Predicting protein-membrane interfaces of peripheral membrane proteins using ensemble machine learning

Motivation: Abnormal protein-membrane attachment is involved in deregulated cellular pathways and in disease. Therefore, the possibility to modulate protein-membrane interactions represents a new promising therapeutic strategy for peripheral membrane proteins that have been considered so far undruggable. A major obstacle in this drug design strategy is that the membrane binding domains of peripheral membrane proteins are usually not known. The development of fast and efficient algorithms predicting the protein-membrane interface would shed light into the accessibility of membrane-protein interfaces by drug-like molecules. Results: Herein, we describe an ensemble machine learning methodology and algorithm for predicting membrane-penetrating residues. We utilize available experimental data in the literature for training 21 machine learning classifiers and a voting classifier. Evaluation of the ensemble classifier accuracy produced a macro-averaged F1 score = 0.92 and an MCC = 0.84 for predicting correctly membrane-penetrating residues on unknown proteins of an independent test set. Availability and implementation: The python code for predicting protein-membrane interfaces of peripheral membrane proteins is available at https://github.com/zoecournia/DREAMM.

Multi-Target Directed Ligands (MTDLs) Binding the σ1 Receptor as Promising Therapeutics: State of the Art and Perspectives

The sigma-1 (σ1) receptor is a ‘pluripotent chaperone’ protein mainly expressed at the mitochondria–endoplasmic reticulum membrane interfaces where it interacts with several client proteins. This feature renders the σ1 receptor an ideal target for the development of multifunctional ligands, whose benefits are now recognized because several pathologies are multifactorial. Indeed, the current therapeutic regimens are based on the administration of different classes of drugs in order to counteract the diverse unbalanced physiological pathways associated with the pathology. Thus, the multi-targeted directed ligand (MTDL) approach, with one molecule that exerts poly-pharmacological actions, may be a winning strategy that overcomes the pharmacokinetic issues linked to the administration of diverse drugs. This review aims to point out the progress in the development of MTDLs directed toward σ1 receptors for the treatment of central nervous system (CNS) and cancer diseases, with a focus on the perspectives that are proper for this strategy. The evidence that some drugs in clinical use unintentionally bind the σ1 protein (as off-target) provides a proof of concept of the potential of this strategy, and it strongly supports the promise that the σ1 receptor holds as a target to be hit in the context of MTDLs for the therapy of multifactorial pathologies.

Folding Behaviour and Antibacterial Activity of Ionic Complementary Peptide EAK-16

Aim: A major challenge in the development of new antibiotics is the biocompatibility within biological environment. Ionic complementary peptide (EAK-16) from amyloid protein, have the ability to adopt secondary structure conformation at membrane interfaces. This study aimed to investigate the effect of membrane on EAK-16 peptide folding and their antibacterial applications. Methodology: We studied secondary structural conformation of EAK-16 using circular dichroism (CD) spectroscopy in an aqueous environment and at membrane bilayers interfaces. Initially, the antibacterial efficacy was investigated against both Gram-positive and Gram-negative bacteria. Membrane mimicking models were synthesised with dimyristoylphosphatidylcholine (DMPC) and dimyristoylphosphatidylserine (DMPS) lipid vesicles using calcein leakage assay. Results: EAK-16 showed transition in secondary structural conformation. In aqueous environment, it was predominantly β-sheets and at membrane interfaces, it was mainly α-helical. EAK-16 peptide was highly active against bacteria (at minimum concentration applied) and membrane leakage was found to be > 60%. This effect was confirmed with both anionic lipids (DMPS) and neutral lipids (DMPC). The helical transition of EAK-16 could be a major factor to disrupt the membrane and bacterial death Conclusion: The secondary structural conformation and calcein leakage data suggest that EAK-16 has potential to kill bacteria by adopting helical tilted conformation and membrane perturbation via lysis. This study revealed structure-function relationship of peptide and lipid bilayers to further investigate the mode of pore formation and mode of action of EAK-16 in membrane perturbation and antibacterial efficacy.