High-resolution cryo-EM structure of urease from the pathogen Yersinia enterocolitica

Urease converts urea into ammonia and carbon dioxide and makes urea available as a nitrogen source for all forms of life except animals. In human bacterial pathogens, ureases also aid in the invasion of acidic environments such as the stomach by raising the surrounding pH. Here, we report the structure of urease from the pathogen Yersinia enterocolitica at 2 Å resolution from cryo-electron microscopy. Y. enterocolitica urease is a dodecameric assembly of a trimer of three protein chains, ureA, ureB and ureC. The high data quality enables detailed visualization of the urease bimetal active site and of the impact of radiation damage. The obtained structure is of sufficient quality to support drug development efforts.

Lines 293-316 A possibility for the short distance between the Ni(II) ions that should be discussed is the deprotonation of the bridging hydroxide to the oxide form, either by some sort of radiation damage or by local pH effects.
Line 616 Change K. aerogenes with S. pasteurii (PDB codes 2UBP and 3UBP are related to S. pasteurii and not K. aerogenes) For all figures The software used for production of the structural representation of protein pictures should be indicated.

Sincerely, Stefano Ciurli
Reviewer #2 (Remarks to the Author): In this manuscript, Righetto et al. describe the structure of Yersinia enterocolitica urease at high resolution, obtained by cryo-EM. Their major findings include the unusual dodecameric organization of the urease complex, the mapping of regions associated with this organization, and the analysis of radiation effects upon the description of metallocenters. So far the dodecameric organization has not been observed outside of the Helicobateriaceae family, making the authors' findings highly attractive. The bacterial survival under acidic conditions seems to drive the recurrence of dodecameric ureases, as speculated by the authors. This high-resolution structure is not only relevant in the context of cryo-EM state of the art applications, but also due to the observations derived from the methods involved in its obtaining and analysis. The conclusions are solidly supported by the results presented, and the methods are described thoroughly, allowing for prompt reproduction of the experiments. This manuscript is well polished in its current form, providing intriguing insights into the evolution of an evolutionarily peculiar enzyme, urease.
Specific points: - Figure 1, panel B, shows the isoform JBURE-II of C. ensiformis urease with 725 aa. It should be replaced by the full length urease, with 840 amino acid residues (doi:10.1016/j.bbapap.2011.07.022) -Supplementary figure 5, as it appears in the manuscript, is blurred . -Supplementary figure 6 is confusing. The same color (gray) is used to indicate sequences that did not align and protein stretches not resolved by the Cryo-EM. Please use different colors to illustrate these details.
-Please check the reference list -some references show errors in the authors' name, such as Kappaun et al, 2018.
-Yersinia enterocolitica urease has been purified and biochemically characterized previously. At least one reference should be given.

Reviewer #3 (Remarks to the Author):
Review of NCOMMS-20-23905 High-resolution cryo-EM structure of urease from the pathogen Yersinia enterocolitica The authors present the structure of the Y. enterocolitica urease by cryoEM. The structure is important, well done and the model is nicely supported by the experimental data. The authors have also provided extensive analysis of the details of the structure determination, and point to some interesting features that are probably related to radiation damage, all of which are very welcome. I recommend publication without delay after some minor revisions.

Specific points to address
It's not clear weather the changes presented in Fig. 6 are due to movement of the sample, global radiation damage, or specific damage to the active site. The authors should address this point in the text, but in a way that does not require additional experiments. Please also include a global B-factor vs frame plot, using the per-frame reconstructions already calculated, in the supplement to help clarify if the changes presented in fig 6 are site specific or part of the global degradation of the map with dose.
Other minor corrections L33 Suggest you remove the "better than" and replace with something less provocative like "sub" or just 2 Å. I think the authors would be hard pressed to demonstrate the difference between a 1.98 vs a 2.00 vs a 2.05 Å map so they shouldn't place such value judgements in the abstract.
L39 "our data" avoid possessive tense here -it's the world's data. Perhaps "The structure" L83 "extremely acidic" please provide pH or range L86 unless the authors wish to demonstrate the significance of the 3 digit of precision, I suggest you round this to 2.0. You are welcome to include as many digits as you like when you deposit it in the database.
L138 not clear what is meant by "annealing". movement? L92 predicate missing L93 Last sentence belongs in the discussion or just omit. You have not actually discovered a drug in this work so should refrain from drug-discovery claims.
L167 to *a* local

REVIEWER COMMENTS
Reviewer #1 (Remarks to the Author): The paper by Timm Maier and co-workers describes the obtainment of a high resolution (< 2 Å) structure of urease from Yersinia enterocolitica using cryo-EM. It is a remarkable result that should be published. The main conclusions of the work are that cryo-EM can be used to obtain valuable structural data for ureases from bacterial human pathogens in the absence of an otherwise critical crystallisation step, with the aim to support drug development for this virulence factor in several diseases caused by ureolytic microorganisms. This result is of extraordinary novelty and potential and will certainly be of interest to the wider scientific community working towards drug design and structure-activity relationships in critical enzymes, potentially changing the future approach to these goals. The proven capability to extract quasi-atomic molecular details using cryo-EM and not X-ray crystal diffraction opens new fields in the pursue of novel drug development. The paper should be published with minor revisions as I indicate below, to improve the manuscript impact: The indicated reference has been added (now line 59).
Lines 59-63 Add that the two Ni ions are bridged by a hydroxide ion that acts as the nucleophile, and add the following two references here to support this claim: Mazzei, L.; Cianci, M.; Benini, S.; Ciurli, S. We have included a sentence in the manuscript to clarify this point, with the indicated references (now lines 60-61).
Lines 60-62 This is not correct: the current hypothesis of the mechanism entails that, after the nucleophilic attack on the carbonyl carbon of urea, a proton is transferred from the bridging hydroxide to the distal NH2 group, not coordinated to Ni(II), yielding a nascent ammonia molecule.
We have corrected our statement to reflect the current hypothesis on mechanism (now lines 63-64). The indicated reference has been added (now line 270).
Lines 262-263 See comment above (lines 60-62) for the description of the mechanism; change accordingly.
The mechanism explanation has been corrected as suggested (now lines 270-273).
Lines 264-265 This is wrong: in the current most accepted mechanistic hypothesis, His325 acts as a base in its neutral form, which on one side allows the flap to close and on the other allows for the stabilization of the C(urea)-NH3+ group to be stabilized after proton transfer from the bridging hydroxide after it has formed the O-C(urea) bond following nucleophilic attack. Lines 282-284 Explain how the protons are visualized; if, as I understand, the position of the protons is deduced from the H-bonds network derived from inter-residue interactions, the statement should be corrected. Only neutron scattering can detect protons.
As stated in lines 355-357 of our current manuscript (lines 342-344 in the initial submission), whereas X-rays are scattered by the electron density distribution of the sample, electrons are scattered by the atomic nuclei and by the electrostatic charge distribution, i.e. "the net charge of the electrons and nucleus of atoms" [1]. Transmission electron microscopy (TEM) images are therefore projections corresponding to the integrated Coulomb potential across the sample, rendering the reconstructed 3D maps from this technique fundamentally different from X-ray crystallography density maps. Protons are then visible by TEM as they do scatter electrons, but not X-rays. While our reconstruction is not of high enough resolution to visualize separated atoms as in true atomic resolution, at ~2 Å "bumps" in the EM Coulomb potential map can already clearly indicate the position of hydrogen atoms, as is the case of the protonated H251 in Fig. 5b, for example. For further clarification, we refer the reviewer to recent atomic resolution protein structures solved by cryo-EM, which provide unambiguous visualization of protons as well as further details on their imaging [2,3]. We consider that our statements regarding the visualization of protonation states do not require correction. We recognize that the wording in this section was misleading and tried to clarify at an earlier point in the manuscript (now lines 199-205 and 303). We are not aiming to provide a full list of all S. pasteurii and K. aerogenes urease structures, but choose the highest resolution urease structure available (PDB 5OL4) and an X-ray structure of similar nominal resolution (PDB 1EJW) to comparatively assess the quality of our reconstruction.
Lines 293-316 A possibility for the short distance between the Ni(II) ions that should be discussed is the deprotonation of the bridging hydroxide to the oxide form, either by some sort of radiation damage or by local pH effects.
This important possibility is now discussed in lines 334-339 and 371-373 of the revised manuscript.
Line 616 Change K. aerogenes with S. pasteurii (PDB codes 2UBP and 3UBP are related to S. pasteurii and not K. aerogenes) The organism name has been corrected accordingly (now line 661).

For all figures
The software used for production of the structural representation of protein pictures should be indicated.
The software was indicated in the section "Structure analysis" of the Methods. We moved this to a new Methods section, "Structural representations and figure generation", for clarity (now lines 527-531).

Sincerely, Stefano Ciurli
Reviewer #2 (Remarks to the Author): In this manuscript, Righetto et al. describe the structure of Yersinia enterocolitica urease at high resolution, obtained by cryo-EM. Their major findings include the unusual dodecameric organization of the urease complex, the mapping of regions associated with this organization, and the analysis of radiation effects upon the description of metallocenters. So far the dodecameric organization has not been observed outside of the Helicobateriaceae family, making the authors' findings highly attractive. The bacterial survival under acidic conditions seems to drive the recurrence of dodecameric ureases, as speculated by the authors. This highresolution structure is not only relevant in the context of cryo-EM state of the art applications, but also due to the observations derived from the methods involved in its obtaining and analysis. The conclusions are solidly supported by the results presented, and the methods are described thoroughly, allowing for prompt reproduction of the experiments. This manuscript is well polished in its current form, providing intriguing insights into the evolution of an evolutionarily peculiar enzyme, urease. The blurred Supp. Figure 5 has been replaced by a high-resolution version.
-Supplementary figure 6 is confusing. The same color (gray) is used to indicate sequences that did not align and protein stretches not resolved by the Cryo-EM. Please use different colors to illustrate these details.
We have changed the figure and the respective legend accordingly. Now, unmatched residues due to differences in sequence are shown in blue, while residues that have not been resolved in the cryo-EM structure are shown in gray.
-Please check the reference list -some references show errors in the authors' name, such as Kappaun et al, 2018.
Reference Kappaun et al, 2018 has been updated with the correct details (now line 579). The reference list has been edited and corrected.
-Yersinia enterocolitica urease has been purified and biochemically characterized previously. At least one reference should be given.
We extended a sentence in the introduction (now lines 88-89) to mention the previous biochemical characterization of Y. enterocolitica urease by Bhagat & Virdi, 2009. Reviewer #3 (Remarks to the Author):

Review of NCOMMS-20-23905
High-resolution cryo-EM structure of urease from the pathogen Yersinia enterocolitica The authors present the structure of the Y. enterocolitica urease by cryoEM. The structure is important, well done and the model is nicely supported by the experimental data. The authors have also provided extensive analysis of the details of the structure determination, and point to some interesting features that are probably related to radiation damage, all of which are very welcome. I recommend publication without delay after some minor revisions.

Specific points to address
It's not clear weather the changes presented in Fig. 6 are due to movement of the sample, global radiation damage, or specific damage to the active site. The authors should address this point in the text, but in a way that does not require additional experiments. Please also include a global B-factor vs frame plot, using the per-frame reconstructions already calculated, in the supplement to help clarify if the changes presented in fig 6 are site specific or part of the global degradation of the map with dose.
We rule out sample movement as a significant source of blurring in this analysis because each reconstruction has been calculated after a new particle polishing run (Zivanov et al, 2019) using 5-frame running averages along the exposure, as explained in the methods (lines 517-519 in the revised manuscript). This is now also clarified in lines 320-323. We have added the global B-factor vs. exposure plot as a black dashed line to Fig. 6c. This plot shows that the nickel ions are damaged more quickly than the rest of the protein structure right from the start of the irradiation as we have implied previously. This is now discussed in lines 339-342 and 373.
Other minor corrections L33 Suggest you remove the "better than" and replace with something less provocative like "sub" or just 2 Å. I think the authors would be hard pressed to demonstrate the difference between a 1.98 vs a 2.00 vs a 2.05 Å map so they shouldn't place such value judgements in the abstract.
The abstract has been changed to simply mention a resolution of 2 Å. L39 "our data" avoid possessive tense here -it's the world's data. Perhaps "The structure" We have replaced "our data" with "The obtained structure" in the abstract. L83 "extremely acidic" please provide pH or range Y. enterocolitica urease has been shown to survive pH as low as 1.5. This is now mentioned explicitly in line 86 of the revised manuscript.
L86 unless the authors wish to demonstrate the significance of the 3 digit of precision, I suggest you round this to 2.0. You are welcome to include as many digits as you like when you deposit it in the database.
The sentence has been changed to simply mention an overall resolution of 2 Å (now line 90).
L138 not clear what is meant by "annealing". movement?
Yes, we meant the movement resulting from the annealing of the vitreous ice layer after being irradiated as described in Brilot et al, 2012. The sentence has been modified to clarify this point (now line 140).

L92 predicate missing
We have reworded the sentence to correct it (now lines 96-97).
L93 Last sentence belongs in the discussion or just omit. You have not actually discovered a drug in this work so should refrain from drug-discovery claims.
We have omitted the last sentence of the introduction. However, we kept the mention of the potential of cryo-EM for drug design in the discussion.
L167 to *a* local The correction has been made (now line 169). The line width was reduced to facilitate visualization. L345 The last sentence is not clear and not supported by the paper. Radiation damage in this context is dose dependent, not time dependent. Reword or omit.
The sentence has been reworded for clarity, omitting the mention of temporal resolution (now lines 374-376).
L386 Include which grids were used for each structure, hole size etc. A citation to Ermantraut et al. 1998 should be added.
We have provided more information about the type of grid used and added the corresponding citation to Quantifoil grids (now line 416).
L587 "(left)" should be moved to before "filtered" The figure legend has been changed as requested (now line 626).

SupL13 typo '
The typo has been corrected.

SupL17 typo '
The typo has been corrected. Sup Fig 2 the B-factor in the lower right is not physical and probably meaningless. Suggest omission.
We changed the figure and the legend to clarify that this is the global sharpening (contrast restoration) B-factor estimated from the Guinier plot as described in Rosenthal & Henderson, 2003.

Sup Fig 9 mesh is again too dense
The line width was reduced to facilitate visualization.
Sup Table 3 "Å/dose" need units of dose here.
The units have been added accordingly (Å/e -).