Design and synthesis of isoxazole-functionalized benzene sulphonamides as novel inhibitors of Mycobacterium tuberculosis β-carbonic anhydrases

Abstract

The escalating prevalence of multidrug-resistant tuberculosis (MDR-TB) underscores the urgent need for new classes of antitubercular agents targeting novel pathways. Carbonic anhydrase, a ubiquitous metalloenzyme, catalyses the reversible hydration of carbon dioxide in the CO₂ + H₂O ⇋ HCO₃⁻ + H⁺ reaction. Suppressing this enzymatic activity has recently been identified as a new pathway for the treatment of Mycobacterium tuberculosis. To address this, a series of isoxazole–sulphonamides was rationally designed, incorporating an isoxazole pharmacophore as the aromatic tail, amide as a linker, and sulphonamide as the zinc-binding group. These compounds were evaluated against Mycobacterium tuberculosis carbonic anhydrases (MtCA 1 and 3) and two human carbonic anhydrases (hCA I and II) to identify selective inhibitors of the bacterial enzymes. The findings indicated that molecules containing an isoxazole pharmacophore with amide-linked benzene-3-sulphonamide were significantly more selective for MtCA 3 than hCA I and II. Among these compounds, 12c, 12e, and 19b had the highest inhibition against the MtCA 3 with K_i values between 0.08–0.09 μM compared to the standard acetazolamide with a K_i value of 0.10 μM. Some of the best compounds exhibited potent and selective inhibition of MtCA 3 over hCA I and II, with the meta- and para-substituted derivatives demonstrating higher selectivity and stronger inhibition. Specifically, compound 19b proved to be 199 and 38 times more selective for MtCA 3 than hCA I and hCA II respectively, compared to the standard drug acetazolamide, which is a non-selective CA inhibitor. The potential of compound 19b as a promising antitubercular agent with a MIC value of 8 μg mL⁻¹ against mc² 6230 was further strengthened by in silico ligand–target interaction studies. Thus, compound 19b is emphasised as a promising lead in the pursuit of new, selective agents targeting MtCA 3.

A series of isoxazole–sulphonamide derivatives were synthesised and tested against MtCA1, 3 and hCA I & II. Among those, 19b showed 199 and 38 fold selectivity for MtCA3 over hCA I & II respectively, compared to the standard drug acetazolamide.

1. Introduction

Tuberculosis (TB), a contagious disease caused by Mycobacterium tuberculosis, is the second-deadliest infectious disease after COVID-19, with a fatality rate surpassing that of AIDS. It is estimated that approximately one-third of the human population is affected by TB.^1,2 According to the World Health Organization (WHO), TB caused 1.8 million deaths annually, and 10.6 million new cases were reported in 2024.^3–5 Additionally, the rise of drug-resistant tuberculosis, especially multidrug-resistant (MDR-TB) and extensively drug-resistant (XDR-TB), poses a significant risk to public health. Therefore, there is a need to explore newer targets and medications to tackle the TB epidemic.⁶ In recent years, significant progress has been made in the research on M. tuberculosis, especially with the decoding of the M. tuberculosis genome, which has led to the discovery of newer targets and novel mechanisms.^7,8 Among these new targets, Mycobacterium tuberculosis carbonic anhydrases (MtCAs) have been identified after the M. tuberculosis genome was deciphered. The carbonic anhydrases (CAs, EC 4.2.1.1) have been cloned, purified, and characterised by Jones' and Supuran's groups.^9,10 To date, eight distinct families of CAs—α-, β-, γ-, δ-, ζ-, η-, θ-, and ι-, each featuring various isoforms, have been recognised in bacteria, archaea, and eukaryotes.⁴ Pathogenic bacteria typically possess carbonic anhydrases (CAs) from the α-, β, and γ-CA families.

Among the eight classes of CAs, the genome of Mycobacterium tuberculosis (Mtb) encodes three β-carbonic anhydrases (β-CAs), MtCA 1, 2, and 3, corresponding to the genes Rv1284, Rv3588c, and Rv3273, respectively.¹¹ These enzymes retain all conserved amino acid residues typical of β-carbonic anhydrases—Cys42, Asp44, His97, and Cys101—which coordinate the active-site zinc ion in the closed form of the enzyme active site.¹⁰ The zinc ion is indispensable for catalytic activity, enabling the reversible hydration of carbon dioxide (CO₂ + H₂O ⇋ HCO₃⁻ + H⁺) through an activated water molecule coordinated to the metal center.^12,13 The activity of CA plays an essential role in several physiological processes, including pH regulation, gluconeogenesis, bone metabolism, and electrolyte homeostasis.¹⁴ J. Rose et al. demonstrated that bicarbonate, generated through the β-CA-catalysed reversible hydration of CO₂, is an essential factor in facilitating extracellular DNA (eDNA) transport and promoting biofilm formation in nontuberculous mycobacteria (NTM) under in vitro conditions. Inhibition of β-CAs using ethoxzolamide (EZA), a CA inhibitor, reduced the eDNA transport and biofilm formation.¹⁵ Additionally, EZA inhibited the PhoPR regulon, a two-component regulatory system in Mtb, as well as the Esx-1 protein secretion system, which is critical for the virulence of Mtb. It also showed efficacy in infected macrophages and mouse models, reinforcing the role of β-CAs in mycobacterial pathogenesis, as further evidenced by studies using M. marinum, a non-tuberculous mycobacterial (NTM) model bacterium. Aspatwar et al. were the first to report that Fc14-584b, a dithiocarbamate β-CA inhibitor, impairs mycobacterial growth in zebrafish larvae in vivo.^16,17 These essential enzymes are thus potential drug targets and are currently under investigation by several groups.^18–20 Similarly, several in vitro studies have shown that all the Mtb β-CAs could be efficiently (K_i in nanomolar ranges) inhibited by sulphonamides/sulfamates.²¹

Crucially, the sulphonamide moiety, known as a key pharmacophore in classical CA inhibitors, was incorporated into these designs.^22,23 Indeed, the majority of the clinical CAIs, such as acetazolamide 6, dichlorphenamide B, dorzolamide C, brinzolamide D, and indisulam E, contain the sulphonamide group as shown in Fig. 1.^24,25 These compounds have been in clinical use or under development for the treatment of various diseases, like cancer, tuberculosis, etc. These primarily target α-CAs but have also demonstrated nanomolar to sub-micromolar activity against the β-CA of mycobacteria. In addition to this, some of the diazenyl benzene sulphonamides,²⁶ are also reported for MtCA inhibition. Our group has been involved in the development of new MtCA inhibitors, e.g., pyrazole-containing sulphonamide F (Fig. 1) demonstrated significant MtCA inhibition, with a K_i value of 0.523 μM against MtCA 3.²⁷

Fig. 1. Reported sulphonamide-based drugs/compounds targeting Mycobacterium tuberculosis carbonic anhydrase (MtCA).

On the other hand, the isoxazole scaffold has attracted significant attention from medicinal chemists due to its unique structural features, which facilitate the synthesis of versatile bioactive molecules with distinct medicinal properties, including antiviral, anti-inflammatory, analgesic, anticancer, antioxidant, antimicrobial, and antitubercular activities.²⁸ Furthermore, the literature review revealed that the tail approach strategy was employed in the design of isoxazole derivatives to achieve improved selectivity and potency towards mycobacterial CA isoforms. Maddipatla et al., reported the antitubercular properties of the (E)-5-(4-morpholinophenyl)-N′-((4-oxo-2-thioxothiazolidin-5-ylidene)methyl)isoxazole-3-carbohydrazide (I) against MtCA 1, 2 and 3, demonstrating good activity with K_i values of 82.1, 63.9 and 42.0, respectively, and a MIC value of 1 μg mL⁻¹ against Mtb mc² 6230 (Fig. 2).²⁹ The antitubercular activity of N-(p-tolyl)-5-(2-(p-tolylamino)thiazol-4-yl)isoxazole-3-carboxamide (II) against Mtb H₃₇Rv with a MIC value of 0.5 μg mL⁻¹ was reported by Girandini et al.³⁰ Huang et al., reported methyl (S)-2-(5-(((2-methylbenzo[d]thiazol-5-yl)oxy)methyl)isoxazole-3-carboxamido)-2-phenylacetate (III) against replicating Mycobacterium tuberculosis (Mtb) H₃₇Rv with a MIC value 1.4 μM.³¹ Naidu et al., identified benzisoxazole-substituted sulfonyl piperazine (IV) which exhibited good inhibitory potential with a MIC value of 3.125 μg mL⁻¹ against Mtb H₃₇Rv.³² Sahoo et al., reported a series of isoxazole carboxylic acid methyl ester-based 2-substituted quinolone (V) with a MIC value 0.12 μg mL⁻¹.³³ Azzali et al., identified 2-aminothiazole linked with isoxazoles (VI and VII) showed good antitubercular activity with MIC₉₀ < 4 μg mL⁻¹.³⁴ Kancharlapalli et al., reported 5-(3,5-dimethoxyphenyl)-3-(isoxazol-5-yl)-4,5-dihydro-1H-pyrazolecarboxamide (VIII) with good antitubercular activity against Mtb H₃₇Rv with a MIC value 0.25 μg mL⁻¹.³⁵

Fig. 2. Representative examples of isoxazole and carboxamide-based molecules as Mtb and carbonic anhydrase inhibitors.

2. Rationale of the work

Carbonic anhydrase (CA) inhibitors are designed using a strategy of a tail approach to ensure specificity, potency, and optimal binding to the enzyme's active site.³⁶ This strategy focuses on Zn²⁺ binding, linkers, and aromatic tails. Sulphonamide groups are used to target the zinc ion, while linkers are incorporated to enhance interactions and ensure a proper fit within the enzyme's active site.³⁷ Additionally, targeted modifications to the aromatic core can improve activity and selectivity for specific mycobacterial carbonic anhydrase inhibitors (MtCAIs). Acetazolamide is a well-established carbonic anhydrase inhibitor (CAI) that targets multiple isoforms of the enzyme with limited selectivity. Due to its non-selective CA inhibition properties, acetazolamide is not ideal for selectively targeting specific carbonic anhydrase isoforms, such as those from Mycobacterium tuberculosis (MtCAs). To address this limitation, we propose structural modifications to acetazolamide and related benzene sulphonamides to enhance their selectivity toward MtCAs. Despite the central role of carbonic anhydrases in the physiology and survival of M. tuberculosis, these enzymes have remained largely unexplored in both drug discovery and clinical development.

To date, isoxazole and sulphonamide compounds have been studied individually as Mtb inhibitors. However, the evaluation of isoxazole sulphonamides as inhibitors of mycobacterial CAs (MtCAs) has not been thoroughly explored. In this study, we designed and synthesised a series of isoxazole-linked sulphonamide derivatives incorporating key pharmacophoric elements: a sulphonamide moiety, an amide linker, and a phenyl–isoxazole core (aromatic tail). As illustrated in Fig. 3, the synthesised compounds were evaluated for their inhibitory activity against two off-target human carbonic anhydrase isoforms (hCA I and hCA II), as well as two β-class CAs from M. tuberculosis (MtCA 1, and MtCA 3) to assess both potency and selectivity.

Fig. 3. Rationale for the design of the target molecule.

3. Results and discussion

3.1. Chemistry

The synthesis of substituted-5-phenylisoxazole-3-carboxylic acid 5a–h was completed in three steps, as outlined in Scheme 1. First, substituted acetophenones 1a–h were reacted with diethyl oxalate 2 in the presence of sodium ethoxide to produce substituted ethyl 2,4-dioxo-4-phenylbutanoates 3a–h. These intermediates underwent cyclisation with hydroxylamine hydrochloride under reflux, yielding substituted ethyl 5-phenylisoxazole-3-carboxylates 4a–h. Next, hydrolysis of these carboxylates with lithium hydroxide monohydrate produced the corresponding 5-(substituted phenyl)isoxazole-3-carboxylic acids 5a–h. This common intermediate served as the diversification point for analogue modification on the newly exposed amine. The first set of analogues was synthesised by 5-amino-1,3,4-thiadiazole-2-sulphonamide (7), which was synthesised from acetazolamide, on treatment with concentrated hydrochloric acid : water (1 : 10), resulting in the formation of a deacetylated acetazolamide intermediate 7 under reflux conditions. This intermediate was reacted with substituted phenyl isoxazole acids to afford the target thiadiazole amide derivatives (8a–c) as outlined in Scheme 1c. Further, the intermediate substituted phenyl isoxazole acids 5a–h were coupled with 4-aminobenzenesulphonamide (9), 3-aminobenzenesulphonamide (11), and other substituted anilines (13a–d) to form the final compounds 10a–f, 12a–f, and 14a–d, respectively.

Scheme 1. a–c. Synthetic route to 5-(substituted phenyl)-N′-phenylisoxazole-3-carboxamide, reagents and conditions: (i) NaOC₂H₅, THF, 0 °C–rt, 2 h (80–85%); (ii) NH₂OH·HCl, C₂H₅OH, reflux, 6 h (65–85%); (iii) LiOH·H₂O, EtOH : H₂O (4 : 1), rt, 4 h, (60–80%); (iv) conc. HCl, H₂O, reflux, 3–4 h (75%); (v) HATU, DMF, DIPEA, 0 °C–rt, 2 h (60–80%).

N-substituted phenyl-5-(thiophen-2-yl)isoxazole-3-carboxamide derivatives were synthesised via a multi-step synthesis, as depicted in Scheme 2. Initially, 1-(thiophen-2-yl)ethan-1-one 15 was treated with diethyl oxalate 2 in sodium ethoxide, yielding ethyl 2,4-dioxo-4-(thiophen-2-yl)butanoate 16. The intermediate 16 was refluxed with hydroxylamine hydrochloride in ethanol to obtain ethyl 5-(thiophen-2-yl)isoxazole-3-carboxylate 17, which was subsequently hydrolysed to the corresponding 5-(thiophen-2-yl)isoxazole-3-carboxylic acid 18 using lithium hydroxide monohydrate as the base. Finally, the carboxylic acid 18 was coupled with various substituted sulphonamides (9 and 11) to generate the final substituted amide compounds 19a–b, which were characterised using various spectral techniques.

Scheme 2. Synthetic route to optimised analogues: reagents and conditions: (i) NaOC₂H₅, THF, 0 °C–rt, 2 h (85%); (ii) NH₂OH·HCl, C₂H₅OH, reflux, 6 h (70%); (iii) LiOH·H₂O, EtOH : H₂O (4 : 1), rt, 4 h, (70%); (iv) HATU, DMF, DIPEA, 0 °C–rt, 2 h (45–50%).

3.2. Pharmacology/biology

3.2.1. Carbonic anhydrase inhibition

All the synthesised compounds (8a–c, 10a–f, 12a–f, 14a–d, and 19a–b) were evaluated for their potential inhibitory activity against carbonic anhydrase with acetazolamide as the positive control. These compounds were screened by using a stopped-flow CO₂ hydrase assay,³⁸ against a Mycobacterium tuberculosis isoform (MtCA 1 and MtCA 3) mentioned in Table 1, in comparison to the inhibitory effect on human isoforms (hCA I and hCA II), which were selected due to their abundance in the human organism and potential for off target effects.³⁹ Based on the acquired in vitro inhibitory constants (K_i), the structure–activity relationships (SARs) were systematically deduced from the data presented in Table 1. This detailed analysis allowed for the identification of critical pharmacophores. All the sulphonamide-containing compounds (8a–c, 10a–f, 12a–f, and 19a–b) exhibited good to moderate CA inhibition, whereas non-sulphonamide compounds (14a–d) did not show any CA inhibition properties. This again indicates the importance of the sulphonamide group for the CA inhibition. Most of the molecules (8a–c, 10a–f, 12a–f, & 19a–b) demonstrated strong inhibitory potentials toward the MtCA isoforms with K_i values ranging from 0.08 μM to 6.96 μM against MtCA 3, as shown in Table 1. The mycobacterial enzyme MtCA1 was inhibited by the compounds (8a–c, 10a–f, 12a–f, & 19a–b) reported here in Table 1, with constant K_i values in the range of 1.07–9.83 μM. Most of the compounds exhibited moderate to good inhibition against β-CA MtCA1 and also showed better and selective inhibition against MtCA 3, with K_i values ranging from 0.08 to 6.96 μM. Notably, compounds 12c, 12e & 19b have shown more potent activity against MtCA 3 with K_i values of 0.09, 0.09 & 0.08 μM, respectively, as compared to the standard acetazolamide (K_i = 0.10 μM). The order of activity towards MtCA 3 is 19b (thiophen-2-yl) > 12e (4-CH₂CH(CH₃)₂) > 12c (4-OCH₃) > 8b (4-CH₂CH(CH₃)₂) > 12f (3-NO₂) > 10a (H) > 12d (4-Cl).

Table 1. The inhibitory effects of compounds (8a–c, 10a–f, 12a–f, 14a–d, and 19a–b) and acetazolamide (positive control) were tested for two human CA isoforms, hCA I, II, and pathogenic Mycobacterium tuberculosis β-CAs MtCA 1 & 3, evaluated by stopped-flow CO₂ hydrase assay.

S. no.	Core	Compound code	R1	R2	K i a (μM)				Consensus log Pb
					hCA I	hCA II	Mt-β CA1	Mt-β CA3
1.	A	8a	–4Cl	—	0.65	0.08	1.07	0.38	1.39
2.	A	8b	–4CH2CH(CH3)2	—	0.57	0.04	1.20	0.18	1.94
3.	A	8c	–3,4diCl	—	0.88	0.11	1.33	0.40	1.69
4.	B	10a	–4H	–4SO2NH2	8.75	0.13	7.63	0.27	1.66
5.	B	10b	–4CH3	–4SO2NH2	10.64	0.12	7.43	0.36	2.00
6.	B	10c	–4OCH3	–4SO2NH2	10.76	0.21	7.62	0.32	1.73
7.	B	10d	–4Cl	–4SO2NH2	12.85	0.31	8.21	0.49	2.34
8.	B	10e	–4CH2CH(CH3)2	–4SO2NH2	6.35	0.11	8.12	1.87	3.01
9.	B	10f	–3NO2	–4SO2NH2	16.69	0.54	9.83	6.96	0.81
10.	B	12a	–4H	–3SO2NH2	18.59	2.68	4.73	1.06	1.58
11.	B	12b	–4CH3	–3SO2NH2	21.98	2.11	3.55	1.28	2.11
12.	B	12c	–4OCH 3	–3SO 2 NH 2	18.5	3.28	4.28	0.09	1.72
13.	B	12d	–4Cl	–3SO2NH2	28.54	4.97	5.86	0.31	2.33
14.	B	12e	–4CH 2 CH(CH 3 ) 2	–3SO 2 NH 2	16.36	2.37	3.28	0.09	3.01
15.	B	12f	–3NO 2	–3SO 2 NH 2	46.29	7.46	5.96	0.21	0.80
16.	B	14a	–3NO2	–4F	>100	>100	>100	>100	2.44
17.	B	14b	–3NO2	–4CH3	>100	>100	>100	>100	2.46
18.	B	14c	–3NO2	–4OCH3	>100	>100	>100	>100	2.09
19.	B	14d	–3NO2	–H	>100	>100	>100	>100	2.07
20.	C	19a	—	–4SO2NH2	5.26	0.10	6.59	0.26	1.57
21.	C	19b	—	–3SO 2 NH 2	16.94	3.27	3.89	0.08	1.56
22.	Acetazolamide		—	—	0.25	0.01	0.48	0.10	−0.98

^aMean from 3 different assays, by a stopped-flow technique (errors were in the range of ±5–10% of the reported values).

^bPredicted values.

Compounds 8a–c & 10a–f demonstrated potent hCA II inhibition (K_i values <1 μM), whereas hCA I showed K_i values >1 μM. Among these, compounds 8a and 8b demonstrated potent hCA II inhibition with K_i values of 0.08 μM and 0.04 μM, respectively. In addition, compounds 8a, 8b, and 8c displayed potent hCA I inhibition with K_i values of 0.62 μM, 0.57 μM, and 0.88 μM, respectively.

The SAR studies indicated that compounds containing thiadiazole-2-yl (8a–c) exhibited non-selective inhibition across all CAs tested—hCA I, hCA II, MtCA 1, and MtCA 3—likely due to the structural features shared with acetazolamide, as shown in Fig. 4. Compounds 10a–f, which feature 4-benzene SO₂NH₂ substitutions, show good selectivity for hCA II over hCA I and for MtCA 3 over MtCA 1. Conversely, compounds 12a–f and 19b, bearing 3-SO₂NH₂ groups, demonstrate a strong preference for MtCA 3 over hCA II and hCA I. Notably, compounds 12c, 12e, 12f, and 19b exhibit the highest selectivity for MtCA 3 among all CAs. In contrast, compounds 14a–d, which contain other substituted anilines, lack a sulphonamide group and demonstrate a complete loss of activity. Further important substitution patterns related to the activity are outlined in Fig. 4. In summary, the best compounds have shown better MtCA3 inhibition and selectivity compared to the standard drug acetazolamide.

Fig. 4. The structure–activity relationships (SARs) derived from assessed compounds on MtCA inhibition.

It may be noted that compounds 12c, 12e, 12f, and 19b, which showed strong inhibitory activity against MtCA 3, exhibited noticeably different consensus log P values: 1.72, 3.01, 0.80, and 1.56, respectively, as depicted in Table 1. This suggests that hydrophilic/hydrophobic balance alone is not the major determinant of inhibition. Instead, we believe that other factors, such as the orientation of the sulfonamide group within the active site, coordination with the zinc ion, hydrogen-bonding interactions, and the fit of the isoxazole scaffold, play a dominant role in showing inhibitory activity. For example, compound 19b, with a higher number of hydrogen bonding interactions involving both NH and S O of the sulfonamide group, showed better inhibition of MtCA 3 and in vitro activity too, when compared to other molecules of this series. However, further optimisation is required to correlate the enzyme inhibition and properties like lipophilicity of the molecules.

3.2.2. Selectivity indices

The selectivity indices (SIs) presented in Table 2 provide clear evidence of potent and selective inhibition of MtCA 3 over human carbonic anhydrase I, II (hCA I & II) by 17 compounds (8a–c, 10a–f, 12a–f, and 19a–b). The compounds displayed substantial selectivity of MtCA 3, with SI values ranging from 1.69 to 217.63 over hCA I. In contrast, all tested compounds exhibited comparatively lower selectivity over hCA II, with SI values ranging from 0.06 to 38.51.

Table 2. Selectivity indices (SIs) of the tested derivatives (8a–c, 10a–f, 12a–g, 19a–b) toward MtCAs 1&3 over hCA I, II.

S. no.	Core	Compound code	R1	R2	SI (μM)
					hCAI/MtCA 1	hCAII/MtCA 1	hCAI/MtCA3	hCAII/MtCA3
1.	A	8a	–4Cl	—	0.60	0.08	1.69	0.22
2.	A	8b	–4CH2CH(CH3)2	—	0.47	0.03	3.09	0.23
3.	A	8c	–3,4diCl	—	0.66	0.08	2.17	0.28
4.	B	10a	–4H	–4SO2NH2	1.14	0.01	31.87	0.49
5.	B	10b	–4CH3	–4SO2NH2	1.43	0.01	29.38	0.33
6.	B	10c	–4OCH3	–4SO2NH2	1.41	0.02	32.68	0.63
7.	B	10d	–4Cl	–4SO2NH2	1.56	0.03	26.13	0.63
8.	B	10e	–4CH2CH(CH3)2	–4SO2NH2	0.78	0.01	3.39	0.06
9.	B	10f	–3NO2	–4SO2NH2	1.69	0.05	2.39	0.07
10.	B	12a	–4H	–3SO2NH2	3.92	0.56	17.45	2.51
11	B	12b	–4CH3	–3SO2NH2	6.19	0.59	17.10	1.64
12.	B	12c	–4OCH 3	–3SO 2 NH 2	4.33	0.76	192.86	34.15
13.	B	12d	–4Cl	–3SO2NH2	4.86	0.84	91.56	15.95
14.	B	12e	–4CH 2 CH(CH 3 ) 2	–3SO 2 NH 2	4.97	0.72	175.81	25.54
15.	B	12f	–3NO 2	–3SO 2 NH 2	7.76	1.25	217.63	35.07
16.	C	19a	—	–4SO2NH2	0.79	0.01	19.84	0.40
17.	C	19b	—	–3SO 2 NH 2	4.35	0.84	199.55	38.51
18.	Acetazolamide		—	—	0.52	0.02	2.40	0.12

Regarding the selectivity for MtCA 3 over hCA I, all compounds except 12a–d demonstrated moderate to potent selectivity. The highest selectivity was observed for four compounds: 12c, 12e, 12f, and 19b, with selectivity index (SI) values ranging from 175.81 to 217.63 (Table 2). Among these, compounds 12f and 19b were the most selective, exhibiting SI values of 217.63 and 199.55, respectively. They were closely followed by 12c (SI = 192.86) and 12e (SI = 175.81). Additionally, compounds 10a–d, 12d, and 19a also exhibited potent selectivity with SI values ranging from SI = 19.84–91.56. In contrast, compounds 8b and 10e demonstrated moderate selectivity, followed by 8a, 8c, and 10f, which exhibited weak selectivity with SI values between 1.69–2.39.

With regards to the selectivity index (SI) of the synthesized compounds (8a–c, 10a–f, 12a–f, 19a–b) for MtCA3 over hCA II (Table 2), compound 19b showed an SI of 38.51, closely followed by compounds 12f (SI = 35.07) and 12c (SI = 34.15). It may be observed that, all tested compounds exhibited comparatively lower selectivity over hCA II, which may cause off target effects. Further optimisation is required to improve the selectivity for MtCA 3. However, 12f (SI = 217.63) and 19b (SI = 199.55), may be considered as the most potent compounds based on their selectivity for MtCA 3 over hCA I, when compared to the reference drug acetazolamide (SI = 2.40), which is a non-selective CA inhibitor.

3.2.3. Structure–activity relationship (SAR) based on selectivity indices

Insights into the SAR of this series of compounds concerning inhibitory activity and selectivity toward MtCA 3 over hCA I, can be discerned. Upon examining, the efficacy and selectivity order is 10c (4-OCH₃) > 10a (H) > 10b (4-CH₃) >10d (4-Cl) > 19a (thiophene-2-yl) > 10e (4-CH₂CH(CH₃)₂) > 12g (3,4-diF) > 10f (3-NO₂) in the case of sulphonamide substitution at the para position. meta-Substituted sulphonamides showed highest selectivity and activity in the following order: 12f (3-NO₂) > 19b (thiophen-2yl) > 12c (4-OCH₃) > 12e (4-CH₂CH(CH₃)₂) > 12d (4-Cl) > 12a (H) > 12b (4-CH₃) > 8b (4-CH₂CH(CH₃)₂-thiadoazol-2yl) > 8c (3,4-diCl-thiadoazol-2yl) > 8a (4-Cl-thiadoazol-2yl). With regards to the selectivity for MtCA 3 over hCA II, the order is as follows: 12g (3,4-diF) > 10d (4-Cl) > 10c (4-OCH₃) > 10a (H) > 19a (thiophen-2-yl) > 10b (4-CH3) > 10c (3,4-diCl) > 8b (4-CH₂CH(CH₃)₂-thiadoazol-2yl) > 8a (4-Cl-thiadoazol-2yl).

Overall, the relatively noteworthy low K_i values observed in the series featuring meta-substituted sulphonamides (19b) against Mtb isoform (MtCA 3), coupled with their impressive selectivity indices, especially those specific to hCA I/MtCA 3, hCA II/ MtCA 3, provide valuable insights for further optimising this chemotype of isoxazole-linked benzene sulphonamides as CAIs in future studies. These broad SAR observations are summarised in Fig. 5.

Fig. 5. Structure–activity relationship (SAR) for the designed compounds based on hCA I/MtCA 3 and hCA II/MtCA 3 selectivity indices.

3.2.4. In vitro antitubercular activity

The lead compounds 12c, 12e, 12f, and 19b were selected to evaluate their potential against the Mtb mc² 6230. The screening procedure was followed as per the Clinical and Laboratory Institute Guidelines, with MIC determined for varying concentrations (8–64 μg mL⁻¹), and was carried out in triplicate for each experiment to ensure reproducibility. The selective MtCA 3 inhibitors 12c, 12e, 12f, and 19b displayed notable Mtb mc² 6230 inhibition, with MICs ranging from 8–64 μg mL⁻¹, as summarised in Table 3. Compounds 19b showed the lowest MIC of 8 μg mL⁻¹, while compounds 12e and 12f exhibited moderate activity with MICs of 32 and 16 μg mL⁻¹, respectively. From the above results, we observed that 19b with the best MtCA 3 inhibition, and K_i of 0.08 μM, showed good inhibition against the mycobacterial cells with an MIC of 8 μg mL⁻¹. Based on these findings, it may be observed that though the compounds are showing very good MtCA inhibition, the in vitro activity is moderate. This may be due to the poor cell permeability or due to efflux. Further optimisation of the scaffold in the direction of enhancing the lipophilicity of the molecules may lead to improved activity.

Table 3. Antimycobacterial activity of compounds 12c, 12e, 12f, and 19b against Mycobacterium tuberculosis mc² 6230 by measuring their MICs in μg mL⁻¹.

S. no.	Compound code	MIC (mc2 6230) μg mL−1
1	12c	64
2	12e	32
3	12f	16
4	19b	8
5	Acetazolamide	>64
Standard	Isoniazid	0.06
	Rifampicin	0.03

3.2.5. In silico studies

In silico drug-likeness evaluation was carried out, using the Qikprop module of Schrödinger molecular modelling. The physicochemical properties relevant to drug-like properties, such as molecular weight, octanol/water coefficient, oral absorption percentage, hydrogen bond donors (<5), and hydrogen bond acceptors (<10), were assessed for the lead compounds 12e, 12f, and 19b. Table 4 presents the key predicted drug-likeness parameters and their recommended ranges. Based on the analysis of the calculated drug likeness parameters, it may be predicted that the lead compounds exhibit acceptable physicochemical properties and adhere to Lipinski's rule of five.

Table 4. Predicted drug-likeness properties for compounds 12e, 12f & 19b.

Parameters	Recommended range	12e	12f	19b
Rule of five	Maximum is 4	No violation	No violation	No violation
Molecular weight	≤500	399.46	388.35	349.38
Donor HB	0.0–6.0	2.0	2	2
Acceptor HB	2.0–20.0	6	8	6
No. of rotatable bonds	≤10	7	6	5
Lipophilicity consensus log Po/w	≤5	3.01	0.80	1.56
Water solubility log S (ESOL)	—	−4.41	−3.31	−3.12
TPSA (Å2)	≤140	123.67	169.49	151.91

3.3. Molecular docking study

Docking studies were performed to unveil the binding modes of 12f and 19b to elucidate the structural features and specific interactions that may underlie their distinct inhibitory profile (Table 1). Since the resolved 3D structure of MtCA3 is not currently available, the homologous model (MtCA3-HM) built for our previous investigations was used.⁴⁰

As observed by docking results, in the active site of the two dimeric MtCAs, residues from both monomers (A and B) are involved in ligand binding, following the literature on β-CAs.⁴⁰ Moreover, the (hetero)aromatic sulphonamide moiety of the ligands was found to bind deeply within the active site, with the deprotonated nitrogen atom (SO₂NH⁻) coordinating directly to the zinc ion. This interaction completes the tetrahedral coordination sphere of the metal, which includes residues C35(A)/C51(A)/C584(A), and C91(A)/C107(A)/C645(A) in MtCA1 and MtCA3, respectively as shown in Fig. 6. The detailed hydrogen bonding and hydrophobic interaction for compounds 12f and 19b are outlined in Table 5.^10,27,40,41

Fig. 6. Predicted binding modes of 12f (magenta) and 19b (orange) within the active sites of M. tuberculosis CA isoforms: (A and B) MtCA1 and (C and D) MtCA3-HM. H-bonds and π–π stacking interactions are shown as black and cyan dashed lines, respectively. Amino acid labels from different chains are colour-coded for distinction.

Table 5. Description of the main ligand 12f and 19b interactions within MtCA1 and MtCA3 active sites.

Complex	Interaction type	Ligand moiety	Residue/zinc ion	Distance
12f-MtCA1 (Fig. 6A)	Coordination	NH−	Zn2+	2.18 Å
	H-bond	NH−	D37 sidechain COO−	2.81 Å
	H-bond	NH−	G92 backbone NH	2.70 Å
	H-bond	C O	T95 sidechain OH	2.46 Å
19b-MtCA1 (Fig. 6B)	Coordination	NH−	Zn2+	2.14 Å
	H-bond	NH−	D37 sidechain COO−	2.82 Å
	H-bond	NH−	G92 backbone NH	2.69 Å
	H-bond	C O	T95 sidechain OH	2.29 Å
12f-MtCA3 (Fig. 6C)	Coordination	NH−	Zn2+	2.36 Å
	H-bond	NH−	D37 (sidechain COO−)	2.85 Å
	H-bond	S O	Q575 (sidechain NH2)	1.98 Å
	H-bond	S O	Y603 (sidechain OH)	2.30 Å
	H-bond	S O	G609 (backbone NH)	2.46 Å
	H-bond	C O	Q575 (sidechain NH2)	2.37 Å
	π–π stacking	Phenyl ring	F627 (sidechain aryl)	3.84 Å
19b-MtCA3 (Fig. 6D)	Coordination	NH−	Zn2+	2.39 Å
	H-bond	NH−	D37 (sidechain COO−)	2.78 Å
	H-bond	S O	Q575 (sidechain NH2)	1.98 Å
	H-bond	S O	Y603 (sidechain OH)	2.30 Å
	H-bond	S O	G609 (backbone NH)	2.45 Å
	H-bond	C O	Q575 (sidechain NH2)	2.23 Å
	π–π stacking	Phenyl ring	F627 (sidechain aryl)	3.84 Å

In the MtCA1 active site, zinc coordination is further stabilised by two H-bonds: one between the sidechain COO⁻ of D37(A) and the sulphonamide NH⁻ of the ligands, and another between the backbone NH of G92(A) and the same sulphonamide group. The lack of aromatic residues in helix 64–79, which could otherwise stabilize the benzene sulphonamide ring of the ligands via π–π interactions, may contribute to the higher K_i values observed for this isoform. Additionally, the amide C O group of both ligands forms an H-bond with the sidechain OH of T95(A), helping to orient the terminal aromatic ring of the tail toward residue P25(A) (Fig. 6A and B). Despite their similar binding modes, the marked difference in inhibitory profiles may be attributed to variations in electronic delocalisation between the π–π systems of the m-nitrobenzene moiety of 12f and the thiophene ring of 19b. These differences likely influence their respective abilities to engage in stable van der Waals (vdW) interactions within the active site.

In MtCA3-HM, the sulphonamide group is involved in an extensive hydrogen-bonding network: the sulphonamide NH and S O groups form four hydrogen interactions with residues Q575(B), D586(A), Y603(B), and G609(A). Additionally, the (hetero)aromatic ring engages in π–π stacking with the side chain of F627(B). Moreover, the amide C O group of both ligands forms an H-bond with the sidechain NH₂ of Q575(B), helping to orient the terminal aromatic ring of the tail toward the entrance of the active site (Fig. 6C and D). As observed with MtCA1, despite a similar binding mode of 12f and 19b, the notable difference in inhibitory activity against MtCA3 can be attributed to variations in electronic delocalisation within the π-systems of the thiophene ring and the m-nitrobenzene moiety.

4. Conclusion

In this study, we successfully synthesized a series of isoxazole-functionalized benzene sulphonamide derivatives and evaluated their inhibitory activity against carbonic anhydrases (CAs) from Mycobacterium tuberculosis (MtCAs) and human cytosolic CAs. Their K_i values for MtCA1 ranged from 1.07 to 9.83 μM, and for MtCA 3, from 0.08 to 6.92 μM, generally indicating potent inhibition. Among these, compound 19b emerged as the most potent inhibitor of MtCA 3, demonstrating a K_i of 3.89 μM for MtCA 1 and a remarkable 0.08 μM for MtCA 3. Compounds 12f and 19b demonstrated high selectivity for MtCA 3, showing 217.63 and 199.55 fold greater selectivity over hCA 1 and 38.07 and 38.51 fold over hCA II, respectively. Among them, compound 19b emerged as the most potent inhibitor of MtCA 3, with K_i values of 3.89 μM for MtCA 1 and 0.08 μM for MtCA 3, respectively. However, compounds 14a–d, which contain a free sulphonamide, exhibited diminished activity and lacked selectivity against all isoforms of carbonic anhydrase. Overall, 19b is the best compound for the inhibition and selectivity compared to the standard drug acetazolamide against MtCA 3. Furthermore, these 12c, 12e, 12f, and 19b active compounds also demonstrated inhibitory activity against Mtb, with MIC values ranging from 8 to 64 μg mL⁻¹, and these compounds comply with Lipinski's rule of five. In silico studies were performed to unveil the peculiar interaction of the most potent compounds 12f and 19b within two target MtCAs. These findings highlight the potential of these compounds as promising antimycobacterial agents, offering alternative mechanisms of action to combat drug-resistant strains.

Author contributions

Bandela Rani – conceptualisation, designing, synthesis, data curation, and writing – original draft. Anuradha Singampalli – review and editing. Sarvan Maddipatla – review and editing. Pardeep Kumar – review and editing. Sri Mounika Bellapukonda – review and editing. Rajendhar Ramavath – synthesis, review and editing. Lina S. Mahajan – review and editing. Srinivas Nanduri – review and editing, Divya Vemula – review and editing, Aman Dalal – biological studies. Nitin Pal Kalia – biological studies. Vasundhra Bhandari – review and editing, Paola Gratteri – molecular docking. Niccolò Paoletti – biological studies. Alessandro Bonardi – molecular docking. Claudiu T. Supuran – biological studies. Y. V. Madhavi – conceptualisation, supervision, writing, review, and editing.

Conflicts of interest

The authors declare no conflict of interest.

Data availability

The data supporting this article have been included as part of the supplementary information (SI).

Supplementary information is available. See DOI: https://doi.org/10.1039/d5cy00628g.