Fig. 3: Antisense targeting of the SARS-CoV-2 FSE.
a, ASOs modified with LNA bases that were designed against the SARS-CoV-2 FSE RNA secondary structure (colored lines). Ambiguous base-pairing partners are indicated with dashed lines. The C13533A mutation between SARS-CoV-1 and SARS-CoV-2 is shown in red. b, Mean in vitro dual luciferase frameshifting assay of initial, pre-structure informed LNAs, as a function of LNA challenge, as shown as percent frameshifting ± standard deviation. Values are normalized as described in the Methods. Fits are derived from a standard binding isotherm fit to the data (Methods) to estimate IC50 values versus SARS-CoV-2 (±standard error in stated units; Methods): S2D 1.9 ± 1.6 µM; S3D-1 500 ± 90 nM; S3D-2 180 ± 20 nM. c,d, FSE-LNA inhibition of SARS-CoV-2-nLuc virus replication in Huh-7 (c) and Vero E6 (d). Cells were treated with 25 nM of FSE-directed or Scrambled (Scr) LNAs 24 h before infection with SARS-CoV-2-nLuc reporter virus. At 48 h post-infection, luciferase expression was measured. The nucleoside analog, EIDD The nucleoside analog, EIDD-1931 (EIDD), was used as a positive control. Results are shown as log10 luciferase expression; n = 2. e, Mean in vitro dual luciferase frameshifting assay of initial, post-structure informed LNAs, as a function of LNA challenge, as shown as percent frameshifting ± standard deviation. Fits are derived as in b. Slp2 320 ± 40 nM; S1D-3 280 ± 90 nM; S1D-1 130 ± 20 nM; S1D-2 280 ± 70 nM. f, FSE-LNA inhibition of SARS-CoV-2-nLuc virus replication in ACE2-A549 cells. LNA treatment and infection were performed as described in c and d with 25 nM and 100 nM LNA concentrations. Results are shown as log10 luciferase expression, n = 4. P values were generated by GraphPad Prism software and computed as an ordinary one-way analysis of variance (ANOVA) using Dunnett’s multiple comparisons test against the Scr. LNA mean. Error bars represent ± s.d. *P < 0.05, **P < 0.01, ***P < 0.001, ****P < 0.0001. NS, not significant.
