Is site preference of N₂O a tool to identify benthic denitrifier N₂O?

Nitrous oxide is a potent greenhouse gas that participates in ozone layer destruction.^[1] Atmospheric N₂O concentrations have increased ~20 % with the rise of anthropogenic activities, such as the intensive use of fertilisers in agriculture areas.^[2] In soils and sediments, the two main processes responsible for N₂O emissions are nitrification^[3] and denitrification.^[4] Mediated by denitrifying bacteria under anoxic conditions, denitrification emits localised and instantaneous N₂O as an intermediate product by reduction of nitrate (NO₃^–) to dinitrogen (N₂).^[5] The linear asymmetric N₂O molecule has a site preference (SP) that is defined by the intramolecular ¹⁵N difference between N^α (central N) and N^β (terminal N).^[6] Determined by natural stable isotope ratios (δ¹⁵N and δ¹⁸O), the N₂O site preference (N₂O-SP) signature has been proposed and used as an emerging tool to define the origin of N₂O.^[7,8] Pure culture studies, using denitrifying bacteria that reduce NO₃^– to N₂O, have shown that the N₂O-SP signature for denitrification ranged between –5 and 0 ‰.^[7,8] In addition to this, several denitrification studies have investigated the N₂O-SP signature originating from soil and water incubations either with N₂O as an end product, substrate or intermediate and found a large variation between –8.5 to 81 ‰.^[9,10] Part of this variation has been shown to be due to the fact that the N₂O consumption during denitrification generates an increase of the SP values.^[10,11]

Even though these investigations have increased the knowledge regarding the N₂O-SP signature associated with denitrification, several questions remain unanswered. The N₂O-SP signature of environmental denitrifying communities is, for example, lacking. In order to bridge the gap between the investigations on denitrification in pure cultures, the focus of this study was to determine N₂O-SP signatures of environmental benthic denitrifying communities under controlled conditions. More specifically, we determined SP values of N₂O produced by an environmental benthic denitrifying community during the different steps of benthic denitrification (either the production of N₂O or during complete denitrification) under denitrifying conditions (anoxia, nitrate supply).

The study was conducted in an open system using flow-through reactor (FTR) experiments.^[12] The use of these reactors allows the determination of denitrification-derived N₂O by nitrate supply and creating anoxic conditions, and thereby exclude nitrification as an additional N₂O source. Nitrate reduction rates and the accompanying N₂O-SP signatures were determined simultaneously, allowing direct comparison.

During benthic denitrification in two types of sediments, nitrate reduction rates and the N₂O-SP values were determined in an open system using FTRs (Fig. 1). The FTRs are described in detail in Laverman et al.^[12] Sediments were collected from the Charmoise River (48°36′37.80″N, 2°9′19.54″E, Essonne, France), which is 7.5 km long, in October and November 2011 and in the Manche-à-Eau lagoon (16°16′38.04″N, 61°33′27.93″W, Guadeloupe, France), which has an area of 260 000 m², in March 2012. In FTR experiments, denitrifying conditions were obtained by supplying anoxic (N₂ bubbled) nitrate solutions (infinite reservoir) to the sediments.

Fig. 1.  Overview of the experimental ‘flow-through reactor’ system; containing sediment of known volume (V = 13.85 cm³), supplied with an anoxic inflow solution containing nitrate ([NO₃^–] in), at a known flow rate (Q). Nitrate reduction rates are determined from the measured concentration difference between inflow and outflow nitrate (ΔC), once the outflow concentration of NO₃^– has stabilised (‘steady state’ conditions). Liquid inflow and outflow solutions were sampled during 3 or 6-h intervals for 3 (river sediment) and 4 days (lagoon sediment) to determine N₂O site preference (N₂O-SP) values or NO₃^– concentrations.

In order to determine the SP value of N₂O as end product during NO₃^– reduction to N₂O, by the environmental benthic denitrifying community, acetylene (C₂H₂) was used to block N₂O reductase, the enzyme responsible for the N₂O reduction to N₂.^[13] In addition to the determination of the SP of N₂O as an end product, the SP of N₂O appearing as an intermediate was measured during different microbial denitrification rates in two types of sediments (river and lagoon). In order to obtain different nitrate reduction rates, diverse nitrate concentrations were supplied to the sediment by the inflow solutions (Fig. 1). During NO₃^– reduction to N₂O (acetylene block) 5 mM nitrate was supplied, while a range of concentrations was applied during benthic denitrification in river sediment (1, 1.5, 3 and 5 mM) and in lagoon sediment (0.83, 1, 3 and 5 mM), in order to achieve different nitrate reduction rates. The denitrification rate depends not only on NO₃^– concentrations, as substrate, but also on organic carbon (C_org) content. Determined by the difference between the total carbon and inorganic carbon contents, the C_org content was on average 6.3 ± 0.07 % (n = 5) in the river sediment and 12.1 ± 0.04 % (n = 6) in the lagoon sediment. These amounts of carbon are sufficient to allow denitrification over the time the sediments were incubated in our experiments.^[14]

Nitrate concentrations were determined by high-performance liquid chromatography (HPLC, Dionex, AS12 column; Thermo Scientific, Sunnyvale, CA, USA) for samples of benthic reduction of NO₃^– to N₂O and by colourimetric measurements using an autoanalyser for samples of complete denitrification within river sediment (Quaatro, Bran & Luebbe, Plaisir, France) and within lagoon sediment (Gallery; Thermo Fisher Scientific, Cergy-Pontoise, France).

Prior to the determination of benthic denitrification derived N₂O, the N₂O-SP signature of the nitrate reduction to nitrous oxide by azide was tested with a laboratory standard. This method, which allowed the isotopic composition of nitrate and nitrite to be obtained, consists of nitrate reduction to nitrite through a column containing activated cadmium; consequently, nitrite is reduced to nitrous oxide by sodium azide solution (NaN₃, δ¹⁵N = –2.9 ± –0.4, n = 10) and the last step (N₂O to N₂) is blocked by the addition of sodium hydroxide.^[15]

All liquid samples were analysed using an isotope ratio mass spectrometer (IRMS, DeltaVplus; Thermo Scientific, Bremen, Germany) in continuous-flow with a purge-and-trap system and coupled with a Finnigan GasBench II system (Thermo Scientific), in order to obtain the stable isotope composition of N₂O (δ¹⁵N–N₂O, δ¹⁸O–N₂O, δ¹⁵N–NO and δ¹⁸O–NO (N^α). The GasBench II allows sampling of the gases and removal of the water contained in the liquid samples, and finally separation of the interfering gases of N₂O using a chromatographic column (CP-poraPLOT U; Thermo Scientific). The N₂O-SP values were calculated based on the raw values of δ¹⁵N–bulk and δ³¹NO (δ¹⁵N^α–N₂O), which were measured by the IRMS from the same N₂O reference gas. Nitrate standards were used to calibrate the isotopic composition of N₂O (USGS34, δ¹⁵N = –1.8 ‰, δ¹⁸O = –27.9 ‰, USGS35, δ¹⁵N = +2.7 ‰, δ¹⁸O = +57.5 ‰ and USGS32, δ¹⁵N = +180 ‰, δ¹⁸O = +25.7 ‰). The precision is 0.5 ‰ for δ¹⁵N and 1 ‰ for δ¹⁸O. Statistical analyses were conducted using SigmaStat with Spearman’s correlation; P < 0.05 was inferred as statistically significant.

Nitrate reduction to N₂O by azide yielded an average N₂O-SP value of 2.2 ± 0.9 ‰ (n = 35). Considering the small variation (±0.9 ‰) of the N₂O-SP values of the laboratory standard, the NO₃^– reduction to N₂O by azide is a safe technique to determine the isotopic composition of nitrate and nitrite.

The δ¹⁵N and δ¹⁸O of N₂O as end product or as intermediate (δ¹⁵N = –44.3 to –21.5 ‰; δ¹⁸O = –10.6 to 30.2 ‰), obtained for 5 mM of nitrate supply, presented a value range consistent with that found in previous studies on soil denitrification (δ¹⁵N = –41 to 20 ‰; δ¹⁸O = –55 to 55 ‰).^[16,17] However, considering these large ranges, the N₂O-SP value determination may bring a stronger proof that denitrification occurs only in these sedimentary studies.

The results concerning the N₂O-SP values for the different experiments at 5 mM of nitrate supply are shown in Fig. 2. Assuming that the acetylene block was complete and N₂O was not further reduced to N₂, the results during benthic denitrification indicated a low N₂O-SP value (6.3 ± 1.0 ‰; n = 16), which was, however, higher than those from pure cultures (–5 to 0 %, Fig. 2).^[7,8] The N₂O-SP value derived from Pseudomonas chlororaphis and P. aureofaciens of 0 ‰^[8] and of –5.1 ± 1.8 ‰ for P. denitrificans^[7] suggest that these values vary among denitrifying species. Our data show that the environmental benthic denitrifying community generates N₂O with a SP value different from those found in pure cultures.^[7,8] Most likely the enzyme implicated in the transformation of NO to N₂O (nitric oxide reductase, NOR) constrains the N₂O-SP values^[7] due to enzyme differences of the environmental benthic denitrifying community and may result in N₂O-SP values deviating from the pure cultures tested so far. Despite the difference compared to pure cultures, the environmental benthic denitrifying community showed N₂O-SP values in a narrow range, which confirms that the SP value of N₂O as end product is constant during this process (NO₃^– to N₂O).^[18]

Fig. 2.  Boxplots of N₂O site preference (N₂O-SP) values during benthic denitrification to N₂O (C₂H₂ block) and complete benthic denitrification for a 5 mM nitrate supply. The N₂O-SP values founded in the NO₃^– reduction to N₂O by azide (dotted area) and in the pure denitrifying culture studies (grey area).^[7,8] Lower limit (i; white squares; i = Q1–1.5(Q3–Q1)), high limit (s; white circles; s = Q3+1.5(Q3–Q1)), quartile 1 (Q1; blue diamonds; n/4 value), quartile 3 (Q3; purples crosses; 3*(n/4) value) and median (red bars).

In the river sediments, N₂O emitted during benthic denitrification showed higher average N₂O-SP values, together with a large variability, (17.4 ± 6.5 ‰, n = 5) compared with those of N₂O as an end product (6.3 ± 1.0 ‰, n = 16, Fig. 2). During the reduction step of N₂O to N₂, an enrichment of ¹⁵N in the central N-position (N^α) of the residual N₂O occurs^[19] and leads to an increase of the N₂O-SP values.^[10,11] This has been explained by the preferential consumption of ¹⁵N^β14N^αO relative to ¹⁴N^β15N^αO because the ¹⁵N^α–O bond is stronger than the ¹⁴N^α–O bond.^[6] Thus, the increase of the N₂O-SP values during the progress of denitrification will lead to a large variability of the N₂O-SP values.

During benthic denitrification, the SP of N₂O produced in the lagoon sediment, exhibiting an average of 6.8 ± 2.8 ‰ (n = 6), was lighter and less variable than that emitted from the river sediment (17.4 ± 6.5 ‰, n = 5, Fig. 2). This difference in N₂O-SP values from the two sediments during complete denitrification is most likely due to the presence and activity of distinct denitrifying communities, similar to the differences between N₂O-SP values obtained during NO₃^– reduction to N₂O between environmental and pure cultures.

During biological NO₃^– reduction to N₂O, the small variation in N₂O-SP values (varying between 4.2 and 7.8 ‰) corresponded to a wide range of nitrate reduction rates (100 to 300 nmol cm^–3 h^–1). Thus, this result confirms that the step of NO₃^– reduction to N₂O results in a specific low N₂O-SP value independent of the nitrate reduction rate. On the contrary, in the two sediments investigated, the N₂O-SP values exhibited a large variation during complete denitrification (Figs 2, 3).

Fig. 3.  N₂O site preference (N₂O-SP) values as a function of the NO₃^– reduction rates during denitrification in river (black diamonds, P = 2.00 × 10^–7, r = 0.835) and lagoon sediments (grey diamonds, P = 1.66 × 10^–3, r = 0.712).

Fig. 3 shows the N₂O-SP values for the two sediments, obtained for the entire range of nitrate supply, indicating a significant correlation between nitrate reduction rates and the N₂O-SP values (P = 0.0086; r = 0.48) during complete denitrification with N₂O appearing as intermediate. In the lagoon sediment, the NO₃^– reduction rates are lower than those found in the river sediment. This statement consolidates the idea of the presence and the activity of distinct species of denitrifying communities in the two sediments. Moreover, the variation of N₂O-SP values related to the variation of the NO₃^– reduction rate, during benthic denitrification in the two sediments, is in good agreement with previous studies that showed that the reduction of N₂O to N₂ leads to an increase of N₂O-SP values.^[10,11] In addition, the denitrification potential by the denitrifying bacteria is enhanced with the higher nitrate concentration and leads to a N₂O reduction to N₂. Thus, the last step of denitrification is crucial in the determination of the N₂O-SP values in field studies where production and reduction of N₂O occur simultaneously. On the contrary, without this last step of denitrification the values of N₂O-SP will be invariant and close to those found by Toyoda et al.,^[7] Sutka et al.^[8] and this study for N₂O production only.

This study showed that the SP of N₂O as an end product of the benthic denitrifying communities is independent of the nitrate reduction rate and constant during the NO₃^– reduction to N₂O step of denitrification. However, the SP value of N₂O tends to increase during complete denitrification, with N₂O as intermediate, and shows a strong positive correlation with the nitrate reduction rate. In fact, a high nitrate concentration enhances the denitrifier activity and increases the N₂O-SP values by ¹⁵N enrichment of residual N₂O during the last step in the denitrification process. Consequently, in field studies, during which production and reduction of N₂O occur simultaneously, the SP of N₂O can show a wide range of values, which does not depend on the composition of the denitrifying community only. This study underscores the importance of the last step of denitrification for the determination of the N₂O-SP values and that care should be taken using the SP of N₂O as the origin for denitrification in the case of field studies. In the identification of nitrous oxide sources, the determination of N₂O-SP values should be associated with other methods such as the use of the natural abundances of isotopes and the isotopic enrichment techniques or the identification of the denitrifying species involved.