Information Technology Reference
In-Depth Information
32-74 in all the SAND C-termini sequences. The residue numbering in this section is taken
from the alignment (Figure 3). This region coincides with the solvent inaccessible D-helix
A2 shown in Figure 3. Further analysis using TopPred2 and MEMSAT2 corroborates the
prediction. A fifth transmembrane prediction algorithm, PredictProtein, does not report a
transmembrane domain for this region. Potential transmembrane regions were noted at
positions 444-464 in SAND1 and 20 residues further towards the C-terminus in SAND2
sequences. These regions are highly conserved within the paralogous groups. They may be
two highly significant hydrophobic regions or potential membrane-spanning regions,
related to the protein interaction with organelles. Domain database searches of SBASE [46]
ProDom [47], BLOCKS [48], PRINTS [49] and PROSITE [50] return predicted features in
many SAND sequences but these were not consistent nor conserved among the individual
members of the SAND sequences submitted with the exception of PRINTS. PRINTS
returned 13 signature elements designated YEAST73DUF across 6 species. These elements
are all found within our alignments.
3.3. Multiple Sequence Alignment
The alignment generated from eleven full-length C-termini of the SAND protein sequences
was used as input for Jpred2 to predict the secondary structure and solvent accessibility.
This alignment is deposited in the EMBL-Align database [51] and can be retrieved using
the accession number ALIGN_000714. The sequence identity between the C-termini varies
from 25.76 and 77.48% (Table 5). There is no significant sequence similarity in the N-
termini (approximately first 100 amino acids) across the SAND subgroups (SAND, SAND1
and SAND2) (data not shown) and no consensus D-helices or E-strands predicted by Jpred2
for this region [25]. There is detectable sequence similarity within the subgroups of SAND1
and SAND2 N-termini sequences (data not shown).
The amino acid sequences of the SAND subgroups have a high degree of amino-acid
sequence identity which is typically >60% (Table 5). However more sequence divergence
is observed between the SAND2 proteins than is seen between the SAND1 proteins (Table
5, Figures 2-3). Several conserved motifs including sixteen invariant amino-acid residues
are highlighted in the alignment (Figure 3). Motif GKP is in L3; an alanine and a serine are
in solvent inaccessible D-helix A2; a leucine is in D-helix A3; an aspartate and a leucine are
in L9; a proline is in L11a; an arginine is in D-helix A5; an aspartate is in D-helix A7; a
PXCXP signature spans E-strand B7 and loop L16; phenylalanine is in D-helix A10 and in
E-strand B13. The alignment shows loops comprising insertions specific to certain SAND
protein groups. Loop L5 contains a plasmodium specific insertion, loop L10 contains an
insect specific insertion, loops L13, L21 and L27 are specific to SAND2 proteins and loop
L20 is plant specific. Conserved plant cysteines are found at alignment positions 98, 102,
133, 215, 448, 473 and 477 (Figure 3).
SeqFold and profiles-3D were used to predict the protein fold of the C-terminal
section of SAND. The structure of the PDB entry 1gw5, a small three layer sandwich with
solvent inaccessible strands and amphipathic helices, was predicted to match the first 100
amino acids of the C-terminal section of SAND. This prediction matches the secondary
structure consensus for that region (Figure 3). 1gw5 is the experimentally determined
protein structure of AP2 (a heterotetrameric clatharin adaptor complex). This protein
mediates endocytosis. SeqFold and profiles-3D predicted a match to annexins, 1a8a, an all
D-helical protein that matches residues 95-190 of the SAND family. This is complementary
to the Jpred2 prediction of 7 consecutive D-helices in this region. This D-bundle structure
has several solvent inaccessible D-helices that tie in with our secondary structure prediction
and solvent accessibility prediction. No conclusive predictions were made for the remaining
region.
Search WWH ::




Custom Search