Fig. 3. Ascidian aPKCs lack a conserved cyteine-rich domain. (A) Domain structure of aPKCs. PB1, Phox and Bem protein interaction domain which binds PAR-6 protein; Kinase, serine threonine kinase domain; C1, cysteine-rich domain which binds InsPtd(3,4,5)P₃. (B) Comparison of aPKC proteins from four ascidian species (Phallusia mammillata, Pm; Ciona intestinalis, Ci; Ciona savigny, Cs; and Halocynthia roretzi, Hr) with aPKCs from human (Hu), frog (Xl), fly (Dm), and worm (Ce). The C1 domain usually present in aPKC but lacking in ascidian aPKCs is underlined. The two boxed areas indicate a normally contiguous region of homology specific to ascidian aPKCs. PmaPKC was obtained by cloning from a cDNA library (this study, GenBank AY987397). CiaPKC was obtained by assembling sequence information from Sasakura et al. (Sasakura et al., 2003) and CLSTR05328r1 from the Ghost genome browser (ghost.zool.kyoto-u.ac.jp). CsaPKC was obtained by analysis of genomic sequence information in http://www2.bioinformatics.tll.org.sg:8082/Ciona_savignyi/ and http://www.broad.mit.edu/annotation/ciona/. HraPKC sequence was obtained from 5' EST data from the Magest genome site accessible via Aniseed (aniseed-ibdm.univ-mrs.fr). The HraPKC protein appears truncated in the kinase domain because the sequence encoding the C-terminal half of the protein is not available.