Hydrazine Synthase From Anammox Is Inhibited by Linear and Aromatic Alkynes

ABSTRACT

N₂O emissions by nitrifiers are often estimated using selective inhibitors, such as 1‐alkynes. However, the effects of these inhibitors on anaerobic ammonium‐oxidising (anammox) bacteria are largely unknown. In this study, we assessed the inhibitory effect of linear and aromatic alkynes on anammox activity and identified their target enzyme. ‘Candidatus Kuenenia stuttgartiensis’ biomass and constructed wetland soil samples were incubated with 10 μM of C₂–C₈ linear alkynes or phenylacetylene for 10 days. Anammox activity was determined using the isotopic tracer ¹⁵N‐nitrite and ²⁹N₂ production. Anammox activity was suppressed by C₂–C₅ alkynes, whereas the larger or aromatic alkynes caused no inhibition. However, hydrazine oxidation activity was not affected, indicating that C₂–C₅ alkynes inactivated enzymes upstream of hydrazine dehydrogenase. In incubations using an NO donor and ¹⁵N‐ammonium, ²⁹N₂ production stopped, suggesting that hydrazine synthase was the target of these alkynes. A comparable trend was observed in the wetland samples, but with a less pronounced reduction in ²⁹N₂ production. Since alkynes >C₅ did not affect anammox, these findings demonstrate the suitability of using 1‐octyne as a selective inhibitor to quantify N₂O contributions from ammonia‐oxidising bacteria (AOB) versus ammonia‐oxidising archaea (AOA) in oxic/anoxic interface environments.

Linear and aromatic alkynes inhibit anammox in culture and in wetlands. The inhibition by alkynes is specific to the hydrazine synthase, a key enzyme in the anammox pathway.

1. Introduction

Bacteria capable of anaerobic ammonium oxidation (anammox) were first identified in a sequencing batch reactor (Strous et al. 1999), two decades after the existence of this metabolism was proposed by Broda (1977). All known anammox bacteria are members of the candidate order ‘Brocadiales’ within the Planctomycetota and comprise five genera: Kuenenia, Brocadia, Jettenia, Anammoxoglobus, and Scalindua (Wang et al. 2012; Kumwimba et al. 2020; Talan et al. 2021; Zhu et al. 2021). To date, no anammox pure cultures are available (Kartal et al. 2013; Versantvoort et al. 2025). The occurrence of anammox bacteria is widespread, with representatives identified in engineered environments such as activated sludge and wastewater treatment (Jetten et al. 2005; Third et al. 2005; Chamchoi and Nitisoravut 2007; Ma et al. 2024) as well as in natural environments including marshes, hydrothermal vents, and brackish and marine ecosystems (Nicholls and Trimmer 2009; Humbert et al. 2010; Teixeira et al. 2012; Russ et al. 2013). It is estimated that anammox accounts for up to 80% of dinitrogen gas (N₂) production in marine ecosystems (Jensen et al. 2007; Tomaszewski et al. 2017) and for half of the global atmospheric nitrogen present in the form of N₂, highlighting the importance of the anammox pathway for the global nitrogen cycle (Kartal et al. 2008; Li et al. 2015).

Whilst anammox plays a key role in the global nitrogen cycle in natural habitats, it has also been widely applied in engineered settings to remediate high concentrations of ammonium typical of landfill leachate. Landfill leachate is formed as water percolates through solid waste over time and is characterised by high ammoniacal concentrations (up to 2.2 g L⁻¹) as well as dissolved and suspended material derived from waste decomposition (Renou et al. 2008; Mak et al. 2018; Parvin and Tareq 2021).

The anammox reaction is a three‐step process that converts ammonium (NH₄ ⁺) and nitrite (NO₂ ⁻) into N₂, without producing nitrous oxide (N₂O) (Kuenen 2008). The first step, the conversion of NO₂ ⁻ to nitric oxide (NO), is mediated by either a NirS or NirK‐type or an octaheme nitrite reductase (Nir) (Vermeir et al. 2025). The second step involves the production of hydrazine (N₂H₄) from NO and NH₄ ⁺, catalysed by hydrazine synthase (Hzs). The final step, the oxidation of N₂H₄ to N₂, is facilitated by the hydrazine dehydrogenase (Hdh; Figure S1). This is fundamentally different from the enzymes responsible for aerobic ammonia oxidation. In ammonia‐oxidising bacteria (AOB) and archaea (AOA), the first step of nitrification is catalysed by the enzyme ammonia monooxygenase (AMO), which converts ammonia (NH₃) to hydroxylamine (NH₂OH) (Radniecki and Lauchnor 2011; Vajrala et al. 2013). Subsequently, hydroxylamine dehydrogenase (HAO) generates NO₂ ⁻ from NH₂OH in AOB. In contrast, AOA lack a canonical HAO, and oxidation of NH₂OH to NO₂ ⁻ in these organisms likely involves alternative enzymes (Walker et al. 2010; Vajrala et al. 2013).

Linear C₂–C₈ alkynes and phenylacetylene have different effects on specific groups of aerobic ammonia‐oxidising microorganisms (Taylor et al. 2015; Wright et al. 2020). The difference in the effects of the inhibitors on the activity of ammonia monooxygenases (AMO) from ammonia‐oxidising archaea (AOA) and bacteria (AOB) has led to the use of 1‐octyne as a specific inhibitor to discriminate between the activities of AOA and AOB (Taylor et al. 2015; Wright et al. 2020). Studies on the effect of linear alkynes on complete ammonia‐oxidising (comammox) bacteria have produced contrasting results. In both arable and pasture soils, the addition of 1‐octyne caused no significant changes in the abundance of clade A comammox Nitrospira over a 28‐day incubation (Li et al. 2019; Tan, Yin, et al. 2022). Conversely, the amendment of 1‐octyne in acidic soils, such as ultisol (Uwah and Iwo 2011), attenuated the growth of comammox clade A across multiple fertiliser treatments, suggesting that C₈ alkyne has the potential to inhibit these ammonia‐oxidising microorganisms in addition to AOB (Lin et al. 2023). Contrastingly, the effects of alkynes on anammox are largely unknown. Anammox is inhibited by the addition of 15–30 μM acetylene (Jensen et al. 2007; Kartal, Geerts, and Jetten 2011), and although the mechanism of inhibition is not understood, it has been suggested to potentially inhibit multiple enzymes in anammox, including hydrazine synthase (Oshiki et al. 2022).

In this study, constructed wetlands receiving landfill leachate were used to evaluate the impacts of alkynes on anammox and other nitrogen cycling processes in environmental samples (Figure S2). The aims of this study were to investigate the influence of linear and aromatic alkynes on anammox activity in culture and in the environment, to determine their suitability as selective inhibitors, and to identify the target of inhibition in the anammox pathway. A 10 μM octyne concentration can reduce the activity of AOB‐driven nitrification without inhibiting AOA activity (Taylor et al. 2015; Wright et al. 2020), but the effect of octyne on anammox activity has not been previously assessed, leaving uncertainties when evaluating nitrogen fluxes. This is pertinent to minimising emissions of N₂O, a potent greenhouse gas with an approximate 270 global warming potential of CO₂ over 100 years (NOAA 2025; Li et al. 2025). Therefore, attenuating the activity of the nitrifiers responsible for N₂O emissions without inhibiting anammox might provide a solution to reduce the greenhouse gas emissions from a range of environments, including habitats associated with landfill remediation. Employing octyne as a tool to assess nitrification activity in soils where anammox coexist with other ammonia oxidisers holds significant potential for application in diverse environmental samples, including those from wastewater treatment plants (WWTP) and marine ecosystems.

2. Materials and Methods

2.1. Biomass Incubations and C₂ –C₈ and Phenylacetylene Preparation

2.1.1. Cultivation of ‘Ca. Kuenenia stuttgartiensis’

A highly enriched culture of ‘Ca. Kuenenia stuttgartiensis’, growing in planktonic cells, originating from Radboud University was cultured in a 2 L bioreactor (Electrolab biotech, Tewkesbury, UK) operated in batch mode, at the University of East Anglia (Kartal, Maalcke, et al. 2011). The bioreactor was filled with 1.5 L of mineral medium containing the following constituents: 0.025 g L⁻¹ KH₂PO₄, 0.5 mL L⁻¹ 1.2 M HCl, 0.15 g L⁻¹ CaCl₂·2H₂O, 0.1 g L⁻¹ MgSO₄·7H₂O, 6.25 mg L⁻¹ FeSO₄·7H₂O, 0.5 μM Na₂EDTA. All compounds were added, and the medium was sterilised by autoclaving at 126°C for 15 min. Subsequently, the reactor was inoculated with 500 mL of ‘Ca. Kuenenia stuttgartiensis’ biomass, containing 8.8 × 10⁷ copies of hdh per mL. The bioreactor was operated at 28°C and a pH 7.1–7.4 and stirred at 50–60 rpm. An oxygen saturation of 0% was maintained by supplying a gas mix of 2% CO₂: 98% N₂ (BOC, Woking, UK) at a flow rate of 25 mL min⁻¹, and monitored using an oxygen electrode (Electrolab biotech, Tewkesbury, UK). The bioreactor was routinely fed with 1 M stocks of NH₄Cl and KNO₂ to maintain concentrations between 2 and 4 mM. Ammonium and nitrite concentrations were determined at least three times per week as described below.

When the biomass was removed for incubation studies, it was replaced with an equal volume of sterilised media. The media was flushed with oxygen‐free nitrogen (OFN) for 10 min in 50 mL Falcon tubes sealed with Suba seals and then injected into the bioreactor through a 0.2 μM filter through the septum of the bioreactor vessel.

2.1.2. Preparation of Biomass for Inhibitor Incubations

Biomass was removed from the bioreactor via the sampling port, collected in a 100–500 mL glass bottle and transferred into 50 mL Falcon tubes sealed using Suba seals under an anoxic atmosphere. The Falcon tubes were then flushed with oxygen‐free N₂ gas for 20 min at a cylinder exit pressure of 5 psi, and transferred to an anaerobic glove box, under an N₂ atmosphere (MBraun, Garching, Germany). Here, 8.8–9.0 mL biomass was aliquoted into 12 mL exetainer tubes (Labco, Ceredigion, UK) and Na¹⁵NO₂ and ¹⁴NH₄Cl, or ¹⁵N‐N₂H₄ solutions were added from 20 mM, 50 mM or 50 mM stock solutions, respectively, to a final concentration of 2 mM (Na¹⁵NO₂ and ¹⁴NH₄Cl) or 1.5 mM (¹⁵N‐N₂H₄). Incubations had an overall volume of 10 mL, with a 2 mL gaseous headspace.

All alkynes were added to a final (aqueous phase) concentration of 10 μM. C₂–C₈ linear 1‐alkynes were added by injecting vapour from 1% v/v stocks, freshly prepared for each experiment, using a gas‐tight syringe. Volumes were calculated using Henry's Law Coefficients (Sander 2015). Phenylacetylene (10 mM) was dissolved in 100% dimethyl sulfoxide (DMSO) and made fresh for each incubation. DMSO‐only controls were included to test the effect of DMSO on anammox activity.

2.1.3. Preparation of the NO‐Donor DEA‐NONOate

Diethylammonium (Z)‐1‐(N,N‐diethylamino)diazen‐1‐ium‐1,2‐diolate (DEA‐NONOate) is an NO donor which dissociates in a pH‐dependent process with a half‐life of 16 min at 22°C–25°C and pH 7.4 (Cambridge Bioscience Ltd., Cambridge, UK). A stock solution of 48.5 mM DEA‐NONOate was prepared in 100% DMSO. DEA‐NONOate was added to biomass incubations of ‘Ca. Kuenenia stuttgartiensis’, pH 7.3–7.4, to a final concentration of 70 μM. The six‐hour incubations were performed in duplicate, with similar results. Exetainers were amended with the gradual addition of DEA‐NONOate over 3 h. For every mole of DEA‐NONOate added, theoretically, 1.5 mol of NO are expected to be liberated. Assuming a 100% conversion of the DEA NONOate, 1.05 μmol of NO was available to anammox. 1 mM ¹⁵NH₄Cl was added to the existing background of 500 μM of ¹⁴NH₄Cl at T0, achieving a 67% atom percentage of ¹⁵NH₄ ⁺.

2.1.4. Construction and Operation of Wetlands

Our study site, Mayton Wood, Norfolk, UK is a decommissioned landfill site which ceased operation in 2005. Current leachate management strategies employed by the Norfolk County Council involve disposing of leachate via haulage to a wastewater treatment plant, a common yet unsustainable practise amongst district councils across the UK. The passive remediation of leachate through constructed wetlands would allow on‐site treatment of leachate, removing the need for haulage to wastewater treatment facilities (Norfolk County Council 2025). Vegetated wetlands encourage the formation of oxic‐anoxic interfaces surrounding the roots, leading to the enrichment of a diverse microbial community comprising both obligate aerobes and anaerobes that exist in close proximity (Bodelier et al. 2006). The rationale behind the proposed treatment relies on the assumption that in wetlands planted with grasses, such as Phragmites australis , the activity within the rhizosphere can provide a source of nitrate which can be reduced by denitrifiers, as observed with Glyceria maxima (Bodelier et al. 1998). The ammonium‐rich leachate dosed onto wetlands would provide the source of ammonium to facilitate the anammox reaction. Raw leachate was diluted with water at a 7:5 ratio (leachate: water) before being applied to the wetlands planted at a density of 12 root plugs per 1 m² (Figure S2). The diluted leachate was pumped from the leachate tank and applied to the wetlands via piping located on the surface soil (Figure S2).

2.2. Sampling Constructed Wetlands and Slurry Incubations

2.2.1. Preparation of Soil Slurries

Soil was extracted from the wetlands at the oxic‐anoxic soil interface layer, approximately 40 cm below the soil surface (Figure S2) using a spade. The soil was stored in 50 mL Falcon tubes, closed with Suba seals, at room temperature for up to 1.5 h. Approximately 200 g of soil was sampled from the wetland for each soil slurry incubation performed.

Soils slurries were prepared in a 1:1 ratio with the mineral medium amended with Na¹⁵NO₂ and ¹⁴NH₄Cl, meaning for every 1 g of soil, 1 mL of the 2 mM Na¹⁵NO₂ and ¹⁴NH₄Cl medium was added. Soil was transferred into the anaerobic glove box, weighed, and added to a 12 mL exetainer. A total of 7 g of soil and 7 mL of medium were added. The stock solutions were flushed with oxygen‐free N₂ gas for 20 min at a cylinder exit pressure of 5 psi and added to the soil in the 12 mL exetainers under an anoxic atmosphere in an anaerobic glove box to prevent introduction of oxygen into the incubations. The incubations had a headspace of 2 mL.

2.3. Nutrient Assays for Biomass and Soil Slurry Incubations

Ammonium concentration was assayed using the indophenol method outlined by Kandeler and Gerber (1988), against 10–250 μM NH₄Cl standards, in duplicate. Nitrite was monitored using Griess reagent in a 96‐well plate format as described in Lehtovirta‐Morley et al. 2014, against 0.78–50 μM NaNO₂ standards, in duplicate. Biomass samples were diluted in deionised water (dH₂O) and measured against standards also prepared in dH₂O. For soil incubations, samples and standards were diluted in 1 M KCl to extract ammonium in a 1:3 mass‐to‐volume ratio of sample (g) to KCl added (mL). Typically, 2 g of soil and 6 mL 1 M KCl were added to a 50 mL Falcon tube and shaken at a speed of 2,800 rpm for 30 min using a Vortex Genie 2 Mixer. To remove soil particles, samples were centrifuged at 15,000g for 1 min, and the supernatant was removed for nutrient assays. All samples were measured in triplicate.

2.4. Molecular Techniques

2.4.1. DNA Extraction and Quantification

DNA was extracted from both environmental soil samples and bioreactor biomass to examine the microbial community structure. The 50 mL of ‘Ca. Kuenenia stuttgartiensis’ biomass was centrifuged in 50 mL Falcon tubes at 5,000g for 5 min at 4°C. The supernatant was removed, and the pellet was resuspended in 500 μL of PBS. For soil samples, 0.25 g was weighed out into a PowerBead Pro Tube (Qiagen, UK). DNA was extracted from both soil and biomass using the DNeasy PowerSoil Pro Kit (Qiagen, UK), following the manufacturer's instructions.

DNA concentrations were determined using a NanoDrop spectrophotometer (Thermo Fisher, Waltham, Massachusetts, US) and Qubit fluorometer (Invitrogen, Waltham, Massachusetts, US) according to the manufacturers' instructions. The purity of the DNA was estimated by the ratio of absorbances at 260 nm and 280 nm wavelengths as well as electrophoresis on a 1% (w/v) agarose gel, run at 85 V for 40 min.

2.4.2. Quantitative Polymerase Chain Reactions (qPCR)

qPCR assays specific to the hdh gene were set up in 96‐well plates using SensiFAST SYBR Hi‐ROX Mix (2×). Reactions contained 100 nmol forward and reverse primers, hzo_1F (5′‐AAGACNTGYCAYTGGGGWAAA‐3′) and hzo_1R (5′‐GACATACCCATACTKGTRTANACNGT‐3′) respectively (Schmid et al. 2008), SensiFAST mastermix and dH₂O in a total reaction volume of 20 μL. For each qPCR assay, ~5 ng of sample DNA was used per reaction. DNA was extracted from ‘Ca. Kuenenia stuttgartiensis' and amplified using hdhCntrl_1F (5′‐CAGGCAACAACCATCAGAA–3′) and hdhCntrl_1R (5′‐ACCGAATGAACCGTCTGAGT–3′). The PCR product was purified using NucleoSpin Gel and PCR Clean‐up Kit (Macherey‐Nagel, Leicestershire, UK) to remove dNTPs and contaminants, and quantified using QuBit (Thermo Fisher Scientific, Waltham, Massachusetts, US). A 10‐fold dilution series was prepared using nuclease‐free water, and a standard range of 10²–10⁷ gene copies was used to generate a standard curve (Stopnišek et al. 2010). Standard curves were constructed by plotting the log (template quantity) against the C_T value. The average qPCR primer efficiency for the hdh primers was 104% (±2.5) and the assay averaged an R ² = 0.997.

2.4.3. Sequencing and Bioinformatic Processing

16S rRNA gene amplicon sequencing was performed for both bioreactor biomass and environmental samples by Novogene, (Cambridge, UK). Sequencing was conducted using the V4 region using the primers 515F, (5′‐GTGCCAGCMGCCGCGGTAA‐3′), and 806R, (5′‐GGACTACHVGGGTWTCTAAT‐3′). Amplicon sequence variant (ASVs) were identified by Novogene. In short, the qiime2 pipeline was applied with DADA2 denoising (including quality filtering, error model learning, denoising and chimaera removal) to obtain the ASVs (Callahan et al. 2016). Demultiplexing was performed and data QC was executed using FASTQC (Andrews 2010), and reads with low quality and chimaeras were removed. The SILVA v138.1 database was used for taxonomic classification (Quast et al. 2012). Data were processed to produce relative abundance plots using the following packages in R Studio v2024.09.0 + 375: gplots (Warnes et al. 2005), wesanderson (Ram et al. 2023), ggtext (Wilke and Wiernik 2020) and ggplot2 (Wickham 2011). Rarefaction analysis was performed to ensure that the diversity was fully sampled (Figure S3).

2.5. Gas Sampling and Isotope Analysis

2.5.1. Measuring N₂ in Samples From Biomass and Slurry Incubations

The anammox process was tracked using ¹⁵N‐labelled substrates and ²⁹N₂ production measured on a GC–MS. ¹⁵NO₂ ⁻ was added at an atom percentage of 98%. ¹⁵N labelled nitrite was replenished to the incubations upon depletion (Figure S4). The 10 μL of gas removed from the headspace of the exetainer was injected into a GC–MS (GCMS‐QP2010S, Shimadzu, Kyoto, Japan), fitted with a PoraBOND Q 25 m length column (Agilent, Santa Clara, US), to measure ²⁹N₂ to determine the activity of ‘Ca. Kuenenia stuttgartiensis’. The GC–MS was operated at an injection temperature of 200°C and a column oven temperature of 35°C. The retention time for N₂ was 2.83 min. The concentration of ³⁰N₂ in the 10 μL gas sample was used to measure the activity of the hydrazine dehydrogenase when incubations were amended with ¹⁵N‐N₂H₄. Gas measurements were taken in triplicate and the concentration of ²⁹N₂ was quantified against known atmospheric concentrations using the ideal gas law.

2.6. Statistical Analysis

All data were presented as the mean of three biological replicates ± standard error, unless otherwise stated in the figure legend. Prior to statistical analysis, Levene's test was performed to assess homogeneity of variances, whilst the Shapiro–Wilk test was used to evaluate normality of the data distributions. Analysis of variance (ANOVA) was used to compare the differences between treatments for incubations supplying anammox biomass with ¹⁵N‐N₂H₄ and ¹⁵NO₂ ⁻ to the soil slurry incubations. Statistical significance was determined at a p‐value of ≤ 0.05. ANOVA was computed using R Studio v2024.12.1 (R Core Team 2022). Tukey's post hoc analysis was used to determine differences amongst treatments. Kruskal‐Wallis was used to compare differences between treatments for incubations where anammox biomass was supplied with ¹⁵NO₂ ⁻ and in the NO donor incubations. Statistical significance was determined at a p‐value of ≤ 0.05. Kruskal‐Wallis was computed using R Studio v2024.12.1 (R Core Team 2022). Conover's post hoc analysis was used to determine differences amongst treatments.

3. Results

3.1. The Impact of C₂ –C₈ Linear Alkynes and Phenylacetylene on Anammox Activity

‘Ca. Kuenenia stuttgartiensis’ biomass from the bioreactor was incubated with ¹⁵NO₂ ⁻ and ¹⁴NH₄ ⁺ and amended with 10 μM of C₂–C₈ linear alkynes or phenylacetylene. Anammox activity was monitored by measuring N₂ production over 10 days. Incubations were supplemented with ¹⁴NH₄ ⁺ and ¹⁵NO₂ ⁻ whenever these substrates were consumed (Figure S4).

Within 48 h, the different chain‐length alkynes had strong and distinct effects on ²⁹N₂ production (Figure 1). After 10 days of incubation, the addition of C₂–C₅ alkynes had a statistically significant inhibitory effect on the activity of anammox, with ²⁹N₂ production dropping to between 1.1 (propyne) and 4.2 (acetylene) μmol ²⁹N₂ compared to 39.6 μmol ²⁹N₂ in the no‐alkyne control (p < 0.05) (Figure 1A). Longer chain alkynes (≥C₆) were less inhibitory to the ‘Ca. Kuenenia’ biomass than the short chain‐length alkynes, with no significant differences observed for ≥C₆ amendments compared to the control (Figure 1A).

FIGURE 1.

Anammox activity quantified by the production of ²⁹N₂ from ¹⁵NO₂ ⁻ and ¹⁴NH₄ ⁺ (A), and denitrification activity quantified by the production of ³⁰N₂ (B). Statistically significant differences (p < 0.05) between the total cumulative ²⁹N₂ production over the 10‐day incubation relative to the untreated controls are indicated by an asterisk. Error bars represent the standard error (n = 3).

The phenylacetylene treatment resulted in a 36.6% reduction of ²⁹N₂ produced compared to the DMSO control (23.6 μmol compared to 37.2 μmol), and 39.5% loss of activity when compared to the unamended control (Figure 1A). As phenylacetylene was dissolved in DMSO due to its low solubility in water, the inhibition of anammox by DMSO alone was determined but was found not to affect ²⁹N₂ production (Figure 1A).

We also determined the production of ³⁰N₂ to test for denitrification activity in anammox biomass, given that the bioreactor used to generate the biomass did not contain a pure culture. ³⁰N₂ production followed similar trends as ²⁹N₂, with the C₈ and DMSO treatments remaining most active and producing 2.2 μmol and 1.8 μmol ³⁰N₂ over 10 days, corresponding to 103% and 86% of the control, respectively (Figure 1B). After 4 days, production of ³⁰N₂ ceased in the C₂–C₅ and phenylacetylene incubations (Figure 1B). There were statistically significant differences in the production of ³⁰N₂ in the ≤C₅ treatments compared to the control, producing 0.38–0.70 and 2.2 μmol of ³⁰N₂, respectively (p < 0.05; Figure 1B).

3.2. Investigating the Target of Inhibition by Alkynes Using ¹⁵N‐N₂H₄

To ascertain which part of the anammox pathway was inhibited by the C₂–C₅ alkynes and phenylacetylene, ¹⁵N‐N₂H₄ was added to the incubations as a substrate instead of ¹⁵NO₂ ⁻ and ¹⁴NH₄ ⁺. As hydrazine is a physiological intermediate in the anammox metabolism, the use of ¹⁵N‐labelled hydrazine allows for determining whether the anammox pathway is affected by the alkynes up‐ or downstream of the hydrazine intermediate (Figure S5). The continuous production of ³⁰N₂ in the presence of 10 μM acetylene over 72 h (Figure S6) indicates that N₂H₄ oxidation at the hydrazine dehydrogenase is not impaired. Furthermore, neither the addition of C₂–C₅ linear alkynes nor phenylacetylene had any inhibitory effect on the ¹⁵N‐N₂H₄ oxidation to ³⁰N₂ (Figure 2). In contrast, alkyne addition had a marginally stimulatory effect on anammox activity. Whereas the unamended control produced 10.7 μmol of ³⁰N₂, the alkyne‐treated incubations yielded 10.6–13.8 μmol ³⁰N₂. Furthermore, the conversion efficiency of ¹⁵N‐N₂H₄ to ³⁰N₂ during the 240 h of incubation increased from 71.3% in the control and phenylacetylene‐amended incubation to 92% in the presence of propyne, based on the theoretical N₂ yield (Figure 2).

FIGURE 2.

Production of ³⁰N₂ from ¹⁵N‐N₂H₄ by ‘Ca. Kuenenia stuttgartiensis’ cells in response to the addition of 10 μM C₂–C₅ alkynes or phenylacetylene. Statistically significant differences between cumulative ³⁰N₂ production in the presence or absence of inhibitors and the abiotic control are denoted by *p < 0.05. Error bars represent the standard error (n = 3).

3.3. Using NO Donor to Explore the Inhibitory Effect of Alkynes on the Anammox Pathway

To identify whether the anammox pathway was inhibited at the nitrite reduction (by Nir) or the hydrazine formation (by Hzs) step by the addition of alkynes, ‘Ca. Kuenenia stuttgartiensis’ biomass was incubated with ¹⁵NH₄Cl and a NO donor (DEA‐NONOate) instead of NO₂ ⁻. After the addition of linear C₂–C₅ alkynes or phenylacetylene, the biomass stopped producing ²⁹N₂ (Figure S7), and this difference was statistically significant for the butyne‐amended sample compared to the unamended control, which produced 0.076 μmol of ²⁹N₂ during the 6‐h incubation (p < 0.05; Figure S7).

3.4. Soil Slurry Incubations Amended With C₂–C₅ and Phenylacetylene

To evaluate alkynes as tools to inhibit anammox in environmental samples and for discriminating between the activities of anammox and other nitrogen‐cycling microorganisms, soil samples were extracted from a depth of 40 cm from a constructed wetland planted with Phragmites. This depth was chosen because it contained a greater abundance of anammox than other depths (2.45 × 10⁸ copies of hdh per g of dry weight soil; Table S1).

Soil slurries using material sampled from the constructed wetlands were incubated with 2 mM ¹⁵NO₂ ⁻ and ¹⁴NH₄ ⁺ for 10 days at 28°C, shaking at 110 rpm. Based on the production of ²⁹N₂, C₂–C₅ alkynes and phenylacetylene had an inhibitory effect on anammox activity also in the soil slurries (Figure 3A). The unamended control incubations produced 8.1 μmol of ²⁹N₂ whereas this dropped to between 1.9 and 4.7 μmol of ²⁹N₂ in the alkyne treatments (corresponding to a 42%–77% reduction in ²⁹N₂ production). Contrastingly, no significant differences in the production of ³⁰N₂ were observed between the control incubations and the butyne and phenylacetylene treatments, which produced 3.7–4.1 μmol ³⁰N₂, respectively, equivalent to 85%–101% of the untreated control. The acetylene and pentyne treatments produced the lowest amount of ³⁰N₂ over 10 days, with 1.3 and 2.3 μmol, respectively. However, there was still no statistically significant difference between the two treatments and the unamended control (Figure 3B).

FIGURE 3.

N₂ production in soil slurry incubations in response to C₂–C₅ alkynes or phenylacetylene amendment. (A) Anammox activity in the slurry based on ²⁹N₂ production from ¹⁵N‐NO₂ ⁻ and ¹⁴N‐NH₄ ⁺, and (B) denitrification activity quantified by the production of ³⁰N₂. Statistically significant differences between incubations in the presence of inhibitors and untreated controls are denoted *p < 0.05. Error bars represent the standard error (n = 3).

3.5. Bacterial Diversity in the Reactor Biomass and Soil Slurries

The anammox bioreactor used in these studies did not contain an axenic culture. To investigate the bacterial diversity and the relative abundance of anammox in the bioreactor, DNA was extracted, and the diversity was analysed by bacterial 16S rRNA gene amplicon sequencing.

Anammox bacteria dominated the bacterial community in the bioreactor, and approximately 70% of the reads were assigned to a single ASV classified as ‘Ca. Kuenenia sp.’ affiliated to phylum Planctomycetota (Figure 4). As expected, the bioreactor also contained microorganisms not involved in anammox. The most abundant taxa included Ignavibacterium sp., a group of obligate anaerobes (7% relative abundance). The putative nitrogen‐fixing bacterium Ensifer sp. also featured amongst the top 4 most abundant ASVs (2% relative abundance). Circa 1% of ASVs were assigned to the genus Pseudomonas. No other anammox bacterium was detected in the bioreactor (Figure 4).

FIGURE 4.

The most abundant bacterial phyla detected by 16S rRNA gene amplicon sequencing from biomass samples. DNA was extracted in triplicate.

DNA was also extracted from the soil slurries, and 16S rRNA gene amplicon sequencing was performed to assess the community structure, relative abundance, and identity of the microorganisms present. As expected, the microbial community in the soil samples used for the soil slurry incubations was more diverse compared to the bioreactor consortium. Two of the most abundant microorganisms detected in the soils were the anammox bacteria, ‘Ca. Scalindua sp.’ and ‘Ca. Kuenenia sp.’ accounting for an average of 2.5% and 1.2% of the total reads from soil samples, respectively, both belonging to phylum Planctomycetota (Figure 5). Ammonia‐oxidising bacteria included Nitrosomonas mobilis (0.1%–0.2% relative abundance), Nitrosococcus sp. (< 0.15%) and Nitrosospira (0.05%–0.07%). The detection of microorganisms typically restricted to marine habitats, such as ‘Ca. Scalindua sp.’ and Nitrosococcus, in the wetland soil samples reflects the oftentimes elevated salinity levels exhibited by landfill leachates, explaining the presence of typically marine species (Setiadi and Fairus 2003; Zhou et al. 2017). Furthermore, a mixed community of facultative and obligate anaerobes affiliated with the Chitinophagaceae family, as well as aerobes including the genus Methylocaldum, was present. Lastly, Nitrospira were not detected in the wetland soils, which is important for considering the key microbial players contributing to nitrous oxide emissions in this wetland ecosystem. Consequently, the impact of comammox Nitrospira activity on nitrogen cycling was not considered in this study.

FIGURE 5.

The most abundant bacterial phyla detected by 16S rRNA gene amplicon sequencing from soil samples. DNA was extracted in triplicate.

4. Discussion

This study set out to investigate the effects of alkynes on anammox, including their target of inhibition, and to evaluate their suitability as selective inhibitors. The sensitivity of ‘Ca. Kuenenia stuttgartiensis’ to 10 μM C₂–C₅ linear alkynes (Figure 1) observed in this study mirrors the responses of the AOA (Taylor et al. 2015; Wright et al. 2020), even if the enzymatic machinery in anammox bacteria is fundamentally different from those in aerobic ammonia oxidisers. In contrast, observations in this study for the ‘Ca. Kuenenia stuttgartiensis’ biomass and wetland soil slurry experiments are different from AOB, which are inhibited by high chain length alkynes of up to C₉ (Hyman et al. 1988; Taylor et al. 2013). Inhibition of anammox by acetylene has been reported previously (Jensen et al. 2007; Kartal, Maalcke, et al. 2011), but this study expands the known range of alkyne inhibitors of anammox to propyne, butyne, pentyne, and phenylacetylene.

Based on our data (Figure 1), the inhibition of ‘Ca. Kuenenia stuttgartiensis’ by phenylacetylene is most akin to the responses observed with ‘Ca. Nitrosocosmicus franklandus’ (Wright et al. 2020). Phenylacetylene is a known inhibitor of both AOA and AOB (Wright et al. 2020). Although the effect of phenylacetylene on anammox had not been evaluated until now, inhibition of anammox by aromatic compounds, including phenol, o‐cresol, p‐nitrophenol, o‐chlorophenol and quinoline has been observed previously (Ramos et al. 2015). Agricultural soils amended with 1 mM phenylacetylene reported steep declines in net nitrification, ranging from 93% to 95% (Rojas‐Pinzon et al. 2024). However, environmental studies in habitats relevant to anammox, such as marine sediments, WWTPs or wetlands using phenylacetylene are very limited, and this represents another avenue for future research.

Hydrazine synthase has been proposed as the target of inhibition by acetylene, but it has also been suggested that multiple targets may exist (Oshiki et al. 2022). Our results indicate that in incubations with ammonium and nitrite as substrates, anammox is inhibited by the addition of short‐chain linear alkynes and, to a lesser extent, phenylacetylene (Figure 1). The production of ³⁰N₂ from ¹⁵N‐N₂H₄ in the presence of these alkynes (Figure 2) demonstrates that hydrazine dehydrogenase is not the target of these inhibitors. In contrast, the addition of alkynes in the presence of ¹⁵NH₄ ⁺ and NO as substrates reduced the activity of ‘Ca. Kuenenia stuttgartiensis’ biomass by > 70% (Figure S7). This suggests that hydrazine synthase is the target of inhibition by the short‐chain linear alkynes (Figures 1, 2, S4, and S7). This conclusion, although strongly suggested by our data, is inferential and future studies should attempt to investigate the target and mechanisms of inhibition using purified enzymes in vitro.

The mechanism of inhibition of hydrazine synthase by alkynes was not identified in this study. The hydrazine synthase is a dimer of heterotrimers, and each heterotrimer contains two c‐type heme containing active sites (Kartal, Geerts, and Jetten 2011; Dietl et al. 2015). The first active site, heme γI, is responsible for reducing NO to hydroxylamine and the second active site, heme αI, catalyses the N–N bond formation between ammonia and hydroxylamine (Dietl et al. 2015). The active sites are connected via an intramolecular tunnel, through which hydroxylamine migrates from the γI site to the αI site. A zinc site is present near the αI site, as well as a short tunnel which allows ammonia to enter (Dietl et al. 2015). The mechanism of N–N bond formation at the αI site has been studied but not fully resolved (Su and Chen 2021; Versantvoort et al. 2025). Whether alkynes interfere with N–N bond formation, NH₂OH formation, or both is uncertain. Inhibition of ammonia and methane monooxygenase is thought to occur via oxidation of acetylene into a ketene intermediate, which covalently binds to an amino acid via a nucleophilic attack (Hyman and Wood 1985; Pham et al. 2015). In cytochrome P450 enzymes, several mechanisms of alkyne inhibition are known including ketene formation and heme alkylation (Ortiz de Montellano 2019). However, these mechanisms involve oxidation of alkynes, and it is not certain whether this would occur with hydrazine synthase. Acetylene also inhibits several anaerobic metalloenzymes including hydrogenase, nitrogenase and nitrous oxide reductase (Smith et al. 1976; Yoshinari and Knowles 1976). Whether alkynes interact with the heme sites within the hydrazine synthase, would require future investigation.

The observed inhibition of anammox by alkynes corroborates environmental studies on the inhibition of anammox by hydrocarbons. For example, ethane has also been shown to suppress anammox activity (Tan, Nie, et al. 2022), and the activity of anammox in sediment slurries is attenuated by crude oil contamination (Ribeiro et al. 2016). Hydrocarbon emissions tend to peak during the early fermentation stage of a life cycle of a landfill (Duan et al. 2021). Thus, a high occurrence of oils in the landfill could result in a higher concentration of hydrocarbons in landfill leachate, which, in turn, affects the activity of the anammox bacteria in downstream constructed wetlands. Still, marine anammox species of the genus ‘Ca. Scalindua sp.’ have been detected in hydrocarbon‐rich seeps in oceanic environments such as the Sea of Cortés (Russ et al. 2013) and may thus be well‐suited for use in bioremediation efforts targeting oil spills, which often create oxygen‐limited environments (Boufadel et al. 2010).

We observed an inhibitory effect for acetylene and propyne, which caused an 89%–97% reduction in ²⁹N₂ production on ‘Ca. Kuenenia stuttgartiensis’ (Figure 1A), and a 60%–78% reduction in the wetland soil slurry (Figure 3A). Given these findings, the presence of the C₂ and C₃ alkynes in landfill leachate could substantially diminish the efficiency of ammonium removal by the anammox process in downstream wetlands.

In our soil slurry incubations, we observed an unexpected response of denitrification activity to the amendment with acetylene (Figure 1B). The addition of 10 μM acetylene resulted in a 62%–65% decrease in ³⁰N₂ production compared to the untreated controls (Figure 1B). Previously, it has been demonstrated that 10 μM acetylene does not inhibit denitrification (Jensen et al. 2007), and its use for determining potential denitrification rates in soils has been proposed (Qin et al. 2012). However, the reason for these contrasting observations remains to be determined for future research.

4.1. Appraising the Use of Long‐Chain Alkynes as Selective Inhibitors Compared to Allylthiourea

Specific nitrification inhibitors, especially allylthiourea (ATU), are often used to discriminate between the activities of anammox and aerobic ammonia oxidation in the environment (Ruser and Schulz 2015). This approach is based on the observations that AOB are very sensitive to ATU, and AOA are also inhibited by ATU, albeit at higher concentrations, but anammox are insensitive to ATU (Shen et al. 2013; Jensen et al. 2007). There are uncertainties associated with the use of ATU, as this inhibitor has been reported to degrade within days (Liu et al. 2018), can stimulate nitrification in a prolonged soil incubation (Lehtovirta‐Morley et al. 2013), and one of the breakdown products is urea, an alternative substrate for comammox, AOA, AOB and anammox (Norton et al. 2008; Koch et al. 2019; Zhao et al. 2023). Based on our study, octyne provides an attractive alternative to ATU for evaluating activities of anammox, AOB and AOA, as octyne in the headspace can be easily vented and replenished to circumvent the potential caveats caused by degradation.

Anammox appears to be only partially inhibited by heptyne, with minimal inhibition during the first few days of incubation and still producing approximately 75% of the total ²⁹N₂ produced by the untreated control after 10 days of incubation (Figure 1A). This highlights another potential candidate for distinguishing between activities of anammox, AOB, and AOA. Heptyne causes full inhibition of NO₂ ⁻ production in Nitrosomonas europaea but did not attenuate the activity of AOA (Taylor et al. 2015). The minimal inhibition by heptyne on anammox activity, especially in short‐term incubations, suggests that heptyne may be a suitable specific inhibitor of AOB activity, allowing for the distinction between their activity and those of AOA and anammox. This enables the specific determination of the contribution of AOB to N₂O emissions during nitrification, which warrants further investigation in the future.

5. Conclusions

This research provides insights into the inhibition of anammox by short‐chain alkynes in both culture of ‘Ca. Kuenenia stuttgartiensis’ and in soil slurries. Furthermore, this study demonstrated that short‐chain alkynes and phenylacetylene inhibited hydrazine synthase. Given that anammox activity was not inhibited by long‐chain alkynes including 1‐octyne (Figure 1A), 1‐octyne is a promising inhibitor for assessing contributions of anammox and aerobic ammonia oxidisers to nitrogen cycling in both oxic and anoxic systems.

Author Contributions

Cerys Maryan: conceptualisation, methodology, software, validation, formal analysis, investigation, writing – original draft and editing and visualisation. Guylaine H. L. Nuijten: methodology, resources. Andrew T. Crombie: methodology and writing – review and editing. Sebastian Lücker: conceptualisation, methodology, resources and writing – review and editing. Victoria Gibson: resources, writing – review and editing, and supervision. J. Colin Murrell: conceptualisation, writing – review and editing and supervision. Marcela Hernández: conceptualisation, software, writing – review and editing, and supervision. Laura E. Lehtovirta‐Morley: conceptualisation, methodology, resources, writing – review and editing, supervision, project administration and funding acquisition.

Funding

This work was supported by Biotechnology and Biological Sciences Research Council (BB/T008717/1), Royal Society (DHF\R1\211076, DH150187), European Research Council (UNITY 852993), and Soehngen Institute of Anaerobic Microbiology.

Conflicts of Interest

The authors declare no conflicts of interest.

Supporting information

Data S1: Supporting Information.

Data Availability Statement

The 16S rRNA gene sequencing data have been deposited in the NCBI Sequence Read Archive (SRA) under the BioProject accession number PRJNA1290922.