Uterine fibroids are conventionally defined as clonally derived benign tumours from the proliferation of a single smooth muscle cell (SMC). We have previously identified fibroblast-like cells in fibroids, the presence of which raises the question as to whether all cells within the fibroid have the same clonal origin. The first aim of this study was to develop a fluorescence-activated cell sorting (FACS)-based method to isolate different cell types from human myometrium and fibroid tissues. Secondly, we aimed to use X chromosome inactivation analysis to determine the clonality of cell subpopulations isolated from myometrial and fibroid tissues. Human myometrium and fibroid tissues were collected from women undergoing hysterectomy. Immunohistochemistry (IHC) and flow cytometry confirmed that in addition to SMCs, fibroblasts constitute a significant proportion of cells in human myometrium and fibroid tissues. FACS based on CD90 and ALDH1 reliably separated cells into three myometrial and four fibroid subpopulations: SMCs, vascular smooth muscle cells and two fibroblast subsets. Clonality was first determined by X chromosome inactivation using the classic DNA methylation-sensitive HUMARA assay. Data from this assay were highly variable, with only a quarter of samples meeting the definition of clonal fibroid and non-clonal myometrium. However, using an RNA-based X chromosome inactivation HUMARA assay, we were able to demonstrate clonality of all cellular constituents of most fibroids. Our results confirm that most fibroids are derived from a single cell, and for the first time demonstrates that these clonal cells differentiate into fibroblast and SMC subpopulation as the fibroid grows.

Gene expression varies markedly across the menstrual cycle and expression levels for many genes are under genetic control. We analyzed gene expression and mapped expression quantitative trait loci (eQTLs) in endometrial tissue samples from 229 women and then analyzed the overlap of endometrial eQTL signals with genomic regions associated with endometriosis and other reproductive traits. We observed a total of 45,923 cis-eQTLs for 417 unique genes and 2,968 trans-eQTLs affecting 82 unique genes. Two eQTLs were located in known risk regions for endometriosis including LINC00339 on chromosome 1 and VEZT on chromosome 12 and there was evidence for eQTLs that may be target genes in genomic regions associated with other reproductive diseases. Dynamic changes in expression of individual genes across cycle include alterations in both mean expression and transcriptional silencing. Significant effects of cycle stage on mean expression levels were observed for (2,427/15,262) probes with detectable expression in at least 90% of samples and for (2,877/9,626) probes expressed in some, but not all samples. Pathway analysis supports similar biological control of both altered expression levels and transcriptional silencing. Taken together, these data identify strong genetic effects on genes with diverse functions in human endometrium and provide a platform for better understanding genetic effects on endometrial-related pathologies.

Introduction: There is a well-established inverse relationship between body mass index and frequency of endometriosis. However, these population-based studies have relied mostly on self-reported cases of endometriosis, rather than surgically confirmed endometriosis where disease severity has been objectively assessed. The aim of the current retrospective study was to establish whether the established relationship between endometriosis and low body mass index was independent of disease severity. Methods: Women with menstrual and/or pelvic pain undergoing laparoscopy for suspected endometriosis were recruited for this retrospective study (n = 509). Women were grouped by body mass index (kg/m²) according to World Health Organization criteria: underweight (<18.5), normal (18.5–24.99), pre-obese (25–29.99) or obese (≥30). Endometriosis was scored according to the revised American Fertility Society system. Data were analysed based on body mass index and endometriosis status to identify any relationship between body mass index and disease. Results: The average body mass index of women with endometriosis was 25.0 kg/m². The body mass index distribution of women with endometriosis differed relative to women in the general population. As expected, fewer obese women had endometriosis than in the lower body mass index categories. However, the obese women who did have endometriosis had significantly higher revised American Fertility Society scores compared to women with normal and pre-obese body mass indices. Discussion: Our results are consistent with the established finding of an inverse relationship between body mass index and endometriosis. The novel finding from this study is that obesity is associated with increased disease severity and reduced frequency of stage I endometriosis. It remains unclear what role body mass index has in the cause or effect of endometriosis; we speculate that body mass index may be useful for sub-classifying the disease.