Fabaceae response to dimethoate insecticide application: investigation of germination, seedling growth and tolerance mechanisms

Plant materials

This study was conducted using Vicia faba L. (var. Saber 02), Trigonella foenum-graecum L., Vicia sativa L., and Medicago truncatula (var. Jemalong A17).

Pesticide treatments

To assess the effects of commonly used pesticide dimethoate on V. faba, T. foenum-graecum, V. sativa and M. truncatula legumes, three different concentrations of dimethoate were prepared.

The tested concentrations of dimethoate were the recommended dose (0.4 g L⁻¹), two times higher than the recommended dose (0.8 g L⁻¹) and three times higher than the recommended dose (1.2 g L⁻¹).

Seed germination and treatment

Seeds of V. faba, V. sativa, and T. foenum-graecum were surface-sterilized with 0.1% mercuric chloride (HgCl₂) solution for 1 min. M. truncatula seeds were additionally chemically scarified with sulfuric acid (H₂SO₄, 95%). All seeds were then thoroughly rinsed three times with sterile distilled water and pre-soaked for 3 h in distilled water. Following pre-soaking, seeds of each species were divided into four treatment groups, namely, distilled water (control) or dimethoate solutions at concentrations of 0.4, 0.8, and 1.2 g L⁻¹.

For the germination test, 20 seeds of each species were evenly placed in Petri dishes (90-mm diamater) lined with two layers of moistened filter paper, with six replicates per treatment. Each dish received 20 mL of either distilled water (control) or the corresponding dimethoate solution. Dishes were incubated in a growth chamber at 25 ± 1°C in the dark for 5 days. Filter papers were kept moist by adding the appropriate treatment solution as needed. A seed was considered germinated when the radicle reached 2 mm in length. Germination percentage was recorded for each treatment after 5 days by using the following formula: (germinated seeds × 100) / total seeds planted.

Seedling growth and dimethoate treatment

Germinated seeds were transferred to square plastic Petri dishes (120 × 120 × 17 mm; Greiner) filled with sterilized sand moistened with 20 mL of either distilled water or dimethoate solution. The seeds were covered with filter paper imbibed with the respective solution to ensure uniform exposure. The experiment was conducted under the following controlled environmental conditions: temperature, 25 ± 1°C; relative humidity, 60–70%; and a photoperiod of 16 h light and 8 h dark. For each treatment, six replicates were used. Each replicate consisted of 20 seeds per species. Seedling length was measured manually with a ruler. On Day 10, the seedlings were harvested and divided into two portions, namely, one for fresh biomass determination and the other for radicle length measurement, whereas the second portion was frozen in liquid N and stored at −80°C for physiological and biochemical analyses. Additionally, a subset of the samples was oven-dried at 60°C for 96 h until a constant weight was achieved, and, subsequently, ground for the determination of total soluble sugar content.

Determination of total soluble sugar content (TSS)

Total soluble sugars were determined by the anthrone sulfuric acid reagent by using L-glucose for the standard curve (Yemm and Willis 1954).

Dry root tissues (25 mg) were homogenized with 5 mL of 80% (v/v) ethanol, incubated at 70°C and centrifuged for 30 min at 3000g, twice. The amount is expressed as milligrams per gram dry weight (mg g⁻¹ DW) through the calibration curve.

Determination of lipid peroxidation (malondialdehyde, MDA)

For MDA content analysis, fresh root tissues (0.5 g) were crushed and homogenized with 1 mL of 5% trichloroacetic acid (TCA) solution (Karabal et al. 2003). The homogenate was then centrifuged at 10.000g for 15 min at 4°C. Next, a 1 mL aliquot of the supernatant was added to 1 mL of a solution containing 20% TCA and 0.5% thiobarbituric acid (TBA). This mixture was incubated at 95°C for 30 min and then cooled immediately in an ice bath. After cooling and centrifuging at 10.000g for 5 min, the specific absorbance of extracts was recorded at 532 nm. Non-specific absorption at 600 nm was recorded at 600 nm and then subtracted from the 532 nm readings. MDA concentration was quantified using the extinction coefficient of 155 mM⁻¹ cm⁻¹.

Extraction of total protein content and determination of antioxidant enzyme activities

Proteins and various antioxidant enzyme activities were investigated for biochemical analysis. Protein content was determined using bovine serum albumin as a standard (Bradford 1976). Fresh tissues (0.5 g) were ground at 4°C in mortar with 2% (w/w) polyvinyl-polypyrrolidone (PVP) and 1 mL phosphate buffer (50 mM; pH 7.8) containing 0.2 mM ethylenediaminetetraacetic (EDTA), 1 mM phenylmethylsulfonyl fluoride (PMSF), as a protease inhibitor, and 0.1% (v/v) Triton X-100. The extract was centrifuged at 12.000g for 20 min at 4°C and the supernatant was used to determine enzyme activities.

Superoxide dismutase (SOD) activity was assessed using the photochemical nitrobluetetrazolium chloride (NBT) (Beauchamp and Fridovich 1971). The reaction mixture contained 50 mM K₂HPO₄/KH₂PO₄ buffer (pH 7.8), 2 mM riboflavin, 13 mM methionine, 75 mM NBT and enzyme extract. One unit of SOD activity was expressed as the amount of enzyme required to inhibit the reduction rate of NBT by 50% at 25°C.

Catalase (CAT) activity was measured by following the decline of absorbance at 240 nm caused by the decomposition of H₂O₂ during 1 min (ε = 36 M⁻¹ cm⁻¹) (Anderson et al. 1995). The assay mixture was composed of 50 mM potassium phosphate buffer (pH 7.0), 10 mM H₂O₂, and 5 μL extract in a total volume of 1 mL.

The activity of GPOX was assessed by the measurement of the kinetic evolution of tetraguaiacol formation from guaiacol (9 mM) at 470 nm during 1 min (ε = 26.6 mM⁻¹ cm⁻¹) in the presence of 19 mM H₂O₂, which was added to initiate the reaction (Anderson et al. 1995).

Statistical analysis

All statistical analyses were conducted using R (ver. 4.3.2; R Core Team 2023). Linear models were fitted for each response variable by using the stats package in R, with species (S), treatment (T), and their interaction (T:S) as fixed effects. Model assumptions were evaluated through the Shapiro–Wilk test of residual normality (Shapiro and Wilk 1965) and Levene’s tests of variance homogeneity (Levene 1960), with both being implemented in the rstatix package (ver. 0.7.2, https://cran.r-project.org/web/packages/rstatix/index.html). These assumptions were met for all variables except germination percentage (GP). Significant effects in two-way ANOVA (α = 0.05) were followed by post hoc pairwise comparisons by using Tukey-adjusted P-values for contrasts via the emmeans package (ver. 1.9.0) (Lenth 2023). The multcomp package (ver. 1.4) (Hothorn et al. 2008) was used to generate compact letter displays for all pairwise comparisons. For GP, we performed aligned rank transformation ANOVA by using the ARTool package (ver. 0.11.1) (Kay et al. 2021), followed by ART-C contrasts (Elkin et al. 2021) for post hoc testing. Results showed that the T:S effect was significant for all studied variables (Table 1). Correlation analyses were conducted using Pearson’s product–moment correlation (Pearson 1895) for normally distributed variables and Spearman’s rank correlation (Spearman 1904) for analyses involving GP, with P-values adjusted using Bonferroni correction. The resulting correlation matrix was visualized using the corrplot package (ver. 0.95) (Wei and Simko 2021). Multivariate patterns were examined through principal component analysis using FactoMineR (ver. 2.11) (Lê et al. 2008), with results being visualized via factoextra (ver. 1.0.7) (Kassambara and Mundt 2020). All graphical representations were created using ggplot2 (ver. 3.5.1) (Wickham 2016), with color optimization through the colorspace package (ver. 2.1-1) (Zeileis et al. 2020).

Effects of dimethoate on seed germination and root length

The germination percentage of all tested species progressively declined with an increasing dimethoate concentration, with M. truncatula exhibiting complete inhibition across all tested concentrations. This extreme sensitivity during germination not only caused a drastic reduction in its germination rates but also necessitated its exclusion from subsequent seedling stage analyses due to its markedly poor performance.

Under control conditions (0 g L⁻¹) and at 0.4 g L⁻¹ dimethoate, both T. foenum-graecum and V. sativa exhibited germination rates greater than 90%. In contrast, V. faba did not achieve 80% germination at the same concentration of 0.4 g L⁻¹ dimethoate (Fig. 1a).

Fig. 1.

Effect of dimethoate on (a) germination percentage, (b) root dry weight, (c) radicle length, and (d) total soluble sugar content in 10-day-old T. foenum-graecum, V. faba, and V. sativa seedlings. (a) Boxplots represent the median of six replicates, with error bars indicating the 95% confidence interval of the median. (b–d) Boxplots display the median (horizontal line), interquartile range (box), individual data points (six replicates), and means (diamonds). Groups sharing the same letter are not significantly different (Tukey-adjusted contrasts, α = 0.05).

At the highest dimethoate concentration of 1.2 g L⁻¹, T. foenum-graecum showed the least significant reduction in germination percentage (17%), followed by V. sativa (20%), whereas V. faba showed a more significant reduction of 50% than did the respective control groups.

In terms of seedling growth, specifically root length and root dry weight, our results showed that dimethoate induced a significant reduction in these parameters in all species, especially at the higher concentration of 1.2 g L⁻¹. The least reduction in root elongation and root dry weight was observed in T. foenum-graecum, with decreases of 53% and 27% respectively, compared with the control. In comparison, V. sativa showed reductions of 59% and 30%, and V. faba showed the most pronounced reductions of 72% and 40% respectively (Fig. 1b, c).

Effects of dimethoate on total soluble sugar content (TSS)

TSS showed a progressive decrease with an increasing dimethoate concentration (Fig. 1d). The degree of reduction was species-dependent, highlighting differences in metabolic resistance to pesticide-induced stress. In T. foenum-graecum, TSS decreased by 18% at 0.8 g L⁻¹ and by 25% at 1.2 g L⁻¹, indicating a relatively higher tolerance to dimethoate. Conversely, V. faba showed a marked sensitivity with a 31% decrease in TSS at 0.8 g L⁻¹ and a 56% decrease at 1.2 g L⁻¹. Similarly, V. sativa showed a reduction of 24% at 0.8 g L⁻¹ and 50% at 1.2 g L⁻¹.

Effects of dimethoate on lipid peroxidation

One of the major biological processes associated with the activity of reactive oxygen species (ROS) is lipid peroxidation, which serves as an indicator of oxidative stress. MDA, a primary by-product of lipid peroxidation and a major metabolite that reacts with thiobarbituric acid (TBA), was significantly elevated in the roots of all three legume species exposed to dimethoate, particularly at the higher concentrations of 0.8 g L⁻¹ and 1.2 g L⁻¹ (Fig. 2a).

Fig. 2.

Effect of dimethoate on (a) malondialdehyde (MDA) content, and the activities of (b) superoxide dismutase (SOD), (c) catalase (CAT), and (d) guaiacol peroxidase (GPOX) in the roots of 10-day-old T. foenum-graecum, V. faba, and V. sativa seedlings. (a–d) Boxplots show the median (horizontal line), interquartile range (box), individual data points (six replicates), and means (diamonds). Groups sharing the same letter are not significantly different (Tukey-adjusted contrasts, α = 0.05).

At a lower concentration (0.4 g L⁻¹), MDA concentration showed a slight decrease in V. faba and V. sativa, while remaining stable in T. foenum-graecum compared with the control. However, at the highest concentration of dimethoate (1.2 g L⁻¹), an increase in MDA content was observed, with increases of 35%, 29% and 35% in V. faba, V. sativa and T. foenum-graecum respectively.

Antioxidant enzymes variation in dimethoate-treated seedlings

Antioxidant enzymes are crucial for the mitigation of oxidative stress. At the seedling stage, the activities of SOD, CAT and GPOX in the three legume species studied showed a significant increase in response to increasing concentrations of dimethoate (Fig. 2b–d). Specifically, at a concentration of 0.4 g L⁻¹, only minor variations were observed in all activities tested for the three species, with the exception of SOD in T. foenum-graecum, which showed a significant induction of about 37%. At the highest concentration of 1.2 g L⁻¹, SOD activity was significantly stimulated (Fig. 2b), with T. foenum-graecum showing an increase of 70%, compared with increases of 41% and 43% in V. faba and V. sativa respectively. In addition, CAT and GPOX activities were significantly increased compared with their controls (Fig. 2c, d), with the most pronounced increase in CAT activity being observed in V. faba, which showed a more than three-fold increase, whereas GPOX activity was significantly higher in V. sativa.

Pearson correlation analysis of growth, and physiological and biochemical traits of V. faba, V. sativa and T. foenum-graecum

The Pearson correlation analysis results showed that all investigated growth parameters (GP, RL and RDW) exhibited a strong positive correlation among themselves. Nevertheless, GP and RL were negatively correlated with SOD (r = −0.7 and r = −0.78 respectively), GPOX (r = −0.67 and r = −0.76 respectively) and CAT (r = −0.63 and r = −0.66 respectively) activities. Additionally, TSS was highly positively correlated with GP (r = 0.52), RL (r = 0.63) and RDW (r = 0.71) and negatively correlated with MDA concentration (r = −0.76), and SOD, GPOX and CAT activities (r = −0.37, r = −0.43 and r = −0.87 respectively) (Supplementary Fig. S1).

Principal component analysis (PCA)

To evaluate the pesticide tolerance capacity of V. faba, V. sativa, and T. foenum-graecum, we conducted a PCA on eight studied variables measured at each dimethoate concentration (0 g L⁻¹, 0.4 g L⁻¹, 0.8 g L⁻¹, and 1.2 g L⁻¹). The PCA results indicated that the first two principal components (PCs) explained more than 66% of the total variance across all treatment levels (Fig. 3a–c). The correlations between the investigated variables and the PCs are detailed in Table 2.

Fig. 3.

Principal component analysis (PCA) of traits in T. foenum-graecum, V. faba, and V. sativa seedlings under dimethoate treatments: 0.4 g L⁻¹ (a), 0.8 g L⁻¹ (b), and 1.2 g L⁻¹ (c). GP, germination percentage (%); RL, root length (cm); DW, root dry weight (mg); TSS, total soluble sugars (mg g⁻¹ DW); MDA, malondialdehyde content (μmol g⁻¹ FW); SOD, superoxide dismutase activity (U SOD mg⁻¹ proteins); CAT, catalase activity (μmol H₂O₂ min⁻¹ mg⁻¹ proteins); GPOX, guaiacol peroxidase activity (nmol H₂O₂ min⁻¹ mg⁻¹ proteins).

At the lowest dimethoate concentration (0.4 g L⁻¹), the PCA biplot explained 66.8% of the total variance. The first axis (PC1 = 45.1%) showed strong correlations with TSS, root RL, and SOD activity, and the second axis (PC2 = 21.7%) was associated with CAT and GPOX activities. Analysis of the score plot showed three distinct groups of species seedlings. Group 1, situated on the right side of the score plot, positively correlated with PC1 and negatively with PC2, and consisted of T. foenum-graecum, characterized by the highest TSS content, SOD induction, and RL. Group 2 was positioned at the upper side of the score plot, negatively correlated with PC1 and positively with PC2, comprising V. faba, which was distinguished by the highest GPOX activity and the lowest GP. Group 3, located on the left side of the score plot, included V. sativa, marked by the highest induction of CAT and the lowest RDW.

At the highest dimethoate concentration (1.2 g L⁻¹), the PCA biplot explained 80.1% of the total variance. Analysis of the score plot showed three groups. Group 1, in the upper-right quadrant, positively correlated with PC1 and negatively with PC2, included T. foenum-graecum, which displayed the highest TSS concentrations, RL, SOD increase, as well as improved GP and RDW. Group 2, located on the left side of the score plot, consisted of V. faba, which exhibited the highest increase in CAT activity. Finally, Group 3, positioned in the lower right quadrant, comprised V. sativa, distinguished by the highest GPOX activity.

Overall, the PCA analysis demonstrated a clear separation among the investigated species. Notably, the biplots for each treatment, even at the highest dimethoate concentration, indicated that T. foenum-graecum was consistently located on the right side of PC1, exhibiting superior values for TSS, SOD activity, and growth indices. This legume displayed greater tolerance to pesticide-induced stress compared to V. faba and V. sativa.