Reassessment of the experimental skin permeability coefficients of polycyclic aromatic hydrocarbons and organophosphorus pesticides

Highlights

• PAHs and OPPs’ k_Ps have been reported from assays with solvent-deposited substances.

• A method to determine k_P from this type of skin permeation assays is not settled.

• The k_P values can be calculated using the maximum flux and the water solubility.

• The method can be easily generalized to vehicle-deposited toxicants and drugs.

• PAHs and OPPs’ fluxes and k_Ps are systematized for researchers and risk assessors.

Abstract

Human exposure to polycyclic aromatic hydrocarbons (PAHs) and organophosphorus pesticides (OPPs) by dermal route is a continuing concern in environmental and occupational toxicology. Diverse authors have measured in vitro the absorption flux and permeability coefficient (k_P) of those compounds delivered on skin surface using volatile solvents. However, there isn’t a harmonized method to obtain k_P when the test substance is deposited on the skin as a solid. Consequently, varied experimental k_Ps have been reported for PAHs and OPPs, most in clear disagreement with the values predicted by well-established mathematical models. In this work, we collected the permeation fluxes reported for these toxicants through human skin and calculated the (aqueous) k_Ps using a method based on the maximum flux and water solubility. The reanalyzed fluxes and recalculated k_Ps show improved consistency between the different experimental works and mathematical models. Notably, the recalculated k_P of benzo[a]pyrene, among others, was approximately 100 times higher than it had been previously considered. Suggestions are given to generalize the method in studies with other solvent-deposited toxicants and drugs.

Introduction

Dermal exposure to polycyclic aromatic hydrocarbons (PAHs) and organophosphorus pesticides (OPPs) is an important concern (Beitel et al., 2020; Fent et al., 2020; Peckham et al., 2017; Thredgold et al., 2019).

Skin permeation of chemicals can be evaluated in vitro using diffusion cells with excised skin from human or animal sources (Franz, 1975). The test substance is applied to the surface of a skin sample separating the two compartments (donor and receptor) of the cell, and the substance or metabolites are monitored in the receptor fluid throughout the experiment (Figure S.1 in Supplementary Material). The absorption rate or flux (J, mg/cm²/h) is calculated from the linear part of the absorption versus time curve (OECD, 2004). The permeability coefficient (k_P, cm/h) is a general representation of the rate at which a chemical penetrates the skin (OECD, 2019) and is used by regulatory agencies for dermal risk assessment (Mitragotri et al., 2011; Stepanov et al., 2020). This important skin absorption parameter is the infinite-dose steady-state flux (J_ss) divided by the difference in concentrations of donor and receptor compartment (ΔC). If sink conditions are maintained in the receptor fluid, ΔC is equal to the substance concentration in the vehicle present at the donor compartment, C_v (Hopf et al., 2020; Mitragotri et al., 2011):

$$J_{\text{S}\text{S}}=k_{\text{P}}\bullet \Delta \text{C}\cong k_{\text{P}}\bullet C_{\text{v}}$$

And k_P is then calculated as:

$$k_{\text{P}}=\frac{J_{\text{S}\text{S}}}{C_{\text{v}}}$$

Assuming the skin as a single pseudo-homogeneous membrane, k_P is defined as:

$$k_{\text{P}}=\frac{K_{\text{S}\text{C}}\bullet D}{h}$$

where K_SC is the stratum corneum/vehicle partition coefficient of the chemical, D is the effective diffusion coefficient and h the thickness of the diffusion barrier (Mitragotri et al., 2011; Zhang et al., 2009).

K_SC is the ratio of solute concentrations in the stratum corneum (C_SC) and in the vehicle (C_v). In the case a saturated solution of the test substance is maintained in the donor, C_v=S_v, and assuming the partition to the stratum corneum is fast, C_SC=S_SC:

$${K_{\text{S}\text{C}}=\frac{C_{\text{S}\text{C}}}{C_{\text{v}}}=\frac{S_{\text{S}\text{C}}}{S_{\text{v}}}}_{}$$

where S_SC and S_v are the solubilities of the permeant in the stratum corneum and in the vehicle contacting the skin, respectively (Zhang et al., 2009).

A common approach in permeation experiments is to load the test substance via a (super)saturated solution or suspension in the donor compartment (Figure S.1), which easily guarantees an infinite-dose of the permeant available for permeation through time (J=J_ss). Importantly, these conditions are also anticipated to saturate the stratum corneum and obtain the maximum flux of the permeant (J=J_max). For these saturating conditions, Eq. 1 can be rewritten (Mitragotri et al., 2011; Zhang et al., 2009):

$$J_{\text{m}\text{a}\text{x}}=k_{\text{P}}\bullet S_{\text{v}}=\frac{D}{h}{\bullet S}_{\text{S}\text{C}}$$

J_max is independent of the vehicle considered (if not affecting skin properties) and a useful parameter for assessing the skin penetration potential of chemicals. D/h and S_SC are difficult to determine experimentally, but, alternatively, J_max of a chemical can be estimated knowing k_P and the solubility of the chemical in the same vehicle (Eq. 5). Since k_P depends on the vehicle (see K_SC in Eq. 3), it is imperative to be combined only with S_v values in the same vehicle, e.g. water (Hopf et al., 2020; Mitragotri et al., 2011).

However, there are different ways to apply the test substances on the skin for dermal absorption studies (Figure S.1). Depending on the purpose of the study, the substance can be applied as solid/granule or liquid formulation in an aqueous or organic solution or suspension, the option that best represents the real exposure scenarios or the most suitable for obtaining the skin absorption parameters of the chemical (e.g., the k_P). When the substance is insoluble in water, such as PAHs and many pesticides, the k_P values measured with aqueous vehicles are not recommended for risk assessment, but the alternative assay methodologies pose extra challenges for determination of the dermal absorption parameters (Hopf et al., 2020).

Sartorelli et al. (1998) provided kinetic curves of permeation of several PAHs and OPPs through monkey skin and calculated the corresponding k_P values (Table 1). The authors have also studied the phenoxy herbicides 2-methyl-4-chlorophenoxyacetic acid (MCPA) and 2,4-dichlorophenoxyacetic acid (2,4-D). In the permeation assays, all the compounds were applied on the skin in a small volume of acetone solutions (dermal loads from 6.9–626 nmol/cm²). Since the donor compartment was left unoccluded, it can be expected that the vehicle solvent evaporated and the compounds became deposited onto skin. In a subsequent work, Sartorelli et al. (2001) used human skin to study the dermal absorption of benzo[a]pyrene (BaP) and other PAHs, but k_Ps were not reported in the article.

More recently, Hopf et al. (2018) presented the permeation kinetics and k_P values of major PAHs measured with human skin. In this study, the compounds were also applied under the form of an acetone mixture with PAH concentrations 5 mg/mL, and the reported k_P values are similar to those in Sartorelli et al. (1998) accepting a tolerance of one order of magnitude (Table 1). However, the k_P values reported in both works seem very high for the polar compounds, such as omethoate, and very low for the more lipophilic compounds such as chlorpyrifos or BaP. Surprisingly, another work (Griffin et al., 2000) reported an even lower k_P for chlorpyrifos (logK_o/w =4.96) (Table 1), and this work serves as dataset supporting query results delivered by the recent skin permeation database HuskinDB (Stepanov et al., 2020).

Facing these apparent inconsistencies in the experimental data, we employed two mathematical models to estimate the k_Ps of the PAHs and OPPs studied. The empirical model of Potts and Guy (Potts and Guy, 1992) and the mechanistic Mitragotri model (Mitragotri, 2002) are relatively simple models that were shown to predict reasonably well the k_P of a great diversity of compounds (Mitragotri et al., 2011), except for highly hydrophilic logK_o/w <-2 compounds (Lian et al., 2008). The values of logK_o/w required to apply these models were obtained by the XLOGP3 method (Cheng et al., 2007) and retrieved at the SwissADME web tool (http://www.swissadme.ch/). It can be observed in Table 1 and Fig. 1 that the k_P values resulting from both mathematical models are substantially different from the values reported in the experimental works, namely for the more polar and apolar compounds. Also noticeable is the fact that the reported k_P values show an inverse correlation with the logK_o/w of the tested substances (Fig. 1), and are in poor accordance with results from other experimental works (Peckham et al., 2017; Thredgold et al., 2019). For an additional example, the top k_P value obtained by Sartorelli et al. (1998) for dimethoate (Table 1) contrasts with the results by Nielsen et al. (2004) which observed no permeation of this pesticide through human skin during 48-h assays.

In the experiments by Sartorelli et al. (1998) and Hopf et al. (2018) the tested substances were loaded onto the skin as acetone solutions, and Griffin et al. (2000) used ethanol to deposit chlorpyrifos. Particulate or dissolved solids can be applied to the skin in acetone or other volatile solvent that rapidly evaporates and has no significant effects on skin permeability (Franz, 1975; Hopf et al., 2020). This procedure is regarded as the best one, mimicking the actual conditions of occupational and environmental exposure to some chemicals (Moore et al., 2014). However, following this experimental method there is no actual solution of the test substance present at the donor in the course of the permeation experiment (Figure S.1), and the guidance documents do not state what should be the C_v or equivalent to consider in the determination of k_P (Eq. 2). In consequence, various authors follow different criteria for a surrogate of C_v, such as using the applied dose per area or the initial concentration in the solvent (before evaporation). A further misunderstanding which gives rise to errors in the determination of k_P are the incoherent units of permeation flux, concentration and k_P. Kissel and Bunge (2003) have previously commented on the incorrect k_P values reported by Sartorelli et al. (1998) and the difficulties in calculating the k_P in experiments in which the test compound is deposited onto the skin instead of being delivered in solution.

Guidelines are available from OECD for dermal absorption studies, namely the Guidance Document 28 for the conduct of skin absorption studies (OECD, 2004), the Test Guideline 428 for in vitro assays and, later, the Guidance Notes 156 document on dermal absorption that is currently under review (OECD, 2019). Meanwhile, varied authors have provided practical insights for revision of the skin absorption guidance documents and defended further harmonization to improve the quality of in vitro permeation studies (Hopf et al., 2020; Sullivan et al., 2017).

For the harmonization of the method of calculation of k_P from experiments where the test substance is deposited on the skin surface, instead of being presented in a liquid maintained in the donor compartment, we suggest the use of the maximum flux and the water solubility of the substance (Eq. 5). This principle will be applied in the following section to recalculate the k_P values of PAHs and OPPs. A broader goal of this work is to reconcile available data on absorption fluxes and k_Ps for these important environmental and occupational toxicants.