In silico screening for human norovirus antivirals reveals a novel non-nucleoside inhibitor of the viral polymerase

Figures

• 

  Figure 1. Norovirus inhibitors in pre-clinical studies. Nitazoxanide (1), 2′-C-methylcytidine (2), valopicitabine (3), PPNDS (4), suramin (5).

• 

  Figure 2. Pharmacophore models used for the structure-based virtual screening. (A) Model based on suramin binding to the MNV RdRp 3UR0 crystal structure. The model consists of three aromatic features (purple, green and turquoise), one hydrogen bond acceptor (pink) and two hydrogen bond acceptors or anionic groups (orange and yellow). Only the interacting portion of suramin has been considered for the pharmacophore creation. (B) Model based on PPNDS binding to the human norovirus RdRp 4LQ3 crystal structure. The model consists of two aromatic features (purple and green), three hydrogen bond acceptors and anionic groups (pink, turquoise and yellow) and one hydrogen bond donor and acceptor. Exclusion volumes are hidden for clarity. The two binding site areas are represented as molecular surface. The MNV RdRp structure is shown as dark green ribbon, while the human norovirus RdRp is represented as salmon ribbon.

• 

  Figure 3. Queries generated using vROCS. (A) Query generated using the partial structure of suramin. (B) Query generated using the structure of PPNDS. The overall molecular shape of both the inhibitors is represented as a grey surface, while the chemical features considered are an H-bond acceptor (red sphere) and hydrophobic/aromatic centres (green spheres).

• 

  Figure 4. Initial screen for inhibition of human norovirus Sydney 2012 RdRp activity. The effects of 62 compounds were examined against norovirus RdRp activity using a fluorescent activity assay at 10 µM (A) and 100 µM (B). Five compounds that exhibited more than 25% inhibition at 10 µM and 50% inhibition at 100 µM (red bars) and three soluble compounds that exhibited more than 25% inhibition at both concentrations (green bars) were chosen for further assessment with a counter-screen gel-shift RdRp assay. The positive control (blue bars) is a previously reported norovirus NNI RdRp inhibitor NIC02 (10 µM in (A) and 100 µM in (B)14. The mean values of triplicate datasets with standard deviation are shown. (C) Structure of the eight most potent compounds with 11 and 13 highlighted. (D) A counter-screen gel-shift assay was used to confirm norovirus RdRp inhibitory activity of hit compounds. The eight compounds were examined for inhibition of primed elongation activity. PE44-NoV RNA templates (32 nucleotides) were extended (44 nucleotides) by the RdRp in the absence of any test compounds (0.5% DMSO [vol/vol] negative control) or with test compounds at a fixed concentration of 100 µM. The…

• 

  Figure 5. General synthetic procedure followed for the preparation of 23a-c, 24a-d, 25a-d. (i) Na2CO3, Pd(PPh3)2, THF/H2O, r.t. 8 h, 60 °C 12 h; (ii) β-alanine, AcOH, 100 °C 1 h; (iii) K2CO3, DMF, r.t. o.n., 60 °C 72 h; (iv) secondary amine, DIPEA, DCM, 0 °C to r.t. 20 min.

• 

  Figure 6. Alternative method for the preparation of the free acid derivative. (i) β-alanine, AcOH, 100 °C 1 h; (ii) K2CO3, DMF, r.t. o.n., 60 °C 72 h.

• 

  Figure 7. Alternative method for the preparation of the free acid derivatives (i) β-alanine, AcOH, 100 °C 1 h; (ii) K2CO3, DMF, r.t. o.n., 60 °C 72 h; (iii) Na2CO3, Pd(PPh3)2, THF/H2O, r.t. 8 h, 60 °C 12 h.

• 

  Figure 8. Preparation of the free acid derivative 28 (i) K2CO3, DMF, r.t. o.n., 60 °C 72 h; (ii) β-alanine, AcOH, 100 °C 1 h.

• 

  Figure 9. General synthetic procedure followed for the preparation of 13, 39d, 40e, 41e, 42a-c. (i) Na2CO3, Pd(PPh3)2, THF/H2O, r.t. 8 h, 60 °C 12 h; (ii) secondary amine, DIPEA, DCM, 0 °C to r.t. 20 min; (iii) K2CO3, DMF, r.t. o.n., 60 °C 72 h or NaH, THF, 60 °C, 24 h; (iv) β-alanine, AcOH, 100 °C 1 h.

• 

  Figure 10. General synthetic procedure followed for the preparation of 54, 55. (i) NEt3, THF, r.t. 3 h; (ii) NaOH, EtOH, r.t. 30 min; (iii) pyridine, r.t. o.n.; (iv) K3PO4, Pd(dppf), H2O/DMF, 130 °C 75 min (microwave); (v) AcOH, 120 °C 3 h.

• 

  Figure 11. Secondary screen for inhibition of human norovirus Sydney 2012 RdRp activity. (A) The inhibitory effects of 26 compounds were examined against norovirus RdRp activity using a fluorescent activity assay at 10 µM. Five compounds that exhibited more than 50% inhibition (red bars) were chosen for further assessment with a counter-screen gel-shift RdRp assay. The positive control (blue bar) is the norovirus NNI RdRp inhibitor NIC0214. The mean values of triplicate datasets with standard deviations are shown. (B) A counter-screen-gel-shift assay was used to confirm norovirus RdRp inhibitory activity of the five compounds identified in panel (A). PE44-NoV RNA templates (32 nucleotides) were extended (44 nucleotides) by the RdRp in the absence of any test compounds (0.5% DMSO [vol/vol] negative control) or with test compounds at a fixed concentration of 100 µM. The nucleoside analogue 3′-deoxyguanosine-5′-triphosphate (3′dGTP) was used as a positive control (10 µM) to demonstrate complete inhibition, and no RdRp was used as a negative control. Full-length gels are presented in Supplementary Fig. S3.

• 

  Figure 12. Broad-spectrum activity of hit compounds against calicivirus RdRps and examination of 54 in cell culture. (A) Effects of four compounds were evaluated against calicivirus RdRp activities (norovirus, MNV, RHDV and sapovirus) using an in vitro fluorescent activity assay at a fixed concentration of 100 µM. The mean values of triplicate datasets from at least two independent experiments are shown with standard deviations. (B) Dose-response effects of 54 (0.1–100 µM) were examined against MNV RdRp activity in vitro. The IC50 value and 95% confidence interval (CI) are shown on the graph. The mean values of triplicate datasets are shown with standard deviations. (C) RAW264.7 mouse macrophage cell viability was assessed with 54 (0.5–100 µM). The mean values of triplicate datasets from three independent experiments are shown with standard deviations. (D) The effects of 54 (0.5–50 µM) were assessed against MNV plaque formation in a plaque reduction assay. The mean values of triplicate datasets are shown with standard deviations.