Taxonomy and Phylogeny of Bacteria

Taxonomy of unicellular organisms has long been a challenge for biologists. Swedish scientist Carl Linnaeus (1707–78) was the first to propose a taxonomic system for classifying living things. Under Linnaeus's original system, all life was categorized into one of two kingdoms, plants or animals. This system changed over time as more and more organisms were discovered that did not easily fit into one of these kingdoms. In 1969, American ecologist Robert Whittaker (1920–80) proposed a five-kingdom system that divided life into five major groups: Animalia, Plantae, Fungi, Protista, and Monera.

Today, Whittaker's system has been replaced by a three-domain system devised by Carl Woese (1928–2012). These domains are Eukarya, Bacteria, and Archaea. The old kingdom Monera was split into the domains Archaea and Bacteria, with the Protista, Animalia, Plantae, and Fungi kingdoms being moved into the domain Eukarya. Domain Archaea encompasses unicellular organisms that lack both nuclei and membrane-bound organelles and do not contain peptidoglycan, a sugar-protein polymer, in their cell walls. Domain Bacteria encompasses unicellular organisms that lack nuclei or membrane bound organelles that contain peptidoglycan in their cell walls. Finally, domain Eukarya includes all eukaryotic organisms, for example animals, fungi, plants, and protists.

Bacteria comprise the largest group of organisms on Earth and have been found in every ecosystem surveyed. They can survive and even thrive in places no other organism can live. Bacteria are tremendously small, ranging in size from 0.2 to 2.0 micrometers (μm). However, aggregates of many bacteria can often be seen with the naked eye because they tend to grow in biofilms, thin films of bacteria that stick to surfaces. Because of the extreme diversity of bacteria, it is necessary to find a way to classify them.

When discussing living things, biologists find it useful to classify them according to a system. Taxonomy is the scientific study of naming and categorizing living organisms based on shared characteristics. For multicellular eukaryotes, such as plants and animals, these classifications tend to be based on relatedness, or how much genetic material the organisms share. Phylogeny refers to the evolutionary history of groups of organisms. It examines common ancestry and branched lines of descent. Many taxonomists use phylogeny to determine how to classify organisms.

However, for bacteria, these methods present unique challenges. Bacteria have generation times measured in minutes or hours, rather than years as plants and animals do. Furthermore, their structures are far less complex than those of multicellular organisms. This means that bacteria rapidly evolve and can quickly develop into new species. In addition, since fossilized bacteria are less common than other fossils, bacterial fossils are more difficult to study. Thus, it becomes extremely challenging to tell how closely related species of bacteria might be.

One way biologists classify bacteria is by using numerical taxonomy, the grouping of organisms by mathematical analyses comparing shared and unshared traits. These states can include morphological, reproductive, and nutritional characteristics, among others. In numerical taxonomy, organisms are compared based on the number of shared characteristics and then grouped according to that number. Operationally, a value of 1 is assigned if a character is present, and a value of zero is assigned if it is absent. Then the number of shared presences and absences between two taxa are tallied and divided by the total number characters analyzed to provide a decimal value ranging between 0 and 1 that indicates their overall similarity. Similar numbers for two organisms indicate high similarity, regardless of whether the numbers are high or low (i.e., two organisms sharing an absence of many traits are likely as similar as two organisms sharing a presence of many traits). Relatedness can also be expressed as percent similarity. The more similar traits two organisms have, the more closely related they are likely to be. This grouping tends to result in classifications based on phenotypic similarities rather than evolutionary relatedness. However, many phenotypic traits are genetically derived, so some evolutionary information can be inferred from numerical taxonomy. Traits that may be used in numerical taxonomy include cell membrane structure, requirement for aerobic or anaerobic conditions, properties of nucleic acids in the cell, and presence or absence of particular enzymes. One characteristic used for grouping bacteria in numerical taxonomy is genetic homology, the analysis of genetic similarities among organisms. It is not necessary to obtain an organism's complete genome in order to evaluate genetic homology. Scientists often look for particular markers, or specific sequences in the genome, rather than comparing entire genomes. Bacteria with a great deal of genetic homology are likely closely related and are thus grouped together. Genetic homology hints at evolutionary relatedness, although it is not conclusive.

Rather than investigating the complete genetic sequence of bacteria, some researchers simply study the base composition—what percentage of the organism's genome is composed of each particular nucleotide. Because adenine (A) pairs with thymine (T) and guanine (G) pairs with cytosine (C), base composition is discussed in terms of A–T and G–C content. Organisms with similar G–C content are grouped more closely than those with differing G–C content.

Researchers may also employ DNA hybridization to analyze genetic homology. In this method, the double-stranded DNA from two organisms are split apart and single strands from each organism are allowed to combine, or anneal. DNA strands will only anneal if they are significantly complementary, so the quantity of annealed DNA gives valuable information about the genetic homology of the two organisms.

Genetic homology research may also include a bacterium's protein profile. A protein profile is a system for scoring the amino acids at each position in a given protein according to the frequency with which they occur. This scoring allows scientists to model protein families and domains based on the proteins cells produce at specific time and set of conditions. It is obtained by isolating all the proteins from a bacterium and visualizing them on a polyacrylamide gel. When electric current is passed through the gel, the proteins separate according to size. This creates a unique protein "fingerprint" for the organism. Because proteins are coded by genes, bacteria with similar protein profiles likely have similar genes.

Bacteria may also be grouped according to phage typing. Phage typing is a method of grouping bacteria killed by the same bacteriophage, or bacteria-infecting virus. There are more strains of bacteriophages than there are types of bacteria. In phage typing, bacteria are spread across an agar plate, making a lawn, or an even distribution of bacteria across the plate. The plate is marked with a grid, and different strains of bacteriophages are placed dropwise into each square of the grid. Bacteriophages are extremely specific, so only the ones that fit the bacteria in question will cause the bacteria to die. After a short time, small gaps in the lawn will be visible where these phage strains were dropped. Plates of different bacteria can be compared to one another. Bacteria affected by the same phages are more likely to be similar.

Bacteria are also regularly classified based on their physical characteristics and growth requirements. Classification based on Gram staining, a method of staining used to differentiate types of bacteria based on cell wall structure, is a simple and ubiquitous example of this type of grouping that has wide utility.