Structure of the tensin PTB domain. (A) Stereo Cα plot of the tensin1 PTB domain with every 10th residue, and N and C termini, labeled. The unstructured β6-β7 loop is shown schematically in green. (B) Ribbon diagram of the tensin1 PTB domain, with secondary structural elements labeled. A model of the integrin peptide is also shown (in red), including the side-chains of the NPxY tyrosine and the tryptophan residue at the –8 position. (C) Structure-based sequence alignment of the PTB domains of tensin family members, the talin PTB-like domain (Garcia-Alvarez et al. 2003), and other PTB domains (X11 [Zhang et al. 1997], Numb [Li et al. 1998], Dab1 [Stolt et al. 2003], Shc [Zhou et al. 1995b], IRS-1 [Eck et al. 1996; Zhou et al. 1996], and Dok1 [Shi et al. 2004]). β-strands are colored red; α-helices are blue. Residues involved in peptide binding are underlined, and those that directly bind phosphotyrosine are boxed. In the tensin homologs, two basic residues from the β6-β7 loop proposed to engage the phosphotyrosine are boxed in green. Numbers in parentheses for X11, Dab1, and Shc indicate the size of sequence insertions that have been removed for clarity. The asterisk in the X11 sequence marks a 19-residue insertion that is also not shown.

PTB-integrin models. Surface representations of the tensin1-integrin model (A) and the talin-integrin crystal structure (B) (Garcia-Alvarez et al. 2003), colored by electrostatic potential (blue for positive, red for negative). The integrin peptides are shown as sticks and colored by atom type. Integrin residue numbers are in black (and numbered with respect to the NPxY tyrosine in parentheses); in A, tensin residue numbers are given in green. Figures were generated with MOLSCRIPT (Kraulis 1991), RASTER3D (Kraulis 1991; Merritt and Murphy 1994), and PyMol (http://www.pymol.org).