Mechanisms responsible for ‘scarless’ tissue repair in the endometrium

Abstract

The endometrium is the inner lining of the uterine cavity, composed of distinct epithelial and stromal cell compartments with the latter containing fibroblasts, a vascular compartment and fluctuating populations of immune cells. The endometrium is a highly dynamic tissue that undergoes cycles of proliferation and stromal cell differentiation (decidualisation) followed by tissue shedding (menstruation) and rapid repair/remodelling, all under control of fluctuating concentrations of steroid hormones secreted from the ovaries. This is known as the menstrual cycle. In response to the ‘injury’ inflicted as a consequence of decidual breakdown and shedding, the endometrium exhibits a unique capacity to restore tissue architecture by rapid tissue repair. This repair process is tightly regulated to ensure that the endometrium heals consistently every month throughout a woman’s reproductive lifespan, without the accumulation of fibrotic scar tissue which could have a negative impact on fertility. The initiation of menstruation is triggered by the withdrawal of progesterone as a consequence of the demise of the corpus luteum within the ovaries which precipitates an increase in production of inflammatory mediators, focal hypoxia and activation of matrix metalloproteinase enzymes culminating in endometrial shedding. In contrast the cellular and molecular mechanisms responsible for the rapid and scar-free tissue repair of endometrium remain poorly understood. Parallels can be drawn between the repair process of the endometrium and that of the foetal skin and oral mucosa which also exhibit ‘scarless’ healing, including rapid reepithelialisation, widespread cellular proliferation and migration and a short time to wound closure. However endometrial repair also shares key features of the wound healing experienced by adult tissues that exhibit scarring including extensive angiogenesis and a substantial inflammatory response. The endometrium appears to be unique, fitting in a gap between tissues that typically undergo ‘scarless’ or ‘scarring’ tissue repair. In women, endometrial shedding is considered an inflammatory event and the culmination of a cascade of inflammatory signals result in the accumulation of a diverse population of immune cells within the tissue. Whilst we believe immune cells play a key role in regulating spatial and temporal tissue breakdown and shedding their role in repair and restoration of tissue homeostasis remains poorly understood. One process essential for endometrial repair is restoration of the luminal epithelial cell layer (re-epithelialisation) and imaging studies have demonstrated that this appears not only to be rapid but also to occur synchronously with tissue degeneration and shedding. Re-epithelialisation was previously thought to be governed by proliferation and migration of glandular epithelial cells in the basal (unshed) tissue compartment, however new data suggest a role for trans-differentiation of stromal cells into epithelial cells which merits further investigation. In addition to role(s) for immune and stromal cells in regulation of endometrial tissue function, a role for somatic stem/progenitor cells capable of differentiating into mature endometrial cells to regenerate the tissue has also been claimed. In summary, whilst progress has been made in understanding the processes governing endometrial decidualisation, breakdown and shedding the regulation and roles of the different cell types that participate in scar-free repair of the tissue remain poorly defined. The studies in this thesis set out to address this gap by addressing three key aims: Aim 1. To investigate the phenotype and location of immune cell populations during scarless tissue repair. Aim 2. To identify and characterise a putative population of mesenchymal stem/progenitor cells in endometrium. Aim 3. To investigate the contribution of putative mesenchymal stem/progenitor cells to endometrial tissue repair. The aims were addressed using a recently refined and extensively characterised mouse model in which endometrial shedding (‘menstruation’) and repair occurs over a 48 hour period following removal of a progesterone stimulus. Importantly the Saunders’ group have already demonstrated that this model recapitulates the key features of human menses including overt vaginal bleeding, immune cell influx, tissue necrosis, transient hypoxia, re-epithelialisation and most importantly simultaneous breakdown and repair. Uterine tissues recovered 12, 24, and 48hrs after removal of progesterone and were investigated using immunohistochemistry (spatial organisation), flow cytometry (quantitation of cell subpopulations), FACS sorting (isolation of subpopulations) and molecular profiling (qPCR, RNAseq and single cell sequencing) with additional insights from bioinformatic analysis. To address Aim 1 endometrial shedding and repair was studied in Macgreen® mice: in this transgenic line all the cells of the mononuclear phagocyte lineage (monocytes, monocyte-derived macrophages) express green fluorescent protein. Immunohistochemistry revealed striking spatio-temporal changes in both numbers and location of GFP+ cells during endometrial breakdown and repair, the most prominent changes occurring 24hrs after removal of progesterone. Flow Cytometry quantified several immune cell populations with a significant increase in GFP+ cells during repair, the majority of which were GR1+F4/80- (inflammatory monocytes). These novel data provided compelling evidence to support a role for inflammatory monocytes in endometrial repair and provide the platform for future studies on the role of these cells in scarless healing. To address Aims 2 and 3 Pdgfrβ-BAC-eGFP® transgenic mice in which GFP was expressed under control of promoter elements of the Pdgfrβ gene was used to identify putative mesenchymal progenitor cells and investigate their role in endometrial repair. GFP+ cells were located exclusively within the endometrial stromal compartment and examination of tissue sections revealed that two subpopulations could be distinguished based on the both the intensity of GFP and expression of CD146 (Mcam). Characterisation by immunohistochemistry, flow cytometry and qPCR identified a GFPbright subpopulation located adjacent to CD31+ endothelial cells that were classified as pericytes based on location and phenotype (NG2+, CD146+, CD31-). When menstruation was stimulated in Pdgfrβ-BAC-eGFP® mice detailed analysis using flow cytometry revealed an increase in the perivascular pericyte subpopulation during active healing (24hrs) and also identified a new previously unidentified subpopulation of GFP+ cells which had a unique phenotype during repair. Evidence that GFP+ cells contribute to restoration of epithelial repair was obtained with an increase in expression of the epithelial cell marker EpCAM and GFP+ cells in the renewed epithelial cell layer. RNAseq and single cell sequencing combined with bioinformatics complemented these findings by identifying novel changes in gene expression in both endometrial fibroblasts and pericyte populations consistent with induction of novel pathways and trans-differentiation of stromal cells by a mesenchymal-to-epithelial transition (MET). In conclusion, using a mouse model of endometrial breakdown and repair a heterogeneous population of myeloid cells and a putative population of endometrial progenitors (pericytes) have been characterised, quantified and novel changes in gene expression identified. Adaptation of these cell types to the insult of endometrial shedding appears to play a key role in temporal and spatial regulation of rapid, scar-free endometrial tissue repair. These novel findings may inform the development of new approaches to treating gynaecological disorders associated with aberrant endometrial repair such as heavy menstrual bleeding, Asherman’s syndrome and endometriosis as well as other disorders associated with excessive fibrosis and scar formation.