Global protein profiling of human milk using pre-enriched RNA-sequence libraries

The complex protein composition of human milk remains challenging to characterize due to technical limitations, yet such insights are crucial as early-life nutrition influences long-term health outcomes and is associated with pathological conditions such as obesity. In this study, we applied APTASHAPE, a high-throughput profiling method based on chemically modified protein-binding RNA molecules, to globally profile protein composition in skimmed human milk and to explore associations with maternal and infant characteristics. A total of 520 human milk samples collected at 3 days, 1 month, 2 months, and 3 months postpartum were analyzed. Ordinary Least Squares regression analysis of the discovery cohort identified discriminatory RNA sequence profiles reflecting the global protein composition and correlating with sampling time, maternal BMI, and parity. These associations were subsequently validated and confirmed in an independent test cohort. Furthermore, 14 candidate aptamers were subjected to protein pull-down assays followed by mass spectrometry, which identified C4b-binding protein and tenascin C as candidate targets associated with maternal BMI and sampling time. Overall, our findings highlight the dynamic nature of the human milk protein composition, with the greatest variation attributable to sampling time, while maternal BMI and parity accounted for more subtle differences.