Microbiology (2003), 149, 1–7
Review
DOI 10.1099/mic.0.25952-0
Prokaryote taxonomy of the 20th century and the
impact of studies on the genus Pseudomonas:
a personal view
Norberto J. Palleroni
Correspondence
Downloaded from www.microbiologyresearch.org by
IP: 130.203.136.75
On: Sun, 11 Sep 2016 21:04:19
[email protected]
Department of Biochemistry and Microbiology, Rutgers University Cook College,
New Brunswick, NJ 08901, USA
The taxonomic studies on the genus Pseudomonas performed in the Department of Bacteriology of
the University of California at Berkeley played a significant rôle in the development of modern
prokaryote taxonomy, which started in the 1960s. This impact was due to a revival of the method of
den Dooren de Jong for the nutritional analysis of chemoorganotrophic organisms and, mostly, to
the introduction of the determination of rRNA as a method of taxonomic analysis. While the
introduction of the nutritional studies facilitated the characterization of Pseudomonas and other
chemoorganotrophs, the applicability of the rRNA studies extended to all prokaryotes.
Overview
The construction of trees representing the similarity of
sequences of one or a few conserved macromolecules (often
the 16S component of the rRNA complex) in different
organisms has become commonplace in publications dealing with taxonomic studies of prokaryotes. The trees are
universally called ‘phylogenetic’, and they are assumed to
play a decisive role in the creation of new taxa. However, it
has been rightly pointed out that the new rRNA phylogeny
is the phylogeny of rRNA genes but not necessarily that of
their hosts (Postgate, 1995), and therefore it should be used
with caution in the characterization of taxonomic units.
In any case, the popularity of molecular taxonomic studies
supplemented by the construction of so-called phylogenetic
trees has contributed to the displacement of most of the
pre-molecular experimental approaches developed during
a long period that goes from the last decades of the 19th
century to the 1960s, a span during which they enjoyed
various degrees of success in determinative bacteriology.
Aside from their historical value, these systems have lost
practically all their interest in the eyes of modern taxonomists. Those who are interested in discussions on the
respective merits of the early classification systems are
referred to the comprehensive articles by Kluyver & van Niel
(1936) and by van Niel (1946), which place them in their
correct perspective. I have taken from these articles many of
the elements presented in the next section.
It is the purpose of this review to give just a glimpse of
these early systems of classification, trying to emphasize
the state of affairs at the time when a radical transformation
in approaches took place in prokaryote taxonomy. The
sequence of events that I shall outline clearly indicates that a
0002-5952 G 2003 SGM
Printed in Great Britain
sudden change in taxonomic methodologies occurred from
the moment of publication of work performed on the genus
Pseudomonas.
The impact of this work was manifested in two steps, the
first of which was the demonstration of the advantage of
an extensive phenotypic characterization of strains with
emphasis on a nutritional analysis, of particular importance
in the characterization of heterotrophic organisms. More
importantly, the second step, a few years later, was the
demonstration of the genomic complexity of Pseudomonas
as shown by significant rRNA sequence differences among
the species then assigned to the genus, and the possibility
of defining ‘RNA homology groups’. The findings clearly
showed the power of rRNA similarities in defining
relationships above the species level, which until that
moment had been elusive in prokaryote taxonomy.
Consequently, true to the subject that has occupied the
largest portion of my scientific career, I dare here to divide
the 20th century with respect to prokaryote taxonomy into
two periods, which (in private and for the sake of simplification) I often call BP and AP (for ‘Before’ and ‘After’
Pseudomonas, respectively).
The first half of the 20th century
The classification of plants and animals can be traced back to
Aristotle in ancient Greece, but modern taxonomy started in
1753 with the publication of the first edition of Systema
Naturae by Carl Linnaeus, who invented the binomial
system of nomenclature in use to this day (Linnaeus, 1753).
For obvious reasons, taxonomic studies of bacteria were
delayed for more than a century, until the compound
microscope reached the resolving power required for the
observation of these organisms.
1
N. J. Palleroni
Downloaded from www.microbiologyresearch.org by
IP: 130.203.136.75
On: Sun, 11 Sep 2016 21:04:19
With the publication of his bacterial classification system,
Ferdinand Cohn is considered to be the father of bacterial
taxonomy (Cohn, 1872). His system was based on the
shape and general appearance of the bacterial cells, a choice
that had its roots in the usefulness of morphology in
the classification of higher organisms, but Cohn clearly
realized the limitation of this approach for defining bacterial
genera, and cautiously referred to the various morphological groups as ‘form genera’, without attaching to
them any phylogenetic significance. We owe also to Cohn
the important concept that bacteria are a special group
of micro-organisms, in spite of which quite a few bacterial names still carry the ending ‘mycetes’, a reminder of
the now discarded hypothesis that bacteria are fission
fungi (schizomycetes). Such is the ‘sticky’ property of
nomenclature.
Cohn’s morphological principle was universally accepted at
the time, and most taxonomists following him used it in
different variations as the basis of their own systems of
classification, occasionally adding new ‘form genera’ to the
inventory (Lehmann & Neumann, 1896, 1927; Migula, 1900;
Pringsheim, 1923; Janke, 1924; Prévot, 1933). However, at
the beginning of the 20th century, Orla-Jensen proposed
using physiological characteristics as the basic criterion for
the definition of the main lines of bacterial systematics
(Orla-Jensen, 1909). It seems clear that this alternative had
its roots in the demonstration of the formidable metabolic
diversity of prokaryotes, mainly in the laboratories of
Beijerinck and Winogradsky, whose extraordinary contributions opened the vast horizon of bacterial metabolic
diversity (Beijerinck, 1921; Winogradsky, 1949).
Anyhow, neither the physiological criterion nor the morphological principle gave a satisfactory answer to additional
conflicting alternatives. Thus if morphology is the primary
criterion, the main lines of the classification often include
organisms of similar physiological properties, and vice
versa, in a physiological system there will be organisms of
similar morphology in each of the main lines. In other
words, evolution seems to have taken its course in parallel in
each of the independent primary lines of each system of
classification. Behind the scene, in each case, is the unproven
assumption of absence of genetic exchange among members
of the various lines of descent.
The system proposed by Orla-Jensen, with its evolutionary
implications, introduced an additional paradox. After the
discovery of chemoautotrophy by Winogradsky, Orla-Jensen
favoured the idea that organisms endowed with these
properties were the most primitive, since they were able to
grow under the simple environmental conditions of a
primitive world. Notwithstanding the originality of this
approach, however, Orla-Jensen failed to take into account
the possibility that there may be an inverse relationship
between the chemical complexity of the medium and the
biochemical sophistication required to grow under such
conditions.
2
Orla-Jensen’s classification included many interesting
points, but eventually morphology was favoured by most
workers as the primary classification criterion. In fact, it was
accepted by most European bacteriologists, who followed
the system of Lehmann and Neumann, first proposed in
1896 and updated in 1927, and it was the basis of the
classification scheme developed by Kluyver and van Niel
in the 1930s (Kluyver & van Niel, 1936). In this system,
morphological characters include the shape and size of
the cells, type of motility, presence of flagella, their number
and type of insertion, mode of reproduction, occurrence of
endospores or gonidia, and various structural peculiarities.
Amongst the physiological properties the most important
are the behaviour towards temperature, oxygen and osmotic
pressure, and various characteristics of the anabolic and
catabolic reactions. Among the latter, a judicious selection
was recommended; for example, the reactions providing
energy were placed at a level of importance above hydrolytic
reactions. Pathogenicity was considered of doubtful value,
and differentiation of genera and even of species on its basis
was objectionable as a taxonomic criterion.
In the meantime, America was on the way to adopting
its own bacterial classification system. This started in 1923
with the publication of the first edition of Bergey’s Manual
of Determinative Bacteriology (Bergey et al., 1923). Its
successive editions were closely followed by all American
workers, thereby acquiring a taste of provincialism that
could not escape Kluyver and van Niel’s objections,
expressed in sharply critical terms. Their analysis of Bergey’s
system is detailed order by order, and many errors are
indicated as a consequence of the use of morphological,
physiological, utilitarian, cultural and pathogenic properties
in the most arbitrary way, with ‘utter disregard for mutual
relationships between natural groups’ (Kluyver & van Niel,
1936).
Unfortunately, the commercial success of the Bergey’s
Manual editions contributed to a disregard of substantial
criticisms formulated mainly by European microbiologists,
particularly those by Rahn (1929, 1937) and others, which
were left aside instead of being incorporated as useful
suggestions in new revisions of the system (see Kluyver &
van Niel, 1936, for references). That was unfortunate, since
the Manual represented ‘the first formal cooperation in the
history of bacterial taxonomy’ (Kluyver & van Niel, 1936).
As mentioned before, Kluyver and van Niel were among
those for whom morphology is the first and most reliable
guide of taxonomic systems but, in designing their own
classification system, they kept in mind Pringsheim’s
warnings that (a) the proposed relationships remain conjectural and (b) a number of links may have been lost in the
course of evolution (Pringsheim, 1923). Accepting the fact
that the scanty morphological variety in bacteria was the
cause of the unsatisfactory state of taxonomy, physiological
properties could be wisely added by the taxonomist when
necessary, but, far from accepting this alternative, Kluyver
and van Niel showed a most determined obstinacy by
Microbiology 149
Taxonomy in the 20th century
Downloaded from www.microbiologyresearch.org by
IP: 130.203.136.75
On: Sun, 11 Sep 2016 21:04:19
remaining truthful to morphology as the leading principle,
and even going to the extreme of considering that physiology is nothing but the expression of submicroscopic
morphology. For them, physiological principles were acceptable in classification as long as they could be subordinated
to morphology.
With time, van Niel abandoned the idea that morphology
has more than a restricted phylogenetic significance and his
previous optimistic attitude did not permeate a communication at one of the Cold Spring Harbor Symposia (van
Niel, 1946), where he seemed inclined to draw empirical
keys for bacteria based on colour, shape, physiology, disease
production, nutrition and other easily determinable characteristics to help in identification. Other bacteriologists
shared van Niel’s pessimistic attitude about the possibility
of designing a definitive natural classification of bacteria.
However, there was also a reaction against the tendency
to use semantic arguments as substitute for original new
approaches, which were badly needed. None of the proposed
systems was wholly convincing, and the arbitrary nature of
the principles on which they were based made them the
target for constant revisions. In fact, it was inevitable that the
basis of a true natural classification of bacteria would remain
unsteady ‘inasmuch as the course of phylogeny will always
remain unknown’ (Kluyver & van Niel, 1936). In the words
of Bruce White, ‘the present call is not for newer, more
ingenious, more pretentious, systems of classification, but
for patient and incisive investigation’ (White, 1937), a
criticism fully supported by van Niel.
The situation of bacterial taxonomy in the
1960s: the ‘Pseudomonas story’
That the unsatisfactory situation of bacterial classification
systems remained essentially unchanged for a long time
is clearly shown in a joint paper by Stanier and van Niel
published as late as 1962, where an analysis of the concept of
a bacterium is presented along with the admission that the
authors ‘have become sceptical about the value of developing formal taxonomic systems for bacteria’ (Stanier & van
Niel, 1962). Very little was anticipated here of the change
that was in store in the development of a Pseudomonas
taxonomic project a few years hence in the laboratory of one
of the authors!
The Pseudomonas story has been told many times, and
the reader interested in the details is referred to the citations included in other reports (Palleroni, 1984, 1993), but
following the opinion of T. S. Eliot, who said that ‘the
maddening thing is that we only have one thing to say, but
must keep on finding new forms of saying it’, a brief
repetition may not be out of place here.
A project was launched in the mid-1960s mainly by my
response to Roger Stanier’s insistence to look for some
improvement in the chaotic state of the definition and
classification of the species of the important genus
Pseudomonas, some of which had been the subjects of
http://mic.sgmjournals.org
intense biochemical research in the Department of
Bacteriology of the University of California at Berkeley for
a number of years.
During the initial stage of this project, strains of different
species were subjected to many phenotypic tests, the most
important of which was the extensive nutritional screening
that had been suggested as a taxonomic tool by den Dooren
de Jong many years before (den Dooren de Jong, 1926). The
results of this study were summarized in a subdivision of
the collection into a number of species and species groups
defined on purely phenotypic grounds (Stanier et al., 1966).
The studies were later supplemented with detailed analysis of
some of the species groups, such as the ‘mallei-pseudomallei’
species (Redfearn et al., 1966), the ‘diminuta-vesicularis’
complex (Ballard et al., 1968), the ‘stutzeri’ group (Palleroni
et al., 1970) and the ‘cepacia’ group (Ballard et al., 1970),
studies on ‘Pseudomonas’ solanacearum (Palleroni &
Doudoroff, 1971) and description of the related species
‘Pseudomonas’ pickettii (Ralston et al., 1973), the ‘fluorescens’
group (Barrett et al., 1986; Champion et al., 1980; Johnson &
Palleroni, 1989; Palleroni et al., 1972), the ‘alcaligenes’ group
(Ralston-Barrett et al., 1976) and the hydrogenomonad
species (Ralston et al., 1972). The interest manifested by
plant pathologists in these studies resulted in some of the
early interesting papers on the phytopathogens assigned
to the genus (Misaghi & Grogan, 1969; Sands et al., 1970;
Pecknold & Grogan, 1973).
DNA–DNA hybridization experiments on the above groups
were performed with DNA immobilized on membranes,
following a methodology developed for other groups of
organisms (Johnson & Ordal, 1968), and the results were a
general confirmation of the phenotypic classification. However, the wide range of ‘DNA homology’ values suggested a
high degree of genomic heterogeneity among the species
assigned to the genus, and this required the exploration of an
area that was quite unfamiliar to most bacterial taxonomists.
While these experiments were in progress, Roger Stanier
decided to leave the country to accept a position in the
Institut Pasteur in Paris, and in my discussions with Mike
Doudoroff I manifested the need for further investigation by
concentrating our attention on conservative regions of the
genome. A few years before, work in two laboratories (Doi &
Igarashi, 1965; Dubnau et al., 1965) had demonstrated
that the genes of rRNA met this criterion. The demonstration of the conservative nature of rRNA genes had been
focused on the properties of strains of Bacillus, and a further
confirmation came from experiments that included strains
of several bacterial genera (Pace & Campbell, 1971a, b).
These two interesting contributions were not, however,
aimed at the solution of taxonomic problems, but rather at
the demonstration of residual rRNA similarities in distantly
related organisms, which, by the way, could be readily
differentiated by following classical methods of taxonomic analysis. Organisms as distant as Escherichia coli and
Bacillus stearothermophilus still preserved detectable levels
of sequence similarity in these genes (Pace & Campbell,
3
N. J. Palleroni
Downloaded from www.microbiologyresearch.org by
IP: 130.203.136.75
On: Sun, 11 Sep 2016 21:04:19
1971b). In contrast, the application of the rRNA hybridization
technique to Pseudomonas species was of a more uncertain
nature, since all the organisms that had been assigned to this
genus shared many basic morphological and physiological
properties, and nothing seemed to guarantee a priori any
degree of success. However, the idea of using it to resolve the
genomic complexity that seems to underline the properties
of Pseudomonas species in the DNA–DNA hybridization
experiments appeared to be particularly attractive.
We adapted to the rRNA–DNA hybridization experiments
the competition method that we had used for our DNA–
DNA experiments, and the results were strikingly clear.
Pseudomonas, as classically defined, became subdivided into
five well-defined rRNA homology groups which deserved
at least five independent generic designations. The main
conclusions were already available by the end of 1970, and
were incorporated by Roger Stanier in the text of his
presentation at a Congress of General Microbiology that
took place in Mexico in 1971 (Stanier, 1971). He was
pleasantly surprised when he heard the success of the line
followed by us and of course, the tone of his comments at the
congress differed significantly from that of his publication in
collaboration with van Niel a few years before. ‘Interestingly
enough’, Stanier said, ‘it was van Niel (1946) who first
clearly pointed out the shortcomings of the deductive phylogenetic approach to bacterial classification. Today, there are
few bacteriologists who still believe in its utility. This does
not necessarily mean, however, that evolutionary considerations should have no place in bacterial taxonomy’. And he
goes on referring to the possibility of defining phylogenetic
relationships by experiments involving rRNA. The actual
results of the rRNA–DNA hybridization experiments were
published later (Palleroni et al., 1973).
Further developments
The work on Pseudomonas has had two important consequences, which will be discussed in the following
paragraphs. One was the proposal of a methodology for
the extensive phenotypic characterization of prokaryotes,
particularly recommended for chemoheterotrophic organisms, and the second, of a far more general applicability,
was the demonstration of the power of rRNA similarity
studies and their introduction as taxonomic tools, since the
strategies adopted for the Pseudomonas species became
extended beyond the limits of a single bacterial genus.
Phenotypic studies
The long phenotypic report (Stanier et al., 1966) was so well
received that soon it reached the level of a ‘citation classic’.
In little more than a decade, the number of citations reached
a figure well above 1000, which is rather uncommon for a
paper on a taxonomic subject. Its success as a descriptive
tool was in part due to its applicability to the study of other
Gram-negative organisms and even selected Gram-positive
groups (Hagedorn & Holt, 1975). The importance of the
contribution was acknowledged by many other workers, and
4
it was cited even recently as a good example of a biodiversity
study (Spiers et al., 2000).
As expected, the methodology of phenotypic studies, which
are very labour demanding, particularly with respect to
nutritional screenings, has been simplified by means of the
use of commercial kits. The simplification did not necessarily represent an improvement. Our nutritional studies
included direct observation of growth on plates, and in
many cases they benefited from the expertise of Mike
Doudoroff in carbohydrate metabolism, the vast knowledge
of Roger Stanier in the degradation of aromatic compounds,
and Ed Adelberg’s familiarity with the degradation of amino
acids by E. coli. Frequently, the results of the nutritional tests
were supplemented with observations on probable pathways
to be included in more detailed studies. It is obvious,
however, that these conditions are no longer found in most
cases. The nutritional properties are reported as results
obtained with commercial kits designed to determine the
utilization of a long list of compounds, supplemented by a
number of enzymic reactions. They are presented without
further discussion on their meaning, and are usually
accepted without discussion by editors and reviewers alike.
Studies on rRNA
Here the situation is just the opposite. With some modifications, the hybridization method was used for more
extensive taxonomic studies of the pseudomonads (De Vos
et al., 1985, 1989; De Vos & De Ley, 1983), but nowadays
studies of rRNA similarities are no longer practised with the
cumbersome hybridization methods, but with the more
direct and informative methodology of sequencing. Starting
with the sequences of oligonucleotides (Fox et al., 1977) and
the determination of ‘signature oligonucleotides’ (Woese
et al., 1985), the total 16S rRNA sequence after PCR
amplification is the procedure of choice.
The paper describing our rRNA experiments (Palleroni et al.,
1973) included, among the most striking findings, the fact
that some of the rRNA groups of pseudomonads were closer
to species of other genera (Escherichia, Xanthomonas) than
to members of other Pseudomonas groups. At that time, the
surprising suggestion that some species of Pseudomonas
appeared to be closer to the enteric bacteria than to other
species of the same genus would have been flatly rejected
by most bacteriologists. I am quite convinced that the
new findings not only helped substantially to unravel the
heterogeneity of a single bacterial genus but, in addition,
they contributed to inspiring spectacular developments in
other laboratories by the use of hybridization techniques
or of more direct refinements, such as oligonucleotide
similarity determinations.
The advent of rRNA sequence comparisons (going from
the crude rRNA–DNA hybridization methods to the actual
16S gene sequencing) had some unfortunate consequences,
because soon they became the main – if not the only –
criterion for the estimation of phylogenetic relationships,
Microbiology 149
Taxonomy in the 20th century
and later for the development of methods purported to
avoid culture isolation procedures in studies of prokaryote
diversity. The unjustified climate of optimism had its roots
in extending the relationships of single macromolecules
to those of the respective host cells. This, added to the
impossibility of inferring the physiological properties of
organisms on the basis of such information, provoked
animated discussions (Gest, 1999a, b; Meyer et al., 1986;
Palleroni, 1994, 1997; Postgate, 1995), in spite of which
the taxonomic literature was enriched by proposals of
many new taxa chiefly on the basis of 16S rRNA sequence
considerations.
Downloaded from www.microbiologyresearch.org by
IP: 130.203.136.75
On: Sun, 11 Sep 2016 21:04:19
Impact of the Pseudomonas studies
The basic methodology used in the Pseudomonas studies
has shaped the procedures used in the taxonomic analysis
of many groups of prokaryotes. This could have been
anticipated by some of the predictions involved in the
reactions shortly after the publication of the results. Thus
Johnson & Francis (1975) manifested that ‘although the
conservation of rRNA cistrons has been evident in all groups
of organisms that have been studied, detailed investigations
of bacterial groups have been limited. The notable exception
of this is the work of Palleroni et al. (1973) on the genus
Pseudomonas. Five species groups were delineated within the
genus on the basis of rRNA homology experiments. The
intra- and interhomologies of the Pseudomonas groups
are similar to those that we have found among the species
in the genus Clostridium. Although arbitrary cut off levels of
homology have been used in defining the homology groups,
a natural array of relatedness among the species is evident.
The correlations of phenotypic traits with the homology
groups suggest that rRNA homology results can be used as
an index of general relatedness and will provide a sound base
on which to construct taxonomic or maybe identifications
schemes.’ (My italics.) It is easy to appreciate the prophetic
quality of these comments at the present time, when taxonomic schemes based on rRNA similarities are commonplace
in prokaryotic systematic work.
rRNA methodology has been remarkably influential
(Woese, 1987) and 16S rDNA sequencing soon confirmed
the basic groupings defined by Palleroni et al. (1973). It
is only fair to mention here that before this time very
important taxonomic research had been done on the basis
of other cell components as subjects of chemotaxonomic
or immunological methods of analysis. These approaches,
however, did not have the general applicability of the nucleic
acid studies and their phylogenetic implications were not
always clear.
Of the valuable pieces of research that followed our work,
but done on a much larger collection of strains, the work
of De Ley and his group mentioned above is outstanding.
The results were essentially comparable with those of the
Berkeley group. The Pseudomonas work example was
followed after a short interval by the beginning of an era
of methodological euphoria so that the methods proposed
http://mic.sgmjournals.org
in recent years represent a magnificent display of ingenuity.
Making use of the enormous amount of information contained in some macromolecules, mainly nucleic acids and
proteins, which was added to a large volume of results of
very valuable studies already available on the cellular lipids
and the components of cell walls and outer membranes, the
phylogenetic relationships among prokaryote groups could
be traced. The list of procedures for these special studies on
Pseudomonas is so extensive that no detailed description will
be included here.
I am sure that up to this point I have been taxing the reader
who has had the tolerance of reading my previous reviews on
Pseudomonas. I guess, however, that in this communication
some sense of temporal perspective has been added that
allows us to see that, in the decade of the 1960s, a change
of mood in the participants occurred as a consequence
of having envisioned for the first time the possibility of
achieving phylogenetic classifications of bacteria. In fact,
one of the consequences of prior methodologies was evident
in the fact that successive editions of Bergey’s Manuals were
unable to define taxonomic hierarchies above the genus
level.
The term ‘pseudomonad’ still applies to all the organisms
that have a group of phenotypic properties common to
strains of the five rRNA homology groups and described
in the early paper of Stanier et al. (1966), and therefore it
makes little sense to try to single out ‘true’ (Busse et al.,
1989) or ‘genuine’ (Amann et al., 1996) members within the
group. In practice, the identification of a new isolate as a
Pseudomonas strain may not necessarily require the use of
methods beyond the capability of the laboratory facilities,
and the worker may consider the use of the excellent
enzymic approach developed by Jensen and his collaborators based on the peculiarities of the biosynthesis of
aromatic amino acids (Byng et al., 1983) or the cellular
fatty acid analysis procedures aptly described by Oyaizu &
Komagata (1983) and Stead (1992).
The Pseudomonas team at Berkeley became dispersed at a
moment when the obvious continuation would have been
to pursue further studies on bacterial phylogeny. Seen in
retrospect, the Pseudomonas example was just the tip of
an iceberg whose real dimensions are still being evaluated.
In his obituary notice of Mike Doudoroff, Stanier considered the projections of the work so significant that it ‘will
certainly remain one of the great landmarks of bacterial
taxonomy’ (Stanier, 1975). The circumstances had prevented him from participating in the final phase of the work,
but he liked to think of it as ‘the culmination of our most
ambitious and rewarding taxonomic enterprise’ (Stanier,
1980).
As to the future of taxonomic studies on the genus
Pseudomonas, there is no doubt that 16S rDNA sequencing
(and the sequencing of other highly conserved genes; see,
for example, Yamamoto et al., 2000) will continue to play
an important role, but for such studies to be biologically
5
N. J. Palleroni
Downloaded from www.microbiologyresearch.org by
IP: 130.203.136.75
On: Sun, 11 Sep 2016 21:04:19
meaningful they need to reflect the underlying genetic
structure of the strains and populations being studied. Until
recently most microbiologists assumed that populations
were strictly clonal and under such an assumption the
phylogeny of a single gene was taken to reflect the phylogeny
of a set of strains – if not a species. Genome sequencing and
rigorous population genetic studies have shown this need
not be true. Lateral gene transfer is far more pervasive than
once thought (Ochman et al., 2000) and for populations
undergoing even limited recombination a phylogeny based
on 16S rDNA may reflect little more than the phylogeny of
the 16S rDNA gene.
The solution of this problem and therefore, the problem of
meaningfully naming Pseudomonas, is not straightforward.
At the very least it requires a much better understanding
of the genetic structure of Pseudomonas populations.
The limited evidence so far indicates that populations of
Pseudomonas do exchange DNA (Haubold & Rainey, 1996;
Stover et al., 2000; Lomholt et al., 2001), but despite this they
are defined by clear genetic and phenotypic groupings.
Interestingly, the groups defined by one of these studies
(Haubold & Rainey, 1996) corresponded precisely to the
fundamental divisions originally described by Stanier et al.
(1966). The challenge for the future is not just to define and
name the biologically meaningful groups within the genus
Pseudomonas, but also to explain why, in the face of lateral
gene transfer, distinct genetic groupings are maintained.
Busse, H.-J., El-Banna, T. & Auling, G. (1989). Evaluation of
different approaches for identification of xenobiotic-degrading
pseudomonads. Appl Environ Microbiol 55, 1578–1583.
Byng, G. L., Johnson, J. L., Whitaker, R. J. & Jensen, R. A. (1983).
The evolutionary pattern of aromatic amino acid biosynthesis and
the phylogeny of pseudomonad bacteria. J Mol Ecol 19, 272–282.
Champion, A. B., Barrett, E. L., Palleroni, N. J., Soderberg, K. L.,
Kunisawa, R., Contopoulou, R., Wilson, A. C. & Doudoroff, M.
(1980). Evolution in Pseudomonas fluorescens. J Gen Microbiol 120,
485–511.
Cohn, F. (1872). Untersuchungen über Bakterien. Beitr Biol Pflanz 1,
127–224.
De Vos, P. & De Ley, J. (1983). Intra- and intergeneric similarities of
Pseudomonas and Xanthomonas ribosomal ribonucleic acid cistrons.
Int J Syst Bacteriol 33, 487–509.
De Vos, P., Goor, M., Gillis, M. & De Ley, J. (1985). Ribosomal
ribonucleic acid cistron similarities of phytopathogenic Pseudomonas
species. Int J Syst Bacteriol 35, 169–184.
De Vos, P., van Landschoot, A., Segers, P. & 9 other authors
(1989). Genotypic relationships and taxonomic localization of
unclassified Pseudomonas and Pseudomonas-like strains by deoxyribonucleic acid : ribosomal ribonucleic acid hybridizations. Int J Syst
Bacteriol 39, 35–49.
den Dooren de Jong, L. E. (1926). Bijdrage tot de kennis van het
Mineralisatieprocess. Rotterdam: Nijgh & Van Ditmar.
Doi, R. H. & Igarashi, R. T. (1965). Conservation of ribosomal and
messenger ribonucleic acid cistrons in Bacillus species. J Bacteriol
90, 384–390.
Dubnau, D., Smith, I., Porell, P. & Marmur, J. (1965). Genetic
conservation in Bacillus species and nucleic acid homologies. Proc
Natl Acad Sci U S A 54, 491–498.
Fox, G. E., Pechman, K. J. & Woese, C. R. (1977). Comparative
Acknowledgements
cataloging of 16S ribosomal ribonucleic acid: molecular approach to
procaryotic systematics. Int J Syst Bacteriol 27, 44–57.
I am grateful to Paul Rainey for his help in many useful discussions.
Gest, H. (1999a). Bacterial classification and taxonomy: a ‘primer’
for the new millennium. Microbiology Today 26, 70–72.
Gest, H. (1999b). Gest’s postulates. ASM News 65, 123.
References
Amann, R., Ludwig, W., Schulze, W., Spring, W., Moore, E. &
Schleifer, K.-H. (1996). r-RNA-targeted oligonucleotide probes for
Hagedorn, C. & Holt, J. G. (1975). A nutritional and taxonomic
survey of Arthrobacter soil isolates. Can J Microbiol 21, 353–361.
Haubold, B. & Rainey, P. B. (1996). Genetic and ecotypic structure
of a fluorescent Pseudomonas population. Mol Ecol 5, 747–761.
the identification of genuine and former pseudomonads. Syst Appl
Microbiol 19, 501–509.
Janke, A. (1924). Allgemeine Technische Mikrobiologie. I Teil: Die
Ballard, R. W., Doudoroff, M., Stanier, R. Y. & Mandel, M. (1968).
Johnson, J. L. & Francis, B. S. (1975). Taxonomy of the clostridia:
Mikroorganismen. Dresden & Leipzig: T. Steinkopf.
Taxonomy of the aerobic pseudomonads: Pseudomonas diminuta and
P. vesiculare. J Gen Microbiol 53, 349–361.
ribosomal ribonucleic acid homologies among the species. J Gen
Microbiol 88, 229–244.
Ballard, R. W., Palleroni, N. J., Doudoroff, M., Stanier, R. Y. &
Mandel, M. (1970). Taxonomy of the aerobic pseudomonads:
Johnson, J. L. & Ordal, E. J. (1968). Deoxyribonucleic acid homology
in bacterial taxonomy: effect of incubation temperature on reaction
specificity. J Bacteriol 95, 893–900.
Pseudomonas cepacia, P. marginata, P. alliicola and P. caryophylli.
J Gen Microbiol 60, 199–214.
Barrett, E. L., Solanes, R. E., Tang, J. S. & Palleroni, N. J. (1986).
Pseudomonas fluorescens biovar V: its resolution into distinct
component groups and the relationship of these groups to other
P. fluorescens biovars, to P. putida, and to psychrotrophic
pseudomonads associated with food spoilage. J Gen Microbiol 132,
2709–2721.
Beijerinck, M. W. (1921). Verzamelde Werken, vols I–V. Gravenhage,
Nijhoff. Vol. VI, ibid. 1940.
Bergey, D. H., Harrison, F. C., Breed, R. S., Hammer, B. W. &
Huntoon, F. M. (1923). Bergey’s Manual of Determinative
Bacteriology, 1st edn. Baltimore: Williams and Wilkins.
6
Johnson, J. L. & Palleroni, N. J. (1989). Deoxyribonucleic acid
similarities among Pseudomonas species. Int J Syst Bacteriol 39,
230–235.
Kluyver, A. J. & van Niel, C. B. (1936). Prospects for a natural
system of classification of bacteria. Zentralbl Bakteriol Parasitenkd
Infektionskr Hyg Abt II 94, 369–403.
Lehmann, K. B. & Neumann, R. O. (1896). Atlas und Grundriss der
Bakteriologie und Lehrbuch der Speciellen Bakteriologischen Diagnostik,
1st edn. München: J. F. Lehmann.
Lehmann, K. B. & Neumann, R. O. (1927). Bakteriologie insbesondere
Bakteriologische Diagnostik. II Allgemeine und spezielle Bakteriologie,
7 Aufl. Munchen: J. F. Lehmann.
Microbiology 149
Taxonomy in the 20th century
Linnaeus, C. (1753). Species Plantarum. Holmiae.
Lomholt, J. A., Poulsen, K. &Kilian, M. (2001). Epidemic poulation
structure of Pseudomonas aeruginosa: evidence for a clone that is
pathogenic to the eye and that has a distinct combination of
virulence factors. Infect Immun 69, 6284–6295.
Pringsheim, E. G. (1923). Zur Kritik der Bakteriensystematik. Lotos
71, 357–377.
Rahn, O. (1929). Contributions to the classification of bacteria. V–X.
Meyer, T. E., Cusanovich, M. A. & Kamen, M. D. (1986). Evidence
Zentralbl Bakteriol Parasitenkd Infektionskr Hyg Abt II 79, 321–343.
against use of bacterial aminoacid sequence data for construction
of all-inclusive phylogenetic trees. Proc Natl Acad Sci U S A 83,
217–220.
Rahn, O. (1937). New principles for the classification of bacteria.
Migula, W. (1900). Specielle Systematik der Bakterien. Jena: G. Fischer.
Misaghi, I. & Grogan, R. G. (1969). Nutritional and biochemical
comparisons of plant-pathogenic and saprophytic fluorescent pseudomonads. Phytopathology 59, 1436–1450.
Downloaded from www.microbiologyresearch.org by
IP: 130.203.136.75
On: Sun, 11 Sep 2016 21:04:19
Prévot, A.-R. (1933). Études de systématique bactérienne. I. Lois
générales. II. Cocci anaérobies. Ann Sci Nat Bot 10th Ser 15, 23–260.
Zentralbl Bakteriol Parasitenkd Infektionskr Hyg Abt II 96, 273–286.
Ralston, E., Palleroni, N. J. & Doudoroff, M. (1972). Deoxyribo-
nucleic acid homologies of some so-called ‘‘Hydrogenomonas’’ species.
J Bacteriol 109, 465–466.
Ralston, E., Palleroni, N. J. & Doudoroff, M. (1973). Pseudomonas
Ochman, H., Lawrence, J. G. & Groisman, E. A. (2000). Lateral gene
pickettii, a new species of clinical origin related to Pseudomonas
solanacearum. Int J Syst Bacteriol 23, 15–19.
transfer and the nature of bacterial innovation. Nature 405, 299–304.
Ralston-Barrett, E., Palleroni, N. J. & Doudoroff, M. (1976).
Orla-Jensen, S. (1909). Die Hauptlinien der natürlichen Bakterien-
systems. Zentralbl Bakteriol Parasitenkd Infektionskr Hyg Abt II 22,
305–346.
Phenotypic characterization and deoxyribonucleic acid homologies
of the Pseudomonas alcaligenes group. Int J Syst Bacteriol 26, 421–
426.
Oyaizu, H. & Komagata, K. (1983). Groupings of Pseudomonas
Redfearn, M. S., Palleroni, N. J. & Stanier, R. Y. (1966). A
species on the basis of cellular fatty acid compositions and the
quinone system with special reference to the existence of 3-hydroxy
fatty acids. J Gen Appl Microbiol 29, 17–40.
Sands, D. C., Schroth, M. N. & Hildebrand, D. C. (1970). Taxonomy
comparative study of Pseudomonas pseudomallei and Bacillus mallei.
J Gen Microbiol 43, 293–313.
Pace, B. & Campbell, L. L. (1971a). Homology of ribosomal
of phytopathogenic pseudomonads. J Bacteriol 101, 9–23.
ribonucleic acid of Desulfovibrio species with Desulfovibrio vulgaris.
J Bacteriol 106, 717–719.
Spiers, A. J., Buckling, A. & Rainey, P. B. (2000). The causes of
Pace, B. & Campbell, L. L. (1971b). Homology of ribosomal
Stanier, R. Y. (1971). Toward an evolutionary taxonomy of the
ribonucleic acid of diverse bacterial species with Escherichia coli and
Bacillus stearothermophilus. J Bacteriol 107, 543–547.
bacteria. In Recent Advances in Microbiology, pp. 595–604. Mexico.
Pseudomonas diversity. Microbiology 146, 2345–2350.
Stanier, R. Y. (1975). Obituary. Michael Doudoroff, 1911–1975.
Palleroni, N. J. (1984). Pseudomonas (Migula). In Bergey’s Manual of
ASM News 41, 737–738.
Systematic Bacteriology, vol. 1, pp. 141–199. Edited by N. R. Krieg &
J. G. Holt. Baltimore, MD: Williams & Wilkins.
Stanier, R. Y. (1980). The journey, not the arrival, matters. Annu Rev
Palleroni, N. J. (1993). Pseudomonas classification. A new case
Stanier, R. Y. & van Niel, C. B. (1962). The concept of a bacterium.
history in the taxonomy of Gram-negative bacteria. Antonie van
Leeuwenhoek 64, 231–251.
Stanier, R. Y., Palleroni, N. J. & Doudoroff, M. (1966). The aerobic
Microbiol 34, 1–48.
Arch Mikrobiol 42, 17–35.
Palleroni, N. J. (1994). Some reflections on bacterial diversity. ASM
pseudomonads: a taxonomic study. J Gen Microbiol 43, 159–271.
News 10, 537–540.
Stead, D. E. (1992). Grouping of plant-pathogenic and some other
Palleroni, N. J. (1997). Prokaryotic diversity and the importance of
culturing. Antonie van Leeuwenhoek 72, 3–19.
Pseudomonas spp. by using cellular fatty acid profiles. Int J Syst
Bacteriol 42, 281–295.
Palleroni, N. J. & Doudoroff, M. (1971). Phenotypic characterization
and deoxyribonucleic acid homologies of Pseudomonas solanacearum.
J Bacteriol 107, 690–696.
Stover, C. K., Pham, X. Q., Erwin, A. L. & 28 other authors (2000).
Palleroni, N. J., Doudoroff, M., Stanier, R. Y., Solanes, R. E. &
Mandel, M. (1970). Taxonomy of the aerobic pseudomonads: the
van Niel, C. B. (1946). The classification and natural relationships of
properties of the Pseudomonas stutzeri group. J Gen Microbiol 60,
215–231.
White, P. B. (1937). Remarks on bacterial taxonomy. Zentralbl
Palleroni, N. J., Ballard, R. W., Ralston, E. & Doudoroff, M. (1972).
Winogradsky, S. (1949). Microbiologie du Sol. Problèmes et Méthodes.
Deoxyribonucleic acid homologies among some Pseudomonas
species. J Bacteriol 110, 1–11.
Woese, C. R. (1987). Bacterial evolution. Microb Evol 51, 221–271.
Palleroni, N. J., Kunisawa, R., Contopoulou, R. & Doudoroff, M.
(1973). Nucleic acid homologies in the genus Pseudomonas. Int J Syst
Bacteriol 23, 333–339.
Pecknold, P. C. & Grogan, R. G. (1973). Deoxyribonucleic acid
homology groups among phytopathogenic Pseudomonas species. Int
J Syst Bacteriol 23, 111–121.
Postgate, J. (1995). Breathless niches. Nature 377, 26.
http://mic.sgmjournals.org
Complete genome sequence of Pseudomonas aeruginosa PAO1,
an opportunistic pathogen. Nature 406, 959–964.
bacteria. Cold Spring Harbor Symp Quant Biol 11, 285–301.
Bakteriol Parasitenkd Infektionskr Hyg Abt II 96, 145–149.
Cinquante Ans de Recherches. Paris: Masson et Cie.
Woese, C. R., Stackebrandt, E., Macke, T. J. & Fox, G. E. (1985).
A phylogenetic definition of the major eubacterial taxa. Syst Appl
Microbiol 6, 143–151.
Yamamoto, S., Kasai, H., Arnold, D. L., Jackson, R. W., Vivian, A. &
Harayama, S. (2000). Phylogeny of the genus Pseudomonas:
intrageneric structure reconstructed from the nucleotide sequences
of gyrB and rpoD genes. Microbiology 146, 2385–2394.
7