Quality Control of RNA Preservation and Extraction from Paraffin-Embedded Tissue: Implications for RT-PCR and Microarray Analysis

Karl Kashofer,Kurt Zatloukal,Martin Pichler,Christian Viertler,Sai Yendamuri

doi:10.1371/journal.pone.0070714

Abstract

Analysis of RNA isolated from fixed and paraffin-embedded tissues is widely used in biomedical research and molecular pathological diagnostics. We have performed a comprehensive and systematic investigation of the impact of factors in the pre-analytical workflow, such as different fixatives, fixation time, RNA extraction method and storage of tissues in paraffin blocks, on several downstream reactions including complementary DNA (cDNA) synthesis, quantitative reverse transcription polymerase chain reaction (qRT-PCR) and microarray hybridization. We compared the effects of routine formalin fixation with the non-crosslinking, alcohol-based Tissue Tek Xpress Molecular Fixative (TTXMF, Sakura Finetek), and cryopreservation as gold standard for molecular analyses. Formalin fixation introduced major changes into microarray gene expression data and led to marked gene-to-gene variations in delta-ct values of qRT-PCR. We found that qRT-PCR efficiency and gene-to-gene variations were mainly attributed to differences in the efficiency of cDNA synthesis as the most sensitive step. These differences could not be reliably detected by quality assessment of total RNA isolated from formalin-fixed tissues by electrophoresis or spectrophotometry. Although RNA from TTXMF fixed samples was as fragmented as RNA from formalin fixed samples, much higher cDNA yield and lower ct-values were obtained in qRT-PCR underlining the negative impact of crosslinking by formalin. In order to better estimate the impact of pre-analytical procedures such as fixation on the reliability of downstream analysis, we applied a qRT-PCR-based assay using amplicons of different length and an assay measuring the efficiency of cDNA generation. Together these two assays allowed better quality assessment of RNA extracted from fixed and paraffin-embedded tissues and should be used to supplement quality scores derived from automated electrophoresis. A better standardization of the pre-analytical workflow, application of additional quality controls and detailed sample information would markedly improve the comparability and reliability of molecular studies based on formalin-fixed and paraffin-embedded tissue samples.

Highlights

Quantitative measurement of messenger RNA levels by quantitative reverse transcription polymerase chain reaction, microarray techniques or the more recently established next-generation sequencing has become an essential tool for studying gene expression and has given new insights into a multitude of pathophysiological processes
It is well known that ct-values generated by quantitative reverse transcription polymerase chain reaction (qRT-PCR) from RNA extracted from formalin-fixed and paraffin-embedded (FFPE) are higher than from cryopreserved tissue (CRYO)
To accurately measure the PCR-ability of RNA from CRYO and formalinfixed paraffin-embedded (FFPE) tissues we designed an assay for quantitative assessment of PCR efficiency for different amplicon lengths (71, 153, 200, 277, 323 and 530 bp) from the housekeeping gene glyceraldehyde-3-phosphate dehydrogenase (GAPDH)

Summary

Introduction

Quantitative measurement of messenger RNA (mRNA) levels by quantitative reverse transcription polymerase chain reaction (qRT-PCR), microarray techniques or the more recently established next-generation sequencing has become an essential tool for studying gene expression and has given new insights into a multitude of pathophysiological processes. Fixation by formalin causes modifications of biomolecules, such as cross-linkage of nucleic acids with proteins, covalent modifications of RNA by the addition of multiple monomethylol moieties to the amino groups of bases or methylene bridging between neighboring bases and fragmentation of the RNA molecules [3,4] These alterations lead to suboptimal performance of RNA extracted from FFPE tissues in subsequent analysis. To overcome this problem, several groups have proposed specific additives to the standard fixative or the use of noncrosslinking organic fixatives to improve RNA yield and quality [5,6,7,8,9,10,11]. All these alternative fixation methods have their limitations, either with respect to preservation of the histopathological diagnostic features or in other applications, like immunohistochemistry [12]

Methods

Results

Conclusion