A multiplex endpoint RT-PCR assay for quality assessment of RNA extracted from formalin-fixed paraffin-embedded tissues

Abstract

Background: RNA extracted from formalin-fixed paraffin-embedded (FFPE) samples is chemically modified and degraded, which compromises its use in gene expression studies. Most of the current approaches for RNA quality assessment are not suitable for FFPE derived RNA.

Results: We have developed a single-tube multiplex endpoint RT-PCR assay specifically designed to evaluate RNA extracted from FFPE tissues for mRNA integrity and performance in reverse transcription - quantitative real-time PCR (RT-qPCR) assays. This single-tube quality control (QC) assay minimises the amount of RNA used in quality control. mRNA integrity and the suitability of RNA for RT-PCR is evaluated by the multiplex endpoint RT-PCR assay using the TBP gene mRNA as the target sequence. The RT-PCR amplicon sizes, 92, 161, 252 and 300 bp, cover a range of amplicon sizes suitable for a wide range of RT-qPCR assays. The QC assay was used to evaluate RNA prepared by two different protocols for extracting total RNA from needle microdissected FFPE breast tumour samples. The amplification products were analysed by gel electrophoresis where the spectrum of amplicon sizes indicated the level of RNA degradation and thus the suitability of the RNA for PCR. The ability of the multiplex endpoint RT-PCR QC assay to identify FFPE samples with an adequate RNA quality was validated by examining the Cq values of an RT-qPCR assay with an 87 bp amplicon.

Conclusions: The multiplex endpoint RT-PCR assay is well suited for the determination of the quality of FFPE derived RNAs, to identify which RT-PCR assays they are suitable for, and is also applicable to assess non-FFPE RNA for gene expression studies. Furthermore, the assay can also be used for the evaluation of RNA extraction protocols from FFPE samples.

Figure 1

Assessment of total RNA yield and purity extracted by the two different RNA extraction protocols. The x-axis shows the sample identification. The y-axis on the left shows the RNA concentration in ng/μL. The RNA concentrations for each of the sample replicates are shown as groups of three bars. The y-axis on the right shows the absorbance ratio A₂₆₀/A₂₈₀. The absorbance ratio A₂₆₀/A₂₈₀ for each of the sample replicates is shown as a scatter graph (■: replicate 1; ▲: replicate 2; ●: replicate 3). Panel A corresponds to protocol 1 (P1) and Panel B corresponds to protocol 2 (P2). Note that for the RNA concentration, panel A is scaled differently to panel B as the RNA extracted by the two different RNA extraction protocols was eluted in different final elution volumes.

Figure 2

Primer placement for the multiplex endpoint RT-PCR and the RT-qPCR assay. Both assays use the TBP mRNA (NM_003194) as the target sequence. The eight exons are shown as rectangles labelled from E1 to E8. The primer locations are shown as horizontal arrow heads. The positions of the primers for the multiplex endpoint RT-PCR assay (92-F, 92-R, 161-F, 161-R, 252-F, 252-R, 300-F and 300-R) and for the RT-qPCR assay (RT-qPCR-F and RT-qPCR-R) are shown. Each primer pair was designed to span at least one intron.

Figure 3

Assessment of RNA degradation and RNA performance in RT-PCR by the multiplex endpoint RT-PCR assay. The sizes of the molecular weight markers (MW) are given on the left, whereas the sizes of the TBP amplicons are indicated on the right. Sizes are given in base pairs. Lane 1 and 10 were loaded with the no template control (NTC). Lane 2 and 9 were loaded with the PCR reaction obtained from a cDNA mixture synthesised from RNA extracted from different cell lines and serves as positive control, showing all four PCR amplification products of the expected size. Lane 8 was loaded with the PCR reaction obtained from genomic DNA and served as negative control. Lanes 3 to 7 were loaded with RT-PCR reactions obtained from cDNAs synthesised from total RNA derived from needle microdissected FFPE breast tumour tissues (Lane 3: P1-98-05, replicate 1; Lane 4: P1-98-06, replicate 1; Lane 5: P1-98-07, replicate 1; Lane 6: P1-98-08, replicate 1; Lane 7: P1-98-09, replicate 1).

Figure 4

Comparison of the multiplex endpoint RT-PCR and RT-qPCR assays for protocol 1 (P1). The bottom axis identifies the samples. The band sizes obtained for each of the sample replicate are shown as bars. The values on the top of the graph provide both the C_q value measured for each sample replicate in the TBP RT-qPCR assay and the mean C_q value. Symbols are as follows: * indicates a standard deviation between replicates that is greater than 0.5; ** one out of two replicates amplified.

Figure 5

Comparison of the multiplex endpoint RT-PCR and RT-qPCR assays for protocol 2 (P2). The bottom axis identifies the samples. The band sizes obtained for each of the sample replicate are shown as bars. The values on the top of the graph provide both the C_q value measured for each sample replicate in the TBP RT-qPCR assay and the mean C_q value. Samples labelled with a * symbol indicate a standard deviation between replicates that is greater than 0.5.

Figure 6

Spearman correlation between fragment size in the multiplex endpoint RT-PCR assay and the appropriate C_q value obtained in the RT-qPCR assay. The x-axis shows the fragment size and the y-axis shows the C_q value. Panel A corresponds to protocol 1 (P1) and Panel B corresponds to protocol 2 (P2).