Background
MicroRNAs (miRNAs) have been linked to several diseases and to regulation of almost every biological process. This together with their stability while freely circulating in blood suggests that they could serve as minimal-invasive biomarkers for a wide range of diseases. Successful miRNA-based biomarker discovery in plasma is dependent on controlling sources of preanalytical variation, such as cellular contamination and hemolysis, as they can be major causes of altered miRNA expression levels. Analysis of plasma quality is therefore a crucial step for the best output when searching for novel miRNA biomarkers.
Conclusions
Our findings suggest that good quality plasma samples suitable for miRNA profiling can be achieved from samples collected and stored by large biobanks. Incorporation of extensive quality control measures, such as those established here, would be beneficial for future projects. The overall low correlation of miRNA expression between cord and maternal samples is an interesting observation, and promising for our future studies on identification of miRNA-based biomarkers in cord blood plasma, considering that these samples were collected at term and some exchange of blood components between cord and maternal blood frequently occur.
Methods
Plasma quality was assessed by three different methods in samples from mother-child duos (maternal and cord blood, N = 2x38), with collection and storage methods comparable to large cohort study biobanks. Total RNA was isolated and the expression profiles of 201 miRNAs was obtained by qPCR to identify differentially expressed miRNAs in cord and maternal plasma samples.
Results
All three methods for quality assurance indicate that the plasma samples used in this study are of high quality with very low levels of contamination, suitable for analysis of circulating miRNAs. We identified 19 significantly differentially expressed miRNAs between cord and maternal plasma samples (paired t-tests, FDR<0.05, and fold change>±1.5), and we observed low correlation of miRNA transcript levels between cord and maternal samples throughout our dataset. Conclusions: Our findings suggest that good quality plasma samples suitable for miRNA profiling can be achieved from samples collected and stored by large biobanks. Incorporation of extensive quality control measures, such as those established here, would be beneficial for future projects. The overall low correlation of miRNA expression between cord and maternal samples is an interesting observation, and promising for our future studies on identification of miRNA-based biomarkers in cord blood plasma, considering that these samples were collected at term and some exchange of blood components between cord and maternal blood frequently occur.