1
Supporting Online Material
2
Supplementary Methods
3
Sample preparation for DNA extraction
4
In 2008, approximately 2 L of fluids were pumped from the bag sampler through
5
a 25 mm-diameter, 0.1 µm pore-sized polyethersulfone membrane filter (Pall
6
Corporation, Port Washington, NY), which was subsequently stored in 0.5 ml of DNA
7
lysis buffer [20 mM Tris-HCl, 2 mM EDTA, 1.2% Triton X-100, 2% lysozyme (w/v), pH 8]
8
at -80°C until further processing. In 2009 and 2010, 2.6 L and 11.2 L of fluids,
9
respectively, were pumped from the bag sampler through a 0.22-µm-pore-sized
10
Sterivex-GP filter cartridge (Millipore Corporation, Billerica, MA), and subsequently
11
stored in 2.0 ml of DNA lysis buffer at -80°C until further processing.
12
Seawater samples were collected in the vicinity of borehole 1301A in order to
13
provide background controls for the borehole fluid samples (Table 1). In 2008, seawater
14
from within the nepheloid layer at 2644 m depth was collected in a 10 L Niskin bottle
15
using a conductivity-temperature-depth (CTD) guided rosette. A total of 2.6 L of fluids
16
were filtered through a 25 mm-diameter, 0.1 µm pore-sized polyethersulfone membrane
17
filter (Pall Corporation), and subsequently stored in 0.5 ml of DNA lysis buffer at -80°C
18
until further processing. Also via CTD rosette, 3.8 L of seawater were collected from a
19
depth of 2660 m in 2009, and 5.0 L of seawater were collected from a depth of 2648 m
20
in 2010. The fluids were filtered through 0.22-µm-pore-sized Sterivex-GP filter
21
cartridges (Millipore Corporation), filled with 2.0 ml of DNA lysis buffer, and stored at -
22
80°C until further processing. In 2010, a 5 L Niskin bottle fitted to the ROV Jason II was
23
used to collect 4.9 L of seawater from a depth of approximately 2660 m in the vicinity of
1
24
CORK 1301A (dive 505) and processed in the same manner as other fluid samples
25
collected in 2010.
26
27
28
DNA extraction
Membrane filters were thawed to room temperature prior to nucleic acid
29
extraction. For the 2008 borehole fluid sample, environmental DNA was extracted using
30
the PowerWater DNA isolation kit (MO BIO Laboratories, Carlsbad, CA) following the
31
manufacturer’s protocol. Environmental DNA was extracted from all other samples
32
(2008-2010 seawater and 2009-2010 borehole fluid) via the PowerSoil DNA isolation kit
33
(MO BIO Laboratories) following the manufacturer’s protocol. Quantification of resulting
34
genomic DNA was performed using a NanoDrop ND-1000 spectrophotometer (Thermo
35
Fisher Scientific, Waltham, MA).
36
37
38
SSU rRNA gene cloning and sequencing
Small subunit ribosomal RNA (SSU rRNA) gene fragments were amplified via the
39
polymerase chain reaction (PCR) using the universal oligonucleotide forward and
40
reverse primers 519F (5’-CAGCMGCCGCGGTAATWC-3’) and 1406R (5’-
41
ACGGGCGGTGTGTRC-3’), respectively. Each 20 µl PCR reaction contained 0.25 U of
42
PicoMaxx high fidelity DNA polymerase (Stratagene, La Jolla, CA), 1x PicoMaxx
43
reaction buffer, 200 µM of each of the four deoxynucleoside triphosphates (dNTPs), 200
44
nM of both forward and reverse primer, and ~3-4 ng of environmental DNA template.
45
PCR cycling conditions consisted of an initial denaturation step at 95°C for 4 minutes,
46
followed by 35 to 38 cycles of 95°C denaturation for 30 sec, 55°C annealing for 1 min,
2
47
72°C extension for 2 min, and a final extension step at 72°C for 20 min. For the 2008
48
borehole fluid sample, a 3-cycle reconditioning PCR was performed in order to help
49
eliminate heteroduplexes (Thompson et al., 2002). Amplification products of the
50
anticipated length were excised from an agarose gel and subsequently purified using
51
the QIAquick gel extraction kit (Qiagen, Valencia, CA). Products were cloned using
52
either the pGEM-T Easy kit (Promega, Madison, WI) or the TOPO TA Cloning kit
53
(Invitrogen, Carlsbad, CA) following the manufacturer’s instructions. Clones were
54
sequenced unidirectionally on an ABI 3730XL DNA Analyzer (Applied Biosystems,
55
Carlsbad, CA).
56
57
58
Phylogenetic analysis
DNA sequences were trimmed of vector sequence and manually curated using
59
Sequencher version 4.9 software (GeneCodes, Ann Arbor, MI) and subsequently
60
checked for chimera formation via Bellerophon (Huber et al., 2004) and
61
CHECK_CHIMERA, available from the Ribosomal Database Project (Cole et al., 2005).
62
Using the ARB software package (Ludwig et al., 2004), curated clone sequences were
63
aligned with version SSURef_106 of the SILVA ARB database (Pruesse et al., 2007)
64
modified to include short (<1200 nucleotides) environmental gene clones that were
65
highly similar to the clone sequences obtained in this study. Phylogenetic analyses were
66
performed with the RAxML maximum likelihood method using a gamma distributed rate
67
variation model for nucleotide substitution (Stamatakis, 2006). Sequences of short
68
length were added to the maximum likelihood-derived phylogenies using the parsimony
69
insertion tool in ARB. Bootstrap analyses were determined by RAxML (Stamatakis et
3
70
al., 2008) via the CIPRES Portal V 3.1 (Miller et al., 2010). All sequences generated in
71
this study have been deposited in GenBank under accession numbers JX194240 -
72
JX194738 and JX255729 - JX255730.
73
74
75
Microbial community analysis
The UniFrac significance test (Lozupone et al., 2005) was used to evaluate
76
whether the microbial community structure was different between the three sample
77
years. A maximum likelihood-based phylogenetic tree containing all borehole 1301A
78
fluid environmental sequences derived in this study was evaluated using an online
79
implementation of the hypothesis testing approach (Lozupone et al., 2006). Significant
80
differences in community structure within pairs of samples were evaluated using 100
81
permutations of both the weighted and unweighted implementations of the UniFrac
82
algorithm. All p-values were corrected for multiple comparisons using the Bonferroni
83
correction. Microbial community α-diversity estimators, rarefaction curves, and
84
community relatedness were generated or assessed using lane-masked (α-diversity
85
estimators, rarefaction curves) or unmasked (community relatedness) clone sequences
86
grouped into operational taxonomic units (OTUs) defined at 99% and 97% SSU rRNA
87
gene sequence similarity cut-off values using the nearest neighbor clustering method as
88
implemented by the mothur software package (Schloss et al., 2009). Microbial richness,
89
evenness, and diversity were assessed by the Chao1 richness estimator (S chao1) (Chao,
90
1984), Simpson evenness index (Esimpson) (Simpson, 1949), and the non-parametric
91
Shannon diversity index (Ĥshannon) (Shannon, 1948), respectively, as implemented in
92
mothur (Schloss et al., 2009).
4
93
94
Sample preparation for microscopy
95
In 2008, 40 ml seawater sub-samples were fixed with a final concentration of 3%
96
formaldehyde (0.2 µm-filtered) for 2 to 4 hours at 4°C, and subsequently filtered through
97
0.2 µm pore-sized polycarbonate membranes (Whatman, Maidstone, United Kingdom).
98
After air-drying, membranes were stored desiccated at -80°C until microscopic analysis.
99
In 2009, 40 ml seawater and borehole fluid sub-samples were fixed as above, but
100
subsequently stored at-80°C until thawed and filtered on 0.2 µm pore-sized
101
polycarbonate membranes (Whatman). Borehole and seawater sub-samples collected
102
in 2010 were fixed, filtered and stored as in 2008.
103
104
Probe design and optimization
105
Using version 106 of the SILVA ARB database (Pruesse et al., 2007) and the
106
‘probe design’ function of the ARB sequence analysis software (Ludwig et al., 2004),
107
taxon-specific oligonucleotide probes were designed to target borehole fluid
108
environmental gene clones within the Candidatus Desulforudis clade of the bacterial
109
phylum Firmicutes (1026B3_590; 5’-TTCACCCGCGACTTCTCGG-3’), and borehole
110
fluid clones related to the uncultivated gammaproteobacterial lineage JTB35
111
(1301A10_105_986; 5’-TTCTGATGATGTCAAGGGAT-3’) (Table S5). Specificity of the
112
probes was validated using the ‘probe match’ tool within the ARB software package.
113
Because no cultivated strains with SSU rRNA genes matching the probe target
114
sequences were available to serve as positive controls, hybridization conditions were
115
optimized using environmental samples: probe melting curves were constructed via
5
116
whole-cell hybridizations using a range of formamide concentrations (20-60%) on high-
117
integrity fluid samples collected from borehole 1301A in 2009 (probe 1026B3_590) and
118
2010 (probe 1301A10_105_986).
119
120
121
CARD-FISH
Catalyzed reporter deposition-fluorescence in situ hybridization (CARD-FISH)
122
was carried out according to the protocol of Grote et al., (2007), modified from Sekar et
123
al., (2003) and Pernthaler et al., (2002). For enumeration of cells of the domain
124
Bacteria, a mix of horseradish peroxidase (HRP)-labeled oligonucleotide probes
125
EUB338, EUB338-II, and EUB338-III were employed, while the HRP-labeled
126
oligonucleotide probe Arch_915 was used to enumerate Archaea (Table S5).
127
Hybridization specificity was evaluated using the HRP-labeled negative control probe
128
NonEUB (Table S5). All HRP-labeled probes were obtained from Biomers.net (Ulm,
129
Germany).
130
The 25 mm-diameter polycarbonate membrane filters were sectioned into one-
131
eighth pieces and embedded in agarose as described by Pernthaler et al., (2002). For
132
hybridization reactions employing the EUB338 bacterial probe suite and probe
133
1026B3_590, cell permeabilization was achieved by incubating filter sections in a
134
solution containing 10 mg ml-1 lysozyme, 0.05 M EDTA (pH 7.6), and 0.1 M Tris-HCl (pH
135
8.3) for 1 hr at 37°C. After washing once in ultrapure water, the filter sections were
136
incubated in an achromopeptidase solution [12 U achromopeptidase, 0.01 M NaCl, and
137
0.01 M Tris-HCl (pH 8.3)] for 30 min at 37°C, followed by a final ultrapure water wash.
138
Hybridization reactions employing HRP-labeled probe 1301A09_077_986 were only
6
139
treated with the lysozyme solution and subsequent ultrapure water wash, while those
140
employing HRP-labeled probe Arch_915 were permeabilized with a proteinase K
141
solution [>120 µAU ml-1 proteinase K, 0.05 M EDTA (pH 7.6), 0.1 M Tris-HCl (pH 8.3)]
142
and washed three times in ultrapure water. Following permeabilization, all filter sections
143
were incubated for 20 min in 0.01 M HCl at room temperature in order to inactivate
144
endogenous peroxidases, washed twice with ultrapure water, and rinsed briefly in 95%
145
ethanol. Hybridization reactions with probes Arch_915, 1026B3_590, and
146
1301A10_105_986 contained a final probe concentration of 90 µM in hybridization
147
buffer [900 mM NaCl, 20 mM Tris-HCl (pH 8.3), 20-55% formamide (Table S5), 0.01%
148
SDS, 0.1 g ml-1 dextran sulfate, 1% blocking reagent (Roche, Basle, Switzerland)], while
149
the EUB338 probe suite was used at a final concentration of 30 µM for each of the three
150
probes. Hybridization reactions were performed for ca. 12 hr at 35°C in the dark on a
151
rotary shaker. Following hybridization, the filter sections were blotted on tissue paper
152
and transferred into pre-warmed (37°C) wash buffer [20 mM Tris-HCl (pH 8.3), 5 mM
153
EDTA (pH 7.6), 0.01% sodium dodecyl sulfate, 13-74 mM NaCl (Table S5) ] for 15 min.
154
After removing excess wash buffer, filter sections were transferred to a substrate mix
155
containing 1 part cyanine 3 (Cy3) labeled tyramides to 50 parts amplification buffer
156
(TSA Plus Cyanine 3 System, Perkin Elmer, Waltham, MA) and incubated for 10 min at
157
room temperature in the dark. Filter sections were subsequently dabbed on blotting
158
paper to remove excess substrate mix and then washed once in phosphate-buffered
159
saline (PBS) solution (23 mM NaH2PO4, 100 mM Na2HPO4, 1.3 M NaCl, pH to 7.7) for
160
15 min at room temperature in the dark. The filter sections were subsequently washed
161
with ultrapure water, rinsed with 95% ethanol, and air-dried.
7
162
163
164
Fluorescence Microscopy
All filter sections were counterstained with a mounting solution [5.5 parts of
165
Citifluor (Citifluor Ltd., London, United Kingdom), 1 part of VectaShield (Vector
166
Laboratories, Burlingame, CA), and 0.5 parts of 1x PBS supplemented with 4',6-
167
diamidino-2-phenylindole (DAPI) to a final concentration of 1 µg ml-1] (Pernthaler et al.,
168
2002) and inspected with a Leica DM5000B epifluorescence microscope (Leica
169
Microsystems, Wetzlar, Germany) equipped with a 100x objective (HCX PL APO CS,
170
Leica Microsystems), filter sets appropriate for Cy3 and DAPI fluorescence, an ORCAII-
171
ER cooled CCD camera (Hamamatsu, Hamamatsu City, Japan) and IPLab imaging
172
software suite version 3.9.5 (Scanalytics, Rockville, MA). Between 500 and 1000 DAPI-
173
positive objects were counted for each sample. Probe-hybridized cells were enumerated
174
directly by overlaying Cy3 images onto corresponding DAPI images and identifying Cy3
175
coincident DAPI signals. Micrograph images were taken with a mounted camera setup
176
using a uniform exposure time of 50 ms for both DAPI and Cy3 images. The detection
177
limit of cell counts determined by DAPI staining was evaluated via enumeration of 40 ml
178
of 0.2 µm filtered ultrapure water (Kallmeyer et al., 2008). The detection limit of CARD-
179
FISH enumeration was determined by the hybridization of a subsurface sample with the
180
non-EUB oligonucleotide probe.
181
182
8
183
Supplementary References
184
Amann RI, Krumholz L, Stahl DA. (1990). Fluorescent-oligonucleotide probing of whole
185
cells for determinative, phylogenetic, and environmental studies in microbiology.
186
J Bacteriol 172: 762-770.
187
188
189
Chao A. (1984). Nonparametric estimation of the number of classes in a population.
Scand J Stat 11: 265-270.
Cole JR, Chai B, Farris RJ, Wang Q, Kulam SA, McGarrell DM et al. (2005). The
190
Ribosomal Database Project (RDP-II): sequences and tools for high-throughput
191
rRNA analysis. Nucleic Acids Res 33: D294-D296.
192
Daims H, Brühl A, Amann R, Schleifer KH, Wagner M (1999). The domain-specific
193
probe EUB338 is insufficient for the detection of all Bacteria: development and
194
evaluation of a more comprehensive probe set. Syst Appl Microbiol 22: 434-444.
195
Grote J, Labrenz M, Pfeiffer B, Jost G, Jürgens K. (2007). Quantitative distributions of
196
Epsilonproteobacteria and a Sulfurimonas subgroup in pelagic redoxclines of the
197
central Baltic Sea. Appl Environ Microbiol 73: 7155-7161.
198
199
200
201
202
203
Huber T, Faulkner G, Hugenholtz P. (2004). Bellerophon: a program to detect chimeric
sequences in multiple sequence alignments. Bioinformatics 20: 2317-2319.
Kallmeyer J, Smith DC, Spivack AJ, D'Hondt S. (2008). New cell extraction procedure
applied to deep subsurface sediments. Limnol Oceanogr-Meth 6: 236-245.
Lozupone C, Knight R. (2005). UniFrac: a new phylogenetic method for comparing
microbial communities. Appl Environ Microbiol 71: 8228-8235.
9
204
Lozupone C, Hamady M, Knight R. (2006). UniFrac - an online tool for comparing
205
microbial community diversity in a phylogenetic context. BMC Bioinformatics 7:
206
371.
207
208
Ludwig W, Strunk O, Westram R, Richter L, Meier H, Yadhukumar et al. (2004). ARB: a
software environment for sequence data. Nucleic Acids Res 32: 1363-1371.
209
Miller MA, Pfeiffer W, Schwartz T. (2010). Creating the CIPRES science gateway for
210
inference of large phylogenetic trees. Proceedings of the Gateway Computing
211
Environments Workshop (GCE): 14 November 2010; New Orleans, LA, pp 1-8.
212
Pernthaler A, Pernthaler J, Amann R. (2002). Fluorescence in situ hybridization and
213
catalyzed reporter deposition for the identification of marine bacteria. Appl
214
Environ Microbiol 68: 3094-3101.
215
Pruesse E, Quast C, Knittel K, Fuchs BM, Ludwig WG, Peplies J et al. (2007). SILVA: a
216
comprehensive online resource for quality checked and aligned ribosomal RNA
217
sequence data compatible with ARB. Nucleic Acids Res 35: 7188-7196.
218
Schloss PD, Westcott SL, Ryabin T, Hall JR, Hartmann M, Hollister EB et al. (2009).
219
Introducing mothur: open-source, platform-independent, community-supported
220
software for describing and comparing microbial communities. Appl Environ
221
Microbiol 75: 7537-7541.
222
Sekar R, Pernthaler A, Pernthaler J, Warnecke F, Posch T, Amann R. (2003). An
223
improved protocol for quantification of freshwater Actinobacteria by fluorescence
224
in situ hybridization. Appl Environ Microbiol 69: 2928-2935.
225
226
Shannon CE (1948). A mathematical theory of communication. Bell Syst Tech J 27:
379-423.
10
227
Simpson EH (1949). Measurement of diversity. Nature 163: 688.
228
Stahl DA, Amann R. (1991). Development and applications of nucleic acid probes in
229
bacterial systematics. In: Stackebrandt E, Goodfellow M (eds). Sequencing and
230
Hybridization Techniques in Bacterial Systematics. John Wiley and Sons:
231
Chichester, England. pp 205-248.
232
Stamatakis A. (2006). RAxML-VI-HPC: maximum likelihood-based phylogenetic
233
analyses with thousands of taxa and mixed models. Bioinformatics 22: 2688-
234
2690.
235
236
237
Stamatakis A, Hoover P, Rougemont J. (2008). A rapid bootstrap algorithm for the
RAxML web servers. Syst Biol 57: 758-771.
Thompson JR, Marcelino LA, Polz MF. (2002). Heteroduplexes in mixed-template
238
amplifications: formation, consequence and elimination by 'reconditioning PCR'.
239
Nucleic Acids Res 30: 2083-2088.
240
Wallner G, Amann R, Beisker W. (1993). Optimizing fluorescent in situ hybridization
241
with rRNA-targeted oligonucleotide probes for flow cytometric identification of
242
microorganisms. Cytometry 14: 136-143.
243
244
11
245
Supplementary Figure Legends
246
Figure S1. Rarefaction curves displaying the number of operational taxonomic units
247
(OTUs) observed given the sequencing effort in each year of sampling, clustered at
248
either 99% or 97% similarity cut-off values.
249
250
Figure S2. Venn diagrams showing microbial community relatedness between borehole
251
1301A fluid-derived and bottom seawater-derived SSU rRNA gene clones from each
252
individual borehole fluid sample and all seawater samples clustered at (A) 99% or (B)
253
97% similarity cut-off values and all borehole fluid years and all seawater samples
254
clustered at (C) 99% or (D) 97% similarity cut-off values.
255
256
Figure S3. Phylogenetic relationships of SSU rRNA gene clones related to the domain
257
Archaea, colored according to source (borehole 1301A fluids vs. rosette-derived
258
surrounding bottom seawater) and total number of years recovered (out of three
259
sampling years). Open circles indicate nodes with bootstrap support between 50-80%,
260
while closed circles indicate bootstrap support >80%, from 1000 replicates. Cultivated
261
Bacteria were used as outgroups (not shown). The scale bar corresponds to 0.10
262
substitutions per nucleotide position.
263
264
Figure S4. Phylogenetic relationships of SSU rRNA gene clones related to the
265
Gammaproteobacteria, colored according to source (borehole 1301A fluids vs. rosette-
266
derived surrounding bottom seawater) and total number of years recovered (out of three
267
sampling years). Open circles indicate nodes with bootstrap support between 50-80%,
12
268
while closed circles indicate bootstrap support >80%, from 1000 replicates. Cultivated
269
Betaproteobacteria were used as outgroups (not shown). The scale bar corresponds to
270
0.10 substitutions per nucleotide position. When applicable, groups are named
271
according to family; however, in some instances a genus name is provided. For clades
272
with no cultivated representatives, the name of the original gene clone representing the
273
lineage is used.
274
275
Figure S5. Phylogenetic relationships of borehole 1301A fluid-derived SSU rRNA gene
276
clones related to (A) the Sulfurimonas lineage of the Epsilonproteobacteria and (B) the
277
Thiomicrospira lineage of the Gammaproteobacteria . Open circles indicate nodes with
278
bootstrap support between 50-80%, while closed circles indicate bootstrap support
279
>80%, from 1000 replicates. Gene clones recovered in this study are highlighted in bold
280
font; the relative abundance of identical clones recovered from the same sampling year
281
is listed in parentheses. Cultivated Epsilonproteobacteria (A) and Thioalkalimicrobium
282
(B) were used as outgroups (not shown). The scale bars correspond to 0.1 substitutions
283
per nucleotide position. Relevant gene clones of short length were added after tree
284
construction and bootstrapping, and are indicated by dashed lines.
285
286
Figure S6. Representative photomicrographs of Bacteria (A), Archaea (B), 1026B3
287
lineage of the Firmicutes (C) and 1301A10_105 lineage of Gammaproteobacteria (D) in
288
fluids from borehole 1301A illuminated via CARD-FISH. Images are false-colored dual
289
image overlays of DNA-containing cells stained with DAPI (blue) and probe-conferred
290
Cy3 fluorescence (red). Panels E-G are DAPI-stained images corresponding to typical
13
291
oligonucleotide probe-positive cells of Archaea (E), 1026B3 lineage of the Firmicutes (F)
292
and 1301A10_105 lineage of Gammaproteobacteria (G). Scale bar, 1 µm.
14