Yamily J. Zavala

High levels of arsenic (As) in rice grain are a potential concern for human health. Variability in total As in rice was evaluated using 204 commercial rice samples purchased mostly in retail stores in upstate New York and supplemented with samples from Canada, France, Venezuela, and other countries. Total As concentration in rice varied from 0.005 to 0.710 mg kg⁻¹. We combined our data set with literature values to derive a global “normal” range of 0.08−0.20 mg kg⁻¹ for As concentration in rice. The mean As concentrations for rice from the U.S. and Europe (both 0.198 mg kg⁻¹) were statistically similar and significantly higher than rice from Asia (0.07 mg kg⁻¹). Using two large data sets from Bangladesh, we showed that As contaminated irrigation water, but not soil, led to increased grain As concentration. Wide variability found in U.S. rice grain was primarily influenced by region of growth rather than commercial type, with rice grown in Texas and Arkansas having significantly higher mean As concentrations than that from California (0.258 and 0.190 versus 0.133 mg kg⁻¹). Rice from one Texas distributor was especially high, with 75% of the samples above the global “normal” range, suggesting production in an As contaminated environment.

Rice is a potentially important route of human exposure to arsenic, especially in populations with rice-based diets. However, arsenic toxicity varies greatly with species. The initial purpose of the present study was to evaluate arsenic speciation in U.S. rice. Twenty-four samples containing high levels of arsenic and produced in different regions of the U.S were selected from a previous market-basket survey. Arsenite and dimethyl arsinic acid (DMA) were the major species detected. DMA increased linearly with increasing total As but arsenite remained fairly constant at ∼0.1 mg kg⁻¹, showing that rice high in As was dominated by DMA. A similar result was obtained when our data was combined with other published speciation studies for U.S. rice. However, when all published speciation data for rice was analyzed a second population dominated by inorganic As and lower levels of DMA was found. We thus categorized rice into DMA and Inorganic As types. Rice from the U.S. was predominantly the DMA type, as were single samples from Australia and China, whereas rice from Asia and Europe was the Inorganic As type. We suggest that methylation of As occurs within rice and that genetic differences lead to the two rice types. Insufficient understanding of DMA toxicity precludes a firm assessment of the relative health risks associated with the two rice types but, based on current knowledge, we suggest that the DMA rice type is likely to be less of a health risk than the Inorganic As rice type and, on this basis, rice from the U.S. may be safer than rice from Asia and Europe.