Candidatus Odyssella Thessalonicensis ' Gen. Nov., Sp. Nov., An Obligate Intracellular Parasite of Acanthamoeba Species
Candidatus Odyssella Thessalonicensis ' Gen. Nov., Sp. Nov., An Obligate Intracellular Parasite of Acanthamoeba Species
Candidatus Odyssella Thessalonicensis ' Gen. Nov., Sp. Nov., An Obligate Intracellular Parasite of Acanthamoeba Species
". The
suspension was incubated at 37 mC for 30 min then 5 l 1 mg
DNase-free RNase ml
3
)
Time (d)
.....................................................................................................
Fig. 2. Effect of temperature on the growth
and survival of uninfected (lled symbols)
and L13-infected (open symbols) A. poly-
phaga. 4, 5, 22 mC; >, =, 30 mC; , , 32 mC;
$, #, 37 mC.
Host range
Of the 23 strains of free-living amoebae in which
cultivation of L13 was attempted, only those belonging
to groups II and III of the genus Acanthamoeba were
susceptible to infection (Table 1). Multiplication of
L13 occurred best in A. polyphaga and Acanthamoeba
castellanii. Acanthamoeba comandoni, a member of the
group I Acanthamoeba species, was the only member
of the genus tested which was not susceptible to
infection.
Effect of temperature on bacterial virulence
A. polyphaga grew at temperatures between 22 and
37 mC, but rapidly encysted at 42 mC. As demonstrated
in Fig. 2, lower temperatures favoured better growth
of the amoeba, with cultures incubated at 22 mC being
over twice as dense as those grown at 37 mC. Higher
temperatures enhanced the virulence of L13. At 37 mC
virtually all amoebae were lysed within 5 d of bacterial
infection, whereas at 22 mC intact amoebae could be
observed over 3 weeks post-infection. Even growth at
32 or 30 mC resulted in destruction of virtually all
amoebae within 10 d of bacterial infection. At all of
the three higher temperatures tested, the rate of decline
of intact amoebae in infected cultures appeared to be
very similar and very rapid (taking 34 d), and
decreasing temperature served only to delay the onset
of this decline. However, at 22 mC the rate at which
amoebae were destroyed was markedly slower; after
an initial sharp rise, the number of intact amoeba fell
gradually throughout the 21 d of the experiment.
Importantly, viable L13-infected amoebae could be
recovered from the cultures incubated at 22 mC over 4
months after the experiment described above was
completed (data not shown).
16S rRNA gene amplication and analysis
A primary 16S rRNA gene sequence of 1454 bp was
determined for L13 and its validity was supported by
secondary structure modelling (data not shown). A
r:s1: search of the GenBank database indicated a
proteobacterial origin for the bacterium. The 16S
rRNA gene of L13 was found to be most similar to
those of members of the -Proteobacteria, sharing
between 80 and 84% similarity. However, within this
subclass, L13 did not share a specically high similarity
with any particular species, nor with the members of
any particular group.
The alignment used as a basis for phylogenetic analysis
comprised the L13 sequence, the sequences of diverse
representatives of the -Proteobacteria and those of
two -Proteobacteria (Escherichia coli and Legionella
brunensis). The latter two were included as out-
grouping species on which trees could be rooted. The
16S rRNA gene sequences available for some of the -
Proteobacteria were not as complete as that obtained
in this study, usually through the lack of data at the 3h
extremity of the gene. Some sequences also included a
relatively large number of base ambiguities. The length
of the alignment was thus limited to 1177 bp. An
evolutionary distance matrix was calculated from
pairwise comparison of sequences in this alignment
and a phylogenetic tree (Fig. 3) was inferred from this
matrix. The inclusion of L13 had no signicant eect
on the overall topology of this tree, which is in
concordance with those previously proposed (e.g. Loy
et al., 1996). The evolutionary position proposed for
L13 was within a cluster of obligate intracellular
bacteria, which, although within the -subclass of the
Proteobacteria, lies apart from any of the four
recognized subgroups (14) and the lineage carrying
International Journal of Systematic and Evolutionary Microbiology 50 67
R. J. Birtles and others
Azospirillum lipoferum
Bartonella bacilliformis
Beijerinckia indica
Afipia felis
Erythrobacter longus
Ehrlichia sennetsu
Rickettsia prowazekii
Rickettsia rickettsii
L13
Holospora obtusa
NHP bacterium
200/200*
151/147*
178/110*
102/155*
144/70*
200/200*
179/183*
173/195*
184/200*
200/200*
199/196*
132/182
131/
110*
200/193*
184/132
184/132*
200/200*
149/184*
51/132*
Magnetospirillum magnetotacticum
Rhodospirillum rubrum
Rhodobacter capsulatus
Rhodobacter sphaeroides
Agrobacterium tumefaciens
Sphingomonas paucimobilis
Wolbachia pipientis
Caedibacter caryophilus
Escherichia coli
Legionella brunensis
1
3
2
4
Ri
.....................................................................................................
Fig. 3. Phylogenetic tree comparing L13
with various members of the - and -
Proteobacteria. The numbers at each node
are the results of bootstrapping; each value
is derived from 200 samples as an indication
of semi-statistical support for the proposed
branching order. The rst value is derived
from distance matrix analysis and the second
value is derived from parsimony analysis.
Values of 140 (70% of samples) are
usually considered as indicating good
support for the proposed branching order at
that node. An asterisk appearing after the
two bootstrap values indicates that the
cluster supported by that node was also
proposed by maximum-likelihood analysis.
The vertical bars to the right of the tree
indicate representatives of various subclasses
of the Proteobacteria (14, ) and
representatives of the Rickettsiales (Ri). Scale
bar, 0n04 K
nuc
.
members of the Rickettsiales (Fig. 3). A specic
evolutionary relatedness between L13 and the three
other members of the identied cluster (H. obtusa,
C. caryophilus and the NHP bacterium) was also
supported by parsimony- and maximum-likelihood-
based phylogenetic inferences, although in none of
these three reconstructions was there overwhelming
support from bootstrapped sampling (see Fig. 3).
Likewise, although all three inferences suggested that
the L13-containing cluster shared specic evolutionary
descent with the Rickettsiales lineage, bootstrap sup-
port was again only moderately strong. However,
within the cluster dened above, L13 branched apart
from the lineage carrying the other three members, a
divergence which was well supported (Fig. 3).
Analysis of cellular proteins
SDS-PAGE proles obtained for L13 are presented in
Fig. 4. The proles obtained for puried bacteria were
distinct from those of uninfected amoebae, demon-
strating the success of the purication protocol used.
Unpuried bacteria yielded a prole which, although
dominated by L13-specic bands, also included
amoebal proteins. The proles yielded by untreated
bacteria included heavy smears at the top of the gel,
indicating the presence of very large proteins that were
unable to enter the gel. These proles also did not have
clearly distinguishable bands larger than 36 kDa with
the exception of one of an estimated 39 kDa. However,
below this size numerous strongly staining bands were
present. There was little dierence between the proles
derived from intact and broken bacteria, although it
appeared that a protein of approximately 33 kDa was
more abundantly liberated from broken cells (Fig. 4).
Heat treatment resulted in denaturation of very high
molecular mass proteins to yield proles more complex
than those derived from untreated organisms. A
number of larger bands appeared in the proles,
including an intensely staining protein of about
50 kDa. Two further particularly strongly staining
bands were present at approximately 25 and 18 kDa,
although at least 12 other bands were also discernible.
Bacteria incubated with Laemmli buer at 37 mC for
68 International Journal of Systematic and Evolutionary Microbiology 50
An intracellular parasite of free-living amoebae
1 2 3 4 5 6 7 8 9 10 11 12 13 14
kDa
107
76
52
36
27
19
kDa
76
52
36
27
19
.....................................................................................................
Fig. 4. SDS-PAGE proles derived from L13
and A. polyphaga. Samples in lanes 25
were boiled prior to loading, those in 69
were held at 37 mC for 30 min and those in
lanes 1013 were untreated. Lanes: 2, 6
and 10, puried intact L13; 3, 7 and 11,
puried non-intact L13; 4, 8 and 12, A.
polyphaga heavily infected with L13; 5, 9
and 13, uninfected A. polyphaga; 1 and 14,
molecular mass markers.
30 min yielded proles which, in general, consisted
both of bands present in proles derived from
untreated bacteria and bands present in proles
derived from boiled bacteria. However, only very few
bands were present in proles derived from both
untreated and heat-treated bacteria (see Fig. 4), with
the co-migrating bands of approximately 33 and
27 kDa being perhaps the best candidates for heat-
stable proteins.
Estimate of mean base composition
The GjC content of L13 was estimated to be
41p1 mol %.
DISCUSSION
The ability to exploit protozoan hosts oers bacteria
access to a protected, nutrient-rich niche suitable for
long-term survival in the environment. That this
opportunity has been seized by a wide spectrum of
bacteria is now becoming more apparent. Protozoa-
incorporating in vitro co-cultivation methods are
known to be able to support an increasing number of
recognized pathogens and, following the introduction
of PCR-based characterization methods, an increasing
number of newly recognized obligate intraprotozoal
organisms are being reported. In this study we add to
this knowledge by reporting the isolation and charac-
terization of another previously unrecognized bac-
terium.
Although the current taxonomic criteria for the cre-
ation of a new species are well-dened (Murray et al.,
1990; Stackebrandt & Goebel, 1994), those required
for the proposal of a new genus are more subjective
(Murray et al., 1990). Thus, whereas the moderate
level of 16S rRNA sequence similarity shared by L13
with previously recognized species can be used to
justify its proposal as a new species, that this new
species warrants placement in a new genus is another
question. Perhaps the best way of resolving this issue is
to assess whether L13 can be satisfactorily accommo-
dated within the genera with which it shares most
evolutionary homology, namely Holospora and
Caedibacter. Although both these genera are com-
prised endosymbionts of protozoa, these have to date
been associated solely with infection of ciliates.
Holospora species specically infect the micronucleus
or macronucleus of their Paramecium hosts and
possess a life cycle comprised two distinct morpho-
logical forms. In a refractile form, the bacteria form
short (13 m) fusiform rods, which undergo binary
ssion and give rise to long infective forms of up
to 20 m (Preer & Preer, 1984). The GjC content
has not been measured for any Holospora species.
Caedibacter are moderately sized (up to 1 m diam.
and 4 m long) rod-shaped or coccobacilli which are
characterized by possession of distinctive refractile
inclusions, termed Rbodies. Fromthese descriptions it
is clear that L13 is somewhat smaller than Holospora
and Caedibacter species and shares none of the
distinctive morphological features of either genus.
Although somewhat pleomorphic, L13 present a typi-
cal Gram-negative rod-like morphology. No elongated
forms comparable to Holospora infective forms were
observed and no refractile bodies characteristic of
Caedibacter were present within L13 bacteria. Fur-
thermore, the host range of L13 was limited to
Acanthamoeba species. Although the GjC content of
L13 was found to be similar to that reported for
Caedibacter species (Preer & Preer, 1984), this cannot
International Journal of Systematic and Evolutionary Microbiology 50 69
R. J. Birtles and others
be used to justify that the organisms share a specic
taxonomic relationship. A low GjC content is in
common with many species both within the -
Proteobacteria and beyond. Thus, from these com-
parisons it is clear that the proposal of a new genus to
accommodate L13 is undoubtedly justied.
Whether L13 is the same organism as any of those
observedin previous studies is dicult to assess. Proca-
Ciobanu et al. (1975) reported Acanthamoeba SN
endosymbionts which were Gram-negative, intra-
cytoplasmic and of a similar size to L13. An
Acanthamoeba endosymbiont (HN-3) described by
Hall & Voelz (1985) was very similar in appearance to
the SN endosymbionts but was surrounded by an
electron-translucent area, similar in appearance to that
of L13. Hall & Voelz (1985) demonstrated that this
translucent area resulted fromthe presence of bacterial
capsular material. Although we have not investigated
if the area surrounding L13 is of the same origin, we
did note that in several electron micrographs L13 did
not possess this translucent halo and that this absence
was more common when L13 was cultured at higher
temperatures (37 mC). More recently, Fritsche et al.
(1993) reported two forms of endosymbionts in
Acanthamoeba. One, a coccal form, appeared indis-
tinguishable from Candidatus Parachlamydia acanth-
amoebae (Amann et al., 1997), whereas the second
bore a marked microscopic and ultrastructural resem-
blance to L13. This organism was, however, reported
to be non-motile. The same workers have subsequently
reported preliminary 16S rRNA sequence data for
these two organisms (Gautom & Fritsche, 1996) ; the
rst form was indeed conrmed as sharing high levels
of sequence similarity with the sequences of Chlamydia
and Parachlamydia species, whereas the second was
found to share highest similarity levels with the
Rickettsiales. Although distinct fromthe Rickettsiales,
we have shown that L13 may well share an evol-
utionary lineage with -proteobacterial members of
this polyphyletic family and that these species are
among those with which L13 shares the highest 16S
rRNA similarity levels.
The results we report add to other recent ndings in
broadening the evolutionary spectrumof bacteria now
recognized as being capable of exploiting Acanth-
amoeba as hosts. The importance of these protozoa to
the natural cycles of legionellae is nowwell-established
(Barker & Brown, 1994) and 16S rRNA methods have
been used to demonstrate a diversity of Legionella
species which can be recovered using in vitro acanth-
amoebal co-cultivation methods, but not axenic media
(Birtles et al., 1996). More recently, Amann et al.
(1997) used phylogenetic methods to demonstrate the
Chlamydia-like nature of another endosymbiont
and named it Candidatus Parachlamydia acanth-
amoebae. Thus, it is now apparent that bacteria
from at least three distinct lineages have evolved
mechanisms for the exploitation of Acanthamoebae. In
addition to these well-characterized organisms, other
amoebal endosymbionts have been observed, notably
an Ehrlichia-like bacteria (Michel et al., 1995b) and
even possible Archaea (Homann et al., 1998). In light
of these ndings, a phylogenetic assessment of such
organisms would be interesting and we are pursuing
these goals.
Evidence for a wider role of protozoa as natural hosts
for pathogenic bacteria is also supported by numerous
laboratory studies. Legionella pneumophila, Listeria
monocytogenes, Vibrio cholerae, Pseudomonas aeru-
ginosa, Burkholderia pickettii and Edwardsiella tarda
have all been shown to be able to multiply within
laboratory protozoal cultures (King & Shotts, 1988;
Ly & Mu$ ller, 1990; Michel et al., 1995a; Michel &
Hauro$ der, 1997; Rowbotham, 1980; Thom et al.,
1992), whereas other species, including Mycobacterium
leprae, Pseudomonas and Bacillus species, some
coliforms and Chlamydia pneumoniae (Jardin,
1975; Tyndall et al., 1991; King et al., 1988; Essig et
al., 1997) have been shown to be able to survive
protozoal uptake and to be viably maintained within
them. Additionally, evidence of the pathogenic po-
tential of Candidatus Parachlamydia acanthamoebae
and intra-amoebal legionellae in atypical pneumonia
has now been presented (Fry et al., 1991; Birtles et al.,
1997). It also appears that such endosymbionts are
common. Fritsche et al. (1993) observed intra-Acanth-
amoeba bacteria in 14 out of 57 (24%) clinical and
environmental isolates examined. Taken together,
these ndings add further weight to previous indi-
cations (Amann et al., 1997) that the role of protozoa,
and possibly small free-living amoebae in particular, in
the epidemiology of human diseases may presently be
signicantly underrated.
Investigation of the eect of temperature on the
virulence of L13 led to some interesting observations.
Higher incubation temperatures resulted in enhanced
virulence, whereas at 22 mC, L13 appeared to form a
far more stable relationship with its host. Whether the
virulence of L13 itself is enhanced at temperatures over
30 mC, or whether its apparent virulence results from
physiological changes within its host is not known.
It is clear from our experiments that uninfected
A. polyphaga did not grow as well at elevated
temperatures and these suboptimal conditions may
lead to disruption of the hostparasite equilibriumand
the onset of bacterial virulence. The longevity of the
L13host relationship at lower temperatures has not
been assessed, although we have observed it to be
ongoing after 4 months of co-cultivation.
Although we have applied polyphasic methods to the
characterization of L13, its obligate intracellular
nature has prevented the unequivocal isolation of a
unique strain for proposal as the type strain. Thus,
according to Murray & Schleifer (1994), we propose
provisional classication of L13 as Candidatus
Odyssella thessalonicensis gen. nov., sp. nov. The
genus name is derived from the eloquent analogy
drawn by Barker & Brown (1993) of protozoa as
Trojan horses of the microbial world. Odysseus is the
70 International Journal of Systematic and Evolutionary Microbiology 50
An intracellular parasite of free-living amoebae
Latin name for the warrior who, according to Greek
mythology, thought up the strategy of penetrating the
walls of Troy by hiding himself and fellow Greek
warriors inside the infamous wooden horse. The
species name reects the location from where the
isolate was obtained, the Greek city of Thessalonika.
Description of Candidatus Odyssella
thessalonicensis gen. nov., sp. nov.
Candidatus Odyssella thessalonicensis (Od.ys.selhla.
dim. fem. n. pertaining to Odysseus; thess.al.lonhi.
cenhsis. M.L. masc. n. adj. pertaining to the Greek
city Thessalonika from where the organism was
isolated).
Obligate intracellular bacterium of Acanthamoeba
species which cannot be cultivated on cell-free media,
only by co-cultivation with a number of Acanthamoeba
species at temperatures at and below 37 mC. At
temperatures over 30 mC the organism multiplies rap-
idly destroying its host, whereas at lower temperatures,
it is far less virulent. Host range appears limited to
group II and III Acanthamoeba species. Organism is
motile and microscopic appearance is typically Gram-
negative and rod-shaped. Electron microscopy reveals
that the organism lies intracytoplasmically and is
usually surrounded by an electron-translucent layer in
which vesicle-like structures derived from the cell
envelope can be observed. The GjC content is
41 mol %. The phylogenetic position, inferred from
comparison of the 16S rRNA gene sequence, is within
the -Proteobacteria.
REFERENCES
Amann, R., Springer, N., Ludwig, W., Go$ rtz, H.-D. & Schleifer,
K.-H. (1991). Identication in situ and phylogeny of uncultured
bacterial endosymbionts. Nature 351, 161164.
Amann, R., Springer, N., Scho$ nhuber, W., Ludwig, W., Schmid,
E. N., Mu$ ller, K.-D. & Michel, R. (1997). Obligate intracellular
bacterial parasites of acanthamoebae related to Chlamydia spp.
Appl Environ Microbiol 63, 115121.
Barker, J. & Brown, M. R. W. (1994). Trojan horses of the
microbial world: protozoa and the survival of bacterial
pathogens in the environment. Microbiology 140, 12531259.
Birtles, R. J., Rowbotham, T. J., Raoult, D. &Harrison, T. G. (1996).
Phylogenetic diversity of intra-amoebal legionellae as revealed
by 16S rRNA gene sequence comparison. Microbiology 142,
35253530.
Birtles, R. J., Rowbotham, T. J., Storey, C., Marrie, T. J. & Raoult,
D. (1997). Chlamydia-like obligate parasite of free-living
amoebae. Lancet 349, 925926.
De Jonckheere, J. (1977). Use of an axenic medium for
dierentiation between pathogenic and non-pathogenic
Naegleria fowleri isolates. Appl Environ Microbiol 33, 751.
Dessen, P., Fondrat, C., Valencien, C. & Mugnier, C. (1990).
nis:Ncr: a French service for access to biomolecular sequence
databases. Comput Appl Biosci 6, 355356.
Essig, A., Heinemann, M., Simnacher, U. & Marre, R. (1997).
Infection of Acanthamoeba castellanii by Chlamydia pneu-
moniae. Appl Environ Microbiol 63, 13961399.
Fallon, R. J. & Rowbotham, T. J. (1990). Microbiological investi-
gations into an outbreak of Pontiac fever due to Legionella
micdadei associated with the use of a whirlpool. J Clin Pathol
43, 479483.
Fitch, W. M. & Margoliash, E. (1967). Construction of phylo-
genetic trees. A method based on mutation distances as
estimated from cytochrome c sequences is of general appli-
cability. Science 155, 279284.
Fritsche, T. R., Gautom, R. K., Seyedirashti, S., Bergeron, D. L. &
Lindquist, T. D. (1993). Occurrence of bacterial endosymbiotes in
Acanthamoeba spp. isolated from corneal and environmental
specimens and contact lenses. J Clin Microbiol 31, 11221126.
Fry, N. K., Rowbotham, T. J., Saunders, N. A. & Embley, T. M.
(1991). Direct amplication and sequencing of the 16S rDNAof
an intracellular Legionella species recovered by amoebal en-
richment from sputum of a patient with pneumonia. FEMS
Microbiol Lett 83, 165168.
Gautom, R. K. & Fritsche, T. R. (1996). Characterisation of
bacterial endosymbiotes of Acanthamoeba spp. In Abstracts of
the 7th International Conference on Small Freeliving Amoebae,
712th January 1996. Womens and Childrens Hospital, North
Adelaide, South Australia.
Hall, J. & Voelz, H. (1985). Bacterial endosymbiotes of
Acanthamoeba sp. J Parasitol 71, 8995.
Higgins, D., Bleasby, A. & Fuchs, R. (1992). ciis1:i v: improved
software for multiple sequence analysis. Comput Appl Biosci 8,
188192.
Hoffmann, R., Michel, R., Mu$ ller, K.-D. & Schmid, E. N. (1998).
Archaea like endocytobiotic organisms isolated from Acanth-
amoeba sp. (gr II). Endocytobiosis Cell Res 12, 185188.
Jardin, J. B. (1975). Amibes limax vecteurs possible de Myco-
bacteries et de M. leprae. Acta Leprol 59, 5767.
Jukes, T. H. & Cantor, C. R. (1969). Evolution of protein
molecules. In Mammalian Protein Metabolism, pp. 21132.
Edited by N. H. Munro. New York: Academic Press.
King, C. H. & Shotts, E. B. (1988). Enhancement of Edwardsiella
tarda and Aeromonas salmonicida through ingestion by the
ciliated protozoan Tetrahymena pyriformis. FEMS Microbiol
Lett 51, 95100.
King, C. H., Shotts, E. B., Wooley, R. E. & Porter, K. G. (1988).
Survival of coliforms and bacterial pathogens within protozoa
during chlorination. Appl Environ Microbiol 54, 30233033.
Laemmli, U. K. (1970). Cleavage of structural proteins during the
assembly of the head of bacteriophage T4. Nature 227, 680685.
Lane, D. J. (1991). 16S\23S rRNA sequencing. In Nucleic Acid
Techniques in Bacterial Systematics, pp. 115176. Edited by E.
Stackebrandt & M. Goodfellow. Chichester: Wiley.
Loy, J. K., Dewhirst, F. E., Weber, W., Frelier, P. F., Garbar, T. L.,
Tasca, S. I. & Templeton, J. W. (1996). Molecular phylogeny and
in situ detection of the etiological agent of necrotizing hepato-
pancreatitis in shrimp. Appl Environ Microbiol 62, 34393445.
Ly, T. M. C. & Mu$ ller, H. E. (1990). Ingested Listeria mono-
cytogenes survive and multiply in protozoa. J Med Microbiol
33, 5154.
Michel, R. & Hauro$ der, B. (1997). Isolation of an Acanthamoeba
strain with intracellular Burkholderia pickettii. Zentbl Bakt 285,
541557.
Michel, R., Burghardt, H. & Bergmann, H. (1995a). Natu$ rliche
intracellula$ re infectionen bei acanthamoeben mit Pseudomonas
aeruginosa nach ihrer isolierung aus einer mikrobiologisch
beanstandeten trinkwasser-hausinstallation eines kranken-
hauses. Zentbl Hyg 196, 532544.
International Journal of Systematic and Evolutionary Microbiology 50 71
R. J. Birtles and others
Michel, R., Mu$ ller, K.-D. & Schmid, E. N. (1995b). Ehrlichia-like
organisms (KSL1) observed as obligate intracellular parasites
of Saccamoeba species. Endocytobiosis Cell Res 11, 6980.
Murray, R. G. E. & Schleifer, K. H. (1994). Taxonomic notes: a
proposal for recording the properties of putative taxa of
procaryotes. Int J Syst Bacteriol 44, 174176.
Murray, R. G. E., Brenner, D. J., Colwell, R. R., De Vos, P.,
Goodfellow, M., Grimont, P. A. D., Pfennig, N., Stackebrandt, E.
& Zavarzin, G. A. (1990). Report of the ad hoc committee on
approaches to taxonomy within the Proteobacteria. Int J Syst
Bacteriol 40, 213215.
Neefs, J.-M., Van de Peer, Y., De Rijk, P., Chapelle, S. & De
Wachter, R. (1993). Compilation of small ribosomal subunit
RNA structures. Nucleic Acids Res 21, 30253049.
Owen, R. J., Legros, R. M. & Lapage, S. P. (1978). Base com-
position, size and sequence similarities of genome deoxy-
ribonucleic acids from clinical isolates of Pseudomonas
putrefaciens. J Gen Microbiol 104, 127138.
Page, F. C. (1976). An Illustrated Key to Freshwater and Soil
Amoeba. Ambleside, Cumbria: Freshwater Biological Associ-
ation.
Preer, J. R., Jr & Preer, L. B. (1984). Endosymbionts of protozoa.
In Bergeys Manual of Systematic Bacteriology, vol. 1, pp.
795811. Edited by N. R. Krieg & J. G. Holt. Baltimore:
Williams & Wilkins.
Proca-Ciobanu, M., Lupascu, G., Petrovici, A. & Ionescu, M. D.
(1975). Electron microscopic study of a pathogenic Acanth-
amoeba castellani strain: the presence of bacterial endo-
symbiotes. Int J Parasitol 5, 4956.
Rowbotham, T. J. (1980). Preliminary report on the patho-
genicity of Legionella pneumophila for freshwater and soil
amoebae. J Clin Pathol 33, 11791183.
Rowbotham, T. J. (1983). Isolation of Legionella pneumophila
from clinical specimens via amoebae, and the interaction of
those and other isolates with amoebae. J Clin Pathol 36,
978986.
Springer, N., Ludwig, W., Amann, R., Schmidt, H. J., Go$ rtz, H.-D.
&Schleifer, K.-H. (1993). Occurrence of fragmented 16S rRNAin
an obligate bacterial endosymbiont of Paramecium caudatum.
Proc Natl Acad Sci USA 90, 98929895.
Stackebrandt, E. & Goebel, B. M. (1994). Taxonomic note: a
place for DNA: DNA reassociation and 16S rDNA sequence
analysis in the present species denition in bacteriology. Int J
Syst Bacteriol 44, 846849.
Thom, S., Warhurst, D. & Drasar, B. S. (1992). Association of
Vibrio cholerae with fresh water amoebae. J Med Microbiol 36,
303306.
Tyndall, R. L., Ironside, K. S., Little, C. D., Katz, D. S. & Kennedy,
J. R. (1991). Free-living amoebae used to isolate consortia
capable of degrading trichloroethylene. Appl Biochem Bio-
technol 28/29, 917925.
72 International Journal of Systematic and Evolutionary Microbiology 50