AIC's 52nd Annual Meeting

Genomics can offer unique perspectives into the creation and history of cultural heritage objects. The possibility of learning more about the makers of objects through genomics is tantalizing: there are examples where an artisan’s intimate contact with the manufacturing of art materials suggest that genetic information may survive. Gofun, a calcium carbonate pigment commonly found in traditional Japanese paintings, is one of these examples, and this abstract describes our attempts to isolate the genetic information of the maker from a modern preparation. The paint is prepared from gofun powder, which is made by an intensive pulverization of air-dried oyster shells and kneaded by the artist or their assistant with an animal binding glue. A small amount of water is added before application to a textile or paper support. The paste is kneaded by hand for at least an hour, thus increasing the opportunity for cell-free DNA from the artisan’s hands, including the artisan’s own DNA, to become incorporated into the paint. This phenomenon sparks intriguing questions about what information genomic tools can provide about the history of a painting such as: Who prepared the paint? Could that correlate to the attribution? Does attribution require a combination of human and microorganism genetic information? What microorganisms and organic materials were the painting exposed to, and how might that impact conservation treatment?

This project, representing a novel collaboration between the Metropolitan Museum of Art and the Mason Laboratory of Weill Cornell Medicine, explores the extent to which these questions can be answered using the smallest paint sample possible. As sample size is the limiting factor for most art analysis, especially Asian art, which typically is painted in very thin layers, the following work describes what might be possible on milligram sample sizes with the hope that miniaturization could be achieved. A modern gofun paint mockup, prepared by a Met conservator, served as our paint source. A protocol optimized for highly-fragmented DNA from calcium-based sources was implemented on four samples, ranging between 0.5 - 2 mg of paint. We selected a protocol that was sensitive to small fragments of DNA as the paint source was exposed over time to light, water, and enzymes that digest DNA, all of which drive DNA degradation reactions. Moreover, calcium ions from the gofun interfere with extraction by tightly binding to DNA, so a protocol that sequesters calcium was essential for DNA recovery. As the extraction yield was too low for detection, we amplified the extracted DNA to reach the minimum concentration required for sequencing. Fluorometry and automated electrophoresis following amplification support the presence of DNA in the paint samples. Moreover, software tools for organism identification reveal the presence of human, bovine, and mollusk DNA, among other species, suggesting the power of genomic tools for material verification. Genomic isolation and analysis from smaller-sized samples of gofun may be possible; however, challenges remain. Repetitive amplification of the sample can create byproducts that interact with other samples run on the same instrument, leading to crosstalk between samples. This can result in the false identification of unexpected species found in other samples. Future studies will probe into better addressing these issues.