Elucidating the role of metal ions in carbonic anhydrase catalysis

Why metalloenzymes often show dramatic changes in their catalytic activity when subjected to chemically similar but non-native metal substitutions is a long-standing puzzle. Here, we report on the catalytic roles of metal ions in a model metalloenzyme system, human carbonic anhydrase II (CA II). Through a comparative study on the intermediate states of the zinc-bound native CA II and non-native metal-substituted CA IIs, we demonstrate that the characteristic metal ion coordination geometries (tetrahedral for Zn2+, tetrahedral to octahedral conversion for Co2+, octahedral for Ni2+, and trigonal bipyramidal for Cu2+) directly modulate the catalytic efficacy. In addition, we reveal that the metal ions have a long-range (~10 Å) electrostatic effect on restructuring water network in the active site. Our study provides evidence that the metal ions in metalloenzymes have a crucial impact on the catalytic mechanism beyond their primary chemical properties.

1 Firstly, we thank the editor and the reviewers for the careful reading of our manuscript and the positive and constructive comments. Below, the reviewers' comments are addressed, with our point-by-point responses in bold.

Reviewer #1 (Remarks to the Author):
The manuscript by J. K. Kim et al. provides many new crystal structures of different metal variants of human carbonic anhydrase II with and without a substrate molecule captured in its catalytic center. Specific knowledge and techniques were employed since the catalyzed reaction of the analyzed enzyme is very fast and the substrate is gaseous. The authors analyzed metal coordination geometry, its effect on catalytic activity and water molecule network. The results will be interesting and useful to the carbonic anhydrase research field, but also could be used for better understanding of other metalloenzyme systems. I believe the information could be employed for more precise computer simulations/ docking procedures of small molecule inhibitors for metalloenzymes. The manuscript is clearly written, with adequate citation and optimal manuscript length. I would suggest adding a short discussion about a possible alternative catalytic activity that may emerge for different metallovariants of the enzyme, as was recently reported by the same group (DOI: 10.1107/S2052252520000986 ) for copper-CA II.
To address the reviewer's pertinent suggestion, we have added a brief description and two new references on the alternative catalytic activities of the CA metallovariants on page 4 of the revised manuscript as below (highlighted in blue).
"The zinc ion can be substituted by other physiologically relevant transition metal ions such as Co 2+ , Ni 2+ , Cu 2+ , Cd 2+ , and Mn 2+ which results in drastic changes in the catalytic activity of CA II (~ 50% active to completely inactive) 21 . It has been also reported that the metal substitutions may induce alternative catalytic activities of CA II other than CO 2 /HCO 3conversion 27 , for instance, reduction of nitrite to nitric oxide in presence of copper. 28

Reviewer #2 (Remarks to the Author):
The paper by Kim et al. reports a detailed study by means of X-ray crystallography on the effect of metal ion substitution in the carbonic anhydrase II active site, showing that the change of coordination geometry modulates the catalytic efficiency and the network of water molecules in the enzyme active site. The paper is rather interesting, technically sound and logically arranged; however, in my opinion it does not possess sufficient novelty to be published in a so high impact factor journal such as Nature Communication.
We thank the reviewer for the kind words and would like to take this opportunity to highlight the novelty of the current manuscript. In metalloenzyme research, a longstanding problem remains as to why metalloenzymes often show dramatic changes in their activity when the native metal ion is substituted by chemically similar, but distinct non-native metal ions.

Reviewer #3 (Remarks to the Author):
This work seeks to systematically evaluate the contribution of the catalytic metal of human carbonic anhydrase II using X-ray crystallography. Sixteen high-resolution structures were obtained under various conditions from apo-enzyme to Zn, Co, Ni, or Cu loaded with and without CO2 pressure at pH 7.8 or 11. This manuscript represents a substantial undertaking which provides high-quality, information-dense data sets for the metallobiochemistry field. My major concern is that the compactness of the presentation may lessen impact for readers who are not already intimately familiar with the CA II system. Also, some conclusions (detailed below) are drawn too definitely considering that this study does not include any biochemical or enzymological characterizations. Major Concerns: Figure 2 condenses too much information and too many ideas. It contains information about i) changes in coordination geometry upon metal binding/substitution (a-e), ii) geometry of CO2 binding (f,g,h), iii) geometry of HCO3 binding (i,m), iv) how pH affects CO2 vs HCO3 binding (h,m), v) effect of Zn on CO2 binding (k), and iv) steric clashes of hypothetical CO2 binding (n,o). While this is certainly space efficient, I think there is enough interesting data here to justify 2 or 3 separate figures with more narrow focus. If this advice is taken, consider moving panels from Supplementary figure 3 to the main text.
To present a surfeit of information condensed in Figure 2 in a more digestible manner, we have rearranged the figure into four new figures (Figures 2, 3, 4  bipyramidal coordination. f-j) At 20 atm of CO 2 pressure, Zn-CA II shows clear binding of CO 2 as in apo-CA II while maintaining tetrahedral metal coordination, but Co-CA II at pH 11.0 shows superposition of CO 2 binding (~ 50% occupancy) with tetrahedral coordination and HCO 3 binding (~ 50% occupancy) with octahedral coordination. Ni-CA II maintains octahedral coordination with HCO 3 binding, but Cu-CA II shows disordered electron density in the CO 2 /HCO 3 binding site. k) Comparison of CO 2 binding at apo-CA II and Zn-CA II (white). l) Upon CO 2 binding (white) in Zn-CA II, W Zn is located at the center of hypothetical tetrahedral structure made up of Zn 2+ ion, Thr199-O2, position (1) (close to W1), and position (2)  Distance between the position (2) and C atom of CO 2 is 0.36, 1.55 and 2.93 Å for Zn-, Ni-and Cu-CA II, respectively.

Supplementary Fig. 3. Substrate and product binding in Co-CA II at pH 7.8 (a-b) and 11.0 (c-d).
At pH 11.0, the active site shows dual binding of CO 2 and HCO 3 when cryocooled under 20 atm CO 2 pressurization. However, at pH 7.8, the active site shows full binding of HCO 3 molecule even when the Co-CA II crystal is untreated with CO 2 gas. It is likely that the captured HCO 3 is converted from the CO 2 absorbed in the crystal from air. The electron density (2F o -F c , blue) and the difference map  showing octahedral coordination even in absence of added CO 2 . It is likely that the captured HCO 3 is converted from the CO 2 absorbed in the crystal from ambient air. pressure, Ni-CA II maintains octahedral coordination with HCO 3 binding. b) Compared to the W Zn geometry in Zn-CA II (Fig. 3d), the nucleophilic attack geometry of W Ni ′ has steric hindrance on CO 2 molecule (adapted from Zn-CA II, 20 atm) and is distorted away. Distance between the position (2) and C atom of CO 2 is 1.55 Å. c) Cu-CA II shows only disordered electron density in the CO 2 /HCO 3 binding site. d) The nucleophilic attack geometry of W Cu has steric hindrance on CO 2 molecule (adapted from Zn-CA II, 20 atm) and is significantly distorted away. Distance between the position (2) and C atom of CO 2 is 2.93 Å.

15
Page 5, lines 89-95: The authors note that carbonate binds Co-CA II in a bidentate fashion, leading to a higher binding affinity which in turn reduces overall activity. While bidentate rather than monodentate binding may correlate with a higher binding affinity, it is not sufficient to draw such a conclusion, especially regarding the solution state behavior of an enzyme.
In the native CA II, the rate limiting step of the overall catalytic reaction is the proton transfer process. If the rate limiting step in the Co-CA II remains same as in the native CA II, we agree that the stronger bidentate binding does not necessarily reduce the overall activity. Accordingly, we removed the conclusive statement in the revised manuscript as: "In the transformed octahedral geometry, the HCO 3 molecule is bound in a bidentate mode to the Co 2+ ion along with an additional water molecule. Compared to the monodentate binding mode in Zn-CA II, the negative charge on the bidentate HCO 3 can be distributed among the two oxygen atoms bound to Co 2+ ion, allowing stronger product binding to the metal ion ( Supplementary Fig.1a-f). The higher affinity for HCO 3 in Co-CA II seems related to its lower catalytic activity, as HCO 3 is more difficult to displace to repeat the catalytic cycle." Supplementary Fig. 1c&f: Unless proton positions have been experimentally determined (Neutron diffraction or spectroscopically) or supported by computational efforts, it is best to leave their positions unassigned.
To ensure rapid CA catalytic activity, it is essential that the negatively charged bicarbonate is released readily from the positively charged Zn 2+ ion. Researchers on CA have studied possible binding configurations of bicarbonate to Zn 2+ ion, and suggested that the specific hydrogen bonding between the monodentate bicarbonate and the OH group of Thr199 (as shown in Supplementary Fig 1c) is essential for the facile release of bicarbonate (the specific configuration shifts the negative charge within the bicarbonate away from the metal ion). This suggestion was supported by the mutational studies on Thr199, in which the catalytic activity of a single Thr199 mutant decreased significantly to just a few fractions of its native form. Thus, the bicarbonate binding configuration in Supplementary Fig 1c has been supported experimentally and is widely accepted in the CA research community. In addition, we believe that the bicarbonate binding configuration in Supplementary Fig 1f could  " Fig. 1.

………….
c) The CO 2 hydration reaction mechanism of Zn-CA II. First, CO 2 binds to the active site, leading to a nucleophilic attack by the zinc-bound hydroxyl ion onto CO 2 . HCO 3 thus formed is subsequently displaced by the water molecule inflowing through EC. The Page 6, line 117: Avoid using "This" as the subject of a sentence. Also, adding a qualifier to this sentence seems appropriate.
As the reviewer suggested, we have changed the corresponding sentence in the revised manuscript as below.
"This explains The inefficient substrate binding and the unfavorable distorted geometry explain the complete enzymatic inactivity of Cu-CA II. " The second and third column of Figure 3 appear to show different angles of the same data set. Such presentation is usually accompanied by a curved arrow labelling the degree of rotation. In this case, following convention would also reduce the number of panel labels needed by 5. Also, the degree to which the conformation of His64 is altered does not seem significant enough to warrant inclusion in the figure. These values may be better presented in a table in the SI.
Based on the reviewer's comments, we have modified Fig. 3 and relabeled it as Fig. 6