Development and validation of a D-loop mtDNA SNP assay for the screening of specimens in forensic casework

Abstract

Mitochondrial DNA (mtDNA) analysis is usually a last resort in routine forensic DNA casework. However, it has become a powerful tool for the analysis of highly degraded samples or samples containing too little or no nuclear DNA, such as old bones and hair shafts. The gold standard methodology still constitutes the direct sequencing of polymerase chain reaction (PCR) products or cloned amplicons from the HVS-1 and HVS-2 (hypervariable segment) control region segments. Identifications using mtDNA are time consuming, expensive and can be very complex, depending on the amount and nature of the material being tested. The main goal of this work is to develop a less labour-intensive and less expensive screening method for mtDNA analysis, in order to aid in the exclusion of non-matching samples and as a presumptive test prior to final confirmatory DNA sequencing. We have selected 14 highly discriminatory single nucleotide polymorphisms (SNPs) based on simulations performed by Salas and Amigo (2010) [1] to be typed using SNaPShotTM (Applied Biosystems, Foster City, CA, USA). The assay was validated by typing more than 100 HVS-1/HVS-2 sequenced samples. No differences were observed between the SNP typing and DNA sequencing when results were compared, with the exception of allelic dropouts observed in a few haplotypes. Haplotype diversity simulations were performed using 172 mtDNA sequences representative of the Brazilian population and a score of 0.9794 was obtained when the 14 SNPs were used, showing that the theoretical prediction approach for the selection of highly discriminatory SNPs suggested by Salas and Amigo (2010) [1] was confirmed in the population studied. As the main goal of the work is to develop a screening assay to skip the sequencing of all samples in a particular case, a pair-wise comparison of the sequences was done using the selected SNPs. When both HVS-1/HVS-2 SNPs were used for simulations, at least two differences were observed in 93.2% of the comparisons performed. The assay was validated with casework samples. Results show that the method is straightforward and can be used for exclusionary purposes, saving time and laboratory resources. The assay confirms the theoretic prediction suggested by Salas and Amigo (2010) [1]. All forensic advantages, such as high sensitivity and power of discrimination, as also the disadvantages, such as the occurrence of allele dropouts, are discussed throughout the article. © 2013 Elsevier B.V.