Date of Award

Spring 5-20-2023

Degree Type

Dissertation

Degree Name

PhD. Chemistry

Department

Chemistry and Biochemistry

Advisor

Monika Raj, PhD

Advisor

Gary E. Martin, PhD

Committee Member

Stephen P. Kelty, PhD

Keywords

NMR, MS sequencing, structure elucidation, ion mobility, DFT, cyclic peptides

Abstract

This dissertation reports on improvements in nuclear magnetic resonance (NMR) and mass spectrometry (MS) structure determination methods of organic compounds, with particular focus on challenging cyclic peptides. A recent and important innovation in NMR spectroscopy is the combination of theoretical property predictions, such as chemical shifts, using density functional theory (DFT) to aid in challenging NMR structure assignments, such as determination of regio- and stereo-configurations. In the first part of this thesis, a comprehensive benchmark study of DFT chemical shift prediction methods was performed using experimental NMR data collected from 50 well curated compounds, which was referred to as the DELTA50 test set. The best methods for proton and carbon chemical shift predictions were GIAO-WP04/6-311G(2d,p)//B3LYP-D3/6-311G(d,p) and GIAO-ωB97X-D/def2-SVP/B3LYP-D3/6-311G(d,p), respectively. Use of an implicit solvent model (i.e., polarizable continuum model) during both geometry optimization and chemical shift calculation improved the accuracy. The best performing DFT methods were then evaluated for 20 probe structures, which included small to medium-sized organic compounds, including natural products, and these were generally found to be more accurate than reported NMR chemical shift predictions in the literature. The calculated root-mean square deviations (RMSDs) for the probe set were 0.07 to 0.19 for 1H and 0.5 to 2.9 for 13C. Next, several challenging case studies, such as sungucine and (+/-)-anisodamine, for determination of relative stereoconfiguration were examined. Peptides were then studied in detail in the final two chapters. A liquid chromatography ion mobility mass spectrometry MS/MS method was developed to rapidly determine the primary structure of cyclic peptides. This approach was facilitated by a site-selective reaction that cleaves peptides at specific residues (viz., serine, threonine, cysteine, and glutamic acid), allowing for application of an automated de novo sequencing software. In the last chapter, the regio- and stereoconfigurations of reaction products from several newly reported peptide reactions were determined by NMR methods and further confirmed using DFT chemical shift predictions.

Share

COinS