A Structural Mimic of Carbonic Anhydrase in a Metal-Organic Framework

Article abstract of DOI:10.1016/j.chempr.2018.09.011

Metal-organic frameworks (MOFs) have exciting potential for biomimetic studies of enzymes, yet construction of high-fidelity models at the metal nodes is challenging. Nonetheless, biomimetic MOFs have significant advantages, such as increased stability and ease of separation, over their enzymatic and homogeneous counterparts, making them particularly attractive for industrial applications. Here, we demonstrate biomimetic behavior of Zn hydroxide moieties inside a MOF with structural and reactivity characteristics of carbonic anhydrase. Similar to the biological system, the MOF binds CO₂ by an insertion mechanism into the Zn–OH bond, leading to significant adsorption of CO₂ (3.41 mmol/g). In reactivity mimicking that of the enzyme, the material also catalyzes the oxygen isotope exchange between water and carbon dioxide. Overall, these results provide the strongest evidence yet of metal nodes in MOFs bearing high structural fidelity to enzymatic active sites. The nodes of metal-organic frameworks are attractive sites for mimicking metalloenzymes, primarily through their site isolation and similar ligand fields. In this article, the metal-organic framework MFU-4l is shown to mimic the active site of carbonic anhydrase with high structural fidelity and reactivity. The material adsorbs high quantities of carbon dioxide at low pressures and mimics critical features of carbonic anhydrase, such as isotopic exchange of oxygen atoms from water and carbon dioxide. Mimicking metalloenzymes at the node of a metal-organic framework (MOF) has the potential to impart enzyme-like catalytic activity within a heterogeneous material. Carbonic anhydrase, one of nature's fastest enzymes, catalyzes the hydrolysis of carbon dioxide into bicarbonate and protons. Notably, carbonic anhydrase mimics have been proposed as potential catalysts for carbon capture and sequestration from the environment. Here, we demonstrate that the metal node of MFU-4l, a MOF featuring a metal node with a N₃ZnX coordination environment, can be functionalized to give a mimic of carbonic anhydrase. This work describes a well-defined example of a metal node within a MOF with high structural fidelity to an enzyme active site. It has potential applicability to applications such as CO₂ capture and sequestration and also important gas separations involving CO₂.

Full text of DOI:10.1016/j.chempr.2018.09.011

Article
A Structural Mimic of Carbonic Anhydrase in a
Metal-Organic Framework
Ashley M. Wright, Zhenwei Wu,
Guanghui Zhang, ...,
Christopher H. Hendon, Jeffrey
ꢀ
T. Miller, Mircea Dinca
mdinca@mit.edu
HIGHLIGHTS
Hydroxylation of the MFU-4l metal
node creates a functioning
carbonic anhydrase mimic
High uptake volumes of adsorbed
carbon dioxide occur at low
pressures
Isotopic exchange of 18O from
H2O into CO2 is catalyzed by
MFU-4l-(OH)
The nodes of metal-organic frameworks are attractive sites for mimicking
metalloenzymes, primarily through their site isolation and similar ligand fields. In
this article, the metal-organic framework MFU-4l is shown to mimic the active site
of carbonic anhydrase with high structural fidelity and reactivity. The material
adsorbs high quantities of carbon dioxide at low pressures and mimics critical
features of carbonic anhydrase, such as isotopic exchange of oxygen atoms from
water and carbon dioxide.
Wright et al., Chem 4, 1–8
December 13, 2018 ª 2018 Elsevier Inc.
https://doi.org/10.1016/j.chempr.2018.09.011

Please cite this article in press as: Wright et al., A Structural Mimic of Carbonic Anhydrase in a Metal-Organic Framework, Chem (2018), https://
doi.org/10.1016/j.chempr.2018.09.011
Article
A Structural Mimic of Carbonic Anhydrase
in a Metal-Organic Framework
Ashley M. Wright,1 Zhenwei Wu,2 Guanghui Zhang,2 Jenna L. Mancuso,3 Robert J. Comito,1
1
3
2
ꢀ1,4,
*
Robert W. Day, Christopher H. Hendon, Jeffrey T. Miller, and Mircea Dinca
SUMMARY
The Bigger Picture
Mimicking metalloenzymes at the
node of a metal-organic
Metal-organic frameworks (MOFs) have exciting potential for biomimetic
studies of enzymes, yet construction of high-fidelity models at the metal nodes
is challenging. Nonetheless, biomimetic MOFs have significant advantages,
such as increased stability and ease of separation, over their enzymatic and ho-
mogeneous counterparts, making them particularly attractive for industrial ap-
plications. Here, we demonstrate biomimetic behavior of Zn hydroxide moieties
inside a MOF with structural and reactivity characteristics of carbonic anhy-
drase. Similar to the biological system, the MOF binds CO2 by an insertion
mechanism into the Zn–OH bond, leading to significant adsorption of CO2
(3.41 mmol/g). In reactivity mimicking that of the enzyme, the material also cat-
alyzes the oxygen isotope exchange between water and carbon dioxide. Over-
all, these results provide the strongest evidence yet of metal nodes in MOFs
bearing high structural fidelity to enzymatic active sites.
framework (MOF) has the
potential to impart enzyme-like
catalytic activity within a
heterogeneous material.
Carbonic anhydrase, one of
nature’s fastest enzymes,
catalyzes the hydrolysis of carbon
dioxide into bicarbonate and
protons. Notably, carbonic
anhydrase mimics have been
proposed as potential catalysts
for carbon capture and
sequestration from the
INTRODUCTION
environment. Here, we
Metal-organicframeworks(MOFs)arepromisingplatformsforbiomimeticstudiesofmet-
alloenzymes.1–4 The connecting ends of organic linkers in MOFs, such as carboxylates
and triazoles, are comparable to the ligands that make up the active sites in metalloen-
zymes. Consequently, the metal nodes are attractive targets for biomimetic chemistry.
The advantages of performing biomimetic chemistry in MOFs over traditional homoge-
neous systems include (1) site isolation, preventing unwanted bimetallic reactivity and in
turnallowingfor stericallyunhinderedmetal sites;5 (2) tunable poreenvironments, afford-
ing hydrophilic or hydrophobic channels, which can mimic the channels found in the
enzymes;3 and (3) higher stability than with enzymes,6–9 potentially leading to broader
use of biomimetic chemistry in industrially relevant applications.
demonstrate that the metal node
of MFU-4l, a MOF featuring a
metal node with a N3ZnX
coordination environment, can be
functionalized to give a mimic of
carbonic anhydrase. This work
describes a well-defined example
of a metal node within a MOF with
high structural fidelity to an
enzyme active site. It has potential
applicability to applications such
as CO2 capture and sequestration
and also important gas
Carbonic anhydrase (CA) is a ubiquitous zinc metalloenzyme that catalyzes the hydration
of carbon dioxide. The active site ofCA features a divalent zinc in a N3ZnOH coordination
environment, where the zinc exhibits a tetrahedral geometry featuring three histidine
groups and a hydroxide (or water) (Figure 1).10 Much has been learned about CA through
modeling the active site with synthetic analogs,11 such as Tp(tBu,Me)ZnOH (Tp = tris(pyra-
zolyl)hydroborate), which reacts reversibly with CO2 and catalyzes oxygen atom ex-
change between CO2 and H2O.12–14 Accessing similar Tp-like environments within
MOFs would afford site-isolated examples of CA. One material whose secondary build-
ing units (SBUs) are almost exact structural mimics of the active site in CA is MFU-4l
(Zn5Cl4(BTDD)3 1, where BTDD2ꢀ = bis(1,2,3-triazolo[4,5-b],[40,50-i])dibenzo[1,4]dioxin)
(Figure 1).15–20 The SBUs, or metal nodes, in this material feature four peripheral zinc
chloride sites that are coordinated by three triazoles in C3 symmetry analogous to the
Tp ligand and, accordingly, CA (Figure 1).
separations involving CO2.
Chem 4, 1–8, December 13, 2018 ª 2018 Elsevier Inc.
1

Please cite this article in press as: Wright et al., A Structural Mimic of Carbonic Anhydrase in a Metal-Organic Framework, Chem (2018), https://
doi.org/10.1016/j.chempr.2018.09.011
Figure 1. Structure of MFU-4l-(OH) (Top) and Comparison of the Active Site in Carbonic
Anhydrase with the Peripheral Zinc in MFU-4l (Bottom)
The structure of MFU-4l-(OH) (2) was modeled from the XYZ coordinates of an extended lattice
DFT model.
In this article, we demonstrate that the metal node of MFU-4l (1) can be modified by
anion exchange to create a biomimetic model of CA. The addition of organic
hydroxide affords MFU-4l-(OH) (2), where original terminal N3Zn–Cl centers are
quantitatively transformed to terminal N3Zn–OH units. Characterization by X-ray
absorption spectroscopy confirms the installation of the hydroxide unit, and subse-
quent reactivity studies show that 2 reacts with CO2 and catalyzes oxygen atom
exchange between H2O and CO2 as well as the hydrolysis of 4-nitrophenyl acetate
(4-NPA), consistent with the activity of CA.
1Department of Chemistry, Massachusetts
Institute of Technology, 77 Massachusetts
Avenue, Cambridge, MA 02139, USA
2Davidson School of Chemical Engineering,
Purdue University, 480 Stadium Mall Drive,
West Lafayette, IN 47907, USA
3Materials Science Institute, Department of
Chemistry and Biochemistry, University of
Oregon, Eugene, OR 97403, USA
RESULTS AND DISCUSSION
4Lead Contact
Although the terminal chlorides in MFU-4l are known to exchange with a variety
of other anions, hydroxide exchange using KOH has been reported to cause
decomposition.21 However, we found that anion metathesis with hydroxide was
*Correspondence: mdinca@mit.edu
https://doi.org/10.1016/j.chempr.2018.09.011
2
Chem 4, 1–8, December 13, 2018

Please cite this article in press as: Wright et al., A Structural Mimic of Carbonic Anhydrase in a Metal-Organic Framework, Chem (2018), https://
doi.org/10.1016/j.chempr.2018.09.011
Figure 2. X-Ray Absorption Plots of MFU-4l (1) and MFU-4l-(OH) (2)
(A) XAS plot of the zinc K-edge XANES spectra for MFU-4l (1, red trace) and MFU-4l-(OH) (2, blue
trace).
(B) Magnitude of Fourier transform of k2-weighted EXAFS spectra of MFU-4l (1, red trace) versus
MFU-4l-(OH) (2, blue trace).
possible through the addition of 10 equiv (2.5 equiv per peripheral zinc) of tetrabuty-
lammonium hydroxide, [TBA][OH], to a suspension of 1 in methanol, which yielded
MFU-4l-(OH) (Zn5(OH)4(BTDD)3, 2) as a beige microcrystalline solid. Analysis of 2 by
powder X-ray diffraction confirmed the retention of crystallinity (Figure S1). Fitting a
77 K N2 adsorption isotherm of 2 to the Brunauer-Emmett-Teller equation gave a sur-
face area of 2,739 m2/g, which, although lower than that of 1 (3,525 m2/g), confirmed
the retention of porosity after treatment with base (Figure S2). Thermogravimetric anal-
ysis of 2 revealed that it is less stable than its parent 1; its extrapolated onset decom-
position temperature occurs at 250ꢁC, approximately 100ꢁC lower than for 1 (Fig-
ure S6). Energy-dispersive X-ray spectroscopy and X-ray photoelectron spectroscopy
of 2 confirmed the quantitative replacement of chloride with hydroxide and revealed
no chlorine-specific peaks (Figures S4 and S5). Finally, diffuse reflectance infrared Four-
ier transform spectroscopy (DRIFTS) of 2 revealed a hydroxide O–H stretching band at
3,699 cmꢀ1, a signal that was not observed in the DRIFTS data for 1 (Figure S9). Indeed,
the frequency of this band is comparable to that of the OH stretch found in the homol-
ogous molecular compound TptBu,MeZn(OH) (3,676 cmꢀ1).11,12
To further characterize the incorporation of a Zn–OH site in 2, we analyzed the ma-
terial by X-ray absorption spectroscopy (XAS). Whereas both 1 and 2 had the same
Zn K-edge energy (9,663.6 eV), indicating a common oxidation state on zinc in both
materials (Figure 2A), a significant difference in the zinc coordination environment
became apparent when we compared the white line and near-edge regions for
the two materials: 2 exhibited a lower and broader white line than 1 and a near-
edge feature (ꢂ9,668 eV) higher than that of 1 (Figure 2A). Moreover, the extended
X-ray absorption fine structure (EXAFS) spectra of 1 and 2 indicated a difference in
the primary coordination sphere of zinc. In 2, the first shell peak shifted to a lower
radial distance and was less intense than the corresponding peak in 1 (Figure 2B),
consistent with a shorter Zn–X bond and scattering by a lighter element, such as ox-
ygen. Quantitative fitting of the EXAFS region for 2 was challenging because there
were two Zn2+ sites (N3ZnX and N6Zn) with distinct coordination environments, and
the small difference between the Zn–N bond length of the two sites made it difficult
to fit them as one path or separate paths. Consequently, we chose to fit the differ-
ence spectrum of 1 and 2; the similar coordination environments of the two Zn sites
removed the disordered Zn–N scattering in the data, and the fit provided informa-
tion regarding the Zn–X bond (see Figures S7 and S8 for further details). As a result,
Chem 4, 1–8, December 13, 2018
3

Please cite this article in press as: Wright et al., A Structural Mimic of Carbonic Anhydrase in a Metal-Organic Framework, Chem (2018), https://
doi.org/10.1016/j.chempr.2018.09.011
Figure 3. CO2 Isotherm at 298 K for MFU-4l (1; Red Diamonds) and MFU-4l-(OH) (2; Blue Circles)
˚
we obtained a satisfactory fit for the Zn–Cl bond distance of 2.17 A in 1 and for the
˚
Zn–O bond distance of 1.93 A in 2. For comparison, an extended lattice calculation
˚
on 2 (2-DFT [density functional theory]) computed a Zn–O bond distance of 1.84 A
˚
(see Supplemental Information), in good agreement with the EXAFS fitting (1.85 A
11
˚
for comparable molecular analogs) and with CA itself (1.9–2.1 A).
A critical step in CO2 hydration by CA is CO2 insertion into the Zn–OH bond. After
confirming the installation of the Zn–OH group, we probed its reactivity with CO2
by measuring its CO2 adsorption isotherm at 25ꢁC (Figures 3 and S3). Notably,
the adsorption isotherm profile featured a significant uptake of CO2 between
0 and 17 Torr, corresponding to 0.86 mmol of CO2 per g of MOF. Importantly,
this is equivalent to 1.0 mmol of CO2 per mmol of MOF or the insertion of CO2
into one of four Zn–OH sites in each SBU. The isotherm profile became shallower,
and uptake of the remaining ꢂ75% CO2 occurred over a large pressure range of
17–800 Torr. The total CO2 adsorption capacity at 800 Torr was 3.41 mmol/g
(4.05 mmol of CO2 per mmol of MOF), which coincides with the total number of
Zn–OH sites (3.36 mmol/g) in the MOF. The initial steep uptake supports a strong
interaction between 2 and the first CO2 equivalent (per SBU), most likely stemming
from the insertion of CO2 into the Zn–OH bond, a unique CO2 capture mechanism
that has been observed only once before in a MOF.22 After the initial steep uptake,
the isotherm profile became shallower with increasing coverage, suggesting a
change in the interaction between the MOF and CO2 or a shifting of the hydrox-
ide-carbonate equilibrium with increasing CO2 pressure.
To gain more insight into the framework-adsorbate interaction, we determined the
coverage-dependent CO2 adsorption enthalpy (Qst) by measuring the CO2 adsorp-
tion isotherms at three different temperatures (288, 298, and 308 K) and fitting the
isotherm data to the virial equation (Figures S13 and S14).23,24 Other fitting routines,
including the Langmuir, Langmuir-Freundlich, and two-site Langmuir models, pro-
vided less satisfactory fits (Figures S11 and S12). With the virial equation, Qst at
zero CO2 coverage is 81 kJ/mol, a value much higher than those typically observed
in MOFs and other microporous materials25 and consistent with a chemisorption
event. Indeed, the Qst for CO2 insertion into metal–hydroxide bonds in
[MnIIMnIII(OH)Cl2(bbta)] and [CoIICoIII(OH)Cl2(bbta)] (H2bbta = 1H,5H-benzo(1,2-
d:4,5-d)bistriazole) is 99 and 110 kJ/mol, respectively.22 Notably, the coverage-
dependent Qs suggests a bimodal adsorption profile (Figure S15). With increasing
4
Chem 4, 1–8, December 13, 2018

Please cite this article in press as: Wright et al., A Structural Mimic of Carbonic Anhydrase in a Metal-Organic Framework, Chem (2018), https://
doi.org/10.1016/j.chempr.2018.09.011
Figure 4. Difference DRIFTS Plots Collected on MFU-4l-(OH) (2) with Increasing Concentrations
of CO2 in an Argon Carrier Gas
The shaded areas denote new bands assigned to vibrations of a bicarbonate ligand (HCO3ꢀ). The
purple trace (0% CO2) was collected after the blue trace (4% CO2).
coverage, Qst stays relatively constant up to ꢂ0.5 equiv of CO2 per SBU and then
gradually decreases until it reaches ranges between 62 and 35 kJ/mol for the second
CO2 equivalent and between 35 and 20 kJ/mol for the third CO2 equivalent. The
large, bimodal variation in Qst suggests that the adsorption mechanism changes
with increasing CO2 coverage.
We modeled CO2 insertion into the Zn–OH bond by using DFT (PBE/6-311G**). To
do so, we used a truncated model of the SBU, Zn5(OH)4(BTA)6 (BTA = benzotria-
˚
zole). Optimization of the cluster model gave a Zn–O(H) bond distance of 1.85 A
˚
and Zn–N bond distances of 2.02–2.05 A, consistent with the EXAFS fit and
extended lattice DFT model. Notably, although the calculated enthalpy of CO2
insertion into the first Zn–OH unit was ꢀ50 kJ/mol, the second, third, and fourth
CO2 molecules were calculated to bind, also by insertion into Zn–OH units, with
a lower energy of approximately ꢀ44 kJ/mol. The trend of decreasing binding
enthalpy is also in qualitative agreement with the experimental data. We postulate
that the conversion of the highly basic –OH to weakly basic –OCO2H perturbs the
coordination environment of the SBU enough to weaken the nucleophilicity of the
remaining three Zn–OH sites within a given metal node. Given that the nucleo-
philic strength of Zn–OH correlates with its reactivity toward CO2,26 less nucleo-
philic Zn–OH units shift the Zn–OH + CO2 # Zn–OCO2H equilibrium to the left.
A comparable trend of initially high heat of adsorption to lower heat of adsorption
was observed in [MIIMIII(OH)Cl2(bbta)].22
To determine the nature of the chemisorption interaction, we monitored the reaction
between 2 and CO2 by DRIFTS. Exposing 2 to incremental increases in [CO2] from
1,000 to 4,000 ppm in an argon carrier gas resulted in the formation of new bands
in the DRIFT spectrum (Figure 4). By contrast, the addition of CO2 to 1 under the
same conditions resulted in no spectral changes (Figures S22 and S23). The new
spectral features in 2 are consistent with the formation of a zinc bicarbonate, which
is also the product of CO2 insertion in CA. Specifically, a new band at 3,626 cmꢀ1 was
assigned to the O–H stretch.13 Bands at 1,660 and 1,245 cmꢀ1 corresponded to the
O=C–O asymmetric and symmetric stretches, respectively.13 Additionally, bands at
1,018 and 1,423 cmꢀ1 were associated with bending and stretching vibrations of the
COH group in bicarbonate. The observed stretches were comparable to those of the
molecular analog tBu,MeTpZn(OCO2H), which featured asymmetric and symmetric
O=C–O stretches at 1,675 and 1,302 cmꢀ1, respectively.13 In addition, the IR
Chem 4, 1–8, December 13, 2018
5

Please cite this article in press as: Wright et al., A Structural Mimic of Carbonic Anhydrase in a Metal-Organic Framework, Chem (2018), https://
doi.org/10.1016/j.chempr.2018.09.011
spectrum of [MnIIMIII(OCO2H)Cl2(bbta)] (M = Mn, Co) featured stretches at 3,682
(O–H), 1,224 (symmetric O–C=O), and 1,050 cmꢀ1 (O–H bending).22 Purging the
DRIFTS setup with 100% argon flow resulted in the loss of the new signals and refor-
mation of the starting spectrum, consistent with a reversible reaction with CO2.
We further illustrated the critical role of the hydroxide ligand in CO2 uptake by
comparing the CO2 sorption properties of 1 and 2. MOF 1, containing a Zn–Cl
bond and a saturated metal center, showed a CO2 adsorption capacity at 800
Torr of 0.98 mmol of CO2/mmol of MOF (1.23 mmol of CO2 per g of MOF), which
is approximately four times smaller than the overall capacity of 2 (Figure 3). The
zero-coverage Qst, 20 kJ/mol, was also significantly lower than the Qst of 2 (Figures
S16–S19). Most notably, at 17 Torr, 2 adsorbed 0.99 mmol of CO2 per mmol of
MOF, which is approximately 300 times more CO2 than 1 (0.0031 mmol of CO2
per mmol of MOF) at the same pressure. The large uptake at low pressures dem-
onstrates the potential utility of CO2 insertion into the M–OH bond for CO2 sorp-
tion applications.
Next, we investigated the ability of 1 and 2 to perform CA-like reactivity. The central
function of CA is to rapidly interconvert H2O and CO2 as well as bicarbonate and
protons. We mimicked the interconversion in vitro by studying the exchange of
18O-labeled H2O with gaseous CO2 and periodically sampling the headspace to
monitor the distribution of CO2 isotopologues by gas chromatography-mass spec-
troscopy. Notably, in the presence of 2, the time to reach an equilibrium mixture of
the three possible CO2 isotopologues (C16O2, C16,18O2, and C18O2) was significantly
reduced to 5 hr from 10 hr without 2 (Figures S28–S31). Consequently, there must be
a rapid equilibrium for the Zn–OH + CO2 # Zn–OCO2H and Zn–OCO2H + H2O #
Zn–OH + H2CO3 reactions. It should be noted that 1 also decreased the time taken
to reach an equilibrium mixture of the isotopologues but was not as effective as 2
because it reduced the equilibrium time to 7 hr. This could be attributed to partial
hydrolysis of the Zn–Cl bond under the reaction conditions.
A second CA-like reaction often mimicked in vitro is the hydrolytic cleavage of
4-NPA. For instance, Jin and co-workers reported that the MOF CFA-1 (Zn5(OAc)4
(bibta)3, where bibta2ꢀ = 5,50-bibenzo[d][1,2,3]triazole, CFA-1)27 replicated the
chemistry of CA through hydration of CO2 and hydrolysis of 4-NPA.28 Using similar
reactivity conditions, we compared 1, 2, and CFA-1 for their activity in the hydrolysis
of 4-NPA (Figures S32–S34). The reactions were performed with ꢂ2.5 mol % (10 mol
% per Zn–X) MFU-4l (1) and MFU-4l-(OH) (2) in 50 mM HEPES buffer solution. They
were sampled after 3, 6, 24, and 48 hr and were monitored by UV-vis spectroscopy
for the formation of 4-nitrophenol (405 nm). A control reaction revealed small
(1.7% G 0.2%) conversion of 4-NPA after 24 hr in the absence of a MOF. Adding
1, 2, or CFA-1 to the reaction mixture increased the conversion yield to 11% G
2%, 15% G 2%, and 14% G 2%, respectively, at the 24 hr time point. In the conver-
sion to mmol of product per mmol of Zn–X sites, this corresponds to low values of
1.2 G 2, 1.8 G 2, and 1.0 G 1 for 1, 2, and CFA-1, respectively (Table S5). MFU-
4l-(OH) 2 outperformed both 1 and CFA-1, most likely because 1 and CFA-1 need
to undergo ligand exchange to generate the Zn–OH active site. The low turnover
number could be caused by pore blocking or slow diffusion through the micropores.
Critically, this demonstrates that the N3ZnOH coordination environment in 2 is a
functional mimic of the CA reactivity.
In summary, we have demonstrated that the metal nodes of MFU-4l can be function-
alized to create a high-fidelity biomimetic model of the CA active site. Reactions
6
Chem 4, 1–8, December 13, 2018

Please cite this article in press as: Wright et al., A Structural Mimic of Carbonic Anhydrase in a Metal-Organic Framework, Chem (2018), https://
doi.org/10.1016/j.chempr.2018.09.011
performed by the enzymes, such as CO2 insertion into the Zn–OH groups of the
metal nodes and hydrolysis of 4-NPA, are mimicked in the MOF. These results
demonstrate a rare example of functionalizing a MOF metal node to mimic the ac-
tivity of an enzyme. The nascent field of biomimetic chemistry in MOFs holds poten-
tial for future understanding of enzyme chemistry and also the potential to perform
enzyme-inspired reactions in stable frameworks. For instance, the insertion of CO2
into a metal-hydroxide bond could lead to CO2 separations with limited energy
cost in regeneration, a concept we are currently exploring through implementing
the chemistry of 2 in membrane composites for gas separations.
SUPPLEMENTAL INFORMATION
Supplemental Information includes Supplemental Experimental Procedures, 36 fig-
ures, and 8 tables and can be found with this article online at https://doi.org/10.
1016/j.chempr.2018.09.011.
ACKNOWLEDGMENTS
This study was supported by the National Science Foundation (grant DMR-1452612)
and used the Extreme Science and Engineering Discovery Environment, which is
supported by National Science Foundation grant ACI-1548562.
AUTHOR CONTRIBUTIONS
A.M.W. and M.D. conceptualized the experiments. A.M.W. synthesized materials
and performed experiments. Z.W., G.Z., R.J.C., and J.T.M. collected and analyzed
X-ray absorption spectroscopy data. R.W.D. and A.M.W. collected and analyzed
X-ray photoelectron spectroscopy and energy-dispersive X-ray spectroscopy data.
J.L.M. and C.H.H. performed computational studies. A.M.W. and M.D. wrote the
original manuscript. All authors proofread, commented on, and approved the final
manuscript for submission.
DECLARATION OF INTERESTS
The authors declare no competing interests.
Received: June 25, 2018
Revised: July 16, 2018
Accepted: September 19, 2018
Published: October 4, 2018
REFERENCES AND NOTES
1. Nath, I., Chakraborty, J., and Verpoort, F.
(2016). Metal organic frameworks mimicking
natural enzymes: a structural and functional
analogy. Chem. Soc. Rev. 45, 4127–4170.
5. Gu, Z.Y., Park, J., Raiff, A., Wei, Z., and Zhou,
H.C. (2014). Metal-organic frameworks as
biomimetic catalysts. ChemCatChem 6, 67–75.
enzyme immobilization techniques: metal-
organic frameworks as novel substrates.
Coord. Chem. Rev. 322, 30–40.
6. Lykourinou, V., Chen, Y., Wang, X.S., Meng, L.,
Hoang, T., Ming, L.J., Musselman, R.L., and Ma,
S. (2011). Immobilization of MP-11 into a
mesoporous metal-organic framework, MP-
11@mesoMOF: A new platform for enzymatic
catalysis. J. Am. Chem. Soc. 133, 10382–10385.
9. Li, P., Moon, S.Y., Guelta, M.A., Lin, L., Go´ mez-
Gualdro´ n, D.A., Snurr, R.Q., Harvey, S.P.,
Hupp, J.T., and Farha, O.K. (2016). Nanosizing
a metal-organic framework enzyme carrier for
accelerating nerve agent hydrolysis. ACS Nano
10, 9174–9182.
2. Chen, Y., and Ma, S. (2016). Biomimetic
catalysis of metal-organic frameworks. Dalton
Trans. 45, 9744–9753.
3. Chen, K., and Wu, C.D. (2018). Designed
fabrication of biomimetic metal–organic
frameworks for catalytic applications.
Coord. Chem. Rev. Published online
February 13, 2018. https://doi.org/10.1016/j.
ccr.2018.01.016.
7. Li, P., Modica, J.A., Howarth, A.J., Vargas, L.E.,
Moghadam, P.Z., Snurr, R.Q., Mrksich, M.,
Hupp, J.T., and Farha, O.K. (2016). Toward
design rules for enzyme immobilization in
hierarchical mesoporous metal-organic
frameworks. Chem 1, 154–169.
10. Liang, J.Y., and Lipscomb, W.N. (1990). Binding
of substrate CO2 to the active site of human
carbonic anhydrase II: a molecular dynamics
study. Proc. Natl. Acad. Sci. USA 87, 3675–
3679.
4. Zhao, M., Ou, S., and Wu, C.D. (2014). Porous
metal-organic frameworks for heterogeneous
biomimetic catalysis. Acc. Chem. Res. 47,
1199–1207.
11. Parkin, G. (2004). Synthetic analogues relevant
to the structure and function of zinc enzymes.
Chem. Rev. 104, 699–767.
8. Mehta, J., Bhardwaj, N., Bhardwaj, S.K., Kim,
K.-H., and Deep, A. (2016). Recent advances in
Chem 4, 1–8, December 13, 2018
7

Please cite this article in press as: Wright et al., A Structural Mimic of Carbonic Anhydrase in a Metal-Organic Framework, Chem (2018), https://
doi.org/10.1016/j.chempr.2018.09.011
12. Alsfasser, R., Vahrenkamp, H., Trofimenko, S.,
Looney, A., and Parkin, G. (1991). A
mononuclear zinc hydroxide complex
stabilized by a highly substituted tris(pyrazolyl)
hydroborato ligand: analogies with the enzyme
carbonic anhydrase. Inorg. Chem. 30, 4098–
4100.
polymerization enabled by cation exchange in
a metal-organic framework. J. Am. Chem. Soc.
138, 10232–10237.
capture from high-humidity flue gas. Energy
Environ. Sci. 8, 1011–1016.
23. Czepirski, L., and JagieŁŁo, J. (1989). Virial-type
thermal equation of gas-solid adsorption.
Chem. Eng. Sci. 44, 797–801.
18. Denysenko, D., Grzywa, M., Jelic, J., Reuter, K.,
and Volkmer, D. (2014). Scorpionate-type
coordination in MFU-4l metal-organic
frameworks: small-molecule binding and
activation upon the thermally activated
formation of open metal sites. Angew. Chem.
Int. Ed. 53, 5832–5836.
ꢀ
24. Dinca, M., Dailly, A., Liu, Y., Brown, C.M.,
Neumann, D.A., and Long, J.R. (2006).
Hydrogen storage in a microporous metal-
organic framework with exposed Mn2+
coordination sites. J. Am. Chem. Soc. 128,
16876–16883.
13. Looney, A., Parkin, G., Alsfasser, R., Ruf, M., and
Vahrenkamp, H. (1992). Zinc pyrazolylborate
complexes relevant to the biological function
of carbonic anhydrase. Angew. Chem. Int. Ed.
31, 92–93.
19. Dubey, R.J., Comito, R.J., Wu, Z., Zhang, G.,
Rieth, A.J., Hendon, C.H., Miller, J.T., and
25. Sumida, K., Rogow, D.L., Mason, J.A.,
McDonald, T.M., Bloch, E.D., Herm, Z.R., Bae,
T.H., and Long, J.R. (2012). Carbon dioxide
capture in metal-organic frameworks. Chem.
Rev. 112, 724–781.
14. Looney, A., Han, R., McNeill, K., and Parkin, G.
(1993). Tris(pyrazolyl)hydroboratozinc
hydroxide complexes as functional models for
carbonic anhydrase: on the nature of the
bicarbonate intermediate. J. Am. Chem. Soc.
115, 4690–4697.
ꢀ
Dinca, M. (2017). Highly stereoselective
heterogeneous diene polymerization by
Co-MFU-4l: a single-site catalyst prepared by
cation exchange. J. Am. Chem. Soc. 139,
12664–12669.
26. Brauer, M., Pe´ rez-Lustres, J.L., Weston, J., and
¨
20. Wright, A.M., Rieth, A.J., Yang, S., Wang, E.N.,
Anders, E. (2002). Quantitative reactivity model
for the hydration of carbon dioxide by
biomimetic zinc complexes. Inorg. Chem. 41,
1454–1463.
15. Biswas, S., Grzywa, M., Nayek, H.P., Dehnen, S.,
Senkovska, I., Kaskel, S., and Volkmer, D.
(2009). A cubic coordination framework
constructed from benzobistriazolate ligands
and zinc ions having selective gas sorption
properties. Dalton Trans. 6487–6495.
ꢀ
and Dinca, M. (2018). Precise control of pore
hydrophilicity enabled by post-synthetic cation
exchange in metal-organic frameworks. Chem.
Sci. (Camb.) 9, 3856–3859.
27. Schmieder, P., Denysenko, D., Grzywa, M.,
21. Denysenko, D., Jelic, J., Reuter, K., and
Volkmer, D. (2015). Postsynthetic metal and
ligand exchange in MFU-4l: a screening
approach toward functional metal-organic
frameworks comprising single-site active
centers. Chemistry 21, 8188–8199.
Baumgartner, B., Senkovska, I., Kaskel, S.,
¨
16. Denysenko, D., Werner, T., Grzywa, M., Puls, A.,
Hagen, V., Eickerling, G., Jelic, J., Reuter, K.,
and Volkmer, D. (2012). Reversible gas-phase
redox processes catalyzed by co-exchanged
MFU-4l(arge). Chem. Commun. (Camb.) 48,
1236–1238.
Sastre, G., van Wullen, L., and Volkmer, D.
¨
(2013). CFA-1: the first chiral metal-organic
framework containing Kuratowski-type
secondary building units. Dalton Trans. 42,
10786–10797.
22. Liao, P.-Q., Chen, H., Zhou, D.-D., Liu, S.-Y., He,
C.-T., Rui, Z., Ji, H., Zhang, J.-P., and Chen,
X.-M. (2015). Monodentate hydroxide as a
super strong yet reversible active site for CO 2
28. Jin, C., Zhang, S., Zhang, Z., and Chen, Y.
(2018). Mimic carbonic anhydrase using metal-
organic frameworks for CO2 capture and
conversion. Inorg. Chem. 57, 2169–2174.
17. Comito, R.J., Fritzsching, K.J., Sundell, B.J.,
ꢀ
Schmidt-Rohr, K., and Dinca, M. (2016). Single-
site heterogeneous catalysts for olefin
8
Chem 4, 1–8, December 13, 2018