ADVERTISEMENTS:
Medical textbooks are the last place to look for definitive taxonomic data. Medical sources are historically the most conservative in keeping abreast of changes in taxonomy and nomenclature. Such nomenclatural anachronisms as “Vibrio comma,” “non-cholera vibrio,” and “Salmonella typhosa” are examples of the ultraconservative approach of some medical sources to changes in nomenclature.
The most comprehensive treatment of bacterial classification, particularly for nomenclature, type strains, description of taxa and references to pertinent literature, is found in Bergey’s Manual of Systematic Bacteriology, vol. 1. and in Bergey’s Manual of Determinative Bacteriology, 8th ed.
These are invaluable reference sources and should be at the desk of every microbiologist. The 8th edition of Bergey’s Manual was published in 1974. Bergey’s Manual of Systematic Bacteriology will be published in four sub-volumes, of which the first appeared in January 1984 and the others will appear at approximately 1-year intervals.
ADVERTISEMENTS:
It is therefore necessary to keep the 8th edition until all four sub-volumes of the Manual are available. New editions of the Bergey’s Manual for Determinative Bacteriology (formerly the shorter Bergey’s Manual) will be published covering sub-volumes 1 and 2, and 3 and 4, of the new Manual and are designed for bench use.
The 8th edition of Bergey’s Manual is out of date for those taxa in which new species have been described or in which nomenclatural changes have been made. Furthermore, space limitations make it impossible to fully describe many species. The interested party should, therefore, begin a taxonomic search with Bergey’s Manual and then augment the information obtained there by searching the International Journal of Systematic Bacteriology—in which all new species must be described or reference must be made to the journal in which they are described-and by contacting authorities in the specific field.
Other journals that publish papers on new species are the Journal of Clinical Microbiology, Current Microbiology, Annales de Microbiologie (Institute Pasteur), and Systematic and Applied Microbiology.
In Bergey’s Manual bacteria are placed in the kingdom Prokaryotae. They are sub-divided into four divisions: Gracilicutes for gram-negative-type cell walls, Firmacutes for gram-positive cell walls, Tenericutes for organisms lacking a cell wall (mycoplasmas), and Mendosicutes for bacteria that have faulty cell walls and presumably lack peptidoglycan.
ADVERTISEMENTS:
Each division is further subdivided into classes. Within each class are orders, and within the orders are families or (if family names are not available) morphological groups that are further subdivided to genera and species.
From a functional standpoint, the bacteria are divided into a number of “sections” (“parts” in the 8th edition) on the basis of Gram reaction, oxygen requirement, spore formation, and metabolic pattern (gram- positive anaerobes; gram-negative, facultatively anaerobic rods; endospore-forming rods; gliding, non-fruiting bacteria; gram-negative heterotrophs; etc.). Each section (part) is further divided to the species level or, where pertinent, to sub-specific categories such as bio-groups and serotypes (serovars).
The following sections are included in subvolume 1 of Bergey’s Manual: spirochetes; gram-negative, aerobic, microaerophilic, motile helical or curved bacteria; gram-negative, non-motile or rarely motile curved bacteria; gram-negative aerobic rods and cocci; gram-negative facultatively anaerobic rods; gram-negative anaerobic rods; gram-negative anaerobic cocci; dissimilatory sulfate- or sulfur-reducing bacteria; rickettsias and chlamydias; mycoplasmas; and unclassified endosymbionts.
Sub-volume 2 will contain gram-positive cocci; endospore-forming rods; gram-positive, regular, non-sporing rods; gram-positive, irregular, non-sporing rods; mycobacteria; and nocardioform bacteria.
Sub-volume 3 will contain gliding, non-fruiting bacteria; anoxygenic photosynthetic bacteria; budding and/or appendaged bacteria; archaeobacteria; sheathed bacteria; gliding, fruiting bacteria; chemolithotrophic bacteria; and cyanobacteria. Sub-volume 4 will contain the streptomycetes and their allies. Until sub-volumes 2 through 4 are published, the reader must depend on the 8th edition of Bergey’s Manual and alternative sources for information on the groups that they will address.
Cowan has referred to “the trinity that is taxonomy” – classification, identification, and nomenclature. Before discussing reasons for nomenclatural and taxonomic changes and the concept of a bacterial species, it is necessary to establish working definitions for these terms.
“Classification” is simply an orderly arrangement of bacteria into groups. There is nothing inherently scientific about classification. Mandel has said that “like cigars, a good species and a good classification is one which satisfies”. Cowan correctly observed that classification is purpose oriented; thus, a successful classification is not necessarily good, and a good classification is not necessarily successful. Very often specialty groups classify the same organisms in a different manner or to a different level.
“Identification” is the practical use of a classification to isolate and distinguish desirable organisms from undesirable ones, to verify the authenticity or utility of a culture, or to isolate and identify the causative organism of a disease, of a desirable reaction, etc. “Nomenclature” is the means through which the characteristics of a species are defined and communicated among microbiologists.
It is essential that a name has the same meaning to all microbiologists, yet there are many names that are defined differently in different parts of the world or by different microbiological specialty groups. Klebsiella pneumoniae is defined differently in England than in most other parts of the world, and Vibrio cholerae has often been equated with a single serotype by epidemiologists and clinical microbiologists.
ADVERTISEMENTS:
Taxonomy is the science of classification. As a science, it is dynamic and subject to change on the basis of available data. New data often necessitate changes in taxonomy. These changes frequently result in changes in the existing classification, in nomenclature, in criteria for identification, and in the recognition of new species.
For eukaryotes the species definition usually stresses the ability of similar organisms to reproduce sexually, with the formation of a zygote, and to produce fertile offspring. Sexual reproduction, in the eucaryotic sense, does not occur in bacteria.
The term species as applied to bacteria has been defined as a distinct kind of organism, having certain distinguishing features, and as a group of organisms which generally bear a close resemblance to one another in the more essential features of their organization. The problem with these definitions is that they are subjective.
What is “a close resemblance”? What are “more essential features”? How many “distinguishing features” are sufficient to create a species? Historically, these questions have been answered arbitrarily. Species were often defined solely on the basis of criteria such as host range, pathogenicity, ability or inability to produce gas in the fermentation of a given sugar, and rapid or delayed fermentation of sugars.
ADVERTISEMENTS:
Since there was no way to devise a single species definition that could be applied to all groups, criteria used to define species were heavily slanted towards the prejudices of the investigators who described the species. For example, Fritz Kauffmann defined a species as “a group of related sero-, biophagotypes” and believed that serology was the ultimate criterion in taxonomy.
To him, each serotype of Salmonella was a separate species. We now know that all serotypes of Salmonella are genetically the same species. These practices probably led Cowan to state that “taxonomy …is the most subjective branch of any biological science, and in many ways is more of an art than a science.”
Edwards and Ewing, in their monumental studies on Enterobacteriaceae, pioneered in establishing the following principles for characterizing, classifying, and identifying organisms:
1. Classification and identification of an organism should be based upon its overall morphological and biochemical pattern. A single characteristic (pathogenicity, host range, biochemical reaction, etc.), regardless of its importance, is not a sufficient basis for classifying or identifying an organism.
ADVERTISEMENTS:
2. To accurately determine the biochemical characteristics of a given species, one must test a large diverse strain sample. The reactions of these strains to any test should be expressed in percentages.
3. Atypical strains, when adequately studied, are often perfectly typical members of a given bio-group within an existing species, and sometimes they are typical members of a new species.
In the 1960s, numerical taxonomy (also called computer or phenetic taxonomy) became widely used. In this method a large number of biochemical, morphological, and cultural characteristics (usually between 50 and 200) are used to determine the degrees of similarity between organisms. More recently, susceptibilities to antibiotics and inorganic compounds have been added to the characteristics used in numerical taxonomy.
Classification and identification of bacteria were significantly improved by the numerical approach to taxonomy. Organisms could be classified on the basis of a large number of characteristics. Many new and some previously ignored biochemical tests were assayed as possible aids in classification.
ADVERTISEMENTS:
Species-specific and genus-specific tests were identified by the numerical approach. Clinical and applied laboratories could then use these tests to help separate specific species and groups of organisms.
In the numerical approach to taxonomy, investigators often calculate the coefficient of similarity or percentage of similarity between strains (for this discussion “strain” refers to a single isolate from a clinical or other specimen). A dendrogram or a similarity matrix is constructed that joins individual strains into groups and joins one group with other groups on the basis of their percentage of similarity.
Group 1 represents three strains that are about 95% similar and join with a fourth strain at the level of 90% similarity. Group 2 is composed of three strains that are 95% similar, and group 3 contains two strains that are 95% interrelated and a third strain to which they are 90% similar. Similarity between groups 1 and 2 occurs at the 70% level, and group 3 is about 50% similar to groups 1 and 2.
In some studies all the characteristics included in the similarity matrix are given equal weight, and in some studies certain characters are weighted (for example, the presence of spores in Clostridium might be weighted more heavily than the organism’s ability to utilize a given carbon source).
A given level of similarity can be and has been equated with relatedness at the species, genus, and, sometimes, the subspecies level. For instance, strains of a given species may cluster at a 90% similarity level, and species within a given genus may cluster at a 70% level.
If these values were applied, the strains in groups 1, 2, and 3 would each represent a separate species. The species represented by groups 1 and 2 would be placed in the same genus, and the species represented by group 3 would be in a separate genus.
ADVERTISEMENTS:
Several problems arise when this approach is used as the sole basis for defining a species:
1. How many and which tests should be chosen?
2. Should the tests be weighted? If so, how?
3. What level of similarity should be chosen to reflect relatedness at the level of species and genus?
4. Is the same level of similarity applicable to all groups?
The molecular weight of DNA in most bacteria is between 1 x 109 and 8 x 109, enough to specify some 1,500 to 6,000 average-sized genes. Therefore, even a battery of 300 tests would, at most, assay only between 5 and 20% of the genetic potential of bacteria.
ADVERTISEMENTS:
It is almost certainly true that tests which are comparatively simple to carry out and assay (such as tests for carbohydrate utilization and for enzymes whose presence can be assayed colorimetrically) predominate over tests for structural genes, for genes involved in macromolecular synthesis, reproduction, or regulatory functions, and for other genes whose presence is difficult to assay.
Additional potential sources of error must be considered when identifying species solely on the basis of phenotype:
1. Different enzymes (specified by different genes) may catalyze the same reaction.
2. Negative reactions can occur when the metabolic gene is present and functional. They can occur through any of several mechanisms, including the inability of the substrate to enter the cell, and a regulatory or suppressor mutation.
3. A negative reaction can occur when the gene is present but not functional because of a mutation in a portion of the gene that is necessary for enzyme activity.
4. The correlation between a reaction and the number of genes (or enzymes) necessary to carry out that reaction is not necessarily one to one. If one assays for the end product, a positive reaction indicates six similar enzymes, whereas a negative reaction can mean the absence or non-function of anywhere from one to six enzymes.
5. Fastidious strains will not cluster with non-fastidious strains from the same species. This is often seen with Escherichia coli and K. pneumoniae. Several other strain characteristics can drastically affect phenotypic characterization. These include slow growth rate, temperature of incubation (Yersinia, Erwinia), salt requirement (marine Vibrio species), and pH (Legionella).
6. Plasmids carrying metabolic genes can enable strains to carry out reactions that are rarely, if ever, seen in the absence of the plasmid. The same set of “definitive” reactions cannot be used to classify all groups of organisms, and there is no magic number of specific reactions that allows one to define a species. Organisms are identified on the basis of phenotype, but from the taxonomic standpoint, identification to the species level based solely on phenotype is subject to error.
The ideal means of identifying bacterial species would be a “black box” which would separate genes and instantly compare the nucleic acid sequences in a given strain with a standard pattern for every known species—something akin to mass spectrophotometry analysis.
Although restriction endonuclease analyses can be done to determine common sequences in isolated genes, we are not at all close to having an appropriate black box, especially one suited for clinical laboratory use. There is, however, a method with which one can compare the total DNA of one organism with that of any other organism.
This method, pioneered by Marmur and Doty, Speigelman, Bolton, and McCarthy, and Britten and Kohne, is called nucleic acid hybridization or DNA hybridization. It measures the amount of DNA sequences held in common between any two organisms. One can also approximate the percentage of divergence or unpaired nucleotide bases within related DNA sequences.
DNA relatedness studies have been done in yeasts, viruses, bacteriophages, and many groups of bacteria. A partial list of these bacteria includes members of the family Enterobacteriaceae, Brucella, Bacillus, Pseudomonas, Lactobacillus, Haemophilus, Mycobacterium, Vibrio, Neisseria, Bacteroides and other anaerobic groups, and Legionella.
Five parameters are now used to determine DNA relatedness: genome size, guanine-plus-cytosine (G + C) content, DNA relatedness under conditions optimal for DNA re-association, thermal stability of related DNA sequences, and DNA relatedness under supraoptimal conditions for DNA re-association.
Genetic parameters are presently impractical for clinical laboratories. We therefore must correlate practical biochemical tests with the genetic data. For example, yellow-pigmented strains of Enterobacter cloacae were shown to be a separate species genetically, but were not designated as such (Enterobacter sakazakii) until there were three practical tests that correlated with the genetic data. The same procedure was followed before designating a number of new species in Klebsiella, Yersinia, and Serratia.
G + C Content:
The G+C content in bacterial DNA ranges from about 25 to 75%. The G + C percentage is specific for a given species, but is not exclusive for that species. For example, E. coli, salmonellae, and Morganella morganii have 50 to 52% G + C, and Bacillus subtilis and Pasteurella spp. have a G + C percentage of about 42%.
Therefore, two strains with a similar G + C percentage may or may not belong to the same species. If the G + C content is very different, however, then the strains cannot be members of the same species. G + C content is useful in placing a strain in the proper ballpark for further testing. A good example is a recently isolated organism that is biochemically between Vibrio and Aeromonas. It was placed in the Vibrio genus because its G + C content was 50%, which is within the range for Vibrio species and is significantly less than the 57 to 60% G + C found in Aeromonas species.
Genome Size:
True bacterial DNAs have molecular weights between 1 x 109 and 8 x 109 (genome size). Genome size determinations can, in certain circumstances, distinguish between groups that have very different genome sizes. They were used to distinguish Legionella pneumophila (the Legionnaires disease bacterium) from Rochalimaea (Rickettsia) quintana. L. pneumophila has a genome size of about 3 x 109 daltons; that of R. quintana is about 1 x 109 daltons.
DNA Relatedness under Conditions Optimal for DNA Re-Association:
DNA relatedness is determined by allowing single-stranded DNA from one strain to re-associate with single-stranded DNA from a second strain to form a double-stranded DNA molecule. DNA re-association is a specific, temperature-dependent reaction. The optimal temperature for DNA re-association is some 25 to 30°C below the temperature at which native double-stranded DNA is denatured into single strands.
Studies with a large number of groups indicate that a bacterial species is composed of strains that are usually 70 to 100% related. Relatedness between species is 0% to about 60%. It is important to emphasize that the term “related” does not necessarily mean “identical” or “homologous.” Similar nucleic acid sequences can re-associate. This fact and its importance are illustrated below.
Thermal Stability of Related DNA Sequences:
It has been shown that each 1% of unpaired nucleotide bases in a double-stranded DNA sequence causes a 1% decrease in the thermal stability of that DNA duplex. By comparing the thermal stability of a control double-stranded molecule in which both strands of DNA are from the same organism with a heteroduplex (DNA strands from two different organisms), we can assess differences in thermal stability.
Decreased thermal stability can be thought of as divergence in nucleotide sequences. In practice, strains that are 70% or more related show from 0% to about 5% divergence in related sequences, whereas sequences held in common between different species show 8 to 20% divergence. It is quite important to know the amount of divergence and the relatedness at supra-optimal conditions when strains are 60 to 70% related.
DNA Relatedness under Supraoptimal Conditions for DNA Re-Association:
When the incubation temperature used for DNA re-association is raised from 25 to 30°C below the renaturation temperature to only 10 to 14°C below the denaturation temperature, only very closely related (and therefore highly thermally stable) DNA sequences can re-associate.
Strains from the same species are 55 to 100% related to supraoptimal incubation temperatures. Strains from different species are 50% or less related. High temperature reactions are especially important in distinguishing between strains that are 60 to 70% related under optimal re-association conditions.
By applying these five parameters, we can generate a species definition based on DNA. E. coli can be defined as a series of strains with a G+C content of 49 to 52%, a genome size of 2.3 x 109 to 3.0 x 109 daltons, relatedness of 70% or more at an optimal re-association temperature with 0 to 4% divergence in related sequences, and relatedness of 60% or more at a supraoptimal re-association temperature.
Experience with more than 100 species resulted in an arbitrary molecular definition of species as a group of strains with similar G+C content and genome size that are 70% or more related with 0 to 5% divergence among related sequences and in which relatedness remains at 55% or more at supraoptimal incubation temperature.
There is general agreement that 70% or more indicates relatedness at the species level. The 70% species relatedness rule has been occasionally ignored when the existing nomenclature is both very deeply ingrained and useful. One such example is E. coli and the four species of Shigella.
These organisms are all 70% or more related and should therefore be grouped into a single species, instead of the present five species in two genera. This change has not been made because the primary consideration was to avoid the confusion that such a change would create among members of the medical community.
Another example is Yersinia pestis and Yersinia pseudotuberculosis, which are the same species genetically. It has been proposed that they be treated as two subspecies of a single species, but clinically it is extremely important to continue to report Y. pestis and to distinguish it from Y. pseudotuberculosis.
DNA relatedness provides one species definition that can be equally applied to all organisms. It is not subject to phenotypic variation, to mutations, or to the presence or absence of metabolic or other plasmids. This is because it measures overall relatedness, and atypical biochemical reactions, mutations, and plasmids affect only a very small percentage of the total DNA.
This point is illustrated by looking at DNA relatedness data obtained from a large number of biochemically atypical E. coli or E. coli-like strains. Many of these biochemically atypical strains were shown to be genetically typical E. coli. The biochemical borders of E. coli, therefore, include all of these bio-groups.
Two groups of strains were shown not to belong to E. coli. These are KCN- and cellobiose-positive, yellow-pigmented strains, which proved to be a new species, and urea-, KCN-, citrate-, and cellobiose-positive strains that were in fact Citrobacter spp.
In practice, the approach to bacterial taxonomy should be polyphasic. The first step is phenotypic grouping of strains by biochemical reactions and any other characteristics of interest. In the second step, phenotype groups are tested for DNA relatedness to determine whether the observed phenotypic homogeneity (or heterogeneity) is reflected by genetic homogeneity or heterogeneity.
The third and most important step for identification is re-examination of the biochemical characteristics of the DNA relatedness groups. This allows a determination of the biochemical borders of each group and of those reactions that are of diagnostic value for the group.
For identification of a given organism, we weight specific tests on the basis of their correlation with DNA results. Occasionally, the commonly used reactions will not totally distinguish between two distinct DNA relatedness groups. In these cases, one must search for other biochemical tests that are of diagnostic value.
Heavily weighted tests are of great value for specific organisms (coagulase for Staphylococcus aureus, DNase for Serratia spp., etc.). We often forget that before these weighted tests are of diagnostic value the organism has been well characterized by growth on selective media, by colonial and cellular morphology, and by other biochemical tests.
It is possible to detect cholera toxin and both heatstable and heat-labile enterotoxins in E. coli by means of specific gene probes. The genes (or genetic regions including all or part of the genes coding for these toxins) are hybridized with purified DNA from suspected toxigenic strains, with colonies lysed directly on nitrocellulose filter paper, or directly with unprocessed stool material lysed on nitrocellulose filter paper.
When stools are used, results are obtained within 48 h. Gene probes are being developed for the identification of toxigenic Yersinia enterocolitica and for identification of salmonellae, shigellae, Neisseria gonorrhoeae, and legionellae.
Detection of a specific plasmid associated with virulence can be used to detect virulent strains of Y. enterocolitica and invasive strains of E. coli.
Molecular analysis of plasmids has recently been extensively used to identify epidemic strains of bacteria. Plasmid DNAs are separated electrophoretically (by molecular weight). Plasmids of similar size can be specifically fragmented, by treatment with one or a series of restriction endonucleases which cleave DNA at specific sites, to determine similarities precisely.
For plasmid profile and restriction endonuclease analysis to be effective in monitoring epidemic strains, the strains must contain plasmids and these plasmids must be different from those of non-epidemic strains. Fortunately, these conditions are met in the majority of Enterobacteriaceae and other bacteria involved in nosocomial disease outbreaks, and in many of the bacteria responsible for food-borne disease outbreaks.
The purpose of classification and identification is to be able to distinguish one organism from another and to group similar organisms on the basis of criteria of interest to all microbiologists or to any single specialty group. The purpose of nomenclature is to provide a convenient system of communication to define an organism without the necessity of listing its characteristics. The most important level of communication is at the species level.
A species name should mean the same thing to everyone, regardless of his or her specialty. We cannot have effective communication if strains of the same species are given different names on the basis of source of isolation, serotype, presence or absence of a converting bacteriophage, or the ability to perform a specific function, such as cause a disease, produce an antibiotic, etc.
Species have been created on the basis of each of these criteria and many others. These criteria may be extremely important for a specialty group, but, by themselves, they are not a sufficient basis for establishing a species. Species should be established based on the polyphasic criteria mentioned above.
Special-interest groups of microbiologists need to communicate, but their needs can and should be met by designations below the species level as “groups” or “types” on the basis of common serological or biochemical reactions, phage or bacteriocin sensitivity, pathogenicity, or other characteristics.
For example, bioserogroup or bioserotype is a group of strains of the same species with common biochemical and serological characteristics that set them apart from other members of the species. Many of these are already commonly used and accepte – serotype, phage type, colicin type, biotype, bioserotype, and pathotype.
The ending “var” (phagovar, etc.) can be substituted for “type.” We prefer “type.” Host specificity has been expressed in many groups by the term “variety.” For example, Erwinia herbicola var. ananas causes rot in pineapple and Erwinia chrysanthemi var. zeae is pathogenic for corn. By using these designations, there can be communication among all microbiologists at the species level and among all special-interest groups at an intrasubspecific (below species) level.
There is no genetic definition of a genus. If there were, an ideal genus would be composed of a series of species that are similar phenotypically and genetically (50 to 65% related). Some genera that contain phenotypically similar species approach this genetic criterion – Citrobacter, Yersinia, and Serratia, to name a few.
More often, the phenotypic similarity is present, but the genetic relatedness is not. Bacillus, Clostridium, Vibrio, Campylobacter, Pseudomonas, and Legionella are good, or at least accepted, phenotypic genera in which relatedness between species is not 50 to 65% but 0 to 65%. When both phenotypic similarity and genetic similarity do not occur, phenotypic similarity should generally be given priority in establishing genera. The reason for phenotypic priority at the genus level is a practical one.
When identifying organisms at the bench, it is desirable to have the most phenotypically similar species in the same genus. Primary consideration for a genus is that it contains biochemically similar species that are convenient or important to consider as a group and that must be separated from one another at the bench.
In some cases, generic names have been changed because the original name was not validly published for one of a variety of reasons. These changes are legislated. More often, name changes are made on the basis of data not available when the original name was proposed.
More than one name for a genus (or species) can also result from two investigators independently publishing different names for the same group. Even though one of the names will have priority, both will often persist for a long time.
A typical example of a current nomenclatural problem based on the interpretation of data is in the family Legionellaceae. Since the first species, Legionella pneumophila, was described by workers at the Centers for Disease Control, 21 additional species have been described or are in the process of being described, all of which were placed in the genus Legionella.
Another group of investigators confirmed the first five species and concluded that two additional genera were necessary on the basis of DNA relatedness. This second group gave precedence to a genetic genus concept, whereas the original investigators argued for a single genus on the basis of phenotypic, pathogenic, and treatment similarities among all of the species.
The first group believed that these similarities, and the present inability to separate most species except by serology, made it impractical to create additional genera. The difference in opinion will ultimately be settled by usage in the scientific and medical community. Thus far the single genus Legionella has by far the greater usage, but the other generic names are occasionally seen in the literature.
Why should species be named, how are they named, and where are they named? According to the International Code of Nomenclature of Bacteria (Bacteriological Code), “the primary purpose of giving a name to a taxon is to supply a means of referring to it.”
In other words, names are to foster communication and to ensure that the description of a given set of characteristics (by using a name to define them) has the same meaning to all scientists, just as it is more meaningful to say Willie Mays or Sean Connery than to describe the greatest center fielder of all time or one of the men who played the role of James Bond.
Species are named in accordance with principles and rules of nomenclature as set forth in the Bacteriological Code. The first principle of bacterial nomenclature is concerned with creating stability, avoiding or rejecting names that cause error or confusion, and avoiding the useless creation of names.
ADVERTISEMENTS:
Scientific names are usually taken from Latin or Greek and, regardless of their origin, are treated as Latin. The correct name of a species or higher taxonomic designation is determined by three criteria: valid publication, legitimacy of the name with regard to the rules of nomenclature, and priority of publication (it is the first validly published name for the taxon). “A name has no status under the rules and no claim to recognition unless it is validly published”.
Until 1 January 1980, priorities for names dated from 1 May 1753. This caused much confusion since it was difficult to search the literature to ensure that a species had not been previously proposed. The early descriptions were often sketchy and were based on fewer and often different tests than are now used.
Furthermore, strains representing the species proposed in the 19th century often were not available, were of uncertain authenticity, or, when tested, did not exactly correspond to the published properties of the species.
Priorities for bacterial names now start as of 1 January 1980. On that date, the Approved Lists of Bacterial Names were published in the International Journal of Systematic Bacteriology (IJSB). Names not on those lists lost all standing nomenclatural status (“Arizona hinshawii” is an example).
Now we can quickly determine whether a species has been previously published by consulting the Approved Lists and the lists of valid names published periodically in IJSB. The Approved Lists and the requirement to search only to 1980 for prior species proposals has removed and will continue to remove much of the past confusion and will preclude many future problems.
To be validly published, a new species proposal must contain the species name, a description of the species, and the designation of a type strain for the species, and the name must be published in IJSB. The proposed name is automatically validly published if the proposal is published in IJSB.
If the proposed name is published in another journal, it is not validly published until it appears in IJSB. It is the author’s responsibility to send reprints of such publications to the editor and to request publication of the new name(s) in IJSB. In this case, the date of valid publication is the date of publication in IJSB.
Most people seem to believe that, once proposed, a name must go through some formal process leading to its official acceptance. The opposite is true: a validly published name is assumed to be correct unless and until it is officially challenged. This is done by publishing a request for an opinion (to the Judicial Commission of the International Union of Microbiological Societies) in IJSB.
The Judicial Commission will then seek advice from appropriate experts and render an opinion. This is done only in cases in which the validity of a name is questioned with respect to compliance with the rules of the Bacteriological Code. A question of classification that is based on scientific data (for example, whether a species, on the basis of biochemical or genetic characteristics or both, should be placed in a new genus or an existing genus) is not settled by the Judicial Commission but by the preference and usage of the scientific community.
This is why generic synonyms exist. Some more recent examples are Citrobacter (Levinea) amalonaticus, Citrobacter diversus (Levinea malonatica), Morganella (Proteus) morganii, Legionella (Tatlockia) micdadei, and Legionella (Fluoribacter) dumoffii. The reader is invited to consult the Bacteriological Code, the Approved Lists, and Bergey’s Manual for further details on bacterial nomenclature.
Taxonomy, albeit far from perfect, is not designed to create confusion in the lives of clinical microbiologists or any other group. The task of the taxonomist is to describe all species, using the best available scientific tools. Newly recognized species have sometimes been masquerading as members of well-known species. Such was the Case with Klebsiella oxytoca (indolepositive Klebsiella pneumoniae), Enterobacter sakazakii (yellow-pigmented, sorbitol-negative, delayed DNase positive Enterobacter cloacae), Vibrio mimicus (sucrose-negative Vibrio cholerae), and Yersinia intermedia (rhamnose- and melibiose-positive Yersinia enterocolitica). Many others, including the Lyme disease spirochete, legionellae. Vibrio hollisae, Yersinia ruckeri, Enterobacter gergoviae, and Escherichia vulneris were not previously recognized.
Clinical microbiologists should keep up with taxonomic changes to know which newly described species present potential clinical problems and which are rarely, if ever, human pathogens. Examples of newly described human pathogens, with an argument that clinical microbiologists should stay abreast of taxonomic changes, were recently given by Farmer.
A final word of caution: you will never see a given organism if you do not use appropriate isolation and enrichment media and do the proper tests for identification. Newly described biochemical tests should be done exactly as described in the literature. Modifications, shortcuts, or any change in media or reagents must not be attempted unless and until these changes have been published or shown to yield comparable results by the individual laboratory.
Several years ago, strains of lactose-positive Salmonella caused an epidemic in Central and South America. These strains were isolated in some United States laboratories, but other laboratories reported that they had never seen them.
Many laboratories pick only lactose-negative colonies from isolation plates; thus, they will never have a problem with “lactose-positive” Salmonella. Similarly, many laboratories never isolated Yersinia enterocolitica, Campylobacter fetus, many Vibrio species, or legionellae until they set up specific procedures for isolating these organisms.
