CD28 facilitates the generation of Foxp3(-) cytokine responsive regulatory T cell precursors

Abstract

The T cell costimulatory molecule CD28 plays an important role in the thymic generation of Foxp3(+) regulatory T cells (Tregs) essential for the maintenance of self-tolerance. In this study, we show that a cell-intrinsic signal from CD28 is involved in the generation of cytokine-responsive Foxp3(-) precursors using studies of mixed bone marrow chimeras as well as TCR-specific generation of Foxp3(+) cells using intrathymic transfer of TCR-transgenic thymocytes expressing a natural Treg TCR. Contrary to a previous report, the analysis of CD28 mutant knockin mice revealed that this cell-intrinsic signal is only partially dependent on the Lck-binding PYAP motif. Surprisingly, even though the absence of CD28 resulted in a 6-fold decrease in thymic Tregs, the TCR repertoires of CD28-deficient and sufficient cells were largely overlapping. Thus, these data suggest that CD28 does not operate by markedly enlarging the repertoire of TCRs available for Treg development, but rather by improving the efficiency of Treg development of thymocytes expressing natural Treg TCRs.

Figure 1

The PYAP, but not Y170, motif of CD28 is required for thymic Treg cell development. (A) Normal CD4/CD8 thymic development in CD28-AYAA and -Y170F knock-in mice. Thymocytes from the indicated strains bred to the Foxp3^gfp reporter were analyzed by flow cytometry. (B) Decreased Foxp3⁺ CD4SP cells in AYAA mice. CD4SP gated cells were assessed for CD25 and Foxp3 by flow cytometry. (C) Decreased CD25^hiFoxp3⁻ frequency in CD28 KO mice. The percentage of Foxp3⁻ CD25^hi and Foxp3⁺ subsets from each strain is summarized (mean ± S.D.; n=12, WT and KO; n=8 AYAA, n=6 Y170F for A, B, and C). (D) Decreased expression of CD122 in CD28 KO CD25^hiFoxp3⁻ cells. Representative FACS plots for CD122 (IL-2Rβ) is shown for CD25^hiFoxp3⁻ and Foxp3⁺ CD4SP cells. The mean percentage and S.D. of CD122^hi cells in the CD25^hiFoxp3⁻ subset is indicated in the upper panels (n=8 WT and KO; n=7 Y170F; n=6 AYAA). ** p<0.01.

Figure 2

CD28 is involved in the generation of Foxp3⁻ Treg cell precursors. (A) Decreased frequency of cytokine responsive Treg cell precursors in CD28 KO and AYAA mice. CD25^hiFoxp3⁻ CD4SP cells were FACS purified from WT and indicated strains, cultured in vitro with hIL-2 (50 U/ml), and analyzed at 24 hours for Foxp3 expression by flow cytometry. Data were normalized to that of WT from the same experiment to account for inter-experimental variation. Each dot represents data from an individual experiment. Bars indicate the average normalized percentage of Treg cell precursors (n=8 WT and KO; n=5 AYAA, n=6 Y170F). (B) The effect of CD28 on the generation of Treg cells is cell-intrinsic. Lethally irradiated congenic hosts (CD45.1) were reconstituted with an equal mixture of bone marrow cells from WT (CD45.1) and WT, KO, or AYAA (all CD45.2) mice. Foxp3 and CD25 expression in CD45.2⁺CD4SP cells were analyzed by flow cytometry six weeks post-reconstitution. (C) Percentages of CD25^hiFoxp3⁻ and Foxp3⁺ cells within the CD45.2⁺CD4SP subset are summarized (n=12 WT and KO; n=8 AYAA; n=6 Y170F). (D) The effect of CD28 on the generation of Treg cell precursors is cell-intrinsic. FACS purified CD45.2⁺ Foxp3⁻ CD4SP cells were cultured in vitro with hIL-2 (50 U/ml) and Foxp3 expression was analyzed 24 hours later by flow cytometry. Data shown were normalized to that of WT cells in the same experiment. Each dot represents data from an individual mouse (n=6 mice, 4 independent experiments) and the bars indicate the average normalized percentage. ** p ≤0.01.

Figure 3

Interaction with B7 is required for TCR-specific Treg cell development. (A) Diminished TCR transgenic Treg cell development in B7 KO mice. Total thymocytes from G113 TCR transgenic Foxp3^gfp Rag1^−/−, which are largely devoid of Treg cells, were mixed with Thy1.1⁺ filler cells, labeled with DDAO-SE and transferred intrathymically into WT, CD80^−/− CD86^−/− (B7 KO), or CD28^−/− (CD28 KO) recipients. Expression of Foxp3 by DDAO⁺Vβ6⁺Thy1.1⁻ donor cells was analyzed by flow cytometry at the indicated time. Each dot represents data from an individual mouse and the bar the average frequency. Data are pooled from at least two independent experiments (total mice: n=4, d1 WT and B7 KO; n=5, d2 WT; n=4, d2 B7 KO; n=13, d3 WT; n=10, d3 B7 KO; n=6, d3 CD28 KO). ** p<0.01.

Figure 4

Large overlap between the CD28 KO and WT TCR repertories. (A and B) The Foxp3⁺ and Foxp3⁻ TCR repertoires are distinct in both WT and CD28 KO mice, with minor overlap. TRAV14 TCRα sequences from CD28 WT (A) and KO (B) mice were obtained as described in the Methods and Table S1. The frequencies of the top 10 Treg TCRs in the pooled data set, as well as their corresponding frequency in the Foxp3⁻ data set, are shown. (C and D) Large overlap between the CD28 WT and KO Treg TCR repertoires. The top 10 Treg TCRs from the CD28 WT (C) or KO (D) data sets are shown, and their percentages in the Foxp3⁺ and Foxp3⁻ data sets are shown. Data from individual data sets are shown in Fig. S5.

Figure 5

Differential requirement of CD28 for thymic selection. TCRs with an average frequency in the Foxp3⁺ CD28 WT and KO data sets of > 0.25% were plotted (58 TCRs, grey dots). The same process was also used to select Foxp3⁻ TCRs (71 TCRs, unfilled dots). These TCRs represent 29.9%, 56.2%, 35.4%, and 39.1% of the total data set for WT Foxp3⁺, KO Foxp3⁺, WT Foxp3⁻, and KO Foxp3⁻ subsets, respectively. The dashed line indicates equal frequency in the CD28 WT and KO data sets. Note that there are 4 and 10 TCRs found only in the CD28 WT or KO Foxp3⁺ data sets, respectively. By contrast, one TCR is found only in the CD28 WT and KO Foxp3⁻ data sets. Using the non-parametric Wilcoxon Rank sum test to compare the TCRs above the 0.25% threshold, TCR usage in Foxp3⁺ CD28 WT and KO cells is different (p < 0.01), whereas TCR usage in Foxp3⁻ cells is not (p = 0.7).