The site of Iita in northwestern Greenland had near-continual occupation from 1000 CE to the mid-20th century, which presents an unparalleled opportunity to examine diachronic changes in the use of marine mammals for food and raw material resources by Pre-Inuit (Late Dorset) and Inuit-Inughuit—two culturally distinct Indigenous groups. Limited wood in High Arctic environments necessitated the use of antler, bone, and ivory for tool making. Choices in the selection and processing of osseous material reflect changes in the way these two groups used the same resources. However, modification of osseous material during the tool-making process is often to such an extent that few diagnostic features remain for species identification. Here, we include Zooarchaeology by Mass Spectrometry (ZooMS) to identify typically indeterminate pieces of manufacturing debris from preserved collagen. The minimally destructive extraction technique using polishing films had a 99% success rate. The results suggest that the selection of raw materials from certain marine mammals, particularly narwhal, by toolmakers at Iita was more frequent than previously understood through traditional faunal analysis alone. There is a paucity of narwhal from the Pre-Inuit (Late Dorset) assemblage, which contrasts starkly with the high frequency of narwhal in the Inuit-Inughuit assemblage. We demonstrate the importance of combining the analysis of bone-tool manufacturing debris with traditional dietary faunal remains to improve our understanding of resource use in coastal environments.

One hundred and thirty specimens were chosen for peptide mass fingerprinting (PMF) to identify previously unknown osseous materials and their taxonomic origin. Identification of these specimens promotes a better understanding of osseous tool production. PMF involves the enzymatic digestion of proteins followed by Matrix Assisted Laser Desorption-Ionization Time of Flight Mass Spectrometry (MALDI) analysis of the resultant peptide mixture. In the case of bone and ivory, Type 1 collagen is the major constitutive protein, and for each mammalian source, the amino acid sequence of Type 1 collagen, albeit highly conserved, may have one or more amino acid substitutions. These substitutions change the mass of some peptides detected in the MALDI spectrum of the protein digest, which are then used as markers for specific mammalian sources. Markers from known reference materials are compared with those from unknown samples for identification. 

The ZooMS (PMF) method was a simplified procedure based on that described in Kirby (2013, 2019). Since there are no cystines in collagen, the cystine reduction and end capping steps were omitted. In addition, spectral quality was sufficient to provide confident identifications in all cases where collagen was detected, so no purification or fractionation was required. Some spectra (~15%) had weak or absent A and/or G markers, but those markers were not required for accurate identifications of these samples. In all cases, only a single sample was digested and analyzed.

The first author sampled the specimens at the UC Davis Zooarchaeology Lab with sample sticks—polystyrene strips with fiber optic polishing film attached on two sides—which are used to abrade and entrap a small amount of material for analysis. A freshly laundered lab coat, KN95 or surgical facemask, and nitrile examination gloves were worn during handling of all osseous materials. To avoid cross-contamination, gloves were changed, and the workstation was thoroughly sterilized with 70% alcohol wipes between each sample. Samples were obtained by rubbing the object lightly in a circular motion for 10 seconds. After sampling, the tip of the sample stick containing the abrasive film and entrapped sample is cut off, placed in an Eppendorf tube, and forwarded for analysis. Polishing films (30um alumina or diamond particles, final size 2.5 X 2.5mm) were from Precision Fiber Products, Inc., www.precisionfiberproducts.com. Polystyrene strips (1.5 X 2.5 X 400mm) were from Walthers, www.walthers.com.

Photographs of each specimen were taken with a DinoLite digital microscope to document the area sampled. The sampling site is generally invisible to the naked eye, and the procedure consumes a miniscule amount of each specimen. Thus, it is considered a minimally destructive sampling technique by most museum curators and conservators. Each specimen and Eppendorf tube was given a unique sample number. Additional samples from the same location were collected on three specimens (KNK912x300; KNK3930x1571, x1822) because they yielded low-quality spectra on the first sample run. These are considered “second passes”, in which collagen was sampled from the same location but from a deeper anatomical component of the specimen, such that a better quality and/or greater quantity of collagen could be obtained.

PMF analysis of the 130 samples collected was then undertaken at the Northeastern University Mass Spectrometry Core Facility.  Digestion: 60µL of 50mM ammonium bicarbonate (AMBI) were added to each sample (on the sample stick tip in a 600µL Eppendorf tube) and heated to 75°C for 60min. After cooling, 8µL Promega Sequence Grade trypsin (0.02µg/µL in 50mM AMBI) was added and digestion proceeded overnight at 37°C. 

MALDI analysis: 2µL of the digest were added to 20µL 40% acetonitrile (ACN), 0.1% trifluoroacetic acid (TFA) with saturated α-Cyano-4-hydroxycinnamic acid (CHCA) matrix. 0.65µL of the mixture was spotted onto the MALDI plate. Spectra were obtained with an Applied Biosystems/Sciex 5800 MALDI-TOF-TOF instrument operated in positive reflector mode. Calibration was done with a standard mixture of peptides: 757.3992Da, 1046.5418Da, 1296.6848Da, 1347.7354Da, 1619.8223Da, 2093.0867Da, 2465.1983Da, and 3147.4710Da. Spectra were coadditions of 1200 – 2000 laser shots. Acquired spectra were exported from Applied Biosystems Data Explorer software as text files and imported into mMass (Strohalm et al. 2010) for analysis. Spectra were manually inspected for markers. Markers used to identify mammalian sources are compiled from Buckley et al. (2009, 2014) and Kirby (2013).

Data is provided in two formats, .msd and .txt files. The text files are organized so that they correspond to the data in Supplementary Table 1. Further details can be found in the README.md file.

The .txt files can be accessed by various software programs to visualize the data. 

The .msd files can be viewed using the mMass software for those who have access. Alternatively, the .msd files can be read by using the packages MALDIQuant and MALDIQuantForeign in R. 

mMass Data Miner: an Open Source Alternative for Mass Spectrometric Data Analysis
Strohalm M, Hassman M, Košata B, Kodíček M
Rapid Commun Mass Spec 22 (6), 905-908 (2008)
DOI: 10.1002/rcm.3444