Technical-grade fipronil (Regent® 0.3% G; 97.5% purity) and its principal metabolites—fipronil sulfone (99.7%), fipronil sulfide (98.8%), fipronil desulfinyl (97.8%), and fipronil amide (99.8%)—were obtained from Bayer Crop Science Ltd., India. All other reagents and solvents used throughout the study were of analytical grade and sourced from HiMedia Laboratories Pvt. Ltd., Mumbai, India. For the isolation of fipronil-degrading bacteria, a mineral salt medium (MSM) was used [23]. The composition of the medium (g L⁻¹) was Na₂HPO₄, (5.8); KH₂PO₄, (3.0); NaCl, (0.5); NH₄Cl, (1.0); MgSO₄, (0.25). The pH was set at 7.0 prior to sterilization at 121°C for 20 minutes. For culturing of individual strains, Luria–Bertani medium (LB) was used. The composition of the medium (g L⁻¹) was tryptone (10.0), yeast extract (5.0), and NaCl (10.0).

Soil samples used for microbial enrichment were collected from the contaminated agricultural fields of Udham Singh Nagar of Uttarakhand, India (29.0369° N, 79.4472° E). A sterilized stainless-steel core auger was used for the collection of soil samples. Soil samples were stored in sterilized plastic bags at low temperature until further analysis. Soil samples used for the isolation purpose were air-dried, sieved, and homogenized. Following this, 10 g of soil sample was used as an enrichment culture for 100 mL of sterile MSM [24]. The strains were further incubated at 30 ± 2°C, 100 rpm, for 24 hours under dark conditions. Potent strains were selected by comparing superior growth of individual strains at a given fipronil concentration. Selected strains were further purified, inoculated in LB medium, and cryopreserved at − 70°C until further use.

Molecular profiling of the fipronil degrading bacterial strains S4 and S6 was accomplished through 16S ribosomal RNA gene sequencing [25]. The nucleotide amplification was performed by using universal bacterial primers viz. 8F (5′-AGAGTTTGATCCTGGCTCAG-3′) and 1512R (5′-ACGGCTACCTTGTTACGACTT-3′). Polymerase chain reaction (PCR) amplification was conducted by employing DNA polymerase in a 25 µL reaction volume. The thermal cycling conditions used for amplification were as follows: 95°C for 1 minute, 30 cycles of 95°C for 30 seconds, 54°C for 60 seconds, and 72°C for 90 seconds, with a final elongation at 72°C for 5 minutes. The amplified products were visualized under UV transillumination. The PCR amplicons were purified using a Gene Jet PCR purification kit (Thermo Scientific) and subsequently submitted for Sanger sequencing (Chromus Biotech Ltd., Bangalore, India).

The 16S rRNA gene sequences were employed to BLASTn programs available on the NCBI GeneBank (https://blast.ncbi.nlm.nih.gov/Blast.cgi). Sequences exhibiting ᵙ 100% similarity with the gene sequence available in the genebank were considered for species-level identification. MEGA X software was further used for phylogenetic characterization. The Kimura 2-parameter model and the neighbor-joining (NJ) method were used to determine the evolutionary distance and construction of the phylogenetic tree. Finally, the sequences were deposited in the GenBank nucleotide database to obtain the accession number.

Maximum tolerance concentration (MTC) of the isolated strains was determined by culturing individual strains at fipronil concentrations ranging from 100 to 500 mg L⁻¹ using broth and plate-based assays [26]. Uninoculated flasks were maintained as control. Culture incubation was performed in the same condition as described before. Growth was assessed spectrophotometrically at 600 nm. The highest concentration at which strains were able to proliferate was determined as the MTC for each strain.

A functionally synergistic bacterial consortium (FP-25) was concocted by cultivation and amalgamation of equal proportions of P. furukawaii S4 and A. pusense—to maintain the same cell count. The growth rate and the fipronil degradation ability of individual bacterial isolates and bacterial consortia were compared by conducting a biodegradation assay in MSM supplemented with fipronil in conditions described in section 2.2 [27–30].

A three-factor, three-level Box–Behnken design (BBD) was availed to ascertain the optimum conditions to maximize the fipronil degradation by bacterial consortium. This statistical approach is pre-eminent for its efficiency in recognizing the most suitable environmental parameters, escalating the required output of a process. The total number of experimental runs was determined using the following equation:

N = 2N_f (N_f-1) + Cp, (1)

N_f denotes the number of independent variables, and Cp as the number of center points. In this study, Design-Expert software (v8.0.0, Stat-Ease Inc., Minneapolis, USA) was used to design the experiment. Three coded levels were used for the assessment of the selected independent variables, i.e., incubation temperature (X1), initial pH (X2), inocula concentration (X3) and fipronil concentration (X4). Interaction among independent variables had a total of 17 experimental runs. The experimental design comprised 12 factorial points and 5 center points. The analysis of variance (ANOVA) and the F-test were deployed to assess the adequacy and significance of the model, as indicated in recent studies on optimization of biodegradation processes [31].

The optimized independent variables were validated by conducting a fipronil biodegradation assay in triplicate. The outcomes were confirmed by analyzing the concentration of the parent compound and it’s metabolites by GC-MS.

The biodegradation of fipronil was studied in real contaminated soil samples from the agricultural region of the Uttarkashi district of Uttarakhand. Soil samples procured from twelve geographically distinct agricultural sites of Barsu (30°52'42"N 78°35'45"E), Raithal (30°49'12"N 78°36'03"E), Bhatwari (30°49'18"N 78°37'02"E), and Kyark (30°49'38"N 78°37'51"E)—designated as L1, L2, L3, L4, L5, L6, L7, L8, L9, L10, L11, and L12, respectively, and a control site—were analytically screened for fipronil residues. All the collected soil samples were extracted and analyzed for the presence of fipronil residues (methodology described in section 2.9.2). The effect of bioaugmentation on fipronil biodegradation in real contaminated soil was further assessed by setting up fourteen microcosm treatments. All the experiments were established in triplicates in 250 mL glass bottles. In non-sterilized treatment, the inoculum concentration of consortium FP-25 was added at the optimized dose, as determined by the RSM method. Soil moisture was reattained around 40% in each setup. A soil microcosm experiment was performed at a previously optimized temperature in dark conditions for 30 days. The residual fipronil concentration was determined at regular time intervals by GC-MS analysis.

Two potent fipronil-degrading bacterial isolates, designated as strains S4 and S6, were identified on the basis of phenotypic, biochemical (Table S1), and molecular characterization. Phenotypic characterization revealed that both S4 and S6 strains are Gram-negative, motile, and aerobic rods. Biochemical analysis further indicated that the strain S4 tested positive for oxidase and catalase activities and exhibited positive reactions for maltose, citrate, and mannitol and displayed negative reactions for glucose, xylose, and lactose. The tests were negative for urea, nitrate reduction, and 3-ketolactose and positive for oxidase and catalase. In contrast, S6 demonstrated positive enzymatic activity for urease, 3-ketolactose, oxidase, and catalase and adroitly availed glucose, xylose, citrate, and lactose as carbon sources. Both phenotypic and biochemical analyses revealed characteristics congruous with the Pseudomonas and Agrobacterium families.

Molecular identification through 16S rRNA gene sequencing established that strains S4 and S6 exhibited 100% sequence identity with P. furukawaii and Agrobacterium pusense, respectively (Fig. 1a, b). The obtained 16S rRNA gene sequences for strains S4 and S6 were deposited to the GenBank database under accession numbers KC736660 and KC736661, respectively.

The maximum growth was observed for both strains at 200 mg L⁻¹ of fipronil concentration. The MTC assay indicated that P. furukawaii and A. pusense exhibited fipronil tolerance at concentrations up to 400 and 350 ppm, respectively (data not shown). Individual bacterial strains and the developed consortium, FP-25, were evaluated for fipronil degradation in MSM media. The results indicated that the consortium showed 80.42% fipronil degradation, followed by 72.6% and 70.2% degradation for P. furukawaii and A. pusense, respectively, after 14 days of incubation (Fig. 2). In addition, the consortium indicated a significantly higher growth rate than the individual strains (p < 0.01). The higher growth and degradation by consortium augmented culture may be accredited to the probable degradation of toxic intermediates, owing to the concerted compatibility of the constituting genera [33, 34]. In a similar study, a significant fipronil degradation by three novel bacterial consortia was observed in paddy soils [35].

Five prominent fipronil metabolites evincing mass-to-charge (m/z) ratios of 438, 421, 370, 229, and 141 were formed and identified via GC-MS analysis after a 14-day incubation period. These metabolites were structurally elucidated via explication of the precursor ions and the fragmentation pattern available in the NIST spectral library. The chemical structure was further confirmed by the heteroatom mass isotopomer pattern and affirmed with previously reported studies (Fig. 3). Retention times for fipronil (1) and metabolites 2, 3, 4, 5, and 6 were found to be 16.592, 18.775, 16.458, 18.542, 13.575, and 6.508 minutes, respectively. The metabolites identified were 1-(2,6-dichloro-4-(trifluoromethyl)phenyl) -3-(iminomethyl) -4-((trifluoromethyl)sulfinyl) -1H-pyrazol-5-amine (2) at m/z 438 (mol. wt.: 438.148 g/mol), 1-(2,6-dichloro-4-(trifluoromethyl)phenyl) -4-((trifluoromethyl)sulfinyl) -1H-pyrazole − 3-carbo nitrile (desamino fipronil) (3) at m/z 421 (mol. wt.: 421.12 g/mol); 5-amino − 1-(2,6-dichloro-4-(trifluoromethyl)phenyl) -4-hydrosulfinyl-3-(iminomethyl)-1H-pyrazole (4) at m/z 370 (mol wt.: 384.16 g/mol); 4-hydrosulfinyl-1H-pyrazole-3-carbonitrile (5) at m/z 141 (mol wt.: 141.14 g/mol); 2,6-dichloro-4-(trifluoromethyl)aniline (6) at m/z 229 (mol wt.: 229.99 g/mol).

Based on the metabolites formed, a fipronil (1) degradation pathway by consortium FP-25 was proposed (Fig. 4). Fipronil was degraded or transformed via two routes; (i) fipronil, through the conversion of the nitrile (–C ≡ N) triple bond, is transformed into 1-(2,6-dichloro-4-(trifluoromethyl)phenyl) -3-(iminomethyl) -4-((trifluoromethyl)sulfinyl) -1H-pyrazol-5-amine (2) via reductive addition of ammonia or enzymatic nucleophilic substitution, where a carbon–nitrogen triple bond (–C ≡ N) is converted to an imine group (HC = NH), resulting in a formamidine moiety at the 3-position of the pyrazole ring. Afterwards, 1-(2,6-dichloro-4-(trifluoromethyl)phenyl) -3-(iminomethyl) -4-((trifluoromethyl)sulfinyl) -1H-pyrazol-5-amine (2) through the cleavage of the carbon–sulfur bond between the sulfinyl sulfur and the trifluoromethyl group is transformed into 5-amino − 1-(2,6-dichloro-4-(trifluoromethyl)phenyl)-4-hydrosulfinyl-3-(iminomethyl) -1H-pyrazole (4) undergoes enzymatic or microbial defluoromethylation, eliminating the trifluoromethyl group (–CF₃) attached to the sulfinyl sulfur and resulting in a free sulfinyl group (–SO–) at the 4-position of the pyrazole ring. Following this, 5-amino − 1-(2,6-dichloro-4-(trifluoromethyl)phenyl) -4-hydrosulfinyl-3-(iminomethyl) -1H-pyrazole (4), through the cleavage of the pyrazole ring and removal of the sulfinyl and carboximidamide substituents, is transformed into 2,6-dichloro-4-(trifluoromethyl)aniline (6) by microbial ring cleavage and reductive deamination, where the C–N and C–C bonds of the pyrazole ring are broken and the aniline amino group (–NH₂) is retained on the substituted phenyl ring. (ii) Fipronil (1), through the cleavage of the amino group (–NH₂) at the 5-position of the pyrazole ring, is transformed into 1-(2,6-dichloro-4-(trifluoromethyl)phenyl) -4-((trifluoromethyl)sulfinyl) -1H-pyrazole − 3-carbonitrile (3) (commonly known as desamino Fipronil) via oxidative deamination, in which the C–NH₂ bond is broken and a hydrogen atom (–H) replaces the amino group, resulting in a de-substituted pyrazole ring. Desamino Fipronil, thus formed, can undergo two pathways:(a) Desamino Fipronil, through C–aryl bond cleavage at the N-1 position of the pyrazole ring and simultaneous nucleophilic substitution of the hydrogen at position 5, is transformed into 4-hydrosulfinyl-1H-pyrazole-3-carbonitrile (5) via microbial dearylation and reductive amination, in which the phenyl ring is removed and an amino group (–NH₂) is introduced at the 5-position of the pyrazole. (b) Desamino Fipronil, through cleavage of the pyrazole ring and removal of the sulfinyl and cyano substituents, is transformed into 2,6-dichloro-4-(trifluoromethyl)aniline (6) via microbial ring-opening degradation, where the pyrazole core is completely broken down, leaving the substituted phenyl amine.

In the present study, the formation of the amide derivative of fipronil sulfide, i.e., 1-(2,6-dichloro-4-(trifluoromethyl)phenyl)-3-(iminomethyl)-4-((trifluoromethyl)sulfinyl) -1H-pyrazol-5-amine (2) is in agreement with other research findings, in which a similar fipronil metabolite, fipronil amide, was observed as a major metabolite of bacterial biodegradation [36]. 2,6-dichloro-4-(trifluoromethyl)-aniline (6), another metabolite that emerged in the present investigation, was also encountered in other fipronil biodegradation studies [37]. The present study highlighted the formation of some new unreported metabolites of fipronil degradation, thereby implying a unique fipronil catabolic pathway, followed by the consortium FP-25 used for degradation.

Response Surface Methodology (RSM) is a potent statistical technique used to optimize a process response dependent on many input variables [38]. Moreover, the RSM is less strenuous and time-consuming in comparison to the optimization study using one variable at a time [39]. Among the available RSM designs, the Box–Behnken Design (BBD) facilitates an efficient prediction of the first- and second-order coefficients, as it frequently has a smaller number of design points [40]. It uses the 12 middle edge points and three central points to fit a second-order equation [41].

In the present study, BBD experimental design was used to recognize optimal input factors, temperature (X1) (15–50°C), pH (X2) (4–10), inoculum biomass (X3) (0.10–0.25 g L⁻¹), and fipronil concentration (X4) (50–350 mg L⁻¹) for the best response (Table 1). For consortium FP-25, a second-order polynomial equation was developed based on the evaluation of experimental data by using multiple regression analysis as follows:

Where Y represents the predicted value of fipronil degradation by consortium FP-25, and X1, X2, X3 and X4 are the coded values of the independent variables.

The analysis of variance (ANOVA) revealed that all three variables—pH, temperature, and biomass—exerted a statistically significant influence on the degradation process (Table 2). The regression model developed for FP-25 was evaluated for goodness-of-fit and predictive validity through several statistical metrics. Generally, a higher F-value, lower p-value, higher R², and a lack of fit p-value are desired to have a significant model [42]. The lower p-value (p < 0.0001) and the high F-values (603.26) further suggested the adequacy and reliability of the developed quadratic model for fipronil degradation by FP-25. Furthermore, R² values of 0.998 and a strong correlation between predicted R² and adjusted R² (R² adj) > 0.98 corroborated the adequacy and satisfactory fit of the developed models [43]. The lack of fit p-values, i.e., insignificant (0.2), indicated that the models adequately represented the system without unexplained variation. The signal-to-noise ratio was measured using Adeq Precision [44]. The adequate precision value (67.10199) indicated an adequate signal. These findings align with the previous research in displaying the resilience of RSM in biodegradation modeling and process optimization [45].

The Distribution of Normal plot of residual, residual vs. predicted, and predicted vs. actual further the fitness of model (Fig. 5). The contour plots were employed to visualize the interaction effects of the independent variables (Fig. 6). The prominent interaction between pH and inoculum biomass is propounded by the elliptical nature of the contours. The model predictions were concurrent with experimental values, indicating the best response with 91.92% degradation at pH 7.0, temperature 32.5°C, fipronil concentration of 200 mg L^− 1 and biomass concentration of 0.175 g L⁻¹. Earlier reports have validated the application of RSM in various optimization studies of pesticide degradation [46, 47].

Soil samples were collected from different locations of the Himalayan highland (Fig. 7). The results indicated significant concentrations of fipronil residues with recorded values of 121.1, 156.7, 111.9, 243.5, 244.3, 217.5, 156.9, 178.2, 197.8, 276.7, 287.6, 187.8, 168.8, and 210.5 µg kg⁻¹ for L1, L2, L3, L4, L5, L6, L6, L7, L8, L9, L10, L11, L12, non-sterilized (NS), and sterilized control, respectively (Fig. S1). Notably, its application rate of 0.6–200 g a.i. ha⁻¹ is lower than that of conventional insecticides [48]. Still, fipronil residue levels in all tested soils indicated a significant concentration of fipronil residues, owing to injudicious application in these agricultural regions, thereby affirming the widespread and persistent nature of fipronil contamination in the studied agroecosystems.

Fipronil residues are reported in different environmental matrices, indicating a probable intrusion in the food chain [49]. For instance, effluent collected from municipal treatment plants in the U.S. in 2015–2016 was evaluated for the presence of fipronil residues, and the results showed their presence at a higher concentration (0.3–112.9 ng L⁻¹) than the permissible limit of 11 ng L⁻¹ by the United States Environmental Protection Agency Office of Prevention, 1996 [50]. The presence of fipronil residues, even in low concentrations in soil, may transcend ecological barriers and thus impair morphological and reproductive features in aquatic species [51, 52].

All real contaminated samples, collected from different locations of Garhwal Himalaya, were further subjected to a soil microcosm experiment, analyzing the applicability of the developed consortium for bioremediation. The collected soil samples were augmented with consortium FP-25 and monitored for the presence of fipronil residues at time intervals of 0, 15, 30, and 45 days (Fig. S1). The results revealed that all the analyzed soil samples demonstrated substantial reductions in fipronil concentrations by day 45. The residual fipronil concentration, after 45 days was 9.3247, 18.3339, 10.6305, 29.707, 21.0098, 37.41, 22.5936, 19.2456, 11.4724, 37.6312, 49.7548, 21.9726, 103.6432, and 158.5065 µg kg^− 1 (Fig. S2). The percent degradation achieved was 92.3, 88.3, 90.5, 87.8, 91.4, 82.8, 85.6, 89.2, 94.2, 86.4, 82.7, and 88.3% for L1, L2, L3, L4, L5, L6, L7, L8, L9, L10, L11, and L12, respectively. However, the non-sterilized and sterilized uninoculated controls showed significantly lesser degradation rates (p < 0.05) with values of 38.6 and 24.7%, respectively. The lower degradation rate in control soil might be ascribed to the meager fipronil degradation capacity of the indigenous microbiota under the evaluated conditions. Similarly, non-sterilized and sterilized control showed statistically significant differences (p < 0.05) in fipronil degradation by 45 days of the time period. The higher fipronil degradation in the non-sterilized control observed might be attributed to the native catabolic potential of indigenous microbes present in soil.

In the present study, bioaugmentation of the developed consortium FP-25 escalated the fipronil degradation rate in real contaminated soil of the Himalayan ecosystem, implying the compatibility of the synthetic consortium with native microbiota [53]. Bioaugmentation is a well-known method in which microbes with the required catabolic ability are inoculated in a contaminated environment to achieve bioremediation goals [54]. This approach works by accelerating the rate of the natural bioremediation process [55]. However, most bioaugmentation studies have been engrossed in its implementation in the culture media and in liquid media at unreal environmental concentrations [56, 57]. Nevertheless, the practical applicability of a bioaugmentation approach can only be validated by analyzing its performance in real contaminated environments. In a study, a synthetic microbial consortium was observed to degrade approximately 90% pesticide residues in contaminated agricultural effluents [58]. Comparable outcomes were perceived in current pesticide degradation studies, in which bacterial inoculation appreciably intensified the degradation rate of neonicotinoids and phenylpyrazole insecticides [26, 35]. In another investigation, the augmentation of an artificial bacterial consortium engendered the absolute degradation of profenofos and chlorpyrifos residues in real contaminated soil [59].

The effect of bioaugmentation on the persistence of fipronil can be assessed by investigating its degradation kinetics. In the present study, ln C₀/C against time (days) was plotted to obtain the rate constant value (k), the inverse of the slope of the best-fit straight line. The soil microcosm experiment revealed a k value of 0.046–0.076 day⁻¹, inferring remarkably high degradation in all the bioaugmented experimental setups (Table S2). In addition. An R² value of 0.97–0.99 further confirmed an excellent fit to first-order kinetics. Contrarily, a lower k value of 0.01 and 0.005 days⁻¹ and a strong R² value of 0.95 and 0.99 were observed for the non-sterilized and sterilized controls, respectively. The results connoted that the lowest degradation rate discerned in the sterilized control was most likely due to the abiotic degradation. However, the higher degradation rate in non-sterilized soil was due to the catalytic activity of indigenous microbes and bioaugmented culture. Earlier research work has implied similar results, where fipronil degradation by bioaugmented culture followed first-order kinetics [60, 61].