Testosterone affects female CD4+ T cells in healthy individuals and autoimmune liver diseases

Abstract

Autoimmune hepatitis (AIH) and primary biliary cholangitis (PBC) are autoimmune liver diseases with strong female predominance. They are caused by T cell-mediated injury of hepatic parenchymal cells, but the mechanisms underlying this sex bias are unknown. Here, we investigated whether testosterone contributes to T cell activation in women with PBC. Compared with sex- and age-matched healthy controls (n = 23), cisgender (cis) women with PBC (n = 24) demonstrated decreased testosterone serum levels and proinflammatory CD4+ T cell profile in peripheral blood. Testosterone suppressed the expression of TNF and IFN-γ by human CD4+ T cells in vitro. In trans men receiving gender-affirming hormone therapy (GAHT) (n = 25), testosterone affected CD4+ T cell function by inhibiting Th1 and Th17 differentiation and by supporting the differentiation into regulatory Treg. Mechanistically, we provide evidence for a direct effect of testosterone on T cells using mice with T cell-specific deletion of the cytosolic androgen receptor. Supporting a role for testosterone in autoimmune liver disease, we observed an improved disease course and profound changes in T cell states in a trans man with AIH/primary sclerosing cholangitis (PSC) variant syndrome receiving GAHT. We here report a direct effect of testosterone on CD4+ T cells that may contribute to future personalized treatment strategies.

Keywords: Autoimmune diseases; Autoimmunity; Hepatology; Immunology; Sex hormones; T cells.

Figure 1. Immune profile of cis women with PBC.

(A and B) Serum cytokine levels determined by flow cytometry–based LEGENDplex analysis from cis women with PBC (n = 24) compared with age- and sex-matched healthy controls (n = 23). (C) Frequencies of CD3⁺, CD4⁺ CD8⁺, Th2, Th1, Th17, and Treg from cis women with PBC (n = 17–21) compared with age- and sex-matched healthy controls (n = 21–22) were analyzed by flow cytometry. (D) Th1 differentiation rate was increased in CD4⁺ T cells derived from cis women with PBC (n = 16) compared with age- and sex-matched healthy controls (n = 21). (E) Serum testosterone levels of cis women with PBC (n = 27) compared with age- and sex-matched healthy controls (n = 23) determined by LC-MS/MS analysis. (F) Correlation analysis of serum testosterone levels and serum cytokine profiles determined using flow cytometry–based LEGENDplex analysis from cis women with PBC (n = 27). (G) Decreased concentrations of IFN-γ and TNF in supernatant of T cell receptor–stimulated T cells derived from healthy cis women in the presence of testosterone. Cytokine secretion was analyzed in supernatants using ELISA, and statistical significance was determined via 1-way ANOVA (G). Statistical analysis was performed using Mann-Whitney U test (B–E). *P ≤ 0.05; **P ≤ 0.01; ***P ≤ 0.001, ****P ≤ 0.0001.

Figure 2. Gender-affirming hormone therapy (GAHT) with a high dose of testosterone alters CD4⁺ T cell phenotype in trans men.

(A) Graphical abstract of the cohort included. (B) Serum testosterone levels of trans men receiving GAHT at baseline (BL) before and 6 months (6M) after therapy start with testosterone (n = 25). (C) Blood parameters from trans men at 6M of GAHT normalized to intrapersonal baseline values (n = 25). (D) UMAP resembling 40,930 peripheral blood T cells of trans men at BL and the 6M time point (n = 4). Cells were subdivided into 15 clusters by Seurat: 6 clusters predominantly expressed CD8⁺ T cell markers (γδ T cells, γδ T cells_CYTOTOXIC, MAIT, CD8⁺ T_NAIVE, CD8⁺ Temra, and CD8⁺ Tem) and 9 clusters expressed CD4⁺ T cell markers (CD4⁺ T_NAIVE, CD4⁺ T_{NAIVE RTE}, CD4⁺ T_{NAIVE ACTIVATED}, CD4⁺ Tcm, CD4⁺ Treg, CD4⁺ Treg _ACTIVATED, CD4⁺ Tem, CD4⁺ Th17_POL, and CD4⁺ Th2_POL). (E) Heatmap showing signature differentially expressed genes (DEG) of each cluster. (F) Fate probability analysis of CD4⁺_NAIVE T cells toward different endpoints is displayed. After 6 months of GAHT (dark blue), the fate probability of naive T cells was increased toward the Treg endpoint and decreased toward Tcm compared with naive CD4⁺ T cells before therapy (light blue). Data are shown as mean ± SD. Statistical analysis was performed using 1-way ANOVA test. ****P ≤ 0.0001.

Figure 3. Shift toward antiinflammatory immune cell subsets in trans men receiving GAHT.

(A) Heatmap of intraindividual serum cytokine changes in trans men (n = 25) 6M after GAHT normalized to baseline before therapy start. Serum cytokine levels were determined by flow cytometry–based LEGENDplex analysis. (B) Heatmap of T cell populations identified as significantly different by flow cytometry–based immunophenotyping 6M after therapy start (n = 22). Red indicates an increase and blue indicates a decrease compared with baseline samples. (C) Frequencies of CD3⁺, CD4⁺, CD8⁺, Th17, Th1, Th2, and Treg from trans men at BL, 3M, and 6M time points analyzed by flow cytometry. (D) Shift in T cell phenotypes toward antiinflammatory immune cell subsets. Data are shown as mean ± SD. Statistical analysis was performed using a 1-way ANOVA with correction for multiple comparisons. *P ≤ 0.05; **P ≤ 0.01.

Figure 4. Testosterone directly acts on murine T cells via the cytosolic AR.

(A) Schematic of breeding strategy for mice lacking AR signaling in mature T cells. (B and C) qRNA expression and protein expression of AR in CD4⁺ T cells isolated from AR^fl/fl Lck^Cre mice (green stroke pattern) compared with WT littermates (green). (D) Immune cell phenotyping in AR^fl/fl Lck^Cre female mice (green stroke pattern) compared with WT female littermates (green) (n = 5). (E and F) In vitro proliferation assay and in vitro differentiation assay of CD4⁺ T cells isolated from AR^fl/fl Lck^Cre female mice (green stroke pattern) compared with WT female littermates (green). Representative figures of 3 independent experiments and mean ± SD are shown. Statistical analysis was performed using Mann-Whitney U test. *P ≤ 0.05; **P ≤ 0.01.

Figure 5. Gender-affirming hormone therapy improves autoimmune liver disease in a trans man.

Single-case study of a trans man with AIH/PSC variant syndrome receiving GAHT. (A) Serum testosterone levels measured using LC-MS/MS analysis. (B) Clinical parameters determined by serum transaminases and IgG levels before and during testosterone therapy. (C) TNF- and IFN-γ–expressing CD4⁺ T cells in blood before and after 6M of GAHT. (D) UMAP reflecting 17,959 peripheral blood CD3⁺ T cells at baseline (BL) and 6M after GAHT showing 9 distinct T cell clusters (5 CD4⁺ and 4 predominantly CD8⁺). (E) Volcano plots of differentially expressed genes (DEGs) of total CD4⁺ and CD4⁺ T_NAIVE cells. (F) Violin plots of the gene expression of CD69, NFKB1, NFKBIA, JUNB, FOS, and DUSP1 at BL and 6M in the CD4⁺ T_NAIVE cluster. (G) Overrepresentation analysis (ORA) of significantly changed pathways according to the Hallmark database in CD4⁺ and CD4⁺ T_NAIVE cells. Indicated P values were corrected for multiple comparisons.