Computational Modeling of Non-Canonical Amino Acid to Determine Photophysical Mechanisms of Experimentally Observed Fluorescence

Jimin working under vent hood

Students

2018
College

Faculty

Associate Professor of Chemistry

Project Summary

One of the most diverse and abundant biological molecules, proteins are involved in virtually every cellular process that takes place in organisms. The linear chains of amino acids fold into specific three-dimensional structure, which allows them to perform their essential functions in cells. Not surprisingly, misfolded proteins are often hallmarks of significant human diseases, making protein folding an important area of study in Biochemistry.

For the past two years, I worked in the Petersson lab to investigate peptides containing a non-canonical amino acid (NCAA) and a thioamide as a fluorophore-quencher pair. Fluorescent probes and quenchers are one of the many methods used to elucidate the molecular mechanism of the misfolding process of intrinsically disordered proteins. Here, the fluorescence intensity dependence on the fluorophore-quencher distance is used to measure the distance between various positions on the protein. My research focused on understanding the mechanism by which the fluorescence is quenched, as the definition of this mechanism directly provides the relationship between probe space and degree of quenching.

With the research grant provided by CURF, I used computational simulation to confirm the energy transfer mechanism between the sidechain thioamide and various NCAA-based fluorophores. I developed parameters that can accurately represent three fluorescent NCAAs: thioacetyl-lysine (LysAcS), p-cyanophenylalanine (Cnf), and 7-methoxycoumarin-4-ylalanine (Mcm), so that they can be used to simulate the structure of peptides containing these groups. In the simulation, the fluorophore and the quencher were separated by varying number of prolines, which form a rigid and stable helical chain. Three types of polyproline peptides were simulated: (1) Tryptophan – (Pro)n – LysAcS (2) Mcm-(Pro)n- LysAcS, and (3) Cnf-(Pro)n- LysAcS, where n was varied from 2 to 6. The simulation was accomplished using Rosetta, a software package that contains algorithms for protein structure prediction and functional de novo design.

This summer, I synthesized two additional series of peptides, one containing thioacetyl-glutamine and the other containing thioacetyl-asparagine instead of LysAcs. I learned techniques involved in solid-phase peptide synthesis and purification, as well as organic synthesis of amino acids containing sidechain thioamide. When the syntheses are completed, the fluorescence of each peptide will be measured and the experimental data will be used as distance constraints to improve the simulation. I especially enjoyed my time learning how to construct and perform experiments that can corroborate and optimize my computational work.

Last, I would like to thank Jack Ferrie, Miklos Szantai-Kis, and the Petersson Lab for the mentorship and support.