Design of two-tail compounds with rotationally fixed benzenesulfonamide ring as inhibitors of carbonic anhydrases

Article abstract of DOI:10.1016/j.ejmech.2018.06.059

Rational design of compounds that would bind specific pockets of the target proteins is a difficult task in drug design. The 12 isoforms of catalytically active human carbonic anhydrases (CAs) have highly similar active sites that make it difficult to des

Full text of DOI:10.1016/j.ejmech.2018.06.059

Accepted Manuscript
Design of two-tail compounds with rotationally fixed benzenesulfonamide ring as
inhibitors of carbonic anhydrases
Audrius Zakšauskas, Edita Čapkauskaitė, Linas Jezepčikas, Vaida Linkuvienė, Miglė
Kišonaitė, Alexey Smirnov, Elena Manakova, Saulius Gražulis, Daumantas Matulis
PII:
S0223-5234(18)30550-6
10.1016/j.ejmech.2018.06.059
EJMECH 10529
DOI:
Reference:
To appear in: European Journal of Medicinal Chemistry
Received Date: 24 April 2018
Revised Date: 11 June 2018
Accepted Date: 23 June 2018
Please cite this article as: A. Zakšauskas, E. Čapkauskaitė, L. Jezepčikas, V. Linkuvienė, Miglė.
Kišonaitė, A. Smirnov, E. Manakova, S. Gražulis, D. Matulis, Design of two-tail compounds with
rotationally fixed benzenesulfonamide ring as inhibitors of carbonic anhydrases, European Journal of
Medicinal Chemistry (2018), doi: 10.1016/j.ejmech.2018.06.059.
This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to
our customers we are providing this early version of the manuscript. The manuscript will undergo
copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please
note that during the production process errors may be discovered which could affect the content, and all
legal disclaimers that apply to the journal pertain.

ACCEPTED MANUSCRIPT
65
8a
8b
CA XII
Kd= 20.8 nM
60
55
Kd= 125 nM
CA XII
7a
7b
50
0
10-6
10-5
Lt(M)
10-4
10-3
Zn(II)

ACCEPTED MANUSCRIPT
Design of two-tail compounds with rotationally fixed
benzenesulfonamide ring as inhibitors of carbonic
anhydrases
Audrius Zakšauskasa, Edita Čapkauskaitėa, Linas Jezepčikasa, Vaida Linkuvienėa, Miglė
Kišonaitėa, Alexey Smirnovb, Elena Manakovab, Saulius Gražulisb, Daumantas Matulisa,*
aDepartment of Biothermodynamics and Drug Design, Institute of Biotechnology, Vilnius
University, Saulėtekio al. 7, Vilnius LT-10257, Lithuania
bDepartment of Protein – DNA Interactions, Institute of Biotechnology, Vilnius University,
Saulėtekio al. 7, Vilnius LT-10257, Lithuania
* Corresponding author: Daumantas Matulis
Address:
Saulėtekio al. 7, Vilnius LT-10257, Lithuania.
Tel. +370-5-223-4364
Fax.+370-5-223-4367
Email: matulis@ibt.lt
Abstract
Rational design of compounds that would bind specific pockets of the target proteins is a
difficult task in drug design. The 12 isoforms of catalytically active human carbonic anhydrases
(CAs) have highly similar active sites that make it difficult to design inhibitors selective for one
or several CA isoforms. A series of CA inhibitors based on 2-chloro/bromo-benzenesulfonamide
that is largely fixed in the CA active site together with one or two tails yielded compounds that
were synthesized and evaluated as inhibitors of CA isoforms. Introduction of a second tail had
significant influence on the binding affinity and two-tailed compounds in most cases provided
high affinity and selectivity for CA IX and CA XIV. The contacts between several compounds
and CA amino acids were determined by X-ray crystallography. Together with the intrinsic
enthalpy and entropy of binding they provided the structure-thermodynamics correlations for this
series of compounds with the insight how to rationally build compounds with desired CA
isoform as a target.

ACCEPTED MANUSCRIPT
Keywords
Fluorescent thermal shift assay; benzenesulfonamide; carbonic anhydrase; inhibitor; structure-
thermodynamics relationships; X-ray crystallography.
1. Introduction
Rational design of small molecular weight compounds that would tightly and specifically bind a
desired binding site on a disease-related target protein is still an elusive task despite efforts of
scientists to find a common path to design pharmaceutical compounds. This is because the
correlations between compound chemical structures and the energetics of their interaction with
protein molecules are still poorly understood.
Here we describe a novel group of substituted aromatic sulfonamides in the attempts to
understand the structure–thermodynamics relationships of their binding to Carbonic anhydrases
(CA, EC 4.2.1.1), therapeutically-relevant protein targets which are zinc-containing
metalloenzymes and catalyze the reversible reaction of carbon dioxide hydration into bicarbonate
and proton[1]. CAs participate in various physiological processes related to respiration,
bicarbonate transport between lungs and metabolizing tissues, pH and CO2 homeostasis,
electrolyte secretion in many tissues, etc.[2,3]. The family of human CAs comprises 15 isoforms
(isozymes) – 12 of them are catalytically active, while the remaining isoforms CA VIII, CA X,
and CA XI are inactive. Twelve active human CA isoforms possess different catalytic activity
and cellular localization – five are cytosolic (CA I, CA II, CA III, CA VII and CA XIII), four are
membrane bound (CA IV, CA IX, CA XII, and CA XIV), two are mitochondrial (CA VA and
CA VB) and CA VI is secreted[4–6].
Numerous diseases are associated with the changes in expression and activity of CAs resulting in
dis-balance of the inter-conversion between carbonic dioxide and bicarbonate causing pH
alteration, disorder of ion transport, fluid secretion, and other processes. Application of selective
CA inhibitors to regulate catalytic activity resulted in numerous approved drugs and still
proposes therapeutic perspectives for numerous ailments.
A large number of different CA inhibitors have been synthesized to date, and used or
investigated for treatment of diseases such as epilepsy[7], glaucoma[8], obesity[9,10], and
sterility[11], to mention a few. Furthermore, inhibitors of CA IX and CA XII are thought to have
potential to be developed as anticancer drugs [5,12]. However, most clinically used CA
inhibiting drugs are insufficiently selective and inhibit not only the target CA isoforms, but also
others causing undesired side effects. Therefore, it is important to build and understand how to
build inhibitors that would selectively inhibit one or more target CAs.

ACCEPTED MANUSCRIPT
The X-ray crystallographic structures have been solved of all isoforms except CA VA and CA
VB. The binding pocket of all CA isoforms is highly similar. It contains the Zn(II) cation held by
three conserved His residues. The fourth coordinated ligand of the Zn(II), a hydroxide anion or a
water molecule depending on pH, catalyzes the hydration reaction of CO2. The mechanism and
the structure of the catalytic site is highly similar among isoforms, making it difficult to design
isoform-selective inhibitors [13]. However, some amino acids are different among isoforms that
make it possible to rationally design highly selective inhibitors.
Despite the numerous recently proposed novel classes of CAs inhibitors, among them coumarins,
phenols, diols, polyamines, carboxylic acids, borols, boronic acids, and other structurally related
compounds, the classical primary sulfonamides have important advantages. First, the
sulfonamide group binds in the active site of CAs in a highly conserved manner making it
possible to predict the tendency of the binding of the remaining part of the sulfonamide-
containing molecule[1,14]. Second, the aromatic sulfonamides usually demonstrate higher
binding affinity toward CAs than other inhibitors that bind to the metal ion e.g. many inorganic
anions, dithio-/monothiocarbamates and xanthates, carboxylates and hydroxamates, boronic
acids and borols, diols, etc. [15,16].
Many aromatic sulfonamide-based CA inhibitors are designed using the "ring" and "tail"
approach [1,17]. The "ring" represents hetero/aromatic ring bearing sulfonamide group, which
binds to the zinc ion, and the electron acceptor substituent, which increases the acidity of the
sulfonamide group and subsequently the binding affinity. The "tail" is a flexible fragment that
participates in additional interaction with the protein parts located further from the zinc ion and
may also be designed to improve aqueous solubility.
In some CA isoforms, the amino acids in the active site cavity are arranged so that there are
hydrophobic and hydrophilic sides situated against each other. The tail interaction with the
Phe131 in CA II, located in the secondary hydrophobic binding site enhances the binding to CA
[18,19]. The hydrophilic wall which is situated opposite to the Phe131 [20] is useful as a target
for inhibitors bearing a hydrophilic tail [21]. A number of compounds have been designed as CA
inhibitors based on this approach. [13,22–24].
A number of compounds bearing two tails attached to the benzene ring have been designed.
William Vernier et al [25] attempted to make good CA II and CA IV inhibitors by balancing
between lipophilicity, which improved the affinity and hydrophilicity, which improved solubility
in water. Congiu et al [26] have synthesized N,N - disubstituted sulfanilamide derivatives which
were selective for CA IX/XII (over CA I/II). Tanpure et al [27] have introduced new approach to
employ hydrophobic and hydrophilic halves simultaneously. Authors modified CA inhibitor
acetazolamide by attaching two tails to the same amino substituent. This approach resulted in
compounds with good CA II/ CA I selectivity. Vaškevičienė et al [28] applied this dual-tail
approach to design para-aminobenzenesulfonamides bearing two distinct tails at N atom, beta-
alanine derivatives and thiazole derivatives. Zhuang Hou et al [29] have proposed an interesting
“two in one” improvement by designing benzenesulfonamide bearing an amino glucosamine

ACCEPTED MANUSCRIPT
fragment as the hydrophilic part with directly attached a cinnamamide fragment as the
hydrophobic part. These compounds showed 1000-fold increased inhibitory affinity against CA
II and CA IX as compared to initial benzenesulfonamide compound. Related 2-aryl-imidazoline
compounds have been synthesized [30], but authors have expressed doubts about participation of
polar 2-imidazoline in the interaction with the protein target.
The work of Dudutienė et al[31] revealed that the introduction of second tail into para-
substituted fluorinated benzenesulfonamides enhanced the binding to CA IX and significantly
improved the selectivity to CA IX. It can be explained by steric clashes in CA II - the deeper CA
IX cavity is more suitable for massive cyclooctyl substituent than a narrow CA II pocket with
bulky Phe131, based on comparison of X-ray crystallographic structures of compound
complexes with CA II and chimeric CA IX.
Here we extend an idea to make inhibitors that have their sulfonamide-bearing ring fixed in
position by the 2-Cl group [32–35]. Such compounds, bearing two tails, could interact with the
sides of CA active site, one being predominantly hydrophobic and another – hydrophilic. This
approach could help design inhibitors that interact strongly with the target CAs and weakly –
with non-target CAs. For the synthesis, the 2-chloro/bromo-benzenesulfonamide scaffold was
selected due to the 2-Cl group’s ability to restrain the benzene ring of the inhibitor molecule in
one known position. One of the tails is made more hydrophobic (alkyl-, aryl- substituents
attached through an amino group) while another as more hydrophilic (substituents containing
hydroxyl, carbonyl, carboxyl, ether groups, and hydrophilic amide linkage).
2. Results and discussion
2.1. Chemistry
The synthesis of designed compounds was achieved using key intermediates 2,4-dihalo-5-
sulfamoylbenzoic acids 1 and 2. Dichloro acid 1 is commercially available, while the synthesis
of dibromo acid 2 is shown in Scheme 1.
Starting from commercially available 3-methylaniline (3) the brominated product 4 was obtained
as reported previously [36], subsequently transformed into sulfonamide 5 using the same
reaction conditions described in [37]. Subsequent oxidation of methyl-group according to the
procedure [38] yielded the acid 2.
The synthesis of dihalosulfamoylbenzamides 6-17 is reported in Scheme 2. The heating of acid 1
and 2 with thionyl chloride at reflux afforded the corresponding acid chlorides, which were
amidated with the appropriate amine using as the base the excess of amine or triethylamine.
The isolation of propanolamide 8 by concentrating ethyl acetate extract was accompanied by the
formation of appreciable amounts of O-acetylated derivative 17. In addition, the compound 17
was synthesized in high yield boiling the compound 8 in ethyl acetate with an acidic catalyst

ACCEPTED MANUSCRIPT
[39]. Interestingly, in case of ethanolamide 7, the purification formed only traces of O-acetylated
compound.
Dichlorobenzamides 6-10, 14 were subsequently subjected to direct amination at 120 °C without
solvent using excess of amine according to the procedure previously reported in [40]. Under the
same reaction conditions, a large amount of byproducts resulted in dibromobenzamides 15 and
16. The formation of byproducts was avoided by performing the reaction in boiling dioxane for 7
days.
Benzoic acid 1c was synthesized from the corresponding benzamide 6c. In order to avoid chloro
substitution, the hydrolysis was performed by boiling in an acidic medium. However, hydrolysis
of ester 11 under these conditions led to the formation of two products (18 and 1) as observed by
TLC, perhaps due to simultaneous hydrolysis of the amide.
2.2. Compound binding to CAs
The synthesized compounds were tested as binders and inhibitors of the twelve catalytically
active human CA isoforms using the fluorescent thermal shift assay (FTSA) and isothermal
titration calorimetry (ITC) (Fig. 1). Since all compounds are primary sulfonamides expected to
bind to the Zn(II) and were demonstrated to interact with the CAs with a stoichiometry of 1:1, it
seemed unnecessary to perform the enzymatic inhibition assay for all compounds. The observed
affinities by FTSA and ITC were in good agreement (Fig. 2). However, there was some
discrepancy reaching almost 10 fold for several compounds. Since FTSA has fewer limitations,
than ITC, the binding constants determined calorimetrically by ITC were not used in the
analysis. The primary limitation of ITC is that it does not accurately determine the affinities
greater than approximately 10 nM. There were several cases when compounds bound CAs with
Kd_obs < 10 nM and the Wiseman c factor was higher than 500, too high for accurate
determination of the affinity. To avoid the discrepancies only FTSA affinities were used for the
thermodynamic analysis. However, there were cases when FTSA yielded unreliable melting
transitions and therefore ITC affinities were used. For example, compound 7a interaction to CA
XII, 8a to CA XIII, and 10a to CA XII and CA XIII could not be determined by FTSA and thus
values obtained by ITC were used in the analysis. The observed affinities of all compounds to
all tested CAs are listed in Table S1.
Sulfonamide – CA binding reaction is linked to several protonation events. First, the sulfonamide
amino group must deprotonate and become negatively charged and second, the Zn(II)-bound
hydroxide must bind proton and become an electrostatically neutral water molecule in order to be
replaced by the sulfonamide. These linked reactions occur simultaneously with the binding
reaction and are hard to observe experimentally. However, they influence the binding energetics
and significantly alter the binding affinities. Therefore, it is important to distinguish the observed
binding affinities that include the linked reactions and the intrinsic affinities that are calculated
by subtracting the linked reactions from the observed parameters.

ACCEPTED MANUSCRIPT
This proton linkage can significantly influence the binding affinity and mislead when comparing
compounds or CA isoforms with different pKas. In such cases, the structure-thermodynamics
correlations may lead to incorrectly assigned reasons for the change in affinity. Therefore, when
compounds with desired affinities are being designed, it is important to show the intrinsic
parameters. They are listed in Table 1. However, despite the advantage in compound design to
have the intrinsic parameters, it is also handy to compare the observed parameters, listed in Table
S1.
The pKas of compound sulfonamide group were determined spectrophotometrically as described
in the methods section and are listed in Table 2. In addition, the protonation enthalpies were also
determined and used to calculate the intrinsic enthalpies of binding. The pKas and protonation
enthalpies of CA hydroxide group were published in previously.
Single-tailed benzenesulfonamides 7-14, 17-18 have similar pKa values, equal to 8.78±0.04. The
pKas of two-tailed benzenesulfonamides 7-8(a, b) differed due to the substituents in para-
position: benzylamino substituted compounds 7b, 8b have the highest pKa values (10.1 – 10.2),
while pKas of cyclohexylamino substituted benzensulfonamides 7a, 8a are equal to 9.7 – 9.8. An
example of pKa determination by spectrophotometric method is shown in Figure 3.
However, the pKas of cyclooctylamino substituted compounds 6c, 1c, and 9c were not
determined due to low solubility of the compounds. For the calculation of intrinsic parameters
we used the same pKas as for cyclohexylamino-substituted benzenesulfonamides (pKa of
compound 6c and 9c were equal to 9.7 and pKa of 1c equal to 10.4).
The enthalpies of sulfonamide amino group protonation were determined by ITC titration of
alkaline inhibitor solution with the nitric acid (HNO3) (Figure 3, Table 2). The enthalpy of the
first transition of the titration curve represents the enthalpy of the reaction between H+ and
excess OH-, while the enthalpy of the second stage represents the enthalpy of sulfonamide
protonation. The protonation enthalpies of two-tailed compounds 7-8(a, b) were in the range
from -36.8 kJ/mol to -39.7 kJ/mol, while ∆Hprot of single-tailed compounds 7-14, 17-18 were in
the range from -27.6 kJ/mol to -31.4 kJ/mol.
The structure-thermodynamics maps of compound intrinsic affinities to 12 CA isoforms are
shown in Figures 4, 5, 6, and 7. Values next to structures show the intrinsic Gibbs energies of
binding to every CA isoform and the values next to arrows show the differences between the
Gibbs energies of binding for compounds linked by the arrow. The difference shows the gain or
loss in intrinsic affinity upon the change of the chemical structure of the compound. The error
margin of the determinations is approximately 1 kJ/mol. The Kd value of 1 nM corresponds to
Gibbs energy of binding of -53.4 kJ/mol. The 10-fold increase of binding affnity corresponds to
the difference in Gibbs energy ∆∆G = -5.93 kJ/mol.
Compound 1 boundd to CAs with nanomolar affinity. After the change of carboxy group to
amido moiety (1‰6) the binding affinity remained essentially the same to all CAs. The
lengthening of the meta tail (clockwise 6‰7, 10, 18, 12, 13, and 14) was analyzed. The aliphatic

ACCEPTED MANUSCRIPT
tails of compounds 7, 10, and 18 interacting with CA I, CA VA, and CA VB diminished the
binding affinity compared to 6. The binding of 12 to CA VA and CA VB became much weaker,
while 13 showed weaker binding affinity to CA III and CA XII. The greatest increase was found
for 10 binding to CA IV (∆G = -61.1 kJ/mol and the difference of Gibbs energies between 6 and
10 interaction with CA IV was ∆∆G = -8.1 kJ/mol). Compound 14 showed the increase in
binding affinity to all CAs compared to 6, except CA I and CA XII, where Gibbs energies
remained the same, and the greatest influence was found for CA IV and CA VB (∆∆G = -6.4
kJ/mol and -6.3 kJ/mol respectively) (Fig. 4). It was surprising that minor structural change, an
addition of a methyl group far away from the sulfonamide group (7 vs 10 to CA IV and 18 vs 11
to CA VB), significantly improved the binding affinities.
Comparing chloro 1, 6-7, 9-14, 17-18 and bromo 2, 15-16 substituted benzenesulfonamides
shows a minor influence to the binding with all CAs (Fig. 4). For example, the difference
between dissociation constants of compounds 1 and 2 is 1.3 to 3.5 times, while for 7 and 15 the
maximum difference between Kds is 2.6 times, except for their interaction with CA IV, where the
difference is 5.8 fold. The difference between binding parameters for compounds 9 and 16 varies
from 0.1 times (for CA VII) to 6.6 (for CA VB) times. These differences are small and show that
Cl and Br substitutes have a similar effect on the binding affinity. However, since 16a and 16b
exhibit low aqueous solubility, their binding affinity to CAs could not be determined accurately,
and therefore the comparison with chlorinated analogs 9a and 9b was not done.
Overall, single-tailed compounds (1-2, 6-7, 9-18) are more effective CA VII inhibitors than other
11 CA isoforms. Compounds 7 and 15 are selective CA VII inhibitors showing nanomolar
binding affinity (Kd = 0.06nM, ∆bG = -60.5 kJ/mol).
To understand the energetic contributions of small functional groups, compounds 7, 8, and 9
which differ in a single methylene group and a presence oh hydroxy group, were put in a triangle
(Figure 5). Consider their interaction with CA IV. The addition of a methylene group increased
the affinity for CA IV by nearly 100-fold, making the compound 8 slightly selective for this
isoform. Such significant increase cannot be explained by simple hydrophobic effect and most
likely involves a hydrogen bond that can be formed for compound 8 but cannot be formed for 7.
This is also confirmed by the fact that the removal of hydroxy group in 9 caused the affinity for
CA IV to drop by over 10-fold. These energetic contributions are in line with the previous
observations for aliphatic compounds [41–43].
The addition of cyclohexylamino tail in para position (Figure 6) highly increases the binding
affinity to CA I, CA IX, CA XII, CA XIII, and CA XIV, except for 9, where addition of a tail
decreased binding affinity to most CAs (except CA I, CA IX, and CA XIV, where affinity
increased). This could be explained by the presence of the hydrophobic tail in the meta position.
Longer benzyl group resulted in greater affinity for almost all CAs comparing with analogs
bearing cyclohexyl group in para position. However, benzyl group diminished compound
selectivity for CA IX and CA XII. For example, the Kd of single tailed 8 binding to CA II was

ACCEPTED MANUSCRIPT
0.07 nM, but the compound is weaker binder to CA IX and CA XII. The addition of cyclohexyl
group diminished the binding affinity to CA II two times, but highly increased the affinity to CA
IX and CA XII making compound 8a selective to these two isofoms. Compound 8b bearing
benzyl group bound CA II and CA IX with the same affinity, decreased the affinity for CA XII
and diminished the selectivity to cancer related CA IX and CA XII. The same tendency was
observed comparing single tailed compounds 7, 9, and 10 with their analogs bearing cyclohexyl
group (7a, 9a, and 10a) as well as benzyl group (7b, 9b, and 10b).
The influence of the changes of meta tail in double-tailed compounds, bearing cyclooctylamino
group in para position, are shown in Figure 7. Compound 1c showed greater binding affinity to
CA I, CA II, CA III, CA VB, CA VI, and CA VII than 6c. Gibbs energies of compound 6c
binding to other CA isoforms could not be determined using FTSA due to low solubility.
The analysis of studied compounds show that two-tailed compounds could be designed as more
potent binders compared with analogs bearing only a single tail in the meta position and
modifications of tails could contribute to the selectivity. The analyzed two-tailed compounds
showed the strongest affinities to CA IX and CA XIV.
In addition to the Gibbs energies (affinities), we studied the enthalpies and entropies of binding.
The observed enthalpies were determined by ITC and then the intrinsic enthalpies were
calculated as explained in the methods part. These intrinsic thermodynamic parameters of
binding are listed in Table 2 and Figure 8, where the ∆G is shown in bold, enthalpy ∆H in
normal font, and the entropy -T∆S is shown in italic. As in previous structural maps, the energies
of binding are shown near the structures and differences of these values between compounds
connected by the arrow are shown on the arrow.
Compounds exhibited enthalpy-driven binding to CA XII (with the partial exception) and CA
XIII, while contribution arising from both enthalpic and entropic reasons was for CA II and CA
VII. Interaction of CA II with the analyzed compounds showed large contributions of both
enthalpy and entropy. The comparison of single-tailed compound 7 with structurally similar 8
showed that ∆Hs were very similar, equal to -32.7 kJ/mol and -30.4 kJ/mol, respectively.
However, -T∆S of compound bearing longer and more flexible tail in meta position 8 was -29.9
kJ/mol, while -T∆S of 7 was -22.2 kJ/mol and due to this greater entropic contribution 8 bound to
CA II with higher affinity.
It was interesting to compare how enthalpic and entropic contribution to the Gibbs energy
changed upon adding and modifying the tails in para position. The addition and modification of
a tail to 7 showed that 7a had the greater entropic than enthalpic influence to ∆Gb_intr while
compound 7b had greater enthalpic than entropic contribution. However, the longer meta tail had
minor influence to thermodynamic parameters when the substituent to para position was added.
Moreover, the addition of cyclohexylamino moiety (7a and 7b) in para-position to compounds 7
and 8 induced similar changes of binding thermodynamics: 1) the unfavorable changes of
enthalpic and favorable changes of entropic gain upon binding to CA II and CA VII, and 2)
favorable changes of enthalpic and entropic gain upon binding to CA XII and CA XIII.

ACCEPTED MANUSCRIPT
Binding of all studied single-tailed chloro-substituted benzenesulfonamides 7-14, 17-18 to CA
VII was entropy driven, except 10, where the contribution arising from enthalpic and entropic
reasons was equal, and 7, where enthalpic contribution was greater than entropic. However,
enthalpic influence became smaller when the second tail in para position was added to
compound 7.
Comparing two-tailed compounds 7b vs 8b, it was clear that longer meta tail by one carbon atom
made protein-ligand interaction entropically driven. However, this tail had minor influence to
thermodynamic parameters of binding to all tested CAs when benzylamino group was added in
para position.
CAs II and VII are structurally close [44], showing also in the tendencies in thermodynamic
parameters that can be seen after adding para tail to compounds 7 and 8. Thermodynamic
parameters of compound binding to CA II change in the same manner as to CA VII.
Modification of 7a to 7b increased the binding affinity by -7.3 kJ/mol to CA II and -9.9 kJ/mol
to CA VII. The ∆∆Hb_intrs of CA II and CA VII binding were -13.3 kJ/mol and -20.2 kJ/mol,
respectively. The –T∆∆Sb_intr were similar as well (6.0 kJ/mol for CA II and 10.3 kJ/mol for CA
VII). Similar differences in thermodynamic parameters were found comparing the changes of
compounds 7b vs 8b and 8a vs 8b binding with CA II and CA VII.
The binding of CA XII to all meta-substituted dichlorobenzenesulfonamides (7-14, 17-18) was
enthalpy driven, except 17, where the contribution arising from enthalpic and entropic reasons
was almost equal. Compounds 11 and 18, bearing quite similar tails, had dominant enthalpic
contribution to the binding affinity. Binding of these compounds and compound 7 to CA XII was
the most enthalpy driven compound when comparing with other studied single-tailed
benzenesulfonamides. The addition of a second tail to the para position increased the binding
affinity to CA VII. The best CA XII binder 7a (∆Gb_intr = -68.3 kJ/mol) had dominant enthalpic
(∆Hb_intr = -53.1 kJ/mol) contribution.
Interaction of CA XIII with single-tailed chlorinated compounds (7-14, 17-18) was enthalpy
driven. Binding of 11 and 18 to CA XIII had an unfavorable entropic contribution. It was
interesting that changes of binding thermodynamics demonstrated distinct effect of enthalpy-
entropy compensation in two pairs of compounds (9 vs 11 and 9 vs 18) in binding to CA XIII.
After the addition of a tail in para position (compounds 7a,b and 8a,b) binding to CA XIII
became stronger and entropic contribution slightly increased. Compound 7b was the strongest
binder of CA XIII compared with all studied benzenesulfonamides and had the greater enthalpic
contribution than other two-tailed compounds.
Figure 9 shows the enthalpy-entropy compensation graph for all studied compound binding to
four CA isoforms. Squares represent the two-tailed benzenesulfonamides and triangles represent
the benzenesulfonamides bearing a single tail in meta position. Different colors show different
CA isoforms. Binding affinities span a relatively narrow range, while the range of enthalpies and
entropies are significantly larger. The enthalpies of single-tailed compound interaction with CA
II varied from -12.8 kJ/mol to -37.5 kJ/mol, while the range of entropies was between -16.4

ACCEPTED MANUSCRIPT
kJ/mol and -39.5 kJ/mol. The range of enthalpies for CA VII was between -8.9 kJ/mol and -37.6
kJ/mol, entropies from -23.0 kJ/mol to -49.0 kJ/mol. The enthalpies for CA XII spanned between
-24.5 kJ/mol and -48.2 kJ/mol, entropies from -6.2 kJ/mol and -26.7 kJ/mol. The range of
enthalpies of single-tailed compounds interacting with CA XIII varied from -32.5 kJ/mol to -59.0
kJ/mol, while entropies span between 11 kJ/mol and -16.2 kJ/mol. The enthalpies of two-tailed
compound interaction with CA II varied from -22.3 kJ/mol to -39.3 kJ/mol, while the range of
entropies was between -27.5 kJ/mol and -34.7 kJ/mol. The range of enthalpies for CA VII was
between -16.8 kJ/mol and -35.1 kJ/mol, for entropies from -25.8 kJ/mol to -37 kJ/mol. The
enthalpies for CA XII spanned between -33.6 kJ/mol and -53.5 kJ/mol, entropies from -8.2
kJ/mol to -31.1 kJ/mol. The range of enthalpies of single-tailed compound interaction with CA
XIII varied from -38.4 kJ/mol to -49.7 kJ/mol, while entropies spanned between -5.7 kJ/mol and
-16.2 kJ/mol. Thermodynamic properties of CA II and CA VII binding was very similar-
enthalpies and entropies spanned equally and thus black and red squares overlaped in the figure.
Equal contribution arising from enthalpic and entropic reasons to CA XII binding affinity also
overlaped red and black datapoints, however part of CA XII datapoints were shifted due to
enthalpically driven interaction. Blue datapoints were the most apart, because CA XIII binding
with these compounds was exclusively enthalpy driven with two compounds exhibiting
unfavorable binding entropy.
2.3. X-Ray crystallographic structures of CA-inhibitor complexes
In search of correlations between the binding thermodynamics and 3D structures of the bound
complexes, four crystal structures of human CAs containing compounds in the active site were
solved by X-Ray crystallography, compound 9 bound to CA II and CA XIII, and CA XII with
bound 7a and 7b. All complexes were obtained by soaking of CA crystals with solution of the
corresponding compound in the crystallization buffer. The crystal structure CA II-9 contains one
protein chain in the asymmetric unit, CA XIII-9 – two, and the crystal structures of CA XII –
four protein chains. Data collection and the refinement statistics of four datasets are presented in
the Table 3. The electron densities of ligands in the crystal structures are shown in Figure 10.
The compound 9 binds to CA II 39 times more strongly than to CA XIII (Kd 0.16 (CA II) vs 6.3
(CA XIII) nM). In the crystal structures, the compound 9 binds to CA II and CA XIII in the
nearly identical positions (Figure 11AB). The chlorine atom interacts in both crystal structures
with the residues that are conservative between two isoforms. The chlorobenzene ring is
anchored by the chlorine in the position that was described previously [33,34,45,46]. The deeper
parts of active sites of CA II and CA XIII are very similar and differ by only three amino acids:
Ile91, Asn62 and Thr200 in CA II are replaced in CA XIII by Arg93, Ser64 and Val202,
respectively. Thus, in CA II the hydroxyl group of Thr200 makes the hydrogen bond with the
nitrogen atom of the meta-substituent of 9. Such binding mode is impossible in CA XIII, since
the threonine side chain is replaced there by the valine. Details of interaction of meta-substituent
with CA XIII and CA II are depicted in Figure 11B. In particular, Thr200 (CA II) makes a direct

ACCEPTED MANUSCRIPT
hydrogen bond with meta-substituent of ligand, whereas in CA XIII there is a water-mediated
interaction. These differences in interaction of 9 with CA II and CA XIII could explain better
binding of compound 9 to CA II.
The binding thermodynamics of compound 9 to CA II and CA XIII exhibit changes of both
thermodynamic parameters that are favorable over CA II:
(CA II): ∆G = -58.2, ∆H = -35.2, -T∆S = -23.0 kJ/mol;
(CA XIII): ∆G = -48.7, ∆H = -32.5, -T∆S = -16.2 kJ/mol;
(CA II vs CA XIII): ∆∆G = -9.5, ∆∆H = -2.7, ∆(T∆S) = -6.8 kJ/mol.
Favorable changes in binding enthalpy could be associated with the direct hydrogen bond
between Thr202 and ligand in CA II-9 complex. The favorable changes in entropy are more
difficult to explain.
Compound 7a and 7b binding mode to CA XII is presented in Figure 11CD. Both ligands are
located in the active site of CA XII in a similar way and demonstrate very similar binding
affinities (Kd of 7a 0.019 nM vs 0.041 nM of 7b). The main differences in the binding modes of
7a and 7b were detected in the positions of the tail of the meta-substituent. In crystal structure of
the complex CA XII-7b the tail of the meta-substituent in all protein subunits was found in the
same well-defined conformation in the electron density. The tail of meta-substituent of 7b is
located deeper in the active site cavity than 7a.The tail of meta-substituent of the compound 7a
in two protein chains of CA XII was found in two alternate conformations. One of the alternate
conformations is similar to the position of meta-tail of the 7b in the complex CA XII – 7b. Other
four conformations are directed towards the entrance of the active site cavity. The slight
displacement in the position of the benzene ring of 7a compared to 7b is caused by the larger
para-substituent of 7b (Figure 11D).
The binding affinities of compounds 7a and 7b demonstrate that compound 7a binds to CA XII
~20 times stronger than to CA II (0.019 nM for CA XII and 0.37 nM for CA II), whereas the
binding affinities of compound 7b to CA II and CA XII are similar (0.022 nM vs 0.041 nM,
Table 1). Interestingly, similar behavior of binding affinities is observed for compounds 10a and
8a which contain the cyclohexylamino substituent in the para-position. In this respect these
ligands are like compound 7a and bind to CA XII significantly stronger than to CA II (10a – 21
times, 8a – 11, Table 1). Compound 7b could be compared with compounds 10b, 8b, and 9b
which also contain the benzylamino substituent in the para-position and have similar binding
affinities to CA XII and CA II, although all compounds show slightly better binding affinities to
CA XII: 10b – 4 times, 7b – 2 times, 8b – 5 times, and 9b – 2 times, Table 1.
The binding of 7a and 7b in CA II and CA XII cannot be directly compared due to the absence
of crystal structures of complexes CA II-7a and CA II-7b. However, it is known from the
crystallographic studies of chlorinated benzensulfonamides, that the binding mode of chlorinated
ring in the active sites of CA II, CA XII, and CA XIII are the same due to conserved residues in

ACCEPTED MANUSCRIPT
these CA isoforms. The subunits from crystal structures of CA XII-7a and CA II-9 can be
superimposed to analyze the possible binding mode of 7a in CA II active site. If the binding
mode of the ligand in CA XII would be retained in CA II, there could be a steric clash between
the cyclohexylamino group of 7a, which is taken from the active site of CA XII, and the side
chain of Phe131 in CA II (Fig. 11E, pink arrow). Therefore, the positions of 7a and 7b in CA II
should differ from those in CA XII. The benzylamino substituent is more flexible and longer
than cyclohexylamino and possibly could adopt a different conformation in the binding to
hydrophobic cavity (CA II) near the Phe131 (blue arrow). We speculate that the longer and more
flexible substituent could escape the collisions with Phe131 in CA II and adopt an optimal
conformation for interaction with the hydrophobic part of active site. Higher conformational
adaptability of this fragment possibly improves the binding affinities to CA II, but seems to have
opposite effect on the selectivity of binding.
3. Conclusions
A series of mono- and di-substituted 2-chloro/bromo-benzenesulfonamides were synthesized and
their binding analysis to all 12 catalytically active human CAs was performed to determine
intrinsic affinities, enthalpies, and entropies of binding. The length and hydrophobicity of the
meta-substituent affected the binding affinity toward each CAs. Single tailed compounds showed
nanomolar affinity and selectivity towards CA VII. However, the introduction of a second tail
had significant increase to the binding affinity and two-tailed compounds in most cases provided
selectivity for CA IX and CA XIV as was designed. This selectivity depended on the length of
the substituent in para position, where longer benzyl group becomes more flexible and suitable
for the active sites of all CA isoforms, making benzylamino substituted dual-tailed sulfonamides
high affinity, but not so selective CA inhibitors as cyclohexylamino substituted analogs. X-ray
crystallographic structures of several compound-CA complexes showed the contacts between the
tails and amino acids providing insights into structure-thermodynamics correlations.
4. Experimental
4.1. Chemistry
All starting materials and reagents were commercial products and were used without further
purification. Melting points of the compounds were determined in open capillaries on a Thermo
1
Scientific 9100 Series and are uncorrected. H and 13C NMR spectra were recorded on a (400
and 100 MHz, respectively) spectrometer in DMSO-d6 using residual DMSO signals (2.52 ppm
1
and 40.21 ppm for H and 13C NMR spectra, respectively) as the internal standard. TLC was
performed with silica gel 60 F254 aluminum plates (Merck) and visualized with UV light.
Column chromatography was performed using silica gel 60 (0.040-0.063 mm, Merck). High-

ACCEPTED MANUSCRIPT
resolution mass spectra (HRMS) were recorded on a Dual-ESI Q-TOF 6520 mass spectrometer
(Agilent Technologies). The purity of final compounds was verified by HPLC to be >95% using
the Agilent 1290 Infinity instrument with a Poroshell 120 SB-C18 (2.1 mm × 100 mm, 2.7 µm)
reversed-phase column. Analytes were eluted using a linear gradient of water/methanol (20 mM
ammonium formate in both phases) from 60:40 to 30:70 over 12 min, then from 30:70 to 20:80
over 1 min, and then 20:80 over 5 min at a flow rate of 0.2 mL/min. UV detection was at 254
nm.
4.1.1. 2,4-dibromo-5-methylaniline hydrochloride (4). N-(3-tolyl)acetamide was synthesized by
the known procedure[47]. Bromine (4.65 mL, 90 mmol) was added dropwise to the solution of
N-(3-tolyl)acetamide (6.12 g, 41.0 mmol) in two portions. The first part was added at room
temperature, the other at 60-70 °C. The reaction mixture was stirred for 15 hours at 70° C,
cooled and washed with 10% Na2SO3. After removal of the solvent, the residue was crystallized
from H2O:MeOH (1:1) to obtain N-(2,4-dibromo-5-methyl-phenyl)acetamide. Yield 11.6g, 92%,
mp 168-169 °C(lit. 170-171 °C[48]).
N-(2,4-dibromo-5-methyl-phenyl)acetamide (11.6 g, 37.8 mmol) was refluxed in methanol (12
mL) and concentrated HCl (aq) (3 mL) solution for 10 hours. After removal of the solvents, the
residue was separated by filtration and washed with ice-cold water. Yield: 7.19 g, 63%, mp 165-
+
167 °C (dec.). 1H NMR δ ppm: 2.20 (3H, s, CH3), 6.81 (3H, br s, NH3 ), 6.91 (1H, s, C6-H), 7.57
(1H, s, C3-H). 13C NMR δ ppm: 22.5, 106.9, 111.7, 118.9, 134.7, 137.6, 143.7.
4.1.2. 2,4-dibromo-5-methyl-benzenesulfonamide (5). A solution of NaNO2 (2.77 g, 40.1 mmol)
in water (7 mL) was added dropwise to a suspension of 4 (7.11 g, 22.8 mmol) in a mixture of
acetic acid (35 mL) and conc. HCl(aq) (14 mL) at 0-5 °C. The reaction mixture was stirred at 5
°C for 1h, then a cold mixture of sulfur dioxide (8.06 g, 126 mmol) and cooper (II) chloride
dihydrate(1.34 g, 7.86 mmol) in acetic acid (20 mL) at -15 °C was rapidly added. The reaction
mixture was stirred for 24h at 0 °C, and then the reaction mixture was diluted with water (40
mL) and extracted with toluene (3x30 mL). The extracts were washed with water and dried over
anhydrous MgSO4. After removal of the solvent, the oily residue was dissolved in 30 ml of THF
and conc. ammonia (53 mL) was added dropwise at 0-5 °C. The mixture was allowed to stand
24h at room temperature. The precipated solid was separated by filtration and purified by
1
recrystallization from H2O:MeOH (1:1). Yield: 5.98 g, 77%, mp 219-221 °C. H NMR δ ppm:
2.38 (3H, s, CH3), 7.64 (2H, s, SO2NH2), 7.96 (1H, s, C6-H), 8.08 (1H, s, C3-H). 13C NMR δ
ppm: 22.4, 116.9, 128.6, 131.2, 137.6, 138.2, 142.4. HRMS calcd. for C7H7Br2NO2S [(M+H)+]:
329.8616 (100%), found: 329.8619 (100%).
4.1.3. 2,4-dibromo-5-sulfamoyl-benzoic acid (2). The solution of KMnO4 (4.74 g, 30.0 mmol) in
water (80 mL) was added dropwise to the suspension of 5 (2.21 g, 6.72 mmol) in water (60 mL)
over a period of 5 hours at 95 °C. The reaction mixture was stirred for 12 hours at 95 °C, cooled
and filtered to remove the MnO2. The excess of KMnO4 is removed by treatment with oxalic
acid and the solution was acidified with hydrochloric acid to pH 1. The resulting precipitate was

ACCEPTED MANUSCRIPT
filtered off, washed with water and crystallized from H2O:MeOH (4:1). Yield: 1.57 g, 65%, mp
237-239 °C.
1H NMR δ ppm: 7.83 (2H, s, SO2NH2), 8.27 (1H, s, C6-H), 8.33 (1H, s, C3-H), 13,96 (1H, br s,
COOH). 13C NMR δ ppm: 122.9, 125.0, 131.3, 133.0, 140.0, 142.7, 166.1. HRMS calcd. for
C7H5Br2NO4S [(M+H)+]: 359.8358 (100%), found: 359.8360 (100%).
4.1.4. General procedure for the syntheses of 6-16. The mixture of 2,4-dihalo-5-
sulfamoylbenzoic acid (1 or 2) (10.0 mmol), SOCl2 (2-3 eq), and 1 drop DMF in toluene (5 mL)
was refluxed for 4 h. Excess SOCl2 and toluene were removed by distillation under reduced
pressure, and the crude acid chloride was used directly in the next step.
The solution of 2,4-dihalo-5-sulfamoylbenzoyl chloride in THF (30 mL) was added dropwise to
a solution of appropriate amine (30.0 mmol) in THF (20 mL) at 0 °C and allowed stirring for 1h
(for compound 11 was used methyl 4-aminobutanoate hydrochloride (1.3 eq), triethylamine (2
eq)). The mixture was warmed to room temperature and stirred for another 1-2 h. THF was
removed under reduced pressure. Water was added to the residue and product was extracted with
EtOAc. The organic layer was washed with 5% HCl(aq), dried over anhydrous MgSO4, filtered
and concentrated.
4.1.4.1. 2,4-dichloro-5-sulfamoyl-benzamide (6, EA1A-1). Recrystallization was accomplished
from H2O. Yield: 2.15 g, 80%, mp 209-210 °C [40].
4.1.4.2. 2,4-dichloro-N-(2-hydroxyethyl)-5-sulfamoyl-benzamide (7, EA3-1). Recrystallization
was accomplished from EtOAc. Yield: 1.69 g, 54%, mp 182-183° C. 1H NMR δ ppm: 3.31 (2H,
q, J = 6.0 Hz, NHCH2), 3.51 (2H, t, J = 6.0 Hz, CH2OH), 4.77 (1H, br s, OH),7.81 (2H, s,
SO2NH2), 7.93 (1H, s, C3-H), 7.95 (1H, s, C6-H), 8.67 (1H, t, J = 5.6 Hz, NH). 13C NMR δ ppm:
42.5, 60.0, 129.3, 132.1, 132.5, 134.5, 136.4, 140.3, 165.0. HRMS calcd. for C9H10Cl2N2O4S
[(M+H)+]: 312.9811, found: 312.9814.
4.1.4.3. 2,4-dichloro-N-(3-hydroxypropyl)-5-sulfamoyl-benzamide (8, EA4-1)[39].
4.1.4.4. N-butyl-2,4-dichloro-5-sulfamoyl-benzamide (9, EA5-1). Recrystallization was
accomplished from toluene:MeOH (5:1). Yield: 3.03 g, 93%, mp 184-186 °C(lit. 180 °C[40]). 1H
NMR δ ppm: 0.91 (3H, t, J = 7.2 Hz, CH3), 1.36 (2H, sext, J = 7.2 Hz, CH2), 1.50 (2H, quint, J
= 7.2 Hz, CH2), 3.24 (2H, q, J = 6.8 Hz, NHCH2), 7.81 (2H, s, SO2NH2), 7.90 (1H, s, C3-H),
7.94 (1H, s, C6-H), 8.62 (1H, t, J = 5.6 Hz, NH). 13C NMR δ ppm: 14.1, 20.0, 31.4, 39.2, 129.1,
132.1, 132.5, 134.4, 136.5, 140.4, 164.8. HRMS calcd. for C11H14Cl2N2O3S [(M+H)+]: 325.0175,
found: 325.0174.
4.1.4.5. 2,4-dichloro-N-(2-methoxyethyl)-5-sulfamoyl-benzamide (10, EA8-1). Recrystallization
1
was accomplished from toluene:MeOH (5:1). Yield: 2.13 g, 65%, mp 137-139 °C. H NMR δ
ppm: 3.29 (3H, s, CH3), 3.38-3.43 (2H, m, NHCH2), 3.45-3.48 (2H, m, OCH2), 7.80 (2H, s,
SO2NH2), 7.91 (1H, s, C3-H), 7.93 (1H, s, C6-H), 8.74 (1H, t, J = 5.6 Hz, NH). 13C NMR δ ppm:

ACCEPTED MANUSCRIPT
39.4, 58.4, 70.7, 129.2, 132.2, 132.5, 134.5, 136.3, 140.3, 165.1. HRMS calcd. for
C10H12Cl2N2O4S[(M+H)+]: 326.9968, found: 326.9967.
4.1.4.6. Methyl 4-[(2,4-dichloro-5-sulfamoyl-benzoyl)amino]butanoate (11, EA11-1). The
product was purified by chromatography on a column of silica gel with EtOAc, Rf = 0.80. Yield:
2.03 g, 55%, mp 138-140 °C. 1H NMR δ ppm: 1.77 (2H, quint, J = 7.2 Hz, CH2), 2.41 (2H, t, J
= 7.6 Hz, COCH2), 3.27 (2H, q, J = 6.8 Hz, NHCH2), 3.61 (3H, m, CH3), 7.82 (2H, s, SO2NH2),
7.92 (1H, s, C3-H), 7.95 (1H, s, C6-H), 8.69 (1H, t, J = 5.6 Hz, NH). 13C NMR δ ppm: 24.7, 31.1,
38.9, 51.8, 129.2, 132.2, 132.5, 134.4, 136.3, 140.4, 165.0, 173.5. HRMS calcd. for
C12H14Cl2N2O5S[(M+H)+]: 369.0073, found: 369.0074.
4.1.4.7.
2,4-dichloro-5-(morpholine-4-carbonyl)benzenesulfonamide
(12,
EA7-1).
1
Recrystallization was accomplished from EtOAc. Yield: 2.34 g, 69%, mp 235-236 °C. H NMR
δ ppm: 3.17-3.20 (2H, m, CH2), 3.54-3.57 (2H, m, CH2), 3.58-3.74 (4H, m, 2CH2), 7.82 (2H, s,
SO2NH2), 7.93 (1H, s, C3-H), 7.98 (1H, s, C6-H). 13C NMR δ ppm: 42.2, 47.1, 66.3, 66.6, 128.6,
132.2, 132.5, 133.8, 134.9, 141.0, 164.3. HRMS calcd. for C11H12Cl2N2O4S[(M+H)+]: 338.9968,
found: 338.9966.
4.1.4.8. 2,4-dichloro-N-cyclohexyl-5-sulfamoyl-benzamide (13, EA9-1). Recrystallization was
accomplished from H2O:MeOH (1:1) and then from toluene:MeOH (5:1). Yield: 2.21 g, 63%,
1
mp 214-215 °C. H NMR δ ppm: 1.17-1.34 (5H, m, Cy-H), 1.57-1.59 (1H, m, Cy-H), 1.71 (2H,
br s, Cy-H), 1.84-1.86 (2H, m, Cy-H), 3.73 (1H, br s, Cy-H), 7.82 (2H, s, SO2NH2), 7.88 (1H, s,
C3-H), 7.93 (1H, s, C6-H), 8.56 (1H, d, J = 7.2 Hz, NH). 13C NMR δ ppm: 24.9, 25.6, 32.5, 48.8,
129.0, 132.0, 132.4, 134.5, 136.7, 140.3, 164.0. HRMS calcd. for C13H16Cl2N2O3S[(M+H)+]:
351.0331, found: 351.0335.
4.1.4.9. N-benzyl-2,4-dichloro-5-sulfamoyl-benzamide (14, EA10-1). Recrystallization was
accomplished from H2O:MeOH (1:1) and then from toluene:MeOH (5:1). Yield: 3.05 g, 85%,
mp 179-181 °C. 1H NMR δ ppm: 4.49 (2H, d, J = 6.0 Hz, CH2), 7.26-7.33 (1H, m, Ph-H), 7.35-
7.40 (4H, m, Ph-H), 7.84 (2H, s, SO2NH2), 7.96 (1H, s, C3-H), 7.97 (1H, s, C6-H), 9.21 (1H, t, J
13
= 6.0 Hz, NH). C NMR δ ppm: 43.1, 127.5, 127.8, 128.9, 129.3, 132.3, 132.6, 134.5, 136.1,
139.2, 140.4, 165.0. HRMS calcd. for C14H12Cl2N2O3S[(M+H)+]: 359.0018, found: 359.0017.
4.1.4.10. 2,4-dibromo-N-(2-hydroxyethyl)-5-sulfamoyl-benzamide (15, LJ14-5). Recrystallization
was accomplished from MeOH. Yield: 1.41 g, 35%, mp 196-198 °C. 1H NMR δ ppm: 3.30 (2H,
q, J = 6.0 Hz, NHCH2), 3.51 (2H, br s, CH2OH), 4.77 (1H, s, OH), 7.74 (2H, s, SO2NH2), 7.91
(1H, s, C6-H), 8.19 (1H, s, C3-H), 8.64 (1H, t, J = 6.0 Hz, NH). 13C NMR δ ppm: 42.5, 59.9,
120.4, 123.6, 129.0, 138.5, 139.1, 142.5, 166.2. HRMS calcd. for C9H10Br2N2O4S[(M+H)+]:
402.8780 (100%), found: 402.8782 (100%).
4.1.4.11. 2,4-dibromo-N-butyl-5-sulfamoyl-benzamide (16, LJ14-6). Recrystallization was
accomplished from MeOH:H2O (1:1). Yield: 1.33 g, 31%, mp 218-220 °C. 1H NMR δ ppm: 0.91
(3H, t, J = 7.2 Hz, CH3), 1.36 (2H, sext, J = 7.2 Hz, CH2), 1.50 (2H, quint, J = 7.2 Hz, CH2),
3.23 (2H, q, J = 6.8 Hz, NHCH2), 7.79 (2H, s, SO2NH2), 7.86 (1H, s, C6-H), 8.20 (1H, s, C3-H),

ACCEPTED MANUSCRIPT
8.61 (1H, t, J = 5.6 Hz, NH). 13C NMR δ ppm: 14.1, 20.0, 31.4, 39.2, 120.3, 123.5, 128.8, 138.5,
139.3, 142.6, 165.9. HRMS calcd. for C12H16Br2N2O3S[(M+H)+]: 428.9301 (100%), found:
428.9297 (100%).
4.1.5. 3-[(2,4-dichloro-5-sulfamoyl-benzoyl)amino]propyl acetate (17, EA4-1-2)[39].
4.1.6. General procedure for the syntheses of (7-10)a, (7-10, 14)b, (6, 9)c. The mixture of
appropriate 2,4-dichloro-N-substituted-5-sulfamoylbenzamide (compounds 6-10, 14) (1.00
mmol) was heated in appropriate amine (6 mmol) at 120 °C for 3-4 h . The mixture was cooled
to room temperature and 2N HCl(aq) (2 mL) was added. The resultant precipitate was washed
with H2O.
4.1.6.1. 4-chloro-2-(cyclooctylamino)-5-sulfamoyl-benzamide (6c, EA1A-9). Recrystallization
was accomplished twice from H2O:2-PrOH (2:1). Yield: 216 mg, 60%, mp 242-244 °C. 1H NMR
δ ppm: 1.44-1.72 (12H, m, cyclooctyl-H), 1.75-1.86 (2H, m, cyclooctyl-H), 3.67 (1H, br s,
cyclooctyl-H), 6.75 (1H, s, C3-H), 7.17 (2H, s, SO2NH2), 7.37 (1H, brs, CONH2), 8.08 (1H, brs,
13
CONH2), 8.14 (1H, s, C6-H), 8.77 (1H, d, J = 8.0 Hz, cyclooctyl-NH). C NMR δ ppm: 23.4,
25.4, 27.3, 31.5, 51.4, 111.8, 113.4, 125.7, 131.3, 135.2, 151.4, 170.8. HRMS calcd. for
C15H22ClN3O3S [(M+H)+]: 360.1143, found: 360.1139.
4.1.6.2. 4-chloro-2-(cyclohexylamino)-N-(2-hydroxyethyl)-5-sulfamoyl-benzamide (7a, EA3-7).
The product was purified by chromatography on a column of silica gel with EtOAc, Rf = 0.65
and then the recrystallization was accomplished from toluene:2-PrOH (5:1). Yield: 90.2 mg,
24%, mp 210-212 °C. 1H NMR δ ppm: 1.20-1.30 (3H, m, Cy-H), 1.38-1.46 (2H, m, Cy-H), 1.55-
1.58 (1H, m, Cy-H), 1.64-1.68 (2H, m, Cy-H), 1.86-1.89 (2H, m, Cy-H), 3.29 (2H, q, J = 6.0 Hz,
NHCH2), 3.48-3.54 (3H, m, CH2OH, Cy-H), 4.72 (1H, t, J = 5.6 Hz, OH), 6.86 (1H, s, C3-H),
7.17 (2H, s, SO2NH2), 8.11 (1H, s, C6-H), 8.51 (1H, d, J = 8.0 Hz, CyNH), 8.56 (1H, t, J = 5.6
Hz, CONH). 13C NMR δ ppm: 24.3, 25.7, 32.4, 40.6, 49.9, 60.0, 112.4, 113.1, 125.8, 130.7,
135.0, 151.3, 168.4. HRMS calcd. for C15H22ClN3O4S [(M+H)+]: 376.1092, found: 376.1092.
4.1.6.3. 2-(benzylamino)-4-chloro-N-(2-hydroxyethyl)-5-sulfamoyl-benzamide (7b, EA3-8).
Recrystallization was accomplished three times from toluene:2-PrOH (8:1). Yield: 61.4 mg,
1
16%, mp 225-228 °C. H NMR δ ppm: 3.31 (2H, q, J = 6.0 Hz, NHCH2), 3.51 (2H, t, J = 6.0
Hz, CH2OH), 4.49 (2H, d, J = 5.6 Hz, NHCH2Ph), 4.73 (1H, br s, OH), 6.76 (1H, s, C3-H), 7.19
(2H, s, SO2NH2), 7.25-7.39 (5H, m, Ph-H), 8.12 (1H, s, C6-H), 8.62 (1H, t, J = 5.6 Hz, NHBn),
13
8.75 (1H, t, J = 5.6 Hz, CONH). C NMR δ ppm: 42.4, 46.2, 60.0, 113.4, 113.5, 126.6, 127.5,
127.6, 129.1, 130.4, 134.7, 138.9, 151.9, 168.2. HRMS calcd. for C16H18ClN3O4S [(M+H)+]:
384.0779, found: 384.0781.
4.1.6.4. 4-chloro-2-(cyclohexylamino)-N-(3-hydroxypropyl)-5-sulfamoyl-benzamide (8a, EA4-7).
Recrystallization was accomplished from: toluene:2-PrOH (1:1). Yield: 179 mg, 46%, mp 192-
194 °C. 1H NMR δ ppm: 1.20-1.30 (3H, m, Cy-H), 1.37-1.49 (2H, m, Cy-H), 1.56-1.58 (1H, m,
Cy-H), 1.63-1.68 (4H, m, Cy-H, CH2), 1.87-1.89 (2H, m, Cy-H), 3.27 (2H, q, J = 6.4 Hz,
NHCH2), 3.46 (2H, t, J = 6.4 Hz, CH2OH), 3.51 (1H, br s, Cy-H), 4.41 (1H, br s, OH), 6.85 (1H,

ACCEPTED MANUSCRIPT
s, C3-H), 7.17 (2H, s, SO2NH2), 8.09 (1H, s, C6-H), 8.49 (1H, d, J = 7.6 Hz, CyNH), 8.61 (1H, t,
J = 5.2 Hz, CONH). 13C NMR δ ppm: 24.3, 25.7, 32.4, 32.7, 39.4, 49.9, 59.1, 112.5, 113.1,
125.8, 130.6, 135.0, 151.3, 168.3. HRMS calcd. for C16H24ClN3O4S [(M+H)+]: 390.1249, found:
390.1252.
4.1.6.5. 2-(benzylamino)-4-chloro-N-(3-hydroxypropyl)-5-sulfamoyl-benzamide (8b, EA4-8). The
product was purified by chromatography on a column of silica gel with EtOAc:CHCl3 (2:1), Rf =
1
0.25. Yield: 171 mg, 43%, mp 179-182 °C. H NMR δ ppm: 1.68 (2H, quint, J = 6.8 Hz, CH2),
3.29 (2H, q, J = 6.4 Hz, NHCH2), 3.45-3.49 (2H, m, CH2OH), 4.48 (3H, d, J = 6.0 Hz,
NHCH2Ph, OH), 6.76 (1H, s, C3-H), 7.20 (2H, s, SO2NH2), 7.26-7.39 (5H, m, Ph-H), 8.10 (1H,
13
s, C6-H), 8.68 (1H, t, J = 5.6 Hz, NHBn), 8.74 (1H, t, J = 5.6 Hz, CONH). C NMR δ ppm:
32.7, 39.6, 46.2, 59.1, 113.4, 113.7, 126.6, 127.5, 127.6, 129.1, 130.3, 134.7, 138.9, 151.9, 168.0.
HRMS calcd. for C17H20ClN3O4S [(M+H)+]: 398.0936, found: 398.0936.
4.1.6.6. N-butyl-4-chloro-2-(cyclohexylamino)-5-sulfamoyl-benzamide (9a, EA5-7). The product
was purified by chromatography on a column of silica gel with CHCl3:EtOAc (5:1), Rf = 0.33.
Yield: 167 mg, 43%, mp 182-184 °C. 1H NMR δ ppm: 0.90 (3H, t, J = 7.2 Hz, CH3), 1.20-1.58
(10H, m, Cy-H, CH2CH2), 1.64-1.67 (2H, m, Cy-H), 1.87-1.89 (2H, m, Cy-H), 3.21 (2H, q, J =
6.4 Hz, NHCH2), 3.50 (1H, br s, Cy-H), 6.85 (1H, s, C3-H), 7.17 (2H, s, SO2NH2), 8.09 (1H, s,
C6-H), 8.47 (1H, d, J = 7.6 Hz, CyNH), 8.62 (1H, t, J = 5.2 Hz, CONH). 13C NMR δ ppm: 14.2,
20.1, 24.2, 25.7, 31.5, 32.4, 39.4, 49.9, 112.7, 113.0, 125.8, 130.6, 134.9, 151.3, 168.2. HRMS
calcd. for C17H26ClN3O3S [(M+H)+]: 388.1456, found: 388.1456.
4.1.6.7. 2-(benzylamino)-N-butyl-4-chloro-5-sulfamoyl-benzamide (9b, EA5-8). Recrystallization
was accomplished from toluene:2-PrOH (8:1). Yield: 91.1 mg, 23%, mp 204-206 °C. 1H NMR δ
ppm: 0.91 (3H, t, J = 7.2 Hz, CH3), 1.33 (2H, sext, J = 7.2 Hz, CH2),1.50 (2H, quint, J = 7.2 Hz,
CH2), 3.23 (2H, q, J = 6.8 Hz, CONHCH2), 4.48 (2H, d, J = 5.6 Hz, NHCH2Ph), 6.76 (1H, s, C3-
H), 7.19 (2H, s, SO2NH2), 7.25-7.39 (5H, m, Ph-H), 8.10 (1H, s, C6-H), 8.68 (1H, t, J = 5.6 Hz,
13
NHBn), 8.72 (1H, t, J = 6.0 Hz, CONH). C NMR δ ppm: 14.2, 20.1, 31.5, 39.1, 46.2, 113.4,
113.8, 126.6, 127.5, 127.6, 129.1, 130.3, 134.6, 138.9, 151.9, 167.9. HRMS calcd. for
C18H22ClN3O3S [(M+H)+]: 396.1143, found: 396.1145.
4.1.6.8. N-butyl-4-chloro-2-(cyclooctylamino)-5-sulfamoyl-benzamide (9c, EA5-9). The product
was purified by chromatography on a column of silica gel with CHCl3:EtOAc (5:1), Rf = 0.35.
Yield: 112 mg, 27%, mp 172-174 °C. 1H NMR δ ppm: 0.90 (3H, t, J = 7.2 Hz, CH3), 1.32 (2H,
sext, J = 7.2 Hz, CH2), 1.45-1.65 (14H, m, cyclooctyl-H, CH2), 1.78-1.83 (2H, m, cyclooctyl-H),
3.19-3.23 (2H, m, NHCH2), 3.66 (1H, br s, cyclooctyl-H), 6.75 (1H, s, C3-H), 7.17 (2H, s,
SO2NH2), 8.10 (1H, s, C6-H), 8.52 (1H, d, J = 7.6 Hz, cyclooctyl-NH), 8.62 (1H, br s, CONH).
13C NMR δ ppm: 14.2, 20.1, 23.4, 25.4, 27.3, 31.4, 31.5, 39.1, 51.4, 112.8, 113.3, 125.8, 130.6,
134.9, 151.1, 168.2. HRMS calcd. for C19H30ClN3O3S [(M+H)+]: 416.1769, found: 416.1770.
4.1.6.9. 4-chloro-2-(cyclohexylamino)-N-(2-methoxyethyl)-5-sulfamoyl-benzamide (10a, EA8-7).
The product was purified by chromatography on a column of silica gel with CHCl3:EtOAc (3:1),
Rf = 0.21. Yield: 129 mg, 33%, mp 214-216 °C. 1H NMR δ ppm: 1.16-1.30 (3H, m, Cy-H), 1.37-

ACCEPTED MANUSCRIPT
1.47 (2H, m, Cy-H), 1.55-1.58 (1H, m, Cy-H), 1.64-1.68 (2H, m, Cy-H), 1.86-1.92 (2H, m, Cy-
H), 3.27 (3H, m, CH3), 3.35-3.40 (2H, m, NHCH2), 3.43-3.46 (2H, m, OCH2), 3.47-3.60 (1H, m,
Cy-H), 6.87 (1H, s, C3-H), 7.19 (2H, s, SO2NH2), 8.11 (1H, s, C6-H), 8.50 (1H, d, J = 8.0 Hz,
CyNH),8.68 (1H, t, J = 5.6 Hz, CONH). 13C NMR δ ppm: 24.3, 25.7, 32.4, 39.2, 49.9, 58.4,
70.7, 112.2, 113.1, 125.8, 130.7, 135.1, 151.3, 168.4. HRMS calcd. for C16H24ClN3O4S
[(M+H)+]: 390.1249, found: 390.1247.
4.1.6.10. 2-(benzylamino)-4-chloro-N-(2-methoxyethyl)-5-sulfamoyl-benzamide (10b, EA8-8).
The product was purified by chromatography on a column of silica gel with CHCl3:EtOAc (1:1),
1
Rf = 0.25. Yield: 163 mg, 41%, mp 194-197 °C. H NMR δ ppm: 3.28 (3H, s, CH3), 3.38-3.42
(2H, m, CONHCH2), 3.45-3.47 (2H, m, OCH2), 4.49 (2H, d, J = 5.6 Hz, NHCH2Ph), 6.77 (1H,
s, C3-H), 7.21 (2H, s, SO2NH2), 7.25-7.39 (5H, m, Ph-H), 8.12 (1H, s, C6-H), 8.75 (2H, br s,
BnNH, CONH). 13C NMR δ ppm: 39.2, 46.2, 58.4, 70.7, 113.3, 113.4, 126.6, 127.5, 127.6,
129.1, 130.4, 134.8, 138.8, 151.9, 168.2. HRMS calcd. for C17H20ClN3O4S [(M+H)+]: 398.0936,
found: 398.0932.
4.1.6.11. N-benzyl-2-(benzylamino)-4-chloro-5-sulfamoyl-benzamide (14b, EA10-8). The product
was purified by chromatography on a column of silica gel with CHCl3:EtOAc (10:1), Rf = 0.13.
1
Yield: 116 mg, 27%, mp 213-216 °C. H NMR δ ppm: 4.45 (2H, d, J = 6.0 Hz, CONHCH2),
4.50 (2H, d, J = 6.0 Hz, NHCH2), 6.80 (1H, s, C3-H), 7.22 (2H, s, SO2NH2), 7.26-7.39 (10H, m,
Ph-H), 8.22 (1H, s, C6-H), 8.82 (1H, t, J = 6.0 Hz, BnNH), 9.30 (1H, t, J = 6.0 Hz, CONH2). 13C
NMR δ ppm: 42.9, 46.2, 113.1, 113.6, 126.6, 127.3, 127.5, 127.6, 127.7, 128.8, 129.1, 130.4,
134.9, 138.8, 139.9, 152.1, 168.1. HRMS calcd. for C21H20ClN3O3S [(M+H)+]: 430.0987, found:
430.0987.
4.1.7. Procedure for the syntheses of 16(a,b). The mixture of 2,4-dibromo-N-butyl-5-sulfamoyl-
benzamide (16) (1.00 mmol), appropriate amine (2.50 mmol), and 1,4-dioxane was refluxed for 7
days. The solvent was removed under reduced pressure and 2N HCl(aq) (2 mL) was added. The
resultant precipitate was washed with H2O.
4.1.7.1. 4-bromo-N-butyl-2-(cyclohexylamino)-5-sulfamoyl-benzamide (16a, LJ15-35). The
product was purified by chromatography on a column of silica gel with EtOAc:CHCl3(5:1), Rf =
0.30. Yield: 151 mg, 35%, mp 203-205 °C. 1H NMR δ ppm: 0.90 (3H, t, J = 7.2, Hz, CH3), 1.16-
1.57 (10H, m, CH3CH2, CH3CH2CH2, Cy-H), 1.64-1.67 (2H, m, Cy-H), 1.86-1.88 (2H, m, Cy-
H), 3.20 (2H, q, J = 6.8 Hz, NHCH2), 3.49-3.51 (1H, m, Cy-H), 7.02 (1H, s, C3-H), 7.13 (2H, s,
SO2NH2), 8.11 (1H, s, C6-H), 8.41 (1H, d, J = 7.6 Hz, NHCy), 8.61 (1H, t, J = 5.6 Hz, NHCH2).
13C NMR δ ppm: 14.2, 20.1, 24.2, 25.7, 31.5, 32.4, 39.1, 49.8, 113.1, 116.6, 123.8, 127.4, 130.6,
151.0, 168.3. HRMS calcd. for C17H26BrN3O3S[(M+H)+]: 434.0932 (100%), found: 434.0933
(100%).
4.1.7.2. 2-(benzylamino)-4-bromo-N-butyl-5-sulfamoyl-benzamide (16b, LJ15-36). The product
was purified by chromatography on a column of silica gel with EtOAc:CHCl3(5:1), Rf = 0.34.
Yield: 176 mg, 40%, mp 207-209 °C. 1H NMR δ ppm: 0.90 (3H, t, J = 7.2, Hz, CH3), 1.32 (2H,
sext, J = 7.2 Hz, CH3CH2), 1.50 (2H, quint, J = 7.2 Hz, CH3CH2CH2), 3.21 (2H, q, J = 6.8 Hz,

ACCEPTED MANUSCRIPT
NHCH2CH2), 4.48 (2H, d, J = 5.6 Hz, NHCH2Ph), 6.95 (1H, s, C3-H), 7.15 (2H, s, SO2NH2),
7.26-7.39 (5H, m, Ph-H), 8.11 (1H, s, C6-H), 8.64-8.66 (2H, m, NHCH2Ph, NHCH2). 13C NMR δ
ppm: 14.2, 20.1, 31.5, 39.1, 46.2, 114.2, 117.0, 123.5, 127.5, 127.6, 128.2, 129.1, 130.3, 138.9,
151.6, 168.0. HRMS calcd. for C18H22BrN3O3S[(M+H)+]: 442.0619 (100%), found: 442.0623
(100%).
4.1.8. Procedure for the syntheses of 1c, 18. Apropriate compound (6c, 11) (0.500 mmol) was
refluxed in methanol (2 mL), H2O (1 mL), and concentrated HCl (aq) (1mL) solution for 12-24
hours. The progress of reaction was monitored by TLC. The reaction mixture was concentrated
under reduced pressure.
4.1.8.1. 4-chloro-2-(cyclooctylamino)-5-sulfamoyl-benzoic acid (1c, EA1-9). Recrystallization
1
was accomplished from H2O:MeOH (5:1). Yield: 115 mg, 64%, mp 255-257 °C. H NMR δ
ppm: 1.44-1.73 (12H, m, cyclooctyl-H), 1.77-1.89 (2H, m, cyclooctyl-H), 3.76 (1H, br s,
cyclooctyl-H), 6.86 (1H, s, C3-H), 7.31 (2H, s, SO2NH2), 8.39 (1H, s, C6-H), 8.43 (1H, d, J = 7.6
Hz, cyclooctyl-NH), 13.26 (1H, br s, COOH). 13C NMR δ ppm: 23.4, 25.4, 27.2, 31.5, 51.5,
107.8, 113.9, 126.3, 134.1, 136.9, 152.1, 169.4. HRMS calcd. for C15H21ClN2O4S [(M+H)+]:
361.0983, found 361.0987.
4.1.8.2.
4-[(2,4-dichloro-5-sulfamoyl-benzoyl)amino]butanoic
acid
(18,
EA12-1).
Recrystallization was accomplished from NaOAc (20.6 mg, 0.251 mmol) solution in H2O. Yield:
107 mg, 60%, mp 136-139 °C. 1H NMR δ ppm: 1.74 (2H, quint, J = 7.2 Hz, CH2), 2.31 (2H, t, J
= 7.2 Hz, COCH2), 3.27 (2H, q, J = 6.8 Hz, NHCH2), 7.82 (2H, s, SO2NH2), 7.92 (1H, s, C3-H),
7.95 (1H, s, C6-H), 8.69 (1H, t, J = 5.6 Hz, NH), 12.11 (1H, br s, CO2H). 13C NMR δ ppm: 24.8,
31.4, 39.0, 129.1, 132.2, 132.5, 134.4, 136.4, 140.4, 164.9, 174.6. HRMS calcd. for
C11H12Cl2N2O5S [(M+H)+]: 354.9917, found: 354.9918.
Patent application has been submitted for the synthesized compounds as CA inhibitors
(WO2017017505 (A1)).
4.2. Protein preparation
All recombinant human CA isoforms used in this study were expressed and purified as
previously described: CA I [43], CA II [49], CA III, CA IV, CA VA, CA VB, CA VI, CA IX and
CA XIV [31], CA VII and CAXIII [50], CA XII [51]. CAs were expressed in Escherichia coli,
except CA IX, which was expressed in mammalian cells, and purified using chromatography.
Their purity, as determined by electrophoretic analysis SDS-PAGE, was >95 %. Protein MW
was confirmed by MS to a precision of 1 Da and the concentrations before experiments were
measured spectrophotometrically.
4.3. Determination of compound binding to CAs
4.3.1. Fluorescent thermal shift assay

ACCEPTED MANUSCRIPT
Fluorescent thermal shift assay experiments were performed in a Corbett Rotor-Gene 6000
(Qiagen Rotor-Gene Q) instrument using the blue channel (excitation λ=365±20 nm, detection
λ=460±15 nm). Protein samples containing various concentrations of inhibitor were heated from
25°C to 99°C while recording extrinsic fluorescence and determining the protein melting
temperatures at each inhibitor concentration. Samples contained 5–10 µM protein, 0–200 µM
ligand, 50 mM solvatochromic dye ANS and 50–100 mM sodium phosphate buffer containing
100 mM NaCl at pH 7.0, with a final DMSO concentration of 2 %. The applied heating rate was
1 °C/min. Binding constants were determined as previously described in [52].
4.3.2. Isothermal titration calorimetry
ITC experiments were performed by using VP-ITC instrument (Microcal, Inc., part of Malvern,
Northampton, MA, USA). For the binding reactions the calorimeter cell contained 4–10 µM
protein, 2 % DMSO, 50–100 mM sodium phosphate buffer (pH 7.0) and 100mM sodium
chloride. Syringe contained 40–100 µM ligand, 2% DMSO, 50–100 mM phosphate buffer (pH
7.0) and 100 mM sodium chloride. Experiments consisted of 25 injections, volume of the first
injection was 1-5 µL, while others were 10 µL, added at 200 s intervals, at 37 °C.
ITC data was analyzed using MicroCal Origin software. The first point in the integrated data
graph was omitted. The binding constants, enthalpies, and entropies of binding were estimated
after fitting the data with the single binding site model.
4.3.3.Protonation enthalpy of the compound
The enthalpy of inhibitor protonation was measured by titrating with 5 mM HNO3 the cell
solution containing the inhibitor (0.25 mM) and 1.5 equivalent (0.375 mM) of NaOH. The
experiment consisted of 56 injections. The volume of each injection was 5 µL, added at 200 s
intervals. DMSO concentrations in the syringe and the sample cell were equal (2.5%).
Experiments were performed at 37 °C.
4.3.4. Spectrophotometric measurements
To determine the pKa of the ligands, the spectrofotometer model “Agilent 89090A” was used.
The constant concentration of DMSO dissolved ligand solution was added to a buffer solution
(100 mM or 200 mM, ranging from pH 5.5 to 12.5 at every half pH unit) and a UV–VIS spectra
were recorded at 37 °C, peltier-controlled isothermal conditions. The normalized ratio of
absorbances (typically 10 nm above and below the isosbestic point) was plotted as a function of
pH.
4.3.5. Calculation of intrinsic parameters

ACCEPTED MANUSCRIPT
The contribution of protonation reactions was subtracted in the calculation of the intrinsic
thermodynamic parameters.
The intrinsic binding constant Kb_intr is equal to the observed binding constant Kb_obs divided by
the fractions of the reactive species of deprotonated inhibitor and protonated Zn(II)-hydroxide
anion:
(1)
Kd_intr can be calculated by inverting the Kb_intr
.
(2)
Fractions of deprotonated compound (
and protonated Zn(II)-bound water molecule
can be calculated if both pKa values are known:
(3)
(4)
The intrinsic enthalpy of binding has contributions from the observed binding enthalpy (∆Hb_obs),
protonation enthalpies of inhibitor (∆Hp_RSO2NH2), CA (∆Hp_CA) and the buffer (∆Hp_buf):
ΔHb_intr= ΔHb_obs-nRSO2NH2ΔHp_RSO2NH2–nCAΔHp_CA+ nbufΔHp_buf
(5)
where nRSO2NH2= fRSO2NH-– 1 is the number of protons binding to the inhibitor, nCA= 1 – fCAZnH2O is
the number of protons bound to Zn(II)-hydroxide, and n = – (nCA+ nRSO2NH2) is the sum of
uptaken or released protons.
pKa_CAZnH2O values of CA isoforms were previously described: CA I and CA II in [53], CA III,
CA IV in [39], CA VA, CA VB in [54], CA VI in [55], CA VII in [56], CA IX in [57], CA XII in
[51], CA XIII in [58], and CA XIV in [59]. The experimental values were observed at pH 25 ºC,
and recalculated to 37 ºC by using van’t Hoff equation.
4.4. Crystallography
4.4.1. Crystallization

ACCEPTED MANUSCRIPT
The proteins were concentrated by ultrafiltration to 20-50 mg/mL. Crystallization conditions
(buffers) are listed in Table S2. The ligand solutions for crystal soaking were made by mixing of
50 µL of corresponding reservoir solution and 0.5-1µL of 50 mM ligand solution (in DMSO).
4.4.2. Data collection and structure determination
Three datasets of X-ray diffraction (CA II–9, CA XII–7a, and CA XII–7b) were collected at the
EMBL beamlines P13 and P14. The CA XIII-9 dataset was collected using Rigaku MicroMax™-
007 HF X-ray generator. The datasets of CA II–9, CA XII–7a, and CA XII–7b were processed
using XDS [60] program, whereas dataset CA XIII–9 – by MOSFLM [61]. The molecular
replacement was made by MOLREP [62] program using the following initial models: 3HLJ for
CA II, 1JD0 for CA XII, and 2NNO for CA XIII. The 3D models of compounds were built using
AVOGADRO [63] program. The library files which contain complete chemical and geometric
descriptions of compounds were created using LIBCHECK program [64,65]. The models were
prepared using COOT [66] and refined by REFMAC [67].
All represented graphics were made using Pymol programs (PyMOL, version 1.8). Coordinates
and structure factors have been deposited to the RCSB Protein Data Bank (PDB). The PDB
access codes are listed in Table 3.
Acknowledgments
This research was supported by the grant S-MIP-17-87 from the Research Council of Lithuania.
Appendix A. Supplementary data
Supplementary data includes:
1. The table of all compound observed affinities to all CA isoforms.
2. The table of crystallization experimental conditions.
3. Correlation maps between the observed affinities and the chemical structures of
compounds arranged in three groups, single-tailed, double-tailed, and compounds bearing
large substituents.
4. The intrinsic thermodynamics, including the enthalpies and entropies of compound
binding to four CA isoforms.
5. NMR spectra of synthesized compounds.
Abbreviations

ACCEPTED MANUSCRIPT
CA
carbonic anhydrase
FTSA fluorescent thermal shift assay
ITC isothermal titration calorimetry
References
[1]
V.M. Krishnamurthy, G.K. Kaufman, A.R. Urbach, I. Gitlin, K.L. Gudiksen, D.B.
Weibel, G.M. Whitesides, Carbonic Anhydrase as a Model for Biophysical and Physical-Organic
Studies of Proteins and Protein−Ligand Binding, Chem Rev. 108 (2008) 946–1051.
[2]
J.M. Purkerson, G.J. Schwartz, The role of carbonic anhydrases in renal physiology.,
Kidney Int. 71 (2007) 103–115.
[3]
[4]
C.T. Supuran, Carbonic anhydrases–an overview, Curr Pharm Des. 14 (2008) 603–14.
M. Aggarwal, C.D. Boone, B. Kondeti, R. McKenna, Structural annotation of human
carbonic anhydrases, J. Enzyme Inhib. Med. Chem. 28 (2013) 267–277.
[5] M. Aggarwal, R. McKenna, Update on carbonic anhydrase inhibitors: a patent review
(2008 - 2011)., Expert Opin Ther Pat. 22 (2012) 903–915.
[6] C.T. Supuran, A. Scozzafava, Carbonic anhydrases as targets for medicinal chemistry,
Bioorg Med Chem. 15 (2007) 4336–4350.
[7] N. Hen, M. Bialer, B. Yagen, A. Maresca, M. Aggarwal, A.H. Robbins, R. McKenna, A.
Scozzafava, C.T. Supuran, Anticonvulsant 4-Aminobenzenesulfonamide Derivatives with
Branched-Alkylamide Moieties: X-ray Crystallography and Inhibition Studies of Human
Carbonic Anhydrase Isoforms I, II, VII, and XIV., J Med Chem. 54 (2011) 3977–3981.
[8]
B.-B. Gao, A. Clermont, S. Rook, S.J. Fonda, V.J. Srinivasan, M. Wojtkowski, J.G.
Fujimoto, R.L. Avery, P.G. Arrigg, S.-E. Bursell, L.P. Aiello, E.P. Feener, Extracellular carbonic
anhydrase mediates hemorrhagic retinal and cerebral vascular permeability through prekallikrein
activation., Nat Med. 13 (2007) 181–188.
[9]
A. Scozzafava, C.T. Supuran, F. Carta, Antiobesity carbonic anhydrase inhibitors: a
literature and patent review, Expert Opin Ther Pat. 23 (2013) 725–735.
[10] G. De Simone, A. Di Fiore, C.T. Supuran, Are Carbonic Anhydrase Inhibitors Suitable
for Obtaining Antiobesity Drugsꢀ?, Curr. Pharm. Des. 14 (2008) 655–660.
[11] J. Lehtonen, B. Shen, M. Vihinen, A. Casini, A. Scozzafava, C.T. Supuran, A.K.
Parkkila, J. Saarnio, A.J. Kivela, A. Waheed, W.S. Sly, S. Parkkila, Characterization of CA XIII,
a novel member of the carbonic anhydrase isozyme family, J Biol Chem. 279 (2004) 2719–27.
[12] O.Ozensoy Guler, G. De Simone, C.T. Supuran, Drug design studies of the novel
antitumor targets carbonic anhydrase IX and XII, Curr. Med. Chem. 17 (2010) 1516–1526.

ACCEPTED MANUSCRIPT
[13] V. Alterio, A. Di Fiore, K. D’Ambrosio, C.T. Supuran, G. De Simone, Multiple Binding
Modes of Inhibitors to Carbonic Anhydrases: How to Design Specific Drugs Targeting 15
Different Isoforms?, Chem Rev. 112 (2012) 4421-4468.
[14] C.T. Supuran, A. Scozzafava, J. Conway, Carbonic anhydrase: its inhibitors and
activators, CRC Press, 2004.
[15] C.T. Supuran, Advances in structure-based drug discovery of carbonic anhydrase
inhibitors, Expert Opin. Drug Discov. 12 (2017) 61–88.
[16] C.T. Supuran, How many carbonic anhydrase inhibition mechanisms exist?, J. Enzyme
Inhib. Med. Chem. 31 (2016) 345–360.
[17] C.T. Supuran, A. Scozzafava, A. Casini, Carbonic anhydrase inhibitors, Med Res Rev. 23
(2003) 146–189.
[18] P.A. Boriack, D.W. Christianson, J. Kingery-Wood, G.M. Whitesides, Secondary
interactions significantly removed from the sulfonamide binding pocket of carbonic anhydrase II
influence inhibitor binding constants, J Med Chem. 38 (1995) 2286–2291.
[19] F. Pacchiano, M. Aggarwal, B.S. Avvaru, A.H. Robbins, A. Scozzafava, R. McKenna,
C.T. Supuran, Selective hydrophobic pocket binding observed within the carbonic anhydrase II
active site accommodate different 4-substituted-ureido-benzenesulfonamides and correlate to
inhibitor potency., Chem Commun Camb. 46 (2010) 8371–8373.
[20] J.F. Domsic, B.S. Avvaru, C.U. Kim, S.M. Gruner, M. Agbandje-McKenna, D.N.
Silverman, Robert McKenna, Entrapment of carbon dioxide in the active site of carbonic
anhydrase II., J Biol Chem. 283 (2008) 30766–30771.
[21] K.M. Jude, A.L. Banerjee, M.K. Haldar, S. Manokaran, B. Roy, S. Mallik, D.K.
Srivastava, D.W. Christianson, Ultrahigh Resolution Crystal Structures of Human Carbonic
Anhydrases I and II Complexed with “Two-Prong” Inhibitors Reveal the Molecular Basis of
High Affinity, J Am Chem Soc. 128 (2006) 3011–3018.
[22] M. Aggarwal, B. Kondeti, R. McKenna, Insights towards sulfonamide drug specificity in
α-carbonic anhydrases., Bioorg Med Chem. 21 (2013) 1526–1533.
[23] G. De Simone, V. Alterio, C.T. Supuran, Exploiting the hydrophobic and hydrophilic
binding sites for designing carbonic anhydrase inhibitors, Expert Opin. Drug Discov. 8 (2013)
793–810.
[24] C.T. Supuran, Carbonic anhydrases: novel therapeutic applications for inhibitors and
activators., Nat Rev Drug Discov. 7 (2008) 168–181.
[25] W. Vernier, W. Chong, D. Rewolinski, S. Greasley, T. Pauly, M. Shaw, D. Dinh, R.A.
Ferre, S. Nukui, M. Ornelas, E. Reyner, Thioether benzenesulfonamide inhibitors of carbonic
anhydrases II and IV: Structure-based drug design, synthesis, and biological evaluation, Bioorg
Med Chem. 18 (2010) 3307–3319.

ACCEPTED MANUSCRIPT
[26] C. Congiu, V. Onnis, G. Balboni, C.T. Supuran, Synthesis and carbonic anhydrase I, II,
IX and XII inhibition studies of 4-N,N-disubstituted sulfanilamides incorporating 4,4,4-trifluoro-
3-oxo-but-1-enyl, phenacylthiourea and imidazol-2(3H)-one/thione moieties., Bioorg Med Chem
Lett. 24 (2014) 1776–1779.
[27] R.P. Tanpure, B. Ren, T.S. Peat, L.F. Bornaghi, D. Vullo, C.T. Supuran, S.-A. Poulsen,
Carbonic Anhydrase Inhibitors with Dual-Tail Moieties To Match the Hydrophobic and
Hydrophilic Halves of the Carbonic Anhydrase Active Site, J. Med. Chem. 58 (2015) 1494–
1501.
[28] I. Vaškevičienė, V. Paketurytė, A. Zubrienė, K. Kantminienė, V. Mickevičius, D.
Matulis, N-Sulfamoylphenyl- and N-sulfamoylphenyl-N-thiazolyl-β-alanines and their
derivatives as inhibitors of human carbonic anhydrases, Bioorg Chem. 75 (2017) 16–29.
[29] Z. Hou, B. Lin, Y. Bao, H. Yan, M. Zhang, X. Chang, X. Zhang, Z. Wang, G. Wei, M.
Cheng, Y. Liu, C. Guo, Dual-tail approach to discovery of novel carbonic anhydrase IX
inhibitors by simultaneously matching the hydrophobic and hydrophilic halves of the active site,
Eur. J. Med. Chem. 132 (2017) 1–10.
[30] C.T. Supuran, S. Kalinin, M. Tanç, P. Sarnpitak, P. Mujumdar, S.-A. Poulsen, M.
Krasavin, Isoform-selective inhibitory profile of 2-imidazoline-substituted benzene sulfonamides
against a panel of human carbonic anhydrases, J. Enzyme Inhib. Med. Chem. 31 (2016) 197–
202.
[31] V. Dudutienė, J. Matulienė, A. Smirnov, D.D. Timm, A. Zubrienė, L. Baranauskienė, V.
Morkūnaite, J. Smirnovienė, V. Michailovienė, V. Juozapaitienė, A. Mickevičiūtė, J. Kazokaitė,
S. Bakšytė, A. Kasiliauskaitė, J. Jachno, J. Revuckienė, M. Kišonaitė, V. Pilipuitytė, E.
Ivanauskaitė, G. Milinavičiūtė, V. Smirnovas, V. Petrikaitė, V. Kairys, V. Petrauskas, P.
Norvaišas, D. Lingė, P. Gibieža, E. Capkauskaitė, A. Zakšauskas, E. Kazlauskas, E. Manakova,
S. Gražulis, J.E. Ladbury, D. Matulis, Discovery and characterization of novel selective
inhibitors of carbonic anhydrase IX, J. Med. Chem. 57 (2014) 9435–9446.
[32] A.D. Scott, C. Phillips, A. Alex, M. Flocco, A. Bent, A. Randall, R. O’Brien, L. Damian,
L.H. Jones, Thermodynamic Optimisation in Drug Discovery: A Case Study using Carbonic
Anhydrase Inhibitors, ChemMedChem. 4 (2009) 1985–1989.
[33] E. Čapkauskaitė, V. Linkuvienė, A. Smirnov, G. Milinavičiūtė, D.D. Timm, A.
Kasiliauskaitė, E. Manakova, S. Gražulis, D. Matulis, Combinatorial Design of Isoform-
Selective N-Alkylated Benzimidazole-Based Inhibitors of Carbonic Anhydrases,
ChemistrySelect. 2 (2017) 5360–5371.
[34] E. Čapkauskaitė, A. Zubrienė, A. Smirnov, J. Torresan, M. Kišonaitė, J. Kazokaitė, J.
Gylytė, V. Michailovienė, V. Jogaitė, E. Manakova, S. Gražulis, S. Tumkevičius, D. Matulis,
Benzenesulfonamides with pyrimidine moiety as inhibitors of human carbonic anhydrases I, II,
VI, VII, XII, and XIII, Bioorg. Med. Chem. 21 (2013) 6937–6947.

ACCEPTED MANUSCRIPT
[35] E. Čapkauskaitė, L. Baranauskienė, D. Golovenko, E. Manakova, S. Gražulis, S.
Tumkevičius, D. Matulis, Indapamide-like benzenesulfonamides as inhibitors of carbonic
anhydrases I, II, VII, and XIII, Bioorg Med Chem. 18 (2010) 7357–7364.
[36] G.D. Parkes, E. d’A. Burney, 385. Derivatives of 6-bromo- and 4ꢀ: 6-dibromo-m-
toluidine, J Chem Soc. (1935) 1619–1621.
[37] T. Fujikura, K. Miigata, S. Hashimoto, K. Imai, T. Takenaka, Studies on
benzenesulfonamide derivatives with alpha- and beta-adrenergic antagonistic and
antihypertensive activities., Chem Pharm Bull Tokyo. 30 (1982) 4092–4101.
[38] P. Francotte, T.P. De, B. Pirotte, L. Danober, P. Lestage, D.-H. Caignard, Cycloalkylated
benzothiadiazines, a process for their preparation and pharmaceutical compostions containing
them, US2010009974 (A1), 2010.
[39] A. Mickevičiūtė, D.D. Timm, M. Gedgaudas, V. Linkuvienė, Z. Chen, A. Waheed, V.
Michailovienė, A. Zubrienė, A. Smirnov, E. Čapkauskaitė, L. Baranauskienė, J. Jachno, J.
Revuckienė, E. Manakova, S. Gražulis, J. Matulienė, E.D. Cera, W.S. Sly, D. Matulis, Intrinsic
thermodynamics of high affinity inhibitor binding to recombinant human carbonic anhydrase IV,
Eur Biophys J. (2017).
[40] K. Sturm, W. Siedel, R. Weyer, H. Ruschig, Zur Chemie des Furosemids, I. Synthesen
von 5‐Sulfamoyl‐anthranilsäure‐Derivaten, Eur. J. Inorg. Chem. 99 (1966) 328–344.
[41] D. Matulis, V.A. Bloomfield, Thermodynamics of the hydrophobic effect. I. Coupling of
aggregation and pKa shifts in solutions of aliphatic amines, Biophys. Chem. 93 (2001) 37–51.
[42] D. Matulis, V.A. Bloomfield, Thermodynamics of the hydrophobic effect. II.
Calorimetric measurement of enthalpy, entropy, and heat capacity of aggregation of alkylamines
and long aliphatic chains, Biophys. Chem. 93 (2001) 53–65.
[43] D. Matulis, Thermodynamics of the hydrophobic effect. III. Condensation and
aggregation of alkanes, alcohols, and alkylamines, Biophys. Chem. 93 (2001) 67–82.
[44] M.A. Pinard, B. Mahon, R. McKenna, Probing the Surface of Human Carbonic
Anhydrase for Clues towards the Design of Isoform Specific Inhibitors, BioMed Res. Int. 2015
(2015) 1–15.
[45] E. Čapkauskaitė, A. Zubrienė, L. Baranauskienė, G. Tamulaitienė, E. Manakova, V.
Kairys,
S.
Gražulis,
S.
Tumkevičius,
D.
Matulis,
Design
of
[(2-
pyrimidinylthio)acetyl]benzenesulfonamides as inhibitors of human carbonic anhydrases, Eur J
Med Chem. 51 (2012) 259–270.
[46] C. Temperini, A. Cecchi, A. Scozzafava, C.T. Supuran, Carbonic anhydrase inhibitors.
Comparison of chlorthalidone and indapamide X-ray crystal structures in adducts with isozyme
II: when three water molecules and the keto-enol tautomerism make the difference, J Med Chem.
52 (2009) 322–328.

ACCEPTED MANUSCRIPT
[47] L.H. Mckendry, M.W. Zettler, Process for the preparation of 2-chloro and 2,6-
dichloroanilines, US5068392 (A), 1991.
[48] T.R. Williams, D.J. Cram, Stereochemistry of sulfur compounds. IV. New ring system of
carbon, nitrogen, and chiral sulfur, J. Org. Chem. 38 (1973) 20–26.
[49] P. Cimmperman, L. Baranauskienė, S. Jachimovičiūtė, J. Jachno, J. Torresan, V.
Michailovienė, J. Matulienė, J. Sereikaitė, V. Bumelis, D. Matulis, A Quantitative Model of
Thermal Stabilization and Destabilization of Proteins by Ligands, Biophys J. 95 (2008) 3222–
3231.
[50] J. Sūdžius, L. Baranauskienė, D. Golovenko, J. Matulienė, V. Michailovienė, J. Torresan,
J. Jachno, R. Sukackaitė, E. Manakova, S. Gražulis, S. Tumkevičius, D. Matulis, 4-[N-
(Substituted 4-pyrimidinyl)amino]benzenesulfonamides as inhibitors of carbonic anhydrase
isozymes I, II, VII, and XIII, Bioorg Med Chem. 18 (2010) 7413–7421.
[51] V. Jogaitė, A. Zubrienė, V. Michailovienė, J. Gylytė, V. Morkūnaitė, D. Matulis,
Characterization of human carbonic anhydrase XII stability and inhibitor binding, Bioorg. Med.
Chem. 21 (2013) 1431–1436.
[52] L. Baranauskienė, M. Hilvo, J. Matulienė, D. Golovenko, E. Manakova, V. Dudutienė, V.
Michailovienė, J. Torresan, J. Jachno, S. Parkkila, A. Maresca, C.T. Supuran, S. Gražulis, D.
Matulis, Inhibition and binding studies of carbonic anhydrase isozymes I, II and IX with
benzimidazo[1,2-c][1,2,3]thiadiazole-7-sulphonamides, J Enzyme Inhib Med Chem. 25 (2010)
863–870.
[53] V. Morkūnaitė, J. Gylytė, A. Zubrienė, L. Baranauskienė, M. Kišonaitė, V.
Michailovienė, V. Juozapaitienė, M.J. Todd, D. Matulis, Intrinsic thermodynamics of
sulfonamide inhibitor binding to human carbonic anhydrases I and II, J. Enzyme Inhib. Med.
Chem. 30 (2015) 204–211.
[54] A. Kasiliauskaitė, V. Časaitė, V. Juozapaitienė, A. Zubrienė, V. Michailovienė, J.
Revuckienė, L. Baranauskienė, R. Meškys, D. Matulis, Thermodynamic characterization of
human carbonic anhydrase VB stability and intrinsic binding of compounds, J. Therm. Anal.
Calorim. 123 (2016) 2191–2200.
[55] J. Kazokaitė, G. Milinavičiūtė, J. Smirnovienė, J. Matulienė, D. Matulis, Intrinsic binding
of 4-substituted-2,3,5,6-tetrafluorobenezenesulfonamides to native and recombinant human
carbonic anhydrase VI, FEBS J. 282 (2015) 972–983.
[56] V. Pilipuitytė, D. Matulis, Intrinsic thermodynamics of trifluoromethanesulfonamide and
ethoxzolamide binding to human carbonic anhydrase VII: Thermodynamics of TFS and EZA
binding to CA VII, J. Mol. Recognit. 28 (2015) 166–172.
[57] V. Linkuvienė, J. Matulienė, V. Juozapaitienė, V. Michailovienė, J. Jachno, D. Matulis,
Intrinsic thermodynamics of inhibitor binding to human carbonic anhydrase IX, Biochim.
Biophys. Acta BBA - Gen. Subj. 1860 (2016) 708–718.

ACCEPTED MANUSCRIPT
[58] L. Baranauskienė, D. Matulis, Intrinsic thermodynamics of ethoxzolamide inhibitor
binding to human carbonic anhydrase XIII, BMC Biophys. 5 (2012) 12.
[59] V. Juozapaitienė, B. Bartkutė, V. Michailovienė, A. Zakšauskas, L. Baranauskienė, S.
Satkūnė, D. Matulis, Purification, enzymatic activity and inhibitor discovery for recombinant
human carbonic anhydrase XIV, J Biotechnol. 240 (2016) 31–42.
[60] W. Kabsch, XDS, Acta Crystallogr. Sect. D. D66 (2010) 125–132.
[61] T.G.G. Battye, L. Kontogiannis, O. Johnson, H.R. Powell, A.G.W. Leslie, iMOSFLM: a
new graphical interface for diffraction-image processing with MOSFLM, Acta Crystallogr. D
Biol. Crystallogr. 67 (2011) 271–281.
[62] A. Vagin, A. Teplyakov, MOLREP: an Automated Program for Molecular Replacement.,
J Appl Cryst. 30 (1997) 1022–1025.
[63] M.D. Hanwell, D.E. Curtis, D.C. Lonie, T. Vandermeersch, E. Zurek, G.R. Hutchison,
Avogadro: An advanced semantic chemical editor, visualization, and analysis platform., J
Cheminform. 4 (2012) 17.
[64] A.A. Vagin, R.A. Steiner, A.A. Lebedev, L. Potterton, S. McNicholas, F. Long, G.N.
Murshudov, REFMAC5 dictionary: organization of prior chemical knowledge and guidelines for
its use, Acta Crystallogr Biol Crystallogr. 60 (2004) 2184–95.
[65] M.D. Winn, C.C. Ballard, K.D. Cowtan, E.J. Dodson, P. Emsley, P.R. Evans, R.M.
Keegan, E.B. Krissinel, A.G.W. Leslie, A. McCoy, S.J. McNicholas, G.N. Murshudov, N.S.
Pannu, E.A. Potterton, H.R. Powell, R.J. Read, A. Vagin, K.S. Wilson, Overview of the CCP4
suite and current developments, Acta Crystallogr. D Biol. Crystallogr. 67 (2011) 235–242.
[66] P. Emsley, B. Lohkamp, W.G. Scott, K. Cowtan, Features and development of ıt Coot,
Acta Crystallogr. Sect. D. 66 (2010) 486–501.
[67] G.N. Murshudov, A.A. Vagin, E.J. Dodson, Refinement of macromolecular structures by
the maximum-likelihood method, Acta Crystallogr Biol Crystallogr. 53 (1997) 240–55.