Assessment of the ameliorative effect of curcumin on pendimethalin-induced genetic and biochemical toxicity

The present study aimed to assess the toxic effects of pendimethalin herbicide and protective role of curcumin using the Allium test on cytological, biochemical and physiological parameters. The effective concentration (EC50) of pendimethalin was determined at 12 mg/L by the root growth inhibition test as the concentration reducing the root length by 50%. The roots of Allium cepa L. was treated with tap water (group I), 5 mg/L curcumin (group II), 10 mg/L curcumin (group III), 12 mg/L pendimethalin (group IV), 12 mg/L pendimethalin + 5 mg/L curcumin (group V) and 12 mg/L pendimethalin + 10 mg/L curcumin (group VI). The cytological (mitotic index, chromosomal abnormalities and DNA damage), physiological (rooting percentage, root length, growth rate and weight gain) and oxidative stress (malondialdehyde level, superoxide dismutase level, catalase level and glutathione reductase level) indicators were determined after 96 h of treatment. The results revealed that pendimethalin treatment reduced rooting percentage, root length, growth rate and weight gain whereas induced chromosomal abnormalities and DNA damage in roots of A. cepa L. Further, pendimethalin exposure elevated malondialdehyde level followed by antioxidant enzymes. The activities of superoxide dismutase and catalase were up-regulated and glutathione reductase was down-regulated. The molecular docking supported the antioxidant enzymes activities result. However, a dose-dependent reduction of pendimethalin toxicity was observed when curcumin was supplied with pendimethalin. The maximum recovery of cytological, physiological and oxidative stress parameters was recorded at 10 mg/L concentration of curcumin. The correlation studies also revealed positive relation of curcumin with rooting percentage, root length, weight gain, mitotic activity and glutathione reductase enzyme level while an inverse correlation was observed with chromosomal abnormalities, DNA damage, superoxide dismutase and catalase enzyme activities, and lipid peroxidation indicating its protective effect.

Pendimethalin (N-(1-ethylpropyl)-2,6-dinitro-3,4-xylidine), a dinitroaniline group-selective herbicide, is widely used to control a wide variety of weeds and broadleaf plants 4 . Generally, it is applied to soil before the emergence of seedlings, planting and sometimes early post-emergence. In light sandy loam soil, pendimethalin can penetrate to root zone and exert phytotoxic effects in presence of adequate moisture 5 . Pendimethalin can inhibit cell division causes chromosomal aberrations and interfere in the formation of cell wall 6 . It also retards root and shoot growth of plants 7 . Pendimethalin is a low mobile and low volatile dinitroaniline compound having poor water solubility 8 . Previously, numerous experiments have been conducted to study the phytotoxicity of pendimethalin in different crop species such as rice 9 , cotton 10 , pigweed species, common lambsquarters, barnyardgrass 11 , paspalum cultivars 12 and various summer season flora 13 .
It has been reported that pendimethalin and other dinitroaniline herbicides might be adsorbed and degraded in the soil [14][15][16] , its lipophilicity, stability and soil adsorption characteristics pose a potential risk to the environment 17,18 . Dinitroaniline compounds are also toxic to non-targeted invertebrates and aquatic organisms. In addition, dinitroaniline can form a carcinogenic compound nitrosamine that is harmful to humans 19 .
Pendimethalin can cause oxidative stress and inhibit the antioxidant system of cells 20 . The excess production of reactive oxygen species (ROS) resulting protein oxidation, cell membrane damage, inactivation of enzymes, RNA and DNA damage 21 . ROS interacts with biomolecules and damages cellular processes that result in the reduction of yield and plant growth. ROS can damage the cell membrane causing lipid peroxidation that is a highly destructive process. Lipid peroxidation may alter the structure and function of cell membranes by oxidizing polyunsaturated fatty acids (PUFAs). Impairment of PUFAs may lead to loss of fluidity and secretary functions. Malondialdehyde (MDA), an important indicator of lipid peroxidation, may be produced as a by-product of PUFA oxidation 22,23 .
Plants have numerous enzymatic and non-enzymatic defense mechanisms to overcome the impact of ROS. They accumulate different metabolites such as osmoprotectants and amino acids to assure from stress. In addition, enzymatic antioxidants are one of the most important mechanisms of plants to counter stress conditions. However, toxic compounds can perturb the activity of antioxidant enzymes such as catalase (CAT), superoxide dismutase (SOD), glutathione reductase (GR), ascorbate peroxidase (APX) and glutathione-S-transferase (GST) reflecting the toxicity level and plant stress tolerance capability. The protein interactions with compounds using molecular docking tool provide information about the predominant mode of interaction and binding efficacy of protein and ligand presenting protein-ligand complexes as 3D-crystal structures 24,25 . Similarly, pendimethalin herbicide may disrupt the natural structure of enzymes/proteins by interacting with their residues.
Turmeric (Curcuma longa L.) is a spice that has increased an interest toward both the medical/scientific and culinary communities. Turmeric is a rhizomatous herbaceous perennial plant of the ginger (Zingiberaceae) family 26 . Turmeric has an important component curcumin a curcuminoid as well as essential oils. Curcuminoids are phenolic chemicals derived from the roots of the C. longa L. and other species of Curcuma. Curcumin (diferuloylmethane) is a low molecular weight polyphenol that was first chemically identified in 1910. It is often regarded as the most active component, comprising about 2-8% of most turmeric formulations 27 . Curcumin (1,7-bis(4-hydroxy-3-methoxyphenyl)-1,6-heptadiene-3,5-dione), a polyphenolic compound, is a brightly yellow color pigment and the main derivative of Curcuma spp. Curcumin has been demonstrated to target several signaling molecules while also displaying action at the cellular level, lending evidence to its numerous health advantages 28 . It has the ability to alter Stat3 phosphorylation and DNA binding activity in human cancer cells 29 . Several in vitro and in vivo studies have shown that the therapeutic potential of curcumin does not cause side effects in animal models or humans, even at very high doses 30 . Curcumin has been linked to several therapeutic effects, including the regulation of cancer cell proliferation via different biological pathways such as apoptosis, mutagenicity, cell cycle regulation, angiogenesis, invasion, and tumorigenesis 31 . It has also antioxidant, antimicrobial, anti-parasitic and anti-inflammatory properties. It has been reported that curcumin scavenges ROS by regulating antioxidant enzyme activities in hyperglycemia 32 . All the effects of curcumin have been reported in animal models whereas no any such evidences are explored in case of plant system so far.
Allium cepa L. is one of the most suitable plant models for investigating the toxic effects of environmental pollutants. It is a very sensitive assay to study chromosome aberrations induced by several toxic chemicals 33,34 . The accuracy of this test is indicated by the similarity in the toxicity results revealed in different experiments using A. cepa L., cell culture (in vitro) and animal tests (in vivo) [35][36][37][38] . Allium test has also been used to explore the harmful effects of environmental pollutants on normal cell division. This test is a suitable technique to elaborate information on toxic substances induced chromosomal alterations, inhibition of mitotic activity and DNA damage with a detoxification mechanism 39,40 .
The present study investigated the effects of pendimethalin on root development (cell division and elongation kinetics) and antioxidant enzymes (SOD, CAT and GR) level as well as the interactive potential of pendimethalin molecule in A. cepa L. In addition, the protective role of curcumin against these effects was also assessed.

Materials and methods
Test organisms. The equal-sized (25-35 mm diameter, untreated) A. cepa L. bulbs purchased from a local market were used as the test material. A. cepa L. (Amaryllidaceae) (2n = 16) was defined using taxonomic characters and approved at the Department of Botany, Faculty of Arts and Sciences, Giresun University. Bulbs were stored in a cool and dry environment. Before application, the external scales as well as brownish base plates were removed from the bulbs without damaging the root primordia.
Determination EC 50 concentration. The

Experimental design.
The experiments based on plant were performed in accordance with international guidelines and regulations. The experiments had been comprised into in six groups. The control group was treated with tap water, application groups were treated with curcumin (5 and 10 mg/L), EC 50 concentration of pendimethalin (12 mg/L) and their combination at 24 °C for 96 h ( Table 1). The roots were exposed to the treatment solution directly in 60 × 42 mm beakers. About 40 bulbs from each group were used for physiological study and 10 bulbs were selected randomly for cytological and biochemical studies.
Physiological parameters. About 50 bulbs from each group were used for the determination of root length based on radical formation using a millimeter ruler. The weight gain was measured using precision scales before and after treatment. The rooting percentage and relative injury rate was calculated using Eqs. (1) and (2) 42 .

Determination of chromosomal abnormalities (CAs), micronucleus (MN) and mitotic index (MI).
The root tips were fixed for 2 h in Clarke solution (ethanol, glacial acetic acid, 3:1) followed by 96% ethanol for 15 min. Then samples were processed in 70% ethanol at 4 °C. For cytological studies, the roots were hydrolyzed in 1 N HCl for 17 min at 60 °C, incubated with 45% acetic acid for 30 min and stained for 24 h in acetocarmine. Preparations of mitotic cells were analyzed under a binocular microscope at ×500 magnification 43 . The evaluation of the MN involvement has been carried out according to Fenech et al 44 .
For each group, 10 slides were prepared from randomly selected bulbs, 1000 cells for MN and CAs frequency and 10,000 cells for MI were counted in each slide (Eq. 3).
Comet assay. Comet assay (alkaline single-cell gel electrophoresis) was performed according to the modified method of Tice et al. 45 The nuclei were isolated from fresh root tips in 600 µL ice-cold nuclear isolation buffer (400 mM 6H 2 O-MgCl 2 , 0.5% w/v Triton X-100, 0.4 M Tris, pH 7.5) using a petri dish with a razor blade and root tips were centrifuged at 1200 rpm for 7 min after being passed through a nylon mesh filter. A mixture of 1:1 nuclear suspension with 1% low melting point (LMPA) in phosphate-buffered saline (PBS) was placed on a pre-coated slide with 1% normal melting point agarose (NMPA). Electrophoresis was performed in chilled electrophoresis buffer at 0.7 V/cm at 4 °C (20 V, 300 mA) for 15 min using a power supply. Slides were rinsed three times with filtered water and neutralized with Tris buffer (0.4 M Tris, pH 7.5). The nuclei were stained for 5 min with ethidium bromide after immersion in cold water for 5 min. The preparations were washed with cold water and eliminated residual stain and coverslip sealed. These steps were taken with low light to avoid DNA degradation and examined with a fluorescence microscope. Comets were analyzed using Comet Assay software version 1.2.3b 46 . About 100 cells per slide were analyzed for DNA damage. The extent of DNA damage was scored from 0 to 4 depending upon the level of DNA damage. The cells were classified into five categories based on tail DNA (%): 0-undamaged (0-2%), 1-low damage (< 2-25%), 2-moderate damage (< 25-45%), 3-high damage (< 45-70%) and 4-extreme damage (< 70%) 47 . The total DNA damage per sample, expressed as arbitrary units, was calculated using Eq. (4).   Evaluation of antimutagenic effects. The antimutagenic effect of curcumin was calculated for each slide by Eq. (5) and mean ± SD (standard deviation) values were calculated 48 . In order to assess the antimutagenic effect, chromosomal abnormalities (CAs) and the arbitrary unit of the comet assay was evaluated.
Lipid peroxidation. Lipid peroxidation was evaluated by measuring the quantity of MDA according to the method of Ünyayar et al. 49 Approximately 0.5 g of root tissue were homogenized with 5% trichloroacetic acid (TCA) and centrifuged at 12,000 rpm at 24 °C for 15 min. The supernatant, TCA solution (20%) and thiobarbituric acid (0.5%) has been transferred to the new tube and incubated at 96 °C for 25 min. The tubes were taken into the ice bath and centrifuged at 10,000 rpm for 5 min. The absorbance was recorded at 532 nm, the extinction coefficient was 155 mM/cm has been used to determine the quantity of MDA content.  www.nature.com/scientificreports/ Data analysis. Root elongation kinetics was characterized using logistic, log-logistic, Gompertz and Weibull models. The selection of the model was based on the best fit in Bayesian Information Criterion (BIC) and Akaike information criterion (AIC). The free parameters namely maximum growth (A), growth rate (µ) and length of lag phase (λ) shared by models were employed as descriptors of the kinetics. The functions of "tidyverse" and "drc" packages for R programming language were applied in root elongation kinetics 63 . SPSS Statistics v22.0 (IBM Corp., USA, 2015) package program was used to perform statistical analyzes. Data were expressed as mean ± SD (standard deviation) in the tables and mean ± SEM (standard error of means) in the graphs. The statistical significance between the means was determined by the method of One-way ANOVA and Duncan's test and the p < 0.05 was deemed statistically significant. Since there were two independent variables in the analysis of mutagenicity inhibitions, independent samples t-test was used and p < 0.05 was deemed statistically significant.

Results
Effect of pendimethalin on root growth and protective role of curcumin. The effects of pendimethalin were evaluated on root length, rooting percentage, injury rate and weight gain ( Table 2). About 100% rooting was recorded in group I (control), group II (5 mg/L curcumin) and group III (10 mg/L curcumin). However, the application of 12 mg/L pendimethalin (group IV) had decreased the rooting percentage to 30%. Further, combination of pendimethalin with curcumin in group V (12 mg/L pendimethalin + 5 mg/L curcumin) and group VI (12 g/L pendimethalin + 10 mg/L curcumin) had recovered rooting percentage to 47.50% and 57.50% respectively.
The relative injury rate represented the severity of the damage. The highest relative injury rate was observed in group IV (0.69). The relative injury rate was low in group VI (0.41). A concentration-dependent improvement in relative injury rate was recorded due to the application of curcumin at different concentrations.
The toxicity of pendimethalin is indicated by retarded root growth. After 96 h of treatment, the mean root length was 16.92, 17.15 and 17.30 cm in group I, group II and group III respectively. About 50% reduction in root growth (8.45 cm) was recorded in pendimethalin-treated roots in group IV. However, the application of curcumin at different doses recovered root growth to 10.21 and 12.22 cm in group V and group VI respectively. Table 3 shows the effects of pendimethalin and curcumin on weight gain. The highest weight gain was observed in group III (10.78 g) followed by group I (10.72 g) and group II (10.26 g). Pendimethalin treatment  Table 2. Effect of pendimethalin and curcumin on rooting percentage and root length. Group I: tap water (control), Group II: 5 mg/L curcumin, Group III: 10 mg/L curcumin, Group IV: 12 mg/L pendimethalin, Group V: 12 mg/L pendimethalin + 5 mg/L curcumin, Group VI: 12 mg/L pendimethalin + 10 mg/L curcumin. Data are shown as mean ± standard deviation (SD). The averages shown with different letters in the same column are statistically significant (P < 0.05).  Table 3. Effect of pendimethalin and curcumin on weight gain. Group I: tap water (control), Group II: 5 mg/L curcumin, Group III: 10 mg/L curcumin, Group IV: 12 mg/L pendimethalin, Group V: 12 mg/L pendimethalin + 5 mg/L curcumin, Group VI: 12 mg/L pendimethalin + 10 mg/L curcumin. Data are shown as mean ± standard deviation (SD). The averages shown with different letters in the same column are statistically significant (P < 0.05). www.nature.com/scientificreports/ decreased the weight gain to 3.09 g in group IV. Further, the application of curcumin recovered weight gain to 4.92 and 6.37 g in group V and group VI respectively. The kinetics of root elongation revealed significant decreases in root length (A), lag phase (λ) and growth rate (µ) of pendimethalin-treated roots in comparison to control (Fig. 1). No significant difference in λ and A of group I, group II and group III was observed. However, a significant difference in µ was noted in all groups.

Impact of pendimethalin on mitotic index and chromosomal abnormalities and protective role of curcumin.
The cytogenetic effects of pendimethalin application and the protective role of curcumin were investigated by mitotic index (MI), chromosomal aberrations (CAs) and micronucleus (MN) formation. According to the data in Table 4, there was no significant difference among the MI rate of group I (9.09%), group II (9.14%) and group III (9.34%). The lowest MI (4.67%) was observed in group IV treated roots. However, a concentration-dependent increase in MI rate with the administration of curcumin (5 mg/L and 10 mg/L) and pendimethalin was recorded in group V (6.00%) and group VI (6.89%).
Different types of chromosomal abnormalities were shown in Table 5. A few abnormal cells were reported in control (group I) showing fragment, sticky chromosome and micronucleus formation. No chromosomal abnormalities were noted in roots in group II and group III treated with 5 mg/L and 10 mg/L curcumin, respectively. There was no significant difference in chromosomal abnormalities observed in group I, group II and group III. A range of chromosomal abnormalities such as fragment, sticky chromosome, bridge, binucleated cells, vagrant chromosome, unequal distribution of chromatin and micronucleus (Tables 4, 5) were recorded (Fig. 2). In addition, micronucleus formation usually indicated chromosomal instability and genotoxic effects. The frequency of chromosomal abnormalities was high in group IV treated root with pendimethalin. Administration of curcumin at 5 mg/L and 10 mg/L in group V and group VI decreased chromosomal abnormalities induced by pendimethalin. This dose-dependent decline in chromosomal abnormalities indicated the antimutagenic effect of curcumin. Mutagenicity inhibition was ascertained as 27.91% and 46.66% in group V and group VI respectively. The results indicated that chromosomal abnormalities were decreased with increasing mutagenicity inhibition.
DNA damage induced by pendimethalin in the roots of A. cepa L. and protecting role of curcumin were determined using single-cell gel electrophoresis test with three scales (head DNA, tail DNA and arbitrary unit) ( Table 6). In the calculation of arbitrary units, cells were analyzed by classifying them into five categories (Fig. 3) according to tail DNA. There was no significant difference in head DNA (%), tail DNA (%) and arbitrary unit between control (group I) and curcumin alone applied groups (group II and group III). However, pendimethalin treatment (group IV) decreased head DNA percentage to 32.43, increased tail DNA percentage to 67.57% and arbitrary unit to 291.90. Further, administration of different doses of curcumin with pendimethalin in group V and group VI reduced the tail DNA percentage and arbitrary unit whereas increased head DNA percentage.  Table 4. Effects of pendimethalin on MN formation and Mitotic index (MI) and protective role of curcumin. Group I: tap water (control), Group II: 5 mg/L curcumin, Group III: 10 mg/L curcumin, Group IV: 12 mg/L pendimethalin, Group V: 12 mg/L pendimethalin + 5 mg/L curcumin, Group VI: 12 mg/L pendimethalin + 10 mg/L curcumin. Data are shown as mean ± standard deviation (SD). The averages shown with different letters in the same column are statistically significant (P < 0.05).  (Fig. 4a). The antioxidant enzymes activities mitigate and repair ROS damage in plants. GR maintains the level of reduced glutathione in the cell. The reduced form of glutathione plays important role in the control of ROS. GR levels in group I, group II and group III was insignificant and noted as 55.16, 54.45 and 53.75 µmol NADPH/g FW respectively. GR levels in group IV, group V and group VI were significantly different from the above groups and recorded as 18.15, 22.48 and 28.56 µmol NADPH/g FW respectively (Fig. 4b).   www.nature.com/scientificreports/ SOD activity was affected by the production of superoxide radicals. SOD levels in control and curcumin alone applied groups were recorded as 35.46, 36.65 and 34.72 U/mg FW respectively. SOD levels in other groups were significantly higher than in the control. SOD levels in group IV, group V and group VI were recorded as 71.56, 59.25 and 46.5 U/mg FW respectively (Fig. 4c).

MN formation Mitotic index (MI) Mitotic index rate (%)
CAT neutralizes the generated hydrogen peroxide in plant cells by detoxifying it into water and oxygen. CAT levels in group I (0.59 OD 240nm /min.gFW), group II (0.55 OD 240nm /min.g FW) and group III (0.52 OD 240nm /min.g FW) were insignificant. However, CAT levels in group IV (2.41 OD 240nm /min . g FW), group V (1.96 OD 240nm / min.gFW) and group VI (1.65 OD 240nm /min . g FW) were significantly different from group I, group II and group III (Fig. 4d).
In addition, the results of molecular docking based on binding energy revealed that pendimethalin has capable to interact with antioxidant enzymes (Fig. 5) as well as DNA molecules (Fig. 6). Table 7 Table 8.
Dose-depended protective effects of curcumin and its correlation with physiological, genetical and antioxidant enzymes. The dose-dependent protecting effects of curcumin against pendimethalin has been shown in Fig. 7. A concentration-dependent recovery in all parameters was observed after synergistic use of curcumin with pendimethalin. The highest protecting effect of curcumin was observed at 10 mg/L con- Table 6. Detection of the effects of pendimethalin and curcumin applications on DNA by comet assay. Group I: tap water (control), Group II: 5 mg/L curcumin, Group III: 10 mg/L curcumin, Group IV: 12 mg/L pendimethalin, Group V: 12 mg/L pendimethalin + 5 mg/L curcumin, Group VI: 12 mg/L pendimethalin + 10 mg/L curcumin. Data are shown as mean ± standard error of means (SEM). The averages shown with different letters in the same column are statistically significant (P < 0.05).

Groups
Head DNA (%) Tail DNA (%) Arbitrary unit Mutagenicity inhibition (%)  The correlation study of all the parameters has been given in Fig. 8. Positive correlations are indicated by blue color and negative correlations by red color. The intensity of color and size of the circle is related to correlation coefficients. A positive correlation of pendimethalin was observed with CAs, micronucleus, DNA damage, SOD and CAT enzyme activities, and lipid peroxidation whereas a negative correlation was noted with rooting percentage, root length, weight gain, MI and GR enzyme activity. Similarly, the positive correlation of curcumin with rooting percentage, root length, weight gain, MI and GR enzyme activity was recorded while a negative correlation was observed with CAs, micronucleus, DNA damage, SOD and CAT enzyme activities, and lipid peroxidation indicating its protective effect.

Discussion
The extensive use of pesticides like pendimethalin has usually contaminated the environment. Therefore, the assessment of action mechanisms of pendimethalin and its effect on mitotic activity, growth chromosomes, genetic material and antioxidant enzymes becomes important. The protective role of curcumin, as performed in this study, is imperative in providing useful information about the recovery of injured roots.
Pendimethalin caused significant decrease in root length, rooting percentage and weight gain but curcumin at 5 mg/L and 10 mg/L showed protecting effect on these parameters when treated alone or in combination with pendimethalin. The root elongation kinetics also supported decreases in lag phase (λ), root length (A) and growth rate (µ) in pendimethalin-treated roots as well as curcumin treatment-induced recovery in µ and A. The elongation kinetics-related parameters are sensitive endpoints to assess phytotoxic responses due to physical interaction between root and pesticide. Earlier studies explained pendimethalin-induced inhibition of germination 64 , reduced seedlings 65 and root growth 66 . The MI may be a reliable endpoint generally used for cytotoxicity measurement. The inhibition of mitotic activity described the cytotoxic potential of pendimethalin 17,67 . In this study, pendimethalin-treated roots of A. cepa showed MI inhibition. However, at a higher concentration of curcumin (10 mg/L), mitotic activity of the cells was improved when applied alone or in combination with pendimethalin. Several studies have been reported mitodepressive effects of herbicides 66,[68][69][70] . Like, pendimethalin exposure inhibits cell division and disturbs the formation of microtubules in root meristems of A. cepa. Microtubules are important for the development of cell wall and spindle fiber leading to cell division and differentiation 71,72 .
Although various types of CAs such as fragrant, sticky chromosome, bridge, binucleated cells, unequal distribution of chromatin and vagrant chromosome were monitored in pendimethalin treated roots, at a higher www.nature.com/scientificreports/ concentration of curcumin (10 mg/L) reduction in CAs were observed. Stickiness is a chromatid type abnormality and is associated with the effect of pollutants on depolymerization/degradation of DNA 73 . The appearance of fragments, bridges and micronuclei indicates the clastogenic potentiality of chemicals 74,75 . The arrest of cytokinesis in the cell cycle results in binucleated cells 76 . Interference of chemicals in spindle formation may lead to chromosomal anomalies like unequal distribution of chromatin and vagrant chromosome. Similar to the results, pendimethalin induced CAs was reported by Verma and Srivastava 17 and Ahmad et al. 67 The findings   78 , prometryne 69 and clopyralid 79 . Environmental contaminants can induce DNA damage that is either repaired or unrepaired. These unrepaired damages may cause alterations in DNA. Presently, the comet assay is extensively used for toxicological studies of environmental contaminants 80 . The results of the comet assay revealed that pendimethalin could induce high levels of DNA damage associated with high toxic effects in roots of A. cepa. However, curcumin having a www.nature.com/scientificreports/ protective effect when supplied with pendimethalin could reduce DNA damage. Previous studies suggested that the antioxidant system is unable to control the activities of ROS and free radicals produced during the treatment of chemicals, in turn that may induce DNA damage 81,82 . Similar response was observed in other organisms Chinese hamster over cells 83 , Biomphalaris alexandrina snails 84 and freshwater fish Clarias batrachus L 85 . The present study ascertained a positive correlation of antioxidant enzyme activities and DNA damage with pendimethalin. However, curcumin treatment increased antioxidant enzymes in the roots when supplied with pendimethalin indicating its protective role. Enhancement in SOD and CAT levels was observed in pendimethalin and curcumin-treated roots. Further, curcumin supply with pendimethalin reduced SOD and CAT levels. In comparison to control, these SOD and CAT levels were still higher. The increased SOD level shows the activation of the antioxidant system. SOD, CAT and GR plays important role in thwarting oxidative stress. SOD, the first line of defense against ROS, catalyzes free radicals to generate hydrogen peroxide and oxygen. Meanwhile, CAT is required to catalyze hydrogen peroxide. In addition, GR utilizes reduced glutathione to play a key role in the defense mechanism 86,87 . This study suggested a reduction in GR level pointed towards the sensitivity of the plants to pendimethalin stress. Similarly, SOD and CAT are less sensitive to pendimethalin stress in comparison   www.nature.com/scientificreports/ to GR. Similar responses were observed under the treatment of some herbicides in hairy fleabane 88 , rice 89 , wheat and maize 86 . The molecular docking studies showed that the 3-D structure of pendimethalin has the interactive capability to antioxidant enzymes and DNA molecules. The higher binding energy (− 8.23 kcal/mol) for CAT-complex indicated that the pendimethalin could highly perturb the 3-D structure of CAT irrespective to other SOD and GR. In contrary to molecular docking, the biochemical analysis showed that the enhanced activity of antioxidant enzymes may lead to increase more enzyme synthesis for reducing oxidative stress.
The nucleic acid interactions showed that the pendimethalin could also alter the DNA structure which was attached between adjacent DNA bases (G10-G11 and C6-A7). Such a binding mode of pendimethalin between bases of nucleic acids may lead to DNA intercalation (Snyder et al. 2004) causing genotoxicity 90,91 . It has been determined that pendimethalin selectively binds to G and C nucleotides in the B-DNA structure. The GC-rich sections of DNA can intercalate by minor groove binders and act as a DNA topoisomerase-I toxin 92 .
MDA level reflects the extent of membrane peroxidation resulting from ROS activity. The results revealed high MDA levels in roots treated with pendimethalin indicating an imbalance between ROS production and antioxidant defense leading to lipid peroxidation and oxidative stress 93 . The severity of lipid peroxidation was reduced when curcumin was supplied with pendimethalin. Similarly, exposure of watermelon to bensulfuronmethyl and penoxsulam herbicides enhanced lipid peroxidation 94 .

Conclusion
The present study elucidates that pendimethalin induced cytotoxic and genotoxic effects, thus inhibited the growth of A. cepa roots. Further, pendimethalin disturbed the normal function of the cell and increased lipid peroxidation and activity of antioxidant enzymes. Molecular docking results suggested that pendimethalin could control the activity of antioxidant enzymes by interacting with their residues. However, the application of curcumin with pendimethalin alleviated the toxicity of herbicide by regulating antioxidant enzyme defense. The correlation study also suggests that curcumin can reduce chromosomal abnormalities, DNA damage, antioxidant enzyme activities (SOD and CAT) and lipid peroxidation indicating its ability to minimize oxidative stress. Hence, the synergistic use of curcumin with herbicide: pendimethalin could minimize the deteriorating impact in plants. However, a detailed study is required to unravel the exact mechanism of curcumin action in the amelioration of herbicide toxicity.