Skip to main content

The focus on sample quality: Influence of colon tissue collection on reliability of qPCR data.

Korenkova V, Slyskova J, Novosadova V, Pizzamiglio S, Langerova L, Bjorkman J, Vycital O, Liska V, Levy M, Veskrna K, Vodicka P, Vodickova L, Kubista M, Verderio P

Publication: Sci Rep (2016) • Vol 6 • Page 29023PMID: 27383461View on PubMed

Purpose

This paper compared colon specimens preserved using the PAXgene Tissue System with those snap-frozen or frozen to identify potential differences in rRNA and mRNA integrity and gene expression profiles.

Conclusion

While mean RQI values were higher for colon tumor specimens than case-matched normal adjacent tissue when specimens were snap-frozen or preserved using PAXgenge, differences in RQI or ΔAMP values were not significant for any preservation method examined. Gene expression profiles were similar for both preservation methods (snap-frozen/frozen versus PAXgenge) for specimens collected at either hospital, but higher variability in gene expression was observed for frozen specimens when compared to PAXgene-preserved specimens. Further analysis of gene expression profiles demonstrated that in both hospitals, PAXgene specimens could be used to discriminate between tumor and normal adjacent tissue but only specimens snap-frozen in liquid nitrogen (as opposed to those frozen at -80°C) could be used for discrimination analysis. However, the overall discrimination between tumor and normal tissues was significantly improved when lower quality specimens (ΔAMP > 1.0 or RQI < 4) were removed from consideration in multivariate gene expression analysis.

Studies (1)

Purpose

This study compared colon specimens preserved using the PAXgene Tissue System with snap-frozen or frozen specimens to identify potential differences in rRNA and mRNA integrity and gene expression profiles. Tumor tissue and normal adjacent colon tissue (5–10 cm distant from the tumor) were collected from patients at two hospitals in the Czech Republic (n=14 and n=16, respectively). Each specimen (tumor or normal adjacent tissue) was divided and placed in PAXgene Tissue FIX solution for 24 h at room temperature or snap-frozen (in liquid nitrogen at Hospital A) or frozen (at -80°C at Hospital B). Both snap-frozen/frozen and PAXgene preserved specimens were stored at −80°C until RNA was isolated using the AllPrep DNA/RNA mini kit or the PAXgene Tissue RNA Kit, respectively. RNA quantity and purity were determined by spectrophotometer and RNA integrity was assessed by electrophoresis and the Experion RNA StdSence Analysis Kit.  Specifically, rRNA integrity was calculated using RNA Quality Indicator (RQI), an electrophoresis-based quality metric that ranges between 1 and 1025, and a RQI of 4 was set as the threshold for acceptable quality. cDNA was synthesized using a RevertAid First strand cDNA synthesis kit. mRNA degradation was quantified using the Differential Amplicons (ΔAMP) Method, which is calculated using the quantification cycle (Cq) value of short, medium, and long amplicons. Expression profiling was performed using high-throughput real-time qRT-PCR for 13 DNA repair-associated genes (OGG1, APEX1, NEIL1, PARP1, XPA, RPA1, RPA2, ERCC3, ERCC2, ERCC1, ERCC6, DDB1, and CCNH) that were normalized to two reference genes (TOP1 and 18S).

Summary of Findings

While no significant differences in rRNA integrity, as determined by RQI values, were observed between case-matched colon tumor and normal adjacent specimens for any of the preservation methods examined, mean RQI values were higher in snap-frozen/frozen tumor specimens than normal adjacent specimens for both hospitals (4.8 versus 6.1 at Hospital A, 3.9 versus 5.4 at Hospital B). When data from both tumor and normal adjacent tissue were pooled, median RQI values were higher in PAXgene specimens than snap-frozen/frozen specimens at both Hospital A (8.05 versus 5.85) and Hospital B (6.25 versus 4.15). Similarly, mRNA integrity, as determined by ΔAMP values, did not differ significantly between case-matched tumor and normal adjacent tissue specimens, regardless of preservation method. However, when data from case-matched tumor and normal adjacent specimens were pooled, ΔAMP values differed significantly between specimens preserved by the PAXgene Tissue system (median ΔAMP = − 0.13 in Hospital A and − 0.15 in Hospital B) and those that were snap-frozen/frozen (median ΔAMP = 0.14 in Hospital A and 0.41 in Hospital B), indicating higher quality in PAXgene than snap-frozen/frozen specimens (P=0.03 in Hospital A and P=0.005 in Hospital B).

Gene expression profiles were similar for specimens preserved by the PAXgene Tissue System and those that were snap-frozen/frozen at either hospital, but differences in individual genes were observed when tumor specimens were compared to normal adjacent specimens. At Hospital A, different levels of expression were observed for two genes (NEIL1 and XPA) in tumor versus normal adjacent tissue specimens preserved by the PAXgene Tissue System and for six genes (APEX1, DDB1, ERCC1, NEIL1, PARP1, RPA2) among snap-frozen specimens at Hospital B, expression of seven genes (CCNH, ERCC2, ERCC6, NEIL1, OGG1, RPA1, XPA)  differed significantly between tumor and normal adjacent tissue preserved by the PAXgene Tissue System, while no significant differences were observed among frozen specimens. Overall, gene expression profiles from snap-frozen/frozen specimens exhibited greater variability compared to PAXgene specimens, especially for those collected at Hospital B. Further analysis of gene expression profiles demonstrated they could be used to discriminate between tumor and normal adjacent tissue when preserved by PAXgene at both Hospital A and B (P=0.0021 and P= 0.0017, respectively) and snap-frozen specimens at Hospital A (P=0.0016); however, gene expression profiles did not discriminate between tumor and normal adjacent specimens that were collected and frozen at Hospital B (P= 0.19). When lower quality specimens (ΔAMP > 1.0 or RQI < 4) were removed from consideration in multivariate gene expression analysis, the ability to discriminate between tumor and normal adjacent tissues was significantly improved, especially when those with an RQI < 4 were excluded.

  • Biospecimens

    TissueColorectal
  • Preservation Types

    FrozenPAXgene
  • Diagnoses

    NeoplasticCarcinoma
    NeoplasticNormal Adjacent
  • Platforms

    RNASpectrophotometry
    RNAReal-time qRT-PCR
    RNAElectrophoresis
  • Pre-analytical Factors

    ClassificationPre-Analytical FactorValue(s)
    PreaquisitionDiagnosis/ patient condition
    Colon cancer
    Normal adjacent
    Biospecimen PreservationType of fixation/preservation
    Frozen
    PAXgene
    Snap frozen
    Real-time qRT-PCR SpecificTargeted nucleic acid
    18S
    APEX1
    CCNH
    DDB1
    ERCC1
    ERCC2
    ERCC3
    ERCC6
    NEIL1
    OGG1
    PARP1
    RPA1
    RPA2
    TOP1
    XPA