(A) Schematic drawing comparing the molecular architecture of human MEGF10 and CED-1. MEGF10 and CED-1 share a highly similar structural pattern. Structural predictions are based on sequence analysis by the Geneworks software and on the SMART website (http://smart.embl-heidelberg.de/smart/show_motifs.pl). The cysteine-rich domains in both proteins were analyzed in term of sequential conservation of EGF-like domains, either 6-Cys EGF-like domains (white pentagons) or laminin-type EGF-like domains (gray pentagons). The Emilin domains [20] (EMI rectangles), the predicted transmembrane segments (white rectangles) and the NPxY motif (black crosses) implicated in the molecular interactions with GULP/CED-6, are also indicated. For comparison a schematic illustration of the predicted structure of LRP-1 is shown; low density lipoprotein receptor domains are indicated by hexagons, in brackets is indicated the number of repetitions. (B) MEGF10 expression pattern. Northern blot analysis of MEGF10 transcript (M10: 7.3 Kb) in RNA samples prepared from adult mouse tissues (H: Heart, A: Adrenals, K: Kidney, Li: Liver, B: Brain, U: Uterus, Lu: Lung), whole mouse embryos (Embryonic day 7, 8, 10, 13, 14 and 15) and a panel of cell lines (Neuro 2a – mouse neuroblastoma, HepG2 – human hepatoma, P388D1, Raw 264.7 and U937 – mouse and human macrophages respectively). Membranes were simultaneously hybridized with an actin probe as a quantitation control.

(A) Positive cooperation between engulfment receptors. Phagocytosis assays were performed on HeLa cells transfected with the indicated molecules. Phagocytic Indexes (PI)±SEM, and number of independent experiments (n) are as follows : Mock: Mock transfected cells PI = 0.46±0.06, n = 11, CD36 PI = 2.83±0.19, n = 6, LRP: LRP1 PI = 1.75±0.06, n = 4, A1: ABCA1 PI = 2.87±0.11, n = 13, M10: MEGF10 PI = 3.15±0.12, n = 11, A1+M10: ABCA1+MEGF10 PI = 4.00±0.14, n = 11, M10+CD36: MEGF10+CD36 PI = 4.22±0.04, n = 3, A1+CD36: ABCA1+CD36 PI = 3.16±0.30, n = 6, A1+LRP: ABCA1+LRP1 PI = 2.82±0.14, n = 2. Coexpressions of receptors were performed by transfection of equal quantities of plasmid cDNA of each species. ECFP or EYFP fusion proteins were used and visualized by confocal microscopy. (B) The inactive ABCA1MM mutant hampers the ability of MEGF10 to promote engulfment. Coexpression of MEGF10 and ABCA1MM (A1MM+M10) reverts the phagocytic index to background levels. The reversion is specific since the coexpression of MEGF10 with ABCA7 (A7) or its inactive mutant (A7MM) is devoid of significant effect on the phagocytic index. Phagocytosis assays were performed on HeLa cells transfected with the indicated molecules. A1MM+M10: ABCA1MM+MEGF10 PI = 1.09±0.10, n = 7, A1MM: ABCA1MM PI = 0.71±0.05, n = 13, A7: ABCA7 PI = 2.90±0.26, n = 7, A7+M10: ABCA7+MEGF10 PI = 3.57±0.37, n = 5, A7MM+M10: ABCA7MM+MEGF10 PI = 2.28±0.27, n = 7, A7MM: ABCA7MM PI = 1.1±0.15, n = 5. ECFP or EYFP fusion proteins were used and visualized by confocal microscopy. (C) ABCA1MM does not affect the trafficking of MEGF10 to the cell surface. Western blot of lysates from cells expressing MEGF10 and ABCA1 or its inactive mutant ABCA1MM were probed with an anti-GFP antibody (upper panel) and compared to blots of biotinylated proteins (lower panel). Mock transfected cells are compared to cells transfected with ABCA1 EYFP (A1), ABCA1MM EYFP (A1MM), MEGF10 ECFP (M10), or the indicated combinations. (D) Molecular competition between wild type and mutant transporter underlies the ABCA1MM effect. ABCA1MM expression levels titrate the ABCA1/MEGF10 cooperation during engulfment. Various ratios of ABCA1/ABCA1MM cDNA (5/0; 4/1; 2.5/2.5; 1/4 and 0/5 from the left to the right) were cotransfected with fixed amounts of MEGF10 and phagocytic assays performed on transfected cells. The phagocytic index is expressed as average of 3 independent experiments±SEM. Values are from the left to the right: MEGF10: 2.81±0.10; 3.99±0.31; 3.03±0.16; 2.00±0.26; 1.02±0.09; 0.89±0.09.

(A) MEGF10 is expressed at the cell surface and clusters around cell corpses during engulfment. Confocal optical X-Y sections show the localization of MEGF10 EYFP (M10Y - pseudo color green) in transfected HeLa cells challenged with 7-AAD labelled apoptotic thymocytes (pseudo color blue). The white arrow indicates the location of the X-Z plan shown on the right. The surface localization was confirmed by the analysis of surface biotinylation as shown in the insert. Western blot of total cell lysates probed with an anti-GFP antibody (TCL, left panel) are compared to blots of biotinylated proteins (BP, right panel). Mock transfected cells (mock), and MEGF10 EYFP transfected (M10) HeLa cells. (B) The expression of MEGF10 enhances the phagocytic ability of HeLa Cells. In vitro phagocytosis assays were carried out on HeLa cells transfected with the indicated engulfment receptors. Results are expressed as distribution histograms. Percentages of cells, scored on ≥100 transfected cells, are plotted against the number of tethered/ engulfed apoptotic thymocytes. The phagocytic index, computed out of at least 4 individual experiments, is indicated in brackets. Mock: mock transfected cells; M10: MEGF10; A1: ABCA1; LRP: LRP-1. (C) MEGF10 can interact with GULP as assessed by GST pull down. Left panel: lysates of HeLa cells transfected with MEGF10 EYFP were incubated with bacterially produced GST or GST-GULP fusion protein before fractionation on SDS-PAGE and blotting onto nitrocellulose membrane. Total proteins were visualized by Ponceau S staining (lower panel) whereas MEGF10 binding to GULP was detected by hybridization with an anti-GFP antibody (upper panel –TCL: Total cell lysate). Right panel: GULP and MEGF10 interact via the NPxY motif since its mutation to APxA abrogates binding (upper panel - NPxY: MEGF10 EYFP-NPxY, APxA: MEGF10 EYFP-APxA). Equivalent amounts of bacterially produced GST-GULP (middle panel) were incubated with equivalent amounts of MEGF10 expressed in transfected cells (lower panel). MEGF10 detection was performed by probing with an anti-GFP antibody on total cell lysate (lower panel) or on pulled down samples (upper panel).

(A) MEGF10 and ABCA1 distribute differentially during engulfment. X-Y confocal sections of transfected HeLa cells evidence that during the engulfment of apoptotic thymocytes (7-AAD in pseudo color blue) MEGF10 (M10: MEGF10 ECFP in green) is recruited at the bottom of the forming phagocytic cup (arrows) while ABCA1 stays at its edges (A1: ABCA1 EYFP in red). X-Z sections of the regions highlighted by the red line confirm the spatial distribution of MEGF10 in the core of the phagocytic cup. (B) Molecular proximity between ABCA1 and MEGF10 assessed by FRET. Imaging FRET analysis of cells expressing simultaneously ABCA1 EYFP and MEGF10 ECFP indicates that the C terminal tailpieces lie at a distance of <100 Å. Transfer efficiency is insensitive to acceptor intensity (YFP RFI – left panel) but sensitive to acceptor: donor ratios (right panel). (C) MEGF10 hampers the oligomerization of ABCA1. Native PAGE profile of lysates of HeLa cells transfected with equivalent amounts of MEGF10 and either ABCA1 (A1 upper panel) or its mutant ABCA1MM (MM lower panel). The immunodetection with anti-GFP and anti-ABCA1 antibodies reveals no evidence of comigration of the two proteins in a single complex in the case of transfection with the wild type functional transporter. However the presence of MEGF10 greatly destabilises the pattern of oligomerization linked to the expression of ABCA1MM. (D) Titration of the transdominant negative effect of ABCA1MM on MEGF10. Cells were transfected with various ratios of ABCA1MM and MEGF10 as shown by the SDS-PAGE pattern (lower panel) and analyzed by phagocytic assays and native PAGE (not shown). Increased ratios of MEGF10 over ABCA1MM lead to increased destabilization of ABCA1MM complexes (not shown) and rescue of phagocytic competence. Phagocytic indexes (averaged from 3 independent experiments)±SEM are from the left to the right: 0.66±0.16, 1.04±0.11, 1.23±0.27, 1.75±0.02, 2.47±0.19, 2.82±0.13, 3.14±0.30.