Postmortem Interval Determination: A Metabolomics / Lipidomics Approach

As submitted by the proposer: The postmortem changes that occur in a corpse are complex, involving metabolic changes in the corpse, in the endogenous microbiome, and in exogenous invasive microbes. While a number of biochemical parameters have been used to estimate short postmortem intervals (PMIs), limited efforts have been applied to investigating longer PMIs. This proposal aims to further study and validate our preliminary observations that time-dependent decrements in the structural glycerophospholipids of skeletal muscle may offer a new forensic tool for providing accurate PMI estimates over a longer postmortem interval (Wood and Shirley, 2013). While previous studies have evaluated biomarkers of short-term PMI and focused on small molecular metabolites not specific to the human corpse, our evaluation of muscle plasmalogens is the first study to focus on biomarkers of longer term PMI that are specific to the corpse and not invading microbes. The primary goal is to provide the forensic community with validated equations that will produce accurate PMI estimates based on biomarkers of long postmortem intervals. A multidisciplinary research team of biochemists, forensic anthropologists, and pathologists will make this project possible: the Metabolomics Unit at Lincoln Memorial University, the University of Tennessee Forensic Anthropology Center (FAC), and the Regional Forensic Center (RFC). Advances in mass spectrometry have enabled the design of high-resolution mass spectrometry lipidomics analytical platforms that can analyze over 700 individual lipids from 26 lipid subclasses. Utilizing this technology we will sample more time points and cadavers to extend our pilot studies that have demonstrated that muscle plasmalogens, a class of lipids not synthesized by invading bacteria, decline over a long PMI (Wood and Shirley, 2013). This study employs a three-tier sampling strategy: (1) the FAC will provide serial tissue samples from 20 bodies placed during Year 1 of the grant; (2) the RFC will collect cross-sectional data from 100 bodies during Year 1; (3) the RFC will sample 50 cases for validation during the first 4 months of Year 2. LOWESS regression will be used to visualize the data and evaluate if linear regression is appropriate. In the event that it is not, multivariate adaptive regression splines (MARS) and/or regression trees will be used to develop protocols for PMI estimation. The proposal is designed to maximize sample sizes over the two-year period while allowing for adequate time to complete grant reporting requirements and deliverables (semi-annual progress reports, the final technical report, abstracts and presentations at scientific meetings, and peer-reviewed publications). ca/ncf