Previous   |   Next

doi:10.2204/iodp.proc.301.105.2005

Microbiology

During Expedition 301, we collected samples from both sediments and basaltic crust. Samples were taken for cultivation assays, polymerase chain reaction (PCR)-based molecular phylogenetic studies, activity measurements, and contamination tests. Prior to Expedition 301, the microbiota of deep subsurface sediments and deep subsurface basalts had been investigated in only a few studies (Cowen et al., 2003; Fisk, et al., 1998; D’Hondt et al., 2002, 2004; Inagaki et al., 2003, 2005; Parkes et al., 1994, 2000). The eastern flank of the Juan de Fuca Ridge is of great interest to microbiologists because of the flow of hydrothermal fluids through basement rocks. Delivery of electron acceptors such as sulfate by hydrothermal fluid flow, combined with potential release of electron donors such as reduced metals and perhaps hydrogen (H₂) from the basalt, may support high microbial activity and biomass tens to hundreds of meters below the top of basement. The moderately hot to very hot temperatures (65° to >100°C) may further promote microbial growth. Moreover, transport of sulfate and heat to the comparatively organic rich sediment column may foster high microbial biomass and activity in sediments near basement. Transport of sulfate into sediments from both the overlying water column and underlying basaltic basement creates two sulfate–methane transition zones, thereby enabling us to examine changes in community composition and activity of sulfate reducers, methanogens, and poorly understood anaerobic methane oxidizers along thermal and geochemical gradients at a single site.

Core handling and sampling

Sediment for cultivation and molecular biological analyses

Whole-round samples taken on the catwalk for microbiological analysis were handled aseptically to minimize contamination. Cores were visually inspected prior to sampling of whole rounds for signs of disturbance such as gas voids, cracks, or drilling disturbance. Core sections suitable for microbiological and molecular biological analyses were quickly transported to the adjacent core laboratory, where they were divided into smaller whole rounds, each used for different analyses. These analyses included cultivation assays, bacterial counts, contamination tests, and molecular biological surveys (all discussed below). To maximize comparability of microbiological and molecular biological information with geochemical and lithologic data, microbiological sampling was coordinated with other shipboard studies.

To prevent oxidation of anaerobic sediments, subsamples used for cultivation experiments were immediately subjected to anaerobic handling techniques. Subsamples used for molecular biological analyses were placed rapidly into sterile bags and frozen at –80°C. Core liners are not sterile, meaning the outer surfaces of cores are contaminated during drilling (Smith et al., 2000) and core cutting may introduce contaminants. Because of this, subsampling for cultivation and later molecular biological analyses was restricted to sediments at or near the center of the whole-round samples (Fig. F12B).

Rock for cultivation and molecular biological analyses

RCB basalt cores were handled aseptically to minimize contamination. Samples were obtained from all representative lithologies and units. Suitable rock samples were chosen with preference for large, intact pieces. From every sampling depth, rocks were chosen for cultivation assays, bacterial counts, contamination tests, and molecular biological surveys (all discussed below).

Most subsurface microbes are sensitive to oxygen exposure. Therefore, rock samples for cultivation experiments were transferred as quickly as possible to anaerobic conditions. In some cases, this involved collection of rock pieces directly on the catwalk and immediate transfer to vials containing anaerobic saline solution (3% NaCl) flushed with N₂, followed by slurry preparation. In other cases, rocks were transferred to an anaerobic glove box (97.5% N₂, 2.5% H₂) within minutes of core arrival, cracked with a hammer and chisel, crushed with a mortar and pestle, and further processed in the glove box. A third approach was to wash and flame rock surfaces in the laboratory to remove microbial contaminants (as described in "Contamination tests with perfluorocarbon tracer"), crack the rock with a hammer and chisel, and then immediately transfer some of the resulting rock fragments to an anaerobic glovebox for crushing and further processing. Rocks used in molecular biological analyses were placed rapidly into sterile bags and frozen at –80°C.

Rock for bioreactor experiments

Samples of basement material were collected to conduct bioreactor experiments. All samples were handled aseptically to minimize contamination. These experiments are a follow-up to previous molecular biological studies by Cowen et al. (2003) in which relatives of nitrate- and sulfate-reducing microorganisms had been detected in a BioColumn deployed on the Circulation Obviation Retrofit Kit (CORK) in ODP Hole 1026B. We will conduct experiments onshore to measure microbial activity and examine the possibility of microbially mediated nitrate and/or sulfate reduction in the basaltic oceanic crust.

We collected 5 cm whole-round samples on the catwalk and, based on size and shape, placed them into N₂-flushed polyethylene terephthalate (PET) bags. Samples were placed in the anaerobic glove box and rinsed twice in sterile anaerobic artificial seawater to remove contaminants. Wash water was saved for analysis. Cleaned samples were transferred to sterile PET bags and submerged in artificial seawater, sealed, and refrigerated.

Water sampling

As part of the drilling process, huge amounts of surface seawater are injected into boreholes. This is the major source of contaminating microorganisms to cores collected for microbiological analyses. As a check on contamination (apart from the tracer tests discussed later), we inspected the microbial composition of drilling fluid. Water samples were collected and handled using sterile equipment. Samples of seawater were collected directly from the injection pipe into sterile bottles with screw caps. Microorganisms present in fluids were extracted by filtration using a vacuum pump, first through 5 µm pore polycarbonate filters and then through 0.2 µm pore polycarbonate filters. Using a scalpel, filters were divided into smaller pieces and frozen separately in small cryotubes at –80°C. Tubes were brought back to the laboratory on shore for deoxyribonucleic acid (DNA) extraction and analysis.

Scale on CORK sampling

Mineral deposits growing on the Leg 168 CORK body recovered from Hole 1026B were sampled. Because the basement fluids at this site are overpressured and the CORK was unsealed, the hole has produced basement fluids for several years (Cowen et al., 2003; Fisher et al., 2003). The flow of 65°C basement fluid has prevented the intrusion of cold seawater from above. As a result, there is a strong biogeochemical interface at the seafloor, where warm, anoxic basement water meets cold, oxygenated seawater.

Using sterile spatulas, we scraped off mineral deposits and biofilms from where they coated the outside of the CORK body. Where necessary, we used a hammer and chisel to break off pieces of these materials. Samples used for cultivation assays were transferred immediately to an anaerobic glove box for further processing. Samples used for molecular biological analyses were placed rapidly into sterile bags and frozen at –80°C. No contamination tests could be performed on these samples, but they were exposed extensively to ordinary seawater during formation and as the CORK body was retrieved from the seafloor.

Storage conditions

All sediment cores used for molecular biological analyses were capped, placed in sterile plastic bags, and stored at –80°C. Some of the sediment samples for ribonucleic acid (RNA) analyses were frozen immediately with liquid nitrogen and stored at –80°C. The slurry samples for cultivation were stored either at 4°C or room temperature. They were kept at the same conditions until they were removed from the ship in Balboa, Panama (following the next drilling expedition), and then shipped under the same conditions to their respective laboratories.

Total cell counts

The most direct method for quantifying the extent of the deep microbial biosphere is by total bacterial cell counts using nucleic acid dyes to distinguish cells from organic matter. These counts have been performed on sediments from a wide range of locations during previous ODP legs (Parkes et al., 1994). A common trend across many sampling locations is an exponential decrease of cell numbers with depth. Despite this reduction, cell densities of 10⁴–10⁵ cells/cm³ have been detected consistently even in the deepest sediments. Total cell counts also can reveal sediment layers of increased cell density, which often coincide with geo- and physicochemical conditions conducive to bacterial growth (Parkes et al., 2000).

During Expedition 301, we performed total cell counts on one sample per sediment core. In geochemical or biogeochemical transition zones (e.g., the sulfate–methane transition zone or in bottom sediments that were near the hydrothermally active, basaltic basement), counts were performed at one sample per section to gain a higher-resolution view of microbial distribution. We counted total bacterial numbers under an epifluorescence microscope using acridine orange as a fluorochrome dye (Fry, 1988; Shipboard Scientific Party, 2003a).

We also tried total cell counts on basaltic basement cores using this method. However, strong interference from autofluorescent rock flour makes it difficult to achieve reliable results.

Contamination tests with perfluorocarbon tracer

The assessment of contamination with microbes from drilling fluid (surface seawater and sepiolite mud) is critical during sample collection for microbiological or molecular biological analyses. Certainty that microbes isolated or DNA sequences extracted from IODP cores are part of the autochthonous microbiota, and not contaminants, is obligatory to draw any conclusions about microbial diversity and metabolism in the subsurface. To quantify sample contamination with drilling fluid, perfluorocarbon tracer (PFT) was injected into the fluid to produce a constant concentration of 1 mg/L.

Concentrations of PFT in sediment cores were monitored with a gas chromatograph with an electron capture detector, flushed with N₂ gas, and followed the settings outlined in Smith et al. (2000). To check for intrusion of drilling fluid from the outside of cores to the inside, we took six 2 cm³ samples of sediment from each whole-round sample using cut-off 3 mL syringes. Two samples were from the outer part near the core liner, two from halfway between the outside and the center, and two from the center of every whole-round sample. We generally expected the PFT content of sediment to decline with increasing distance from the core liner, although APC coring can also operate cleanly enough to prevent any (measurable) sample contamination with drilling fluid.

Each sample was placed in a 10 mL vacuum-seal tube. We used these tubes instead of the 20 mL GC vials used by Smith et al. (2000) because we found that the GC vials' septa leaked after they had been punctured. This became especially problematic with samples where measurements had to be repeated—for instance, when the rubber septum of the GC leaked and required replacement, or when pieces of septum got stuck in syringes during injection of sample into the GC, thereby preventing full sample injection. Since sufficient overpressure to push off rubber caps of vacuum-seal tubes could develop during incubation, the rubber cap of each tube was secured by wrapping electric tape around it several times. The secured tube with sample was then incubated at an oven temperature of 75°C for ~10 min. Using a disposable, preheated 1 mL syringe, 0.5 mL of headspace gas was injected into a GC. We found this technique to be more practical for making replicate analyses than repeated use of the same gas-tight syringe (e.g., Smith et al., 2000). We found that reusing the same gas-tight syringe required thorough and repeated flushing with methanol and incubation in the oven, with needle and plunger separated from syringes, for ≥30 min to remove all PFT residue. Even though disposable syringes were not gas tight, loss between transfer from storage tube to GC was minimal. Subsamples for PFT measurement were taken adjacent to all subsamples used for microbiological and molecular biological analyses.

Contamination of basalt samples and the effectiveness of decontamination treatments were examined in small pieces of rock from the exterior and interior. Rock pieces had been removed using a hammer and chisel and placed rapidly into 10 mL vacuum-seal tubes. To remove PFT, and hence microbial contaminants, from rock surfaces, each rock sample underwent a series of cleaning treatments. Removal of PFT from rock surfaces was essential to measure contamination in the interior. To prevent cross contamination, all surfaces and tools used were washed twice with ethanol and flamed between treatments. To evaluate the success of cleaning treatments, we took samples at each step. First, pieces of the untreated rock exterior were removed. The rock was washed twice by placement into a sterile bag containing sterile saline solution (deionized water; 3% NaCl) followed by vigorous shaking. For further PFT removal, the rock was flamed thoroughly using a propane torch until the exterior was completely dry. Finally, rocks were cracked open and subsamples taken from the rock interior. Where possible, only pieces that had not been in contact with the outside of the rock sample were analyzed. In cases where we could not obtain pieces that had not been in contact with the rock exterior, we chose pieces that appeared to have the lowest contact to the outside relative to their volume. Incubations and injections followed the same method as for sediment. The absence of PFT in the untreated exterior of rocks was interpreted to indicate failure of the PFT delivery system because the RCB system used in basalt drilling should have led to contact of drilling fluid with essentially all samples. Greatly reduced PFT values on rock surfaces after cleanup treatments indicated successful removal. Zero to very low PFT concentrations in rock interiors were interpreted to indicate minimal contamination.

Cultivation

Based on culture-independent molecular analyses using subseafloor sediment cores, most microbial communities in deep sediments appear to be composed of previously unidentified, yet-to-be-cultured microorganisms (Marchesi et al., 2001; Reed et al., 2002; Inagaki et al., 2003, 2005; Kormas et al., 2003; Newberry et al., 2004). There are, however, examples of successful microbial isolations from deep subsurface environments. For instance, Desulfovibrio profundus was isolated from the deep sediment of the Japan Sea at a depth of 500 mbsf (ODP Leg 128, Site 798B) (Bale et al., 1997). Methanoculleus submarinus was isolated from deep methane hydrate-bearing sediments in the Nankai Trough (Mikucki et al., 2003). In an attempt to increase knowledge of physiology of subseafloor microorganisms and to better understand the role of these microbes in the biogeochemical cycles of their environment, we tried to enrich for various types of microorganisms using a wide range of media and culture conditions (Tables T13, T14, T15, T16, T17, T18, T19, T20, T21, T22). We prepared (1) culture media without organic carbon substrates to select for chemolithoautotrophic microbes and (2) media with organic carbon compounds to select for chemohetero- and chemoorganoautotrophic microorganisms.

Slurry preparation

Slurries were prepared in 200 mL flasks closed by rubber stoppers that allowed aseptic flushing with N₂. Nitrogen gas was supplied with stainless steel tubes equipped with gas filters. Side arms could be closed by sterile stopcocks. Flasks contained 50 mL of sterilized artificial seawater that had been autoclaved and cooled under N₂. Slurries were prepared by aseptically adding sediment contents from 50 mL syringes under an N₂ counterflow. Slurries were homogenized by repeated shaking and vortexing. Subsamples of this "master slurry" were used for cell counts, cultivation, and fluorescence in situ hybridization (FISH) analyses.

Most probable number

Using classical cultivation techniques such as the most probable number (MPN) cultivation method, various physiological types of microorganisms have been enriched and their numbers have been determined from marine sediments (e.g., Parkes et al., 2000; D’Hondt et al., 2004; Inagaki et al., 2005; Inagaki and Nealson, in press). For MPN counts, a series of tenfold dilutions is prepared from initial slurry and incubated (e.g., in 96 well microtiter plates) (Fig. F13). A sufficient number of dilutions are performed until a concentration is reached where no growth is detected and cells are presumably absent. Based on the number of dilutions necessary for no growth to occur, it is possible to estimate cell concentrations of the initial slurry and, hence, the environmental sample. MPN population counts obtained from the sediment column of the Peru margin (ODP Leg 201) ranged from 0 to 10⁵ cells/cm³, showing a general decrease with depth. In general, <0.6% of total cell numbers in deep sediments are detected by MPN. Therefore, MPN counts only yield limited quantitative information about deep subsurface microbiology. However, isolation of microorganisms via high dilutions is an effective tool to select for phylotypes that are numerically abundant in the initial sample. Fast growing, opportunistic cells that are numerically insignificant in the environment, but cope best with culture conditions and therefore dominate many enrichments, can be eliminated.

Gradient cultures

To examine the response of microbial communities from undisturbed sediment samples to low substrate concentrations, gradient cultures were inoculated with nonhomogenized sediment samples (Fig. F14). A mixture of monomers (4 mM each; 0.5 mL) was placed on the bottom of a glass tube and fixed with 0.5 mL of agar (4% in marine salt medium [MM]; 60°C). After the agar had solidified, another 2.5 mL of the same agar was added as a spacer and allowed to solidify. A 2 mL sediment sample in 5 mL MM was placed on top and the tube was gassed with N₂, closed by a rubber stopper, and cooled on ice.

Molecular biological surveys

Culture-independent molecular biological surveys are becoming an indispensable and powerful approach to investigate microbial diversity in the environment. Several kinds of molecular biological analyses will be performed in postcruise investigations and used for comparison with results from culture experiments. The most commonly used approach to obtain DNA sequences of genes uses PCR, an enzymatic reaction that produces replica of DNA. This reaction is repeated until target DNA has reached high enough concentrations to be cloned and sequenced. Molecular techniques using PCR allow rapid, sensitive, and extensive surveys of present (or past) marine subsurface microbial communities (Marchesi et al., 2001; Inagaki et al., 2001, 2003, 2005; Reed et al., 2002; Kormas et al., 2003). Numerous molecular approaches rely on PCR. Following, we list techniques that we will use as part of our postcruise work.

FISH

FISH direct count (FISH-DC) and catalyzed reporter deposition-FISH (CARD-FISH) will be performed to estimate total microbial biomass as well as specific biomass of targeted phylogenetic groups. In general, CARD-FISH shows a higher sensitivity than the commonly used FISH technique. It is based on the use of horseradish peroxidase-labeled oligonucleotide probes and tyramide signal amplification (Bobrow et al., 1989). This novel technique was introduced into microbial ecology for the identification of bacterial cells in marine plankton and benthos by Pernthaler et al. (2002). The signal amplification is essential to detect metabolically active cells with low numbers of ribosomes, as is typical for deep subsurface microorganisms.

Terminal restriction fragment length polymorphism

Terminal restriction fragment length polymorphism allows identification of PCR products based on length variations after digestion with restriction endonucleases. It is used as a fingerprinting technique for nucleic acids of mixed microbial communities (Inagaki et al., 2001).

Gel electrophoresis

Denaturing gradient gel electrophoresis and temperature gradient gel electrophoresis are gel electrophoretic methods that separate mixed PCR products in denaturing polyacrylamide gradients and are based on melting domain structure of the DNA double strand (Muyzer and Smalla, 1998). Bands appearing on the gel can be extracted and sequenced for phylogenetic identification of dominant members of the microbial community.

Real-time PCR

Real-time PCR, a modification of classical PCR, is essentially a fluorogenic assay used to quantify the number of target genes, and hence cells, in a given environmental sample (Suzuki et al., 2000; Takai and Horikoshi, 2000).

Copy DNA PCR

Copy DNA (cDNA)-PCR of reverse-transcribed RNA is a technique by which the 16S ribosomal RNA molecule, or the unstable messenger RNA copy of a gene, is reverse-transcribed to its respective cDNA (Inagaki et al., 2002). Because prokaryotic DNA is likely to be preserved in anoxic sediments for geological timescales (Inagaki et al, 2004; Inagaki and Nealson, in press), we try to distinguish metabolically active from inactive, buried cells by RNA-based molecular techniques. The cDNA is then used for qualitative and real-time PCR (Wilson et al., 1999). Using qualitative PCR, we can identify phylotypes of metabolically active cells, whereas with real-time PCR we cannot only identify phylotypes of active cells but also quantify metabolic activity of those phylotypes.

Gene sequencing

Gene sequencing is used to obtain the primary information (base sequence) of the gene itself and is essential for phylogenetic analysis and identification. Genes that we will analyze include 16S ribosomal DNA, based on which sequence libraries of microbial diversity will be established and compared downhole. Furthermore, genes encoding key enzymes of different metabolic pathways will be PCR-amplified, sequenced, and analyzed with respect to their composition and distribution along environmental gradients. We will examine functional genes involved in sulfate reduction (dissimilatory sulfite reductase, adenosine-S′-phosphosulfate [APS] reductase), methanogenesis (methyl coenzyme-M reductase), acetogenesis (formyl tetrahydrofolate synthase), complex organic compound degradation (benzoyl-CoA reductase; group I/II dehalogenase), the Calvin-Benson-Bassham cycle (ribulose bisphosphate carboxylase I/II/III[?]), the acetyl-CoA pathway (carbon monoxide dehydrogenase), the reverse citric acid cycle (ATP synthase), and other processes.

Top of page   |   Previous   |   Next