Heterologous Expression of Human Norovirus GII.4 VP1 Leads to Assembly of T=4 Virus-Like Particles

Abstract

Human noroviruses are a leading cause of acute gastroenteritis, yet there are still no vaccines or antivirals available. Expression of the norovirus capsid protein (VP1) in insect cells typically results in the formation of virus-like particles (VLPs) that are morphologically and antigenically comparable to native virions. Previous structural analysis of norovirus VLPs showed that the capsid has a T=3 icosahedral symmetry and is composed of 180 copies of VP1 that are folded into three quasi-equivalent subunits (A, B, and C). In this study, we determined the cryo-EM VLP structures of two GII.4 variants, termed CHDC-1974 and NSW-2012. Surprisingly, we found that greater than 95% of these GII.4 VLPs were larger than virions and 3D reconstruction showed that these VLPs exhibited T=4 icosahedral symmetry. We found that the T=4 VLPs showed several structural differences to the T=3 VLPs. The T=4 particles assemble from 240 copies of VP1 that adopt four quasi-equivalent conformations (A, B, C, and D) that form two distinct dimers, A/B and C/D. The T=4 protruding domains were elevated [~]21-[A] off the capsid shell, which was [~]7-[A] more than the previously studied GII.10 T=3 VLPs. A small cavity and flap-like structure at the icosahedral twofold axis disrupted the contiguous T=4 shell, a consequence of the D-subunit S-domains having smaller contact interfaces with neighboring dimers. Overall, our findings that old and new GII.4 VP1 sequences assemble T=4 VLPs might have implications for the design of potential future vaccines.\n\nIMPORTANCEThe discovery that the GII.4 VLPs have a T=4 symmetry is of significance, since this represents the first known T=4 calicivirus structure. Interestingly, the GII.4 2012 variant shares 96% amino acid identity with a current GII.4 VLP vaccine candidate sequence, which suggests that this vaccine might also have a T=4 symmetry. Our previous results with these GII.4 VLPs showed functional binding properties to antibodies and Nanobodies that were raised against T=3 (GII.10) VLPs. This suggests that the T=4 VLPs were antigenically comparable to T=3 particles, despite the obvious structural and size differences. On the other hand, these larger T=4 VLPs with novel structural features and possibly new epitopes might elicit antibodies that do not recognize equivalent epitopes on the T=3 VLPs. Further structural and binding studies using a library of GII.4-specific Nanobodies are planned in order to precisely investigate whether new epitopes are formed.