Peeling Back the Layers of Escherichia coli Chromatin Architecture

Abstract

Understanding how chromosomes are organized and compacted has been a longstanding question in biology. How does a cell as small as two microns fit an entire genome 1000x its length inside? As most prokaryotes are not known to contain histones, and cellular processes, such as transcription and translation, are co-occurring in the same compartment as the DNA, there needs to be some coordinated manner and organized structure in which the DNA is compacted but still allows processes to occur. Escherichia coli, a gram-negative bacterium, is the most studied organism in the world. While there is an immense amount of knowledge on E. coli, the smaller nucleoid of this bacterium simply has not been as systematically studied yet as eukaryotic chromatin structure, in terms of DNA contacts and DNA accessibility. In this thesis, I revisit previously answered questions about the E. coli chromosome, through a different lens, reconciling discrepancies and filling knowledge gaps. I worked to develop a massively multiplexed sequencing method (Mu-seq) that measures chromosomal contacts, without the use of external crosslinkers, as is the current standard in the field. This method maps DNA:DNA contacts, made physically inside the cell, with the use of a barcoded version of transposon Mu. This method yields a finer-grain understanding of long-range contacts, those made over kilobase scales, and how contacts correlate with other cellular processes, such as transcription. I find that long-range interactions, which have been previously undetected by crosslinking-based methods, may form due to shared gene functions, and the need for coordinated regulation. I further demonstrate that chromatin accessibility shapes the interaction landscape through DNA-bound protein occupancy and supercoiling dynamics. Accessibility plays an important role in the organization of the E. coli nucleoid, as RNA polymerase needs to access tightly supercoiled and organized DNA for transcription. We chose to investigate what specific factors might influence the DNA contacts and organization patterns I detected from the Mu-seq method. I adapted an existing eukaryotic assay to function in a prokaryotic system, to measure DNA accessibility and protein occupancy at the single-molecule level, employing long-read nanopore sequencing. The approach uses methylation of DNA unoccupied by proteins, followed by long-read sequencing to detect methylated adenines along individual molecules -- I call it SHREK-seq (Single-molecule High-REsolution occupancy of proKaryotic chromosomes sequencing). Using this method, I was able to show the methylation signal, which labels open, unoccupied chromatin, to overlap with known extended protein occupancy domains (EPODs). I also showed how SHREK-seq method could detect more specific protein signals from Lrp, a global regulator in E. coli, and genetic perturbations resulted in an expected increase in methylation signal, corresponding to a loss of protein binding at known sites. My aim with this method was to obtain single-molecule binding patterns, which would resolve questions such as the phasing of the protein, how various oligomeric states bind differently at the same locus, and whether we resolve other aspects of chromosome dynamics which are currently undetected by population-averaged methods. Altogether, this dissertation provides two useful tools for measuring DNA contacts within the E. coli genome and detecting protein-occupied DNA regions, both of which not only advance our knowledge of the E. coli genome but may also be used in other systems.