Estradiol-mediated enhancement of the human ectocervical epithelial barrier correlates with desmoglein-1 expression in the follicular menstrual phase

Introduction: Structural integrity of the cervical epithelial barrier is crucial for defending the female reproductive tract against sexually transmitted infections, including HIV. Female sex hormones, estradiol and progesterone, play an important role in modulating this epithelial barrier. However, the influence of fluctuating estradiol and progesterone on ectocervical tissue gene and protein expression in naturally cycling women at high risk for sexually transmitted infections is not well understood. Methods: Ectocervical biopsies, cervicovaginal lavage fluid, and venous blood samples were collected from naturally cycling Kenyan female sex workers at two time points, separated by 2 weeks. The first time point represented the luteal phase of the menstrual cycle and the second time point represented the follicular phase. Plasma estradiol and progesterone levels were measured at each time point. Ectocervical tissues were analyzed by RNA-sequencing and in situ immunofluorescence staining. Cervicovaginal lavage samples were evaluated using antibody-based protein profiling. Results: Employing a systems biology approach, we demonstrated that high plasma estradiol levels enhanced ectocervical epithelial integrity. These findings were more pronounced in samples from the follicular phase (when progesterone and estradiol levels were significantly lower) and included increased expression, and more intact distribution, of the desmosomal cadherin DSG1. The effects of progesterone on gene and protein levels, as well as on intact tissue (as visualized by in situ staining), were modest throughout the menstrual cycle. Both estradiol and progesterone levels had limited influence on mucosal immune factors. Conclusion: Estradiol levels were associated with modulation of cervical epithelial barrier integrity, including the expression and distribution of DSG1, during the follicular phase of natural menstrual cycles in Kenyan sex workers. In this study we correlated gene expression with plasma estradiol and progesterone at two time points in the menstrual cycle. Ectocervical tissue were obtained from Kenyan female sex workers, the first sample was obtained during the luteal phase (n=58), and the second sample was obtained during the follicular phase (n=66) of the menstrual cycle (Table 1). RNA used for sequencing was extracted from the ectocervical biopsies. The data from the follicular phase can be accessed at Omnibus public repository, SuperSeries ID GSE217237. The records represent total 113 Luteal samples. The counts-norm-LUT.n=58.csv contains the normalized data for the selected 58 Luteal samples. The counts-norm-FOL.n=66.csv contains the normalized data for the re-analyzed 64 samples from GSE183513: GSM5558523, GSM5558524, GSM5558525, GSM5558526, GSM5558527, GSM5558528, GSM5558529, GSM5558530, GSM5558531, GSM5558532, GSM5558533, GSM5558534, GSM5558535, GSM5558536, GSM5558537, GSM5558538, GSM5558539, GSM5558540, GSM5558541, GSM5558542, GSM5558543, GSM5558544, GSM5558545, GSM5558546, GSM5558547, GSM5558548, GSM5558549, GSM5558550, GSM5558551, GSM5558552, GSM5558553, GSM5558554, GSM5558555, GSM5558556, GSM5558557, GSM5558558, GSM5558559, GSM5558560, GSM5558561, GSM5558562, GSM5558563, GSM5558564, GSM5558565, GSM5558566, GSM5558567, GSM5558568, GSM5558569, GSM5558570, GSM5558571, GSM5558572, GSM5558573, GSM5558574, GSM5558575, GSM5558576, GSM5558577, GSM5558578, GSM5558579, GSM5558580, GSM5558581, GSM5558582, GSM5558583, GSM5558584, GSM5558585, GSM5558586 and two (GSM8436015, GSM8436014) samples from the GSE194276 records *** Submitters state that the raw sequencing data from the study participants cannot be held in a public repository due to the sensitive nature of such personal data***