Expression of miR-99 family microRNAs and the clustered miR-125b in the subepithelial stroma of endometriotic lesions (#102)
Although the pathogenesis of endometriosis remains poorly understood, Sampson’s theory of retrograde menstruation prevails as the favoured hypothesis. Mouse models of endometriosis demonstrate that infiltration of host-derived cells of haematopoietic and mesenchymal origin occurs, and may play an important role in lesion pathogenesis. Our laboratory previously demonstrated that all three members of the human miR-99 family of microRNAs – miR-99a, miR-99b, and miR-100 – are upregulated in ectopic lesions compared to eutopic endometrium in patients with endometriosis. Furthermore, miR-125a and miR-125b, which are clustered with the miR-99 family genes, are also upregulated in ectopic tissue. Emerging evidence suggests that these related / clustered miRNAs are associated with haematopoietic progenitor cells. These genes may also be expressed in the recently-characterised endometrial stem cell. We hypothesised that miR-99 and miR-125 family genes are located in the stromal compartment of eutopic and ectopic tissues, where stem/progenitor cells are likely to be present. We localised miR-99a, miR-100, and miR-125b in a panel of paired eutopic and ectopic tissues from women with endometriosis (n=4 proliferative phase, n=4 secretory phase) using Locked Nucleic Acid – In Situ Hybridisation (LNA-ISH). We demonstrated that miR-99a, miR-100 and miR-125b are expressed in eutopic endometrial stromal cells, and absent from epithelial cells. In ectopic lesions, expression is confined predominantly to sub-epithelial stromal cells. A similar localisation pattern was observed for all three miRNAs, but the relative abundance of each transcript clearly differed (miR-99a < miR-100 << miR-125b). There was no obvious difference between the proliferative and secretory phase. Immunohistochemical analysis of serial sections revealed that miR-99a, miR-100, and miR-125b localisation is distinct from regions of alpha smooth muscle actin expression, suggesting that the cells expressing these miRNA are not (myo)fibroblasts. However, we observed partial co-localisation with CD10 and CD68, markers of endometrial stromal cells and macrophages, respectively, suggesting an endometrial or haematopoietic origin for these cells. Modifying the behaviour of infiltrating stromal cells by manipulating miRNA expression may provide a therapeutic avenue to prevent glandular remodelling and facilitate endometriotic lesion resolution.