Effects of arsenite and dimethylarsenic on the growth and health of hydroponically grown commercial Doongara rice

Hydroponic system

The system was a self-contained reservoir, with individual plants growing in individual reservoirs (Fig. 1). The system consisted of three main components: a reservoir, plant housing and aerating system. The reservoir was constructed using 90-mm diameter polyvinyl chloride (PVC) pipes cut to 370-mm lengths. Along the edge, five holes (22-mm diameter) were drilled 55 mm apart. The five large holes were used to insert the plant housing units. A sixth hole was also bored, with a diameter of 10 mm. The smaller hole was used to insert an air hose to aerate the nutrient solution and as an access point to take pH measurements. The capacity of the system was 1.6 L per reservoir.

Fig. 1.

Hydroponics set up. (a) The complete system with growing rice plants. (b) An example of the plant housing tube with a rice plant placed in the styrofoam disc.

The plant housing units were constructed using 20-mm diameter piping cut to 80 mm in length to fit into the reservoirs. Each housing unit held a single plant, and the spacing between the plants allowed the plant to grow without the roots becoming entangled with neighbouring plants. In previous hydroponic systems, it has been found that if the plants were not separated, direct competition occurred, leading to roots becoming entangled, resulting in significant experimental variation within treatments. Plants were held in place using styrofoam discs. Styrofoam was used due to its inert properties, thus avoiding the leaching of unwanted contaminants into the nutrient solution. Each nutrient solution and treatment was aerated by pumping compressed air through Teflon tubing and bubbling it into the nutrient solution by air stones through the sixth hole.

The plants were grown in a temperature-controlled laboratory between 20 and 24°C with two ATI T5 power module light systems using T5 full-spectrum fluorescent bulbs suspended from the roof above the plants. The light field was measured using an irradiance sensor (Biospherical Instruments QSL 2102) across the surface of the hydroponic growing tubes to ensure an even light distribution across the surface.

Seed germination and growth conditions

Rice seeds were surface sterilised utilising a technique adapted from Kim et al. (2005). The seeds were sterilised by rinsing with 10% v/v H₂O₂ for 10 min, followed by 70% w/w ethanol for 5 min and a final rinse with deionised water. After surface sterilisation, the seeds were placed in deionised water and then placed in an incubator at 30°C for 48 h to break dormancy and promote germination of the seedlings.

Seedlings were transplanted into the hydroponic system after germination. For the first 14 days following germination, the plants were exposed to a half-strength nutrient treatment to allow the plants to acclimate, followed by full strength. The nutrient media composition was as follows: 396 μmol L⁻¹ of KNO₃, 360 μmol L⁻¹ of Ca(Cl)₂, 290 μmol L⁻¹ of MgSO₄·H₂O, 38 μmol L⁻¹ of NaH₂PO₄2H₂O, 230 μmol L⁻¹ of K₂SO₄, 2.5 μmol L⁻¹ of MnCl₂·4H₂O, 3.2 μmol L⁻¹ of H₃BO₃, 0.04 μmol L⁻¹ of (NH₄)₆Mo₇O₂₄·4H₂O, 0.035 μmol L⁻¹ of ZnSO₄·7H₂O, 0.04 μmol L⁻¹ of CuSO₄·5H₂O, 0.037 μmol L⁻¹ of FeCl₃·6H₂O and 0.04 μmol L⁻¹ of Na₂SiO₂·5H₂O. The pH of the nutrient solution was 5.5. The nutrient solution was similar to that used by Yoshida et al. (1976), with the addition of silicon at 0.04 μmol L⁻¹ and EDTA at 3.4 μmol L⁻¹. The ETDA was added to prevent the precipitation of iron oxyhydroxide and plaque formation on roots (Jacobson 1951). The nutrient solution and As species were renewed every 7 days. This involved discarding the old nutrient solution and cleaning the tubes to ensure no build-up of mould or algae. The tubes were then topped up with 1.6 L of nutrient solution and appropriate arsenic species. Plants were grown on a 14:10-h day:night cycle at a light intensity of 180 µE m⁻² s⁻¹.

Experimental design

Two hydroponic growth experiments were conducted with Doongara, a long-grain rice variety that was selected due to its susceptibility to As and straighthead disease (Martin et al. 2023). The rice plants were exposed to As^III and DMA at four different concentrations: 0.0, 0.8, 1.6, 3.5 and 6.7 µmol L⁻¹. The arsenic concentrations selected were based on the study by Shaibur et al. (2006), where As_i toxicity in hydroponically cultivated rice plants was found to be severe above 6.7 µmol L⁻¹ of As^III. No experiments have reported direct toxicity for DMA in rice, so for this study, the DMA concentrations were matched to that of As^III to allow comparison between the two As species. For quality control, the highest arsenic treatment (6.7 µmol L⁻¹) of the alternative arsenic species was included in the As^III and DMA experiments.

Treatments were carried out in triplicate for a total of 38 days to allow for substantial growth to occur and to observe As accumulation throughout the vegetative growth stage. The experiment was terminated before the plants reached flowering, thus allowing us to observe the early accumulation of As because studies have shown that the uptake of nutrients and As changes at different growth stages for rice (Zheng et al. 2011).

Harvested plants were lightly rinsed with deionised water, and height was recorded from the base of the stem to the tip of the highest leaf. Plant biomass was determined from plant dry mass. Plant roots were also inspected for iron plaque formation; however, no plaque was observed on any of the root systems (Fig. 2).

Fig. 2.

Photos of rice plants being processed at the end of the experiments. (a) A rice plant from the hydroponic experiments. Note the white root structure. (b) Plants grown in the field (Martin et al. 2023) – the top trays hold the above-ground biomass, and the bottom trays are the roots of the plant that show iron plaque (Chen Z et al. 2005). Panel (a) illustrates that the rice grown hydroponically exhibited no iron plaque formation (typically an orange colour) on the roots.

Arsenic measurement

Arsenic accumulation

Plants exposed to As_i accumulated more As as exposure concentrations increased, and the amount of As accumulated was much higher than in plants exposed to a comparable concentration of DMA (Fig. 3). Because plants were grown for the same period of time, 38 days, accumulated As amounts are directly comparable. When the plant was exposed to either As species, the majority of the As was in the roots. Specifically, 92 ± 5% of As_i and 52 ± 15% of DMA were found in the roots. Plants exposed to As_i or DMA both showed a linear response to increasing As concentration. The response to increasing As exposure, however, showed two distinct accumulation patterns. The roots of the rice plants accumulated As_i to much higher concentrations than the shoots, whereas plants exposed to DMA had a similar distribution of As between shoots and roots with increasing As concentration (Fig. 3). The differences in uptake observed between As species are in agreement with previously published data (Abedin et al. 2002).

Fig. 3.

Arsenic accumulation within different plant segments of rice grown hydroponically with increasing exposure to As_i (a) and DMA (b). Linear regressions were fitted, as shown by dotted lines, with 95% confidence intervals represented by the shaded areas.

Plants exposed to As_i contained up to 13–19 fold higher As concentrations in the roots and 1.6–4 fold higher As concentrations in the shoots than when exposed to DMA. The lower uptake rate of DMA is in accordance with results published for other rice species (Abedin et al. 2002). Visual observations at the end of the experiment show that iron plaque did not form on the roots (Fig. 2). In the absence of iron plaque, the high As concentrations measured in the roots for the As_i treatments suggest active sequestration by root cells or As absorption followed by incorporation into root cells (Vázquez Reina et al. 2005; Moore et al. 2011).

Although root cells took up more As_i, the translocation of As from roots to shoots was significantly lower for As_i than DMA. The ratios of shoot to root As concentrations were 0.06–0.08 and 0.3–0.6 for As_i and DMA respectively. The accumulation of As in the shoots of the plants exposed to As_i remained fairly constant at low As_i concentrations (0.8–3.5 µmol L⁻¹) (Fig. 4). It was only through exposure to the highest As_i treatment (6.7 µmol L⁻¹) that the As concentration significantly increased within shoots (P < 0.05). The As concentrations in the shoots increased from 21 ± 14 to 45 ± 14 µg g⁻¹ for the 3.5 to 6.7 µmol L⁻¹ treatments, whereas for the lower As_i treatments (0.8 and 1.6 µmol L⁻¹), mean As concentrations were 11 ± 3 and 21 ± 14 µg g⁻¹ respectively. This poses the question, is there a threshold for As_i exposure before plants lose the ability to sequester As in roots (Hartley-Whitaker et al. 2001)?

Fig. 4.

Mean arsenic concentrations throughout the rice plant for plants exposed to As_i (a) and DMA (b). Error bars represent 1 s.d.

As previously described, detoxification of As_i involves either the sequestration of As by the formation of As^III–thiol complexes with glutathione (GSH) or phytochelatins (PCs) (Raab et al. 2005) in vacuoles (Song et al. 2010, 2014) or As being pumped out of cells into the external medium via efflux pathways (Xu J et al. 2017). An efficient As_i efflux transporter within rice plants has not yet been identified; however, Lsi1 has been identified as a bi-directional transporter that can efflux a small amount of As^III (Zhao et al. 2010). Sequestration of As_i in the roots is an effective way to limit the transport of As_i to above-ground tissues, although As_i still has some mobility throughout the plant (Zhao et al. 2012). Unlike Lsi1, Lsi2 can control the efflux of As^III towards the stele and restrict xylem loading (Ma et al. 2008). Further transport is limited by additional vascular sequestration (Chen Y et al. 2015). These plant nodes play a critical role in storing As_i and controlling further As_i distribution (Yamaji and Ma 2014; Zhao et al. 2014; Chen Y et al. 2015).

In plants exposed to DMA, the As concentrations in the roots and shoots are fairly consistent across all As exposures (as shown in Fig. 3 and 4). Specifically, the roots contain 10–23 µg g⁻¹, and the shoots contain 3–13 µg g⁻¹ for a DMA treatment concentration range from 0.8 to 3.5 µmol L⁻¹. Approximately half of the arsenic is translocated from the roots to the shoots. Similar to the highest As_i concentration tested, significantly greater amounts of DMA (28 µg g⁻¹) are accumulated in shoots. Our current understanding is that rice plants lack the ability to either efflux or sequester DMA into vacuoles to reduce its mobility within the plant. Mishra et al. (2017) exposed rice plants to As^V, monomethyl arsenic (MA) and DMA for 7 days, and DMA had the lowest shoot and root translocation factors, and no DMA–PC complexes were detected (Raab et al. 2007a). DMA, however, was efficiently translocated between the roots and shoots (Raab et al. 2007b) by both xylem and phloem (Carey et al. 2010), with phloem transport believed to be the main pathway for As transport to the grain (Carey et al. 2010; Zhao et al. 2012; Kumarathilaka et al. 2018).

A peptide transporter (OsPRT7) has been identified as being potentially involved in the translocation of DMA (Tang et al. 2017). Peptide transporters play an essential role in the transport and remobilisation of nitrogen throughout the plant (Tsay et al. 2007) and can affect germination, plant growth and grain yield (Fang et al. 2013, 2017). If DMA is transported by peptide transporters, this can offer a plausible mechanism for the translocation of DMA within rice plants (Tang et al. 2017). During growth and nitrogen utilisation, DMA could continually accumulate to high concentrations throughout the plant, leading to DMA stress and, in turn, straighthead in rice plants.

Plant growth

Plants exposed to As_i showed a significant decrease in plant height (F_(5,47) = 13.04, P = 5.65 × 10⁻⁸) over the As concentration range tested. At As_i concentrations below 1.6 µmol L⁻¹, plant height was not affected (Fig. 5); however, when As_i concentrations exceeded this value, a significant decrease in plant height occurred (1.6 µmol L⁻¹ = 155 ± 84 mm, 3.5 µmol L⁻¹ = 61 ± 23 mm, 6.7 µmol L⁻¹ = 95 ± 31) when compared to the control, (230 ± 73 mm). For the plants exposed to DMA, there was no significant change in plant height across treatments (F_(5,46) = 2.37, P = 0.053); however, when DMA concentrations exceeded 1.6 µmol L⁻¹, the mean plant heights decreased (Fig. 5).

Fig. 5.

Plant height at the completion of the growth experiments. Error bars represent 1 standard deviation, n = 45 (As_i), n = 44 (DMA).

The effect of exposure to As_i and DMA on rice plant biomass showed a similar trend to that observed for plant heights (Fig. 6). Plants exposed to As_i showed a significant decrease (P < 0.05) in plant mass when exposed to increasing As_i concentrations (F_(5,84) = 11.04, P = 3.43 × 10⁻⁸). Like plant heights, exposure to the lower As_i concentrations resulted in no significant decrease in mean plant biomass, whereas the higher As_i exposures showed a significant decrease in plant biomass relative to controls.

Fig. 6.

Plant dry mass at the completion of the growth experiments Error bars represent 1 standard deviation, n = 45 (As_i), n = 44 (DMA).

Plants exposed to DMA also showed a significant decrease in plant mass when exposed to increasing DMA concentrations (F_(5,66) = 3.133, P = 0.0134). Again, when DMA concentrations exceeded 1.6 µmol L⁻¹, the mean plant biomass decreased (Fig. 6).

To determine if exposure to the As species had any effect on plant health, the final plant heights were used as a measure of plant fitness. Both modelled dose–response curves and Z scores were used to investigate the effects on plant health (see Eqn 1):

where X̅ and s respectively represent the mean and standard deviation (s.d.) of the control group for each experiment, and X is the individual value. Z scores were used to normalise plant height to the control in each experiment. The percentage of plants that fell below the Z-scores for each treatment were deemed to be unhealthy (Table 2).

Plant height significantly decreased with increasing As_i dosage with r² = 0.89 when a second-order polynomial was fitted (Fig. 7). The DMA treatments displayed no significant change with increasing DMA concentration, r² = 0.21, when a second-order polynomial was fitted (Fig. 7). A 10% reduction in plant height (EC₁₀) is predicted when plants are exposed to 0.7 mmol L⁻¹ As_i, and a 50% reduction in height (EC₅₀) is predicted for As_i concentrations above 2.5 mmol L⁻¹ As_i (Fig. 7). The rice plants treated with DMA demonstrated no overall consistent reduction in height with increasing DMA exposure (Fig. 7); however, due to the high variability within each DMA concentration treatment, it was not possible to calculate meaningful EC₁₀ and EC₅₀ results.

Fig. 7.

Dose–response curve for rice plants treated with As_i (blue) and DMA (orange). Plant height (%) relative to the control v. arsenic concentration. Error bars represent 1 standard deviation, n = 45 (As_i), n = 44 (DMA).

Plants exposed to an As_i concentration greater than 1.6 µmol L⁻¹ showed an increasing number of unhealthy plants (Fig. 8). Plants exposed to DMA had a consistent number of unhealthy plants across all DMA exposures (Fig. 8) with a greater number of unhealthy plants at the higher exposure concentrations that contain higher As concentrations.

Fig. 8.

Arsenic concentration in the shoots of healthy and unhealthy rice plants exposed to (a) As_i (n = 45) and (b) DMA (n = 44). Divided by healthy (blue) and unhealthy (orange) plants.

Arsenic exposure has been reported to cause a reduction in both plant growth and root elongation (Han et al. 2015; Seneviratne et al. 2019). In this study, a delayed response has been observed with plant heights and mass decreasing after exposure to As concentrations greater than 1.6 µmol L⁻¹, indicating either that the rice plant has some tolerance to As_i (Song et al. 2014; Xu J et al. 2017) or that As_i is interrupting key metabolic pathways where the effects are not immediately observed (Kamiya et al. 2013). The rice plants exposed to the lower concentrations of As_i appear to display a degree of tolerance. As_i is most likely sequestered into the vacuoles as part of a detoxification mechanism (Song et al. 2014). Although As_i can cause oxidative stress to the rice plant, rice appears to handle low-level exposure, between 0.8 and 1.6 µmol L⁻¹. When exposed to between 1.6 and 3.5 µmol L⁻¹, reduction in growth and plant mass are observed; however, these plants still show the ability to regulate and limit As transport to above-ground tissues. A significant reduction in growth and increased As concentrations in the shoots was measured for the 6.7 µmol L⁻¹ As_i treatment, corresponding to a substantial increase of As transport from the roots to shoots (and other plant tissues). Thus, the higher As_i concentrations used in this study could be approaching concentrations that are toxic for hydroponically grown rice. Shaibur et al. (2006) also found As_i toxicity to be induced in hydroponically grown rice at 6.7 µmol L⁻¹.

It is well documented that As^III uses silicon transporters (Ma et al. 2008; Katsuhara et al. 2014; Chen Y et al. 2017) and As^V uses phosphate transporters (Wang P et al. 2016) for uptake and translocation (Kumarathilaka et al. 2018), thus As can lead to the disruption of signalling or metabolic pathways. As^III also has a high affinity to sulfhydryl groups (–SH), and readily reacts with enzymes and proteins (Dixit et al. 2015), inhibiting enzyme activity (Chen W et al. 2010) affecting plant growth and metabolism (Jha and Dubey 2004). As^V can cause a reduction in photosynthetic activity and may delay the effects of arsenic on the health of the plant (Mateos-Naranjo et al. 2012; Abbas et al. 2018). The reduction in height and biomass (Fig. 5 and 6) observed may also be caused by an increase and imbalance in reactive oxygen species (ROS), resulting in oxidative stress throughout the plant (Stoeva et al. 2005; Shri et al. 2009). ROS are signalling compounds and intermediates produced by metabolic pathways in cells that play essential roles throughout the plants’ lifecycle, including plant growth, germination and grain development (Mhamdi and Van Breusegem 2018). Other studies have reported how plants respond to high As_i concentrations, with both As^V and As^III inducing the production of ROS (Finnegan and Chen 2012) and inducing oxidative stress in the plant and eventual cell death (Hartley‐Whitaker et al. 2001; Tripathi et al. 2012), with As^III typically having a more pronounced effect (Pessarakli and Tan 2010). Oxidative stress can cause a range of effects in plants (Finnegan and Chen 2012). Generally, this is a function of excess ROS production, leading to DNA damage, protein modification or lipid peroxidation, which results in impaired cellular function and potential cell death (Pessarakli and Tan 2010).

The effects DMA has on rice health are less clear and have not been thoroughly studied. Elevated DMA concentrations in soil have been linked to straighthead disease in rice (Yan et al. 2005); however, the critical role DMA plays has not been established (Meharg and Zhao 2012). Recently, several studies have found that DMA is more phytotoxic to plants than As_i (Tang et al. 2016a, 2016b), primarily due to its mobility within plants (Raab et al. 2007b; Carey et al. 2011) and the plant’s inability to detoxify DMA (Tang et al. 2016a). For each DMA exposure, many plants displayed a significant reduction in growth (Fig. 8). At the higher DMA treatment concentrations, the plants with reduced growth had higher DMA concentrations in their shoots. Tang et al. (2016a) found that plants exposed to DMA exhibited significantly more oxidative stress (lipid peroxidation), particularly in the shoots, compared to plants exposed to As^V or MA.

The higher translocation of DMA could cause stress to plants by localised accumulation of DMA in different sections of the plant. This could also explain the varying effects of DMA on plant health documented in the literature (Finnegan and Chen 2012). In this study, when plant height was used as a proxy for plant health, we observed that unhealthy plants had higher DMA concentrations compared to healthy plants (Fig. 8). This trend was not observed for As_i-exposed plants. This may indicate that different As species cause different types of stress within plants, or it could demonstrate that some plants can detoxify As_i to a certain extent compared to DMA. Once a specific As concentration is reached, the production of ROS results in oxidative stress and, in some cases, cell death. Unlike As_i, which has limited translocation throughout the plant, the high mobility of DMA could result in many different parts of the plant being susceptible to DMA-induced oxidative stress, which has the potential to disrupt vital metabolic pathways.

The exact mechanism of how DMA induces toxicity in plants is still unclear. Typically, DMA^V is being analysed when DMA is quantified. DMA^III is unstable under aerobic conditions and is oxidised to the more stable DMA^V (Jiang et al. 2003). The toxicity of DMA may be induced by DMA^III; however, changes in DMA^III concentrations cannot be detected using standard methods (Garbinski et al. 2019; Kerl et al. 2019). In animal cells, the trivalent organic arsenic species (DMA^III and MA^III) are more cytotoxic than As^III and As^V (Petrick et al. 2000; Styblo et al. 2000). Oxidative stress can be induced through the redox cycling of DMA^III and DMA^V (Naranmandura et al. 2007). In this hydroponic experiment, due to the nutrient solution being continuously aerated DMA^III was unlikely to be present.

DMA may be the major cause of straighthead disease. Zheng et al. (2013) found that DMA was toxic to reproductive tissues, and Carey et al. (2011) showed that DMA is remobilised and transported to reproductive tissues at the beginning of grain formation. Under these conditions, highly localised DMA accumulation is likely to occur and cause stress to the plant.