Skip to main content

Quality control of RNA preservation and extraction from paraffin-embedded tissue: implications for rt-PCR and microarray analysis.

Kashofer K, Viertler C, Pichler M, Zatloukal K

Publication: PLoS One (2013) • Vol 8 • Page e70714PMID: 23936242View on PubMed

Purpose

The purpose of this paper was to compare the efficacy of RNA quantity, quality, and expression assays among tissue types preserved using formalin, a non-cross linking fixative (Tissue-Tek Xpress Molecular Fixative), or by snap-freezing in liquid nitrogen-cooled isopentane.

Conclusion

QRT-PCR success was dependent upon fragment length and was influenced by tissue type in formalin-fixed paraffin-embedded specimens (FFPE) and Tissue-Tek Xpress Molecular Fixative and paraffin-embedded (TFPE) specimens, but amplification of all fragment sizes and from all tissue types was possible with snap-frozen specimens. For FFPE specimens, time in fixative also impacted successful RNA amplification. RNA integrity numbers (RIN) and 260/280 OD ratios were not reliable predictors of successful amplification of RNA isolated from FFPE specimens. In general, cycle threshold (Ct) values were 10 and 4 cycles higher in RNA from FFPE and TFPE specimens, respectively, compared to snap frozen specimens. The authors determined that the efficacy of reverse transcription was adversely impacted in FFPE, but not TFPE, specimens in comparison to snap frozen specimens. When results of a qRT-PCR assay were directly compared to those obtained using an Applied Biosystems Whole Genome 2.0 microarray, expression patterns were markedly different for the same sample for FFPE and TFPE, but not snap frozen, specimens. Based on comparisons with snap frozen specimens, the authors conclude that qRT-PCR is more accurate than whole genome microarray for FFPE and TFPE specimens. Storage of FFPE specimens for a year resulted in increased Ct values and variability compared to those stored for 6 months.

Studies (6)

Purpose

This study compared qRT-PCR performance in four different tissue types preserved by FFPE, TFPE, or by snap-freezing in liquid nitrogen-cooled isopentane. Glyceraldehyde 3-phosphate dehydrogenase (GAPDH) amplicons of different sizes (71-530 bp) were assessed in 99 samples collected from 28 patients by raw Ct value and gel electrophoresis of the end product. RIN was also evaluated as a predictor of qRT-PCR performance in snap frozen and FFPE liver specimens. RNA was extracted from snap frozen specimens with Trizol and from FFPE specimens using a modified phenol/chloroform method that had previously been reported.

Summary of Findings

Amplification of all GAPDH fragments was successful using RNA isolated from snap-frozen thyroid, kidney, and duodenum specimens. For RNA from FFPE and TFPE specimens qRT-PCR success was influenced by tissue type and fragment size. Amplification of GAPDH was successful using RNA from FFPE kidney, thyroid and duodenum specimens when the GAPDH fragments were less than or equal to 150 bp, 200 bp, and 277 bp, respectively. Amplification was successful for all sized fragments of GAPDH using RNA from TFPE kidney and of GAPDH fragments less than or equal to 323 bp using RNA from TFPE thyroid and duodenum. When raw Ct values were plotted against amplicon size, the slope of the regression line was greater for RNA from FFPE specimens (2.4) compared to TFPE (0.55) or snap frozen (0.34) liver specimens. Raw Ct values also differed significantly when RNA from FFPE and snap frozen liver specimens were compared using a different specimen set (p<0.01). For RNA from snap frozen liver specimens, raw Ct values were modestly correlated to RIN (r= -0.05 to -0.6) while the correlation was weak (r= -0.25 to -0.28) for FFPE specimens. The authors also compared the efficacy of reverse transcription for each fixation method. For snap frozen and TFPE specimens, a positive linear relationship was observed between the amount of RNA template and cDNA product. For FFPE specimens, the slope associated with RNA template and cDNA product was greatly reduced, as 1 ug of RNA template from FFPE specimens was necessary to generate the same amount of cDNA as 0.125 ug of RNA template from snap frozen or TFPE specimens. The authors note that while snap frozen and TFPE specimens had similar reverse transcription efficiencies, qRT-PCR analysis of TFPE specimens generated raw Ct values that were 4 cycles higher than that for case-matched snap frozen specimens.

  • Biospecimens

    TissueThyroid Gland
    TissueKidney
    TissueSmall Bowel
    TissueLiver
  • Preservation Types

    FrozenOther PreservativeFormalin
  • Diagnoses

    NeoplasticCarcinoma
    NeoplasticBenign
  • Platforms

    RNASpectrophotometry
    RNAAutomated electrophoresis/Bioanalyzer
    RNAReal-time qRT-PCR
  • Pre-analytical Factors

    ClassificationPre-Analytical FactorValue(s)
    Biospecimen AcquisitionBiospecimen location
    Duodenum
    Kidney
    Liver
    Thyroid gland
    Biospecimen PreservationType of fixation/preservation
    Formalin (buffered)
    Snap frozen
    Xpress Molecular Fixative
    Real-time qRT-PCR SpecificLength of gene fragment
    153 bp
    200 bp
    277 bp
    323 bp
    530 bp
    71 bp
    Real-time qRT-PCR SpecificTargeted nucleic acid
    GAPDH
    Real-time qRT-PCR SpecificTemplate/input amount
    0 ug
    0.125 ug
    0.25 ug
    0.50 ug
    1 ug