Chapter 1 Advances in Assessing Bioavailability of Metal(Loid)s in Contaminated Soils

Abstract

The term bioavailability has many different meanings across various disciplines of toxicology and pharmacology. Often bioavailability is concerned with human health aspects such in the case of lead (Pb) ingestion by children. However, some of the most contaminated sites are found in nonpublic access facilities (Department of Defense or Energy) or in remote regions as a result of mining or industrial practices in which ecoreceptors such as plants, animals, and soil organisms are the primary concerns as well as the potential for food-chain transfer. In all cases, the endpoint requires movement of the element across a biological barrier. The still utilized approach to base risk assessment on total metal content in soils is an outdated endeavor and has never been proved to be scientifically sound. Yet to reverse this trend, much work is required to establish baseline bioavailability measurements and to develop complementary methods that are capable of predicting bioavailability across a whole range of impacted media in a cost-efficient manner. Thus, regulators have recognized site-specific human health risk assessments play a key role in decision-making processes at contaminated sites.

Bioavailability issues surrounding metal-contaminated soils and media have been an area of intense research. For obvious ethical reasons, we cannot solicit humans, in particular the sensitive population of children, from the general population for experimental purposes to examine the long-term harmful effect of metals in soils. However, some adult human feeding studies have been accomplished under tight medical supervision and with very small doses. One option to understand and relate bioavailability in humans is to employ animal surrogates; however, the physiology of most animals is different than that of humans but good correlations have been achieved despite the dose–response paradigm not being identical. The biggest drawback of in vivo studies to examine metal bioavailability to an appropriate ecoreceptor, be it human, plant, or soil organism, is the tremendous cost and time involved relative to chemical and physical surrogates. Chemical surrogate methods generally only require knowledge of the total metal content so that a percent bioaccessible number can be generated from in vitro extractions that simulate digestive systems or mimic responses to sensitive ecoreceptors. However, there is not a consensus as to which of the many in vitro methods is the best analogy to an ecoreceptor uptake and the same can be said for in vivo animal models to mimic human response as well. Further, there is yet to be a single in vitro method that can account for more than a few elements for a specific exposure pathway (e.g., Pb and/or arsenic (As) for human health). These in vitro tests require honest and accurate validation against in vivo bioavailability measurements, but most of all would benefit from metal speciation methods to identify the forms of metals allowing their release. Adaptation of spectroscopic speciation techniques to identify metal(loid) phases is extremely beneficial in bioavailability research to understand the variability of biologically available metal uptake, to manipulate the ecosystem to reduce bioavailability via in situ amendments, to monitor the long-term stability of elements to ensure bioavailability indicators do not change over time, and to develop comprehensive predictive models based on speciation.

Introduction

In most cases, the toxicity of contaminants depends on how much of it is absorbed into the body or taken up by plants. For soil contaminants where human exposure is by ingestion of soil or plants and organisms produced on the soil, toxicity depends on absorption into the gastrointestinal (GI) system. Information on how well a contaminant is absorbed into the GI system is important to determining how much of a contaminant humans can be exposed to before health effects occur. Because typical health effect dose–response assessments (and resulting oral reference doses (RfDs) and cancer slope factors (CSFs)) are generally expressed in terms of ingested dose (rather than absorbed dose into the organism), accounting for potential differences in absorption between different exposure media can be important to site risk assessment (USEPA, 1989). Thus, if the oral RfD for a particular metal is based on bioavailability studies in water, risks from ingestion of the metal in soil or plant produced on the soil might be (likely is) overestimated. Minor adjustments in oral bioavailability based on nonrelevant exposure pathways can have significant impacts on estimated risks and cleanup goals for hazardous waste sites (USEPA, 1989).

It is increasingly recognized that the response of an at-risk population is not controlled by the total metal concentration, but instead is controlled by only the biologically available portion, which is dependent on the route of exposure, the pharmacokinetics of the organism, and the speciation of the contaminant.

In spite of the earlier understanding, the complexity of metal-contaminated sites has and continues to be simplified to a measure of the total metal content. Regulations on the fate and effects of metals in the environment based solely on total concentrations are no longer (perhaps never has been) valid, state-of-the-art, or scientifically sound. A vast amount of knowledge clearly illustrates the decisive role of metal speciation when metal bioavailability and phytoavailability in the environment have to be assessed (McNear et al., 2007, Ryan et al., 2004). While total metal content is a critical regulatory measure in assessing risk of a contaminated site, total metal content alone does not provide predictive insights on the bioavailability, mobility, and fate of the metal contaminants. Thus, a better understanding of the nature of the chemical and physical interactions of contaminants with soil constituents can increase the scientific understanding and lead to regulatory and public confidence in the use of bioavailability adjustments. Predictions of long-term stability rely on a mechanistic understanding of how contaminants are stored or sequestered within the soil.

Bioavailability processes are defined as the individual physical, chemical, and biological interactions that determine the transfer of chemicals associated with soils to plants and animals. Bioavailability processes are embedded within existing human health and ecological risk frameworks to reduce uncertainty in exposure estimates and improve risk assessment (USEPA, 2007b). In both ecological and human health risk assessment, bioavailability is usually reflected in default values or site-specific data that are inserted into exposure equations. Although a multitude of processes can affect bioavailability, a typical risk assessment generates one value that is used to adjust the applied dose. For this reason, many bioavailability processes are hidden within risk assessment, and assumptions made about these processes are sometimes not clear. Although long employed in toxicology and agricultural sciences, the concept of bioavailability has recently sparked the interest of the hazardous waste industry as an important consideration in deciding how much waste to clean up. This interest stems from observations that some contaminants in soils appear to be less available to cause harm to humans and ecological receptors than is suggested by their total concentration, such that cleanup levels expressed as total concentrations poorly correlate with actual risk. Correct characterization of bioavailability in contaminated soils and sediments may indicate that greater levels of contamination can be left untouched without increased risks, thus, reducing cleanup costs and reducing volumes of contaminated media requiring intrusive remedial options (USEPA, 2007c). However, in order to pursue this concept in risk assessment critical knowledge of bioavailability processes and spectroscopic speciation techniques are required to develop a mechanistic understanding of the bioavailability processes to improve the science of risk assessment to develop predictive models derived from sound research. Further, chemical, environmental, and regulatory factors must align in considering bioavailability processes that influence risk-based decision-making (NRC, 2003).

Because the fraction of a soil element which can actually be absorbed by an organism to cause harm depends on the chemical forms present and physical/chemical properties of the soil, in both risk assessment and remediation evaluation, the fraction of a soil element which can actually cause harm must be identified. This fraction is ultimately defined as the bioavailable fraction, and because measurement of the bioavailable fraction is time-consuming and expensive via in vivo animal feeding studies, in vitro chemical methods are being developed to estimate the bioavailable fraction. In the case of ingestion of soil, the in vitro or chemical estimation method has been labeled “bioaccessible” (to avoid confusion with “bioavailable”) and is a measure of the amount of metal that can be liberated from the soil matrix, thus not a measure of the amount of metal that moves across the GI epithelium to harm internal target tissues and organs. Extensive progress has been made in development of soil Pb and As bioavailability testing in conjunction with bioaccessibility methods (Drexler and Brattin, 2007, Rodriguez et al., 2003, Ruby et al., 1993, Ruby et al., 1996). Additionally, great effort has been wasted in planning inconsequential research efforts to develop bioaccessibility methods, which try to match all digestion processes without a valid bioavailability endpoint as a comparison. In the end, an in vitro bioaccessibility method only needs to be well correlated with an acceptable in vivo bioavailability model. Actually, the simpler and less expensive the bioaccessibility method can be made, the better, as long as the correlation with bioavailability is high. Further, it is necessary for the tests to be reproducible in laboratories across the globe, which has not been the case for many of the bioaccessibility methods available today. Further, for such methods to be relevant to testing of remediation methods, changes in bioavailability due to field treatments should be reflected in the bioaccessibility test results. In the case of soil Pb, in situ remediation using phosphate and other treatments have been proved to reduce bioavailability to pigs, rats, and humans, but the bioaccessibility test conducted at pH 1.5 does not measure this 69% reduction in bioavailability to human adults while testing at pH 2.2 or 2.5 does reflect the effectiveness of the soil treatment (Ryan et al., 2004). Other simple chemical tests have been shown to suffer significant flaws in that the extraction causes changes in chemical speciation during the test, and have not been shown to correlate with bioavailability changes due to soil treatments. Further, it is necessary to have a valid measure of why the bioavailability or bioaccessibility of samples are different and whether the changes are persistent; thus, the need for metal speciation.

For sensitive ecoreceptors (plants, animals, and soil organisms), where testing with the organism to be protected is more readily conducted, chemical methods have been developed which integrate potential toxicity across soil properties including pH which often strongly affects bioavailability. Mild neutral salt extractions (similar to the first extraction step of a sequential extraction procedure) are often found to be effective methods. However, assessment of potential toxicity by adding metal salts to uncontaminated soils substantially fails to mimic field contaminated soils because elements react with soils, and metal salt additions alter soil pH and do not account for the aging effect of metals in soils. Traditional toxicology approaches of adding element salts and immediately measuring toxicity are clearly inappropriate, and can cause serious artifacts due to pH change resulting from the metal salt addition, or formation of soluble metal complexes which temporarily increase or decrease element bioavailability. Thus, testing of potential toxicity has as many problems as testing of bioaccessibility. It seems clear that by taking present knowledge into account, effective toxicity testing, bioaccessibility evaluation, and risk assessment can provide massive savings to the public in dealing with contaminated soils.

The extent to which metals are bioavailable has significant implications on human and ecological health following exposure and on potential remediation of contaminated sites. Characterization via speciation of insoluble metal phases in contaminated soils and sediments may indicate that greater levels of contamination can be left untouched without increased risks, thus, driving reduced cleanup costs and limited volumes of contaminated media through less intrusive remedial options. A mechanistic understanding of the bioavailability process in relation to metal speciation will allow development of predictive models and improvement of risk assessment. Further, chemical, environmental, and regulatory factors must align in considering bioavailability processes that influence risk-based decision-making (NRC, 2003).

In both ecological and human health risk assessment, bioavailability is usually reflected in default values or site-specific data that are inserted into exposure equations. Although a multitude of processes can affect bioavailability, a typical bioavailability assessment generates one value that is used to adjust the applied dose. The Risk Assessment Guidance for Superfund, Volume I: Human Health Evaluation (Part A) (RAGS) (USEPA, 1989) supports the consideration of bioavailability in the determination of site-specific human health and environmental risks. This guidance has been used to support bioavailability adjustments across different routes of exposure at contaminated sites. However, the use of bioavailability information in site-specific risk assessment has not been widespread (due to limited data, uncertain methodologies, and lack of method validation). The primary impediment to the broad use of bioavailability data in risk assessment and decision-making is the absence of rapid and inexpensive tools that can generate reliable relative bioavailability (RBA) estimates in the receptors of concern. It is in this context that coupling in vivo bioavailability, in vitro bioaccessibility, and speciation research can fill many data gaps to aid in understanding and predicting bioavailability.

The speciation, or chemical form, of metals governs their fate, toxicity, mobility, and bioavailability in contaminated soils, sediments, and water. Different chemical forms of metals, for example, can differ greatly in the amounts taken up by organisms. The varying bioavailability values of different metal species is a large reason for the wide range of bioaccessibility values measured using standardized in vitro analyses of different soils. Other interactions between metals and soil components also govern speciation and affect bioavailability. The influence of the soil matrix on metal(loid) availability is in constant dynamic equilibria with multiple independent variables such as solid mineral phases, exchangeable ions and surface adsorption, nutrient uptake by plants, soil air, organic matter, and microorganisms, and water flux.

However, determining speciation is not a trivial task, particularly at low concentrations in a complex matrix such as soil. To assess these chemical properties and to accurately gauge their impact on human health and the environment we need to characterize metals at the atomic level with spectroscopic techniques. This research must move beyond operationally defined sequential extraction methods and utilize analytical instruments that are capable of identifying metal species (D'Amore et al., 2005) Researchers have used advanced synchrotron radiation methods to elucidate the true, in situ speciation of metal contaminants. Synchrotron techniques include X-ray absorption near-edge spectroscopy (XANES), which identifies the oxidation state and first coordination shell and X-ray absorption fine structure (XAFS) spectroscopy provides information on coordination environment of a selected element as well as interatomic bond distances and identity of nearest-neighboring atoms to determine speciation. These methods can also be used in conjunction with statistical methods (principal component analysis and linear combination fitting) to determine chemical phases via a finger printing process with a library of known reference standards. Although most soil criteria and regulations for metals are still based on the total concentration of the metal in question, it is becoming more and more evident that spectroscopic speciation is vital for regulatory risk assessment of environmentally relevant metals in conjunction with in vivo animal data and validated in vitro extractions for human health effects and plant uptake/food-chain transfer for sensitive ecoreceptors. These innovative research tools are expanding our ability to directly identify the role of metal speciation on many dynamic processes that influence bioavailability and risk.

The application of synchrotron techniques for the speciation of metals to assess bioavailability seems logical to this chapter, but to the common regulator a simpler approach has been to pick a fractional number relative to the total concentration of a metal in order to establish a cleanup standard. Fortunately from a human health perspective, the common regulator approach is significantly conservative almost to a point that hinders common sense for site remediation. A good example of this is arsenic which is regulated assuming 100% bioavailability, yet several studies have demonstrated that absolute bioavailability of arsenic at most sites can be as low as 20% through matrix effects or natural attenuation processes. If a lower bioavailability value can be utilized at a site, then the effective cleanup standard is raised resulting in significant savings in remedial clean up costs without harm to human health or the natural environment. However, few speciation studies have truly taken on the task of addressing bioavailability from start to finish—meaning many synchrotron-based studies will broadly state that their results support an understanding or prediction of bioavailability but provide no real data on bioavailability to support the claim. There is much speciation research needed to complement in vivo and in vitro research on metal bioavailability that can lead to effective predictive models on the long-term fate of contaminants.