The proteomics of pottery and its residues: applications to ancient foodways

Food is one of the most important parts of our daily lives yet one of the most difficult to detect in the archaeological record. Proteomic approaches to food have enormous potential in the detection of cuisine as, due to the genetically coded nature of protein sequences, they can provide evidence for specific ingredients and ingredient mixtures. Recently, proteomic analysis of pottery and its residues has revealed the ingredients and cuisine processed in a range of archaeological periods and contexts. Yet we remain largely ignorant of the impact of intentional and incidental processes on the preservation and composition of ancient protein residues in and on pottery. Therefore, in this thesis I seek to address the question “to what extent do proteomic results from ancient pottery reflect the input ingredients and cuisine processed in that pot?” Through case studies, I investigate the impact of different input ingredients, food preparation practices and deposition on proteomic results in pottery and pottery residues. First, a large-scale proteomic analysis is undertaken on 100 Romano-British pots from the site of modern day Northstowe. Here I explore the utility of proteomics in detecting specific ingredients. I report on the ingredients identified by proteomics throughout the site’s occupation, revealing direct evidence for a transition from multi-taxa dairying to a focus on cattle milk. I then appraise the utility of proteomics in ingredient detection, by comparing the protein results to lipid analysis and other dietary evidence. Lastly, I explore trends in protein identification, establishing biases that exist in proteomic detection of different ingredients. The results reveal that, in line with previous research, particular milk peptides survive remarkably well, while plant proteins were detected infrequently. Next, I explore the impact of food preparation practices on protein detection through a case study of the detection of cheesemaking in Late Neolithic pottery from central Poland. In this chapter I investigate the capability of proteomics to detect food preparation practices, rather than simply ingredients themselves. The results reveal that a high proportion of casein peptides in calcified residues may provide evidence for curd-enriched dairy products such as cheese, and that both sheep and cows’ dairy was exploited at Sławęcinek and were processed in the same vessels. Comparison with the lipid data revealed discrepancies in ingredient detection, potentially indicating aspects of vessel use-life. I then assess the impact of cooking and burial on proteins in ceramics and foodcrust, using experimental samples containing deer meat, salmon meat and chestnut flour. While calcified residues have received more attention, proteomic analysis of charred residues may yield valuable evidence of ancient foodways. The results reveal that foodcrust is a better sample type for proteomics than ceramic, and that ingredients themselves impact preservation (and therefore detection) in the cooked, buried samples. I explore the physicochemical properties of the proteins and peptides which persist through cooking and burial in foodcrusts, revealing that surviving proteins are more likely to be hydrophobic, and that complex physicochemical interactions are likely influencing protein preservation. Finally, I scrutinise the biases observed across the Northstowe sample against a range of factors which may explain them. I first investigate the mechanisms of limescale formation, which reveal that the unique scale forming characteristics of milk may explain its dominance. An analysis of the physicochemical properties of milk proteins further reveals that specific properties such as antimicrobial tendency may play a role in preservation. The low level of plant detection at Northstowe is explored, revealing that the inferior scale forming capacity of plants, and the use of inappropriate enzymes in their extraction may play a role, and that certain protein tertiary structures are more likely to survive. This thesis therefore demonstrates that proteomic results from pottery are indeed impacted by ingredient selection – and the physicochemical properties of those ingredients, food preparation practices, and burial. While protein residues can yield powerful insights into ingredients and food preparation practices, given the preservation biases and variability observed in this thesis, they should not be viewed as representative of all foodways employed on ancient sites. This thesis reveals that proteomics should be integrated with lipid analysis and other methods to provide the most complete window into ancient foodways.

Name dc:type.qualificationname

Doctor of Philosophy (PhD)

Level dc:type.qualificationlevel

Doctoral

Grantor dc:publisher.institution

University of Cambridge

Year dc:date.issued

2024

Author dc:creator

• Evans, Miranda

Advisors dc:contributor.advisor

• O'Connell, Tamsin
• Collins, Matthew
• Millett, Martin
• Hendy, Jessica

• Palaeoproteomics
• Archaeology
• Roman Britain
• Late Neolithic
• Experimental Archaeology
• Archaeological Science

Licence

• https://apollo8-f-pro.lib.cam.ac.uk/bitstreams/46a83eca-e8df-421a-be0e-f553914cb0bd/download
• https://creativecommons.org/licenses/by/4.0/

Language dc:language

eng

Author Identifier

0000-0002-9284-9268

OAI identifier oai:identifier

oai:www.repository.cam.ac.uk:1810/379809