摘要Background:Ovarian follicular fluid influences follicle and oocyte growth,but the fluctuation of its protein content during folliculogenesis has not been comprehensively analyzed.Here we used a shotgun approach and bioinformatics analyses to investigate and compare the proteomes of porcine follicular fluid (pFF) obtained from small (＜ 4 mm),medium (4-6 mm) and large (＞ 6-12 mm) follicles.Results:Follicular fluid samples containing highest estrogen levels were selected as non-atretic from small (SNA:26.1 ± 15 ng/mL),medium (MNA:162 ± 54 ng/mL),and large (LNA:290 ± 37 ng/mL) follicles for proteomic analyses.We detected 1627,1699,and 1756 proteins in SNA,MNA,and LNA samples,respectively.Nearly 60-63％ of total proteins were specific to each sample,11-13％ were shared in pairwise comparisons,and 247 proteins were shared among all samples.Functional categorization indicated comparable gene ontology (GO) terms distribution per cellular component,molecular function,and biological process categories across samples;however,the ranking of highly significantly enriched GO terms per category revealed differences between samples.The patterns of protein-to-protein interactions varied throughout follicle development,and proteins such as serine protease inhibitor,clade E (SERPINE);plasminogen activator,urokinase (PLAU);and plasminogen activator,urokinase receptor (PLAUR) appeared stage-specific to SNA,MNA,and LNA,respectively.The "complement and coagulation cascades" was the common major pathway.Besides,properdin and fibulin-1 were abundant proteins that appeared absent in LNA samples.Conclusion:This study provides extensive and functional analyses of the pFF proteome changes during folliculogenesis and offers the potential for novel biomarker discovery in pFF for oocyte quality assessment.