Perinatal androgens organize sex differences in mast cells and attenuate anaphylaxis severity into adulthood

Abstract

Mast cell (MC)-associated diseases, including allergy/anaphylaxis and neuroinflammatory pain disorders, exhibit a sex bias, with females at increase risk. While much attention has been directed toward adult sex hormones as drivers of sex differences, that female sex bias in MC-associated diseases is evident in prepubertal children, suggesting early-life origins of sex differences which have yet to be explored. Utilizing rodent models of MC-mediated anaphylaxis, our data here reveal that, 1) compared with females, males exhibit significantly reduced severity of MC-mediated anaphylactic responses that emerge prior to puberty and persist into adulthood, 2) reduced severity of MC-mediated anaphylaxis in males is linked with the naturally high level of perinatal androgens and can be recapitulated in females by perinatal exposure to testosterone proprionate, 3) perinatal androgen exposure guides bone marrow MC progenitors toward a masculinized tissue MC phenotype characterized by decreased concentration of prestored MC granule mediators (e.g., histamine, serotonin, and proteases) and reduced mediator release upon degranulation, and 4) engraftment of MC-deficient Kit ^W-sh/W-sh mice with adult male, female, or perinatally androgenized female MCs results in MC-mediated anaphylaxis response that reflects the MC sex and not host sex. Together, these data present evidence that sex differences in MC phenotype and resulting disease severity are established in early life by perinatal androgens. Thus, factors affecting levels of perinatal androgens could have a significant impact on MC development and MC-associated disease risk across the life span.

Keywords: histamine; immune; inflammation; mast cell; sex differences.

Fig. 1.

MC-mediated anaphylaxis is sexually dimorphic without the presence of adult gonadal hormones. Male and female mice were GDX or SHAM at 12 wk of age. Mice (13 wk to 14 wk old) were sensitized with DNP-specific IgE (10 μg i.p.) and challenged the following day with DNP-HSA (500 μg i.p.; n = 5 to 7 per group). (A) Male and female GDX mice had sexually dimorphic hypothermia responses that were not different from male and female SHAM mice. Repeated measures two-way ANOVA with Tukey’s multiple comparisons test. (B) Peak temperature change was greater in female mice regardless of group. Two-way ANOVA with Sidak’s multiple comparisons test. (C) Serum histamine levels 30 min after anaphylaxis were greater in female mice, regardless of GDX status, compared with male cohorts. No difference in serum histamine levels was found between GDX and SHAM groups within their sex. Two-way ANOVA with Sidak’s multiple comparisons test. (D) Gonadectomizing did not affect the number of MCs in mesenteric tissue MCs in male or female mice, and MC number was not different between the sexes. Two-way ANOVA with Sidak’s multiple comparisons test. (E) Gonadectomizing had no overall effect on tissue MC degranulation measured as the percentage of degranulated MCs 30 min after induction of anaphylaxis. Two-way ANOVA with Sidak’s multiple comparisons test. MCs were counted in 10 randomly chosen fields (200× magnification) per mouse. Unpaired Student’s t test. Data represent mean ± SEM; ns: nonsignificant; ****P < 0.0001.

Fig. 2.

Prepubertal mice exhibit sexually dimorphic MC-mediated anaphylaxis. Male and female prepubertal mice (14 d of age) were sensitized with DNP-specific IgE (10 μg i.p.) and challenged the following day with DNP-HSA (500 μg i.p., n = 8 to 10 per sex) or PBS (n = 6 to 8 per sex). (A) Prepubertal male and female mice exhibited a decrease in body temperature in response to PSA. Prepubertal female mice had more severe hypothermia responses starting at 15 min and lasting for the remainder of the experiment. RM two-way ANOVA with Sidak’s multiple comparisons test. (B) Prepubertal female mice had greater drop in temperature after DNP injection compared to prepubertal male mice. Two-way ANOVA with Sidak’s multiple comparisons test. (C) PSA increased serum histamine levels in both prepubertal male and female mice. Prepubertal female mice exhibited higher serum histamine levels 15 min after DNP or PBS injection compared to prepubertal males. Two-way ANOVA with Sidak’s multiple comparisons test. (D) DNP injection increased the percentage of degranulated MCs to a similar level in male and female prepubertal mice; n = 7 or 8 per sex. No difference in MC degranulation was observed between the sexes in IgE-PBS treated groups; n = 5 or 6 per sex. (E) MC number per high powered field (hpf) in intestinal mesenteric windows was similar in prepubertal male and female mice; n = 12 or 13 per sex; unpaired Student’s t test. Experimental animals within each treatment were derived from three independent animal cohorts and dams to establish reproducibility; data represent means ± SEM, *P < 0.05, **P < 0.01, ***P < 0.001, ****P < 0.0001.

Fig. 3.

Perinatal exposure to the antiandrogen, DEHP in males increases the severity of MC-mediated anaphylaxis into adulthood. (A) Schematic of DHEP exposure. Created with BioRender.com. Male (8 to 10 wk old) perinatally DEHP-treated or CO-treated mice were sensitized with DNP-specific IgE (5 μg i.p.) and challenged the following day with DNP-HSA (500 μg i.p., n = 6 to 8) or PBS (n = 3 or 4 per group). (B) In response to IgE-DNP, male mice exposed to perinatal DEHP had greater serum histamine levels compared to CM; two-way ANOVA with Sidak’s multiple comparisons test. (C) Male mice exposed to DEHP perinatally had more severe hypothermia responses compared to controls in response to PSA. RM two-way ANOVA with Dunnett’s multiple comparisons test. (D) Peak temperature change was greatest after high-dose perinatal DEHP exposure. One-way ANOVA with Dunnett’s multiple comparisons test. (E) Perinatal DEHP exposure did not change intestinal mesenteric MC number; one-way ANOVA with Tukey’s multiple comparisons test. (F) PSA increased MC degranulation to similar levels in all groups; two-way ANOVA with Sidak’s multiple comparisons test. Data represent means ± SEM. Experimental animals within each treatment were derived from six dams and conducted in two independent cohorts to establish reproducibility; #P < 0.10, *P < 0.05, **P < 0.01, ***P < 0.001, ****P < 0.0001.

Fig. 4.

Perinatal androgen exposure in females masculinizes MC-mediated anaphylactic responses into adulthood. (A) Schematic of perinatal androgenization protocol created with BioRender.com. Adult (8 wk old) male and female mice that were perinatally TP-treated or SO-treated were sensitized with DNP-specific IgE (10 μg i.p.) and challenged the following day with DNP-HSA (500 μg i.p., n = 6 to 9 per group) or PBS (n = 3 per sex). (B) Control (CF) mice had more severe hypothermia responses compared to AF mice. CM and AF mice had similar hypothermia responses. Androgenized male (AM) mice had reduced hypothermia compared to all other groups. RM two-way ANOVA with Tukey’s multiple comparisons test. (C) Peak temperature change was greatest in CF mice. One-way ANOVA with Tukey’s multiple comparisons test. (D) Serum histamine levels were increased in all groups 30 min after DNP injection. Perinatal testosterone exposure greatly reduced histamine levels in female mice to the level of CM. Two-way ANOVA with Sidak’s multiple comparisons test. (E) Serum MCPT-1 levels were increased in all groups 30 min after DNP injection. Serum MCPT-1 levels were below the limit of detection in mice treated with PBS. Perinatal androgenization greatly reduced MCPT-1 levels in female mice to the level of CM. (F) One-way ANOVA with Tukey’s multiple comparisons test. MC number in intestinal mesenteric windows was not different between groups; n = 5 to 7 per group; one-way ANOVA with Tukey’s multiple comparisons test. (G) The percentage of mesenteric MCs that exhibited a degranulated appearance was increased following induction of PSA in intestinal mesentery, but there were no differences between control and androgenized mice. Two-way ANOVA with Sidak’s multiple comparisons test. (H) Meningeal MC number was similar between groups; n = 7 to 10 per group. One-way ANOVA with Tukey’s multiple comparisons test. (I) PSA increased MC degranulation in the meninges to similar levels in all groups. Two-way ANOVA with Sidak’s multiple comparisons test. Data represent means ± SEM. Experimental animals within each treatment were derived from at least seven dams and studies conducted in at least three independent cohorts to establish reproducibility. *P < 0.05, **P < 0.01, ***P < 0.001, ****P < 0.0001.

Fig. 5.

Perinatal androgen exposure programs bone marrow MC progenitors for reduced MC granule mediator concentrations in adulthood. (A) Schematic depicting derivation of BMMCs from femoral bone marrow of 7- to 8-wk-old GDX CM, CF, and AF mice. Created with BioRender.com. (B) Total histamine concentration was reduced in perinatally AF BMMCs compared to CF BMMCs. (C) Hdc gene expression in BMMCs. One-way ANOVA with Tukey’s multiple comparisons test. (D) BMMCs were sensitized with monoclonal anti-DNP IgE (0.5 µg/mL) overnight and later stimulated with 0, 15, 31, and 62 ng/mL DNP-HSA for 1 h. Supernatants from AF BMMCs had reduced concentrations of histamine after FcεRI stimulus compared with CF BMMCs, and had similar responses to CM BMMCs, with the exception of 15 ng/mL DNP stimulus. Two-way ANOVA with Tukey’s multiple comparisons test. (E) BMMCs were stained with fluorescent-conjugated c-kit (Pacific Blue) and FcεRI (PE) antibodies, and representative images of two-color flow cytometric analysis of FcεRI (horizontal axis) and c-kit (vertical axis) receptor expression from CM, CF, and AF adult BMMCs are shown. Flow cytometry geometric means (bar graphs) for c-kit, FcεRI, and c-kit/FcεRI double positive cell percentages showed no difference between BMMC groups. One-way ANOVA with Tukey’s multiple comparisons test. (F) Androgenized BMMCS exhibited reduced levels of MC proteases MCPT4 and CPA (Western blot analysis), and 5HT (ELISA) in BMMC pellets. (G) Relative mRNA expression for Mcpt4, Cpa3, and Tph in BMMC pellets was not different across experimental groups. One-way ANOVA with Tukey’s multiple comparisons test. Data are representative of experiments from five or six independent BMMC cultures in each experimental group to establish reproducibility. #P < 0.10, *P < 0.05, **P < 0.01, ***P < 0.001, ****P < 0.0001.

Fig. 6.

MC-deficient Kit^W-sh/W-sh mice engrafted with BMMCs from adult male, female, and perinatally AF exhibited serum histamine and hypothermia responses that represented the BMMC phenotype and not host sex. (A) Schematic depicting engraftment protocol. Created with BioRender.com. Adult female MC-deficient Kit^W-sh/W-sh mice were engrafted with BMMCs derived from adult GDX male, female, or perinatally AF mice (i.p., 1 × 10⁷ BMMCs). Adult male MC-deficient Kit^W-sh/W-sh mice were engrafted with BMMCs derived from adult GDX male or female mice (i.p., 1 × 10⁷ BMMCs). Eight to twelve weeks after engraftment, mice were exposed to PSA for 30 min. (B) Female Kit^W-sh/W-sh mice engrafted with female BMMCs had more severe hypothermic responses to PSA compared to male and perinatally AF BMMCs. Engrafted mice exposed to IgE-PBS treatment had no change in temperature; n = 7 to 10 per group; RM two-way ANOVA with Dunnett’s multiple comparisons test. (C) Female Kit^W-sh/W-sh mice engrafted with female BMMCs had greatest change in peak temperature after PSA. One-way ANOVA with Dunnett’s multiple comparisons test. (D) Kit^W-sh/W-sh mice engrafted with BMMCs had increased serum histamine levels after DNP injection compared to PBS injection. Female Kit^W-sh/W-sh mice engrafted with female BMMCs had greater serum histamine levels after PSA compared to female Kit^W-sh/W-sh mice engrafted with AF BMMCs and female Kit^W-sh/W-sh mice engrafted with CM BMMCs; IgE-PBS; n = 3 or 4 per group, IgE-DNP; n = 6 to 8 per group; two-way ANOVA with Dunnett’s multiple comparisons test. (E) MC degranulation percentage was similar in all groups at baseline and after PSA; IgE-PBS; n = 4 or 5 per group, IgE-DNP; n = 5 or 6 per group; two-way ANOVA with Dunnett’s multiple comparisons test. (F) intestinal mesenteric MC numbers were similar in female Kit^W-sh/W-sh mice engrafted with BMMCs; n = 8 to 11 per group; one-way ANOVA with Tukey’s multiple comparisons test. *P < 0.05, **P < 0.01, ****P < 0.0001.

Fig. 7.

A model for the role of perinatal androgen levels in sexual dimorphism of MC phenotype and disease severity throughout the lifespan. (A) In the presence of high perinatal androgen levels (e.g., normal perinatal androgen surge in males or androgenization of females with TP), bone marrow MC progenitors develop into tissue MCs that have decreased storage of preformed granule mediators (e.g., histamine, proteases, serotonin), resulting in lower amounts of released mediators upon activation and decreased the severity of anaphylaxis from neonatal period into adulthood. (B) In the absence of high perinatal androgens (e.g., biological female or exposure to antiandrogen compounds such as phthalates [DHEP] in males), bone marrow MCs develop into tissue MCs that have increased concentrations of MC granule mediator, resulting in higher amounts of released mediators upon activation thus enhancing the severity of anaphylaxis. Created with BioRender.com.