Fig 2: C9ORF72 localizes in part to late phagosomes and phagolysosomes.(A) Immunofluorescence of C9ORF72 (GTX632041, magenta) and F-actin staining (cyan) in human MDMs. The inset is a higher magnification of the region indicated by the arrow and shows a grayscale image of C9ORF72. Cell is outlined with a white dashed line. Scale bar = 5 µm (full size) and 2 µm (inset). (B) Quantitative immunoblot of RAW264.7 cell lysates treated with 1 µg/ml LPS for the indicated time points. The total protein stained transfer is shown as loading control. C9ORF72 is detected using GTX634482. The C9ORF72 protein signal as a ratio to total protein was determined, normalized to RAW264.7 lysates at time 0 (red), and presented as fold change (blue numbers). C9ORF72 KO MEFs are included as a specificity control. (C) Quantification of C9ORF72 intensity from immunoblots performed as in (B) at 0 hr and 24 hr time points, presented as normalized values. Four individual experiments were quantified. *=p value<0.05 (D) Pulse-chase (P–C) experiments in LPS-treated RAW264.7 cells using phagocytosed beads. Beads were incubated for a 30 min pulse followed by a 0 min (30-0), 30 min (30-30) or 90 min (30-90) chase. Brightfield imaging reveals the phagocytosed beads. Staining of C9ORF72 (GTX632041; magenta), LAMP1 (yellow) and F-actin (cyan) were performed and imaged using confocal microscopy together with LIGHTNING image processing as described previously. Single focal plane images are shown. Insets are higher magnification views of the internalized beads indicated with arrowheads in the large images and show the merged channel together with grayscale images of F-actin, LAMP1 and C9ORF72. White arrows point toward C9ORF72 and LAMP1 co-localization. All images are the result of LIGHTNING applied to the confocal image. Cells are outlined with white dashed lines. Scale bar = 8 µm (full size) and 2 µm (inset).

Fig 3: Analysis of 16 C9ORF72 antibodies by immunofluorescence.(A) Parental and KO cells were transfected with LAMP1-YFP or LAMP1-RFP, respectively. Parental and KO cells were combined and incubated with buffer only or stained with the indicated GeneTex C9ORF72 monoclonal antibodies. Grayscale images of the green, red and far-red channels are shown. Parental and KO cells are outlined with green and red dashed line, respectively. Representative images are shown. Bars = 40 µm. (B) Parental and KO cells were prepared as in (A) and stained with the indicated C9ORF72 antibodies. Grayscale images of the far-red channel are shown. Parental and KO cells are outlined with green and red dashed line, respectively. Representative images are shown. Bars = 40 µm.

Fig 4: RELCH, OSBP, and Rab11 depletion results in less cholesterol accumulation in the TGN and ER. (A and B) The homogenates from the shRNA-expressing HeLa cells were fractionated using a Histodenz step density gradient. (A) The fractions were immunoblotted (IB) with antibodies against TGN46, calnexin, and Lamp2. (B) Percentages of cholesterol (µg/mg protein) in the TGN (fractions 4–6), LEs/lysosomes (LE/LY; 7–9), or ER (10 and 11) in the total fractions are shown in the bar graph. (C) HeLa cells were treated with 2 µg/ml U18666A for 16 h and stained with Filipin. Bar, 20 µm. (D–G) Immunoisolation of ER, TGN, and LE/lysosome from the PNS derived from the U18666A-treated (D and E) or RELCH-, Rab11a/b-, and OSBP-depleted cells by shRNAs (F and G). ER, TGN, and LE/lysosome membranes were isolated using calnexin, TGN46, and Lamp1 antibodies, respectively. (D and F) The isolated samples were immunoblotted with calnexin, TGN46, and Lamp2 antibodies. (E and G) Quantification of the cholesterol content in the isolated membranes (relative to the control samples). Data are expressed as means ± SEM from at least three independent experiments. *, P < 0.05; **, P < 0.01; ***, P < 0.001 (relative to the control; two tailed Student’s t tests). Representative data from three independent experiments are shown in A, D, and F.