Fig 1: Split NanoBiT proximity complementation assay. (A) Schematic showing the design of the system. Misfolded membrane protein (ABCA4; blue) is tagged at the cytoplasmic C-terminus with SmBiT part of NanoLuc (dark red) and the LgBiT subunit (light red) is targeted to the plasma membrane by the N-terminus of RP2 (magenta). When misfolded protein is retained in the ER the SmBiT (dark red) and LgBiT (light red) cannot complement (upper panel). Small molecule rescue of ABCA4 plasma membrane traffic enables the SmBiT–LgBiT system to produce luminescence through proximity induced complementation of the NanoLuc (lower panel). (B) WT-RHO fused with SmBiT localises to the plasma membrane. The staining was performed with antibody 4D2 without a permeabilization step, as the 4D2 epitope is located on the extracellular N-terminus the protein. (C) P23H fused to SmBiT staining is observed in a perinuclear and reticular pattern, consistent with the ER, following a detergent permeabilization step to reveal 4D2 immunoreactivity. (D) P23H-SmBiT is not detected with 4D2 using a no-permeabilization protocol. (E) The RP2-LgBiT localised to the plasma membrane, when stained with anti-LgBiT antibody in permeabilized cells. (F) Intracellular WT-ABCA4-SmBiT was detected with the ABCA4 3F4 ABCAM antibody (yellow) and WT-ABCA4-SmBiT localising to the plasma membrane was detected using an Abbexa antibody (magenta) 48 h post-transfection analysis. Scale bars = 10 μm. (G) 24 h post-transfection with RHO-SmBiT + RP2-LgBiT and P23H-SmBiT + RP2-LgBiT plasmids luminescence signal was measured. Raw luminescence values are shown. Error bars are ± SD. n = 3. two-tailed Student’s t-test for RHO vs. P23H. (H) Luminescence signal was analysed 48 h post-transfection with ABCA4 variants + RP2-LgBiT plasmids in live HEK293T cells. Mean of fold change relative to WT ± SD. n = 4 independent experiments. One-way ANOVA and post-hoc analysis comparisons were performed only against WT-ABCA4-SmBit + RP2-LgBit sample. Significant differences are displayed.
Fig 2: Structure of ABCA4 in the unbound state. The structure is represented as a cartoon with the N- and C-terminal halves coloured in silver and gold, respectively. The selected residues are represented as spheres and colour coded. The structure has been modelled in Pymol (The PyMOL Molecular Graphics System, Version 2.5.5 Schrödinger LLC, New York, NY, USA) using the cryo-EM map of substrate-free ABCA4 [42]. The α-ABCA4 Abbexa antibody antigen is highlighted with a black curve.
Fig 3: The effect of AICAR on ABCA4 protein. (A,B) Western blot analysis and respective quantification of 10 µg of protein lysate from HEK293T cells expressing ABCA4 variants showing that AICAR treatment increases the steady state of the ABCA4 protein. n = 3. Error bars are mean of fold change ± SD. Black dashed line represents baseline vehicle value, 1.00. Two-tailed Student’s t-test for treated vs. vehicle. Significant differences are displayed. (C) CHO cells transfected with ABCA4 variants were treated with 0.5 mM AICAR for 24 h. The sub-cellular localisation was analysed by confocal microscopy using two different antibodies recognising ABCA4 intra-(1D4) and extra-(Abbexa) cellular epitopes. Blue arrows indicate vesicles, red arrows indicate ABCA4 at the plasma membrane. Scale bars = 10 μm.
Fig 4: The effect of temperature on ABCA4 protein level and localisation. (A) HEK293T cells were transfected with plasmids expressing WT-ABCA4 and the indicated variants and subsequently incubated at 37 °C or 30 °C. WT ABCA4 and variant protein levels detected by western blot (10 µg of protein lysate). Mock represents a non-transfected control. (B) Quantification of band intensity using ImageJ. Data were normalised to GAPDH reference protein intensity and 37 °C level (dashed line, 1.00). Error bars are mean ± SD. n = 3 independent experiments. Significant differences are displayed. (C) CHO cells transfected with WT-ABCA4 and variants incubated at 37 °C or 30 °C and stained for cell surface accessible ABCA4 (Abbexa) and total ABCA4 (1D4). Confocal images showing the localisation of ABCA4 in vesicle structures (blue arrows) and plasma membrane (red arrows). Cells treated at 30 °C show an increase in the plasma membrane and vesicle-positive cells. (D) ABCA4 vesicle-positive cells and plasma membrane-positive cells were scored among all transfected cells and expressed as a percentage of total transfected cells. n = 10 fields of view (×40 objective). Mean ± SD. Scale bars = 10 μm. Two-tailed Student’s t-test for 37 °C vs. 30 °C.
Fig 5: Missense ABCA4 variants present partially impaired trafficking and maturation in cultured cells. (A) CHO cells were transfected with plasmids expressing WT-ABCA4 and selected mutants. At 48 h post-transfection, cells were double-labelled with the 1D4 antibody to stain the intracellular ABCA4 (yellow) and with the ABCA4 Abbexa antibody (magenta, upper panels; white, lower panels) to stain the membrane-exposed extracellular epitope. Nuclei were stained with DAPI (cyan). WT-ABCA4 was localised both in the ER and at the plasma membrane, whereas missense variants were mainly trapped in the ER. Scale bars = 10 μm. (B) HEK293T cells were transfected with plasmids expressing WT-ABCA4, T983A and R2077W. At 48 h post-transfection, cell lysates were collected, and 6 µg of protein lysate was treated with glycosidases Endo H and PNGase-F, or buffer and temperature-only protocol control (T°). The untreated sample migrated as two species, band A and band B. The treated samples only showed band A. (C) The B/A band ratio was quantified using ImageJ (Version 2.14.0/1.54f). Fold changes relative to WT are shown. T983A and R2077W band ratios were reduced with respect to WT protein. Error bars are mean of fold change ± SD. n = 3 independent experiments. One-way ANOVAs and post-hoc analysis comparisons were performed only against WT ABCA4 sample.
Supplier Page from Abcam for Anti-ABCA4 antibody [3F4]