Targeting the SH3 domain of human osteoclast‐stimulating factor with rationally designed peptoid inhibitors

Abstract

Human osteoclast-stimulating factor (hOSF) is an intracellular protein produced by osteoclasts that induces osteoclast formation and bone resorption. The protein contains a modular Src homology 3 (SH3) domain that mediates the intermolecular recognition and interaction of hOSF with its biological partners. Here, we proposed targeting the hOSF SH3 domain to disrupt hOSF-partner interactions for bone disease therapy by using SH3 inhibitors. In the procedure, the primary sequences of three known hOSF-interacting proteins (c-Src, SMN and Sam68) were parsed, from which totally 31 octapeptide segments that contain the core SH3-binding motif PXXP were extracted, and their binding behavior to hOSF SH3 domain was investigated at structural level using a biomolecular modeling protocol. Several SH3-binding candidates were identified theoretically and then determined to have high or moderate affinity for the domain using fluorescence spectroscopy assays. One potent peptide (425) APPARPVK(432) (Kd = 3.2 μM), which corresponds to the residues 425-432 of Sam68 protein, was used as template to derive N substitution of peptides (peptoids). Considering that proline is the only endogenous N-substituted amino acid that plays a critical role in SH3-peptide binding, the substitution was addressed at the two key proline residues (Pro427 and Pro430) of the template peptide with nine N-substituted amino acid types. By systematically evaluating the structural and energetic effects of different N-substituted amino acids presenting at the two proline sites on peptide binding, we rationally designed five peptoid inhibitors and then determined in vitro their binding affinity to hOSF SH3 domain. Consequently, two designed peptoids APPAR(N-Clp)VK and APPAR(N-Ffa)VK with Pro430 replaced by N-Clp and N-Ffa were confirmed to have increased (Kd = 0.87 μM) and comparable (Kd = 2.9 μM) affinities relative to the template, respectively. In addition, we also found that the Pro427 residue plays an essential role in restricting peptide/peptoid conformations to polyproline II (PPII) helix as the basic requirement of SH3 binding so that the residue cannot be modified. Copyright © 2016 European Peptide Society and John Wiley & Sons, Ltd.

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

“…The OSF can interact with an array of its cognate protein partners through its SH3 domain to orchestrate cellular signaling networks. Previously, Han et al identified 3 interacting partners of OSF SH3 domain, ie, proto‐oncogene tyrosine‐protein kinase ( c ‐Src); survival motor neuron (SMN); and Src‐associated in mitosis, 68 kD (Sam68), from which totally 31 octapeptide segments that contain linear polyproline motif PXXP recognized by SH3 domain were determined . Considering that the CICSR‐based QSAR model built in this study was based on SH3‐binding decapeptide ligands, we herein extended these 8‐ to 10‐mer peptides by adding a residue separately at N‐ and C‐termini of each peptide.…”

“…Short polyproline peptide segments have been widely used as the OSF SH3 ligands, as they can mimic the interacting sites of the domain's cognate partner proteins. However, these peptides show a broad specificity and high cross‐reactivity when binding to the domain, largely limiting the application of designing therapeutic peptides to disrupt the peptide‐mediated interactions of between human OSF and its partners in SO therapy . Recently, computational modeling and statistical analysis have been widely used to characterize the binding behavior and energetics of peptide ligands to their cognate domain receptors .…”

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

“…A total of 1472 peptides with known binding potency to SH3 domain were collected from the literatures and databases, from which we only selected those of 8 amino acids long (octopeptides) with determined affinity values and, consequently, 573 samples were extracted for further analysis. Compared to other public databases, this is a much more homogenous set of data, as all of it was generated in 1 of only 2 assay systems.…”

“…A combination of template‐directed peptide grafting and virtual mutagenesis described in our previous work was used to computationally model the complex structures of hOSF SH3 domain with its peptide ligands. First, the primary sequence of hOSF SH3 domainwas retrieved from the UniProt database with code Q92882, which was searched against the PDB database by using NCBI BLAST server to identify those homologous SH3 domains in complex with peptide ligands.…”

“…Previously, we proposed targeting hOSF SH3 domain to disrupt hOSF‐partner interactions for bone disease therapy by using SH3 inhibitors. We also successfully designed and structurally optimized a number of peptod inhibitors by replacing key proline residues in these peptides with unnatural N ‐substituted amino acids on the basis of the modeled complex structures of crystal hOSF SH3 domain with its cognate peptide ligands . The SH3 domain is a critical structural module, which allows hOSF to bind a variety of plasmatic proteins to orchestrate diverse cellular processes.…”