Exploring aliphatic sulfonamides as multiclass inhibitors of the carbonic anhydrases from the pathogen bacterium Vibrio cholerae

Abstract

This study investigates aliphatic sulfonamide derivatives as inhibitors of the α‐, β‐, and γ‐class carbonic anhydrase (CA) isoforms from Vibrio cholerae (VchCAs). A series of 26 compounds bearing a triazole linker and urea‐ or ether‐based tails were described and evaluated for their inhibitory action using a stopped‐flow CO₂ hydrase technique. These inhibitors demonstrated a preferential efficacy against VchCAβ. Specifically, the ureido derivatives showed the highest inhibitory potency with inhibition constants (K _Is) in the submicromolar range (0.67–0.93 µM). Selectivity indices were calculated to assess the selective inhibition of VchCAβ over human CA I and II, as well as other VchCA isozymes. Urea‐linked compounds demonstrated a significant 25‐ to 125‐fold selectivity for VchCAβ over hCAs and 14‐ to 26‐fold over other VchCAs. Molecular modeling elucidated the interactions contributing to the efficacy and selectivity of aliphatic sulfonamides as VchCA inhibitors, aligning with and reinforcing the experimental results. The latter suggests that aliphatic sulfonamides could serve as valid targeted therapeutics to treat V. cholerae infections.

Aliphatic sulfonamide derivatives were evaluated as inhibitors of Vibrio cholerae carbonic anhydrase isoforms (VchCAs). These compounds exhibited significant selectivity for VchCAβ over other VchCAs, suggesting their potential as targeted therapeutics for V. cholerae infections.

1. INTRODUCTION

Cholera is a highly contagious acute diarrheal disease caused by the Gram‐negative bacillus Vibrio cholerae.^{[ 1 ]} Morbidity and mortality associated with this disease remain significant issues in developing countries. With appropriate treatment, the mortality rate is less than 0.2%; however, without treatment, the mortality rate can be as high as 50%–70%.^{[ 1} , ² , ³ , ^{4 ]} It is estimated that cholera causes between 21,000 and 143,000 deaths worldwide annually.^{[ 3 ]} The key symptoms of cholera include profuse watery diarrhea, vomiting, and leg cramps.^{[ 2 ]} Cholera treatment involves rapid rehydration through oral or intravenous fluids, which is crucial in managing cholera cases to prevent severe dehydration, electrolyte imbalances, and potentially fatal outcomes. Adjunctive treatment with antibiotics, such as tetracyclines, fluoroquinolones, and macrolides, is beneficial in moderate to severe cases. However, several V. cholerae strains are resistant to tetracycline and sometimes to multiple antibiotics.^{[ 5} , ⁶ , ⁷ , ^{8 ]} Anyhow, considering that Vibrio spp. are bacteria present in freshwater, estuarine and marine environments, improved sanitation, access to clean water, and public health education could play an essential role in cholera prevention and control strategies in at‐risk and developing regions.^{[ 2} , ^{4 ]}

Up to now, two oral inactivated cholera vaccines are also available: WC/rBS (licensed in the USA) and BivWC (used in endemic areas), which are administered in two or three doses and provide 60%–85% protection for 2–3 years.^{[ 9} , ^{10 ]} To colonize human hosts, V. cholerae secretes virulence factors, the most significant being cholera toxin (CT), which is essential for the manifestation of cholera effects resulting in massive diarrhea.^{[ 11} , ^{12 ]} In fact, new therapies targeting virulence genes, toxin production, and colonization by V. cholerae could be beneficial, especially as alternatives to antibiotics become increasingly necessary.^{[ 12} , ¹³ , ^{14 ]}

One of the key physiological reactions for the life cycle of most organisms, including bacteria, is the hydration/dehydration of CO₂. This reaction is linked to numerous metabolic pathways, such as those requiring CO₂ or HCO₃ ^–, and processes including pH homeostasis, electrolyte secretion, and CO₂ and bicarbonate transport.^{[ 15} , ^{16 ]} The reversible hydration of CO₂ is catalyzed by essential enzymes called carbonic anhydrases (CAs), which are a superfamily of metalloenzymes, preferentially including a Zinc(II) atom. The genome of Vibrio cholerae encodes for three CAs including Vibrio cholerae carbonic anhydrase (VchCA)α, VchCAβ, and VchCAγ from the α‐CA, β‐CA, and γ‐CA classes, respectively.^{[ 17} , ^{18 ]} These enzymes play important roles in bacterial cellular metabolism and pH homeostasis and could represent potential targets for developing anti‐infective therapies with the aim of preventing V. cholerae colonization. The three‐dimensional (3D) structure of VchCAβ (PDB 5CXK), solved by X‐ray crystallography in 2015 by Supuran and co‐workers,^{[ 19 ]} revealed a tetrameric type II β‐CA. In 2021, the same group proposed models built by homology for VchCAα and VchCAγ.^{[ 18 ]}

Searching for new inhibitor chemotypes for the design of CA inhibitors, we developed a series of aliphatic sulfonamide derivatives, which demonstrated very potent inhibitory activity against human CA III, but mild inhibitory effects against a plethora of other human isoforms, among which hCA II, IV, VII, and XII. Therefore, considering the different active site architectures among α‐, β‐, and γ‐CAs, the purpose of our work is to provide new information about the potential of aliphatic sulfonamides as inhibitors of three CA‐classes belonging to Vibrio cholerae, adopting in silico methodologies to study the structure–activity relationship.

2. RESULTS AND DISCUSSIONS

Vchα was the first isoform from V. cholerae cloned, purified, and characterized in 2012 by Capasso and co‐workers. This enzyme showed significant catalytic activity. Moreover, a study with sulfonamides and sulfamates led to the discovery of a variety of low nanomolar inhibitors, including methazolamide (I), acetazolamide (AAZ) (II), and ethoxzolamide (III), with K _I values ranging from 0.69 to 8.1 nM (Figure 1).^{[ 20 ]} Subsequently, in 2014, Ceruso et al. and Alafeefy et al. expanded the set of sulfonamide‐based inhibitors, discovering nanomolar inhibitors of pathogenic bacteria. Among these, compounds IV and V, a quinazoline‐based inhibitor, exhibited promising potency and selectivity for Vchα over human carbonic anhydrase (hCA) I and II, with K _I values of 8.8 nM (selectivity index [SI] > 550) and 8.5 nM (SI > 25.7), respectively.^{[ 21} , ^{22 ]}

Figure 1.

Structure of carbonic anhydrase (CA) inhibitors I–IX.

Vchβ was cloned, purified, and structurally characterized in 2015 by Ferraroni et al.,^{[ 19 ]} revealing a tetrameric type II β‐CA. AAZ (II) was found to be a weak inhibitor of this isozyme, with a K _I in the micromolar range (452 μM). In 2016, Capasso and co‐workers improved the series of arylsulfonamide‐based inhibitors, confirming the limited inhibitory capacity of the β isozyme.^{[ 23 ]} Few months later, the third VchCA, VchCAγ, was discovered and characterized in 2016 by Del Prete et al., confirming its classification in the γ‐class. In contrast to the β isoform, VchCAγ was strongly inhibited by sulfonamide‐based compounds, such as Ethoxzolamide (III), which exhibited a low nanomolar inhibition with a K _I of 85.1 nM.^{[ 24 ]} Continued research in subsequent years led to the discovery of new VchCA inhibitors with various CA inhibitory scaffolds. In 2017, Supuran and co‐workers tested imidazole‐based and secondary sulfonamide‐based inhibitors against VchCAα, identifying nanomolar inhibitors such as compounds VI and VII with K _Is of 8.5 and 5.4 nM, respectively (Figure 1).^{[ 25 ]} In the same year, Angeli et al.^{[ 26 ]} evaluated acyl selenoureido benzenesulfonamide, showing subnanomolar inhibition for isozyme α and micromolar inhibition for β isoform. A similar trend was observed in 2018 by Bua et al., who introduced a novel series of sulfimide‐based compounds as VchCA inhibitors, favoring nanomolar inhibition against VchCAα and micromolar inhibition against VchCAβ, as demonstrated by derivative VIII (VchCAα: K _I = 91.4 nM, VchCAβ: K _I > 10000 nM) (Figure 1).^{[ 27 ]} Another CA inhibitor chemotype was also evaluated in 2018 by Berrino et al., which provided a kinetic study against all VchCA with a panel of N’‐aryl‐N‐hydroxy‐ureas.^{[ 28 ]} This set of peculiar inhibitors showed a better inhibition against the VchCAβ isozyme with respect to the α and γ isoforms, for example, compound IX (VchCAα: K _I = 4000 nM, VchCAβ: K _I > 60.3 nM, VchCAγ: K _I > 10000 nM) (Figure 1).^{[ 28 ]} In 2019, Angeli et al. assessed famotidine, an antiulcer drug belonging to the H2‐antagonists class of pharmacological agents, against three VchCAs. Following the behavior of several sulfonamide‐based inhibitors, it resulted to be a nanomolar modulator of VchCAα, K _I of 72.3 nM, and a micromolar inhibitor of VchCAβ, K _I of 83917 nM, and VchCAγ, K _I of 5321 nM.^{[ 29 ]}

It should be stressed that, since their discovery, VchCAs have been extensively examined for their modulation to offer new perspectives in the CA domain. In fact, alongside the previously discussed inhibitors I–IX (Figure 1), additional CA modulators have recently emerged as promising regulators of V. cholerae. These include benzoxaboroles,^{[ 30 ]} poly(amidoamine) derivatives,^{[ 31 ]} phenols,^{[ 32 ]} coumarins (only against VchCAα, the singular VchCA with hydrolyzing activity),^{[ 33 ]} and hydrazones.^{[ 34 ]} In this work, we provide fresh insights into the neglected aliphatic sulfonamides, to overcome the selectivity profile of the parent arylsulfonamide derivatives in VChCA inhibition.

The development of aliphatic sulfonamides was facilitated by a scaffold resizing approach (Figure 2). This method was initially employed in 2023 to uncover potent and selective hCA III inhibitors.^{[ 35 ]}

Figure 2.

The adopted strategy for the synthesis of aliphatic sulfonamide‐based derivatives.

Briefly, the synthesis initially involved the incorporation of azide and sulfonic acid moieties by the cleavage of 5‐ and 6‐membered sultones with sodium azide (NaN₃). These formed sulfonic acid moieties, which were converted into the corresponding sulfonyl chlorides using a suitable chlorinating agent, and subsequently transformed into sulfonamides with an ammonia solution. For building the counterpart of the products, several alkyne‐based intermediates were synthesized starting from substituted phenols or isocyanates which were respectively reacted with propargyl bromide and propargyl amine under mild conditions. After that, the Cu‐catalyzed Huisgen‐cycloaddition reaction was applied to form the triazole as a linker between aliphatic sulfonamides and different tails, achieving compounds 1–26.^{[ 35 ]}

2.1. CA inhibition studies

The CA inhibition profile of compounds 1–26 was thus evaluated against three VchCA isoforms by a stopped‐flow CO₂ hydrase assay,^{[ 36 ]} using AAZ as a reference inhibitor (Table 1). The data against physiologically relevant hCA I and II, as well as SAR discussion, were already reported.^{[ 35 ]}

Table 1.

 Inhibition activity of derivatives 1−26 and the standard CA inhibitor Acetazolamide (AAZ) against human isoforms hCA I and II and bacterial CAs VchCAα, VchCAβ and VchCAγ belonging to Vibrio cholerae by the CO₂ hydrase stopped‐flowAssay.

Code	R	n	K I (µM)a
			hCA		VchCA
			I	II	α	β	γ
1		1	0.75	2.72	29.1	4.16	37.8
2		1	4.89	2.02	21.3	3.74	33.1
3		1	1.16	2.44	16.5	2.96	25.6
4		1	1.91	4.10	14.9	3.38	23.8
5		1	>100	2.89	13.4	0.93	12.6
6		1	>100	2.96	14.6	0.82	10.9
7		1	>100	4.16	11.2	0.67	17.3
8		1	>100	2.50	8.43	2.59	15.1
9		1	>100	0.93	17.5	1.72	9.4
10		1	>100	0.87	10.6	4.03	11.5
11		1	0.62	0.65	22.6	7.44	53.0
12		1	0.60	0.71	24.5	7.18	49.3
13		1	0.86	3.14	30.7	9.67	41.9
14		2	6.53	7.76	34.4	6.91	43.5
15		2	6.42	8.12	25.1	5.89	39.7
16		2	8.54	3.93	28.7	6.13	40.3
17		2	2.31	5.97	26.6	4.97	36.1
18		2	5.38	7.57	16.9	3.41	20.6
19		2	5.29	4.93	18.3	2.86	15.3
20		2	5.38	0.78	16.8	3.00	16.8
21		2	6.52	0.56	20.4	5.08	18.4
22		2	5.51	4.60	14.0	1.97	22.2
23		2	>100	8.15	12.6	2.14	14.6
24		2	0.91	0.77	27.8	8.35	62.3
25		2	3.27	0.80	31.4	8.04	60.6
26		2	0.77	8.13	30.1	11.3	75.0
AAZ	‐	‐	0.25	0.012	0.0068	0.452	0.473

Abbreviations: AAZ, acetazolamide; hCA, human carbonic anhydrase; VchCA, Vibrio cholerae carbonic anhydrase.

^a Mean from three different assays, by a Stopped‐Flow technique (errors were in the range of ±5%–10% of the reported values).

The following structure–activity relationship (SAR) can be gathered from the inhibition data reported in Table 1.

VchCAα is minimally affected by all derivatives 1–26, with K _I values falling within the medium micromolar range. Particularly, K _I values range from 8.43 µM up to 34.4 µM. Notably, compound 8 (n = 1), featuring a urea‐based link, emerged as the most potent compound with a K _I of 8.43 µM, surpassing its counterpart 21 (n = 2) by at least twofold in efficacy. This pattern persisted with other congeners like 5–7 and 10, which exhibited greater potency than 18–20 and 23, respectively. On the other hand, the 3‐methoxyphenylureido derivative 22, with a K _I of 14.0 µM, was the sole compound in this selection to demonstrate more effective inhibition of VchCAα than its lower analog 9, with a K _I of 17.5 µM. The substitution on the phenyl had a marginal impact on the potency of these compounds. For instance, compounds 5–7 displayed K _Is between 11.2 and 13.4 µM, while 18–20 showed K _Is between 16.8 and 18.3 µM. The ether‐based link derivatives 1–4 were detected as poor inhibitors of this isoform, exhibiting K _Is of 14.9–29.1 µM. In particular, the presence of a methoxy group on the phenyl moiety improves the potency, as seen with compounds 3 and 4 with K _Is of 16.5 and 14.9 µM, respectively, while the superior analogs 16 and 17 exhibited a lower activity, with K _Is of 28.7 and 26.6 µM, but comparable to that of their congener 15, K _I of 25.1 µM, featuring a fluorine atom as a substituent. Compounds with nonsubstituted phenyl moiety, such as 1, K _I of 29.1 µM, and 14, K _I of 34.4 µM, demonstrated a diminished potency, with 14 being the least effective inhibitor against this isoform. Finally, compounds 11–13 and 24–26 exhibited comparable activity against VchCAα ranging from 22.6 to 31.4 µM. Compound 11, K _I of 22.6 µM, demonstrated a more potent inhibition than the superior analog 24, K _I of 27.8 µM, as well as 12 and 25, with K _Is of 24.5 and 31.4 µM, respectively. On the other side, the length of the chain did not significantly affect the K _Is of 13 and 26, which were 30.7 and 30.1 µM, respectively.

VchCAβ is the most inhibited isoform by the CA inhibitors studied 1–26, with K _I values ranging from high nanomolar to medium micromolar, specifically between 0.67 and 11.3 µM. The urea‐based compounds were the most effective inhibitors, exhibiting K _Is even in the nanomolar range. In fact, the most potent inhibitor identified is derivative 7, with a K _I of 0.67 µM. Compounds 5 and 6 also showed impressive inhibition profiles, with K _I values of 0.93 and 0.82 µM, respectively. All these compounds feature a para‐substituted phenyl with lipophilic substituents such as fluorine (5), methyl (6), or trifluoromethyl (7). However, this promising activity was not observed in their superior analogs 18–20, which exhibited K _I values in the low micromolar range, 3.41, 2.86, and 5.08 µM, respectively. The inhibitory potential of ureidic derivatives was generally influenced by the length of the aliphatic linker, except for the analogs 9, K _I of 1.72 µM, and 22, K _I of 1.97 µM. The ether‐based linked derivatives also showed variable inhibition, particularly for the p‐methoxyphenyl substituted compounds 3 (n = 1) and 16 (n = 2), with K _Is of 2.96 and 6.13 µM, respectively. The methoxy substituent's position did not significantly affect the inhibitory potency of compounds 4 (n = 1), K _I of 3.38 µM, and 17 (n = 2), K _I of 4.97 µM. Moreover, the absence of substituents on the phenyl produced poor inhibitors of this isoform, 1, K _I of 4.16 µM, and 14, K _I of 6.91 µM. Finally, compounds 11–13 and 24–26, lacking a specific linker between the CAI chemotype and tail, were the least effective inhibitors, with K _Is ranging from 7.18 to 11.3 µM. It should be noted that the absence of phenyl significantly decreased the potential inhibitory activity of this series, as demonstrated by inhibitors 13 (n = 1), K _I of 9.67 µM, and 26 (n = 2), K _I of 11.3 µM.

The third and final assessed isoform is VchCAγ, which is the least inhibited by aliphatic sulfonamides 1–26, with K _I values ranging from 9.4 to 75 µM. Compound 9 (n = 1), bearing a 3‐methoxypheylureidic tail, is the most potent inhibitor of this class showcasing a K _I value in the low micromolar range, 9.4 µM; meanwhile, other derivatives pointed out K _Is in the medium micromolar range. For instance, compounds 5–8 and 10 (n = 1) are among the poorest inhibitors, with K _I values ranging from 10.9 to 17.3 µM. In comparison with their superior homologs, 18–23 (n = 2), showing K _Is between 14.6 and 22.2 µM, a shift in the inhibitory activity was observed. In fact, within this subset, compound 23 is the most effective with a K _I of 14.6 µM, while compound 22 is the least potent, with a K _I of 22.2 µM. The ether‐based derivatives exhibited a lower inhibitory activity against this isoform, confirming the pattern observed for the previous bacterial α‐ and β‐CAs. In fact, The K _I values for compounds 1–4 (n = 1) range from 23.8 to 37.8 µM, while their superior homologs 14–17 (n = 2) range from 36.1 to 43.5 µM. Particularly, the presence of a substituted phenyl in the tail part increased the activity of each derivative compared with the nonsubstituted ones, as demonstrated by compounds 2–4 (K _Is of 23.8–33.1 µM) over derivative 1 (K _I of 37.8 µM) and compounds 15–17 (K _Is of 36.1–40.3 µM) over derivative 14 (K _I of 43.5 µM). Finally, compounds 11–13 and 24–26 resulted to be the poorest inhibitors of this isoform, with K _I values spanning toward the high part of the medium micromolar range, 41.9–75.0 µM.

Considering that VchCAβ was identified as the most inhibited isozyme among the three classes of CAs in V. cholerae, we further present the SI for human CAs I and II, as well as for V. cholerae α and γ isozymes, relative to VchCAβ. This analysis aims to clarify the preferential inhibition of these compounds and highlight their selectivity for VchCAβ over both the human isoforms and the other bacterial isozymes.

Regarding the target/off‐target CA selectivity ratios of these aliphatic sulfonamides, several compounds exhibited notable I/VchCAβ and II/VchCAβ SIs, ranging from 0.083 to 149.2 and 0.087 to 6.20, respectively (Table 2). Additionally, to demonstrate the preferential inhibition of VchCAβ, the VchCAα/VchCAβ and VchCAγ/VchCAβ selectivity ratios were calculated, with ranges of 2.63–17.8 and 2.85–25.8, respectively. Generally, a random specificity based on the type of linker was observed for human CAs I and II against VchCAβ, while a marked preference for VchCAβ over the α and γ isozymes was noted. Specifically, a similar pattern among the SI values can be observed based on the type of linker and tail. Urea‐linked derivatives strongly preferred VchCAβ over other VchCAs and human CAs. Notably, derivatives 5–10 and 23 showed superior SI values for hCA I/VchCAβ (Table 2), reaching at least two orders of magnitude selectivity, with SI values ranging from 24.8 to 149.2. Furthermore, some compounds also demonstrated significant selectivity for VchCAβ against the α and γ isoforms. For example, compounds 5–7 maintained one order of magnitude selectivity for the β‐class over the α (SI: 14.4–17.8) and γ (SI: 13.2–25.8) isoforms. In contrast, the ether‐based linked derivatives did not show preferential inhibition for VchCAβ, as indicated by their SI values for hCA I/VchCAβ and hCA II/VchCAβ, with most SI values being less than 1. Specifically, compounds 1 (SI: hCA I/VchCAβ, 0.18), 3 (SI: hCA I/VchCAβ, 0.39), 4 (SI: hCA I/VchCAβ, 0.56), and 17 (SI: hCA I/VchCAβ, 0.46) showed good selectivity for hCA I, which then decreased for hCA II, indicating comparable inhibition to that against VchCAβ. Considering the three VchCAs, compounds 1–4 and 14–17 exhibited low selectivity for the β‐class, with SI values ranging from 4.26 to 6.99 (VchCAα/VchCAβ) and 6.29 to 9.08 (VchCAγ/VchCAβ). Finally, compounds 11–13 and 24–26 exhibited strong selectivity for the human CAs over VchCAβ, with some being up to 100‐fold more selective. This is demonstrated by compounds 11–13 (SI: hCA I/VchCAβ, 0.083–0.088) and 26 (SI: hCA I/VchCAβ, 0.068) for hCA I, and compounds 11 (SI: hCA II/VchCAβ, 0.087), 12 (SI: hCA II/VchCAβ, 0.099), 24 (SI: hCA II/VchCAβ, 0.092), and 25 (SI: hCA II/VchCAβ, 0.099) for hCA II. Among the VchCAs, these compounds showed low selectivity for the β‐class over the α and γ types, with SI values greater than 1.

Table 2.

The selectivity index (SI) of derivatives 1–26 has been assessed between human CAs I and II against VchCAβ, as well as between VchCAα and VchCAγ against VchCAβ.

Code	Selectivity index (SI)
	hCA I/VchCAβ	hCA II/VchCAβ	VchCAα/VchCAβ	VchCAγ/VchCAβ
1	0.18	0.65	6.99	9.08
2	1.30	0.54	5.69	8.85
3	0.39	0.82	5.57	8.64
4	0.56	1.21	4.40	7.04
5	>107.5	3.10	14.4	13.5
6	>121.9	3.60	17.8	13.2
7	>149.2	6.20	16.7	25.8
8	>38.6	0.96	3.25	5.83
9	>58.3	0.54	10.1	5.46
10	>24.8	0.21	2.63	2.85
11	0.083	0.087	3.03	7.12
12	0.084	0.099	3.41	6.86
13	0.088	0.32	3.17	4.33
14	0.94	1.12	4.97	6.29
15	1.08	1.37	4.26	6.74
16	1.39	0.64	4.68	6.57
17	0.46	1.20	5.35	7.26
18	1.57	2.21	4.95	6.04
19	1.84	1.72	6.39	5.34
20	1.79	0.26	5.6	5.60
21	1.28	0.11	4.01	3.62
22	2.79	2.33	7.10	11.2
23	>46.7	3.80	5.88	6.82
24	0.11	0.092	3.32	7.46
25	0.40	0.099	3.90	7.53
26	0.068	0.72	2.66	6.63
AAZ	0.55	0.026	0.015	1.04

Abbreviations: AAZ, acetazolamide; hCA, human carbonic anhydrase; VchCA, Vibrio cholerae carbonic anhydrase.

2.2. In silico studies

In silico studies were performed to clarify the binding mode of aliphatic sulfonamides within the active sites of VchCAs. Currently, no solved crystal structures are available in the Protein Data Bank (PDB)^{[ 37 ]} for VchCAα, VchCAγ, and type I/open VchCAβ. Therefore, docking simulations were carried out using homology‐built models of the three Vibrio cholerae CAs, according to the previously reported procedures.^{[ 30 ]} To assess the impact of the linker type and aromatic tail substitution on the ligand‐target interaction, compounds 2, 7, and 12 were selected as representative ligands for this study. According to the literature,^{[ 18 ]} all docking solutions found the deprotonated nitrogen atom of the sulfonamide group (SO₂NH^–) bound to the zinc ion, completing the tetragonal coordination geometry of the metal in all three VchCAs.

In the VchCAα active site, two H‐bonds are formed between the zinc‐bound sulfonamide NH^– and S═O groups with the side chain OH and the backbone NH of T199, respectively, which further stabilizes the ligands within the active site (Figure 3a–c). The aliphatic linker between the sulfonamide group and the triazole is centrally located within the cavity, allowing the five‐membered cycle to form van der Waals (vdW) interactions solely with the L188 residue. The aromatic tail of compound 2 forms a π–π stacking interaction with the indole of W23 (Figure 3a). However, the shorter tail of compound 12, compared with ligand 2, prevents the formation of such an interaction (Figure 3c). In contrast, the urea linker in compound 7 forms two H‐bonds between its C═O group and the NH₂ side chains of N77 and Q82, causing the aromatic tail to be located in a different pocket of the active site (Figure 3b).

Figure 3.

Predicted binding mode of compounds 2 (purple), 7 (orange), and 12 (green) within VchCAα (a–c), VchCAβ (d–f), and VchCAγ (g–i) active sites. H‐bonds and π–π stacking interactions are depicted as black and cyan dashed lines, respectively. Amino acid labels are colored according to the a (black) and b (blue) chains (d–i). VchCAs, Vibrio cholerae carbonic anhydrase.

In the dimeric and narrow VchCAβ‐binding pocket, the sulfonamide zinc‐binding motif of the ligands participates in two H‐bonds with the residues D44 (chain A) and Q33’ (chain B), reinforcing the stability of the complex. The aliphatic chain is involved in several vdW contacts with V66, Y83’, and G102, and in addition, the triazole is stacked with the phenolic side chain of Y83’, which is typically involved in the stabilization of aromatic sulfonamides, forming π–π interactions (Figure 3d,e). In compound 7, the carbonyl group of the urea linker is in H‐bond distance to the Y83’ side chain, contributing to the stability of the binding pose and also orienting the aromatic tail differently within the binding site (Figure 3e).

Within the dimeric VchCAγ active site, the coordination of the ligand to the zinc ion is further stabilized by two H‐bonds between the sulfonamide group and the side chain NH₂ and OH of Q59 and Y166, respectively (Figure 3g–i). However, the docking solutions predict a linear arrangement of the three compounds at the center of the cylindrical VchCAγ active site, which limits both polar interactions and hydrophobic contacts. In this arrangement, the carbonyl C═O in the urea linker of compound 7 forms an H‐bond with the backbone NH of A124 (Figure 3h).

Overall, the greater inhibitory ability of compound 7 compared with compounds 2 and 12, observed across the three isoforms (Table 1), may be related to the urea linker's ability to engage in polar interactions. Additionally, the enhanced inhibition of all three ligands toward VchCAβ, as opposed to VchCAα and VchCAγ, can be explained by the smaller size of the VchCAβ active site, particularly near the catalytic zinc ion, which allows for maximized hydrophobic contacts with the ligands’ aliphatic linker.

3. CONCLUSION

This study presents a series of aliphatic sulfonamide derivatives acting as potential inhibitors of the V. Cholerae‐expressed CAs, α‐, β‐ and γ‐VchCA. The design of these inhibitors, featuring a triazole linker and urea‐ or ether‐based tails, exhibited a preferential inhibition of the VchCAβ over the VchCAα and VchCAγ, pointing out K _I values ranging from high nanomolar to medium micromolar concentrations. Among them, the urea‐based derivatives 5–10 and 18–23 stood out as the most potent subset demonstrating a superior inhibitory activity. The SIs emphasized such preferential inhibition toward the VchCAβ over the other two microbial isozymes and also over the human isoforms I and II, which are considered the main off‐targets. The predicted in silico analyses on compounds 2, 7, and 12 provided a deeper understanding of the molecular interactions that explain this selectivity, further reinforcing the experimental observations. In fact, such predictions suggested that compounds belonging to the urea‐based subsets, 5–10 and 18–23, form more extensive interactions within the active site of the VchCAβ, which may explain their major inhibitory efficacy. Overall, the findings present a novel intriguing chemotype for the inhibition of CAs expressed by V. cholerae with a primary specificity for the β isozyme over the α and γ isozymes. Accordingly, this could pave the way for the selective inhibition of a specific family of CAs, thereby reducing the risk of side effects and unwanted consequences associated with pan‐CA inhibition.

4. MATERIALS AND METHODS

4.1. Chemistry

Synthetic procedures and compound characterization are reported in Giovannuzzi et al.^{[ 35 ]}

4.2. CA inhibition

The CA‐catalyzed CO₂ hydration activity has been measured with an Applied Photophysics stopped‐flow instrument.^{[ 36 ]} The used pH indicator was phenol red (at a concentration of 0.2 mM), working at the absorbance maximum of 557 nm. A volume of 10 mM HEPES (pH 7.4) was employed as a buffer, in the presence of 10 mM NaClO₄ to maintain the ionic strength constant. The initial rates of the CA‐catalyzed CO₂ hydration reaction were followed up for a period of 10–100 s. The substrate CO₂ concentrations ranged from 1.7 to 17 mM for determining the inhibition constants. For each inhibitor, at least six traces of the initial 5%–10% of the reaction were used to determine the initial velocity. The uncatalyzed rates were determined in the same manner and subtracted from the total observed rates. Stock solutions of inhibitors (10 mM) were prepared in distilled‐deionized water with a maximum of 5% DMSO, and dilutions up to 10 nM were done thereafter with the assay buffer. Inhibitor and enzyme solutions were preincubated together for 1–6 h before the assay, to allow for the formation of the enzyme‐inhibitor complex. The inhibition constants were obtained by nonlinear least‐squares methods using Prism 3 and the Cheng‐Prusoff equation, as reported previously, and represent the mean from at least three different determinations. All enzymes employed were recombinant and obtained in‐house as reported,^{[ 35 ]} with concentrations in the assay ranging from 5 to 12 nM. The human/protozoan enzymes were recombinant proteins obtained in‐house, as described earlier.^{[ 30} , ^{32 ]}

4.3. In silico studies

The experimental protocol used for the development of the VchCAα, VchCAβ, and VchCAγ homology models was described in Bonardi et al.^{[ 30 ]} The primary sequences of VchCAα and VchCAγ were retrieved from the UniProt Consortium. The crystal structure of α‐CA from Photobacterium profundum (PDB 5HPJ)^{[ 38 ]} and γ‐CA homologous protein from Escherichia coli (PDB 3TIO)^{[ 39 ]} were used as templates in the homology modeling procedure. Multiple models were generated using the Prime module of Schrödinger^[40a] and the SwissModel platform (https://swissmodel.expasy.org/) and submitted to loop refinement and quality evaluation procedures (Supporting Information S1: Figures S1–S9 and Tables S1–S3).^{[ 30 ]} The best‐scored structures of VchCAα and VchCAγ and the VchCAβ crystal structure (retrieved as PDB 5CXK)^{[ 19 ]} were prepared for docking using the Protein Preparation Wizard tool, using the OPLS4 force field^{[ 41} , ⁴² , ^{43 ]} and applying an energy minimization protocol with a root mean square deviation (RMSD) value of 0.30 Å. Ligand structures were generated using Maestro^[40b] and their ionization states were evaluated at pH 7.4 ± 0.5 using Epik.^[40c] Energy minimization of the ligands was carried out using the conjugate gradient method in Macromodel,^[40e] with a maximum of 2500 iterations and a convergence criterion of 0.05 kcal mol^‐1 Å^‐1. Docking was performed with Glide,^[40f] centering grids on the centroids of the zinc‐coordinating residues, and ligands were docked using the standard precision (SP) mode. Figures were generated with Maestro and Chimera.^{[ 40} , ^{44 ]}

CONFLICTS OF INTEREST STATEMENT

The authors declare no conflicts of interest.

Supporting information

Supporting information.

Paoletti N., Giovannuzzi S., Bonardi A., De Luca V., Capasso C., Nocentini A., Gratteri P., Supuran C. T., Arch. Pharm. 2025, 358, e2400814. 10.1002/ardp.202400814

DATA AVAILABILITY STATEMENT

The data that support the findings of this study are available from the corresponding author upon reasonable request.