Skip to main content

Comparison of Automated and Manual DNA Isolation Methods for DNA Methylation Analysis of Biopsy, Fresh Frozen, and Formalin-Fixed, Paraffin-Embedded Colorectal Cancer Samples.

Kalmár A, Péterfia B, Wichmann B, Patai ÁV, Barták BK, Nagy ZB, Furi I, Tulassay Z, Molnár B

Publication: J Lab Autom (2015) • Vol 20 • Page 642-51PMID: 25576093View on PubMed

Purpose

The paper compared DNA yield, purity, integrity and methylation rates from frozen surgically-resected specimens, frozen RNAlater-preserved biopsies and FFPE biopsies from colorectal carcinoma (CRC) patients and healthy patients using automated or manual extraction methods.

Conclusion

Although maximum amplicon size and methylation rates in frozen surgically-resected or RNAlater-preserved biopsy specimens were generally comparable between extraction methods, manual extraction of DNA from FFPE specimens resulted in larger maximum amplicon size and less variability in methylation rates of the T-cell differentiation protein MAL; however, hypermethylation of MAL in FFPE specimens in comparison to frozen or RNAlater specimens was still observed. Interestingly, discordance in methylation status between manually and automatically extracted DNA was limited to one of the three genes examined and was only present in FFPE specimen. Manual DNA extraction also resulted in higher DNA yields from FFPE and frozen RNAlater-preserved biopsy specimens and higher optical density (OD)260/280 and OD260/230 ratios from FFPE specimens, but lower OD260/230 ratios from frozen surgically-resected and RNAlater-preserved biopsy specimens. 

Studies (1)

Purpose

The study compared DNA yield, purity, integrity and methylation rates from of frozen surgically-resected specimens, frozen RNAlater-preserved biopsies and FFPE biopsies from CRC and healthy patients on using automated or manual extraction methods. Surgically-resected colorectal carcinoma and normal adjacent specimens from 10 patients were snap-frozen and stored at -80˚C. Biopsy specimens were obtained from 10 patients with CRC and 10 healthy patients during colonoscopy, placed in RNAlater and stored at -80˚C.  Sections from FFPE blocks from an additional 10 patients with CRC and 10 without were stored for <6 months, but no other details of FFPE processing were provided. DNA was extracted from frozen specimens and xylene-deparaffinized FFPE sections using the manual QIAamp DNA Mini Kit and the automated MagNA Pure DNA and Viral NA SV kit.

Summary of Findings

Higher DNA yields were obtained from FFPE and frozen RNAlater-preserved biopsy specimens when DNA was extracted using the manual method rather than the automated method (p<0.01, both), but no difference between extraction methods was observed for frozen surgically-resected specimens. While the average OD260/280 in DNA from frozen surgically-resected and RNAlater-preserved biopsy specimens were comparable between extraction methods, the OD260/280 was higher in FFPE specimens extracted with the manual method rather than the automated method (2.00 versus 1.83, p<0.01). OD260/230 ratios were higher in frozen surgically-resected and RNAlater-preserved biopsy specimens extracted automatically rather than manually, but higher OD260/230 ratios from FFPE tumor specimens were obtained using the manual rather than automated extraction method. DNA integrity as determined by multiplex PCR and methylation rates were generally comparable between extraction methods for frozen and RNAlater-preserved specimens, albeit with some small differences. For 10 of the 20 FFPE specimens, a larger maximum amplicon size was possible following manual extraction than automated extraction and for only one specimen was the reverse true. Similarly, less variability in MAL methylation was observed among specimens that were manually extracted compared to those that were automatically extracted. Overall, FFPE specimens of normal tissue were more likely to have MAL hypermethylation and increased variability in methylation compared to frozen normal adjacent or RNAlater-preserved healthy specimens. Methylation results for each of the three genes analyzed were strongly to very strongly correlated between extraction methods in frozen surgically-resected (r=0.85-0.89)and frozen RNAlater-preserved biopsy specimens (r=0.78-1.00) and the correlations between methods for Secreted frizzled-related protein(SFRP)1 and 2 methylation in FFPE specimens were very strong (r=0.94 and r=0.95, respectively), but the correlation for MAL methylation in FFPE specimens between automated and manual methods was weak (r=0.3065).

  • Biospecimens

    TissueColorectal
  • Preservation Types

    FormalinFrozenRNAlater
  • Diagnoses

    NeoplasticCarcinoma
    NeoplasticNormal Adjacent
    Normal
  • Platforms

    DNABisulfite conversion assay
    DNAPCR
    DNASpectrophotometry
  • Pre-analytical Factors

    ClassificationPre-Analytical FactorValue(s)
    Biospecimen AcquisitionMethod of tissue acquisition
    Biopsy
    Surgical resection
    Biospecimen PreservationType of fixation/preservation
    Formalin (buffered)
    Frozen
    RNAlater
    Analyte Extraction and PurificationAnalyte isolation method
    MagNA Pure DNA and Viral NA SV kit (automated)
    QIAamp DNA Mini Kit (manual)
    PCR SpecificLength of gene fragment
    100 bp
    200 bp
    300 bp
    400 bp
    Bisulfite conversion assay SpecificTargeted nucleic acid
    MAL
    SFRP1
    SFRP2