Fig. 5: Contactin 1–neurofascin 155 expression mediates cell co-clustering.

a Representative cell clustering images of K562 cells expressing contactin 1 (mCherry; magenta; CNTN1) and neurofascin 155 (GFP; green; NF155). Wildtype contactin 1–neurofascin 155 co-clusters form when contactin 1 is expressed in the presence of kifunensine (Kif; 10 µM). Neurofascin 155 but not contactin 1-expressing cells exhibit homophilic clustering. Mutation of the competitive interface residues on either protein abolishes co-clustering. Each experiment was repeated three times independently with similar results. Scale bar, 100 µm. Contactin 1 Phe177Asp, Phe180Asp and Phe212Asp is CNTN1^Mut, neurofascin 155 Phe168Asp, Met170Asp, Met174Asp and Ile217Asp is NF155^Mut. b Location of the mutations. Contactin 1–neurofascin 155 interface residues, shown in stick representation, at the “bottom side” of the Ig2 super β-sheet (left panel) are mutated to aspartates and prevent heterophilic trans interactions. The same neurofascin 155 residues are also located in the neurofascin 155 homodimer interface (right panel) and mutating them to aspartates prevents homophilic trans interactions. c Clustering index; the proportion of the total segmented cell area classified as clusters. p values are: 0.9544 for NF155 vs. CNTN1mut + Kif/NF155, <0.0001 for CNTN1 + Kif/NF155 vs. CNTN1 + Kif/NF155mut, 0.0003 for CNTN1 + Kif/NF155 vs. CNTN1mut + Kif/NF155 and <0.0001 for CNTN1 + Kif/NF155 vs. CNTN1mut + Kif/NF155mut. d Co-clustering ratio determined as the mean mCherry signal over the mean GFP signal per cluster. The increase of this ratio for contactin 1 (kifunensine-treated wildtype)–neurofascin 155 (wildtype) indicates co-clustering, whereas the low ratio values for the other conditions indicate presence of only neurofascin 155 homophilic clusters. Single data points in c and d represent the average values from the n = 15 (n = 14 for contactin 1^Mut–neurofascin 155^Mut) images from N = 3 independent experiments (each experiment 5 images). Error bars indicate the mean ± SEM. Statistical significance was determined by performing a one-way ANOVA followed by a Tukey’s multiple comparison test, and results are indicated using the following conventions: n.s. not significant, ***p < 0.001. In absence of clusters for analysis “n.a.” is indicated, and “X” denote single-channel conditions in which no fluorescence ratio could be calculated. p value is <0.0001 for CNTN1 + Kif/NF155 vs. CNTN1mut + Kif/NF155. Source data are provided as a Source Data file.