Adaptation in Salmonella Typhimurium during infection

Abstract

Salmonella infects millions of people worldwide every year. To cause disease, Salmonella Typhimurium (S. Tm) and other enteric bacterial pathogens must adapt to ever-changing conditions during and between infection. This study aims to provide insights into how these pathogens adapt, survive, and evolve under various conditions, which can help in developing new therapeutic strategies to combat infections and antibiotic resistance. Many bacterial virulence factors are recognized by immune systems. As a result, the fitness effect of expression by the pathogen, greatly depends on the environmental context. One previously known example is that of fliC. Expression of fliC enables flagella-mediated motility and is crucial for S. Tm to access nutrients and reach the intestinal epithelial layer which separates it from the lamina propria. At the same time, flagella are recognized by several players of the hosts immune system. As a result, fliC expression in the intestinal lumen was selected, but is halted at other sites such as the lamina propria, perhaps because the risk of detection would make expression a disadvantage. The conflicting selective pressures acting on fliC are detectable in the form of an increased frequency of premature stop codons in pangenomes. In the first research chapter we investigated another gene that has an increased frequency in premature stop codons, and mutated in 70 % of clones in a murine longterm infection model. Mutants deficient for the gene showed a competitive advantage for the colonization of the murine intestinal lumen. The gene, glpT, encodes for a sn-Glycerol-3-phosphate (Gly3P) and phosphate antiporter. Using competitive infections, we found that glpT expression is harmful during growth in the intestinal lumen, where the phosphate concentration is high. We show that when the extracellular Pi concentration exceeds that of Gly3P, GlpT can import phosphate, and that additional Pi import can reduce the fitness of S. Tm growing in the intestinal lumen. In contrast, glpT expression seems to increase S. Tm’s intracellular replication inside macrophages, a niche that is crucial for virulence for S. Tm and many other facultative intracellular pathogens. Our work demonstrates that metabolic genes experience antagonistic pleiotropy, due to the strong changes in metabolic landscapes and could explain why the frequency of different variants of enteric pathogens change over the course of infection. The understanding of antagonistic pleiotropy acting on virulence genes has greatly increased our understanding of the evolution of pathogens and contributed to the design of a therapeutic approach that directs the evolution of an attenuated S. Tm variant. We envision similar approaches for the future, were enteric bacterial pathogens are directed to evolve variants that are metabolically vulnerable. GlpT is one of three routes of Gly3P import in S. Tm. Once imported, the important model organism S. Tm SL1344 has been shown to have a high catabolic potential for glycerol, due to high protein concentrations and potential turnover rates. Multiple glycerol metabolism intermediates are toxic, including Gly3P, methylglyoxal and acrolein, which raised the question if SL1344’s high catabolic potential for glycerol could be exploited by creating conditions that favor intoxication. In the second research chapter, we show that strains deficient for S. Tm’s Gly3P dehydrogenases are attenuated during colonization of the murine intestinal tract, which is exacerbated by glycerol supplementation. Additionally, we find that pduC attenuates S. Tm in vitro and in vivo, possibly due to the production of acrolein from glycerol. Our findings suggest that S. Tm SL1344 is susceptible to intoxication by glycerol intermediates and that therapeutic strategies aimed at exploiting the metabolic vulnerabilities of pathogens have the potential to be valuable complementary therapeutic approaches. Despite our knowledge that adaptation is crucial for enteric bacterial pathogens, capturing their transcriptome is challenging as high levels of inflammation result in the degradation of RNA. In the third research chapter, we explore the potential of a novel technology called Record-seq to capture S. Tm’s transcriptomic history. The technology relies on a CRISPR system found in a human commensal that features a fusion of a reverse transcriptase to the Cas1—Cas2 complex. Heterologous overexpression of the RT-Cas1–Cas2 complex, enables recording information about endogenous transcript abundance, stored in the form of DNA spacers within CRISPR spacer arrays. We show that Record-seq can be used to detect S. Tm’s transcriptional histories during infection. We validate its accuracy by comparing transcriptional data for strains with known transcriptional consequences: a hilA-deficient strain, and a strain deficient for S. Tm two type-3-secretion systems. We further demonstrate Record-seq’s potential to assign transcriptional information to the strain of origin in the presence of highly related strains. Doing so is not possible with RNA-seq as mRNA transcripts are too similar to be able to unambiguously assign them to one of the strains. Here, we employ Record-seq to investigate the transcriptional adaptations of S. Tm in competition of E. coli 8178, which can effectively compete against S. Tm and thus has potential as a probiotic strain. Collectively we believe that Record-seq has potential to further our understanding of microbe-host interactions. For the use in S. Tm basic research, Record-seq should be optimized by increasing spacer acquisition efficiency and decreasing the cost of carrying the construct. Beyond S. Tm and microbe-host interactions, sentinel cells have longterm potential to serve as diagnostic tools in variety of disease indications such as gastrointestinal cancers, chronic inflammation and metabolic diseases, which could help in clinical decision-making and drug development. Throughout the thesis the focus lies in understanding the dynamic adaptations and interactions of enteric bacterial pathogens in various infection-associated environments. Our main findings are 1. There are examples of antagonistic pleiotropy in metabolic genes due to the differences in metabolic environments, 2. It is possible to create conditions that result in S. Tm’s in vitro and in vivo intoxication on glycerol metabolism intermediates and 3. Record-seq has the potential of capturing S. Tm’s transcriptional history during infection and in competition with the probiotic E. coli 8178, where spacer acquisition efficiency should be improved and its accuracy further validated. The studies emphasize the importance of considering environmental fluctuations, genetic mutations, and metabolic adaptations as factors that shape the evolutionary trajectories and pathogenicity of enteric pathogens. In the longterm, our aim is to develop new therapeutic and diagnostic strategies that can effectively combat infection by enteric bacterial pathogens and mitigate the risk posed by the antimicrobial resistance crisis.