Cloning and Expression of the Organophosphate Pesticide-Degrading α-β Hydrolase Gene in Plasmid pMK-07 to Confer Cross-Resistance to Antibiotics

Pesticide residual persistence in agriculture soil selectively increases the pesticide-degrading population and transfers the pesticide-degrading gene to other populations, leading to cross-resistance to a wide range of antibiotics. The enzymes that degrade pesticides can also catabolize the antibiotics by inducing changes in the gene or protein structure through induced mutations. The present work focuses on the pesticide-degrading bacteria isolated from an agricultural field that develop cross-resistance to antibiotics. This cross-resistance is developed through catabolic gene clusters present in an extrachromosomal plasmid. A larger plasmid (236.7 Kbp) isolated from Bacillus sp. was sequenced by next-generation sequencing, and important features such as α-β hydrolase, DNA topoisomerase, DNA polymerase III subunit beta, reverse transcriptase, plasmid replication rep X, recombination U, transposase, and S-formylglutathione hydrolase were found in this plasmid. Among these, the α-β hydrolase enzyme is known for the degradation of organophosphate pesticides. The cloning and expression of the α-β hydrolase gene imply nonspecific cleavage of antibiotics through a cross-resistance phenomenon in the host. The docking of α-β hydrolase with a spectrum of antibiotics showed a high G-score against chloramphenicol (−3.793), streptomycin (−2.865), cefotaxime (−5.885), ampicillin (−4.316), and tetracycline (−3.972). This study concludes that continuous exposure to pesticide residues may lead to the emergence of multidrug-resistant strains among the wild microbial flora.


Introduction
Plasmid-borne drug resistance and its persistence among soil bacteria cause great public health hazards. Although other genetic elements such as nonconjugative, mobilizable plasmids contribute to multidrug resistance [1][2][3], bacterial plasmid DNA also confers drug resistance [4,5]. Multidrug resistance is very common among soil bacteria that are exposed to pesticides [6,7]. Moreover, the indiscriminate use of pesticides favours the microbial population that can metabolize those pesticides. Higher exposure to pesticides enables bacteria to produce suitable enzymes for the degradation of pollutants. Consequently, pesticide-metabolizing populations tend to overgrow [8][9][10]. Pesticide-degrading bacteria metabolize the pesticides through hydrolytic cleavage for their carbon and energy sources [11,12]. Degradation 2 BioMed Research International of pesticides may also occur through oxidation, reduction, hydrolysis, peroxidise, or oxygenase mechanisms [13,14].
Enzymes produced by pesticide-degrading bacteria can catabolize drugs and other xenobiotics. Breaking C=C bonds in the drug by nonspecific cleavage leads to cross-resistance to antibiotics [15,16]. This could be an explanation regarding how susceptible soil bacteria become multidrug resistant. A previous study on formaldehyde resistance with Enterobacteriaceae has shown high formaldehyde dehydrogenase activity leading to multidrug resistance through nonspecific activity [17,18]. A study conducted on formaldehyde dehydrogenase enzyme-mediated multidrug resistance provided the notion that cross-resistance is possible via nonspecific degradation. Another study proved that enzymes produced by soil bacteria in pesticide-contaminated soil can degrade other xenobiotics [19][20][21]. Assessing the catabolic properties ofhydrolase, an enzyme-degrading pesticide, would answer whether these enzymes play a role in multidrug resistance. It is very common among soil bacteria to adapt themselves to changing environmental conditions. Evolutionarily, bacteria in a stressed environment tend to produce putative enzymes via induced mutations that can degrade a wide range of xenobiotics [22]. Therefore, the present study aimed to understand the bacterial isolates with an extrachromosomal plasmid carrying pesticide-resistant genes that also confer the cross-resistance of multidrug resistance. In silico analyses were performed to understand the mechanism of cross-resistance via thehydrolase enzyme.

Sample Collection.
Soil samples were collected from up to 5 cm of the upper layer from different pesticides applied to an agriculture field located in the Salem district (11.7794 ∘ N, 78.2034 ∘ E), Tamil Nadu, India. The collected samples were transported to the laboratory and stored at 4 ∘ C until further analysis.

Plasmid Isolation and DNA Sequencing.
Owing to the increased use of the plasmid, the sample was kept for plasmid enrichment by incubating with 1% pesticide (monocrotophos) supplemented medium for two days. The bacterial strains were isolated from the enriched sample, and it was found that Bacillus sp. MK-07 (KU510395.1) was predominant. Plasmid DNA (pMK-07) was isolated by the alkaline lysis method from the isolate [23]. The plasmid was treated with plasmid-safe, adenosine 5-triphosphate-(ATP-) dependent DNase to remove any genomic DNA contamination. The plasmid DNA was processed for library preparation using an Illumina Nextera XT DNA library preparation kit. SnapGene version 4.0.2 software was used to create a plasmid DNA map [24]. The library was sequenced on MiSeq using 2 × 300 bp to generate approximately 1 GB of data. The Draft assemblies of short Illumina sequence reads (2 × 300 MiSeq library) were analysed with a 4200-tape station system, Eurofins Genomics, Bangalore, India (Agilent Technologies, USA) [25].

2.3.
Assembling the Plasmid DNA Sequence. Raw data were processed using Trimmomatic v 0.35 to remove adapter sequences, ambiguous reads (reads with unknown nucleotides "N" larger than 5%), and low-quality sequences (reads with more than 10% quality thresholds (QV) < 20 phred score). Clear sequences with a size of 1,073,566 (2 × 300 bp) highquality reads were retained for further analysis and were used for de novo assembly [2,26].

Gene Prediction and Functional
Annotation. Sequences were predicted using prodigal with default parameters. In total, 225 genes were predicted with an average gene size of 816 bp, while the maximum and minimum sizes of the genes were 15,033 bp and 105 bp, respectively [27,28]. Gene ontology annotations of the predicted [29] genes were determined by the Blast2GO program (https://www.Blast2GO.com). Gene ontology assignments were used to classify the functions of the predicted genes. Functional annotations of the genes were performed using BLASTx, part of the NCBI-Blast-2.3.0 standalone tool. BLASTx was used to find the homologous sequence of genes against NR (nonredundant protein database) within Bacillus cereus (MK-07).

Phylogenetic Distinct Clades and Cloning of -Hydrolase
Gene. The scaffold sample of plasmid (pMK-07) was aligned against plasmids of all the Bacillus species using BlastN. A Newick file was downloaded from the Blast Tree View and plotted further using an interactive tree of life (http://itol.embl.de/upload.cgi). Different parameters were adjusted according to the visualization requirements and were exported [29]. The -hydrolase gene was amplified using a gene-specific primer that was designed by the net primer (Premier Biosoft): -hydrolase MK-FP: ATGGC-TAAAGAAATGTTTGTGC and MK-RP: CGCACTAAC-TACTACTTCTGGT. The polymerase chain reaction mixtures (50 l) contained 10 M of each primer, PCR Invitrogen Master Mix (PCR buffer, 5 U of Taq polymerase, 10 M of BSA and 2 l of DNA). The thermocycling conditions included a denaturation step at 94 ∘ C for 3 min, 34 amplification cycles of 94 ∘ C for 1 min, 57 ∘ C for 30 sec and 72 ∘ C for 1 min, and a final extension step for 8 min using an Eppendorf thermocycler (Eppendorf AG 22331). Electrophoresis was continued for 30 min at 100 V (Tarson electrophoresis unit). The size of the fragment was determined by comparing it with a 1 kb marker (NEB). The gene product was inserted into the pXcm vector using a ligation (Fermentas) enzyme.
The -hydrolase gene was released from the pXcm vector using Bam H1 and Hind III. Expression of -hydrolase in E. coli DH5 was achieved by subcloning it into pET-20. Transformation of recombinant DNA into E.coli/DH5 : pET-20b was carried out by standard methods [30]. Preliminary screening was performed based on the blue-white colonies on x-gal medium, followed by PCR amplification of thehydrolase gene.
The recombinant bacterial strains were cultured overnight. The cells were harvested by centrifugation, andhydrolase was recovered by sonication (10-15 min). Crude enzyme was electrophoresed by slope gel electrophoresis along with marker protein (SERVA) and then analysed. Next, 100 l of crude enzyme was mixed with 1 ml of 30 g/ml chloramphenicol, followed by incubation at 37 ∘ C for 48 hours. After the incubation, the metabolites were purified with twofold ethyl acetate and were evaporated under vacuum conditions. The extracted residues were dissolved in methanol to a volume of 2 ml and were stored at 4 ∘ C until GC-FID analysis. The extract was analysed in an Agilent gas chromatograph (Model 7820A Series USA) equipped with a flame ionization detector [31].
2.5. Docking with Ligands. The Crystal Structure ofhydrolase (PDB ID: 1I6W) was retrieved from the protein data bank, and the ligands were downloaded from PubChem (http://www.ncbi.nlm.nih.gov/pccompound) with a Pub-Chem ID ( Table 1). The ligands were retrieved from the PubChem (http://www.ncbi.nlm.nih.gov/pccompound) database based on a literature survey. These compounds were subjected to ligand preparation by the Ligprep wizard application of the Maestro 9.2. Corrections such as the addition of hydrogen, 2D to 3D conversion, corrected bond lengths and bond angles, low energy structure, stereochemistries, and ring conformation, followed by minimization and optimization in the optimized potential for the liquid simulation force field [32][33][34] were performed. One conformation for each ligand was generated with other parameters used as the default in Maestro 9.2. Protein-ligand binding sites were predicted by the core-attachment based method (COACH) (http://zhanglab.ccmb.med.umich.edu/COACH/) using the meta-server approach. Complementary ligand binding site predictions were achieved using two comparative methods, TM-SITE and S-SITE, which recognize ligand binding templates from the BioLiP protein function database by binding-specific substructure and sequence profile comparisons. Docking was performed using the Glide software package (http://www.schrodinger.com/), which searches for favourable interactions between one or more typically small ligand molecules and a larger receptor molecule, usually a protein. The retrieved structures were subjected to the removal of water up to 5-Å distances, assigning lone pair electron atoms using a protein preparation wizard. The receptor grid was set up and generated to specify the binding pocket where the ligand binds using the receptor grid generation panel. Molecular docking of the prepared protein and ligand was carried out using Glide.

Results and Discussion
Continuous usage and accumulation of pesticide in the agricultural field lead to the development of cross-resistance to antibiotics among soil bacteria. Plasmid DNA (pMK-07) from Bacillus sp. was sequenced and analysed using in silico tools, revealing that the plasmid DNA sequences and their relatedness lead to cross-resistance to pesticide and antibiotics. A triclosan-resistant bacterial population showing resistance to antimicrobial agents [35,36] due to self-transmissible genes that can jump between plasmids and chromosomes [37,38] and the accumulation of multidrug resistance genes in the soil bacterial community through horizontal gene transfer were common among pesticide degraders [39,40]. Thus, these studies proved the phenomenon of cross-resistance in bacteria.
In total, 225 genes were annotated from the plasmid, among which 221 genes found hits in the nucleotide database and four genes did not have a matching sequence in the database. Based on gene ontology annotation, the genes from the plasmid were categorized into three domains: biological process, cellular component, and molecular function. Similar categorization was performed for a plasmid (pNUC and p11601MD) that includes cellular and molecular component genes of a clinical multidrug resistance in S. typhimurium and Campylobacter jejuni strain 11601MD [2,3] (Table 2).

Biological and Molecular Function of Genes.
The genes responsible for the biological process of the bacteria, including the genes for spore formation, germination, sporulationspecific N-acetylmuramoyl-L-alanine amidase, small, acidsoluble spore protein C 5 , and germination protein-Ger (x) C family protein, were found. These genes enable the bacteria to withstand the adverse conditions. The genes that are essential for DNA recombination were also present in the plasmid: site-specific recombinase, resolvase family, Tn1546 resolvase recombination protein U, and integrase (Bacillus cereus) [4]. Thus, the plasmid underwent random recombination with different strains of Bacillus sp. The presence of genes such as thetraG/traD family, Flp pilus assembly protein, and cpaB determines the horizontal gene transfer through conjugation [41,42]. The presence of the ars R regulatory element makes the bacteria sense the presence of metal ions in the surroundings [43] and develop tolerance against the metal ions. The gene rep X present in the plasmid is known for plasmid DNA replication. The presence of the IS3 and IS605 transposase families allows the DNA-mediated recombination and insertion of random sequences in the bacterial genome and extrachromosomal plasmid DNA. The presence of RNA-mediated DNA polymerase (reverse transcriptase)    genes indicated a history of involvement of viral transduction in the route of de novo plasmid generation. DNA sequence analysis of the plasmid (pMK-07) DNA revealed the genes harboured in the novel de novo plasmid pMK-07 ( Figure 2). Previous studies have shown thathydrolase can hydrolyse a wide range of pesticides [44], and glutathione S-transferases (GSTs) were found to hydrolyse DDT [45,46], organochlorine, and organophosphorus insecticides [47,48]. Hydrolases and hypothetical protein existence in the plasmid suggest that it was a degradative plasmid, especially pesticides [49].

Cellular Components.
It was observed that approximately 10% of genes are present in the plasmid codes for membrane components. The bacteria possess an LPXTG anchoring domain and sortase enzyme genes, whose coexistence affirms that the bacteria carrying this pMK-07 plasmid might also be pathogenic [50].

Evolution of Newer Characters.
The increased uses of pesticides in the agricultural field serve as the selection pressure for the evolution of soil microbial flora. The bacteria in the soil tend to develop tolerance by acquiring new genes or plasmids from other bacterial sources by either vertical or horizontal gene transfer. Surprisingly, plasmids carry all the essential genes required for survival under adverse or stressed conditions, a finding that has been confirmed in C. jejuni and E. coli [51]. A similar observation was noted among plant pathogenic Gram-negative bacteria carrying genes essential for their infection in plants [52]. The results from our study agree with those in previous studies.

Cloning of the -Hydrolase
Gene. The -hydrolase gene (700 bp) was cloned into pXcm and was confirmed for their presence by running it on a 1.0% agarose gel. The expression of the gene for the -hydrolase enzyme was also verified by SDS-PAGE, with the protein size corresponding to 45 kDa (Figure 3). The -hydrolase gene from the pXcm vector was then excised and subcloned into pET-20b. After transformation, the bacterial cells were screened on LB agar medium supplemented with ampicillin, IPTG, and X-gal.
Plates showing white colonies (transformants; pET-hydrolase plasmid) were picked and processed for further use.

Nonspecific Degradation of Chloramphenicol.
It was predicted that -hydrolase could degrade chloramphenicol by nonspecific cleavage and break the C=C bond in the ring    structure. Similar phenomena were observed in our study, when the cell lysate was mixed with 30 g/ml of chloramphenicol, the lysate degraded the antibiotic, an observation that was proven through GC-MS analysis (Figure 4). GC-MS analysis revealed the breakdown compounds present in the pET -hydrolase-treated sample (methane, oxybis dichloro, phenol, indole-2-one) ( Figure 5). Based on previous works on the characterization of the catabolic ability of -hydrolase, the side chain of the nucleophilic amino acid residue of the enzyme attacks the electropositive carbon atom of the substrate [53][54][55]. The findings of the present study suggest that this bacterial strain, Bacillus sp. MK-07, which survived all the sublethal concentration of pesticides, potentially has the cross-resistance property to degrade the antibiotic chloramphenicol. The cross-resistance mechanisms may be due to ribosomal gene alteration to evolve cross-resistance [7].

Protein-Ligand Binding Site Prediction.
The -hydrolase ligand binding sites were predicted by COACH. The number of templates as the Cluster size was 69, the confidence score (c-score) was 0.96, and the binding residues were VAL9,   (Table 3).

Protein-Ligand
Interaction. -hydrolase (PDB Id: 1I6W) was docked with antibiotics (chloramphenicol, streptomycin, cefotaxime, ampicillin, and tetracycline) and the pesticide monocrotophos using Glide Maestro 9.2. Identification of the best-fit antibiotic was performed based on the G-score and number of hydrogen bonds involved. A similar study showed that a sublethal concentration of herbicides would result in the development of multidrug resistance among soil bacteria [56]. Because of the toxicity of the pesticide, the bacteria develop resistance, which allows them to adapt to such components [57]. The strong interaction of -hydrolase with chloramphenicol showed a Glide score of −3.793 Kcal/mol. Chloramphenicol interacts with ILE 12, GLY 11, and SER 77 with distances of 2.214Å and 2.176Å, 2.255Å, and 2.430Å, respectively, at the active site of the enzyme (Figure 6. I.a). In the surface view of the -hydrolase-chloramphenicol complex, chloramphenicol is highlighted with green ( Figure 6. I.b). In the 2D interaction of the -hydrolase-chloramphenicol complex, the purple dotted line represents the hydrogen bond with the side chain (Figure 6. I.c and Table 3).
The surface view of -hydrolase with streptomycin showed a Glide score of −2.865 Kcal/mol and interaction of streptomycin with ILE 12, GLY 13, and HIS 76 at the active site (Figure 6. II.a). In the surface view of the -hydrolasestreptomycin complex, streptomycin is highlighted in green ( Figure 6. II.b). In the 2D interaction of the -hydrolasestreptomycin complex, the purple dotted line represents the hydrogen bond with the side chain ( Figure 6. II.c and Table 3).
The interaction formed between -hydrolase and cefotaxime showed a Glide score of −5.885 Kcal/mol. This interaction of cefotaxime at the active site of the enzyme with ASN 18 and with SER 77 revealed distances of 2.077Å and 2.022Å, respectively (Figure 6. III.a). In the surface view of the interaction of the -hydrolase-cefotaxime complex, cefotaxime is highlighted in green (Figure 6. III.b). In the 2D interaction of the -hydrolase-cefotaxime complex, the purple dotted line represents the hydrogen bond with the side chain ( Figure 6. III.c and Table 3).
The interaction formed between -hydrolase and ampicillin showed a Glide score of −4.316 Kcal/mol. This interaction involves two hydroxyl bonds between the hydrogen atom of ampicillin with ASN 18 and an oxygen atom of ampicillin with SER 77 with a distance of 1.987Å and 2.181Å (Figure 6. IV.a). In the surface view of the -hydrolase-ampicillin complex, ampicillin is highlighted in green (Figure 6. IV.b). In the 2D interaction of the -hydrolase-ampicillin complex, the purple dotted line represents the hydrogen bond with the side chain (Figure 6. IV.c and Table 3). Figure 6: Docking of -hydrolase with antibiotics. Interaction of GLY 11 and SER 77 with the OH group. Similarly, ILE 12 binds to NH and OH; II: interaction of ILE 12 with the OH group. HIE 76 binds to NH, and GLY 13 binds to both the NH and OH groups; III: binding interaction of ASN 18 with the OH group and SER 77 with the NH group; IV: interaction of ASN 18 with the NH 3 group. SER 77 binds to the OH group; V: docking interaction of ILE 12 binding to the OH group; VI: docking interaction of ILE 12 binding to the NH group. SER 77 interacts with the OH group.
The interaction formed between -hydrolase and tetracycline showed a Glide score of −3.972 Kcal/mol. This interaction involved ILE 12 hydroxyl bonds between tetracycline and ILE 12 with a distance of 1.826Å (Figure 6. V.a). In the surface view of the -hydrolase-tetracycline complex, tetracycline is highlighted in green (Figure 6. V.b). In the 2D interaction of the -hydrolase-tetracycline complex, the purple dotted line represents the hydrogen bond with the side chain ( Figure 6. V.c and Table 3).
The interaction of -hydrolase with monocrotophos showed a Glide score of −4.464 Kcal/mol. The monocrotophos interacts with ILE 12 and with SER 77 with a distance of 1.865Å and 2.210Å at the active site of the enzyme (Figure 6. VI.a). In the surface view of the -hydrolasemonocrotophos complex, monocrotophos is highlighted in green (Figure 6. VI.b). In the 2D interaction of thehydrolase-monocrotophos complex, the pink dotted line represents the hydrogen bond with the side chain ( Figure 6 VI.c and Table 3). Among these antibiotics, based on the docking scores, it can be concluded that all five antibiotics can be degraded through nonspecific cleavage by -hydrolase.
A similar observation in the present study has proven that the hydrolase enzymes could bind with chloramphenicol and hydrolyse it into a nontoxic substance [58,59]. Excessive pesticide usage resulted in the accumulation of pesticide residues in crops, soils, and the biosphere, creating ecological stress [60,61].

Conclusion
The present work focuses on the pesticide-degrading bacteria isolated from an agricultural field that develop crossresistance to antibiotics. This cross-resistance is developed through catabolic gene clusters present in an extra chromosomal plasmid. It can be concluded from the current study that existence of pesticide-resistant plasmids among soil bacteria can also confer cross-resistance to antibiotics through natural selection exerted by pesticide accumulation in the agriculture field. The enzymes that degrade pesticides can also catabolize the antibiotics by inducing changes in the gene or protein structure through induced mutations. Hence, an alternate way to control pests may pave the way for limiting the emergence of multidrug resistance.

Conflicts of Interest
All the authors declare no conflicts of interest.