The structural basis for release factor activation during translation termination revealed by time-resolved cryogenic electron microscopy

Background

Determination of the three-dimensional structure of macromolecules can give us insight into their biological function. Crystallography or cryogenic electron microscopy (cryoEM) can provide us one or more snapshots of a particular reaction cycle; however, in complex multi-step reactions, it is often difficult to determine the order of particular intermediates. In addition to this, some transient, short-lived steps may be missed all together. By adding a time dimension to our 3D structures, we can get a clearer picture as to the true mechanism.

CryoEM, unlike X-ray crystallography, does not require sample crystallisation. Analysis is performed in aqueous solution, which is rapidly frozen and therefore is thought to more closely mimic in vivo conditions. Further to this, time resolved cryoEM can be used to measure changes in millisecond or longer timescales, which may be required for complex biological processes. In recent years, time-resolved cryoEM has become a much more feasible experiment, as detectors and software in the field has vastly improved (1), enabling the throughput required to measure several time points. One final important point to note with cryoEM is that, using suitable software, multiple structures can be solved in one experiment. This allows the identification and quantification of differentially populated states, which might be used to estimate reaction kinetics.

In cryoEM samples are added to pre-prepared grids, blotted to remove excess sample and frozen in liquid ethane. This process takes several seconds and so is not compatible with the capture of millisecond states. For this reason, several alternative strategies have been used to reduce the time before plunge freezing. In one example, a chemical can be “caged” such that it is only active under certain conditions, for example, when it is exposed to light. In this example, samples would be applied to grids, blotted, exposed to light and immediately frozen (2). Another strategy is to apply sample 1 to the grid, blot, then to spray the second sample on this grid and immediately freeze (3). One problem with this set up is that it is very difficult to tell if all areas of the grid are sprayed with sample 2. The final method used in this preprint mixes the two samples by microfluidics, and then sprays on the empty grid and freezes the combined samples. This ensures that wherever there is sample, both samples are present, which simplifies analysis. In addition, the length of the chamber post mixing and/or delay before freezing can be used to determine the reaction time to be measured.

The ribosome is a complex machine made up of multiple components. This machine has been extensively characterised by cryoEM and so makes a good test case for the time-resolved experiment in this preprint. Further to this, there is indirect (non-structural) data to suggest the existence of a variety of short-lived states. In this work, the authors measure states during translation termination.

Results

The bacterial ribosomal release complex (Met-Phe-Phe-tRNA in P site and STOP codon in A site) was mixed with either release factor (RF) 1 or 2, sprayed and plunge frozen. Two millisecond-scale time points were characterised by cryoEM: 24 and 60 ms.

The two time points showed different populations of states, particularly the pre- and post-accomodation state. The pre-accomodation state contained a compacted release factor, whereas the post-accomodation state showed an extended release factor. In the experiments with RF1, after 24 ms 25% of ribosomes were found in the pre-accomodation state, with this number reduced at 60 ms.

In the pre-accomodation state, the tripeptide product (MFF) is still attached to the tRNA in the P site, but in the post-accomodation state the tripeptide is observed in the exit tunnel (at the 5 hour time point the tripeptide is not observed).

Conclusions and comments

This preprint describes the quickest time-resolved mixing/spraying cryoEM published to date. There are many biological processes that occur in the low millisecond timescale, which could be analysed with this method. At present, there are very few time-resolved EM set ups available; however the case can now be made for the more widespread use of time-resolved cryoEM.

It is satisfying how the authors have used orthogonal techniques to choose the time points for study. In this case they compare their results to quench-flow and FRET methods used by other authors.

One point worthy of note regarding the data in this preprint involves the differences between RF1 and RF2. The data suggest that in RF2-mediated release the pre-accomodation state is even more short-lived, as at 60 ms there is no observed pre-accomodation state with RF2, but there is around 10% present with RF1. The authors do not mention this in their article; however it remains to be seen if this difference has any significance in vivo.

Why I chose this article

I chose this article both because of the scientific content and the methodology used. Time-resolved cryoEM is a field that has been in existence for decades (3); however, with the improvements in data collection and handling, we now have the throughput and resolution to really gain insight into biologically relevant questions. Fu et al. are able to see different ribosomal sub-states at 24 ms, but by waiting until 60 ms, these states are not observed. Without this technology we would be unable to structurally characterise these states in the typical cryoEM timescales. It is clear that this strategy could greatly benefit structural biology, particularly when a system has previously been characterised by cryoEM and there are transitions to measure in the millisecond timescale.

References

1. Kühlbrandt W. The Resolution Revolution. Science. 2014;343(6178):1443.
2. Shaikh TR, Barnard D, Meng X, Wagenknecht T. Implementation of a flash-photolysis system for time-resolved cryo-electron microscopy. Journal of structural biology. 2009;165(3):184-9. PubMed PMID: 19114106. Epub 12/10.
3. Berriman J, Unwin N. Analysis of transient structures by cryo-microscopy combined with rapid mixing of spray droplets. Ultramicroscopy. 1994 1994/12/01/;56(4):241-52.

 

doi: https://doi.org/10.1242/prelights.6216

Read preprint (No Ratings Yet)