Fig 2: Modeled structure of the SRCR domain of human SR-AI and MACRO. a Schematic illustration of human MCRCO and SR-AI. The MARCO domains: cytoplamic domain (cyto), transmembrane domain (TM), spacer domain (Spa), coiled coil domain (coiled coil), triple-helical collagenous domain (collagenous), and cysteine-rich C-terminal domain (SRCR). The predicted N-glycosylation sides and amino acid sequence number was marked above and below the schematic illustration, respectively. b The structure of the SRCR domain of SR-AI and MACRO is composed of six β-strands (β1-6, yellow), one α-helix (α1, red), two 310 helices (η1, η2, blue), turns (indigo), and loops (gray). Disulfide bonds are shown in orange. The conformation folds with three β-strands (β1, β2, β3) at the N-terminal region followed by a short turn, α-helix (α1), and β4. Then, there are three long loops, located between β4 and β5, between β5 and η1, and between η1 and η2, ending with β6. Surface electrostatic potential representation of the SRCR domain of SR-AI and MACRO with basic (blue) and acidic (red) clusters. c Sequence alignment of human SR-AI and MACRO were displayed using ClustalW2 and ESPrip. The β-sheets are shown as arrows, and helixes are shown as a saw tooth pattern. Turns are marked as TT. The three pairs of disulfide bonds are labeled as numbers in green

Fig 3: Identification of the potential N-glycosylation sites of SR-AI. a and b COS 7 cells were transfected with human SR-AI, the single mutation of asparagine to glutamine at residues 82, 102, 143, 184, 221, 249 and 267. Western blot analysis was performed on the total cell lysates before (a) and after (b) treated by PNGase F, and avidin pull-down of biotinylated cell lysates (c). d and e COS-7 cells were transfected with hSR-AI and the double mutation of asparagine (N) to glutamine (Q) residues at 102-143, 102-184 and 143-184. Western blot analysis was performed on the total cell lysates (d) and the PNGase F or Endo H treated cell lysates (e). The result was repeated for three times (N = 3) and the representative blot was shown

Fig 4: Schematic representation shows the SRCR mutations in SR-AI and MARCO. The mutations sites at loop structure are shown in blue and those at α-helix, β-sheet, η-helix and disulfide structure are shown in red. The black line indicates the sequence of SRCR; α-helix is indicated as red spring, and disulfide bonds are shown as green line

Fig 5: Critical residues in SRCR domain of human SR-AI for the surface targeting and N-glycosylation were identified by point mutation. COS-7 cells were transfected with wild type SR-AI and SR-AI variants. a Surface-targeted SR-AI and its variants were detected by live immunostaining (green). Cytosolic SR-A and its variants were detected by immunocytochemistry (red). Nuclei were counterstained with Hoechst 33258 (blue). b Cells were incubated with FAM-labeled oAβ. Representative merged confocal images of surface-targeted SR-AI (red) and internalized FAM-labeled oAβ (green) were shown. Nuclei were counterstained with Hoechst 33258 (blue). Scale bar, 10 μm. The result was repeated for four times (N = 4) and the representative image was shown