BackgroundFemale mice and rats injected with estrogen perinatally become anovulatory and develop follicular cysts. The current consensus is that this adverse response to estrogen involves the hypothalamus and occurs because of an estrogen-induced alteration in the GnRH delivery system. Whether or not this is true has yet to be firmly established. The present study examined an alternate possibility in which anovulation and cyst development occurs through an estrogen-induced disruption in the immune system, achieved through the intermediation of the thymus gland.Methods, Results and ConclusionA putative role for the thymus in estrogen-induced anovulation and follicular cyst formation (a model of PCOS) was examined in female mice by removing the gland prior to estrogen injection. Whereas all intact, female mice injected with 20 ug estrogen at 5–7 days of age had ovaries with follicular cysts, no cysts were observed in animals in which thymectomy at 3 days of age preceded estrogen injection. In fact, after restoring immune function by thymocyte replacement, the majority of thymectomized, estrogen-injected mice had ovaries with corpora lutea. Thus, when estrogen is unable to act on the thymus, ovulation occurs and follicular cysts do not develop. This implicates the thymus in the cysts' genesis and discounts the role of the hypothalamus. Subsequent research established that the disease is transferable by lymphocyte infusion. Transfer took place between 100-day-old estrogen-injected and 15-day-old naïve mice only when recipients were thymectomized at 3 days of age. Thus, a prerequisite for cyst formation is the absence of regulatory T cells. Their absence in donor mice was judged to be the result of an estrogen-induced increase in the thymus' vascular permeability, causing de facto circumvention of the final stages of regulatory T cell development. The human thymus has a similar vulnerability to steroid action during the fetal stage. We propose that in utero exposure to excessive levels of steroids such as estrogen has a long-term effect on the ability of the thymus to produce regulatory T cells. In female offspring this can lead to PCOS.

Day 3 thymectomy (D3Tx) leads to a paucity of CD4+CD25+ suppressor T cells, a loss of peripheral tolerance, and the development of organ-specific autoimmune disease in adult mice. Importantly, D3Tx does not lead to autoimmune disease in all mouse strains, indicating that this process is genetically controlled. Previously, we reported linkage of D3Tx-induced autoimmune ovarian dysgenesis (AOD) and its intermediate phenotypes, antiovarian autoantibody responsiveness, oophoritis, and atrophy, to five quantitative trait loci (QTL), designated Aod1 through Aod5. We also showed interaction between these QTL and H2 as well as Gasa2, a QTL controlling susceptibility to D3Tx-induced autoimmune gastritis. To physically map Aod1, interval-specific bidirectional recombinant congenic strains of mice were generated and studied for susceptibility to D3Tx-induced AOD. Congenic mapping studies revealed that Aod1 controls susceptibility to oophoritis and comprises two linked QTL with opposing allelic effects. Aod1a resides between D16Mit211 (23.3 cM) and D16Mit51 (66.75 cM) on chromosome 16. Aod1b maps proximal of Aod1a between D16Mit89 (20.9 cM) and D16Mit211 (23.3 cM) and includes the candidate genes stefin A1, A2, and A3 (Stfa1-Stfa3), inhibitors of cathepsin S, a cysteine protease required for autoantigen presentation, and the development of autoimmune disease of the salivary and lacrimal glands following D3Tx. cDNA sequencing revealed the existence of structural polymorphisms for both Stfa1 and Stfa2. Given the roles of cathepsins in Ag processing and presentation, Stfa1 and Stfa2 alleles have the potential to control susceptibility to autoimmune disease at the level of both CD4+CD25+ suppressor and CD4+CD25− effector T cells.