Overview of recent researches on nitrifying microorganisms in soil

ABSTRACT

Nitrogen cycle, a most important elemental cycle in earth ecosystem, is carried out by three major microbial processes including nitrogen fixation, nitrification and denitrification. Nitrification is particularly important in agricultural soil ecosystems because it is involved in nitrogen loss, groundwater pollution by nitrogen leaching, and greenhouse gas nitrous oxide emission. Recent genomic, metagenomic and physiological analysis of nitrifying microorganisms gives new insights into their ecology and functions in soils. In the last decade, there have been a number of important findings regarding nitrification. The discovery of nitrifying archaea and complete nitrifying bacteria has overturned the conventional theory that nitrification is driven by ammonia-oxidizing and nitrite-oxidizing bacteria. Each of the newly discovered nitrifying microorganisms has unique properties, for example, a different affinity for ammonia. These nitrifying microorganisms have been shown to actually play important roles in nitrification in agricultural soils. Soil conditions and agricultural practices such as fertilization and tillage have been shown to influence the contribution of each nitrifying microorganism to the activity of nitrification. In addition, nitrification inhibition technologies have been developed to prevent the loss of nitrogen fertilizer and the of nitrous oxide emission. This review provides an overview of recent researches on the diversity and characteristics of nitrifying microorganisms, soil factors affecting their ecology in soil, their involvement in nitrous oxide emissions, and nitrification control technologies.

KEY WORDS:

• Ammonia oxidization
• nitrite oxidization
• complete ammonia oxidization
• anammox
• agricultural soils

1. Introduction

In recent decades, anthropogenic nitrogen input, either directly or indirectly, has dramatically increased due to agricultural and industrial activities. The high-production agricultural systems rely on large fertilizer inputs to sustain productivity. Global inputs of N fertilizer and N fixation by crops are currently estimated to exceed the natural N inputs to terrestrial systems by biological N₂ fixation (Fowler et al. 2013). Anthropogenic nitrogen input disturbs the nitrogen cycle in the earth, which exceeds planetary boundaries (Steffen et al. 2015). Therefore, we need to improve the efficiency of anthropogenic nitrogen utilization in crop production while reducing the negative effects of anthropogenic nitrogen input on the environment (Galloway et al. 2008).

The type and amount of nitrogen fertilizer have a significant impact on the microbial community structure involved in nitrification, denitrification, and nitrogen fixation. Nitrifying microorganisms are greatly influenced by nitrogen fertilizers because they obtain energy from ammonia oxidation or nitrite oxidation (Norton and Ouyang 2019). Nitrification is directly involved in the emission of the greenhouse gas nitrous oxide, nitrate contamination of freshwater resources, eutrophication, and loss of nitrogen from agricultural fields (Robertson et al. 2013; Norton and Ouyang 2019; Prosser et al. 2020). Therefore, we need to have a better understanding of nitrification processes to improve the utilization efficiency of nitrogen fertilizers and minimize negative impact of nitrification on the environment.

Since the discovery of nitrifying microorganisms by Sergei Winogradsky at the end of the 19th century (Winogradsky 1891), nitrification, which is the oxidation of ammonia to nitrate via nitrite, has been considered a two-step microbial reaction process performed by two distinct functional groups: ammonia-oxidizing bacteria (AOB) and nitrite-oxidizing bacteria (NOB). However, metagenomic analysis revealed the existence of a novel group of archaea that can oxidize ammonia in several environments (Treusch et al. 2005), and since then a number of pure cultures of ammonia-oxidizing archaea have been isolated from the various several environments including soil and ocean (Könneke et al. 2005; Lehtovirta-Morley 2018). Recently, it was found that a new complete ammonia oxidation (comammox) bacterium, Nitrospira, can oxidize ammonia to nitrate in a single microorganism (Daims et al. 2015; van Kessel et al. 2015). Thus, knowledge about nitrifying microorganisms is rapidly increasing. In this review, we focus on the latest findings on the following five microbial groups: (1) autotrophic ammonia oxidizing bacteria (AOB), (2) autotrophic nitrite oxidizing bacteria (NOB), (3) complete ammonia oxidizing bacteria (Comammox), (4) autotrophic ammonia oxidizing archaea (AOA), and (5) heterotrophic nitrifiers. We also briefly describe anaerobic ammonia oxidation (anammox), which plays an important role in the dynamics of ammonia in the soil. Then, we discuss the characteristics of these microorganisms and the reactions that they mediate in the soil. Finally, we introduce the control technology for them, referring to recent research results obtained using advanced methods such as metagenomics and genomics.

2. Diversity of nitrifiers

2.1. Autotrophic ammonia oxidizing bacteria (AOB)

Ammonia oxidizing bacteria (AOB) are classified into the classes Betaproteobacteria or Gammaproteobacteria. Nitrosomonas and Nitrosospira have been found in various environments such as soils, oceans, lakes, and activated sludges (Prosser and Embley 2002). Nitrosospira (encompassing Nitrosolobus and Nitrosovibrio) (Head et al. 1993; Stephen et al. 1996) is the dominant AOB genus in agricultural and forest soils (Kowalchuk and Stephen 2001). In contrast, γ-AOB of the Nitrosococcus genus, including N. oceani, N. watosonii, and N. halophilus, have been found in saline or hyper-saline aquatic environments such as oceans and salt lakes (Campbell et al. 2011). Recently, gammaproteobacterial AOB were isolated from the strongly acidic soil of the tea fields. The isolated γ-AOB, Candidatus Nitrosoglobus terrae, is acid-tolerant and non-halophilic, whose genome is considerably smaller than those of other γ-AOB and lacks several genes associated with salt tolerance that are unnecessary for survival in soil (Hayatsu et al. 2017). The potential role of Ca. Nitrosoglobus in wastewater treatment systems has recently been demonstrated (Wang et al. 2021b). The acid-tolerant γ-AOB, Ca. Nitrosacidococcus tergens, was also isolated from bioreactor cultivation using an acidic air biofilter (Picone et al. 2020).

In soil ecosystems, β-AOB, Nitrosospira, and Nitrosomonas are two dominant genera of terrestrial AOB, as revealed by culture-and metagenome-based investigations (Kowalchuk and Stephen 2001; Jiang et al. 2014). The lineages of β-AOB are defined and can be divided into clusters 0 to 8 and cluster 0 to 13 based on 16S rRNA gene sequences (Norton 2011) and ammonia monooxygenase subunit A (amoA) sequences, respectively, from pure cultures and a variety of environments (Avrahami and Conrad 2003). Each β-AOB cluster responds to various environmental factors, such as soil pH (Nicol et al. 2008), temperature (Fierer et al. 2009), land use (Fierer et al. 2009) and fertilization (Lin et al. 2018; Yang et al. 2020). Different AOB species have functional differentiation and do not equally contribute to nitrification in soil ecosystems. For example, Nitrosospira cluster 8a played a significant role in the nitrification in acidic soil under long-term inorganic and organic fertilization (Lin et al. 2018). The dominant AOB community structure was changed from Nitrosospira cluster 2 to cluster 3 by treatment with biochar (He et al. 2018).

Nitrous oxide gas (N₂O) is generated by two metabolic pathways, ammonia oxidation and nitrifier-denitrification but the mechanisms involved in this process are not completely understood. In AOB, ammonia monooxygenase (AMO) oxidizes ammonia to hydroxylamine (NH₂OH), which is subsequently converted to nitrite by hydroxylamine dehydrogenase (HAO) under oxic conditions. Nitrous oxide is formed during NH₂OH oxidation by chemical abiotic or biochemical reactions (Prosser et al. 2020; Stein 2019). Intermediate compounds such as NH₂OH, nitrite, and NO produced during biological ammonia oxidation are converted to nitrous oxide by an abiotic reactions (Liu et al. 2017). On the other hand, there are two biochemical N₂O production pathways. One is called nitrifier-denitrification, in which nitrite (NO₂⁻) is reduced to NO and N₂O by nitrite reductase (NIR) and NO reductase (NOR), respectively, under O₂ limitation and low pH conditions (Shaw et al. 2006; Wrage-Mönnig et al. 2018). The second is the HAO-mediated incomplete oxidation of NH₂OH to NO, which is subsequently reduced to N₂O by NO reductase (Kozlowski, Price, and Stein 2014). A recent study revealed a novel pathway for N₂O production in N. europaea through the oxidation of NH₂OH by cytochrome P460 (Caranto and Lancaster 2017; Caranto, Vilbert, and Lancaster 2016).

2.2. Autotrophic nitrite oxidizing bacteria (NOB)

Nitrite is an important intermediate in many nitrogen metabolism processes mediated by microbes, including nitrification and denitrification. Moreover, nitrite is reactive and toxic to some microorganisms and eukaryotes. Therefore, NOB have a significant regulatory function in the nitrogen cycle and microbial ecology in the soil. Nitrite oxidizers are a phylogenetically diverse functional group and classified into seven genera within four phylum in parentheses: Nitrobacter (α-proteobacteria), ‘Candidatus Nitrotoga’ (β-proteobacteria), Nitrococcus (γ-proteobacteria), Nitrospira (Nitrospirae), ‘Nitrolancetus’ (Chloroflexi), Nitrospina (Nitrospinae) and ‘Candidatus Nitromaritima’ (Nitrospinae) (Daims, Lücker, and Wagner 2016). NOB are widely distributed in natural and urban environments, including seawater, freshwater systems, sewage treatment plants, and various soil ecosystems (Daims et al. 2011).

Nitrobacter, Nitrospira, Candidatus Nitrotoga, Nitrospina, ‘Nitrolancetus’ and Nitrococcus have been identified in the soil by metagenomic analysis (Wang et al. 2021a). Since the abundance of Nitrobacter and Nitrospira is higher than those of the other four genera and they are more responsive to agricultural management, the two genera are considered to be the more important functional players in the NOB community in soil (Han et al. 2018). Nitrite oxidation kinetic analysis of six pure cultures, including three Nitrobacter and three Nitrospira species, showed that Nitrospira spp. are adapted to a highly substrate-limited environment and that Nitrobacter spp. have affinity to a wide range of substrates and an advantage under nitrite-rich environments (Nowka, Daims, and Spieck 2015). Therefore, Nitrobacter is commonly found in nitrogen-rich environments such as wastewater. In tillage/no-tillage agricultural soil with high nitrogen availability, potential nitrite oxidation exhibited a strong positive correlation with the abundance of Nitrobacter and a weak negative correlation with the abundance of Nitrospira-like NOB (Attard et al. 2010). Fertilization of lodgepole pine and spruce stands in forest soil increased Nitrobacter-like NOB abundance but did not affect Nitrospira-like NOB abundance. Nitrobacter-like NOB abundance correlated with potential nitrification activity and soil nitrate concentration, but no such correlations were observed for Nitrospira-like NOB (Wertz, Leigh, and Grayston 2012). In the paddy field soil, Nitrobacter-like NOB preferred the rhizosphere compartment and responded positively to increased nitrogen levels, whereas Nitrospira-like NOB were dominant in surface soil, which is a nutrient-limited environment (Ke et al. 2013). Both Nitrospira and Nitrobacter-like NOB contributed to the nitrite-oxidation process in red soil, and these communities were shaped by pH, total nitrogen, and soil organic carbon (Han et al. 2018). Ca. Nitrotoga are distributed in various environments, including soil and sediment, and has been enriched and characterized for its genome and physiology (Kitzinger et al. 2018; Ishii et al. 2017, 2020). Ca. Nitrotoga can tolerate lower nitrite and oxygen concentrations (Nowka, Daims, and Spieck 2015). However, further studies are needed to clarify the role of Ca. Nitrotoga in soil nitrification.

2.3. Complete ammonia oxidizing bacteria (Comammox)

Complete nitrification has been known to be mediated by AOB and NOB for over a century (Costa, Pérez, and Kreft 2006). However, a century after the discovery of AOB and NOB, AOA archaea were discovered as a second major group of ammonia oxidizers (Könneke et al. 2005). Recently, it was discovered that a single microorganism belonging to the genus Nitrospira can mediate complete nitrification of ammonia to nitrate via nitrite, a process termed as comammox (complete oxidation of ammonia to nitrate) (Daims et al. 2015; van Kessel et al. 2015). The thermodynamic feasibility of the comammox process was predicted by Costa, Pérez, and Kreft (2006). Comammox yields more energy (Equation 1(1) $\mathrm{N}{{\mathrm{H}}_4}^{+}+2{\mathrm{O}}_2\to \mathrm{N}{{\mathrm{O}}_3}^{-}+{\mathrm{H}}_2\mathrm{O}+2{\mathrm{H}}^{+}\left({\mathrm{\Delta }{\mathrm{G}}^{\circ }}^{\prime }=-348.9\mathrm{k}\mathrm{J}\mathrm{m}\mathrm{o}{\mathrm{l}}^{-1}\right)$(1) ) than does ammonia oxidation to nitrite (Equation 2(2) $\mathrm{N}{\mathrm{H}}_4++1.5{\mathrm{O}}_2\to \mathrm{N}{{\mathrm{O}}_2}^{-}+{\mathrm{H}}_2\mathrm{O}+2{\mathrm{H}}^{+}\left({\mathrm{\Delta }{\mathrm{G}}^{\circ }}^{\prime }=-274.7\mathrm{k}\mathrm{J}\mathrm{m}\mathrm{o}{\mathrm{l}}^{-1}\right)$(2) ) and nitrite oxidation to nitrate (Equation 3(3) $\mathrm{N}{{\mathrm{O}}_2}^{-}+0.5{\mathrm{O}}_2\to \mathrm{N}{{\mathrm{O}}_3}^{-}\left({\mathrm{\Delta }{\mathrm{G}}^{\circ }}^{\prime }=-74.1\mathrm{k}\mathrm{J}\mathrm{m}\mathrm{o}{\mathrm{l}}^{-1}\right)$(3) ). Thus, comammox is more advantageous than other nitrifying bacteria and archaea in environments with low substrate concentrations (Costa, Pérez, and Kreft 2006).

(1) $\mathrm{N}{{\mathrm{H}}_4}^{+}+2{\mathrm{O}}_2\to \mathrm{N}{{\mathrm{O}}_3}^{-}+{\mathrm{H}}_2\mathrm{O}+2{\mathrm{H}}^{+}\left({\mathrm{\Delta }{\mathrm{G}}^{\circ }}^{\prime }=-348.9\mathrm{k}\mathrm{J}\mathrm{m}\mathrm{o}{\mathrm{l}}^{-1}\right)$(1)

(2) $\mathrm{N}{\mathrm{H}}_4++1.5{\mathrm{O}}_2\to \mathrm{N}{{\mathrm{O}}_2}^{-}+{\mathrm{H}}_2\mathrm{O}+2{\mathrm{H}}^{+}\left({\mathrm{\Delta }{\mathrm{G}}^{\circ }}^{\prime }=-274.7\mathrm{k}\mathrm{J}\mathrm{m}\mathrm{o}{\mathrm{l}}^{-1}\right)$(2)

(3) $\mathrm{N}{{\mathrm{O}}_2}^{-}+0.5{\mathrm{O}}_2\to \mathrm{N}{{\mathrm{O}}_3}^{-}\left({\mathrm{\Delta }{\mathrm{G}}^{\circ }}^{\prime }=-74.1\mathrm{k}\mathrm{J}\mathrm{m}\mathrm{o}{\mathrm{l}}^{-1}\right)$(3)

The Nitrospira genus is widely distributed in both natural and engineered ecosystems (Daims et al. 2011) and is divided in six sublineages based on the DNA sequences of 16 rRNA genes (Lebedeva et al. 2011). All known comammox organisms belong to Nitrospira sublineage II (Lawson and Lücker 2018; Takahashi et al. 2020). Canonical NOB in sublineage II lacks the activity and genes involved in ammonia oxidation.

Nitrospira inopinata in pure culture have a higher substrate affinity for ammonia than AOA and AOB in soil, as the Km of N. inopinata for NH₃ (63 nM) is 4- to 2500-fold lower than that of AOB; thus, comammox bacteria could have a competitive advantage in soil with a very low ammonia concentration (Kits et al. 2017). On the other hand, in comparison with AOA and especially AOB, N. inopinata produces N₂O at lower levels (Kits et al. 2019). Comammox are distributed in various environments (Shi et al. 2020) and significantly contribute to soil nitrification (Osburn and Barrett 2020).

2.4. Autotrophic ammonia oxidizing archaea (AOA)

AOA belong to the phylum Thaumarchaeota and are divided into five main clusters: Nitrososphaera, Nitrosocosmicus (previously Nitrososphaera-sister lineage), Nitrosocaldus, Nitrosotalea, and Nitrosopumilus (Pester et al. 2012). Recently, Alves et al. (2018) reconstructed a robust and highly resolved archaeal amoA gene phylogeny and defined a novel multilevel taxonomy (Alves et al. 2018). After the isolation of Nitrosopumilus maritimus from seawater, many AOA have been isolated from various environments. Nitrososphaera viennensis was isolated from soil at pH 8.0, and their optimal growth pH was 7.5 (Tourna et al. 2011; Stieglmeier et al. 2014a). Ca. Nitrosotalea devanaterra is an obligately acidophilic AOA that grows in the pH range of 4.0 to 5.5 (Lehtovirta-Morley et al. 2011, 2016b). Members of the Nitrosotalea lineage are widely distributed and abundant in acidic soils (Gubry-Rangin et al. 2011).

Kinetic analysis revealed a higher affinity of AOA to NH₃ compared with that of AOB isolated from soil samples. Thus, growth and ammonia-oxidizing activity of AOA are more competitively advantageous than are those of AOB under lower ammonia concentration conditions (Martens-Habbena et al. 2009). Indeed, Nitrosopumilus maritimus SCM1 revealed the lowest reported Km value (~3 nM) for NH₃ among all the cultured ammonia oxidizers. However, most recently, this substantially higher affinity of AOA to NH₃ in comparison with that of AOB has been questioned by several reports (Kits et al. 2017), as the affinity of several isolated AOA for ammonia was not significantly different from that of AOB.

AOA significantly contribute to N₂O production in various environments, based on the AOA amoA abundance in environments and the N₂O isotopic signature (Santoro et al. 2011; Jung et al. 2014). However, N₂O yields (N₂O-N per NO₃-N produced) from AOA cultures are generally lower than those from AOB culture (Hink et al. 2017, 2018). In AOA, N₂O is considered to be produced via abiotic reactions because there is no genomic or physiological evidence for the enzymatic production of N₂O through NH₃ oxidation (Kozlowski et al. 2016; Stieglmeier et al. 2014b). However, Jung et al. (2019) indicated that Nitrosocosmicus oleophilus might be able to enzymatically denitrify nitrite to N₂O at low pH conditions (Jung et al. 2019).

2.5. Heterotrophic nitrification

Heterotrophic nitrification is generally defined as the oxidation of inorganic and organic reduced forms of nitrogen, to nitrite or nitrate, by a wide range of eukaryotes (mostly fungi) and heterotrophic bacteria (De Boer and Kowalchuk 2001). Heterotrophic nitrification is not linked to energy conservation (i.e., ATP production), and its cell-based rates are significantly lower than those in cultivated autotrophic nitrifiers (Verstraete and Focht 1977). However, heterotrophic nitrification might have an important role in soils under particular conditions that are unfavorable for autotrophic nitrifiers (Li et al. 2018b). Little is known about the prevalence and diversity of heterotrophic nitrifiers in different soil environments. In some cases, heterotrophic nitrification has a low efficiency in oxidizing ammonia to nitrite or nitrate. Some heterotrophic nitrification organisms metabolize organic nitrogen to produce nitrite and nitrate (Islam, Chen, and White 2007). It has also been shown that this process can produce nitrous oxide (Zhang et al. 2018; Zhang, Mueller, and Cai 2015).

Generally, ¹⁵N-labeled NH₄⁺, NO₃⁻, were applied with C₂H₂, which are specific inhibitors of denitrification and autotrophic nitrification at different concentrations, to distinguish the relative contributions of heterotrophic nitrification, autotrophic nitrification, and denitrification to total NO₃⁻ and NO₂ production. In most cases, heterotrophic nitrification refers to the oxidation of organic N compounds to NO₃⁻ (Killham 1990; Zhang, Mueller, and Cai 2015). This type of heterotrophic nitrification occurs commonly in forest soils that have low soil pH and high soil C/N ratio (Huygens et al. 2008; Zhu et al. 2015). In the coniferous forest soil, heterotrophic nitrification is the dominant NO₃⁻ production pathway and contributes with 95% of the total nitrification (Zhang et al. 2013). Heterotrophic nitrification is an important pathway involved in N₂O emission from the soil (Lan et al. 2020; Zhang et al. 2018). The contribution of heterotrophic nitrification to N₂O emission is more significant than is autotrophic nitrification in acidic (Stange et al. 2013) and neutral soils (Müller et al. 2014).

2.6. Environmental factors affecting nitrification

2.6.1. Occurrence of AOB, AOA, and comammox bacteria in agricultural soils

The abundance, community structure, and relative contributions to nitrification of AOA, AOB, and comammox are influenced by correlated physical and chemical factors such as pH, ammonia concentration, temperature, and water content in soils (Taylor et al. 2012; Giguere et al. 2015; Li et al. 2019a). Since the discovery of AOA (Könneke et al. 2005), extensive research has been conducted to elucidate the contribution of AOA and AOB to nitrification and their niche specialization in soil ecosystems (Prosser and Nicol 2012; Lehtovirta-Morley 2018). These studies show, in many cases, that AOA is often more abundant than AOB in soils (Clark et al. 2020; Leininger et al. 2006; Schleper and Nicol 2010; He et al. 2012). Since the abundance of AOB, AOA, and comammox varies greatly depending on the soil management and soil conditions, it is necessary to confirm the relationship between the abundance of AOB, AOA, and comammox and the nitrification activity by referring to the original paper. Specific inhibitors of AOA and AOB and stable isotope probing techniques have been used to directly link the activities of AOA and AOB to the total nitrification activity in soil (Taylor et al. 2013). Although there is not enough information on the ecology of comammox organisms, there has been increasing numbers of reports on their abundance and their contribution to nitrification in various environments (Li et al. 2019a; Osburn and Barrett 2020). According to these reports, comammox has been detected in many soils, and their abundance may even exceed that of AOB. The abundance of comammox amoA genes was comparable to that of AOA and AOB amoA genes, suggesting that comammox is also involved in nitrification activity in soil ecosystems. Given that nitrification is considered to be driven by AOA and AOB, if the contribution of comammox to nitrification becomes clear, it will be necessary to reconsider the nitrogen dynamics and nitrogen fertilization techniques in arable soil. This review outlines the relationship between these nitrifying bacteria and some important environmental factors.

2.6.2. Soil pH

Soil pH is one of the most important factors affecting activity, population, and diversity of nitrifying microorganisms (Li et al. 2018b; Aigle, Prosser, and Gubry-Rangin 2019). Nitrification activity was observed in a wide range of soil pH values (Booth, Stark, and Rastetter 2005). The substrate of ammonia monooxygenase Amo, which mediates the first step of ammonia oxidation in autotrophic ammonia oxidizer, is ammonia rather than ammonium (Suzuki, Dular, and Kwok 1974). Ammonia concentration decreases exponentially with decreasing pH due to the ionization of ammonia to ammonium. Therefore, the low ammonia availability at low pH is a major factor limiting AOB growth in acidic soils. However, many studies have shown the contribution of AOB to nitrification activity in acidic soils (Aigle, Prosser, and Gubry-Rangin 2019; De Boer, Gunnewiek, and Laanbroek 1995; Allison and Prosser 1991; Long et al. 2012; Jiang and Bakken 1999; Petersen et al. 2012; Wertz, Leigh, and Grayston 2012). Several reports indicate that the contribution of AOA to nitrification activity is more important than that of AOB in acidic soil (Hu, Xu, and He 2014; Li et al. 2018b; He et al. 2020; Li et al. 2019b). AOA amoA gene abundance decreased with increasing soil pH, while AOB amoA gene abundance increased with increasing pH (Nicol et al. 2008). A biogeographical study on AOA and AOB showed that their distribution was driven by soil pH, and the AOA/AOB ratio decreased with an increase in soil pH in 65 soil samples collected from different regions (Hu et al. 2013b).

The discovery of the acidophilic ammonia oxidizer, Nitrosotalea devanaterra (Lehtovirta-Morley et al. 2011, 2014, 2016b; Yang et al. 2020) has significantly expanded our understanding of nitrification in acidic environments because AOB and AOA isolated from the environment before the discovery was neutrophilic and not able to grow below pH 6.5. Moreover, an acid-tolerant ammonia-oxidizing gamma proteobacterium, Candidatus Nitrosoglobus terrae, was isolated from strongly acidic tea field soils (Hayatsu et al. 2017). Recently, an enrichment culture of acid-adapted comammox Nitrospira was obtained from the tea field acidic soil from which Candidatus Nitrosoglobus terrae was isolated (Takahashi et al. 2020), suggesting that comammox Nitrospira may also play an important role in acidic soils. The possibility that acidophilic or acid-tolerant nitrifying bacteria are still overlooked should be considered.

The significant contribution of heterotrophic nitrification has been observed in restricted to natural or semi-natural ecosystems, particularly acidic coniferous forest soils (Killham 1990; Li et al. 2018b). As a typical example, in soils from coniferous forests, fungal heterotrophic nitrification was responsible for most of the nitrification activity (Zhu et al. 2015). Besides coniferous forest soil, significant heterotrophic nitrification activity was detected in acidic arable (Zhang et al. 2018, 2020) and tea field soils (Yokoyama et al. 2012) although not the primary nitrifier.

2.6.3. Ammonia concentration

AOA, AOB, and comammox co-exist in various soils, including agricultural soils (Pjevac et al. 2017; Liu et al. 2019b; Li et al. 2019a; Wang et al. 2020). The concentration of ammonia, which is an energy source of autotrophic ammonia oxidizers, is one of the most important environmental factors influencing niche differentiation (Prosser and Nicol 2012). Numerous studies including field and microcosm experiments determined the effect of ammonia concentration on AOA and AOB, indicating that AOA and AOB were more abundant in conditions with low and high ammonia concentrations, respectively (Prosser and Nicol 2012). Although the application of mineral nitrogen fertilizer alters the abundance and diversity of AOB, its effect on AOA is not significant in several agricultural soils (Ouyang et al. 2016; Ai et al. 2013). On the other hand, AOA abundance increased with ammonia supplementation through mineralization of compost manure or soil organic matter (Levičnik-Höfferle et al. 2012; Liu et al. 2019a; Clark et al. 2020). These studies indicate that AOA and AOB prefer low concentrations of ammonia generated by mineralization of organic N and high concentrations of inorganic ammonia, respectively (Ouyang et al. 2016; Ouyang, Norton, and Stark 2017).

There are few reports on the effects of ammonia on abundance and diversity of comammox, since these bacteria were discovered only a few years ago. Stable isotope probing experiments revealed that the abundance of comammox Nitrospira clade A was significantly higher than that of AOA and AOB and might contribute to nitrification in the tested agricultural soils amended with N fertilizers (Li et al. 2019a). Comammox Nitrospira clade B also appears to play an active role in nitrification and to outcompete AOB at low concentrations of ammonia supplied from mineralized organic nitrogen (Wang et al. 2019). The abundance of comammox Nitrospira clade A was significantly increased by the addition of nitrogen fertilizers and was significantly suppressed by nitrification inhibitors, suggesting their potential contribution to nitrification in the soil (Li et al. 2020). Soil pH-dependent ammonia availability, rather than soil N content itself, determines the niche separation of comammox Nitrospira and canonical nitrifiers in forest ecosystems under long-term N deposition (Shi et al. 2018).

Differences in affinity for ammonia are a major factor in determining the dominant ammonia oxidizers and their community composition (Prosser and Nicol 2012). All AOA had two- or three-fold higher affinity for NH₃ than did AOB based on the Km values obtained from N. maritimus SCM1 (Martens-Habbena et al. 2009). Thus, AOA grow favorably under low ammonia conditions and becomes the dominant ammonia oxidizer in various soils. However, recently, important studies have been opposed these interpretations. Nitrosocosmicus, which was isolated from arable soil and was proposed as a new cluster, is adapted to high ammonia concentrations (Lehtovirta-Morley et al. 2016a; Sauder et al. 2017). AOA have a wide range of affinity for ammonia, and its affinity for ammonia is similar to that of AOB belonging to Nitrosomonas cluster 6a (Kits et al. 2017). On the other hand, the affinity for ammonia of the comammox Nitrospira inopinata was lower than that of marine AOA but higher than that of AOA isolated from non-marine environments (Kits et al. 2017). The affinity of ammonia oxidizers for ammonia is more diverse than previously considered. Further isolation of ammonia oxidizers and physiological studies are necessary to evaluate the niche separation of AOA, AOB, and comammox Nitrospira by ammonia concentration.

2.6.4. Temperature

Most of the pure cultured AOB strains isolated from soils have optimum growth at temperatures between 25 and 30°C (Jiang and Bakken 1999). Although enriched or pure cultured AOA from various environments have optimum growth at temperatures ranging from 25 to 74°C, Nitrosotalea devanaterra (Lehtovirta-Morley et al. 2011) and Nitrosocosmicus franklandus (Lehtovirta-Morley et al. 2016a) isolated from soil have optimal growth at 25°C.

The effect of temperature on the community structure and activity of AOA and AOB has been investigated in soil samples (Tourna et al. 2008; Taylor et al. 2017). The contribution of AOA to nitrification increases with increasing temperature, and the optimum temperature for AOA growth is higher than that for AOB growth (Ouyang, Norton, and Stark 2017). Szukics et al. (2010) revealed that the abundance of AOB increased with increasing soil temperature and that the abundance of AOA decreased in a pristine forest soil (Szukics et al. 2010). Soil nitrification activities of eight soil samples from different sites were assessed across a temperature gradient ranging from 4 to 42°C, and the results showed that the optimal temperatures for AOA growth were 12°C greater than those for AOB growth (Taylor et al. 2017).

2.6.5. Moisture and aeration

Since oxygen is an essential substrate for energy production by the oxidation of ammonia or nitrite, oxygen concentration has a great influence on the ecology of nitrifying bacteria. Nitrifiers have a wide range of affinity for oxygen, and AOA have the highest affinity, followed by AOB and NOB. Km for oxygen of AOA Candidatus Nitrososphaera sp. JG1 and Candidatus Nitrosoarchaeum koreensis MY1 isolated from soils was 4.67 µM and 10.38 µM, respectively (Kim et al. 2012; Jung et al. 2011). The Km for oxygen of AOB Nitrosomonas oligotropha NL7 and Nitrosomonas europaea C-31 was 76.3 µM and 186 µM, respectively (Park and Noguera 2007; Park et al. 2010).

2.7. Microbial N₂O emission

Nitrification-related pathways, bacterial denitrification, and fungal denitrification are major pathways that generate N₂O in soil ecosystems. N₂O is produced by nitrifiers mainly through two pathways: the oxidation of NH₂OH by AOB and the abiotic reaction of nitric oxide (NO) (Lehtovirta-Morley 2018), namely nitrifier nitrification. Another pathway is nitrifier denitrification, in which NO₂⁻ is reduced to N₂O by AOB (Wrage-Mönnig et al. 2018). In hydroxylamine oxidation, the N₂O production potential of AOB is much higher than that of AOA and comammox at the culture experiments (Kits et al. 2019; Prosser et al. 2020). On the other hand, AOB can also produce N₂O under low oxygen concentrations through nitrifier denitrification, the reduction of NO₂⁻ to N₂O via NO, by the NO₂⁻ and NO reductases, a process that was thought to be restricted to AOB (Kozlowski, Kits, and Stein 2016; Wrage-Mönnig et al. 2018). AOA and comammox bacteria lack NO reductase genes (Daims et al. 2015; Kits et al. 2019; Kim et al. 2011; Spang et al. 2012; Abbas et al. 2019).

The relative contribution of AOA and AOB to N₂O production has been investigated by quantification of amoA genes, transcripts, and/or specific inhibitors, such as the AOA-specific inhibitors, PTIO (Shen et al. 2013) and simvastatin (Zhao et al. 2020), and the AOB specific inhibitor, 1-octyne (Hink, Nicol, and Prosser 2017) to selectively inhibit and distinguish them. The relative contribution of AOA and AOB to N₂O generation was evaluated in soils with different ammonia concentrations using microcosm experiments and selective inhibitors (Hink, Nicol, and Prosser 2017; Wang et al. 2016) indicating that AOB, not AOA, dominate in both N₂O production and nitrification in soils amended with ammonium. AOB-dominated N₂O production has also been reported in several soils, including the bioenergy crop soil (Meinhardt et al. 2018; Pannu et al. 2019), urea fertilized soil (Chen et al. 2020) and paddy soils (Fu et al. 2020). On the other hand, Giguere et al. (2017) reported no significant differences in contribution of N₂O production between AOA and AOB in non-cropped Oregon soils (Giguere et al. 2017). AOA-dominated N₂O production has been observed depending on conditions such as urea levels. Duan et al. (2019) indicated that the dominant contribution of AOA and AOB to N₂O production was affected by the levels of urea application to the vegetable fields (Duan et al. 2019). Addition of urea at conventional or lower application levels induced higher AOB-dependent N₂O production than AOA, while excess urea application led to higher AOA-dependent N₂O production than AOB.

2.8. Control of nitrification

N₂O emissions and nitrogen leaching can be prevented by controlling nitrification. Nitrogen in agricultural fields has been managed through several indirect and direct nitrification control methods, such as controlled-release N fertilizers, nitrification inhibitors, and urease inhibitors (Norton and Ouyang 2019). Nitrification inhibitors, which are the most effective nitrification control method, directly inhibit the conversion of ammonium to nitrite by AOB by suppressing AMO activity and reducing the risk of nitrate leaching and N₂O emissions (Subbarao et al. 2006; Beeckman, Motte, and Beeckman 2018). Meta-analyses have reported that the application of nitrification inhibitors significantly reduces dissolved inorganic N leaching and N₂O emission (Akiyama, Yan, and Yagi 2010; Qiao et al. 2015; Xia et al. 2017). The N₂O mitigation effects of 2-chloro-6-(trichloromethyl) pyridine (nitrapyrin), dicyandiamide (DCD), and 3,4-dimethylepyrazole phosphate (DMPP), which are widely used in agriculture, were evaluated as 50%, 30%, and 50%, respectively (Akiyama, Yan, and Yagi 2010). On the other hand, the application of nitrification inhibitors with nitrogen fertilizer or animal urine could potentially increase NH₃ emission in some soils (Qiao et al. 2015). DCD residues have been detected in New Zealand milk products due to DCD spray application in pastures (Pal, McMillan, and Saggar 2016). Nitrapyrin was detected in water samples from streams across Iowa (Woodward et al. 2019). Therefore, from the viewpoint of environmental pollution, nitrification inhibitors should be fully considered before its use in the agricultural field.

The efficacy of nitrification inhibitors can be influenced by various environmental factors and, thus, vary across soils (Subbarao et al. 2006). In the case of DMPP, DMPP-amended urea fertilizer had no effect on N₂O emission of Australian commercial crops (Nauer et al. 2018), whereas DMPP reduced N₂O from 37% to 75% in vegetable soils (Scheer et al. 2014; Lam et al. 2018). Four contrasting arable soils including gray desert soil, alluvial paddy soil, loess formed paddy soil, and red soil in China were studied to understand the effects of soil properties on the efficacy of DMPP (Fan et al. 2019). DMPP significantly inhibited nitrification and reduced N₂O emissions, with an inhibitory rate ranging from 41.7% in red soil to 90.0% in gray desert soil. Reduction of N₂O emissions by DMPP is correlated with AOB abundance, which is a major contributor to N₂O emission rather than AOA. This study concluded that DMPP effectively inhibits nitrification by suppressing the abundance of AOB across soil types, and abiotic factors play a more important role than biotic factors in soil N₂O emissions. The response of AOA, AOB, and comammox to nitrification inhibitors has been evaluated using pasture and arable soils (Li et al. 2020). Nitrapyrin, DMPP, DCD, and allylthiourea markedly inhibited AOB abundance in both soils. Furthermore, the three nitrification inhibitors, except for DMPP, effectively suppressed AOA abundances in both soils. Comammox bacteria respond differently to the four nitrification inhibitors in both soils. The authors speculated that the different responses may be due to the different soil properties that can influence the absorption and degradation of nitrogen inhibitors. Moreover, AOA, AOB, and comammox bacteria have different susceptibilities to DMPP, nitrapyrin, and DCD, as evaluated using stable isotope labeling methods (Zhou et al. 2020).

Most commercially available synthetic nitrification inhibitors target and inhibit AMO, which mediates the first step of ammonia oxidation. However, AMO has not been purified so far, and there is limited information on this enzyme, such as a three-dimensional structure, which is necessary for the development of more effective nitrification inhibitors. Nishigaya et al. (2016) proposed the development of nitrification inhibitors targeting HAO (Nishigaya, Fujimoto, and Yamazaki 2016). The three-dimensional structure of HAO was determined based on structure-guided drug design using the information from 3D structures of the target molecule and the target-ligand complex for drug discovery. Alkyl- and aryl-hydrazine derivatives can irreversibly inactivate HAO by covalent modification at the active site of the enzyme (Logan and Hooper 1995). One of these derivatives, phenylhydrazine hydrochloride, effectively inhibits ammonia oxidization in soil slurry (Wu et al. 2012) and soil incubation experiments (Yang et al. 2017).

Biological nitrification inhibition (BNI) is defined as the ability of certain plant roots to produce and release nitrification inhibitors to suppress soil-nitrifier activity (Subbarao et al. 2015). Screening for BNI potential has been performed on various crops, and BNI ability was found for many crops such as rice, wheat, and forage grass (Sun et al. 2016; Nuñez et al. 2018; O’Sullivan et al. 2016). Recently, several plants, such as Megathyrsus maximus and Brachiaria hybrid cv., were shown to be able to suppress not only soil nitrification but also N₂O emission (Byrnes et al. 2017; Villegas et al. 2020). BNI holds promise as a strategy to conserve N in nutrient-poor environments (Subbarao et al. 2013).

3. Anammox

Anammox, discovered by Mulder et al. (Mulder et al. 1995) and Van de Graaf et al. (Van de Graaf et al. 1996) is a microbial process that oxidizes NH₄⁺ to N₂ with NO₂⁻ as an electron acceptor. The anammox process consists of three steps: NO reduction from NO₂⁻, hydrazine synthesis from NO and NH₄⁺, and N₂ gas dehydration from hydrazine (Jetten et al. 2009). Anammox has been detected in various environments, including marine, freshwater, and estuarine sediments, oxygen-depleted zones, wetland soils, and riparian and agricultural soil (Nie et al. 2019). The occurrence and activity of anammox bacteria have also been detected in paddy soils, and the anammox process has been reported to contribute ~37% to N₂ production in soil (Zhu et al. 2011). Furthermore, the anammox process is an important nitrogen loss pathway in paddy fields.

3.1. Anammox bacteria

Strous et al. (1999) obtained anammox bacterial suspensions (99.6% pure) physically separated from an enrichment culture through density-gradient centrifugation (Strous et al. 1999). The bacterium was responsible for anammox and CO₂ assimilation by incubating cell suspensions with ammonium, nitrite, and CO₂ under anaerobic conditions and was named Candidatus Brocadia anammoxidans and classified into the phylum Planctomycetes (Jetten et al. 2001; Strous et al. 1999). Although no pure cultures have been obtained, there are at least five candidate genera of anammox bacteria classified into the phylum Planctomycetes: Candidatus Brocadia, Candidatus Kuenenia, Candidatus Anammoxoglobus, Candidatus Jettenia and Candidatus Scalindua (Kuenen 2008). Candidatus Scalindua has been detected in marine environments and the other genera have mainly been sampled from freshwater activated sludge.

Essential functional genes for nitrogen transformation related to the anammox process are cytochrome cd₁-type NO-forming nitrite reductase (nirS), copper-containing NO-forming nitrite reductase (nirK), hydrazine synthase (hzs), and hydrazine dehydrogenase/ hydroxylamine dehydrogenase (hdh/hao). Neither nirS nor nirK was found in Ca. Brocadia sinica and related genomes (Lawson et al. 2017; Oshiki et al. 2015; Speth et al. 2016; Zhao et al. 2019). The hzs genes (hazA, hzsB, and hzsC) are unique for anammox bacteria and, thus, have been used as specific biomarkers. The hzsA and hzsB genes have often been used for the quantification of the abundance of anammox bacteria in environmental samples.

3.2. Anammox in soil environment

The anammox bacteria are more diverse in soil than in seawater water columns, as evidenced by molecular biology techniques (Humbert et al. 2010; Nie et al. 2019; Sonthiphand, Hall, and Neufeld 2014). Brocadia was most commonly detected in soil, followed by Kuenenia (Nie et al. 2019). The community composition of anammox bacteria in soil is affected by soil depth, fertilization treatment, agricultural land use, and cultivation history (Nie et al. 2019). Candidatus Jettenia and Candidatus Brocadia have been enriched from a nitrogen-loaded peat soil (Hu et al. 2011), and Candidatus Anammoxoglobus and Candidatus Jettenia have been enriched from a flooded paddy soil (Hu et al. 2013a) with 40–50% of the enrichment cultures.

Anammox potential in soil samples has been detected using ¹⁵N tracer techniques. These techniques have been used to detect the anammox process in various environments by the addition of ¹⁵NH₄⁺ and ¹⁴NO_x⁻ or ¹⁴NH₄⁺ and ¹⁵NO_x⁻ to produce ¹⁵⁺¹⁴N₂ (m/z 29) consisting of one N from NH₄⁺ and the other N from NO₂⁻, which can be distinguished from N₂ from the denitrification process. To date, many studies have reported anammox potential in natural and agricultural soils. The anammox process was detected in an enrichment culture from a nitrogen-loaded peat soil (Hu et al. 2011) and was detected directly in paddy soil samples (Sato et al. 2012; Zhu et al. 2011).

Anammox activity is most frequently detected in paddy soils, which presents superior anammox in comparison with other soils (Nie et al. 2019). Anammox activity is higher in oxic-anoxic interfaces in wetlands loaded with nitrogen fertilizer, suggesting the occurrence of coupled nitrification-anammox-denitrification (Nie et al. 2019). Anammox activity has also been detected in brackish natural environments such as wetland sediments (Zhao et al. 2021) and mangrove sediments (Amano et al. 2011; Li et al. 2018a). The anammox activity and hazA abundance of rice rhizosphere in paddy soil were higher than those in bulk soil (Nie et al. 2015). Thus, the rhizosphere in paddy soil could provide the environment to enable the occurrence of anammox activity.

Anammox potential was also detected in agricultural soil, such as vegetable soil (Long et al. 2013; Shen et al. 2015, 2017). Ammonium concentration in soil correlates with the abundance and activity of anammox bacteria (Shen et al. 2013a, Shen et al. 2015, 2017). On the other hand, there is also a report that nitrate concentration is negatively correlated with anammox abundance (Qin et al. 2020). Further studies are needed to elucidate the relationship between N content in arable soil and the abundance and activity of anammox bacteria. Anammox activity was also detected in temperate upland forest soil, and the contribution of anammox activity to total N₂ production was 1–7% in mixed forest (Xi et al. 2016). Total soil C and N, NO₃⁻ concentration and soil water content were positively correlated with the N₂ production rates by anammox plus codenitrification (Xi et al. 2016). Upland forests may contribute less to N₂ production by anammox than do N-loaded farmland soils. In acidic red soil, anammox rate was correlated with NO₃⁻, and hzsB and anammox bacterial 16S rRNA gene abundances were correlated with total nitrogen (Wu et al. 2018). The contribution of the anammox process to N₂ production ranged from less than 1% (Bai et al. 2015) to 53% (Wu et al. 2018) in natural and agricultural soil samples. Because anammox occurs under anaerobic conditions, anammox tends to be more active in wet soil where oxygen is scarce. Anammox activity is often correlated with the concentration of ammonia or nitrate in the soil, indicating that the supply of inorganic nitrogen has a great influence on anammox (Abbas et al. 2020).

4. Conclusions

In the last 15 years, there has been several breakthroughs regarding nitrifying microorganisms. For more than 100 years, nitrification has been thought to be performed only by AOB and NOB. The discovery of AOA and comammox overturned this theory, and these nitrifying microorganisms have been identified in various environments such as oceans and soils and have been shown to play an important role in the nitrogen cycle. On the other hand, the classical nitrifying microorganisms have been found to be more diverse, as shown by the classification of novel γ-AOB and NOBs into various phyla. In this review, we have outlined the diversity and functional characteristics of these novel and classic nitrifying microorganisms and presented published studies on the relationship between these nitrifying microorganisms and environmental factors in the soil. Genome and metagenome analyses have provided important and remarkable insights into the ecology and function of nitrifying microorganisms. However, more researches considering physiology, biochemistry, and enzymology of nitrifying microorganisms are needed for the development of technologies to control these organisms in different environments.

Disclosure statement

No potential conflict of interest was reported by the author(s).

Additional information

Funding

This work was financially supported by the project JPNP18016, commissioned by the New Energy and Industrial Technology Development Organization (NEDO), and KAKENHI Grant Numbers 19H01156 and 18K19194.