Zinc-containing enzymes known as human carbonic anhydrases (CAs) facilitate a reversible reaction between carbon dioxide (CO₂) and proton/bicarbonate ions.¹ These enzymes govern extracellular and intracellular pH homeostasis, contributing to critical pathophysiological processes.² The existing literature has identified 12 active human CAs with distinct tissue-specific and sub-cellular expression.³ Mounting evidence suggests a connection between irregular expression or activity of CAs and various human disorders, including cancer,⁴ glaucoma,⁵ obesity,⁶ and epilepsy.⁷ Consequently, researchers are working to develop CA inhibitors and activators with diverse potential medical applications.⁸ Among the CA subtypes, CA VI exists exclusively in serum,⁹ saliva,¹⁰ milk,¹¹ the alimentary canal,¹² and respiratory airways.¹³ Earlier investigations have associated CA VI dysregulation with certain cancers, such as salivary gland carcinoma,¹⁴ highlighting its potential as a cancer biomarker.¹⁵ Furthermore, studies have noted a correlation between heightened CA VI activity and dental caries.¹⁶ For example, Picco et al.¹⁷ demonstrated elevated CA VI activity in the dental biofilms of children with caries compared with caries-free children. Moreover, Al-Mahdi et al.¹⁸ established a positive link between CA VI copy numbers and increased caries on both smooth and occlusal surfaces, indicating that CA VI potentially plays a role in the progression of dental caries.

For primary care, the use of herbal remedies has become increasingly prevalent during the past decade.^19,20 Flavonoids, a category of polyphenolic compounds, naturally accumulate in numerous plants and are found in fruits, beverages, and vegetables. These secondary metabolites have a range of health advantages, including antioxidative, anti-inflammatory, anticancer, and anti-tooth-caries properties.²¹ Cinnamic acid derivatives are a cluster of herbal compounds with a foundational C6–C3 structure. These aromatic carboxylic acids are prominently present in the cellular walls of plants and play a role in plant growth, development, and defense against diseases.²² A body of literature underscores the characterization of cinnamic acids as compounds with anticancer and anti-tooth-caries attributes.²³ Anthraquinones (AQs), a substantial group of secondary metabolites, are primarily discovered in fungi, Rubiaceae, Polygonaceae, and Rhamnaceae.²⁴ Prior investigations have linked AQs with antioxidant, anti-inflammatory, antimicrobial, anticancer, and anti-tooth-caries qualities.^25,26

Within this context, we wondered about the potential CA VI inhibitory properties of flavonoids, cinnamic acid derivatives, and AQs and their implications for therapeutic interventions to treat various human disorders. Therefore, the binding affinity between selected herbal compounds and the active site of CA VI was scrutinized using AutoDock 4.0 software (http://autodock.scripps.edu). Additionally, the stability of the docked poses for the most potent flavonoid, cinnamic acid derivative, and AQ was investigated through molecular dynamics (MD) simulations. Then we explored plausible interactions between the most potent inhibitors and residues within the catalytic site of CA VI. Validation of such computational predictions commonly depends on in vitro tests with relevant cell lines. Among cancers, colorectal cancer (CRC) has the third-highest prevalence and fourth-highest mortality rate.²⁷ In this study, we performed an in vitro investigation with the HCT-116 human CRC cell line, which was originally derived from a malignant colon tumor, to assess the cytotoxic effects of the most potent CA VI inhibitor that we identified. The 2,5-diphenyl-2H-tetrazolium bromide (MTT) method—an extensively used technique for estimating cell viability and ascertaining compound cytotoxicity—was chosen to gauge those effects.

Structure preparation

The three-dimensional coordinates of the CA VI structure were downloaded from the Protein Data Bank (PDB;https://www.rcsb.org) at 1.3 Å X-ray resolution (PDB ID: 6QL2).¹⁴ The 6QL2 file included one polypeptide chain (chain A) with 258 residues and an ethoxzolamide inhibitor (PDB ID, EZL; PubChem ID, 3295). The inhibitor compound was removed from the CA VI structure, and energy optimization of the enzyme was performed using Discovery Studio Client version 16.1.0.15350 (BIOVIA). The binding affinity between 79 small molecules (45 flavonoids, 21 AQs, and 12 cinnamic acids), with EZL as a positive control inhibitor, and the CA VI active site was examined using AutoDock 4.0 software. The structures of the ligands were obtained as spatial data file (SDF) files, converted into PDB format, and then energy-minimized following procedures from our previous reports.^24,28

Molecular docking

A Windows-based system with a 64-bit operating system, 32 GB random access memory (RAM), and Intel Core i7 central processing unit (CPU) (Intel) was used for molecular docking and dynamics simulations.²⁹ The docking procedures were conducted using AutoDock 4.0 software, and the MD simulations were executed in Discovery Studio Client version 16.1.0.15350. The AutoDock 4.0 tool uses the Lamarckian genetic algorithm to compute the Gibbs free binding energy between ligands and receptors. Preparation of the CA VI structure involved the inclusion of polar hydrogen atoms and Kollman charges. The protein and ligand structures were readied as PDB partial charge (Q) & atom type (T) (PDBQT) files to facilitate the molecular docking processes.

The identification of central residues within the active site of CA VI was achieved by analyzing the two-dimensional structures of several inhibitors within the catalytic domain of the protein. These investigations were conducted across multiple studies led by Kazokaitė et al.¹⁴ Noteworthy among these inhibitors were V50 (PubChem ID, 71299336), found within the 6QL1 file; EZL, incorporated in the 6QL2 file; and V14 (PubChem ID, 73774785), located within the 6QL3 file. As a result of those studies, a set of 18 amino acids was recognized within the CA VI active site: Asn62 (asparagine), His64 (histidine), Thr65 (threonine), Gln67 (glutamine), Gln92, His94, His96, Glu106 (glutamic acid), His119, Val121 (valine), Tyr131 (tyrosine), Gln135, Leu198 (leucine), Thr199, Thr200, Pro202 (proline), Thr204, and Trp209 (tryptophan). Based on the coordinates of these identified residues, the specifications for the grid box were established as follows: spacing, 0.375 Å; X-dimension, 56; Y-dimension, 66; Z-dimension, 62; X-center, 15.924; Y-center, –4.825; and Z-center, 15.029.

Next, 46 flavonoids, 21 AQs, and 12 cinnamic acids, all ligands, were subjected to 50 docking runs each. Among the runs, the most negative ΔG_binding value within the largest cluster of the root mean square deviation (RMSD) table is reported as the binding affinity between the molecules. In this study, compounds exhibiting ΔG_binding values higher than –10 kcal/mol were categorized as the top-ranked CA VI inhibitors. To uncover potential interactions between these top-ranked inhibitors and the residues integrated within the active site of CA VI, we used BIOVIA Discovery Studio Visualizer version 19.1.0.18287.

MD computer simulations

Among the array of flavonoids, AQs, and cinnamic acids, the most potent inhibitors of CA VI were singled out for MD analyses in 40-ns computer simulations. The parameters for the MD simulations were aligned with the advanced settings outlined in our earlier report.²⁸ From the MD analysis, we gleaned insights into the interactions between the compounds and amino acids within the CA VI active site. Throughout the simulation, both the RMSD of the backbone atoms and the root mean square fluctuation (RMSF) were scrutinized to understand their dynamics.

Ethical approval

This study was approved by the Ethics Committee/Institutional Review Board of Hamadan University of Medical Sciences, Hamadan-Iran (IRB no. IR.UMSHA.REC.1401.141). Informed consent was waived by the board.

Cell viability studies

Human HCT-116 epithelial CRC cells were procured from the Pasteur Institute of Iran Cell Bank. Subsequently, the HCT-116 cells were plated in a 96-well format at a density of 10,000 cells per well. The cells were cultivated until they reached 70% to 80% confluence. Then they were exposed to varying concentrations of the most potent CA VI inhibitor (0, 10, 20, 40, 80, 120, and 160 µM) for 24, 48, and 72 hours. The MTT reagent was added to each well during the assay and incubated with the cells at 37 °C with 5% CO₂ for 4 hours. Viable cells converted the MTT into insoluble purple formazan crystals. Post-incubation, dimethyl sulfoxide was added, and a microplate reader was used to measure the absorbance at a wavelength of 570 nm to determine the percentage of viable cells. The obtained data were analyzed using both Excel (Microsoft) and GraphPad Prism 5 software (GraphPad Software Inc.).

Binding affinity assessment

The analysis conducted through AutoDock 4.0 revealed that 14 flavonoids, three AQs, and three cinnamic acid derivatives had ΔG_binding values below –10 kcal/mol. These were designated as the top-ranked herbal inhibitors of CA VI in this study. The flavonoids were kaempferol 3-rutinoside-4-glucoside, orientin, kaempferol 3-rutinoside-7-sophoroside, amentoflavone, rutin, isoquercitrin, vicenin-2, vitexin, quercitrin, astragalin, quercetin-3-rhamnoside, kaempferol 7-O-glucoside, nicotiflorin, and cynaroside. Among the AQs, aloe-emodin 8-glucoside, hypericin, and pulmatin (chrysophanol-8-O-glucoside) emerged as the top CA VI inhibitors. Cynarin, chlorogenic acid, and rosmarinic acid had the most negative ΔG_binding values among the cinnamic acid derivatives.

Remarkably, five compounds, kaempferol 3-rutinoside-4-glucoside, orientin, kaempferol 3-rutinoside-7-sophoroside, cynarin, and chlorogenic acid, were estimated to establish attachments with the CA VI catalytic domain at picomolar (pM) concentrations. They were consequently identified as the most potent CA VI inhibitors in this study; the ΔG_binding values between them and the CA VI catalytic domain were –15.60, –13.92, –12.70, –14.93, and –12.42 kcal/mol, respectively, and their calculated inhibition constant (Ki) values were 3.69, 62.48, 493.64, 11.33, and 783.13 pM. A comprehensive list of the ΔG_binding and Ki values between the studied ligands and the CA VI active site is presented in Table 1. Fig. 1 compares the ΔG_binding values between the top-ranked components and the positive control inhibitor and the CA VI catalytic domain.

Protein-ligand interactions

The interactions between the top-ranked compounds identified in this study (and the positive control inhibitor) and residues within the CA VI active site were assessed using BIOVIA Discovery Studio Visualizer. Additionally, comparisons were performed before and after 40 ns of MD simulations for three compounds: kaempferol 3-rutinoside-4'-glucoside, aloe-emodin 8-glucoside, and cynarin. Those findings are presented in Table 2 and illustrated in Fig. 2.

Among the flavonoids, orientin exhibited the highest number of hydrogen bonds with residues inside the CA VI active site. Specifically, the interactions involved Asn62, His64, Gln135, and Pro201. Among the AQs, aloe-emodin 8-glucoside established four hydrogen bonds with Glu69, Leu198, Thr199, and Thr200. Similarly, rosmarinic acid had three hydrogen bonds, forming interactions with Pro201 and Pro202.

Before the MD analysis, kaempferol 3-rutinoside-4'-glucoside, aloe-emodin 8-glucoside, and cynarin exhibited one, four, and one hydrogen bonds, respectively. Following 40 ns of MD simulation, the number of hydrogen bonds for kaempferol 3-rutinoside-4'-glucoside and cynarin increased to seven and two, respectively. Notably, the count of hydrogen bonds for aloe-emodin 8-glucoside remained unchanged after the MD simulation. Hydrogen bonds with distances exceeding 5 Å were omitted from Table 2 for clarity and relevance.

MD computer simulations

The RMSD ranges for CA VI in complex with kaempferol 3-rutinoside-4-glucoside, aloe-emodin 8-glucoside, and cynarin were 1.37 to 2.05, 1.25 to 1.85, and 1.07 to 1.54 Å, respectively. These compounds’ corresponding average RMSF values were 1.70, 1.57, and 1.34 Å. Notably, the enzyme complexes involving kaempferol 3-rutinoside-4-glucoside, aloe-emodin 8-glucoside, and cynarin appeared to attain a state of relative stability at approximately 30, 25, and 35 ns, respectively, as illustrated in Fig. 3A.

Furthermore, the mean RMSF values for the CA VI residues complexed with kaempferol 3-rutinoside-4-glucoside, aloe-emodin 8-glucoside, and cynarin were 1.72, 1.57, and 1.32 Å, respectively. Most notably, the residues integrated within the CA VI active site exhibited lower fluctuations than the other residues, as depicted in Fig. 3B.

The structural alignment of CA VI in complex with aloe-emodin 8-glucoside, kaempferol 3-rutinoside-4-glucoside, and cynarin, both before and after 40 ns of MD simulations, is visually presented in Fig. 4.

MTT assay

Cell viability was assessed across various concentrations of cynarin, the most potent CA VI inhibitor. The results from the MTT assays indicate that cynarin induced cytotoxic effects in HCT-116 cells, with an observed half maximal inhibitory concentration of 199.9 µg/mL, as depicted in Supplementary Fig. 1.

This study employed a comprehensive computational approach to identify potential inhibitors of CA VI from a pool of 79 plant-based compounds: 46 flavonoids, 21 AQs, and 12 cinnamic acid derivatives. In that analysis, kaempferol 3-rutinoside-4-glucoside, aloe-emodin 8-glucoside, and cynarin emerged as the strongest CA VI inhibitors from the categories of flavonoids, AQs, and cinnamic acid derivatives, respectively. The ΔG_binding value between kaempferol and the CA VI active site was –7.99 kcal/mol. In contrast, the binding affinities for kaempferol 7-O-glucoside, kaempferol 3-rutinoside-7-sophoroside, and kaempferol 3-rutinoside-4'-glucoside were –10.60, –12.70, and –15.60 kcal/mol, respectively. This suggests that appending a sugar moiety to kaempferol enhances its binding affinity to the CA VI catalytic domain. Notably, kaempferol 3-rutinoside-4-glucoside established one hydrogen bond and three hydrophobic interactions with residues within the CA VI active site (Trp5, His64, Tyr131, and Thr200) before the MD simulation. After 40 ns of MD computer simulation, the compound formed seven hydrogen bonds with Asn62, Gln67, Glu69, Gln92, Tyr131, and Pro201. Kaempferol is recognized as a metabolite and has demonstrated efficacy in reducing the incidence of breast³⁰ and colon³¹ cancers. Its anticancer properties predominantly involve inducing apoptosis and suppressing cell proliferation in cancer cells.³² Notably, kaempferol has also been identified as an anti-tooth-caries agent. Research has indicated its antibacterial activity against leading oral pathogens that contribute to dental caries, including Porphyromonas gingivalis and Streptococcus mutans.³³

Before the MD simulation, aloe-emodin 8-glucoside established four hydrogen bonds and two hydrophobic interactions with residues within the CA VI active site (Glu69, His94, Val121, Leu198, Thr199, and Thr200). After the MD analysis, the ligand exhibited four hydrogen bonds and one hydrophobic interaction with residues within the CA VI catalytic site (His94, Thr200, and Pro201). In a related study, Mohammed et al.³⁴ conducted an integrated investigation combining in silico and in vitro analyses to explore the anticancer and antibacterial properties of metabolites extracted from the Lycium shawii shrub. They identified five metabolites in L. shawii: costunolide, catechin, lyciumaside, emodin, and aloe-emodin 11-O-rhamnoside. Notably, emodin demonstrated good binding affinity to CA IX, indicating its potential for anticancer and antibacterial activity against pathogens such as methicillin-resistant Staphylococcus aureus, S. mutans, Escherichia coli, and Klebsiella pneumoniae.

Cynarin, known chemically as 1,3-O-dicaffeoylquinic acid, is a plant-derived compound primarily sourced from artichokes.³⁵ It has garnered attention for its range of advantageous attributes, particularly its antioxidant effects, antihypertensive properties, and ability to lower cholesterol.³⁶ Before the MD simulation, cynarin established one hydrogen bond and one hydrophobic interaction with Leu198 and Pro201 within the CA VI active site. Following the MD simulation, it exhibited an augmented interaction profile, forming two hydrogen bonds and two hydrophobic interactions with His64,Gln67, Leu198, and Pro201 inside the CA VI catalytic site.

Within the confines of the current study, the outcomes from the MTT assay substantiate the cytotoxic attributes of cynarin against the HCT-116 cell line. Notably, the investigation unveiled a correlation between increasing cynarin concentration and inhibition of cancer cell survival. Thus, the idea that cynarin and structurally analogous compounds could serve as therapeutic agents in conjunction with conventional treatment modalities is a plausible hypothesis. Nonetheless, future investigations should comprehensively elucidate the interactions between cynarin and the human serum albumin (HSA) protein. HSA is the primary transport vehicle for an extensive array of molecules in the bloodstream and significantly influences the effectiveness, metabolism, distribution, and clearance of pharmaceutical compounds.³⁷

Angelini et al.³⁸ showed that cynarin notably amplified cytotoxicity and the doxorubicin concentration in doxorubicin-resistant uterine sarcoma cells (MES-SA/Dx5) compared with control cells exposed solely to doxorubicin. In summation, we conclude that kaempferol 3-rutinoside-4-glucoside, orientin, kaempferol 3-rutinoside-7-sophoroside, cynarin, and chlorogenic acid can inhibit CA VI activity at the pM scale. Those compounds emerged as the most robust CA VI inhibitors among the metabolites investigated here. Furthermore, the docked conformations of kaempferol 3-rutinoside-4-glucoside, aloe-emodin 8-glucoside, and cynarin maintained their stability in 40-ns computer simulations, suggesting the potential suitability of these metabolites for medicinal development targeting obesity, salivary gland carcinoma, and dental caries. However, comprehensive in vitro, in vivo, and clinical trials are needed to validate and explore these results.

Each residue within the enzymatic active site is pivotal in orchestrating intricate biochemical activities, and some of those roles have been definitively characterized. For example, Asn62 and Gln92 participate in binding interactions with specific CA VI activators, notably histamine and L-histidine. Concomitantly, His64 serves as a proton donor and acceptor, mediating vital hydrogen bonding interactions. His94, His96, and His119 collectively contribute to the meticulous maintenance of Zn⁺² coordination within the catalytic pocket of CA VI. Furthermore, Thr199 has a critical responsibility in establishing binding interactions with the enzyme substrate.^39,40

Significant insights from previous studies underscore the pronounced effects of mutations on specific residues. Notably, mutagenesis at Asn62 markedly attenuates enzyme activity, whereas analogous manipulations of His64 severely impede CO₂ hydrase functionality. Similarly, mutagenesis targeting His94 substantially reduces CO₂ hydrase and p-nitrophenyl acetate esterase activity, concurrently destabilizing zinc binding. Furthermore, perturbations in His119 led to a profound reduction in CA VI activity.^39,40

Therefore, it is conceivable that binding between the identified top-ranked CA VI inhibitors and residues within the active site of the enzyme could precipitate multifaceted mechanisms of enzymatic attenuation, including (1) decreasing the binding affinity of histamine and L-histidine to the CA VI allosteric site, (2) attenuating CO₂ hydrase activity, and (3) compromising the stability of Zn⁺² coordination.

Supplementary materials can be found online at https://doi.org/10.7570/jomes23029.

The authors declare no conflict of interest.

The authors thank the support of the Dental Research Center, Research Center for Molecular Medicine, Hamadan University of Medical Sciences, Hamadan, Iran. This research received no specific grant from any funding agency in the public, commercial, or not-for-profit sectors.

Study concept and design: EY, AT; acquisition of data: MK, MS, AT; analysis and interpretation of data: EY, MK, ZK, MS, AT; drafting of the manuscript: AT; critical revision of the manuscript: EY, ZK; statistical analysis: MS; administrative, technical, or material support: MS; and study supervision: EY, AT.