My current research involves simulating cell membrane permeabilization under an external electric field (electroporation or EP) using molecular dynamics (MD) simulations and quantifying the mechanisms which drive this phenomenon by connecting biological experiments to interdisciplinary theories. Electroporation can induce cell apoptosis (programmed cell death) or enhance the treatment of tumor cells (as is the case with electrochemotherapy) by facilitating the entry of therapeutic molecules across a normally impermeant cell membrane. This technique can also be used to improve the efficacy of electro gene therapy and gene expression in cells by increasing DNA transfection across a transiently permeabilized membrane. These simulations require optimization of massively-paralleled programming algorithms, physical characterization of biological and chemical processes, and relating current continuum models to our observations in MD.

Recent advances in parallel programming paradigms and high-performance computing have facilitated a new era in biological computation through the use of MD simulations. Although there has been considerable efforts to incorporate quantum mechanical properties in physical and chemical simulations (e.g. through density functional theory), biological simulations require both a large number of atoms and long, physiologically relevant timescales which are currently incompatible with existing quantum simulations (which are inefficient at best). Classical MD simulations however offer unparalleled computational speed using a fraction of the resources and at a reduced cost while still incorporating all major atomic properties such as charge, mass, Van der Waals forces, and chemical bond constraints. This has allowed us to simulate a wide variety of lipids, solvents, macromolecules, and other large-scale structures for microseconds and beyond, regimes which begin to approach what is observed in experiments. Classical MD allows for the direct simulation of these events from an atomistic point of view and gives researchers precise physical frameworks to explain their experimental observations and relate them to existing analytical models.

Our group at USC conducts biological experiments with Jurkat T Lymphocytes to confirm our theoretical observations in MD, and using these techniques I was able to simulate and confirm that peroxidized lipid bilayers electropermeabilize more readily than unoxidized lipid bilayers due to the attractions between additional acyl oxygen atoms located on the oxidized lipid tail and corresponding interfacial water molecules. This made the formation of water channels and subsequent electropores in oxidized bilayers much more likely than in systems without oxidized lipid tails. We confirmed this observation through experiments by showing that oxidized cells fluoresce with increased intensity compared to unoxidized cells, using the flourescent dye YO-PRO-1 in the buffer (which flouresces only inside of permeabilized cells) and after an external electric pulse was applied. This was signficant because it showed us that tumors could be locally and selectively oxidized, and this would allow for the application of smaller external electric fields during electrochemotherapy to target and permeabilize only a subset of tumor cells while allowing surrounding healthy tissue to survive.


Other projects I have worked on include the identification and characterization of electropore life cycles from pore initiation to pore dissolution, in order to allow for the objective comparison between two separate electropores. Using this metric I was able to identify that anionic lipids (e.g. phosphatidylserine) or cations (e.g. Ca2+) impede or extend the time to pore formation while also reducing the time for pore resealing after the external field had been removed. Once again the significance of this finding is that small heterogeneities in lipid composition (e.g. neutral versus charged lipids) and electrolyte composition (e.g. Ca2+ versus distilled water) make a dramatic difference in the permeability of cell membranes. Real cell membranes are composed of hundreds of different lipid types, ion channels, and membrane proteins, and thus their conformation and effect on membrane area per lipid are of paramount importance when determining the breakdown voltages of real cells.


More of my work can be found under the publications section of my CV. Also, some graphics of our work can be found in the gallery. If you would like to use any of our images, please let us know beforehand.

Moving forward I would like to connect existing experiments of EP in the lab with current continuum theories in order to explain the primary contributors and mechanisms of electric field-induced membrane permeabilization by running systematic and increasingly detailed MD simulations of lipid EP using highly-parallel computing architectures and free-energy analyses. This work could immediately improve the efficacy of current electrochemotherapy treatments by providing researchers with systems which enhance or reduce the effective electric fields deep in the membrane interior, in addition to achieving a full descriptive model of EP which is fundamentally important in understanding basic biology through simple first-principle physics.



I'm sorry, Dave. I'm afraid I can't go any further than that.