Oral and Dermal Bioavailability Studies of Polycyclic Aromatic Hydrocarbons from Soils Containing Weathered Fragments of Clay Shooting Targets.

The relative oral bioavailability and dermal absorption of chemical substances from environmental media are key factors that are needed to accurately estimate site-specific risks and manage human exposures. This study evaluated the in vivo relative oral bioavailability and in vitro dermal absorption of several polycyclic aromatic hydrocarbons (PAHs) found in soils collected from two formerly used Department of Defense sites impacted by weathered fragments of clay shooting targets. Concentrations of individual carcinogenic PAHs in the ≤250 µm fraction of soil ranged from approximately 0.1 to 100 mg/kg. A novel sample preparation method was developed to produce accurate and precise test diets for oral studies. The resulting test diets showed consistent concentrations of PAHs in soil- and soil-extract-amended diets and a consistent PAH concentration profile. Mean oral relative bioavailability factors (RBAFs) and dermal absorption fractions (ABSd) for benzo(a)pyrene ranged from 8 to 14% and 0.58 to 1.3%, respectively. Using the RBAF and ABSd values, measured here, for benzo(a)pyrene in USEPA's regional screening level equations yields concentrations for residential soils that are approximately eight times higher than those when default values are used (e.g., 9.6 vs 1.2 mg/kg at a target excess risk of 1 × 10-5).

Risk assessment implications of site-specific oral relative bioavailability factors and dermal absorption fractions for polycyclic aromatic hydrocarbons in surface soils impacted by clay skeet target fragments.

Risk assessment conclusions for a site may differ when using site-specific versus default values for the relative bioavailability factor (RBAF) and dermal absorption fraction (ABS.d), because these inputs affect both surface soil screening levels and risk/hazard estimates. Indeed, our case study demonstrates that different conclusions may be reached as to regulatory need for remedial action to protect human health when evaluating soil sampling data for seven carcinogenic polycyclic aromatic hydrocarbons (PAHs) using site-specific versus default TCEQ and USEPA residential soil screening levels. Use of site-specific RBAF and ABS.d values increased carcinogenicity-based TCEQ and USEPA surface soil screening levels for PAHs by 4.4- and 6-fold on average, respectively. Soil screening levels for PAHs were more sensitive to changes in ingestion exposure route parameters than to changes in dermal exposure route parameters. Accordingly, site-specific RBAF and ABS.d information has important implications for screening chemicals at PAH-impacted sites, and in addition provides more realistic estimates of risks/hazards posed by PAHs in soil with reduced uncertainty compared to estimates based on default RBAF and ABS.d values. Although default values are generally deemed acceptable by regulatory agencies, use of risk/hazard estimates based on these default values may compel insufficiently justified remedial action in some instances.

Quantitative determination of hydroxy polycylic aromatic hydrocarbons as a biomarker of exposure to carcinogenic polycyclic aromatic hydrocarbons.

A high-resolution gas chromatography/high-resolution mass spectrometry (HRGC/HRMS) method was developed for quantitative analysis of hydroxy polycyclic aromatic hydrocarbons (OH-PAHs). Four hydroxy metabolites of known and suspected carcinogenic PAHs (benzo[a]pyrene (B[a]P), benz[a]anthracene (B[a]A), and chrysene (CRY)) were selected as suitable biomarkers of PAH exposure and associated risks to human health. The analytical method included enzymatic deconjugation, liquid - liquid extraction, followed by derivatization with methyl-N-(trimethylsilyl) trifluoroacetamide and instrumental analysis. Photo-induced oxidation of target analytes - which has plagued previously published methods - was controlled by a combination of minimizing exposure to light, employing an antioxidant (2-mercaptoethanol) and utilizing a nitrogen atmosphere. Stability investigations also indicated that conjugated forms of the analytes are more stable than the non-conjugated forms. Accuracy and precision of the method were 77.4-101% (<4.9% RSD) in synthetic urine and 92.3-117% (<15% RSD) in human urine, respectively. Method detection limits, determined using eight replicates of low-level spiked human urine, ranged from 13 to 24pg/mL. The method was successfully applied for analysis of a pooled human urine sample and 78 mouse urine samples collected from mice fed with PAH-contaminated diets. In mouse urine, greater than 94% of each analyte was present in its conjugated form.