Abstract: A total of 34 pelagic water samples from the upper-, mid-and lower-Oxygen Minimum Zones (OMZs) in the eastern Pacific Ocean were collected. Particles suspended therein were captured by 0.2 μm pore size filters to extract bulk genomic DNA for PCR amplification of the genes relevant to methane oxidation. The genes encoding particulate methane monooxygenase and its relative enzyme ammonia monooxygenase (β), pmoA and amoA(β), respectively, were amplified and analyzed. Five OMZ samples (from four sites) showed amplification of pmoA. This finding provide the bases for consideration that pelagic methane oxidation may occur in OMZs. Interestingly, one pmoA sequence was identical (based on amino acids) to pmoA of the methanotrophic endosymbiont.
INTRODUCTION
Oxygen Minimum Zones (OMZs) are the pelagic regions characterized by dissolved oxygen concentration as low as <0.5 mL L-1 (<22 μM, or <7.5% of saturation; Levin, 2003). Large-scale OMZs are found in the southeast Atlantic off West Africa, the northern Indian Ocean and the eastern Pacific Ocean (Helly and Levin, 2004) and hypoxic to anoxic conditions are maintained in the middle parts of OMZs, namely OMZ cores. Denitrification is a typical anaerobic biogeochemical process in OMZs (Codispoti et al., 2005; Deutsch et al., 2007) and releases nitrogen suboxide (N2O) that contributes to global warming along with carbon dioxide and methane (Houghton et al., 2001).
Occurrence of a methane pool has been reported in the eastern tropical North Pacific Ocean (Burke et al., 1983; Sansone et al., 2001; Sansone et al., 2004). The methane pool is a water column region characterized by >10 nM methane, centered at water depths of 250-650 m with >500 m thickness and >1000 km horizontal extension in the OMZ and contributes to methane release into atmosphere. Similar methane pool is proposed for the off-Chile OMZ in the eastern South Pacific Ocean (Sansone et al., 2001). Stable carbon isotope analysis suggested that the methane in OMZs are subject to aerobic microbial oxidation (Sansone et al., 2001; Sansone et al., 2004).
The microbial agents that oxidize methane to obtain metabolic energy are also known to assimilate part of methane as the carbon source via either the ribulose monophosphate pathway (type I) or serine cycle (type II) (Hanson and Hanson, 1996) and thus termed methanotrophs. The type I methanotrophs possess the enzymes, particulate methane monooxygenase (pMMO), while the type II methanotrophs possess both pMMO and soluble methane monooxygenase (sMMO). Active centers of pMMO and sMMO are encoded by the genes, pmoA and mmoX, respectively. Thus, pmoA and mmoX are often used as genetic markers to study diversity and distribution of methanotrophs in natural environments (Dumont and Murrell, 2005). The enzyme, ammonia monooxygenase (AMO), is also responsible for aerobic methane oxidation, despite lower activity than pMMO and sMMO (Bedard and Knowles, 1989). The gene amoA encodes the AMO active center, with two subclasses of amoA(β) and amoA(γ) according to their carriers β- and γ-Proteobacteria, respectively (Ward and O’Mullan, 2005).
Therefore, four genes, pmoA, mmoX, amoA(β) and amoA(γ), are available as genetic markers. However, these genetic markers have not been applied to detect and characterize the distribution and diversity of methane-oxidizing enzymes in OMZs. This study aims at qualitative cataloging of pmoA, mmoX, amoA(β) and amoA(γ) by analysis of the PCR-amplified clones and reports the occurrence and diversity of these genes in the OMZs of eastern Pacific Ocean, covering from eastern tropical North Pacific Ocean to off-Peru South Pacific Ocean.
MATERIALS AND METHODS
Collection of Water Samples and Suspended Particles
Water samples from OMZs in the eastern Pacific Ocean were collected at the
sites HY01 to HY10A and H15A (Table 1) during the Legs 3 and
4 of the KH-03 cruise (June to August 2003) by RV Hakuhomaru, University
of Tokyo (currently operated by Japan Agency for Marine-Earth Science and Technology).
The Dissolved Oxygen (DO) profiles (Fig. 1) were obtained
at each site by CTD equipped with a DO sensor. The sampled-water depth was determined
by DO, which is corresponding to approximately 50% of surface water and minima.
In this study, these samples were regarded as the representative water of OMZs.
Figure 2 shows the vertical section of dissolved oxygen (mL
L-1) along 95°W (8°N to 8°S, which correspond to the
site HY03-HY09). The site 15A showed little development of OMZ (Fig.
1) and was thus regarded as the negative control. Water samples were collected
with Niskin bottles and three- to five-liter aliquots were filtered through
Sterivex filters (pore size, 0.22 μm; Millipore Corp., Bedford, Massachusetts)
to capture DNA-containing particles suspended therein. The used Sterivex filters
were washed with 1.8 mL of STE buffer (20% sucrose, 50 mM EDTA, 50 mM Tris-HCl,
pH 8.0) and kept frozen at -20°C on board until DNA extraction in the onshore
laboratory, according to Somerville et al. (1989).
Extraction and Amplification of Bulk Genomic DNA
The sterivex filters were added with 1.8 mL of SET buffer and 62 μL
of lysozyme solution (5 mg mL-1 in TE buffer containing 10 mM Tris-HCl,
1 mM EDTA and 10 mM NaCl, pH 8.0) and incubated on ice for 15 min, to extract
bulk genomic DNAs. The primary lysates were further incubated with 16 μL
of 25% SDS at room temperature for 1 h. Finally the lysates were incubated with
50 μL of proteinase K (20 mg mL-1) at room temperature for 3
h.
The lysates were separately collected from the Sterivex filter cassettes using sterilized syringes and the residual lysates were collected by adding 1 mL each of SET buffer and then a collection with fresh syringes. The original and residual lysates from a Sterivex filter were mixed and used for DNA precipitation by the phenol-chlorofom-isopropanol method. The DNA precipitates were immediately used for amplification of the genetic marker pmoA, mmoX, amoA(β) and amoA(γ) sequences.
PCR Amplification of Marker Gene Sequences
The amplified bulk genomic DNAs were purified using the QIAquick PCR Purification
Kit (Qiagen Inc., Valencia, Calif.) and prepared at 70 ng μL-1
each, which were tested by PCR with the primers for bacterial 16S rRNA gene
(Lane, 1991).
| Table 1: | Sites of OMZ water sample collection, sample codes, physico-chemical parameters and detected genetic markers |
| Fig. 1: | Dissolved oxygen profile at each site. The black circles show the sampled-water depth |
Then the bulk DNAs were used to amplify about 500 bp each of the pmoA (full
length, 744 bp), mmoX (1584 bp), amoA(β) (831 bp) and amoA(γ)
(744 bp) sequences using ExTaq DNA polymerase (TaKaRa Bio Inc., Otsu, Japan)
with a TaKaRa Cycler PERSONAL TP240, according to the PCR primers and conditions
(Costello and Lidstrom, 1999; Miguez et al., 1997; Purkhold et al.,
2000; Rotthauwe et al., 1997) listed in Table 2.
| Table 2: | PCR primer sets and conditions for amplification of the genetic markers, pmoA, mmoX, amoA(β) and amoA(γ) sequences |
| (a), Costello and Lidstrom (1999) (b), Miguez et al. (1997) (c), Rotthauwe et al. (1997) and (d) Purkhold et al. (2000) | |
| Fig. 2: | Vertical section of dissolved oxygen (mL L-1) along 95°W |
Cloning, Sequencingand Molecular Analyses
The PCR products of the expected sizes were excised after agarose gel electrophoresis,
purified with the QIAquick PCR Purification Kit and cloned using the TOPO Cloning
Kit with One Shot TOP10 E. coli (Invitrogen Corp., Carlsbad, Calif.).
Twenty four transformants per sample, if PCR-positive, were randomly collected
and sequenced bi-directionally by the dideoxy method (Sanger et al.,
1977) on a 3730xl DNA Analyzer (Applied Biosystems, Foster City, Calif.). Retrieved
sequences were searched for homology based on both nucleotides and amino acids
by FASTA at the DNA Data Bank of Japan (DDBJ; www.ddbj.hig.ac.jp).
Sequences non-homologous to target genes were excluded from further analyses.
The sequences having >97% nucleotide similarities were grouped into an operational unit. The most equidistant sequence within a unit was chosen to represent the unit and the representative sequences were deposited to DDBJ under the accession numbers AB276025 to AB276029. Each representative was converted to amino acid sequences to construct a phylogenetic tree along with known closely related sequences using the MEGA3 program (Kumar et al., 2004).
The sequences were checked for chimeras by bisecting and drawing two sub-phylogenetic trees from the bisects of each sequence. The sequences that showed different topologies among the two sub-trees were regarded as chimeric and removed from analyses. Transmembrane-spanning regions and topology of the deduced pMMO and AMO proteins were estimated using the TMHMM tools (br.expasy.org/tools/).
RESULTS AND DISCUSSION
PCR Clone Libraries of Amplified pmoA and amoA(β) Sequences
A total of 90 pmoA and 45 amoA(β) PCR clones were obtained
from five and two samples, respectively, out of total 34 samples, were obtained
(Table 3), while mmoX and amoA(γ) sequences
were not amplified by PCR despite repeated trials with the standard (Table
2) and modified thermal cycles.
The chimera-checked clones having≥ 97% nucleotide sequence similarities were grouped into an operational pmoA or amoA(β) units, OPU and OAU, respectively, which represent the equidistant (least deviated) sequences of component clones. As a result, four OPUs (OPU1 to OPU4, in the order of clone numbers shown in Table 3) and one OAU (OAU1) were formed. The four OPUs showed nucleotide similarities of 66.0 to 94.6%.
The pmoA and amoA(β) sequences yielded putative transmembrane regions of the corresponding enzymes pMMO and AMO, respectively and the numbers and lengths of putative transmembrane regions were compared with known counterparts of Methylococcus capsulatus and Nitrosomonas europaea (Table 4). The OPU1 to OPU4 encoded 169 amino acids, in which four transmembrane regions were identified at the amino acid positions at 13-35, 40-57, 64-86 and 96-118 (OPU1) or 90-112 (OPU2 to OPU4; identical to that of M. capsulatus). Similarly, the OAU1 encoded 163 amino acids, in which three transmembrane regions were identified at the amino acid positions at 10-32, 37-59, 107-129, nearly identical to that of N. europaea.
Thus, the obtained OPU and OAU sequences were highly likely corresponding to pmoA and amoA(β), respectively and contributed to archiving of marine pmoA and amoA genes that have been underrated. The pmoA sequences in particular are the first from the vast oceanic water column that is the global largest habitat.
Phylogenetic Analysis of Amplified pmoA and amoA(β) Sequences
The FASTA nucleotide homology-search showed that the most closely related
at a 73.5% (nucleotides) similarity to the pmoA of the thermophilic methanotroph,
Methylocaldum szegediense OR2 (U89303; Bodrossy et al., 1997),
which was originally isolated from a Hungarian geothermal fluid and grows up
to 62°C.
It should be noted that the OPU2 is the most closely related at a 99.8% similarity
to the pmoA of the methanotrophic symbiont of the vent mussel Bathymodiolus
sp. (AB062137). The high score of boot strap values supports the phylogenetic
positioning of OPU2 (Fig. 3). The mussel specimens were collected
from a 1035 m deep hydrothermal vent in the mid-Okinawa Trough, western North
Pacific Ocean and known to harbor a unique endosymbiont, namely, type X methanotroph,
that possesses the genes encoding both methanotrophic and autotrophic enzymes,
pmoA and cbbL, simultaneously (Elsaied and Naganuma, 2001; Elsaied
et al., 2006).
| Table 3: | Distribution of retrieved pmoA and amoA(β) clones in seven OMZ water samples (from five sites) of the eastern Pacific Ocean. Sample codes are listed in Table 1, The number of representative sequence is given in parentheses |
| Table 4: | Numbers of amino acid residues, numbers of transmembrane (TM) regions and TM amino acid positions inferred from the pmoA and amoA(β) sequences of the OMZ clones (this study) and the selected species, Methylococcus capsulatus and Nitrosomonas europaea |
The OPU3 and OPU4 were most closely but only weakly related at 78.3 and 76.6% similarities, respectively, to the pmoA of Methylomicrobium sp. NI (AB253367), a marine methanotroph that notably possesses particulate and soluble methane monooxygenase genes simultaneously (Nakamura et al., unpublished).
As to the deduced amino acids, FASTA resulted in the closest homology of OPU1 at an 81.5% similarity to an environmental clone (Q75NB8) from the 650 m deep methane seep sediment in the western North Pacific Ocean (Inagaki et al., 2004); OPU2 at 100% identity to an environmental clone (Q8KZJ5) from hydrothermal chimney fragments of the TAG mound, 3655 m deep, Mid-Atlantic Ridge (Elsaied et al., 2004) and OPU3 and OPU4 both at a 93.9% similarity to an environmental clone (Q19PD7) from the 540 m deep hydrocarbon seep sediment in the Gulf of Mexico (Yan et al., 2006).
As mentioned above, the OPU2 is also showed the 100% identity (amino acids; 99.5% similarity based on nucleotides) to the pmoA environmental clone from the TAG hydrothermal mound, Mid-Atlantic Ridge (Q8KZJ5; Elasied et al., 2004). The shortest distance of about 15000 km lies between the OPU2-positive sites and the mid-Okinawa Trough; about 6000 km between the OPU2-positive sites and the TAG mound and about 14500 km between the mid-Okinawa Trough and the TAG mounds. It should be noted that the OPU2 was the second abundant operational pmoA unit (Table 3). Therefore, the OPU2 may be widely shared (via lateral gene transfer) among diverse methanotrophs or represent a cosmopolitan but low O2-adapted methanotroph in the global oceanic regimes.
The nucleotide sequence of the amoA(β) OAU1 was the most closely related at a 86.9% similarity to the amoA(β) of the environmental isolate bacterium amoA.12.V-frei.kultur (AY795821) from the inland saline soil of Hannoversches Wendl and Schreyahn, Germany (unpublished). The deduced amino acid sequence of the OAU1 was the most closely related at 99.3% identity to the environmental clone (Q66UX2) from the Plum Island Sound estuary sediment, Massachusetts (Bernhard et al., 2005). No marine amoA(β) counterparts were related to the OAU1 at >90% similarities.
Distribution of pmoA and amoA(β) in Eastern Pacific Ocean OMZs
Eastern Pacific OMZs are regarded to host methane pools (Sansone et al.,
2004), whatever the sources and extents are and thus it is implied that peripheries
(i.e., redox boundaries) of the OMZs may correspond to the sites of aerobic
methane oxidation. Methane oxidation in water column should require a redox
boundary in which both methane and molecular oxygen exist and such redox gradients
are likely formed in the OMZ peripheries or in association with OMZ-trapped
marine snows, for example. Five OMZ waters (0.09-0.14 mL L-1 O2)
yielded PCR amplification of pmoA sequences, which suggests that aerobic
methane oxidation occurs in these anoxic waters. In other words, distribution
of pmoA may be influenced by the occurrence of methane pools that are
formed in association with OMZs. The sites of two pmoA-positive waters,
HY02 and HY03, corresponded to the periphery of the eastern tropical North Pacific
methane pool observed in 1983 and 2001 (Burke et al., 1983; Sansone et
al., 2001) and the sites of HY08B and HY10A were possibly located within
the off-Peru methane pool (Sansone et al., 2001).
| Fig. 3: | The neighbor-joining phylogenetic tree based on the deduced pmoA and amoA amino acid sequences (102 amino acids) of the OMZ clones and cultured species. Bootstrap values (percentage of 1000 replications) greater than 50 are shown on or horizontally to nodes. Scale bar, 0.05 substitutions per site |
The observed inconsistence of pmoA distribution and DO concentration may be ascribed to: 1) time-lags between the formation/disappearance of methane-oxidizing redox conditions and corresponding microflora and/or 2) involvement of other parameters such as concentration of copper that is required for the expression of pMMO enzymatic activity (Hanson and Hanson, 1996).
In contrast, the equatorial sites HY04 to HY08, where OMZs (with <0.2 mL L-1 O2) were likely less developed (Fig. 2), yielded no amplification of pmoA. Equatorial water columns, particularly >250 m deep, have higher dissolved oxygen concentrations and less developed OMZs (Helly and Levin, 2004), due to migration of the O2-rich eastward-moving Equatorial Undercurrent from 150°E to off-Ecuador (Tsuchiya, 1968). The O2-rich water mass serves as capping and depressing the upper OMZ boundaries in the equatorial (5° N to 5° S, Fig. 2) eastern Pacific Ocean (Helly and Levin., 2004) and thus may have hindered the formation of m ethane-oxidizing conditions and thus occurrence of pmoA there.
The amoA sequence, an additional genetic marker for potential aerobic methane oxidation, was detected in two shallower hypoxic, not anoxic, water samples (HY08B-8Bh at 96 m deep with 1.93 mL L-1 O2 and HY09-9h at 83 m deep with 1.75 mL L-1 O2). This may suggest that the amoA-encoded enzyme, ammonia monooxygenase (AMO) catalyzes aerobic methane oxidation in these non-mid-OMZ waters with different enzymological featuers such as sensitivity to O2, substrate affinity Km and velocity Vmax, or that AMO catalyzes oxidation of ammonia rather than methane to yield nitrite. Part of the regenerated nitrate via ammonia oxidation might be recycled by nitrate reduction or denitrification that is active in the eastern Pacific Ocean (Zehr and Ward, 2002).
CONCLUSIONS
This study displayed the occurrence and diversity of pelagic pmoA and amoA(β), the genes coding for enzymes probably involved in pelagic methane oxidation in the OMZs of eastern Pacific Ocean. In future study, quantitative analysis using messenger RNA (mRNA) instead of DNA may allow us to determine whether these gene sequences are actually from active, predominant cells or not, although in situ or on-board detection of mRNA requires more expertise and sophisticated set-ups with thorough anti-contamination measures.
ACKNOWLEDGMENTS
We are obliged to the crew of the RV Hakuhomaru, University of Tokyo (currently operated by Japan Agency for Marine and Earth Agency), as well as the shipboard party of the Legs 3 and 4 of the KH-03 cruise, for assistance in sample/data collection. Part of this study was supported by Grants-in-Aid for Scientific Research (14340268) from the Japan Society for the Promotion of Science.