Sulfur Induces As Tolerance in Barley Plants

Abstract

The use of sulfur (S) in polluted soils can reduce metal(loid) toxicity and enhance phytoremediation effectiveness. Here we studied the response of barley plants to As in soil amended with sulfate or elemental sulfur throughout the growing cycle. A greenhouse experiment was carried out using 4-L pots filled with clay-loam soil spiked with 60 mg kg⁻¹ As (Na₂HAsO₄·7H₂O). Two chemical forms of sulfur (elemental sulfur (S⁰) or sulfate (CaSO₄·2H₂O)) were applied at a dose of 1 and 3 Mg ha⁻¹, respectively, and two previously seeded barley plants were transplanted in each pot, using eight pots per treatment. At the end of the growing cycle, the biomass, nutrients, and metal(loid) content, as well as several physiological and biochemical parameters of the plants were analyzed. Moreover, the effect of the treatments on soil characteristics was also evaluated, including soil pore water. The treatment with sulfur promoted the growth of barley plants through their vegetative cycle, enhancing photosynthesis, although biomass did not significantly increase. Both sources of S promoted the accumulation of As in the root, thereby limiting its translocation to the aerial part of the plant, sulfate being more effective (an increase of 300%) than elemental S (an increase of 82%). The addition of S decreased soil pH. Furthermore, both treatments, but particularly sulfate, increased soluble sulfate and stimulated soil biological properties. In conclusion, the application of sulfate to As-polluted soil can enhance As phytostabilization by barley plants while simultaneously improving the biological properties of the soil.

1. Introduction

Soil contamination is a global issue. Indeed, worldwide, more than 10 million sites are considered to be affected by soil pollution, and in the European Union alone, there are an estimated 2.5 million potentially contaminated sites [1]. Among the many contaminants responsible for this pollution, metal(loid)s are one of the most concerning groups due to their abundance and toxicity. Unlike organic contaminants, metal(loid)s cannot be degraded either biologically or chemically and they generally persist in the soil environment for extended periods after their introduction, thereby posing a risk to human health and the ecosystem [2,3]. Non-essential metal(loid)s like As, Cd, Hg, and Pb, are those most related to phytotoxicity [4]. Arsenic is a metalloid that can covalently bond to most metals and non-metals. In the soil, As is adsorbed to clay, hydroxides, and organic matter (Kabata-Pendias, 2011). Soil concentrations of this pollutant are increased by industrial activities and As-containing pesticides [4]. This metalloid poses a significant risk to public health due to its involvement in major systemic diseases in humans. This is a critical issue in highly contaminated areas of South and Southeast Asia, where rice (Oryza sativa L.) grown in affected soils has been found to contain elevated levels of As, thus raising pressing concerns [5,6,7,8,9,10].

The presence of metal(loid)s in plants causes oxidative stress through the generation of reactive oxygen species (ROS). Some metal(loid)s trigger the direct synthesis of ROS, while others promote their accumulation by interfering with antioxidative systems, the electron transport chain, or certain metabolic pathways [11]. ROS induces oxidative damage to lipids, proteins, and DNA. Plants have several oxidative stress defense mechanisms, including low molecular weight thiols like cysteine and glutathione, and phytochelatins [12,13].

Phytoremediation emerged at the end of the last century as a promising approach to recover contaminated soils since it is a cost-effective, efficient, and environmentally friendly strategy that also prevents soil erosion [14]. Among the phytoremediation strategies available, phytoextraction leverages the ability of plants to take up contaminants from the environment and translocate them to their organs [2,15]. To guarantee the performance of this technique, it is important to use species that tolerate the contaminant and have high biomass production and rapid growth. In this context, the focus has been placed on crops, as they have been bred for fast growth and maximum biomass production. Cereal crops meet these criteria, demonstrating high adaptability to various soil types and climatic conditions. In addition, they are tolerant to metal(loid)s including As, Cd, and Zn [16,17,18,19,20,21,22].

Several strategies have been tested to improve phytoremediation, such as the addition of chelating agents and exogenous phytohormones to soil [23,24]. However, these methods are often expensive and can cause secondary contamination [24,25,26]. A more environmentally friendly alternative is the addition of nutrients such as S, Se, Si, B, or P, which reduce metal(loid) toxicity and can sustainably enhance the effectiveness of phytoremediation while acting as a fertilizer [25,27,28,29,30,31,32]. However, many of those experiments were performed under hydroponic conditions, which do not always reflect those present in soil [33,34,35,36]. In this regard, S is an essential nutrient that plays a major role in plant metabolism, protein synthesis, and plant defense against abiotic stresses, particularly metal(loid)s [25]. Soil amendments of S, including elemental S, CaSO₄ (gypsum), Na₂SO₄, (NH₄)₂SO₄, Na₂SO₃, Na₂S₂O₃, and (NH₄)₂S₂O₃, can act as fertilizers or pesticides [25,37]. Sulfur-containing defense compounds (SDCs) are low molecular weight metabolites of key importance in biotic and abiotic stress responses. H₂S, glutathione, glucosinolates, phytochelatins, phytoalexins, and S-rich proteins (SRPs) are examples of SDCs. Among SDCs, glutathione and phytochelatins play a major role in metal(loid) detoxification [38].

In this context, the treatment with S has been shown to enhance As tolerance and accumulation in Hydrilla verticillata plants under hydroponic conditions [34]. Arsenic accumulation in the roots of rice plants under hydroponic conditions is promoted under treatment with high concentrations of S, with a consequent decline in the root-to-shoot translocation factor of As [36]. In soil conditions, wheat (Triticum aestivum L.) grown in polluted soil amended with S showed enhanced translocation of Pb from roots to shoots, together with an increase in grain yield [39]. In another study, treatment of an As-polluted soil with ammonium sulfate led to a decrease in toxicity for barley plants [40]. A recent study revealed that sulfur fertilization enhanced the efficiency of Cd phytoremediation in Sedum alfredii Hance plants [31]. Furthermore, sulfate application on As and Cd accumulation in wheat plants grown in a polluted soil has been reported to lead to an increase in biomass, a decrease in oxidative stress, a higher concentration of both metal(loid)s in the root, but a decrease in grain [41]. Other authors compared the effects of two forms of S fertilizer (elemental S and gypsum) on As accumulation in rice [28] and reported differences in As mobility in the paddy soil (higher for the treatment with elemental S), although both treatments favored the formation of Fe/Mn plaques and inhibited the translocation of As from roots to grains. Elemental S is gradually oxidized to sulfate by S-oxidizing bacteria according to Equation (1), releasing H⁺, which reduces soil pH [42,43].

S⁰ + 3/2O₂ + H₂O → SO₄²⁻ + 2H⁺

(1)

Thus, the application of elemental S slowly releases sulfate to the soil, minimizing losses of this nutrient through leaching; therefore, it can be considered as a slow-release method for sulfate application. However, the results obtained for rice cannot be extrapolated for other cereals like barley or wheat, due to the very different growing conditions. In this regard, little data are available on the tolerance of barley plants to the application of different sources of S on As-polluted soils as well as their impact on soil properties. Thus, here we performed a greenhouse experiment in which we studied the response of barley (Hordeum vulgare L.) throughout its growth cycle to As in soil after amendment with sulfate (SO₄²⁻) or elemental S to evaluate their potential use in phytoremediation strategies. In addition, we evaluated the impact of these amendments on the chemical and biological properties of the soil.

2. Materials and Methods

2.1. Experimental Design

The experiment was conducted in a greenhouse. A barley (Hordeum vulgare L.) cultivar “Pedrezuela” was grown in clay-loam soil from central Spain (Alcalá de Henares, Madrid, Spain). The soil characteristics were analyzed following MAPA [44]. In brief, pH and electrical conductivity were determined in a 1:2.5 soil-to-water ratio; organic matter (OM) was measured using the Walkley–Black method; available phosphorus was measured by the Olsen method using sodium bicarbonate at pH 8.5 as extractant; total nitrogen content was measured by the Kjeldahl method; percentage of carbonates was measured using a Bernard calcimeter; and exchangeable Ca, Mg, Na, and K were measured by extraction with 0.1 N ammonium acetate and quantification using flame atomic absorption spectrometry (FAAS) (AA240FS, Varian, Victoria, Australia). The soil texture was analyzed using the Robinson pipette method. The total metal(loid) concentration was determined by FAAS after the acid digestion of samples (0.5000 g) with a mixture of HNO₃ (6 mL, 69%) and HCl (2 mL, 37%) in a microwave (Multiwave GO, Anton Paar GmbH, Graz, Austria). Heavy metals were quantified by inductively coupled plasma optical emission spectroscopy (ICP-OES) (Liberty AX Sequential, Varian, Victoria, Australia), whereas As was determined with a graphite furnace atomic absorption spectrometer (GFAAS) (AA240Z, Varian, Victoria, Australia). The certified reference soil SQC001 (Sigma-Aldrich, Laramie, WY, USA) was used to verify accuracy, with recoveries consistently greater than 90%. Water-soluble anions in soil were determined after shaking with Milli-Q water (1:10, weight:vol) for 1 h, followed by filtration and measurement in an ionic chromatograph (ICS-1100, Dionex, Sunnyvale, CA, USA) [45]. The mean values of the physico-chemical soil parameters are shown in

Table 1. Soil properties.

Parameter		Mean Value
pH		8.22
electrical conductivity (dS m−1)		0.33
CO32− (%)		5.75
N (%)		0.20
Organic matter (%)		1.40
Available nutrients (mg kg−1)	P	19
	Ca2+	3076
	Mg2+	469
	Na+	56
	K+	347
Total metal(loid)s (mg kg−1)	As	8.5
	Cd	<LD
	Cr	31
	Cu	14
	Ni	10
	Pb	8.5
	Zn	49
Anions (mg kg−1)	Cl−	82
	NO2−	0.80
	NO3−	132
	PO43−	5.20
	SO42−	33

.

In the greenhouse experiment, eight pots (4 L) were used per treatment. The soil was spiked with As using Na₂HAsO₄·7H₂O solutions to obtain concentrations of 60 mg As kg⁻¹ in each pot (420 mL of the arsenate salt at 2.26 g L⁻¹). This concentration of As is above the regional legislative threshold. Once spiked, samples were incubated for forty days at greenhouse conditions, irrigated every three days to reach a soil’s water-holding capacity of approximately 60%. Then, sulfate amendment (CaSO₄·2H₂O, Panreac, Castellar del Vallès, Spain) was added to reach 3 Mg ha⁻¹ and elemental S (Agrodan, Madrid, Spain) at 1 Mg ha⁻¹. In summary, the treatments were as follows: As without amendment (As); As + elemental S (As-S); and As + sulfate (As-SO₄), using eight pots per treatment. In addition, three positive controls were included: unpolluted soil (C), unpolluted soil + elemental sulfur (S), and unpolluted soil + sulfate (SO₄). The doses of S amendment were chosen according to previous experiments [28,40,41,43,46]. A week later, 3 g of fertilizer (N:P:K, 16:16:16) was added per pot. Fifteen days after the amendments had been applied, two plants (at stage 13 on the Zadoks growth scale) per pot were transplanted [47]. Pots were irrigated with tap water during the growing period up to 60% of the soil’s water-holding capacity. The characteristics of tap water are shown in Table S1 (Supplementary Materials). When the crop had reached the end of its cycle (130 days after transplanting), plants were cut, dividing the ears, stems, and roots to obtain the dry weight of each of the parts (grain, aerial part, and roots). The roots were washed with tap water and then with distilled water. The plant material was dried in an oven at 80 °C for 48 h until a constant weight was obtained.

Pore water samples were collected using rhizon samplers (Rhizosphere Research Products, Wageningen, The Netherlands) installed in each pot at the beginning of flowering (state 58–60 of Zadoks growth scale). Electrical conductivity, pH, and As concentration were determined in the pore water samples.

2.2. Plant Analysis

The crops were monitored throughout the growth cycle. Samples and measurements were taken at the same phenological stage for all plants. The following ranges were used according to Zadoks growth scale [47]: M1, stages 21–25 (tillering); M2, stages 40–42 (flag leaf); M3, stages 60–63 (anthesis); and M4, stages 70–75 (grain filling). Finally, measurements were taken at the end of the crop cycle (stage 92).

2.2.1. Physiological Parameters

Chlorophyll content

A portable chlorophyll meter SPAD 502 (Minolta, Osaka, Japan) was used to determine the chlorophyll content using a specific calibration to relate SPAD units to chlorophyll content [48]. Measurements were taken on the flag leaf of the main shoot.

Chlorophyll fluorescence

The chlorophyll fluorescence was determined using a fluorimeter FMS2 from Hansatech Instruments Ltd. (Pentney, UK). Maximum photosynthetic efficiency of PSII (Fv/Fm) was obtained using the fluorimeter in the central part of the flag leaf of the main stem after the zone had been in the dark for 30 min [49]. Measurements were carried out in the same stage and the same leaf as SPAD measurements.

2.2.2. Analysis of Nutrients and As Content

Dried grains, leaves, stems, and roots were powdered and mineralized following the official method [44]. Briefly, powdered samples were ashed at 500 °C overnight; the ashes were then heated with 1 mL of HCl 37% and 5 mL of Milli-Q water for 30 min in boiling, and then filtrated and diluted with Milli-Q water to 50 mL. Nitrogen was determined following the Kjeldahl digestion method (FOSS Tecator-Kjeltec 8400, Hillerød, Denmark). The concentration of Ca, Na, Mg, and K was determined using FAAS (AA 240 FS, Varian). The arsenic concentration was measured by GFAAS (AA240Z, Varian). The translocation factor (TF) was calculated using the expression: TF = C_shoots/C_root, where C_shoots is the concentration of As in the shoots and C_roots is the concentration of As in the roots. The sulfur content was determined after digestion (0.5000 g) in a heating block with HNO₃ and HClO₄ followed by quantification by ICP-MS (Agilent 7500CE, Waldbronn, Germany) [50]. The accuracy of the proposed method was evaluated by the analysis of a reference material (Virginia tobacco leaves, CTA-VTL-2, Warszawa, Poland), which presented recoveries for all elements ranging 95–103%.

2.2.3. Oxidative Stress Parameters

Malondialdehyde (MDA) concentration was used as an oxidative stress indicator. It was determined following the method described by [51]. To this end, 0.1 g of fresh leaf was homogenized in 5 mL of a 10% trichloroacetic acid (TCA) solution. The extract was centrifuged at 12,000× g for 10 min. Next, 2 mL of the supernatant was transferred to a vial, and 4 mL of a 0.6% thiobarbituric acid solution (dissolved in 10% TCA) was then added to each vial. Samples were incubated at 100 °C in a water bath for 15 min. The reaction was stopped using an ice bath. Tubes were centrifuged at 12,000× g for 10 min. Finally, supernatant absorbance was measured at 450, 532, and 600 nm (Thermo Spectronic HEλIOS α, Cambridge, UK). MDA concentration was calculated using the following formula:

[MDA] (μmol L⁻¹) = 6.45 (DO₅₃₂ − DO₆₀₀) − 0.56DO₄₅₀

2.3. Soil Analysis

2.3.1. Chemical Analysis

The properties of the soils collected after harvest were analyzed following MAPA, (1994) as previously explained in Section 2.1.

Available As in soil was determined using the TCLP method (Toxicity Characteristic Leaching Procedure) [52] at two time points: (i) at tillering; and (ii) after harvest.

2.3.2. Biological Properties

The impact of the treatments on soil biological properties was evaluated for the following: glucosidase (EC 3.2.1.21) and galactosidase (EC 3.2.1.23) activity for the C cycle; urease activity (EC 3.5.1.5) for the N cycle; alkaline and acid phosphatase activity (EC 3.1.3.1 and EC 3.1.3.2) for the P cycle; and arylsulfatase activity (EC 3.1.6.1) for the S cycle, according to ISO 20130:2018 methodology [53].

Soil respiration was determined by the glucose-induced soil respiration method (ISO-17155-2012) using the BacTrac µ-Trac 4200 system (SY-LAB, Neupurkersdorf, Austria) [54]. Briefly, 15 g of soil samples at 60% of water-holding capacity was incubated at 22 °C for 72 h and then mixed with glucose-talc and introduced in the BacTrac system [55].

2.4. Statistical Analyses

Data were analyzed using the IBM SPSS statistical package for Windows, version 23.0.0.0 (Chicago, IL, USA). Differences among treatments were determined by one-way analysis of variance at a significance level of p < 0.05, followed by a Duncan post-hoc test.

3. Results and Discussion

3.1. Impact on Soil Chemical Properties

Table 2. Mean values of soil properties in samples collected after plant harvest. Mean values followed by different letters indicate significant differences (p < 0.05).

	C	S	SO4	As	As-S	As-SO4
pH	8.43 a	8.06 c	8.23 b	8.34 ab	8.00 c	8.06 c
EC (dS m−1)	0.27 c	1.10 a	0.65 b	0.36 c	1.06 a	1.03 a
N (%)	0.09 a	0.10 a	0.09 a	0.09 a	0.10 a	0.10 a
OM (%)	1.54 a	1.49 a	1.42 a	1.56 a	1.59 a	1.54 a
Ca (mg kg−1)	3427 ab	3790 a	3306 b	3152 b	3772 a	3718 a
Mg (mg kg−1)	390 a	378 a	363 a	372 a	364 a	376 a
Na (mg kg−1)	76 a	70 a	65 a	68 a	69 a	101 a
K (mg kg−1)	195 b	172 b	178 b	282 a	279 a	318 a
Cl− (mg kg−1)	22.2 b	29.6 b	41.4 b	23.9 b	24.6 b	71.6 ab
NO2− (mg kg−1)	<LQ	<LQ	<LQ	<LQ	<LQ	<LQ
NO3− (mg kg−1)	279 bc	381 b	415 b	313 b	139 c	741 a
PO43− (mg kg−1)	0.82	5.44	<LQ	2.43	5.53	<LQ
SO42− (mg kg−1)	36.1 c	718 b	1453 a	38.8 c	246 b	1192 a

shows the main physico-chemical properties of the soil samples collected after plant harvest. Table S2 shows mean values, standard deviation and statistical significance of the analyzed soil properties. The pH decreased when sulfate or elemental S was applied. However, the reduction was moderate, from 8.34 in untreated As-polluted soil to 8.06 and 8.00 for the As-SO4 and As-S treatments, respectively. As previously commented, elemental S was gradually oxidized to H₂SO₄, which lowered the soil pH. However, this decrease was minimal due to the soil’s high alkalinity and buffering capacity. Other studies also reported a decrease in soil pH after the addition of sulfate or elemental S [31,41,43]. Electrical conductivity significantly increased with the S treatments due to the release of soluble salts, although the values obtained were within the normal range for soils. Of note, the availability of K in As-polluted soils increased, regardless of the S treatment applied. This observation could be explained by the spiking process, which involved a sodium arsenate salt; Na can release K from exchangeable sites. High variability was observed for the available Na content in soil, and no clear trend was identified. Soil amendment with sulfate may promote the formation of Na₂SO₄, which is highly soluble and can be leached.

Both forms of S significantly increased soluble sulfate in soil, especially sulfate treatment, as expected. This can be explained by the progressive oxidization of elemental S to sulfate by microorganisms. Therefore, elemental S can be considered a slow source of sulfate to soil, as previously explained. In this regard, ref. [46] observed that microbial oxidation of elemental S is enhanced close to root surfaces. In addition, the availability of elemental S in soil has been reported to be conditioned by its hydrophobicity [56]. It would be of interest to determine the long-term effects of elemental S application.

Figure 1 shows As availability in the soil, as determined by the TCLP test at two sampling times (one month after transplanting and at the end of the experiment). Arsenic availability decreased over time in all the treatments. At the first sampling time, S addition did not affect the availability of this metalloid and no significant differences were observed between treatments. In the samples collected after harvest, As availability was significantly higher in the As-SO₄ treatment than in the control and As-S treatments, which showed a similar content. These observations can be explained by anionic competitive adsorption with arsenate for binding sites due to the presence of a significant amount of sulfate. No effect of sodium sulfate addition on As mobility was observed using CaCl₂ as an extractant [41]. In contrast, sulfate was found to alter As distribution in flooded contaminated mangrove sediment [57].

3.2. Impact on Rhizosphere

3.2.1. Pore Water

Figure 2 shows the As concentration, pH, and electrical conductivity of pore water samples collected at the flowering stage. The addition of elemental S or sulfate did not significantly affect As concentration in pore water and the concentration was similar to that in untreated soil at this sampling time. This result is consistent with the As concentration in the TCLP extract performed at tillering. Indeed, the content of As in pore water was variable. The mean pH values were between 8.27 and 7.69, the lowest values corresponding to the As-S and As-SO₄ treatments—findings that are consistent with the results observed in soil after harvest (Table 2). The rhizosphere can be slightly more acidic than bulk soil due to the excretion of organic acids by plant roots. As previously explained, elemental S in soil is gradually oxidized to sulfuric acid, as described by Equation (1), which could reduce soil pH. The soil used in this experiment showed high alkalinity, with a pH of 8.22 and carbonate content exceeding 5% (Table 1), suggesting a considerable acid buffering capacity. As expected, the electrical conductivity of the pore water increased significantly upon S amendment, both as elemental S and sulfate, since both forms increased soluble salts in soil. In this regard, at the end of the experiment, soil samples from the S treatments also showed higher electrical conductivity.

3.2.2. Soil Biological Properties

Figure 3 shows the mean values of soil respiration and enzyme activities in rhizosphere samples collected after harvest. As expected, contamination with As negatively affected the biological properties of the soil. However, the addition of elemental S or sulfate stimulated microbial populations, although with varying degrees of effectiveness. In general, sulfate treatment showed a more pronounced effect on enzyme activity than elemental S. In this regard, soil respiration significantly increased in treated As-polluted soil, with a 26% and 41% increase for the elemental S and sulfate treatment, respectively. Arylsulfatase is an enzyme that serves as a representative of the S cycle. The impact of the addition of S depended on the chemical form used. In this regard, elemental S did not affect this enzyme, whereas sulfate stimulated its production (almost 12%). The enzymes for the C cycle showed a different behavior. On the one hand, β-galactosidase was not stimulated by the S treatments, and a similar trend was observed in treated and untreated polluted soils. On the other hand, β-glucosidase was significantly enhanced by S treatment, increasing 12% and 24% in the As-S and As-SO₄ treatment, respectively, compared with untreated polluted soil. Therefore, our results reveal that β-galactosidase is more sensitive to As contamination than β-glucosidase and the presence of S does not enhance its toxic effects. In this regard, previous studies concluded that β-galactosidase can be used as an indicator of metal(loid) pollution [58]. Similarly, the enzymes related to the P cycle also showed a different behavior. In this regard, the addition of sulfate stimulated the production of alkaline phosphatase, showing a similar trend to that observed for β-glucosidase and respiration, whereas acidic phosphatase was not affected by the treatments. The activity of urease, a representative enzyme of the N cycle, also increased in response to S treatment, especially when applied as sulfate. The enhanced biological properties of the soil may be because S is a nutrient for soil microorganisms, and it can therefore stimulate microbial population while mitigating contamination. In addition, the reduction in soil alkalinity can promote microbial activity. Given all the above observations, under the experimental conditions tested, our results indicate that S addition, especially in the form of sulfate, enhanced the biological properties of the As-polluted soil, which is particularly interesting from the perspective of restoring soil functionality. To the best of our knowledge, this is the first study to evaluate the impact of S addition on the biological properties of As-polluted soil. In this context, it has been reported that the response of enzymatic activities to different doses of S treatment in Cd-polluted soil depends on both the specific enzyme and the dose of S applied [59]. In the same way, Sun et al. observed that the elemental sulfur rate could be one of the main factors affecting bacterial community structure in a Cd-polluted soil [31].

3.3. Impact on Barley Plants

Figure 4 shows the mean values of height, SPAD, chlorophyll fluorescence, MDA, and biomass of barley plants under the different treatments. As expected, positive controls (unpolluted soil, C, S, and SO₄) showed better development than plants grown in untreated As-polluted soils. Barley plants grown in polluted soils showed a similar height (no significant differences at p < 0.05) at tillering, flag leaf, and anthesis, whereas at the last sampling time (grain filling), plants grown in As-polluted soil treated with elemental S and sulfate showed higher height than the controls. Similarly, the chlorophyll content measured as SPAD and Fv/Fm were more affected at the beginning of the experiment (M1, tillering), and the addition of S led to a reduction in toxicity and consequently faster recovery of photosynthetic activity, regardless of the chemical form of S applied. In relation to oxidative stress, the highest MDA content was detected at the tillering stage, especially for the As and As-S treatments, and the addition of sulfate (As-SO₄) reduced this parameter to similar levels to those found in unpolluted soil (C and S treatments). Then, MDA decreased, and at the end of the experiment, all the treatments showed a similar MDA content, approximately 0.050 µmol g⁻¹ FW. This reduction in toxicity symptoms and oxidative damage can be attributed to the addition of S, which activates various ROS-scavenging mechanisms, including the synthesis of glutathione and phytochelatins. Glutathione is a non-enzymatic antioxidant that can scavenge excessive ROS, balance redox homeostasis, and reduce membrane damage through the ascorbate cycle [25,36,38]. Phytochelatins can sequester As through complexation and store it in vacuoles [25,60]. Sulfur is taken up as sulfate by plants and used for the synthesis of S-metabolites, including cysteine (an amino acid of primary metabolism), and glutathione and phytochelatins, two common cysteine-rich peptides [25,38,61]. Furthermore, plants can adapt to the presence of pollutants within a specific concentration range over time. In the present experiment, the treatment with elemental S or sulfate induced faster adaptation. Gonzalez et al. [62] also found the highest values of MDA in the first 15 days of growth of barley and wheat plants in As-polluted soil. In addition, the decrease in As availability over time contributed to the reduction in toxicity (Figure 1). The capacity of plants to adapt to moderately contaminated soils is a key factor for phytoremediation but may introduce pollutants into the food chain. However, the reduction in toxicity was not enough to significantly increase biomass, and similar values were obtained in the As, As-S, and As-SO₄ treatments both for roots and aerial parts. In this regard, Thouin et al. [40] added different doses of ammonium sulfate to an As-polluted soil and detected a decrease in the toxicity in barley plants, although no differences in biomass were observed. In contrast, Shi et al. [41] reported an increase in biomass and a reduction in MDA content in wheat plants grown in As- and Cd-polluted soil treated with Na₂SO₄.

3.3.1. As Accumulation in Barley Plants

The concentration of As in the different parts of the plants is shown in Figure 5A. The roots showed the highest concentrations, followed by the shoots and grains. Of note, the addition of S favored As accumulation in the root, especially when applied in the form of sulfate (almost 300% and 82% higher for sulfate and elemental S than the control, respectively). However, this As accumulation in the root did not induce As translocation to the aerial part. Moreover, the translocation index decreased significantly in the S treatments (Figure 5B). The presence of a higher content of S promoted the immobilization of As in the root as a defense mechanism (Figure 6). This observation may be attributed to the ability of root cells to compartmentalize As into vacuoles or to restrict its translocation to aerial parts by complexing it with glutathione and phytochelatins [36,40,63]. Phytochelatins are synthesized from S-containing amino acids and consequently their accumulation, as well as plant detoxification mechanisms, is significantly influenced by S availability [40], which was higher in the As-SO₄ treatment (Table 2). Thus, the application of sulfate in As-polluted soil can favor the phytostabilization of As with barley plants. Under hydroponic conditions, Dixit et al. [36] concluded that S supply immobilized As in rice roots, limiting its translocation to the shoot possibly by As complexation through enhanced synthesis of thiolic ligands, such as non-protein thiols and phytochelatins. Shi et al. [41] also reported an increase in the As content of roots in wheat grown in an As-polluted soil treated with Na₂SO₄. Zhang et al. [28] observed a reduction in As accumulation in rice plants (root, stem, leaf, and grain) after the addition of CaSO₄ to an As-polluted soil. However, rice is grown under very different conditions (generally managed under flooded conditions) from those used for barley and wheat (dry crops). In this regard, Zhang et al. [28] observed an increase in As in the stem and leaf of rice plants at the tillering stage when treated with elemental S. However, this effect decreased over time. Elemental S slowly oxides to sulfate and the oxidation rate is influenced by the microbial communities responsible for sulfur oxidation and the bioavailability of elemental S, which is conditioned by its hydrophobicity [56]. In this regard, it would be of interest to evaluate the application of elemental S earlier in the crop cycle to allow the gradual oxidation of S to the sulfate form or various applications of sulfate throughout the cycle. Our results, together with the available literature on this topic, reveal that As accumulation under S amendment is influenced by the plant species and the chemical form of S applied. Longer-term studies, including other species and different types of soil, are required to define application strategies.

3.3.2. Nutrient Accumulation in Barley Plants

Figure 6 and

Table 3. Nutrient concentration in root, stem, and grain of barley plants grown under the different treatments. Mean values followed by different letters indicate significant differences (p < 0.05).

Part of the Plant	Nutrient	C	S	SO4	As	As-S	As-SO4
root	K (g kg−1)	4.84 b	4.86 b	4.39 b	8.16 a	6.10 ab	7.93 a
	Na (g kg−1)	0.85 a	1.09 a	0.58 a	1.05 a	0.66 a	1.30 a
	Ca (g kg−1)	20.3 a	12.7 b	15.8 ab	5.29 c	3.88 c	3.25 c
	Mg (g kg−1)	1.65 ab	1.52 b	2.42 a	1.33 b	0.81 b	1.39 b
stem	K (g kg−1)	24.9 c	23.3 c	24.6 c	36.1 b	32.8 b	43.3 a
	Na (g kg−1)	2.64 ab	2.63 ab	1.44 b	3.35 a	3.59 a	3.36 a
	Ca (g kg−1)	5.77 b	7.18 b	5.94 b	9.09 ab	8.47 ab	11.1 a
	Mg (g kg−1)	2.51 a	2.63 a	2.73 a	2.47 a	3.10 a	2.65 a
	N (%)	0.77 a	0.84 a	0.84 a	1.41 a	1.16 a	1.49 a
grain	K (g kg−1)	4.73 ab	3.60 b	4.14 ab	4.52 b	3.83 b	5.07 a
	Na (g kg−1)	0.044 c	0.046 c	0.045 c	0.230 a	0.069 bc	0.108 b
	Ca (g kg−1)	0.057 ab	0.039 c	0.049 bc	0.039 c	0.039 c	0.047 bc
	Mg (g kg−1)	0.84 a	0.82 a	0.69 a	0.84 a	0.92 a	0.85 a
	N (%)	2.25 a	2.47 a	2.18 a	2.67 a	2.63 a	2.58 a

show the nutrient contents of the different parts of the plants. The accumulation of S depended on the form of S applied (Figure 6). In this regard, S treatments favored S accumulation in barley plants in agreement with sulfate availability in soil. In the root, the highest S content was found in plants grown in polluted soil amended with elemental S followed by sulfate treatment. This observation may reflect the plant defense mechanism of taking up S to synthesize cysteine, and from that, gluatathione and phytochelatins. The shoots of plants in the S treatments showed a higher concentration of S regardless of the presence of As, whereas no significant differences in the S concentration in grain were observed between treatments. Shi et al. [41] also observed an increase in S uptake by wheat plants grown in As-polluted soil treated with sulfate. In contrast, a recent study with Arabis alpine plants grown in a Pb-polluted soil found a higher S accumulation in root and shoot (especially in cell walls and soluble components) for Na₂SO₄ treatment compared to elemental S [32]. These data highlight the importance of the species.

We also evaluated the content of nutrients other than S (Table 3 and Table S3). To the best of our knowledge, little data are available about the impact of sulfur addition on the accumulation of nutrients in barley plants grown in As-polluted conditions. Calcium accumulated mainly in roots, followed by stems and grains; K was more concentrated in stems than in roots and grains, which showed similar levels; and Mg and Na showed a concentration gradient: stem > root > grain. The presence of As affected nutrient accumulation; however, the reasons for this effect are not clearly understood and further information is needed to elucidate these changes. In this regard, the presence of As did not promote the accumulation of Ca in roots but appeared to stimulate its translocation to stems. However, the treatment with S did not affect Ca accumulation. Potassium was higher in plants grown in untreated polluted soil. Amendment with sulfate induced a significant increase in K in shoots and grains. This increase in K is in agreement with the amount of available K in soil, which was higher in As-polluted soil (Table 2). The accumulation of Na in grain was enhanced in soil polluted with As, especially in untreated soil. The content of N in grain was similar regardless of the treatment, being higher in the grain than the stem. In summary, soil amendment with S led to an increase in its concentration in roots and stems and minor changes in grains. Nutrient accumulation in barley plants was conditioned by the presence of As contamination in soil and the applications of S treatment.

4. Conclusions

The application of elemental S or sulfate to an As-polluted soil under the experimental conditions tested promoted the growth of barley plants throughout their cycle, enhancing photosynthesis; although, it was not sufficient to significantly increase biomass. The addition of sulfate was more effective than elemental S at reducing oxidative damage. Both sources of S increased the accumulation of As in roots (300% for the sulfate treatment and 82% for the elemental S treatment) but its translocation to aerial parts was not enhanced. Sulfur addition also increased the content of this nutrient in roots and stems, but not in grains.

Regarding the impact of S addition on the physico-chemical properties of the soil, it is worth highlighting the decrease in soil pH and the increase in electrical conductivity and soluble sulfate. The availability of As in soil decreased with time, being higher in the As-SO₄ treatment than in the untreated and As-S ones at the end of the experiment. In relation to soil biological properties, the addition of S, particularly in its sulfate form, stimulated soil respiration and enzymatic activities (arylsulfatase, β-glucosidase, alkaline phosphatase, and urease). Thus, the amendment of As-polluted soil with sulfate can promote phytostabilization by barley plants while improving the biological properties of soil. Further studies are needed to explore other doses of sulfate or elemental S, varying application times (including successive applications throughout the plant growth cycle), or combined applications with other nutrients such as Si, Se, and/or B to improve the phytomanagement of As-polluted soil and elucidate the mechanisms involved.