Nonobese diabetic (NOD) mice spontaneously develop diabetes with a strong female prevalence; however, the mechanisms for this gender difference in susceptibility to T cell-mediated autoimmune diabetes are poorly understood. This investigation was initiated to find mechanisms by which sex hormones might affect the development of autoimmune diabetes in NOD mice. We examined the expression of IFN-γ, a characteristic Th1 cytokine, and IL-4, a characteristic Th2 cytokine, in islet infiltrates of female and male NOD mice at various ages. We found that the most significant difference in cytokine production between sexes was during the early stages of insulitis at 4 wk of age. IFN-γ was significantly higher in young females, whereas IL-4 was higher in young males. CD4+ T cells isolated from lymph nodes of female mice and activated with anti-CD3 and anti-CD28 Abs produced more IFN-γ, but less IL-4, as compared with males. Treatment of CD4+ T cells with estrogen significantly increased, whereas testosterone treatment decreased the IL-12-induced production of IFN-γ. We then examined whether the change in IL-12-induced IFN-γ production by treatment with sex hormones was due to the regulation of STAT4 activation. We found that estrogen treatment increased the phosphorylation of STAT4 in IL-12-stimulated T cells. We conclude that the increased susceptibility of female NOD mice to the development of autoimmune diabetes could be due to the enhancement of the Th1 immune response through the increase of IL-12-induced STAT4 activation by estrogen.
Type 1 diabetes, also known as insulin-dependent diabetes mellitus, results from the loss of insulin-producing pancreatic β cells by β cell-specific, cell-mediated autoimmune responses (1, 2, 3). Nonobese diabetic (NOD)4 mice spontaneously develop autoimmune type 1 diabetes and are considered to be one of the best animal models for human insulin-dependent diabetes mellitus (4). The destruction of pancreatic β cells in NOD mice is preceded by infiltration of dendritic cells/macrophages and then T and B cells into the pancreatic islets (5, 6, 7, 8, 9, 10). Infiltration of immunocytes into the periislet region of the pancreas (periinsulitis) begins at 3–4 wk of age, followed by the slow, progressive, and selective destruction of insulin-producing β cells at 4–6 mo of age (11). Both female and male NOD mice show insulitis; however, females exhibit more invasive and destructive insulitis, leading to an earlier onset (12 wk of age) and higher incidence (80–90%) of diabetes as compared with males (20 wk of age, 10–30%).
It has been suggested that sex hormones are associated with the sexual dimorphism in the onset of autoimmune diabetes in NOD mice. The incidence of diabetes was significantly increased in male NOD mice, but decreased in females, by castration at the time of weaning (12). The basal circulating levels of estrogen were found to be about twice as high in female NOD mice as in other strains of mice (13). In addition, the long-term administration of androgen or its derivatives to young female NOD mice resulted in a decrease in the incidence of diabetes (14, 15, 16). However, it is poorly understood how sex hormones modulate the incidence of diabetes in NOD mice. This investigation was initiated to find mechanisms by which sex hormones affect the development of autoimmune diabetes in NOD mice. We found that estrogen increased the IL-12-induced activation of STAT4, which enhanced the Th1 immune response, whereas testosterone (TS) did not significantly change the activation of STAT4. Thus, sex hormones may modulate the Th1/Th2 immune balance through the regulation of IL-12-induced STAT4 activation in the early stages of the T cell-mediated autoimmune process and affect the development of diabetes in NOD mice.
Materials and Methods
All mice used in this study were purchased from Taconic Farms (Germantown, NY) and maintained in the pathogen-free facility in the Health Science Center of the University of Calgary (Calgary, Alberta, Canada). The use and care of the animals in this study were approved by the Animal Care Committee, Faculty of Medicine, University of Calgary.
Isolation of islets
Islets were isolated from female and male NOD mice at 4, 8, 16, and 20 wk of age and from C57BL/6 mice at 20 wk of age by collagenase digestion and Ficoll gradient centrifugation (Sigma-Aldrich, St. Louis, MO), as described elsewhere (17), and pooled for RNA extraction. Each group contains pooled islet RNA from four to six mice.
Isolation of CD4+ T cells
Lymphocytes were prepared from the lymph nodes of female and male NOD mice at 4 and 10 wk of age, and CD4+ T cells were purified using the MACS system anti-CD4 beads (Miltenyi Biotec, München, Germany).The purity was >90% when determined by flow cytometric analysis.
RT-PCR analysis of gene expression
Total RNA was extracted from islets or cultured lymphocytes using TRIzol reagent (Life Technologies, Burlington, Ontario, Canada), according to the manufacturer’s protocol. A total of 3 μg RNA was subjected to the first-strand cDNA synthesis in 20 μl reaction mixture containing 50 mM Tris-HCl (pH 8.3), 75 mM KCl, 3 mM MgCl2, 10 μg/ml oligo(dT), 10 mM DTT, 0.5 mM of each nucleotide, 2 U/ml RNase inhibitor, and 2.5 U Superscript II reverse transcriptase (Life Technologies) at 37°C for 60 min. The samples were then heated at 95°C for 5 min to inactivate the enzymes and diluted to 100 μl with distilled water. Five microliters of serially diluted cDNA samples were added to 15 μl reaction mixture containing 0.2 U Taq DNA polymerase (Sigma-Aldrich) for PCR. The reaction conditions were optimized for each pair of primers. Primers were as follows: β-actin sense (5′-GTTACCAACTGGGACGACA-3′) and antisense (5′-TGGCCATCTCCTGCTCGAA-3′); IFN-γ sense (5′-AGCTCTGAGACAATGAACGC-3′) and antisense (5′-GGACAATCTCTTCCCCACCC-3′); IL-4 sense (5′-TCTTTCTCGAATGTACCAGG-3′) and antisense (5′-CATGGTGGCTCAGTACTACG-3′); estrogen receptor (ER)-α sense (5′-GAGACTGTCCAGCAGTAACGAGAA-3′) and antisense (5′-GGACAAGGCAGGGCTATTC-3′); and TS receptor (TSR) sense (5′-TCTCAAGAGTTTGAATGGCTCC-3′) and antisense (5′-GAGATGATCTCTGCCATCATTTC-3′). The PCR products were visualized on a 1.5% agarose gel by ethidium bromide staining, and the densitometric analysis of the PCR products was performed by Adobe Photoshop 4.1 software (Adobe Systems, Mountain View, CA).
T cell proliferation assay
The isolated CD4+ T cells from lymph nodes were washed and resuspended in serum-free RPMI 1640 medium. The cells (5 × 105 cells/well) were seeded into 96-well round-bottom culture plates coated with 0, 1.25, 2.5, or 5 μg/ml anti-CD3 Ab (145.2C11; BD PharMingen, San Diego, CA) and incubated for 72 h in the presence of 2 μg/ml anti-CD28 Ab (BD PharMingen). [3H]Thymidine (1 μCi) was added to each well, and the cells were cultured for an additional 16 h. The cells were harvested to determine the incorporation of [3H]thymidine.
ELISA for cytokine production
CD4+ T cells were activated with anti-CD3 and anti-CD28 Ab for 72 h, as described above. The culture supernatant was collected, and the production of IFN-γ and IL-4 was determined by ELISA using Duoset ELISA development system (R&D Systems, Minneapolis, MN), according to the manufacturer’s protocol.
IL-4- and IL-12-induced activation of CD4+ T cells
CD4+ T cells (2.5 × 106/ml) isolated from lymph nodes of 4- and 10-wk-old female or male NOD mice were activated with Con A (2.5 μg/ml; Sigma-Aldrich) and IL-2 (20 U/ml; Takeda Chemical Industries, Osaka, Japan) for 3 days. The cells were then washed and reincubated in serum-free RPMI medium in the presence of IL-2 for another 4 days. The cells were washed again and incubated in serum-free culture medium with or without 17β-estradiol (E2, 25 ng/ml; Sigma-Aldrich) or TS (25 ng/ml) for 20 h. IL-12 (20 ng/ml; PeproTech, Rocky Hill, NJ) was added to the culture medium, and the cells were incubated for an additional 48 h. The supernatant was collected for cytokine ELISA, and cells were harvested for RNA extraction to perform RT-PCR analysis. For immunoprecipitation and immunoblot analysis of STAT4 activation, the cells (5 × 106/ml) were incubated for 20 min, instead of 48 h, in the presence of IL-12 (20 ng/ml). For analysis of STAT6 activation, cells were incubated with IL-4 (40 ng/ml), instead of IL-12, for 40 min after sex hormone treatment.
Flow cytometric analysis of IL-12R expression in CD4+ T cells
CD4+ T cells were activated with Con A, rested, and treated with E2 or TS, as described above. The cells were harvested and incubated with purified anti-mouse IL-12R β1 Ab (BD PharMingen). The cells were washed and incubated with biotinylated anti-mouse IgG, followed by incubation with streptavidin-PerCP. The expression of IL-12R was measured by flow cytometry.
Immunoprecipitation and immunoblotting
The cells were washed and lysed in ice-cold lysis buffer containing 1% Triton X-100, 0.5% sodium deoxycholate, 0.1% SDS, 10 mg/ml PMSF, 30 mM aprotinin, and 100 mM sodium orthovanadate. The lysates were left on ice for 20 min and centrifuged at 12,000 × g for 20 min at 4°C. The cleared lysates were then immunoprecipitated with anti-STAT4 or anti-STAT6 Ab (Upstate Biotechnology, Lake Placid, NY). The precipitated protein was resolved on 10% SDS-PAGE and transferred to polyvinylidene difluoride membranes (Bio-Rad, Hercules, CA). The membranes were first probed with anti-phosphotyrosine Ab (4G10; Upstate Biotechnology) and visualized by the ECL system (Amersham Life Science, Arlington Heights, IL). The membranes were subsequently stripped using a solution containing 100 mM 2-ME, 2% SDS, and 62.5 mM Tris-HCl (pH 8), and reprobed with anti-STAT4 or anti-STAT6 Ab (Upstate Biotechnology).
Student’s t test was used to calculate statistical significance in all experiments. A value of p < 0.05 was considered to be significant.
Preferential expression of Th1 cytokines in islet infiltrates of young female NOD mice
To determine whether there is a difference in the level of expression of Th1 (IFN-γ) and Th2 (IL-4) cytokines in islet-infiltrating lymphocytes between female and male NOD mice, we isolated islets from female or male NOD mice at 4, 8, 16, and 20 wk of age and examined the expression of IFN-γ and IL-4 by RT-PCR analysis. The expression of IFN-γ and IL-4 was readily detected in the islets from all ages tested of NOD mice, whereas the expression of these cytokines was not detected in the islets from C57BL/6 mice (Fig. 1,A). The expression of IFN-γ increased with increasing age in both female and male NOD mice up to 16 wk of age, with females showing significantly higher expression of IFN-γ than males at 4 wk of age (Fig. 1, A and B). The expression of IL-4 increased between 4 and 8 wk of age and plateaued thereafter in both males and females. At all ages tested, the expression of IL-4 was significantly higher in males than in females (Fig. 1, A and C). When we examined the ratio of IFN-γ and IL-4 gene expression, it was found to be significantly higher in females than in males at 4 and 8 wk of age (Fig. 1 D).
Differential Th1/Th2 cytokine production in CD4+ T cells of young female and male NOD mice
To determine whether there is a difference in Th1/Th2 cytokine production in T cells of female and male NOD mice, we isolated CD4+ T cells from lymph nodes of young (4-wk-old) and adult (10-wk-old) NOD mice of both sexes and activated them with anti-CD3 and anti-CD28 Abs in serum-free medium, to avoid any contamination by sex hormones. We then examined the proliferative response of the cells and the production of IFN-γ and IL-4. We found that CD4+ T cells from both female and male mice at both ages proliferated equally well at all concentrations of anti-CD3 Ab (Fig. 2, A and B), but T cells from both female and male NOD mice produced more IFN-γ and IL-4 at 10 wk of age compared with 4 wk of age (Fig. 2, C–F). There were considerable differences in the production of IFN-γ and IL-4 between T cells from female and male NOD mice. At 4 wk of age, CD4+ T cells from female NOD mice produced significantly higher amounts of IFN-γ (Fig. 2,C), but lower amounts of IL-4 (Fig. 2,E), compared with CD4+ T cells from males. This difference was more pronounced when CD4+ T cells were activated with higher concentrations of anti-CD3 Ab. At 10 wk of age, the amount of IFN-γ produced by CD4+ T cells from female NOD mice was still higher than that from males (Fig. 2,D), but the production of IL-4 was almost equal between females and males (Fig. 2,F). When we calculated these results as a ratio of IFN-γ:IL-4, we found that this ratio was significantly higher in female mice than in male mice at 4 wk of age (Fig. 2,G), whereas it was not different between females and males at 10 wk of age (Fig. 2 H). These results suggest that the gender-associated activation of the Th1 response, as indicated by the up-regulation of IFN-γ, occurs predominantly in the T cells of young mice as compared with adult mice.
The expression of IFN-γ and IL-4 in CD4+ T cells was also examined at the mRNA level by semiquantitative RT-PCR. The expression of IFN-γ was higher in females than in males at both 4 and 10 wk of age. However, the expression of IL-4 was higher in males than in females at 4 wk of age (Fig. 3, A and C). The ratio of IFN-γ:IL-4 mRNA was significantly higher in females than in males at 4 wk of age (Fig. 3,B), but was not different at 10 wk of age (Fig. 3,D), consistent with the production of these cytokines at the protein level. This result indicates that gender factors influence the expression of IFN-γ and IL-4 at the transcriptional level. To determine whether this differential expression in T cells between female and male mice is due to different levels of expression of sex hormone receptors, we examined the expression of ER and TSR in CD4+ T cells from 4- and 10-wk-old NOD mice. We found no difference in the expression level of sex hormone receptors in the T cells between males and females at any age (Fig. 4).
Increase of IL-12-induced production of IFN-γ in T cells by estrogen treatment and decrease by TS treatment
As we found that there was a gender difference in the production of IFN-γ at 4 wk of age, but not at 10 wk of age, we asked whether exposure to TS or E2 in vitro might reveal a gender difference in cytokine production by CD4+ T cells from 10-wk-old NOD mice. Purified CD4+ T cells from lymph nodes from 10-wk-old female or male NOD mice were activated with Con A or anti-CD3 Ab in the presence of E2 or TS. However, we found no difference in the production of IFN-γ and IL-4 between T cells from female or male mice treated with or without E2 or TS (data not shown). We then asked whether prolonged cell proliferation and rest in the absence of sex steroids would result in a response to subsequent sex steroid treatment of CD4+ T cells from 10-wk-old mice. In this experiment, CD4+ T cells from 10-wk-old female NOD mice were first activated with Con A for 3 days, rested in IL-2-containing serum-free medium for another 4 days, and exposed to E2 or TS for 20 h before being stimulated with IL-12, a potent inducer of Th1 cytokines. We found that steroid treatment did not affect the proliferative response of the T cells to Con A and IL-12 in male or female NOD mice (data not shown). However, we found that E2-treated T cells from both males and females clearly showed an increased production of IFN-γ, whereas TS-treated T cells from females, but not males, showed a decreased production of IFN-γ compared with untreated IL-12-stimulated T cells (Fig. 5,A). IFN-γ mRNA expression was also increased by estrogen treatment and decreased by TS treatment in female NOD mice (Fig. 5, B and C). Similar results were obtained for CD4+ T cells from both female and male NOD mice; therefore, only the data from female mice are shown in Fig. 5, B and C. These results indicate that sex hormones directly regulate IL-12-induced cytokine production of T cells.
To determine whether the IL-12-induced change in IFN-γ production in E2- or TS-treated T cells was due to a change in IL-12R expression, which is up-regulated in Th1 cells, we examined the gene expression of IL-12R β1 in T cells activated by Con A and IL-2 and treated with E2 or TS by flow cytometric analysis. We found that treatment with E2 or TS did not significantly change the expression of the IL-12R (Fig. 6,A). In addition, we examined the expression of sex hormone receptors by RT-PCR and found that treatment with sex hormones did not affect the expression of E2 or TS receptors in these cells (Fig. 6 B).
Increase of IL-12-induced STAT4 activation in T cells by estrogen treatment
To determine whether the differential expression of IFN-γ in IL-12-stimulated T cells results from the modulation of the activation of STAT4, which is known to play a key role in IL-12-induced IFN-γ expression (18, 19), CD4+ T cells were activated with Con A, rested, and then incubated with E2 or TS for 20 h. After incubation of these cells with IL-12 for 20 min, we examined the level of phosphorylated STAT4 by Western blot. We found that T cells treated with E2 had increased phosphorylation of STAT4, whereas this was not significantly different in T cells treated with TS as compared with untreated T cells (Fig. 7, A and B). When we examined the effect of sex hormones on IL-4-induced activation of STAT6, which play a pivotal role in Th2 differentiation (20, 21), we found no change in the phosphorylation of STAT6 in IL-4-stimulated T cells in the presence of estrogen or TS (Fig. 7, C and D).
It has been known that gender influences the susceptibility to various autoimmune diseases (22). Females are generally more susceptible to diseases such as multiple sclerosis, rheumatoid arthritis, systemic lupus erythematosus, Sjögren’s syndrome, myasthenia gravis, and experimental allergic encephalomyelitis. Although the incidence of type 1 diabetes is similar in both men and women, the NOD mouse, an animal model of type 1 diabetes, shows female prevalence of the disease. Castration of male NOD mice increases, whereas castration of female NOD mice decreases the frequency of autoimmune diabetes, and treatment of female NOD mice with TS prevented the development of diabetes (14, 15, 16). However, the mechanisms for the gender differences in susceptibility to autoimmune diabetes are poorly understood.
The immune systems of females and males have different characteristics. For example, higher Ab production after immunization was observed in female mice, women have higher absolute numbers of CD4+ T lymphocytes (23), and higher Th1 cytokine production was observed in female mice (24). In addition, sex hormones were shown to modulate the cytokine profiles in immunocytes (25) and T cell lines and clones (26, 27) in vitro. As well, the expression of cytokine genes was found to be influenced by sex hormones in an animal model of experimental autoimmune encephalitis. The production of Th1 cytokines such as IL-12 and IFN-γ was significantly less in lymph node cells from male as compared with female mice after Ag-specific stimulation (28). Therefore, it is conceivable that sex hormones may contribute to the gender difference in the development of diabetes in NOD mice by influencing the balance of Th1/Th2 cytokine production, as it is known that the Th1 response plays a pathogenic role and the Th2 response plays a preventive role in the development of autoimmune diabetes in NOD mice (29, 30, 31, 32, 33).
We first examined the expression of IFN-γ, as a characteristic Th1 cytokine, and IL-4, as a characteristic Th2 cytokine, in the islet infiltrates in female and male NOD mice at various ages. We found that the most significant difference between sexes was during the early stages of insulitis at 4 wk of age; the expression of IFN-γ was high in females, whereas the expression of IL-4 was high in males. A similar result was also reported in a previous study (34). Mice approach sexual maturation at 4 wk of age, and the levels of sex hormones increase in the circulation at this time. The increased levels of sex hormones at this age probably influence cytokine gene expression.
Second, we examined the production of Th1/Th2 cytokines in CD4+ T cells activated with anti-CD3 Ab in female and male NOD mice. We found that T cells from 10-wk-old mice produced more IFN-γ and IL-4 than T cells from 4-wk-old mice. This might be due to the relatively poor activation of T cells from young NOD mice as compared with adults (32, 35). However, female mice produced significantly higher amounts of IFN-γ, but lesser amounts of IL-4 at 4 wk of age, whereas only a slight difference between females and males was observed at 10 wk of age. Therefore, the major impact of gender factors on the pathogenesis of autoimmune diabetes in NOD mice may be at an early, rather than later, age. Consistent with this notion, castration at weaning drastically changed the course of the development of autoimmune diabetes in both female and male NOD mice (12). One possible explanation for this difference in cytokine production between T cells from 4- and 10-wk-old NOD mice is that the T cells might express different levels of sex hormone receptors and respond to sex hormones according to the numbers of receptors in the cells. Thus, we examined the expression of ER and TSR on CD4+ T cells of young (4-wk-old) and adult (10-wk-old) mice. We found no difference in the level of expression of these receptors at any age, indicating that sex hormone receptors on the CD4+ T cells are not involved in the effect of age on the production of cytokines from these cells.
Third, we examined whether treatment with estrogen or TS would alter the Th1/Th2 (IFN-γ/IL-4) cytokine profile of activated CD4+ T cells from female and male NOD mice. We could not find any difference in the production of these cytokines when we activated T cells with Con A or anti-CD3 Ab in the presence of estrogen or TS. Therefore, we activated T cells with Con A, rested them in IL-2 for 4 days in the absence of sex hormones, exposed them to estrogen or TS, and then activated them with IL-12. We found that estrogen treatment clearly increased, whereas TS treatment decreased IFN-γ mRNA expression and IFN-γ production. These results suggest that transcriptional regulation of cytokine gene expression is programmed by an initial exposure to sex hormones, and reprogram of the gene expression by a different sex hormone requires a long time. An earlier study found that female recipient mice implanted with dihydrotestosterone pellets for 10–14 days before the transfer of encephalitogenic T cells showed a significantly less severe course of encephalitis as compared with control placebo pellet-implanted female mice (36). This may have been due to the enhancement of the Th2 shift by the TS treatment. In addition, Ag-specific T cells from male donors induced experimental autoimmune encephalomyelitis in female recipients with a lower severity than that induced by T cells from female donors (37), as female hormones in the recipients could not change the cytokine gene expression of T cells from male donors (Th2 biased) during the disease induction period, ∼10 days.
Fourth, we determined whether the change in IL-12-induced IFN-γ production by treatment with sex hormones may result from a modulation of IL-12R β1 expression on the T cells. The high affinity IL-12R complex requires both β1 and β2 subunits, and these two subunits are up-regulated by T cell activation (38, 39, 40, 41). We found that there was no change in the expression of IL-12R β1 by treatment with sex hormones.
Fifth, we then examined whether the change in IL-12-induced IFN-γ production by treatment with sex hormones is due to the regulation of STAT4 activation. We found that estrogen treatment increased the phosphorylation of STAT4 in IL-12-stimulated T cells. It was reported that the sequences similar to the estrogen-responsive element were identified in the IFN-γ gene promoter region (42). However, there is no evidence showing that estrogen/ER complex directly binds to the IFN-γ gene promoter. STAT4 is a key transcription factor for IL-12-induced IFN-γ gene expression (18, 19). Therefore, it is conceivable that sex hormones may modulate IFN-γ gene expression possibly by regulating the activation of STAT4, although the mechanisms remain to be elucidated.
In conclusion, we have shown that there are gender differences in Th1/Th2 cytokine expression in T cells of NOD mice. The expression of IFN-γ was significantly higher in T cells from young female mice, whereas the expression of IL-4 was significantly higher in T cells from young male mice. This differential expression was found to be due to the up-regulation of IL-12-induced STAT4 activation by estrogen. Therefore, sex hormones may affect the development of autoimmune diabetes by modulating Th1/Th2 cytokine production via the regulation of IL-12-induced STAT4 activation.
We are grateful to Dr. Merle Olsen for advice and help with animal care, Lori Bryan for assistance with flow cytometry, and Dr. Ann L. Kyle for editorial assistance.
This work was supported by a grant from the Canadian Institutes of Health Research (MA 9584). Y.Y. is a recipient of the Career Development Award from the Juvenile Diabetes Research Foundation International. J.-W.Y. holds a Canada Research Chair in Diabetes.
Abbreviations used in this paper: NOD, nonobese diabetic; E2, 17β-estradiol; ER, estrogen receptor; TS, testosterone; TSR, TS receptor.