- Article
- Open Access
- Published:
Structural dynamics of single SARS-CoV-2 pseudoknot molecules reveal topologically distinct conformers
Nature Communications volume 12, Article number: 4749 (2021)
Discussion
These results confirm the suggestion from simulations and cryo-EM imaging that the SARS-CoV-2 frameshift signal can form a variety of different structures. The state N, which was by far the most common conformation under physiological-like conditions (handle far from stimulatory structure, with Mg2+ but without oligos), unfolded through the full length of the pseudoknot at moderately high force. This conformation was suppressed significantly by occlusion of the 5′ end by the duplex handle, precisely as would be expected for a 5′-end threaded structure such as those seen in cryo-EM images on and off the ribosome10,15 or predicted from simulations18. In contrast, the conformation unfolding at lower force, N′, while occurring ten-fold less frequently than N under normal conditions, increasingly replaced N as occlusion of the 5′ end suppressed the occupancy of N, as would be expected for a conformation in which the 5′ end remains unthreaded. Unthreaded conformers have been predicted computationally18 but not yet characterized structurally in experiments, although some individual cryo-EM images show the straight morphology expected for unthreaded conformers, in contrast to the bent shape of threaded conformers10.
The picture of the pseudoknot folding and unfolding that emerges from this work is illustrated in Fig. 5. Stem 1 always folds first, followed by sequential folding of stem 3 and then stem 2. The orientation of the 5′ end at the moment of stem 2 formation leads to two distinct fold topologies that cannot interconvert: 5′-threaded or -unthreaded. These two topologies give rise to distinct unfolding behaviors: higher forces for the threaded fold, lower forces for the unthreaded fold. The partitioning of the folding at the point when stem 2 forms—depending on whether or not the 5′ end is lying across the stem 1/stem 3 junction, as required for threading—ensures the presence of both threaded and unthreaded conformers, with the minority unthreaded state populated at some finite level, similar to what was seen in the folding of the Zika exoribonuclease-resistant RNA (xrRNA)21. This folding mechanism is dependent on stem 2 folding last; as it happens, stem 2 is also predicted by mfold47 to be the least stable thermodynamically, whereas stem 1 is expected to be the most stable, so that the folding is ordered by the relative stabilities of the stems as seen previously for two-stem pseudoknots34,48. Intriguingly, this order is also the same one in which the stems would refold co-translationally as the ribosome leaves the frameshift signal: stem 1 is at the 5′ end and would be expected to refold first, whereas stem 2 is at the 3′ end and would not be able to refold until after the whole pseudoknot emerged from the ribosome. Notably absent from this picture of the folding, however, is a third fold that was predicted computationally with the 3′ threaded through the stem 2/stem 3 junction18, which makes sense given that this fold requires stem 1 to form last instead of first.
Inga kommentarer:
Skicka en kommentar