EPI-peptide designer: a tool for designing peptide ligand libraries based on epitope-paratope interactions.

MOTIVATION: Antibodies are an important class of biological drugs, but with limitations, such as inadequate pharmacokinetics, adverse immunogenicity and high production costs. Synthetic peptides for the desired target represent an important alternative to antibodies. However, no computational tool exists to guide the design of these peptides. RESULTS: To identify the interacting residues in a given antibody-antigen (Ab-Ag) interface we used Interface Interacting Residue (I2R), a selection method based on computed molecular interactions. The aggregation of all the molecular interactions between epitope and paratope residues allowed us to transform the 3D Ab-Ag complex structures into interface graphs. Based on these data and the probability of molecular interaction we developed EPI-Peptide Designer tool that uses predicted paratope residues for an epitope of interest to generate targeted peptide ligand libraries. EPI-Peptide Designer successfully predicted 301 peptides able to bind to LiD1 target protein (65% of the experimentally tested peptides), an enrichment of 22% compared to randomly generated peptides. This tool should enable the development of a new generation of synthetic interacting peptides that could be very useful in the biosensor, diagnostic and therapeutic fields. AVAILABILITY AND IMPLEMENTATION: All software developed in this work are available at http://www.biocomp.icb.ufmg.br/biocomp/ CONTACT: liza@icb.ufmg.br SUPPLEMENTARY INFORMATION: Supplementary data are available at Bioinformatics online.

Protein-Protein and Peptide-Protein Interactions of NudE-Like 1 (Ndel1): A Protein Involved in Schizophrenia.

Schizophrenia (SCZ) is a devastating chronic mental disease determined by genetic and environmental factors, which susceptibility may involve an impaired neural migration during the neurodevelopmental process. Several candidate risk genes potentially associated with SCZ were related to the formation of protein complexes that ultimately mediate alterations in the neuroplasticity. The most studied SCZ risk gene is the Disrupted-in-Schizophrenia 1 (DISC1) gene, which functions seem to depend on the binding with cytoskeleton proteins, as the Nuclear-distribution gene E homolog like-1 (Ndel1) protein among others. Interestingly, Ndel1 is the only binding partner of DISC1 proteins with oligopeptidase activity, besides playing roles in multiple processes, including cytoskeletal organization, cell signaling, neuron migration, and neurite outgrowth. It is still not clear if the protein-protein interaction between Ndel1 and DISC1 is enough to explain all cellular functions attributed to these proteins, but there are several lines of evidence suggesting the importance of the catalytic activity of Ndel1 for the neurite outgrowth and neuron migration during embryogenesis. Recent works of the group have demonstrated the modulation of Ndel1 activity by DISC1, which is hypothetically impaired in SCZ patients. In fact, more recently, we also showed a lower Ndel1 activity in the plasma of SCZ patients compared to control health subjects, but the physiopathological significance of this feature is still unknown. Here we discuss Ndel1 ligands involved in protein-protein complex formations related to neurodevelopmental diseases, as (1) lissencephaly or Miller-Dieker Syndrome (MDS), which is characterized by the typical craniofacial features and abnormal smooth cerebral surface, and as (2) SCZ, since they both seem to be determined by defects in neuronal migration. Although impaired lissencephaly protein Lis1 complex formation with Ndel1 is the leading cause of lissencephaly, this binding does not affect Ndel1 oligopeptidase activity. On the other hand, although MDS and SCZ may be both determined by an abnormal neuronal migration, DISC1 complex formation with Ndel1 was shown to inhibit Ndel1 activity. Also differently of MDS, SCZ needs inputs from environmental factors, while lissencephaly is not likely dependent or affected by the environment. Several other proteins and peptide ligands were described for Ndel1, Lis1 and DISC1, thanks to the employment of biochemical, immunochemical, and biological (using cells or living animals) assays, including heterologous expression and also simply by purification from nature of these proteins in the complex form. Effects of the post-translational modifications of these proteins are also discussed here. Taken together, the data presented here show in essence how protein-protein and proteinpeptide interactions can underlie fundamental processes as cell division, maturation and migration, necessary for adequate formation of a complex structured tissue as the brain. A special attention was given to Ndel1 as this protein binds to either proteins or peptides, besides having proteolytic activity. Moreover, Ndel1 seems to be the key protein underlying two seemingly unrelated diseases with highly complex etiology, as lissencephaly and SCZ.