LICs 1 and 2 act redundantly in Golgi apparatus positioning, with helix 1 being essential. (A–C) HeLaM cells were depleted of LIC1, LIC2, or both LICs using 5 nM siRNA for each subunit, then analyzed by immunoblotting of lysates with antibodies to LIC1 and 2 (A) or fixed and labeled with antibodies to GM130 and imaged on a DeltaVision microscope to reveal the Golgi apparatus (B). Z-projections of deconvolved images stacks are shown. Quantitation of knock-down efficiency is shown in Fig. S2 C. (C) Golgi morphology was scored manually for 100 cells per condition in three independent experiments. Mean ± SD values are plotted. Statistical analysis was performed using multinomial logistic regression (see Table S2 for full results): comparisons vs. control samples are shown on the graph (P ≤ 0.0001 = ****). (D and E) HeLaM cells were depleted of both LICs and then transfected with RNAi-resistant LIC1-mKate or LIC2-mKate (D) or GFP-LIC1-FL, GFP-LIC1-CT2, GFP-LIC1-CT3, and GFP constructs (E). Cells treated with control siRNAs and transfected with GFP were used as controls. GM130 labeling was used to reveal Golgi apparatus morphology (wide-field images: asterisks mark cells expressing the constructs). Golgi morphology was scored for ∼100 cells per experimental condition in three independent experiments (E). Mean ± SD values are plotted. GFP and GFP-LIC1-CT3 rescue data are significantly different to control knockdown (P ≤ 0.0001), whereas rescue with GFP-LIC1-FL or GFP-LIC1-CT2 is not (multinomial logistic regression, see Table S2). All scale bars = 10 µm. Source data are available for this figure: SourceData F5.