Environ. Sci. Technol. 2005, 39, 3723-3731
Percutaneous Absorption of 4-Cyanophenol from Freshly Contaminated Soil in Vitro: Effects of Soil Loading and Contamination Concentration G . D . T O U R A I L L E , †,§ K . D . M C C A R L E Y , ‡ A. L. BUNGE,‡ J.-P. MARTY,§ AND R . H . G U Y * ,†,| Departments of Biopharmaceutical Sciences & Pharmaceutical Chemistry, University of California, San Francisco, California, Chemical Engineering and Petroleum Refining Department, Colorado School of Mines, Golden, Colorado, Faculte´ de Pharmacie, Universite´ de Paris-Sud, Chaˆtenay-Malabry, France, and Department of Pharmacy & Pharmacology, University of Bath, Claverton Down, Bath BA2 7AY, U.K.
Despite the skin’s excellent barrier function, dermal exposure to soil contaminated with toxic chemicals can represent a significant health hazard (e.g., via multiple work related contacts in the farming and waste disposal industries). The development of environmental standards or limits for chemical levels in soil has been impeded because quantification of percutaneous uptake from this medium has not been well-defined. The objective of the research described here, therefore, was to better characterize the rate and extent of dermal penetration as a function of soil loading and degree of soil contamination. The absorption of a model compound (4-cyanophenol, CP) across hairless mouse skin in vitro has been determined at four different soil loadings (5, 11, 38 and 148 mg cm-2) and at six levels of soil contamination (concentrations ranging from 0.19 to 38 mg/g soil). Following 8 h of exposure, the amount of CP absorbed was independent of soil loading when CP concentration was constant, implying that the quantity of soil present was always sufficient to provide at least a single layer of tightly packed particles. At the lowest loadings, however, with increasing times of exposure, the CP transport rate fell off due to depletion of chemical from the soil. At constant soil loading (38 mg cm-2), CP flux (Jss) across the skin was linearly proportional to the level of contamination (Cosoil) over the range 0.19 to 23.5 mg of CP per gram of soil: Jss (µg cm-2 h-1) ) (1.1 × 10-5 g cm-2 h-1) × Cosoil (µg/g soil). At the highest CP contamination concentration, however, the transport rate was about an order of magnitude higher than expected, possibly due to the presence of pure CP crystals. In conclusion, these results provide new quantifications of the characteristics of dermal uptake from chemically contaminated soils and * Corresponding author phone: +44.1225.384901; fax: +44.1225.386114; e-mail:
[email protected]. Corresponding author address: Department of Pharmacy & Pharmacology, University of Bath, Claverton Down, Bath BA2 7AY, U.K. † University of California. ‡ Colorado School of Mines. § Universite ´ de Paris-Sud. | University of Bath. 10.1021/es0494454 CCC: $30.25 Published on Web 04/05/2005
2005 American Chemical Society
important information with which to develop and verify predictive models of dermal absorption.
Introduction As a major interface between man and his environment, the skin is an important route of entry for potentially hazardous agents. Exposure to toxic substances can occur via contact with various contaminated media, and, depending on the magnitude of exposure, adverse public health consequences may be realized. Recently, soil has been recognized as a potentially important medium of exposure, but experimental studies are limited. Because most of the available data were measured by applying large amounts of soil (e.g., 40 mg cm-2 (1-5)), there are reasons to believe that these data may significantly under-represent dermal absorption (6). The need for accurate evaluation of skin absorption of chemicals from soil is necessary because the estimated risk attributable to dermal contact with contaminated soil can be significant. In a review of risk assessments for U.S. Superfund sites, Johnson and Kissel (7) found that the estimated cancer risk from dermal exposure to surface soil contamination was greater than 10-4 for 1 of 6 sites and that, for 1 of 12 sites, dermal exposure to soil presented a larger risk than exposure by any other route. Quantitative estimates of dermal uptake of chemicals from soil involve several uncertainties (8, 9). These arise from diverse and poorly understood exposure variables including the dynamic conditions by which soil contacts the skin, the interactions of the soil with the skin surface, the transport of chemicals within the skin and the chemical’s interaction with the soil (10, 11). Soil is a mineralogically, chemically and reactively heterogeneous material. Organic matter in the soil plays a major role in the binding of organic chemicals, including many pesticides (12). The mechanisms of release of bound xenobiotics from soil are not well understood, although clearly the type of interaction between the xenobiotic and the soil affects both the stability of the residue and its potential for release (13). The skin consists of two distinct layers: the epidermis and the dermis. The outermost part of the epidermis comprises a 10-20 µm “brick and mortar” structure called the stratum corneum, SC (14). The SC behaves as a hydrophobic membrane and generally presents the rate-controlling barrier in the overall skin permeation of organic nonelectrolytes (15-17). It has been further established that a chemical’s permeability through the SC from water (but not necessarily from soil) is principally determined by its size and its relative affinity for lipid and aqueous environments (measured, for example, by the octanol-water partition coefficient and the ability to form hydrogen bonds) (18-20). Two scenarios for the dermal absorption of a chemical from contaminated soil may be postulated: either (a) the SC controls mass transfer of chemical from contaminated soil and the kinetics of dermal absorption are slow compared to the kinetics of soil desorption, or (b) the situation is reversed, in which case slow soil release kinetics controls absorption through the skin (6). With respect to skin exposure to soil, we envisage two extreme situations: (i) a single skin-soil contact, after which the soil adheres to the skin until removed; or (ii) frequent skin-soil contacts where fresh soil regularly replaces soil from each previous contact. In the latter case, the chemical concentration contacting the skin remains essentially constant, whereas, in the former case, this surface level is subject to depletion because of absorption or VOL. 39, NO. 10, 2005 / ENVIRONMENTAL SCIENCE & TECHNOLOGY
9
3723
evaporation (6). Actual soil exposures are probably some combination of these two extremes Systemic human health risk from chemical exposure depends on the mass of chemical that is absorbed into the skin. Clearly, this quantity cannot exceed the mass of the chemical to which the organism is exposed. Because skin presents a formidable barrier to the penetration of many chemicals, the difference between the amount absorbed and that to which exposure occurred can be quite large for dermal contact, and the true risk may be much less than represented by assuming that all of the chemical contacting the skin is absorbed. Furthermore, it seems intuitively reasonable that the amount absorbed from the chemically contaminated soil is probably less than would be absorbed if the chemical was presented in its neat form in the same amount and for the same time exposure. That is, the bioavailability of a chemical absorbed following skin contact with soil is likely to be lower relative to that from the neat chemical. The intention of the work presented here, therefore, is to shed light upon some of these issues. As an initial step, we have measured the permeation of a model compound, 4-cyanophenol (CP), from freshly contaminated soil through hairless mouse skin. The octanol-water partition coefficient (Kow) of CP (log Kow ) 1.60) is similar to compounds in soils contaminated with munitions, such as trinitrotoluene (log Kow ) 1.60), 2,6-dinitrotoluene (log Kow ) 2.10), and trinitrobenzene (log Kow ) 1.18). Hairless mouse skin is a frequently used and readily available model for the human tissue. Its permeability is typically (and consistently) 3-5 times greater than that of human skin, making it an appropriate and conservative test system for risk assessment (21). Transport was determined as functions of CP concentration in the soil and of soil loading on the skin. For comparison, percutaneous absorption was also measured following the application of solid CP crystals and of a saturated aqueous solution of CP. Based on the data obtained, a simple membrane model has been developed to calculate the skin permeability coefficient of CP absorption following dermal exposure to contaminated soil at different loadings.
Materials and Methods The soil for these experiments was a clay loam from Fort Collins, Colorado. Only the e 250 µm sieve fraction was used (36.2% 125-250 µm, 38.5% 63-125 µm, 14.9% 38-63 µm, 4.6% 25-38 µm, and 5.8%
