Extinct species identification from late middle Pleistocene and earlier Upper Pleistocene bone fragments and tools not recognizable from their osteomorphological study by an enhanced proteomics protocol

Abstract

Ancient preserved molecules offer the opportunity of gaining a deeper knowledge on their biological past. However, the development of a proteomic workflow remains a challenge. The analysis of fossils must involve a low quantity of material to avoid damaging the samples. In this study an enhanced proteomic protocol was applied to 5‐milligram samples of about 130,000‐year‐old mammalian bones ranging from the end of the Middle Pleistocene up to the earlier Upper Pleistocene, excavated from Scladina Cave (Sclayn, Belgium). Using sequence homology with modern sequences, a biological classification was successfully achieved and the associated taxonomic ranks to each bone were identified consistently with the information gained from osteomorphological studies and palaeoenvironmental and palaeodietary data. Amino acid substitutions on collagens were identified, thus providing new information on extinct species sequences and helping in taxonomy‐based clustering. Considering samples with no osteomorphological information, such as two fragments of bone retouchers, proteomics successfully identified the families providing paleontologists new information on these objects. Combining osteomorphology studies and amino acid variations identified by proteomics, one of the retouchers was potentially identified as belonging to the Ursus spelaeus species.

Robust High-Throughput Proteomics Identification and Deamidation Quantitation of Extinct Species up to Pleistocene with Ultrahigh-Resolution MALDI-FTICR Mass Spectrometry

“…33 To avoid the induction of chemical modifications to the proteins and artifactual changes, we used TFA instead of hydrochloric acid and a short incubation time of 4 h instead of the common overnight protocol during the demineralization step. 16 The bone powder was washed with ABC. Collagen was subsequently extracted under mild denaturation conditions at 65 °C.…”

“…More sensitive but time-consuming LC−MS/MS studies required also a few milligrams. 15,16 Comparison of ZooMS and Protein Plate Digestion. The demineralization solution and bone powder fraction of eight modern samples (1−8) and eight Caours samples (99, 101, 102, 8, 18, 26, 52, and 58) were analyzed with the ZooMS standard protocol and the protein plate digestion described here.…”

“…20,42,45,46,54,55 The deamidation value from the same period may vary among publications. 6,16,38,56 This difference can be explained also by the various experimental conditions, from demineralization time to buffer composition and protein extraction procedures, inducing variable deamidation. 41 Nevertheless, for the very homogeneous and well-preserved bones of the Caours site, the average deamidation value found by MALDI-FTICR was 44.3% with a standard error of ± 8.4% (Table S1, Supporting Information S4).…”

“…Liquid chromatography−mass spectrometry/mass spectrometry (LC−MS/MS) analyses were performed on an Orbitrap Q Exactive plus mass spectrometer hyphenated to a U3000 RSLC Microfluidic HPLC system (ThermoFisher Scientific, Waltham, Massachusetts, USA) as described previously. 16 Bioinformatics for MALDI Data. MS raw data from MALDI-TOF were processed using Data Explorer 4800.…”

“…If ZooMS has a high probability of success, its destructive nature is undesirable when performing analytical studies of complete or rare bone artifacts. We describe here an improved sampling protocol yielding identifications from much reduced amounts down to 1 mg, similar to the sensitivity of nanoLC nanoESI analysis. , The second one is the absence of a high-throughput protocol. Indeed, 400 to more than 2000 bones were analyzed by the ZooMS method, using a classical in tube-preparation from bone quantities ranging from 15 to 50 mg. , The method was miniaturized using a low-protein-binding polystyrene 96-well plate.…”