Abstract:

Chronic infections place a significant burden on healthcare systems, requiring over $20 billion in treatment annually in the United States alone. Notably, chronic infections are frequently polymicrobial and are often recalcitrant to antibiotic treatment, through a process termed “synergy”. Despite the clinical importance, many key features of bacterial physiology in chronic infections, including the molecular mechanisms and impacts of microbe-microbe interactions, remain understudied, in part due to the challenge of assessing bacterial physiology in the human host. My work aims to address this challenge, focused on the human pathogen Staphylococcus aureus, a leading cause of human infection worldwide and a significant cause of both morbidity and mortality. Specifically, I ask foundational questions about how S. aureus, causes disease and persists in human infection, both by using validated experimental models of infection and by directly analyzing human clinical samples. In my previous work, I found co-infection altered the requirement for ~6% (192 genes) of the S. aureus genome when compared to mono-infection in three different infection models, indicating global changes in S. aureus metabolism in response to another microbe, highlighting the importance of microbe-microbe interactions in mediating bacterial physiology in vivo. In addition, I recently performed the first large scale assessment of S. aureus physiology in situ in chronic human infection and found remarkable conservation of S. aureus gene expression in the cystic fibrosis (CF) lung across patients using RNA-seq, despite numerous epidemiological differences. With this data, I inferred the metabolic state and nutritional environment of S. aureus in CF sputum including iron scarcity, carbohydrate use, and virulence factor production. Further, through a machine learning framework, I defined a ‘human CF lung transcriptome signature’ – 32 genes whose transcription distinguishes human CF sputum from other in vivo and in vitro environments, primarily consisting of genes involved in metabolism and virulence – and was able to apply these findings to make an in vitro model more similar to CF infection. Ongoing work further explores mechanisms that allow S. aureus to establish and persist in chronic polymicrobial infections, including CF, osteomyelitis, and chronic wounds, centered on the role of microbe-microbe interactions.

Abstract:

Protein glycosylation, one of the most complex post-translational modifications, plays central roles in a variety of cellular processes in prokaryotes. Elucidating its involvement in biofilm formation is crucial for a detailed understanding of antibiotic resistance and pathogenicity mechanisms. Yet the complexity and variability of glycoproteins in prokaryotes have made their system-wide analysis, thus far, virtually unachievable. Here, I will present an interdisciplinary approach that combines bioinformatics, comprehensive glycoproteomics and phenotypic characterizations for the functional analysis of prokaryotic glycosylation. This includes the development of universally applicable bioinformatic tools that are suitable to analyze the plethora of glycan compositions existing in prokaryotes, as well as the initiation of the Archaeal Proteome Project, a community-effort that combines proteomics datasets across a broad range of experimental conditions to harvest the wealth of information buried within them. Together with an in-depth proteomic analysis of glycosylation pathway mutants from the model archaeon Haloferax volcanii, this has led to the identification of the largest archaeal glycoproteome described so far. It also revealed the concurrence of two independent N-glycosylation pathways that can modify the same glycosylation sites. A variety of phenotypic assays of mutants defective in N-glycosylation pathways or glycosylated proteins were further used to show the involvement of glycosylation in crucial cellular processes such as biofilm formation and cell shape. The establishment of these approaches for H. volcanii not only provided new insights into the extent, complexity, and roles of glycosylation in archaea, but their applicability to a multitude of prokaryotes also paves the way for functional glycoproteomics in biofilm forming pathogens such as Pseudomonas aeruginosa.

Abstract:

Microbial populations can adapt rapidly due to their vast supply of mutations.  To understand what makes these mutations adaptive, we need a systems-level approach that accounts for the wide range of biological traits that mutations can affect.  Using a combination of high-throughput experimental methods and computational modeling, I will discuss two systems-level approaches to microbial evolution.  First, I will show that both genetic and non-genetic mechanisms lead to substantial covariation in microbial life-history traits, such as the maximum growth rate in an environment and the lag time when transitioning between environments.  In particular, I will demonstrate the potential for tradeoffs in these traits and how such tradeoffs produce rich evolutionary and ecological phenomena, including stable coexistence of multiple strains and higher-order ecological effects.  Second, I will describe our development of a high-diversity DNA barcode library to track adaptation at high-resolution in bacterial populations with large numbers of simultaneously segregating mutations.  We apply this method to study adaptation to ultra-low concentrations of antibiotics, which are believed to be common in both clinical and natural environments.  We show that these ultra-low concentrations have distinct effects on evolution even without detectable resistance, and that adaptation to these conditions is highly predictable at the population level.  Furthermore, the fate of individual lineages can also be highly predictable in some conditions, depending on the relative contributions of standing genetic variation and de novo mutations to adaptation.  Altogether these examples demonstrate the power of systems-level approaches to elucidating the evolution of microbes.

Abstract:

Bacterial gene expression is highly heterogeneous even in isogenic bacteria grown in the same conditions; bacteria differentiate into subpopulations that may assume different roles for the survival of community. Population-level gene expression measurements are insufficient to resolve such phenotypic states which have been only discovered through single-cell methods. Using quantitative single-cell time-lapse microscopy, we discovered a novel microbial gene regulatory strategy in the model organism Bacillus subtilis accomplished through chromosomal arrangement of key genes regulating the heterogeneous sporulation cell fate. Methods such as fluorescence microscopy, however, are typically based on reporters allowing to measure only a limited set of genes at a time and requiring tractable model organisms. To address these limitations, I developed microSPLiT, a scalable single-cell RNA sequencing method tailored for bacteria. MicroSPLiT revealed a plethora of gene expression states in >25,000 single B. subtilis cells, including rare and unexpected cell states that remained hidden at a population level. With high scalability and resolution, microSPLiT is an emergent technology for single-cell gene expression studies of complex natural and engineered microbial communities.

Please email mibcoord@uga.edu for meeting info.

Abstract:

Many environmentally and clinically important fungi are suspectable to assault from bacterially-produced, redox-active molecules called phenazines. Despite being vulnerable to phenazine-assault, many fungi are found living within microbial communities that contain phenazine producers. Because many fungi cannot withstand phenazine challenge, but some bacterial species can, I hypothesized that a bacterial partner may be responsible for protecting fungi in these communities. In the first soil sample collected, I co-isolated several such physically associated pairings that appear to represent this hypothetical partnership class. I discovered the novel species Paraburkholderia edwinii and demonstrated that, when in the presence of a co-isolated Aspergillus species, the bacterium will protect its partner fungus from phenazine assault by sequestering the molecule and acting as a toxin sponge. When challenged with phenazines, P. edwinii changes its morphology to create bacterial aggregates within the growing fungal colony, and the bacterium creates an anoxic and reducing environment, conditions that would be expected to limit the toxicity of phenazines. A mutagenic screen revealed this program to be partially regulated by the stress-inducible transcriptional repressor HrcA, and the deletion of the hrcA gene results in a strain more capable of providing protection against phenazine assault. One relevant stressor is fungal acidification in response to phenazine challenge, and when challenged with acid P. edwinii can be made to sequester phenazines, triggering the protection response as though its fungal partner were present even when absent.

Paraburkholderia species collected from geographically diverse sites also demonstrate this protective ability toward several groups of fungi when paired in the lab, including plant and human pathogens. Finally, efforts to reproduce co-isolations of this class from the rhizosphere of citrus trees revealed such protective interactions to be common, highlighting the potential widespread nature of such mechanisms. These results have consequential implications for how microbial communities in the rhizosphere as well as plant and human infection sites are policed for membership to include or exclude certain fungal groups.

See department email for meeting link and password.

See department email for meeting link and password.

"Chlamydomonas reinhardtii as a model to study cilia-related disease"

by Dr. Karl Lechtreck, Department of Cellular Biology, University of Georgia

Abstract:

Bacteria use extracellular appendages called type IV pili (T4P) for diverse behaviors including DNA uptake, surface sensing, virulence, protein secretion, and biofilm formation. Dynamic extension and retraction of T4P is essential for their function in these behaviors, yet little is known about the molecular mechanisms controlling these dynamics. Furthermore, due to difficulties in visualizing T4P in live cells, their exact function in many of these processes has remained unclear. Through the development of a labeling method to visualize T4P in live cells in real time, our work has addressed multiple outstanding questions related to T4P biology. Using this labeling method, we defined the mechanism by which individual cells use T4P to take up DNA during natural transformation. We furthermore developed the highly naturally transformable species Acinetobacter baylyi as a new model to dissect the molecular mechanisms of T4P dynamics and T4P localization, with implications for how these structures may contribute to multicellular interactions. This work provides insight into the mechanisms that govern diverse microbial behaviors important for bacterial physiology through direct observation of the T4P appendages that mediate them.