Structural proteins such as collagens are well preserved in mineralised tissues such as bones, teeth, and ivory, and can remain a source of taxonomic information long after the degradation of DNA. For this reason, the analysis of ancient proteins can be used to enable species identification of morphologically unidentifiable objects. This presentation will describe a study of ivory objects from the collection of the Metropolitan Museum of Art. Here, the question from the curators was to identify the species from which the ivory was sourced, but very little sample was available for analysis.
A bottom-up proteomics protocol, based on filter-aided digestion, was minimised and simplified to prevent sample loss and contamination. Samples were taken by sequential rubbing microabrasive films on the ivory surface, removing a small amount of “dust” from the object surface. Hippopotamus and elephant references were obtained by the American Museum of Natural History, and the validated method applied to speciate objects from the Medieval and Egyptian collections at the Metropolitan Museum of Art. Data acquisition methods using an Orbitrap Eclipse mass spectrometer were adapted for low sample amounts, including the development of methods targeting species-specific peptides, and MS3 methods performing I/L discrimination. A unique data analysis approach was used focusing on the high collagen heterogeneity and allow for species ID. These new methods are essential to establish minimally invasive sampling with microabrasive films as a standardised procedure both in the context of cultural heritage and forensic investigations.
Since the species considered in this study (namely hippopotamus and elephant) have been understudied using proteomics, modern exemplar samples were used to validate the collagen sequences published on public databases, and to build home-made proteomics-based databases, which were then used to develop targeted data acquisition methods.
In the analysis of samples from the museum, in many cases several hundred peptides attributable to collagen type-I were identified, with 70% collagen coverage from the studied objects up to 3200 years old, allowing a confident species identification based on the identification of several peptides corresponding to informative sections of the collagen sequence with different modifications and unspecific/missed cleavages, and with y- and b-fragment ions over the full sequence.
Finally, a targeted MS3 methodology was developed for I/L discrimination, where in some cases only these two amino acids can differentiate two species. Informative portions of the collagen sequence containing I/L were identified and several peptides with different modifications or cleavages were targeted to obtain an unequivocal assignment.
https://zoom.us/j/98578019100?pwd=VkI3Vk5pVUY1amlCR0ZCV0hqNFQ5dz09
