Spatiotemporal changes in mechanical matrisome components of the human ovary from prepuberty to menopause.

How do elastic matrisome components change during the lifetime of the human ovary? The deposition and remodeling of mechanical matrisome components (collagen, elastin, elastin microfibril interface-located protein 1 (EMILIN-1), fibrillin-1 and glycosaminoglycans (GAGs)) that play key roles in signaling pathways related to follicle activation and development evolve in an age- and follicle stage-related manner. The mechanobiology of the human ovary and dynamic reciprocity that exists between ovarian cells and their microenvironment is of high importance. Indeed, while the localization of primordial follicles in the collagen-rich ovarian cortex offers a rigid physical environment that supports follicle architecture and probably plays a role in their survival, ovarian extracellular matrix (ECM) stiffness limits follicle expansion and hence oocyte maturation, maintaining follicles in their quiescent state. As growing follicles migrate to the medulla of the ovary, they encounter a softer, more pliant ECM, allowing expansion and development. Thus, changes in the rigidity of the ovarian ECM have a direct effect on follicle behavior. Evidence supporting a role for the physical environment in follicle activation was provided in clinical practice by ovarian tissue fragmentation, which promoted actin polymerization and disrupted ovarian Hippo signaling, leading to increased expression of downstream growth factors, promotion of follicle growth and generation of mature oocytes. We investigated quantitative spatiotemporal changes in collagen, elastin, EMILIN-1, fibrillin-1 and GAGs from prepuberty to menopause, before conducting a closer analysis of the ECM surrounding follicles, from primordial to secondary stages, in both prepubertal and tissue from women of reproductive age. The study included ovarian tissue (cortex) from 68 patients of different ages: prepubertal (n = 16; mean age [±SD]=8 ± 2 years); reproductive (n = 21; mean age [±SD]=27 ± 4 years); menopausal with estrogen-based HRT (n = 7; mean age [±SD]=58 ± 4 years); and menopausal without HRT (n = 24; mean age [±SD]=61 ± 5 years). Quantitative investigations of collagen and GAG deposition in ovarian tissue throughout a woman's lifetime were conducted by analyzing brightfield images. Characteristic features of collagen fiber content were based on polarized light microscopy, since polarized light changes with fiber thickness. To evaluate the deposition and distribution of elastin, fibrillin-1 and EMILIN-1, multiplex immunofluorescence was used on at least three sections from each patient. Image processing and tailored bioinformatic analysis were applied to enable spatiotemporal quantitative evaluation of elastic system component deposition in the human ovary over its lifetime. While collagen levels increased with age, fibrillin-1 and EMILIN-1 declined. Interestingly, collagen and elastin reached their peak in reproductive-age women compared to prepubertal (P < 0.01; P = 0.262) and menopausal subjects with (P = 0.706; P < 0.01) and without (P = 0.987; P = 0.610) HRT, indicating a positive impact of secreted estrogen and hormone treatment on collagen and elastin preservation. Interestingly, HRT appears to affect elastin presence in ovarian tissue, since a significantly higher (P < 0.05) proportion of elastin was detected in biopsies from menopausal women taking HRT compared to those not. Higher GAG levels were found in adult ovaries compared to prepubertal ovaries (P < 0.05), suggesting changes in tissue ultrastructure and elasticity with age. In this context, elevated GAG values are suspected to participate in hampering formation of the fibrillin-1 network (r = -0.2475; P = 0.04687), which explains its decline over time. This decline partially accounts for the decrease in EMILIN-1 (r = 0.4149; P = 0.00059). Closer examination of the ECM surrounding follicles from the primordial to the secondary stage, both before and after puberty, points to high levels of mechanical stress placed on prepubertal follicles compared to the more compliant ECM around reproductive-age follicles, as suggested by the higher collagen levels and lower elastin content detected mainly around primordial (P < 0.0001; P < 0.0001, respectively) and primary (P < 0.0001; P < 0.001, respectively) follicles. Such a stiff niche is nonpermissive to prepubertal follicle activation and growth, and is more inclined to quiescence. Not applicable. The duration and form of administered HRT were not considered when studying the menopausal patient group undergoing treatment. Moreover, we cannot exclude interference from other nongynecological medications taken by the study patients on ovarian ECM properties since there is no information in the literature describing the impact of each medication on the ECM. Finally, since the ECM is by definition a very heterogeneous meshwork of proteins, the use of two-dimensional histology could be a limitation. Single time points on fixed tissues could also present limitations, since following ovary dynamics from prepuberty to menopause in the same patient is not feasible. From a biomechanical perspective, our study revealed important changes to ECM properties dictating the mechanical features of ovarian tissue, in line with the existing literature. Our findings pave the way for possible therapeutic targets at the ECM level in the context of female fertility and ovarian rejuvenation, such as mechanical stimulation, antifibrotic treatments, and prevention or reversion of elastic ECM degradation. Our study also sheds light on the follicle-specific ECM composition that is dependent on follicle stage and age. These data will prove very useful in designing biomimetic scaffolds and tissue-engineered models like the artificial ovary. Indeed, they emphasize the importance of encapsulating each type of isolated follicle in an appropriate biomaterial that must replicate the corresponding functional perifollicular ECM and respect ovarian tissue heterogeneity in order to guarantee its biomimicry. This study was supported by grants from the Fonds National de la Recherche Scientifique de Belgique (FNRS) (C.A.A. is an FRS-FNRS research associate; grant 5/4/150/5 awarded to M.M.D.) and the Université Catholique de Louvain (PhD grant 'Coopération au développement' awarded to E.O.). None of the authors have any competing interests to declare.