Deep-time phylogenetic inference by paleoproteomic analysis of dental enamel

Abstract

In temperate and subtropical regions, ancient proteins are reported to survive up to about 2 million years, far beyond the known limits of ancient DNA preservation in the same areas. Accordingly, their amino acid sequences currently represent the only source of genetic information available to pursue phylogenetic inference involving species that went extinct too long ago to be amenable for ancient DNA analysis. Here we present a complete workflow, including sample preparation, mass spectrometric data acquisition and computational analysis, to recover and interpret million-year-old dental enamel protein sequences. During sample preparation, the proteolytic digestion step, usually an integral part of conventional bottom-up proteomics, is omitted to increase the recovery of the randomly degraded peptides spontaneously generated by extensive diagenetic hydrolysis of ancient proteins over geological time. Similarly, we describe other solutions we have adopted to (1) authenticate the endogenous origin of the protein traces we identify, (2) detect and validate amino acid variation in the ancient protein sequences and (3) attempt phylogenetic inference. Sample preparation and data acquisition can be completed in 3–4 working days, while subsequent data analysis usually takes 2–5 days. The workflow described requires basic expertise in ancient biomolecules analysis, mass spectrometry-based proteomics and molecular phylogeny. Finally, we describe the limits of this approach and its potential for the reconstruction of evolutionary relationships in paleontology and paleoanthropology.

Key points

Paleoproteomics has shown that it is possible to obtain useful phylogenetic information from dental enamel proteins up to 2 million years old. They are heavily fragmented and chemically modified, making their recovery and analysis challenging.

The protocol describes how to (1) extract million-year-old dental enamel protein remains while minimizing contamination, (2) sequence them using high-resolution tandem mass spectrometry and (3) attempt otherwise so far impossible molecular-based phylogenetic inference.