SNX5, SNX6, and SNX32 interact with the CI-MPR tail via the WLM sorting motif. (A) Consistent with SILAC-based proteomics, GFP-tagged SNX5, SNX6, and SNX32 but not SNX1, SNX2, or retromer immunoprecipitate (IP) CI-MPR in a GFP-trap experiment. Under the blots is a summary of CI-MPR binding ability from quantitative fluorescence-based Western blotting and SILAC-based enrichment and peptide counts. (B) Coexpression of heterodimeric combinations of GFP-tagged and mCherry-tagged SNX1–SNX5 and SNX2–SNX5 in HEK293T cells. GFP trap of GFP-SNX1 and GFP-SNX2 revealed an enhanced association with CI-MPR when the formation of SNX1–SNX5 and SNX2–SNX5 was favored by coexpression of the binding partner SNX5. (C) Summary of constructs used. Retrograde sorting motifs in brackets correspond with the black boxed regions. (D) GFP trap of GFP-tagged cargo tails transiently transfected in HEK293T cells showing that of the tails’ constructs; only CI-MPR pulls down the retromer-linked SNX-BARs but little, if any, of the retromer. SNX30 was used as a negative control. (E) The WLM motif within the tail of CI-MPR is an endosome-to-TGN sorting motif. HeLa cells were cotransfected with CRISPR-Cas9 plasmids against CI-MPR and a puromycin resistance–expressing plasmid before puromycin selection 24 h later. Cells were then transfected with full-length (FL) CI-MPR or full-length CI-MPR WLM-AAA. 48 h after transfection, cells were incubated with an antibody targeting the CI-MPR extracellular domain, and its trafficking was analyzed after 40 min. Zoomed images on right are marked by boxes in main images. Bars: (main images) 20 µm; (zooms) 10 µm. (F) The CI-MPR WLM sorting motif is necessary for interaction with the retromer-linked SNX-BARs. Summary of CI-MPR mutant binding ability from quantitative fluorescence–based Western blotting. SNX30 was used as negative control. n = 3 independent experiments (means ± SEM; one-way ANOVA compared with GFP–CI-MPR tail 1–164. *, P < 0.05; **, P < 0.01; ***, P < 0.001; ****, P < 0.0001).