Figure 3: Solution Structure of the M. brevicollis crka1 SH2 Domain.
From: A pre-metazoan origin of the CRK gene family and co-opted signaling network
(a) The panel shows an amino-acid sequence alignment based on the solution structures of the SH2 domain of M. brevicollis crka1 (Mb-crka1), human CRKL and CRK (Hs-CRKL and Hs-CRK). Secondary structure elements of Mb-crka1 are indicated below the alignment. The basic residues surrounding the phosphotyrosine binding site and the hydrophobic residues located at the +3 residue binding pocket are indicated as gray and black shades, respectively. An asterisk indicates the i+1 position of β turn between βE and βF. (b,c) Panels b and c show ribbon and surface models of the mean structure of the Mb-crka1 SH2 domain, respectively. The surface model also shows the electrostatic potential. The basic and acidic residues are shown in blue and red, respectively. The phosphotyrosine binding site (pY) and the +3 residue binding site (+3) are indicated by purple letters. Consistent with previous structural studies of the mammalian CRK SH2 domain, two highly conserved basic residues (R15 and R33) lie in the pY binding pocket and interact with the phosphate group of phosphotyrosine in ligand proteins. In Mb-crka1 SH2, another basic residue (K56) also faces the pY binding pocket. This residue is conserved among the choanozoa crka proteins, but not in the other species (Supplementary Figure S6b). Another pocket lies on the other side of the three anti-parallel beta sheets and appears to be structurally sufficient to accommodate a proline residue identified in mammalian CRK/CRKL SH2 binding motif (the +3 position from a tyrosine residue). (d) A wire model of Mb-crka1 SH2 domain (black) is shown in stereo views, overlaid with human CRK and CRKL SH2 domains based on PDB IDs 2EO3 and 2EYV (red and blue, respectively). While human CRK has an extruded DE loop (see also Panel a), the other parts of the SH2 backbones are well-aligned.
