Antagonistic effects of EDTA against biochemical toxicity induced by Cr(VI) in Hordeum vulgare L. seedlings

Abstract

The present study aims at the amelioration of chromium Cr(VI) toxicity using ethylenediaminetetraacetic acid (EDTA), and to understand the interactive effects of Cr(VI) and EDTA with respect to seedling growth, lipid peroxidation as assessed from malondialdehyde, pigments and antioxidative enzymes in Hordeum vulgare L. Following multivariate statistical techniques were used to study binary interactions between Cr(VI) and EDTA: 2-way ANOVA, Tukey’s multiple comparison test, multiple regression with interaction between Cr an EDTA, beta coefficients, path analysis and non-metric multidimensional scaling (NMDS). The present study revealed that the EDTA decreases lipid peroxidation induced by Cr(VI) and ameliorates the antioxidative defence system and pigment constitution of seedlings grown in Cr(VI) containing media. EDTA–Cr(VI) interaction decreased the Cr content in the seedlings which may be attributed to the chelating effect of EDTA. The root and shoot bioconcentration factors, the ratio of Cr content in the plant to that in the medium, were decreased by addition of EDTA to Cr(VI), indicating a decrease in the uptake of Cr by the seedlings from the medium. NMDS revealed that the ranking of the studied parameters is maintained by ordination on two axes. The study established that EDTA is antagonistic to Cr(VI) induced biochemical toxicity, and improves the antioxidative defence system, increases the chlorophyll content, and decreases Cr uptake in barley seedlings.

Introduction

Incessant burning of fossil fuels along with uncontrolled release of industrial wastes has resulted in accruing of heavy metals (HMs) in the ecosystems (Govindasamy et al. 2011a, b; Kumar et al. 2019a, b). Some of the HMs are vital for most of the redox reactions which are important for cellular metabolism, although, their levels above tolerable limits lead to the generation of reactive oxygen species (ROS). The ROS are extremely cytotoxic and oxidize various macromolecules, i.e., proteins, lipids and nucleic acids, thus disorganizing the permeability of membranes (Kumar et al. 2020). Consequently, the pile up of ROS results in imbalance of antioxidative defense system in plants leading to oxidative stress (Sharma et al. 2011). Amongst the heavy metals, chromium (Cr) is ubiquitously dispersed in nature, and its harmful effects on morphological and biochemical parameters of plants have been well studied (Kanwar et al. 2015). The compounds of Cr are extensively employed in industry for multifarious purposes, such as leather tanning, steel industry and electroplating, electrical batteries etc. The effluents discharged from industries contaminate our water bodies and soil resources and may affect flora, fauna and human health through biomagnification. Cr is one of the most toxic heavy metals with a natural abundance of 83.8%. Among the stable states of chromium, Cr(III) is more stable than Cr(VI) (Wuana and Okieimen 2011). Cr(III) compounds form octahedral complexes and are the dominant form of Cr (Wuana and Okieimen 2011). Lower doses of biologically active form of Cr(III) complex interact with insulin, and help in the regulation of normal blood sugar levels. On the contrary, Cr(VI) compounds are 1000 times more toxic than Cr(III) compounds for the flora and fauna in the ecosystems due to their solubility in water and are powerful oxidants at low and neutral pH (Vajpayee et al. 2000). Cr(VI) is known to impair the integrity of cell membranes, increases lipid peroxidation, affects the action of enzymes and is genotoxic. Manganese (Mn) present in soil oxidizes Cr(III) to Cr(VI) which remains in the soil for a long time to affect plant growth and development (Sharma et al. 2019).

The process of phytoremediation of HMs is reduced by various factors such as formation of complexes with organic matter and soil properties, i.e., pH, clay and cation exchange capacity (Bali et al. 2020). Diverse chelating agents may be applied to increase the solubility of metals in the soil. Ethylenediaminetetraacetic acid (EDTA) is one of the most efficient chelating agents for enhancing the solubility, complexation and transport of Cr and other heavy metals (Bareen 2012; Chigbo and Batty 2013; Kambhampati 2013; Mani et al. 2015; Habiba et al. 2015; Mahey et al. 2020). EDTA is a chelate ligand with a high affinity constant and forms metal-EDTA complexes. It is purposely added to the soils to sequester metal ions. However, treatment of EDTA stimulates the transport and movement of heavy metals from the roots to shoots of the plants (Meers et al. 2008; Sonia et al. 2019). EDTA may prevent the precipitation of heavy metals in the soil, or alternatively may dissolve heavy metals adsorbed in the sediments. EDTA also inhibits the cellular division and chlorophyll synthesis. Bloem et al. (2017) studied the effects of EDTA in agricultural soils on nutrient uptake. Hordeum vulgare commonly known as Barley was used as a phytoremediation plant in this study. It is a hardy crop and requires less rainfall and grows well where other crops, like wheat, fail (Morel 1997), and can be used for enhanced uptake and translocation of heavy metals. Therefore, this study was planned to analyse the effects of Cr(VI) on the growth and antioxidative enzymes of H. vulgare seedlings in binary combinations with EDTA, and to study the uptake of Cr(VI) by H. vulgare seedlings. The hypothesis tested was that soil amendment of Cr(VI) containing soils with EDTA will chelate with Cr(VI), and make it unavailable to the plants, thus decreasing Cr(VI) induced toxicity by improving the antioxidative potential of seedlings cultured in Cr(VI) containing media.

Materials and methods

Chemicals and reagents

Potassium chromate (K₂CrO₄) for Cr(VI), and other chemicals used in the study were of AR grade, and procured from Sigma-Aldrich, Hi-Media, Merck, S.D. Fine and Qualigens brands.

Seed material and sterilization

The seeds of H. vulgare L. cv. PL-426 were purchased from Punjab Agricultural University, Ludhiana, Punjab. The seeds were surface sterilized by soaking them in 0.5% sodium hypochlorite (NaOCl) solution for 15 min followed by 5 min soaking in percentage of tween-20? and then continuous washing with distilled water to elute NaOCl and tween-20.

Growth of H. vulgare seedlings in laboratory

The seeds were germinated in plastic boxes of diameter 11.7 cm, containing 500 g of sand, which had been acid washed (0.1 N HCl) and rinsed with distilled water. Each treatment comprised of thirty seedlings grown in three boxes with each box containing ten seedlings. An experiment was performed to find the 50% inhibitory concentration (IC₅₀) to define Cr(VI) treatments. Seeds were sown separately for study on each of the growth and biochemical parameter. Seedling growth was studied in all the seedlings, and biochemical analyses were done in triplicate for each treatment.

Chromium treatment

The treatment doses of Cr(VI) were defined by in vitro toxicity experiments on H. vulgare in the laboratory. IC₅₀ was calculated for Cr(VI) by germinating seeds of H. vulgare in sand culture for 7 days. The treatment doses were defined as 0, 0.5, 1.0 and 1.5 mM Cr(VI), and equimolar concentrations of EDTA (Table 1). Observations were made on 7-day old seedlings for growth and biochemical analysis.

Table 1.

Binary combinations of EDTA and Cr(VI) for the growth of H. vulgare seedlings

EDTA (mM)	Cr(VI) (mM)
	0	0.5	1	1.5
0	0/0	0.5/0	1/0	1.5/0
0.5	0/0.5	0.5/0.5	1/0.5	1.5/0.5
1	0/1	0.5/1	1/1	1.5/1
1.5	0/1.5	0.5/1.5	1/1.5	1.5/1.5

Growth parameters

For growth parameters, thirty seedlings were harvested for each treatment. Root and shoot lengths were measured each seedling and observations on fresh weight of roots and shoots were made. The root and shoot samples were dried in hot air oven at 80 °C till constant weight was achieved, and dry weight was recorded. Experiments were performed in triplicate.

Preparation of plant extracts

2 g of 7-day old H. vulgare seedlings were homogenized using liquid nitrogen in ice-chilled 6 ml of 50 mM phosphate buffer (pH 7.8) containing 1 mM EDTA, 1 mM phenylmethanesulfonyl fluoride (PMSF), 0.5% (v/v) triton X-100 and 3% (w/v) PVP 40 in a pre-chilled pestle and mortar. The homogenized mixture was centrifuged at 12,000 g, and 4 °C for 20 min. The supernatant was collected and stored at − 80 °C in 100 µl aliquots for protein content estimation and enzyme assays.

Biochemical parameters

Protein content (Bradford 1976): Coomassie Brilliant Blue G-250 binds with proteins, and the maximum absorbance shifts from 465 to 595 nm in acidic medium. Bradford reagent was prepared by dissolving 100 mg of the dye Coomassie brilliant blue G-250 in 50 ml of ethanol (95% v/v). 100 ml of 85% (v/v) ortho-phosphoric acid was added and the volume was made up to 1l. The dye solution was kept in dark for 12 h, filtered and OD was taken at 595 nm. The protein content (mg g⁻¹ fw) was determined from linear regression equation between bovine serum albumin (BSA) concentration and absorbance.

Ascorbate peroxidase (Nakano and Asada 1981): APX uses ascorbic acid to reduce H₂O₂ to produce monodehydroascorbate and water. APX activity was estimated in terms of rate of oxidation of ascorbate which is determined by decrease in absorbance at 290 nm. The reaction was carried out at 25 °C by addition of 60 μl of the enzyme extract to 50 mM potassium phosphate buffer (pH 7.0) containing EDTA (0.1 mM), ascorbate (0.5 mM) and H₂O₂ (1 mM).

Guaiacol peroxidase (Pütter 1974): GPX catalyses the oxidation of H₂O₂ to H₂O with guaiacol as a substrate. The reaction mixture contains 50 mM phosphate buffer (pH-7), 20 mM guaiacol and 12.3 mM H₂O₂. The rate of formation of guaiacol dehydrogenation product was determined by increase in absorbance at 436 nm for 1 min at 6 s intervals.

Superoxide dismutase (Kono 1978): Hydroxylamine hydrochloride is autooxidized to nitrite by superoxide radicals. The addition of nitro blue tetrazolium chloride (NBT) induces an increase in the absorbance at 540 nm due to the production of blue coloured formazon. With the addition of SOD enzyme, the superoxide radicals get trapped, and hence there is an inhibition of reduction of NBT to blue formazon formation. Three ml of the reaction mixture contained 50 mM sodium carbonate buffer (pH 10.2), 24 μM NBT, 0.03% Triton X-100, 0.1 mM EDTA and enzyme extract (60 μl). Unit enzyme activity is defined as the amount of enzyme which inhibits 50% of NBT reduction at 25 °C.

$$SOD=\frac{Changeinabsorbance{\min }^{-1}\left(\mathit{blank}\right)-Changeinabsorbance{\min }^{-1}\left(\mathit{test}\right)}{Changeinabsorbance{\min }^{-1}\left(\mathit{blank}\right)}$$

where the percent inhibition produced by 60 μl of the sample. Hence, 50% inhibition is produced by (50 × 60)/x = y μl of the sample.

Catalase (Aebi 1984): Catalase causes the decomposition of H₂O₂ to H₂O. The assay utilises decrease in absorbance at 240 nm due to decomposition of H₂O₂ for 30 s at intervals of 3 s. The reaction mixture contains 60 µl of enzyme extract, 15 mM H₂O₂ in 50 mM phosphate buffer. One unit of enzyme activity is the amount of enzyme required to release half of the peroxide oxygen from H₂O₂.

Glutathione reductase (Carlberg and Mannervik 1975): GR catalyzes the reduction of glutathione disulphide (GSSG) accompanied with the oxidation of NADPH. 3 ml of reaction mixture contained 50 mM, phosphate buffer (pH 7.8), 1 mM of EDTA, 0.1 mM of NADPH, 1 mM of GSSG, and 75μl of enzyme sample. Decrease in absorbance per minute was determined at 340 nm. Unit enzyme activity is the amount of enzyme required to catalyze the formation of 1 mM of NADPH min⁻¹ g⁻¹ fw. The specific enzyme activities were calculated as follows:

$$\begin{array}{cc}& Unitenzymeactivity\left(\text{unit}{\min }^{-1}{\text{g}}^{-1}\text{fw}\right)\hfill \\ \hfill & =\frac{Changeinabsorbance{\min }^{-1}\times Totalvol.\left(\text{ml}\right)}{Extinctioncoeff.\times vol.ofthesampletaken\left(\text{ml}\right)\times wt.ofplant\left(\text{g}\right)}\hfill \\ \hfill \end{array}$$

$$Specificenzymeactivity\left(\text{units}{\min }^{-1}{\text{mg}}^{-1}\text{protein}\right)=\frac{Unitenzymeactivity}{proteincontent\left(\text{mg}{\text{g}}^{-1}\right)}$$

The extinction coefficients taken for calculation of enzyme activities were as follows: APX = 2.8 mM⁻¹ cm⁻¹, GPX = 25.5 mM⁻¹ cm⁻¹, CAT = 39.4 M⁻¹ cm⁻¹, GR = 6.22 mM⁻¹ cm⁻¹.

Chlorophyll content (Arnon 1949): 200 mg of fresh leaves from each of the samples were homogenized with liquid nitrogen and 5 ml of 80% acetone added to it. The reaction mixture was incubated in dark for 1 h at 25 °C, and centrifuged at 14,000 rpm at 4 °C for 15 min in Eltek centrifuge. Absorbance was taken at 645 and 663 nm on UV–Visible spectrophotometer (Systronics 2202) using 80% acetone as blank.

Carotenoid content (Lichtenthaler 1987): 200 mg of fresh leaves from each sample were homogenized with liquid nitrogen and 5 ml of 80% acetone. The mixture was centrifuged at 14,000 rpm in Eltek centrifuge at 4 °C for 15 min. Absorbance was taken at 470 nm on UV–Vis spectrophotometer (Systronics 2202) against 80% acetone as blank. The equations used for determination of pigments are given below:

$$Chl\left(\text{mg}{\text{g}}^{-1}\text{fw}\right)=(20.2O{D}_{645}+8.02O{D}_{663})\left(\frac{v}{1000w}\right)$$

$$Chl\text{`}a\text{'}\left(\text{mg}{\text{g}}^{-1}\text{fw}\right)=(12.7O{D}_{663}-2.69O{D}_{645})\left(\frac{v}{1000w}\right)$$

$$Chl\text{`}b\text{'}\left(\text{mg}{\text{g}}^{-1}\text{fw}\right)=(22.9O{D}_{645}-4.68O{D}_{663})\left(\frac{v}{1000w}\right)$$

$$Car\left(\text{mg}{\text{g}}^{-1}\text{fw}\right)=\frac{1000OD\left(470\right)-1.82Chla-85.02Chlb}{198}\left(\frac{v}{1000w}\right)$$

where OD = optical density, w = fresh weight of the sample in g, v = volume of the plant extract in ml.

Malondialdehyde content (MDA) (Heath and Packer 1968): One g of H. vulgare seedlings was homogenized in 3 ml pre-chilled 0.1% TCA and subsequently 3 ml of 0.5% thiobarbituric acid (TBA) in 20% trichloroacetic acid (TCA) solution was added. After centrifugation, the absorbance of the supernatant was measured at 532 nm, and from it, absorbance at 600 nm was subtracted to account for nonspecific absorption. The extinction coefficient used was 155 mM⁻¹ cm^−1.

Cr uptake: 0.2 g of dried plant samples were weighed and digested in a mixture of H₂SO₄:HNO₃:HClO₄ digestion mixture in the ratio of 1:2:1 as proposed by Allen et al. (1976). The digests were diluted with double distilled water to 20 ml. The Cr concentration was determined by atomic absorption spectrophotometer (Shimadzu AA-6200). Shoot to root ratio of metal uptake was calculated to evaluate translocation factor as given by Gautam and Agrawal (2017).

Statistical analysis

All the experiments were performed in triplicate and the results were expressed in mean ± standard error. The data were analysed using the following techniques:

ANOVA and Tukey’s test: ANOVA is a multivariate statistical technique to compare the significance of differences among the means of samples. Normal distribution of the data was checked using Shapiro–Wilk test (statskingdom.com 2019) and the normally distributed data were subjected to two-way analysis of variance (ANOVA) using self-coded software in MS-Excel. The null hypothesis was checked that there are no significant differences between the means, and the alternate hypothesis was that the means of at least two samples are significantly different from each other. p value was calculated using statistical tables and online software (Strangroom 2018). ANOVA gives the significance of differences among the treatments (main effects) i.e., Cr(VI) and EDTA, and the interaction between them, but does not describe the differences among pairs of means. The means of the samples were compared using post hoc Tukey’s multiple comparison test. Self-coded Tukey’s HSD (Honestly significant difference) statistic was calculated for each set of treatments for a given parameter. Differences among the means which are more than the HSD value; are significantly different from each other at p ≤ 0.05. In order to compare the means of each treatment, HSD values were calculated to compare the differences among the means, both for rows and columns for a given parameter using self-coded software. This does not require separate Tukey’s tests to compare the means separately for rows and columns.

Multiple regression analysis: The data were analysed using self-coded software for multiple regression analysis for trend analysis (Sokal and Rohlf 1981). In this technique the dependent variable, for example the chlorophyll content, is regressed on the concentrations of Cr(VI) and EDTA in binary combinations. In order to understand if EDTA affects the toxicity of Cr(VI), interaction between Cr(VI) × EDTA was also included in the multiple regression analysis with interaction (MRI). Using the MRI equation, we can determine the quantitative effects of independent variables and their interaction on the dependent variable. MRI can be used for interpolation studies and gives absolute change in the dependent variable due to the independent variables. The generalised equation for MRI is given as below:

$$Y=a+b_1{X}_1+b_2{X}_2+b_3{X}_1{X}_2$$

where Y = dependent variable; X₁ = Cr(VI) concentration, X₂ = EDTA concentration; X₁X₂ = interaction between independent variables; a = y-intercept; b_i = partial regression coefficient of X_i eliminating the effects of the other independent variables. The partial regression coefficients b₁ and b₂ represent the rate of change of dependent variable, Y, per unit increase in Cr(VI) and EDTA, and b₃ represents partial regression coefficient due to their interaction.

β-regression coefficients: The values of β-regression coefficients were computed to determine the relative effects of independent variables on the dependent variables using self-coded software. Beta coefficients are unitless quantities which provide insight into the mechanism of different reactions and the effects of EDTA on Cr(VI). β₁, β₂ and β₃ represent the relative effects of Cr(VI), EDTA and Cr(VI) x EDTA on the dependent variable. Higher the value of β, more is the effect of that variable on the response variable irrespective of the units of variables.

Path analysis: The data were also subjected to path analysis (Sokal and Rohlf 1981) to estimate the direct and indirect effects of independent variables on the dependent variable using self-coded software. Path analysis is a technique to determine as to what extent the independent variable affects directly the dependent variable, and to what extent it affects the dependent variable indirectly through the other independent variable. Contour graphs were plotted with the help of Minitab version 14.0 (Pennsylvania, USA) computer software programs.

Non-metric multidimensional scaling (NMDS): The data was subjected to NMDS using PAST-3 software as given by Taguchi and Oono (2004). NMDS is a non-parametric technique in which data in n-dimensions is compressed to 2-dimensions. A stress level of less than 0.05 implies that the ranking of the variables is maintained when they are compressed into 2- or 3 dimensions. The environmental variables declared were Cr(VI) and EDTA. The measure of similarity used was Euclidean distance.

Results

Growth parameters for 7-day old H. vulgare seedlings

Relative changes in shoot lengths of H. vulgare seedlings cultured in sand containing different concentrations of Cr(VI) and EDTA are given in Table 2. Reduction in the shoot and root lengths of 7-day old seedlings was observed with an increase in Cr(VI) and EDTA concentrations in sand culture. Except for shoot dry weight and interaction of EDTA and Cr(VI) for shoot dry weight, two-way ANOVA for root and shoot length and dry weight showed statistically significant differences among the mean values with all the sources of variations. Multiple regression analysis for Cr(VI) and EDTA (Table 3) for shoot length revealed significant correlation for all seedling growth parameters with Cr(VI) in binary combinations with EDTA. Negative β₁ for all the growth parameters showed the harmful effects of Cr(VI) on the growth of seedlings. β₂ coefficients for EDTA showed harmful effects of EDTA on growth of seedlings with increasing concentrations. EDTA interaction with Cr(VI) slightly improved root length as revealed from β₃ coefficients. Path analysis was performed on the growth parameters for binary combination of Cr(VI) and EDTA (Table 4). It was found that Cr(VI) has damaging effects on growth of 7-day old H. vulgare seedlings for all the parameters as the direct effects of Cr(VI) have negative values. The positive values of indirect effects obtained in path analysis for root length in the seedlings grown in Cr(VI) containing sand revealed that EDTA has some positive effect on their growth. Similarly, EDTA had direct negative effects for all the growth parameters of the 7-day old H. vulgare seedlings.

Table 2.

Seedling growth (mean ± SE, n = 30) of 7-day old H. vulgare seedlings in binary combinations of Cr(VI) and EDTA

EDTA (mM)	Cr(VI) (mM)
	0	0.5	1	1.5
Shoot length (cm)
0	18.3 ± 0.363a	15.6 ± 0.446b	9.0 ± 0.211ef	7.6 ± 0.091gh
0.5	18.4 ± 0.171a	12.9 ± 0.292c	9.1 ± 0.088ef	6.6 ± 0.139h
1	17.7 ± 0.096a	11.8 ± 0.261cd	9.2 ± 0.069e	7.2 ± 0.044gh
1.5	17.8 ± 0.069a	11.2 ± 0.356d	8.0 ± 0.184 fg	5.3 ± 0.171i
Root length (cm)
0	12.3 ± 0.685b	5.5 ± 0.441d	1.2 ± 0.168gh	1.2 ± 0.186gh
0.5	14.4 ± 0.611a	3.1 ± 0.498e	1.2 ± 0.058gh	0.7 ± 0.033i
1	12.1 ± 0.250b	2.1 ± 0.236efg	1.7 ± 0.315fgh	0.8 ± 0.133h
1.5	10.6 ± 0.107c	2.6 ± 0.077ef	1.4 ± 0.102gh	0.8 ± 0.096h
Shoot dry weight (mg/seedling)
0	12.7 ± 0.404a	11.5 ± 0.992ab	8.7 ± 0.208abcd	8.3 ± 0.416bcd
0.5	11.9 ± 0.721ab	10.9 ± 0.153abcd	9.2 ± 0.208abcd	7.9 ± 0.624bcd
1	10.8 ± 0.833abcd	11.0 ± 2.179abcd	8.6 ± 0.656abcd	7.3 ± 0.321d
1.5	11.8 ± 1.193abc	9.7 ± 0.346abcd	7.4 ± 0.379cd	7.0 ± 0.321d
Root dry weight (mg/seedling)
0	9.8 ± 0.231b	6.2 ± 0.346c	4.0 ± 0.416e	2.3 ± 0.208fg
0.5	12.1 ± 0.351a	5.8 ± 0.208cd	4.1 ± 0.153e	2.2 ± 0.252fg
1	12.4 ± 0.265a	4.8 ± 0.321cde	3.8 ± 0.416ef	1.4 ± 0.153g
1.5	9.6 ± 0.666b	4.3 ± 0.208de	3.6 ± 0.306ef	1.5 ± 0.208g

Source of variation	Shoot length	Root length	Shoot dry weight	Root dry weight
Two-way ANOVA F-ratios and Tukey’s multiple comparison test (HSD)
Between Cr(VI)	2025.3***	3386.3***	28.2***	599.5***
Between EDTA	61.9***	36.5***	4.9ns	11.5***
Interaction Cr(VI) × EDTA	16.5***	34.6***	1.3ns	6.7***
HSD	1.173	0.958	4.16	1.678

***p ≤ 0.001, ns: not significant at p < 0.05. There is no significant difference (p < 0.05) between the means of the samples followed by the same letter both across rows and columns of a parameter using Tukey’s HSD

Table 3.

Multiple regression equations for seedling growth in binary combination of Cr(VI) (X₁, mM) and EDTA (X₂, mM)

Treatments	Multiple regression with interaction	r	β regression coefficients
			β1	β2	Interaction β3
Shoot length (cm)	Y = 18.3 − 7.6 X1 − 1.3 X2 − 0.08 X1X2	0.9736***	− 0.95	− 0.16	− 0.01
Root length (cm)	Y = 11.3 − 8.2 X1 − 1.7 X2 + 1.2 X1X2	0.8697***	− 0.97	− 0.21	0.17
Shoot dry weight (mg/seedling)	Y = 12.5 − 2.9 X1 − 0.87 X2 − 0.052 X1X2	0.9630***	− 0.91	− 0.27	− 0.02
Root dry weight (mg/seedling)	Y = 10.2 − 5.7 X1 − 0.48 X2 − 0.14 X1X2	0.9296***	− 0.90	− 0.08	− 0.03

***p ≤ 0.001

Table 4.

Path analysis for the effects of Cr(VI) and EDTA on seedling growth of H. vulgare seedlings

Parameter	Independent variables	Direct effects	Indirect effects	Total effects Ryx
Shoot length	Cr(VI)	− 0.95	− 0.007	− 0.96
	EDTA	− 0.16	− 0.007	− 0.17
Root length	Cr(VI)	− 0.97	0.11	− 0.86
	EDTA	− 0.21	0.11	− 0.1
Shoot dry weight	Cr(VI)	− 0.91	− 0.01	− 0.92
	EDTA	− 0.27	− 0.01	− 0.28
Root dry weight	Cr(VI)	− 0.91	− 0.02	− 0.92
	EDTA	− 0.08	− 0.02	− 0.1

Biochemical parameters for 7-day old H. vulgare seedlings

Variations in protein content, specific activities of antioxidative enzymes and MDA content for 7-day old seedlings of H. vulgare in binary combinations of Cr(VI) and EDTA are given in Fig. 1. F-ratios of two-way ANOVA (Table 5) revealed that except for ascorbate peroxidase, the other parameters like protein content, antioxidative enzymes and MDA were significantly affected by Cr(VI). Similarly, except for SOD all other parameters were significantly affected by EDTA. The interaction between Cr(VI) and EDTA was significant for protein content, catalase, glutathione reductase, and root MDA content. Effects of Cr(VI) in binary combination with EDTA on pigments are given in Fig. 2. Except for interaction of Cr(VI) and EDTA for chlorophyll ‘b’, F-ratios for all other parameters were significant in two-way ANOVA (Table 6). Multiple regression analysis with interaction is given in Table 7. All correlation coefficients, excluding catalase, were found to be significant. Positive values of β₁ coefficient revealed increments in the levels of protein content, CAT, GR, shoot and root MDA contents, total chlorophyll, chlorophyll ‘a’, chlorophyll ‘b’, and carotenoid contents. On the other hand, APX, GPX, SOD, GR, root MDA and chlorophyll ‘b’ were positively affected by EDTA as revealed by the β₂ coefficient. The interaction between Cr(VI) and EDTA decreases the shoot and root MDA, but increases chlorophyll ‘b’ and carotenoid contents indicating the stress-protective effects of EDTA. Path analysis was also performed on the biochemical parameters (Table 8). Path analysis is a statistical technique which describes the causal effects of two or more independent variables, Cr(VI) and EDTA in the present study, on a response variable, such as an enzyme. The effect of each independent variable can be decomposed into two components, direct effects and indirect effects. Direct effects of an independent variable are attributable to its direct interaction with the response variable, whereas the indirect effects imply that the independent variable is acting on the response variable through the other independent variable. Positive direct or indirect effects imply an enhancement in the studied response, and the negative values of path coefficients reveal negative effects. It was found that Cr(VI) had direct negative effects on the biochemical parameters of H. vulgare seedlings for ascorbate peroxidase, guaiacol peroxidase, total chlorophyll content, chlorophyll ‘a’ and ‘b’, and carotenoid contents. The positive values of indirect effects in path analysis for Cr(VI) revealed that EDTA has positive effects on the biochemical parameters of the seedlings when Cr(VI) interacted with EDTA. Similarly, the values of direct effects for EDTA were positive for ascorbate peroxidase, guaiacol peroxidase, SOD, catalase, glutathione reductase, total chlorophyll, chlorophyll ‘b’, carotenoid, and root MDA contents.

Fig. 1.

Protein content, antioxidative enzyme activities and MDA contents of 7-day old H. vulgare seedlings in binary combination of Cr(VI) and EDTA

Table 5.

Two-way ANOVA F-ratios and Tukey’s multiple comparison test (HSD) for protein content, antioxidative enzymes and MDA content of 7-day old H. vulgare seedlings in binary combinations of Cr(VI) and EDTA

Sources of variation	Protein content (mg/g FW)	Ascrbate peroxidase (µ mole min−1 mg−1 protein)	Guaiacol peroxidase (µ mole min−1 mg−1 protein)	Superoxide dismutase (UA min−1 mg−1 protein)	Catalase (µ mole min−1 mg−1 protein)	Glutathione reductase (µ mole min−1 mg−1 protein)	Shoot MDA (µ mole g−1 fw)	Root MDA (µ mole g−1 fw)
Between Cr(VI)	286.45***	1.40ns	10.60***	10.54***	14.96***	5.635**	14.84***	10.03***
Between EDTA	247.03***	4.27**	100.54***	0.5054ns	1.02**	5.530**	18.59***	3.49*
Cr(VI) x EDTA	8.51***	0.35ns	1.61ns	1.672ns	3.45***	7.289***	1.45ns	6.73***
HSD	1.964	7.617	31.43	0.328	0.160	2.855	0.923	0.406

Fig. 2.

Photosynthetic pigment contents of 7-day old H. vulgare seedlings in binary combinations of Cr(VI) and EDTA

Table 6.

Two-way ANOVA F-ratios and Tukey’s multiple comparison test (HSD) for photosynthetic pigments in H. vulgare seedlings in binary combinations of Cr(VI) and EDTA

Source of variation	Total chlorophyll (mg g−1 fw)	Chlorophyll ‘a’ (mg g−1 fw)	Chlorophyll ‘b’ (mg g−1 fw)	Carotenoid content (mg g−1 fw)
Between Cr(VI)	465.85***	169.75***	75.50***	3030.15***
Between EDTA	242.65***	107.55***	27.22***	3680.91***
Interaction Cr(VI) × EDTA	38.93***	30.16***	1.77ns	975.49***
HSD	0.055	0.052	0.030	0.303

***p ≤ 0.001, **p ≤ 0.01, *p ≤ 0.05. There is no significant difference (p < 0.05) between the means of the samples followed by the same letter both across rows and columns of a parameter using Tukey’s HSD

Table 7.

Multiple regression with interaction equations for the antioxidative enzymes, MDA content and pigments for the 7-day old H. vulgare seedlings treated with Cr(VI) (X₁, mM) and EDTA (X₂, mM)

Treatments	Multiple regression with interaction	r	β regression coefficients
			β1	β2	Interaction β3
Protein content	Y = 19.34 + 5.53 X1 − 3.59 X2 – 0.94 X1X2	0.9499***	0.81	− 0.53	− 0.16
Ascorbate peroxidase	Y = 11.24 − 1.65 X1 + 1.83 X2 + 0.547 X1X2	0.8815***	− 0.56	0.62	0.22
Guaiacol peroxidase	Y = 43.30 − 5.68 X1 + 48.68 X2 − 10.55 X1X2	0.8822***	− 0.12	0.99	− 0.25
Superoxide dismutase	Y = 0.3377 − 0.056 X1 + 0.121 X2 − 0.159 X1X2	0.6041*	0.31	0.67	− 1.05
Catalase	Y = 0.136 + 0.10 X1 + 0.07 X2 − 0.097 X1X2	0.5120	0.85	0.56	− 0.94
Glutathione reductase	Y = 2.68 + 3.49 X1 + 3.47 X2 − 3.41 X1X2	0.9544***	1.44	1.43	− 1.68
Shoot MDA content	Y = 2.80 + 0.72 X1 − 0.296 X2 − 0.373 X1X2	0.9029***	0.86	− 0.35	− 0.53
Root MDA content	Y = 1.55 + 0.54 X1 + 0.34 X2 − 0.46 X1X2	0.9009***	1.50	0.96	− 1.54
Total chlorophyll	Y = 0.718 − 0.207 X1 + 0.016 X2 + 0.1031 X1X2	0.7925***	− 0.96	0.07	0.57
Chlorophyll ‘a’	Y = 0.588 − 0.166 X1 − 0.023 X2 + 0.1146 X1X2	0.7291**	− 1.0	− 0.14	0.83
Chlorophyll ‘b’	Y = 0.13 − 0.04 X1 + 0.04 X2 − 0.012 X1X2	0.9348***	− 0.64	0.63	− 0.21
Carotenoid content	Y = 0.185 − 0.07 X1 + 0.0006 X2 + 0.051 X1X2	0.8900***	− 1.07	0.009	0.94

***p ≤ 0.001; **p ≤ 0.01; *p ≤ 0.05

Table 8.

Path analysis for antioxidative enzymes, MDA content and pigments for the effects of Cr(VI) and EDTA on H. vulgare seedlings

Parameter	Independent variable	Direct effects	Indirect effects	Total effects Ryx
Protein content	Cr(VI)	0.81	− 0.103	0.71
	EDTA	− 0.53	− 0.103	− 0.63
Ascorbate peroxidase	Cr(VI)	− 0.563	0.140	− 0.422
	EDTA	0.626	0.140	0.766
Guaiacol peroxidase	Cr(VI)	− 0.116	− 0.161	− 0.276
	EDTA	0.990	− 0.161	0.829
Superoxide dismutase	Cr(VI)	0.311	− 0.661	− 0.350
	EDTA	0.671	− 0.661	0.011
Catalase	Cr(VI)	0.850	− 0.592	0.258
	EDTA	0.567	− 0.592	− 0.026
Glutathione reductase	Cr(VI)	1.440	− 1.056	0.384
	EDTA	1.435	− 1.056	0.379
Shoot MDA content	Cr(VI)	0.865	− 0.334	0.531
	EDTA	− 0.353	− 0.334	− 0.687
Root MDA content	Cr(VI)	1.507	− 0.963	0.544
	EDTA	0.964	− 0.963	0.0006
Total chlorophyll content	Cr(VI)	− 0.964	0.360	− 0.603
	EDTA	0.077	0.360	0.438
Chlorophyll ‘a’	Cr(VI)	− 1.00	0.521	− 0.484
	EDTA	− 0.14	0.521	0.381
Chlorophyll ‘b’	Cr(VI)	− 0.650	− 0.137	− 0.787
	EDTA	0.631	− 0.137	0.493
Carotenoid content	Cr(VI)	− 1.077	0.591	− 0.485
	EDTA	0.009	0.591	0.601

Metal uptake for 7-day old H. vulgare

A significant increase (85.3%) in the mean shoot Cr content of 7-day old H. vulgare seedlings was observed with respect to 1.5 mM Cr(VI) concentration in sand culture (Fig. 3a). Seedlings grown in 0.5 mM Cr(VI) showed a significant increase in Cr uptake content (61.76%) on the application of 1.5 mM EDTA. Mean root Cr content increased significantly by 2950% at 1.5 mM Cr(VI) concentration with respect to the control (Fig. 3b). Seedlings grown in 1 and 1.5 mM Cr(VI) showed a reduction in their mean root Cr contents by 56.1 and 76.8% on application of 1.5 mM EDTA with respect to 1 and 1.5 mM Cr(VI) treatments. It was observed that with increase in the concentration of EDTA, there was a significant decrease in the shoot bioconcentration factor (BCF) of the seedlings by 63.56% in a binary combination of 0.5 mM Cr(VI) and 1.5 mM EDTA (Fig. 3c). A significant increase in the root BCF (78.85%) was found with an increase in Cr(VI) concentration. At binary combination of 0.5 mM Cr(VI) and 1.5 mM EDTA, the root BCF of the seedlings increased by 62.3% with respect to 0.5 mM Cr(VI) only (Fig. 3d). At higher Cr(VI) concentration (1.5 mM), root BCF decreased by 43.5% with respect to 1.5 mM Cr(VI) treatment. There was no significant effect of binary treatments on Cr translocation factor in H. vulgare seedlings (Fig. 3e). Except for EDTA, shoot Cr content, and shoot BCF, the F-ratios of two-way ANOVA are statistically significant for all sources of variation (Table 9). Correlation coefficients were observed to be statistically significant for all metal uptake parameters (Table 10). Positive β₁ and β₂ coefficients for Cr(VI) and EDTA showed their positive effects on the metal uptake with increasing concentrations of Cr(VI) and EDTA. Interactions between Cr(VI) and EDTA (β₃) were found to be negative for all parameters indicating that application of EDTA hinders the uptake of Cr in H. vulgare seedlings. Shepard plots for seedlings grown in binary combinations of Cr(VI) and EDTA are given in Fig. 4. Shepard plot is a plot between the observed and target ranks using NMDS. In NMDS the distances between the observed points in n-dimensions are ranked and arranged in 2-dimensions. Path analysis (Table 11) revealed that Cr(VI) and EDTA have direct positive effects on shoot and root Cr contents, root and shoot BCF values. However, the indirect effects of Cr(VI) and EDTA were negative for shoot and root Cr contents, and shoot and root BCF.

Fig. 3.

Cr content (mg/g dw) (mean, n = 3) of 7-day old H. vulgare seedlings in binary combinations of ascorbic acid (AA) and Cr(VI). a Shoot metal content (SMU), b root metal content (RMU), c shoot BCF, d root BCF (RBCF), and e translocation factor (TF)

Table 9.

Two-way ANOVA F-ratios and Tukey’s multiple comparison test for metal uptake content of 7-day old H. vulgare seedlings in binary combinations of Cr(VI) and EDTA

Cr + EDTA (mM)	Shoot Cr content (mg g−1 fw)	Root Cr content (mg g−1 fw)	Shoot BCF (mg g−1 fw)	Root BCF (mg g−1 fw)	Translocation factor
Cr(VI)	367.6***	434.0***	171.1***	17.6***	28.1***
EDTA	1.38ns	28.7***	2.69ns	23.2***	3.85*
Cr(VI) × EDTA	12.67***	29.9***	12.4***	31.2***	5.16**
HSD	0.018	0.043	0.485	0.827	0.218

***p ≤ 0.001, **p ≤ 0.01, *p ≤ 0.05, ns: not significant at p < 0.05

Table 10.

Multiple regression with interaction equations for root and shoot Cr uptake contents for the 7-day old H. vulgare seedlings treated with Cr(VI) (X₁, mM) and EDTA (X₂, mM)

Treatments	Multiple regression with interaction	r	β regression coefficients
			β1	β2	Interaction β3
Shoot Cr content	Y = 0.006 + 0.042 X1 + 0.009 X2 − 0.014 X1X2	0.8104***	1.04	0.23	− 0.40
Root Cr content	Y = − 0.004 + 0.17 X1 + 0.024 X2 − 0.07 X1X2	0.9658***	1.29	0.18	− 0.64
Shoot bioconcentration factor	Y = 0.19 + 0.65 X1 + 1.96 X2 − 1.7 X1X2	0.8956***	0.59	1.8	− 1.83
Root bioconcentration factor	Y = 0.22 + 2.5 X1 + 2.3 X2 − 2.7 X1X2	0.9080***	1.13	1.08	− 1.45
Translocation factor	Y = 0.86 − 0.45 X1 + 0.05 X2 + 0.009 X1X2	0.9312***	− 0.93	0.15	0.03

***p ≤ 0.001

Fig. 4.

Shepard plot for seedlings grown in binary combinations of Cr(VI) and EDTA. a Biochemical analysis, b pigments, c root and shoot MDA content

Table 11.

Path analysis for root and shoot Cr contents for the effects of Cr(VI) and EDTA on H. vulgare seedlings

Parameter	Independent variables	Direct effect	Indirect effects	Total effect Ryx
Shoot Cr content	Cr(VI)	1.04	− 0.25	0.788
	EDTA	0.228	− 0.25	− 0.024
Root Cr content	Cr(VI)	1.29	− 0.402	0.892
	EDTA	0.183	− 0.402	− 0.219
Shoot bioconcentration factor	Cr(VI)	0.588	− 0.827	0.234
	EDTA	− 1.83	− 1.51	0.494
Root bioconcentration factor	Cr(VI)	1.13	− 0.658	0.687
	EDTA	1.08	− 1.20	0.303

Discussion

Chromium is highly toxic heavy metal and its concentration is hastily growing in the ecosystem owing to incessant fossil fuels burning and exoneration of industrial wastes (Govindasamy et al. 2011a, b). Chromium stress results in alteration of biochemical parameters in H. vulgare L. such as chlorophyll content, biomass, uptake of important elements, reduction in root and shoot growth, chlorosis and reduction in water potential (Sharma et al. 2019). In this paper, we attempted to assess how EDTA controls the Cr(VI) mediated changes in barley plant growth and biochemical alterations. EDTA was chosen for the present study because it is well known chelating agent and binds with metals. It is also used in medicine to treat several human ailments like heavy metal poisoning, stroke, high blood pressure, atherosclerosis, angina etc. (WebMD 2019). In areas inundated with effluents containing heavy metals, EDTA would be a better candidate for assisted Cr(VI) phytoremediation studies, for the reasons of its being relatively safe due to its proven medical uses. It is inferred from the results, that Cr(VI) causes sharp reduction in seedling growth and EDTA does not improve seedling growth. The reduction in root and shoot length and their respective weights with Cr(VI) treated H. vulgare seedlings in the present study can be attributed to the oxidative stress, alteration in germination process and deleterious effects on metabolic processes such as photosynthesis, mineral nutrition imbalance caused by Cr(VI) in roots and shoots of plants as reported by various workers (Shanker et al. 2005; Zou et al. 2006; Stambulska et al. 2018).

In the present work, an increase in the protein content per g fresh weight of the seedlings was observed on treatment with Cr(VI). Since protein content is a ratio of proteins to the fresh weight of the seedlings, the decline in fresh weight of the seedlings increased the protein content of the seedlings on fresh weight basis in Cr(VI) treatments. Under Cr(VI) stress, plant cells use antioxidant defense mechanism to reduce the oxidative damage. The antioxidant enzymes such as SOD, CAT, POD, APX and GR control the cellular superoxide (O₂⁻) and hydrogen peroxide (H₂O₂) contents, thereby reducing the generation of ^–OH radicals (Rucinska-Sobkowiak and Pukacki 2006). SOD and CAT are substantive in eliminating oxidative stress (Gomes-Junior et al. 2006). Gwózdz et al. (1997) reported that at lower heavy metal concentrations the activities of antioxidant enzymes increased, whereas at higher concentrations there was no further increase in the SOD activity, although the CAT activity decreased. The findings on the antioxidative enzymes in the present study indicated that Cr(VI) stress increased the GR activity, and increased the root and shoot MDA contents in H. vulgare seedlings. However, EDTA amendment to Cr(VI) containing sand enhanced the activity of GPX, and decreased the shoot and root MDA contents. The beta regression coefficients (β₃) for interaction between Cr(VI) and EDTA, except for GPX, were opposite to the effects of Cr(VI) (β₁), thus revealing antagonistic effects of EDTA against Cr(VI). Decrease in the MDA contents is indicative of stress amelioration by EDTA, and can be attributed to the chelation of Cr(VI) by EDTA, thus protecting the seedlings. Pandey et al. (2005) observed the same effects on the growth of Brassica juncea when treated with hexavalent chromium. The authors also reported the elevation in CAT and GR activities with increase in Cr(VI) concentration. According to Shanker (2003), induction and activation of SOD and CAT are some of the major metal detoxification mechanisms in plants. Kanwar et al. (2015) reported elevation in enzymatic activity in B. juncea when exposed to Cr(VI). These results support our findings showing the same enzymatic trends with chromium exposed barley seedlings. Enhancement in the GPX activity with increase in Cr(VI) concentration is also coherent with the findings of previous researchers. Pigments play an important role in photosynthetic activities and metabolic processes associated with plant growth. In the present study, increasing Cr(VI) concentration significantly decreased the chlorophyll content, but on application of EDTA, augmentation in the level of chlorophyll content was observed at lower concentration of Cr(VI). β₃ coefficients for interaction between EDTA and Cr(VI) for Chl a and total chlorophyll and carotenoids were positive, whereas β₁ for Cr(VI) for these pigments were negative. It indicates that EDTA application improves the pigment constitution of the barley seedlings. Disorganization of chloroplast ultrastructure and inhibition of electron transport processes due to Cr, and diversion of electrons from the electron-donating side of photosystem (PS1) to Cr(VI) is a possible explanation for Cr-induced decrease in photosynthetic rate (Shanker et al. 2005). Our results are in consonance with Das et al. (2014) on wheat seedlings in binary combination with Cr(VI) and EDTA. The authors reported a decrease in chlorophyll content in hydroponically cultured wheat seedlings on Cr(VI) treatment, and that EDTA increased the chlorophyll content in Cr(VI) treated seedlings.

It was observed that shoot and root MDA contents were significantly increased with elevation of Cr(VI) concentration. Same results were reported by Panda and Choudhury (2005) in the leaves and roots of different plants, and in germinating kiwi fruit pollen under chromium stress (Scoccianti et al. 2008). Increase in MDA content is a measure of lipid peroxidation. Similar observations were made by Bala and Thukral (2011) that MDA content increases with increase in Cr(VI) concentration in Spirodela polyrrhiza. Cr(VI) increased the MDA content in wheat leaves which reflects the level of lipid peroxidation resulting from oxidative stress induced damage to the membranes. Lipid peroxidation can be induced via free radicals of reactive oxygen species that are generated as a result of heavy metal toxicity in plants (Panda and Choudhury 2005). This increase in the MDA content is due to the oxidising nature of Cr(VI) inside the plant cells and causes the production of stress inducing free radicals (Kotas and Stasicka 2000; Subrahmanyam 2008; Bala and Thukral 2011).

Multiple regression analysis and beta regression coefficients (Table 10) between Cr content in the roots and shoots of seedlings under the combined effects of Cr(VI) and EDTA revealed that both Cr and EDTA treatments increased Cr content in the seedlings. However, negative interaction beta coefficients (β₃) imply that EDTA chelates with Cr(VI) making it unavailable to the seedlings, thus decreasing Cr uptake. Similar observations were made on BCF. The rate of translocation of Cr from root to shoot in binary treatments was not affected. Path analysis also proved (Table 11) that both Cr and EDTA increase the uptake of Cr, but when present together, both indirectly hamper the uptake of Cr. The uptake of heavy metals and chelating agents has been reported by various workers under controlled conditions. Some workers observed that EDTA enhanced the heavy metal uptake in plants (Blaylock et al. 1997; Huang et al. 1997; Grčman et al. 2001), whereas some inferred no increment (Athalye et al. 1995). The application of EDTA also decreases the uptake of heavy metals in the plants (Robinson 1997). High EDTA levels damage the cell membranes which enhances the permeability of the membranes (Grcman et al. 2001). Uptake of significant quantity of chromium by the roots of H. vulgare seedlings with respect to the control may be due to the reason that plant roots excrete low-molecular weight organic acids as root exudates in the rhizosphere for mineral solubility by acidification and formation of organic–mineral complexes and function as natural chelators (Meers et al. 2004; Yu and Gu 2008). Shanker et al. (2005) described the poor translocation of Cr(VI) due to sequestration of the metal in root vacuoles. It was observed in the present study that the accumulation of Cr(VI) in seedlings of H. vulgare roots is more than that of shoots. Gupta et al. (1994) studied uptake and toxicity of chromium in single and combined combinations with Cu in root and shoot of Bacopa monnieri L. and Scirpus lacustris L, and their inferences indicated more uptake of chromium in roots as compared to shoots. Furthermore, Cu uptake was inhibited whereas that of Cr was increased. Chromium ameliorated the Cu toxicity in combined metal treatments. The findings of this study revealed that these plants may be used in bio-monitoring and mitigation of metal contamination. Similar trends were found with Typha latifolia (Baudo et al. 1985). In the present study NMDS stress level for antioxidative enzymes, MDA and pigments was 0.0078, 0.019, 0 respectively. A stress level of less than or equal to 0.05 indicates a good fit of the data.

Conclusions

It is inferred from the multiple regression analysis, beta coefficients and path analysis that application of EDTA to Cr(VI) containing sand is antagonistic to the toxic effects of Cr(VI) on protein content, antioxidative enzymes (APX, SOD, CAT and GR), lipid peroxidation, total chlorophyll, Chl-a, and carotenoids in H. vulgare seedlings. EDTA also interacted with Cr(VI) to decrease Cr uptake in the seedlings. This antagonistic interaction between EDTA and Cr may be attributed to its capability to chelate with heavy metals preventing them from causing damage to the plants.

Compliance with ethical standards

Conflict of interest

There is no conflict of interest.