The exact molecular mechanism by which β-catenin up-regulates IL-7Rα and γ_c expression and down modulates SOCS-1 expression requires further investigation. However, as up-regulation of IL-7R and down-regulation of SOCS-1 expression are both positive selection induced events, β-catenin expression may enhance these events by lowering the threshold of positive selection. In addition, β-catenin may affect gene transcription or RNA stability of IL-7R and SOCS-1. This affect is supported by the observation that expression of β-catenin in peripheral CD4 T cells using the retroviral technology leads to increased expression of IL-7Rα-chain expression. These data also show that β-catenin expression induces IL-7Rα independently of positive selection signal.

Ab blocking of IL-7R in FTOC has been used to study the role of IL-7R in positive selection because the requirement of IL-7 at the double-negative stage interferes with the analysis of its role in positive selection in IL-7- or IL-7R-deficient mice. Blocking IL-7R signaling by anti-IL-7Rα and anti-γ_c Ab in FTOC from CAT-Tg and wild-type mice shows that very few CD8SP thymocytes are generated in the absence of IL-7R signals, indicating that β-catenin expression enhances but cannot replace IL-7R signal for the development of CD8SP thymocytes. Importantly, the fact that the small number of IL-7R-independent CD8SP thymocytes is not increased by β-catenin expression demonstrates that the effect of β-catenin in promoting generation of CD8SP thymocytes is dependent on IL-7R signal.

Even though CAT-KO thymocytes show increased SOCS-1 expression, CAT-KO mice do not show a significant decrease in CD8SP thymocytes. Furthermore, CAT-KO thymocytes do not show a decrease in ex vivo IL-7R signaling or in vitro IL-7 responses. This result is perhaps due to redundancy with γ-catenin, a β-catenin homolog that is expressed in thymocytes, binds to T cell factor-1, and has been shown to be redundant in other biological systems (23, 26). β- and γ-catenin double knockout mice, not yet available for analysis, will hopefully reveal the independent requirement of these molecules in cytokine signaling and positive selection in the future. Nevertheless, the finding that β-catenin regulates IL-7R signaling in thymocytes provides mechanisms for its function in thymocyte positive selection. The observation that β-catenin expression in mature T cells up-regulates IL-7R expression suggests that β-catenin may influence other cytokine-mediated processes.