Skin sensitisation prediction using read-across, an illustrative next generation risk assessment (NGRA) case study for vanillin

Skin sensitisation is a key adverse human health effect to be addressed in the safety assessment of cosmetic ingredients. Regulatory demands and scientific progress have led to the development of a Next Generation Risk Assessment (NGRA) framework, relying on the use of New Approach Methodologies (NAM) Defined Approaches (DA) and read-across instead of generating animal data. This case study illustrates the application of read-across for the prediction of the skin sensitisation potential of vanillin at the hypothetical use concentration of 0.5% in a shower gel and face cream. A three-step process was applied to select the most suitable analogues based on their protein reactivity, structural characteristics, physicochemical properties, skin metabolism profile and availability of skin sensitisation data. The applied read-across approach predicted a weak skin sensitiser potential for vanillin corresponding with a Local Lymph Node Assay EC3 value of 10%. Based on this EC3 value a point of departure of 2500 μg/cm2 was derived, resulting in an acceptable exposure level (AEL) of 25 μg/cm2. Because the consumer exposure levels (CEL) for the face cream (13.5 μg/cm2) and shower gel (0.05 μg/cm2) scenarios were lower than the AEL, the NGRA concluded both uses as safe.

Safety assessment of cosmetics by read across applied to metabolomics data of in vitro skin and liver models

As a result of the cosmetics testing ban, safety evaluations of cosmetics ingredients must now be conducted using animal-free methods. A common approach is read across, which is mainly based on structural similarities but can also be conducted using biological endpoints. Here, metabolomics was used to assess biological effects to enable a read across between a candidate cosmetic ingredient, DIV665, only studied using in vitro assays, and a structurally similar reference compound, PA102, previously investigated using traditional in vivo toxicity methods. The (1) cutaneous distribution after topical application, (2) skin metabolism, (3) liver metabolism and (4) effect on the intracellular metabolomic profiles of in vitro skin and hepatic models, SkinEthic^®RHE model and HepaRG^® cells were investigated. The compounds exhibited similar skin penetration and skin and liver metabolism, with small differences attributed to their physicochemical properties. The effects of both compounds on the metabolome of RHE and HepaRG^® cells were similarly small, both in terms of the metabolites modulated and the magnitude of changes. The patterns of metabolome changes did not fit with any known signature relating to a mode of action known to be linked to liver toxicity e.g. modification of the Krebs cycle, urea synthesis and lipid metabolism, were more reflective of transient adaptive responses. Overall, these studies indicate that PA102 is biologically similar to DIV665, allowing read across of safety endpoints, such as in vivo sub-chronic (but not reproduction toxicity) studies, for the former to be applied to DIV665. Based on this, in the absence of animal data (which is prohibited for new chemicals), it could be concluded that DIV665 applied according to the consumer topical use scenario, is similar to PA102, and is predicted to exhibit low local skin and systemic toxicity.

In vitro investigation of metabolic fate of α-mangostin and gartanin via skin permeation by LC-MS/MS and in silico evaluation of the metabolites by ADMET predictor™

BACKGROUND

Mangosteen, Garciniam angostana L., is a juicy fruit commonly found in Thailand. The rinds of Garciniam angostana L.have been used as a traditional medicine for the treatment of trauma, diarrhea and skin infection. It is also used in dermatological product such as in cosmetics. The mangosteen pericarp can be used to extract valuable bioactive xanthone compounds such as α-mangostin and gartanin. This study is aimed to predict the metabolism of α-mangostin and gartanin using in silico and in vitro skin permeation strategies.

METHODS

Based on their 2D molecular structures, metabolites of those compounds were predicted in silico using ADMET Predictor™. The K_m and V_max, for 5 important recombinant CYP isozymes 1A2, 2C9, 2C19, 2D6 and 3A4 were predicted. Moreover, the in vitro investigation of metabolites produced during skin permeation using human epidermal keratinocyte cells, neonatal (HEKn cells) was performed by LC-MS/MS.

RESULTS

It was found that the results derived from in silico were in excellent alignment with those obtained from in vitro studies for both compounds. The prediction referred that gartanin and α-mangostin were the substrate of CYP1A2, 2C9, 2C19 and 3A. In the investigation of α-mangostin metabolites by LC-MS/MS system, the MW of the parent compound was increased from 411.200 to 459.185 Da. Therefore, α-mangostin might be metabolized via tri-oxidation process. The increased molecular weight of parent compound (397.200 to 477.157 Da) illustrated that gartanin might be conjugated to sulfated derivatives.

CONCLUSIONS

In all the studies, α-mangostin and gartanin were predicted to be. metabolized via phase I and phase II metabolism (sulfation), respectively.

Skin Metabolism: Relevance of Skin Enzymes for Rational Drug Design

Transdermal therapeutic systems (TTS) have numerous pharmacological benefits. Drug release, for example, is independent of whether a patient is in a fed or a fasted state, and lower doses can be given as gastrointestinal and hepatic first-pass metabolism is avoided. In addition, inter- and intrapatient variability is minimized as the release of the drug is mainly controlled by the system. This makes TTS interesting as alternative systems to the most common dosage form of oral tablets. The difficulty with the dermal administration route is transporting the drug through the skin, since the skin is an efficient barrier against foreign bodies. Various strategies have been reported in the literature of how drug penetration can be improved. Most of them, however, focus on overcoming the stratum corneum as the first (mechanical) skin barrier. However, penetration is much more complex, and the skin's barrier function does not only depend on the stratum corneum; what has been underestimated is the second (biological) skin barrier formed of enzymes. Compared to the stratum corneum, very little is known about these enzymes, e.g., which enzymes are present in the skin and where exactly they are localized. Hence, very few strategies can be found for how to bypass or even use the skin enzyme barrier for TTS development. This review article provides an overview of the skin enzymes considered to be relevant for the biotransformation of dermally applied drugs. Also, we discuss the use of dermal prodrugs and soft drugs and give the stereoselectivity of skin metabolism careful consideration. Finally, we provide suggestions on how to make use of the current knowledge about skin enzymes for rational TTS design.

Successful proof of concept of a micronucleus genotoxicity assay on reconstructed epidermis exhibiting intrinsic metabolic activity

We investigated the commercially available Episkin LM™ reconstructed epidermis test system as a potential 3D model for human genotoxicity assessment by cytokinesis-block micronucleus assay to mitigate limitations of the currently accepted micronucleus test. We established appropriate culture conditions for cytokinesis-block micronucleus assay in maximizing the frequency of binucleated cells by choice of culture medium and calibration of the system exposure to the cytokinesis inhibitor Cytochalasin B, without affecting the basal frequency of micronuclei in the model. We confirmed that the application of the classic solvents had no significant effect on this basal level of micronuclei. We determined the performance of cytokinesis-block micronucleus assay in Episkin LM™ reconstructed epidermis to predict in vivo genotoxins by testing the genotoxicity potential of 17 well known in vivo genotoxic, progenotoxic and non-genotoxic reference chemicals over a 48 h and 72 h exposure period. We found that cytokinesis-block micronucleus assays in Episkin™ reconstructed epidermis following the 48 h-topical regimen had a specificity of 60-75% and a sensitivity of 83-85%, resulting in an overall accuracy of 76-82% for genotoxicity assessment in tissues depending on the assessment of the reference chemicals with equivocal genotoxic profiles in the literature. The positive micronucleus test results obtained without addition of any exogenous metabolic activation system confirmed the ability of Episkin LM™ reconstructed epidermis to intrinsically bioactivate progenotoxic chemicals. The evidence showed that the 72-h exposure protocol significantly improved the detection of progenotoxins. Taken together, our data demonstrated that the Episkin LM™ reconstructed epidermis system is a relevant in vitro tool in the study of genetic toxicology.

Glutathione metabolism in the HaCaT cell line as a model for the detoxification of the model sensitisers 2,4-dinitrohalobenzenes in human skin

Glutathione (GSH) is the most prominent antioxidant in cells and the co-factor of an important set of enzymes involved in the skin metabolic clearance system, glutathione S-transferases (GST). Here, we describe an LC-MS (liquid chromatography-mass spectroscopy) method to measure GSH and its disulfide form (GSSG) in HaCaT cells and a 3D Reconstructed Human Epidermis (RHE) model. In our assay, the basal level of GSH in both systems was in the low nmol/mg soluble protein range, while the level of GSSG was systematically below our limit of quantification (0.1 μM). We found that 2,4-dinitrohalobenzenes deplete the GSH present in HaCaT cells within the first hour of exposure, in a dose dependent manner. The level of GSH in HaCaT cells treated with a single non-toxic dose of 10 μM of dinitrohalobenzene was also shown to increase after two hours. While cells treated with 1-chloro-2,4-dinitrobenzene (DNCB) and 1-fluoro-2,4-dinitrobenzene (DNFB) repleted GSH to levels similar to untreated control cells within 24h, 1-bromo-2,4-dinitrobenzene (DNBB) seemed to prevent such a repletion and appeared to be the most toxic compound in all assays. A mathematical modelling of experimental results was performed to further rationalise the differences observed between test chemicals. For this purpose the biological phenomena observed were simplified into two sequential events: the initial depletion of the GSH stock after chemical treatment followed by the repletion of the GSH once the chemical was cleared. Activation of the nuclear factor E2-related factor 2 (Nrf2) pathway was observed with all compounds within two hours, and at concentrations less than 10 μM. These data show that GSH depletion and repletion occur rapidly in skin cells and emphasize the importance of conducting kinetic studies when performing in vitro experiments exploring skin sensitization.

Introduction of a methoxymethyl side chain into p-phenylenediamine attenuates its sensitizing potency and reduces the risk of allergy induction

The strong sensitizing potencies of the most important primary intermediates of oxidative hair dyes, p-phenylenediamine (PPD) and p-toluylenediamine (PTD, i.e. 2-methyl-PPD) are well established. They are considered as the key sensitizers in hair dye allergic contact dermatitis. While modification of their molecular structure is expected to alter their sensitizing properties, it may also impair their color performance. With introduction of a methoxymethyl side chain we found the primary intermediate 2-methoxymethyl-p-phenylenediamine (ME-PPD) with excellent hair coloring performance but significantly reduced sensitizing properties compared to PPD and PTD: In vitro, ME-PPD showed an attenuated innate immune response when analyzed for its protein reactivity and dendritic cell activation potential. In vivo, the effective concentration of ME-PPD necessary to induce an immune response 3-fold above vehicle control (EC3 value) in the local lymph node assay (LLNA) was 4.3%, indicating a moderate skin sensitizing potency compared to values of 0.1 and 0.17% for PPD and PTD, respectively. Finally, assessing the skin sensitizing potency of ME-PPD under consumer hair dye usage conditions through a quantitative risk assessment (QRA) indicated an allergy induction risk negligible compared to PPD or PTD.