SEARCH

SEARCH BY CITATION

Keywords:

  • 16S rRNA gene;
  • PCR;
  • Archaea;
  • terrestrial acidic hot spring;
  • thermoacidophiles;
  • Thermoplasma

Abstract

  1. Top of page
  2. Abstract
  3. Introduction
  4. Materials and methods
  5. Results and discussion
  6. References
  7. Supporting Information

The phylogenetic diversity of archaeal 16S rRNA genes in a thermoacidic spring field of Ohwakudani, Hakone, Japan, was investigated by PCR-based analysis using a novel Archaea-specific primer designed in the present study. Clone libraries of archaeal 16S rRNA genes were constructed from hot water (78 °C) and mud (28 °C) samples by PCR using a newly designed forward primer and a previously reported forward primer with reverse primers. Most phylotypes found in the libraries from the hot water sample were related to cultured (hyper)thermophiles. The phylotypes and their detection frequencies from the hot water sample were similar for the libraries amplified with the two different primer sets. In contrast, phylotypes having a low similarity (<95%) to cultured Archaea were found in the libraries from the mud sample. Some of the phylotypes were relatively close to members of Thermoplasmata (80–93% similarity) and the others were not clearly affiliated with Crenarchaeota and Euryarchaeota, but related to Thaumarchaeota and Korarchaeota. The phylotypes and their detection frequencies were significantly different between the two libraries of the mud sample. Our results from the PCR-based analysis using the redesigned primer suggest that more diverse, uncultured Archaea are present in acidic environments at a low temperature than previously recognized.


Introduction

  1. Top of page
  2. Abstract
  3. Introduction
  4. Materials and methods
  5. Results and discussion
  6. References
  7. Supporting Information

PCR-based analysis targeting the 16S rRNA gene has revealed that diverse yet-uncultivated prokaryotes are present in natural environments (Pace, 1997; Schleper et al., 2005). It is assumed that cultured species account for <1% of the total prokaryotes living on Earth (Amann et al., 1995). ‘Universal’ oligonucleotide primers for the domain Bacteria or Archaea have been used for gene amplification with PCR (Lane, 1991; Delong, 1992). The nucleotide sequences of the primers have been modified based on the accumulated 16S rRNA gene sequences from various environments (Baker et al., 2003). These ‘universal’ primers have been designed based on the conserved regions among most archaeal and/or bacterial 16S rRNA genes. It should be noted that the detectable members are constrained by the nucleotide sequence of the PCR primers used, meaning that these ‘universal’ primers may not be completely universal.

Diverse Archaea have been detected in terrestrial and marine environments (Robertson et al., 2005; Schleper et al., 2005). Archaeal diversity in natural environments has often been investigated by PCR-based analysis using Arch21F as a forward primer as reported previously (Delong, 1992) or other primers (Massana et al., 1997; Dojka et al., 1998; Eder et al., 1999; Reysenbach et al., 2000). These primers were designed based on the conserved regions of archaeal 16S rRNA gene sequences between positions 7 and 26 in the Escherichia coli numbering system (Brosius et al., 1981); this region corresponds to positions 2–21 of the 16S rRNA gene sequence (rrnB) of Methanocaldococcus jannaschii (L77117) (Bult et al., 1996). Whole-metagenome sequencing and direct cultivation have shown that some Archaea are not detected when using general Archaea-specific primers, including Nanoarchaeum (Huber et al., 2002) and the ARMAN group (Baker et al., 2006). It is important to assess, redesign and use PCR primers that can amplify more sequences as well as longer sequences for the study of the diversity, distribution and evolution of Archaea.

Terrestrial hot springs are an extreme environment where (hyper)thermophilic and/or acidophilic Archaea thrive. The study of the hyperthermophilic Archaea is important to understand the early evolution of life because hyperthermophilic archaeal groups are one of the deepest lineages of all life (Woese, 1987). In fact, the deeper lineage Korarchaeota was detected in a terrestrial hot spring field (Barns et al., 1994; Barns et al., 1996) and the genome analysis has provided an insight into the early evolution of Archaea (Elkins et al., 2008). Several (hyper)thermophilic Archaea have been cultured from a terrestrial thermoacidic spring in Ohwakudani, Hakone, Japan (Itoh et al., 2002, 2003a, 2007). However, the molecular characterization of this spring field has not been performed. Here, we report the diversity of archaeal 16S rRNA genes in this spring by PCR-based analysis using a novel Archaea-specific primer modified from Arch21F.

Materials and methods

  1. Top of page
  2. Abstract
  3. Introduction
  4. Materials and methods
  5. Results and discussion
  6. References
  7. Supporting Information

Site description and sampling

Hot water and mud samples were obtained from a thermoacidic spring field that is located in Ohwakudani, Hakone, Japan (35°14.40′N, 139°01.12′E; Supporting Information, Fig. S1), in September 2009. The hot water sample was collected in clean 20-L polypropylene tanks at a hot water pool (78 °C, pH 3.5). The mud sample was collected from a depth of 0–5 cm from the bottom of a warm water pool (28 °C, pH 2.5) that is located downstream of the hot water pool. The mud sample was stored in a sterile 50-mL plastic tube. The geochemical characteristics of the hot water in this field have been reported previously (Ohba et al., 2007): in brief, the water is hot (up to 97 °C) and acidic (pH 2.0–3.3), and contains H2S (0.1–5.6 mg L−1) and Fe2+ (1.6–144 mg L−1). In this field, the fumarolic gas contains H2 (41.6–500 μmol mol−1), H2S (135–3310 μmol mol−1), SO2 (9.2–123 μmol mol−1), CH4 (up to 22 μmol mol−1) and CO2 (2030–20 600 μmol mol−1) (Ohba et al., 2007).

Primer design

To design a primer for the 5′ end of archaeal 16S rRNA genes, a total 82 of archaeal 16S rRNA gene sequences were extracted from whole-genome sequences and fosmid library data in Genbank and were aligned (Fig. 1). Arc9F (5′-CYGGTYGATCCYGCCRG-3′) was redesigned by modifying Arch21F (Delong, 1992) based on the alignment (Fig. 1). It is shown that Arch21F could not cover 39 of the 82 archaeal 16S rRNA genes (47.6%) (Fig. 1). In particular, Arch21F has several mismatches to taxonomic groups related to Methanomicrobia and Thermoplasma. This could cause a low PCR amplification efficiency of the 16S rRNA gene from these groups. Arc9F was designed to cover Methanomicrobia- and Thermoplasma-related groups. It should be noted that Arc9F still has two to four mismatches to several members, for example, Methanobacteria, Nanoarchaeum (Huber et al., 2002) and the ARMAN group (Baker et al., 2006) (Fig. 1). One of the mixed sequences of Arc9F (5′-CTGGTTGATCCTGCCAG-3′) has no mismatches to several eukaryotic 18S rRNA genes as confirmed by probecheck (Loy et al., 2008), suggesting that an alternative forward primer should be used in PCR if eukaryotes were detected with the original Arc9F. In the present study, we did not need to modify Arc9F because no eukaryotic 18S rRNA genes were detected.

image

Figure 1.  Comparison of the sequences of Archaea-specific primers with archaeal 16S rRNA genes. The phylogenetic tree was constructed from archaeal 16S rRNA genes derived from whole-genome sequences and fosmid clones. The ML tree was inferred using 1131 homologous positions in the alignment dataset. Bootstrap values (>70%) based on 100 replicates are shown at the branch points as filled circles. The scale bar represents 0.1 nucleotide substitutions per sequence position. Numbers in parentheses indicate the phylotype number included in each taxonomic group. Alignment of the sequences of the Methanocaldococcus jannaschii positions 1–23 is shown. The nucleotide sequences matching Arch21F are shown as ‘=’. Taxonomic groups shown in bold contain one or more phylotypes that have one or more mismatch to Arch21F.

Download figure to PowerPoint

16S rRNA gene analysis

The hot water sample was centrifuged at 3000 g to collect particles including microbial cells. The total weight of the particles precipitated from the 27-L hot water sample was 38 g. The mud sample was used directly for DNA extraction. Genomic DNA was extracted from a part of the precipitate (0.2 g) and the mud (0.3 g) using a Fast DNA kit for soil (Qbiogene Inc., Irvine, CA). Partial 16S rRNA genes were amplified by PCR using TaKaRa EX Taq Hot Start Version (Takara Bio, Shiga, Japan) with the following oligonucleotide primer sets Arch21F–Arch958R (Delong, 1992) and Arc9F–Uni1406R. A variety of archaeal groups can be detected using the reverse primer Uni1406R (Kato et al., 2009a, b, 2010). The PCR was performed for 25 cycles of the following thermal cycle (94 °C for 30 s, 60 °C for 30 s and 72 °C for 120 s) with each primer set. The PCR products were cloned using a TOPO TA cloning kit (Invitrogen, CA). The nucleotide sequences of randomly selected clones were determined with a BigDye Terminator v3.1 cycle sequencing kit (Applied Biosystems, CA) using M13 forward and reverse primers (Invitrogen) and the internal primers 519r, 530f, 907r and 926f (Lane, 1991) on an ABI Prism 3130xl genetic analyzer (Applied Biosystems). We checked all sequences to remove chimeric sequences with blast analysis of sequence segments (Altschul et al., 1990) and the mallard program (Ashelford et al., 2006). The nucleotide sequences of the clones without chimeric sequences were aligned using muscle (Edgar, 2004). Putative introns observed in the sequences of clones were removed by judging from the alignments. Clones having 97% sequence similarity or higher were treated as a phylotype using dotur (Schloss & Handelsman, 2005). The exon sequences of the phylotypes were realigned with other published sequences including the closest one determined by blast searches (Altschul et al., 1990). The construction of phylogenetic trees and the diversity analysis were performed as described previously (Kato et al., 2009a).

Accession numbers

The nucleotide sequences of the phylotypes reported in this paper have been deposited in the DDBJ database under accession numbers AB600328AB600387.

Results and discussion

  1. Top of page
  2. Abstract
  3. Introduction
  4. Materials and methods
  5. Results and discussion
  6. References
  7. Supporting Information

Archaea in hot water

The phylotypes detected in the hot water sample were related to cultured (hyper)thermophilic members of Crenarchaeota (Figs 2a, b and 3), i.e. Vulcanisaeta, Caldivirga, Thermoproteus, Acidilobus and Stygiolobus (Zillig et al., 1981; Segerer et al., 1991; Itoh et al., 1999, 2002; Prokofeva et al., 2000) with 97–99% similarity. These members have been isolated from terrestrial hot springs and include thermoacidophiles for which the optimum growth temperatures and pH are 80–90 °C and 2.5–6.8, respectively, as summarized in the previous report (Itoh, 2003). The detection of phylotypes related to these thermoacidophiles in the hot water sample is consistent with the high-temperature (78 °C) and acidic environment (pH 3.5).

image

Figure 2.  The detection frequency of the phylotypes in the clone libraries for each sample and each primer set. The pie charts indicate the results from the hot water sample using the primer set Arc9F–Uni1406R (a) or Arch21F–Arch958R (b) and those from the mud sample using the primer set Arc9F–Uni1406R (c) or Arch21F–Arch958R (d).

Download figure to PowerPoint

image

Figure 3.  Phylogenetic tree of 16S rRNA genes in the Crenarchaeaota. The ML tree was inferred using 748 homologous positions in the alignment dataset. Ferroglobus placidus (AF220166), Archaeoglobus profundus (AF322392), Thermococcus hydrothermalis (Z70244), Methanocaldococcus jannaschii (L77117) and Methanopyrus kandleri (AE009439) were used as outgroups. Bootstrap values (>50%) based on 100 replicates are shown at the branch points. The scale bar represents 0.1 nucleotide substitutions per sequence position. Bold letters indicate the clones detected in this study. Each color indicates the clones from the hot water sample using the primer set Arc9F–Uni1406R (red) or Arch21F–Arch958R (yellow) and those from the mud sample using the primer set Arc9F–Uni1406R (green) or Arch21F–Arch958R (blue). THSC, terrestrial hot spring Crenarchaeota MHVG-I, marine hydrothermal vent group I; UTSCG, uncultured thermoacidic spring clone group; UTRCG, uncultured Thaumarchaeota-related clone group.

Download figure to PowerPoint

One clone (HO78W9A61, AB600380) detected in the hot water was related to the environmental clone OP-9, which is affiliated with the Nanoarchaeota (Hohn et al., 2002) (94% similarity). We excluded this clone in the construction of the phylogenetic tree and the statistical analysis because of the short length of the sequence (190 bp). Phylotypes affiliated with Nanoarchaeota have been detected in other hot spring fields (Hohn et al., 2002; Casanueva et al., 2008).

Archaea in solfataric mud

All euryarchaeotic phylotypes detected in the mud sample were related to members of the Thermoplasmata, a thermoacidophilic group. In this study, these phylotypes were clustered in four groups: Thermoplasma-related groups I to IV (TRG-I to IV) (Fig. 4). Cultured species related to Thermoplasma and other acidic environmental clones were included in the TRG-I (Fig. 4). The phylotypes in TRG-I, except HO28S9A75, were closely related to a thermoacidophilic archaeon, Thermogymnomonas acidicola, which belongs to a recently reported Thermoplasma-related genus (Itoh et al., 2007) (90–93% similarity). This archaeon, which grows in the range 38–68 °C and at pH 1.8–4.0, was isolated from a solfataric soil in Hakone, Japan (Itoh et al., 2007). In contrast, TRG-II, III and IV include no cultured species (Fig. 4). The detected phylotypes in TRG-II have high similarity (96–98%) to environmental clones recovered from the Lassen Volcanic National Park (Wilson et al., 2008) (not shown in Fig. 4 because of the short sequences). The phylotypes in TRG-III were related to environmental clones recovered from acidic wetlands, river water and a mine (Jennifer et al., 2002; Garcia-Moyano et al., 2007; Rowe et al., 2007). TRG-IV includes environmental clones from terrestrial hot springs (Jackson et al., 2001; Ng et al., 2005; Spear et al., 2005). These uncultured phylotypes in the TRGs detected in the present study may represent acidophiles, as supported by the environmental characteristics of the present study field and other environments where related clones were detected, and the physiology of the cultured members of the Thermoplasmata (Reysenbach, 2001).

image

Figure 4.  Phylogenetic tree of 16S rRNA genes in the Euryarchaeaota. The ML tree was inferred using 771 homologous positions in the alignment dataset. Ferroglobus placidus (AF220166), Archaeoglobus profundus (AF322392) and Thermococcus hydrothermalis (Z70244) were used as outgroups. Bootstrap values (>50%) based on 100 replicates are shown at the branch points. The scale bar represents 0.1 nucleotide substitutions per sequence position. Bold letters indicate the clones detected in this study. Each color indicates the clones from the mud sample using the primer set Arc9F–Uni1406R (green) or Arch21F–Arch958R (blue). TRG-I to IV: Thermoplasma-related groups I to IV.

Download figure to PowerPoint

Crenarchaeotic phylotypes related to cultured thermoacidophiles, such as Thermocladium, Caldisphaera, Metallosphaera, Sulfolobus and Acidianus, were detected in the 28 °C mud sample (Fig. 3). These cultured thermoacidophiles have been isolated from hot springs (Brock et al., 1972; Segerer et al., 1986; Huber et al., 1989; Itoh et al., 1998, 2003b). These members can grow at a relatively low temperature (45–50 °C) compared with members of Vulcanisaeta, Caldivirga and Stygiolobus (Itoh, 2003), phylotypes of which were detected in hot water samples and also in the mud sample. Nevertheless, the temperature (28 °C) of the solfataric mud does not provide a suitable growth condition for (hyper)thermophiles. Therefore, these phylotypes related to (hyper)thermophiles that were detected in the mud sample are possibly remnant DNA derived from the high-temperature environments in the hot water pool and/or the stream between the hot water pool and the solfataric mud pool.

Phylotypes that did not clearly belong to the cultured thermophilic Crenarchaeota and Euryarchaeota were detected in the mud sample (Fig. 3). These phylotypes were affiliated with the terrestrial hot spring Crenarchaeota (THSC) (Takai & Horikoshi, 1999; Takai & Sako, 1999), Uncultured thermoacidic Spring Clone Group (UTSCG) or Uncultured Thaumarchaeota-related clone group (UTRCG). The latter two groups are defined in the present study. These phylotypes were relatively close to the recently proposed Thaumarchaeota (Brochier-Armanet et al., 2008) and Korarchaeota (Barns et al., 1994; Barns et al., 1996) rather than thermophilic cultured Crenarchaeota (Fig. 3). The phylotypes in the THSC (the representative clones are HO28S21A13 and HO28S9A51) were related to environmental clones A14 and A1 (Jackson et al., 2001) and pUWA2 and pUWA36 (Takai & Sako, 1999), which were detected in thermoacidic springs. The phylotype (the representative clone is HO28S9A21) in the UTSCG was related to environmental clones A6 and A13 (Jackson et al., 2001). The phylotypes (the representative clone is HO28S21A56) in the UTRCG were related to soil clone ArcB_cB07 (Hansel et al., 2008) and groundwater clone SWA13 (Shimizu et al., 2007). These uncultured phylotypes detected in the present study seem to represent acidophiles as inferred from the environmental characterization. Further cultivation efforts are needed to determine their physiology.

Differences in community structures depend on the primer sets used

The archaeal phylotypes detected in the hot water sample were similar between the libraries, i.e. HO78W9 vs. HO78W21, amplified with the primer sets Arc9F–Uni1046R and Arch21F–Arch958R (Fig. 2a and b), respectively. The coverage values were 99–100% (Table 1) and the rarefaction curves leveled off (Fig. S2). These results indicate that almost all of the archaeal phylotypes in the hot water were recoverable by the primer sets used. However, we are not able to exclude the possibility that there are novel Archaea that could not be recovered with either primer set, such as the ARMAN group (Baker et al., 2006). Archaeal diversity in the hot water (78 °C) is likely to be lower than that in the solfataric mud (28 °C), which is supported by the Chao1 species richness estimates, Shannon diversity index (Table 1) and rarefaction curves (Fig. S2) for both clone libraries HO78W9 and HO78W21.

Table 1.   Statistical analyses of 16S rRNA gene clone libraries
Library nameSamplePrimer set usedTotal clone numberPhylotype numberChao1 species richness*Shannon score*Coverage (%)
  • *

    Numbers in parentheses indicate the 95% confidence interval.

HO28S9MudArc9F-Uni1406R932862 (38–142)2.59 (2.33–2.85)82
HO28S21MudArch21F-Arch958R942048 (27–133)2.29 (2.06–2.52)88
HO78W9Hot waterArc9F-Uni1406R6444 (4–4)1.04 (0.88–1.21)100
HO78W21Hot waterArch21F-Arch958R9177 (7–7)1.33 (1.13–1.53)99

Differences in the community structures of the mud sample determined using Arc9F–Uni1406R and Arch21F–Arch958R, i.e. HO28S9 vs. HO28S21, were observed (Fig. 2c and d). The detection frequency of the TRG-I to -IV clones (57.0% of the total clone number) was higher in the HO28S9 library than in the HO28S21 library (12.8%) (Fig. 2c and d), although both libraries were derived from the same DNA extract. In contrast, the detection frequencies of Vulcanisaeta, Thermocladium and UTRCG were relatively higher in HO28S21 (23.4%, 9.6% and 9.6%, respectively, Fig. 2d) than in HO28S9 (1.1%, 2.2% and 1.1%, respectively, Fig. 2c). The differences most likely resulted from the efficient annealing of the forward primer Arch9F used with the 16S rRNA gene of the phylotypes in the TRGs (Fig. 5). In fact, the 16S rRNA gene sequences of most phylotypes in the TRGs have C at position 21 of the 16S rRNA gene sequence (rrnB) of M. jannaschii (L77117) (Fig. S3). Arch21F has A at its 3′ final end (the M. jannaschii position is 21), which could cause a low amplification efficiency of the 16S rRNA gene of the TRGs. Actually, the phylotypes of the TRGs represented over half of the total clone number of the HO28S9 library (Fig. 2c). Such phylotypes were also detected in the HO28S21 library (<10% of the total number of clones, Fig. 2d) despite the mismatch. This is probably because of the non-Watson–Crick base pairing of A–G and/or loss of the 3′ end of A during storage of the primer. A larger number of unique phylotypes in the TRGs was detected in the HO28S9 library than in the HO28S21 library (Fig. S4). This may also reflect the differences in the sequences of the primer sets used.

image

Figure 5.  Probability of the nucleotide sequences of 16S rRNA genes in the clone libraries at the 5′ end position targeted by the forward primers. The probability of the consensus sequence in all clones for each library is shown as a sequence logo. (a) The comparison of clone libraries of HO78W9 and HO78W21 from the hot water sample and (b) the comparison of clone libraries of HO28S9 and HO28S21 from the solfataric mud sample. Numbers at the bottom axis of each figure indicate the positions of the 16S rRNA gene sequence (rrnB) of Methanocaldococcus jannaschii (L77117).

Download figure to PowerPoint

The phylotypes related to Vulcanisaeta and Thermocladium accounted for higher percentages of the HO28S21 library (23.4% and 9.6%, respectively, Fig. 2d) than the HO28S9 library (both 1.1%, Fig. 2c). The phylotypes related to Metallosphaera were dominant in both the HO28S9 and the HO28S21 libraries (Fig. 2c and d). These phylotypes may represent thermophiles as supported by the optimum growth temperature estimation based on the GC content of the 16S rRNA gene (Kimura et al., 2007) and the physiology of the cultured members. The optimum growth temperatures estimated for the phylotypes related to Vulcanisaeta, Thermocladium and Metallosphaera are 94.6, 79.0 and 76.8 °C, respectively. These estimates are compatible with the optimum growth temperatures of members of each genus (Huber et al., 1989; Itoh et al., 1998, 2002). The optimum growth temperatures for the phylotypes related to UTSCG and UTRCG are estimated to be 58.0 and 61.0 °C. These phylotypes related to cultured (hyper)thermophiles, UTSCG and UTRCG that were detected in the mud sample may be remnant DNA that originated from the higher temperature environments as described above. In contrast, the optimum growth temperatures estimated for the TRG-I to IV phylotypes detected are 36.8, 38.6, 45.0 and 46.0 °C, respectively. These temperatures are relatively comparable to the low temperature of the solfataric mud environment. Overall, the archaeal community structure represented in the HO28S9 library is more consistent with the environment than that represented in the HO28S21 library.

More archaeal phylotypes are likely to be obtained in acidic spring fields using the primer set Arc9F–Uni1406R than using the set Arch21F–Arch958R, based on comparative analysis of the archaeal phylotypes obtained with the two primer sets. The number of phylotypes observed was larger in HO28S9 than HO28S21 (Table 1), even though the total number of clones was very similar in each library. Accordingly, the number of unique phylotypes found in the HO28S9 library was more than those in HO28S21 (Fig. S4). The analysis of the Chao1 richness estimators of shared phylotypes suggests that the phylotypes in the HO28S9 library would cover all phylotypes in HO28S21 (Fig. S4) if the coverage of the clone library for each primer set had reached 100% of the total archaeal phylotypes. Modification of the primer sequence of the Arch21F to Arc9F was expected to match more phylotypes (Fig. 1). In addition to the M. jannaschii position 21 as described above, the modification at positions 5 and 9 may have also contributed to the increased efficiency of hybridization and amplification (Fig. 5). Furthermore, the reverse primers used may contribute to efficient amplification. In fact, the sequences of some phylotypes that were recovered using Arc9F–Uni1406R have mismatches to the primer sequence of Arch958R at the position targeted by this reverse primer (Fig. S3). We conclude that a more diverse archaeal community in acidic environments at a low temperature was revealed by 16S rRNA gene clone library construction using the Arc9F–Uni1406R primer set.

References

  1. Top of page
  2. Abstract
  3. Introduction
  4. Materials and methods
  5. Results and discussion
  6. References
  7. Supporting Information

Supporting Information

  1. Top of page
  2. Abstract
  3. Introduction
  4. Materials and methods
  5. Results and discussion
  6. References
  7. Supporting Information

Fig. S1. Photos of the sampling points.

Fig. S2. Rarefaction curves for each clone library.

Fig. S3. Alignment of the 16S rRNA gene sequences of the phylotypes in the HO28S9 library.

Fig. S4. Venn diagrams comparing (a) the phylotype numbers and (b) the Chao1 species richness estimates in the archaeal clone libraries HO28S9 and HO28S21.

Please note: Wiley-Blackwell is not responsible for the content or functionality of any supporting materials supplied by the authors. Any queries (other than missing material) should be directed to the corresponding author for the article.

FilenameFormatSizeDescription
FML_2267_sm_supplfig.pdf5989KSupporting info item

Please note: Wiley Blackwell is not responsible for the content or functionality of any supporting information supplied by the authors. Any queries (other than missing content) should be directed to the corresponding author for the article.