Multi Targeted Ligands for Potential Inhibition of Dipeptidyl Peptidase

4, Acetylcholinesterase and Cyclooxygenase 2

 

Minhajul Arfeen*, Ruba Alqasem, Mashal Alwahabi

Department of Medicinal Chemistry and Pharmacognosy, College of Pharmacy,

Qassim University, Buraydah, 51452, Saudi Arabia.

*Corresponding Author E-mail: m.arfeen@qu.edu.sa

 

ABSTRACT:

Two series of novel compounds were designed by combining indomethacin and ibuprofen with sixteen sulfa drugs. These compounds were systematically evaluated through target fishing using the Pharm Mapper, leading to the identification of DPP-4, AChE, and COX-2 as potential targets. Molecular docking was performed to evaluate the binding affinity of designed compounds against the identified three target proteins. The results revealed that the designed compounds exhibited binding affinities ranging from ~8 to -12kcal/mol, 12 to 13 kcal/mol and 8 to 11kcal/mol for DPP-4, AChE and COX-2 respectively. The binding affinities were found to be comparable or higher than binding affinity of co-crystallized ligand, which was found to be ~10, 12 and 9 kcal/mol respectively. Further investigation into the binding modes of these compounds was carried out. Notably, for DPP-4 complexes, interactions with Arg125, Glu205, and Glu206 were observed which are essential for substrate and inhibitor binding. For AChE complexes, interactions involved crucial His447 residues, essential for acetylcholine hydrolysis. In the case of COX-2, hydrogen bond interaction was noted with Arg120 located at the entrance of the hydrophobic channel. Despite favorable binding potentials, ADME profiling highlighted five compounds (1A, 1F, 1G, 1H, and 1O) with drug-like characteristics but lacking blood-brain barrier permeation ability. Out of five compounds, 1H stood out, demonstrating superior binding affinity and interactions vital residues necessary for catalytic activity of three enzymes. Thus, 1H emerges as a promising candidate for Multi-Targeted Drug-Like (MTDL) development aimed at addressing diabetes mellitus related dementia.

 

KEYWORDS: Molecular Docking simulation, AChE, COX-2, Diabetes, Dementia, Inflammation.

 

 


INTRODUCTION: 

Diabetes, a widespread metabolic disorder characterized by chronic hyperglycemia, exerts a significant global health impact. Beyond its influence on glucose regulation, diabetes affects various organs and cognitive function. Individuals with diabetes face an elevated risk of cognitive impairment1. Research indicates that Type 2 diabetes (T2D) alone escalates dementia risk by up to 60%, with poorly managed blood sugar exacerbating this association specially in adults2.

 

The merging of nonsteroidal anti-inflammatory drugs (NSAIDs) with sulfa drugs marks an exciting frontier in medicinal chemistry,3,4 driven by the goal of synergizing the distinct therapeutic potentials of these compound classes5,6. NSAIDs, such as ibuprofen and indomethacin, are recognized for their ability to alleviate inflammation and provide pain relief through COX enzyme inhibition. Moreover, the intriguing potential of sulfa drugs to contribute to hypoglycemic effects adds an intriguing dimension to this hybridization approach7,8. Sulfa drugs have shown promise in aiding glycemic control, potentially offering a novel angle for addressing both inflammatory conditions and diabetes. Through the deliberate integration of the strengths inherent in NSAIDs, sulfa drugs, and their possible antidiabetic properties, this hybridization strategy holds the potential to deliver fresh therapeutic agents with heightened effectiveness and adaptability. In this work, ibuprofen and indomethacin was hybridized with sixteen sulfa drugs to generate thirty-two compounds in two series (figure 1). The sulfa drugs used for hybridization of pharmacophore were sulfanilamide, sulfamethoxazole, sulfisoxazole, sulfametrole, sulfathiazole, sulfapyridine, sulfadiazine, sulfamerazine, sulfamethazine, sulfadimethoxine, sulfamethoxypyridazine, sulfachloropyridazine, sulfalene, sulfaquinoxaline, sulfacetamide and sulfaguanidine The designed compounds were investigated for binding potential against COX-2, DPP-4 and AChE using molecular docking. It should be noted that the dementia is characterized by elevated levels of COX-2,9 while DPP-4 and AChE is a molecular target of treatment of diabetes10 and dementia respectively.11,12 Besides drug like properties were also evaluated for newly designed molecules.

 

Figure 1: Structure of designed molecules obtained through hybridization of ibuprofen and indomethacin with sulfa drugs.

 

MATERIALS AND METHODS:

Molecular docking experiments were conducted using AutoDock Vina, building upon methods outlined in prior studies.13,1,14 The input files were generated using MGL tools (ver 1.5.6) Grids were established using AutoDock's autogrid4 program (version 1.5.6)15 with specific considerations for each protein target. The grids for the proteins was generated by targeting the active binding site corresponding to ligand position. The proteins with codes 2OQV, 5IKR and 4EY7 and 4TPK corresponding to DPP-4, COX-2, AChE and hBuChE were downloaded from protein data bank. The proteins were prepared by deleting chain B, water and other metal ions. Polar hydrogens were added, missing atoms were added, Kollman charges were added, and input files were saved in pdbqt format. Similarly, for ligand preparation, the structures were generated using chemdraw, minimized using MM2 forcefield and saved into mol2 format. The corresponding mol2 files were imported using AutoDock tools, root and torsions were defined, Gasteiger charges were added followed by saving input files in pdbqt format.  A configuration text file (config.txt) was generated to execute molecular docking. For DPP 4, the X, Y and Z grid coordinates were set at 15.296, 54.304, and 22.388 respectively. For COX-2, the grids coordinates were centered at 38.042, 2.131, 61.28. For AChE, the coordinates determined were -13.988, -43.906, 27.109 and for BuChE, the coordinates used were 4.401, 10.0, 13.912. The box size of the four proteins DPP-4, COX-2, AChE and BuChE used in this study were 30 X 30 X 30, 40 X 40 X 40, 40 X 40 X 40 and 25 X 25 X 25 Å respectively. The configuration file encompassed critical components, including receptor and ligand PDBQT files and the central grid box coordinates. The "--score_only" flag facilitated re-scoring using AutoDock Vina 1.1.2, leading to docking scores recorded in resulting .log files. These scores denoted binding affinities measured in Kcal/mol. Subsequent investigation into diverse protein-ligand interactions, such as hydrogen bonds, cation-π bonds, π-π bonds, and alkyl-π interactions within active site amino acids and hybrid compounds were performed using Discovery Studio. Druglikeness for the designed molecules were evaluated using webserver SwissADME 16.

 

 

Figure 2: Comparison of Co-crystalized and redocked ligand (A) DPP-4, (B) AChE and (C) COX-2.

 

RESULTS:

The designed hybrid compounds were evaluated for their binding potential using molecular docking. For each of the hybrid compounds ten binding modes were generated. Only top three binding modes were evaluated and the binding modes which displayed interactions with key residues was considered for discussion. Before, initiating the molecular docking of designed compounds, methodology was validated by extracting and re-docking co-crystallized ligand. The binding affinity for the docked ligand was compared with the orientation of co crystalized ligand. The RMSD values for the re-docked ligands were found to be ~1 in all three complexes (figure 2).


 

Table 1: Binding Scores of designed compounds for three enzymes considered in the study.

Sr. No.

Code

DPP-4

AChE

COX-2

Sr. No.

Code

DPP-4

AChE

COX-2

Co-L

-10.0

-12.0

-9.1

Co-L

-10.0

-12.0

-9.1

1

1A

-7.6

-10.1

-7.6

17

2A

-10.2

-11.0

-9.5

2

1B

-9.3

-11.9

-8.6

18

2B

-11.1

-13.1

-9.9

3

1C

-8.8

-11.3

-8.9

19

2C

-11.8

-13.2

-10.2

4

1D

-8.2

-11.5

-8.9

20

2D

-10.8

-12.5

-9.3

5

1E

-8.4

-11.2

-8.3

21

2E

-11.3

-11.4

-9.2

6

1F

-9.2

-11.9

-8.9

22

2F

-11.7

-13.1

-8.8

7

1G

-9.2

-11.4

-8.4

23

2G

-11.8

-11.5

-9.3

8

1H

-9.2

-12.0

-9.0

24

2H

-10.9

-12.2

-9.8

9

1I

-9.4

-12.0

-9.3

25

2I

-11.1

-13.0

-9.2

10

1J

-9.3

-11.6

-8.8

26

2J

-10.7

-12.4

-8.9

11

1K

-9.1

-11.6

-8.3

27

2K

-10.8

-12.8

-10.7

12

1L

-9.4

-12.3

-8.9

28

2L

-11.4

-13.2

-10.8

13

1M

-9.0

-11.7

-8.3

29

2M

-10.9

-11.8

-9.5

14

1N

-9.8

-12.2

-10.0

30

2N

-11.7

-13.0

-10.0

15

1O

-8.4

-10.9

-8.3

31

2O

-11.0

-11.3

-9.6

16

1P

-8.3

-11.6

-8.3

32

2P

-10.5

-12.9

-9.1

 


DPP-4: The molecular docking of the designed ligands on DPP-4 using crystal structure 2OQV showed good to high binding affinity. The calculated binding affinity for designed ligand varied in the range of ~7.6 to ~12 kcal/mol (Table 1). It should be noted that indomethacin hybridized compounds displayed high binding affinity as compared to the ibuprofen counterparts. The highest binding affinity was noted for compounds 2C, 2G and 2N (~11.8kcal/mol). Further, the indomethacin derived compounds displayed binding affinities either higher or comparable to co-crystallized ligand (~10kcal/mol). The ibuprofen-derived compounds displayed either marginally lower or comparable binding affinities to co-crystallized ligand. The observed binding affinity for the indomethacin derived and ibuprofen derived compounds was in the range of ~10 to 12 and ~8 to 10kcal/mol respectively. Further, to note that designed compounds bearing sulfanilamide (1A and 2A) and quinoxaline rings (1N and 2N) displayed lowest and highest binding potential respectively. Considering, the binding scores displayed in table 1, it can be safely stated that indomethacin derived compounds displayed high binding affinity for DPP-4 enzyme, while ibuprofen derived compounds displayed moderate to significant binding affinity.

 

Figure 3: Bar diagram displaying % frequency of residue wise interaction of DPP-4 com-plexes. A and B displays the polar and hydrophobic interactions for ibuprofen-derived compounds while C and D displays polar hydrophobic interactions of indomethacin compounds.


 

Figure 4: Binding modes of Compounds 1B (A, sub site S1, S2, S1’ and S2’), 1F (B, S1, S2 and extended S2 region) and 2C (C, S1, S2 and extended S2 region) and 2C (C).

 


The small molecule binding site of DP-4 can be categorized into five sub-pockets i.e. S1, S2, S1’ S2’ and S2 extended. The sub pocket S1 is made up of residues Ser630, Tyr662 and Tyr666 1718 . The sub-pocket S2 is made up of residues Glu205, Glu206, and side chains of Arg125 and Tyr662. The S1’ and S2’ sub-site is maintained by the side chain of Tyr547 and residue Trp629 respectively while the S2 extended sub-site is formed due to Arg358. The binding mode analysis of ibuprofen-derived compounds showed two types of orientations, i.e. 1A, 1B, 1H-L and 1P occupied sub site S1, S2, S1’ and S2’ (figure 4A), whereas 1C, 1D, 1E, 1F, 1G, 1M, 1N and 1O, occupied S1, S2 and extended S2 region (figure 4B). In terms of polar interactions, two sulfonyl oxygen atoms of the compounds occupying S1, S2, S1’ and S2’ region showed anion-pi interactions with Glu205, Glu206. In addition to this cation-pi) interaction between the sulfonyl oxygen with Arg125 and a hydrogen bond between amide nitrogen of ligands with side chain of Tyr547 was observed. The residues showing hydrophobic interactions with the first set of compounds are Tyr547, Trp627, Tyr662 and Tyr666. For the second set of compounds, the sulfonyl oxygen atoms showed polar interactions with Arg125 and Ser630, the sulfonamide nitrogen showed hydrogen bond interactions with Tyr547 and amide carbonyl oxygen atom displayed anion-pi interaction with backbone carbonyl oxygen of Glu205. The residues showing hydrophobic interactions with the second set of compounds are Arg356, Phe357, Arg358, Val656 and Ty662. Compounds 1D and 1M, though they occupied S1, S2 and extended S2 region of the active site, but their orientation was different as compared to the other compounds of same set and hence were not discussed. The percentage frequency levels (figure 3A) of polar interactions for the residues Arg125, Glu205, Glu206, Tyr547 and Tyr662 are ~ 170, 150, 120, 120 and 110 respectively. Other notable polar contribution was found to be from Ser630 (~80%). The hydrophobic residues displaying notable contributions are Tyr666, Tyr662 and Phe357. Their percentage frequency levels are ~100, 80 and 80 respectively. Other notable contributions were from residues Arg356 (~43 %), Tyr547 (~50 %), Tyr631 (~30 %) and Val656 (38 %) (Figure 3B). The binding mode analysis of indomethacin-derived compounds (2) occupied S1, S2 and extended S2 sub-site of the active site (figure 4C). It should be noted that binding modes of the docked ligands were selected based on the polar interaction with two catalytic residues Glu205 and Glu206. The amide carbonyl oxygen of 2A to 2O displayed pi-anion interaction with carbonyl oxygen of Glu205. Further, the amide NH of 2C, 2F, 2H, 2I, and 2J displayed hydrogen bond interaction with side chain of Glu206. Besides, the two sulfonyl oxygen atoms from the 2A to 2O displayed hydrogen bond interactions with Arg125 and Tyr547. The Arg125 also displayed cation pi interaction with benzene ring of the ligands. The percentage frequency (figure 3C) for the Glu205, Glu206, Arg125 and Tyr547 were ~100, 56, 190 and 110 respectively. In addition to these residues, polar contributions from Ser630 and Tyr662 was also observed in significant number of complexes. Their percentage frequency was found to be ~90 and 120 respectively. Other notable contributions observed are Arg358 (~44%). The hydrophobic interactions mainly involved residues Arg356, Phe357 and Arg358. The percentage frequencies among the indomethacin derived compounds were found to be ~190, 200 and 90 respectively (figure 3D). Other notable hydrophobic contributions were from residues His740 (56%), Trp629 (~37%), Tyr662 (~37%), Tyr666 (~32%) and Val656 (~38%). 2N, the two sulfonyl oxygen atoms and guanidine group displayed hydrogen bond interactions with Arg125, Glu205 and Glu206, while hydrophobic interaction involved residue Arg356.

 

Figure 5: Bar diagram displaying % frequency of residue wise interaction of AChE complexes. A and B displays the polar and hydrophobic interactions for ibuprofen derived compounds. C and D displays polar hydrophobic interactions of indomethacin compounds.

 

AChE: Similar, to the results for DPP-4, the designed ligands showed significant binding potential for AChE. The binding affinity for the co-crystallized ligand was found to be 12 kcal/mol (table 1). The binding affinity of indomethacin-derived compounds were observed to be higher than ibuprofen compounds. The binding affinities for ibuprofen-derived compounds varied in the range of ~11 to 12kcal/mol, comparable to co-crystalized ligand. The compounds 1H, 1I, 1L and 1N displayed the binding scores of 12.0, 12.0, 12.3 and 12.2kcal/mol, while other ibuprofen compounds showed slightly lower binding potential than the co-crystallized ligand. For indomethacin derived compounds the binding affinity varied in the range of 11 to 13kcal/mol. The compounds 2B, 2C, 2F, 2L, 2N and 2P displayed binding affinity higher than co-crystallized ligand. The observed binding affinities for above mentioned compounds are 13.1, 13.2, 13.1, 13.2, 13.3 and 12.9kcal/mol respectively. The other indomethacin derived compounds showed binding scores comparable to co-crystallized ligand except for 2A. As noted from the results of DPP-4, the lowest binding affinity was observed for sulfanilamide derived compounds from ibuprofen and indomethacin i.e 1A and 2A (10.7 and 11kcal/mol respectively).

 

The binding site of AChE is divided into five sub regions (i) Anionic site, (ii) catalytic triad, (ii) oxyanion hole (iv) peripheral aromatic site and (v) acyl binding pocket. The anionic site is made up of residues Trp86, Tyr337 and Phe338. The catalytic triad is composed of the residues Ser203, Glu334 and His447 while, the oxyanion hole is made up of Gly120, Gly121 and Ala204. The acyl binding pocket is made up of residues Trp236, Phe295 and Phe297 and The peripheral binding site (P-site) is composed of residues Tyr72, Asp74, Tyr124, Val282, Glu285, Trp286 and Tyr341 19. The analysis of binding modes revealed similar orientation of all designed molecules. The designed molecules majorly occupied anionic site and P-site in addition to being in close proximity with catalytic triad Figure 6A and 6B). In the case of ibuprofen-derived molecules, the benzene ring associated with isopropyl group was noticed to occupy P-site, while the sulfonyl group and the heterocyclic ring associated with sulfonamide functional group occupied the anionic site. In case of indomethacin derived compounds indole ring coupled with p-chloro benzoyl group was observed to occupy anionic site, while the benzene ring associated with sulfonamide connected to heterocyclic ring system occupied the P-site except for 1A, 2M, 2P and 2O. For 1A, 2M, 2P and 2O the indole ring coupled with p-chloro benzoyl group occupied P-site and benzene ring associated with sulfonamide connected to heterocyclic ring system occupied the anionic site. The residue wise interaction analysis of ibuprofen derived compounds showed Tyr124, His447 displaying polar interactions in all the complexes. From the Figure 5A, it is evident that residues Ty124 and His447 were involved in two polar interactions for some of the complexes (percentage frequency of 12 and 160 respectively). In addition to the above mentioned residues, Glys121, Ser125 and Trp86 were also involved in polar interactions in significant number of complexes. In the case indomethacin derived compounds Tyr124 showed hydrogen bond interactions in all the complexes (figure 5C). The percentage frequency for Tyr124 was found to be ~170. In addition to Tyr124, Ser293 also displayed hydrogen bond interactions in significant number of designed ligands. The percentage frequency was found to be ~80. In addition to these, others residues displaying polar interactions and worth mentioning are Tyr341 and His447. Their percentage frequency are ~30. It should be noted that amide nitrogen and amide carbonyl oxygen was involved in hydrogen bond interactions with Tyr124 and Tyr341 respectively. while Benzoylic oxygen atom and heterocyclic ring from the ligand was involved in polar interactions with Ser293 and His447 respectively. With respect to the hydrophobic interactions, among the ibuprofen derived compounds Trp286, Trp86, Tyr341 and His447 displayed the percentage frequency levels of ~250, 190 and 75 respectively (figure 5B). In addition to these other residues displaying significant contribution to hydrophobic interactions are Tyr124 and Tyr72. Their percentage frequency levels were ~43 and 55 respectively. For the indomethacin derived compounds and as observed for the ibuprofen derived compounds Trp286, Trp86 and His 447 displayed interactions in all the complexes (figure 5D). Their percentage frequencies are 260, 180 and 130 respectively. In addition to above mentioned residues, Tyr341 displayed the frequency level 100 percent indicating hydrophobic interaction in all the complexes. Besides, Leu130, Leu289, Tyr124 and Tyr337 displayed hydrophobic contribution in 80% of the designed indomethacin compounds.

 

 

Figure 6: Binding modes of Compounds 1H (A) and 2L (B) in the active site of AChE

 

COX-2: As observed for DPP-4 and AChE, the designed compounds showed good binding potential for COX-2. The calculated binding affinity for the co-crystallized ligand was found to be ~9 kcal/mol. The binding affinity for the designed compounds varied in the range of ~7.5 to ~11kcal/mol (table 1). Similar, to the results observed for DPP-4, the indomethacin hybridized compounds showed higher binding affinity as compared to their ibuprofen derived compounds. The highest binding scores were observed for 2C, 2K and 2L. Their binding scores varied in the range of ~10 to ~11kcal/mol and hence indicating good binding potential for COX-2 enzyme. Other indomethacin derived compounds displayed binding scores comparable to the co-crystallized ligand indicating considerable binding potential for COX-2 enzyme. Among ibuprofen derived series of compounds 1N displayed good binding potential, while all other designed compounds displayed marginally lower binding potential than co crystallized ligand. The observed binding affinity range for ibuprofen derived compound was ~ ~8kcal/mol to 10 kcal/mol, with most of the compounds displaying binding potential of ~9kcal/mol. For the indomethacin derived compounds, the average value of binding affinity was found to ~10kcal/mol. The ibuprofen hybridized with sulfanilamide displayed lowest binding potential (7.6 kcal/mol) for COX-2. Similar to results from DPP-4, it can be safely envisaged that all the designed compounds displayed moderate to good binding potential for COX-2.

 

The cyclooxygenase enzyme consists of two active sites cyclooxygenase site and peroxidase site. The two active site of COX are connected through a hydrophobic channel. The substrate binds to the cyclooxygenase site and access to the peroxidase site through the hydrophobic channel. The hydrophobic channel is made up of non-polar residues such as Val116, Tyr348, Val349, Leu352, Leu359, Phe381, Leu384, Tyr385, Trp387, Phe518, Ile523, Ala527 and Leu531 with Arg120 and Tyr355 acting as the guard residues at the entrance of hydrophobic channel 20. The binding mode analysis of ibuprofen hybridized compounds showed sulfonyl oxygen atoms interacting with the side chain of Arg120 through two hydrogen bonds in all the compounds (figure 8). The percentage frequency for Arg120 among the ibuprofen derived compound was found to be ~190 (figure 7A). The ibuprofen derived compounds also displayed hydrogen bonds with Ser119 among all ibuprofen derived compounds. While other 50 percent of the compounds displayed hydrogen bond interaction with Tyr355. In addition, these observed polar interaction, the other residues which displayed hydrogen bond interaction in few of the compounds are Tyr115. The percentage frequency for Ser119 and Tyr355 was found to be ~100 and ~50% respectively The residues displaying hydrophobic interactions for the majority of ibuprofen derived compounds are Val89, Ile92, Leu93, Phe96, Phe99, Tyr115, Val116 and Arg120. Their percentage frequency levels are ~106, 206, 75, 125, 125, 81, 62 and 57 respectively (figure 7B). Other notable residue displaying hydrophobic bond interaction is Leu123. Its percentage frequency among the designed compound was found to ~44. For the indomethacin-sulfa derived compounds, the hydrogen bond interaction with Ser119 and Arg120 was noticed at the percentage frequency levels of ~ 100 and 112 respectively while hydrogen bond interaction with Tyr115 was observed among 50% of the compounds (figure 7C). The other residues displaying hydrogen bond interactions were Glu524 (30%) and Lys79 (25%). The residues displaying hydrophobic interactions among majority of the complexes are Leu112, Ile92, Met471, Phe96, Phe99, Trp100, Tyr115 and Val89. Their percentage frequency levels among the indomethacin derived compounds was found to be ~ 118, 43, 62, 56, 112, 43, 112 and 144 respectively (figure 7D). Other notable hydrophobic contribution was observed from residue Val116 (~30%).

 

 

Figure 7: Bar diagram displaying % frequency of residue wise interaction of COX-2 complexes. A and B displays the polar and hydrophobic interactions for ibuprofen-derived compounds. C and D displays polar hydrophobic interactions of indomethacin compounds.

 

Figure 8: Binding mode of compound 1H and 2B in the active site of COX-2.


 

Table 2: ADME profile for the designed compounds.

M. ID.

MW

FCsp3

RB

HBA

HBD

TPSA

MLogP

Log S

ESOL Class

GIA

BBBP

1A

360.47

0.32

7

4

2

97.64

2.59

-4.12

Moderately soluble

High

No

1B

441.54

0.3

9

5

2

109.68

2.78

-5.22

Moderately soluble

Low

No

1C

455.57

0.33

9

5

2

109.68

2.99

-5.37

Moderately soluble

Low

No

1D

474.6

0.32

10

6

2

146.9

2.04

-5.33

Moderately soluble

Low

No

1E

443.58

0.27

9

4

2

124.78

2.55

-4.72

Moderately soluble

Low

No

1F

437.55

0.25

9

4

2

96.54

2.94

-4.66

Moderately soluble

High

No

1G

438.54

0.26

9

5

2

109.43

2.33

-4.91

Moderately soluble

High

No

1H

452.57

0.29

9

5

2

109.43

2.55

-4.83

Moderately soluble

High

No

1I

466.6

0.32

9

5

2

109.43

2.75

-5.56

Moderately soluble

Low

No

1J

498.59

0.32

11

7

2

127.89

2.53

-5.89

Moderately soluble

Low

No

1K

468.57

0.29

10

6

2

118.66

2.84

-4.96

Moderately soluble

Low

No

1L

472.99

0.26

9

5

2

109.43

3.22

-5.49

Moderately soluble

Low

No

1M

468.57

0.29

10

6

2

118.66

1.62

-5.2

Moderately soluble

Low

No

1N

488.6

0.22

9

5

2

109.43

3.12

-6.07

Poorly soluble

Low

No

1O

402.51

0.33

9

4

2

100.72

2.16

-4.33

Moderately soluble

High

No

1P

402.51

0.3

8

4

3

136.02

3.16

-3.96

Soluble

Low

No

2A

481.95

0.08

7

5

2

119.64

3.08

-5.46

Moderately soluble

Low

No

2B

563.02

0.11

9

6

2

131.68

3.23

-6.41

Poorly soluble

Low

No

2C

577.05

0.14

9

6

2

131.68

3.83

-6.71

Poorly soluble

Low

No

2D

596.08

0.11

10

7

2

168.9

2.67

-6.48

Poorly soluble

Low

No

2E

596.08

0.11

10

7

2

168.9

2.67

-6.48

Poorly soluble

Low

No

2F

559.04

0.07

9

5

2

118.54

3.49

-6.46

Poorly soluble

Low

No

2G

560.02

0.07

9

6

2

131.43

2.93

-6.22

Poorly soluble

Low

No

2H

574.05

0.1

9

6

2

131.43

3.12

-6.38

Poorly soluble

Low

No

2I

588.08

0.13

9

6

2

131.43

3.31

-6.88

Poorly soluble

Low

No

2J

620.08

0.13

11

8

2

149.89

3.13

-6.63

Poorly soluble

Low

No

2K

590.05

0.1

10

7

2

140.66

3.43

-6.11

Poorly soluble

Low

No

2L

594.47

0.07

9

6

2

131.43

3.80

-6.63

Poorly soluble

Low

No

2M

590.05

0.1

10

7

2

140.66

2.21

-6.07

Poorly soluble

Low

No

2N

610.08

0.06

9

6

2

131.43

3.65

-7.01

Poorly soluble

Low

No

2O

523.99

0.12

9

5

2

122.72

3.04

-5.67

Moderately soluble

Low

No

2P

523.99

0.08

8

5

3

158.02

3.68

-5.3

Moderately soluble

Low

No

 


ADME Properties:

Understanding the pharmacokinetic behavior (ADME) of new molecule, holds a crucial role in drug development. To assess potential drug candidates, researchers carefully analyze physicochemical attributes like molecular weight (MW), hydrogen bonding (HB) capabilities, and lipophilicity (LP). Our results (table 2) from the in silico-ADME profiling showed that none of the ibuprofen derived compounds disobeyed Lipinski rule, while indomethacin derived compounds, except for 2A showed one violation which is MW higher than 500g/mol. The drug and non-drug like character of ibuprofen and indomethacin derived compounds is further evident from higher fraction of sp3 hybridized carbon over total count of the carbon (F Csp3) which should not be less than 0.25. The results indicated the F Csp3 value in the range of 0.22 to 0.33 and 0.07 to 0.12 for ibuprofen and indomethacin derived compounds respectively. The rotatable bonds were within the acceptable range excepts for  1D, 1J, 1K, 1M, 2D, 2E, 2J, 2K and 2M where it was noted to be more than nine. The designed compounds also showed less than five hydrogen bond donors and no more than ten hydrogen bond acceptors. The total polar surface area (TPSA) for the designed compounds were also found be in the optimal range i.e within 20 to 130 Å2 except for 2J, 2K, 2M and 2P. It should be noted that TPSA is an important parameter during drug development phase as it helps in assessing biovalaiblity, absorption, permeability, transporter interactions, hydrogen bonding and drug interactions. In terms of LP, the designed compounds, both ibuprofen and indomethacin derived compounds showed moderate LP as indicated from the value of MLogP (~2 to 4). The solubility of the designed compounds was estimated by calculating Log S, which predicted ibuprofen and indomethacin derived compounds as moderately and poorly soluble respectively. In terms of intestinal absorption, 1A, 1F, 1G and 1H are predicted to have high absorption, while all other compounds were predicted to have low absorption and further and no compounds showed the ability to cross blood brain barrier. The designed compounds were also predicted to be inhibitor of cytochrome 2C19, 2C9 and 3A4 and thus may show drug interactions. The designed compounds showed bioavailability value of 0.55, implying only 55% of the administered dose will reach blood, while remaining 45% will be lost.

DISCUSSION:

Cognitive impairment is one of the many implications from log term Diabetes mellitus. DPP-4 is a very well established target for the treatment of diabetic conditions. Further, AChE is one of the very well established target for the treatment of cognitive impairment. Further, both diabetic and cognitive impairment is characterized by elevated levels of      COX-2 21. In this work, thirty-two molecules were designed by hybridizing ibuprofen and indomethacin with sixteen sulfa drugs through amide linkage. The idea of hybridizing two molecules was based on the fact that ibuprofen and indomethacin are well established anti-inflammatory drugs and sulfa drugs are reported for hypo-glycemic action. The designed molecules subjected to target fishing using pharmmapper lead to identification of DPP-4, AChE and COX-2 as potential molecular targets. Thus, designed molecules were evaluated for binding potential against DPP-4, AChE and COX-2 using molecular docking. The designed compounds were also evaluated for ADME using Swiss ADME. The results from molecular docking displayed high binding potential to all three molecular targets for the two series of compounds. The indomethacin derived compounds displayed high binding potential as compared to ibuprofen derived compounds. The high binding affinity of the indomethacin derived compounds can be attributed to presence of indole ring associated with the 4-chloro benzene ring through amide linkage thus providing more opportunity for hydrophobic and polar. Further, it was observed that sulfanilamide derived compounds displayed low binding potential as compared to its counterparts in two designed series. This variance in the observed binding affinity is due to the presence of additional heterocyclic ring system on the compounds other than sulfanilamide derivatives. For each of the compounds, three binding modes were generated and considered for evaluation. It should be noted that there was very marginal energy difference among top three binding modes and hence final binding mode for discussion was selected depending on the interactions with key residues. The binding mode analysis of designed compounds into the active site of DPP-4 showed polar interactions with residues Arg125, Glu205, Glu206, Ser630, Tyr547 and Tyr662, while hydrophobic interactions were observed with Arg356 and Phe357. It should be noted that Arg125 interacts with the carbonyl group of P1’ prime residue from the substrate and thus helps stabilizing the substrate in the substrate binding site22. Further, the above mentioned residues displaying polar interactions are critical for catalytic activity23, and also to be noted that reported crystal structures of the marketed drugs have shown, that DPP-4 inhibitors interact strongly with the Glu205 and Glu20624. Therefore, it can be envisaged that the designed compounds have high potential to inhibit DPP-4. The binding mode analysis reveals that all indomethacin derived and few ibuprofen derived 1C, 1D, 1E, 1G, 1M, 1N and 1O have the potential to act as class I DPP-4 inhibitors as it occupies S1, S2 and extended S2 region. While 1A, 1B, 1H-L and 1P has the potential to act as class III inhibitor (occupies S1, S2, S1’ and S2’)25. The binding mode analysis of the docked complexes of AChE showed hydrogen bond interaction with His447 and Tyr124 among ibuprofen derived compounds while among indomethacin derived compounds Tyr124 showed polar interaction, while His447 displayed hydrophobic interaction. It should be noted His447 is an important residue for hydrolysis of acetylcholine26,27, while Tyr124 is the most frequently found interactions among the reported AChE-inhibitor28. Additionally, the designed compounds also showed hydrophobic interactions with Trp86 (anionic site) and Trp286 (P-site) and are important for inhibitor binding29. With respect to the binding mode analysis of designed ligands among the complex of COX-2, both ibuprofen and indomethacin derived compounds displayed polar interactions with Arg120 and Ser119. Arg120 is an important residue from the perspective of substrate binding as it acts as gate keeper residue located at the aperture of hydrophobic tunnel and helps aligning of bisallylic carbon of substrate below Tyr385 for catalysis30,31. In addition to this, the residues, Val89 and Ser119 play major roles in exit/entry of substrates to the COX32. Therefore, considering the above mentioned evidences obtained through molecular docking, it can be safely stated that the designed ligands has high potential act as MTDL against DPP-4, AChE and COX-2.

 

The designed compounds were also evaluated for pharmacokinetic properties using Swiss ADME. The drug like properties were evaluated by calculating Molecular weight, FCSp3, rotatable bonds, hydrogen bond acceptors, hydrogen bond donors, TPSA, MLogP, Log S, gastrointestinal absorption and blood brain permeation. The results showed that ibuprofen derived compounds has more drug like properties as compared to the indomethacin derived compounds. The indomethacin derived compounds showed MW higher than 500g/mol, low values of FCsp3, high values of TPSA, poor solublity and low gastrointestinal absorption. The molecular weight of ibuprofen derived was found to be less than 500g/mol, Fsp3 values not less than 0.25, number of rotatable bonds less than 9 except for 1D, HBA less than 10, HBD not more than 5, TPSA values in the range of 20 to 130 Å2, moderate solubility. It should be noted that majority of ibuprofen derived compounds also showed low gastrointestinal absorption except for 1A, 1F, 1G, 1H and 1O which showed high GI absorption and hence are more suitable candidates for being developed into drug molecule suitable for oral administration as compared to rest of the designed molecules. Both ibuprofen and indomethacin derived compounds showed moderate lipophilicity and no indication BBB permeation. Therefore, taken together i.e. calculated binding affinity, binding mode analysis and ADME profiling, it can be safely stated that all the designed molecules have considerable potential of inhibiting DPP-4, AChE and COX-2 enzyme, if evaluated under in vitro conditions. However, if a suitable formulation is developed to improve the BBB penetration capacity and to increase the gastrointestinal absorption, the designed compounds can behave as MTDL against targeted enzymes under in vivo condition. Further, 1A, 1F, 1G, 1H and 1O has the potential to be developed into molecules suitable for oral administration. More specifically 1H has substantial potential for being developed as MTDL against DPP-4, AChE and COX-2 as it has displayed good binding affinity supported by interactions with critical residues and favorable drug like attributes.

 

CONCLUSIONS:

Thirty-two new compounds were designed by hybridizing indomethacin and ibuprofen with sixteen sulfa drugs. Target fishing using pharmmapper for designed compounds identified three potential targets DPP-4, AChE and COX-2. The binding affinity of designed compounds against DPP-4, AChE and COX-2 was evaluated using molecular docking. The designed compounds showed equivalent to high binding affinity as compared to co-crystallized ligands against all three target proteins. To further support binding affinity, binding mode analysis was performed to identify interactions with key residues. The docked complexes showed interaction with Arg125 (residue important for substrate binding) and Glu205 and Glu206 (residues important for inhibitor binding as documented in protein-inhibitor crystal structure reports) in the case of DPP-4. Similarly, in case of AChE, the docked ligand showed polar as well as hydrophobic interactions with Tyr124 and His447 (important for hydrolysis of acetylcholine) and in case of COX-2 docked complexes, the designed compounds showed hydrogen bond interactions with Arg120 located at the aperture of hydrophobic channel used by substrate arachidonic acid to access peroxidase and cyclooxygenase site. Considering the molecular docking study all designed compounds showed good binding potential with three target enzymes in this study. The designed compounds when subjected to ADME profiling showed only five compounds i.e 1A, 1F, 1G, 1H and 1O with drug like character but with no BBB permeation capacity. 1H has also displayed superior binding affinity clubbed by interaction with residues important from catalytic perspective in all three enzymes as compared to other four molecules. Therefore, 1H is most suitable candidate to be taken forward as the MTDL for the DM related dementia.

 

FUNDING INFORMATION:

This research was funded by the Deanship of Scientific Research, Qassim University, Saudi Arabia, under the project number (pharmacy-2022-1-3-J-31826) during the academic year 1444 AH/2022 AD.

 

ACKNOWLEDGEMENT:

The authors gratefully acknowledge Qassim University, represented by the Deanship of Scientific Research, on the financial support for this research under the number (pharmacy-2022-1-3-J-31826) during the academic year 1444 AH/2022 AD

 

CONFLICTS OF INTEREST:

The authors declare no conflict of interest.

 

REFERENCES:

1.      Mani V, Arfeen M, Mohammed HA, et al. Sukkari dates seed improves type-2 diabetes mellitus-induced memory impairment by reducing blood glucose levels and enhancing brain cholinergic transmission: In vivo and molecular modeling studies. Saudi Pharm J. 2022; 30(6): 750-763. doi:https://doi.org/10.1016/j.jsps.2022.03.016

2.      Cholerton B, Baker LD, Montine TJ, Craft S. Type 2 Diabetes, Cognition, and Dementia in Older Adults: Toward a Precision Health Approach. Diabetes Spectr. 2016; 29(4): 210-219. doi:10.2337/ds16-0041

3.      Srivastava A, Mishra A, Rai AK. Synthesis, characterization and evaluation of ulcerogenic potential for NSAIDs-alendronate based prodrug. Res J Pharm Technol. 2020; 13(5): 2107-2111. doi:10.5958/0974-360X.2020.00379.0

4.      Srivastava A, Mishra A, Rai AK. Nsaids-alendronate based prodrug for bone specific drug targeting. Res J Pharm Technol. 2020; 13(7): 3520-3523. doi:10.5958/0974-360X.2020.00623.X

5.      Akgul O, Di Cesare Mannelli L, Vullo D, et al. Discovery of Novel Nonsteroidal Anti-Inflammatory Drugs and Carbonic Anhydrase Inhibitors Hybrids (NSAIDs–CAIs) for the Management of Rheumatoid Arthritis. J Med Chem. 2018;61(11):4961-4977. doi:10.1021/acs.jmedchem.8b00420

6.      Lucarini L, Durante M, Sgambellone S, et al. Effects of New NSAID-CAI Hybrid Compounds in Inflammation and Lung Fibrosis. Biomolecules. 2020; 10(9): 1307.

7.      Markowicz-Piasecka M, Sadkowska A, Sikora J, Broncel M, Huttunen KM. Novel sulfonamide-based analogs of metformin exert promising anti-coagulant effects without compromising glucose-lowering activity. Pharmaceuticals. 2020; 13(10): 1-28. doi:10.3390/ph13100323

8.      Sharma P, Dayma V, Dwivedi A, et al. Synthesis of sulpha drug based hydroxytriazene derivatives: Anti-diabetic, antioxidant, anti-inflammatory activity and their molecular docking studies. Bioorg Chem. 2020;96:103642. doi:https://doi.org/10.1016/j.bioorg.2020.103642

9.      Prafulla S, Lata P, Priya R, Vidya S. Novel Curcumin Derivatives: Targeted for Anti-Inflammatory Activity. Asian J Res Chem. 2019; 12(2): 49. doi:10.5958/0974-4150.2019.00011.7

10.    Vijaya Nagini D, Krishna MSR, Karthikeyan S. Identification of novel dipeptidyl peptidase-iv inhibitors from ferula asafoetida through gc-ms and molecular docking studies. Res J Pharm Technol. 2020; 13(11): 5072-5076. doi:10.5958/0974-360X.2020.00888.4

11.    Lakshmi Kanthamma S, Kriszhnamurthi J, Hemalatha CN, Vijey Aanandhi M. Insilico approach of interaction studies in Bacopa monnieri compounds targeting multi-proteins for Alzheimer’s disease. Res J Pharm Technol. 2018; 11(4): 1522-1526. doi:10.5958/0974-360X.2018.00283.4

12.    Sravani M, Sai Lakshmi P, Amuktha Reddy K, Naga Deepthi N, Kantheti USK. Insilico analysis and docking of tacrine and donepezil derivatives targeting histamine-N-methyltransferase and acetyl cholinesterase protein respectively for Alzheimer’s disease. Res J Pharm Technol. 2013; 6(1): 86-89.

13.    Trott O, Olson AJ. AutoDock Vina: improving the speed and accuracy of docking with a new scoring function, efficient optimization, and multithreading. J Comput Chem. 2010; 31(2): 455-461.

14.    Satpute UM, Rohane SH. Efficiency of AUTODOCK: Insilico study of Pharmaceutical Drug Molecules. Asian J Res Chem. 2021;14(1):1-5. doi:10.5958/0974-4150.2021.00016.x

15.    Morris GM, Huey R, Lindstrom W, et al. AutoDock4 and AutoDockTools4: Automated docking with selective receptor flexibility. J Comput Chem. 2009; 30(16): 2785-2791.

16.    Daina A, Michielin O, Zoete V. SwissADME: a free web tool to evaluate pharmacokinetics, drug-likeness and medicinal chemistry friendliness of small molecules. Sci Rep. 2017; 7(1): 42717. doi:10.1038/srep42717

17.    Nojima H, Kanou K, Terashi G, et al. Comprehensive analysis of the Co-structures of dipeptidyl peptidase IV and its inhibitor. BMC Struct Biol. 2016; 16(1): 1-14. doi:10.1186/s12900-016-0062-8

18.    Bajda M, Wiȩckowska A, Hebda M, Guzior N, Sotriffer CA, Malawska B. Structure-based search for new inhibitors of cholinesterases. Int J Mol Sci. 2013; 14(3): 5608-5632. doi:10.3390/ijms14035608

19.    Macdonald IR, Martin E, Rosenberry TL, Darvesh S. Probing the Peripheral Site of Human Butyrylcholinesterase. Biochemistry. 2012; 51(36): 7046-7053. doi:10.1021/bi300955k

20.    Kiefer JR, Pawiitz JL, Moreland KT, et al. Structural insights into the stereochemistry of the cyclooxygenase reaction. Nature. 2000; 405(6782): 97-101. doi:10.1038/35011103

21.    Tabatabaie T, Waldon AM, Jacob JM, Floyd RA, Kotake Y. COX-2 Inhibition Prevents Insulin-Dependent Diabetes in Low-Dose Streptozotocin-Treated Mice. Biochem Biophys Res Commun. 2000; 273(2): 699-704. doi:https://doi.org/10.1006/bbrc.2000.2959

22.    Roppongi S, Suzuki Y, Tateoka C, et al. Crystal structures of a bacterial dipeptidyl peptidase IV reveal a novel substrate recognition mechanism distinct from that of mammalian orthologues. Sci Rep. 2018; 8(1): 1-5. doi:10.1038/s41598-018-21056-y

23.    Rodhi AM, Yap PG, Abayomi OO, Gan CY. A review on the types of amino acid at ultimate, penultimate and antepenultimate position in some dipeptidyl-peptidase IV inhibitory peptides based on molecular docking analysis. Food Chem Adv. 2023; 2(October). doi:10.1016/j.focha.2023.100244

24.    Macalalad MAB, Gonzales AA. In Silico Screening and Identification of Antidiabetic Inhibitors Sourced from Phytochemicals of Philippine Plants against Four Protein Targets of Diabetes (PTP1B, DPP-4, SGLT-2, and FBPase). Molecules. 2023; 28(14): 1-33. doi:10.3390/molecules28145301

25.    Mathur V, Alam O, Siddiqui N, et al. Insight into Structure Activity Relationship of DPP-4 Inhibitors for Development of Antidiabetic Agents. Molecules. 2023; 28(15). doi:10.3390/molecules28155860

26.    Zhang Y, Kua J, McCammon JA. Role of the Catalytic Triad and Oxyanion Hole in Acetylcholinesterase Catalysis:  An ab initio QM/MM Study. J Am Chem Soc. 2002; 124(35): 10572-10577. doi:10.1021/ja020243m

27.    Mani V, Arfeen M, Rabbani SI, Shariq A, Amirthalingam P. Levetiracetam Ameliorates Doxorubicin-Induced Chemobrain by Enhancing Cholinergic Transmission and Reducing Neuroinflammation Using an Experimental Rat Model and Molecular Docking Study. Molecules. 2022; 27(21). doi:10.3390/molecules27217364

28.    Lu SH, Wu JW, Liu HL, et al. The discovery of potential acetylcholinesterase inhibitors: A combination of pharmacophore modeling, virtual screening, and molecular docking studies. J Biomed Sci. 2011; 18(1): 8. doi:10.1186/1423-0127-18-8

29.    Peitzika SC, Pontiki E. A Review on Recent Approaches on Molecular Docking Studies of Novel Compounds Targeting Acetylcholinesterase in Alzheimer Disease. Molecules. 2023; 28(3). doi:10.3390/molecules28031084

30.    Vecchio AJ, Orlando BJ, Nandagiri R, Malkowski MG. Investigating Substrate Promiscuity in Cyclooxygenase-2: The role of ARG-120 and residues lining the hydrophobic groove. J Biol Chem. 2012; 287(29): 24619-24630. doi:https://doi.org/10.1074/jbc.M112.372243

31.    Vecchio AJ, Orlando BJ, Nandagiri R, Malkowski MG. Investigating substrate promiscuity in cyclooxygenase-2 the role of Arg-120 and residues lining the hydrophobic groove. J Biol Chem. 2012; 287(29): 24619-24630. doi:10.1074/jbc.M112.372243

32.    Blobaum AL, Xu S, Rowlinson SW, et al. Action at a distance: Mutations of peripheral residues transform rapid reversible inhibitors to slow, tight binders of cyclooxygenase-2. J Biol Chem. 2015; 290(20): 12793-12803. doi:10.1074/jbc.M114.635987

 

 

 

 

 

Received on 09.09.2023            Modified on 19.10.2023

Accepted on 14.11.2023           © RJPT All right reserved

Research J. Pharm. and Tech 2024; 17(4):1611-1620.

DOI: 10.52711/0974-360X.2024.00255