Vol. 37, No. 2
JOURNALOF SYSTEMATIC BACTERIOLOGY,Apr. 1987, p. 116-122
0020-7713/87/020116-07$02.00/0
Calpyright 0 1987, International Union of Microbiological Societies
INTERNATIONAL
Phylogenetic Analysis of Certain Sulfide-Oxidizing and Related
Morphologically Conspicuous Bacteria by 5s Ribosomal Ribonucleic
Acid Sequences
DAVID A. STAHL,l DAVID J. LANE,* GARY J. OLSEN,2 DEBRA J. HELLER,? THOMAS M.
AND NORMAN R. PACE2*
Department of Veterinary Pathobiology, University of Illinois, Urbana, Illinois 61801 Department of Biology and
Institute for Molecular and Cellular Biology, Indiana University, Bloomington, Indiana 474052; and Program of
Environmental Biology, The Ohio State University, Columbus, Ohio 432103
’;
Comparisons of 5s ribosomal ribonucleic acid (rRNA) nucleotide sequences were used for the phylogenetic
characterization of a number of ‘‘morphologically conspicuous” sulfide-oxidizing eubacteria and certain of
their non-sulfide oxidizing counterparts. Nucleotide sequences were determined for 5s rRNAs isolated from
Thiovulum sp., Beggiatoa alba B18LD, Thwthrix nivea JP2,strain 30 (a Thwthrix-like organism), Vitreoscilla
beggiatoides B23SS, Vitreoscillafiliformis L1401-7, V .filiformis ATCC 15551, Vitreoscilla stercoraria VT1, and
Leptothrix discophora (Stokes). All 5s rRNAs characterized in this study, except that of Thiovulum sp., belong
to either the gamma or beta subdivisions of the “purple bacteria,” which form 1 of the 10 major eubacterial
divisions so far defined by 16s rRNA oligonucleotide catalog analysis. Although its 5s rRNA is clearly
eubacterial in character, Thiovulum sp. appears to have no specific relatives among the eubacteria presently
represented in the 5s rRNA data collection.
The colorless sulfur bacteria are a diverse collection of
microorganisms that are capable of oxidizing reduced sulfur
compounds (16). Thiobacillus is the best-characterized genus of colorless sulfur bacteria, but also included are members of the genera Beggiatoa, Thiothrix, Thiovulum,
Achromatium, Macromonas, Thiobacteria, Thiospira, and
Thioploca (19). These latter have been referred to collectively as the “morphologically conspicuous sulfor-oxidizing
b<acteria”(19). Of these, only a few Thiothrix and Beggiatoa
strains are in pure cultures. Consequently, little is known
albout the physiologies of organisms belonging to these
genera. The morphologically conspicuous sulfur bacteria
aire, in fact, an ad hoc assemblage of novel morphologies
which are united by the restriction of their occurrence to
transition zones where sulfide and oxygen coexist, and by
their intracellular deposition of sulfur. The genera Beggiatoa, Thiothrix, Thioploca, and Achromatium have been
assigned to various families of the gliding bacteria (order 11,
Cytophagales);Macromonas, Thiobacteria, Thiovolum, and
Thiospira are treated as gram-negative, sulfur-metabolizing
organisms of uncertain affiliation (22). Because of striking
morphological similarities, members of the genus Beggiatoa
have been considered by some workers to be apochlorotic
relatives of the oscillatorian cyanobacteria (30).
The use of ribosomal ribonucleic acid (rRNA) sequence
comparisons to infer natural relationships is now a practical
criterion for characterizing microorganisms (32). So far, 10
major divisions (termed “phyla” [40]) within the eubacteria
have been defined, using partial sequences (oligonucleotide
catalogs) of 16s rRNAs (9, 39, 40). (No officially sanctioned
consensus exists regarding terminology for phylogeneticallydefined “groupings.” In this paper we follow the usage of
FVoese et a1 [40] and define, in order of descending rank:
phylum [division], subdivision [group, in some earlier cita-
* Corresponding author.
t Present address: 11325 E. Vassar Dr.,
Denver, CO 80206.
$ Present address: Center for Great Lakes Study, Milwaukee, WI
53104.
tions], subgroup, cluster, genus, species, and strain.) However, some known eubacteria do not fall into any of these
phylogenetic divisions. The largest of the defined divisions,
which contains the majority of familiar, gram-negative bacteria, has been called by Woese the “purple bacteria and
relatives” to reflect its inferred photosynthetic ancestry
(40-43). As first revealed by comparative 16s rRNA sequencing (9, 32), this assemblage encompasses a diverse
collection of microorganisms, differing in many ranking
taxonomic features (photosynthetic, chemoautotrophic, and
heterotrophic metabolisms; gliding and flagellated motilities;
filaments and single cells; the presence and absence of
sheaths), but sharing a gram-negative cell wall and a common ancestral root within the eubacteria. Analyses of 16s
rRNA oligonucleotide catalog data have defined three distinct subdivisions of the purple bacteria, termed the alpha,
beta, and gamma subdivisions (originally called groups I, 11,
and 111, respectively) (39-43).
As a complement to 16s rRNA analysis, comparisons of
complete 5s rRNA sequences have proven useful for the
rapid assessment of phylogenetic relationships within the
various eubacterial phyla. For instance, 5s rRNA sequence
comparisons have been used to assign a number of thiobacilli
(18) and symbiotic sulfide-oxidizers (33) to specific subdivisions of the purple bacteria. In the present study, members
of the genera Beggiatoa, Thiothrix, and Thiovulum are
characterized by 5s rRNA nucleotide sequencing. Representatives of the related genera Leptothrix and Vitreoscilla are
also included. Vitreoscilla stercoraria and a Beggiatoa sp.
previously had been characterized by 16s rRNA oligonucleotide catalog analysis (42); these results are related to the
present 5s rRNA findings.
MATERIALS AND METHODS
Cultivation and isolation of organisms. The organisms used
in this study are listed in Table 1. Their cultivation was as
described by Strohl et al. (36).
Thiovulum enrichment. The enrichment culture of the
Thiovulum sp. was developed as previously described (38).
116
Downloaded from www.microbiologyresearch.org
by
IP: 88.99.165.207
On: Sun, 18 Jun 2017 04:58:46
VOL. 37, 1987
PHYLOGENY OF SULFIDE-OXIDIZING BACTERIA
117
TABLE 1. Summary of strains
Strain
Organism
Sequence
available
Source
16s
5s
~~~
Subdivision
f:?:
Sulfide
oxidation
Beta
Beta
Beta
Beta
Beta
-
G+C
(mol%)
Filament
diam
(Pm)
~
Vitreoscilla filiforrnis
Vitreoscilla filiforrnis
Vitreoscilla stercoraria
Vitreoscilla stercoraria
Leptothrix discophora
ATCC 15551
L 1401-7
VT1
Vitreoscilla beggiatoides
Beggiatoa alba
Beggiatoa sp.
Thiothrix nivea
Thiothrix-like
B23SS
B18LD
Stokes
JP2
30
Thiovulum sp.
W. R. Strohl
W. R. Strohl
W. R. Strohl
H. Reichenback
W. R. Strohl
-
+a
+a
+a
+=
-
-
+a
W. R. Strohl
W. R. Strohl
J. Gibson
W. R. Strohl'
W. R. Strohl'
-
+a
-
Enrichment
+a
+k
+a
-
-
Gamma
Gamma
Gamma
Gamma
Gamma
-
+a
+g
-
63
59b
51b
N A ~
71'
42b
41-51b
NA
52'
NA
-
+
+
+
+
+
?
1.C1.5'
1J-1.5'
1.Ob
NA
O.M.8f
2.5-3 .Ob
1.5-2.0b
1.o-2 .Oh
1.0-1.5'
1.0-2.v
NA
This study.
See reference 36.
See reference 43.
NA, Not available.
See reference 4.
f See reference 23.
g See reference 42.
Inferred from the original description of the strain as B. leptomidiformis (see reference 42).
Originally isolated by J. M. Larkin (see reference 20).
j W. R. Strohl; personal communication.
D. A. Stahl, unpublished data.
Thiovulum veils were harvested repeatedly over several
days. After each harvest, the cell suspension was largely
freed from comtaminating microorganisms by repeated concentration on an 8-pm-pore Nucleopore filter. The essential
purity of the suspension was confirmed by epifluorescent
microscopy and by recovery of only one species of 5s
rRNA. Most cells observed in the enrichment were approx-
imately 9 to 10 pm in diameter, possibly corresponding to
Thiovulum minus (19, 38).
Isolation and sequencing of 5s rRNAs. Total RNA was
isolated from cells by phenol-sodium dodecyl sulfate (SDS)
extraction and then fractionated on 10% polyacrylamide slab
gels. The various RNA species were visualized by ultraviolet
shadowing (ll),and the portion of the gel containing the 5s
I
I1
I11
Pseudomonas aeruginosa CCE0401
UGCUUGACGAUCAUAGAGCGUUGGAACCACCUGAUCCCUUCCCGAAC
Thiothrix nivea JP2
UUUGCCUGGUGUCCAUAGAGCACUGGAACCACCUGAUCCCAUCCCGAAC
Thiothrix-like s t r a i n 30
UGCCUGGUGUCCAUAGAGCACUGGAACCACCUGAUCCCAUCCCGAAC
Beggiatoa alba B18LD
UUCUUGGCGACCAUAGCAAAUAGGAACCACCCGACCCCAUCCCGAAC
V i t r e o s c i l l a beggiatoides B23SS
CGCCUGACGACCACAGCGACUGUGAACCACCCGACCCCAUCUCGAAC
UGUUUGACGACCAUAGCGAGUUGGUCCCACGCCUUCCCAUCCCGAAC
V i t r e o s c i l l a stercoraria VT i
V i t r e o s c i l l a f i l i f o r m i s ATCCl5551 G L1401-7
GCCUGAUGACCAUAGCAAGGUGGUCCCACUCCUUCCCAUCCCGAAC
Leptothrix discophora (Stokes)
aUGCCUGACGACCAUAGCGAGGUGGUCCCACUCCUUCCCAUCCCGAAC
UUUGGUUGGUGAUUACAGAGAAAAGGUCACACUCAGCUCCAUUUCGAAC
Thiovulum sp.
I
I
aeruginosa
T. n i vea
strain 30
B. alba
V. beggiatoides
V. stercoraria
V. f i 1i formi s
L. discophora
Thiovulum sp.
0.
111'-
I
I
20
10
40
30
---
11
V
IV
IV '
V'
I'
UCAGAAGUGAAACGACGCAUCGCCGAUGGUAGUGUGGGGUCUCCCCAUGUGAGAGUA~UCAUCGUCAAGCUC
UCAGAAGUGAAACGGUGCAUCGCCGAUGGUAGUGUGGGGCCUCCCCAUGUGAGAGUAGGUCAACGCCAGGCGC
UCAGAAGUGAAACGGUGCAUCGCCGAUGGUAGUGUGGGGCCUCCCCAUGUGAGAGUAGGUCAACGCCAGGCGC
UCGGUAGUGAAACUGUUCUGCGCCGAUGAUAGUGUGGAUACUCUCCAUGUGAAAGUAGGUUAUCGCCAAGAGc
UCGGUAGUGAAACCAGUCAGCGCCGAUGAUAGUGUGGC-AUAUGCCAUGUGAAAGUAGGUCAUCGUCAGGCU
AGGACCGUGAAACGACUUAGCGCCGAUGAUAGUGUGGA-UUA-CCCAUGUGAAAGUAGGUCAUCGUCAAACGC
AGGACAGUGAAACGCCUUAGCGCCGAUGAUAGUGCGGU-UCU-CCCGUGUGAAAGUAGGACAUCGUCAGGCU
AGGACAGUGAAACGCCUCAGCGCCGAUGAUAGUGCGCA-UU--CGCGUGUGAAAGUAGGUCAUCGUCAGGC~
CUGAAA~UAAGCUUUUCUUCG~GAUAAU~CUGCCCCCUACGGGGGUGGGACGGUA~UCGUUGCCAACCAUu
I
I
I
I
I
1
I
5Q
60
70
80
90
100
110
I
120
FIG. 1. Alignment of 5s rRNA nucleotide sequences. The nucleotide sequences of 5s rRNAs characterized in this study are shown aligned
on the 5s rRNA sequence of Pseudornonas aeruginosa (1).Regions of base pairing as defined by the 5s rRNA consensus secondary structure
are indicated by horizontal bars and labeled with roman numerals. Positions of terminal length variation are indicated by lowercase letters.
Downloaded from www.microbiologyresearch.org by
IP: 88.99.165.207
On: Sun, 18 Jun 2017 04:58:46
118
STAHL ET AL.
INT. J . SYST.BACTERIOL.
0.1
FIG. 2. 5s rRNA phylogeny of the beta and gamma subdivisions. The illustrated phylogenetic relationships were inferred using weighted
5s rRNA sequence homologies as described in the text. 5s rRNA sequences not given in Fig. 1 are available elsewhere (6, 18, 33). The tree
shown has been abstracted from a larger tree which included additional gamma division representatives. Based on relative distances from the
various sequences, the root of this tree lies within the dotted segments; the beta and gamma subdivisions are below and above the root,
respectively. The scale bar represents an evolutionary distance of 0.1 fixed mutation per sequence position.
rRNA was excised. The 5s rRNA was eluted by soaking in
0.5 M NH40Ac-10 mM MgC12-1.0 mM EDTA-0.1% SDS,
recovered as an ethanol precipitate, and labeled at one or the
other of its termini with 32Pfor nucleotide sequence determinations. The 5’ termini were labeled following
dephosphorylation with bacterial alkaline phosphatase plus
polynucleotide kinase and [Y-~~PIATP;
the 3’ termini were
labeled with RNA ligase and [5’-32P]pCp.The end-labeled 5s
r RNAs were repurified on high-resolution polyacrylamide
gels (sequencing gels) to resolve molecules with end-length
variation. All 5s rRNAs were sequenced from both termini,
using chemical and enzymatic partial digestion procedures
(3, 28). Terminal nucleotides were determined by thin-layer
chromatography or high-voltage paper electrophoresis of
exhaustive nuclease P1 digestion (5‘ end labeled) and complete alkaline hydrolysis (3’ end labeled) products (33). The
derived sequences and alignments are shown in Fig. 1.
Phylogenetic analyses. The phylogeny of the bacteria
shown in Fig. 2 was deduced by a distance matrix method ( 5 ,
8). Briefly, the sequences are first aligned, as in Fig. 1, so
that evolutionarily homologous sequence positions are
brought into correspondence. For each pair of sequences,
the evolutionary distance (average number of nucleotide
substitutions per sequence position) was estimated as - 3/4
ln(1 - 4L/3), were L is the observed fractional sequence
difference (14). This treatment partially corrects the observed number of sequence differences for multiple and back
mutations (14). Regions of terminal length variation (positions 1to 3 and 121 to 123 [Fig. 11) and ambiguous sequence
alignment (positions 87 to 93 [Fig. 11) were omitted from
sequence comparisons (25). Phylogenetic calculations were
performed both with the “unweighted” homology data just
described, and also with data “weighted” as follows. Regions of conserved base pairing which, in essence, represent
only one (rather than two) independently varying positions
per base pair were given half the weight assigned to unpaired
positions. In general, we find that using this positional
weighting with 5s rRNA sequence data improves the internal consistency of the data without appreciably affecting the
pattern (topology) of inferred relationships.
For phylogenetic tree construction, these evolutionary
distance estimates (weighted or unweighted) are fit to an
initial (assumed) branching order, the optimal segment
lengths are determined, and then the goodness of fit of this
topology to the input evolutionary distance estimates is
calculated. We define the error of the tree as the sum of the
Downloaded from www.microbiologyresearch.org by
IP: 88.99.165.207
On: Sun, 18 Jun 2017 04:58:46
VOL. 37. 1987
PHYLOGENY OF SULFIDE-OXIDIZING BACTERIA
squares of the differences between the pairwise evolutionary
distance estimates and the corresponding tree reconstructions of those distances (i.e., the sums of the tree branches
which connect the pairs of organisms), with each difference
being weighted by its corresponding statistical uncertainty.
The uncertainty (i.e., the expected counting error due to
finite sequence length) associated with the evolutionary
distance estimates (as defined above) has been calculated by
Kimura and Ohta (15).
Given an initial tree topology (branching order) and a
measure of its goodness of fit to the input homology data, an
optimal topology is sought. There is as yet no known process
which is guaranteed to find the best branching order of a
large phylogenetic tree in a practical amount of computational time. Therefore, methods are employed which look for
better branching orders by systematically testing a tractable
number of alternatives to the current best-known branching
order. We use a computer program which examines all
rearrangements of a tree which can be achieved either by
moving any single group of organisms (subtree) to every
alternative location in the tree or by interchanging any pair
of groups (subtrees). This optimization program examines all
of the above rearrangements and then takes the best of the
examined alternatives as the starting point for another cycle
of searching for better branching orders (G. J. Olsen, Ph.D.
thesis, University of Colorado Health Sciences Center,
Denver, 1983). When a cycle of optimization fails to find a
better tree (i.e., a branching order with a lower error than the
starting tree), then the program is assumed to have found the
best tree.
During optimization of the tree shown in Fig. 2, approximately 4,000 alternative topologies were inspected during
each cycle. Some 30,000 topologies were inspected overall.
In addition to testing alternative topologies of the illustrated
tree, other trees of slightly different organism composition
were also tested to ensure that the final, optimized topology
was “robust” with respect to any potential organismspecific (i.e., sequence-specific artifacts. These operations
were performed on a Digital Equipment Corporation
VAX- 11/780 computer.
RESULTS
Overview. The 5s rRNA nucleotide sequences derived in
this study and the alignments used for phylogenetic analysis
are given in Fig. 1. The relationships inferred from comparisons of these sequences and others available (6) are presented as a phylogenetic tree shown in Fig. 2. The morphologically conspicuous sulfur-oxidizingbacteria characterized
in this study, with the exception of Thiovulum sp., fall into
two of the subdivisions of the purple bacteria. Thiothrix
nivea, Beggiatoa alba, and the non-sulfur organism
Vitreoscilla beggiatoides are associated with the gamma
subdivision, whose representatives in Fig. 2 include Escherichia coli, Photobacterium phosphoreum, Pseudomonas
aeruginosa, Thiomicrospira sp. strain L-12, and certain
thiobacilli. Leptothrix discophora, Vitreoscilla Jiliformis,
and Vitreoscilla stercoraria are assigned to the beta subdivision, which also includes a number of thiobacilli (18,43) as
well a s bacteria such a s R h o d o c y c l u s (formerly
Rhodopseudomonas) gelatinosa (13) and Pseudomonas
cepacia. The nature and implications of these inferred relationships are discussed below.
Beggiatoa and Vitreoscilla spp. The Beggiatoa species
characterized by 5s rRNA sequencing (B. alba B18LD) and
its closest identified relative, V . beggiatoides, are the type
119
species of their respective genera. Together they define an
early, and possibly peripheral, branch of the gamma subdivision of the purple bacteria. In contast, V .Jiliformis and V .
stercoraria fall within the beta subdivision. The strain of V .
stercoraria characterized by 16s cataloging also has been
assigned to the beta subdivision (41; also see below). Thus,
the genus Vitreoscilla, as presently constituted, does not
represent a phylogenetically coherent unit; it spans at least
two of the major subdivisions of the purple bacteria.
Thiothrix and Leptothrix spp. The Thiothrix-like strain 30
differs from T. nivea JP2 by the inability of the former strain
to form rosettes. It may also be devoid of holdfast material
(W. R. Strohl, personal communication). However, the two
strains have identical 5s rRNA nucleotide sequences and are
members of the gamma subdivision. More specifically, they
appear, by 5s homology, to be loosely affiliated with the
fluorescent pseudomonads (represented by P. fluorescens
and P. aeruginosa in Fig. 2). The closest identified relative of
T. nivea is the sulfide-oxidizing, procaryotic symbiont of the
marine bivalve Solemya velum. L . discophora falls, by 5s
analysis, into the beta subdivision of purple bacteria and is,
therefore, as distantly related to Thiothrix spp. as V .
filiformis and V . stercoraria are to V . beggiatoides.
Thiovulum sp. The final 5s rRNA sequence presented here
is that of a Thiovulum sp. Although this organism has yet to
be grown axenically, studies of highly enriched Thiovulum
cultures are consistent with a chemoautotrophic character
(38). The Thiovulum 5s rRNA is typically eubacterial(2), but
it is not specifically related to any other sequence in our
reference collection. It has uniformly low sequence homologies (average, 57%) with all other eubacterial 5s rRNAs
represented in the collection (64% to members of the
Spirochaetaceae, including representatives of the genera
Borrellia, Treponema, and Spirochaeta [data not shown]).
These homology levels are similar to those separating the
major eubacterial phyla. However, the definition of a new
eubacterial phylum on this basis would be premature. The
relatively small 5s rRNA molecule is not a particularly
reliable “chronometer” at these low homology levels. A
preliminary analysis of recently obtained, partial 16s rRNA
nucleotide sequence information from this organism suggests that Thiovulum sp. may be loosely affiliated with the
purple bacteria, although not with any of the three subdivisions so far defined (data not shown).
DISCUSSION
Beta subdivision. The beta subdivision presents a striking
variety of phenotypic features which belies the closeness of
the evolutionary relationships of the organisms included. L .
discophora is a sheathed bacterium that deposits iron hydroxide extracellularly and effects the biological oxidation of
manganese; R . gelatinosa is photosynthetic; and P. cepacia
is an archetypal heterotroph. The filaments of Vitreoscilla
spp. are motile by gliding, while other bacteria of the beta
subdivision (e.g., R . gelatinosa) are unicellular and motile
by flagella. By 16s rRNA cluster analysis, two subgroups (in
the nomenclature of Woese et al. [40]) within the beta
subdivision have been defined (43): the beta-1 subgroup is
represented in Fig. 2 by R. gelatinosa, and the beta-2
subgroup is represented by P. cepacia. However, the beta
subdivision is not well represented in the 5s rRNA sequence
collection, so closer scrutiny of the concordance between
the beta subgroup topologies inferred by using these two
molecules is not possible.
By 5s rRNA analysis, L . discophora would appear to
belong to the beta-2 subgroup, although its location in the
Downloaded from www.microbiologyresearch.org by
IP: 88.99.165.207
On: Sun, 18 Jun 2017 04:58:46
,120
INT. J. SYST.BACTERIOL.
STAHL ET AL.
tree (very near the point of divergence between the beta-1
and beta-2 subgroups) makes the assignment of such a
:specific affiliation uncertain. Sphaerotilus natans, an orgainism that, like Leptothrix spp., deposits oxidized salts of
imanganese extracellularly, is by 16s rRNA analysis a close
irelative of R . gelatinosa (43). The V. Jiliformis strains
(ATCC 15551 and L1401-7) also exhibit a clear affiliation
\with R . gelatinosa, and therefore the beta-1 subgroup. No
16s rRNA data are presently available for these organisms
((Table1). V. stercoraria is represented in both the 5s and
16s rRNA sequence collections (Table 1).16s rRNA analy!sisplaces it as a deep but specific associate of the beta-2 ( P .
cepacia) subgroup (43). The 5s data do not permit such a
!specific assignment. Rather, as shown in Fig. 2, V.
stercoraria VT1 would appear to be a deeper branching of
the subdivision than the beta-1 and beta-2 subgroups. This
could be a reflection of strain differences between the two V.
stercoraria cultures inspected or of the limited precision of
phylogenies inferred from 5s rRNA sequences, A peripheral
;associationof V. stercoraria with either the beta-1 or beta-2
:subgroupscannot be excluded on the basis of the 5s rRNA
‘sequences.
Gamma subdivision. The diversity in morphology and
imetabolism characteristic of the purple bacteria is evident as
.well in the gamma subdivision. This diversity is further
(extendedby the present inclusion of members of the genera
Beggiatoa, Thiothrix, and Vitreoscilla. Three major subgroups of the gamma subdivision have been defined by 16s
irRNA analysis (42). Only the gamma-3 subgroup is significantly represented in the 5s rRNA sequence collection (Fig.
‘2). By 16s rRNA comparisons, the gamma-3 subgroup is
seen to contain a number of moderately distinct clusters
I(e.g., the enterics and vibrios, the oceanospirilla, and the
thorescent pseudomonads). Most of these clusters are also
evident by 5s rRNA analysis, although the precise branching
orders linking the various clusters are difficult to assign.
Members of the genus thiothrix are ensheathed, filamentous organisms that produce gliding gonidia as part of their
life cycle. All strains so far examined accumulate sulfur
.when grown on organic carbon in the presence of sulfide or
thiosulfate. Some are probably mixotrophic, requiring reduced inorganic sulfur for growth, whereas others will grow
lheterotrophically (37). The capacity for autotrophic carbon
fixation has not been studied in sulfide-oxygen gradients
such as those which are necessary for the autotrophic
igrowth of a marine Beggiatoa strain (24). T. nivea JP2 and
Thiothrix-like strain 30 are peripherally affiliated with the
fluorescent pseudomonad cluster. As noted above, the closest identified relative of the thiothrix sp. is the bacterial
endosymbiont of Solemya velum, one of a number of recently discovered “gutless” marine invertebrates whose
lbacterial symbionts effectively confer sulfur-based
chemoautotrophy on their hosts. Although the physiology of
these symbionts cannot as yet be studied in cultures, key
enzymes of sulfur-based autotrophy can be detected in
inaturally available material (7).
Beggiatoa spp. are taxonomically distinguished from
Vitreoscilla spp. by their ability to accumulate sulfur when
grown in the presence of sulfide. Only recently has detailed
physiological characterization of any Beggiatoa sp. been
possible (21); all strains available in pure cultures are capable of heterotrophic growth. Apparent mixotrophic growth
of the B18LD strain has been reported (lo), and facultative
chemoautotrophy has been demonstrated for one marine
isolate (24). The placement of B . alba near the base of the
gamma subdivision is qualitatively similar to the placement
of another Beggiatoa sp. by 16s rRNA oligonucleotide
signature analysis (42). However, this does not necessarily
indicate that these two Beggiatoa species are specific relatives of one another. V. beggiatoides is specifically, but
distantly, related to B . alba. For comparison, the evolutionary distance separating B. alba and V. beggiatoides is
greater than that separating Escherichia coli from Proteus
vulgaris. Therefore, differences between B . alba and V .
beggiatoides likely are more profound than loss of sulfuroxidative capacity by the latter organism.
Gliding and flagellar motility. Notable in the present study
are close evolutionary ties between certain filamentious
gliding organisms (V. Jiliformis and V . stercoraria) and those
motile by polar flagella (e.g., R . gelatinosa). Other gliding
organisms, such as the cytophagas and cyanobacteria, fall
well outside the purple bacteria by comparisons of 16s
rRNA catalogs (39). Thus, as with nutritional themes, modes
of motility also fail to define exclusive phylogenetic groupings. However, the close evolutionary relationships observed between flagellated and gliding bacteria may indicate
that one mode of motility can readily be derived from the
other, supporting previous suggestions of a common, underlying mechanism (27).
Phylogeneticplacement of Beggiatoa spp. The present study
addresses the long-standing controversy over the suggested
relationships of some of the morphologically conspicuous
bacteria to the cyanobacteria. Close morphological similarities between certain oscillatorian cyanobacteria and the
beggiatoas have often been cited as indicating evolutionary
relatedness, the beggiatoas sometimes referred to as
apochlorotic derivatives of the former (29, 31, 35). Less
persuasive analogies between Thiothrix spp. and cyanobacteria of the Rivulariaceae have also been raised (30).
Other authors have pointed out that these morphological
similarities more likely are superficial and not indicative of
close evolutionary relationships (21). 16s rRNA sequence
comparisons have shown that the cyanobacteria are a cohesive, natural assemblage, distinct from the purple bacteria
(32, 39). The two Thiothrix strains, the Beggiatoa type
species (B. alba BlSLD), and the Beggiatoa strain characterized by 16s rRNA cataloging are specifically related to the
purple bacteria, not to the cyanobacteria.
5s rRNA in contrast to 16s rRNA as a phylogenetic tool. As
components of the translation apparatus, the 5s and 16s
rRNAs presumably are subject to similar evolutionary pressures. Indeed, phylogenetic trees constructed by using 5s or
16s rRNA sequences are generally in good agreement (12,
18, 33, 34), as are relative levels of “similarity” between 5s
rRNA sequences and 16s rRNA oligonucleotide catalogs
(26). Past usage of 5s or 16s rRNA for establishing
phylogenetic relationships has largely been dictated by the
relative convenience of obtaining their sequences. Until
recently, oligonucleotide cataloging, a relatively cumbersome and expensive process, provided the only access to
16s rRNA sequences (9). In comparison, 5s rRNA can be
sequenced relatively easily, but the small size of that molecule, about 120 nucleotides, compromises the precision with
which relationships between organisms can be established.
This is because the statistical uncertainty in sequence homologies is dependent upon the number of residues employed. Thus, while 5s rRNA is useful for the assignment of
organisms to broad phylogenetic domains, for instance the
alpha, beta, and gamma subdivisions of the purple bacteria,
finer relationships often are difficult to resolve (see, for
example, the above discussion regarding L . discophora).
Sequence comparisons of 16s rRNA, because of the larger
Downloaded from www.microbiologyresearch.org by
IP: 88.99.165.207
On: Sun, 18 Jun 2017 04:58:46
PHYLOGENY OF SULFIDE-OXIDIZING BACTERIA
VOL. 37, 1987
molecular size (>1,000 nucleotides), yield more accurate
relationships than those of 5s rRNA. This consideration,
together with recent developments in rapid rRNA sequencing techniques (17), mean that future analyses likely will
emphasize the larger rRNAs, the 16s and 23s rRNAs.
Nonetheless, the accumulated 5s rRNA sequence data base,
as used here, will remain a valuable reference for outlining
phylogenetic diversity.
ACKNOWLEDGMENTS
We thank William R. Strohl for many helpful discussions and
critical readings of the manuscript, and Holger Jannasch and Carl
Wirsen for assistance with the Thiovulum enrichments.
This project was partially funded by National Institutes of Health
grant GM34527 to N.R.P.
LITERATURE CITED
1. Dams, E., A. Vandenberghe, and R. De Wachter. 1983. Nucle-
otide sequences of the 5s rRNAs of Azotobucter vinelandii,
Pseudomonas aeruginosa, and Pseudomonas fluorescens with
some notes on 5s RNA secondary structure. Nucleic Acids Res.
11:1245-1252.
2. Delihas, N., and J. Andersen. 1982. Generalized structures of the
5s ribosomal RNAs. Nucleic Acids Res. 10:7323-7344.
3. Donis-Keller, H., A. Maxam, and W. Gilbert. 1977. Mapping
adenines, guanines, and pyrimidines in RNA. Nucleic Acids
Res. 4:2527-2538.
4. Crombach, W. H., W. L. Van Veen, A. W. van der Vlies, and
W. C. Bob. 1974. DNA base composition of some sheathed
bacteria. Antonie van Leeuwenhoek J. Microbiol. Serol. 40:
217-220.
5 . Elwood, H. J., G. J. Olsen, and M. L. Sogin. 1985. The smallsubunit ribosomal RNA gene sequences from the hypotrichous
ciliates Oxytricha nova and Stylonychia pustulata. Mol. Biol.
Evol. 2:399410.
6. Erdmann, V. A,, J. Wolters, E. Huysmans, and R. De Wachter.
1985. Collection of published 5 S , 5.8S, and 4.5s ribosomal RNA
sequences. Nucleic Acids Res. 13:r105-r153.
7. Felbeck, H., J. J. Childress, and G. M. Somero. 1981. CalvinBenson cycle and sulphide oxidation enzymes in animals from
sulphide-rich habitats. Science 293:291-293.
8. Fitch, W. M., and E. Margoliash. 1967. Construction of
phylogenetic trees: a method based on mutational distances as
estimated from cytochrome c sequences is of general applicability. Science 155279-284.
9. Fox, G. E., E. Stackebrandt, R. B. Hespell, J. Gibson, J.
Maniloff, T. A. Dyer, R. S. Wolfe, W. E. Balch, R. S. Tanner,
L. J. Magrum, L. B. Zablen, R. Blakemore, R. Gupta, L. Bonen,
B. J. Lewis, D. A. Stahl, K. R. Luehrsen, K. N. Chen, and C. R.
Woese. 1980. The phylogeny of prokaryotes. Science 209:
457-463.
10. Gude, H., W. R. Strohl, and J. M. Larkin. 1979. Mixotrophic
and heterotrophic growth of Beggiatoa alba in continuous
culture. Arch. Microbiol. 129:357-360.
11. Hassur, S. M., and H. W. Whitlock. 1974. UV shadowing-a
new and convenient method for the location of ultravioletabsorbing species in polyacrylamide gels. Anal. Biochem.
59: 162-1 64.
12. Hori, H., and S. Osawa. 1979. Evolutionary change in 5s RNA
secondary structure and a phylogenetic tree of 54 5s rRNA
species. Proc. Natl. Acad. Sci. USA 76:381-385.
13. Imhoff, J. R., H. G. Triiper, and N. Pfennig. 1984. Rearrangement of the species and genera of the phototrophic “purple
nonsulfur bacteria.” Int. J. Syst. Bacteriol. 34:340-343.
14. Jukes, T. M., and C. R. Cantor. 1969. Evolution of protein
molecules, p. 21-132. In H. N. Munro (ed.), Mammalian protein
metabolism. Academic Press, Inc., New York.
15. Kimura, M., and T. Ohta. 1972. On the stoachastic model for
estimation of mutational distance between homologous proteins. J. Mol. Evol. 2:87-90.
121
16. Kuenen, J. G. 1975. Colourless sulfur bacteria and their role in
the sulfur cycle. Plant Soil 43:49-76.
17. Lane, D. J., B. Pace, G. J. Olsen, D. A. Stahl, M. L. Sogin, and
N. R. Pace. 1985. Rapid determination of 16s ribosomal RNA
sequences of phylogenetic analyses. Proc. Natl. Acad. Sci.
USA 82:6955-6959.
18. Lane, D. J., D. A. Stahl, G. J. Olsen, D. J. Heller, and N. R.
Pace. 1985. Phylogenetic analysis of the genera Thiobacillus and
Thiomicrospira by 5s ribosomal RNA sequences. J. Bacteriol.
163:75-81.
19. la Riviere, J. W. M., and K. Schmidt. 1981. Morphologically
conspicuous sulfur-oxidizing eubacteria, p. 1037-1048. In M. P.
Starr, H. Stolp, H. G. Triiper, A. Balows, and H. G . Schlegel
(ed.), The prokaryotes-a handbook on habitats, isolation, and
identification of bacteria, vol. I. Springer-Verlag, New York.
20. Larkin, J. M., and D. L. Shinabarger. 1983. Characterization of
Thiothrix nivea. Int. J. Syst. Bacteriol. 33:841-846.
21. Larkin, J. M., and W. R. Strohl. 1983. Beggiatoa, Thiothrix,
and Thioploca. Annu. Rev. Microbiol. 37:341-367.
22. Leadbetter, E. R. 1974. Order 11. Cytophagales nomen novum,
p. 99. In R. E. Buchanan and N. E. Gibbons (ed.), Bergey’s
manual of determinative bacteriology, 8th ed. The Williams and
Wilkins Co., Baltimore.
23. Mulder, E. G., and M. H. Deinema. 1981. The sheathed bacteria, p. 425-440. In M. P. Starr, H. Stolp, H. G. Triiper, A.
Balows, and H. G. Schlegel (ed.), The prokaryotes-a handbook on habitats, isolation, and identification of bacteria, vol. I.
Springer-Verlag, New York.
24. Nelson, D. C., and H. W. Jannasch. 1983. Chemoautotrophic
growth of a marine Beggiatoa in sulfide-gradient cultures. Arch.
Microbiol. 136:262-269.
25. Olsen, G. J., D. J. Lane, S. J. Giovannoni, N. R. Pace, and D. A.
Stahl. 1986. Microbial ecology and evolution: a ribosomal RNA
approach. Annu. Rev. Microbiol. 40:337-365.
26. Pace, N. R., D. A. Stahl, D. J. Lane, and G. J. Olsen. 1986. The
analysis of natural microbial populations by ribosomal RNA
sequences. Adv. Microb. Ecol. 9:l-55.
27. Pate, J. L., and L. E. Chang. 1979. Evidence that gliding
motility in prokaryotic cells is driven by rotary assemblies in the
cell envelopes. Curr. Microbiol. 259-64.
28. Peattie, D. A. 1979. Direct chemical method for sequencing
RNA. Proc. Natl. Acad. Sci. USA 76:1760-1764.
29. Pringsheim, E. G. 1949. The relationship between bacteria and
Myxophyceae. Bacteriol. Rev. 13:47-91.
30. Reichenbach, H. 1981. Taxonomy of the gliding bacteria. Annu.
Rev. Microbiol. 35339-364.
31. Reichenbach, H., and M. Dworkin. 1981. Introduction to the
gliding bacteria, p. 315-327. In M. P. Starr, H. Stolp, H. G.
Trtiper, A. Balows, and H. G. Schlegel (ed.), The prokaryotesa handbook on habitats, isolation, and identification of bacteria,
vol. I. Springer-Verlag, New York.
32. Stackebrandt, E., and C. R. Woese. 1981. The evolution of
prokaryotes, p. 1-31. In M. J. Carlisle, J. R. Collins, and
B. E. B. Moseley (ed.), Molecular and cellular aspects of microbial evolution. Cambridge University Press, Cambridge.
33. Stahl, D. A., D. J. Lane, G. J. Olsen, and N. R. Pace. 1984.
Analysis of hydrothermal vent-associated symbionts by ribosomal RNA sequences. Science 224:409411.
34. Stahl, D. A., D. J. Lane, G. J. Olsen, and N. R. Pace. 1985.
Characterization of a Yellowstone hot spring microbial community by 5s ribosomal RNA sequences. Appl. Environ. Microbiol. 49:1379-1384.
35. Strohl, W. R., and J. M. Larkin. 1978. Cell division and
trichome breakage in beggiatoa. Curr. Microbiol. 1:151-155.
36. Strohl, W. R., T. M. Schmidt, N. H. Lawry, M. J. Mezzino, and
J. M. Larkin. 1986. Characterization of Vitreoscilla beggiatoides and V .jiliformis sp. nov., nom. rev., and comparison to
V . stercoraria and B e g g i a t o a . Int. J. Syst. Bacteriol.
36:302-313.
37. Williams, T. M., and R. F. Unz. 1985. Filamentous sulfur
bacteria of activated sludge: characterization of Thiothrix, Beggiatoa, and Eikelboom type 021N strains. Appl. Environ.
Microbiol. 49:887-898.
Downloaded from www.microbiologyresearch.org by
IP: 88.99.165.207
On: Sun, 18 Jun 2017 04:58:46
I22
STAHL ET AL.
INT.J. SYST.BACTERIOL.
38. Wirsen, C. O., and H. W. Jannasch. 1978. Physiological and
morphological observations on Thiovulum sp. J. Bacteriol.
136:765-774.
39. Woese, C. R. 1985. Why study evolutionary relationships among
bacteria?, p. 1-30. In K. H. Schleifer and E. Stackebrandt (ed.),
Evolution of prokaryotes. Academic Press, Inc., New York.
40. Woese, C. R., E. Stackebrandt, T. J. Macke, and G. E. Fox.
1985. A phylogenetic definition of the major eubacterial taxa.
syst. Appl. Microbiol. 6:143-151.
41. Woese, C. R., E. Stackebrandt, W. G. Weisburg, B. J. Paster,
M. R. Madigan, V. J. Fowler, C. M. Hahn, P. Blanz, R. Gupta,
K. H. Nealson, and G. E. Fox. 1984. The phylogeny of purple
bacteria: the alpha subdivision. Syst. Appl. Microbiol.
5: 315-326.
42. Woese, C. R., W. Weisburg, C. M. Hahn, B. Paster, L. B.
Zablen, B. J. Lewis, W. Ludwig, and E. Stackebrandt. 1984. The
phylogeny of purple bacteria: the gamma subdivision. Syst.
Appl. Microbiol. 6:25-33.
43. Woese, C. R., W. G. Weisburg, B. J. Paster, C. M. Hahn, R. S.
Tanner, N. R. Krieg, H. P. KOOPS,H. Harms, and E.
Stackebrandt. 1984. The phylogeny of purple bacteria: the beta
subdivision. Syst. Appl. Microbiol. 5:327-336.
Downloaded from www.microbiologyresearch.org by
IP: 88.99.165.207
On: Sun, 18 Jun 2017 04:58:46