(326g) Osmotic Second Virial Coefficient as a Useful Tool to Optimize Self Assembly of Virus-like Particles

A virus-like particle (VLP) is the non-infectious protein coat of a virus. A VLP vaccine for cervical cancer and genital warts, based on the structure of human papillomavirus, has been developed and has been shown to give 100% immunoprotection in clinical trials [1]. Such high levels of protection are rare, highlighting the potential health benefits realizable through the large-scale manufacture of VLPs. This vaccine is manufactured through the self-assembly processing of VLPs obtained from a yeast-based expression system [2]. This breakthrough process heralds a new era in biomolecular engineering necessitating detailed engineering studies into the scale-up of self-assembly processes for VLP manufacture [3]. There is a clear need to understand the quantitative fundamentals underpinning this new class of process technology, to facilitate the delivery of new VLP products to society in the quickest possible time. Emerging potential products requiring urgent process engineering input include vaccines for pandemic influenza [4] and gastroenteritis [5], as well as potential novel products for gene delivery [6].

Work related to murine polyomavirus, which is closely related to human papillomavirus, is reported here. Murine polyomavirus is being “engineered” as a DNA delivery vehicle based on self-assembly processing [6]. The polyomavirus major capsid protein (VP1) can be expressed as a recombinant protein and purified with affinity chromatography. After subsequent cleavage of the affinity tags, pentamers of the VP1 protein have the ability to self-assemble in vitro into VLPs, provided the correct physicochemical process environment is imposed. In vitro self assembly of polyomavirus-like particles from recombinant VP1 pentamers often involves incubation with calcium or/and ammonium sulfate at high ionic strength and low protein concentration. This method, however, usually yields polymorphic capsid-like structures of varying size [7]. The reasons for this process inefficiency are not well understood, necessitating optimization studies.

The self assembly of VLPs can be considered somewhat analogous to protein crystallization because both processes rely on delicate approach of the precursors, in correct configuration, to form higher-order structures. Studies relating protein crystallization to osmotic second virial coefficient (B22), which is a measure of protein-protein interactions in a dilute solution, have been undertaken [8]. For B22 more positive than -1x10-4 mol.mL/g2, solubilities of proteins in the solution are too high for the formation of crystals. For values of B22 less than -8x10-4 mol.mL/g2, the protein-protein attractive interactions are excessively strong that there is insufficient time for protein monomers to orient into a crystallisation form, resulting in the formation of amorphous precipitates [9].

In this study, we investigated the relationship between osmotic second virial coefficient and self assembly of polyomavirus-like particles. By coupling size exclusion chromatography with multiangle light-scattering, B22 of the VP1 pentamers was screened efficiently under various solution conditions which are known to favor self assembly of VLPs. We related the B22 data to yield and quality of VLPs assembled under the same set of solution conditions and constructed a “self assembly slot” to optimize the chemical self-assembly processing of VLPs.

1. Villa, L.L., et al., Prophylactic quadrivalent human papillomavirus (types 6, 11, 16, and 18) L1 virus-like particle vaccine in young women: a randomised double-blind placebo-controlled multicentre phase II efficacy trial. Lancet Oncology, 2005. 6(5): p. 271-278. 2. Volkin, D.B. and H. Mach, Human papilloma virus vaccine with disassembled and reassembled virus-like particles (US 6,245,568). 2001: US Patent office. 3. Pattenden, L.K., et al., Towards the preparative and large-scale precision manufacture of virus-like particles. Trends in Biotechnology, 2005. 23(10): p. 523-529. 4. Pushko, P., et al., Influenza virus-like particles comprised of the HA, NA, and M1 proteins of H9N2 influenza virus induce protective immune responses in BALB/c mice. Vaccine, 2005. 23: p. 5751-5759. 5. Tacket, C.O., et al., Humoral, mucosal, and cellular immune responses to oral Norwalk virus-like particles in volunteers. Clinical Immunology, 2003. 108(3): p. 241-247. 6. May, T., S. Gleiter, and H. Lilie, Assessment of cell type specific gene transfer of polyoma virus like particles presenting a tumor specific antibody Fv fragment. Journal of Virological Methods, 2002. 105(1): p. 147-157. 7. Salunke, D.M., D.L.D. Caspar, and R.L. Garcea, Polymorphism in the assembly of polyomavirus capsid protein VP1. Biophysical Journal, 1989. 56: p. 887-900. 8. George, A., et al., Second virial coefficient as predictor in protein crystal growth. Methods Enzymol, 1997. 276: p. 100-110. 9. Curtis, R.A., J.M. Prausnitz, and H.W. Blanch, Protein-protein and protein-salt interactions in aqueous protein solutions containing concentrated electrolytes. Biotechnology and Bioengineering, 1998. 57(1): p. 11-21.