Becoming self-aware: the thymic education of regulatory T cells

Abstract

The generation of Foxp3(+) regulatory T (Treg) cells in the thymus is essential for immune homeostasis. In the past several years, substantial progress has been made in understanding the mechanisms by which a minor portion of developing thymocytes are selected to become Treg cells. Although previously controversial, recent data support the importance of TCR specificity as a primary determinant for selecting self-reactive thymocytes to become Treg cells in a multi-step process involving cytokines, co-stimulatory molecules, and a variety of antigen-presenting cells. Importantly, the antigenic niche for Treg cell development appears to be typically quite small, implying the recognition of tissue-specific, rather than ubiquitous, self-antigens. Finally, it appears that an NF-κB transcription factor, c-Rel, may be the link between TCR recognition and the induction of Foxp3 expression, which is required for the function and stability of the natural Treg cell population.

Figure 1

A probabilistic model for TCR specificity and Treg cell development. For the TCRs reported so far, the clonal frequency of cells expressing a specific TCR (x-axis, plot on right) is inversely related with the percentage of the cells that are Foxp3⁺ (y-axis). These data illustrate that there are quantitative differences between TCRs, primarily at the y-intercept. Three general classes of TCRs are shown (primarily Treg, incomplete Treg, and naïve). It is expected that TCR affinity, abundance of antigen, and APC costimulatory molecules, contribute to these quantitative differences. Note that using the entire CD4SP subset includes thymocytes that have not yet encountered antigen, potentially resulting in an under estimation of the efficiency of Treg cell development. Future analyses may be improved by restricting the analyses to the CD4SP subset ready to exit the thymus, perhaps by using molecular markers such as KLF2. A hypothetical extrapolation of the experimental data to the low clonal frequencies in the normal polyclonal setting (dashed lines) is shown on the left. Green area depicts the Treg cell subset.

Figure 2

Summary of thymic Treg cell development. After positive selection, immature CD4SP cells migrate from thymic cortex to the medulla. The CD4SP cells then interact extensively with APCs, such as mTECs, as well as DCs that originated from the thymus (Sirpα⁺ cDC) or periphery (Sirpα⁻ cDC and pDC), to screen for potential self-reactivity before emigrating to the periphery. All of these APC subsets are capable of directing Treg cell development, but probably present differing antigenic repertoires (depicted), of which some can be peripherally derived. Aire-dependent antigens can be transferred from mTECs to DCs. Chronic TCR/CD28 engagement with ubiquitous antigens is hypothesized to favor negative selection. Cells that express TCRs recognizing uncommon antigens may become Foxp3⁻ cytokine-responsive Treg cell precursors, in which IL-2 subsequently induces Foxp3. Cells without overt self-reactivity will be exported as naïve T cells. Quantitative thresholds for these different cell fates are currently not available.