Preferential Amplification of Apoptotic DNA from Plasma: Potential for Enhancing Detection of Minor DNA Alterations in Circulating DNA

Tumors release genomic DNA into the circulation of cancer patients after cellular necrosis and apoptosis. Isolation of the apoptotic fraction of plasma-circulating DNA can enhance detection of low-level mutations that can serve as tumor biomarkers (1). Because the amount of DNA circulating in the plasma of cancer patients is low, on the order of a few nanograms per milliliter of blood, the number of genes that can be examined for tumor-specific alterations is limited, a situation that reduces biomarker sensitivity. We recently applied whole-genome amplification of plasma-circulating DNA to increase the number of targets that can be analyzed from each sample, thus potentially increasing biomarker sensitivity(2). This approach yields highly-expanded DNA amounts for performing genetic screening; however, there is no preferential enrichment of smaller sized DNA fragments. We report a new method for whole-genome amplification of plasma-circulating DNA, based on ligation-mediated PCR of blunted DNA fragments (BLM-PCR)1 , which results in preferential amplification of smaller size, apoptotic DNA fragments. Plasma-circulating DNA was extracted from blood obtained from radiation therapy patients after the patients gave informed consent and the study received institutional review board approval. Within 2–3 h of collection, whole blood was centrifuged at 2000g for 15–30 min, plasma was separated, and plasma-circulating DNA was purified by use of a QIAamp™ MinElute Virus Spin Kit (Qiagen) and quantified via Taqman real-time-PCR. To test for v-Ki-ras2 Kirsten rat sarcoma viral oncogene homolog (KRAS) codon 12 mutations, we used a simplified version of the recently reported fluorescent amplicon-generation method (3), which employs the highly thermostable PspGI enzyme to destroy wild-type alleles during PCR. Of the 15 plasma samples studied, plasma samples from 3 patients (#5, #6, and #15) were positive for KRAS mutations when unamplified DNA was used. Tumor-derived DNA from these patients also contained the plasma-identified KRAS nucleotide changes. These KRAS-positive samples (#5, #6, and #15) plus 2 KRAS-negative samples (#12 and #13) were chosen for further study. To apply BLM-PCR to these plasma-circulating-DNA samples, we generated blunt ends on 2–5 ng plasma-circulating DNA using 0.6 U T4 DNA polymerase at 12 °C for 15 min in 5 μL ligase buffer supplemented with dNTP at a final concentration of 100 μmol/L. T4 DNA polymerase was then heat inactivated. Double-stranded adaptors were prepared by annealing the following primers at 55 °C: 5′-TTCCCTCGGATA-3′ and 5′-AGGCAACTGTGCTATCCGAGGGAA-3′. 5 μL of blunted DNA and 0.8 μL adaptors were then ligated via T4 DNA ligase. 60-μL Final reagent concentrations of the PCR reactions were as follows: 1X GoTaq Flexi™ buffer, 1.5 mmol/L MgCL2, 2 mmol/L each dNTP, 0.2 μmol/L 24-mer primer, and 6 μL adaptor-ligated product. The reaction was incubated at 72 °C for 3 min followed by rapid cooling on ice. 1.3 U GoTaq Flexi DNA polymerase (Promega) was used. PCR-cycling was: 72 °C, 5 min; 95 °C, 2 min; (95 °C, 15 s; 72 °C, 15 s) × 25 cycles. When we examined the BLM-PCR product via gel electrophoresis, a pattern of discrete DNA sizes of approximately 200 and approximately 400 bp was observed uniformly for all samples (Fig. 1A ). Plasma-circulating DNA consists of a mix of small fragments consistent with apoptotic DNA the size of mono- or dinucleosomes and of large DNA fragments (1)(4)(5). As we demonstrated using larger size DNA, the conditions applied for BLM-PCR preclude amplification of DNA in excess of a few hundred base pairs. We observed a similar ladder-like pattern when we amplified plasma-circulating DNA from healthy volunteers via BLM-PCR (data not shown). When BLM-PCR samples were tested for KRAS mutations, we found that KRAS mutations were absent for samples #12 and #13 (no PCR product in the presence of PspGI), but these mutations were clearly present in samples #5, #6 and #15 (Fig. 1B ). Two additional methods of whole-genome-amplification were also applied to the 5 plasma-circulating DNA samples, BL-whole-genome amplification (which amplifies all DNA fragment-sizes), and multidisplacement amplification (which amplifies only the large DNA fragment-sizes)(2); KRAS-mutation detection was also repeated. The KRAS mutation load (fraction of DNA product containing KRAS mutations) was semiquantitatively estimated by measuring the fraction of endpoint PCR product surviving PspGI digestion. For this purpose, samples were amplified either in the presence or in the absence of PspGI, and relative PCR product was quantified via denaturing HPLC. After BLM-PCR, results for the 3 samples that were KRAS-mutation positive demonstrated an increased KRAS mutation load relative to results obtained with other methods of amplification or relative to unamplified plasma-circulating DNA, indicating an enhanced ability to detect KRAS mutations (Fig. 1C). Amplification of short (apoptotic) DNA fragments from plasma and KRAS mutation detection. Panel A. Blunt-ended ligation-mediated PCR of plasma-circulating DNA (BLM-PCR). This 3-step procedure involves (a) blunting of the ends of double-stranded plasma-circulating DNA using T4 DNA polymerase, (b) ligation of blunt-ended adaptors, and (c) PCR amplification. The DNA amplicons resulting from amplification of DNA from samples #12, #13, #5, #6, and #15 is indicative of a DNA-ladder–like pattern consistent with the amplification of mono- and dinucleosomal DNA. Panel B. Detection of KRAS mutations by the fluorescent amplicon generation assay on BLM-PCR–amplified plasma-circulating DNA from patients #12, #13, #5, #6, and #15, using gel electrophoresis. All samples generate PCR product in the absence (−) of PspGI enzyme; however, only samples with KRAS mutations generate products in the presence (+) of PspGI. Samples #5, #6, and #15 present a substantial amount of residual PCR product in the presence of PspGI. Normal human male DNA was used as a negative control. Panel C. Comparison of KRAS mutation load (percentage PCR product surviving PspGI digestion) in plasma-circulating DNA amplified using 3 different whole-genome amplification methods, and in unamplified DNA. The KRAS mutation load in patient #5 is below detection level following multidisplacement amplification (MDA) (which amplifies large DNA fragments) or BL–whole-genome amplification (WGA) (which amplifies all DNA fragments). BLM-PCR–amplified DNA results in a significant increase in the KRAS mutation load in all 3 samples that are KRAS positive in unamplified plasma-circulating DNA (#5, #6, and #15). Normal human male DNA was used as a negative control. The histograms depict the mean and SD of 3 independent experiments, starting from unamplified plasma each time. In summary, BLM-PCR provides selective amplification of apoptotic DNA and potentially a method to improve identification of rare alleles in plasma-circulating DNA while also providing ample DNA for testing an almost unlimited number of biomarkers for monitoring cancer patients. This technique is anticipated to be equally applicable to other applications such as prenatal diagnosis. Grant/Funding Support: This work was supported in part by NIH grants CA-115439 and CA-111994, by NIH training grant 5 T32 CA09078 (JL), and by the Joint Center for Radiation Therapy Foundation. Financial Disclosures: None declared.