Seeds of S. nigrum were collected from plants grown in Campo Alegre, Porto, Portugal (41°9′55, 79″N and 8°38′19, 48″O) and surface sterilized with 70% (v/v) ethanol for 3 min, followed by 20% (v/v) sodium hypochlorite containing 0.02% (v/v) of Tween-20 for 3 min. Then, the seeds were washed in sterile deionized water in 3 cycles of 1 min each, and placed in filter papers in a flow hood until the seeds and the filter papers were completely dried. Seeds were placed in sterile Petri dishes (diameter 10 cm) containing 8 mL of 1X Hoagland solution (Hoagland and Arnon, 1950) supplemented with 0.625% (w/v) agar, and maintained at 4°C under dark conditions for 48 h (stratification). After stratification seeds were transferred to a growth chamber at 25°C, under a 16 h/8 h day/night photoperiod with a photosynthetically active radiation (PAR) of 60 μmol m^-2 s^-1.

Twenty eight days after germination, seedlings were randomly selected and transferred to plastic pots and grown hydroponically in a substrate mixture of vermiculite:perlite (2:1) for another 28 days, in a growth chamber under the described conditions. Four seedlings were placed in each pot and 3 sets of a total of 6 pots were treated with 1x Hoagland solution supplemented with 0 (control), 12.5 and 25 mg L^-1 of metalaxyl (supplied as a fungicide solution commercially available as Ridomil^® by Syngenta Agro S.A., Portugal). Metalaxyl concentrations were selected based on preliminary data obtained from biometric parameters of plants exposed to: 0, 12.5, 25, 50, 75, 100 mg L^-1. The concentrations chosen to conduct this study (12.5- and 25 mg L^-1) allowed us to: (1) assess the differential physiological and biochemical responses of plants to the agrochemical, and (2), produce enough biomass for all assays. At the end of the treatment, plants were harvested and separated into roots and leaves and fully expanded intact leaves at the third leaf node from the apex were immediately used for biochemical determinations, or grounded to a fine powder in liquid nitrogen (N₂) and stored at -80°C for further studies. The data represent the average of four independent replicates.

Leaves and roots were washed several times with deionized water and, after drying; tissue samples were immersed in 10 mL of deionized water and incubated at 25°C in a rotary shaker (100 rpm). Electrical conductivity of the bathing solution (T1) was recorded after 24 h. Samples were immediately autoclaved at 120°C for 20 min and a last conductivity reading (T2) was obtained when the solutions reached 25°C. EL was expressed following the formula: [I (%) = [1-(1-T1-T2)/(1-(C1-C2))] × 100], where C corresponds to the control situations and T to the treated samples.

Superoxide production was determined by a spectrophometric method based on the nitrobluetetrazolium (NBT) assay, where formazan is produced upon the reduction of NBT by superoxide (Gajewska and Skłodowska, 2007). A total of 0.1 g of S. nigrum leaves and roots (fresh material) was immersed in 0.01M of sodium phosphate buffer pH = 7.8, containing 0.05% (w/v) of NBT and 0.01M of sodium azide (NaN₃). Samples were placed for 60 min in a rotary shaker (250 rpm), in the dark, at room temperature. After the reaction mixture was heated at 85°C for 15 min and cooled in an ice bath during 10 min, the A₅₈₀ was measured. The reducing activity of NBT was expressed as the increase in the A₅₈₀ per gram of fresh weight. Histochemical detection of O2•– was performed in leaves of control and metalaxyl-treated plants according to (Romero-Puertas et al., 2004).

Frozen tissues (0.1 g) from S. nigrum roots and leaves were homogenized at 4°C with a MagNALyser (Roche, Vilvoorde, Belgium) in 1 mL of extraction buffer. Superoxide dismutase (SOD, EC1.15.1.1) and catalase (CAT, EC1.11.1.6) were extracted in a 50 mM potassium phosphate (pH 7.0) containing 0.4 mM EDTA, 0.2 mM PMSF, 2% (w/v) insoluble polyvinylpolypyrrolidone (PVPP) and 1 mM ascorbic acid. Other enzymes such as ascorbate peroxidase (APX, EC1.11.1.11), monodehydroascorbate reductase (MDHAR, EC1.6.5.4), dehydroascorbate reductase (DHAR, EC1.8.5.1), glutathione reductase (GR, EC1.8.1.7) and phenol peroxidase (GPOX, EC1.11.1.7) were extracted in 50 mM MES/KOH buffer (pH 6.0), containing 2 mM CaCl₂, 40 mM KCl and 1 mM ascorbic acid. Ribulose 1,6-biphosphate carboxylase/oxygenase (RuB EC 4.1.1.39) was isolated from rye leaves using a modified method as described by (Wang and Portis, 1992). SOD activity was assayed in a 1.0 ml reaction mixture containing 50 mM potassium phosphate (pH 7.8) buffer, 13 mM methionine, 75 M nitro blue tetrazolium (NBT), 0.1 mM EDTA and 2 M riboflavin. Its activity was measured by the inhibition of the photochemical reduction of NBT at 560 nm, as described by (Dhindsa et al., 1981). CAT activity was determined spectrophotometrically at 240 nm by monitoring the rate of H₂O₂ decomposition at pH 7.0 (Aebi, 1984). APX, MDHAR, DHAR and GR activities were determined according to (Murshed et al., 2008). APX activity was measured in reaction mixtures consisting of 50 mM potassium phosphate buffer (pH 7.0), 0.25 mM AsA, and 5 mM H₂O₂. Activity was determined by measuring the decrease in reaction rate at A₂₉₀. The fraction buffer of DHAR activity was 50 mM HEPES buffer (pH 7.0), contained 0.1 mM EDTA, 2.5 mM GSH, 0.2 mM DHA, and the activity was determined by measuring the increase in reaction rate at A₂₆₅. MDHAR and GR activities were measured in HEPES buffer (100 mM, pH 7.6) and HEPES buffer (100 mM, pH 8), respectively. Their activities were determined by measuring the decrease in absorption at A₃₄₀ and calculated from the 6.22 Mm^-1_.cm^-1 extinction coefficient. Peroxidase (POX) activity was determined by the oxidation of pyrogallol (𝜀₃₄₀ = 2.47 mM^-1 cm^-1) (Kumar and Khan, 1982). The assay mixture of POX contained 0.05M phosphate buffer (pH 6.8), 0.01 M pyrogallol, 10 μl of 20 μL of enzyme extract and 0.01 M H₂O₂. Activity was determined by measuring the decrease in absorption at A₄₃₀ and calculated from the 2.47 Mm^-1cm^-1. Soluble protein content for was obtained according to Lowry et al. (1951). RuBisCO was determined spectrophotometrically at 340 nm by monitoring the rate of NADH oxidation (Racker, 1962). The reaction mixture (1 mL) contained 1 M Tris-HCl (pH 7.8), 6 mM NADH, 0.1 M GSH, 0.5% (w/v) glyceraldehyde-3-phosphate, 25 mM 3-phosphoglycerate kinase, 0.05% (w/v) α-glycro-phosphate dehydrogenase-triose phosphate isomerase, 25 mM RuBP, 0.2 M ATP, 0.5 M MgCl₂, 0.5 M KHCO and isolated RuBisCO extract. One unit of enzyme activity was defined as the amount of enzyme producing 1 μM of RuBisCO per min. All activity measurements were scaled down for semi-high throughput measurement using a micro-plate reader (Synergy Mx, Biotech Instruments Inc., Winooski, VT, United States). The photorespiration enzyme glycolate oxidase (GO, EC1.1.3.1) was determined according to (Feierabend and Beevers, 1972). GO was measured by the formation of a glyoxylate complex with phenylhydrazine (𝜀324 = 17 mM^-1 cm^-1).

Total soluble protein in the leaves was determined by the Bradford method (Lowry et al., 1951). SDS-PAGE of soluble proteins was performed as described by (Laemmli, 1970) and the gels were stained with Coomassie Brilliant Blue R250. Relative RuBisCO content was determined according to Soares et al. (2016) after extraction of rubisco small and large subunit bands in formamide. Rubisco content was expressed relative to total protein content, after extraction and spectrophometric determination of all proteins in the gel.

The soluble sugar content was analyzed by the anthrone method (Irigoyen et al., 1992). Briefly, 0.2 g (FW) of plant material (leaves and roots) was extracted in 80% ethanol at 80°C for 60 min. Total soluble sugars were analyzed by adding 3 mL of freshly prepared anthrone reagent (150 mg anthrone in 100 mL 72% H₂SO₄) to 0.1 mL of the alcoholic extract. This mixture was placed in a water bath at 100°C for 10 min and then cooled in an ice bath for 5 min. The absorbance of samples was measured at 625 nm. The sugar content was calculated from a standard curve prepared with known concentrations of D-glucose in a range of 0–100 μg mL^-1. Following the extraction of soluble sugars, the solid fraction was used for the extraction of starch in 30% perchloric acid at 60°C for 60 min. In this procedure 0.3 mL of the acidic extract were added to 3 mL of anthrone. Test tubes were placed in a water bath at 90°C for 10 min and cooled down in an ice bath until the mixture reached room temperature. D-glucose was also used as standard (in a range of 0–100 μg mL^-1) and the optical density of the starch solution read at 625 nm (Osaki et al., 1991).

Ethanolic root and shoot extracts of S. nigrum plants were used for assaying free amino acids (FAAs) levels using a Waters Acquity UPLC-tqd system (Milford, MA, United States) equipped with a BEH amide 2.1 × 50 column according to (Sinha et al., 2013) with the minor modification previously described by (AbdElgawad et al., 2015).

Metalaxyl levels were determined in both organs of 25 mg L^-1-treated plants as previously described (Teixeira et al., 2011).

Results were expressed as mean ± SD (standard deviation) and analyzed by two-way ANOVA using IBM SPSS Statistica 23 software package (SPSS^® Inc; Chicago, IL, United States) for windows, with organs and concentrations used as fixed variables. Data were tested for normal distribution and homogeneity and normalized when necessary. A significance level of 0.05 was used for rejection of the null-hypothesis. In cases of significant interactions between factors, one-way ANOVA analysis was performed for each factor, and Tukey’s multiple range test was used to determine significant differences among means. All experiments were carried out in quadruplicate (n = 4).

As shown in Figure 1, 25 mg L^-1-treated plants accumulate metalaxyl, with leaves containing seven times more metalaxyl than roots (Figure 1A). Total biomass of 25 mg L^-1- treated plants decreased 50%, when compared to control plants (Figure 1B). Total soluble protein content increased in both roots and shoots in plants treated with the lowest concentration of the fungicide (Figure 1C).

To estimate impact on plant photosynthetic capacity, we measured rubisco activity and rubisco content (relative to soluble protein content), as well as soluble sugar and starch content (Figure 2). Relative rubisco content (%) decreased by 28 and 46% in leaves of plants exposed to 12.5 and 25 mg L^-1 of metalaxyl, respectively (Figure 2A) and rubisco activity decreased by 16 and 37% (Figure 2C). Leaf and root carbohydrate and starch levels decreased under stress. TSS and starch levels decreased as the concentration of metalaxyl increased for both leaves and roots (Figures 2B,D).

Various parameters were measured to determine the oxidative stress status of the metalaxyl-treated plants. First, electrolyte leakage (EL) did not present significant fluctuations for both roots and shoots at all metalaxyl concentrations tested (Figure 3A). Superoxide was determined spectrophotometrically in leave and root extracts (Figure 3B), and through histochemical staining (Figure 3C) of plants supplemented with 0, 12.5, and 25 mg L^-1 metalaxyl. Leaves contain considerably higher O2•–, compared to roots. This ROS significantly decreased by 29 and 39% in shoots and roots, respectively, in 25 mg L^-1- treated plants (Figure 3B). Histochemical detection in leaves confirmed these results, with leaves of control plants showing higher O2•–-induced staining (blades and margins), compared to leaves of metalaxyl-treated plants (Figure 3C). As another indicator of oxidative stress-related changes, we estimated leaf photorespiration, which is an important source of hydrogen peroxide. Increased Gly/Ser ratios (11 and 18% increase) and higher GO activity (11 and 9% increase) were observed in leaves after exposure to 12.5- and 25 mg L^-1 of metalaxyl respectively (Figures 3D,E).

Molecular and enzymatic antioxidants, protect plants largely against ROS impact. These are partially organized in the ASC/GSH cycle (see below). One estimate of the molecular capacity to scavenge cellular ROS is through determination of the total antioxidant capacity (TAC). This parameter significantly increased, in leaves (about 40%), and roots (by 27 and 53%), in response to metalaxyl (Figure 4A). As the TAC assay only measures overall changes in the small molecular antioxidant pool, predominant antioxidant, i.e., polyphenols, flavonoids, tocopherols, ascorbate and gluthatione, were separately quantified. Polyphenol levels were higher in leaves then in roots, and were significantly induced in leaves under the metalaxyl treatment (Figure 4B). Other plant phenols, such as flavonoids, also act as powerful antioxidants. Flavonoid levels were about 49 and 87% in leaves of 12.5- and 25 mg L^-1- treated plants, respectively. Roots behaved in a similar way, with increases of 45 and 62% (Figure 4C). Tocopherol, protecting against lipid oxidation, levels were higher in leaves than in roots, and increased in the metalaxyl treatment (45 and 70% for leaves) (Figure 4D).

Among the enzymatic antioxidants, SOD is essential in controlling O2•– levels. Metalaxyl increased SOD activity in leaves (21 and 83%) and roots (34 and 35%) (Figure 5A). Several enzymes, are dedicated to H₂O₂ scavenging. CAT activity was always higher in roots than in leaves, independent of the experimental situation. Metalaxyl exposure, increased CAT in roots (73% in 25 mg L^-1), and leaves (37%) (Figure 5B). Other enzymes important in controlling H₂O₂ levels are the peroxidases. General peroxidase activity (Figure 5D), was not induced by metalaxyl exposure, but rather decreased, in leaves (14 and 24%) and roots (56 and 7%). Another enzyme involved in defense against H₂O₂ damage, APX, is a key enzyme in the ASC-GSH cycle. It is supplied with electrons by oxidation of ascorbate, which is recycled in a series of redox reactions. Plants exposed to the highest concentration of the pesticide exhibited a significant increase in APX activity of 51 and 33% in leaves and roots, respectively (Figure 5C).

Other parameters of the ASC-GSH cycle, include the ascorbate and glutathione levels and redox status, and the recycling enzymes, MDHAR, DHAR, GR were measured (Figures 6A–E). With respect to ASC, its levels were increased in leaves (59 and 81%), but reduced in roots by 62% in 25 mg L^-1-treated plants (Figure 6A). The ascorbate redox status decreased in leaves and roots (Figure 6C). MDHAR remains relatively unchanged (Figure 6E), whereas DHAR increased slightly, predominantly at 25 mg L^-1 (Figure 6G). Also the GSH concentration was increased predominantly at 25 mg L^-1 (Figure 6B). The glutathione redox status, decreased in leaves and roots (Figure 6D), whereas the GR activity increased in leaves and roots (Figure 6F).

Rubisco can be a rate-limiting enzyme for photosynthesis, especially when plants are exposed to stressful conditions (Wingler et al., 2000). Here, we observe that metalaxyl decreased rubisco activity and amount, probably leading to the inhibition of photosynthesis and abrupt decrease in plant growth. Consistently, the lower content of TSS and starch in metalaxyl-treated leaves reflects the decrease in photosynthesis. Under metalaxyl exposure, we also observed increases in the photorespiration parameters Gly/Ser ratio and glycolate oxidase (GO) activity. Photorespiratory rates increase as a result of the oxygenase activity of rubisco at low CO₂ and high O₂ conditions. The increased photorespiration, strongly indicates elevated cellular levels of H₂O₂, which is also consistent with previous results (Teixeira et al., 2011).

Our results suggest that metalaxyl treatment affected the water status of S. nigrum plants, since leaves of treated plants showed a tendency to wilt that became more severe in plants treated with the highest concentration of the fungicide (25 mg L^-1). Consistently, in a previous work, we observed a significant increase in proline levels which is widely known to act as an osmoprotectant in water-stressed plants (Teixeira et al., 2011). This proteinogenic amino acid may also contribute to reducing the oxidative damage induced by metalaxyl.

To inspect for oxidative stress signs, we investigated membrane integrity (electrolyte leakage), and O2•– levels. Notably, there was no apparent increase in membrane damage, indicating that the stress exposure was at least not very severe. This is consistent with our previous finding of reduced MDA levels in metalaxyl-treated plants (Teixeira et al., 2011). This considerable resistance is remarkably consistent with the bioremediation capacity of this species. Also consistent with the lack in membrane damage, are the lowered O2•– levels, which in turn can be explained by the enhancement of the antioxidant defenses. In fact, SOD is described as a first line of defense against ROS and higher SOD activity often appears to enhance plant tolerance to oxidative damage induced by multiple biotic and abiotic stress factors (Gill and Tuteja, 2010). Our previous studies showed that metalaxyl stimulated the accumulation of hydrogen peroxide (H₂O₂) in shoots and roots of S. nigrum, at similar metalaxyl concentrations. Therefore, as a consequence, the ratio of O2•–/H₂O₂ decreases in metalaxyl-treated plants, lowering the rate production of OH^-, probably also contributing to reduced oxidative damage.

Finally, reduced membrane damage observed in metalaxyl-treated plants could also be explained by enhanced tocopherols levels observed under stress conditions. These lipophilic membrane-embeded antioxidants, prevent chain propagation in lipid peroxidation and efficiently scavenge singlet oxygen radicals (¹O₂). Consistent with the increased tocopherols, metalaxyl exposure also induced significant increases in total antioxidant capacity (TAC), polyphenols and flavonoid contents, especially in leaves. This suggests that non-enzymatic antioxidants may contribute considerably in helping leaf tissues accumulating higher metalaxyl levels with few collateral effects on the oxidative metabolism.

Along with the non-enzymatic antioxidants, H₂O₂-scavenging enzymes are involved in the fine tuning H₂O₂ levels. Catalase (CAT) is a peroxisomal matrix enzyme protecting peroxisomal proteins and membranes from oxidative damage (Mhamdi et al., 2010). CAT activity significantly increased in S. nigrum plants submitted to 25 mg L^-1 of metalaxyl. CAT activity may be regulated by H₂O₂ levels. Here, we found that enhanced H₂O₂ levels quantified in previous studies (Teixeira et al., 2011) induced antioxidant protection by increasing CAT catalytic activity. This behavior can be interpreted as a compensatory mechanism to confront the mild oxidative stress imposed by metalaxyl treatment in plants. Similar results were found in suspension cells of S. nigrum propagated in 20- and 40 mg L^-1 of metalaxyl (de Sousa et al., 2013). In addition to CAT, also POX and APX activities contribute significantly to H₂O₂-level control. Peroxidases are heme-containing glycoproteins encoded by a large multigene family, involved in several physiological processes and are well known to play a key role in oxidative stress conditions (Kawano, 2003). Some peroxidases use aromatic electron donors as substrates to reduce H₂O₂ (Hiraga et al., 2001). POX activities were reduced after metalaxyl exposure in both roots and shoots of S. nigrum. This suggests that this enzyme does not play a pivotal role in defense mechanisms against mild oxidative stress induced by the fungicide.

The ASC/GSH cycle is a combination of reactions in which APX reduces H₂O₂ at the expense of reduced ascorbate. Oxidized ascorbate is regenerated through MDHAR, DHAR and GR, using electrons from GSH and NADPH (Foyer and Noctor, 2011). Although ASC plays several physiological roles, its function in the ASC/GSH cycle puts it central in the maintenance of the cellular redox status. Under metalaxyl treatment, MDHAR and DHAR showed only moderate increases, somewhat more pronounced in roots. We also observed a decreasing ascorbate redox status in metalaxyl exposure, which is consistent with the mild oxidative stress in the plants, and a largely unaltered recycling capacity (Teixeira et al., 2011). As for GSH, the GSH pool increased at mild metalaxyl doses. However, the glutathione redox status decreased despite an increase in GR activity. The decreased GSH levels and redox status indicate relatively high glutathione consumption levels. In particular at the higher metalaxyl concentrations, this GSH consumption could also be related to elevated toxin conjugation through GST activity. This hypothesis is in agreement with the increased root GST activity reported previously in S. nigrum (Teixeira et al., 2011).