Cancer Diagnosis Program | Biorepositories and Biospecimen Research Branch

Search BRD Papers

Comparison of enzymatic and bisulfite conversion of circulating cell-free tumor DNA for DNA methylation analyses.

Author(s): Kresse SH, Thorkildsen EM, Brandt-Winge S, Pharo H, Vedeld HM, Lind GE

Publication: Clin Epigenetics, 2025, Vol. 17, Page 93

PubMed ID: 40468374 PubMed Review Paper? No

Purpose of Paper

Conclusion of Paper

Studies

1. Study Purpose

   This study compared DNA recovery, fragment size and methylation analysis between enzymatic and bisulfite converted cfDNA from pooled plasma and cell line DNA; investigated the effects of using just the conversion module rather than the full kit; and assessed the effects of using different ratios and manufacturers of magnetic beads during the clean-up step. DNA recovery, fragment size and methylation results were also compared between enzymatic and bisulfite converted cfDNA from plasma collected from five patients with colorectal cancer with known BCAT1 methylation. Plasma was obtained from the EDTA blood of 8 healthy volunteers by centrifugation at 1600 g for 10 min, followed by 7000 g for 10 min and from the plasma of 5 colorectal cancer patients with known BCAT1 methylation by centrifugation at 1600 g for 10 min at 4 °C.  The plasma from healthy volunteers was pooled, and all plasma was subsequently stored at -80°C until DNA was isolated from the pooled plasma and the plasma from colorectal cancer patients using the QIAamp Circulating Nucleic Acid Kit. To compare conversion methods, 50 ng of cfDNA isolated from pooled EDTA plasma and 100 ng of fragmented DNA from the colorectal cancer cell line RKO were bisulfite converted using the EpiTect Plus Bisulfite Kit or enzymatically converted using the full NEBNext Enzymatic Methyl-seq Kit or the NEBNext Enzymatic Methyl-seq Conversion Module. To test the effect of magnetic bead type and ratio on DNA loss during enzymatic conversion, 20 ng fragmented RKO DNA was cleaned using five different bead brands (AMPure XP, Mag-Bind TotalPure NGS, NEBNext Sample Purification Beads, ProNex Size-Selective Purification System and SPRIselect) at a bead to sample ratio of 1.8x (normal) and 3.0x. Effects on DNA recovery of bead ratio were investigated during a second clean-up step during the enzymatic conversion using AMPure XP beads at a ratio of 1.0x, 1.8x, and 3.0x and cell-line DNA. The effect of bead ratio on cfDNA recovery during enzymatic conversion was tested using 20 ng of cfDNA from pooled plasma and the AMPure XP beads at a ratio of 1.8x and 3.0x in the first clean-up and a ratio of 1.0x, 1.8x, and 3.0x during the second clean-up. The efficiency of the original enzymatic conversion protocol was compared to bisulfite conversion using cfDNA isolated from five colorectal cancer patients with known BCAT1 methylation. Fragment length was analyzed using the Cell-free DNA ScreenTape Assay and a 4200 TapeStation Instrument and the RNA 6000 Pico Kit with a 2100 Bioanalyzer Instrument.  Conversion was detected by ddPCR amplification of chromosome 3 (unconverted DNA), MYOD1 (converted DNA), CD01, and a 4Plex assay (EPHA3, KBTBD4, PLEKHF1 and SYT10).

   Summary of Findings:

   The conversion efficiency with the bisulfite conversion EpiTect Kit was 100% for both cfDNA from plasma and cell line DNA. The conversion efficiency when the full kit versus only the conversion module of the enzymatic conversion NEBNext Kit was 99.6% and 99.9%, respectively, for plasma cfDNA and 99.2% and 97.1%, respectively, for cell line DNA. The DNA fragment length was longer for enzymatically converted DNA compared to bisulfite converted DNA, as enzymatically converted DNA had slightly longer fragments in each peak, even after compensating for adapters and loss of many of the small fragments when enzymatically converted.  DNA recovery for cfDNA from plasma and cell-line DNA was superior with the bisulfite conversion method (68% and 51%, respectively) compared to the enzymatic conversion method using either the full kit (38% and 21%, respectively) or the conversion module alone (30% and 5%, respectively). The average normalized CDO1 methylation level was highest when DNA was converted with the enzymatic conversion module alone (113), followed by the full enzymatic conversion kit (73), and was lowest with bisulfite conversion (61), but there were many fewer positive control and target droplets when enzymatic conversion was used compared to bisulfite conversion; the difference was even more pronounced when the conversion module alone was used. Increasing the ratio of magnetic beads during the clean-up step of enzymatic conversion from 1.8 to 3.0 improved DNA recovery by 9-17%, depending on bead manufacturer. The highest DNA recovery was with beads from AMPure XP, followed by Mag-Bind TotalPure NGS, NEBNext, and ProNex beads, and the lowest recovery occurred when SPRI beads were used. In a second step clean-up step with AMPure XP beads, the recovery improved from 22% when the recommended 1.0x bead ratio was used to 52% and 59% when 1.8x and 3.0x beads were used, respectively. Importantly, the higher bead ratio increased the retention of smaller fragments. For plasma cfDNA, the recovery during the first clean-up was higher when a bead ratio of 3.0x was used compared to 1.8x(82% versus 79%); the recovery during the second clean-up was highest when a bead ratio of 3.0x was used followed by 1.8x beads and 1.0x beads (87%, 86% and 73%, respectively), but there was no change in fragment size. Using 1.8x beads for both clean-up steps instead of the kit-specified 1.8x for the first and 1.0x for the second slightly, decreased the DNA recovery for the full enzymatic conversion kit from 38% to 36%, but slightly increased the DNA recovery from 30% to 33% when the enzymatic conversion module was used alone. Regardless of bead ratio, the conversion efficiency was 99.7-99.9% and the fragment length profile was unaffected. Based on these results, the authors chose to compare the full enzymatic conversion kit using the original bead ratio to bisulfite conversion using specimens from colorectal cancer patients. The conversion efficiency was slightly higher (but was still ≥99.0% for all five specimens) and the fragment sizes were slightly shorter when the bisulfite conversion kit was used rather than the enzymatic kit. Although using the bisulfite conversion rather than the enzymatic conversion led to higher DNA recovery (61-81% versus 34–47%) and more positive droplets for both the target and control genes, BCAT1 methylation was detected at similar levels (0.6–28.8 for bisulfite conversion and 0.7–36.8 for enzymatic conversion) in the cfDNA of 4 of the 5 patients. 

   Biospecimens

   • Bodily Fluid - Plasma

   Preservative Types

   • Frozen

   Diagnoses:

   • Normal
   • Neoplastic - Carcinoma

   Platform:

   Analyte	Technology Platform
   DNA	Bisulfite conversion assay
   DNA	Automated electrophoresis/Bioanalyzer
   DNA	Digital PCR

   Pre-analytical Factors:

   Classification	Pre-analytical Factor	Value(s)
   Bisulfite conversion assay Specific	Technology platform	EpiTect Plus Bisulfite Kit NEBNext Enzymatic Methyl-seq Kit NEBNext Enzymatic Methyl-seq Conversion Module only
   Bisulfite conversion assay Specific	Template modification	AMPure XP beads Mag-Bind TotalPure NGS beads NEBNext Sample Purification beads (included in the kit) ProNex Size-Selective Purification System SPRI select beads 1.0x beads to sample 1.8x beads to sample 3.0x beads to sample 1.8x beads to sample for both steps 1.8x beads to sample for first step and 1.0x beads to sample for second step

   View more