Accepted Article
Received Date : 07-Jun-2013 Revised Date : 26-Jul-2013 Accepted Date : 26-Jul-2013 Article type : Research Paper Editor : Angela Sessitsch
Different behavior of methanogenic archaea and Thaumarchaeota in rice field microcosms
Xiubin Ke1, 2, 3, Yahai Lu1 and Ralf Conrad2*
1, College of Resources and Environment Sciences, China Agricultural University, Beijing 100193, China 2, Max-Planck-Institute for Terrestrial Microbiology, 35043 Marburg, Germany 3, Biotechnology Research Institute, Chinese Academy of Agricultural Sciences, Key Laboratory of Crop Biotechnology, Ministry of Agriculture, Beijing, China
*Correspondence: R Conrad, Max-Planck-Institute for Terrestrial Microbiology, Department Biogeochemistry, Karl-von-Frisch-Strasse10, D-35043 Marburg, Germany. E-mail: [email protected]
Running title: Methanogens and Thaumarchaeota in rice fields
This article has been accepted for publication and undergone full peer review but has not been through the copyediting, typesetting, pagination and proofreading process, which may lead to differences between this version and the Version of Record. Please cite this article as doi: 10.1111/1574-6941.12188 This article is protected by copyright. All rights reserved.
Accepted Article
Key words: Archaea; Methanogens; Thaumarchaeota; rice field soil; T-RFLP; qPCR
Abstract Archaea in rice fields play an important role in carbon and nitrogen cycling. They comprise methane-producing Euryarchaeota as well as ammonia-oxidizing Thaumarchaeota, but their community structures and population dynamics have not yet been studied in the same system. Different soil compartments (surface, bulk, rhizospheric soil) and ages of roots (young and old roots) at two N fertilization levels and at three time points (the panicle initiation, heading and maturity periods) of the season were assayed by determining the abundance (using qPCR) and composition (using T-RFLP and cloning/sequencing) of archaeal genes (mcrA, amoA, 16S rRNA). The community of total Archaea in soil and root samples mainly consisted of the methanogens and the Thaumarchaeota and their abundance increased over the season. Methanogens proliferated everywhere, but Thaumarchaeota only on the roots and in response to nitrogen fertilization. The community structures of Archaea, methanogens and Thaumarchaeota, were different in soil and root samples indicating niche differentiation. While Methanobacteriales were generally present, Methanosarcinaceae and Methanocellales were the dominant methanogens in soil and root samples, respectively. The results emphasize the specific colonization of roots by two ecophysiologically different groups of archaea which may belong to the core root biome.
One sentence summary The community structures of Archaea, methanogens and Thaumarchaeota, were different in soil and roots of rice microcosms indicating niche differentiation and the specific colonization of roots.
This article is protected by copyright. All rights reserved.
Accepted Article
Introduction Rice paddy fields are an important source of methane and contribute about 20% to the global methane emission (Conrad, 2009). Rice fields are managed wetland ecosystems with bulk soil, surface soil, rhizosphere soil and roots as distinct habitats (Liesack et al., 2000). The gas vascular system of the rice plants allows the ventilation of gases (e.g., O2, CH4) between soil and atmosphere. Therefore, O2 is only available in the uppermost surface layer of the soil and in a shallow layer at the root surface, while bulk soil is completely devoid of O2. Methanogenic Euryarchaeota play an important role in the anaerobic degradation of organic matter and the production of CH4, and occupy all major habitats in the rice field ecosystem. Under flooded anoxic conditions, CH4 is mainly produced by aceticlastic and hydrogenotrophic methanogenesis after fermentative degradation of organic matter or of photosynthates exuded by rice roots (Le Mer and Roger, 2001). Recently, a particular group of methanogens (Methanocellales) was found to be responsible for the production of CH4 from rice photosynthates (Lu and Conrad, 2005). Meanwhile, several strains of Methanocella have been cultivated (Sakai et al., 2007; Lü and Lu, 2012). The Methanocellales together with Methanosarcinaceae, Methanosaetaceae, Methanbacteriales and Methanomicrobiales were also found in the anoxic bulk soil of virtually all rice field soils studied (Ramakrishnan et al., 2001; Watanabe et al., 2010; Wu et al., 2009; Liu et al., 2012).
Besides anaerobic methanogenic Euryarchaeota rice paddy fields are also a habitat of aerobic Thaumarchaeota. Thaumarchaeota are the most recently identified kingdom of Archaea (previous Crenarchaeota 1.1a/b; Pester et al., 2011), quite often characterized by the possession of the ammonia monooxygenase (amoA) gene indicating a potential role in aerobic ammonia oxidation (Könneke et al., 2005; Treusch et al., 2005). Thaumarchaeota populations were found in many soils and sediments at high abundance (Leininger et al., 2006) and also occur in rice fields in surface soil, bulk soil, rhizosphere This article is protected by copyright. All rights reserved.
Accepted Article
soil and on the rice roots (Chen et al., 2011; Ke et al., 2013). Since nitrogen is a limiting factor for plant growth, ammonium-based fertilizers are rapidly depleted in rice field ecosystems. Like in most other soil systems amoA-encoding Thaumarchaeota were usually found to outnumber bacterial ammonia oxidizers (Chen et al., 2008; Ke et al., 2013). The ecological niche of the amoA-encoding Thaumarchaeota may be shaped by a mixotrophic lifestyle (Mussmann et al., 2011; Xu et al., 2012) or a high affinity to ammonia (Herrmann et al., 2008, 2009; Martens-Habbena et al., 2009). On the other hand, nitrogen fertilization was also shown to affect CH4 emission from rice fields (Schütz et al., 1989; Cai et al., 1997; Bodelier et al., 2000; Lindau, 1994; Lindau et al., 1997). Oxidized nitrogen species (nitrate, nitrite, NO, N2O) are acting as inhibitors of CH4 production and of methanogens (Achtnich et al., 1995; Klüber and Conrad 1998a; 1998b; Scheid et al., 2003). Since Thaumarchaeota potentially produce oxidized nitrogen species and coexist with methanogenic archaea in the rice field ecosystem, these two archaeal groups may interact and link carbon and nitrogen metabolism (Bodelier, 2011). However, the coexistence of these two archaeal groups has to our knowledge not yet been studied in one and the same experimental system.
Recently, we analyzed the abundance of amoA-encoding Thaumarchaeota in different soil compartments (surface soil, bulk soil, rhizospheric soil, young roots, old roots) of rice field microcosms as function of season and extent of N fertilization (Ke et al., 2013). Since an analogous study of methanogens is missing so far, we now have also analyzed the abundance and composition of the methanogenic community using the same samples. We asked the following questions: (i) What is the relative abundance of total Archaea, methanogenic archaea and Thaumarchaeota in the different soil compartments, at different nitrogen fertilization levels and sampling time (samplings at different rice growing season); and (ii) how is the community structure of the methanogens affected? We hypothesized the populations of methanogens and Thaumarchaeota display different This article is protected by copyright. All rights reserved.
Accepted Article
dynamics within the rice field ecosystem. For this purpose, we investigated the gene abundance and composition of Archaea, methanogens, and Thaumarchaeota using quantitative PCR, terminal restriction fragment length polymorphism (T-RFLP) fingerprinting and cloning/sequencing targeting archaeal 16S rRNA genes, archaeal amoA and methanogenic mcrA genes.
Experimental Microcosm setup and sample collection Rice microcosms were established in triplicate in the greenhouse using soil from paddy fields of the Chinese Rice Research Institute in Hangzhou. The details of microcosm setup and soil characteristics have been described previously (Ke et al., 2013). Briefly, microcosms were prepared in black polyethylene containers in which a self-made nylon mesh bag was placed into the center of the pot to create two independent soil compartments, the central rooted compartment and the outside non-rooted compartment. Nitrogen treatments included two levels of N (urea) application: 100 and 500 mg N per kg soil. One-third of the total N was applied as basic fertilizer at the beginning of planting and the two other one-third portions were used for topdressing at the beginning and middle of tillering of rice growing season (35 d and 55 d), respectively.
Soil and roots were sampled at three time-points at 50, 70 and 90 d which corresponded to panicle initiation, heading and maturity of rice. Destructive sampling was used to collect surface soil, bulk soil (non-rooted), rhizospheric soil (rooted) and young and old roots. The definition of each compartment of soil and roots and sampling method were described in detail in earlier study (Ke et al., 2013). After sampling, each soil fraction was homogenized and frozen in liquid nitrogen. Root samples were treated in the same way except that they were first pulverized with mortar and pestle after freezing with liquid nitrogen. Soils and roots were stored at - 80 °C for later analysis. This article is protected by copyright. All rights reserved.
Accepted Article
Molecular analyses Genomic DNA was extracted from 0.5 g (wet weight) of soil or 0.3 g of roots with a Fast DNA SPIN kit (MP Biomedicals, LLC) according to the manufacturer’s instructions. The quality and concentration of DNA was determined in a nanodrop spectrophotometer (NanoDrop Technologies, Wilmington, USA). The analysis of terminal restriction fragment length polymorphism (T-RFLP) was done as described earlier (Lueders and Friedrich, 2003; Ke et al., 2013) using the following primer combinations: for archaeal 16S rRNA genes Ar109f/Ar915r, with the reverse primer labeled with FAM (6caboxyfluorescein); and for methanogenic mcrA genes MCRf/MCRr, with the forward primer labeled with FAM. The 16S rRNA gene amplicons were digested with TaqI (Fermentas, St. Leon Rot, Germany), the mcrA gene amplicons with Sau96I (Fermentas), and the products were size-separated in an ABI 373 DNA sequencer (Applera, Darmstadt, Germany). Correspondence analysis of T-RFLP patterns of 16S rRNA gene and mcrA genes was done using CANOCO 4.0 software (Microcomputer Power, Ithaca, NY, USA) as described by Noll et al. (2005).
For cloning and sequencing, PCR amplification used the same primers without FAM label. The mcrA gene amplicons were ligated into pGEM-T-easy vector system (Promega) following the manufacture’s instructions. A total of 179 randomly selected clones of the mcrA gene were sequenced. Nine clone libraries were constructed for analyses of the mcrA gene fragments using DNA samples retrieved from the surface soil, rhizospheric soil and roots at 50 d, 70 d and 90 d of rice growing season, respectively. The gene sequences obtained in this study have been deposited in GenBank under the accession numbers KF181999 to KF182177 for mcrA genes. Quantitative PCR of archaeal 16S rRNA genes, mcrA genes and amoA genes was done according to published protocols (Stubner, 2004; Kemnitz et al., 2005; Jia and Conrad, 2009). The copy numbers of Methanocellales/ Methanomicrobiales methanogens were This article is protected by copyright. All rights reserved.
Accepted Article
calculated by multiplying the copy numbers of archaeal 16S rRNA genes with the relative abundance of the T-RF with 395 bp length representing these groups. Minitab software version 16 (Minitab Inc. PA, USA) and SPSS software version 13.0 (SPASS Inc. Chicago, USA) were used to perform the statistical analysis. Data were log-transformed (using the natural logarithm), or an additional box-cox transformation to meet criteria for parametric ANOVA analyses.
Results Structure of archaeal communities (16S rRNA gene) The composition of the archaeal communities was assessed by T-RFLP analyses targeting archaeal 16S rRNA genes. Thirteen different T-RFs (75, 78, 83, 92, 184, 285, 382, 395, 490, 697, 738, 796, and 814 bp) were detected in the T-RFLP profiles (Fig .1). Most of the T-RFs could be assigned to phylogenetic groups of Archaea according to clone sequences determined in previous studies using the same soil (Peng et al., 2008; Wu et al., 2009; K.Ma, personal communication). Accordingly, the following assignment of individual T-RFs was achieved: 75 bp to Thaumarchaeota or Crenarchaeota; 78 bp to euryarchaeotal RC-V; 83 bp to Methanomicrobiales, Thaumarchaeota or Crenarchaeota; 92 bp to Methanobacteriales, Thaumarchaeota or Crenarchaeota; 184 bp T-RF to Methanosarcinaceae, Thaumarchaeota or Crenarchaeota; 285 bp to Methanosaetaceae; 382 bp to RC-III; and 395 bp to Methanocellales, Methanomicrobiales or Crenarchaeota; 814 bp to Thaumarchaeota or Crenarchaeota. Other fragments (490, 697, 738, and 796 bp) belonged to other archaeal subgroups that were presumably non-methanogenic. T-RFLP patterns were similar among the different soil samples (surface soil, bulk soil, and rhizospheric soil) and among the young and old roots. However, the T-RFLP patterns of soil samples versus roots samples were clearly different (Fig. 1). Correspondence analysis of the T-RFLP profiles confirmed the separation of soil and root samples (Fig. 2a). While the different compartments of soil were all dominated (about 40% relative This article is protected by copyright. All rights reserved.
Accepted Article
abundance) by the 184 bp T-RF (Methanosarcinaceae/ Thaumarchaeota or Crenarchaeota), the roots showed a relatively large abundance of the 92 bp T-RF (Methanobacteriales/ Thaumarchaeota or Crenarchaeota) and the 395 bp T-RF (Methanocellales/ Methanomicrobiales/ Crenarchaeota). Correspondence analysis also showed that the level of nitrogen fertilization and the time point of sampling had no significant effect on the T-RFLP profiles in both soil and root samples. Note, that the assignment of T-RFs was based on literature data rather than on clone sequences retrieved. Several of the T-RFs represent more than one archaeal group, and the assignment may be different for soil compartments and roots samples.
Composition of methanogenic communities (mcrA) The composition of the methanogenic communities was assessed by T-RFLP analyses targeting mcrA genes. Eleven different T-RFs (91, 147, 237, 393, 408, 418, 426, 469, 491, 503 and 534 bp) were detected in the T-RFLP profiles (Fig. 3). To assign the different TRFs to phylogenetic groups of methanoges, nine clone libraries of mcrA genes were constructed from the soil and roots samples at three time-points of rice season (Table 1). From a total of 179 mcrA clone sequences the following T-RFs could be assigned to particular phylogenetic groups: 147 bp to Methanosaetaceae; 237 bp to Methanocellales; 270 bp to unidentified methanogens; 393 bp to Methanosarcinaceae; 408 bp to Methanomicrobiales; 469 bp to Methanobacteriaceae; 503 bp to unidentified methanogens. Most of the clone sequences belonged to Methanosarcinaceae (44%) which were mainly found in soil rather than on roots; to Methanocellales (32%) which were mainly found on roots and in surface soil; and to Methanobacteriaceae (12%) which were mainly found on roots and in rhizospheric soil (Table 1). Sorting the clone sequences by time point of sampling showed no obvious effect (p = 0.993) on the relative distribution of the phylogenetic groups of methanogens (Table 1).
This article is protected by copyright. All rights reserved.
Accepted Article
The mcrA gene T-RFLP patterns also showed that differences appeared between the soil samples and the root samples (Fig. 3). The 393 bp T-RF (Methanosarcinaceae) was predominant in all soil samples (accounting for about 60% relative abundance), while the 237 bp T-RF (Methanocellales) and 503 bp T-RF (unidentified methanogens) were predominant in all roots samples. Moreover, the 147 bp T-RF (Methanosaetaceae) was only detected in soil samples, and the 408 bp T-RF (Methanomicrobiales) was mainly found in the early season (at day 50). Correspondence analysis of mcrA genes T-RFLP profiles showed a clear separation between the soil samples and root samples (Fig. 2b), similarly as found for the archaeal 16S rRNA gene T-RFLP (Fig. 2a).
Abundance of Archaea, Methanogens and Thaumarchaeota Abundance of Archaea, methanogens and Thaumarchaeota were determined by quantifying the copy numbers of 16S rRNA genes, mcrA, and amoA, respectively (Fig. 4). For statistical analysis, the raw data were subject to logarithmic and box-cox transformations, which gave basically the same results. In addition, data from soil samples and roots samples were analyzed separately. ANOVA analysis showed that archaeal 16S rRNA gene and mcrA copy numbers in the paddy soil compartments significantly increased with time, while copy numbers of Thaumarchaeota amoA stayed constant with time (Fig. 4, Table 2). On the rice roots, however, Thaumarchaeota amoA copy numbers also increased with time (Fig. 4, Table 2). The highest copy numbers were always reached at maturity of rice (d 90). In contrast, compartments of soil, age of roots or nitrogen fertilization level did not significantly affect copy numbers in soil or roots and accounted for only a small fraction of the variance (Table 2).
Relative abundance of Methanogens and Thaumarchaeota For the relative abundance of methanogens to Thaumarchaeota we used the ratio of mcrA to amoA gene copy numbers as an index, meanwhile assuming that one copy of This article is protected by copyright. All rights reserved.
Accepted Article
each gene existed per genome. The data show that this ratio generally increased by a factor of 2-7 during the rice growing season in all soil compartments (Fig. 5a). In addition, the data showed that the ratios were significantly higher in the soil than on the roots (pF)
Nitrogen
1
6.1
0.018
6.9
0.012
0.01
0.924
Compartment (Soil)
2
1.9
0.165
4.1
0.026
3.9
0.027
Time Point
2
44.1
F )
F
Pr(>F)
Nitrogen
1
5.1
0.033
6.4
0.018
22.2
Received Date : 07-Jun-2013 Revised Date : 26-Jul-2013 Accepted Date : 26-Jul-2013 Article type : Research Paper Editor : Angela Sessitsch
Different behavior of methanogenic archaea and Thaumarchaeota in rice field microcosms
Xiubin Ke1, 2, 3, Yahai Lu1 and Ralf Conrad2*
1, College of Resources and Environment Sciences, China Agricultural University, Beijing 100193, China 2, Max-Planck-Institute for Terrestrial Microbiology, 35043 Marburg, Germany 3, Biotechnology Research Institute, Chinese Academy of Agricultural Sciences, Key Laboratory of Crop Biotechnology, Ministry of Agriculture, Beijing, China
*Correspondence: R Conrad, Max-Planck-Institute for Terrestrial Microbiology, Department Biogeochemistry, Karl-von-Frisch-Strasse10, D-35043 Marburg, Germany. E-mail: [email protected]
Running title: Methanogens and Thaumarchaeota in rice fields
This article has been accepted for publication and undergone full peer review but has not been through the copyediting, typesetting, pagination and proofreading process, which may lead to differences between this version and the Version of Record. Please cite this article as doi: 10.1111/1574-6941.12188 This article is protected by copyright. All rights reserved.
Accepted Article
Key words: Archaea; Methanogens; Thaumarchaeota; rice field soil; T-RFLP; qPCR
Abstract Archaea in rice fields play an important role in carbon and nitrogen cycling. They comprise methane-producing Euryarchaeota as well as ammonia-oxidizing Thaumarchaeota, but their community structures and population dynamics have not yet been studied in the same system. Different soil compartments (surface, bulk, rhizospheric soil) and ages of roots (young and old roots) at two N fertilization levels and at three time points (the panicle initiation, heading and maturity periods) of the season were assayed by determining the abundance (using qPCR) and composition (using T-RFLP and cloning/sequencing) of archaeal genes (mcrA, amoA, 16S rRNA). The community of total Archaea in soil and root samples mainly consisted of the methanogens and the Thaumarchaeota and their abundance increased over the season. Methanogens proliferated everywhere, but Thaumarchaeota only on the roots and in response to nitrogen fertilization. The community structures of Archaea, methanogens and Thaumarchaeota, were different in soil and root samples indicating niche differentiation. While Methanobacteriales were generally present, Methanosarcinaceae and Methanocellales were the dominant methanogens in soil and root samples, respectively. The results emphasize the specific colonization of roots by two ecophysiologically different groups of archaea which may belong to the core root biome.
One sentence summary The community structures of Archaea, methanogens and Thaumarchaeota, were different in soil and roots of rice microcosms indicating niche differentiation and the specific colonization of roots.
This article is protected by copyright. All rights reserved.
Accepted Article
Introduction Rice paddy fields are an important source of methane and contribute about 20% to the global methane emission (Conrad, 2009). Rice fields are managed wetland ecosystems with bulk soil, surface soil, rhizosphere soil and roots as distinct habitats (Liesack et al., 2000). The gas vascular system of the rice plants allows the ventilation of gases (e.g., O2, CH4) between soil and atmosphere. Therefore, O2 is only available in the uppermost surface layer of the soil and in a shallow layer at the root surface, while bulk soil is completely devoid of O2. Methanogenic Euryarchaeota play an important role in the anaerobic degradation of organic matter and the production of CH4, and occupy all major habitats in the rice field ecosystem. Under flooded anoxic conditions, CH4 is mainly produced by aceticlastic and hydrogenotrophic methanogenesis after fermentative degradation of organic matter or of photosynthates exuded by rice roots (Le Mer and Roger, 2001). Recently, a particular group of methanogens (Methanocellales) was found to be responsible for the production of CH4 from rice photosynthates (Lu and Conrad, 2005). Meanwhile, several strains of Methanocella have been cultivated (Sakai et al., 2007; Lü and Lu, 2012). The Methanocellales together with Methanosarcinaceae, Methanosaetaceae, Methanbacteriales and Methanomicrobiales were also found in the anoxic bulk soil of virtually all rice field soils studied (Ramakrishnan et al., 2001; Watanabe et al., 2010; Wu et al., 2009; Liu et al., 2012).
Besides anaerobic methanogenic Euryarchaeota rice paddy fields are also a habitat of aerobic Thaumarchaeota. Thaumarchaeota are the most recently identified kingdom of Archaea (previous Crenarchaeota 1.1a/b; Pester et al., 2011), quite often characterized by the possession of the ammonia monooxygenase (amoA) gene indicating a potential role in aerobic ammonia oxidation (Könneke et al., 2005; Treusch et al., 2005). Thaumarchaeota populations were found in many soils and sediments at high abundance (Leininger et al., 2006) and also occur in rice fields in surface soil, bulk soil, rhizosphere This article is protected by copyright. All rights reserved.
Accepted Article
soil and on the rice roots (Chen et al., 2011; Ke et al., 2013). Since nitrogen is a limiting factor for plant growth, ammonium-based fertilizers are rapidly depleted in rice field ecosystems. Like in most other soil systems amoA-encoding Thaumarchaeota were usually found to outnumber bacterial ammonia oxidizers (Chen et al., 2008; Ke et al., 2013). The ecological niche of the amoA-encoding Thaumarchaeota may be shaped by a mixotrophic lifestyle (Mussmann et al., 2011; Xu et al., 2012) or a high affinity to ammonia (Herrmann et al., 2008, 2009; Martens-Habbena et al., 2009). On the other hand, nitrogen fertilization was also shown to affect CH4 emission from rice fields (Schütz et al., 1989; Cai et al., 1997; Bodelier et al., 2000; Lindau, 1994; Lindau et al., 1997). Oxidized nitrogen species (nitrate, nitrite, NO, N2O) are acting as inhibitors of CH4 production and of methanogens (Achtnich et al., 1995; Klüber and Conrad 1998a; 1998b; Scheid et al., 2003). Since Thaumarchaeota potentially produce oxidized nitrogen species and coexist with methanogenic archaea in the rice field ecosystem, these two archaeal groups may interact and link carbon and nitrogen metabolism (Bodelier, 2011). However, the coexistence of these two archaeal groups has to our knowledge not yet been studied in one and the same experimental system.
Recently, we analyzed the abundance of amoA-encoding Thaumarchaeota in different soil compartments (surface soil, bulk soil, rhizospheric soil, young roots, old roots) of rice field microcosms as function of season and extent of N fertilization (Ke et al., 2013). Since an analogous study of methanogens is missing so far, we now have also analyzed the abundance and composition of the methanogenic community using the same samples. We asked the following questions: (i) What is the relative abundance of total Archaea, methanogenic archaea and Thaumarchaeota in the different soil compartments, at different nitrogen fertilization levels and sampling time (samplings at different rice growing season); and (ii) how is the community structure of the methanogens affected? We hypothesized the populations of methanogens and Thaumarchaeota display different This article is protected by copyright. All rights reserved.
Accepted Article
dynamics within the rice field ecosystem. For this purpose, we investigated the gene abundance and composition of Archaea, methanogens, and Thaumarchaeota using quantitative PCR, terminal restriction fragment length polymorphism (T-RFLP) fingerprinting and cloning/sequencing targeting archaeal 16S rRNA genes, archaeal amoA and methanogenic mcrA genes.
Experimental Microcosm setup and sample collection Rice microcosms were established in triplicate in the greenhouse using soil from paddy fields of the Chinese Rice Research Institute in Hangzhou. The details of microcosm setup and soil characteristics have been described previously (Ke et al., 2013). Briefly, microcosms were prepared in black polyethylene containers in which a self-made nylon mesh bag was placed into the center of the pot to create two independent soil compartments, the central rooted compartment and the outside non-rooted compartment. Nitrogen treatments included two levels of N (urea) application: 100 and 500 mg N per kg soil. One-third of the total N was applied as basic fertilizer at the beginning of planting and the two other one-third portions were used for topdressing at the beginning and middle of tillering of rice growing season (35 d and 55 d), respectively.
Soil and roots were sampled at three time-points at 50, 70 and 90 d which corresponded to panicle initiation, heading and maturity of rice. Destructive sampling was used to collect surface soil, bulk soil (non-rooted), rhizospheric soil (rooted) and young and old roots. The definition of each compartment of soil and roots and sampling method were described in detail in earlier study (Ke et al., 2013). After sampling, each soil fraction was homogenized and frozen in liquid nitrogen. Root samples were treated in the same way except that they were first pulverized with mortar and pestle after freezing with liquid nitrogen. Soils and roots were stored at - 80 °C for later analysis. This article is protected by copyright. All rights reserved.
Accepted Article
Molecular analyses Genomic DNA was extracted from 0.5 g (wet weight) of soil or 0.3 g of roots with a Fast DNA SPIN kit (MP Biomedicals, LLC) according to the manufacturer’s instructions. The quality and concentration of DNA was determined in a nanodrop spectrophotometer (NanoDrop Technologies, Wilmington, USA). The analysis of terminal restriction fragment length polymorphism (T-RFLP) was done as described earlier (Lueders and Friedrich, 2003; Ke et al., 2013) using the following primer combinations: for archaeal 16S rRNA genes Ar109f/Ar915r, with the reverse primer labeled with FAM (6caboxyfluorescein); and for methanogenic mcrA genes MCRf/MCRr, with the forward primer labeled with FAM. The 16S rRNA gene amplicons were digested with TaqI (Fermentas, St. Leon Rot, Germany), the mcrA gene amplicons with Sau96I (Fermentas), and the products were size-separated in an ABI 373 DNA sequencer (Applera, Darmstadt, Germany). Correspondence analysis of T-RFLP patterns of 16S rRNA gene and mcrA genes was done using CANOCO 4.0 software (Microcomputer Power, Ithaca, NY, USA) as described by Noll et al. (2005).
For cloning and sequencing, PCR amplification used the same primers without FAM label. The mcrA gene amplicons were ligated into pGEM-T-easy vector system (Promega) following the manufacture’s instructions. A total of 179 randomly selected clones of the mcrA gene were sequenced. Nine clone libraries were constructed for analyses of the mcrA gene fragments using DNA samples retrieved from the surface soil, rhizospheric soil and roots at 50 d, 70 d and 90 d of rice growing season, respectively. The gene sequences obtained in this study have been deposited in GenBank under the accession numbers KF181999 to KF182177 for mcrA genes. Quantitative PCR of archaeal 16S rRNA genes, mcrA genes and amoA genes was done according to published protocols (Stubner, 2004; Kemnitz et al., 2005; Jia and Conrad, 2009). The copy numbers of Methanocellales/ Methanomicrobiales methanogens were This article is protected by copyright. All rights reserved.
Accepted Article
calculated by multiplying the copy numbers of archaeal 16S rRNA genes with the relative abundance of the T-RF with 395 bp length representing these groups. Minitab software version 16 (Minitab Inc. PA, USA) and SPSS software version 13.0 (SPASS Inc. Chicago, USA) were used to perform the statistical analysis. Data were log-transformed (using the natural logarithm), or an additional box-cox transformation to meet criteria for parametric ANOVA analyses.
Results Structure of archaeal communities (16S rRNA gene) The composition of the archaeal communities was assessed by T-RFLP analyses targeting archaeal 16S rRNA genes. Thirteen different T-RFs (75, 78, 83, 92, 184, 285, 382, 395, 490, 697, 738, 796, and 814 bp) were detected in the T-RFLP profiles (Fig .1). Most of the T-RFs could be assigned to phylogenetic groups of Archaea according to clone sequences determined in previous studies using the same soil (Peng et al., 2008; Wu et al., 2009; K.Ma, personal communication). Accordingly, the following assignment of individual T-RFs was achieved: 75 bp to Thaumarchaeota or Crenarchaeota; 78 bp to euryarchaeotal RC-V; 83 bp to Methanomicrobiales, Thaumarchaeota or Crenarchaeota; 92 bp to Methanobacteriales, Thaumarchaeota or Crenarchaeota; 184 bp T-RF to Methanosarcinaceae, Thaumarchaeota or Crenarchaeota; 285 bp to Methanosaetaceae; 382 bp to RC-III; and 395 bp to Methanocellales, Methanomicrobiales or Crenarchaeota; 814 bp to Thaumarchaeota or Crenarchaeota. Other fragments (490, 697, 738, and 796 bp) belonged to other archaeal subgroups that were presumably non-methanogenic. T-RFLP patterns were similar among the different soil samples (surface soil, bulk soil, and rhizospheric soil) and among the young and old roots. However, the T-RFLP patterns of soil samples versus roots samples were clearly different (Fig. 1). Correspondence analysis of the T-RFLP profiles confirmed the separation of soil and root samples (Fig. 2a). While the different compartments of soil were all dominated (about 40% relative This article is protected by copyright. All rights reserved.
Accepted Article
abundance) by the 184 bp T-RF (Methanosarcinaceae/ Thaumarchaeota or Crenarchaeota), the roots showed a relatively large abundance of the 92 bp T-RF (Methanobacteriales/ Thaumarchaeota or Crenarchaeota) and the 395 bp T-RF (Methanocellales/ Methanomicrobiales/ Crenarchaeota). Correspondence analysis also showed that the level of nitrogen fertilization and the time point of sampling had no significant effect on the T-RFLP profiles in both soil and root samples. Note, that the assignment of T-RFs was based on literature data rather than on clone sequences retrieved. Several of the T-RFs represent more than one archaeal group, and the assignment may be different for soil compartments and roots samples.
Composition of methanogenic communities (mcrA) The composition of the methanogenic communities was assessed by T-RFLP analyses targeting mcrA genes. Eleven different T-RFs (91, 147, 237, 393, 408, 418, 426, 469, 491, 503 and 534 bp) were detected in the T-RFLP profiles (Fig. 3). To assign the different TRFs to phylogenetic groups of methanoges, nine clone libraries of mcrA genes were constructed from the soil and roots samples at three time-points of rice season (Table 1). From a total of 179 mcrA clone sequences the following T-RFs could be assigned to particular phylogenetic groups: 147 bp to Methanosaetaceae; 237 bp to Methanocellales; 270 bp to unidentified methanogens; 393 bp to Methanosarcinaceae; 408 bp to Methanomicrobiales; 469 bp to Methanobacteriaceae; 503 bp to unidentified methanogens. Most of the clone sequences belonged to Methanosarcinaceae (44%) which were mainly found in soil rather than on roots; to Methanocellales (32%) which were mainly found on roots and in surface soil; and to Methanobacteriaceae (12%) which were mainly found on roots and in rhizospheric soil (Table 1). Sorting the clone sequences by time point of sampling showed no obvious effect (p = 0.993) on the relative distribution of the phylogenetic groups of methanogens (Table 1).
This article is protected by copyright. All rights reserved.
Accepted Article
The mcrA gene T-RFLP patterns also showed that differences appeared between the soil samples and the root samples (Fig. 3). The 393 bp T-RF (Methanosarcinaceae) was predominant in all soil samples (accounting for about 60% relative abundance), while the 237 bp T-RF (Methanocellales) and 503 bp T-RF (unidentified methanogens) were predominant in all roots samples. Moreover, the 147 bp T-RF (Methanosaetaceae) was only detected in soil samples, and the 408 bp T-RF (Methanomicrobiales) was mainly found in the early season (at day 50). Correspondence analysis of mcrA genes T-RFLP profiles showed a clear separation between the soil samples and root samples (Fig. 2b), similarly as found for the archaeal 16S rRNA gene T-RFLP (Fig. 2a).
Abundance of Archaea, Methanogens and Thaumarchaeota Abundance of Archaea, methanogens and Thaumarchaeota were determined by quantifying the copy numbers of 16S rRNA genes, mcrA, and amoA, respectively (Fig. 4). For statistical analysis, the raw data were subject to logarithmic and box-cox transformations, which gave basically the same results. In addition, data from soil samples and roots samples were analyzed separately. ANOVA analysis showed that archaeal 16S rRNA gene and mcrA copy numbers in the paddy soil compartments significantly increased with time, while copy numbers of Thaumarchaeota amoA stayed constant with time (Fig. 4, Table 2). On the rice roots, however, Thaumarchaeota amoA copy numbers also increased with time (Fig. 4, Table 2). The highest copy numbers were always reached at maturity of rice (d 90). In contrast, compartments of soil, age of roots or nitrogen fertilization level did not significantly affect copy numbers in soil or roots and accounted for only a small fraction of the variance (Table 2).
Relative abundance of Methanogens and Thaumarchaeota For the relative abundance of methanogens to Thaumarchaeota we used the ratio of mcrA to amoA gene copy numbers as an index, meanwhile assuming that one copy of This article is protected by copyright. All rights reserved.
Accepted Article
each gene existed per genome. The data show that this ratio generally increased by a factor of 2-7 during the rice growing season in all soil compartments (Fig. 5a). In addition, the data showed that the ratios were significantly higher in the soil than on the roots (pF)
Nitrogen
1
6.1
0.018
6.9
0.012
0.01
0.924
Compartment (Soil)
2
1.9
0.165
4.1
0.026
3.9
0.027
Time Point
2
44.1
F )
F
Pr(>F)
Nitrogen
1
5.1
0.033
6.4
0.018
22.2
