SARS-CoV-2 Inhibitors Identified by Phenotypic Analysis of a Collection of Viral RNA-Binding Molecules

Figures

• 

  Figure 1. Phenotypic strategy used for identifying RNA-binding small molecules with anti-SCoV2 activity.

• 

  Figure 2. Chemical structure, SCoV2 antiviral activity, and toxicity of the best three RNA-binding SCoV2 inhibitors selected from the phenotypic study. (A) Chemical structure and antiviral and toxicity data of clomiphene citrate, trim, and qz2, obtained in Vero E6 and human 293T-ACE2 cells. The dose–response graphs correspond to 293T-ACE2 experiments. (B) Virus titers determined in supernatants by TCID50 at 48 h post-infection, with DAPI cell count from the well indicated. The limit of detection for viral titers is indicated with a dotted line. (C) Time-of-addition assays analyzing the antiviral action of clomiphene citrate, trim, and qz2. The data were normalized to the mean of DMSO-treated cells at each time point, and remdesivir was used as a positive control.

• 

  Figure 3. SCoV2 genome and viral RNA elements analyzed in this study. (A) Secondary structure representation of the SCoV2 genome, indicating the domains examined as possible targets for clomiphene citrate, trim, and qz2. Blue rectangles indicate secondary structure domains analyzed with ligand-based NMR experiments. Red rectangles mark domains studied with ligand-based and RNA-based NMR experiments. The orange rectangle around the 3_S2m hairpin indicates that this domain was additionally analyzed with fluorescence experiments. The small ATTH domain was investigated with RNA-based NMR experiments only. (B) Secondary structure of the SCoV2 RNA constructs evaluated in this report: 32-nt 5_SL2+3 domain, 124-nt full-length PRF domain, 56-nt PRF ATTL subdomain, 26-nt PRF ATTH hairpin, 176-nt ORF7b segment, 115-nt 3_SL1+2 domain, and 45-nt 3_SL2m stem–loop. The blue-colored nt mark changes relative to the wild-type sequence introduced to increase transcription yield. The viral nt encompassed by each domain are indicated using genomic numbering.

• 

  Figure 4. Interaction of clomiphene citrate, qz2 and trim with SCoV2 RNA elements analyzed by ligand-based NMR spectroscopy experiments. (A) Representative 1H one-dimensional (1D) CPMG, and wLOGSY spectra of trim in the absence (red) and presence (blue) of 5_SL2+3. RNA binding was assessed by quantifying chemical shift perturbation (CSP) or line broadening (LB) in 1D spectra, peak area variations in CPMG spectra, and LOGSY factors in wLOGSY spectra. (B) Quantification of CPMG and wLOGSY perturbations of aromatic protons as a function of compound and RNA element. RNA elements are ordered according to their relative locations in the virus genome. For RNA targets of similar size, increased binding translates into greater variations of CPMG areas and greater LOGSY factors. Note that both CPMG and LOGSY perturbations intensify with RNA target size. To account for this effect, the CPMG and LOGSY bars are colored according to the size of each RNA element as depicted in the image. The error bars represent the standard deviation of the average perturbation detected for the aromatic protons of each ligand. Conditions for (A,B): 300 μM compound; 3 μM RNA (1:100 molar ratio; clomiphene) or 6 μM RNA…

• 

  Figure 5. Recognition of the SCoV2 3’UTR S2m stem–loop by trim. (A) Interaction between 3_S2m and trim monitored by NMR spectroscopy. The pyrimidine H5-H6 region of the TOCSY spectrum (60 ms mixing time, 25 °C) of 3_S2m (blue) is superposed on the spectra of complexes with increasing RNA–ligand molar ratios, color-coded as indicated in the graph. On the right, the location of the ligand binding sites in the 3_S2m hairpin are indicated by highlighting the nt whose protons underwent broadening or chemical shift variations greater than two (orange) and three (red) standard deviations from the mean perturbation (0.028 and 0.042 ppm) upon the addition of four equivalents of ligand. To map the binding sites, we monitored the chemical shift variations of 63 aromatic, H1’, and H2’ 3_S2m RNA resonances with TOCSY and NOESY experiments. Conditions: 120 μM 3_S2m, 50 mM KCl, 25 °C. (B) Interaction between 3_S2m-23fl and trim monitored by fluorescence intensity. The experiments were carried out in the absence (blue) and presence (red) of a 100-fold molar excess of unlabeled competitor tRNA. The error bars represent standard deviations of three independent experiments. Conditions: 100 nM 3_S2m-23fl, 150…