Dr. Evelyne Deplazes

Evelyne Deplazes
NHMRC Early Career Research Fellow

Contact Details

Dr. Evelyne Deplazes
Institute for Molecular Bioscience (King group)
School of Chemistry and Molecular Biosciences (Molecular Dynamics group)
The University of Queensland
St. Lucia, Brisbane
QLD 4072
Australia

email e.deplazes|at|uq.edu.au
Phone: +61 7 3365 7562

Biography

I completed a double degree in Chemistry and Computer Science at Curtin University in Perth (2003-2006). During my undergraduate research project I used quantum mechanical calculations to simulate the IR and Raman spectra of histidine in solution. After that I completed an Honours in Computational Chemistry (2007) focusing on quantum mechanical calculations of transition metals for modelling the chemical and physical properties of materials.

In 2012 I was awarded my PhD in Computational Biophysics from the University of Western Australia, Perth. As part of my PhD I have developed methods for integrating structural data from spectroscopy experiments into simulations to model the gating behavior of mechanosensitive ion channels that predicted the role of specific structural domains in the gating mechanism of the channel. Furthermore, my PhD project focused on using molecular dynamics and Monte Carlo simulations to aid in the analysis and interpretation of FRET spectroscopy experiments. After finishing my PhD I joined the research groups of Prof. Alan Mark (School of Chemistry and Molecular Biosciences, UQ) and Prof. Glenn King (Institute for Molecular Bioscience, UQ) to work on a collaborative project that focuses on peptides derived from animal venoms for the development of new drugs to treat chronic pain.

Research Interests and Projects

My main research area is the use of computational chemistry for investigating the structure-activity relationship of ion channels and their interactions with drug molecules. This allows me to integrate my knowledge and skills in chemistry, structural biology and computer science. I am experienced in a range of molecular modelling techniques and theoretical approaches including quantum mechanical calculations, Monte Carlo simulations, molecular dynamics simulations, homology modelling and docking. In addition I am interested in the developing computational methods that enables researchers to make better use of the data from spectroscopy experiment, with a particular focus on FRET and EPR.

Development of venom-based drugs for treating chronic pain
The main focus of this project is to develop potent and selective blockers of acid sensing ion channels (ASICs). These ion channels are the primary sensors of acidic conditions in humans and recent studies have shown that ASICs are involved in a range of physiological processes and medical conditions including pain, inflammation, neurological and psychiatric diseases. We are particularly interested in the ASIC3 and ASIC1a subtype as these have been shown to be involved in the perception and transmission of pain in the central and peripheral nervous system. This project combines molecular dynamics simulations and docking with experimental validation to characterize the interactions between a range of venom peptides and ASIC1a and ASIC3. The results from the simulations will aid the optimization of the peptides into small-molecule analgesics that might be more useful therapeutically.
This project is a collaboration between the MD group at the School of Chemistry and Molecular Biosciences (UQ) and the King group at the Institute for Molecular Biosciences (UQ) and is part of the venom-based drug design program in the King group to develop novel therapeutics to treat chronic pain. (See for example this article in ABC Science).

Balancing protons: The gating of acids sensing ion channels
The regulation of cellular pH is of central importance in human health. Consequently, acidosis, the over-acidification of the cells in our body, is associated with a range of disease states and illnesses. Our neurons, the cells in our central and peripheral nervous system, contain specific membrane proteins called acid sensing ion channels (ASICs) whose main function is to respond to changes in extracellular acid; they are essentially “balancing protons”. Due to the central importance of extracellular pH in both physiological and pathological conditions, ASICs play a central role in processes as diverse as touch sensation, perception of pain, gastrointestinal mechanosensory functions, fear conditioning and the synaptic plasticity involved in learning and memory formation. In addition, various studies demonstrated the role of ASICs in inflammatory and neuropathic pain, psychiatric illnesses, seizure termination, ischemic stroke and neuronal death during spinal cord injuries. Despite their importance we know very little about the workings of ASICs at the molecular level. Questions include: How do ASICs “sense protons”? How can the binding of a few tiny protons induce a cascade of structural changes in the ASIC protein that result in the channel switching from a non-conductive to a conductive state? What part of the protein makes some ASICs mechanosensitive while others are not? I use a molecular modelling approach, in particular Molecular Dynamics (MD) simulations, to address these questions. MD simulations allow me to observe the structural changes involved in the gating mechanism of ASICs at the molecular level. The structural insights can be combined with experimental validation to establish a structure-function relationship of ASICs. Understanding how ASICs “sense” protons will help us in building the fundamental knowledge required to understand the physiological and pathological conditions in which ASICs are involved.

Developing computational methods to improve the accuracy of structural data obtained from DEER spectroscopy
Our ability to understand physiological processes relies on the availability of structural data to create accurate 3D models of the proteins involved. Techniques to gather and process structural data are thus important tools for structural biology and related fields such as rational drug design. Double electron-electron resonance (DEER) spectroscopy is a powerful technique that can be used to measure distances between specific sites in a protein. However, relating the distance distributions from the DEER experiments to the structural model of the protein is challenging. Current methods are often inadequate and limit the accuracy of the resulting structural model. This project employs molecular dynamics simulations to develop new and robust protocols to accurately predict DEER spectra and the corresponding distance distributions. The outcomes of this project are computational methods that will significantly improve the interpretation of DEER experiments and thus increase our ability to obtain accurate structural models of proteins.
This project is a collaboration with Dr. Megan O'Mara at the School of Chemistry and Molecular Biosciences (UQ).

Publications

Links to all my publications can be found on my Google Scholar profile.

Book Chapters

1. O'Mara M., Deplazes E.: Polypeptide and Protein Modeling for Drug Design. In: Jaeger D., Jung R. (Ed.) Encyclopedia of Computational Neuroscience, 2013, available online

Refereed Journal Articles

1. Deplazes E. , Louhivuori M., Jayatilaka D., Marrink SJ., Corry B., 2012, Structural investigation of MscL gating using experimental data and coarse grained MD simulations, PloS Computational Biology, vol 8, issue 9

2. Martinac B. , Petrov E., Cranfield CG, Nomura T., Rhode PR., Battle AR., Landsberg JL, Foo A. Constatine M, Rothnagel R., Carne S., Chi G., Deplazes E., Cronell B., Hankammer, B., 2012, Bacterial mechansosensitive ion channels (Review), accepted for publication in Antioxidant & Redox signalling

3. Nomura T., Cranfield CG, Deplazes E., Owen DM., Macmillan A., Battle AR, Constantine M., Sokabe M., Martinac B. 2012, Differential effects of lipids and lyso-lipids on the mechanosensitivity of the mechanosensitive channels MscL and MscS, PNAS, vol 109 (122), 8770-8775

4. Deplazes E., Jayatilaka D and Corry B., 2011, Testing the use of molecular dynamics to simulate fluorophore motions and FRET, Physcial Chemistry Chemcial Physics, 13 (23), 11045-11054

5. Deplazes E., Jayatilaka D and Corry B., 2011, ExiFRET: A flexible tool for understanding FRET in complex geometries, Journal of Biomedical Optics, 17, 011005

6. Deplazes E., van Bronswijk W., Zhu F., Barron L.D. , Ma S., Nafi A. and Jalkanen K.J., 2010, A combined theoretical and experimental study of the structure and vibrational absorption, vibrational circular dichroism, Raman and Raman optical activity spectra of the L -histidine zwitterion, Theoretical Chemistry Accounts, Vol. 19, no. 1-3, pp. 155 – 176

7. Varganov S.A., Gilbert A.T., Deplazes E., Gill P.M., 2008, Resolution of the Coloumb operator, Journal of Chemical Physics, Vol. 128, no. 20, pp. 201104

Student Projects

The following are projects that are suitable for 3rd and 4th year undergraduate students, Honours or Masters students. Feel free to contact me for more information, and also with your own research ideas.

Gating modifiers vs pore blockers: Investigating the membrane partitioning of ion channel toxins using molecular dynamics simulations
Voltage-gated ion channels ion channels control the excitability of neurons and are responsible for the generation and transmission of nerve signals in the central nervous system. These ion channels are thus important drug targets for a range of diseases including ischemic stroke, chronic pain, sleep disorders, anxiety and epilepsy. Spider venoms are a rich source of pharmacologically active peptides, many of which modulate the activity of voltage-gated ion channels (VGIC).
VGICs are integral membrane proteins that consist of a central pore domain, providing a pathway for ions to travel across the cell membrane, and peripheral voltage-sensing domains that “sense” changes in membrane potential. Toxins that modulate the activity of VGICs do so by either physically occluding the central pore and thus prevent ions from passing through the channel (pore blockers) or by binding to the voltage-sensing domains and thereby altering the channel’s response to changes in membrane potential (gating modifiers). Recent experiments of gating-modifier toxins suggest that they partition into the membrane and it has been proposed that this contributes to their mechanism of action.
In this project we use molecular dynamics (MD) simulations to compare the interaction of gating modifiers and pore blockers with lipid bilayers. Specifically, we aim to test the hypothesis that gating modifiers show different specific peptide-lipid interactions and / or increased membrane partitioning in comparison to pore blockers. The project can be combined with surface plasmon resonance (SPR) spectroscopy experiments to compare the lipid binding activity of the different toxins. Characterising the membrane partitioning and lipid binding of these toxins can help us understand their mechanism of action and thus contribute to future rational drug design.

Molecular Dynamics simulations of a spider toxin with neuroprotective properties
Spider venoms are a rich source of peptides with various pharmacological properties. For example, the spider venom peptide Hi1a, recently discovered in the King lab at the Institute for Molecular Biosciences, shows neuroprotective activity and is thus a valuable drug lead for the treatment of ischemic stroke. Hi1a is works by inhibiting acid sensing ion channels (ASICs), proteins that are found in the central and peripheral nervous system.
The first step in using Hi1a as a lead molecule in rational drug design is to determine the toxin’s three-dimensional structure. The NMR structure of Hi1a revealed a unique structural motif in which two peptides (Hi1a monomers) are joined end-to-end to form a “double-knot toxin”. Each of these Hi1a monomers shows significantly reduced activity on ASICs, suggesting that the double-knot motif plays an important role in the toxin’s mechanism of action. The aim of this project is to further refine the high-resolution structure of Hi1a, with a particular focus on the relative orientation of the two Hi1a monomers. To achieve this we will combine the experimental data from NMR with molecular dynamics (MD) simulations to investigate the conformational flexibility of the peptide in solution. The resulting structural models of Hi1a will help us understand how this toxin binds to ASICs and thus facilitate future drug design.

Molecular Dynamics simulations for refining the NMR solution structure of a molt-inhibiting hormone
Molt-inhibiting hormones (MIH) are peptides that control molting in crustaceans (shedding of the shell). Due to their sequence similarity with the crustacean hyperglycemic hormone (CHH), MIH and CHH form a peptide family called CHH family. The CHH peptides are of considerable interest in aquaculture as growth promoters for crustaceans. In addition, genomic analysis suggests that MIH are the structural ancestor of peptides found in the venom of spiders and centipedes, some of which show analgesic or insecticidal properties.
The solution structure of the MIH from the Kuruma Prawn Marsupenaeus japonicas was solved by NMR in 2003. The structure revealed a novel structural motif that consists of 5 α-helices. Since then, the NMR structure has been used for structure-activity relationship studies and as a template for homology modelling of other peptides in the CHH family. Our recent molecular dynamics (MD) simulations, however, suggest that the first α-helix is not stable in solution and rather than 5 α-helices, the peptide has only 4 stable α-helices and a large disordered N-terminal region.
The aim of this project is to use MD simulations for the refinement of the MIH solution structure. The data from the MD simulations will be used for the (back)calculation of the NOE data for comparison to the experimental NMR data. The long-timescale MD simulations will also allow us to predict the conformational flexibility of the peptide in solution. The project can be extended to include the homology modelling and MD refinement of other peptides in the CHH family.