Microbial metabolic exchange mediates microbial interactions and plays key roles in regulating biology and has shaped modern healthcare, agriculture and other commercial processes. In this thesis, cutting edge mass spectrometry techniques, new genome mining approaches, and innovative bioinformatics tools were coupled and developed into investigating microbial metabolic exchange and led to the identification and characterization of biological active molecules that may have therapeutic values. The thesis begins by informing the reader current understanding of microbial metabolic exchange as well as the emergence of new technologies that are allowing us to eavesdrop on microbial conversations. Chapter 2 describes the use of imaging mass spectrometry (IMS) to study a microbial intraspecies interaction which led to the identification and structural elucidation of two complex metabolites, sporulation killing factor (SKF) and sporulation delaying protein (SDP), that are active in a Bacillus subtilis cannibalism system. Further bioassays investigate the cannibalistic activity of SDP and also reveal SDP inhibits Staphylococci pathogens with comparable IC50 to vancomycin. Chapter 3 introduces a new genome-mining approach (peptidogenomics), and when combined with IMS leads the discovery of anti-infective agent arylomycin and its biosynthetic gene cluster in S. roseosporus. Chapter 4 goes beyond studying one or two metabolites at a time, presents the systematic characterization of microbial metabolic profile by the combination of peptidogenomics and an innovative bioinformatics tool (MS/MS spectral molecular networking) that is capable of clustering molecules of similar structure classes which leads to efficient characterization of the molecular universe of S. roseosporus. One of the molecules, stenothricin, after further investigations suggest it appears to be unique in its mechanism of antibiotic activity. Lastly, chapter 5 puts forth future directions envisioning how these tools and approaches described in the thesis can be further extended into investigating more complex microbial interactions that are of significant biological or clinical importance. Here, a proposal for investigating the molecular insight of fecal transplantation for the treatment of Clostridium difficile-associated diarrhea was presented. Similar approach could be undertaken to investigate other important human microbiota associated diseases/disorders such as Crohn's diseases, obesity, and diabetes

Natural products are a crucial component in drug discovery because of their considerable pharmaceutical properties. Cyclic non-ribosomally peptides are one category of natural products featured by containing non-standard amino acids and lactam or lactone structures, thus increasing the complexity of the resulting tandem mass spectrometry data. Cyclosporin, microcystins and nodularins all are well-known examples and have notable pharmacologically importance. In this current work, by collaborating with Dr. Pevzner bioinformatics group, a program was developed for the annotation and characterization of tandem mass spectra obtained from cyclic peptides. This program, MS- CPA is available as a web tool (http://lol.ucsd.edu/ms- cpa_v1/Input.py). The second half of this thesis will be focused on a ribosomally encoded peptide antibiotic trifolitoxin produced by a rhizobium strain. This peptide undergoes multiple unknown post-translational modifications results in a loss of 24 Da masses. Mass spectrometry based strategies as well as in vivo reconstitution experiments were used to partially characterize trifolitoxin structure.

A simple but effective rare earth metal (RE) confinement structure is demonstrated to suppress surface accumulation of Yb generally encountered in the RE metal incorporated Ni silicide system. The confinement structure is realized by inserting a Ti diffusion barrier layer between Yb and Ni layers. Yb atoms can be constrained in a specified reaction region during silicidation as evidenced by Auger electron spectroscopy and cross-section transmission electron microscopy analysis. The RE metal confinement structure provides a complementary metal-oxide-semiconductor compatible approach for further Schottky barrier height engineering and is a promising method for future technology nodes.

In recent years, hybrid nanostructure consisted of plasmonic metals and different dielectrics have attracted much attention for their intriguing plasmonic properties. Recent studies have also shown that by introducing plasmonic metals, the photocatalytic efficiency of semiconductor can improve via plasmon-enhanced light absorption. In this work, we have demonstrated excellent photocatalytic properties for hydrogen production using hexagonal close-packed core-shell Au/TiO2 hybrid nanocrystal arrays. Under both ultra-violet and visible light, significant increase in the hydrogen production from methanol (20%) solution splitting was achieved with the hybrid Au/TiO2 nanocrystal arrays in comparison with bare TiO2 thin film as well as randomly distributed Au/TiO2 nanocrystals. From the finite difference time domain (FDTD) simulation, the significant increase (up to 60%) in hydrogen production can be correlated to strong and optimum coupling of the enhanced electric field from localized surface plasmon resonance in Au/TiO2 nanocrystal arrays. In addition to allowing more accurate measurement of plasmonic enhancement, the ordered nanostructures have been shown to be especially amenable to the systematic analysis of lateral coupling of plasmonically enhanced electric field. As a result, optimal structures with appropriate spacing of core-shell metal-dielectric nanocrystals, metal core size and dielectric shell thickness for maximum enhancement can be designed.

Although most clinically used antibiotics are derived from natural products, identifying new antibacterial molecules from natural product extracts is difficult due to the complexity of these extracts and the limited tools to correlate biological activity with specific molecules. Here, we show that bacterial cytological profiling (BCP) provides a rapid method for mechanism of action determination on plates and in complex natural product extracts and for activity-guided purification. We prepared an extract from Bacillus subtilis 3610 that killed the Escherichia coli lptD mutant and used BCP to observe two types of bioactivities in the unfractionated extract: inhibition of translation and permeablization of the cytoplasmic membrane. We used BCP to guide purification of the molecules responsible for each activity, identifying the translation inhibitors bacillaene and bacillaene B (glycosylated bacillaene) and demonstrating that two molecules contribute to cell permeabilitization, the bacteriocin subtilosin and the cyclic peptide sporulation killing factor. Our results suggest that bacillaene mediates translational arrest, and show that bacillaene B has a minimum inhibitory concentration 10 × higher than unmodified bacillaene. Finally, we show that BCP can be used to screen strains on an agar plate without the need for extract preparation, greatly saving time and improving throughput. Thus, BCP simplifies the isolation of novel natural products, by identifying strains, crude extracts and fractions with interesting bioactivities even when multiple activities are present, allowing investigators to focus labor-intensive steps on those with desired activities.

Most (75%) of the anti-infectives that save countless lives and enormously improve quality of life originate from microbes found in nature. Herein, we described a global visualization of the detectable molecules produced from a single microorganism, which we define as the 'molecular network' of that organism, followed by studies to characterize the cellular effects of antibacterial molecules. We demonstrate that Streptomyces roseosporus produces at least four non-ribosomal peptide synthetase-derived molecular families and their gene subnetworks (daptomycin, arylomycin, napsamycin and stenothricin) were identified with different modes of action. A number of previously unreported analogs involving truncation, glycosylation, hydrolysis and biosynthetic intermediates and/or shunt products were also captured and visualized by creation of a map through MS/MS networking. The diversity of antibacterial compounds produced by S. roseosporus highlights the importance of developing new approaches to characterize the molecular capacity of an organism in a more global manner. This allows one to more deeply interrogate the biosynthetic capacities of microorganisms with the goal to streamline the discovery pipeline for biotechnological applications in agriculture and medicine. This is a contribution to a special issue to honor Chris Walsh's amazing career.

Coral reefs are intricate ecosystems that harbor diverse organisms, including 25% of all marine fish. Healthy corals exhibit a complex symbiosis between coral polyps, endosymbiotic alga, and an array of microorganisms, called the coral holobiont. Secretion of specialized metabolites by coral microbiota is thought to contribute to the defense of this sessile organism against harmful biotic and abiotic factors. While few causative agents of coral diseases have been unequivocally identified, fungi have been implicated in the massive destruction of some soft corals worldwide. Because corals are nocturnal feeders, they may be more vulnerable to fungal infection at night, and we hypothesized that the coral microbiota would have the capability to enhance their defenses against fungi in the dark. A Pseudoalteromonas sp. isolated from a healthy octocoral displayed light-dependent antifungal properties when grown adjacent to Penicillium citrinum (P. citrinum) isolated from a diseased Gorgonian octocoral. Microbial MALDI-imaging mass spectrometry (IMS) coupled with molecular network analyses revealed that Pseudoalteromonas produced higher levels of antifungal polyketide alteramides in the dark than in the light. The alteramides were inactivated by light through a photoinduced intramolecular cyclization. Further NMR studies led to a revision of the stereochemical structure of the alteramides. Alteramide A exhibited antifungal properties and elicited changes in fungal metabolite distributions of mycotoxin citrinin and citrinadins. These data support the hypothesis that coral microbiota use abiotic factors such as light to regulate the production of metabolites with specialized functions to combat opportunistic pathogens at night.

The potential of the diverse chemistries present in natural products (NP) for biotechnology and medicine remains untapped because NP databases are not searchable with raw data and the NP community has no way to share data other than in published papers. Although mass spectrometry (MS) techniques are well-suited to high-throughput characterization of NP, there is a pressing need for an infrastructure to enable sharing and curation of data. We present Global Natural Products Social Molecular Networking (GNPS; http://gnps.ucsd.edu), an open-access knowledge base for community-wide organization and sharing of raw, processed or identified tandem mass (MS/MS) spectrometry data. In GNPS, crowdsourced curation of freely available community-wide reference MS libraries will underpin improved annotations. Data-driven social-networking should facilitate identification of spectra and foster collaborations. We also introduce the concept of 'living data' through continuous reanalysis of deposited data.

An amendment to this paper has been published and can be accessed via a link at the top of the paper.