University of Southampton
Structure-based modelling of LK peptides in complex with antagomiR-138 RNA
Abstract
dc:description.abstractRNA is a multipotent polymer of great biological and engineering interest. microRNA mimics have revolutionized the manipulation of gene expression and are being explored as next-generation therapeutics. Despite their high versatility and specificity, their weak pharmacokinetic profile poses serious limitations, that could be bypassed by using biocompatible delivery vectors, such as cell-penetrating peptides (CPPs). Recently, the potent therapeutic microRNA antagonist antagomiR-138 has been successfully delivered using the pH-responsive, histidine-containing CPP LKH-stEK, both in in vivo and in vitro models of glioma. This CPP vector confers high target silencing, albeit with suboptimal complex formation efficiency and cell-penetrating profile, requiring the use of higher concentrations of peptide to achieve the optimal therapeutic effect. In this work, a novel multistage workflow is presented for the sequence-to-structure investigation of antagomiR-138—LKH-stEK nanocomplexes, and the design of new LKH-stEK analogues.<br/>In the first stage, the ab initio modelling of the currently unknown RNA and CCP structures was undertaken, using extensive multi-μs enhanced sampling simulations, in combination with knowledge-based and homology modelling tools, to thoroughly investigate the conformational spaces of interest. This process revealed the existence of multiple α-helical, and partially helical conformations for LKH-stEK-like peptides, that are stabilized to various degrees depending on the type of residues at positions 3, 6, and 9 of the 16-mer CPP. Additionally, the protonation of specific histidines of LKH-stEK was found to regulate the helical content of the CPP, suggesting a potential mechanisms for pH-triggered RNA release. For the 23-mer RNA, a parsimonious ensemble of alternative conformations, representing the highly variable hairpin-like predicted structure, was derived after μs-long enhanced sampling simulations under a carefully selected all-atom force field.<br/>In the second stage, RNA-CPP aggregates were produced through self-assembling simulations of large-scale systems containing a range of experimentally relevant RNA:CPP ratios (1:5 to 1:50). This thermodynamically-consistent process resulted in the formation of potential nanocomplex precursors with diverse interaction patterns between monomers, shedding light on previously unexplored structural details of these complex delivery systems. To address this complexity, a novel Python-based interaction analysis tool, capable of summarizing widely different intermolecular interactions within aggregates, was developed. This tool evaluates i) the importance of individual residues in relation to properties of interest (e.g. monomer binding) and ii) the different binding modes exhibited by monomers within aggregates, accounting both for local (molecule-based) and global (aggregate-wide) interaction patterns observed in simulation trajectories.<br/>Finally, based on the interaction patterns and hotspots of LKH-stEK-like peptides correlated with high complex formation efficiency and cell-penetrating ability, as well as prior information from the CPP design literature, three new targeted designs of LKH-stEK -namely L1W, G16W, and G16A- were proposed to improve specific aspects of the RNA- and peptide-binding capabilities of the lead CPP. Subsequent modelling and simulations of the new designs confirmed the initial hypotheses, further resulting in aggregates with new structural characteristics, making them interesting targets for further experimental investigation and development.
Degree
thesis:*- Name dc:type.qualificationname
- Ph.D.
- Level dc:type.qualificationlevel
- doctoral
- Grantor dc:publisher.institution
- University of Southampton
- Year dc:date.issued
- 2024
Author and committee
dc:creator, dc:contributor.*- Author dc:creator
-
- Stamatis, Dimitrios
- Advisors dc:contributor.advisor
-
- Essex, Jonathan
- Verma, Chandra
- Stulz, Eugen
- ORCID person:identifier.orcid
Rights
- Language dc:language
- English, eng