New strategies to identify bioactive compounds from complex matrices to combat drug-resistant bacterial infections

UNCG Author/Contributor (non-UNCG co-authors, if there are any, appear on document)
Emily R. Britton (Creator)
Institution
The University of North Carolina at Greensboro (UNCG )
Web Site: http://library.uncg.edu/
Advisor
Nadja Cech

Abstract: The use of plants as medicine for human health has been documented on 5000-year-old Sumerian clay slabs, which also serve as the first evidence of written language. From 1940 to 2014, 83 % of small molecule anti-cancer treatments and 67 % of small-molecule anti-infective drugs were discovered from or modeled after compounds found in nature. Furthermore, botanical medicines constitute a large portion of dietary supplements, with sales in the United States exceeding $7 billion in 2016. Natural product mixtures are chemically complex with multifaceted bioactivity, which further complicates the goal of understanding how these mixtures interact with specific biological systems. High resolving-power mass spectrometry coupled to ultra-high pressure liquid chromatography has played a significant role in making natural product research more efficient and reliable by improving the quality of biological measurements and by assisting in guiding isolation. The goals of this project were to: (1) develop a mass spectrometry-based assay that could be used to discover compounds that inhibit the spread of Staphylococcus aureus in the skin and to (2) use Hydrastis canadensis as a case study for investigating how biochemometrics, a methodology that combines chemical and biological data using multivariate statistics, can be best used to identify synergists and additives in complex mixtures. The first goal was achieved by adapting a crude disk diffusion assay where bacterial hyaluronidase degrades polymeric hyaluronan into disaccharide units, into a quenchable, small volume aqueous reaction in lo-binding tubes that could be directly analyzed via LC-MS. This method was validated by screening a known inhibitor of Streptococcus agalactiae hyaluronidase, ascorbic acid 6-palmitate, resulting in values within the same order of magnitude as literature reports. Instrumental method validation was also performed by evaluating repeatability, and intermediate precision, and determining the limit of detection and limit of quantification. This assay was then used to identify two fungal metabolites, leotiomycene A and leotiomycene B as the most potent Streptococcus agalactiae hyaluronidase inhibitors to date, and leotiomycene B as the first reported inhibitor of Staphylococcus aureus hyaluronidase. The concentrations at which these compounds inhibited hyaluronidase were found to not have any antimicrobial effects, and are therefore potential candidates for the development of anti-hyaluronidase treatments for Streptococcus agalactiae and Staphylococcus aureus infections. The second goal was achieved by creating a database of isolated Hydrastis canadensis constituents using LC-MS that included retention time, accurate mass in both polarities, and high mass accuracy fragmentation patterns in both polarities. The raw file for each pure compound was uploaded into the Global Natural Product Social Molecular Networking database for dereplication purposes and to facilitate open access data sharing. To evaluate the effectiveness of biochemometrics for identifying synergists and additives, an organic extract of aerial material from H. canadensis underwent various stages of fractionation. Each stage of resultant fractions was profiled using the previously described LC-MS method, and their ability to enhance the antimicrobial activity of berberine against S. aureus was evaluated. These data were uploaded into the program Sirius, where multivariate statistical analyses were applied to create selectivity ratio (SR) plots. Retention time-mass pairs with low SR values indicate that a specific compound correlates with the observed biological activity, and should therefore guide isolation efforts. It was found that three stages of sample purification were needed to obtain SR plots that correctly predicted known additives and synergists. Interestingly, a m/z that was not yet known to H. canadensis was predicted to be responsible for bioactivity, and upon isolation, structure elucidation and confirmatory testing, it was shown that the new flavonoid 3,3'-dihydroxy-5,7,4'-trimethoxy-6,8 C-dimethylflavone was indeed a synergist, thus showing that this workflow can be useful for understanding the complex biological activities of botanical medicines.

Additional Information

Publication
Dissertation
Language: English
Date: 2018
Keywords
Biochemometrics, Goldenseal, Mass spectrometry, Metabolomics, MRSA, Spreading factors
Subjects
Mass spectrometry
Chemometrics
Goldenseal $x Therapeutic use
Staphylococcus aureus infections
Streptococcus agalactiae

Email this document to