The endophytic strain HSTU-ASh6 was isolated from healthy fresh rice plant roots. Two months old rice plants were collected from farmers’ fields near Basherhat (25° 37′ 59.88″ N; 88° 39′ 0.00″ E), Sadar, Dinajpur, Bangladesh. The roots of rice plants were separated from whole plants. To get rid of soil and dust, the root samples were gently washed in distilled water. Following this, twenty root samples were then surface sterilized with 75% ethanol for 3 min and shaken in a 2.5% (w/v) NaOCl solution for 10 min at 120 rpm, which was adopted from Khan et al. (2016) and Das et al. (2022). The root samples underwent additional washings with autoclaved distilled water while they were agitated for 20 min. The roots were rolled on nutrient agar plates, and a 0.1 mL aliquot from the final wash was inoculated into 10 mL of nutrient broth for sterility testing. Samples were discarded if any growth of bacteria was detected in the sterility check (Das et al., 2022). Finally, sterilized rice plant roots were ground using a mortar and pestle and placed in a sterilized test tube (Haque et al., 2015). The squeezed roots were cultured in chlorpyrifos-enriched minimal nutrient media to isolate the chlorpyrifos mineralizing endophytic bacteria, as described (Das et al., 2022). Selected bacteria with prompt growth potentialities were sub-cultured in the same medium to obtain pure colonies, as described previously (Das et al., 2022). The isolate was preserved in 80% (v/v) glycerol at –20°C for further analysis. Biochemical tests, including methyl red, Voges–Proskauer, catalase, KOH string, oxidase, triple sugar iron, citrate utilization, motility indole urease, and urease, and individual sugar fermentation tests such as dextrose, lactose, maltose, and sucrose utilization (Adedayo et al., 2004; Das et al., 2022), were performed to facilitate the identification of the bacterium. In addition, the activities of extracellular cellulase, amylase, protease, and xylanase were determined as described elsewhere (Abdullah-Al-Mamun et al., 2022; Das et al., 2022).

The ACC deaminase activity of the bacterium was determined according to the modified methods of Shaharoona et al. (2006) and Belimov et al. (2015), which measures the amount of α-ketobutyrate produced upon the hydrolysis of ACC. The endophytic bacterium strain Enterobacter sp. strain HSTU-ASh6 was grown in Tryptic Soy Broth medium (TSB) for 18 h at 28°C to determine ACC deaminase activity. The cells were then harvested by centrifugation, washed with 0.1 M Tris–HCl (pH 7.5), and incubated for another 18 h in minimal medium containing 3 mM ACC as the sole source of nitrogen. The bacterial cells were collected by centrifugation and suspended in 5 mL of 0.1 moL L^–1 of Tris–HCl, pH 7.6, and transferred to microcentrifuge tube. The contents of the tubes were centrifuged at 16,000 rpm for 5 min, and supernatant was removed. The pellets were suspended in 2 mL of 0.1 moL L^–1 Tris HCl, pH 8.5. Next, 30 μL of toluene was added to the cell suspension and vortexed for 30 s. After 200 μL of the toluenized cells were placed in a fresh microcentrifuge tube, 20 μL of 0.5 moL L^–1 ACC was added to the suspension, vortexed, and then incubated at 30°C for 15 min. Following the addition of 1 mL of 0.56 moL L^–1 HCl, the mixture was vortexed and centrifuged for 5 min at 13,000 rpm at room temperature. Two mL of the supernatant was vortexed together with 1 mL of 0.56 moL L^–1 HCl. Thereupon, 2 mL of the 2, 4-dinitrophenylhydrazine reagent (0.2% 2, 4-dinitrophenylhydrazine in 2 moL L^–1 HCl) was added to the glass tube, and the contents were vortexed and then incubated at 30°C for 30 min. Following the addition and mixing of 2 mL of 2 moL L^–1 of NaOH, the absorbance of the mixture was measured by using a spectrophotometer at 540 nm. The cell suspension without ACC was used as a negative control and with (NH₄)₂SO₄ (0.2% w/v) as the positive control. The number of μmol of α-ketobutyrate produced by this reaction was determined by comparing the absorbance at 540 nm of a sample to a standard curve of α-ketobutyrate ranging between 10 and 200 μmoL (Honma and Shimomura, 1978; Shaharoona et al., 2006; Belimov et al., 2015).

The phosphate solubilization ability of the isolate was determined qualitatively by streaking a strain on PVK agar media in a Petri plate. The plate was placed at 30°C for 72 h under aerobic conditions. The colonies that could grow and create a zone on the insoluble phosphate media were selected as the phosphate-solubilizing bacteria (Madhusmita et al., 2017).

The nitrogen fixation capability of the endophytic strain HSTU-ASh6 was determined by streaking the strain on completely nitrogen-free Jensen’s media (Sulistiyani and Meliah, 2017). The strain was grown for 5–7 days at 37°C in an incubator under aerobic conditions.

The Enterobacter sp. HSTU-ASh6 effect on tomato plant growth promotion was assessed in two different ways. Firstly, its effect at seed germination stage was performed in laboratory conditions. Secondly, its effect on tomato plant seedlings development to vegetative and reproductive stages was investigated in fields at epiphytic conditions.

Tomato seedlings aged 32 days were collected from a local nursery and planted in the research field using a randomized complete block design method. The experiment was divided into the following five treatments to observe the effect of using different concentrations (30–100%) of urea doses and bacterial strains: T1 (30% urea with bacterium), T2 (70% urea with bacterium), T3 (100% urea with bacterium), Tc (100% urea without bacterium), and T0 (no urea and no bacterium/control). The experimental unit area (plot) was 1.80 m² and consisted of four rows measuring 1.32 m in width and 1.37 m in length, and the seedlings were planted using a hand drill method, maintaining a row-to-row distance of 40 cm and a plant-to-plant distance of 10 cm in a line. The plants were irrigated regularly according to requirement. The treatments were applied to the plants after 10, 20, 30, and 40 days of plantation. The root length, shoot length, leaf size, fresh weight, and dry weight were measured two times after 30 and 45 days of plantation.

Firstly, the growth of the strain was checked in chlorpyrifos-enriched minimal salt broth media, in which chlorpyrifos (1 g/L) was the sole source of carbon (Lee et al., 2021; Das et al., 2022). Next, 5 μL of the stock bacterium was inoculated into the medium and kept for 14 days, after which 5 mL of chlorpyrifos-containing MSM was transferred into a separating funnel. Secondly, 25 mL of distilled water and 5 mL of n-hexane were added to the funnel and shaken vigorously for 5–10 min. The desired solvents in the n-hexane layer appeared on the top layer, whereas the unwanted layer remained on the bottom layer. The bottom layer was separated and kept in a bottle. Finally, to examine the biodegradation of chlorpyrifos, the n-hexane with appropriate solvent-containing layer solution was run on Gas Chromatography – Tandem Mass Spectrometry (GC–MS/MS) (Shimadzu QP2010, Japan). The GC–MS/MS instrumentation was set up according to Al Mansur et al. (2018), and following examination of each compound’s mass spectrum were determined using the NIST11 library (Zaman et al., 2021).

The preparation of the genomic library of Enterobacter sp. HSTU-ASh6 and the sequencing of the complete genome were performed using the Illumina MiniSeq System (Illumina, Inc., San Diego, CA, USA), as described previously (Abdullah-Al-Mamun et al., 2022; Das et al., 2022). After complete genome sequencing, the quality test was conducted using the fast QC analysis and Illumina Base Space sequence analysis Hub. SPAdes assembler version 3.5 was used to assemble the complete genome of Enterobacter sp. HSTU-ASh6, and the resulting assembled genome was annotated and analyzed using NCBI Prokaryotic Genome Annotation Pipeline (PGAP) version 4.5, Prokka. The circular genome map of Enterobacter sp. HSTU-ASh6 was prepared using CGView Comparison Tool. Next, different types of genes, e.g., those responsible for plant growth promotion (nitrogen fixation, phosphate solubilization, IAA production, ACC deaminase production, and sulfur assimilation), biofilm formation, root colonization (chemotaxis and mortality), pesticide degradation, and stress tolerance (heat shock and cold shock), were predicted from the PGAP file.

A total of 105 various types of organophosphorus pesticides, weed killers, and nerve agent small molecules were collected from the PubChem database³. After ordering, the ligands were optimized and minimized using the mmff94 force field and the steepest descent algorithm in the PyRx software environment. The 3D structures of small molecules were stored in a file in .sdf format.

A total of 105 pesticides (small molecules) with 28 valid protein structures were used individually and screened with 105 ligands for virtual screening. Virtual screening and molecular docking were performed using the PyRx software (Dallakyan and Olson, 2015). Protein–ligand complexes were constructed using the Chimera software, and 2D and 3D structure visualization was performed using Discover Studio 2021 Client. The results of virtual screening were displayed by boxplot analysis performed using the Origin Pro 8.0 software. Furthermore, protein and ligand non-bonded interactions were recognized using Discovery Studio 2021 Client. At the time of virtual screening, the grid box sizes were fixed at 61.6411, 61.6072, and 61.8787, which was the maximum size of the grid box, respectively for X, Y, and Z. However, in the case of molecular docking, the grid box sizes were not fixed, and they were dissimilar as for (i) α/β fold hydrolase (GM298_18675) with crotoxyphos, (ii) carboxylesterase with coumaphos, (iii) α/β fold hydrolase (GM298_08590) with cypermethrin, (iv) α/β fold hydrolase (GM298_00815) protein with diazinon, (v) amidohydrolase family protein (GM298_20245) with chlorpyrifos, and the X, Y, and Z sizes were 61.6411, 61.6072, and 61.8787, respectively. The binding affinity of protein with ligands was calculated in kcal/mole for a negative score.

To determine the potential pesticide-degrading stability of proteins with pesticides, a molecular dynamics simulation was run using the Desmond program (Schrodinger) that has an explicit solvent MD package linked to the OPLS 2005 force field. The protein preparation wizard was used to create the protein-ligand complexes, and the system builder panel was solvated them using the orthorhombic simple point-charge water model. The counter ions Na+ or Cl^– were added as necessary to charge and neutralize the solvated system. In order to relax the system to the lowest local energy, minimization tasks were carried out. Next, this model system was run to MD simulation steps using the OPLS 2005 force field parameter (Lee et al., 2021; Patel et al., 2021). The MD simulation was performed up to 100 ns using the NPT ensemble (isothermal-isobaric ensemble, constant temperature, constant pressure, and constant number of particles) at 300 K temperature and 1.013 bar pressure (Jorgensen et al., 1996; Lee et al., 2021). For controlling temperature and pressure, the Noose-Hover chain thermostat algorithm and the Martyna-Tobias-Klein barostat algorithm were employed, respectively. The particle mesh Ewald method was used to measure long-range electrostatic interactions, while the other parameters were left at their default values (Patel et al., 2021). The behavior and interactions between the ligands and protein were examined using the Simulation Interaction Diagram tool. The output.cms file was imported, and root mean square deviation (RMSD) and root mean square fluctuation (RMSF) was chosen to produce the revealed plots.

The biochemical characterization of HSTU-ASh6 is summarized in Supplementary Figure 1. The strain was gram-negative; KOH dissolves the thin peptidoglycan layer of the cell wall of gram-negative bacteria, and they showed viscously and stuck in the loop within 30 s after the addition of KOH to the bacteria cultured for 24 h. Conversely, the isolate demonstrated positive results in the Voges–Proskauer, catalase, triple sugar iron, citrate utilization, motility indole urease, urease, dextrose, lactose, maltose, and sucrose utilization tests. However, the strain showed negative results in the indole, MR, and oxidase tests. It also exhibited mild activities of cellulase, amylase, and xylanase but a vigorous protease activity in the plate assay.

The results of phosphate solubilization, nitrogen fixation, and IAA production are summarized in Supplementary Figure 2. The phosphate solubilization test was positive because growth was detected and a hollow zone appeared in the PVK plate. The nitrogen fixation test was also positive based on the detection of the growth of HSTU-ASh6 in Jensen’s medium. Moreover, the strain was capable of producing IAA as demonstrated by the reddish pink color, which amounted to 4.8 mg/mL compared to that in the control. In addition, the ACC deaminase activity of the strain was 0.0076 μg/mL/h.

The GC–MS/MS analysis spectra revealed the presence of phorate sulfoxide, phorate sulfone, chlorpyrifos methyl, carbophenothion sulfoxide, oxydisulfoton, carbonochloridic acid, 2,4,5-trichlorophenyl ester, thionodemeton sulfone, 3-(2-thienyl)-DL-alanine, chlorpyrifos oxon, diethyl methanephosphonate (DEMP), and 3,5,6-trichloro-2-pyridinol (TCP), etc. (Table 3). On the basis of GC–MS/MS results, a pathway of chlorpyrifos biodegradation by HSTU-ASh6 was proposed as illustrated in Figure 1. Chlorpyrifos was initially degraded by hydrolysis to generate chlorpyrifos oxon that was further broken down to generate TCP and DEMP (spectrum = 8). Subsequently, TCP was again degraded by ring breakage, resulting in its complete detoxification.

The genome size of Enterobacter sp. HSTU-ASh6 was 4,857,424 bp with 55.1% GC, 4478 protein-coding sequences (CDS), and 72 tRNAs. The subsystem of the RAST annotation is shown in Supplementary Figure 4. All proteins of Enterobacter sp. HSTU-ASh6 were revealed by CGview analysis in a proper genomic map (Figure 2). The genomic map of proteins from Enterobacter sp. HSTU-ASh6 was created successfully, where GC-skew positivity indicates the presence of CDS downstream and GC-skew negativity shows the presence of CDS upstream. Although GC-skew is predominantly introduced by RNA synthesis in local genomic areas, it was first used to estimate the ori and ter positions computationally by looking at known genome sequences.

The genome of Enterobacter sp. HSTU-ASh6 harbors various types of PGP genes, namely those encoding nitrogen fixation (iscU, nifJ, iscA, iscR, iscS, iscX, sufABCDES, and fdx), nitrosative stress (ntrB, norR, norV, nsrR, and glnK), nitrogen metabolism regulation (glnD, glnB, and ptsN), ammonia assimilation (gltB), ACC deaminase (dcyD and rimM), siderophore enterobactin (fes, entFSD, fhuABCD, and tonB), IAA production (trpCFBDS and ipdC), phosphate metabolism (pitA, pstABCS, phoU, ugpABE, phoABERHQ, and pntAB), biofilm formation (tomB, luxS, efp, flgABCDGHIJKLMN, motAB, and hfq), sulfur assimilation (cysACHIJKMNTWZ), sulfur metabolism (fdxH and cysACDEHIJKMNQSWZ), root colonization (cheZYBRWA, malE, rbsB, fliZDSTFZGHIJKMPQR, hofC, pgaABCD, motAB, and murJ), superoxide dismutase (sodABC), and trehalose metabolism (treBCR, otsAB, and lamb) (Table 5). Furthermore, numerous types of genes associated with stress tolerance e.g., heat shock (smpB, ibpAB, hspQ, dnaJK, rpoH, lepA, and grpE), cold shock (cspADE), and drought resistance (nhaA, chaAB, proABQVWXPS, betABT, trkAH, and kdpABCF) were detected in the HSTU-ASh6 genome (Table 5). These genes are essential for the plant to survive and grow in a harsh environment.

A significant number of genes whose products were responsible for pesticide degradation (ampD, glpABQ, pdeHR, pepABDQ, phnFDGHJKLMOP, paaC, hpxKW, amidohydrolase, and alpha/beta fold hydrolase) were identified in the genome of HSTU-ASh6 (Supplementary Table 2).

For model validation, 39 modeled proteins were analyzed using the SAVESv6.0-structure validation server, among which 28 were validated according to VERIFY (3D-1D score), with their score recorded as 80.05–97.97%. Among these, alpha/beta fold hydrolase family protein (GM298_09990) had the maximum ERRAT quality score of 97.35 (Table 6). The individual protein average TM score, RMSD, IDEN, and Cov were 0.95, 0.75, 0.51, and 0.97, respectively, based on the I-TASSER result (Table 6). Moreover, the Ramachandran plot regions, including the most favored regions (A, B, L), additional allowed regions (a, b, l, p), generously allowed regions (∼, ∼b, ∼l, ∼p), and disallowed regions of the model proteins, are presented in Table 6. Consequently, the 28 validated model proteins were selected for virtual screening with 105 small molecules.

A total of 105 different pesticides, including weed killers and organophosphorus nerve agents, were screened with the 28 validated model proteins of Enterobacter sp. HSTU-ASh6 (Figure 5). The score ranged from –8.8 to –3.1 (kcal/mol). The highest negative score was recorded for cypermethrin with phnL (–8.8 kcal/mol), and the lowest negative score was found for demephion-O with amidohydrolase (GM298_10355) (–3.1 kcal/mol). The box plot analysis revealed that each box with the bold line represents median values, and the square sign indicates mean values. Moreover, the upper and lower lines in the box plots represent minimum and maximum values, respectively. Interestingly, most values were found between the 25th and 75th percentile. The 25th and 75th percentile quartile differences indicated the interquartile range. Surprisingly, in this dataset, some outlier data were observed, indicating the presence of some small molecules possessing an excellent binding affinity with respective proteins. It was also observed that the nine proteins’ outlier binding affinities values were outside of IQR1.5.

As shown in Figure 6, the alpha/beta fold hydrolase (GM298_18675) of Enterobacter sp. HSTU-ASh6 anchored with crotoxyphos and formed a potential catalytic triad in the binding pocket region with residues Ser153-His277-Asp152. Ser153 and Gly81 specifically mediated the attractive charge connection with the +P-atom of crotoxyphos and the typical H-bond contact with the O atom (Figure 6A). Besides other interactions, Val201 and Phe14 directly interacted with the phosphodiester bond of crotoxyphos via π-sigma and π-π stacked interaction. The interaction distances among the residues of the catalytic site were recorded within < 3 Å, except for Asp152, which was 5.36 Å.

Surprisingly, the residues Ser195, His400, and Glu194 constituted a putative catalytic triad in the binding pocket region of the Enterobacter sp. HSTU-ASh6 carboxylesterase. Ser195 and Arg199 interacted with the O atom to form the traditional H-bond, and Glu194 interacted with the +P atom to form phosphodiester bonds and the H atom to interact with the carbon-hydrogen bond of coumaphos (Figure 6B). Moreover, the π-π stacked bond was formed by the His400 residue of carboxylesterase with coumaphos ligand, and it directly interacted with the phosphodiester bond of ligand. The interaction distances among the residues of the catalytic site were recorded within <4 Å.

The alpha/beta fold hydrolase (GM298_08590) formed a potential catalytic triad Ser157-His279-Asp156 in the binding pocket region with cypermethrin (Figure 6C). The interaction distances among the catalytic site residues were recorded within <5.25 Å. Furthermore, the alpha/beta fold hydrolase (GM298_00815) protein–diazinon docked complex demonstrated interaction with multiple residues (Figure 6D). In particular, conventional H-bonds were made by Ser99, His231 with the O atom of the diazinon compound. In addition, the amidohydrolase family protein (GM298_20245) anchored with chlorpyrifos and showed a significant interaction (Figure 6E). For instance, Arg408 interacted with the O atom of chlorpyrifos by a conventional hydrogen bond. Moreover, alkyl and pi-alkyl bonds were made by Val547, Arg408, Ala550, Lys549, His250, His467, and Tyr545 with chlorpyrifos, respectively. Interestingly, all of them were observed with two bonds, except His467 and Tyr545. Only one amino acid made the attractive charge and was interconnected with +P-atom. His250 made pi-sulfur and pi-alkyl bond at a time.

The RMSD plot of the α/β hydrolase (GM298_18675)–crotoxyphos complex was sharply increased from 0 to 4 ns (Figure 7Aa), and it continued up to 30 ns with an RMSD ranging from 2.6 to 3.45 Å. After a slight decrease in the RMSD of the protein, at 35 ns, an increasing parallel trend continued up to 100 ns, with an RMSD ranging from 2.5 to 3.35 Å, except a slight decrease at 83–86 ns of the ligand. In the RMSF plot of the alpha/beta hydrolase (GM298_18675)–crotoxyphos complex (Figure 7Ab), the favorable residues of alpha/beta hydrolase (GM298_18675) with crotoxyphos were observed at Leu10 (0.93 Å), Phe14 (1.161 Å), Lys19 (2.806 Å), Ser21 (3.243 Å), Ser22 (3.639 Å), Arg30 (0.798 Å), Gly33 (0.714 Å), Thr34 (0.636 Å), Ser37 (0.842 Å), Leu40 (0.975 Å), Gly81 (0.43 Å), Cys82 (0.416 Å), Ser85 (0.449 Å), Asp152 (0.378 Å), Tyr181 (0.425 Å), Val201 (0.70 Å), Ile202 (0.659 Å), Thr206 (0.644 Å), Leu207 (0.51 Å), Ile276 (0.542 Å), His277 (0.493 Å), Gly278 (0.456 Å), Gln281 (0.452 Å), Leu282 (0.539 Å), and Ile285 (1.005 Å). The Rg of the α/β hydrolase (GM298_18675)–crotoxyphos complex fluctuated throughout the simulation, with an average value of 3.960 Å. The MolSA, solvent-accessible surface area (SASA), and PSA of the alpha/beta hydrolase (GM298_18675)–crotoxyphos complex insignificantly varied throughout the simulation period. The average MolSA, SASA, and PSA were 310.304, 56.5454, and 81.3709 Ų, respectively (Figure 7Ac).

The RMSD plot (Figure 7Ba) of the carboxylesterase–coumaphos complex was sharply increased from 0 to 3 ns, and it continued up to 23 ns, with an RMSD ranging from 2.3 to 3.45 Å. After a slight decrease in the RMSD of the protein, at 35 ns, a parallel increasing trend continued up to 100 ns, with an RMSD ranging from 2.5 to 3.80 Å, except a slight decrease at 83–86 ns of the ligand. In the RMSF plot of the carboxylesterase–coumaphos complex (Figure 7Bb), the favorable residues of carboxylesterase with coumaphos were observed at Glu69 (1.491 Å), Thr73 (1.723 Å), Gly75 (1.948 Å), Gly76 (1.807 Å), Trp106 (0.456 Å), His108 (0.787 Å), Ile114 (0.918 Å), Leu119 (0.923 Å), Pro121 (0.736 Å), Tyr122 (0.66 Å), Phe192 (0.437 Å), Ser195 (0.64 Å), Ala196 (0.829 Å), Arg199 (0.71 Å), Ser221 (0.717 Å), Thr224 (2.112 Å), Leu225 (1.79 Å), Glu308 (0.59 Å), Val311 (0.716 Å), Met312 (0.884 Å), Val314 (1.919 Å), Phe315 (1.881 Å), Ile317 (1.698 Å), Ala359 (0.803 Å), Phe360 (0.643 Å), His391 (1.243 Å), Trp399 (0.976 Å), His400 (0.734 Å), and Val404 (0.761 Å). The Rg of the carboxylesterase–coumaphos complex fluctuated throughout the simulation, with an average value of 4.147 Å. The MolSA, SASA, and PSA of the carboxylesterase–coumaphos complex insignificantly varied throughout the simulation period. The average MolSA, SASA, and PSA were 317.102, 38.885, and 82.488 Ų, respectively (Figure 7Bc).

The RMSD plot of the α/β hydrolase (GM298_08590)–cypermethrin complex was sharply increased from 0 to 3 ns (Figure 7Ca), and it continued up to 100 ns with the exception of a few irregularities. The overall RMSD difference ranged from 3.5 to 4.01 Å. In the RMSF plot of the alpha/beta hydrolase (GM298_08590)–cypermethrin complex (Figure 7Cb), the favorable residues of alpha/beta hydrolase (GM298_08590) with cypermethrin were observed at Ala17 (3.112 Å), Pro18 (2.65 Å), Leu26 (0.956 Å), Ala29 (1.522 Å), Thr30 (1.435 Å), Leu34 (1.069 Å), Leu37 (1.135 Å), His83 (0.858 Å), Gly85 (1.275 Å), Gly86 (1.409 Å), Trp87 (1.15 Å), Cys88 (1.027 Å), Leu89 (0.969 Å), Asp156 (0.639 Å), Ser157 (0.688 Å), Ala158 (0.771 Å), Gly159 (0.63 Å), Tyr185 (0.628 Å), Ala187 (0.845 Å), Phe205 (0.93 Å), Leu206 (1.228 Å), Asp209 (0.995 Å), Ala210 (0.974 Å), Met211 (0.878 Å), Phe213 (0.867 Å), Cys214 (0.883 Å), Thr215 (0.929 Å), Leu251 (0.61 Å), His279 (0.561 Å), Ala280 (0.652 Å), and His283 (0.856 Å). The Rg of the alpha/beta hydrolase (GM298_08590)–cypermethrin complex fluctuated throughout the simulation, with an average value of 4.512 Å. The MolSA, SASA, and PSA of the alpha/beta hydrolase (GM298_08590)–cypermethrin complex insignificantly varied throughout the simulation period. The average MolSA, SASA, and PSA were 378.775, 26.194, and 93.0372 Ų, respectively (Figure 7Cc).

The RMSD plot of the α/β hydrolase (GM298_00815)–diazinon complex was sharply increased from 0 to 17 ns (Figure 7Da), and it continued up to 75 ns. Next, a slight parallel downward trend appeared from 75 to 100 ns. The overall RMSD difference ranged from 1.2 to 1.7 Å. In the RMSF plot of the alpha/beta fold hydrolase (GM298_00815)–diazinon complex (Figure 7Db), the favorable residues of alpha/beta fold hydrolase (GM298_00815) with diazinon were observed at Gly30 (0.689 Å), Leu31 (0.553 Å), Gly32 (0.767 Å), Cys33 (0.934 Å), Ala34 (0.947 Å), Ala35 (1.37 Å), Ser99 (0.656 Å), Met100 (0.76 Å), Ser103 (0.906 Å), Glu123 (0.66 Å), Pro124 (0.801 Å), Asn125 (1.033 Å), His127 (1.774 Å), Met132 (1.439 Å), Phe 133 (1.301 Å), Ser136 (1.316 Å), Ile137 (0.994 Å), Phe145 (1.217 Å), Gln148 (1.277 Å), Gly149 (1.105 Å), Asp151 (1.429 Å), Met153 (1.143 Å), Val181 (0.822Å), Val184 (0.792 Å), Ser207 (1.098 Å), Leu208 (1.684 Å), and His231 (1.028 Å). The Rg of the alpha/beta fold hydrolase (GM298_00815)–diazinon complex fluctuated throughout the simulation, with an average value of 3.29 Å. The MolSA, SASA, and PSA of the alpha/beta fold hydrolase (GM298_00815)–diazinon complex insignificantly varied throughout the simulation period. The average MolSA, SASA, and PSA were 304.144, 88.6872, and 33.733 Ų, respectively (Figure 7Dc).

The RMSD plot of the amidohydrolase (GM298_20245)–chlorpyrifos complex (Figure 7Ea) showed an increasing trend for the ligand protein from 0 to 14 ns. After 17 ns, an increasing parallel trend continued up to 44 ns, with an RMSD ranging from 2.5 to 3.5 Å. Next, a steady and stable RMSD remained up to 80 ns. Then, the RMSD difference was below ∼1.2 Å up to 93 ns, which indicated the stable protein–ligand complex. In the RMSF plot of the amidohydrolase–chlorpyrifos complex (Figure 7Eb), the favorable residues of amidohydrolase with chlorpyrifos were observed at His81 (0.826 Å), Ser84 (0.99 Å), His109 (0.872 Å), Glu110 (0.888 Å), Glu174 (0.906 Å), Met176 (1.848 Å), Arg207 (1.4 Å), Asp276 (1.373 Å), Asp277 (1.118 Å), Ala407 (1.276 Å), Arg408 (1.631 Å), Tyr465 (0.683 Å), Ser466 (1.006 Å), His467 (1.215 Å), Asp468 (0.95 Å), Lys549 (1.393 Å), Ala550 (1.131 Å), Glu552 (1.336 Å), Gly553 (1.338 Å), and Leu556 (1.327 Å). The Rg of the amidohydrolase–chlorpyrifos complex fluctuated throughout the simulation, with an average value of 3.48 Å. The MolSA, SASA, and PSA of the amidohydrolase–chlorpyrifos complex insignificantly varied throughout the simulation period. The average MolSA, SASA, and PSA were 278.25, 156.69, and 37.03 Ų, respectively (Figure 7Ec).