Atoms to Phenotypes: Molecular Design Principles of Cellular Energy Metabolism

Abstract

Bioenergetic membranes are the key cellular structures responsible for coupled energy-conversion processes, which supply ATP and important metabolites to the cell. Here, we report the first 100million atom-scale model of an entire photosynthetic organelle, a chromatophore membrane vesicle from a purple bacterium, which reveals the rate-determining steps of membrane-mediated energy conversion. Molecular dynamics simulations of this bioenergetic organelle elucidate how the network of bioenergetic proteins influences membrane curvature and demonstrates the impact of thermal disorder on photosynthetic excitation transfer. Brownian dynamics simulations of the quinone and cytochrome c2 charge carriers within the chromatophore interior probe the mechanisms of nanoscale charge transport under various pH and salinity conditions. Reproducing phenotypic properties from atomistic details, a rate-kinetic model evinces that low-light adaptations of the bacterium emerge as a spontaneous outcome of optimizing the balance between the chromatophore's structural integrity and robust energy conversion. Put together, the hybrid structure determination and systems-level modeling of the chromatophore, in conjunction with optical spectroscopy, illuminate the chemical and organizational design principles of biological membranes that foster energy storage and transduction in living cells. Parallels are drawn with the more universal mitochondrial bioenergetic machinery, from whence molecular-scale insights on the mechanism of cellular aging are inferred. This endeavor made feasible through the advent of petascale supercomputers, paves the way to first-principles modeling of whole living cells.

Molecular Basis of SARS-CoV-2 Infection and Rational Design of Potential Antiviral Agents: Modeling and Simulation Approaches

“…They have proven to be extremely efficient in modeling the interactions and the related conformational reorganization between proteins, nucleic acids, and lipid membranes. 29 − 31 Molecular simulations have allowed us to resolve the complex processes related to, among others, enzymatic catalysis 32 − 35 and DNA lesion production and repair 36 − 40 and to unravel the key mechanisms of passive and active membrane transporters. 41 − 46 These successes have also been made possible by the impressive development of computational algorithms, including enhanced sampling and free-energy methods, 47 which nowadays allow us to simulate the behavior of systems of hundreds of thousands of atoms up to the microsecond time scale.…”

“… 41 − 46 These successes have also been made possible by the impressive development of computational algorithms, including enhanced sampling and free-energy methods, 47 which nowadays allow us to simulate the behavior of systems of hundreds of thousands of atoms up to the microsecond time scale. 31 , 48 …”

Simulations of Asymmetric Membranes Illustrate Cooperative Leaflet Coupling and Lipid Adaptability

“…Advances in computer simulations of lipid bilayers have made simulating complex and more biologically relevant membranes possible. This is exemplified by recent simulations on a realistic plasma membrane (Marrink et al, 2019), a neuronal membrane (Ingólfsson et al, 2017), bacterial membranes (Khalid et al, 2015), and entire organelles, such as a chromatophore vesicle (Singharoy et al, 2019). While at the outset it seems trivial for one to set up an asymmetric membrane in silico, as one can simply place lipids on one side or the other, there are problems that can create artifacts (Park et al, 2015;Huber et al, 2019).…”

Mind reading of the proteins: Deep-learning to forecast molecular dynamics

“…In MD simulations, the chronological evolution of an N -particle system is computed by solving the Newton's equations of motion. Methodological developments in MD has pushed the limits of computable system-sizes to hundreds of millions of interacting particles, and timescales from femtoseconds (10 −15 second) to microseconds (10 −6 second), allowing all-atom simulations of an entire cell organelle [23]. High performance computing, parallelized architecture, speciality hardware and GPUaccelerated simulations have made notable contributions towards this progress.…”

“…Thus, the resolution of our MD data sets will result in novel training strategies that decrypt an inhomogeneously evolving time series. As a publicly accessible resource, our MD simulations trajectories of even larger systems (10 5 -10 7 particles) [23] will be provided in the future to seek generalizable big-data solutions of fundamental Physics problems.…”