An assessment of the impact of environmental factors on the quality of post-mortem DNA profiling.

Abstract

DNA profiling has ignited public interest and consequently their expectations for the capabilities of forensic criminal and science investigations. The prospect of characterising the genetic makeup of individuals or trace samples from a wide variety of depositional and post-mortem circumstances raises the question of how reliable the methods are given the potential for prolonged exposure to variation in environmental factors, i.e. temperature, pH, UV irradiation and humidity, that are known to induce damage to DNA. Thus, it is crucial to verify the validity of the DNA profiling for characterising the genetic makeup of post-mortem tissues. This project aimed to assess the reliability of sequence and microsatellite based genotyping of tissues (muscle, hair and bone) sampled from carcasses over a two year post-mortem period. This assessment investigated the impact of environment induced DNA degradation in the local geographic region that is typical of the circumstances that confront forensic practitioners in southern Australia and to utilise rigorous controls by studying animals whose time of death and burial was known and for which we had pre-decay tissue samples available. A ‘body farm’ with 12 pig carcasses on the northern Adelaide plains, ~60km north of Adelaide, which has a typical southern Australian Mediterranean climate, i.e. cold wet winters and hot dry summers. Pigs (Sus scrofa) were used as an experimental analogue for human subjects because of the logistical and ethical reasons. The pig carcasses were allocated among three treatments: four were left on the surface, four were buried at 1m depth, and four were buried at 2 m depth. These ‘burial’ conditions mimic a range of conditions encountered typically in forensic and archaeological studies. Cortical bone samples were taken from each pig carcass at one week, one month, three months, six months, one year and two years post-mortem and muscle and hair over the same sampling period for as long as those tissue types were present. A set of PCR primers to amplify two (short and a long) fragments from the hypervariable part of the mitochondrial control region (HVRI) that is used in forensic and evolutionary studies of humans and many other mammal species were developed. Also a panel of four pig microsatellite loci with fluorescent labels to facilitate automated multiplex genotyping. These loci matched as closely as possible the core motifs and allele lengths typical of the commercially available microsatellite marker kits used in Australian forensic science labs so that our experiments were as good a model as possible of the human forensic DNA technology. In this study it was possible to retrieve samples from muscle tissue up to 90 days, hair up to one year and bone at two years post-mortem. The analyses showed that the long and short HVRI region PCR fragments were only amplifiable up to 30 days from muscle tissue and that these fragments were amplifiable up to one year from hair. In contrast, in cortical bone both PCR fragments were amplifiable up to two years. The long fragment disappeared in muscle tissue completely after 30 days and in hair after six months. However, the long fragment was present in cortical bone even at two years. Overall, there was a general trend of loss of concentration of both the long and short fragments over time. Comparisons of the HVRI nucleotide sequences among tissues sampled from individual animals showed substitution changes in muscles as early as 30 days (3 out of 6 individuals) and hair at six months (1 out of 6 individuals). In contrast, in cortical bone substitutions first appeared at 365 days (1 out of 6 individuals). The most common substitution observed in all tissues types was the C-T transition, with A-G transversions observed in two episodes and C-A transversion observed in one episode. Analyses of microsatellite genotypes in muscle tissues showed high allele peaks on chromatograms up to day seven samples. However, by three months PCR was not successful from muscle tissue. While, bone tissue had lower allele peak heights compared to the muscle tissues, alleles were detectable up to six months. Allele drop out occurred for one animal (at 2 meters) in muscle tissue at the dinucleotide locus and for another animal (kept on surface) also in muscle tissue at a tetranucleotide locus. Stuttering was observed for a single animal at dinucleotide locus in muscle tissue (buried sample 2 meter depth). No stuttering or allele drop outs were seen in the bone tissue. Overall the four loci completely disappeared after 30 days in muscle tissue and after 180 days in bone tissue. In summary, analyses showed that post-mortem DNA degradation was present in all the three tissue types (muscle, hair and bone). The types of damage identified were DNA fragmentation, nucleotide substitutions and DNA loss, which resulted in a diminished frequency of successful PCR for mitochondrial and nuclear markers over time and stuttering and allele drop out in microsatellite genotyping. In addition, two nucleotide substitutions were concentrated in ‘hotspots’ that correlate with sites of elevated mutation rate in vivo. Also the frequency of successful PCR of longer nuclear and mitochondrial PCR products declined markedly more quickly than for shorter products. These changes were first observed at much shorter post mortem intervals in muscle and much longer post mortem intervals in hair and bone tissue. When considering the carcass deposition treatments, tissues that were retrieved from buried carcases showed higher levels of DNA degradation compared to tissues retrieved from carcases left on the surface. Overall, muscle tissue is a good source for DNA analysis in immediate post mortem samples, whereas hair and bone tissue are good source for DNA analysis from older samples. When comparing the microsatellite genotyping and mtDNA analyses, mtDNA is a reliable source for DNA analysis from tissue recovered from bodies that had decayed for longer post-mortem durations such as months to years, whereas microsatellite genotyping gives reliable results for tissue from shorter post mortem intervals (hours to few days). Therefore it is recommended that when analysing mtDNA sequences, cloning and sequencing PCR products can help to identify the base pair substitutions especially for tissue retrieved from longer post mortem intervals. In addition, increasing the template DNA concentrations and "neutralising" co-extracted DNA inhibitors should be considered when dealing with tissue from longer post mortem intervals. Finally, the more stringent protocols used in ancient DNA studies should be considered when dealing with tissue with much longer post mortem intervals in forensic settings.