Surface Plasmon Resonance as an Analytical Tool for Bioprocessing

Well established as an analytical tool for biomolecular interaction analysis, surface plasmon resonance (SPR) is a spectroscopic method that has entered the biopharmaceutical toolkit as a technology for supporting process-development and quality-control operations. Among its virtues, the technology offers real-time, label-free detection, which may prove useful in the future as a process analytical technology (PAT). Pharmaceutical Technology spoke with Fredrik Sundberg, director of Strategic Market Development at GE Healthcare, which produces an SPR system under the name Biacore, to learn how this analytical tool can be incorporated into biopharmaceutical manufacturing.

PharmTech: What is SPR used for?

Sundberg: SPR exploits the binding between two molecules to measure the interaction between an analyte and ligand. In the case of Biacore, the interaction takes place on a chip surface, where the ligand is immobilized. SPR provides more comprehensive information about binding events than any other technology on the market today such as affinity, kinetics, specificity, and active concentration of a particular protein in solution. For example, you can detect protein concentrations in a purified sample or in a crude sample.

PharmTech: What do you mean by active concentration?

Sundberg: You get real-time binding data about how much of your protein of interest is actually actively binding to the target. In fact, you can see the binding taking place as it happens in the processing unit by looking at the system-controller computer. This allows you to characterize the product and detect certain impurities during purification and process development. Due to the sensitivity of the SPR detection principle, it is possible to very effectively assess comparability of products by looking at binding kinetics and conducting epitope mapping. For example, you may be able to detect various protein modifications by looking at changes in binding patterns. SPR detection not only measures total protein content in an unknown sample, like you would do with A280 (optical absorbance at a wavelength of 280 nm). In fact, with SPR, it is possible to very accurately measure the active concentration without using a calibration curve. This allows companies to improve their assays in general by using SPR for raw material control of critical reagents used in, for example, ELISA-based immunoassays.

PharmTech: Where do you see SPR being useful in a bioprocessing setting?

Sundberg: It is very useful during process development. Because you can detect concentrations of your product in real time, you can determine how various processing variables are affecting your product yield and product integrity. It could also be used during commercial bioprocessing for quality control to detect product and product changes. In process development, it has been used for detection of protein aggregates, for example, or host-cell-proteins during product purification. SPR is often used in combination with other biophysical methods such as microcalorimetry, size-exclusion chromatography (SEC), and liquid chromatography/mass spectometry. And as companies move into follow-on biologics, they will find they can configure the system to determine the immunogenicity of product variants. It's even possible to design experiments to identify parts of a molecule that may trigger immunogenicity so a company can go back and re-engineer the product and process to reduce that product's immunogenicity.

PharmTech: How does it work?

Sundberg: The sample analysis is fully automated and supports both 96-well and 384-well plates. First, the ligand is injected by an autosampler through the microfluidics system in the processing unit over a chip onto which it is immobilized. After the chip is coated with the desired density of ligand, the analyte is injected, and binding is measured by an optical unit. Accurate concentrations of the unknown samples can be calculated based on the amount bound, which corresponds to the signal measured from changes in refractive index. For example, coating the chip with Protein A would allow you to measure the concentration of antibodies (since Protein A would attract the constant fragment [Fc] of the antibody). To achieve greater specificity, or to combine measurements using the multiple flow cells in the system, one might also use a specific antigen as ligand to measure the concentration of only those antibodies in the sample with the antigen-binding (Fab) region specifically capable of recognizing the ligand.

PharmTech: What are the advantages of SPR over existing analytic technologies?

Sundberg: Because SPR detects binding in real time, and sample prep is minimal, there's a tremendous time-saving over, say, ELISA or high-performance liquid chromatograhy. Variability is low, and the coefficient of variability often falls in the range of 1–5%, even for very crude samples (e.g., cell-culture supernatants). Apart from the improved productivity, SPR also resolves affinity into its kinetic parameters (on-and off-rates) which enables scientists to assess more clinically relevant parameters such as recognition and stability of binding to target-binding site.

PharmTech: Could you give us an idea of what sort of time saving you're talking about?

Sundberg: You would get an answer from an ELISA assay in a matter of hours, but SPR could give you an answer within a few minutes (3–10 minutes). For process engineers, this can shave considerable time off process development. Also, unlike HPLC, SPR does not require you to produce a standard curve, so that essentially SPR enables calibration-free concentration analysis. This is of great advantage when you have an unstable standard, or if a standard is not available.

PharmTech: What is the future potential of SPR as a PAT tool?

Sundberg: The real-time capability and critical information from SPR would provide a great advantage in PAT applications in the future. On-line or in-line detection of product with SPR would provide critical information of quality attributes and how these are affected by process modifications. For example, there have been some academic initiatives that have proven the possibility of connecting Biacore to bioreactors for continuous monitoring of expression levels over time under various conditions.