Associate Research Scientist, The Metropolitan Museum of Art
Columbia University in the City of New York
Postdoctoral Fellow, The Metropolitan Museum of Art
This project is supported by a grant from the National Science Foundation to The Metropolitan Museum of Art and Columbia University (NSF Award CHE 1041839)
In the age of the modern museum, dissimilar artworks come into contact and conversation with one another, often in contrast to the artists' or craftsmen's original intentions. We may now encounter artworks as disparate as a medieval Italian polychrome sculpture and an eighteenth-century Syrian reception room beneath the same roof. The combined study of diverse cultural objects may provide new insights about the exchange of imagery, artistic techniques, technology, and materials. What's more, that the condition and preservation of distinctive works have made it possible to collect them in a single time and place remains truly wondrous.
Scientific research often works by a similar mode, wherein serendipitous meetings of seemingly unrelated fields of knowledge result in groundbreaking discoveries. These meetings increase our amazement at the reach of human understanding, adding to our delight in contemplation of the beauty and complexity of both the natural and man-made world.
The discovery of antibodies by the German physician Paul Ehrlich is an example of one such meeting. Earning him the Nobel Prize in Physiology and Medicine in 1908, Ehrlich's foundational research in the field of immunology can be traced to his earlier experiments with cellular dyes, chemical compounds not unlike those dyes used for color in tapestries and other textile arts. Within the cell, the unique coloring properties of dyes result from chemical reactions between cellular material and the specific structure of the dye. By analogy, Ehrlich formulated his side-chain theory of immunological responsiveness.  Ehrlich surmised that the chemical makeup of different cell types brings about the production of proteins able to interact with foreign materials, called antigens, in the bloodstream. These proteins, known as antibodies, possess a wide range of variations in chemical structure. This variation accounts for the specificity of antibodies to bind and react with one kind of material rather than another.
Ehrlich's findings set the stage for subsequent investigations in the field of cellular immunology. Enormous varieties of distinctive antibodies have been shown to exist within the proteomes of living organisms, each sensitive to the presence of an antigenic material, and highly specific for a particular type of antigen. These features help to explain how antibodies accomplish their function, quickly eliciting a targeted response from other actors in the immune process.
Antibody Methods for Conservation Science: ELISA and Immuno-SERS
Outside the context of living organisms, scientific technologies have successfully exploited the highly specific and sensitive nature of antibodies in a variety of applications. As a result, antibody-based assays for the identification and/or quantification of organic matter in samples have become staples of the scientific toolbox. The Enzyme-Linked ImmunoSorbent Assay (ELISA) represents one such technique (fig. 1). A single or multicomponent mixture of antigenic (e.g., protein) and nonantigenic (e.g., oils) materials is exposed to a primary antibody specific for one antigen. A second antibody specific for the primary antibody and carrying an observable marker, known as a "reporting system," is introduced. On the basis of differential intensities signaled by the reporting system, the kind of protein in the original sample can be easily determined.
The ELISA technique presents several emerging applications for the conservation sciences as well. The most basic paints are made from inorganic mineral pigments and polysaccharide (e.g., fruit tree gums) or protein-based binders, such as egg, milk, and animal glues. Each binder contains molecules unique to their respective sources: ovalbumin in eggs, casein in milk, and collagen in animal glues. Researchers interested in the makeup of an artist's palette or craftsperson's materials—such as tree gums or glues composed of collagen for use as adhesives—would have the ability to detect the presence of organic components in an art sample at low concentrations. Whether applied to paintings, sculptures, or decorative objects, small samples of paint may be taken from a work and analyzed to address specific questions. Once established, this technique can be performed at relatively low cost to cultural institutions and without the need for high-tech instrumentation.
ELISA encounters issues particular to the conservation and analysis of artworks. Cultural heritage applications of this technique must overcome degradation of the antigen due to aging, heat and humidity, and light exposure, as well as the potential for diminished binding efficiencies and responsiveness of the reporting system due to pigment interaction, impurities, or heterogeneous sample conditions. However, the power of ELISA to identify unambiguously the presence of a specific kind of binder within a sample containing many proteins, gums, or other binding media represents the most significant advantage of antibody-based techniques over other methods. Further optimization of the ELISA technique against extreme sample variability will strengthen the interpretation of analytic results. ELISA will increasingly aid researchers in the investigation of processes faced by artworks and cultural objects in both museum and field settings.
In addition to ELISA, combination immunological and nanomaterial-based techniques show great promise for the specific identification and localization of proteins in artworks. Taking advantage of established spectroscopic methods in the conservation sciences, Immunological-Surface Enhanced Raman Spectroscopy (Immuno-SERS) also exploits the layered antibody/visualization approach shared by ELISA (fig. 2).
First, a small multilayered sample taken from an object of interest is mounted intact in resin and polished to reveal the stratigraphy. This sample may be composed of layers made from one or more organic binders and inorganic pigments. Exposed proteins located just on the surface of the sample present epitopes for primary antibody attachment. Next, the sample is incubated with a primary antibody highly specific for one kind of binder. Finally, gold nanoparticles attached to secondary antibodies are applied to the cross-section. The gold nanoparticles are also complexed with a small organic molecule sensitive to excitation with a laser, which acts here as the reporting system. The nanotag-complexed secondary antibodies attach to primary antibody in the sample. After multiple washings with detergent, the cross-section can be analyzed for presence of the reporter. A positive (or negative) result is associated with presence (or absence) of a protein-based binding medium.
Detection of the reporting tag is accomplished using Raman spectroscopy. A spectrum relating wavelength and signal intensity is produced for analysis (fig. 3). This spectrum will include a signature spectral pattern specific to the reporting system just in those places where tag is present in the cross-section. However, many pigments and binders of interest in the conservation sciences inherently fluoresce when exposed to light. If spectral patterns from these materials overlap with signal from the reporter, meaningful analysis of an artwork may be confounded. The SERS approach cleverly overcomes this limitation via coupling of the reporter to a gold nanoparticle. Enhancement of electromagnetic fields near the reporting molecule amplifies the signal above any overlapping spectral peaks. Moreover, when the reporter molecule links two gold nanoparticles, "hot spots" amplify the reporter signal, allowing up to 1010 orders of magnitude greater sensitivity.  The occurrence of "hot spots" results in greater likelihood of identifying the reporter, even when surrounded by an inherently fluorescent environment. In order to increase individual signal intensities and number of "hot spots," further development of SERS includes experimentation with various reporting molecules and controlled manufacture of nanoparticles of diverse shapes.
In the context of museum study, the Immuno-SERS assay shares several benefits with ELISA. These include the use of low-cost materials, or equipment common to most conservation science labs, as well as the specific detection of a variety of binding media (either simultaneously or in-series). Most importantly, Immuno-SERS functions in the localization and mapping of proteinaceous binders within individual cross-sections, offering data similar to highly sophisticated techniques such as ATR-FTIR. This is in contrast to ELISA, which detects the presence of proteins across an entire sample. Cross-sections may also be probed for organic binding media in situ, as opposed to ELISA experiments requiring lengthier extraction protocols. Finally, the large size of each nanotag-antibody complex restricts penetration beyond the exposed surface of the sample. Following each trial, any remaining nanoparticles or primary antibodies may be polished off the cross-section, allowing experiments to be repeated multiple times on the same sample.
Though possessing great symbolic significance, works of art also possess the material properties of their creation. Heritage objects are complex combinations of naturally occurring and/or synthetic materials. The artist or craftsperson may have selected materials for their availability—or scarcity and value—or on the basis of numerous functional properties such as color, texture, sheen, color-fastness, drying time, compatibility, etc. Correct identification of materials in a given work represents a key towards advancing and enhancing art-historical interpretations of objects. Knowledge of art materials may help us to answer the questions: When and where was the work made? How was the work made? Who made this object? Why has the object changed in a particular way?
Binders from animal and plant sources such as proteins and gums, respectively, are elemental components in a majority of artworks dating from prehistory to the present. In addition to aiding scholars in curatorial and other art-historical pursuits, the ability to determine various kinds of binding media can help in the preservation and conservation of artworks. Without clear working knowledge of the materials from which an object is constructed, conservators may be unable to repair a damaged work in optimal fashion, and curators may not recognize the proper environment for a given work to be displayed and stored. Insights concerning the composition of binding media are especially relevant in these respects, since inorganic pigments and other colorants may interact with a given binding medium over time, as is discussed in "Investigating the Formation and Structure of Lead Soaps in Traditional Oil Paintings." The use of more favorable, reparative substances or diminished exposure to unfavorable conditions may help in minimizing these reactions.
Case Study One: An Italian Polychrome Sculpture of St. John
In order to identify the proteinaceous material used as an adhesive for the application of gilding on a thirteenth-century Italian polychrome sculpture of St. John (part of The Cloisters Collection) (fig. 4), both ELISA and Immuno-SERS assays were applied as complimentary methods of investigation.  Applied directly to the gesso preparation along the figure's garments, this tin-leaf gilding was reportedly held in place by either animal glue (collagen) or egg white (ovalbumin). Though examination with ATR-FTIR verified the presence of protein, analysis of the gilding adhesive proved inconclusive as to the material's exact identity. This provided us the opportunity to explore antibody-based methods as a means for the specific identification of a protein adhesive in an actual artwork.
ELISA studies on samples extracted from just below the gilding showed high signal responsiveness when exposed to ovalbumin-specific primary antibodies (fig. 5). Samples were extracted from just below the tin leaf, and from farther into the gesso layer. The signal from ovalbumin-specific primary antibodies was maximal in sample from below the tin leaf, and significantly reduced in samples from the gesso layer. In contrast, responsiveness to collagen-specific primary antibodies increased in the gesso layer. In both samples the presence of milk (casein) or gum (polysaccharides) was not detected. Thus, the ELISA technique revealed the identity of the adhesive—located just below the tin-leaf application—to be ovalbumin, and the identity of the binder in the gesso ground layer as collagen.
Immuno-SERS was also performed on intact cross-section samples of gesso, gilding, paint, and glaze (fig. 6). After exposing the sample to ovalbumin-specific primary antibody, the most intense signal from the Raman-active reporting system (trans-1,2-bis(4-pyridyl)-ethylene) when exposed to laser light (λ = 785nm) arose just below the gilding, indicating the presence of ovalbumin. Signal from the reporting system was not found farther away from the gilding in the gesso preparation layer, indicative of the absence of ovalbumin in this layer.
The Immuno-SERS results were in excellent agreement with those ascertained from ELISA studies of the adhesive used in the creation of the St. John polychrome. Together, these analyses unambiguously identified the adhesive as having been sourced primarily from egg whites (ovalbumin) and the gesso binder as animal glue (collagen). Moreover, Immuno-SERS permitted the simultaneous identification and localization of ovalbumin in the polychromy stratigraphy, a question not easily answerable through other methods common to the practice of conservation science.
Since Paul Ehrlich's discovery of antibodies through experimentation with colored dyes, immunological research has once again progressed with the recent application of antibody-based methods for the identification of protein materials in artworks. Further development and optimization of ELISA and Immuno-SERS methods promise significant contributions in the field of conservation science, most notably in the unambiguous identification of specific organic binding media in complex sample mixtures. Moreover, the capability to perform such experiments at low costs to museum laboratories heightens the attractiveness of these methods for active researchers.
In the future, research in the conservation sciences will greatly benefit from focused study of how antibodies interact with organic media found in artworks over time. With further research, the ELISA method may be used to investigate specific effects of aging, such as the loss of signal responsiveness in different paints over time. With respect to the Immuno-SERS method, progress will be made towards characterization of novel Raman-active reporting systems and the reliable fabrication of SERS nanoparticles of various shapes. Antibody-based methods carry the potential to revolutionize not only the practices of researchers working in laboratories or in the field, but of the technical and interpretive appreciation of artworks in general.
1 Prüll, C.-R. "Part of a scientific master plan? Paul Ehrlich and the origins of his receptor concept." Medical History, 47, 332–356 (2003).
2 Indrasekara, A. S. D. S., Paladini, B. J., Naczynski, D. J., Starovoytov, V., Moghe, P. V. & Fabris, L. "Dimeric gold nanoparticle assemblies as tags for SERS-based cancer detection." Adv. Healthcare Mater, 2, 1370–1376 (2013).
3 Arslanoglu, J., Zaleski, S. & Loike, J. "An improved method of protein localization in artworks through SERS nanotag-complexed antibodies." Anal Bioanal Chem, 399, 2997–3010 (2011).