Forensic Species Identification Using Phylogenetic Proteomics

Wildlife trafficking is a global issue with devastating ecological effects and synergistic relationships with other forms of international illicit networks. Furs and pelts are a major component of this trade and are more practical to transport. Proper identification of such items is important to be able to prosecute and punish poachers. Wildlife forensic investigators depend on reliable tools and methods to either include or exclude illegal versus legal species. However, due to the chemically and physically harsh production process, DNA-based methods often fail because the sample DNA is degraded. Morphological-based methods have poor species resolution, are time-consuming, and are dependent on a limited number of highly trained individuals. However, these challenges may be by-passed by focusing on the phylogenetic information in protein. DNA changes between closely related species are reflected at the protein level, allowing the phylogenetic information between species to lead to a species-level identification. Proteins are chemically more stable than DNA. Therefore, proteomic analysis of heavily processed items may provide an alternative, and robust, method of species identification. However, determining the amino acid sequences in mass spectrometry data requires the precise sequences to be present in a reference protein database. The hypothesis is that the protein sequences, or reference proteome, of a given species would match more than proteomes from other species and be more efficient at identifying peptide sequences in fur digests from the same species. This research focuses on the phylogenetic relationship between six cat species, lion (Panthera leo), leopard (Panthera pardus), tiger (Panthera tigris), cheetah (Acinonyx jubatus), Canada lynx (Lynx canadensis), and domestic cat (Felis catus); of which the first four had good-quality, morphologically-identified materials that were selected for analysis. The remaining two species, Canada lynx and domestic cat, were included in the phylogenetic comparison using their reference proteome databases. The selected fur or pelt reference material items were digested based on a previously-published method for protein extraction from human hair and analyzed by LC-MS/MS and PEAKS DB software. Species identification was achieved at three different levels: total peptide yield, protein coverage patterns, and presence of peptides containing species-specific markers. For each item sampled, the maximum number of total peptides identified using the PEAKS scoring algorithm was obtained using the corresponding theoretical database. For example, the raw peptides generated from the lion paw pelt yielded 13928 total identified peptides using the lion database. As evolutionary distance between the lion and related felid species increases, the total peptide yield decreases. Using the leopard, tiger, Canada lynx, domestic cat, and cheetah databases, the total peptide yields were less - 13766, 13758, 12604, 12307, and 11774, respectively. Similar results were obtained with the leopard pelt, tiger pelt, and cheetah coat sampled, demonstrating how species identification can be achieved through comparison of total peptide yields. Protein coverage patterns also indicated species of origin. Many keratin and keratin-associated proteins were identified from the hair and skin samples, including, but not limited to, Keratin 14, Keratin 16, Keratin 32, Keratin 35, Keratin 38, Keratin 75, Keratin 82, Keratin 85, Desmoplakin, and Corneodesmosin. Coverage patterns demonstrated the highest percent coverage when analyzing the raw peptides from a sample with its corresponding database. For example, the raw peptides generated from the lion sample that matched with Keratin 35 resulted in 84% coverage using the lion database, while the tiger, leopard, domestic cat, Canada lynx, and cheetah databases resulted in 77%, 75%, 77%, 75%, and 65%, respectively. Species-specific marker locations were identified by protein alignments between the sequences of all six species. Peptides containing these species-specific markers were then identified in the raw data and found to be present only in the samples that correlated with the species of origin. This paper demonstrates four individual examples of species-specific markers within Keratin 35 corresponding to the respective species of origin for all four sampled items: lion, leopard, tiger, and cheetah. In summary, the three approaches used in this project – the total peptide approach, the protein coverage pattern approach, and the species-specific peptide approach – individually demonstrated that species identification is plausible using phylogenetic proteomics. This data demonstrates the robustness and reliability of proteomic approaches to species identification from fur and lays the foundation for future research, such as the development of targeted proteomic assays, for broader application to wildlife forensic casework.

Fetched live from OpenAlex and de-inverted. Abstracts are not stored in this database: the inverted indexes are 8.6 GB of the frame’s 9.3 GB of text, and the host has 13 GB free.