Using multiscale modeling to discover protein-sugar interactions and harness them for renewable energy and improved health

Nature has evolved a wide range of proteins responsible for storing energy in carbohydrates, transporting them, and cleaving their bonds to release energy, each exquisitely tuned to the unique stereochemistry of different sugars. These sugars also modify protein structure and function through post-translational attachment carbohydrates to proteins, with varied effects based on glycan composition and binding location. The study of carbohydrate-protein interactions is industrially important for efforts to harness biotechnology to create renewable fuels and chemicals from non-food biomass. Applications of this research also include human health, as defects in carbohydrate-active enzymes and protein glycosylation are implicated in human diseases including cancer, muscular dystrophy, and autoimmune disorders.

My group uses computational tools to probe these interactions at a wide range of length and time scales to answer questions ranging from fundamental understanding to industrial feasibility. We employ computational chemistry tools including quantum mechanics (QM), molecular dynamics (MD), and rare-event sampling methods to uncover fundamental understandings of protein-carbohydrate structure-function relationships, opening opportunities for rational design of enzymes and diagnostic tools. Collaborating with experimental groups, we aim to understand past and guide future wet-lab studies to advance renewable chemicals and fuels as well as health.

Currently, these interests fuel the following projects:

What if? Computational investigations of enzyme mutants to learn from nature’s design strategies, Led by Tucker Burgin

In collaboration with structural biologists, we seek to use simulations to rationalize and harness the ‘design’ strategies used by nature to synthesize, digest, and modify polysaccharides across all kingdoms of life. The ultimate research goal is to elucidate the molecular foundations underlying the behavioral changes imparted to enzymes by mutations in a generalizable way, to aid rational enzyme design towards renewable biofuels, novel medical research tools and biologics, and more.

Computational Tools for Protein Design, Led by Alex Adams with assistance from Rohith Pentaparthy

A challenge for utilizing biomass for renewable fuels and chemicals is achieving co-utilization of the primary constituent sugars: glucose and xylose. Typically, microorganisms selectively catabolize all available glucose before consuming xylose, leading to two distinct growth phases which complicates the design of an economical continuous process. We aim to investigate the molecular basis of glucose inhibition of xylose transport into cells through multi-scale atomistic modeling. We will apply this understanding to the rational design of improved protein transporters that advance biotechnology and future protein design.

Size-based sorting of Exosomes, Led by Emma Purcell

Exosomes are lipid bilayer membrane vesicles that are known to enhance tumor metastasis by stimulating both tumor growth and tumor cell migration. It is crucial to be able to sort exosomes out of blood to further elucidate their role in cancer. Current techniques for exosomal sorting in microfluidics are expensive and have low throughput, due to their use of either antibody capture or other active techniques. By designing a high-throughput, size-based exclusion method for the separation of exosomes using inertial microfluidics, we aim to facilitate the further study of exosome biology.

Conformational cartography: mapping and comparing ring conformational landscapes, Led by Justin Huber

with assistance from Sam Schwartz and Stephen Vicchio

Researchers utilizing Quantum Mechanic/Molecular Mechanic (QM/MM) simulations are able to provide fundamental insight to crucial molecular phenomena, such as the cleavage of the cellulose glycosidic bonds during biofuel production. The QM method captures essential electronic distortions in the cellulose structure initiated by the enzyme, which provides fundamental knowledge about the mechanism that is experimentally elusive. However, the expense of QM is often a limiting factor in the amount of conformational sampling that is feasible. We are investigating the role of the computational method used in the the results found through simulations of key phenomena.