Mining biomarkers for infertility-associated conditions. Studies on polycystic ovary syndrome and recurrent implantation failure through microbiome and AI-based approaches
Date
2024-08-30
Authors
Journal Title
Journal ISSN
Volume Title
Publisher
Abstract
Naiste viljatuse esinemissagedus on viimastel aastakümnetel suurenenud. Reproduktiivsed häired, nagu polütsüstiline munasarjade sündroom (PCOS) ja korduvad implanteerimise ebaõnnestumised (RIF), on olulised naise viljatuse põhjustajad. Varasemad uuringud on näidatud, et nii PCOS kui ka RIF patsientide endomeetriumi immuunprofiil on muutunud, mis võib aidata kaasa endomeetriumi talitluse häiretele ja viljatuse kujunemisele. Lisaks on nende seisundite puhul täheldatud ka reproduktiivse trakti ja/või soolestiku mikrobioomi muutuseid, mis võivad mõjutada patsiendi üldist tervist ja tema viljakust.
Käesolevas doktoritöös uuriti reproduktiivse trakti mikrobioomi, sealhulgas vaginaalseid ja endomeetriumi proove, et iseloomustada mikroobide kooslust menstruaaltsükli erinevates faasides PCOS-i diagnoosiga naistel ja tuvastada PCOS-iga seotud mikroobide profiil. Seejärel analüüsisime PCOS diagnoosiga naiste soolestiku mikrobioomi ja selle seost patsientidel esinevate meeleoluhäiretega. Viimase teemana arendati välja tehisintellektil (AI) baseeruv mudel, mis aitab anda histoloogilist hinnangut PCOS-i ja RIF diagnoosiga naiste endomeetriumi koe proovidele. AI-mudeliga uuriti menstruaaltsükli faasist sõltuvaid muutusi endomeetriumi näärmete proportsioonides ja põletikulises seisundis, mida iseloomustas CD138 positiivsete plasmarakkude esinemine.
Väitekirjas läbi viidud uuringud näitasid, et endomeetriumi mikrobioomi alfa mitmekesisus oli oluliselt suurem PCOS-iga naistel varases sekretsioonifaasis võrreldes naistega, kellel PCOS-i ei olnud. Lisaks tuvastati kolm bakteritaksonit, mille sagedused olid oluliselt erinevad PCOS-i diagnoosiga naistel. Lisaks varieerus soolestiku mikrobioom PCOS-ga naistel sõltuvalt nendel esinevast meeleoluhäirete diagnoosist ning spetsiifilised bakterid korreleerusid nii PCOS-i kui ka meeleoluhäirete ühiste kliiniliste tunnustega. Viimaks näitas AI analüüs
PCOS naistel muutusi CD138+ rakkude proportsioonis. Samas ei mõjutanud endomeetriumi retseptiivsus RIF patsientidel endomeetriumi koe epiteeli näärmete arengut ega CD138+ rakkude esinemist.
Tulemused sillutavad teed tulevastele uuringutele reproduktiivse trakti mikrobioomi ja viljatuse omavaheliste seoste kohta PCOS ja RIF patsientidel ning rõhutavad soolestiku mikrobioomi muutuste võimalikku rolli PCOS diagnoosiga naistel esinevate meeleoluhäirete tekkes. Lisaks näitavad tulemused, et AI võiks olla kliinilises praktikas kasutuses põletikuliste seisundite tuvastamisel emaka koes ja antud meetod on paljulubav endomeetriumi funktsionaalsete häirete tõhusamateks teadusuuringuteks.
The prevalence of female infertility has increased over the last decades. Reproductive disorders, such as polycystic ovary syndrome (PCOS) and recurrent implantation failure (RIF), are important contributors to female infertility. PCOS and RIF endometria have been shown to have altered immune profiles that may contribute to endometrial dysfunction. Furthermore, changes in the microbiome of the reproductive tract (RT) or the gut have also been observed in these conditions, potentially influencing host health as well as reproductive outcomes. In this thesis, the RT microbiome, including the vagina and the endometrium, was investigated to characterize the landscape of microbial community across the menstrual cycle phases in women with PCOS and to identify microbial signatures associated with PCOS. Next, this thesis explored the gut microbiome among PCOS women who also experienced mood disorders (MDs), aiming to understand its association with both PCOS and MDs. Lastly, an artificial intelligence (AI) model was developed to perform histological assessments on the endometria of women with PCOS and women with RIF. The AI model examined endometrial gland proportions and inflammatory status evidenced as CD138+ plasma cell aggregation. The studies conducted in this thesis revealed that the alpha diversity of the endometrial microbiome was significantly higher in women with PCOS during the early secretory phase compared to women without PCOS. Additionally, three bacterial taxa exhibited significantly different abundances in relation to PCOS. The gut microbiome varied based on MD status in women with PCOS, and specific bacteria correlated with common clinical traits of both PCOS and MDs. Last, the AI analysis showed variations in CD138+ cell percentages based on PCOS phenotypes. On the other hand, endometrial receptivity did not affect either epithelial gland development or CD138+ cell aggregation in RIF patients. These findings pave the way for future research on RT microbiome and adverse reproductive outcomes in PCOS and underline the possible roles of altered gut microbiome on MDs among women with PCOS. Additionally, the results indicate that AI could serve as a tool in clinical practice for screening inflammatory status and hold promise for large-scale sample analysis due to quick and accurate diagnostic potential.
The prevalence of female infertility has increased over the last decades. Reproductive disorders, such as polycystic ovary syndrome (PCOS) and recurrent implantation failure (RIF), are important contributors to female infertility. PCOS and RIF endometria have been shown to have altered immune profiles that may contribute to endometrial dysfunction. Furthermore, changes in the microbiome of the reproductive tract (RT) or the gut have also been observed in these conditions, potentially influencing host health as well as reproductive outcomes. In this thesis, the RT microbiome, including the vagina and the endometrium, was investigated to characterize the landscape of microbial community across the menstrual cycle phases in women with PCOS and to identify microbial signatures associated with PCOS. Next, this thesis explored the gut microbiome among PCOS women who also experienced mood disorders (MDs), aiming to understand its association with both PCOS and MDs. Lastly, an artificial intelligence (AI) model was developed to perform histological assessments on the endometria of women with PCOS and women with RIF. The AI model examined endometrial gland proportions and inflammatory status evidenced as CD138+ plasma cell aggregation. The studies conducted in this thesis revealed that the alpha diversity of the endometrial microbiome was significantly higher in women with PCOS during the early secretory phase compared to women without PCOS. Additionally, three bacterial taxa exhibited significantly different abundances in relation to PCOS. The gut microbiome varied based on MD status in women with PCOS, and specific bacteria correlated with common clinical traits of both PCOS and MDs. Last, the AI analysis showed variations in CD138+ cell percentages based on PCOS phenotypes. On the other hand, endometrial receptivity did not affect either epithelial gland development or CD138+ cell aggregation in RIF patients. These findings pave the way for future research on RT microbiome and adverse reproductive outcomes in PCOS and underline the possible roles of altered gut microbiome on MDs among women with PCOS. Additionally, the results indicate that AI could serve as a tool in clinical practice for screening inflammatory status and hold promise for large-scale sample analysis due to quick and accurate diagnostic potential.
Description
Dissertatsiooni kaitmine toimus Oulu Ülikoolis