MK-571, a Cysteinyl Leukotriene Receptor 1 Antagonist, Inhibits Hepatitis C Virus Replication

The quinoline MK-571 is the most commonly used inhibitor of multidrug resistance protein-1 (MRP-1) but was originally developed as a cysteinyl leukotriene receptor 1 (CysLTR1) antagonist. While studying the modulatory effect of MRP-1 on anti-hepatitis C virus (HCV) direct-acting antiviral (DAA) efficiency, we observed an unexpected anti-HCV effect of compound MK-571 alone. This anti-HCV activity was characterized in Huh7.5 cells stably harboring a subgenomic genotype 1b replicon. A dose-dependent decrease of HCV RNA levels was observed upon MK-571 administration, with a 50% effective concentration (EC50 ± standard deviation) of 9 ± 0.

Despite the spectacular virological results of current anti-HCV therapies, several issues remain. In patients who fail to achieve a cure of the infection, HCV variants carrying resistance-associated substitutions (RASs) on their genome, i.e., substitutions that confer reduced susceptibility to the administered drugs, are generally selected (6). Their long-term persistence after treatment raises issues as to subsequent retreatment. Although the global rate of treatment failure is low with current DAA combinations, the absolute number of patients requiring retreatment is high. This number will further increase due to the large number of patients who will be treated, in the context of the World Health Organization endeavor to eliminate HCV as a major public health threat by 2030 (3). Importantly, some regions of the world (e.g., central Africa and Southeast Asia) harbor unusual subtypes of known genotypes that are inherently resistant to commonly administered DAAs (7,8). In addition, the high cost of last-generation DAA regimens limits access to care in low-income areas, while the management of special patient groups, such as those with advanced liver disease or renal failure, may be problematic with current drugs.
Multidrug resistance (MDR), i.e., cell ability to acquire drug resistance, is mainly mediated by the overexpression of membrane drug transporters, such as P᎑glycoprotein (P᎑gp), breast cancer resistance protein (BCRP), or multidrug resistance protein-1 (MRP-1), which belong to the ATP-binding cassette (ABC) transporter superfamily (9,10). These transporters influence drug pharmacokinetics, particularly their distribution, thereby modifying their concentrations in cells (11). Drug-drug interactions may occur at the transporter level and modulate drug efficacy and/or toxicity (12). Functional interactions between anti-HCV DAAs and ABC transporters have been reported (4,13). Indeed, almost all of the approved HCV inhibitors, including sofosbuvir, daclatasvir, ledipasvir, velpatasvir, voxilaprevir, paritaprevir, dasabuvir, glecaprevir, and pibrentasvir, are substrates and/or inhibitors of at least one ABC transporter (4,14).
To investigate the involvement of ABC transporters in the efflux of HCV protease inhibitors, we had tested the in vitro anti-HCV activity of the NS3-4A protease inhibitor telaprevir, alone or in combination with LY335979 (15), KO143 (16), or MK-571 (17,18), inhibitors of P-gp, BCRP, and MRP-1, respectively. In the control experiments, we observed an unexpected antiviral effect of MK-571 alone, a result that prompted us to characterize the anti-HCV activity of this compound and identify its target.
In the present study, based on our serendipitous discovery of the anti-HCV activity of MK-571, we characterized the antiviral activity of this compound in cellular HCV models and identified its target.

RESULTS
MK-571 inhibits HCV replication. With the goal to assess the effect of inhibiting ABC transporter (e.g., MRP-1) efflux on the antiviral activity of the HCV protease inhibitor telaprevir, we measured HCV RNA levels in Huh7.5 cells stably harboring a genotype 1b HCV-SGR treated with 1 M of telaprevir in combination with 50 M of MK-571. In our control experiments with MK-571 in the absence of telaprevir, a 12-fold decrease of HCV RNA levels was observed, suggesting that MK-571 bears anti-HCV activity. The antiviral effect of 50 M of MK-571 was equivalent to the effect of 1 M of telaprevir (Fig. 1). In the presence of telaprevir, MK-571 increased by approximately 3-fold the anti-HCV effect of telaprevir ( Fig. 1).
We next examined HCV-SGR clearance during 2 weeks of treatment with MK-571. Huh7.5 cells harboring an HCV-SGR containing the neomycin resistance gene were grown in the presence of increasing concentrations of MK-571 under G418 selective pressure; under such pressure, only cells efficiently replicating the HCV-SGR were able to grow. As shown in Fig. 2C, cell growth was not significantly affected by 100 M of MK-571 in the absence of G418. In contrast, in the presence of G418, MK-571 inhibited cell growth in a dose-dependent manner, as a result of its inhibition of HCV-SGR replication (Fig. 2C). These results indicate that MK-571 blocks an intracellular step of the HCV life cycle.
The anti-HCV activity of MK-571 was then confirmed in a chimeric genotype 2a/2a HCV full-length infectious model (HCV-J6/JFH1) containing a luciferase reporter gene. Huh7.5 cells were infected with HCV-J6/JFH1 in the presence of an increasing concentration of MK-571, and luciferase activity was quantified 48 h after infection. As with HCV-SGR, MK-571 induced a dose-dependent decrease of HCV-J6/JFH1 RNA replication. However, MK-571 anti-HCV activity appeared higher in the genotype 1b HCV-SGR than the genotype 2a HCV full-length infectious model (Fig. 2D).
MK-571 anti-HCV activity is related to its CysLTR1 antagonist property rather than MRP-1 inhibition. Both MRP-1 and CysLTR1 are known targets of MK-571. To identify which of them is associated with MK-571 anti-HCV activity, we first measured the anti-HCV effect of two other MRP-1 inhibitors, including probenecid (1 mM) (28) and apigenin homodimer (APN; 0.1 M) (29). As shown in Fig. 3A, none of these two compounds reduced HCV-SGR RNA levels. Furthermore, the addition of 1 mM of probenecid to MK-571 did not modify its anti-HCV activity (data not shown). Together, these results suggest that MRP-1 is not the MK-571 target associated with anti-HCV activity.
Next, the antiviral effect of CysLTR1 natural agonist LTD4 on HCV-SGR was assessed. LTD4 alone did not exert any effect on HCV replication (Fig. 3B). However, increasing We examined whether leukotriene biosynthesis is necessary for the HCV life cycle. The antiviral effect of two potent inhibitors of the 5-lypoxygenase (5-LO), MK-886 and zileuton, was assessed. None of them affected HCV-SGR replication ( Fig. 3D and E), suggesting that leukotriene biosynthesis is not required for HCV replication.
Together, these results indicate that the anti-HCV activity of MK-571 is related to CysLTR1 and that it can be reverted by the addition of the CysLTR1 natural ligand.
Anti-HCV activity of CysLTR1 antagonists zafirlukast, cinalukast, and SR2640 and their interaction with MK-571 antiviral activity. The effect of increasing concentrations of CysLTR1 antagonists zafirlukast, cinalukast, and SR2640 was tested in the absence or in the presence of 50 M MK-571 in Huh7.5 HCV-SGR cells. As shown in Fig. 4A and B, respectively, zafirlukast and cinalukast alone had a modest proviral effect on HCV RNA replication. SR2640 alone increased HCV-SGR RNA levels in a dose-  dependent manner, with a maximum increase of 11-fold achieved at a concentration of 20 M (Fig. 4C). Both zafirlukast and cinalukast reversed the anti-HCV activity of MK-571 in a dosedependent manner, with full reversion achieved at the highest dose tested (Fig. 4D, G,  E, and H). The addition of SR2640 to MK-571 not only reversed its anti-HCV activity but also increased HCV replication at the highest concentrations tested (Fig. 4F and I). Together, these results suggest that CysLTR1 regulates HCV replication, in a positive or negative way depending on the CysLTR1 antagonist used.

DISCUSSION
The approval of highly efficient, well-tolerated DAAs has profoundly changed the HCV treatment landscape. HCV infection can now be cured in most DAA-treated patients within 8 to 12 weeks (4). Despite this unprecedented therapeutic revolution, HCV remains an interesting model to better understand RNA virus life cycles and their interactions with infected hosts.
While studying the modulatory effect of ABC transporter inhibition on the anti-HCV activity of first-generation HCV protease inhibitors, we fortuitously observed an inhibitory effect of MK-571 on HCV replication in a hepatoma cell line harboring an HCV replicon. We further characterized the anti-HCV activity of MK-571 in Huh7.5 cells stably harboring a genotype 1b subgenomic replicon and observed a dose-dependent effect with an EC 50 of 9.0 Ϯ 0.3 M and a maximum HCV RNA level reduction of approximatively 11-fold at a concentration of 50 M of MK-571. Extended treatment with 100 M of MK-571 cured almost all Huh7.5-SGR-containing cells from their replicons. These results suggest that MK-571 inhibits HCV RNA replication. Although the MK-571 antiviral effect was relatively modest, it points out a new, thus far unknown mechanism involving an interaction between HCV and a host cellular pathway that influences its replication.
The lack of anti-HCV activity of APN and probenecid, two specific MRP-1 inhibitors, and their lack of effect on MK-571 antiviral activity suggested that another target of MK-571 is involved in the interaction with the HCV life cycle. CysLTR1 has been described as another binding target for MK-571 (18). CysLTR1 belongs to the G protein-coupled receptor (GPCR) family. Its activation is linked to the metabolism of phosphatidylinositol and intracellular calcium mobilization (26). CysLTR1 has been shown to activate mitogen-activated protein (MAP) kinases to induce cellular differentiation and proliferation, chemotaxis, and actin reorganization to release mediators of inflammation and to regulate hematopoietic stem cells (25). CysLTR1 has been implicated in a number of inflammatory diseases, including asthma and allergic rhinitis (21,23). Several CysLTR1-specific antagonists, including montelukast, cinalukast, and zafirlukast, have been developed and are used in clinical practice for the treatment of these diseases (30). Recent studies suggested that CysLTR1 is also involved in various types of liver diseases (31). The human CysLTR1 gene encodes a 337-amino acid protein with a calculated molecular mass of 38 kDa, reported to migrate at a molecular weight of approximately 44 kDa in its monomeric form, which contains a nuclear localization signal at its C-terminal end (25).
Our results demonstrate that the antiviral activity of MK-571 is related to the CysLTR1. Indeed, LTD4, the natural agonist of CysLTR1, reversed the anti-HCV effect of MK-571 in a dose-dependent manner, suggesting that MK-571 displacement from CysLTR1 is sufficient to lose anti-HCV activity. Unlike MK-571, LTD4 and 5-LO inhibitors alone had no effect on HCV replication, suggesting that the presence of CysLTR1 rather than its ligand-induced activation is exploited by HCV for its replication. Interestingly, none of the other CysLTR1 antagonists tested, including zafirlukast, cinalukast, and SR2640, reduced HCV RNA replication; however, all of them reversed the anti-HCV activity of MK-571. SR2640 induced HCV replication in a dose-dependent manner, suggesting that CysLTR1 has an effect on the HCV life cycle that is not univocal, according to the compound and its binding mode. This differential effect could be explained by the fact that receptor binders may act as partial agonists, neutral agonists, or inverse agonists (32,33), thereby inducing different biological responses.
In summary, we serendipitously identified MK-571 as an inhibitor of HCV replication in hepatoma cell lines harboring an HCV replicon. MK-571 anti-HCV activity was not related to its well-known effect as an MRP-1 inhibitor but rather to its CysLTR1 antagonist property. Our results highlight, for the first time, the implication of the CysLTR1 in the HCV life cycle. Overall, our study demonstrates that CysLTR1 is involved in HCV replication, with an impact on both the HCV life cycle and infected cell biology.
Assessment of anti-HCV activity in the subgenomic replicon model. Huh7.5 cells stably harboring an HCV genotype 1b bicistronic replicon (I389-Neo/NS3-3=/5.1) were seeded at the density of 5,000 cells/well in a 96-well plate. The cells were treated for 48 h with the tested compounds diluted in complete DMEM containing 1% dimethyl sulfoxide (DMSO) with 0.5 mg/ml of G418. Total RNA was extracted using the RNeasy 96 kit (Qiagen, Hilden, Germany). HCV-SGR RNA levels were measured by reverse transcriptase quantitative PCR (RT-qPCR) using the TaqMan technology with HCV-specific primers (forward primer, 5=-CGC CCA AAC CAG AAT ACG A-3=; and reverse primer, 5=-AGA TAG TAC ACC CTT TTG CCA GAT G-3=) and a qPCR probe (5=-6-carboxyfluorescein [FAM]-CAA TGT GTC AGT CGG-6carboxytetramethylrhodamine [TAMRA]-0.3=). The data were analyzed with the 2 ϪΔΔCT method, with all samples normalized to glyceraldehyde-3-phosphate dehydrogenase (GAPDH) mRNA. All experiments were performed in triplicate. HCV-SGR RNA relative quantities were plotted against compound concentrations and fitted with a four-parameter logistic curve with SigmaPlot v11 software. EC 50 s were determined from the curves.