Using genetic tools and high-throughput screening to understand proteolysis in bacteria
Date
2023
Authors
Journal Title
Journal ISSN
Volume Title
Publisher
University of Delaware
Abstract
All bacteria possess multiple ATP-dependent proteases that aid in the regulated destruction of cytosolic proteins. These enzymes regulate discrete cellular pathways and enforce protein quality control by degrading damaged or misfolded proteins. In many pathogenic bacteria ATP-dependent proteases regulate virulence factors or are strictly essential for viability. Consequently, these enzymes have emerged as attractive drug targets, particularly in the global pathogen Mycobacterium tuberculosis. Therefore, it is of utmost importance to elucidate the function and physiological roles of ATP-dependent proteases in to inform drug discovery efforts. One important aspect of proteolytic function is the regulation of substrate recognition. ATP-dependent proteases selectively degrade substrate proteins and minimize wasteful or off-target proteolysis by ignoring non-substrate proteins through a variety of paradigms. Many substrates are directly recognized via short, unstructured sequences termed degrons. While a handful of degrons have been discovered and are well-characterized, the overarching rules governing the sequence-specific recognition of substrates are poorly understood. There remains a vast sequence space that has been unexplored by prior unbiased screening attempts to define substrate specificity. In this dissertation, we seek to develop and implement novel, high-throughput screening approaches in vivo to systematically interrogate degron specificity and protease function in bacteria. ☐ In Chapter 2, we screened millions of possible C-terminal degron sequences in E. coli using a toxin-based selection approach that coupled proteolysis to cell survival. This strategy ties degron strength to fold-enrichment in a counter-selective competitive growth assay. Using Illumina sequencing and bioinformatic analysis, we identified over 1000 tags that promoted proteolysis of the toxin and permitted cell growth in wild-type cells. Surprisingly, the majority of these tags bore striking resemblance to the ssrA tag, a well-characterized degron that is present throughout bacteria that is recognized by the ClpXP protease. Moreover, these sequences were not enriched in E. coli strains in which the ClpXP was disrupted. Together, this work highlights the ssrA tag as uniquely potent degron and provides a novel degron screening method that may be applied in clinically relevant bacteria. ☐ In Chapter 3, we aimed to overcome limitations identified in our toxin-based strategy from Chapter 2 by instead coupling proteolysis to cellular fluorescence. By replacing the toxin with a fluorescent mCherry substrate, we demonstrated the use of fluorescence-activated cell sorting (FACS) to select for degron-dependent proteolysis in E. coli. Proof-of-concept experiments in chapter 2 that tied degron strength to fold-enrichment were recapitulated here over several rounds of cell sorting. Additionally, we showed that the fluorescent signal resolution can be improved in a dose-dependent manner by supplementing ClpX and ClpA on an additional inducible plasmid. While efforts to implement this strategy on a large scale were unsuccessful, this work nonetheless lays the foundation for a robust screening platform that requires further development. ☐ In Chapter 4, we sought to probe the sequence-function relationship in ClpX by inverting the screening logic of the two-plasmid expression system described in Chapter 3. Recent cryo-EM structures of ClpXP bound to a ssrA-tagged substrate have underscored the importance of axial pore loops in ssrA recognition. We utilized our FACS-based workflow to screen libraries of ClpX in which two of these loops – RKH and GYVG – were randomized separately. Interestingly, our screening approach generally selected for ClpX variants with weakened proteolytic activity. While we inadvertently biased selection away from the wild-type sequence, our results highlight critical residues within these loops and provide insight into the effect that subtle side chain variations have on proteolysis of ssrA-tagged substrates. With more gentle selection pressure, this strategy has the potential to be applied for engineering unfoldases with altered or enhanced degron specificity. ☐ In Chapter 5, we described the use of CRISPR-based techniques to aid in the study of the ClpC1P1P2 protease in Mycolicibacterium smegmatis, a non-pathogenic surrogate for Mycobacterium tuberculosis. Using an established platform for CRISPR interference (CRISPRi) in mycobacteria, we showed that transcriptional knockdown of ClpC1 arrests cell growth. We sought to appropriate CRISPRi to test variants of ClpC1 for function in vivo. We showed that introducing a second copy of ClpC1 that is immune to targeting by CRISPRi can rescue the cells from transcriptional interference of the genomic copy. This altered CRISPRi strategy enables rapid screening of ClpC1 mutants and other essential genes in mycobacteria.
Description
Keywords
Bacterial proteolysis, Flow cytometry, Gene editing, High-throughput screening, Mycolicibacterium smegmatis, Protein engineering