Applied Microbiology and Biotechnology

, Volume 86, Issue 4, pp 1043–1055

Potential roles of anaerobic ammonium and methane oxidation in the nitrogen cycle of wetland ecosystems

Authors

    • State Key Laboratory of Environmental Aquatic QualityResearch Center for Eco-Environmental Sciences, Chinese Academy of Sciences
  • Mike S. M. Jetten
    • Institute for Water and Wetland Research, Department of MicrobiologyRadboud University Nijmegen
  • Peter Kuschk
    • Helmholtz-Centre for Environmental Research–UFZ
  • Katharina F. Ettwig
    • Institute for Water and Wetland Research, Department of MicrobiologyRadboud University Nijmegen
  • Chengqing Yin
    • State Key Laboratory of Environmental Aquatic QualityResearch Center for Eco-Environmental Sciences, Chinese Academy of Sciences
Mini-Review

DOI: 10.1007/s00253-010-2451-4

Cite this article as:
Zhu, G., Jetten, M.S.M., Kuschk, P. et al. Appl Microbiol Biotechnol (2010) 86: 1043. doi:10.1007/s00253-010-2451-4

Abstract

Anaerobic ammonium oxidation (anammox) and anaerobic methane oxidation (ANME coupled to denitrification) with nitrite as electron acceptor are two of the most recent discoveries in the microbial nitrogen cycle. Currently the anammox process has been relatively well investigated in a number of natural and man-made ecosystems, while ANME coupled to denitrification has only been observed in a limited number of freshwater ecosystems. The ubiquitous presence of anammox bacteria in marine ecosystems has changed our knowledge of the global nitrogen cycle. Up to 50% of N2 production in marine sediments and oxygen-depleted zones may be attributed to anammox bacteria. However, there are only few indications of anammox in natural and constructed freshwater wetlands. In this paper, the potential role of anammox and denitrifying methanotrophic bacteria in natural and artificial wetlands is discussed in relation to global warming. The focus of the review is to explore and analyze if suitable environmental conditions exist for anammox and denitrifying methanotrophic bacteria in nitrogen-rich freshwater wetlands.

Keywords

AnammoxAnaerobic methane oxidationWetlandsNitrogen cycle

Background and introduction

Anammox, the anaerobic oxidation of ammonium coupled to nitrite reduction with N2 as the end product, is a newly discovered microbial transformation pathway in the global nitrogen cycle (Devol 2003; Strous and Jetten 2004; Douglas and Angela 2007). The anammox reaction is performed by chemolithoautotrophic or mixotrophic bacteria that use nitrite or nitrate as electron acceptor (Kartal et al. 2008). The anammox bacteria have been identified as a distinct phylogenetic order, the Brocadiales, which is part of the phylum Planctomycetes (Jetten et al. 2001). Anammox bacteria were first observed in wastewater treatment systems (Mulder et al. 1995) before their presence and activity was observed in natural ecosystems such as marine sediments (Thamdrup and Dalsgaard 2002), the water column of an anoxic basin (Kuypers et al. 2003), and oxygen minimum zones (Kuypers et al. 2005; Hamersley et al. 2007). This discovery and potential contribution of anammox bacteria is important because it may necessitate a re-evaluation of the major fluxes in the global nitrogen budget and regulation of the processes leading to losses of fixed nitrogen. The addition of nitrite-dependent anaerobic methane oxidation (ANME coupled to denitrification) to the nitrogen cycle is even more recent (Raghoebarsing et al. 2006; Ettwig et al. 2008). Anoxic sediment from a fresh water canal in The Netherlands was used as an inoculum and supplied with methane, nitrate, and nitrite in a bioreactor with efficient biomass retention. The reactor was operational for more than 16 months before significant methane and nitrite consumption could be measured (Raghoebarsing et al. 2006). Molecular analysis of the biomass indicated the presence of two groups of microbes; about 15% of the biomass was made up of archaea distinctly related to the ANME archaea found in marine ecosystems, and about 80% of the cells were related to the NC10 phylum that has no isolated or cultured relatives. These finding were recently confirmed in a study using fresh water sediment and wastewater sludge as inoculum in Australia (Hu et al. 2009). However, many 16S rRNA gene sequences similar to the sequence of denitrifying methanotrophs have now been retrieved from different anoxic freshwater sediments, indicating that denitrifying methanotrophic bacteria may be more widespread than previously assumed. In a very recent study (Ettwig et al. 2009), sediment from a eutrophic fresh water ditch harboring a diverse community of NC10 bacteria was incubated in a bioreactor with efficient biomass retention and a constant supply of methane and nitrite. In this case already after 6 months, the denitrifying methanotrophic bacteria dominated the reactor community as visualized by fluorescence in situ hybridization with probes specifically designed to detect these bacteria. The enrichment culture oxidized methane and reduced nitrite to dinitrogen gas in the expected stoichiometry. The diversity of the methanotrophic NC10-bacteria was assessed in both the inoculum and the final biomass of the enrichment culture. After compilation of the currently available sequences of this bacterial phylum, it was shown that only members of one particular subgroup had been enriched. By designing primers suitable for Q-PCR, the growth of this NC10 subgroup was monitored retrospectively and could be correlated to nitrite-reducing activity and total biomass of the culture (Ettwig et al. 2009). The availability of suitable primers makes it possible to survey more anoxic freshwater sediments that are suspected to provide suitable environmental conditions for nitrite-dependent anaerobic methane oxidation. Together with stable isotopic methods, the environmental distribution and importance of these bacteria can now be estimated.

In the last 10 years, anammox research has bloomed with more than 787 publications and 13,460 citations listed in the ISI Web of Science (Fig. 1), and it is estimated that more than 70 research groups worldwide are investigating different aspects of this process. In spite of this effort many questions on anammox and ANME coupled to denitrification remain and need to be addressed.
https://static-content.springer.com/image/art%3A10.1007%2Fs00253-010-2451-4/MediaObjects/253_2010_2451_Fig1_HTML.gif
Fig. 1

Publications and citations of anammox research (source: ISI Web of Science)

  • Nitrification links N regeneration and N loss in nature. The relative efficiency of this connection appears to differ between coastal marine or estuarine and freshwater sediments. Estuarine and marine sediments release similar amounts of ammonium and dinitrogen (N2), whereas freshwater sediments are assumed to release mainly N2 with coupled nitrification–denitrification. The factors that control this pattern are unclear. Is it related to presence or absence of anammox bacteria and denitrifying methanotrophic bacteria?

  • Are anammox bacteria present and active in other natural ecosystems besides marine ecosystems and wastewater treatment plants? What is the extent of the biodiversity of anammox bacteria in the other natural ecosystems?

  • The overall importance of the various nitrogen converting processes in freshwater ecosystems are still largely unknown due to the lack of sufficient data. How many of the budgets of the freshwater nitrogen cycle can be attributed to anammox or ANME coupled to denitrification?

  • Presently there are no indications for N2O as an intermediate of the anammox or ANME coupled to denitrification process. This may call for a recalculation of N2O fluxes and budgets for marine and continental surfaces.

Based on the presently known 16S rRNA gene sequences of anammox and denitrifying methanotrophic bacteria, specific oligonucleotide probes and primers have been developed to document the presence of anammox and denitrifying methanotrophic cells in natural and engineered ecosystems (Schmid et al. 2005, 2007; Ettwig et al. 2009). However, there is relatively little literature about anammox in natural and constructed wetlands (Shipin et al. 2005; Paredes et al. 2007; Erler et al. 2008), although wetlands are an important part of the ecological systems on earth and may harbor enough suitable habitats for anammox bacteria and other nitrogen cycle bacteria (Fig. 2). In this overview, we will address the following two issues: Do freshwater wetlands provide suitable environmental conditions for anammox bacteria? and if so, where on earth would wetlands have specific zones and key environmental circumstances favorable for anammox bacteria? Moreover, what is the potential role of anammox and ANME coupled to denitrification in the nitrogen cycle of wetlands?
https://static-content.springer.com/image/art%3A10.1007%2Fs00253-010-2451-4/MediaObjects/253_2010_2451_Fig2_HTML.gif
Fig. 2

Microbial nitrogen transformations in wetland ecosystems. DNRA dissimilatory nitrate reductase to ammonium; AOB ammonia-oxidizing bacteria; AOA ammonia-oxidizing archaea

Opportunities for anammox and ANME coupled to denitrification in wetlands

Favorable environmental conditions

Experimental results with bioreactors and environmental surveys in marine ecosystems have shown that suitable growth conditions for anammox and NC10 bacteria include: low dissolved oxygen concentration <2 µM, low nitrite concentration <5 mM, moderate nitrogen loading rates 0.02–0.3 kg NH4+-N/kg dry weight per day, relatively long solid retention times (>11 days), temperature between −2°C and 80°C, and a pH between 6.7 and 8.3 (Jetten et al. 1999; Jetten et al. 2001; Ettwig et al. 2009). Most of these conditions can be found in either polluted, natural, or constructed wetlands.

The anammox and NC10 bacteria

The group of anammox bacteria, placed in the order Brocadiales, is currently made up of five genera, “Candidatus Brocadia” “Candidatus Kuenenia” “Candidatus Anammoxoglobus” “Candidatus Jettenia” and “Candidatus Scalindua” as shown in Table 1. Various anoxic marine ecosystems have been monitored for anammox activity and the presence of anammox cells (Kuypers et al. 2003, 2005; Rysgaard et al. 2004; Risgaard-Petersen et al. 2004; Schmid et al. 2007). The anammox bacteria in marine environments seem to be restricted to the Scalindua branch with relatively little biodiversity as reflected in very similar 16S rRNA gene sequences (Fig. 3a). In order to increase the resolution, Woebken et al. (2008) used the intergenic spacer region of marine anammox bacteria and showed the Scalindua anammox bacteria in the oxygen minimum zone of the Arabian Sea were most distinct from other oxygen-limited marine ecosystems.
Table 1

Recently described bacteria capable to oxidize ammonium anaerobically

Order

Species

Field

References

Brocadiales

Candidatus Brocadia anammoxdians

WR

Kuenen and Jetten 2001

Candidatus Kuenenia stuttgartiensis

WR

Schmid et al. 2000

Candidatus Brocadia fulgida

WR

Kartal et al. 2008

Candidatus Jettenia asiatica

WR

Quan et al. 2008

Candidatus Anammoxoglobus propionicus

WR

Kartal et al. 2007

Candidatus Scalindua brodae

WR and ME

Schmid et al. 2003

Candidatus Scalindua wagneri

WR and ME

Schmid et al. 2003

Candidatus Scalindua sorokinii

ME

Kuypers et al. 2003

Candidatus Scalindua arabica

ME

Woebken et al. 2008

WR wastewater treatment, ME marine ecosystem

https://static-content.springer.com/image/art%3A10.1007%2Fs00253-010-2451-4/MediaObjects/253_2010_2451_Fig3_HTML.gif
Fig. 3

The diversity of anammox bacteria (a) and NC10 bacteria (b) in marine and freshwater nitrogen cycle. In (a) all detected marine anammox samples belong to the Candidatus “Scalindua” genus, while there is high diversity in freshwater ecosystem. The red triangles labeled sequences are newly detected by authors in freshwater ecosystems. In (b) all phylogenetic tree is based on the 16S rRNA gene of the NC10 phylum with acidobacteria as the outgroup

Different from marine ecosystems, wetlands are the most active zones in biogeochemistry with great regional heterogeneity. The biogeochemical N cycle in wetland ecosystems features a combination of chemical transformations and transport processes not shared by many other ecosystems (Morris 1991; Mitsch and Gosselink 2007). This may usually lead to a high microbial biodiversity (Birol et al. 2006). The results of our anammox research also showed that there were relative high biodiversity of anammox bacteria in various freshwater wetlands (unpublished results). Beside Scalindua, the other genera “Brocadia,” “Kuenenia,” “Anammoxoglobus,” and “Jettenia” were all detected in wetlands (Fig. 3a). The apparent possible reasons may be the variability of the wetlands with respect to nitrogen load, oxygen concentration, and salinity, but elucidating the niche differentiation of various anammox bacteria will need much more research.

So far, nitrite-dependent anaerobic methane oxidation has been restricted to bacteria of the NC10 phylum as outlined above and shown in Fig. 3b.

Based on phylogenetic and genomic analysis, anammox bacteria appear to have a common ancestor, despite the fact that there is a large evolutionary difference between the five genera. Furthermore, all species share similar physiological traits (Kartal et al. 2007; van de Vossenberg et al. 2008). Anammox bacteria contain a separated intracytoplasmic compartment named the anammoxosome, where anammox catabolism was shown to take place (Strous et al. 1999; Van Niftrik et al. 2004, 2008a, b). The membrane of this ‘‘organelle” consists of unusual ladderane lipids forming a dense barrier which is thought to reduce the permeability of the membrane to small molecules (Sinninghe et al. 2002, 2005). Ladderane lipids occur in a variety of different forms either ester or ether bound to the glycerol backbone and contain three or five linearly fused cyclobutane rings (Boumann et al. 2006; Rattray et al. 2008), which is unprecedented in nature. Ladderanes have been applied as biomarkers for anammox abundance in anoxic waters of the Black Sea (Kuypers et al. 2003), oxygen depleted zones of the ocean (Kuypers et al. 2003; Hamersley et al. 2007; Jaeschke et al. 2009a), and in marine sediments (Hopmans et al. 2006). One of the major octaheme cyctochrome c proteins of anammox bacteria (hydroxylamine or hydrazine oxidoreductase) was found to reside in the anammoxosome by immunogold labeling (Lindsay et al. 2001) and cytochemical staining (van Niftrik et al. 2008a, b). Recently, the genome of the wastewater anammox bacterium Kuenenia stuttgartiensis was elucidated using a metagenomic approach. The genome analysis showed that anammox bacteria might be more versatile than previously assumed. In addition, laboratory enrichment cultures showed that anammox bacteria may be able to use acetate and propionate as energy source (Kartal et al. 2007, 2008) in addition to ammonium and nitrite. The use of alternative electron donors (ammonium, formate, acetate, propionate, or Fe(II)) and acceptors (nitrate, nitrite, Fe(III), and Mn(IV)) might also be advantageous for anammox bacteria in wetland ecosystems. The construction of the metagenome of the NC10 bacteria responsible for ANME coupled to denitrification is currently in progress and will undoubtedly yield useful data to unravel the biochemical pathway of this intriguing process.

Temperature

Anammox bacteria have been found in many different ecosystems with a large temperature range from −2°C to 80°C. The optimal temperature for most anammox bacteria from wastewater treatment systems is around 35°C (Kuenen and Jetten 2001). However, the temperature optima for marine anammox populations are much lower and appear to reflect their local environment, as shown in Table 2. For example, in the permanently cold sediment of Young Sound, Greenland, with a temperature of <1°C year round, the temperature optimum for anammox was 12°C (Rysgaard et al. 2004), and in the Skagerrak, with annual temperatures between 4°C and 6°C, the optimum temperature was 15°C (Dalsgaard and Thamdrup 2002). Recently, anammox activity, ladderane lipids, and 16S rRNA genes were observed at hydrothermal vents and in hot springs which have expanded the temperature ranges significantly (Jaeschke et al. 2009b; Byrne et al. 2009). The wide range in growth temperature of anammox bacteria may promote their existence in wetlands. So far, the NC10 bacteria have only been detected in moderate fresh water sediments with temperatures between 10°C and 30°C (Ettwig et al. 2008, 2009; Hu et al. 2009).
Table 2

Anammox and ANME coupled to denitrification in marine and fresh water ecosystems

Location

sample

Anammox rate (nmol cm−3 h−1)

Anammox cell count (cell ml−1; ×108)

16S rRNA affiliation

[NOX] (μM)

[NH4+] (μM)

[DO] (μM)

Depth (m)

In situ temperature (°C)

Reference

Arabian Sea

Water column

ND

ND

Candidatus Scalindua” spp.

∼22

ND

<3

221

ND

Woebken et al. 2008

Celtic Sea

Sediments (0-8 cm)

1.3–2.8 nmol 29N2 ml−1 wet sediment

ND

Candidatus S. sorokinii/wagneri

0–59

5–120

1.2 to 2.0 cm

500–2,000

ND

Jaeschke et al. 2009a,b

Barents Sea

Sediment

0.8

0.8

Candidatus Scalindua” spp.

8

ND

ND

206

2

Schmid et al. 2007

Black Sea

Water column

ND

0.000019 ± 0.000008

Candidatus S. sorokinii

3–5

ND

ND

92

10

Kuypers et al. 2003

Baltimore

Sediment

0.008 µM 29N2 and 8.3 × 10−5 µM 30N2 g−1 h−1

0.000003 ± 0.000001

Candidatus Brocadia/Kuenenia/Scalindua

ND

ND

ND

9.1

ND

Tal et al. 2005

Benguela OMZ

Water column

ND

0.0003

Candidatus S. sorokinii/brodae

25

0.71

0.6

110

11

Kuypers et al. 2005

Cape Fear Estuary

Sediment

0.065 ± 0.0127 ∼0.66 ± 0.099 nmol N2 g−1 h−1

1.3 × 105 to 8.4 × 106 g−1

Candidatus Brocadia

5.5–6.9

4–5.2

ND

1

19.3–19.6

Dale et al. 2009

Scalindu/Kuenenia/Jettenia

Baltic Sea

Near-bottom water

10–30 µmol N m−2 day−1

ND

ND

0.29–2.26

ND

95–390

0.05

2.0–5.9

Hietanen and Kuparinen 2008

Chesapeake Bay

Sediment

ND

ND

Candidatus Scalindua” spp.

74.4–93.6

9.8

ND

10

15–25.3

Rich et al. 2008

Chilean OMZ

Water

8.79 × 10−5 to 2.4 × 10−4

0.0000275

Candidatus Scalindua” spp.

13

0

ND

50

14°C

Galan et al. 2009

Disko Bay

Sediment

0.63

0.7

ND

10.4

1.7

0.75

50

−1.5

Schmid et al. 2007; Rysgaard et al. 2004

Lake Rassnitzer

Water column

6 ∼ 504 nmolN2 L−1 day−1

0.00027–0.00052

Candidatus S. sorokinii/brodae/wagneri

0–200

5–600

0–400

0–36

4–23

Hamersley et al. 2009

Candidatus B. fulgida

Greenland Sea

Sediment

0.22

0.2

ND

3.4

ND

0.3–0.5

36

0.7

Schmid et al. 2007; Rysgaard et al. 2004

Gullmarsfjorden

Sediment (upper 2 cm)

 

0.4

Candidatus S. sorokinii/brodae

2

7

ND

67

7

van de Vossenberg et al. 2008

Golfo Dulce

Water

0.02

ND

ND

1.5

1.5

ND

180

ND

Dalsgaard et al. 2003

Gullmarsfjord

Sediment

0.86–2.75

ND

ND

ND

ND

ND

116

6

Jensen et al. 2007

Hot springs

Water

ND

ND

Brocadia/Kuenenia

0.8–27.7

0–13.3

ND

ND

23.7–96.6

Jaeschke et al. 2009a,b

Irish Sea

Sediments (0-8 cm)

2.1–25.7 nmol 29 N2 ml−1 wet sediment

ND

Candidatus S. sorokinii/wagneri

0–14

0–120

0.7 to 1.6 cm

50–100

ND

Jaeschke et al. 2009a,b

Logan/Albert River

Sediment

0–9

ND

ND

0–120

ND

0–107

0–7 mm

ND

Meyer et al. 2005

Mariager Fjord

Water

0.00125 ∼ 0.01875

ND

ND

0–14.4

ND

8–190

13–21

ND

Jensen et al. 2009

Mid Atlantic Ridge

Sediment

0–0.03 µM day−1

ND

Brocadia/Kuenenia/Scalindua

ND

ND

ND

750–3650

4–153

Byrne et al. 2009

North Atlantic

Sediment (5 cm)

0.1–2.5 mmol N m−2 h−1

ND

ND

1.1–20.8

ND

ND

50–2,000

3.9–14.4

Trimmer and Nicholls 2009

Namibian OMZ

Water

ND

ND

Candidatus S. Brodae/sorokinii

ND

ND

18

52

ND

Woebken et al. 2007

Peruvian OMZs

Water column

18–290 nM N day−1

0.2

Candidatus S. sorokinii

10–30

<1

ND

96–2,400

ND

Lam et al. 2009

Plum Island Sound Estuary

Sediment (3–4 cm)

0 ∼ 0.2 nmol N g −1 h−1

ND

ND

200

ND

ND

ND

 

Koop-Jakobsen and Giblin 2009

Randers Fjord

Sediment (30 cm)

3.2–11.2

ND

Candidatus S. sorokinii

15–300

ND

0.3–0.5

0.5

5–20

Risgaard-Petersen et al. 2004

Skagerrak and Aarhus Bay

Sediment

1.17 (80% of the total N2)

ND

ND

19–35

ND

0.3–1.7 cm

16–695

5.5–6.5

Thamdrup and Dalsgaard 2002

Tanganika lake

Water

10 nM N2 h−1

0.00013

Candidatus S. brodae

0 ∼ 10

0–1.6

<1

100–110

ND

Schubert et al. 2006

Washington bay

Sediment (0–15 cm)

0.065 ∼1.7 nmol N mL−1 h−1

ND

Candidatus S. sorokinii

0–6

0–40

ND

2,800–3,100

2

Engstrom et al. 2009

Yodo estuary

Sediment (0–2 cm)

0.69 nmol 29N2 cm−3 h−1

ND

Candidatus S. wagneri

ND

ND

ND

10.5–42

ND

Amano et al. 2007

Young Sound

Sediment

0.51

0.38 ± 0.27

ND

3.2

0.5

1.3–1.7

36

−1.3

Schmid et al. 2007; Rysgaard et al. 2004

inoculum source

Sample

NO2-/NO3-reduction [nmol min-1mg-1(protein)]

Degree of enrichment (NC10 bacteria & archaea)

Type of enriched microorganisms

NOx supplied

[CH4] µMa

DO

Depth

Enr. T °C

 

Twentekanaal, Netherlands

Canal sediment

3.7 (NO2-)

90 %

80 % NC10, 10 % archaea

NO2- and NO3-

1320

0

ND

25

Raghoebarsing et al. 2006; Ettwig et al. 2008

Ooijpolder, Netherlands

Ditch sediment

3.4–5.6 (NO2-)

80 %

80 % NC10, no archaea

NO2- and NO3-

1210

0

ND

30

Ettwig et al. 2009

Brisbane, Australia

lake sediment

0.05 (NO3-)

15 %

15 % NC10, no archaea

NO3-

1160

0

ND

22

Hu et al. 2009

Brisbane, Australia

Wastewater sludge

2.6 (NO2-) 3 (NO3-)

70 %

30 % NC10, 40 % archaea

NO3-

900

0

ND

35

Hu et al. 2009

ND no data

acalculated from headspace concentration; saturation 1275 µM at 30 °C and 0 % salinity

The aerobic, anoxic, and anaerobic interface

Due to its positive redox potential, molecular oxygen is one of the most important reactants in biogeochemical cycles (Brune et al. 2000). However, the low solubility of oxygen in water (about 0.2–0.3 mM in equilibration with air), in combination with rapid consumption leads to the development of oxic–anoxic interfaces, which separate aerobic from anaerobic processes in virtually all environments, ranging in scale from oceanic sediments to the fecal pellets of small soil invertebrates, as shown in Fig. 4. In various ecosystems, anammox and NC10 bacteria will be dependent on the activity of aerobic ammonia-oxidizing bacteria under oxygen-limited conditions, e.g., at the oxic/anoxic interface (Dalsgaard et al. 2003; Schmid et al. 2007; Lam et al. 2007, 2009). Oxic/anoxic interfaces are abundant in nature, for example in the rhizosphere of helophytes, biofilms, and marine sediments (Murray et al. 1989). Moreover, oxygen supply may be a major rate-limiting factor for aquatic plant-based wastewater treatment processes such as constructed wetlands, ponds, and lagoons. In addition, some studies reported that biomass at the interface is three orders of magnitude higher than that in the superstratum body of water (Stottmeister et al. 2003).
https://static-content.springer.com/image/art%3A10.1007%2Fs00253-010-2451-4/MediaObjects/253_2010_2451_Fig4_HTML.gif
Fig. 4

Microprofiles of pH, ORP and DO near the lateral root of bulrush (Scirpus validus). (Bezbaruah and Zhang 2004)

In wetland ecosystems, specially adapted higher plants (helophytes) rooting in waterlogged soils/sediments have a well-developed system of air lacunae permitting efficient intra-plant transport of gases. Oxygen transport to the roots maintains aerobic metabolism and frequently results in oxygen release to the surrounding waterlogged soil (Stottmeister et al. 2003). Diverse and dense microbial populations may be found associated with the plant roots because of a supply of organic matter from root exudates or decomposing roots, coupled with a continuous supply of oxygen from the roots and reduced ionic species from the anoxic surroundings (Christensen et al. 1994).

The critical limit oxygen concentration for the known anammox organisms is around 2 μmol/l (Jetten et al. 2001), but the inhibition is reversible as soon as oxygen is removed (Strous et al. 1997; Third et al. 2001). The influence of oxygen on the ANME coupled to denitrification process is currently being investigated. Where sulfate reduction occurs, the presence of sulfide inhibits conventional coupled nitrification/denitrification (Joye and Hollibaugh 1995), thus removing the oxidant required for anammox, although higher sulfide concentration may also inhibit anammox bacteria (van de Graaf et al. 1995). These limits constrain the depth distribution of anammox observed in the Black Sea and the Golfo Dulce and implied that the extent of oceanic oxygen minimum zones in which anammox might occur is restricted to a few meters. In wetland sediments, the oxic–anoxic boundary may be limited to a few millimeters, thereby restricting the zone favorable for anammox and NC10 bacteria.

Role of nitrite formation

After the microbial nature of the anammox process was verified, nitrite was shown to be the preferred electron acceptor over nitrate. Hydrazine was identified as an important intermediate (Jetten et al. 2008), and based on genomic analysis, it was recently speculated that nitric oxide could also be an intermediate in the anammox reaction (Strous et al. 2006). Nitrite can be formed in two processes: either during incomplete oxygen-limited ammonia oxidation or during substrate limited nitrate reduction in denitrification (Lam et al. 2007, 2009).

Complete nitrification is a sequential biological oxidation process, which involves two different groups of bacteria. The first step of nitrification is the oxidation of ammonia to nitrite via hydroxylamine, involving the membrane-bound ammonia mono-oxygenase. This reaction is carried out by ammonia-oxidizing bacteria (AOB). Recently also, crenarchaea (AOA) have been shown to oxidize ammonium to nitrite (Konnecke et al. 2005) but no indications for the involvement of NH2OH or HAO proteins have been obtained yet. The next group of nitrifiers consists of nitrite-oxidizing bacteria (NOB) that further oxidize nitrite to nitrate. Under most environmental conditions, ammonia oxidation is the rate-limiting step, so that nitrite seldom accumulates in nitrifying systems.

From the end of the nineteenth century, researchers have continued to investigate the diversity of AOB and NOB, and at present, at least 31 different species of AOB and four genera with eight described species of NOB are widespread in different ecosystems (Alawi et al. 2007; Koops et al. 1991; Koops and Pommerening-Röser 2001). Generation and maintenance of a nitritation reaction requires that nitrite-oxidizing bacteria are removed from the biomass in biological treatment systems (van Dongen et al. 2001). However, ammonia- and nitrite-oxidizing bacteria can be found in close proximity in many ecosystems and therefore it might be difficult to find conditions that favor one over the other. Two exceptions have been reported at high temperature and at low oxygen concentrations. Since AOB have higher growth rates at elevated temperatures and NOB have lower affinities for oxygen (Peng and Zhu 2006), ecosystems with low DO concentration and high temperatures may be more restrictive to the growth of NOB, which may result in net nitrite accumulation. The radial oxygen loss from plant roots described above may favor such conditions.

Low concentrations of electron donors or a high concentration of DO can lead to incomplete denitrification or DNRA, and thus result in nitrite accumulation. Denitrification proceeds in a stepwise manner in which nitrate is sequentially reduced to nitrite, nitric oxide, nitrous oxide and nitrogen gas by different catalyzing NOx reductases. Electrons originating from, e.g., organic matter, reduced sulfur compounds, or molecular hydrogen are transferred to oxidized nitrogen compounds instead of oxygen in order to build up a proton motive force. In a recent study combining biogeochemical data and molecular surveys of functional genes Lam et al. (2009) showed that in the oxygen minimum zone of Peru, anammox bacteria obtained at least 67 % of their nitrite from nitrate reduction, and less than 33% from aerobic ammonium oxidation. The ammonium for anammox could be supplied by oxygen-limited remineralization of organic matter or DNRA.

Cooperation of anaerobic ammonia and methane oxidizing bacteria with other environmental microbes

The prerequisites for anaerobic ammonia or methane oxidation to proceed are the coexistence of ammonia/methane, nitrite and anoxic conditions. Also in this case there are two possibilities. Either anammox or NC10 bacteria cooperate with AOB that consume oxygen and provide nitrite, or anammox and NC10 bacteria coexist with nitrate reducing bacteria that supply nitrite as exemplified above under limitation of an organic electron donor (Third et al. 2001; Lam et al. 2009).

AOB and anammox bacteria may be natural partners in ecosystems with limited oxygen supply. In oxygen-limited environments, the AOB would oxidize ammonium to nitrite and keep the oxygen concentration low, while anammox bacteria would convert the produced nitrite and the remaining ammonium to dinitrogen gas. Such conditions have been established in many different reactor systems (Third et al. 2001; Pynaert et al. 2003). The NOB cannot compete in obtaining oxygen with AOB and also cannot compete in obtaining nitrite with anammox bacteria. During such conditions, a stable community can be formed by AOB and anammox bacteria, not only in reactor systems, but also maybe in natural oxic/anoxic interfaces. Our ongoing research showed the widespread distribution of anammox bacteria in a typical eutrophied lake (Baiyangdian, People’s Republic of China), moreover the anammox bacteria coexist with AOB in all of the 17 samples investigated (unpublished results). The natural partners of NC10 bacteria have not been investigated so far.

An interesting matter is to note that archaeal nitrification may provide new thinking to anammox and anaerobic methane oxidation research. The recent observation showed that AOA are more abundant than AOB in marine ecosystems (Francis et al. 2005) and soil ecosystems (Leininger et al. 2006) and might be responsible for large part of ammonium oxidation. So the existence scope of anammox and ANME coupled to denitrification maybe much larger than our previous view from anoxic/subanoxic zone to more anaerobic zone as long as alternative electron donors (ammonium, formate, acetate, propionate, or Fe(II)) and acceptors (nitrate, nitrite, Fe(III), and Mn(IV)) are available (Strous et al. 2006). Our ongoing research on the coexistence of AOA and anammox bacteria has also been verified in deep ground of sediments (−12 ∼ −15 m in North Canal) and soils (−1 m in Jiaxing Paddyfield; unpublished results) similar to the observation of Woebken et al. (2008) in marine snow particles.

Potential impact of Anammox and ANME coupled to denitrification in wetlands on the global nitrogen cycle

Potential impact on the N cycle

The discovery of anammox in marine ecosystems has distinctly influenced our knowledge of the global nitrogen cycle (Ward 2003; Brandes et al. 2007; Lam et al. 2009). Up to 50% of N2 production in oxygen-limited marine ecosystems is now attributed to anammox bacteria (Devol 2003). Wetlands are one of the most active zones of biogeochemistry on the continental surface and have great regional heterogeneity. In many countries, wetlands act as a receiving body for point and non-point pollutants. Large quantities of ammonia from fertilizer application, nitrogen-fixing crops, landfill leachate, sewage, and industrial wastewater may create oxygen-limiting conditions in wetlands which will further increase ammonia and methane liberation during anaerobic degradation processes. Moreover, wetlands act as an influx or treatment unit for point and diffused nitrogen pollutants. Therefore, an even larger contribution of anammox and ANME coupled to denitrification is possible in wetlands where N-rich organic matter is decomposed or where NH4+ is supplied from elsewhere. Also in previous studies on CH4 and N2O emissions of a typical hypereutrophic lake (Taihu Lake, People’s Republic of China), it was shown that the littoral zone is an important potential “hotspot” of N2O and CH4 emission source, and their emission zone did not overlap (Wang et al. 2006a, 2007a). Research on spatial–temporal nitrogen distribution in the littoral zone showed that during the growing season, NO3-N concentrations increased by up to three to five times from open water to reed belt, while NH4+-N concentrations decreased. Maximum total dissolved nitrogen and NH4+-N concentrations occurred in January and maximum NO3-N concentrations in March. Minimum NH4+-N and NO3-N concentrations occurred in July and August, respectively (Wang et al. 2007b). As indicated by this example, the continental surface may offer great possibilities and environmental conditions for anammox and NC10 bacteria in freshwater wetland ecosystems. If the role of anammox or NC10 has been underestimated, then re-evaluation of the N cycle must also focus on the nitrogen fixation potentials to understand whether the nitrogen budget is in balance or not.

Potential relationship to nitrous oxide (N2O) emissions

It is known that a large quantity of nitrous oxide (N2O) can be emitted during the process of biological denitrification if organic carbon is limited or during nitrification when oxygen is limited (Itokawa et al. 2001; Kampschreur et al. 2006). Recent studies showed that natural and artificial wetlands could be important emission sources of N2O (Wang et al. 2006b; Søvik et al. 2006). In this context, it is important to note that N2O has an atmospheric lifetime of 120 years and a global warming potential 296 greater than CO2 (over a 100 year time horizon) and is anticipated to be responsible for about 5% of global warming. Nitrous oxide also contributes to the depletion of stratospheric ozone. Furthermore, the concentration of N2O in the atmosphere is increasing at a rate of 0.3% per year (IPCC 2001). Along with increasing N2O emissions from different technical processes (combustion of fossil fuels etc.), large application of broader constructed wetlands for water treatment could also contribute to more severe global warming if unsuitable conditions for nitrification or denitrification are to be expected.

As we understand the processes now, there is no, or at least lower, N2O emission via the anammox or ANME coupled to denitrification pathway, which is another great advantage over conventional nitrification and denitrification (Kampschreur et al. 2008; Zhu et al. 2008). Therefore, the ratio of anammox to conventional nitrification/denitrification could influence net fluxes of these trace gases, with ramifications for global warming and the catalytic destruction of ozone in the stratosphere.

Summary

The properties of anammox and NC10 bacteria make them very suitable inhabitants of freshwater wetlands with abundant oxic/anoxic interfaces. All molecular and microbial tools have been developed for the environmental detection of anammox and NC10 bacteria, and suitable reactor systems exist for the enrichment of both groups of bacteria from these wetlands. The discovery of anammox in marine ecosystems has distinctly influenced our knowledge about global nitrogen cycle. Discovery of large-scale anammox and ANME coupled to denitrification in wetland systems could have a dramatic impact on conventional nitrogen cycle theory. Moreover, it could also provide an incentive for research on anammox and ANME coupled to denitrification in other continental surface ecosystems and groundwater systems. This research into wetland systems may be extremely important for further understanding the natural nitrogen cycle, reducing greenhouse gas emissions and global warming.

Acknowledgements

This research is financially jointly supported by the National Natural Science Foundation of China (No. 20877086), National Basic Research Program of China (No. 2009CB421103), Knowledge Innovation Program of the Chinese Academy of Sciences (RCEES-QN-200706; KZCXI-YW-06-02), and special fund of the State Key Laboratory of Environmental Aquatic Chemistry (08Y04ESPCR). The anammox research of MJ is support by ERC Advanced grant 232937. Moreover, the Guibing Zhu wants to give special appreciation to German Academic Exchange Service (DAAD) for their financially supporting the opportunity to visit the Helmholtz-Centre for Environmental Research–UFZ Leipzig-Halle, Germany and the Department of Microbiology, IWWR, Nijmegen, The Netherlands.

Copyright information

© Springer-Verlag 2010