Synthesis of new 1,2-disubstituted benzimidazole analogs as potent inhibitors of β-Glucuronidase and in silico study

2.1 Material and method

Melting points were measured in open glass capillaries using Stuart Scientific SMP11 Analog melting point apparatus. NMR spectra of benzimidazole derivatives which had been synthesized were collected using a Bruker Ultra Shield FT NMR 500 MHz spectrometer operating at 500 MHz. EI-MS was used to determine mass and fragmentation patterns were determined by analysis carried out using Finnigan-MAT-311-A instrument. Thin layer chromatography was used to monitor progress of reactions (Merck, Kieselgel 60F-254, 0.20 mm) and visualized using UV lamp at 254 nm (UVGL-58; Upland, USA).

2.2 General procedure for preparation of benzimidazole derivatives (I-X)

The compounds (I-X) intermediate was synthesized as reported previously (Taha et al., 2015). All spectroscopic data was matched with reported in recent past (Bonacci et al., 2020; Ghosh & Subba, 2015; Liang et al., 2016; Nori et al., 2020; Roudsari et al., 2020) and found same that why not reporting their data.

2.3 General procedure for preparation of benzimidazole derivatives (1–18)

The benzimidazole (1 mmol), various benzyl bromide derivatives (1 mmol), and pyridine (10 mL) were stirred at room temperature for overnight. The completion of reaction was examined by TLC. After the completion of reaction, the reaction mixture was dried, washed and recrystallized in methanol. The structures of all synthesized compounds were determined by various methods.

2.4 β-Glucuronidase assay

β-Glucuronidase activity was determined following the previously used method (Taha et al., 2016) by measuring absorbance at 405 nm of the p-nitrophenol formed substrate by a spectrophotometric method. 250 µL was the volume of the complete reaction. The reaction mixture containing 5 µL of test compound solution, 185 µL of 0.1 M acetate buffer, and 10 µL of enzyme solution were incubated for 30 min at 37 °C. At 405 nm the plates were recorded on a multiplate reader (SpectaMax plus 384) after the addition of 50 µL of 0.4 mM p-nitrophenyl-β-D-glucuronide. Experiments were performed for triplicate (Jamil et al., 2014; Mohammed Khan et al., 2012; Taha et al., 2020). To avoid molecules precipitation, compound concentrations were decreased, and the volume of the reaction was increased (200 µL). Precipitation probability was less; thus, the addition of detergents was not needed.

2.5 Methodology

For molecular docking, the method by Taha et. al. (Taha et al., 2016)] had been used. Molecular docking was carried out using GOLD software package version 4.0 (Genetic Optimization for Ligand Docking) to study the interactions between the synthesized compounds and the active site of the enzyme. In this study, receptor was treated as a rigid molecule together with flexible ligands. Three-dimensional X-ray crystal structure of human-β-glucuronidase was downloaded from protein data bank [www.rcsb.org/pdb] (PDB ID: 1BHG). Heteroatoms and chain B were removed from the dimer using Discovery Studio Visualizer and was saved as pdb file. The search space was defined by the coordinates x = 81.733, y = 82.591 and z = 89.436. The docking results obtained were analyzed using Discovery Studio Visualizer.

2.6 Cytotoxicity activity

The cytotoxicity assay was performed as we did in our previous paper (Taha et al., 2015).

4.4 General procedure for the synthesis of comp

3.1 Chemistry

We started synthesis of 2-arylbenzimidazole (I-X) compounds by treating 1,2-diaminobenzne with various aldehydes according to know procedure already reported (Taha et al., 2015). Scheme 1 Table 1. All synthesized compounds are reported which are cited (Bonacci et al., 2020; Ghosh & Subba, 2015; Liang et al., 2016; Nori et al., 2020; Roudsari et al., 2020; Z. Wang et al., 2019). The yield of all synthesized compounds ranges 87–93%.

Close

Scheme 1

Synthesis of 2-arylbenzimidazole intermediates (I-X).

Export to PPT

Table 1 Synthesis of 2-arylbenzimidazole (I-X) and their percentage yield.

Compound	R1	Yield %	Compound	R1	Yield %
I		91	VI		91
II		88	VII		93
III		92	VIII		88
IV		93	IX		87
V		89	X		90

The compounds (I-X) were treated benzyl bromide in the presence of pyridine to get our desired product (1–18). The reaction was monitored by TLC and after completion of reaction the reaction mixture was poured in cold water for rapid precipitation which was then filtered, dried and recrystallized in methanol. The structures of all (1–18) were confirmed by different spectroscopic methods Scheme 2, Table 2.

Close

Scheme 2

N-substitution on benzimidazole ring with various benzyl bromide.

Export to PPT

Table 2 New benzimidazole derivatives (1–18) and their β-glucuronidase inhibitory potential.

Compound	R1	R2	IC50 ± SEMa [μM]
1			6.20 ± 020
2			15.20 ± 0.30
3			22.10 ± 0.30
4			39.60 ± 0.70
5			25.80 ± 0.40
6			5.30 ± 0.20
7			35.90 ± 0.70
8			19.50 ± 0.40
9			2.30 ± 0.10
10			22.70 ± 0.40
11			1.70 ± 0.10
12			35.80 ± 0.70
13			8.10 ± 0.20
14			17.50 ± 0.30
15			31.40 ± 0.50
16			17.30 ± 0.30
17			1.10 ± 0.10
18			15.30 ± 0.30
Standard D-saccharic acid-1,4- lactone		48.30 ± 1.25 µM

3.2 β-Glucuronidase activity

We have synthesized benzimidazole analogues (1–18) which have diverse degree of β-glucuronidase inhibitory potential ranging between 1.10 ± 0.10 to 39.60 ± 0.70 μM when compared with standard _D-saccharic acid 1,4 lactone as standard inhibitor having IC₅₀ value 48.30 μM. All synthesized analogues (1–18) showed potent β-glucuronidase inhibitory potential better than the standard _D-saccharic acid 1,4 lactone. Structure-activity relationship studies reveal that the β-glucuronidase inhibitory potential of this class of compounds are mostly reliant upon the substitutions pattern on the benzyl ring present at N-1 and phenyl ring present at C-2 of benzimidazole skeleton. We also tested all compounds 1–18 for cytotoxicity and found that none of them is toxic.

Analog 17 displayed highest β-glucuronidase inhibitory potential among the series (Fig. 2) having chloro group on benzyl ring at 2,5-position and Nitro group on phenyl ring at para-position. To interact with enzyme, the two chloro group at 2,5-position on benzyl ring and Nitro group at para-position on phenyl ring demonstrated most appropriate arrangement. The elevated potential of analogue 17 might be due to the involvement of oxygen of nitro group in hydrogen bonding with active site of enzymes.

Close

Fig. 2

SAR study of most potent analogue.

Export to PPT

By comparing analog 17 having IC₅₀ value 1.10 ± 0.10 μM with analog 11 having IC₅₀ value 1.70 ± 0.10 μM analog 17 was found to be more potent than analog 11 (Fig. 3). In analog 17 the two chloro groups are present at ortho–meta position on phenyl ring and Nitro group at para-position on benzyl ring while in analog 11 the two chloro groups are present at ortho–para positions on benzyl ring and chloro group at para-position on phenyl ring. The little bit difference in the inhibitory potential of these analogs might be owing to the difference in position of the substituents on the benzyl and phenyl ring.

Close

Fig. 3

SAR study of analogues 17 and 11.

Export to PPT

If we compare analog 9 (IC₅₀ value 2.30 ± 0.10 μM) having para-Nitro group both on benzyl as well as on phenyl ring with analog 13 (IC₅₀ values 8.10 ± 0.20 μM) having para-chloro groups both on benzyl and phenyl ring, the analog 9 showed better β-glucuronidase inhibitory potential than analog 13 owing to the strong electron withdrawing effect offered by nitro group (Fig. 4). Similar effect of substituent position was also observed in other analogs like 2 and 10 having methoxy and methyl group.

Close

Fig. 4

SAR study of analogues 9, 13, 10 and 2.

Export to PPT

There is another interesting observation revealed when para-chloro group present on phenyl ring of analog 6 was replaced with strong electron withdrawing group like nitro group as in compound 9, the para-nitro group containing analog 9 was found to be a potential inhibitor of β-glucuronidase with an IC₅₀ value of 2.30 ± 0.10 µM (Fig. 5). Para-chloro group containing analog 6, however, found to be a week inhibitor with an IC₅₀ value of 5.30 ± 0.20 µM. Similar pattern was also observed in other analogs.

Close

Fig. 5

SAR study of analogues 9 and 6.

Export to PPT

On the basis of aforementioned observation, it was concluded from this study that the nature, number, as well as position of substituents affect the inhibitory potential of these analogs.

3.3 Molecular docking studies

The active compounds alignment showed that not much variation in the position of extended benzyl moiety with most of the variation coming from the benzene ring of the main benzimidazole scaffold. The overlaid docking results (Fig. 6) also showed that the binding cavity has no issue to deal with linear bulky groups like nitrobenzene, methylbenzene, and extended biphenyl moieties as being observed for compounds 1 (6.2 µM), 18 (25.8 µM), 9 (2.3 µM), and 10 (22.7 µM). The ability of these linear bulky moieties to fit well in the binding cavity does not necessarily correlate to better activity. This can be observed for analogs showing decent activity like compounds 2 (15.2 µM), 3 (22.10 µM), and 19 (15.3 µM) despite having bulky moieties. Even though compounds with moderately bulky substituents displayed decent inhibition activities, the results suggest that having less bulky moieties would significantly improve the activity. It was observed for compound 3 that its bulky non-linear naphthalene moiety prevents it from binding well in the binding cavity like active compound 18.

Close

Fig. 6

(a) Binding positions of active compounds in the active site of β-glucuronidase; (b) Comparison between binding position of compound 3 (blue) and 18 (yellow).

Export to PPT

Besides that, compounds with extended benzene ring having substitution at ortho (C-2) and meta (C-3) are not able to fit well in the binding site due to lack of cavity width (Fig. 7). This can be clearly observed for compounds having methoxy at meta position like compound 2 (15.2 µM) and compound with methyl substitution at ortho (C-2) like compound 19 (15.3 µM). The results for compounds 2 and 19 also indicate that the binding cavity can accommodate substituents at meta position better than substituents at ortho position. It can be clearly seen that despite no major structural differences, compound 19 has difficulty to fit the methylbenzene moiety in the binding cavity.

Close

Fig. 7

(a) Comparison between binding position of compound 2 (red) and 18 (yellow); (b) Comparison between binding position of compound 19 (blue) and 18 (yellow).

Export to PPT

In a different observation (Fig. 8), compounds with bulky substituents on the benzyl moiety, causes major decrease in inhibition activity. This was observed for least active analogs in the series, compounds 4 (39.6 µM), 7 (35.9 µM), and 12 (35.8 µM). Besides being bulky, docking results suggest that these substituents were not able to form any interaction with surrounding residues in the active site. Despite being able to fit in the binding cavity, these compounds are not able to bind the same way as observed for the most active compound 18.

Close

Fig. 8

Comparison between binding positions of compounds 4 (green), 7 (blue), 12 (red), and 18 (yellow).

Export to PPT

The molecular docking result for compound 18 is correlating well with proposed structure–activity relationship (SAR). As mentioned earlier, the nitro and dichloro substituents are expected to be the main reason for the excellent inhibitory activity displayed by compound 18. Molecular docking result for compound 18 suggests that both, nitro group and chlorine atoms, can form interactions with crucial amino acids (Fig. 9). The result showed that both chlorine atoms can form hydrophobic π-alkyl interactions with the π-orbital of phenol sidechain of Tyr508 while being by a π-π stacking interaction between the benzene ring and phenol sidechain of Tyr508. On the other end, the oxygen on the nitro group can act as a hydrogen bond acceptor to form a carbon hydrogen bond with the hydrogen on the backbone (HA) of Tyr205 (2.51 A). Another carbon-hydrogen bond was observed between benzylic hydrogen of compound 18 and the oxygen on backbone (Oe1) of Glu451 (2.90 A). Other supporting interactions include electrostatic π-anion interactions between imidazole scaffold and two active site residues, Asp207 and Glu451.

Close

Fig. 9

(a) 2D interaction diagram of compound 18; (b) Binding position of compound 18 in the active site of β-glucuronidase.

Export to PPT

The results (Fig. 10) displayed comparison between docking results for compound 11 and 18 to confirm that replacing of nitro substituent with dichloro reduces the number of interaction for compound 11. It was found that the dichloro substituents on the benzylic moiety of compound 11 were not able to form any interaction with the active site as compared to nitro substituent for compounds 18. It was also observed that replacing dichloro substitution on compound 18 with a single chlorine substitution at para position does not affect the original interaction much. The single chlorine atom on compound 11 is still able to form hydrophobic π-alkyl with π-orbital of phenol sidechain of Tyr508. Besides that, compound 11 is still able to maintain the carbon-hydrogen interaction between the benzylic hydrogen of compound 11 and the backbone (Oe1) of Glu451. Other interactions available includes support through electrostatic π-anion interaction between imidazole scaffold and the two active residues, Asp207 and Glu451.

Close

Fig. 10

(a) 2D interaction diagram of compound 11; (b) Binding position of compound 11 in the active site of β-glucuronidase.

Export to PPT

Comparison between docking results for compounds 9 (2.30 µM) and 13 (8.10 µM) suggest that the presence of nitro group at para (C-4) position on the benzylic ring is very crucial (Fig. 11). The result indicates that the nitro on the benzylic ring of compound 9 can form electrostatic attractive charge interaction with the backbone (OD1) of Asp207 while chlorine on the benzylic ring of compound 13 was not observed to form any interaction. As for the substituents on the imidazole scaffold, nitro and chloro substituents on both compounds, 9 and 13, are interacting well with Tyr508. The nitro group of compound 9 can form an electrostatic π-cation interaction with phenol sidechain of Tyr508 while the chloro substituent of compound 13 forms a hydrophobic π-alkyl interaction with the same residue. For both compounds, the imidazole moieties are supported by electrostatic π-anion interactions with Asp207 and Glu451. This pattern was also observed for compounds 1 and 10 in which replacing nitro substituent with methoxy prevents the compound from interacting with catalytic residue Asp207, causing decrease in inhibition activity.

Close

Fig. 11

The 2D interaction diagrams for compounds 9, 13, 1, and 10.

Export to PPT

Comparison between docking results for compounds 9 and 6 showed that replacing chloro substituent with a strong electro withdrawing group like nitro will allow formation of additional potential interactions (Fig. 12). It can be observed that the nitro of compound 28 is able to form stronger electrostatic π-anion interactions with the phenolic sidechain of Tyr508. On the other hand, chloro substituent can only form less effective hydrophobic π-alkyl interaction with Tyr508. Despite showing better electrostatic interactions compared to compound 6, the bulky features of compound 9 prevents it from forming a crucial carbon-hydrogen bond that can be observed between benzylic hydrogen of compound 6 and the backbone (Oe1) of Tyr508 (2.85 A).

Close

Fig. 12

The 2D interaction diagrams for compounds 9 and 6.

Export to PPT

Comparison between compound 1 and 14 showed how replacing nitro substituent to nitrile affects the binding interaction. The structure overlay of compounds 1 and 14 (Fig. 13), showed that the binding position and binding interaction of the main benzimidazole scaffold are the same for both compounds. Further analysis showed that replacing the nitro substituent on compound 1 with nitrile causes the benzyl moiety flip in a different direction as compared to benzyl ring with nitro substitution. This flip might be due to the rigidity of nitrile group for compound 14 which in return prevents the nitrile group from binding well in active site. The 2D interaction diagram of compound 14 also suggests that nitrile group was not able to form interaction with residues in the binding cavity, therefore causing compound 14 to displayed lower inhibition activity compared to compound 1.

Close

Fig. 13

(a) The 2D interaction diagrams for compounds 14; (b) Binding position of compounds 1 and 14 in the active site of β-glucuronidase.

Export to PPT

3.4 In silico ADME analysis

In silico ADME analysis had been performed on all compounds. Based on Fig. 14, it was predicted that nearly all compounds can be absorbed through the human intestinal tract. Two compounds, 7 and 11, are not predicted to be within the 99% of well-absorbed compounds (Absorption-99 elipse), indicating potentially low absorption through the human intestinal tract. Within the Absorption-95 elipse, indicating 95% of well-absorbed compounds, only 9 compounds (1, 2, 6, 8, 10, 14, 15, 17, and 19) were predicted to be able to get through the human intestinal tract. These compounds (1, 2, 6, 8, 10, 14, 15, 17, and 19) were also predicted as compounds that are within the 99% of compounds which can pass through the blood brain barrier (BBB). However, only 5 compounds (2, 8, 10, 14, and 15) were found to among the 95% of compounds that can be well-absorbed through the BBB. In Table 3, the aqueous solubility of most compounds were predicted to have low solubility with two compounds (7 and 11) displaying extremely low aqueous solubility. It was also predicted that all compounds are not CYP2D6 inhibitor except for compound 7.

Close

Fig. 14

Plot of Polar Surface Area (PSA) vs. LogP of all compounds corresponding to the Blood Brain Barrier and Intestinal Absorption models.

Export to PPT