Extended Data Fig. 3: A naturally occurring variant in LXRα exerts dominant negative effects in cultured hepatocytes.

A: To demonstrate dominant negative effects of LXRα^W443R in a relevant cell type we generated HepG2 cell lines stably transfected with a Doxycycline-inducible transgene expressing either a wild-type LXRα isoform (WT) or a mutant LXRα^W443R (MT or W443R). The experimental paradigm is summarized in (A), 24 hours after seeding cells were treated with doxycycline to induce LXRα expression, 24 hours later LXR agonist GW3965 (1000 nM) was added and cells were lysed 24 hours later. B: Expression of NR1H3 mRNA (encoding LXRα) in the paradigm described in (A). N = 5 independent experiments. Two-sided P values are displayed from 3-way ANOVA with post hoc Holm-Šídak test. Error bars represent mean ± SEM C: Representative immunofluorescent images of non-transduced HepG2 cells and WT or MT/W443R HepG2 cells treated with doxycycline stained for Beta-Tubulin, LXRα and DAPI. Enhanced LXRα nuclear immunoreactivity can be seen in the transduced cells consistent with induction of LXRα. This experiment was performed twice with consistent results D: Heatmap illustrating expression of GW3965 regulated genes in an RNA-seq experiment after treatment of cells in the paradigm outlined in (A). Ligand treated cells with overexpression of LXRα^W443R cluster with non-ligand treated conditions, indicative of the inhibitory effects of LXRα^W443R on LXR-dependent signalling. Furthermore, induction of canonical LXR-regulated genes, including FADS1, SCD, ANGPTL3, LPCAT3, FASN, MYLIP and SREBF1, is impaired by induction of LXRα^W443R, consistent with dominant negative actions of LXRα^W443R. N = 3 independent experiments. E: We further validated the dominant negative actions of LXRα^W443R in hepatocytes in an LDL uptake paradigm. Following the experimental paradigm illustrated in (A), cells were treated with Dil-labelled LDL for 4 hours before cells were lysed and fluorescence measured as an index of LDL uptake. LXRα^W443R induction was capable of increasing LDL uptake in the basal state, counteracting the previously described effects of LXRα in this setting. Error bars represent mean ± SEM. Analysis was conducted by three-way ANOVA with post hoc testing undertaken using the Holm-Šídak method. *P = 0.01 Induced vehicle treated wildype versus Induced vehicle treated W443R, P = 0.02 Not-induced vehicle treated W443R versus Induced W443R Vehicle treated, N = 4 independent experiments. Figure 3a was created with Biorender.

Source data