WGA has become an important tool to generate additional DNA when working with limited biological sample. WGA is especially pertinent for genetic susceptibility studies of pediatric cancer, when buccal brushes may be the preferred sample collection method due to the age or ability of the child. Recent studies have applied WGA to buccal brush DNA using the improved primer extension preamplification (I-PEP)-PCR method. Although DNA has been successfully amplified, there have been reports of discrepancies between WGA DNA and non-WGA DNA. We applied two commercially available kits to proteinase K-digested DNA isolated from buccal brush samples from 14 individuals and to 8 buccal brush samples directly to evaluate genotypes for four selected single nucleotide polymorphisms (SNPs) including: excision repair cross-complementing 2 (ERCC2) (Lys751Gln), uncoupling protein-3 (UCP3) (C-55T), paraoxonase 1 (PON1) (Gln192Arg), and methionine synthase (MTR) (A2756G). Selected SNPs were low to medium in GC content (36-57%) and none occurred in the telomere. For the 14 samples, an aliquot of purified DNA was set aside (non-WGA DNA), while another was diluted to 20ng/μl and processed as recommended for WGA. For the 8 samples, WGA was applied directly to the brush. Following WGA, samples were purified and resuspended. DNA concentrations for the 14 WGA reactions ranged from 169-357ng/μl compared to 30-1735ng/μl for non-WGA DNA. For the 8 brush samples applied directly to the WGA reactions, concentrations ranged from 5-895ng/μl. Samples were diluted for use in genotyping assays using two platforms: TaqMan polymerase chain reaction and restriction-fragment length polymorphism. Results were verified using direct sequencing. A range of discrepancies was found for the buccal brush WGA DNA compared to the non-WGA DNA for 3 of 4 SNPs tested (all samples matched for PON1, while 3/14 (21%) conflicted for ERCC2). Similar discrepancies were observed with WGA applied directly to buccal brushes. Importantly, discrepancies in WGA DNA samples manifested as a perceived loss of an allele informative for a heterozygote individual. Repeat WGA produced similar results. Given the discrepancies observed through repetitive testing using different platforms, we recommend caution in using commercially available WGA kits for SNP assays using buccal brush DNA. This work was supported by NIH R01 CA79940 and the Children’s Cancer Research Fund.

[Proc Amer Assoc Cancer Res, Volume 46, 2005]