Establishment of preanalytical conditions for microRNA profile analysis of clinical plasma samples.
Author(s): Suzuki K, Yamaguchi T, Kohda M, Tanaka M, Takemura H, Wakita M, Tabe Y, Kato S, Nasu M, Hashimoto T, Mine S, Serizawa N, Tomishima K, Nagahara A, Matsuda T, Yamaji T, Tsugane S, Saito Y, Daiko H, Yoshikawa T, Kato K, Okusaka T, Ochiya T, Yamamoto Y, Yotsui S, Yamamoto T, Yamasaki T, Miyata H, Yasui M, Omori T, Ohkawa K, Ikezawa K, Nakabori T, Sugimoto N, Kudo T, Yoshida K, Ohue M, Nishizawa T
Publication: PLoS One, 2022, Vol. 17, Page e0278927
PubMed ID: 36516194 PubMed Review Paper? No
Purpose of Paper
The purpose of this paper was to compare sequencing metrics and miRNA profiles among: sodium (Na2) EDTA, potassium (K2) EDTA, sodium fluoride (NaF), sodium citrate, and Na2EDTA plasma from blood subjected to pre-centrifugation storage (1 h, 6 h, 1 days, 2 days and 3 days at 4°C), and from Na2EDTA plasma stored at 4°C (1 h versus 1, 3, 5, 7 and 30 days), -20°C (1 h versus 1, 3, 5, 7 and 30 days) or at -80°C for 3 y versus 4 or 5 years. The reproducibility of the assay was investigated by comparing sequencing metrics and expression of platelet and hemolysis markers among plasma specimens from cancer patients collected at three different sites.
Conclusion of Paper
Library concentrations were highest and the degree of hemolysis was lowest (when assessed by miR-451a) in Na2EDTA or K2EDTA plasma specimens. Further, Na2EDTA, K2EDTA and sodium fluoride (NaF) plasma had a higher percentage of reads that mapped to the human genome and a higher number of unique molecular identifiers (UMI) than sodium citrate plasma. Nevertheless, levels of the platelet markers miR-126-3p, miR-223-3p, let-7f-5p and miR-16-5p were comparable among the anticoagulants evaluated. When Na2EDTA plasma was used as reference, sodium citrate plasma had the most miRNAs with >2-fold differences in expression (62 miRNAs).
Storage of Na2EDTA blood at 4°C did not affect the number of UMIs or the expression of platelet markers (miR-126-3p, miR-223-3p, let-7f-5p and miR-16-5p), but led to a progressive decline in library concentration and a significant decrease in the percentage of reads that mapped to the human genome; an increase in the hemolysis marker miR-451a was also observed when Na2EDTA blood was stored at 4°C for 3 days. Changes in miRNA levels were limited to <2-fold when centrifugation was delayed by 6 h, but longer delays led to an increased number of miRNAs with >2-fold changes. A non-significant trend toward decreased library concentration, number of UMIs, and percentage of reads mapping to the human genome was found with longer storage durations of plasma at 4°C, but there was no effect of storing plasma for up to 30 days at -20°C; and, only library concentration was found to decline when plasma was stored at -80°C for 5 years versus 3 years. Further, while ≤ 5 miRNAs were differentially expressed between plasma stored for 1 h and ≤ 30 days at 4°C, ≤ 1 miRNA were differentially expressed between plasma stored for 1 h and ≤ 30 days at -20°C or 3 and 5 years at -80°C.
Blood from cancer patients collected and processed at three different sites did not differ with regards to library concentration, the number of UMIs, percentage of reads that mapped to the human genome, or levels of platelet miRNAs (miR-126-3p miR-223-3p, let-7f-5p and miR-16-5p). While levels of miR-451a differed by up to 2.3-fold among specimens from the three sites, the variability that occurred between sites was not larger than the variability that occurred among specimens at a single site.
Studies
-
Study Purpose
The purpose of this study was to compare sequencing metrics and the miRNA profile among: Na2EDTA, K2EDTA, NaF, and sodium citrate plasma collected from healthy volunteers. Blood was collected from eighteen healthy volunteers into Na2EDTA, K2EDTA, NaF and sodium citrate tubes. Plasma was separated by centrifugation at 1,200 g at 4°C within 2 h of collection unless otherwise specified. miRNA was isolated from plasma using the Maxwell RSC miRNA Plasma and Serum Kit and quantified using the QuantiFluor RNA System. Libraries were constructed using the QIAseq miRNA Library Kit and sequenced using the NextSeq 550Dx Kit.
Summary of Findings:
Sequencing library concentrations were highest when blood was collected in Na2EDTA or K2EDTA tubes. Significantly higher library concentrations were obtained from K2EDTA than NaF (P<0.05) or sodium citrate (P<0.001) blood, Na2EDTA blood than sodium citrate (P<0.001) blood, and NaF blood than sodium citrate blood (P<0.05). The percentage of reads that mapped to the human genome and the number of unique molecular identifiers (UMI) were both lower when blood was collected in sodium citrate tubes than any other tube type (P<0.01, all). In comparison to Na2EDTA plasma, K2EDTA plasma had higher (> 2-fold) levels of three miRNAs and lower (<1/2-fold) levels of two miRNAs, NaF plasma had higher levels of twenty-five miRNAs and lower levels of two miRNAs, and sodium citrate plasma had higher levels of fifty-two miRNAs and lower levels of ten miRNAs. Importantly, 22 of the 23 miRNAs with differential expression between Na2EDTA and NaF plasma were also differentially expressed when Na2EDTA and sodium citrate plasma were compared. While levels of the platelet markers miR-126-3p, miR-223-3p, let-7f-5p and miR-16-5p were comparable among the plasma types evaluated, levels of miR-451a were significantly lower in Na2EDTA and K2EDTA plasma compared to NaF plasma (P<0.001 and P<0.05, respectively) or sodium citrate plasma (P<0.01 and P<0.05, respectively).
Biospecimens
Preservative Types
- Frozen
Diagnoses:
- Normal
Platform:
Analyte Technology Platform RNA Next generation sequencing Pre-analytical Factors:
Classification Pre-analytical Factor Value(s) Biospecimen Acquisition Anticoagulant Sodium EDTA
Potassium EDTA
Sodium fluoride
Sodium citrate
-
Study Purpose
The purpose of this study was to compare sequencing metrics and miRNA profiles among Na2EDTA plasma subjected to pre-centrifugation storage (1 h, 6 h, 1 days, 2 days and 3 days at 4°C), and post-centrifugation storage at 4°C (1 h versus 1, 3, 5, 7 and 30 days), -20°C (1 h versus 1, 3, 5, 7 and 30 days), or at -80°C for 3 y versus 4 or 5 y. The reproducibility of the assay was investigated by comparing sequencing metrics and expression of platelet and hemolytic markers among plasma specimens collected from cancer patients at three different sites. Blood was collected from eighteen healthy volunteers into Na2EDTA tubes. Plasma was separated by centrifugation at 1,200 g at 4°C within 2 h unless otherwise specified. miRNA was isolated from plasma using the Maxwell RSC miRNA Plasma and Serum Kit and quantified using the QuantiFluor RNA System. Libraries were constructed using the QIAseq miRNA Library Kit and sequenced using the NextSeq 550Dx Kit. To investigate potential effects of delayed processing, case-matched Na2EDTA blood tubes from five volunteers were stored for 1 h and 6 h at room temperature, and 1 day, 2 days and 3 days at 4°C before plasma separation. To investigate the effects of plasma storage, Na2EDTA plasma was stored for 1 h, 1 day, 3 days, 5 days, 7 days and 30 days at 4°C and at -20°C and for 3, 4, and 5 years at -80°C. To investigate inter-laboratory reproducibility, Na2EDTA blood was collected from 20 cancer patients (gastric, esophageal or pancreatic cancer) at each of three sites, then was centrifuged to obtain plasma within 12 h and the plasma was then stored at -80°C.
Summary of Findings:
Pre-centrifugation storage of Na2EDTA blood at 4°C did not affect RNA concentration or the number of UMIs but led to a progressive decline in library concentration (P<0.05 after ≥ 2 days compared to 1 h) and a significantly lower percentage of reads mapping to the human genome when stored for 3 days rather than 1 h (P<0.01). Relative to plasma obtained within 1 h of blood collection,, no miRNA were differentially expressed (<0.5 or > 2-fold) in plasma from blood centrifuged after 6 h, but when centrifugation was delayed further an increasing number of miRNA were expressed at >2- fold higher (5 miRNAs after 1 day, 16 after 2 days and 40 after 3 days) or at <1/2 - fold (4 at 2 days and 8 at 3 days); most miRNAs affected by a 2 day delay to centrifugation were also affected by a 3 day delay. Among the miRNAs that declined significantly when blood was subjected to a 3-day centrifugation delay at 4°C were miR-197-3p (P<0.01), miR-485-3p (P<0.01) and miR-130b-5p (P<0.01). While the platelet markers miR-126-3p, miR-223-3p, let-7f-5p and miR-16-5p were unaffected by a ≤3 days at 4°C centrifugation delay, levels of miR-451a were significantly higher when blood was stored for 3 days at 4°C compared to a 1 h delay to centrifugation.
A non-significant trend toward decreased library concentration, number of UMIs, and percentage of reads mapping to the human genome was observed with longer plasma storage durations at 4°C. No miRNAs were differentially expressed between plasma stored for 1 h and 1 day at 4°C, but a few miRNAs were found to increase (>2-fold) or decrease (<1/2-fold) when plasma was stored at 4°C for 3 days (1 decreased), 5 days (3 increased and 2 decreased), 7 days (3 decreased) and 30 days (3 increased and 2 decreased) compared to plasma stored for 1 h. There was no effect of storing plasma for up to 30 days at -20°C on library concentration, number of UMIs, or the percentage of reads mapping to the human genome; and compared to plasma stored for 1 h at -20°C, plasma stored for ≤ 30 days at -20°C had increases/decreases in ≤ 1 miRNA. Library concentrations were comparable from plasma stored at -80°C for 3 and 4 years but were significantly lower in plasma stored for 5 years than 3 years (P<0.05). There was no effect of storing plasma for 4 or 5 years versus 3 years at -80°C on the number of UMIs, percentage of reads mapping to the human genome, or miR-451a and miR-126-3p levels; and compared to plasma stored for 3 years at -80°C, plasma stored for 5 years at -80°C had increased levels of only one miRNA.
When blood from cancer patients was collected and processed at three different sites, there were no differences in the library concentration, number of UMIs, percentage of reads mapping to the human genome or levels of platelet miRNAs (miR-126-3p miR-223-3p, let-7f-5p and miR-16-5p); and while levels of miR-451a differed by up to 2.3-fold among the sites, the variability between sites was not larger than the variability between specimens at a single site.
Biospecimens
Preservative Types
- None (Fresh)
- Frozen
Diagnoses:
- Neoplastic - Carcinoma
- Normal
Platform:
Analyte Technology Platform RNA Next generation sequencing Pre-analytical Factors:
Classification Pre-analytical Factor Value(s) Biospecimen Aliquots and Components Centrifugation Centrifugation delays investigated
Storage Storage temperature 4°C
-20°C
-80°C
Storage Storage duration 1 h
6 h
1 day
2 days
3 days
5 days
7 days
30 days
3 years
4 years
5 years
Biospecimen Acquisition Locale of biospecimen collection Between site variability