Extremophiles

, 12:605

Comparative analysis of archaeal 16S rRNA and amoA genes to estimate the abundance and diversity of ammonia-oxidizing archaea in marine sediments

Authors

  • Soo-Je Park
    • Department of MicrobiologyChungbuk National University
  • Byoung-Joon Park
    • Department of MicrobiologyChungbuk National University
    • Department of MicrobiologyChungbuk National University
Original Paper

DOI: 10.1007/s00792-008-0165-7

Cite this article as:
Park, S., Park, B. & Rhee, S. Extremophiles (2008) 12: 605. doi:10.1007/s00792-008-0165-7

Abstract

Considering their abundance and broad distribution, non-extremophilic Crenarchaeota are likely to play important roles in global organic and inorganic matter cycles. The diversity and abundance of archaeal 16S rRNA and putative ammonia monooxygenase α-subunit (amoA) genes were comparatively analyzed to study genetic potential for nitrification of ammonia-oxidizing archaea (AOA) in the surface layers (0–1 cm) of four marine sediments of the East Sea, Korea. After analysis of a 16S rRNA gene clone library, we found various archaeal groups that include the crenarchaeotal group (CG) I.1a (54.8%) and CG I.1b (5.8%), both of which are known to harbor ammonia oxidizers. Notably, the 16S rRNA gene of CG I.1b has only previously been observed in terrestrial environments. The 16S rRNA gene sequence data revealed a distinct difference in archaeal community among sites of marine sediments. Most of the obtained amoA sequences were not closely related to those of the clones retrieved from estuarine sediments and marine water columns. Furthermore, clades of unique amoA sequences were likely to cluster according to sampling sites. Using real-time PCR, quantitative analysis of amoA copy numbers showed that the copy numbers of archaeal amoA ranged from 1.1 × 107 to 4.9 × 107 per gram of sediment and were more numerous than those of bacterial amoA, with ratios ranging from 11 to 28. In conclusion, diverse CG I.1a and CG I.1b AOA inhabit surface layers of marine sediments and AOA, and especially, CG I.1a are more numerous than other ammonia-oxidizing bacteria.

Keywords

Ammonia-oxidizing archaeaamoA16S rRNAMarine sediment

Introduction

Cultivation-independent analysis of diversity using the PCR-amplification and sequencing of 16S rRNA gene is a useful method for identifying novel bacterial and archaeal phylogenetic lineages from different environments. The first detection of crenarchaeotal 16S rRNA genes in ocean environments was surprising (DeLong 1992; Fuhrman et al. 1992) because they had previously only been cultivated from thermal and/or acidic environments. Thereafter, members of the phylum Crenarchaeota have been detected in most moderate and cold environments such as forest and agricultural soils (Bintrim et al. 1997; Buckley et al. 1998; Jurgens et al. 1997; Jurgens and Sanno 1999; Kudo et al. 1997; Reed et al. 2002; Sandaa et al. 1999; Ueda et al. 1995), freshwater sediments (Abreu et al. 2001; Hershberger et al. 1996; Jurgens et al. 2000; MacGregor et al. 1997; Schleper et al. 1997), marine water columns and sediments (Inagaki et al. 2003; Karner et al. 2001; Li et al. 1999; Massana et al. 1997; Vetriani et al. 1999), subsurface environments (Takai et al. 2001), and cold springs (Koch et al. 2006). When the molecular phylogeny of the 16S rRNA gene was analyzed, members of the Crenarchaeota phylum could be classified into several groups (Schleper et al. 2005): I.1a (marine, freshwater, subsurface), I.1b (soil, freshwater, subsurface), I.1c (forest soil, freshwater), I.2 (marine/freshwater sediment), I.3a (marine, hotspring, freshwater), I.3b (freshwater, wastewater, soil), the South Africa Gold Mine Crenarchaeotal Group (SAGMCG), and the Marine Hydrothermal Vent Group (MHVG). In addition to these groups, members of the Marine Benthic Groups (MBG) A, B and C, and the Miscellaneous Crenarchaeotal Group (MCG) were widely reported to be present in marine sediments (Inagaki et al. 2003, 2006).

Considering their abundance and broad distribution, it is likely that members of the Crenarchaeota phylum have important ecological roles in biogeochemical cycles. The results of metagenomic studies have revealed the existence of ammonia monooxygenase α-subunit (amoA) genes in members of the soil Crenarchaeotal Group I.1b (CG I.1b) (Treusch et al. 2005) and Marine Group I (MG I) of Crenarchaeotal Group I.1a (CG I.1a) subgroup (Hallam et al. 2006; Venter et al. 2004). Recently, an ammonia-oxidizing archaeon of MG I has been isolated and verified to contain the amoA gene (Konneke et al. 2005). Furthermore, a variety of archaeal amoA-like genes from marine environments have been isolated using PCR (Francis et al. 2005). Increased expression of archaeal amoA occurs after spiking soils with ammonia, and such results indicate that these amoA genes are actively involved in ammonia oxidation (Treusch et al. 2005). Furthermore, dominance of ammonia-oxidizing archaea (AOA) over ammonia-oxidizing bacteria (AOB) in a variety of soil types (Leininger et al. 2006) and North Sea water has been reported (Wuchter et al. 2006).

Microorganisms in marine sediments contribute significantly to global cycles of organic and inorganic matters because of their abundance (Whitman et al. 1998). Although MG I is the only major group of Crenarchaeota that is found below the photic zone of the ocean water column (Karner et al. 2001), a large variety of other crenarchaeotal groups have been found in marine sediments (Hershberger et al. 1996; Inagaki et al. 2006, 2003). In this study, we investigated AOA diversity and abundance by conducting a comparative analysis of archaeal 16S rRNA and amoA genes to assess the genetic potential for nitrification by Crenarchaeota from marine sediments. The results of this molecular survey showed that AOA include various members of CG I.1b as well as CG I.1a, all of which inhabit the surface layers of marine sediments. The results of a quantitative analysis of amoA copy numbers using real-time PCR suggested that AOA are numerically dominant over AOB in marine sediments.

Materials and methods

Site descriptions and sample collection

Marine sediments were collected from four different sites of the East Sea, Korea during July and August, 2005 using a core sampler. The sampling locations and water depth were as follows: ST3 (E 128° 35′, N 38° 20′; 650 m), JD (E 129° 11′, N 37° 46′; 500 m), UJ (E 129° 49′, N 37° 04′; 600 m), and DD (E 130° 56′, N 37° 28′; 500 m). The core samples were sectioned and each section was transferred to sterile plastic tubes and then stored at −80°C until analysis. The properties of the bottom water at the four sampling sites were similar and described as follows: temperature, 0.5–0.6°C; dissolved oxygen, 6.7–6.8 mg l−1; salinity, 34‰, chlorophyll, 0.2–0.02 mg l−1; nitrate, 1.5–1.6 mg l−1; ammonia 0.052–0.067 mg l−1; phosphate, 0.02 mg l−1.

Extraction of genomic DNA and cloning of archaeal 16S rRNA sequences

A Power Soil DNA kit (Mo Bio Laboratories, Solana Beach, CA, USA) was used to extract genomic DNA from the frozen sediment samples. After three independent extractions of DNA from the surface layers (0–1 cm) of each sample, the DNA samples from each site were pooled for molecular analysis. The genomic DNA concentrations were determined by spectrophotometry (Nanodrop Technologies, Rockland, DE, USA). The 16S rRNA gene sequences were amplified by PCR using EF-Taq DNA polymerase (Solgent, Korea) and the Arch-20F and Arch-958R primers (DeLong 1992) (Table 1) to generate a product for the preparation of clone libraries. The conditions of the PCR were: 5 min at 94°C; 30 cycles of 30 s at 94°C, 30 s at 55°C, 45 s at 72°C; 7 min at 72°C. Duplicate PCR products were pooled and purified using the Gel&PCR Purification System Kit® (Solgent, Korea), then ligated using the T&A Cloning Vector Kit (Real Biotech Corporation, Taiwan), and thereafter transformed into Escherichia coli DH5α cells, according to the manufacturer’s instructions. Putative positive clones were then transferred to a 96-well plate that contained Luria broth supplemented with ampicillin (100 μg/ml). The clones were grown overnight at 37°C and then stored at −70°C before screening.
Table 1

Primers used for PCR amplification of archaeal 16S rRNA gene sequences, and archaeal and bacterial amoA gene sequences

Primer name

Target

Positiona

Sequence (5′-3′)

Reference

Archaeal 16S rRNA gene

Arch-20F

Archaeal 16S rRNA

2–20

TTCCGGTTGATCCYGCCRG

Delong (1992)

Arch-958R

958–976

TCCGGCGTTGAMTCCAATT

CG I.1b-270F

CG I.1b 16S rRNA

265–284

TGGATTGGACTGCGKCCGAT

This study

CG I.1b-750R

772–791

GTCGAGCGCRTTCTGGMAAG

DGGE 340F

Archaeal 16S rRNA

340–357

GC-clampb-CCCTACGGGGYGCASCAG

Ovreas et al. (1997)

DGGE 519R

519–533

TTACCGCGGCKGCTG

amoA gene

A

Arch-amoA26F

Archaeal amoA

91–116

GACTACATMTTCTAYACWGAYTGGGC

This study

Arch-amoA417R

481–506

GGKGTCATRTATGGWGGYAAYGTTGG

B

CG I.1b-amoAF

CG I.1b amoA

43–65

ATAGTTGTAGTTGCTGTAAATAG

This study

CG I.1b-amoAR

439–459

CTCTAGAGGGTCTCTGACCAG

amoA1F

Bacterial amoA

332–349

GGGGTTTCTACTGGTGGT

Rotthauwe et al. (1997) (real-time PCR)

amoA2R

802–822

CCCCTCKGSAAAGCCTTCTTC

Arch-amoAF

Archaeal amoA

211–231

GCTCTAATTATGACAGTATAC

This study (real-time PCR)

Arch-amoAR

393–416

AYCATGTTGAAYAATGGTAATGAC

CG I.1b-amoAF

CG I.1b amoA

53–74

GTACATTATTGACAATCAACGC

This study (real-time PCR)

CG I.1b-amoAR

248–268

ATCCTARYGCAAACCAAGCTC

aNumbering is based on the 16S rRNA gene of E. coli, archaeal amoA of metagenome clone 54d9 (AJ627422), and bacterial amoA of Nitrosomonas europaea (L08050), respectively

bGC-clamp sequence: 5′-CGCCCGCCGCGCGCGGCGGGCGGGGCGGGGGCACGGGGGG-3′

PCR-denaturing gradient gel electrophoresis (DGGE) of CG I.1b-specific 16S rRNA genes

To analyze the diversity of CG I.1b, the 16S rRNA gene sequence was specifically amplified and analyzed using PCR-DGGE. Genomic DNAs from surface layers of each sample site were amplified by PCR. We used a nested PCR approach for the specific amplification of CG I.1b-specific 16S rRNA sequences using a CG I.1b-specific primer set (Table 1). The amplified PCR products were used as templates for a second round of amplification using a DGGE universal primer set (DGGE 340F and DGGE 519R) for archaeal 16S rRNA genes (Ovreas et al. 1997). The conditions for the first round of PCR amplification were: 5 min at 95°C; then 30 cycles of 30 s at 95°C, 30 s at 60°C, and 30 s at 72°C. The conditions for the second round of PCR amplification were: 5 min at 95°C; then 20 cycles of 30 s at 95°C, 30 s at 55°C, and 30 s at 72°C. Final PCR products were purified using a PCR Purification Kit (Solgent, Korea) and then used for the DGGE analysis. DGGE was performed with the D-code system (Bio-Rad Laboratories, Hercules, CA, USA). The polyacrylamide concentration and denaturant gradient for the 16S rRNA sequence were optimized as 8% polyacrylamide (37.5:1 = acrylamide:bisacrylamide) and 30–60% denaturant. Gels were run for 14 h at 80 V in 1× Tris-Acetate-EDTA buffer at a constant temperature of 60°C. The gels were stained for 30 min in 1× TAE buffer containing a 1:10,000 dilution of the dye, SYBR green (Bioneer, Korea). All visible bands were recovered from the gel and re-amplified using the DGGE 340F, without GC-clamp, and DGGE 519R primers. The purified PCR products were then sequenced using the amplification primers.

Amplification and cloning of archaeal amoA gene sequences

Archaeal amoA gene sequences were amplified from genomic DNA by PCR using EF-Taq DNA polymerase (Solgent, Korea) and amoA primers (Table 1). Two different primer sets were used to amplify the various archaeal amoA sequences. PCR primer set A was designed on the basis of alignments of all archaeal amoA sequences from Sargasso Sea water column (AACY01007942, AACY01575171, AACY01435967, and AACY01075168), various marine water column and sediments (Francis et al. 2005) and the soil metagenomic clone 54d9 (AJ627422). The design of PCR primer set B (Table 1), which is specific for the detection of soil (CG I.1b-specific) archaeal amoA, was based on the amoA sequences from the soil metagenomic clone 54d9 (AJ627422) and Oak Ridge soil clones (Francis et al. 2005).

The protocols for detection of various amoA gene sequences using these two primer sets were as follows: primer set A (5 min at 95°C; then 35 cycles consisting of 30 s at 95°C, 30 s at 60°C, 45 s at 72°C; and 7 min at 72°C); primer set B (5 min at 95°C; then 35 cycles consisting of 30 s at 95°C, 30 s at 55°C, 45 s at 72°C; and 7 min at 72°C). Aliquots from the PCR products were analyzed on a 1.5% agarose gel. Duplicate PCR products were pooled and purified using a PCR Purification Kit (Solgent, Korea), then ligated into the T&A Cloning Vector (Real Biotech Corporation, Taiwan), and thereafter transformed into E. coli DH5α cells, according to the manufacturer’s instructions. Putative positive clones were transferred to a 96-well plate that contained Luria broth supplemented with ampicillin (100 μg/ml). The clones were grown overnight at 37°C and then stored at −70°C before screening.

Sequencing and phylogeny analysis

Library clones were screened directly by PCR for the presence of inserts using two M13 universal primers, M13F (5′-GTTTCCCAGTCACGAC-3′) and M13R (5′-TCACACAGGAAACAGCTATGAC-3′). The conditions for PCR were: 5 min at 94°C; 30 cycles of 30 s at 94°C, 30 s at 55°C, 30 s at 72°C; 5 min at 72°C. The positive clones from each library were randomly selected and the PCR products were purified using a PCR Purification Kit (Solgent, Korea).

PCR products were sequenced directly using ABI PRISM® BigDye™ Terminator Cycle Sequencing Ready Reaction kits (Applied Biosystems, Foster City, CA, USA) and an ABI PRISM 3730xl DNA Analyzer (Applied Biosystems). The sequences were checked for possible chimeras using the CHIMERA_CHECK program on the Ribosomal Database Project web site (http://rdp8.cme.msu.edu) and subsequently submitted to a BLAST analysis (http://www.ncbi.nlm.nih.gov) to identify related 16S rRNA and amoA genes. The 16S rRNA and amoA sequences of related taxa were obtained from the GenBank database, and multiple alignments were performed using the Clustal X program (Thompson et al. 1997). The evolutionary distances were calculated using the Kimura two-parameter model (Kimura 1983), while the phylogenetic trees were constructed by using the neighbor-joining routine (Saitou and Nei 1987) in the MEGA 3 computer software program (Kumar et al. 2004), with bootstrap values based on 1,000 replications (Felsenstein 1985).

Rarefaction analysis and estimation of microbial diversity

The estimation of species richness, diversity indices, and rarefaction curves of archaeal 16S rRNA and amoA genes in each library were determined using the ESTIMATES software program (http://viceroy.eeb.uconn.edu/EstimateS). The species rarefaction curve of the entire data set from each group was computed using the individual-based Coleman method with the PAST software program (http://folk.uio.no/ohammer/past). The bias-corrected Chao1 estimator of species richness was calculated after 1,000 randomizations of sampling, without replacement. The percentage of coverage was calculated by Good’s method using the formula [1 − (n/N)] × 100, where n is the number of phylotypes in a sample represented by one clone (singletons) and N is the total number of sequences in that sample (Good 1953). The diversity of the sampled sequence set was estimated using the Simpson and Shannon indices within the ESTIMATES application. The Shannon index of evenness was calculated using the formula E = eD/N, where D is the Shannon diversity index. Operational taxonomic units (OTUs) of the 16S rRNA and amoA genes were defined as sequence groups in which sequences differed by 3 and 5%, respectively (Beman and Francis 2006; Hughes et al. 2001). Rarefaction Curves were plotted using Sigmaplot (V7.101, SPSS).

Quantitative real-time PCR of amoA and 16S rRNA gene

All real-time PCR experiments were done using an Exicycler thermocycler (Bioneer, Korea) and built-in Exicycler analysis software version 3.0 (Bioneer, Korea). The general conditions for real-time PCR were: 25 μl Greenstar PCR Master Mix (Bioneer, Korea), 0.5 μM of each primer (final concentration), 1 μl of template (20 ng of genomic DNA) and water added to a final volume of 50 μl. All reactions were performed in 8-strip PCR tubes with optical sealing tape (Bioneer, Korea). Specificity for real-time PCR reactions was tested by electrophoresis through a 1.5% agarose gel and melting curve analyses. Copy numbers of amoA were determined using external standard curves. An external standard curve that describes the relationship between archaeal and bacterial amoA copy numbers and cycle threshold (CT) values was generated using serial dilutions of a known copy number of the archaeal and bacterial amoA genes of the plasmid DNA: archaeal amoA, EF534487; CG I.1b-specific amoA, EF534411; and bacterial amoA, EF617305. We calculated the copy numbers of archaeal and bacterial amoA directly from the concentration of extracted plasmid DNA which was determined by spectrophotometry (Nanodrop Technologies, Rockland, DE, USA). The identical primers used for library construction of archaeal amoA and CG I.1b were used (Table 1). Real-time PCR detection of bacterial amoA was carried out using a previously described primer set amoA1F and amoA2R (Rotthauwe et al. 1997). For the amplification of all amoA sequences, the following thermal cycling parameters were used: 15 min at 95°C; 40 cycles of 30 s at 95°C, 30 s at 55°C, and 30 s at 72°C, readings were taken between each cycle. Melting curve analysis was performed from 55 to 95°C with a reading made every 1°C and the samples held for 1 s between readings.

For real-time PCR of 16S rRNA genes of CG I.1a and I.1b, primers specific for the CG I.1a (1a-F CTCAAGTGGTCAGGATG and 1a-R GCTTTCRTCCCTCACCGT) and I.1b (1b-F CCCGAGTGGTCGGGGCG and 1b-R GCTTTCRTCCCTCACCGT) were designed using the sequences obtained in this study. Primers were checked for specificity using the primer check program from the RDP II project (http://rdp8.cme.msu.edu). An external standard curve that describes the relationship between 16S rRNA genes of CG I.1a and I.1b copy numbers and cycle threshold (CT) values was generated using serial dilutions of a known copy number of the 16S rRNA genes of those of the plasmid DNA: CG I.1a, DQ831586; CG I.1b, EU332111. For the amplification of 16S rRNA genes, we used the same thermal cycling parameters which were used in the amplification of amoA sequences.

Nucleotide sequence accession numbers

All partial 16S rRNA and amoA gene sequences that were determined in this study were deposited in the GenBank database under the accession numbers: DQ831583–DQ831601, DQ831636–DQ831655, DQ866045–DQ866048, EF534379–EF591472, EF617305, and EU332040–EU332138.

Results and discussion

Diversity of archaeal 16S rRNA genes in marine sediments

In ocean water columns, below >150 m (euphotic zone), archaeal abundance increased to 20–40% of total microbial counts, and most Archaea belonged to the MG I subgroup (Karner et al. 2001), which might harbor amoA and comprise autotrophic nitrifiers. In contrast, a greater diversity of Crenarchaeota, which includes members of the MG I, MCG and MBG A, B and C has been reported in samples of marine sediments than that in marine water columns (Inagaki et al. 2003, 2006; Vetriani et al. 1999). To analyze the diversity and distribution of marine sediment AOA candidates, we constructed 16S rRNA gene libraries from the genomic DNA of the surface layer of marine sediments. For this purpose, approximately 160 clones were randomly selected and sequenced from each of the ST3, UJ, JD and DD libraries. Phylogenetic analysis of the clone sequences based on approximately 900 bp showed that the majority of clones from the surface layer (54.8%) belonged to CG I.1a (MG I) (Fig. 1), and had high sequence similarity (≥98%) to each other. Clones of CG I.1 comprise 30–97% of total clones. Interestingly, we obtained CG I.1b-related clones (5.8%) from the surface layer of the sediment of ST3 and DD sampling sites (Fig. 1), although these had previously been found exclusively in terrestrial or freshwater environments. We found that 16S rRNA sequence similarity was >96% in our CG I.1b-related clone sequences. Furthermore, the 16S rRNA sequence similarity was >94% when compared with clones previously detected in calcerous grassland soils (Ochsenreiter et al. 2003; Treusch et al. 2005), glacier foreland soils (Nicol et al. 2005), the rhizosphere (Nicol et al. 2003), and agricultural field soils (Bintrim et al. 1997; Sliwinski and Goodman 2004). The results of our analysis on archaeal 16S rRNA gene diversity revealed that marine sediments and terrestrial environments share certain crenarchaeotal phylotypes. For example, fluorescence in situ hybridization and 16S rRNA gene library analysis showed that MG I-related Crenarchaeota were detected in a terrestrial sulfidic spring in Germany (Koch et al. 2006).
https://static-content.springer.com/image/art%3A10.1007%2Fs00792-008-0165-7/MediaObjects/792_2008_165_Fig1_HTML.gif
Fig. 1

Diversity and abundance of archaea from marine sediments. a Relative abundance of archaeal groups of marine sediments is shown in the bar chart. b Tree showing phylogenetic positions of archaeal 16S rRNA sequences obtained from clone libraries of marine sediments of the East Sea. Phylotype names refer to the surface layer (0–1 cm) of the sediments, ES2 (ST3), EU1 (UJ), EJ1 (JD), and ED1 (DD). Numbers in parentheses following clone names indicate the number of times the sequences were found in the clone libraries from the four sampling sites ST3, UJ, JD, and DD, respectively. Clones from this study are indicated in boldface. The reference sequences were chosen to show the diversity of the sequences and to indicate the closest relatives to the sequences found in our study. The scale bar represents five substitutions per 100 nucleotide positions. Significant bootstrap values (≥50%; 1,000 replicates) are indicated by filled circles at branch points

We also detected other crenarchaeotal groups which were affiliated with the MCG (8.4%), MBGB and MBGC groups (20.0 and 1.9%, respectively) at all the sampling sites except ST3. In addition, groups of Euryarchaeota, such as VAL III (1.9%), Terrestrial Miscellaneous Euryarchaeotal Group (TMEG, 2.6%), and MBG Group D (4.5%), were also found in the surface layers of the JD and DD sampling sites. The results of the BLAST analysis indicated that those groups that were present in the surface layers at the JD and DD sampling sites were related to the groups that have been found in anoxic marine sediments, such as subsurface cold seep environments (Knittel et al. 2005) and methane-hydrate-bearing deep sediments (Inagaki et al. 2006; Inagaki et al. 2003). Our estimations of species coverage, richness, evenness, diversity (Table 2) and analysis of rarefaction (sFig. 1) indicate that archaeal diversities at the JD and DD sampling sites were relatively higher than those at the other two sites. The high archaeal diversity at the JD sampling site might be caused by compression of the oxic and anoxic zones into a short depth profile which results in increased niche diversity in 0–1 cm surface sediments.
Table 2

Estimation of sequence diversity and phylotype coverage of 16S rRNA gene libraries

 

Archaeal 16S rRNA genea

ST3

UJ

JD

DD

No. of clones

23

39

60

36

Phylotypes

2

4

18

13

Singletons

0

1

8

3

Chao1 estimated richness

2

4

22.67

13.5

Chao1 standard deviation

0

0.25

4.49

1.02

Shannon’s index for diversity

0.57

0.81

2.31

2.37

Simpson’s index for diversity

1.67

1.82

7.02

11.45

Evenness

0.89

0.56

0.56

0.82

Fisher’s diversity

0.53

1.12

8.72

7.30

Good’s estimator of coverage, %

100

97.4

86.7

91.7

Diversity indices and richness estimators were calculated using the ESTIMATES software program

aDiversity was estimated using Operational Taxonomic Units and defined as groups with ≥97% sequence similarity

Diversity of CG I.1b in marine sediments

To further explore the diversity of CG I.1b in marine sediments, we analyzed the CG I.1b 16S rRNA gene using PCR-DGGE. We amplified 16S rRNA sequences from the sediment samples from the four sites using newly designed PCR primers that are specific to CG I.1b (Table 1). As shown in Fig. 2a, diverse DGGE banding patterns were observed in the different samples. When the DGGE bands were excised and sequenced, all were found to be from the CG I.1b-related group. The phylogenetic tree (Fig. 2b), which was built from these DGGE band sequences, indicated that CG I.1b from marine sediments were diverse. Most of these marine CG I.1b-related sequences are closely related to sequences found in freshwater and terrestrial environments. Our results show that diverse CG I.1b inhabits marine sediments.
https://static-content.springer.com/image/art%3A10.1007%2Fs00792-008-0165-7/MediaObjects/792_2008_165_Fig2_HTML.gif
Fig. 2

Diversity of CG I.1b in marine sediments. a DGGE profiles of 16S rRNA gene sequences from CG I.1b in marine sediments: The four different sample sites are indicated at the top of figure. Bands marked with a dot were excised from the gel, re-amplified and sequenced. b Phylogenetic tree based on the 16S rRNA gene sequences of CG I.1b recovered from the DGGE bands. Clones closely related to a representative clone are shown in parentheses. The reference sequences were chosen to show the diversity of the sequences and to indicate the closest relatives to the sequences found in our study. For comparison, ES2 and ED1 library clones are included (see Fig. 1). The scale bar represents two substitutions per 100 nucleotide positions. Significant bootstrap values (≥50%; 1,000 replicates) are indicated by filled circles at branch points

Archaeal amoA diversity in marine sediments

The detection of CG I.1a and CG I.1b 16S rRNA gene sequences in marine sediments suggested that amoA sequences from these two groups might be present also in the sediment samples. To compare the diversity of archaeal amoA with archaeal 16S rRNA genes in marine sediments, we constructed archaeal amoA libraries from the surface layers of the four marine sediments using a universal archaeal amoA primer set A (see Table 1). However, we were unable to amplify archaeal amoA sequences from the subsurface layers (>9 cm deep) (data not shown). This suggests that AOA might prefer oxic condition although 16S rRNA gene of MG I was reported to be found at deep subsurface marine sediments (Teske 2006). Analysis of the library sequences showed that the average similarity of the nucleotide sequences was 90.7%. The phylogenetic trees which were constructed with environmental clones of each group from the four sampling sites are shown in Fig. 3. We found that most clones that were identified from the four different libraries using the universal archaeal amoA primer set A were related to amoA of CG I.1a, which contains an isolated crenarchaeotal strain, Nitrosopumilus maritimus (≥82.7% similarity) (Fig. 3). A clone, DDTO9, which was detected in the surface layers from the DD sampling site, was related to CG I.1b amoA genes, although its similarity to that of the soil fosmid clone 54d9 was also quite low (78.8%) (Fig. 3). If we assumed that our amoA clones were related to the 16S rRNA gene phylotypes of CG I.1a that were retrieved in this study, we found that the sequence similarity of the 16S rRNA gene was >98%. Moreover, the average similarity of the nucleotide sequences in our clones was 90.7% to that of amoA. In addition, those AOBs with more than 98% similarity with respect to 16S rRNA also showed more than 90% amoA similarity (Purkhold et al. 2000). We also found that most of the sequences in our amoA clones were not closely related to sequences retrieved from estuarine sediments (Beman and Francis 2006) and marine water columns (Francis et al. 2005) (see Fig. 3).
https://static-content.springer.com/image/art%3A10.1007%2Fs00792-008-0165-7/MediaObjects/792_2008_165_Fig3_HTML.gif
Fig. 3

Phylogenetic tree constructed from archaeal amoA gene sequences that were obtained from marine sediments. Phylotype names refer to the name of the sampling site (ST3, UJ, JD, and DD). The lettersO and S that are written before the digit in the clone name indicate clones from a library produced using PCR primer set A or B, respectively. Numbers in parentheses following the clone names indicate the number of times that the sequences were found in the clone libraries from the four sampling sites ST3, UJ, JD, and DD, respectively. Clones from this study are indicated in boldface. The reference sequences were chosen to show the diversity of the sequences and to indicate the closest relatives to the sequences found in our study. The scale bar represents five substitutions per 100 nucleotide positions. Significant bootstrap values (≥50%; 1,000 replicates) are indicated by filled circles at branch points

To analyze further the diversity of CG I.1b-like amoA, we designed a set of PCR primers (PCR primer set B), which were specific for the detection of CG I.1b-like amoA for use in this study (see Table 1). Using this set of primers, we were able to detect various amoA sequences in all samples of marine sediment. Furthermore, these sequences were related to CG I.1b-like amoA, and the average nucleotide sequence similarity was 88.7% (≥89.1% identity to that of 54d9). Clades of our CG I.1b-like amoA sequences were also distantly related to amoA sequences isolated from estuarine sediments (Beman and Francis 2006) but rather related to those from soils (Leininger et al. 2006) and wastewater plant (Park et al. 2006).

Our data tend to show that clades of unique amoA sequences were likely to cluster according to sampling sites. For example, most of the CG I.1b clones (95%) from the UJ sampling site were closely related to each other and constituted a clade with the representative clone, UJTS11. The distribution of amoA subgroups might reflect adaptation of specific AOA to provinces of marine sediments. If we could analyze microdiversity of Crenarchaeota using 16S-23S intergenic spacer region, we may observe clustering of specific AOA to provinces of marine sediments. We estimated the coverage, richness, evenness, and diversity of the phylotypes in each library (Table 3). We recovered 57 unique phylotypes (primer set A: 40; primer set B: 17) from approximately 180 individual sequences in which the cutoff value for nucleotide sequence similarity was 95%. Phylotype richness at each sampling site varied among the libraries of archaeal CG I.1a amoA (DD > UJ > JD > ST3) and CG I.1b amoA (DD = ST3 > JD > UJ). Conventional diversity and evenness indices were highest for the DD sampling site at which archaeal amoA sequences from groups CG I.1b and I.1a were detected using the universal PCR primer set. The results of the analysis of the rarefaction curves, which were calculated for each library using the individual-based Coleman method, also supported the results obtained after determining diversity indices (sFig. 2). Overall, the estimates of species richness using Chao1 and the percentage of coverage using Good’s method indicated that diversity was higher in libraries that were constructed using the universal primer set than those constructed using the CG I.1b-specific primer set. This contrasts with the previous result that the diversity of the CG I.1b 16S rRNA gene is higher than that of the CG I.1a 16S rRNA gene when determined in the clone library and by 16S rRNA gene-DGGE analysis. This suggests that only some subgroups of CG I.1b might harbor archaeal amoA.
Table 3

Estimation of sequence diversity and phylotype coverage of amoA gene libraries

 

Universal amoAa

CG I.1b specific amoAa

ST3

UJ

JD

DD

ST3

UJ

JD

DD

No. of clones

21

20

21

21

22

21

21

22

Phylotypes

6

10

9

12

6

1

3

6

Singletons

3

6

5

7

5

2

0

3

Chao1 estimated richness

7.5

17.5

14

17.25

16

2

3

9

Chao1 standard deviation

2.6

8.18

6.05

5.37

10.07

0.35

0.05

4.42

Shannon’s index for diversity

1.29

2.06

1.93

2.30

0.90

0.19

0.93

1.33

Simpson’s index for diversity

2.88

8.64

7.24

1.40

1.7

1.11

2.50

3.08

Evenness

0.61

0.78

0.76

0.84

0.41

0.61

0.85

0.63

Fisher’s diversity

2.81

7.96

5.97

11.64

2.72

0.54

0.96

2.08

Good’s estimator of coverage, %

85.7

70.0

76.2

66.7

77.3

90.5

100

86.4

Diversity and richness estimators were determined using the ESTIMATES software program. Each group is subdivided according to the sampling sites and the individual estimates are displayed in columns for each group and sampling site

aDiversity was estimated using Operational Taxonomic Units defined as groups with ≥95% sequence similarity

Abundance of AOA in marine sediments

Archaea are known to predominate among ammonia-oxidizing microorganisms in a variety of soils (Leininger et al. 2006) and in North Sea waters (Wuchter et al. 2006). In this study, we found that CG I.1a-related and CG I.1b-related AOA are diverse and ubiquitous in marine sediments using sets of PCR primers specific for archaeal amoA and 16S rRNA genes. To determine the relative abundance of the AOA and AOB groups, we quantified copy numbers of archaeal and bacterial amoA in marine sediments using real-time PCR. To this end, sets of PCR primers designed in this study were used for amplification of amoA sequences from CG I.1a and I.1b and AOB (Table 1). PCR amplification of archaeal amoA appeared to diminish according to sediment depth and could not be achieved at sediment depths greater than 9 cm. The archaeal amoA copy numbers ranged from 1.1 × 107 to 4.9 × 107 per gram of surface layer sediments. The copy numbers of archaeal amoA were greater than those of bacterial amoA, with ratios ranging from 11 to 28 (Fig. 4). This range of ratios is relatively low when compared to that of the soil or North Sea water, in which the ratio was as great as 1,000. CG I.1a-like amoA dominated over CG I.1b-like amoA, with the ratios ranging from 17 to 270. The ranges of copy number of 16S rRNA gene of CG I.1a and I.1b are 1.3 × 107–1.3 × 108 and 2.5 × 105–2.7 × 106, respectively (Fig. 4). The copy number of archaeal 16S rRNA genes within the surface layers of typical marine sediments ranges from 107 to 108 copies per gram of sediment (Inagaki et al. 2004; Schippers and Neretin 2006). This indicates that the copy number of archaeal amoA determined in this study roughly corresponds with that of archaeal 16S rRNA gene with ratios of CG I.1a- and CG I.1b rRNA gene to amoA ranging from 0.9 to 2.6 and from 0.7 to 2.2, respectively. The results of our comparative study on the diversity and abundance of archaeal 16S rRNA and amoA genes suggests that the AOA, especially CG I.1a, might be dominant ammonia oxidizers in oxic environments of marine sediments. Beman and Francis (2006) have identified CG I.1b-related archaeal amoA as a major constituent at two of the five sampling sites of estuarine sediments. This indicates that the AOA community composition of estuarine environments is situated between those of freshwater (or soil) and marine sediments. Accordingly, the contribution of CG I.1b to nitrification might be higher in estuarine than in marine sediments.
https://static-content.springer.com/image/art%3A10.1007%2Fs00792-008-0165-7/MediaObjects/792_2008_165_Fig4_HTML.gif
Fig. 4

Relative abundance of the groups of AOA and AOB found in the East Sea of Korea, as determined by estimating the copy number of the amoA gene. The copy number of the amoA gene was quantified by real-time PCR. The error bar represents the standard deviation of triplicate PCR reactions. Ratios of amoA gene copies are shown in boxes above the chart

Despite the importance of the nitrogen cycle in marine sediments, there are insufficient data about the potential ecological role of Archaea in the nitrification of marine sediments. In this study, we tentatively identified two groups of Archaea as possible AOA, using comparative analysis of 16S rRNA and amoA gene libraries. Quantitative analysis suggested that CG I.1a was the dominant group among ammonia-oxidizing microorganisms in marine sediments. Furthermore, the observation of some amoA sequences distantly related to known archaeal amoA genes indicates that novel AOAs still remain unidentified in marine sediments. Metagenomic and/or single cell analysis (Ottesen et al. 2006; Stepanauskas and Sieracki 2007; Treusch et al. 2005) might be required for the identification of Archaea with specific amoA genes.

Acknowledgments

This research was supported by grants PE07010 (KOPRI Project) and the 21C Frontier Microbial Genomics and Application Center Program from the Ministry of Science and Technology, Republic of Korea.

Supplementary material

792_2008_165_MOESM1_ESM.pdf (304 kb)
sFig. 1 (PDF 304 kb)
792_2008_165_MOESM2_ESM.pdf (258 kb)
sFig. 2 (PDF 257 kb)

Copyright information

© Springer 2008