Skip to main content

Bisulfite-Based DNA Methylation Analysis from Recent and Archived Formalin-Fixed, Paraffin Embedded Colorectal Tissue Samples.

Kalmár A, Péterfia B, Hollósi P, Wichmann B, Bodor A, Patai ÁV, Schöller A, Krenács T, Tulassay Z, Molnár B

Publication: Pathol Oncol Res (2015) • Vol 21 • Page 1149-56PMID: 25991403View on PubMed

Purpose

This paper compared DNA yield, bisulfate conversion success, and methylation status in recent (<6 months) and archival (>5 years) tumor and normal adjacent specimens extracted using the QIAamp and HighPure DNA isolation kits.

Conclusion

Extraction of DNA using the High Pure kit rather than QIAamp resulted in higher yields of bisulfate-converted DNA from tumor specimens, higher success in determination of methylation levels in archival specimens, a smaller increase in methylation in archival specimens, and comparable yields and methylation success in archival specimens; however, the QIAamp kit resulted in higher OD260/280 and OD 260/230 ratios and higher DNA yields from recent specimens. Importantly, the methylation status (hyper or hypo in comparison to normal adjacent) of the three genes differed between extraction methods used with concordance between methods only obtained in 3 of 5 cases for MAL and SFRP1 and 4 of 5 cases for SFRP2 in archival specimens. The promoters of MAL, SFRP1, and SFRP2 were hypermethylated in tumor specimens when compared to the matched normal adjacent specimen. 

Studies (1)

Purpose

This study compared DNA yield, bisulfate conversion success, and methylation status in recent (<6 months) and archival (>5 years) tumor and normal adjacent specimens extracted using the QIAamp and HighPure DNA isolation kits. Ten matched colorectal adenocarcinomas and normal adjacent specimens were fixed in formalin for an unspecified duration, embedded in formalin, and stored for less than 6 months (5 pairs) or more than 5 years (5 pairs) before use. Sections (10 µm) were deparaffinized in two changes of xylene and DNA was extracted from four serial sections using the High Pure FFPET DNA Isolation kit or the QIAamp DNA FFPE Tissue kit. The digestion was extended to 2 h at 56˚C for all archival specimens and for specimens stored <6 months when extracted with the High Pure kit. Archival specimens were eluted in 50µL, but specimens stored < 6 months were eluted in 100µL. Resultant DNA was quantified by spectrophotometer and were analyzed by spectrophotometer. DNA was bisulfate converted with the Zymo EZ DNA Methylation Kit and analyzed by high resolution melting of real-time PCR amplified DNA. DNA was sequenced using the GS Junior Titanium Sequencing Kit.  HumanMethylation450k BeadChip data from 38 colorectal tumor specimens were downloaded from The Cancer Genome Atlas (TCGA).

Summary of Findings

Average spectrophotometric DNA yield from recent specimens was higher when the QIAamp kit was used for extraction rather than the High Pure kit (11.51 µg versus 4.10 µg,) but the yield from archival specimens was not affected by kit choice. The authors report the OD260/280 and OD 260/230 ratios were significantly higher when extraction was with the QIAamp kit rather than the High Pure kit; however, there was significantly more DNA in tumor specimens extracted using High Pure than QIAamp kit after bisulfate conversion, a difference that the authors state could be partially due to overestimation of initial DNA concentration in QIAamp-extracted specimens by spectrophotmetry. Although methylation of all three genes could be evaluated in all recent specimens, DNA extracted from archival specimens with QIAamp kit had higher failure rates than those extracted with the HighPure kit (25% versus 2.5% for MAL, 10% versus 0% for SFRP1, and 12.5% versus 0% for SFRP2). Methylation levels were higher in archived than recent specimens and this was particularly true for normal adjacent specimens in which methylation was not expected and when extraction was with the QIAamp kit. Importantly, the methylation status (hyper or hypo in comparison to normal adjacent) of the three genes differed between extraction methods used with concordance between methods only obtained in 3 of 5 cases for MAL and SFRP1 and 4 of 5 cases for SFRP2 in archival specimens. Similar to what was observed in the TCGA dataset, GS Junior sequencing showed that the promoters of MAL, SFRP1, and SFRP2 were hypermethylated in tumor specimens when compared to the matched normal adjacent specimen.

  • Biospecimens

    TissueColorectal
  • Preservation Types

    Formalin
  • Diagnoses

    NeoplasticCarcinoma
    NeoplasticNormal Adjacent
  • Platforms

    DNASpectrophotometry
    DNABisulfite conversion assay
    DNADNA sequencing
    DNADNA microarray
    DNAReal-time qPCR
  • Pre-analytical Factors

    ClassificationPre-Analytical FactorValue(s)
    Biospecimen AcquisitionBiospecimen location
    Normal adjacent
    Tumor
    StorageStorage duration
    <6 months
    >5 years
    Analyte Extraction and PurificationAnalyte isolation method
    High Pure FFPET DNA Isolation kit
    QIAamp DNA FFPE Tissue kit
    DNA sequencing SpecificTargeted nucleic acid
    MAL
    SFRP1
    SFRP2
    Real-time qPCR SpecificTargeted nucleic acid
    MAL
    SFRP1
    SFRP2