Insights into the pathways of iron- and sulfur-oxidation ...

Research in Microbiology 165 (2014) 753e760 www.elsevier.com/locate/resmic

Insights into the pathways of iron- and sulfur-oxidation, and biofilm formation from the chemolithotrophic acidophile Acidithiobacillus ferrivorans CF27 Emmanuel Talla a, Sabrina Hedrich b,1, Sophie Mangenot c, Boyang Ji a,2, D. Barrie Johnson b, Valerie Barbe c, Violaine Bonnefoy a,* a

Aix-Marseille Universite and Centre National de la Recherche Scientifique, LCB UMR7283, 31 chemin J. Aiguier, 13402 Marseilles Cedex 20, France b College of Natural Sciences, Bangor University, Brambell Building, Deiniol Road, LL57 2UW Bangor Gwynedd, UK c CEA/IG/Genoscope, Laboratoire de finition, 2 rue Gaston Cremieux, CP5706, 91057 Evry Cedex, France Received 14 April 2014; accepted 7 August 2014 Available online 19 August 2014

Abstract The iron-oxidizing acidithiobacilli cluster into at least four groups, three of which (Acidithiobacillus ferrooxidans, Acidithiobacillus ferridurans and Acidithiobacillus ferrivorans) have been designated as separate species. While these have many physiological traits in common, they differ in some phenotypic characteristics including motility, and pH and temperature minima. In contrast to At. ferrooxidans and At. ferridurans, all At. ferrivorans strains analysed to date possess the iro gene (encoding an iron oxidase) and, with the exception of strain CF27, the rusB gene encoding an iso-rusticyanin whose exact function is uncertain. Strain CF27 differs from other acidithiobacilli by its marked propensity to form macroscopic biofilms in liquid media. To identify the genetic determinants responsible for the oxidation of ferrous iron and sulfur and for the formation of extracellular polymeric substances, the genome of At. ferrivorans CF27 strain was sequenced and comparative genomic studies carried out with other Acidithiobacillus spp.. Genetic disparities were detected that indicate possible differences in ferrous iron and reduced inorganic sulfur compounds oxidation pathways among iron-oxidizing acidithiobacilli. In addition, strain CF27 is the only sequenced Acidithiobacillus spp. to possess genes involved in the biosynthesis of fucose, a sugar known to confer high thickening and flocculating properties to extracellular polymeric substances. © 2014 Institut Pasteur. Published by Elsevier Masson SAS. All rights reserved.

Keywords: Acidithiobacillus ferrivorans; Genome analysis; Iron oxidation; Sulfur oxidation; Biofilm; Extracellular polymeric substances

1. Introduction

* Corresponding author. E-mail addresses: [email protected] (E. Talla), [email protected] (S. Hedrich), [email protected] (S. Mangenot), [email protected] (B. Ji), [email protected] (D.B. Johnson), [email protected] (V. Barbe), [email protected] (V. Bonnefoy). 1 Present address: Federal Institute for Geosciences and Natural Resources (BGR), Stilleweg 2, 30655 Hannover, Germany. 2 Present address: Department of Chemical and Biological Engineering, Chalmers University of Technology, Gothenburg, Sweden.

Autotrophic acidophilic iron- and sulfur-oxidizing bacteria of the genus Acidithiobacillus catalyse the oxidative dissolution of sulfide minerals and play major roles in biomining and the genesis of acid mine drainage. They form at least four monophyletic groups [1], three of which (Acidithiobacillus ferrooxidans [22,38], Acidithiobacillus ferrivorans [17] and Acidithiobacillus ferridurans [19]) have been validated as distinct species. The main physiological differences between strains of At. ferrooxidans (and At. ferridurans, which display very similar traits) and At. ferrivorans are in their response to

http://dx.doi.org/10.1016/j.resmic.2014.08.002 0923-2508/© 2014 Institut Pasteur. Published by Elsevier Masson SAS. All rights reserved.

754

E. Talla et al. / Research in Microbiology 165 (2014) 753e760

temperature and pH, and cell motility, and oxidation of hydrogen [17,20]. The pathways involved in the oxidation of ferrous iron (Fe(II)) and reduced inorganic sulfur compounds (RISCs) by At. ferrooxidans and At. ferridurans have been well studied [2,5e7,34]. In the case of At. ferrivorans, preliminary bioinformatic analysis of the genome of strain SS3 suggests a similar Fe(II) oxidation pathway but a less clear RISC oxidation pathway, with redundant genes from both At. ferrooxidans and Acidithiobacillus caldus present [28,29]. Previous data suggested that another Fe(II) oxidation pathway may have evolved in At. ferrivorans, since all the strains studied to date possess the iro gene [1] which encodes the periplasmic high potential ironesulfur (HiPIP) protein Iro (for iron oxidase) that has been described as the first electron acceptor involved in Fe(II) oxidation in bacteria that were designated at the time as strains of At. ferrooxidans, but which are actually Group IV acidithiobacilli [11,15,25]. Furthermore, the rusB gene encoding rusticyanin B, an isoform of rusticyanin A which is known to play a key role in Fe(II) oxidation [2,5,6,34], was detected in all the strains of At. ferrivorans examined, apart from strain CF27 [1]. The kinetic rate constant for electron transfer between Fe(II) and RusB was shown to be approximately one half that of RusA [21] and its exact function in aerobic Fe(II) oxidation has not yet been clarified. It was also noted that strain CF27 often clustered away from the four other strains of At. ferrivorans studied by multilocus sequence analysis [1]. In addition, strain CF27 forms large macroscopic aggregates of mineral grains enmeshed with bacterial biomass in liquid media, suggesting a greater propensity for extracellular polymeric substances (EPS) biofilm formation than in other strains (Fig. 1). To identify genetic determinants likely responsible for these physiological differences, the genome of At. ferrivorans CF27 was sequenced. Here we report the initial findings obtained from analysing and comparing the data with the genomes of other strains of Acidithiobacillus spp.

2. Materials and methods 2.1. Strains and cultivation conditions Four strains of At. ferrivorans were used in the present study: NO-37T, Peru6, OP14, and CF27 [1]. These were grown routinely in a liquid medium containing 20 mM ferrous iron, basal salts and trace elements [29] at an initial pH of 1.9 (adjusted with sulfuric acid), at 30 C. The type strains of At. ferrooxidans (ATCC 23270T) and At. ferridurans (ATCC 33020T) used in some experiments, were cultivated under the same conditions. 2.2. General DNA manipulations DNA from At. ferrivorans strains NO-37T, OP14, Peru6, and the type strains of At. ferrooxidans and At. ferridurans, was extracted from 5 mL of Fe(II)-grown cells to serve as a template in PCR reactions [32]. As the primers previously used to amplify the hip and rusA genes in strains of At. ferrooxidans and At. ferridurans could not amplify these genes in strains of At. ferrivorans [1], new primer sets (Table S1) were designed based on alignments of the hip and rusA genes of At. ferrooxidansT [40], At. ferriduransT [8,3] and At. ferrivorans strains CF27 and SS3. PCR were carried out in 20 ml reactions consisting of 1 GoTaq buffer (Promega, UK), 0.2 mM dNTPs, 1.5 mM MgCl2, 25 pmol each primer and 1 U Hotstart Taq-Polymerase (Promega) using genomic DNA from At. ferrivorans NO-37T, OP14, Peru6 as a template. The PCR program was as follows: initial denaturation at 95 C for 5 min, 40 cycles of (i) denaturation for 30 s at 95 C, (ii) annealing for 30 s at 58 C for rusA and 62 C for hip, and (iii) elongation for 30 s at 72 C and a final elongation step of 5 min at 72 C before the temperature was reduced to 4 C. PCR products were purified using SureClean (Bioline Ltd., UK) and sequenced by Macrogen (Macrogen Inc., Korea). 2.3. DNA preparation, whole-genome sequencing, and genome analysis

Fig. 1. Macroscopic image of biofilm growth of At. ferrivorans strain CF27, encapsulating fine grains of pyrite. The diameter of the image shown is ~5 mm.

In order to obtain sufficient DNA for genome sequencing, At. ferrivorans CF27 was grown in 1 L of 20 mM ferrous iron medium supplemented with 1.0% (w/v) elemental sulfur. Total genomic DNA was extracted from cells using Wizard Genomic DNA purification kits (Promega, UK). Wholegenome sequencing of At. ferrivorans CF27 was performed using Illumina technology. A mate-paired (MP) and pairedend (PE) libraries were created with 5 kbp and 330 bp insert size, respectively. Sequence data were then assembled using Velvet (http://www.ebi.ac.uk/~zerbino/velvet). To reduce the number of undetermined bases, GapCloser (http://soap. genomics.org.cn/soapdenovo.html) was performed onto the scaffold sequences with the PE reads. All general aspects of the library construction, sequencing and assembly were performed at Genoscope (www.genoscope.cns.fr, Evry, France). Computational prediction of coding sequences (CDS) and


other genome features (RNA encoding genes, ribosome binding sites, signal sequences etc.), together with functional assignments were performed using the annotation pipeline implemented in the MicroScope platform [42]. The accession number of the CF27 draft genome is CCS020000001 eCCCS020000082 (Bioproject PRJEB5721). Other genomes of Acidithiobacillus spp. (At. ferrivorans SS3 (NC_015942) [29], At. ferrooxidansT (NC_011761) [40], At. ferrooxidans ATCC 53993 (NC_011206), At. caldus ATCC 51756T (NC_ACVD00000000) [41], At. caldus SM-1 (NC_015850) [12] and At. thiooxidans ATCC 19377T (AFOH01000000) [39]) were retrieved from the NCBI website and integrated within the MicroScope platform for further analysis. Comparative analysis of genes involved in pathways of the oxidation of Fe(II) and RISC oxidation, and also those implicated in EPS synthesis, was performed using the MicroScope website facilities [42]. 2.4. Phylogenetic analysis The sequences of the phylogenetic markers (rusA/B and iro/ hip genes) were obtained either from published gene sequences or from entire genomes [1,12,29,40,41], retrieved from the non-redundant database [4]. For At. ferrivorans strains NO-37T, OP14 and Peru6, rusA and hip sequences were determined as described above, and the corresponding accession numbers listed below. For each marker, sequences were aligned using ClustalW [26] followed by the selection of unambiguous parts with the Gblocks 0.91b program [10]. Next, phylogenetic trees were reconstructed using the Bayesian approach implemented in MrBayes 3.2 program [36], with the K80 þ G model for rusA/B and HKY þ G model for iro/hip. The Markov chain Monte Carlo search was run with 4 chains for 1,000,000 generations. Trees were sampled every 100 generations with the first 1500 trees being discarded as “burnin”.

755

markers (rrs, ITS1, recA and hip) [1]. However, the primers used failed to amplify rusA and hip genes from all five strains of At. ferrivorans that were analysed [1], while these genes were subsequently detected in the genome of strain SS3 [28,29] and, in this study, that of strain CF27. New primers designed using the published genome of At. ferrivorans SS3 [28,29] and that of CF27 (this study), were used successfully to amplify the internal fragment of rusA and hip genes from At. ferrivorans strains NO-37T, OP14 and Peru6. Phylogenetic relationships based on rusA/rusB and hip/iro markers (Fig. 2) showed the same clusters (iron-oxidizing acidithiobacilli Groups IeIV) to those found previously using other marker genes [1]. Again the additional sequences of rusA and hip genes from OP14, NO-37T, Peru6, SS3 and CF27 strains cluster together, separately from At. ferrooxidans (Group I), At. ferridurans (Group II) and Group IV, though interestingly, the iro gene of CF27 seems to be atypical since it is emerged as an external branch of Group IV, separated from Group III iro genes. The results from this part of the study confirmed that phylogenetic analysis using rusA/B and iro/hip is an appropriate method to delineate species of iron-oxidizing acidithiobacilli. Since CF27 clustered away from the other At. ferrivorans strains (Group III) not only with rrs, ITS1, recA and hip but also with iro and rusA, this strain could be the first member of a new subgroup within Group III. 3.2. Overview of the draft genome of At. ferrivorans CF27

The nucleotide sequences determined in this paper have been deposited at GenBank database under the following accession numbers: KC533886eKC53388 for hip from At. ferrivorans Peru6, OP14 and NO-37T, and KC533889eKC53391 for rusA from At. ferrivorans OP14, NO-37T and Peru6, respectively. The accession numbers of the other sequences are given in Ref. [1].

The draft genome of At. ferrivorans CF27 comprises 82 contigs (81 scaffolds) spanning 3.44 Mbp with a G þ C content of 56.4% (Table 1), which is similar to the G þ C content of At. ferrivorans SS3 (56.6%) [29] but lower than values reported from sequenced genomes for strains of At. ferrooxidans (~58%) [40] or At. caldus (>61%) [12,41]. The current version of the chromosome contains 3933 predicted coding sequences (CDS), one rRNA operon organized with the order 16Se23Se5S and 71 tRNA genes that cover all the 20 amino acids. While 2246 CDS (57.1%) could be assigned to putative functions, 948 CDS (24.1%) are conserved hypothetical proteins of unknown function, and the remaining 739 CDS (18.8%) show no sequence similarity to any previously reported sequences. The high number of tRNA genes in CF27 strain as well as in At. ferrooxidansT may reflect the physiological importance of protein translation metabolism in these organisms.

3. Results and discussion

3.3. Iron oxidation

3.1. Phylogenetic analysis of rusA/B and iro/hip genes

The rus operon encoding an outer membrane cytochrome c (cyc2), a periplasmic membrane-bound c4-type cytochrome (cyc1), a protein (cup or acoP), which function is matter of debate [9,34], the four subunits of an aa3 cytochrome oxidase (coxBACD) and the blue copper protein rusticyanin (rus) have been shown to be involved in the electron transfer from Fe(II) to O2 through an electron nanowire spanning the outer and the inner membranes in At. ferrooxidans [34,35] and At.

2.5. Nucleotide sequence accession numbers

In a previous study, it was shown that the types of rusticyanin (rusA/B) and HiPIP (iro/hip) genes that the ironoxidizing acidithiobacilli harbour correlate with their division in four monophyletic groups [1]. In addition, strain CF27, while clearly belonging to Group III (At. ferrivorans) clustered away from the other strains analysed with several phylogenetic

756


Fig. 2. Bayesian unrooted phylogenetic trees of Fe(II)-oxidizing Acidithiobacillus iro/hip (a) and rusA/rusB (b) genes. Italicized numbers at nodes represent posterior probabilities (PPs). Stars highlight the two different copies of rusB identified in JCM 3865 and SS3. For clarity, only PPs greater than 0.5 are shown. The scale bars represent the average number of substitutions per site. Groups I, II, III, and IV are labelled according to Ref. [1].

ferridurans [2,43]. The electrons coming from the Fe(II) oxidation are used to reduce O2 to water, consuming the protons entering the cells through the ATP synthetase due to the huge proton gradient across the inner membrane. In addition, iron-oxidizing acidithiobacilli have to regenerate the NAD(P)H pool necessary for CO2 fixation and other anabolic processes. In that case, the electrons have to be pushed “uphill” against the redox potential gradient from Fe(II)/Fe(III) (Em,2 ¼ þ0.77 V) to NAD (Em,7 ¼ 0.32 V). A bc1 complex functioning in reverse encoded by the petI operon has been shown to be involved in this pathway in At. ferrooxidans [34,35] and At. ferridurans [7,27]. The rus and petI operons have also been detected in other iron-oxidizing acidithiobacilli: At. ferrooxidans ATCC 53993, and At. ferrivorans strains SS3 [28,29] and CF27 (this study; Table 2a and Table

Table 1 General features of the Acidithiobacillus ferrivorans CF27 discontinuous draft genome sequence. At. ferrivorans CF27 Genome size (bp) Number of contigs Number of scaffolds Predicted CDS G þ C content (%) Coding density (%) Mean CDS length (bp) Maximal CDS length (bp) tRNAs 5/16/23S IS/Transposases Proteins assigned to COGs Proteins involved in Replication, recombination and repair Amino acid transport and metabolism Inorganic ion transport metabolism Proteins with predicted function

3,433,672 82 81 3933 56.4 89.2 802.3 5304 71 (20) 1/1/1 56 2535 (64.5%) 287 (7.3%) 280 (7.1%) 237 (6.0%) 2333 (59.3%)

S2). The gene order in both operons is conserved (data not shown), and it can be inferred that At. ferrivorans utilizes the same ferrous iron oxidation pathway as At. ferrooxidans and At. ferridurans. However, the possibility that another pathway of iron oxidation has evolved in At. ferrivorans cannot be excluded, since the iron oxidase putatively encoded by iro [11,15,25] has been detected in the genomes of both At. ferrivorans SS3 and CF27 [28,1] though not in At. ferrooxidans (ATCC 23270T and ATCC 53993), or Acidithiobacillus spp. (At. caldus and At. thiooxidans) that do not oxidize iron (Table 2a and Table S2). In addition, while rusB genes encoding the iso-rusticyanin B are present in At. ferrivorans strains NO-37T, OP14, Peru6 and SS3, they have not been found in other iron-oxidizing acidithiobacilli, or (as confirmed in the present study) At. ferrivorans CF27 [1]. The environmental conditions in which these genes are up-regulated could help elucidate the pathways in which they are involved. By comparing the iron oxidation efficiencies of At. ferrooxidans (rusA), At. ferrivorans SS3 (rusA, rusB1, rusB2 and iro) and At. ferrivorans CF27 (rusA and iro) when grown under different regimes it would be possible to identify conditions where these additional genes might confer an advantage. 3.4. Reduced inorganic sulfur compound oxidation Both At. ferrivorans strains CF27 and SS3 have most of the genes shown to encode enzymes and redox proteins involved in the dissimilatory RISC oxidation in At. ferrooxidansT [34,35], At. caldus [30,41] and At. thiooxidans [39] (hdr locus: heterodisulfide reductase; sqr: sulfide quinone reductase; tetH: tetrathionate hydrolase; doxDA locus: thiosulfate quinone reductase; sat/cytC: ATP sulfurylase/APS kinase, cyoABCD: bo3 cytochrome oxidase; Table 2b and Table S3). Interestingly, At. ferrivorans strains CF27 and SS3 seem not to have the cydAB operon (bd quinol oxidase) while strains


757

Table 2 Analysis of the sequenced genomes of Acidithiobacillus caldus SM-1 (1) and ATCC 51756T (2), At. thiooxidans ATCC 19377T (3), At. ferrooxidans ATCC 23270T (4) and ATCC 53993 (5), and At. ferrivorans SS3 (6) and CF27 (7) with reference to genes involved in iron-(a) and sulfur-(b) oxidation, and in biofilm biogenesis (c). White cells: absence of the gene/operon; black cells: presence of the gene/operon, the number of copies is indicated in white letterings; grey cells: incomplete pathway (pseudogene (ps) or gene(s) missing (i)).

-

-

-

-

-

-

-

-

- - - -

- -

- - -

-

- - - -

of At. caldus and At. thiooxidans have six and five copies, respectively (Table 2b and Table S3). At. ferrivorans strains CF27 and SS3 have the soxXYZAehypesoxB locus (Sox system involved in thiosulfate, sulfur, sulfite and sulfide oxidation [14]) present in At. caldus [12,30,41] and At. thiooxidans [39] but it seems not functional since soxX (SS3 and CF27) and soxA (SS3) are pseudogenes (Table 2b and Table S3). The sor gene sulfur oxygenase reductase), previously

detected in At. caldus ATCC 51756T [41,30] and MTH-04 [12], is also present in both At. ferrivorans strains (Table 2b and Table S3). Other iron-oxidizing Acidithiobacillus spp. (At. ferrooxidans and At. ferridurans) have the tsd gene encoding the thiosulfate dehydrogenase [23]. Interestingly, the petII operon, proposed recently to be involved in the dissimilatory anoxic ferric iron reduction in At. ferrooxidans [33] is only present in At. ferrooxidans, At. ferridurans and At. ferrivorans

758


Fig. 3. Synteny of GDP-L-fucose biosynthesis genes between the different acidithiobacilli. The upper panel shows At. ferrivorans CF27 while the lower panel shows the other acidithiobacilli. In between the two panels are given numbers corresponding to the genes with the following functional description: 1, mannose-1phosphate guanylytransferase; 2, GDP-D-mannose dehydratase; 3, GDP-6-deoxy-D-mannose reductase; 4, bifunctional GDP-fucose synthetase; 5, 9, 10, 11, 12, putative glycosyl transferase; 6, putative glycose-acyl transferase; 7, transposase; 8, UDP-sulfoquinovose synthase; 13, polysaccharide biosynthesis protein; 14, putative protein-tyrosine phosphatase; 15, putative exopolysaccharide synthesis protein; 16, polysaccharide export protein; 17, phosphoesterase PA-phosphatase related protein; 18, undecaprenyl-phosphate galactose phosphotransferase; 19, 20, putative RNA-directed DNA polymerase; 21, shufflon-specific DNA recombinase (fragment); 22, integrase family protein (fragment); 23, putative NADPH dehydrogenase; 24, dienelactone hydrolase; 25, regulatory protein TetR; 26, nicotinate-nucleotide pyrophosphorylase (NadC); 27, Na(þ)/H(þ) antiporter. Underlined, are indicated the proteins predicted to be involved in EPS biosynthesis and in bold, the proteins which gene is found only in CF27. In italics, proteins which could play a function in DNA rearrangement. The genes encoding proteins of unknown function are not indicated. Same grey tone indicates the same localization.

strains, which are the only acidithiobacilli that can grow under anaerobic conditions by respiring Fe(III) with sulfur as an electron donor [17] and under aerobic conditions with Fe(II) as an electron donor. At. ferrivorans CF27 differs from strain SS3 by having two doxDA genes (Table 2b and Table S3). In addition, the cyoD gene of one cyoABCD operon of SS3 presents a frameshift suggesting that this bo3 quinol oxidase is not functional since CyoD has been proposed to assist the CuB binding to subunit I during biosynthesis of the oxidase complex [37]. Noteworthy is that this cyo operon is located in the same locus as doxDA, tetH, sor and sqr, suggesting that this bo3 quinol oxidase is involved in the RISC oxidation pathways. Apart from At. ferrooxidans, all the other acidithiobacilli have several copies of the bo3 quinol oxidase operons (Table 2b and Table S3). These genetic disparities provide at least partial explanations for the different propensities among Acidithiobacillus spp. for oxidizing RISCs. 3.5. EPS biosynthesis As mentioned previously, At. ferrivorans CF27 is particularly adept at producing EPS and, in contrast to other acidithiobacilli, forms conspicuous biofilms in liquid cultures (Fig. 1). To help understand why this is so, genes known to be involved in biofilm formation, including genes encoding proteins necessary for lipopolysaccharide and EPS formation, for biosynthesis and assembly of type IV pilus and for tight adherence, were looked for in genomes of Acidithiobacillus

spp. (data summarized in Table 2c). Of particular note are: (i) the genes necessary for tight adherence were detected only in At. ferrooxidans; (ii) in At. ferrooxidans ATCC 53993, pilM has a frame-shift suggesting that this strain might be unable to assemble type IV pili; (iii) the gene encoding cellulose synthase is present in all the acidithiobacilli except At. ferrooxidans; (iv) the gene involved in UDP-D-galacturonate biosynthesis is only present in At. ferrooxidans and At. ferrivorans SS3; (v) only At. ferrivorans CF27 has the fcl gene involved in the biosynthesis of GDP-L-fucose. This indicates that the composition of EPS in Acidithiobacillus spp. can display differences that are both species-related and, in the case of At. ferrivorans, strain-specific. In line with this assumption, variations in the adhesion to solid surfaces by cells belonging to different Acidithiobacillus spp. and also to different strains of the same species have been previously reported [16,18,24]. Interestingly, fucose was detected in EPS synthesized by the iron-oxidizing Acidithiobacillus strain R1 (one of the Group IV acidithiobacilli [1]) which has also been reported to form dense biofilms [18,24]. Fucose is a relatively uncommon sugar, and fucose-containing EPS have been shown to have high thickening and flocculating properties as well as considerable viscosity and high emulsion-stabilizing capacity [13]. The propensity of At. ferrivorans CF27 to form macroscopic biofilms in liquid media might therefore result from the production of fucose-containing EPS. Interestingly, the gene encoding the bifunctional GDP-fucose synthetase co-localized with genes encoding proteins known to be involved in EPS biosynthesis (Fig. 3). Among them,


those encoding six putative glycosyl transferases are absent in the other acidithiobacilli (Fig. 3). In addition, the sulfoquinovose synthase sqdB gene is present in At. ferrivorans strain CF27, At. caldus SM-1 and ATCC 51756T but not in At. thiooxidans, At ferrooxidans (ATCC 23270T and ATCC 53993) or in At. ferrivorans SS3 (Fig. 3). This enzyme has also been shown to be important in the N-glycosylation of the surfacelayer glycoprotein of the acidophilic archaeon Sulfolobus acidocaldarius, a process necessary for the maintenance of an intact and stable envelope, cell adherence and biofilm formation [31]. 4. Concluding remarks The data presented in this paper clearly show that At. ferrivorans CF27 shares a large amount of communality with other genome-sequenced Acidithiobacillus spp. with regard to pathways of Fe(II) and RISC oxidation and, to some extent, biofilm formation. However, it has specific features that clearly distinguish it from the other iron-oxidizing acidithiobacilli and also from At. ferrivorans SS3. The presence of the iro gene in CF27 (and all other strains of At. ferrivorans) and its absence in At. ferrooxidans and At. ferridurans might suggest an alternative mechanism for oxidizing ferrous iron (i.e. not necessarily involving rusticyanin) in Group III iron-oxidizing acidithiobacilli, as previously suggested [1]. The significance of the absence of rusB in CF27 (a unique trait among strains of At. ferrivorans) is not known, but does clearly not impair its ability to oxidize ferrous iron. As with strain SS3, At. ferrivorans CF27 possesses all of the “core” genes involved in RISC oxidation detected in all Acidithiobacillus spp. examined (sqr, hdr, tetH, doxDA, sat/ cytC, and cyoABCD allowing sulfide, sulfur, tetrathionate, thiosulfate and sulfite oxidation with O2 reduction), but also some genes found only in At. ferrooxidans (tsd and petII ) and others in At. caldus (sor and soxXYZAB, even if the Sox system is predicted to be deficient in At. ferrivorans CF27 and SS3). These disparities might reflect genetic rearrangements by horizontal gene transfers and/or a reductionist evolution by gene mutation(s) and deletion(s). At. ferrivorans CF27 possesses genes involved in the glycosylation of the surface-layer glycoprotein as well as the fcl gene necessary for fucose biosynthesis, neither of which have been detected in other Acidithiobacillus spp.. Fucose confers distinct rheological properties to EPS [13] that could explain the propensity of At. ferrivorans CF27 to form gelatinous macroscopic growths in liquid media in contrast to other Acidithiobacillus spp. Conflict of interest The authors declare no conflict of interest. Acknowledgements The authors acknowledge the LABGeM (Genoscope, Evry, France) for the automatic annotation. B.J. was a Ph.D. student

759

with a China Scholarship Council fellowship. We are grateful to Dr. Barry Grail (Bangor University) for help in preparing biomass of At. ferrivorans CF27.

Appendix A. Supplementary data Supplementary data related to this article can be found at http://dx.doi.org/10.1016/j.resmic.2014.08.002. References [1] Amouric A, Brochier-Armanet C, Johnson DB, Bonnefoy V, Hallberg KB. Phylogenetic and genetic variation among Fe(II)-oxidizing acidithiobacilli supports the view that these comprise multiple species with different ferrous iron oxidation pathways. Microbiology 2011;157:111e22. [2] Appia-Ayme C, Guiliani N, Ratouchniak J, Bonnefoy V. Characterization of an operon encoding two c-type cytochromes, an aa(3)-type cytochrome oxidase, and rusticyanin in Thiobacillus ferrooxidans ATCC 33020. Appl Environ Microbiol 1999;65:4781e7. [3] Bengrine A, Guiliani N, Appia-Ayme C, Jedlicki E, Holmes DS, Chippaux M, et al. Sequence and expression of the rusticyanin structural gene from Thiobacillus ferrooxidans ATCC33020 strain. Biochim Biophys Acta 1998;1443:99e112. [4] Benson DA, Clark K, Karsch-Mizrachi I, Lipman DJ, Ostell J, Sayers EW. GenBank. Nucleic Acids Res 2014;42:D32e7. [5] Bird LJ, Bonnefoy V, Newman DK. Bioenergetic challenges of microbial iron metabolisms. Trends Microbiol 2011;19:330e40. [6] Bonnefoy V, Holmes DS. Genomic insights into microbial iron oxidation and iron homeostasis in extremely acidic environments. Environ Microbiol 2012;14:1597e611. [7] Bruscella P, Appia-Ayme C, Levican G, Ratouchniak J, Jedlicki E, Holmes DS, et al. Differential expression of two bc(1) complexes in the strict acidophilic chemolithoautotrophic bacterium Acidithiobacillus ferrooxidans suggests a model for their respective roles in iron or sulfur oxidation. Microbiology 2007;153:102e10. [8] Bruscella P, Cassagnaud L, Ratouchniak J, Brasseur G, Lojou E, Amils R, et al. The HiPIP from the acidophilic Acidithiobacillus ferrooxidans is correctly processed and translocated in Escherichia coli, in spite of the periplasm pH difference between these two micro-organisms. Microbiology 2005;151:1421e31. [9] Castelle C, Ilbert M, Infossi P, Leroy G, Giudici-Orticoni MT. An unconventional copper protein required for cytochrome c oxidase respiratory function under extreme acidic conditions. J Biol Chem 2010;285:21519e25. [10] Castresana J. Selection of conserved blocks from multiple alignments for their use in phylogenetic analysis. Mol Biol Evol 2000;17: 540e52. [11] Cavazza C, Guigliarelli B, Bertrand P, Bruschi M. Biochemical and EPR characterization of a high potential iron-sulfur protein in Thiobacillus ferrooxidans. FEMS Microbiol Lett 1995;130:193e200. [12] Chen L, Ren Y, Lin J, Liu X, Pang X, Lin J. Acidithiobacillus caldus sulfur oxidation model based on transcriptome analysis between the wild type and sulfur oxygenase reductase defective mutant. PLoS One 2012;7:e39470. [13] Freitas F, Alves VD, Torres CAV, Cruz M, Sousa I, Melo MJ, et al. Fucose-containing exopolysaccharide produced by the newly isolated Enterobacter strain A47 DSM 23139. Carbohydr Polym 2011;83:159e65. [14] Friedrich CG, Bardischewsky F, Rother D, Quentmeier A, Fischer J. Prokaryotic sulfur oxidation. Curr Opin Microbiol 2005;8:253e9. [15] Fukumori Y, Yano T, Sato A, Yamanaka T. Fe(II) oxidizing enzyme purified from Thiobacillus ferrooxidans. FEMS Microbiol Lett 1988;50:169e72.

760


[16] Ghauri MA, Okibe N, Johnson DB. Attachment of acidophilic bacteria to solid surfaces: the significance of species and strain variations. Hydrometallurgy 2007;85:72e80. [17] Hallberg KB, Gonzalez-Toril E, Johnson DB. Acidithiobacillus ferrivorans, sp. nov.; facultatively anaerobic, psychrotolerant iron-, and sulfur-oxidizing acidophiles isolated from metal mine-impacted environments. Extremophiles 2010;14:9e19. [18] Harneit K, G€ oksel A, Klock J-H, Gehrke T, Sand W. Adhesion to metal sulfide surfaces by cells of Acidithiobacillus ferrooxidans, Acidithiobacillus thiooxidans and Leptospirillum ferrooxidans. Hydrometallurgy 2006;83:245e54. [19] Hedrich S, Johnson DB. Acidithiobacillus ferridurans sp. nov., an acidophilic iron-, sulfur- and hydrogen-metabolizing chemolithotrophic gammaproteobacterium. Int J Syst Evol Microbiol 2013;63:4018e25. [20] Hedrich S, Johnson DB. Aerobic and anaerobic oxidation of hydrogen by acidophilic bacteria. FEMS Microbiol Lett 2013;349:40e5. [21] Ida C, Sasaki K, Ando K, Blake 2nd RC, Saiki H, Ohmura N. Kinetic rate constant for electron transfer between ferrous ions and novel rusticyanin isoform in Acidithiobacillus ferrooxidans. J Biosci Bioeng 2003;95:534e7. [22] Kelly DP, Wood AP. Reclassification of some species of Thiobacillus to the newly designated genera Acidithiobacillus gen. nov., Halothiobacillus gen. nov. and Thermithiobacillus gen. nov. Int J Syst Evol Microbiol 2000;50(Pt 2):511e6. [23] Kikumoto M, Nogami S, Kanao T, Takada J, Kamimura K. Tetrathionateforming thiosulfate dehydrogenase from the acidophilic, chemolithoautotrophic bacterium Acidithiobacillus ferrooxidans. Appl Environ Microbiol 2013;79:113e20. [24] Kinzler K, Gehrke T, Telegdi J, Sand W. Bioleaching e a result of interfacial processes caused by extracellular polymeric substances (EPS). Hydrometallurgy 2003;71:83e8. [25] Kusano T, Takeshima T, Sugawara K, Inoue C, Shiratori T, Yano T, et al. Molecular cloning of the gene encoding Thiobacillus ferrooxidans Fe(II) oxidase. High homology of the gene product with HiPIP. J Biol Chem 1992;267:11242e7. [26] Larkin MA, Blackshields G, Brown NP, Chenna R, McGettigan PA, McWilliam H, et al. Clustal W and Clustal X version 2.0. Bioinformatics 2007;23:2947e8. [27] Levican G, Bruscella P, Guacunano M, Inostroza C, Bonnefoy V, Holmes DS, et al. Characterization of the petI and res operons of Acidithiobacillus ferrooxidans. J Bacteriol 2002;184:1498e501. [28] Liljeqvist M, Rzhepishevska OI, Dopson M. Gene identification and substrate regulation provide insights into sulfur accumulation during bioleaching with the psychrotolerant acidophile Acidithiobacillus ferrivorans. Appl Environ Microbiol 2013;79:951e7. [29] Liljeqvist M, Valdes J, Holmes DS, Dopson M. Draft genome of the psychrotolerant acidophile Acidithiobacillus ferrivorans SS3. J Bacteriol 2011;193:4304e5.

[30] Mangold S, Valdes J, Holmes DS, Dopson M. Sulfur metabolism in the extreme acidophile Acidithiobacillus caldus. Front Microbiol 2011;2:17. [31] Meyer BH, Zolghadr B, Peyfoon E, Pabst M, Panico M, Morris HR, et al. Sulfoquinovose synthase e an important enzyme in the N-glycosylation pathway of Sulfolobus acidocaldarius. Mol Microbiol 2011;82:1150e63. [32] Okibe N, Gericke M, Hallberg KB, Johnson DB. Enumeration and characterization of acidophilic microorganisms isolated from a pilot plant stirred-tank bioleaching operation. Appl Environ Microbiol 2003;69:1936e43. [33] Osorio H, Mangold S, Denis Y, Nancucheo I, Esparza M, Johnson DB, et al. Anaerobic sulfur metabolism coupled to dissimilatory iron reduction in the extremophile Acidithiobacillus ferrooxidans. Appl Environ Microbiol 2013;79:2172e81. [34] Quatrini R, Appia-Ayme C, Denis Y, Jedlicki E, Holmes DS, Bonnefoy V. Extending the models for iron and sulfur oxidation in the extreme acidophile Acidithiobacillus ferrooxidans. BMC Genomics 2009;10:394. [35] Quatrini R, Appia-Ayme C, Denis Y, Ratouchniak J, Veloso F, Valdes J, et al. Insights into the iron and sulfur energetic metabolism of Acidithiobacillus ferrooxidans by microarray transcriptome profiling. Hydrometallurgy 2006;83:263e72. [36] Ronquist F, Huelsenbeck JP. MrBayes 3: Bayesian phylogenetic inference under mixed models. Bioinformatics 2003;19:1572e4. [37] Saiki K, Nakamura H, Mogi T, Anraku Y. Probing a role of subunit IV of the Escherichia coli bo-type ubiquinol oxidase by deletion and crosslinking analyses. J Biol Chem 1996;271:15336e40. [38] Temple KL, Colmer AR. The autotrophic oxidation of iron by a new bacterium, Thiobacillus ferrooxidans. J Bacteriol 1951;62:605e11. [39] Valdes J, Ossandon F, Quatrini R, Dopson M, Holmes DS. Draft genome sequence of the extremely acidophilic biomining bacterium Acidithiobacillus thiooxidans ATCC 19377 provides insights into the evolution of the Acidithiobacillus genus. J Bacteriol 2011;193: 7003e4. [40] Valdes J, Pedroso I, Quatrini R, Dodson RJ, Tettelin H, Blake 2nd R, et al. Acidithiobacillus ferrooxidans metabolism: from genome sequence to industrial applications. BMC Genomics 2008;9:597. [41] Valdes J, Quatrini R, Hallberg K, Dopson M, Valenzuela PD, Holmes DS. Draft genome sequence of the extremely acidophilic bacterium Acidithiobacillus caldus ATCC 51756 reveals metabolic versatility in the genus Acidithiobacillus. J Bacteriol 2009;191:5877e8. [42] Vallenet D, Belda E, Calteau A, Cruveiller S, Engelen S, Lajus A, et al. MicroScope e an integrated microbial resource for the curation and comparative analysis of genomic and metabolic data. Nucleic Acids Res 2013;41:D636e47. [43] Yarzabal A, Brasseur G, Ratouchniak J, Lund K, Lemesle-Meunier D, DeMoss JA, et al. The high-molecular-weight cytochrome c Cyc2 of Acidithiobacillus ferrooxidans is an outer membrane protein. J Bacteriol 2002;184:313e7.