Influence of biological nitrification inhibition by forest tree species on soil denitrifiers and N₂O emissions

Highlights

• Trees that inhibit nitrification increased the limitation of denitrification by N.

• Denitrification and in situ N₂O emissions were not affected by soil NO₃⁻ levels.

• Denitrification was more limited by available C than NO₃⁻ levels under some trees.

• In situ N₂O emissions were related to ratio of nir-to-nos genes, pH and moisture.

Abstract

Some forest tree species are able to carry out a process known as biological nitrification inhibition, BNI, i.e. they inhibit nitrifiers through the production of specific compounds. We tested the hypothesis that, by restricting N supply to NO₂⁻- and N₂O-reducers, BNI would decrease potential N₂O production and consumption and in situ N₂O emissions. as compared to soils under trees without BNI capacity. Soils were collected from long-term monocultures (>43 ys) of three tree species without BNI capacity (Fagus sylvatica, Pinus nigra and Pseudotsuga menziesii) and two tree species with BNI capacity (Abies nordmanniana and Picea abies). The level of limitation of denitrification by NO₃⁻ was high for species with BNI capacity and low for species without BNI capacity, and was correlated with potential nitrification rates and the abundances of genes specifically harboured by ammonia oxidizing archaea and Nitrobacter. However, potential denitrification and actual N₂O emissions did not reflect the tree BNI status, and denitrification limitation by soil carbon was higher than limitation by N under three tree species. Structural equation modelling revealed that the ratio between the gene copy abundances of nitrite-reducers and N₂O-reducers was the microbial variable that best explained N₂O emissions, along with soil pH and moisture. In addition, the NO₃⁻ concentration in the soil solution at 60 cm depth increased with the potential nitrification-to-denitrification ratio, suggesting a higher risk of NO₃⁻ leaching under some tree species like Douglas fir.

Introduction

Nitrification, i.e. the oxidation of NH₄⁺ to NO₂⁻ and NO₃⁻, and denitrification, i.e. the sequential reduction of NO₃⁻ and NO₂⁻ into the gaseous compounds NO, N₂O and N₂, represent key processes determining the availability and form of nitrogen, N, in soils. They are major determinants of soil N forms available for plants (Stuart Chapin et al., 2011) and N losses from ecosystems through NO₃⁻ leaching and emissions of NO/N₂O (Firestone and Davidson, 1989), two potent greenhouse gases (Baggs, 2011).

In soils with low pH and low N availability, typical for most forests (He et al., 2012; Hu et al., 2015), plants have developed diverse strategies to effectively utilize the scarce N resource (Bardon et al., 2018; Boudsocq et al., 2009; Chapman et al., 2006; Vitousek and Sanford, 1986). In particular, some plant species are able to inhibit nitrifiers through a process known as biological nitrification inhibition, BNI (Laffite et al., 2020; Lata et al., 2004; Srikanthasamy et al., 2018; Subbarao et al., 2009) in relation with the production of specific compounds by roots or litter decomposition (Coskun et al., 2017; Subbarao et al, 2006, 2015). In forest ecosystems, it has been shown that particular tree species are associated with very low soil nitrification rates (Andrianarisoa et al., 2010) and that these species inhibit the growth and abundance of specific nitrifier groups, i.e. Nitrobacter, also influencing ammonia-oxidizing archaea, AOA (Laffite et al., 2020). However, it is unknown whether the low NO₃⁻ amounts associated to species with BNI capacity also affect denitrification and the main soil microbial groups determining the balance between N₂O production and consumption, i.e. nirK- and nirS-harbouring NO₂⁻-reducers, and nosZ1-or nosZ2-harbouring N₂O-reducers (Assémien et al., 2019; Domeignoz-Horta et al., 2017; Florio et al., 2019), which would explain a cascading effect of BNI on N₂O emission from soil. Such an influence of plant species on N₂O emissions has been reported for herbaceous plants (Niklaus et al., 2016).

So far, little information is available on N₂O emissions from soils and denitrifier activities under forest plantations with contrasting BNI capacity. In general in situ N₂O emissions are higher for soils under deciduous than coniferous forests (Ambus et al., 2006). For instance, nitrate production and reduction were tightly coupled in soils under coniferous forests (Stark and Hart, 1997), and both processes are related to soil C and N statuses (Menyailo and Huwe, 1999) and influence soil N₂O-reducers activity (Hénault et al., 1998). This suggests that N availability could be one of the main determinants of denitrification and N₂O emissions in forest ecosystems. A wide range of limitation of denitrification by soil N has been observed in different terrestrial ecosystems (Peterjohn and Schlesinger, 1991; Florio et al., 2017). However, to what extent NO₂⁻- and N₂O-reducers are sensitive to the modification of N availability linked to BNI capacity of tree species, and whether this induces cascading effects on N₂O emissions from soil remains unclear.

Here, we identified the major determinants of the activities and gene copy abundances of NO₂⁻- and N₂O- reducers and of in situ N₂O emissions across different forest tree species with contrasting BNI capacity. A straightforward hypothesis is that low soil NO₃⁻ availability associated to species with BNI capacity would decrease denitrifier abundances and activities, and that without major change in the balance between NO₂⁻-reducers and N₂O-reducers this would decrease N₂O emissions from soil (Fig. 1). However, denitrification can also depend on other soil environmental variables, including the level of anaerobiosis (largely modulated by soil moisture), soil pH, and the amount of readily available C sources (Firestone and Davidson, 1989; Zumft, 1997). Would one of these additional factors be more limiting for denitrifiers than NO₃⁻ availability, denitrification and N₂O emission rates could be largely decoupled from the BNI status of trees. To test these hypotheses, we studied three tree species identified as without any BNI capacity, i.e. beech, Corsican pine and Douglas fir, and two tree species identified as having BNI capacity, i.e. Nordmann fir and spruce (Laffite et al., 2020) at the Breuil-Chenue experimental site (Legout et al., 2016). We measured the potential activities and gene copy abundances of the two nitrifier groups driving nitrification in these forest soils, i.e. archaeal ammonia oxidizers, AOA, and Nitrobacter (Laffite et al., 2020), and of denitrifiers (nirK- and nirS-harbouring NO₂⁻-reducers, and nosZ1-and nosZ2-N₂O-reducers), along with in situ N₂O emissions from soils. Because bacterial ammonia oxidizers, AOB, might contribute to N₂O emission rates (Tzanakakis et al., 2019), we also quantified copy abundance for the AOB-amoA gene. In addition, we measured soil environmental parameters (pH, moisture, NH₄⁺ and NO₃⁻ concentrations) and the level of denitrification limitation by N or C. Structural equation modelling, SEM, was used to explore the links between N₂O emissions, microbial activities, microbial gene copy abundances, and soil environmental parameters.