Elimination of heparin interference during microarray processing of fresh and biobank-archived blood samples

• Published in: Environmental and molecular mutagenesis 2014-07, Vol.55 (6), p.482-491
• Main Authors: Hebels, Dennie G.A.J., van Herwijnen, Marcel H.M., Brauers, Karen J.J., de Kok, Theo M.C.M., Chalkiadaki, Georgia, Kyrtopoulos, Soterios A., Kleinjans, Jos C.S.
• Format: Article
• Language: English
• Subjects: biobank samples; blood; Blood Banking - methods; Blood Chemical Analysis - methods; Fluorescent Dyes - metabolism; fluorophore PCR labelling; Heparin - analysis; Heparin - metabolism; heparin interference; Humans; Lithium Chloride - pharmacology; lithium chloride treatment; Microarray Analysis - methods; Quality Control; RNA - blood; Temperature; transcriptomics
• ISSN: 0893-6692; 1098-2280
• DOI: 10.1002/em.21869

In the context of environmental health research, biobank blood samples have recently been identified as suitable for high‐throughput omics analyses enabling the identification of new biomarkers of exposure and disease. However, blood samples containing the anti‐coagulant heparin could complicate transcriptomic analysis because heparin may inhibit RNA polymerase causing inefficient cRNA synthesis and fluorophore labelling. We investigated the inhibitory effect of heparin and the influence of storage conditions (0 or 3 hr bench times, storage at room temperature or −80°C) on fluorophore labelling in heparinized fresh human buffy coat and whole blood biobank samples during the mRNA work‐up protocol for microarray analysis. Subsequently, we removed heparin by lithium chloride (LiCl) treatment and performed a quality control analysis of LiCl‐treated biobank sample microarrays to prove their suitability for downstream data analysis. Both fresh and biobank samples experienced varying degrees of heparin‐induced inhibition of fluorophore labelling, making most samples unusable for microarray analysis. RNA derived from EDTA and citrate blood was not inhibited. No effect of bench time was observed but room temperature storage gave slightly better results. Strong correlations were observed between original blood sample RNA yield and the amount of synthesized cRNA. LiCl treatment restored sample quality to normal standards in both fresh and biobank samples and the previously identified correlations disappeared. Microarrays hybridized with LiCl‐treated biobank samples were of excellent quality with no identifiable influence of heparin. We conclude that, to obtain high quality results, in most cases heparin removal is essential in blood‐derived RNA samples intended for microarray analysis. Environ. Mol. Mutagen. 55:482–491, 2014. © 2014 Wiley Periodicals, Inc.