iSPOT: a web tool for the analysis and recognition of protein domain specificity

Reduced false positives in PDZ binding prediction using sequence and structural descriptors

Abstract—Identifying the binding partners of proteins is a problem of fundamental importance in computational biology. The PDZ is one of the most common and well-studied protein binding domains, hence it is a perfect model system for designing protein binding predictors. The standard approach to identifying the binding partners of PDZ domains uses multiple sequence alignments to infer the set of contact residues that are used in a predictive model. We expand on the sequence alignment approach by incorporating structural information to generate descriptors of the binding site geometry. Furthermore, we generate a real-value score for binary predictions by applying a filter based on models that predict the probability distributions of contact residues at each of the canonical PDZ ligand binding positions. Under training cross validation, our model produced an order of magnitude more predictions at a false positive proportion (FPP) of 10 percent than our benchmark model chosen from the literature. Evaluated using an independent cross validation, with computationally predicted structures, our model was able to make five times as many predictions as the benchmark model, with a Matthews' correlation coefficient (MCC) of 0.33. In addition, our model achieved a false positive proportion of 0.14, while the benchmark model had a 0.25 false positive proportion.

Putting into practice domain-linear motif interaction predictions for exploration of protein networks

PDZ domains recognise short sequence motifs at the extreme C-termini of proteins. A model based on microarray data has been recently published for predicting the binding preferences of PDZ domains to five residue long C-terminal sequences. Here we investigated the potential of this predictor for discovering novel protein interactions that involve PDZ domains. When tested on real negative data assembled from published literature, the predictor displayed a high false positive rate (FPR). We predicted and experimentally validated interactions between four PDZ domains derived from the human proteins MAGI1 and SCRIB and 19 peptides derived from human and viral C-termini of proteins. Measured binding intensities did not correlate with prediction scores, and the high FPR of the predictor was confirmed. Results indicate that limitations of the predictor may arise from an incomplete model definition and improper training of the model. Taking into account these limitations, we identified several novel putative interactions between PDZ domains of MAGI1 and SCRIB and the C-termini of the proteins FZD4, ARHGAP6, NET1, TANC1, GLUT7, MARCH3, MAS, ABC1, DLL1, TMEM215 and CYSLTR2. These proteins are localised to the membrane or suggested to act close to it and are often involved in G protein signalling. Furthermore, we showed that, while extension of minimal interacting domains or peptides toward tandem constructs or longer peptides never suppressed their ability to interact, the measured affinities and inferred specificity patterns often changed significantly. This suggests that if protein fragments interact, the full length proteins are also likely to interact, albeit possibly with altered affinities and specificities. Therefore, predictors dealing with protein fragments are promising tools for discovering protein interaction networks but their application to predict binding preferences within networks may be limited.

Regulation of mu opioid receptor internalization by the scaffold protein RanBPM

Mu opioid receptors (MOP) are transducers of the pharmacological effects of many opioid drugs, including analgesia and tolerance/dependence. Previously, we observed increased MOP signaling during postnatal development that was not associated with increased MOP or G protein expression. A yeast two-hybrid screen of a human brain cDNA library using the MOP C-terminus as bait identified RanBPM as a potential MOP-interacting protein. RanBPM has been recognized as a multi-functional scaffold protein that interacts with a variety of signaling receptors/proteins. Co-immunoprecipitation studies in HEK293 cells indicated that RanBPM constitutively associates with MOP. Functionally, RanBPM had no effect on MOP-mediated inhibition of adenylyl cyclase, yet reduced agonist-induced endocytosis of MOP. Mechanistically, RanBPM interfered with beta arrestin2-GFP translocation stimulated by MOP but not alpha(1B)-adrenergic receptor activation, indicating selectivity of action. Our findings suggest that RanBPM is a novel MOP-interacting protein that negatively regulates receptor internalization without altering MOP signaling through adenylyl cyclase.

RanBPM, a scaffolding protein in the immune and nervous systems

We review the literature for Ran Binding Protein in the Microtubule-Organizing Center (RanBPM; RanBP9), a 90-kDa protein that possesses many characteristics of a scaffolding protein, including protein-interaction motifs, a cytoskeletal-binding domain, and multiple canonical docking sites for signaling intermediates. We focus on studies that have examined functional interactions between RanBPM and other proteins. These studies suggest that RanBPM provides a platform for the interaction of a variety of signaling proteins, including cell surface receptors, nuclear receptors, nuclear transcription factors, and cytosolic kinases. These studies indicate that RanBPM acts as a scaffolding protein and is important in regulating cellular function in both the immune system and the nervous system.

Solution structure of a Hck SH3 domain ligand complex reveals novel interaction modes

We studied the interaction of hematopoietic cell kinase SH3 domain (HckSH3) with an artificial 12-residue proline-rich peptide PD1 (HSKYPLPPLPSL) identified as high affinity ligand (K(D)=0.2 muM). PD1 shows an unusual ligand sequence for SH3 binding in type I orientation because it lacks the typical basic anchor residue at position P(-3), but instead has a tyrosine residue at this position. A basic lysine residue, however, is present at position P(-4). The solution structure of the HckSH3:PD1 complex, which is the first HckSH3 complex structure available, clearly reveals that the P(-3) tyrosine residue of PD1 does not take the position of the typical anchor residue but rather forms additional van der Waals interactions with the HckSH3 RT loop. Instead, lysine at position P(-4) of PD1 substitutes the function of the P(-3) anchor residue. This finding expands the well known ligand consensus sequence +xxPpxP by +xxxPpxP. Thus, software tools like iSPOT fail to identify PD1 as a high-affinity HckSH3 ligand so far. In addition, a short antiparallel beta-sheet in the RT loop of HckSH3 is observed upon PD1 binding. The structure of the HckSH3:PD1 complex reveals novel features of SH3 ligand binding and yields new insights into the structural basics of SH3-ligand interactions. Consequences for computational prediction tools adressing SH3-ligand interactions as well as the biological relevance of our findings are discussed.

Identification and sequence analysis of six new members of the NIMA-related kinase family in Chlamydomonas

The NIMA kinases are an evolutionarily conserved protein family with enigmatic roles in the regulation of mitosis. We report six new members of this family in Chlamydomonas, in addition to the previously identified NIMA-related kinase, Fa2p. Chlamydomonas NIMA-related kinases (CNKs) 1-6 were sequenced from subclones generated by RT-PCR using information from EST libraries and the recently sequenced Chlamydomonas genome. Phylogenetic and bioinformatic approaches were used to determine the relationships of the six new members with known members of the NIMA-related kinase family. Although humans express at least eleven NIMA-related kinases, the eukaryotic microbes that have been studied to date express only one or two members of the family. Thus, the discovery that Chlamydomonas expresses a total of at least seven NIMA-related kinases is intriguing. Our analyses suggest that members of this family may play roles in the assembly and function of cilia.

iSPOT: A web tool to infer the interaction specificity of families of protein modules

iSPOT (http://cbm.bio.uniroma2.it/ispot) is a web tool developed to infer the recognition specificity of protein module families; it is based on the SPOT procedure that utilizes information from position-specific contacts, derived from the available domain/ligand complexes of known structure, and experimental interaction data to build a database of residue-residue contact frequencies. iSPOT is available to infer the interaction specificity of PDZ, SH3 and WW domains. For each family of protein domains, iSPOT evaluates the probability of interaction between a query domain of the specified families and an input protein/peptide sequence and makes it possible to search for potential binding partners of a given domain within the SWISS-PROT database. The experimentally derived interaction data utilized to build the PDZ, SH3 and WW databases of residue-residue contact frequencies are also accessible. Here we describe the application to the WW family of protein modules.