Accepted Manuscript

Realistic risk assessment of arsenic in rice

Randa A. Althobiti, Nausheen Sadiq, Diane Beauchemin

PII: S0308-8146(18)30431-X
DOI: https://doi.org/10.1016/j.foodchem.2018.03.015
Reference: FOCH 22558

To appear in: Food Chemistry

Received Date: 20 November 2017

Revised Date: 22 February 2018
Accepted Date: 6 March 2018

Please cite this article as: Althobiti, R.A., Sadiq, N., Beauchemin, D., Realistic risk assessment of arsenic in rice,
Food Chemistry (2018), doi: https://doi.org/10.1016/j.foodchem.2018.03.015

This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers
we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and
review of the resulting proof before it is published in its final form. Please note that during the production process
errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.
Realistic risk assessment of arsenic in rice

Randa A. ALTHOBITI, Nausheen SADIQ† and Diane BEAUCHEMIN‡*

Queen’s University, Department of Chemistry, 90 Bader Lane, Kingston, ON K7L

3N6, Canada

ABSTRACT

Over 3 billion people share a diet consisting mainly of rice, which may contain

significant amounts of arsenic. Because the toxicity of arsenic is dependent on its

chemical form and that it may be in a form that is not bio-accessible (i.e. dissolved in

the gastrointestinal tract) and can thus not become bio-available (i.e. end up in the

blood stream, where it may exert its toxic effect), the bio-accessibility of arsenic was

determined in thirteen different types of rice. The effects of washing and cooking were

also studied. The total concentration of arsenic ranged from 93 to 989 µg kg−1 and its

bio-accessibility ranged from 16 to 93%. Cooking only changed arsenic speciation in a

few cases. However, simply washing rice with arsenic-free water before cooking

removed 3-43% of the arsenic, resulting in all the rice tested except the most

contaminated one being safe to consume by adults.

Key words: bio-accessibility, rice, arsenic, speciation analysis, inductively coupled

plasma mass spectrometry

†
Current address: McGill University, Department of Food Science and Agricultural
Chemistry, 21111 Lakeshore, Ste Anne de Bellevue, QC H9X 3V9.
‡
E-mail: diane.beauchemin@chem.queensu.ca; telephone: 1-613-533-2619; fax: 1-613-
533-6669.

1
1. Introduction

Rice is a food consumed worldwide by over half of the world’s population

(Champagne, 2004; Mohanty, 2013). It is often recommended as the main form of

carbohydrate in the diet of those who have celiac disease (Green et al., 2015). Yet,

several varieties of rice have been reported to contain the most toxic form of arsenic

(As), i.e. inorganic As (iAs) (Meharg and Zhao, 2012; Mandal and Suzuki, 2002). The

Food and Agriculture Organization/World Health Organization’s provisional tolerable

daily intake (PTDI) of iAs in rice is 2.1 µg kg−1 of body weight per day (Joint

FAO/WHO Expert committee for Food additives, 1989). However, the European Food

Safety Authority recently indicated that the iAs regulation standard was no longer

suitable (Joint FAO/WHO Expert committee for Food additives, 2011; Meharg and

Raab, 2010). Indeed, a positive correlation between urinary As and the As level in

consumers' rice requires re-consideration of regulatory limits in order to minimize

public exposure to As through this route (Banerjee et al., 2013; Gilbert-Diamond et al.,

2011).

With well over 40000 different rice varieties (Meharg and Zhao, 2012; Mandal and

Suzuki, 2002), there is a need for a fast and simple risk assessment method. Ideally,

bio-availability, defined as the fraction reaching the blood stream (Van de Wiele et al.,

2007), should be measured. However, only the fraction dissolved in the gastro-

intestinal tract, called bio-accessibility, is available for absorption into the circulatory

system. Hence, in the worst case scenario, bio-availability (the fraction entering the

blood stream) is equal to bio-accessibility (Guerra et al., 2012). For realistic risk

assessment, speciation analysis of the bio-accessible fraction of elements whose

2
toxicity depends on their chemical form, such as As, should be done to determine the

portion that is actually toxic and to take into account any species transformation during

the gastro-intestinal digestion process. For instance, the hydrochloric acid in gastric

juice is a mild reducing agent. Yet, few of the past studies on rice (for example, Table

1, which contains a few examples out of the over 500 papers that have appeared on As

in rice) considered bio-accessibility and speciation analysis of the bio-accessible

fraction, despite the more realistic information that they would provide on the risk

present. Furthermore, except for previous work by this group (Horner and

Beauchemin, 2012, 2013), previous studies that considered bio-accessibility mostly

used a gastric phase in vitro batch extraction (Bolan et al., 2017, and references

therein). Yet, valuable information on the accessibility of As can be obtained when

saliva is included. It is part of the first step in the gastro-intestinal process.

An on-line leaching method was previously developed to mimic the gastro-

intestinal tract while providing results in a fraction of the time taken by batch methods

(Horner and Beauchemin, 2012, 2013). It involves sequential leaching of rice by

artificial saliva, gastric juice and intestinal juice maintained at body temperature to

determine the bio-accessible fraction, followed by speciation analysis of that fraction

by ion exchange chromatography (IEC) with detection by inductively coupled plasma

mass spectrometry (ICPMS) (Dufailly et al., 2008). Continuous exposure of the sample

to fresh reagent drives the dissolution equilibrium to the right, thus decreasing the

leaching time compared to that used with batch methods. This on-line method was

validated using rice flour certified reference material (Horner and Beauchemin, 2012).

However, it has so far only been applied to a bag each of brown rice and white rice

3
from California. Application to a greater variety of rice samples is required to verify

that similar results to those of batch methods are systematically obtained.

The safety of rice consumption depends on the rice variety (brown or white,

organic or not, etc.), purity of the water used for cooking, cooking conditions and

chemical forms of As present (Jitaru et al., 2016). For instance, brown rice (not

polished, with bran) may contain higher levels of both total As and iAs than white rice

(polished, without bran) (Yim et al., 2017; Zhu et al., 2008), the bran sometimes

containing as much As as the grain (Welna et al., 2015; Ma et al., 2014). Simply

washing rice prior to cooking was reported to remove 30% of As (Jitaru et al., 2016),

28% of As (Sengupta et al., 2006) or 13% of As (Horner and Beauchemin, 2013),

depending on the rice. Even the cooking step may convert some As(V) and

dimethylarsinic acid (DMA) to the more toxic As(III), with only a small change in the

portion of monomethylarsonic acid (MMA) (Horner and Beauchemin, 2013).

The goal of this study was to investigate the level of As and its bio-accessibility in

thirteen different types of rice, which have not been extensively studied before. The

study focused on white and some brown rice from ten Middle Eastern countries,

basmati rice from India, as well as organically grown white and brown rice from the

United States because they have never been examined in such a realistic risk

assessment. The total concentration and bio-accessibility of As were determined, and

speciation analysis of the bio-accessible As fraction released by saliva and gastric juice

was performed to determine the toxic portion of bio-accessible As. Bio-accessibility

was measured by both the above-mentioned continuous on-line leaching method and a

conventional batch method to verify that the on-line method systematically provided

similar results to those by the batch method but in a much shorter time. Another aim of

4
this work was to study the effect of washing rice with water at room temperature prior

to cooking on the toxicity level, as this would constitute a simple way for consumers to

protect themselves if it decreases the remaining bio-accessible As concentration to safe

levels.

2. Materials and methods

2.1. Instrumentation

Analyses were carried out using a Varian 820MS ICPMS instrument (Mulgrave,

Victoria, Australia) equipped with a MicroMist concentric nebulizer (Glass Expansion),

Peltier-cooled double-pass Scott spray chamber (at 0°C) and collision-reaction interface

(CRI). The Ar plasma, auxiliary, nebulizer and sheathing gases flow rates were

respectively 19, 1.8, 0.99 and 0.15 l min-1. Because gastric juice contains a high

chloride concentration, resulting in 40Ar35Cl+ spectroscopic interference on the only As

isotope, 80 ml min-1 H2 was added through the tip of the CRI skimmer cone to reduce

the interference without a great sacrifice in analyte sensitivity (Versantvoort et al.,

2005; Amiard et al., 2008; Moreda-Piñeiro et al., 2011). Data acquisition was done in

time-resolved mode with 3 points/peak, 80-ms dwell time, 1 scan/replicate and 0.025

a.m.u. spacing for on-line leaching and speciation analysis, and in steady-state mode

with 10-s integration time for batch leaching and the analysis of digests.

A previously optimized DX600/BioLC ion chromatography system equipped with

GS50 gradient pump (Dionex, Oakville, Canada), Ion Pac AG7 and AS7 (25-cm long,

4-mm diameter) (Dionex, Oakville, Canada) guard and analytical anion exchange

columns respectively (Horner and Beauchemin, 2012), was used for speciation analysis

of 25-µl aliquots of 5-fold diluted saliva leachates. A PEEK tubing (0.17-mm internal

5
diameter, 40-cm long) connected the analytical column to the ICPMS nebulizer. The

system was stabilized by pumping 0.5 mM HNO3 in 1% (v/v) methanol at 1.5 ml min-1

for 20 min prior to each series of chromatographic separations. This mobile phase

separated As(III), MMA and DMA within 3 min. Then, 50 mM HNO3 in 1% (v/v)

methanol (to increase As sensitivity) eluted As(V) in 2.5 min. Identification of As

species in leachates was based on retention time comparison with standard solutions of

the four As species. Saliva leachates were diluted 5-fold prior to analysis, whereas

gastric juice leachates were analysed without dilution (Dufailly et al., 2008).

2.2. Chemicals

For the determination of total As concentration, standard and internal standard

solutions were respectively prepared from 1000 mg l-1 As and In solutions (SCP

Science, Baie d'Urfé, Québec, Canada). For speciation analysis, 1000 mg l-1 solutions

of each chemical species were prepared from As(III) oxide (99.999%), As(V) oxide

(99.9%) (all Alfa Aesar, Ward Hill, USA), disodium methyl arsenate (MMA) (97.5%)

(ChemService, West Chester, USA), and cacodylic acid (DMA) (≥98%) (Sigma–

Aldrich, St. Louis, USA). They were diluted to 10 mg l-1 and stored at 4oC in the dark

to prevent decomposition or oxidation. For external calibration, 5 to 100 μg l-1 standard

solutions of the different species were prepared daily by appropriate dilution. For the

mobile phase, doubly deionized water (DDW) (18.2 MΩ cm-1) purified using an Arium

Pro UVDI water purification system (Sartorius Stedim Biotech, Göttingen, Germany),

sub-boiled HNO3 and methanol were used (ACS grade; Fisher Scientific, Ottawa,

Canada).

6
 Artificial saliva was prepared with 6.8 g of KH2PO4 (ACS grade; Fisher Scientific,

NJ, USA) and 77 ml of 0.2 M NaOH (ACS grade; BioShop, Burlington, Canada),

adjusting pH to 6.5 using 0.2 M NaOH and diluting to 1 l using DDW. For artificial

gastric juice, 2.0 g of NaCl (ACS grade; BioShop, Burlington Canada), 3.2 g of pepsin

(Sigma-Aldrich, Oakville, Canada) and 7.0 ml of sub-boiled HCl (ACS grade; Fisher

Scientific, Ottawa, Canada) were mixed and diluted to 1 l using DDW. For intestinal

fluid, 6.8 g of KH2PO4, 10 g of pancreatin (Sigma-Aldrich, St.Louis, USA) and 77 ml

of 0.2 M NaOH were mixed and diluted to 1 l using DDW, with pH adjusted to 6.8.

HNO3 and HCl were purified with a DST-1000 sub-boiling distillation system

(Savillex, Minnetonka, USA). For the digestion of rice and residues from the bio-

accessibility studies, sub-boiled HNO3 and H2O2 (J.T. Baker, Phillipsburg, USA) were

used.

2.3. Rice samples

One bag each of thirteen different types of rice was used for this work. Bags of rice

from Middle-Eastern countries were purchased at a grocery store in Saudi Arabia,

whereas organic rice and basmati rice were obtained at a natural food store in Kingston,

ON, Canada. These bags originated, and represent common brands. A raw rice portion

from each bag was ground with ceramic mortar and pestle. Another portion was

washed with DDW at room temperature (25ºC) before being placed in DDW (volume

ratio of 2:1 DDW:rice) in a glass beaker and brought to a boil, cooking it until all the

water was absorbed (usually 1 h). The cooked rice sample was then ground. All

samples were kept in Ziploc bags at room temperature (25ºC). Generally, five or six

replicates were analyzed in each case and the mean was used in calculations.

7
2.4. Continuous on-line leaching procedure

A mini-column was prepared as described in previous papers (Horner and

Beauchemin, 2012, 2013) by packing about 0.2 g of rice wrapped in quartz wool into a

PTFE tube (10-cm long, 3/15-in outer diameter, 1/8-in inner diameter), with a quartz

wool plug inserted at each end. Two quartz wool plugs in an empty tube served as a

blank mini-column. Artificial saliva, gastric juice, and intestinal fluid were sequentially

pumped for 5 min each through each mini-column that was directly connected to the

nebulizer of the ICPMS instrument because nothing was released from the samples

beyond 5 min. The leaching reagents and the mini-column were maintained at 37 ◦C

(human body temperature) using a thermostatically controlled water bath. For washed

and cooked rice, raw rice was first placed in a mini-column and DDW at room

temperature (25◦C) pumped through it until no As was detected. The washed rice was

then removed and placed in boiling DDW until all water was absorbed before being

ground and packed in a mini-column for sequential leaching by gastro-intestinal fluids

as described above.

2.5. Data processing and calibration with on-line leaching

The As signal versus time profiles were integrated to obtain peak area. The peak

area for the blank mini-column was then subtracted from the peak area for each

leachate. The same process was repeated for the external calibration that was

performed by flow injection of standard solutions and a blank prepared in each leaching

reagent, which were injected in order of increasing concentration through a 100-μl

injection loop attached to a universal automatic actuator (Anachem Ltd., Luton,

England). Linear regression analysis of the blank-subtracted peak areas of the

8
standards as a function of concentration was then performed to obtain the line of best

fit. The latter was finally used to convert the blank-subtracted peak areas obtained for

leachates into concentrations.

2.6. Batch method

For each replicate, 1 g of ground rice was placed in a 50-ml falcon tube with,

sequentially, saliva, gastric juice and intestinal fluid. Between each addition, the test

tube was shaken for 10 min for saliva and 2 h for other juices at 37 ◦C, followed by

centrifuging the samples and collecting the supernatant. The leaching time in saliva was

the middle of the time range (5-15 min) used by other studies that measured the bio-

accessibility of elements from food (Versantvoort et al., 2005; Amiard et al., 2008;

Moreda-Piñeiro et al., 2011). For washed cooked rice, a 5-min wash with DDW at

room temperature (25 ◦C), where the raw rice was shaken in DDW for 5 min, with

collection of the supernatant preceded the above gastro-intestinal treatment. All

supernatants were quantitatively analyzed by matrix-matched external calibration with

internal standardization using 5 µg l-1 In in 4% (v/v) HNO3 added through a Y-

connector.

2.7. Digestion of rice and of rice residues following leaching

Rice aliquots as well as residues left after on-line or batch leaching procedures were

digested with 2.5 ml of concentrated HNO3 and 0.5 ml of H2O2 in closed

perfluoroalkoxy digestion vessels (Savillex, Minnetonka, USA) on a hotplate at 50°C,

until the rice or residue was fully digested (around 1 h). After cooling to room

temperature (25◦C), around 20 ml DDW were added to each digest, which was then

9
filtered through Whatman 541 ashless filter paper into a polyethylene tube and diluted

to 25 ml with DDW. (The filtration step was only preventative, i.e. to ensure that

nothing could clog the micro-nebulizer; it may be omitted.) A blank was prepared

following the same procedure but without sample. Digests were analyzed by matrix-

matched external calibration with internal standardization using 5 µg l-1 In in 4% (v/v)

HNO3 added through a Y-connector. Digestion of the residues, which is not required

for risk assessment, was performed to verify mass balance for each method.

2.8. Comparison of data sets

The sum of bio-accessible and residue As concentrations was compared to the total

concentration measured after digestion of the rice itself. Similarly, the sum of As

species concentrations in the saliva leachate was compared to the total bio-accessible

concentration measured separately by leaching with saliva. In each case, an F test was

first done at the 95% confidence level to determine if there was a significant difference

in variance between the two data sets being compared. The appropriate Student`s t test

was then performed at the 95% confidence level to establish if there was a significant

difference between the results.

3. Results and discussion

3.1. Total concentration and bio-accessibility of arsenic

In contrast to previous studies that mostly involved batch gastric phase extraction,

both batch and on-line methods involving the sequential leaching of rice with artificial

saliva, gastric juice and intestinal juice were used to measure bio-accessibility. To

assess the worst case scenario, the recipe used for gastric juice represented fasting

10
conditions, when the hydrochloric acid concentration is the highest. Fig. 1 shows that

the total concentration of As and its bio-accessibility varied between rice samples.

Whereas many of them contained around 300 µg kg−1 As, some contained down to

around 100 µg kg−1 and one contained almost 1000 µg kg−1. All rice samples except

that from Iraq Tamn exceed the EU recommended value of 100 µg kg−1 As for young

children (Islam et al., 2017), with only the samples from Iraq Tamn, Saudi Arabia,

Morocco and Iran not exceeding the EU maximum level of 200 µg kg−1 for adults

(Islam et al., 2017).

In general, the bio-accessibility results obtained by both methods were similar (Fig.

1) despite the fact that the on-line method only takes 15 min (5 min per gastro-intestinal

reagent) versus over 4 h for the batch method. The drastically shortened time results

from continuously pumping fresh reagent through the sample, which drives the

dissolution equilibrium to the right. The on-line method was previously validated using

standard reference material SRM 1568a rice flour from the National Institute of

Standards and Technology (Horner and Beauchemin, 2012). The Iran rice batch results

in Fig. 1, where the bio-accessible fraction, excluding the portion left in the residue, is

greater than the total As concentration in the rice, also demonstrate another limitation

of the batch method: it is more susceptible to contamination because of the multiple

sample manipulation steps involved. In contrast, all the leaching is performed in a

closed system with the on-line method. Hence, the following discussion will focus on

the results obtained by the on-line method, for which mass balance was verified in all

cases, i.e. the sum of the concentration leached and that left in the residue was, within

error according to a Student’s t test at the 95% confidence level, the same as the total

concentration independently obtained following total digestion of the sample.

11
3.2. Effect of washing rice

For easier comparison, the bio-accessibility of As in rice from the Middle East is

expressed as a percentage of the total As concentration in Fig. 2. For raw unwashed

rice, 16-88% of the As was bio-accessible whereas for washed, cooked rice, the range

of bio-accessible As (excluding that released by washing with high-purity water) was

18-73% because water removed some of it. In fact, depending on the rice, water

washed away 3-43% of the As (Table 2). Unfortunately, the fraction of As released is

not commensurate with the total As concentration in the rice. For example, about 33%

of As was washed away from rice from Morocco, which contained the next-to-lowest

amount of As of the 13 rice samples analyzed, whereas 22% was removed from the Iraq

rice that contained the most As (nearly 1000 µg kg−1). The fraction of As removed by

washing is also not proportional to the percentage of bio-accessible As from raw

unwashed rice (Table 2).

Nonetheless, the amount of bio-accessible As left following a water wash (Table 2)

is sufficiently reduced so that all rice samples, except the one from Iraq, would now be

safe for adults according to the EU maximum value of 200 µg kg−1 (Islam et al., 2017).

In the case of young children, the rice from Lebanon, Saudi Arabia, Turkey, Sudan and

Morocco would now be safe in addition to that from Iraq Tamn. As there may be a

concern for the other rice samples, for instance most of the As left in the Iraq rice is

bio-accessible (Fig. 1), speciation analysis of the bio-accessible fraction was thus

carried out to determine the form(s) in which As is present.

3.3. Speciation analysis of arsenic in saliva leachates

12
 Given that saliva generally released As the most, speciation analysis of the saliva

leachates was systematically carried out, which revealed that As(V) was prevalent in

most cases (Fig. 3). Moreover, cooking generally had no effect on As speciation,

except in three cases. For the rice from Iraq, some As(V) was converted to the less

toxic MA. However, not enough was converted to make the rice safe, even for adults.

For the rice from Egypt, because most of the iAs left after washing the rice was

converted to the less toxic DMA following cooking, this rice can be added to the list of

rice samples that are safe for consumption by either adults or young children. For the

rice from Lebanon, where less toxic species were converted into the most toxic As(III),

the rice is still safe for both adults and young children, because the bio-accessibility of

As from this rice is the lowest of all rice samples tested (Table 2).

To better assess the effect of cooking, some rice samples were cooked without first

being washed, and speciation analysis of both the saliva and gastric juice leachates was

carried out for a more complete picture. Figure 4 confirms what was observed with the

Middle East rice samples: depending on the rice, cooking had no significant effect on

bio-accessible species (Arkansas rice), converted some iAs to the less toxic DMA

(California rice) or converted some of the less toxic DMA into the more toxic iAs

(basmati rice). While these rice samples are safe for adults, they should not be

regularly given to young children, as the total amount of bio-accessible iAs exceeds

100 µg kg−1 (Islam et al., 2017).

4. Conclusions

13
 Both the total concentration of As and its bio-accessibility vary greatly depending

on the rice and its origin. Speciation analysis of the bio-accessible fraction of As is

important for realistic risk assessment. In all cases, simply washing rice prior to

cooking made the rice safer to eat. It may be possible to further decrease the fraction of

bio-accessible As left in rice after washing by cooking rice in excess water and then

discarding the excess water, in addition to washing the rice with As-free water prior to

cooking (Zhu et al, 2008). For example, washing rice until the water is clear, cooking it

in a 1:6 rice:water ratio and then discarding the excess water could remove 57% of As

(Sengupta et al., 2006). This approach would likely make the rice samples from

California, Arkansas, India, Yemen and Iran safe to eat on a regular basis by young

children. Future work will study the effect of the cooking method (such as sautéing

before adding water) on the bio-accessibility of As and its speciation.

Acknowledgements

The authors gratefully acknowledge the financial support of the Saudi Arabia

government and of the Natural Sciences and Engineering Research Council of Canada

(grant number 39487-2013). NWS is grateful for a Marie Mottashed scholarship and

graduate awards from Queen’s School of Graduate Studies and Research.

References

Amiard, J.-C., Amiard-Triquet, C., Charbonnier, L., Mesnil, A., Rainbow, P.S., & Wan,

W.-X. (2008). Bioaccessibility of essential and non-essential metals in

commercial shellfish from Western Europe and Asia. Food and Chemical

Toxicology, 46, 2010–2022.

14
Banerjee, M., Banerjee, N., Bhattacharjee, P., Mondal, D., Lythgoe, P.R., Martínez, M.

Pan, J., Polya, D.A., & Giri, A.K. (2013). High Arsenic in Rice is Associated with

Elevated Genotoxic Effects in Humans. Scientific Reports, 3, 2195.

Bolan, S., Kunhikrishnan, A., Chowdhury, S., Seshadri, B., Naidu, R., & Ok, Y.S.

(2017). Comparative analysis of speciation and bioacessibility of arsenic in rice

grains and complementary medicines, Chemosphere, 182, 433-440.

Champagne, E., Ed. (2004). Rice: Chemistry and Technology. American Association of

Cereal Chemists, St. Paul, Minnesota, USA, ed. 3.

Dufailly, V., Guérin, T., Noël, L., Frémy, J.-M., & Beauchemin, D. (2008). A simple

method for the speciation analysis of bio-accessible arsenic in seafood using on-

line continuous leaching and ion exchange chromatography coupled to

inductively coupled plasma mass spectrometry. Journal of Analytical Atomic

Spectrometry, 23, 1263-1268.

Fu, Q.L., Li, L., Achal, V., Jiao, A.Y., & Liu, Y. (2015). Concentrations of heavy metals

and arsenic in market rice grain and their potential health risks to the population

of Fuzhou, China. Human and Ecological Risk Assessment, 21, 117-128.

Gilbert-Diamond, D., Cottingham, K.L., Gruber, J.F., Punshon, T., Sayarath, V.,

Gandolfi, A.J., Baker, E.R., Jackson, B.P., Folt, C.L., & Karagas, M.R. (2011).

Rice Consumption Contributes to Arsenic Exposure in US Women. Proceedings

of the National Academy of Science, U. S. A., 108, 20656–20660.

Green, P.H.R., Lebwohl, B., & Greywoode, R. (2015). Celiac disease. The Journal of

Allergy and Clinical Immunology, 135, 1099-1106.

15
Guerra, A., Etienne-Mesmin, L., Livrelli, V., Denis, S., Blanquet-Diot, S., & Alric, M.

(2012). Relevance and challenges in modeling human gastric and small intestinal

digestion. Trends in Biotechnology, 30, 591-600.

Heikens, A. (2006). Arsenic contamination of irrigation water, soil and crops in

Bangladesh: risk implications for sustainable agriculture and food safety in Asia.

RAP Publication, 20, 1-38.

Horner, N.S., & Beauchemin, D. (2012). A Simple Method Using on-Line Continuous

Leaching and Ion Exchange Chromatography Coupled to Inductively Coupled

Plasma Mass Spectrometry for the Speciation Analysis of Bio-Accessible Arsenic

in Rice. Analytica Chimica Acta, 717, 1–6.

Horner, N.S., & Beauchemin, D. (2013). The Effect of Cooking and Washing Rice on

the Bio-Accessibility of As, Cu, Fe, V and Zn Using an on-Line Continuous

Leaching Method. Analytica Chimica Acta, 758, 28–35.

Huq, S.M.I., Joardar, J.C., Parvin, S., Correll, R., & Naidu, R. (2006). Arsenic

contamination in food-chain: transfer of arsenic into food materials through

groundwater irrigation. Journal of Health Population and Nutrition, 24, 305-316.

Islam, S., Rahman, M.M., Rahman, M.A., & Naidu, R. (2017). Inorganic arsenic in rice

and rice-based diets: Health risk assessment. Food Control, 82, 196-202.

Jitaru, P., Millour, S., Roman, M., El Koulali, K., Noël, L., & Guérin, T. ( 2016).

Exposure assessment of arsenic speciation in different rice types depending on

the cooking mode. Journal of Food Compostion and Analysis, 54, 37-47.

Joint FAO/WHO Expert committee for Food additives (1989). Evaluation of Certain

Food Additives and Contaminants. WHO Food Additives Series, 45–94.

16
Joint FAO/WHO Expert committee for Food additives (2011). Evaluation of Certain

Contaminants in Food, WHO Technical Report Series, No. 959, 1–105.

Lan, D.Z., Lei, M., Zhou, S., Liao, B.H., Cui, Y.S., Yin, N.Y., & Shen, Y. (2014). Health

risk assessment of heavy metals in rice grains from a mining- impacted area in

south hunan by in vitro simulation method. Journal of Agro-Environment

Science, 33, 1897-1903.

Ma, R., Shen, J., Wu, J., Tang, Z., Shen, Q., & Zhao, F.J. (2014). Impact of Agronomic

Practices on Arsenic Accumulation and Speciation in Rice Grain. Environmental

Pollution, 194, 217–223.

Mandal, B.K., & Suzuki, K.T. (2002). Arsenic round the world: a review. Talanta, 58,

201-235.

Meharg, A.A., & Raab, A. (2010). Getting to the Bottom of Arsenic Standards and

Guidelines. Environmental Science & Technology, 44, 4395–4399.

Meharg, A.A., & Zhao, F.-J. (2012). Arsenic & Rice. Springer, Dordrecht, Netherlands.

Mohanty, S. (2013). Trends in Global Rice Consumption. Rice Facts, 12, 44-45.

Moreda-Piñeiro, J., Moreda-Piñeiro, A., Romarís-Hortas, V., Moscoso-Perez, C.,

Lopez-Mahia, P., Muniategui-Lorenzo, S., Bermejo-Barrera, P., & Prada-

Rodríguez, D. (2011). In-vivo and in-vitro testing to assess the bioaccessibility

and the bioavailability of arsenic, selenium and mercury species in food samples.

Trends in Analytical Chemistry, 30, 324-345.

Seguar, F.R., Souza, J.M.O., Paula, E.S., Martins, A.C., Paulelli, A.C.C., Barbosa, F., &

Batista, B.L. (2016). Arsenic speciation in brazilian rice grains organically and

traditionally cultivated: is there any difference in arsenic content? Food Research

International, 89, 169-176.

17
Sengupta, M.K., Hossain, M. A., Mukherjee, A., Ahamed, S., Das, B., Nayak, B., Pal,

A., & Chakraborti, D. (2006). Arsenic burden of cooked rice: Traditional and

modern methods. Food and Chemical Toxicology, 44, 1823–1829.

Van de Wiele, T., Oomen, A., Wragg, J., Cave, M., Minekus, M. Hack, A., Cornelis, C.,

Rompelberg, C., Zwart, L., Klinck, B., Wijen, J., Verstraete, W., & Sips, A.

(2007). Comparison of five in vitro digestion models to in vivo experimental

results: Lead bio-accessibility in the human gastrointestinal tract. Journal of

Environmental Science and Health, Part A, 42, 1203-1211.

Versantvoort, C.H.M., Oomen, A.G., Van de Kamp, E., Rompelberg, C.J.M., & Sips,

A.J.A.M. (2005). Applicability of an in vitro digestion model in assessing the

bioaccessibility of mycotoxins from food. Food and Chemical Toxicology, 43,

31–40.

Welna, M., Szymczycha-Madeja, A., & Pohl, P. (2015). Comparison of Strategies for

Sample Preparation prior to Spectrometric Measurements for Determination and

Speciation of Arsenic in Rice. Trends in Analytical Chemistry, 65, 122–136.

Yim, S.R., Park, G.Y., Lee, K.W., Chung, M.-S., & Shim, S.-M. (2017).

Determination of total arsenic content and arsenic speciation in different types of

rice. Food Science and Biotechnology, 26, 293-298.

Zhu, Y.G., Williams, P.N., & Meharg, A.A. (2008). Exposure to Inorganic Arsenic from

Rice: A Global Health Issue? Environmental Pollution, 15, 169–171.

18
 Table 1. Selected studies previously conducted on arsenic in rice
Type of rice Country Extractant Analysis Observations Ref.
technique*
White rice USA Artificial IEC-ICPMS Over 90% of As Horner and
saliva was bio- Beauchemin,
accessible; 2012
cooking rice
converted As(V)
to As(III)
White and USA Water, IEC-ICPMS Washing rice Horner and
brown rice artificial saliva removed much As Beauchemin,
and gastric 2013
juice
16 rice China ICPMS The As level was Fu et al.,
samples below Chinese 2015
threshold
standards
Rice from a China Simulated GFAFS Contaminated Lan el al.,
mining gastric and areas had 3.8 vs 2014
region intestinal juice 0.3 mg kg-1 As,
which exceeded
the allowance in a
child’s diet
Various rice Spain, Trifluoroacetic IEC-ICPMS The total As Heikens,
samples Italy, USA, acid concentration 2006
India, ranged from 0.03-
China, 0.4 mg kg-1, of
Thailand, which 11-90%
Bangladesh was iAs
Rice Bangladesh Hydride As content can be Huq et al.,
generation as high as 1.7 mg 2006
AAS kg-1
Organic and Brazil 0.28 M HNO3 IEC-ICPMS The level of iAs Seguar et al.,
regular rice at 95°C for 2 h was higher in 2016
organic polished
rice than
conventional
polished rice
Rice grown Bangladesh Gastric phase ICPMS As bio- Bolan et al.,
with , Korea and in vitro batch accessibility 2017
irrigation India extraction ranged from 23 to
water
32%
containing
Na2HAsO4·7
H2O

19
Brown and South 5 mM IEC- Total As and Yim et al.,
white rice Korea malonic acid ICPMS iAs were higher 2017
(pH 5.6) in brown rice
As(III)
predominant in
brown and white
rice
* IEC-ICPMS= ion exchange chromatography coupled to inductively coupled plasma
mass spectrometry; AAS= atomic absorption spectrophotometry; GFAFS= graphite
furnace atomic fluorescence spectrometry.

20
 Table 2. Average percentage (± standard deviation, n=5-6) of bio-accessible and
washable As for rice from various countries (one bag of rice per country) along with
the average amount (± standard deviation, n=5-6) of bio-accessible As left after rice
washing and the effect of cooking rice on As speciation

Country City or Fraction of Fraction of Amount of Type Change

of region of bio- As bio-accessible of rice of As
origin origin accessible removed As left after and speciation
As from by washing washing grain upon
raw rice raw rice (µg kg-1) cooking
(%) (%)
Iraq Mosul 73 ± 13 22.1 ± 9.1 493 ± 17 To less
toxic
Iraq Alnajaf 56.0 ± 3.6 24.8 ± 9.9 41 ± 17 No
Tamn
Lebanon Brucke 16.0 ± 5.3 6.5 ± 4.6 59.0 ± 3.7 To more
White toxic
Yemen Qishn 60.7 ± 5.2 19.5 ± 2.8 120 ± 10 long No
Turkey Samsun 43.5 ± 1.9 30.7 ± 6.9 97.0 ± 8.2 No
U.S.A. California 93 ± 14 42.9 ± 7.8 130 ± 11 To less
toxic
India 81 ± 13 36 ± 16 160 ± 22 To more
toxic
Saudi Al Hasa 56.4 ± 8.7 29.2 ± 5.6 49.0 ± 4.9
Arabia Brown
No
U.S.A. Arkansas 86 ± 16 41.4 ± 7.5 140 ± 14 long
Sudan Aweil 43.4 ± 7.2 13.9 ± 4.6 94.0 ± 2.3
Iran Rasht 88.4 ± 5.0 15 ± 11 130.0 ± 5.4 No
Morocco Tangier 80 ± 19 34 ± 19 17.5 ± 2.3 White No
Egypt Giza 54.7 ± 9.2 3.3 ± 2.0 100.7 ± 3.4 short To less
toxic

21
 1000
Intestinal juice Gastric juice Saliva Total concentration
Concentration (µg kg-1)

800

600

400

200

0
On-line

On-line

On-line

On-line

On-line

On-line

On-line

On-line

On-line

On-line

On-line

On-line
On-line

Batch
Batch

Batch

Batch

Batch

Batch

Batch

Batch

Batch

Batch

Batch

Batch

Batch
Iraq Egypt Lebanon Saudi Morocco Sudan Turkey Yemen Iran Iraq California Arkansas India
Arabia Tamn
Source of rice
Fig. 1. Comparison of the bio-accessible concentration of As (leached by different
artificial gastro-intestinal fluids) in raw rice from various countries obtained by on-line
and batch leaching methods with the total concentration of As measured following
digestion of raw rice. Error bars are standard deviations for 5 or 6 replicate analyses.

22
 100 Intestinal juice Gastric juice Saliva Water
90
Fraction of As released (%)
80
70
60
50
40
30
20
10
0
Raw unwashed

Raw unwashed

Raw unwashed

Raw unwashed

Raw unwashed

Raw unwashed

Raw unwashed

Raw unwashed

Raw unwashed
Raw unwashed

Washed, cooked
Washed, cooked

Washed, cooked

Washed, cooked

Washed, cooked

Washed, cooked

Washed, cooked

Washed, cooked

Washed, cooked

Washed, cooked
Iraq Egypt Lebanon Saudi Morocco Sudan Turkey Yemen Iran Iraq
Arabia Tamn
Source of rice

Fig. 2. Comparison of the fraction of As released from raw unwashed rice and washed,
cooked rice from various Middle-Eastern countries. Error bars are standard deviations
for 5 replicate analyses.

23
 350 DMA As(V) MMA As(III)
As species concentration (µg kg-1)

300
250
200
150
100
50
0
Washed, cooked
Raw unwashed

Raw unwashed

Raw unwashed

Raw unwashed

Raw unwashed

Raw unwashed

Raw unwashed

Raw unwashed

Raw unwashed
Washed, cooked

Washed, cooked

Washed, cooked

Washed, cooked

Washed, cooked
Raw unwashed
Washed, cooked

Washed, cooked

Washed, cooked

Washed, cooked
Iraq Egypt Lebanon Saudi Morocco Sudan Turkey Yemen Iran Iraq
Arabia Tamn
Source of rice

Fig. 3. Speciation of arsenic in the saliva leachate of raw unwashed rice and washed,
cooked rice from various countries. Error bars are standard deviations for 5 replicate
analyses.

24
 300
As(III) MMA DMA As(V)

250
As concentration (µg kg-1)

200

150

100

50

0
saliva gastric saliva gastric saliva gastric saliva gastric saliva gastric saliva gastric
juice juice juice juice juice juice
California raw rice California cooked Arkansas raw rice Arkansas cooked India raw rice India cooked rice
rice rice
Source of rice

Fig. 4. Comparison of the concentrations of bio-accessible As species released by

artificial saliva and gastric juice from three different unwashed rice samples, before and
after cooking. Error bars are standard deviations for 6 replicate analyses.

25
Highlights

 Arsenic’s total concentration and bio-accessibility depend on rice type and

origin.
 Washing rice with arsenic-free water before cooking removes 3-43% of arsenic.
 Cooking may have no effect, decrease or increase the toxicity of arsenic
species.
 On-line leaching measures bio-accessibility in 15 min instead of several hours.
 Over 100 µg kg-1 bio-accessible inorganic arsenic can remain after washing
rice.

26