Research ArticleHORMONES

Estrogen receptor α contributes to T cell–mediated autoimmune inflammation by promoting T cell activation and proliferation

See allHide authors and affiliations

Science Signaling  17 Apr 2018:
Vol. 11, Issue 526, eaap9415
DOI: 10.1126/scisignal.aap9415

Estrogen receptor stimulates T cells

Sex-linked susceptibility has been described for nearly 75% of all autoimmune disorders. More than any other risk factor discovered, being female confers the greatest risk of developing these diseases. Mohammad et al. identify a direct role for the female sex hormone estrogen in the development of autoimmune T cell responses. Deletion of estrogen receptor α (ERα) in T cells reduced disease burden in a mouse model of colitis. ERα-expressing T cells were more activated after stimulation, proliferated more, and expressed more proinflammatory cytokines than T cells lacking this receptor. Conversely, ERα-deficient T cells were more readily skewed to a regulatory T cell phenotype. Together, these data identify a role for direct sex hormone–dependent activation of T cells in autoimmune responses.

Abstract

It has long been appreciated that most autoimmune disorders are characterized by increased prevalence in females, suggesting a potential role for sex hormones in the etiology of autoimmunity. To study how estrogen receptor α (ERα) contributes to autoimmune diseases, we generated mice in which ERα was deleted specifically in T lymphocytes. We found that ERα deletion in T cells reduced their pathogenic potential in a mouse model of colitis and correlated with transcriptomic changes that affected T cell activation. ERα deletion in T cells contributed to multiple aspects of T cell function, including reducing T cell activation and proliferation and increasing the expression of Foxp3, which encodes a critical transcription factor for the differentiation and function of regulatory T cells. Thus, these data demonstrate that ERα in T cells plays an important role in inflammation and suggest that ERα-targeted immunotherapies could be used to treat autoimmune disorders.

INTRODUCTION

More than 75% of autoimmune disorders are characterized by increased prevalence in females, and both sex hormones and genetics have been implicated in this sex bias (1). Estrogen, in addition to regulating reproductive functions, also acts on a wide range of other systems and tissues such as immune, central nervous, cardiovascular, and skeletal systems, as well as liver, kidney, and skin (2, 3). Estrogen functions primarily through binding to two intracellular receptors, estrogen receptor α (ERα, encoded by the gene Esr1) and ERβ, members of the nuclear receptor (NR) family. These two ERs are ubiquitously expressed in many cell types and tissues (4, 5). The NR family consists of a class of intracellular proteins responsible for sensing steroid and thyroid hormones, as well as vitamins and certain other molecules. NRs directly bind to DNA and regulate the expression of adjacent genes. However, NRs usually work together with co-regulators that enable precise regulation of the transcriptional machinery.

Basic immune responses differ between females and males. Females respond to infection and vaccination with increased antibody production and a T helper 2 (TH2) cell–dominant immune response, whereas males usually have TH1 cell–biased immune responses (6, 7). Moreover, the disease severity of both multiple sclerosis and rheumatoid arthritis is usually reduced during pregnancy, especially during the third trimester when estrogen and progesterone reach their highest concentrations (8, 9). Autoimmune sex bias has also been observed in inflammatory intestinal disorders and colon cancer, suggesting that estrogen may play a role in the modulation of intestinal health (1012). Notably, the development of inflammatory bowel disease (IBD) has been associated with oral contraceptive use (13). In a prospective cohort study of 117,375 U.S. women, oral contraceptive use was associated with an increased risk of Crohn’s disease (CD) (11). In another study, hormone replacement therapy in postmenopausal women was associated with an increased risk of ulcerative colitis (10). In contrast, another study indicated that hormone replacement therapy provided a protective effect in women with IBD (14). These studies suggest that estrogen plays a role in the pathogenesis of IBD. However, estrogen exerts both immunostimulatory and immunosuppressive functions that are cell type–specific and context-dependent (1517).

In response to antigen presentation by antigen-presenting cells and dependent on the cytokine microenvironment, naïve CD4+ T cells differentiate into distinct lineages with different cytokine production profiles, namely TH1, TH2, TH17, and induced regulatory T (iTreg) cells (18, 19). These TH subsets play essential roles in host defense against pathogens, immune system homeostasis, and tumor immunity. Dysfunction of these cells contributes to asthma, allergy, tumor, and autoimmune inflammation, including IBD. Especially crucial for maintaining self-tolerance and immune homeostasis, FOXP3+ Treg cells can be induced centrally in the thymus [natural Treg (nTreg) cells] or in the periphery by cytokines such as transforming growth factor–β (TGFβ) [inducible Treg (iTreg) cells] (20, 21). Genetic mutations of FOXP3 lead to the development of autoimmunity, including immune dysregulation, polyendocrinopathy, enteropathy, X-linked (IPEX) syndrome in humans, and the scurfy phenotype in mice. Previously, it was believed that TH1 cells, through the production of interferon-γ (IFN-γ), were crucial for the pathogenesis of IBD (22). However, since the discovery of TH17 cells, studies have demonstrated that the proinflammatory cytokine interleukin-23 (IL-23) contributes to IBD by inducing both IL-17A and IFN-γ production and inhibiting Treg cell proliferation (2326). Therefore, therapeutics targeting both TH1 and TH17 cell responses are already under development for treating IBD (27, 28).

The effect of estrogen has on TH17 cell differentiation remains controversial. For example, exogenous estrogen causes adult male cystic fibrosis mice to develop more severe bacterial pneumonia through enhanced TH17 cell–regulated inflammation and reduced antimicrobial peptide production (29). In contrast, another study showed that estrogen deficiency in postmenopausal women is associated with increased IL-17A abundance (30). Moreover, administration of the estrogen estradiol (E2) to mice activated T lymphocytes through ERα and protected against the development of experimental autoimmune encephalomyelitis by modulating TH1 and TH17 cell responses (31). Despite these foundational studies, the precise role of estrogen and its receptors in T cell differentiation is not well understood.

The goal of this study is to unveil the role of ERα in the development of T cell–mediated autoimmune inflammation using a T cell transfer model of colitis in mice. To dissect the effect of estrogen through its receptor ERα in TH cells, we generated tissue-specific CD4-creERαfl/fl mice, in which ERα is deleted specifically in T lymphocytes. We found that ERα ablation had marked effects on T cell activation, proliferation, survival, and Foxp3 expression, which substantially limited their inflammatory and colitogenic potential. Thus, our data demonstrate the importance of ERα in regulating T cell functions, which suggests that ERα may be a potential therapeutic target for autoimmune disorders.

RESULTS

ERα in T cells is required for the development of T cell–dependent colitis

ERα is ubiquitously expressed in many cell types, including T cells (32). To assess whether ERα plays any role in T cells in autoimmune disease, we crossed CD4-cre mice with ERαfl/fl mice (33) to generate CD4-creERαfl/fl mice in which ERα is specifically deleted within T cells. Deletion of Esr1 was validated by reverse transcription polymerase chain reaction (RT-PCR) using RNA from CD4+ and CD4 T cells from the spleen (fig. S1). CD4-creERαfl/fl mice had similar numbers of thymocytes, lymphocytes, and splenocytes as littermate control ERαfl/fl mice. CD4-creERαfl/fl mice had normal proportions of CD4, CD8, and CD3 cells in the thymus and spleen (fig. S2). Furthermore, CD4-creERαfl/fl mice also had a similar proportion of Foxp3+ cells within CD3+CD4+ thymocytes and within peripheral CD4+ cells compared with littermate control ERαfl/fl mice (fig. S2). These results suggest that ERα is not essential for T cell development.

To clarify how ERα in T cells contributes to colitis, we transferred sorted naïve (CD4+CD25CD45RBhi) T cells from CD4-creERαfl/fl or littermate control ERαfl/fl mice to Rag1−/− recipients and monitored colitis development. Female mice were specifically used for this study because of their increased estrogen levels relative to male mice. We observed that 4 weeks after transfer, Rag1−/− mice reconstituted with control T cells stopped gaining weight and even started to lose weight by week 5, whereas mice reconstituted with CD4-creERαfl/fl T cells continued to gain weight (Fig. 1A). We measured inflammatory cytokines in peripheral blood samples from these mice at the time of sacrifice (about 9 weeks). Consistently, mice that received CD4-creERαfl/fl T cells had significantly decreased circulating concentrations of the inflammatory cytokines TNFα (tumor necrosis factor–α), IL-6, and IFN-γ compared to mice that received control T cells (Fig. 1B). Histological evaluation of Rag1−/− mice that received control T cells showed the development of diffuse inflammation in the mucosa that extended multifocally to the submucosa (Fig. 1C). The lamina propria was also diffusely infiltrated with leukocytes that displaced colonic glands (fig. S3). The surface epithelium also displayed alterations, including cuboidal and pseudostratified, instead of tall columnar and single-layered, epithelium and erosion of the epithelium (fig. S3). In contrast, in mice that received CD4-creERαfl/fl T cells, the colons appeared essentially normal histologically (Fig. 1C and fig. S3). Clinical scoring of colon sections indicated that mice that received CD4-creERαfl/fl T cells showed reduced distribution of inflammation, degree of inflammation, and extent of erosion/ulceration. Their cumulative histopathologic scores were significantly less than those of mice that received control ERαfl/fl T cells (Fig. 1D). These results demonstrate that CD4-creERαfl/fl T cells were less colitogenic than control T cells, suggesting that ERα in T cells is intrinsically required for colitis development.

Fig. 1 ERα deletion within T cells limits their pathogenic potential in a murine colitis model.

(A to D) Rag1−/− mice received 400,000 sorted naïve CD4+CD45RBhiCD25 T cells isolated from ERαfl/fl (filled squares) or CD4-creERαfl/fl (open circles) mice. (A) Mice were weighed weekly to monitor the onset of colitis. Data are pooled from 22 ERαfl/fl mice and 23 CD4-creERαfl/fl mice from five independent experiments. (B) Serum tumor necrosis factor (TNF), interleukin-6 (IL-6), interferon-γ (IFN-γ), IL-12p70, monocyte chemoattractant protein-1 (MCP-1), and IL-10 were measured using a murine inflammation cytokine bead array at the time of sacrifice. Data are pooled from 15 ERαfl/fl mice and 18 CD4-creERαfl/fl mice from four independent experiments. (C) Colon sections from mice that received the indicated T cells were subjected to hematoxylin and eosin staining and clinical scoring. Data are histological images representative of 15 ERαfl/fl mice (filled squares) and 18 CD4-creERαfl/fl mice (open circles) from four independent experiments. In mice receiving ERαfl/fl T cells, there is diffuse inflammation in the mucosa, and the inflammation extends multifocally to the submucosa (arrows). The lamina propria (bar) contains colonic glands, and within each gland, there are numerous goblet cells (arrows). Original magnification, ×40; scale bar, 200 μm. (D) Development of colitis was assessed by monitoring the distribution of inflammation, degree of inflammation, and extent of erosion/ulceration. Cumulative histopathologic scores are also shown. *P < 0.05 and **P < 0.01 by two-tailed Student’s t test (A and B) or two-tailed Mann-Whitney test (C).

Adoptive transfer of naïve T cells from ERαfl/fl mice also led to the development of dermatitis and a psoriasis-like disease, which was quantified by scoring of primary histopathologic features, including dermatitis, folliculitis, and furunculosis, as well as secondary histopathologic features, including epidermal hyperplasia with hyperkeratosis and dermal fibrosis (fig. S4). In addition, mice that received CD4-creERαfl/fl T cells also developed significantly less severe dermatitis compared to those that received control ERαfl/fl T cells (fig. S4). Together, these data demonstrate that ERα in T cells contributes to the development of T cell–mediated inflammation.

ERα deletion within effector T cells limits T cell accumulation and TH17 and TH1 cell responses in a transfer model of colitis

Because Rag1−/− mice are lymphopenic (lack lymphocytes), the transferred naïve T cells undergo massive proliferation and induce inflammation (22). We therefore questioned whether the lack of inflammation in Rag1−/− recipients of CD4-creERαfl/fl T cells was due to a defect in T cell proliferation. To this end, we quantitated the accumulation of T cells in the recipient mice. As expected, transfer of control ERαfl/fl T cells resulted in enlarged spleens and greater accumulation of CD4+ T cells in spleens and mesenteric lymph nodes (MLNs) compared to transfer of CD4-creERαfl/fl T cells (Fig. 2, A to C). Because both TH1 and TH17 cells play important roles in the pathogenesis of IBD (22, 26), we examined whether recipient Rag1−/− mice had impaired TH1 or TH17 cell differentiation. We performed intracellular cytokine staining to detect the proportion of IFN-γ– and IL-17A–producing CD4+ T cells in spleens and MLNs from recipient Rag1−/− mice. Transfer of CD4-creERαfl/fl T cells resulted in significantly reduced numbers of IFN-γ–positive T cells in both the spleen and MLNs and reduced numbers of IL-17A–positive T cells in MLNs (Fig. 2, D to F). These results indicate that ERα is required for the accumulation of inflammatory TH1 and TH17 cells in vivo in a lymphopenic environment.

Fig. 2 ERα deletion within T cells limits their accumulation and inflammatory cytokine production in a T cell transfer model of colitis.

(A to F) Rag1−/− mice received 400,000 sorted naïve CD4+CD45RBhiCD25 T cells isolated from ERαfl/fl mice or CD4-creERαfl/fl mice. (A) Spleens were harvested at about 9 weeks after T cell transfer and weighed. Data are means ± SEM from 11 ERαfl/fl mice and 13 CD4-creERαfl/fl mice from three independent experiments. Percentage (B) and total cell numbers (C) of spleen and MLN CD4 T cells (TCRβ+CD4+) were quantitated. Data are means ± SEM from 19 ERαfl/fl mice and 23 CD4-creERαfl/fl mice from five independent experiments. (D to F) Spleens and MLNs were harvested, and the proportions of IFN-γ+ and IL-17A+ effectors within the gated TCRβ+CD4+ population were analyzed. Representative flow cytometry plots (D) gated CD4+TCRβ+ T cells from the spleens of colitic mice from five independent experiments. Quantified data (E and F) are means ± SEM of 19 ERαfl/fl mice and 23 CD4-creERαfl/fl mice. *P < 0.05 and **P < 0.01 by two-tailed Student’s t test. SSC, side scatter.

Transcriptomics analysis reveals defective activation of CD4-creERαfl/fl–deficient T cells

To gain further insight into the possible molecular mechanism by which ERα in T cells contributes to colitis development, we isolated CD4+ T cells from the spleens of recipient Rag1−/− mice 9 weeks after T cell transfer. We extracted RNA from these samples and performed transcriptomics analysis using RNA sequencing (RNA-seq). By comparing transcriptomes of transferred CD4+ T cells, we observed that 788 genes were increased in expression, and 577 genes were decreased in expression in CD4-creERαfl/fl T cells relative to control ERαfl/fl cells (Fig. 3A and table S1). We further performed pathway enrichment analysis of differentially expressed genes and found that T cell receptor (TCR) signaling was among the most enriched pathways (Fig. 3B). Notable genes in this pathway included Cd3e, Cd3g, Cd247, Cd4, Lat, Zap70, Nfatc2, and Cd69 (Fig. 3C). Compared to transferred control CD4+ T cells, expression of all these genes was reduced in CD4-creERαfl/fl T cells, suggesting that ERα may have a role in T cell activation. RNA-seq data also demonstrated reduced expression of Ifng in CD4-creERαfl/fl T cells, which was consistent with the reduced serum IFN-γ concentrations and reduced IFN-γ–producing T cells in the spleen and lymph nodes (Figs. 1B and 2, D and E).

Fig. 3 Transcriptomics analysis reveals reduced expression of genes involved in T cell activation in CD4-creERαfl/fl T cells.

(A to C) RNA sequencing (RNA-seq) analysis of transcripts detected in ERαfl/fl and CD4-creERαfl/fl CD4+CD3+ T cells isolated from the spleens of colitic Rag1−/− recipient mice. Data are representative of two biological replicates. (A) Scatter plot of differentially expressed genes where orange triangles represent the genes that were increased in expression and blue diamonds represent genes that were decreased in expression in CD4+ T cells from mice receiving CD4-creERαfl/fl T cells, as compared to ERαfl/fl T cells. (B) KEGG (Kyoto Encyclopedia of Genes and Genomes) pathway gene enrichment of differentially expressed genes from ERαfl/fl and CD4-creERαfl/fl cells that were isolated from spleens of Rag1−/− recipient mice. Rich factor is measured as the ratio of number of differentially expressed genes annotated in this pathway term to number of all genes annotated in this pathway term. (C) Heatmap displaying the log fold change of differentially expressed genes from CD4-creERαfl/fl versus ERαfl/fl CD4+ T cells.

ERα deletion reduces T cell activation

We next investigated whether ERα contributes to T cell activation in vitro. We activated control or CD4-creERαfl/fl naïve CD4+ T cells with various concentrations of anti-CD3 and anti-CD28 for 24 hours and analyzed cells for the cell surface expression of CD69 and CD25, two widely used markers for T cell activation. When CD4+ T cells were activated with high-dose anti-CD3 alone (5 μg/ml), CD4-creERαfl/fl T cells showed significantly reduced induction of CD69 (Fig. 4, A and B). However, no differences were noted at lower doses. In addition, increasing concentrations of anti-CD28 increased the proportion of CD69+ cells in control T cells, but not in CD4-creERαfl/fl T cells (Fig. 4, A and B). However, a high anti-CD3 concentration (5 μg/ml), in combination with increasing concentrations of anti-CD28, largely compensated for the T cell activation defect in CD4-creERαfl/fl T cells (fig. S5). Although we also observed a consistent trend for reduced expression of CD25, the trend was not statistically significant (Fig. 4B). These data suggest that in vitro cultured CD4-creERαfl/fl T cells have a clear defect in TCR-stimulated T cell activation, which can be compensated by strong TCR and costimulatory signals. Furthermore, they are consistent with the RNA-seq data showing reduced CD69 mRNA expression in ex vivo isolated CD4-creERαfl/fl T cells adoptively transferred into Rag1−/− mice (Fig. 3C).

Fig. 4 ERα regulates T cell activation.

(A) CD69 expression was assessed by flow cytometry on naïve T cells activated for 1 day with the indicated concentrations of anti-CD3 alone (left), anti-CD3, and the indicated concentrations of anti-CD28 (middle and right). Data are means ± SEM of three biological replicates. (B) Flow cytometry plots of CD69 and CD25 expression on ERαfl/fl and CD4-creERαfl/fl T cells after 1 day of culture with the indicated concentrations of anti-CD3 and anti-CD28. Data are representative of three independent experiments. (C) Western blot analysis of the indicated proteins in ERαfl/fl and CD4-creERαfl/fl T cells that were activated for 3 days. Left: Blots are representative of three experiments. Right: Normalized band intensity data are means ± SEM. (D) Western blot analysis of the indicated proteins in CD4-creERαfl/fl T cells transduced with retrovirus-expressing control vector (pRV) or ERα-expressing plasmid (pRV-Esr1) that were activated for 3 days. Left: Blots are representative of three independent experiments. Right: Normalized band intensity data are means ± SEM. *P < 0.1 and **P < 0.05 by paired Student’s t test.

To validate the gene expression changes identified by RNA-seq, we isolated naïve CD4-creERαfl/fl and control CD4+ T cells, activated them in vitro for 3 days with anti-CD3 and anti-CD28, and analyzed expression of target proteins by Western blot. In response to CD3 and anti-CD28 activation in vitro, the amounts of NFAT1 (nuclear factor of activated T cells 1; product of the Nfatc2 gene), Zap70, and STAT5 (signal transducer and activator of transcription 5) protein were reduced in CD4-creERαfl/fl T cells compared to control T cells (Fig. 4C). Conversely, retroviral expression of Esr1 in CD4-creERαfl/fl CD4+ T cells resulted in increased amounts of STAT5b and NFAT1 compared to that in CD4-creERαfl/fl CD4+ T cells transduced with a retroviral vector control (Fig. 4D). Collectively, these data suggest that ERα in T cells influences the expression of many genes known to be important for TCR signaling and subsequent events.

ERα deletion in T cells influences T cell survival

Because the RNA-seq data identified programmed cell death 4 (Pdcd4), a gene related to apoptosis, as being reduced in expression in the absence of ERα (Fig. 5A), we next investigated whether increased cell death could account for the reduced inflammatory cytokine production and impaired TH1 and TH17 cell differentiation by CD4-creERαfl/fl cells. To determine whether ERα plays any role in T cell survival, CD4-creERαfl/fl and control CD4+ T cells were activated with a combination of anti-CD3 and anti-CD28 for 3 days, followed by annexin V and propidium iodide (PI) staining. After T cell activation, fewer PI- and/or annexin V–positive cells were detected from CD4-creERαfl/fl T cells (Fig. 5B), indicating that, compared to ERαfl/fl T cells, CD4-creERαfl/fl cells underwent less activation-induced T cell death. B cell lymphoma 2 (Bcl2) is decreased by T cell activation and is a prosurvival protein that maintains lymphocyte survival (34, 35). Despite the fact that, using transcriptomics, we found no change in Bcl2 mRNA expression, and the proportion of Bcl2-expressing T cells was significantly increased in CD4-creERαfl/fl T cells in response to T cell activation, as determined by intracellular staining (Fig. 5C). In addition, we detected mitochondrial membrane potential with tetramethylrhodamine, ethyl ester (TMRE) to label active mitochondria in live cells. Consistent with the changes in annexin V and PI staining and Bcl2 expression, a greater proportion of TMREhi cells was detected in CD4-creERαfl/fl T cells after 40 min in response to T cell activation, indicating that these cells had more active mitochondria. Collectively, our data suggest that ERα enhances T cell activation–induced apoptosis.

Fig. 5 ERα influences T cell survival.

(A) RNA-seq of Pdcd4 transcript in ERαfl/fl and CD4-creERαfl/fl splenic CD4+ T cells from colitic Rag1−/− recipient mice. Data are means ± SEM of fragments per kilobase of transcript per million mapped reads (FPKM) values from two biological replicates. (B and C) ERαfl/fl and CD4-creERαfl/fl CD4 T cells were activated for 3 days. (B) Cellular viability was assessed by annexin V and propidium iodide (PI) staining. Top: Flow cytometry plots are representative of three independent experiments. Bottom: Quantified data are means ± SEM. (C) Bcl2 expression was assessed by intracellular staining. Left: Flow cytometry plots are representative of three independent experiments. Right: Quantified data are means ± SEM. (D) Mitochondrial membrane potential was determined by staining ERαfl/fl and CD4-creERαfl/fl CD4 T cells that were activated for 40 min or 1 day with TMRE. Top: Flow cytometry plots are representative of three independent experiments. Bottom: Quantified data are means ± SEM. **P < 0.05 by paired Student’s t test.

ERα deletion within T cells limits their proliferative potential

In addition to reduced expression of TCR signaling molecules, we also observed reduced expression of Cd28 and Ctla4 by RNA-seq from CD4-creERαfl/fl CD4+ T cells harvested ex vivo from colitis mice (fig. S6). TCR and costimulatory signaling molecules such as CD28 and cytotoxic T lymphocyte–associated protein 4 (CTLA4) are critical for T cell proliferation in immunodeficient hosts (3639). We therefore examined whether ERα deficiency influenced T cell proliferation. We found earlier by RNA-seq that several genes involved in regulating the cell cycle, such as Cdk1 and Ccnb1, were reduced in expression in CD4-creERαfl/fl T cells (fig. S6). CD4-creERαfl/fl CD4+ T cells exhibited reduced cyclin-dependent kinase 1 (CDK1) abundance as compared to control cells activated in vitro (Fig. 4C). Conversely, retroviral overexpression of ERα in CD4-creERαfl/fl cells increased the abundance of CDK1 (Fig. 4D). Next, we evaluated whether ERα contributes to T cell proliferation. CD4-creERαfl/fl or control ERαfl/fl naïve CD4+ T cells were labeled with CellTrace Violet dye and activated with immobilized anti-CD3 (1 μg/ml) and anti-CD28 (1 μg/ml). After 4 days, CD4-creERαfl/fl T cells exhibited less proliferation as compared to ERα control T cells, as detected by fluorescent dye dilution (Fig. 6A). The reduced proliferation of CD4-creERαfl/fl T cells was also observed after 5 days in vitro (Fig. 6A). However, when we labeled CD4-creERαfl/fl naïve CD4+ T cells and control cells separately with two different dyes, mixed, and cultured them together at a 1:1 ratio, the CD4-creERαfl/fl CD4+ T cells proliferated similarly to control cells (Fig. 6B), suggesting that either soluble or contact-dependent factors produced by the control T cells could compensate for the proliferative defect in CD4-creERαfl/fl CD4+ T cells.

Fig. 6 ERα contributes to T cell proliferation and is required in lymphopenia-induced expansion of CD4+ T Cells.

(A) In vitro proliferation of CellTrace Violet–labeled ERαfl/fl and CD4-creERαfl/fl T cells stimulated with anti-CD3 (1 μg/ml) and anti-CD28 (1 μg/ml) for 4 and 5 days. Right: Representative flow cytometry plots from three independent experiments. Left: Quantified data are means ± SEM. (B) Naïve CD4+ T cells were isolated from CD4-creERαfl/fl and littermate control mice. Cells were labeled with CellTrace Violet or CellTrace Yellow dye, mixed at a 1:1 ratio, and cultured together under iTreg conditions. Dye dilution was detected on day 4. Data are representative flow cytometry plots of one mouse per group from three independent experiments. (C) In vivo proliferation of carboxyfluorescein succinimidyl ester (CFSE)–labeled wild-type (WT) CD45.1+ mice and CD45.2+ CD4-creERαfl/fl CD4+ T cells transferred into Rag1−/− recipient mice. An aliquot of input cells was retained to confirm the starting ratio (input). Spleens and MLNs of recipients were harvested for analysis on day 10 or 14 after injection. Right: Flow cytometry plots of CFSE dilution for WT (CD45.1+) and CD4-creERαfl/fl (CD45.2+) T cells within the same host on day 10 are representative of three independent experiments. Left: Ratio of WT and CD4-creERαfl/fl CD4+ T cells was quantified by staining for CD45.1 and CD45.2. Representative data from one experiment are means ± SEM of three mice per group on day 10 and five mice per group on day 14. *P < 0.1 by paired Student’s t test.

To further investigate whether the defect in T cell activation and proliferation observed in CD4-creERαfl/fl CD4+ T cells causes metabolic changes in response to TCR activation, CD4-creERαfl/fl and control T cells were activated with anti-CD3 (1 μg/ml) and anti-CD28 (1 μg/ml) for 2 days in vitro. The Seahorse XF Cell Mito Stress Test was applied to detect key parameters of mitochondrial function by directly measuring the oxygen consumption rate (OCR) of the cells. In comparison to that of control T cells, the spare respiratory capacity of the CD4-creERαfl/fl T cells was almost 50% reduced (fig. S7), suggesting that the CD4-creERαfl/fl T cells have impaired metabolism. This is consistent with the observed defects in activation and proliferation of CD4-creERαfl/fl T cells.

Last, to determine whether ERα deletion in T cells influences T cell proliferation in vivo in a lymphopenic environment, we isolated naïve CD4+ T cells from CD45.1+ C57BL6/J congenic mice and CD45.2+ CD4-creERαfl/fl mice. These cells were mixed at a 1:1 ratio before labeling with carboxyfluorescein succinimidyl ester (CFSE). The mixture of labeled cells was injected into Rag1−/− recipient mice. Ten to 14 days later, we analyzed CFSE dilution in adoptively transferred wild-type (WT) (CD45.1+) and CD4-creERαfl/fl (CD45.2+) T cells within the same host. WT T cells isolated from both the spleen and MLNs proliferated more than CD4-creERαfl/fl T cells, as indicated by the proportion of recovered cells (Fig. 6C) and CFSE dilution (Fig. 6C). Together, these data show that ERα deficiency within CD4+ T cells reduced T cell proliferation both in vitro and in vivo, demonstrating that ERα is required for TH cell proliferation and T cell expansion.

ERα modulates iTreg cell differentiation and function

Previous studies have demonstrated that weak TCR activation favors iTreg differentiation by regulating the phosphatidylinositol 3-kinase (PI3K)–Akt–mammalian target of rapamycin (mTOR) pathway that suppresses Foxp3 expression (40). In addition, by modulating Foxp3 locus methylation, DNA methyltransferase 1 (DNMT1) suppresses Foxp3 expression (41, 42). We investigated whether the impaired TH1 and TH17 cell response and reduced accumulation of CD4+ T cells observed in Rag1−/− mice receiving CD4-creERαfl/fl T cells were due to increased iTreg differentiation. To test this, we first cultured CD4-creERαfl/fl and control T cells isolated from female mice in vitro under TH0 (anti-CD3 and anti-CD28 only) and iTreg-polarizing (anti-CD3 and anti-CD28 with TGFβ and IL-2) conditions for 3 days and measured Foxp3 expression by intracellular staining. Under iTreg conditions, the addition of E2 enhanced the frequency of Foxp3+ cells in control T cells (Fig. 7A). In addition, in CD4-creERαfl/fl T cells, there was an increased proportion of Foxp3+ cells compared to control cultures; however, the addition of E2 did not further increase the percentage of Foxp3+ cells (Fig. 7A). These results suggest that in the absence of estrogen, ERα inhibits Foxp3 expression and upon the addition of the ligand, estrogen releases the inhibitory effect in an ERα-dependent manner.

Fig. 7 ERα modulates iTreg cell differentiation and function.

(A) FoxP3 expression was assessed by flow cytometry of CD4-creERαfl/fl and control ERαfl/fl naïve CD4+ T cells from female mice that were activated with anti-CD3 and anti-CD28 alone (TH0) or under iTreg-inducing conditions for 3 days with or without exogenous 17β-estradiol (E2; 10 nM). Left: Flow cytometry plots are representative of three independent experiments. Right: Quantified data are means ± SEM. (B) Western blot analysis of the indicated proteins in ERαfl/fl and CD4-creERαfl/fl T cells activated for 3 days. Left: Blots are representative of three experiments. Right: Normalized band intensity data are means ± SEM. (C) Western blot analysis of CD4-creERαfl/fl CD4 T cells transduced with either control vector (pRV) or ERα-expressing vector (pRV-Esr1) and cultured under iTreg-polarizing conditions. Left: Blots are representative of three experiments. Right: Normalized band intensity data are means ± SEM. (D) RNA-seq analysis of the indicated transcripts in ERαfl/fl and CD4-creERαfl/fl splenic CD4+ T cells from colitic Rag1−/− recipient mice. FPKM data are means ± SEM from two biological replicates. (E) Intracellular Foxp3 staining of ERαfl/fl and CD4-creERαfl/fl CD4+ T cells that were harvested from the lamina propria (LP) of colitic Rag1−/− recipient mice. Right: Flow cytometry plots are representative of two independent experiments. Left: Quantified data are means ± SEM of one mouse per group. *P < 0.1 and **P < 0.05 by paired Student’s t test.

We next characterized which signaling pathways contribute to the increased expression of Foxp3 in CD4-creERαfl/fl T cells. In response to T cell stimulation, in addition to the reduced abundance of NFAT1, STAT5, and CDK1 in CD4-creERαfl/fl T cells (Fig. 4C), we also detected reduced abundance of mTOR, phosphorylated AKT (pAKT), and DNMT1 after 3 days (Fig. 7B). Retroviral overexpression of ERα in CD4-creERαfl/fl T cells, followed by in vitro culture under iTreg conditions, led to increased amounts of pAKT and DNMT1, whereas Foxp3 abundance was reduced (Fig. 7C), indicating that ERα may suppress Foxp3 expression.

Furthermore, comparing gene expression changes in CD4+ T cells adoptively transferred into Rag1−/− recipient mice, CD4-creERαfl/fl T cells expressed twofold more Foxp3 than control T cells (Fig. 7D). Treg cells express more Irf8 and Tgfb than other CD4+ T cell subsets (20, 43) and that Foxp3 suppresses expression of Satb1 and Tcf7 (44, 45). In agreement with the enhanced Foxp3 expression in CD4-creERαfl/fl T cells, our RNA-seq data revealed increased expression of Tgfb and Irf8 but reduced expression of Satb1 and Tcf7 (Fig. 7D). Intracellular staining of cells from colitic Rag1−/− mice showed an increased frequency of Foxp3+CD4+ T cells from the spleens and lamina propria of CD4-creERαfl/fl compared to control T cells (Fig. 7E). Together, these results demonstrate that ERα suppresses iTreg cell differentiation and suggest that the ERα deficiency biases T cells toward an iTreg phenotype when adoptively transferred to lymphopenic Rag1−/− mice.

DISCUSSION

Upon antigen activation in the periphery, naïve CD4 T cells undergo functional differentiation toward different subsets such as TH1, TH2, TH17, and iTreg cells with distinct cytokine production profiles and functions. In the T cell transfer model of colitis, disease development is dependent on differentiated TH1 and TH17 cells. Here, we found that ERα deficiency in T cells is protective against colitis development because of reduced TH1 and TH17 cell responses and increased iTreg differentiation. To further investigate how ERα influences TH1 and TH17 cell differentiation in vivo, we showed that the reduced production of TH1 and TH17 cytokines, IFN-γ and IL-17A, is the consequence of impaired T cell activation and proliferation. An activation defect in CD4-creERαfl/fl T cells was confirmed by impaired expression of markers of T cell activation in vitro and alterations in T cell transcriptomics using RNA-seq. Notably, the defect in T cell activation in CD4-creERαfl/fl T cells in response to TCR and/or costimulatory signaling was compensated for by a high concentration of anti-CD3 and anti-CD28.

Previous studies have indicated that TCR signaling strength plays an essential role in TH subset differentiation, such that intermediate antigen doses favor TH1 cell–dominated responses, whereas reduced or increased doses favor TH2 cell responses (46, 47). For Treg cell differentiation and proliferation, TCR signaling can activate Akt through PI3K. One of the downstream effects of Akt activation is to activate mTOR, which inhibits expression of Foxp3 (48). Here, because of an intrinsic activation defect in CD4-creERαfl/fl T cells, suboptimal T cell activation favored iTreg cell differentiation in vitro and in vivo that may contribute to the suppression of effector T cell accumulation and colitis development. This is in agreement with previous studies showing that low peptide doses presented by dendritic cells induce weak TCR signaling that inhibits the Akt-mTOR pathway, which consequently enhances the proliferation of Foxp3+ Treg cells. In contrast, high peptide doses induce strong Akt-mTOR signaling, suppress the expression of Foxp3, and therefore favor the proliferation of effector TH cells (49). Our data suggest that ERα may inhibit Foxp3 expression indirectly through its effects on DNMT1. Therefore, our data suggest that ERα may promote Foxp3 expression through two mechanisms: One is through the activation of the TCR-Akt-mTOR pathway, and the other is via the induction of DNMT1 expression, which may modulate the chromatin status at the Foxp3 locus.

Last, we observed that ERα deficiency in T cells led to reduced T cell proliferation. We also found that CDK1 abundance was reduced in the absence of ERα in T cells. In most tissues, CDK1 is required to regulate cell cycle progression (50). Notably, CDK substrate motifs have been identified within the N-terminal repressor domain of murine Foxp3. CDK2-deficient Treg cells have increased suppressive activity in vitro and ameliorate disease in a mouse model of colitis (51). CDK2 can also phosphorylate Foxp3 and regulates its stability (52). Furthermore, a CDK inhibitor, Kenpaullone, enhances TGFβ-induced iTreg differentiation (53, 54). These studies demonstrate that CDK2 deficiency or inhibiting CDK activity leads to reduced cell proliferation and enhanced iTreg induction. However, it needs to be further investigated whether and how ERα, through regulating CDK1, contributes to T cell proliferation and iTreg differentiation.

Previous in vitro studies have suggested that estrogen induces Treg cell differentiation and expansion and increases the expression of Foxp3, programmed cell death 1 (PD-1), and CTLA4 in an ERα-dependent manner (5558). In our in vitro cultures, we also observed that E2 induced Foxp3 expression through ERα. When ERα was deleted, E2 failed to further increase Foxp3 expression (Fig. 7A). In CD4-creERαfl/fl T cells, Foxp3 expression was already enhanced without exogenous E2 (Fig. 7A). Estrogen functions mostly through its receptors, ERα and ERβ. However, ERs can also be activated in the absence of a ligand (59). ERα can interact with many other transcription factors that are involved in chromatin remodeling, histone modifications, and transcription regulation. Therefore, the response to ERα with or without ligand may depend on different mechanisms. Sex hormones contribute to sex bias in autoimmune diseases. In addition to estrogen, testosterone, androgen, and its receptor may also play a role. A study using lupus-prone Esr1−/− mice reports that ERα deficiency does not protect mice from lupus, suggesting that another hormone, such as testosterone, may play a role (60). Our study suggests that T cell–expressed ERα contributes to sex-biased autoimmune inflammation through its effects on T cell activation, proliferation, and Foxp3 expression.

Here, we found that ERα deficiency in T cells within a lymphopenic environment led to increased expression of Foxp3 and several other genes preferentially expressed in Treg cells, such as Lgals3, Tgfb1, Eno3, and Irf8, and reduced expression of Foxp3-suppressed genes, such as Tcf7 and Satb1. As a predominantly T cell–expressed transcription factor, Tcf7 is a member of the T cell factor/lymphoid enhancer-binding factor family of high-mobility group box transcriptional activators. In mice, deletion of Tcf7 results in an increased proportion of Foxp3+ Treg cells in CD4+CD8+ cells in the thymus, whereas reduced expression of Tcf7 leads to increased Treg-generating capacity. Furthermore, expression of Tcf7 is repressed by Foxp3 (45). Tcf7 binds to the T lymphocyte–specific enhancer motif (5′-WWCAAAG-3′) in the promoter of the Cd3e gene and plays an important role in the formation of the TCR-CD3 complex (61). Here, we observed that in peripheral T cells, Tcf7 and CD3e expression was reduced in CD4-creERαfl/fl T cells, and Foxp3 expression was increased. Although we found that ERα is involved in T cell activation both from in vitro culture and from in vivo T cell proliferation in a lymphopenic environment, surprisingly, we did not observe severe defects on T cell development in the thymus or periphery. ERα-mediated gene expression can be achieved by its interaction with other DNA binding proteins, such as activator protein-1 (AP-1), which includes the FOS and JUN family members (62). In breast cancer cells, knockdown of c-Fos attenuates the expression of 37% of all estrogen-regulated genes, suggesting that c-Fos is a fundamental partner for ERα-mediated transcription (63). The cytokine IL-2 has an essential role in T cell proliferation, and the expression of IL-2 is induced by T cell activation (64, 65). The transcription factors AP-1 and NFAT1 are required for Il2 expression in T cells (64, 65). We found that CD4-creERαfl/fl T cells showed reduced T cell activation, proliferation, and abundance of NFAT1. The proliferative defect in CD4-creERαfl/fl T cells was abrogated when the cells were cocultured with control T cells, suggesting that IL-2 may affect the proliferation of CD4-creERαfl/fl T cells. However, we cannot exclude the possibility that other secretary or contact-dependent factors may contribute to this effect. Because of the low abundance of ERα in T cells and the lack of a high-quality antibody for chromatin immunoprecipitation (ChIP) in murine T cells, we were unable to perform ChIP sequencing to globally map ERα-binding sites. Therefore, it is still unclear how ERα regulates expression of genes involved in T cell activation and Treg cell differentiation and function. It remains to be further clarified whether ERα exerts its effects through directly binding to these gene promoters or through interacting with other transcription factors, such as AP-1/c-Jun, c-Fos, ATF-2, Sp1, and Sp3, to mediate estrogen response element–independent signaling.

Given the important role of ERα in the pathogenesis of breast cancer, its effects on cell cycle control and apoptosis have been extensively studied, but the effects of ER seem to be cell type–dependent. In mammary glands and ER+ breast cancer cells, E2 promotes cell proliferation (66). One of the mechanisms is through increasing the expression of c-myc, which activates CDK to stimulate cell growth (67). E2 also has an antiapoptotic effect through regulating the expression of several apoptotic proteins, including BCL2, in breast cancer cells (68). However, in contrast with its ability to stimulate growth and inhibit apoptosis, in some studies, physiologic E2 induces apoptosis in other cell types, such as osteoclasts (69), neuronal cells (70), and thymocytes (71). Here, we showed that CD4-creERαfl/fl T cells expressed less CDK1 and more Bcl2 and proliferated less than control T cells but exhibited less apoptosis. These effects were rescued by reconstitution with ERα, confirming that expression of these genes in T cells depends on ERα.

In conclusion, our data suggest that ERα in T cells influences multiple aspects of T cell function. Through its effects on T cell activation, proliferation, survival, and TH subset differentiation, ERα contributes to T cell–mediated inflammation, as demonstrated in a colitis model. Our data demonstrate the importance of ERα in regulating T cell functions and suggest that ERα could be exploited as a potential therapeutic target for autoimmune disorders.

MATERIALS AND METHODS

Generation of CD4-creERαfl/fl mice

CD4-cre mice (Taconic) were bred with mice bearing conditional alleles of ERα in which exon 3 is flanked by loxP sites (33, 72) to yield mice with ERα conditionally ablated in T cells, CD4-creERαfl/fl mice, or cre-littermate controls ERαfl/fl. Inbred C57BL/6 mice expressing the CD45.1 congenic marker and Rag1−/− mice were obtained from the Jackson Laboratory. Age-matched female mice were used at 6 to 16 weeks of age for colitis experiments. Mice were maintained in the Central Animal Laboratory at the Turku University or the Central Animal Facility of the College of Veterinary Medicine at the University of Georgia (UGA). All experiments were carried out in accordance with appropriate guidelines for the care and use of laboratory animals and were approved by the UGA Institutional Animal Care and Use Committee and the Finnish Animal Ethics Committee.

T cell transfer model of colitis

Spleens and lymph nodes were disaggregated by pressing through a 70-μm filter. For spleens, red blood cells were lysed with ACK lysing buffer (Invitrogen). Naïve CD4+ T cells were isolated from spleens and lymph nodes of mice as follows. First, CD4+ T cells were enriched using magnetic separation with a CD4+ T Cell Isolation kit (Miltenyi Biotec). Naïve T cells (CD4+CD44loCD62LhiCD25) were further purified by fluorescence-activated cell sorting (FACS) using antibodies against CD4, CD44, CD62L, and CD25 (eBioscience). For colitis experiments, 6- to 7-week-old female Rag1−/− mice were injected intravenously with 400,000 FACS-sorted naïve CD4+CD45RBhiCD25 T cells isolated from either ERαfl/fl or CD4-creERαfl/fl mice. Mice were weighed before injection and then weekly thereafter. Blood was collected from the tail vein at the indicated intervals after transfer and by terminal cardiac puncture at the time of euthanasia (about 9 weeks). Serum cytokines were quantified using a murine inflammation cytokine bead array (BD Biosciences).

RNA-seq and analysis

CD4+ cells were FACS-sorted using antibodies against CD4 and CD3 (BD Biosciences) from spleens of Rag1−/− mice that received either naïve control ERαfl/fl or CD4-creERαfl/fl T cells after the induction of colitis. Total RNA was isolated with an RNeasy Isolation kit (Qiagen). The RNA-seq libraries were prepared using Illumina TruSeq RNA sample preparation kit according to the manufacturer’s instructions. The library was sequenced by Illumina HiSeq 2500 instrument at the Turku Center for Biotechnology. Data analysis was performed by BGI Genomics (New Territories). After filtering and cleaning, the sequencing reads were mapped to a reference genome using the HISAT/Bowtie2 tool. Gene expression level was calculated by RNA-seq by Expectation Maximization (RSEM). Differentially expressed genes were screened using Poisson distribution. Corrected P values were obtained using Bonferroni correction. To further understand ERα biological functions, the KEGG database was used to perform pathway enrichment analysis.

In vitro T cell activation and proliferation and in vivo T cell proliferation

For in vitro T cell activation and proliferation assay, naïve CD4+ T cells were isolated with a CD4+CD62L+ T Cell Isolation Kit II (Miltenyi Biotec) from ERαfl/fl and CD4-CreERαfl/fl mice. For T cell activation assay, T cells were activated with plate-bound anti-CD3 (BD Pharmingen) at 0.5, 1, 2, and 5 μg/ml with 0, 0.5, 1, 2, and 5 μg/ml of plate-bound anti-CD28 (BD Pharmingen) for 1 day. The expressions of CD69 and CD25 were stained with antibodies against CD69 and CD25 (both were from BD Biosciences) and were analyzed by flow cytometry. For in vitro T cell proliferation assay, the isolated naïve CD4+ T cells were stained with 2.5 μM CellTrace Violet or CellTrace Yellow dye (Invitrogen) for 20 min at 37°C. After labeling, cells were washed with 10 ml of complete culture medium [Iscove’s modified Dulbecco’s medium supplemented with 10% charcoal-stripped fetal bovine serum (FBS), 2 mM glutamine, penicillin (100 IU/ml), streptomycin (0.1 mg/ml; Sigma-Aldrich), and 2.5 μM β-mercaptoethanol and without phenol red]. CD4+ T cells were cultured in the presence of anti-CD3 (1 μg/ml; BD Pharmingen) and anti-CD28 (1 μg/ml; eBioscience) for 4 or 5 days and acquired on an LSR II flow cytometer (BD Biosciences), and data were analyzed using FlowJo software (V10, Tree Star). For iTreg polarization, naïve CD4+ T cells were isolated with a CD4+CD62L+ T Cell Isolation Kit II (Miltenyi Biotec) and were cultured with anti-CD3 (1 μg/ml) and anti-CD28 (1 μg/ml) in the presence of recombinant mouse IL-2 (10 ng/ml; R&D Systems) and recombinant human TGFβ1 (10 ng/ml; PeproTech) in complete culture medium. For in vivo T cell proliferation, T cells were isolated from CD45.1+ C57BL6/J congenic mice and CD4-creERαfl/fl mice using a Pan T Cell Isolation kit (STEMCELL Technologies), and naïve CD4 T cells (CD4+CD62LhiCD44lowCD25) were further sorted to purity using antibodies to CD4, CD62L, CD44, and CD25. WT (CD45.1+ C57BL6/J) and CD45.2+CD4-creERαfl/fl naïve T cells were mixed at a 1:1 ratio before labeling with CFSE. The mixture of labeled cells was washed twice with phosphate-buffered saline (PBS), and 1 × 106 total cells were injected intravenously in Rag1−/− recipient mice in a volume of 200 μl. An aliquot of input cells was retained to confirm the starting ratio (input). Mice were euthanized on day 10 or 14, and T cells isolated from the spleens and MLNs were analyzed for CD45.1, CD45.2, and CFSE dilution.

Flow cytometry

For colitis experiments, the spleens and MLNs were harvested from mice and quantified before restimulation for 4 hours in the presence of phorbol 12-myristate 13-acetate (PMA) and ionomycin plus Golgi inhibitor. For the analysis of surface markers, cells were stained in PBS containing either 5% (v/v) FBS for cell sorting or 0.1% (w/v) bovine serum albumin for analysis along with anti-CD4 and anti-TCRβ (eBioscience). Cells were then fixed in 4% formalin for 10 min at room temperature, followed by permeabilization in Perm/Wash buffer (eBioscience). Intracellular staining for IFN-γ and IL-17A was performed with antibodies obtained from eBioscience for 30 min in Perm/Wash buffer according to the manufacturer’s instructions, and cells were acquired using an LSR II flow cytometer. As a measure of T cell activation, cultured T cells were stained with antibodies against the following surface markers: CD4, CD69, and CD25 (BD Biosciences). Intracellular staining for Foxp3 and Bcl2 (BD Biosciences) was carried out using a Foxp3 Staining kit (eBioscience) according to the manufacturer’s instructions. To detect apoptotic and proapoptotic CD4+ T cells, annexin V (BD Biosciences) and PI (Sigma-Aldrich) staining was performed. After activation with anti-CD3 (1 μg/ml) and anti-CD28 (1 μg/ml) for 3 days, CD4+ T cells were washed and stained with annexin V in binding buffer [10 mM Hepes, 140 mM NaCl, and 5 mM CaCl2 (pH 7.4)] for 20 min in the dark at room temperature. The cells were washed and stained with PI. The stained cells were analyzed using an LSR II flow cytometer, and data were analyzed using FlowJo software (V10, Tree Star).

Histopathology

Colonic sections from mice were collected and fixed in 10% neutral buffered formalin for 24 hours at room temperature. Formalin-fixed intestinal tissue was embedded in paraffin, and 4-μm cross sections were mounted on glass slides and stained with hematoxylin and eosin. Histological sections were evaluated by a veterinary pathologist (T.N.) and scored according to the following criteria: (i) distribution of the inflammation: 0, none; 1, focal; 2, multifocal; 3, diffuse; (ii) degree of inflammation: 0, none; 1, mild; 2, moderate; 3, severe; and (iii) extent of erosion and/or ulceration: 0, none; 1, superficial (lamina propria only); 2, moderate (extends to the submucosa); 3, severe (transmural). The cumulative score represents the sum of these three independent criteria.

Retroviral transduction

CD4+ T cells were isolated from spleens of CD4-creERαfl/fl mice using MACS (magnetic-activated cell sorting) CD4+ T Cell Positive Selection kit (Miltenyi Biotec). T cells were activated with anti-CD3 (1 μg/ml) and anti-CD28 (2 μg/ml) for 48 hours. The construct for pMy-Esr1biotag-T2A-mOrange was made from a vector provided by R. Casellas at the National Institute of Arthritis and Musculoskeletal and Skin Diseases (NIAMS), National Institutes of Health. Plat-E packaging cells were transfected with pMy-Esr1biotag-T2A-mOrange or vector control using calcium phosphate precipitation method. After 48-hour transfection and T cell activation, the supernatants containing retrovirus were harvested and spin-inoculated onto activated CD4+ T cells with polybrene (8 μg/ml) at 2000 rpm for 60 min at 30°C. After 3 days, the cells were collected and analyzed by Western blot or flow cytometry.

Immunoblotting

CD4+ T cells were lysed by adding Triton X-100 lysis buffer [50 mM tris-HCl (pH 7.5), 0.5% Triton X-100, 150 mM NaCl, 5% glycerol, and 1% SDS] containing proteinase and phosphatase inhibitors (both from Roche). Protein quantification was carried out with a Detergent-Compatible Protein Assay kit (Bio-Rad), and 6× loading dye [330 mM tris-HCl (pH 6.8), 330 mM SDS, 170 μM bromophenol blue, 6% β-mercaptoethanol, and 30% glycerol] was added to lysate. Thirty to 50 μg of protein was separated on precast 4–15% gels (Bio-Rad). Antibodies used for immunoblotting were NFAT1 (5861S), STAT5 (9363S), Foxp3 (12653S), CDK1 (4688; all from Cell Signaling Technology), Zap70 (Z24820, BD Biosciences), and β-actin (A5441, Sigma-Aldrich).

Seahorse flux analysis

CD4+ T cells from ERαfl/fl and CD4-creERαfl/fl were cultured with plate-bound anti-CD3 and anti-CD28 (both 1 μg/ml) in RPMI without phenol red with 10% fetal calf serum, 2 mM glutamine, penicillin (100 IU/ml), streptomycin (0.1 mg/ml; Sigma-Aldrich), and 2.5 μM β-mercaptoethanol (pH 7.4) at 37°C for 2 days. For analysis, 300,000 cells per well were added to poly-d-lysine–coated plates, and the Seahorse XF Cell Mito Stress Test kit was used on a Seahorse XFp analyzer (Agilent Technologies) according to the manufacturer’s protocol, except that FCCP (carbonyl cyanide p-trifluoromethoxyphenylhydrazone) concentrations of 0.5 and 1.0 μM were used. Briefly, OCR and extracellular acidification rate (ECAR) values were monitored under basal conditions and also after sequential injections with oligomycin (1 μM), FCCP (0.5 and 1 μM), and antimycin A/rotenone (0.5 μM) to the same wells containing cells. Control wells only with media (without cells) were used for background measurements. Cell numbers were obtained using a flow cytometer, and all the values were normalized according to cell numbers. OCR and ECAR results were analyzed in Wave Desktop software (Agilent Technologies).

Quantitative RT-PCR

RNA was purified from CD4+ T cells isolated with Mouse CD4+ T Cell Isolation kit (Miltenyi Biotec) with an RNeasy kit (Qiagen) according to the manufacturer’s instructions. A SuperScript VILO cDNA Synthesis kit (Invitrogen) was used for complementary DNA (cDNA) synthesis and PROBE FAST ABI Prism 2X qPCR Master Mix (Kapa Biosystems) for the TaqMan Master Mix. Primers for murine Esr1 (5′-ACCCGCCGCCGCAGCTGTCTCCTT-3′ and 5′-TGCCGCCTTTCATCATGCCCACTT-3′) were obtained from Sigma-Aldrich. Gene-specific primers and probe were designed from the Universal ProbeLibrary (Roche Applied Science), and probe #93 was used to detect Esr1. The relative quantification of gene expression was carried out with QuantStudio (Applied Biosystems). Relative gene expression (2−ΔΔCt) was calculated using the endogenous control gene Hprt (Applied Biosystems) to normalize the target gene.

Statistical analysis

P values were calculated using Student’s t test or two-tailed Mann-Whitney test, where appropriate. Error bars represent means ± SEM, unless otherwise indicated. All the statistical tests used to analyze the data in this study have been confirmed by a statistician from the Department of Mathematics and Statistics, Faculty of Mathematics and Natural Sciences, University of Turku.

SUPPLEMENTARY MATERIALS

www.sciencesignaling.org/cgi/content/full/11/526/eaap9415/DC1

Fig. S1. Detection of Esr1 deletion or overexpression.

Fig. S2. ERα is not essential for T cell development.

Fig. S3. ERα deletion within T cells limits colonic inflammation in a murine colitis model.

Fig. S4. ERα deletion within T cells limits skin inflammation in a murine T cell transfer model.

Fig. S5. Activation of CD4-creERαfl/fl CD4+ T cells and control T cells with a high concentration of anti-CD3 and varying doses of anti-CD28.

Fig. S6. Analysis of selected mRNAs in splenocytes.

Fig. S7. Oxygen consumption by ERαfl/fl and CD4-creERαfl/fl CD4+ T cells.

Table S1. Transcriptomics analysis of gene expression changes in CD4-creERαfl/fl and control CD4+ T cells.

REFERENCES AND NOTES

Acknowledgments: We thank R. Kirkland at the Department of Infectious Diseases, College of Veterinary Medicine, University of Georgia for excellent technical assistance. We also thank R. Casellas at the NIAMS, NIH for providing retroviral plasmids. We acknowledge M. Georgiadou from the Turku Center for Biotechnology for helping with the Seahorse experiment. We also acknowledge the Finnish Functional Genomics Centre and the Cell Imaging Core at the Turku Center for Biotechnology for technical assistance. Funding: This work was supported by the Academy of Finland the grant 258313 (to Z.C.), the Centre for International Mobility (CIMO) Foundation (to I.S.), and the University of Turku Foundation (to I.M.) and by institutional funds provided (to W.T.W.) by the Office of the Vice President for Research at the University of Georgia. W.T.W. was supported by NIH grant R01AI099058. Author contributions: I.M. performed the experiments, analyzed the data, and wrote part of the manuscript. I.S. performed the experiments and analyzed the data. T.N. performed pathology analysis. J.G. prepared retroviral constructs. E.Y. provided expertise. K.V. provided expertise and guidance. W.T. W. and Z.C. designed the study, performed the experiments, analyzed the data, and wrote the manuscript. Competing interests: The authors declare that they have no competing interests. Data and materials availability: RNA-seq data can be found at the National Center for Biotechnology Information Gene Expression Omnibus with accession number GSE107284. All other data needed to evaluate the conclusions in the paper are present in the paper or the Supplementary Materials.
View Abstract

Navigate This Article