Evaluation of methods for assessing the oral bioavailability of inorganic mercury in soil.

The risks associated with environmental exposures to inorganic mercury are typically assessed based on toxicity studies conducted with the soluble salt, mercuric chloride (HgCl2). Evidence indicates, however, that inorganic mercury is present in soil as a variety of compounds and that oral absorption of inorganic mercury decreases with a decrease in the solubility of the mercury compound being studied. Thus, while HgCl2 is approximately 15-20% bioavailable, the bioavailability of cinnabar (HgS) may be 30- to 60-fold less. The solubility and, hence, bioavailability of inorganic mercury in soil is expected to be substantially less than that of HgCl2 due to the presence of less soluble compounds and their interactions with soil constituents. Quantification of this difference in bioavailability is important in assessing potential risks associated with exposure to mercury-containing soil. A review of available studies supports the expectation that mercury bioavailability in soils will be reduced. This paper reviews methods for assessing soil metal absorption with consideration of the characteristics of the oral absorption of elemental and inorganic mercury that should be evaluated in designing additional studies. Because of the very slow elimination of mercury in some species, it is recommended that a repeated-dose study be conducted. Such a study would yield an estimate of relative bioavailability based on a comparison of tissue mercury concentrations in animals ingesting soil with those of animals receiving HgCl2. The dose, age, gender, and species of animal selected are not expected to affect relative bioavailability estimates; however, it is recommended that studies be conducted in two animal species. Rats should be used because they have been used in many studies of mercury absorption and toxicity. A species of large animals such as monkeys, swine, or dogs should also be used to provide confirmation in a species with greater similarities to humans in gastrointestinal physiology and anatomy. Other critical factors in designing these studies, such as selection and characterization of soil samples, are also addressed.