Cellular and molecular characterisation of pelvic ligaments fibroblasts : effects of gonadal steroids
The high incidence of pelvic organ prolapse (POP) observed in postmenopausal women and in those used levormeloxifene suggested an aetiological role for the hypoestrogenic state. Further, the oestrogen-related molecular and structural changes in hip and knee ligaments, regulation of oestrogen action by extracellular matrix (ECM) proteins, and the high expression of steroid receptors in the mechanically failed connective tissue were previously reported. These observations led to the hypotheses of changed ECM composition with compensatory increase in steroid receptors expression in the prolapsed cardinal ligaments. A series of immunohistochemical studies proved the assumption that the prolapsed cardinal ligaments would have higher expression of collagen III (p=0.001), tenascin (p=0.001), oestrogen receptor α (p=0.09), androgen receptor (p=0.004), progesterone receptor (p=0.03), and lower expression of elastin (p=0.004) when compared to non-prolapsed ligaments, irrespective of menopausal status. Unpredictably, collagen I expression was directly related to the menopausal status rather than prolapse, and the prolapsed ligaments showed lower expression of oestrogen receptor β (p=0.02). HRT was assumed to rectify the prolapse-related changes in postmenopausal women, but it had only partial ameliorating effect by reducing collagen III (p=0.001) and androgen receptor (p=0.06) expression in the prolapsed ligaments to levels similar to the normal ones. Given the immunohistochemical findings, it was hypothesised that a number of genes representing the altered ECM proteins and gonadal steroids receptors would be modified. The stretch model was used assuming that the ffect of long duration mechanical stretch on primary cultures of fibroblasts derived from cardinal ligaments in vitro would be similar to the effect of chronic stretch associated with prolapse in vivo. cDNA microarray identified genes coding for regulation of actin remodelling, but it fell short of identifying alterations in gene expression commensurate with the findings of immunohistochemistry. This might be attributed to the differences between the in vivo and in vitro environment and/or the design of the stretch model used in this experiment. The cDNA microarray findings led to the hypothesis that mechanical strain e.g. in cases of increased intra-abdominal pressure and levormeloxifene could cause POP by destroying the cytoskeleton of the fibroblasts. To test this hypothesis, the effect of mechanical stretch and levormeloxifene on the morphology of the cytoskeleton of those fibroblasts grown in primary cultures was studied using fluroprobe technology. The effect of 17β-oestradiol was also investigated assuming that it would prevent and/or reverse the effect of levormeloxifene and stretch. Stretch caused major phenotypic alterations in actin morphology (p=0.0001), and levormeloxifene caused similar changes in the static fibroblasts (p=0.0001). Nonetheless, the use of oestradiol did not protect the cytoskeleton, but significantly increased cell proliferation (p=0.02), which was reduced by stretch (p=0.001), suggesting a beneficial role in the healing process. By submitting this thesis and the publication of this set of articles, it is hoped that the area of prolapse will be opened up for further objective assessment of cell-matrix interactions, and additional opportunities for creative exploration will be catalysed.