Journal of Applied Microbiology ISSN 1364-5072

1 ORIGINAL ARTICLE 2 3 Sedimentary arsenite-oxidizing and arsenate-reducing 4 bacteria associated with high arsenic groundwater from 5 6 Shanyin, Northwestern 7 H. Fan1,C.Su2, Y. Wang1, J. Yao2, K. Zhao1, Y. Wang2 and G. Wang1 8 9 1 State Key Laboratory of Agricultural Microbiology, College of Life Science and Technology, Huazhong Agricultural University, Wuhan, China 10 2 Key Laboratory of Biogeology and Environmental Geology of Ministry of Education, China University of Geosciences, Wuhan, China 11 12 13 14 15 Keywords Abstract aoxB 16 , arsenate-reducing bacteria, arsenic, arsenite-oxidizing bacteria, groundwater. Aims: Shanyin County is one of the most severe endemic arsenism affected 17 areas in China but micro-organisms that potentially release arsenic from sedi- 18 Correspondence ments to groundwater have not been studied. Our aim was to identify bacteria 19 Gejiao Wang PhD, State Key Laboratory of with the potential to metabolize or transform arsenic in the sediments. 20 Agricultural Microbiology, College of Life Methods and Results: Culture and nonculture-based molecular methods were Science and Technology, Huazhong 21 performed to identify arsenite-oxidizing bacteria, arsenate-reducing bacteria Agricultural University, Wuhan 430070, 22 and arsenite oxidase genes. Arsenite-oxidizing bacteria were identified only 23 China. E-mail: [email protected]; [email protected] from the land surface to 7 metres (m)-deep underground that were affiliated to 24 a- and b-Proteobacteria. Arsenate-reducing bacteria were found in almost all 25 2007/1589: received 29 September 2007, the sediment samples with different depths (0–41 m) and mainly belong to 26 revised 18 January 2008 and accepted 19 c-Proteobacteria. Several novel arsenite oxidase genes (aoxBs) were identified 27 January 2008 from the upper layers of the sediments (0–7 m) and were found to be specific 28 for arsenite-oxidizing bacteria. doi:10.1111/j.1365-2672.2008.03790.x 29 Conclusions: The distribution of arsenite-oxidizing bacteria in upper layers and 30 arsenate-reducing bacteria in different depths of the sediments may impact the 31 arsenic release into the nearby tubewell groundwater. 32 Significance and Impact of the Study: This study provides valuable sources of 33 micro-organisms (and genes) that may contribute to groundwater arsenic 34 abnormality and may be useful to clean arsenic contaminated groundwater. 35 36 37 38 found in Shanyin county of province where Introduction 39 groundwater is the only water supply. About 50 000 40 Arsenic (As) is widely distributed and toxic to humans people drink groundwater with arsenic concentration ) 41 and animals. Arsenic pollution has become a severe reaching up to 1Æ93 mg l 1 resulting in severe clinical 42 worldwide problem in recent years, especially in Bangla- symptoms of arsenism (Guo et al. 2003). 43 desh, India and China (Chowdhury et al. 2000; Sun There is an urgent need to develop low cost and effi- 44 2004). Chronic intake of groundwater with high levels of cient technologies to clean arsenic from groundwater in 45 arsenic has caused arsenism in several . this region. For this goal, we first have to gain knowledge 46 The arsenic contaminated area of China was found to be of biotic and abiotic factors that contribute to the arsenic 47 larger than that of Bangladesh and new endemic regions enrichment in such environment. Shanyin is located in 48 were continuously emerging (Xia and Liu 2004). Shanxi the south of the Basin. The arsenic rich metamor- 49 province is a geological abnormal region for arsenic. Well phic rocks and coal bearing strata around the margin of 50 water in all the investigated villages showed elevated con- the basin are the natural origins of arsenic sources. Lacus- 51 centration of arsenic and 52% of the tubewell water is trine deposits with high organic materials are the second- 52 unsafe (Sun 2004). Endemic arsenism cases caused by ary media containing arsenic. High arsenic is mainly 53 drinking arsenic contaminated groundwater were mainly found in tubewell groundwater between 10 and 50 m

ª 2008 The Authors Journal compilation ª 2008 The Society for Applied Microbiology, Journal of Applied Microbiology 1 JAM 3790 Dispatch: 14.3.08 Journal: JAM CE: Ulagammal Journal Name Manuscript No. B Author Received: No. of pages: 11 PE: Karuppiah Arsenic resistant bacteria in groundwater H. Fan et al.

1 below land surface (Pei et al. 2005). A previous study tion of groundwater (Islam et al. 2004). The effect of 2 showed that chemical reducing condition was a key factor microbial oxidation on groundwater environment is still 3 affecting the dissolution of arsenic from sediments to unclear. Especially in China, there is no study of micro- 4 groundwater and the authors pointed out that the trans- bial species related to the enrichment and transportation 5 formation of arsenic may mainly caused by microbial of arsenic in groundwater. 6 metabolisms (Guo et al. 2003). The aim of this study was to investigate sedimentary 7 Many micro-organisms that interact and transform arsenite-oxidizing bacteria, arsenate-reducing bacteria and 8 inorganic arsenic species have been identified. Certain arsenite oxidase genes (aoxBs) from a well characterized 9 bacteria could oxidize arsenite [As(III)] to arsenate arsenic rich site of Shanyin County. Results of this study 10 [As(V)] and they are termed arsenite-oxidizing bacteria will provide both valuable information on microbial spe- 11 or arsenite oxidizers. Such oxidation process is considered cies with arsenic redox metabolisms that may contribute 12 as a detoxification metabolism because As(V) is much less to the arsenic abnormality of groundwater in this region 13 toxic than As(III). Furthermore, As(V) is negative charged and valuable sources of arsenic resistant micro-organisms 14 and easy to be absorbed, thus arsenite-oxidizing bacteria that may be useful for bioremediation of arsenic contami- 15 have been used in cleaning system of arsenic contami- nated groundwater. 16 nated water (Lie`vremont et al. 2003). Some arsenite- 17 oxidizing bacteria have been isolated from arsenic Materials and methods 18 contaminated environments and classified as Achromo- 19 bacter, Agrobacterium, Alcaligenes, Hydrogenophaga, Pseu- Site description and sampling 20 domonas and Thermus etc. (Osborne and Enrlich 1976; 21 Philips and Taylor 1976; Abdrashitova et al. 1985; Santini Sediment and water samples were collected at Gucheng 22 et al. 2000; Oremland et al. 2002; Salmassi et al. 2002). Village, Shanyin County (Fig. 1). Shanyin is an arid- 23 The As(III) oxidation reaction was catalysed by a peri- semiarid region which is located in the south of Datong 24 plasmic arsenite oxidase. This enzyme contains two Cenozoic Basin, Shanxi Province. Alluvial and lacustrine 25 subunits encoded by the genes aoxA (small Fe-S Rieske sediments were accumulated in the Basin. It was reported 26 subunit) and aoxB (large Mo-pterin subunit) respectively that arsenic is concentrated in the bottomland between 27 (Mukhopadhyay et al. 2002; Silver and Phung 2005). Shangan River and Huangshui River (Fig. 1), as well as in 28 Recently the aoxB gene encoding the large subunit of the downfold connecting with pluvial plain and alluvial- 29 arsenite oxidase has been found in different soil and laustrine plain (Pei et al. 2005). Aquifers in the sediments 30 water systems containing arsenic (Inskeep et al. 2007), are mainly composed of alluvial–lacustrine silty sands and 31 but has not been identified in Chinese arsenic contami- silts that are rich in organic matters and arsenic in such 32 nated groundwater environments so far. enclosed-semienclosed geological environment (Guo et al. 33 Many arsenate-reduction bacteria (or arsenate reducers) 2003). 34 capable to reduce As(V) to As(III) have also been identi- Sediment samples were collected using the rotary dril- 35 fied and classified as Bacillus, Clostridium, Citrobacter, ling method to make a borehole between two ground- 36 Desulfomicrobium, Sulfurospirillum and Wolinella etc. water tubewells (W1 and W2) that were 6 and 10 m away 37 (Silver and Phung 2005). The reduction of arsenate by respectively (Table1). The water surfaces of W1 and W2 38 micro-organisms occurs via two mechanisms, dissimila- were about 34 and 59 m deep underground respectively. 39 tory reduction and detoxification. Dissimilatory reduc- The borehole was drilled from the surface to 41 m under- 40 tions of arsenate are carried out by microbes either strict ground. The water and sediments were kept in sealed 41 anaerobic or facultative anaerobic that couple growth plastic bags at 4°C during transport and storage. The ) 42 with arsenate as the terminal electron acceptor (Malasarn water samples were filtered through 0Æ45 lmol l 1 mem- 43 et al. 2004). The arsenic detoxification was first catalysed branes for chemical analysis using AFS (AFS-830, Titan 44 by the cytoplasmic arsenate reductase ArsC to reduce Instruments, Beijing, China). Analyses of the sediment 45 As(V) to As(III) and later used an efflux pump to extrude samples were performed as described by Guo et al. 46 As(III) outside the cell (Silver and Phung 2005). (2003). Sediment samples were stored at 4°C for 1 week 47 The impacts of microbial arsenite oxidation and arse- before bacterial and DNA isolation. 48 nate reduction have been reported to associate with 49 arsenic cycle of soil and saline lakes (Ahmann et al. 1997; Isolation and characterization of As(III)-oxidizing 50 Macur et al. 2004; Oremland et al. 2004). It was reported bacteria and As(V)-reducing bacteria 51 that in Bengal delta plain sediments, dissimilatory arse- 52 nate reduction caused arsenic transformation from solid A total of nine arsenic contaminated sediment samples 53 phase into aqueous phase, resulting in arsenic contamina- with different depths were used in this study (4–41 m,

ª 2008 The Authors 2 Journal compilation ª 2008 The Society for Applied Microbiology, Journal of Applied Microbiology H. Fan et al. Arsenic resistant bacteria in groundwater

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 Figure 1 Sketch GPS map of Datong Basin and approximate location of the sampling site (marked as ¯Gc). 33 34 ) 35 Table 1 Inorganic component concentrations (mg l 1) of the groundwater samples in two tubewells near the sediment collecting borehole site in 36 Shanyin, Datong Basin 37 Distance from Well ) ) ) 2) ) 2) + 2+ 38 Well the borehole depth* pH Alkalinity F Cl NO3 SO4 HCO3 HPO4 Na Mg As 39 W1 6 m 35 m 8Æ11 481Æ28 1Æ63 410 1Æ97 75Æ8 586Æ58 2Æ84 339Æ31 74Æ39 1Æ09 40 W2 10 m 60 m 8Æ38 300Æ80 0Æ40 17Æ60Æ06 0Æ67 366Æ61 0Æ88 87Æ78 29Æ66 0Æ20 41 42 *The water surfaces are about 34 and 59 m deep underground, for W1 and W2 respectively. 43 Total arsenic. 44 45 Table 2). For enrichment, 100 g of each sediment sample were incubated at 27°C for 1 week. Single colonies were

46 was amended with Na-As(III) (NaAsO2) at a final con- re-streaked several times to obtain pure isolates. ) 47 centration of 500 mg l 1 and kept for 1 week at 27°C. The obtained arsenic resistant bacteria were tested for

48 Arsenic resistant bacteria were isolated from each sedi- their abilities to oxidize As(III) (NaAsO2) or reduce 49 ment sample by adding 10 g sediment (triplicates) to As(V) (Na3AsO4) using a qualitative KMnO4 screening 50 100 ml of 0Æ85% NaCl solution and shaking for 10 min. method (Salmassi et al. 2002). Each arsenic resistant bac- 51 The extraction solution was serially diluted and plated terium was cultured to 0Æ4 OD in CDM liquid medium )1 52 onto chemically defined medium (CDM) plates (Weeger containing 100 mg l either NaAsO2 or Na3AsO4. Then )1 )1 53 1et al. 1999) containing 100 mg l NaAsO2. The plates 20 llof0Æ01 mol l KMnO4 was added to 1 ml of the

ª 2008 The Authors Journal compilation ª 2008 The Society for Applied Microbiology, Journal of Applied Microbiology 3 Arsenic resistant bacteria in groundwater H. Fan et al.

1 Table 2 Characteristics of the arsenic contaminated sediments with different depths 2 Name of 3 sediment Depth Texture of Arsenic TOC* No. of As(III)-oxidizing No. of As(III)-oxidase No. of As(V)-reducing No. of nonoxidizing ⁄ 4 sample (m) (m) sediments pH (mg kg)1) (%) bacteria genes bacteria reducing bacteria 5 6 4 0–4 Silty soil 6Æ31 7Æ99 0Æ28 7 10 2 1 7 4–7 Fine sand 6Æ78 3Æ09 0Æ15 4 4 4 9 7 10 7–10 Silt 8Æ40 4Æ21 0Æ18 0 0 17 3 8 15 10–15 Clay 8Æ60 3Æ68 0Æ12 0 0 2 4 9 19 15–19 Fine sand 7Æ98 12Æ29 0Æ37 0 0 0 0 10 23 19–23 Clay 7Æ30 12Æ46 0Æ14 0 0 14 6 11 28 23–28 Clay 8Æ11 4Æ98 0Æ20 0 0 16 4 12 34 28–34 Silt 7Æ93 26Æ25 0Æ10 0 0 8 4 13 41 34–41 Clay 8Æ91 4Æ34 0Æ31 0 0 7 0 14 *Total organic carbon. 15 16 culture. A pink colour of the mixture indicated positive amplified by PCR using universal primers Uni-27F and 17 arsenite oxidation reaction [formation of As(V)], and a Uni-1492R (Wilson et al. 1990). PCR amplification was 18 yellow colour indicated positive arsenate reduction reac- performed in a ATC 201 Thermal Cycler (Apollo, San 19 tion [formation of As(III)]. Bacterial arsenite oxidation or Diego, CA, USA) in a 50 ll volume containing 10 ng ) ) 20 arsenate reduction was further determined by measuring DNA, 2 lmol l 1 of each primer, 200 lmol l 1 of each )1 21 arsenate quantum with a spectrophotometer (DU800; dNTP, 50 mmol l MgCl2,2Æ5 ll dimethyl sulfoxide, ) 22 2BeckMan, CA, USA) using the molybdene blue method 5 ll of 10x PCR buffer [100 mmol l 1 Tris–HCl (pH 3) )1 23 (Lenoble et al. 2003). Arsenate reacts with MoO4 to 8Æ3), 100 mmol l KCl] and 0Æ5 units of Taq DNA poly- 24 form an arsenate-molybdate complex. This complex reacts merase (Fermentas, Hanover, MD, USA). The PCR pro- 25 with ascorbic acid to produce a blue colour solution that gramme consisted of an initial denaturation for 5 min at 26 can be measured at 846 nm. For arsenite oxidation tests, 94°C; 30 cycles of 1 min at 94°C, 1 min at 50°C, 1 min 27 a single colony of each isolate was inoculated into 100 ml at 72°C; and a final extension step at 72°C for 5 min.

28 CDM medium containing NaAsO2 and incubated at 27°C Sequencing analyses of the PCR products were performed ) 29 with 150 rev min 1 shaking. Five millilitres of the culture as described below. 30 was taken every hr, 3 ml was used to check the OD values 31 and the remaining centrifuged. Three-hundred microlitres Identification of aoxB gene encoding arsenite oxidase 32 of the supernatant was mixed with 5 ml ddH O, 400 ll 2 large subunit from the arsenic resistant strains and the 33 of 50% H SO , 200 ll of 3% ascorbic acid, 400 llof3% 2 4 sediments 34 Na3MoO4 and boiled at 100°C for 10 min. This mixture 35 3was then made up to 10 ml with ddH2O and measured Total DNA from each sediment sample (Table 2) was 36 for the OD value at 846 nm. A standard curve with extracted using the Ultra Clean Soil DNA Kit as described )1 37 Na3AsO4 concentration from 0 to 400 lmol l was used. in the manufacturer’s instruction (MoBio Laboratories, 38 An As(III)-oxidizing bacteria Agrobacterium sp. C13 was Solana Beach, CA, USA). DNA of the strains with or 39 used as a positive control (Xiong et al. 2006) and the without arsenite oxidation abilities was isolated using a 40 nonbacterial inoculated CDM medium containing the modified protocol (Sambrook et al. 1989). One millilitre

41 same amount of NaAsO2 was used as an abiotic control. of middle-log-phase cells was centrifuged at 12 000 ) 42 A similar method was used to detect the arsenate reduc- rev min 1 for 5 min and resuspended in 400 ll Lysis buf- )1 )1 43 tion. The difference was adding Na3AsO4 to the medium fer (40 mmol l Tris–HCl, pH 8Æ0, 20 mmol l NaAC, )1 44 instead of NaAsO2. 1 mmol l EDTA and 2% SDS). After brief mixing, the 45 Minimum inhibitory concentrations (MICs) of the mixture was incubated at 65°C for 30 min. Then incuba- ) 46 arsenic resistant bacteria were determined by inoculating tion, 100 ll of 5 mol l 1 NaCl was added and mixed ) 47 a single colony of each isolate in triplicate into liquid before centrifuging at 12 000 rev min 1 for 5 min. The 48 CDM medium supplemented with increasing concentra- supernatant was extracted twice with equal volume of phe-

49 tions of NaAsO2 and shaking at 27°C for 1 week. The nol ⁄ chloroform ⁄ isoamyl alcohol (25 : 24 : 1) and centri- ) 50 MIC is defined as the lowest As (III) concentration that fuged at 12 000 rev min 1 for 5 min. DNA was 51 completely inhibited the growth of each bacterium. precipitated with 2 volumes of 100% ethanol and washed 52 The nearly full-length 16S rRNA genes of the As(III)- with 70% ethanol. After centrifugation at 12 000 ) 53 oxidizing strains and As(V)-reducing strains were rev min 1 for 20 min, the pellet was air dried and

ª 2008 The Authors 4 Journal compilation ª 2008 The Society for Applied Microbiology, Journal of Applied Microbiology H. Fan et al. Arsenic resistant bacteria in groundwater

) 1 resuspended in 50 ll TE buffer (10 mmol l 1 Tris, rRNA gene and aoxB sequences were checked manually ) 2 1 mmol l 1 EDTA, pH 8Æ0). The DNA was incubated at and edited for phylogenetic analyses. Sequence alignments 3 4°C for overnight and later stored at )20°C. were performed using the clustalW program (Thomp- 4 All DNA samples were further purified using the Ultra- son et al. 1994). BlastN (for 16S rRNA gene) and 5 PureTM PCR Kit (SBS Genetech, Shanghai, China). Four BlastX searching programs (for deduced aoxB amino 6 degenerate primers for aoxB were designed using premier acid sequences) were used to analyze similarities (http:// 7 5 program (http://www.premierbiosoft.com/primerdesign/ 6www.ncbi.nlm.nih.gov/BLAST). Phylogenetic trees were 8 4primerdesign.html) based on five large subunit sequences generated from alignments by the neighbour-joining 9 of arsenite oxidases (AAQ19838, ABB51928, CAD53341, method and the reliability of inferred trees was tested 10 AAR05656, AAN05581) from NCBI GenBank (http:// with bootstrap test using the mega3 program (http:// 11 5www.ncbi.nlm.nih.gov). They are AoxA-F: 5¢-ACV TTC 7www.megasoftware.net). Some reference sequences from 12 AAS TGY CCH KGY CAY TTC-3¢; AoxB-R1: 5¢-TGR the GenBank were used in generating phylogenetic trees 13 TTN AGR AAR TAR TTN GTY TG-3¢; AoxB-F: 5¢-TGY for clarification. 14 CAY TTY TGY ATH GTN GGN TG-3¢ and AoxB-R2: 15 5¢-TAN GCN GGN CGR TTR TGD AT-3¢. Nucleotide sequence accession numbers 16 PCR amplification of the aoxB gene fragment was per- 17 formed using primers AoxA-F and AoxB-R1 for the first The NCBI GenBank accession numbers for the 16S rRNA 18 PCR, and primers AoxB-F and AoxB-R2 for the nested gene sequences are: arsenite-oxidizing bacteria GW1–4, 19 PCR. A touch-down program was used: after initial dena- EF550171–EF550174; arsenate-reducing bacteria GW5–14, 20 turation at 94°C for 5 min, 34 cycles at 94°C for 1 min, EF550155–EF550164; The accession numbers of the 21 55–46°C for 45 s (decreased by 1°C from every cycle in aoxB1–aoxB14 gene sequences are EF550141–EF550154. 22 the first nine cycles) and 72°C for 1 min were performed, 23 and then the reaction mixture was kept at 72°C for 5 min. Results 24 The PCR products (c. 500 bp) were separated in a 1Æ5% 25 agarose gel and visualized by ethidium bromide staining. Characterization of the water and the sediment 26 To construct aoxB gene libraries of 4 and 7 m sedi- 27 ments, the PCR products were purified using the Gel The sediment samples were taken from a borehole between 28 Extraction Kit (Watson Biotechnologies, Beijing, China). two tubewells (W1 and W2). The groundwater of the two 29 The purified PCR products were ligated with pGEM-T tubewells was characterized as high pH, weak alkaline, high ) 2) 30 (Promega, Madison, WI, USA) and the ligation products content of HCO3 and low concentration of HPO4 and ) 2) 31 were used to transform E. coli DH5a by electroporation. NO3 . The SO4 concentration was lower than that of ) 32 The transformants were grown on LB agar containing Cl . The groundwater of W1 tubewell (6 m far from the 33 Ampicillin, X-Gal and IPTG at 37°C for 16 h according borehole site) showed higher As concentration ) ) 34 to manufacturer’s recommendations. PCR-RFLP analysis (1Æ09 mg l 1) than that of W2 tubewell (0Æ2mgl 1) which 35 was performed for 50 randomly selected clones from the is located 10 m away from the borehole site (Table 1). 36 two aoxB gene libraries (4 and 7 m). Each bacterial col- The depth of the borehole was 41 m underground. The 37 ony was picked from the plates and suspended in 50 ll texture types included silt, fine sand and clay that varied

38 sterile H2O in a PCR tube. The tube was kept at 100°C among different layers of the sediments. The sediments 39 for 5 min and later centrifuged for 5 min at were rich in organic carbon (>0Æ1%). The highest As con- ) 40 10 000 rev min 1. Five microlitres of above mixture was centration was detected in the 34 m sediment sample ) 41 used in PCR using primers T7 and SP6 (Promega) using (26Æ25 mg kg 1). The average As concentration of the ) 42 the same condition as described before. Enzyme digestion sediment samples from this borehole was 13Æ37 mg kg 1. 43 was performed at 37°C for 3 h in a 20 ll volume con- Generally, the concentration of As(III) was higher than 44 taining 1· enzyme buffer, 0Æ5 units of HaeIII or AIuI and that of As(V) and the ratio of As(III) ⁄ As(V) increased 45 15 ll of the PCR products. The digested DNA fragments with the depth of the sediments (Guo et al. 2003). The 46 were separated in 2% agarose gels and the digestion top two upper layer sediments were weakly acidic (pH 47 patterns were grouped by DNA fingerprinting profiles. 6Æ31 and 6Æ78). The pH value of the down layer sediments 48 from 7 to 41 m were all higher than 7Æ3 (Table 2). 49 DNA sequencing and phylogenetic analysis 50 The As(III)-oxidizing and As(V)-reducing bacteria 51 The PCR products were purified using the UltraPureTM 52 PCR Kit (SBS Genetech). DNA sequencing was performed A total of 112 arsenic resistant bacterial strains were iso- 53 by SUNBIO Company (Beijing, China). All of the 16S lated from sediment samples of different depths except

ª 2008 The Authors Journal compilation ª 2008 The Society for Applied Microbiology, Journal of Applied Microbiology 5 Arsenic resistant bacteria in groundwater H. Fan et al.

1 for the 19 m sediment sample. Using KMnO4 biochemical method and the results of arsenite oxidation are shown in 2 analysis, 11 As(III) oxidizers (11 ⁄ 112 = 9Æ8%) and 70 Fig. 3. 3 As(V) reducers (70 ⁄ 112 = 62Æ5%) were identified. The The As(III)-oxidizing bacteria were isolated only from 4 rest 31 arsenic resistant strains (27Æ7%) did not show the first two upper layer sediments (0–7 m, Table 2). 5 As(III) oxidation or As(V) reduction abilities. The num- BlastN searches of the16S rRNA genes of our As(III)- 6 bers of the isolates from different layers of the sediments oxidizing strains showed that the closest matches for 7 are shown in Table 2. After screening for the bacterial GW1, GW2, GW3 and GW4 were members of Achromo- 8 morphologies and PCR-RFLP fingerprinting patterns, the bacter (EF025350, 99%), Acidovorax (AF235013, 98%), 9 11 As(III) oxidizers were later identified as 4 different Sinorhizobium (DQ337550, 99%) and Agrobacterium tum- 10 As(III)-oxidizing strains (GW1, GW2, GW3 and GW4) efaciens (DQ647054, 99%) respectively (Fig. 2). Phyloge- 11 and the 70 As(V) reducers were identified as 10 different netic analysis of the 16S rRNA genes showed that the 12 As(V) reducing strains (GW5–GW14, Fig. 2). The quanti- four As(III) oxidizers were divided into two clusters rep- 13 tative As(III) oxidation or As(V) reduction tests of these resenting a-Proteobacteria (GW3 and GW4) and b-Prote- 14 14 strains were later performed using the molybdene blue obacteria (GW1 and GW2) (Fig. 2). 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49

50 Figure 2 A 16S rRNA gene neighbour-joining tree showing the genetic relationship of the As(III)-oxidizing strains (GW1–GW4), As(V)-reducing 51 strains (GW5–GW14) and some published microbial species in the GenBank. Numbers at branch nodes show bootstrap values obtained from 52 1000 re-samplings. Scale bar 0Æ05 = 5% difference among the nucleotide sequences. Tree construction was performed as described in ‘Materials 53 and methods’.

ª 2008 The Authors 6 Journal compilation ª 2008 The Society for Applied Microbiology, Journal of Applied Microbiology H. Fan et al. Arsenic resistant bacteria in groundwater

) 1 tant bacteria included GW1 (15 mmol l 1), GW3 ) ) ) 2 (9 mmol l 1), GW4 (25 mmol l 1), GW6 (14 mmol l 1), ) ) 3 GW9 (14 mmol l 1), GW11 ⁄ 12 (7 mmol l 1) and GW13 ) 4 (8 mmol l 1). 5 The 31 arsenic resistant isolates without As(III) oxida- 6 tion or As(V) reduction abilities were identified as three 7 bacterial strains, one is Bacillus cereus and two are Pseu- 8 domonas spp. (data not shown). 9 10 The arsenite oxidation efficiencies of the 11 arsenite-oxidizing bacteria 12 13 All of the four arsenite-oxidizing bacteria were heterotro- 14 phic. They did not grow with arsenite as a sole energy 15 source, and their growth were not enhanced in the pres- 16 ence of arsenite (data not shown). As shown in Fig. 3, 17 As(III) oxidation of all the four arsenite oxidizers 18 occurred during logarithmic growth that started at 14Æ5, 19 38Æ0, 48Æ0 and 20Æ5 h after inoculation and finished at 20 19Æ5, 59Æ0, 62Æ0 and 32Æ5 h, for GW1, GW2, GW3 and 21 GW4 respectively. The arsenite oxidation rates were cal- ) ) 22 8culated as 270Æ2, 30Æ3, 47Æ9 and 110Æ2 lmol l 1 h 1, for 23 GW1, GW2, GW3 and GW4 respectively. No obvious 24 oxidation was found in abiotic controls (Fig. 3). During 25 the incubation period, the concentration of As (V) 26 increased with the increase in the cell density (data not Figure 3 (a) Arsenite oxidation analysis of the arsenite-oxidizing bac- 27 teria GW2 ( ), GW3(h) and the abiotic control ( ). (b) Arsenite oxi- shown). The sensitivity of the molybdene blue method is )1 28 dation analysis of the arsenite-oxidizing bacteria GW1 ( ), GW4 (h) 0Æ1 lmol l . 29 and the abiotic control ( ). Fresh culture of each strain was adjusted 30 to the same ODs (0Æ4), diluted 1 : 500, and inoculated into each CDM )1 )1 The aoxB genes were identified from As(III)-oxidizing 31 broth containing 0Æ67 mmol l NaAsO2 for (a), and 1Æ33 mmol l bacteria and from the two upper layers sediments 32 NaAsO2 for (b). The concentration of As(V) was detected at OD846 as described in ‘Materials and methods’. Error bars represent the varia- 33 Each DNA of the nine sediment samples from different tion of the results from three replicate experiments. 34 depths (Table 2) was amplified by PCR using degenerate 35 primers for aoxB partial sequences. Meanwhile, same PCR 36 As(V)-reducing bacteria were identified in almost all was also performed using each DNA of 20 arsenic resis- 37 of the sediment samples (0–41 m, Table 2), except for tant bacteria without arsenite oxidation abilities and the 38 the 19 m sample, from which, no arsenic resistant bacte- DNA of the four arsenite oxidizers. The aoxB gene frag- 39 ria were obtained after repeated attempts. The 16S rRNA ments were amplified from the four arsenite oxidizers 40 genes of As(V) reducing strains GW6–13 were affiliated (GW1–4) and from the two upper layers sediments (4 41 to c-Proteobacteria group including Pseudomonas and and 7 m samples, Table 2). No aoxB products were 42 Acinetobacter. The 16S rRNA genes of As(V) reducing amplified from the arsenic resistant bacteria without 43 strains GW5 and GW14 showed 98% and 99% nucleo- As(III) oxidation abilities and from the deeper sediments 44 tide identities with those of Sphingomonas parapaucimo- (10–41 m samples, Table 2). The PCR amplification of 45 bilis (D84525) (a-Proteobacteria group) and Arthrobacter each sediment DNA was successful using 16S rRNA gene 46 sp. (X83408) (Actinobacteria group) respectively (Fig. 2). primers. This confirmed the good quality of the sediment 47 Members of Pseudomonas (GW9, 10, 11 and 12) were DNAs (data not shown). 48 the major As(V) reducers that were identified in most of The aoxB PCR products from the two upper layer sedi- 49 the sediment samples with different depths (data not ments were then cloned into E. coli to construct two aoxB 50 shown). gene libraries. After PCR-RFLP analysis, 10 different aoxB 51 All of the four As(III)-oxidizing bacteria and the 10 clones were identified and sequenced (aoxB1–10). Mean- 52 As(V)-reducing bacteria were resistant to arsenite (MICs: while, the aoxB fragments of the four arsenite oxidizers ) 53 2–25 mmol l 1). The higher MICs of the arsenic resis- were also sequenced [aoxB-GW(1–4)]. Four of these 10

ª 2008 The Authors Journal compilation ª 2008 The Society for Applied Microbiology, Journal of Applied Microbiology 7 Arsenic resistant bacteria in groundwater H. Fan et al.

1 aoxB sequences from the aoxB gene libraries, aoxB1–4, (ABD35886). The aa sequence of aoxB5 showed 59% aa 2 showed identical nucleotide sequences with the aoxB identity with the arsenite oxidase Mo-pterin subunit of 3 sequences of the four cultured As(III)-oxidizers aoxB- an environment sequence (ABD35897). 4 GW(1–4) respectively. Six aoxB sequences (aoxB6–10) The a-Proteobacteria cluster included aoxB3, 4, 8, 9 5 were new environmental clones (Fig. 4). The distribution and aox10 (Fig.4). The aoxB8 sequence had 88% aa 6 of aoxBs in 4 and 7 m sediment samples was shown in identity with the large subunit of the arsenite oxidase 7 Table 2. of Xanthobacter autotrophicus (ESA15672). The aoxB4 8 Phylogenetic analysis divided the aoxB sequences into (=aoxB-GW4) and aoxB3(=aoxB-GW3) sequences 9 two clades that were affiliated to a- and b-Proteobacteria showed 98% and 96% aa identities, respectively, to AroA 10 (Fig. 4). The b-Proteobacteria group includes aoxB1, 2, 5, subunit of the gold mine chemoautotrophic arsenite 11 6 and aox7. The aoxB1(=aoxB-GW1) sequence had 80% 9oxidizer NT-26 (Santini et al. 2001). The aoxB3(=aoxB- 12 amino acid (aa) identity with the arsenite oxidase GW3) had 84% aa identity with the large arsenite oxi- 13 Mo-pterin subunit of Alcaligenes faecalis (AAQ19838). dase subunit of A tumefaciens (ABB51928). The aoxB10 14 The aoxB2(=aoxB-GW2), aoxB6, and aoxB7 showed sequence had 77% aa identity with the corresponding 15 76%, 77% and 86% aa identities, respectively, to an region of the aoxB from an environment sequence 16 arsenite oxidase Mo-pterin subunit of Variovorox sp. (ABD35951) (Fig. 4). 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49

50 Figure 4 A aoxB gene based neighbor-joining tree showing the deduced amino acid similarities of the aoxBs of our study and some published 51 aoxB sequences. aoxB-GW(1–4) represent aoxBs of the four arsenite oxidizers (GW1–4). aoxB 1–10 are aoxBs from the first two upper layers of 52 the sediments. Numbers at nodes show bootstrap values obtained from 1000 re-samplings. Scale bar 0Æ1 = 10% difference among the amino acid 53 sequences.

ª 2008 The Authors 8 Journal compilation ª 2008 The Society for Applied Microbiology, Journal of Applied Microbiology H. Fan et al. Arsenic resistant bacteria in groundwater

1 As(III) to be the prevalent chemical species in such con- Discussion 2 dition, and because As(III) absorbs to fewer minerals, it is 3 The investigation of microbial species with arsenic redox much easier (and notably more toxic) to be transported 4 biotransformation abilities is very useful for studying the into aqueous phase. This is probably the reason for the 5 impact of sedimentary micro-organisms on the release of formation of the high arsenic concentration in the nearby 6 arsenic in groundwater and bioremediation. In this study, tubewell groundwater. 7 we described the distribution and characterization of Arsenite oxidation reactions of all the four As(III) oxi- 8 As(III)-oxidizing and As(V)-reducing bacteria in sedi- dizers GW1–4 began when bacteria grew to the logarith- 9 ments containing high levels of naturally occurring mic phase. Such phenomenon was in agreement with that 10 arsenic. As(III)-oxidizing bacteria were identified only of several reported arsenite oxidizers (Santini et al. 2000; 11 from the first two upper layers of the sediments (0–7 m), Salmassi et al. 2002). This may indicate the involvement 12 while As(V)-reducing bacteria were found in almost all of of the arsenite induced quorum-sensing signal transduc- 13 the sediment samples (0–41 m depth). To our knowledge, tion mechanism (Gihring and Banfield 2001; Gihring 14 this is the first study to examine both arsenite-oxidizing et al. 2001; Kashyap et al. 2006). The arsenite oxidation ) ) 15 and arsenate-reducing micro-organisms from a Chinese rates of GW2 (30Æ3 lmol l 1 h 1) and GW3 (47Æ9 lmol ) ) 16 arsenic groundwater contaminated environment. l 1 h 1) seem similar to that of Agrobacterium sp. C13 ) ) 17 The presence of both As(III)-oxidizing and As(V)- (29 lmol l 1 h 1, Xiong et al. 2006) and Agrobacterium ) ) 18 reducing bacteria in the upper layer sediments suggests albertimagni AOL15 (9Æ2 lmol l 1 h 1, Scudlark and ) ) 19 that the aerobic ⁄ facultative-aerobic upper sediments may 10Johnson 1982); The rate of GW4 (110Æ2 lmol l 1 h 1) 20 permit the microbial activities of both arsenite oxidizers appears to be close to that of an arsenite-oxidizing con- ) ) 21 and arsenate reducers. In natural sediments, arsenic is sortium (160 lmol l 1 h 1, Battaglia-Brunet et al. 2002); ) ) 22 generally present in anions, including arsenite and arse- GW1 (270Æ2 lmol l 1 h 1) and appears to have higher 23 nate. After digging tubewells, certain oxidation condition arsenite oxidation rate comparable to those isolated 24 could be established through transportation of aerobic previously. 25 groundwater and diffusion of oxygen through vadose Among the four arsenite oxidizers (GW1–4), GW1, 26 zone. The As(III)-oxidizing bacteria may utilize sediment GW3 and GW4 that belong to three genera identified in 27 organic carbon to grow and oxidize As(III). The oxida- this study were also previously identified as arsenite oxi- 28 tion product As(V) tends to be immobilized by absorp- dizers by other researchers (Osborne and Enrlich 1976; 29 tion onto the solid phase, such as Fe oxide, goethite and Philips and Taylor 1976; Salmassi et al. 2002). To our 30 gibbsite, ferrihydrite and clay minerals (Goldberg 1986; knowledge, strain GW2 that affiliated to Acidovorax is a 31 Waychunas et al. 1993; Guo et al. 2003). In some reduc- newly identified arsenite oxidizer. 32 ing micro-environments of the upper layer sediments, the Using nonculture-based aoxB library method, we iden- 33 As(V) -reducing bacteria may transform As(V) to soluble tified six more environmental aoxB sequences. Further, 34 As(III) form, release arsenic from the minerals and clays, the aoxB library method revealed four aoxB sequences 35 and may further transport arsenic to the deeper layers. identical with that of the four cultured As(III) oxidizers. 36 In the deeper sediments, no As(III)-oxidizing bacteria These results confirmed the reliability of the gene library 37 were identified and the bacteria affiliated with As(V)- approach and proved that nonculture-based molecular 38 reducing groups were dominant. This phenomenon may method can detect more micro-organisms from environ- 39 associate with the reduction environment (higher pH in ments. Phylogenetic analyses of the arsenite oxidizers 40 the deeper sediments etc.). A previous study indicated generated from both the 16S rRNA genes and the aoxBs 41 that dissimilatory As(V) reduction was related to the revealed that a- and b-Proteobacteria are the major arse- 42 arsenic release from sediments to groundwater in India nite oxidizers in this environment. 43 (Islam et al. 2004). In this study, we found that aero- The aoxB genes were detected from the four arsenite 44 bic ⁄ facultative-aerobic As(V) reducers were also dominant oxidizers and from the first two upper layer sediments 45 in the deeper sediments. As our bacterial isolation was only. No aoxB genes were detected in bacteria without 46 performed in aerobic condition, anaerobic As(V) reducers arsenite oxidation abilities and from deeper layers. The 47 have not been determined in this study. It was reported aoxB primers used in this study have also been tested in 48 that certain facultative-aerobic conditions may still exist other arsenic resistant bacteria in our laboratory, and 49 in deeper sediment aquifers (Guo et al. 2003). Our results aoxB has never been amplified from any nonarsenite oxi- 50 suggest that not only microbial dissimilatory As(V) dizing bacteria. These results indicate that aoxBs may be 51 reduction but also aerobic ⁄ facultative-aerobic As(V) specific for certain arsenite-oxidizing bacteria and useful 52 reduction may contribute to the As(III) enrichment in in detecting arsenite-oxidizing micro-organisms in the 53 deeper sediments. The reduction reaction resulting in environment. Inskeep et al. (2007) reported that arsenite-

ª 2008 The Authors Journal compilation ª 2008 The Society for Applied Microbiology, Journal of Applied Microbiology 9 Arsenic resistant bacteria in groundwater H. Fan et al.

1 oxidase genes are widely present in different arsenite Goldberg, S. (1986) Chemical modeling of arsenate adsorption 2 oxidizers and widespread in soil-water systems. However, on aluminum and iron oxide minerals. Soil Sci Soc Am J 3 an exception was found, genome annotation of a faculta- 50, 1154–1157. 4 tive arsenite oxidizer MLHE-1 (Alkalilimnicola erhlichii) Guo, H., Wang, Y., Shpeizer, G.M. and Yan, S. (2003) Natural 5 showed that it does not have a similar aoxB gene as aero- occurrence of arsenic in shallow groundwater, Shanyin, 6 bic As(III) oxidizers have (Hoeft et al. 2007). Datong Basin, China. J Environ Sci Health A. Tox Hazard 7 Our results indicate that the high arsenic groundwater Subst Environ Eng 38, 2565–2580. 8 environments permit the selection of highly arsenic resis- Hoeft, S.E., Blum, J.S., Stolz, J.F., Tabita, F.R., Witte, B., King, 9 tant bacterial population. The distribution of different G.M., Santini, J.M. and Oremland, R.S. (2007) Alkalilimni- cola ehrlichii sp. nov., a novel, arsenite-oxidizing 10 sedimentary arsenic resistant bacteria species may associ- haloalkaliphilic gamma proteobacterium capable of 11 ate with the geological factors and impact arsenic cycling chemoautotrophic or heterotrophic growth with nitrate or 12 between sediments and groundwater. The bacterial oxida- oxygen as the electron acceptor. Int J Syst Evol Microbiol 13 tion of As(III) to As(V) decreases arsenic toxicity and 57, 504–512. 14 increases absorption. Such bacteria have been applied in Inskeep, W.P., Macur, R.E., Hamamura, N., Warelow, T.P., 15 batch reactors together with immobilizing materials for Ward, S.A. and Santini, J.M. (2007) Detection, diversity 16 removing arsenic from wastewater (Lie`vremont et al. and expression of aerobic bacterial arsenite oxidase genes. 17 2003; Simeonova et al. 2005). Arsenite oxidizers therefore Environ Microbiol 9, 934–943. 18 are good candidates for cleaning of arsenic contaminated Islam, F.S., Gault, A.G., Boothman, C., Polya, D.A., Charnock, 19 groundwater. J.M., Chatterjee, D. and Lloyd, J.R. (2004) Role of metal- 20 reducing bacteria in arsenic release from Bengal delta sedi- 21 Acknowledgements ments. Nature 430, 68–71. 22 Kashyap, D.R., Botero, L.M., Lehr, C., Hassett, D.J. and 23 We thank Huaiqing Liu for sample collections. This work McDermott, T.R. (2006) Complex regulation of arsenite 24 was funded by the National Natural Science Foundation oxidation in Agrobacterium tumefaciens. J Bacteriol 188, 25 of China, No. 30570058 for G.W., No. 40425001 for Y.W. 1081–1088. 26 and No. 40673065 for J.Y. Lenoble, V., Deluchat, V., Serpaud, B. and Bollinger, J.C. 27 (2003) Arsenite oxidation and arsenate determination by 28 the molybdene blue method. Talanta 61, 267–276. References 29 Lie`vremont, D., N’negue, M.A., Behra, P. and Lett, M.C. 30 Abdrashitova, S.A., Abdullina, G.G. and Ilialetdinov, A.N. (2003) Biological oxidation of arsenite: batch reactor 31 (1985) Glucose consumption and dehydrogenase activity experiments in presence of kutnahorite and chabazite. 32 of the cells of the arsenite-oxidizing bacterium Pseudomo- Chemosphere 51, 419–428. Macur, R.E., Langner, H.W., Kocar, B.D. and Inskeep, W.P. 33 nas putida. Mikrobiologiia 54, 679–681. (2004) Linking geochemical processes with microbial com- 34 Ahmann, D., Krumholz, L.R., Hemond, H.F., Lovley, D.R. and munity analysis: successional dynamics in an arsenic-rich, 35 Morel, F.M.M. (1997) Microbial mobilization of arsenic acid-sulphate-chloride geothermal spring. Geobiology 2, 36 from sediments of the Aberjona Watershed. Environ Sci 163–177. 37 Technol 31, 2923–2930. Battaglia-Brunet, F., Dictor, M.C., Garrido, F., Crouzet, C., Malasarn, D., Saltikov, C.W., Campbell, K.M., Santini, J.M., 38 Hering, J.G. and Newman, D.K. (2004) arrA is a reliable 39 Morin, D., Dekeyser, K., Clarens, M. and Baranger, P. (2002) An arsenic(III)-oxidizing bacterial population: marker for As(V) respiration. Science 306, 455. 40 selection, characterization and performance in reactors. Mukhopadhyay, R., Rosen, B.P., Phung, L.T. and Silver, S. 41 J Appl Microbiol 93, 656–667. (2002) Microbial arsenic: from geocycles to genes and 42 Chowdhury, U.K., Biswas, B.K., Chowdhury, T.R., Samanta, enzymes. FEMS Microbiol Rev 26, 311–325. 43 G., Mandal, B.K., Basu, G.C., Chanda, C.R., Lodh, D. et al. Oremland, R.S., Hoeft, S.E., Santini, J.M., Bano, N., Hollib- 44 (2000) Groundwater arsenic contamination in Bangladesh augh, R.A. and Hollibaugh, J.T. (2002) Anaerobic oxida- 45 and West Bengal, India. Environ Health Perspect 108, 393– tion of arsenite in Mono Lake water and by a facultative, 46 397. arsenite-oxidizing chemoautotroph, strain MLHE-1. Appl 47 Gihring, T.M. and Banfield, J.F. (2001) Arsenite oxidation and Environ Microbiol 68, 4795–4802. 48 arsenate respiration by a new Thermus isolate. FEMS Oremland, R.S., Stolz, J.F. and Hollibaugh, J.T. (2004) The 49 Microbiol Lett 204, 335–340. microbial arsenic cycle in Mono Lake, California. FEMS 50 Gihring, T.M., Druschel, G.K., McCleskey, R.B., Hamers, R.J. Microbiol Ecol 48, 15–27. 51 and Banfield, J.F. (2001) Rapid arsenite oxidation by Ther- Osborne, F.H. and Enrlich, H.L. (1976) Oxidation of arsenite 52 mus aquaticus and Thermus thermophilus: field and labora- by a soil isolate of Alcaligenes. J Appl Bacteriol 41, 295– 53 tory investigations. Environ Sci Technol 35, 3857–3862. 305.

ª 2008 The Authors 10 Journal compilation ª 2008 The Society for Applied Microbiology, Journal of Applied Microbiology H. Fan et al. Arsenic resistant bacteria in groundwater

1 Pei, H.H., Liang, S.X. and Ning, L.Y. (2005) A discussion of Sun, G. (2004) Arsenic contamination and arsenicosis in 2 the enrichment and formation of arsenic in groundwater China. Toxicol Appl Pharmacol 198, 268–271. 3 in Datong Basin. Hydrogeol Eng Geol 4, 65–69. Thompson, J.D., Higgins, D.G. and Gibson, T.J. (1994) CLUS- 4 Philips, S.E. and Taylor, M.L. (1976) Oxidation of arsenite to TAL W: improving the sensitivity of progressive multiple 5 arsenate by Alcaligenes faecalis. Appl Environ Microbiol 32, sequence alignment through sequence weighting, position- 6 392–399. specific gap penalties and weight matrix choice. Nucleic 7 Salmassi, T.M., Venkateswaren, K., Satomi, M., Nealson, K.H., Acids Res 22, 4673–4680. 8 Newman, D.K. and Hering, J.G. (2002) Oxidation of arse- Waychunas, G.A., Rea, B.A., Fuller, C.C. and David, J.A. 9 nite by Agrobacterium albertimagni, AOL15, sp. nov., (1993) Surface chemistry of ferrihydrite: part I. isolated from Hot Creek, California. Geomicrobiol J 19, EXAFS studies of the geometry of coprecipitated and 10 53–66. adsorbed arsenate. Geochim Cosmochim Acta 57, 2251– 11 Sambrook, J., Fritsch, E.F. and Maniatis, T. (1989) Molecular 2269. 12 Cloning: A Laboratory Manual, 2nd edn. Cold Spring Har- Weeger, W., Lievremont, D., Perret, M., Lagarde, F., Hubert, 13 bor, NY: Cold Spring Harbor Laboratory Press. J.C., Leroy, M. and Lett, M.C. (1999) Oxidation of arsenite 14 Santini, J.M., Sly, L.I., Schnagl, R.D. and Macy, J.M. (2000) A to arsenate by a bacterium isolated from an aquatic envi- 15 new chemolithoautotrophic arsenite-oxidizing bacterium ronment. Biometals 12, 141–149. 16 isolated from a gold mine: phylogenetic, physiological, and Wilson, K.H., Blitchington, R.B. and Greene, R.C. (1990) 17 preliminary biochemical studies. Appl Environ Microbiol Amplification of bacterial 16S ribosomal DNA with poly- 18 66, 92–97. merase chain reaction. J Clin Microbiol 28, 1942–1946. 19 Scudlark, J.R. and Johnson, D.L. (1982) Biological oxidation of Xia, Y.J. and Liu, J. (2004) An overview on chronic arse- 20 arsenite in seawater. Estuar Coast Shelf Sci 14, 693–706. nism via drinking water in PR China. Toxicology 198, 21 Silver, S. and Phung, L.T. (2005) Genes and enzymes involved 25–29. 22 in bacterial oxidation and reduction of inorganic arsenic. Xiong, J., Wang, W., Fan, H., Cai, L. and Wang, G. (2006) 23 Appl Environ Microbiol 71, 599–608. Arsenic resistant bacteria in mining wastes from Shangrao 24 Simeonova, D.D., Micheva, K., Muller, D.A., Lagarde, F., Lett, coal mine of China. Environmental Science and Technol- 25 M.C., Groudeva, V.I. and Lie`vremont, D. (2005) Arsenite ogy (I). In Environmental Science and Technology (I) ed. 26 oxidation in batch reactors with alginate-immobilized Lyon, W.G., Starrett, S.K. and Hong, J. pp. 536–540. Cam- 27 ULPAs1 strain. Biotechnol Bioeng 91, 441–446. bridge, MA, USA: American Science Press. 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53

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

Journal: JAM

Article: 3790

Dear Author, During the copy-editing of your paper, the following queries arose. Please respond to these by marking up your proofs with the necessary changes/additions. Please write your answers on the query sheet if there is insufficient space on the page proofs. Please write clearly and follow the conventions shown on the attached corrections sheet. If returning the proof by fax do not write too close to the paper’s edge. Please remember that illegible mark-ups may delay publication.

Many thanks for your assistance.

Query reference Query Remarks 1 Au: Weeger and Lievremont 1999 has been changed to Weeger et al. 1999 so that this citation matches the list. 2 Au: Please provide city name here. 3 Au: The term ‘brought’ has been changed to ‘made up’. Please check. 4 Au: Please check this website address and confirm that it is correct. 5 Au: Please check this website address and confirm that it is correct. 6 Au: Please check this website address and confirm that it is correct. 7 Au: Please check this website address and confirm that it is correct. ) ) 8 Au: Please check the unit ‘lmol l 1 h 1’ here and subsequent occurrences. 9 Au: Santini et al. 2001 not found in the list. Please provide publication details. 10 Au: Scudlark 1982 has been changed to Scudlark and Johnson 1982 so that this citation matches the list. MARKED PROOF Please correct and return this set

Please use the proof correction marks shown below for all alterations and corrections. If you wish to return your proof by fax you should ensure that all amendments are written clearly in dark ink and are made well within the page margins.

Instruction to printer Textual mark Marginal mark Leave unchanged under matter to remain Insert in text the matter New matter followed by indicated in the margin or Delete through single character, rule or underline or or through all characters to be deleted Substitute character or substitute part of one or through letter or new character or more word(s) through characters new characters Change to italics under matter to be changed Change to capitals under matter to be changed Change to small capitals under matter to be changed Change to bold type under matter to be changed Change to bold italic under matter to be changed Change to lower case Encircle matter to be changed Change italic to upright type (As above) Change bold to non-bold type (As above) or Insert ‘superior’ character through character or under character where required e.g. or

Insert ‘inferior’ character (As above) over character e.g. Insert full stop (As above) Insert comma (As above) or and/or Insert single quotation marks (As above) or

or and/or Insert double quotation marks (As above) or Insert hyphen (As above) Start new paragraph No new paragraph Transpose Close up linking characters

Insert or substitute space through character or between characters or words where required

Reduce space between between characters or characters or words words affected