You are here:  > Browse Articles

Original article

Antivirally active ribavirin analogues – 4,5-disubstituted 1,2,3-triazole nucleosides: biological evaluation against certain respiratory viruses and computational modelling

Anna Krajczyk1, Katarzyna Kulinska1, Tadeusz Kulinski1, Brett L Hurst2, Craig W Day2, Donald F Smee2, Tomasz Ostrowski1, Piotr Januszczyk1, Joanna Zeidler1,*

1Institute of Bioorganic Chemistry, Polish Academy of Sciences, Poznan, Poland
2Institute for Antiviral Research, Department of Animal, Dairy and Veterinary Sciences, Utah State University, Logan, UT, USA

*Corresponding author e-mail: joandler@ibch.poznan.pl

Citation: Antiviral Chemistry & Chemotherapy 2014; 23:161-171
doi: 10.3851/IMP2564

Date accepted: 12 March 2013
Date published online: 28 March 2013

Copyright (c) 2014 International Medical Press, all rights reserved.

Abstract

Background: Ribavirin is a broad-spectrum antiviral agent that derives some of its activity from inhibition of cellular inosine monophosphate dehydrogenase (IMPDH), resulting in lower guanosine triphosphate (GTP) levels. Here we report the biological activities of three ribavirin analogues.

Methods: Antiviral activities of test compounds were performed by in vitro cytopathic effect inhibition assays against influenza A (H1N1, H3N2 and H5N1), influenza B, measles, parainfluenza type 3 (PIV-3) and respiratory syncytial viruses. Compounds were modelled into the ribavirin 5′-monophosphate binding site of the crystallographic structure of the human type II IMPDH (hIMPDH2) ternary complex. Effects of compounds on intracellular GTP levels were performed by strong anion exchange HPLC analysis.

Results: Of the three compounds evaluated, the 5-ethynyl nucleoside (ETCAR) exhibited virus-inhibitory activities (at 1.2–20 μM, depending upon the virus) against most of the viruses, except for weak activity against PIV-3 (62 μM). Antiviral activity of ETCAR was similar to ribavirin; however, cytotoxicity of ETCAR was greater than ribavirin. Replacing the 5-ethynyl group with a 5-propynyl or bromo substituent (BrCAR) considerably reduced antiviral activity. Computational studies of ternary complexes of hIMPDH2 enzyme with 5′-monophosphates of the compounds helped rationalize the observed differences in biological activity. All compounds suppressed GTP levels in cells; additionally, BrCAR suppressed adenosine triphosphate and elevated uridine triphosphate levels.

Conclusions: Three compounds related to ribavirin inhibited IMPDH and had weak to moderate antiviral activity. Cytotoxicity adversely affected the antiviral selectivity of ETCAR. As with ribavirin, reduction in intracellular GTP may play a role in virus inhibition.

Introduction

For many years the antiviral agent ribavirin (1-β-D-ribofuranosyl-1,2,4-triazole-3-carboxamide, 1; Figure 1) [1,2], has been used clinically to treat respiratory syncytial virus (RSV) infections in children [3]. More recently, the combination of ribavirin and interferon-α has become a mainstay in the treatment of HCV infections in humans [4]. The mode of action of 1 against different pathogenic RNA viruses has been a complex issue, since it may involve several mechanisms that contribute to the overall effect, depending on virus species [5,6]. Inosine monophosphate dehydrogenase (IMPDH) inhibition by ribavirin 5′-monophosphate (RMP; 2) was the first proposed mechanism [7] and proved essential for antiviral activity against flaviviruses [8], paramyxoviruses [8] and orthopoxviruses [9]. Being the substrate analogue, RMP competitively inhibits IMPDH, resulting in low levels of intracellular guanosine triphosphate (GTP) that no longer sustain viral replication. The major metabolite of 1, ribavirin 5′-triphosphate may inhibit GTP-dependent RNA capping enzymes causing interference in the viral messenger RNA maturation process [10]. Moreover, ribavirin 5′-triphosphate interacts with viral RNA-dependent RNA polymerases of different viral species [6,11]. After incorporation into viral RNA, ribavirin can either terminate the synthesis or induce mutations. Accumulation of mutations in viral genomic RNA may result in error catastrophe for a virus. This distinctive mechanism was postulated to explain antiviral activity of ribavirin against poliovirus [12]. To date, mutagenic activity of the drug has been found to be responsible for error-prone replication of other viruses such as GB virus B [13], HCV [14], Hantaan virus [15] and West Nile virus [16]. Recently, studies to elucidate cell-based resistance after continuous treatment with this antiviral agent have shown that, depending on its intracellular metabolism, ribavirin inhibits replication of the same virus via different mechanisms in different cell types [17].

-
Figure 1.
Figure 1. Chemical structures of triazole compounds related to ribavirin and ribavirin 5′-monophosphate

Along with its application as a drug, ribavirin has become a lead compound for many other nucleoside analogues. One of them, 5-ethynyl-1-β-D-ribofuranosylimidazole-4-carboxamide (EICAR; 3), demonstrates broad-spectrum antiviral activity against DNA and RNA viruses comparable to that of ribavirin, although the former compound exceeds the latter in potency by 10- to 100-fold [18]. When metabolized to 5′-monophosphate (EICARMP; 4), it acts primarily as a mechanism-based inhibitor of IMPDH [19], covalently binding to a cysteine residue of the enzyme active site [20]. It may be further phosphorylated to 5′-O-triphosphate. Due to its toxicity, EICAR has no longer been investigated as a drug candidate for treatment of viral infections in humans. Nevertheless, EICAR has shown efficacy against animal viruses such as infectious pancreatic necrosis virus (the pathogen of many marine species) [21] and canine distemper virus [22]. Recently, newly synthesized 3-ethynyl-1-β-D-ribofuranosyl-1,2,4-triazole (ETAR; 5), which lacks pseudo-base pair capacity, has shown a higher level of activity against hantaviruses in vitro and in vivo [23], and flaviviruses (dengue, Langat and Modoc viruses) in vitro [24], when compared to that of ribavirin. Also, studies of another binding factor showed that appending a π-conjugated system onto the 1,2,4-triazole ring (compound 6), improved the antiviral activity of 6 against HCV [25]. Additionally, with a hydrophobic substituent attached at the C-4 position, new pyrazole analogues of 1 (79) have been targeted to the poliovirus RNA-dependent RNA polymerase [26]. In this report we describe the synthesis of a novel 4,5-disubstituted 1,2,3-triazole nucleoside and compare it with related compounds for antiviral activity against several respiratory RNA viruses in vitro, perform computational modelling with IMPDH, and demonstrate effects on GTP levels in treated cells.

Materials and methods

Chemistry

General methods and techniques were applied to the synthesis of ETCAR (10) and ProTCAR (11) according to prior publications [27,28]. Specifics for the synthesis of 5-bromo-1-(β-D-ribofuranosyl)-1H- [1–3]triazole-4-carboxamide (BrCAR; 12) are given below. Compound 12 and its intermediates were characterized by 1 H and 13 C NMR spectra recorded on a Bruker 400 spectrometer (Bruker Optics, Poznan, Poland) operating at 400 MHz and 100.6 MHz, respectively. Mass spectra were recorded using BrukermicroTOF-q mass spectrometer (Bruker Optics). Ribavirin and 3-deazaguanine were obtained from the former ICN Pharmaceuticals (Costa Mesa, CA, USA).

Synthesis of 5-bromo-1-(ß-D-ribofuranosyl)-1H- [1–3]triazole-4-carboxamide, 12

To a suspension of 1-(5-O-acetyl-2,3-O-isopropylidene-β-D-ribofuranosyl)-5-amino-1H- [1–3]triazole-4-carboxylic acid amide [29] (790 mg, 2.31 mmol, dried in vacuo over P2O5, at 90°C, for 4 h) in bromoform (34.4 g, 136 mmol) was added isoamyl nitrite (949 mg, 8.1 mmol). The mixture was stirred at 100°C for 2 h, then it was applied onto a silica gel column. The column was eluted with CH2Cl2 (in order to collect the bromoform) followed by CH2Cl2-MeOH (95:5) to give crude product. It was next re-chromatographed using ethyl acetate/hexane (1:1→2:1) to afford 537 mg (55% yield) of 5-bromo-1-(5-O-acetyl-2,3-O-isopropylidene-β-D-ribofuranosyl)-1H- [1–3]triazole-4-carboxylic acid amide: 1 H NMR (DMSO-d6) δ 7.98 and 7.65 (2×s, 2H, CO-NH2), 6.22 (s, 1H, 1′-H), 5.63 (d, J2′,3′ =6.0 Hz, 1H, 2′-H), 5.04 (dd, J3′,2′ =6.0Hz, J3′,4′ =2.0 Hz, 1H, 3′-H), 4.47 (m, 1H, 4′-H), 4.00 and 3.81 (2×dd, J5′,5′ =11.6 Hz, J5′,4′ =7.2 Hz, 2H, 5′-H), 1.94 (s, 3H, OAc), 1.53 and 1.36 (2×s, 6H, 2×CH3). 13 C NMR (DMSO-d6) δ 169.84 (CO), 160.40 (CO-NH2), 139.55 (C-4), 115.19 (C-5), 112.93 (CH3)2C, 91.60 (C-1′), 85.90 (C-4′), 83.37 (C-2′), 81.40 (C-3′), 63.10 (C-5′), 26.64, 24.97 (CH3)2C, 20.38 (CO-CH3). HRMS [M+Na]+ calcd for C13H17BrN4NaO6: 427.0224; found: 427.0219.

Deacetylation of the compound (435 mg, 1.04 mmol) was performed with sodium methoxide (83 mmol) in anhydrous methanol (13 ml) at 0°C. After the reaction was complete (3 h), neutralization with Dowex H+ (100 mg) prewashed with anhydrous methanol followed. The filtrate was evaporated and the residual oily product was dried in vacuo to give 370 mg (98% yield) of 5-bromo-1-(2,3-O-isopropylidene-β-D-ribofuranosyl)-1H- [1–3]triazole-4-carboxylic acid amide. 1 H NMR (DMSO-d6) δ 7.97 and 7.65 (2×s, 2H, CO-NH2), 6.17 (s, 1H, 1′-H), 5.62 (d, J2′,3′ =6.0 Hz, 1H, 2′-H), 4.98 (dd, J3′,2′ =6.0 Hz, J3′,4′ =1.6 Hz, 1H, 3′-H). 4.91 (t, J5′OH,5′ =5.6 Hz, 1H, 5′-OH), 4.21 (dt, J4′,5′ =6.0 Hz, J4′,3′ =1.6 Hz, 1H, 4′-H), 3.18 (m, 2H, 5′-H), 1.52 and 1.36 (2×s, 6H, 2×CH3). 13 C NMR (DMSO-d6) δ 160.45 (CO-NH2), 139.52 (C-4), 115.16 (C-5), 112.62 (CH3)2C, 91.57 (C-1′), 89.04 (C-4′), 83.19 (C-2), 81.78 (C-3′), 60.86 (C-5′), 26.65, 24.95 (CH3)2C. HRMS [M+Na]+ calcd for C11H15BrN4 NaO5: 385.0118; found: 385. 0126.

Isopropylidene blockade was cleaved with an ice-cooled 80% aqueous trifluoroacetic acid (4 ml/1 mmol of substrate). After the reaction was complete (0°C, 2 h), the solution was co-evaporated several times with EtOH, then with anhydrous toluene. The oily residue was dissolved in EtOAc/MeOH (10:1) and evaporated slowly to enforce precipitation of 12. It was then separated and dried in vacuo to a white powder (mp 126°C, 80% yield). 1 H NMR (DMSO-d6) δ 7.96 and 7.63 (2×s, 2H, CO-NH2), 5.86 (d, J1′,2′ =4.0 Hz, 1H, 1′-H), 5.66 (d, J2′OH, 2′ =5.6 Hz, 1H, 2′-OH), 5.31 (d,J3′OH, 3′ =6.0 Hz, 1H, 3′-OH), 4.78 and 4.75 (t overlap. dd, J5′OH,5′ =5.6 Hz, 2H, 5′-OH, 2′-H), 4.26 (dd, J3′,2′ =4.8 Hz, J3′,3′OH =6.0 Hz, 1H, 3′-H), 3.99 (dd, J4′, 5′ =7.2 Hz, J4′, 5′ =5.6 Hz, 1H, 4′-H), 3.53 and 3.41 (2×m, 2H, 5′-H). 13 C NMR (DMSO-d6) δ160.60 (CO-NH2), 139.29 (C-4), 115.30 (C-5), 90.34 (C-1′), 86.37 (C-4′), 73.58 (C-2′), 70.53 (C-3′), 61.58 (C-5′). HRMS [M+Na]+ calcd for C8H11BrN4 NaO5: 344.9805; found: 344.9802.

In vitro antiviral evaluations

Viruses and cells

Influenza A/California/07/2009 (H1N1), A/Perth/16/2009 (H3N2) and B/Florida/4/2006 viruses were obtained from the Centers for Disease Control and Prevention (Atlanta, GA, USA). Influenza A/Duck/MN/1525/81(H5N1), a low pathogenic avian virus isolated in the United States, was kindly provided by Robert Webster (St Jude Children’s Research Hospital, Memphis, TN, USA). Experiments with the above influenza virus strains were carried out in Madin–Darby canine kidney (MDCK) cells. Parainfluenza type 3 (PIV3) virus strain 14702, RSV strain A2 and measles virus strain CC were obtained from the American Type Culture Collection (ATCC; Manassas, VA, USA). Experiments with PIV-3 and RSV were performed in African green monkey kidney (MA-104) cells, whereas measles virus was assayed in the Vero 76 line of African green monkey kidney cells. The cell lines were all obtained from ATCC.

Antiviral assays

Inhibition of viral cytopathic effect assays were performed as previously described using confluent monolayers of cells in 96-well microplates [30], with cytotoxicity determined in parallel in uninfected cells. Trypsin (10 units/ml; Sigma–Aldrich, St Louis, MO, USA) in serum-free medium was required to facilitate the replication and cytopathology of influenza A and B viruses. Assays with the other viruses were performed in medium containing 2% fetal bovine serum. Quantitation of percent cytopathic effect was accomplished by treating cells with 0.011% neutral red dye for 2 h, then determining dye uptake into viable cells [31]. Unincorporated dye was removed from cells by aspiration and rinsing with phosphate-buffered saline, then the absorbed dye was eluted from the cells for 30 min with 0.1 ml of 50% Sörensen’s citrate buffer (pH 4.2)/50% ethanol. The plates were read for optical density determinations at 540 nm with 405 nm reference. Readings were converted to percentages of uninfected control and normalized to the virus control. The 50% virus inhibitory concentration (IC50) values and 50% cytotoxic concentration (CC50) values were determined by linear regression analysis. Selectivity index (SI) values were calculated as CC50/IC50.

Computational methods with IMPDH

Analysed 5′-monophosphates of 1,2,3-triazole nucleosides: 13, 14 or 15 were modelled into the substrate binding site of the solved X-ray crystallographic structure of hIMPDH2 complexed with RMP (compound 2) as substrate and C2-mycophenolic adenine dinucleotide as a cofactor analogue. The crystal structure of hIMPDH2 was taken from the Protein Data Bank: 1NF7 [32]. To our knowledge this crystal structure has never been published in a scientific journal. Modelling was performed with the Accelrys Discovery Studio 3.0 suite of programmes (Accelrys Software, San Diego, CA, USA).

All water molecules were removed and hydrogen atoms were added. While coordinates of all heavy atoms of the enzyme were kept frozen, hydrogen atoms positions were optimized with the use of adopted basis Newton-Raphson minimization algorithm (max steps =500, RMS gradient =0.01, distance-dependent dielectrics as implicit solvent model). Compounds 13, 14 or 15 were initially modelled into the binding site via super-positioning onto the RMP binding site of the hIMPDH2 crystal structure.

For the docking procedure the dock ligands (CDOCKER) protocol, an implementation of the CDOCKER algorithm [33] within the Discovery Studio 3.0 package was used. Random ligand conformations were generated from the initial ligand structure through high temperature molecular dynamics at 700 K. The binding site was defined from the volume of a co-crystallized ligand (RMP). The random conformations were refined by grid-based simulated annealing and a final full force field minimization. In all calculations for proteins and ligands, CHARM forcefield with MMFF94 partial charge rules was used. To account for protein flexibility, the ligand binding site of each model was further refined by restraint energy minimization.

HPLC analysis of intracellular nucleotide pools

Confluent MDCK cell cultures in 6-well microplates were treated with compounds in serum-free medium for 24 h. The medium was aspirated from each well, followed by precipitation and breaking open of the cells with 3.5% perchloric acid and neutralization of the acid with 1 N KOH/1 M imidazole. The supernates were collected from each well and frozen at -80°C until analysed by HPLC.

HPLC analysis of the samples was conducted as previously published [34]. Briefly, chromatographic separations were performed using an HPLC apparatus (Waters Corp., Milford, MA, USA) fitted with a 10×250 mm column packed with Whatman partisil SAX resin (Phenomenex, Torrance, CA, USA). A linear gradient from 10 mM to 1 M potassium phosphate (Sigma–Aldrich) at pH 5.0 was run over 30 min at 1 ml/min. The 1 M buffer ran an additional 2 min before re-equilibrating the column in low salt buffer. Detection of nucleotides was made at 260 nm with a Waters LC spectrophotometer and the peaks analysed using an integrator (Shimadzu Scientific Instruments, Columbia, MD, USA). Relative peak areas of D282-treated cells were compared to untreated cells and expressed as percentages of control levels.

Results

Synthesis of compounds

Alkynyl-1,2,3-triazole nucleosides 10 and 11 (Figure 1) were prepared by Stille and copper-free Sonogashira reactions, respectively, as described [27,35]. Other authors synthesized 10 according to a standard Sonogashira coupling [36]. Bromotriazole nucleoside 12 (Figure 1) was prepared by diazotization of a protected 5-amino-precursor with organic nitrite in bromoform, followed by a two-step deprotection. Importantly, several groups reported on halogenation to take place at C-5 of 1,4-disubstituted 1,2,3-triazoles during reactions of 5-cuprated 1,2,3-triazole derivatives (key intermediates in CuAAC reactions) with electrophiles: I+ , Br+ and Cl+ [3640]. In addition, O-protected N6 -substituted derivative of 12 could be found among compounds prepared to exemplify this new and versatile method [37,38].

In vitro antiviral and cytotoxic activities

Primary in vitro screening tests were conducted against a broad panel of 12 RNA viruses. Further assays allowed us to select compounds 1012 (Figure 1) for interesting antiviral activity profiles. ETCAR was found to be the most active of the three compounds evaluated, inhibiting both orthomyxo- and paramyxoviruses (Table 1). ProTCAR was only weakly inhibitory to influenza, measles and PIV- 3. A greater effect was seen against RSV. BrCAR exhibited its greatest inhibitory effect against influenza B and RSV. Concurrently run cytotoxicity assays demonstrated that ETCAR exhibited greater cytotoxicity than the other compounds. This made its SI values smaller than those of ribavirin, which exhibits less toxicity. Otherwise, the potencies of ETCAR and ribavirin (as demonstrated by IC50 values) were quite similar against the various viruses.

-
Table 1.  Antiviral activities of triazole compounds and ribavirin against certain respiratory viruses in cell culture
Table 1. Antiviral activities of triazole compounds and ribavirin against certain respiratory viruses in cell culture

Viral cytopathic effect inhibition assays were conducted by a neutral red dye uptake method. a Ribavirin was the positive control for testing influenza, parainfluenza type 3 (PIV-3) virus, and respiratory syncytial virus (RSV); 3-deazaguanine was the positive control for testing measles. b Data for 50% inhibitory concentration (IC50) and 50% cytotoxic concentration (CC50) are mean (±sd) of 2–3 independent assays. c The selectivity index (SI) was calculated as CC50/IC50. MDCK, Madin–Darby canine kidney.

Interaction of compounds with IMPDH

The structural resemblances of 10 and 11 to ribavirin and EICAR might imply several plausible mechanisms of antiviral action. First of all, IMPDH inhibition should be considered, assuming activation of nucleosides 10 and 11 to corresponding 5′-O-phosphates 13 and 14 (Figure 1). To gain insight into possible interactions with the enzyme and to correlate observed differences to biological activity of compounds, computational modelling studies on the complexes of hIMPDH2 with substrate analogues 13, 14 or 15, were undertaken as a complementary study.

Compound 13 (ETCARMP) was modelled into the RMP (2) binding site in the crystallographic structure (PDB: 1NF7) of the hIMPDH2 ternary complex. The phosphate moiety and ribose of 13 show consistent overlap with these parts of RMP and form the analogous hydrogen-bonding network with the conserved residues of the IMP binding pocket, as was observed in the crystal structures of IMPDH complexes with IMP, XMP or XMP covalently bound to the key catalytic cysteine residue. The ribose ring hydroxyl groups form a hydrogen-bonding network to Ser68 and Asp364. The phosphate moiety forms bifurcated hydrogen bonds with Gly387, two hydrogen bonds with Ser388 and one with Gly326 (Figure 2A).

-
Figure 2.
Figure 2. Computational modelling and docking of the compound 13 (ETCARMP) into the ribavirin 5′-monophosphate (2) binding site in the crystallographic structure of the hIMPDH2

Crystallographic structure from Protein Data Bank: 1NF7. (A) Schematic representation of the structure and hydrogen-bonding network of ETCARMP in the substrate binding site. (B) Visualization of the substrate binding site in the ternary complex of human type II inosine monophosphate dehydrogenase (hIMPDH2) with NAD+ and ETCARMP. The atoms of NAD+, the catalytic Cys331 and Thr333 residues are highlighted by a coloured space-filling model representation. (C) Superimposition of the dominant poses of docked ETCARMP to the hIMPDH2 substrate binding site. The catalytic Cys331 is highlighted.

The 4,5-disubstituted 1,2,3-triazole ring is well accommodated in the substrate binding site of the hIMPDH2 enzyme and makes three hydrogen bonds with the active site flap: N(2) and N(3) of triazole form bifurcated hydrogen bonds with the Met414 backbone amide NH group, N(6)H atoms of 4-carboxamide group form a hydrogen bond with Gly415 backbone carbonyl. The heterocyclic N(3) and amide N(6) atoms are on the same side of the triazole moiety. In this configuration N(6) and O(6) atoms are spatially similar to the N(6) and N(1) atoms of adenosine, which can be of great importance for the biological activity of ETCARMP (13), because its antiviral activity might be exerted at least in part through its mutagenic activity. This configuration that depends on the rotation of the carboxamide group in 5-amido 1,2,3-triazoles shows lower relative energy than the opposite one (10.891 kJ/mol versus 49.268 kJ/mol) [41] as was found in the crystallographic structure of 5-fluoromethyl-1-β-D-ribofuranosyl)-1H- [1–3]triazole-4-carboxamide [28].

By contrast, examination of the X-ray structures of the RMP complexes with IMPDH revealed that in all of them ribavirin (1) is an excellent substrate mimic and its O(7) and N(8) atoms are spatially similar to the O(6) and N(1) atoms of inosine and guanosine [42,43]. This configuration of 3-carboxamide group was also found in two polymorphic ribavirin crystalline forms in the X-ray structure of 1-β-D-ribofuranosyl-1,2,4-triazole-3-carboxamide [44]. The rotation of ribavirin’s (1) carboxamide group has much lower energy barrier (25.593 kJ/mol versus 20.229 kJ/mol calculated on the level of the CONH2 substituted 1,2,4-triazoles) [41]. In consequence, ribavirin can be incorporated into RNA as a base analogue of either adenine or guanine and it pairs equally well with either uracil or cytosine, inducing mutations in the viral genome [45,46]. One can expect that ETCARMP (13) may act as mutagenic agent as well and base-pair with uracil.

Similar to EICARMP (4), ETCARMP (13) may be a potent irreversible inactivator of the hIMPDH2 enzyme [20]. Because of the anti conformation of the glycosidic bond (χ=245°), the 5′-ethynyl substituent lies away from the sugar ring and proximal to the catalytic Cys331 assisted by Thr333. It has been shown that Thr333 residue is completely conserved among IMPDHs [47] and its mutation decreases the rate of hydride transfer, suggesting that Thr333 may influence the reactivity of the catalytic Cys by activating the thiol [48]. The sulfur atom of Cys331 and the electrophilic carbon atom of the 5-ethynyl group are separated by 2.6 Å (Figure 2A and 2B). This is an excellent position for a nucleophilic attack and formation of a covalent adduct. Although no stacking interaction between the cofactor nicotinamide ring and the triazole is observed, a distance between the C4 of nicotinamide moiety and the ethynyl group is very short (2.6 Å; Figure 2B). The reactivity of the 5-ethynyl is enhanced by hydrogen-bonds between the triazole moiety and the main chain at residues Met414 and Gly415.

CDOCKER docking results are shown in Figure 2C. One can see that dominant poses of ETCARMP are similarly located at the IMP binding pocket, as it was observed in the crystal structures of IMPDH complexes with IMP, XMP* or XMP covalently bound to the catalytic cysteine residue.

Compound 14 (ProTCARMP), when modelled into the substrate binding site of hIMPDH2 enzyme in the 1NF7 crystallographic structure using the same procedure as for 13, does not have enough manoeuvring room. The methyl of the 5-propynyl substituent comes into collision with Gly326 residue of the active site loop and the nicotinamide portion of NAD+ . In the absence of the cofactor, ProTCARMP is located in the substrate binding site similar to ETCARMP, with slight reorientation of the 5-propynyl substituent toward the empty NAD+ site (Figure 3A). In consequence, the electrophilic carbon atom of the 5-propynyl group shifts away from the nucleophilic sulfur atom of Cys331. The glycosidic torsion angle, which describes the relative orientation of the base with respect to the sugar, is similar to compound 13, in the anti conformation, close to high anti region (χ=254°). It has been shown previously, that the antiviral effectiveness of several analogues of ribavirin is related to their stability in the high anti/anti conformation [49].

-
Figure 3.
Figure 3. Computational modelling and docking of the compound 14 (ProTCARMP) into the ribavirin 5′-monophosphate (2) binding site in the crystallographic structure of the hIMPDH2

Crystallographic structure from Protein Data Bank: 1NF7. (A) Schematic representation of the structure and hydrogen-bonding network of ProTCARMP in the substrate binding site. (B) Superimposition of the dominant poses of docked ProTCARMP the human type II inosine monophosphate dehydrogenase (hIMPDH2) binding site. The catalytic Cys331 and Gly326 residues are highlighted by a coloured space-filling model representation.

Figure 3B presents dominant poses of ProTCARMP obtained in the CDOCKER docking procedure. The orientation of ProTCARMP in the substrate binding site is like ETCARMP with the reorientation of the 5-propynyl group shifting away from the Gly326 residue.

The bulky 5-bromo substituent in compound 15 substantially influences its binding properties. When modelled into the substrate binding site of hIMPDH2 enzyme, 15 does not have enough room and the 5-bromo substituent comes into collision with neighbouring residues (Figure 4). Docking analysis with CDOCKER procedure shows poses with different orientations of the base within the binding pocket (Figure 4B and 4C). The predominant two groups of poses were found, characterized by the glycosidic torsion angle χ in the region of 173° and 224°, corresponding to anti conformation. Our results of antiviral screening of compound 12 showing that it is weakly active are consistent with previous reports of other ribavirin analogues containing bulky substituent in the 5-position [50].

-
Figure 4.
Figure 4. Computational modelling and docking of the BrCAR with a 5′-bromo substituent modelled into the substrate binding site of hIMPDH2 enzyme

Crystallographic structure from Protein Data Bank: 1NF7. (A) Schematic representation of the structure of the compound 15 modelled into the substrate binding site of human type II inosine monophosphate dehydrogenase (hIMPDH2) enzyme. (B) The predominant two groups of poses of BrCAR found by CDOCKER docking with different orientations of the base within the binding pocket. (C) The poses with the glycosidic torsion angle c in the region of 173. (D) The poses with the glycosidic torsion angle c in the region of 224.

Effects of compounds on intracellular nucleotide pools

The IMPDH assays described above do not directly prove whether treatment of cells with the compounds leads to reductions in intracellular GTP concentrations as a result of inhibition of IMPDH. For a direct demonstration we treated MDCK cells with compounds 1012 and ribavirin, followed 24 h later by an analysis of intracellular nucleotide pools (Figure 5). Treatment of cells with all compounds resulted in suppression of intracellular GTP levels, but higher extracellular concentrations of compounds 11 and 12 were required to achieve inhibition comparable to compound 10 and ribavirin. Compound 10 had greater inhibitory effects on GTP pools than ribavirin at 3.2 to 100 μM. Minor inhibitory effects on cytidine triphosphate (CTP) levels occurred at higher concentrations of compound 11 and ribavirin. Compounds 10, 11 and ribavirin had minimal effects on uridine triphosphate (UTP) and adenosine triphosphate (ATP) levels. Compound 12 was most unusual because treated cells exhibited high levels of UTP, some elevations in CTP, and decreases in ATP levels. These results suggest that 12 has more than one cellular activity besides inhibition of IMPDH.

-
Figure 5.
Figure 5. Nucleotide pool changes in Madin–Darby canine kidney cells treated with varying micromolar concentrations of compound 10 (ETCAR), compound 11 (ProTCAR), compound 12 (BrCAR) or ribavirin for 24 h

Uridine triphosphate (UTP), adenosine triphosphate (ATP), cytidine triphosphate (CTP) and guanosine triphosphate (GTP) levels were determined by strong anion exchange HPLC analysis. The results represent averages of three independent experiments ±sd. CPD, compound; RBV, ribavirin.

Discussion

In these studies, we demonstrated how small or profound changes at the C-5 position of 1,2,3-triazole-4-carboxamide, a pseudobase isomeric to that of ribavirin, modulate the antiviral activity of nucleosides 1012. It was determined that the smaller the substituent (ethynyl, 10), the greater the antiviral potency. The presence of the bulky 5-bromo group greatly diminished the antiviral activity of 12. We could rationalize the observed activity of 10 and 11 against orthomyxo- and/or paramyxoviruses enabling their 5′-monophosphates 13 and 14, respectively, to inhibit IMP dehydrogenase in host cells. Assuming that a cellular enzyme, adenosine kinase effectively transforms 1012 into metabolites 1315, we modelled each of the latter compounds into the substrate binding site of hIMPDH2 enzyme in the 1NF7 crystallographic structure. Only the least hindered compound 13 fitted perfectly, whereas 14 and 15 located into the site in the absence of NAD+ . The docking analysis of 15 showed that the compound preferentially adopted the opposite orientation in the substrate binding side with respect to the compounds 13 and 14, and in consequence would be useless as an inhibitor of the enzyme. On the contrary, the docking analyses of 13 and 14 supported our presumption that both compounds would be inhibitory to the enzyme hIMPDH2, although with different ability.

Results of studying effects of compounds on intracellular nucleotide pools revealed a more complex picture than simple reduction in GTP levels. Compounds 11 and 12 were approximately equally suppressive of GTP, yet compound 12 had additional effects on UTP, CTP and ATP. Inhibition of ATP pools has negative consequences on cell function that may manifest as toxicity, such as in an animal. In cell culture, overt toxicity sometimes is not manifest, particularly if the assays are performed in stationary monolayers as was done here. The two compounds (11 and 12) had similar but not identical antiviral sensitivities. Whether compounds 10, 11 or 12 phosphorylate up to the triphosphate level is not known. Evidence for phosphorylation to the monophosphate level is provided by the fact that treatment of cells results in GTP suppression.

It has been pointed out that a drug combination therapy of influenza virus infections would be feasible to prevent an emergence of new drug-resistant strains of the virus [51]. Antiviral drugs used in combination should exert their activities by different modes of action [52]. Thus, ribavirin or a selection of its analogues could be important components of such combinations.

Acknowledgements

Financial support from the Polish Ministry of Science and Higher Education (project numbers N N405 2516 33 and N N519 4050 37) is gratefully acknowledged. Cell culture studies were funded by contract N01-AI-15435 from the Virology Branch, Division of Microbiology and Infectious Diseases, National Institute of Allergy and Infectious Diseases, National Institutes of Health, Bethesda, MD, USA.

Disclosure statement

The authors declare no competing interests.

References

-
1. Sidwell RW, Huffman JH, Khare GP, Allen LB, Witkowski JT, Robins RK. Broad-spectrum antiviral activity of virazole: 1-β-D-ribofuranosyl-1,2,4-triazole-3-carboxamide. Science 1972; 177: 705-706. Medline doi:10.1126/science.177.4050.705
-
2. Witkowski JT, Robins RK, Sidwell RW, Simon LN. Design, synthesis, and broad spectrum antiviral activity of 1-β-D-ribofuranosyl-1,2,4-triazole-3-carboxamide and related nucleosides. J Med Chem 1972; 15: 1150-1154. Medline doi:10.1021/jm00281a014
-
3. Krilov LR. Respiratory syncytial virus disease: update on treatment and prevention. Expert Rev Anti Infect Ther 2011; 9: 27-32. Medline doi:10.1586/eri.10.140
-
4. Gaetano JN, Reau N. Hepatitis C: management of side effects in the era of direct-acting antivirals. Curr Gastroenterol Rep 2013; 15: 305 Medline doi:10.1007/s11894-012-0305-1
-
5. Parker WB. Metabolism and antiviral activity of ribavirin. Virus Res 2005; 107: 165-171. Medline doi:10.1016/j.virusres.2004.11.006
-
6. Leyssen P, De Clercq E, Neyts J. Molecular strategies to inhibit the replication of RNA viruses. Antiviral Res 2008; 78: 9-25. Medline doi:10.1016/j.antiviral.2008.01.004
-
7. Streeter DG, Witkowski JT, Khare GP, et al. Mechanism of action of 1-β-D-ribofuranosyl-1,2,4-triazole-3-carboxamide (virazole), a new broad-spectrum antiviral agent. Proc Natl Acad Sci U S A 1973; 70: 1174-1178. Medline doi:10.1073/pnas.70.4.1174
-
8. Leyssen P, Balzarini J, De Clercq E, Neyts J. The predominant mechanism by which ribavirin exerts its antiviral activity in vitro against flaviviruses and paramyxoviruses is mediated by inhibition of IMP dehydrogenase. J Virol 2005; 79: 1943-1947. Medline doi:10.1128/JVI.79.3.1943-1947.2005
-
9. Smee DF, Bray M, Huggins JW. Antiviral activity and mode of action studies of ribavirin and mycophenolic acid against orthopoxviruses in vitro. Antivir Chem Chemother 2001; 12: 327-335. Medline
-
10. Bougie I, Bisaillon M. The broad spectrum antiviral nucleoside ribavirin as a substrate for a viral RNA capping enzyme. J Biol Chem 2004; 279: 22124-22130. Medline doi:10.1074/jbc.M400908200
-
11. Eriksson B, Helgstrand E, Johansson NG, et al. Inhibition of influenza virus ribonucleic acid polymerase by ribavirin triphosphate. Antimicrob Agents Chemother 1977; 11: 946-951. Medline doi:10.1128/AAC.11.6.946
-
12. Crotty S, Cameron CE, Andino R. RNA virus error catastrophe: direct molecular test by using ribavirin. Proc Natl Acad Sci U S A 2001; 98: 6895-6900. Medline doi:10.1073/pnas.111085598
-
13. Lanford RE, Chavez D, Guerra B, et al. Ribavirin induces error-prone replication of GB virus B in primary tamarin hepatocytes. J Virol 2001; 75: 8074-8081. Medline doi:10.1128/JVI.75.17.8074-8081.2001
-
14. Contreras AM, Hiasa Y, He W, Terella A, Schmidt EV, Chung RT. Viral RNA mutations are region specific and increased by ribavirin in a full-length hepatitis C virus replication system. J Virol 2002; 76: 8505-8517. Medline doi:10.1128/JVI.76.17.8505-8517.2002
-
15. Severson WE, Schmaljohn CS, Javadian A, Jonsson CB. Ribavirin causes error catastrophe during Hantaan virus replication. J Virol 2003; 77: 481-488. Medline doi:10.1128/JVI.77.1.481-488.2003
-
16. Day CW, Smee DF, Julander JG, Yamshchikov VF, Sidwell RW, Morrey JD. Error-prone replication of West Nile virus caused by ribavirin. Antiviral Res 2005; 67: 38-45. Medline doi:10.1016/j.antiviral.2005.04.002
-
17. Shah NR, Sunderland A, Grdzelishvili VZ. Cell type mediated resistance of vesicular stomatitis virus and Sendai virus to ribavirin. PLoS ONE 2010; 5: e11265 Medline doi:10.1371/journal.pone.0011265
-
18. De Clercq E, Cools M, Balzarini J, et al. Antiviral activities of 5-ethynyl-1-β-D-ribofuranosylimidazole-4- carboxamide and related compounds. Antimicrob Agents Chemother 1991; 35: 679-684. Medline doi:10.1128/AAC.35.4.679
-
19. Balzarini J, Karlsson A, Wang L, et al. Eicar (5-ethynyl-1-β-D-ribofuranosylimidazole-4-carboxamide). A novel potent inhibitor of inosinate dehydrogenase activity and guanylate biosynthesis. J Biol Chem 1993; 268: 24591-24598. Medline
-
20. Wang W, Papov VV, Minakawa N, Matsuda A, Biemann K, Hedstrom L. Inactivation of inosine 5′-monophosphate dehydrogenase by the antiviral agent 5-ethynyl-1-β-D-ribofuranosylimidazole-4-carboxamide 5′-monophosphate. Biochemistry 1996; 35: 95-101. Medline doi:10.1021/bi951499q
-
21. Moya J, Pizarro H, Jashes M, De Clercq E, Sandino AM. In vivo effect of EICAR (5-ethynyl-1-β-D-ribofuranosylimidazole-carboxamide) on experimental infected rainbow trout (Oncorhynchus mykiss) and coho salmon (Oncorhynchus kisutch) fry with infectious pancreatic necrosis virus. Antiviral Res 2000; 48: 125-130. Medline doi:10.1016/S0166-3542(00)00122-4
-
22. Dal Pozzo F, Galligioni V, Vaccari F, Gallina L, Battilani M, Scagliarini A. Antiviral efficacy of EICAR against canine distemper virus (CDV) in vitro. Res Vet Sci 2010; 88: 339-344. Medline doi:10.1016/j.rvsc.2009.08.010
-
23. Chung DH, Kumarapperuma SC, Sun Y, et al. Synthesis of 1-β-D-ribofuranosyl-3-ethynyl-[1,2,4]triazole and its in vitro and in vivo efficacy against Hantavirus. Antiviral Res 2008; 79: 19-27. Medline doi:10.1016/j.antiviral.2008.02.003
-
24. McDowell M, Gonzales SR, Kumarapperuma SC, Jeselnik M, Arterburn JB, Hanley KA. A novel nucleoside analog, 1-β-D-ribofuranosyl-3-ethynyl-[1,2,4]triazole (ETAR), exhibits efficacy against a broad range of flaviviruses in vitro. Antiviral Res 2010; 87: 78-80. Medline doi:10.1016/j.antiviral.2010.04.007
-
25. Wan J, Xia Y, Liu Y, et al. Discovery of novel arylethynyltriazole ribonucleosides with selective and effective antiviral and antiproliferative activity. J Med Chem 2009; 52: 1144-1155. Medline doi:10.1021/jm800927r
-
26. Moriyama K, Suzuki T, Negishi K, et al. Effects of introduction of hydrophobic group on ribavirin base on mutation induction and anti-RNA viral activity. J Med Chem 2008; 51: 159-166. Medline doi:10.1021/jm7009952
-
27. Ostrowski T, Zeidler J. Synthesis of 5-ethynyl-1-β-D-ribofuranosyl-1H- [[1–3]]triazole-4-carboxylic acid amide (isosteric to EICAR) and its derivatives. Nucleic Acids Symp Ser 2008; 52: 585-586. doi:10.1093/nass/nrn296
-
28. Ostrowski T, Januszczyk P, Cieslak M, et al. 5-Ethynyl-1-β-D-ribofuranosyl-1H-[1–3]triazole-4-carboxylic acid amide (ETCAR) and its analogues: synthesis and cytotoxic properties. Bioorg Med Chem 2011; 19: 4386-4398. Medline doi:10.1016/j.bmc.2011.05.050
-
29. Biagi G, Giorgi I, Livi O, Scartoni V. Synthesis of new 2-substituted 9-β-D-ribofuranosyl-8-azahypoxanthines. Farmaco 1992; 47: 525-536.
-
30. Smee DF, Huffman JH, Morrison AC, Barnard DL, Sidwell RW. Cyclopentane neuraminidase inhibitors with potent in vitro anti-influenza virus activities. Antimicrob Agents Chemother 2001; 45: 743-748. Medline doi:10.1128/AAC.45.3.743-748.2001
-
31. Smee DF, Morrison AC, Barnard DL, Sidwell RW. Comparison of colorimetric, fluorometric, and visual methods for determining anti-influenza (H1N1 and H3N2) virus activities and toxicities of compounds. J Virol Methods 2002; 106: 71-79. Medline doi:10.1016/S0166-0934(02)00137-4
-
32. Risel D, Strickler MD, Goldstein BM. Crystal structure of human inosine monophosphate dehydrogenase type II complexed with the MPA/NAD analog C2-MAD. RCSB Protein Data Bank 2004; doi: 10.2210/pdb1nf7/pdb.10.2210/pdb1nf7/pdb
-
33. Wu G, Robertson DH, Brooks CL, Vieth M. Detailed analysis of grid-based molecular docking: a case study of CDOCKER-A CHARMm-based MD docking algorithm. J Comput Chem 2003; 24: 1549-1562. Medline doi:10.1002/jcc.10306
-
34. Smee DF, Hurst BL, Day CW. D282, a non-nucleoside inhibitor of influenza virus infection that interferes with de novo pyrimidine biosynthesis. Antivir Chem Chemother 2012; 22: 263-272. Medline doi:10.3851/IMP2105
-
35. Ostrowski T, Januszczyk P, Cieslak M, et al. 5-Ethynyl-1-β-D-ribofuranosyl-1H-[1–3]triazole-4-carboxylic acid amide (ETCAR) and its analogues: synthesis and cytotoxic properties. Bioorg Med Chem 2011; 19: 4386-4398. Medline doi:10.1016/j.bmc.2011.05.050
-
36. Malnuit V, Duca M, Mantuo A, Bougrin K, Benhida R. Tandem azide-alkyne 1,3-dipolar cycloaddition/electrophilic addition: a concise three-component route to 4,5-disubstituted triazolyl-nucleosides. Synlett 2009; 2009: 2123-2126. doi:10.1055/s-0029-1217560
-
37. Li L, Li Y, Li R, Zhu A, Zhang G. A new synthetic protocol for one-pot preparations of 5-halo-1,4-disubstituted-1,2,3-triazoles. Aust J Chem 2011; 64: 1383-1389. doi:10.1071/CH11067
-
38. Li L, Li R, Zhu A, Zhang G, Zhang L. CuBr-NCS-mediated multicomponent azide-alkyne cycloaddition: mild and efficient synthesis of 5-bromo-1,4-disubstituted-1,2,3-triazoles. Synlett 2011; 2011: 874-878. doi:10.1055/s-0030-1259908
-
39. Ackermann L, Potukuchi HK. Regioselective syntheses of fully-substituted 1,2,3-triazoles: the CuAAC/C-H bond functionalization nexus. Org Biomol Chem 2010; 8: 4503-4513. Medline doi:10.1039/c0ob00212g
-
40. Li L, Zhang G, Zhu A, Zhang L. A convenient preparation of 5-iodo-1,4-disubstituted-1,2,3-triazole: multicomponent one-pot reaction of azide and alkyne mediated by CuI-NBS. J Org Chem 2008; 73: 3630-3633. Medline doi:10.1021/jo800035v
-
41. Oziminski WP, Dobrowolski JC, Mazurek AP. DFT studies on tautomerism of C5-substituted 1,2,4-triazoles. J Mol Struct THEOCHEM 2004; 680: 107-115. doi:10.1016/j.theochem.2004.05.005
-
42. Prosise GL, Wu JZ, Luecke H. Crystal structure of Tritrichomonas foetus inosine monophosphate dehydrogenase in complex with the inhibitor ribavirin monophosphate reveals a catalysis-dependent ion-binding site. J Biol Chem 2002; 277: 50654-50659. Medline doi:10.1074/jbc.M208330200
-
43. Sintchak MD, Nimmesgern E. The structure of inosine 5′-monophosphate dehydrogenase and the design of novel inhibitors. Immunopharmacology 2000; 47: 163-184. Medline doi:10.1016/S0162-3109(00)00193-4
-
44. Prusiner P, Sundaralingam M. The crystal and molecular structures of two polymorphic crystalline forms of virazole (1-β-D-ribofuranosyl-1,2,4-triazole-3-carboxamide). A new synthetic broad sprectrum antiviral agent. Acta Crystallogr 1976; B32: 415-419.
-
45. Maag D, Castro C, Hong Z, Cameron CE. Hepatitis C virus RNA-dependent RNA polymerase (NS5B) as a mediator of the antiviral activity of ribavirin. J Biol Chem 2001; 276: 46094-46098. Medline doi:10.1074/jbc.C100349200
-
46. Vo NV, Young KC, Lai MM. Mutagenic and inhibitory effects of ribavirin on hepatitis C virus RNA polymerase. Biochemistry 2003; 42: 10462-10471. Medline doi:10.1021/bi0344681
-
47. Patton GC, Stenmark P, Gollapalli DR, et al. Cofactor mobility determines reaction outcome in the IMPDH and GMPR (β-α)8 barrel enzymes. Nat Chem Biol 2011; 7: 950-958. Medline doi:10.1038/nchembio.693
-
48. Guillén Schlippe YV, Hedstrom L. Is Arg418 the catalytic base required for the hydrolysis step of the IMP dehydrogenase reaction? Biochemistry 2005; 44: 11700-11707. Medline doi:10.1021/bi048342v
-
49. Miles DL, Miles DW, Eyring H. A conformational basis for the antiviral inactivity of tetrazole ribonucleosides. Biochim Biophys Acta 1978; 518: 17-30. Medline doi:10.1016/0005-2787(78)90112-0
-
50. Naik SR, Witkowski JT, Robins RK. Synthesis of nucleosides of 5-substituted-1,2,4-triazole-3-carboxamides. J Heterocycl Chem 1974; 11: 57-61. doi:10.1002/jhet.5570110112
-
51. Nguyen JT, Hoopes JD, Le MH, et al. Triple combination of amantadine, ribavirin, and oseltamivir is highly active and synergistic against drug resistant influenza virus strains in vitro. PLoS ONE 2010; 5: e9332 Medline doi:10.1371/journal.pone.0009332
-
52. De Clercq E. Antiviral agents active against influenza A viruses. Nat Rev Drug Discov 2006; 5: 1015-1025. Medline doi:10.1038/nrd2175

Copyright © 2019 Nucleus Holdings Ltd. Part of Nucleus Global.
Design and Technology by Nucleus Global
Company registration No. 321 0712 (England & Wales). Registered Office address: Admiral House 76-78 Old Street, London EC1V 9AZ.