Reprint Scries
15 May 1987, Volume 236, pp. 822-824
SdENCE
Isolation of Extremely Thermophilic Sulfate Reducers:
Evidence for a Novel Brandl of Archaebacteria
K A R L O.
STETTER, GERTA L A U E R E R , M I C H A E L T H O M M , A N D A N N E M A R I E
NEUNER
Copyright © 1987 by thc American Association for the Advanccmcnt of Science
Isolation of Extremely Thermophilic Sulfate Reducers:
Evidence for a Novel Branch of Archaebacteria
sulfate reduetion has been observed up to
now exclusively within some eubacteria (5).
Eubacterial sulfate reducers are mesophiles
or moderate thermophiles and play an imK A R L O . STETTER, G E R T A L A U E R E R , M I C H A E L T H O M M ,
portant role in the global sulfur cycle (5).
ANNEMARIE N E U N E R
We describe here a novel group of extremely
thermophilic sulfate reducers that belong to
Extremely thermophilic archaebacteria are known to be metabolizers of elemental
the archaebacterial kingdom.
sulfur and the methanogens. A novel group of extremely thermophilic archaebacteria is
To study microbial life at extremely high
described, which consists of sulfate-respiring organisms that contain pure factor 420
temperatures, we collected 45 anaerobic
and that have been isolated f r o m marine hydrothermal Systems in Italy. They possess a
samples (6) of hot Sediments from marine
third type of archaebacterial R N A polymerase structure previously unknown, indicathydrothermal Systems in the vicinities of
ing an exceptional phylogenetic position. Most likely, this group represents a third
Vulcano and Stufe di Nerone, Italy. The
major branch within the archaebacteria. The existence of sulfate reducers at extremely
original temperatures of the Sediments
high temperatures could explain hydrogen sulfide formation in hot sulfate-containing
ranged between 70° and 100°C. In the
environments, such as submarine hydrothermal Systems and deep oil wells.
laboratory, oxygen-free marine mineral medium supplemented with 0.1% (w/v) yeast
of R N A polymerase structures (2), 165 riWO MAIN SUBDIVISIONS OF ARextract (Difco) was inoculated with portions
bosomal R N A (rRNA) sequences (7), and
chaebacteria are distinguished: one
metabolic properties. Members of the group of each sample (7, 8) and then ineubated at
comprising the methanogenic bacte85°C (gas phase: N / C 0 = 80/20). After 1
of archaebacterial sulfur metabolizers are
ria and the extreme halophiles, and a second
week we observed coeeoid bacterial cells in
able to grow by either oxidizing or reducing
consisting of the sulfur-metabolizing ex28 of the 45 cultures that we attempted to
elemental sulfur, depending on the genera
treme thermophiles (1). These branches can
(3, 4). Energy conservation by dissimilatory form. They showed a strong blue-green
be recognized on the basis of a comparison
fluorescence under the ultraviolet (UV) microscope at 420 nm characteristic of methTable 1. Substrates for growth of isolatc VC-16. Portions of 20-ml anaerobic sulfate- or thiosulfateanogenic bacteria (8).
containing mineral medium (MGG) (7) were supplemented with possible Substrates (2 g/liter) and then
Surprisingly, however, only traces of
inoculated (1% inoculation) with strain VC-16. Incubation was usually at 85° and 65°C in the case of
glucose and pyruvate, respectively. Substrate utilization was determined after four transfers in sequence mediane (about 10~ times that produced
into the samc culture medium (1% inoculation), cach foüowcd by an incubation period. Acids were
by a methanogen culture of comparable cell
added as sodium salts.
density) were detected. The novel organisms
were cloned by plating on culture medium
Class of nutrients
solidified by 1.5% agar (Oxoid). The plates
Intcrmediates in
were ineubated anaerobically at 70°C (8).
Sugar
Complex Substrates
anaerobic degradation
Isolate VC-16 from Vulcano was the first
Yeast extract, casamino acids,* beef extract, peptone,
Glucose
Molecular hydrogen,* formate,
cell homogenates of E. coli, Lactobacillus bavaricus
formamide, L( + )- and
and Mcthanothermus fervidus
D(-)-lactate, pyruvate
Institut für Biochemie, Genetik und Mikrobiologie, Uni-
T
2
2
4
•Growth ortfy in thiosulfatc-containing medium.
822
versität Regensburg, Universitätsstraße 31, D-8400 Regensburg, FederaJ Republic of Germany.
SCIENCE, VOL. 236
1 2 3
4
5
6
7
r R N A sequence (1, 9), (ii) the existence o f
an adenosine diphosphate-ribosylable protein in the crude extract (10, 77), (iii) the
presence o f phytanyl ether lipids (72, 13),
(iv) a cell wall consisting o f glycoprotein
instead o f murein (14), and (v) the presence
of an R N A polymerase with eight subunits
resistant to 200 (xg/ml o f rifampicin and
streptolydigin (2). D N A from isolate V C 16 had a content o f guanosine plus cytosine
of 46 m o l % (4, 75). Spectra o f the enriched
Compound that fluoresced blue-green in the
F i g . 3. SDS-polyacrylamide gel electrophoresis
U V light showed the same excitation and
and immunochallenging (Western blots) o f
emission maxima (419 and 467.5 n m , reDNA-dependent R N A polymerascs. R N A polyspectively) as pure factor 420 ( F ) from
merases o f 1, Escherichia coli; 2 and 4a to 7a, M.
Methanobacterium thermoautotrophicum and
thermolithotrophicus; 3 and 4 b to 7b, isolate V C as enriched F
from Methanococcus thermo- 16. Coomassie staining o f 1 to 3. The positions o f
molecular weight Standards in the Polyacrylamide
lithotrophicus (Fig. 2) (16). This finding suggel are o n the left side. Escherichia coli I W A
gests that the unknown Compound was an polymerase was obtained from Boehringer M a n n 8-OH-5-deazaflavin similar to F
in methheim. R N A polymerases o f M. thermolithotrophiF i g . 1. Electron micrographs o f isolate V C - 1 6 ,
cus and V C - 1 6 were purified by Standard methods
anogenic bacteria (77). Cell extracts o f V C platinum-shadowed. ( A ) Cells and empty enve(23). Western blots o f components o f R N A polylope, ( B ) cell with two flagella protruding from a
16 also contained substances showing the
merases o f M. thermolithotrophicus and o f isolate
Single origin (arrow). Scale bars, 1 u.m.
same ultraviolet/visual spectrophotometry
V C - 1 6 were challenged (2, 26) with antibodies
spectrum as methanopterin and 5,10-methdirected against single components A , B ' , B", and
strain obtained in pure culture and was enyltetrahydro methanopterin (18, 19). In
C (lanes 4 to 7, respectively) o f the R N A polymerase o f M. thermoautotrophicum (2). B o u n d
therefore studied in detail.
contrast to methanogens, no substance
antibodies were visualized with peroxidase-couCultures o f V C - 1 6 consisted o f irregulär showing the spectrum o f factor 430 ( F )
pled antibodies to immunoglobulin G .
spherical cells with envelopes composed o f (18) could be detected. Moreover, coensubunits in hexagonal array (Fig. 1A). The
zyme M (2-mercaptoethanesulfonic acid) as
cells were motile and possessed flagella prodetermined by the bioassay described by
duetase exist in cell extracts o f V C - 1 6 and
truding from a common location (Fig. 1B,
Balch and Wolfe (20) was not evident. The
are under investigation (22).
arrow). The new organisms grew within a
formation o f mediane without coenzyme M
The phylogenetic position o f the newly
temperature ränge between 64° and 92°C,
and F o could be analogous to the " m i n i
isolated sulfate reducers within the archaewith an optimum at approximately 83°C mediane production" described for some
bacteria was investigated by comparing
(corresponding to a doubling time o f about
eubacteria (27). The essential enzymes o f the
DNA-dependent R N A polymerases. Within
4 hours). Growth depended on the presence dissimilatory sulfate reduction pathway,
archaebacteria, only two distinet types o f
o f sulfate, which was supplied in the culture
namely, adenosine triphosphate sulfurylase,
R N A polymerase struetures are known that
medium (14 fimol/ml). Large amounts o f adenylyl-sulfate reductase, and bisulfite reparallel the two phylogenetic branches: the
H S (up to 6 jimol/ml) and C 0 were
BAC-type o f the S°-metabolizing and the
1.0 i
formed during growth, indicating that the
A B ' B " C - t y p e o f the methanogenic-halophilnew organisms carried out dissimilatory sulic archaebacteria. The basis o f this difference
f\
»
0.8
fate reduction (5). In addition, mediane (up
is the occurrence o f two Polypeptides, B '
/
\
to 0.01 jxmol/ml) was formed. Sulfate could
and B", in the A B ' B " C - t y p e , which show
i /
1 /> !
/ /
be replaced by thiosulfate (14 |xmol/ml) and
homology to component B o f the other type
0.6
» / v
>
by sulfite (1 |xmol/ml), but not by elemental
(2, 23). The R N A polymerase purified from
' V
\ \
'7
\ \
sulfur. Molecular hydrogen and some simple
V C - 1 6 exhibited eight identifiable compo/
'
/
0.4
V
organic molecules served as Substrates (Tanents, each having a different molecular
ble 1). In contrast to eubacterial sulfate
weight (Fig. 3, lane 3). Structural homolo0.2
reducers (5), complex organic Substrates and
gies between the R N A polymerases o f the
V
"
"1
even glucose were used (Table 1). As expectnovel sulfate reducer and other archaebacV
0
ed, crude oil (Höhenrain; Preussag, Hannoteria were revealed by immunological cross440
460
380
420
ver) did not support growth. However, it
reactions. Antibodies directed against the
Wavelength of
Wavelength of
excitation (nm)
emission (nm)
was not inhibitory when present with Subsingle components o f M. thermoautotrophistrates. This finding suggests that the new F i g . 2. Excitation and emission spectra o f the cum R N A polymerase were challenged with
organisms could be responsible for the for- fluorescent Compound o f the new isolate V C - 1 6 isolated components o f the enzymes o f V C (—) and o f F o from M. thermolithotrophicus 16 (Fig. 3, lanes 4b to 7b) and, as a control,
mation o f H S in geothermally heated o i l
( — ) , and M. thermoautotrophicum (
). T h e
wells ("sour oil") when suitable Substrates
of M. thermolithotrophicus (Fig. 3, lanes 4a to
fluorescent Compound o f isolate V C - 1 6 and F o
and sulfate-containing water are present. In
from M. thermolithotrophicus was extracted from 7a). The three heaviest components o f the
batch culture with L( + )-lactate as the SubV C - 1 6 enzyme showed serological crossthe cells with acetone and was purified by chromatography o n a QAE-Sephadex A 25 column
strate, 140-g cells (wet weight) were obreaction with components A , B ' , and B " o f
(16). The excitation and emission spectra were
tained in a 300-liter, enamel-protected ferthe methanogen enzyme, so far indicating
measured in a Cary 118 C spectrophotometer and
mentor (Pfaudler, West Germany).
homology to the A B ' B " C strueture. A n t i a Hitachi-Perkin-Elmer fluorescence spectrophoThe isolate V C - 1 6 was recognized as an tometer M P F - 4 4 A , respectively (buffer: 50 mAf bodies directed against the fourth largest
tris,/>H 7.5, and IM N a C l ) .
archaebacterium o n the basis o f (i) its 16S
component (C) o f the enzyme o f methano420
4 2 0
4 2 0
430
43
2
Relative fluorescence
2
2
\
42
42
15 M A Y 1987
REPORTS
823
genic bacteria (Fig. 3, lane 7a) exhibited a
strong cross-reaction with the heaviest component (A) of the VC-16 R N A polymerase
(Fig. 3, lane 7b). This cross-reaction demonstrates that the largest component of the
R N A polymerase of the sulfate reducer contained structural elements of components A
and C of the methanogen enzyme. The new
sulfate reducer thus exhibits a previously
unknown third type of R N A polymerase
(A + C)B'B"
within
the
archaebacteria,
which is phylogenetically removed from
those of the two accepted archaebacterial
branches. This finding provides evidence for
the existence of a third phylogenetic branch
within the archaebacterial kingdom.
On the basis of their phylogenetic uniqueness, archaebacterial sulfate reducers may
have existed since ancient times. Early Archaean ocean waters were generally poor in
sulfate and were thus unfavorable for the
presence
of sulfate
reducers
that
were
thought to have originated later (24). This
theory is in line with the lack of significant
biogenic sulfur isotope fractionation within
such Sediments (24). However, there should
824
Zentralbl. Bakteriol. Mikrobiol. Hyq. 1 Abt. Orig. C 3,
228 (1982).
13. T. A. Langworthy, personal communication.
14.
H.
König, personal communication.
chaean hydrothermal Systems (25). There15. J. Marmur and P. DotvJ. Mol. Biol. 5, 109 (1962).
fore, sulfate reduction by extremely thermo16. P. Schönheit, H. Kcweloh, R. K. Thauer, FEMS
(Ted. Eur. Microbiol. Soc.) Microbiol. Leu. 12, 347
philic archaebacteria could have existed
(1981).
since early Archaean times and may be an
17. P. Cheescman, A. Toms-Woods, R. S. Wolfe, /.
Bacteriol. 112, 527 (1972).
ancient type of metabolism.
18. A. Pfaltz, personal communication.
19. J. T. Keltjens, M. J. Huberts, W. H. Laarhoven, G.
D. Vogels, Eur.]. Biochem. 130, 537 (1983).
20. W. E. Balch and R. S. Wolfe,/. Bacteriol. 137, 256
REFERENCES AND NOTES
(1979).
1. C R . Wocse and G. J. Olsen, Syst. Appl. Microbiol. 7,
21. R. Schauder, B. Eikmanns, R. K. Thauer, F. Wid161 (1986).
del, G. Fuchs, Arch. Microbiol 145, 162 (1986).
2. W. Zillig, K. O. Stetten, R. Schnabel, M. Thomm,
22. H. G. Trüper, personal communication.
in The Bacteria, C. R. Woese and R. S. Wolfe, Eds. 23. M. Thomm, J. Madon, K. O. Stetter, Biol. Chem.
(Academic Press, Orlando, FL, 1985), vol. 8, pp.
Hoppe-SeyUr 367, 473 (1986).
499-524.
24. E. M. Cameron, Nature (London) 296, 145 (1982).
3. T. D. Brock, K. M. Brock, R. T. Bellv, R. L. Weiss,
25. K. Hattori and E. M. Cameron, ibid. 319, 45
Arch. Microbiol. 84, 54 (1972).
(1986).
4. A. Segerer, K. O. Stetter, F. Klink, Nature (London) 26. H. Towbin, T. Staehelin, J. Gordon, Proc. Natl.
313, 787 (1985).
Acad. Sei. U.SA. 76, 4350 (1979).
5. F. Widdel, in Anaerobic Bacteria in Habitats Other
27. We thank F. Klink, H. König, T. A. Langworthy, A.
Than Men, E. M. Barnes and G. C. Mead, Eds.
Pfaltz, H. G. Trüper, and C. R. Woese for commu(Blackwell, Oxford, 1986), pp. 157-184.
nicating unpublisned results; R. Thauer for the gift
6. K. O. Stetter, Nature (London) 300, 258 (1982).
of F o and for discussions; R. Jaenicke for spectros7. H. Huber, M. Thomm, H. König, G. Thies, K. O.
copy; N. Pfennig, D. Ubben, and P. Ziegler for
Stetter, Arch. Microbiol. 132, 47 (1982).
discussions; and W. Kleinitz for oil samples. Sup8. W. E. Balch, G. E. Fox, L. J. Magrum, C. R. Woese,
ported by grants from the Deutsche ForschungsgeR. S. Wolfe, Microbiol. Rer. 43, 260 (1979).
meinschaft (SFB 43 and Schwerpunkt: Methano9. C. R. Woese, personal communication.
gene Bakterien) and the Fonds der Chemischen
10. M. Kessel and F. Klink, Nature (London) 287, 250
Industrie to K.O.S.
(1980).
11. F. Klink, personal communication.
12. T. A. Langworthy, T. G. Tornabene, G. Holzer,
1 December 1986; accepted 20 February 1987
have been reasonable quantities of sulfate of
magmatic origin present locally within Ar-
42
SCIENCE, VOL. 236