Previous   |   Next

doi:10.2204/iodp.proc.301.106.2005

Microbiology

Site U1301 lies on a sediment-buried basement ridge on the eastern flank of the Juan de Fuca Ridge. The basement at this location hosts vigorously connecting hydrothermal fluid from the crust that enters the overlying sediment. This presents an opportunity to investigate microbial communities in distinct biogeochemical zones within the sediment column as well as within igneous basement. The Expedition 301 microbiology program was designed to collect sediment and igneous rock samples to attempt cultivation of indigenous microbes, study phylogenetic diversity, estimate the total and active microbial population densities, quantify microbial biomass, measure potential rates of microbial metabolism, and investigate microbial alteration of primary igneous minerals.

Sediment microbiology

The lithology and geochemical/thermal gradients of the sediment section at Site U1301 (Holes U1301C and U1301D) are nearly identical to those at Site 1026, which is located only 1–2 km north and has been characterized previously (Shipboard Scientific Party, 1997). Site U1301 has two sulfate–methane transition zones at ~30–50 and ~100–120 mbsf. Sulfate reduction is the dominant terminal electron-accepting process of organic matter remineralization above and below the two transition zones, whereas methanogenesis is the dominant terminal electron-accepting process of organic matter remineralization between the two transition zones. Sulfate in the upper part of the sediment column comes from ocean bottom seawater, whereas sulfate in the lower part diffuses up from young hydrothermal crustal fluids (altered seawater). The two sulfate transition zones are exposed to different temperatures (~15° and ~40°C). The geochemistry of the upper zone is influenced by water column processes (e.g., photosynthetic primary production, deposition of turbidites, or volcanic ash), whereas that of the lower zone is additionally controlled by the composition of hydrothermal fluids that flow within the uppermost oceanic crust. The differences in thermal and geochemical conditions between the zones make for an interesting comparison of their respective microbial communities. Because of operational and time constraints, the sediment section at Site U1301 was not continuously cored in Hole U1301C. Near the end of the expedition, we had time to return to the site to fill a gap in Hole U1301D. However, microbiological samples from Hole U1301D were only collected, processed, and preserved for shore-based analyses. In addition to the dual sulfate–methane transition zones at Site U1301, the sediment/basement interface at ~265 mbsf is where basaltic oceanic crust inundated by warm, electron-acceptor-rich fluid (~64°C; ~15 mM SO₄^2–) meets organic matter—and, hence, electron-donor-rich sediment. These conditions are likely to support high microbial activity and biomass. Moreover, the sediment column at Site U1301 has a wide temperature range (~2°–64°C), making it well suited for the study of microbial community composition and activity along temperature gradients in deep-sea sediments. The lithologic heterogeneity with sharp interfaces between layers (e.g., between hemipelagic clay and turbidite sand) allows us to study changes in microbial communities across different lithologies.

Sample preparation and distribution

Samples were taken as listed in Table T13. Nineteen APC sediment cores were taken over the depth interval of 0–265.3 mbsf in Hole U1301C. We obtained samples from all cores except 301-U1301C-14R, which had poor recovery. The majority of cores (14) were taken above 130 mbsf. Additional cores were obtained from the middle (180–198 mbsf) and bottom (238–265 mbsf) of the hole. Additional coring was conducted in Hole U1301D over 120–177 mbsf.

Estimates of contamination with drilling fluid

We measured PFT concentrations along cross-sectional transects in cores. Three general locations were sampled: (1) the exterior adjacent to core liner, (2) halfway between the core liner and the center of the core, and (3) the center of the core. For each location, duplicate samples from opposite sides of the core were taken and their PFT concentrations averaged. Contamination was generally low, averaging for all cores at 0.23, 0.03, and 0.02 µL/g for the exterior, halfway, and center, respectively (Fig. F52), and ranging 0–2.12, 0–0.30, and 0–0.39 µL/g for individual samples in the exterior, halfway, and center, respectively (Table T14). PFT was absent in measurable amounts from 9 of the 23 samples analyzed and could only be detected in 12, 10, and 4 of the samples from the exterior, halfway, and center, respectively (Table T14). Direct cell counts estimated an average of 1.85 × 10⁷ ± 0.26 × 10⁷ cells/L for surface seawater (drilling fluid). Based on this mean, contamination with cells from seawater ranged 0–39.1 cells/g (mean = 4.3 ± 10.1 cells/g) in the exterior part of each core, 0–5.5 cells/g (mean = 0.6 ± 1.3 cells/g) in the halfway part of each core, and 0–7.1 cells/g (mean = 0.4 ± 1.5 cells/g) in the center part of each core (Table T14). Total cell counts, conducted in parallel for each sample (see "Total cell counts," below), were used to estimate ratios of contaminant to indigenous cells. Ratios ranged from 0 to 114.6 × 10^–9 (mean = 21.4 × 10^–9 ± 36.1 × 10^–9) in the exterior part of each core, 0 to 19.7 × 10^–9 (mean = 3.2 × 10^–9 ± 5.9 × 10^–9) in the halfway part of each core, and 0 to 17.4 × 10^–9 (mean = 1.2 × 10^–9 ± 3.8 × 10^–9) in the center part of each core (Table T14). This is a conservative estimate because PFT molecules have high volatility and a much smaller size than bacterial molecules, so they are likely to diffuse and hence penetrate cores much faster than contaminant cells.

We examined potential relationships between drilling fluid contamination and variables such as core depth (mbsf), location within the core, and lithology (sand versus clay). Our data suggested no relationship between drilling fluid contamination and core depth or lithology (Table T14). In contrast, although our data set is small and biased toward shallow sediments, there appears to be a relationship between contamination and sample location within a core: four out of five samples from the first section of each core (Cores 301-U1301C-1H, 5H, 11H, 12H, and 19H) and eight out of eleven samples from the second section of each core (Cores 1H, 3H, 6H, 7H, 8H, 9H, 10H, 13H, 15H, 16H, and 18H) were contaminated. By contrast, none of the three samples from the third section of each core (Cores 301-U1301C-2H, 4H, and 19H) were measurably contaminated, and only one of the two samples from the fourth and fifth sections of each core (Cores 17H, 18H, and 19H) was measurably contaminated (Table T14).

Total cell counts

We obtained a vertical depth profile of prokaryotic cells from 24 sediment samples (code = AODC; Table T15) by direct counts of fixed, acridine orange–stained cells. The cell numbers were counted independently by two shipboard microbiologists to balance individual counting biases. Cells were detectable at all depths analyzed. Staining alone with acridine orange or the alternative fluorescent dye DAPI did not provide satisfactory results. Samples from surface sediments in particular contained high amounts of diatom ooze, which emitted a strong fluorescent signal after staining (Fig. F53). To improve our ability to distinguish cells from background, we prepared several dilutions and used the one that optimized the signal-to-noise ratio prior to counting cells (Fig. F54).

Total cell counts decreased slightly with depth, from near-surface concentrations of 7.5 × 10⁸ to concentrations of 1.8 × 10⁷ cells/cm³ at 248 mbsf. Overall, the profile of microbial cell densities (Fig. F55) followed a similar trend as for other ODP sites (Parkes et al., 1994). Our cell counts over the upper 70 m were slightly higher than those reported for Site 1026 during Leg 168 (Mather and Parkes, 2000). One sample (Section 301-U1301C-19R-1; 259 mbsf) was judged as an outlier: cell counts were much higher than in samples from surrounding locations. Examination of PFT (Table T14) and geochemical data indicated strong contamination of this sample.

Although microscopic examination only provides rough estimates of microbial diversity, we were able to distinguish morphotypes along the depth profile (Table T15). Tiny coccoid-shaped cells dominated throughout the sediment column. Numbers of rod-shaped cells fluctuated strongly. Aggregates of up to 30 microbial cells were detected in four horizons between 63 and 90 mbsf.

Interestingly, an increase in cell numbers was observed near the sediment/basement interface. This increase in biomass may be supported by upward flux of electron acceptors from hydrothermal fluids in the underlying basement. Lowest cell densities of 1.8 × 10⁷ to 4.1 × 10⁷ cells/cm³ were observed in the zone where hydrothermally introduced sulfate was almost depleted (240–252 mbsf). Sulfate may play an important role as an electron acceptor in the deepest zone of the sediment column and stimulate microbial growth. The continuous increase in cell numbers from 7.5 × 10⁷ to 2.3 × 10⁸ cells/cm³ near the basement (256–262 mbsf) supports this interpretation and provides an example of how the deep biosphere may be fueled by hydrothermal fluids.

Cultivation experiments

Geochemical analyses indicate the occurrence of methanogenesis and sulfate reduction within the sediment column, yet nothing is known about the identity and respective metabolism of microbes present. We performed cultivation experiments during Expedition 301 using three different approaches. In total, ~1000 enrichment cultures of indigenous microorganisms were inoculated on board.

In the first approach, samples from inner parts of cores (from a total of 25 horizons; described as CLTI in Table T13) were used. Cultivation media were designed to target (1) methanogens, (2) sulfate reducers, (3) hydrogen/sulfur oxidizers, and (4) fermenters (see "Microbiology" in the "Methods" chapter). Slurries were made by mixing 25 cm³ of sample with 50 mL of synthetic seawater containing 0.05% (w/v) sodium sulfide under an N₂ atmosphere (200 kPa). Aliquots of 500 µL from these slurries were inoculated in tubes containing 3 mL of medium and incubated at 20°, 37°, 55°, 70°, and 85°C. The total number of enrichments was 336. The remainder of slurries were stored at 4°C for shore-based studies. In addition, using Mono medium (see "Microbiology" in the "Methods" chapter) as an inoculum, 25 cm³ aliquots of sediment samples were incubated at different temperatures: (1) sample codes 4-2, 8-2, 11-1, 12-1, and 13-2 at 20°C; (2) sample code 15-2 at 37°C; (3) sample code 17-4 at 55°C; and (4) sample codes 18-2, 19-1, 19-2, 19-4, and 19-5 at 70°C (for more information see "Microbiology" in the "Methods" chapter).

A second cultivation approach was performed on samples from 11 depth intervals (sample code BERT; Table T13). Samples were taken from the innermost part of whole-round cores and transferred to three series of sterile 100 mL flasks. Two series were flushed with nitrogen, one of which additionally contained a mild reducing agent. The third series was flushed with a mixture of 80% hydrogen and 20% carbon dioxide. Samples will serve as inocculi for further shore-based cultivation experiments. Onboard cultivation was performed in Mono medium (see "Microbiology" in the "Methods" chapter) under oxic and anoxic conditions at room temperature. Sediment samples from the 11 depth intervals were cultivated in gradient cultures, dilution series in test tubes, and most probable number (MPN) enrichment series in microtiter plates (see "Microbiology" in the "Methods" chapter). The total number of enrichments was 372 cultures. One week after sampling, growth was detected only in some of the aerobic enrichments.

In a third approach (sample code DNASS; Table T13), samples were collected from the center part of each core with a 60 mL syringe and stored for shore-based deoxyribonucleic acid (DNA) extraction. Samples for cultivation were taken along the inner edges of the remaining part of the core and transferred immediately to sterile 50 mL serum bottles under constant flushing with N₂. Bottles were sealed with butyl rubber stoppers, and a slight overpressure was applied. Finally, 2–3 mL of sulfide-reduced (2 mM Na₂S) artificial seawater was added to each of the bottles. The bottles were kept refrigerated and returned to shore for further cultivation experiments. From each core, 3 cm³ of sediment was collected with a 3 mL syringe and transferred to a 50 mL serum bottle containing 30 mL of Met3-medium (see Table T13 in the "Methods" chapter). The bottles were shaken vigorously until the sediment was evenly distributed. These slurries were used as an inoculum in cultivation experiments. Enrichments targeting methanogens were made from all sampled cores. From the last three cores sampled (Cores 301-U1301C-17R through 19R), additional enrichments selecting for sulfur reducers and chemolithotrophic nitrate reducers were started. Incubation temperatures were 5°, 20°, 37°, 55°, and 70°C, corresponding to known in situ temperatures at Site 1026. A total of 44 enrichment cultures were inoculated using this cultivation approach.

None of the anaerobically incubated enrichments showed growth during Expedition 301. The incubation time was probably too short for most of the microorganisms to grow. All shipboard microbiologists will continue the incubation experiments on shore. Additional enrichments will be started that will focus on cultivation of methanogens and homoacetogens at different temperature and substrate treatments and measure potential rates of methanogenesis (sample codes = CULT-L and AODC). We expect anaerobically incubated enrichments to start showing signs of growth within weeks after end of expedition.

Basalt microbiology

Previous studies of basaltic rock obtained by ODP drilling have suggested a potentially extensive microbial biosphere in oceanic crust (Giovannoni et al., 1996; Fisk et al., 1998). Microscopic observations and nucleic acid staining indicated low densities of cells in basaltic rocks within volcanic glass and along fractures (Fisk et al., 1998). To date, there are no extensive studies on microbes inhabiting ocean crustal basalt. No cultured members exist, and nothing is known about phylogenetic diversity or microbial metabolism.

Total cell counts

We counted total cell numbers in samples fixed in 60% ethanol and containing (1) small pieces of basalt and (2) basalt that had been crushed to powder. Unfortunately, the material itself showed a high amount of cell-like structures (small crystals and needles) with high fluorescent signals. Even after testing a variety of dilutions that had been filtered and stained, it was impossible to distinguish cells because of the background fluorescence.

Estimates of contamination with drilling fluid

We measured PFT concentrations on rock exteriors and interiors. PFT measurements were conducted on 15 rock samples, each from a different core. PFT was detected in all but one sample (Section 301-U1301B-33R-1). To remove PFT, and thus potential microbial contaminants, from rock surfaces, we washed rocks twice in sterile bags containing sterile saline solution (ultrananopure deionized H₂O and 3% NaCl) and flamed rocks using a propane torch until the exterior was completely dry. We then took samples from the rock interior. To compare PFT removal, we measured PFT content in pieces of (1) untreated rock exterior, (2) 2× washed rock exterior, (3) 2× washed + flamed rock exterior, and (4) rock interior (Table T16; Fig. F56). Our PFT removal method was highly effective for the 14 samples with measurable PFT concentrations: on average 83% of PFT was removed via washing, and 99% was removed after additional flaming. PFT content of the rock interior was as low as on the outside after flaming (Fig. F56). In 4 of the 14 rock samples where PFT was detectable on the outside, no PFT was measured in the interior. In the 9 rock samples with measurable PFT values, drilling fluid contamination estimated from PFT concentrations ranged from 0.003 to 0.081 µL/g. Based on acridine orange direct cell counts during this cruise, microbial cell concentrations in surface seawater (drilling fluid) were on the order of 1.85 × 10⁷ ± 0.26³ × 10⁷ cells/L. This equates to 0.1–1.5 contaminant cells/g basalt in our samples. Contamination of basalt with microbes from drilling fluid therefore appears to be very low. Unfortunately, due to the difficulty of quantifying densities of basalt-inhabiting microbes, we could not compare numbers of contaminant cells to densities of indigenous cells. Nonetheless, contamination of rock interiors with drilling fluid is so low that it probably is insignificant and even low compared to the minute but inevitable contamination during the "aseptic" handling process of samples.

Cultivation experiments

Using slurries or rock pieces, we inoculated ~300 test tubes in 12 different growth media at five different temperatures (20°, 37°, 55°, 70°, and 85°C; see "Microbiology" in the "Methods" chapter). After 2 weeks of incubation, we observed cell growth in several culture media (<10% of total cultures) (Table T17). We obtained cells that could grow at near in situ temperature, potentially suggesting successful enrichment of indigenous microbes from the warm, shallow basalt aquifer (Table T17). Microscopic observations of DAPI-stained cells revealed coccoid-shaped cells attached to iron sulfide particles. These particles were part of the growth medium. Curiously, in these enrichments no cells were found in association with basalt particles (Fig. F57). Considering the chemical composition of the Mono medium (see "Microbiology" in the "Methods" chapter), these microorganisms probably grow with the provided substrates as carbon sources and ferrous iron as an electron donor. In other enrichments at room temperature, we found anaerobic mesophilic microbes, likely to be fermenters and/or heterotrophic sulfate reducers. There are three conceivable explanations for the retrieval of mesophilic strains: microbes might be (1) derived from sediment above basement, (2) contaminants imported by drilling fluid, or (3) relics transferred to the basaltic oceanic crust by hydrothermal circulation. Further physiological and phylogenetic characterizations of retrieved microbes will be performed as part of shore-based studies. We will compare the results from culture experiments to the results obtained from culture-independent molecular analyses.

Top of page   |   Previous   |   Next