Genetic material is highly dynamic. Genome stability is constantly challenged by endogenous and environmental agents that induce DNA lesions, genome rearrangements and other types of genotoxic stress. At the same time, localized and global mutation events and horizontal gene transfer (HGT) contribute to alterations in DNA. The balancing act between these forces has a major impact on the fitness of cells. While many of the changes generated can be deleterious, the variations they create are the bread and butter of evolutionary processes.

During infection of animals, bacterial pathogens encounter DNA-damaging compounds generated by the host as antibacterial defenses. Nitric oxide and reactive oxygen species are particularly important genotoxic agents and generate many DNA lesions. DNA lesions may also accumulate in genomes over time as byproducts of normal metabolic processes. Most such lesions are mended by DNA repair. However, there is still the potential for cumulative unrepaired DNA damage in bacterial genomes, which can play a critical role in generating mutator activity.

Our view of microbial DNA metabolism has to a large extent derived from analyses of the model organism Escherichia coli, which has served as a paradigm for DNA repair. However, genome projects have revealed a remarkable variation in gene content among different bacterial species, therefore reflecting differential needs and aims for genetic stability in disparate organisms. Mycobacterium tuberculosis, for instance, compensates for its genetic isolation by allowing twice as many nonsynonymous polymorphisms in DNA repair, recombination and replication genes as in other genes. Notably, no homologues of mismatch repair genes have been found in M. tuberculosis. This suggests a link between heightened genome-wide mutability and the development of adaptive traits such as antibiotic resistance in Mycobacterium, where HGT is rare.

Many bacterial genomes are beholden to the rapid generation of genome instability as a means of adaptation. A range of mechanisms, such as phase variation (promoter or gene slippage events generating modulation or ON–OFF regulation of gene expression), generate instability in specific, highly localized regions of the genome producing phenotypic variation, as opposed to the generalized mutator phenotypes affecting the genome globally, that can be induced by defects in some DNA repair genes. However, we are still in the early stages of appreciating how these pathways may shape bacterial adaptation, especially as seen through the narrow aperture of the commensal and pathogenic behaviour of bacterial species.

The lack of separation between germinal and somatic lines in prokaryotes implies that chromosomal changes are systematically passed to the offspring. Hence, rearrangements, mutations and HGT will all durably imprint the fitness of the bacterium and its descendants. This may impose complex evolutionary trade-offs, one of which pervades several articles in this issue: how adaptive is the increase in generation of genetic variation by mutation, intrachromosomal recombination or lateral transfer, knowing that many such changes are expected to be deleterious. Adaptive processes come with the eventual cost of a burden of deleterious mutations in other functions, making it unclear if such lineages survive in the long term. Alleviation of the cost of mutation might render possible a range of scenarios between the following two extremes. Damaged functions could be recovered by recombination (e.g. from DNA acquired by natural transformation), resulting in a descendant with improved fitness. Alternatively, the adaptive changes may be transmitted into a genetic background with higher fitness by HGT, in which case the mutator lineage disappears. From a biomedical point of view, in both cases repression of HGT (e.g. by targeting conjugation or transformation machineries) might result in bacteria with reduced fitness.

DNA repeats are both causes and consequences of genome plasticity. They are consequences of amplifications of genetic material and HGT. However, because they are targeted by recombination mechanisms, they are subsequently motors of genome dynamics, inducing deletions, amplifications, rearrangements and gene conversion. They also show the impact of recombination in genome evolution, not only by producing genome changes, but also because repeats are selectively recruited to allow targeted sequence variation in many bacteria, most notably in some important human pathogens. The association of repeats with certain types of genes (e.g. surface-exposed proteins) provides important clues to how pathogens diversify to escape the immune system, to out-compete other microorganisms or to adapt to new niches.

Transformation is the ability of many bacterial species to take up naked DNA from the environment and permanently incorporate the DNA into their genome by recombination. The well-developed molecular machines undertaking this process are of a diverse nature in Gram-positive and Gram-negative bacterial species, representing an excellent example of convergent evolution. Natural transformation in Streptococcus pneumoniae and Bacillus subtilis relies on a quorum-sensing circuit that regulates competence development by monitoring the concentration of species-specific peptide pheromones in the environment. New discoveries in the processing of internalized single-stranded DNA fragments into recombination products at the bacterial poles have led to advances in the understanding of the last steps of this dynamic machinery. In the pathogenic Neisseria species and Haemophilus influenzae, competence is less strictly regulated and DNA uptake requires the presence of short, specific DNA-uptake sequences distributed frequently throughout their respective genomes. Although a completely different strategy from the quorum-sensing systems, this mechanism also ensures that competent cells preferentially take up homologous conspecific DNA. The contrasting strategies employed by Gram-negative and Gram-positive organisms for gaining access to transforming DNA, leading to the same outcome, namely preferential uptake of its own DNA, represent an exciting level of mechanistic diversity for analogous biological systems. From being viewed as a marginal phenomenon, it has become increasingly clear that HGT by natural transformation is a powerful mechanism for generating genetic diversity, potentially causing problems for treatment of infectious diseases.

For many scientists interested in DNA metabolism, the ultimate goal of research in this field is to bridge the gap between basic science and clinical medicine. In this regard, ample findings provide evidence that DNA repair plays a direct role in aging and disease pathology in humans, including cancer and neurodegeneration. This thematic issue of FEMS Microbiology Reviews: Genome Dynamics focuses on current research on DNA metabolism in microorganisms. The link between genome (in)stability, DNA repair and HGT was also the topic of a recent conference, the Third Genome Maintenance Meeting (GMM3) ( held in Oslo, Norway, on August 30–September 2, 2008. This meeting was among the first multidisciplinary international initiatives to focus on and promote this emerging area of research. It also signalled an exciting new scientific era, in which researchers in complementary fields of science are working together to elucidate the mechanisms and aetiology of infectious diseases. The articles in this thematic issue demonstrate increased understanding of how genome instability, defects in DNA repair and HGT may contribute to antibiotic resistance, antigenic variation and virulence. We believe this research may help identify viable approaches to diagnose, treat and prevent infectious diseases.