Abstract
The tumor suppressor p53 plays a central role in tumor suppression by inducing apoptosis, cell cycle arrest, senescence, and DNA repair. In addition to the antitumor functions of p53, accumulating evidence using systemic p53-deficient mice suggests that p53 suppresses autoimmunity. However, it remains unknown how p53 suppresses autoimmunity. In this study, we generated T cell–specific p53-deficient mice (CD4-Cre p53fl/fl mice, or p53 conditional knockout [cKO] mice) and found that aged p53-cKO mice spontaneously developed inflammatory lesions in various organs, including lung, liver, stomach, thyroid gland, submandibular gland, and kidney. Additionally, anti-nuclear Abs and autoantibodies against gastric parietal cells were detected in p53-cKO mice but not in control p53fl/fl mice (p53 wild-type mice). Importantly, the number of Foxp3+CD4+ regulatory T cells (Tregs) in the spleen and lung as well as in vitro differentiation of induced Tregs was significantly reduced in p53-cKO mice as compared with that in p53 wild-type mice. Regarding the mechanisms underlying p53-mediated Treg induction, p53 enhanced the transcription of Foxp3 by binding to the promoter and the conserved noncoding DNA sequence-2 of the Foxp3 gene. Taken together, these results suggest that p53 expressed in T cells functions as a suppressor for autoimmunity by inducing Treg differentiation.
This article is featured in In This Issue, p.3487
Introduction
The tumor suppressor p53 protein, a transcription factor that is activated by various DNA damage and cellular stress, plays a central role in the suppression of tumorigenesis (1, 2). A number of p53 target genes that regulate apoptosis and cell cycle progression have been identified, and a number of mechanisms underlying antitumor effects of p53 have been identified in various cells (1, 2). Hence, p53 has been considered as the guardian of the genome.
Recently, p53 has been implicated in the suppression of autoimmune and inflammatory diseases in both humans and mice. It has been shown that patients with systemic lupus erythematosus have an autoantibody to the C-terminal domain of p53, which can inhibit p53 functions (3). Additionally, Hara et al. (4) have reported that serum levels of anti-p53 Ab are elevated in patients with systemic sclerosis. Moreover, some patients with rheumatoid arthritis have been shown to have p53 gene mutations in synovial cells (5, 6). In mice, it has been demonstrated that systemic p53-deficient mice exhibit accelerated collagen-induced arthritis and Ag-induced arthritis (7, 8). Additionally, systemic p53-deficient mice are more susceptible to experimental autoimmune encephalomyelitis and autoimmune diabetes than wild-type (WT) mice (9, 10). Moreover, a recent study has reported that systemic p53-deficient CD45.1 mice develop an autoimmune glomerulonephritis-like disease with elevated serum levels of proinflammatory cytokines, including IL-17 and IL-6 (11). Although these findings suggest that p53 is involved in the suppression of autoimmunity, it remains unknown how p53 suppresses autoimmunity.
More recently, it has been shown that the expression of Foxp3, a master regulator of regulatory T cells (Tregs) (12–14), is upregulated by p53 in breast and colon carcinoma cells in humans (15). Furthermore, p53 has been shown to be expressed in CD4+ T cells after TCR-mediated stimulation (16). Based on these findings, we hypothesized that p53 is involved in the regulation of Foxp3 expression in CD4+ T cells and thus plays a role in the maintenance of self-tolerance.
In this study, we tested the hypothesis by generating T cell–specific p53-deficient (CD4-Cre p53fl/fl, or p53-conditional knockout [cKO]) mice. We found that aged p53-cKO mice developed inflammatory diseases such as interstitial pneumonitis, hepatitis, gastritis, thyroiditis, sialadenitis, and glomerulonephritis spontaneously, and that the number of Tregs was significantly decreased in the spleen of p53-cKO mice as compared with that in control p53fl/fl mice (p53-WT mice). Additionally, we found that p53 enhanced the transcription of Foxp3 by binding to the promoter and the conserved noncoding DNA sequence (CNS)-2 of the Foxp3 gene. These findings indicate that p53 expressed in T cells induces Foxp3+ Tregs and suppresses autoimmunity.
Materials and Methods
Mice
CD4-Cre mice (17) and p53fl/fl mice (18) were backcrossed onto C57BL/6 mice (Japan SLC, Shizuoka, Japan) for ten generations. CD4-Cre mice were bred to p53fl/fl mice to generate T cell–specific p53-cKO (CD4-Cre p53fl/fl) mice. Littermate p53fl/fl mice (p53-WT mice) were used as controls. All mice were housed in microisolator cages under specific pathogen-free conditions and all experiments were performed according to the guidelines of Chiba University.
Reagents
Histological analysis
Samples of the lung, liver, stomach, thyroid gland, submandibular gland, kidney, pancreas, and small and large intestine were fixed in 10% buffered formalin and embedded in paraffin. Sections (3 μm thick) were stained with H&E. Histological analysis was performed using light microscopy and AxioVision software (Carl Zeiss Microimaging, Oberkochen, Germany).
Flow cytometric analysis
Measurement of serum Igs and anti-nuclear Ab
Sera were obtained from p53-cKO mice and littermate p53-WT mice at 6 and 9 mo of age. The levels of anti-nuclear Ab (ANA) were determined by a Mesacup ANA kit (MBL, Nagoya, Japan). The levels of total IgG were determined by an SBA clonotyping system/HRP kit (SouthernBiotech, Birmingham, AL).
Detection of autoantibodies against gastric parietal cells
Stomach of C57BL/6 mice was snap-frozen in OCT compound (Sakura Finetek USA, Torrance, CA), and cryosections (5 μm thick) were incubated with sera from p53-cKO mice and littermate p53-WT mice (9 mo of age) for 30 min at room temperature. After washing, sections were incubated with FITC-labeled goat anti-mouse IgG (Santa Cruz Biotechnology, Santa Cruz, CA) in PBS buffer (5% FCS) for 30 min at room temperature. After washing, the presence of autoantibodies against gastric parietal cells was visualized using LSM 710 confocal laser microscopy and ZEN 2011 software (Carl-Zeiss Microimaging) immediately after staining.
Cell isolation
Single-cell suspensions of splenocytes were prepared from 8-wk-old p53-cKO mice and p53-WT mice. CD4+CD62L+CD25−TCRγδ− naive T cells were isolated from splenocytes by using a CD4+CD62L+ T cell isolation kit II (Miltenyi Biotec, Auburn, CA). The resultant cells were >98% pure CD4+CD62L+ T cells by flow cytometry.
Cell culture
Naive CD4+ T cells (1 × 106 cells/ml) were stimulated with plate-bound anti-CD3 mAb (1 μg/ml, 145-2C11) in the presence of anti-CD28 mAb (1 μg/ml, 37.51) in RPMI 1640 medium supplemented with 10% heat-inactivated FCS and 50 μM 2-ME (complete medium). Where indicated, cells were cultured under either neutral (Th0) conditions (anti–IFN-γ mAb [5 μg/ml, XMG1.2] and anti–IL-4 mAb (5 μg/ml, 11B11]), Th17 conditions (IL-6 [100 ng/ml], TGF-β [1 ng/ml], anti–IFN-γ mAb, and anti–IL-4 mAb), IL-6 conditions (IL-6 [100 ng/ml], anti–IFN-γ mAb, and anti–IL-4 mAb), Treg conditions (TGF-β [1 ng/ml], anti–IFN-γ mAb, and anti–IL-4 mAb), Th1 conditions (IL-12 [10 ng/ml] and anti–IL-4 mAb), or Th2 conditions (IL-4 [10 ng/ml] and anti–IFN-γ mAb).
Proliferation of CD4+ T cells
Naive CD4+ T cells (4 × 105 cells) were isolated from spleen of 8-wk-old p53-cKO mice and p53-WT mice and stimulated with anti-CD3 mAb/anti-CD28 mAb for 3 d. Cell proliferation was examined by using a premix WST-1 cell proliferation assay system (TaKaRa Bio, Shiga, Japan) according to the manufacturer’s instructions.
Intracellular staining for Foxp3, p53, and CTLA-4
Intracellular staining of Foxp3, p53, and CTLA-4 was performed using anti–Foxp3-FITC (259D/C7; BD Biosciences), anti–p53-A647 (1C12; Cell Signaling Technology), and anti–CTLA-4-allophycocyanin (UC10-4B9; BD Biosciences) as described previously (19).
Intracellular cytokine staining
19).
ELISA for IL-6
The levels of IL-6 were measured by a mouse IL-6 ELISA Ready-SET-Go! kit (eBioscience).
Isolation of lung leukocytes
Lung was cut into small pieces and incubated in RPMI 1640 medium containing collagenase A (1 mg/ml; Roche, Basel, Switzerland) for 10 min at 37°C. A suspension of single cells was separated from undigested tissue by passing through a 0.07-mm nylon mesh. Viable lymphocytes were collected from the cell suspension using Lympholyte-M solution (Cedarlane Laboratories, Burlington, ON, Canada) according to the manufacturer’s instructions.
Isolation of lamina propria lymphocytes from the colon
Lamina propria lymphocytes were isolated from the colon of 8-wk-old p53-cKO mice and p53-WT mice as described previously (20).
Induction and assessment of colitis
p53-cKO mice and p53-WT mice at the age of 8 wk were given 1.5% dextran sulfate sodium (molecular mass 36,000–50,000 Da; MP Biomedicals, Solon, OH) dissolved in drinking water provided ad libitum for 7 d. Changes in body weight were calculated as follows: body weight change (%) = [(weight on a given day (days 0–7) − weight on day 0)/weight on day 0] × 100. Histological scoring of colitis was performed by a pathologist in a blinded fashion as described previously (21).
Suppression assay
CD4+CD25+ T cells were isolated from spleen with a CD4+CD25+ Treg isolation kit (Miltenyi Biotec). The resultant cells were >98% pure CD4+CD25+ T cells by flow cytometry. Naive CD4+ T cells from C57BL/6 WT mice were labeled with CFSE (0.5 μM, Vybrant CFSE cell tracer kit; Invitrogen, Carlsbad, CA) and used as responder cells. CFSE-labeled naive CD4+ T cells (1 × 105 cells/well) were stimulated with soluble anti-CD3 mAb (1 μg/ml) in the presence of CD3-depleted irradiated splenocytes for 3 d. Where indicated, isolated CD4+CD25+ T cells were added to the culture at the indicated ratios. CFSE dilution profiles of responder cells were assessed by flow cytometry and data were transformed into the division index using FlowJo software.
Quantitative real-time PCR analysis
Total RNA was prepared and reverse transcription was carried out as described previously (22). The mRNA expression levels of retinoic acid–related orphan receptor (ROR)γt, Foxp3, T-bet, Gata3, and IL-10 were determined by quantitative real-time PCR analysis (qPCR) with a standard protocol on an ABI Prism 7300 instrument (Applied Biosystems, Foster City, CA) using a Power SYBR Green PCR Master Mix (Applied Bioscience). The levels of each expression were normalized to the levels of β-actin. The sequences of PCR primers were as follows: RORγt, forward, 5′-TGCAAGACTCATCGACAAGGC-3′, reverse, 5′-AGCTTTTCCACATGTTGGCT-3′; Foxp3, forward, 5′-ATCCAGCCTGCCTCTGACAAGAACC-3′, reverse, 5′-GGGTTGTCCAGTGGACGCACTTGGAGC-3′; T-bet, forward, 5′-CAACAACCCCTTTGCCAAAG-3′, reverse, 5′-TCCCCCAAGCAGTTGACAGT-3′; Gata3, forward, 5′-GCGGGCTCT ATCACAAAATGA-3′, reverse, 5′-GCTCTCCTGGCTGCAGACAGC-3′; IL-10, forward 5′- ACCTGCTCCACTGCCTTGCT-3′, reverse, 5′-GGTTGCCAAGCCTTATCGGA-3′; β-actin, forward, 5′-TGTTACCAACTGGGACGACA-3′, reverse, 5′-CCATCACAATGCCTGTGGTA-3′.
Western blotting
Whole-cell extracts were prepared and immunoblotting was performed as described previously (22) using antisera to p53 (Cell Signaling Technology) and HSP90 (Santa Cruz Biotechnology).
Reporter constructs
pGL3-Foxp3 promoter (Foxp3P)-Luc (−500 to +100) (pFoxp3pro2; promoter 2), pGL3-Foxp3P/CNS-1 (+2022 to +2721)-Luc (pFoxp3pro2/CNS-1; CNS-1), and pGL3-Foxp3P/CNS-2 (+4001 to +4820)-Luc (pFoxp3pro2/CNS-2; CNS-2) (23) were gifts from Dr. Warren Leonard (National Heart, Lung, and Blood Institute, National Institutes of Health, Bethesda, MD). pGL4-Foxp3P-Luc (−1702 to +174) (pFoxp3pro1; promoter 1) (24) was provided by Dr. Masahide Tone (University of Pennsylvania, Philadelphia, PA).
Luciferase reporter assay
Jurkat cells were transfected with the indicated plasmids using a Gene Pulser (Bio-Rad, Hercules, CA). After cells were cultured at 37°C for 12 h in the complete medium, aliquots of cells were left untreated or treated for another 12 h with PMA (50 ng/ml) and ionomycin (1 μg/ml). Cells were lysed, and relative light units were assessed with a dual luciferase assay system (Promega Biotech, Madison, WI). Firefly luciferase activity of reporter constructs was normalized by Renilla luciferase activity of pRL-TK. All values were obtained from experiments carried out in triplicate.
Chromatin immunoprecipitation-qPCR assays
Chromatin immunoprecipitation (ChIP) assays were performed using a ChIP assay kit (Upstate Biotechnology, Lake Placid, NY) as described previously (25). In brief, after cross-linking with 1% formaldehyde for 10 min at 37°C, cells were lysed in SDS buffer (50 mM Tris-Cl [pH 8.1], 10 mM EDTA, 1% SDS, and protease inhibitors). Lysates were sonicated using a Bioruptor (Cosmo Bio, Tokyo, Japan) and diluted with ChIP dilution buffer (20 mM Tris-Cl [pH 8.1], 1 mM EDTA, 150 mM NaCl, and 0.3% Triton X-100). Lysates were then precleared with protein A/G agarose for 30 min and incubated with anti-p53 mAb (1C12) or control rabbit IgG at 4°C overnight. Protein A/G agarose was then added to the lysates and incubated for 2 h at 4°C. Ab/protein/DNA complexes were eluted and incubated at 65°C overnight to reverse formaldehyde cross-linking. DNA fragments were amplified and DNA contents were measured by qPCR with specific primers. Primer sequences for ChIP-qPCR analyses were as follows: promoter of Foxp3 gene, forward, 5′-CTGAGGTTTGGAGCAGAAGGA-3′, reverse, 5′-TCTGAAGCCTGCCATGTGAA-3′; CNS-1 of Foxp3 gene, forward, 5′-GTTTTGTGTTTTAAGTCTTTTGCACTTG-3′, reverse, 5′-CAGTAAATGGAAAAAATGAAGCCATA-3′; CNS-2 of Foxp3 gene, forward, 5′-CATCCGCTAGCACCCACAT-3′, reverse, 5′-CTACCCCACAGGTTTCGTT -3′. Data were expressed as the percentage input for each ChIP fraction.
Data analysis
Data are summarized as means ± SD. The statistical analysis of the results was performed by unpaired t test. A p value < 0.05 was considered statistically significant.
Results
Spontaneous development of inflammatory diseases accompanied by autoantibodies in p53-cKO mice
The precise roles of p53 in autoimmunity are difficult to be addressed by using systemic p53-deficient mice because such mice die of malignancies at earlier ages (26). Accordingly, to address the roles of p53 in immune homeostasis and the development of autoimmune diseases, we generated p53-cKO mice and compared immunological phenotypes of p53-cKO mice with those in control p53-WT mice. As predicted by the previous study using systemic p53-deficient mice (26), 20% of p53-cKO mice developed lymphoma-like diseases with marked splenomegaly (Supplemental Fig. 1A), increased numbers of splenocytes (Supplemental Fig. 1B), and increased numbers of CD3+CD4−CD8− cells with high forward and side scatter characteristics (Supplemental Fig. 1C) by the age of 9 mo. Consequently, p53-cKO mice having lymphoma-like diseases were excluded from subsequent analyses.
Even in p53-cKO mice lacking lymphoma-like diseases, such mice exhibited reduced body weight gain as compared with p53-WT mice (Fig. 1A). Histological analyses of the organs revealed that p53-cKO mice had inflammatory lesions in various organs, including lung, liver, stomach, thyroid gland, kidney, submandibular gland, and pancreas at the age of 9 mo (Fig. 1B, Table I). As expected, no inflammatory lesions were found in age-matched p53-WT mice (Table I). An incidence of inflammatory diseases was higher in 9-mo-old p53-cKO mice than that in 6-mo-old p53-cKO mice (100 versus 50%) (Table I). Additionally, serum levels of ANA and total IgG were significantly increased in p53-cKO mice as compared with those in age-matched p53-WT mice (Fig. 1C). Furthermore, indirect immunofluorescence revealed the presence of autoantibodies against gastric parietal cells in the sera of p53-cKO mice but not p53-WT mice (Fig. 1D). Alternatively, IL-6, which is well known to enhance IgG production (27), was not detected in sera of 9-mo-old p53-cKO mice and p53-WT mice by ELISA (n = 4).
p53-cKO mice develop inflammatory diseases in multiple organs. (A) p53-cKO mice exhibit impaired growth as compared with p53-WT mice. Shown are means ± SD of body weight (BW) of p53-WT mice and p53-cKO mice without lymphoma-like diseases (n = 20, each). **p < 0.005 versus the mean value of p53-WT mice. (B) Representative photomicrographs (H&E staining) of indicated organs of p53-WT mice and p53-cKO mice at 9 mo of age. Arrowheads indicate areas of mononuclear cell infiltration. Scale bars, 100 μm. (C) Elevated ANA and total IgG in p53-cKO mice. Sera were obtained from p53-WT mice and p53-cKO mice at 6 and 9 mo of age and subjected to ELISA for antinuclear Ab and IgG. Data are means ± SD (n = 6, each). *p < 0.05, **p < 0.005. (D) Presence of autoantibodies against gastric parietal cells in sera of p53-cKO mice. Cryosections of the stomach of C57BL/6 WT mice were incubated with sera of p53-cKO mice or p53-WT mice at 9 mo of age, followed by incubation with FITC-conjugated anti-mouse IgG. Shown are representative of five independent experiments. Scale bars, 100 μm.
To further elucidate the characteristics of inflammatory diseases in p53-cKO mice, cellular phenotype of CD4+ T cells in the lung of 6-mo-old p53-cKO mice with inflammatory diseases was examined. The number of CD44+CD62low memory CD4+ T cells was significantly increased in the lung of p53-cKO mice as compared with that in p53-WT mice (Fig. 2A). Additionally, the frequencies of IL-17A–producing CD4+ T cells and TNF-α–producing CD4+ T cells were significantly increased in the lung of p53-cKO mice (Fig. 2B), consistent with a recent report showing the increased Th17 cell responses in systemic p53-deficient mice (11). Alternatively, the frequencies of IFN-γ– or IL-13–producing CD4+ T cells were not significantly increased in the lung of p53-cKO mice (Fig. 2B). These results suggest that p53 expressed in T cells plays an indispensable role in the prevention of Th17 cell responses and autoimmunity.
p53-cKO mice develop Th17 type inflammation in the lung. (A) Memory CD4+ cells are increased in the lung of p53-cKO mice. Shown are representative dot plots of CD44 versus CD62L staining gating on CD4+CD3+ cells and means ± SD of the numbers of CD44+CD62lowCD4+ cells and CD44−CD62highCD4+ cells in the lung of 6-mo-old p53-WT mice and p53-cKO mice (n = 4, each). **p < 0.005. (B) Th17 cells are increased in the lung in p53-cKO mice. Lymphocytes isolated from the lung of 6-mo-old p53-WT mice and p53-cKO mice were stimulated with PMA/I for 4 h and the expression of IFN-γ, IL-13, IL-17A, IL-17F, TNF-α, and IL-6 in CD4+ cells was analyzed by intracellular staining. Shown are representative dot plots of IFN-γ versus IL-13, IL-17A versus IL-17F, and TNF-α versus IL-6 of CD4+ T cells and means ± SD of the frequencies of IL-17A+CD4+ cells and TNF-α+CD4+ cells (n = 4, each). *p < 0.05, **p < 0.005.
The development of Tregs in the periphery is reduced in p53-cKO mice
We next examined the differentiation of T cells in young healthy p53-cKO mice (2 mo old). The numbers of thymocytes and splenocytes as well as the frequency of CD4+ T cells and CD8+ T cells in p53-cKO mice were indistinguishable from those in p53-WT mice (Fig. 3A–C). We thus examined Treg and effector CD4+ T cell differentiation in p53-cKO mice. As shown in Fig. 3D and 3E, no significant difference was observed in the frequency and number of thymic Foxp3+CD4+CD8− naturally occurring Tregs (nTregs) between p53-WT mice and p53-cKO mice. In contrast, the numbers of Foxp3+CD4+CD3+ cells (Tregs) in the spleen, lung, and lamina propria of the colon were significantly reduced in p53-cKO mice as compared with those in p53-WT mice (Fig. 3D, 3E). Consistent with the reduced number of Tregs, the susceptibility to dextran sulfate sodium–induced colitis, in which iTregs have been shown to play an inhibitory role (28), was significantly enhanced in p53-cKO mice as compared with that in p53-WT mice (Supplemental Fig. 2).
The numbers of Tregs in the periphery are decreased in p53-cKO mice. (A) Shown are the numbers of thymocytes and splenocytes in p53-WT mice and p53-cKO mice at 8 wk of age. Data are means ± SD (n = 5, each). (B and C) Thymocytes (B) and splenocytes (C) from p53-WT mice and p53-cKO mice were stained with Abs for CD4, CD8, CD3, and CD19 and analyzed by flow cytometry. Shown are representative dot plots of three independent experiments. (D and E) Single-cell suspensions of thymus, spleen, lung, and lamina propria of the colon of p53-WT mice and p53-cKO mice (n = 5, each) at 8 wk of age were stained with Abs against CD3, CD4, CD8, and Foxp3 and analyzed by flow cytometry. (D) Shown are representative dot plots of CD4 versus Foxp3 staining of CD8− cells (thymus) and CD3+CD4+ cells (spleen, lung, and colon) of five independent experiments. (E) Shown are means ± SD (n = 5, each) of the numbers of Foxp3+CD4+ cells in each organ. *p < 0.05.
To address the role of p53 in effector T cell differentiation in detail, we next stimulated naive CD4+ T cells from p53-cKO mice and p53-WT mice with anti-CD3 mAb/anti-CD28 mAb in Th0, Th1, Th2, Th17, and Treg conditions. Naive CD4+ T cells in p53-cKO mice showed similar levels of anti-CD3 mAb/anti-CD28 mAb–induced proliferation to those in p53-WT mice under nonpolarizing Th0 conditions (Supplemental Fig. 3). Consistent with the reduced numbers of Tregs in p53-cKO mice in vivo (Fig. 3D, 3E), mRNA levels of Foxp3 were significantly lower in p53-cKO mice than those in p53-WT mice under Treg conditions (Fig. 4A). In contrast, mRNA levels of RORγt were significantly increased in CD4+ T cells in p53-cKO mice as compared with those in p53-WT mice under Th17 conditions (Fig. 4A). Alternatively, no significant difference was observed in mRNA levels of T-bet or Gata3 between p53-cKO mice and p53-WT mice (Fig. 4A). Moreover, intracellular staining revealed that the frequency of Foxp3+ CD4+ T cells under Treg conditions was significantly decreased and the frequency of IL-17A–producing CD4+ T cells under Th17 conditions was significantly increased in p53-cKO mice as compared with that in p53-WT mice (Fig. 4B). Furthermore, even when anti–IL-6 Ab and IL-21R/Fc were added to the culture to block endogenously produced IL-6 and IL-21 that could inhibit Treg differentiation (29), the frequency of Foxp3+CD4+ T cells under Treg conditions was still significantly decreased in p53-cKO mice as compared with that in p53-WT mice (Fig. 4C). These results suggest that in addition to the previously described roles of p53 in the suppression of Th17 cells (11), p53 may be involved in the induction of Tregs in the periphery.
The differentiation of iTregs is impaired in p53-cKO mice. (A) Naive CD4+ T cells isolated from p53-WT mice or p53-cKO mice (8 wk old) were stimulated with anti-CD3 mAb/anti-CD28 mAb in neutral (Th0), Treg, Th17, Th1, or Th2 conditions for 48 h. Total RNA was isolated and mRNA expression of Foxp3, RORγt, T-bet, and Gata3 was measured by qPCR. Data are means ± SD (n = 4, each) of the relative expression to β-actin, **p < 0.005. (B) Naive CD4+ T cells isolated from p53-WT mice or p53-cKO mice were stimulated with anti-CD3 mAb/anti-CD28 mAb in Th0, IL-6, Treg, or Th17 conditions for 48 h. The expression of Foxp3 and IL-17A in CD4+ cells was analyzed by flow cytometry. Shown are representative dot plots of IL-17A versus Foxp3 on CD4+ cells and means ± SD (n = 5, each) of the frequency of IL-17A+Foxp3− cells and IL-17A−Foxp3+ cells. (C) Naive CD4+ T cells isolated from p53-WT mice or p53-cKO mice were stimulated with anti-CD3 mAb/anti-CD28 mAb in the absence (control IgG) or presence of anti–IL-6 Ab and IL-21R/Fc in Treg conditions for 48 h. The expression of Foxp3 and IL-17A in CD4+ cells was analyzed by flow cytometry. Shown are means ± SD (n = 3, each) of the frequency of IL-17A−Foxp3+ cells in CD4+ cells. *p < 0.05.
The function of Tregs is not impaired in p53-cKO mice at the individual cell level
We next examined whether p53 regulated the suppression activity of Tregs. Because the expression of CTLA-4, GITR, and CD25 has been shown to correlate with the suppressive activity of Tregs (30), we first examined the expression levels of these molecules on Foxp3+CD4+ T cells in the spleen in p53-cKO mice and p53-WT mice. As shown in Fig. 5A, the expression levels of GITR, CD25, and CTLA-4 on Foxp3+CD4+ T cells in p53-cKO mice were comparable to those in p53-WT mice. Moreover, the number of IL-10–producing cells in Foxp3+CD4+ T cells (Fig. 5B) as well as the level of IL-10 mRNA in CD4+ CD25+ T cells in p53-cKO mice (Supplemental Fig. 4) were similar to thise in p53-WT mice. Consistently, suppressive activity of isolated CD4+CD25+ T cells in p53-cKO mice was similar to that in p53-WT mice (Fig. 5C). These results indicate that suppression activity of Tregs at the individual cell level is not affected by the absence of p53.
Tregs in p53-cKO mice exhibit normal suppressive activities. (A) Splenocytes from p53-WT mice and p53-cKO mice were first stained for surface markers (CD4 and either GITR or CD25) and then stained with intracellular Foxp3 and CTLA-4. Shown are representative histograms and means ± SD (n = 6, each) of mean fluorescence intensity (MFI) of anti-GITR, anti-CD25, and anti–CTLA-4 staining of Foxp3+CD4+ T cells. Gray histograms indicate isotype-matched control staining. (B) Splenocytes from 8-wk-old p53-WT mice and p53-cKO mice were stimulated with PMA/I for 4 h and the expression of IL-10 was analyzed by intracellular staining. Shown are representative dot plots of IL-10 versus CD4 staining of Foxp3+CD4+ T cells and means ± SD of the frequency of IL-10–producing Foxp3+CD4+ cells (n = 3, each). (C) CFSE-labeled naive CD4+ T cells (responder T cells [Tresp]) were stimulated with soluble anti-CD3 mAb in the presence of T cell–depleted irradiated splenocytes. Where indicated, CD4+CD25+ cells isolated from either p53-WT mice or p53-cKO mice were added to the culture at indicated ratios. CFSE dilution profiles of Tresps were assessed by flow cytometry. and data were transformed into a division index using FlowJo software. Data are means ± SD (n = 5, each). Data are representative of two independent experiments.
TCR-mediated signaling induces p53 expression in CD4+ T cells
We next examined the regulation of p53 expression in CD4+ T cells. As shown in Fig. 6A, TCR stimulation induced p53 expression in CD4+ T cells in p53-WT mice but not in p53-cKO mice. Intracellular staining of p53 confirmed TCR-mediated induction of p53 in CD4+ T cells (Fig. 6B). Among CD4+ T cells, the expression levels of p53 were modestly but reproducibly higher in Foxp3+ Tregs than those in Foxp3- non-Tregs (Fig. 6B). Additionally, rottlerin, a inhibitor for protein kinase Cθ (31), which is required for the activation of NF-κB and AP-1 under TCR-mediated signaling (32), inhibited TCR-mediated p53 expression in CD4+ T cells in a dose-dependent manner (Fig. 6C). Alternatively, TGF-β did not affect p53 expression in CD4+ T cells in the presence or absence of TCR stimulation (Fig. 6D). These results suggest that TCR-mediated signaling but not TGF-β–mediated signaling induces p53 expression in CD4+ T cells.
TCR-mediated signaling induces p53 expression in CD4+ T cells. (A) Naive CD4+ T cells isolated from p53-WT mice or p53-cKO mice (8 wk old) were stimulated with or without anti-CD3 mAb/anti-CD28 mAb (TCR) for 6 h. Whole-cell extracts were prepared and subjected to Western blotting. Shown are representative immunoblottings with anti-p53 Ab and anti-HSP Ab (as a control) from three independent experiments. (B) Naive CD4+ T cells isolated from C57BL/6 mice were stimulated with TCR for indicated time periods, and the expression levels of p53 in Foxp3+CD4+ cells and Foxp3−CD4+ cells were assessed by intracellular staining. Gray histograms indicate isotype-matched control staining. Shown are representative histograms of p53 staining with mean fluorescence intensity (MFI) of five independent experiments. (C) Naive CD4+ T cells from C57BL/6 mice were stimulated with TCR for 24 h in the absence (DMSO) or presence of rottlerin (protein kinase Cθ inhibitor) at indicated concentrations. Shown are representative histograms of p53 staining of CD4+ T cells with MFI of three independent experiments. (D) Naive CD4+ T cells from C57BL/6 mice were stimulated with or without TCR in the absence (PBS) or presence of TGF-β (1 ng/ml) for 6 or 24 h. Shown are representative histograms of p53 staining of CD4+ T cells with MFI of four independent experiments.
p53 Directly enhances the transcription of Foxp3
It has been reported that Foxp3 is expressed in various tumor cells and functions as a transcriptional repressor of oncogenes in some tumor cells (33–36). Additionally, it has been shown that Foxp3 is induced by p53 in breast and colon carcinoma cells in humans (15). Moreover, it has recently been shown that the CNS elements at the Foxp3 locus encode information defining the size, composition, and stability of the Treg population (37). Because TCR stimulation induced p53 expression in CD4+ T cells (Fig. 6), we next examined whether p53 directly regulates the activity of the promoter and the CNS regions of the Foxp3 gene. To determine the effects of p53 on the transcriptional control of the Foxp3 gene, we first performed a luciferase reporter assay. As shown in Fig. 7A, p53 enhanced the activity of Foxp3 promoter (promoter 1, −1702 to +174) in the presence of PMA plus ionomycin (PMA/I), consistent with a previous report in tumor cells (15). Alternatively, small fragments of Foxp3 promoter (promoter 2, −500 to +100) did not respond to p53 in the presence or absence of PMA/I (Fig. 7B), suggesting that the upstream region (−1702 to −500) of the promoter contains a responsive element for p53. Importantly, luciferase reporter containing promoter 2 and CNS-2 (+4001 to +4820) but not promoter 2 and CNS-1 (+2022 to +2721) responded strongly to p53 in the presence of PMA/I (Fig. 7B), suggesting that the CNS-2 contains a responsive element for p53. Finally, to determine whether p53 binds to the promoter and the CNS-2 of the Foxp3 gene in CD4+ T cells, ChIP-qPCR assays for corresponding regions in the Foxp3 gene were performed. As shown in Fig. 7C, p53 bound to the Foxp3 promoter in CD4+ T cells, consistent with a previous finding in tumor cells (15). Importantly, strong association between p53 and the CNS-2 was detected in CD4+ T cells by ChIP-qPCR assays (Fig. 7C). Taken together, these results indicate that p53 directly induces Foxp3 expression by binding to the promoter and the CNS-2 of the Foxp3 gene in TCR-stimulated CD4+ T cells.
p53 enhances the transcription of the Foxp3 gene. (A) Jurkat cells were transfected with pFoxp3pro1 (promoter-1) or control pGL4 in the presence or absence of pcDNA3 p53 or pcDNA3 (empty), along with a Renilla luciferase vector pRL-TK. Cells were rested for 12 h and then stimulated with or without PMA/I for 6 h. Data are means ± SD of fold induction relative to pGL4- and pcDNA3 (empty)-transfected cells without PMA/I stimulation (n = 3). *p < 0.05. (B) Jurkat cells were transfected with pFoxp3pro2 (promoter-2), pFoxp3pro2/CNS-1 (CNS-1), pFoxp3pro2/CNS-2 (CNS-2), or control pGL3 in the presence or absence of pcDNA3 p53 or pcDNA3 (empty), along with pRL-TK. Cells were rested for 12 h and then stimulated with or without PMA/I for 6 h. Data are means ± SD of fold induction relative to pGL3- and pcDNA3 (empty)-transfected cells without PMA/I stimulation (n = 3). **p < 0.005. (C) Naive CD4+ T cells isolated from C57BL/6 mice (8 wk old) were stimulated with TCR in Th0 conditions for 12 h. ChIP-qPCR assays for promoter, CNS-1, and CNS-2 of the Foxp3 gene were performed with anti-p53 Ab or control rabbit IgG. Results are expressed as the percentage input for each ChIP fraction. Data are representative of three independent experiments.
Discussion
In this study, we show that p53 expressed in T cells plays an indispensable role in the induction of Treg differentiation and the prevention of autoimmunity. We found that aged p53-cKO mice spontaneously developed inflammatory lesions in various organs (Fig. 1B, Table I). Additionally, ANA and autoantibodies against gastric parietal cells were detected in sera in p53-cKO mice but not in p53-WT mice (Fig. 1C, 1D). Not only the numbers of Foxp3+CD4+ Tregs in the spleen, lung, and colon (Fig. 3D, 3E) but also in vitro differentiation of Tregs (Fig. 4A, 4B) were significantly decreased in p53-cKO mice as compared with age-matched p53-WT mice. Regarding the mechanism underlying p53-mediated induction of Tregs, we found that TCR-mediated signaling but not TGF-β signaling induced p53 expression in CD4+ T cells (Fig. 6), and subsequently p53 enhanced the transcription of Foxp3 by directly binding to the promoter and the CNS-2 regions of the Foxp3 gene (Fig. 7). Taken together, these results suggest that p53 deficiency in T cells causes the breakdown of self-tolerance and the development of autoimmunity owing to impaired Treg differentiation.
We show that the tumor suppressor p53 expressed in T cells suppresses autoimmunity. It is well established that p53 plays a central role in tumor suppression by inducing apoptosis, cell cycle arrest, senescence, and DNA repair (1, 2). In addition to these conventional functions of p53, several studies using systemic p53-deficient mice have suggested that p53 regulates the development of autoimmunity (6–8, 10, 11). Regarding autoimmune diseases in humans, a recent clinical study has shown an association between the polymorphisms in p53 codon 72 and the susceptibility to systemic lupus erythematosus in an Asian population (38). In agreement with these findings, it has been shown that genetic deletion of p53 target genes such as p21 and Gadd45 is associated with lupus-like syndromes in mice (39–41). In this study, we found that aged p53-cKO mice, which lack p53 in T cells, spontaneously developed inflammatory lesions in multiple organs and autoantibody production (Fig. 1B, Table I). These findings indicate that p53 expressed in T cells is at least in part responsible for the suppression of autoimmunity.
We also show that p53 is involved in the differentiation of Tregs. It has been shown that Tregs play a central role in keeping the homeostasis of immune systems (12–14). Accumulating evidence has shown that Tregs are divided into at least two major subgroups: thymus-derived nTregs and extrathymically derived inducible Tregs (iTregs) (14). In this study, we found that the numbers of Foxp3+CD4+ T cells in the spleen, lung, and colon but not in the thymus were decreased in p53-cKO mice (Fig. 3D, 3E), suggesting that p53 is required for the development and/or maintenance of iTregs but not of nTregs. Alternatively, we found that the suppressive activity of Tregs on a per cell basis was normal in p53-cKO mice (Fig. 5C). Additionally, we found that Tregs in p53-cKO mice produced IL-10 at the levels similar to those in p53-WT mice (Fig. 5B, Supplemental Fig. 4). These data suggest that p53 is involved in the development and/or maintenance of iTregs but not in the suppressive function of iTregs on a per cell basis.
In agreement with the above assumption, we show that p53 directly activates the promoter and the CNS-2 regions of the Foxp3 gene (Fig. 7). Recently, in addition to the promoter, the importance of three Foxp3 CNS elements (CNS-1, CNS-2, and CNS-3) has been identified for Treg differentiation (37). Among them, mice lacking CNS-2 have been shown to exhibit reduced numbers of iTregs with normal numbers of nTregs (37), similar to p53-cKO mice (Fig. 3D). Foxp3 itself has been shown to directly bind to CNS-2 in a Runx1- and Cbf-β–dependent manner to maintain the active state of the Foxp3 locus in the progeny of dividing Tregs (37). It has also been reported that the binding of CREB/ATF and STAT5 to CNS-2 is essential for Foxp3 induction (23, 42). In this study, we demonstrated that p53 enhanced the transcriptional activity of the CNS-2 region of the Foxp3 gene (Fig. 7B), and that p53 bound to this region in TCR-stimulated CD4+ T cells (Fig. 7C). Additionally, we found putative p53 binding sites in murine Foxp3 promoter (−941 to −917 bp and −620 to −598 bp from transcription start site) as well as in CNS-2 (+4216 to +4237 bp and +4270 to +4290 bp from transcription start site) by in silico analysis. Taken together, it is suggested that in addition to the already known effect of p53 on the promoter, p53 enhances Foxp3 expression through the activation of CNS-2 presumably by acting cooperatively with transcription factors such as Foxp3, Runx1, Cbf-β, CREB/ATF, and STAT5.
Consistent with a previous report (11), we found that the number of IL-17A–producing CD4+ T cells was increased in the lung in p53-cKO mice (Fig. 2B), and that mRNA expression of RORγt (Fig. 4A) as well as the production of IL-17A (Fig. 4B) were significantly enhanced in CD4+ T cells under Th17 conditions in p53-cKO mice as compared with p53-WT mice. These findings raise the possibility that the enhanced Th17 cell differentiation is primarily involved in the development of autoimmune phenotypes in aged p53-cKO mice. In this regard, it has been demonstrated that Th17 cells promote autoimmune diseases by triggering a positive feedback loop of IL-6 production (43). Alternatively, we found that p53-cKO mice did not exhibit elevated levels of IL-6 but developed inflammatory diseases in various organs such as stomach, thyroid gland, submandibular gland, and kidney with the production of ANA and autoantibodies against gastric parietal cells (Fig. 1C, 1D). Moreover, the phenotypes of p53-cKO mice are similar to those in mice that are transferred with Treg-depleted CD4+ T cells (44). Therefore, we suppose that the autoimmune phenotypes of p53-cKO mice primarily result from the impaired differentiation of Tregs. Further investigation using mice lacking p53 expression specifically in Tregs (e.g., Foxp3-Cre p53fl/fl mice) will provide important information to argue against the possible involvement of p53 expressed in effector CD4+ T cells, including Th17 cells and/or CD8+ T cells, for the development of autoimmune diseases in p53-cKO mice.
Given the well-documented reciprocal relationship between Th17 cells and Tregs (29), it is possible that the enhanced Th17 cell differentiation results in the impaired iTreg generation in p53-cKO mice, and vice versa. In this regard, we found that even when anti–IL-6 Ab and IL-21R/Fc were added to the culture to block endogenously produced IL-6 and/or IL-21 that could inhibit Treg differentiation (29), the frequency of Foxp3+ CD4+ T cells was still significantly decreased in p53-cKO mice as compared with that in p53-WT mice (Fig. 4C). Additionally, we found that p53 directly activated the promoter and the CNS2 of Foxp3 gene (Fig. 7), whereas p53 did not affect the transcriptional activity of RORγt promoter (−2048 to +92 from transcription start site) (data not shown). Alternatively, Zhang et al. (11) have shown that the enhanced Th17 response in the absence of p53 results from an increased sensitivity to IL-6–induced STAT3 phosphorylation. Taken together, these findings suggest that the impaired Treg differentiation and the enhanced Th17 cell differentiation are, at least in part, regulated independently by the absence of p53 in CD4+ T cells.
It has recently been reported that Foxp3 CNS1-deficient mice, which exhibit impaired iTreg generation with normal nTreg development, spontaneously develop Th2 type inflammation in the lung and intestine (45). Alternatively, we found that p53-cKO mice, which also exhibit impaired iTreg generation with normal nTreg development, did not develop Th2 type inflammation (e.g., IL-13–producing CD4+ T cells [Fig. 2B] and eosinophil recruitment [data not shown]) in the lung. Additionally, we found that p53-cKO mice did not spontaneously develop inflammatory diseases in intestine (data not shown). These findings suggest that the impaired generation of iTregs by itself could not explain all the phenotypes of p53-cKO mice and that the accompanied enhancement of Th17 cells might be involved in the difference in the phenotypes between Foxp3 CNS1-deficient mice and p53-cKO mice.
In conclusion, we have shown that p53 deficiency in T cells results in the spontaneous development of autoimmunity presumably by the impaired differentiation of Tregs in the periphery. Our results highlight a novel function of the tumor suppressor p53 expressed in T cells in the maintenance of immune homeostasis.
Disclosures
The authors have no financial conflicts of interest.
Acknowledgments
We thank Dr. Warren Leonard (National Heart, Lung, and Blood Institute, National Institutes of Health) and Dr. Masahide Tone (University of Pennsylvania) for reporter constructs.
Footnotes
This work was supported in part by grants-in-aids for scientific research from the Ministry of Education, Culture, Sports, Science and Technology, the Japanese Government, and by the Global Centers of Excellence Program (Global Center for Education and Research in Immune System Regulation and Treatment) of the Ministry of Education, Culture, Sports, Science and Technology, Japan.
The online version of this article contains supplemental material.
Abbreviations used in this article:
- ANA
- anti-nuclear Ab
- ChIP
- chromatin immunoprecipitation
- cKO
- conditional knockout
- CNS
- conserved noncoding DNA sequence
- GITR
- glucocorticoid-induced TNFR
- IL-21R/Fc
- recombinant mouse IL-21R/Fc chimera
- iTreg
- induced regulatory T cell
- nTreg
- naturally occurring regulatory T cell
- PMA/I
- PMA plus ionomycin
- qPCR
- quantitative real-time PCR
- ROR
- retinoic acid–related orphan receptor
- WT
- wild-type.
- Received February 20, 2013.
- Accepted August 6, 2013.
- Copyright © 2013 by The American Association of Immunologists, Inc.

















