Proteome-Wide Inference of Human Endophilin 1-Binding Peptides

Abstract

Human endophilin 1 (hEndo1) is a multifunctional protein that was found to bind a wide spectrum of prolinerich endocytic proteins through its Src homology 3 (SH3) domain. In order to elucidate the unknown biological functions of hEndo1, it is essential to find out the cytoplasmic components that hEndo1 recognizes and binds. However, it is too time-consuming and expensive to synthesize all peptide candidates found in the human proteome and to perform hEndo1 SH3-peptide affinity assay to identify the hEndo1-binding partners. In the present work, we describe a structure/ sequence-hybrid approach to perform proteome-wide inference of human hEndo1-binding peptides using the information gained from both the primary sequence of affinity-known peptides and the interaction profile involved in hEndo1 SH3-peptide complex three-dimensional structures. Modeling results show that (i) different residue positions contribute distinctly to peptide affinity and specificity; P-1, P2 and P4 are most important, P1 and P3 are also effective, and P-3, P-2, P0, P5 and P6 are relatively insignificant, (ii) the consensus core PXXP motif is necessary but not sufficient for determining high affinity of peptides, and some other positions must be also essential in the hEndo1 SH3-peptide binding, and (iii) the alternating arrangement of polar and nonpolar amino acids along peptide sequence is critical for the high specificity of peptide recognition by hEndo1 SH3 domain. In addition, we also find that the residue type at a specific position of hEndo1-binding peptides is not stringently invariable; amino acids that possess similar polarity could replace each other without substantial influence on peptide affinity. In this way, hEndo1 presents a broad specificity in the peptide ligands that it binds.

Chemometrics‐based identification, characterization, and modeling of the intermolecular interaction between human osteosarcoma SH3 domain and its peptide ligands

“…Therefore, we herein employed PLS regression to ascertain the linear relationship between the structure and affinities of hOSF SH3‐binding peptides. The structure‐based statistical modeling of SH3‐peptide binding described by Wu et al was used to characterize the nonbonded interaction between the domain and peptide based on their molded complex structure, resulting in an array of structural descriptors, which was then correlated to experimentally measured affinity of SH3‐binding peptides. Consequently, 5 significant principal components were extracted by PLS modeling that could explain a majority of variance over the dependent variable and considerable variance with cross‐validation.…”

“…A structure‐based statistical modeling method proposed by Wu et al was employed here to perform statistical regression and prediction of SH3‐binding peptides. In the modeling, nonnonbonded interactions between domain and peptide were classified into 3 types, namely, electrostatic, steric, and hydrophobic; they were characterized between the domain and each residue type at each position of octapeptide described quantitatively using quasichemical potential approach, where the 3 nonbonded potentials were calculated by Coulomb's law, Lennard‐Jones equation, and empirical hydrophobic potential, respectively.…”

“…Here, we employed a chemometrics‐based approach to model and predict the hOSF binding partners. The method is a new solution that uses the structural knowledge of SH3 domain–peptide interactions to implement large‐scale inference of SH3‐binding peptides based on a chemometrics framework . Three matrices were first established on the basis of the virtual mutagenesis and interaction energy analysis to characterize complete interaction profile between the hOSF SH3 domain and all residue positions of peptide ligands.…”

Structure‐based statistical modeling and analysis of peptide affinity and cross‐reactivity to human senile osteoporosis OSF SH3 domain

“…Computational modeling of SH3 domain complex structures with its peptide ligands was performed using a protocol described previously . The NMR solution structure of amphiphysin SH3 in complex with chikungunya virus nsP3 peptide was retrieved from the protein data bank (PDB) database with accession ID 5I22.…”

Biology of endophilin and it’s role in disease

“…The SH3 structural domain recognizes and binds proline-rich structural domain, including synaptic proteins and dynamin proteins ( 15 , 19 , 20 ). N-BAR structural domain is required for binding lipid bilayer and inducing membrane bending ( 21 , 22 ), and can induce and stabilize membrane curvature upon dimerization ( 23 – 25 ). N-BAR has three major functional regions, including N-terminal amphiphilic helix (H0), amphiphilic helix inserted into helix 1 (H1), and crescent-shaped body formed by the dimerized BAR structural domain.…”