Pre-analytic Methodological Considerations and Sample Quality Control of Circulating miRNAs.
Author(s): Chan SF, Cheng H, Kai-Rui KG, Zou R
Publication: J Mol Diagn, 2023, Vol. , Page
PubMed ID: 37030398 PubMed Review Paper? No
Purpose of Paper
This paper identified microRNA (miRNA, miR) markers of hemolysis and platelet contamination in serum, plasma, and platelet poor plasma (PPP) and established threshold for hemolysis and platelet scores based on the relative expression of these markers. Platelet and hemolysis scores were then compared in plasma obtained in 6 versus 10 mL tubes, in PPP obtained by different centrifugation protocols, in PPP obtained after storage of blood for 0 or 7 h at room temperature versus on wet ice, and in PPP and serum stored at 25°C for up to 7 days, at 4°C for up to 30 days, and at -20°C and -80°C for up to 360 days.
Conclusion of Paper
Hemoglobin levels were higher in plasma and PPP than in serum regardless of clotting time but remained below the threshold of acceptable hemolysis (< or <=500 ug/mL) in all specimens. As expected, platelet activation (measured by thromboxane B2 levels) was higher in serum than plasma or PPP and increased progressively with clotting time. The mean expression of targeted miRNAs was highest in plasma and lowest in PPP and in serum; however, mean miRNA expression increased with clotting time along with thromboxane B2 levels. Principal component analysis (PCA) of miRNA expression profiles revealed that specimens clustered by specimen type (serum, plasma, PPP) and to a lesser extent based on clotting time, rather than by patient. The authors identified miR-1973 and miR-28-5p as markers of platelet contamination and miR-20b-5p, miR-363-3p, and miR-451a as markers of red blood cell (RBC) contamination; the authors also established thresholds for hemolysis and platelet scores based on the average cycle of these miRNAs relative to miRNAs either found at low levels in platelets or not expressed in RBC lysate, respectively.
In plasma, hemoglobin levels were higher in specimens from the initial blood draw (typically the discard tube) than in subsequent tubes, but tube volume did not affect hemolysis or platelet scores. When centrifuging blood to obtain PPP, increasing the speed of the first centrifugation from 1000 x g to 1500 x g had a non-significant trend toward a decline in RBC- and platelet-contamination (increased hemolysis and platelet scores, respectively). While the speed of the second centrifugation to produce PPP had no effect on hemolysis scores, platelet contamination declined (higher platelet scores) as centrifugation speed increased (up to 3000 x g), at which point faster centrifugation speeds (5000 x g) led to an increase in platelet contamination.. Based on these results, the authors stated that the ideal processing workflow to reduce RBC and platelet contamination is centrifugation at 1500 x g for 15 minutes followed by centrifugation at 2500 x g for 15 minutes. The platelet score increased (less platelet contamination) when a second centrifugation was performed, but the magnitude of the increase was higher when performed prior to frozen storage. The timing of the second centrifugation relative to frozen storage did not affect hemolysis score. While no difference in platelet or hemolysis score were noted between PPP obtained from blood immediately after collection or after 7 h at room temperature, a significant but modest increase in platelet contamination was observed following storage of blood for 7 h on wet ice. The mean expression of all detectable miRNAs decreased when serum was stored at 25°C for 3 days (P<0.0001) or when stored at 4°C for 30 days (P<0. 001), but only nonsignificant declines were noted when serum was stored at -20°C or -80°C for up to a year. Significant declines in the mean expression of miRNAs were observed when PPP was stored at 25°C for ≥3 days, 4°C for ≥14 days, and -20°C for 270 days but not 360 days. The degradation in miRNA mean expression was significantly albeit poorly correlated with the expression of miRNAs in serum (R=-0.3464, P<0.0001) and PPP (R=-0.2970, P<0.0001), the percentage GC (R=0. 0.1812, P=0.0017), and the length of the miRNA (R=-0.1906, P=0.0009).
Studies
-
Study Purpose
This study identified microRNA markers of hemolysis and platelet contamination in serum, plasma, and platelet poor plasma (PPP) and established hemolysis and platelet scores based on the relative expression of these markers. Platelet and hemolysis scores were then compared in plasma collected in 6 versus 10 mL tubes; in PPP obtained by different centrifugation protocols; in PPP obtained after storage of blood for 0 or 7 h at room temperature versus on wet ice; and in PPP and serum stored at stored at 25°C for up to 7 days, at 4°C for up to 30 days, and at -20°C and -80°C for up to 360 days. Blood was collected from 10 healthy volunteers into serum separator and K2EDTA plasma tubes using a 21-gauge Vacutainer Safety-Lok blood collection set. Serum separator tubes were allowed to clot upright for 30, 60, and 90 min at room temperature before separation of the serum by centrifugation at 1500 g for 15 min and storage at -80°C. Plasma was obtained by centrifugation of K2EDTA blood at 1500 g for 15 min and stored at -80°C. PPP was obtained by centrifugation of plasma at 2500 g for 15 min and stored at -80°C. Hemolysis was assessed by quantification of hemoglobin by ELISA and platelet activation was evaluated by quantification of thromboxane B2 (TXB2) by ELISA. RNA was isolated from serum and plasma spiked with synthetic controls using the Maxwell RSC miRNA Plasma and Serum Kit. Levels of 356 miRNA were preamplified for 14-cycles and quantified using real-time RT-PCR assays. To identify red blood cell markers, PPP was spiked with 0.01%, 0.05% 0.2% and 1% hemolysate. To investigate the effects of tube volume, blood was collected from four healthy volunteers into 6 and 10 mL EDTA tubes and plasma was obtained by centrifugation of K2EDTA blood at 1500 x g for 15 min and stored at -80°C. To investigate the effects of centrifugation protocol, blood was collected from five healthy volunteers into three 6 mL EDTA tubes. Plasma was obtained by centrifugation of one tube at 1000 x g for 15 min and the other remaining two tubes by centrifugation at 1500 x g for 15 min. Plasma was then recentrifuged at 1500 x g for 20 minutes, 2000 x g for 15 minutes, 3000 x g for 15 minutes, or 5000 g for 5 minutes. To investigate the effects of delayed processing, blood from four volunteers was stored on wet ice or at room temperature for 0 and 7 h before processing to obtain plasma and serum. To investigate the stability of miRNA in PPP and serum, specimens from four volunteers were stored at 25°C for 0, 3 days and 7 days; at 4°C for 0, 3 ,7, and 30 days; and at -20°C and -80°C for 0, 3 ,7, 30, 90, 180 and 360 days.
Summary of Findings:
Hemoglobin levels were higher in plasma and PPP than in serum regardless of clotting time but remained below the acceptable threshold of 500 µg/mL in all specimens. Platelet activation (measured by thromboxane B2 levels) was higher in serum than plasma or PPP and increased with clotting time. The mean expression of the miRNAs evaluated was highest in plasma, lowest in PPP, and increased with clotting time in serum specimens. Mean miRNA expression was modestly correlated with thromboxane B2 levels in serum (r=0.6425, P=0.00013). PCA of the miRNA expression profile clustered specimens by specimen type (serum, plasma and PPP) and to a lesser extent based on clotting time, rather than by patient. In serum, levels of miR-1973 and miR-28-5p were modestly correlated with thromboxane B2 levels (r=0.5693, P=0.00102 and r= 0.5917, P= 0.0006, respectively) and, as expected, levels of miR-1973 and miR-28-5p were highest in plasma and lowest in PPP. By normalizing levels of the two platelet miRNAs (miR-1973 and miR-28-5p) to two miRNAs not affected by platelet contamination (1290 and miR10b-5p), the authors established a score for platelet contamination. Levels of miR-20b-5p, miR-363-3p, and miR-451a had >8-fold difference in expression between RBC lysate and PPP. miR-1290 and miR10b-5p were found to have higher expression in PPP than RBC lysate, making them suitable reference genes for RBC markers. Thus, the authors established a hemolysis score by normalizing expression of miR-20b-5p, miR-363-3p, and miR-451a to miR-1290 and miR10b-5p.
Hemoglobin levels were higher in the initial blood draw specimen (discard tube) than in subsequent tubes, but tube volume did not affect the hemolysis or platelet scores of plasma. When centrifuging blood to obtain PPP, increasing the speed of the first centrifugation from 1000 x g to 1500 x g had a non-significant trend toward a decline in RBC and platelet contamination (increased hemolysis and platelet scores). While the speed of the second centrifugation to produce PPP had no effect on hemolysis scores, platelet contamination decreased with increasing centrifugation speed (up to 3000 x g), with increased platelet contamination noted when centrifugation speed was further increased to 5000 x g. Based on this, the authors stated the ideal processing to reduce RBC and platelet contamination is centrifugation at 1500 x g for 15 minutes followed by 2500 x g for 15 minutes. The platelet score increased (less platelet contamination) when a second centrifugation was performed(P<0.001), but the magnitude of the increase was higher when performed prior to frozen storage (P<0.01). Hemolysis score was not affected by whether the second centrifugation step occurred before or after freezing. While there were no differences in platelet or hemolysis score between PPP obtained from blood immediately after collection and after 7 h of storage at room temperature, a significant increase in hemolysis (elevated hemoglobin and decreased hemolysis score, P<0.05) and a slight increase in platelet contamination was observed following storage of blood for 7 h on wet ice. The mean expression of all detectable miRNAs decreased following storage of serum at 25°C for 3 days (P≤0.0001) and at 4°C for 30 days (P≤0.001), but only nonsignificant declines were noted following storage at -20°C and -80°C for up to a year. Significant declines in the mean expression of miRNAs were observed following storage of PPP at 25°C for ≥3 days (P≤0.05), 4°C for ≥14 days (P≤0.05), and -20°C for 270 days (P≤0.01) but not 360 days. The degradation in miRNA mean expression was significantly albeit poorly correlated with the expression of miRNAs in serum (R=-0.3464, P<0.0001) and PPP (R=-0.2970, P<0.0001), the percentage GC (R=0. 0.1812, P=0.0017), and the length of the miRNA (R=-0.1906, P=0.0009).
Biospecimens
Preservative Types
- Frozen
Diagnoses:
- Normal
Platform:
Analyte Technology Platform RNA Real-time qRT-PCR Protein ELISA Pre-analytical Factors:
Classification Pre-analytical Factor Value(s) Biospecimen Acquisition Type of collection container/solution 6 mL tube
10 mL EDTA tube
Biospecimen Aliquots and Components Aliquot size/volume 6 mL
10 mL
Biospecimen Aliquots and Components Blood and blood products Platelet-poor plasma
Plasma
Serum
Biospecimen Aliquots and Components Aliquot sequential collection Serial collections compared
Biospecimen Aliquots and Components Centrifugation Multiple durations compared
Multiple speeds compared
Biospecimen Aliquots and Components Hemolysis Hemolysate added
Storage Storage temperature 25°C
4°C
-20°C
-80°C
Room temperature
On ice
Storage Storage duration 0 h
7 h
0 days
3 days
7 days
30 days
90 days
180 days
360 days
Storage Storage conditions As plasma
As PPP