However, for the development of pharmaceutical products based upon virus-like particles (VLPs), the application of reliable analytical tools is of great importance. For quality control purposes, the analytical methods need to be sufficiently sensitive to detect and quantify even small varieties between different active pharmaceutical ingredient (API) bulk materials, and varying formulations upon manufacture and storage. For the assessment of physical properties of viruses and VLPs, three main techniques are established: transmission electron microscopy (TEM) (4–6); dynamic light scattering (DLS), often referred to as photon correlation spectroscopy (PCS) or quasi-elastic light scattering (QELS) (7–9); and size-exclusion chromatography (SEC) (10–12). The utilization of TEM can be ascribed to the high resolution enabling accurate particle analysis. Inherent drawbacks of this technique are random sampling — instead of an overall sample analysis — and time-consuming preparation and measurement procedures.

DLS is described as a powerful, fast, nondestructive method for the determination of the average hydrodynamic radii and size distributions of particles in their natural liquid environment. Thereby, increasing average sizes and polydispersity indices indicate aggregation of the particles (13). However, the size resolution is rather low and the obtained size distribution is rather inaccurate; the particles must differ in radius by about twofold to be resolved. Furthermore, the accurate quantification of different particle fractions is not feasible (14,15). SEC represents the gold standard for the analysis of the physical stability of proteins and also is used for the characterization of large biomolecules like VLPs. The drawbacks of SEC are its limited resolution capacity of especially high molecular weight molecules and the fact that only the analysis of soluble aggregates is possible (16). Consequently, for formulation development studies, there is a strong need for a more sensitive analytical tool that allows separation, characterization, and quantification of different VLP fractions.

Figure 1: Separation principle of asymmetrical flow FFF.

A sample is injected into the hollow channel, which is at the downside ("accumulation wall") limited by an ultrafiltration membrane with a certain molecular weight cut-off (open for solvent, but not for species to be analyzed). After a focusing step, samples are eluted by a parabolic channel flow. At the same time, a cross-flow perpendicular to the carrier flow is applied that "pushes" the dispersed particles in the direction of the accumulation wall, where the cross-flow exits the channel via the membrane. Thereby, analytes are partitioned into regions with different flow velocities in dependence of their diffusion properties. In essence, the larger a particle, the smaller its diffusion coefficient acting in opposite direction than the convection from the cross-flow. Thus, the larger a particle, the stronger it is influenced by the cross-flow and the longer it is retained in the channel. As a consequence, smaller particles elute faster than larger ones. Due to the open architecture of the separation channel, particles in the size range of several nanometers up to a few microns are accessible for separation by field-flow fractionation (23).

During the past decades, asymmetrical flow FFF has gained more and more attention and was applied successfully for example for the analysis of the size and size distribution of monoclonal antibodies (16), liposomes (24), lipid–DNA complexes (25), nanoparticles (26,27), and viruses (28). Additionally, by coupling this technique with multiangle laser light scattering (MALLS) detection, it became possible to obtain the molecular weight distributions of the fractionated species (28–30).

In this context, it was the objective of the present work to develop an asymmetrical flow FFF method as an alternative tool for the analysis of VLP formulations and to investigate whether asymmetrical flow FFF can provide a better insight into VLP compositions than DLS and SEC.