Distribution and diversity of thermophilic sulfate-reducing bacteriawithin a Cu-Pb-Zn mine (Toyoha, Japan)

Tatsunori Nakagawa a, Satoshi Hanada b, Akihiko Maruyama b, Katsumi Marumo c,Tetsuro Urabe d, Manabu Fukui a;�

a Department of Biological Science, Graduate School of Science, Tokyo Metropolitan University, Minami-ohsawa 1-1, Hachioji, Tokyo 192-0397, Japanb Research Institute of Biological Resources, National Institute of Advanced Industrial Science and Technology (AIST), Tsukuba Central 6, 1-1-1 Higashi,

Tsukuba, Ibaraki 305-8566, Japanc Marine Resources and Environment Institute, National Institute of Advanced Industrial Science and Technology (AIST), Tsukuba Central 7, 1-1-1 Higashi,

Tsukuba, Ibaraki 305-8567, Japand Department of Earth and Planetary Science, University of Tokyo, Hongo, Bunkyo-ku, Tokyo 113-0033, Japan

Received 9 February 2002; received in revised form 2 May 2002; accepted 3 May 2002

First published online 10 July 2002

Abstract

The distribution and diversity of thermophilic sulfate-reducing bacteria at the Cu-Pb-Zn Toyoha underground mine, Japan, wereinvestigated using denaturing gradient gel electrophoresis analysis based on the 16S rRNA gene, and sequence analysis of the dissimilatorysulfite reductase gene. Hydrothermal waters from different boreholes penetrating the Cu-Pb-Zn sulfide veins were collected andconcentrated with a sterile filter (pore size: 0.2 Wm) at sites A (64‡C), B (71‡C), and C (48‡C). Microbial mats developed at sites A (53‡C),B (66‡C), and D (73‡C) were harvested. The denaturing gel electrophoresis analysis showed 17 bacterial and three archaeal bandsincluding two of spore-forming, Gram-positive sulfate-reducing bacteria, Desulfotomaculum-like 16S rDNA sequences from site B. Thephylogenetic analysis of 16 clone families of dissimilatory sulfite reductase genes indicated that they are Desulfotomaculum-,Thermodesulforhabdus-like sequences, and unresolved sequences. We obtained evidence of the diversity and distribution of microbesrelated to thermophilic sulfate-reducing bacteria within effluent-hydrothermal groundwater and microbial mats in the thermophilicsubsurface environment of the Toyoha Mine. = 2002 Federation of European Microbiological Societies. Published by Elsevier ScienceB.V. All rights reserved.

Keywords: Thermophilic sulfate-reducing bacterium; Dissimilatory sul¢te reductase; Cu-Pb-Zn mine; Denaturing gradient gel electrophoresis ;Desulfotomaculum ; Thermodesulforhabdus

1. Introduction

Sulfate-reducing bacteria (SRB) play an important rolein the formation of ferric sul¢des (FeS and FeS2) withinorganic and sulfate-rich marine anaerobic sediments [1].However, relatively little is known about sul¢de formationdue to the action of thermophilic SRB in high-temperaturehydrothermal environments. Currently identi¢ed thermo-philic SRB and sulfate-reducing archaea (SRA) are found

within the genera Thermodesulforhabdus [3], Desulfotomac-ulum [4^6], Thermodesulfobacterium [7^10], Thermodesulfo-vibrio [7,11], and Archaeoglobus [12].Dissimilatory sul¢te reductase (DSR) is a key enzyme

that catalyzes the reduction of sul¢te to sul¢de duringanaerobic sulfate respiration. Recent studies on the diver-sity of SRB indicate that sequence analysis of DSR genesis e¡ective for the detection of SRB within a complicatedmicrobial community structure such as sediments [13,14],microbial mats [15], the dorsal surface of a polychaeteannelid in a high-temperature environment of activedeep-sea hydrothermal vents [17], and uranium tailing sites[18]. Furthermore, a DSR gene-based molecular ecologicalapproach can be a powerful tool for obtaining data onpotential sul¢de production through anaerobic sulfate res-piration by a sulfate reducer, since it is not necessary to

0168-6496 / 02 / $22.00 = 2002 Federation of European Microbiological Societies. Published by Elsevier Science B.V. All rights reserved.PII: S 0 1 6 8 - 6 4 9 6 ( 0 2 ) 0 0 2 9 3 - 3

* Corresponding author. Tel. : +81 (426) 77-2581;Fax: +81 (426) 77-2559.E-mail address: fukui-manabu@c.metro-u.ac.jp (M. Fukui).

FEMSEC 1378 7-8-02

FEMS Microbiology Ecology 41 (2002) 199^209

www.fems-microbiology.org

relate closely to known pure cultures, such as in the 16SrDNA-based approach.The Toyoha Mine surveyed in the present study oper-

ates at depths of 80^600 m below ground level. Thewall rocks maintain a high temperature, because themining site is in an active geothermal system. Chalco-pyrite, galena, sphalerite and pyrite are obtained as majorore minerals from Cu-Pb-Zn sul¢de veins which were pre-cipitated at 250^300‡C during very early mineralization(6 3 million years ago) at the Toyoha Mine. The sul¢deveins contain some pores ¢lled with anoxic hydrothermalwater, suggesting that the veins were conduits for ore-forming £uids and that the anoxic hydrothermal water isa relic of ore-forming £uids. Moreover, dense microbialmats attach to a wall of driftway under running geother-mal water. Laser ablation mass spectroscopy analyses ofthe sulfur isotopic composition of abundant ¢ne-grainedpyrite indicated that the pyrites in the aragonite sinter,which developed in the discharging area of hydrother-mal £uids, were formed by microbial activity [19]. Simi-larly, 34S depletion of sul¢des (FeS and FeS2) withinmicrobial mats was stimulated through dissimilatory sul-fate reduction by sulfate reducers [20]. Hence, it was es-sential to detect the sulfate-reducing prokaryotes withine¥uent-hydrothermal groundwater and microbial mats inthe thermophilic subsurface environment of the ToyohaMine.The aim of this study is to clarify the distribution and

phylogenetic diversity of thermophilic sulfate reducerswithin e¥uent-hydrothermal groundwater and microbialmats developed at the Toyoha underground mine, Japan.We used sequence analysis of 16S rDNA fragments ondenaturing gradient gel electrophoresis (DGGE), andDSR genes obtained by cloning technique to detect ther-mophilic sulfate-reducing prokaryotes in hydrothermalgroundwater and microbial mats at the mine.

2. Materials and methods

2.1. Study area and collection of hydrothermal water andmicrobial mats

The Toyoha Mine is located about 42 km southwest ofSapporo, Hokkaido, Japan. Samples were collected fromthe subsurface mining sites : site A (3500 m level from thesurface), site B (3550 m level), site C (3550 m level), andsite D (3550 m level) within the mine in December 2000.Approximately 1 l of hydrothermal water at sites A, B,and C was ¢ltered with Sterivex1 (pore size: 0.2 Wm;Millipore, Bedford, MA, USA) to harvest microorganismsin the hydrothermal water. The ¢lters were kept in anaer-obic bags (AnaeroPack0, Mitsubishi Gas Chemical, To-kyo, Japan) to prevent the oxidation of samples. White-colored microbial mats developed on the wall of driftwaynear sites A, B, and D. The biomats at these sites, desig-

nated Am, Bm, and Dm, were kept in sterile plastic tubes.All samples were transferred to our laboratory under coolconditions (4‡C). A part of the ¢lter was used for nucleicacid extraction of the microorganisms. The ¢lters and bio-mats for a molecular approach were placed into 1.5-mlmicrocentrifuge tubes and 2-ml screw-cap microcentrifugetubes containing 0.5 g of glass beads, and then stored at380‡C.

2.2. Extraction of nucleic acids

Nucleic acids of microorganisms on ¢lters were ex-tracted as described by Ishii et al. [21]. Extraction ofnucleic acids from microbial mats was performed usingthe HTP spin column method described by Purdy et al.[22].

2.3. PCR ampli¢cation of 16S rDNA and dsrAB

DNA fragments encoding the 16S rRNA gene of thedomain Bacteria and the domain Archaea were ampli¢edusing two sets of primers as follows: 341F with a GC-clamp (341F, 5P-CCTACGGGAGGCAGCAG-3P ; GC-clamp, 5P-CGCCCGCCGCGCCCCGCGCCCGTCCCG-CCGCCCCCGCCCG-3P) and 907R (5P-CCGTCAATTC-CTTTRAGTTT-3P) for the domain Bacteria [23,24], andARC344F with GC-clamp (5P-ACGGGGYGCAGCAG-GCGCGA-3P) and ARC915R (5P-GTGCTCCCCCGCC-AATTCCT-3P) for the domain Archaea [25,26]. PCR con-ditions for the bacterial and archaeal primers were carriedout as described by Muyzer et al. [23,24], and Casamayoret al. [27], respectively. PCR ampli¢cations were per-formed with 100-Wl volumes containing 5^100 ng of tem-plate DNA, 1UEX Taq bu¡er (Takara Shuzo, Kyoto,Japan), 250 WM of each deoxynucleoside triphosphate,25 pmol of each primer, 2.5 U of EX Taq DNA polymer-ase (Takara), and 2 drops of mineral oil (Sigma). PCRproducts were analyzed by electrophoresis in 2% w/v Nu-sieve 3:1 agarose (FMC, Rockland, ME, USA) gels con-taining ethidium bromide (1 Wg ml31).DSR genes were ampli¢ed with primer set DSR1F (5P-

ACSCACTGGAAGCACG-3P) and DSR4R (5P-GTGT-AGCAGTTACCGCA-3P) [28]. PCR conditions for DSRgene ampli¢cation consisted of 30 cycles of 94‡C for 1 min,54‡C for 1 min and 72‡C for 3 min. The reaction wascompleted by a ¢nal extension at 72‡C for 10 min. There-ampli¢cation of PCR products with 20 and 30 cycleswas performed with 1 Wl of the ampli¢cations as templateDNA to obtain DSR gene fragments. For the DSR am-pli¢cation, 1UEX Taq bu¡er (Takara), 1^100 ng of tem-plate DNA, 250 WM each deoxynucleoside triphosphates,25 pmol of each primer, 2.5 U of EX Taq DNA polymer-ase (Takara) were combined in a ¢nal 50-Wl volume. TheDSR gene fragments were analyzed by electrophoresis in1.1% w/v agarose S gels (Nippon Gene, Japan) containingethidium bromide (1 Wg ml31).

FEMSEC 1378 7-8-02

T. Nakagawa et al. / FEMS Microbiology Ecology 41 (2002) 199^209200

2.4. DGGE

DGGE was performed as described by Muyzer et al.[23,24] using the D-gene and D-code systems (Bio-RadLaboratories, Hercules, CA, USA) with a 1.5-mm gel.PCR products were applied directly onto 6% w/v polyac-rylamide gels with denaturing gradients from 20 to 60%(100% denaturant is 7 M urea and 40% v/v formamide).DGGE bands were excised from the gels, and reampli¢edusing primer sets 341F with GC-clamp, 907R, andARC344F with GC-clamp, ARC915R. After PCR, prod-ucts of the second ampli¢cation were electrophoresedagain in the denaturing gel to check the purity of thebands, the PCR ampli¢cations were puri¢ed with CON-CERT1 Rapid PCR puri¢cation kit (Invitrogen, Carls-bad, CA, USA).

2.5. Cloning and restriction digestion

PCR products (ca. 1.9 kb) encoding the DSR geneswere excised from the gels and extracted with QIAquick1Gel Extraction kit (Qiagen, Hilden, Germany). Puri¢edfragments were cloned with the vector pCR0-XL-TOPO0

and competent TOP10 cells (Invitrogen). From each li-brary, 8^36 colonies were randomly selected, and the in-serts were reampli¢ed with the vector primers M13 reverseand M13 forward (320). Portions (3 Wl) of PCR productscontaining the correct-size inserts were digested at 37‡Cfor 1 h with restriction endonuclease MspI according tothe manufacturer’s instructions (Takara). Speci¢c patternsof di¡erently sized DNA fragments were analyzed by elec-trophoresis in 3% w/v Nusieve 3:1 agarose gels (FMC,Rockland, ME, USA) containing ethidium bromide (1Wg ml31).

2.6. Sequencing and phylogenetic analysis

Nucleotide sequencing was performed with an ABIPrism BigDye Terminator Cycle Sequencing Ready Reac-tion Kit and an ABI model 377 automated sequencer (Ap-plied Biosystems) according to the manufacturer’s instruc-tions. 16S rDNA fragments from DGGE bands weresequenced using primers 341F and 907R. Partial sequences

of DSR ampli¢cation products were sequenced usingprimers DSR1F and an internal primer TNdsr1150r1 (5P-GTTRATGCAGTGCATGCA-3P) for clones. This num-ber corresponds to the DSR sequence of Desulfovibriovulgaris [29], and priming sites of DSR PCR productsre-ampli¢ed with the vector primers M13 reverse andM13 forward (320). The sequences from DGGE bandsand the deduced amino acid sequences of K-subunits ofDSR genes were entered into the BLAST programs [30]and FASTA programs [31] of the National Center forBiotechnology Information and the DNA Data Bank ofJapan (DDBJ) in order to identify phylogenetic relatives.Sequence alignments with parts from 16S rRNA and de-

Fig. 1. DGGE pro¢les derived from hydrothermal water (sites A, B,and C), and microbial mats collected from sites A, B, and D (Am, Bm,and Dm) with the primer sets for the domains Bacteria and Archaea.The bands labeled with numbers were sequenced. 1, TOYO-1-A; 2,TOYO-2-A; 3, TOYO-3-B; 4, TOYO-4-B; 5, TOYO-5-B; 6, TOYO-6-C; 7, TOYO-7-C; 8, TOYO-8-C; 9, TOYO-9-C; 10, TOYO-10-Am; 11,TOYO-11-Am; 12, TOYO-12-Bm; 13, TOYO-13-Bm; 14, TOYO-14-Bm; 15, TOYO-15-Bm; 16, TOYO-16-Dm; 17, TOYO-17-Dm; 18,TOYO-18-Am; 19, TOYO-19-Bm; 20, TOYO-20-Dm.

Table 1Environmental characteristics of sampling sites in the Toyoha Mine

Sample Temperature (‡C) pH SO234(mg l31)

Cl3

(mg l31)HCO3

3

(mg l31)Naþ

(mg l31)Mg2þ

(mg l31)Depth(m)

WaterSite A 64 6.6 1652.2 295.5 309.9 281.0 113.9 500Site B 71 6.3 239.8 1208.1 213.5 472.0 80.5 550Site C 48 6.8 85.8 712.7 385.6 478.0 27.3 550BiomatsAm 53 7.0 N.D. N.D. N.D. N.D. N.D. 500Bm 66 N.D. N.D. N.D. N.D. N.D. N.D. 550Dm 73 6.6 80.4 790.9 523.32 497.0 32.7 550

N.D. indicates not determined.

FEMSEC 1378 7-8-02

T. Nakagawa et al. / FEMS Microbiology Ecology 41 (2002) 199^209 201

duced DSR amino acid sequences of reference prokaryotesfrom the DDBJ/EMBL/GenBank with the CLUSTAL Wprogram [32], and matrices of evolutionary distance wereconstructed by the neighbor-joining method [33]. Phyloge-netic trees were constructed from the evolutionary distan-ces using Tree View software [34]. Bootstrap resamplinganalysis for 1000 replicates was performed to estimate thecon¢dence of tree topologies.

2.7. Chemical analysis

In situ temperature and pH of hydrothermal water were

measured by the electrode method using a pH meter(WM-22EP, TOA Electronics, Tokyo, Japan). Chemicaldata in December 2001 were provided by the ToyohaMine.

2.8. Nucleotide sequence accession numbers

The rRNA and DSR genes sequences were submittedto DDBJ/EMBL/GenBank and have been assigned thefollowing accession numbers: AB079462^AB079481(16S rRNA genes), and AB079482^AB079497 (DSRgenes).

Fig. 2. Phylogenetic relationships of bacterial 16S rDNA fragments (approximately 320 bp) (A) and archaeal 16S rDNA fragments (approximately 450bp) (B) as determined by neighbor-joining analysis. The scale bar represents an estimated 10% sequence divergence. Numbers beside branching pointsindicate bootstrap values determined from 1000 iterations.

FEMSEC 1378 7-8-02

T. Nakagawa et al. / FEMS Microbiology Ecology 41 (2002) 199^209202

3. Results

3.1. Environmental conditions at the Toyoha Mine

Table 1 shows the environmental characteristics of sam-pling sites at the Toyoha Mine. The in situ water temper-atures of the samplings ranged from 48 to 73‡C. The con-centration of SO234 at site A was much higher than atother sites. The concentrations of SO234 , Cl

3 and Mg2þ

at site C were similar to those at site D, due to the closedlocation at both sites (ca. 5 m).

3.2. DGGE analysis and phylogenetic analysis

DGGE analysis of 16S rDNA fragments shows the dif-ference in microbial community structure between hydro-thermal water samples, and between hydrothermal waterand microbial mats (Fig. 1). Bacterial bands were detectedin both hydrothermal water and microbial mats; however,no archaeal band was retrieved from hydrothermal water.As shown in Fig. 2A, phylogenetic analysis of the bandsshows that bands TOYO-1-A, -2-A, and -8-C were closelyrelated to Aqui¢cales clones derived from sulfur mats

Fig. 2 (Continued).

FEMSEC 1378 7-8-02

T. Nakagawa et al. / FEMS Microbiology Ecology 41 (2002) 199^209 203

[35,36]. Gram-positive bacteria bands, TOYO-3-B, -5-B,and TOYO-6-C were closely related to Desulfotomaculumkuznetsovii [4], and Syntrophothermus lipocalidus [37], re-spectively. TOYO-7-C showed the highest sequence simi-larity to ‘Ferribacter thermoautotrophicus’, an autotrophic,thermophilic, anaerobic dissimilatory Fe(III)-reducingbacterium isolated from a calcite spring in YellowstoneNational Park (published only in database). The bandsTOYO-14-Bm and TOYO-16-Dm belonged to the Ther-modesulfobacterium group, and the Thermus group, respec-tively. TOYO-9-C, -10-Am, -13-Bm and TOYO-4-B,-11-Am, -12-Bm belonged to L-Proteobacteria and N-Pro-teobacteria, respectively. TOYO-15-Bm was distant fromany known sequences with sequence similarities of lessthan 87%. Archaeal DGGE bands TOYO-18-Am, andTOY-19-Bm, -20-Dm were closely related to an unculturedcrenarchaeote clone SUBT-14 derived from a subterraneanhot spring in Iceland [38], and an uncultured crenar-chaeote clone NAB25 derived from a hot spring in Japan[39], respectively (Fig. 2B).

3.3. Occurrence of dissimilatory sul¢te reductase genes

To increase the sensitivity of sulfate-reducing prokary-otes detection in our samples, we used a set of DSR geneprimers. Genes encoding DSR (ca. 1.9 kb) were detectedin the ¢ltrated hydrothermal water at sites A^C and themicrobial mats Bm at site B and Dm at site D (Fig. 3).DSR PCR products from sites A and B, and those fromsite C were obtained after re-PCR with 20 and 30 cycles,respectively. No DSR PCR product was detected from theDNA sample obtained within the microbial mats Am atsite A.

3.4. Clone library characterization

A clone library of the PCR products of DSR gene frag-ments was used to investigate the diversity of DSR genesof the sulfate-reducing microorganisms within hydrother-mal water and microbial mats (Fig. 4). A total of 123

clones were assembled into 16 clone families based onMspI restriction fragment banding patterns.

3.5. Phylogenetic analysis of DSR genes

Fig. 5 shows the phylogenetic relationships between thededuced amino acid sequences of K-subunits of DSR genefragments of clone families and that of reference cultures.Three di¡erent branches of DSR sequences (branchesI^III) were found in the Toyoha samples. Branch I wasdivided into three sub-branches, I-1 to I-3 (Fig. 5A). Clonefamilies TOYOdsr-7 and -10 within branch I-1 showed thehighest nucleotide sequence similarity to Desulfotomacu-lum kuznetsovii, a thermophilic, Gram-positive SRB [4],with 96% and 92% similarity. However, the similaritiesof clone sequences within branches I-2 and I-3 to D. kuz-netsovii were 83^84% and 76^77%, respectively. TOYOdsr-2 to -6 (branch II) were related to Thermodesulforhabdusnorvegicus, a thermophilic, Gram-negative SRB which wasisolated from a North Sea oil ¢eld [3], with 80% nucleotide

Fig. 3. Agarose gel electrophoresis of dissimilatory sul¢te reductase(DSR) gene fragments (approximately 1.9 kb) ampli¢ed with the DSRgene primer set, DSR1F and DSR4R. DSR PCR products at sites Aand B were obtained with additional PCR cycles (20) after 30 cycles ofPCR. DSR PCR products at site C were obtained with additional PCRcycles (30) after 30 cycles of PCR.

Fig. 4. MspI restriction patterns of the 16 clone families identi¢ed in the clone library of DSR gene fragments. DSR gene clone family 1, TOYOdsr-1;2, TOYOdsr-2; 3, TOYOdsr-3; 4, TOYOdsr-4 ; 5, TOYOdsr-5; 6, TOYOdsr-6; 7, TOYOdsr-7 ; 8, TOYOdsr-8; 9, TOYOdsr-9; 10, TOYOdsr-10; 11,TOYOdsr-11; 12, TOYOdsr-12; 13, TOYOdsr-13; 14, TOYOdsr-14; 15, TOYOdsr-15; 16, TOYOdsr-16.

FEMSEC 1378 7-8-02

T. Nakagawa et al. / FEMS Microbiology Ecology 41 (2002) 199^209204

sequence similarity. TOYOdsr-1 was less closely related toknown species, the closest relative being Desulfotomaculumalkaliphilum [40] with 68% amino acid sequence similarity.As shown in Fig. 6, TOYOdsr-1 had a major deletionbetween positions 259 and 273 of K-subunits of the DSRgene (numbered according to Desulfovibrio vulgaris [29]) aswell as archaeal sulfate reducers and Gram-positive SRBas previously described by Klein et al. [41].

3.6. Comparison of abundance of DSR gene branchesbetween hydrothermal water and microbial mats

Table 2 shows a comparison of the frequency ratios ofdi¡erent DSR gene branches (Fig. 5) from hydrothermalwater and microbial mats at distinct sites within the Toyo-ha Mine. Branch I showed a higher frequency from hydro-thermal water at site A (83%) and site B (76%). On the

Fig. 5. Phylogenetic relationships among the 15 clone families of DSR genes and the K-subunits of DSR genes of the known cultures within N-SRB andxenologous SRB [18] (A), and among the one clone family of DSR genes and the K-subunits of DSR genes of the known cultures within Thermodesul-fovibrio, Archaeoglobus, authentic Gram-positive SRB [18], and the clones (c63 and b31) retrieved from a marine sediments with anaerobic methane oxi-dation [38] (B) as determined by neighbor-joining analysis. The scale bar represents an estimated 10% sequence divergence. Numbers beside branchingpoints indicate bootstrap values determined from 1000 iterations.

FEMSEC 1378 7-8-02

T. Nakagawa et al. / FEMS Microbiology Ecology 41 (2002) 199^209 205

other hand, branch II showed a higher frequency frommicrobial mats at site Bm (79%) and site Dm (65%) com-pared to hydrothermal water at site A (0%) and site B(20%). Branch III was only detected from hydrothermalwater at site A and B.

4. Discussion

4.1. Distribution and diversity of Desulfotomaculum andThermodesulfobacterium-like sequences

In contrast to non-spore forming SRB, spore-formingsulfate reducers, e.g., Desulfotomaculum species, are con-sidered to be robust and widely distributed in environ-

ments due to the durability of spores relative to dryness,aerobic conditions, and heat habitats [6]. Ishii et al. [21]retrieved several 16S rDNA sequences related to both me-sophilic and thermophilic Desulfotomaculum species fromthe cold-deep groundwater (11^15‡C) at the KamaishiMine, Japan. In contrast, TOYO-3-B and -5-B, 16SrDNA PCR products, and TOYOdsr-7^16, the K-subunitsof DSR PCR products, retrieved from hydrothermal waterwere related to Desulfotomaculum kuznetsovii, a thermo-philic Gram-positive sulfate reducer [4]. Presumably, theSRB belonged to DSR branch I-1^3 (Fig. 5) might be thespore-forming, thermophilic SRB distributed widely in hy-drothermal water within the cracks and boreholes of theToyoha Mine.Hydrogen is an important electron and energy source

Fig. 5 (Continued).

Table 2Abundance of clone families in the library of DSR gene fragments retrieved from hydrothermal water and microbial mats within the Toyoha Mine

Toyoha DSR branch Clone family Number of clones (% of clones at each site)

Hydrothermal water Biomats

Site A Site B Site C Site Bm Site Dm

Branch I-1 TOYOdsr-7 3 (14)TOYOdsr-10 1 (5)

Branch I-2 TOYOdsr-8 12 (55) 7 (23) 1 (13) 5 (17)TOYOdsr-9 2 (9) 5 (17)TOYOdsr-11 2 (7)TOYOdsr-12 9 (30) 3 (9)TOYOdsr-14 1 (3)

Branch I-3 TOYOdsr-13 4 (13)TOYOdsr-15 2 (6)TOYOdsr-16 2 (6)

Branch II TOYOdsr-2 6 (20) 5 (63) 26 (76) 13 (45)TOYOdsr-3 2 (25) 1 (3)TOYOdsr-4 4 (14)TOYOdsr-5 1 (3)TOYOdsr-6 1 (3)

Branch III TOYOdsr-1 4 (18) 1 (3)

FEMSEC 1378 7-8-02

T. Nakagawa et al. / FEMS Microbiology Ecology 41 (2002) 199^209206

for deep subsurface microbial life [42]. It seems likely thathydrothermal water contained hydrogen because of thepresence of DGGE band TOYO-13-Bm, which is closelyrelated to Hydrogenophilus thermoluteolus, a thermophilic,facultatively chemolithoautotrophic, hydrogen-oxidizingbacterium (Fig. 2A) [43]. Desulfotomaculum kuznetsovii[4] and Desulfotomaculum thermocisternum [5] can growautotrophically with H2 as electron donors by sulfate re-duction. This function is probably essential for the SRB tosurvive and distribute in deep-rock aquifers besides thephenotype of high optimum temperature and spore forma-tion.We successfully obtained the DSR gene fragments (DSR

branch II) related to Thermodesulforhabdus norvegicusfrom microbial mats Bm (66‡C) and Dm (73‡C) withinthe mine environment (Table 2). The temperature rangeof T. norvegicus for growth is 44^74‡C [3]. Thus, DSRBranch II may possibly be a thermophilic, Gram-negativeSRB. As shown in Figs. 1 and 2A, microbial mats Bm andDm contained species of Thermus (TOYO-16-Dm) andGram-positive bacteria (TOYO-17-Dm). These speciesseem to supply the low molecular mass of organic matterto the thermophilic SRB within DSR branch II.DSR primers made it possible for us to access the di-

versity of thermophilic SRB and the potential sul¢de pro-duction through dissimilatory sulfate reduction within amicrobial community containing several non-SRB at theToyoha Mine. However, we could not retrieve any DSRgene related to Thermodesulfobacterium commune andThermodesulfobacterium thermophilum in spite of the ap-pearance of a DGGE band (TOYO-14-Bm) belonging tothe Thermodesulfobacterium group at site Bm (Fig. 1). In

addition, Desulfotomaculum-like DSR gene sequences weredetected from the same site. These results indicate thedi⁄culty of assuming the biological sul¢de productionby SRB using only phylogenetic information based on16S rDNA.DGGE and cloning are based on PCR ampli¢cation of

16S rRNA and the DSR gene, respectively. Several pitfallsand potential biases of PCR-dependent techniques havebeen previously described [44], and PCR biases might oc-cur in the present study due to the re-ampli¢cation of the¢rst PCR products from hydrothermal water, i.e., the am-pli¢cation with a primer set for the DSR gene that doesnot target all SRB [21]. Hence, it is necessary to study inmore detail the abundance of thermophilic SRB withinhigh-temperature hydrothermal environments through hy-bridization techniques, and to develop speci¢c probes forthermophilic microbes.

4.2. Biological sul¢de production

DSR genes related to known thermophilic SRB weredetected together with decomposers such as TOYO-16-Dm and TOYO-17-Dm from the microbial mats. The pro-vision of low molecular organic carbons to sulfate reduc-ers is responsible for the high rates of dissimilatory sulfatereduction. Hence, thermophilic SRB possibly producedsul¢de through sulfate respiration within microbial matsBm and Dm. Oxidation of organic carbon with dissimila-tory sulfate reduction by sulfate reducers caused 34S de-pletion of sul¢des (FeS and FeS2) such as the sul¢deswithin the microbial mat containing SRB in Solar Lake[20]. Indeed, ¢ne-grained pyrites within microbial matsDm of the Toyoha Mine were depleted in 34S by 0^11x, compared to the isotope composition of hydrother-mal water sulfate (ca. 9^10x) [19]. However, relativelylittle is known about sul¢de formation via the action ofthermophilic sulfate-reducing prokaryotes in high-temper-ature mine habitats. Hence, it is essential to investigate thesulfate reduction rate of thermophilic sulfate-reducing mi-crobes with 35SO234 , combined with quantitative data ofSRB and/or SRA for a better understanding of in situbiological sul¢de production through dissimilatory sulfatereduction.

4.3. Conclusions

Our study with DSR primers suggests that the distribu-tion and diversity of spore-forming (Desulfotomaculum-like DSR sequences) and non-spore-forming (Thermode-sulforhabdus-like DSR sequences) thermophilic SRB with-in e¥uent-hydrothermal groundwater and microbial matsin high-temperature mine environments. Moreover, a nov-el DSR gene obtained from this study indicates the pres-ence of thermophilic sulfate-reducing prokaryotes, whichwere neither found nor cultured, in hydrothermal waterwithin high-temperature mine habitats. Further study, as

Fig. 6. Deduced amino acid alignment of K-subunits of DSR genesshowing that TOYOdsr-1 has a deletion between positions 259 and 273of K-subunits of DSR genes. Amino acid positions correspond to the K-subunit of the DSR gene of Desulfovibrio vulgaris [17].

FEMSEC 1378 7-8-02

T. Nakagawa et al. / FEMS Microbiology Ecology 41 (2002) 199^209 207

well as developing tools for quanti¢cation of the thermo-philic microbes combined with physiological characteriza-tion (sul¢de production and sulfate reduction rate), shouldprovide a more comprehensive understanding of thermo-philic sulfate-reducing prokaryotes within e¥uent-hydro-thermal groundwater and microbial mats developed in thethermophilic subsurface environment of the Toyoha Mine.

Acknowledgements

We are indebted to Dr. T. Kakegawa for critical readingof the manuscript, and Toyoha Mines Co., Ltd. for thecontribution of samples and information about mine con-ditions. This work was supported by a grant from theMinistry of Education, Culture, Sports, Science and Tech-nology through the Special Coordination Fund ‘ArchaeanPark Project’ : International Research Project on Interac-tion Between Sub-Vent Biosphere and Geo-Environments.

References

[1] Berner, R.A. (1985) Sedimentary pyrite formation: An update. Geo-chim. Cosmochim. Acta 48, 605^615.

[3] Beeder, J., Torsvik, T. and Lien, T. (1995) Thermodesulforhabdusnorvegicus gen. nov., sp. nov., a novel thermophilic sulfate-reducingbacterium from oil ¢eld water. Arch. Miclobiol. 164, 331^336.

[4] Nazina, T.N., Ivanova, A.E., Kanchaveli, L.P. and Rozanova, E.P.(1988) Desulfotomaculum kuznetsovii sp. nov., a new spore-forming,thermophilic, methylotrophic, sulfate-reducing bacterium. Mikrobio-logija 57, 659^663.

[5] Nilsen, R.K., Torsvik, T. and Lien, T. (1996) Desulfotomaculum ther-mocisternum sp. nov., a sulfate reducer isolated from a hot North Seaoil reservoir. Int. J. Syst. Bacteriol. 46, 397^402.

[6] Widdel, F. (1992) The genus Desulfotomaculum. In: The Prokaryotes,2nd edn. (Balows, A., Tru«per, H.G., Dworkin, M., Harder, W. andSchleifer, K.-H., Eds.), pp. 1793^1799. Springer Verlag, New York.

[7] Sonne-Hansen, J. and Ahring, B.K. (1999) Thermodesulfobacteriumhveragerdense sp. nov. and Thermodesulfovibrio islandicus sp. nov.,two thermophilic sulfate reducing bacteria isolated from a Icelandichot spring. Syst. Appl. Microbiol. 22, 559^564.

[9] Widdel, F. (1992) The genus Thermodesulfobacterium. In: The Pro-karyotes, 2nd edn. (Balows, A., Tru«per, H.G., Dworkin, M., Harder,W. and Schleifer, K.-H., Eds.), pp. 3390^3392. Springer Verlag, NewYork.

[10] Zeikus, J.G., Dawson, M.A., Thompson, T.E., Ingvorsen, K. andHatchikian, E.C. (1983) Microbial ecology of volcanic sulphidogen-esis : Isolation and characterization of Thermodesulfobacteruim com-mune gen. nov. and sp. nov. J. Gen. Microbiol. 129, 1159^1169.

[11] Henry, E.A., Devereux, R., Maki, J.S., Gilmour, C.C., Woese, C.R.,Mandelco, L., Schauder, R., Remsen, C.C. and Mitchell, R. (1994)Characterization of a new thermophilic sulfate-reducing bacteriumThermodesulfovibrio yellowstonii, gen. nov. and sp. nov. its phyloge-netic relationship to Thermodesulfobacterium commune and their ori-gins deep within the bacterial domain. Arch. Microbiol. 161, 62^69.

[12] Stetter, K.O. (1992) The genus Archaeoglobus. In: The Prokaryotes,2nd edn. (Balows, A., Tru«per, H.G., Dworkin, M., Harder, W. andSchleifer, K.-H., Eds.), pp. 707^711. Springer Verlag, New York.

[13] Joulian, C., Ramsing, N.B. and Ingvorsen, K. (2001) Congruentphylogenies of most common small-subunit rRNA and dissimilatory

sul¢te reductase gene sequences retrieved from estuarine sediments.Appl. Environ. Microbiol. 67, 3314^3318.

[14] Thomsen, T.R., Finster, K. and Ramsing, N.B. (2001) Biogeochem-ical and molecular signatures of anaerobic methane oxidation in amarine sediment. Appl. Environ. Microbiol. 67, 1646^1656.

[15] Minz, D., Flax, J.L., Green, S.J., Muyzer, G., Cohen, Y., Wagner,M., Rittmann, B.E. and Stahl, D.A. (1999) Diversity of sulfate-re-ducing bacteria in oxic and anoxic regions of a microbial mat char-acterized by comparative analysis of dissimilatory sul¢te reductasegenes. Appl. Environ. Microbiol. 65, 4666^4671.

[17] Cottrell, M.T. and Craig Cary, S. (1999) Diversity of dissimilatorybisul¢te reductase genes of bacteria associated with the deep-sea hy-drothermal vent polychaete annelid Alvinella pompejana. Appl. Envi-ron. Microbiol. 65, 1127^1132.

[18] Chang, Y.-J., Peacock, A.D., Long, P.E., Stephen, J.R., McKinley,J.P., Macnaughton, S.J., Anwar Hussain, A.K.M., Saxton, A.M. andWhite, D.C. (2001) Diversity and characterization of sulfate-reducingbacteria in groundwater at a uranium mill tailings site. Appl. Envi-ron. Microbiol. 67, 3149^3160.

[19] Kakegawa, T. (2001) Biological and non-biological sul¢de precipita-tion from hydrothermal £uids at the Toyoha mine, Hokkaido. TheJoint Meeting of Japan Earth and Planetary Science. Cm-011.

[20] Habicht, K.S. and Can¢eld, D.E. (1996) Sulphur isotope fractiona-tion in modern microbial mats and the evolution of the sulphur cycle.Nature 382, 342^343.

[21] Ishii, K., Takii, S., Fukunaga, S. and Aoki, K. (2000) Characteriza-tion by denaturing gradient gel electrophoresis of bacterial commun-ities in deep groundwater at the Kamaishi Mine, Japan. J. Gen. Appl.Microbiol. 46, 85^93.

[22] Purdy, K.J., Embley, T.M., Takii, S. and Nedwell, D.B. (1996) Rapidextraction of DNA and rRNA from sediments by a novel hydroxy-apatite spin-column method. Appl. Environ. Microbial. 62, 3905^3907.

[23] Muyzer, G., De Waal, E.C. and Uitterlinden, A.G. (1993) Pro¢ling ofcomplex microbial populations by denaturing gradient gel electro-phoresis analysis of polymerase chain reaction-ampli¢ed gene codingfor 16S rRNA. Appl. Environ. Microbiol. 59, 695^700.

[24] Muyzer, G., Teske, A., Wirsen, C.O. and Jannasch, H.W. (1995)Phylogenetic relationships of Thiomicrospira species and their identi-¢cation in deep-sea hydrothermal vent samples by denaturing gra-dient gel electrophoresis of 16S rDNA fragments. Arch. Microbiol.164, 165^172.

[25] Raskin, L., Stromley, J.M., Rittmann, B.E. and Stahl, D.A. (1994)Group-speci¢c 16S rRNA hybridization probes to describe naturalcommunities of methanogens. Appl. Environ. Microbiol. 60, 1232^1240.

[26] Stahl, D.A. and Amann, R.I. (1991) Development and application ofnucleic acid probes. In: Nucleic Acid Techniques in Bacterial System-atics (Stackebrant, E. and Goodfellow, M. Eds.), pp. 205^248. Wiley,New York.

[27] Casamayor, E.O., Scha«fer, H., Ban‹eras, L., Pedro¤s-Alio¤ , C. andMuyzer, G. (2000) Identi¢cation of and spatio-temporal di¡erencebetween microbial assemblages from two neighboring sulfur lakes:comparison by microscopy and denaturing gradient gel electrophore-sis. Appl. Environ. Microbiol. 66, 499^508.

[28] Wagner, M., Roger, A., Flax, J., Brusseau, G. and Stahl, D. (1998)Phylogeny of dissimilatory sul¢te reductases supports an early originof sulfate respiration. J. Bacteriol. 180, 2975^2982.

[29] Karkho¡-Schweizer, R.R., Huber, D.P. and Voordouw, G. (1995)Conservation of the genes for dissimilatory sul¢te reductase fromDesulfovibrio vulgaris and Archaeoglobus fulgidus allows their detec-tion by PCR. Appl. Environ. Microbiol. 61, 290^296.

[30] Altschul, S.F., Madden, T.L., Scha«fer, A.A., Zhang, J., Zhang, Z.,Miller, W. and Lipman, D.J. (1997) Gapped BLAST and PSI-BLAST: A new generation of protein database search programs.Nucleic Acids Res. 25, 3389^3402.

FEMSEC 1378 7-8-02

T. Nakagawa et al. / FEMS Microbiology Ecology 41 (2002) 199^209208

[31] Lipman, D.J. and Pearson, W.R. (1985) Rapid and sensitive proteinsimilarity searches. Science 227, 1435^1441.

[32] Thompson, J.D., Higgins, D.G. and Gibson, T.J. (1994) CLUSTALW: Improving the sensitivity of progressive multiple sequence align-ment through sequence weighting, position-speci¢c gap penalties andweight matrix choice. Nucleic Acids Res. 22, 4673^4680.

[33] Saito, N. and Nei, M. (1987) The neighbor-joining method: a newmethod for constructing phylogenetic trees.Mol. Biol. Evol. 4, 406^425.

[34] Page, R.D.M. (1996) TREEVIEW: An application to display phylo-genetic trees on personal computers. CABIOS 12, 357^358.

[35] Skirnisdottir, S., Hreggvidsson, G.O., Hjo«rleifsdottir, S., Marteins-son, V.T., Retursdottir, S.K., Holst, O. and Kristjansson, J.K.(2000) In£uence of sul¢de and temperature on species compositionand community structure of hot spring microbial mats. Appl. Envi-ron. Microbiol. 66, 2835^2841.

[36] Yamamoto, H., Hiraishi, A., Kato, K., Chiura, H.X., Maki, Y. andShimizu, A. (1998) Phylogenetic evidence for the existence of novelthermophilic bacteria in hot spring sulfur-turf microbial mats in Ja-pan. Appl. Environ. Microbiol. 44, 844^851.

[37] Sekiguchi, Y., Kamagata, Y., Nakamura, K., Ohashi, A. and Hara-da, H. (2000) Syntrophothermus lipocalidus gen. nov., sp. nov., anovel thermophilic, syntrophic, fatty-acid-oxidizing anaerobe whichutilizes isobutyrate. Int. J. Syst. Evol. Microbiol. 50, 771^779.

[38] Marteinsson, V.T., Hauksdottir, S., Hobel, C.F.V., Kristmannsdottir,

H., Hreggvidsson, G.O. and Kristjansson, J.K. (2001) Phylogeneticdiversity analysis of subterranean hot springs in Iceland. Appl. En-viron. Microbiol. 67, 4242^4248.

[39] Nakagawa, T. and Fukui, M. (2002) Phylogenetic characterization ofmicrobial mats and streamers in a Japanese alkaline hot spring with athermal gradient. Submitted.

[40] Friedrich, M.W. (2002) Phylogenetic analysis reveals multiple lateraltransfers of adenosine-5P-phosphosulfate reductase genes among sul-fate-reducing microorganisms. J. Bacteriol. 184, 278^289.

[41] Klein, M., Friedrich, M., Roger, A.J., Hugenholtz, P., Fishbain, S.,Abicht, H., Blackall, L.L., Stahl, D.A. and Wagner, M. (2001) Multi-ple lateral transfers of dissimilatory sul¢te reductase genes betweenmajor lineages of sulfate-reducing prokaryotes. J. Bacteriol. 183,6028^6035.

[42] Pedersen, K. (2000) Exploration of deep intraterrestrial microbiallife : current perspectives. FEMS Microbiol. Lett. 185, 9^16.

[43] Hayashi, N.R., Ishida, T., Yokota, A., Kodama, T. and Igarashi, Y.(1999) Hydrogenophilus thermoluteolus gen. nov., sp. nov., a thermo-philic, facultatively chemolithoautotrophic, hydrogen-oxidizing bacte-rium. Int. J. Syst. Bacteriol. 49, 783^786.

[44] Von Wintzingerode, F., Go«bel, U.B. and Stackebrandt, E. (1997)Determination of microbial diversity in environmental samples: pit-falls of PCR-based rRNA analysis. FEMS Microbiol. Rev. 21, 213^229.

FEMSEC 1378 7-8-02

T. Nakagawa et al. / FEMS Microbiology Ecology 41 (2002) 199^209 209