A Molecular Docking of New 9β-Halogenated Prostaglandin Analogues ^†

Abstract

Prostaglandins (PGs) with cytoprotective activity were studied for a long time, and a few PGE₁ and PGE₂ stable analogues were promoted as drugs: arbaprostil, enprostil, misoprostol, and rioptostol. Similarly, nocloprost, a 9β-chlorine prostaglandin analogue, and many 9β- and 11β-substituted prostaglandins were synthesized and studied for their biological activity. We previously synthesized new 9β-halogenated prostaglandins with an ester group at the carbon atom 6 (PGs numbering) by the reaction of a δ-lactone intermediate with diols in acid catalysis. These compounds were used in the current molecular docking study to determine their potential cytoprotective (anti-ulcer) activity. The current study was done with the CLC Drug Discovery Workbench 2.4. software and an oxidoreductase enzyme receptor, chosen from the Protein Data Bank, ID: 4KEW (www.rcsb.org). We used two recognized drugs, omeprazole (co-crystallized with the enzyme) and nocloprost, as the standard. The 9β-halogenated prostaglandin analogs were docked. Nocloprost and all 9β-halogenated compounds had docking scores greater than that of omeprazole. The majority of the 9β-halogenated analogs had docking scores even greater than that of nocloprost, indicating that these compounds could have potential cytoprotective (anti-ulcer) activity. A few correlations between docking score and substituents on the prostaglandin skeleton were found.

Prostaglandins (PGs) with cytoprotective activity were studied for a long time, and a few PGE₁ and PGE₂ stable analogues were promoted as drugs: arbaprostil, enprostil, misoprostol, and rioptostol. Similarly, nocloprost, a 9β-chlorine prostaglandin analogue, and many 9β- and 11β-substituted prostaglandins were synthesized and studied for their biological activity. In the same direction, we have synthesized 9β-halogenated prostaglandins of types 2 and 3, starting from the δ-lactone compound 1 with diols (n = 0 to 4) or 2-butyn-1,4-diol, catalyzed by toluenesulfonic acid [1,2] (Scheme 1).

The new compounds have not only a 9β-halogen (Cl, Br, and F in the Scheme 1), but also an ester group at the carbon atom 6 (PGs numbering). Other compounds with the same ester group have been presented in the literature and showed cytoprotective activity [3]; the new compounds synthesized as in Scheme 1 have not been tested for cytoprotective activity.

We completed a molecular docking study, using CLC Drug Discovery Workbench 2.4. software (QIAGEN Bioinformatics, Aarhus, Denmark) and an oxidoreductase enzyme receptor, chosen from the Protein Data Bank, ID: 4KEW (www.rcsb.org), to determine the predicted cytoprotective activity of the new compounds. In the study, we use as a standard two recognized drugs, omeprazole (co-crystallized with the enzyme) and nocloprost. The co-crystallized omeprazole with a protein receptor was used for docking, defining the binding site and binding pockets, validating docking, and identifying hydrogen bonds between co-crystallized omeprazole and amino acid residues of the receptor. Then, the prostaglandin analog drug nocloprost, also used as a standard in this study, was docked (no docking studies about nocloprost were found in the literature) and the results are presented in

Table 1. Docking scores and the molecular properties of the ligands omeprazole, nocloprost, and 9β-halogenated prostaglandin analogs 2a–2l and 3a–3d, calculated with CLC Drug Discovery Workbench 2.4 software.

Ligand	Score	RMSD *	Atoms No.	Weight	Flexible Bonds	Lipinski Violation	HD **	HA ***	Log P
Omeprazole (co-crystallized)	−58.11	0.06	41	328.41	5	0	0	5	3.28
Nocloprost	−71.25	1.32	64	400.98	12	1	3	4	5.38
2a	−66.93	1.02	49	417.28	10	0	2	6	2.35
2b	−73.00	0.83	52	431.31	11	0	2	6	2.71
2c	−70.67	1.18	52	475.76	11	0	2	6	2.88
2d	−74.92	0.78	52	414.85	11	0	2	6	2.46
2e	−74.67	0.83	55	445,33	12	0	2	6	3.06
2f	−74.92	0.43	58	478.89	13	0	2	6	3.32
2g	−67.73	1.67	55	489.78	12	0	2	6	3.23
2h	−73.63	0.92	55	428.88	12	0	2	6	2.82
2i	−72.29	1.60	58	459.36	13	0	2	6	3.42
2j	−71.13	1.36	61	473.39	14	0	2	8	3.78
2k	−75.03	0.76	57	447.35	12	0	3	6	2.72
2l	−77.57	1.25	60	480.90	13	0	3	6	2.97
3a	−75.09	1.42	51	441.30	11	0	2	6	2.41
3b	−82.27	0.63	54	474.85	12	0	2	6	2.66
3c	−77.47	0.41	53	443.32	11	0	3	6	2.06
3d	−84.91	1.28	56	476.87	12	0	3	6	2.32

* RMSD: root-mean-square deviation; RMSD should be <2, **HD = hydrogen donors, ***HA = hydrogen acceptors.

(in red) docking poses of the interactions between nocloprost and the amino acid residues are presented in Figure 1.

The 9β-halogenated prostaglandin analogs 2a–2l and 3a–3d, presented in Scheme 1, were docked, and the results of the calculated properties (flexible bonds, Lipinski violations, the number of hydrogen bond donors, the number of hydrogen bond acceptors, and log P) [4] are presented in Table 1 and Figure 2. The calculated parameters can predict if a molecule possesses properties that might turn it into an active drug, according to the Lipinski’s rule of five [4].

An expressive presentation of the docking score is presented in Figure 2.

According to the data presented in Table 1, all 9β-halogenated compounds comply with the Lipinski rules (Lipinski violation is 0), and the nocloprost drug has one violation. In Table 1, the docking score and RMSD, which is <2, are also presented. All 9β-halogenated analogs and nocloprost have docking scores greater than that of omeprazole (−58.11, RMSD 0.06). The majority of the 9β-halogenated analogs have a docking score greater than that of the 9β-chlorine nocloprost prostaglandin-recognized drug (−71.25, RMSD 1.32), used as a standard in the study, with the exception of the compounds 2a (−66.93, RMSD 1.01), 2c (−70.67, RMSD 0.83), 2g (−67.73, RMSD 1.67), and 2j (−71.13, RMSD 1.36) (Figure 2). Based on the docking scores, this study shows that the predicted cytoprotective (anti-ulcer) activities of the compounds 2b, 2d–2f, 2h–2i, 2k–2l, 3a–3d are greater than that of nocloprost. The compound 3d had the best score (−84.91), followed by compound 3b (−82.27), both those compounds have a 2-butyn synthon in α-side chain; for compounds with a normal α-side chain, the best score is for the compound 2l (−77.57).

In addition to the parameters mentioned in Table 1, group interaction and hydrogen bonds of the ligands with amino acid residues were determined and hydrogen bond length was calculated.

A few correlations between docking score and substituents on the prostaglandin skeleton were also found.

The docking poses of the ligands with the best scores interacting with the amino acid residues of the protein, with the 2-butyn scaffold of compound 3d and with a linear chain in the α-side chain of 2l, are presented in Figure 3b and Figure 4b: