The ISME Journal (2010) 4, 1 0 3 1 - 1 0 4 3© 2 0 1 0 International Society for Microbial Ecology All rights reserved 1 7 5 1 -7 3 6 2 /1 0 $ 3 2 .00www.nature.com/ismej

ORIGINAL ARTICLENovel observations of Thiobaa sulfur-storing Gammaproteobacterium producing gelatinous matsStefanie Grünke1,2, Anna Lichtschlag2, Dirk de Beer2, Marcel Kuypers2,Tina Lösekann-Behrens3, Alban Ramette2 and Antje Boetius1,21HGF-MPG Joint Research Group fo r Deep Sea Ecology and Technology A lfred Wegener Institu te fo r Polar a nd M arine Research, Rrem erhaven, Germany; 2M ax P lanck Institu te fo r M arine M icrobiology Rremen, G erm any and 3D epartm ent o f M icrobiology and Im m uno logy Stanford U niversity Stanford, CA, USA

The genus Thiobacterium includes uncultivated rod-shaped microbes containing several spherical grains of elemental sulfur and forming conspicuous gelatinous mats. Owing to the fragility of mats and cells, their 16S ribosomal RNA genes have not been phylogenetically classified. This study examined the occurrence of Thiobacterium mats in three different sulfidic marine habitats: a submerged whale bone, deep-water seafloor and a submarine cave. All three mats contained massive amounts of Thiobacterium cells and were highly enriched in sulfur. Microsensor measurements and other biogeochemistry data suggest chemoautotrophic growth of Thiobacterium. Sulfide and oxygen microprofiles confirmed the dependence of Thiobacterium on hydrogen sulfide as energy source. Fluorescence in situ hybridization indicated that Thiobacterium spp. belong to the Gammaproteobacteria, a class that harbors many mat-forming sulfide-oxidizing bacteria. Further phylogenetic characterization of the mats led to the discovery of an unexpected microbial diversity associated with Thiobacterium.The ISME Journal (2010) 4, 1031-1043; doi:10.1038/ismej.2010.23; published online 11 March 2010 Subject Category: geomlcroblology and microbial contributions to geochemical cycles Keywords: gelatinous mats; microsensor; sulfur oxidizer; Thiobacterium

IntroductionThe genus Thiobacterium was first described by M olisch in 1912 (M olisch, 1912). In the years after this discovery, the conspicuous sulfur-storing, non-m otile rods em bedded in a gelatinous m atrix w ere found in other m arine and continental loca­tions w orldw ide. They occur in therm al and sulfur springs (Dévidé, 1952; Lackey and Lackey, 1961; Lackey et al., 1965; Vouk et al., 1967; A nagnostidis, 1968; Schem inzky et al., 1972; Fjerdingstad, 1979), bu t also in sulfid ic m arine and brackish waters (M olisch, 1912; Lackey et al., 1965; Seki and N aganum a, 1989). So far two different m orphologies of the gelatinous mats have been described: a spherical or bladder-like shape (M olisch, 1912; Dévidé, 1952; Vouk et al., 1967; A nagnostidis, 1968; Schem inzky et al., 1972; Seki and Naganum a, 1989) and a dendro id shape (Lackey and Lackey, 1961; Lackey et a l , 1965). F irst experim ents on the chem ical and elem ental com position of the

C orrespondence: S Grünke, M ax Planck Institu te for M arine M icrobiology, C elsiusstrasse 1, 28359 Bremen, Germany.E-mail: sgruenke@ m pi-brem en.deR eceived 10 Decem ber 2009; revised 5 February 2010; accepted 6 February 2010; pub lish ed online 11 M arch 20Í0

gelatinous m aterial led to the assum ption that it consists of an a llo p hane-su lfu r-hydrogel (Vouk et ah, 1967; allophane is an am orphous hydrous alum inum silicate). In contrast to the varying shapes and sizes of the gelatinous m aterial, the m orphology of the rod-shaped m icrobes was very consisten t in m ost studies. An experim ent targeting the physiology of Thiobacterium suggested that u nder aerobic conditions these organism s m ay express a euryther- m ally m esophilic and slightly haloph ilic behavior (Seki and Naganum a, 1989). M ost im portantly, the enrichm ent study ind icated that Thiobacterium cells are them selves forming the gelatinous m asses, m ost likely to retain a favorable spatial position in their hab itat w ith access to the sulfide and oxygen sources.

In sp ite of all past observations, know ledge on the genus Thiobacterium is still ra ther poor, because none of its m em bers have been cultivated. By m orphological analogy and ecological context, it was associated w ith the family T hiotrichaceae of the G am m aproteobacteria in Bergey’s m anual of system atic bacteriology (Kuenen, 2005) and the E ncyclopedia of Life (http ://w w w .eol.org/pages/ 97513). Its classification as one genus Thiobacterium is based on consisten t m orphological observations of rod-shaped cells w ith chain-like inclusions of up to

Thiobacterium matsS G rünke e f a l

103220 spherical sulfur granules (Lackey and Lackey, 1961), em bedded in gelatinous m atter (Kuenen,2005). However, no 16S ribosom al RNA (rRNA) gene sequence has yet been attributed to a Thiobac- terium sp., and the taxonom ic positioning of this genus w ith in the group of sulfur bacteria is uncertain . Furtherm ore, the energy sources, n iche- selection and ecological role of Thiobacterium rem ain unknow n. The m ain problem s in the investigation of these m at-form ing bacteria are their rarity and the fragility of m ats, colonies and cells (M olisch, 1912; Lackey and Lackey, 1961; Vouk et ah, 1967; Seki and Naganum a, 1989).

This study investigated Thiobacterium m ats from three sulfid ic m arine habitats, includ ing a m inke w hale bone collected offshore Sw eden, deep-sea sedim ents from the Storegga Seeps off N orw ay and a shallow -w ater cave in Greece. To broaden our know ledge on the genus Thiobacterium and its ecological role, (i) the geochem ical gradients w ith in the gelatinous m asses were analyzed, (ii) the cells were m icroscopically and phylogenetically charac­terized, and (iii) the overall m icrobial com m unity com position associated w ith the gelatinous mats was exam ined.

Materials and methodsSite description and sample collection All sam pling locations and prevailing environ­m ental conditions are described in Table 1. W hale bones were recovered from the carcass of a female m inke w hale that was previously im plan ted at 125 m depth in Kosterfjord, Sw eden (58°53.1'N, 11°6.4'E; Glover et ah, 2005; Dahlgren et ah, 2006). Since their recovery, the bones had been m ain tained at 7 -8 °C in aquaria flushed w ith filtered seawater (Glover et ah, 2005). Sam pling of a sm all spherical Thiobacterium m at was achieved by pipetting.

Deep-sea Thiobacterium m ats w ere obtained from the Storegga area off Norway during the ‘VICKING’ expedition w ith the RV Pourquoi Pas? and the ROV Victor 60 0 0 (IFREMER) in June-Ju ly 2006 from a sm all seep structure covered w ith a w h itish mat (Dive 275-05; core CT-02; 64°45.27'N, 4°58.87'E). Aboard the ship, the Thiobacterium -containing core was im m ediately transferred to a cold room. Sam pling occurred directly after the dive through m echanical d isrup tion w ith a pipette.

The gelatinous m ats of the shallow -w ater cave (‘Blue Pot Cave’; 39°10.66'N, 20°12.54'E) off Paxos (Greece) were first discovered by Paul Bowbeer (Oasi-Sub-Diving) and Dr Thom as Beer. In itial sam ples of the spheres, native and fixed in 4% formaldehyde, reached the MPI Bremen in September 2006. A second sam pling in Paxos took place in A ugust 2007. The partia lly sun-lit cave is located at approxim ately 23 m depth. W hole gelatinous spheres and cave w ater w ere sam pled into sterile tubes and transported in a cool cham ber to the

on-site laboratory. Subsam pling of the gelatinous m ats was carried out either by preserving w hole spheres for biogeochem ical analyses, or by d issect­ing single spheres w ith a sterile scalpel for m icro­scopic, phylogenetic and fluorescence in situ hybrid ization (FISH) analyses.

MicroscopySubsam ples of the sam e gelatinous mats sam pled for DNA extraction w ere analyzed directly after sam ­pling by bright field and phase contrast microscopy. V isualization of subsam ples stained w ith different fluorochrom es was achieved by using a Zeiss LSM 510 and the appropriate software (Cari Zeiss Micro- Imaging GmbH, Göttingen, Germany).

StainingSubsam ples of the gelatinous mats w ere preserved in either 2% or 4% form aldehyde/seaw ater at room tem perature, 4 °C or —20 °C. W hen applying the protein-targeting fluorochrom e SYPRO Red (Molecular Probes, Invitrogen Corporation, Karlsruhe, Germany), staining was conducted at room tem pera­ture for at least 4 5 m in (4 h m axim um ). The dye was d ilu ted beforehand in artificial seaw ater (salinity 34%o) to a 5 X concentrated solution. Slides were directly subjected to m icroscopy. The DNA-targeting fluorescent stain 4 ',6-diam idino-2-phenylindole (DAPI) was also app lied to fixed m aterial of the gelatinous spheres. Subsam ples of the structures were placed onto ind iv idua l spots of Teflon-coated slides and were dried at 46 °C for 1 h. S taining was conducted for 7 m in at room tem perature w ith 15 pi of a 2 .5 p g m U 1 DAPI solution. To remove excess DAPI, the preparations were shortly rinsed w ith d istilled water, follow ed by rinsing in 96% ethanol and subsequent drying at room tem perature. Pre­parations were finally m ounted in a 2:3 m ix of Vecta Shield (Vector Laboratories Inc., Burlingam e, CA, USA) and C itifluor (Agar Scientific Ltd, Essex, UK), stored at —20 °C and subjected to m icroscopy the following day.

Microsensor measurementsH igh-resolution geochem ical gradients were m ea­sured on a Thiobacterium m at during the ‘VICKING’ expedition in 2006 (IFREMER) w ith a laboratory set up. M icrosensors for pH, 0 2 and H2S were used, and sensors were calibrated as previously described (Revsbech and Ward, 1983; Jeroschew ski et ah, 1996; de Beer et al., 1997). The total sulfide (H2S + HS~ + S2~) was calculated from the H 2S concentrations and the local pH using equilibrium constants. M icrosensors were m ounted on a motor- driven m icrom anipulator and data acquisition was perform ed using a DAQ-PAD 6015 (National Instru ­ments Corporation, Austin, TX, USA) and a computer. Relative to the m icrosensor tips, the surface of

The ISME Journal

Thiobacterium matsS Grünke e t al

1033T ab le 1 Physicochemical characteristics of habitats in w hich Thiobacterium spp.-resembling microbes and gelatinous mats occurred

Locationa H abitat H 2Sh p H Temp. S a lin ity D eptlT R eference

Trieste [Italy) M arine w ater A ssum ed _ _ _ l m M olisch [1912)D ubrovnik [Croatia) Sulfur springd Detected — — — l m Dévidé [1952)Venice [Florida) Sulfur springd ~ 5 pM 7.2 29 C 17% l m Lackey and Lackey [1961);

Lackey e t al. [1965)T itusville [Florida) P ollu ted w aterd — — 12.5 C 17% Shallow Lackey and Lackey [1961);

Lackey e t al. [1965)C edar Keys [Florida) Tide poolsd — — — 17% l m Lackey e t al. [1965)Bad Gastein [Austria) T herm al spring ~ 15—23 pM 8.7-9.3 43.5-45.5 C — 0 m Vouk et al. [1967)Am élie-Les-Bains T herm al spring ~ 103pM e 9.1e 22 G :45 G — 0 m Schem inzky e t al. [1972)[France)Greece H ot springs Detected — — — 0 m A nagnostid is [1968)N orth Fiji B asin1 H ydrotherm al vent — 7.5 — 27.8% 2671m Seki and N aganum a [1989)Tjärnö [Sweden) W hale bone Detected — 7-8 C® 34.3-34.7% s aq This studyStoregga Seeps M arine Detected 7.8—7.9 - 1 C 34.1-35.5% 745 m This study[Norway) sed im entPaxos [Greece) M arine cave b.d .h CO 0 1 CO to 15-20 C 39% 23m This study

in c lu d es observations on Thiobacterium that could clearly be attributed to the genus under investigation. Publications providing only scarce morphological information (Miyoshi, 1897; Caldwell and Caldwell, 1974; Gugliandolo and Maugeri, 1993; Yang et al., 2008), or describing microbes that appear morphologically deviant from Thiobacterium (Miyoshi, 1897; Naganuma et al., 1990; Hedoin et al., 1996; Mo s so et al., 2002) are not listed.bHydrogen sulfide was occasionally detected only qualitatively, or its presence was m entioned w ithout presenting data on measured concentrations. Hydrogen sulfide concentrations were estimated for Venice (Florida) from 0.162 p.p.m. H2S (Lackey and Lackey, 1961), for Bad Gastein (Austria) from 0.5-0.8 mg kg-1 H2S (Vouk et al., 1967) and for Amélie-Les-Bains (France) from 3.5 mg I-1 H2S (Scheminzky et aí., 1972). cIf no explicit value for depth was given, this is now indicated by a substitute value of 1 m. A value of 0 m resulted from the observation of gelatinous structures directly on rock surface overflown by water from thermal springs (Vouk et al., 1967; Scheminzky et al., 1972), or where structures were found on the water surface (Anagnostidis, 1968). The minke whale bones from Tjärnö were kept in an aquarium, w hich is indicated by aq.dMixture of seawater/freshwater. eValues determined for a nearby source.fOne Thiobacterium sp. was isolated from w ater as single rods, and formation of the gelatinous matrix was observed during a laboratory life cycle experiment. Salinity and pH were determined by Seki and Naganuma (1989) for the seawater used in enrichm ent cultures (collected from the Strait of Georgia, Canada).8Values obtained from Glover et al. (2005).hHydrogen sulfide was detected qualitatively, bu t measurements yielded values below detection lim it (b.d.).

the sed im ent or of the gelatinous sphere was determ ined w ith the help of a dissection m icro­scope. During the analyses, the core was kept in an aquarium w ith w ater cooled to approxim ately 1 °C to sim ulate in situ tem perature conditions. A gentle jet stream , po in ting at the w ater surface of the core, stirred the overlying w ater and assured the develop­m ent of a d istinct diffusive boundary layer. Flux calculations were perform ed as in Lichtschlag et al. ( 2 0 1 0 ).

Composition o f the gelatinous matrix Three spheres from the cave m ats were separated from sam ple w ater and cleaned from attached sed im ent particles as efficiently as possible. They w ere then com bined and stored in a sterile po ly ­propylene tube at —20 °C u n til further processing. Wet w eight of the spheres was determ ined as 13.35 g. After freeze-drying for 4 days, total dry w eight was determ ined as 0.61 g.

Three replicate subsam ples of the dried m aterial, w ith in itia l w eights betw een 24 and 25 mg, were used for m easuring total carbon, nitrogen and sulfur conten t of the m aterial w ith a Therm o Fisher Scientific FlashEA 1112 (Thermo Fisher Scientific Inc., W altham, MA, USA). No acidification was

perform ed before the m easurem ents. In addition, inorganic carbon was m easured w ith a CM5012 C 0 2 C oulom eter (UIC Inc., Joliet, IL, USA) for three separate replicate subsam ples w ith in itia l w eights betw een 21 and 22 mg. Finally, organic carbon, organic nitrogen and organic/elem ental sulfur conten t w ere calculated accordingly. All values w ere corrected to a norm alized dry w eight, w hich excluded the total sea salt content of approxim ately3 9 % o .

Analyses of carbon and nitrogen isotopic com po­sition w ere conducted w ith tw o replicate subsam ­ples of 26 and 2 7 mg of the freeze-dried m aterial, using a Therm o F isher Scientific MS DELTAplus XP gas isotope ratio m ass spectrom eter (Thermo F isher Scientific Inc.). Isotopic ratios w ere corrected against air-N2 and the in terna tional Vienna Pee Dee Belem nite (VPDB) standard for obtaining 815N or 813C, respectively. No acidification was perform ed before the m easurem ents.

Fluorescence in situ hybridization Subsam ples of the cave m ats chosen for FISH analyses were fixed in 1 m l of 4% form aldehyde/ seaw ater for 1 h at room tem perature. They were then directly transferred to W hatm an N uclepore

The ISME Journal

Thiobacterium matsS G r i in k e e f a l

1034Track-Etch M em branes (PC MB; 25 mm; 0.2 gm; W hatm an, Schleicher and Schuell, M aidstone, UK), the latter being captured in an appropriate filtration apparatus and supported by W hatm an ME25 m em brane filters (m ixed cellulose ester; 25 mm; 0.45 pm; W hatm an). After rem oval of excess fixative, the filters were dried for 2 m in at 50 °C on a heating plate and w ere finally stored separately at —20 °C in sterile Petri dishes.

FISH was perform ed as described earlier (Snaidr et al., 1997). M ono-labeled probes (Biomers, Ulm, Germany) and hybrid ization conditions are listed in Table 2. All filter sections w ere countersta ined w ith 10 pi of a 1 or 2 .5pgm l~1 DAPI so lution before rinsing w ith d istilled w ater and air-drying. Finally,

the filters were m ounted in separate spots of Teflon-covered slides w ith a 2:3 m ix of Vecta Shield (Vector Laboratories Inc.) and Citifluor (Agar Scientific Ltd) and were stored at —20 °C before microscopy.

Construction o f 16S rRNA gene libraries and sequencingSubsam ples for phylogenetic analyses were stored at —20 °C, preserved in either PCR-grade w ater (Sigma- A ldrich B iochem ie GmbH, Hamburg, Germany), 1 x TE buffer (Promega Corporation, M adison, WI, USA) or w ithou t any fixative. Three different approaches for obtaining total com m unity DNA w ere applied ,

T ab le 2 O l ig o n u c le o t id e p ro b e s a n d h y b r id i z a t io n c o n d i t io n s u s e d fo r F IS H a n a ly s e s i n th i s s tu d y

Probe Probe specificity Probe sequence (5'-3') Target site within

16S/23S rRNA genea

FAh(%)

Thc/ T j(°c)

Positive hybridization

with Thiobac terium

cells

Reference

EUB33 8(1-111) Most bacteria Equimolar m ixture of three probes

16S, 338-355 35 46/48 Yes Daims et al. (1999)

EUB338-I Most bacteria GCTGCCTCCCGTAGGAGT 16S, 338-355 35 46/48 Am ann et al. (1990)EUB338-11 PI anc tomyc etales GCAGCCACCCGTAGGTGT 1 6 S ,338-355 35 46/48 Daims et al. (1999)EUB338-III Verrue omi cr obiales GCTGCCACCCGTAGGTGT 1 6 S ,338-355 35 46/48 Daims et al. (1999)NON338 Negative control ACTCCTACGGGAGGCAGC 1 6 S ,338-355 10 46/48 No W allner et al.

(1993)GAM42ae Gammapro teobac teri a GCCTTCCCACATCGTTT 23S, 1027-1043 35 46/48 Yes Manz e t al. (1992)BET42ae B etapr oteobacter ia GCCTTCCCACTTCGTTT 23S, 1027-1043 35 46/48 No Manz e t al. (1992)GAM 6 60 Free-living or

endosymbiotic sulfur- oxi dizing bacteria in the Gammapro teobac teri a

TCCACTTCCCTCTAC 16S, 660-674 35 46/48 No Ravenschlag e t al. (2001)

Gm.705 M ethanotrophs in the Gammaproteobacteria except Me thy local d um

CTGGTGTTCCTTCAGATC 16S, 705-722 10 46/48 No Gui ledge et al. (2001)

My669 Methylobacter sp p. and M ethylomonas sp p.

GCTACACCTGAAATTCCACTC 16S, 669-690 10 46/48 No Eller et al. (2001)

LMU Leucothrix m ucor CCCCTCTCCCAAACTCTA 16S, 652-669 10 46/48 No Wagner et al. (1994)G123T6 Thiothrix sp p. CCTTCCGATCTCTATGCA 16S, 697-714 10 46/48 No Kanagawa et al.

(2000)829-Thioploca M arine Thioploca sp p.

and Beggiatoa sp.GGATTAATTTCCCCCAACATC 16S, 829-849 10 46/48 No Teske et al. (1995)

Blim575 M arine Beggiatoa spp. CTAGCCGCCTACATACGC 16S, 575-592 10 46/48 No Mußmann et al. (2003)

B lim l93 M arine Beggiatoa spp. AAAAGACGCCCTCTTTCC 16S, 193-210 10 46/48 No Mußmann et al. (2003)

VS0673 Unattached, vacuolate, sulfur-oxi dizing spp.

CGCTTCCCTCTACTGTAC 16S, 656-673 10 46/48 No Kalanetra e t al. (2004)

WPF464 Attached, vacuolate, filamentous spp.

AGCTTTAAGTTTTTCTTCCC 16S, 445-464 10 46/48 No Kalanetra e t al. (2004)

Thm.465 Thiomargarita spp. Amon m ud volcano

GTCAAGACTCTAGGGTAT 16S, 465-482 10 46/48 No Girnth e t al. (submitted)

Thm482f Thiomargarita spp. CTTCTTCTATTGCTGATG 16S, 482-499 10 46/48 No This studyBiwa829 Freshwater Thioploca

spp.AGGTATACCCTTCCAACGTC 16S, 829-849 10 46/48 No Kojima et al. (2003)

ThioBar894f Th ioba cill u s baregensis TGAGTTTCAACTTCCGGC 16S, 894-911 10 46/48 No This studyThioBar272f Th ioba cill u s baregensis CTACTGTATCGTCGCCTT 1 6 S ,272-289 10 46/48 No This studyTHIOl A cidith ioba cill us spp. GCGCTTTCTGGGGTCTGC 16S, 1275-1292 10 46/48 No González-Toril

et al.(2003)

Thio820 A cidith ioba cill us spp. ACCAAACATCTAGTATTCATCG 16S, 816-837 10 46/48 No Peccia et al. (2000)LaSp60 Tubeworm

en dosymbionts, uncult.Gammapro teobac teri a

CCATCGTTACCGTTCGAC 16S, 60-77 10 46/48 No Duperron e t al. (2009)

RifTOl47 Bif/Tev/Oas symbiont G ATTTC TCCG AGTT GT CC 16S, 147-164 10 46/48 No Nussbaumer e t al. (2006)

RifT0445 Bif/Tev/Oas symbiont TCCTCAGGCTTTTCTTCC 16S, 445-462 10 46/48 No Nussbaumer e t al. (2006)

Ohaal-65 O. haakonm osbiensis symbiont

AGCTCTTGCTGTTACCGT 16S, 65-82 10 46/48 No Lösekann e t al. (2008)

The ISME Journal

Thiobacterium matsS Grünke e t al

T able 2 (C ontinued)

Probe Probe specificity Probe sequence (5'-3') Target site within

16S/23S rRNA genea

FAh(%)

Thc/ T j(°C)

Positive hybridiza tion

with Thiobacterium

cells

Reference

O haal-77 O. haakonm osbiensis GCCAAGAGCAAGCTCTTG 16S, 77-94 10 46/48 No Lösekann et aí.symbiont (2008)

Ohaa2-60 O. haakonm osbiensis GCATCGTTACCGTTCGAC 16S, 60-77 10 46/48 No Lösekann et al.symbiont (2008)

Ohaa2-77 O. haakonm osbiensis CCTGCTAGCAAGCTAGCA 16S, 77-94 10 46/48 No Lösekann et al.symbiont (2008)

Abbreviations: FISH, fluorescence in situ hybridization; rRNA, ribosomal RNA. aEscherichia coli positions.bFormamide concentration in the hybridization buffer. cHybridization temperature. dW ashing temperature.eMixed in a 1:1 ratio w ith unlabeled competitor oligonucleotides, to assure specific hybridization (Manz et al., 1992; Amann and Fuchs, 2008). fThese probes represent newly developed oligonucleotides w hich have not been tested in corresponding positive control experiments because of inavailability of the respective organisms/16S rRNA gene sequence.Probe specificity is given for the original design. In addition to the listed probes, nine other, but so far unpublished probes, were used: two designed for a hypersaline Beggiatoa sp., one designed for sulfur-oxidizing endosymbionts, one designed for Thiomargarita spp.f and five designed for smaller subgroups w ithin the Gammaproteobacteria. None of them positively hybridized w ith the target organisms, w hy publication of the probe sequences was reserved for the original developers. Specific gammaproteobacteria! probes were used at lower stringency than published, that is, 10% formamide, to create favorable hybridization conditions. For probe Gm705 specificity had been achieved by the original publishers at Td = 51 °C (midpoint dissociation temperature).

1035

includ ing (i) freezing and thaw ing, (ii) an extraction m ethod developed for m arine m icroorganism s em ­bedded in exopolysaccharides (El; N arváez-Zapata et al., 2005) and (iii) the U ltraClean Soil DNA Isolation Kit (E2; MO BIO Laboratories Inc., Carlsbad, CA, USA).

U niversal bacterial prim ers GM3F (5'-AGAGTTTG ATCMTGGC-3'; M uyzer et al., 1995) and GM4R (5'-T ACCTTGTTACGACTT-3'; M uyzer et al., 1995) were used to am plify a nearly com plete region of the 16S rRNA gene. PCR products were purified using the QIAquick PCR Purification Kit (Qiagen, H ilden, Germany), ligated into pGEM-T Easy (Promega Corporation) and transform ed into One Shot TOP 10 C hem ically Com petent Escherichia coli (Invitrogen Corporation). After p lasm id preparation (Montage P lasm id M in ip repHTs Kit; M illipore GmbH, Schw albach, Germany), inserts w ere sub­jected to Taq cycle sequencing w ith an ABI Prism 3130x Genetic A nalyzer (A pplied Biosystems, Foster City, CA, USA). Sequencing was either conducted w ith the M13R vector prim er, or by using the bacterial prim ers 907R (5'-CCGTCAATTCCTT TRAGTTT-3'; M uyzer et al., 1995) and prim er2 (5'-ATTACCGCGGCTGCTGG-3'; M uyzer et ah, 1993). Both sequencing approaches y ielded good- quality fragments of the 16S rRNA gene w ith lengths betw een 400 and 700 bp.

Sequence analysesPartial sequences derived from 907R and prim er2 w ere assem bled w ith Sequencher 4.6 software (Gene Codes Corporation, A nn Arbor, MI, USA) and m anually checked. In addition , all libraries were screened for pu tative chim eras w ith the M allard program at a cut-off of 99.9% (Ashelford et ah,

2006). A nom alous sequences w ere further investi­gated by using P intail (Ashelford et ah, 2005), w ith nearest neighbors of the sequences obtained by using the SILVA-based SINA W ebaligner (Pruesse et ah, 2007). Sequences w ith genuine chim eric signals w ere excluded from further analyses.

To search for a sequence com m on to either all libraries or all sam ple sites, gam m aproteobacteria! sequences, as previously determ ined w ith the RDP Classifier (Wang et ah, 2007), and their nearest neighbors (obtained by using the SILVA-based SINA W ebaligner and ARB database) were used for tree reconstruction in ARB (Ludwig et ah, 2004) by applying m axim um likelihood (RAxML).

Nucleotide sequence accession numbers A ll sequences have been subm itted to the EMBL database u n d er accession no. FN597289 to FN597418 (Gam maproteobacteria) and FN662907 to FN663128 (all other sequences).

ResultsVisual description o f the gelatinous mats A t all three sam pling sites the gelatinous mats w ere spherically shaped masses. A second, less dom inant, layer-like m at type was found only in the shallow -w ater cave. The environm ental settings prevailing at these three sites and in previously described habitats are sum m arized in Table 1.

On the w hale bone a y e llow ish -w hite gelatinous sphere was observed (Figure la). Its diam eter was approxim ately 4 mm. The sphere had evolved w ith in 1 to 2 days in an area that was scraped free

The ISME Journal

Thiobacterium matsS G r i in k e e f a l

1036of bacterial mats of Beggiatoa and Arcobacter (as ind icated by m icroscopy). During sam pling the entire gelatinous sphere was rem oved and in the next few days no thing alike grew back on the bone.

The m ats on top of the deep-sea m ethane seep sed im ent had an approxim ate diam eter of 1.5 cm (Figure lb). U nder bright light conditions the gelatinous spheres seem ed to be covered by a th in w hite layer w ith some th icker w hite filam ents. The w hite color of the gelatinous spheres is an in d ica­tion for the presence of elem ental sulfur, and m icroscopic analyses later revealed a high density of light-reflecting granules typical for the sulfur inclusions of giant sulfide-oxidizing bacteria (Teske and Nelson, 2006). The w hite filam ents were also observed surrounding the gelatinous m asses on top of the sed im ent and w ere m orphologically attributed to bacteria of the genus Beggiatoa, m at-form ing giant su lh d e oxidizers know n to occur on highly reduced sulfidic sedim ents (Ahm ad et al., 1999; M ills et al., 2004; de Beer et al., 2006). The underly ing sedim ent was characterized by a beige-brow n top layer above a dark gray to blackish horizon.

The largest accum ulation of gelatinous m ats in this study was found in a shallow -w ater cave off Paxos in Greece (Figure le ; Supplem entary Figure 1; Supplem entary Results and D iscussion). The cave walls were alm ost com pletely covered by gelatinous mats over an estim ated area of 5 to 7.5 m 2, w ith a sharp boundary below the top of the cave (Dr Thom as Beer, personal com m unication). Two m orphological types of m ats could be observed w ithou t a particu lar zonation in the cave (Figure le). The dom inating type consisted of spherical structures show ing diam eters betw een less than 1cm and more than 10 cm, the second type was characterized by a rather th in layering of the gelatinous m aterial. Both spherical- and layer- like structures of the gelatinous m ats appeared w h itish sim ilar to the deep-sea spheres. The p resence of elem ental su lfur and w hite filam entous bacteria was verified by m icroscopy. The m ats were present in sum m er 2006 and 2007, bu t had vanished in January 2007 w hen they w ere rep laced by a dense filam entous m at on the cave w alls (Paul Bowbeer, personal com m unication).

Microscopic description o f Thiobacterium microbes M icroscopic exam ination of the gelatinous spheres from all three sites revealed the presence of high num bers of rod-shaped, som etim es slightly curved chains of sulfur granules. The Thiobacterium spp. em bedded in the gelatinous m aterial found on the w hale bone had an approxim ate size of 3 -8 pm X 0 .9-2 pm (Figure 2a), those from the deep-sea ranged from 3-11 pm x 0.7-2 pm (Figure 2b), and the ones from the shallow -w ater cave were from 2-9 pm x 0.3-2 pm in size (Figure 2c). The num ber of sulfur granules per cell

Figure 1 Thiobacterium gelatinous m ats analyzed in th is study, [a) G elatinous sphere w ith an estim ated d iam eter of 4 mm, found on top of a m inke w hale bone kept in an aquarium . In add ition , the bone was covered by dense m ats of th io troph ic bacteria [Beggiatoa and Arcobacter), [b) G elatinous sphere found on top of a sed im ent core collected at the Storegga Seeps, [c) A ssem blage of spherically shaped and layer-like gelatinous m asses coating the w alls of a shallow -w ater cave off Paxos in Greece.

varied betw een 3 and 5 (whale bone), 3 and 11 (deep-sea sedim ent) as w ell as 3 and 10 (cave). In addition , single and pairs of sulfur granules were

The ISME Journal

Thiobacterium matsS Grünke e t s /

Figure 2 M icroscopic observations on Thiobacterium cells em bedded in the gelatinous spheres, [a) W hale bone sam ple [bright field), [b) deep-sea sam ple [phase contrast) and [c) shallow -w ater cave sam ple [phase contrast). In all sam ples, chains of elem ental su lfu r granules w ere visible, [d) True cell ou tlines a round the granules w ere no t visible by phase contrast and bright field m icroscopy, bu t by staining w ith the protein-targeting fluorochrom e SYPRO Red using epifluorescence m icroscopy [staining was conducted w ith fixed m aterial recovered from the cave in 2006).

observed, bu t it was not clear w hether those belonged to living cells or rem nants of cell decay. Except for the Thiobacterium cells from the w hale bone, we found that even w ith in one cell the size of the sulfur inclusions could vary by up to a factor of 5. By phase contrast m icroscopy cell outlines around the granules could no t be observed, bu t becam e visible after staining fixed m aterial w ith the protein-targeting fluorochrom e SYPRO Red (Figure 2d). We observed a loss of the in ternal sulfur deposits in ethanol fixation, because ethanol d is­solves elem ental sulfur. However, even w ith form aldehyde/seaw ater fixation, the sulfur granules vanished w ith time. At all three sites, im m ediately after sam pling, the Thiobacterium cells em bedded in the gelatinous m ats appeared to be non-m otile. However, m icroscopic exam ination of gelatinous m aterial obtained from the underside of a red leaf-like structure (cave) revealed a few rod-shaped cells show ing directed or tum bling m ovem ent ou t­side of the matrix.

In add ition to Thiobacterium , m icroscopic ana­lyses of the gelatinous spheres from all three sam ple sites revealed the presence of m any other organism s em bedded in and associated w ith the gelatinous m ats. This unexpected ly high m icrobial diversity is described separately in the S upplem entary Results and D iscussion and in S upplem entary Figure 2.

Microsensor measurementsM icrosensors w ere used to m easure h igh-resolution profiles of oxygen and sulfide concentration, as

w ell as pH, (i) through a gelatinous deep-sea sphere (diam eter approxim ately 1.5 cm) and the first 1.5 cm of the underly ing sed im ent (Figure 3a) and (ii) through the top 4 cm of sed im ent im m ediately next to the sphere (Figure 3b). Oxygen penetrated the gelatinous m atrix to a dep th of approxim ately 3 to 4 mm. Sulfide from the sed im ent diffused into the underside of the structure, bu t no local overlap of oxygen and sulfide was observed in the sphere. The pH profiles showed several m inim a and maxima, b u t generally decreased in the sphere, possibly as a consequence of su lh d e oxidation. In the u n d er­lying sedim ent, no oxygen could be detected and sulfide concentrations reached a m axim um of 3.2 mM at a dep th of 1.5 cm below the sphere. The sulfide and oxygen fluxes into the sphere were 5 and 11 m m olm ~2 d-1, respectively, m atching the stoichiom etry of sulfide oxidation w ith oxygen as electron acceptor. Next to the sphere, oxygen penetrated approxim ately 2 m m into the sedim ent. Sulfide was not detectable u n til 2.7 cm below seafloor and then increased linearly w ith depth. At 2 cm below seafloor, the pH reached a m inim um of 7.6, below w hich it increased w ith depth. The oxygen uptake of the sed im ent next to the sphere was com parable to that in the sphere w ith 10 m m ol n r 2 d _1.

Elemental and isotopic composition o f the matrix The dried m atter com prised 5.48% (w/w) organic carbon, 1.06% (w/w) nitrogen and 3.62% (w/w) organic and elem ental sulfur. The m olar C:N:S ratio

1037

The ISME Journal

Thiobacterium matsS Grün k e e t al

1038

80

SPH ER E

pH

8 5 90

0.0 -0.03 -t—

oxygen (mM)

0.1 0.2 0.3

bottom water

-0.01

sphere

E£ 0.01 -

0 02 -

0 03 -—o— oxygen — sulfide pH1

pH 2005

0 1 2 3 4

sulfide (mM)

80

SEDIMENT

PH

8.5 90

oxygen (mM)

0.1 0.2 0.3

bottom water

0 2 31 4

sulfide (mM)

Figure 3 Oxygen, su lfide and pH m icroprofiles m easured [a) th rough a gelatinous sphere w ith a d iam eter of approxim ately 1.5 cm and [b) w ith in reference sed im ent nex t to the sphere [recovered from deep-w ater seafloor at the Storegga Seeps). Single, ex situ m easured profiles are show n. R eplicate pH profiles 1 [black) and 2 [gray) w ere re trieved from [a) the sam e gelatinous sphere and [b) the adjacent sedim ent. The m icroprofiles clearly show that the gelatinous spheres had form ed in an area in w h ich hydrogen sulfide reached the sed im ent surface. Sulfide and oxygen d id no t overlap w ith in the sphere, ra ising the question of how electron transport is m ediated w ith in the sphere during sulfide oxidation.

of 6:1:1.5 reflects the enrichment of the cells and matrix w ith elemental sulfur. The 813C measured for two replicates of the freeze-dried cave material resulted in values o f—16.1 and —18.3%o. The 815N was determined for the same replicates as —6.0 and —6.2%o. More details on the composition of the matrix are given in the Supplementary Results and Discussion.

Ribosomal RNA gene analysesBacterial 16S rRNA gene libraries were constructed w ith subsam ples of the gelatinous spheres from all three sam pling sites. Gamma-, Delta- and E psilon­proteobacteria dom inated the m at sam ples from all three sites. O ther abundan t classes of bacterial 16S rRNA genes inc lu d ed the A lphaproteobacteria, as w ell as Flavobacteria, Sphingobacteria, P lanctom y­cetacia and C lostridia (Supplem entary Table 1; Supplem entary Results and D iscussion). In spite of the observed abundance of Thiobacterium cells in all sam ples, it was not possible to identify com m on phylotypes dom inating the clone libraries of all three sites (all bacterial libraries per site com bined). A substantial diversity of 16S rRNA gene sequences belonging to the G am m aproteobacteria was retrieved from all three m ats (Figure 4), includ ing m any sequences of sulfur-oxidizing bacteria, such as different types of Beggiatoa, sulfide-oxidizing sym bionts of tubew orm s, and bacteria from hot springs, hydrotherm al vents, cold seeps and lava. Subsequent FISH analyses w ith the GAM42a probe on fixed subsam ples of the cave spheres revealed

that the Thiobacterium cells belong to the Gamma­proteobacteria (Figure 5). Unfortunately, further attem pts to confine phylogenetic relatedness using more specific oligonucleotide probes for FISH were not successful (Table 2). A probe includ ing all sulfide-oxidizing and sulfur-storing m em bers of the Thiotrichaceae has no t been developed so far. Specific gam m aproteobacteria! probes, if not o ther­w ise indicated , w ere used at lower stringency than published , that is, 10% form am ide, to create favor­able hybrid ization conditions. However, none of the established FISH probes specific for certain sub­groups w ith in the G am m aproteobacteria resulted in positive hybrid ization w ith Thiobacterium spp.-resem bling cells, includ ing those for the Thio- trichaceae-associated genera Beggiatoa, Thioploca, Thiomargarita, Thiothrix and Leucothrix, as w ell as probes for other organism s involved in the m icrobial cycling of sulfur like A cidith iobacillus, or for endosym bionts of seep and vent fauna. Further approaches to identify a candidate 16S rRNA gene sequence for Thiobacterium by single cell sorting were also not successful (data not shown) and detailed sequence analyses of Thiobacterium cells w ill require special atten tion in future experim ents.

Discussion

Thiobacterium classificationSince its first description by Molisch in 1912 (Molisch, 1912), no representative of this genus has been

The ISME Journal

Thiobacterium matsS Grünke e f s /

1039C3170 (0, 0, 10), FN597360

----------- une bacterium from oyster shell EU369126une bactenum from Newport Harbour (RI). EU799857

C633, FN597347Thiobacillus baregensis, Y09280

Acidithiobacillus thiooxidans, AY552087Nitrosococcus oceani. M96395

DS179, FN597397

une bactenum from natural gas field. OQ867047 DS433. FN597302

r une bactenum from seafloor lavas. EU491420 ~ 1--------- WB1373, FN597290

DS183 (0, 3, 0). FN597400-------------------- une m anne Beggiatoa sp . AF532775

une. m anne Beggiatoa sp Bay of Concepción', AF035956 une m anne Beggiatoa sp , AF5Ü2773----------------- freshwater Thioploca ingnca. EU718071

DS118 (0, 3, 0), FN597376 une. filamentous bactenum White Point'. AY496953

DS1243 (0. 20 .0 ). FN597317 DS136. FN597388DS1299. FN597335

m anne Beggiatoa sp MS-81-6'. AF110277 DS1279 (0, 12 ,0 ), FN597305

hypersaline Beggiatoa sp Chiprana'. EF428583 freshwater Beggiatoa sp AA5A AF 110275

freshwater Beggiatoa alba B18LD'. AF 110274 r une bactenum from seafloor lavas ËU491607

___________f — DS182, FN597399u une bactenum from arctic surface sediment. EU287165

DS1287, FN5973271------------------------- une bactenum from d eep-sea sediment. AY373399

i C3098, FN5973521----------- une bacterium from subm anne hot spnng mat. AB294964

Thiohalomonas demtnftcans, EF 117909 une bactenum from hydrothermal sediment. AY225631------------------- une bacterium from Gulf of Mexico seafloor sediment. AY542567 C496. FN597342une bactenum from arctic surface sediment. EU287176

DS119. FN597385 une bacterium from seafloor lavas. EU491111

une bactenum from deep-sea sediment, AJ966594 une bacterium from Yellow S ea sediment. EU617749

C3140, FN597357 une bacterium from seafloor lavas. EU491811

C3227, FN597373■ une bacterium from mud volcano. AY592169 une bacterium from subm anne hot spnng mat. AB294933

C3185. FN597363 ectosymbiont of Robbea sp . EU711428Thiorhodovibno winogradskyi. AB016986

— C3118. FN597355 Candidatus Thiobios zoothamnicoli'. AJ879933

une bactenum from subm anne hot spnng mat. AB294952 Lamellibrachia columna endosymbiont. U77481 DS331 (1. 18. 0). FN597410Escarpia spicata endosymbiont Alvin #2837', AF 165909

Vibno Sp TJD753'. DQ993344C636. FN597348

^— Colwellia\ psychrerythi 597393

raea. AF 173964DS150. FN5j »

. Halomonas sp MÁN K9', AB166927 WB1065. FN597293

DS149. FN597392 — une bacterium from hydrothermal field carbonate chimney. DQ270628

r* une bactenum from hydrate ridge m anne sediment. AJ535246 DS396, FN597417

DS1283. FN597307WB1459, FN597289

une bacterium from deep-sea sulfide oxidizer mat. EF687362 WB1374, FN597291--------------- une bacterium from arctic surface sediment. EU287338DS146, FN597379

une bacterium from arctic sediment. EU050811 DS372, FN597404

i— DS151, FN5973801------------- une bactenum from arctic sediment, EU050810

une bacterium from crab surface, EU265799 OS1222, FN597314

Leucothrix mucor. X87277Thiothrix eikelboomn. L79965

une bacterium from arctic sediment. EU050807— une bactenum from hydrothermal field carbonate chimney. DQ270618

OS412, FN597299Thiomicrospira crunogena. AF069959 Thiovibno halophilus. DQ390450

C631, FN597341 C735, FN597340

C3100. FN597353 DS336, FN597403

une bactenum from deep-sea iron oxidizer mat, EF687204DS147. FN597390- Maorithyas hadalis gili thioautotrophic symbiont I AB188776

DS176. FN597382_ P DS267, FN597409

^ ------- DS117, FN597384une bactenum from Namibian sediment. EU290725

Desulfonema magnum. U45989

Figure 4 M axim um likelihood tree show ing gam m aproteobacteria! 16S rRNA gene sequences obtained in th is study [bold type; WB = w hale bone m at, DS = deep-sea sed im ent m at, C = cave mat) to reference sequences w ith in the G am m aproteobacteria. O nly selected sequences are show n. For large c lusters of sequences obtained in th is study, the num ber of sequences associated w ith the given represen tative is ind ica ted in paren theses, follow ing the order WB, DS and C. In itia l calculations w ere conducted w ith nearly fu ll-length sequences [Escherichia coli positions 99-1331). Partial sequences [bold type) w ere ad ded subsequently to the reconstruc ted tree by app ly ing parsim ony criteria. D eltaproteobacteria! sequences w ere used as outgroup.

cultivated or genomically classified. The visual ob­servation of internally stored sulfur granules and the biogeochemical analyses discussed below suggest that Thiobacterium is closely related to other known sulfide-oxidizing and sulfur-storing members of the Thiotrichaceae. Positive hybridization of Thiobacter-

ium cells w ith the probe GAM42a (76% group cover­age; Am ann and Fuchs, 2008) confirmed their phylogenetic affiliation w ith the class of Gammapro­teobacteria (Figure 5), as previously suggested based on morphological analogies (Kuenen, 2005). However, no common gammaproteobacteria! phylotypes (Figure 4)

The ISME Journal

Thiobacterium matsS G r i in k e e f a l

1040

Figure 5 FISH images of Thiobacterium cells [cave sam ples 2007). [a) Positive signal after staining w ith the Gammaproteobacteria-targeting probe GAM42a [unlabeled competitor BET42a). [b) Corresponding DAPI stain, [c) Overlay of [a, b). In addition, staining of the Thiobacterium cells w ith the probe BET42a [unlabeled competitor GAM42a) did not result in any positive signal. Autofluorescence of the cells was found to be negligible.

dominating the clone libraries of all three sites could be identified. Reasons might include that one of the universal primers does not have full access to its target site on the Thiobacterium spp.-16S rRNA gene, or that the amplification of this gene is underrepresented compared w ith that of other bacteria w ith the applied PCR conditions, a problem also known from the giant sulfide oxidizer Thiomargarita namibiensis (Schulz,2006). However, considering the broad range of environmental conditions from w hich it was described, the genus Thiobacterium could be composed of rather different phylotypes, as it is known for the genus Beggiatoa (Figure 4; Ahm ad et a l, 2006; Teske and Nelson, 2006).

Niches o f ThiobacteriumEnvironm ental conditions u n d er w hich Thio­bacterium m ats have been observed include

pH values betw een 7.2 and 9.3, tem peratures ranging from —1 °C to 45.5 °C and w ater depths betw een 0 and 2700m (Table 1). Low 815N values of —6.0 and —6.2% o m easured for the cave m ats in this study suggest au to trophic growth of the associated organism s (see also S upplem entary Results and D iscussion). The m ain environm ental factor select­ing for Thiobacterium seems to be the availability of the electron donor hydrogen sulfide. Their ability to store m assive am ounts of sulfur, and their prefer­ence for sulfidic habitats suggests that Thiobacterium could be a sulfide oxidizer, belonging at least functionally into the same group as other mat- forming m em bers of the Thiotrichaceae such as Beggiatoa, Thiomargarita, Thioploca and Thiothrix. W ith m icrosensor m easurem ents it could be show n that the gelatinous spheres had exclusively formed in an area in w hich hydrogen sulfide reached the sed im ent surface and that th is hydrogen sulfide was

The ISME Journal

Thiobacterium matsS Grünke e f s /

depleted w ith in the sphere (Figure 3a). The oxida­tion of sulfide occurs in several steps, from sulfide to sulfur, and from sulfur to sulfate. Sulfide ox ida­tion can be driven by oxygen, n itrate and Mn- and Fe-oxides. The m icrosensor profiles (Figure 3a) ind icate that the m atrix had consum ed sto ich io­m etric am ounts of oxygen and sulfide (2:1 in aerobic sulfide oxidation), bu t sulfide and oxygen d id not overlap. A pronounced gap betw een oxygen and sulfide has previously also been observed for Thioploca and Beggiatoa spp. that are capable of bridging this gap by m eans of their gliding m otility and the use of in terna lly stored n itrate as additional electron acceptor u nder anoxic conditions (Fossing et al., 1995; H uettel et al., 1996; M ußm ann et al., 2003; Sayama et al., 2005; Lichtschlag et al., 2010). However, Thiobacterium are neither know n to express a gliding behavior w ith in the gelatinous m atrix, nor to possess an in ternal vacuole for n itrate storage. The nature of a po ten tial in term ediate electron shuttle therefore rem ains unknow n. The existence of a m icrooxic and m icrosulfidic environm ent w ith in the m atrix can be ru led out, as at undetectab ly low levels of reactants, the diffusive transport w ill be too low for significant m etabolic activity. Also m etal cycling, as in the suboxic zone, can be excluded as no solid phase transport seems possible in these gel-like structures. The potential involvem ent of electrical currents in the oxidation of sulfide rem ains a possib ility (Nielsen et al., 2008).

It is unknow n how and w hen Thiobacterium can outcom pete other sulfide-oxidizing bacteria for energy and space. On the w hale bone, it grew after Beggiatoa and A rcobacter mats had been physically rem oved, ind icating that it m ay only be able to com pete w ith others after specific disturbances. W hen it grew, Beggiatoa was always observed, either growing th in ly on the gelatinous m atrix or more densely next to it, ind icating that it m ay com pete w ith Thiobacterium for the same energy source and space.

Conclusion and further questions M orphological and ecological characteristics of Thiobacterium cells and m ats suggest that this genus is closely related to o ther sulfide-oxidizing and sulfur-storing bacteria of the family T h io tricha­ceae. M ost interestingly, we found a stoichiom etrical oxidation of sulfide to sulfate w ith oxygen w ithou t them overlapping. M otile m icrobes resem bling the m orphology of Thiobacterium have been reported few tim es (Lackey and Lackey, 1961; Lackey et ah, 1965; this study), bu t so far were never truly accounted to this genus. M olisch described sim ilar m otile organism s separately as Bacillus thiogenus (M olisch, 1912), hence it rem ains an im portan t question w hether they can move inside their gels. It also rem ains hypothetical w hich factors trigger the attachm ent of free cells of Thiobacterium in a

1041certain habitat and w hether m at form ation may undergo a succession in w hich other p ioneer m icrobes m ay be needed. As often reported , the gelatinous m asses of Thiobacterium w ere found to be attached to a substratum , for exam ple, the algae Lyngbya (Lackey and Lackey, 1961), rock surface (Vouk et ah, 1967), sed im ent (this study) or a w hale bone (this study), suggesting that the production of a gel m ay not only be used for com peting against other m icrobes, b u t also functions as an anchor in areas w ith favorable environm ental conditions (Vouk et al., 1967; Seki and Naganum a, 1989).Further it is no t clear w hat role quorum sensing, density -dependen t ce ll-ce ll com m unication, may have in the form ation, m aturation and final d is in ­tegration of the gelatinous mats. In this respect, future studies need to assess (i) the m echanism s triggering the appearance of Thiobacterium in conspicuous gelatinous m ats, (ii) the access to electron donor and acceptor by the cells em bedded in the th ick m atrix, (iii) the 16S rRNA gene-based taxonom ic affiliation of Thiobacterium spp. and (iv) w hether and w hat role other m icrobes m ay have in the developm ent of the Thiobacterium m ats in natu ral environm ents.

AcknowledgementsW e th a n k T h o m as D ah lg ren from th e T järnö M arine B io log ical L aborato ry for g iv ing u s th e o p p o rtu n ity to s tu d y th e w h a le b ones, as w e ll as H ans R oy from th e C en te r for G eom icrob io logy at th e U n iv e rs ity o f A arh u s (D enm ark), for h e lp in g w ith sam p le recovery an d first m ic ro sco p ic analy ses. We also th a n k th e crew s of th e RV P o urquo i Pas? an d th e ROV V ictor 6000 (IFREMER) for th e ir great su p p o r t w ith w o rk at sea. S p ec ia l th an k s go to P au l B ow beer an d Dr T h o m as B eer a n d h is fam ily, w ho d isco v e red th e sh a llo w -w a te r cave off P axos an d h e lp e d reco v erin g a ll sam p les. We th a n k Jö rn P M eyer, H eide Schulz-V ogt, Bo B Jorgensen , K atrin K n itte l, G u n te r W egener, S u sa n n e H inck , A n ja K am p, S ab in e Lenk, V erena S a lm an , E rika W eiz, M artin a A lisch , T h o m as M ax, M artin a M eyer, A n d rea S ch ip p er , T om as W ilkop , th e S ym biosis G roup at M PI an d R utger de W it for th e ir te c h n ic a l su p p o r t an d h e lp fu l su ggestions a n d d iscu ss io n s . W e fu r th e r th a n k tw o an o n y m o u s rev iew ers for th e ir co n s tru c tiv e co m m en ts th a t h e lp e d to im p ro v e th e paper. T he w o rk of SG a n d AB w as fin an c ia lly s u p p o r te d b y th e EU 6 th FP HERM ES p ro jec t (GOCE-CT- 2005-511234-1) a n d th e M ax P lan ck Society.

ReferencesA h m ad A, B arry JP, N elso n DC. (1999). P h y lo g en e tic

a ffin ity of a w id e , vacu o la te , n itra te -a c c u m u la tin g B eggiatoa sp . from M o n terey C anyon , C alifo rn ia , w ith T h io p lo ca sp p . A p p l E nv iron M icrob io l 65: 270-277 .

A h m ad A, K alan e tra KM, N elso n DC. (2006). C u ltiv a ted B eggiatoa spp . d efin e th e p h y lo g en e tic roo t o f m o r­p h o lo g ica lly d iverse , n o n c u ltu re d , v acu o la te su lfu r bac te ria . Can J M icrob io l 52: 591 -598 .

The ISME Journal

Thiobacterium matsS G r i in k e e f a l

1042A m an n R, F u ch s BM. (2008). S ing le -ce ll id e n tif ic a tio n in

m ic ro b ia l c o m m u n itie s by im p ro v e d flu o rescen ce in situ h y b rid iz a tio n te ch n iq u es . N a t R e v M icrobio l 6: 339-348 .

A m an n RI, B in d e r BJ, O lso n RJ, C h ish o lm SW, D evereux R, S tah l DA. (1990). C o m b in a tio n of 16S rRNA- ta rg e ted o lig o n u c leo tid e p ro b es w ith flow cy tom etry for an a ly z in g m ix ed m ic ro b ia l p o p u la tio n s . A p p l E nviron M icrob io l 56: 1919-1925 .

A n a g n o stid is K. (1968). U n te rsu c h u n g e n ü b e r d ie S alz- u n d S ü ssw asse r-T h io b io cö n o sen (S u lp h u re tu n r) G riechenlands. W issenschaftliches Jahrbuch der Physiko- M athem atischen Fakultät A ristoteles-Universität, vol. 10. T h essa lo n ik i, p p 551 -552 .

A sh e lfo rd KE, C h u zh an o v a NA, F ry JC, Jones AJ, W eight- rnan AJ. (2005). A t leas t 1 in 20 16S rRN A seq u en ce reco rd s cu rre n tly h e ld in p u b lic rep o s ito r ie s is e s tim a ted to co n ta in su b s ta n tia l an o m alie s . A p p l E nviron M icrob io l 71: 7724-7736 .

A sh e lfo rd KE, C h u zh an o v a NA, F ry JC, Jones AJ, W eightm an AJ. (2006). N ew screening softw are show s tha t m ost recent large 16S rRNA gene clone libraries con ta in chim eras. A p p l E nviron M icrobiol 72: 5734-5741.

C a ld w ell DE, C ald w e ll SJ. (1974). T he re sp o n se of litto ra l co m m u n itie s of b ac te ria to v a ria tio n s in su lfid e an d th io su lfa te . A b stra c ts A n n M ee tin g A m S o c M icrobio l 74: 59.

D ah lg ren TG, W ik lu n d H, K älls tröm B, L u n d ä lv T, S m ith CR, G lover AG. (2006). A sh a llo w -w a te r w h ale -fa ll ex p e rim en t in th e n o r th A tlan tic . Cah B io l M ar 47: 385 -389 .

D aim s H, B rü h l A, A m an n R, S ch le ife r KH, W agner M.(1999). T he d o m ain -sp ec ific p ro b e EUB338 is in s u ff i­c ien t for th e d e te c tio n of a ll B acteria : d ev e lo p m en t an d ev a lu a tio n o f a m ore co m p reh en s iv e p robe set. S y s t A p p l M icrob io l 22: 4 3 4 -4 4 4 .

de B eer D, S au te r E, N iem an n H, K au l N , F o u ch e r J-P, W itte U e t al. (2006). In s itu f luxes a n d z o n a tio n o f m ic ro b ia l ac tiv ity in su rface se d im e n ts o f th e H âkon M osby M u d V olcano. L im n o l O ceanogr 51: 1315-1331 .

de B eer D, S ch ram m A, S an teg o ed s CM , K üh l M. (1997). A n itr i te m ic ro sen so r for p ro filin g e n v iro n m e n ta l b io film s. A p p l E nv iron M icrob io l 63: 973 -9 7 7 .

D évidé Z. (1952). Z w ei n e u e farb lose S chw efe lbak te rien : Thiogloea ru ttn er i n .gen ., n .sp . u n d Thiogloea ragusina n .sp . S c h w e iz Z H ydro l 14: 446-455 .

D u p e rro n S, de B eer D, Z b in d e n M, B oetiu s A, S c h ip a n i V, K ah il N e t al. (2009). M o lecu la r ch a ra c te riz a tio n of b ac te ria asso c ia ted w ith th e tro p h o so m e a n d th e tu b e of L a m ellib rach ia sp ., a s ib o g lin id a n n e lid from co ld seep s in th e eas te rn M ed ite rran ean . F EM S M icrob E col 69: 395 -409 .

E ller G, S tu b n e r S, F ren ze l P. (2001). G roup -spec ific 16S rR N A ta rg e ted p ro b es for th e d e te c tio n of ty p e I an d ty p e II m e th a n o tro p h s by flu o rescen ce in situ h y b rid isa tio n . F EM S M icrob io l L e tt 198: 9 1 -97 .

F je rd in g stad E. (1979). S u lfu r bac te ria . A m S o c Test M a t p 77.

F oss ing H, G alla rdo VA, Jo rgensen BB, H ü tte l M, N ie lsen LP, S ch u lz H e t al. (1995). C o n cen tra tio n a n d tra n sp o rt o f n itra te by th e m a t-fo rm ing s u lp h u r b ac te riu m T h iop loca . N a tu re 374: 713 -715 .

G irn th A-C, G riinke S, L ich tsch lag A, F e ld e n J, K n itte l K, W enzhö fer F e t al. (su b m itted ) A n o ve l, m a t-fo rm in g

Th iom argarita p o p u la tio n a sso c ia ted w ith a su lfid ic f lu id f lo w fro m a deep -sea m u d volcano .

G lover AG, K älls tröm B, S m ith CR, D ah lg ren TG. (2005). W o rld -w ide w h a le w orm s? A n e w sp ec ie s o f O sedax from th e sh a llo w n o rth A tlan tic . P roc R S o c B 272: 2587-2592 .

G onzález-T oril E, L lo b e t-B rossa E, C asam ayor EO, A m an n R, A m ils R. (2003). M icrob ia l ecology o f an ex trem e ac id ic e n v iro n m en t, th e T in to River. A p p l E nviron M icrob io l 69: 4 853 -4865 .

G u g lian d o lo C, M augeri TL. (1993). C h em o lith o tro p h ic , su lfu r-o x id iz in g b ac te ria from a m arin e , sh a llo w h y d ro th e rm a l v en t o f V u lcano (Italy). G eom icrobio l J 11: 109 -120 .

G u lledge J, A h m ad A, S te u d le r PA, P o m eran tz WJ, C avanaugh CM. (2001). F am ily - a n d gen u s-lev e l 16S rR N A -targeted o lig o n u c leo tid e p ro b es for eco log ical s tu d ie s of m e th a n o tro p h ic b ac te ria . A p p l E nviron M icrob io l 67: 4 726 -4733 .

H ed o in H, C oû té A, K aiser P, L aug ier R. (1996). N a tu re an d o ccu rren ce o f su lfo x id iz in g b ac te ria in B arégine d ev e lo p in g in su lfu ra te d th e rm a l w a te rs at B areges (France). H ydrob io log ia 323: 75 -81 .

H u e tte l M , F o rs te r S, K löser S, F oss ing H. (1996). V ertical m ig ra tio n in th e se d im e n t-d w e llin g su lfu r b ac te ria T h io p lo ca sp p . in overcom ing d iffu s io n lim ita tio n s . A p p l E nv iron M icrob io l 62: 1863-1872 .

Je ro schew sk i P, S teu ck art C, K ü h l M. (1996). A n arnpero- rne tric m ic ro sen so r for th e d e te rm in a tio n of H 2S in aq u a tic e n v iro n m en ts . A n a l C hem 68: 4351 -4 3 5 7 .

K alan e tra KM, H u sto n SL, N e lso n DC. (2004). N ovel, a tta ch ed , su lfu r-o x id iz in g b ac te ria at sh a llo w hyd ro - th e rm a l v en ts p o ssess v acu o le s n o t in v o lv e d in re sp ira to ry n itra te a ccu m u la tio n . A p p l E nviron M icrob io l 70: 7487-7496 .

K anagaw a T, K am agata Y, A ruga S, K ohno T, H orn M, W agner M. (2000). P h y lo g en e tic an a ly s is of an d o lig o n u c leo tid e p robe d ev e lo p m en t for E ikelboom ty p e 021N filam en to u s b ac te ria iso la te d from b u lk in g a c tiv a ted sludge. A p p l E nv iron M icrob io l 66: 5043 -5052 .

K ojim a H, Teske A, F u k u i M. (2003). M orpho log ica l a n d p h y lo g en e tic ch a rac te riz a tio n s o f fre sh w a te r T h io p lo ca sp ec ie s from Lake B iw a, Japan , a n d Lake C o n stan ce , G erm any. A p p l E nv iron M icrob io l 69: 390 -398 .

K u en en JG. (2005). G enus V. T h io b ac te riu m (ex Janke 1924) la R iv iere a n d K u en en 1989b, 496w (Effective p u b lic a tio n : la R iv iere a n d K u en en 1989a, 1838). In: B ren n e r DJ, K rieg NR, S ta ley JT, G arrity GM (eds). B ergey’s M a n u a l o f S y s te m a tic B ac terio lo g y 2 n d edn . S p ringer: N ew York, NY. p 169.

L ackey JB, L ackey EW. (1961). T he h a b ita t a n d d e sc r ip tio n of a n e w g en u s of s u lp h u r b ac te riu m . ƒ Gen M icrobio l 26: 2 9 -39 .

L ackey JB, L ackey EW, M organ GB. (1965). T axonom y an d eco logy of th e su lfu r bac te ria . F lorida E ng ineering In d u s tr ia l E xp er im en ta tio n S ta tio n , S ta te U n ivers ity o f F lorida, B u lle tin Series N o 119, G ainesville , FL 29: 1 -23 .

L ich tsch lag A , F e ld e n J, B rü ch e rt V, B oetiu s A, de B eer D. (2010). G eo ch em ica l p ro cesses a n d c h e m o sy n th e tic p r im a ry p ro d u c tio n in d iffe ren t th io tro p h ic m a ts o f th e H a ko n M o sb y M u d V olcano (B aren ts Sea) L im nol O ceanogr XX.

L ö sek an n T, R obado r A, N iem an n H, K n itte l K, B oetiu s A, D u b ilie r N. (2008). E n d o sy m b io ses b e tw een b ac te ria

The ISME Journal

Thiobacterium matsS Grünke e t s /

an d d eep -sea s ib o g lin id tu b ew o rm s from an A rctic co ld seep (H aakon M osby M u d V olcano, B aren ts Sea). E nviron M icrob io l 10: 3237-3254 .

L u d w ig W, S tru n k O, W estram R, R ich te r L, M eier H, Y ad h u k u m ar e t al. (2004). ARB: a so ftw areen v iro n m e n t for seq u en ce data . N u c le ic A c id s R es 32: 1363-1371 .

M anz W, A m an n R, L udw ig W, W agner M, S ch le ife r KH. (1992). P h y lo g en e tic o lig o d eo x y n u c leo tid e p ro b es for th e m ajo r su b c lasses o f p ro teo b ac te ria : p ro b lem s a n d so lu tio n s . S y s t A p p l M icrob io l 15: 593 -600 .

M ills HJ, M artin ez RJ, S tory S, S obecky PA. (2004). Id en tif ic a tio n of m em bers of th e m e tab o lica lly ac tive m ic ro b ia l p o p u la tio n s a sso c ia te d w ith B eggiatoa spec ies m at co m m u n itie s from G u lf o f M exico cold- seep sed im en ts . A p p l E nviron M icrobiol 70: 5447-5458 .

M iy o sh i M. (1897). S tu d ie n ü b e r d ie S ch w efe lra sen b il- d u n g u n d d ie S ch w efe lb ac te rien d e r T h e rm e n v o n Y urnoto b e i N ikko. J College Sei, Im p Univ, Japan, T okyo 10: 143 -173 .

M o lisch H. (1912). N eue farb lose S chw efe lbak te rien . Z en tra lb la tt fü r B akterio log ie , P a ra siten ku n d e , In fe k ­tio n sk ra n k h e ite n u n d H yg iene , A b t 2 33: 5 5 -62 .

M osso M A , S án ch ez MC, de la R osa MC. (2002). M icrob ia l d iv e rs ity o f th e m in e ra l w a te r sp rin g s o f A lharna de G ran ad a S pa tow n . A n a l R ea l A c a d N a c F arm 6 8 : 4 6 -7 5 .

M u ß m an n M, S ch u lz H N , S tro tm an n B, K jæ r T, N ie lsen LP, R osselló -M ora RA e t al. (2003). P h y lo g en y a n d d is tr ib u tio n of n itra te -s to rin g B eggiatoa spp . in coasta l m a rin e sed im en ts . E nviron M icrob io l 5: 523 -533 .

M u y zer G, de W aal EC, U itte r l in d e n AG. (1993). P ro filing of co m p lex m ic ro b ia l p o p u la tio n s by d e n a tu r in g g rad ien t gel e lec tro p h o re s is an a ly s is o f po ly m erase ch a in reac tio n -am p lified genes co d in g for 16S rRNA. A p p l E nv iron M icrob io l 59: 695 -7 0 0 .

M u y zer G, Teske A, W irsen CO, Jan n asch HW. (1995). P h y lo g en e tic re la tio n sh ip s of T h io m icro sp ira sp ec ies an d th e ir id e n tif ic a tio n in d eep -sea h y d ro th e rm a l v en t sam p le s by d e n a tu r in g g rad ien t gel e lec tro ­p h o re s is o f 16S rD NA fragm en ts. A rc h M icrob io l 164: 165-172 .

N ag an u m a T, Ikem oto E, S u k izak i S, T suji Y, H o tta H. (1990). S u lfu r b ac te ria o rig in a tin g from th e w a te r a n d organism s in a h y d ro therm ally active area of th e M id- O kinaw a Trough. J O ceanograph Soc Japan 46 :1 1 1 -1 1 7 .

N arváez-Z apa ta JA, R odríguez-A vila N, O rtega-M orales BO. (2005). M eth o d for recovery of in ta c t DNA for co m m u n ity a n a ly s is of m arin e in te r t id a l m ic ro b ia l b io film s. M ol B io tech n o l 30: 5 1 -56 .

N ie lsen LP, R isg aard -P e te rsen N , F oss ing H, C h ris ten sen PB, S ayam a M. (2008). E lec tric cu rre n ts in m a rin e sed im en t. T h e 12th In te rn a tio n a l S y m p o s iu m on M icrob ia l E co logy—ISM E12: M icrob ia l D ivers ity— S u s ta in in g th e B lu e P la n e t C airns , A u stra lia .

N u ssb au m er AD, F ish e r CR, B righ t M. (2006). H orizo n ta l en d o sy m b io n t tra n sm iss io n in h y d ro th e rm a l v en t tu b ew o rm s. N a tu re 441: 345-348 .

P ecc ia J, M arch an d EA, S ilv e rs te in J, H e rn an d ez M.(2000). D ev e lo p m en t a n d a p p lic a tio n of sm a ll-su b u n it rRN A p ro b es for a sse ssm en t of se lec ted T h io b a c illu s sp ec ie s a n d m em b ers of th e g en u s A c id ip h iliu m . A p p l E nviron M icrob io l 6 6 : 3065-3072 .

P ru esse E, Q u ast C, K n itte l K, F u c h s BM, L udw ig W, P ep lie s J e t al. (2007). SILVA: a c o m p reh en siv e o n lin e re so u rce for q u a lity ch eck ed a n d a lig n ed ribosonra l RNA seq u en ce d a ta co m p a tib le w ith ARB. N u c le ic A c id s R es 35: 7188-7196 .

R av ensch lag K, Sahrn K, A m an n R. (2001). Q u an tita tiv e m o lecu la r a n a ly s is o f th e m ic ro b ia l co m m u n ity in m a rin e a rc tic se d im e n ts (Svalbard). A p p l E nviron M icrob io l 67: 387-395 .

R evsbech NP, W ard DM. (1983). O xygen n ric ro e lec tro d e th a t is in se n s itiv e to m e d iu m ch em ica l co m p o sitio n : u se in an ac id m ic ro b ia l m a t d o m in a te d by C ya n id iu m ca ldarium . A p p l E nv iron M icrob io l 45: 755-759 .

S ayam a M, R isgaard -P e te rsen N, N ie lsen LP, F oss ing H, C h ris te n se n PB. (2005). Im p ac t o f b ac te ria l N O 3 tra n sp o rt on se d im e n t b iogeochenris try . A p p l E nviron M icrob io l 71: 7575-7577 .

S ch em in zk y F, K las Z, Job C. (1972). Ü ber d as V orkom m en vo n T h io b a c ter iu m b o v is ta in T h erm a lw ässe rn . In t R evu e ges H ydrob io l 57: 801 -813 .

S ch u lz HN. (2006). T he G enus Thiom argarita . In: D w ork in M , F a lkow S, R osenberg E, S ch le ife r KH, S tack eb ran d t E (eds). T h e P rokaryotes, 3 rd ed n . S pringer: N ew York, p p 1156-1163 .

S ek i H, N ag an u m a T. (1989). G row th ch a rac te ris tic s of T h io b a c ter iu m sp . from th e p lu m e o f h y d ro th e rm a l ven ts o f th e N o rth F iji B asin . M ar E col Prog S er 54: 199-202 .

S n a id r J, A m an n R, H u b er I, L u d w ig W, S ch le ife r KH. (1997). P h y lo g en e tic a n a ly s is a n d in s itu id e n tif ic a tio n of b ac te ria in a c tiv a ted sludge . A p p l E nv iron M icrobio l 63: 2884-2896 .

Teske A , N elso n DC. (2006). T he G enera B eggiatoa a n d T hiop loca . In: D w ork in M , F a lk o w S, R osenberg E, S ch le ife r KH, S tack eb ran d t E (eds). T h e P rokaryo tes, 3 rd ed n . S pringer: N ew York, p p 784-810 .

Teske A, R am sing NB, K üver J, F oss ing H. (1995). P h y lo g en y of T h io p lo ca a n d re la te d filam en to u s su lfid e -o x id iz in g bac te ria . S y s t A p p l M icrob io l 18: 517 -526 .

V ouk V, K las Z, S ch em in zk y F. (1967). K renob io log ische U n te rsu c h u n g e n an d e r G aste in e r T h erm e I. Die S ch w efe lb ak te rien des A llo p h an -H y d ro g e ls . S itz ber O steri' A k a d W iss, M a th n a t Kl, A b t 1 176: 21 -43 .

W agner M , A m an n R, K äm pfer P, A ssm u s B, H a rtm a n n A, H u tz ie r P e t al. (1994). Id en tif ic a tio n a n d in situ d e te c tio n of g ram -negative filam en to u s b ac te ria in a c tiv a ted sludge. S y s t A p p l M icrob io l 17: 4 0 5 ^ 1 7 .

W alln er G, A m an n R, B eisker W. (1993). O p tim iz in g flu o re scen t in s itu h y b rid iz a tio n w ith rR N A -targeted o lig o n u c leo tid e p ro b es for flow cy tom etric id e n tif ic a tio n of m icroo rgan ism s. C yto m etry 14: 136-143 .

W ang Q, G arrity GM, T ied je JM, C ole JR. (2007). N aïve B ayesian c la ssif ie r for ra p id a ss ig n m en t o f rRNA seq u en ces in to th e n ew b ac te ria l taxonom y. A p p l E nviron M icrob io l 73: 5261 -5267 .

Yang E, S c h a p e rd o th I, A lb rech t H, F reem an KH, M aca lady JL. (2008). G eonricrob io logy a n d h o p a n o id co n ten t o f su lfid ic su b su rface v en t b io film s, L ittle Salt S pring , F lo rid a . A b str Eos T rans AGU, 89 (53), F all M eet S u p p l, ab str B53C-0517.

Supplem entary Information accompanies the paper on The ISME Journal website (http://w w w .nature.com /ism ej)

1043

The ISME Journal