Cancer Diagnosis Program | Biorepositories and Biospecimen Research Branch

Search BRD Papers

Haemolysis Detection in MicroRNA-Seq from Clinical Plasma Samples.

Author(s): Smith MD, Leemaqz SY, Jankovic-Karasoulos T, McAninch D, McCullough D, Breen J, Roberts CT, Pillman KA

Publication: Genes (Basel), 2022, Vol. 13, Page

PubMed ID: 35886071 PubMed Review Paper? No

Purpose of Paper

Conclusion of Paper

Studies

1. Study Purpose

   The purpose of this study was to identify a miRNA hemolysis signature of plasma with known levels of hemolysis by high throughput sequencing and verify the signature in pregnant and non-pregnant women, in males with prostate cancer and patients (male and female) with radiographic knee osteoarthritis. The signature was compared to hemolysis detection by spectrophotometry (absorbance at 414 nm) and the ratio of miR-23a-3p to miR-451a. Blood was collected from 111 pregnant and 10 non-pregnant women into EDTA tubes and stored on ice until plasma separation. Plasma was separated by dual-centrifugation at 800 g for 15 min at 4°C (second centrifugation speed not specified) or 1015 g for 10 min at 4°C (second centrifugation speed not specified).  Plasma aliquots were stored at -80°C.  RNA was extracted from plasma with the Qiagen miRNA Serum/Plasma Kit and stored at -80°C. Sequencing libraries were prepared with the QIAseq miRNA Library and QIAseq miRNA 48 Index IL Kits and assessed by Illumina NovaSeq 75 bp single-end read sequencing. Hemolysis was quantified by real-time PCR using the ratio of miR-23a-3p to miR-451a, a threshold difference of >7 cycles was indicative of hemolysis.  The hemolysis metric was further validated using publicly available datasets from patients with prostate cancer and those with early radiographic knee osteoarthritic.

   Summary of Findings:

   A total of 12 specimens were hemolyzed based on a ratio of miR-23a-3p to miR-451a that was ≥7. No difference in the percentage of specimens excluded due to library size (numbers not included) was observed between hemolyzed and non-hemolyzed plasma. There were significantly fewer mature miRNAs detected at a given read-depth in hemolyzed specimens than non-hemolyzed specimens (P=1.68 x 10^-9).  However, hemolyzed plasma had higher levels of 100 (Log₂ fold change > 0) and 60 (Log₂ fold change > 0.9) miRNAs than non-hemolyzed specimens (false discovery rate <0.05); from the affected miRNA,20 miRNAs with high confidence were selected for the hemolysis miRNA signature. A total of 127 miRNAs were differentially expressed in plasma from pregnant and non-pregnant women. Interestingly, miR-451a, a common marker of hemolysis, was strongly correlated with pregnancy status, as were nine other miRNAs in the 20 miRNA hemolysis signature. Exclusion of the 10 hemolysis markers that were significantly affected by pregnancy resulted in a hemolysis signature of the 10 remaining miRNAs.  A ≥ 1.9 difference in the geometric mean of the 10 miRNA signature relative to all other miRNA was indicative of hemolysis (classified as “Caution”). Twelve specimens were identified as hemolyzed based on a ratio of miR-23a-3p to miR-451a ≥ 7, and 13 additional specimens were identified using the 10 miRNA signature. The geometric mean of the 10 miRNA signature was strongly correlated with the ratio of miR-23a-3p to miR-451a (R=0.90).  Importantly, a slightly higher correlation with absorbance at 414 nm was observed for the new metric than the ratio of miR-23a-3p to miR-451a (R =0.87 versus R=0.82).  After exclusion of miRNAs with evidence of changes in prostate cancer (2 miRNAs) or osteoarthritis (1 miRNA), the new metric was found to be correlated with the ratio of miR-23a-3p to miR-451a in prostate cancer and osteoarthritis datasets, respectively. Further there was no difference in the relationship between the new metric and the ratio of miR-23a-3p to miR-451a between specimens from males and females.

   Biospecimens

   • Bodily Fluid - Blood
   • Bodily Fluid - Plasma

   Preservative Types

   • Frozen

   Diagnoses:

   • Neoplastic - Carcinoma
   • Pregnant
   • Osteoarthritis
   • Normal

   Platform:

   Analyte	Technology Platform
   Protein	Spectrophotometry
   RNA	Real-time qRT-PCR
   RNA	Next generation sequencing

   Pre-analytical Factors:

   Classification	Pre-analytical Factor	Value(s)
   Preaquisition	Patient gender	Female Male
   Real-time qRT-PCR Specific	Targeted nucleic acid	miR-23a-3p miR-451
   Real-time qRT-PCR Specific	Technology platform	Spectrophotometer NGS of 10 miRNA signature Ratio of miR-23a-3p to miR-451
   Preaquisition	Diagnosis/ patient condition	Pregnant Not pregnant
   Biospecimen Aliquots and Components	Hemolysis	Absent Present

   View more