Biochemical and physiological responses of soybean [ Glycine max ( L . ) Merrill ] to nickel toxicity

Despite the crucial role of nickel (Ni) in the plant metabolism, small increases in its concentration can cause leaf tissues injury. This study identified the highest dose of Ni foliar-applied that does not cause toxicity to soybean plants. Plants were sprayed with five Ni doses (0, 30, 60, 120, and 240 g·ha-1). At 1, 3, and 5 days after spray (DAS), the malondialdehyde (MDA), superoxide (O2 -), hydrogen peroxide (H2O2), and photosynthetic pigments concentrations, antioxidant enzymes activities, and gas exchange and chlorophyll (Chl) a fluorescence parameters were determined. Symptoms of Ni toxicity started at 120 g·ha-1 Ni and intense foliar necrosis occurred at 3 DAS. The concentrations of O2 -, H2O2, and MDA were significantly higher by 49% at 3 DAS, 47% at 3 DAS, and 19% at 5 DAS, respectively, for plants sprayed with 120 g·ha-1 Ni and by 48% at 3 DAS, 48% at 3 DAS and 18% at 5 DAS, respectively, for plants sprayed with 240 g·ha-1 Ni. Higher antioxidant enzymes activities and lower Chl a and Chl b concentrations occurred for plants sprayed with either 120 and 240 g·ha-1 Ni compared to the other Ni doses. Decrease on energy destined to photochemical process [Y(II)] (8 and 8% at 5 DAS) and increase on the dissipation of energy by the nonregulated process [Y(NO)] (15 and 15% at 5 DAS) occurred for plants sprayed with 120 and 240 g·ha-1 Ni, respectively. The Ni doses above 120 g·ha-1 promoted oxidative stress to the plants and affected the functionality of their photosynthetic apparatus. Doses below 60 g·ha-1 had a low risk of toxicity to plants without causing any biochemical or physiological damage.

A.M. Einhardt et al. (Chen et al. 2009;Reis et al. 2017). High concentrations of Ni can promote an increase in the accumulation of reactive oxygen species (ROS), which can cause damage to cell membranes (Chen et al. 2009).
Considering the pivotal role of Ni in the metabolism of soybean plants, including the increase of their resistance against diseases and the risk of its toxicity, this study aimed to identify the maximum Ni dose that could not cause toxicity to soybean leaf tissues by examining some alterations at the biochemical and physiological levels.
Soybean seeds from the cultivar TMG135, containing 0.54 ± 0.03 µg Ni per seed (extracted by the nitric-perchloric digestion method [Zasoski and Burau 1977] and determined by inductive coupled plasma optical emission spectrometry [ICP-OES]) were sown in sand washed with HCl 1N. Seven days after sowing, five seedlings were transplanted to each 5 L plastic pots and cultivated in hydroponic system containing nutrient solution of Hoagland and Arnon (1950) with aeration. The nutrient solution was changed every four days and the pH adjusted to 6.0 daily. Plants were kept in a greenhouse (temperature of 28 ± 3 °C, relative humidity of 80 ± 5% and natural radiation). Plants at the V4 growth stage (three trifoliate expanded leaves) were sprayed with solutions (9 mL per plant) of NiSO 4 ·6H 2 O at the concentrations of 0, 74.6, 149.2, 298.4 and 596.8 mg·L -1 , equivalent to 0, 30, 60, 120 and 240 g·ha -1 Ni, considering a plant density of 200,000 plants per hectare. For biochemical analysis, the first three trifoliate leaves of two plants from the replication of each treatment were collected at 1, 3 and 5 days after spray (DAS). Leaf samples were collected in liquid nitrogen and stored at -80 °C until further analysis. Cellular oxidative damage was estimated based on the production of total 2-thiobarbituric acid reactive substances and expressed as equivalents of malondialdehyde (MDA) according to Hodges et al. (1999). The concentrations of superoxide (O 2 -) and hydrogen peroxide (H 2 O 2 ) were quantified according to Chaitanya and Naithani (1994) and Debona et al. (2012), respectively. The activities of superoxide dismutase (SOD, EC 1.15.1.1), catalase (CAT, EC 1.11.1.6), and ascorbate peroxidase (APX, EC 1.11.1.1) were determined according to the methodologies described by Debona et al. (2012). The gas exchange and chlorophyll (Chl) a fluorescence parameters, as well as the concentration of photosynthetic pigments, were evaluated on the second trifoliate leaf of each plant at 1, 3, and 5 DAS. The photosynthetic pigments of six leaf-discs (0.8 cm³ each) were extracted with dimethyl sulfoxide saturated with calcium carbonate (Santos et al. 2008). The concentrations of Chl a, Chl b and carotenoids were calculated based on the absorbance at 480, 649 and 665 nm (Sumanta et al. 2014). The Chl a fluorescence parameters maximum photosystem II quantum efficiency (F v /F m ), photochemical yield [Y(II)], yield for dissipation by down-regulation [Y(NPQ)], yield for other nonphotochemical (nonregulated) losses [Y(NO)] and apparent electron transport rate (ETR) were obtained by using the Imaging-PAM image fluorometer and the Imaging Win software MAXI version (Heinz Walz GmbH, Effeltrich, Germany) following the procedures described by Fagundes-Nacarath et al. (2018) fixing in 5 min the time of actinic photon irradiance to obtain the steady-state fluorescence yield. The gas exchange parameters rate of net CO 2 assimilation (A), stomatal conductance to water vapor (g s ), transpiration rate (E) and internal CO 2 concentration (C i ) were determined in the lateral leaflet of the second trifoliate leaf between 09:00 and 12:00 h, when A was at its maximum, under artificial and saturating photon irradiance (1200 μmol·m -2 ·s -1 ) and an external CO 2 concentration of 400 μmol·mol -1 using a portable open-system infrared gas analyzer (LI-6400, LI-COR Inc., Lincoln, NE, USA). All measurements were performed by setting the block temperature at 25 °C.
The experiment was arranged in a completely randomized design with five treatments (control [plants sprayed with water] and plants sprayed with 30, 60, 120 and 240 g·ha -1 Ni) with four replications. Each experimental unit consisted of a plastic pot containing five plants. Data from the variables and parameters evaluated were checked for normality and homogeneity of variance and then submitted to analysis of variance. Means of treatments were compared by Tukey's test (p ≤ 0.05) by using the Minitab software v.18.
No symptoms of Ni toxicity were reported from plants sprayed with doses up to 60 g·ha -1 Ni ( Fig. 1a-c). Irrespective of the evaluation time, soybean plants exhibited symptoms of Ni toxicity starting at 120 g·ha -1 Ni. Necrotic areas were observed on the leaflets of plants sprayed with 120 and 240 g·ha -1 Ni, mostly at 5 DAS ( Fig. 1d-e). The O 2 concentration was significantly higher by 49 and 48%, respectively, for plants sprayed with 120 and 240 g·ha -1 Ni at 3 DAS and by 42% for plants sprayed with 240 g·ha -1 Ni in comparison to that plants sprayed with water ( Fig. 2a). At 3 DAS, H 2 O 2 concentration was higher by 47 and 48%, respectively, for plants sprayed with 120 and 240 g·ha -1 Ni in comparison to plants sprayed with water (Fig. 2b). The MDA concentration at 5 DAS was higher by 19 and 18%, respectively, for plants sprayed with 120 and 240 g·ha -1 Ni in comparison to plants sprayed with water ( Fig. 2c). Superoxide dismutase activity significantly increased at 3 DAS by 57 and 54% for plants sprayed with 120 and 240 g·ha -1 Ni, respectively, and at 5 DAS by 37% for plants sprayed with 120 g·ha -1 Ni in comparison to plants sprayed with water ( Fig. 2d). Catalase activity significantly increased at 1 DAS by 26 and 25% for plants sprayed with 120 and 240 g·ha -1 Ni, respectively, and at 3 DAS by 38% for plants sprayed with 240 g·ha -1 Ni in comparison to plants sprayed with water ( Fig. 2e). Ascorbate peroxidase activity significantly increased at 3 DAS by 30 and 32% and at 5 DAS by 32 and 31% for plants sprayed with 120 and 240 g·ha -1 Ni, respectively, in comparison to that plants sprayed with water ( Fig. 2f). Irrespective of the evaluation time, the Chl a, Chl b, and Chl a + b concentrations decreased for plants sprayed with 120 and 240 g·ha -1 Ni, except for Chl b at 1 DAS for the dose of 120 g·ha -1 Ni (Fig. 3a-c). Carotenoid concentration decreased by 19 and 14% at 3 and 5 DAS, respectively, for plants sprayed with 120 g·ha -1 Ni in comparison to plants sprayed with water ( Fig. 3d). Images of Chl a fluorescence on the leaflets obtained from plants sprayed with water or sprayed with 30 and 60 g·ha -1 Ni did not show any difference between them regarding color patterns for the parameters maximum photosystem II F v /F m , Y(II), Y(NPQ) and Y(NO) (Fig. 4). Alterations in the images of Chl a fluorescence parameters were observed on the leaflets of plants sprayed with 120 and 240 g·ha -1 Ni in comparison to leaflets of plants sprayed with water (Fig. 4). The F v /F m was significantly lower by 8 and 9% at 1 DAS, by 8 and 9% at 3 DAS and by 8 and 8% at 5 DAS for plants sprayed with 120 and 240 g·ha -1 Ni, respectively, in comparison to plants sprayed with water (Fig. 5a). Similarly to F v /F m , Y(II) significantly decreased by 6 and 7% at 1 DAS, 8 and 8% at 3 DAS and 8 and 8% at 5 DAS, respectively, for plants sprayed with 120 and 240 g·ha -1 Ni in comparison to plants sprayed with water (Fig. 5b). No significant difference was observed for Y(NPQ) among the treatments (Fig. 5c). Regarding Y(NO), there were significant increases of 8 and 7% at 1 DAS, 12 and 14% at 3 DAS and 15 and 15% at 5 DAS, respectively, for plants sprayed with 120 and 240 g·ha -1 Ni in comparison to plants sprayed with water ( Fig. 5d). Electron transport rate was significantly lower by 10 and 11% at 1 DAS, 10 and 11% at 3 DAS and 9 and 10% at 5 DAS, respectively, for plants sprayed with 120 and 240 g·ha -1 Ni in comparison to plants sprayed with water (Fig. 5e). Mainly, after 3 DAS, there were decreases for A, g s and E values and increase of C i values for plants sprayed with 120 and 240 g·ha -1 Ni in comparison to plants sprayed with water or sprayed with 30 and 60 g·ha -1 Ni (Fig. 6A-D). Net CO 2 assimilation values were lower by 21 and 18% at 3 DAS and by 19 and 17% at 5 DAS, respectively, for plants sprayed with 120 and 240 g·ha -1 Ni in comparison to plants sprayed with water (Fig. 6a). Regarding g s , there were significant decreases of 18 and 20% at 3 DAS and 27 and 26% at 5 DAS, respectively, for plants sprayed with 120 and 240 g·ha -1 Ni in comparison to plants sprayed with water (Fig. 6b). Internal CO 2 concentration values were higher by 8 and 8% at 3 DAS and by 8 and 9% at 5 DAS, respectively, for plants sprayed with 120 and 240 g·ha -1 Ni in comparison to plants sprayed with water (Fig. 6c). The E values were lower by 11 and 11% at 3 DAS and by 21 and 19% at 5 DAS, respectively, for plants sprayed with 120 and 240 g·ha -1 Ni in comparison to plants sprayed with water (Fig. 6d).
The results of the present study demonstrated that the foliar spray of Ni at doses up to 60 g·ha -1 was no toxic while doses above 120 g·ha -1 caused oxidative stress and damage to the photosynthetic apparatus of soybean leaf tissues. Plants sprayed with doses above 120 g·ha -1 Ni presented chlorosis of leaves followed by tissue necrosis corroborating with the symptoms of Ni toxicity reported by Gajewska et al. (2006) and Antonkiewicz et al. (2016). The ROS are continuously produced in plant tissues as byproducts of the metabolic process (Dat et al. 2000). However, abiotic and biotic stresses in plants promote overproduction and accumulation of ROS. The O 2 and H 2 O 2 are two moderate ROS and may be converted to hydroxyl (OH) and hydroperoxyl radicals (HO 2 -) causing cellular damage by lipid peroxidation (Demidchik 2015;Zhang et al. 2010 240 g·ha -1 Ni indicated that these doses caused oxidative stress in the leaves resulting in lipid peroxidation. Moreover, the increased antioxidant enzyme activities were not sufficient to detoxify the ROS on the cells of plant tissues sprayed with 120 and 240 g·ha -1 Ni. Similarly, the major cause of Ni toxicity in the roots of grapevine (Pavlovkin et al. 2016) and in the leaves of wheat (Gajewska and Skłodowska 2007) was the oxidative stress resulting from an imbalance in the generation and or removal of ROS.
Days after Spray Days after Spray The Ni does not generate ROS directly because it is not a redox-active metal (Chen et al. 2009). However, Ni interferes indirectly with several antioxidant enzymes, increasing or decreasing their activities (Baccouch et al. 1998;Gajewska and Skłodowska 2005). In the present study, the effect of high doses of Ni, increasing the accumulation of ROS, cannot be associated with a decrease in the activity of antioxidant enzymes. However, it is reasonable to relate the increase in ROS with the toxic effect of Ni on photosynthesis. Several indirect and direct paths are known, which lead to nonspecific metal inhibition of photosynthesis (Yusuf et al. 2011). The Ni excess damages the structure of thylakoid membranes and the structure of grana (Szalontai et al. 1999;Molas 2002), reducing the size of grana and increasing the number of nonappressed lamellae (Molas 1997). In fact, in the present study, high doses of Ni (120 and 240 g·ha -1 ) caused damage to the photosynthetic apparatus based on the low values of F v /F m , Y(II) and ETR and high Y(NO) values. An increase on the dissipation of energy by nonregulated process [Y(NO)] generally leads to an increase in ROS production that may have damaged the photosystems and other cellular constituents (Huang et al. 2018).
Distinctly from what was reported by Barcelos et al. (2017), we did not observe an increase on the concentration of photosynthetic pigments in the leaves of plants sprayed with nontoxic Ni concentrations in comparison to plants sprayed with water. These results were distinct probably due to the early plant growth stage evaluated in this study and the advanced plant growth stage reported by Barcelos et al. (2017). Considering the low Ni concentration demanded by plants, it is plausible to take into account that the Ni content in the seeds was sufficient to supply the plants demand at their growth stage evaluated with no effect on the synthesis of photosynthetic pigments. The decreases on Chl concentrations in the leaves of plants sprayed with toxic Ni doses (above 120 g·ha -1 Ni) are in agreement with the findings of Gopal et al. (2002), Gajewska et al. (2006) and Freitas et al. (2019) and can be attributed to disturbances in the synthesis of pigments (Stobart et al. 1985), as well as an increase on their degradation (Somashekaraiah et al. 1992).

ACKNOWLEDGMENTS
Prof. Rodrigues thanks to the Conselho Nacional de Desenvolvimento Científico e Tecnológico (CNPq) for his fellowship.
The limitations in the capacity of the photosynthetic apparatus to capture light and direct energy to photochemical process associated with limitation at stomatal conductance level on leaves of plants sprayed with 120 and 240 g·ha -1 Ni resulted in a decrease on the capacity to process CO 2 . Plant leaf tissues exposed to heavy metals frequently decrease the stomatal aperture (Rucińska-Sobkowiak 2016). Reduction in stomatal conductance in leaves of soybean plants under Ni stress was observed by Reis et al. (2017). Rauser and Dumbroff (1981) reported that leaf tissues of common bean plants treated with Ni showed an increased level of abscisic acid, which is known to cause stomatal closure. Decreased stomatal conductance, associated with the limitations of light capture and use of its energy, already mentioned, may explain the increased C i values observed for plants sprayed whit 120 and 240 g·ha -1 Ni and probably not associated with biochemical limitations.
In conclusion, the spray of Ni doses above 120 g·ha -1 promoted oxidative stress on the leaf tissues of soybean plants and affected the functionality of their photosynthetic apparatus. The Ni doses below 60 g·ha -1 had a low risk to cause toxicity to soybean plants considering the absence of biochemical or physiological damage. It is imperative to consider that Ni concentration supplied by the soil when plants are grown in the field may decrease the threshold line of toxicity at lower levels than what is reported in the present study.