Rho GTPase is required for Rho-kinase localization and planar polarity in intercalating cells. (A–F) Localization of HA:Rho, Venus:Rok^K116A, and Myo:GFP in stage 7 embryos expressing wild-type Rho (Rho WT; A–C) or dominant-negative Rho (Rho^N19; D–F). (G and H) Localization of E-cadherin and Baz/Par-3 in stage 7 embryos expressing wild-type Rho (G) or Rho^N19 (H). Anterior is left, and ventral is down. (I–L) Localization of Venus:Rok^K116A (I and J) or E-cadherin (K and L) in stage 7 embryos expressing wild-type Rho (I and K) or Rho^N19 (J and L). Cross sections are shown, with apical up. (M) Planar polarized enrichment of Rho-kinase and myosin II at AP cell boundaries (75–90° with respect to the AP axis) relative to DV cell boundaries (0–15°). Rho-kinase and myosin II were significantly less planar polarized in embryos expressing Rho^N19 compared with embryos expressing wild-type Rho (P < 0.01). (N) Rok RB:PH was significantly more enriched at AP cell boundaries relative to DV cell boundaries than total Rho protein (P < 0.001). No planar polarized enrichment was detected for total Rho, GFP:Rho, or the PKN Rho probe. (O) Localization of total Rho protein (red) and the PKN Rho probe (green) at different z planes (0 µm is at the level of adherens junctions, and −3 µm is basal to the junctions). Anterior is left, and ventral is down. (P) Total Rho and the PKN Rho probe cortical levels normalized to the apical junctional intensity. PKN Rho probe signal at the basolateral membrane was reduced compared with total Rho (P < 0.002 at −0.75 µm; P < 0.0001 at −1.5, −2.75, and −3.0 µm). A single value was obtained for each image by averaging 100–200 edges/image; 4–16 images in 4–8 embryos were analyzed for each genotype at each z plane. Means ± SEM between images are shown. Bars, 10 µm.