Arsenic Sequestration in Iron Plaque, Its Accumulation and Speciation

of a Native Wetland Plant from the Savannah River Site under Reducing Conditions ..... Arsenic in cereals, their relation with human health risk, ...
0 downloads 0 Views 165KB Size
Download PDF
Environ. Sci. Technol. 2006, 40, 5730-5736

Arsenic Sequestration in Iron Plaque, Its Accumulation and Speciation in Mature Rice Plants (Oryza Sativa L.) W . J . L I U , †,‡ Y . G . Z H U , * ,† Y . H U , † P. N. WILLIAMS,§ A. G. GAULT,⊥ A . A . M E H A R G , § J . M . C H A R N O C K , ⊥,| A N D F. A. SMITH# Research Center for Eco-environmental Sciences, Chinese Academy of Sciences, China, and College of Resources and Environmental Sciences, Hebei Agricultural University, Baoding, Hebei Province, China, and School of Biological Sciences, University of Aberdeen, Aberdeen, AB24 3UU, UK, and School of Earth, Atmospheric and Environmental Sciences and Williamson Research Centre for Molecular Environmental Science, University of Manchester, Oxford Road, Manchester, M13 9PL, UK, and CCLRC Daresbury Laboratory, Daresbury, Warrington, WA4 4AD, UK, and Soil and Land Systems, School of Earth and Environmental Sciences, The University of Adelaide, SA5005, Australia

A compartmented soil-glass bead culture system was used to investigate characteristics of iron plaque and arsenic accumulation and speciation in mature rice plants with different capacities of forming iron plaque on their roots. X-ray absorption near-edge structure spectra and extended X-ray absorption fine structure were utilized to identify the mineralogical characteristics of iron plaque and arsenic sequestration in plaque on the rice roots. Iron plaque was dominated by (oxyhydr)oxides, which were composed of ferrihydrite (81-100%), with a minor amount of goethite (19%) fitted in one of the samples. Sequential extraction and XANES data showed that arsenic in iron plaque was sequestered mainly with amorphous and crystalline iron (oxyhydr)oxides, and that arsenate was the predominant species. There was significant variation in iron plaque formation between genotypes, and the distribution of arsenic in different components of mature rice plants followed the following order: iron plaque > root > straw > husk > grain for all genotypes. Arsenic accumulation in grain differed significantly among genotypes. Inorganic arsenic and dimethylarsinic acid (DMA) were the main arsenic species in rice grain for six genotypes, and there were large genotypic differences in levels of DMA and inorganic arsenic in grain.

Introduction Arsenic (As) is an ubiquitous metalloid, widely distributed in the environment through both natural and anthropogenic * Corresponding author phone: +86 10 6293 6940; fax: +86 10 6292 3563; e-mail: [email protected]. † Chinese Academy of Sciences. ‡ Hebei Agricultural University. § University of Aberdeen. ⊥ University of Manchester. | CCLRC Daresbury Laboratory. # The University of Adelaide. 5730

9

ENVIRONMENTAL SCIENCE & TECHNOLOGY / VOL. 40, NO. 18, 2006

pathways. Rice grown on As-contaminated paddy soils can accumulate As to high levels in grain posing a health risk to consumers (1, 2). Thus the transfer of As from soils or irrigation water to rice can amplify the health risk of As to people living in those areas. Therefore, it is necessary to assess the impact of As-contaminated irrigation water on As accumulation and distribution in different parts of mature rice plants. Iron plaque is commonly formed on the surfaces of rice roots as a result of the release of oxygen and oxidants into the rhizosphere (3). The structure of iron plaque is characterized as amorphous or crystalline iron (oxyhydr)oxides (35). Iron plaque induced artificially through adding ferrous iron in solutions has a strong affinity for arsenate (6-7). In paddy soils, it has been shown that As in irrigation water is sequestered in iron (oxyhydr)oxides in soil during the nonflooded period (8), and arsenate and arsenite coexist in the paddy soil solution, with arsenite being predominant in flooded conditions (9). However, the physio-chemical environments of the rhizosphere in paddy soil are very different from those in bulk soil because of the oxidation of rice roots and the formation of iron plaque on the root surface. The oxidative rhizosphere in wetland (paddy) soil is likely to alter the speciation of arsenic associated with the rice root surface (10-11). Arsenic on iron plaque of wetland plants is comprised predominantly (80 or 82%) of arsenate with lesser amounts (18 or 20%) of arsenite-iron (oxyhydr)oxide complexes, although the predominant species of arsenic in flooded soil is arsenite (10, 11). However, there is little information available regarding the speciation of arsenic in iron plaque formed on roots of rice plants. The potential risk of arsenic in rice grains should not be ignored (12). Our previous hydroponic experiments have revealed that iron plaque plays an important role in the behavior of arsenic on rice root surface (6, 13). However, in these studies we only considered the situation in solution culture (6, 13, 14). Because the rhizosphere does not exist in a solution culture, the oxygen released from root to growth medium is diluted and equilibrated quickly. The process of oxygen release from the rice root to the rhizosphere has already been shown in field paddy soil (15). Accordingly, in addition to solution culture studies, it is necessary to examine the effect of iron plaque on arsenic biogeochemistry in paddy soils and arsenic accumulation in grains. In addition, the toxicity of As in crops to humans depends not only on its total content but also on its chemical speciation. Inorganic arsenic compounds are thought to be more harmful than organic forms (1, 16). Although arsenic speciation has been studied thoroughly in seafood (17), the arsenic species in food of terrestrial origin are not well characterized (18). In order to understand the relationship between iron plaque formation and the dynamics of arsenic in rice-soil systems and arsenic accumulation in grain, the aim of the present study was to investigate (i) the distribution of arsenic in different components of mature rice plants; (ii) arsenic speciation in and sequestration with iron plaque formed naturally during the whole growth period in different rice genotypes, and (iii) arsenic accumulation and speciation in rice grain.

Materials and Methods Preparation of Rice Seedlings. Seeds of six rice (Oryza sativa L.) genotypes: YY-1, 94D-64, KY1360, Gui630, 94D-54, and 94D-22 (with different oxidation capacities) were obtained from Professor Li Damo, Institute of Subtropical Regional Agriculture, Chinese Academy of Sciences. Seeds were 10.1021/es060800v CCC: $33.50

 2006 American Chemical Society Published on Web 08/17/2006

disinfected in 30% H2O2 (w/w) solution for 15 min, followed by thorough washing with deionized water. They were germinated in moist perlite and were allowed to grow for 6 weeks in a growth chamber, with automatically controlled 14 h light period (8:00 am to 22:00 pm, 260-350 µE m-2 s-1), temperature (28 °C day and 20 °C night) and relative humidity (60-70%). Compartmented Rhizo-Bag Experimental Setup. The experimental soil was collected from Fuyang, Zhejiang province, China. The principal soil properties were measured following standard procedures, and were as follows. Texture loam; pH (H2O) 6.49; cation exchange capacity (CEC): 10 cmol kg-1; organic matter content: 22.6 g kg-1; Fe2O3: 1.87%; available P: 14.3 mg kg-1; available K: 137.4 mg kg-1; available N: 38.5 mg kg-1; total As: 13.8 mg kg-1. The soil was airdried and passed through a 2 mm sieve. Phosphorus as CaH2PO4 H2O at 0.15 g P2O5 kg-1, K as KCl at 0.2 g K2O kg-1, and N as CO (NH2) 2 at 0.2 g N kg-1 were thoroughly mixed as solid with soil at the start of the experiment to ensure adequate mineral nutrition for the growth of rice seedlings. A compartmented soil-glass bead culture system was set up. Rice seedlings were grown in the rhizo-bag (made of 35 µm nylon mesh) with glass beads, and the rhizo-bag was placed into a bigger pot (outer compartment) filled with soil. This system was used to reduce damage to rice roots with iron plaque when they are harvested at maturity (with grain). After 2 weeks of equilibrium, the 1.25 kg soil mixture was put into a 2.5 L porcelain pot (7.5 cm radii and 22 cm height) without a drainage hole at the bottom. Uniform 6-week old rice seedlings were then transplanted into rhizo-bags (4 cm radii and 18 cm height, one plant per bag) filled with glass beads (1 kg per bag). Finally the rhizo-bags with plants were inserted into the soil in the porcelain pots. Experimental Treatment and Management. After transplanting, the seedlings were grown under flooded conditions a layer of water about 2-3 cm above the soil surface. The pots with rice plants were placed in a greenhouse with temperature of 22 °C night and 35 °C day, and with ambient light intensity. Arsenic was supplied as a solution of Na3AsO4‚12H2O at tillering, stem elongation, booting, flowering, and grain filling stages. For each growth stage, plants were irrigated with 200 mL of a solution of 0.4 mg As L-1 for three times, this concentration was chosen based on reported As concentrations in contaminated environments (1, 13). The concentration of arsenic in the irrigating water was in accordance with As levels in wastewater or groundwater in arsenic-contaminated areas in China. The choice of arsenate in irrigation water was based on the fact that arsenate and arsenite (and methylated species) interchange, depending on microsite redox and microbial activity (8), in groundwater (19) and rhizosphere solution (9). During the experiment, 2-morpholinoethanesulfonic acid, monohydrate (MES, 0.5 g L-1) buffer solution was added twice into rhizobags to adjust and maintain pH values of the growth medium, as it tended to increase during the growth period. Mature rice was harvested after nearly 6 months. All pots were arranged randomly, and rearranged every week. Sampling Procedure. At maturity, plants were harvested and were separated into grain, husk, straw, and root. First, the ear was cut with stainless steel scissors and put them into nylon mesh bags to air-dry in the greenhouse. After 3 weeks, dry ears were washed thoroughly using tap water, rinsed with deionized (DI) water and oven-dried at 70 °C for 72 h. Dry spikelets were divided into brown rice and husks using a pestle and mortar. Grains and husks were milled in a ceramic mortar to homogenized powders and stored at -20 °C for analysis. Straw was harvested using stainless steel scissors. The base of each straw that was soaked in Ascontaminated irrigation water was discarded in order to

reduce error caused by contamination from irrigation water. The washing and drying procedures for rice straw were the same as rice ears. Dry straw samples were ground using a stainless steel mill, to fine homogenized powders and stored in sealed bags at room temperature. Root samples were taken from rhizobags gently to avoid damage to roots and iron plaque on the root surface. The root samples were washed with tap water to remove glass beads on the root surface and rinsed using DI water more than three times. The clean roots were stored immediately at -20 °C before analysis. One part of the roots was freeze-dried for analysis of arsenic speciation in iron plaque using extended X-ray absorption fine structure (EXAFS) spectroscopy and X-ray absorption near edge structure (XANES) spectra (20). Roots with intact iron plaque were purged in nitrogen during transport. The other subsample of roots was extracted using dithionite-citratebicarbonate (DCB) to measure Fe and As concentrations accumulated on the root surface. DCB Extraction of Iron Plaque. During harvesting, we observed that the color of rice roots was different from root tip to base. Therefore, root subsamples were cut into three sections: tip, middle, and base. Iron plaque on fresh root surfaces of three root sections was extracted with DCB (21, 22), for details refer to our previous publications (6, 7). After DCB extraction, roots were oven dried at 70 °C for 3 days and weighed. The concentrations of As and Fe in the DCB-extracts were measured by ICP-OES (inductively coupled plasma optical emission spectrometer, Optima 2000 DV, PerkinElmer, USA). Analysis of Arsenic Speciation in Iron Plaque Using X-ray Absorption Spectroscopy. Iron and Arsenic Model Compounds. Sodium arsenite (NaAsO2) and disodium arsenate heptahydrate (Na2HAsO4‚7H2O) were obtained from BDH Chemicals (Merck, UK). Natural samples of orpiment (As2S3), arsenopyrite (FeAsS), vivianite (Fe3(PO4)2‚8H2O), and siderite (FeCO3) were obtained from the Harwood mineral collection at the University of Manchester. Two-line ferrihydrite (Fe(OH)3), goethite (R-FeOOH), lepidocrocite (γ-FeOOH), and hematite (Fe2O3) were prepared according to the method of Schwertmann and Cornell (2000) (23). Immediately after synthesis, the iron (oxyhydr) oxide phases were washed with deionised water and centrifuged (4700 rpm, 20 min, 15 °C). This washing procedure was repeated a further five times prior to freeze drying or overnight oven drying at 50 °C (goethite only). Green rust (carbonate form, (Fe(II)4,Fe(III)2)(OH)12CO3) was synthesized following the procedure of Taylor et al. (24). The identity of the model compounds used was confirmed by powder X-ray diffraction (XRD). Details of on the preparation of arsenic-bearing iron (oxyhydr)oxide can be found in the Supporting Information. X-ray Absorption Spectroscopy. Arsenic K-edge X-ray absorption spectra were obtained on station 16.5 at the UK CCLRC Daresbury Synchrotron Radiation Source (SRS) operating at 2 GeV with a beam current of between 130 and 240 mA. Station 16.5 is equipped with a Si (220) double crystal monochromator, with harmonic contamination of the beam minimized by a vertically focusing mirror in addition to detuning to 70%. The monochromator was calibrated using the L(III) edge of a gold foil. The freeze-dried rice root and As(III)- or As(V)-bearing iron (oxyhydr)oxide wet pastes were mounted in an aluminum sample holder with Sellotape windows. Data were collected at liquid nitrogen temperature with the station operating in fluorescence mode using an Ortec 30 element solid-state Ge detector. Four scans were collected for each sample and summed to improve signalto-noise. Standards of sodium arsenite, disodium arsenate heptahydrate, orpiment, and arsenopyrite were diluted with boron nitride and collected at room temperature in transmission mode. VOL. 40, NO. 18, 2006 / ENVIRONMENTAL SCIENCE & TECHNOLOGY

9

5731

Unless otherwise stated, iron K-edge X-ray absorption spectra were obtained on station 7.1 at the UK CCLRC Daresbury SRS. A sagitally bent focusing Si (111) double crystal monochromator was utilized, with harmonic contamination of the beam minimized by a vertically collimating plane mirror as well as to detuning to 70%. Before collecting data, the monochromator was calibrated using a 5 µm iron foil. The rice root was presented to the beam in an aluminum sample holder with Sellotape windows. Sample data were collected in a liquid nitrogen cooled cryostat in either transmission or fluorescence mode; a three element solidstate Ge detector was used to collect data in the latter mode. Four scans were collected and summed for each sample. Lepidocrocite, hematite, 2-line ferrihydrite, vivianite, and siderite were collected at liquid nitrogen temperature in transmission mode after appropriate dilution with boron nitride for the Fe XAS study. The green rust sample was analyzed in transmission mode at 80 K using a liquid nitrogen cooled cryostat on station 16.5. The XAS analysis of the goethite standard was conducted at cryogenic temperature (80 K) on station 8.1 of the Daresbury SRS. A Si (220) monochromator was used, detuned to 50% of the maximum intensity to minimize harmonic contamination. The monochromator energy was calibrated using a 5 µm iron foil. The proportion of different arsenic or iron phases in each sample was established by fitting the summed sample X-ray absorption near edge structure (XANES) or extended X-ray absorption fine structure (EXAFS) spectra to a linear combination of end-member standard spectra using the Solver package included in Microsoft Excel, with the relative contribution of each standard determined by minimizing a least-squares residual. In addition, the EXAFS spectra (χ) were background-subtracted and plotted as a function of the photoelectron wave number, k. The EXAFS were k3 weighted in order to amplify the weak EXAFS oscillations at high k. EXAFS data analysis was performed with EXCURV98 using full curved wave theory (25, 26), with phase shifts calculated ab initio using Hedin-Lundqvist potentials and von Barth ground states (27). The experimental data were fitted by defining a theoretical model and comparing the calculated EXAFS spectrum with the experimental data. Shells of backscatterers were added around the central absorber atom and the absorber-scatterer distance (r), Fermi energy and Debye-Waller factor (2σ2) were refined until a least-squares residual was minimized. For each shell of scatterers, the number of atoms in the shell was chosen as the integer or half integer to give the best fit, but was not further refined. Additional shells of scatterers were only considered justified if they improved the final fit of the data significantly. Sequential Extraction of Arsenic in Iron Plaque. Sequential extraction schemes are commonly used to evaluate the solid-phase partitioning of arsenic (e.g., ref 28), however, it should be noted that the phases of interest are operationally defined by the reagents and experimental conditions used (e.g., “poorly crystalline” refers not to a direct measure of crystallinity but to a fraction that is dissolved in a particular reagent). Arsenic in iron plaque was extracted sequentially using a method modified from Wenzel et al. (28) (Table S1, Supporting Information). Segments of root (Gui630 and 94D22 were selected) coated with iron plaque (0.3 g) were placed in 50 mL centrifuge tubes and 12 mL of the extraction reagents (Merck, Germany) were added sequentially according to Table 1 (Supporting Information). Shaking speed was controlled at 150 rpm. After each extraction step, the tube containing the root with iron plaque and the extractant were centrifuged for 15 min at 3500 rpm. The solution was filtered through a membrane filter (0.45 µm, MFS, USA). A solution from each step was kept in a clean plastic tube for analysis. Arsenic concentrations in extractants were determined using 5732

9

ENVIRONMENTAL SCIENCE & TECHNOLOGY / VOL. 40, NO. 18, 2006

TABLE 1. Amounts of Iron Plaque On Different Root Zones (Base, Middle and Tip) for Six Rice Genotypes (means ( SE, n ) 3) and Analysis of Variance by Two-Way ANOVA Fe concentrations in iron plaques (g kg-1) genotypes YY-1 94D-64 KY1360 Gui630 94D-54 94D-22 genotype (G) zone (Z) GXZ

root base

middle section

68.4 ( 10.0 173.5 ( 10.9 25.9 ( 13.1 68.8 ( 18.3 22.3 ( 3.1 49.1 ( 3.2 76.6 ( 6.8 106.1 ( 4.5 41.6 ( 4.3 78.7 ( 5.9 66.0 ( 6.6 98.7 ( 6.3 analysis of variance base. Comparison of the iron K-edge XANES spectra of the rice root samples with a suite of iron reference compounds indicates that iron mineralogy of the root plaque was dominated by (oxyhydr)oxides (data not shown), however, the lack of spectral features unique to different ferric (oxyhydr)oxides limits the utility of XANES analysis to delineate contributions from such phases (29). Data collected into the iron EXAFS energy region show clear differences between iron (oxyhydr)oxide minerals, allowing the proportion of such phases in unknown samples to be calculated (Figure 1). The EXAFS data further showed that the iron plaque on the rice root was dominated by ferrihydrite (81100%), with a minor amount of goethite (19%) fitted in one of the samples. Arsenic Sequestration in Iron Plaque and its Speciation. The predominant fractions of arsenic were associated with amorphous and poor-crystalline hydrous oxides of iron (Gui630: 274 mg kg-1; 94D-22: 210 mg kg-1) (Table 2). The proportions of this fraction to total arsenic of the four fractions were 59.8% and 57.0% for Gui630 and 94D-22, respectively. The second most abundant arsenic fraction was wellcrystallized iron (oxyhydr)oxides (genotype Gui630: 174 mg

FIGURE 2. As K-edge XANES spectra of As(III)- and As(V)-sorbed on 2-line ferrihydrite and two rice root samples. The experimental spectra and best fit of the rice root data are displayed as solid and dotted lines respectively (root 1: 71% As(V)-ferrihydrite, 29% As(III)-ferrihydrite; root 2: 74% As(V)-ferrihydrite, 26% As(III)-ferrihydrite).

FIGURE 1. Normalized Fe K-edge EXAFS spectra of Fe (oxyhydr)oxide model compounds and two rice root samples. The experimental spectra and best fit of the rice root data are displayed as solid and dotted lines respectively (root 1: 81% 2-line ferrihydrite, 19% goethite; root 2: 100% 2-line ferrihydrite). kg-1; Genotype 94D-22: 148 mg kg-1). Arsenic concentrations in the fractions of nonspecifically sorbed and specifically sorbed were very low; i.e., most arsenic in iron plaque was sequestered in amorphous and well-crystalline hydrous oxides of iron (Table 2). To further understand the redox state of arsenic on the root surface, XANES and EXAFS spectra were used to analyze arsenic speciation in the iron plaque. Arsenic K-edge XANES and EXAFS spectra obtained from analysis of two different root samples were near identical (Figure 2 and Figure S1 in the Supporting Information, respectively). The XANES spectra for two of the rice root samples was dominated by As(V), with the pronounced shoulder at lower energy indicating the presence of a smaller proportion of As(III) (Figure 2). A two component mixture of 71-74% As(V) and 26-29% As(III), sorbed to ferrihydrite, matched the XANES spectra well; the incorporation of further As species components, either orpiment or arsenopyrite, did not improve the fit. The oscillations giving rise to the first peak in the Fourier transform (Figure 3a) were best fitted with a shell of four oxygen atoms at 1.69-1.70 Å (Table 3), consistent with the XANES assignment that the arsenic in the samples is dominated by As(V). The inclusion of multiple scattering of the outgoing photoelectron within the arsenate tetrahedron in the fit further improved the least-squares residual. Outer shell iron scatterers were identified at 3.26-3.28 Å, suggesting that the arsenic associated with the rice root samples was sorbed on iron minerals, most likely iron oxyhydroxides.

FIGURE 3. Normalized As K-edge EXAFS spectra (a) and radial distribution function (b) of two rice root samples. Black lines represent experimental data and gray lines the leasy squares best fit based on the parameters listed in Table 3. Arsenic Accumulation in Iron Plaque and Mature Rice. The average distribution of arsenic in mature plants (all in mg kg-1) of the six genotypes was 515 ( 67.61 in iron plaque, 98.0 ( 2.86 in roots, 18.1 ( 3.95 in straw, 2.3 ( 0.03 in husks, and 0.5 ( 0.04 in grains. Detailed information can be found in Table S2 (Supporting Information). Concentrations of As in iron plaque formed during the entire growth period ranged from 300 mg kg-1 to 800 mg kg-1 for all genotypes.

TABLE 2. Arsenic Concentrations and Distributions of Different Fractions after Sequential Extraction for Two Genotypes Gui630 and 94D-22 rice genotypes Gui630 step

fractions

1a 2b 3c 4d

NH4 (SO4) 2(0.05M) NH4H2PO4 (0.05M) NH4-oxalate buffer (0.2 M); pH 3.25 NH4-oxalate 177408n (page.2 M) + ascorbic acid (0.1 M); pH 3.25

94D-22

arsenic concentrations (mg kg-1)

percentagee (%)

arsenic concentrations (mg kg-1)

percentage (%)

2.42 ( 0.72 7.45 ( 1.32 274 ( 2.25 174 ( 5.16

0.5 1.6 59.8 38.0

2.67 ( 0.76 7.44 ( 1.26 210 ( 7.61 148 ( 8.67

0.2 2.0 57.0 40.2

a NH (SO ) b NH H PO specifically sorbed c NH -oxalate: amorphous and poor-crystalline hydrous oxides of iron 4 4 2: nonspecifically sorbed 4 2 4: 4 NH4-oxalate + ascorbic acid: well-crystalline hydrous oxides of iron e Percentage was calculated based on the totral amount of Fe extracted in these procedures. Residual Fe as compared to the total Fe extracted by DCB was around 40% for both roots. d

VOL. 40, NO. 18, 2006 / ENVIRONMENTAL SCIENCE & TECHNOLOGY

9

5733

have the highest oxidation capacity, i.e., high radial oxygen loss as found in other wetland plants (31-33), this may also be due to relatively large surface area of the root tips as they had much smaller diameters. Although As concentration in the soil used here was only around 15 mg kg-1, concentrations of arsenic in iron plaque were very high (up to 800 mg kg-1) demonstrating that the iron plaque formed naturally in the soil has strong affinity for arsenic similar to that induced in nutrient solution (14). The results of XANES and EXAFS showed that arsenate predominated over arsenite in iron plaque. This further suggests that part of arsenite in the rhizosphere of wetland plants may be oxidized to arsenate due to oxygen or oxidants released by roots and accumulate in iron plaque on the root surface (10). The existence of iron plaque may create a gradient for arsenic diffusion toward root surface due to the adsorption of arsenic species to iron plaque. The present result with XANES also showed that the arsenic associated with roots was mostly sorbed on iron (oxyhydr)oxides. This As-Fe interaction was consistent with bidentate arsenic oxyanions occupying corner-sharing positions at the apexes of iron oxyhydroxide octahedra (34-35). Sequential extraction showed that most arsenic was sequestered in amorphous and crystalline of iron (oxyhydr)oxides. Our study demonstrated that with uniform soil arsenic concentration, there was a large variation in total arsenic in grains of different genotypes (YY-1: 0.69 mg kg-1; 94D-64: 0.32 mg kg-1). This large genotypic variation in grain As concentrations should encourage breeding new rice cultivars (genotypes) with low As accumulation for areas with Ascontaining irrigation water and soils. Human exposure to arsenic through rice ingestion has not been widely investigated. However, recently Williams et al. (36) reported that a grain arsenic level of 1.16 and 0.26 g As g-1 would equate to 45% and 75%, respectively, of the WHO’s provisional maximum tolerable daily intake (MTDI). Therefore, rice may represent an important As exposure pathway for subsistence rice consumers. Arsenic speciation in rice grain is an important factor in risk assessment of As in rice on human health, as it is generally accepted that organic arsenic species are less toxic to humans, and that only inorganic species are considered in risk assessment (1, 36). The results obtained here revealed that the six genotypes not only had different total grain arsenic concentrations, but also different arsenic speciation (Table 4); this observation is in agreement with the finding by Williams et al. (18). Mechanisms responsible for the difference in arsenic speciation in rice grain are not yet known, but based on the general pathway of arsenic metabolism, the production of organic arsenic requires the reduction of arsenate to arsenite, and subsequent methylation (37). Our results indicated that the presence of iron plaque might affect arsenic speciation in grains. The amounts of iron plaque were the lowest for KY1360 and 94D-64, and the proportions of inorganic arsenic were the highest (69.4% and 64.5%) (Table 4). In contrast, 94D-54 and 94D-22 had higher capacities to form iron plaque

TABLE 3. Parameters Obtained from Fitting As K-edge EXAFS Spectra for Roots of Rice Genotypes: Gui630 and 94D-22) sample

scatterer

Na

r (Å)

2s2 (Å2)

root 1 (Gui630)

O MSb Fe

4 12 2

1.70 3.09 3.28

0.010 0.016

root 2 (94D-22)

O MS Fe

4 12 2

1.69 3.07 3.26

0.010 0.015

a N is the coordination number ((25%), r is the interatomic distance ((0.02 Å for the first shell, (0.05 Å for more distant shells) and 2σ2 is the Debye-Waller factor ((25%). b A multiple scattering (MS) contribution to the outer shell is included in the fit (As-O-O’-As paths length equivalent to 3.09 and 3.07 for root 1 and root 2, respectively).

Arsenic Speciation in Rice Grains. Arsenic species, MMA, DMA, As(III), and As(V) were analyzed in rice grain for all genotypes. During extraction TFA might reduce arsenate to arsenite, so the levels of total inorganic arsenic were adopted instead of the concentrations of As(III) and As(V). The main species were inorganic arsenic (40.8-69.4%) and DMA (30.654.7%) for all genotypes (Table 4). The concentrations of MMA were very low (0.01-0.03 µg g-1), and not detected in KY1360. There were genotypic differences in levels of DMA and inorganic arsenic in the grain. The concentrations of inorganic As were higher than those of DMA for genotypes YY-1, Gui630, KY1360, and 94D-64. For 94D-54 and 94D-22, the opposite was observed. In short, the levels of inorganic As in grains followed the order of KY-1360 > 94D-64 > Gui 630 > YY-1 > 94D-54 and 94D-22; while the order for DMA levels was YY-1 > 94D-54 > 94D-22 > Gui 630 > KY-1360 and 94D-64 (Table 4).

Discussion In general, the formation of iron plaque depends on number of factors, such as ferrous ion concentration in a soil solution and oxygen or oxidants released by roots. Although the different genotypes of rice were grown in the same medium containing uniform ferrous ion, the amounts of iron plaque formed on the root surface were different among genotypes. This difference in iron plaque formation could reflect, at least partly, the genotypic difference in the oxidation capacity of roots during the growth period. It is generally believed that under field conditions the processes occurring in the rhizosphere are more complicated, as considerably more Fe(III) accumulates in the rhizosphere soil (30), this is also evident as glass beads in the mesh bag turned reddish during the experimental period (picture not shown). It is admitted that the culture system we used in the present study is to mimic the complex rhizosphere processes, and is a step closer to the field conditions as compared to solution culture systems reported (13, 14). In addition, the amounts of iron plaque (mg Fe per unit root biomass) varied significantly with the sections of the root. The ranking (root tip > middle root > root base) may be related to the fact that the root tips

TABLE 4. Proportions of Arsenic Species in Grain for Six Rice Genotypes Using TFA Extraction and HPLC-ICP-MS Measurement genotypes

inorganic As (µg/g)

inorgani As (%)

DMAb (µg/g)

DMA (%)

MMAc (µg/g)

recoveryd (%)

YY-1 94D-64 KY-1360 Gui 630 94D-54 94D-22

0.35 ( 0.002 0.21 ( 0.02 0.22 ( 0.001 0.24 ( 0.06 0.17 ( 0.01 0.15 ( 0.001

52.3 64.5 69.4 56.2 41.2 40.8

0.29 ( 0.02 0.10 ( 0.00 0.10 ( 0.00 0.18 ( 0.04 0.22 ( 0.03 0.19 ( 0.003

43.8 32.0 30.6 42.6 54.7 53.7

0.03 ( 0.002 0.01 ( 0.00 nda 0.01 ( 0.00 0.02 ( 0.004 0.02 ( 0.002

67 ( 1.58 72 ( 4.66 64 ( 0.501 85 ( 18.7 68 ( 7.39 58 ( 0.27

a nd: not detected. b DMA, dimethyl arsinic acid. c MMA, monomethyl arsinic acid. d The % recovery refers to the recovery of the postcolumn arsenic extracted by 2 M TFA compared with total As from a nitric acid digestion, not total As from the TFA extraction. % Recovery ) ([As] sum of species TFA extraction/[As] hot conc. nitric acid digestion) × 100.

5734

9

ENVIRONMENTAL SCIENCE & TECHNOLOGY / VOL. 40, NO. 18, 2006

than KY1360 and 94D-64, and the percentage of inorganic arsenic (41%) and concentrations of inorganic As were the lowest among the six genotypes. Our previous study showed that the presence of iron plaque may favor the influx of arsenite relative to arsenate (12), due mainly to the lower affinity between molecular As(OH)3 and iron oxide. The relative abundance of arsenite influx may be conducive to the formation of organic arsenic (37). The results of our study and Williams et al. (18) indicated that inorganic arsenic and DMA were dominant species in rice grain. The market basket survey in some contaminated areas of Bangladesh showed that the concentrations of inorganic arsenic were comparably high and the proportion of arsenic species varied greatly depending on the kinds and the sources of grains (38). In conclusion, this detailed study provides holistic information regarding the distribution and speciation of arsenic in mature rice plants, and shows that arsenic in the iron plaque of rice plants is mainly associated with amorphous and crystalline of iron (oxyhydr)oxides, and arsenic exists mainly as arsenate. There is significant difference in iron plaque formation and accumulation, and arsenic speciation between different genotypes, suggesting the possibility of breeding rice cultivars with less arsenic and relatively high DMA in rice grains. Further research is needed to clarify the relationship between iron plaque formation and arsenic accumulation/speciation in rice plants.

Acknowledgments The work was financially supported by the Natural Science Foundation of China (40225002 and 20477055), Ministry of Science and Technology (2002CB410808). We are grateful to CCLRC for the provision of beam time at Daresbury, and thank Mr. Bob Bilsborrow and Dr. Lorrie Murphy for assistance with the XAS data collection.

Supporting Information Available Details of grain As speciation, sequential extraction of arsenic from iron plaque, preparation of arsenic-bearing iron (oxyhydr)oxide, and arsenic concentrations in plants. This material is available free of charge via the Internet at http:// pubs.acs.org.

Literature Cited (1) Abedin, M. J.; Cresser, M. S.; Meharg, A. A.; Feldmann, J.; CotterHowells, J. Arsenic accumulation and metabolism in rice (Oryza sativa L.). Environ. Sci. Technol. 2002a, 36, 962-968. (2) Xie, Z. M.; Huang, C. Y. Control of arsenic toxicity in rice plants grown on an arsenic-polluted paddy soil. Commun. Soil Sci. Plant Anal. 1998, 29, 2471-2477. (3) Chen, C. C.; Dixon, J. B.; Turner, F. T. Iron coatings on rice roots: Morphology and models of development. Soil Sci. Soc. Am. J. 1980, 44, 1113-1119. (4) Bacha, R. E.; Hossner, L. R. Characteristics formed on rice roots as affected by Fe and Mn additions. Soil Sci. Soc. Am. J. 1977, 41, 931-935. (5) Hansel, C. M.; Fendorf, S.; Sutton, S.; Newville, M. Characterization of Fe plaque and associated metals on the roots of minewaste impacted aquatic plants. Environ. Sci. Technol. 2001, 35, 3863-3868. (6) Liu, W. J.; Zhu, Y. G.; Smith, F. A.; Smith, S. E. Do iron plaque and genotypes affect arsenate uptake and translocation by rice seedlings (Oryza sativa L.) grown in solution culture?. J. Exp. Bot. 2004a, 55, 1707-1713. (7) Liu, W. J.; Zhu, Y. G.; Smith, F. A.; Smith, S. E. Do phosphorus nutrition and iron plaque alter arsenate (As) uptake by rice seedlings in hydroponic culture?. New Phytol. 2004b, 162, 481488. (8) Takahashi, Y.; Minamikawa, R.; Hattori, K.; Kurishima, K.; Kihou, N.; Yuita, K. Arsenic behavior in paddy fields during the cycle of flooded and non-flooded periods. Environ. Sci. Technol. 2004, 38, 1038-1044. (9) Abedin, M. J.; Feldmann, J.; Meharg, A. A. Uptake kinetics of arsenic species in rice plants. Plant Physiol. 2002b, 128, 11201128.

(10) Hansel, C. M.; La Force, M. J.; Fendorf, S.; Sutton, S. Spatial and temporal association of As and Fe species on aquatic plant roots. Environ. Sci. Technol. 2002, 36, 1988-1994. (11) Blute, N. K.; Brabander, D. J.; Hemond, H. F.; Sutton, S. R.; Newville, M. G.; Rivers, M. L. Arsenic sequestration by ferric iron plaque on cattail roots. Environ. Sci. Technol. 2004, 38, 6047-6077. (12) Meharg, A. A.; Rahman, M. Arsenic contamination of Bangladesh paddy soils:Implications for rice contribution to arsenic consumption. Environ. Sci. Technol. 2003, 37, 229-234. (13) Chen, Z.; Zhu, Y. G.; Liu, W. J.; Meharg, A. A. Direct evidence showing the effect of root surface iron plaque on arsenite and arsenate uptake into rice (Oryza sativa L.) roots. New Phytol. 2005, 165, 91-97. (14) Liu, W. J.; Zhu, Y. G.; Smith, F. A. Effects of iron and manganese plaques on arsenic uptake of by rice seedlings (Oryza sativa L.) grown in solution culture supplied with arsenate and arsenite. Plant Soil. 2005, 277, 127-138. (15) Armstrong, J.; Armstrong, W. Rice and Phragmites: effects of organic acids on growth, root permeability, and radial oxygen loss to the rhizosphere. Am. J. Bot. 2001, 88, 1359-1370. (16) Cullen, W. R.; Reimer, K. J. Arsenic speciation in the environment. Chem. Rev. 1989, 89, 713-764. (17) Francesconi, K. A.; Edmonds, J. S. Biotransformation of arsenic in the marine environment. In Arsenic in the Environment, Part I: Cycling and Characterization; Nriagu, J. O., Ed.; John Wiley & Sons: New York, 1994; pp 221-263. (18) Williams, P. N.; Price, A. H.; Raab, A.; Hossain, S. A.; Feldmann, J.; Meharg, A. A. Variation in arsenic speciation and concentration in paddy rice related to dietary exposure. Environ. Sci. Technol. 2005, 39, 5531-5540. (19) Samanta, G.; Clifford, D. A. Preservation of inorganic arsenic species in groundwater. Environ. Sci. Technol. 2005, 39, 88778882. (20) Lombi, E.; Zhao, F. J.; Fuhrmann, M.; Ma, L. Q.; McGrath, S. P. Arsenic distribution and speciation in the fronds of the hyperaccumulator Pteris vittata. New Phytol. 2002, 156, 195203. (21) Taylor, G. J.; Crowder, A. A. Use of DCB technique for extraction of hydrous iron oxides from roots of wetland plants. Am. J. Bot. 1983, 70, 1254-1257. (22) Otte, M. L.; Dekkers, M. J.; Rozema, J.; Broekman, R. A. Uptake of arsenic by Aster tripolium in relation to rhizosphere oxidation. Can. J. Bot. 1991, 69, 2670-2677. (23) Schwertmann, U.; Cornell, R. M. Iron Oxides in the Laboratory, 2nd ed.; Wiley-VCH: Weinham, Germany. 2000. (24) Taylor, R. M.; Schwertmann, U.; Fechter, H. A rapid method for the formation of Fe(II), Fe(III) hydroxycarbonate. Clay Miner. 1985, 20, 147-151. (25) Gurman, S. J.; Binsted, N.; Ross, I. A rapid, exact curved-wave theory for EXAFS calculations. J. Phys. C. 1984, 17, 143-151. (26) Binsted, N. CLRC Daresbury Laboratory EXCURV98 program; CLRC Daresbury Laboratory: Warrington, UK, 1998. (27) Hedin, L.; Lundqvist, S. Effects of electron-electron and electronphonon interactions on the one-electron states of solids. Solid State Phys. 1969, 23, 1-181. (28) Wenzel, W. W.; Kirchbaumer, N.; Prohaska, T.; Stingeder, G.; Lombi, E.; Adriano, D. C. Arsenic fractionation in soils using an improved sequential extraction procedure. Anal. Chimica Acta. 2001, 436, 309-323. (29) O’Day, P. A.; Rivera, N.; Root, R.; Carroll, S. A. X. -ray absorption spectroscopic study of Fe reference compounds for the analysis of natural sediments. Am. Miner. 2004, 89, 572-585. (30) Kirk, G. J. D. The biogeochemistry of submerged soil; Wiley: Chichester, UK, 2004; p 291. (31) Conlin, T. S. S.; Crowder, A. A. Location of radial oxygen loss and zones of potential iron uptake in a grass and two non-grass emergent species. Can. J. Bot. 1989, 67, 717-722. (32) Wang, T. G.; Peverly, J. H. Iron oxidation states on root surfaces of a wetland plant (Phragmites australis). Soil Sci. Soc. Am. J. 1999, 63, 247-252. (33) Connell, E. L.; Colmer, T. D.; Walker, D. I. Radial oxygen loss from intact roots of Halophila ovalis as a function of distance behind the root tip and shoot illumination. Aquat. Bot. 1999, 63, 219-228. (34) Farquhar, M. L.; Charnock, J. M.; Livens, F. R.; Vaughan, D. J. Mechanisms of arsenic uptake from aqueous solution by interaction with goethite, lepidocrocite, mackinawite, and pyrite: An X-ray absorption spectroscopy study. Environ. Sci. Technol. 2002, 36, 1757-1762. (35) Sherman, D. M.; Randall, S. R. Surface complexation of arsenic (V) to iron (III) (hydr)oxides: Structural mechanism from ab VOL. 40, NO. 18, 2006 / ENVIRONMENTAL SCIENCE & TECHNOLOGY

9

5735

initio molecular geometries and EXAFS spectroscopy. Geochim. Cosmochim. Acta. 2003, 67, 4223-4230. (36) Williams, P. N.; Islam, M. R.; Adomako, E. E.; Raab, A.; Hossain, A. A.; Zhu, Y. G.; Feldmann, J.; Meharg, A. A. Increase in rice grain arsenic for regions of Bangladesh irrigating paddies with elevated arsenic in groundwaters. Environ. Sci. Tech. 2006, in press. (37) Meharg, A. A.; Hartley-Whitaker, J. Arsenic uptake and metabolism in arsenic resistant and non-resistant plant species. New Phytol. 2002, 154, 29-43.

5736

9

ENVIRONMENTAL SCIENCE & TECHNOLOGY / VOL. 40, NO. 18, 2006

(38) Schoof, R. A.; Yost, L. J.; Eickhoff, J.; Crecelius, E. A.; Cragin, D. W.; Meacher, D. M.; Menzel, D. B. A Market basket survey of inorganic arsenic in food. Food Chem. Toxicol. 1999, 37, 839846.

Received for review April 4, 2006. Revised manuscript received July 16, 2006. Accepted July 20, 2006. ES060800V