Advanced Approaches for Metabolite Identification by Mass Spectrometry
Loading...
Author(s)
Asef, Carter K.
Advisor(s)
Editor(s)
Collections
Supplementary to:
Permanent Link
Abstract
The “omics cascade” begins with genomics and proceeds through the central dogma of biology to transcriptomics, proteomics, and ends with metabolomics. Similarly, the information gathered by these omics fields begins with the nature of an organism and proceeds to investigate nurture as one progresses through the cascade. Being at the bottom of the cascade, metabolomics offers the greatest capability for exploring changes resulting from environment, diet, or disease state, making it a central tool for understanding biological systems. Whereas the analytes of genomics, transcriptomics, and proteomics are assembled from of a limited number of building blocks (e.g. nucleotides and amino acids), metabolites span across a far greater chemical space, precluding the development of a singular analytical technique which can properly characterize all metabolites. Additionally, amplification of metabolites cannot be accomplished in the same manner as genetic material (e.g. polymerase chain reaction) as metabolites are not self-replicating. These challenging chemical traits necessitate a broad set of highly sensitive tools to fully probe the depths of the metabolome.
Mass spectrometry (MS) and nuclear magnetic resonance spectroscopy (NMR) are the two most used analytical techniques in metabolomics for their sensitivity and ability to characterize complex mixtures of small molecules. Generally, MS offers wider coverage of the metabolome at the cost of reduced capabilities for structural elucidation. This process of connecting spectral signals to specific molecular structures remains the greatest challenge to MS based metabolomics studies. The work presented in this thesis serves to evaluate and develop new technologies for MS based structural annotation of small molecules, and to synthesize these technologies into workflows which can be implemented in metabolomic studies.
Chapter 1 provides an overview of the techniques traditionally used for structural annotation of measured ion features in MS based metabolomics studies. This overview provides details is provided for ion mobility spectrometry (IMS), a high-speed separation technique which provides an additional dimension to MS analysis. The unique challenges of lipid identification is also described. Background information is provided for the field of newborn screening, a field of public health well suited for metabolomic analyses.
Chapter 2 describes the evaluation of two recently developed in silico techniques for predicting collision cross section (CCS), the two-dimensional rotationally averaged projection of molecular size and shape which is responsible for driving separation in IMS. This chapter explores the capability for IMS to serve as an added dimension for identification when paired to MS, as well as the degree of CCS prediction and measurement accuracy necessary for effective filtering of structural candidates.
Chapter 3 presents a metabolomics study on metabolic mutant strains of C. elegans worms which incorporated non-targeted IMS data. The study leveraged in silico approaches to generate large lists of candidate structures for ion features of interest from the tandem MS data collected in the non-targeted data set. A CCS prediction technique evaluated in chapter 2 was incorporated into the data processing workflow to filter candidate structures by their predicted vs. measured CCS. Further analysis of this data set produced new insights into best practices for utilizing non-targeted IMS data for metabolite identification.
Chapter 4 introduces a redesigned ion source which produces electrical current by the triboelectric effect. Previous work has shown the utility of triboelectric driven ionization for improving sensitivity and for aiding structural elucidation of lipid species. The low-current/high-voltage nature of triboelectricity offers the unique ability to maintain a corona discharge on the nanoelectrospray (nanoESI) emitter without damaging the orifice. The corona discharge is notably useful for driving epoxidation reactions, introducing new fragmentation sites on lipid fatty acid chains. The redesigned device described in this chapter retains these unique abilities while being easily reproduced and an order of magnitude smaller. Full structural elucidation of a phosphatidylcholine lipid found in human blood serum is accomplished with the redesigned triboelectric ion source.
Chapter 5 presents a non-targeted metabolomics workflow designed to explore correlations between birthweight and the metabolome, enhanced by the technologies described throughout this thesis. 810 newborn-derived dried bloodspot specimens were analyzed by liquid chromatography-MS. The results of this study provided new insights into the links between newborn birthweight and metabolic pathways related to the inborn errors of metabolism monitored by newborn screening (NBS). Additional analysis yielded a novel biomarker for the detection of total-parenteral nutrition, an intravenous nutritional supplement given to premature infants which often confounds NBS results. Lastly, this study investigated the effects of maternal nicotine exposure on embryonic development.
Chapter 6 gives a summary of the major findings in this thesis and discusses future directions which may result from the research described. The current state of CCS prediction and measurement is discussed as it relates to the ability for IMS to act as an additional dimension for metabolite identification. The advancements to triboelectric ionization made in this thesis are outlined, with comments on future experiments which may be enabled by the new ion-source design. Lastly, the major results of a metabolomics study on newborn DBS samples are described, highlighting the important findings which are relevant to current NBS practices.
Sponsor
Date
2024-08-20
Extent
Resource Type
Text
Resource Subtype
Dissertation