Introduction

Organophosphate (OP) pesticides are a group of chemicals (phosphoric acid esters) broadly used as insecticides to control pests in agriculture and household due to their wide action spectrum and high effectiveness (Karasali and Maragou, 2016; Sharma et al., 2020). These pesticides cause the irreversible inactivation of the acetylcholinesterase activity (AChE, EC 3.1.1.7), essential for nervous action in insects and vertebrates. Therefore, these chemicals are classified as highly toxic. As a consequence of the intensive usage of OP pesticides, several environmental pollution events in agricultural soils and water bodies have been reported (Fosu-Mensah et al., 2016; Okoli et al., 2017). The environmental accumulation of such toxic chemicals is recognized as a menace for fauna (Siddiqa et al., 2016; Dorneles et al., 2017; Poirier et al., 2018) and human health (Upadhayay et al., 2020; Nicolopoulou-Stamati et al., 2016; Jokanović, 2018).

However, even though the use of OP pesticides has been banned in several countries due to their toxicity, some of them, especially parathion and methyl parathion (MP), continue to be widely used for pest control in crops like cotton, rice, bean, and corn, among others (Urióstegui-Acosta et al., 2020). Due to its high toxicity to mammals and the environmental concern associated with the accumulation of OP pesticide wastes in soil, water supplies, and food, it is necessary to develop reliable and economic strategies for OP pesticides detoxification, and elimination (Mulla et al., 2020; Pailan et al., 2020). Microbial biodegradation of pesticides has received particular attention; different bacterial strains capable of biodegrading OP pesticides have been isolated from agricultural soils from different regions worldwide.

Bacterial biodegradation of OP pesticides is related to the activity of enzymes that catalyze the hydrolysis of phosphoester bonds (Bhatt et al., 2021). Among the most studied OP pesticide-hydrolyzing enzymes are phosphotriesterases, as the organophosphorus hydrolase (OPH) encoded in the opd (organophosphate-degrading) gene in Flavobacterium ATCC 27551 (Mulbry and Karns, 1989), the coroxon hydrolase (HOCA) coded by the gene hocA (coroxon hydrolysis) in Pseudomonas moteilli (Horne et al., 2002), and methyl parathion hydrolase (MPH) coded by the gene mpd (methyl parathion-degrading) in Plesiomonas sp. strain M6 (Zhongli et al., 2001).

Most of the bacteria capable of hydrolyzing MP express the enzyme methyl parathion hydrolase (MPH EC 3.1.8.1.), a triesterase coded in the mpd gene. This enzyme catalyzes the hydrolysis of the phosphotriester bond of the MP, releasing p-Nitrophenol (PNP) and dimethyl-thiophosphoric acid. These metabolites are not toxic to the nervous system because they are unable to inhibit the AChE activity (Siddavattam et al., 2003; Pakala et al., 2007; Popoca-Ursino et al., 2017). However, PNP is a chemical compound considered an environmental pollutant due to its high toxicity over soil microbiota (Megharaj et al., 1991; Subashchandrabose et al., 2012).

Several bacteria can hydrolyze MP (Kulkarni and Chaudhari, 2006; Zhang et al., 2009a; Zheng et al., 2009; Min et al., 2017; Zhang et al., 2018). However, reports on bacteria capable of efficiently degrading PNP are scarce. Bacterial-mediated degradation of PNP has been studied, reporting two metabolic pathways. The first, denominated as the hydroquinone pathway (HQ), has been reported in Gram-negative bacteria as Moraxella sp. (Spain and Gibson 1991), while the second, the hydroxyquinol or benzenetriol pathway (BT), has been reported in Gram-positive bacteria as Arthrobacter sp. (Jain et al., 1994) and Rhodococcus opacus SAO101 (Kitagawa et al., 2004). The presence of both PNP biodegradation pathways has also been reported in some Gram-negative bacteria, as Burkholderia sp. SJ98 (Vikram et al., 2013), Burkholderia zhejiangensis CEIB S4-3 (Castrejón-Godínez et al., 2019), and Pseudomonas sp. WBC-3 (Zhang et al., 2009b). In these strains, the catabolic genes that encode the enzymes related to PNP degradation (monooxygenases, dioxygenases, reductases, and dehydrogenases) are organized as a gene cluster.

The bacterial strain Burkholderia cenocepacia CEIB S5-2 was characterized with the capability of hydrolyzing MP and degrading PNP completely (Fernández-López et al., 2017). The analysis of its draft genome reveals the presence of a cluster of genes (pnpABA’E1E2FDC) related to both known degrading PNP pathways (Martínez-Ocampo et al., 2016). Evaluation of gene expression levels and transcriptional regulation in B. cenocepacia CEIB S5-2 may conduct a better understanding of the bacterial response to the presence of both compounds MP and PNP and the biological processes involved in their biodegradation. The present study aimed to evaluate the transcriptional changes generated by the exposure to MP on the bacterial strain B. cenocepacia CEIB S5-2, as well as the differential expression of genes related to the hydrolysis of MP and the biodegradation of PNP, through the transcriptome analysis generated by the next-generation sequencing technology RNA-seq.

Materials and methods

Bacterial strain

B. cenocepacia CEIB S5-2 was previously isolated from agricultural soils from the state of Morelos, Mexico. This strain can efficiently hydrolyze the pesticide MP and degrade its main hydrolysis product PNP (Fernández-López et al., 2017); its genome has a total length of around 8,976,170 bp and a G + C content of 65.7%. The PNP degradation profile of this strain was correlated to the presence of a catabolic cluster denominated pnpABAʼE1E2FDC (Martínez-Ocampo et al., 2016).

Chemical reagents

Methyl parathion (O,O-dimethyl O-4-nitrophenyl phosphorothioate) analytical grade (99% purity) was purchased from Chem Service Inc. (West Chester, PA, USA). Pesticide working solution (10,000 mg/L) used in all experiments was prepared using HPLC-grade methanol (Merck, Darmstadt, DEU) as solvent.

Culture media, pre-inoculum, and bacterial growth conditions

For the maintenance of the strain and the production of the pre-inoculum of the B. cenocepacia CEIB S5-2 for subsequent transcriptomic experiments, three 250-mL independent cultures were incubated (30 °C, 150 rpm) for 12 h using tripticasein soy broth (Bioxon, Becton Dickinson, Mexico State, Mexico), to reach an optical density (OD600 nm) of 1.0. Subsequently, the cell pellet from each culture was collected by centrifugation at 10,000 rpm (4 °C, 10 min) and washed three times with mineral salts medium (MSM), with the following composition in g/L, 2.92 of K2HPO4; 0.20 of KH2PO4; 2.0 of MgSO4·7H2O; 0.99 of KNO3; and 0.99 of (NH4)2SO4; and after sterilization, 2 mL of trace elements solution (pH 7.00 ± 0.05) composed in g/L by 2.8 of H3BO3; 2.55 of MnSO4·H2O; 0.20 of CuSO4·5H2O; 2.43 of CoCl2·6H2O; and 0.25 of ZnSO4·7H2O. Finally, the cell pellets were resuspended in 5 mL of MSM, and the resulting cellular suspension was used for the inoculation of the experimental cultures.

Growth kinetics in the presence of MP were performed to establish adequate sampling points for RNA extraction for subsequent transcriptome sequencing (Fig. 1). Sampling times were selected according to the following considerations; the sampling time 0 h was selected due to the fast MP hydrolysis observed at the beginning of the growth kinetics. The sampling time of 2 h corresponded to the highest PNP concentration observed over the growth kinetics. Finally, the sampling time of 5 h was selected due to the important decreasing in the PNP concentration observed compared to the sampling time of 2 h, thus indicating the biodegradation of this metabolite by the strain.

Fig. 1
figure 1

Hydrolysis of MP and degradation of PNP by B. cenocepacia S5-2

Full size image

In the experimental cultures, B. cenocepacia CEIB S5-2 (OD600nm = 0.5) was grown in 250 mL Erlenmeyer flasks (30 °C, 120 rpm, 5 h) containing 50 mL of previously sterilized MSM according to Fernández-López et al. (2017). Eighteen culture flasks were inoculated with the strain, nine culture flask were supplemented with 50 mg/L of MP (pesticide exposed condition), and nine without the pesticide (control condition). In the cultures of both experimental conditions, bacterial growth was followed through the OD at 600 nm record, while the release of PNP, resulting from the MP hydrolysis, was followed through the absorbance at 410 nm. At each sampling time (0, 2, and 5 h), the cells of three culture flask were collected by centrifugation (3500 rpm, 4 °C, 10 min) in the presence of 400 μL of RNAlater® (Ambion, Foster City, CA, USA) solution. The obtained cellular pellets were stored at − 70 °C for subsequent ARN extraction.

RNA extraction

Total RNA extraction of each sample was performed according to the TRIzol-chloroform method, following the TRI Reagent® Kit (Sigma-Aldrich, Sheboygan Falls, WI, USA) protocol. Per each sample, an amount of 30 μg of total RNA was processed with 5 μL of DNaseI enzyme solution (Thermo Scientific, Waltham, MA, USA) to eliminate DNA debris remaining in the samples. Subsequently, samples were cleaned and concentrated using “RNA Clean and ConcentratorTM” columns (Zymo Research, Irvine, CA, USA). Finally, the purified total RNA concentration from each sample (5 μL) was determined according to its absorbance at 260 nm in a Nanodrop 200c (Thermo Scientific, Waltham, MA, USA), while the RNA integrity was evaluated by electrophoresis in 1% agarose gels. Subsequently, the samples were sent to the GENEWIZ company (https://www.genewiz.com/) (South Plainfield, NJ, USA) for quality evaluation, ribosomal RNA elimination, construction of the cDNA libraries, and transcriptome sequencing through the Illumina HiSeq 2 × 150 bp 2000 System (RNA-seq). The quality and concentration of the samples were evaluated by fluorescence (QubitTM), and the integrity of the samples was evaluated by capillary electrophoresis (Agilent 2100 Bioanalyzer). In this method, the resulting electropherogram provides the RNA integrity number (RIN), a numerical index ranging from 1 to 10, where the minimum acceptable value is 8 and a value of 10 indicates that the RNA sample has the highest integrity (Schroeder et al., 2006).

Analysis and mapping of RNA-seq data

The RNA-seq transcriptomic data were processed using the FastQC software (Babraham, Bioinformatics, UK) to get the quality values of the sequences. Subsequently, sequences were processed in the software Cutadapt 2.9 (Martin, 2011) to eliminate the adapters and carried out the low-quality sequences filtering. After processing, the sequences were aligned against the B. cenocepacia CEIB S5-2 draft genome (GenBank, LNCR00000000), as the reference genome, using the CLC Genomics Workbench v. 10.0.1 server (Qiagen, Hilden, DEU) for inference of the expressed transcripts; alignments were visualized in the IGV software (Broad Institute, CA, USA) according to Thorvaldsdóttir et al. (2013). The normalization of the obtained readings for the differential expression analysis between both conditions was performed using the DeSeq2 method (Love et al., 2014). Differentially expressed genes (DEGs) were defined as those that showed Log2 (fold-change) values ≥ 1.5 between both conditions and a p-value ≤ 0.05. A principal component analysis (PCA) was performed under PAST version 4.03 software (Hammer et al., 2001) to evaluate the ordination of the samples over sampling times (0, 2, and 5 h) according to their expression profiles in both experimental conditions WMP and MP. This analysis was performed using the log of normalized data corresponding to the 467 genes identified with differential expression (fold-change value ≥ 1.5 and p ≤ 0.05).

Functional classification

To identify the corresponding assigned function for the DEGs identified and the metabolic process, which they are implicated, an analysis in the Kyoto Encyclopedia of Genes and Genomes (KEGG) pathway color was performed (http://www.genome.jp/kegg/tool/map_pathway2.html). The identification of orthologous (KO, KEGG Orthology) for the B. cenocepacia CEIB S5-2 DEGs was performed in the Integrated Microbial Genomes an Microbiomes database (JGI, IMG/M) (https://img.jgi.doe.gov). The respective KO gene functions were assigned using as reference the metabolic pathways dataset of Burkholderia sp. RPE67 (Takeshita et al., 2014) in the KEGG database.

DEGs’ classification by clusters of orthologous groups (COG)

For the evaluation of the functional categories corresponding to the identified DEGs, a cluster analysis of orthologous groups (COGs) of proteins was performed using the gene database of B. cenocepacia CEIB S5-2 from the IMG/M database. The DEGs with a minimum fold-change value of 1.5 and p < 0.05 were classified into functional categories (Galperin et al., 2014).

qRT-PCR analyses

For evaluating the transcriptional expression of the set of genes involved in the MP hydrolysis and the degradation of PNP, qRT-PCR analyses were performed. cDNA libraries were synthesized for each sampling time (0, 2, and 5 h) and experimental condition. cDNA libraries synthesis was carried out from 2 μg of the total RNA, previously treated with DNaseI, using the RevertAid H Minus First Strand cDNA Synthesis Kit (Thermo Scientific, Waltham, MA, USA). Each qRT-PCR reaction was carried out in biological and technical triplicates. The cDNA of each sample was used for the transcriptional expression analysis of the interest genes (mpd and pnpABA'E1E2FDC) between experimental conditions. The RNA polymerase sigma factor 70 (rpoD) gene was used as the endogenous expression control due to its constitutive expression. Subsequently, the expression levels of the selected genes were calculated through the 2−ΔΔCT method (Livak and Schmittgen, 2001). The primers used for qRT-PCR gene validation were taken from Castrejón-Godínez et al. (2019) and are shown in Table S1. The data have equal sample sizes, and each group presents a normal distribution according to Shapiro–Wilk statistical test (W test). In addition, the Levene test showed homogeneity of variances (Levene test). Therefore, we proceeded to perform a one-way analysis of variance. Subsequently, to compare between pairs of averages, a post hoc Tukey test was performed (p < 0.05) (Zar, 2010). Average ± standard deviation and one-way analysis of variance were used to evaluate the effect of the sampling time (0, 2, and 5 h) on relative gene expression (2−ΔΔCT).

Experimental determination of the operon organization of the cluster pnpABA’E1E2FDC

For the experimental determination of the possible organization of the pnpABA’E1E2FDC gene cluster as an operon, PCR reactions were performed using the cDNA library synthesized from the B. cenocepacia CEIB S5-2 RNA extracted 2 h post-culture inoculation in the presence and absence of the pesticide MP. For PCR reaction, the following oligonucleotide pairs were employed, pnpAf-pnpBr, pnpE2f-pnpE1r, pnpE1f-pnpFr, pnpFf-pnpDr, and pnpCf-fdxDr; this set of oligonucleotides were used to amplify adjacent and overlapping sequences in the gene cluster to determine the possible operon arrangement of the genes related with the PNP degradation and to evaluate differences in the expression levels of these genes between the two experimental conditions. The PCR products were evaluated in 1.5% agarose gels, using B. cenocepacia CEIB S5-2 genomic DNA as the positive control and total RNA as the negative control.

Results

B. cenocepacia CEIB S5-2 growth and PNP degradation kinetics

Growth and MP degradation kinetics were performed in MSM supplemented with 50 mg/L of MP. As observed in Fig. 1, at the beginning of the kinetics (0 h), MP is fast hydrolyzed by B. cenocepacia CEIB S5-2, releasing PNP in the culture (8.2 mg/L). After 2 h of culture, the PNP concentration reached its higher value in the kinetics (33.4 mg/L); subsequently, the PNP concentration starts to decrease. Finally, after 5 incubation hours, the PNP concentration in the culture diminished to 6.5 mg/L. On the other hand, the PNP release causes slight adverse effects over the bacterial population, mainly after 2 h of culture when the highest PNP concentration was also observed in the culture. Finally, after 5 incubation hours, a slight increase in DO600nm was observed. This finding suggests that B. cenocepacia CEIB S5-2 not only can efficiently degrade the PNP, but also it could use the PNP as a carbon source.

Total RNA extraction, yield, and quality

After the total RNA extraction and the treatment of the samples with DNaseI, the total RNA concentration in the samples ranged from 500 to 2500 ng/μL (Table S2). After evaluating the quality, concentration, and integrity of the samples, most of the samples showed acceptable RIN values (Fig. S1), being 6.8 the lowest value, corresponding to the second replicate of the experimental condition with MP at time 5 h (14 Bc T5 R2 WMP), and the highest value 8.9, corresponding to the second replicate of the control condition at time 0 h (5 Bc T0 R2 MP).

Gene expression analysis

The fold-change analysis allowed the identification of the set of genes that showed a differential expression for each sampling time in both experimental conditions, in the absence of pesticide (WMP) as well as in the presence of 50 mg/L of MP. At the initial time (T0) in the control condition without pesticide, no DEGs were observed (Fig. 2). However, in the same condition, at sampling time 2 h, 90 DEGs were observed, while at sampling time of 2 h, 143 genes showed differential expression. Of the DEGs identified, 67 (31.9%) were exclusive of the sampling time of 2 h, and 120 DEGs (57.1%) were observed exclusively in the sampling time of 5 h. Twenty-three DEGs were shared between these two sampling times.

Fig. 2
figure 2

Venn diagram showing the distribution of differentially expressed genes detected by DESeq2 from each time and genes shared between the three times of each condition. a Genes expressed in the absence of MP and b genes expressed in the presence of MP. No genes shared between conditions were found. Bc, Burkholderia cenocepacia CEIB S5-2; T0, time zero h; T2, time 2 h; T5, time 5 h. WMP, without methyl parathion; MP, methyl parathion

Full size image

On the other hand, in the experimental condition in the presence of pesticide, at the beginning of the kinetics (T0), seven DEGs were observed; the numbers of differentially expressed genes increased with time; at the sampling time of 2 h (T2), 166 genes showed differential expression, while at 5 h (T5), 232 DEGs were observed. All DEGs observed at T0 were shared with the subsequent sampling times, and 141 DEGs were shared between the sampling times 2 and 5 h. The highest number of exclusive DEGs was observed in the sampling time of 5 h, 91 DEGs (35.4%). Meanwhile, at sampling time T2, only 25 exclusive DEGs (9.7%) were observed. In the presence of MP, several of the identified DEGs encode hypothetical or unknown function proteins. None of the DEGs identified were shared between control and MP-exposed condition, thus indicating that B. cenocepacia CEIB S5-2 responds in a differential way to the stress generated by the absence of a carbon source in the control condition and the generated by the exposure to the pesticide MP. Table S3 shows some examples of the DEGs identified in the different sampling times in the strain B. cenocepacia CEIB S5-2.

Principal component analyses

In general, the principal component analysis (PCA) separated the sampling time per treatment (MP and WMP). In sampling time of 0 h, the PCA explained 99.97% of the variation in the first two components. In the principal component 1 (PC1), which explained 98.50% of the variation, the genes that encoded for p-nitrophenol 4-monooxygenase (pnpA) and the hydroxyquinol 1,2-dioxygenase (pnpE1/E2) were highly correlated. The genes of the histidine kinase GAF domain-containing protein and a transcriptional regulator of the TetR family showed the highest correlations with the principal component 2 (PC2) which explained 1.46% of the variation (Fig. 3a). In sampling time of 2 h (Fig. 3b), the PC1 explained 69.63% of the variation; the negatively correlated genes were two transcriptional regulators of the TetR family, and the gene of a universal stress protein and the positively correlated genes were those who encode proteins related with the cell motility as the flagellar basal-body rod protein (FlgC), the flagellar basal-body rod modification protein (FlgD), and the flagellar basal-body rod protein (FlgF). The PC2 at T2 explained the 20.95% of the variation. The negatively correlated genes were the outer membrane protein (porin) gene and the gene that encodes the glutaryl-CoA dehydrogenase enzyme. In contrast, the positively correlated genes were a transcriptional regulator, the TetR family, a universal stress protein, and the gene of the (R,R)-butanediol dehydrogenase/meso-butanediol dehydrogenase/diacetyl reductase; the two components have an accumulative correlation of 90.58%. Finally, at 5 h sampling time (Fig. 3c), the PCA analysis explained 97.73% of the variation in the two components. The PC1 explained 89.24% of the variation; the genes positively correlated were the 2-polyprenyl-6-methoxyphenol hydroxylase gene, a gene that encodes alcohol dehydrogenase, class IV, the ferredoxin subunit of nitrite reductase, or a ring-hydroxylating dioxygenase gene. The PC2 explained only 8.49% of the variation; the genes negatively correlated were a BON domain-containing protein gene, the histidine kinase GAF domain-containing protein, and a signal transduction histidine kinase gene. In contrast, the genes positively correlated were the uroporphyrinogen-III C-methyltransferase gene, the sulfate adenylyltransferase subunit 1 gene, and a gene that encodes a hypothetical protein.

Fig. 3
figure 3

Principal component analysis of RNA-Seq data. Data corresponding to the control condition are shown in green (T0 filled circles, T2 open squares, and T5 filled diamonds), while data corresponding to the experimental condition with 50 mg/L of MP are shown in red (T0 filled circles, T2 open squares, and T5 filled diamonds). The PCA analyses were performed using normalized RNA-Seq data of a set of 467 with differential expression (fold-change value ≥ 1.5 and p ≤ 0.05)

Full size image

Identification of COGs

Figure 4 shows the DEGs identified for each sampling time and experimental condition which were classified according to their functional category, with the help of the cluster of orthologous group (COG) database (Galperin et al., 2014). At the initial time (T0) in the control condition, no gene showed a significant differential expression. Subsequently, at the sampling time of 2 h, of the 90 DEGs identified, 24 (26.7%) genes couldn’t be associated with a COG; six genes were grouped into poorly characterized categories, four (4.4%) genes in [S] function unknown, and two (2.2%) in [R] general function prediction only. Most of the remaining 60 DEGs identified in T2 were grouped in the following COGs: (1) cell motility [N] with 25 genes (27.8%); (2) amino acid transport and metabolism [E] with seven genes (7.8%); energy production and conversion [C] with six genes (6.7%); cell wall/membrane/envelope biogenesis [M]; and inorganic ion transport and metabolism [P] with four genes (4.4%) each. Meanwhile, in the sampling time of 5 h, 143 DEGs were identified; of these, 52 (36.4%) weren’t grouped in a COG; 14 genes were grouped into poorly characterized categories, eight (5.6%) genes in [S] function unknown, and six (4.2%) in [R] general function prediction only. Most of the remaining 77 genes identified in T5 were grouped in the following COGs: (1) carbohydrate transport and metabolism [G] with 19 genes (13.3%); (2) inorganic ion transport and metabolism [P] with ten genes (7%); (3) lipid transport and metabolism [I] with eight genes (5.6%); (4) amino acid transport and metabolism [E] with seven genes (5%); (5) energy production and conversion [C] with six genes (4.2%); and (6) cell wall/membrane/envelope biogenesis [M] with six genes (4.2%).

Fig. 4
figure 4

Distribution of COG’s in B. cenocepacia CEIB S5-2. T0, time 0 h; T2, time 2 h; T5, time 5 h

Full size image

In the presence of MP, at the initial sampling time (T0), 7 DEGs were observed, four (57.1%) of these genes weren’t associated to COGs, and the remaining three genes were grouped into an equal number of COGs: (1) transcription [K], (2) transport and metabolism of carbohydrates [G], and (3) transport and metabolism of coenzymes [H]. After 2 h of culture, 166 DEGs were identified, and 57 of these genes (34.3%) weren’t associated with a COG, while 15 of the DEGs were grouped in general function prediction only [R] and six in function unknown [S] categories. For the remaining 88 DEGs, the five principal functional categories identified were (1) production and conservation of energy [C] with 16 genes (9.6%); (2) transcription [K] with 13 genes (7.8%); (3) transport and metabolism of coenzymes [H] with 10 genes (6%); (4) carbohydrate transport and metabolism [G] with 9 genes (5.4%); and (5) inorganic ion transport and metabolism [P] with 8 genes (4.8%). At the sampling time of 5 h, of the 232 DEGs, 73 weren’t grouped into a COG, while 15 of the DEGs were grouped in the general function prediction only [R] and nine in function unknown [S] categories. The main functional categories that group the remaining 136 DEGs were (1) energy production and conversion [C] with 21 genes (9.05%); (2) inorganic ion transport and metabolism [P] with 18 genes (7.76%); (3) carbohydrate transport and metabolism [G] with 17 genes (7.33%); (4) transcription [K] with 17 genes (7.33%); and (5) transport and metabolism of coenzymes [H] with 11 genes (4.74%).

Functional annotation by Kyoto Encyclopedia of Genes and Genomes (KEGG)

The DEGs identified in both conditions were submitted to an analysis in the KEGG database (Ogata et al., 1999), using the genome of the strain Burkholderia sp. RPE67 (Takeshita et al., 2014) as reference. In the absence of the pesticide, 53 of 90 DEGs identified at the sampling time of 2 h (59%) were located in metabolic pathways by KEGG; the exclusive metabolic pathways in T2 were (1) ABC transporters, (2) methane metabolism, (3) biosynthesis of antibiotics, and (4) biosynthesis of amino acids. While in the sampling time of 5 h, 54 of the 143 DEGs identified (38%) were located in metabolic pathways by KEGG. In this sampling time, the exclusive metabolic pathways identified were (1) carbon fixation in photosynthetic organisms, (2) pentose phosphate pathway, (3) methane metabolism, and (4) starch and sucrose metabolism (Fig. 5).

Fig. 5
figure 5

Distribution of the DEGs of B. cenocepacia CEIB S5-2 in the KEGG maps at 2 h and five of incubation

Full size image

In the presence of MP, at the initial time (T0), through the analysis in the KEGG database, only one DEG was located in a metabolic pathway related to the degradation of aromatic compounds. After 2 h of culture (T2), 53 of the 166 DEGs identified were located in metabolic pathways in KEEG. The seven exclusive metabolic pathways corresponding to T2 were (1) glutathione metabolism, (2) pyruvate metabolism, (3) fatty acid degradation, (4) ketone body synthesis and degradation, (5) valine degradation, leucine and isoleucine, (6) propanoate metabolism, and (7) degradation of aromatic compounds. The principal overrepresented metabolic pathways were the (1) two-component system, (2) benzoate degradation, (3) tryptophan metabolism, (4) glutathione metabolism, and (5) metabolism of butanoate. Finally, at the sampling time of 5 h, 107 of the 232 DEGs identified were located in metabolic pathways in KEEG; the exclusive metabolic pathways observed in T5 were (1) lysine degradation; (2) benzoate degradation; (3) glutathione metabolism; (4) sulfur metabolism; (5) alanine, aspartate metabolism, and glutamate; (6) biosynthesis of monobactam; (7) synthesis and degradation of ketone bodies; (8) degradation of fatty acids; and (9) metabolism of fatty acids. The overrepresented metabolic pathways in this condition were the following: (1) two-component system, (2) metabolism of butanoate, (3) metabolism of glyoxylate and dicarboxylate, (4) metabolism of tryptophan, and (5) metabolism of propanoate. For the control condition, the exclusive metabolic pathways observed were (1) carbon fixation in photosynthetic organisms, (2) pentose phosphate pathway, (3) methane metabolism, and (4) starch and sucrose metabolism (Fig. 5).

Transcriptional analysis of cluster pnpABA’E1E2FDC

According to the observed fold-change values, all genes that integrate the catabolic cluster pnpABA’E1E2FDC increased their expression levels in the presence of MP. This catabolic cluster includes the genes implicated in both PNP-degrading pathways. Regarding the expression levels of the HQ pathway genes, reported for PNP degradation in Gram-negative bacteria, at the beginning of the kinetics (T0), the genes pnpA (p-nitrophenol monooxygenase), pnpE1 (hydroquinone 1,2-dioxygenase, subunit I), and pnpE2 (hydroquinone 1,2-dioxygenase, subunit II) were the only genes of the catabolic cluster that showed differential expression with fold-change values of 1.8, 1.7, and 1.7 respectively. Subsequently, after 2 h of culture (T2), all the genes related with the PNP degradation through the HQ pathway presented a significant increase in their expression levels, showing the following fold-change values: pnpA, 9.3; pnpB (p-benzoquinone reductase), 9.1; pnpA´, 9.6; pnpE1, 9.1; pnpE2, 9.4; and pnpF (4-hydroxymuconic semialdehyde dehydrogenase), 8.7. After 5 h of culture (T5), all these genes reached their highest fold-change values: pnpA, 9.8; pnpB, 9.8; pnpA´, 9.6; pnpE1, 9.5; pnpE2, 9.9; and pnpF, 9.6 (Fig. 6).

Fig. 6
figure 6

Metabolic pathway used by Burkholderia cenocepacia CEIB S5-2 for MP hydrolysis and PNP degradation

Full size image

On the other hand, after 2 h of culture (T2), the genes implicated in the BT pathway, reported for PNP degradation in Gram-positive bacteria, showed increases in their expression levels. The gene pnpC (catechol 1,2-dioxygenase), which catalyzes the breakdown of hydroxyquinol (BT) for obtaining maleylacetate, showed a fold-change value of 8.1. The gene of the ferredoxin dioxygenase subunit of the nitrate reductase, also known as ring-hydroxylating dioxygenase, showed a fold-change value of 7.8 in T2. The amino acid sequence encoded in this gene belongs to the Burkholderiaceae Rieske 2Fe-2S domain-containing proteins (Pfam PF00355). In bacteria, this group of redox proteins is implicated in the hydroxylation of different aromatic compounds (Haigler and Gibson, 1990; Martin and Mohn, 1999; Fishman et al., 2004). A gene with 97.7% similarity with the enzyme 4-nitrocatechol monooxygenase ferredoxin (Genebank: EKS70305.1) from Burkholderia sp. SJ98, a well-characterized strain capable of degrading nitroaromatic compounds, as PNP (Kumar et al., 2012), showed an increase in expression. 4-Nitrocatechol 4-monooxygenases are implicated in the catalytic hydroxylation of PNP to p-nitrocatechol and its subsequent transformation to 1,2,4 benzenetriol, the two initial enzymatic steps in the BT pathway. The gene of 4-nitrocatechol 4-monooxygenases is situated next to the sequence of the pnpC gene. Due to its proximity and enzymatic activity, this gene could be part of the PNP catabolic cluster in B. cenocepacia CEIB S5-2. After 5 h of culture (T5), both genes showed fold-change values of 9.0. Finally, both PNP-degrading metabolic pathways converge in the production of maleylacetate; the gene pnpD, which also is present in the catabolic cluster pnpABA’E1E2FDC, encodes for maleylacetate reductase, an enzyme that catalyzes the conversion of maleylacetate to β-ketoadipate. This gene also showed differential expression at T2 and T5, with respective fold-change values of 8.4 and 9.2. These findings indicate that B. cenocepacia CEIB S5-2 can overexpress all the genes of both PNP-degrading pathways (Fig. 6).

Relative expression of genes involved in MP hydrolysis and PNP degradation

The expression levels of the mpd gene, involved in the MP hydrolysis and the cluster pnpABA’E1E2FDC, which groups the genes related to PNP degradation, were evaluated through qRT-PCR. The relative expression of this set of genes in the experimental condition without MP was around 1.0 in all sampling times. However, as shown in Table 1, the mpd gene and the genes of the pnpABA’E1E2FDC cluster increased their expression levels over time. The genes pnpA´ and pnpE2 showed higher increases in relative expression levels (2−ΔΔCT). In the sampling time of 5 h, the pnpA´ gene showed relative expression values over 10,000, while the relative expression values of the pnpE2 gene were around 7000. The overexpression of this set of genes observed in the qRT-PCR experiments is in concordance with the results of the differential expression analysis (fold-change) of the transcriptomic data.

Table 1 Relative expression of mpd gene and pnpABA’E1E2FDC cluster. Average ± standard deviation and one-way analysis of variance to evaluate the effect to the sampling time (0, 2, and 5 h) on relative gene expression (2−ΔΔCT)
Full size table

Operon organization of the cluster pnpABA’E1E2FDC

The strain B. cenocepacia CEIB S5-2 presents all the genes that encode the enzymes related to the PNP degradation through the HQ (Gram-negative bacteria) and BT (Gram-positive bacteria) pathways. These genes are organized in a cluster pnpABA’E1E2FDC. As is shown in Fig. 7, all genes included in the cluster are translated in the same direction and show separations between genes lower than 470 bp. This distribution and direction of the genes included in the pnpABA’E1E2FDC cluster suggest that they could be organized in an operon. The results of the sequence analysis indicated the sequential organization of the genes of the PNP biodegradation pathway, with the presence of putative sites for translation promoters, situated in the − 10 and − 35 regions. The pnpABA’E1E2FDC cluster pnpABA’E1E2FDC from B. cenocepacia CEIB S5-2 was compared with similar gene cluster previously reported, as Burkholderia sp. SJ98 (Kumar et al., 2012), B. zhejiangensis CEIB S4-3 (Hernández-Mendoza et al. 2014), Pseudomonas sp. 1–7 (Zhang et al. 2012), Pseudomonas sp. WBC3 (Zhang et al., 2009a), Pseudomonas sp. NyZ402 (Wei et al. 2010), and Pseudomonas putida DLL-E4 (Shen et al., 2010). The analysis revealed the high sequence conservation of this gene cluster among species as Burkholderia sp. and Pseudomonas sp. (Fig. S2), displaying the higher similarity with the clusters from Burkholderia sp. SJ98 (bit score 8446) and B. zhejiangensis CEIB S4-3 (bit score 8278). To evaluate the operon organization of the gene cluster pnpABA’E1E2FDC, different sequential and overlapping segments were PCR amplified using the following oligonucleotide pairs pnpAf-pnpBr, pnpE1f-pnpFr, pnpFf-pnpDr, and pnpCf-fdxDr; in Table S1, the sequence of the oligonucleotide employed and the expected amplicon size are shown; the PCR amplification products showed molecular weights in the range of the theoretical expected values for each oligonucleotide pairs (Fig. S3). Additionally, the co-expression of the evaluated genes was verified, indicating the operon conformation of the genetic cluster.

Fig. 7
figure 7

Operon conformation of the pnpABA´E1E2FDC gene cluster. The genome location of the PNP-degrading genes is shown; arrows indicate the direction of the translation. Op, predicted operons; hp, genes encoding hypothetical proteins; −10 and −35, transcription promoter regions; pnpA, gene encoding p-nitrophenol-4-monoxygenase; pnpB, gene encoding p-benzoquinone reductase; pnpE2/E1, gene encoding hydroquinone-1,2-dioxygenase; pnpF, gene encoding 4-hdroxymuconic semialdehyde dehydrogenase; pnpD, gene coding for maleylacetate reductase; pnpC, gene encoding 1,2,4-benzenetriol dioxygenase; and fdxD, gene coding for ferredoxin dioxygenase subunit of the nitrate reductase. The image was built in SnapGene viewer (SnapGene software, Insightful Science, San Diego, CA, USA)

Full size image

Discussion

According to the genomic data, B. cenocepacia CEIB S5-2 presents all the genes implicated in the biodegradation of PNP through the HQ pathway, which is a characteristic of Gram-negative bacteria. This strain also presents genes implicated in the PNP biodegradation through the BT pathway, mainly characterized in Gram-positive bacteria. Previous works have described the presence of both PNP-degrading pathways in Gram-negative bacteria as Burkholderia sp. SJ98 (Vikram et al., 2013), Burkholderia zhejiangensis CEIB S4-3 (Castrejón-Godínez et al., 2019), and Pseudomonas sp. strain WBC-3 (Zhang et al., 2009b). However, it is important to highlight that B. cenocepacia CEIB S5-2 can biodegrade PNP in only 5 h, while Burkholderia sp. SJ98 can biodegrade PNP in soil (45 mg/Kg) in 16 days (Min et al., 2017), and Burkholderia zhejiangensis CEIB S4-3 can biodegrade PNP in solution (50 mg/L) in 12 h.

The first step in the BT pathway is the conversion of PNP to p-nitrocatechol; subsequently this metabolite is converted to 1,2,4-benzenetriol in the second step of the pathway; these reactions are catalyzed by the enzyme p-nitrophenol monooxygenase-II, in Gram-positive bacteria as Arthrobacter sp. (Jain et al., 1994), Bacillus sphaericus JS905 (Kadiyala et al., 1998), and Rhodococcus opacus SAO101 (Kitagawa et al., 2004). However, the gene which encodes for p-nitrophenol monooxygenase-II has not been identified in the B. cenocepacia CEIB S5-2 genome. The presence of all genes implicated in the HQ pathway and the presence of the pnpC gene from the BT pathway suggest that the strain could degrade PNP and other aromatic compounds through both metabolic pathways. Vikram et al. (2012) studied the PNP degradation pathway present in Burkholderia sp. SJ98, a bacterial strain that can use PNP as the sole carbon source. In the study, all genes of the HQ degradation pathway were identified. However, as in the present work, just the gene pnpC of the BT pathway was identified; the authors could not establish if Burkholderia sp. SJ98 can use the BT pathway simultaneously with the HQ pathway for PNP degradation. The genomic data of Burkholderia sp. SJ98 revealed the presence of the gene EKS70305.1 which encodes for a 4-nitrocatechol monooxygenase ferredoxin protein; this enzyme can catalyze the two initial enzymatic steps in the BT pathway; in B. cenocepacia CEIB S5-2 was identified a gene with 97.7% sequence similarity with the enzyme 4-nitrocatechol monooxygenase ferredoxin from Burkholderia sp. SJ98.

The genomic conformation and sequences of the catalytic gene clusters implicated in the PNP biodegradation are highly conserved in bacterial strains isolated from PNP-contaminated sites. The catalytic cluster pnpABA’E1E2FDC of B. cenocepacia CEIB S5-2 shows great similarity with the PNP catabolic clusters identified in Burkholderia sp. SJ98 (Vikram et al., 2012) and B. zhejiangensis CEIB S4-3 (Castrejón-Godínez et al., 2019). The strain B. zhejiangensis CEIB S4-3 was isolated from agricultural soils from Tepoztlán, in the state of Morelos, Mexico, at the same sampling location as B. cenocepacia CEIB S5-2. The high sequence similarity between the PNP catabolic clusters of both strains suggest events of horizontal transference of the PNP-degrading genes between the strains. In Burkholderia sp. SJ98, the genes pnpE1 and pnpE2 encoded for an Fe(II)-dependent heterodimeric protein that catalyzes the hydroquinone transformation to 4-hydroxymuconic semialdehyde, the genes pnpE1 and pnpE2 from B. cenocepacia CEIB S5-2 show great sequence similarity with the genes from Burkholderia sp. SJ98, suggesting the formation of E1/E2 heterodimers.

According to the results of the DEGs and PCA analyses, the presence of MP in the culture media induces the expression of genes that encode transcriptional regulators in the B. cenocepacia CEIB S5-2 strain; 9 and 13 DEGs with this function were identified at the sampling times of 2 and 5 h, respectively; these genes corresponded to the 1 and 2% of the total DEGs identified at such sampling times. These findings suggest that the MP hydrolysis and the PNP degradation require complex transcriptional regulation. As was previously reported by Chakka et al. (2015), Chen et al. (2016), and Castrejón-Godínez et al. (2019), the transcriptional regulation is not limited to the genes involved in the PNP degradation but also require the regulation of transcriptional and metabolic processes.

The exposure to MP induces the overexpression of genes from different metabolic pathways. The transcriptional profile of B. cenocepacia CEIB S5-2 in the presence of MP shows an overexpression in genes from the TCA pathway at 5 h; in prior sampling times, these genes did not show differences in the expression with respect to the control condition. Similar findings were observed with glyoxylate metabolism. In limited carbon source conditions, the glyoxylate cycle plays an important role in the carbohydrate synthesis from oxalacetate, a molecule generated by the joining of two acetyl-CoA molecules derived from fatty acid reserve degradation. The PNP degradation generates two intermediate metabolites of the TCA pathway, acetyl-CoA and succinyl-CoA. In the presence of MP, after 2 and 5 post-inoculation hours, different genes related to the degradation of the aromatic compounds showed a significant overexpression. In the study of Chen et al. (2016), the transcriptional profile of P. putida DLL-E4 in the presence of PNP and glucose as carbon sources revealed the overexpression of key enzymes from the TCA cycle, in contrast to the control condition in the presence of glucose as the sole carbon source. On the other side, Lu et al. (2013) evaluated the adverse effects related to the organophosphorus herbicide glyphosate exposure in E. coli; the transcriptional analysis revealed that the inhibition of essential enzymes from the pentose phosphate pathway, TCA cycle, and gluconeogenesis caused an inhibition of bacterial growth. The overexpression of genes from the TCA cycle seems to play a key role in the PNP degradation capabilities in bacteria.

At the beginning of the kinetics in the presence of MP, the transcriptional analysis shows the overexpression of a gene that encodes the dominium-denominated GAF from a histidine kinase, a key protein in the two-component signal transduction system. These signal transduction systems are predominant in sensing environmental changes and coupling the cellular physiological responses in bacteria (Capra and Laub, 2012). In the report of Lau et al. (1997), the exposure of P. putida F1 to the aromatic hydrocarbon toluene induced the overexpression of the genes todS and todT. These genes encode for a histidine kinase sensor and its transcriptional regulator, respectively, integrating a two-component signal transduction system. In recent work, it was reported that the expression of the tod operon is related to the regulation of the toluene degradation pathway in P. putida (Koh et al., 2016). Also, through proteome and qRT-PCR analyses, Tiwari et al. (2018) proposed that a two-component signal transduction system triggers the induction of proteins related to oxidative stress contention and biodegradation MP in Fischerella sp. Another early overexpressed gene by B. cenocepacia CEIB S5-2 in the presence of MP was a translational regulator of the family TetR, commonly overexpressed in bacterial strains as a response to the stress caused by the use of alternative carbon sources in the cultures. The TetR transcriptional regulators are also implicated in the environmental stimuli response, multidrug-resistance, quorum-sensing, osmotic stress, and biofilm formation (Cuthbertson and Nodwell, 2013; Le Minh et al., 2015; Colclough et al., 2019). In cultures of Fischerella sp., the presence of MP induced the overexpression of the TetR gene (Tiwari et al., 2018). According to these findings, it could be possible that the TetR gene plays a key role in the MP recognizing, activation of the PNP catabolic cluster, and modulation of the physiological stress response in B. cenocepacia CEIB S5-2.

The participation of different transcriptional regulators has been reported for the degradation of PNP and its derived metabolites. Takeo et al. (2008) reported an AraC type (NphR) transcriptional regulator implicated in the PNP oxidation in Rhodococcus sp. PN1. Shen et al. (2010) identified a LysR-type transcriptional regulator implicated in the hydroquinone degradation in P. putida DLL-E4; the mutation-mediated elimination of the LTTR gene caused the loss of the ability for degrading hydroquinone in the strain. Zhang et al. (2009b) described the importance of the pnpR gene, a LysR-type transcriptional regulator, to activate the expression of the operon that mediates the PNP degradation in Pseudomonas sp. WBC-3. In the present study, at the sampling time of 5 h in the presence of MP, B. cenocepacia CEIB S5-2 showed the overexpression of genes that encode for AraC- and LysR-type transcriptional regulators. Genes that encode for other types of transcriptional regulators as LuxR, reported as a quorum-sensing regulator (Chen and Xie, 2011); GntR, implicated in the regulation of biofilm formation and multiple sugar transport (Li et al., 2019); and ArsR, implicated in metal sensing in different bacteria (Osman and Cavet 2010), also showed an overexpression in the presence of MP; however, the role of these genes in pesticide degradation is still unknown.

The presence of MP in the culture caused the early (T0) overexpression of two genes that encode for a permease and a cellular transporter, both belonging to the mayor facilitator superfamily (MFS) (InterPro: IPR036259). According to the BLAST analysis, the amino acid sequences of such genes showed high similarity, above 90%, with transporters of the MFS family in Burkholderia sp. SJ98 and Burkholderia sp. YI23, respectively. These bacterial strains were isolated from contaminated environments and are capable of degrading the organophosphate insecticide fenitrothion and PNP; the expression of these cellular transporters could play an essential role for the uptake of these xenobiotics during the degradation process in these bacterial strains (Lim et al., 2012; Kumar et al., 2012). In the transcriptomic study of B. zhejiangensis, CEIB S4-3 in the presence of MP observed the overexpression of genes that encode for different transporters such as GntR, AraC, and MerR (Castrejón-Godínez et al., 2019). Transporters are membrane proteins essential for nutrient uptake and the exporting of metabolites and toxic substances from the cell. The MFS transporter family are of the most preponderant proteins in the transport of drugs, ions, sugar, sugar phosphates, and hydrophobic compounds (Jiang et al., 2013; Zomot et al., 2018; Huang et al., 2003). The overexpression of cellular transporters in the presence of MP in B. cenocepacia CEIB S5-2 could be implicated in the pesticide uptake at the beginning of the culture.

After 2 incubation hours in the presence of MP, the highest PNP concentration in the culture was observed. At this time, the overexpression of a gene that encodes for a protein of the universal stress protein family (Usp) was observed; these are cytoplasmic proteins whose expression is altered as a consequence of different internal or external stress conditions (Kvint et al., 2003; Vollmer and Bark, 2018). The Usp-proteins are essential to overcome oxidative stress in E. coli. The absence of the UspA gene in a mutant strain caused premature cellular death, while its exacerbated expression causes and bacterial growth arrest (Nachin et al., 2005). Exposure to pesticides causes diverse stressor conditions over microorganisms (Stanley and Preetha 2016). At the same sampling time (T2), it was observed the overexpression of a gene that encodes for an RNA polymerase sigma-70 factor. The sigma factor family members have the function of directing the general transcription in bacteria, serving as contact points for regulatory proteins, and are also expressed in response to stressing conditions that menace the cell wall and the membrane integrity (Paget and Helmann, 2003). The overexpression of the Usp and the RNA polymerase sigma-70 factor genes at the time of 2 h indicates that the high concentration of PNP induces cellular stress in B. cenocepacia CEIB S5-2; the bacterial capability to tolerate or overcome the adverse effects of exposure to pesticides is a vital characteristic in bacterial strains with high pesticide degradation profiles.

At the sampling times 2 and 5 h, the overexpression of different genes that encode for efflux proteins were observed. As the gene that encodes for a subunit of the outer membrane protein (oprM), the gene that encodes for the permease subunit of a resistance-nodulation-division (RND)-type efflux pump, and the gene that encode for the adapter subunit of an RND-type efflux pump. The efflux systems are key proteins for the resistance to different toxic chemical compounds, mostly in multidrug-resistant bacteria, because these systems actively expulse these toxic compounds to the extracellular media (Akama et al., 2004; Symmons et al., 2009; Phan et al., 2010; Blanco et al., 2016). Moreover, it has been reported that efflux pumps are also important in the tolerance and resistance to pesticides (Nikaido et al., 2011) and oxidative stress contention (Bogomolnaya et al., 2013).

At the sampling time of 5 h, different genes that encode for proteins implicated in the oxidative stress damage contention were overexpressed. Two genes that encode for the C and D subunits of the enzyme alkyl-hydroperoxide reductase (Ahp) were observed. This kind of proteins protect the cell from oxidative stress by reducing the hydrogen peroxide (Bryk et al., 2002; Nunn et al., 2002). In Fischerella sp., it has been reported that the exposition to MP causes the release of reactive oxygen species after 2 and 8 exposure days (Tiwari et al., 2018). Chen et al. (2016) observed the overexpression of genes related to the contention of oxidative stress in P. putida DLL-E4 exposed to PNP.

In the transcriptomic analysis with MP, different genes implicated in the PNP degradation were identified. As a gene that showed a 100% amino acid sequence identity and coverture with the enzyme p-nitrophenol monooxygenase from the Burkholderiaceae family (pnpA gene), an FDA-dependent monooxygenase catalyzes the hydroxylation of PNP to generate p-benzoquinone and release of nitrate. In the genes that encode the sequences of the large subunit of the same enzyme hydroquinone 1,2 dioxygenase (pnpE1) and the small subunit of the same enzyme (pnpE2), both sequences showed high amino acid sequence similitude, 82.2 and 99.4%, respectively, with pnpE1 and pnpE2 from Burkholderia sp. SJ98. Hydroquinone 1,2 dioxygenase catalyzes the aromatic ring opening of hydroquinone to produce 4-hydroxy muconic semialdehyde in the hydroquinone pathway for PNP degradation (Liu et al., 2015; Chen et al., 2016). In the transcriptomic analysis, all genes implicated in the PNP degradation showed differential expression across the kinetics supplemented with MP; in a similar way all genes of the pnpABA’E1E2FDC cluster showed a significant overexpression in the qRT-PCR experiments.

Besides, genes from several metabolic pathways show differential expression in the presence of MP, such as oxidative phosphorylation, sulfur metabolism, amino acids biosynthesis, and the TCA cycle; the overexpression of these metabolic pathways suggests that the strain can use MP and PNP as carbon source and for energy generation, necessary for the maintenance of the bacterial population and the slightly bacterial growth observed during the kinetics. Similar findings were observed with the bacterial strain B. zhejiangensis CEIB S4-3 (Castrejón-Godínez et al., 2019).

Conclusions

B. cenocepacia CEIB S5-2 is a strain with the capability of hydrolyze MP and degrade PNP in 5 h. This capability is related with the expression of hydrolytic enzymes encoded in the mpd gene and the PNP degradation cluster pnpABA’E1E2FDC organized as an operon.

Regarding the transcriptome analysis, in general, the observed expression changes in B. cenocepacia CEIB S5-2 genes are related to the ability to fight with the stress caused by the presence of MP in the media. The observed overexpression of transporters, permeases, and porins in the MP condition suggests an increase in the transport and exchange of molecules with the surrounding media; the overexpression of these kind of proteins could be an essential mechanism to lie with the toxicity caused by the presence of MP, PNP, and the metabolites derived from the PNP degradation. The observed capability of the strain for degrading PNP in the range of 5 h could be related to the presence of genes from both HQ and BT PNP degradation pathways, a significant overexpression in all the genes of the pnpABA’E1E2FDC cluster, as was observed in both transcriptome and qRT-PCR experiments, thus increasing the efficiency of the biodegradation process.

At the end of the kinetics, 5 h, the transcriptional analysis revealed the overexpression of exclusive genes related to the oxidative stress response, probably as a defense mechanism towards the toxic effects caused by PNP. On the other hand, the overexpression of genes related with aromatic compounds and amino acid metabolism, as well as oxidative phosphorylation, suggests that the strain could use the pesticide as a carbon and energy source and for the maintenance of the bacterial population. B. cenocepacia CEIB S5-2 has a great potential for its application in pesticide bioremediation approaches due to its high PNP biodegradation capability and the expression of resistance mechanisms to avoid the stress generated by the PNP exposure. In future studies, the genes related to the PNP biodegradation could be cloned in suitable organisms, and the enzymes can be produced by the use of a heterologous expression system for their subsequent application in pesticide bioremediation strategies.