Effects of high nitrate input in the denitrification-DNRA activities in the sediment of a constructed wetland under varying C/N ratios

Abstract

In constructed wetlands (CW), denitrification usually accounts for >60% of nitrogen removal and is supposedly affected by the inflow water and the wetland management practices. Fluctuations in nutrient concentration of inflow water can cause an impact in sediment properties and microbial communities living therein. We have estimated the effects of a high input of nitrate or nitrite (simulating an eutrophication event) on dissimilatory nitrite reduction by analysing the activities of nitrite reducing bacteria (denitrification and dissimilatory nitrite reduction to ammonia) at different C/N ratios in compartmentalized microcosms. Denitrification was always the predominant pathway for nitrite removal (>60%) and eventually led to the complete removal of nitrate. Dissimilatory nitrite reduction to ammonia was negatively affected by the input of nitrogen, and more severely due to a transient increase of nitrite. Analyses of the nir genes sequences based on DNA and cDNA analyses revealed the importance of uncultured phylotypes as main contributors to nitrite reduction in wetlands. Our results highlight a high recovery rate of the ecosystem service after a severe event of potential eutrophication and point to metabolic redundancy of denitrifiers.

Introduction

The increased concentration of N-pollutants such as NH₄⁺, NO₃⁻ and NO₂⁻ in water resources and the direct and indirect effects of these compounds on environmental health makes nitrogen removal a critical step in water treatment processes (Schnobrich et al., 2007; Shrimali and Singh, 2001). Constructed wetlands have traditionally been considered as water polishing systems (Kadlec and Wallace, 2009; Vymazal, 2007) and over the past three decades they have gained a lot of popularity for small-scale conveniences as engineered water treatment ecosystems (Brix, 1994; Vymazal, 2013). Constructed wetlands basically mimic the functioning of natural systems, but processes for the improvement of water quality are regulated and forced to occur at higher rates compared to natural conditions. In CWs, nitrogen removal is achieved through a combination of microbial activities (coupled nitrification and denitrification), and plant assimilation (Vymazal, 2007). Nitrogen fixed as plant biomass has to be artificially removed (plant harvesting) in order to ensure elimination form the system. Microbial activity cause a net N elimination via dissimilatory reduction to gaseous compounds in a process called denitrification. During denitrification, nitrate (NO₃⁻) is anaerobically reduced to nitrogen gas (N₂) in a step-wise manner (Tiedje, 1988; Zumft, 1997). Although denitrification activity is the predominant pathway in most environments, other metabolisms, such as dissimilatory nitrite reduction to ammonia (DNRA) and anaerobic ammonia oxidation (anammox), may also have a role in nitrogen removal (Burgin and Hamilton, 2007; Koop-Jakobsen and Giblin, 2010; Tiedje, 1988). Collectively, nitrification and denitrification help preventing negative effects of eutrophication on constructed wetlands by an effective removal of nutrients. Similar to what may occur in natural environments, excessive eutrophication in constructed wetlands leads to accelerated oxygen consumption and organic matter accumulation in biofilms facilitating methanogenesis (Sánchez-Carrillo et al., 2010; Yang et al., 2020). In this sense, a balanced nitrogen removal capacity preventing ammonia accumulation in the water, is advisable for an effective performance of the water-treatment system.

Denitrification consists in three sequential enzymatic reactions catalysed by metalloproteins that differ between organisms (Park and Yoo, 2009). First, nitrite reduction to nitric oxide (NO) is catalysed by two structurally different but metabolically equivalent enzymes, the periplasmic copper containing nitrite reductase (NirK or Cu-NIR) and the haem containing nitrite reductase (NirS or cd1-NIR) (Jones et al., 2008). Nitrite reduction (Eq. 1), by either NirS or NirK-type nitrite reductases is a widespread metabolic trait among bacteria of very different taxonomic groups (Burghate and Ingole, 2013; Ligi et al., 2014; Philippot and Hallin, 2005). DNRA (Eq. 2), in contrast, has a significant relevance in the N cycle since it blocks true denitrification bypassing N removal by transiently producing a highly soluble compound, ammonia (Burgin and Hamilton, 2008; Tiedje, 1988; Yin et al., 2017). The reaction is catalysed by a molecularly distinct cytochrome c nitrite reductase, the NrfA protein (Simon, 2002; Welsh et al., 2014). DNRA ability in bacteria is rather diverse, and may function as a nitrite detoxification reaction under certain environmental conditions (Simon, 2002; Welsh et al., 2014).

Both denitrification and DNRA, although competing for the same substrate, are mainly anaerobic processes and require organic matter as electron donor (Eqs. 1 and 2). There are many environmental factors which influence the competition between Nir and NrfA containing bacteria, including labile organic carbon, nitrate availability, the ratio of electron donor/acceptor (carbon/nitrate), sulphide concentration, sand content, pH, microbial generation time, NO₃⁻/NO₂⁻, and temperature (Burgin and Hamilton, 2007; Friedl et al., 2018; Nizzoli et al., 2010; Papaspyrou et al., 2014). One of the main factors controlling the dominance of Nir over NrfA containing bacteria is the C/N ratio. Lindemann et al. (2015) showed that in eutrophic conditions, DNRA was disfavoured over denitrification in a study carried out in an urban estuary (Lindemann et al., 2015). In agreement, van den Berg et al. (2016) found a coexistence of DNRA and denitrification in environments at a relative wide range of C/N supply ratios, being DNRA favoured more than three fold in batch cultures from activated sludges (van den Berg et al., 2015). More, Palacin-Lizarbe et al. (2019) and Hardison et al. (2015) showed the importance of NO₃⁻ limitation to avoid an increase of DNRA activity specially in anoxic environments from marine and lake sediments (Hardison et al., 2015; Palacin-Lizarbe et al., 2019).

$$\text{Nitrite reduction}\mathrm{via}\text{denitrification}:{{\text{6NO}}_2}^{-}+{\text{3CH}}_3\mathrm{OH}➔3{\mathrm{N}}_2+{\text{3CO}}_2+{\text{3H}}_2\mathrm{O}+{\mathrm{OH}}^{-}$$

$$\text{Nitrite reduction}\mathrm{via}\text{DNRA}:{\text{2CH}}_2\mathrm{O}+{{\mathrm{NO}}_3}^{-}+{\mathrm{H}}_2\mathrm{O}➔{{\mathrm{NH}}_4}^{+}+{\text{2HCO}}_3$$

Measurements of metabolic rates under controlled laboratory conditions provides an estimate of the inherent activity of a given environment. Despite being limited for the extrapolation of the obtained results to the whole system, laboratory experiments can be determinant for several aspects, for instance confirmation of predicted metabolisms (Hernández-del Amo et al., 2019; Kim et al., 2016), or assessing plant litter decomposition in wetlands (Yarwood, 2018). In general, laboratory-controlled experiments performed in either meso- or microcosms, provide necessary data to evaluate the individual effects of different environmental parameters and to compare potential activities in simulated situations that may be difficult to test in the real system. Experimental activity analysis allows for a precise comparison of samples without the side effects of uncontrolled environmental variables and so may be useful to determine which are the best conditions for different microbial N cycle pathways. Potential activities can be measured by mass-balance calculations which can be easily done in laboratory microcosms. Time measurements of NO₃⁻, NO₂⁻, and ammonia concentrations can be used to assess nitrate and nitrite reducing activities (including denitrification and anammox) and NH₄⁺ production due to DNRA (Caffrey et al., 2019; Kim et al., 2016; Ruiz-Rueda et al., 2009; Song et al., 2014). In most CWs, and in spite of large variations according to prevalent environmental conditions, microbial N-reduction processes constitute 60–80% of the total nitrogen removal (Jahangir et al., 2014; Lee et al., 2009). We have previously shown (García-Lledó et al., 2011a; Hernández-del Amo et al., 2019) denitrification as the main pathway responsible for nitrite reduction in the Empuriabrava free water surface constructed wetlands (FWS-CW), though DNRA accounted for approximately 10% of the total nitrite reduction. Other studies have also pointed to a similar importance of DNRA, accounting for approximately 5 to 10% of NO₃⁻ removal (Scott et al., 2008; van Oostrom and Russell, 1994).

In this study, we aimed to assess the effect of TOC/TN ratio on nitrate and nitrite reduction activities after a sustained level-up of nitrate in a constructed wetland. Experiments were performed in non-mixed sediment-water interphases using samples from the Empuriabrava FWS-CW (NE, Spain). We combined activity measurements with molecular and culturing methods to study resident and active microbial communities responsible for the N removal in the system. We hypothesized that small changes in C/N fed in order to mimic an eutrophic situation could alter nitrite reduction activity. Our results pointed to a functionally redundant denitrifying community with a low number of active microorganisms able to reduce nitrite to gaseous compounds. DNRA activity, although being relatively low, was detected in almost all conditions and changed according to C/N ratios.