Arsenic Accumulation in Rice Grains: Effects of Cultivars and Water Management Practices

Abstract

Arsenic (As) accumulation in rice grains is a threat to human health and marketability of rice products. In an effort to minimize As uptake by rice grains, field experiments were conducted to investigate As accumulation in rice grains of three cultivars in monosodium methanearsonate-treated soil under saturated and flooded water management practices. Results indicated that As concentrations in rice grains were cultivar-dependent and influenced by water management. Soil flooding would substantially enhance the As accumulation with a great variation among cultivars. Extractable As in the soil was positively correlated with sodium dithionite-sodium citrate-sodium bicarbonate solution-extractable Fe, suggesting a strong association of As with ferric (hydr)oxide. Additional laboratory studies showed a strong affinity of synthetic ferric (hydr)oxide with monosodium methanearsonate. This study demonstrated that selection of less As-responsive rice cultivars and use of saturated water management in paddy fields could be an effective means to minimize As accumulation in rice grains.

Introduction

Arsenic (As) is listed as one of the top health hazards by the International Agency for Research on Cancer, which is linked to bladder, lung, skin, and prostate cancers (National Research Council, 2001). Human exposure to As is primarily through the intake of drinking water and foods, such as rice grains with elevated As contents (Marin et al., 1993).

Health concern of As in rice grains is due to a large quantity of average daily consumption of rice per person (0.4 kg day⁻¹), especially in Asian countries (FAO, 2002). Accumulation of As in rice plant tissues and grains was reported resulting from the soils or irrigation waters containing an elevated level of As and varying among cultivars (Duxbury et al., 2003; Meharg, 2004; Zavala and Duxbury, 2008; Zavala et al., 2008; Hossain et al., 2009). In Bangladesh, irrigation with As-contaminated groundwater had caused As accumulation in soils as high as 83 mg/kg soil (Ullah, 1998), and the As concentrations in rice grains, husks, stalks, and roots were positively correlated with arsenate contents in the irrigation water (Abedin et al., 2002). Analyses of 204 commercial rice samples showed that the total As concentration in rice ranged from 5 to 710 μm/kg (Zavala and Duxbury, 2008). In addition to As accumulation in the tissues and grains, high As levels in soil and irrigation water would also hinder root growth, decrease plant height, and reduce grain yields (Abedin et al., 2002). In Arkansas, elevated As in the soils resulting from a long historical application of As-containing pesticides, such as monosodium methanearsonate (MSMA), could induce the straighthead or straighhead-like symptom, a physiological disorder of rice with which rice heads remain straight at maturity because of lack of grain development (Gilmour and Wells, 1980; Yan et al., 2005, 2008). As a result, As accumulation in rice plants would have profound, adverse impacts on the quality, security, marketability, and profitability of rice products.

The interactions of As species with naturally occurring minerals in soil have strong influences on the As bioavailability and mobility in the environment. Ferric (hydr)oxide-based sorbents have been used to remove aqueous As for drinking water treatment because of the strong affinity between As and ferric (hydr)oxide (Gu and Deng, 2007; Sylvester et al., 2007). Hossain et al. (2009) found the application of ferrous iron to pot-culture rice plants could decrease As uptake by rice grains and increase grain yields, while the addition of phosphates led to reduced rice yield and increased As contents in grain and straw. However, the interactions of organic As such as MSMA with ferric (hydr)oxide in paddy soil are largely unknown and little studied in terms of organic As phytoavailability and uptake. Understanding such interactions would be critical to minimize As uptake by rice plants in the fields containing elevated MSMA. In contrast to most recent studies focusing on inorganic As, this study was to investigate and compare As uptake by rice cultivars in the soils with elevated MSMA. The objectives were focused on evaluating As uptake of three rice cultivars in MSMA-elevated soil under two water management conditions, and testing the hypothesis that iron (Fe) redox transformation controls the MSMA-As solubility and extractability in soil.

Materials and Methods

Experimental procedures

The study site was at the USDA-ARS Dale Bumpers National Rice Research Center located near Stuttgart, Arkansas. Soil at the site was Dewitt silt loam (fine, smectitic, thermic Typic Albaqualf). Selected physiochemical properties of the pretreated soil were listed in Table 1.

Table 1.

Selected Physiochemical Properties of the Pretreated Soil

Soil properties	Value ^a
pHs	5.9
Organic matter (%)	1.7
Cation exchange capacity (meq/100 g)	9.5
Neutralizable acidity (meq/100 g)	4.3
Electrical conductivity (Mmho/cm)	0.2
NO₃-N (mg/kg)	4.2
NH₄-N (mg/kg)	3.5
Exchangeable potassium (mg/kg)	180
Available phosphorous (mg/kg)	31
Total As (mg/kg)	5.0
Sand (%)	13.8
Silt (%)	70
Clay (%)	16.5

Average of the composited samples. More data can be found at Yan et al. (2008).

As, Arsenic.

The experiment was conducted in 2009, consisting of 1.8-×1.5-m plots arranged in split-split plot design where soil As levels were main plots and water management practices sub-plots. Within each sub-plot, three rice cultivars were completely randomized in each of the four replications or blocks. Three rice cultivars (Zhe 733, Cocodrie, and Rondo) were selected according to their straighthead susceptibility (resistant-Zhe 733, intermediate-Rondo, and susceptible-Cocodrie) (Yan et al., 2008; Yan and McClung, 2010). Soil As treatment included low and high As levels. Low As level was the native soil that is common in the southern rice belt where As-containing chemicals have never been applied, while high As level was the soil that had been applied with MSMA in alternate years since 1980s for rice straighthead disease evaluation (Yan et al., 2005). Prior to planting, the high As soil was applied with MSMA at a rate of 6.7 kg MSMA/ha to bring total soil As to ∼20 mg/kg. Predetermined amount of the MSMA solution was sprayed on the soil surface and incorporated into 20-cm soil depth by a fine tiller as described by Yan et al. (2005). Weed control and nitrogen fertilization (46-0-0) followed a recommended procedure (Yan et al., 2005). Each selected cultivar was then drill-seeded in six rows each 2 cm deep at a seeding rate of 65 kg/ha⁻¹ (Slaton and Cartwright, 2001). Field water management included flooded and saturated water practices. In the flooded practice, water stand was maintained all the time during the growing season from about the five-leaf stage to full maturity, while the saturated practice maintained soil moisture at or above the field capacity. When the soil surface became naturally dry or soil moisture dropped below the field capacity, the field was irrigated up to 10 cm of water stand. The irrigation water (pH ∼7.0, <2 μg As/L) was primarily from rainfall collected in a reservoir near the study site.

Sampling procedures

Soil samples were collected bi-weekly from each plot starting from the time of seed planting. Each time, at least 100 g of soil was taken individually from the center and four corners of each plot at 10-cm soil depth where most rice roots distribute. The collected samples were composited per plot, then air-dried, ground to pass a 0.25-mm sieve, and stored at 4°C prior to analysis. At maturity, rice grains were harvested per plot, oven-dried for 12 h at 65°C, and then crushed, separated from the husk, ground into powder, and stored at room temperature prior to analysis.

Analytical procedures

All chemicals used in this study were of at least reagent grade. Nitric acid (HNO₃, trace metal grade), hydrochloric acid (HCl), potassium phosphate dibasic (K₂HPO₄), sodium hydroxide (NaOH), sodium bicarbonate (NaHCO₃), sodium citrate (Na₃C₆H₅O₇·2H₂O), iron(II) sulfate heptahydrate (FeSO₄·7H₂O), hydrogen peroxide (H₂O₂, 30%), and inductive-coupled plasma (ICP) As, iron, or phosphate standard solutions were purchased from Fisher Scientific Co., and sodium dithionite (Na₂S₂O₄) from Sigma-Aldrich Co. MSMA used for the laboratory study was purchased from Sigma-Aldrich Co., while MSMA used for the field was purchased in Almyra Farmer Co. in a 2.5 gallon jar (3.00 kg of MSMA per gallon).

Total As analyses

About 1.0 g of soil or rice powder was digested with 10 mL of concentrated HNO₃ and 5 mL 30% H₂O₂ in a microwave oven digester (Ethos EZ; Milestone Co.) for 20 min at 180°C. The digested solution was filtered through a Whatman 40 filter paper, diluted with deionized water to 25 mL, and filtered again through a 0.45 μm syringe filter prior to analysis.

Sodium dithionite-sodium citrate-sodium bicarbonate solution extractable As, Fe, P analyses

1.0 g of soil was extracted by 15 mL of sodium dithionite-sodium citrate-sodium bicarbonate solution (DCB) solution overnight at room temperature (Liu et al., 2004). The extracted solution was filtered through a Whatman 40 filter paper and a 0.45 μm syringe filter before analysis.

The digested or extracted solutions were prepared in duplicate and analyzed within one week. The concentrations of Fe and P in the solutions were determined by an ICP-optical emission spectrometer (Varian Vista-PRO). A graphite furnace atomic adsorption spectrometer (Varian, 220F) was used to analyze As in soils, while ICP/mass spectrometers (VG Axiom high-resolution and ICP-mass spectrometer; Perkin-Elmer Elan) were used for determining As in rice flour. A standard rice flour (NIST, SRM 1568a) and a standard Montana soil (NIST, SRM 2711) were included in the extraction and analytical procedures for data quality control/assurance (QC/QA), with spiked recovery being in the range of 93%–98%.

To test the hypothesis that the iron redox status in soil controls As availability to rice plants, ferric (hydr)oxide was synthesized by oxidizing ferrous irons in the presence of MSMA to assess its ability to immobilize MSMA. The immobilized As by ferric (hydr)oxide was determined by the extraction using DCB, K₂HPO₃, or HCl solutions, respectively. The procedures were as follows: (i) transferred 10 g of Fe as FeSO₄ and 0.1 g of As as MSMA into a 1-L beaker; (ii) dissolved the solids with 300 mL Millipore water (18.2 MΩ·cm), and bubbled the solution with air for 24 h; (iii) neutralized the solution to pH 7.0 with NaOH solution, and diluted to 500 mL with Millipore water; (iv) filtered the solid using a Whatman 40 filter paper and washed the solid with 20 mL Millipore water three times; (v) oven-dried the solid at 70°C overnight; (vi) weighed 0.50 g of the solid in three centrifuge tubes containing 15 mL DCB, 0.5 M K₂HPO₃, or 1.0 M HCl, respectively, and shook the tubes for 24 h; (vii) filtered the suspensions with 0.45 μm syringe filter and measured As in the filtrates. The As:Fe ratio of 0.1:10 was selected to mimic the As:Fe ratio in the field soil.

Results and Discussion

As in rice grains

Total As concentrations in grains of the three selected rice cultivars as influenced by soil As levels and water managements were presented in Fig. 1 and analysis of variance in Table 2. The analysis of variance results in Table 2 indicated that the most significant factor affecting the As accumulation was water management practice (p<0.003), followed by soil MSMA level (p<0.008). Rondo and Cocodrie had statistically similar As contents (p<0.98), while Zhe 733 contained much less As (p<0.07 or 0.12), suggesting a cultivar-dependent resistance to elevated As in soil. Higher As level in the soil (18.3±1.2 μg As/g), in all cases, resulted in a significantly higher As content in grains as compared with the native soil (5.0±0.3 μg As/g) (p<0.008). Rondo or Cocodrie was found to be more susceptible to elevated soil As level, while Zhe 733 was less susceptible. Data were consistent with the survey showing the As content range of 5 to 710 μg/kg among 204 rice cultivars (Zavala and Duxbury, 2008). Under the two water management practices, significantly higher As concentrations in the grains were measured in the flooded than the saturated soils regardless of the As level and cultivar (p<0.003). In addition, Rondo and Cocodrie showed a consistent behavior in response to flooded water management in both MSMA-treated and native soils, while Zhe responded more strongly to flooded water management in the native soil. In the flooded conditions, the soil might be in a more reduced status (Takahashi et al., 2004). This implied that soil redox reactions, especially iron transformation from ferric to ferrous forms as induced by flooding, might enhance the As release from solid phases, thus increasing As availability for rice uptake. Preliminary field data also found that As content in rice grains was positively related to the straighthead rating (the higher the rating, the more severe the symptom) and negatively related to the grain yield for Rondo and Cocodrie (data not shown).

FIG. 1.

As concentration in grains of three rice cultivars as influenced by soil As levels and water management practices (each bar represents average of duplicate of samples; error bars represent standard deviations). As, Arsenic.

Table 2.

Analysis of Variance for Arsenic Content in Rice Grains as Effected by Soil Arsenic Level, Water Management, and Cultivar

Factors	F_calculated	F_critical	α	p
As concentration (MSMA vs. No MSMA)	8.43	4.30	0.05	0.008
Water management practice (Flooded vs. Saturated)	11.47	4.30	0.05	0.003
Cultivar I (Zhe 733 vs. Rondon)	3.79	3.10	0.1	0.07
Cultivar II (Zhe 733 vs. Cocodrie)	2.77	3.10	0.1	0.12
Cultivar III (Rondon vs. Cocodrie)	0.0009	3.10	0.1	0.98

F_calculated, the quotient of the square of standard deviations; F_critical, tabulated F value; α, significant level; p, probability; MSMA, monosodium methanearsonate.

Relationships among soil As, Fe, and P

In an effort to confirm our speculation that As solubility be closely associated with ferric (hydr)oxide in the soil and the redox status of iron species control the As availability to rice plants, soil samples were analyzed for DCB-extractable As, Fe, and P. As shown in Fig. 2, DCB-extractable As in the soils was found positively correlated with DCB-extractable Fe. Total As concentration in the native soil was measured at 5.0±0.3 μg/g; while the As in the MSMA-treated soil was 18.3±1.2 μg/g. The strong positive correlation between DCB-extractable As and Fe indicated that As in the soils was closely associated with amorphous ferric (hydr)oxide. When soil iron was reduced as induced by the flooded water management practice, the As associated with the solid phase of ferric (hydr)oxide would be released into the aqueous phase, leading to the enhanced As uptake by rice plants. In addition, data in Fig. 2 also implied that the presence of relatively high As concentrations in the soil would lead to more DCB-extractable Fe. This was consistent with the findings by Violante et al. (2007, 2009), where arsenate was found strongly to retard the crystallization of goethite and hematite, leading to the formation of more poorly crystalline and relatively soluble precipitates.

FIG. 2.

Correlation between DCB-extractable Fe and As in soil (each point represents average of six samples; error bars are standard deviations). DCB, sodium dithionite-sodium citrate-sodium bicarbonate solution.

Measurements of DCB-extractable As, Fe and P presented in Fig. 3 illustrated that there was a positive correlation between P and Fe (Fig. 3a), and DCB-extractable P concentrations were generally higher in the soils with high As and lower in the soils with low As (Fig. 3b). It is well known that formation of ferric (hydr)oxide due to changes in soil redox conditions is an important process controlling soil P solubility and availability (Manahan, 1991). Under oxidized conditions, soluble P can be easily immobilized through sorption by ferric (hydr)oxide and/or formation of relatively insoluble iron phosphate. As a result of ferric reduction to ferrous form under reduced conditions, insoluble iron-associated P fraction could be released into the aqueous phase, facilitating plant P uptake. The positive correlations among Fe, P and As suggested that there was no apparent inhibition between As and P for sorption by ferric (hydr)oxide in the soil. It was reported that arsenate has similar adsorptive behavior as phosphate due to their structural similarities as well as acid dissociation constants (Zeng et al., 2008). Zhang et al. (2008) demonstrated that monomethylarsonate, arsenate, and phosphate had comparable affinities to goethite. It is therefore not surprising that the soil As, Fe, and P were closely related and the formation of ferric (hydr)oxide/iron phosphate controlled the solubility or availability of both P and As in the soils. Whether phosphate application could reduce As accumulation in rice plants is dependent on experimental conditions. Abedin et al. (2002) reported that there was no significant difference in As concentrations in plants due to phosphate application. However, Liu et al. (2004) found that phosphate addition to hydroponic rice plants (CDR22) significantly decreased As concentrations in iron plaque and increased As concentrations in the shoots.

FIG. 3.

Correlation between DCB-extractable P and Fe (a) and between DCB-extractable P and As (b) in soil (each point represents average of duplicate of samples; error bars are standard deviations).

Interactions of MSMA with ferric (hydr)oxide

The adsorption of inorganic As on ferric (hydr)oxides and the adsorptive competition between inorganic As and phosphate have been extensively studied. The interactions of methyl As such as MSMA with ferric (hydr)oxides, however, are largely unknown. In an effort to substantiate our hypothesis that the formation of ferric (hydr)oxide would immobilize MSMA and reduce As availability for plant uptake, ferric (hydr)oxide was synthesized through the oxidation of aqueous Fe(II) in the presence of MSMA, and iron-associated As fractions were determined afterward by selective extraction procedures. Results presented in Fig. 4 confirmed that 8.5 mg As/g Fe (85% of total As added) was extracted by DCB from the synthesized ferric (hydr)oxide, among which 7.6 mg As/g Fe was 1.0 M HCl-extractable, suggesting a strong affinity of MSMA to the ferric (hydr)oxide. It was also found that 3.5 mg As/g Fe, about 41% of the DCB extractable fraction, could be extracted by 0.5 M K₂HPO₄ (pH=9.7). Possible explanation for this partial extraction by phosphate would be the incomplete As desorption from the ferric (hydr)oxide. Jackson and Miller (2000) and Lafferty and Loeppert (2005) found that adsorbed DMA on goethite could not be completely displaced by phosphate. As incorporation into ferric (hydr)oxide structure or formation of solid solution could be another mechanism responsible for the partial extraction. Violante et al. (2009) demonstrated that arsenate co-precipitated with ferric/aluminum (hydr)oxide was partially occluded into the structure.

FIG. 4.

As extracted from synthetic ferric (hydr)oxide by selective extraction procedures (each bar represents average of duplicate of samples; error bars are standard deviations).

Conclusions

This field study has demonstrated that As concentration in rice grains depends on soil As level, cultivar, and water management. Among three cultivars studied, Zhe 733 is less responsive, while Rondo and Cocodrie are more responsive to elevated MSMA-As in the soil. Water flooding could lead to enhanced As uptake by rice plants, which is linked to reduced conditions that induce the reductive dissolution of ferric (hydr)oxide and the mobilization of As from solid phase. Formation of ferric (hydr)oxide under the water-saturated management practice is able to immobilize soluble As in the soil, reducing As availability to rice plants. Results suggest that a significant reduction of As accumulation in rice grains could be achieved through appropriate selections of As-less responsive cultivars and saturated water management practices, which would minimize human health risk of oral As intake.

Footnotes

Acknowledgments

This research was funded by the USDA-NIFA through the grant No. 0215149 to Lincoln University of Missouri. Authors would like to thank Tiffany Sookaserm, Yao Zhou and Biaolin Hu for sample collection and preparation, Brandid Clark and Joy Pyles for their assistance in sample analyses. Contribution from the Cooperative Research Program, Lincoln University of Missouri, No. 2010002.

Author Disclosure Statements

No competing financial interests exist.

References

Abedin

M.J.

, Cotter-Howells

, Meharg

A.A.

2002. Arsenic uptake and accumulation in rice (Oryza sativa L.) irrigated with contaminated water. Plant Soil, 240:311.

Duxbury

J.M.

, Mayer

A.B.

, Lauren

J.G.

, Hassan

2003. Food chain aspects of arsenic contamination in Bangladesh: effects on quality and productivity of rice. J. Environ. Sci. Health, 38:61.

FAO. 2002. Rice Information, 3. Rome, Italy: FAO.

Gilmour

J.T.

, Wells

B.R.

1980. Residual effects of MSMA on sterility in rice cultivars. Agron. J., 72:1066.

, Deng

2007. Use of iron-containing mesoporous carbon (IMC) for arsenic removal from drinking water. Environ. Eng. Sci., 24:113.

Hossain

M.B.

, Jahiruddin

, Loeppert

R.H.

, Panaullah

G.M.

, Islam

M.R.

, Duxbury

J.M.

2009. The effects of iron plaque and phosphorus on yield and arsenic accumulation in rice. Plant Soil, 317:167.

Jackson

B.P.

, Miller

W.P.

2000. Effectiveness of phosphate and hydroxide for desorption of arsenic and selenium species from iron oxides. Soil Sci. Soc. Am. J., 64:1616.

Lafferty

B.J.

, Loeppert

R.H.

2005. Methyl arsenic adsorption and desorption behavior on iron oxides. Environ. Sci. Technol., 39:2120.

Liu

W.J.

, Zhu

Y.G.

, Smith

F.A.

, Smith

S.E.

2004. Do phosphorus nutrition and iron plaque alter arsenate (As) uptake by rice seedlings in hydroponic culture? New Phytol., 162:481.

10.

Manahan

S.E.

1991. Environmental Chemistry, 5th. Chelsea, MI: Lewis Publishers, Inc.

11.

Marin

A.R.

, Masscheleyn

P.H.

, Patrick

W.H.

Jr.

1993. Soil redox-pH stability of arsenic species and its influence on arsenic uptake by rice. Plant Soil, 152:245.

12.

Meharg

A.A.

2004. Arsenic in rice—understanding a new disaster for South-East Asia. Tr. Plant Sci., 9:415.

13.

National Research Council. 2001. Arsenic in Drinking Water Update. Washington, DC: National Academies Press.

14.

Slaton

N.A.

, Cartwright

2001. Rice stand establishment. Slaton

N.A.

Rice Production Handbook. Little Rock, AR: University of Arkansas Cooperative Extension Service. MP 192-10M-1-01RV21–28.

15.

Sylvester

, Westerhoff

, Moller

, Badruzzaman

, Boyd

2007. A hybrid sorbent utilizing nanoparticles of hydrous iron oxide for arsenic removal from drinking water. Environ. Eng. Sci., 24:104.

16.

Takahashi

, Minamikawa

, Hattori

H.K.

, Kurishima

, Kihou

, Yuita

2004. Arsenic behavior in paddy fields during the cycle of flooded and non-flooded periods. Environ. Sci. Technol., 38:1038.

17.

Ullah

S.M.

1998. Arsenic contamination of groundwater and irrigated soils of Bangladesh. In International Conference on Arsenic Pollution of Groundwater in Bangladesh: Causes, Effects and Remedies. DCH, Dhaka: 133 (abstract).

18.

Violante

, Del Gaudio

, Pigna

, Ricciardella

, Banerjee

2007. Coprecipitation of Arsenate with metal oxides, 2. Nature, mineralogy, and reactivity of iron(III) precipitates. Environ. Sci. Technol., 41:8275.

19.

Violante

, Pigna

, Del Gaudio

, Cozzolino

, Banerjee

2009. Coprecipitation of arsenate with metal oxides, 3. Nature, mineralogy, and reactivity of iron(III)-aluminum precipitates. Environ. Sci. Technol., 43:1515.

20.

Yan

, Agrama

A.H.

, Slaton

A.N.

, Gibbons

W.J.

2008. Soil and plant minerals associated with rice straighthead disorder induced by arsenic. Agron. J., 100:1655.

21.

Yan

, Dilday

H.R.

, Tai

H.T.

, Gibbons

W.J.

, McNew

W.R.

, Rutger

J.N.

2005. Differential response of rice germplasm to straighthead induced by arsenic. Crop Sci., 45:1223.

22.

Yan

, McClung

A.M.

2010. ‘Rondo’, a long-grain indica rice with resistances to multiple diseases. J. Plant Registrations, 4:131.

23.

Zavala

Y.J.

, Duxbury

J.M.

2008. Arsenic in rice: I. Estimating normal levels of total arsenic in rice grain. Environ. Sci. Technol., 42:3856.

24.

Zavala

Y.J.

, Gerads

, Gurleyuk

, Duxbury

J.M.

2008. Arsenic in rice: II. Arsenic speciation in USA grain and implications for human health. Environ. Sci. Technol., 42:3861.

25.

Zeng

, Fisher

, Giammar

D.E.

2008. Individual and competitive adsorption of arsenate and phosphate to a high-surface-area iron oxide-based sorbent. Environ. Sci. Technol., 42:147.

26.

Zhang

J.S.

, Stanforth

, Pehkonen

S.O.

2008. Irreversible adsorption of methyl arsenic, arsenate, and phosphate onto goethite in arsenic and phosphate binary systems. J. Colloid Interface Sci., 317:35.