Autoimmune liver diseases predominantly affect women. In this study, we aimed to elucidate how sex affects autoimmune hepatic inflammation. Acute experimental cholangitis was induced by adoptive transfer of OVA-specific CD8+ T cells into mice, which express the cognate Ag on cholangiocytes. In contrast to previous mouse models of cholangitis, this model displayed a strong sexual dimorphism: female mice developed marked cholangitis, whereas male mice were resistant to cholangitis induction. The recruitment of endogenous CD4+ T cells, but not transferred CD8+ T cells into female livers was strongly increased. These cells expressed higher amounts of the proinflammatory cytokine IL-17, which was at least in part responsible for the liver inflammation observed. The recruitment of endogenous CD4+ T cells was associated with increased expression of the chemokines CXCL-9 and CXCL-10 in female livers. The sex-specific factor responsible for the observed differences was found to be testosterone: male mice could be rendered susceptible to liver inflammation by castration, and testosterone treatment was sufficient to completely suppress liver inflammation in female mice. Accordingly, testosterone treatment of female mice significantly reduced the expression of IL-17A, CXCL-9, and CXCL-10 within the liver. Serum testosterone levels of untreated mice negatively correlated with the IL-17, CXCL-9, and CXCL-10 expression in the liver, further supporting a role for testosterone in hepatic immune homeostasis. In conclusion, testosterone was found to be the major determinant of the observed sexual dimorphism. Further study into the role of testosterone for liver inflammation could lead to novel treatment targets in human autoimmune liver diseases.

Most autoimmune diseases occur more frequently in women than in men (1). The autoimmune liver diseases primary biliary cirrhosis (PBC) and autoimmune hepatitis (AIH) range among the diseases with the highest female predominance. In PBC, the ratio of affected women to men can be as high as 10 to 1 (2), and in AIH the ratio is approximately 4 to 1 (3).

Sex hormones have been shown to shape the differentiation and functioning of the immune system significantly (4). Estrogen and androgen receptors have been described in T and B cells (5). Whereas estrogens seem to direct the immune system toward a T helper type 2 (Th2) differentiation (6), androgens seem to favor the development of a T helper type 1 (Th1) response (1, 7), characterized by the expression of IFN-γ. From clinical observations, the influence of sex hormones on the activity of autoimmune liver inflammation can be seen by the strong influence of pregnancy on the course of AIH (8, 9) and, to lesser extent, of primary sclerosing cholangitis (PSC) (10). Sex hormones can not only influence T cell differentiation; they can also modulate the homing of lymphocytes to a target organ and the process of Ag presentation (4). Along this line, estrogen has been described to upregulate IL-6 and to induce the secretion of proinflammatory chemokines in immature dendritic cells (11). A protective role of testosterone, however, has been demonstrated in several models of autoimmune diseases, such as experimental autoimmune encephalomyelitis (EAE), diabetes in nonobese diabetes mice, thyroiditis, and adjuvant arthritis (12).

Previously described mouse models of cholangitis lack the female predominance seen in autoimmune liver inflammation in humans (13). We describe an Ag-induced experimental cholangitis model with a strong female predominance. The observed sex-dependent differences seemed to depend on the presence of testosterone, which was able to suppress cholangitis in female mice. To our knowledge, this study is the first to demonstrate the ability of testosterone to suppress liver inflammation. These results should stimulate further research into the role of testosterone for the maintenance of immune homeostasis in the liver.

K14-OVAp mice (C57BL/6 background) express the antigenic MHC class I restricted SIINFEKL peptide (residues 257–264 of OVA) under control of the human keratin 14 promoter and were provided by K. Hogquist (Minneapolis, MN). Although these mice were developed to study the thymic development of T cells, they were shown to express OVA on biliary epithelial cells (BECs) and to develop liver inflammation when crossed to OT-1 transgenic mice (14, 15). K14-OVAp were crossed to IL-17af−/− mice to investigate the role of endogenous IL-17 on liver inflammation. IL-17af−/− mice were described previously (16). CD8+ T cells from OT-1 mice (Jackson Laboratory, Bar Harbor, ME) express a transgenic receptor specific for the SIINFEKL peptide bound on H-2Kb. In our experiments, CD45.1 allele-congenic OT-1 donor mice were used to distinguish transferred from recipient lymphocytes of CD45.2+ K14-OVAp mice. All animals were age matched (8–10 wk). Animal care was in accordance with the governmental and institutional guidelines and all experiments were approved by the animal experimentation committee of the State of Hamburg (G10/03; G12/113).

Castration of male mice and pellet implantations were conducted on 4-wk-old mice as described previously (17). Sixty-day release pellets of testosterone (12.5 mg/pellet) or β-estrogen (0.36 mg/pellet) were purchased from Innovative Research of America (Sarasota, FL) and implanted s.c. in the neck.

For the induction of cholangitis, freshly isolated congenic OT-1 CD8+ T cells from female donors were injected i.p. (4 × 106 to 20 × 106 T cells) into K14-OVAp male and female recipient mice as well as K14-OVApxIL-17a/f knockout female mice. OT-1 CD8+ T cells were isolated with anti–CD8-FITC Ab (BioLegend, Fell, Germany) and anti-FITC immunomagnetic beads (Miltenyi Biotec, Bergisch Gladbach, Germany) according to the manufacturer’s instructions.

For the isolation of liver nonparenchymal cells (NPCs), mouse livers were perfused with PBS and dissected mechanically. Hepatocytes and debris were sedimented twice at 40 × g, and nonparenchymal cells were recovered by centrifugation over a 35% Optiprep (Sigma-Aldrich, Taufkirchen, Germany) gradient at 400 × g.

Spleen cells and liver NPCs (5 × 105 per well) were restimulated for 24 h with 2 μg/ml anti-CD3 and 2 μg/ml anti-CD28 (BD Biosciences, Heidelberg, Germany) Abs. Supernatants were analyzed for IFN-γ and IL-17 levels by ELISA using DuoSet mouse ELISA kits (R&D Systems, Wiesbaden, Germany). The results shown are representative for at least three similar experiments.

CD4+ T cells were isolated from spleens of C57/Bl6J WT mice using magnetic cell separation (Miltenyi, Bergisch Gladbach, Germany) according to manufacturer’s description. CD4+ T cells were cultured in the presence of irradiated CD3 APCs (4 × 105, 30 Gray, 11 min) for five days in Clicks medium supplemented with TGFβ1, IL-1β, IL-6, IL-23, anti–IFN-γ and anti–IL-4 (R&D Systems, Wiesbaden, Germany) (18). Ten percent testosterone-enriched serum (derived from mice supplemented with testosterone pellets) was added as indicated. FCS and serum derived from female mice were used as control.

Serum liver enzymes were measured with a Hitachi Modular E170 analyzer (Boehringer Mannheim, Mannheim, Germany). Serum anti-mitochondrial Abs (AMAs) were measured using the QUANTA LiteM2 EP (MIT3) ELISA kit (INOVA Diagnostic, San Diego, CA).

Immunofluorescent surface staining of liver NPCs and spleen cells was performed with Abs to CD3, CD8, CD4, CD25, CD45.1, and CD45.2. For intracellular cytokine staining, cells were treated with Golgi Plug (1 μl/ml; BD Biosciences, Heidelberg, Germany) and restimulated with PMA (50 ng/ml) and ionomycin (1 μg/ml; Sigma-Aldrich, Taufkirchen, Germany) for 4 h. Fixed cells were then perforated in buffer containing Saponin/BSA (0.5%/2%; Sigma-Aldrich, Taufkirchen, Germany) and stained for IFN-γ and IL-17. All Abs for T cell staining were purchased from BioLegend. Intracellular Foxp3 staining was performed with Foxp3 staining buffer and anti-Foxp3 Ab (both from eBioscience, Frankfurt, Germany). Flow cytometry data were analyzed with FACS Diva software (BD Biosciences, Heidelberg, Germany).

Total RNA was isolated from total liver tissue using Nucleospin Kit (Macherey Nagel, Düren, Germany). cDNA was reverse-transcribed from total RNA (1st strand RNA cDNA Synthesis Kit; Roche, Mannheim, Germany) followed by RT-PCR using the Bio-Rad CFX96 real-time system (Bio-Rad, München, Germany) and absolute PCR SYBR mix (Thermo Fisher, Schwerte, Germany). Sequence-specific primers for CXCL-9 (forward 5′ CTT GAG CCT AGT CGT GAT AAC 3′; reverse 5′ CCA GCT TGG TGA GGT CTA TC 3′), CXCL-10 (forward 5′ CCA CGT GTT GAG ATC ATT GC 3′; reverse 5′ AGT AGC AGC TGA TGT GAC C 3′) and CCL-20 (forward 5′ CGA CTG TTG CCT CTC GTA CA 3′; reverse 5′ AGG AGG TTC ACA GCC CTT TT 3′) were used. Relative mRNA levels were calculated after normalization to β-actin with male mice as calibrator (=1), using the CFX96 Manager software. The expression of RORc mRNA was measured using the Taqman Universal PCR master mix (Applied Biosystems, Darmstadt, Germany) and Taqman gene expression kits (Applied Biosystems) were used for amplification, namely Rorc (RORγτ, Mm01261022_m1). Target gene expression was normalized to Hprt (Mm00446968_m1).

H&E staining (Roth, Karlsruhe, Germany) was performed on formalin-fixed liver sections. SIINFEKL (eBioscience, Frankfurt, Germany) staining was performed on cryofrozen tissue. All histologic scorings were performed by a pathologist in a blinded fashion. Scoring of liver inflammation was analyzed according to the modified hepatitis activity index (mHAI), which consists of the four subgroups A–D, with D representing the grade of portal inflammation (19). Scoring of skin and esophagus was analyzed according to an internal standard: 0 = no inflammation, 1 = minimal inflammation, 2 = mild inflammation, 3 = moderate inflammation, 4 = strong inflammation.

Differences between two experimental groups were assessed for statistical significance with Student t test (p < 0.05 was considered significant). Differences between more than two groups were assessed using the one-way ANOVA Test and Tukey post test (p < 0.05 was considered significant). Correlation was calculated using the Pearson correlation coefficient (r), and p < 0.05 was considered significant. The p values are as follows: *p < 0.05, **p < 0.01, ***p < 0.001.

There is a strong female preponderance in autoimmune liver diseases. We therefore sought to establish a model, in which the mechanisms behind this sexual dimorphism could be investigated. Experimental cholangitis was induced by the transfer of OVA-specific CD8+ T cells from OT-1 transgenic mice into male or female K14-OVAp recipient mice. In this model, the adoptive transfer of 4 × 106 OVA specific CD8+ T cells induced an increase in serum aminotransferase levels in female, but not in male recipient mice (Fig. 1A). Liver enzymes from female mice were significantly increased on day 8 and day 9 after OVA-specific CD8+ T cell transfer (Fig. 1B). Male mice were resistant to liver inflammation even after transfer of 10 × 106 OVA specific CD8+ T cells (Fig. 1C). Histologically, cholangitis and lymphocytic portal tract inflammation was only observed in female recipient mice (Fig. 1D, 1E). Signs of autoimmune hepatitis such as interface hepatitis, lymphoplasmacytic infiltrate, rosetting, emperipolesis or lobular inflammation were missing. In other models described, experimental cholangitis has been shown to be associated with the development of AMA, in humans a hallmark of PBC (2), but in mice a nonspecific sign of cholangitis (20, 21). Supporting the female predilection described in this model, we could detect AMA in 50% of K14-OVAp female but not in male mice after the induction of cholangitis (female 4/7 [mean ± 58.10 U/ml]; male 0/7 [mean ± 22.45 U/ml]; cutoff, 30 U/ml]. We did not find any differences in the hepatic expression levels of OVA between female and male mice that could account for the observed sexual dimorphism (Supplemental Fig. 1). As in this model, the expression of OVA is not restricted to BEC, we analyzed whether there is a sexual dimorphism also in the inflammatory activity in esophagus and skin of male and female recipient mice after OVA-specific CD8+ T cell transfer. In contrast to the prominent sex-dependent differences observed in the liver, there were no significant differences between male and female mice in skin and esophagus (Supplemental Fig. 2).

FIGURE 1.

Cholangitis developed in female but not in male mice. (A) Alanine aminotransferase (ALT) levels in female and male mice on day 8 after transfer of 4 × 106 OT-1 CD8+ T cells (mean ± SD; n = 4–5; one representative result from four independently performed experiments). (B) ALT serum levels in female mice after transfer of 4 × 106 OT-1 CD8+ T cells (in black) and PBS injection (in gray) (mean ± SD; n = 4; one representative result from two independently performed experiments). (C) ALT serum levels in male mice after transfer of 10 × 106 OT-1 CD8+ T cells (in black) and PBS injection (in gray) (mean ± SD; n = 4; one representative result from two independently performed experiments). (D) H&E staining and (E) mHAI histologic activity score of liver sections from K14-OVAp female and male mice on day 8 after transfer of 4 × 106 OT-1 CD8+ T cells (mean ± SD; n = 5; one representative result from four independently performed experiments). *p < 0.05, **p < 0.01, ***p < 0.001.

FIGURE 1.

Cholangitis developed in female but not in male mice. (A) Alanine aminotransferase (ALT) levels in female and male mice on day 8 after transfer of 4 × 106 OT-1 CD8+ T cells (mean ± SD; n = 4–5; one representative result from four independently performed experiments). (B) ALT serum levels in female mice after transfer of 4 × 106 OT-1 CD8+ T cells (in black) and PBS injection (in gray) (mean ± SD; n = 4; one representative result from two independently performed experiments). (C) ALT serum levels in male mice after transfer of 10 × 106 OT-1 CD8+ T cells (in black) and PBS injection (in gray) (mean ± SD; n = 4; one representative result from two independently performed experiments). (D) H&E staining and (E) mHAI histologic activity score of liver sections from K14-OVAp female and male mice on day 8 after transfer of 4 × 106 OT-1 CD8+ T cells (mean ± SD; n = 5; one representative result from four independently performed experiments). *p < 0.05, **p < 0.01, ***p < 0.001.

Close modal

As cholangitis was induced by the adoptive transfer of OT-1 CD8+ T cells, we analyzed whether the transferred OT-1 cells were differentially recruited to female or male livers. The differences in liver inflammation between male and female mice were not associated with altered frequencies of transferred OVA specific CD45.1+CD8+ T cells in the livers of recipient male or female mice (Fig. 2 A; 55% in males versus 59% in females of CD3+CD8+ T cells on day 8 after transfer). To exclude differences in the activation status of OVA-specific CD8+ T cells migrating into female and male livers, intracellular IFN-γ levels were analyzed in CD45.1+ OT-1 T cells reisolated from livers on day 8 after transfer. The vast majority of transferred CD8+ T cells expressed IFN-γ without differences between cells isolated from female or male livers (Fig. 2B).

FIGURE 2.

The intrahepatic frequency of transferred CD8+ T cells and regulatory T cells is not different between male and female mice. (A) OT-1 CD8+CD45.1+ T cells (4 × 106) were injected into female and male K14-OVAp mice. On day 8 after injection, mice were sacrificed and the percentage of CD8+CD45.1+ T cells within the liver was determined with flow cytometry (mean ± SD; n = 4). (B) The activation status of transferred OT-1 CD8+CD45.1+ T cells in the liver of K14-OVAp female and K14-OVAp male mice on day 8 after transfer was determined by flow cytometry of intracellular IFN-γ expression (mean ± SD; n = 4). (C) On day 8 after OT-1 CD8+ T cell transfer, CD4+CD25+Foxp3+ Treg frequencies in spleen and liver were analyzed with flow cytometry (mean ± SD; n = 8).

FIGURE 2.

The intrahepatic frequency of transferred CD8+ T cells and regulatory T cells is not different between male and female mice. (A) OT-1 CD8+CD45.1+ T cells (4 × 106) were injected into female and male K14-OVAp mice. On day 8 after injection, mice were sacrificed and the percentage of CD8+CD45.1+ T cells within the liver was determined with flow cytometry (mean ± SD; n = 4). (B) The activation status of transferred OT-1 CD8+CD45.1+ T cells in the liver of K14-OVAp female and K14-OVAp male mice on day 8 after transfer was determined by flow cytometry of intracellular IFN-γ expression (mean ± SD; n = 4). (C) On day 8 after OT-1 CD8+ T cell transfer, CD4+CD25+Foxp3+ Treg frequencies in spleen and liver were analyzed with flow cytometry (mean ± SD; n = 8).

Close modal

Differences in regulatory T cell (Treg) numbers have been described to contribute to sexual differences observed in an experimental hepatitis model (22). We therefore investigated the CD4+CD25+Foxp3+ Treg frequencies in the livers of male and female recipient mice on day 8 after cell transfer. There was no significant difference in hepatic or splenic Treg frequencies between female and male mice (Fig. 2C), suggesting that Tregs are not responsible for the observed sexual dimorphism.

To elucidate the mechanisms behind the sex dimorphism of liver inflammation, the recruitment of endogenous lymphocytes of the recipient mice into the liver was investigated. The recruitment of endogenous CD45.2+CD4+ T cells was significantly increased in female as compared with male livers (28 × 105 cells per 1 g female liver versus 9.4 × 105 cells per 1 g male liver tissue; p < 0.0001; Fig. 3A), indicating a proinflammatory role for endogenous CD4+ T cells in our cholangitis model. To investigate the functional role of endogenous CD4+ T cells we performed transfer experiments of hepatic CD4+ T cells isolated from K14-OVAp mice after cholangitis induction into C57BL76J wild-type animals. Twenty-four hours after CD4+ T cell transfer the largest fraction of K14-OVAp CD4+ T cells was detected in the liver (24%) or in liver-infiltrating lymph nodes (5%) of the wild-type recipient mice. We could hardly detect any K14-OVAp CD4+ T cells in the spleens of recipient mice (0.2%). As control, we analyzed the liver homing of activated CD4+ T cells derived from Smarta mice, which express a transgene TCR with specificity for LCMV (lymphocytic choriomeningitis virus) and which were activated in an Ag-specific manner. Smarta CD4+-derived T cells were mainly localized in spleen (17%) and peripheral lymph nodes (4%), and almost no transferred cells were detected in livers (Fig. 3B). Moreover, the transfer of CD4+ T cells isolated from inflamed livers of K14-OVAp mice could cause liver inflammation in K14-OVAp recipient mice even in the absence of OT-1 cells (Fig. 3C). Thus, liver-derived CD4+ T cells from cholangitis mice are able to target the liver and cause hepatic inflammation even in the absence of Ag-specific OT-1 cells.

FIGURE 3.

Increased recruitment of endogenous CD4+ T cells into and increased expression of the chemokines CXCL-9 and CXCL-10 in the livers of female mice. (A) Total number of CD4+ T cells per gram of liver tissue in female and male recipient mice on day 8 after transfer of 4 × 106 OT-1 CD8+ T cells (mean ± SD; n = 5; one representative result from four independently performed experiments). (B) Activated Smarta CD4+ T liver cells and CD4+ T cells isolated from K14-OVAp mice livers after cholangitis induction were adoptively transferred into wild-type mice. The recruitment of transferred cells was analyzed 12 h after transfer (mean ± SD; n = 4). (C) ALT serum levels in female K14-OVAp mice after transfer of 2 × 106 CD4+ T cells isolated from K14-OVAp mice after cholangitis induction (mean ± SD; n = 3–4). (D) Liver NPCs and spleen cells were isolated from female and male mice on day 8 after transfer of 4 × 106 OT-1 CD8+ T cells and were restimulated for 24 h with anti-CD3 and anti-CD28. IL-17 levels were measured in the supernatants by ELISA (mean ± SD; n = 4; four independently performed experiments). Percentage of IL-17–producing CD4+ T cells were detected by flow cytometric analysis (mean ± SD; n = 4). Relative mRNA expression of RORc in liver tissue from male and female recipient mice on day 8 after cell transfer of 4 × 106 OT-1 CD8+ T cells (mean ± SD; n = 4; four independently performed experiments). (E) ALT serum levels and grading of inflammatory activity in K14-OVAp × IL17A+F knockout and K14-OVAp female recipient mice after transfer of OT-1 CD8 T cells on day 8 after transfer (mean ± SD; n = 8–12; performed twice independently). (F) Relative mRNA expression of chemokine ligands in liver tissue from female and male mice on day 8 after transfer of 4 × 106 OT-1 CD8+ T cells (mean ± SD; n = 5; four independently performed experiments). *p < 0.05, **p < 0.01, ***p < 0.001.

FIGURE 3.

Increased recruitment of endogenous CD4+ T cells into and increased expression of the chemokines CXCL-9 and CXCL-10 in the livers of female mice. (A) Total number of CD4+ T cells per gram of liver tissue in female and male recipient mice on day 8 after transfer of 4 × 106 OT-1 CD8+ T cells (mean ± SD; n = 5; one representative result from four independently performed experiments). (B) Activated Smarta CD4+ T liver cells and CD4+ T cells isolated from K14-OVAp mice livers after cholangitis induction were adoptively transferred into wild-type mice. The recruitment of transferred cells was analyzed 12 h after transfer (mean ± SD; n = 4). (C) ALT serum levels in female K14-OVAp mice after transfer of 2 × 106 CD4+ T cells isolated from K14-OVAp mice after cholangitis induction (mean ± SD; n = 3–4). (D) Liver NPCs and spleen cells were isolated from female and male mice on day 8 after transfer of 4 × 106 OT-1 CD8+ T cells and were restimulated for 24 h with anti-CD3 and anti-CD28. IL-17 levels were measured in the supernatants by ELISA (mean ± SD; n = 4; four independently performed experiments). Percentage of IL-17–producing CD4+ T cells were detected by flow cytometric analysis (mean ± SD; n = 4). Relative mRNA expression of RORc in liver tissue from male and female recipient mice on day 8 after cell transfer of 4 × 106 OT-1 CD8+ T cells (mean ± SD; n = 4; four independently performed experiments). (E) ALT serum levels and grading of inflammatory activity in K14-OVAp × IL17A+F knockout and K14-OVAp female recipient mice after transfer of OT-1 CD8 T cells on day 8 after transfer (mean ± SD; n = 8–12; performed twice independently). (F) Relative mRNA expression of chemokine ligands in liver tissue from female and male mice on day 8 after transfer of 4 × 106 OT-1 CD8+ T cells (mean ± SD; n = 5; four independently performed experiments). *p < 0.05, **p < 0.01, ***p < 0.001.

Close modal

As we found sex-dependent differences in the recruitment of endogenous CD4+ T cells into inflamed livers, we next wanted to know whether IL-17 expression was also affected by sex. Of note, the cytokine IL-17 has been implicated in the pathogenesis of PBC, as well as PSC and experimental cholangitis models (2326). To that end, spleen cells and liver NPCs were isolated from female and male mice on day 8 after transfer of OVA-specific CD8+ T cells. After restimulation of male or female spleen cells, IL-17 production was equally low; in contrast, lymphocytes isolated from female livers produced significantly more IL-17 as compared with cells isolated from male livers (220.8 pg/ml by female liver lymphocytes versus 57.5 pg/ml by male liver lymphocytes; p < 0.017). In accordance with the enhanced ex vivo IL-17 production by female liver-infiltrating lymphocytes, we detected an increased frequency of IL-17–expressing CD4+ T cells and higher amounts of RORc mRNA in female livers, as compared with male livers (Fig. 3D). The RORc gene encodes the lineage-specific master transcription factor that regulates the differentiation and function of Th17 cells. The role of endogenously produced IL-17 as a modulator of cholangitis in our model was confirmed by cell transfer experiments into K14-OVAp female recipient mice that lack endogenous IL-17A and F. OT-1 CD8 T cell transfer into female K14-OVApxIL-17a/f knockout mice led to significantly decreased ALT levels as compared with female IL-17 K14-OVAp wild-type mice. Histologically, the severity of cholangitis in K14-OVApxIL-17a/f knockout mice was significantly decreased compared with female K14-OVAp IL-17 wild-type mice (Fig. 3E).

Because we observed sex-dependent differences in the recruitment of endogenous CD4+ T cells expressing IL-17 into the liver, we next investigated whether the expression of lymphotropic chemokines differed between female and male livers. The recruitment of effector T cell subsets into the inflamed liver is regulated by secretion of chemokine ligands (27, 28). The proinflammatory chemokines CXCL-9-11 are involved in the recruitment of CXCR3+, CD4+, and CD8+ T cells and are known to be expressed by hepatocytes, cholangiocytes, and stellate cells (29, 30). Furthermore, IL-17 has been demonstrated to increase the expression of chemokines, such as CXCL-9, CXCL-11, and CCL-20, by biliary epithelial cells (2931). Therefore, it was possible that the increased IL-17 expression by female lymphocytes could feed chemokine secretion by liver cells, which in turn could lead to the recruitment of lymphocytes into portal tracts (31). In female livers, the expression of CXCL-9 and CXCL-10 was found to be significantly increased as compared with male livers. Differences in the hepatic expression of CCL-20 were less pronounced (Fig. 3F). These results suggested that increased expression of specific chemokines in female livers might stimulate the recruitment of endogenous proinflammatory lymphocytes to the site of biliary inflammation.

Sexual dimorphisms are to a large extent dependent on sex hormones. We therefore investigated whether the sexual differences observed in our cholangitis model depended on the activities of estrogen or testosterone. To that end, the susceptibility for cholangitis was analyzed in castrated male mice without estrogen supplementation. We observed that castration alone increased cholangitis severity to a degree that was similar to female mice (Fig. 4A). Pretreatment of castrated male mice with estrogen could not further enhance the effect observed with castration alone (data not shown). These observations suggested that the higher androgen levels in male mice protect against disease development. To investigate the potential suppressive effect of androgens on portal inflammation, female mice were pretreated with testosterone pellets before the transfer of OT-1 T cells. Testosterone significantly suppressed inflammation in female mice, marked by decreased histologic scores (Fig. 4B, 4C) and decreased numbers of CD4+ liver-infiltrating T cells (Fig. 4D), as compared with female mice without testosterone treatment. Indeed, the portal inflammation of testosterone treated females was as low as that of male recipient mice.

FIGURE 4.

Testosterone suppresses cholangitis and T cell recruitment into the liver. Grading of liver inflammation of (A) castrated male mice and (B) testosterone-treated K14-OVAp female mice was performed using the mHAI score on day 8 after cell transfer (mean ± SD; n = 4–5). (C) Representative H&E staining of liver sections and (D) total number of CD4+ T cells per gram of liver tissue in testosterone-pretreated female mice after the transfer of 4 × 106 OT-1 CD8+ T cells in comparison with hormone-untreated controls (mean ± SD; n = 4–5). All experiments are representative results from three independently performed experiments. *p < 0.05, **p < 0.01.

FIGURE 4.

Testosterone suppresses cholangitis and T cell recruitment into the liver. Grading of liver inflammation of (A) castrated male mice and (B) testosterone-treated K14-OVAp female mice was performed using the mHAI score on day 8 after cell transfer (mean ± SD; n = 4–5). (C) Representative H&E staining of liver sections and (D) total number of CD4+ T cells per gram of liver tissue in testosterone-pretreated female mice after the transfer of 4 × 106 OT-1 CD8+ T cells in comparison with hormone-untreated controls (mean ± SD; n = 4–5). All experiments are representative results from three independently performed experiments. *p < 0.05, **p < 0.01.

Close modal

Because we hypothesized that chemokine expression is key to the sexual dimorphism of this model, we next investigated whether testosterone pretreatment could modulate the hepatic expression of the proinflammatory chemokine ligands, which we have found to be increased in female cholangitis. Indeed, testosterone pretreatment of female mice significantly decreased the hepatic mRNA expression of CXCL-9 and CXCL-10 (Fig. 5A, 5B) to mRNA levels similar to those observed in male livers.

FIGURE 5.

Testosterone suppresses the hepatic expression of CXCL-9 and CXCL-10 and the expression of IL-17 in liver infiltrating lymphocytes. Relative hepatic mRNA expression of (A) CXCL-9 and (B) CXCL-10 from male and female K14-OVAp recipient mice with and without prior testosterone treatment 8 d after transfer of 4 × 106 OT-1 CD8+ T cells (mean ± SD; n = 4). (C) IL-17 levels in supernatants of liver NPCs isolated from testosterone treated K14-OVAp female mice in comparison with controls on day 8 after cell transfer. Cells were restimulated with anti-CD3 and anti-CD28 (2 μg/ml; mean ± SD; n = 4). (D) Relative mRNA expression of RORc in liver tissue from testosterone-pretreated females compared with male and female recipient mice without hormone treatment on day 8 after cell transfer (mean ± SD; n = 4). Each experiment was performed twice independently. *p < 0.05, **p < 0.01.

FIGURE 5.

Testosterone suppresses the hepatic expression of CXCL-9 and CXCL-10 and the expression of IL-17 in liver infiltrating lymphocytes. Relative hepatic mRNA expression of (A) CXCL-9 and (B) CXCL-10 from male and female K14-OVAp recipient mice with and without prior testosterone treatment 8 d after transfer of 4 × 106 OT-1 CD8+ T cells (mean ± SD; n = 4). (C) IL-17 levels in supernatants of liver NPCs isolated from testosterone treated K14-OVAp female mice in comparison with controls on day 8 after cell transfer. Cells were restimulated with anti-CD3 and anti-CD28 (2 μg/ml; mean ± SD; n = 4). (D) Relative mRNA expression of RORc in liver tissue from testosterone-pretreated females compared with male and female recipient mice without hormone treatment on day 8 after cell transfer (mean ± SD; n = 4). Each experiment was performed twice independently. *p < 0.05, **p < 0.01.

Close modal

Accordingly, a significant decrease in IL-17 levels and RORc mRNA expression (Fig. 5C, 5D) was observed in female mice after testosterone treatment.

Our results suggested that hepatic expression levels of CXCL-9 and CXCL-10 and the production of IL-17 by liver lymphocytes were controlled by testosterone. To support this notion further, we analyzed whether serum testosterone levels in untreated wild-type animals correlated with IL-17 production of liver lymphocytes and hepatic chemokine expression. Indeed, liver lymphocytes isolated from untreated female mice produced significantly more IL-17 in response to anti-CD3/anti-CD28 stimulation than lymphocytes from untreated male mice did (Fig. 6A). Moreover, a significant negative correlation was observed between serum testosterone levels and the expression of IL-17 by liver-derived lymphocytes (r = −0.6697; p = 0.0003), as well as the relative hepatic expression of CXCL-9 and CXCL-10 (Fig. 6B–D). To investigate whether testosterone could act on CD4+ T cells directly, we isolated CD4+ T cells from male and female mice and analyzed them for androgen receptor expression. We found that male and female CD4+ T cells similarly express the testosterone receptor (Supplemental Fig. 3). Furthermore, we performed in vitro TH17 differentiation assays in the presence or absence of testosterone. We found that the addition of testosterone enriched serum, which was derived from mice supplemented with testosterone pellets, decreased TH17 differentiation compared with estrogen-enriched serum and FCS serum control (Fig. 6E). Taken together, these results identified testosterone as a novel and important modulator of autoimmune liver inflammation in experimental cholangitis.

FIGURE 6.

Serum testosterone levels negatively correlate to IL-17 production by liver-derived lymphocytes and the hepatic expression of CXCL-9 and CXCL-10 in untreated wild-type mice. (A) Liver NPCs were isolated from healthy C57BL/6 female and male mice and restimulated in vitro with anti-CD3 and anti-CD28. IL-17 levels were determined in culture supernatants by ELISA (mean ± SD; n = 4; one representative result from four independently performed experiments). Serum testosterone levels were correlated with (B) IL-17 levels of restimulated liver lymphocytes and the hepatic expression levels of (C) CXCL-9 and (D) CXCL-10 relative to HPRT. Experiments (B) and (C) were pooled from three independently performed experiments. (E) TH17 in vitro differentiation of CD4 T cells in the presence or absence of testosterone enriched serum. Expression of IL-17 was analyzed using flow cytometry on day 5 (mean ± SD; n = 5; experiment was performed twice independently). **p < 0.01, ***p < 0.001.

FIGURE 6.

Serum testosterone levels negatively correlate to IL-17 production by liver-derived lymphocytes and the hepatic expression of CXCL-9 and CXCL-10 in untreated wild-type mice. (A) Liver NPCs were isolated from healthy C57BL/6 female and male mice and restimulated in vitro with anti-CD3 and anti-CD28. IL-17 levels were determined in culture supernatants by ELISA (mean ± SD; n = 4; one representative result from four independently performed experiments). Serum testosterone levels were correlated with (B) IL-17 levels of restimulated liver lymphocytes and the hepatic expression levels of (C) CXCL-9 and (D) CXCL-10 relative to HPRT. Experiments (B) and (C) were pooled from three independently performed experiments. (E) TH17 in vitro differentiation of CD4 T cells in the presence or absence of testosterone enriched serum. Expression of IL-17 was analyzed using flow cytometry on day 5 (mean ± SD; n = 5; experiment was performed twice independently). **p < 0.01, ***p < 0.001.

Close modal

Sexual differences in autoimmune diseases are poorly understood. We describe an Ag-driven inducible model of cholangitis that features a similar female predominance as observed in human autoimmune liver diseases (2).

Because of the lack of a cholangiocyte-specific promoter, we chose a previously developed model that relies on the expression of an OVA peptide driven by the human K14 promoter, to target the immune response against cholangiocytes. The keratin-14 promoter is expressed in several tissues, including liver, skin, and thymus (14). However, we could detect only mild inflammation in extrahepatic organs (e.g., skin and esophagus) without differences between male and female mice after transfer of Ag-specific CD8+ T cells. Thus, the sexual dimorphism described in this study was specific to the liver.

We aimed to investigate the mechanisms behind the observed sexual differences in liver inflammation. Differences in Treg numbers have recently been described to account for the sex bias observed in a murine model of AIH (22, 32). In our model, Treg numbers were similar between female and male livers. The different modes of disease induction (i.e., hepatitis versus cholangitis) may in part explain these findings. Our data suggest that Ag-specific CD8+ T cells initiate cholangiocyte activation and subsequent recruitment of endogenous CD4+ T cells, which exert their proinflammatory activities at least in part by the production of IL-17. This activity was suppressed by testosterone in a sex-dependent fashion. These findings are in accordance with earlier findings that TH17 cells may play a crucial role in human PBC, PSC, and other inflammatory liver diseases (26, 33), as well as in a murine mouse model of PBC (13). Recently, it has been shown that in the inflamed human liver, IL-17 stimulation of BEC induces the expression of the chemokine ligands CXCL-9-11 and CCL-20 (31). Indeed, we observed a significantly higher expression of the proinflammatory chemokines CXCL-9 and CXCL-10 in the livers of female recipient mice after OVA-specific CD8+ T cell transfer. Thus, enhanced CD4+ T cell recruitment into female livers with CD8+ T cell–induced cholangitis can be driven by a positive feedback loop involving the IL-17–induced upregulation of chemokines by BECs (31). In the original publication of the K14-OVAp model using OT-1/ K14-OVAp double-transgenic mice (14), serum levels of TNF-α and IL-1β were found to be increased. In our transfer model, these cytokines were not increased in male mice after castration (data not shown), which rendered the mice susceptible to cholangitis induction; therefore, our subsequent studies focused on IL-17. However, an additional effect of these cytokines on liver inflammation cannot be excluded.

In castration and hormone supplementation experiments, we could clearly show that the sexual differences observed in our cholangitis model depended on the activities of testosterone. It has previously been described that female sex hormones, such as estrogen, drive autoimmune diseases rather than androgens being suppressive (34). In the model described in this study, testosterone supplementation clearly induced resistance to disease in female mice, and castration alone increased susceptibility of male mice to levels observed in female mice (Fig. 4A, 4B). In support of our findings, a protective effect of testosterone has been previously described in EAE, which is also driven by IL-17. In that model, castration of male mice worsened EAE (35), whereas ovariectomy did not have a major effect on disease (36, 37). In human multiple sclerosis, it has been shown that 24% of affected men present with reduced serum testosterone levels (38). In addition, it was demonstrated in a recent publication that testosterone activity was essential for the protection from islet inflammation in the nonobese diabetic mouse model (39). To our knowledge, testosterone levels in male patients with autoimmune liver diseases have not been studied.

Whereas protection from EAE may be mediated by the cytokines IL-10 and IL-4 (36, 37), there were no differences in the expression of these cytokines in our model (data not shown). In our study, the protective effects of testosterone seemed to be mediated by downregulation of the inflammatory TH17 response; females rendered resistant by testosterone supplementation exhibited decreased IL-17 production of liver-infiltrating lymphocytes (Fig. 5B). Androgen receptors have been identified in CD4+ T cells (40), and androgens have been shown to suppress IL-17 expression in T cells, as demonstrated by an increased expression of IL-17 in male T cells after castration (40). Accordingly, we found that testosterone acts directly on CD4+ T cells by suppressing TH17 differentiation in vitro (Fig. 6). However, the IL-17 response of T cells could also depend on their anatomic location, as female CD4+ T cells isolated from lymph nodes expressed more IL-17 than male cells did after in vitro restimulation, whereas female CD4+ T cells isolated from spleen expressed less IL-17 than male cells did (17). The role of androgens in regulating the IL-17 response in the liver has not been evaluated.

Of interest, the nuclear receptor peroxisome proliferator activated receptor (PPAR)α has been identified as an androgen-sensitive suppressor of NF-κB activation (17). PPARα expression was higher in male than in female T cells, and lack of PPARα increased EAE severity in male, but not female mice, suggesting that an androgen-dependent expression of PPARα protects from T cell–mediated autoimmune disease. Future experiments will have to show to which extent the effects of testosterone in experimental cholangitis are mediated directly via T cells or alternatively via the expression of chemokines or nuclear receptors within parenchymal cells of the liver.

In conclusion, we describe an inducible mouse model of autoimmune cholangitis that, in contrast to previously described cholangitis models, showed a female preponderance similar to human autoimmune liver diseases. We could identify an important immunosuppressive effect of testosterone on Ag-induced cholangitis. These findings should stimulate further research into the mechanisms involved in sexual differences in human autoimmune liver diseases and could lead to novel treatment options for liver inflammation.

We thank Kristin Hogquist for providing K14-OVAp animals and Agnes Malotta, Christina Trabandt, and Marko Hilken for technical assistance.

This work was supported by the Deutsche Forschungsgemeinschaft (SFB 841) and Hamburgische Stiftung für Wissenschaften, Entwicklung und Kultur Helmut und Hannelore Greve.

The online version of this article contains supplemental material.

Abbreviations used in this article:

AIH

autoimmune hepatitis

AMA

anti-mitochondrial Ab

BEC

biliary epithelial cell

EAE

experimental autoimmune encephalomyelitis

mHAI

modified hepatitis activity index

NPC

liver nonparenchymal cell

PBC

primary biliary cirrhosis

PPARα

peroxisome proliferator activated receptor

PSC

primary sclerosing cholangitis

Treg

regulatory T cell.

1
Gleicher
N.
,
Barad
D. H.
.
2007
.
Gender as risk factor for autoimmune diseases.
J. Autoimmun.
28
:
1
6
.
2
Kaplan
M. M.
,
Gershwin
M. E.
.
2005
.
Primary biliary cirrhosis.
N. Engl. J. Med.
353
:
1261
1273
.
3
Mieli-Vergani
G.
,
Vergani
D.
.
2011
.
Autoimmune hepatitis. Nature reviews.
Gastroenterol. Hepatol.
8
:
320
329
.
4
Lleo
A.
,
Battezzati
P. M.
,
Selmi
C.
,
Gershwin
M. E.
,
Podda
M.
.
2008
.
Is autoimmunity a matter of sex?
Autoimmun. Rev.
7
:
626
630
.
5
Cohen
J. H.
,
Danel
L.
,
Cordier
G.
,
Saez
S.
,
Revillard
J. P.
.
1983
.
Sex steroid receptors in peripheral T cells: absence of androgen receptors and restriction of estrogen receptors to OKT8-positive cells.
J. Immunol.
131
:
2767
2771
.
6
McCarthy
M.
2000
.
The “gender gap” in autoimmune disease.
Lancet
356
:
1088
.
7
Beagley
K. W.
,
Gockel
C. M.
.
2003
.
Regulation of innate and adaptive immunity by the female sex hormones oestradiol and progesterone.
FEMS Immunol. Med. Microbiol.
38
:
13
22
.
8
Heneghan
M. A.
,
Norris
S. M.
,
O’Grady
J. G.
,
Harrison
P. M.
,
McFarlane
I. G.
.
2001
.
Management and outcome of pregnancy in autoimmune hepatitis.
Gut
48
:
97
102
.
9
Schramm
C.
,
Herkel
J.
,
Beuers
U.
,
Kanzler
S.
,
Galle
P. R.
,
Lohse
A. W.
.
2006
.
Pregnancy in autoimmune hepatitis: outcome and risk factors.
Am. J. Gastroenterol.
101
:
556
560
.
10
Wellge
B. E.
,
Sterneck
M.
,
Teufel
A.
,
Rust
C.
,
Franke
A.
,
Schreiber
S.
,
Berg
T.
,
Günther
R.
,
Kreisel
W.
,
Zu Eulenburg
C.
, et al
.
2011
.
Pregnancy in primary sclerosing cholangitis.
Gut
60
:
1117
1121
.
11
Bengtsson
A. K.
,
Ryan
E. J.
,
Giordano
D.
,
Magaletti
D. M.
,
Clark
E. A.
.
2004
.
17beta-estradiol (E2) modulates cytokine and chemokine expression in human monocyte-derived dendritic cells.
Blood
104
:
1404
1410
.
12
Gold
S. M.
,
Voskuhl
R. R.
.
2009
.
Estrogen and testosterone therapies in multiple sclerosis.
Prog. Brain Res.
175
:
239
251
.
13
Chuang
Y. H.
,
Ridgway
W. M.
,
Ueno
Y.
,
Gershwin
M. E.
.
2008
.
Animal models of primary biliary cirrhosis.
Clin. Liver Dis.
12
:
333
347
,
ix (ix.)
.
14
McGargill
M. A.
,
Mayerova
D.
,
Stefanski
H. E.
,
Koehn
B.
,
Parke
E. A.
,
Jameson
S. C.
,
Panoskaltsis-Mortari
A.
,
Hogquist
K. A.
.
2002
.
A spontaneous CD8 T cell-dependent autoimmune disease to an antigen expressed under the human keratin 14 promoter.
J. Immunol.
169
:
2141
2147
.
15
Stefanski
H. E.
,
Mayerova
D.
,
Jameson
S. C.
,
Hogquist
K. A.
.
2001
.
A low affinity TCR ligand restores positive selection of CD8+ T cells in vivo.
J. Immunol.
166
:
6602
6607
.
16
Haas
J. D.
,
Ravens
S.
,
Düber
S.
,
Sandrock
I.
,
Oberdörfer
L.
,
Kashani
E.
,
Chennupati
V.
,
Föhse
L.
,
Naumann
R.
,
Weiss
S.
, et al
.
2012
.
Development of interleukin-17-producing γδ T cells is restricted to a functional embryonic wave.
Immunity
37
:
48
59
.
17
Dunn
S. E.
,
Ousman
S. S.
,
Sobel
R. A.
,
Zuniga
L.
,
Baranzini
S. E.
,
Youssef
S.
,
Crowell
A.
,
Loh
J.
,
Oksenberg
J.
,
Steinman
L.
.
2007
.
Peroxisome proliferator-activated receptor (PPAR)alpha expression in T cells mediates gender differences in development of T cell-mediated autoimmunity.
J. Exp. Med.
204
:
321
330
.
18
Harrington
L. E.
,
Hatton
R. D.
,
Mangan
P. R.
,
Turner
H.
,
Murphy
T. L.
,
Murphy
K. M.
,
Weaver
C. T.
.
2005
.
Interleukin 17-producing CD4+ effector T cells develop via a lineage distinct from the T helper type 1 and 2 lineages.
Nat. Immunol.
6
:
1123
1132
.
19
Ishak
K.
,
Baptista
A.
,
Bianchi
L.
,
Callea
F.
,
De Groote
J.
,
Gudat
F.
,
Denk
H.
,
Desmet
V.
,
Korb
G.
,
MacSween
R. N.
, et al
.
1995
.
Histological grading and staging of chronic hepatitis.
J. Hepatol.
22
:
696
699
.
20
Hohenester
S.
,
Beuers
U.
,
Medina
J. F.
,
Elferink
R. P.
.
2013
.
Antimitochondrial antibodies may be insufficiently specific to define primary biliary cirrhosis-like disease in mouse models.
Hepatology
58
:
828
830
.
21
Leung
P. S.
,
Yang
G. X.
,
Dhirapong
A.
,
Tsuneyama
K.
,
Ridgway
W. M.
,
Gershwin
M. E.
.
2012
.
Animal models of primary biliary cirrhosis: materials and methods.
Methods Mol. Biol.
900
:
291
316
.
22
Lapierre
P.
,
Béland
K.
,
Martin
C.
,
Alvarez
F.
 Jr.
,
Alvarez
F.
.
2010
.
Forkhead box p3+ regulatory T cell underlies male resistance to experimental type 2 autoimmune hepatitis.
Hepatology
51
:
1789
1798
.
23
Hammerich
L.
,
Heymann
F.
,
Tacke
F.
.
2011
.
Role of IL-17 and Th17 cells in liver diseases.
Clin. Dev. Immunol.
2011
:
345803
.
24
Lan
R. Y.
,
Salunga
T. L.
,
Tsuneyama
K.
,
Lian
Z. X.
,
Yang
G. X.
,
Hsu
W.
,
Moritoki
Y.
,
Ansari
A. A.
,
Kemper
C.
,
Price
J.
, et al
.
2009
.
Hepatic IL-17 responses in human and murine primary biliary cirrhosis.
J. Autoimmun.
32
:
43
51
.
25
Rong
G.
,
Zhou
Y.
,
Xiong
Y.
,
Zhou
L.
,
Geng
H.
,
Jiang
T.
,
Zhu
Y.
,
Lu
H.
,
Zhang
S.
,
Wang
P.
, et al
.
2009
.
Imbalance between T helper type 17 and T regulatory cells in patients with primary biliary cirrhosis: the serum cytokine profile and peripheral cell population.
Clin. Exp. Immunol.
156
:
217
225
.
26
Katt
J.
,
Schwinge
D.
,
Schoknecht
T.
,
Quaas
A.
,
Sobottka
I.
,
Burandt
E.
,
Becker
C.
,
Neurath
M. F.
,
Lohse
A. W.
,
Herkel
J.
,
Schramm
C.
.
2013
.
Increased T helper type 17 response to pathogen stimulation in patients with primary sclerosing cholangitis.
Hepatology
58
:
1084
1093
.
27
Oo
Y. H.
,
Shetty
S.
,
Adams
D. H.
.
2010
.
The role of chemokines in the recruitment of lymphocytes to the liver.
Dig. Dis.
28
:
31
44
.
28
Rot
A.
,
von Andrian
U. H.
.
2004
.
Chemokines in innate and adaptive host defense: basic chemokinese grammar for immune cells.
Annu. Rev. Immunol.
22
:
891
928
.
29
Cox
M. A.
,
Jenh
C. H.
,
Gonsiorek
W.
,
Fine
J.
,
Narula
S. K.
,
Zavodny
P. J.
,
Hipkin
R. W.
.
2001
.
Human interferon-inducible 10-kDa protein and human interferon-inducible T cell alpha chemoattractant are allotopic ligands for human CXCR3: differential binding to receptor states.
Mol. Pharmacol.
59
:
707
715
.
30
Curbishley
S. M.
,
Eksteen
B.
,
Gladue
R. P.
,
Lalor
P.
,
Adams
D. H.
.
2005
.
CXCR 3 activation promotes lymphocyte transendothelial migration across human hepatic endothelium under fluid flow.
Am. J. Pathol.
167
:
887
899
.
31
Oo
Y. H.
,
Banz
V.
,
Kavanagh
D.
,
Liaskou
E.
,
Withers
D. R.
,
Humphreys
E.
,
Reynolds
G. M.
,
Lee-Turner
L.
,
Kalia
N.
,
Hubscher
S. G.
, et al
.
2012
.
CXCR3-dependent recruitment and CCR6-mediated positioning of Th-17 cells in the inflamed liver.
J. Hepatol.
57
:
1044
1051
.
32
Lapierre
P.
,
Béland
K.
,
Yang
R.
,
Alvarez
F.
.
2013
.
Adoptive transfer of ex vivo expanded regulatory T cells in an autoimmune hepatitis murine model restores peripheral tolerance.
Hepatology
57
:
217
227
.
33
Rowan
A. G.
,
Fletcher
J. M.
,
Ryan
E. J.
,
Moran
B.
,
Hegarty
J. E.
,
O’Farrelly
C.
,
Mills
K. H.
.
2008
.
Hepatitis C virus-specific Th17 cells are suppressed by virus-induced TGF-beta.
J. Immunol.
181
:
4485
4494
.
34
Quintero
O. L.
,
Amador-Patarroyo
M. J.
,
Montoya-Ortiz
G.
,
Rojas-Villarraga
A.
,
Anaya
J. M.
.
2012
.
Autoimmune disease and gender: plausible mechanisms for the female predominance of autoimmunity.
J. Autoimmun.
38
:
J109
J119
.
35
Bebo
B. F.
 Jr.
,
Zelinka-Vincent
E.
,
Adamus
G.
,
Amundson
D.
,
Vandenbark
A. A.
,
Offner
H.
.
1998
.
Gonadal hormones influence the immune response to PLP 139-151 and the clinical course of relapsing experimental autoimmune encephalomyelitis.
J. Neuroimmunol.
84
:
122
130
.
36
Matejuk
A.
,
Adlard
K.
,
Zamora
A.
,
Silverman
M.
,
Vandenbark
A. A.
,
Offner
H.
.
2001
.
17 beta-estradiol inhibits cytokine, chemokine, and chemokine receptor mRNA expression in the central nervous system of female mice with experimental autoimmune encephalomyelitis.
J. Neurosci. Res.
65
:
529
542
.
37
Voskuhl
R. R.
,
Palaszynski
K.
.
2001
.
Sex hormones in experimental autoimmune encephalomyelitis: implications for multiple sclerosis.
Neuroscientist
7
:
258
270
.
38
Wei
T.
,
Lightman
S. L.
.
1997
.
The neuroendocrine axis in patients with multiple sclerosis.
Brain
120
:
1067
1076
.
39
Markle
J. G.
,
Frank
D. N.
,
Mortin-Toth
S.
,
Robertson
C. E.
,
Feazel
L. M.
,
Rolle-Kampczyk
U.
,
von Bergen
M.
,
McCoy
K. D.
,
Macpherson
A. J.
,
Danska
J. S.
.
2013
.
Sex differences in the gut microbiome drive hormone-dependent regulation of autoimmunity.
Science
339
:
1084
1088
.
40
Liva
S. M.
,
Voskuhl
R. R.
.
2001
.
Testosterone acts directly on CD4+ T lymphocytes to increase IL-10 production.
J. Immunol.
167
:
2060
2067
.

The authors have no financial conflicts of interest.

Supplementary data