The biomolecular characterization of a finger ring contextually dated to the emergence of the Early Neolithic from Syltholm, Denmark

Abstract

We present the analysis of an osseous finger ring from a predominantly early Neolithic context in Denmark. To characterize the artefact and identify the raw material used for its manufacture, we performed micro-computed tomography scanning, zooarchaeology by mass spectrometry (ZooMS) peptide mass fingerprinting, as well as protein sequencing by liquid chromatography tandem mass spectrometry (LC-MS/MS). We conclude that the ring was made from long bone or antler due to the presence of osteons (Haversian canals). Subsequent ZooMS analysis of collagen I and II indicated that it was made from Alces alces or Cervus elaphus material. We then used LC-MS/MS analysis to refine our species identification, confirming that the ring was made from Cervus elaphus, and to examine the rest of the proteome. This study demonstrates the potential of ancient proteomics for species identification of prehistoric artefacts made from osseous material.

Keywords: collagen; liquid chromatography tandem mass spectrometry; mass spectrometry; micro-computed tomography scanning; zooarchaeology; zooarchaeology by mass spectrometry.

Figure 1.

(a) Location of the site on the southern part of Lolland, Denmark. (b) Overview of site MLF906-II where the ring was found in the northern part. (c) Digitized archaeological wood and stones found in a small section of the site, from where the ring was found. Digitization based on seven three-dimensional models obtained by Structure from Motion. (d) Photograph of the ring.

Figure 2.

Probability distributions of the 15 radiocarbon samples found in close proximity to the ring. The coloured probability distributions are the result of a simple Bayesian model assuming all samples to originate from the same phase of activity. Onset and termination of the phase are indicated with black probability distributions. The light green probability distribution is the date we propose for the ring as well, based on the proximity of the spear to the ring. KDE model of all 70 ¹⁴C dates indicating a single inferred period of archaeological activity at site in general.

Figure 3.

Micro-CT scan. (a) Full field of view of volume rendering, (b) transverse slice showing a few small black pores assumed to be Volkmann's canals, (c) cut along the middle of the volume rendering, (d) slice along showing several small black pores assumed to be osteons, (e) network of osteons arranged longitudinally and Volkmann's canals aligned perpendicular to the latter (in blue).

Figure 4.

ZooMS results. MALDI-TOF-MS spectral output visualized in mMass (v.5.5.0) Extraction 1 and 2 flipped. Peptides unique to Cervus and Alces are highlighted in red.

Figure 5.

Example of two peptides (a,b) from the MS/MS output from MaxQuant, containing SAPs unique for Cervus (marked in grey). Panel (a) is located on the collagen 1 α-1 sequence, while (b) is located on the collagen 1 α-2 and both can confidently, based on the y and b ion series, be identified as Cervus.

Figure 6.

Venn diagram demonstrating 1 protein shared between the ring and antler, and 0 proteins shared between the ring and bone.

Figure 7.

Deamidation comparison of the ring, heated modern reference samples as well as contaminant proteins. Numbers in each column denote number of peptides used for the calculation. Deamidated asparagine (N) and glutamine (Q) residues from the ring and known contaminant proteins, showing a considerably higher deamidation rate in the sample, which is evidence of authentically ancient proteins.