Deciphering the Mechanism of Human Carbonic Anhydrases Inhibition with Sulfocoumarins: Computational and Experimental Studies

Article abstract of DOI:10.1002/chem.201800941

The reaction mechanism of the carbonic anhydrase-mediated hydrolysis of sulfocoumarins to sulfonic acids has been investigated on an enzyme cluster model using the B3LYP hybrid density functional theory (DFT) and the QST procedure for the Transition State (TS) search. A multistep process was highlighted, with the rate-determining step identified in the initial dual nucleophilic/acidic attack of the zinc-bound hydroxide ion to the sulfocoumarin sulfur atom and to the C3=C4 double bond. The reported multi-step process, combined to SAR analysis on a new set of derivatives, highlighted unprecedented mechanistic aspects of the CA-mediated prodrug activation, which in turn possess relevant consequences to the isoforms-selective inhibition profiles reported by such a class of compounds.

Full text of DOI:10.1002/chem.201800941

A Journal of
Accepted Article
Title: Deciphering the mechanism of human carbonic anhydrases
inhibition with sulfocoumarins: computational and experimental
studies.
Authors: Alessio Nocentini, Fabrizio Carta, Muhammet Tanc, Silvia
Selleri, Claudiu T. Supuran, Carla Bazzicalupi, and Paola
Gratteri
This manuscript has been accepted after peer review and appears as an
Accepted Article online prior to editing, proofing, and formal publication
of the final Version of Record (VoR). This work is currently citable by
using the Digital Object Identifier (DOI) given below. The VoR will be
published online in Early View as soon as possible and may be different
to this Accepted Article as a result of editing. Readers should obtain
the VoR from the journal website shown below when it is published
to ensure accuracy of information. The authors are responsible for the
content of this Accepted Article.
To be cited as: Chem. Eur. J. 10.1002/chem.201800941
Link to VoR: http://dx.doi.org/10.1002/chem.201800941
Supported by

10.1002/chem.201800941
Chemistry - A European Journal
COMMUNICATION
Deciphering the mechanism of human carbonic anhydrases
inhibition with sulfocoumarins: computational and experimental
studies.
Alessio Nocentini,[a,b] Fabrizio Carta,[b] Muhammet Tanc,[b] Silvia Selleri,[b] Claudiu T. Supuran,[b] Carla
Bazzicalupi,*[c] Paola Gratteri*[a]
ion by means of a proton transfer reaction from the Zn2+-
Abstract: The reaction mechanism of the carbonic anhydrase-
mediated hydrolysis of sulfocoumarins to sulfonic acids has been
bound water molecule to an external proton acceptor or to
investigated on an enzyme cluster model using the B3LYP hybrid
an active site residue.[7]
density functional theory (DFT) and the QST procedure for the
Besides the hydration on CO2, some CAs are endowed with
a certain substrate versatility, which includes COS, CS2
cyanamides and aldehydes hydration reactions.[5] α-CAs
are also reported to possess hydrolytic activities towards
esters of carboxylic acids, sulfonic acids, phosphates and
thioesters.[5,8]
Transition State (TS) search. A multistep process was highlighted,
with the rate determining step identified in the initial dual
nucleophilic/acidic attack of the zinc bound-hydroxide ion to the
sulfocoumarin sulfur atom and to the C3-C4 double bond. The
reported multi-step process, combined to SAR analysis on a new set
of derivatives, highlighted unprecedentedly considered mechanistic
aspects of the CA-mediated prodrug activation, which in turn do
possess relevant consequences to the isoforms-selective inhibition
profiles reported by such a class of compounds.
The esterase activity shown by the α-CAs, was the
enzymatic blueprint which allowed us to make use of
prodrugs of the coumarin and sulfocoumarin type as novel
classes of CA Inhibitors (CAIs).[1,9-12] In analogy to the
coumarins, the sulfocoumarins are not per se able to inhibit
the CA activity. Instead they interfere with the enzyme by
acting as efficient substrates for the esterase (sulfatase)
activity, generating the actual inhibitor in situ.[1-3] Thereafter
the formed sulfonic acid inhibit the enzyme by means of a
stable adduct and a particular inhibition mechanism, i.e.,
anchoring to the non-protein zinc ligand.[1-3] The prodrug
features of the sulfocoumarin scaffold was also consistent
The discovery of the coumarin bioisosters 1,2-
benzoxathiine-2,2-dioxides, also named sulfocoumarins, as
novel and selective inhibitors of the zinc metalloenzymes α-
carbonic anhydrases (CAs, EC 4.2.1.1) is the result of
intense efforts towards the development of innovative
compounds as potential tools for the treatment of various
diseases.[1-4] The role of CAs in cells as well as in complex
organisms, is primarily related to their ability to reversibly
catalyze the carbon dioxide hydration reaction to afford
bicarbonate and a proton.[5,6] Such an equilibrium is crucial
for a vast array of physio/pathological processes.[5-7] The
CA mediated catalytic pathway proceeds at first with the
nucleophilic attack of a Zn2+-bound hydroxide ion towards
the CO2 substrate with consequent formation of a CA-
−
HCO3 adduct, which is then displaced from the active site
by an additional water molecule. The final, rate limiting step,
regenerates the catalytically active Zn2+-bound hydroxide
Figure 1. A) Structure of the 6-bromosulfocoumarin; B) The corresponding
sulfonic acid was obtained from in situ hydrolysis and binds into the hCA II/IX
mimic active site, by anchoring of the sulfonic acid group to the zinc-
coordinated water molecule (PDB 4BCW).[1]
[a]
Dr. A. Nocentini, Prof. P. Gratteri
Department NEUROFARBA – Pharmaceutical and nutraceutical
section; Laboratory of Molecular Modeling Cheminformatics &
QSAR, University of Firenze, via Ugo Schiff 6, 50019 Sesto
Fiorentino (Italy).
E-mail: paola.gratteri@unifi.it
Dr. A. Nocentini , Dr. F. Carta, Dr. M. Tanc, Prof. S. Selleri, Prof.
C.T. Supuran
Department NEUROFARBA – Pharmaceutical and nutraceutical
section, University of Firenze, via Ugo Schiff 6, 50019 Sesto
Fiorentino (Italy)
with the longer time of incubation required to exert in vitro
inhibition against the α-CAs (6h vs 15 minutes for the
classical zinc binders CAIs of the sulfonamide type).[1,9,12]
Evidences for the CA promoted intramolecular sulfonic
ester cleavage was obtained by means of X-ray co-
crystallographic experiments conducted by soaking a hCA
[b]
[c]
Prof. C. Bazzicalupi
Department of Chemistry "U. Schiff", University of Florence , Via
della Lastruccia 3, 50019 Sesto Fiorentino (Italy).
E-mail: carla.bazzicalupi@unifi.it
II/IX mimic isoform with the 6-bromo-sulfocoumarin
(Figure 1).[1]
1
The 2Fo–Fc density maps showed a well-defined density for
the in situ formed sulfonate, which anchors to the Zn-bound
Supporting information for this article is given via a link at the end of the
document.((Please delete this text if not appropriate))
This article is protected by copyright. All rights reserved.

10.1002/chem.201800941
Chemistry - A European Journal
COMMUNICATION
water within the hCA II/IX mimic active site.[1] The adduct is
further stabilized by means of a network of hydrogen bonds
occurring between the ligand and the Pro202/Thr200 amino
acid residues from the enzyme active site.[1]
The final compounds 10
-
14 were screened in vitro for the
inhibition against four physiologically relevant hCA
isoforms, the cytosolic hCAI and II and the trasmembrane
tumor-associated hCAIX and XII.
To date a detailed reaction mechanism for the CA-mediated
intramolecular sulfonic esters hydrolysis has not yet been
reported. In recent studies, we investigated a variety of
sulfocoumarins bearing electro-withdrawing (EWG) or
electro-donating (EDG) groups placed at the 6- and 7-
position.[1-3,13,14] Herein, as extension of our previous
investigations, we report novel analogues decorated at the
6- and 7-position with chemical moieties endowed with
different electronic properties (EWG and EDG, scheme 1)
and their inhibition profiles against the most relevant human
(h)CAs are reported. Since computational studies are
validated tools to provide deep insights into enzymatic
mechanisms,[15-20] we made use of the B3LYP hybrid
density functional theory (DFT) to study the hydrolysis of
the unsubstituted sulfocoumarin (compound 10 – scheme
1) upon the CA-esterase activity, by using the X-Ray crystal
structure of CA II/IX mimic in complex with the hydrolysed
6-bromosulfocoumarin as a model.
Table 1: Inhibition data of hCA isoforms I, II, IX and XII with sulfocoumarins 10-
14 compared to the standard sulfonamide inhibitor acetazolamide (AAZ) by the
stopped flow CO2 hydrase assay.[21]
KI (nM)a
Cmp
R
hCA I
hCA II
hCA IX
hCA XII
H
>100000
92000±4800
6780±560
>100000
>100000
>100000
>100000
250±15
>100000
>100000
8890±710
>100000
>100000
>100000
>100000
12±0.8
432.2±33.1
3770±240
46.0±3.1
136.3±7.5
37.0±2.1
33.5±1.9
33.7±1.8
25±1.4
69.5±5.1
3160±210
23.0±1.2
89.5±5.4
81.7±4.3
38.4±2.2
60.9±4.6
5.7±0.5
10
Ab
6-NO2
6-NH2
6-Cl
7-NO2
7-NH2
7-Cl
Bc
11
12
13
14
AAZ
-
a. Inhibition data are expressed as means ± SEM of 3 different assays; b.
compound 10 in ref. 1; c. compound 11 in ref. 1.
The following Structure-Activity-Relationships (SARs) can
be draw from the inhibition data of Table 1:
(i) In agreement with previous literature reports,[1-4] the
sulfocoumarins 10-14 were ineffective in inhibiting the
isoform hCA I and II. The only exception was the compound
The novel sulfocoumarins 10-14 were obtained by applying
the general synthetic strategy firstly reported by
Zalubovski’s group (Scheme 1).[4,13]
B
which was effective against both cytosolic isoforms at low
micromolar concentrations (KIs 6780±560 and 8890±710
μM respectively).
(ii) The transmembrane hCA isoforms resulted inhibited by
the sulfocoumarins here reported, with KI values spanning
between 33.5±1.9 - 3770±240 nM and 23.0±1.2 - 3160±210
nM against the hCA IX and hCA XII respectively. The
introduction of substituents of the NO2 or NH2 type on the 6
Scheme 1. General synthetic procedure of sulfocoumarins 10-14.
Reaction of the proper 2-hydroxybenzaldehydes 1-3 with
mesyl chloride afforded the mesylate derivates 4-6, which
were treated with 1,8-diazabicyclo[5.4.0]undec-7-ene
(DBU), with formation of the racemic 4-hydroxy-3,4-
dihydrosulfocoumarins 7-9 in mixture with their dehydration
products 10-12 The dehydration was further pushed
forwards by treating the mixtures with POCl3 in pyridine. 7-
NH2-sulfocoumarin 13 was obtained by reducing the
correspondent nitro-derivative 12 with iron (0), which in turn
was converted to the chlorine compound 14 through
Sandmeyer reaction with CuCl2 in the presence of tert-butyl
nitrite.
position of the sulfocoumarin ring, as in
A and B,
Figure 2. Superimposition of the docked pose of sulfocoumarin 10 (blue)
and the hydrolized form of 6-bromosulfocoumarin (black) in hCA II/IX
mimic active site (4BCW).
This article is protected by copyright. All rights reserved.

10.1002/chem.201800941
Chemistry - A European Journal
COMMUNICATION
Figure 3. A. Reaction path for step I. Thr199 omitted for clarity; B. Drawings of the resonant structures for II. Zinc coordination bonds are represented
as hashed lines; C. Energy profile for the calculated multi-step reaction path of the unsubstituted sulfocoumarin CA-mediated hydrolysis. Red dashed
lines
represent
reaction
energies,
black
dashed
lines
activation
energies
(energy
values
in
kcal/mol).
led to opposite changes in the hCA IX and XII inhibitory
the crystal structure established a H-bond with the sulfonate
moiety, and (v) an additional water molecule extrapolated
from the X-ray structure. In the crystal structure, the
hydrolysed 6-substituted sulfocoumarin locates its sulfonate
oxygen 2.8 Å apart from the Zn ion. The unsubstituted
sulfocoumarin 10 was considered as the best candidate to
initially identify the steps through which the process takes
place, owing to minor electronic effects by absence of
appended substituents. Compound 10 was docked into the
4BCW active site (Figure 2), and subsequently the complex
was superposed with the built cluster model to obtain the
starting position of 10 into the model system. The transition
state (TS) search undertaken (ESI, Theoretical studies)
highlighted the presence of a multi-step process (Figure 3).
(i) In the optimized docked pose of the sulfocoumarin 10
within the cluster model (I, cfr ESI for the procedure used to
obtain the I, II and TS I systems, i.e. adduct, structures), the
distance between the negative oxygen atom of the Zn-
bound hydroxide ion and the sulfonate ester sulfur atom
was of 3.87 Å (Figure 3A). Such a distance makes the
hydroxide ion suitable to elicit a nucleophilic attack towards
the electrophilic sulfonic ester, thus inducing the cycle to
open by lengthening the distance between sulfur and the
endocyclic oxygen. At the same time the hydroxide
hydrogen protonates the double bond of the forming
ethenylsulfonic acid, with the formation of the stable
carbocation locating the positive charge on C4 of the
starting sulfocoumarin. The simultaneous occurring of these
two events provides the molecule with the degree of
properties in comparison to the unsubstituted 10 (KIs of
432.2±33.1 and 69.5±5.1 nM, respectively). Indeed,
inhibit both isozymes with KIs ranging between 3160±210
and 3770±240 nM, whereas exhibited enhanced
A
B
inhibition profiles in the range 23.0±1.2 - 46.0±3.1 nM.
Among 7-substituted derivatives, all functionalizations do
increase hCA IX inhibition potency with respect to 10, with
hCA XII being instead more significantly affected uniquely
in the case of 13 (7-NH2, KI of 38.4±2.2 nM).
The cluster-based approach was considered to overcome
the complexity in treating the overall enzyme systems with
the quantum-mechanical (QM) transition state search
methods.[17-19] It is worth stressing the importance of the
identification of
a proper cluster model, which was
propaedeutical to unveil the reaction mechanism.[17-19]
Indeed, considering that the differently substituted
sulfocoumarins possessed diverse activity on the CA
isoforms, we had to look for the common elements deemed
to be involved in the esterase process among the different
isozymes, whose active site cavities differ for residues
composition and molecular architectures.[5,7,22] As a result,
these latter define lightly narrower binding pockets for hCA
I, II in comparison to hCA IX and XII.[22]
Thus, a model of the active site was built using the PDB
4BCW coordinates, which includes the essential parts
involved in the reaction: (i) the Zn ion; (ii) the Zn-bound OH-
(as present in the active form of the enzyme); (iii) three
imidazole rings representative of the three side chains of
the His residues; (iv) the truncated Thr199 residue, that in
This article is protected by copyright. All rights reserved.

10.1002/chem.201800941
Chemistry - A European Journal
COMMUNICATION
Figure 4. Reaction path for steps II, III and IV. Thr199 omitted for clarity. Intermediate III reported as III (H2O) for clarity.
freedom necessary to rearrange, leading to the first reaction
intermediate (II). The bond lengths featuring II, are
consistent with its stabilization by means of two main
resonance structures of the carbocation and the phenoxide
ion type (Figure 3B). The activation barrier for the first step
was 36.8 kcal/mol (Figure 3C).
restoring of the zinc bound hydroxide ion. Noteworthy, it
should be mentioned that in the native CA II crystal
structure (PDB 2CBA)[23] a zinc-bound water molecule has
been observed.
The energy profile shown in Figure 3C evidences that
product V is thermodynamically unstable with respect to its
precursor IV by 7.3 kcal/mol. These data are in agreement
with the cluster model herein considered where the
bipyramidal coordination of IV is expected to prevail on the
pseudo tetrahedral one of V. Nevertheless, it is noteworthy
that even in the 4BCW adduct, the coordination geometry of
the zinc ion is actually better described as a distorted
tetrahedron rather than as a trigonal bipyramid, with an
oxygen atom from the hydrolysed inhibitor at only 2.8 Å
from the metal centre. Indeed, the relative instability
ascribed to the pseudo-trigonal bipyramid coordination
pattern found for the adduct IV is compensated by
additional non-bonded interactions (i.e. H-bonds and
hydrophobic contacts) occurring between the ligand and the
amino acid residues of the enzymatic cleft, thus contributing
to further stabilize the orientation assumed by the ligand
(Figure 2).
(ii) The second step pertains the Z to E isomerization
(Figure 4). In this step an intramolecular acid-base reaction
occurs, where the phenoxide moiety acts as a base to
deprotonate the CH2 moiety (C3 position of the starting
sulfocoumarin 10). As for step (i), also here a simultaneous
intramolecular rearrangement takes place, providing the
less hindered isomer E (III). Overall, the II
 III reaction
energy is -7.4 kcal/mol (Figure 3C), while the activation
barrier is 16.1 kcal/mol.
(iii) As known, the displacement of a zinc-bound ligand in
the enzyme active site proceeds through the so-called
water-exchange pathway.[5] Thus, an additional water
molecule was added to III (ESI, Theoretical Studies). The
observed TS evidences that the added water molecule
approaches the Zn ion, eliciting a reorganization of the
coordination sphere, by which a trigonal bipyramid is
formed having the sulfonate oxygen, and its opposite
histidine nitrogen in apical positions (Figure 4). The
activation barrier is only 0.4 kcal/mol (Figure 3C).
In conclusion, our theoretical model indicated the first step, (i.e.
the nucleophilic attack of the zinc-hydroxide ion on the cyclic
sulfonate), as the rate-determining step for the entire hydrolytic
reaction. This finding is not surprising, as the cycle opening and
the overall ligand rearrangement required a dual attack of the
zinc-bound O-H group both to the sulfocoumarin sulfur atom and
to the C3-C4 double bond. Structure-activity-relationship (SAR)
analysis based on the obtained inhibition assay was rather flat.
We propose that the CA isoform-selective inhibition of
sulfocoumarins may be properly ascribed to the mechanism
(iv) Finally, the sulfonate moiety slightly moves away from
the Zn ion and the remaining components of the trigonal
bipyramid rearrange forming
a
pseudo tetrahedrical
coordination system (Figure 4). The activation barrier for
this step was 9.8 kcal/mol. The anchoring of the sulfonate
to the zinc coordinated water molecule is expected to
prevent its deprotonation and, as a consequence, the
This article is protected by copyright. All rights reserved.

10.1002/chem.201800941
Chemistry - A European Journal
COMMUNICATION
[11] C. T. Supuran, J. Enzyme Inhib. Med. Chem., 2016, 31, 345-360.
[12] F. Carta, A. Maresca, A. Scozzafava, C. T. Supuran, Bioorg.
Chem. Lett. 2012, 22, 267-270.
[13] M. Tanc, F. Carta, A. Scozzafava, C. T. Supuran, Org. Biomol.
Chem., 2015, 13, 77-80.
[14] A. Nocentini, M. Ceruso, F. Carta,C. T. Supuran, J. Enzyme Inhib.
Med. Chem. 2016, 31, 1226-1233.
[15] G. Tian, Y. Liu, Phys. Chem. Chem. Phys., 2017, 19, 7733- 7742.
[16] P. Janoš, T. Trnka, S. Kozmon, I. Tvaroška, J. Koča. J. Chem.
Theory Comput., 2016, 12, 6062-6076.
[17] P. Georgieva and F. Himo, Chem. Phys. Lett. 2008, 463, 214-218.
[18] V. Pelmenschikov, M. R. Blomberg, J. Siegbahn, Biol. Inorg. Chem.,
2002, 7, 284-298.
[19] V. Navrátil, V. Klusák, L. Rulíšek, Chemistry, 2013, 19, 16634-
16645.
[20] R. Z. Liao, J. G. Yu, F. Himo, J. Phys. Chem. B. 2010, 114, 2533-
2540.
through each α-CA hydrolyzes different substrates rather to
depend from conventional enzyme-ligand interactions occurring
between the amino acids of the enzymatic cleft and the product
obtained from the hydrolytic process. In this context, it is
reasonable to hypothesize that the ligand rearrangements
described in the reaction mechanism (cluster model) strictly
depend on the active sites architectures when taking place
within the overall binding cavities. As a results, the overall
molecular architectures as well as the different isoforms binding
site residues could allow or impair the hydrolytic process, with
the roomier active site of hCA IX and XII in comparison to hCA I
and II being more suitable to accommodate the ligand
rearrangements required to start and complete the process.[5,7,22]
Therefore, the knowledge of such mechanism might further aid
in the design of novel specific enzymatic modulators and
currently further studies are ongoing in this direction.
Med.
[21] R. G. Khalifah, J. Biol. Chem. 1971, 246, 2561-2573.
[22] V. Alterio, M. Hilvo, A. Di Fiore, C. T. Supuran, P. Pan, Parkkila, A.
Scaloni, J. Pastorek, S. Pastorekova, C. Pedone, A. Scozzafava, S.
M. Monti, G. De Simone, Proc. Natl. Acad. Sci. U.S.A. 2009, 106,
16233-16238.
[23] K. Hakansson, M. Carlsson, L. A. Svensson, A. Liljas, J. Mol. Biol.
1992, 227, 1192-1204.
Acknowledgements
This work was supported by the EU FP7 ITN Project
Dynano (to M.T. and C.T.S.) and by grants from the
University of Florence (Ateneo 2016). The PhD fellowship
for A.N. was funded by University of Florence.
Conflicts of interest
The authors state no conflicts of interest.
Keywords: Carbonic Anhydrase, sulfocoumarin, selective
inhibition, reaction mechanism, quantum-mechanics.
[1]
[2]
K. Tars, D. Vullo, A. Kazaks, J. Leitans, A. Lends, A. Grandane,
Zalubovskis, A. Scozzafava, C. T. Supuran, J. Med. Chem. 2013, 56,
293-300.
A. Grandane, M. Tanc, L. Di Cesare Mannelli, F. Carta, C.
Ghelardini, R. Zalubovskis, C. T. Supuran, J. Med. Chem. 2015, 58,
3975-3983.
R.
[3]
[4]
[5]
A. Grandane, M. Tanc, R. Zalubovski, C. T. Supuran, Bioorg.
Chem. 2014, 22,1522-1528.
A. Grandane, S. Belyakov, P. Trapencieris, R. Zalubovskis,
Tetrahedron, 2012, 68, 5541-5546.
V. Alterio, A. Di Fiore, K. D’Ambrosio, C. T. Supuran, G. De Simone,
Chem. Rev. 2012, 112, 4421-4468.
Med.
[6]
[7]
[8]
C. T. Supuran, Nat. Rev. Drug Discovery, 2008, 7, 168-181.
C. T. Supuran. Biochem. J. 2016, 473, 2023-2032.
M. Tanc, F. Carta, A. Scozzafava, C. T. Supuran, ACS Med. Chem.
Lett. 2015, 6, 292-295.
[9]
A. Maresca, C. Temperini, H. Vu, N. B. Pham, S. A. Poulsen, A.
Scozzafava, R. J. Quinn, R. J., C. T. Supuran, J. Am. Chem. Soc.
2009, 131, 3057-3062.
[10] A. Maresca, C. Temperini, L. Pochet, B. Masereel, A.Scozzafava, C.
T. Supuran, J. Med. Chem. 2010, 53, 335-344.
This article is protected by copyright. All rights reserved.

10.1002/chem.201800941
Chemistry - A European Journal
COMMUNICATION
COMMUNICATION
The reported multistep process for the
carbonic anhydrase-mediated
hydrolysis of sulfocoumarins
Alessio Nocentini,Fabrizio Carta,
Muhammet Tanc, Silvia Selleri, Claudiu
T. Supuran,Carla Bazzicalupi,*Paola
Gratteri*
highlighted unprecedentedly
considered mechanistic aspects of the
CA-mediated prodrug activation.
Page No. – Page No.
Deciphering the mechanism of
human carbonic anhydrases
inhibition with sulfocoumarins:
computational and experimental
studies.
This article is protected by copyright. All rights reserved.