SEARCH

SEARCH BY CITATION

Keywords:

  • ammonia-oxidizing archaea;
  • ammonia-oxidizing bacteria;
  • nitrification;
  • soil pH

Abstract

  1. Top of page
  2. Abstract
  3. Introduction
  4. Material and methods
  5. Results
  6. Discussion
  7. Acknowledgements
  8. References
  9. Supporting Information

Nitrification is a central component of the global nitrogen cycle. Ammonia oxidation, the first step of nitrification, is performed in terrestrial ecosystems by both ammonia-oxidizing bacteria (AOB) and ammonia-oxidizing archaea (AOA). Published studies indicate that soil pH may be a critical factor controlling the relative abundances of AOA and AOB communities. In order to determine the relative contributions of AOA and AOB to ammonia oxidation in two agricultural acidic Scottish soils (pH 4.5 and 6), the influence of acetylene (a nitrification inhibitor) was investigated during incubation of soil microcosms at 20 °C for 1 month. High rates of nitrification were observed in both soils in the absence of acetylene. Quantification of respective amoA genes (a key functional gene for ammonia oxidizers) demonstrated significant growth of AOA, but not AOB. A significant positive relationship was found between nitrification rate and AOA, but not AOB growth. AOA growth was inhibited in the acetylene-containing microcosms. Moreover, AOA transcriptional activity decreased significantly in the acetylene-containing microcosms compared with the control, whereas no difference was observed for the AOB transcriptional activity. Consequently, growth and activity of only archaeal but not bacterial ammonia oxidizer communities strongly suggest that AOA, but not AOB, control nitrification in these two acidic soils.


Introduction

  1. Top of page
  2. Abstract
  3. Introduction
  4. Material and methods
  5. Results
  6. Discussion
  7. Acknowledgements
  8. References
  9. Supporting Information

Nitrification plays an important role in the global nitrogen cycle and in the majority of terrestrial environments. The first and rate-limiting stage of nitrification is the oxidation of ammonia to nitrite via intermediates (Prosser & Embley, 2002). Until recently, autotrophic ammonia-oxidizing bacteria (AOB) were considered to be the major contributors to ammonia oxidation in soil with all cultivated strains belonging to a monophyletic group within the Betaproteobacteria. More recently, substantial discoveries, including metagenomic studies (Venter et al., 2004; Treusch et al., 2005), laboratory cultivation and strain isolation (Könneke et al., 2005; de la Torre et al., 2008; Hatzenpichler et al., 2008), nitrification inhibition coupled with molecular analysis (Offre et al., 2009) and physiological studies (Martens-Habbena et al., 2009) provide evidence that archaea belonging to the Thaumarchaeota lineage (Brochier-Armanet et al., 2008; Spang et al., 2010) can also perform ammonia oxidation.

A key functional enzyme in both bacterial and archaeal ammonia oxidizers is ammonia monooxygenase (AMO), which oxidizes ammonia to hydroxylamine (in AOB). Acetylene (C2H2) is a rapid (10–15 min) and efficient inhibitor of bacterial ammonia oxidation (Hynes & Knowles, 1978), acting as a suicide substrate of AMO. It is commonly used as an inhibitor of autotrophic nitrification in experimental studies (de Boer & Kowalchuk, 2001) and in most soils inhibits nitrification at a low concentration (e.g. 10 Pa) (Berg et al., 1982; de Boer et al., 1991; Offre et al., 2009) under aerobic conditions (Hyman & Wood, 1985; Schmidt & Bock, 1998). There is currently no published evidence for inhibition of cultivated archaeal ammonia oxidizers by acetylene, but microcosm experiments have demonstrated inhibition in soils in which ammonia oxidation is dominated by archaea (Offre et al., 2009).

Both archaeal and bacterial ammonia oxidizers are found in the majority of terrestrial ecosystems, including agricultural, grassland, forest and alpine soils (Leininger et al., 2006; Boyle-Yarwood et al., 2008; Nicol et al., 2008), and quantification of respective amoA genes indicates greater abundance of archaeal ammonia oxidizers in many soils (Leininger et al., 2006; He et al., 2007; Shen et al., 2008; Hai et al., 2009; Jia & Conrad, 2009). Soil typically contains 104–106 bacterial ammonia oxidizers g−1 (Phillips et al., 2000; Okano et al., 2004; Kolb et al., 2005) and >107archaeal ammonia oxidizers g−1 (Leininger et al., 2006), but particular conditions, such as heavy metal contamination (Mertens et al., 2009), plant community composition (Boyle-Yarwood et al., 2008) or high nutrient concentration (Di et al., 2009) may lead to greater relative abundance of bacterial ammonia oxidizers. Soil pH is also believed to influence the distribution and activity of archaeal and bacterial ammonia oxidizers (Nicol et al., 2008) and is an important determinant of bacterial diversity and community structure on a global scale (Fierer & Jackson, 2006). Bacterial and archaeal amoA genes have been recovered from soils with pH values ranging from 3.7 (He et al., 2007) to 8.65 (Shen et al., 2008). Soil pH also appears to lead to selection for different bacterial and archaeal communities (Stephen et al., 1998; Nugroho et al., 2007; Nicol et al., 2008; Lehtovirta et al., 2009), suggesting adaptation and selection of particular phylogenetic groups in acidic soils (see de Boer & Kowalchuk, 2001).

One mechanism by which soil pH can influence ammonia oxidizer activity and community structure is by decreasing the availability of ammonia by increasing ionization to ammonium as pH is decreased (de Boer & Kowalchuk, 2001), although ammonia availability will be low even in neutral soils. Reduced ammonia availability could increase selection for archaea, as physiological studies of the isolated archaeal ammonia oxidizer (Nitrosopumilus maritimus) indicate a half-saturation constant for ammonia that is significantly lower than those for cultivated bacterial ammonia oxidizers (Martens-Habbena et al., 2009). The goal of the study was therefore to determine the relative contributions of archaea and bacteria to ammonia oxidation in acidic soils, where ammonia availability will be reduced. Ammonia oxidizer activity, abundance and community structure were investigated in microcosms containing soil from two acidic plots of a Scottish agricultural soil that has been the subject of previous investigations into the influence of soil pH on bacterial and archaeal ammonia oxidizer communities (Stephen et al., 1996, 1998; Nicol et al., 2008) and nitrification activity (Killham, 1990).

Material and methods

  1. Top of page
  2. Abstract
  3. Introduction
  4. Material and methods
  5. Results
  6. Discussion
  7. Acknowledgements
  8. References
  9. Supporting Information

Sampling and microcosms

Sandy loam soil samples (Countesswells series) with a pH of 4.5 and 6.0 were collected from a pH gradient of an experimental agricultural soil (Scottish Agricultural College, Craibstone, Scotland, Grid reference NJ872104) that had been maintained for 49 years (Kemp et al., 1992). Three 200-g soil samples were collected from each plot and then pooled, sieved (3.35-mm mesh size) and stored at 4 °C for 1 week before construction of microcosms. Water content, determined as weight loss after drying at 100 °C for 24 h, was 24.4% (SE 0.28) and 24.8% (SE 0.09) for pH 4.5 and 6 soils, respectively. Individual microcosms consisted of 125-mL sterile serum bottles containing 10 g of fresh soil sealed with rubber stoppers and metal crimp tops. Acetylene was added to the headspace of half of the microcosms at a final partial pressure of 10 Pa (0.01%) and microcosms were incubated at 20 °C in the dark. Three replicate microcosms for each treatment were destructively sampled after incubation for 10, 20 and 27 days. Aerobic conditions were maintained by briefly removing seals at intervals of 3–4 days, resealing and re-establishing acetylene partial pressure.

Nitrification measurement

Soil mineral N (ammonia and combined nitrite+nitrate concentrations) was determined colorimetrically by flow injection analysis (FIA star 5010 Analyser, Foss Tecator AB, Höganäs, Sweden) (Allen, 1989) after extraction from 4 g of wet soil in 32 mL of 1M KCl. As nitrite concentration was negligible, results are expressed as nitrate concentration only, and as μg NO3-N g−1 dry soil. Net nitrification rate, defined as the net production of nitrate without amendment of soil with ammonium (Killham, 1990), was calculated for the two soils by linear regression of nitrate concentration with time.

Nucleic acid extraction, cDNA synthesis and denaturing gradient gel electrophoresis (DGGE)

Nucleic acids were extracted from 0.5 g of soil as described by Griffiths et al. (2000). RNA reverse transcription was performed using SuperScript II reverse transcriptase (Invitrogen, Paisley, UK) and random hexamer primers (Invitrogen) as described previously (Nicol et al., 2008) except that the RNA column purification step was omitted, analysis was performed on 1 μL nondiluted extract and 1 μL of RNasin (Promega, Southampton, UK) was added during reverse transcription. PCR assays and DGGE analysis were as described by Nicol et al. (2008). Briefly, PCR amplification of archaeal and bacterial amoA genes was performed using primers crenamoA23F and crenamoA616R without the requirement of a GC clamp (Tourna et al., 2008) and primers amoA-1R and amoA-2R-GC (Rotthauwe et al., 1997), respectively. A nested PCR strategy was used for DGGE analysis of bacterial ammonia oxidizer and thaumarchaeal 16S rRNA gene sequences. Bacterial ammonia oxidizer 16S rRNA gene primary amplification used primers CTO189f and CTO654r (Kowalchuk et al., 1997) and secondary amplification used primers P3 (357f-GC) and P2 (518r) (Muyzer et al., 1993). Archaeal 16S rRNA gene primary amplification used primers A109f (Großkopf et al., 1998) and a modified version of 1492r (Nicol et al., 2008) and secondary amplification used thaumarchaeal primers 771f and 957r (Ochsenreiter et al., 2003) with primer 957r containing additionally the GC clamp of primer P3 (Muyzer et al., 1993). DGGE gels contained 8% (w/v) polyacrylamide and a linear gradient of 35–70% denaturant for 16S rRNA gene assays and 15–55% for amoA assays; electrophoresis was performed at 60 °C for 900 min at 100 V.

Quantitative PCR analysis of amoA genes and transcripts

Abundance of bacterial amoA genes and transcripts was determined as described by Leininger et al. (2006). Each reaction was performed in a 25-μL volume containing 0.2 mg mL−1 bovine serum albumin (BSA), 1.5 μM of each primer (amoA-1R and amoA-2R, Rotthauwe et al., 1997), 10 μL of QuantiTect SYBR Green PCR Master Mix (Qiagen, Crawley, UK) and 5 μL of nucleic acid extract. Either 25 ng of DNA or 20 × diluted cDNA was added per reaction. A dilution series (101–107amoA copies) consisting of a known amount of Nitrosospira multiformis ATCC25196 genomic DNA was used as a standard. Amplification efficiencies of 86.3–98% were obtained, with r2 values >0.98. Cycling conditions were 15 min at 95 °C, 46 cycles of 1 min at 95 °C, 1 min at 54 °C, 1 min at 72 °C, followed by a plate read, incubation for 5 s at 76 °C and a second plate read followed by a final extension reaction for 10 min at 72 °C. Amplification was performed in a DNA Engine OPTICON 2 System (GRI Ltd, Braintree, UK). Amplification of bacterial amoA genes was also performed using the QuantiFastTM qPCR master mix (Qiagen) with the same reaction mix (except for the primers) and the PCR cycling conditions described below for the archaeal amoA genes. Because of the similarity between the two methods, only the first set of data is presented.

Archaeal amoA genes and transcripts were quantified using primers crenamoA23F and crenamoA616R (Tourna et al., 2008) for amplification. Each reaction was performed in a 25-μL volume containing 0.2 mg mL−1 BSA, 1 μM of each primer and 12.5 μL of QuantiFastTM qPCR master mix (Qiagen) and 5 μL of nucleic acid. Either 25 ng of DNA or 5 × or 20 × diluted cDNA was added per reaction. Cycling conditions were 15 min at 95 °C, 35 cycles of 10 s at 95 °C, 30 s at 60 °C, an incubation at 82 °C for 6 s, a plate read followed by final extension reaction for 10 min at 60 °C. Standards consisted of a dilution series (102–108amoA copies) of a 1934-bp PCR product containing the amoA gene of the fosmid 54d9 (Treusch et al., 2005) amplified with primers 54d9 : orf37f (TGT CCT TCA GCA GTT TG) and 54d9 : orf40r (TTT GCA GCC GAA TCT ACA CCA). Amplification efficiencies for archaeal amoA qPCR ranged from 102% to 105% and r2 values were >0.98. Amplification was performed in a BioRad MyIQ Single-Color Real-Time PCR Detection System (BioRad, Hertfordshire, UK). For all qPCR assays, melting curve analysis and standard 1% agarose gel electrophoresis of amplicons were checked at the end of each run.

Statistical analysis

Nitrification rates were compared by ancova analysis of slopes using minitab 15 (Minitab Ltd, Coventry, UK). The effects of treatment (presence or absence of acetylene) and incubation time on gene abundance and soil mineral N were tested by comparison of means by anova using statistica 6 (StatSoft Ltd, Bedford, UK) and least-significant-difference Fisher post hoc tests on each measurement were then used to assess significant differences shown on the graphs. When necessary, measurements were log-transformed to normalize residuals and maintain homoscedasticity.

Results

  1. Top of page
  2. Abstract
  3. Introduction
  4. Material and methods
  5. Results
  6. Discussion
  7. Acknowledgements
  8. References
  9. Supporting Information

Soil mineral nitrogen measurements and net nitrification activity

In microcosms with no acetylene added, nitrate concentration increased continuously in both soils throughout incubation for 1 month, without amendment of ammonia, while ammonia concentration showed no significant change from initial values of 0.86 (SE 0.04) and 0.59 (SE 0.10) μg NH4+-N g−1 soil for pH 4.5 and 6 soils, respectively (Fig. 1). Nitrate concentration increased linearly and net nitrification rate, determined by linear regression of nitrate concentration with time, was 0.76 (r2=0.996) and 0.56 (r2=0.997) μg NO3-N g−1 soil per day for pH 4.5 and 6 soils, respectively. The net nitrification rate in the pH 4.5 soil was slightly higher than in the pH 6 soil (P<0.001). In contrast, addition of acetylene to microcosms led to accumulation of ammonia in both soils and constant levels of nitrate, which were not significantly different from the initial values of 3.36 (SE 0.08) and 4.40 (SE 0.10) μg NO3-N g−1 soil for pH 4.5 and 6 soils, respectively.

image

Figure 1.  Changes in ammonia and nitrate concentrations during incubation of microcosms containing soils of pH 4.5 (a, b) or pH 6 (c, d) for 27 days in the presence (dotted line) or absence (solid line) of 10 Pa acetylene. Plotted values are means and SEs from triplicate microcosms.

Download figure to PowerPoint

DGGE analysis of archaeal and bacterial community structure and transcriptional activity

Putative archaeal ammonia oxidizer communities were characterized by molecular analysis of respective amoA and 16S rRNA genes, while putative bacterial ammonia oxidizers were characterized by analysis of 16S rRNA genes only, due to difficulties in obtaining suitable amoA products with the GC clamp on primers used for DGGE analysis. Reproducible DGGE profiles were obtained from replicate microcosms but the community structure of archaeal and bacterial ammonia oxidizers did not change during incubation and was unaffected by addition of acetylene (Supporting Information, Fig. S1).

Archaeal amoA gene abundance increased significantly (1.6-fold) in both soils during incubation for 1 month, with the highest level being reached after incubation for 20 days (Fig. 2a and c). There was no detectable increase in archaeal amoA gene abundance in the presence of acetylene in either soil (Fig. 2a and c). There was no significant increase in bacterial amoA gene abundance in either pH 4.5 and 6 soils in the absence or presence of acetylene (Fig. 2b and d). Addition of acetylene significantly reduced archaeal amoA gene transcript abundance (P<10−3 and <10−1 for pH 4.5 and 6 soils, respectively; Fig. 2e and g), but not bacterial transcript abundance (P=0.407 and 0.952 for pH 4.5 and 6 soils, respectively, Fig. 2f and h). The ratio of archaeal : bacterial abundance increased from initial values of 1.8 and 0.7 to final ratios (day 27) of 3.7 and 3.5 for pH 4.5 and 6 soils, respectively.

image

Figure 2.  Changes in abundances of archaeal (a, c, e, g) and bacterial (b, d, f, h) amoA genes (a–d) and transcripts (e–h) during incubation of microcosms containing soils of pH 4.5 (a, b, e, f) or pH 6 (c, d, g, h) for 27 days in the presence or absence of 10 Pa acetylene. Abundance is expressed per g dry soil. Within each graph, different letters above bars indicate significantly different means (P<0.05). Means and SEs were calculated from triplicate microcosms.

Download figure to PowerPoint

Relationship between nitrification and ammonia oxidizer abundance

A significant positive relationship was observed between nitrate concentration and archaeal but not bacterial ammonia oxidizer abundance (Fig. 3). The following linear regression equations were obtained for the archaeal communities and tested against the null hypothesis (pH 4.5: y=1.4 × 106x+4.2 × 107; r2=0.56; P=0.005) (pH 6: y=5 × 105x+1.4 × 107; r2=0.47; P=0.01). There was no significant correlation (pH 4.5 soil) or a significant negative correlation (pH 6 soil) for the bacterial communities. No relationship was observed between nitrification and transcript abundance or transcript : gene abundance ratio for both ammonia oxidizer archaea and bacteria (data not shown).

image

Figure 3.  Relationship between nitrate concentration and archaeal (a, c) or bacterial (b, d) amoA gene abundance using data plotted in Figs 1 and 2. Means and SEs were calculated from triplicate microcosms.

Download figure to PowerPoint

Discussion

  1. Top of page
  2. Abstract
  3. Introduction
  4. Material and methods
  5. Results
  6. Discussion
  7. Acknowledgements
  8. References
  9. Supporting Information

Nitrification rates

High rates of nitrification were observed in plots of an agricultural soil that have been maintained at pH 4.5 and 6 since 1961. Rates were similar to those observed previously in this soil (Tourna et al., 2008) and nitrification was inhibited by acetylene, implying AMO-dependent (and presumably autotrophic) nitrification, as frequently observed in many acid soils (de Boer & Kowalchuk, 2001). The net nitrification rate was higher in the more acidic soil, which is surprising, as ammonia oxidation in pure cultures decreases with pH (Allison & Prosser, 1991). In addition, nitrification potential decreased with decreasing pH in these soil plots (Killham, 1990), but was determined in soil slurries buffered at pH 6 (Killham, 1990), which might not favour organisms selected at pH 4.5, possibly because they prefer low ammonia concentration. However, Booth et al. (2005), in a survey of a range of forest, agricultural and grassland soils, reported a slight but significant negative relationship between net nitrification rate and soil pH, and a positive relationship between net nitrification and mineralization rates. Microcosms in this study were not amended with ammonia, and nitrate will therefore have been produced through oxidation of ammonia generated by mineralization of organic matter. Ammonia concentration was always low and, under these conditions, the influence of pH on mineralization may be of greater importance for nitrification rate than direct effects of soil pH on ammonia oxidizer activity.

Archaeal and bacterial ammonia oxidizer community structure and abundance

As observed previously (Nicol et al., 2008), archaeal ammonia oxidizers outnumbered bacterial ammonia oxidizers in these soils, particularly after incubation for 1 month, but the archaea : bacteria ratio was lower than reported previously (Nicol et al., 2008). Ammonia oxidizer abundance will vary temporally, seasonally and following fertilization (e.g. He et al., 2007), and ratios fall within those found in a range of soils (1.02–232; Leininger et al., 2006; He et al., 2007).

Nitrification was associated with archaeal ammonia oxidizer growth (increases in amoA gene abundance) during incubation of both soils and nitrification in both was completely inhibited by acetylene. There was no evidence of growth of bacterial ammonia oxidizers, in the presence or absence of acetylene, which is consistent with lack of growth of AOB in laboratory culture at pH values <6.5 (Allison & Prosser, 1991; Jiang & Bakken, 1999), although some strains are able to adapt to lower pH (de Boer et al., 1995). In addition, bacterial ammonia oxidizer amoA transcript abundances were unaffected by acetylene, suggesting that cellular mRNA levels were associated with survival and basal metabolism, rather than with nitrification. Furthermore, the turnover rate of these transcripts may be extremely low as acetylene blocks all ATP generation in AOB. This contrasts with archaeal amoA transcripts, which decreased in abundance in the presence of acetylene in both soils.

There was no evidence for changes in bacterial ammonia oxidizer community structure during nitrification, which is consistent with the lack of growth and activity. Growth of archaeal ammonia oxidizers was also not associated with changes in community structure. However, it is possible that bacterial and/or archaeal community changes occurred below the limits of detection for relative abundance and taxonomic precision provided by DGGE. Nevertheless, environmental conditions used in this microcosm study were potentially not highly selective for both ammonia oxidizer communities, as suggested by the archaeal community stability during growth. Changes in ammonia oxidizer communities in these soils have generally been associated with higher incubation temperature (Tourna et al., 2008; Offre et al., 2009) or manipulation of pH (Nicol et al., 2008).

Relative contribution of archaea and bacteria to nitrification rate

Nitrate production was strongly correlated with abundance of archaeal but not bacterial ammonia oxidizers. The only other report in which a contribution of AOB to nitrification could not be detected was an investigation of nitrification in Craibstone pH 7.5 soil (Offre et al., 2009). Other studies have reported a correlation between net nitrification rate and bacterial amoA abundance (Di et al., 2009; Jia & Conrad, 2009) or between potential nitrification rate and archaeal and bacterial amoA abundance (He et al., 2007). Such differences in the relative contributions of bacterial and archaeal ammonia oxidizers may be due to a number of factors (Prosser & Nicol, 2008), but there is evidence for preference of archaeal ammonia oxidizers for low ammonia concentration (Hatzenpichler et al., 2008; Martens-Habbena et al., 2009). Dominance of ammonia oxidation by archaea in acidic soils in this study, in which the low measured concentration of ammonium reflects even lower availability of ammonia, suggest that archaeal ammonia oxidizers may dominate where ammonia is produced at continuous low rates, rather than through input of high concentrations as inorganic fertilizer, or in higher pH soils.

In conclusion, increases in archaeal amoA gene and transcript abundance during nitrification, inhibition of these increases by acetylene and the absence of changes in bacterial ammonia oxidizer abundance provide strong evidence for domination of ammonia oxidation in these soils by archaea. Although previously observed in neutral pH soil at this site (Offre et al., 2009), this study is the first to show such dominance in low pH soils and the differences between communities in pH 4.5 and 6 soils (Nicol et al., 2008; Fig. S1) suggest selection for pH-adapted phylotypes. The mechanisms for such selection have still to be elucidated but results suggest that preference for low ammonia concentration may be an important factor.

Acknowledgements

  1. Top of page
  2. Abstract
  3. Introduction
  4. Material and methods
  5. Results
  6. Discussion
  7. Acknowledgements
  8. References
  9. Supporting Information

C.G.-R. is funded by an NERC research grant (NE/F021909/1). G.W.N. is funded by an NERC Advanced Fellowship (NE/D010195/1). The authors would like to thank Mr Lawrence Maurice and the SAC Craibstone Estate (Aberdeen) for access to the Woodlands Field pH plots.

References

  1. Top of page
  2. Abstract
  3. Introduction
  4. Material and methods
  5. Results
  6. Discussion
  7. Acknowledgements
  8. References
  9. Supporting Information

Supporting Information

  1. Top of page
  2. Abstract
  3. Introduction
  4. Material and methods
  5. Results
  6. Discussion
  7. Acknowledgements
  8. References
  9. Supporting Information

Fig. S1. Denaturing gradient gel electrophoresis of bacterial (a) and archaeal (c) 16S rRNA genes and archaeal amoA genes (b) during incubation of microcosms containing soils of pH 4.5 or 6 in the presence or absence of acetylene.

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
FEM_971_sm_suppl-s1.ppt6607KSupporting 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.