Cyclohexane, naphthalene, and diesel fuel increase oxidative stress, CYP153, sodA, and recA gene expression in Rhodococcus erythropolis

Abstract In this study, we compared the expression of CYP153, sodA, sodC, and recA genes and ROS generation in hydrocarbon‐degrading Rhodococcus erythropolis in the presence of cyclohexane, naphthalene, and diesel fuel. The expression of cytochrome P450, sodA (encoding Fe/Mn superoxide dismutase), recA, and superoxide anion radical generation rate increased after the addition of all studied hydrocarbons. The peak of CYP153, sodA, and recA gene expression was registered in the presence of naphthalene. The same substrate upregulated the Cu/Zn superoxide dismutase gene, sodC. Cyclohexane generated the highest level of superoxide anion radical production. Hydrogen peroxide accumulated in the medium enriched with diesel fuel. Taken together, hydrocarbon biotransformation leads to oxidative stress and upregulation of antioxidant enzymes and CYP153 genes, and increases DNA reparation levels in R. erythropolis cells.

ancestral form, but acquire expanded substrate specificity and capability to effectively oxidize new substrates (Pérez-Pantoja et al., 2013). This process facilitates emergence of bacteria with xenobiotic-oxidizing enzymes. Whether the processes detrimental for a single cell can be beneficial for a bacterial population and whether the oxidative stress can be an adaptive mechanism to increase nutritional substrate variety? The aim of this study was to evaluate the expression of genes, cytochrome P450, superoxide dismutases A and C, recA, as well as generation of superoxide anion radical and hydrogen peroxide under the influence of hydrocarbons on Rhodococcus.

| Cultivation of microorganisms
Hydrocarbon-degrading Rhodococcus erythropolis was isolated from a technogenically polluted soil containing polycyclic aromatic hydro- Bacteria cultivated in Luria-Bertani (LB) and basic mineral salt medium described before (Sazykin et al., 2016). Fifty-microliter Erlenmeyer flasks containing 20 ml of medium were cultivated at 170 rpm and 30°C in an orbital shaker incubator Innova 40R (New Brunswick, USA).
Such hydrocarbons as cyclohexane, naphthalene (analytical grade, "Aquatest," Russia), and commercial diesel fuel were used in the experiments as an additional carbon source.

| Superoxide anion radical generation assay
For superoxide anion radical assay, microorganisms were grown overnight (18 hr) in the basic mineral salt medium complemented of 0.5% of yeast extract and 0.5% of tryptone. Suspension of microorganism was triply washed and diluted with basic mineral salt medium to the concentration of 1 × 10 8 cells per ml. Hundred microliters of culture suspension, 80 μl of basic mineral salt medium, 10 μl of 4 mM deionized water solution of lucigenin (Sigma-Aldrich, USA), and 10 μl of the hydrocarbon were added to each well of 96-well microplate COSTAR 3632 (USA).The control sample contained 100 μl of the suspension culture, 90 μl of basic mineral salt medium with the addition of 1% of glucose and 10 μl of 4 mM solution of lucigenin in deionized water.
The plate was incubated for 24 hr in the Luminoskan Ascent microplate luminometer (Thermo Scientific, USA) at 30°C with simultaneous chemiluminescence (CL) measurement every 30 min (48 measurements in total) (Sazykin et al., 2016(Sazykin et al., , 2018. Three independent experiments were carried out and repeated 8 times.

| Hydrogen peroxide generation assay
Bacteria were cultivated in 20 ml of basic mineral salt medium with the addition of 2% (400 μl) of hydrocarbons. Basic mineral salt medium with the addition of 2% of glucose was used as the control. The suspension of bacterial cells in the medium before incubation was diluted to the concentration of 1 × 10 6 cells per ml. Microorganisms were incubated in an orbital incubator for 30 days.
Cultural liquid samples were taken from a flask and centrifuged for 5 min at 14,100 g. For hydrogen peroxide assay, 60 µl of the supernatant of culture medium, 100 µl of PBS, and 20 µl of luminol solution were introduced into a well of a plate. CL measurements were carried out 8 times using a Luminoscan Ascent microplate luminometer. The luminescence of each well was measured within 100 s with the interval of 1 s. Twenty microliters of horseradish peroxidase solution (0.01 u/µl) were added to each plate by means of a built-in dispenser immediately after the beginning of measurement.
Luminescence level in a well was determined within 100 s after addition of peroxidase with an interval of 1 s. For each measurement, the average value of CL intensity was calculated and subsequently, the biggest average CL value was used (Sazykin et al., 2016(Sazykin et al., , 2018.
Three independent experiments were carried out 8 times.

| Expression of CYP153, recA, sodA, and sodC genes
Bacteria were cultivated in a basic mineral salt medium with the addition 0.5% yeast extract. Approximately 2% w/v hydrocarbons were added to the medium. Basic mineral salt medium with the addition of 0.5% yeast extract and 2% of glucose was used as a control.
Strains were cultivated to the late logarithmic growth phase and cells were pelleted at 4,000 g for 2 min.

| RNA extraction and cDNA synthesis
RNA was isolated from 25 mg of sample (app. 10 9 CFU). Eight biological replicates were used for each sample. Samples were thoroughly homogenized with mortar and pestle in the presence of liquid nitrogen.
Total RNA extraction was performed with Extract RNA kit (Evrogen, Russia), based on acid guanidinium thiocyanate-phenolchloroform extraction method (Chomczynski & Sacchi, 2006). RNA First-strand cDNA was synthesized using MMLV RT kit (Evrogen, Russia) with random primers. For each sample, we also included a negative control-the same cDNA reaction mix, including RNA, except MMLV.

| Quantitative PCR
The qPCR was performed with the designed primers (Table 1) and hot start PCR kit with EvaGreen dye (Syntol, Russia) using the CFX96 Real-Time PCR Detection System (Bio-Rad, USA). Each sample was analyzed in triplicate qPCR reactions. The reaction parameters were as follows: 94°C for 5 min (the polymerase activation step); 35 cycles of 94°C for 15 s, 60°C for 20 s, 72°C for 30 s, and 72°C for 5 min (the final elongation step) followed by a melting analysis (0.5°C increment from 60 to 95°C; 10 s per cycle). We designed primers using Primer-BLAST tool (https ://www.ncbi.nlm.nih.gov/tools/ primerblast ). The nucleotide sequences of the studied genes were received from the NCBI database. Reaction specificity was controlled using the melting curve analysis and 1.5% agarose gel electrophoresis for each primer pair. No abnormal products were detected.

| Relative gene expression analysis
The normalization of RT-qPCR results should be performed with more than one validated reference gene (Bustin et al., 2009), and we used gyrA, map, recA, rpoB, and rpoC as some of the recommended bacterial reference genes (Rocha, Santos, & Pacheco, 2015). All reference genes had similar expression levels in the control and experimental groups except recA. The recA expression significantly varied between groups, and so we studied recA as the gene of interest. The other genes of interest were CYP153, sodA, and sodC. The PCR efficiency was determined with standard curve analysis and it counted 90%-100%. The relative levels of genes expression were calculated using ΔΔCt method (Bustin et al., 2009;Rao, Huang, Zhou, & Lin, 2013) taking into account PCR efficiency.

| Statistics
Data statistical analysis was conducted using R-studio version 3.4.1 (https ://www.rstud io.com/). The Shapiro-Wilk test was used to check the normality of the data. For data comparison, unpaired Student's t-test and Mann-Whitney U test were utilized. Differences were considered statistically significant at p < 0.05.

| R. erythropolis growth in the presence of hydrocarbons
For culture of R. erythropolis grown in the basic mineral salt medium complemented of 0.5% of yeast extract with the addition of 2% w/v of the investigated hydrocarbons or glucose (as control), the growth curves were built (Appendix 1, Figure A1). Analysis of the growth curves demonstrated that lag phase time was the same at cultivating R. erythropolis on different substrates. Similarly, the exponential phase time was the same for different substrates, as well as the stationary phase time. But R. erythropolis had the lowest increase in biomass (turbidity of suspension) when cultured on the medium with cyclohexane addition. The greatest turbidity of suspension was found in R. erythropolis cultured on the medium with the addition of glucose. After 20-22 hr of cultivation, the difference was threefold. Taken together, based on the similarity of time of various growth phases of R. erythropolis on different substrates, it can be assumed that bacterial cells mainly used the most complete substrate (0.5% yeast extract).

| Superoxide generation by R. erythropolis in the presence of hydrocarbons
The data showing the influence of incubation time with hydrocarbons on lucigenin-activated CL intensity of R. erythropolis are presented in Figure 1.

| Hydrogen peroxide generation by R. erythropolis in the presence of hydrocarbons
Hydrogen peroxide accumulation in the culture medium was es- Microorganisms incubated with the naphthalene produced 2.15 times less H 2 O 2 in the culture medium than that in the control group.
No significant differences were documented between the experimental and control groups starting from the eighth day of incubation in the presence of naphthalene. The production of the hydrogen peroxide by R. erythropolis incubated with the diesel fuel statistically exceeded the same in the control group throughout the experiment. On the first day, the H 2 O 2 concentration in the experimental group was 2.19 times higher than in the control group and gradually F I G U R E 1 Intensity of superoxide anion radical generation (measured by lucigenin-activated CL) upon incubation of R. erythropolis within 24 hr in basal mineral salt medium with the addition of hydrocarbons as a carbon source. Mineral salt medium with the addition of 1% of glucose was used as a control. Error bars are confidence interval limits. Differences between experimental and control measurements are statistically significant (p < 0.05) for all the data sets.
F I G U R E 2 Intensity of hydrogen peroxide production (measured by luminol-activated CL) upon incubation of R. erythropolis within 30 days in basal mineral salt medium with the addition of hydrocarbons as a carbon source. Mineral salt medium with the addition of 2% of glucose was used as a control. Error bars are confidence interval limits. Differences between experimental and control measurements are statistically significant (p < 0.05) for all the data sets.
decreased toward the middle of incubation. From 22nd day, the concentration of the peroxide increased and exceeded by 24.7 times the value in the control group by the end of incubation.
First, it should be mentioned that for more clarity the gene relative expression levels were shown in Figure 3 according to 2-∆Ct data, but for an accurate comparison, ∆∆Ct formula was used. and naphthalene by 9.8 times, as it was in case of CYP153 expression. It is important to note that recA expression is usually stable and it is often used for normalization of other genes (Rocha et al., 2015).

Significant differences in
In this study, we obtained quite opposite result.
It is well known that in the course of enzymatic reactions of cytochrome P450 the so-called "disjunction" of the cycle may occur when the flow of electrons derived from NAD(P)H to P450 molecules leads to the generation of superoxide anion radical and/ or hydrogen peroxide instead of the products of monooxygenase reaction (Guengerich, 2001;Goeptar, Scheerens, & Vermeulen, 1995).
In this study, we demonstrated the increase of CYP153 transcription in R. erythropolis cultivated with hydrocarbons. The naphthalene addition caused the greatest induction (20.7-fold). Cyclohexane and diesel fuel caused a weaker effect-6.0-and 8.2-fold, respectively.
Since bacterial P450 cytochromes could be a source of superoxide anion radicals, we observed the simultaneous induction of sodA and CYP153 genes. Cyclohexane increased the transcription of CYP153 by 6 times and sodA by 3.1 times, diesel fuel-8.2 and 5.4 times, and naphthalene-20.7 and 16.1 times, respectively.
Besides, only naphthalene introduction led to upregulation of the sodC expression.
The present data obtained for R. erythropolis coincided with our previous results of reactive oxygen species production by Acinetobacter calcoaceticus (Sazykin et al., 2016) and Achromobacter xylosoxidans (Sazykin et al., 2018). Namely, O 2 ·− production increased during first 12 hr of bacteria incubation with hydrocarbons, and then it decreased. However, microorganisms of different bacterial taxa differ in ROS generation depending on hydrocarbons type. In R. erythropolis and A. xylosoxidans (Sazykin et al., 2018), the maximum production of O 2 ·− was caused by cyclohexane, and in A. calcoaceticus-by diesel fuel and PAHs (Sazykin et al., 2018). Most likely, it is associated with great importance of O 2 ·− production in a prokaryotic cell when shunting an enzymatic cycle by suboptimum substrates, and, respectively, presence of hydrocarbons oxidases with different specificities in various microorganisms. In prokaryotes, the contribution of ROS caused by oxidized PAHs derivatives is much lower compared to the eukaryotic cell, possibly, due to the fact that PAHs immediately disrupt electron transport chain in the eukaryotic cell.
Decrease in O 2 ·− in the final stage of the 24-hr incubation, presumably, is not due to the reduction of superoxide production in a cell, but due to the increase of the bacterial superoxide dismutase Higher sodA in comparison to sodC expression is important for bacterial cell protection against O 2 ·− radicals as it was shown for P.
aeruginosa (Hassett, Schweizer, & Ohman, 1995). Significant increase in SOD enzymatic activity caused by hydrocarbons was demonstrated for such bacteria as A. calcoaceticus and A. xylosoxidans earlier (Sazykin et al., 2016(Sazykin et al., , 2018. In this case, hydrocarbon degraders are protected from the reactive type of ROS-superoxide anion radical generated at the initial stages of hydrocarbons oxidation, transforming it into much stable form-hydrogen peroxide.  (Gogoleva, Nemtseva, & Bukharin, 2012;Sazykin et al., 2016Sazykin et al., , 2018. These processes do not cause only damages to cells but represent a mechanism of procrustean adaptation to changing nutritional conditions of the environment.
In case of Pseudomonas putida, high level of intracellular antioxidants protects bacterial cells from negative effects of the oxidative stress, but it might also reduce the rate of changes in the cell's enzyme systems and its adaptability to new substrates. Taken together, the antioxidant systems possibly determine plasticity and diversification of bacterial population. The balance between bacterial cells survival and the ability of microorganisms to colonize new ecological niches is still an extremely intriguing problem.

| CON CLUS IONS
Addition of cyclohexane, diesel fuel, or naphthalene increases the CYP153 gene expression and production of superoxide anion radical in hydrocarbon-degrading R. erythropolis. The expression of Fe/ Mn superoxide dismutase (sodA) and recA genes proliferates also.
Further, the quantity of sodC (Cu-Zn superoxide dismutase) mRNA elevates in the presence of naphthalene. Bacteria incubated with diesel fuel accumulate hydrogen peroxide in the culture medium.
Therefore, biotransformation of such hydrocarbons as alkanes, cycloalkanes, and aromatic hydrocarbons leads to oxidative stress and intensifies enzymatic antioxidant protection and DNA reparation in R. erythropolis cells.

CO N FLI C T O F I NTE R E S T S
The authors declare that they have no conflict of interest. All authors read and contributed to the manuscript.

E TH I C S S TATEM ENT
None required.

DATA ACCE SS I B I LIT Y
The 16S