Heat shock protein HSP90 immunoexpression in equine endometrium during oestrus, dioestrus and anoestrus

Abstract

Heat shock proteins play a crucial role in cellular development, proliferation, differentiation and apoptosis. Heat shock protein 90 (HSP90) has been localised in the human endometrium, where its immunoexpression changes during the menstrual cycle. Similar studies have not been done for the equid species, so the present study aimed to describe endometrial HSP90 immunoexpression in mare endometrium. Endometrial biopsies were formalin‐fixed and paraffin‐embedded, and sections were stained with haematoxylin–eosin in preparation for HSP90 immunohistochemistry. Immunostaining and morphometric analyses were performed on the epithelial lining, endometrial glands and connective stroma during oestrus, dioestrus phase and anoestrus period (n = 7 per phase or period). Immunoexpression was localised in the basal region of the epithelial cells lining the lumen. Immunoexpression was greater during oestrus than during either dioestrus or anoestrus. During anoestrus, there was little immunostaining in the endometrium, suggesting that HSP90 is involved in the functional modulation of sex steroid receptors in cyclic mares. Indeed, the function of HSP90 as a chaperone in the folding of proteins, such as steroid receptors, might explain the greater intensity of immunostaining during the oestrus and dioestrus phases, compared the anoestrus period. We conclude that, in the mare, HSP90 plays a role in endometrial function and that further studies are needed to test whether it is important in pathological conditions as endometritis.

Dynamic changes in AI-based analysis of endometrial cellular composition: Analysis of PCOS and RIF endometrium

“…Previous protocols for the histological assessment of the endometrium employed additional staining with the p16 antibody to differentiate epithelial and stromal cell compartments by selectively staining only epithelial cells 20 , 21 and using traditional image analysis software, which requires pre-processing steps, such as converting original images into grayscale or binary images for detecting staining intensity. 20 , 22 These measurements were limited to specifically chosen tissue areas, 23 , 24 and intra- and inter-observer variability was not considered. 23 In contrast, here we showed that our AI model, which does not require additional staining or image conversion processes, analyzes a whole scanned slide within 1 min while producing highly accurate and reproducible results.…”

“… 20 , 22 These measurements were limited to specifically chosen tissue areas, 23 , 24 and intra- and inter-observer variability was not considered. 23 In contrast, here we showed that our AI model, which does not require additional staining or image conversion processes, analyzes a whole scanned slide within 1 min while producing highly accurate and reproducible results. The AI algorithm, the functionality and reliability of which were validated in our previous study, 14 coincided with the pathologists’ assessment.…”