Phylogenetic position of the North American isolate of Pasteuria that parasitizes the soybean cyst nematode, Heterodera glycines, as inferred from 16S rDNA sequence analysis.

Abstract

A 1341 bp sequence of the 16S rDNA of an undescribed species of Pasteuria that parasitizes the soybean cyst nematode, Heterodera glycines, was determined and then compared with a homologous sequence of Pasteuria ramosa, a parasite of cladoceran water fleas of the family Daphnidae. The two Pasteuria sequences, which diverged from each other by a dissimilarity index of 7%, also were compared with the 16S rDNA sequences of 30 other bacterial species to determine the phylogenetic position of the genus Pasteuria among the Gram-positive eubacteria. Phylogenetic analyses using maximum-likelihood, maximum-parsimony and neighbour-joining methods showed that the Heterodera glycines-infecting Pasteuria and its sister species, P. ramosa, form a distinct line of descent within the Alicyclobacillus group of the Bacillaceae. These results are consistent with the view that the genus Pasteuria is a deeply rooted member of the Clostridium-Bacillus-Streptococcus branch of the Gram-positive eubacteria, neither related to the actinomycetes nor closely related to true endospore-forming bacteria.

Full text

Page 1

International Journal of Systematic and Evolutionary Microbiology (2000), 50, 605–613
Printed in Great Britain
Phylogenetic position of the North American
isolate of Pasteuria that parasitizes the
soybean cyst nematode, Heterodera glycines,
as inferred from 16S rDNA sequence analysis
N. Atibalentja,1G. R. Noel1,2and L. L. Domier1,2
Author for correspondence: G. R. Noel. e-mail: g-noel1?uiuc.edu
1Department of Crop
Sciences, University of
Illinois at
Urbana–Champaign,
Urbana, IL 61801, USA
2Crop Protection Research
Unit, USDA ARS, Urbana,
IL 61801, USA
A 1341 bp sequence of the 16S rDNA of an undescribed species of Pasteuria
that parasitizes the soybean cyst nematode, Heterodera glycines, was
determined and then compared with a homologous sequence of Pasteuria
ramosa, a parasite of cladoceran water fleas of the family Daphnidae. The two
Pasteuria sequences, which diverged from each other by a dissimilarity index
of 7%, also were compared with the 16S rDNA sequences of 30 other bacterial
species to determine the phylogenetic position of the genus Pasteuria among
the Gram-positive eubacteria. Phylogenetic analyses using maximum-
likelihood, maximum-parsimony and neighbour-joining methods showed that
the Heterodera glycines-infecting Pasteuria and its sister species, P. ramosa,
form a distinct line of descent within the Alicyclobacillus group of the
Bacillaceae. These results are consistent with the view that the genus Pasteuria
is a deeply rooted member of the Clostridium–Bacillus–Streptococcus branch of
the Gram-positive eubacteria, neither related to the actinomycetes nor closely
related to true endospore-forming bacteria.
Keywords: Heterodera glycines, Pasteuria sp., phylogeny, 16S rDNA sequence,
soybean cyst nematode
INTRODUCTION
Pasteuria spp. are Gram-positive, mycelial and
endospore-forming bacteria traditionally classified in
theActinomycetales(Sayre&Starr,1989).Fourspecies
of Pasteuria have been described, the first of which,
Pasteuria ramosa (the type species), parasitizes clado-
ceran water fleas of the family Daphnidae (Ebert et al.,
1996;Metchnikoff,1888;Sayreetal.,1979,1983).The
other three species of Pasteuria are parasites of plant-
parasitic nematodes, against which they have shown
great potential as biological control agents (Atiba-
lentjaetal.,1998;Brownetal.,1985;Chenetal.,1996,
1997; Duponnois & Ba, 1998; Giblin-Davis, 1990;
Gowen et al., 1998; Nishizawa, 1987; Weibelzahl-
Fulton et al., 1996). The three species of nematode-
infecting Pasteuria are as follows: Pasteuria penetrans,
.................................................................................................................................................
Abbreviations: ML, maximum likelihood; MP, maximum parsimony; NJ,
neighbour-joining; RDP, Ribosomal Database Project.
The GenBank accession number for the 16S rDNA sequence of the North
American isolate of Pasteuria reported in this paper is AF134868.
which occurs on root-knot nematodes, i.e. Meloido-
gyne spp. (Sayre & Starr, 1985; Starr & Sayre, 1988);
Pasteuriathornei,whichaffectsroot-lesionnematodes,
i.e. Pratylenchus spp. (Starr & Sayre, 1988); and
Pasteuria nishizawae, which is parasitic on cyst nem-
atodes of the genera Heterodera and Globodera (Sayre
et al., 1991a, b). Members of the genus Pasteuria are
widespread and have been reported from 323 nema-
tode species belonging to 116 genera including plant-
parasitic,entomopathogenic,predatoryandfree-living
nematodes (Chen & Dickson, 1998; Ciancio et al.,
1994; Sayre & Starr, 1988; Sturhan, 1988).
Attempts to culture Pasteuria spp. in vitro have not
been successful (Bishop & Ellar, 1991; Williams et al.,
1989).Asaresult,thetaxonomyofthegenusPasteuria
has been marked by a series of errors and confusion
(Sayre & Starr, 1989), as it relies mainly on mor-
phological, developmental and pathological charac-
teristicsincludingthesizeandshapeofthesporangium
andendospore,ultrastructure,lifecycleandhostrange
(Davies et al., 1990; Giblin-Davis et al., 1990; Metch-
nikoff, 1888; Noel & Stanger, 1994; Sayre & Starr,
01240 ? 2000 IUMS
605

Page 2

N. Atibalentja and others
1989; Sayre et al., 1991a, b; Starr & Sayre, 1988;
Sturhan et al., 1994). For the nematode-infecting
Pasteuria, the taxonomic value of endospore morpho-
metrics and host range was questioned when these
characteristics were reported to reflect an evolutionary
adaptation that occurred during host–nematode
speciation (Ciancio, 1995). In addition, many of the
Pasteuria isolates reported display a high degree of
morphological similarity and others exhibit cross-
generic parasitism of nematodes (Chen & Dickson,
1998). Clearly, morphological, developmental and
pathological characteristics are not sufficient to ac-
commodate the increasing number of Pasteuria iso-
lates collected from soil and plant nematodes.
During the last few years, the analysis of 16S rDNA
and rRNA sequences has been used for the identi-
fication and classification of bacteria, including those
that cannot be cultured (Relman et al., 1990, 1992;
Woese, 1987). Recently, Ebert et al. (1996) questioned
the position of the genus Pasteuria in the Actino-
mycetales based on the analysis of the 16S rDNA
sequence of P. ramosa. Like other investigators (Sayre
& Starr, 1985; Sayre & Wergin, 1977), those authors
observed some morphological similarities between P.
ramosa and Thermoactinomyces spp. Unfortunately,
they could not resolve the phylogenetic relationships
between these organisms and Pasteuria spp., since
there was no Thermoactinomyces sequence for com-
parison. Apart from that of P. ramosa, no 16S rDNA
sequence has been reported for Pasteuria spp., es-
pecially those infecting nematodes. Our objectives,
therefore,were as follows: to determine the 16S rDNA
sequence of a nematode-infecting Pasteuria, namely
the North American isolate of Pasteuria that para-
sitizes the soybean cyst nematode, Heterodera glycines
Ichinohe (Noel & Stanger, 1994); to compare this
sequence with that of P. ramosa (Ebert et al., 1996);
and to resolve the phylogenetic position of the genus
Pasteuria.
METHODS
Sourceof endospores. Pasteuria-infectedH. glycines females
and cysts were selected, on the basis of their opaque
appearance, from the nematodes extracted from the rhizo-
sphere of 3-month-old soybean plants grown in naturally
infested soil in a greenhouse. The selected nematodes were
transferred individually into 100 µl tap water in 1?6 ml
microfuge tubes in which they were crushed with a tissue
grinder; 10 µl of the resulting suspension was examined
microscopically for the presence of Pasteuria endospores.
Positive fractions were pooled in a 1?6 ml microfuge tube to
form the stock suspension, which was then forced through a
5-µm-pore polycarbonate membrane filter (Poretics) to
remove nematode debris. The endospore concentration of
the suspension was determined with a Levy–Hausser count-
ing chamber (Arthur H. Thomas) and the material was
stored at 4 ?C until used.
DNA extraction from Pasteuria endospores. DNA extraction
was based on a modification of a procedure described by
Ebert et al. (1996). Specifically, 500 µl of a stock suspension
containing 2?10? endospores ml−? was transferred into a
clean 1?6 ml microfuge tube, which was then incubated for
10 min in a boiling water-bath. After cooling to room
temperature (24 ?C), the suspension was centrifuged for
15 min in an Eppendorf microcentrifuge at maximum speed
(approx. 16000 g) and the pellet was resuspended with
200 µl 25 mg lysozyme ml−? solution in 10 mM Tris?HCl
buffer, pH 8?0. The mixture was incubated at 37 ?C for
30 min, at which point a 10% solution of SDS was added to
a final concentration of 2% (v?v); this was followed by an
additional 30 min incubation at 37 ?C. The endospores,
which are resistant to heat, lysozyme and detergent treat-
ments, were pelleted as before and washed twice by re-
suspending the pellet in 500 µl 10 mM Tris?HCl buffer,
pH 7?0; this was followed by 15 min centrifugation at high
speed. The pellet from the last wash was resuspended in
500 µl buffered solution (40 mM Tris?HCl, pH 8?0, 10 mM
NaCl, 6 mM MgCl?and 10 mM CaCl?) containing 10 µg
(50 U ml−?) RQ1 RNase-free DNase ml−? (Promega) to
destroy any DNA released from contaminating micro-
organisms. The DNase reaction was conducted at 37 ?C for
30 min, at which point the endospores were pelleted, washed
twice and resuspended with 500 µl of a solution containing
50 U RNase One ml−? (Promega) in RNase One buffer
(10 mM Tris?HCl, pH 7?5, 5 mM EDTA and 200 mM
sodium acetate). After 30 min incubation at 37 ?C, pro-
teinase K (50 µg ml−?) and SDS (0?5%, v?v) were added and
the mixture was incubated further at 55 ?C for 2 h. The
endospores were pelleted again, washed twice and re-
suspended in 500 µl TE buffer (10 mM Tris?HCl, 1 mM
EDTA, pH 8?0) to produce the pre-lysis suspension.
To ascertain the absence of vegetative cells in the pre-lysis
suspension, 15 µl was examined microscopically and 100 µl
ofa1:10dilutioninsteriledistilledwaterwasspreadoneach
of two Luria agar plates, which were incubated at 30 ?C for
a week. As positive controls, two Luria agar plates were
spread with equivalent amounts of the stock suspension that
had not been subjected to either heat or enzyme treatments
and incubated at 30 ?C. Aliquots (10 µl) of the pre-lysis
suspension and the suspensions derived from representative
coloniesofthebacteriathatgrewoncontrolplateswereused
inseparatePCRreactions(seebelow),theproductsofwhich
were sequenced to determine the identity of the DNAs in the
suspensions. The remaining 455 µl of the pre-lysis sus-
pension was transferred into a fresh screw-capped 1?6 ml
microfuge tube containing 860 mg acid-washed glass beads
(0?150–0?212 mm in diameter) and 500 µl Tris buffer
(pH 8?0)-equilibrated phenol?chloroform?isoamyl alcohol
(25:24:1, by vol.). Lysis of endospores was achieved with
four 30 s pulses of bead-beating in a Mini-BeadBeater
(BiosPec Products) operated at high speed in a cold room
(4 ?C). The mixture was incubated on ice for 5 min between
pulses of bead-beating. The lysate was centrifuged at
maximum speed for 10 min and the aqueous phase was
transferred into a fresh 1?6 ml microfuge tube. The handling
of the supernatant, thereafter, was according to standard
procedures (Sambrook et al., 1989), which included chloro-
form:isoamyl alcohol (24:1, v?v) extraction followed by
DNA precipitation with 3 M sodium acetate (pH 5?2) and 2
vols ice-cold ethanol (100%), then washing with 70%
ethanol. The final pellet was resuspended with 50 µl TE
buffertomakethepost-lysissuspension,whichwasstoredat
?20 ?C until used.
DNA amplification and sequencing. The PCR primers
(PrDNA-1 and PrDNA-2) were designed (Table 1) to
amplify an approximately 1400 bp region of the 16S rDNA,
606
International Journal of Systematic and Evolutionary Microbiology 50

Page 3

Heterodera glycines–Pasteuria sp. phylogeny
Table 1. Primers used for PCR amplification and sequencing of the 16S rDNA of the
Heterodera glycines-infecting Pasteuria
Name*Direction5??3? sequence Position of 5? nucleotide†
1
2
3
4
5
6
7
8
Forward
Reverse
Forward
Reverse
Forward
Reverse
Forward
Reverse
GCGGCGTGCCTAATACA
ACGGGCGGTGTGTACAAG
TACGGGAGGCAGCAGTA
TACTGCTGCCTCCCGTA
CCCTGGTAGTCCACGCG
CGCGTGGACTACCAGGG
TCGAGAGAGTGCTGTGCCC
GGGCACAGCACTCTCTCGA
39
1388‡
342
342‡
794
794‡
1011
1011‡
*Primers were designated by the generic name ‘PrDNA-X’, where X is a number from 1 to 8.
†Numbers refer to equivalent positions in the E. coli sequence (Brosius et al., 1978).
‡The position of the 5? nucleotide in the complementary sequence.
based on a comparison of published sequences from related
Gram-positive bacteria including Bacillus subtilis, Bacillus
caldolyticus, Bacillus validus, Thermoactinomyces vulgaris
and Thermoactinomyces candidus (accession numbers of
publishedsequences usedinthis studyarelisted attheend of
this section). Additional primers were designed (Table 1) on
the basis of preliminary sequencing results. The 50 µl PCR
reactions [10 µl sample, 1? buffer (20 mM Tris?HCl,
pH 8?4, 50 mM KCl), 1?5 mM MgCl?, 0?2 µM each primer,
0?2 mM each dNTP and 2?5 U Taq DNA Polymerase (Life
Technologies)] were carried out in a GeneAmp PCR System
9700 (Perkin-Elmer, Applied Biosystems) using the fol-
lowing regimen: 94 ?C for 10 min; 30 cycles each of 1 min
denaturation at 94 ?C, 1 min primer annealing at 52 ?C and
2 min extension at 72 ?C; final extension at 72 ?C for 5 min;
and incubation at 4 ?C. The sizes of the amplicons were
determined by loading 10 µl on to a 1% agarose gel
alongside a 1 kbp DNA ladder (Life Technologies). Un-
incorporated primers, dNTPs and salts were removed from
the amplicons using the QIAquick PCR purification kit
(QIAgen). The concentration of the eluted DNA (50 µl
DNA in 10 mM Tris?HCl buffer, pH 8?5) was estimated by
comparing band intensities on agarose gel and by UV-
spectrophotometry.
The purified amplicons were sequenced using the ABI
PRISM dye terminator cycle sequencing ready reaction kit
with AmpliTaq DNA Polymerase, FS (Perkin-Elmer, Ap-
plied Biosystems), according to a protocol suggested by the
manufacturers (except that the annealing temperatures were
adjusted for each primer). The 20 µl sequencing reactions
(8 µl terminator ready reaction mix, 75 ng DNA template
and 3?2 µM primer) were conducted in the same GeneAmp
PCR System 9700 used for PCR amplification. Sequencing
products were purified prior to electrophoresis using Centri-
Sep spin columns (Princeton Separations) according to the
manufacturer’s instructions. The sequencing products were
dried in a vacuum microcentrifuge and forwarded to the
Genetic Engineering Laboratory of theUniversity of Illinois
at Urbana–Champaign, where the pellets were resuspended
and loaded into an automated ABI PRISM 377 sequencer
according to the manufacturers’ instructions.
Phylogenetic analysis. The various fragments of DNA
sequence obtained from both strands were assembled with
the multiple sequence editor  (Gene Codes). The
consensus sequence, hereafter referred to as Hg Pasteuria
(Heterodera glycines-infecting Pasteuria), was submitted
for sequence-similarity searches using the  program
(Altschuletal.,1990)andthe?featureofthe
Ribosomal Database Project (RDP; Maidak et al., 1999).
The ? option provided an approximate place-
ment of Hg Pasteuria in the RDP maximum-likelihood tree.
In addition to the sequences suggested by the above search
methods, a set of sequences from a previous alignment
involving P. ramosa (Ebert et al., 1996) and sequences from
Anacystis nidulans and Escherichia coli (the latter two
sequences were used as the outgroup) were retrieved from
public databases and aligned with Hg Pasteuria, using the
graphical multiple alignment software  1.64b
(Thompson et al., 1997). The multiple alignment was edited
manually to remove regions with long stretches of leading or
trailing gaps. A measure of signal content was obtained
using Relative Apparent Synapomorphy Analysis (RASA)
withthe program  2.2(Lyons-Weiler et al., 1996) before
the data were subjected to phylogenetic analyses using
maximum-likelihood(ML),maximum-parsimony(MP)and
distance-based methods. The robustness of trees was
assessed through bootstrap analysis (Felsenstein, 1985). The
fastDNAml?loop script of fastDNAml 1.1.1a (Felsenstein,
1981; Olsen et al., 1994) was used for ML analysis with the
following option settings: empirical base frequencies, a
transition:transversion ratio of 2?0, global rearrangement,
? 3 and  10. The fastDNAml?boot script of
the same software was also used to generate 100 bootstrap
trees, which were fed into the  program in the
 package (Felsenstein, 1993) to compute bootstrap
proportions. MP analyses were based on informative sites
only and were conducted with  3.1 (Swofford, 1993),
using a heuristic search method with 100 random sequence-
addition replicates, steepest descent,  and 
branch-swapping option settings. The same search method
and optionswereusedforbootstrapanalysisoftheMP trees
based on 100 replicates with 10 random sequence addition
replicates each. The ? program (Galtier et al.,
1996) was used to compute pair-wise evolutionary distances
either from all sites or from parsimony sites only and to
construct neighbour-joining (NJ) trees (Saitou & Nei, 1987)
basedonLogDet-correcteddistances(Lake,1994;Lockhart
et al., 1994). The NJ trees were evaluated through bootstrap
resampling, with 1000 replicates each.
International Journal of Systematic and Evolutionary Microbiology 50
607

Page 4

N. Atibalentja and others
Accession numbers.Theaccessionnumbersofpublished 16S
rDNA sequences used in this study are as follows: Actino-
mycesbovis,X81061;
Alicyclobacillus
X60742; Alicyclobacillus acidoterrestris, X60743; Alicyclo-
bacillus cycloheptanicus, X51928; strain ALV, M79375,
M79376, M80290; Anacystis nidulans, X03538, K01982,
X01296; Aureobacterium liquefaciens, X77444; Bacillus
caldolyticus, Z26924; Bacillus methanoliticus, X64465,
S42879; Bacillus schlegelii, Z26934; Bacillus subtilis,
D64126; Bacillus thermocatenulatus, Z26926; Bacillus
thermocloacae, Z26939; Bacillus thermoruber, Z26921;
Bacillus thuringiensis (strain IAM12077), D16281; Bacillus
tusciae, Z26933; Bacillus validus, M77489; Clostridium
difficile,X73450;
Clostridium
X71858; Escherichia coli,
M24833–M24837, M24911, M24996; Flavobacterium hepa-
rinum, M11657, M61766, M81326; Heliobacterium chlorum,
M11212; strain HTA1417, AB002649; Lactobacillus brevis
(isolate L63), D37785; Megasphaera elsdenii, M26493;
Oxalophagus oxalicus, Y14581; Paenibacillus apiarius,
U49247; Paenibacillus validus (strain DSM3037), D78320;
Pasteuria ramosa (Daphnia endosymbiontic bacterium),
U34688; Planococcus citreus, X62172, S49897; strain
Rt8B.4, L35151; Saccharococcus thermophilus, X70430;
Thermoactinomyces candidus, M77490; Thermoactinomyces
vulgaris, M77491; type 0803 unidentified filamentous
bacterium, X86071.
acidocaldarius,
polysaccharolyticum,
K02555,J01859,M24828,
RESULTS
PCR amplification and sequencing of Pasteuria DNA
Microscopic examinations revealed that the stock
suspension of Pasteuria endospores contained many
contaminating bacteria. In contrast, no contaminants
wereobservedin thepre-lysis suspension. Likewise, no
bacterialgrowthwasobservedontheLuriaagarplates
that were spread with the pre-lysis suspension and
incubated for a week at 30 ?C. Conversely, the control
plates spread with the stock suspension contained
several bacterial colonies after 48 h incubation under
similar conditions. PCR reactions using the sus-
pensions of representative bacterial colonies that grew
on control plates as the template produced a single
band,eachofwhichwasoftheexpectedsize(1400 bp).
DNA sequences of those products matched the 16S
rDNA of F. heparinum, A. liquefaciens or type 0803
of an unidentified filamentous bacterium following
sequence-similarity searches with the  program
(Altschul et al., 1990). PCR reactions with either the
pre- or the post-lysis suspension as the template also
yielded a single product, each of which was of the
expected size (1400 bp). DNA sequences of the pro-
ducts originating from the pre- or the post-lysis
suspensions were identical. At least two fragments of
DNA sequence were generated with each of the eight
primers used to amplify and sequence DNA from
either the pre- or the post-lysis suspension, for a total
of 18 fragments derived from both DNA strands.
When each of the 18 fragments was submitted for
sequence-similarity searches using the  program
(Altschul et al., 1990), the 16S rDNA sequence of P.
ramosa was always retrieved as the highest scoring
segment with the query sequence. The 18 fragments
were assembled with the multiple sequence editor
 (Gene Codes) to produce a 1341 bp
consensus sequence referred to as the Hg Pasteuria
sequence. This sequence not only included the
Pasteuria-specific oligonucleotide, 5?-CATTTCTTC-
TTCCCGATG-3? (Ebert et al., 1996), but also con-
tained two oligonucleotides (5?-TACACCCCAGA-
GGATGCA-3? and 5?-GGGCACAGCACTCTCT-
CGA-3?) that searches of the RDP database of 16S
rDNA probes revealed to be unique to the North
Americanisolate of Pasteuria [the above sequences are
from the minus strand and the positions of the 5?
residues in the complementary strand are 440, 69 and
1011, respectively, according to the E. coli numbering
system (Brosius et al., 1978)].
Phylogenetic relationships between Hg Pasteuria and
other bacterial species
The edited DNA sequence alignment comprised 1336
sites, including 531 informative sites under parsimony,
which were unambiguously aligned for 32 taxa. On
the basis of observed (uncorrected) distances, Hg
Pasteuria shared 93?2% sequence identity with P.
ramosa, as opposed to the 87?3, 84?7, 83?3, 80?4 and
75?1% sequence identity found between Hg Pasteuria
and T. vulgaris, A. cycloheptanicus, B. subtilis, A. bovis
and E. coli, respectively. For comparison, sequence
similarities between A. acidocaldarius and A. acido-
terrestris and between T. vulgaris and T. candidus were
99?4 and 98?5%, respectively. The analysis of signal
content,afterremovalofinvariantandautapomorphic
sites, indicated a significant (tRASA?12?992, P?
0?001) cladistic structure in the data. Phylogenetic
inferenceusingMLplacedHgPasteuriaandP.ramosa
at the base of a clade that contained Alicyclobacillus
spp., B. tusciae and a strain (ALV) of a facultatively
thermophilic iron-oxidizer (Fig. 1). Six equally par-
simonious trees, 2972 steps long with a rescaled
consistency index of 0?2 each, were recovered by MP
analysis. Four of the six trees (and, hence, the
consensus tree) concurred with the ML tree in placing
Hg Pasteuria and P. ramosa in the same branch as
Alicyclobacillus spp., B. tusciae and ALV, there being
similar bootstrap values for these relationships (result
not shown). The other two most parsimonious trees
recovered the Pasteuria (Hg Pasteuria and P. ramosa)
as a separate cluster between the branch leading to
Alicyclobacillus spp. and B. tusciae and the branch
leading to Thermoactinomyces spp. and HTA1417, a
strain of an unidentified low-G?C-DNA Gram-
positive bacterium (result not shown). Similar top-
ology was obtained when the NJ tree was constructed
from LogDet-transformed distances based on all sites
(result not shown). However, when LogDet-trans-
formeddistances were calculated from parsimony sites
only, the NJ tree was consistent with both the ML and
MPconsensustreesregardingthepositionofPasteuria
spp. (Fig. 2). Unlike the ML tree, the NJ and MP
consensus trees placed B. schlegelii at the base of the
branch leading to the Alicyclobacillus group, although
608
International Journal of Systematic and Evolutionary Microbiology 50

Page 5

Heterodera glycines–Pasteuria sp. phylogeny
.................................................................................................................................................................................................................................................................................................................
Fig. 1. Phylogenetic reconstruction based on maximum-likelihood analysis of 16S rDNA sequences from 32 bacterial
species including Hg Pasteuria (Heterodera glycines-infecting Pasteuria). A. nidulans (cyanobacteria) and E. coli
(Proteobacteria) were used as outgroup taxa. Bootstrap proportions (100 replicates) are given for relevant branches only;
the scale represents the mean number of nucleotide substitutions per site.
the bootstrap support for this position was poor (18
and 11%, respectively, for the NJ and the MP
consensus trees). Strain ALV was not recovered with
Pasteuria spp. in any of the NJ trees.
DISCUSSION
Provision of clean Pasteuria DNA for PCR
amplification
The absence of vegetative cells in the pre-lysis sus-
pension of endospores, as evidenced by both micro-
scopic and cultural observations, shows that the heat
and lysozyme treatments were effective in destroying
contaminating bacteria present in the stock suspen-
sion. The identity of the DNA amplified from the pre-
and post-lysis suspensions of endospores indicates
that, following destruction (by DNase and RNase) of
the nucleic acids released (mostly from contaminating
bacteria) under the heat and lysozyme treatments,
proteinase K and SDS were able to destabilize
Pasteuria endospores sufficiently to enable release of
the DNA amplified during the PCR reaction. The
assumption that the amplified DNA originated from
Pasteuria rather than from contaminating bacteria is
supported by the fact that the resulting sequence was
not only very similar to that of P. ramosa, the only
Pasteuria sequence currently available in the public
domain, but also contained the Pasteuria-specific
oligonucleotide (Ebert et al., 1996). The identity
between the amplicons from the pre- and post-lysis
suspensions also suggests that the task of extracting
and sequencing DNA from Pasteuria endospores can
now be simplified by eliminating unnecessary steps
such as glass-bead beating and subsequent DNA
purification.
Hg Pasteuria and P. ramosa are closely related but
distinct species
According to current criteria for delineating bacterial
species (Stackebrandt & Goebel, 1994; Wayne et al.,
1987) Hg Pasteuria and P. ramosa are closely related
butdifferentspecies.BothPasteuriabelongtoacluster
of thermophilic endospore-forming bacilli that differ
enough from the other members of the genus Bacillus
towarranttheirassignmentintoseparategenerawithin
the Bacillaceae (Wisotzkey et al., 1992). The genus
International Journal of Systematic and Evolutionary Microbiology 50
609