Bacteria are classified into genera and species according to a binomial Linnean scheme similar to that used for higher organisms. For example, in the case of Staphylococcus au-reus, Staphylococcus is the name of the genus and aureus is the species designation. Some genera with common characteristics are further grouped into families. However, bacterial classification has posed many problems. Morphologic descriptors are not as abundant as in higher plants and animals, there is little readily interpreted fossil record to help establish phylogeny, and there is no elaborate developmental process (ontogeny) to recapitulate the evolutionary path from ancestral forms (phylogeny). These problems are minor compared with others: bacteria mutate and evolve rapidly, they reproduce asexu-ally, and they exchange genetic material over wide boundaries. The single most important test of species, the ability of individuals within a species to reproduce sexually by mating and exchanging genetic material, cannot be applied to bacteria. As a result, bacterial tax-onomy developed pragmatically by determining multiple characteristics and weighting them according to which seemed most fundamental; for example, shape, spore formation, Gram reaction, aerobic or anaerobic growth, and temperature for growth were given spe-cial weighting in defining genera. Such properties as ability to ferment particular carbo-hydrates, production of specific enzymes and toxins, and antigenic composition of cell surface components were often used in defining species. As presented, such properties and their weighting continue to be of central importance in identification of un-known isolates in the clinical laboratory, and the use of determinative keys is based on the concept of such weighted characteristics. These approaches are much less sound in estab-lishing taxonomic relationships based on phylogenetic principles.
The recognition that sound taxonomy ought to be based on the genetic similarity of organ-isms and to reflect their phylogenetic relatedness has led in recent years to the use of new methods and new principles in taxonomy. The first approach was to apply Adansonian or numeric taxonomy, which gives equal weighting to a large number of independent char-acteristics and allocates bacteria to groups according to the proportion of shared character-istics as determined statistically. Theoretically, a significant correspondence of a large number of phenotypic characteristics could be considered to reflect genetic relatedness.
A more direct approach available in recent years involves analysis of chromosomal DNA. Analysis can be somewhat crude, such as the overall ratio of A – T to G – C base pairs; differences of greater than 10% in G – C content are taken to indicate unrelatedness, but closely similar content does not imply relatedness. Closer relationships can be as-sessed by determining base sequence similarity, as by DNA – DNA hybridization, in which single strands of DNA from one organism are allowed to anneal with single strands of another. Some clinical laboratory tests have been devised based on the ability of DNA from a reference strain to undergo homologous recombination with DNA from an un-known isolate . However, overwhelmingly the molecular genetic tech-nique that is introducing the greatest change in infectious disease diagnosis is the com-parison of nucleotide sequences of genes highly conserved in evolution, such as 16 S ribosomal RNA genes. So successful have been the deductions of phylogenetic related-ness based on these sequences that the absence of a fossil record is now regarded as in-significant. Part of the excitement in this field is that the use of polymerase chain reaction to amplify the DNA of cells has made it possible to identify even infectious organisms that cannot be cultivated in the laboratory.
The most startling recent advance in medical microbiology is indicated by the fact that in the few years since the printing of the previous edition of this book, the complete nu-cleotide sequences of the genomes of several dozen medically significant bacteria have been determined. Furthermore, advances in the technology of DNA sequencing promise the rapid determination of many more genomes in the next few years. It is difficult to overstate the significance of the present situation. First, comparison of virulent with non-virulent species of closely related bacteria is providing means to identify virulence genes, that is, genes responsible for the disease-producing capability of these bacteria. Second, thanks to the sequence information, the products of these genes can readily be produced and studied, and mutants can be prepared for genetic and functional analysis. Among the genes being discovered in this way are many organisms of hitherto unknown virulence, providing new information about the many molecular processes involved in pathogene-sis. Third, new information on virulence factors and how they work is suggesting new, ra-tional design of therapeutic and prophylactic agents to replace our current overreliance on natural antimicrobics and their chemical derivatives.
Finally, detailed genomic analysis of pathogens involves suggesting pathways of the evolution of important human and animal pathogens. Already, molecular genetic studies have uncovered the existence of pathogenicity islands (PAIs) within genomes — that is, groups of adjacent genes that encode functions important for colonization, invasion, avoidance of host defenses, and production of tissue damage. These PAIs exist not only within the chromosome of pathogens but also within the plasmids that assist in confer-ring virulence properties on the bacteria. As described, analysis of PAIs provides important clues to the origin of these gene clusters and to their transmis-sion between species. This should be a fertile area for understanding the evolution of pathogens.