Physiological and biochemical mechanisms of grain yield loss in fumitory (Fumaria parviflora Lam.) exposed to copper and drought stress

Soil contamination with heavy metals adversely affects plants growth, development and metabolism in many parts of the world including arid and semi-arid regions. The aim of this study was to investigate the single and combined effects of drought and copper (Cu) stresses on seed yield, and biochemical traits of Fumaria parviflora in a split – factorial experiment at Research Field of Payam-E-Noor university of Kerman during 2019. The collected seeds from two Cu contaminated regions were evaluated under drought and Cu (0, 50, 150, 300, and 400 mg/kg) stresses. Drought stress levels were depletion of 50% (D1), 70% (D2) and 85% (D3) soil available water. The individual effects of drought and copper stresses were similar to each other as both reduced seed yield. The highest seed yield was observed at Cu concentration of 50 mg/kg under non-drought stress conditions. The maximum values of malondialdehyde (0.47 µmol/g), proline (2.45 µmol/g FW), total phenolics (188.99 mg GAE/g DW) and total flavonoids (22.1 mg QE/g DW) were observed at 400 mg/kg Cu treatment. However, the strongest antioxidant activity (83.95%) through DPPH assay, and the highest total soluble carbohydrate (115.23 mg/g DW) content were observed at 300 and 150 mg/kg Cu concentration under severe drought stress, respectively. The highest amount of anthocyanin (2.18 µmol/g FW) was observed at 300 mg/kg Cu and moderate drought stress. The findings of this study showed a high tolerance of F. parviflora plant to moderate drought stress and Cu exposure up to 150 mg/kg by modulating defense mechanisms, where grain yield was slightly lower than that of control. The results could also provide a criterion for the selection of tolerance species like F. parviflora for better acclimatization under Cu mines and/or agricultural contaminated soils subjected to drought stress.


Results
The results of ANOVA showed highly significant (p < 0.01) effect of drought stress on all the studied traits (Table 1).The region had not significant effect on each of the traits, but there was a highly significant effect (p < 0.01) for zones on grain yield, proline, TSC, anthocyanins, TPC and TFD (Table 1).According to results of Table 1.Analysis of variance for the interaction effects of zones, Cu and drought stresses on F. parviflora under field condition.MDA malondialdehyde, TSC total soluble carbohydrate, TPC total phenolics content, TFD total flavonoids.*and ** are significant at p < 0.05 and p < 0.01, respectively.

Effects of zone on studied traits
The originated plants from Z 3 and Z 4 showed lower level of grain yield and higher levels of biochemical traits than Z 1 and Z 2 (Fig. 1).The grain yield decreased by an average of 115.43% in Z 3 and Z 4 compared to Z 1 and Z 2 (Fig. 1).The biochemical traits increased by an average of 116.44% for proline 115.63% for TSC, 119.34% for anthocyanin, 114.68% for TPC and 115.98% for TFD in Z 3 and Z 4 compared to Z 1 and Z 2 (Fig. 1).

Grain yield
The highest grain yield was observed at Cu 0 and Cu 50 (56.16g/m 2 in Z 1 , 56.51 g/m 2 in Z 2 , 55.98 g/m 2 in Z 3 and 54.67 g/m 2 in Z 4 ; Fig. 2) while the plants in the soil treated with Cu 400 showed the lowest grain yield in all zones (33.94 g/m 2 in Z 1 , 33.1 g/m 2 in Z 2 , 36.45 g/m 2 in Z 3 and 36.08 g/m 2 in Z 4 ; Fig. 2A).The concentrations of Cu 0 and Cu 50 were appropriate for grain yield, but with increasing Cu concentration, a reduction in grain yield was observed, which was significant (p ≤ 0.01).This reduction in Cu 150 (10%), was less than concentrations of Cu 300 (40% in Z 1 and Z 2 and 29% in Z 3 and Z 4 ) and Cu 400 (41% in Z 1 and Z 2 and 28% in Z 3 and Z 4 ) compared to the control (Fig. 2A).The regions of Z 3 and Z 4 showed a better grain yield under high concentrations of Cu (Fig. 2A).
The best zones for collecting F. parviflora seeds (tolerance to high concentrations of copper) for grain yield were identified as Z 3 and Z 4 zones.
In particular, exposure to Cu 50 was better for F. parviflora to withstand the drought stress tensions.The reduction of grain yield at this concentration (Cu 50 ) was less than other concentrations of copper (Cu 150 , Cu 300 , and Cu 400 ) in response to drought stress (D 2 and D 3 ).At Cu 50 , the grain yield in D 2 and D 3 decreased (on average in four zones), respectively, 24% and 40% (Fig. 2B) compared to the control, while the reduction of this trait in moderate and severe drought (on average in four zones) was 28% and 45% in Cu 150 and 43% and 55% in Cu 300 and Cu 400 (Fig. 2B) compared to the control.The highest value of grain yield in a combination of Cu and drought stresses was recorded at Cu 0 and Cu 50 under D 1 treatment in all zones (55.87 g/m 2 ).The lowest value was observed at Cu concentration of 400 mg/kg under D 3 treatment in four zones (17.02 g/m 2 ; Fig. 2B).

Malondialdehyde
Oxidative stress in term of high contents of MDA was observed in this study (Fig. 3).High concentrations of Cu in the soil significantly (p < 0.05) increased contents of MDA of F. parviflora leaves compared with control (Fig. 3A).MDA contents were increased averaged by 102% in Cu 50 , 104% in Cu 150 , 123% in Cu 300 and 125% in Cu 400 over all zones in the leaves (Fig. 3A).The maximum contents of MDA in leaves were in Cu 300 and Cu 400 (0.3 µmol/g in Z 1 , 0.28 µmol/g in Z 2 , 0.28 µmol/g in Z 3 and 0.31 µmol/g in Z 4 ; Fig. 3A) and the minimum contents were in Cu 0 and Cu 50 (0.21 µmol/g in Z 1 , 0.21 µmol/g in Z 2 , 0.24 µmol/g in Z 3 and 0.26 µmol/g in Z 4 ; Fig. 3A).Elevated Cu led to significant increase in MDA content by 108% at Cu 0 , 114% at Cu 50 , 116% at Cu 150, 114% at Cu 300 and 116% at Cu 400 (average in four zone; Fig. 3B) at moderate drought and 117% at Cu 0 , 123% at Cu 50 , 133% at Cu 150, 146% at Cu 300 and 152% at Cu 400 at severe drought (average in four zone; Fig. 3B).Cu 400 and severe drought showed the highest amount of MDA in all zones (0.47 µmol/g; Fig. 3B) and Cu 0 and Cu 50 under D 1 treatment showed the lowest one (0.21 µmol/g; Fig. 3B).

Total soluble carbohydrates
Increasing Cu concentration up to Cu 300 was associated with a significant (p < 0.05) increase in the contents of TSC (Fig. 4) of F. parviflora leaves.Compared to the control, the maximum TSC were in Cu 150 at all zones (65.81 mg/g DW in Z 1 , 66.53 mg/g DW in Z 2 , 67.74 mg/g DW in Z 3 and 65.69 mg/g DW in Z 4 ; Fig. 4A) and the minimum was at Cu 400 over all zones (39.95 mg/g DW in Z 1 , 42.16 mg/g DW in Z 2 , 45.17 mg/g DW in Z 3 and 44.56 mg/g DW in Z 4 ; Fig. 4A) .At 150 mg/kg of Cu, TSC increased by 151% compared to the control (Fig. 4A).Interaction of the stresses (Cu and drought) increased TSC content, by 142% at Cu 0 , 154% at Cu 50 , 155% at Cu 150 , 110% at Cu 300 and 113% at Cu 400 (average in four zone; Fig. 4B) in moderate drought and 161% at Cu 0 , 170% at Cu 50 , 172% at Cu 150 and 96% at Cu 300 and 90% at Cu 400 (average in four zone; Fig. 4B) in severe drought stress.Cu 150 under severe drought showed the highest amount of TSC in all zones (114.83mg/g DW; Fig. 4B) and Cu 400 under severe drought showed the lowest one over four zones (40.62 mg/g DW; Fig. 4B).

Proline
It was noticed that highest content of proline was observed under Cu 400 concentration in all zones (1.45 µmol/g FW in Z 1 , 1.42 µmol/g FW in Z 2 , 1.38 µmol/g FW in Z 3 and 1.39 µmol/g FW in Z 4 ; Fig. 5A) and the least content of proline was in Cu 0 in all zones (0.83 µmol/g FW in Z 1 , 0.85 µmol/g FW in Z 2 , 0.87 µmol/g FW in Z 3 and 0.9   A Cu₀(mg/kg) Cu₅₀(mg/kg) Cu₁₅₀(mg/kg) Cu₃₀₀(mg/kg) Cu₄₀₀(mg/kg) www.nature.com/scientificreports/µmol/g FW gm -2 in Z 4 ; Fig. 5A).Increasing Cu level till Cu 400 , significantly increased contents of proline in the leaves of F. parviflora.The averaged content of proline under Cu 300 and Cu 400 treatments was less (17%) than the averaged values for proline in Z 1 and Z 2 (Fig. 5A).Both drought and Cu treatments had an increasing effect on proline content (Fig. 5B).Moderate drought increased proline content by 123% at Cu 0 , 106% at Cu 50 , 119% at Cu 150, 132% at Cu 300 and 137% at Cu 400 compared to the control (average in four zone (Fig. 5B).Severe drought also increased proline content by 126% at Cu 0 , 108% at Cu 50 , 145% at Cu 150 , 157% at Cu 300 and 161% at Cu 400 compared to the control (average in four zone; Fig. 5B).The highest value of proline content in a combination of Cu and drought stresses, recorded at Cu 400 under D 3 treatment in all zones (2.27 µmol/g FW; Fig. 5B) but the lowest one was observed at Cu 0 under D 1 treatment in four zones (0.86 µmol/g FW; Fig. 5B).

Anthocyanin
In the present study, increasing levels of Cu concentrations in soil up to Cu 400 , enhanced anthocyanin content in the leaves (Fig. 6A).Compared to the control, the highest content of the anthocyanin was in Cu 400 over all zones (1.27 µmol/g FW in Z 1 , 1.29 µmol/g FW in Z 2 , 1.23 µmol/g FW in Z 3 and 1.23 µmol/g FW in Z 4 ; Fig. 6A) and the least one was in Cu 0 (0.86 µmol/g FW in Z1, 0.84 µmol/g FW in Z2, 0.88 µmol/g FW in Z3 and 0.90 µmol/g FW in Z4; Fig. 6A) .Anthocyanin content, increased 117% at Cu 0 , 104% at Cu 50 , 134% at Cu 150 , 168% at Cu 300 and 130% at Cu 400 in moderate drought and 126% at Cu 0, 112% at Cu 50 , 174% at Cu 150 , 140% at Cu 300 and 137% at Cu 400 in severe drought (averaged in all zones; Fig. 6B).The highest value of anthocyanin content was recorded at Cu 300 and moderate drought in all zones (2.06 µmol/g FW; Fig. 6B) and the minimum value was recorded at Cu 0 without drought in all zones (0.87 µmol/g FW, Fig. 6B).www.nature.com/scientificreports/

Total phenolics and total flavonoids contents
In the present study, TPC and TFD were also measured from the leaves of F. parviflora under elevated levels of Cu in the soil (Figs. 7 and 8).These results suggested that TPC and TFD significantly (p < 0.01) increased as the Cu level in the soil rises compared with the control.www.nature.com/scientificreports/Contents of total phenolic and total flavonoids in Z 1 and Z 2 were 11.76% and 15.5% respectively more than Z 3 and Z 4 under high concentration of Cu (300 and 400 mg/kg; Figs.7A and 8A).The minimum values were observed in the plants which grown without the Cu concentration in the soil (94.88 mg GAE/g DW in Z 1 , 94.57 mg GAE/g DW in Z 2 , 97.43 mg GAE/g DW in Z 3 and 98.83 mg GAE/g DW in Z 4 for TPC (Fig. 7A) and 9.42 mg QE/g DW in Z 1 , 9.37 mg QE/g DW in Z 2 , 9.91 mg QE/g DW in Z 3 and 10.13 mg QE/g DW in Z 4 for TFD (Fig. 8A).But these values increased continuously as the Cu level increases in the soil and the maximum contents of them were observed in Cu 400 (129.95mg GAE/g DW in Z 1 , 129.2 mg GAE/g DW in Z 2 , 122.35 mg GAE/g DW in Z 3 and 122.92 mg GAE/g DW in Z 4 for TPC (Fig. 7A) and 12.25 mg QE/g DW in Z 1 , 12.16 mg QE/g DW in Z 2 , 11.36 mg QE/g DW in Z 3 and 11.08 mg QE/g DW in Z 4 for TFD (Fig. 8A).In Cu 400 the contents were increased by 137% for TPC and 127% for TFD (compared to the control).
The effects of two stresses (Cu and drought) increased TPC in all zones and this increase was significant (Fig. 7).Elevated Cu increased TPC at moderate drought stress by averaged 124% in Cu 0, 123% in Cu 50, 127% in Cu 150, 129% Cu 300 and 129% in Cu 400 over all zones (Fig. 7B) and in severe drought by averaged 131% in Cu 0, 129% in Cu 50, 131% in Cu 150, 139% in Cu 300 and 140% in Cu 400 over all zones (Fig. 7B).It was noticed that minimum values of TPC was recorded in a stress-free environment (Cu 0 and D 1 ; 96.42 mg GAE/g DW; Fig. 7B).The highest TPC was in Cu 400 under severe drought over all zones (173.96mg GAE/g DW; Fig. 7B).
The contents of TFD in the leaves of F. parviflora was increased with moderate drought by 119% at Cu 0 , 130% at Cu 50 , 137% at Cu 150, 152% at Cu 300 and 148% at Cu 400 and increased with severe drought by 150% at Cu 0 , 145% at Cu 50 , 169% at Cu 150 , 171% at Cu 300 and 174% at Cu 400 (Fig. 8B).The highest value of TFD in a combination of Cu and drought stresses in all zones was recorded at Cu 400 under D 3 treatment (20.23 mg QE/g DW ; Fig. 8B).The lowest TFD was observed at Cu concentration of 0 mg/kg under D 1 treatment in four zones (9.7 mgQE/g DW; Fig. 8B).

Antioxidant activity
Antioxidant activity of F. parviflora leaves were evaluated under various levels of Cu (0, 50, 150, 300 and 400 mg/kg; Fig. 9A).It was noticed that the antioxidant capacity increased when plants were subjected to high concentration of copper (400 and 500 mg/kg).Our results depicted that antioxidant capacity increased in the leaves by 102% in Cu 50 , 109% in Cu 150 , 114% in Cu 300 and 113% in Cu 400 compared to the control.The maximum antioxidant capacity was observed in the plant grown under Cu level of 300 and 400 mg/kg (66.72% in Z 1 , 68.26% in Z 2 , 67.61% in Z 3 and 67.12% in Z 4 ; Fig. 9A) and the minimum antioxidant capacity was observed in a stress-free environment (58.71% in Z 1 , 59.81% in Z 2 , 59.11% in Z 3 and 58.11% in Z 4 ; Fig. 9A).Antioxidant capacity was increased by interaction of drought and Cu hence the interaction between the two stresses was significant (Fig. 9B).Antioxidant capacity was increased by 110% in Cu 0, 108% in Cu 50, 115% in Cu 150, 116% in Cu 300 and 115% in Cu 400 in moderate drought and 110% at Cu 0, 112% in Cu 50, 117% in Cu 150, 118% in Cu 300 and 116% in Cu 400 in severe drought (average in four zone, Fig. 9B).The lowest value of antioxidant capacity was recorded at Cu 0 and D 1 treatment (58.93%; Fig. 9B) while the highest one was recorded at Cu 300 and severe drought (78.34%;Fig. 9B).

Heat map analysis
The Heat Mam graph was drawn to gain a better understanding of the treatments clustering (Cu × Z) based on their different studied traits in two different non-drought (A) and drought-stress (B) environments (Fig. 10).
Clearly, the treatments (Cu × Z) may be can be categorized into 7 different ones, which according map legend, seven different color represented different ranges for traits values from Z 1 Cu 1 to Z 4 Cu 5 , demonstrating the relative value of each under these 20 different treatments.As presented in heat map graphs (Fig. 10 A and B), the synergism effects of drought stress were significantly obvious on different biochemical traits under Cu stress.The comparison between two condition including non-drought (Fig. 10A) and drought-stress (Fig. 10B) implied at significant increase in MDA, DPPH, TPC, TFD, and anthocyanin content, specially under Cu 300 and Cu 400 (mg/kg) drought stress (Fig. 10B) Vice versa, a sharp decrease observed in grain yield under higher mentioned Cu concentrations (Fig. 10B).

Principle component analysis
A biplot, as a result of Principle Component Analysis (PCA) done as an efficient multivariate tool for the interpretation of data, was constructed based on the first (PC 1 ) and (PC 2 ) principal components for different studied traits under the interaction of Cu × Z (Zone) treatments under non-drought and drought-stress conditions (Fig. 11A,B).Based on the studied traits represented in the biplot, the first (PC 1 ) and second (PC 2 ) components explained about 72.2% and 16.3% of the total variance of the traits under non-drought condition, respectively (Fig. 11A); thus, both PCs cumulatively described vc 88.5% of the total variance of all the traits analyzed.The highest values for TFD, TPC, DPPH, anthocyanin and proline was identified under the Cu × Z treatments in blue color croup (Fig. 11A).On the other hand, the treatment in red group had the highest positive values for grain yield (Fig. 11A).Finally, the Cu × Z treatments in green color had the highest values for TSC (Fig. 11A).Under water-deficit condition, in 82.6% and 12.8% of the total variance was explained by PC 1 and PC 2 , respectively (Fig. 11B).Therefore, both PCs explained 95.4% of the total variance in all the traits investigated.Below drought stress, the plant under the Cu × Z treatments in blue color group, had the highest TFD, TPC, MDA and proline content (Fig. 11B).The plants under Cu × Z treatments in red color group had the highest TSC (Fig. 11B).Finally, the highest grain yield was observed under Cu × Z treatments in green color (Fig. 11B).

Discussion
Abiotic stresses are major constraints for plants growth worldwide 27,28 and exacerbate yield loss under changing climatic conditions 28 .Heavy metals and drought are generally considered important stresses in plant production systems 29 .Copper, as micronutrient is essentially required for redox-active transition and as a cofactor for several enzymes which involved in many biochemical processes including photosynthesis, respiration and cell wall metabolism 30,31 .It has been acknowledged that Cu plays an important role in drought tolerance in different plant species through promoting the photosynthesis pigments and contribute to organic matter accumulation under water deficit conditions 30,32 .The soil pollution with heavy metals has become a serious environmental problem around the world due to industrialization and soil pollution 26 , specially under arid and semi-arid regions.Metal-rich soils often have low organic matter content, resulting in low water-holding capacity and subsequent reduction in hydraulic and stomatal conductance 33 .Therefore, the collection of these responses may aggravate the drought stress 24 .In this respect, substantial efforts have widely been made in previous years to identify the independent effects of heavy metals and drought stress on plants, but little attention has been paid to the interactive effects of these two stresses on plants.Here, the single and combined effects of Cu and drought stress were evaluated on different biochemical traits of F. parviflora which their seeds were collected from four different zones (Z 1 , Z 2 , Z 3 and Z 4 ) from two copper mines.Regarding the effects of different zones (Z 1 , Z 2 , Z 3 and Z 4 ) on grain yield and biochemical traits of the plants grown in the field, it should be acknowledged that under both drought and non-drought conditions, the originated plants from Z 3 and Z 4 regions showed lower levels of grain yield and higher levels of physiological traits than Z 1 and Z 2 under high concentrations of Cu (Cu 300 and Cu 400 ).This finding can be resulted from the higher genetic and/or edaphic acclimatization of F. parviflora seeds originated from Z 3 and Z 4 regions to combat the Cu toxicity stress.It the other hand, Z 3 and Z 4 were identified as the best areas for collecting F. parviflora seeds and cultivation on contaminated soils with Cu to achieve expected grain yield.In the present study, the effects of Cu stress showed adverse effects on grain yield of the F. parviflora.The application of Cu 0 and Cu 50 were suitable for grain yield, but with increasing Cu concentrations, the grain yield significantly (p < 0.01) was decreased.Under Cu 50 × drought stress interactions, Cu showed adverse effect on drought stress compared to the control.Here, low concentrations of Cu (50 and 150 mg/kg) were able to improve grain yield loss under different levels of drought stress (D 2 and D 3 ).It could be concluded that under low Cu concentrations, it acts as a micronutrient, with triggering effects to increase plant tolerance to drought stress through the increase in water holding capacity and biomass accumulation 30,32 .This finding was similar to positive effects of Zn and Fe microelements on increase of grain yield in Thymus vulgaris 34 and Ocimum basilicum 35 .Combined effects of severe drought and high concentrations of Cu (300 and 400 mg/kg) may affect nutrients uptake and allocation within fumitory tissues due to disruption of water homeostasis, ionic allocation and cell permeability barriers, which led to a significant decrease in grain yield 12 .Also, under such condition, defense mechanisms could only sustain plant until the end of the growing season and grain yield at these levels of stress severely reduced.High concentrations of heavy metals reduce the synthesis of chlorophyll, photosynthetic activities, inhibiting activities of the Calvin cycle and the activity of enzymes related to soluble carbohydrates production, which causes a decrease in grain yield in areas with high concentrations of Cu 36,37 .
Malondialdehyde (MDA) often used as a marker for measure of oxidative stress in cell membranes 38 .Under different oxidative stresses, membrane peroxidation led to an increase in the content of lipid peroxidation which produces MDA in the tissues 38 .Here, Cu toxicity may exert in electron transport and respiration like subcellular organelle functions, which generate hydroxyl radical due to the decomposition of H 2 O 2 and subsequent increase in lipid peroxidation of cell membranes 39 .It was observed that F. parviflora had no significant increase in MDA content under interactions of lower Cu (Cu 50 and Cu 150 ) and drought stresses (D 2 and D 3 ), but the plants under drought and high Cu concentrations (Cu 300 and Cu 400 ) showed higher MDA content rather than their content under single stresses of drought (D 1 and D 2 ) and or Cu (50, 150, 300 and 400 mg/kg), which demonstrate a synergism effects of Cu and drought stress on lipid peroxidation of membranes rather than singular stresses.
Plant adaptation to oxidative stress is associated with different metabolic adjustments such as accumulation of soluble carbohydrates 40 , which reveal two defensive mechanisms including osmotic adjustment and cellular compatibility 41 .According to the results, increasing Cu concentration up to 150 mg/kg under moderate and severe drought stress showed modulatory roles on increase in TSC under drought stress conditions, however, exposure to higher concentration (Cu 300 and Cu 400 ) aggravates drought stress in an additive manner, which led to decrease in the production of TSC, making the plants more vulnerable to drought stress.This phenomenon may be linked to this fact that under low-medium concentrations of heavy metals like Cu, plants produce osmolytes such as TSC 14,42 , but at higher concentration (Cu 400 ) increase generation of toxic free radicals which induce oxidative stress and subsequent inhibition in the production of TSC 43 .This could be due to the reduction of CO 2 fixation in heavy metals treated plants at high concentrations, but in lower concentrations, the photosynthetic measurements were not affected following exposure, and subsequently cause a general increase in carbohydrate content 44 .Similarly, the decrease in TSC was previously reported under the high concentrations of Cu stress in Trigonella foenum-graecum 39 and Kandelia obovate 45 .In concentration of Cu 150 , the content of TSC was significantly higher than single stresses of drought and Cu and control treatment (Cu 0 D 1 ), which led to transfer of more carbohydrates to grains and produce greater seed yield under combined stresses rather than single and/or control conditions.All morpho-physiological and biochemical changes under stress conditions are occurred for better compatibility of F. parviflora to employed stresses, which are associated with synthesis of TSC compounds.
Proline as an important compatible osmolyte in plants, plays multifarious roles in plant compromising adaptation, recovery, and signaling when it comes to combating stress 46 .Higher accumulation of proline, protects plants with maintaining cell turgor under stress conditions.Under drought stresses the occurrence of water imbalance leads to increase in the content of proline through osmotic regulation function 14,46,47 , but under heavy metals stress, proline increase the tolerance of plants through detoxifying the heavy metal effects in the cytoplasm and regulating the intracellular redox homeostasis potential by osmotic adjustments 48 .
In the present study, a significant (p < 0.01) increase was occurred in proline content with increasing the levels of Cu concentration, similar to the increase in proline content under Cu stress for Cinnamomum camphora 27 and Astragalus tragacantha 49 .Here, the proline content of F. parviflora showed a significant increase after exposure to Cu treatment, because the applied concentration of Cu (up to 400 mg/kg) is under the tolerance threshold of fumitory ecotype.The secondary effects of drought (D 1 and D 2 ) and Cu (up to 400 mg/kg) stresses led to more accumulation of proline compared to the control.
Anthocyanins as a subclass of flavonoids 50 have an important role in inducing and/or modulating array of different environmental agents including heavy metals 1,51 for ROS quenching 52 , photo-protection and stress signaling 53 .Here, single effects of Cu (50 to 400 mg/kg) stress led to increase in anthocyanins content, which was similar to the other heavy metals application such as arsenic in Lemna gibba L. 54 Cu, Zn, Mn, Pb and Hg in Arabidopsis 55 , Cu and Zn in Capsicum annum L. 56 , and in contrast with the report in annual Halophytes 51 .The chemical reactivity of anthocyanins depends on their capacity to metal chelation 51 .The most common metals that can form complexes with anthocyanins are Cu, Fe, Mg, Sn and K 57 .The combined effect of drought stress and Cu has not been previously reported on anthocyanin content in medicinal plant species.Under the combination of all drought stress levels with Cu 50 (D 2 × Cu 50 and D 3 × Cu 50 ) the anthocyanin content was more than that of single effects of drought stress (D 1 × Cu 50 ), showing that Cu could mitigate the negative effects of drought in plants (Waraich et al., 2011).Under interaction of higher concentrations of Cu (Cu 150 , Cu 300 , and Cu 400 ) and drought stress (D 2 and D 3 ), the dose-dependent trend was observed on the changes of anthocyanin.Hovered, under moderate and severe drought conditions, the highest content of anthocyanin was observed at 300 mg/kg (2.1 µmol/g FW) and 150 mg/kg (2.1 µmol/g FW) of Cu application, respectively.Although the greater degradation of chlorophyll molecules caused an increase in protective pigments of anthocyanin under drought stress 58  www.nature.com/scientificreports/ the suppressing effects of high concentration of Cu (300 and 400 mg/kg) on anthocyanin content was observed under drought stresses treatments (D 2 and D 3 ).Plants increase the contents of two major groups of SMs (phenolics and flavonoids) as effective antioxidants to avoid the oxidative damage which inhibit the over-generation of ROS under heavy metal stresses 52,59,60 or in combination with other environmental stresses 12 .Phenoilcs and flavonoids biosynthesis pathways provide compounds involved in defense responses against a certain type of stresses such as heavy metals 52,61 .Phenolic compounds, could be also substrates for different peroxidases, which are the first line of defense against Cu toxicity 37 .Under heavy metal stresses, flavonoids enhance the process of metal chelation, which may reduce the level of harmful hydroxyl radicals in plant cells 62 .In the present study, increasing drought stress intensity led to an increase in TPC and TFD, which were consistent with the previous reports in Carthamus tinctorius 63 and Eruca sativa 64 .Also, the increase in Cu concentrations up to Cu 400 , resulted in an increase in TPC and TFD.These findings were similar to the previous reports on tomato 65 , Mentha spicata 12 , wheat 61 , Withania somnifera L. 66 , Gymnema sylvestre 67 and Erica andevalensis 68 .In contrary, a dose-dependent decrease was reported in TPC under the heavy metal stresses (Pb and Cu) in Citrus aurantium L. and cadmium in Prosopis glandulosa 69 .These discrepancies trends could be attributed to the differences in heavy metal type and its intensity, plant developmental stages and differences in plant species 63 .Here, the TPC and TFD of the F. parviflora showed upward trend under different interaction of Cu and drought stress compared to the single effects of drought and Cu levels.Therefore, incidence of combined stresses of Cu and drought have not necessarily adverse effects on SMs metabolism, because F. parviflora plants were able to activate the production of SMs to defend oxidative stress damage via ROS scavenging.
The effects of moderate drought stress on antioxidant activity under all levels of Cu treatment were not statistically significant compared to the control, but under severe drought (D 3 ), the increase in antioxidant activity (DPPH%) of the F. parviflora leaves was higher under the both drought and Cu stresses than single treatments of Cu and/or drought, which demonstrated at synergism effects of severe drought and higher Cu concentrations on increasing the antioxidant activity.Conclusively, the findings of this study showed that Cu (at 50 mg/kg) as a micronutrient reduced the deleterious effects of drought stress in F. parviflora.Under interaction of D × Cu (150 mg/kg), and D × Cu (300 mg/kg), F. parviflora could withstand the stress by using some defense mechanisms such as increase in the contents of total soluble carbohydrates and anthocyanins, respectively.Under higher concentrations (Cu 300 and Cu 400 ), Cu addition could increase tolerance of F. parviflora to the osmotic stress caused by drought mainly by increasing the contents of total phenolics and total flavonoids.Furthermore, the ascending trend in the values of total phenolics and flavonoids was observed (from Cu 0 to Cu 400 ), which demonstrated at high threshold and pre-dominant role of non-enzymatic antioxidants in fumitory tolerance under combined stresses of Cu (up to 500 mg/kg) and drought.

Conclusions
The results from this research could help our understanding of how medicinal plants act to polluted regions with Cu as a heavy metal under drought conditions.The findings supported the use of F. parviflora as a suitable species for contaminated soils with low and moderate Cu (50-150 mg/kg) under moderate drought stress.Based on these findings, it can be concluded that F. parviflora can cope with Cu stress due to active antioxidant defense system.The results could provide a criterion for the selection of tolerance species for better acclimatization under Cu mines and/or agricultural contaminated soils.More researches are required to evaluate the phyto-extraction of Cu in plants grown in the contaminated soils of arid regions.

Plant materials and treatments
Two mineral regions (R 1 : Askary and R 2 : Rabor) including copper were studied from Kerman, Iran (Fig. S1), which their climatic and soil characteristics is presented in Table 2.Each mine (R 1 and R 2 ) were divided into four zones based on the Cu concentration in these four zones (Z 1 :50 mg/kg, Z 2 : 150 mg/kg, Z 3 : 300 mg/kg and Z 4 : 400 mg/kg).Then, seeds of the F. parviflora were collected from the growing plants four different zones for the aim of field experiment.The experimental design was conducted as a split factorial experiment based on a randomized complete block design (RCBD) design (replication = 3) was carried out at the research field of Payam -E-Noor University of Kerman (longitude 57° 06ʹ 21.22ʺ and latitude 30° 15ʹ 32.89ʺ, 1766 m ASL), Iran in 2019.Different levels of drought stresses (D 1 : non-stress, D 2 : moderate stress and D 3 : severe stress) were considered as ) were considered as sub factors.Three number of rows were planted in each plot (2.5 m × 2 m).The distance between and within rows were 50 cm and 15 cm, respectively.Plants were irrigated uniformly under different conditions (D 1 , D 2 and D 3 ) until the 6-8 leaves stage and after that time drought and Cu stresses were established.The irrigation time was estimated according to cumulative evaporation (CE) (mm) from CLASS A pan.Under D 1 treatment, the irrigation was done when cumulative evaporation (CE) reached to 75 (± 4) from evaporation pan (Class A).Under D 2 treatment, the irrigation was done after 110 (± 4) mm CE.Under D 3 treatment, the irrigation was done after 135 (± 4) mm CE.A pumping station via polyethylene pipes and a volumetric counter were used for irrigation system.The soil available water (SAW) is equivalent to difference between the volumetric water content at the permanent wilting point ( θ PWP ) and at field capacity ( θ FC ).In this study, volumetric soil water content at FC (-0.03 MPa) and PWP (-1.5 MPa) were 25% and 15%, respectively.The levels of irrigation treatments including D1, D2 and D3 were equivalent to 50%, 70% and 85% of depletion from SAW.The amount of irrigation water applied calculated as I = (θFC −θ PWP )Z•ρ b 100 , which, I is irrigation depth (cm), θ FC is soil gravimetric moisture percentage at field capacity (%), θ PWP is soil gravimetric moisture percentage at irrigation time (20%), Z is root zone depth (cm) (0.3 m), and ρ b is soil bulk density in the root zone (1.4 g cm -3 ) 70 .For applying Cu stress, final concentrations of Cu (0, 50, 150, 300, and 400 mg kg -1 ) were prepared by dissolving different amounts of CuSO 4 .5H 2 O (Merck.Com., Germany) in a specific amount of distilled water and then sprayed on the soil surface.Then, the soil of each plot was mixed uniformly.Different biochemical traits were measured at full maturity stage based on three replications for each treatment.

Malondialdehyde determination
For estimation malondialdehyde (MDA) content, 3 mL of 0.5% TBA and 2 mL of extraction solution were aggressively mixed together for this experiment 63 .The mixture was heated for 30 min at 95 °C in a water bath of constant temperature before cooling to room temperature with ice.The supernatant was discovered at wavelengths of 450, 532, and 600 nm after 15 min of centrifugation at 5000×g.

Proline determination
The content of proline was measured according to Bates et al. 71 .The fresh leaf samples (0.2 g) were first given 3 ml of sulphosalicylic acid (3% w/v).The mixture was then centrifuged for 15 min at 18,000×g.Glacial acetic (CH 3 COOH) (2ml) and the ninhydrin (C 9 H 6 O 4 ) (2ml) reagent were added to the test tubes after the supernatant (2 ml) was transferred to a fresh tube.The reaction mixture was heated for 60 min in a water bath before cooling on ice.Then 4 ml of toluene was then added, and it was left at room temperature for 30 min.After the separation of upper phase by shaking the tubes for 15 s, a spectrophotometer (Unico-UV 2100, USA (was used to detect its absorbance at 520 nm.

Total soluble carbohydrates assay
A test tube containing 2 mL of a carbohydrate solution and 1 mL of a 5% aqueous phenol solution is used to measure the total soluble carbohydrates (TSC).The liquid is then quickly added to 5 mL of pure sulfuric acid.The test tubes are vortexed for 30 s after standing for 10 min, and then they are placed in a water bath at room temperature for 20 min to allow for color development.After that, a spectrophotometer records the light absorption at 490 nm 72 .

Anthocyanins determination
Fresh leaf tissues (0.1 g) were homogenized in 2 mL of acidified methanol (1% HCI) in the initial stage.This was done at room temperature 73 .The entire extract was centrifuged for 25 min at 4000 rpm after one day.Then, using a spectrophotometer and a wave length of 511 nm, the anthocyanin (Ant) content was calculated using the extinction coefficient of raphanusin (33,000 M −1 cm −1 ).

Total phenolics content determination
The Folin-Ciocalteu technique was used to estimate the total phenolics content (TPC) 74 .Briefly, 2.3 mL of deionized water was added to a 15-mL tube before 0.1 mL of the extract sample was added.Folin-Ciocalteu reagent (1:1) was added to the mixture in a volume of 0.4 mL.Then, 1.2 mL of 20% sodium carbonate was then added after the mixture had been allowed to stand at room temperature for 7 min.For 60 min, the response was permitted to proceed.A Unico-UV 2100 (USA) spectrophotometer calibrated with gallic acid (GA) was used to measure absorbance at 765 nm against a blank.

Total flavonoids determination
The colorimetric AlCl 3 method was used to determine each fraction's total flavonoid (TFD) concentration 75 .Separately, 1.5 mL of methanol, 0.1 mL of 10% aluminum chloride, 0.1 mL of 1 M potassium acetate, and 2.8 mL of deionized water were combined with 0.5 mL of the properly diluted sample solutions.This mixture was then left at room temperature for 30 min.A Unico-UV 2100 (USA) spectrophotometer was used to test the reaction mixture's absorbance at 415 nm.Quercetin equivalents (QE) per gram of dry extract were used to measure the results.

AFigure 4 .AFigure 5 .
Figure 4.The effect of Cu stress on total soluble carbohydrates of F. parviflora under field conditions (A).The means for the interaction effects of Cu and drought stresses on total soluble carbohydrates (B).Different letters for each figure indicate significant differences at p < 0.05.Copper concentration (mg/kg).Seed originated zones Z 1 : Zone 1. Z 2 : Zone 2. Z 3 : Zone 3. Z 4 : Zone 4. D 1 : Control, D 2 : moderate drought stress; D 3 : severe drought stress.

AFigure 8 .
Figure 8.The effect of Cu stress on total flavonoids of F. parviflora under field conditions (A).The means for the interaction effects of Cu and drought stresses on total flavonoids (B).Different letters for each figure indicate significant differences at p < 0.05.Copper concentration (mg/kg).Seed originated zones Z 1 : Zone 1. Z 2 : Zone 2. Z 3 : Zone 3. Z 4 : Zone 4. D 1 : Control, D 2 : moderate drought stress; D 3 : severe drought stress.

AFigure 9 .
Figure 9.The effect of Cu stress on antioxidant activity of F. parviflora under field conditions (A).The means for the interaction effects of Cu and drought stresses on antioxidant activity (B).Different letters for each figure indicate significant differences at p < 0.05.Copper concentration (mg/kg).Seed originated zones Z 1 : Zone 1. Z 2 : Zone 2. Z 3 : Zone 3. Z 4 : Zone 4. D 1 : Control, D 2 : moderate drought stress; D 3 : severe drought stress.