Microbial Diversity in an in situ Reactor System Treating onochlorobenzene Contaminated Groundwater as Revealed by 16S Ribosomal DNA Analysis

System. Appl. Microbiol. 25, 232–240 (2002) © Urban & Fischer Verlag http://www.urbanfischer.de/journals/sam

Microbial Diversity in an in situ Reactor System Treating Monochlorobenzene Contaminated Groundwater as Revealed by 16S Ribosomal DNA Analysis ALBIN ALFREIDER, CARSTEN VOGT, and WOLFGANG BABEL UFZ-Umweltforschungszentrum, Sektion Umweltmikrobiologie, Leipzig, Germany Received: March 18, 2002

Summary A molecular approach based on the construction of 16S ribosomal DNA clone libraries was used to investigate the microbial diversity of an underground in situ reactor system filled with the original aquifer sediments. After chemical steady state was reached in the monochlorobenzene concentration between the original inflowing groundwater and the reactor outflow, samples from different reactor locations and from inflowing and outflowing groundwater were taken for DNA extraction. Small-subunit rRNA genes were PCR-amplified with primers specific for Bacteria, subsequently cloned and screened for variation by restriction fragment length polymorphism (RFLP). A total of 87 bacterial 16S rDNA genes were sequenced and subjected to phylogenetic analysis. The original groundwater was found to be dominated by a bacterial consortium affiliated with various members of the class of Proteobacteria, by phylotypes not affiliated with currently recognized bacterial phyla, and also by sporulating and non-sporulating sulfatereducing bacteria. The most occurring clone types obtained from the sediment samples of the reactor were related to the β-Proteobacteria, dominated by sequences almost identical to the widespread bacterium Alcaligenes faecalis, to low G+C gram-positive bacteria and to Acidithiobacillus ferrooxidans (formerly Thiobacillus ferrooxidans) within the γ subclass of Proteobacteria in the upper reactor sector. Although bacterial phylotypes originating from the groundwater outflow of the reactors also grouped within different subdivisions of Proteobacteria and low G+C gram-positive bacteria, most of the 16S rDNA sequences were not associated with the sequence types observed in the reactor samples. Our results suggest that the different environments were inhabited by distinct microbial communities in respect to their taxonomic diversity, particular pronounced between sediment attached microbial communities from the reactor samples and free-living bacteria from the groundwater in- and outflow. Key words: 16S rRNA – Microbial diversity – Polluted groundwater – Chlorobenzene – In situ reactor

Introduction As a result of industrial activity in the 20th century, the aquifer in the Bitterfeld region (Saxony-Anhalt, Germany) is contaminated by large amounts of chlorinated aliphatic and aromatic hydrocarbons. The SAFIRA (Sanierungs-Forschung In Regional kontaminierten Aquiferen) on-site underground reactor facility was set up to develop and implement in situ techniques for the remediation of the polluted groundwater and to evaluate the implementation of in situ reactive barriers in regionally contaminated aquifers (WEISS et al., 1998; MERKEL et al., 2000). Bioremediation processes using the degradative potential of the autochthonous microbial community of the contaminated aquifer are one key component within the in situ technologies applied. For this purpose, two 0723-2020/02/25/02-232 $ 15.00/0

reactors, both filled with original aquifer material, were designed for the investigation and optimization of microbiological in situ remediation processes of the contaminated groundwater by the indigenous microbial consortia. In this context, studying the microorganisms colonizing the reactor and, in particular, assessing the microbial community composition are of fundamental importance for the design and evaluation of a suitable bioremediation strategy. However, investigation of the microbial communities from environmental samples taking part in in situ biodegradation processes has proved a difficult challenge for microbiologists. This is because the traditional approach based on cultivation of bacteria from environmental samples usually only include a minor part of the

Microbial diversity in an in situ reactor system

total bacterial diversity or often produce biased results. (TORSVIK et al., 1990; WAGNER et al., 1992). DNA and RNA-based approaches have helped to overcome many previous limitations and allow the accurate identification of specific microorganisms. In particular, developments based on the use of the polymerase chain reaction (PCR) and 16S rRNA sequence analysis have inspired numerous recent studies in environmental microbiology. Several molecular tools exist nowadays to characterize microbial communities from various habitats (THERON and CLOETE, 2000) and within recent years an increasing number of studies have also applied these methods to investigate the microbial community structure in polluted subsurface environments (DOJKA et al., 2000; RÖLING et al., 2001; WATANABE, 2001). The aim of this study was to investigate the microbial colonization of two analogous in situ reactors filled with original aquifer sediments after chemical steady state had been reached in the monochlorobenzene content between the inflowing (original) and outflowing groundwater, and to compare the bacteria detected with the microbial community structure analyzed in the original groundwater, which is characterized by a long history of contamination with chlorinated aromates, and the reactor outflow. To avoid sampling-induced artifacts which are common in aquifer samples, i.e. variations in the sediment structure and particle size and the occurrence of biological hot spots, 3 subsamples at different reactor zones were investigated. In order to determine the microbial diversity, PCR products of 16S rDNA were used to construct a clone library that was analyzed by grouping clone inserts by RFLP and sequencing dominant clones.

Materials and Methods Site description and sample collection The hydrogeological and physicochemical characteristics of the Bitterfeld aquifer and test site are described in detail elsewhere (MERKEL et al., 2000). The reactors of shaft 5 (A and B) are 12 m tall and each have a diameter of 600 mm (Fig. 1). They are made of stainless steel and native aquifer sediment is used as reactor filling material. Both reactors are operated in flowthrough mode from bottom to top, with a flow rate of 4.7 l h–1, resulting in a retention time of approximately 10 days for water, according to a conductivity tracer test. The reactors were put into operation in June 1999. The physicochemical characteristics of the inflowing ground-water between June 1999 and March 2000 (until operating day 285) are listed in Table 1. Samples were taken on operating day 244, after chemical steady state for chlorobenzene had been reached. Sediment material from three different reactor zones (2.5 m = R7, 5.5 m = R9, 11.5 m = R13 above reactor bottom, see Fig. 1) were sampled with a sample lance, immediately transferred in an anaerobic jar (Anaerocult® A, Merck, Germany), transported at 4 °C and stored at –20 °C before further processing (see below). The inflowing (original) groundwater, which is collected at a depth of 19.5 m by a horizontal well, and the outflowing reactor water were collected in sterile bottles, and 300–500 ml of each water sample was filtered through 0.22 µm pore size filters (Durapore, Millipore, Bedford, MA, USA). After processing, the filters were immediately frozen and stored at –20 °C until extraction.

233

DNA extraction and PCR Sample preparation of reactor sediment samples was performed by ultrasonic separation of cells followed by DNA extraction with FastDNA® Spin Kit for soil (Qbiogene Inc., Carlsbad, CA) according to the instructions supplied by the manufacturer. For the extraction of the water samples, the filters were cut into small pieces using an aseptic technique. The pieces were placed in a capped tube with extraction buffer and beads as provided by the FastDNA® Spin Kit for soil. After processing the samples in a bead beater to lyse the cells, total genomic DNA was extracted according to the protocol. Almost full-length bacterial 16S rDNA fragments were amplified by PCR using universal bacterial primers 27F and 1525R (LANE, 1991). Amplification was carried out in 50 µl reactions containing 25 µl Taq PCR Master Mix (Qiagen GmbH, Valencia, CA, USA), 10 pmol of each primer and 5–50 ng template DNA with a Thermal Cycler PTC200 (MJ Research, Inc Waltham, MA, USA). The cycle parameters were as follows: 5 min at 95 °C; 30 cycles of 45 sec at 95 °C, 45 sec at 55 °C, and 1 min 15 sec at 72 °C. These cycles were followed by 10-minutes’ incubation at 72 °C. Products were run on a 0.8% agarose gel and visualized with ethidium bromide. Bands with the proper size range were cut out of the gel and purified (QIAquick gel extraction kit; Qiagen, Valencia, CA). Cloning and variant screening of PCR amplification products PCR products were TA-cloned into pCR2.1 (Invitrogen Corp., Carlsbad, CA) according to the protocols provided by the manufacturers. Plasmid DNA was isolated (Qiagen plasmid kit, Qiagen Inc., Valencia, CA) and clones were screened for the presence of inserts by PCR using vector-specific primers. The rDNA amplicons were digested with restriction endonucleases HaeIII and/or AluI for 3h. The reaction was stopped by incubating the samples at 65 °C for 20 min. 10 µl of the restriction digests was separated using Metaphor agarose (FMC Corp., Rockland, ME). Between 15 and 25 clones per sampling station were selected and screened for variation. RFLP patterns of different samples were grouped visually, and representative clones were selected for sequencing analysis. DNA sequencing and phylogenetic analysis Double strand sequencing was carried out with an ABI PRISM 310 Genetic analyzer (Applied Biosystems) as previously described by MUELLER et al. (1999). The CHECK_CHIMERA program (MAIDAK et al., 2001) was used to identify potential chimeric sequences among the environmental clones. Since some of the sequences investigated in this study are only distantly related to known ones, phylogenetic trees of partial sequences were constructed to identify the occurrence of chimeric sequences. Closest relatives to 16S rDNA sequences were obtained by using NCBI’s sequence similarity search tool BLASTN 2.1.1. (Basic Local Alignment Search Tool; ALTSCHUL et al., 1990). The phylogenetic analysis of sequences was accomplished with ARB software (STRUNK et al.) The sequences were initially aligned by the ARB automatic aligner and then verified and corrected manually. Maximum-parsimony trees were constructed using the program contained in the ARB software package. A 50% invariance criterion for the inclusion of individual nucleotide sequence positions in the analysis was used to exclude highly variable positions and to avoid possible treeing artifacts. Nucleotide sequence accession numbers The 16S rDNA sequences were deposited at the NCBI databases under the following accession numbers: AY050576–AY050607; AF407193–AF407207; AF407379–AF407417.

234

A. ALFREIDER et al.

Results and Discussion The chemical composition of the inflowing and outflowing groundwater between July 1999 and March 2000 (until operating day 285) is listed in Table 1. Chlorobenzene (CB) was the principal contaminant and the inflowing concentrations varied between 19 and 33 mg l–l. After 210–260 days operating, the situation was considered to have reached chemical steady state, since the CB concentrations in the effluent stopped increasing. Sulfate concentrations did not decrease during the reactor flow path, indicating that in situ sulfate reduction was negligible, although sulfate concentrations were very high (Table 1). Ammonia and phosphate were always detectable in the inflow but while ammonia showed similar concentrations in the outflow (Table 1), phosphate concentrations were significantly lower in the outflow, which may have been due to the uptake of microorganisms. Samples for DNA extraction were taken on operating day 244 (details are given in Materials and Methods). RFLP patterns were used as an initial measure of diversity in the different samples. From each clone library, between 15 and 25 clones were screened by RFLP and numbers of identical patterns are listed in Table 2. Clone patterns that occurred more than once and most representatives with unique RFLP types were selected and sequenced from both sites. The prefix number of the clone designation indicates the sampling station from which the samples were obtained (Fig. 1). 87 sequences were obtained from different sampling locations and at first analyzed with a similarity search program from GenBank, which implements the BLAST algorithm (ALTSCHUL et al., 1990; Table 1).

Fig. 1. Schematic diagram and sample locations (R7; R9; R13) of reactors A and B in shaft 5 filled with aquifer material. Locations of the inflowing and the outflowing groundwater are represented by GIF and GOUT respectively. The clone designation indicates the sampling station. For example, clone RA9C4 was obtained from sampling port 9 of reactor B. (C)4 indicates the consecutive numbering of the clones.

Phylogenetic analysis of the clone sequences Table 1. Chemical and physical characteristics of the inflowing (original) groundwater from horizontal well 5 and the reactor outflow from both reactors of the SAFIRA pilot plant. Data were measured from operation day 60 till 285. Parameter

Inflow

Outflow

Chlorobenzene (mg × l–1) 1,4-Dichlorobenzene (mg × l–1) 1,2-Dichlorobenzene (mg × l–1) TOC (mg × l–1) TOCnv (non-volatile substances, mg × l–1) NO3– (mg × l–1) SO42– (mg × l–1) Cl– (mg × l–1) PO43– (mg × l–1) NH4+ (mg v l–1) H2S (mg × l–1) Dissolved O2 (mg × l–1) pH Temperature (°C)

19–33 ≤ 0.4 ≤ 0.2 20–30 ≤8

0–262 ≤ 0.7 ≤ 0.6 not measured not measured

b.d.1 690–840 390–420 8–11 7–8 b.d. ≤ 0.03 6.6–6.8 13.5–15.8

b.d. 680–870 400–470 0,2–1.7 5–5.4 b.d. b.d. 6.3–6.7 not measured

1

below detection limit The aquifer reactor systems approached chemical steady state for chlorobenzene approximately 210 days after start-up. 2

Although there are no exact sequence similarity limits for the phylogenetic resolution of the 16S rDNA approach and for the definition of specific taxa such as genus and species (LUDWIG et al., 1998), we decided to group sequences with similarities greater than 98% in the phylogenetic tree. Five sequences for which a chimeric structure cannot be ruled out were removed from further phylogenetic analysis (Table 2). From the 6 clone libraries of the reactor material, Proteobacteria with over 80% comprised the most abundant group of bacteria. 16S rDNA sequences within the Proteobacteria include two major groups forming several closely related subclusters belonging to Rhodoferax and Alcaligenes (Table 2; Fig. 2). Especially in reactor A, and to a lesser extent in reactor B, sample stations R7 and R9 were dominated by sequences almost identical to the widespread and physiologically versatile bacterium Alcaligenes faecalis (Table 2). The microdiversity observed within several clone clusters in the β-Proteobacteria seems to be a common occurrence in environmental samples (FIELD et al., 1997; GARCIA-MARTINEZ and RODRIGUEZ-VALERA 2000; SCHULZE et al., 1999) and is argued to be a strategy for adaptation (or “speciation”) to

Microbial diversity in an in situ reactor system

235

Table 2. Summary of the 16S rDNA clone sequences from different samples and nearest relatives from GenBank. Name of clones

No. of clonesa

Size (bp)

Closest relativeb

Similarity (%)

Sediments Reactor A RA7C9 RA7C16 RA9C1 RA9C2 RA9C4 RA9C8 RA13C1 RA13C2 RA13C3 RA13C4 RA13C5 RA13C6 RA13C7 RA13C8 RA13C9 RA13C10 RA13C21 RA13C22

11 5 1 8 1 1 1 1 1 1 1 1 1 1 1 1 2 1

1504 1483 1452 1475 1446 1451 1496 1481 1520 1513 1530 1528 1505 1516 1501 1468 1511 1512

Alcaligenes faecalis Alcaligenes faecalis Clostridium thermosuccinogenes Alcaligenes faecalis Dechlorisoma sp. Uncultured bacterium strain:1028 Uncultured bacterium SHA-61 Unidentified bacterium clone BSV72

99.6 99.5 92.1 99.5 91.4 97.7 93.6 91.7 <90 99.9

Sediments Reactor B RB7C2 RB7C3 RB7C4 RB7C5 RB7C6 RB7C8 RB7C10 RB9C1 RB9C2 RB9C3 RB9C5 RB9C6 RB9C10 RB13C1 RB13C3 RB13C4 RB13C5 RB13C6 RB13C10 RB13C11 RB13C12 RB13C13 RB13C16

1 5 2 1 1 1 1 1 1 1 1 4 1 1 1 1 1 1 1 1 1 1 2

1448 1464 1433 1481 1467 1496 479 1502 1517 1443 1495 1456 1468 1460 1016 1489 1497 1486 1505 1492 1492 1463 1468

Inflowing Groundwater GIF1 GIF2 GIF3 GIF4 GIF5 GIF7 GIF8 GIF9 GIF10 GIF11 GIF12 GIF15 GIF16 GIF17 GIF19

1 2 1 1 1 1 2 1 1 1 1 1 1 1 1

1468 1460 1440 1035 1483 1479 1468 1542 1446 466 1475 1465 1094 1504 1491

Acidithiobacillus sp. NO-37 Chimeric Beta proteobacterium Wuba72 Uncultured bacterium SHA-219 Chimeric Uncultured bacterium GKS16 Thiobacillus ferrooxidans Thiobacillus ferrooxidans Rhodoferax fermentans Alcaligenes faecalis Unidentified bacterium, UP1 Unidentified bacterium, UP1 Unidentified bacterium, TBW3 Unidentified bacterium, clone SJA-62 Uncultured beta proteobacterium SBR1021 Chimeric Uncultured bacterium GR-WP33–36 Uncultured bacterium SHA-61 Rhodoferax fermentans Rhodoferax fermentans Uncultured bacterium SHA-61 Beta proteobacterium Wuba72 Thiobacillus ferrooxidans Uncultured bacterium BPC087 Uncultured proteobacterium 1016 Uncultered proteobacterium 1016 Alcaligenes faecalis Uncultured bacterium SHA-61 Uncultured bacterium GKS16 Beta proteobacterium Wuba72

Unidentified bacterium (strain:rJ1) Desulfosporosinus sp. PFB Unidentified bacterium (strain: rJ1) Pasture soil clone HPS-64 Alcaligenes sp. Desulfovibrio sp. Mlhm Unidentified bacterium (strain: rJ1) Uncultured beta proteobacterium clone 8–11 Desulfotomaculum thermobenzoicum

99.0 <90 97.3 97.0 99.3 99.3 97.6 99.0 98.0 97.4 96.7 97.1 92.9 <90 97.8 94.3 97.5 97.6 93.5 97.0 99.3 96.7 98.2 98.8 99.7 93.6 96.9 99.0 <90 99.8 <90 97.0 99.9 95,4 99.7 <90 <90 99.7 <90 99.9 96.8 93.4 <90

236

A. ALFREIDER et al.

Table 2. (Continued). Name of clones

No. of clonesa

Outflowing Groundwater; Reactor A GOUTA1 1 GOUTA3 1 GOUTA4 1 GOUTA6 1 GOUTA7 1 GOUTA8 1 GOUTA9 1 GOUTA10 1 GOUTA11 1 GOUTA12 1 GOUTA13 1 GOUTA14 1 GOUTA17 1 GOUTA18 1 GOUTA19 1555 Outflowing Groundwater Reactor B GOUTB1 1 GOUTB2 1 GOUTB3 1 GOUTB4 1 GOUTB5 1 GOUTB6 1 GOUTB7 1 GOUTB8 1 GOUTB10 1 GOUTB13 1 GOUTB15 1 GOUTB16 1 GOUTB17 1 GOUTB18 1 GOUTB19 1 GOUTB20 1 a b

Size (bp)

Closest relativeb

Similarity (%)

1482 593 1485 1499 1463 1452 1496 1471 1459 1491 1024 1459 1494 569

Unidentified bacterium

97.8 <90

Uncultured bacterium GR-WP33-36 Chimeric

97.6

902 1491 1491 1501 762 938 1501 1481 1455 1024 1506 1267 1482 1460 1442 1496

Uncultured bacterium Legionella sp. strain LLAP11 Unidentified bacterium (strain rJ1) Uncultured beta proteobacterium, clone 4-2 Chimeric Denitrifying Fe-oxidizing bacterium Ultramicrobacterium strain 12-3

Desulfofaba gelida Legionella lytica Uncultured bacterium clone 36-28 Acidovorax sp. BSB421 Desulfosporosinus meridiei Acidovorax sp. UFZ-B517 Uncultured proteobacterium 1016 Uncultured eubacterium WD254

Acidovorax sp. UFZ-B517 Unidentified bacterium Gallionella ferruginea Pseudomonas lanceolata Uncultured delta proteobacterium Sva0081

<90 99.3 96.9 99.8 97 99.7 <90 98.8 <90 92.7 92.9 99.4 99.7 98.2 99.0 99.9 92.1 <90 <90 <90 99.9 97.6 96.3 99.6 93.3

Number of identical RFLP patterns Closest relatives below 90% sequence identity are neglected

environmental conditions and the evolution of genotypes (PAUL et al., 2000). On the other hand the genetic variability could also result from artifacts caused by, for example, polymerase errors and rRNA operon heterogeneity, and may partly explain the high degree of microheterogeneity typical of sequence clusters detected in environmental clone clusters (NÜBEL et al., 1996; PUKALL et al., 1999; QIU et al., 2001; SPEKSNIJDER et al., 2001; VON WINTZINGERODE et al., 1997). Representatives of closely related genera of the family Comamonadaceae within the β-Proteobacteria are metabolically rather diverse, including phototrophic, facultatively chemolithoautotrophic and obligately chemoorganoheterotrophic bacteria (SCHULZE et al., 1999). Considering the physiological heterogeneity of these bacteria we cannot derive their possible function or impact in the system based exclusively on the analysis of its 16S rRNA sequences. Other sequences within the β-subgroup were affiliated with Thiobacillus sp., an uncultured bacterium from uranium mining waste pile (unpublished), and clones only distantly related to

other sequences within the β-subgroup of Proteobacteria (Fig. 2). A substantial part of the clones from sampling station R13A formed a cluster within the group of Proteobacteria closely related to Acidithiobacillus ferrooxidans (Table 2; Fig. 2). One clone found in sampling station R9 of reactor B showed 97% similarity with sequences of the β subclass of the Proteobacteria found in petroleum-contaminated cavity groundwater (WATANABE et al., 2000). Within the α subgroup of Proteobacteria two clones retrieved from reactor B were closely related to methanol-utilizing bacteria from soils characterized by stable isotope probing and sequencing (RADAJEWSKI et al., 2000). The remaining sequences, which do not belong to the phylum of Proteobacteria, were predominantly associated with low G+C gram-positive bacteria, forming one major cluster related to uncultured bacteria of an anaerobic 1,2-dichloropropane-dechlorinating mixed culture (SCHLOTELBURG et al., 2000) and three clones only distantly related to members of the low G+C gram-positive bacteria. One sequence (RA13C7) shows no specific asso-

Fig. 2. Dendogram depicting relationships among the predominant community members of the groundwater and reactor samples as revealed by comparative analysis of 16S rRNA sequences and those stored in the ARB database and Genbank. 16S rDNA data determined in this study are shown in bold. Sequences with similarities >98% were grouped in the phylogenetic tree. Scale-bar represents 10% estimated change.

238

A. ALFREIDER et al.

ciation with currently recognized bacterial divisions. Since the precise depth and position of unaffiliated or only modestly related sequences cannot be calculated based on only one or a few examples (HUGENHOLTZ et al., 1998), the placement of the sequence in the phylogenetic tree (Fig. 2) should be considered, for the time being, as an estimate. Bacterial phylotypes detected in the inflowing groundwater show a distinct pattern and a broad degree of diversity. Three sequences belonging to the β-Proteobacteria were affiliated with the genus Aquaspirillum. Other clones found within the β-subgroup included one clone (GIF8) closely associated with Alcaligenes sp., and the clone GIF16 which is less than 97% homologous to other genera (Dechloromonas, Rhodocyclus) within the Rhodocyclus group. Sequences found in other subgroups of Proteobacteria include one clone (GIF7) affiliated with isolates or environmental clones retrieved from soils, and within the δ-subgroup of Proteobacteria clone GIF11 clearly belongs to the genus Desulfovibrio. Other clones related to sulfate-reducing bacteria belong to the low G+C gram-positive bacteria with GIF4 clustered with sequences of Desulfosporosinus and clone GIF17, which shows a phylogenetic relationship to the genus Desulfotomaculum. A considerable number of clones (GIF 1; 3; 9) were deeply branched in the pylogenetic tree and cannot be affiliated to known rDNA sequences or clearly placed with currently recognized bacterial phyla. The sequences are distantly related to previously described clones recovered from a low-biomass deep subsurface paleosol (CHANDLER et al., 1998). Two sequences (GIF 10; GIF19) clearly belong to candidate division OP3, a recently proposed novel bacterial division for which there exists no reported cultivated representative (HUGENHOLTZ et al., 2001). The deeply branched clone GIF12 is the sole representative found in this study belonging to the phylum of Actinomycetes. Our results suggest that in the original groundwater several bacterial clones were distantly related to previously characterized sequences and branch deeply within the universal phylogenetic tree. Since these sequences are not specifically related to cultivated organisms with common physiological properties, it is impossible to infer any phenotypic characteristics of these microorganisms based on their phylogenetic placement. Several of the described clone sequences from both reactor outflows were found within the γ-subgroup of Proteobacteria, with two clones which are closely related to Legionella, and clone GOUTA18 with an ultramicrobacterium as its closest relative. In the outflow of reactor B three sequences clustered within the genus Acidovorax and the sequences were very closely related with bacterial strains (B517; B530), which were isolated from the outflow of a laboratory column filled with aquifer sediment from the pilot plant (VOGT et al., 2000). These strains are able to grow aerobically on chlorobenzene as the sole source of carbon and energy. Another clone among the βProteobacteria subgroub was GOUTA14, which was almost identical (99.7%) to a denitrifying Fe-oxidizing bacterial strain, and three clones affiliated with miscella-

neous sequences found in the reactor sediments and the original groundwater. One deeply branched assemblage with a low in-cluster similarity containing 3 clones is most likely to be related to sulfate-reducing genera within the subclass of δ-Proteobacteria. A distinct group of sequences with very high similarity values within the cluster, but < 90% identical to other available sequences, was found within the low G+C gram-positive bacteria. Other sequences within this phylum include clone GOUTB5 affiliated with the clone GIF4, and two almost identical clones which were related to uncultured bacteria recently found in coal tar waste-contaminated groundwater (BAKERMANS and MADSEN, not published). Within the Acidobacterium division, a newly recognized phylum with only three cultivated representatives, clone GOUTB8 was assigned to subdivision 3, which is well represented by environmental clones (HUGENHOLTZ et al., 1998). Clones GOUTB15 and GOUTA3 are deeply branched within the candidate division OP3, which was already described in the context of the inflowing groundwater clone sequences. GOUTA19 was affiliated with members of the Nitrospira group, while one bacterial sequence type (GOUTA4) found in the outflow of reactor had less than 85% sequence similarity to known sequences and could not be clearly associated with any sequences currently recognized in the bacterial domain. Comparison of clone libraries: Although monitoring indicated that the aquifer reactor systems approached chemical steady state of chlorobenzene after 210–260 days after start-up, our investigations suggest that the different habitats were inhabited by distinct microbial communities in regard to their taxonomical diversity. The original groundwater was populated by a bacterial consortium related to different members of the class of Proteobacteria, by phylotypes distantly affiliated with other taxa and also by sporulating and non-sporulating sulfate-reducing bacteria (Fig. 2). In detail, clone libraries from sampling stations R7 and R9 of the reactor sediment samples were rich in β-Proteobacteria, dominated by sequences very closely related to Alcaligenes faecalis. In the upper reactor sector, sequence diversity increased with a number of clones clustering with Acidithiobacillus ferrooxidans in the γ-subclass of Proteobacteria and with low G+C gram-positive bacteria. Although most bacterial phylotypes detected in the groundwater outflow of the reactors also grouped within different subdivisions of Proteobacteria and low G+C grampositive bacteria, these sequences were not associated with the sequence types obtained from the groundwater inflow and the reactor samples. For example GOUTA1, GOUTB7 and GOUTB17 are grouped within the Rhodoferax group, but these sequences are clearly separated from clones dominating the sediment samples. Another cluster of sequences closely related to Acidovorax is found in the outflowing groundwater of reactor B (Fig. 2). Several clone sequences obtained from the inflowing groundwater were also affiliated with the family of Comamonadaceae within the β-Proteobacteria, al-

Microbial diversity in an in situ reactor system

though the most frequently occurring clones formed a separated cluster related to Aquaspirillum, which was not detected in the reactor sediments. A similar arrangement of the clone sequences, with distinct clusters from samples of reactor samples and the groundwater outflow, was also found within the low G+C gram-positive bacteria. The number of clones screened and sequenced in this study are certainly not sufficient to reveal the ‘total’ 16S rDNA diversity or to allow the application of reliable statistical approaches to detect significant differences in diversity between comparable samples (HUGHES et al., 2001). Furthermore inherent biases of PCR can distort library data and the distribution of molecular isolates from clone libraries may not the reflect the original composition of the microbial assemblages in the sample. Nevertheless, the sample sizes were appropriate to obtain the predominating taxa. The differences observed in the microbial community structure from the reactor sediments and the groundwater samples may be explained by the differences between these two habitats, attached and free-living bacteria. Although numerous studies have investigated the effect of different environmental components and mechanisms by which selected bacterial isolates partition between aqueous and solid phases, much less is known about the differences in the microbial community structure between original sediment and groundwater (KÖLBEL-BOELKE et al., 1988; CRUMP et al., 1999). In a recent study, RÖLING et al. (2001) found that pollution in a landfill leachate-contaminated aquifer did not affect the particle-bound microorganisms, but that the groundwater community structure was clearly affected in relation to pollution and redox processes, supporting the hypothesis that bacteria attached to sediment particles and forming biofilms usually consists of stable communities which are less influenced by changing environmental factors. The significance of specific pollutants for the structure within the bacterial assemblages in contaminated groundwater ecosystems is hard to assess, because various physical, chemical and biological factors may often mask anthropogenic effects. In this context it is also of interest that the clone sequences affiliated with sulfate reducing bacteria were exclusively found in the groundwater inflow and outflow, but not in the sediment samples of both reactors. Although sulfate concentrations were very high and the oxygen concentration in the inflowing groundwater and reactor water was below detection limit, sulfate concentrations did not decrease during the reactor flow path, indicating that sulfate was not reduced in situ. The sulfate reduction seems to be electron-donor limited. Therefore sulfate-reducing bacteria present in our system may not develop selective advantages by becoming attached to environmental surfaces because the particle association of bacteria usually only occurs when cells are actively growing and not when they are starved. This assumption was supported by the outcome from a stimulation experiment of the autochthonous bacteria in the reactor with hydrogen peroxide as the oxygen-releasing compound (data not shown), which caused the development of an extensive biofilm also inhabited by sulfate-reducing bacteria.

239

Ultimately, questions of temporal and spatial changes in the microbial community composition caused by planned stimulation experiments of the autochthonous chlorobenzene-degrading bacteria in the reactor sediments, will require quantitative methods. The sequence information obtained in this study are essential for the development of specific gene probes for quantitative slotblot hybridization of RNA extracted or whole cell in situ hybridization technique. Further investigations in the reactor system will also include the detection and analysis of key catabolic genes coding for the (micro)aerobic chlorobenzene degradation pathways based on the analysis of the messenger-RNA. Acknowledgements The authors would like to thank Anett Heidtmann for her technical assistance in the laboratory and Doreen Hoffmann for critical comments on the manuscript. We thank the personnel at the SAFIRA pilot plant for their collaboration. This work was kindly funded by the Federal Ministry of Education and Research (BMBF) within the groundwater remediation project SAFIRA.

References ALTSCHUL, S. F., GISH, W., MILLER, W., MYERS, E. W., LIPMAN, D. J.: Basic local alignment search tool. J. Mol. Biol. 215, 403–410 (1990). BAKERMANS, C., MADSEN, E. L.: Accession AF351223. Unpublished sequence submitted to Genbank. CHANDLER, D. P., BROCKMAN, F. J., BAILEY, T. J., FREDRICKSON, J. K.: Phylogenetic diversity of Archaea and Bacteria in a deep subsurface paleosol. Microb. Ecol. 36, 37–20 (1998). CRUMP, B. C., ARMBRUST, E. V., BAROSS, J. A.: Phylogenetic analysis of particle-attached and free-living bacterial communities in the Columbia river, its estuary, and the adjacent coastal ocean. Appl. Environ. Microbiol. 65, 3192–3204 (1999). DOJKA, M. A., HUGENHOLTZ, P., HAACK, S. K., PACE, N. R.: Microbial diversity in a hydrocarbon- and chlorinated-solventcontaminated aquifer undergoing intrinsic bioremediation. Appl. Environ. Microbiol. 64, 3869–3877 (1998). FIELD, K. G., GORDON, D., WRIGHT, T., RAPPE, M., URBACK, E., VERGIN, K., GIOVANNONI, S. J.: Diversity and depth-specific distribution of SAR11 cluster rRNA genes from marine planktonic bacteria. Appl. Environ. Microbiol. 63, 63–70 (1997). GARCIA-MARTINEZ, J., RODRIGUEZ-VALERA, F.: Microdiversity of uncultured marine prokaryotes: the SAR11 cluster and the marine Archaea of Group I. Mol. Ecol. 9, 935–48 (2000). HUGENHOLTZ, P., GOEBEL, B. M., PACE, N. R.: Impact of cultureindependent studies on the emerging phylogenetic view of bacterial diversity. J. Bacteriol. 180, 4765–4774 (1998). HUGENHOLTZ, P, PITULLE, C, HERSHBERGER, K. L., PACE, N. R.: Novel division level bacterial diversity in a Yellowstone hot spring. J. Bacteriol. 180, 366–376 (1998). HUGENHOLTZ, P., TYSON, G. W., WEBB, R. I., WAGNER, A. M., BLACKALL, L. L.: Investigation of candidate division TM7, a recently recognized major lineage of the domain bacteria with no known pure-culture representatives. Appl. Environ. Microb. 67, 411–419 (2001). HUGHES, J. B. HELLMANN, J. J., RICKETTS, T. H., BOHANNAN, B. J. M.: Counting the uncountable: Statistical approaches to estimating microbial diversity. Appl. Environ. Microbiol. 67, 4399–4406 (2001).

240

A. ALFREIDER et al.

KÖLBEL-BOELKE, J.; ANDERS, E. M., NEHRKORN.: A. Microbial communities in the satured groundwater environment: II. Diversity of bacterial communities and in-vitro-activities in a pleistocene sandy aquifer. Microb. Ecol. 16, 31–48 (1988). LANE, D.J.: 16S/23S sequencing, pp 148–163. In: Nucleic Acid Techniques in Bacterial Systematics (E. STACKEBRANDT, M. GOODFELLOW eds.) New York. John Wiley and Sons. 1991. LUDWIG, W., STRUNK, O., KLUGBAUER, S., KLUGBAUER, N., WEIZENEGGER, M., NEUMAIER, J., BACHLEITNER, M., SCHLEIFER K.-H.: Bacterial phylogeny based on comparative sequence analysis. Electrophoresis 19, 554–568 (1998). MAIDAK, B. L., COLE, J. R., LILBURN, T. G. PARKER, C. T. Jr., SAXMAN, P. R., FARRIS, R. J., GARRITY, G. M., OLSEN, G. J., SCHMIDT, T. M., TIEDJE, J. M.: The RDP-II (Ribosomal Database Project). Nucleic Acids Res. 29, 173–174 (2001). MERKEL, P., TEUTSCH, G., WEISS, H., RIJNAART, H. H.: Innovative reactive barrier technologies for regional contaminated groundwater, pp 532–540. In: Proceedings of the Seventh International FZK/TNO Conference on Contaminated Soil (ConSoil 2000), Vol. 1, Leipzig, Germany, 2000. MUELLER, R. H., JORKS, S., KLEINSTEUBER, S., BABEL, W.: Comamonas acidovorans strain MC1: a new isolate capable of degrading the chiral herbicides dichloroprop and mecoprop and the herbicides 2,4-D and MCPA. Microbiol. Res. 154, 241–246 (1999). NÜBEL, U., ENGELEN, B., FELSKE, A., SNAIDR, J., WIESHUBER, A., AMANN, R. I., LUDWIG, W., BACKHAUS, H.: Sequence heterogeneities of genes encoding 16S rRNAs in Paenibacillus polymyxa detected by temperature gradient gel electrophoresis. J. Bacteriol. 178, 5636–5643 (1996). PAUL, J. H., ALFREIDER, A., WAWRIK, B.: Micro- and macrodiversity in rbcL sequences in ambient phytoplankton populations from the southeastern Gulf of Mexico. MEPS 198, 9–18 (2000). PUKALL, R., PAEUKER, O., BUNTEFUß, D., ULRICHS, G., LEBARON, P., BERNARD, L., GUINDULAIN, T., VIVES-REGO, J., STACKEBRANDT, E.: High sequence diversity of Alteromonas macleodii-related cloned and cellular 16S rDNAs from a Mediterranean seawater mesocosm experiment. FEMS Microb. Ecol. 28, 335–344 (1999). QIU, X., WU, L., HUANG, H., MCDONEL, P. E., PALUMBO, A. V., TIEDJE, J. M., ZHOU, J.: Evaluation of PCR-generated chimeras, mutations, and heteroduplexes with 16S rRNA gene-based cloning. Appl. Environ. Microbiol. 67, 880–887 (2001). RADAJEWSKI, S., INESON, P., PAREKH, N. R., MURRELL, J. C.: Stable-isotope probing as a tool in microbial ecology. Nature 403, 646–649 (2000). RÖLING, W. F, VAN BREUKELEN, B. M., BRASTER, M., LIN, B., VAN VERSEVELD, H. W.: Relationships between microbial community structure and hydrochemistry in a landfill leachate-polluted aquifer. Appl. Environ. Microbiol. 67, 4619–4629 (2001). SCHLOTELBURG, C., VON WINTZINGERODE, F., HAUCK, R., HEGEMANN, W., GOBEL, U. B.: Bacteria of an anaerobic 1,2dichloropropane-dechlorinating mixed culture are phylogenetically related to those of other anaerobic dechlorinating consortia. Int. J. Syst. Evol. Microbiol. 50, 1505–1511 (2000).

SCHULZE, R., SPRING, S., AMANN, R., HUBER, I., LUDWIG, W., SCHLEIFER, K.-H., KÄMPFER, P.: Genotypic diversity of Acidovorax strains isolated from activated sludge and description of Acidovorax defluvii sp. nov. Syst. Appl. Microbiol. 22, 205–214 (1999). SPEKSNIJDER, A. G., KOWALCHUK, G. A., DE JONG, S., KLINE, E., STEPHEN, J. R., LAANBROEK, H. J.: Microvariation artifacts introduced by PCR and cloning of closely related 16S rRNA gene sequences. Appl. Environ. Microbiol. 67, 469–472 (2001). STRUNK O., GROSS O., REICHEL B., MAY M., HERMANN S., STUCKMANN N., NONHOFF B., GINHART T., VIBIG A., LENKE M., LUDWIG T., BODE A, SCHLEIFER K.-H., LUDWIG W.: ARB – a software environment for sequence data. http://www.arbhome.de. Department of Microbiology, Technical University of Munich, Munich, Germany. THERON, J., CLOETE, T. E.: Molecular techniques for determining microbial diversity and community structure in natural environments. Crit. Rev. Microbiol. 26, 37–57 (2000). TORSVIK V, SALTE K, SORHEIM R., GOKSOYR J.: Comparison of phenotypic diversity and DNA heterogeneity in a population of soil bacteria. Appl. Environ. Microbiol. 56, 776–781 (1990). VOGT, C., LORBEER, H., WUENSCHE, L., BABEL, W.: Transformation of chlorobenzene in a laboratory aquifer column, pp. 133–142. In: Bioremediation & Phytoremediation of Chlorinated & Recalcitrant Compounds (A. R. GAVASKAR, B. C. ALLEMAN, G. B. WICKRAMANAYAKE, V. S. MAGA, eds.) Columbus, Richland, Battelle press 2001. VON WINTZINGERODE, T., GOEBEL, U. B., STACKEBRANDT, E.: Determination of microbial diversity in environmental samples: pitfalls of PCR-based rRNA analysis. FEMS Microbiol. Rev. 21, 213–229 (1997). WATANABE, K.: Microorganisms relevant to bioremediation. Curr. Opin. Biotechnol. 12, 237–241 (2001). WATANABE, K., WATANABE, K., KODAMA, Y., SYUTSUBO, K., HARAYAMA, S.: Molecular characterization of bacterial populations in petroleum-contaminated groundwater discharged from underground crude oil storage cavities. Appl. Environ Microbiol. 66, 4803–4809 (2000). WAGNER, M., AMANN, R., LEMMER, H., SCHLEIFER, K. H.: Probing activated sludge with oligonucleotides specific for proteobacteria: inadequacy of culture-dependent methods for describing microbial community structure. Appl. Environ. Microbiol. 59, 1520–1525 (1993). WEISS, H., DAUS, B., FRITZ, P., KOPINKE, F.-D., POPP, P., WÜNSCHE, L.: In situ groundwater remediation research in the Bitterfeld region in eastern Germany (SAFIRA); pp 443–450. In: Groundwater Quality: Remediation and Protection. (M. Herbert, K. Kovar, eds.)., IAHS Publication 250, 1998.

Corresponding author: ALBIN ALFREIDER, UFZ-Umweltforschungszentrum, Sektion Umweltmikrobiologie, Permoserstr. 15, D-04318 Leipzig, Germany Tel.: ++49-341-235 2367; Fax: ++49-341-235 2247; e-mail: [email protected]