Skip to main content

Microarray-Based Identification of Time-Sensitive miRNAs as Serum Quality Indicators in Liquid Biopsy.

Takizawa S, Mizoguchi A, Sudo H, Nakadera Y, Akiyama H, Kondou S

Publication: Genes to cells : devoted to molecular & cellular mechanisms (2026) • Vol 31 • Page e70122PMID: 42246163View on PubMed

Purpose

This paper compared microRNA (miRNA, miR) levels in case-matched serum from blood stored for 0.5, 1 or 6 h pre-centrifugation, and from serum stored for 0, 0.5 and 1 h post-centrifugation. Levels of the affected miRNAs were then compared between clinical specimens with a 2 h (1 h pre- and 1 h post- centrifugation) and a 6 h (unspecified breakdown) processing time; two affected miRNAs were then selected for analysis in extracellular vesicles.

Conclusion

Levels of all miRNAs evaluated in serum remained within 5% of the baseline control (centrifuged after 30 min and immediately frozen) when blood was stored for 1 h before centrifugation, but levels of miR-486-5p and miR- increased significantly when blood was stored for 6 h before centrifugation. Levels of 14 miRNAs increased and 3 decreased by more than 5% when serum was stored for 0.5 h post-centrifugation prior to transfer and freezing compared to controls that were immediately transferred and frozen. Of these, 4 miRNAs displayed significant changes when serum was stored for 0.5 h post-centrifugation (miR-92a-3p, miR-486-5p, miR-1470, and miR-4787-3p). When serum was stored for 1 h post-centrifugation, significant (q<0.05) changes were found in 12 miRNAs (miR-371b-5p, miR-885-3p, miR-940, miR-1260a, miR-1260b, miR-1470, miR-4429, miR-4787-3p, miR-6511-5p, miR-6792-5p, miR-6872-3p, and miR-6880-3p) compared to controls. While storage of serum (post-centrifugation) generally had similar effects on specimens from three different volunteers, effects on miR-4435 and miR-486-5p levels were not observed in specimens from all volunteers. Nevertheless, in a mixed-effects model only miR-4435 was no longer significant (P>0.05). In an independent validation set, miR-4429 was 1.03 fold and miR-6511b-5p 1.49 fold higher in specimens with a 2 h processing time (1 h pre-centrifugation, 1 h post-centrifugation) than those with a 6 h processing time (unspecified pre- versus post- centrifugation).  Neither miR-4429 nor miR-6511b-5 were detected in EVs.

Studies (1)

Purpose

This study compared microRNA (miRNA, miR) levels in case-matched serum from blood stored for 0.5, 1 or 6 h pre-centrifugation, and from serum stored for 0, 0.5 and 1 h post-centrifugation. Levels of the affected miRNAs were then compared between clinical specimens with a 2 h (1 h pre- and 1 h post- centrifugation) and 6 h (unspecified breakdown) processing time and of these 2 affected miRNAs were selected for analysis in extracellular vesicles.  Blood was collected from nine healthy volunteers and allowed to clot at room temperature for 30 min (control), 1 h or 6 h (experimental) before serum separation. Serum was separated by centrifugation at 25°C at 2,300 g for 10 min. Unless otherwise specified, serum was removed immediately and stored at -80°C. To investigate the effects of serum storage, some tubes were left at room temperature for 0, 0.5 and 1 h after centrifugation but before removal, freezing, and frozen storage of serum. EVs were isolated by ultracentrifugation (10,000 g at 4°C for 0.5 h followed by two 4°C centrifugations at 230,000 g for 70 min), diluted in phosphate buffered saline and stored at -80°C. RNA was isolated from blood, EVs and serum using the 3D-Gene RNA extraction reagent. Levels of 2,588 miRNAs were quantified using 3D-Gene Human microRNA Oligo Chip. Variability of the assay was evaluated using promptly processed specimens from 3 healthy volunteers. For validation, 150 serum specimens from blood centrifuged after 1 h and then stored for 1 h, and 60 specimens frozen within 6 h of collection (storage details not clear) were compared.

Summary of Findings

Based on plots of the log2 signal and coefficient of variance (CV), the authors determined that a signal of 6 or higher would have a CV <5%; thus, miRNAs with a signal >6 could be analyzed without further correction between measurements. In baseline specimens, 445 miRNAs in serum and 411 miRNAs in blood had a signal >6 and were included in subsequent analysis. Level of all miRNAs in serum remained within 5% of baseline (centrifuged after 30 min) when blood was stored for 1 h before centrifugation. Serum from blood that experienced a 6 h delay before centrifugation had higher levels of miR-486-5p and miR-92a-3p than serum isolated from specimens centrifuged after 30 min (1.52-fold, q < 0.05 and 1.17-fold, q < 0.05, respectively). Levels of miR-371b-5p, miR-375-5p, miR-885-3p, miR-940, miR-1260a, miR-1260b, miR-1470, miR-4429, miR-4435, miR-4787-3p, miR-6511-5p, miR-6792-5p, miR-6872-3p, and miR-6880-3p were >5% lower, and levels of miR-92a-3p, miR-486-5p, and miR-4476 were >5% higher when serum stored for 0.5 h post-centrifugation before transfer and freezing compared to controls immediately transferred and frozen. Of these, the changes in miR-92a-3p (0.48-fold), miR-486-5p (0.49-fold), miR-1470 (-0.41-fold), and miR-4787-3p (-0.46-fold) were significant (q<0.05). When serum was stored post-centrifugation for 1 h, significant (q<0.05) changes were found in miR-371b-5p (-1.05-fold), miR-885-3p (-1.61- fold) miR-940 (-2.42-fold), miR-1260a (-0.95-fold), miR-1260b (-1.25-fold), miR-1470 (-0.76-fold), miR-4429 (-1.80-fold), miR-4787-3p (-0.51 fold), miR-6511-5p (-1.49 fold), miR-6792-5p (-1.60-fold), miR-6872-3p (-0.93-fold), and miR-6880-3p (-0.43-fold). While storage of serum (post-centrifugation) generally had similar effects on specimens from the three different volunteers, the storage effects on miR-4435 and miR-486-5p levels were not observed in specimens from all volunteers. Nevertheless, in a mixed- effects model only miR-4435 was no longer significant (P<0.05). In an independent, validation set miR-4429 was 1.03-fold and miR-6511b-5p 1.49-fold higher in specimens with a 2 h processing time (1 h pre-centrifugation, 1 h post-centrifugation) than those with a 6 h processing time (unspecified pre- versus post- centrifugation).  Neither miR-4429 nor miR-6511b-5 were detected in EVs.

  • Biospecimens

    FluidSerum
    FluidBlood
  • Preservation Types

    Frozen
  • Diagnoses

    Normal
  • Platforms

    RNADNA microarray
  • Pre-analytical Factors

    ClassificationPre-Analytical FactorValue(s)
    Biospecimen Aliquots and ComponentsBlood and blood products
    Extracellular vesicles
    Serum
    Biospecimen Aliquots and ComponentsCentrifugation
    Centrifugation delays investigated
    StorageStorage conditions
    As blood
    As serum
    StorageStorage duration
    0, 0.5 and 1 h (post-centrifugation)
    0.5 h, 1 h or 6 h (precentrifugation)
    2 h processing time (1 h pre- and 1 h post-centrifugation)
    6 h processing time