Recognition of Unique Carboxyl-Terminal Motifs by Distinct PDZ Domains

Abstract

The oriented peptide library technique was used to investigate the peptide-binding specificities of nine PDZ domains. Each PDZ domain selected peptides with hydrophobic residues at the carboxyl terminus. Individual PDZ domains selected unique optimal motifs defined primarily by the carboxyl terminal three to seven residues of the peptides. One family of PDZ domains, including those of the Discs Large protein, selected peptides with the consensus motif Glu-(Ser/Thr)-Xxx-(Val/Ile) (where Xxx represents any amino acid) at the carboxyl terminus. In contrast, another family of PDZ domains, including those of LIN-2, p55, and Tiam-1, selected peptides with hydrophobic or aromatic side chains at the carboxyl terminal three residues. On the basis of crystal structures of the PSD-95-3 PDZ domain, the specificities observed with the peptide library can be rationalized.

The C-terminus of the HTLV-1 Tax oncoprotein mediates interaction with the PDZ domain of cellular proteins

“…The resolution of the structure of a complex associating PDZ3 of PSD-95 and a peptide has determined the molecular basis of such interactions (Doyle et al, 1996). These ®ndings have been also supported by a recent study of the motifs interacting with various PDZ domains by the oriented peptide library technique (Songyang et al, 1997). The structural data show that the terminal valine is engaged in a hydrophobic pocket¯anked by the b sheet B and the a helix B.…”

“…For clones 2 and 43, the phenylalanine is changed to leucine and isoleucine, respectively. These changes maintain the hydrophobic nature of the residues at these positions and the study of Songyang et al (1997) indicates that these changes also correspond to a C-terminal valine in the bound peptide. The threonine at position 72 establishes hydrogen bonds with an isoleucine in bB and with an histidine in aB.…”

“…This is also the case for the histidine residue, with the notable exception of clone 40. From the results of the study with the peptide library, this could mean that the product of this clone belongs to a group of PDZ domains preferring an hydrophobic amino acid residue at position 72 (Songyang et al, 1997). This would explain why the C-terminal tail of Tax was not observed to interact with this protein in our assays.…”

Potassium channel Kir4.1 macromolecular complex in retinal glial cells

“…This idea is lent credence by the findings that PDZ domains can discriminate interaction partners based on single amino acid substitutions (Songyang et al, 1997;Gee et al, 2000;Vaccaro et al, 2001). Therefore, the differences in the PDZ domain-binding C-termini of Kir4.1 (-R-I-S-N-V) and AQP4 (-V-L-S-S-V) may facilitate the preferential binding of AQP4 to a particular syntrophin isoform and Kir4.1 to another.…”

The HtrA family of serine proteases

“…Prc recognizes protein substrates possessing a short C-terminal tail or tag sequence (Keiler et al, 1995;. This is reminiscent of the way in which many eukaryotic PDZ domains recognize their targets (Fanning et al, 1996;Songyang et al, 1997;Stricker et al, 1997). Given the presence of two PDZ domains in HtrA, we have hypothesized that HtrA, like Prc, recognizes its targets by binding to a short sequence, either within or at the C-terminus of its target.…”