A human counterpart to the oocyte autoimmunity that develops in the neonatally thymectomized mouse model of autoimmune primary ovarian insufficiency (hypergonadotropic hypogonadism)

Abstract

OBJECTIVE: Neonatal thymectomy (NTX) performed on B6A mice leads to autoimmunity against oocyte proteins and total destruction of the ovary leaving a fibrous band of tissue. We test the hypothesis that the oocyte autoimmunity (OA) that develops in the NTX model of autoimmune primary ovarian insufficiency (POI) has a human counterpart (OA-POI). We also test the hypothesis that women with OA-POI have a distinct ovarian phenotype as compared to women who have steroidogenic cell autoimmunity as the mechanism of POI (SCA-POI).DESIGN: Cross-sectional, controlled, IRB approved.MATERIALS AND METHODS: We developed an ELISA to detect antibodies against hrMATER, an oocyte protein that we previously demonstrated to be a target of autoimmunity in the NTX mouse model. We defined normal based on findings in control women with regular menstrual cycles (N=56). We tested 222 consecutive women with 46,XX spontaneous POI for OA-POI (abnormal level of antibodies against hrMATER) and SCA-POI (abnormal levels of antibodies detected by indirect immunofluorescence against monkey adrenal or by immunoabsorption assay using I-125 hr 21-hydroxylase). We calculated ovarian volume using transvaginal ultrasound. Analysis: Kruskal-Wallis test, SAS version 9.0 (Cary, NC).RESULTS: OA-POI (anti-MATER antibodies) and SCA-POI (anti-steroidogenic cell antibodies) were detected in 10.4% (23/222) and 5.9% (13/222) of women with POI, respectively. Three women were positive for both antibodies. Women exclusively with OA-POI had smaller mean (SD) ovarian volume (mm3) than women exclusively with SCA-POI [4.4 (4.4) vs 9.4 (7.9), p<0.02]. As compared to women with idiopathic POI (without OA or SCA), women with OA-POI had similar ovarian volumes [4.4 (4.4) vs 4.7 (5.8), p=0.64] and were more likely to be positive for anti-thyroglobin antibodies (5/20, 25% vs. 17/189, 9%, p=0.03). There were no differences in the demographics, menstrual history, endocrine profile or other autoimmune antibodies between the exclusive OA-POI and SCA-POI groups.CONCLUSIONS: There is a human counterpart to the oocyte autoimmunity that develops in the NTX model of autoimmune POI. Furthermore, the ovarian phenotype of OA-POI is distinct from SCA-POI. A report of MATER autoimmunity associated with the hypogonadism that develops in autoimmune polyglandular failure type 1 (NEJM 358:1018, 2008) supports these findings. OBJECTIVE: Neonatal thymectomy (NTX) performed on B6A mice leads to autoimmunity against oocyte proteins and total destruction of the ovary leaving a fibrous band of tissue. We test the hypothesis that the oocyte autoimmunity (OA) that develops in the NTX model of autoimmune primary ovarian insufficiency (POI) has a human counterpart (OA-POI). We also test the hypothesis that women with OA-POI have a distinct ovarian phenotype as compared to women who have steroidogenic cell autoimmunity as the mechanism of POI (SCA-POI). DESIGN: Cross-sectional, controlled, IRB approved. MATERIALS AND METHODS: We developed an ELISA to detect antibodies against hrMATER, an oocyte protein that we previously demonstrated to be a target of autoimmunity in the NTX mouse model. We defined normal based on findings in control women with regular menstrual cycles (N=56). We tested 222 consecutive women with 46,XX spontaneous POI for OA-POI (abnormal level of antibodies against hrMATER) and SCA-POI (abnormal levels of antibodies detected by indirect immunofluorescence against monkey adrenal or by immunoabsorption assay using I-125 hr 21-hydroxylase). We calculated ovarian volume using transvaginal ultrasound. Analysis: Kruskal-Wallis test, SAS version 9.0 (Cary, NC). RESULTS: OA-POI (anti-MATER antibodies) and SCA-POI (anti-steroidogenic cell antibodies) were detected in 10.4% (23/222) and 5.9% (13/222) of women with POI, respectively. Three women were positive for both antibodies. Women exclusively with OA-POI had smaller mean (SD) ovarian volume (mm3) than women exclusively with SCA-POI [4.4 (4.4) vs 9.4 (7.9), p<0.02]. As compared to women with idiopathic POI (without OA or SCA), women with OA-POI had similar ovarian volumes [4.4 (4.4) vs 4.7 (5.8), p=0.64] and were more likely to be positive for anti-thyroglobin antibodies (5/20, 25% vs. 17/189, 9%, p=0.03). There were no differences in the demographics, menstrual history, endocrine profile or other autoimmune antibodies between the exclusive OA-POI and SCA-POI groups. CONCLUSIONS: There is a human counterpart to the oocyte autoimmunity that develops in the NTX model of autoimmune POI. Furthermore, the ovarian phenotype of OA-POI is distinct from SCA-POI. A report of MATER autoimmunity associated with the hypogonadism that develops in autoimmune polyglandular failure type 1 (NEJM 358:1018, 2008) supports these findings.