Stability of Circulating Blood-Based MicroRNAs - Pre-Analytic Methodological Considerations

Abstract

Background and aim: The potential of microRNAs (miRNA) as non-invasive diagnostic, prognostic, and predictive biomarkers, as well as therapeutic targets, has recently been recognized. Previous studies have highlighted the importance of consistency in the methodology used, but to our knowledge, no study has described the methodology of sample preparation and storage systematically with respect to miRNAs as blood biomarkers. The aim of this study was to investigate the stability of miRNAs in blood under various relevant clinical and research conditions: different collection tubes, storage at different temperatures, physical disturbance, as well as serial freeze-thaw cycles.

Methods: Blood samples were collected from 12 healthy donors into different collection tubes containing anticoagulants, including EDTA, citrate and lithium-heparin, as well as into serum collection tubes. MiRNA stability was evaluated by measuring expression changes of miR-1, miR-21 and miR-29b at different conditions: varying processing time of whole blood (up to 72 hours (h)), long-term storage (9 months at -80°C), physical disturbance (1 and 8 h), as well as in a series of freeze/thaw cycles (1 and 4 times).

Results: Different collection tubes revealed comparable concentrations of miR-1, miR-21 and miR-29b. Tubes with lithium-heparin were found unsuitable for miRNA quantification. MiRNA levels were stable for at least 24 h at room temperature in whole blood, while separated fractions did show alterations within 24 h. There were significant changes in the miR-21 and miR-29b levels after 72 h incubation of whole blood at room temperature (p<0.01 for both). Both miR-1 and miR-21 showed decreased levels after physical disturbance for 8 h in separated plasma and miR-1 in serum whole blood, while after 1 h of disturbance no changes were observed. Storage of samples at -80°C extended the miRNA stability remarkably, however, miRNA levels in long-term stored (9 months) whole blood samples were significantly changed, which is in contrast to the plasma samples, where miR-21 or miR-29b levels were found to be stable. Repetitive (n = 4) freeze-thaw cycles resulted in a significant reduction of miRNA concentration both in plasma and serum samples.

Conclusion: This study highlights the importance of proper and systematic sample collection and preparation when measuring circulating miRNAs, e.g., in context of clinical trials. We demonstrated that the type of collection tubes, preparation, handling and storage of samples should be standardized to avoid confounding variables influencing the results.

Fig 1. Different blood tubes for miRNA analysis.

miRNA was isolated and levels of miR-1, miR-21 and miR-29b from either; EDTA-plasma, citrate-plasma, lithium-heparin-plasma, or serum fraction, were detected by RT-qPCR. Values were normalized to a spike-in control, cel-miR-39. Normalized average ΔC_T values for each are shown, demonstrating amplification from all sources, except lithium-heparin-plasma. In the remaining samples, the levels of miR-21 and miR-29b did not vary significantly between the blood tubes. Li = lithium, ND = not detectable. A) Copenhagen Cohort. B) Munich Cohort.

Fig 2. Stability of miRNA in whole blood incubated at room temperature.

Whole blood was collected into EDTA containing tubes and incubated for 0, 4, 8, 12, 24 and 72 h at room temperature before processed into plasma. qRT-PCR was performed. Statistically significant differences in miRNA expression are marked with asterisks (**). **p<0.01. A) miR-21. B) miR-29b.

Fig 3. Impact of delayed processing on separated blood fractions.

Whole blood was collected into EDTA or serum separator containing tubes and incubated at room temperature for 0h, 24h, and 4 days before processed into plasma (A) and serum (B) or processed into plasma (C) and serum (D) immediately and then incubated at room temperature for 0h, 24h, and 4d. Data presented as normalized average ΔC_Tvalues±SEM. *p<0.05, **p<0.01, ***p<0.001.

Fig 4. Impact of physical disturbance.

Whole blood was collected into EDTA or serum separator containing tubes and incubated on a shaker at 30 rpm at room temperature for 0h, 1h, and 8h before processed into plasma (A) and serum (B) or processed into plasma (C) and serum (D) immediately and then incubated on a shaker at 30 rpm at room temperature for 0 h, 1 h, and 8 h. Data presented as normalized average ΔC_Tvalues±SEM. *p<0.05, **p<0.01, One-way ANOVA with Dunnet post test.

Fig 5. Incubation of plasma at -80°C for up to 9 months.

There was no significant change in miRNA concentration up to 9 months after freezing for either miR-21 (A) or miR-29b (B) in plasma (p = 0.11 and p = 0.96). However, the concentration of both miRNAs in whole blood increased during storage at -80°C. *p<0.05, **p<0.01.

Fig 6. MiRNA stability after repetitive freeze-thaw cycles within the separated fractions.

Whole blood was collected into EDTA or serum separator containing tubes, immediately processed into plasma (A) and serum (B) and frozen at -80°C. Samples were thawed one or four times before RNA was isolated and RT-qPCR was performed. *p<0.05, **p<0.01, unpaired student t-test.

Fig 7. MiRNA levels in the 3 blood fractions; plasma, buffy coat and red blood cells (RBC).

The two miRNAs, miR-21 and miR-29b, were differently distributed in the fractions; there are significant higher levels of miR-21 in the buffy coat and RBC compared to plasma, and miR-29b plasma levels are significantly lower than in the buffy coat. Statistically significant differences in miRNA expression are marked with asterisks (*).**p<0.01, ***p<0.001.