Organization of the Prokaryotic Genome

This chapter summarizes the recent findings of bacterial genomics and comments on the themes and trends which are emerging. A variety of techniques and methods are available to construct physical maps, and those most commonly employed involve pulsed field gel electrophoresis (PFGE) of macrorestriction fragments generated by digesting intact genomic DNA, immobilized in agarose plugs, with rare-cutting enzymes. Hybridization techniques are often used to construct a map and to deduce the positions of genetic markers. In recent years significant effort has been devoted to developing direct-mapping techniques for large DNA molecules that do not require gel electrophoresis. Among the more promising of these are two new methods known as DNA combing and optical mapping, both of which make use of fluorescence microscopy and image analysis to visualize single DNA molecules. Overall, bacterial genomes range in size from about 0.6 to 9.4 Mb. In a recent review, it was suggested that there may be a relationship between the genome sizes and the lifestyles of bacteria. The genome size of Bacillus cereus strains varies from 5.5 to 6.3 Mb, and great diversity is seen in the number and organization of the chromosomes. Higher levels of gene expression might result from altered topological constraints acting on the DNA or from a more favorable location with respect to the origin of replication, leading to an increase in gene dosage.

This chapter begins by comparing the organization of DNA in bacteria and eukaryotes. The possibility that similar mechanisms are utilized by both eukaryotes and prokaryotes to separate or disentangle sister chromatids or daughter strands is outlined. It is suggested that transcriptional activity and resulting RNA-protein (hnRNP) particles may play a role in DNA strand separation in eukaryotic cells, just as coupled transcription-translation does in prokaryotes. Two characteristics of a bacterial cell, such as Escherichia coli, may contribute to the difficulty in understanding its mechanism of genome segregation: first, the occurrence of DNA synthesis throughout the whole cell cycle during rapid growth, and second, the lack of a unique centromere sequence, which is characteristic of the eukaryotic chromosome. In vitro studies have shown that phenomena like phase separation, monomolecular collapse, and intermolecular aggregation of the DNA into a condensed state can be induced by high concentrations of proteins (macromolecular crowded solution) and ions. Visualization of RNA transcription and DNA replication has shown that these two processes occur in hundreds of different domains scattered throughout the S-phase nucleus. The chapter discusses possible mechanisms used in bacteria for movement of DNA and for nucleoid segregation. It also focuses on separation of daughter strands in bacteria, and separation of sister chromatids in eukaryotes. The genomic organization may help in chromosome segregation, the expression of genes fulfilling a role in the fundamental process of daughter strand separation.

This chapter talks about bacterial genomes, primarily those of Escherichia coli and Salmonella typhimurium (proper name, Salmonella enterica serovar Typhimurium). For these bacteria, abundant information is available on evolutionary relationships, high-quality genetic maps exist (E. coli is completely sequenced), and there is extensive knowledge of mechanisms of recombination. Comparing the genomes of E. coli and S. typhimurium is therefore a natural starting point for discussing the forces which determine genome organization and stability in general. The degree to which ectopic exchanges between directly repeated sequences are RecA dependent varies with size and distance. Large chromosomal duplications are genetically unstable but are stabilized by recA mutations, implicating homologous recombination in their formation and loss. Genes expressed at high levels are generally located in the origin-proximal half of the chromosome, presumably because closeness to the origin of DNA replication gives a gene dosage effect. Tandem duplications, and their associated deletions and translocations, create novel sequence join points which potentially have selective value for the cell. Recombination between directly oriented repeat sequences can create a DNA fragment (linear or circular, depending on the mechanism of recombination) which can recombine with the chromosome. Inversions can occur between homologous short and long sequences. Most inversions isolated in E. coli and S. typhimurium are reported as having no significant effects on growth rate, but the few translocations made and tested in E. coli are associated with decreases in growth rate of up to a few percent.

Transposable elements (TEs) have been defined as DNA sequences able to insert at many sites in the genome. At present there are about 500 TEs identified in many different bacterial species. Most insertion sequence (IS) elements and transposons were discovered after their transposition into genes of interest. In bacteria the relative juxtapositions of genes are not necessarily important because most will be involved in producing tows-acting factors, which are able to fulfill their function irrespective of their location or arrangement in the genome. Nonreplicative (or "cut-and-paste") transposons are excised from the donor site by double-strand breaks and inserted at the target site without the duplication of the transposon sequences (e.g., IS10 and IS50). Some bacteriophages are also considered to group with the replicative transposons because they use transposition to replicate during the lytic phase of their life cycles. It should be noted that the transposons in the portable-homology rearrangements are no longer flanked by the same direct repeats of target sequences as they were before the rearrangements. Usually composite transposons have their IS elements in the inverted-repeat configuration so that homologous recombination only causes the inversion of the markers within the transposon. Conjugative transposons harbor within the same sequence the cellular (transposition) and the intercellular (conjugation) mobilities. In the few experiments where selection has been maintained for many generations, IS elements have been found to have a major effect on the genetic structure of the population.

Gyrase was the first type II enzyme discovered, and it remains unique for its ability to introduce negative supercoils into relaxed, positively or negatively supercoiled DNA at the expense of ATP binding and hydrolysis. An essential enzyme in bacteria, gyrase is critical for nearly all complex transactions that involve DNA, including recombination, replication, transcription, and chromosome segregation. The homeostatic supercoil regulation model was inspired by two observations. First, many promoters sense supercoiling levels. Second, expression of gyrase increases when the chromosome becomes relaxed, whereas the expression of topo I requires high supercoiling levels. Chromosome replication has three critical stages (initiation, elongation, and segregation), each with different supercoiling problems. For initiation, the two strands of oriC must separate to allow assembly of the replication machinery (the replisome). Transcription encompasses supercoiling problems similar in two respects to those of replication. First, transcription initiation requires unpairing of the DNA duplex, and supercoiling can influence this step as it does in initiation of DNA replication. Second, transcription is similar to replication in that movement of RNA polymerase generates temporary positive supercoiling ahead of, and negative supercoiling behind, the DNA segment that is being transcribed. The torsional effects of transcription have been studied with supercoil-sensitive promoters by measuring the formation of Z-DNA and by monitoring the extrusion of cruciforms. Homologous and site-specific genetic recombination, adaptive mutation, "supercoil regulated" gene transcription, gene order on chromosomes, and plasmid-chromosome replication segregation are all phenomena that are likely to be influenced by DNA dynamics.

This chapter evaluates the present knowledge about the degree of stability of the genome of enteric bacteria. It discusses about the forces which have contributed to maintaining stability, and considers the types of rearrangements which can and do occur even in those genomes generally considered to be stable. With the use of methods of physical analysis of DNA, and especially the introduction of the use of pulsed-field gel electrophoresis (PFGE), used first in Escherichia coli by Smith and colleagues, the genomes of many strains were determined and conservation of gene order was shown to be the rule. It talks about two modifications of PFGE methods, which allow the determination of genome structure in many strains, have been used in representative enteric bacteria. It focuses on the forces which would be expected to act in a conservative way to maintain gene order. It is possible that transspecies recombination due to conjugation or transduction followed by homologous recombination, though very rare, is so important that colinearity is an important advantage. The author concludes that the overall conservation of gene order within the enteric bacteria which was reported many years ago in comparisons of E. coli and Salmonella typhimurium has been confirmed by the analysis of the physical maps of many strains of Salmonella and other enteric bacteria determined by PFGE and by the comparison of nucleotide sequences.

This chapter describes methods for assessing the frequency of successful gene acquisition and the fraction of modern genomes that has been acquired by horizontal transfer of useful phenotypic information. Although both loss and acquisition strongly influence genome evolution, the authors suggest that the two processes are synergistic due to the limits on genome expansion. To assess the contribution of genomic flux to genome evolution and speciation, one must measure rates of gene loss and acquisition. All novel phenotypes were conferred by horizontally transferred genes. Therefore, genes for the diverse metabolic pathways have formed slowly at earlier times and not during the course of competitive invasion of novel ecological niches. Regardless of the relative rates of these two processes, it is clear that gene loss and acquisition have facilitated exploration of novel environments and allowed more rapid divergence of bacterial types in competitive situations. The authors propose that the prevalence of gene clusters stands as evidence that genomic flux has historically been a primary contributor to genome evolution. They have outlined a model for the evolution of bacterial genomes through the synergistic processes of gene acquisition and gene loss. The organization of genes into operons reflects the important role in bacterial evolution and speciation played by genomic flux—the development of bacterial genomes by gene loss and gene acquisition.

The delineation of groups of genes and proteins that trace back to common ancestors derives legitimacy and depth when functions are known and can be assessed for relatedness within any sequence related group. The goal in understanding protein evolution is the reconstruction of past events that have given rise to the inventory of extant proteins. Early ancestral organisms that existed before the separation of the three branches of the tree of life are believed to have possessed many of the functions of cell physiology and metabolism that are found in all living forms today. Molecular phylogeny tries to trace all the speciation events back to the last universal common ancestor. A few Escherichia coli proteins have been rearranged and have fused since their initial duplication and divergence, and the separate elements within complex proteins need to be identified and separated. In E. coli many of the paralogous modules were clearly distinguishable by inspection of the DARWIN output. To inquire into the descent of the proteins of a family from their ancestral sequence, the authors have studied further the history of individual families. With the many whole-genome-sequencing projects under way today and planned for the future, there will be an abundance of data to be analyzed further along these lines, with the ambitious aim of eventually reconstructing the paths of evolution of all proteins.

This chapter outlines how complete genome sequences promote the development of broad genome-wide or functional-genomics approaches for investigating individual genes to entire gene sets. Classical approaches had focused on a small portion of the Saccharomyces cerevisiae, Bacillus subtilis, and Escherichia coli genes. The power of genetically tractable model systems is that they facilitate the analysis of gene function at the level of the intact organism. In various adaptations of this method, PCR-generated replacement DNAs are used for constructing strains harboring simple point mutations or large deletions, making reporter gene constructs for any gene, inserting epitope or fluorescent tags in any protein, and studying the regulatory information controlling gene expression. Regulatable promoter replacement cassettes in combination with the availability of complete genome sequences will be important resources for studying gene function in yeasts and other eukaryotic systems. New methods for assessing gene expression, protein-protein interactions, and protein-nucleic acid interactions mean that it is now possible to use genome-wide approaches to study gene function. These global approaches should significantly accelerate progress toward a detailed understanding of the model organisms B. subtilis, E. coli, and S. cerevisiae. Genome-wide methods will generate exponentially increasing amounts of biological data. Effective systems of data storage, management, and analysis must keep pace with data generation so that the data and the tools necessary to apply these resources to individual research programs are accessible to all researchers.