Fig. 1.

PPIP5K2 but not PPIP5K1 contains a candidate nuclear localization sequence in a highly-conserved context. In Panel a, ClustalW was used to generate an alignment of human PPIP5K2 [Swiss-Prot: O43314] with homologues from the mouse, Mus musculus [GenBank: XP_006529559]; big brown bat, Eptesicus fuscus [Genbank: XP_008142724); minke whale, Balaenoptera acutorostrata scammoni [Genbank: XP_007191888.1); ghostshark, Callorhinchus milii [GenBank: XP_007891317]; chicken, Gallus gallus [GenBank: XP_424859]; frog, Xenopus tropicalis [Xenbase: Xetro.A00468.1); king cobra, Ophiophagus hannah [GenBank: ETE65579.1]. The candidate nuclear localization sequence (NLS) is boxed. Related amino-acids are grouped into classes - A/G, D/E, F/Y, I/L/V/M, N/Q and S/T - and then color coded based upon their relative prevalence at each position (orange, most prevalent; green, second most prevalent; blue, third most prevalent). Where two groups of amino acids are equally prevalent, priority defaults to alphabetical order. Using these same amino-acid classes, panel b depicts the percentage similarities of aligned regions of hPPIP5K2 and hPPIP5K1 [Swiss-Prot: Q6PFW1]; the proteins can be divided into regions that are either >85% similar (coloured blue) or no more than 30% similar (colored pink). The graphics also depict the kinase domains and the PBD (PtdIns(3,4,5)P₃-binding domains [22]) that are present in both hPPIP5K1 and hPPIP5K2. The position of the 63-residue domain in hPPIP5K2 that Contains Penta-Arginine (CPA) is also indicated; note how it is more conserved than the regions immediately flanking it. PPIP5K1 does not contain sequence that is homologous to the CPA domain, and so is also missing both the penta-Arg NLS and the adjacent Ser residue that is phosphorylated (see text for details)