Glass-like relaxation dynamics during the disorder-order transition of viral nucleocapsids*

Assemby is a crucial step in the life cycle of all viruses. It occurs within the host cell via a complex albeit accurate self-assembly process following the synthesis of viral components by the cellular machinery. The simplest viruses are made up of a protein shell called capsid protecting genome in the form of nucleic acids, as is the case for example with cowpea cholortic mottle virus (CCMV), a nonenveloped, icosahedral, single-stranded (ss)RNA virus infecting a variety of bean. Upon in vitro mixing, purified CCMV capsid proteins rapidly bind on RNA, and the resulting amorphous nucleoprotein complexes slowly relax through a disorder-order transition into 30-nm sphere-like nucleocapsids. The relaxation dynamics is ill known to date and experimental studies are hampered by the lack of techniques with sufficiently high spatiotemporal resolution.

By using time-resolved small-angle X-ray scattering with synchrotron source, we investigated the relaxation dynamics of CCMV-based nucleoprotein complexes with various RNA. Quite interestingly, we found out that the mean square radius of gyration〈Rg²〉decayed according to an exponential integral function rather than a single exponential function. This evolution could be analytically recovered by assuming that capsid subunits overcome uniformly-distributed energy barriers to acquire the right orientation and position in order to fit the final ordered structure. The maximal relaxation timescale τ with genomic RNA was estimated to be 2.2±0.2 s at 25 µM of capsid subunits and 45±15 s at 60 µM. With nongenomic RNA, τ remained within a range of a few seconds. By fitting the scattering intensities with a polydisperse mass fractal model, we observed in all cases that correlation length was stabilized slightly earlier than fractal dimension, which suggested that even though the complexes reached their final size after ~0.7 s, their compaction and/or ordering still took a little longer. We hope this study will initiate further investigations on relaxation dynamics, notably through numerical simulations, in order to understand quantitatively virus self-assembly.