Supplemental Materials for Colonization of Subsurface Microbial Observatories Deployed

Supplemental Materials for “Colonization of subsurface microbial observatories deployed in young ocean crust”

Beth N. Orcutt1*, Wolfgang Bach2, Keir Becker3, Andrew T. Fisher4, Michael Hentscher2, Brandy M. Toner5, C. Geoffrey Wheat6, Katrina J. Edwards1†

1: University of Southern California, Los Angeles, CA, USA

2: University of Bremen, 28334 Bremen, Germany

3: University of Miami, Miami, FL, USA

4: Earth and Planetary Science Department, University of California Santa Cruz, Santa Cruz, CA, 95064, USA

5: University of Minnesota, Twin Cities, St. Paul, MN, USA

6: Global Undersea Research Unit, University of Alaska, Fairbanks, Moss Landing, CA, 95039 USA

*: now at the Center for Geomicrobiology, Aarhus University, DK-8000, Aarhus C, Denmark;

†: to whom correspondence should be addressed. Email: ; phone: 213-821-439

Supplemental methods:

The basalt used in the colonization experiments originated from the 9oN site on the East Pacific Rise, the harzburgite was collected from the Mariana forearc (courtesy of Sherm Bloomer), the pyrite/hematite sample was collected from a sulfide mine in northern California (Iron Mountain), and the biotite was purchased commercially (Wards Geology).

Following CORK instrument string assembly, the experiments would have been exposed to (minimally circulating) ambient atmosphere for less than a day, and then suspended for several hours inside the drill string and in contact with seawater during installation. The instruments were protected in sleeves of 7.62 cm outer diameter high-density polyethylene (HDPE) plastic and mounted on 1.25 cm diameter steel support rods through the center of the sleeves. Sleeve assemblies were connected end-to-end via galvanized steel shackles and subsequently to a weighted bar (sinker bar). Instrument strings were deployed within the low-alloy steel casing of the CORKs; at Hole 1301A, the instrument string hung within a section of slotted 11.76 cm diameter casing to allow exchange with the formation fluids in the open hole (Figure 1), whereas at Hole 1026B the instrument string resided in closed 17.78 cm diameter casing a few meters above the section exposed to basement fluids.

For recovery of the CORK instrument string, a latching tool connected to flotation was attached to the top of the instrument string at the CORK platform; the flotation and string were then recovered on deck by a specially-designed wench system using Plasma® lines. Once the instrument string was pulled free of the borehole, the string was pulled through the water column (~2660 m) at about 60 m min-1. The instrument string rested on the deck for less than one hour until the HDPE sleeve containing the microbial experiments were cut from the string and brought in the shipboard lab for processing. The experiments were removed from the string assembly and transferred to an ethanol- and flame-sterilized tray for deconstruction using sterilized tools. The screws holding the grids to the hanger were removed and the grids transferred wholesale into various fixative solutions in 50 ml sterile centrifuge tube. One grid of each material was transferred and processed as follows: 40 ml of cold, sterile 4% [w/v] paraformaldehyde in phosphate buffered saline (1xPBS, 10 mM sodium phosphate, 150 mM NaCl) for four hours, then rinsed for one hour in cold 1xPBS and subsequently stored at 4 oC in 1:1 1xPBS:ethanol; 4 oC in sterile 1:1 1xPBS:ethanol; 4 oC in sterile seawater (Sigma Aldrich) for 1 week, then frozen; directly frozen (-20 oC) without any treatment. Following grid removal, precipitates on the experiment hangers were scraped off using sterile tweezers and stored frozen in 1.5ml sterile centrifuge tubes.

Chips that had been preserved in either paraformaldehyde/PBS:ethanol or in PBS:ethanol were used for the epifluorescence and SEM microscopy analyses, with no noticeable difference between the two fixation methods (data not shown). Rock chip grids were transferred to a sterilized processing tray, and small (~4 mm2) fragments of incubated rock chips were carefully removed using cooled ethanol- and flame-sterilized chisels. Using sterilized tweezers, chip fragments were transferred into sterile 1-ml centrifuge tubes containing 0.5 ml of 1:5,000X diluted SYBR Green I stain (Invitrogen) and 300 M propidium iodide in sterile 1x TE buffer (10 mM Tris-HCl, 1 mM ethylenediaminetetraacetic acid (EDTA), pH 8.0). Chip fragments were incubated in the staining solution in the dark for 15 minutes; then carefully rinsed in a series of 1xPBS buffer, milliQ water, and then 80% ethanol before allowing the chip to dry briefly; the chip fragment was then mounted upside down in a drop of Citifluor immersion oil on a glass coverslip. Images shown in Figure 2A-C are compressions of 3-10 m vertical stacks taken at ~0.5 m intervals and were processed using the Leica LAF software. The density of stalk particles on chip fragments was measured by manually counting the percentage of squares on a field-of-view grid filled or partially filled with stalk particles (n=6 random fields-of-view of the same size for each sample type). Following epifluorescence viewing, the chips were then prepared for SEM by immersing in 80% ethanol to remove the Citifluor oil, then dehydrating in an ethanol series (5 min. in 80 % ethanol, 5 min. in 90 % ethanol, 5 min. in 100% ethanol) and stored in 100 % ethanol. Ethanol-dehydrated chip fragments were critical point dried, mounted on carbon conductive tape on aluminum stubs, sputter coated with gold, and analyzed with a JEOL JSM-7001 field emission SEM (operated with a 10 kV accelerating voltage) at the Center for Electron Microscope and Microanalysis at the University of Southern California.

The reason for the green fluorescent appearance of the stalk particles on the mineral chips (Figure 1) is unclear, as the fluorescent stains used in this study should only target DNA molecules; however, stalks were not visible on unstained chip fragments. Pipe dope, a lubricant applied generously between the steel joints in the CORK casing, was used during the installation of the CORK(Fisher et al., 2005), and has been observed on some recovered CORK instruments. Hence, pipe dope compounds were investigated as a potential substance that could explain the fluorescent properties of the stalks, as greases and oils tend to have strong autofluorescent properties(Orcutt et al., 2005). The pipe dope samples tested did have strong autofluorescence under blue and green light excitation. However, when the pipe dope compounds were mixed with a culture of the stalk forming Mariprofundus ferrooxydans and then stained with the same dyes, no fluorescence was observed on the stalks (data not shown). Thus, it is unclear if the fluorescence of the stalks on the Hole U1301A experimental chips is due to interaction of the fluorescent stains, particularly the SYBR Green I dye, with another substance coating the stalks. Fluorescence staining of stalk particles has also been observed in other iron oxide mat environments from the Loihi Seamount (E. Fleming, personal communication), and non-specific binding of SYBR Green I dye has been observed elsewhere (Morono et al., 2009).

For micro X-ray absorption analysis, a fragment of the pyrite/hematite chip frm the Hole 1301A subsurface experiment was dehydrated in 80% ethanol for 5 min., air dried, and then embedded in quick-set epoxy under vacuum to create a ‘thin’ section for microXAS analysis, similar in principle to methods described elsewhere (Toner et al., 2009). After the embedding resin hardened, the chip fragment and resin was cut perpendicular to the chip surface in 0.8 mm-thick sections on an ultraslow speed saw (Buehler) using a diamond wafering blade at 200 revolutions per minute with distilled water as cutting fluid. A cut section was mounted over the analytical window on an aluminum sample holder using aluminum tape and analyzed at the Advanced Light Sourceon the 10.3.2 beamline(Marcus et al., 2004).

X-ray fluorescence (XRF) elemental maps were collected at multiple energies to distinguish among elements having overlapping or interfering XRF signals. Specifically, maps were collected at PbL2+50 eV, PbL3-50 eV, and MnK+50 eV with a 6 × 4 m beam spot on a sample and a pixel size of 5 × 5 m using a 7-element germanium detector. XRF maps were deadtime corrected and registered, and individual channels of interest were selected to build a single composite map using custom beamline software. Regions within and below the surface of the chip were then analyzed for Fe- and Mn- X-ray absorption near-edge structure (XANES) spectroscopy in fluorescence and transmission modes simultaneously at the K-edge. The energy calibration of the monochromator was accomplished with a Mn foil (inflection point set to 6537.57 eV) and subsequently, an Fe foil (inflection point set to 7110.75 eV). Mn(III,IV) minerals are particularly prone to photoreduction during synchrotron radiation measurements. All spectra were monitored closely for any evidence of beam damage (changes in pre-edge and white line shape, amplitude, or energy position among individual scans), and none was observed for these Fe and Mn XANES measurements. In addition, photon doses to the sample were minimized as much as possible, and Mn and Fe measurements were never taken from the exact same location. All XANES scans were deadtime corrected (if fluorescence mode), energy calibrated, averaged, and compared to reference standard spectra(Bargar et al., 2000; Hansel et al., 2003; Marcus et al., 2008).

Original DNA extracts were purified using the QIAquick PCR purification kit (Qiagen) according to the manufacturer’s protocols and eluted in 30 l of EB buffer. To augment this analysis, whole genome amplification/multiple displacement amplification methods(Dean et al., 2002) were also used for phylogenetic analysis. Here, an aliquot of cleaned DNA extract (as described above) was used as the template in either an Illustra ™ GenomiPhi ™ V2 DNA Amplificiation Kit reaction or in a Qiagen REPLI-gTM Mini Kit reaction conducted according to manufacturer protocols. The bacterial 16S rRNA gene was amplified by polymerase chain reaction (PCR) in 50 l reactions using the primers 27F (5’ AGA GTT TGA TCC TGG CTC AG) and 1492R (5’ GGT TAC CTT GTT ACG ACT T; Lane, 1991) and the 5-Prime MasterTaq kit with the following reaction conditions: 10 l original DNA template (presumably 10-100 ng DNA based on spectroscopic absorption at 260 nm wavelength) or 5 l of GenomiPhi™ reaction as template, 95 oC denaturation for 4 min.; 28 cycles of 95 oC for 30 sec., 50 oC for 30 sec. for primer annealing, and 72 oC for 2 min. for extension; and a final extension at 72 oC for 10 min. The primer set Uni341F (5’ CCT AYG GGR BGC ASC AG)and Arch915R (5’ GTG CTC CCC CGC CAA TTC CT; Stahl & Amand, 1991) was used to amplify archaeal 16S rRNA genes, although bacterial 16S rRNA genes were also recovered with this set likely due to cross specificity of the Uni341F primer. PCR reactions for the Uni341F/Arch915R primer reactions were created using the Speedstar Taq (TaKaRa) kit in 25 l reactions with the following conditions: 2 l REPLI-g™ reaction as template, 94 oC denaturation for 2 min.; 31 cycles of 98 oC for 10 sec., 55 oC for 15 sec. for primer annealing, and 72 oC for 20 sec. for extension; and a final extension at 72 oC for 10 min. PCR product cleanup, cloning, sequencing, and alignment were performed as described elsewhere(Orcutt et al., 2010).

A negative control for DNA extraction, which was followed through all extraction, amplification, and cloning steps, produced two viable TOPO TA colonies with the 27f/1492r primer pair from the multiple agar plates streaked with this reaction; both were sequenced and had identical sequences that poorly matched (75% sequence similarity) with environmental sequences belong to the uncultivated OP1 candidate phyla. Any OP1-related sequences recovered from the Hole 1301A sample cloning (33 clones from the Hole 1301A basalt chip, 38 clones from the Hole 1301A pyrite chip, and 3 clones from the Hole 1301A washer) were thus removed from the dataset and are not discussed further, leaving 85 high-quality, robust sequences. The source of the OP1-related contaminant sequences is not known.

Thermodynamic modeling

To evaluate the thermodynamic affinity of various redox reactions within the observatory environment, water-rock path models (using Geochemist’s workbench, GWB; Bethke, 1996) were calculated, assuming heating from 2-65 oC and the interaction of 30 g of basalt reacting with 1 kg of seawater. The composition of the basalt was 3.0 g olivine (Fo70), 15.6 g plagioclase (An80), 11.4 g clinopyroxene (Wo50En40Fs10), and 0.05 g MnO, approximating typical mid-ocean ridge basaltic composition. Seawater composition was taken from published data near the Baby Bare outcrop and Hole 1301A(Wheat & Mottl, 2000). In the model, all redox couples (except H2/H+) and Fe2+/Fe3+ were suppressed. A GWB thermodynamic database was constructed for 250 bars and temperatures from 2-65oC using SUPCRT92(Johnson et al., 1992) with data from Wolery(2004) for minerals and the OBIGT database ( for solutes. To examine the potential energy yield of Fe2+ oxidation, the formation of all Fe(III) oxides was suppressed.

Supplemental References:

Bargar JR, Tebo BM, Villinski JE. (2000). In situ characterization of Mn(II) oxidation by spores of the marine Bacillus sp. Strain SG-1. Geochim. Cosmochim. Acta64: 2775-2778.

Bethke CM. (1996). Geochemical reaction modeling. Oxford University Press, New York, NY.

Dean FB, Hosono S, Fang L, Wu X, Faruqi AF, Bray-Ward P, Sun Z, Zong Q, Du J, Driscoll M, Song W, Kingsmore SF, Egholm M, Lasken RS. (2002). Comprehensive human genome amplification using multiple displacement amplification. Proc. Nat. Acad. Sci. USA, 99: 5261-5266.

Hansel CM, Benner SG, Neiss J, Dohnalkova A, Kukkadapu RK, Fendorf S. (2003). Secondary mineralization pathways induced by dissimilatory iron reduction of ferrihydrite under advective flow. Geochim. Cosmochim. Acta67: 2977-2992728.

Johnson JW, Oelkers EH, Helgeson HC. (1992). SUPCRT92: A software package for calculating the standard molal thermodynamic properties of minerals, gases, aqueous species, and reactions from 1 to 5000 bar and 0-1000oC. Computers Geosci.18: 899-947.

Marcus MA, MacDowell A, Celestre R, Manceau A, Miller T, Padmore HA, Sublett RE. (2004). Beamline 10.3.2 at ALS: a hard X-ray microprobe for environmental and material sciences. J. Synchrotron Rad.11: 239-247.

Marcus MA, Westphal AJ, Fakra S. (2008). Classification of Fe-bearing species from K-edge XANES data using two-parameter correlation plots. J Synchrotron Rad.15: 463-468.

Morono Y, Terada T, Masui N, Inagaki F. (2009). Discriminative detection and enumeration of microbial life in marine subsurface sediments. ISME J. 3: 503-511.

Orcutt BN, Boetius A, Elvert M, Samarkin V, Joye SB. (2005). Molecular biogeochemistry of sulfate reduction, methanogenesis and the anaerobic oxidation of methane at Gulf of Mexico cold seeps. Geochim. Cosmochim. Acta, 69: 4267-4281.

Orcutt B, Wheat CG, Edwards KJ. (2010). Subseafloor ocean crust microbial observatories: Development of FLOCS (FLow-through Osmo Colonization System) and evaluation of borehole construction materials. Geomicrobiol. J.27(2): 143-157

Stahl DA, Amann R. (1991) In: Nucleic Acid Techniques in Bacterial Systematics. Stackebrandt E, Goodfellow M, editors. Chichester, UK: John Wiley & Sons; 1991. pp 205-248.

Toner BM, Santelli CM, Marcus MA, Wirth R, Chan CS, McCollom TM, Bach W, Edwards KJ. (2009). Biogenic iron oxide formation at Mid-Ocean Ridge hydrothermal vents: Juan de Fuca Ridge. Geochim. Cosmochim. Acta73: 388-403.

Wolery TJ. (2004). Qualification of Thermodynamic Data for Geochemical Modeling of Mineral-Water Interactions in Dilute Systems. In: U.S.D.O. Energy (Editor). Bechtel SAIC Company, LLC.

Figure Captions for Supplemental Figures

Supplemental Figure 1.(a) Map of field area, eastern flank of the Juan de Fuca Ridge, showing location of IODP Expedition 301 sites. (b) Seafloor bathymetric contour map of area around IODP Expedition 301 sites (highlighted with yellow box in (a)) showing spatial relations between CORK observatories (filled circles) in Holes 1026B, 1027C, U1301A, and U1301B, and nearby basement outcrops (gold bathymetric contours). Depth contours in meters. (c) Schematic of casing and CORK system deployed in Hole U1301A, not drawn to scale. Primary CORK casing is 4-1/2 inches in diameter and is sealed with two plugs, one at depth and one at the top. Additional seals are provided by casing packers, but Hole U1301A CORK was not sealed between 10-3/4 inch and 16 inch casing strings, initially allowing cold seawater to enter the observatory at depth. Excess fluid pressure in basement eventually overcame the flow of cold water, and warm formation fluid flowed into the borehole and then up to the seafloor. Microbial colonization experiment was placed near the base of the inner CORK casing, above open hole.

Supplemental Figure 2. Comparison images of biogenic stalks formed on the Hole U1301A experiment to those formed at the seafloor by iron oxidizing bacteria. (a-c) SEM images of the surfaces of the Hole U1301A experiment samples, highlighting the secondary alteration features of the stalks and chip surfaces. (d) SEM image and (e) transmission electron micrograph images of biogenic, short-range order iron oxyhydroxide stalks formed on the surface of seafloor-reacted minerals(Edwards et al., 2003; Toner et al., 2009), and (f) a scanning transmission x-ray microscopy image of a biogenic stalk particle from an iron oxide mat from Loihi seamount (K. Edwards, unpublished data). Scale bars are 5 m for a-d; 500 nm for e; and 1 m for f. Panels d and e reprinted with permission from Elsevier(Edwards et al., 2003).

Supplemental Figure 3. Example field-emission SEM images of mineral precipitates from Hole U1301A (panels a and b) and Hole 1026B (panels c and d). (a) crystalline pyrite with amorphous Fe-Si oxides. (b) framboidal pyrite with Fe-Si-oxide and Fe-oxides. (c) metal (Fe, Zn) sulfides and Fe-Si oxides coating an aragonite prism. (d) metal (Fe, Zn) sulfides coating amorphous Fe-Si oxide (note grey box in center of image is an artifact of image processing). Scale bar lengths provide in micrometers.

Supplemental Figure 4. Phylogenetic tree of subset of bacterial 16S rRNA gene sequences (Proteobacteria, Bacteriodetes, OP8 and Verrucomicrobia phyla). from Hole U1301A colonized rock chips. Sequences from this investigation in red, sequences from Hole 1026B black rust(Nakagawa et al., 2006a) and fluid(Cowen et al., 2003) in blue; and isolates in italics.