Homeostatic maintenance of natural Foxp3(+) CD25(+) CD4(+) regulatory T cells by interleukin (IL)-2 and induction of autoimmune disease by IL-2 neutralization

Abstract

Interleukin (IL)-2 plays a crucial role in the maintenance of natural immunologic self-tolerance. Neutralization of circulating IL-2 by anti-IL-2 monoclonal antibody for a limited period elicits autoimmune gastritis in BALB/c mice. Similar treatment of diabetes-prone nonobese diabetic mice triggers early onset of diabetes and produces a wide spectrum of T cell-mediated autoimmune diseases, including gastritis, thyroiditis, sialadenitis, and notably, severe neuropathy. Such treatment selectively reduces the number of Foxp3-expressing CD25(+) CD4(+) T cells, but not CD25(-) CD4(+) T cells, in the thymus and periphery of normal and thymectomized mice. IL-2 neutralization inhibits physiological proliferation of peripheral CD25(+) CD4(+) T cells that are presumably responding to normal self-antigens, whereas it is unable to inhibit their lymphopenia-induced homeostatic expansion in a T cell-deficient environment. In normal naive mice, CD25(low) CD4(+) nonregulatory T cells actively transcribe the IL-2 gene and secrete IL-2 protein in the physiological state. IL-2 is thus indispensable for the peripheral maintenance of natural CD25(+) CD4(+) regulatory T cells (T reg cells). The principal physiological source of IL-2 for the maintenance of T reg cells appears to be other T cells, especially CD25(low) CD4(+) activated T cells, which include self-reactive T cells. Furthermore, impairment of this negative feedback loop via IL-2 can be a cause and a predisposing factor for autoimmune disease.

Figure 1.

Neutralization of IL-2 by specific mAbs reduces the number of Foxp3 ⁺ CD25⁺ CD4⁺ T reg cells in normal naive mice. (A) Expression of CD122 (IL-2Rβ) and CD132 (IL-2Rγc) by CD25⁺ or CD25⁻ CD4⁺ T cells in the thymus or spleen of normal BALB/c mice. Dotted lines represent control staining with isotype-matched mAb. (B) Representative flow cytometric profiles of thymus and spleen cells from 8-wk-old BALB/c mice 14 d after injection of anti–IL-2 mAb or saline. Kinetics of the number of CD25⁺ CD4⁺ T cells and CD25⁻ CD4⁺ T cells after injection of anti–IL-2 mAb (black circles) or saline (white circles). Asterisks indicate significant differences (P < 0.05). (C) Relative Foxp3 mRNA levels assessed by real-time quantitative PCR with CD4⁺ CD8⁻ thymocytes or CD4⁺ splenocytes from mice treated with anti–IL-2 mAbs. Quantity of Foxp3 normalized to HPRT. (D) BALB/c nude mice were i.v. transferred with spleen cells (3 × 10⁷) from BALB/c mice injected with anti–IL-2 mAbs or saline 14 d earlier with or without CD25⁺ CD4⁺ T cells (10⁶) from nonmanipulated mice and examined histologically and serologically 3 mo later. Titers of anti-parietal cell autoantibodies were assessed by ELISA. Black circles, macroscopically and histologically evident grade 2 gastritis; gray circles, histologically evident grade 1 gastritis; white circles, intact gastritis mucosa. See reference for histological grading.

Figure 2.

IL-2 neutralization decreases the number of CD25⁺ CD4⁺ T cells in Tx mice. Adult BALB/c mice were Tx (Tx, +) or sham operated (Tx, −). 3 wk later, animals were injected with anti–IL-2 mAb or saline and the percentages of CD25⁺ CD4⁺ T cells among the CD4⁺ T cells (A) and their absolute numbers (B) were determined on day 14 after antibody injection.

Figure 3.

Neutralization of IL-2 inhibits natural in vivo proliferation of peripheral CD25⁺ CD4⁺ T cells in normal mice. BALB/c mice received one injection of 1 mg anti–IL-2 mAb or saline and were treated with BrdU for the next 3 d. Percentage of BrdU-labeled CD25⁺ or CD25⁻ CD4⁺ T cells from anti–IL-2–treated mice or saline-treated mice was assessed by flow cytometry using FITC-labeled anti-BrdU mAb. (A) Representative flow cytometric profiles. (B) Percent of BrdU⁺ cells among CD25⁺ or CD25⁻ CD4⁺ T cells in mice treated with anti–IL-2 mAb or saline. NS, not significant.

Figure 4.

IL-2–independent lymphopenia-driven expansion of CD25⁺ CD4⁺ T cells. (A) CD25⁺ CD4⁺ T cells (3 × 10⁵) purified from Thy-1.1 BALB/c mice were labeled with CFSE and injected into RAG-2^−/− mice that were treated with anti–IL-2 mAb or control rat IgG and assessed for the percentage of CFSE-labeled cells 4 d later. (B) CD25⁺ CD4⁺ T cells (3 × 10⁵) highly purified from BALB/c-Thy1.1 mice were transferred to BALB/c IL-2^+/+ or IL-2^−/− BALB/c RAG-2^−/− mice. A group of IL-2^−/− RAG-2^−/− BALB/c mice transferred with CD25⁺ CD4⁺ T cells were treated with anti–IL-2 mAb. Another group of IL-2^−/− RAG-2^−/− BALB/c mice received a cotransfer of an equal number (3 × 10⁵) of CD25⁺ CD4⁺ T cells and CD25⁻ CD4⁺ T cells sorted from the lymph nodes of normal BALB/c mice. The numbers of Thy1.1⁺ TCRβ⁺ cells recovered in the spleen of individual recipients were assessed 4 wk after transfer.

Figure 5.

Active transcription of the IL-2 gene in CD25^low CD4⁺ nonregulatory T cells in the spleens of normal BALB/c mice. (A) Splenic CD4⁺ T cells were sorted into CD25^high, CD25^low, and CD25⁻ subpopulations as shown and compared with other cell populations in the spleen (see Results) for expression levels of IL-2 mRNA assessed by real-time quantitative RT-PCR (normalized to HPRT). The average IL-2 mRNA levels of total unseparated spleen cells was defined to be 1. The level of IL-2 mRNA in in vitro–activated CD25⁻ CD4⁺ T cells (stimulated for 24 h with precoated 1 μg/ml anti-CD3 and 1 μg/ml of soluble anti-CD28) was 1,000–1,500 (not depicted). Sorted CD25^high, CD25^low, and CD25⁻ CD4⁺ T cells were cultured for 24 h, and the IL-2 concentrations in the supernatant were determined by ELISA. The results of three independent experiments are shown. (B) CD25^high, CD25^low, and CD25⁻ CD4⁺ populations from DO11.10 TCR transgenic mice (DO.RAG⁺) or the CD25⁻ CD4⁺ population from RAG^−/− DO11.10 transgenic mice (DO.RAG⁻) were assessed for the expression of IL-2 mRNA by real-time PCR and for the expression of transgenic TCR by FACS analysis using the clonotypic KJ1-26 mAb. (C) Relative Foxp3 mRNA levels in CD25^high, CD25^low, or CD25⁻ CD4⁺ T cells from the spleens of DO.RAG⁺, DO.RAG⁻, or normal BALB/c mice. (D) In vitro responsiveness and suppressive activity of splenic CD25^high or CD25^low CD4⁺ T cells from normal BALB/c mice. These populations were stimulated for 3 d with anti-CD3 mAb and APC either alone or in the presence of an equal number of CD25⁻ CD4⁺ responder T cells, and [³H]thymidine incorporation was measured. A representative of three independent experiments is shown in A–D. N.D., none detected.

Figure 6.

Induction of autoimmune gastritis in BALB/c mice by IL-2 neutralization. BALB/c mice were i.p. injected with 1 mg purified anti–IL-2 mAb or saline on days 10 and 20 after birth and were examined at 3 mo of age for histological and serological evidence for gastritis. (A) Hematoxylin and eosin staining of gastric mucosa of anti–IL-2–treated and saline-treated mice (a magnification of 20). (B) Titers of anti-parietal cell autoantibodies were determined by ELISA. The gastric lesion was histologically graded as described in Fig. 1 D. (C) 3 × 10⁷ spleen cells from anti–IL-2–treated mice showing gastritis or from saline-treated control mice were i.v. transferred to BALB/c athymic nude mice, which were histologically and serologically examined 3 mo later. Gastritis was scored as in Fig. 1 D.

Figure 7.

Exacerbation of diabetes and induction of other organ-specific autoimmune diseases in NOD mice by IL-2 neutralization. Female NOD mice were injected i.p. with 1 mg of purified anti–IL-2 mAbs (n = 16) or rat IgG (n = 16) on days 10 and 20 after birth. Their blood glucose levels were measured every week starting from 7 wk of age. (A) Incidence of diabetes. (B) Hematoxylin and eosin staining of pancreas, thyroid, and lachrymal glands (a magnification of 5–10). (C) Severity of insulitis. Grade 1, peri-insulitis; grade 2, moderate insulitis; grade 3, severe insulitis (reference 59). (D and E) Titers of anti-parietal cell autoantibodies (D) and anti-thyroglobulin autoantibodies (E) in the sera of anti–IL-2–treated mice (n = 11) and rat IgG-treated mice (n = 12) assessed by ELISA.

Figure 8.

Spontaneous development of autoimmune neuropathy in NOD mice treated with anti–IL-2 mAb. Female NOD mice treated as shown in Fig. 7 were checked weekly for clinical signs of neuropathy (hind and front leg paralysis). (A) Incidence of macroscopically evident neuropathy. (B) A mouse with paralysis of hind limbs. (C) Hematoxylin and eosin staining of the sections of the peripheral nerves (left; a magnification of 20) and sciatic nerves (middle; a magnification of 40). Luxol fast blue staining of peripheral nerves (right; a magnification of 40). (D) A NOD.SCID mouse showing paralysis of hind legs after receiving CD4⁺ T cells from anti–IL-2–treated NOD mice with neuropathy (left). Hematoxylin and eosin staining of sciatic nerve from the NOD.SCID mouse (right; a magnification of 40). (E) Spleen cells from NOD mice with neuropathy were sorted to CD4⁺ and CD4⁻ cells by MACS, and 4–6 × 10⁶ cells of each population were transferred to NOD.SCID mice. Grade 1, lymphocytic infiltration in the peripheral nerves without clinical signs; grade 2, clinical signs of neuropathy with lymphocytic infiltration in peripheral nerves.

Figure 9.

Development of various autoimmune diseases in NOD mice after anti–IL-2 treatment. Individual NOD female mice that were treated as shown in Fig. 7 and individual mice that survived to 20 wk of age (11 anti–IL-2–treated mice and 12 control mice) were assessed clinically and histologically. DM, diabetes mellitus; NP, neuropathy; Neur, neuritis; Ins, insulitis; Gas, gastritis; Thyr, thyroiditis; Sial, sialadenitis; lacr, lacrimal gland adenitis.