Gestational age-associated differences in amniotic fluid extracellular vesicle-enriched preparations during...

Protein profiles of amniotic fluid extracellular vesicle-enriched preparations may provide information on the fetal and gestational intrauterine environment, but their normal variation remains incompletely characterized. We used data-independent acquisition-based quantitative proteomics to compare AF-EV-enriched preparations sampled at approximately 19 and 24 weeks from pregnancies with normal fetal cytogenetic findings and uncomplicated outcomes. Of 3321 quantified proteins, 2611 detected in at least three of four samples in each group were included in the primary analysis without imputation. Primary analysis identified 904 candidates. Of these, 698 were also supported by the KNN sensitivity analysis with concordant direction, including 385 higher at 19 W and 313 higher at 24 W. A separate complete-case analysis identified 596 candidates, all within the 698-protein set. Nominal over-representation analysis using the 2611 eligible protein groups as the background highlighted metabolic/digestion and apical/brush-border annotations among 19 W-higher candidates and proteasome-centered annotations among 24 W-higher candidates. These enrichment findings are exploratory because the displayed P values were not adjusted for multiple comparisons. Given the small cohort and absence of independent validation and comprehensive EV characterization, the results are hypothesis-generating and cannot establish EV-specific gestational remodeling. SIGNIFICANCE: Gestational age is a recognized source of variation in total amniotic fluid, but its influence on proteins recovered from AF-EV-enriched preparations during mid-pregnancy is less well defined. Using DIA proteomics, we compared preparations sampled at approximately 19 and 24 weeks from pregnancies with normal fetal cytogenetic findings and uncomplicated outcomes. Of 3321 quantified proteins, 698 were supported by both the primary analysis and KNN sensitivity analysis and were retained as hypothesis-generating candidates. Nominal over-representation analysis using the 2611 protein groups eligible for the primary analysis as the background highlighted metabolic/digestion and apical/brush-border annotations among 19 W-higher candidates and proteasome-centered annotations among 24 W-higher candidates. These unadjusted enrichment results are descriptive and do not establish EV-specific remodeling. The findings nevertheless indicate that gestational age should be considered when designing and interpreting AF-EV proteomic studies and provide candidates for validation in larger independent cohorts.