Physical mapping of bacterial genomes

Abstract

The elucidation of gene order in chromosomes, influencing both linkage and coregulation, is at the heart of genetic analysis. Indeed, the characterization of any mutation or cloned gene traditionally includes the determination of its position on the chromosomal map. Further, questions about the structural maintenance of the chromosome itself can be approached by comparative physical and genetic mapping of chromosomes. However, viewing the genome in one dimension may not be sufficient to understand genome organization comprehensively, if the object of evolutionary conservation is its threedimensional structure. Beginning in the early 1980s, several experimental approaches made it possible to map megabase-size DNAs: pulsed-field gel electrophoresis (PFGE)-related methods, gene encyclopedia construction, and total DNA sequencing. The ideal targets for these physical approaches are bacterial genomes, with sizes ranging from 600 kb for Mycoplasma genitalium (22) to 9.5 Mb for Myxococcus xanthus (14). Aligning genes to physical maps generated by these approaches has surpassed traditional techniques for constructing bacterial genetic maps. Many chromosomal maps of different bacteria have been produced by traditional genetic methods. Some of these maps have hundreds of genetic markers: there are 1,717 on the last version of the Escherichia coli map, of which 300 were located by physical mapping (71). Purely genetic maps have a few serious disadvantages. First, their construction is labor-intensive. Also, because genetic maps provide distances measured in units derived from recombination frequencies, they are sensitive to recombination hot spots. The physical distance equivalent to 1% of the E. coli linkage map varies from 33 to 60 kb (80). The genetically linear chromosomal map of Caulobacter crescentus turned out to be physically circular, with proximal (unlinked) markers being only 400 kb apart. This genetic separation is ascribed to a recombination hot spot in the Ter region (31). The basic questions mentioned above, plus the availability of several convenient mapping techniques, have resulted in the construction of physical maps of many bacterial genomes. More than 90 studies dealing with physical mapping of bacterial genomes were published in the last 7 years. The five purely genetic maps produced during the same period is a fair reflection of their relative difficulty. In this minireview, we summarize the principal methods used for map construction, salient features of the maps produced, and applications of the gene encyclopedias for questions of genome evolution, rapid mapping of cloned genes, and global studies of gene expression. We also refer readers to a recent article by Cole and Saint Girons (21), published as this minireview was being completed. They cover much of the same ground and include tabular compilations of methods used and genome sizes. Their article should also be consulted for speculations on the evolution of genome size and the conservation of genome architecture.