Samples were collected during ODP Leg 187 to the north of the Australian Antarctic Discordance (AAD), Southeast Indian Ridge (Christie, Pedersen, Miller, et al., 2001; Thorseth et al., 2003). Samples of basalt, crystalline or breccia, with or without glassy margins were collected from the 18- to 28-Ma crust (Table T1). The water depth at the sites was 4000-5000 m, and the thickness of the sediment layer was up to 233 m. The maximum penetration was 374.2 meters below seafloor (mbsf). Igneous rock and sediment were recovered by using a rotary core barrel (RCB). The sediment cores were called "wash cores" because the RCB was pushed into the sediment layer and an unknown quantity of sample material passed through the core. Polycarbonate core liners were inserted into the core barrels, and after the core arrived on the deck the core liner was split longitudinally. The sample collection and preparation is described further in Christie, Pedersen, Miller, et al. (2001). The samples were placed under anaerobic and cold conditions within 30 min after the core arrived on deck. Sediment samples, surface seawater, collected by lowering a sterile bottle from the ship, and sepiolite (drilling mud) were collected for comparison. Surface seawater was used as drilling fluid. The seawater and the sepiolite samples were used for molecular analysis only. A list of all samples taken for microbial studies is given in Tables T1 and T2.
The samples for DNA analysis were transported frozen to Bergen, Norway, on dry ice. The enrichment cultures were transported at ~0°C, but were unfortunately stored at ~19°C for 1 week during transportation from Fremantle, Australia, to Bergen. This may have led to a loss of the original microbial diversity in the enrichments.
Seventeen types of bacterial culture media, twelve anaerobic and five aerobic, were used to enrich viable microbial populations from the rock and sediment samples. Eight different media based on filtered (0.2-µm pore size) and autoclaved natural seawater (anoxic or oxic) either without additions or with the addition of methanol, lactate, succinate, glucose, or yeast extract as carbon sources, were used. In addition, the following media were used: for iron reducers (iron basal culture medium: Lovley and Phillips, 1988, and Fe-TSB [iron-tryptone soya broth] medium: Küsel et al., 1999), manganese oxidizers (PYGV medium: Staley, 1968), sulfate reducers (W20: Widdel and Bak, 1992), methanogenic Archaea (methanogenic medium 2: Jones et al., 1983), and methanotrophic bacteria (NMS medium: Hanson et al., 1992). Approximately 1 g of crushed rock sample was added to each 10-mL bottle of microbial enrichment medium.
In addition, 4-5 g of crushed rock was added to two different types of microcosms, aimed at enriching microbes participating in the cycling of iron and manganese, in 20-mL tubes filled with anaerobic seawater. One contained Mn4+ as an electron acceptor, and the other contained Fe3+. Chitin, pectin, and acetate were added as carbon sources. The setup was designed to establish a gradient from anaerobic conditions at the bottom of the tube to aerobic conditions at the top. Small amounts of oxygen were added at the top of this gradient.
Sediment enrichment cultures, used as controls, were set up on the same microbial enrichment media as basalt, except that no microcosms were started with sediment inoculum.
All enrichment cultures were incubated at 4°C. Growth was determined by phase-contrast microscopy after ~1 month of incubation.
Sulfide was measured in sulfate-reducing media (W20) according to the quick colorimetric method described by Cord-Ruwisch (1985). Methane was measured in cultures of methanogenic medium by gas chromatography. Reduction of ferric iron (Fe3+) to ferrous iron (Fe2+) in cultures for iron-reducing bacteria (IRB) were observed as a color change from reddish brown to light gray. All analyses of metabolic products were performed after ~1 month of incubation.
PCR amplification (95°C for 15 min, 30-36 cycles of denaturation at 92°C for 1 min; annealing at 55°C for 30 s; extension at 72°C for 1 min; and a final extension at 72°C for 10 min) of the V3-region of 16S rDNA using HotStar Taq DNA polymerase (Qiagen) was performed in a GeneAmp2400 thermal cycler (PerkinElmer Applied Biosystems, USA). The bacterial primers PRUN518r and PRBA338f (Øvreås et al., 1997) were used. This primer set amplifies a 236 base-pair DNA segment.
Aliquots of 5 µL of PCR products were analyzed by electrophoresis on 1.5% agarose gels stained with ethidium bromide. No DNA analysis was performed for Archaea.
PCR products were further analyzed by DGGE. For DGGE analysis, 1-mm-thick polyacrylamide gels (8% [weight/volume] acrylamide-bisacrylamide; Bio-Rad) were prepared with and electrophoresed in 0.5x TAE (0.04-M Tris base, 0.02-M sodium acetate, 1-mM EDTA [pH 7.4]). Next, 15 µL of PCR product mixed with 3 µL of loading buffer (blue-orange loading buffer; Promega) was added to the DGGE wells and a prerun (20 V for 10 min) was performed to concentrate the DNA on the bottom of the wells. The DGGE conditions were 25%-65% urea gradient, 18 hr, 70 V, and 65°C.
The gels were stained with a 1:10,000 dilution of SYBR Gold (Molecular Probes) in 1x TAE for 30 min and examined under ultraviolet light.
Samples of individual DGGE bands were obtained by excising small cores of the gel with sterile 1000-µL pipette tips. These gel cores were added to sterile 1.5-mL microcentrifuge tubes with 20 µL of sterile double-distilled water for elution of the DNA from the gel (diffusion of DNA into the water overnight at 4°C). Subsequently, these samples were reamplified and the DNA was sequenced.
Reamplified PCR products of excised DGGE bands were purified with the QIAquick PCR Purification kit (Qiagen). The BigDye Terminator Cycle Sequencing kit (PerkinElmer Applied Biosystems, USA) was used for direct sequencing in the PCR machine (25 cycles of denaturation at 96°C for 15 s, annealing at 50°C for 10 s, and extension at 60°C for 2 min). Sequencing reactions were analyzed with an ABI 377 DNA sequencer (PerkinElmer), and the obtained DNA sequences were analyzed using the BLAST tool (Karlin and Altschul, 1990) at the National Center for Biotechnology Information (NCBI) World Wide Web site (www.ncbi.nlm.nih.gov).
Evolutionary distances of the DNA sequences were calculated using Clustal_X software (Thompson et al., 1997), and phylogenetic trees were constructed using the neighbor-joining algorithm (Saitou and Nei, 1987). The phylogenetic trees were bootstrapped (1000 bootstrap replicates) using the Clustal_X program. For simplifying reasons, identical or near-identical (>97% similarity) sequences are presented as one operational taxonomic unit (OTU).
The PCR-DGGE analysis of partial 16S rDNA was chosen because it is practical to use with a large number of samples, even though phylogenetic information is lost because of a short fragment length. The DGGE fingerprinting method also provides useful information on community structure and can be used to view the similarities and differences between samples.
The sequences reported in this paper have been deposited into GenBank (Benson et al., 2000) under accession numbers AY129833 through AY129946.
To examine possible drilling-induced contamination, yellow-green fluorescent carboxylate microspheres (Polysciences, Inc.) 0.518 µm (±0.021 µm) in diameter were used as a particulate tracer (Smith et al., 2000). The microsphere suspension was diluted to a concentration of ~1010 mL-1, which was then poured into a plastic bag and secured inside the core barrel. The first core to enter the barrel ruptured the bag and dispersed the microspheres into the core barrel. After splitting the core liner, rock surfaces were washed with distilled water and the water was filtered onto polycarbonate filters. To check for microspheres within rocks, thin sections were prepared without any special precautions. For sediment cores, smear slides were prepared from both the outer part and the interior of the cores. Filters, thin sections, and smear slides were examined under an epifluorescence microscope with a blue filter set. The contamination tracer test is further described in Smith et al. (2000) and in Christie, Pedersen, Miller, et al. (2001).