NEW YORK (GenomeWeb) – The use of high temperature to fragment genomic DNA prior to digital PCR analysis has been shown to induce mutations that are detected as false-positive results for some rare alleles, according to newly published research.
Digital PCR demands fragmentation of DNA in order to ensure uniform droplet formation. The methods used to accomplish this include restriction digestion, microwaving irradiation, biopolymer-based shredding, acoustic shearing, and treatment with high heat.
While there isn't any data on what proportion of researchers use each method, Muneesh Tewari, senior author on the study and a professor at the University of Michigan Health System, told GenomeWeb in an interview that users new to dPCR might consider heat fragmentation because of its low cost and ease.
"The point of our paper was to bring awareness to a potential pitfall of heat fragmentation of DNA for rare allele detection of dPCR," he said.
In the study, published earlier this month in BioTechniques, the researchers heated genomic DNA purchased from a vendor to 95° C for more than 10 minutes prior to quantification with RainDance Technologies' RainDrop Digital PCR System, which led to increased detection of G12D KRAS mutations.
This effect was not seen using sonication or when examining BRAF V600E mutations. The researchers also found that recovery efficiency of heat-fragmented genomic DNA was higher than expected, suggesting single-stranded DNA was also incorporated into droplets.
"Testing was done with and without heat shearing, so we can be fairly certain that the increase in mutation counts associated with heat treatment arises from the heating, and not from pre-existing mutations in the DNA," Tewari noted.
The authors suggested that high temperature could be causing deamination of cytosine to uracil. They designed other KRAS mutation assays with different point mutations and found evidence that supported this hypothesis, which was also substantiated by a decades-old paper in Biochemistry showing high heat can cause deamination of cytosines in DNA.
First author Qing Kang also said in an email that the group focused on the impact of heat fragmentation on dPCR specifically, but there remains "a concern whether heat would have a broader impact on the quality of DNA samples used in multiple settings."
In the BioTechniques study, the researchers also examined the effect of heat incubation performed during genomic DNA extraction from blood.
In these experiments they found that moderate heating, such as is typically used for DNA extraction, did not cause any mutagenic effects as measured by dPCR.
With increasing sensitivity of sequencing, however, it may be important to study whether steps involving heating could be affecting results.
"During sequencing library preparation, although heating is not routinely used for the purpose of DNA fragmentation it is commonly used in steps such as DNA denaturation and enzyme inactivation, which could increase the DNA mutation background that is not originally present in the DNA samples," Kang said.
She noted the group had not experimentally tested this, but, knowing the potential mutation-causing factors during library prep could be crucial for some analyses. "We think the mutagenic effect of heating could be applied to any other G-to-A mutation detection," she said.
The group is now using both restriction enzyme digestion and a shearing method from Covaris, "which work equally well" for DNA fragmentation, Kang said.
They have not tried these experiments with other digital PCR platforms, but Tewari predicted that similar results would be obtained, "because the heat-induced changes in the DNA happen before the digital PCR step."