New Carbonic Anhydrase-II Inhibitors from Marine Macro Brown Alga Dictyopteris hoytii Supported by In Silico Studies

In continuation of phytochemical investigations of the methanolic extract of Dictyopteris hoytii, we have obtained twelve compounds (1–12) through column chromatography. Herein, three compounds, namely, dimethyl 2-bromoterepthalate (3), dimethyl 2,6-dibromoterepthalate (4), and (E)-3-(4-(dimethoxymethyl)phenyl) acrylic acid (5) are isolated for the first time as a natural product, while the rest of the compounds (1, 2, 6–12) are known and isolated for the first time from this source. The structures of the isolated compounds were elucidated by advanced spectroscopic 1D and 2D NMR techniques including 1H, 13C, DEPT, HSQC, HMBC, COSY, NEOSY, and HR-MS and comparison with the reported literature. Furthermore, eight compounds (13–20) previously isolated by our group from the same source along with the currently isolated compounds (1–12) were screened against the CA-II enzyme. All compounds, except 6, 8, 14, and 17, were evaluated for in vitro bovine carbonic anhydrase-II (CA-II) inhibitory activity. Eventually, eleven compounds (1, 4, 5, 7, 9, 10, 12, 13, 15, 18, and 19) exhibited significant inhibitory activity against CA-II with IC50 values ranging from 13.4 to 71.6 μM. Additionally, the active molecules were subjected to molecular docking studies to predict the binding behavior of those compounds. It was observed that the compounds exhibit the inhibitory potential by specifically interacting with the ZN ion present in the active site of CA-II. In addition to ZN ion, two residues (His94 and Thr199) play an important role in binding with the compounds that possess a carboxylate group in their structure.


Introduction
In recent years, more than 200,000 eukaryotic marine species have been validated, among which, algae contributed nearly 44,000 well described species [1]. Among algae, macroalgae (seaweeds) serve as a novel source of compounds, due to their use in traditional medicine and nutritional importance [2]. The chemical composition varies considerably due to both environmental condition and genetic differences among species [3]. These are valuable due to their high content in compounds with different biological activities, including both complex organic compounds and primary and secondary metabolites [3,4]. Moreover, these seaweeds can be used as a source of biofuels due to their high growth rate, high photosynthetic efficiency, no competition with food crops, high carbohydrate

Structure Elucidation of Compounds 1-12
The molecular formula (C10H9BrO4) of compound 3 was determined from high resolution-electrospray ionization (HR-ESI)-MS spectrum having a pseudomolecular ion at m/z 294.9581 [ 79 BrM + Na] + (Calcd. 294.9582), and 296.9560 [ 81 BrM+Na] + (Cald. 296.9562). The 1 H-and 13 C-NMR spectra of 3 were identical to those of a 2-bromoethylmethylbenzene-1,4-dioate (13), previously isolated from the same species [15] with the exception of a methoxy in 3 instead of ethoxy group in 13. Compound 3 was isolated as colorless powder ( Figure 2) from the methanolic extract of D. hoytii. The IR spectrum exhibited the presence of carbonyl ester (1735, 1654 cm −1 ), and aromatic ring (1598, 1510, and 1430 cm −1 ) in the molecule. The data obtained from 1 H NMR spectrum of 1 displayed signals assigned to three aromatic protons at δH 8.29 (1H, d = 1.8 Hz, H-3), 7.98 (1H, dd = 7.8, 1.8 Hz, H-5), and 7.79 (1H, d = 7.8 Hz, H-6), which was further confirmed by 13 C NMR spectra at δ 135.1 (C-3), 128.0 (C-5), and 131.0 (C-6). The proton at H-5 was found ortho to H-6 due to their large coupling constants (J > 7 Hz), while the low coupling constant of 1.8 Hz between H-3 and H-5 revealed that H-3 was meta to H-5, which was further confirmed by the 1 H-1 H COSY correlations between H-5 and H-6 ( Figure 2). In addition, 1 H-NMR spectrum also displayed one singlet at 3.93 (6H, s, OCH3) corresponding to the presence of two methoxy groups and their HMBC correlation with δC 166.1 (C-7), and 164.9 (C-8) further confirmed the presence of ester groups. The 13 C-NMR spectrum disclosed 10 signals which were classified with the aid of distortionless enhancement by polarization transfer (DEPT) and heteronuclear single-quantum correlation (HSQC) experiments as a presence of two methoxy, three aromatic methines, two carbonyls, and three quaternary carbons. The downfield signals observed in the spectra at δC 166.1, 164.9, 52.7, and 52.6 support the presence of these functional groups in the molecule. Furthermore, the correlations observed in HMBC of H-3 with C-2, C-4, and C-8; H-5 with C-4, C-6, and C-8; H-6 with C-5, C-4, C-7, and C-1 further supported the presence of methyl esters at the para positions. The above data were compatible with a benzyl skeleton substituted with two methyl esters at para

Structure Elucidation of Compounds 1-12
The molecular formula (C 10 H 9 BrO 4 ) of compound 3 was determined from high resolution-electrospray ionization (HR-ESI)-MS spectrum having a pseudomolecular ion at m/z 294.9581 [ 79 BrM + Na] + (Calcd. 294.9582), and 296.9560 [ 81 BrM+Na] + (Cald. 296.9562). The 1 H-and 13 C-NMR spectra of 3 were identical to those of a 2-bromoethylmethylbenzene-1,4-dioate (13), previously isolated from the same species [15] with the exception of a methoxy in 3 instead of ethoxy group in 13. Compound 3 was isolated as colorless powder ( Figure 2) from the methanolic extract of D. hoytii. The IR spectrum exhibited the presence of carbonyl ester (1735, 1654 cm −1 ), and aromatic ring (1598, 1510, and 1430 cm −1 ) in the molecule. The data obtained from 1 H NMR spectrum of 1 displayed signals assigned to three aromatic protons at δ H 8.29 (1H, d = 1.8 Hz, H-3), 7.98 (1H, dd = 7.8, 1.8 Hz, H-5), and 7.79 (1H, d = 7.8 Hz, H-6), which was further confirmed by 13 C NMR spectra at δ 135.1 (C-3), 128.0 (C-5), and 131.0 (C-6). The proton at H-5 was found ortho to H-6 due to their large coupling constants (J > 7 Hz), while the low coupling constant of 1.8 Hz between H-3 and H-5 revealed that H-3 was meta to H-5, which was further confirmed by the 1 H-1 H COSY correlations between H-5 and H-6 ( Figure 2). In addition, 1 H-NMR spectrum also displayed one singlet at 3.93 (6H, s, OCH 3 ) corresponding to the presence of two methoxy groups and their HMBC correlation with δ C 166.1 (C-7), and 164.9 (C-8) further confirmed the presence of ester groups. The 13 C-NMR spectrum disclosed 10 signals which were classified with the aid of distortionless enhancement by polarization transfer (DEPT) and heteronuclear single-quantum correlation (HSQC) experiments as a presence of two methoxy, three aromatic methines, two carbonyls, and three quaternary carbons. The downfield signals observed in the spectra at δ C 166.1, 164.9, 52.7, and 52.6 support the presence of these functional groups in the molecule. Furthermore, the correlations observed in HMBC of H-3 with C-2, C-4, and C-8; H-5 with C-4, C-6, and C-8; H-6 with C-5, C-4, C-7, and C-1 further supported the presence of methyl esters at the para positions. The above data were compatible with a benzyl skeleton substituted with two methyl esters at para positions and one ortho substituted bromine, which was further supported by 1 H-1 H correlation positions and one ortho substituted bromine, which was further supported by 1 H-1 H correlation spectroscopy (COSY) and heteronuclear multiple bond correlation (HMBC) experiment. Based on these facts, compound 3 was assigned as dimethyl 2-bromoterepthalate.  Figure  2). The 1 H-and 13 C-NMR spectra of 4 closely resembled those of compound 3 with the only difference being additional bromine at C-4 attached to the benzyl ring. Due to symmetry in the molecule, one downfield signal at δ 8.04 with double integration (2H, s, H-3/6) was observed in 1 H-NMR spectra assigned to two aromatic protons H-3/6 and this was further confirmed by 13 C-NMR spectra at δ 166.3 (C-7/8). Moreover, the 1 H-NMR spectrum displayed two signals at 3.96 (3H, s), and 3.88 (3H, s), which corresponded to the presence of two methoxy groups and their HMBC correlations with C-7/8 further confirmed the presence of dimethyl esters ( Figure 2). The HMBC correlations of H-3 to C-2, C-4, C-1, and C-8 and H-5 to C-6, C-4, C-8, and C-1 further indicated the presence and position of ester groups at C-1 and C-4 in the molecule. Because of these advance The 1 H-and 13 C-NMR spectra of 4 closely resembled those of compound 3 with the only difference being additional bromine at C-4 attached to the benzyl ring. Due to symmetry in the molecule, one downfield signal at δ 8.04 with double integration (2H, s, H-3/6) was observed in 1 H-NMR spectra assigned to two aromatic protons H-3/6 and this was further confirmed by 13 C-NMR spectra at δ 166.3 (C-7/8). Moreover, the 1 H-NMR spectrum displayed two signals at 3.96 (3H, s), and 3.88 (3H, s), which corresponded to the presence of two methoxy groups and their HMBC correlations with C-7/8 further confirmed the presence of dimethyl esters ( Figure 2). The HMBC correlations of H-3 to C-2, C-4, C-1, and C-8 and H-5 to C-6, C-4, C-8, and C-1 further indicated the presence and position of ester groups at C-1 and C-4 in the molecule. Because of these advance spectroscopic techniques and above observations, the structure of compound 4 was designated as dimethyl 2,6-dibromoterepthalate.

Carbonic Anhydrase-II Inhibition and Structural-Activity Relationship
In the fields of medicine, drug discovery is under increased pressure to recognize more appropriate secondary metabolites as starting points for drug development and searching for new compounds as chemical starting points for optimization is one of the key challenges in drug discovery research. To search novel inhibitors, all the isolated com pounds (except compounds 6, 8, 14, and 17) were evaluated for CA-II inhibitory potential

Carbonic Anhydrase-II Inhibition and Structural-Activity Relationship
In the fields of medicine, drug discovery is under increased pressure to recognize more appropriate secondary metabolites as starting points for drug development and searching for new compounds as chemical starting points for optimization is one of the key challenges in drug discovery research. To search novel inhibitors, all the isolated compounds (except compounds 6, 8, 14, and 17 4, 5, 7, 9, 10, 12, 13, 15, 18, and 19) were found active, while compounds 2, 3, 11, 16, and 20 displayed less than 50% inhibition, however, due to paucity of compounds 6, 8, 14, and 17, they were not evaluated. The compounds 4, 5, 7, 12, and 18 exhibited excellent inhibitory activity, whereas compounds 13, 15, and 19 displayed comparable inhibition as compared to standard, while compounds 1, 9, and 10 showed moderate inhibitory activities (Table 1).
When we compared the activity of most active compounds with clinically standard inhibitor acetazolamide (IC 50 =18.2 ± 1.23 µM), the compounds 4, 5, 7, 12, and 18 showed better potency against carbonic anhydrase II enzyme. Comparing compound 1 with 5, higher inhibition of 5 (17.7 ± 2.01 µM) than 1 (44.9 ± 1.10 µM) may be due to the presence of dimethoxymethyl at para position to the acrylic acid. Our results match with the reported data of Innocenti et al. [53], who studied some phenolic compounds including p-coumaric acid, caffeic acid, and ferulic acid, similar in structure to compounds 1 and 5, against CA-II and showed excellent activity in the range of 0.98-2.4 µM. Higher inhibition of these compounds than 1 and 5 may be due to the replacement of -OH with -OCH 3

Molecular Docking of Carbonic Anhydrase-II Inhibitors
Eleven compounds (1, 5, 7, 9-10, 12-13, 15, 18, and 19) exhibited significant inhibitory potential against CA-II in an in vitro experiment. Therefore, the binding mechanism of the active hits was predicted in the active site of CA-II through a molecular docking strategy. When docked at the active site of CA-II, the most active compound, 12, demonstrated excellent binding with ZN ion in the center of the active site. The carboxylate moiety of lacceroic acid (12) formed strong ionic interactions with ZN ion and accepted a hydrogen bond from the side chains of His94, Thr199, and Thr200. Additionally, the main chain amino nitrogen of Thr199 also mediated the H-bond with the carboxylate group of the compound. This suggests that the carboxylate moiety is responsible for the binding, in turn, for inhibitory activity of the compound. The docked conformation of 12 depicted that the long alkyl chain of the compound covered the entrance of the active site completely; however, the solvent was found partially exposed. After compound 12, compound 4 demonstrated good inhibitory activity. The binding mode of compound 4 revealed that one of the carbonyl moieties of the compound binds with ZN ion through ionic interaction and with His94 via the H-bond; however, the other carbonyl group did not show any interaction. Due to the lack of a polar group or warhead in the structure of compound 7, this molecule only interacted with the side chain of His94 through hydrophobic interaction; however, the compound was also found to be accommodated nicely in the active site. The binding mode of compound 18 (cerotic acid) resembled the docked orientation of compound 12. The carboxyl group of 18 also formed ionic interaction with ZN ion, and H-bonds with the side chain and main chain nitrogen of His94 and Thr199, respectively, while the alkyl chain of this compound remained solvent exposed. Similarly, the carboxylate group of 5 was involved in the bidentate interaction with ZN ion and H-bonding with the side chains of His94, His96, and Thr199.
The docked views of compounds 19, 15, and 13 revealed that the carboxylate group of 19 bounds with ZN ion by ionic interactions, and additionally formed H-bonds with the side chain of Thr199; however, like compound 18, the alkyl chain of 19 also remained solvent exposed. The mode of binding of 15 (fucosterol) showed that the -OH group of 15 mediated the H-bond with the side chain of His94, and ionic interaction with ZN ion, whereas the rest of the structure of 15 remained surface exposed. The docked conformation of compound 13 was found quite similar to that of the docked orientation of compound 4. Like compound 4, one of the carbonyl groups of 13 formed ionic interaction with ZN ion, and an H-bond with the side chain of His94. Similarly, Thr199 provided the H-bond to the formyl and the carboxylate groups of 1, one of the aldehyde groups of compound 9, and the carboxylate group of compound 10, while these compounds interacted with ZN ion through ionic interaction. Moreover, the least active molecule, 10, also demonstrated an H-bond with the side chain of His94. However, the compound 10 was found mostly solvent exposed. The binding mode of all these compounds revealed that His94 and Thr199 play an important roles in the stabilization of these compounds in the active site of CA-II. However, several residues (His4, Trp5, Leu60, Asn62, His64, Asn67, and Phe131) at the entrance of the active site, give support to the surface-exposed atoms of these ligands.
The positive control, acetazolamide, a high affinity inhibitor of CA-II was also docked at the active site of CA-II among other inhibitors. The docked view of the compound depicted that the sulfonamide group of acetazolamide interacted with ZN ion through metallic and ionic interactions, and with His94, His96, and Thr199 via H-bonds, while the carbonyl moiety of the control interacted with a surrounding solvent molecule (Figure 4). The docking score of acetazolamide (docking score = −9.47 kcal/mol) was found to be lower than the docking scores of compounds 12 (−7.40 kcal/mol) and 19 (−9.66 kcal/mol), while higher than the docking scores of rests of the compounds. The docking scores and binding interactions of the docked compounds are shown in Table 2. We observed that the docking scores of the structurally similar compounds with long alkyl chains (12, 7, 18, 19, and 20) were higher (in range of >−9 kcal/mol) than that of the docking scores of small molecules with ring structures (1, 4, 5, 9, and 13, docking score in range of >−5kcal/mol) inhibitors, showing a good trend of correlation with the biological activities of the compounds. The binding modes of all the compounds are presented in Figure 5.  1, 4, 5, 7, 9-10, 12-13, 15, 18-19, and acetazolamide.

Material and Methods
This study comprises of extraction and isolation of bioactive compounds from D. hyotii and determination of CA-II inhibitory activity. The molecular docking studies and prediction of pharmacokinetic properties were performed computationally. The computational results were validated by in vitro testing of active compounds against most suitable predicted target.

General Instrumentation
High-resolution electrospray ionization mass spectrometry (HR-ESI-MS) was documented on Agilent Technologies (6530 Accurate Mass, Q-TOF LC/MS). 1 H NMR (Nuclear Magnetic Resonance) spectra were recorded on 600 MHz and 13 C NMR on 150 MHz spectrometers using the solvent peak as internal reference (CDCl 3 , δ H : 7.26; δ C : 77.0) (CD 3 OD, δ H : 4.87; δ C : 48.5). NMR data was reported as chemical shift (δ) in ppm; multiplicities are as singlet (s), doublet (d), triplet (t), doublet of doublet (dd), multiplet (m); coupling constants (J) are in hertz (Hz). Optical rotation ([α] 25 D ) was measured on a KRUSS P3000 polarimeter (A. Kruss Optronic, Germany). Column chromatography (CC) was carried out by using silica gel of the selected particle size of 100-200 mesh. Visualizations of the thin layer chromatographic plates were attained under the UV light at 254 and 366 nm, and by spraying with the ceric sulphate reagent. Solvents for column chromatography (ethyl acetate, EtOAc), methanol, dichloromethane, and n-hexane) were of technical grade and distilled prior to use. Microplate reader (Bio-Rad, Molecular Devices, CA, USA) was used for carbonic anhydrase activity.

Sample Collection and Identification
The seaweed fresh samples of D. hoytii were collected (April, 2017) from coastal region of Raysut (16 • 59 10.04" N, 54 • 1 58.59" E), Dhofar, Oman, and were provided by Oman Animal and Plant Genetic Resource Center. The voucher specimen (DHS-04/2017) was deposited in the herbarium unit at Natural and Medical Sciences Research Center (NMRSC), University of Nizwa, Nizwa, Oman. The provided samples were washed with distilled water (DW) and transported in cool boxes to the laboratory. The cleaned material was freeze dried and ground to make powder, later stored at −20 • C until further analysis.

Carbonic Anhydrase II Inhibition
The total reaction volume of 200 µL including 20 µL of test compounds, isolated from Dictyopteris hoytii, and prepared in DMSO, followed by the addition of 140 µL of HEPEStris buffer, 20 µL of purified bovine erythrocyte CA-II (0.1 mg/mL) prepared in buffer, and 20 µL of a solution of 4-nitrophenyl acetate [53]. Then, 20 µL of the test compound was incubated with the enzyme carbonic anhydrase (EC 4.2.1.1) for 15 min in 96-well flat bottom plates. The rate of product formation was monitored with the addition of 20 µL of 4-NPA as substrate, prepared in methanol at the final concentration of 0.7 mM, at 25 • C for 30 min with regular intervals of 1 min, by using microplate readers (Bio-Rad, Molecular Devices, CA, USA). HEPES-tris was used as a buffer for the reaction at a final concentration of 20 mM at pH 7.4. Acetazolamide was used as standard inhibitor (IC 50 = 18.2 ± 1.23 µM).

Molecular Docking and ADMET Prediction
The docking was performed by MOE docking program [54]. The human CA-II structure in complex with ligand (acetazolamide derivative) was chosen from Protein Databank (PDB code 4IWZ [55], resolution = 1.60Å). The 3D structure of protein was prepared for docking by removing heteroatoms other than the co-crystallized ligand, ZN ion, and the water molecules within the 3Å region of the active site. Hydrogen atoms were added on protein structure and partial charges were applied using MMFF94x force field through MOE Protonate 3D option. The modified protein file was saved in PDB format. The 2D-structures of compounds 1, 3, 5, 7, 9-10, 12-13, 15, 18, and 19 were sketched on ChemDraw Ultra 8.0 and imported into MOE database in which structures were converted into 3D-form by applying minimization (RMS Gradient = 0.1 Kcal/mol/Å). Partial charges (MMFF94x) and hydrogen atoms were added during minimization automatically. After preparation of protein and ligand files, docking was performed with Induced Fit docking protocol of MOE using the Triangle Matcher docking algorithm and London dG scoring function. Induced Fit protocol is appropriate when protein sidechains are to be included in the refinement stage. The co-crystallized ligand and acetazolamide were added in the ligand database as positive controls. The co-crystallized ligand adapts similar docked conformation like in the X-ray crystal structure and showed almost similar binding pattern after docking with RMSD value of 0.96Å ( Figure S2). Afterwards, the physicochemical and pharmacokinetic properties (absorption, distribution, metabolism, excretion, and toxicity) of the compounds were predicted through Swiss ADME server (http://www.swissadme.ch, accessed on 1 August 2021)

Conclusions
Twelve compounds (1-12) were isolated from the methanolic extract of D. hoytii through column chromatography. Compounds 3-5 were identified as natural products for the first time in this study while rest of the compounds (1, 2, 6-12) were known, and were isolated for the first time from the natural source. Among the tested compounds, compounds 4, 5, 7, 12, and 18 were identified as having the most potential hits and may be excellent inhibitors against CA-II. These natural products can eventually be used as lead scaffolds for future drug discovery design and development against cancer. Furthermore, the results of the current investigation also suggest that the selected marine seaweed may serve as a source of novel bioactive natural products and provide a basis to search an alternative source of therapeutic agents.
Supplementary Materials: The following are available online, Figure S1: 1 H, 13 C NMR, and HR-ESI-MS of the compounds 1-12; Figure S2: Superimposed view of co-crystalized ligand in 4IZW. The docked and the X-ray conformations of compound are shown in gray and green colors, respectively; Table S1: The predicted ADMET properties of compounds.