RNA Amplification Protocol Leads to Biased Polymerase Chain Reaction Results Especially for Low-Copy Transcripts of Human Bone Marrow-Derived Stromal Cells.

Carolin Coenen,Stefanie Liedtke,Gesine Kogler

doi:10.1371/journal.pone.0141070

Abstract

The amplification of RNA is becoming increasingly important, as often only limited amounts of cells are available for gene expression analysis. In this study, the gene expression profile of the 39 human homeobox (HOX) genes was analyzed in bone marrow-derived multipotent stromal cells (BM-MSCs) by reverse transcription (RT-) and quantitative polymerase chain reaction (qPCR). For further unlimited gene expression analysis, Whole Transcriptome Amplification (WTA) was used to amplify RNA from human BM-MSCs. However, WTA led to biased RT- and qPCR results, and even non-detectability of HOX transcripts compared to non-amplified BM-MSC samples which instead revealed transcription. It is important to note that the same RNA of the respective human BM-MSC line was used for normal cDNA synthesis by standard reverse transcription (non-amplified RT samples) and for cDNA synthesis by WTA (amplified WTA samples). On this account, the different RT- and qPCR results were unexpected applying WTA. The significantly reduced detection of HOX transcripts after WTA has been demonstrated for numerous BM-MSC lines (n = 26) by RT-PCR analysis. Furthermore, undetectable HOX transcripts meaning HOX transcripts of human BM-MSCs that were detected after normal cDNA synthesis, but were no longer detectable after WTA, were consistently observed by qPCR analysis. Finally, qPCR experiments revealed a possible explanation for the differences between amplified and non-amplified BM-MSC samples: an inverse correlation between the biased qPCR results and the low expression level of the respective HOX gene. The PCR analysis of high-copy transcripts like GAPDH or RPL13A revealed unchanged qPCR results after WTA compared to corresponding non-amplified BM-MSC samples. In contrast, WTA led to biased qPCR results for medium-copy HOX transcripts, and even non-detectability of low-copy HOX transcripts of human BM-MSCs resulting in false negative RT- and qPCR data applying WTA.

Highlights

The analysis of transcriptome profiling provides understanding of important biological processes in cancer or other diseases
The Whole Transcriptome Amplification (WTA) method is modified from the Whole Genome Amplification (WGA) method based on Multiple Displacement Amplification (MDA) [5, 6]
The analysis of quantitative polymerase chain reaction (qPCR) revealed that the expression level of HOX genes comprise a wide range in human bone marrow-derived multipotent stromal cells (BM-MSCs) (Fig 1)

Summary

Introduction

The analysis of transcriptome profiling provides understanding of important biological processes in cancer or other diseases. Often only limited amounts of cells and correlating RNA are available for molecular biological analysis like cDNA microarrays [1], generation sequencing [2] or polymerase chain reactions [3]. The DNA amplification without thermal cycling is based on the extraordinary processibility of the enzyme phi polymerase synthesizing DNA fragments up to a length of 100 kb [7]. After phi polymerase tightly binds to random primers, the enzyme replicates the DNA template without dissociating from the DNA template strand. The hybridization of further primers at the displaced DNA strands leads to additional replication resulting in DNA products larger than 10 kb. The long and branched DNA products synthesized by phi polymerase ensure that MDA enables the representative amplification of human genomic DNA [5]. An advantage of phi polymerase compared to thermostable Taq polymerase usually used in PCR-based methods, is their 3’− 5’ exonuclease proofreading activity resulting in significantly lower error rate of 1 in 106−107 [8] in contrast to normal Taq polymerase with an error rate of 3 in 104 [9]

Methods

Results

Conclusion