VEGFA, B, C: Implications of the C-Terminal Sequence Variations for the Interaction with Neuropilins.

Vascular endothelial growth factors (VEGFs) are the key regulators of blood and lymphatic vessels' formation and function. Each of the proteins from the homologous family VEGFA, VEGFB, VEGFC and VEGFD employs a core cysteine-knot structural domain for the specific interaction with one or more of the cognate tyrosine kinase receptors. Additional diversity is exhibited by the involvement of neuropilins-transmembrane co-receptors, whose b1 domain contains the binding site for the C-terminal sequence of VEGFs. Although all relevant isoforms of VEGFs that interact with neuropilins contain the required C-terminal Arg residue, there is selectivity of neuropilins and VEGF receptors for the VEGF proteins, which is reflected in the physiological roles that they mediate. To decipher the contribution made by the C-terminal sequences of the individual VEGF proteins to that functional differentiation, we determined structures of molecular complexes of neuropilins and VEGF-derived peptides and examined binding interactions for all neuropilin-VEGF pairs experimentally and computationally. While X-ray crystal structures and ligand-binding experiments highlighted similarities between the ligands, the molecular dynamics simulations uncovered conformational preferences of VEGF-derived peptides beyond the C-terminal arginine that contribute to the ligand selectivity of neuropilins. The implications for the design of the selective antagonists of neuropilins' functions are discussed.

Deep Learning for Novel Antimicrobial Peptide Design.

Antimicrobial resistance is an increasing issue in healthcare as the overuse of antibacterial agents rises during the COVID-19 pandemic. The need for new antibiotics is high, while the arsenal of available agents is decreasing, especially for the treatment of infections by Gram-negative bacteria like Escherichia coli. Antimicrobial peptides (AMPs) are offering a promising route for novel antibiotic development and deep learning techniques can be utilised for successful AMP design. In this study, a long short-term memory (LSTM) generative model and a bidirectional LSTM classification model were constructed to design short novel AMP sequences with potential antibacterial activity against E. coli. Two versions of the generative model and six versions of the classification model were trained and optimised using Bayesian hyperparameter optimisation. These models were used to generate sets of short novel sequences that were classified as antimicrobial or non-antimicrobial. The validation accuracies of the classification models were 81.6-88.9% and the novel AMPs were classified as antimicrobial with accuracies of 70.6-91.7%. Predicted three-dimensional conformations of selected short AMPs exhibited the alpha-helical structure with amphipathic surfaces. This demonstrates that LSTMs are effective tools for generating novel AMPs against targeted bacteria and could be utilised in the search for new antibiotics leads.