Characterization of Microbial Arsenate Reduction in ...

This article was downloaded by: [Ronald Oremland] On: 23 January 2013, At: 05:59 Publisher: Taylor & Francis Informa Ltd Registered in England and Wales Registered Number: 1072954 Registered office: Mortimer House, 37-41 Mortimer Street, London W1T 3JH, UK

Geomicrobiology Journal Publication details, including instructions for authors and subscription information: http://www.tandfonline.com/loi/ugmb20

Characterization of Microbial Arsenate Reduction in the Anoxic Bottom Waters of Mono Lake, California Shelley E. Hoeft , Franc¸oise Lucas , James T. Hollibaugh & Ronald S. Oremland Version of record first published: 10 Nov 2010.

To cite this article: Shelley E. Hoeft , Franc¸oise Lucas , James T. Hollibaugh & Ronald S. Oremland (2002): Characterization of Microbial Arsenate Reduction in the Anoxic Bottom Waters of Mono Lake, California, Geomicrobiology Journal, 19:1, 23-40 To link to this article: http://dx.doi.org/10.1080/014904502317246147

PLEASE SCROLL DOWN FOR ARTICLE Full terms and conditions of use: http://www.tandfonline.com/page/ terms-and-conditions This article may be used for research, teaching, and private study purposes. Any substantial or systematic reproduction, redistribution, reselling, loan, sub-licensing, systematic supply, or distribution in any form to anyone is expressly forbidden. The publisher does not give any warranty express or implied or make any representation that the contents will be complete or accurate or up to date. The accuracy of any instructions, formulae, and drug doses should be independently verified with primary sources. The publisher shall not be liable for any loss, actions, claims, proceedings, demand, or costs or damages whatsoever or howsoever caused arising directly or indirectly in connection with or arising out of the use of this material.

Geomicrobiology Journal, 19:23 – 40, 2002 C 2002 Taylor & Francis Copyright ° 0149-0451 /02 $12.00 + .00

Characterization of Microbial Arsenate Reduction in the Anoxic Bottom Waters of Mono Lake, California SHELLEY E. HOEFT

Downloaded by [Ronald Oremland] at 05:59 23 January 2013

U.S. Geological Survey Menlo Park, California, USA

FRANC ¸ OISE LUCAS JAMES T. HOLLIBAUGH Department of Marine Sciences University of Georgia Athens, Georgia, USA

RONALD S. OREMLAND U.S. Geological Survey Menlo Park, California, USA Dissimilatory reduction of arsenate (DAsR) occurs in the arsenic-rich, anoxic water column of Mono Lake, California, yet the microorganisms responsible for this observed in situ activity have not been identi ed. To gain insight as to which microorganisms mediate this phenomenon, as well as to some of the biogeochemical constraints on this activity, we conducted incubations of arsenate-enriched bottom water coupled with inhibition/amendment studies and Denaturing Gradient Gel Electrophoresis (DGGE) characterization techniques. DAsR was totally inhibited by lter-sterilization and by nitrate, partially inhibited (»50%) by selenate, but only slightly (»25%) inhibited by oxyanions that block sulfate-reduction (molybdate and tungstate). The apparent inhibition by nitrate, however, was not due to action as a preferred electron acceptor to arsenate. Rather, nitrate addition caused a rapid, microbial re-oxidation of arsenite to arsenate, which gave the overall appearance of no arsenate loss. A similar microbial oxidation of As(III) was also found with Fe(III), a fact that has implications for the recycling of As(V) in Mono Lake’s anoxic bottom waters. DAsR could be slightly (10%) stimulated by substrate amendments of lactate, succinate, malate, or glucose, but not by acetate, suggesting that the DAsR micro ora is not electron donor limited. DGGE analysis of ampli ed 16S rDNA gene fragments from incubated arsenate-enriched bottom waters revealed the presence of two bands that were not present in controls without added arsenate. The resolved sequences of these excised bands indicated the presence of members of the epsilon (Sulfurospirillum) and delta (Desulfovibrio) subgroups of the Received 1 April 2001; accepted 3 July 2001. This research was supported by the Of ce of Naval Research grant N00014-99-1-010 7 and the Superfund Basic Research Program (NIEHS) grant, ES100337, from the National Institutes of Health. We are grateful to R. Jellison for providing us with water samples. This work was supported by the U.S. Geological Survey (RSO and SHE), and by an NSF Microbial Observatories grant (#MCB 99-77886) to JTH. Address correspondenc e to Ronald S. Oremland. E-mail: [email protected]

23

24

S. E. Hoeft et al. Proteobacteria; both of which have representative species that are capable of anaerobic growth using arsenate as their electron acceptor. Keywords

arsenate respiration, anaerobes, soda lakes


Introduction Dissimilatory reduction of arsenate (DAsR) to arsenite is a means of anaerobic respiration whereby certain prokaryotes couple the reduction of arsenate [As(V)] to the gain in energy from oxidation of organic electron donors or H2 to achieve growth. Respiratory arsenate reductases are located in the cell envelope (Krafft and Macy 1998; Stolz and Oremland 1999; Oremland et al. 2002) and are thus distinguished from the arsC detoxifying reductases which are located in the cytoplasm and do not provide cells a means of energy generation (Cervantes et al. 1994). Currently, seven novel species from the domain Eubacteria have been isolated that can achieve growth by DAsR, including both Gram-positive (Newman et al. 1997) and Gram-negative (Macy et al. 2000) sulfate-reducers. In addition, two representatives of the Crenoarchaea, Pyrobaculum arsenaticum and P. aerophilum achieve chemoautotrophic growth by DAsR using H2 as their electron donor (Huber et al. 2000). It is now evident that the phenomenon of DAsR, which was discovered somewhat recently (Ahmann et al. 1994), occurs among widely diverse representatives of both the Eubacterial and Archaeal Domains. Included are the freshwater strains Sulfurospirillum barnesii and S. arsenophilum (Stolz et al. 1999), as well as the haloalkaliphiles Bacillus arsenicoselenatis and B. selenitireducens (Switzer Blum et al. 1998). The last two organisms were isolated from the bottom sediments of Mono Lake, California, an alkaline (pH D 9.8), hypersaline (salinity D 75 – 90 g/l), soda lake. These isolates proved well adapted to the harsh chemical conditions occurring therein. However, the ability to culture microorganisms retrieved from environmental samples that can achieve DAsR does not prove that they are also the primary causative agents for this reaction under in situ conditions. We chose Mono Lake as a site to investigate biogeochemical arsenic cycling because its waters contain exceptionally high concentrations (»200 ¹M) of inorganic arsenic. These high concentrations are a consequence of large cumulative inputs from hydrothermal springs coupled with the high evaporation rates characteristic of this arid region. The chemical speciation of arsenic completely changes from arsenate [As(V)] to the more reduced arsenite [As(III)] form with vertical transition from the lake’s surface oxic waters (mixolimnion) to its unmixed, anoxic bottom waters (monimolimnion ) (Maest et al. 1992). Incubation experiments conducted with lakewater samples amended with 73 As(V) revealed that this reductive speciation change was a microbial rather than a chemical process, and that significant populations of As(V)- and sulfate-respiring bacteria were present in the anoxic waters (Oremland et al. 2000). Indeed, As(V) is the second most abundant anaerobic electron acceptor present in the lake after sulfate (95– 130 mM). During meromixis, DAsR mineralizes 8 to 14% of annual phytoplankton productivity in the strati ed water column (Oremland et al. 2000). In this paper we describe inhibition experiments with arsenate-amended bottom waters from Mono Lake conducted in an attempt to de ne the physiological group(s) of microbes involved in DAsR, and to identify some of the factors constraining this activity. We have further coupled these incubations with molecular techniques (PCR/DGGE) to characterize micro ora that responded to As(V) amendment, a technique that has already been employed to characterize the microbes associated with selenate reduction in estuarine sediments (Lucas and Hollibaugh 2001). We also report on the observation of microbial oxidation of As(III) with nitrate or Fe(III) as the electron acceptor, a process that is of signi cance to the internal regeneration of arsenate ions in the anoxic monimolimnion waters of this meromictic lake.

Arsenate Reduction in Mono Lake

25

Materials and Methods


Water Column Sampling and Incubation Experiments Details of the methodology have been published previously (Oremland et al. 2000), but are reviewed brie y here. Water samples were taken from 24 or 28 m depth in the anoxic monimolimnion (dissolved sul de >1 mM) during September to December 2000 at a station located in the western basin of the lake (Oremland et al. 1987). Samples were recovered with a Van Dorn sampler, drained to over owing into 1-liter glass bottles, and sealed using a frosted glass plug, avoiding any entrapped air bubbles. Samples were shipped on ice overnight to the USGS laboratory in Menlo Park, California. The samples were stored at »6± C in the dark until the experiments commenced (usually within 1– 4 weeks of receipt). Bottles were opened under N2 , and water was drawn from the bottom of the bottle into a 50-ml syringe. Duplicate or triplicate subsamples (8 ml) were dispensed into 10-ml Glaspak syringes, capped, and sealed with rubber- lled cutoff hubs. Electron acceptors, inhibitors, and electron donors from concentrated stock solutions were added by syringe injection (vol. D 0.1 – 0.3 ml). Ferric iron was added as Fe(III)-nitrolotriacetic acid (NTA). The samples were incubated in the dark (double wrapped in Al foil) at room temperature (»20± C). Discrete samples were expelled from the syringes at various time intervals to generate progress curves. For the DGGE experiments, larger volumes of lakewater were required in order to obtain suf cient DNA for extraction. Therefore, we employed 500-ml serum bottles that contained 350 ml of lakewater sample. The gas phase was N2 and bottles were closed with crimp-sealed butyl rubber stoppers. Amendments were made from stock solutions. Duplicate samples were incubated with the following amendments: without arsenate (NAR1 and 2), with 1 mM arsenate (ASR), with 75 mM tungsate (TUN), and with 1 mM arsenate plus 75 mM tungstate (ASTUN). To more closely mimic eld conditions, samples were incubated in the dark at »6± C rather than at 20± C. Subsamples for As(V) analyses were periodically removed from the liquid phase by syringe to generate progress curves. After »3 weeks incubation, asks were opened, the water ltered, and 1.8 ml of lysis buffer was added to the lter (Ferrari and Hollibaugh 1999). Filters were stored at ¡60± C and shipped overnight to the University of Georgia for DGGE analysis (see next). DNA Extraction, Ampli cation, and Denaturing Gel Gradient Electrophoresis (DGGE ) Analysis Nucleic acids were extracted from lter cartridges according to Ferrari and Hollibaugh (1999) followed by centrifugation through Centricon 100 (Amicon) spin lters. The variable region V3 of the16S rRNA gene was ampli ed with primer sequences bracketing positions 341 to 358f (primer 358, Eubacterial) and positions 517 to 534 (primer 517r, universal) of Escherichia coli gene (Brosius et al. 1978). A 40-bp GC clamp (Myers et al. 1985) was added to the 50 -end of the 358f primer. Fluorescein label was coupled to the 50 end of the primer 517r. DNA, cDNA samples, and RT-controls were ampli ed under PCR conditions given by Ferrari and Hollibaugh (1999). DGGE was performed according to Bano and Hollibaugh (2000) using a CBS Scienti c DGGE system. After electrophoresis, band positions were determined by measuring dye uorescence using a Hitachi FMBIO II gel scanner equipped with a 505-nm lter. The gel image was then processed using image analysis software (Molecular Analyst version 1.12, Bio-Rad). Similarities of band patterns were calculated (Jacard Index; Jaccard 1908) and compared by cluster analysis (UPMGA: unweighted pair group with mathematical averages). Bands of interest and containing suf cient quantities of DNA were excised from the gels and frozen at ¡20± C

26

S. E. Hoeft et al.


for 1 h. DNA was eluted in 60 ¹l of MilliQ water at 65± C for 1 h, then puri ed using Qiaquick PCR puri cation kits (Qiagen). All sequences were obtained with 358f primer using an automated sequencer (Applied Biosystem 377) at the University of Georgia Molecular Genetics core sequencing facility. Nucleotide sequences were used to search for homologous sequences in GenBank using the BLAST algorithm blastn and the databases nr. Sequences were aligned and compared using the Genetics Computer Group Inc. package (Madison, Wisconsin). Phylogenetic trees were inferred from Jukes-Cantor distances using the neighbor-joining method (PHYLIP 3.572; Felsenstein et al. 1988). The branching pattern was checked by 100 bootstrap replicates. To ensure reliable phylogenetic positioning, a complete sequence (at least 400 bp) is desirable; however, it is possible to use partial sequences to identify organisms or to assign them to well-established phylogenetic groups, as long as the database contains sequences of close relatives (Ludwig et al. 1998). Analytical. Arsenate was analyzed by ion chromatography after a 50-fold dilution into MilliQ water to eliminate interference with HS¡ peaks. We employed a Dionex DX-300 system equipped with a model 1, pulsed electrochemical detector set in the conductivity mode. The column was a Dionex Ionpac AS4-SC (4 mm) anion exchange column in series with an AG4A-SC guard column and an ASRS ultra (4 mm) anion self-regenerating suppressor. The detection limit in lakewater samples was 25 ¹M, which accounts for the effect of sample dilution. Arsenite, although separated, could not be detected by this method because of interferences from the lakewater matrix; however, previous studies have demonstrated quantitative recovery of As(III) from dissimilatory reduction of arsenate in Mono Lake water samples (Oremland et al. 2000). Acetate was determined by high performance liquid chromatography (Culbertson et al. 1988; Oremland et al. 1994). Direct counts of bacterial cell densities in the water samples were achieved with acridine orange (Hobbie et al. 1977).

Results Effect of Inhibitors on As(V) Reduction DAsR occurred steadily in water samples amended with 1 mM As(V) (rate D »7 ¹mol l¡1 h¡1 ), with reduction being essentially complete by 120 h, yet no reduction occurred in lter-sterilized controls (Figure 1). In samples containing 25 mM tungstate or molybdate, the rate of DAsR was slightly slower (»25%) compared with the uninhibited samples. Similar inhibitory effects were achieved at lower (5 mM) concentrations of these two Group VIA oxyanions (data not shown). Selenate provided a more noticeable disruption of DAsR, initially causing a small increase in As(V), then a delayed inhibitory response before As(V) loss was apparent (Figure 2). In contrast, nitrate appeared to completely inhibit DAsR, being essentially as effective as lter-sterilization (Figure 3). Samples amended with As(III) and nitrate demonstrated clear production of As(V), and activity was much greater in live versus lter sterilized controls (Figure 4). The added 5 mM nitrate was entirely consumed in the live samples and was balanced by the accumulation of 5 mM nitrite, which was not metabolized further (data not shown). No nitrate loss or nitrite accumulation was noted in lter sterilized controls. In addition, live and ltered controls with As(III) but without nitrate produced only a small amount of As(V). In another experiment (not shown), oxidation of As(III) occurred in live samples amended with 5 mmol/L Fe(III)NTA, nitrate, or nitrite, which after 355 h of incubation respectively produced the following levels of As(V): 325 § 54 ¹M (average of 2; § 12 range), 1136 § 48 ¹M (n D 3; § std. dev.), and 775 § 68 ¹M (n D 3; § std. dev.). No signi cant levels of As(V) were noted in unamended controls (75 § 59 ¹M) or in a lter sterilized sample incubated with Fe(III) (80 ¹M).



27

FIGURE 1 Reduction of 1 mM added As(V) in anoxic Mono Lake water samples incubated at 20± C. Symbols: , no additions; ¤, with 25 mM molybdate; ¥, with 25 mM tungstate; ", lter sterilized controls. Symbols represent the mean of 3 samples and bars indicate §1 std. dev. In addition, no signi cant biological oxidation was noted in samples amended with Mn(IV) or Se(VI). Effect of Electron Donors on As(V ) Reduction Addition of lactate to arsenate-amended water samples caused a slight stimulation of DAsR, which essentially shifted the most rapid reduction rates forward by 24 h compared with controls without lactate (Figure 5). In contrast, no stimulation of DAsR was evident

FIGURE 2 Reduction of 1 mM added As(V) in anoxic Mono Lake water samples incubated at 20± C. Symbols: , no additions; 1, with 5 mM selenate; ", lter sterilized controls. Symbols represent the mean of 3 samples and bars indicate §1 std. dev.


28

S. E. Hoeft et al.

FIGURE 3 Inhibition of the reduction of 1 mM As(V) by nitrate in anoxic Mono Lake samples incubated at 20± C. Symbols: , no additions; N, with 5 mM nitrate; ", lter sterilized controls. Symbols represent the mean of 3 samples and bars indicate §1 std. dev. with acetate amendment. Lactate was quantitatively oxidized to acetate in arsenate-amended samples, as it was in samples without As(V), although acetate production lagged in the latter case (Figure 6). No acetate production was evident in samples lacking lactate amendment. Three other electron donors were also examined to determine if they could accelerate the endogenous rate of DAsR (Figure 7). As was the case for lactate, glucose, succinate, and malate did not stimulate the initial rate of DAsR, although a slight stimulation was notable at »135 h, near the end of the incubation for each of these electron donors. Bacterial growth was noted in all incubated water samples. Initial cell counts were 2.3 § 0.7 £ 106

FIGURE 4 Oxidation of 1 mM As(III) in anoxic Mono Lake water incubated at 20± C. Symbols: ¥, no additions; ", with 5 mM nitrate; 1, lter sterilized with 5 mM nitrate; , lter sterilized. Symbols represent the mean of 3 samples and bars indicate §1 std. dev.



29

FIGURE 5 Effect of electron donors on the reduction of 2 mM As(V) anoxic in Mono Lake waters incubated at 20± C. Symbols: , no additions; ¯, with 1 mM acetate; , with 1 mM lactate; ", lter sterilized control. Symbols represent the mean of 3 samples and bars indicate §1 std. dev. cell ml¡1 (n D 3; § std. dev.). After 212 h of incubation, cell counts increased with electron donor/acceptor amendments as follows (£106 cells ml¡1 ; § std. dev.): no additions (3.8 § 1.2); As(V) (19.3 § 7.2); lactate (9.8 § 1.5); As(V) C lactate (29.8 § 6.6); As(V) C glucose (19.5 § 4.2); As(V) C succinate (22.1 § 2.1); As(V) C malate (17.6 § 6.2). DGGE Analysis of Incubated Water Samples Arsenate was reduced at roughly linear rates in samples containing As(V) or As(V) plus 75 mM tungstate additions (data not shown). The rate of DAsR at the 6± C incubation

FIGURE 6 Formation of acetate from added 1 mM lactate in anoxic Mono Lake water samples incubated at 20± C. Symbols: ¥, with 2 mM As(V); ¨, no additions; ", lter sterilized controls. Symbols represent the mean of 3 samples and bars indicate §1 std. dev.


30

S. E. Hoeft et al.

FIGURE 7 Effect of electron donors on the reduction of 2 mM As(V) in anoxic Mono Lake waters samples incubated at 20± C. Symbols: , no additions; r, with 1 mM glucose; H, with 2 mM succinate; , with 2 mM malate; ", lter sterilized control. Symbols represent the mean of 3 samples and bars indicate §1 std. dev.

¦

temperature of this experiment (1.2 ¹mol l¡1 h¡1 ) was about 6-fold slower than that observed at 20± C, and was even less with 75 mM tungstate (»0.5 ¹mol l¡1 h¡1 ). After 32 days incubation at this temperature, not all the initial 1.1 mM As(V) was consumed, although the average residual As(V) concentration in the uninhibited samples (0.28 mM) was lower than in tungstate-inhibited samples (0.67 mM). Clear differences between the bacterial assemblages in these treatments were evident in the DGGE gels (Figure 8a). Analysis of the DGGE banding patterns (Figure 8b) indicates that replicate treatments responded similarly. Replicates were always more similar to each other (74 – 99% similarity) than to other treatments. Between-treatment similarities ranged from 35 to 70% with treatments incorporating tungstate amendments clustering together and being least similar to controls. We obtained sequences for the DNA fragments in the 9 DGGE bands indicated on Figure 8b. Similarities of these sequences to others in Genbank are given in Table 2. Figure 9 displays the phylogenetic relationship between sequences from ASR, TUN, ASTUN treatments, and 16S rDNA sequences from the database. Bands 1, 3, and 6, obtained from TUN, ASTUN, and ASR amendments, aligned with Picocystis salinarium, a photosynthetic eukaryotic picoplankter that occurs throughout the water column of Mono Lake at cell abundances of »105 cells ml¡1 (Roesler et al. 2002). A dense band (#4) was evident in all samples amended with tungstate, including ASTUN, which had 90.2% sequence similarity with Propionigenium modestum. Of major interest was ASR2band 9 from the arsenate-enriched samples that yielded a sequence with 89.2% similarity to Desulfovibrio. ASR2band 7 had 100% similarity to Thiomicrospiria sp.

Discussion Mono Lake represents an environmental extreme because its waters are highly alkaline and hypersaline, and clearly it is not typical of aquatic systems in which the mobility of arsenic is a serious health concern, such as within drinking water aquifers of Bangladesh (Nickson et al. 2000) or western Nevada (Welch et al. 2000). Nonetheless, this lake offers an excellent system in which to study aspects of the biogeochemistry of arsenic in an

31



FIGURE 8 A. Image of a denaturing gradient gel containing rDNA fragments ampli ed using template DNA extracted from Mono Lake water samples incubated at 6± C for 3 weeks with: 75 mM tungstate (lanes 1 and 2); 75 mM tungstate plus 1 mM As(V) (lanes 3 and 4); 1 mM As(V) (lanes 5 and 6); and no additions (lanes 7 and 8). Lanes marked “std” contain fragments ampli ed from Clostridium perfringens and Bacillus thuringeniesis genomic DNA. B. Processed image of the gel in A with the same lane designations. The distributions of bands in lanes were compared using the Jacard Index, the resulting similarity matrix was used to cluster (UPGMA) the samples, the dendrogram to the left indicates the similarity of different samples. Numbers on the image indicate bands that were excised for sequencing. aquatic milieu. This is due, in part, to the lake’s extraordinarily high content of inorganic arsenic, a factor that magni es the importance of processes such as DAsR, especially with regard to its role in the carbon cycle (Oremland et al. 2000). Furthermore, because these reductive transformations of arsenic occur within the anoxic waters of a strati ed water

Downloaded by [Ronald Oremland] at 05:59 23 January 2013 32 S. E. Hoeft et al.



33

column undergoing prolonged meromixis (Jellison et al. 1998), these experiments can be conducted using aqueous samples, thereby circumventing the more problematic conditions encountered in experiments with sediments. Addition of 5 mM As(V) to Mono Lake water samples previously resulted in a strong (71%) inhibition of 35 S-sulfate reduction (Oremland et al. 2000). It is likely that at this concentration, As(V) acted as a preferred electron acceptor to sulfate by the sulfate-respiring ora (Newman et al. 1997). This interpretation is supported by the much higher energy yields associated with use of As(V) as an oxidant as opposed to sulfate (Newman et al. 1998). Therefore, as a practical compromise, we chose a concentration of 1 mM As(V) to be added to our Mono Lake water samples. We believed this concentration would allow activity of sulfate-reducing bacteria as well as modest growth of any other As(V)-respiring prokaryotes. It is important to note that this concentration represents As(V) levels that are only 5-fold greater than the highest levels present in the oxic surface waters (Oremland et al. 2000), and therefore are sometimes present in the bottom waters after lakewide mixing events. Measurable rates of sulfate reduction (1 – 2 ¹mol l¡1 d¡1 ) as well as culturable sulfaterespiring bacteria occur in the monimolimnion of Mono Lake (Oremland et al. 2000). These facts, coupled with the proven ability of sulfate-reducers from diverse phylogenies of the Eubacteria to achieve growth via DAsR (Newman et al. 1997; Macy et al. 2000), suggest that microorganisms of this type should be important biogeochemical agents of DAsR in these sulfate-rich anoxic waters. We have attempted to disrupt the activity of sulfatereducing bacteria by adding either molybdate or tungstate as speci c inhibitors of this process (Oremland and Capone 1988). Previous work indicated that 5 mM tungstate caused a 26% inhibition of 35 S-sulfate-reduction in monimolimnion water (Oremland et al. 2000), while 50 mM tungstate resulted in an 81% inhibition of this activity in bottom sediments

FIGURE 9 Phylogenetic relationships between 16S rDNA sequences from DGGE bands and sequences from GenBank. Name and accession numbers of database sequences used in the tree are: Sulfurospirillum arsenophilus MIT-13 (U85964), Sulfurospirillum barnesii SES-3 (U41564), Thiomicrospora denitri cans DSM1251 (L40808), Sulfurospirillum arcahonense DSM29755 (Y11561), uncultured epsilon proteobacterium KNIB01 (AF249343 ), uncultured bacterium VC2.1 Bac32 (AF068806), Thiomicrospira sp. CVO (U46506), hydrothermal vent eubacterium PVB OTU 8 (U15107), uncultured bacterium ODPB-B3 (AF121088 ), Klebsiella planticola ATCC33531T (AF129443 ), Klebsiella ornithinolytica ATCC31898 (AF129441 ), Klebsiella terrigena ATCC33257T (AF129442), Klebsiella treviscanii ATCC33558T (AF129444 ), Enterobacter aerogenes NCTC10006T (EAE251468), Erwinia carotovora subsp. Carotova ATCC15713 (ECU80197), Erwinia alni DSM11811 (EAL233409 ), Escherichia coli ATCC43895 (Z83205), Propionigenium modestum (X54275), Propionigenium maris ML-1 (Y16800), Fusobacterium mortiferum ATCC25557 (M58680), Clostridium rectum NCIMB10651 (X77850 ), Fusobacterium gonidoformans ATCC25563 (M58679), Fusobacterium sp. M-3333 (X78419), Fusobacterium perfoetens ATCC29250 (M58684), Desulfovibrio termitidis KSS1 (Y12255), Desulfovibrio longrachii isolate 16910a (Z24450), uncultured bacterium ODPB-U9 (AF121083 ), Desulfovibrio desulfuricans subsp. desulfuricans MB ATCC27774 (AF192154), Prochlorococcus marinus subsp. pastoris PCC9511 (AF180967 ), Prochorococcus sp. NATL1 (AF133834), Prochlorococcus sp. CCMP1426 (AF133833), uncultured marine bacterium ENATL7 (AF099997 ), Synechococcus sp. PCC9005 (AF216950 ), Synechococcus leopoliensis PCC6301 (Z82780), Picocystis salinarium SSFB (AF125173 ), Flavobacterium aquatile ATCC11947 (M62797).

34

S. E. Hoeft et al.


TABLE 1 Name of the band sequences, similarity to most closely related database sequences and number of different base pairs Band

Base pairs

Related species

TUN2band1 ASTUN2band3 ASR2band7

120/120 120/120 120/137

ASR2band8

134/135

ASTUN2band5

121/129

TUN2band2 ASTUN2band4 ASR2band9

145/158 145/158 140/157

Picocystis salinarum SSFB chloroplast Picocystis salinarum SSFB Uncultured bacterium VC2.1 Bac32 Hydr. Vent eubacterium PVB OTU 8 Thiomicrospira sp. CVO Uncultured bacterium ODP-B3 Prochlorococcus marinus Prochlorococcus sp. NATL1 Prochlorococcus sp. CCMP1426 Uncultured marine bacterium ENATL7 Synechococcus sp. PCC9005 Klebsiella planticola ATCC33531T K. ornithinolytica ATCC 31898 K. terrigena ATCC 33257T Erwinia carotovora Erwinia alni DSM 11811 Propionigenium modestum taxon 2333 Propionigenium modestum taxon 2333 Desulfovibrio termiditis KSS1 Desulfovibrio longreachii 16910a

% similarity 100 100 100

99.2

94.4

90.2 90.2 89.2

(R. L. Smith 1986, unpublished data), and thus we expected these concentrations to have similar effects on DAsR if it was carried out by sulfate-reducers. Both 25 mM molybdate and tungstate achieved a partial (» 25%) inhibition of DAsR in water samples incubated at 20± C (Figure 1), yet 75 mM tungstate caused greater (»59%) inhibition in samples incubated at 6± C (see text). Collectively, these results imply that sulfate-respiring bacteria make a substantial contribution, perhaps 25 to 60%, of the observed DAsR activity. The respiratory As(V) reductase of Chrysiogenes arsenatis contains molybdenum (Krafft and Macy 1998). It should therefore be inhibited by tungstate but not by molybdate, while our water samples exhibited similar inhibition by both of these anions. The enhanced signal of a DGGE band containing DNA aligning within the Desulfovibrio clade (ASRband9) in samples amended with 1 mM As(V) provides further evidence for the involvement of sulfate-reducers from the ± subgroup of the Proteobacteria (Figures 8 and 9; Tables 1 and 2). However, because 75 mM molybdate or tungstate caused only a partial inhibition of DAsR, microorganisms other than sulfate-reducers must also contribute to the observed DAsR activity. Another band of interest that was observed with addition of As(V) was ASRband7, which grouped as a member of the "-Proteobacteria (Figure 9) with 100% sequence similarity to Thiomicrospira sp. as well as to three other uncultured bacteria (Table 2). Although members of the "-Proteobacteria include the recognized As(V)-respiring species Sulfurospirillum barnesii and S. arsenophilus, these organisms have an overall metabolism characterized as being facultative anaerobes that use intermediates of the sulfur cycle (e.g., thiosulfate, elemental sulfur) as their terminal electron acceptors. In this regard, they differ from the known species of Thiomicrospira that are characterized by being small (1.5 mM) in the bottom water with As(V) as its terminal electron acceptor in lieu of nitrate. We were, however, surprised that we did not detect any 16S rRNA gene sequences that aligned with the low GCC Gram-positive bacilli, such as Bacillus arsenicoselenatis and B. selenitireducens, the two As(V)-respiring haloalkaliphiles isolated previously from Mono Lake sediments (Switzer Blum et al. 1998). Gram-positive bacilli are commonly found in enrichments from a diversity of alkaline soda lakes (Duckworth et al. 1996). The fact that we detected a representative of the Gram-positive Fusobacteria (Figure 9) in samples incubated with tungstate (TUN2band2; Tables 1 and 2), and that DNA was readily extractable from cell pellets of B. arsenicoselenatis and B. selenitireducens (J. T. Hollibaugh, unpublished data) further indicates that our extraction techniques were not biased by the different cell wall types. Other Gram-positive organisms, including bacilli, are abundant in Mono Lake (J. T. Hollibaugh, unpublished data). This suggests that these two organisms were not present in signi cant numbers in the incubated waters, although they could be abundant in the bottom muds that were the source of their initial enrichment (Switzer Blum et al. 1998). We did not, however, conduct DNA ampli cation of any Archaeal genes, and therefore our work does not extend to probing this domain for DAsR candidates. One curious response was the repeated occurrence of a dense band (TUN2band2 and ASTUN2band2 ) in all the samples containing tungstate, which included both those amended with As(V) and those without As(V) (Figure 9). One possible interpretation of this result is that the disruption of sulfate-reduction by tungstate created a niche for Gram-positive representatives of the Fusobacteria to carry out DAsR. Alternatively, an enrichment of a population of organisms capable of dissimilatory reduction of 75 mM tungstate itself may have been achieved. Other speculative interpretations are also possible, but are beyond the scope of this discussion. Although these bands aligned with various members of the Fusobacteria, none of these fermentative anaerobes has been characterized as having a capacity for anaerobic respiration, let alone being able to respire arsenate or tungstate. Similar reasoning can be extended to ASTUN2band5 that aligned with the Klebsiella. However, it should be noted that the expressed phenotypes used to characterize bacterial orders and genera held in culture collections may not typify all members of these taxonomic assemblages. This is especially likely in the case of nonculturable organisms detected only by ampli cation of DNA extracted from environmental samples. In some cases, new isolates that align within established taxonomic groups characterized by a fermentative metabolism exhibit anaerobic respiration previously not described for that division. For example, Selenihalanaerobacter shriftii is a halophilic anaerobe that respires selenate, trimethylamine-N-oxide, or nitrate, but does not grow by fermentation typi ed by members of the order Halanaerobiales (Switzer Blum et al. 2001). The reduction of As(V) by the microorganisms present in these water samples did not appear to be severely constrained by the availability of electron donors required to fuel DAsR. Although amendment with lactate provided some modest stimulation over the endogenous rate, this was only evident after 100 h incubation (Figure 5), and in the case of glucose, succinate, and malate the response was even less pronounced (Figure 7). The only electron donor that clearly did not in uence the rate of DAsR was acetate (Figure 5). The lactate added in these experiments was quantitatively oxidized to acetate plus CO2 , and this oxidation was speeded by As(V) (Figure 6). However, this also occurred in the absence of any added As(V), and strongly suggests involvement of sulfate-reducing bacteria. Sulfate was the only potential electron acceptor present in suf cient abundance in these waters (»130 mM) that could allow for the oxidation of 1 mM lactate. These observations are reinforced by the data on cell growth (see Results) that indicated only a 65% increase in yield in the absence of any amendments, which contrasted with the


38

S. E. Hoeft et al.

426% and 839% increases with additions of lactate and arsenate, respectively. The source of the naturally occurring electron donor is not clear. Mono Lake contains very high levels (»90 mg L¡1 ) of dissolved organic carbon (Oremland et al. 1987), although this material usually consists of refractory organic compounds. Another, more labile source could be derived from lysis of the abundantly occurring Picocystis sp. phototrophic eurkaryote as well as prokaryotic phototrophs (e.g., ASR2band8) that sink into the bottom waters (Figure 9; Tables 1 and 2) and are present at high concentrations therein (e.g., Picocystsis sp. D »105 cells ml¡1 ). The biological oxidation of As(III) to As(V) with nitrate as the electron acceptor was an unexpected nding (Figure 4). A similar phenomenon may have occurred in the early part of the selenate-amendment experiments (Figure 2), possibly caused by selenate-linked oxidation of the ambient arsenite (0.2 mM) present in the bottom water. Oxidation of As(III) by aerobic bacteria is a well-known phenomenon (Ehrlich 1990), and its manifestations are clear along hydrological ow paths emanating from reducing hot springs (Wilkie and Hering 1998). Some aerobic arsenite oxidizers are chemoautotrophs (Santini et al. 2000), but there are no reports of As(III) oxidation by anaerobes. Chemical oxidation of As(III) can be ef ciently achieved with Mn(IV) (Oscarson et al. 1981; Chiu and Hering 2000), but is barely perceptible with Fe(III) and has not been shown for nitrate. It would be of interest to isolate a chemoautotroph that obtains its energy via oxidation of As(III) with nitrate as its electron acceptor and we have initiated enrichment cultures with this goal in mind. However, from a geochemical standpoint, nitrate is unimportant in Mono Lake, essentially being present only at trace concentrations (·1 ¹M) throughout the water column (Maest et al. 1992). Fe(III) has a 10-fold greater abundance in Mono Lake than does nitrate (Maest et al. 1992) and we also observed biological As(III) oxidation linked to Fe(III). The bottom waters of Mono Lake contain low (5 – 10 ¹M), but signi cant concentrations of As(V) that are not in equilibrium with the highly reducing nature of the surrounding chemical milieu (Oremland et al. 2000). Advective mixing of these strati ed waters is insuf cient to sustain these observed levels of dissolved As(V), so in situ regeneration of As(V) must occur. It is possible that some bacteria in the anoxic water column employ sinking particles of ferrihydrite as an electron acceptor, thereby achieving a re-oxidation of As(III).

References Ahmann D, Roberts AL, Krumholz LR, Morel FMM. 1994. Microbe grows by reducing arsenic. Nature 371:750. Bano N, Hollibaugh JT. 2000. Diversity and distribution of DNA sequences with af nity to ammonia-oxidizing bacteria of the ¯ subdivision of the class Proteobacteria in the Arctic Ocean. Appl Environ Microbiol 66:1960 – 1969. Brosius J, Palmer L, Poindexter JK, Noller HF. 1978. Complete nucleotide sequence of a 16S ribosomal RNA gene from Escherichia coli. Proc Natl Acad Sci USA 75:4801 – 4805. Cervantes C, Ji G, Ramirez JL, Silver S. 1994. Resistance to arsenic compounds in microorganisms. FEMS Microbiol Rev 15:355 – 367. Chiu VQ, Hering JG. 2000. Arsenic adsorption and oxidation at magnanite surfaces. 1. Method for simultaneous determination of adsorbed and dissolved arsenic species. Environ Sci Technol 34:2029 – 2034. Culbertson CW, Strohmaier FE, Oremland RS. 1988. Acetylene as a substrate in the development of primordial bacterial communities. Orig Life Evol Biosphere 18:397 – 407. Duckworth AW, Grant WD, Jones BE, van Steenbergen R. 1996. Phylogenetic diversity of soda lake alkaliphiles. FEMS Microbiol Ecol 19:181 – 191. Ehrlich HL. 1990. Geomicrobiology. 2nd edition. New York: Marcel Dekker. p 252 – 256.



39

Felsenstein J. 1988. Phylogenies from molecular sequences: interference and reliability. Annu Rev Genetics 22:521 – 526. Ferrari VC, Hollibaugh JT. 1999. Distribution of microbial assemblages in the Central Arctic Ocean basin studied by PCR/DGGE: analysis of a large data set. Hydrobiologia 401: 55– 68. Hobbie JE, Daley RL, Jaspar S. 1977. Use of Nuclepore lters for counting bacteria for uorescence microscopy. Appl Environ Microbiol 33:1225 – 1228. Huber R, Sacher M, Vollmann A, Huber H, Rose D. 2000. Respiration of arsenate and selenate by hyperthermophilic archaea. Syst Appl Microbiol 23:305– 314. Jaccard P. 1908. Nouvelles researches sur la distribution orale. Bull Soc Vaud Sci Nat 44:223 – 270. Jellison R, Romero J, Melack JM. 1998. The onset of meromixis during restoration of Mono Lake, California: Unintended consequences of reducing water diversions. Limnol Oceanogr 43:706 – 711. Krafft T, Macy JM. 1998. Puri cation and characterization of the respiratory arsenate reductase of Chrysiogenes arsenatis. Eur J Biochem 255:647 – 653. Kuenen JG, Touvinen OH. 1981. The genera Thiobacillus and Thiomicrospora. In: Balows A, Schlegel HG, editors. The prokaryotes, a handbook of habitats, isolation and identi cation of bacteria. Berlin, Germany: Springer-Verlag. p 1023 – 1036. Lucas FS, Hollibaugh JT. 2001. Response of sediment bacterial assemblages to selenate and acetate amendments. Environ Sci Technol 35:528 – 534. Ludwig W, Strunk O, Klugbauer S, Weizenegger M, Neumaler J, Bachleitner M, Schleifer KH.1998. Bacterial phylogeny based on comparative sequence analysis. Electrophoresis 19:554 – 568. Macy JM, Santini JM, Pauling BV, O’Neill AH, Sly LI. 2000. Two new arsenate/sulfate-reducing bacteria: mechanisms of arsenate reduction. Arch Microbiol 173:49 – 57. Maest AS, Pasilis SP, Miller LG, Nordstrom DK. 1992. Redox geochemistry of arsenic and iron in Mono Lake, California. In: Kharaka YK, Maest AS, editors. Water-rock interaction. Rotterdam, Netherlands: AA Balkema. p 507 – 511. Miller LG, Ellison R, Oremland RS, Culbertson CW. 1993. Meromixis in hypersaline Mono Lake, California. 3. Biogeochemical response to strati cation and overturn. Limnol Oceanogr 38:1040 – 1051. Myers RM, Fischer SG, Lerman LS, Maniatis T. 1985. Nearly all single base substitutions in DNA fragments joined to a GC-clamp can be detected by denaturing gradient gel electrophoresis. Nuc Acids Res 13:3131– 3145. Newman DK, Ahmann AM, Morel FMM. 1998. A brief review of microbial arsenate respiration. Geomicrobiol J 15:255 – 268. Newman DK, Kennedy EK, Coates JD, Ahmann D, Ellis DJ, Lovley DR, Morel FMM. 1997. Dissimilatory arsenate and sulfate reduction in Desulfotomaculum auripigmentum sp. nov. Arch Microbiol 168:380 – 388. Nickson RT, McArthur JM, Ravenscroft P, Burgess WG, Ahmed KM. 2000. Mechanism of arsenic release to groundwater, Bangladesh and West Bengal. Appl Geochem 15:403 – 413. Oremland RS, Capone DG. 1988. Use of “speci c” inhibitors in biogeochemistry and microbial ecology. Adv Microbial Ecol 10:285 – 383. Oremland RS, Dowdle PR, Hoeft S, Sharp JO, Schaefer JK, Miller LG, Switzer Blum J, Smith RL, Bloom NS, Wallschlaeger D. 2000. Bacterial dissimilatory reduction of arsenate and sulfate in meromictic Mono Lake, California. Geochim Cosmochim Acta 64:3073 – 3084. Oremland RS, Miller LG, Whiticar MJ. 1987. Sources and ux of natural gases from Mono Lake, California. Geochim Cosmochim Acta 51:2915 – 2929. Oremland RS, Newman DK, Kail BW, Stolz JF. 2002. Bacterial respiration of arsenate and its signi cance in nature. In: Frankenberger WT Jr, editor. Environmental chemistry of arsenic. New York: Marcel Dekker. In press. Oremland RS, Switzer Blum J, Culbertson CW, Visscher PT, Miller LG, Dowdle P, Strohmaier FE. 1994. Isolation, growth and metabolism of an obligately anaerobic, selenate-respiring bacterium, strain SES-3. Appl Environ Microbiol 60:3011 – 3019.


40

S. E. Hoeft et al.

Oscarson DW, Huang PM, Defosse C, Herbillon A. 1981. Oxidative power of Mn(IV) and Fe(III) oxides with respect to As(III) in terretrial and aquatic environments. Nature 291:50 – 51. Roesler CS, Culbertson CW, Etheridge SM, Goericke R, Kiene RP, Oremland RS. 2002. Distribution, production, and ecophysiology of Picocystis strain ML in Mono Lake, CA. Limnology & Oceanography. In press. Santini JM, Sly LI, Schnagl RD, Macy JM. 2000. A new chemoautotrophic arsenite-oxidizing bacterium isolated from a gold mine: phylogenetic, physiological, and preliminary biochemical studies. Appl Environ Microbiol 66:92– 97. Stolz JF, Ellis DJ, Switzer Blum J, Ahmann D, Lovley DR, Oremland RS. 1999. Sulfurospirillum barnesii sp. nov., Sulfurospirillum arsenophilum sp. nov., and the Sulfurospirillum clade in the Epsilon Proteobacteria. Int J Syst Bacteriol 49:1177 – 1180. Stolz JF, Oremland RS. 1999. Bacterial respiration of arsenic and selenium. FEMS Microbiol Rev 23:615 – 627. Switzer Blum J, Burns Bindi A, Buzelli J, Stolz JF, Oremland RS. 1998. Bacillus arsenicoselenatis sp. nov., and Bacillus selenitireducens sp. nov.: two haloalkaliphiles from Mono Lake, California that respire oxyanions of selenium and arsenic. Arch Microbiol 171:19 – 30. Switzer Blum J, Stolz JF, Oren A, Oremland RS. 2001. Selenihalanaerobacter shriftii gen. nov., sp. nov., a halophilic anaerobe from Dead Sea sediments that respires selenate. Arch Microbiol 175:208 – 219. Welch AH, Westjohn DB, Helsel DR, Wanty DR. 2000. Arsenic in ground water of the United States: occurrence and geochemistry. Groundwater 38:589 – 604. Wilkie JA, Hering JG. 1998. Rapid oxidation of geothermal arsenic (III) in streamwaters of the eastern Sierra Nevada. Environ Sci Technol 32:657– 662.