Optimizing intermittent micro-aeration as a strategy for enhancing aniline anaerobic biodegradation: kinetic, ecotoxicity, and microbial community dynamics analyses

Aromatic amines (AAs) are toxic and carcinogenic pollutants present in the industrial manufacture of many organic chemicals (dyes, agrochemicals, and pharmaceuticals) (Fujii et al. 1997; Hu et al. 2014; Arora 2015). The inadequate use and disposal of industrial chemicals, their waste, and effluents are a growing source of aromatic compound contamination in soil and groundwater as a consequence of technological progress (Amaral et al. 2014; Melo et al. 2023). Even though they are a product of the anaerobic reduction of electron withdrawing groups (as the nitro group and azo bond) in substituted aromatic compounds, they are recalcitrant to anaerobic degradation (Field et al. 1995), causing anaerobic zones, that naturally occur in groundwater and soil, to accumulate them.

The persistence of AAs demands urgent attention, from environmental and public health perspectives. These compounds are known to be toxic to plants, affecting their growth and development by inducing genetic damage in their cells (Tao et al. 2017); they also exhibit neurotoxic effects in laboratory animals such as rats (Makhdoumi et al. 2018), and are recognized to have carcinogenic effects in humans, leading to conditions such as bladder and liver cancer (Wang et al. 2019, 2021; Czubacka & Czerczak 2020). Thus, AAs pose a threat to both the environment and humans, as they bioaccumulate in higher organisms and can also be incorporated into the soil through bonding with humic and fulvic acids and into colloidal organic matter, further complicating their biodegradation in the environment (Menezes et al. 2021).

The use of microorganisms for the biodegradation of complex organic compounds has been a subject of great interest, as it is a low-cost and environmentally friendly process. Typically, both aerobic and anaerobic steps are required to completely mineralize xenobiotic pollutants from industrial effluents. The combination of these steps is well-established when performed in reactors arranged in series (Amaral et al. 2014; Tomei et al. 2016) or in single reactors using sequenced operational periods (Menezes et al. 2019; Carvalho et al. 2020). Anaerobic processes, although low-cost and highly effective for treating high organic loads and xenobiotics, are slow and may result in the production of even more toxic byproducts, such as AAs, which are recalcitrant in anaerobic environments but readily degraded in aerobic ones (Spain 1995; Arora et al. 2012; Azeez & Al-Zuhairi 2022). Furthermore, relying solely on aerobic processes would be significantly more costly and generate a larger volume of residual sludge.

Recently, micro-aeration has been studied as a novel technology to improve the anaerobic digestion process for the treatment of diverse effluents, from domestic to pharmaceutical, and petrochemical wastewater (Su et al. 2020; Chu et al. 2021; Fu et al. 2023; Buakaew & Ratanatamskul 2024). The use of intermittent micro-aeration in anaerobic reactors for the degradation of aromatic compounds brings light to the possibility of accelerating anaerobic degradation by favoring the aromatic ring opening process (Siqueira et al. 2018). Compared with the conventional activated sludge, the reduction in aeration in this process significantly decreases the production of excess sludge. Consequently, this leads to cost savings associated with aeration, as it is nearly eliminated. Micro-aeration can stabilize the anaerobic digestion process by maintaining COD and NH₄-N removal rates (Peng et al. 2022), increasing microbial diversity with an increase of facultative bacteria population, scavenging hydrogen sulfide, and improving syntrophic interactions between different microbial groups (Field et al. 1995; Nguyen & Khanal 2018; Su et al. 2020; Buakaew & Ratanatamskul 2024). The consumption of oxygen by the aerobes is enough to create anaerobic microniches, protecting the strict anaerobes from oxygen toxicity (Field et al. 1995; Chen et al. 2020).

Since aniline degradation under aerobic (Takeo et al. 2013; Arora 2015) and anaerobic (Schnell & Schink 1991; Kahng et al. 2000) conditions are well-established, it was selected as a model compound of AA for this study. As a hazardous AA, aniline presents toxicity to microcrustaceans (Abe et al. 2001), growth inhibition and genotoxicity to higher plants (Tao et al. 2017), induces spleen toxicity and neurotoxicity in rats (Makhdoumi et al. 2018), and presents carcinogenic effects in humans (Mohanta & Mishra 2020). However, to our best knowledge, there is little report so far on the micro-aeration or intermittent aeration contribution on the anaerobic biodegradation of aniline and its effects on the microbiome (Cheng et al. 2015; Shi et al. 2023). Also, this study can help in understanding the performance parameters that can be applied when using intermittent micro-aeration to improve the anaerobic degradation of AA.

Considering the harmful effects caused by aromatic pollutants and the obstacle of removing them from nature, this study evaluated the biodegradation of aniline in an intermittently micro-aerated anaerobic environment. The composition and dynamics of the active microbial community in the degradation of aniline tell the story of how the reactor functioned at each condition studied.

Aniline (CAS# 62-53-3, purity >99%), catechol (CAS# 120-80-9, purity >99%), and benzoic acid (CAS# 65-85-0, purity >99%) were purchased from Sigma-Aldrich (St. Louis, MO, USA).

The inoculum used was a 1:2 (v:v) mixture of petrochemical sludge and sludge from a pilot-scale UASB reactor that treated real textile wastewater and produced aniline as a byproduct, operated by Amaral et al. (2014). The initial mixed liquor suspended solids (MLSS) concentration was 7.5 g·L⁻¹.

The mineral media was adapted from Florencio et al. (1993) and Carvalho et al. (2020) and it contained (mg·L⁻¹): NH₄Cl (280), K₂HPO₄ (252), MgSO₄·7H₂O (100), CaCl₂ (70), and 1 mL·L⁻¹ of micronutrient solution, in mg·L⁻¹: FeCl₂·4H₂O (2,000), H₃BO₃ (50), ZnCl₂ (50), MnCl₂·4H₂O (500), CuCl₂·2H₂O (38), (NH₄)Mo₇O₂₄·4H₂O (50), AlCl₃·6H₂O (90), CoCl₂·6H₂O (2,000), NiCl₂·6H₂O (142), EDTA (1,000), and HCl 36% (1 mL·L⁻¹). NaHCO₃ was used in condition T10, at 400 mg·L⁻¹ as a buffer.

Following previous studies by this group, aniline was used as the only carbon source at a concentration of 0.14 mM (13.2 mg·L⁻¹), based on the composition of real textile wastewater (Amorim et al. 2013). Based on the success reported by Menezes et al. (2019) in treating more complex AAs using intermittent pulses of 30 min of micro-aeration every 2 h, this study tested three operating conditions for the degradation of aniline: under single batch strategy, micro-aeration was applied to the anaerobic reactor for either 30 min (T30), 15 min (T15), or 10 min (T10) during a 2-h cycle (micro-aeration/anaerobic), until complete aniline degradation was achieved. These conditions corresponded to oxygens dosages of 4.7, 2.4, and 1.6 L O₂/L reactor/day, respectively, since the range of micro-aeration dosing in an anaerobic system varies from 0.005 to 5 L O₂/L reactor/day (Nguyen & Khanal 2018). Each operating condition had three aniline spikes. Sequential operation of T30, T15, and T10 involved a 2-h settle period and liquid phase disposal followed by refilling when switching between conditions.

A 48-h period without micro-aeration was required before starting the micro-aeration pulses every 2 h to allow the filling of adsorption sites. 2 mg·L⁻¹ of the aniline concentration was adsorbed and the initial concentration was reestablished before starting the operation.

Detailed profiles of the 2-h cycles were performed by collecting samples every 2 min during the micro-aeration period and then every 10 min during the anaerobic period, which lasted 105 min for T15 and 110 min for T10. Both conditions were chosen for profiling due to lower aeration costs among the tested conditions and to the promising results obtained. T15 had 19 samples collected and T10 had 17 samples.

Dissolved oxygen (DO) concentration, pH, temperature (T), redox potential (ORP), and concentrations of ammonium, nitrite, and nitrate were measured according to the Standard Methods (APHA 2005). 1 mL of samples were filtered on 0.22 μm membranes. Aniline, catechol, and benzoic acid concentrations were monitored using high-pressure liquid chromatography (HPLC). The HPLC was equipped with a Shimadzu™ diode array detector (DAD) (LC-20AT, Japan) and a Zorbax™ Eclipse XDB-AT308 column (5 μm, 4.6 × 250 nm) from Agilent™ (Santa Clara, CA, USA), in an isocratic gradient. For aniline, the mobile phase was 30% methanol (HPLC grade) and 70% phosphate buffer solution (5 mM, pH = 7.0). For catechol and benzoic acid, the mobile phase was 30% acetonitrile (HPLC grade) and 70% phosphoric acid solution. The lower quantification limit (QL) of the curve for the benzoic acid was 0.3 mg·L⁻¹ and for the catechol, 0.075 mg·L⁻¹.

All tests with Allium cepa seed system were done according to Leme et al. (2008). Samples from the mixed liquor (20 mL each) were taken from the influent of the first spike of aniline in each condition (T30_in, T15_in, T10_in) and the effluent, after the degradation of the third spike (T30_ef, T15_ef, T10_ef). Allium cepa seeds were germinated in 9 cm Petri dishes covered with two sheets of filter paper with 5 mL of each sample, sealed with plastic film to prevent evaporation losses. Four Petri dishes with 70 seeds per dish were analyzed for each treatment. The experiment was conducted in a randomized design. Mineral water was used as a negative control and as a positive control, methyl methanesulfonate (MMS, 0.04 mM) and Trifluralin 0.84 ppm were used. Germination occurred in a dark room at 24 ± 1 °C for 72 h. The roots were fixed in Carnoy's fixative 3:1 (ethanol:acetic acid, v/v) and stored at −20 °C for posterior analysis. For slide preparation, the root tips were washed three times in distilled water, for 5 min each time, and hydrolyzed in 5 N HCl for 8 min at 60 °C. Then, the root tips were washed again, stained in Schiff's reactive for 2 h in a dark room and washed for complete removal of the reagent. The meristems were squashed with a drop of 2% acetic carmine onto a slide, and mounted with Entellan^®.

Toxicity potential was determined according to the germination index and the average root length of 15 roots per Petri dish, after the 72 h of germination. The experimental unit consisted of one Petri dish for both analyses. Cytotoxic and genotoxic potentials were analyzed by the average mitotic index (MI) and the average genotoxicity index (IGen), respectively. The latter was calculated based on the percentage of nuclear and chromosomal changes per experimental unit, including multipolar anaphases, C-metaphases, nuclear buds, chromosome adherences, chromosome breaks, chromosome losses, chromosome bridges, and micronuclei (MN). Five thousand cells were analyzed per sample and controls under a light microscope. The experimental unit consisted of one slide with 500 cells, with 10 slides per treatment.

Samples were collected for DNA extraction from the inoculum and from the end of each operational condition, following manufacturer's instructions. PowerSoil™ isolation kit (Mo Bio Laboratories Inc. Carlsbad, CA, USA) was used for DNA extraction. Nanodrop 2000™ DNA spectrophotometer was used to measure DNA concentration and purity. The samples were sent for sequencing of the V3/V4 regions of the 16S rRNA gene, using MiSeq Sequencing System (Illumina Inc., USA) platform, performed by Neoprospecta Microbiome Technologies (Florianopolis, Brazil). Alpha diversity indices were calculated using the Past^® software. The sequences identified by Illumina MiSeq^® were submitted to the National Center for Biotechnology Information (NCBI) under project access number PRJNA717023 and the Sequence Read Archive (SRA) are under access numbers as follows (sample#, experiment#, run#): Inoculum (SAMN18475451, SRX10438143, SRR14063417), Condition 1 (SAMN18475452, SRX10438144, SRR14063416), Condition 2 (SAMN18475453, SRX10438145, SRR14063415), and Condition 3 (SAMN18475454, SRX10438146, SRR14063414).

Kinetic parameters were analyzed using Origin™ software (OriginLab, Northampton, Massachusetts, USA) and aniline degradation data were fitted to the first-order decay mathematical model. For toxicity and cytogenotoxicity assays, data were transformed using the formula arc sine ⁠, and analyzed using the BioEstat 5.3 software (AnalystSoft Inc., Chapel Hill, NC, USA). Statistical analyses to evaluate normal distribution were carried out by Lilliefors, while the variance homogeneity was evaluated by the Cochran test. As the data did neither present normal distribution nor homogeneity, they were analyzed by the Kruskal–Wallis non-parametric test, followed by the Student–Newman–Keuls as a posteriori test (p < 0.05 or p < 0.01).

The kinetic decay coefficient (k) increased from 0.20 ± 0.03 in the first spike of T30 to 1.55 ± 0.11 in the third spike. In T15, it rose from 1.51 ± 0.21 to 14.56 ± 3.38. These increases in the kinetic coefficients, of 7.8-fold in T30 and 9.6-fold in T15, reflect the adaptation of the microbial community to the xenobiotic compound under the new experimental conditions. Even though T10 had a shorter total degradation time, the kinetic coefficient did not show a significative change between spikes, suggesting a stabilization of the community coming from T15 cycles. Toräng et al. (2002) verified aerobic biodegradation of aniline, which fitted to a first-order kinetic model. They demonstrated that this result aligned with aniline depletion rates observed in natural water bodies. The model was applied to data from a river contaminated with industrial effluent treated in a treatment plant, revealing a residual aniline concentration of approximately 160 μg·L⁻¹, with a decay kinetic coefficient of 1.8 day⁻¹. Zhu et al. (2012) studied the degradation of aniline and chloroanilines using aerobic granules with DO concentrations above 2 mg·L⁻¹, and found that the kinetics followed the Haldane equation. The authors reported that aniline showed inhibitory concentrations for bacterial growth above 500 mg·L⁻¹ and a maximum specific degradation rate of 111.6 mg·gSSV^−1·L⁻¹. In this study, the degradation of aniline occurred similarly to Toräng et al. (2002), fitting a first-order kinetic model. The concentration used in our experiments is within the range that promotes bacterial growth, according to Zhu et al. (2012).

Parameters t₁ and t_1/2 also indicate microbial adaptation; for the third aniline spike, both parameters were 9-fold and 2-fold lower in T15 and T10 than T30, respectively. These results are promising for practical applications of aniline removal, as lower oxygen demand while maintaining pollutant removal quality implies reduced costs for aeration supply in treatment (Peng et al. 2022).

These results also demonstrate that microorganisms degraded aniline, despite its recalcitrant characteristics, with much less oxygen. However, the kinetic parameters indicate limiting factors in the operation cycle with 10 min of micro-aeration (T10). The decrease in t₁ and t_1/2 parameters by 4-fold from T15 to T10 demonstrate that the microorganisms needed more time to reach maximum growth rates in an environment with less oxygen, highlighting the compound recalcitrance characteristics under anaerobic conditions, which is typically slower than in aerobic growth, since the rate-limiting step of anaerobic degradation is the hydrolysis of the compounds into simple and soluble organics (Lim & Wang 2013; Chen et al. 2020; Wei et al. 2024).

In T30 and T15, 16% of the ammonium present was converted to nitrite, but there was no evidence of nitrate formation (Supplementary Table S1). This may be due to the slow growth and sensitivity of nitrifying bacteria (Abe et al. 2017), once there was no oxygen limitation in T30 or T15. Nevertheless, the total nitrogen (TN) removal increased by 1.7-fold and removal efficiency went from 12.2 to 20.4%, from T30 to T15, indicating the community's ability to adapt to environmental stress, still degrading aniline at faster rates and nitrifying. In T10, nitrite production increased over 10-fold, while the TN removal efficiency decreased to 9.6% and nitrate was also detected. Nitrogen may have been introduced in the system by an N₂-fixing bacteria in this condition since their relative abundance (RA) increased by 4-fold from T15 to T10 (Section 3.5), as reported in the literature (Fitzgerald et al. 2015; Tao et al. 2021).

Under anaerobic conditions, aniline is reported to be first carboxylated to 4-aminobenzoic acid, followed by reductive deamination to benzoic acid, and entering the benzoate route. Then, 3 Acetyl-CoA is produced before entering in the tricarboxylic (TCA) cycle (Schnell & Schink 1991; Kahng et al. 2000). This intermediary byproduct in the aniline degradation pathway was detected in all three operational conditions, with highest concentrations of 4.7 mg·L⁻¹, 3.1 mg·L⁻¹, and 2.5 mg·L⁻¹ in T30, T15, and T10, respectively, showing there was anaerobic degradation in all conditions, despite the energetic preference of the microbes for the aerobic pathway.

Catechol, the aerobic byproduct, was also detected in T30 and T15, with highest concentrations of 0.2 mg·L⁻¹ and 0.3 mg·L⁻¹, respectively. However, there was no detection of this intermediary in T10. The formation of catechol in the aerobic aniline degradation is oxygen-dependent due to the action of mono and dioxygenase-type enzymes (Takeo et al. 2013). The non-detection of catechol, associated with the limited available oxygen in T10 point to the accumulation of metabolites in the peripheral aniline aerobic pathway (aniline to catechol), such as γ-glutamylanilide (ɣ-GA) (Takeo et al. 2013).

The DO accumulated during micro-aeration in P15, which increased ORP while aniline was oxidized (Figure 3(b)). The DO reached its maximum, 1.69 mg·L⁻¹ (21% of saturation), at 15 min and, once micro-aeration ceased, it decreased to less than 10% of saturation at 30 min, reaching a steady value at 65 min. Notice that even after most oxygen had been consumed, ORP continued increasing. The slow increase of ORP in this condition reflects the oxidation of aniline and its byproducts (Figure 3(c)), as oxygen slowly decreased. The steady state of ORP at 65 min is due to the presence of residual oxygen. The same cannot be said for the ORP at P10 (Figure 3(e)). The prompt consumption of DO at P10 diminished the ORP, as reported by Menezes et al. (2019). When the oxygen was no longer consumed (at 65 min), aniline degradation became slower in T15, indicating that there is a dependence on oxygen to perform initial steps of aniline degradation, whether completely aerobic or hybrid pathway.

Aerobic and anaerobic intermediates (Figure 3(c)) followed the DO gradient: during micro-aeration until the end of oxygen consumption the concentration of catechol increased, and the concentration of benzoic acid were low. After oxygen consumption, catechol reached the QL (0.075 mg·L⁻¹) and benzoic acid concentration increased. The detection of both metabolites suggests a concomitant degradation of aniline through aerobic (Schnell & Schink 1991) and anaerobic (Fujii et al. 1997) processes or a degradation by coupling the micro-aeration, to start the process, followed by the anaerobic degradation in a hybrid pathway (Fuchs et al. 2011).

During P10, there was a 46.5 μM (4.3 mg·L⁻¹) decay in the concentration of aniline (Figure 3(d)), without distinction from periods with and without micro-aeration, and aniline consumption was 7 times faster in P10 than in P15. The DO in P10 varied similarly to P15 (Figure 3(e)). It increased during micro-aeration and reached a peak of 1.61 mg·L⁻¹ (also 21% of saturation). However, it decayed faster. In just 10 min after micro-aeration ceased, it reached 12% of saturation and was completely consumed, leaving no residual. The ORP (Figure 3(e)) dropped to negative values and oscillated around zero. The aerobic intermediate, catechol, was not detected in P10 profile (Figure 3(f)) − or at any aniline spikes in T10. Notice that the anaerobic byproduct was only detected after the oxygen was consumed and when the ORP was close to zero. The greatest concentrations of benzoic acid were detected when the ORP was negative.

These results point to the following: partial aerobic degradation and anaerobic degradation occurred to enhance aniline removal in T10. According to Takeo et al. (2013), the first step toward the aerobic degradation of aniline through the catechol pathway is the accumulation of ɣ-GA. It is followed by the oxygenation of the aromatic ring, producing catechol as the final product. Since ɣ-GA has cytotoxic effects when in excess, it can prevent dioxygenase enzymes from converting it to catechol. Considering catechol was not detected in T10 or P10 and this condition was the only one whose effluent presented cytogenotoxic effects (discussed in the next section), it is likely that the intermediate that accumulated in the aerobic pathway was ɣ-GA, suggesting that the aniline aerobic conversion did not occur completely in condition T10 and the main pathway used by microorganisms in P10 and T10 was the anaerobic one. This is supported by the fact that the aniline degradation progressed at the same rate before and after the oxygen ended (Figure 3(d) and 3(f)). The O₂ supplied was mostly used to perform nitrification, once were nitrified at T10 (Supplementary Table S1).

The influent of all conditions induced toxicity, cytotoxicity, and genotoxicity (Table 2; Supplementary Figure S1(a)). Average root lengths, germination, and mitotic indices statistically decreased, while the rates of chromosomal changes during the mitosis process statistically increased, when compared with the negative control. Similar results were reported previously. Aniline also reduced the root growth of wheat seeds and inhibited their germination, reduced the growth of higher plant seedlings, such as Suaeda salsa, and was cytogenotoxic for animal and plant cells (Tao et al. 2017; Makhdoumi et al. 2018; Xu et al. 2020). These data confirm A. cepa sensitivity and its efficiency for detecting wastewaters toxic and cytogenotoxic effects (Tao et al. 2017; Xu et al. 2020).

Additionally, germination index, root length, MI, and genotoxicity of T30 and T15 effluents did not show the statistical difference to the negative control (Table 2), which indicates the absence of toxic and cytogenotoxic metabolites after both treatments. The presence of benzoic acid and catechol at T15 effluent suggests that both byproducts are non-toxic and non-cytogenotoxic using the A. cepa test system. These results ratify biological treatment efficiency in removing toxicity (Amaral et al. 2014; Balapure et al. 2016; Menezes et al. 2019). The effluent of T10 also did not present toxicity. However, cytotoxicity and genotoxicity potentials were observed (Table 2). The decrease in MI in the effluent of T10 can be associated to changes in cell cycle regulation, both at the G1/S and/or G2/M checkpoints, which delayed or prevented cell mitosis (Fioresi et al. 2020). Additionally, when the substrate is highly cytotoxic as observed for T10_ef, its genotoxic effect can be partially masked (da Silva Souza et al. 2018). The genotoxic damage was mainly observed in the form of C-metaphases (Supplementary Table S2 and Figure S1(b)), which are the result of aneugenic agents that inactivate the mitotic spindle of the cell (Fiskesjö 1993). Non-significant micronucleus formation was also visualized in the present work, even though neither chromosomal fragments nor lost chromosomes have been observed in the analyzed cells. The cytogenotoxic effects in T10 effluent support the hypothesis that the oxygen shortage leads to the non-formation of catechol but to the accumulation of ɣ-GA, which is a cytogenotoxic effect-inducer compound in cells, as reported by Takeo et al. (2013).

The diversity indices (Supplementary Table S3) indicate the complexity of the community, considering not only the richness of populations in the sample, but also their proportion in biological reactors, as the RA. The most abundant microbial phyla found in all conditions (Supplementary Figure S2) were Proteobacteria, Bacteroidetes, Chloroflexi, Firmicutes, Euryarcheota, and Actinobacteria, which are commonly reported with RA higher than 1% in anaerobic systems, exhibiting good organic matter removal performances (Köchling et al. 2017; Carvalho et al. 2020). The phylum Proteobacteria stood out with highest RA in the inoculum (58.13%), in T15 (50.99%), and in T10 (80.74%). However, in T30, Bacteroidetes (34.88%) and Proteobacteria (32.49%) were both in evidence. The phylum Actinobacteria also stood out as the second most abundant in T10 (12.21%), whose RA in T15 was only 1.49%. The change in phyla RAs implies their adaptability to aniline and to the different oxygen supply over time. Proteobacteria, Bacteroidetes, and Actinobacteria have also been reported by Li et al. (2020a) in a sequential aerobic batch reactor subjected to high concentrations of aniline, 600 mg·L⁻¹ (6.44 mM), as main characters in aniline and nitrogen removal.

According to the literature (Supplementary Table S4), 18 genera present in the microbial community in all conditions were directly related to the degradation of aniline and its byproducts. Nine of these genera had RA ≥ 1% in at least one of the conditions (Figure 4): Comamonas, Clostridium, Delftia, Flavobacterium, Novosphingobium, Phenylobacterium, Pseudomonas, Pseudoxanthomonas, and Thiobacillus. The 18 genera, all facultative aerobic or fermentative anaerobes, represented 24.7, 42.0, and 41.8% of the microorganisms in T30, T15, and T10, respectively. In the inoculum, only Syntrophus and Rhizobium represented 47% of the microbiota. The presence of different genera performing the same function indicates that the community is more robust and the reactor is more stable, supporting greater organic loads and toxic compounds. This happens because a system that has more metabolic pathways available for the degradation of a compound is more functionally stable, since it depends on metabolic processes happening in parallel, not in series (Nguyen & Khanal 2018).

Ten genera with RA ≥ 1% in at least one of the conditions were also present in biodegradation studies (Supplementary Table S4) for other xenobiotic compounds such as methanol and methanesulfonate (Afipia), phenol (Alicycliphilus), nitroanilines (Bradyrhizobium), sulfonamides (Leucobacter, Oligotropha), polycyclic aromatic hydrocarbons (Longilinea and Thermomonas), triphenyl phosphate (Petrimonas), pyrene (Pseudoxanthomonas), and phenylacetic acid (Sphingopyxis).

Comamonas (3.33; 2.74%), Pseudoxanthomonas (3.28; 2.29%), Petrimonas (34.78; 28.95%), Logilinea (16.57; 8.11%), and Thiobacillus (0.40; 21.03%), were present in T30 and T15 (respectively). All these genera are known for degrading aniline (Boon et al. 2000; Jiang et al. 2019; Shi et al. 2023), benzoic acid (Taylor et al. 1969; Thierry et al. 2004), or other aromatic hydrocarbons (Hou et al. 2019). Thiobacillus is also known for its ability to oxidize H₂S or thiosulfate to sulfate, using oxygen or nitrate as an electron acceptor (Kelly & Wood 2000). On the other hand, elemental sulfur or sulfate can stimulate Petrimonas genus growth, which reduces them to sulfide (H₂S) (Grabowski et al. 2005). Petrimonas was the most abundant population in T30 and T15.

The presence of Clostridium with RA ≥ 1% in all conditions is worth noticing. As an anaerobic fermentative, it participated in the anaerobic degradation of xenobiotic pollutants, textile effluents, and toxic organic compounds (Köchling et al. 2017; Carvalho et al. 2020). Furthermore, Yohda et al. (2017) demonstrated that Petrimonas consortiates with Clostridium for trichloroethylene hydrocarbon degradation, supporting the hypothesis of cooperation between these genera. The consumption of 50 μM of aniline at the P10 profile along with an accumulation of 20 μM of benzoic acid was another evidence of the anaerobic degradation of aniline in T10 when there was no more oxygen available. The studies by Carvalho et al. (2020), Peng et al. (2022), and Shi et al. (2023) also revealed the importance of syntrophic interactions between anaerobic and facultative genera when degrading AA.

The emergence of strict and facultative aerobic microorganisms, in addition to favoring aniline degradation speed, also allowed the permanence of the strict anaerobic genera such as Petrimonas, Longilinea, Clostridium, and the methanogenic archaea. This is possible because in one-stage aerobic–anaerobic reactors the facultative aerobic bacteria consume oxygen quickly, protecting strict anaerobes (Field et al. 1995; Nguyen & Khanal 2018). The fact that the genera Petrimonas, Longilinea, Clostridium, Thiobacillus, Methanosaeta, and Methanobacterium together corresponded to 59 and 65% of the microorganisms present in T30 and T15, respectively, indicate the strong participation of these microorganisms in the AD process and evidence the existing cooperation for the effective removal of aniline and its byproducts toward methane formation (not quantified in this study).

As possible dominant metabolic pathways in T10 (Figure 5), aniline degradation mainly happened anaerobically and aerobic degradation was interrupted by lack of oxygen. Oxygen was disputed for other metabolic activities as nitrification and carbon fixation by the genus Oligotropha. In addition, Pseudomonas, Delftia, Leucobacter, and Thermomonas deserve attention for future research, since they are reported as aniline denitrifiers and together were present with about 50% RA in this study. These denitrifying genera occurrence is supported by the detection of nitrite and nitrate in the mixed liquor (Supplementary Tables S1 and S4), even though no nitrate or nitrite were added. Nevertheless, the presence of nitrogen-fixing genera, such as Bradyrhizobium, Bosea, and Rhizobium, support the N-content increment in T10 (Fitzgerald et al. 2015; Tao et al. 2021).

Detection of benzoic acid and catechol throughout conditions reveals that aniline was degraded by both anaerobic and aerobic pathways. The identified microbial community supports these findings and they are also corroborated by literature on aniline-degrading genera (Schnell & Schink 1991; Alexandra De et al. 1994; Boon et al. 2000; Kahng et al. 2000; Chu et al. 2015; Shi et al. 2023).

A clear cooperative interaction occurred between microbes in T30 and T15, while in T10, there was a competitive environment. In T30 and T15, enough oxygen was available to allow the coexistence of facultative aerobes, supporting the presence of fermentative microorganisms by creating anaerobic microniches (Field et al. 1995; Chen et al. 2020). In T10, however, there was a competition for oxygen that led to competition for organic carbon, making aniline's total degradation time and kinetic coefficients comparable to those in T15. Even though the anoxic environment favored the anaerobic respiration of aniline, the competition for oxygen hindered its complete aerobic respiration in T10, leading to the accumulation of cytotoxicity and indicating that the cooperative microbiome of T15 would be a preferential strategy for enhancing anaerobic aniline degradation without accumulating harmful metabolites.

The use of micro-aeration to improve anaerobic processes has been recently tested and validated as a tool to enhance anaerobic digestion of various wastes (Chen et al. 2020; Su et al. 2020; Chu et al. 2021; Zhang et al. 2022; Fu et al. 2023; Shi et al. 2023; Buakaew & Ratanatamskul 2024; Wei et al. 2024). From the hydrolyzation of complex organic polymers in the conventional anaerobic digestion to biodegradation of xenobiotic pollutants, different methods for oxygen distribution and dosing have been applied. To our knowledge, this is the first report on the use and optimization of intermittent micro-aeration to degrade aniline, one of the most common AA intermediaries of the anaerobic degradation of industrial pollutants. Even though these results’ applicability may be restrained to specific chemicals or residual concentrations, this optimized method may not be limited to liquid effluents, as it can also be a viable design for the application in naturally occurring anaerobic zones in soil and groundwater in bioremediation areas contaminated with diverse organic industrial chemicals.

Recent studies have demonstrated the effectiveness of micro-aeration in enhancing anaerobic digestion processes, including the degradation of complex organic polymers and xenobiotic pollutants. These substances are significant as hazardous byproducts of industrial effluents, posing risks to both the environment and human health. In this study, through the analysis of kinetic coefficients and identification of metabolic intermediates from both aerobic and anaerobic pathways, assessment of the shifts in microbial community, and evaluation of toxicity effects on the A. cepa test system during aniline degradation in each operational configuration, it became clear that intermittent micro-aeration can enhance the anaerobic degradation of aniline. The reactor removed aniline efficiently in all three conditions tested by stimulating cooperation (as in 30 and 15 min of intermittent micro-aeration) or competition (as in 10 min of intermittent micro-aeration) between genera. Although the efficient removal of aniline using 10 min of intermittent micro-aeration, the competition for oxygen in T10 hindered completion of aerobic respiration, leading to cytotoxic and genotoxic effects on A. cepa seeds. Hence, the condition with 15 min of micro-aeration every 2 h was the one with the shortest period tested that removed toxic metabolites efficiently. While this work specifically applies to reactors, it demonstrates that micro-aeration can effectively enhance the anaerobic digestion of AAs. We anticipate this study will set a precedent for future research and applications of micro-aeration in this field. Subsequent works will focus on the role of main aniline degraders, their influence in the overall community, and enzymatic apparel.

This work was supported by FACEPE (Science and Technology Foundation of the State of Pernambuco, Brazil [grant numbers IBPG-0653-3.07/18, APQ-0456.3-07/20]; CNPq (National Council for Scientific and Technological Development [grant number 409165/2021-2]; and CAPES (Coordination for the Improvement of Higher Education Personnel) [master scholarship granted to V.V.S. and Proap and PrInt program number 88887.311967/2018-00].

All relevant data are included in the paper or its Supplementary Information.

The authors declare there is no conflict.