A Proteomics Method for Presumptive Identification of Human Tissue

Background: The positive identification of a source of tissue as human plays an important role in various contexts. It is particularly important for investigations concerning tissue and organ trafficking, since unequivocal confirmation is required for legal proceedings involving such cases. While deoxyribonucleic (DNA) methods are considered the gold standard for tissue identification, issues such as degraded DNA or the presence of chemical preservatives can hinder performance and positive identification using DNA techniques. Objectives: The aim of this study was to develop a simple method for presumptive identification of human tissue using standard bottom-up proteomics data. Methods: We identified proteins isolated from human kidney, lung and spleen tissues by bottom-up proteomics and database search using Proteome Discoverer and Sequest HT algorithms. The list of identified proteins was sorted based on liquid chromatography (LC)–mass spectrometry (MS) data metrics such as the number of unique peptides used to identify each protein and the % sequence coverage of an identified protein to determine if any parameter would cluster proteins annotated as human in a distinct category. We found that eliminating proteins identified with fewer than two unique peptides and those with less than 5% sequence coverage resulted in a final list where at least half of the remaining proteins are annotated as human. We applied this data filtration process to blinded LC–MS/MS data from 26 previous experiments to assess accuracy. Results: Using bottom-up proteomics data and the filtration rules established, we identified tissue samples (n = 10), including kidney, spleen, lung, formalin-fixed paraffin-embedded uterus, frozen breast tissue, dry blood and dry saliva as human, and tissue (n = 16) from rat, mouse, bovine, and sheep as non-human, resulting in 100% sensitivity and specificity. Conclusions: The results demonstrate that the list of identified proteins following a standard bottom-up proteomics experiment could be filtered and potentially used as a fast and simple method for presumptive human tissue identification.