Cyanate is a low abundance but actively cycled nitrogen compound in soil

Cyanate analysis

To test soil extractants for cyanate analysis, three soils (0–15 cm depth) differing in soil pH were collected in Austria, sieved to 2 mm and stored at 4 °C. An alkaline grassland soil was collected in the National Park Seewinkel (47° 46′ 32′′ N, 16° 46′ 20′′ E; 116 m a.s.l.), a neutral mixed forest soil in Lower Austria (N 48° 20′ 29′′ N, 16° 12′ 48′′ E; 171 m a.s.l.) and an acidic grassland soil at the Agricultural Research and Education Centre Raumberg-Gumpenstein (47° 29′ 45′′ N, 14° 5′ 53′′ E; 700 m a.s.l.). The recovery of cyanate was assessed by using cyanate-spiked (15 nM potassium cyanate added) and unspiked extraction solutions. We used water (Milli-Q, >18.2 MOhm, Millipore), 10 mM CaSO₄ and 1 M KCl as extractants. The three soils (n = 4) were extracted using a soil:extractant ratio of 1:10 (w:v), shaken for 10 min, and centrifuged (5 min at 14,000 × g). The supernatant was stored at −80 °C until analysis, as it has been shown that cyanate is stable at −80 °C over a period of 270 days²⁷. In our study, the storage time of samples ranged from a few days to a few months.

To explore soil cyanate concentrations across different soil and land management types, and along a climatic gradient, we collected 42 soils from Europe. Sites ranged from Southern France to Northern Scandinavia and included forests (F), pastures (P), and arable fields (A) (Fig. 2a). At each site five soil cores (5 cm diameter, 15 cm depth) were collected, after removal of litter and organic horizons. Soil samples were shipped to Vienna and aliquots of the five mineral soil samples of each site were mixed to one composite sample per site and the fresh soil was sieved to 2 mm. In addition to those 42 samples, we collected a rice paddy soil in Southern France (sample code A1; four replicates) and three grassland soils (G) in close vicinity of Vienna, Austria (G1 and G2 from saline grassland, three replicates; G3, one soil sample). Soil samples were stored at 4 °C and extracted within a few days. All sampling sites with their location, soil pH, and cyanate, ammonium, and nitrate concentrations are listed in Supplementary Data 1. For cyanate and ammonium analysis, soils (2 g fresh soil) were extracted with 15 mL 1 M KCl, shaken for 30 min and centrifuged (2 min at 10,000 × g). The supernatants were transferred to disposable 30 mL syringes and filtered through an attached filter holder (Swinnex, Millipore) containing a disc of glass microfiber filter (GF/C, Whatman). To reduce abiotic decay of cyanate to ammonium during extraction, the extraction was performed at 4 °C with the extracting solution (1 M KCl) cooled to 4 °C prior to extraction. Soil extracts were stored at −80 °C until analysis.

To compare cyanate availability across different environments, we analyzed cyanate in salt marsh sediments and activated sludge from municipal wastewater treatment plants, and, additionally, we collected published data on cyanate concentrations in the ocean. We collected sediment samples (0-10 cm, n = 4) from a high and low salt marsh dominated by Spartina alterniflora Loisel in New Hampshire, USA (43° 2′ 26′′ N, 70° 55′ 36′′ W), and from a S. alterniflora and a S. patens (Aiton) Muhl salt marsh in Maine, USA (43° 6′ 31′′ N, 70° 39′ 56′′ W). We chose these types of salt marsh because they have been shown to accumulate cyanide⁴⁴, which potentially could be oxidized to cyanate. Sediment samples were stored at 4 °C and extracted within a few days after collection using 2 M KCl at a sediment:extractant ratio of 1:10 (w:v) for 30 min at room temperature. The supernatants were filtered through glass microfibre filters as described above for soil samples. Pore water was extracted with Rhizon samplers (Rhizon CSS, 3 cm long, 2.5 mm diameter, Rhizosphere Research Products, Netherlands) with a filter pore size of 0.15 µm. Triplicate samples of activated sludge were collected from four municipal Austrian wastewater treatment plants (WWTPs), i.e., from Alland (48° 2′ 30′′ N, 16° 6′ 1′′ E), Bruck an der Leitha (48° 2’ 4” N, 16° 49′ 7′′ E), Wolkersdorf (48° 21′ 31′′ N, 16° 33′ 31′′ E) and Klosterneuburg (48° 17′ 39′′ N, 16° 20′ 30′′ E). Samples from the discharge were also collected from the first three listed WWTPs. Samples were cooled on gel ice packs during the transport to Vienna. Upon arrival in Vienna, samples were transferred to disposable 30 mL syringes and filtered through an attached filter holder (Swinnex, Millipore) containing a disc of glass microfiber filter (GF/C, Whatman). All samples were immediately stored at −80 °C until analysis.

Cyanate concentrations were determined using high performance liquid chromatography (HPLC) with fluorescence detection, after conversion to 2,4(1H,3H)-quinazolinedione²⁷. Briefly, a 230 µL aliquot of the sample was transferred to a 1.5 mL amber glass vial, 95 µL of 30 mM 2-aminobenzoic acid (prepared in 50 mM sodium acetate buffer, pH = 4.8) were added, and samples were incubated at 37 °C for 30 min. The reaction was stopped by the addition of 325 µL of 12 M HCl. Standards (KOCN) were prepared fresh daily and derivatized with samples in the same matrix. Derivatized samples were frozen at −20 °C until analysis. Just before analysis samples were neutralized with 10 M NaOH. The average detection limit was 1.2 nM (±0.2 SE). Ammonium concentrations were quantified by the Berthelot colorimetric reaction. As direct comparison of cyanate concentrations was not possible across the different environments and matrices, we normalized cyanate concentrations relative to ammonium concentrations, by calculating ammonium-to-cyanate ratios. Data on marine cyanate and ammonium concentrations were taken from Widner et al.²². For marine samples where cyanate was detectable but ammonium was below detection limit, we used the reported limit of detection of 40 nM for ammonium. The presented soil and sediment data are biased toward higher cyanate availabilities (i.e., low NH₄⁺/NCO⁻ ratios), due to the exclusion of samples where cyanate was possibly present but was below detection limit. Soil pH was measured in 1:5 (w:v) suspensions of fresh soil in 0.01 M CaCl₂ and water.

Dynamics of cyanate consumption in soil using stable isotope tracer

For the determination of half-life of cyanate, we used two soils: a grassland soil (G3) and a rice paddy soil (A1). Both soils had a pH of 7.4 (determined in 0.01 M CaCl₂). The grassland soil had a soil organic C concentration of 37 mg g⁻¹, soil N concentration of 1.92 mg g⁻¹, molar C:N ratio of 22.4, ammonium concentration of 5.60 nmol g⁻¹ d.w., nitrate concentration of 1.03 µmol g⁻¹ d.w., and an electrical conductivity of 82.0 mS m⁻¹. The rice paddy soil had a soil organic C concentration of 10 mg g⁻¹, soil N concentration of 0.98 mg g⁻¹, molar C:N ratio of 11.9, ammonium concentration of 2.47 nmol g⁻¹ d.w., nitrate concentration of 0.91 µmol g⁻¹ d.w., and an electrical conductivity of 21.7 mS m⁻¹. To equilibrate soil samples after storage at 4 °C, soil water content was adjusted to 55% water holding capacity (WHC; gravimetric water content of water saturated soil) and soils incubated at 20 °C for one week prior to the start of the experiment. To correct for abiotic reactions of cyanate, a duplicate set of soil samples was prepared and one set of them was sterilized by autoclaving prior to label addition while the other set was left under ambient conditions. Soil samples were autoclaved three times at 121 °C for 30 min with 48 h-incubations at 20 °C between autoclaving cycles to allow spores to germinate prior to the next autoclaving cycle and to inactivate enzymes⁴⁵.

Preliminary experiments indicated rapid consumption of added cyanate. Thus, to avoid fast depletion of the added cyanate pool, we added 5 nmol ¹³C¹⁵N-KOCN g⁻¹ f.w. (¹³C: 99 atom%; ¹⁵N: 98 atom%), which equals to approximately 250-fold the in-situ cyanate concentration. With the tracer addition the soil water content was adjusted to 70% WHC. After tracer addition, non-sterile and sterile soil samples were incubated at 20 °C for a period of 0, 10, 20, 30, 45, 60 and 90 min (n = 3) before stopping the incubation by extraction. Soil extractions were performed with 1 M KCl as described above for the 46 soil samples. Soil extracts were stored at −80 °C until analysis.

As no method for compound-specific isotope analysis of cyanate existed, we developed a method to measure isotopically labeled and unlabeled forms of cyanate in soil extracts using hydrophilic interaction chromatography coupled to high-resolution electrospray ionization mass spectrometry (HILIC-LC-MS). For this analysis, cyanate was converted to 2,4(1H,3H)-quinazolinedione as described above for the RP-HPLC method but with some modifications. Aliquots of 280 µL of each sample were transferred to 2 mL plastic reaction vials, and 20 µL of internal standard solution (4 µM ¹³C-KOCN, 98 atom%) were added. To start the reaction, 120 µL of 30 mM 2-aminobenzoic acid (prepared in ultrapure water) were added, and samples were incubated at 37 °C for 30 min. The reaction was stopped by the addition of 420 µL 12 M HCl. To remove HCl and bring the target compound into an organic solvent that can be easily evaporated, we performed liquid-liquid extractions using a mixture of ethyl acetate/toluene (85/15 (v/v)). Each sample was extracted 3 times with 1 mL organic solvent mixture. For extraction, samples were thoroughly mixed by vortexing and the tubes were briefly spun down to separate the two phases. The organic phases of each extraction were combined in a 10 mL amber glass vial and dried under a stream of N₂. Before analysis, samples were redissolved in 200 µL mobile phase. Samples were analyzed on a UPLC Ultimate 3000 system (Thermo Fisher Scientific, Bremen, Germany) coupled to an Orbitrap Exactive MS (Thermo Fisher Scientific). 2,4(1H,3H)-quinazolinedione was separated using an Accucore HILIC column (150 mm × 2.1 mm, 2.6 µm particle size) with a preparative guard column (10 mm × 2.1 mm, 3 µm particle size; Thermo Fisher Scientific). We used isocratic elution with 90/5/5 (v/v/v) acetonitrile/methanol/ammonium acetate, with a final concentration of ammonium acetate of 2 mM (pH = 8). The sample injection volume was 7 µL, and the flow rate 0.2 mL min⁻¹. The Orbitrap system was used in negative ion mode and in full scan mode at a resolution of 50,000. The source conditions were: spray voltage 4 kV, capillary temperature 275 °C, sheath gas 45 units, and AUX gas 18 units. The instrument was calibrated in negative ion mode before sample acquisition using Pierce LTQ ESI Negative Ion Calibration Solution (Thermo Fisher Scientific). To improve the accuracy of absolute quantification, external calibration (concentration standards and ¹³C¹⁵N-KOCN standards) was paired with an internal calibrant (¹³C-potassium cyanate) to correct for deviations in liquid-liquid extraction efficiency, ionization efficiency and ion suppression. ¹³C-KOCN (98 atom%) and ¹³C¹⁵N-KOCN (¹³C: 99 atom%; ¹⁵N: 98 atom%) were purchased from ICON Isotopes. The mass-to-charge (m/z) ratio of unlabeled, ¹³C- and ¹³C¹⁵N-labeled cyanate was 161.0357, 162.0391, and 163.0361, respectively, and the retention time was 2.2 min. The limit of detection was 9.7 nM.

To obtain biotic cyanate consumption rates, the non-sterile samples were corrected for abiotic decomposition of cyanate derived from the sterile (autoclaved) samples. Dynamics of cyanate consumption over time for the corrected non-sterile soils were then described by fitting a first order exponential decay curve:

Where C(t) is the remaining ¹³C¹⁵N-cyanate concentration at time t, C₀ is the initial concentration of ¹³C¹⁵N-cyanate and k is the exponential coefficient for ¹³C¹⁵N-cyanate consumption. The half-life (t_1/2) of the ¹³C¹⁵N-cyanate pool was calculated as:

Abiotic reactions of cyanate and isocyanic acid

Urea (CO(NH₂)₂) exists in chemical equilibrium with ammonium cyanate (NH₄CNO) in aqueous solution:

The rate constant for the decomposition of urea (k_1a) and for the conversion of ammonium cyanate into urea (k_1b) were taken from Hagel et al.⁴⁶, and temperature dependence was calculated by using the Arrhenius equation:

where T is temperature in Kelvin.

Cyanate is the anionic form of isocyanic acid. The latter exists as two isomers in aqueous solution, where isocyanic acid is the dominant species. Thus, the acid will be referred to as isocyanic acid. The decomposition of isocyanic acid and cyanate in aqueous solution was found to take place according to three simultaneous reactions:

Eq. (6) is for the hydronium ion catalyzed hydrolysis of isocyanic acid (rate constant k_2a; dominant reaction at low pH), Eq. (7) is for the direct hydrolysis of isocyanic acid (k_2b), and Eq. (8) is for the direct hydrolysis of cyanate (k_2c; dominant reaction at high pH). The rate constants are as follows⁴⁶:

Isocyanic acid reacts with amino groups of proteins, in a process called carbamoylation¹⁹:

We used glycine as an example for an amino acid, with the following rate constant⁴⁷:

Urea-derived cyanate formation in a fertilized agricultural soil

For studying the formation and consumption of cyanate after urea addition, we used a rice paddy soil (A1; the same soil as used in the stable isotope tracer experiment), which was cultivated with rice once every second year with a urea application rate of 180 kg N ha⁻¹ y⁻¹. Treatment of the soil samples was the same as for the stable isotope tracer experiment. Briefly, soil water content was adjusted to 55% water holding capacity (WHC) and soil samples (4 g of fresh soil in a 5 mL centrifugation tube) were incubated at 20 °C for one week prior to the start of the experiment. With the addition of the urea solution, the soil water content was adjusted to 70% WHC. We added 140 µg urea g⁻¹ soil d.w., which corresponds to ~180 kg N ha⁻¹. Soil samples were incubated at 20 °C for a period of 0, 6, 12, 24, and 30 h (n = 4). At each sampling, we collected the soil solution. For this a hole was pierced in the bottom of the 5 mL centrifugation tube containing the soil sample. This tube was then placed into another, intact, 15 mL centrifugation tube and this assembly was then centrifuged at 12,000 × g for 20 min at 4 °C to collect the soil solution. Soil solution samples were stored at −80 °C until analysis. For comparative analysis, we converted rates based on nmol L⁻¹ soil solution to rates based on a dry soil mass basis. For the conversion, we recorded the volume of the soil solution collected and determined the water content of the soil samples after centrifugation.

Cyanate concentrations in soil solution were determined as described above using HPLC. Urea was quantified by the diacetyl monoxime colorimetric method, ammonium by the Berthelot colorimetric reaction and ammonium, and nitrite and nitrate by the Griess colorimetric procedure. For cyanate analysis, aliquots of two replicates were pooled because of insufficient sample volume.

We used the well-established rate constants for the equilibrium reaction of urea in aqueous solution and decomposition of cyanate to ammonia/ammonium and carbon dioxide/bicarbonate, to model gross cyanate production and consumption after urea amendment from observed changes in urea, ammonium and cyanate concentrations over time. Cyanate accumulation was calculated as cyanate formation from urea (rate constant k_1a, Eq. (4)) minus the conversion of ammonium cyanate into urea (rate constant k_1b, Eq. (5)), and minus abiotic cyanate hydrolysis to ammonium and carbon dioxide (rate constants k_2a, k_2b, k_2c, Eqs. (9)–(11)). It has been found that only the ionic species (i.e., NCO⁻ and NH₄⁺) are involved in the reaction of ammonium cyanate to urea. The difference between cyanate accumulation and the net change in cyanate concentration over time gives then cyanate consumption, as follows:

where [NCO^–] represents the concentration of cyanate and isocyanic acid, [NH₄⁺] is the sum of ammonium and ammonia, K_HNCO and K_NH3 is the acid dissociation constant of isocyanic acid and ammonia, respectively, and [H₃O⁺] is the hydronium ion concentration. Urea concentration over time was described by a first order reaction (Eq. (15); unit of rate constant is min⁻¹), and ammonium and cyanate concentrations were fitted with a third and fourth degree polynomial function, respectively (Eqs. (16) and (17), respectively), as follows:

where t is time in min and concentrations are mol/L soil solution.

The input parameters were 7.4 for pH (pH of solution: 7.4 ± 0.1 SD) and 20 °C for temperature. As rate constant k_1b is dependent on the ionic strength, we corrected the rate constant (given at I = 0.25⁴⁶) using the Extended Debye–Hückel expression:

Where f is the activity coefficient, A and B are constants that vary with temperature (at 20 °C, A = 0.5044 and B = 3.28 × 10⁸), z is the integer charge of the ion, and a is the effective diameter of the ion (a = 5 Å⁴⁶). We used an ionic strength I = 0.01, which is within the range observed for soils.

Statistical analysis

Statistical significance of the difference between extractants within each soil type was analyzed by one-way ANOVA followed by Tukey HSD post-hoc test. Levene’s Test was used to test equality of variances and QQ plot and Kolmogorov Smirnov Test were used to assess normal distribution of residuals. For each extractant, statistical significance of the difference between added and recovered cyanate was tested using t test on raw data, where F-test was used for testing equality of variances. To analyze the effect of type of environment on relative cyanate availability (i.e., NH₄⁺/NCO⁻), we used the Kruskal-Wallis test (assumption for parametric procedure were not met) followed by a non-parametric multiple comparison test (Dunn’s test). For solving differential equations in the model, we used the “deSolve” package in R⁴⁸.

Source: Ecology - nature.com