Fluoride effect indicators in Phaseolus vulgaris seeds and seedlings

Plant material and experimental design

Seeds of the super-early cycle carioca bean (P. vulgaris) cultivar BRS FC104 were used in all experiments. This cultivar is widely cultivated in Brazil and indicated for cultivation in irrigation systems. The experiments were carried out in a completely randomized design, consisting of four treatments: 0 (control), 10, 20 and 30 mg L⁻¹ of potassium fluoride (KF), with four replicates.

Germination test

Fifty P. vulgaris seeds by replicate were placed on two germitest sheets moistened with 2.5 times their dry weight with a liquid solution of KF (pH 6.0) at 0 (control), 10, 20 and 30 mg L⁻¹ in order to simulate F concentrations found in irrigation water both in Brazil and worldwide (Abiye, Bybee & Leshomo, 2018; Martins et al., 2018). Then, the seeds were covered with a third germitest sheet already moistened with treatment solutions, rolled up, packed in transparent plastic bags, and maintained in a B.O.D incubator at a constant temperature of 25 °C (±0.5 °C), as described in the Rules for Seed Analysis (2009), for 7 days. Germination was evaluated daily, and the seeds were considered germinated where there was a two mm root protrusion to determine the total germination (TG, %). The germination rate index (GRI) was determined as described by Maguire (1962) as: ${\displaystyle \mathrm{GRI}=\sum _{i=1}^n\left(\frac{ni}{i}\right)}$

where:

ni = number of seeds germinated on day i;

i = number of days.

The stem and root length (cm), and stem diameter of the plants were measured in fourteen seedlings randomly collected per replicate using a digital caliper, 7 days after the beginning of the experiments. In addition, the seedlings were classified as normal, abnormal or dead.

Electrical conductivity test

Seeds cell membrane stability was measured using the electrical conductivity of 50 seeds per replicate. The seeds previously weighed using an analytical balance were immersed in 75 mL of each KF solution (0, 10, 20 and 30 mg L⁻¹) and maintained in a B.O.D incubator at 25 °C. After 24 h, the electrical conductivity of each solution was determined using a conductivity meter (Tec-4MP, Tecnal, Piracicaba, Brazil). The obtained values were divided by the initial sample weight and expressed as µS cm⁻¹ g⁻¹ (Krzyzanowski et al., 2020).

Tetrazolium test

The seeds viability and vigor were performed by the tetrazolium test using 50 seeds for each replicate. The samples were pre-moistened in germitest paper with 2.5 times the dry seed matter with KF solutions (0, 10, 20 and 30 mg L⁻¹) in plastic bags to prevent dehydration, at 25 °C, for 16 h. Subsequently, the seeds were totally immersed in a 0.075% tetrazolium solution in plastic cups and maintained in the dark, followed by heating in an oven at 40 °C for 150 min (Krzyzanowski et al., 2020). After staining, the samples were washed with water and seed vigor was evaluated using a table magnifer lamp (RT301 model; Ritek Electronics Co., China). Each seed was classified according to its types of damage (mechanical, moisture and bed bug damage), and damage location as: highest vigor (class 1), high vigor (class 2), medium vigor (class 3), low vigor (class 4), very low vigor (class 5), unviable (class 6) and dead (class 8), according to Krzyzanowski et al. (2020). Vigor seeds data were obtained by the sum of seeds categorized as classes 1, 2 and 3.

Seed X-ray test

The X-ray test is a non-destructive test that allows checking the presence of internal damage to the seeds caused by insects or mechanical damage. Sixty-four seeds per replicate were immersed in KF solutions (0, 10, 20, and 30 mg L⁻¹) for 48 h. After this time, the seeds were placed on transparent acrylic plates on double-sided adhesive tape and exposed to radiation using a Faxitron HPX-ray system (43855A model, Faxitron X-ray Corp, Wheeling, USA) at an intensity of 30 Kv for 15 s. Digital images were obtained and qualitatively analyzed to assess internal seed morphology.

Morphoanatomical and histochemical characterization

Seed material was collected from the endosperm region 48 h after the seeds be immersed in KF solution (0, 10, 20 and 30 mg L⁻¹), and fixed in Karnovsky’s solution (1965), for 24 h. For morphoanatomical analyses, the seed material was prepared as described by Rodrigues et al. (2020). Transverse sections were stained with toluidine blue-polychromatic (0.05%) in phosphate buffer (0,1 M, pH 6,8) (O’Brien, Feder & McCully, 1964). For histochemical analyses, sample sections were stained with lugol 10 g L⁻¹ (Jensen, 1962) and Xylidine Ponceau (XP) for starch and total protein determinations, respectively (O’Brien & McCully, 1981). Qualitative morphoanatomical and histochemical observations were performed according to Rodrigues et al. (2020).

Malonaldehyde (MDA) and hydrogen peroxide (H₂O₂) concentration

The level of lipid peroxidation was measured by quantifying MDA concentration according to Cakmak & Horst (1991). Seed samples (0.1 g) obtained after 4 days of the germination test were homogenized in 10 mL of trichloroacetic acid (TCA; 0.1% w/v), and centrifuged at 15,000 g, at 4 °C, for 10 min. The supernatant (one mL) was mixed with four mL of thiobarbituric acid solution (TBA; 0.5% w/v of TBA in 20% TCA). The reaction mixture was heated at 95 °C, for 30 min in a water bath. The reaction was stopped in a subsequent ice bath. The samples were centrifuged at 10,000 g, at 4 °C, for 10 min, and the supernatant absorbances were determined at 532 and 600 nm (Heath & Packer, 1968) using a UV-visible spectrophotometer (Evolution 60S model; Thermo Fisher Scientific Inc., MA, EUA). The concentration of MDA was calculated using the molar extinction coefficient of 155 mM⁻¹ cm⁻¹ (Heath & Packer, 1968) according to the following equation: MDA (nmol mL⁻¹) = [(Abs ₅₃₂ −Abs ₆₀₀)/155,000] × 10⁶, and expressed as nmol MDA g⁻¹ fresh weight.

The production of reactive oxygen species was estimated by the H₂O₂ concentration also after 4 days of germination in the F solutions. Seeds samples (0,2 g) were homogenized in liquid nitrogen and then in two mL of potassium phosphate buffer (50 mM, pH 6.5, containing 1 mM hydroxylamine). After filtration, the homogenate was centrifuged at 10,000 g, at 4 °C, for 15 min (Kuo & Kao, 2003). Supernatant Aliquots (50 µL) of the supernatant were added to a reaction medium consisting of ammoniacal ferrous sulfate FeNH₄(SO₄) (100 µM), sulfuric acid (25 mM), xylenol orange (250 µM) and sorbitol (100 mM), in a final volume of two mL (Gay & Gebicki, 2000). After 30 min in the dark, the absorbance of the samples was determined at 560 nm in a UV–VIS spectrophotometer. In parallel, a blank sample were obtained for each sample and subtracted from sample absorbance values. The H₂O₂ concentration was estimated based on an H₂O₂ standard curve, and expressed as µmol g⁻¹ fresh weight.

Fluoride and macronutrients content

Four days after germination teste, F and macronutrients content were analysed. Fluoride determinations was carried out in previously dried and ground seeds soaked with different KF concentrations. Fluoride determination was performed according to Silva et al. (2020).

For macronutrients analyses, seed samples were washed in distilled water and dried in paper bags at 60 °C under induced air circulation conditions until constant mass. The material was ground in a Willey type mill (20-mesh sieve), ashed in a muffle and the minerals extracted by nitroperchloric digestion according to Embrapa (2009). The nutrients calcium (Ca) and magnesium (Mg) were analyzed by atomic absorption spectrophotometry, and potassium (K) using a flame photometry.

Chlorophyll a seedling fluorescence

Photosynthetic efficiency and physiological changes were assessed by the chlorophyll a fluorescence measurement. Variables of chlorophyll fluorescence were measured 7 days after KF solution (0, 10, 20, and 30 mg L⁻¹) on the germination test using a modulated Imaging-PAM fluorometer (Maxi version; Heinz Walz GmbH, Effeltrich, Germany). Image capture and equipment calibration were performed according to Lima et al. (2017); from these, the maximum photosystem II (PSII) quantum yield was calculated [F_v/F_m = (F_m−F₀)/F_m] (Genty, Briantais & Baker, 1989). After sample illumination, saturation pulses were applied to determine the initial fluorescence (F), and the maximum fluorescence (F_m′) in light-acclimated leaves. These variables were used to estimate the apparent electron transport rate [ETR = (F_m′ −F)/F_m′ × PAR × LeafABS × 0.5] (Bilger, Schreiber & Bock, 1995), where PAR is the photon flux (µmol m⁻² s⁻¹) in the leaves, LeafABS is the fraction of incident light absorbed by the leaves, and 0.5 is the fraction of excitation energy directed to the PSII (Laisk & Loreto, 1996). The quenching of regulated non-photochemical dissipation [Y_NPQ = (F_s/F_m′) − (F/F_m)] and the quenching of non-regulated non-photochemical dissipation [Y_NO = F_s/F_m] were calculated according to Genty, Briantais & Baker (1989) and to Hendrickson, Furbank & Chow (2004).

Statistical analyses

The obtained data were submitted to previous analyzes of homogeneity (Levene test) and normality (Shapiro–Wilk test) of the error. Following data normality confirmation, ANOVA was performed, and the treatment means were compared to the control using the Dunnett test, considering 5% (*) and 1% (**) of probability. All analyses were performed using R software version 3.1.2 (R Core Team, 2021). For multivariate analysis, data were initially standardized by square root transformation and scaled by mean centering. Principal component analysis (PCA) and score plots were obtained using the MetaboAnalyst platform version 5.0 (https://www.metaboanalyst.ca).

Germination parameters and electrical conductivity

P. vulgaris seeds did not show statistical differences in total germination (∼93%) when exposed to different KF solutions, compared to the control (Table 1). Seeds vigor level and viability were not affected by KF treatments. However, germination rate index reduced (10%) and electrical conductivity increased (17%) when the seeds were exposed to the highest KF doses (30 mg L⁻¹) when compared to the control (Table 1). Also, a reduction in normal seedlings (7%) and an increase in abnormal seedling (33%) was observed at the highest KF dose (Table 2). Dead seedling was not altered by KF treatments.

X-ray and tetrazolium seed analyses, and initial growth

The X-ray seed images and tetrazolium test revealed ideal morphological P. vulgaris structures and healthy cotyledons and endosperm, with no morphological damage with increasing KF concentrations (Fig. S1, Supplementary Material). However, root length was reduced between 10% and 15%, compared to the control, when exposed to the lowest (10 mg L⁻¹) and the highest (30 mg L⁻¹) KF solutions (Table 2). The stem length was not significantly affected by KF treatments (Table 2).

Anatomical and histochemical seed characterizations

Increasing KF concentrations did not caused cell changes in P. vulgaris endosperms (Figs. 1D, 1G and 1J), compared to the control treatment (Fig. 1A). Starch accumulation, i.e., large areas stained in black by lugol, were observed both in the control (Fig. 1B) and the KF-exposed seeds (Figs. 1E, 1H and 1K), showing no qualitative differences between treatments. Protein-stained areas also indicate that no protein leakage from the endosperm cell occurred after KF treatments (Figs. 1C, 1F, 1I and 1L).

MDA and H₂O₂ contents

KF exposure did not affect the concentration of MDA and H₂O₂ in P. vulgaris seedlings (Fig. S2).

Seed fluoride and macronutrient contents

F content increased in P. vulgaris seeds in 74% and 85% when exposed to 20 and 30 mg L⁻¹ KF, respectively, when compared to the control treatment (Table 3). Calcium (Ca) and potassium (K) seed contents were not significantly altered by KF treatments (Table 3). Mg content decreased by 10% at the highest KF concentration (30 mg L⁻¹), when compared to the control (Table 3).

Chlorophyll a fluorescence in P. vulgaris seedlings

The potential quantum yield of photosystem II (F_v/ F_m) and the electron transport rate (ETR) showed a slight reduction with increasing the KF dose, but it was not this was not statistically significant (Fig. 2). Seed exposure to 30 mg L⁻¹ KF resulted in a reduction in the quenching of regulated non-photochemical dissipation (Y_NPQ) and did not affect the quenching of non-regulated non-photochemical dissipation (Y_NO) in P. vulgaris seedlings (Fig. 2).

Principal component analysis

Analysis of the first three principal components explained in total 74.6% of the total variation observed (Fig. 3). The greatest contribution to the first component (PC1) was observed by abnormal seeds, EC, GRI, and vigor. Dead and abnormal seeds, and also H₂O₂ contributed to PC2 (Fig. 3A). The score plot indicated that there was a high degree of overlap between 10 and 20 mg KF L⁻¹ treatments. However, a clear separation between control treatment and the highest KF dose (30 mg L⁻¹) was observed (Fig. 3B).