Structural snapshots of V/A-ATPase reveal the rotary catalytic mechanism of rotary ATPases

V/A-ATPase is a motor protein that shares a common rotary catalytic mechanism with FoF1 ATP synthase. When powered by ATP hydrolysis, the V1 domain rotates the central rotor against the A3B3 hexamer, composed of three catalytic AB dimers adopting different conformations (ABopen, ABsemi, and ABclosed). Here, we report the atomic models of 18 catalytic intermediates of the V1 domain of V/A-ATPase under different reaction conditions, determined by single particle cryo-EM. The models reveal that the rotor does not rotate immediately after binding of ATP to the V1. Instead, three events proceed simultaneously with the 120˚ rotation of the shaft: hydrolysis of ATP in ABsemi, zipper movement in ABopen by the binding ATP, and unzipper movement in ABclosed with release of both ADP and Pi. This indicates the unidirectional rotation of V/A-ATPase by a ratchet-like mechanism owing to ATP hydrolysis in ABsemi, rather than the power stroke model proposed previously for F1-ATPase.

It is good to see a nanodisc being used to recreate a more native like membrane bilayer. Why is Vo more poorly resolved? Although a lower resolution why is the analysis of the different states in V1 not carried through into Vo. Are there any interesting observations in Vo which reflect the different nucleotide bound states?
It would be good to see a broader analysis for what has been shown with the F and V ATPase family to expand on the impact.
Line 121, states 1 and 2 were improved, what about state 3?
Line 151 although high resolution, ~3A is not typically referred to as "atomic" resolution.
Line 288, how likely is it that this system has 3 ATP's bound at any one time, how often would it naturally experience "saturation" concentrations? The primary source of ATP in archaea or the thermophilic bacterium, Thermus thermophilus (Tth) is the A-ATP synthase (in the manuscript called V/A-ATPase), which shares several structural features with eukaryotic V-ATPases, and is evolutionarily distant from the F-ATP synthases that fulfill this role in other bacteria and eukaryotes. Metabolism in archaea is coupled to the generation of a H+-and/or Na+potential across the membrane, both of which can provide the energy for ATP synthesis by the A-ATP synthase. Whereas the F-ATP synthases in prokaryotes and eukaryotes catalyze ATP synthesis at the expense of an electrochemical ion potential, the evolutionarily related V-ATPases function as ATP-driven ion pumps, and are unable to synthesize ATP under physiological condition. In recent years, the Yokoyama lab has contributed significant to the structural and mechanistic understanding of the Tth V/A-ATPase.
In the study presented by Kishikawa et al., the authors describe the atomic cryo-EM structures of four different reactions conditions using the mutant enzyme A/S232A, T235S, which enabled the team to generate a nucleotide-free enzyme, and to design three respective defined nucleotide-conditions. The studies are described in detail and are state of the art. The resolution of the structures resolved, allow the interpretation of the data described in the manuscript with the focus of the three catalytic AB dimers during rotation. Such focus differs from the rotary studies of single molecule rotation studies, in which the rotary elements subunit γ and ε are in focus of the published studies.
The structures presented reveal, that the conformational changes in the three catalytic AB-dimers occur simultaneously, rather than in sequence, as has been shown for F-ATP synthases. The authors propose that rotation of the Tth V/A-ATPase proceeds via the tri-site model with the protein progressing through a two nucleotide bound state and a three nucleotide bound state. They also reveal that the unidirectional rotation occurs via a ratched-like mechanism driven bt ATP cleavage instead of a powerstroke mechanism as described for F-ATPases. Finally, the data confirm that only 120° degree steps occur during rotation of the Tth V/A-ATPase as described for the A-type F-ATPase (Sielaff et al., J. Biol. Chem. (2016) Dec 2;291(49):25351-25363) and as in contrast to F-ATPases, for which substeps have been described. It is a pity that the detailed publication of Sielaff et al., which described in detail the differences and similarities of archaeal A-ATP synthases, thermophilic and mesophilic ATP synthase molecular motors, is not cited and discussed in the manuscript.
In summary, the studies described in the manuscript by Kishikawa et al., are well designed and performed, and contribute to novel insights into the rotary mechanism of this molecular engine.

Our responses to Reviewer #1 comments,
We are very grateful to your detailed evaluation of our research and for your very positive comments. We have closely read the comments you have given to us and responded to each of them, as below. We hope that these responses will meet your requirements.

REVIEWER COMMENTS
Reviewer #1 (Remarks to the Author): The manuscript by Kishikawa et al., provides significant new structural details on the V/A-ATPase and is an impressive study in terms of the data collected and the number of structures determined. This work will be significant in the field and should provide new insights into the V/A-ATPase and its comparison to other members of the field. With advances in cryoEM has come a depth of new information on rotary ATPases and their mechanism and this work will contribute to those studies. There are a few areas which it would be good to see considered and discussed within the manuscript; How relevant are the concentrations of nucleotides used compared to what would be expected in the cell and how might differences influence the structure.

Response
The concentration of ATP in the cell has been reported to be on the order of millimolar [ Imamura et al PNAS 2009 15651-6]. Therefore, the structure obtained at 6 mM ATP should represent the intracellular state. It is likely that the structure at 50 mM ATP does not exist in the cell. However, as mentioned in the text, the structure of the V/A-ATPases are almost unchanged by nucleotide binding, so it is likely that the determined structures in this study are relevant regardless of the higher ATP concentration found in the cell.
It is good to see a nanodisc being used to recreate a more native like membrane bilayer. Why is Vo more poorly resolved? Although a lower resolution why is the analysis of the different states in V1 not carried through into Vo. Are there any interesting observations in Vo which reflect the different nucleotide bound states?

Response
The lower resolution of the Vo part may be due to the fluctuation of the Vo domain relative to the V1 domain. Indeed, the resolution of the DF subdomain connecting V1 and Vo becomes lower towards the Vo side (see local resolution maps of holo-V/A-ATPase in Supplementary information). We expect that it is possible to obtain a high-resolution structure of the Vo by focusing on the Vo domain. We will report the results elsewhere if we obtain the Vo structures under various reaction conditions. It would be good to see a broader analysis for what has been shown with the F and V ATPase family to expand on the impact.

Response
I agree with the reviewer's point. A similar analysis of FoF1 will be performed in the future and compared with our findings on V/A-ATPase in order to provide a broader context to the function of the rotary ATPases. Such a detailed analysis is not possible here.
Line 121, states 1 and 2 were improved, what about state 3?

Response
As the reviewer suggested the resolution of the V1 domain of states1 and 2 were improved using the focused refinement with signal subtraction techniques. However, in the case of state3, the focused refinement did not work well. Because the signal subtraction was applied to a very small number of particles (about 5,000 particles), it is assumed that the refinement process did not converge. Since the structure of the V1 domain in state 3 is expected to be similar to that in states 1 and 2, it does not affect the conclusions of our paper.
Line 151 although high resolution, ~3A is not typically referred to as "atomic" resolution.

Response
Accordingly, we removed "atomic" and rewrote this sentence as follows:

Line 151
We identified a cryoEM structure of the original....... Line 288, how likely is it that this system has 3 ATP's bound at any one time, how often would it naturally experience "saturation" concentrations?

Response
As mentioned earlier, since the cell is at a saturating concentration of ATP, this enzyme is likely to always have three nucleotides bound to the catalytic sites. In this state, the ATP binding dwell time is very short, therefore, the V/A-ATPase adopts V3nuc structure at ATP saturated condition.
Line 306 demonstrates not demonstrate Line 648, space after Cryo-EM

Response
We have modified the sentences as indicated.

Responses to Reviewer #2 comments,
We are very grateful for the appreciation of the significance of our research. In addition, we would like to thank you for your very helpful and positive comments. We believe that the comments we received have enhanced our research. The responses to each comment is given below in red. We hope that these responses will meet your requirements. The primary source of ATP in archaea or the thermophilic bacterium, Thermus thermophilus (Tth) is the A-ATP synthase (in the manuscript called V/A-ATPase), which shares several structural features with eukaryotic V-ATPases, and is evolutionarily distant from the F-ATP synthases that fulfill this role in other bacteria and eukaryotes. Metabolism in archaea is coupled to the generation of a H+-and/or Na+-potential across the membrane, both of which can provide the energy for ATP synthesis by the A-ATP synthase. Whereas the F-ATP synthases in prokaryotes and eukaryotes catalyze ATP synthesis at the expense of an electrochemical ion potential, the evolutionarily related V-ATPases function as ATP-driven ion pumps, and are unable to synthesize ATP under physiological condition. In recent years, the Yokoyama lab has contributed significant to the structural and mechanistic understanding of the Tth V/A-ATPase. The structures presented reveal, that the conformational changes in the three catalytic ABdimers occur simultaneously, rather than in sequence, as has been shown for F-ATP synthases.
The authors propose that rotation of the Tth V/A-ATPase proceeds via the tri-site model with the protein progressing through a two nucleotide bound state and a three nucleotide bound state.
They also reveal that the unidirectional rotation occurs via a ratched-like mechanism driven bt ATP cleavage instead of a power-stroke mechanism as described for F-ATPases. Finally, the data confirm that only 120° degree steps occur during rotation of the Tth V/A-ATPase as described for the A-type F-ATPase (Sielaff et al., J. Biol. Chem. (2016) Dec 2;291(49):25351-25363) and as in contrast to F-ATPases, for which substeps have been described. It is a pity that the detailed publication of Sielaff et al., which described in detail the differences and similarities of archaeal A-ATP synthases, thermophilic and mesophilic ATP synthase molecular motors, is not cited and discussed in the manuscript.
In summary, the studies described in the manuscript by Kishikawa et al., are well designed and performed, and contribute to novel insights into the rotary mechanism of this molecular engine.

Response
Thanks to the valuable comments from reviewer 2. By using a gold nano rod for high time resolution rotational experiments, Sielaff et al. showed that hydrolysis-driven rotational motion occurs after the catalytic dwell, consistent with the findings from our study. Therefore, we now cite this paper and have added the following sentence to the discussion. Line 368 ...are likely to share the same rotary mechanism. In fact, observation of rotation with high time resolved rotation analysis using a tiny gold rod showed that bacterial F-ATPase and an A-ATPase share the same rotation mechanism. For both rotary ATPases, a 120° rotation step together with ATP hydrolysis occurs after the catalytic dwell under ATP-saturated conditions.