Nitratereducing processes


Denitrification is the reduction of NO3~ or NO2~ to N2 under anaerobic conditions with N2O and NO emitted as intermediary gaseous products (Robertson and Tiedje, 1987; Bremner, 1997). The process is catalysed by the enzymes nitrate reductase, nitrite reductase, nitric oxide reductase and nitrous oxide reductase, and the transport of electrons to NO3~ or other N oxides is coupled to the synthesis of ATP (adenosine triphosphate) (Hochstein and Tomlinson, 1988). The nature and regulation of the reductase enzymes involved in denitrification have been well characterized (Zumft, 1997).

Denitrifiers are predominantly heterotrophic bacteria with the denitrification trait being widespread in more than 50 genera (Smith and Zimmerman, 1981; Knowles, 1982). Some archaea and fungi, such as Fusarium (Shoun et al, 1992), were also shown to be capable of denitrifying.

Since the ability to denitrify is sporadically distributed both within and between different genera, the genes encoding the catalytic sub-unit of the different denitrification reductases are commonly used as molecular markers in studies investigating the diversity of denitrifiers. Thus membrane-bound and periplasmic nitrate reductase encoding genes (napA, narG) (Gregory et al, 2000; Philippot et al, 2002), cytochrome cd1 and Cu-containing nitrite reductase encoding genes (nirS, nirK) (Hallin and Lindgren, 1999) and N2O reductase encoding genes (nosZ) (Stres et al, 2004) have successfully been used in terrestrial environments for fingerprinting analyses to identify the environmental factors driving the denitrifier community composition.

The activity of denitrifying bacteria is controlled by C availability, O2 concentration (mainly through soil water content), N availability, pH and temperature. The contribution of denitrification to N2O emissions from soils is traditionally thought to be greatest under sub-oxic conditions, and in soils at a water-filled pore space (WFPS) >70 per cent where soil NO3~ and available C are non-limiting (Davidson, 1991; Bateman and Baggs, 2005). High O2 concentrations are known to suppress the activity and synthesis of the denitrification reductases. The N2O reductase is thought to be the most sensitive to O2 (Otte et al, 1996), reflected in a higher N2O/N2 ratio with increasing O2 availability (Weier et al, 1993). When re-exposed to O2 after an anaerobic phase, all denitrification enzymes but the N2O reductase remain active (Morley et al, 2008), and conversely when aerobic soils become anaerobic, such as following heavy rainfall, the NO3~ and NO2~ reductases are typically activated sooner than the N2O reductase, so that the denitrifier N2O/N2 ratio is higher up to one or two days after rainfall (Knowles, 1982). Several bacteria isolated from soils and sediment are capable of denitrifying in the presence of O2 (Lloyd et al, 1987; Patureau et al, 2000). Although N2 and N2O production is common during aerobic denitrification by cultured isolates (Patureau et al, 2000; Takaya et al, 2003) and aerobic denitrifiers are thought to be present in high numbers in soils (Pa-tureau et al, 2000), their contribution to N2O emissions from soils has yet to be proven, and in the meantime, denitrification is still modelled as an anaerobic process.

Denitrification rates are influenced by the availability of N electron acceptors, and this is why they are stimulated by addition of N, usually in the form of inorganic fertilizers (Eichner, 1990). N availability also affects the N2O/N2 ratio, which can be lowered with NO3~ concentrations >10pg g_1 soil, where NO3~ is preferred over N2O as an electron acceptor (Blackmer and Bremner, 1978; Baggs et al, 2003). The enzyme kinetics and affinity for substrates by denitrifiers is highly variable, possibly reflecting the wide phylogenetic diversity of this group, with Km values for N2O reduction of between 0.1 and 100^M (Conrad, 1996; Holtan-Hartwig et al, 2000).

As most denitrifiers are heterotrophic, the availability of organic C, which can be used as an electron donor, is also a major factor determining denitrification rates (Knowles, 1982; Bremner, 1997). However, organic C availability is the least well-understood factor controlling denitrifier N2O production and reduction, despite the fact that it has traditionally been thought a reliable index for predicting a soil's denitrification capacity (Burford and Bremner, 1975). The effects of C are not only mediated directly, but also indirectly through the creation of anaerobiosis generated as a result of microbial respiration (Azam et al, 2002). With regards to direct effects, there is likely to be strong coupling between plant exudate-C and denitrification, through the provision of the reductant for denitrification, and changing the size, structure and activity of the denitrifier community. Indeed, Smith and Tiedje (1979) found a positive effect of roots on denitrification when soil NO3~ concentrations were high, which was thought to be due to release of organic C from the roots, with denitrifying activity decreasing rapidly in the first few millimetres away from the roots. Denitrification has been shown to increase following addition of easily oxidizable organic matter to soils, often very rapidly, and significant correlations have been reported between total organic C (Baggs and Blum, 2004), 'available' C (Stanford et al, 1975), water-soluble and 'mineralizable' C (Burford and Bremner, 1975; Paul and Beauchamp, 1989) and soluble C-to-N ratio of applied residues (Millar and Baggs, 2005). However, little is known about the effect of individual C compounds on the regulation of the enzymes producing or reducing N2O, and we are still unsure whether the type of C compounds is an important driver of the denitrifier community. In a recent study, Henry et al (2008) applied artificial root exudates composed of different combinations of sugars, organic acids and amino acids to soil microcosms, and showed that a higher proportion of sugars resulted in a lower N2O/N2 ratio. Such results warrant further attention before we can propose that manipulating plant-derived C flow has the potential to increase reduction of N2O to N2 (Philippot et al, 2009b).

Among the other factors influencing denitrification, special emphasis has been placed on soil pH. Denitrification rates are usually thought to decrease with pH, but significant denitrification rates can be still observed below pH 4.9 (Ellis et al, 1998). This can reflect differences in the composition of the denitrifier community (Enwall et al, 2005) and adaptation of denitrifiers to low soil pH (Parkin et al, 1985). Interestingly, soil pH does not only affect denitrification rates but also the N2O/N2 ratio. Thus, several studies reported that the N2O reductase enzyme is more sensitive to low pH than the other denitrification reductases, resulting in higher ratios of N2O to N2 as pH declines (Firestone et al, 1980; Nägele and Conrad, 1990; Thomsen et al, 1994; Simek and Cooper, 2002).

Fungal denitrification

The potential for fungi to produce N2O has been demonstrated in woodland

(Laverman et al, 2000) and in grassland soils (Laughlin and Stevens, 2002). N2O release from acid coniferous forest soils is often low and attributed to nitrification (Martikainen et al, 1993; Sitaula and Bakken, 1993), but there is evidence that fungal denitrification may actually be important in these systems (Kester et al, 1996; Laverman et al, 2000). Several fungal strains have been shown to possess cytochromes P450mr or P450nor that enable them to denitrify and release N2O when anaerobically incubated with NO2~ in culture (Shoun et al, 1992; Takaya and Shoun, 2000). Fungal nir appears to be expressed in the mitochondrion and is analogous to the bacterial copper-containing enzyme (Kobayashi and Shoun, 1995), whereas the fungal nor (P450nor) gene differs from that found in bacteria. In fungal denitrification the enzymes accept two electrons from NADH or NADPH without involving a membrane-bound electron transfer chain as in bacterial denitrification. In most cases the main product of fungal denitrification is N2O, as many fungi lack a N2O reductase (Shoun et al, 1992), although it is possible that the P450nor gene has two functions and may also facilitate N2 production (Takaya and Shoun, 2000).

Nitrifier denitrification

In nitrifier denitrification, the oxidation of NH3 to NO2~ by ammonia-oxidizing bacteria (AOB) is followed by the reduction of NO2~ to N2O and possibly also to N2(Wrage et al, 2001), although there is still no direct evidence that ammonia oxidizers can produce N2 during this process. The entire process is undertaken by ammonia-oxidizing bacteria (Kuai and Verstraete, 1998) and therefore contrasts with 'coupled nitrification-denitrification' where coexisting nitrifiers and denitrifiers can together transform NH3 to NO3~ and NO3~ to N2, respectively. The enzymes involved in nitrifier denitrification are essentially those for ammonia oxidation and denitrification, namely ammonia monooxy-genase, hydroxylamine oxidoreductase, nitrite reductase, nitric oxide reductase and, possibly, nitrous oxide reductase (Jiang and Bakken, 1999). This process has been demonstrated in a number of studies on mixed and pure cultures of Nitrosomonas, with NH3 converted under oxygen limitation to N2O and N2, with NO2~ and NO as intermediates (Abeliovich and Vonshak, 1992; Bock et al, 1995; Kuai and Verstraete, 1998). Shaw et al (2006) provided direct evidence for reduction of exogenously applied 15N-NO2~ to 15N-N2O in seven strains of ammonia-oxidizing bacteria representative of clusters 0, 2 and 3 in the cultured Nitrosospira lineage, with up to 13.5 per cent of measured N2O derived from the applied 15N-NO2~ (Table 2.1). This suggests that the ability to denitrify may be a widespread trait amongst betaproteobacterial ammonia-oxidizing bacteria, but its significance as a N2O source in soil has still to be proven as it requires advances in techniques for its quantification (see below). Ammonia-oxidizing archaea have been shown to possess a nitrite reductase enzyme, and although they have been demonstrated to oxidize ammonia (Prosser and Nicol, 2008), any ability to denitrify is as yet unproven, but may well be akin to nitrifier denitrification.

Table 2.1 Ammonia oxidizer 14+15N-N2O and 15N-N2O production rates and molar yields expressed as a percentage of the NO2~ and 14+15N-N2O production rate, respectively

AOB strain

Total N2O

Yield of


Yield of




15N2O on a

rate (amol*

on a nitrite

rate (amol*

total 14+15N-

14+15N-N2O h-1

basis* (%)

15N-N2O h-1

N2O basis*




N. europaea ATCC 19718





N. europaea ATCC 25978

1 5.5b




N. briensis strain 128





N. multiformis ATCC 25196





N. tenuis strain NV12





Nitrosospira sp strain 40KI





Nitrosospira sp strain En13





Nitrosospira sp strain NpAV





Note: Mean values in columns superscripted by different letters are significantly different (P<0.05). *amol = attomol = 10-18 mol.

t Values are the means of three independent experiments. $ Values are the means (n = 4) for cells harvested from a single independent flask. **Rate of nitrite production by Nitrosospira sp 40KI was not significantly different from 0 (P>0.05); therefore percentage yield of 14+15N-N2O on a nitrite basis was not calculated. Source: Shaw et al (2006)

Note: Mean values in columns superscripted by different letters are significantly different (P<0.05). *amol = attomol = 10-18 mol.

t Values are the means of three independent experiments. $ Values are the means (n = 4) for cells harvested from a single independent flask. **Rate of nitrite production by Nitrosospira sp 40KI was not significantly different from 0 (P>0.05); therefore percentage yield of 14+15N-N2O on a nitrite basis was not calculated. Source: Shaw et al (2006)

Little is known about the environmental regulation of nitrifier denitrification and effects on N2O production during this process. Several studies provide evidence for greater N2O production through nitrifier denitrification as oxygen concentration decreases (Goreau et al, 1980; Lipschultz et al, 1981; Hynes and Knowles, 1984; Kester et al, 1996; Jiang and Bakken, 1999; Dundee and Hopkins, 2001), most likely because NO2- is used as an alternative to O2 as an electron acceptor for microorganisms temporarily subjected to anaerobic conditions (Ritchie and Nicholas, 1972). Decreasing pH may also influence nitrifier denitrification (Jiang and Bakken, 1999; Wrage et al, 2001). From our ongoing experiments we estimate that nitrifier denitrifier-N2O accounts for a daily loss of fertilizer-N of up to 0.2 per cent and have found that although N2O emissions were raised following amendment of soil with amino acids and artificial root exudates, the proportional contribution of nitrifier denitrifi-cation did not vary with the different C amendments.


Here one N atom from NO or N2O combines with one atom from another source (a co-substrate) forming a hybrid product (Tanimoto et al, 1992; Su et al, 2004). N2 can also be produced solely from the co-substrate (Tanimoto et al, 1992) although it is not known how common this is. Co-denitrification has been shown to occur in both fungi (Tanimoto et al, 1992) and bacteria (Garber and Hollocher, 1982) but the reasons why, or the conditions under which, an organism would utilize a co-substrate are unknown. Co-denitrification and conventional denitrification have been shown to occur simultaneously in some denitrifying bacteria such as Pseudomonas stutzeri and denitrifying fungi (Shoun et al, 1992; Tanimoto et al, 1992). Su et al (2004) demonstrated the P450nor in Fusarium oxysporum to be multifunctional and able to catalyse denitrification or co-denitrification. There is some evidence for co-substrate preference between different fungal strains, which may indicate different mechanisms. For example, Fusarium oxysporum only produced N2O during co-denitrification when azide or salicylhydroxamic acid was available (Tanimoto et al, 1992; Su et al, 2004) but other fungi, for example Fusarium solani and Cylindrocarpon tonkinense (now Fusarium lichenicola), have been shown to emit N2 as a co-denitrification product when amino acids are available (Shoun et al, 1992).

The contribution of co-denitrification to N2O emission from soils is unknown, as there has been no targeted source determination for this process in situ, in part due to difficulties in distinguishing co-denitrification from denitrification. The only existing evidence from soil is from Laughlin and Stevens (2002) who estimated 92 per cent of measured N2 production in a grassland soil (sandy loam, pH 6.3) amended with glucose (up to 15mg per g soil) to result from co-denitrification, with the remaining 8 per cent attributed to conventional denitrification.

Nitrate ammonification ordissimilatorynitrate reduction to ammonium (DNRA)

During nitrate-dependent ammonification NO3~ is reduced to NO2~ and NH4+ (Figure 2.1) thereby providing a 'short circuit' in the nitrogen cycle bypassing denitrification and N2 fixation (Mohan et al, 2004). The process of nitrate ammonification is in many cases coupled to a respiratory electron transport system through which energy conservation (and thereby ATP synthesis) can be achieved. However, a range of electron donors can be used by ammonifiers, including fermentation products such as lactate, formate and hydrogen, as well as NADH produced from intermediary metabolism, and the nature of electron donor influences the level of energy conservation. This process is undertaken by both Gram-negative and Gram-positive bacteria, including obligate anaerobes (e.g. Clostridium spp), facultative anaerobes (e.g. Enterobacter spp) and aerobes (e.g. Bacillus spp) (Fazzolari et al, 1990), which are widely distributed in many different environments. Some of these bacteria are capable of reducing NO2~ to N2O during nitrate ammonification, as well as to NH4+ (Kelso et al, 1997). Two biochemically distinct nitrate reductases, the membrane-bound nitrate reductase (Nar) and the periplasmic nitrate reductase (Nap), are found in ammonifying bacteria. These two enzymes can also catalyse the first step of the denitrification pathway. Nitrite is then reduced to ammonium by NrfA, which can be N2O-genic (Jackson et al, 1991). However, the biochemistry that determines the amount of N2O formed is still unknown. The contribution of nitrate ammonification to N2O emissions from soils is also pretty much unknown, mainly because of technical limitations in its determination, which are discussed below.

It is possible that both nitrate ammonification and denitrification can occur simultaneously in anaerobic micro-sites in soils (Fazzolari et al, 1990) and produce N2O at the same time (Smith and Zimmerman, 1981; Stevens et al, 1998) and may compete for NO3~, depending on environmental conditions as well as the soil microbial population. Compared to denitrification, very little is known about the regulation of nitrate ammonification. Intensively reduced and C-rich environments are traditionally thought to favour ammonification, as it is a more efficient electron acceptor where C availability is non-limiting (Buresh and Patrick, 1978; Tiedje et al, 1982). Addition of glucose to anoxic soil has been shown to result in ammonification of 43 per cent of nitrate applied (Caskey and Tiedje, 1979), and there is evidence that nitrate ammoni-fication may be significant in agricultural soils, even those to which no C has been added (Chen et al, 1995a, 1995b).

0 0

Post a comment