Microfluidic Diffusional Sizing (MDS) is a novel technology which can be used to measure binding affinity and characterise protein interactions. MDS is complementary to current approaches such as surface plasmon resonance (SPR) and bio-layer interferometry (BLI) as interactions are monitored in solution, rather than on a sensor surface. Performing measurements in solution enables the analysis of samples that cannot be readily analysed using surface-based approaches.
In this article, Dr Sean Devenish, the Chief Scientist at Fluidic Sciences, discusses how the MDS technology can enhance the assay development process.
There is a growing industry recognition of the importance of working in authentic biological matrices
Protein binding affinity assays in modern biopharma research
Biopharmaceutical research involves working with a number of challenging samples; this is usually because of the nature of the protein target itself, as well as the context by which it functions within the body. Proteins are tremendously diverse; while around 40% of them adopt well folded globular structures that are easy to study, a majority of them present a range of challenges. For example, proteins can be intrinsically disordered, present in macromolecular complexes, or integrated into lipid membranes.
There is a growing industry recognition of the importance of working in authentic biological matrices; this is because they can help drug developers to determine the effects of their therapeutic as they might be in a real-world setting. Simply assuming that measurements made in simple buffers will reflect behaviour in complex biological settings — such as in the serum or plasma — is no longer sufficient. Therefore, being able to perform measurements directly in biofluids is an important benefit to the sector.
Current binding affinity measures not ideal for all samples
To conduct SPR and BLI, proteins are immobilised on a surface through their natural attraction to said surface under appropriate conditions. This technique has enabled a huge amount of research — though this comes at a cost. Firstly, the immobilisation process itself can create challenges, such as the steric occlusion of binding sites, or the detection of avidity rather than affinity if target molecules are too closely spaced. Relying on detection at a surface presents even more issues, as non-specific binding events on a surface cannot be discriminated from target-mediated specific binding.
This prevents the use of undiluted biofluids such as serum or plasma, because the complexity and variety of components in these matrices inevitably leads to a significant amount of non-specific binding. Optimisation of immobilisation and subsequent surface passivation can require significant work — diverting time and effort away from more valuable activities. 
Dr Sean Devenish, Chief Scientist at Fluidic Sciences
Solution-phase analysis using MDS
By working in solution, MDS immediately avoids many of the problems associated with surface-detected approaches. The large and insoluble sensor surface is replaced with a small (~ 1 kDa) and soluble fluorescent tag. This type of labelling is widely used in modern biological research, and is also generally well tolerated. Non-specific binding of the analyte or matrix components is irrelevant in this case, as only interactions with the labelled probe molecules can be detected, making the technology specific to the probe molecule of interest.
MDS operates by measuring the diffusion range of the fluorescent probe molecule, and relies on understood physical principles. When the probe molecule interacts with its binding partner, diffusion slows because of the larger size of the complex. In the instrument readout, this diffusion data is converted to hydrodynamic or Stokes radii to ensure ease of interpretation.
Working in solution ... means that researchers know the concentration of the probe molecule
It’s worth noting that the hydrodynamic radii can be used to better understand the sample — as the probe molecule’s radius can detect aggregation or degradation of the protein species. Likewise, if the size of the complex formed in the reaction differs from that expected, it can reveal off-target binding, aggregate formation or unanticipated stoichiometry of a reaction.
Working in solution also means that researchers know the concentration of the probe molecule, which can also be altered to suit the experiment. When using surface-based interaction technologies, the loading of the probe molecule on a surface can be altered by applying different concentrations of probe during the labelling step; though the amount of problem that is actually immobilised is unpredictable and hard to determine.
Being able to set the concentration of probe to suit the experiment is useful even for standard binding curves, where it’s important for the probe concentration to be below the dissociation constant (KD). The ability to use different and known concentrations of probe in an experiment becomes particularly powerful in samples that have an unknown concentration. 
Using MDS with a powerful analytical suite allows researchers to fit datasets comprising differing concentrations of probe and sample globally to determine both the binding affinity of interactions and the concentrations of binding sites in the sample. For example, this can be used to determine the concentration of antigen-specific antibodies in a serum or plasma sample. It can also determine binding stoichiometry where the nominal binder concentration is known, but the ratio at which the binder interacts with the probe is unknown.
In-solution analyses using MDS in biopharmaceutical research
MDS has the capability to impact a range of factors across the biopharmaceutical workflow. A selection of applications that would benefit from the in-solution analytical capabilities of MDS are:
- Reagent quality control: Modern biopharmaceutical research frequently makes use of fluorescent-labelled reagents, and MDS can provide a mechanism to test these molecules —ensuring they are suitable for use in other assays. Tests used could be as simple as a sizing experiment to ensure that the labeled protein is not aggregated or degraded, or could include determination of function through a binding assay.
- Antibody discovery: Selecting lead candidates by assessing their fundamental properties (affinity and concentration) rather than ELISA titres could streamline selection of suitable candidates, accelerating subsequent development work. This approach would allow researchers to focus on the parameter of most value when choosing candidates, even allowing them to generate different pools focused on a range of properties.
- Developing antibodies: Many antibody drug candidates struggle or fail late in development work due to aggregation or other non-specific binding behaviour. Examining lead molecules early in the development workflow and identifying those with a propensity for non-specific interaction could avoid late-stage failure. This assay can readily be carried out in solution using MDS technology to monitor the size of antibody therapeutics in the presence of non-specific interactors such as nucleic acids, or even testing directly in complex matrices such as serum or plasma.
- Serum stability: Current serum stability assays rely on ELISA titres — which are a composite value of different sample properties. These data are not able to shed light on the mechanism by which performance decreases. By examining binding affinity and concentration as independent parameters, MDS can reveal whether antibodies are being degraded entirely, or merely suffering oxidative damage or other assaults that reduce the affinity of binding. More quickly understanding the mechanism of degradation will accelerate the identification of solutions and progress towards the clinic.
- Vaccine immune responses: A good vaccine induces the production of high affinity antibodies against an antigen. Being able to directly examine affinity as a property independently from concentration could help vaccine developers more quickly identify constructs or formulations that lead to desirable immunological responses. Monitoring binding affinity in serological samples during repeat dosing could readily identify vaccines that are able to induce robust affinity maturation, and therefore, promise lifelong protection.
- Anti-drug antibody detection and characterisation: Biologics are an incredibly powerful and versatile class of therapeutic agents, although one of the risks with protein-based therapies is that they can induce immune responses in patients. Testing for the production of these anti-drug antibodies is essential during biotherapeutic development, and MDS provides a tool to collect powerful data that provides a fundamental understanding of responses. By being able to distinguish between high affinity, specific, binding and low affinity binding that is potentially non-specific or cross-reactive and therefore of lower concern, MDS offers a path to faster and more efficient testing for anti-drug antibodies.
Determining binding affinities for challenging proteins and/or in complex matrices can be difficult when using classical techniques that depend on immobilisation. Performing these analyses in solution avoids the drawbacks associated with surface-based detection and can greatly simplify assay development.
Microfluidic Diffusional Sizing is a recently developed technology that offers powerful assay capabilities and can be usefully applied to a range of experiments throughout the biopharmaceutical development workflow.