Introduction

Cadmium (Cd) is a highly toxic environmental pollutant that has high mobility in the soil-plant system and negatively impacts plants and animals. It is a relatively rare metal in nature, but in recent decades, its concentration in the environment has increased due to the intensification of anthropic activities, mainly of industry and mining, but also in agricultural practices (He et al. 2015; Kubier and Pichler 2019). One of the main effects of Cd toxicity on plants is oxidative stress, originated from excessive production of reactive oxygen species (ROS), which results in damage to lipid membranes (lipid peroxidation), proteins, and nucleic acids (Awan et al. 2020). Moreover, Cd stress can influence physiological processes, such as inhibition of photosynthesis in various ways, including reduction in stomatal conductance, chlorophyll content, and quantum yield of photosystem II (Dobrikova et al. 2021). To protect against Cd-induced oxidative stress, plants have antioxidant mechanisms (enzymatic and non-enzymatic), which have the function of eliminating ROS, transforming them into non-toxic compounds (Kaya et al. 2020a).

In recent years, tanzania guinea grass has been studied under conditions of phytotoxicity and recovery of areas contaminated by heavy metals (Cardoso and Monteiro 2021; Silva et al. 2016; Souza Junior et al. 2019). This grass has shown promise for phytoextraction of metals due to its high shoot mass production, fast growth, large root system, and great adaptability in addition to allowing regrowth (Gilabel et al. 2014; Rabêlo et al. 2017). However, the toxic effects of Cd can compromise biomass production and even grass survival (Leite and Monteiro 2019a; Rabêlo et al. 2017). The efficiency of phytoextraction depends on mechanisms of plant resistance to stresses caused by metals (Ke et al. 2021). The optimization of nutritional management positively interferes in the tolerance capacity of plants exposed to Cd (Rabêlo et al. 2018a; Sebastian and Prasad 2016).

Previous studies have shown that K acts on processes, including the protein synthesis, enzymatic activity, and photosynthesis, that can attenuate the severity of heavy metal toxicity in plants (Li et al. 2020; Naciri et al. 2021; Song et al. 2015; Wu et al. 2020). K supplementation can reduce the deleterious effects of Cd toxicity by increasing photosynthesis and promoting chlorophyll synthesis in tomato (Solanum lycopersicum L.) plants (Naciri et al. 2021). Yasin et al. (2018) reported that the application of K alleviated toxic effects of Cd and improved plant growth, photosynthetic pigments, proline content, and activity of antioxidant enzymes in gladiolus (Gladiolus grandiflora L.). However, few studies have investigated the physiological and biochemical processes involved in the role of potassium in alleviating Cd toxicity in plants, especially in grass, such as tanzania guinea grass; therefore, it needs to be further studied. Furthermore, K metabolism is related to the contents of low-molecular-weight organic nitrogen compounds such as putrescine (PUT), spermidine (SPD), and spermine (SPM) polyamines (Santos et al. 2017). Polyamines increase plant tolerance to abiotic stresses, including Cd toxicity (Howladar et al. 2018; Rady and Hemida 2015), by acting primarily as stress signaling molecules that activate the antioxidant system of plants and stabilize lipid membranes (Anjum et al. 2015). Therefore, it is identified a gap in the knowledge of the relationship between K supply and Cd exposure on polyamines accumulation in plants, and particularly in tropical grasses, that needs to be elucidated.

This study was carried out to evaluate the effect of K on photosynthesis and changes in the antioxidant system of tanzania guinea grass under Cd toxicity. The hypothesis is that improving K nutrition can reduce the Cd-induced oxidative stress and enhance photosynthetic activity by upregulating the antioxidant system (antioxidant enzymes, proline, and polyamines), increasing the tolerance of tanzania guinea grass to this heavy metal.

Materials and methods

Experimental conditions

Tanzania guinea grass was grown in nutrient solution in 3.6-L pots filled with ground quartz used as substrate. The pots were arranged in a greenhouse, in randomized complete block design, in a 3 × 4 factorial, with three K levels (0.4, 6.0, and 11.6 mmol L-1) and four Cd levels (0.0, 0.5, 1.0, and 1.5 mmol L-1), with three replications. The nutrient solutions were prepared from the solution of Hoagland and Arnon (1950). K and Cd levels were defined according to Monteiro and Consolmagno Neto (2008) and Rabêlo et al. (2017), respectively.

The grass seeds were germinated in sand washed with deionized water. Fifteen seedlings were transplanted to each pot, and the Hoagland and Arnon nutrient solution (modified to provide the 0.4 mmol L-1 K level) with 20% of ionic strength was immediately used. Seven days after transplantation, a solution with full ionic strength was applied, modified to provide K levels (0.4, 6.0, and 11.6 mmol L-1). The nutrient solution was replaced every two weeks. For aeration of the root system, the nutrient solution was circulated six to three times a day according to the growth of the grass. Deionized water was used every day to complete the volume of nutrient solution lost. During the seedling adaptation period (15 days after transplanting), thinning was carried out and only five plants per pot remained. Other details on seedlings and nutrient solution management were presented by Anicésio and Monteiro (2019).

Plants were allowed to grow for a month, as such period favors the development of Panicum maximum grown under K rates (Lavres Júnior et al. 2010), and plants can reach complete metabolism, with the antioxidant system able to tolerate the oxidative stress. At 30 days of plant growth, Cd levels were applied and there was a five-day Cd exposure period (until the first harvest).

The plants were grown in two periods. The first harvest (shoots) was done upon the occurrence of severe symptoms of Cd toxicity and occurred at 35 days after transplantation. After the first harvest, the contaminated nutrient solution was replaced, and Cd was not reapplied, because the high Cd levels caused serious toxic effects on the first growth, so its reapplication could lead to the death of plants. Then, Cd response at the second harvest was evaluated based on the residual effect of this metal. The second harvest was done 22 days after the first one, and shoots and roots were collected.

For the metabolic evaluations, in both harvests, the plant material was packed in aluminum bags and frozen in liquid N to cease metabolic reactions, and then stored in a freezer at – 80 °C. Later on, it was ground in a mill containing liquid N.

Gas exchange and fluorescence assessments

At the end of each growth period, the gas exchange parameters were evaluated as net photosynthesis (A) and stomatal conductance (gs), as well as the quantum efficiency of photosystem II (QEP). To perform these determinations, plants were kept for 24 h under controlled conditions of light, temperature, and humidity, in the growth chamber. Readings were taken on the middle third of two newly expanded leaves using the WALZ-GFS-3000 IRGA@ photosynthesis analyzer. The parameters were set at 380 μmol mol-1, 20,000 μmol mol-1, 750 μmol s−1, 1000 μmol m−2 s−1, and 27 °C for CO2 concentration, H2O concentration, flow rate, light, and the cuvette temperature respectively, according to the methods proposed by Long and Bernacchi (2003), Magney et al. (2017) and Rabêlo et al. (2018b), with adaptations.

Lipid peroxidation and hydrogen peroxide

Lipid peroxidation of membranes was measured by malondialdehyde (MDA) concentration (Heath and Packer 1968). Samples of frozen plant tissue (0.2 g) were homogenized with 2 mL of trichloroacetic acid (TCA, 1 g L-1) and polyvinylpolypyrrolidone (PVPP) in an ice bath and then centrifuged at 10,000 × g for 5 min at 4 °C. The supernatant was used to determine MDA and H2O2 concentrations. For determination of MDA concentration, an aliquot of the supernatant was taken and TBA (5 g L-1 TBA diluted with 200 g L-1 TCA) was added. The solution was incubated in a dry bath at 95 °C for 30 min. Spectrophotometer absorbance readings were taken at the wavelengths of 535 and 600 nm. MDA concentration was calculated using the extinction coefficient of 155 mM-1 L-1. For the determination of H2O2 concentration, an aliquot of the supernatant was collected and phosphate buffer (100 mmol L-1, pH 7.5) and potassium iodide (1 mol L-1) were added. The mixture was incubated on ice in the dark for 60 min. The absorbance reading was taken in a spectrophotometer at 390 nm and the hydrogen peroxide concentration was determined based on a standard curve (Velikova et al. 2000).

Extraction and quantification of protein

Frozen plant tissue samples (1 g) were homogenized with 3 or 2 mL (shoots or roots, respectively) of extraction buffer containing 100 mmol L-1 potassium phosphate (pH 7.5), 1 mmol L-1 ethylenediaminetetraacetic acid (EDTA), and 3 mmol L-1 dithiothreitol (DTT) and macerated with polyvinylpolypyrrolidone (PVPP). The homogenate was centrifuged at 10,000 × g for 30 min at 4 °C (Monteiro et al. 2011). The supernatant was collected and divided into aliquots. An aliquot was used for quantification of protein and the others were stored in a freezer at − 80 °C and used to determine the activities of antioxidant enzymes. Soluble protein quantification was performed in accordance with Bradford (1976), using a standard curve prepared with bovine serum albumin (BSA).

Antioxidant enzymes activities

Superoxide dismutase (SOD - EC 1.15.1.1)

To determine superoxide dismutase activity, protein separation by electrophoresis was initially performed using non-denaturing polyacrylamide gels (PAGE), as described by Azevedo et al. (1998). After separation, the gels were incubated for 30 min, in the dark box, in a 100 mmol L-1 potassium phosphate buffer solution (pH 7.8) containing 1 mmol L-1 EDTA, 0.1 mmol L-1 nitroblue tetrazolium chloride (NBT), N, N, N ', N'-tetramethylethylenediamine (TEMED), and 0.05 g L-1 riboflavin. Then, the gels were placed under light for 15 min until the appearance of the achromatic SOD bands on a purple/blue background. The SOD bands for each sample were qualitatively compared with the bovine SOD band (standard).

Catalase (CAT - EC 1.11.1.6)

Catalase activity was determined at 25 °C in a reaction medium composed of 100 mmol L-1 potassium phosphate buffer solution (pH 7.5) containing 300 g L-1 H2O2. The reaction was initiated by the addition of a 25μL aliquot of the protein extract. Reading was performed in a spectrophotometer at 240 nm after 1 min H2O2 decomposition (Monteiro et al. 2011). For the calculation of CAT activity, a molar extinction coefficient of 39.4 M-1 cm-1 was used in the following formula:

$$ \mathrm{CAT}=\left\{\left(\mathrm{Abs}/\upvarepsilon \right)/1000\right)/\left[\left(\mathrm{protein}\ \left(\mathrm{mg}/\mathrm{mL}\right)\ \mathrm{x}\ \mathrm{volume}\ \left(\upmu \mathrm{L}\right)\right)/1000\right)\left]\right\}\ \mathrm{x}\ 1000000 $$

Where: Abs = Absorbance, Ɛ = molar extinction coefficient, and volume = volume of protein extract (25 μL).

The results were expressed in μmol/min/mg protein.

Ascorbate peroxidase (APX – EC 1.11.1.11)

The determination of ascorbate peroxidase activity was based on the oxidation of ascorbate at 30 °C (water bath) for 1 min according to Nakano and Asada (1981). In the reaction medium composed of 80 mmol L-1 potassium phosphate buffer (pH 7.0), 5 mmol L-1 ascorbic acid and 1 mmol L-1 EDTA, H2O2 (1 mmol L-1) and the protein extract (40 μL aliquot) were added at the time of reading. The absorbance was measured by spectrophotometer at 290 nm. APX activity was calculated using an extinction coefficient of 2.8 mM-1 cm-1 and the formula:

$$ \mathrm{APX}=\left\{\left(\mathrm{Abs}/\upvarepsilon \right)/1000\right)/\left[\left(\mathrm{protein}\ \left(\mathrm{mg}/\mathrm{mL}\right)\ \mathrm{x}\ \mathrm{volume}\ \left(\upmu \mathrm{L}\right)\right)/1000\right)\left]\right\}\ \mathrm{x}\ 1000000 $$

Where: Abs = Absorbance, Ɛ = molar extinction coefficient, and volume = volume of protein extract (40 μL).

The results were expressed in nmol/min/mg protein.

Guaiacol peroxidase (GPX - EC. 1.11.1.7)

The method described by Matsuno and Uritani (1972) was used to determine guaiacol peroxidase activity. A reaction medium composed of pH 5.0 citrate phosphate buffer [dibasic sodium phosphate (0.2 mmol L-1) plus citric acid (0.1 mmol L-1), guaiacol (5 g L-1) and H2O2 (30 g L-1)] was used. The protein extract (20 μL aliquot) was added, mixed, and incubated at 30 °C (water bath) for 15 min. To stop the reaction, sodium metabisulfite (0.1 mol L-1) was added and the reading was performed in a spectrophotometer at 450 nm. The extinction coefficient of 25.5 mM-1 cm-1 and the formula below were used to calculate GPX activity, which was expressed in μmol/min/mg protein.

$$ \mathrm{GPX}=\left(\mathrm{Abs}/\upvarepsilon \right)\times \left(1000/\mathrm{volume}\left(\mathrm{mL}\right)\right)\times \left(1/15\;\min \right)\times \left(1/\mathrm{protein}\;\left(\mathrm{mg}/\mathrm{mL}\right)\right) $$

Where: Abs = Absorbance, Ɛ = molar extinction coefficient, and volume = volume of protein extract (0.02 mL).

Glutathione reductase (GR - EC 1.6.4.2)

To determine glutathione reductase activity, a reaction medium containing 100 mmol L-1 potassium phosphate buffer (pH 7.5) and 1 mmol L-1 2-nitrobenzoic acid (DTNB) was prepared and placed in a water bath at 30 °C. At the time of reading, 0.1 mmol L-1 reduced adenine dinucleotide nicotinamide (NADPH), 1 mmol L-1 oxidized glutathione (GSSG), and protein extract (50 μL aliquot) were added to initiate the reaction. The reading was performed in a spectrophotometer at 412 nm (Azevedo et al. 1998; Smith et al. 1988). A standard curve was prepared from known concentrations of GR. From the GR concentration values, GR activity, expressed in μmol/min/mg protein, was calculated for the ratio of the concentration and the protein content according to the formula:

$$ \mathrm{GR}=\left(\upmu \mathrm{mol}\ \mathrm{GR}/\mathrm{mL}\right)/\left[\left(\mathrm{protein}\ \left(\mathrm{mg}/\mathrm{mL}\right)\ \mathrm{x}\ \mathrm{volume}\ \left(\upmu \mathrm{L}\right)\right)/1000\right)\Big] $$

Where: volume = volume of protein extract (50 μL).

Proline concentration

Proline concentration was determined by colorimetry, according to Bates et al. (1973). Frozen plant tissue samples (0.3 g) were homogenized with 10 mL of sulfosalicylic acid (30 g L-1). The mixture was centrifuged at 10,000 × g at 15 °C for 20 min. Acid ninhydrin solution and glacial acetic acid were added to the supernatant (1 mL). It was placed in a water bath at 100 °C for 60 min. Toluene (2 mL) was added to each sample and mixed. The organic phase (toluene) was collected and placed in glass cuvettes for absorbance reading in a spectrophotometer at 520 nm. Concentrations were obtained using a standard proline curve, inserting the absorbances of the samples. Proline concentration, expressed in μmol g-1 fresh tissue, was calculated according to the following formula:

$$ \mathrm{PRO}=\left[\left(\upmu \mathrm{g}\ \mathrm{proline}/\mathrm{mL}\ \mathrm{x}\ \mathrm{vol}.\mathrm{toluene}\ \left(\mathrm{mL}\right)\right)/115.13\ \upmu \mathrm{g}/\upmu \mathrm{mol}\right]/\left[\left(\mathrm{sample}\ \mathrm{mass}\ \left(\mathrm{g}\right)\right)/10\ \left(\mathrm{DF}\right)\right] $$

Where: 115.13 μg/μmol (molar mass of proline) and 10 (DF- dilution factor).

Polyamines concentrations

The putrescine (PUT), spermidine (SPD), and spermine (SPM) polyamines concentrations were determined according to Flores and Galston (1982) with adaptations by Lima et al. (2008). Frozen plant tissue samples (0.2 g) were homogenized with 2 mL of 50 mL L-1 perchloric acid and subsequently centrifuged at 10,000 × g at 4 °C for 10 min. The dansylation reaction was carried out by adding saturated sodium carbonate and dansyl chloride (5-[N, N-dimethylamino] naphthalene-1-sulfonyl chloride) diluted with acetone (5 mg mL-1). The samples were kept at room temperature in the dark for 16 h. Then, proline (100 mg L-1) was added and allowed to stand for a further 30 min in the dark. To extract the dansylated polyamines, toluene was added and the organic phase (dansylated extract) was collected. Polyamines were separated by thin-layer chromatography. Dansylated extract was applied to glass chromatographic plates (20 × 20 cm) covered with 60G-Merck silica gel (250 μm thick) previously activated at 160 °C for 90 min. These plates were placed in chromatography tank and chromatography was performed using chloroform:triethylamine (12:2 v/v) as the mobile phase. A mixture containing the three amine standards was processed parallel to the samples. At the end of chromatography, the dansylated compounds were activated with UV light at 365 nm and fluorescence intensity readings at 507 nm were taken on the GelDoc-It Imaging System UVP photodocumenter using VisionWorksLS Acquisition and Analysis software. Polyamines were quantified by relating the fluorescence intensity of the standards (also applied to the plates) to those of the samples.

Statistical analysis

Results were statistically analyzed by using the “Statistical Analysis System” software (SAS-Institute 2011). Initially, the analysis of variance was performed by the GLM procedure, and when the K levels × Cd levels interaction was significant, the polynomial regression (response surface) was performed by the RSREG procedure. When the K levels × Cd levels interaction was not significant, regression analysis was performed for the Cd levels and means were compared by the Tukey test for K levels. Significance level of 0.05 was used for all analyses.

Results

Gas exchange and fluorescence

At the first harvest, the highest values of stomatal conductance (gs) as well as the net photosynthesis (A) in the leaf blades were observed at K levels of 6.0 and 11.6 mmol L-1 (p ≤ 0.05). The increases were two- to three-fold for gs and three- to four-fold for A compared with the K deficiency conditions (0.4 mmol L-1; Table 1). For the Cd levels, significant (p ≤ 0.05) decreases of 58 and 76% were observed for gs and A, respectively, when comparing the no-Cd application in the nutrient solution with the highest level of this heavy metal (1.5 mmol L-1; Fig. 1a, c).

Table 1 Net photosynthesis (A), stomatal conductance (gs), and malondialdehyde concentration (MDA) in tanzania guinea grass shoots, at the first harvest, as related to K levels
Fig. 1
figure 1

Net photosynthesis and stomatal conductance in tanzania guinea grass, at the first harvest as related to Cd levels (a and c), at the second harvest as related to K and Cd levels combinations (b and d), and quantum efficiency of photosystem II, at the first and second harvest as related to K and Cd levels combinations (e and f). Bars (a and c) indicate standard errors

At the second harvest, the K levels × Cd levels interaction was significant (p ≤ 0.05) for A and gs (Fig. 1b, d). Despite the deleterious effect of Cd at the first harvest, application of 1.0 mmol L-1 Cd induced increases of 57 and 42%, respectively, in A and gs compared to control (no Cd added). Moreover, plants grown with 1.0 mmol L-1 Cd associated with intermediate and high K availability (6.0 and 11.6 mmol L- 1) had values of A around 15% higher than under conditions of low K availability (0.4 mmol L-1; Fig. 1b).

The interaction of K levels × Cd levels in the nutrient solution showed significant (p ≤ 0.05) effect on the quantum efficiency of photosystem II (QEP), at both harvests (Fig. 1e, f). At the first harvest, the QEP values decreased by 40% with the increase in Cd levels, but by supplying K such problem was alleviated. For plants exposed to 0.5 mmol L-1 Cd, the high availability of K (11.6 mmol L-1) promoted a 35% increase in QEP values, when compared with the low availability of K (0.4 mmol L-1; Fig. 1e). Conversely, at the second harvest, plants exposed to Cd increased the QEP by 35% compared to control. Plants grown at low Cd levels (up to 0.5 mmol L-1) combined with high K levels (6.0 and 11.6 mmol L- 1) in nutrient solution showed values of QEP around 15% higher than those under conditions of low K availability (0.4 mmol L-1; Fig. 1f).

Lipid peroxidation and hydrogen peroxide concentration

In the shoots collected at the first grass harvest, the highest concentration of malondialdehyde (MDA; lipid peroxidation) was observed under K deficiency conditions (0.4 mmol L-1; p ≤ 0.05; Table 1). Exposure to Cd promoted a dose-dependent increase (p ≤ 0.001) in the concentration of MDA in the shoots, with an increase of approximately 35% when comparing the 1.5 mmol L-1 Cd level with no-application of this metal (Fig. 2a). At the second harvest, Cd stress significantly (p ≤ 0.05) increased the concentration of MDA by 50% compared to control; however, the highest K supply (11.6 mmol L-1) led to a 36% reduction in MDA concentrations in the shoots of tanzania guinea grass grown with high Cd level (1.5 mmol L-1) solution (Fig. 2b). There was no significant effect of K and Cd levels on MDA concentration in tanzania guinea grass roots, and the average MDA value was 11.7 nmol g-1 fresh tissue.

Fig. 2
figure 2

Tanzania guinea grass lipid peroxidation (MDA) in the shoots, at the first harvest as related to Cd levels (a) and at the second harvest as related to the K and Cd levels combinations (b); hydrogen peroxide (H2O2) in the shoots, at the first and second harvests as related to Cd levels (c); superoxide dismutase (SOD) activity in PAGE in the shoots at the first (d) and second (e) harvests and in the roots (f) as related to the K and Cd levels combinations. Arrows (d, e, and f) indicate SOD isoforms identified. S, Standard (Bovine SOD). Bars (a and c) indicate standard errors

At the first harvest, the highest Cd level (1.5 mmol L-1) promoted a significant (p ≤ 0.01) 25% increase in the concentration of hydrogen peroxide (H2O2) in the shoots when compared to the non-availability of Cd in the growth medium. At the second harvest, the H2O2 concentration in the shoots was significantly (p ≤ 0.05) increased non-proportionally with Cd levels. Exposure to Cd increased H2O2 concentration in tanzania guinea grass shoots by 30% compared to unstressed plants (no Cd added; Fig. 2c). The H2O2 concentration in the roots was not significant for the studied factors and showed an average of 1.2 μmol g-1 fresh tissue.

Antioxidant enzymes activities

The superoxide dismutase (SOD) activity in PAGE in the tanzania guinea grass shoots collected at the first harvest showed different band intensities due to K and Cd levels, especially in isoform II (Fig. 2d). It was found, especially through isoforms II and III, that the bands were more intense with the lowest K supply (0.4 mmol L-1). At the K level of 11.6 mmol L-1, the increased availability of Cd caused an increase in SOD activity, which was verified by the increasing intensity of the isoforms I and II bands.

At the second harvest, K level of 11.6 mmol L-1 was responsible for the highest intensities of SOD bands in the three isoforms (and isoform III could only be detected under this condition). The high availability of residual Cd (1.0 and 1.5 mmol L-1) in the nutrient solution promoted an increase in band intensity, especially of isoform II (Fig. 2e). In the roots, only the presence of the isoforms I and II was verified. The intensity differences in SOD bands between K levels and Cd levels were clear in isoform II. Only roots grown under K level of 11.6 mmol L-1 showed clear and non-variable intensity SOD bands for all Cd levels (Fig. 2f).

Catalase (CAT), guaiacol peroxidase (GPX), and ascorbate peroxidase (APX) activities in the grass shoots collected at the first harvest showed no significant effects of K and Cd levels, and the respective average values were 55.08 μmol min-1 mg-1 protein, 29.37 μmol min-1 mg-1 protein, and 54.65 nmol min-1 mg-1 protein. For glutathione reductase (GR) activity, a significant (p ≤ 0.05) response to K levels was found and the high K levels (6.0 and 11.6 mmol L-1) promoted increases of 1.8 and 1.6-fold, respectively, in the activity of this enzyme compared with the K deficiency conditions (0.4 mmol L-1; Fig. 4a).

At the second harvest, a significant (p ≤ 0.05) interaction of K levels × Cd levels was found for CAT, GPX, APX, and GR activities (Fig. 3a, c, d and 4b). Cd stress decreased the activities of these enzymes in the grass shoots by 80–100% compared to control plants. The lowest activities of these enzymes in shoots were obtained with low K availability (0.4 mmol L-1) and high residual Cd (1.5 mmol L-1). Conversely, the increase in the supply of K to plants exposed to Cd promoted an increase in the activities of antioxidant enzymes. For plants exposed to 0.5 mmol L-1 Cd, the high K supply (11.6 mmol L-1) promoted increase in CAT (75%), GPX (50%), APX (66%), and GR (66%) activities.

Fig. 3
figure 3

Tanzania guinea grass activities of catalase (CAT), guaiacol peroxidase (GPX), and ascorbate peroxidase (APX), in the shoots at the second harvest as related to K and Cd levels combinations (a, c, and d); catalase activity in the roots as related to Cd levels (b), and ascorbate peroxidase in the roots as related to K levels (e) and Cd levels (f). Bars (b, e, and f) indicate standard errors

Fig. 4
figure 4

Tanzania guinea grass shoots: glutathione reductase (GR) activity at the first harvest, as related to K levels (a) and, at the second harvest, as related to K and Cd levels combinations (b), and proline concentration at the first harvest, as related to K levels (c) and Cd levels (d) and, at the second harvest, as related to K and Cd levels combinations (e). Bars (a, c, and d) indicate standard errors

APX activity in the roots was increased significantly (p ≤ 0.05) under K deficiency (0.4 mmol L-1; Fig. 3e). For Cd levels, CAT and APX activities in the roots showed a significant (p ≤ 0.01; p ≤ 0.05, respectively) dose-dependent increase, as opposed to the responses in the shoots (Fig. 3b, f). GPX and GR activities in the roots were not significantly affected by K and Cd levels and showed average values of 160.15 and 3.51 μmol min-1 mg-1 protein, respectively.

Proline concentration

At the first harvest, there were significant independent effects of K (p ≤ 0.05) and Cd (p ≤ 0.01) levels on proline concentration. The 0.4 mmol L-1 K level increased proline concentration in shoots of tanzania guinea grass by 2.7-fold when compared to the 11.6 mmol L-1 K level in the nutrient solution (Fig. 4c). Concentration of this amino acid showed a dose-dependent increase with the Cd levels, and an increase of 65% was observed when compared high Cd level (1.5 mmol L-1) with unstressed plants (no Cd added; Fig. 4d).

Conversely, at the second harvest, the interaction K levels × Cd levels was significant (p ≤ 0.05) for the concentration of proline. The increase in the supply of K to plants exposed to Cd promoted an increase in the concentration of proline. Plants grown with high K level (11.6 mmol L-1) and exposed to 1.0 mmol L-1 residual Cd increased their proline concentration, which was 2-fold higher than that observed with the K deficiency (0.4 mmol L-1; Fig. 4e). Proline concentration in the roots was not changed by K and Cd levels and had an average of 0.28 μmol g-1 fresh tissue.

Polyamines concentrations

Significant (p ≤ 0.05) increase was found in the concentration of putrescine in the grass shoots, at the first harvest, in plants under K deficiency (0.4 mmol L-1), which was about 34% higher than in those receiving 6.0 and 11.6 mmol L-1 of this nutrient (Table 2). In the material collected at the second harvest, the interaction K levels × Cd levels was significant (p ≤ 0.05) for the concentration of putrescine, both in shoots and roots. The highest putrescine concentration was found under conditions of low K availability (0.4 mmol L-1) associated with the non-availability of Cd in the nutrient solution. High K supply (11.6 mmol L-1) resulted in low putrescine concentration in shoots and roots (Fig. 5a, b).

Table 2 Polyamines concentrations in the shoot samples collected at the first harvest of tanzania guinea grass, as related to K levels
Fig. 5
figure 5

Putrescine concentration at the second harvest in the grass shoots (a) and roots (b), as related to K and Cd levels combinations, and spermidine (c) and spermine (d) concentrations at the second harvest in the shoots of tanzania guinea grass, as related to K and Cd levels combinations

At the first harvest of the grass, the concentrations of spermidine and spermine were significantly (p ≤ 0.05) higher in the shoots of plants grown at 6.0 and 11.6 mmol L-1 K levels than at 0.4 mmol L-1 (Table 2). At the second harvest, the concentrations of spermidine and spermine in shoots showed a significant (p ≤ 0.05) interaction K levels × Cd levels. Cd stress decreased spermidine and spermine concentrations in shoots; however, when plants were grown under high K availability (11.6 mmol L-1), a 25% increase was observed (Fig. 5c, d). There was no significant effect of K and Cd levels on spermidine and spermine concentrations in tanzania guinea grass roots, whose averages were 363 and 304 nmols g-1 fresh tissue, respectively.

Discussion

Cd-induced plant toxicity interferes with various plant physiological and biochemical processes, causing damage to cell structure and reducing photosynthetic rate due to the oxidative stress caused by this metal (He et al. 2015). Few studies show the positive interference of K in plants’ tolerance to this heavy metal. In a previous study, it was found that K promotes high growth and shoot dry mass production of tanzania guinea grass exposed to Cd, and therefore, K increases extraction of Cd from the growing medium (Anicésio and Monteiro 2019). However, there was no information on the physiological and biochemical mechanisms responsible for increasing Cd phytoextraction capacity of this grass supplied with high K levels.

At the end of the first growth period, a short period of exposure (five days) to Cd led to low stomatal conductance (gs) and quantum efficiency of photosystem II (QEP), thus resulting in low net photosynthesis (A) in tanzania guinea grass. These results corroborate those found by Zhang et al. (2014) in Pennisetum americanum (L.) K. Schum. x Pennisetum purpureum Schum., by Ahanger et al. (2020) in Vigna angularis (L.), and by Imran et al. (2021) in rice (Oryza sativa L.). The reduction in photosynthetic activity caused by Cd may be related to stomatal and non-stomatal limitations (Varone et al. 2012). There was a decrease in gs, possibly due to stomatal closure, which resulted in lower CO2 diffusion (Rabêlo et al. 2018a). Non-stomatal limitations are related to the change caused by Cd in chloroplasts structure, in chlorophyll content, in the composition of thylakoid membranes, and in the electron transport, resulting in low QEP (Araújo et al. 2017; Kaya et al. 2020b). These non-stomatal limitations under Cd toxicity conditions occur when there is an excess in accumulation of ROS (demonstrated by high H2O2 and MDA concentrations), which cause oxidative stress in chloroplasts, one of the main organelles where ROS are produced (Kaya et al. 2019; Kaya et al. 2020a).

At the second harvest, there were increases in A, gs, and QEP in plants exposed to Cd (up to 1.0 mmol L-1), which suggest a eustress state, as previously evidenced by Leite and Monteiro (2019b). This could be related to the high proline concentration and its protective role in photosynthesis. Among the functions of proline in plants are osmotic regulation and prevention against the degradation of photosynthetic pigments induced by Cd (Zouari et al. 2016). This beneficial effect of proline on the photosynthetic activity has been previously reported in date palm (Phoenix dactylifera L.) (Zouari et al. 2016) and eggplant (Solanum melongena L.) (Singh et al. 2015). However, high K availability in nutrient solution further improved A under Cd toxicity (1.0 mmol L-1). Interestingly, plants exposed to 1.0 mmol L-1 Cd and supplied with high K levels showed the same value of A (15 μmol m-2 s-1) as control plants, at the first harvest. These results indicate the positive role of K in increasing the photosynthetic efficiency of plants under Cd toxicity. These findings agree with previous reports that K supplementation improved photosynthesis, enhanced chlorophyll synthesis, and maintained chlorophyll fluorescence in plants under Cd stress (Naciri et al. 2021). This could be associated with the high proline concentration and upregulation of antioxidant enzymes, due to the important role of this amino acid and of these enzymes in reducing oxidative stress and increasing photosynthetic activity (Gill et al. 2012; Zhang et al. 2020).

Oxidative stress in tanzania guinea grass due to Cd toxicity increased H2O2 and MDA concentrations in its shoots. Cd is a non-redox metal, so it cannot produce hydroxyl radicals (OH-) through the Haber-Weis and Fenton reactions (Clemens 2006). However, this heavy metal indirectly induces oxidative stress by accumulating other types of ROS through its interference with electron transport chains (blocking) and its negative relationship with compounds of the antioxidant defense system (inhibition of antioxidant enzymes due to the high Cd affinity for thiol groups) (Gallego et al. 2012). Cd inhibited the antioxidant enzymes activities in tanzania guinea grass and may be responsible for the oxidative stress. However, high K supply reduced the concentration of MDA and thus the oxidative stress under Cd toxicity. One reason for the reduction caused by K in lipid peroxidation might be due to enhanced antioxidant enzymes activities as previously reported by Yasin et al. (2018).

Three SOD isoforms were detected in the grass tissue under Cd stress. The changes observed in the profile of this enzyme indicate its involvement in the detoxification mechanisms of tanzania guinea grass. Plants grown under the highest K level (11.6 mmol L-1) maintained high SOD activity in the tissue collected at the second harvest, even under high Cd levels. This suggests high efficiency in converting O2- into H2O2 (Clemens 2006), that was eliminated by peroxidases, which also had high activities under high K availability.

In the plant tissue collected at the second harvest, CAT and GPX peroxidases as well as enzymes of the AsA-GSH cycle (APX and GR) were inhibited at high Cd levels, which resulted in high H2O2 accumulation and lipid peroxidation. Therefore, they play a crucial role in the resistance to Cd-induced stress. This metal causes inhibition of the enzymes because of its high affinity with cysteine residues (thiol groups) of their structures, as well as compromising the transcription and translation of genes responsible for their synthesis (DalCorso et al. 2013). The decrease in APX and GR activities due to Cd toxicity may also be due to disturbances in chloroplasts, where these enzymes are found (Lou et al. 2017). Wu et al. (2017) found low activity of CAT, APX, and GR and Wang et al. (2013) found low activity of CAT and APX in mustard (Brassica campestris L.) and ryegrass (Lolium perenne L.) plants, respectively, exposed to Cd. However, the supply of K to plants under Cd toxicity increased the activity of antioxidant enzymes SOD, CAT, and GPX as well as AsA-GSH cycle enzymes (APX and GR) in shoots of tanzania guinea grass, reducing lipid peroxidation (demonstrated by the low MDA concentration). Positive K responses in the antioxidant system of plants exposed to Cd toxicity were also found by Ahmad et al. (2016b), who reported high activity of the enzymes SOD, APX, and GR in chickpea (Cicer arietinum L.) and by Zaheer et al. (2018), who reported increase in CAT and APX activities in gladiolus (Gladiolus grandiflora L.). Increases in antioxidant enzymes activity may be due to the recognized role of K as enzyme activator (Marschner 2012). Under stress conditions, K plays an important role in the synthesis of proteins, such as thioredoxins, glutaredoxins, among others. These proteins act as reducing agents, regenerating the catalytically active form of peroxiredoxins, which participate in reducing ROS formation in plants under environmental stress (Tripathi et al. 2009). Increasing antioxidant capacity is especially important for the survival of plants grown in metal-contaminated environments (Habiba et al. 2015). Therefore, the application of K increases the tolerance of this grass to Cd stress, improving the enzymatic antioxidant defense system. Thus, K promoted high growth and shoot dry mass production in tanzania guinea grass under Cd toxicity, as verified by Anicésio and Monteiro (2019).

Proline concentration in the shoots of tanzania guinea grass was high when Cd increased in the nutrient solution, indicating that Cd stress stimulates osmoprotectant production. This may be due to proline, which, besides acting on the plant water balance, eliminates the excess of ROS (avoiding the degradation of photosynthetic pigments), stabilizes the structure of cell membranes, among other functions that help plants survive under Cd stress (Zouari et al. 2016). Higher proline concentration due to Cd stress has also been observed in mustard (Brassica juncea L.) (Ahmad et al. 2016a) and wheat (Triticum aestivum L.) (Repkina et al. 2019). At the second harvest, Cd toxicity (up to level 1.0 mmol L-1) also increased the concentration of proline but it was observed that the high K level (11.6 mmol L-1) further improved the concentration this amino acid. The reason might be that proline is derived from precursors like glutamic acid, arginine, and ornithine (Fichman et al. 2015). Genes related to enzymes involved in the biosynthesis of proline from glutamic acid showed a slight increase with K (Li et al. 2020). Moreover, K enhanced the conversion of arginine into proline, possibly due to increased arginase activity (Nageswararao et al. 1981). K supplementation also enhanced the concentration of proline in Brassica oleracea L. (Ahmad et al. 2019) and Gossypium hirsutum L. (Zahoor et al. 2017), stressed by salt and drought, respectively.

Overall, Cd stress causes disruption in polyamines’ homeostasis due to increased PUT concentration and decreased SPD and SPM, impairing plant development (Qiao et al. 2015). However, it was found that Cd stress reduced SPD, SPM and PUT concentrations. This can be attributed to high proline synthesis under Cd toxicity, as it occurs through activation of precursors in common with PUT (arginine and ornithine) (Szabados and Savoure 2010), so the increase in proline synthesis resulted in low PUT concentration in the grass. The decrease in PUT concentration resulted in low SPD and SPM concentrations because PUT is a precursor molecule of these polyamines (SPD and SPM). In addition, polyamine oxidase (PAO) activity was possibly increased in plants with Cd supply, and consequently SPD and SPM concentrations were low. However, high K supply (11.6 mmol L-1) to plants under Cd stress resulted in high SPD and SPM concentrations, reaching values close to those found in control plants and showing the importance of K in the synthesis of polyamines. The reason might be that the biosynthesis of SPD and SPM needs S adenosyl methioninamine (SAE), which is produced from S-adenosyl methionine (SAM) decarboxylation, and K+ acts as a cofactor for SAM synthetase (Takusagawa et al. 1996). SPD and SPM polyamines play an important protective role in plants exposed to stress by regulating cell redox homeostasis, by acting as metal chelators and components of stress signaling pathways, by eliminating ROS, and by activating the antioxidant defense system of the cells (Anjum et al. 2015; Kebert et al. 2017). Thus, the decrease in MDA concentration promoted by K may be a result of the increases in SPD and SPM concentrations caused by this nutrient in plants exposed to Cd. K also can alleviate Cd stress by increasing the polyamines (SPD and SPM) concentrations in tanzania guinea grass.

Conclusions

Cd reduced photosynthetic activity in tanzania guinea grass by causing low stomatal conductance and losses in quantum efficiency of photosystem II. However, the increase in proline synthesis under 1.0 mmol L-1 Cd and high K level conditions resulted in a eustress state that restored the net photosynthesis of this grass. Cd increased lipid peroxidation by inhibiting antioxidant enzymes activities and reducing polyamines concentrations. However, the abundant supply of K can reduce the Cd-induced oxidative stress in tanzania guinea grass, which is evidenced by the decrease in MDA concentration in the shoots. Mitigation of Cd toxicity by increasing K availability results from increased antioxidant enzymes activities and proline and polyamines (SPD and SPM) synthesis. Therefore, K increases tanzania guinea grass tolerance to Cd.