Design, synthesis, and molecular modeling of novel thiazolopyridine-based inhibitors of enoyl acyl carrier protein reductase (InhA) as anti-Mycobacterium tuberculosis agents

Abstract

The updated guidelines from the World Health Organization highlight that treatment options for multidrug-resistant tuberculosis (MDR-TB) remain limited due to the scarcity of effective drugs. As a result, there is an urgent necessity to develop novel or repurposed drugs that demonstrate efficacy against multidrug-resistant (MDR) strains. In this study, a new series of thiazole-pyridine hybrids were thoughtfully designed and synthesized to assess their potential as antitubercular agents. These compounds were specifically created to target enoyl acyl carrier protein reductase (InhA), a crucial enzyme in the pathogenesis of Mycobacterium tuberculosis. The majority of the compounds evaluated demonstrated substantial antitubercular activity, with minimum inhibitory concentrations (MIC) ranging from 0.5 to 3.9 μg mL⁻¹ against Mycobacterium tuberculosis H37Rv. Among them, compound 5a was the most effective, with an MIC of 0.5 μg mL⁻¹. Further evaluations of compound 5a demonstrated its ability to disrupt bacterial biofilms and its strong inhibition of InhA, with an IC₅₀ of 0.19 ± 0.008 μg ml⁻¹, demonstrating superior efficacy compared to triclosan, which was employed as the reference drug. Molecular docking and dynamics analyses demonstrated that the pyridine ring and thiazole group are essential for binding affinity, and the pyridine-thiazole framework in compound 5a formed stable interactions within the active site of InhA.

The updated guidelines from the World Health Organization highlight that treatment options for multidrug-resistant tuberculosis (MDR-TB) remain limited due to the scarcity of effective drugs.

1. Introduction

Tuberculosis (TB) is a long-lasting infectious illness resulting from the bacterium Mycobacterium tuberculosis (Mtb), historically known as the “great white plague”.^1,2 Mtb is a rod-shaped bacterium that requires oxygen to survive and characterized by a cell wall rich in lipids, which makes it highly lipophilic and resistant to alcohols, acids, alkalis, and various disinfectants.³ According to the latest World Health Organization (WHO) Global Tuberculosis Report 2025, the global incidence of tuberculosis (TB) decreased by approximately 2% between 2023 and 2024, while TB-related mortality declined by 3%. These reductions indicate a sustained recovery of critical health services following the disruptions caused by the COVID-19 pandemic. Despite these improvements, tuberculosis remains one of the leading causes of death worldwide, disproportionately impacting low-income countries. Furthermore, TB continues to be the foremost cause of mortality among individuals living with HIV and constitutes a significant contributor to deaths associated with antimicrobial resistance.^4–6

In spite of the extensive research efforts in microbiology over recent years, infection with Mycobacterium tuberculosis (Mtb) remains a significant worldwide health threat. This persistent issue is largely attributable to the increasing prevalence of drug-resistant Mtb strains and the limited effectiveness of current tuberculosis (TB) therapies, both of which have intensified the worldwide TB burden. The administration of complex multidrug regimens frequently results in severe adverse effects, underscoring the urgent necessity for innovative therapeutic approaches. Conventional antitubercular agents comprising isoniazid, ethambutol, rifampicin, fluoroquinolones, and linezolid are constrained by problems such as the development of resistance to pharmaceutical agents and the requirement for prolonged treatment courses.^7–9

In response to the escalating threat posed by multidrug-resistant (MDR-TB) and extensively drug-resistant tuberculosis (XDR-TB), a combination therapy comprising linezolid, bedaquiline, and pretomanid received approval in 2019.¹⁰ In recent times, Dorman and coauthors showed that a four-month treatment regimen incorporating rifapentine and moxifloxacin is non-inferior to the conventional six-month protocol, thereby presenting the possibility of reduced treatment duration.¹¹ Moreover, the fatty acid biosynthesis pathway is essential for Mycobacterium tuberculosis survival, directly producing the mycolic acids that form the cell wall's impermeable, protective barrier.^12,13 Inhibiting this pathway is a proven strategy; isoniazid, a frontline drug, acts by blocking the InhA enzyme within this system.^14,15 However, resistance frequently arises from mutations that prevent isoniazid's activation. Therefore, developing direct InhA inhibitors, which bypass this activation step, represents a promising approach to overcome resistance and treat drug-resistant TB.^16–19

The nitrogen-containing heterocyclic compound pyridine, along with its various analogues, holds a significant role as a valuable source of pharmacologically active agents within the domain of medicinal chemistry research;^20–23 considerable research efforts have been devoted to investigating its potential as a structural framework for antitubercular agents.²⁴ These investigations have predominantly concentrated on the design of molecular architectures capable of eliciting potent anti-tuberculosis effects. A notable example is isoniazid I (Fig. 1), a pyridine-based molecule developed and evaluated as an InhA inhibitor, which has demonstrated significant antitubercular efficacy.²⁵ NITD-916 II (Fig. 1), a derivative of 4-hydroxy-2-pyridone, functions as an orally bioavailable inhibitor of InhA, with the value of IC₅₀ = 570 nM.²⁶ Furthermore, the natural product pyridomycin III (Fig. 1), which contains a pyridine moiety, displays strong antimycobacterial activity primarily through the specific inhibition of the essential bacterial enzyme InhA.²⁷ Thiazole and its derivatives represent a significant class of compounds with considerable potential as antitubercular agents, primarily due to their target specificity.²⁸ In addition to their application in tuberculosis therapy, compounds containing the thiazole moiety have exhibited a wide range of pharmacological activities.^29–35 From a structural perspective, the thiazole ring is analogous to that found in nitazoxanide (NTZ) IV (Fig. 1), an orally administered, FDA-approved medication used to treat protozoal infections, which has also demonstrated substantial inhibition of intracellular Mycobacterium tuberculosis (Mtb) growth.³⁶ Furthermore, faropenem V (Fig. 1), an orally bioavailable β-lactam antibiotic, has been shown to effectively eradicate Mtb independently of β-lactamase inhibitors.³⁷ Concurrently, the pharmacological relevance of hydrazone functional group R₁R₁C NNH₂VI (Fig. 1)³⁸ and azo functionalized (–N N–) derivatives VII (Fig. 1)³⁹ has been explored with respect to their antibacterial and antitubercular properties, particularly their capacity to inhibit InhA.

Fig. 1. Previously documented antitubercular agents containing bioactive scaffolds, including pyridine derivatives (I–III), thiazole derivatives (IV & V), and hydrazone moieties (VI). Additionally, the figure depicts the newly synthesized target compounds of the current study.

Building upon these insights, our previous work in this area^40–42 and expanding upon investigations into naturally derived bioactive compounds, we developed novel anti-tuberculosis agents utilizing pyridine-based drugs, such as pyridomycin III, as structural templates. This study centers on the chemical modification of thiosemicarbazide (TSC) derivatives—compounds noted for their efficacy against Mycobacterium tuberculosis, including drug-resistant strains and biofilm formations—through the strategic introduction of halogen substituents aimed at enhancing antimicrobial activity while minimizing toxicity. Additionally, an azo functional group, known for its anti-mycobacterial properties, was incorporated and subsequently hybridized with a bioactive thiazole moiety to synthesize innovative hybrid molecules (see Fig. 1). This approach was informed by established pharmacophoric features associated with InhA inhibition as it can form a polar contact with the NAD cofactor (Fig. 2). The synthesized hybrid molecules were subsequently assessed for their antimycobacterial activity and their potential to inhibit the InhA enzyme. Additionally, molecular docking and molecular dynamics simulations were performed to elucidate the binding interactions and stability of the most promising candidates within the enzyme's active site.

Fig. 2. The pharmacophoric characteristics of the most potent compounds and their binding interactions within the active site of the InhA enzyme.

2. Results and discussion

2.1. Chemistry

The synthesis of the target compounds 5a–f is illustrated in Scheme 1. Initially, thiosemicarbazone 3 was obtained by reacting 2-acetylpyridine (1) with thiosemicarbazide (2). Subsequently, the key intermediate thiosemicarbazone 3 underwent heterocyclization by reacting with hydrazonoyl halide derivatives 4a–f in dioxane, using triethylamine (TEA) as a catalyst, leading to the creation of the corresponding thiazole derivatives 5a–f. the new synthesized compounds were verified through detailed spectroscopic analyses, including proton nuclear magnetic resonance (¹H-NMR), carbon-13 nuclear magnetic resonance (¹³C-NMR), and high-resolution time-of-flight electrospray ionization mass spectrometry (HR-ESI-MS), as described in the Experimental section, with supporting data provided in the SI.

Scheme 1. Synthesis of thiazole derivatives 5a–f.

2.2. Biological activity

2.2.1. Evaluation of antimycobacterial properties

The synthesized compounds 5a–f were investigated for their antitubercular efficacy and cytotoxicity against Mycobacterium tuberculosis H37Rv, and against rapidly growing non-tuberculous mycobacterial species, including Mycobacterium abscessus. The initial screening was performed at a constant concentration of 125 μg mL⁻¹ to evaluate their inhibitory potential. The compounds exhibited no activity against M. abscessus;⁴³ however, these compounds exhibited inhibitory effects against M. tuberculosis when compared to the reference values of standard drugs ETH and INH, which were obtained exclusively from the literature and not from the present study,⁴⁴ as detailed in Table 1. For those derivatives exhibiting activity against M. tuberculosis, the minimum inhibitory concentration (MIC) was determined using the microplate alamar blue assay (MABA). Results indicated that the synthesized compounds effectively inhibited M. tuberculosis growth at concentrations ranging from 0.5 to 3.9 μg mL⁻¹. Subsequent to the primary screening, selected compounds were subjected to further cytotoxicity evaluation in accordance with international standards (ISO 10993-5:2009(E)), utilizing L929 fibroblast cells and the MTT assay.⁴⁵ For compound 5a, the IC₅₀ value was determined at 11.5 μg mL⁻¹, for compounds 5b and 5c at approximately 40 μg mL⁻¹, while compound 5e showed no toxicity at the tested concentrations up to 100 μg mL⁻¹. The IC₅₀ values were 10.3-fold (5b), 23-fold (5a), 19.1-fold (5c), or even more than 52.6-fold (5e) higher than the determined MIC values for the same compounds, as presented in Table 1.

Table 1. The antimycobacterial effect of target compounds 5a–f.

Compound	MIC μg ml−1M. tuberculosis	MIC [μg ml−1] M. abscessus	IC50 - L929 [μg ml−1]	IC50/MIC Mtb	BC [μg mL−1]
5a	0.5	>125	11.5	23	3
5b	3.9	>125	40	10	40
5c	1.9	>125	36.2	19	20
5d	>125	>125	—	—	—
5e	1.9	>125	>100	>50	40
5f	>125	>125	—	—	—
INH	0.05	—	—		—
ETH	0.25–16	—	—	—	—

Next, minimum bactericidal concentrations (BC) were determined for selected compounds. For this purpose, a recombinant M. tuberculosis strain with stable expression of the luciferase gene (Mtb_Lux) was prepared. In control studies, the relationship between the number of viable bacteria (CFU) and luminescence level was analyzed. Using basic antitubercular drugs (rifampicin), it was demonstrated that bactericidal action leading to a reduction in the number of viable mycobacteria causes a decrease in luminescence. This prepared and validated strain was used to determine the minimum bactericidal concentrations of the tested compounds based on luminescence level analysis. Among the tested compounds, compound 5a showed bactericidal effect at the lowest concentrations (3 μg mL⁻¹). Compounds 5b, 5c, and 5e showed bactericidal effects against tuberculosis mycobacteria at concentrations of 40, 20, and 40 μg mL⁻¹, respectively (Fig. 3).

Fig. 3. Evaluation of the bactericidal effect against M. tuberculosis of the tested compounds through assessment of luminescence levels. The tested compounds and concentrations used are provided in the legend description. Statistical analysis was conducted using the independent two-sample Student's t-test (Welch's) to compare the untreated M. tuberculosis control group (M.tb_control) with bacilli treated with the indicated compounds at specified time intervals. The statistical analysis yielded a p-value of p < 0.0001 for all pair comparisons in 24, 48, 72, and 168 time intervals. The visualization was conducted utilizing GraphPad Prism 9 version 9.3.1.

Furthermore, for compound 5a, the BC value was additionally verified using the colony-forming unit (CFU) counting method (CFU), Fig. 4.

Fig. 4. Bactericidal effect of compound 5a. Compound 5a was used at concentrations of 3 μg mL⁻¹ for 7 and 14 days. The quantity of viable bacteria was evaluated through colony-forming unit (CFU) analysis. Statistical analysis was conducted using the independent two-sample Student's t-test (Welch's) to compare the untreated M. tuberculosis control group (Rv_control) with bacilli treated with the indicated compounds at specified time intervals. The corrected p-value was less than 0.0001 for paired comparisons denoted by asterisks. The visualization was conducted utilizing GraphPad Prism 9 version 9.3.1.

2.2.2. Evaluating the efficacy of substances on mycobacterial biofilms

Mycobacterial biofilms represent a critical defense mechanism that poses significant challenges to therapeutic interventions, forming robust microbial communities with enhanced resistance to both antimicrobial agents and host immune responses.⁴⁶ We assessed the antimicrobial efficacy of all selected compounds against M. tuberculosis biofilms. Bacterial cultures were grown in Sauton's medium for five weeks to establish mature biofilms, which were subsequently treated with compounds at their respective bactericidal concentrations. Viability analysis demonstrated statistically significant but limited bactericidal activity. The findings show that compounds 5a, 5b, and 5c exhibit modest antibiofilm potential; however, they demonstrate limited ability to effectively penetrate these complex microbial architectures (Fig. 5). In contrast, compound 5e applied at 10× MIC concentration failed to produce statistically significant bactericidal effects against biofilm-associated bacilli.

Fig. 5. The antibiofilm activity of selected compounds was evaluated against established M. tuberculosis biofilms. The bacterial load in treated biofilms was compared to untreated controls (M. tuberculosis biofilm without compound treatment) using ordinary one-way ANOVA. Compounds tested included 5a, 5b, 5c, and 5e in concentrations 5, 20, 10, and 20 μg mL⁻¹, respectively. Statistical significance (P < 0.05) is indicated on the graph where applicable. Data analysis and visualization were performed using GraphPad Prism version 9.3.1.

2.2.3. Evaluation of drug interactions between a selected compound and rifampicin

Tuberculosis treatment relies on multi-drug regimens, making the assessment of interactions between potential therapeutic agents and existing anti-tuberculosis drugs critically important. Given that rifampicin (RMP) is a basis of tuberculosis therapy, we investigated whether concurrent administration of compound 5a with rifampicin would result in synergistic, additive, or antagonistic effects, or demonstrate no interaction. Drug interactions were evaluated using the checkerboard assay method. The analysis revealed no significant interaction between compound 5a and rifampicin, with a fractional inhibitory concentration index (FICI) value of 1.25, indicating an indifferent relationship between the two agents.

2.2.4. Suppression of the Mycobacterium tuberculosis InhA enzyme activity

To evaluate the effectiveness of the most promising candidate, derivative 5a underwent additional testing to determine its inhibitory activity against the InhA enzyme. The results displayed that this derivative 5a unveiled substantial inhibition of InhA, demonstrating approximately threefold greater potency compared to the reference inhibitor, triclosan. In particular, the thiazole pyridine derivative 5a showed a potent inhibitory effect, with an IC₅₀ value of 0.19 ± 0.008 μg ml⁻¹, which is significantly lower than the IC₅₀ of triclosan at 0.62 ± 0.02 μg ml⁻¹, thereby indicating enhanced inhibitory efficacy, Table 2.

Table 2. The inhibitory effect of compound 5a on InhA.

Compound	IC50 (μg ml−1)
5a	0.19 ± 0.008
Triclosan	0.62 ± 0.02

Structure–activity relationship (SAR)

The structure–activity relationship (SAR) analysis demonstrated a consistent pattern wherein para-unsubstituted (H) derivatives exhibited enhanced biological activity. Additionally, para-substitution with electron-donating groups such as methyl (CH₃) and methoxy (OCH₃) further improved efficacy. In contrast, electron-withdrawing substituents showed variable effects: fluorine (F) substitution enhanced activity, whereas bromine (Br) and chlorine (Cl) substitutions led to decreased activity. These findings suggest that an increased electron density on the aromatic ring facilitates favorable π–π stacking and hydrogen-bonding interactions within the enzyme active sites, thereby augmenting target binding affinity. This interpretation is corroborated by both experimental inhibition data and theoretical molecular docking results.

All synthesized derivatives exhibited varying levels of anti-tubercular activity, with a clear correlation between biological efficacy and the electronic nature of the aromatic substituents. Notably, compound 5a, bearing a para-hydrogen substituent (4-H), demonstrated the highest potency with a minimum inhibitory concentration (MIC) of 0.5 μg mL⁻¹. Compounds substituted with electron-donating groups at the para position, specifically 5b (4-CH₃) and 5e (4-OCH₃), also showed significant activity, with MIC values of 3.9 and 1.9 μg mL⁻¹, respectively. Conversely, derivatives containing electron-withdrawing groups such as 5c (4-F), 5d (4-Br), and 5f (2,4-Cl) generally exhibited reduced anti-TB efficacy, with the exception of 5c, which displayed excellent activity (MIC = 1.9 μg mL⁻¹). These results imply that increased electronegativity at the para position tends to diminish anti-mycobacterium potency.

2.3. Molecular docking study

Our newly designed compounds were synthesized to be promising M. tuberculosis InhA inhibitors; in this study the candidates were screened against InhA to evaluate their possible affinity and anti-TB activity. The results of the top-fitted compound 5a is illustrated in Table 3.

Table 3. Molecular docking analysis of the newly synthesized 5a and reference ligand triclosan against M. tuberculosis enoyl reductase (InhA).

Extracted compounds	M. tuberculosis enoyl reductase (InhA)
	RMSD value (Å)	Affinity score (kcal mol−1)	(InhA) IC50 μg ml−1
Compound 5a	1.21	−8.03	0.19 ± 0.008
Triclosan	1.54	−7.18	0.62 ± 0.026
The co-crystallized ligand	0.38	−7.96	NA

In the present study, the proposed interaction mode of compound 5a exhibited a promising binding interaction against InhA with an affinity score of −8.03 kcal mol⁻¹. 5a interacted with a pharmacophore amino acid Tyr158 by two hydrophobic π-sulfur and π-alkyl interactions. Additionally, the interaction was established by forming an extra eight hydrophobic π-alkyl interactions with Ala157, Ile215, Met103, Leu218, Ile202, Met199, Phe149, and Ile194. Moreover, a hydrogen bond was observed with essential amino acid Pro156 with a distance of 2.92 Å, Fig. 6. Meanwhile, the diaryl ether inhibitor triclosan exhibited an affinity score of −7.18 kcal mol⁻¹ against enoyl reductase (InhA). Ten hydrophobic π-alkyl and π–π stacked interactions were formed with Ala191, Ile202, Met199, MET103, Met155, Ile215, and Phe149. Additionally, a halogen interaction was noted between the chloro group and Pro156 amino acid side chain, Fig. 7.

Fig. 6. 3D figure of the proposed binding mode of derivative 5a against M. tuberculosis enoyl reductase (InhA). Amino acids colored yellow and the ligand colored turquoise.

Fig. 7. 3D figure of the proposed binding mode of Triclosan against M. tuberculosis enoyl reductase (InhA). Amino acids colored yellow and triclosan rose.

Furthermore, the co-crystallized ligand complexes with enoyl reductase (InhA) formed some hydrophobic interactions with hydrophobic side chains of amino acids in the target pocket and the pattern of the interaction was supported by one hydrogen bond with pharmacophoric amino acid Tyr158 with a distance of 1.71 Å, Fig. 8. Also, the RMSD value for the re-docked co-crystal ligand compared to the crystal pose was 0.38 Å, Fig. 9.

Fig. 8. 3D figure of the proposed binding mode of the co-crystallized ligand against M. tuberculosis enoyl reductase (InhA). Amino acids colored yellow and ligand colored rose.

Fig. 9. The re-docked pose of the co-crystallized ligand of enoyl reductase (InhA). Original pose was colored green and re-docked pose was colored rose with a RMSD value of 0.38 Å.

From the previous illustration, our synthesized compounds, especially, compound 5a demonstrated a high binding affinity against M. tuberculosis enoyl reductase (InhA), which play main roles in bacterium's survival, as it is essential for the synthesis of mycolic acids, a key component of the mycobacterial cell wall. Moreover, compound 5a exhibited notable fitting and promising binding affinity toward enoyl reductase (InhA) with an affinity score of −9.36 Kcal mol⁻¹ and InhA/IC₅₀ 0.19 ± 0.008 μg ml⁻¹, in which the results revealed that our candidate consistently displayed high binding affinities and activity compared to their respective reference inhibitor as shown in Table 3.

2.4. Investigation using molecular dynamics (MD) simulation study

A molecular dynamics simulation was conducted for a duration of 100 nanoseconds to assess the stability of molecules for compound 5a within the active site of the enoyl reductase (InhA) target. The movements of ligand within the target pocket were measured by root mean square deviation (RMSD) to evaluate the complex. Finally, candidate 5a interactions were also analyzed and evaluated in detail.

2.4.1. Analysis of protein and ligand RMSD and RMSF

In this study, compound 5a demonstrated the highest binding affinity and inhibitory activity against enoyl reductase (InhA). The compound established multiple hydrophobic and hydrogen bond interactions with the target protein, contributing to its strong binding stability. Therefore, 5a was selected for further molecular dynamics (MD) simulation studies. The structural stability of the protein was assessed by tracking the positions of the Cα atoms (represented by the blue line) relative to their initial coordinates. As depicted in Fig. 10, the 5a compound demonstrated minimal fluctuations within the InhA binding pocket during the initial 0–30 ns interval. Concurrently, the target protein exhibited significant loop movements in the regions encompassing amino acid residues 50–70, 150–160, and 190–220, resulting in substantial conformational alterations of the protein backbone but no effect on the protein target interactions; after 30 ns compound 5a showed stability until the end of simulation time with a RMSD value between 5.6 Å and 6.5 Å, Fig. 10.

Fig. 10. The RMSD and RMSF of 5a against InhA over 100 ns.

2.4.2. Analysis of protein ligand interactions

2.4.2.1. Histogram of protein ligand interaction analysis

The target compound 5a displayed notable stability with InhA, and many interactions were observed with the target pocket pharmacophoric amino acids such as Tyr185. Moreover, 5a formed π-interactions with the following residues: Ile16 (∼10%), Ile21 (∼20%), Phe97 (∼20%), Met147 (∼10%), Phe149 (∼40%), Tyr158 (∼20%), and Ala198 (∼10%), as shown in Fig. 11. Additionally, the target compound 5a interacted with InhA by many polar hydrogen bonds and a water-bridged H-bond, where crystal water molecules form a link between the protein residues and ligands as illustrated with residues Ser94 (∼5%), Gly96 (∼20%), Met98 (∼10%), Tyr158 (∼35%), Lys165 (∼30%), Ile194 (∼50%), and Thr196 (∼40%).

Fig. 11. Histogram analysis illustrating the binding interactions between compound 5a and the InhA enzyme throughout the 100-nanosecond simulation period.

To analyze the frequency of interactions, a heat map was employed to visualize the number of interactions over time, as illustrated in Fig. 12, where darker colors correspond to a higher frequency of interactions. The heat map analysis revealed that the protein InhA exhibited up to eight interactions in its most prevalent conformations. The amino acid residues of InhA that demonstrated the highest interaction frequency with compound 5a were identified as Ile21, Gly96, Phe97, Phe149, Tyr158, Lys165, and Ile194.

Fig. 12. Heat map illustrating the cumulative interactions between 5a and InhA throughout the duration of the simulation.

3. Conclusion

In summary, this study synthesized a series of small molecules comprising pyridine moieties tethered to thiazole derivatives, aiming to identify potential antitubercular agents. The synthesized compounds were evaluated for their bactericidal activity against Mycobacterium tuberculosis and Mycobacterium abscessus. Notably, derivatives 5a, 5b, 5c, and 5e exhibited minimum inhibitory concentration (MIC) values of 0.5, 3.9, 1.9, and 1.9 μg mL⁻¹, respectively, against M. tuberculosis but not M. abscessus. Nontuberculous mycobacteria, including opportunistic pathogens like M. abscessus, are known for their phenotypic drug resistance, even to drugs effective in treating tuberculosis. Therefore, screening new bactericidal compounds for mycobacterium should encompass not only M. tuberculosis but also nontuberculous mycobacteria (NTM). Subsequent assessments of antibiofilm activity revealed that compounds 5a, 5b, and 5c possess moderate antibiofilm effects, although their capacity to penetrate complex biofilm structures was limited. Furthermore, a strain overexpressing the InhA enzyme demonstrated sensitivity to the majority of active compound, with compound 5a exhibiting an IC₅₀ value of 0.19 μg mL⁻¹. To elucidate the molecular interactions underlying enzyme inhibition, docking studies identified critical hydrogen bond interactions between the compound and residues Tyr158 and Met161 within the InhA active site. Molecular docking simulations indicated that compound 5a shares comparable binding affinities and interaction profiles with previously characterized InhA inhibitors, underscoring the significance of hydrogen and halogen bonding, alongside other stabilizing interactions, in maintaining ligand–enzyme complex stability. Molecular dynamics (MD) analyses further corroborated the binding mode predicted by docking, supporting the structural stability of the InhA–compound complex. Collectively, these findings provide a foundation for future research aimed at optimizing these derivatives to enhance their potency and selectivity against mycobacterial InhA. Prospective studies may also explore combination therapies to evaluate potential synergistic effects between these novel compounds and established antitubercular agents. Additionally, in vivo investigations are imperative to assess the safety, pharmacokinetics, and therapeutic efficacy of the most promising candidates within relevant tuberculosis models. Such research endeavors hold considerable promise for advancing the treatment of mycobacterial infections.

4. Experimental section

4.1. Chemistry

Melting points were measured using the Electrothermal IA 9000 apparatus, and no corrections were applied to these values. High-resolution mass spectrometry (HR-EI-MS) data for all compounds were acquired utilizing the JEOL JMS-700 instrument, (Tokyo, Japan). Nuclear magnetic resonance (NMR) analysis, encompassing both ¹H and ¹³C-NMR spectra, was conducted with Bruker 500 NMR spectrometers located at the Faculty of Pharmaceutical Science, Tokushima Bunri University, Japan. Chemical shifts are reported in δ (ppm), while coupling constants are expressed in Hz. Thin-layer chromatography (TLC) was employed to monitor the reactions, utilizing silica gel on aluminum sheets (60 F254, Merck) with a chloroform/methanol (9.8 : 0.2 v/v) eluent, which was subsequently visualized using iodine–potassium spray. Compound 3 was synthesized according to previously established methods.⁴⁷

Methodology for the synthesis of the target compounds 5a–f

A solution containing equimolar amounts (0.001 mol each) of thiosemicarbazone 3 and hydrazonyl chloride derivatives 4a–f (commercially procured) was prepared in dioxane (15 mL). A catalytic quantity of triethylamine, about 3 to 5 drops, was added to the reaction mixture, which was then refluxed for 8 hours. After cooling, the precipitated product was isolated via filtration and purified by recrystallization from ethanol, affording the desired derivatives 5a–f in pure form.

4-Methyl-5-(phenyldiazenyl)-2-(2-(1-(pyridin-2-yl)ethylidene)hydrazinyl)thiazole (5a)

Red powder, m.p. 127–128 °C, yield (74%), HPLC: R_T 6.09 min (purity: 99.70%); ¹H NMR (500 MHz, CDCl₃): δ = 2.44 (s, 3H, CH₃), 2.67 (s, 3H, CH₃-thiazole), 7.03 (t, 1H, J = 7.5 Hz, H–Ar), 7.21 (d, 2H, J = 8.0 Hz, H–Ar), 7.33–7.38 (m, 3H, H–Ar), 7.74–7.77 (m, 1H, H–Ar), 8.27 (d, 1H, J = 8.0 Hz, H–Ar), 8.68 (d, 1H, J = 8.5 Hz, H–Ar), 11.90 (s, 1H, NH). ¹³C NMR (126 MHz, CDCl₃) δ 14.70 (CH₃), 16.80 (CH₃), 114.17, 121.73, 123.05, 124.69, 129.57, 136.19, 140.53, 142.54, 148.94, 155.43, 166.98, 178.19 (C N). LREIMS: 336.0; HREIMS: 336.1159 (calcd. for C₁₇H₁₆N₆S; 336.1157).

4-Methyl-2-(2-(1-(pyridin-2-yl)ethylidene)hydrazinyl)-5-(p-tolyldiazenyl)thiazole (5b)

Red powder, m.p. 182–184 °C, yield (81%). ¹H NMR (500 MHz, CDCl₃): δ = 2.32 (s, 3H, CH₃), 2.42 (s, 3H, CH₃), 2.67 (s, 3H, CH₃-thiazole), 7.10–7.15 (m, 4H, H–Ar), 7.31–7.34 (m, 1H, H–Ar), 7.73–7.76 (m, 1H, H–Ar), 8.28 (d, 1H, J = 8.0 Hz, H–Ar), 8.66 (d, 1H, J = 8.5 Hz, H–Ar), 12.05 (s, 1H, NH). ¹³C NMR (126 MHz, CDCl₃) δ = 14.13 (CH₃), 16.78 (CH₃), 20.78 (CH₃-ph), 114.15, 121.70, 124.62, 129.67, 130.07, 132.70, 136.14, 139.95, 140.26, 148.91, 155.50, 166.78, 178.04 (C N). LREIMS: 350; HREIMS: 350.1317 (calcd. for C₁₈H₁₈N₆S; 350.1314).

5-((4-Fluorophenyl)diazenyl)-4-methyl-2-(2-(1-(pyridin-2-yl)ethylidene)hydrazinyl)thiazole (5c)

Red powder, m.p. 151–152 °C, yield (88%), HPLC: R_T 7.66 min (purity: 99.83%); ¹H NMR (500 MHz, CDCl₃): δ = 2.43 (s, 3H, CH₃), 2.68 (s, 3H, CH₃-thiazole), 7.03–7.06 (m, 2H, H–Ar), 7.14–7.19 (m, 2H, H–Ar), 7.32–7.39 (m, 1H, H–Ar), 7.33–7.39 (m, 1H, H–Ar), 8.24 (d, 1H, J = 8.0 Hz, H–Ar), 8.66 (d, 1H, J = 7.5 Hz, H–Ar), 11.94 (s, 1H, NH). ¹³C NMR (126 MHz, CDCl₃) = δ 14.70 (CH₃), 16.78 (CH₃), 116.19, 116.37, 121.72, 124.70, 136.17, 138.95, 140.61, 148.92, 155.42, 157.90, 159.83, 166.98, 178.09 (C N). LREIMS: 354; HREIMS: 354.1070 (calcd. for C₁₇H₁₅FN₆S; 354.1063).

5-((4-Bromophenyl)diazenyl)-4-methyl-2-(2-(1-(pyridin-2-yl)ethylidene)hydrazinyl)thiazole (5d)

Red powder, m.p. 165–166 °C, yield (79%). ¹H NMR (500 MHz, CDCl₃): δ = 2.49 (s, 3H, CH₃), 2.69 (s, 3H, CH₃-thiazole), 7.09 (d, 2H, J = 9.0 Hz, H–Ar), 7.33–7.35 (m, 1H, H–Ar), 7.43 (d, 2H, J = 8.5 Hz, H–Ar), 7.73–7.77 (m, 1H, H–Ar), 8.24 (d, 1H, J = 8.0 Hz, H–Ar), 8.68 (d, 1H, J = 7.5 Hz, H–Ar), 11.76 (s, 1H, NH). ¹³C NMR (126 MHz, CDCl₃) δ 14.75 (CH₃), 16.80 (CH₃), 115.24, 115.66, 121.76, 124.76, 132.44, 136.21, 141.34, 141.73, 148.92, 155.36, 167.16, 178.13 (C N). LREIMS: 416; HREIMS: 414.0262 (calcd. for C₁₇H₁₅BrN₆S; 414.0262).

5-((4-Methoxyphenyl)diazenyl)-4-methyl-2-(2-(1-(pyridin-2-yl)ethylidene)hydrazinyl)thiazole (5e)

Red powder, m.p. 124–126 °C, yield (76%), HPLC: R_T 8.25 min (purity: 99.77%); ¹H NMR (500 MHz, CDCl₃): δ = 2.44 (s, 3H, CH₃), 2.68 (s, 3H, CH₃-thiazole), 3.80 (s, 3H, OCH₃), 6.89 (d, 2H, J = 9.0 Hz, H–Ar), 7.15 (d, 2H, J = 8.5 Hz, H–Ar), 7.31–7.34 (m, 1H, H–Ar), 7.71–7.78 (m, 1H, H–Ar), 8.27 (d, 1H, J = 8.0 Hz, H–Ar), 8.66 (d, 1H, J = 7.5 Hz, H–Ar), 11.81 (s, 1H, NH). ¹³C NMR (126 MHz, CDCl₃) δ = 14.64 (CH₃), 16.76 (CH₃), 55.65 (OCH₃), 114.91, 115.44, 121.68, 123.88, 124.61, 136.14, 136.31, 139.63, 148.92, 155.91, 160.89, 166.71, 177.91 (C N). LREIMS: 366; HREIMS: 366.1287 (calcd. for C₁₈H₁₈N₆OS; 366.1263).

5-((2,4-Dichlorophenyl)diazenyl)-4-methyl-2-(2-(1-(pyridin-2-yl)ethylidene)hydrazinyl)thiazole (5f)

Reddish powder, m.p. 135–137 °C, yield (87%). ¹H NMR (500 MHz, CDCl₃): δ = 2.49 (s, 3H, CH₃), 2.70 (s, 3H, CH₃-thiazole), 7.34–7.35 (m, 2H, H–Ar), 7.50 (d, 1H, J = 9.0 Hz, H–Ar), 7.68 (s, 1H, H-Ar), 7.77–7.80 (m, 1H, H–Ar), 8.30 (d, 1H, J = 8.0 Hz, H–Ar), 8.67 (d, 1H, J = 4.0 Hz, H-Ar), 10.97 (s, 1H, NH). ¹³C NMR (126 MHz, CDCl₃) δ = 14.81, 16.81, 115.83, 119.30, 121.87, 121.96, 124.85, 127.37, 128.00, 128.42, 129.06, 136.27, 137.48, 143.78, 149.00, 155.18, 178.05. LREIMS: 404; HREIMS: 404.0386 (calcd. for C₁₇H₁₄Cl₂N₆S; 404.0378).

4.2. Biological activity

4.2.1. Antitubercular activity

4.2.1.1. MIC determination

The minimum inhibitory concentration (MIC) against M. tuberculosis H37Rv and M. abscessus was measured using liquid 7H9/OADC culture medium (Middlebrook, Difco, Baltimore, MD, USA) supplemented with different concentrations of the test compounds. Before being added to the growth medium, the test compounds were dissolved in dimethyl sulfoxide (DMSO). To prevent any effect on bacterial growth, the final DMSO concentration in the medium was kept at a maximum of 0.1% (v/v). MIC values were determined using the micro-plate alamar blue assay (MABA) method as described by Franzblau et al.⁴⁸ Antimicrobial activity was assessed by visually monitoring the color change of the indicator from blue to pink. Each experiment included appropriate control wells containing only the bacteria, only the medium, or only the compound, and the MABA procedure was conducted in triplicate.

4.2.1.2. In vitro cytotoxicity assay

The in vitro cytotoxicity evaluation of the tested compounds was performed after 24-hour incubation in compliance with international guidelines (ISO 10993-5:2009(E)) utilizing L929 cell lines (L929 is ATCC-CCL-1) (Manassas, VA, U.S.A.), and the MTT assay technique.

4.2.1.3. Bactericidal activity

The assessment of bactericidal activity was conducted using a recombinant M. tuberculosis strain with stable luciferase expression (M. tuberculosis attB::pMV306LuxABCD) and/or by determining colony-forming unit (CFU) counts. The strain with stable luciferase expression had been previously validated using rifampicin to evaluate the relationship between inhibitor presence and luminescence levels. Additionally, for compound 5a, bactericidal activity assessment was performed using both the strain with stable luminescence expression and the CFU enumeration method. The bactericidal activity of the examined compounds was determined by tracking changes in optical density (OD600), measuring luminescence intensity, and counting colony-forming units (CFU) in M. tuberculosis H37Rv cultures exposed to the test substances. M. tuberculosis cultures were standardized to an OD600 of 0.1 using liquid 7H9/OADC medium (Middlebrook, Difco, Baltimore, MD, United States) supplemented with 0.05% Tween 80. Test compounds were added at different concentrations in duplicate, along with untreated control cultures. The bacterial cultures were maintained at 37 °C, with optical density readings recorded at 7 and 14 days' post-treatment. For CFU determination, Middlebrook 7H10/OADC agar (Difco, Baltimore, MD, USA) enriched with 0.5% glycerol was utilized. Bacterial suspensions from culture bottles were serially diluted in 7H9/OADC broth containing 0.05% Tween 80, then spread onto solid agar medium on days 1, 7, and 14 of the experiment. Colony enumeration was performed after 3–5 weeks of incubation at 37 °C to determine CFU values.

4.2.1.4. Biofilm formation

M. tuberculosis biofilm development was performed following standard methodologies with minor adaptations.⁴⁹Mycobacterium tuberculosis cultures were grown in 7H9/OADC medium supplemented with 0.05% tyloxapol until an optical density at 600 nm (OD600) of 1.0 was attained. Subsequently, the bacterial suspension was diluted 1 : 100 (v/v) in Sauton's medium and aliquot into 24-well plates at a volume of 2.5 mL per well. Then, the plates were covered with parafilm and maintained at 37 °C in a humidified environment for five weeks to promote biofilm development. Following biofilm maturation, the growth medium was replaced with fresh medium containing 0.1% casitone and varying concentrations of the tested compounds, and then incubated for 48 hours at 37 °C. Cell viability within the biofilms was assessed using resazurin-based fluorometric analysis, where 375 μL of 0.02% resazurin solution was introduced per well and allowed to incubate for 90 minutes. Fluorescence measurements were obtained using a SpectraMax® i3 multi-mode microplate reader (Syngen) with excitation and emission wavelengths configured at 550 nm and 590 nm, respectively. Results were presented as the mean fluorescence values from four biological replicates for both tested and control samples.

4.2.1.5. Quantifying the interaction between rifampicin (RMP) and 5a using a two-dimensional dilution matrix

Combination drug studies were performed using the checkerboard method to assess possible interactions between rifampicin (RMP) and the compound 5a. Sequential dilutions of rifampicin (0.04; 0.02; 0.01; 0.005; 0.0025; 0.00125 μg mL⁻¹) were arranged along one dimension of 96-well plates, while the compound 5a was progressively diluted along the orthogonal dimension (1.0; 0.5; 0.25; 0.125 μg mL⁻¹), generating a bidimensional concentration grid. The plates were then seeded with M. tuberculosis and maintained under standard incubation conditions. Bacterial proliferation was quantified using resazurin-based detection techniques. The fractional inhibitory concentration index (FICI) was determined by initially establishing individual minimum inhibitory concentrations (MICs) for each agent, followed by identification of the minimal effective combination concentrations from the checkerboard matrix. FICI values were computed as the summation of FIC_A and FIC_B, where FIC_A denotes the rifampicin concentration in the active combination relative to its standalone MIC, and FIC_B denotes the test compound concentration in the active combination relative to its standalone MIC. Interaction types were classified according to the fractional inhibitory concentration index (FICI) as follows: synergistic for FICI ≤ 0.5, additive for FICI between 0.5 and 1, indifferent for FICI > 1 but ≤ 4, and antagonistic for FICI > 4.

4.2.2. Evaluation of InhA inhibitory activity

The inhibitory effect on the enoyl-acyl carrier protein reductase InhA was assessed using a colorimetric assay, conducted in accordance with the methodology described in previous research.⁵⁰

4.3. Molecular docking analysis

In this study, our candidates were docked against the M. tuberculosis enoyl reductase (InhA) target protein obtained from the protein data bank (PDB ID: 4TZK) to evaluate their potential affinity and pattern of the interactions.⁵¹ At first, the target protein was prepared by removing water molecules and unnecessary particles, and then protons were added and unfilled valence atoms were corrected.⁵² The protein structure was minimized and saved as pdbqt file. With the same procedure, the tested compounds were prepared, protonated, energy minimized and saved as pdbqt file;⁵³ the prepared ligands were docked against the previous target site, and then the best scoring candidates were chosen. The process was conducted using Autodock Vina 1.5.7 software.⁵⁴ The docking scores (affinity energy) of the best-fitted poses with the active site were recorded, and 3D and 2D figures were generated using Discovery Studio 2024 visualizer.⁵⁵

4.4. Molecular dynamics (MD) simulation

Molecular dynamics (MD) simulations were conducted utilizing the Desmond simulation package (Schrödinger LLC).⁵⁶ All simulations were performed under the NPT ensemble conditions, maintaining a constant temperature of 300 K and a pressure of 1 bar. Each simulation was executed for a duration of 100 ns, with a relaxation time of 1 ps applied to the ligands under investigation. The OPLS_2005 force field parameters were consistently applied throughout the simulations. Long-range electrostatic interactions were calculated using the particle mesh Ewald (PME) method, employing a cutoff radius of 9.0 Å for coulombic interactions.⁵⁷

Explicit solvation was modeled using the simple point charge (SPC) water model. Pressure and temperature were controlled using the Martyna–Tuckerman–Klein chain coupling scheme, with a coupling constant of 2.0 ps, and the Nosé–Hoover chain coupling scheme, respectively. The r-RESPA multiple time-step integrator was utilized to evaluate non-bonded interactions, updating short-range forces at every integration step and long-range forces every three steps. Trajectory data were recorded at intervals of 4.8 ps for subsequent analysis. Ligand-protein binding patterns and interaction profiles were analyzed using the simulation interaction diagram tool available within the Desmond MD suite. The stability of the simulations was evaluated by tracking the root mean square deviation (RMSD) of atomic positions for both ligand and protein over the simulation time course.⁵⁸

Author contributions

Ahmed Sabt: writing – review & editing, writing – original draft, supervision, methodology, investigation, data curation, conceptualization. Małgorzata Korycka-Machala: methodology, data curation. Asmaa F. Kassem: validation, software, methodology. Abdulrahman M. Saleh: writing – original draft, software, methodology. Hanaa Farag: methodology, data curation, Moataz A. Shaldam: writing – original draft, software, methodology, Mohamed G. Thabit: investigation, data curation, Anna Brzostek: methodology, data curation. Malwina Kawka: methodology, data curation. Bozena Dziadek: methodology, data curation. Hoda Atef Abdelsattar Ibrahim; investigation, data curation, methodology. Xinsheng Lei: investigation, data curation, dupervision. Jarosław Dziadek: writing – review & editing, supervision, formal analysis, conceptualization, Mohamed A. Abdelrahman: investigation, data curation, methodology.

Conflicts of interest

The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper. The authors alone are responsible for the content and writing of the paper.

Data availability

ALL required data are inserted in the manuscript and supplementary information (SI).

Supplementary information is available. See DOI: https://doi.org/10.1039/d5md00997a.