Microbial potential for denitrification in the hyperarid Atacama Desert soils

Highlights

• Denitrification was shown to occur in the hyperarid Atacama Desert.

• Denitrification potential and associated gene abundance was not affected by aridity.

• Increasing aridity reduced soil bacterial richness.

• Fungal and bacterial denitrification co-contributed to N₂O production.

• Bacterial denitrification dominated N₂O production with increasing hyperaridity.

Abstract

The hyperarid soils of the Atacama Desert, Chile, contain the largest known nitrate deposits in the world. They also represent one of the most hostile environments for microbial life anywhere in the terrestrial biosphere. Despite known for its extreme dryness, several heavy rainfall events causing localised flash flooding have struck Atacama Desert core regions during the last five years. It remains unclear, however, whether these soils can support microbial denitrification. To answer this, we sampled soils along a hyperaridity gradient in the Atacama Desert and conducted incubation experiments using a robotized continuous flow system under a He/O₂ atmosphere. The impacts of four successive extreme weather events on soil-borne N₂O and N₂ emissions were investigated, i) water addition, ii) NO₃⁻ addition, iii) labile carbon (C) addition, and iv) oxygen depletion. The ¹⁵N–N₂O site-preference (SP) approach was further used to examine the source of N₂O produced. Extremely low N₂O fluxes were detected shortly after water and NO₃⁻ addition, whereas pronounced N₂O and N₂ emissions were recorded after labile-C (glucose) amendment in all soils. Under anoxia, N₂ emissions increased drastically while N₂O emissions decreased concomitantly, indicating the potential for complete denitrification at all sites. Although increasing aridity significantly reduced soil bacterial richness, microbial potential for denitrification and associated gene abundance (i.e., napA, narG, nirS, nirK, cnorB, qnorB and nosZ) was not affected. The N₂O¹⁵N site preference values based on two end-member model suggested that fungal and bacterial denitrification co-contributed to N₂O production in less arid sites, whereas bacterial denitrification dominated with increasing aridity. We conclude that soil denitrification functionality is preserved even with lowered microbial richness in the extreme hyperarid Atacama Desert. Future changes in land-use or extreme climate events therefore have a potential to destabilize the immense reserves of nitrate and induce significant N₂O losses in the region.

Introduction

The Atacama Desert represents one of the most hostile environments for microbial life on Earth due to its hyperarid moisture regime, thermal extremes, high concentrations of salt in the soil, and intense UV radiation at the soil surface (Armando Azua-Bustos et al., 2012; Calderón et al., 2014). This makes the Atacama Desert an ideal location to study the potential for life to exist on other planets, e.g. Mars, which are thought to possess similar soil properties (McKay et al., 2003; Valdivia-Silva et al., 2012). Nevertheless, the Atacama Desert is not totally devoid of life. Short term water inputs caused by fog or rare rainfall events may provide temporary favourable conditions for microorganisms and plants (i.e. “desert blooms”; Orlando et al., 2010). Recent studies have shown that microbial communities in these soils respond rapidly to the addition of available carbon (C) when moisture limitation is removed (Jones et al., 2018). Conversely, the addition of water has also been shown to compromise organisms adapted to a xerophilic lifestyle (Azua-Bustos et al., 2018).

Biological denitrification, which is the major nitrogen (N) loss mechanism in terrestrial ecosystems, occurs via the sequential reduction of NO₃⁻ to NO₂⁻, NO, N₂O and N₂ (Firestone, 1982). Denitrification was generally considered to be a prokaryotic process for a century, while it becomes clear nowadays that both bacteria and fungi possess the ability to produce N₂O during denitrification (Philippot et al., 2007; Laughlin and Stevens, 2002). Nevertheless, the relative contribution of fungi to soil N₂O production remains poorly understood, especially in natural ecosystems (Senbayram et al., 2018; Xu et al., 2019). Although the occurrence of denitrification is widespread in moist soils, its functional significance in hyperarid ecosystem remains unknown. In the last five years, several extreme rainfall events occurred in the Atacama Desert, causing localised flash flooding (Wilcox et al., 2016; Azua-Bustos et al., 2018; Schulze-Makuch et al., 2018). Since the Atacama Desert contains the largest known nitrate deposits in the world, the susceptibility of these reserves to denitrification after land-use change or extreme climate events remains unclear (Michalski et al., 2004; Walvoord et al., 2003). Despite the fact that microbial communities in the Atacama Desert are of low abundance and low diversity, recently denitrification–related genes were detected (Knief et al., 2020; Orlando et al., 2012). Nevertheless, it is still not clear whether the environmental conditions in the Atacama Desert can support denitrifier activity, or whether those denitrification–related genes were introduced by atmospheric long-distance transport and deposition of bacterial cells (Mayol et al., 2017). In this study, we hypothesized that i) soils along a hyperarid gradient in the Atacama region will differ in microbial community composition, driven by differences in moisture availability; (ii) the capacity for complete denitrification will be preserved in the Atacama Desert soils but will decline with increasing aridity; ii) the response of denitrification-derived N₂O and N₂ emissions to highly ephemeral weather events (i.e. rainfall) will be slow and that this response will be bacterially driven.

The combination of a δ¹⁸O and site preference (δ¹⁵N^sp) approach has been widely used during the last decade to distinguish the different sources of N₂O production pathways [+34‰ to +40‰ for nitrification (Ni) and fungal denitrification (fD), −9‰ to +9‰ for bacterial denitrification (bD)] (Decock and Six, 2013; Toyoda et al., 2017). In this study we combined this isotopic mapping approach with qPCR and high-throughput 16S rRNA gene sequencing to (i) quantify rates of denitrification in a contrasting range of hyperarid soils, (ii) measure the abundance of functional genes involved in denitrification, (iii) determine whether denitrification could be attributed to fungi or bacteria, (iv) identify key abiotic factors that may limit denitrification, and (v) evaluate the theoretical potential for microbial-mediated loss of nitrate reserves in the Atacama region.