β-Catenin expression in thymocytes accelerates thymic involution

2.1 Expression of stabilized β-catenin in thymocytes accelerates thymic involution

To study the Wnt-β-catenin-TCF signaling pathway in T cell development, we expressed stabilized β-catenin in thymocytes using the proximal Lck promoter cassette. Stabilized β-catenin was expressed in DN, DP and SP thymocytes (data not shown). To study the effect of β-catenin expression on thymic aging, the number of thymocytes was assessed in mice at various ages. Control (non-transgenic littermates) and ΔCat-Tg mice showed comparable numbers of thymocytes at birth, suggesting that similar number of thymocytes were generated in young mice. As the mice aged, we noticed that thymocyte numbers decreased dramatically in ΔCat-Tg mice compared with age-matched littermates (Fig. 1). Thymic involution was evident in ΔCat-Tg mice by 3 months of age, unlike in littermate control mice and was accelerated as the mice aged. The striking conclusion is that 4–8-month-old ΔCat-Tg mice resemble 12–18-month-old control mice (Fig. 1).

Data from large numbers of ΔCat-Tg and littermate control mice at various ages are summarized in Fig. 2. These data suggest that thymic involution is augmented earlier in adulthood in ΔCat-Tg mice and the process continues to be accelerated as the mice age (Fig. 2a).

The phenotype of thymocytes in ΔCat-Tg mice remains comparable as they age (Fig. 2b). In 1–3-month-old ΔCat-Tg mice (right top panel) the proportion of DP thymocytes was decreased compared with littermate controls (left top panel), the proportion of CD4 SP thymocytes was slightly increased compared with controls and the percentage of CD8 SP thymocytes was significantly increased in ΔCat-Tg mice compared with controls. As the mice aged this phenotype remained consistent (Fig. 2b, middle and bottom panels). Therefore, no significant difference in phenotype of thymocytes could be attributed to age in ΔCat-Tg mice.

These data indicate that an agent that stabilizes β-catenin in thymocytes may be important in thymic involution under normal circumstances. The effect of this agent is augmented in ΔCat-Tg mice, leading to faster thymic involution. There is evidence of a hormonal basis for thymic involution 16, 17 and, since β-catenin has been shown to participate in signaling from hormone receptors, expression of stabilized β-catenin may mimic hormonal signaling in thymocytes. Alternately, we have recently shown that TCR and CD28 signals stabilize and target β-catenin to the nucleus; therefore, another possibility is that enhanced thymic involution in ΔCat-Tg mice mimics increased chronic signaling by the TCR and CD28 in thymocytes. Finally, the canonical Wnt-β-catenin-TCF signaling pathway may play a role in thymic function and involution in ways not yet defined.

2.2 Expression of stabilized β-catenin in thymocytes does not lead to a significant change in splenic T and B cells

Age-dependent thymic involution is not accompanied by a drastic reduction in the number of peripheral T cells. Splenic T cells in aged animals express high levels of CD44, indicating a memory phenotype 7. However, we did not address this issue because ΔCat-Tg thymocytes and T cells express intermediate levels of CD44 in the absence of any other memory markers (Mulroy et al., manuscript submitted). It appears that β-catenin expression up-regulates CD44 expression directly.

We characterized splenic T and B cells in young and aged ΔCat-Tg mice (Fig. 3). The number of splenocytes was comparable in age-matched ΔCat-Tg and control mice (Fig. 3a). This showed that, unlike the observations in other tissues such as the breast and colon, expression of stabilized β-catenin did not enhance cellular proliferation in lymphoid tissue. The percentage of T cells was not significantly different in ΔCat-Tg and control mice; neither was the percentages of B cells significantly different in ΔCat-Tg and control mice (Fig. 3b). The number of splenic T cells was also found to be similar; so was the number of B cells (Fig. 3c). Finally, the percentage (Fig. 3d) and number (Fig. 3d and e) of CD4 cells were also comparable in age-matched ΔCat-Tg and control mice; similarly, the percentage and number of CD8 T cells were comparable in control and ΔCat-Tg mice (Fig. 3e). Therefore, aging of the immune system in ΔCat-Tg mice was accelerated but the phenotype of peripheral T cells remained normal. Thus the ΔCat-Tg mouse model system retains an essential feature of normal aging of the immune system.

2.3 The mechanism of thymic involution due to expression of stabilized β-catenin

One hypothesis is that age-dependent changes in both the thymic microenvironment and thymocytes affect thymic involution. This suggests that age-related changes in both compartments are responsible for thymic involution. Stabilized β-catenin was expressed only in thymocytes in ΔCat-Tg mice; therefore, the primary effect of this transgene has to be on thymocytes.

We further examined thymocyte sub-populations in young and aged control and ΔCat-Tg mice (Fig. 4). In 1–3-month-old ΔCat-Tg and control mice the percentage of DN populations was comparable but by 4–8 months the proportion of DN thymocytes was higher in the ΔCat-Tg mice (Fig. 4a). The absolute number of DN cells at all ages was lower in ΔCat-Tg mice compared with control littermates. This reflects the decrease in total number of thymocytes starting at 3 months of age. The increase in proportion of DN thymocytes is reflective of the effective decrease in the proportion of DP thymocytes described below.

The proportion of DP population was significantly decreased in ΔCat-Tg mice, starting at 1–3 months of age and it continued to decline sharply throughout the lifetime of ΔCat-Tg mice (Fig. 4b, left panels). The absolute number of DP thymocytes was also significantly reduced in ΔCat-Tg mice compared with control littermates (Fig. 4b, right panels). However, we have ruled out the possibility that the decrease in DP thymocytes was due to enhanced apoptosis of ΔCat-Tg thymocytes. We have previously shown that expression of ΔCat-Tg did not enhance apoptosis in DP thymocytes. It also did not protect DP thymocytes from spontaneous death in in vitro culture or from dexamethasone-mediated death (Mulroy et al., manuscript submitted). Indeed, in light of the observation that β-catenin was important for DP thymocyte survival 11, 12 it was unexpected that expression of stabilized β-catenin did not provide extra protection from cell death in vitro. One explanation is that, although β-catenin may be required for the survival of DP thymocytes, more is not better.

The proportion of CD4 SP thymocytes was comparable in ΔCat-Tg and control mice at 1–3 months of age but increased thereafter (Fig. 4c). The percentage of CD8 SP thymocytes was high in mice at a young age in ΔCat-Tg mice, as has been described previously, and remained high at all ages (Fig. 4c, right panels). The absolute numbers of CD4 and CD8 SP thymocytes declined with age, commensurate with the decrease in the number of thymocytes in aged mice (Fig. 4d).

These data point to the DP population as being the target of β-catenin mediated involution signals. We have previously shown that the DP thymocytes in ΔCat-Tg mice were not more susceptible to apoptosis than control thymocytes when assayed freshly ex vivo, cultured in vitro or treated with dexamethasone (see above). Thus the decrease in the number of DP thymocytes does not reflect a direct effect of stabilized β-catenin on the viability of DP thymocytes. One interpretation is that the effect of an unknown agent that drives thymic involution in normal mice is enhanced in ΔCat-Tg mice leading to age-related decline in this population by an unknown mechanism.

In conclusion, ΔCat-Tg transgenic mice expressing stabilized β-catenin in thymocytes undergo accelerated thymic involution. Thymocyte maturation, survival and proliferation are controlled by several factors, many of which may be produced by the thymic microenvironment in response to interactions between thymocytes and thymic stromal cells. This paper is one of the first to describe that a protein expressed exclusively in the thymocyte can modulate cellular fate in aging. It shows that thymic involution can be uncoupled from the bone marrow derived precursor input, which would be normal in ΔCat-Tg mice. In normal mice, agents driving β-catenin-mediated involution of the thymus could be TCR and CD28 signaling, hormonal effects or Wnt proteins produced by the thymocytes or by thymic stromal cells. In ΔCat-Tg mice, β-catenin expression in thymocytes could lead to a change in the dynamics of the interaction between thymocytes and stromal cells and/or to enhanced production of factors that accelerate thymic involution.

3.1 Generation and analysis of transgenic mice

The ΔCat construct was generated by cloning the BamH I fragment containing the mutant ΔN87βCat gene 18 into the BamH I site in p1017 19. Not-Icut DNA containing the ΔN87βCat gene was injected into FVB recipient mice. Transgenic mice were identified by Southern-blot analysis of DNA extracted from tail cuts using standard protocols. The probe consisted of the BamH I fragment containing the ΔN87βCat gene. Genotyping of transgenic mice will been described elsewhere (Mulroy et al., manuscript submitted).

3.2 FACS analysis

For three-color analyses, thymocytes were immunostained with FITC-, Cy-chrome- or PE-labeled anti-CD4, anti-CD8, anti-TCR, anti-CD5, anti-CD69, anti-HSA, anti-CD62L or anti-Qa-2 antibodies in PBS containing 2% FCS and 0.2% sodium azide in the presence of anti-CD16/CD32 antibodies. All antibodies were purchased from PharMingen (San Diego, CA). Immunostained cells were analyzed using a FACScan flow cytometer and CellQuest software (Becton Dickinson, Mountain View, CA).