RETRACTED ARTICLE: Thiophen urea derivatives as a new class of hepatitis C virus entry inhibitors

Abstract To develop unique small-molecule inhibitors of hepatitis C virus (HCV), thiophen urea (TU) derivatives were synthesised and screened for HCV entry inhibitory activities. Among them, seven TU compounds exhibited portent anti-viral activities against genotypes 1/2 (EC50 < 30 nM) and subsequently, they were further investigated; based on the pharmacological, metabolic, pharmacokinetic, and safety profiles, J2H-1701 was selected as the optimised lead compound as an HCV entry inhibitor. J2H-1701 possesses effective multi-genotypic antiviral activity. The docking results suggested the potential interaction of J2H-1701 with the HCV E2 glycoprotein. These results suggest that J2H-1701 can be a potential candidate drug for the development of HCV entry inhibitors.


Introduction
Approximately 200 million people worldwide are chronically infected with hepatitis C virus (HCV). This pathogen is the major cause of acute hepatitis and chronic liver disease, including cirrhosis and liver cancer 1 . HCV is an enveloped, positive-stranded RNA virus and a member of the Flaviviridae family in the hepacivirus genus. HCV is closely related to the flavivirus genus, which includes a number of viruses implicated in human disease, such as dengue virus and yellow fever virus 2 . Seven major HCV genotypes and numerous subtypes have been described and differ as much as 30% in nucleotide sequence 3,4 .
Despite increasing efforts to develop novel drugs effective against HCV, patients are mainly being treated with a virus-unspecific combination therapy of pegylated interferon alpha (PEG-IFN) and ribavirin 5,6 . This treatment is expensive and can be associated with severe side effects; in addition, it is effective in only 50-60% of patients infected with HCV genotype 1 7 . Since early 2011, two direct acting antivirals (DAAs) targeting the viral NS3 protease have been approved by the United States Food and Drug Administration. Unfortunately, both drugs induce severe side effects and have a low resistance barrier as well as an inconvenient administration regime 8 . Viral resistance may eventually become an issue; due to potential HCV genotype specificity, it is unclear whether all seven HCV genotypes and their subtypes are covered 9 . Therefore, it is a primary goal to identify targets with a significantly higher genetic barrier of resistance that cover all HCV genotypes. Other challenges include appearance of escape mutants, high costs of current therapy regimens, and negative side effects. The mechanisms of resistance to interferon (IFN)-based therapy through immune system interception 10 and also to DAAs like telaprevir and boceprevir through resistance-conferring mutations could be avoided by combinatorial treatment. Although drugs that target the virus are currently in clinical trials, a medical need remains for new HCV therapeutic agents. There is a particular need for HCV therapeutic agents that have broad activity against the majority of HCV genotypes and their subtypes (e.g. The entry process of HCV into hepatocytes has been recognised as one of the potential targets for therapeutic intervention to treat or prevent HCV infection. To develop new types of smallmolecule inhibitors for HCV, we first sought to screen HCV entry inhibitory molecules using an infectious cell culture system (HCVcc). During the preliminary screening 11 , thiophen urea (TU) derivatives were found to possess useful activity against all major HCV genotypes. Accordingly, we further tested 296 TU derivatives to determine the lead optimisation process. Herein, we present a new class of compounds that exert antiviral activity against HCV.

General procedure for synthesis of TU derivatives
The target products were prepared in three steps as follows (Scheme 1). anhydrous CH 2 Cl 2 at 0 C. The reaction mixture was warmed to room temperature and stirred for 2 h. After the reaction was completed, the reaction mixture was quenched with H 2 O and extracted using CH 2 Cl 2 . The organic layer was dried over anhydrous MgSO 4 , filtered, and concentrated. The crude product was precipitated by adding diethyl ether and kept at 4 C for few hours to complete the precipitation. The precipitate was filtered and washed with diethyl ether and dried in vacuum to produce a 4-nitrophenoxycarbonylamino derivative (2) as a yellow solid.

R E T R
A C T E D 0 C. The reaction mixture was stirred at room temperature for 5 h. After the reaction was complete, the mixture was diluted with CH 2 Cl 2 and washed with brine (br). Next, 2N NaOH (aq.) was slowly added to the separated organic solution at 10 C, and then the suspension was stirred for 1.5 h. The precipitate was filtered and agitated with methyl tertbutyl ether for 2 h. The solid was filtered and washed with diethyl ether and dried under vacuum to obtain the product (5) as a white powder. The final product (6) could be obtained as an organic or inorganic salt by further process to get more stable solid state.

Molecular docking analysis
The HCV (genotype: 1a strain H77) E2 X-ray diffraction structure was obtained from the Protein Data Bank (PDB: 4MWF) 12 . Of the two enveloped glycoprotein E2 chains that are involved, chain D was selected for docking as its secondary structure was welldefined compared to chain C. The residues of T542RPPLG547 were not missing from this chain. The selected D chain of glycoprotein E2 was prepared for docking using the Protein Preparation tool. Molecular docking was carried out with Glide 13 . All possible binding pockets were analysed according to the binding cavities of the protein as well as the biologically meaningful regions. Four grids were generated for each assigned binding site, and the prepared three-dimensional ligand conformation of J2H-1701 was

R E T R A C T E D
docked at each site. The best docked pose was determined based on the highest docking score and also any favourable interactions with the targeted residues. The selected docked complex was energy minimised with MacroModel using OPLS_2005 force field in a water environment (until converging at a termination gradient of 0.0001 kJ/mol Å) referring to the report by Di Maio et al. 14 .

Results and discussion
Thiophen urea derivatives were synthesised in three steps as follows: a reaction of 4-nitrophenyl chloroformate, urea synthesis by amine substitution, and final deprotection (Scheme 1). In each step all products were easily obtained as a pure solid by a simple work-up method (precipitation or filtration) without further purification. Two starting materials that have a thiophen ring with various substituents were utilised to achieve structural diversity. Several compounds including J2H-1701, HCV-2602, and HCV-2694 were prepared as an acid salt to enhance aqueous solubility using an additional method. For example, J2H-1701 was prepared as a HCl salt by treatment of HCl solution in 1,4-dioxane. Almost all the test compounds showed good chemical stability and crystallinity. J2H-1701, a representative compound as a HCl salt, has a long-term stability (purity >98% for 3 months) at room temperature, and so do the other compounds. This stability is supposed to come from its crystallinity that would be intrinsic property of thiophen urea structure, based on their high melting point and differential scanning calorimetry data (data not shown). When the drug-like properties of TU derivatives were assessed by Lipinski's rule of five 15 , most of the compounds were shown to be applicable for orally available drugs; for example, J2H-1701, it has 3 hydrogen bond donors, 6 hydrogen bond acceptors, logP (experimental) of 4.21, and molecular weight of 515.62 and the number of violation is 1. Therefore, J2H-1701 is considered as an orally available drug. Initially, we screened 296 TU derivatives for anti-viral activity against genotypes 1/2 using the infectious HCVcc system and seven selected compounds with potent activities for further in vitro ADME/toxicity studies ( Figure 1, Table 1). Accordingly, in vitro ADME and cytotoxicity evaluations were conducted for the seven compounds to select the best potential drug candidate ( Table 2). The compounds showed good metabolic stability in mouse and human liver microsomes, and they also were found to be stable in mouse and human plasma. In addition, the compounds did not reveal any negligible effects for CYP3A4 and CYP2A9 inhibition. However, some compounds showed significant cytotoxicity with human cell lines, including HEK293 (embryonic kidney cells), SK-N-SH (neuroblastomas), HepG2 (hepatomas), and Jurkat (immortalised T lymphocytes) (data not shown). Accordingly, based on these in vitro test results, J2H-1701 was selected for further characterisation as an anti-HCV drug candidate.
J2H-1701 was tested for cross-genotype activities against seven genosubtypes (nine subtypes) of the virus using a dose-response assay in the HCVcc system. The representative dose-response plots for HCV inhibition and cytotoxicity are provided in Figure 2. J2H-1701 generally exhibited potent inhibition against the six HCV genotypes (HCV genotype 1a, 2a, 2 b, 3a, 4a, and 7a) whereas no significant cytotoxicity was observed for all tested genosubtypes. J2H-1701 showed IC 50 values in the nano-subnanomolar range in several HCV subgenotypes, which is much lower than those of a well-known HCV inhibitor, sofosbuvir (Table 3). This demonstrates that J2H-1701 possesses effective multi-genotypic antiviral activity.

R E T R A C T E D
Since J2H-1701 was shown to be an HCV entry inhibitor, we investigated the relevant mechanisms for HCV entry inhibition and found that J2H-1701 inhibits HCV envelope proteins (data not shown). Accordingly, the HCV envelope glycoproteins were selected as a target protein for docking studies and we performed a molecular docking analysis on the crystal structures of the HCV E2 glycoprotein 12 to predict whether J2H-1701 would interact with E1/E2 viral glycoproteins. Out of the four potential binding sites of the E2 monomer, J2H-1701 fit snugly into the binding site (Figure 3(A)). This site is located in the loop region and is also a surface on the b sandwich of the E2 glycoprotein (Figure 3(B)), which has been identified by the mutagenesis study as being involved in interacting with CD81 12 . The interactions of J2H-1701 with the residues of the E2 monomer indicated strong binding of J2H-1701 within the E2 monomer. There were four hydrogen (H) bonds: 1) two H bonds formed with both nitrogens in the urea group and Val515; 2) one bond was made between the oxygen of the urea group and His421; and 3) another H bond formed with the nitrogen in piperidine and Thr425. The trifluoro group was docked around the threonines and valines in the b5 region of the b sandwich. The van der Waals interactions of the trimethyl group were observed with Thr510, Asn548, and Trp549 of the b sandwich (Figure 3(C)). The docking energy was À4.437 kcal/mol, which was the lowest among all docked poses. The docking score (GlideScore) was À4.437 kcal/mol, which was the lowest among all docked poses. The minimisation energy difference between the unbound protein structure (À28224.178 kJ/mol) and the J2H-1701 docked structure (À28598.234) was À374.056 kJ/mol.
The docked site of J2H-1701 has been reported to participate in CD81 receptor binding and also viral entry 12 . Recently, docking studies with berberine showed that it may potentially bind to the site on the E2 glycoprotein 16,17 . This site is located in the back layer of E2, which is about 11 Å away from the front layeran area reported to participate in CD81 receptor binding and viral entry. J2H-1701 demonstrated weaker docking activity to the back layer of the E2 glycoprotein (data not shown). Overall, the binding pose of J2H-1701, which favourably interacts with those residues known to participate in CD81 receptor binding and viral entry, indicates that J2H-1701 most likely interacts with the HCV envelope glycoprotein E2.
The essential players in HCV virus entry are HCV glycoproteins E1 and E2, which form a heterodimer and constitute viral envelope proteins [18][19][20] . HCV E1 does not directly interact with host receptors, whereas E2 directly binds with host receptors, such as the CD81 receptor 21 . Therefore, we propose that J2H-1701 would

Conclusions
In this study, TU derivatives were synthesised and tested for their potential as HCV entry inhibitors. The seven selected TU compounds exhibited potent anti-viral activities against genotypes 1/2 with EC 50 values ranging from 0.04 to 27 nM. The compounds also showed good pharmacokinetic properties. However, some compounds significant cytotoxicity; accordingly, J2H-1701 was selected as a potential anti-HCV drug candidate. The J2H-1701 possesses effective multi-genotypic antiviral activity. The docking analysis results suggest that J2H-1701 would bind to E2 to inhibit the entry of HCV. These results demonstrate that J2H-1701 can be a potential candidate drug for the development of HCV entry inhibitors.