This special edition of the BioExcel webinar series features student speakers who were awarded poster prizes at the BioExcel Summer School 2022. Read along to find out more about our speakers and their research.
After obtaining the bachelor’s degree in Physics at the Physics Department of the University of Milan, Costanza enrolled in the master’s degree course at the same institution, choosing the Complex Systems curriculum. There, Costanza could deepen her knowledge in statistical physics with applications in condensed matter and biophysics. She developed a particular interest in the physical modeling of the structure and dynamics of biological macromolecules, learning the most common simulation techniques for macromolecules.
Currently, Costanza is a first-year PhD student at the Physics Department of the University of Trento in the Statistical and Biological Physics group under the supervision of Prof. Gianluca Lattanzi, where she employs molecular dynamics simulations for the investigation of proteins and other biological macromolecules of medical interest.
Pore-forming toxins: investigating the action mechanism with molecular dynamics simulations
In the Statistical and Biological Physics group of the University of Trento, we are studying the action mechanism of the pore-forming toxin γ-hemolysin, composed of the LukF and Hlg2 components. These proteins are expressed by Staphylococcus aureus as water-soluble monomers which cause the lysis of the leukocyte cells of the host by assembling into an oligomer pore on the cytoplasmic membrane. Several stages of the pore-forming mechanism have already been captured by experiments, but little is known about the complex dynamical process and the conformational changes involved. Through molecular dynamics simulations at different resolution scales, we have investigated the first steps of pore formation. We have tested the effect of membrane composition on the ability of γ-hemolysin components LukF and Hlg2 to steadily adhere to the lipid, finding agreement with experimental data of γ-hemolysin pore formation on model membranes. We have then employed these results to investigate the subsequent stage of inter-monomer interaction and dimerization on the membrane, gaining insights on the LukF and Hlg2 interaction interface and on the putative mechanism of transition between soluble monomers and the membrane bound dimer.
Rachael is a 3rd year PhD student at Newcastle University. She completed her MChem in Computational Chemistry at Heriot-Watt University, graduating in 2019 with First Class honours. Her dissertation project investigated the use of CP2K to predict the formation of unknown σ-alkane complexes. Rachael’s PhD research (with the Cole group) sits at the interface between chemistry, mathematics and data science, with a focus on the development of a new method for determining 3D shape similarity between molecules for use in drug discovery. If successful, these methods will be used in projects of interest to the Newcastle Centre for Cancer. In her 3rd year she was awarded a 6-month Enrichment placement at the Alan Turing Institute to investigate whether better outcomes can be reached by combining this work with machine learning methodologies.
RGMolSA: Riemannian Geometry for Molecular Surface Approximation
3D shape similarity between molecules is a useful tool to predict structurally diverse hits in Ligand-Based Virtual Screening. Compared to other similarity measures, 3D shape has the advantages of allowing for consideration of different conformers and enables scaffold hopping, which can aid rescue of candidates with undesirable properties. As there is no absolute definition of molecular shape, vector descriptors produced using mathematical approximations are used in comparison studies.
RGMolSA1 is a new molecular shape descriptor that uses the Riemannian metric to describe the geometry of the molecular surface. This gives a simple vector descriptor constructed of the weighted surface area and eight non-zero eigenvalues. The eigenvalues are obtained by considering the spectrum of the Laplacian associated with the surface. These descriptors are alignment-free and mesh-free and are quick to calculate, facilitating screening of large databases on a reasonable timescale. An initial case study using a series of PDE5 inhibitors known to have similar shape revealed the method compared well to existing methods in the field. The descriptors were also found to handle the consideration of different conformers accessible to the molecules well, with consistently high similarity amongst conformers of the same molecule as expected. The code and data used to produce the results are available via GitHub: https://github.com/RPirie96/RGMolSA.
 Cole, D. J., Hall, S. J. & Pirie, R. Riemannian Geometry and Molecular Surfaces I: Spectrum of the Laplacian. arXiv:2201.04230 [math, q-bio] (2022).
Serena has achieved their Bachelor’s and Master’s degrees in Physics at the University of Trento, Italy. Last October, Serena took part in the interdisciplinary doctoral program offered by the International Max Plank Research School for Cellular Biophysics (IMPRS-CBP). Serena is developing the PhD project in Dr Roberto Covino’s lab at the Frankfurt Institute for Advanced Studies (FIAS) of Germany in collaboration with experimental colleagues from Prof. Mike Heliemann’s group. In their research, Serena uses MD simulations to study the activation mechanism of a human tyrosine kinase receptor involved in a complex interaction network on the plasma membrane.
Mechanistic Insight into the Early Events of the Activation of the c-Met Receptor during Listeria Invasion
The human receptor tyrosine kinase Met is a transmembrane glycoprotein placed on the plasma membrane. The Met receptor is crucial in regulating cell migration, replication, and growth. Its dysregulation leads to a spectrum of diseases featuring cancer and bacterial invasion. In particular, the intracellular bacterium Listeria monocytogenes infects cells via a Met-mediated internalization through the invasion protein Internalin-B (InlB). Here, we focused on investigating the activation mechanism of Met’s ectodomain triggered upon binding with InlB. We used atomistic molecular dynamics simulations to study the conformational plasticity of the receptor in different scenarios: isolated or in complex with InlB and with different glycosylation patterns. Our simulations reveal the signalling-competent and inactive conformations of the receptor. In addition, we observed alternating bridging interactions among the glycans that stabilize the inactive conformation of the receptor. In the future, we will explore how the complex membrane environment modules MET’s activation.