a Schematic representation of oligonucleotides used to explore the structure–activity relationship for preferential targeting of V3. Measured melting temperatures (Tm) for the oligonucleotide-C9 surrogate RNA heteroduplexes are indicated. b Relative expression of V3 (top) or the ratio of V3 to all variants (bottom) with respect to HPRT in human ALS motor neurons after gymnotic treatment with 10 µM of the indicated oligonucleotides (stereorandom, navy; stereopure, aqua; stereopure with asymmetric wing chemistry, coral; controls, beige). Dotted horizontal line denotes 0.5-fold expression level for V3. Data are presented as mean ± SD, n = 3. One-way ANOVA with Dunnett’s multiple comparison test. ****P < 0.0001, ***P < 0.001, **P < 0.01, ns non-significant. c Relative expression of the indicated C9orf72 transcript (V3, left; All variants, right) with respect to HPRT in human ALS motor neurons treated with increasing concentrations (0.001–10 µM) of C9orf72-630 or C9orf72-631. Lines represent four-parameter least-squares fit. Half-maximal inhibitory concentrations (IC50s) for V3 are indicated (n = 2). d Relative expression of V3 with respect to HPRT in human ALS motor neurons treated with increasing concentrations (3 nM–50 μM) of the indicated oligonucleotide under gymnotic conditions. Lines represent four-parameter non-linear fit of data. Data are presented as mean ± SD, n = 3. ASO1 was toxic to cells at concentrations >10 μM, precluding activity evaluations at high concentrations. Source data, including exact P values, are provided as a Source Data file. NTC non-targeting control; All V All variants, PS phosphorothioate, PO phosphodiester, C cytosine, OMe O-methyl, MOE methoxyethyl, V variant.