A new strategy for greenhouse gas emission reduction in the anaerobic/anoxic/oxic biological treatment process using exogenous N-acyl-homoserine lactones, a quorum-sensing signaling molecules

Methane (CH₄), nitrous oxide (N₂O), and carbon dioxide (CO₂) stand as pivotal greenhouse gases (GHGs), contributing 15–20, 6, and 49% to the greenhouse effect, respectively (Su et al. 2003). Over the preceding century, the global warming potentials (GWPs) of CH₄ and N₂O per unit mass have been noted as 25 and 298 times that of CO₂, respectively (Edenhofer et al. 2014), playing a critical role in the dynamics of climate change. Wastewater treatment plants (WWTPs) are estimated to account for nearly 5% of global non-CO₂ GHG emissions, with projections indicating a 22% rise by 2030 (Valkova et al. 2021; Ren et al. 2022).

The GHGs in WWTPs originate from the biochemical reaction stages and the energy consumption of the equipment during the wastewater treatment process. CH₄ generated during the biochemical reaction process is the primary GHG emission source. Direct emissions of CH₄ during the wastewater treatment process mainly occur in the anaerobic and aerobic zones of the bioreactor. Under anaerobic conditions, microorganisms (methanogens) metabolize organic matter to produce CH₄. Depending on the substrate, this process can be classified into three categories: (1) CO₂ reduction type mainly utilizes H₂ and formic acid as electron donors to reduce CO₂ and generate methane, for example, Methanobrevibacter (Leadbetter & Breznak 1996); (2) methylotrophic type produces methane by either reducing the methyl group in methyl compounds with H₂ or through the bifunctional activity of methyl compounds themselves, for example, Methanococcus (Whitman et al. 1982); (3) acetate type, which decomposes acetate, oxidizing the carboxyl group of acetate into CO₂ and reducing the methyl group to methane, for example, Methanosaeta (Garcia 1990; Ferry 1999). The release of CH₄ in the aerobic and anoxic zones is mainly associated with the disturbance of water flow caused by aeration or mechanical agitation, which accelerates the release of dissolved methane gas from water to the atmosphere (Ma et al. 2023).

N₂O is primarily generated during the biological nitrification and denitrification stages in the wastewater treatment process. During the nitrification stage, there are two pathways for N₂O production in the nitrification process: under conditions of high and low ⁠, ammonium nitrogen is oxidized to hydroxylamine (NH₂OH), which is further oxidized to nitryl group (NOH), while simultaneously generating N₂O. During the denitrification stage, nitrate nitrogen is ultimately converted into N₂ through the traditional four-step heterotrophic denitrification process, accompanied by the production of N₂O (Ding et al. 2018). Additionally, N₂O is also generated during the processes of aerobic denitrification (Zhang et al. 2023a) and heterotrophic nitrification (Lu et al. 2024). Under the circumstance of low dissolved oxygen (DO) concentration, ammonia-oxidizing bacteria employ ammonia or hydrogen as the electron donor to reduce to N₂O. During the denitrification process, when the activity of nitrous oxide reductase is inhibited, N₂O cannot be further reduced to N₂. Consequently, it accumulates and is emitted into the atmosphere.

Quorum sensing (QS) can regulate the growth, reproduction, and trait expression of bacteria through N-acyl-homoserine lactones (AHLs) and affect the biological behavior of bacteria. Since Valle et al. found AHLs in activated sludge (Valle et al. 2004), a large number of researchers have focused on the QS in activated sludge. In terms of denitrification, the increase in C8-HSL levels can increase the oxygen transfer resistance of the biofilm sludge layer, thereby reducing the oxygen mass transfer efficiency and inhibiting the growth of the nitrite-oxidizing bacteria (NOB) population within the biofilm (Yan et al. 2020). The addition of AHL has been found to increase the abundance and diversity of functional bacterial communities, and the relative abundance of key functional microbial groups, including bacteria involved in carbon and nitrogen removal as well as extracellular polymeric substances (EPS) production, also undergoes corresponding changes. Signal molecules can regulate bacterial physiological behaviors (Ren et al. 2010), influencing the synthesis of EPS and the formation of biofilms. They also impact the microbial composition, functional capabilities, and proliferation rate of sludge, subsequently affecting the structure and dominance of bacteria (Li et al. 2014).

In addition, exogenous AHLs can also change the microbial community structure in sewage, thereby affecting the production of GHG. Cheng et al. found that exogenous C6-HSL can significantly affect the transcription of nitrite reductase and nitric oxide reductase genes in denitrifying bacteria (DNB), inhibit the production of N₂O by DNB in aerobic conditions, and stimulate the production of N₂O in anaerobic conditions (Cheng et al. 2017). Lv et al. found that exogenous AHLs can improve the performance of anaerobic granular sludge to degrade organic matter and increase the specific methanogenic activity by 17.1% (Lv et al. 2018).

There are few relative studies on the relationship between QS and GHG emission in the anaerobic/anoxic/oxic (A/A/O) process, which is the popular typical process to remove nitrogen and phosphorus simultaneously in sewage. With the deepening of research on GHG emissions in the sewage treatment process, it is urgent to study how QS affects GHG emission characteristics in this process.

This study involved the addition of two exogenous AHL signal molecules, C6-HSL and C8-HSL, into the activated sludge process. The effects of these signal molecules on pollutant removal efficiency and GHG emission characteristics, as well as their impact on denitrification performance, sludge properties, and microbial communities in the A/A/O process, were investigated. The aim was to elucidate the mechanisms underlying GHG emissions and verify the following hypotheses: (1) AHL signal molecules exhibit a phenomenon of high-concentration inhibition and low-concentration promotion regarding GHG emissions, and (2) AHL signal molecules influence GHG emissions by altering the structure of functional microbial communities.

Artificially simulated wastewater was used as an influent in this experiment. The specific parameters were as follows: chemical oxygen demand (COD) of 200–250 mg/L, total nitrogen (TN) of 35–40 mg/L, total phosphorus (TP) of 3.5–4.0 mg/L, pH of 6–8. The concentrations of CaCl₂, MgSO₄·7H₂O, and trace elements were 30, 27, and 1 mL/L, respectively. The returned sludge from the Longwangzui Sewage Treatment Plant in Wuhan was taken as the inoculated sludge. Prior to the formal experiment, batch experiments were conducted to compare and select three signaling molecules, namely C6-HSL, C8-HSL, and C12-HSL. After comparison, it was found that C8-HSL and C12-HSL exhibited favorable promoting effects on the removal of ammonia nitrogen and the generation of nitrate nitrogen. The following synthetic AHLs (>97%) were purchased from Sigma-Aldrich (China-Mainland): N-octanoyl-DL-homoserine lactone (C8-HSL), N-dodecanoyl-L-homoserine lactone (C12-HSL).

A laboratory-scale A/A/O reactor was used in this study (Figure S1). The effective volumes of A/A/O and secondary sedimentation tanks were 2.2, 2.2, 7.6, and 7 L, respectively. During the test, control the inlet flow rate to 1 L/h. The temperature was controlled at about 25 °C and DO in the oxic tank was controlled at about 3.0 mg/L. The returned sludge rate was 70%, and the inner refluxed nitrification ratio was 300%; the sludge retention time was 15 days. Based on the research of Hu et al. (2016), three doses (5, 25, and 50 nM) were selected as experimental conditions for the subsequent A/A/O system operation. A total of seven operating conditions were set up, and each operating cycle was 30 days. The operating parameters of different conditions are shown in Table S1.

During the stable operation, the concentrations of COD, ⁠, TN, and TP in the influent and effluent in A/A/O were measured every other day. In order to explore the effect of exogenous AHLs on the GHG emission of the system, the GHG emissions and GHG-related biological activities in the anaerobic tank, anoxic tank, and oxic tank were detected at the end of each condition. The activity of coenzyme F₄₂₀ was detected by spectrophotometry at 420 nm (Lv et al. 2018). The denitrifying bacteria activity was characterized by a specific nitrite reduction rate (SNIRR) and a specific nitrate reduction rate (SNRR). The specific oxic rate (SOUR) was used to directly reflect the activity of microorganisms and their degradation rate of organic matter (Zubir et al. 2024). A specific ammonia oxidation rate (SAOR) and a nitrite oxidation rate (SNOR) (Yu et al. 2019; Zheng et al. 2019) were used to directly reflect the activity of nitrifying bacteria. The specific detection methods of microbial activity are in the Supplementary Materials. The concentrations of N₂O and CH₄ in the samples were measured using a Clarus 680 gas chromatograph (PerkinElmer, USA).

In order to analyze the community structure and diversity of the A/A/O system after the addition of exogenous signaling molecules, samples obtained from stable operation days were preserved in a −80 °C refrigerator. The genomic deoxyribonucleic acid (DNA) was extracted using the FastDNA™ Spin Kit (MP Biomedicals, Santa Ana, CA, USA) according to the manufacturer's instructions. The purity and concentration of nucleic acid were accurately determined using a Qubit 3.0 fluorescent agent. To purify the polymerase chain reaction (PCR) products, an ion-exchange purification kit was utilized. The V3–V4 region of DNA was amplified using the positive primers 515F (5′-GTGCCAGCMGCCCCGCGG-3′) and reverse primers 806R (5′-GGACTACHVGGGTWTCTAAT-3′). Finally, high-throughput sequencing was performed on the Illumina MiSeq platform to obtain more accurate results. Alpha diversity analysis and Beta diversity analysis were performed on the obtained data according to the Illumina Miseq sequencing library. The species abundance of each sample was counted at different taxonomic levels. According to the results of high-throughput sequencing, the functional genes of different samples were predicted, the abundance information and annotation information of each function in different samples were obtained, and the functional changes of the system were explained.

The removal efficiency of in the A/A/O system under different conditions is shown in Figure 1(b). It can be seen that both C8-HSL and C12-HSL improved the removal efficiency of and increased with the increase of dosage. In the early stage of adding AHLs, the balance between AHLs and flora in the original QS system was destroyed, which decreased the ammonia oxidation capacity of the system. After about 1 week, the system re-balanced and the removal efficiency gradually recovered and then improved over 2 weeks. In P1, the average effluent concentration was 1.07 ± 0.29 mg/L, and the average removal efficiency was 97.08%. In P2, P3, and P4, the average concentrations in effluent were 0.56 ± 0.14, 0.47 ± 0.12, and 0.38 ± 0.09 mg/L, and the average removal efficiencies were 97.63, 98.46, 98.96%, respectively. Studies have shown that C8-HSL enhanced the ammonia oxidation capacity and nitrification capacity of the system by increasing the proportion of ammonia-oxidizing bacteria (AOB) and promoting the secretion of EPS (Liu et al. 2021; Li et al. 2024), which caused the increase of the removal efficiency of ⁠. C12-HSL was also beneficial to the enrichment of AOB, the secretion of EPS, the increase of proportion of protein (PN) in EPS, increased the compactness and adsorption capacity of the zoogloea, promoted the secretion of endogenous AHLs, which further strengthened the mechanism (Zhang et al. 2023b).

The TN removal efficiency of the A/A/O system under different conditions is shown in Figure 1(c). Both C8-HSL and C12-HSL promoted the removal efficiency of TN in the system, and the effluent TN concentration decreased with the increase in dosage. In P1, the average TN concentration of effluent was 14.80 ± 0.83 mg/L, and the average removal efficiency was 60.38%. In P2, P3, and P4, the average TN concentrations of effluent were 14.45 ± 0.64, 13.65 ± 0.45, and 12.93 ± 0.54 mg/L, and the average removal efficiencies were 62.39, 64.01, and 65.56%, respectively, indicating that C8-HSL improved the TN removal performance of the system. It was found that the removal efficiency of TN was low at the initial stage. After a period of time, the TN concentration of effluent began to decline until it was stable. It was speculated that C8-HLS temporarily decreased the TN removal performance in the early stage, as time went by, the sludge re-adapted to the environment and promoted the enrichment of AOB, NOB, and DNB (Yan et al. 2020), which enhanced the nitrification and denitrification capacity of the system.

In P5, P6, and P7, the average TN concentrations of effluent were 13.34 ± 0.51, 12.57 ± 0.39, and 11.79 ± 0.66 mg/L, respectively, and the average removal efficiencies were 63.93, 66.31, and 68.24%, respectively. It can be seen that C12-HSL had a stronger effect on the nitrogen removal performance than C8-HSL. There were two reasons for this phenomenon: (1) C12-HSL could improve the biological activity of AOB and strengthen the nitrification ability in the anoxic section, shorten the nitrogen removal pathway, and form short-cut nitrification and denitrification locally (De Clippeleir et al. 2011), which was conducive to nitrogen removal. (2) C12-HSL could promote the secretion of EPS (Tan et al. 2014) and increase the consumption of carbon and nitrogen because of the high content of PN and polysaccharide in EPS, which was conducive to the attachment and retention of microbial cells (Shi et al. 2017; Lin et al. 2018), and improve the denitrification performance.

The TP removal efficiency of the A/A/O system under different conditions is shown in Figure 1(d). It can be seen that C8-HSL and C12-HSL reduced the TP removal performance of the system. In P1, the average TP concentration of effluent was 0.31 ± 0.07 mg/L, and the removal efficiency was 91.24%. On the 30th day of P2, P3, and P4, the TP concentrations of effluent were 2.64 ± 0.16, 1.97 ± 0.17, 1.92 ± 0.16 mg/L, and the removal efficiencies were 28.46, 46.61, 47.79%, respectively. On the 30th day of P5, P6, and P7, the TP concentrations of effluent were 1.20 ± 0.08, 1.05 ± 0.06, and 0.83 ± 0.03 mg/L, respectively, and the removal efficiencies were 67.31, 71.13, and 76.56%, respectively. It was shown that the TP removal performance in the initial stage of every dosage condition was poor, and then recovered slowly, but all of them were not better than P1 in the end. The reasons were as follows: after the addition of AHLs, the balance between the AHLs and receptors in the liquid was destroyed, which caused the phosphorus absorption capacity of the sludge to become worse; after a period of time, the system re-balanced, and the phosphorus removal capacity of the system recovered to a certain extent. In addition, the reason for the decrease of phosphorus removal capacity of the system also may be that C8-HSL and C12-HSL increased in the reactor. was easy to combine with H⁺ in wastewater to produce HNO₂, which can penetrate the cell membrane of bacteria into cells by free diffusion, reduce the pH in cells, inhibit the synthesis of adenosine triphosphate (ATP) (Weon et al. 2002), hinder the conversion of ATP/adenosine diphosphate (ADP), and reduce the phosphorus uptake capacity of sludge.

The N₂O emission under different conditions is shown in Figure 2(b). It can be seen that the characteristics of N₂O emission were similar to those of CH₄. In P1, the N₂O emission in the oxic tank was significantly higher than that of the anaerobic tank and anoxic tank, and the N₂O emissions were 0.72 ± 0.04, 1.03 ± 0.02, and 1.23 ± 0.06 mg/day, respectively. After C8-HSL addition, the N₂O emission was almost higher than that of P1 in the whole A/A/O process and increased with the increase of C8-HSL dosage. In P4, the N₂O emissions reached the maximum, which were 1.16 ± 0.04, 1.40 ± 0.03, and 1.58 ± 0.02 mg/day, respectively, and the total emission was 1.39 times that of P1. At 5 nM concentration, C8-HSL significantly impacts N₂O emissions (p < 0.05), with a 95% confidence interval of [1.13–1.22], [1.36–1.46], and [1.53–1.63]. After C12-HSL was added, the N₂O emission was almost lower than that of P1 in the whole process, but with the increase of C12-HSL dosage, the N₂O emission in the reactor increased gradually. In P5, the N₂O emissions reached the minimum, which were 0.38 ± 0.02, 0.59 ± 0.01, and 0.57 ± 0.03 mg/day respectively, the total emissions were reduced by 48.32% compared with P1. At 5 nM concentration, C12-HSL significantly impacts N₂O emissions (p < 0.05), with a 95% confidence interval of [0.34–0.42], [0.54–0.63], and [0.51–0.62]. In P7, the N₂O emissions of the system were 0.65 ± 0.03, 0.65 ± 0.04, and 0.74 ± 0.03 mg/day, respectively. The total emissions were reduced by 31.55% compared with P1. At 50 nM concentration, C12-HSL significantly impacts N₂O emissions (p < 0.05), with a 95% confidence interval of [0.58–0.70], [0.62–0.72], and [0.68–0.80]. Studies have shown that C8-HSL and C12-HSL were conducive to the enrichment of AOB and NOB, thereby promoting N₂O production. In the oxic environment, AOB was dominant in the process when competed with NOB for DO, which leaded to the nitrification reaction staying in the nitrosation stage for a certain period of time, and the accumulation of NO₂⁻-N was beneficial to the formation of N₂O in the oxic tank (Wunderlin et al. 2012). In addition, AOB can reduce NO₂⁻-N to N₂O through denitrification in the anoxic environment, so the NO₂⁻-N flowing into the anoxic tank through internal reflux promoted the production of N₂O in the anoxic tank (Jiang et al. 2017). In addition, N₂O was a gas that was slightly soluble in liquid, part of the N₂O produced in the anaerobic and anoxic tanks was dissolved in liquid, which escaped from the wastewater under the aeration in the oxic tank, resulting in higher N₂O emissions in the oxic tank than in the anaerobic and anoxic tanks.

CO₂ emission is an important indicator of organic matter decomposition and microbial activity (Awasthi et al. 2017). The CO₂ emissions under different conditions are shown in Figure 2(c). It can be seen that the CO₂ emission in the oxic tank was the highest among all tanks, C8-HSL promoted the increase of CO₂ emissions in the system, while C12-HSL reduced the CO₂ emissions. In P1, the CO₂ emissions were 292.39 ± 12.27, 333.25 ± 13.53, and 343.65 ± 12.37 mg/day, respectively. In P3, the CO₂ emissions reached the maximum, which were 305.98 ± 9.76, 354.52 ± 10.65, and 371.66 ± 13.29 mg/day, respectively. The total emissions were 1.06 times that of P1. At 25 nM concentration, C8-HSL significantly impacts CO₂ emissions (p < 0.05), with a 95% confidence interval of [298.11–323.58], [339.80–359.86], and [340.61–372.90]. After adding C12-HSL, the CO₂ emission of the anaerobic tank and anoxic tank decreased significantly, and the CO₂ emission of the oxic tank increased gradually. In P5, the CO₂ emissions were 132.63 ± 6.58, 107.46 ± 3.27, and 362.65 ± 11.75 mg/day, respectively. Total emissions were reduced by 31.55% compared with P1. At 5 nM concentration, C12-HSL significantly impacts CO₂ emissions (p < 0.05), with a 95% confidence interval of [125.13–142.62], [102.78–112.82], and [355.30–381.08]. In P7, the CO₂ emissions were 156.80 ± 5.76, 212.90 ± 4.81, and 438.80 ± 14.57 mg/day, respectively. The total emissions were reduced by 16.59% compared with P1. At 50 nM concentration, C12-HSL significantly impacts CO₂ emissions (p < 0.05), with a 95% confidence interval of [147.75–166.94], [202.84–219.19], and [419.53–464.86]. C8-SHL promoted the CO₂ emission of the system increased, indicating that C8-HSL was beneficial to the improvement of the degradation rate of organic matter in the system. After the addition of C12-HSL, the increased rate of the CO₂ emission in the oxic tank was the highest, CO₂ emissions of the anaerobic tank and anoxic tank were greatly reduced, indicating that C12-HSL promoted the increase in the organic matter degradation rate in the oxic stage. There were two possible reasons to explain this phenomenon. On the one hand, C12-HSL was conducive to the proliferation of sludge microorganisms and the secretion of EPS (Chen et al. 2019), so that most of the carbon sources in influent were converted into sludge and the production of CO₂ was reduced. On the other hand, the CO₂ generated in the anaerobic tank and the anoxic tank was resolved into liquid and released by aeration after entering the oxic tank (Yan et al. 2014), which increased the CO₂ emission in the oxic tank.

The GWP of different conditions is shown in Figure 2(d). C8-HSL increased the GWP in the A/A/O system and reached the maximum in P4, which was 5.09 ± 0.13 gCO₂-eq/m²/h, 30.85% higher than that in P1 (3.89 ± 0.11 gCO₂-eq/m²/h). The low-dose C12-HSL greatly reduced the GWP in the system and reached the minimum in P5, which was 2.07 ± 0.07 gCO₂-eq/m²/h, 46.79% lower than in P1. In addition, the GWP of the A/A/O system also increased with the increase of the dosage of C12-HSL and reached 3.00 ± 0.10 gCO₂-eq/m²/h in P7. Observing the proportion of three GHGs to GWP, it was found that N₂O and CO₂ were the main GHGs contributing to GWP, which accounted for more than 39.58 and 39.94%, respectively.

The particle size of sludge is related to the mass transfer efficiency and microbial community composition. The smaller particle size of sludge was more conducive to the transport of substances due to its larger specific surface area, while the larger particle size of sludge had an environment suitable for the living of multiple functional flora, with higher diversity and richness of microbial community.

The formation of the activated sludge flocs was a self-protection mechanism for bacteria to resist harsh environments, which directly affected the treatment effect of the system (Li et al. 2020). In order to explore the effect of AHLs on the microstructure of sludge, the sludge floc structure at the end of P1, P4, P5, and P7 was elected to be observed by scanning electron microscopy. The results are shown in Figure 3(b). It can be seen that the sludge floc structure was relatively integrated, dense, and less porous in P1. In P4, the sludge floc particles were clear, the particle size was significantly larger than in P1, the sludge surface was rough, the bacterial micelle structure was dense, and the cocci and bacilli adhered to the sludge surface. In P5, the sludge floc particles were clear, and the sludge particle size was between P1 and P4. In P7, the extracellular polymer on the sludge surface increased, which contributed to the adsorption and fixation of the sludge to the zoogloea and small flocs and helped to reduce the sludge flocs and debris in the liquid. The rough sludge surface, loose sludge structure, and bacteria interlaced on the sludge surface were conducive to the transfer and transportation of nutrients in the sludge and improved the utilization efficiency of macromolecular organic matter (Tran et al. 2015).

N₂O was mainly produced in the nitrification and denitrification stages of the biological denitrification process. It was significant to explore the nitrification and denitrification activity of sludge to study the production of N₂O. The SNIRR and SNRR of sludge under these different conditions are shown in Figure 4(b). It was seen that C8-HSL and C12-HSL were beneficial to the improvement of anoxic sludge denitrification activity. In P1, SNIRR and SNRR were 6.25 ± 0.17 and 7.68 ± 0.11 mg N/(g MLVSS·h), respectively. In P4, SNIRR and SNRR increased to 7.93 ± 0.21 and 9.86 ± 0.09 mg N/(g MLVSS·h), which were 26.88 and 28.39% higher than P1, respectively. In P7, SNIRR and SNRR increased to 8.24 ± 0.19 and 10.37 ± 0.16 mg N/(g MLVSS·h), which increased by 31.84 and 35.03%, indicating that both C8-HSL and C12-HSL can improve the denitrification rate, and the improvement effect on SNRR was higher than that of SNIRR, resulting in accumulation.

The SOUR, SAOR, and SNOR of the oxic tank sludge under these different seven conditions are shown in Figure 4(c). It can be seen that C8-HSL and C12-HSL were beneficial to the improvement of nitrification activity. In P1, P4 and P7, SOUR was 12.62 ± 0.27, 16.28 ± 0.23, and 17.69 ± 0.13 mg O₂/(g MLVSS·h). It was obvious that both C8-HSL and C12-HSL promoted the material exchange and information transfer rate among different bacteria in sludge, accelerated the metabolism of microorganisms, and improved sludge activity. The nitrification rate in the oxic tank has also been improved, and the variation characteristics were basically consistent with SOUR. The SAOR and SNOR were 8.48 ± 0.13 and 13.56 ± 0.27 mg N/(g MLVSS·h) in P1. SAOR and SNOR increased to 10.92 ± 0.17 and 16.83 ± 0.19 mg N/(g MLVSS·h) in P4, which were 28.77 and 24.12% higher than P1. In P7, the SAOR and SNOR were 11.36 ± 0.15 and 17.61 ± 0.13 mg N/(g MLVSS·h), which increased by 33.96 and 29.87%. The results showed that both C8-HSL and C12-HSL promoted the conversion of to and the conversion of to ⁠, and the conversion rate of the latter was higher than that of the former, so that the nitrification reaction remained in the nitrosation stage for a period, resulting in the accumulation of ⁠.

Combined with the changes of nitrogen forms along the A/A/O (as shown in Figure 4(d)), it can be found that in the anaerobic tank was extremely few, and in the anoxic tank and the aerobic tank increased after adding C8-HSL and C12-HSL. In P1, the in the anoxic tank and oxic tank was 0.17 ± 0.01 and 0.44 ± 0.09 mg/L and reached the highest in P4, which was 0.30 ± 0.03 and 0.75 ± 0.12 mg/L, respectively. In P7, was 23.53 and 38.64% lower than that in P1, which was 0.13 ± 0.01 and 0.27 ± 0.06 mg/L, respectively. Higher means higher N₂O emissions, and the accumulation of in the nitrification process was conducive to the denitrification of AOB to produce N₂O (Colliver & Stephenson 2000). In the process of denitrification, the accumulation of would lead to a decrease in the denitrification rate and the accumulation of N₂O (Lu et al. 2014). In addition to the above conditions conducive to the production of N₂O, the increase of sludge particle size after AHLs addition further promoted the production of N₂O. In theory, after the addition of C8-HSL and C12-HSL, the change in denitrification and nitrification activity and the increase of sludge particle size were beneficial to the generation of N₂O. However, the change of concentration in our test was little different from the theory, indicating that the influence mechanism of AHLs on the N₂O emission needs to be further explored.

The limitations of using artificial simulation of domestic sewage lie in the simplicity of sewage components and the fact that the cultivated microbial community is single, while the actual sewage components are complex, the cultivated microbial population is diverse, and there are complex ecological relationships between various microorganisms (Mir-Tutusaus et al. 2018). The actual sewage is complex in composition and contains various rich trace elements, while simulated sewage, despite the addition of some nutrient solutions, still cannot compare with actual sewage in terms of the completeness of mineral elements. The lack of substances required for microbial growth in simulated wastewater and the use of sodium acetate as a small-molecule organic compound in simulated wastewater are more conducive to the growth and absorption of filamentous bacteria (Iorhemen & Liu 2021). Due to the morphological advantage of filamentous bacteria with a large specific surface area (Wilén et al. 2003), the ecological balance between them and flocculent bacteria will be disrupted, thereby affecting the morphology of sludge flocs. This leads to high sludge activity, high EPS content, loose flocs, small particle size, irregular shape, and poor settling performance in simulated sewage systems. The microbial community is single and the species diversity is poor, which can lead to poor GHG emission reduction compared to actual wastewater.

The Alpha diversity index can reflect the number of species and relative abundance of ecosystems and reflect the diversity of microbial communities in different samples. Alpha diversity analysis was performed on the above high-throughput sequencing data, and the results are shown in Table S2. The coverage of 12 samples was greater than 0.99, indicating that the sequencing results covered a wide range and could well reflect the real situation of microbial community structure in sludge samples. It can be seen that the abundance-based coverage estimator (ACE), Chao1, and Sobs of sludge samples in P4, P5, and P7 conditions were higher than those in P1, indicating that these sludge samples had higher richness. It also indicated that C8-HSL and C12-HSL promoted the growth and reproduction of microorganisms in activated sludge of each stage in the A/A/O system, which was beneficial to the improvement of microbial community richness. Comparing the Shannon and Simpson of sludge samples, those of P4, P5, and P7 were also higher than P1, which indicated that C8-HSL and C12-HSL were also beneficial to the improvement of microbial diversity in activated sludge. The higher the richness and diversity of microbial communities in the wastewater biological treatment process, the better the treatment effect of the reactor (Wang et al. 2022). The above results show that C8-HSL and C12-HSL can improve the richness and diversity of the microbial community in the sludge and promote the improvement of the operation effect of the A/A/O system. In addition, due to the different microbial taxonomic composition in different samples, the degree of influence on the operation effect of the A/A/O system was also different.

Beta diversity analysis investigated the similarity and difference of community composition between different samples by comparing the species diversity of microbial communities. In order to compare the similarities and differences of microbial community structure in each unit in the A/A/O system by different AHLs, principal component analysis (PCA) was performed on the species abundance information of each sample microbial community. The analysis results are shown in Figure 5(b). The closer the two sample points were, the more similar the species composition was. The contribution of the first and second components in the PCA was 25.12 and 19.61%, indicating that the selected principal components can fully explain the differences in sample data. It can be seen that the distance between samples in the same system was relatively close, and the distance between samples in different systems was relatively far. The distance between P5 and P7 samples was relatively close, indicating that AHLs could have a significant impact on the microbial community of the A/A/O system, and the difference between different AHLs was large; the difference between the same AHLs was small.

Figure 6(b) shows the microbial community composition of the top 30 bacteria at the genus level. Trichococcus was a denitrifying phosphorus-accumulating bacteria (DNPAOs) with the highest relative abundance in this study (Han et al. 2020). The relative abundance of this bacteria in P4 and P7 was significantly higher than that of other conditions, which provided a biological basis for C8-HSL and C12-HSL to improve the nitrogen removal performance of the system. As a DNPAOs, the relative abundance of Dechloromonas increased in P5 and P7, which was beneficial to the improvement of nitrogen and phosphorus removal performance. As an important DNB and anaerobic fermentation bacteria, Azospira can hydrolyze and ferment macromolecular organic matter into small molecular compounds (Xu et al. 2018) and improve the nitrogen removal performance of the system. In P4, P5, and P7, the relative abundance of Nitrospira as an AOB was improved, which was beneficial to the improvement of the nitrification performance of the system. As a DNBs, Acinetobacter can oxidize to in anoxic conditions and conduct denitrification in oxic conditions (Zhang et al. 2019a; b; c), which was beneficial to the ammonia oxidation in the anaerobic tank and the denitrification in the oxic tank, promoted the accumulation of and the production of N₂O. The relative abundance of Acinetobacter increased in P4 and P7, and decreased in P5, which were consistent with the changes in N₂O emission, indicating that Acinetobacter was relative to N₂O emission in the system. Flavobacterium, a DNBs, without the ability to reduce but can reduce to N₂O in anoxic conditions (Horn et al. 2005). The relative abundance of Flavobacterium increased in P4, and decreased in P5 and P7, indicating that Flavobacterium was closely related to N₂O emissions.

In this study, the main genera of methanogens are shown in Figure 6(c). Methanocella, Methanoculleus, and Methanobacterium were H₂/CO₂-reducing methanogens, Methanosarcina, Methanosaeta, and Methanolinea were acetate methanogens. Methanoculleus had a strong adaptability to the environment and can decompose and utilize macromolecular organic matter such as starch and cellulose to produce CH₄ (Bharathi & Chellapandi 2017). The relative abundance showed P4 > P1 > P5 > P7. The relative abundance of the three methanogens showed P4 > P7 > P1 > P5, which was consistent with the characteristics of CH₄ emission in the system, indicating that the three methanogens were the main sources of CH₄. It can be seen from Figure 6(c) that the relative abundance of acetate methanogens in these four conditions was higher than that of H₂/CO₂-reducing methanogens, indicating that acetate methanogens were the main bearers of methane synthesis in this study, and two kinds of methanogens were affected by exogenous AHLs which led to changes in methane emissions in the system.

The main functional enzymes related to methanogenesis in this study included Mcr, acetyl-CoA synthase (EC: 6.2.1.1), acetate kinase (EC: 2.7.2.1), phosphate acetyltransferase (PTA, EC: 2.3.1.8), and methenyltetrahydromethanopterin cyclohydrolase (EC: 3.5.4.27). In addition, the change of Mcr can better reflect the change in the methanogenic capacity of the system because these three methanogenic pathways will eventually form methyl-CoM and then form CH₄, the relative abundance of Mcr of these four conditions was P4 > P7 > P1 > P5. Acetyl-CoA synthase, acetate kinase, and PTA were important functional enzymes in the biosynthesis of CH₄ with acetic acid as metabolic substrate. These enzymes can convert CH₃COOH into acetyl-CoA (CH₃CO-S-CoA), and then CH₃CO-S-CoA traversed tetrahydromethylpterin (H₄MPT) to generate methyl tetrahydropterin (CH₃-H₄MPT), and CH₃-H₄MPT was catalyzed to generate methyl-CoM (CH₃-S-CoM) by coenzyme M. CH₃-S-CoM was finally reduced to CH₄ under the catalysis of Mcr. acetate kinase in different systems did not change significantly, but the relative abundance of acetyl-CoA synthase and PTA remained in the same order as Mcr, showing that the CH₃COOH methanogenic pathway was closely related to methane production. As an important enzyme in the H₂/CO₂ methanogenesis pathway, formyltetrahydromethotrexate cyclase can reduce CO₂ to CH₄, and the relative abundance of this enzyme maintained the order of P4 > P7 > P1 > P5.

Owing to the absence of functional enzymes in the methyl compounds methanogenic pathways, two methanogenic pathways were selected to compare the relative abundance of functional enzymes in different conditions. It was found that the relative abundance of enzymes in the CH₃COOH methanogenesis pathway was always much larger than that in the H₂/CO₂ methanogenesis pathway. In summary, the CH₃COOH methanogenesis pathway was the main methanogenesis pathway in this study, and the addition of AHLs had a significant impact on the methanogenesis pathway. C8-HSL increased the relative abundance of important functional enzymes in the CH₃COOH and H₂/CO₂ methanogenesis pathways to promote the production of CH₄, while low-dose of C12-HSL reduced the abundance of functional enzymes related to these two methanogenesis pathways and reduced CH₄ emissions in the system. With the increase in dosage, C12-HSL increased functional enzymes related to the two methanogenesis pathways and increased CH₄ emissions.

The biological denitrogenation process was accompanied by the production of N₂O, and its production path is shown in Figure 8(b). The biological enzymes related to N₂O production were mainly ammonia monooxygenase (Amo, EC:1.14.99.39), hydroxylamine oxidoreductase (Hao, EC:1.7.2.6) and nitrite oxidoreductase (Nor, EC:1.7.2.5) in the nitrification process (Zhou et al. 2023). In order to further analyze the mechanism of the effect of AHLs on N₂O production, the above enzyme genes were screened from the Tax4Fun function prediction results, and their relative abundance is shown in Figure 8(a). As shown in Figure 8(b), Amo mainly acted on the conversion of to NH₂OH by AOB, while Hao mainly acted on the oxidation of NH₂OH to NOH. In this process, a high concentration of would lead to the accumulation of NH₂OH and the production of N₂O under the catalyze of Hao, then N₂O₂H₂ was hydrolyzed to produce N₂O after NOH polymerization (Soler-Jofraa et al. 2021). After adding AHLs, the relative abundance of Amo increased, and the order was P7 > P4 > P5 > P1, while Hao was P7 > P1 > P4 > P5, basically consistent with N₂O emissions, indicating that Hao was the key enzyme for N₂O production in the nitrification process.

In anoxic conditions, DNBs consumed organic matter as substrates and relied on denitrification functional enzymes such as nitrate reductase (Nar, EC:1.7.5.1), nitrite reductase (Nir, EC:1.7.2.1), nitric oxide reductase (Nor, EC:1.7.2.5) and nitrous oxide reductase (Nos, EC:1.7.2.4) to reduce or to NO_x or N₂, while N₂O was a byproduct of incomplete denitrification (Guo et al. 2025). It can be seen from Figure 7(a) that the relative abundance of Nar, Nir, and Nor in the system increased after adding C8-HSL and C12-HSL, indicating that AHLs can promote the reduction of to ⁠, the reduction of to NO, and the reduction of NO to N₂O in the denitrification process, and improve the denitrification capability of the system. Nos mainly acted on the process of N₂O reduction to N₂, which was the key step to restrict N₂O production in the denitrification process. The relative abundance of Nos in the system was P5 > P7 > P1 > P4, indicating that C8-HSL could reduce Nos to increasing the N₂O accumulation of the system, while C12-HSL had the opposite effect to reducing the N₂O emission, and the lower the dosage was, the less the N₂O emission was. In summary, C8-HSL and C12-HSL increased the abundance of Nar, Nir, and Nor and promoted N₂O synthesis in the denitrification stage. However, C8-HSL reduced the abundance of Nos and inhibited the reduction of N₂O to N₂, resulting in the accumulation of N₂O in this stage. C12-HSL increased the abundance of Nos, promoted the reduction of N₂O to N₂, and reduced N₂O emissions.

The addition of signal molecules plays a significant role in reducing GHG emissions, primarily manifested in the decrease of nitrous oxide emissions. These signal molecules enhance the enrichment and activity of NOB by increasing sludge particle size, regulating microbial community structure, and impacting the abundance of functional enzymes to mitigate nitrous oxide emissions.

This study demonstrates the positive impact of low concentrations of C12-HSL on GHG emission reduction. It identifies the key functional bacteria influenced by signal molecules as acetate methanogens, Acinetobacter and Flavobacterium, with the main affected functional enzymes being Amo, Hao, and Nos. This research provides theoretical insights into the targeted regulation of GHG emission reduction through signal molecules. However, the limitation of this study is that no signal molecules were found to reduce methane emissions, and further research is needed on how signal molecules affect sludge aggregation. Future research should focus on studying the mechanisms of signaling molecules under different environmental conditions and utilizing cellular mechanics to investigate the effects of signaling molecules on sludge aggregation, in order to reveal the relationship between signaling molecules and GHG emissions reduction.

The main findings were:

• The removal efficiency of pollutants such as ⁠, COD, and TN increases with the higher dosage of exogenous signal molecules. However, the addition of signal molecules reduces the removal efficiency of TP.

• C8-HSL and high-dose C12-HSL promoted the GHG emissions of the system, while low-dose C12-HSL reduced the GHG emissions. When 5 nM C12-HSL was added, the GWP of the system was significantly reduced by 46.79%.

• Both C8-HSL and C12-HSL can create an environment conducive to the production of CH₄ and N₂O by promoting the accumulation of and the increase of sludge particle size.

• C8-SHL and C12-HSL increased the richness and diversity of the microbial community and promoted the enrichment of NOBs and DNBs. Acinetobacter and Flavobacterium was an important factors affecting the emissions of N₂O.

• The CH₃COOH methanogenesis pathway was the main methanogenesis pathway in A/A/O. C8-HSL and high-dose C12-HSL had a promoting effect on the CH₃COOH and the H₂/CO₂ methanogenesis pathway, while low-dose C12-HSL had an inhibiting effect.

• C8-HSL and C12-HSL promoted the synthesis of N₂O, while C8-HSL inhibited the reduction of N₂O to N₂ by reducing the abundance of Nos, resulting in the accumulation of N₂O, while C12-HSL increased the abundance of Nos, thereby reducing N₂O emissions.

This work was supported by the National Natural Science Foundation of China (Grant Nos. 51108360 and 51208397) and the Fundamental Research Funds for the Central Universities (Grant Nos. 2023-vb-030).

Data cannot be made publicly available; readers should contact the corresponding author for details.

The authors declare there is no conflict.