The kinase activity of DPak3 is required for PLS invasion. (A) Stage 15 wild-type (a and e, wt) or dpak3^zyg mutant (b–d) embryos carrying indicated transgenes were labeled with α-MHC. V5-DPak3^K322A expression at 25°C failed to rescue the fusion defect in dpak3^zyg mutant embryos (b) and at 30°C resulted in a more severe fusion defect (c). DPak3^K322A-V5 expression at 25°C enhanced the fusion defect in dpak3^zyg mutant embryos (d, compare with b), and resulted in a minor fusion defect in wild-type embryos (e, compare with a). Arrowheads indicate randomly selected unfused FCMs. (B) Quantification of the fusion defects in the genotypes shown in A. Statistical significance was determined by unpaired student’s t test (***, P < 0.001). Error bars: standard deviations. (C) Localization of overexpressed DPak3^K322A to muscle cell contact sites. Stage 14 embryos triple labeled with phalloidin (green), α-V5 (red), and α-Duf (blue). V5-DPak3^K322A localized to the muscle cell contact sites (arrowheads) indicated by F-actin foci and Duf enrichment at 25°C (a), and accumulated at a higher level to these sites at 30°C (b). The C-terminally tagged DPak3^K322A-V5 showed an even higher accumulation to these sites at 25°C (c) than V5-DPak3^K322A at 30°C (b). All images were acquired by the same confocal settings. Note that the F-actin foci did not cause V-shaped curvatures in the founder cell membranes marked by Duf enrichment. Bars: (A) 25 µm; (C) 5 µm.