TJ membrane tortuosity is regulated by CGN interaction with NM2B. (Related to Figs. 8 and 9.) (A and D) IF microscopy localization of endogenous occludin and GFP-tagged exogenous CGN constructs (A) and quantification of ZI (D) in CGN-KO MDCK cells. Cells were rescued with either full-length (FL) CGN (top panel), or CGN lacking the NM2BR (middle panel), or GFP alone (bottom panel, negative control). Lines in occludin channel trace membranes to highlight tortuosity. Data are represented as mean ± SD (n = 67–80 from three independent experiments). Statistical significance was determined by unpaired Mann–Whitney test, ***P < 0.001. (B, C, and E) IF microscopy localization (B and C) of endogenous occludin in either CGN-KO (B) or WT (C) MDCK cells overexpressing either HA-tagged CGN (top panels), or HA-tagged ZO-1 (middle panels), or HA-CFP (bottom panels, negative control) and quantification of ZI (E). Dots in E show replicates (n = 45 junctional segments from three independent experiments), and data are represented as mean ± SD. Statistical significance was determined by unpaired Mann–Whitney test, ****P < 0.0001. Scale bars (A–C), 10 μm. (F) IB analysis of the expression of endogenous and exogenous CGN constructs (either CGN-FL orCGN-Δ1003-1190) either in WT MDCK cells, or CGN-KO cells, or CGN-KO cells inducibly rescued with the indicated constructs for the measurement of apical membrane stiffness (Fig. 9). Source data are available for this figure: SourceData FS7.