Assessing Mesolithic hunting strategies using ZooMS and zooarchaeology at La Grande Rivoire rockshelter (Isère, France)

Out of the total remains analysed , 196 were selected for ZooMS analysis. The selection of samples was based on faunal that had previously undergone archaeozoological and taphonomic analysis. Most of them are identified at the anatomical level and attributed to the mammal-size class 1 and 1-2. Priority was also given to various bone types (cancellous and cortical). Out of them, 64 remains of foetal and infant individuals, whose taxonomic identification is often challenging due to fragmentation and lack of typical morphological characteristics, were also selected. The selection was limited by issues related to burning, as the analysis employed cannot yet be applied to bones that have been altered by fire. This limitation is even more pronounced at La Grande Rivoire, as the coloration of numerous bones remains exhibits hues similar to those produced by heating, making it impossible to distinguish between the two processes. This coloration is due to the presence of organic matter in the sediments (Brochier 2017). Protein digestion on plate. Bone proteins were analyzed by Bray et al. 2023 method . All solutions were eluted through the wells using a vacuum manifold (Merck KGaA, Darmstadt, Germany) pumped by a DS 102 rotary vane pump (Agilent, Santa Clara, USA). The wells of a 0.45 μm hydrophobic-high protein binding Immobilon-P membrane plate (MSIPS4510, Millipore, Billerica, MA, USA) were prepared by washing them 3 times with 100 µL of 70% ethanol flowed through the filter. In each well, 5 mg of bone powder was deposited and 100 µL of demineralization solution (5% v/v TFA (trifluoroacetic acid), pH 0.8) were added. The plate was incubated at 4°C for 16 h without shaking. The demineralization solution in TFA was recovered in a V-bottom well collecting plate by applying vacuum. Then 6 µL of NaOH 6 M were added to neutralize the TFA and 100 µL of 100 mM ABC (ammonium bicarbonate) pH 8.8. The plate was kept at 4 °C for digestion. The bone powder in each well was washed three times with 100 µL of digestion buffer (50 mM ABC, pH 8.8). The plate was incubated at 65 °C for 1 h on a heating stirrer MHR23 (Hettich, Tuttlingen, Germany) for collagen gelatinization. The digestion of bone powder and demineralization solution was performed by adding 0.5 μg of sequencing grade trypsin (Promega, Madison, WI, USA) and incubating the mixture at 37 °C for 16h while shaking on a heating stirrer MHR23. Peptides from bone powder were recovered in a V-bottom well collecting plate by applying vacuum and the plate was washed one time with elution buffer (50 mM ABC pH 8.8). Tryptic peptides from bones powder and demineralization solution were acidified with 1 µL of acetic acid (final concentration 0.5% of acetic acid). Tryptic peptides were desalted on 96 wells plates C18 (Affinisep, Petit-Couronne, France). Briefly, the plate was washed two times with 200 µL of ACN (acetonitrile) followed by a washing step for 2 successive times with 100 µL of 80% ACN, H2O 0.5% acetic acid and another washing repeated for 3 times with 100 µL of H2O 0.5% acetic acid. Tryptic peptides form both solutions were transferred to C18 96 wells plate and eluted with vacuum manifold. The plate was washed 6 times with 100 µL of H2O, 0.5% acetic acid. Peptides were recovered in a V-bottom well collecting plate using 100 µL of 80% ACN, 0.1% acetic acid and 100 µL of ACN. The plate was evaporated on TurboVap 96 Evaporator. Before analysis, each sample was resuspended in 10 µL of H2O, 0.1% formic acid. The concentration of peptide was measured with DS-11+ (Denovix, Wilmington, USA) at 215 nm. MALDI analysis. Desalted peptides (1 µL) were deposited on 384 Ground steel MALDI plates (Bruker Daltonics, Bremen, Germany) or 384-well AB Sciex MALDI plates, then 1 µL of HCCA matrix at 10 mg/mL in ACN/H2O 80:20 v/v 0.1% TFA was added for each sample spot and dried at ambient temperature. MALDI-FTICR experiments were carried out on a Bruker 9.4 Tesla SolariX XR FTICR mass spectrometer controlled by FTMSControl software and equipped with a CombiSource and a ParaCell (Bruker Daltonics, Bremen, Germany). A Bruker Smartbeam-II Laser System was used for irradiation at a frequency of 1,000 Hz and using the “Minimum” predefined shot pattern. MALDI-FTICR spectra were generated from 500 laser shots in the m/z range from 693.01 to 5,000 with 2 M data points (i.e., transient length of 5.0332 s). Twenty spectra were average. The transfer time of the ICR cell was set to 1.2 ms and the quadrupole mass filter operating in RF-only mode was set at m/z 600. Bioinformatics for MALDI data. MS raw data from MALDI FTICR were processed using DataAnalysis 5.0. SNAP algorithm was employed with the following parameters of S/N > 3 and quality 0.6. The procedure for the deamidation value calculation from MALDI FTICR was based on Bray et al. 2023 and the identification of taxonomic rank was realized with PAMPA Touzet et al. 2025. Funding ZooMS analyses were caried out at the IMEC laboratory. Financial support from the IR INFRANALYTICS FR2054 for conducting the research is gratefully acknowledged. This research was supported by a research grant from the « Archéologies » association, based in Montauban. We would particularly like to thank Claire Fradet for her key role in the implementation and funding of this research. The research problematics and preliminary results were presented at the European Associasion of Archaeologists Annual Meeting 2024, for which the TRACES Laboratory – and more specifically the PSH and SMP3C research teams – provided financial support. We also thank the Occit’animaux Research Challenge for its financial contribution. The Auvergne-Rhône-Alpes region and the French Department of Isère also contributed to the funding of the excavations at La Grande Rivoire.