Skip to main content

Analysis of the longitudinal stability of human plasma miRNAs and implications for disease biomarkers.

Sandau US, Wiedrick JT, McFarland TJ, Galasko DR, Fanning Z, Quinn JF, Saugstad JA

Publication: Sci Rep (2024) • Vol 14 • Page 2148PMID: 38272952View on PubMed

Purpose

This paper compared levels of microRNAs (miRNA, miR) among specimens collected over a 3 month period to investigate intra-individual variance and evaluate the impact of hemolysis, tobacco use, stress-level, and sleep quality.

Conclusion

In total, 36 miRNAs were observed in the plasma of all participants on each visit. Spikes in the quantification cycle (Cq) values of these 36 miRNAs were not related to hemolysis scores. Cq values of the spike-in control (cel-miR-39-3p) were strongly correlated with Cq values of the 36 miRNAs (r=0.89) and Cq values for the endogenous control (miR-16-5p) (r=0.92), indicating the variance is likely attributable to variability in extraction or real-time PCR efficiency. Consequently, the authors chose to normalize expression to cel-miR-39-3p to account for technical variability and miR-16-5p to account for between-subject variability. Subsequent normalization for batch effects had little effect on Cq values. Of the 134 miRNAs that passed filtering, 74 had a retest standard deviation <1.0, including 11 that were previously identified as markers of Alzheimer’s Disease. Hemolysis led to increased variability in more than half of the 134 miRNAs, including 27 that increased >1.5-fold per unit increase in hemolysis score and 4 that decreased >1.5-fold for each unit increase in hemolysis score. Recent tobacco use affected (>1.5-fold change) the levels of 22 miRNAs, but none of the miRNAs evaluated were affected by fasting status or change in reported stress levels or sleep quality. Importantly, tobacco use was also associated with increased test-retest variability. There were 3 miRNAs that showed >1.5-fold variance in retest specimens from males, but no miRNAs showed >1.5-fold variance in retest specimens from females.

Studies (1)

Purpose

This study investigated intra-individual variance of and the impacts of hemolysis, tobacco use, stress level, and sleep quality on levels of miRNAs over a 3-month period. Blood was collected biweekly for 3 months from 22 healthy volunteers (15 females, 7 males) into BD Vacutainer K2EDTA tubes that were stored for 30 min to 1 h at room temperature before plasma separation. Plasma was obtained by centrifugation at 450 g for 10 min followed by two subsequent centrifugations at 2500 g for 15 min and frozen at -80°C. RNA was isolated using the miRNeasy Serum/Plasma Advanced Kit; the miRNA cel-miR-39-3p was added following lysis. miRNA was quantified using the Qubit miRNA Assay Kit. miRNAs were quantified by real-time PCR using the TaqMan Advanced miRNA Human A Cards (375 miRNAs, and controls). Hemolysis was assessed using the difference in Cq values between miR-23a-3p (not affected by hemolysis) and miR-451a (hemolytic marker).

Summary of Findings

In total, 36 miRNAs were observed in the plasma of all participants on each visit. Spikes in the Cq values of these 36 miRNAs were not related to hemolysis scores. Cq values of the spike-in control (cel-miR-39-3p) (r=0.89) were strongly correlated with the Cq values of the 36 miRNAs evaluated (r=0.89) and Cq values for the endogenous control (miR-16-5p) (r=0.92), indicating the variance is likely attributable to variability in extraction or real-time PCR efficiency. Consequently, the authors chose to normalize expression to cel-miR-39-3p to account for technical variability and miR-16-5p to account for between subject-variability. Subsequent normalization for batch effects had little effect on Cq values. Of the 134 miRNAs that passed filtering, 74 had a retest standard deviation <1.0, including 11 that were previously identified as markers of Alzheimer’s Disease. Importantly, these 74 miRNAs showed low longitudinal drift. Hemolysis led to increased variability in more than half of the 134 miRNAs, including 27 that increased >1.5-fold per unit increase in hemolysis score and 4 that decreased >1.5- fold for each unit increase in hemolysis score. Recent tobacco use affected (>1.5-fold change) the levels of 22 miRNAs, but none of the miRNAs were affected by fasting status or a change in reported stress levels or sleep quality. Importantly, tobacco use was also associated with increased test-retest variability. There were 3 miRNAs that showed >1.5-fold variance in retest specimens from males, but no miRNAs showed >1.5-fold variance in retest specimens from females.

  • Biospecimens

    FluidBlood
    FluidPlasma
  • Preservation Types

    Frozen
  • Diagnoses

    Normal
  • Platforms

    RNAReal-time qRT-PCR
  • Pre-analytical Factors

    ClassificationPre-Analytical FactorValue(s)
    PreaquisitionDiagnosis/ patient condition
    Fasting
    Multiple levels of self-reported sleep quality
    Multiple levels of self-reported stress
    Non-fasting
    PreaquisitionPatient gender
    Female
    Male
    Biospecimen AcquisitionTime of biospecimen collection
    Fasting
    No recent tobacco use
    Not fasting
    Recent tobacco use
    Biospecimen Aliquots and ComponentsHemolysis
    A range of hemolysis scores investigated