http://www.paper.edu.cn

Journal of Applied Microbiology ISSN 1364-5072

ORIGINAL ARTICLE Isolation and characterization of thermophilum sp. nov., a novel extremely acidophilic, moderately thermophilic archaeon and its role in bioleaching of chalcopyrite H. Zhou1,2, R. Zhang1,P.Hu1, W. Zeng1, Y. Xie1,C.Wu1,3 and G. Qiu1,2

1 School of Minerals Processing and Bioengineering, Central South University, Changsha, P.R. China 2 Key Laboratory of Biometallurgy, Ministry of Education, Central South University, Changsha, P.R. China 3 China Ocean Mineral Resources R&D Association, Beijing, P.R. China

Keywords Abstract 16S rRNA gene, archaeon, chalcopyrite, T Ferroplasma sp., ferrous iron-oxidizing. Aims: To isolate Ferroplasma thermophilum L1 from a low pH environment and to understand its role in bioleaching of chalcopyrite. Correspondence Methods and Results: Using serial dilution method, a moderately thermophilic Guanzhou Qiu, School of Minerals Processing and acidophilic ferrous iron-oxidizing archaeon, named L1T, was isolated from and Bioengineering, Central South University, a chalcopyrite-leaching bioreactor. The morphological, biochemical and physio- Changsha, 410083, P.R. China. logical characteristics of strain L1T and its role in bioleaching of chalcopyrite E-mail: [email protected] were studied. Strain L1T was a nonmotile coccus that lacked cell wall. Strain T 2007 ⁄ 1566: received 26 September 2007, L1 had a temperature optimum of 45°C and the optimum pH for growth was T revised and accepted 24 January 2008 1Æ0. Strain L1 was capable of chemomixotrophic growth on ferrous iron and yeast extract. Results of fatty acid analysis, DNA–DNA hybridization, G+C con- doi:10.1111/j.1365-2672.2008.03807.x tent, and analysis based on 16S rRNA gene sequence indicated that strain L1T should be grouped in the genus Ferroplasma, and represented a new species, Ferroplasma thermophilum. Ferroplasma thermophilum in combination with caldus and Leptospirillum ferriphilum could improve the copper dissolution in bioleaching of chalcopyrite. Conclusions: A novel extremely acidophilic, moderately thermophilic archaeon isolated from a bioleaching reactor has been identified as F. thermophilum that played an important role in bioleaching of chalcopyrite at low pH. Significance and Impact of the Study: This study contributes to understand the characteristics of F. thermophilum L1T and its role in bioleaching of sulfide ores.

techniques have been applied to research the ecology of Introduction the mineral leaching environments, and the results sug- The use of micro-organisms to recover metals from gest that other species of moderately thermophilic bacte- low-grade ores has developed into a successful commer- ria, like Leptospirillum spp. and Acidithiobacillus caldus, cial biotechnology. For many years, it has been thought are dominant at certain mineral leaching environments that mesophilic Acidithiobacillus ferrooxidans (Temple (Goebel and Stackebrandt 1994; Norris et al. 2000). Okibe and Colmer 1951; Rawlings 2002), which grows opti- et al. (2003) found that Ferroplasma species were mally at 25–30°C is the most significant micro-organism dominant in the last bioleaching tank in a series of three. in the leaching of metal sulfides and a major contribu- This was probably the result of the low pH as a tor to (Goebel et al. 2000; Rawlings result of sulfur compound oxidation, the high metal 2002). However, in recent years, molecular phylogenetic concentration and an increased organic carbon

ª 2008 The Authors Journal compilation ª 2008 The Society for Applied Microbiology, Journal of Applied Microbiology 1

转载 中国科技论文在线 http://www.paper.edu.cn

Isolation of Ferroplasma thermophilum sp. nov. H. Zhou et al.

concentration from dead cells from the previous leaching Materials and methods 1tanks in the series. Hawkes et al. (2006a,b) also found that Ferroplasma dominated in a chalcocite heap in Enrichment and isolation Myanmar. The microbial community inhabiting a low- grade copper sulfide run-of-mine test heap was analysed Leachate solution sample was taken from a column reac- 2by PCR-DGGE, and the result indicates that Ferroplasma tor processing chalcopyrite (Key Laboratory of Biometal- groups are quantitatively dominant at certain phases lurgy, Ministry of Education, China). The reactor was during the bioleaching process (Demergasso et al. 2005). originally inoculated with acidic mine drainage sample Ferroplasma groups are often flourishing at low pH, high collected from the Daye copper mine in Hubei province, amounts of total iron, ferrous iron and other heavy met- China. The operating temperature of the reactor was als (Johnson and Hallberg 2003). Reports suggest that 45°C and the pH was 1Æ3–1Æ9. Leachate solution was Ferroplasma spp. increase the sulfide’s leaching efficiency enriched with modified 9K medium (see later) containing at low pH and remove the dead biomass and cell 0Æ02% (w ⁄ v) yeast extract and trace elements (see later). secretions (Okibe and Johnson 2004). Therefore, this Strain L1T was isolated by serial dilution of the enrich- extreme micro-organism has potential commercial ment solution. Pure culture A. ferrooxidans type strain applications. ATCC23270 (AF465604) was purchased from ATCC. Pure culture of this type of organism was first Acidithiobacillus caldus strain s2 (DQ256484) and L. ferr- described in 2000 by Golyshina et al., designated Ferropl- iphilum strain YSK (DQ343299) used in the experiment asma acidiphilum strain YT. The strain is obligately auto- were isolated and conserved by our laboratory. Ferroplas- trophic and was isolated from a -leaching ma cupricumulans BH2T (AY907888) was kindly provided bioreactor in Kazakhstan (Golyshina et al. 2000). Second by Rebecca B. Hawkes (Hawkes et al. 2006a,b). species of the same genus, ‘Ferroplasma acidarmanus’ Fer1T, was subsequently isolated from a high concentra- Growth conditions tion of metals and extremely acidic mine drainage site at Iron Mountain and it is capable of chemo-organo- Unless otherwise stated, strain L1T was cultivated in basal trophic growth on yeast extract or sugars and chemo- salts of modified 9K medium (Silverman and Lundgren mixotrophic growth on ferrous iron and yeast extract or 1959), trace elements (Dopson and Lindstro¨m 1999), 30 g

sugars (Edwards et al. 2000). In 2006, a new species of FeSO4Æ7H2O and 0Æ02% yeast extract. Modified 9K T the genus, Ferroplasma cupricumulans BH2 , was isolated medium contained (per litre): 3Æ0 g (NH4)2SO4,0Æ5g from an industrial mineral sulfide bioleach heap and KH2PO4,0Æ1 g KCl, 0Æ5 g MgSO4Æ7H2O, 0Æ01 g Ca(NO3)2. was only capable of chemomixotrophic growth on fer- Trace elements (per litre) comprised: 11Æ0 mg FeCl3Æ rous iron and yeast extract (Hawkes et al. 2006a,b). 6H2O, 0Æ5 mg CuSO4Æ5H2O, 50 mg Na2SO4,2Æ0mg Unlike mesophilic F. acidiphilum and ‘F. acidarmanus’, H3BO3,2Æ0 mg MnSO4,0Æ8mgNa2MoO4Æ2H2O, 0Æ6mg F. cupricumulans is a moderately thermophilic acido- CoCl2Æ6H2O, 0Æ9 mg ZnSO4Æ7H2O and 0Æ1mgNa2SeO4. phile. The pH of the medium was adjusted to 1Æ0 by adding

Defined consortia of sulfur-oxidizing bacteria A. caldus 50% (v ⁄ v) H2SO4, was autoclaved, and then filter-steril- and iron-oxidizing bacteria, including Sulfobacillus ther- ized (0Æ2 lm filter paper) trace elements and FeSO4Æ7H2O 3mosulfidooxidans and A. ferrooxidans have been used to were added in the medium. Cultures were incubated in study the bioleaching of sulfide minerals (Dopson and rotary shakers at the indicated temperatures. Lindstro¨m 1999; Mcguire et al. 2001). The consortium The optimum temperature and pH for growth of the that comprises autotrophs Leptospirillum MT6 and A. cal- isolated strain L1T were determined by temperature- and dus KU, and the heterotroph Ferroplasma MT17 is the pH-controlled cultures in 250 ml shaking flasks with most efficient among all of those tested in a bioleaching 100 ml medium. Pure culture of strain L1T was inocu- of pyrite experiment (Okibe and Johnson 2004). lated to the liquid medium and the final cell density was In this study, a moderately thermophilic and acido- 1 · 106 cells per ml after inoculation. The incubation was philic ferrous iron-oxidizing archaeon Ferroplasma ther- performed at initial pH 1Æ0 and different temperatures or mophilum L1T was isolated from a chalcopyrite-leaching at 45°C and different initial pH values. After being incu- bioreactor and characterized. To further understand the bated for 4 days, the cell densities of cultures in the flasks role of genus Ferroplasma in mineral bioleaching, our were determined by a cell counting chamber. study was focussed on the copper dissolution from chal- The growth curve of strain L1T was determined by copyrite using pure culture A. ferrooxidans and mixed counting cell numbers in the culture after incubation at cultures A. caldus in combination with F. thermophilum an optimal growth temperature and growth pH value. L1T or Leptospirillum ferriphilum. Strain L1T was cultured on solid medium which contained

ª 2008 The Authors 2 Journal compilation ª 2008 The Society for Applied Microbiology, Journal of Applied Microbiology 中国科技论文在线 http://www.paper.edu.cn

H. Zhou et al. Isolation of Ferroplasma thermophilum sp. nov.

1Æ0% agarose, and the plates were incubated at 45°C The growth of strain L1T on each substrate was moni- for 23 days. tored using a phase contrast microscope after three suc- To determine the optimum concentration of yeast cessive subcultures. extract, the basal salts medium with trace element and ) 30 g l 1 FeSO Æ7H O was supplemented with the follow- 4 2 Anaerobic growth ing different concentrations of yeast extract (w ⁄ v): 0Æ005%, 0Æ01%, 0Æ02%, 0Æ03%, 0Æ05%, 0Æ08%. All cultures Anaerobic growth was tested in an anaerobic tank. The were incubated at 45°C for 4 days, and the cell densities media contained basal salts, trace elements and yeast were determined. extract (0Æ02%) with ferrous iron or ferric iron. The

atmosphere consisted of 21 kPa CO2 and 79 kPa N2. Prior to inoculation, harvested cells were washed in basal Morphology observation salts medium to remove any ferrous iron and ferric iron, The morphology of strain L1T grown on ferrous iron and and the initial cell densities of all cultures were 1 · 106 yeast extract was observed by light microscopy (Olympus cells per ml. To ensure the exclusion of oxygen, the media

4BX-FLA-3 Epifluorescence Microscope; Olympus, Japan), were sparged with N2 for 5 min and the oxygen indica- scanning electron microscope (SEM; JEOL JSM-35 Scan- tors were placed in the anaerobic tank. Lead acetate paper 5ning Electron Microscope, USA) and transmission elec- was placed in the anaerobic tank to indicate whether tron microscope (TEM; JEOL JEM-2200FS Transmission sulfate could be reduced in anaerobic conditions. The Electron Microscope). For SEM preparations, cells were concentration of total iron was analysed by atomic fixed overnight in 2Æ5% (v ⁄ v) glutaraldehyde, dehydrated absorption spectrometry, and ferric iron was calculated by in acetone and sputter-coated with gold (Geng et al. the concentrations of total iron and ferrous iron. Cell 2006), and then viewed with SEM at 15–20 kV. For TEM growth was determined by counting the cell densities after preparations, the method of Holmes et al. (1995) was being incubated stationary at 45°C for 15 days. adopted, and the sample was viewed with TEM at 80 kV. Sensitivity to antibiotics and tolerance to heavy metals Oxidation of ferrous iron The sensitivity to antibiotics and tolerance to some heavy Oxidation of ferrous iron with strain L1T was assessed by metals of strain L1T were monitored by inoculating strain determining the concentrations of ferrous iron with L1T in the media containing basal salts, trace elements,

potassium dichromate (K2Cr2O7) at regular intervals, and ferrous iron, yeast extract (as described earlier). Different the oxidation rates were calculated by the concentration concentrations of ampicillin, chloramphenicol, neomycin, of ferrous iron in liquid medium. kanamycin, streptomycin, tetracycline and erythromycin

or varying concentrations of CuSO4, AgNO3, Pb(NO3)2 and CdCl Æ2Æ5H O were added in the media. Chlorides Growth substrates 2 2 were omitted from the medium when silver tolerance was To determine what kinds of energy source strain L1T uti- assessed. Cell growth was determined by counting the cell lizes, the basal salts medium with trace element and one densities after being incubated at 45°C rotary shaker for 4 of the following compounds (per litre) supplemented days.

were tested, with or without the addition of FeSO4Æ7H2O. The possible substrates: 1 g yeast extract; 1 g peptone; 1 g Whole-cell lipid fatty acid analysis glucose; 1 g sucrose; 1 g galactose; 1 g lactose; 1 g fruc- tose; 1 g maltose; 1 g thiamine; 1 g asparagine; 1 g tyro- Cells of strain L1T and F. cupricumulans BH2T were

sine; 30 g FeSO4Æ7H2O and 1 g yeast extract; respectively harvested by centrifugation, and then trans- 30 g FeSO4Æ7H2O and 1 g peptone; 30 g FeSO4Æ7H2O and ferred directly to a screw-cap vial. The fatty acid methyl 1 g glucose; 30 g FeSO4Æ7H2O; 10 g sodium sulfide; 5 g esters (FAME) were obtained by methylation, saponifica- sulfur; 5 g sodium thiosulfate; 10 g pyrite and 10 g chal- tion and extraction, as described previously (Jantzen et al. copyrite. Sodium thiosulfate, sodium sulfide and FeS- 1993). The separation of FAME was performed by using a

O4Æ7H2O were sterilized by filtration, and sulfur was 6gas chromatography (model DNAI 6500-HR) equipped sterilized by tyndallization. Inocula were prepared by har- with a flame ionization detector (FID) and a 25-m fused vesting cells via centrifugation and washing twice with silica column cross-linked with an SE-30 liquid phase

basal salts medium. The initial cell densities of all cultures (SGE; Ringwood, Victoria, Australia), a carrier gas (H2) ) were 1 · 106 cells per ml, and all cultures were adjusted flow rate of 1Æ2 ml min 1, an injector temperature of

to pH 1Æ0 with H2SO4 and incubated at 45°C for 4 days. 250°C, a detector temperature of 300°C, and an oven

ª 2008 The Authors Journal compilation ª 2008 The Society for Applied Microbiology, Journal of Applied Microbiology 3 中国科技论文在线 http://www.paper.edu.cn

Isolation of Ferroplasma thermophilum sp. nov. H. Zhou et al.

) temperature program of 120–280°Cat4°C min 1. The Table 1 Chemical composition of the chalcopyrite concentrate FAME were identified by comparison with the retention Elements Cu S Fe Pb Zn times of known standards, and their relative proportions were calculated on a Chromatography Data System. Mass% 28Æ49 27Æ36 20Æ13 23Æ14 0Æ88

performed using the following inocula: pure culture of 16S rRNA gene sequencing and phylogenetic analysis A. ferrooxidans operating at 30°C, mixed cultures of A. The total genomic DNA was isolated from 50 ml of late- caldus ⁄ L. ferriphilum, A. caldus ⁄ L1T, A. caldus ⁄ L1T with 7exponential-phase cells, using the CTAB miniprep proto- 0Æ02% (w ⁄ v) yeast extract and A. caldus ⁄ L. ferriphi- col for bacterial genomic DNA preparations (Wilson lum ⁄ L1T operating at 45°C. In the experiment, the min- 1987). The 16S rRNA gene was amplified by PCR using eral concentration was 2% (w ⁄ v), and abiotic controls forward primer 16FpgA and the reverse primer 16RpgA were also designed. The basal salts medium was used in (Golyshina et al. 2000). The PCR program was 94°C for bioleaching experiments and adjusted to pH 1Æ3 (at 45°C)

5 min followed by 32 cycles of 94°C for 45 s, 55°C or 1Æ8 (at 30°C) with H2SO4. The inocula were obtained for 40 s and 72°C for 90 s, and a final extension at 72°C by centrifugation, and the cell densities in the culture for 10 min. The PCR product was purified by gel electro- media were about 1Æ0 · 107 cells per ml after inoculation. phoresis with a 1% agarose gel and was recovered using Bioleaching tests were carried out in 250 ml flasks con- the E.Z.N.A Gel Extraction Kit. The purified PCR product taining 100 ml medium and flasks were shaken at ) was sequenced using an automatic ABI 3730XL DNA ana- 170 rev min 1. Leachate was sampled at regular intervals lyzer (Sunbiotech co. Ltd., Beijing, China). To construct a and the concentration of dissolved copper was analysed phylogenetic tree which could show the relationship of by atomic absorption spectrometry. The leached residues strain L1T with other acidophilic strains, the 16S rRNA were filtered and dried using a freeze drier, and their sequences of some type strains were downloaded from chemical composition was analysed by X-ray diffraction NCBI databases. These downloaded sequences were (XRD). All tests were conducted in triplicate. aligned with the sequence from strain L1T using ClustalX1.80, and the alignment was used to make a Results distance matrix, followed by a neighbor joining tree. Bootstrap analysis was carried out on 1000 replicate input Morphological characterization data sets. The phylogenetic tree was viewed using Tree- view software (Johnson et al. 2006). Cells of the isolated strain L1T appeared as cocci under SEM and TEM (Fig. 1). The diameter of strain L1T ran- ged from 0Æ4to1Æ0 lm and strain L1T was nonmotile. Genomic DNA G+C content and DNA–DNA Flagellum and cell wall were not observed by electron hybridization microscopy. No growth was observed on solid medium The G+C content of strain L1T was determined by using after being incubated at 45°C for 23 days. an HPLC with a Nucleosil 100-5 C-18 column, according to the method described previously (Tamaoka and Koma- Temperature and pH optimum gata 1984; Mesbah et al. 1989). Purified non-methylated k phage DNA (Sigma) was used as a control. Strain L1T could grow at temperatures between 30 and Genomic DNA were also sheared by sonication to 60°C, and the optimum temperature for growth was 45°C produce an average fragment with size of 1 kb for (Fig. 2a). No obvious growth occurred when temperature DNA–DNA hybridization experiments in which the spec- was below 30°C or above 60°C (data not shown). At the trophotometric renaturation rate procedure was used optimum temperature, growth occurred at initial pH 0Æ2 (Huss et al. 1983). The buffer solution used for the DNA– to 2Æ5, with an optimum pH at 1Æ0 (Fig. 2b). DNA hybridization experiments was 33 · SSC (0Æ45 ) ) mol l 1 NaCl plus 0Æ045 mol l 1 trisodium citrate, pH 7Æ0). Growth rates and ferrous oxidation The growth curve of strain L1T is shown in Fig. 3. The Bioleaching experiments generation time of strain L1T grown on ferrous iron with Chalcopyrite used in the experiments was passed through 0Æ02% yeast extract was 6Æ65 h and the maximum specific )1 T a sieve with a pore size of 75 lm. The chemical composi- growth rate (lmax) was 0Æ104 h . Strain L1 oxidized fer- tion of the chalcopyrite concentrate used in the experi- rous iron in the presence of yeast extract, and about 90% ments is shown in Table 1. Bioleaching experiments were ferrous iron was oxidized in the initial 80 h (Fig. 3).

ª 2008 The Authors 4 Journal compilation ª 2008 The Society for Applied Microbiology, Journal of Applied Microbiology 中国科技论文在线 http://www.paper.edu.cn

H. Zhou et al. Isolation of Ferroplasma thermophilum sp. nov.

(a) (b)

Figure 1 Scanning electron (a) and transmis- sion electron (b) micrographs of strain L1T grown chemomixotrophically on ferrous and yeast extract liquid medium at 45°C and pH 1Æ0.

(a) 8·0 8·0 7·8 100 7·8 ) –1 7·6 )

–1 80 7·6 7·4 7·2 60 7·4 7·0 6·8 7·2 40 6·6 oxidation ratio (%) ratio oxidation

7·0 6·4 20 2+

6·2 Fe 6·8 Lg cells density (cells ml 6·0 0 Lg cells density (cells ml 5·8 6·6 0 20 40 60 80 100 120 Time (h) 6·4 30 3540 45 50 55 60 Temperature (°C) ) Growth curve and (.) ferrous iron oxidation curve of strain L1T grown under conditions of optimum pH (1Æ0) and tempera- ° (b) 8·0 ture (45 C).

7·8 ) for growth. However, in the presence of ferrous iron, the –1 7·6 strain L1T could grow in the media containing yeast 7·4 extract, peptone or glucose. In the presence of ferrous iron, the cell density of strain L1T grown on yeast extract 7·2 was higher, compared with the cell densities on peptone 7·0 and glucose. No growth occurred when ferrous iron pre- sented as the sole energy source. Furthermore, growth 6·8 was not detected using the following inorganic energy Lg cells density (cells ml 6·6 sources: sodium sulfide, sulfur, sodium thiosulfate, pyrite or chalcopyrite as the sole energy source. 6·4 T 0·0 0·51·0 1·5 2·0 2·5 3·0 The growth of strain L1 at different concentrations of pH value yeast extract was determined, and the optimum concen- tration of yeast extract was 0Æ02%. Growth of strain L1T Figure 2 Effects of temperature (a) and initial pH values (b) on the was strongly inhibited by the presence of yeast extract in T growth of strain L1 . amounts greater than 0Æ10%. Strain L1T was observed to grow under anaerobic con- dition after being incubated for 15 days. Growth occurred ) on 0Æ02% (w ⁄ v) yeast extract plus 6 g l 1 Fe (SO ) , and Growth on organic and inorganic substrates 2 4 3 the concentration of ferric iron decreased from 1Æ68 to ) Strain L1T was not capable of growth on any of the 1Æ12 g l 1. Meanwhile, the lead acetate paper became organic substrates (described in Materials and Methods) black, which indicated that sulfate was reduced to sulfu- alone, although the addition of yeast extract was essential rated hydrogen under anaerobic condition.

ª 2008 The Authors Journal compilation ª 2008 The Society for Applied Microbiology, Journal of Applied Microbiology 5 中国科技论文在线 http://www.paper.edu.cn

Isolation of Ferroplasma thermophilum sp. nov. H. Zhou et al.

Table 2 Sensitivity of strain L1T to antibiotics Analysis of fatty acid Concentration Growth of the T Æ of antibiotic strain L1T Strain L1 had high levels of 15 : 0 anteiso (10 28%), ) Æ Æ Æ Antibiotic (mg l 1) (Lg cells per ml)* 16 : 0 (21 06%), 18 : 0 (15 96%), 18 : 1 x9c (14 16%) and 18 : 1 x7c (13Æ43%), whereas strain F. cupricumulans Æ Æ No antibiotic 0 7 732 ± 0 031 BH2T had high levels of 16 : 0 (21Æ45%), 18 : 0 (6Æ67%), Ampicillin 10 7Æ299 ± 0Æ038 18 : 1x7c (46Æ36%), and 19 : 0 cyclo x8c (5Æ27%). The 100 6Æ918 ± 0Æ036 Chloramphenicol 10 7Æ171 ± 0Æ036 results indicated that the fatty acid contents of two strains 50 6Æ819 ± 0Æ040 were different, and the content of 18 : 1 x7c in strain T Neomycin 10 7Æ179 ± 0Æ038 F. cupricumulans BH2 was obviously higher. 100 6Æ892 ± 0Æ040 Kanamycin 10 7Æ090 ± 0Æ038 100 6Æ980 ± 0Æ036 G+C content of genomic DNA Streptomycin 10 7Æ068 ± 0Æ041 The G+C content of strain L1T was 34Æ1 mol%. This was 100 6Æ832 ± 0Æ037 Tetracycline 5 6Æ664 ± 0Æ039 consistent with the G+C content of the type strain F. cu- T 10 6Æ101 ± 0Æ039 pricumulans BH2 34Æ0 mol%, and was also close to the T Erythromycin 10 7Æ020 ± 0Æ037 type strains F. acidiphilum Y 36Æ5 mol% (Golyshina et al. 100 6Æ880 ± 0Æ038 2000) and ‘F. acidarmanus’ Fer1T 36Æ8 mol% (Dopson et al. 2004). *Growth was measured by counting cell densities after 4 days. Values are means ± SD (n = 3). Molecular phylogenetic analysis and DNA hybridization

T Antibiotic sensitivity and heavy metal tolerance The 16S rRNA gene sequence of strain L1 (1431 bp) was submitted to GenBank with the accession number T Strain L1 was sensitive to all antibiotics in some EF062309. Phylogenetic relationships based on 16S rRNA T degree at the tested concentration (Table 2). Strain L1 gene sequences were depicted (Fig. 4). The 16S rRNA had slight sensitivities to the following antibiotics: gene sequence of strain L1T was similar to F. cupricumu- ampicillin, chloramphenicol, neomycin, kanamycin, lans BH2T, F. acidiphilum YT and ‘F. acidarmanus’ Fer1T streptomycin and erythromycin that inhibited cell wall with 99%, 96% and 96% sequence similarities, respec- formation in micro-organisms. Tetracycline, which tively. inhibited the synthesis of protein, obviously affected the The DNA–DNA similarity hybridization between F. cupri- T growth of strain L1 . cumulans BH2T and strain L1T was 46Æ3%. To some extent, the growth of strain L1T was inhibited with the increase of heavy metal concentrations (Table 3). T However, strain L1T was hardly inhibited in the presence The role of strain L1 in bioleaching of chalcopyrite of low concentration of silver. In the leaching system inoculated with A. ferrooxidans, 38Æ2% of copper was released from chalcopyrite after 30 T T Table 3 Tolerance of the strain L1T to some heavy metals days. 45Æ4% (A. caldus ⁄ L1 ), 63Æ1% (A. caldus ⁄ L1 with 0Æ02% yeast extract), 69Æ1% (A. caldus ⁄ L. ferriphilum) and Growth of the 73Æ8% (A. caldus ⁄ L. ferriphilum ⁄ L1T) of copper were Concentration strain L1T Metal of metal (g l)1) (Lg cells per ml)* released from chalcopyrite after 30 days in mixed culture leaching systems (Fig. 5a). The ratio of released copper in No metal 0 7Æ732 ± 0Æ031 the bioleaching experiment was much greater than that 2+ Æ Æ Cu 16996 ± 0 028 observed in the sterile experiment (5Æ1% at 30°Cor 26Æ643 ± 0Æ036 13Æ7% at 45°C). The results showed that pure culture A. 46Æ151 ± 0Æ032 Ag+ 0Æ17Æ591 ± 0Æ029 ferrooxidans was not as efficient as mixed cultures except T 0Æ37Æ119 ± 0Æ033 for mixed culture of A. caldus ⁄ L1 without yeast extract Pb2+ 0Æ17Æ374 ± 0Æ026 supplemented. The percentage of released copper was 0Æ56Æ968 ± 0Æ031 greatly enhanced in mixed culture of A. caldus ⁄ L1T with 2+ Cd 0Æ17Æ245 ± 0Æ028 0Æ02% (w ⁄ v) yeast extract (63Æ1%) compared with mixed Æ Æ Æ 0 56986 ± 0 033 culture of A. caldus ⁄ L1T (45Æ4%). The consortium A. cal- T *Growth was measured by counting cell densities after 4 days. Values dus ⁄ L. ferriphilum ⁄ L1 had the strongest ability of leach- are means ± SD (n = 3). ing chalcopyrite (73Æ8%) among all the defined consortia.

ª 2008 The Authors 6 Journal compilation ª 2008 The Society for Applied Microbiology, Journal of Applied Microbiology 中国科技论文在线 http://www.paper.edu.cn

H. Zhou et al. Isolation of Ferroplasma thermophilum sp. nov.

Picrophilus oshimee (X84901)

T 999 Ferroplasma acidiphilum Y (AJ224936) 1000 Ferroplasma sp. MT17 (AF513710) Ferroplasma acidarmanus Fer1T (AF145441) Figure 4 Phylogenetic tree based on frag- ment of 16S rRNA sequences. The tree rooted 1000 L1 (EF062309) with L1T was constructed by the neighbor- 1000 Ferroplasma cupricumulans BH2T (AY907888) joining method with bootstrap values calcu- 740 lated from 1000 trees. The numbers at each Ferroplasma sp. JTC3 (AY830840) clustering node indicate the percentage of Thermoplasma acidophilum (M20822) bootstrap supporting, and in the brackets 1000 after each bacterial name are 16S rDNA Thermoplasma volcanium (AF339746) accession numbers in GenBank. The scale bar 0Æ005 indicates evolutionary distance. 0·005

(a) 4·5 Initial increase of pH was observed in pure and mixed culture leaching systems except consortium A. caldus ⁄ L1T 4·0

) after 8–12 days, the pH started decreasing. However, the –1 3·5 pH variation experienced three phases in leaching system T (g l containing A. caldus and strain L1 . The pH decreased in

2+ 3·0 the leaching system at first, followed by increase after 4 2·5 days, and then decreased again after 12 days (Fig. 5b). 2·0 The leached residues were analysed by XRD (Fig. 6). 1·5 The results indicated that the leached residues from mixed cultures (A. caldus ⁄ L. ferriphilum and A. caldus ⁄ L. 1·0 ferriphilum ⁄ L1T) primarily comprised chalcopyrite, jaro- Concentration of Cu Concentration 0·5 site, anglesite and lead dioxide. However, jarosite was not detected in the leached residues from the abiotic control 0·0 0 4 8 12 16 20 24 28 3 and mixed culture of A. caldus ⁄ L1T with 0Æ02% yeast Time (days) extract. The leached residue of abiotic control mainly contained chalcopyrite, anglesite and lead dioxide, while (b) 2·4 leached residue of mixed culture A. caldus ⁄ L1T mainly 2·2 contained chalcopyrite, pyrite, anglesite and lead dioxide.

2·0 Discussion 1·8 The isolated strain L1T had many similar characteristics as other species of genus Ferroplasma, for example, they 1·6 pH value had no cell wall and were extremely acidophilic. The bio- T 1·4 chemical and physiological characteristics of strain L1 also resembled other species of genus Ferroplasma T 1·2 (Table 4). Strain L1 had 96–99% 16S rRNA gene sequence similarities with other strains of genus Ferroplas- 1·0 ma, and the results were in accordance with the proposal 0 4 8 12 16 20 24 28 32 Time (days) of 93% or 95% 16S rRNA gene sequence similarity as a genus border (Fox et al. 1992; Wiik et al. 1995; Wagner- Figure 5 Copper concentration during the bioleaching of chalcopy- Dobler et al. 2004). The difference of G+C content rite by Acidithiobacillus ferrooxidans and mixed cultures of Acidithio- between strain L1T and other strains of genus Ferroplasma T bacillus caldus, Leptospirillum ferriphilum and L1 (a) and pH changes was less than 3 mol% (Table 4), which was less than the during the bioleaching of chalcopyrite by A. ferrooxidans and mixed defined genus border 10%. According to the results of cultures of A. caldus, L. ferriphilum and L1T (b). ( ) Sterile at 30°C; (d) A. ferrooxidans;( ) sterile at 45°C; (.) A. caldus ⁄ L1T;(b) A. cal- 16S rRNA gene sequence and G+C content analysis, T T dus ⁄ L1T with 0Æ02% yeast extract; (c) A. caldus ⁄ L. ferriphilum;(¤) strain L1 was most close to F. cupricumulans BH2 . T A. caldus ⁄ L. ferriphilum ⁄ L1T. However, DNA–DNA similarity value of strain L1 with

ª 2008 The Authors Journal compilation ª 2008 The Society for Applied Microbiology, Journal of Applied Microbiology 7 中国科技论文在线 http://www.paper.edu.cn

Isolation of Ferroplasma thermophilum sp. nov. H. Zhou et al.

(a) (b) 40 000 25 000

20 000 30 000

15 000 20 000 10 000 Intensity (CPS) Intensity (CPS)

10 000 5000

0 0 10 20 30 40 50 60 70 80 10 20 30 40 50 60 70 80 2-Theta(°) 2-Theta(°)

(c) 60 000 (d) 60 000

50 000 50 000

40 000 40 000

30 000 30 000

Intensity (CPS) 20 000 Intensity (CPS) 20 000

10 000 10 000

0 0 10 20 30 40 50 60 70 80 10 20 30 40 50 60 70 80 2-Theta(°) 2-Theta(°)

Figure 6 X-ray diffraction images of chalcopyrite residues from abiotic control (a), consortium of Acidithiobacillus caldus ⁄ Leptospirillum ferriphi- T T lum (b), consortium of A. caldus ⁄ L1 with 0Æ02% yeast extract (c) and consortium of A. caldus ⁄ L. ferriphilum ⁄ L1 (d). (d) CuFeS2;(¤) PbSO4;( )

PbO2;(.) FeS2;( ) KFe3(SO4)2(OH)6.

F. cupricumulans BH2T was only 46Æ3%, which was less extract may offer growth factors for growth of Ferroplas- than the defined species border of 70% (Wayne et al. ma spp. (Dopson et al. 2004). In its natural environ- 1987). The DNA–DNA similarity data was currently ments, other micro-organisms like fungi might provide viewed as the most reliable method of defining taxonomic this growth factor; hence, the growth of strains of genus relationships (Wayne et al. 1987); therefore, strain L1T Ferroplasma occurred. Strain L1T also could make use of should represent a new species in the genus Ferroplasma. glucose, peptone in the presence of ferrous iron; hence, Based on morphology, physiology, fatty acid, phylogen- the strain was chemomixotrophic and could remove the esis and DNA–DNA similarity, strain L1T represents a organic substances produced by other chemoautotrophic new species of within the genus Ferroplasma and micro-organisms in mixed cultures. The growth of some we propose the name F. thermophilum. chemoautotrophic micro-organisms can be inhibited by Strain L1T grew at pH 0Æ2–2Æ5 with an optimum pH the organic substances that accumulate in mineral leach- of 1Æ0, which means that it was extremely acidophilic and ates (Johnson 2001). This indicates that the presence of could adapt to low pH of bioleaching environments. It strain L1T in the bioleaching system is of significance. has been reported that cell membranes of Ferroplasma As an archaeon, strain L1T does not possess cell wall. spp. contain novel caldarchaetidylglycerol tetraether lip- Therefore, it was not suppressed by ampicillin that could ids, which have extremely low proton permeability, and inhibit cell wall formation. Ferroplasma acidiphilum YT therefore, it could survive at low pH (Golyshina and and ‘F. acidarmanus’ Fer1T had similar results with strain Timmis 2005). 8,9L1T (Golyshina et al. 2000; Dopson et al. 2004). Under aerobic condition, strain L1T grew by oxidizing Under anaerobic condition, strain L1T grew by reduc- ferrous iron in the presence of yeast extract. The yeast ing ferric iron and sulfate, which means that it could play

ª 2008 The Authors 8 Journal compilation ª 2008 The Society for Applied Microbiology, Journal of Applied Microbiology 中国科技论文在线 http://www.paper.edu.cn

H. Zhou et al. Isolation of Ferroplasma thermophilum sp. nov.

Table 4 Comparison of characteristics of strain L1T and other species of Ferroplasma (F.) genus

Characteristic L1T F. cupricumulans BH2T F. acidiphilum YT ‘F. acidarmanus’ Fer1T

Cellular morphology Cocci Irregular cocci Pleomorphic Pleomorphic growth temperature Range (°C) 30–60 22–63 15–45 23–46 Optimum (°C) 45 53 37 42 pH Range 0Æ2–2Æ50Æ4–1Æ81Æ3–2Æ2 0–1Æ5 Optimum 1Æ01Æ0–1Æ21Æ71Æ2 Heterotrophic growth - - + + Under anaerobic condition Reduction of Fe3+ +NR + + 2- Reduction of SO4 +NRNRNR Oxidation of Fe2+ ++ + + DNA G+C content (mol%) 34Æ134Æ036Æ536Æ8 16S rDNA similarity with 99% 96% 96% indicated type strains

Data were taken from Hawkes et al. (2006, 2006), Golyshina et al. (2000) and Dopson et al. (2004). +, positive; -, negative; NR, not reported.

an important role in iron and sulfur cycle in bioleaching 3þ 2À þ þ 3Fe þ2SO4 þ6H2O þ K À! KFe3ðSO4Þ2ðOHÞ6þ6H systems, and sustained or even enhanced the biomass in ð4Þ some zones where oxygen was scarce. Strain L1T is the first strain of genus Ferroplasma reported to be able to Copper concentration was lower in pure culture (at anaerobically reduce sulfate. The high abundances of Fer- 30°C) than those in mixed cultures (at 45°C) except con- T roplasma spp. in pyrite-dominated ores and in mining sortium A. caldus ⁄ L1 , which shows the advantage of environments also suggest that the strains of the genus using moderately thermophilic acidophiles. Compared T play a significant role in the iron and sulfur cycle, and with consortium A. caldus ⁄ L. ferriphilum ⁄ L1 , consortia T are more environmentally important than other iron- and of A. caldus ⁄ L. ferriphilum and A. caldus ⁄ L1 with 0Æ02% sulfur-oxidizing bacteria (Baker and Banfield 2003). yeast extract seemed to be less effective in leaching of For aforementioned reasons, we can explain why Fer- copper from chalcopyrite, which is consistent with other roplasma spp. often flourish in low pH, high concentra- reports (Okibe and Johnson 2004). This indicates that T tions of total iron, ferrous iron and other metals, and strain L1 played an important role in the bioleaching of T moderately elevated temperatures (Bond et al. 2000; chalcopyrite. It was possible that strain L1 removed the Edwards et al. 2000; Johnson and Hallberg 2003). There- organic matter produced by A. caldus and L. ferriphilum. T fore, Ferroplasma are often dominant members in the Furthermore, the iron-oxidizing strain L1 increased the later phase with low pH and high concentrations of heavy ferric iron concentration and enhanced the indirect oxidi- metals during bioleaching (Okibe et al. 2003; Okibe and zation of chalcopyrite derived by ferric iron. Therefore, T Johnson 2004; Hawkes et al. 2006a,b). strain L1 in combination with other moderate thermo- The increase of pH in leaching systems except consor- philes have potential industrial application in bioleaching tium A. caldus ⁄ L1T resulted from the acid dissolution of sulfide ores. [eqn (1)] and ferrous iron oxidation [eqn (2)]. The Description of F. thermophilum sp. nov. Zhou et al. F. decrease of pH after 8–12 days was related to the bacterial thermophilum (ther’mo.phi.lum Gr. n. therm heat; Gr. oxidation of sulfur [eqn (3)] and the precipitation of Adj. philus loving; M. L. adj. thermophilum heat-loving). jarosite [eqn (4)], because these two steps were both acid- 10Cells are sphere-shaped, nonspore-forming archaea, and producing reactions. lack cell wall. Motility is not observed. The diameter of the strain ranges from 0Æ4to1Æ0 lm in modified 9K med- þ 2þ 2þ CuFeS2þ2H þ0Á5O2 À! Cu þ2S þ Fe þH2O ð1Þ ium containing 0Æ02% yeast extract and trace elements. The temperature range for growth is from 30 to 60°C, 2þ þ A:f ;L:f ;Ferroplasma sp: 3þ ° 4Fe þ4H þO2 ƒƒƒƒƒƒƒƒƒƒ! 4Fe þ2H2O ð2Þ with an optimum temperature of 45 C. It grows at pH 0Æ2–2Æ5 with an optimum pH 1Æ0. Strain L1T is A:caldus chemomixotrophic by oxidizing ferrous iron in the pres- 2S þ 3O þ2H O ƒƒƒƒ! 2SO2Àþ4Hþ ð3Þ 2 2 4 ence of yeast extract under aerobic condition and reduced

ª 2008 The Authors Journal compilation ª 2008 The Society for Applied Microbiology, Journal of Applied Microbiology 9 中国科技论文在线 http://www.paper.edu.cn

Isolation of Ferroplasma thermophilum sp. nov. H. Zhou et al.

ferric iron under anaerobic condition. Strain L1T is not natural and commercial bioleaching environments. Appl capable of growth on the inorganic substrates or organic Environ Microb 60, 1614–1621. substrates alone. Major cell membrane fatty acids (>5%) Goebel, B.M., Norris, P.R. and Burton, N.P. (2000) Acido- are: 16 : 0 (21Æ06%), 18 : 0 (15Æ96%), 18 : 1 x9c philes in biomining. In Applied Microbial Systematics ed. (14Æ16%) and 18 : 1 x7c (13Æ43%), 15 : 0 anteiso Priest, F.G. and Goodfellow, M.The Netherlands: Kulwer (10.Æ28%). The G+C content of DNA is 34Æ1% (HPLC). Academic. DNA–DNA hybridization between strain L1T and the type Golyshina, O.V. and Timmis, K.N. (2005) Ferroplasma and rel- strain BH2 of F. cupricumulans is 46Æ3%. The strain has atives, recently discovered cell wall-lacking archaea making been deposited within the China Center for Type Culture a living in extremely acid, heavy metal-rich environments. Environ Microbiol 7, 1277–1288. Collection (CCTCC) (AB 207143T). Golyshina, O.V., Pivovarova, T.A., Karavaiko, G.L., Kon- drat’eva, T.F., Moore, E.R.B., Abraham, W.R., Lunsdorf, Acknowledgements H., Timmis, K.N. et al. (2000) Ferroplasma acidiphilum gen. nov., sp. nov., an acidophilic autotrophic, ferrous- We would like to thank Prof Wenjun Li and Prof Chen- iron-oxidizing, cell-wall-lacking, mesophilic member of gxiang Fang for the latin name of Ferroplasma thermophi- the fam. Nov., comprising a distinct lum. This work was supported by the National Nature lineage of the Archaea. Int J Syst Evol Micr 50, 997– Science Foundation of China (50621063), the China 1006. National Basic Research Program (2004CB619204) and Hawkes, R.B., Franzmann, P.D. and Plumb, J.J. (2006a) the Program for New Century Excellent Talents in Uni- Moderate thermophiles including ‘Ferroplasma cupricu- versity of China (NCET-06-0691). mulans’ sp. nov. dominate an industrial-scale chalcocite heap bioleaching operation. Hydrometallurgy 83, 229– 236. References Hawkes, R.B., Franzmann, P.D., O’hara, G. and Plumb, J.J. Baker, B.J. and Banfield, J.F. (2003) Microbial communities in (2006b) Ferroplasma cupricumulans sp. nov., a novel mod- acid mine drainage. FEMS Microbiol Ecol 44, 139–152. erately thermophilic, acidophilic archaea isolated from an Bond, P.L., Druschel, G.K. and Banfield, J.F. (2000) Compari- industrial-scale chalcocite bioleach heap. Extremophiles 10, son of acid mine drainage microbial communities in phys- 525–530. ically and geochemically distinct ecosystems. Appl Environ Holmes, J.D., Smith, P.R., Evans-Gowing, R. and Richardson, Microb 66, 4962–4971. D.J. (1995) Bacterial photoprotection through extracellular Demergasso, C.S., Galleguillos, P.A., Escudero, L.V., Zepeda, cadmium sulfide crystallites. Photochem Photobiol 62, V.J., Castillo, D. and Casamayor, E.O. (2005) Molecular 1022–1026. characterization of microbial populations in a low-grade Huss, V.A.R., Festl, H. and Schleifer, K.H. (1983) Studies on copper ore bioleaching test heap. Hydrometallurgy 80, 241– spectrophotometric determination of DNA hybridiza- 253. tion from renaturation rates. Syst Appl Microbiol 4, 184– Dopson, M. and Lindstro¨m, E.B. (1999) Potential role of Thio- 192. bacillus caldus in arsenopyrite bioleaching. Appl Environ Jantzen, E., Sonesson, A., Tangen, T. and Eng, J. (1993) Microb 65, 36–40. Hydroxy-fatty acid profiles of legionella species: diagnostic Dopson, M., Baker-Austin, C., Hind, A., Bowman, J.P. and usefulness assessed by principal component analysis. J Clin Bond, P.L. (2004) Characterization of Ferroplasma isolates Microbiol 31, 1413–1419. and Ferroplasma acidarmanus sp. Nov., extreme acido- Johnson, D.B. (2001) Genus Leptospirillum.InBergey’s Manual philes from acid mine drainage and industrial bioleaching of Systematic Bacteriology ed. Boone, D.R., Castenholz, environments. Appl Environ Microb 70, 2079–2088. R.W. and Garrity, G.M.2nd edn pp. 453–456. New York: Edwards, K.J., Bond, P.L., Gihring, T.M. and Banfield, J.F. Springer-Verlag, Vol. 1. (2000) An archaeal iron-oxidizing extreme acidophile Johnson, D.B. and Hallberg, K.B. (2003) The microbiology of important in acid mine drainage. Science 287, 1796–1799. acidic mine waters. Res Microbiol 154, 466–473. Fox, G.E., Wisotzkey, J.D. and Jurtshuk, P. (1992) How close is Johnson, D.B., Kimura, B.S. and Hallberg, K.B. (2006) Isola- close: 16S rRNA sequence identity may not be sufficient to tion and characterization of Acidicaldus organivorus, gen. guarantee species identity. Int J Syst Bacteriol 42, 166–170. nov., sp. nov.: a novel sulfur-oxidizing, ferric iron-reduc- Geng, A., Soh, A.E.W., Lim, C.J. and Loke, L.C.T. (2006) Iso- ing thermo-acidophilic heterotrophic Proteobacterium. Arch lation and characterization of a phenol-degrading bacte- Microbiol 185, 212–221. rium from an industry activated sludge. Appl microbiol Mcguire, M.M., Edwards, K.J., Banfield, J.F. and Hamers, R.J. Biot 71, 728–735. (2001) Kinetics, surface chemistry, and structural evolution Goebel, B.M. and Stackebrandt, E. (1994) Cultural and phylo- of microbially mediated sulfide mineral dissolution. genetic analysis of mixed microbial populations found in Geochim Cosmochim Ac 65, 1243–1258.

ª 2008 The Authors 10 Journal compilation ª 2008 The Society for Applied Microbiology, Journal of Applied Microbiology 中国科技论文在线 http://www.paper.edu.cn

H. Zhou et al. Isolation of Ferroplasma thermophilum sp. nov.

Mesbah, M., Premachandran, U. and Whitman, W. (1989) Temple, K.L. and Colmer, A.R. (1951) The autotrophic oxida- Precise measurement of the G+C content of desoxyribonu- tion of iron by a new bacterium: Thiobacillus ferrooxidans. cleic acid by high performance liquid chromatography. J Bacteriol 62, 605–611. Int J Syst Bacteriol 39, 159–167. Wagner-Dobler, I., Rheims, H., Felske, A., El-Ghezal, A., Norris, P.R., Burton, N.P. and Foulis, N.A.M. (2000) Acido- Flade-Schroder, D., Laatsch, H., Lang, S., Pukall, R. et al. philes in bioreactor mineral processing. Extremophiles 4, (2004) Oceanibulbus indolifex gen. nov., sp. nov., a 71–76. North Sea Alphaproteo-bacterium producing Okibe, N. and Johnson, D.B. (2004) Biooxidation of pyrite by bioactive metabo-lites. Int J Syst Evol Micr 54, 1177– defined mixed culture of moderately thermophilic acido- 1184. philes in pH-controlled bioreactor: significance of micro- Wayne, L.G., Brenner, D.J., Colwell, R.R., Grimont, P.A.D., bial interaction. Biotechnol Bioeng 87, 574–583. Kandler, O., Krichevsky, M.I., Moore, L.H., Moore, W.E. Okibe, N., Gericke, M., Hallberg, K.B. and Johnson, D.B. C. et al. (1987) Report of the ad-hoc-committee on recon- (2003) Enumeration and characterization of acidophilic ciliation of approaches to bacterial systematics. Int J Syst microorganisms isolated from a Pilot plant stirred-tank Bacteriol 37, 463–464. bioleaching operation. Appl Environ Microb 69, 1936–1943. Wiik, R., Stackebrandt, E., Valle, O., Daae, F.L., Rodseth, O.M. Rawlings, D.E. (2002) Heavy metal mining using microbes. and Andersen, K. (1995) Classification of fish-pathogenic Annu Rev Microbiol 56, 65–91. vibrios based on comparative 16S rRNA analysis. Int J Syst Silverman, M.P. and Lundgren, D.G. (1959) Studies on the Bacteriol 45, 421–428. chemoautotrophic iron bacterium Ferrobacillus ferrooxi- Wilson, K.(1987) Preparation of genomic DNA from bacte- dans: I. An improved medium and harvesting procedure ria. In Current Protocols in Molecular Biology ed. Ausu- for securing high cell yields. J Bacteriol 77, 642–647. bel, F.M., Brent, R., Kingston, R.E., Moore, D.D., Tamaoka, J. and Komagata, K. (1984) Determination of DNA Seidman, J.G., Smith, J.A. and Struhl, K.New York: base composition by reversed-phase high-performance Wiley. liquid chromatography. FEMS Microbiol Lett 25, 125–128.

ª 2008 The Authors Journal compilation ª 2008 The Society for Applied Microbiology, Journal of Applied Microbiology 11