Functional genomics of microbial pathogens

Sequencing bacterial genomes over the past two decades has opened a new era in the analysis of pathogenic bacteria. Analyses of genomic sequences provided important insights into evolution of pathogenicity, antibiotic resistance and epidemicity of important human, animal or plant pathogens. Genomic comparison between pathogenic bacteria and their less harmful relatives showed that virulence factors may be acquired by horizontal transfer of pathogenicity islands. Evidence has been provided that virulence, host tropism and adaptability may be further enhanced by gene loss as shown for genera Mycobacteria, Rickettsia, Yersinia and Shigella [1]. Likewise, it has been experimentally confirmed that micro-evolutionary adaptation of Staphylococcus aureus during its transition from a colonizer to invasive pathogen includes accumulation of mutations that lead to loss of protein function [2]. Therefore, in addition to the presence of virulence and anti-virulence genes, several features of the bacterial genome such as the G+C content, genome size and the proportion of genes encoding specific functions could help identify pathogens and assess their intrinsic virulence. Genomic information provides a background for additional high-throughput functional studies such as in silico metabolic modelling [3] and wet-lab experimentation at the RNA and protein levels. Conversely, metabolic modelling and metabolomics may suggest the presence of gene functions not identified by sequence comparisons. Transcriptome studies of microbial pathogens using microarrays and RNA sequencing are particularly focused on changes in gene expression (i) in response to different exogenous stimuli, and (ii) due to mutations affecting bacterial virulence, gene regulation and resistance to antibiotics. Construction of an ORFeome showed great potential in proteomics research, by providing a basis to study protein localization, secretion, translocation as well as interactions between proteins or between proteins and other ligands. Overexpressed proteins are used to immunize mice, and the resulting immune sera are tested for their potential to provide protection against the pathogen. Such an approach resulted in the development of an effective vaccine against Neisseria meningitidis serogroup B [4]. Genome-wide targeted gene inactivation has been used to construct mutants and identify those with an altered phenotype. Targeted inactivation has been difficult to do in bacteria showing barriers to transformation. Two papers in this issue reported advancements in transformation of Staphylococcus aureus [5] and Chlamydia trachomatis [6], two important pathogens that were difficult to genetically engineer until recently. Although the role of bacteriophages in the evolution, virulence and antibiotic resistance of bacteria has been recognized, only a few functional genomics studies of these viruses have been performed so far. Remarkable progress has been made in predicting prophage regions in bacterial genomic sequences [7] but significant work remains to be done to elucidate the life cycle of bacteriophages. Such studies are important for use of bacteriophages and their proteins in bacterial identification and infection control. For the investigation of genotype–phenotype associations, it would be of great interest to systematically perform standardized phenotypic tests for each sequenced strain. Correlations between the genotype and phenotype may be improved by considering these entities as continuous instead of categorical values [8]. On the other hand, because of the simultaneous assessment of a high number of genetic variables, for certain analyses it may be advantageous to exclude non-informative observations, while those which are strongly correlated may be combined. Several authors point to a need to integrate data from different –omics approaches, e.g. transcriptomics, proteomics and metabolomics, for both the pathogen and the host, in order to better understand their interactions and their co-evolution. To mimic conditions encountered by a pathogen during infection, functional genomics uses animal models, human blood or cell culture. Studies of bacterial pathogens during human infections present challenges such as low bacterial load and, as a consequence, presence of host molecules that may heavily outweigh the bacterial ones. This limitation may be overcome by using high-throughput techniques such as RNA-seq, which generates a sufficient number of sequence reads so that those of bacterial origin can be detected within an ocean of host-derived sequence reads. Then, the sequence reads may be mapped to bacterial and human genomic sequences, thus reflecting the host–pathogen interaction. Holistic approaches similar to those used to study individual pathogens may also be applied to bacterial communities colonizing the human body. It has been shown that several human diseases and disorders correlate with changes in microbiota profiles [9], suggesting that bacterial communities may play a causative role of disease. In conclusion, the expanding field of functional genomics provides powerful tools and insights for assessing the contribution of pathogens and microbial communities to disease, as well as the characteristics of the host response, and will contribute to development of new prevention and therapeutic strategies.