Bicarbonate Inhibition of Carbonic Anhydrase Mimics Hinders Catalytic Efficiency: Elucidating the Mechanism and Gaining Insight toward Improving Speed and Efficiency

Article abstract of DOI:10.1021/acscatal.8b04077

Carbonic anhydrase (CA) mimics are often studied with a focus on the hydration of CO₂ for atmospheric carbon capture. Consequently, the reverse reaction (dehydration of HCO₃^-) has received minimal attention, so much so tha

Full text of DOI:10.1021/acscatal.8b04077

Subscriber access provided by University of Kansas Libraries
Article
Bicarbonate inhibition of Carbonic Anhydrase mimics
hinders catalytic efficiency: Elucidating the mechanism and
gaining in-sight towards improving speed and efficiency
Jonathan Rains, Kerry O'Donnelly, Thomas Oliver, Rudiger
Woscholski, Nicholas J. Long, and Laura M. C. Barter
ACS Catal., Just Accepted Manuscript • DOI: 10.1021/acscatal.8b04077 • Publication Date (Web): 03 Jan 2019
Downloaded from http://pubs.acs.org on January 5, 2019
Just Accepted
“Just Accepted” manuscripts have been peer-reviewed and accepted for publication. They are posted
online prior to technical editing, formatting for publication and author proofing. The American Chemical
Society provides “Just Accepted” as a service to the research community to expedite the dissemination
of scientific material as soon as possible after acceptance. “Just Accepted” manuscripts appear in
full in PDF format accompanied by an HTML abstract. “Just Accepted” manuscripts have been fully
peer reviewed, but should not be considered the official version of record. They are citable by the
Digital Object Identifier (DOI®). “Just Accepted” is an optional service offered to authors. Therefore,
the “Just Accepted” Web site may not include all articles that will be published in the journal. After
a manuscript is technically edited and formatted, it will be removed from the “Just Accepted” Web
site and published as an ASAP article. Note that technical editing may introduce minor changes
to the manuscript text and/or graphics which could affect content, and all legal disclaimers and
ethical guidelines that apply to the journal pertain. ACS cannot be held responsible for errors or
consequences arising from the use of information contained in these “Just Accepted” manuscripts.
is published by the American Chemical Society. 1155 Sixteenth Street N.W.,
Washington, DC 20036
Published by American Chemical Society. Copyright © American Chemical Society.
However, no copyright claim is made to original U.S. Government works, or works
produced by employees of any Commonwealth realm Crown government in the course
of their duties.

Page 1 of 16
ACS Catalysis
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
Bicarbonate Inhibition of Carbonic Anhydrase Mimics Hinders Catalytic
Efficiency: Elucidating the Mechanism and Gaining Insight Towards
Improving Speed and Efficiency
Jonathan G. D. Rains, Kerry O’Donnelly, Thomas Oliver, Rudiger Woscholski*, Nicholas J. Long*,
Laura M. C. Barter*
Institute of Chemical Biology, Department of Chemistry, Imperial College, Molecular Science Research Hub, White City
Campus, London, W12 0BZ, UK.
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
*Email: r.woscholski@imperial.ac.uk; n.long@imperial.ac.uk; l.barter@imperial.ac.uk
Abstract. Carbonic Anhydrase (CA) mimics are often studied with a focus on the hydration of CO
2
for atmospheric carbon
-
capture. Consequently, the reverse reaction (dehydration of HCO
3
) has received minimal attention, so much so, that the rate-
limiting step of the dehydration reaction in CA mimics is currently unknown. The rate-limiting step of the hydration reaction
is reported to be the bicarbonate-bound intermediate step, and thus is susceptible to product inhibition. It is not, however,
clear if this inhibition is a consequence of an increase in the rate of the competing dehydration reaction or resulting from the
strong affinity of bicarbonate to the mimic. To address this, insight into the dehydration reaction kinetics is needed. We
therefore report the most comprehensive study of a CA mimic to date. The dehydration profile of the fastest small-molecule
CA mimic, ZnL1S, was characterised and consequently evidence for the rate-limiting step for the dehydration reaction was
seen to be the bicarbonate-bound intermediate step, much like the hydration reaction. This experimental validation of the
rate-limiting step was achieved through a variety of methods including NMR experiments and the effect of inhibitors,
substrate concentration and metal centre on activity. With this understanding, an improvement in the favourability of the
rate-limiting step was achieved resulting in decreased bicarbonate inhibition. Thus, an increase in the mimic’s kcat for both
reactions was observed, resulting in the largest rate constants of any small-molecule CA mimic reported to date (28,093 and
-
1
-1
5
-1
5
s
79 M
s
for hydration and dehydration respectively). Enzyme-like kcat / k
m
values were obtained for ZnL1S (5.9 x 10 M
-
1
for CO
2
hydration) and notably there is only a difference of 2.5 orders of magnitude with the enzyme, the closest of any CA
mimic reported in the literature. The results from this work can be applied to the development and improvement of future
and existing mimics towards attaining increased activities.
-
Keywords. Carbonic Anhydrase, Biomimetic, HCO
Introduction. The capture of CO
3
dehydration, CO
2
hydration, rate-limiting step, stopped-flow
2
produced during
thus increasing hydration rates.24 While the CA mimics
and the natural CA enzyme have the same catalytic
industrial processes, or directly from the atmosphere, are
important processes and have received significant
attention in order to reduce the impact of climate change
4,
25
mechanism (Scheme 1)
significantly.
, their kinetics differ
1
-3
resulting from increased atmospheric CO
2
.
Frequently,
synthetic mimics of the enzyme Carbonic Anhydrase
4-6
SO H
(
CA), which catalyse the reversible reaction between
carbon dioxide hydration and bicarbonate dehydration,
are investigated as CO capture agents, for storage or
Most synthetic CA mimics replicate the
3
H
2
3, 7-15
other uses.
O
N
H
active site of the α class of mammalian CA’s and thus are
composed of a hetero-macrocycle scaffold coordinating
NH
SO H
Zn
3
N
1
6-19
zinc,
and are often designed for use in organic
N
9
solvents. There are reports of CA mimics that are
catalytically active in aqueous solutions at similar rates to
those that work only in organic solvents, albeit with rates
around 4-5 orders of magnitude slower than the natural
N
N
H
NH
HO S
3
1
8, 20-23
enzyme.
Chart 1. ZnL1S, a water-soluble small-molecule CA
mimic.
An example of a CA mimic, ZnL1S (Chart 1), that
works in aqueous environments and possesses one of the
2
4
fastest rates of CO
2
hydration to date. It coordinates a
The rate-limiting step of the natural enzyme is the
deprotonation of the bound water molecule, which, due to
the shuttling of protons, is essentially diffusion limited
zinc ion through a scaffold containing three histidine-like
imidazole ligands and a water molecule. Additionally, the
authors state that the ligand generates a hydrophobic
pocket within the structure, facilitating the direction of
26
(Step 1 in Scheme 1 highlighted in blue). For this reason,
it has one of the highest known enzymatic catalytic
activities, with a rate of CO hydration of ~10 molecules
2
per second per enzyme. . In contrast, the rate-limiting
6
CO
2
approach to the reactive zinc centre, thereby
hydration and
4, 25
overcoming the activation barrier of CO
2
1
ACS Paragon Plus Environment

ACS Catalysis
step in the hydration catalytic cycle of CA mimics is the
Page 2 of 16
Here, we present an extensive characterisation of the
often-ignored dehydration reaction for the fastest small-
molecule water-compatible CA mimic (ZnL1S). Kinetic
measurements using stopped-flow and other relevant
techniques were employed, providing the necessary
insights to improve the next generation environment-
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
bicarbonate release step (Step 4 in Scheme 1 highlighted
in red). This is due to the favourable low energy of the
bidentate bicarbonate-bound intermediate, and has been
9
-11, 17, 20, 27
observed in mimics using macrocycle ligands.
The consequence of the CA mimics’ rate-limiting step is
that they are susceptible to product inhibition by the
formed bicarbonate ion, thus decreasing the hydration
2
friendly CO capture catalysts. We report the rate-limiting
step for the dehydration reaction by a CA mimic for the
first time. Furthermore, this study achieves the highest
reported activities by a CA mimic. Finally, we show that
under specific conditions both the dehydration and
hydration reaction activities can be improved significantly
beyond what is currently reported. Thus, ZnL1S has the
potential to be the fastest mimic of CA known to date.
Following the evidence of the bicarbonate inhibition, this
work concludes with a comparison between the mimic
and the enzyme, including their analysis methodologies,
and consequently the true potential of ZnL1S is revealed.
9
-11, 17, 20, 27
catalytic rates.
In contrast, the surrounding
amino acid environment in the natural enzyme’s active
site likely facilitates bicarbonate release and an easier
water approach by forming an extensive hydrogen
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
28-30
bonding network.
Rate-limiting step for
H
H
Carbonic Anhydrase Enzyme
H+
H
O
O
R
Zn2+
NH
R
Zn2+
NH
N
N
N
N
R
1
R
R
N
R
N
N
R
R
N
R
H
R
H
R
NH
Experimental. Experimental information including
instrument details, compound characterisation and
reagent sources, are detailed in Supporting Information
1a.
NH
R
HCO3-
Rate-limiting step
for Mimics
4
2
CO2
H O
2
Synthesis of the ligand L1S and its metal
O
C
complexes. The ligand L1S was synthesised using an
O C O
H
38
H
adaption of the literature method
(detailed in
O
O
O
3
R
Zn2+
Supporting Information 1b) which resulted in an
improved purified yield (92% yield) and reaction time
R
Zn2+
NH
NH
N
N
N
N
R
R
R
(overnight) from the original synthesis procedure for the
R
N
R
N
N
H
N
R
24
R
ligand and complex by Nakata et.al.
In brief, 3,4-
H
R
NH
R
NH
diaminobenzenesulfonic acid was reacted with
nitrilotriacetic acid in a double condensation reaction in
ethylene glycol to form the ligand L1S. The complexes
R
Scheme 1. The mechanism of Carbonic Anhydrase and
its mimics. Step 1 (Blue): Deprotonation of a zinc-
bound water to generate (His)
Nucleophilic attack of CO
intermediate formed. Step 4 (Red): Displacement of
bicarbonate by water to generate the complex initial
(
ZnL1S, CuL1S, NiL1S and CoL1S) were synthesised by
-
reacting the ligand L1S with the appropriate hydrated
metal perchlorate salt in water and precipitating upon the
addition of ethanol. See Supporting Information 1c for
synthetic details and characterisation.
3
-Zn-OH . Step 2:
. Step 3: Bicarbonate-bound
2
-
state. The
R
groups on the zinc-coordinating
2 3
Kinetic measurement of CO hydration and HCO
imidazole’s represent either an attached benzene
moiety to form benzimidazole, like in ZnL1S, or as
part of a histidine amino acid, as is the case for the
active site of the mammalian α-Carbonic Anhydrase.
dehydration. The hydration and dehydration rate
measurements were undertaken by detecting the
associated pH change through the use of an appropriate
colorimetric pH indicator via stopped-flow equipment
24
4
(Applied
Spectrometer).
hydration or HCO3 dehydration was observed via the
absorbance change of a buffered indicator solution in
order to measure the change in Vhydr/dehydr over a
concentration range of catalyst, and to determine the
observed rate constant (kobs) for each reaction.
Photophysics
SX-20
Stopped-flow
3
9
For the application of industrial scale CO
from the atmosphere, efficient and environment-friendly
catalysts are needed.
2
capture
The pH time-dependence of CO2
-
3
, 7-15
To generate the next generation
of CA mimics the intrinsic product inhibition of the
hydration reaction needs to be understood before this
obstacle can be overcome. Given the reversibility of the
mechanism, it is not clear if the product inhibition is a
consequence of an increase in the dehydration reaction,
favoured by the presence of the bicarbonate complex, or
caused by a strong affinity of bicarbonate to the CA
mimetic. To address this more insight into the
dehydration reaction kinetics is needed. With the
24
Following the previously reported procedure, a 0.1
-5
M buffer solution containing 10 x 10 M indicator, 0.2 M
4
NaClO and varying concentrations of the zinc complex
(
ZnL1S) in water were loaded into one syringe and a CO
2
-
or HCO
3
solution in the other. CO
2
saturated solutions
(
2
33.8 mM) were prepared by bubbling CO gas through
emphasis of research focusing on CO
there is however a huge void of understanding
concerning the dehydration of HCO
2
hydration catalysts,
o
water for 1 hour at 25 C. The saturated solution was
3, 7-15
diluted appropriately to give the desired concentration of
-
3
by these CA mimics
-
CO
2
. HCO
3
solutions were prepared by dissolving NaHCO
3
1
8, 22, 31-37
and their rate limiting step.
-
3
in water to give the desired concentration of HCO . The
2
ACS Paragon Plus Environment

Page 3 of 16
ACS Catalysis
for 5 minutes in a 96-well microplate. 50 µl of 5 mM 4-
buffer-indicator pairs were chosen based upon their
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
similar pKa’s, and the pH of the buffer-indicator solution
was adjusted to match these pKa’s to maximize sensitivity
to pH change (Table S1).
nitrophenyl acetate was added to start the reaction. The
formation of 4-nitrophenolate was monitored at 348 nm
in
a Thermo Scientific Varioskan Flash Multimode
microplate reader for 2000 seconds. Final concentrations
of CO or HCO in 100 µl were 0 mM, 1.69 mM, 3.38 mM,
Upon mixing of the two syringes, the change in
-
o
absorbance over time was measured at 25 C. For each
2
3
6
.76 mM, 10.14 mM, 13.52 mM and 16.9 mM. Δ
buffer solution employed over the tested pH range, the
buffer factor (change in absorbance vs concentration of
absorbance / min was calculated by subtracting the final
absorbance from the starting absorbance and then
dividing by the rate constant as calculated from the slope
of time against absorbance.
-
HCl/ NaOH) and linearity of CO
absorbance vs concentration of CO
2
and HCO
3
(change in
-
2
/
HCO ) were
3
determined as controls (Figure S5 and S6 for the buffer
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
factor and substrate linearity respectively). Based upon
Results and Discussion.
-
2 3
the linearity of the response between CO or HCO
Synthesis and Catalytic Validation of ZnL1S. In this
study, the previously reported fastest water-compatible
concentration and change in absorbance, appropriate
-
2 3
concentrations of CO or HCO were chosen within the
small-molecule CA active site mimic ZnL1S, which uses the
linear range for the subsequent complex runs. At each
concentration of complex, the Vhydr (rate of hydration) and
24
nitrilotris(2-benzimidazolylmethyl)
ligand,
was
successfully synthesised and its catalytic properties
validated. The first step in the generation of the water-
soluble metal complex ZnL1S was the synthesis of the
ligand, L1S, as described previously in the experimental
and in more detail in the Supporting Information (Scheme
S1 and Figures S7 and S8). The L1S ligand was then
complexed with zinc as described in the experimental. The
synthesized ligand and the resulting complex, ZnL1S, were
V
dehyd (rate of dehydration) was measured and kobs
(
observed rate constant) was calculated as per the
procedure of Nakata et al. It is of note that not all of the
24
observed rate constants for hydration were obtained from
2
a
linear correlation (R >0.9) between catalyst
concentration and rate. In contrast, all dehydration kobs
were obtained from linear plots. kcat (catalytic rate
constant) was calculated as the pH independent catalyst
rate constant as per Nakata et al. and Koziol et al. for the
1
13
fully characterised by H and C NMR spectroscopy, mass
spectrometry and elemental analysis (Scheme S2 and
Figure S9). It is of note that due to the new ligand synthesis
procedure, as described in the methods, only the 665
+
20, 24
hydration reaction (kcat =
k
obs x (([H ]/Ka) + 1)) ,
and
+
Sun et al. for the dehydration reaction (kcat = (kobs x ([H ] +
Ka)) / [H ]). Each run of 3 repeats was undertaken 3
times and the average taken, to give N = 9.
+
35
isomer was seen in the product (whereas Nakata et al. saw
24
the 665 and 666 isomers
)which is of a benefit for
NMR studies of bicarbonate-bound intermediate.
characterisation and subsequent studies. See Chart S1 for
structures.
Studies of the catalysis mechanism of reaction were
1
13
o
carried out and monitored via H and C NMR at 25 C. For
Previously, CO
measured at temperatures of 5, 10 and 15 C. This data was
used to extrapolate rates to 25 C, resulting in a predicted
2
hydration rates of ZnL1S were
13
the dehydration reaction using C-labelled sodium
o
bicarbonate, three solutions were prepared in NMR tubes;
o
0
.01 M complex (ZnL1S), 0.01 M substrate (NaHCO
0.01 M complex mixed with 0.01 M substrate (ZnL1S +
NaHCO ). All three solutions were made up twice, once in
deuterated DMSO and once in D O (both H O-free). The
solutions were then analysed by NMR. For the hydration
3
) and
-1
-1 24
catalytic rate constant of 10,000 M s . While this
predicted rate constant makes ZnL1S the fastest reported
hydration small-molecule CA mimic, there is clearly a lack
of experimental data at “greener” ambient temperatures,
3
2
2
o
i.e. at 25 C, which would require no heating or cooling
1
3
reaction using C-labelled CO
2
, 1 atm (approximately 0.2
was added to the appropriate Young’s tap
NMR tubes using a Toepler pump. As before, the three
solutions (ZnL1S complex, CO substrate and ZnL1S
complex + CO substrate) were made up twice, once in
deuterated DMSO and once in D O. The subsequent
hydration NMR studies were carried out in D O with
or C labelled
(0.001 M, 0.01 M & 0.1 M) and one concentration
of NaNO (0.01 M) in the appropriate Young’s tap NMR
tube containing 0.01 M ZnL1S and 1 atm (approximately
inputs during the industrial process. Interestingly, the
dehydration properties of ZnL1S have been previously
1
3
mmol) of CO
2
o
-1 -1 37
studied at 25 C at pH 6 (kcat of 363 M s ), however only
using the less sensitive and less accurate weight loss
method, and not the stopped-flow method more
commonly reported for the hydration reaction.
-
2
2
2
2
13
2 3
To address this, both the CO hydration and HCO
increasing concentrations of NaHCO
NaHCO
3
dehydration of ZnL1S were analysed using the stopped-
3
o
flow method up to 25 C. (Table S2). We found a rate
3
-
1
-1
constant for CO
2
hydration of 9358 M
s for ZnL1S at
o
1
3
25 C, which is within reasonable agreement with the
previously predicted value (10,000 M s ). For the
dehydration reaction, we determined a catalytic
at 25 C, making ZnL1S the
fastest small-molecule CA mimics for HCO
0.2 mmol) CO
Bruker AMX-400 spectrometer at room temperature.
-Nitrophenyl acetate hydrolysis assay. The
2
. The solutions were all analysed using a
-
1
-1 24
-
HCO
3
4
-1 -1
o
rate constant of 103.1 M
s
hydrolysis of 4-nitrophenyl acetate to 4-nitrophenolate
with the addition of ZnL1S was monitored following the
-
3
dehydration
reported to date. It should be noted that the dehydration
rate constant for ZnL1S had been previously measured
using the less accurate weight loss method, and a catalytic
o
40
previously reported procedure at 25 C. 50 µl of 1 mM
ZnL1S (100 mM TAPS pH 8.2, 200 mM NaClO ) was
incubated with varying concentrations of CO or NaHCO
4
2
3
-1 -1
37
rate constant of 363 M
s was found. This variation in
3
ACS Paragon Plus Environment

ACS Catalysis
Page 4 of 16
measured values can be accounted for by the difference in hydration) were studied across multiple pH’s and
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
accuracy of the methodology employed. It is of note that
there was agreement with the literature for the hydration
reaction as the same stopped-flow methodology was
utilized. In conclusion, ZnL1S is confirmed to be the fastest
water-soluble small-molecule CA mimic for both
hydration and dehydration reactions.
substrate concentrations.
The dehydration reaction showed a linear decrease in
-
rate with increasing HCO
Plotting the ratio
concentration generated linear trends for all HCO
3
concentration (Figure S11).
-
V
dehyd
/
HCO
3
against catalyst
-
3
concentrations. Additionally, there was an inhibitive effect
with increasing substrate concentration and the effect was
greater at the higher concentration of catalyst (as seen by
the greater slope at higher mimic concentrations in Figure
S11). Furthermore, the kobs decreased linearly with
increasing substrate concentration (Figure 1 E-G).
Effect of Substrate Concentration and pH on
ZnL1S CA Activity. To investigate the rate-limiting step of
the dehydration reaction, the kinetic properties of ZnL1S
for both reactions (dehydration, and for comparison
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
4
ACS Paragon Plus Environment

Page 5 of 16
ACS Catalysis
1
2
3
4
5
6
7
8
9
A
B
5
4
3
2
1
00000
00000
00000
00000
00000
0
40000
3
2
1
0000
0000
0000
0
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
0.000
0.005
0.010
0.015
0.020
0.020
0.020
0.000
0.005
0.010
0.015
0.020
0.020
0.020
CO2 Concentration (M)
CO2 Concentration (M)
C
D
2000
1500
600
4
00
1
000
200
5
00
0
0
0.000
0.005
0.010
0.015
0.000
0.005
0.010
0.015
CO2 Concentration (M)
CO2 Concentration (M)
E
F
8
00
200
150
100
50
600
400
200
0
0
0.000
0.005
0.010
0.015
0.000
0.005
0.010
0.015
HCO3- Concentration (M)
HCO - Concentration (M)
3
G
6
0
0
0
0
4
2
0.000
0.005
0.010
0.015
0.020
HCO - Concentration (M)
3
Figure 1. The non-linear relationship between the kobs of ZnL1S for CO
and 7.4 respectively). At pH 7.4 (D), concentrations below 0.00676 M CO
between kobs of ZnL1S (0.5, 5 and 50 µM ZnL1S) for HCO
respectively). At pH 8.2 (G), concentrations below 0.005 M HCO
2
hydration across the studied pH range (A-D pH 10.8, 9.5, 8.2
were below the limit of detection. The linear relationship
dehydration across the studied pH range (E-G pH 6.3, 7.4 and 8.2
2
-
3
-
3
were below the limit of detection. kobs data is the slopes of the data
shown in Figure S10 B, D, F and H and Figure S11 B, D and F using 0.5, 5 and 50 µM ZnL1S (and 500 µM ZnL1S at pH 10.8). Error bars
are the error associated with the fit to generate kobs
-
Table 1. kobs values for the hydration of CO
and substrate concentrations.
2
and the dehydration of HCO
3
by ZnL1S across the studied range of pH values
5
ACS Paragon Plus Environment

ACS Catalysis
Page 6 of 16
pH 6.3 kobs
pH 7.4 kobs
(M-1 s-1)
pH 8.2 kobs
(M-1 s-1)
pH 9.5 kobs
(M-1 s-1)
pH 10.8 kobs
(M-1 s-1)
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
(
M-1 s-1)
0
.00169
0.00338
.00676
1282 ± 125.4
916.1 ± 54.16
466.6 ± 110.1 *
Not Measured
239.8 ± 90.00 *
Not Measured
150.5 ± 83.00 *
86.88 ± 97.82*
26426 ± 2021
17465 ± 1044
9397 ± 1358
258753 ± 4242
145698 ± 2563
77679 ± 2969
Not Measured
34739 ± 3224
Not Measured
18157 ± 1546
13346 ± 1545
Below LOD
0
154.10 ± 60.48 *
119.30 ± 48.41*
84.48 ± 36.36 *
70.26 ± 30.59*
56.05 ± 24.79 *
31.92 ± 12.84 *
132.5 ± 9.56
Hydration
CO ] (M)
0.00845
0.01014
0.01183
Not Measured
4333 ± 1609 *
Not Measured
1911 ± 962.5 *
1105 ± 769.7 *
Below LOD
[
2
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
0.01352
0
.0169
0
.0015
572.7 ± 21.39
491.6 ± 16.89
361.9 ± 15.77
Not Measured
255.6 ± 17.97
Not Measured
126.1 ± 7.998
51.88 ± 0.7247
Below LOD
0.003
.006
.0075
.009
0.0105
105.7 ± 11.57
88.02 ± 13.25
Not Measured
63.53 ± 7.114
Not Measured
51.90 ± 12.82
30.53 ± 10.26
0
49.53 ± 3.689
45.36 ± 3.865
40.77 ± 4.061
38.22 ± 3.140
35.67 ± 2.245
31.14 ± 2.601
Dehydratio
0
n
-
Below LOD
Below LOD
0
[HCO ] (M)
3
0
.012
.015
0
2
The data is the average of three repeats. * indicates that the plot of V/[CO
data is the slopes of the data shown in Figure S10 B, D, F and H and Figure S11 B, D and F using 0.5, 5 and 50 µM ZnL1S (and 500 µM ZnL1S at pH
0.8). Extra substrate concentrations were studied for pH 7.4 Hydration and pH 8.2 Dehydration to compensate for the loss of data points at lower
2
] vs [ZnL1S] for that [substrate] had an R <0.9. LOD= limit of detection. kobs
1
substrate concentrations.
For the hydration reaction, there was an exponential
decrease in rate with increasing CO concentration, across
2
all conditions (Figure S10). The inhibitory effect was
greater at higher concentrations of catalyst (as seen by the
greater slope at higher mimic concentrations in Figure
S10). Additionally, the catalyst’s observed rate constant
decreased exponentially with increasing CO
concentration (Figure 1 A-D). Upon plotting of Vhyd/CO
against catalyst concentration, the linearity of this trend
improved with decreasing concentrations of CO
indicating less bicarbonate inhibition (i.e. the R value of
the linear plot increased towards 1, Table S3). The
catalysts appear more inhibited by bicarbonate with
increasing catalyst concentration, due to the subsequent
increased bicarbonate produced. Thus, there is a decrease
in the expected improvement in activity with increased
first time any inhibition for the dehydration reaction by a
CA mimic has been observed.
Table 1 summarises the calculated kobs values for both
reactions across the pH and substrate concentration
ranges studied. As mentioned previously, a decrease in
obs is observed with increasing substrate concentration
for both reactions. Additionally, the expected increase in
hydration kobs and expected increase in dehydration kobs
with increasing and decreasing pH respectively, was seen.
k
(
k
obs
)
2
2
,
2
,
18, 33-34, 37
As generally observed in previous studies,
the
2
hydration reaction was faster than dehydration at pH 8.2
and above, whereas the dehydration reaction was faster at
pH 6.3. Interestingly, at pH 7.4, the two reactions had very
similar observed rate constants at higher substrate
concentrations. The hydration reaction was, however,
slightly faster at the same substrate concentration. This is
the first time that a CA mimic has been shown to have
similar observed rate constants for both reactions at the
same pH value. Furthermore, this is a change from the
enzyme. CA only exhibits faster catalysis for the
catalysts concentration, resulting in
decreasing linearity. significant improvement in
a plateau and
A
linearity was seen, however, at pH 10.8. This change in
inhibition at pH 10.8, is likely the result of the easier
release of carbonate, which is three times more abundant
than bicarbonate at this higher pH. This was previously
17, 22,
dehydration reaction over hydration at pH’s below 7,
34, 36, 41
whereas the mimic ZnL1S demonstrates this at
11
reported for a different CA mimic, thus validating the
neutral pH.
results seen.
In previous studies,18, 22, 33-37 much larger HCO
-
3
The differences between the effect of substrate
concentration on the hydration or dehydration reactions
suggests that the subsequent bicarbonate inhibition is
more prevalent in the hydration reaction than the
dehydration reaction. This is evident by the exponential
slope for the hydration reaction compared to the linear
slope for the dehydration reaction (Figure 1). This is the
-3
concentrations were used (of the order of x 10 M) and as
the data here shows, this leads to catalyst inhibition.
Therefore, previous studies concluded that hydration is
much faster at pH 7.4. This study shows that under certain
substrate conditions (i.e. at higher substrate
concentrations), however, it is possible for ZnL1S to
catalyse the dehydration reaction faster than the
6
ACS Paragon Plus Environment

Page 7 of 16
ACS Catalysis
-
hydration reaction at neutral pH values. This conclusion
The rate-limiting step of HCO
3
dehydration is however
currently unknown. Given the natural equilibrium
between HCO and CO , it is also unclear which species is
3 2
responsible for the observed inhibition of the dehydration
reaction.
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
could likely be applied to other CA small molecule mimics
as they share the same mechanism. The specific substrate
concentrations required would vary depending upon the
catalysts’ affinity for bicarbonate.
-
Based upon its mechanism, it is expected that the
relationship between hydration kcat (the catalyst’s rate
Capturing the Bicarbonate-bound Intermediate
via NMR. To provide evidence for the dehydration
reaction’s rate-limiting step, an NMR study was
undertaken to see if a bicarbonate-bound intermediate
+
constant independent of pH) and H concentration should
be linear. This is due to the activation of the catalyst with
increasing pH as the zinc-bound water molecule is
2
+
could be captured. Employing the Zn complex in aqueous
and non-aqueous solvents (D and/or deuterated
3
3
deprotonated (Scheme 1 Step 1). Surprisingly, however,
this relationship instead followed a power regression
model (Figure S12A). The non-linearity of this trend
2
O
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
DMSO), both the hydration and dehydration reactions
were studied. The study also allowed for a comparison of
each reaction’s NMR profile in each solvent.
suggests that there is a loss of inhibition with decreasing
+
H
concentration (increasing pH). This leads to greater
There was no bound bicarbonate intermediate
detected as part of the dehydration reaction in D O, due to
2
the speed of the reaction relative to a typical NMR scan
duration. It was apparent, however, that the reaction had
proceeded to completion, as observed though a loss of
improvement of the kcat values than expected from
activation of the catalyst alone. At higher pH values, as
-
well as the expected increase in hydration-active OH -
bound catalyst, there is an increased conversion to the
weaker binding carbonate over bicarbonate, resulting in a
faster release of product and thus decreasing inhibition.
-
HCO
3
and production of CO
2
and H
2
O (Figures S13 and
2+
S14). In DMSO, however, a bound Zn bicarbonate
intermediate (Figure 2), as well as the production of CO
and H O (Figures S15-17), could be observed as the
-
In contrast, the catalyst’s HCO
3
dehydration activity
2
+
increased linearly with increasing
H
concentration
2
(decreasing pH) (Figure S12B). This agrees with the
catalyst’s mechanism and previous investigations of other
mimic’s turnover was slowed in this solvent. This suggests
that the rate-limiting step of the dehydration reaction is
also at the bicarbonate-bound intermediate stage, as is the
case with the hydration reaction.
mimics, as more dehydration-active H
2
O-bound catalyst
We found that bicarbonate inhibition
significant limiting factor affecting the
2
2, 35-36
was formed.
was not
a
Evidence of the hydration reaction was not detected
in DMSO (Figures S18 and S19), due to the inability of the
bound water molecule to deprotonate as per the first step
of the catalyst’s mechanism (Scheme 1), and as seen for a
O, however, production of free
HCO by the hydration reaction was observed, although
3
the intermediate of the hydration reaction was not
detected. The latter is not surprising, given the fast
reaction times for the hydration reaction as compared to
the speed at which a NMR scan reaches completion
(Figures S20 and S21).
dehydration reaction. This contrasts with the hydration
reaction, which was significantly impacted by the
produced bicarbonate concentration, as seen by its non-
linear slopes in Figures 1 and Figure S12.
42
different CA mimic. In D
2
-
The results thus far have shown that the catalysis of
both the hydration and the dehydration reactions, by
ZnL1S, are inhibited at higher concentrations of substrate.
As pointed out previously (see introduction), the CO
2
hydration by ZnL1S and other mimics is rate-limited by
9-11, 17, 20, 27
the bicarbonate-bound state of the catalysts.
7
ACS Paragon Plus Environment

ACS Catalysis
Page 8 of 16
-
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
HCO3
ZnL1S
-
ZnL1S + HCO3
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
Figure 2. 13C NMR spectrum in DMSO comparing free bicarbonate, ZnL1S and ZnL1S with bicarbonate illustrating the ability
of ZnL1S to bind to bicarbonate. Free bicarbonate is seen at 160 ppm and bound bicarbonate at 167 ppm. ZnL1S peaks at 180,
165, 135, 117, 116, 114, 59 and 54 ppm. Peak at 57 ppm is due to ethanol used to enhance the solubility of sodium bicarbonate
with the complex in d-DMSO.
-), the
To visualise the bicarbonate-bound intermediate
through NMR during the hydration reaction, attempts
were made to slow down the reaction through the
addition of the hydration product HCO . By adding HCO
3 3
into the hydration reaction as catalysed by ZnL1S, the
activity of the catalyst should slow down as the
confirm the likely inhibitory species, (CO or HCO
inhibitory effect of CO and HCO mimics on ZnL1S
2 3
hydration activity was undertaken. It is well established
that CA enzymes can be inhibited by a range of anions such
as Cl , Br , NCS . These anions bind via replacement of the
coordinated water molecule or via direct insertion into the
active site, thus increasing the coordination number
around the zinc centre.
tested on other CA mimics, such as the well-studied mimic
zinc cyclen as well as ZnL1S.
2 3
-
-
-
-
-
-
equilibrium is pushed back towards the substrate CO
2
.
-
21-24
This effect on the hydration rate by addition of HCO
3
has
They have previously been
been seen previously with a different CA mimic, although
not via NMR but through the observed effect on the
9
-10, 17, 24
1
1
mimic’s kinetics via stopped-flow. Furthermore, the
In this study, a wider range of potential inhibitors
were chosen, including those previously studied. Potential
inhibitors were selected that are structurally and
-
3
addition of a product mimic, such as NO , should inhibit
ZnL1S sufficiently to allow the bicarbonate-bound
intermediate to be captured. With increasing
concentrations of HCO , evidence of increased zinc-bound
3
-
electronically related to either CO
2
or HCO
3
, the reactants
for the CA catalysed reactions. The commercially available
CA inhibitor acetazolamide was also tested to compare its
binding to the enzyme and its synthetic mimic. Charge,
geometry and localisation of electrons were all key factors
in choosing the inhibitors employed (Table S4 for details).
As can been seen from the data presented in Figure 3,
neutral species such as the linear CS
resembling CO ) and trigonal planar urea were not
inhibitory. Charged tetrahedral species showed a mixed
response. Phosphate was the only tetrahedral species
revealing significant inhibition, whereas sulfate and
bicarbonate was seen with an increasing upfield shift of
1
the aromatic H NMR peaks of the complex ligand (Figure
-
S22). Additionally, upon introduction of NO
3
, a slight
upfield shift was again seen indicating zinc-bound
bicarbonate or nitrate, although it is unknown which one
is bound. Regardless, this demonstrates that when the
hydration reaction is slowed, the intermediate rate-
limiting bicarbonate-release step can be captured.
2
or bulky DIC (both
2
In summary, it was shown that the catalyst behaves
as expected by undergoing both reactions under the
appropriate conditions. Most importantly however,
evidence of the intermediates of both reactions were
captured demonstrating the complex’s ability to strongly
bind bicarbonate.
perchlorate did not inhibit the ZnL1S catalysed CO
2
hydration. Single-charge inhibitors with planar geometry
(e.g. chloride, thiocyanate, nitrite, bicarbonate, formate
and nitrate) were all inhibitory, although nitrate and
thiocyanate were the most potent. Interestingly, the well-
known sulfonamide CA enzyme inhibitor acetazolamide
was the most effective inhibitor of the tested compounds.
Effect of CO
Mimics on ZnL1S CO
2
Hydration Substrate and Product
Hydration Activity. To further
2
8
ACS Paragon Plus Environment

Page 9 of 16
ACS Catalysis
The inhibitory constants for the most potent ionic
inhibitors, nitrate and thiocyanate, were therefore
surrounding amino acids in the enzyme’s active site, is
sufficient for the interaction of this important drug
molecule within the catalytic cavity of CA enzymes.
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
-
-5
24, 44-46
determined. The IC50 for NO
3
of 4.236 x 10 M is about two
-
-5
times smaller than that observed for SCN 8.791 x 10 M.
A two-sample t-test (16) = 27.34, p= <0.0001 shows that
there is a statistically significant difference between the
The effect of pH (9.5 versus 10.8) was examined via
the inhibition of ZnL1S by the tested compounds to see
how it affects the inhibitor profile. Specifically, the effect
that the ratio of bicarbonate to carbonate has either side
-
-
two IC50 values. Both NO
3
and SCN share a single
negatively charged planar geometry, which may reflect
the transition state of the Zn-bicarbonate complex (see
of their pKa of 10.3 was observed. Only NaHCO
significant effect, with a stronger inhibitory potency at pH
.5 than at pH 10.8.The increased concentration of the
3
showed a
-
Scheme 1 between step 3 and 4). In particular, NO
charge and geometry with HCO
3
shares
9
-
3
, and hence has been used
previously to explore bicarbonate binding to CA mimics.
weaker binding carbonate, over the stronger binding
bicarbonate, at higher pH was the reason for this
difference, as discussed earlier.
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
4
3
-
SCN shares only its linear shape with CO
2
and thus, may
with the shape of the
, making it a transition state mimic. Since the
-
well comprise the charge of the HCO
CO
3
The effect of inhibitors on the dehydration reaction
Figure S23) yields similar results to that of the hydration
2
(
inhibitory potency correlates with its affinity for the
catalyst, one can conclude that bicarbonate mimicking
inhibitors (e.g. nitrate) are better in binding to the catalyst
than the carbon dioxide mimicking test compounds (e.g.
reaction. It was observed that those compounds that most
similarly mimic bicarbonate (i.e. charged species) inhibit
ZnL1S dehydration most significantly at pH 6.3.
2
Interestingly, the addition of CO resulted in a decrease in
CS
2
; DIC). This in turn indicates a preference of the ZnL1S
. This inhibitor
activity of 27%. This can be attributed to the push back of
the equilibrium by introducing the product into the
reaction mixture, resulting in slower dehydration activity.
Overall, the dehydration reaction seems to be more
susceptible to inhibition by these mimics than the
hydration reaction, although the sequence of inhibitor
strength and compound is the same for both reactions.
-
catalyst for binding bicarbonate over CO
2
profiling, demonstrates that bicarbonate mimics are
stronger inhibitors of, and thus bind more strongly to,
ZnL1S than CO mimics. This corroborates the data
2
presented so far and provides further evidence that the
bicarbonate-bound intermediate is the most likely rate-
limiting step. This is in agreement with similar smaller
9-11, 17, 20, 27
2 3
Following this, study of the effect of CO and HCO
investigations for other CA mimics.
values for the CA enzyme inhibitor acetazolamide were
determined for the inhibition of the CO hydration
The IC50 and
mimics on the typical CA reactions and the effect of CO
2
k
i
-
3
and HCO on non-CA catalysis by ZnL1S was explored for
2
-6
-6
further evidence of the rate-limiting species.
reaction (6.99 x 10 M and 3.14 x 10 M respectively).
This is evidence that the Zn-complex, without the
9
ACS Paragon Plus Environment

ACS Catalysis
Page 10 of 16
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
B
A
1
00
50
NO3-
SCN-
100
Acetazolamide
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
5
0
0
1
0-10
10-8
10-6
10-4
10-2
100
Inhibitor Concentration
50 M Inhibitor at pH 9.5
0 M Inhibitor at pH 10.8
500 M Inhibitor at pH 10.8
0
5
r
l
e
3
H
3
2
4
4
4
N
S2
N
a
z
C
k
o
C
e
I
d
i
t
O
2
O
O
O
a
lO
O
C
n
i
a
r
D
a
C
O
N
N
P
S
C
l
b
C
S
U
m
i
H
C
a
a
H
a2
a
a
B
h
a
a
N
N
a2
l
N
N
N
n
o
I
N
N
N
o
a
t
N
e
c
A
Figure 3. (A) The effect of different substrate mimics and known CA inhibitors on the % activity of CO
2
hydration by 50 µM ZnL1S
, SCN and acetazolamide on the activity
(1.69 mM) Hydration by 5 µM ZnL1S at pH 10.8. All data is the average of three repeats of three. Error bars are the standard
deviation (N = 9).
-
-
over a range of pH and inhibitor concentrations. (B) The IC50 curves for the inhibitors NO
of CO
3
2
activities.23, 46-48 The effect of increasing concentrations of
on the esterase activity of ZnL1S
Figure 4) (measured by the conversion of 4-nitrophenyl
acetate into 4-nitrophenolate and acetate) provides an
alternative method of finding evidence for the rate-
showed no significant inhibition,
inhibited the esterase activity with
increasing concentrations from 1.69 mM to 16.9 mM
3
. There was an initial activity rise from the
associated increase in pH from bicarbonate. The results
demonstrate that HCO binds to the zinc centre causing an
3
inhibition of the esterase activity of the catalyst. In
contrast, no binding was observed for CO , thus this data
supports the conclusion drawn from the NMR
experiments described above, and provides evidence for
previous predictions made by other investigations for
both ZnL1S and other CA mimics.
this is the first time bicarbonate inhibition of esterase
activity has been shown for ZnL1S, which complements
similar data obtained with a different CA mimic.
1
50
00
-
either CO
2
or HCO
3
(
1
limiting step. CO
whereas HCO
2
-
3
50
CO2
-
HCO
HCO -
3
-
0
0
5
10
15
20
2
CO / HCO3- Concentration (mM)
2
9
-11, 17, 20, 27
-
Interestingly,
Figure 4. Effect of addition of CO
2
and HCO
3
on the
hydrolysis of 4-nitrophenyl acetate by 500 µM ZnL1S.
Reactions were monitored at 348 nm. Data is the average of
three repeats of three. Error bars represent the standard
46
-
deviation (N = 9). The linear fit excludes 0 mM CO
2
/ HCO
3
.
Effect of Metal Centre. The experimental evidence in
this study has shown that the rate-limiting step for both
the hydration and dehydration reactions by ZnL1S is at the
-
The effect of CO and HCO on ZnL1S Esterase
2 3
Activity. It has been well documented that CA synthetic
mimics demonstrate esterase activity by catalysing the
hydrolysis of esters, typically 4-nitrophenyl acetate, into
their respective alcohols and carboxylic acids (or their
respective conjugate bases), although at much lower rates
bicarbonate-bound intermediate step.
To better
understand the reasons behind the mechanism, and thus
to see if the preference for either reaction can be changed,
we investigated the role of the metal centre on the rate of
hydration and dehydration of ZnL1S. To achieve this, the
-
than their CO
2
hydration and HCO
3
dehydration
1
0
ACS Paragon Plus Environment

Page 11 of 16
ACS Catalysis
L1S ligand was complexed to several different transition
and six-coordination states,49 as well as the similar ionic
radii (88 for Zn(II) cf. 88.5 for Co(II)). This trend of
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
metals, Cu, Ni and Co, instead of Zn (Scheme S3). While
previous investigations with other mimics confirmed
50
ZnL1S > CoL1S > CuL1S > NiL1S directly correlates with
the likelihood of the metal centres to form bidentate
coordination to bicarbonate, Ni > Cu > Co > Zn. As
1
6-19
zinc’s superiority for catalytic behaviour,
the L1S
ligand chelation to different metal centres had yet to be
explored. Furthermore, the effect of metal centre on
dehydration activity has not been investigated before for
any CA mimic. The three novel transition metal complexes
of L1S were successfully synthesised (Supporting
Information section 1c), which were subsequently tested
for their ability to catalyse both the hydration reaction and
dehydration reactions.
4
3
described by Han et al., Zn(II) derivatives bind to
bicarbonate with unidentate coordination, and the Co(II)
derivatives show binding that was intermediate between
unidentate and bidentate. In contrast, the Cu(II) and Ni(II)
derivatives show a higher tendency to exhibit bidentate
4
3
coordination. Furthermore, the similar pKa values (8.28
for Co and Ni and 8.29 for Cu and Zn) exclude the pKa of
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
the metal-bound water as a determining factor for the
The experimentally observed rate constants for CO
hydration were measured for each transition metal
complex at pH 9.5, 8.2, and 7.4 using a typical CO
2
-
2 3
differences seen in the interconversion of CO /HCO
between the metal complexes. The tighter binding of
bicarbonate results in the slower release of product,
resulting in slower rates of hydration. This is especially
true at lower pH’s as illustrated by the non-linear pH
dependency. Thus, the importance of overcoming the rate-
liming step by reducing bicarbonate affinity for the metal
centre is emphasised.
2
hydration measurement (Figure S24 and Table 2 for
results). As expected from the mechanism, the catalysts
were most active for CO
2
hydration at pH 9.5. There was,
however, non-linear trend between proton
a
concentration and kcat, demonstrating that all the catalysts
suffered from bicarbonate inhibition regardless of metal
centre. The order of activity for the transition metal L1S
complexes at pH 9.5 was ZnL1S > CoL1S > CuL1S > NiL1S.
-
The initial activity for HCO
3
dehydration was
measured for each transition metal complex at pH 6.3, 7.4
and 8.2 (Figure S25 and Table 2 for results). As expected
Table 2. Summary of kobs for the hydration of CO
2
and
-
from the mechanism, the catalysts were most active for
dehydration of HCO
3
by CA mimics at multiple pH's.
-
HCO
3
dehydration at pH 6.3. The linear trend between
Metal
Centr
e
proton concentration and kcat for all active catalysts,
demonstrated minimal bicarbonate inhibition for these
Zn
Cu
Ni
Co
mimics. It can be observed from the data that ZnL1S is the
-
8
803
±
50.
3
superior catalyst for HCO
3
dehydration at pH 6.3 and pH
4
478
±
3489
±
5766
±
7.4, followed closely by CoL1S. A similar trend in rate
constant was observed at pH 6.3 and pH 7.4, as seen for
the hydration reaction at pH 9.5. This also correlated with
strength of bicarbonate binding. Tighter binding would
pH 9.5
pH 8.2
pH 7.4
pH 6.3
5
5
23.6 347.7 670.6
CO
Hydration
2
574.
5 ±
34.5
9
9
3.6
±
35.1
±
102.4
±
result in a slower catalytic rate, for both CO
2
hydration and
dehydration, and could explain why no activity was
observed for NiL1S.
k
obs
-
HCO
3
2
0.70 15.77 19.98
-
1 -1
(
M s )
1
3
1
28.
44.3
Belo
w
LOD
Belo
w
LOD
This study showed that the zinc metal centre
generated the fastest rates and kobs for both the hydration
and dehydration reactions, followed by cobalt, copper and
then nickel. The zinc centre was therefore key to the
±
4.4
3
±
1
2.55
102.
1 ±
4
4.3
±
Belo
w
LOD
79.6
±
10.08
ability of the catalyst to efficiently hydrate CO
2
and
-
dehydrate HCO
3
due to its preferred weaker monodentate
1
2.9
8
-
HCO
3
9.91
binding to bicarbonate. This demonstrates just how
important bicarbonate affinity is in determining the
mimic’s activities and overcoming the rate-limiting step.
Dehydratio
n
3
.01
±
.49
Belo
w
LOD
Belo
w
8.5 ±
0.57
5.9 ±
0.63
pH 7.4
pH 8.2
k
obs
-1
0
Belo
w
-
1
Future Improvements in Catalyst Activity through
Strategic Ligand Design. As discussed, it is desired that
rates as fast as that of the natural CA enzyme can be
achieved by its synthetic mimics. For this goal to be
reached, the effect of bicarbonate inhibition needs to be
addressed. The zinc centre within CA mimics is highly
electron deficient and therefore easily coordinates to
anions, such as bicarbonate. Although it is not the case for
ZnL1S, this effect is often enhanced further in many
reported CA mimic complexes, due to their cationic
nature, which attract anionic species into their secondary
coordination sphere to maintain charge balance, which
(
M s )
Belo
w
LOD
5
0
.1 ±
.52
LOD
LOD
The data is the average of three repeats. kobs data calculated from the
fit of the data in Figure S24 and S25. LOD= limit of detection.
As predicted, ZnL1S was shown to be the fastest
2
catalyst for CO hydration, in alignment with other CA
mimic studies investigating the effect of alternative metal
16-19
centres.
The higher observed rate constant for CoL1S
over CuL1S and NiL1S, was attributed to the similar
coordination chemistry of Zn(II) and Co(II), and the fast
interconversion of these two metals between four-, five-
1
0
drives the equilibrium to the bicarbonate-bound species.
1
1
ACS Paragon Plus Environment

ACS Catalysis
Page 12 of 16
Our work has shown that the limitations of bicarbonate At pH 9.5 however, the metalloenzyme was faster. This
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
binding, resulting in slower activity, should be addressed
in the development of future CA mimics. There are two
potential ways in which this could be achieved; (i)
reducing the affinity of the anion for the zinc centre to
discourage strong bicarbonate binding and (ii) facilitating
the approach of water into the hydrophobic pocket
around the zinc atom to displace the bicarbonate anion.
Both can be primarily achieved through intelligent ligand
design.
metalloenzyme is, however, significantly larger and
protein-like, and therefore suffers from a lower thermal
stability and more complex synthesis than a small-
molecule mimic like ZnL1S.
CA mimics have also struggled to match the
dehydration activity of the natural enzyme (e.g. hCA II
7
-1 -1 53
dehydration kcat / k
cat for ZnL1S described here has shown
improvement of that previously reported for the mimic, as
m
of 1.2 x 10 M s ). The dehydration
k
a
60%
3
7
To reduce anion affinity for the metal centre, more
electron donating ligands should be used. The ligand L1S,
employed in this study, is a good example of this, as it is an
anionic, water-soluble, salen-like ligand, that has the
ability to donate electron density into the metal centre and
measured via the weight-loss method. ZnL1S has a
s which is over ten times that
of the next fastest mimic Zn Cyclen, with a kcat of 55 M
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
-1
-1
dehydration kcat of 579 M
-1
-1
s
(Table S5). ZnL1S has therefore shown the fastest
dehydration kcat reported for a small molecule CA mimic to
date. Consequently, this work represents a significant
step towards matching the activity demonstrated by CA.
therefore facilitates dissociation of bicarbonate and thus
9-10
improves the rate of CO
2
hydration.
This is part of the
reason why ZnL1S is the fastest small molecule CA mimic
seen in the literature.
Comparing the kcat values for the hydration and
dehydration reactions, ZnL1S shows preference for the
hydration reaction rather than dehydration, as is the case
for both the enzyme and other CA mimics published
previously.
The other reason why ZnL1S is highly active for CO
hydration is that Nakata et al. designed the mimic to
contain hydrophobic pocket which facilitates the
approach of CO
2
a
2
4
2
for hydration. The approach of water to
Simply by reducing the concentration of substrate
used for both the hydration and dehydration reactions, we
have reported increases in catalytic rates due to weaker
bicarbonate binding. This has been without the need for
any modifications to the complex’s structure. It is
expected that this principle could be applied to any CA
mimic used for CO capture, that follows the same
2
mechanism as ZnL1S and therefore has the same rate-
limiting step.
displace the formed bicarbonate is however, consequently
unfavourable, thus rate-limiting the catalyst. To overcome
this, a ligand should include the ability to be able to induce
an extensive hydrogen bonding network in the proximity
of the metal centre, much like the natural enzyme CA,
which achieves this by utilising three hydrogen bonding
2
8-30
amino acids.
For this reason, carbonic anhydrase is
limited not by bicarbonate release but by the shuttle of
2
6
protons, and therefore is significantly faster than any
Comparison of Inhibitor Effects on ZnL1S and CA.
In the previously discussed inhibitor study, it was
observed that the negatively charged species were more
2
7
synthetic mimic reported to date. With the structural
changes recommended, even greater improvements in
catalytic activities could be achieved.
i
inhibitory for the mimic (μM range K ) than for the enzyme
Comparison of ZnL1S Kinetic Properties to CA and
other CA mimics. Historically, CA mimics have struggled
to match the hydration activity of the natural enzyme (e.g.
(mM range K ) (Table 3). It is likely that it is easier for the
anions to enter the active site of the complex and compete
with substrates in the more solvent exposed centre. The
opposite was true for acetazolamide where it was more
i
8
-1
1 51
hCA II hydration kcat / k
m
of 1.5 x 10 M s ). Using a
), we have
lower concentration of substrate (1.69 mM CO
2
inhibitory for the enzyme (nM range K ) over the mimic
i
however, achieved a step in the right direction towards
this goal. The hydration kcat achieved by ZnL1S in this
study is the highest reported by a small-molecule CA
(μM range K ) (Table 3). This is probably caused by lack of
i
acetazolamide binding amino acids for the mimic, as well
as an increased steric hindrance when acetazolamide
-1
-1
24
mimic catalyst to date. With a kcat of 28,093 M s , ZnL1S
has a catalytic rate that is over five times faster than that
of the next best small molecule mimic, Zn Cyclam, with a
binds to ZnL1S. The application of inhibitor profiling
(which is commonly used to characterise enzymes) to
probe the hydration mechanism of CA mimics revealed
new insights into how binders interact differently
between native CA and its mimics.
-1
-1
k
cat of 5040 M
s
(Table S5). By employing a lower
concentration of substrate, ZnL1S was found to be almost
three times faster than had been previously predicted
from the low temperature study reported by Nakata et
Table 3. Comparison of inhibition data for the
complex ZnL1S with data from different isoforms of
human CA (hCA).
24
al. and thus, a better idea of the true potential of the
catalyst for CO
possible that decreasing the CO
2
hydration was seen. Furthermore, it is
concentration further,
2
IC50
K
i
could improve the kcat beyond the value reported here, but
this concentration falls below our current methodology’s
limit of detection. At pH 10.8, the kcat of ZnL1S is similar to
that of the recent synthetic metalloenzyme developed by
hCA
I
hCA
II
hCA
VI
ZnL1S
Acetazolamid 6.99 ±
0.50 µM
3.14
µM
250
nM
12
nM
11
nM
5
2
Zastrow et al. , which at the time demonstrated the
e
5
2
2
highest recorded rate of CO hydration for a CA mimic.
1
2
ACS Paragon Plus Environment

Page 13 of 16
ACS Catalysis
closeness/ difference between the activities of each
4
2.36 ±
27.9
µM
7
mM
35
mM
0.76
mM
-
e
1
2
3
4
5
NO
3
cat
species. The hydration catalytic efficiency (k / km ) of
2
.93 µM
8
-1 -1
hCA II at pH 9.0 is 1.5 x 10 M s , whereas the highest
e
catalytic efficiency (kcat / k
m
) of ZnL1S, at a similar pH, in
8
7.91 ±
59.8
µM
0.2
mM
1.6
mM
0.89
mM
SCN-
5
this study was found to be 5.9 x 10 at pH 9.5. At pH 10.8,
an even greater catalytic efficiency was found (2.0 x 10 M
s ). The maximum efficiency of ZnL1S hydration, i.e. with
4
.05 µM
7
-
ZnL1S CO
mM CO at pH 10.8 and 25 C. Data is the average of three repeats.
Enzyme data is from Bertucci et. al. and Supuran et. al., measured
2
hydration data was measured using 5 µM ZnL1S and 1.69
1 -1
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
o
2
no inhibition, is therefore only 2.5 magnitudes smaller
than that of hCA II at similar pH. This is a significant
o
54-55
using 235 nM hCA I and 1.5 nM hCA II and VI, at pH 7.5 and 20 C.
c
improvement when compared to the kcat calculated
ZnL1S only showed inhibition by bicarbonate and no
inhibition by carbonate. Whereas it has been shown that
hCA II is inhibited by both bicarbonate and carbonate,
although the inhibition is significantly weaker for the
i
= 85 and 73 mM for bicarbonate and carbonate
respectively) than for ZnL1S (nearly 50% reduction in
previously using the small-molecule methodology, which
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
4
-1 -1
gave a value of 2.8 x 10 M
s at pH 9.5 using 1.69 mM
CO2.
The previously discussed metalloenzyme mimic
enzyme (K
developed by Zastrow et al., which had a catalytic
5
6
e
5
-1 -1
52
m
efficiency (kcat / k ) of 1.8 x 10 M s at pH 9.5. ZnL1S
-
3
activity with 50 µM HCO ). It should be noted that there is
is now over three times more efficient at the same pH.
When not inhibited by bicarbonate, ZnL1S therefore has
no observable difference between inhibition of CA by
bicarbonate or carbonate, as bicarbonate release is not the
rate-limiting step for the natural enzyme, highlighting
further the different rate limiting steps between the native
enzyme and its mimics. It is also worth noting that our
data reveal a much less potent effect of chloride on ZnL1S
the potential to be the fastest mimic of CA ever
e
synthesised. Furthermore, the kcat/ k
m
of ZnL1S increased
with decreasing concentration of mimic and therefore the
maximal catalytic efficiency could be even greater than
reported here. A lower concentration of mimic could not
be measured in this study due to the limit of detection of
our current experimental methodology. Nevertheless, it is
entirely feasible that with a lower concentration of mimic,
catalytic efficiencies closer to that of CA could be achieved.
2
4
activity than previously reported,
which could be due
to the ZnL1S complex being generated in situ in the
stopped-flow by Nakata et. al., potentially resulting in
inhibition of the initial formation of the catalyst.
Comparison of Small-Molecule Mimic and Enzyme
Kinetic Analysis Methodologies. Interestingly, there is a
difference observed in the literature between the analysis
methods employed to study rates of catalysis by enzymes
with that used for their mimics. For more information on
the different analysis methodologies see Supporting
Information Section 2g. In brief, the kinetic data for ZnL1S
was analysed via the typical enzyme analysis methodology
generating “enzyme-like” kinetic parameters. The result of
this is described below and in Figure S26 and Table S6. To
avoid confusion, the kinetic parameters generated by the
enzyme and catalysis analysis methods will be labelled
with a superscript e (for enzyme) or c (for catalyst)
respectively.
The same conclusions can be drawn for the
dehydration reaction. The dehydration catalytic efficiency
e
7
-1 -1
53
(
k
cat / k
m
) of hCA II is 1.2 x 10 M
s
at pH 5.5, whereas
m
e
the highest catalytic efficiency (kcat / k
study was found to be 1.7 x 10
) of ZnL1S in this
4
-1 -1
M s at pH 6.3. The
maximum efficiency of ZnL1S dehydration, i.e. with no
inhibition, is therefore only 3 orders of magnitude smaller
than that of hCA II. This is significantly higher when
c
compared to the value of kcat calculated previously, 579
-1 -1
-
M
s
3
at pH 6.3 using 1.5 mM HCO . Our work has
revealed that ZnL1S is the fastest small-molecule water-
soluble CA mimic for dehydration reported to date. As
with the case for the hydration reaction, ZnL1S has the
-
potential to be an even greater catalyst for HCO
3
e
Typically, the kcat/k
cat of synthetic mimics, as they have the same units (M
s ) and both quantify catalytic efficiency. Upon
comparison of the kcat/ k
m
of enzymes are compared to the
dehydration with a lower concentration of mimic and
potentially reach catalytic efficiencies similar to that of CA.
c
-1
k
-1
Conclusions. We report for the first time the rate-
limiting step of the dehydration reaction by a CA mimic to
be at the bicarbonate-bound step, much like the hydration
reaction. This confirms that it is the mimic’s affinity for
bicarbonate that results in the previously observed
hydration inhibition, and not the occurrence of the reverse
dehydration reaction. Each reaction is however inhibited
by bicarbonate by differing degrees, with the hydration
reaction demonstrating significantly greater bicarbonate
inhibition than the dehydration reaction. With this
knowledge, the catalyst’s preference for either reaction
could be controlled via substrate concentration.
e
c
m
of ZnL1S to the kcat of ZnL1S, a
difference in the values can be seen. For both reactions
e
and across all conditions studied, the calculated kcat/ k
m
values were greater than their respective mimic-like kcatc,
e
often by orders of magnitude. As the kcat/ k
m
considers the
effect of substrate concentration and the subsequent
inhibition, it can be said that these values demonstrate a
better representation of the mimic’s ability to catalyse
each reaction. These values show the mimic’s maximal
catalytic efficiency when substrate concentration is
negligible (at a particular concentration of catalyst). This
is rather than the catalytic efficiency per molar of catalyst
at a specific substrate concentration, as often reported.
e
Consequently, a significant increase in the activity of
the previously published CA mimic ZnL1S, which is the
fastest small-molecule mimic published to date, was
achieved without the need for modification of the catalyst
Consequently, a comparison of the kcat/ k
to that of hCA II allows for a better indication of the
m
of ZnL1S
1
3
ACS Paragon Plus Environment

ACS Catalysis
Page 14 of 16
structure. Rather, by simply lowering the concentration of
the substrate used for either the hydration of CO or
, we were able to achieve an increase
in kcat by a factor of 3 and 1.6 respectively. By lowering the
concentration of substrate, especially for the hydration
reaction, the energetically unfavourable rate-limiting
bicarbonate-intermediate step was overcome due to the
subsequently lower concentration of bicarbonate. This
information will be critical for optimising the use of such
mimics in industrial carbon capture.
important in medicine, such as is the case with the CA
studied here.
Future work will look at how these findings,
especially the effect of substrate concentration to
manipulate preference for either reaction at physiological
pH and the potential to improve catalytic activity further,
can be applied to uses of CA and its mimics for wider-
world applications.
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
2
-
dehydration of HCO
3
ASSOCIATED CONTENT
Conversely, by increasing the substrate concentration
used, similar kobs could be seen for both reactions at the
same pH (pH 7.4) for the first time. This opens the
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
All ligand and complex spectra, as well as data from the
stopped-flow and all other assays can be found in the
attached Supporting Information. This material is available
free of charge via the Internet at http://pubs.acs.org
potential to manipulate mimics to enable them to show
-
preference to catalyse HCO
3
dehydration over CO
2
hydration at physiological pH values. As demonstrated by
ACKNOWLEDGEMENT
e
m
the maximal catalytic efficiencies (kcat / k ) of ZnL1S at
Thank you to Dan Scott and Andy Ashley at Imperial
College London for their assistance with the Topeler pump
negligible substrate concentration, the mimic could reach
activities of only 2.5-3 orders of magnitude slower than
that of the natural CA enzyme itself. This finding could
likely be applied to any CA, and thus these mimics have the
potential to demonstrate significantly higher catalytic
activities in the future. With future careful ligand design,
to facilitate bicarbonate release at a faster rate, even
greater activities could also be reached.
1
3
2
for the C CO NMR studies. Thank you to Lisa Haigh for
operation of the Imperial College London Chemistry Mass
Spectrometry service. Thank you to Stephen Boyer at the
Elemental Analysis Service at London Metropolitan
University. This work was supported by an EPSRC Centre
for Doctoral Training Studentship from the Institute of
Chemical Biology (Imperial College London) awarded to
Jonathan Rains.
Additionally, within this study, the observed rate
c
-
constants (kobs ) for CO
catalysed by ZnL1S, CuL1S, NiL1S and CoL1S, (except
ZnL1S for CO hydration), were reported for the first time
2 3
hydration and HCO dehydration
REFERENCES
1. Bouzalakos, S.; Mercedes, M. Overview of Carbon Dioxide
Capture and Storage Technology. In Developments and Innovation
2
in Carbon Dioxide (CO ) Capture and Storage Technology, 1 ed.;
Maroto-Valer, M. M., Ed. Woodhead Publishing Ltd.: University of
Nottingham, UK, 2010; Vol. 2.
2. Ma, J.; Sun, N.; Zhang, X.; Zhao, N.; Xiao, F.; Wei, W.; Sun, Y., A
2
Short Review of Catalysis for CO Conversion; Catalysis Today
2009, 148, 221-231.
2
via stopped-flow. A comparison of the kinetic activity of
this suite of CA mimics with differing metal centres
revealed that ZnL1S was a superior catalyst for both the
-
CO
2
hydration and HCO
3
dehydration reactions as
expected. The kinetic profile of ZnL1S for both hydration
and dehydration activities was studied across a large pH,
o
complex and substrate concentration range at 25 C.
3
. Ranjan, M.; Herzog, H. J., Feasibility of Air Capture; Energy
Procedia 2011, 4, 2869-2876.
. Lindskog, S., Structure and Mechanism of Carbonic
Anhydrase; Pharmacol. Ther. 1997, 74, 1-20.
. Domsic, J. F.; Avvaru, B. S.; Kim, C. U.; Gruner, S. M.; Agbandje-
Consequently, substrate inhibition kinetics, caused by
increased bicarbonate, were seen for the first time for a CA
mimic. A comparison between analytical methodologies of
catalysts and enzymes was carried out, and the same
trends could be seen, and subsequent conclusions drawn.
4
5
McKenna, M.; Silverman, D. N.; McKenna, R., Entrapment of
Carbon Dioxide in the Active Site of Carbonic Anhydrase Ii; J. Biol.
Chem. 2008, 283, 30766-30771.
Finally, this work contains the largest inhibitor study
of any CA mimic reported in the literature. The first IC50
and K values for the hydration of CO by CA mimic
i 2
6
. Supuran, C. T., Structure and Function of Carbonic
Anhydrases; Biochem. J. 2016, 473, 2023-2032.
inhibitors are also reported, for both anionic inhibitors
and the commercial CA inhibitor acetazolamide. The use
of substrate and product mimics to probe the mechanism
of enzyme mimics, as shown here for CA and its small
molecule mimic ZnL1S, allows for a subsequent deeper
understanding revealing differences between the
substrate binding in the enzyme and its mimics, as well as
the difference in mechanisms that this alludes to. In the
case of the natural CA enzyme and the ZnL1S mimic, small
anions were found to be more inhibiting for the mimic,
suggesting a greater importance for anion binding affinity
in the mechanism of CA mimics than the enzyme. This
approach could be used to investigate any other chemical
mimic of a biological enzyme, especially those that are
7. Nath, I.; Chakraborty, J.; Verpoort, F., Metal Organic
Frameworks Mimicking Natural Enzymes: A Structural and
Functional Analogy; Chem. Soc. Rev. 2016, 45, 4127-4170.
8. Saeed, M.; Deng, L., CO
Promoted by Mimic Enzyme; J. Membr. Sci. 2015, 494, 196-204.
. Kelsey, R. A.; Miller, D. A.; Parkin, S. R.; Liu, K.; Remias, J. E.;
Yang, Y.; Lightstone, F. C.; Liu, K.; Lippert, C. A.; Odom, S. A.,
Carbonic Anhydrase Mimics for Enhanced CO Absorption in an
Amine-Based Capture Solvent; Dalton Trans. 2015, 45, 324-333.
0. Lippert, C. A.; Liu, K.; Sarma, M.; Parkin, S. R.; Remias, J. E.;
2
Facilitated Transport Membrane
9
2
1
Brandewie, C. M.; Odom, S. A.; Liu, K., Improving Carbon Capture
from Power Plant Emissions with Zinc- and Cobalt-Based
Catalysts; Catal. Sci. Technol. 2014, 4, 3620-3625.
11. Floyd, W. C., 3rd; Baker, S. E.; Valdez, C. A.; Stolaroff, J. K.;
Bearinger, J. P.; Satcher, J. H., Jr.; Aines, R. D., Evaluation of a
Carbonic Anhydrase Mimic for Industrial Carbon Capture;
Environ. Sci. Technol. 2013, 47, 10049-10055.
1
4
ACS Paragon Plus Environment

Page 15 of 16
ACS Catalysis
1
2. Wong, S. E.; Lau, E. Y.; Kulik, H. J.; Satcher, J. H.; Valdez, C.;
Worsely, M.; Lightstone, F. C.; Aines, R., Designing Small-Molecule
Catalysts for CO Capture; Energy Procedia 2011, 4, 817-823.
3. Ibrahim, M. M.; Shaban, S. Y.; Ichikawa, K., A Promising
Structural Zinc Enzyme Model for CO Fixation and Calcification;
Tetrahedron Lett. 2008, 49, 7303-7306.
Catalysis; Acta Crystallogr., Sect. C: Struct. Chem. 2014, 70, 123-
131.
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
2
28. Bandeira, N. A.; Garai, S.; Muller, A.; Bo, C., The Mechanism
of Co2 Hydration: A Porous Metal Oxide Nanocapsule Catalyst
Can Mimic the Biological Carbonic Anhydrase Role; Chem.
Commun. (Cambridge, U. K.) 2015, 51, 15596-15599.
29. Piazzetta, P.; Marino, T.; Russo, N.; Salahub, D. R., Direct
Hydrogenation of Carbon Dioxide by an Artificial Reductase
Obtained by Substituting Rhodium for Zinc in the Carbonic
Anhydrase Catalytic Center. A Mechanistic Study; ACS Catal.
2015, 5, 5397-5409.
1
2
14. Ibrahim, M. M.; Amin, M. A.; Ichikawa, K., Synthesis and
Characterization of Benzimidazole-Based Zinc Complexes as
Structural Carbonic Anhydrase Models and Their Applications
Towards Co2 Hydration; J. Mol. Struct. 2011, 985, 191-201.
1
5. Jin, C.; Zhang, S.; Zhang, Z.; Chen, Y., Mimic Carbonic
Anhydrase Using Metal-Organic Frameworks for CO
and Conversion; Inorg. Chem. 2018, 57, 2169-2174.
16. Bergquist, C.; Fillebeen, T.; Morlok, M. M.; Parkin, G.,
Protonation and Reactivity Towards Carbon Dioxide of the
Mononuclear Tetrahedral Zinc and Cobalt Hydroxide Complexes,
2
Capture
30. Yu, F.; Cangelosi, V. M.; Zastrow, M. L.; Tegoni, M.; Plegaria,
J. S.; Tebo, A. G.; Mocny, C. S.; Ruckthong, L.; Qayyum, H.; Pecoraro,
V. L., Protein Design: Toward Functional Metalloenzymes; Chem.
Rev. 2014, 114, 3495-3578.
31. Pocker, Y.; Bjorkquist, D. W., Stopped-Flow Studies of
Carbon Dioxide Hydration and Bicarbonate Dehydration in
Water and Water-D2. Acid-Base and Metal Ion Catalysis; J. Am.
Chem. Soc. 1977, 99, 6537-6543.
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
[Tp(Bu)T(,Me)]Znoh and [Tp(Bu)T(,Me)]Cooh: Comparison of
the Reactivity of the Metal Hydroxide Function in Synthetic
Analogues of Carbonic Anhydrase; J. Am. Chem. Soc. 2003, 125,
6189-6199.
32. Brown, R. S.; Curtis, N. J.; Huguet, J., Tris(4,5-
+
1
7. Lau, E. Y.; Wong, S. E.; Baker, S. E.; Bearinger, J. P.; Koziol,
Diisopropylimidazol-2-Yl)Phosphine-Zinc(2 ) - a Catalytically
L.; Valdez, C. A.; Satcher, J. H., Jr.; Aines, R. D.; Lightstone, F. C.,
Comparison and Analysis of Zinc and Cobalt-Based Systems as
Catalytic Entities for the Hydration of Carbon Dioxide; PloS one
2013, 8, e66187.
Active Model for Carbonic-Anhydrase; J. Am. Chem. Soc. 1981,
103, 6953-6959.
33. Zhang, X.; van Eldik, R.; Koike, T.; Kimura, E., Kinetics and
Mechanism of the Hydration of Carbon Dioxide and Dehydration
of Bicarbonate Catalyzed by a Zinc (II) Complex of 1,5,9-
Triazacyclododecane as a Model for Carbonic Anhydrase; Inorg.
Chem. 1993, 32, 5749-5755.
1
8. Davy, R.; Shanks, R. A.; Periasamy, S.; Gustafason, M. P.;
Zambergs, B. M., Development of High Stability Catalysts to
Facilitate CO Capture into Water–an Alternative to
Monoethanolamine and Amine Solvents; Energy Procedia 2011,
2
34. Zhang, X.; van Eldik, R., A Functional Model for Carbonic
4
, 1691-1698.
Anhydrase: Thermodynamic and Kinetic Study of
a
19. Keum, C.; Kim, M.-C.; Lee, S.-Y., Effects of Transition Metal
Tetraazacyclododecane Complex of Zinc(II); Inorg. Chem. 1995,
34, 5606-5614.
Ions on the Catalytic Activity of Carbonic Anhydrase Mimics; J.
Mol. Catal. A: Chem. 2015, 408, 69-74.
35. Sun, Y.-J.; Zhang, L. Z.; Cheng, P.; Lin, H.-K.; Yan, S.-P.; Liao,
D.-Z.; Jiang, Z.-H.; Shen, P.-W., Dehydration Kinetic Studies of
Hco3− Catalyzed by Three Half-Sandwich Nickel(Ii) Complexes in
2
0. Koziol, L.; Valdez, C. A.; Baker, S. E.; Lau, E. Y.; Floyd, W. C.,
3
rd; Wong, S. E.; Satcher, J. H., Jr.; Lightstone, F. C.; Aines, R. D.,
-
-
-
Toward a Small Molecule, Biomimetic Carbonic Anhydrase
Model: Theoretical and Experimental Investigations of a Panel of
Zinc(Ii) Aza-Macrocyclic Catalysts; Inorg. Chem. 2012, 51, 6803-
2 3
the Presence of Inhibitors NO , N and NCS ; Inorg. Chem.
Commun. 2004, 7, 165-168.
36. Sun, Y.-J.; Zhang, L. Z.; Cheng, P.; Lin, H.-K.; Yan, S.-P.; Liao,
D.-Z.; Jiang, Z.-H.; Shen, P.-W., Experimental and Theoretical
Studies of the Dehydration Kinetics of Two Inhibitor-Containing
Half-Sandwich Cobalt(Ii) Complexes; J. Mol. Catal. A: Chem. 2004,
208, 83-90.
37. Davy, R., Development of Catalysts for Fast, Energy
2
Efficient Post Combustion Capture of CO into Water; an
6
812.
1. Satcher, J. H.; Baker, S. E.; Kulik, H. J.; Valdez, C. A.; Krueger,
2
R. L.; Lightstone, F. C.; Aines, R. D., Modeling, Synthesis and
Characterization of Zinc Containing Carbonic Anhydrase Active
Site Mimics; Energy Procedia 2011, 4, 2090-2095.
2
2. Sun, Y.-J.; Zhang, L. Z.; Sun, W.; Cheng, P.; Lin, H.-K.; Yan, S.-
P.; Liao, D.-Z.; Jiang, Z.-H.; Shen, P.-W., Kinetics and Mechanism of
the Bicarbonate Dehydration of the Half-Sandwich Zinc(II)
Alternative to Monoethanolamine (Mea) Solvents; Energy
Procedia 2009, 1, 885-892.
Complexes
[Tpph]Znx
([Tpph]
=
Hydrotris(3-
38. Rodionov, V. O.; Presolski, S. I.; Gardinier, S.; Lim, Y. H.; Finn,
M. G., Benzimidazole and Related Ligands for Cu-Catalyzed Azide-
Alkyne Cycloaddition; J. Am. Chem. Soc. 2007, 129, 12696-12704.
39. Gibbons, B. H.; Edsall, J. T., Rate of Hydration of Carbon
Dioxide and Dehydration of Carbonic Acid at 25°C; J. Biol. Chem.
1963, 238, 3502-3507.
40. Tu, C. K.; Thomas, H. G.; Wynns, G. C.; Silverman, D. N.,
Hydrolysis of 4-Nitrophenyl Acetate Catalyzed by Carbonic
Anhydrase-III from Bovine Skeletal-Muscle; J. Biol. Chem. 1986,
261, 100-103.
41. Bräuer, M.; Pérez-Lustres, J. L.; Weston, J.; Anders, E.,
Quantitative Reactivity Model for the Hydration of Carbon
Dioxide by Biomimetic Zinc Complexes⊥; Inorg. Chem. 2002, 41,
1454-1463.
42. Echizen, T.; Ibrahim, M. M.; Nakata, K.; Izumi, M.; Ichikawa,
K.; Shiro, M., Nucleophilic Reaction by Carbonic Anhydrase Model
Phenylpyrazolyl)Borate; X− = Oh−, N3−, Ncs−); J. Mol. Catal. A:
Chem. 2003, 198, 99-106.
2
3. Kimura, E.; Shiota, T.; Koike, T.; Shiro, M.; Kodama, M., A
3
Zinc(II) Complex of 1,5,9-Triazacyclododecane ([12]Anen ) as a
Model for Carbonic Anhydrase; J. Am. Chem. Soc. 1990, 112, 5805-
5
811.
24. Nakata, K.; Shimomura, N.; Shiina, N.; Izumi, M.; Ichikawa,
K.; Shiro, M., Kinetic Study of Catalytic CO Hydration by Water-
Soluble Model Compound of Carbonic Anhydrase and Anion
Inhibition Effect on CO Hydration; J. Inorg. Biochem. 2002, 89,
55-266.
5. Rowlett, R. S., Structure and Catalytic Mechanism of the
2
2
2
2
Beta-Carbonic Anhydrases; Biochim. Biophys. Acta 2010, 1804,
362-373.
2
6. Silverman, D. N.; Vincent, S. H., Proton Transfer in the
Catalytic Mechanism of Carbonic Anhydras; Crit. Rev. Biochem.
Mol. Biol. 1983, 14, 207-255.
Zinc Compound: Characterization of Intermediates for CO
2
Hydration and Phosphoester Hydrolysis; J. Inorg. Biochem. 2004,
98, 1347-1360.
43. Han, R., Structural and Spectroscopic Studies on Four-,
Five-, and Six-Coordinate Complexes of Zinc, Copper, Nickel, and
2
7. Kulik, H. J.; Wong, S. E.; Baker, S. E.; Valdez, C. A.; Satcher, J.
H., Jr.; Aines, R. D.; Lightstone, F. C., Developing an Approach for
First-Principles Catalyst Design: Application to Carbon-Capture
1
5
ACS Paragon Plus Environment

ACS Catalysis
Page 16 of 16
Cobalt: Structural Models for the Bicarbonate Intermediate of the
Carbonic Anhydrase Catalytic Cycle; J. Inorg. Biochem. 1993, 49,
05-121.
4. Alzuet, G.; Casella, L.; Perotti, A.; Borrás, J., Acetazolamide
Binding to Zinc(II), Cobalt(II) and Copper(II) Model Complexes of
Carbonic Anhydrase; J. Chem. Soc., Dalton Trans. 1994, 0, 2347-
2351.
45. Chufán, E. E.; García-granda, S.; Díaz, M. R.; Borrás, J.;
Pedregosa, J. C., Several Coordination Modes of 5-Amino-1,3,4-
Thiadiazole-2-Sulfonamide (Hats) with Cu(II), Ni(II) and Zn(II):
Mimetic Ternary Complexes of Carbonic Anhydrase-Inhibitor; J.
Coord. Chem. 2001, 54, 303-312.
46. Koike, T.; Kimura, E.; Nakamura, I.; Hashimoto, Y.; Shiro, M.,
The First Anionic Sulfonamide-Binding Zinc(II) Complexes with
a Macrocyclic Triamine: Chemical Verification of the Sulfonamide
Inhibition of Carbonic Anhydrase; J. Am. Chem. Soc. 1992, 114,
51. Di Fiore, A.; Monti, S. M.; Hilvo, M.; Parkkila, S.; Romano, V.;
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
Scaloni, A.; Pedone, C.; Scozzafava, A.; Supuran, C. T.; De Simone,
G., Crystal Structure of Human Carbonic Anhydrase Xiii and Its
Complex with the Inhibitor Acetazolamide; Proteins: Struct.,
Funct., Bioinf. 2009, 74, 164-175.
52. Zastrow, M. L.; Peacock, A. F.; Stuckey, J. A.; Pecoraro, V. L.,
Hydrolytic Catalysis and Structural Stabilization in a Designed
Metalloprotein; Nat. Chem. 2011, 4, 118-123.
53. Baird, T. T., Jr.; Waheed, A.; Okuyama, T.; Sly, W. S.; Fierke,
C. A., Catalysis and Inhibition of Human Carbonic Anhydrase Iv;
Biochemistry 1997, 36, 2669-2678.
54. Bertucci, A.; Innocenti, A.; Zoccola, D.; Scozzafava, A.;
Allemand, D.; Tambutte, S.; Supuran, C. T., Carbonic Anhydrase
Inhibitors: Inhibition Studies of a Coral Secretory Isoform with
Inorganic Anions; Bioorg. Med. Chem. Lett. 2009, 19, 650-653.
55. Supuran, C. T.; Popescu, A.; Ilisiu, M.; Costandache, A.;
Banciu, M. D., Carbonic Anhydrase Inhibitors. Part 36. Inhibition
of Isozymes I and Ii with Schiff Bases Derived from Chalkones and
Aromatic/Heterocyclic Sulfonamides; Eur. J. Med. Chem. 1996,
31, 439-447.
56. Rusconi, S.; Innocenti, A.; Vullo, D.; Mastrolorenzo, A.;
Scozzafava, A.; Supuran, C. T., Carbonic Anhydrase Inhibitors.
Interaction of Isozymes I, II, IV, V, and IX with Phosphates,
Carbamoyl Phosphate, and the Phosphonate Antiviral Drug
Foscarnet; Bioorg. Med. Chem. Lett. 2004, 14, 5763-5767.
1
4
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
7
338-7345.
7. Lin, Y.-W., Rational Design of Metalloenzymes: From Single
4
to Multiple Active Sites; Coord. Chem. Rev. 2017, 336, 1-27.
48. Lopez, M.; Vu, H.; Wang, C. K.; Wolf, M. G.; Groenhof, G.;
Innocenti, A.; Supuran, C. T.; Poulsen, S. A., Promiscuity of
Carbonic Anhydrase Ii. Unexpected Ester Hydrolysis of
Carbohydrate-Based Sulfamate Inhibitors; J. Am. Chem. Soc.
2
011, 133, 18452-18462.
49. Williams, R. J. P., Bio-Inorganic Chemistry: Its Conceptual
Evolution; Coord. Chem. Rev. 1990, 100, 573-610.
0. Woolley, P., Models for Metal Ion Function in Carbonic
5
Anhydrase; Nature (London, U. K.) 1975, 258, 677-682.
Graphical Abstract
Rate-limiting step for
Carbonic Anhydrase Enzyme
H
H
H
O
O
Zn2+
R
Zn2+
H+
R
NH
NH
N
N
N
N
R
R
R
R
N
N
R
N
1
N
R
R
H
R
H
NH
R
NH
R
HCO3-
H2O
Rate-limiting step
for Mimics
4
CO2
[Substrate]
kcat
2
O
C
O
C
O
H
O
H
O
O
R
Zn2+
3
R
Zn2+
NH
NH
N
R
N
N
N
R
R
R
N
R
N
N
N
R
R
H
H
R
NH
NH
R
R
1
6
ACS Paragon Plus Environment