DNA-based approaches used in the identification and classification of species of bacteria include DNA-DNA hybridization, DNA fingerprinting, and DNA sequencing. DNA-DNA hybridization, initially developed in the 1980s, is used to determine the similarity of DNA sequences from different organisms. The degree of similarity is reflected in the degree to which a strand of DNA from the organism of interest passively hybridizes with (attaches to) a single strand of DNA from a known organism. The less stable the hybridization is, the more quickly the DNA strands will dissociate when heated; hence, low DNA melting temperatures typically suggest low degrees of sequence similarity. DNA-DNA hybridization is most valuable for determining genetic relatedness at the genus and species levels.

DNA fingerprinting methods for bacterial identification center primarily on the use of the polymerase chain reaction (PCR). Repetitive element-PCR, for example, targets specific DNA segments that are repeated at random in the bacterial genome. The identification of repetitive elements is powerful, capable of resolving bacteria at intraspecies levels. DNA sequencing methods, including whole genome sequencing and multilocus sequence analysis (MLSA), likewise have proved useful in the identification of bacteria. MLSA, which entails DNA sequencing of subsets of so-called housekeeping (or conserved) genes, has been shown to provide resolution down to intraspecies levels.

16S rRNA analysis

Evolutionary relatedness can also be assessed through the sequencing of 16S rRNA, the gene that encodes the RNA component of the smaller subunit of the bacterial ribosome (16S refers to the rate of sedimentation, in Svedberg units, of the RNA molecule in a centrifugal field). The 16S rRNA gene is present in all bacteria, and a related form occurs in all cells. The 16S rRNA gene of E. coli is 1,542 nucleotides long, and some of its regions are double-stranded, while other regions are single-stranded. Single-stranded regions often form loops, because there is a lack of complementary bases on the opposing strand. Since 16S rRNA makes very specific contacts with many different ribosomal proteins and with other parts of itself, the pace at which spontaneous random mutation can change the sequence of the bases in the rRNA is slow. Any change in sequence at one site must be compensated for by another change elsewhere within the rRNA or in a ribosomal protein, lest the ribosome fail to assemble properly or to function in protein synthesis and the cell die.

Analysis of the 16S rRNA sequences from many organisms has revealed that some portions of the molecule undergo rapid genetic changes, thereby distinguishing between different species within the same genus. Other positions change very slowly, allowing much broader taxonomic levels to be distinguished. The comparison of 16S rRNA sequences between organisms is quantitative and is based on a defined set of assumptions. The assumption that the rate at which base changes occur and are established within a species is constant is unlikely to be true. Changes in Earth’s environment are expected to alter the ecological niches or selective pressures that affect the rate of mutation and the rate at which various species are able to evolve.

The radical differences between Archaea and Bacteria, which are evident in the composition of their lipids and cell walls and in the utilization of different metabolic pathways, enzymes, and enzyme cofactors, are also reflected in the rRNA sequences. The rRNAs of Bacteria and Archaea are as different from each other as they are from eukaryotic rRNA. That suggests that the bacterial and archaeal lines diverged from a common precursor somewhat before eukaryotic cells developed. That proposal also implies that the eukaryotic line is quite ancient and probably did not arise from any currently known bacteria. It had been previously believed that eukaryotic cells arose when some bacterial cells engulfed another type of bacterium. Those bacteria might have formed a symbiotic relationship in which the engulfed cell continued to survive but gradually lost its independence and took on the properties of an organelle. Although the original eukaryotic cell may or may not be derived from bacteria, it remains likely, if not certain, that eukaryotic organelles (e.g., mitochondria and chloroplasts) are descendants of bacteria that were acquired by eukaryotic cells in an example of symbiotic parasitism.

Early hypotheses about the origins of life suggested that the first cells obtained their energy from the breakdown of nutrients in a rich organic liquid environment proposed to have formed in the early oceans by the action of light and intense solar radiation on the early, anaerobic atmosphere. The process of photosynthesis might have evolved much later in response to the gradual depletion of those rich nutrient sources. On the other hand, rRNA sequence analysis places photosynthetic capability in almost all of the major bacterial divisions and shows that photosynthetic genera are closely related to nonphotosynthetic genera. Since photosynthesis is such a highly conserved, mechanistically complex process, it is unlikely that the ability to carry out photosynthesis could have evolved at different times in so many different organisms. Even more widely distributed among prokaryotes is lithotrophy (from the Greek word lithos, meaning “stone”), the ability to obtain energy by the transfer of electrons from hydrogen gas to inorganic acceptors. It has been proposed that the earliest forms of life on Earth used lithotrophic metabolism and that photosynthesis was a later addition to the early bacterial progenitors. The nonlithotrophic and nonphotosynthetic forms found today arose from the earliest forms of Bacteria, although they have lost their capacities for lithotrophy and photosynthesis.

The proposal that lithotrophy was widely distributed among bacterial organisms before photosynthesis developed suggests that the Archaea came from a different line of descent from that of Bacteria. The only photosynthetic archaeon, Halobacterium, has a completely different type of photosynthesis that does not use chlorophyll in large protein complexes to activate an electron, as in plants and bacteria. Rather, it uses a single protein, bacteriorhodopsin, in which light energy is absorbed by retinal, a form of vitamin A, to activate a proton (hydrogen ion).

The analysis of rRNA sequences from bacteria that are closely related to one another has revealed several surprising relationships between those organisms. For example, Mycoplasma, which appear to be different from other bacteria—in that they are very small, lack a cell wall, have a very small genome, and have sterols in their cell membranes—actually are related to some gram-positive clostridia on the basis of their nucleic acid sequences. That circumstance underscores the hazard of relying on phenotypic traits (observable characteristics such as the absence of a cell wall) for the assignment of evolutionary or genetic relationships. In fact, there are many groupings of bacteria that are not supported by RNA sequence analysis.

A limitation of 16S rRNA sequence analysis, however, is its poor resolution below the genus level. Populations of organisms in a given genus that reside within the same habitats can have unique 16S rRNA genotypes. However, whether those unique genotypes are indicative of distinct species typically cannot be determined from rRNA information alone.