CD4+ regulatory T cells require CTLA-4 for the maintenance of systemic tolerance

Abstract

Cytotoxic T lymphocyte antigen-4 (CTLA-4) plays a critical role in negatively regulating T cell responses and has also been implicated in the development and function of natural FOXP3(+) regulatory T cells. CTLA-4-deficient mice develop fatal, early onset lymphoproliferative disease. However, chimeric mice containing both CTLA-4-deficient and -sufficient bone marrow (BM)-derived cells do not develop disease, indicating that CTLA-4 can act in trans to maintain T cell self-tolerance. Using genetically mixed blastocyst and BM chimaeras as well as in vivo T cell transfer systems, we demonstrate that in vivo regulation of Ctla4(-/-) T cells in trans by CTLA-4-sufficient T cells is a reversible process that requires the persistent presence of FOXP3(+) regulatory T cells with a diverse TCR repertoire. Based on gene expression studies, the regulatory T cells do not appear to act directly on T cells, suggesting they may instead modulate the stimulatory activities of antigen-presenting cells. These results demonstrate that CTLA-4 is absolutely required for FOXP3(+) regulatory T cell function in vivo.

Figure 1.

Ctla4^+/+ cells can regulate Ctla4^−/− T cells in mixed BM and blastocyst chimaeras. (A) Healthy Rag1^−/− mice received a total of ∼4 × 10⁶ T cell–depleted BM cells from 2–3-wk-old Ctla4^−/− mice (KO only) or a 1:1 mix from Ctla4^−/− and WT mice (KO:WT). Blastocyst chimaeras were generated as per Materials and methods, and all chimaeras were monitored for disease incidence. Mice were considered healthy (% Healthy) if no evidence of prolonged weight loss (>2 wk), ruffled fur, hunched posture, or skin or eye inflammation were visible. Mice were also bled periodically to measure relative KO:WT T cell ratios. Ctla4^−/−-only recipients were killed for ethical reasons by 8 wk of age. (B) 12 blastocyst chimaeras, each line representing individual mice, were periodically bled over several years, and the frequency of Ctla4^−/− T cells among total T cells enumerated via flow cytometry. (C) 6–8-wk-old KO only or KO:WT chimaera groups were killed, and LN cells were isolated and analyzed for expression of activation markers CD44, CD62L, CD69, and incorporation of the thymidine analogue BrdU, by flow cytometry. Results gated on CD4⁺ LN T cells from representative chimaeras are shown. Data shown in A and C are representative of over a dozen experiments. Data in B are the cumulative results of at least three experiments.

Figure 2.

A diverse repertoire of Ctla4^+/+ T cells are required to regulate Ctla4^−/− T cells in vivo. Rag1^−/− recipient mice received a total of ∼4 × 10⁶ T cell–depleted BM cells from a 1:1 mix of Ctla4^−/− and WT mice (KO:WT), or Ctla4^−/− and 2CT, OTI, or OTII TCR transgenic Rag1/2-sufficient mice. Some mice received a 1:3 ratio of Ctla4^−/−:OTII BM cells. Mice were monitored for disease incidence (% Healthy), as described in Fig. 1 A. Data are pooled from two independent experiments.

Figure 3.

Ctla4^−/− T cell quiescence depends on the persistent presence of WT T cells. (A) WT:WT or Ctla4^−/−:WT (KO:WT) mixed BM chimaeras received depleting anti-CD90.1 mAb (no treatment control, PBS injected only, and depleted) as indicated in Materials and methods. 8 wk after initiating depletion, chimaeras were killed and analyzed. LN cells were isolated and stained for CD44, CD62L, and CD69. Incorporation of the thymidine analogue BrdU as a measure of cell turnover was also assayed. Data are gated on CD90.2⁺ CD4⁺ T cells. (B) Enumeration of T cells from the LN and spleens (SN) of depleted mice. Each circle represents individual mice. Data are representative of two independent experiments with at least three chimaeras per group.

Figure 4.

WT CD4⁺ T cells regulate Ctla4^−/− T cells in vivo. CD4⁺ or CD8⁺ T cells from WT mice and CD4⁺ Ctla4^−/− T cells from mixed BM chimaeras were sorted and transferred (10⁷ cells total) to Rag1^−/− recipients. (A) Representative cytometric data showing the activation profile of CD4⁺ T cells and the frequencies of CD25⁺ CD4⁺ T cells from each of the three recipient groups (WT CD4⁺ T cells alone, Ctla4^−/− CD4⁺ and WT CD8⁺, and Ctla4^−/− CD4⁺ and WT CD4⁺) 3 wk after transfer. (B) Splenic CD4⁺ T cell numbers from each group in A. Each group consisted of a minimum of two mice each and is representative of three experiments. Error bars denote the SD.

Figure 5.

CD4⁺25⁺ WT T cells are able to regulate Ctla4^−/− T cells in vivo. (A) WT T cell–depleted LN and spleen cells (KO) from healthy Ctla4^−/−:WT mixed BM chimaeras were transferred (2 × 10⁷ cells/mouse) either separately or with purified WT CD4⁺25⁺ T cells (2 × 10⁶ cells/mouse) into Rag1^−/− recipients. Frequency and activation status of CD4⁺ (A) and CD8⁺ (C) T cells from the LNs were analyzed in recipients 12 wk after transfer. (B) Quantitation of total, CD4⁺, and CD8⁺ T cell numbers in the spleen and LN of recipients at the time of analysis. Data are representative of three individual experiments with at least three mice per group. Error bars denote the SD.

Figure 6.

Expression of Gadd45β is decreased in naive Ctla4^−/− T cells from mixed BM chimaeras relative to the WT T cells. Semiquantitative RT-PCR analysis of the expression of select genes in regulated Ctla4^−/− T cells from mixed BM chimaeras with inconsistent results from the array experiments. Threefold dilutions of cDNA prepared from naive (CD44⁻CD62L⁺) WT or Ctla4^−/− T cells were analyzed for the genes indicated with documented functions in T cell activation and tolerance. Expression of β-actin was used as loading control. Data are representative of three independent experiments.

Figure 7.

Regulation of Ctla4^−/− T cells does not absolutely require signaling through the IL-10 receptor. Rag1^−/− mice received a 1:1 mix of 4 × 10⁶ BM cells from WT and either Ctla4^−/− or Il10ra^−/−Ctla4^−/− mice. Mice were killed at 12 wk after reconstitution, and CD4⁺FoxP3⁻ (A) or CD8⁺ (B) splenic T cells were analyzed for the activation markers CD44 and CD62L in WT and KO T cell populations. FoxP3⁺ frequency was also assessed in the CD4⁺ T cell subset (A, bottom). Data shown are representative flow cytometric results from the spleen of reconstituted mice (three mice per group) in two independent experiments.