Fig. 2. Alignment of the three class V myosins in vertebrates. Bolded letters denote identical residues in all three sequences. The head, ATP binding site and globular tail are marked. Solid boxes mark each of six IQ motifs in the neck domain. Regions predicted (Lupas et al., 1991) to form a coiled-coil with greater than 90% probability are denoted by dashed underlines. Carets indicate the boundaries of Myo5a exons A through F that have been mapped by PCR (Lambert et al., 1998; Seperack et al., 1995), and shading indicates the three exons known to undergo tissue-specific alternative splicing. Note that the human Myo5a sequence reported to bind the dynein light chain (dotted bracket) does not contain exons D and F (Naisbitt et al., 2000). The sequence corresponding to the mouse Myo5a construct reported to bind kinesin (Huang et al., 1999) is also indicated (solid bracket). The position of a conserved serine in the globular tail (`serine 1650' for mouse Myo5a), whose phosphorylation regulates binding of Myo5a to melanosomes, is marked by a diamond (Karcher et al., 2001). Rab11a-binding by rabbit Myo5b is reported to require the sequence corresponding to aa 1397-1418 and 1797-1811 of rat Myo5b (Lapierre et al., 2001). The sequence and location of a putative vacuole-binding site in the globular tail of Myo2p is also shown; asterisks denote functionally crucial residues (Catlett et al., 2000). The GenBank accession number for Myo5c is AF272390.