Anisotropic surface-plasmon- resonance imaging biosensor

Biochips promise a range of potential applications in the life sciences: for example, the diagnosis of genetic diseases, the study of protein structure and function, and the discovery of new drugs. These devices usually consist of an array of biosensing locations—probes deposited onto the chip's surface using specialized chemistry—that interact with biomolecular targets such as short gene sequences or protein fragments. Many biochemical interactions can be screened in parallel and then translated into intelligible data using different techniques. One of the most common, called microarray technology, uses targets that are modified: e.g., by attaching fluorescent dyes to the molecules. Scanning the sites of interaction reveals only the spots where probes have stuck or `bound.’ Mapping these locations provides valuable data about gene expression and protein binding strength, among other things.

A second technique relies on surface-plasmon-resonance imaging (SPRI) transduction. SPR is a physical phenomenon that occurs when light is coupled to a thin layer of metal—usually silver or gold—under certain conditions (specific angle of incidence, wavelength, polarization, and metal thickness). An evanescent plasmon wave propagates along the metal-dielectric interface (see Figure 1). Changes in either the refractive index or biomolecular layer thickness in the vicinity of the metallic interface can be monitored as a shift in the resonance curve. SPRI allows direct measurement of the optical constants of the dielectric layer above the metal. This gives an indirect measure of the increase in mass resulting from adsorption of biomolecular targets.

Figure 1. Diagram of a typical surface plasmon resonance imaging (SPRI) system. The transverse magnetic (TM)–polarized incident beam of wavelength λ is refracted through a high-index glass prism of angle θ.

SPRI enables us to evaluate precisely, in real time, the adsorption and desorption kinetics¹ of each spot throughout the chip's surface (2D array). Moreover, additional dimensions can be exploited, such as angle of incidence, wavelength, and polarization. This dynamic information provides a multidimensional `image’ of the interactions.

Figure 2. Picture of the anisotropic SPRI system.

Figure 3. Evolution of the average reflectivity for each arm when sequentially injecting (1) 1% glycerol in water, (2) 2% glycerol in water (repeated two times), and finally (3) 1×PBS (phosphate-buffered saline), then (4) 1.25×PBS.

Changes in the shape of biomolecules are often important in analyzing biological processes. For example, different conformations of a protein could result in different functions, or in disease. The system we developed (shown in Figure 2) is based on two orthogonal SPRI-sensing arms. Each arm is able to retrieve real-time reflectivity data about every spot in the 2D array. We can then perform a differential measurement and extract the average anisotropy—i.e., how physical properties vary with direction—at each point of the array. With this information, we can recover shape and directional features for the biomolecular layer (with known optical constants), such as order parameters or the average orientation of the biomolecules tethered to the chip.

Determining anisotropy from two sensing arms implies that both give similar results accurately and consistently for isotropic samples. In order to verify the stability of our setup, we conducted an experiment using a bare gold chip covered with isotropic mixtures of different refractive indexes laid down one after the other. The two cameras recorded the reflectivity information at each location on the surface during the course of the experiment (see Figure 3). We determined that the results are identical within a 3.10⁻²% differential noise limit.²

Over the short term we will focus on validating the SPRI system for anisotropic samples. For instance, we plan on binding magnetic beads to the biomolecules on the chip's gold surface and to record changes in their orientation induced by a magnetic field. We believe that this type of system will be useful for detecting local flow differences within the same sample, enabling more accurate calibration of the results. In the long term, the conformational data will aid in discriminating variants of proteins. That is important, for example, in prion-type diseases, which are caused by a mutated protein in which only a handful of atomic bonds have changed orientation.