Structural analysis of herpes simplex virus by optical super-resolution imaging

Laine, Romain F.; Albecka, Anna; van de Linde, Sebastian; Rees, Eric J.; Crump, Colin M.; Kaminski, Clemens F.

We were also able to resolve and quantify the capsid spatial offset in HSV-1 particles, using precise localization of both the centre of the capsid and those of the other protein layers (tegument or envelope). We observed offsets in good agreement with data previously obtained by cryo-ET 45 . Our method is based on the superposition of two-dimensional (2D) dSTORM data obtained from several particles, and the analysis of radial distributions of the offsets taking the three-dimensional (3D) structure of the particle into account. The spatial offset between the capsid and several tegument proteins in a related alphaherpesvirus, pseudorabies virus, has also been reported in an elegant study by Bohannon et al. 46 Our analysis expands on this approach, which uses conventional fluorescence imaging, through the application of super-resolution microscopy. Super-resolution microscopy provides detailed images of individual virus particles, and so it is possible to determine their centre with greater confidence and accuracy. For example, non-uniformity in the fluorescence signal that may arise from incomplete labelling, or geometric defects, are directly observable by SMLM but are hidden when using diffraction-limited imaging techniques. Although we did use conventional fluorescence data for the determination of the capsid centre, the high copy number of VP26 and the rigid structure of the capsid ensure that an isotropic signal is obtained from mTurquoise; therefore, the centroid of the ensemble localization provides an accurate determination of the capsid centre. Surprisingly, our measurements of capsid offset relative to pUL37 gave a larger value (24 nm) than relative to VP16 (7 nm), which is difficult to reconcile with the notion that these two tegument proteins are both co-ordinated by interaction with a similar region of VP1/2. However, it is well established that there is a significantly greater number of VP16 proteins per virion (4600) than either pUL37 or VP1/2 (both o200) 53 . Furthermore, VP16 interacts with several other tegument and envelope proteins in addition to VP1/ 2 (ref. 54 ). As the structure and orientation of pUL37 and VP16 within the tegument have yet to be determined, whether bound to VP1/2 or not, it is unclear whether this discrepancy in spatial offset for VP16 and pUL37 is a genuine reflection of a different intra-virion distribution of the proteins or not. Future studies with other specific antibodies against VP16 and pUL37 should clarify this.