PCTP-like START domains are conserved in Arabidopsis and rice. (a) Branch of a neighbor-joining phylogenetic tree that was constructed assuming the Poisson correction model and a complete deletion algorithm (bootstrapped 2,000 replicates). Six START domains from Arabidopsis (green lettering), and four from rice (white lettering in green boxes) are similar to human PCTP (white lettering in red box). (b) Alignment of the 10 PCTP-like START domains from Arabidopsis and rice against the START domain of human PCTP (highlighted in red). Yellow highlighting indicates PCTP residues contacting the sn-1 or sn-2 acyl chains or the glycerol-3-phosphorylcholine headgroup of phosphatidylcholine in any of the three crystallized structures of PCTP (PDB ID: 1LN1, 1LN2, 1LN3) , and also points to plant residues that are predicted to be involved in contact with bound ligand. Additional amino acids that are similarly conserved in PCTP and plant PCTP-like START domains are indicated by gray shading. (c-f) RIBBONS drawing of the START domain from Arabidopsis protein At3g13062 generated by homology modeling. Amino and carboxy termini and secondary structural elements (α helices, red; β sheets, gold) are indicated. The START domain contains a hydrophobic tunnel that extends the length of the protein with openings at both ends. The backbone of the structure shown overlaps with the X-ray crystal structure of PCTP . (c) A nine-stranded antiparallel β sheet and two α helices (α2 and α3) form an interior hydrophobic chamber that can accommodate a single ligand molecule. (d) Top of the hydrophobic tunnel with a clear view of α4, the carboxy-terminal α helix comprising the lid. α4 contains a conserved kink at glycine 280. (e) Lateral view showing the basket structure formed by the antiparallel β sheets β4, β5, and β6 on one end of the hydrophobic cavity while β1, β2, β3, β7, β8, and β9 form the other side. (f) View through the empty START domain cavity, showing that the amino-terminal α1 helix is not predicted to contact the ligand within the hydrophobic interior.
Schrick et al. Genome Biology 2004 5:R41 doi:10.1186/gb-2004-5-6-r41