(A) Model of the p120-catenin–bound N-cadherin D774–H791 fragment, based on the p120-catenin/E-cadherin structure (PDB: 3L6X), showing an extended conformation with a COOH-terminal 3-10 helix. (B and C) When the phosphomimic E778 is included in molecular dynamic simulations of this p120-catenin/N-cadherin complex, the 3-10 helix that supports hydrophobic interactions with p120-catenin loses multiple stabilizing hydrogen bonds. (D) This loss of hydrogen bonds destabilizes the helix, resulting in more motion and weaker interactions with the corresponding hydrophobic patch on p120-catenin, relative to simulations performed on the WT complex. (E) Representative immunoblot using protein lysates prepared from HEK293 cells transfected with control GFP or full-length WT (GFP-Ncad-WT), phosphomimic (GFP-Ncad-SE) or phosphoablated (GFP-Ncad-SA) N-cadherin constructs indicate similar levels of total RhoA across all groups; GAPDH was used as loading control (n = 3). Data are expressed as mean ± SEM and 1-way ANOVA followed by Fisher’s LSD test was used for statistical evaluation. (F) Measurement of GTP-RhoA levels (i.e., active RhoA) using luminescence-based G-LISA indicated that overexpression of phosphoablated GFP-Ncad-SA in HEK293 cells results in significantly increased RhoA activity compared with GFP and GFP-Ncad-SE; of note, GFP-Ncad-WT behaves similarly to GFP-Ncad-SA; n = 4 experiments performed in duplicate. Data are expressed as mean ± SEM and statistical evaluation was done with 1-way ANOVA followed by a Fisher’s LSD test, *P = 0.0104 for GFP vs. GFP-Ncad-WT, *P = 0.0189 for GFP-Ncad-WT vs. GFP-Ncad-SE, *P = 0.006 for GFP-Ncad-SE vs. GFP-Ncad-SA, and *P = 0.0033 for GFP vs. GFP-Ncad-SA.