Use light-scattering techniques to predict how your early development candidates will behave at high concentrations for the clinic

Self-association, or the tendency for a molecule to be attracted to itself, is an important assessment criterium for those concerned about how their biologic will behave in the clinic.

Researchers in early development generally use low-volume, low-concentration biologics preps for stability and quality assessment; however, once these therapeutics reach the clinic, they are generally at much higher concentrations, which alters physical properties such as viscosity and aggregation propensity.

However, getting enough material to assess quality attributes at those high concentrations is often prohibitive in early development when there are lots of candidates to screen. Light scattering techniques are used to predict high-concentration behavior of candidates early in the development process. Let’s take a look at what they tell you, how to set them up, and how they’re used for developability assessment.

Light scattering techniques enable you to perform predictive assessments on a biologic’s self-association behaviors

Dynamic light scattering, or DLS, uses the movement of molecules to find out information about their size and whether there are different size populations in the sample.

Static light scattering, SLS, uses the total amount of light deflected by the sample to inform on the average molecular weight of the sample, as well as its colloidal stability.

Large particles scatter more light for SLS and move more slowly for DLS. Each light scattering technique is used to calculate a different parameter for self-association.

DLS gives the kD of the interaction. k_D is a measure of the dissociation constant; that is, the strength of the interaction between two molecules, in this case, two identical biologic molecules in solution. A lower kD value indicates a stronger attraction between molecules and is therefore less desirable.

SLS gives the second virial coefficient, or B₂₂. B₂₂ derives from the concentration dependence of osmotic pressure, which informs on the strength and nature of protein-protein interactions more globally. A negative B₂₂ indicates attractive interactions and should be avoided.

Experiments for B22 and k_D measurements have a similar set-up procedure

Without getting too deep into the mathematics required for the calculations, both k_D and B₂₂ experiments require replicates of your sample at about five different concentrations, from 0.5 mg/ml to 10 or 20 mg/ml.

A light scattering measurement is conducted on each sample, and – depending on which type of light scattering you’re talking about – a graph is produced.

For k_D, the graph is concentration vs diffusion coefficient, D. For B₂₂, the graph is concentration vs Kc/r_Θ. The resultant slope of the linear regression for the measurement is used to find the k_D or B₂₂.
It’s worth noting that it’s possible to set up one set of dilution samples and obtain both the k_D and B₂₂, depending on the instrument you’re using.

Careful sample handling is critical to get reliable data

Meticulous sample preparation is necessary to get reliable data. For SLS in particular, at these low concentrations there can be a lot of measured variability from the scattered light, which makes it difficult to confidently fit a line.

• Filter your samples first, to ensure you are only measuring the effects of your protein sample and its buffer
• Prepare the filter by first pushing some buffer through it, before filtering and collecting your sample
• Always make measurements of at least duplicates, if not triplicates
• Mix your samples thoroughly but gently via pipette – never vortex – as you dilute out
• Be cautious that you’re not picking up extra sample on the outside of your pipette tip
• Create a concentration range with at least 4 different concentrations, and be sure you can reach at least 10 mg/ml for the highest concentration

A sample k_D measurement informing on optimal formulation buffer

Salt often helps with protein solubility when a protein is at lower concentrations, as it interrupts the electrostatic interactions between two protein molecules. However, as the protein gets higher in concentration, that electrostatic protection may become a detriment as more protein molecules start competing against the salt for water molecules to remain solvated.

The interplay between salt concentration, pH, and additives has a dramatic impact on protein solvation, but varies from protein to protein. This is why it’s so important to assess self-association prior to scale-up.

In the above example, the sample with salt has a negative slope, and therefore k_D, while the low salt sample has a positive slope/k_D. When choosing between these two formulation buffers, it is most ideal to select the buffer without high salt concentration.

Self-association measurements are critical at early stages, as they have an outsize impact on the long-term developability of a therapy destined for patient use. Though self-association experiments are a bit more time- and sample- consuming to set up, they are well worth it in terms of determining if your biologic will not be aggregated or viscous upon clinical administration.

Watch this lunch and learn video to get an overview of self-association prediction strategies.