RESEARCH ARTICLE

Analysis of community composition of sulfur-oxidizing bacteria in hypersaline and soda lakes using soxB as a functional molecular marker Tatjana P. Tourova1, Natalija V. Slobodova2, Boris K. Bumazhkin2, Tatjana V. Kolganova2, Gerard Muyzer3 & Dimitry Y. Sorokin1,4 1

Winogradsky Institute of Microbiology, Russian Academy of Sciences, Moscow, Russia; 2Centre ‘Bioengineering’, Russian Academy of Sciences, Moscow, Russia; 3Department of Aquatic Microbiology, Institute for Biodiversity and Ecosystem Dynamics, University of Amsterdam, Amsterdam, The Netherlands; and 4Department of Biotechnology, Delft University of Technology, Delft, The Netherlands

Correspondence: Tatjana Tourova, Winogradsky Institute of Microbiology, Russian Academy of Sciences, P-t 60-Letiya Oktyabrya, b. 7/2, Moscow 117312, Russia. Tel.: +79 036 882 605; fax: +74 991 356 530; e-mail: [email protected]

MICROBIOLOGY ECOLOGY

Received 1 November 2012; revised 6 December 2012; accepted 10 December 2012. Final version published online 31 December 2012. DOI: 10.1111/1574-6941.12056 Editor: Riks Laanbroek

Abstract The diversity of soxB gene encoding a key enzyme of the Sox pathway sulfate thiohydrolase has been investigated in pure cultures of various halophilic and haloalkaliphilic sulfur-oxidizing bacteria (SOB) and in salt and soda lakes in southwestern Siberia and Egypt. The gene was detected in the majority of strains belonging to eleven SOB genera excluding members of genera Thiohalospira and Thioalkalimicrobium. The uncultured diversity of soxB in salt and soda lakes was low with a majority of detected sequences belonging to autotrophic SOB from the Gammaproteobacteria. In addition, the soxB analysis allowed detection of putative heterotrophic Gamma- and Alphaproteobacterial SOB yet unknown in culture. All clone libraries obtained from soda lakes contained soxB belonging to the genus Thioalkalivibrio in agreement with the cultivation results. Besides, representatives of the genera Halothiobacillus, Marinobacter, and Halochromatium and of the family Rhodobacteraceae have been detected in both type of saline lakes.

Keywords sulfur oxidation; phylogeny; diversity; soxB; soda lakes; Thioalkalivibrio.

Introduction The last decade of intensive microbiological investigation of the sulfur cycle in hypersaline inland lakes with neutral and alkaline pH revealed an unexpectedly rich diversity of novel halophilic and haloalkaliphilic chemo- and phototrophic sulfur-oxidizing bacteria (SOB). In particular, five new genera of halophilic (Thiohalomonas, Thiohalobacter, Thiohalophilus, Thiohalorhabdus, and Thiohalospira) and four genera of haloalkaliphilic obligately chemolithoautotrophic SOB (Thioalkalimicrobium, Thioalkalivibrio, Thioalkalibacter, and Thioalkalispira) have been described (Sorokin et al., 2006; Sorokin, 2008; Grant & Sorokin, 2011; Sorokin et al., 2011) as well as several heterotrophic (Sorokin, 2003) and purple anoxygenic phototrophic SOB (Gorlenko, 2007). In parallel, evidences on the presence of various groups of prokaryotes in saline lakes were accumulating using a culture-independent approach. ª 2012 Federation of European Microbiological Societies Published by Blackwell Publishing Ltd. All rights reserved

Most commonly used analysis of microbial community based on the 16S rRNA gene revealed substantial diversity in hypersaline alkaline lakes from Kenya (Rees et al., 2004), Inner Mongolia (Ma et al., 2004), Egypt (Wadi Natrun; Mesbah et al., 2007), and California (Mono Lake; Humayoun et al., 2003). However, in many cases, functionally very important groups of bacteria and archaea are present in very low numbers (either because of sharp population dynamics or because of low growth yield), rendering them undetectable by the 16S rRNA gene–based phylogenetic surveys. One such example is the SOB, detection of which in situ might be more successful using functional molecular markers encoding key enzymes of specific metabolic pathways. The general detection of autotrophs in salt lakes has already been attempted using genes encoding autotrophic carbon assimilation by the Calvin–Benson cycle (RuBisCO large subunite, cbbL/M). The results of this analysis indicated a domination of FEMS Microbiol Ecol 84 (2013) 280–289

281

Sulfur-oxidizing bacteria in hypersaline and soda lakes

Gammaproteobacterial autotrophic SOB in hypersaline neutral and soda lakes in Kulunda Steppe (Russia; Tourova et al., 2010, 2011; Kovaleva et al., 2011) and in alkaline Mono Lake (California; Giri et al., 2004). More specific detection of SOB might be achieved using functional genes of the sulfur-oxidation pathways as molecular markers. The multi-enzyme Sox pathway has been demonstrated to operate in a wide range of photo- and chemotrophic SOB that oxidize sulfide and thiosulfate to sulfate. A fully functional Sox complex involves SoxB, SoxXA, SoxYZ, and SoxCD components (Friedrich et al., 2005). While many proteobacterial SOB (especially within the Gammaproteobacteria) with confirmed phenotype, as well as the sequenced genomes that contain the Sox genes, may lack the SoxCD component (functioning as a sulfur dehydrogenase), and SoxB is an universally indispensible part of this complex (Ghosh & Dam, 2009). It is represented by a unique di-manganese containing enzyme acting as sulfate thiohydrolase at the final stage of thiosulfate oxidation to sulfate. Therefore, it is considered as an appropriate functional molecular marker for the detection of SOB directly in the environment. A recent development of a primer system based on soxB gene as a functional marker for the detection of SOB has allowed a broad view on the distribution of this gene among SOB and in the genomes of other bacteria, which were not known to be able to oxidize sulfur compounds (Petri et al., 2001; Meyer et al., 2007). Furthermore, the system has been used for functional analyses of SOB directly in some microbial communities, such as coastal aquaculture (Krishnani et al., 2010), marine sediments (Lenk et al., 2012), bioreactors (Luo et al., 2011), and hydrothermal vents (Hugler et al., 2010). In this work, we tested a collection of pure cultures of halophilic and haloalkaliphilic SOB strains isolated from hypersaline chloride–sulfate and soda lakes for the presence of soxB genes. In addition, a culture-independent study of the diversity of soxB genes in the sediments of typical inland hypersaline chloride–sulfate and of soda lakes was performed to assess the uncultured diversity of halo(alkali)philic SOB populations.

Materials and methods Bacterial strains and environmental samples

Twenty-eight strains of halo(alkali)philic SOB used in this study and represented by 11 genera of the Alpha- and Gammaproteobacteria were described previously (Gorlenko et al., 2004; Sorokin et al., 2006; Sorokin, 2008) and are maintained in our active culture collection (Table 1). For the environmental detection of soxB genes, mixed sediment samples from various hypersaline neutral and alkaline (soda) lakes from Kulunda Steppe (Altai, Russia) FEMS Microbiol Ecol 84 (2013) 280–289

and Wadi Natrun (Libyan desert, Egypt) were used. The lakes in Kulunda Steppe were sampled in 2005, 2007, and 2009, and the lakes in Egypt, in 2001. General characteristics of the lakes are given in Table 2. After arrival to the laboratory, the sediments from soda lakes were washed in neutral NaCl solutions with the salinity corresponding to the native until the pH was brought down to neutral. The sediments from salt lakes were washed with 4 M NaCl once. Finally, 0.5 cm3 of fine colloidal fraction of the sediments was separated from the residual course sediments by low-speed centrifugation and frozen at 80 °C until further DNA extraction. DNA extraction

Genomic DNA from pure SOB cultures was extracted with a modified Promega Wizard technology (Promega, Madison, WI). DNA extraction from the sediments was performed with Power DNA Isolation Kit (Mo Bio Laboratories Inc., Carlsbad, CA) according to the manufacture’s instructions except that before the DNA isolation, 200 lL of 0.1 M AlNH4(SO4)2 was added to remove potential PCR inhibitors, such as humic acids (Braid et al., 2003). Amplification of the soxB genes from pure cultures and environmental samples

The soxB gene was amplified with three primer pairs: soxB432F–1446R, soxB693F–1446R, and soxB693F–1164R (Petri et al., 2001; see Table 1 for details of PCR results; potential contamination of the examined reference strains was excluded by preliminary partial 16S rRNA gene sequence analysis). PCR amplification was performed as a two-step PCR in a total volume of 25 lL PCR mixture containing DNA-buffer (67 lM Tris–HCl, pH 8.8; 17 lM (NH4)2SO4); 1.5 lM MgCl2, 5 nmol each of dNTP, 0.5 lM of each primer, 2.5 U Taq DNA polymerase, and 10–100 ng genomic DNA from the reference strains as template. The PCR conditions were as follows: initial step 2 min 94 °C, 10 cycles of 94 °C for 30 s, 55 °C for 30 s, 72 °C for 40 s, followed by 25 cycles of 94 °C for 30 s, 47 °C for 30 s, 72 °C for 40 s, and final extension at 72 °C for 7 min. The PCR products of expected size were purified through low-melting agarose using Wizard PCR Preps kit and sequenced directly (for pure cultures) or used for cloning (sediments). Clone library construction

Clone libraries were constructed with the pGEM-T vector system (Promega) and Escherichia coli DH10b competent cells according to the manufacture’s protocol. From each clone library, 50 clones were selected randomly and ª 2012 Federation of European Microbiological Societies Published by Blackwell Publishing Ltd. All rights reserved

282

T.P. Tourova et al.

Table 1. List of pure cultures of halo(alkali)philic SOB tested for the presence of soxB PCR product SOB Species Halophilic SOB Thiohalorhabdus denitrificans Thiohalospira halophila

Thiohalobacter thiocyanaticus Thiohalophilus thiocyanatoxidans Thiohalomonas denitrificans Halothiobacillus sp. Haloalkaliphilic SOB Thioalkalvibrio nitriatis Thioalkalivibrio paradoxus Thioalkalivibrio thiocyanoxidans Thioalkalivibrio thiocyanoxidans Thioalkalivibrio versutus Thioalkalivibrio denitrificans Thioalkalivibrio nitratireducens Thioalkalivibrio halophilus Thioalkalivibrio jannaschii Thioalkalibacter halophilus Thioalkalimicrobium sibiricum Thioalkalimicrobium aerophilum Thioalkalimicrobium cyclicum Ectothiorhodosinus mongolicus Roseinatronobacter thiooxidans

Strain

soxB432–soxB1446

HLD10 HLD18 HL3 HL21 HL23 HL25 HL10 HL11 HLgr2 HRh1 HRhD2 HLD15 HL6 HL7

+ +

ALJ12 ARh1 ARh2 ARh4 AL2 ALJD ALEN2 HL17 ALM2 ALCO1 AL7 AL3 ALM1 M9 ALG1

soxB693–soxB1446

soxB693–soxB1164

Product size (bp)

926 796 – – – – – – – 665 738 998 696 704

w w +

+ + nd + +

+ + nd + +

w +

+ w + + w + w + + +

+

w w w + + w w −

701 791 667 736 446 677 777 823 687 757 – – – 755 740

+ + + + + + +

w + +

nd nd

+, signal; −, no signal; w, weak signal; nd, not determined. Table 2. Characteristics of the lakes Brine characteristics Sample name

Sample date

05-3

2005

4KL 6KL 2KL 14KL WN

2009 2009 2009 2007 2001

Area

Lakes

Kulunda Steppe (Altai, Russia)

Bitter Lake System, mix from 3 lakes Bitter-1 Tanatar-5 Cock Salt Lake Burlinskoe Mix from 8 lakes

Wadi Natrun (Egypt)

screened for the presence of inserts by PCR using the vector-specific M13-specific primer pair. Comparative sequence analysis

Preliminary analysis of the sequences was carried out with BLAST (www.ncbi.nlm.nih.gov/blast/). The nucleotide and ª 2012 Federation of European Microbiological Societies Published by Blackwell Publishing Ltd. All rights reserved

Salinity (g L 1)

Soluble carbonate alkalinity (M)

Lake type

pH

Soda lakes

10.2–10.4

30–130

0.2–1.2

Hypersaline chloride–sulfate lakes Alkaline hypersaline lakes

10.16 10.1 7.7 7.45 9.5–10.1

300 180 300 360 200–360

4.70 1.60 – – 0.2–1.5

inferred amino acid sequences were aligned with sequences from GenBank using CLUSTALW (Thompson et al., 1994). Phylogenetic trees were reconstructed using two different algorithms: neighbor-joining in the TREECONW program package (Van de Peer & De Wachter, 1994) and maximum-likelihood using MEGA5 software (Tamura et al., 2011). A homologous coverage of clone FEMS Microbiol Ecol 84 (2013) 280–289

Sulfur-oxidizing bacteria in hypersaline and soda lakes

libraries was estimated according to Singleton et al., 2001. The GenBank/EMBL/DDBJ accession numbers for the sequences reported in this article are as follows: KC017785–KC017824.

Results and discussion Detection of soxB genes in pure cultures of halo(alkali)philic SOB

In general, the amplification with soxB693F/soxB1446R and soxB693F/soxB1164R primer pairs resulted in a single, correct-sized PCR product (c. 750 and 470 bp, respectively), whereas the primer pair soxB432F/soxB1446R frequently generated no amplicons, weak amplicons, or amplicons with ambiguous sequencing results (Table 1). The truly successful amplification with this primer pair was shown only for Thiohalorhabdus strains. Negative amplification results with all three primer sets were obtained for several proven SOB species including Thiohalospira halophila (7 strains) and Thioalkalimicrobium (3 strains), which may be caused either by inhibited primer annealing or by the absence of this gene in those SOB. For Thioalkalimicrobium, it was definitely caused by the first reason, because soxB genes were detected in the complete genomes of two Thioalkalimicrobium species (T. aerophilum and T. cyclicum). The nucleotide sequences of their soxB genes are highly divergent from those used to develop the existing primer sets, which was probably the main reason for the amplification failure. In case of Thiohalospira, however, it is possible that soxB is absent in the genome, similar to another member of the family Ectothiorhodospiraceae–Thiorhodospira sibirica. Phylogenetic analysis of soxB genes in pure cultures of halo(alkali)philic SOB

The SoxB-based phylogeny significantly overlaps with the classical 16S rRNA gene–based phylogeny, although obvious examples of horizontal transfer are well documented (Petri et al., 2001; Meyer et al., 2007). According to the 16S rRNA gene–based phylogeny, the most known halo (alkali)philic SOB genera represent deep lineages within the Gammaproteobacteria not related to each other or to other groups (Fig. 1). Among them, only the genera Halothiobacillus and Thioalkalibacter (Halothiobacillaceae) and Thioalkalivibrio and Ectothiorhodosinus (Ectothiorhodospiraceae) are currently classified as members of the order Chromatiales, while the other groups remain unassigned. In particular, most groups of recently discovered halophilic SOB, such as Thiohalorhabdus, Thiohalobacter, and Thiohalophilus branched as novel deep lineages within the Gammaproteobacteria in the SoxB tree (Fig. 2) FEMS Microbiol Ecol 84 (2013) 280–289

283

with no relatives in the soxB gene database, while Thiohalobacter thiocyanaticus formed a single cluster with an nondescribed marine SOB isolate NDII1.2. Two salt lake isolates Halothiobacillus sp. HL6 and HL7 (with 98.7% of 16S-rRNA gene interstrain sequence similarities and 97.8– 97.9% with type strains of H. halophilus and H. hydrothermalis) formed a new branch within the Halothiobacillus SoxB cluster (88.2% of amino acid identity with the soxB sequence of H. hydrothermalis). Also in line with the 16S rRNA gene-based tree, Thioalkalibacter halophilus belonged to a separate branch of Halothiobacillus SoxB cluster (Fig. 2). The haloalkaliphilic phototrophic SOB E. mongolicum formed a separate branch within the family Ectothirhodospiraceae both on the 16S rRNA gene–based and SoxBbased trees, in contrast to the chemolithotrophic genus Thioalkalivibrio – a dominant culturable taxon of haloalkaliphilic SOB in soda lakes, which exemplified a case of incongruence between SoxB- and 16S rRNA gene–based phylogenies. According to the latter, the Thioalkalivibrio species formed a monophyletic cluster within Ectothirhodospiraceae divided into three subclusters: (1) T. versutus, T. jannaschii, T. nitratis, T. thiocyanoxidans, and T. halophilum; (2) T. nitratireducens and T. paradoxus; and (3) T. denitrificans, T. thiocyanodenitrificans, and T. sulfidophilus (Fig. 1). At the same time, soxB genes of the genus Thioalkalivibrio were not monophyletic, although their division into independent clusters partly correlated with the phylogenetic divergence revealed by the 16S rRNA gene analysis. Clusters 2 and 3 and partly the largest cluster 1 formed separated branches within Ectothirhodospiraceae, but some strains (including the type strain of the type species T. versutus) of cluster 1 formed an additional cluster of uncertain position within the Gammaproteobacteria (Fig. 2). One of the three published genomes of Thioalkalivibrio, Thioalkalivibrio sp. K90mix, contains two significantly different copies of soxB gene (Muyzer et al., 2011) – a so far unique example – although such multiplicity of other functional genes is widespread in prokaryotes. For example, in autotrophs, three complete cbbL/M operons in one genome coding fully functional RuBisCO isoforms with different specialization are known (Yoshizawa et al., 2004). The soxB genes of the Thioalkalivibrio cluster 1 were grouped around one of the two different soxB genes of Thioalkalivibrio sp. K90mix. Judging from the similarity of nucleotide composition of soxB gene copies between each other and with the total genome (G + C 62–65 mol%), it looks as a gene duplication rather than a lateral transfer. Assuming further that in the other Thioalkalivibrio species of cluster 1, one of the two ancestral soxB genes has been lost, the two resulting clusters of species with a single soxB copy have phylogeny incongruent with their position on the 16S rRNA gene ª 2012 Federation of European Microbiological Societies Published by Blackwell Publishing Ltd. All rights reserved

284

T.P. Tourova et al.

Cupriavidus taiwanensis LMG 19424, NR_028800

0.05 100 (99)

96 (98)

Comamonas testosteroni LMG 1800, EU024141 Ramlibacter tataouinensis TTB310, NR_027534

PB

Methylibium petroleiphilum PM1 strain PM1, NR_041768

100 (99)

Sphaerotilus natans D-507, FJ871053

92

Leptothrix cholodnii SP-6, NC_010524

98 (63) 100 (99)

Acidithiobacillus ferrivorans SS3, NC_015942 Acidithiobacillus thiooxidans ATCC 19377, AJ459803

Halothiobacillus sp. HL6 (KC017785) Halothiobacillus sp. HL7 (KC017786)

94 (67) 100 (99) 98 (76)

Halothiobacillaceae

Halothiobacillus hydrothermalis R3, NR_025943

100 (99)

Thioalkalibacter halophilus ALCO1, EU124668 Halothiobacillus neapolitanus DSM 581, AF173169 Halothiobacillus kellyi BII-1, NR_025030

88 (69)

100 (99) 100 (99)

Thioalkalimicrobium cyclicum ALM1, AF329082 Thioalcalimicrobium aerophilum AL3, AF126548

100 (99)

Thiomicrospira pelophila DSM 1534, L40809 Thiomicrospira crunogena ATCC 35932T, L40810

86 100 (99)

Symbiont Candidatus Ruthia magnifica str. Cm, NC_008610 Symbiont Candidatus Vesicomyosocius okutanii HA, NC_009465

Thiothrix nivea DSM 5205, HQ823668 Leucothrix mucor ATCC 25107, HQ897925 Marinobacter sp. HY-106, AJ294336

100 (99)

Marinobacter hydrocarbonoclasticus ATCC 49840, X67022 Marinobacter sp. NP40, EU196314 Congregibacter litoralis KT71, AY007676 Endosymbiont of Riftia pachyptila (vent Ph05), NZ_AFOC01000137 Endosymbiont of Ifremeria nautilei PM-2, AB238959

74

Halochromatium salexigens DSM 4395, NR_036810

Chromatiaceae

Thiorhodococcus minor DSM 11518, FN293057

100 (99)

Thiocapsa roseopersicina DSM 217, AF113000

71 (60)

Allochromatium vinosum DSM 180, NR_044605 Thiocystis violacea DSM 207, FN293059

83 (69)

PB

Sulfur-oxidizing bacterium NDII1.1, AF170424

86 (75)

Thiohalophilus thiocyanatoxydans HRhD 2, DQ469584 Thiohalobacter thiocyanaticus HRh1, FJ482231 Thiohalomonas denitrificans HLD 15, EF455919 Ectothiorhodospira shaposhnikovi DSM243, M59151 100 (95) 100 (89)

Ectothiorhodospira mobilis DSM 4180, X93482

100 (99)

Ectothiorhodosinus mongolicus M9, AY298904 Thioalkalivibrio denitrificans ALJD, NR_028745

Ectothiorhodospiraceae-1

Cluster 3

Thioalkalivibrio sulfidophilus HL-EbGr7, EU709878 100 (99)

Thialkalivibrio nitratreducens ALEN 2, AY079010 Thioalkalivibrio paradoxus ARh 1, NR_025014 Thioalkalivibrio halophilus HL17, NR_042855 Thioalkalivibrio versutus AL2, AF126546 Thioalkalivibrio jannaschii ALM2, NR_028807 Thioalkalivibrio thiocyanoxidans ARh 2, AF302081 Thioalkalivibrio nitratis ALJ12, NR_024991

97 (70) 99 (60) 100 (99) 99 (95)

Cluster 2

Thioalkalivibrio Cluster 1

89 (79)

Thioalkalivibrio sp. K90mix, CP001905 Halorhodospira halophila SL1, NC_008789 Halorhodospira abdelmalekii DSM 2110, X93477

100 (99)

93 (91)

100 (99)

Ectothiorhodospiraceae-2

Halorhodospira halochloris DSM 1059, FR749892 100 (99)

Thiohalorhabdus denitrificans HLD 10, EU374712 Thiohalorhabdus denitrificans HLD 18, EU374711

100 (96) 99 (99) 100 (99)

Roseobacter denitrificans OCh 114, M96746 Roseovarius nubinhibens ISM, NR_028728 Sagittula stellata E-37, NR_026016 Paracoccus denitrificans PD1222, NC_008686

Rhodobacteraceae

PB

Roseinatronobacter thiooxidans ALG 1, AF249749 Rhodobacter sphaeroides ATCC 17025, CP000661 98 (77)

Rhodovulum sulfidophilum W-1S, U55277 Anaeromyxobacter dehalogenans 2CP-1, NR_027547 Sulfurimonas denitrificans DSM 1251, NC_007575

PB

100 (99) 100 (99)

Sulfurimonas autotrophica DSM 16294, NC_014506

100 (99)

Sulfuricurvum kujiense DSM 16994, CP002355

PB

Nitratiruptor sp. SB155-2, NC_009662 Pelodictyon phaeoclathratiforme BU-1, NR_044914

100 (63)

100 (99)

Chlorobium limicola ATCC 8327, M31769

100

Chlorobi

Chlorobium tepidum TLS, NC_002932 99 (97)

Chlorobaculum limnaeum DSM 1677, HE582773 Aquifex aeolicus VF5, NC_000918

100 (99)

Thermocrinis albus DSM 14484, NR_025414 100 (97)

Aquifica

Hydrogenobacter thermophilus TK-6, Z30214

Meiothermus silvanus DSM 9946, NR_027600

77 (98)

Meiothermus ruber DSM 1279, NR_027614

Deinococci

Thermus thermophilus HB8, NR_037066

100 (99) 100 (99)

Thermus scotoductus SA-01, EU330195

Fig. 1. Phylogenetic tree based on the 16S rRNA gene sequences of the soxB gene-containing SOB investigated halo(alkali)philic and reference strains. The sequences obtained in this study are in bold, and the strains for which soxB gene sequences were determined in this study are underlined. The sequence of Thioalkalivibrio sp. K90mix is in frame. Tree topography and evolutionary distances are given by the neighbor-joining method with Jukes–Cantor distances. All positions with < 95% site coverage were eliminated. That is, fewer than 5% alignment gaps, missing data, and ambiguous bases were allowed at any position. There were a total of 1209 positions in the final data set. Numbers at the nodes indicate the percentage of bootstrap values for the clade in 1000 replications (the values for the maximum-likelihood method are given in parentheses). Only values above 70% are shown.

ª 2012 Federation of European Microbiological Societies Published by Blackwell Publishing Ltd. All rights reserved

FEMS Microbiol Ecol 84 (2013) 280–289

285

Sulfur-oxidizing bacteria in hypersaline and soda lakes

100 (95) 95 (78) 100 (99)

Chlorobaculum limnaeum 1677, EF618591 Chlorobium limicola DSM 257, EF618579

Chlorobi

Chlorobium tepidum TLS, AAM72255 Pelodictyon phaeoclathratiforme BU-1, YP_002017723 Congregibacter litoralis KT71, ZP_01102556 100 (99) Thioalkalimicrobium cyclicum ALM1, YP_004537084

0.1

90 (96)

Thioalkalimicrobium aerophilum AL3, 82801 Thiomicrospira crunogena XCL-2, YP_391815 Symbiont Candidatus Vesicomyosocius okutanii HA, YP_001219016 Symbiont Candidatus Ruthia magnifica Cm, ABL01948

99 (99)

100 (99)

Marinobacter sp. NP40, EU196348 100 (99)

Hipersaline lake sediment clones 2KL-otu5 (n = 2) (KC017815) Hipersaline lake sediment clones 2KL-otu3 (n = 11) (KC017817) Hipersaline lake sediment clones 2KL-otu1 (n = 13) (KC017819)

70 (70)

Marinobacter sp. HY-106, CAC82473

Hipersaline lake sediment clones 2KL2-otu4 (n = 4) (KC017816) Thiocystis violacea DSM 207, EF618573

PB

Thiocapsa roseopersicina DSM 217, EF618576 Thiorhodococcus minor DSM 11518, EF618606 Allochromatium vinosum DSM 180, NC_013851

Chromatiaceae

Soda lake sediment clone 05-3-otu5 (KC017806)_

95 (84)

77 (81)

79 (77)

70 (74)

Halochromatium salexigens DSM 4395, EF618598 Halochromatium glycolicum DSM 11080, EF618605 Endosymbiont of Ifremeria nautilei PM-2 , ABR67384 Endosymbiont of Riftia pachyptila , (vent Ph05), ZP 08830136

72 (67)

Soda lake sediment clones 05-3-otu1 (n = 45) (KC017810) Thiohalomonas denitrificans HLD15 (KC017787) Thiohalophilus thiocyanoxidans HRhD2 (KC017805) Soda lake sediment clones WN-otu4 (n = 2) (KC017821) Leptothrix cholodnii SP-6, ACB36032 Sphaerotilus natans 507, AEI59687

100 (99) 70 (60)

Methylibium petroleiphilum PM1, ABM95387 Ramlibacter tataouinensis TTB310, YP_004618489 Comamonas testosteroni ATCC 11996, EHN65669 Cupriavidus taiwanensis LMG 19424, NC_010528 Anaeromyxobacter dehalogenans 2CP-1, YP_002492081 Leucothrix mucor DSM 621, EF618580

87 (90) 79 (76)

100 (99)

PB PB

Thiothrix nivea DSM 5205, EF618600 Sulfur-oxidizing bacterium NDII1.2, ABR67386

97 (88) 70 (63)

Thiohalobacter thiocyanaticus HRh1 (KC017804) Sulfur-oxidizing bacterium str. manganese crust, ABR67385

Soda lake sediment clones WN-otu2 (n = 8) (KC017823) Soda lake sediment clones WN-otu3 (n = 5) (KC017822) Thioalkalivibrio thiocyanoxidans ARH4 (KC017789)

100 (99)

Thioalkalivibrio sp. K90mix, soxB-2, ADC71659

83 (77) 95 (63) 95 (91)

Thioalkalivibrio sp. K90mix, soxB-1, ADC71142

100 (93) 98 94

100 (97)

100 (98)

98 (97)

Thioalkalivibrio jannaschii ALM2 (KC017790) Thioalkalivibrio versutus AL2 (KC017793) Thioalkalivibrio halophilus HL17 (KC017800) Soda lake sediment clones WN-otu1 (n = 10) (KC017824)

Thioalkalivibrio thiocyanoxidans ARH2 (KC017796) Thioalkalvibrio nitriatis ALJ12 (KC017797) Soda lake sediment clones 6KL-otu1 (n = 34) (KC017820) Soda lake sediment clone 05-3-otu4 (KC017807) Thioalkalivibrio nitratireducens ALEN2 (KC017801) Thioalkalivibrio paradoxus ARh 1 (KC017788)

Cluster 1b

Cluster 1a

Ectothiorhodospiraceae-1

Cluster 2

PB

Ectothiorhodospira mobilis DSM 4180, EF618594 Ectothiorhodospira shaposhnikovii DSM 243, EF618578 84 (76)

Ectothiorhodosinus mongolicus M9 (KC017792) Thioalkalivibrio sulfidophilus HL-EbGr7, NC_011901

Thioalkalivibrio denitrificans ALJD (KC017802) Soda lake sediment clones 05-3-otu3 (n = 2) (KC017808) Thioalkalibacter halophilus ALCO1 (KC017794)

100 (98) 100 (99)

Halothiobacillus kellyi DSM 13162, EF618609 Halothiobacillus neapolitanus DSM 581, AJ294332 Halothiobacillus hydrothermalis DSM 7121, AJ294325

76 (76) 97 (61) 70 (75) 99 (99) 96 (67)

82

100 (99)

Cluster 3

Halothiobacillaceae

Hipersaline lake sediment clones 2KL-otu2 (n = 12) (KC017818) Halothiobacillus sp. HL7 (KC017795) Halothiobacillus sp. HL6 (KC017803) Acidithiobacillus ferrivorans SS3, AEM48586 Acidithiobacillus thiooxidans ATCC 19377, ZP_09995367

100 (99) 100 100 (99)

Hipersaline lake sediment clones 14KL-long-otu1 (n = 15) (KC017814) Thiohalorhabdus denitrificans HLD10 (KC017799) Thiohalorhabdus denitrificans HLD18 (KC017798) Ectothiorhodospiraceae-2 Halorhodospira halophila SL1, ABM62703 Rhodovulum sulfidophilum, AAF99435

100 (95) 74 100 (99)

100 80 (82)

86 (92)

Roseinatronobacter thiooxidans ALG1 (KC017791) Hipersaline lake sediment clone 14KL-short-otu3 (KC017811) Soda lake sediment clones 05-3-otu2 (n = 8) (KC017809) Rhodobacter sphaeroides ATCC 17025, ABP72768 Paracoccus denitrificans PD1222, ABL72218 Sagittula stellata E-37, EBA06075

Rhodobacteraceae

PB

Hipersaline lake sediment clones 14KL-short-otu1 (n = 41) (KC017813) 100 (95) 82 (97) 83 (65)

Roseovarius nubinhibens ISM, EAP75534 Roseobacter denitrificans OCh 114, ABG31148

Hipersaline lake sediment clones 14KL-short-otu2 (n = 4) (KC017812)

90 (85) 100 (99)

70 (61)

100 (99)

100 (99)

76 (92)

100 (99)

Hydrogenobacter thermophilus TK-6, YP_003433170 Thermocrinis albus DSM 14484, ADC89396 Aquifex aeolicus VF5, NP_214237 Nitratiruptor sp. SB155-2, BAF70932 Sulfuricurvum kujiense DSM 16994, YP_004059316 Sulfurimonas autotrophica DSM 16294, YP_003892056 Sulfurimonas denitrificans DSM 1251, YP_392780 100 (99) Meiothermus ruber DSM 1279, ADD29608 Meiothermus silvanus DSM 9946, ADH65259

Aquificae PB Deinococci

Thermus scotoductus SA-01, YP_004203260 Thermus thermophilus HB8, YP_144683

Fig. 2. Phylogenetic tree based on partial soxB gene translated sequences showing position of pure cultures of halophilic and haloalkaliphilic SOB and of environmental clones retrieved from hypersaline salt and soda lakes. The sequences obtained in this study are in bold (pure cultures) and highlighted in gray (environmental clones). Two sequences of Thioalkalivibrio sp. K90mix are in frame. Tree topology and evolutionary distances are given by the neighbor-joining method with Poisson distances. All positions with < 95% site coverage were eliminated. That is, fewer than 5% alignment gaps, missing data, and ambiguous bases were allowed at any position. There were a total of 130 positions in the final data set. Numbers at the nodes indicate the percentage of bootstrap values in 1000 replications (the values for maximum-likelihood method are given in parentheses). Only values above 70% are shown.

FEMS Microbiol Ecol 84 (2013) 280–289

ª 2012 Federation of European Microbiological Societies Published by Blackwell Publishing Ltd. All rights reserved

286

T.P. Tourova et al.

tree. Furthermore, apparently, evolution of species of clusters 1a, 2, and 3 was in line with the rest of the SOB in the Ectothiorhodobacteriaceae, while in the cluster 1b, it was different, resulting in the formation of a separate soxB cluster outside the Ectothiorhodobacteriaceae. An interesting question is whether the two types of soxB present in the Thioalkalivibrio species encode enzymes with different functions. The phylogenetic position of a single investigated Alphaproteobacterial SOB Roseinatronobacter thiooxidans in the SoxB tree corresponded to its 16S-rRNA gene-based phylogeny within the Rhodobacteraceae being related to the genera Rhodobacter and Rhodovulum (Fig. 2). Detection and diversity of soxB genes in sediments of hypersaline lakes

05-3 14KL 4KL 6KL 2KL WN +, signal;

soxB693– soxB1446

+

+ +

+

soxB693–soxB1164

Product size (bp)

w w w + + +

725 876, 696 – 476 720 477

, no signal; w, weak signal.

ª 2012 Federation of European Microbiological Societies Published by Blackwell Publishing Ltd. All rights reserved

Soda lake clade I Soda lake clade II Soda lake clade III

100%

80%

60%

40%

20%

05-3

Table 3. Results of soxB amplification from environmental samples soxB432– soxB1446

Rhodobacteraceae Halothiobacillus Thioalaklivibrio

0%

The soxB gene primer system yielded positive amplification results with DNA extracted from five of six sediment samples of hypersaline (14KL and 2KL) and soda (05-3, 4KL, and WN) lakes, but the efficiency of each primer pair was different for different samples (Table 3). Therefore, clone libraries (about 50 for each sample) were constructed with PCR products obtained with the most efficient primer pairs, that is, with soxB693F/soxB1446R in case of samples 05-3, 14KL, and 2KL, and with soxB693F/soxB1164B for samples 6KL and WN. Only for 14KL, an additional library was obtained with soxB432F/ soxB1446R taking into account the results obtained with pure cultures. In each of these clone libraries, only one to five soxB phylotypes (OTUs based on a nucleotide sequence identity cut-off of 97%) could be detected. The coverage of the libraries was estimated to be 96–100%. The composition of the clone libraries constructed for each sample differed considerably. The library from the sample 05-3 (integrated sediments from moderately saline soda lakes) was dominated by a single phylotype (05-3otu1, up to 73% of the total number of clones) related to the halophilic SOB Thiohalomonas denitrificans. Such a domination of an apparently halophilic organism in soda lakes is difficult to interpret, unless it belongs to a haloalkaliphile closely related to the halophilic Thiohalomonas.

Sample

Thiohalorhabdus Halochromatium Marinobacter

6 KL

Soda lakes

WN

14 KL 14 KL long short

2 KL

Salt lakes

Fig. 3. Comparative representation of uncultured diversity of soxB gene detected in hypersaline and soda lakes.

Second in abundance (up to 18% of the total number of clones), phylotype was 05-3-otu2 belonging to the family Rhodobacteraceae within the Alphaproteobacteria. The other three minor phylotypes were related to the genera Halochromatium, Thioalkalivibrio, and Halothiobacillus within the Gammaproteobacteria (Figs 2 and 3). In the clone library from sample 6KL (surface sediments from a moderately saline soda lake), only a single phylotype belonging to the genus Thioalkalivibrio has been detected. The representatives of Thioalkalivibrio were also dominating (two phylotypes making up to 60% of the total number of clones) the clone library from sample WN (integrated surface sediment sample from hypersaline alkaline lakes). Two other phylotypes from this library belonged to two novel SoxB branches within the Gammaproteobacteria. So, all clone libraries obtained from soda lakes contained SoxB phylotypes belonging to the genus Thioalkalivibrio, which corresponds well to the cultivation results. On the other hand, the SoxB phylotypes of Thioalkalivibrio retrieved directly from the soda lake sediments were more diverse and significantly divergent from those present in pure cultures, apparently representing several new uncultured lineages within the genus Thioalkalivibrio and indicating that the cultivation efforts only touched the enormous diversity of this dominant group of soda lake chemolithoautotrophs. The soxB clone library from surface sediments of a hypersaline chloride–sulfate lake 2KL contained two groups: a dominant group (72% of the total) included four closely related phylotypes belonging to the genus Marinobacter (Petri et al., 2001; Perreault et al., 2008) FEMS Microbiol Ecol 84 (2013) 280–289

287

Sulfur-oxidizing bacteria in hypersaline and soda lakes

and the rest of the clones were closely related to Halothiobacillus sp. strains HL6 and HL7 isolated from hypersaline lakes in the same area. Two clone libraries from another hypersaline lake with neutral pH (14KL) gave significantly different SOB compositions: the short-fragment library included three phylotypes forming separate branches within the family Rhodobacteraceae, while the long-fragment library captured a single phylotype belonging to the extremely halophilic SOB of genus Thiohalorhabdus. In contrast to soda lakes, in the two chloride– sulfate lakes analyzed, common SoxB phylotypes were not observed. Comparative analysis of bacterial community in sediments of hypersaline lakes using different functional markers

The samples analyzed in this work were previously used for detection of the cbbL/M genes as markers of autotrophy (Tourova et al., 2010, 2011; Kovaleva et al., 2011). This allows a direct comparison of the results obtained with two different functional molecular markers – one for carbon assimilation pathway and another for energy metabolism. One of the interesting examples in this respect is the dominant SoxB phylotype detected in soda lakes 05-3 which, together with the halophilic SOB T. denitrificans, formed a deep-lineage cluster within the Gammaproteobacteria only distantly related to the family Chromatiaceae. A similar type of a remote cluster has been detected in the same sample using the cbbL probing (Kovaleva et al., 2011). This tempts to speculate that those two clusters based on two different functional markers belong, in fact, to the same autotrophic SOB related to the genus Thiohalomonas. However, so far, culturable representatives of this genus have been found only in chloride–sulfate lakes with neutral pH. Thus, the cluster detected by molecular methods might represent an unknown haloalkaliphilic relative of this genus. This example, once again, demonstrated that there are still many novel SOB in hypersaline lakes, which have not been isolated in pure culture. The soxB library obtained from hypersaline alkaline lakes of Wadi Natrun, similar to the cbbL library, was dominated by the Thioalkalivibrio phylotypes. On the other hand, several soxB lineages detected in WN belonged to unknown Gammaproteobacteria and were not present in the cbbL library. Either the primer systems for different markers have different efficiency, or, which seems more likely, those soxB belonged to heterotrophic SOB. The same might be true for sample 14KL, in which the soxB approach detected representatives of the Rhodobacteraceae (which contains many heterotrophic SOB; Sorokin, 2003; Lenk et al., 2012), while the cbb approach did not, FEMS Microbiol Ecol 84 (2013) 280–289

whereas the presence of lithoautotrophic Thiohalorhabdus phylotypes was shown by the both functional markers. Overall comparison of the results obtained with the two different functional markers demonstrated significant overlapping for the dominant bacteria in most of the analyzed samples, but also differences in the composition of minor phylotypes. For example, the genus Ectothiorhodosinus was detected only by the cbb approach, while the genus Halochromatium, only by the soxB probing. Both markers indicated an obligatory presence of the genus Thioalkalivibrio in soda lakes and the genus Halothiobacillus in salt lakes, in line with the cultivation approach. Absence of the other pair of closely related Gammaproteobacterial SOB (Thiomicrospira and Thioalkalimicrobium) in the soxB and cbbL libraries, while heavily represented in culture, most probably indicates their minor presence at in situ conditions. According to physiological studies, these two groups of halo(alkali)philic SOB represent fastgrowing r-strategists characterized by sharp population fluctuations (Sorokin et al., 2006), making them a difficult target for a single time-point detection. In case of Halorhodospira, however, the situation is different, because it has been detected previously by the cbb probing as one of the dominant autotrophs in samples 05-3, 14KL, and 2KL. Most probably, the failure can be explained by the substantial divergence of the soxB gene sequence of H. halophila (genome data), which forms a deep lineage within the Gammaproteobacteria outside of its position established by the traditional 16S rRNA gene– based phylogeny (Fig. 2). The same might be true for another failed case of soxB detection of extremely halophilic SOB of the genus Thiohalospira – a dominant cultured SOB in hypersaline chloride–sulfate lakes, which was also detectable with the cbb probing. It is clear that the soxB primers need updating and maybe, in case of a substantial divergence, group-specific primers would be necessary to develop. Overall, the analysis of soxB diversity in hypersaline salt and soda lakes confirmed the domination of SOB among the autotrophic bacterial communities. In addition, the soxB analysis allowed the detection of putative heterotrophic SOB yet unknown in culture. Such SOB may most probably be represented by lithoheterotrophs, similar to, for example, R. thiooxidans (Sorokin et al., 2000). Their current lack in culture may be explained by the less selective culture conditions for isolation in comparison with the autotrophic SOB. So, more focused efforts are necessary to identify this yet unknown group of halo(alkali)philic SOB.

Acknowledgements This work was supported by the RFBR (grants 12-04-00003 and 10-04-00152) to TT and DS, by the ‘Living Nature’ ª 2012 Federation of European Microbiological Societies Published by Blackwell Publishing Ltd. All rights reserved

288

Program of RAS to DS, and by the Russian Ministry of Education and Sciences to NS, TK, and BB. GM was supported by the European Research Council.

References Braid MD, Daniels ML & Kitts CL (2003) Removal of PCR inhibitors from soil DNA by chemical flocculation. J Microbiol Methods 52: 389–393. Friedrich CG, Bardischewsky F, Rother D, Quentmeier A & Fischer J (2005) Prokaryotic sulfur oxidation. Curr Opin Microbiol 8: 253–259. Ghosh W & Dam B (2009) Biochemistry and molecular biology of lithotrophic sulfur oxidation by taxonomically and ecologically diverse bacteria and archaea. FEMS Microbiol Rev 33: 999–1043. Giri BJ, Bano N & Hollibaugh JT (2004) Distribution of RuBisCO genotypes along a redox gradient in Mono Lake, California. Appl Environ Microbiol 70: 3443–3448. Gorlenko VM (2007) Anoxygenic phototrophic bacteria from soda lakes. Transactions of the Winogradsky Institute of Microbiology, V. XIV, pp. 159–183. Nauka, Moscow (in Russian). Gorlenko VM, Bryntseva IA, Panteleeva EE, Tourova TP, Kolganova TV, Makhneva ZK & Moskalenko AA (2004) Ectothiorhodosinus mongolicum gen. nov., sp. nov., a new purple bacterium from a soda lake in Mongolia. Microbiology (Moscow, English translation) 73: 66–73. Grant WD & Sorokin DY (2011) Distribution and diversity of soda lake alkaliphiles. Extremophiles Handbook, vol. 1 (Horikoshi K, Antranikian G, Bull AT, Robb FT & Stetter KO, eds), pp. 27–54. Springer, Tokyo. Hugler M, Gartner A & Imhoff JF (2010) Functional genes as markers for sulfur cycling and CO2 fixation in microbial communities of hydrothermal vents of the Logatchev field. FEMS Microbiol Ecol 73: 526–537. Humayoun SB, Bano N & Hollibaugh JT (2003) Depth distribution of microbial diversity in Mono Lake, a meromictic soda lake in California. Appl Environ Microbiol 69: 1030–1042. Kovaleva OL, Tourova TP, Muyzer G, Kolganova TV & Sorokin DY (2011) Diversity of RuBisCO and ATP citrate lyase genes in soda lake sediments. FEMS Microbiol Ecol 75: 37–47. Krishnani KK, Gopikrishna G, Pillai SM & Gupta BP (2010) Abundance of sulphur-oxidizing bacteria in coastal aquaculture using soxB gene analyses. Aquac Res 41: 1290–1301. Lenk S, Moraru C, Hahnke S et al. (2012) Roseobacter clade bacteria are abundant in coastal sediments and encode a novel combination of sulfur oxidation genes. ISME J 6: 2178–2187. Luo J-F, Lin W-T & Guo Y (2011) Functional genes based analysis of sulfur-oxidizing bacteria community in sulfide removing bioreactor. Appl Microbiol Biotechnol 90: 769–778. Ma Y, Zhang W, Xue YP, Zhou P, Ventosa A & Grant WD (2004) Bacterial diversity of the Inner Mongolian Baer Soda

ª 2012 Federation of European Microbiological Societies Published by Blackwell Publishing Ltd. All rights reserved

T.P. Tourova et al.

Lake as revealed by 16S rRNA gene sequence analyses. Extremophiles 8: 45–51. Mesbah N, Abou-El-Ela S & Wiegel J (2007) Novel and unexpected prokaryotic diversity in water and sediments of the alkaline, hypersaline lakes of the Wadi an Natrun. Microb Ecol 54: 598–617. Meyer B, Imhoff JF & Kuever J (2007) Molecular analysis of the distribution and phylogeny of the soxB gene among sulfur-oxidizing bacteria – evolution of the Sox sulfur oxidation enzyme system. Environ Microbiol 9: 2957–2977. Muyzer G, Sorokin DY, Mavromatis K et al. (2011) Complete genome sequence of Thioalkalivibrio sp. K90mix. Stand Genomic Sci 5: 341–355. Perreault NN, Greer CW, Andersen DT, Tille S, LacrampeCouloume G, Lollar BS & Whyte LG (2008) Heterotrophic and autotrophic microbial populations in cold perennial springs of the high arctic. Appl Environ Microbiol 74: 6898–6907. Petri R, Podgorsek L & Imhoff JF (2001) Phylogeny and distribution of the soxB gene among thiosulfate-oxidizing bacteria. FEMS Microbiol Lett 197: 171–178. Rees HC, Grant WD, Jones BE & Heaphy S (2004) Diversity of Kenyan soda lake alkaliphiles assessed by molecular methods. Extremophiles 8: 63–71. Singleton DR, Furlong MA, Rathbun SL & Whitman WB (2001) Quantitative comparisons of 16S rRNA gene sequence libraries from environmental samples. Appl Environ Microbiol 67: 4374–4376. Sorokin DY (2003) Oxidation of inorganic sulfur compounds by obligately organotrophic bacteria. Microbiology (Moscow, English Translation) 72:725–739. Sorokin DY (2008) Diversity of halophilic sulfur-oxidizing bacteria in hypersaline habitats. Microbial Sulfur Metabolism (Dahl C & Friedrich CG eds). Proceedings of the International Symposium on Microbial Sulfur Metabolism 29.06.2006–02.07.2006, pp. 225–237. Springer-Berlin, M€ unster, Germany. Sorokin DY, Tourova TP, Kuznetsov BB, Bryatseva IA & Gorlenko VM (2000) Roseinatronobacter thiooxidans gen. nov., sp.nov. – a new alkaliphilic aerobic bacteriochlorophyll a-containing bacterium from soda lake. Microbiology (Moscow, English Translation) 69:1–9. Sorokin DY, Banciu H, Robertson LA & Kuenen JG (2006) Haloalkaliphilic sulfur-oxidizing bacteria. Prokaryotes. Vol. 2: Ecophysiology and Biochemistry (Dworkin M, Falkow S, Rosenberg E, Schleifer KH & Stackebrandt E eds), pp. 969– 984. Springer, New York. Sorokin DY, Kuenen JG & Muyzer G (2011) The microbial sulfur cycle in soda lakes. Front Microbiol 2: 44. Tamura K, Peterson D, Peterson N, Stecher G, Nei M & Kumar S (2011) MEGA5: molecular evolutionary genetics analysis using maximum likelihood, evolutionary distance, and maximum parsimony methods. Mol Biol Evol 28: 2731–2739. Thompson JD, Higgins DG & Gibson TJ (1994) CLUSTAL W: improving the sensitivity of progressive multiple sequence alignment through sequence weighting, positions-specific

FEMS Microbiol Ecol 84 (2013) 280–289

Sulfur-oxidizing bacteria in hypersaline and soda lakes

gap penalties and weight matrix choice. Nucleic Acids Res 9: 3251–3270. Tourova TP, Kovaleva OL, Sorokin DY & Muyzer G (2010) Ribulose-1,5-bisphosphate carboxylase/oxygenase genes as a functional marker for chemolithoautotrophic halophilic sulfur-oxidizing bacteria in hypersaline habitats. Microbiology 156: 2016–2025. Tourova TP, Kovaleva OL, Bumazhkin BK, Patutina EO, Kuznetsov BB, Briantseva IA, Gorlenko VM & Sorokin DY (2011). Utility of Ribulose-1,5-Biphosphate Carboxylase as a molecular marker for evaluation of autotrophic microbial consortia biodiversity in the surface layers of the Kulundine

FEMS Microbiol Ecol 84 (2013) 280–289

289

steppe soda lakes. Microbiology (Moscow, English translation) 80: 803–817. Van de Peer Y & De Wachter R (1994) TREECON for Windows: a software package for the construction and drawing of evolutionary trees for the Microsoft Windows environment. Comput Appl Biosci 10: 569–570. Yoshizawa Y, Toyoda K, Arai H, Ishii M & Igarashi Y (2004) CO2- responsive expression and gene organization of three ribulose-1,5-bisphosphate carboxylase/oxygenase enzymes and carboxysomes in Hydrogenovibrio marinus strain MH110. J Bacteriol 186: 5685–5691.

ª 2012 Federation of European Microbiological Societies Published by Blackwell Publishing Ltd. All rights reserved