Introduction

Petroleum hydrocarbon contamination in soil fields is a serious problem, particularly in countries that produce, transport, and refine oil (Lu et al. 2014). Phytoremediation is an effective technique used for the long-term in situ rehabilitation of petroleum hydrocarbon contamination (Vidali 2001). The removal of pollutants from soil is a disruptive and expensive physical process (Trellua et al. 2016). Therefore, phytoremediation is an efficient and promising cleanup approach for extracting or inactivating organic and inorganic pollutants from soil.

Different plant groups have investigated to reveal specific mechanisms to reduce oil contamination in soil (Jones et al. 2004; Liste and Prutz 2006; Kechavarzi et al. 2007). Plant species vary considerably in their phytoremediation capacity (Liste and Alexander 1999; Liste and Prutz 2006). Wheat, rye, oat, maize, and other agricultural crops have been reported to tolerate crude oil pollutants in soil (Aprill and Sims 1990; Wild et al. 1992; Liste and Alexander 1999). The bioremediation capability of plant species has been established in different studies investigating oil-contaminated soil (Jones et al. 2004; Kaimi et al. 2006; White et al. 2006; Yergeau et al. 2012). Even though effective phytoremediation of hydrocarbon pollution has been demonstrated, the specific adaptations responsible for this ability and the mechanisms involved in managing the resistance of particular plants to pollutants are not fully understood.

The specific adaptation mechanisms regulating the resistance of specific plants to contaminants are very important for the efficient phytoremediation of organic compounds. Researchers are increasingly focused on this area, investigating factors such as pollutant-specific tolerance mechanisms and structural and functional changes in the rhizosphere under pollution. The identification of the most effective remediating species for particular compounds is a priority in phytoremediation experiments. During the phytoremediation of crude oil-contaminated soil, the effect of plant species on colonization pattern and metabolic activity of the inoculated endophytes was evaluated in different studies (Fatima et al. 2016).

Polycyclic-aromatic hydrocarbons (PAHs) are major components in petroleum-contaminated soils, are resistant to degradation processes, have strong carcinogenic effects on human health, and pose a serious risk to ecosystems (Totsche et al. 2006; Wehrer and Totsche 2009). PAH treatments in plants lead to different stresses. PAHs cause leaf deformations, accumulation of H2O2, oxidative stress, up-regulation of antioxidant systems, cell death, and reduced plant growth (Gao and Zhu 2004; Alkio et al. 2005; Burritt 2008; Liu et al. 2009). Phytotoxicity differs according to the specific PAH applied and the plant species (Baek et al. 2004). However, information regarding the mechanisms of PAH toxicity in plants, the genes responsible for PAH degradation and stress signaling mechanisms, is not fully understood.

cDNA microarrays have been reported to be effective tools for the identification of stress-responsive genes involved in tolerance and stress responses (Seki et al. 2004; Shinozaki et al. 2003). The present study aims to contribute fundamental information regarding the mechanisms underlying tolerance to the toxic effects of petroleum hydrocarbons on plant leaves and roots using the Affymetrix Zea mays GeneChip. The mechanism by which petroleum hydrocarbon exposure activates stress signaling pathways common to other abiotic or biotic stresses has not been well characterized, and whether the identified genes are restricted to petroleum hydrocarbon stress is unclear.

We report the common and distinct regulatory mechanisms involved in the response to petroleum hydrocarbon contaminated soil in leaf and root tissues of Z. mays. Molecular knowledge regarding the stress response and defense mechanisms is imperative for finding new approaches to promote plant tolerance to PAHs.

Materials and methods

Plant material and petroleum treatment

Zea mays L. Shemal seeds were surface sterilized for 5 min in 1% (w/v) sodium hypochlorite and washed with distilled water. After soaking the sterilized seeds in distilled water for 12 h, a pre-germination step was performed in Petri dishes for 10 days in the dark. Three independent experiments were performed for the effect of petroleum on Z. mays growth and each petroleum concentration was examined in triplicate. The germinated seeds were sown in pots containing a soil and sand mixture contaminated with 0 (control), 1, 2.5, and 5% Siberian Light (SBL) crude petroleum. The equal amount of ¼ Hoagland’s solution was applied twice a week to all the pots and grown under 25/22 ± 2 °C and 60 ± 5% % RH with the 8-h light/16-h dark cycle. After 30 days, the leaves and roots from the petroleum-treated (1, 2.5, and 5%) and untreated control plants were harvested and the roots were rinsed in double distilled water for 3 min to remove any surface contamination. Total fresh weights of leaves and roots were measured and frozen in liquid nitrogen for RNA isolation.

RNA isolation

Total RNA was isolated from the leaf and root tissues (100 mg) using the RNeasy Plant Kit (Qiagen, Germany). Each RNA sample was incubated with RNase-free DNaseI (Recombinant, Roche Applied Science GmbH, Germany) to eliminate the genomic DNA and purified according to a previously described method (Keskin et al. 2010). The RNA integrity and the quantity of the purified RNA samples were determined using a spectrophotometer (Nanodrop, ThermoFisher Scientific, USA) and 1.0% (w/v) formaldehyde agarose gel electrophoresis.

cDNA synthesis for quantitative real time PCR

The cDNA synthesis was performed using MMLV reverse transcriptase (Roche Transcriptor High Fidelity cDNA Synthesis Kit), according to the manufacturer’s instructions. The efficiencies of the cDNA synthesis reactions were assessed by performing conventional PCR using a housekeeping gene-specific primer designed from Z. mays β-actin. The QRT-PCR experiments were performed in triplicate of each sample for three biological replicates.

Microarray analysis

GeneChip Maize Genome Arrays were used to identify the differentially expressed genes. The arrays contain approximately 14,850 Zea mays transcripts that correspond to 13,339 genes. One microgram DNA-free RNA was used to generate biotin-labeled cDNA. The leaves and roots of Z. mays seedlings that were grown for 30 days in different petroleum hydrocarbon contaminated soils (control, 1, and 5%) were individually sampled, and each sample included three independent biological replicates for the microarray analysis. The hybridization, washing, staining, and scanning steps were performed according to the Affymetrix GeneChip standard protocol, and all protocols were performed at AY-KA Ltd. (Ankara, Turkey). Three biological replicates of each sample were performed to assess the quality and reproducibility of the chip hybridization. GeneChip suite 4.0 (Affymetrix) was used for the data normalization, and the fold changes were determined using GENESPRING 5.0. The Affymetrix CEL data analysis was performed using R Studio Version 0.98.1091 (http://www.bioconductor.org/biocLite.R). The data pre-processing included a background correction performed using the Robust Multiarray Average (RMA) algorithm. The normalization and PM correction were performed using the LOESS method and MAS5 algorithm, respectively, and the median polish method was applied for the summary.

Quantitative real-time PCR analysis

To independently confirm the microarray data, qRT-PCR experiments were performed using an IQ5 System (BioRad Laboratories, Hercules, USA). The qRT-PCR reactions were performed in 25 μl total volumes in 96 well polypropylene plates containing 200 ng of the cDNA samples, 10 pmole of each primer, and 12.5 μl SYBR Green Mix (Roche FastStart Universal SYBR Green Master), as described by Cevher-Keskin et al. (2012). The IQ5 System (BioRad Laboratories, Hercules, USA) was used for all qRT-PCR reactions in cycles of 3 min at 95 °C, followed by 50 cycles at 95 °C for 30 s, 55 °C for 30 s, and 72 °C for 40 s. To rule out possible contamination, we subjected a negative control containing no template cDNA to the same procedure in all experiments. To accurately quantify the transcript level, we performed three technical replicates for each experiment. The relative abundance levels of Cys protease (NM_001112009.1), CBL Interacting Protein Kinase (CIPK) (NM_001115006.2), 12-oxophytodienoic acid reductase (OPR) (NM_001112429.1), polyamine oxidase1 (pao1) (NM_001111636.1), early nodulin 93 (Enod93) (NM_001153581.1), and glutamine synthetase2 (GS2) (NM_001111973.1) were detected. The primer sequences of these genes are shown in Table 1. The comparative CT method was used to calculate the fold changes in expression (Schmittgen et al. 2000). The ΔCT values of all mRNA transcripts were averaged across all treatments and experimental replicates. The mRNA expression levels of the genes were normalized to the levels of β-actin in all qRT-PCR experiments.

Table 1
Full size table

Results

To determine the differences in the transcript levels among the petroleum hydrocarbon-treated Z. mays, we performed microarray experiments using whole plants grown for 30 days in pots containing 0, 1, 2.5, and 5% petroleum contaminated soil. Petroleum hydrocarbons in soils negatively impact plant growth and development (Fig. 1). In all tested concentrations, chlorosis symptoms were detected in the leaves of Z. mays. Significant decreases were observed in the shoot and root lengths and the number of leaves, nodes, and branches (Fig. 2).

Fig. 1
figure 1

Growth of Z. mays in soils polluted with the following concentrations of petroleum: 0 (control), 1, 2.5, and 5% (v/v)

Full size image
Fig. 2
figure 2

Petroleum hydrocarbon inhibits root and shoot growth. Z. mays plants were grown in soil contaminated with different percentages (0, 1, 2.5, and 5%) of petroleum hydrocarbon for 30 days

Full size image

The GeneChip Maize Genome Array (Affymetrix, California, USA) was used to characterize the petroleum hydrocarbon responsive genes in maize. Figures 3 and 4 show the distribution of the different expression levels of the genes in leaf and root tissues treated with two different petroleum hydrocarbon percentages (1 and 5%). The Venn diagrams show the number of genes that are responsive to petroleum hydrocarbon in the leaf (Fig. 3) and root (Fig. 4) tissues in Z. mays.

Fig. 3
figure 3

Venn diagram demonstrating the number of up- and down-regulated genes in leaf tissues under petroleum hydrocarbon stress. Genes whose expression was significantly changed (p < 0.05) by twofold are shown. The maximum gene expression difference (2668 genes) was observed in the leaf tissues of the plants treated with 5% petroleum hydrocarbon. a Differentially expressed genes in leaf tissues exposed to 1% oil are shown in pink. b Differentially expressed genes in leaf tissues exposed to 5% oil are shown in green. c Differentially expressed genes regulated in leaf tissue by 1 and 5% oil are shown in blue

Full size image
Fig. 4
figure 4

Number of differentially expressed probe sets under petroleum hydrocarbon stress. Venn diagrams of up-regulated and down-regulated genes in root tissue. Genes that changed their expression significantly (p < 0.05) by twofold are given. The maximum gene expression difference (2162 genes) was observed in root tissues of 5% petroleum hydrocarbon treated plants. a Genes that change their expression in root tissue exposed to 1% oil are shown in pink. b Genes that change their expression in root tissue exposed to 5% oil are shown in green. c Genes that are commonly regulated by 1 and 5% oil are shown in blue

Full size image

The distribution of the expression levels of the genes in the leaf tissues was more variable than that in the root tissues (Fig. 3). Expectedly, the maximum gene expression difference in both the root and leaf tissues was observed in the 5% petroleum-treated Z. mays plants (Figs. 3 and 4). In the 5% petroleum hydrocarbon-treated plants, 2668 and 2162 up- and down-regulated genes, respectively, with greater than a twofold change were identified in the leaf and root tissues (Suppl. Data 1 and 2). After the pre-processing step, the farthest neighbor algorithm was applied to select ten highly differentially expressed genes and the 20 genes clustered in the diagrams (Suppl. Data 3 and 4). Table 2 shows the gene IDs and functional descriptions of these 20 highly differentially expressed genes.

Table 2 The gen ID and functional description of highly differentially expressed 20 genes
Full size table

Molecular functional categories of the differentially regulated genes in the root and leaf tissues in Z. mays

The differentially expressed genes in the root and leaf tissues of the petroleum hydrocarbon-treated Z. mays were categorized according to their biological and molecular functions and cellular location using the agriGO (http://bioinfo.cau.edu.cn/ agriGO/) program. To identify the genes that were regulated by petroleum hydrocarbon, we classified the up- and down-regulated genes according to their functions, including binding, catalytic activity, electron carrier, transporter, structural molecule, transcription regulator activity, and molecular transducer activity. Notably, most differentially expressed genes in the leaf tissues belonged to the structural molecular activity group (Fig. 5). This group includes basic vital functional genes, such as new ribosome biogenesis genes. This result was consistent with the observation that stress-related gene expression activation induces protein synthesis in leaf tissues. Because petroleum hydrocarbon treatments induce responses similar to those induced by drought stress, increased mRNA levels of osmotic stress- and antioxidant enzyme-related genes were observed in the root tissues (Fig. 6).

Fig. 5
figure 5

In leaf tissues, the differentially expressed genes with greater than twofold changes were classified according to their molecular functions. Most differentially expressed genes in the leaf tissues belonged to the structural molecular activity group. This group includes basic vital functional genes, such as new ribosome biogenesis genes. This result was consistent with the observation that stress-related gene expression activation induces protein synthesis in leaf tissues

Full size image
Fig. 6
figure 6

Differentially expressed genes with greater than twofold changes in the root tissues were classified according to their molecular function

Full size image

The expression levels of catalytic activator genes in the root tissues were up-regulated but were not higher than those in the leaf tissues (Fig. 7). The expression level of catalytic activator genes, such as hormone biosynthesis-related genes (abscisic acid and gibberellic acid), was increased in the petroleum hydrocarbon-treated leaf tissues of Z. mays (Fig. 7). The structural molecular activity in the leaf tissues was also up-regulated by petroleum hydrocarbons (Fig. 7).

Fig. 7
figure 7

Differentially expressed genes with greater than twofold changes in three leaf tissue groups (0–1%, 0–5%, and 1–5%) were classified according to their molecular functions. The expression levels of catalytic activator genes, such as hormone biosynthesis-related genes (abscisic acid and gibberellic acid), and the structural molecular activity were increased in the petroleum hydrocarbon-treated leaf tissues of Z. mays. The antioxidant and electron carrier activity was up-regulated in the root tissues of the 1 and 5% petroleum hydrocarbon-treated Z. mays (Fig. 8)

Full size image
Fig. 8
figure 8

Differentially expressed genes with greater than twofold changes in three root tissue groups (0–1%, 0–5%, and 1–5%) were classified according to their molecular functions

Full size image

Regarding the biological categories of genes, the expression of cell component biogenesis-related genes in the leaf tissues was higher than that in the root tissues of the petroleum hydrocarbon-treated plants (Figs. 9 and 10). Biological regulation and the regulation of biological processes were suppressed in the root and leaf tissues of the petroleum-treated Z. mays (Figs. 9 and 10). The response to stimulus, multi-organism, and developmental process was up-regulated in the root tissues of the petroleum-treated plants (Fig. 10).

Fig. 9
figure 9

In the leaf tissues, the differentially expressed genes with greater than twofold changes were classified according to their biological function. The expression of cell component biogenesis-related genes in the leaf tissues was higher than that in the root tissues of petroleum hydrocarbon-treated plants

Full size image
Fig. 10
figure 10

In the root tissues, the differentially expressed genes with greater than twofold changes were classified according to their biological functions

Full size image

Quantitative RT-PCR

To confirm the microarray data, differentially expressed six genes that might be implicated in different stress responses were selected for qRT-PCR experiments (Table 1). qRT-PCR analyses of these transcripts were performed in Z. mays roots and leaves grown in soils polluted with different concentrations of petroleum (0, 1, 2.5, and 5%).

Cys proteases are involved in cellular regulation (Wagstaff et al. 2002), and the maximal Cys expression was observed in the 2.5% petroleum hydrocarbon-treated leaf tissues. In the 1% petroleum hydrocarbon-treated leaf tissues, the mRNA expression of Cys was higher than that in the control tissue. The maximal Cys expression was observed in the 2.5% treated-plants, and the Cys expression was lower in the 5% petroleum-treated plant leaves (Fig. 11a). The mRNA expression levels of the Cys gene were decreased by petroleum in the root tissues of Z. mays (Fig. 11b).

Fig. 11
figure 11

Total RNA was isolated from 30-day-old Z. mays grown on 0, 1, 2.5, and 5% petroleum hydrocarbon-contaminated soil and used for qRT-PCR experiments. The mRNA regulation of Cys protease, CIPK, and Opr in root and leaf tissues of Z. mays are shown. Error bars represent the standard deviation of the qRT-PCRs performed in triplicate. The gene expression was normalized using β-actin as a housekeeping gene. a Maximal CIPK expression was observed in the 2.5% petroleum hydrocarbon-treated plant leaves. b In the root tissues, the expression levels decreased as the petroleum hydrocarbon concentration increased. c Maximal CIPK expression was observed in the 2.5% petroleum hydrocarbon-treated plant leaves. d In the root tissues, the expression levels were decreased by the petroleum hydrocarbon treatment. e Opr mRNA from plants grown in soil with 2.5 and 5% petroleum hydrocarbon showed 25- and 30-fold changes in expression level relative to the control tissues, respectively. f Opr expression in the root tissues was not dramatically affected by the different petroleum concentrations

Full size image

The CBLs constitute a family of Ca2+ sensors that are only found in plants. The responses of CIPK (CBL-interacting protein kinase) to the petroleum hydrocarbon treatment were different between the root and leaf tissues (Fig. 11c, d). The maximal CIPK expression was observed in the 2.5% petroleum hydrocarbon-treated plant leaves (fourfold). In contrast, the expression levels were decreased by the petroleum hydrocarbon treatment in the root tissues (Fig. 11d).

The 12-oxo-phytodienoic acid reductases (OPRs) are the enzymes that are important for the detoxification process. In leaves, Opr mRNAs from plants grown in soil contaminated by 2.5 and 5% petroleum hydrocarbon showed 25- and 30-fold expression level changes relative to the control tissues, respectively (Fig. 11e, f). The Opr expression in the root tissues was not dramatically affected by the different petroleum concentrations (Fig. 11f).

Pao is involved in ABA-induced cytosolic antioxidant defense through H2O2 in maize leaves. H2O2 is derived from polyamine oxidation due to cell death, the hypersensitivity response, and the expression of defense genes (Yoda et al. 2006; Xue et al. 2009). According to our qRT-PCR analysis, the highest Pao1 gene expression was observed in the 1% petroleum hydrocarbon-treated leaf tissues. Decreasing expression patterns were observed in the leaf tissues as the concentration of petroleum hydrocarbon increased (Fig. 12a). In the root tissues, significant differences were observed among the 1, 2.5 and 5% petroleum hydrocarbon-treated Z. mays, but there was no correlation between the Pao gene expression and the concentration of petroleum hydrocarbon (Fig. 12b).

Fig. 12
figure 12

mRNA expression patterns of Pao, GS2, and Enod93 in the root and leaf tissues of Z. mays grown in soil contaminated with 0, 1, 2.5, and 5% petroleum hydrocarbon. Gene expression was normalized to β-actin, which was used as a housekeeping gene. Error bars represent the standard deviation of the qRT-PCRs performed in triplicate. a The highest Pao gene expression was observed in the 1% petroleum hydrocarbon-treated leaf tissue. Decreasing expression patterns were observed in the leaf tissues as the concentration of petroleum hydrocarbon increased. b In the root tissues, no correlation was observed between the Pao gene expression and the concentration of petroleum hydrocarbon. c Maximal GS2 gene expression was observed in the 1% petroleum hydrocarbon-treated leaf tissues (20-fold). Decreasing expression pattern was observed as the concentration of petroleum hydrocarbon increased. d However, no significant difference was observed in the root tissues. f Enod93 mRNA expression in the root tissues increased as the concentration of petroleum hydrocarbons increased. Maximal expression was observed in the 5% petroleum hydrocarbon-treated root tissues. e In contrast, the opposite Enod93 expression profile was observed in the leaf tissue

Full size image

In plant cells, GS2 plays an important role in the flow of nitrogen into nitrogenous organic compounds and is involved in glutamate synthesis. Maximal GS2 gene expression was observed in the 1% petroleum hydrocarbon-treated leaf tissue (20-fold). Decreasing expression patterns were observed in the leaf tissues as the concentration of petroleum hydrocarbon increased. In contrast, no significant difference was observed in the root tissues (Fig. 12c, d).

The ENOD93 proteins are assumed to play roles in cell structure, the control of nodule ontogeny, and carbon metabolism (Okubara et al. 2000). According to our qRT-PCR experiments, Enod93 mRNA expression increased in the root tissues as the concentration of petroleum hydrocarbon increased. Maximal expression was observed in the 5% petroleum hydrocarbon-treated root tissues. However, the opposite Enod93 expression profile was observed in the leaf tissues (Fig. 12e, f).

Discussion

The main goal of this study was to identify the global expression profile in response to petroleum hydrocarbon stress in Z. mays. The reduced growth of Z. mays plants due to the petroleum-contaminated soil might be the cause of the reduction in leaf growth and/or delay in cell expansion (Agbogidi et al. 2007; Athar et al. 2016). Organic chemical pollution in soil and water has been demonstrated to extensively affect microbial populations (Vilchez-Vargas et al. 2013; Gonzalez et al. 2015). According to the microarray analysis, the expression of genes related to general stress response mechanisms (such as GST) was increased as the petroleum hydrocarbon concentration was increased. The increases in the percentage of the oil prevented plant water intake and, therefore, incurred transcriptional changes that are observed following drought stress. The increased expression levels of certain genes, such as those in the Derlin general secretion pathway, may influence the provision of microorganisms in the rhizosphere. In addition, the suppression of certain root meristematic stem cell differentiation genes, such as Ocl4, may retard the growth of plant root tissue.

Cys

Cys protease expression was up-regulated in the root tissues as the concentrations of petroleum hydrocarbon increased, closely correlating with the microarray data (Fig. 11b). Cys proteases are important for cellular regulation and several developmental events associated with programmed cell death (Elbaz et al. 2002; Estelle 2001; Matarasso et al. 2005). Cys protease has also been shown to be expressed during drought, senescence ripening, and wounding (Alonso and Granell 1995; Harrak et al. 2001).

Cysteine proteases are involved in signaling pathways and the response to biotic and abiotic stresses. Four different cysteine protease probes were induced in the leaf tissues of Z. mays plants.

CBL–CIPK complexes regulate abiotic stress responses, such as responses to drought, cold, and salt and ABA signaling (Li et al. 2006; Xu et al. 2006). According to the qRT-PCR analysis, the responses of CIPK to the petroleum hydrocarbon treatment differed between the root and leaf tissues (Fig. 11c, d). Decreasing expression was observed with increasing concentrations of petroleum in the root tissues of Z. mays (Fig. 11d).

According to the qRT-PCR experiments, the Opr expression levels in the root tissues increased with increasing concentrations of petroleum (Fig. 11f). The OPRs catalyze the decrease in the double bonds adjacent to an oxo group in a, b-unsaturated aldehydes or ketones. Therefore, these enzymes are very important for the detoxification process. Previous reports (Zhang et al. 2005) suggest that different Opr genes are also differentially regulated in response to stress hormones, wounding, and biotic stress, such as pathogen infection (Sobajima et al. 2007; Zhang et al. 2005).

In plants, polyamine oxidation-derived hydrogen peroxide (H2O2) production has been associated with cell wall development, ABA-induced antioxidant mechanism, wound healing, cell death, and biotic stress (Angelini et al. 2008; Yoda et al. 2006; Xue et al. 2009). Abiotic stress, such as drought, heat, and salt induced Pao gene expression in plants and bacteria (Aziz et al. 1998; Suzuki et al. 2004). Pao expression has also been reported in the ABA-related antioxidant mechanism in maize leaves (Cona et al. 2006). According to our microarray data, Pao expression in the leaf tissues increased as the oil concentration increased. The qRT-PCR results confirmed the petroleum hydrocarbon stress responsive behavior of Pao (Fig. 12a, b).

GS is an important enzyme for nitrogen metabolism, which converts toxic ammonium to nontoxic glutamine (Gln) in the presence of glutamate and ATP. Genetic transformations of GS genes have been performed in different systems, such as transgenic lotus plants, tobacco, wheat, and poplar trees, to enhance nitrogen assimilation (Fuentes et al. 2001; Habash et al. 2001; Oliveira et al. 2002; Pascual et al. 2008). Maximal GS2 gene expression was observed in the 1% petroleum hydrocarbon-treated leaf tissues (20-fold) according to the qRT-PCR analysis (Fig. 12c, d). Decreased expression patterns were observed in leaf tissues as the petroleum hydrocarbon concentrations increased.

According to the qRT-PCR analysis, Enod93 mRNA expression was up-regulated in the root tissues and down-regulated in the leaf tissues with increasing petroleum hydrocarbon concentrations (Fig. 12e, f) “Nodulation (nod)” genes play an important role in the induction of nodules in leguminous plant roots. The development of the nodule depends on the coordinated expression of plant and bacterial genes. The nodulation process is initiated by the stimulation of root cortical cell division (Tam et al. 1997). Enod40 is rapidly triggered by rhizobia in the root pericycle and dividing cortical cells of the nodule primordium (Yang et al. 1993; Crespi 1994).

Antioxidant enzymes

Under biotic or abiotic stress, major reactive oxygen species (ROS) (superoxide oxidase and H2O2) are produced in plant cells (Moller et al. 2007), and antioxidant and detoxification enzymes are up-regulated under these stresses. In maize, CAT (Zm.6357.3.S1_a_at), APX Zm.5762.1.A1_at), SOD (Zm.241.1.S1_at), and GR (Zm.7168.1.S1_at) were up-regulated under petroleum hydrocarbon pollution. Ten microarray probes representing peroxidase were up-regulated in our microarray data, particularly in the root tissues. The APX1 gene transcript was highly abundant in willow trees grown in petroleum-contaminated soil (Gonzalez et al. 2015).

Versicolorin reductase

Oxidoreductase activity was increased in the leaf tissues. In the presence of oxygen and water, aldehyde oxidase produces carboxylic acids from aldehydes and catalyzes the conversion of an aldehyde to an acid and hydrogen peroxide. According to our microarray data, two probes representing aldehyde oxidase, four probes representing 12-oxo-phytodienoic acid reductase, and four probes representing peroxidase were induced in the leaf tissues by petroleum hydrocarbon.

Lipoxygenase gene expression is involved in different stress responses, such as wounding, abiotic (water deficiency), or biotic (pathogen attack) stress responses (Melan et al. 1993; Porta et al. 1999). The LOX5 and LOX8 probes, representing lipoxygenase, were up-regulated in the leaf tissues. Our comparative microarray approach found that Z. mays MAP kinase4 (MPK4) mRNA was up-regulated by 2.6- and 1.5-fold in the leaf tissue grown in the 1 and 5% petroleum hydrocarbon-contaminated soils, respectively. According to a DDRT-PCR analysis in T. aestivum performed by Keskin et al. (2010), the expression of MPK4 mRNA was induced by ABA treatment. Water stress induced MPK4 mRNA expression in T. aestivum, according to an RNAseq analysis (Cevher-Keskin et al. 2015).

Pyruvate decarboxylase catalyzes the oxidation of decarboxylation of pyruvate to acetaldehyde. We observed that pyruvate decarboxylase (pcd2) mRNA was up-regulated in our microarray experiments (twofold by 1% petroleum and 1.9-fold by 5% petroleum). Pyruvate decarboxylase (pcd3) was also up-regulated in the leaf tissues (2.3-fold by 1% and 4.1-fold by 5% petroleum). In contrast, in the root tissue, Pcd1 was down-regulated (25-fold by 1% and 11-fold by 5% petroleum). In the absence of oxygen, an increase in the activity of pyruvate decarboxylase was observed in wheat and maize root tips (Waters et al. 1991; Drew et al. 1994). Thus, increased pyruvate decarboxylase activity is likely a marker of tissue hypoxia or anoxia (Mohanty and Ong 2003).

Carbonyl reductase 1 (cbr1) has oxidoreductase activity and was induced 28.5-fold and 9.8-fold in the leaf tissues of maize grown in 1 and 5% petroleum hydrocarbon-contaminated soil, respectively. Versicolor in reductase, which is another oxidoreductase, was up-regulated in the leaf (18.3-fold by 1% and 11.0-fold by 5% petroleum) and root (2.2-fold by 1% and 2.0-fold by 5% petroleum) tissues.

Transport- and cell wall-related genes

In cellular transport processes, the ion transporters support homeostasis in the plant cytoplasm. The “ion transporter genes” are up-regulated by high salinity stress (Hasegawa et al. 2000). The cell wall protein expansin is involved in cell wall loosening, enlargement, and other developmental processes (Sampedro and Cosgrove 2005). Beta expansins (exB4, exB5, exB7, and exB8) and α-expansins (exA3 and exA5) were up-regulated in the root tissues of Z. mays by petroleum hydrocarbon according to the microarray data. β-Expansin mRNA is induced by ABA treatment in leaf tissues of T. aestivum (Keskin et al. 2010), and six different probes representing glutathione S-transferase12 (gst6, 9, 11, 12, 13, 14) were increased by petroleum in the Z. mays root tissues. Twenty-two probes representing glutathione S-transferase 2 and three different glutamine synthetase2 (gln2) probes were up-regulated in the leaf tissues. The gene expression of gln2 was induced by petroleum hydrocarbon, but no significant difference was found in the root tissues (Fig. 12c, d). The microarray data are consistent with the data obtained from the qRT-PCR experiments.

Abiotic stresses, such as salt and water, stimulate proteolytic activity and decrease protein synthesis, leading to free amino acid accumulation in the cells. Excessive amino acids reduce the need for ammonium incorporation to form ammonium acids and indirectly cause ammonium accumulation. Higher GS1 and GS2 mRNA expression levels induce more ammonium absorption in roots, causing ammonium accumulation and eventually inducing cell senescence.

Tonoplast-membrane-integral protein ZmTIP2-2 (NM_001111561) was induced (threefold) in the root tissues by 1% petroleum hydrocarbon. TIP1 (TaAQP6) participates in internal water redistribution by regulating water transfer under drought stress (Zhang et al. 2008). The overexpression of plasma membrane aquaporin BnPIP1 in Nicotiana tabacum caused an enhanced tolerance to drought stress (Yu et al. 2005). In contrast, ABA-induced TIP1 mRNA expression was observed in the leaf tissues of T. aestivum (Keskin et al. 2010). TIP1 expression is differentially regulated in leaf and root tissues under drought conditions in drought tolerant and non-tolerant T. aestivum cultivars (Cevher-Keskin et al. 2015).

Hormone signaling network

ABA-induced plasma membrane protein PM19 and ABA-induced protein were up-regulated in response to the petroleum hydrocarbon treatment in the root tissues of Z. mays. ABA-treated T. aestivum immature embryo cells were closely related to the considerable accumulation of PM19 (Koike et al. 1997). However, in leaf tissues, auxin inducible protein gene expression was up-regulated approximately 2.3- and 3.5-fold by the 1 and 5% petroleum contaminated-soils, respectively.

Transporters of ion or sugar function by adjusting the osmotic potential between two sides of the membrane. According to our microarray results, inorganic phosphate transporter 3 and polyol transporter protein 4 were down- and up-regulated, respectively, in the root tissues (Suppl. Data 5 and 6). However, inorganic phosphate transporter 3 and plastidic 2-oxoglutarate/malate transporters were down-regulated in the leaf tissues, and zinc transporter 4 was up-regulated (Suppl. Data 7 and 8). Bray (2004) demonstrated that genes related to cell wall metabolism were down-regulated during osmotic stress in Arabidopsis.

Two different probes representing transglutaminase (TGase) were up-regulated by petroleum hydrocarbon. Calcium-dependent TGase activity is regulated by senescence, salt stress, programmed cell death, differentiation, and light in different plants (Serafini-Fracassini and Del Duca 2002; Dondini et al. 2003).

A significant response to oil contamination was observed in the gene expression of chitinase. In the leaf and root tissues, the functional genes encoding chitinase were dramatically down-regulated as the concentration of petroleum hydrocarbon increased. Regarding the nitrogen fixation-related functional genes, the early nodulin 93 genes were significantly down-regulated in the root tissue by the 1 and 5% petroleum hydrocarbon contamination. Nodulin-like protein, which is responsible for transmembrane transport, was up-regulated in the leaf tissues grown in the 1 and 5% petroleum hydrocarbon-contaminated soil.

Because of the hydrophobicity of petroleum-polluted soils, petroleum contamination induces drought stress (Li et al. 1997). Our comparative transcriptome analyses in both root and leaf tissues of Z. mays show that the metabolism of water uptake and osmotic homeostasis-related gene expression is predominantly influenced by the petroleum hydrocarbon treatment.

Conclusion

In conclusion, transcriptome studies performed in maize provided data regarding the mechanism of tolerance to petroleum hydrocarbon. Plants grown in petroleum-contaminated soils showed increased protein and gene expression that induced microorganisms. Because of the toxic crude oil, related genes are activated in plants. Since plant water uptake is affected, the expression levels of genes associated with osmotic balance are also affected. The present results are highly valuable for breeding and engineering plants for the phytoremediation of petroleum hydrocarbons.