Suitability of different reconstructed human skin models in the skin and liver Chip2 microphysiological model to investigate the kinetics and first-pass skin metabolism of the hair dye, 4-amino-2-hydroxytoluene

The HUMIMIC skin-liver Chip2 microphysiological systems model using the epidermal model, EpiDerm™, was reported previously to mimic application route-dependent metabolism of the hair dye, 4-amino-2-hydroxytoluene (AHT). Therefore, we evaluated the use of alternative skin models-SkinEthic™, EpiDermFT™ and PhenionFT™-for the same purpose. In static incubations, AHT permeation was similar using SkinEthic™ and EpiDerm™ models. Older Day 21 (D21) SkinEthic™ models with a thicker stratum corneum did not exhibit a greater barrier to AHT (overall permeation was the same in D17 and D21 models). All epidermal models metabolised AHT, with the EpiDerm™ exhibiting higher N-acetylation than SkinEthic™ models. AHT metabolism by D21 SkinEthic™ models was lower than that by D17 SkinEthic™ and EpiDerm™ models, thus a thicker stratum corneum was associated with fewer viable cells and a lower metabolic activity. AHT permeation was much slower using PhenionFT™ compared to epidermal models and better reflected permeation of AHT through native human skin. This model also extensively metabolised AHT to N-acetyl-AHT. After a single topical or systemic application of AHT to Chip2 model with PhenionFT™, medium was analysed for parent and metabolites over 5 days. The first-pass metabolism of AHT was demonstrated, and the introduction of a wash step after 30 min decreased the exposure to AHT and its metabolites by 33% and 40%-43%, respectively. In conclusion, epidermal and FT skin models used in the Chip2 can mimic the first-pass skin metabolism of AHT. This highlights the flexibility of the Chip2 to incorporate different skin models according to the purpose.

Application of a skin and liver Chip2 microphysiological model to investigate the route-dependent toxicokinetics and toxicodynamics of consumer-relevant doses of genistein

The HUMMIC skin-liver Chip2 microphysiological system using EpiDerm™ and HepaRG and stellate liver spheroids was used to evaluate the route-specific metabolism and toxicodynamic effects of genistein. Human-relevant exposure levels were compared: 60 nM representing the plasma concentration expected after topical application of a cosmetic product and 1 μM representing measured plasma concentrations after ingesting soya products. Genistein was applied as single and repeated topical and/or systemic doses. The kinetics of genistein and its metabolites were measured over 5 days. Toxicodynamic effects were measured using transcriptional analyses of skin and liver organoids harvested on Days 2 and 5. Route-specific differences in genistein's bioavailability were observed, with first-pass metabolism (sulfation) occurring in the skin after topical application. Only repeated application of 1 μM, resembling daily oral intake of soya products, induced statistically significant changes in gene expression in liver organoids only. This was concomitant with a much higher systemic concentration of genistein which was not reached in any other dosing scenario. This suggests that single or low doses of genistein are rapidly metabolised which limits its toxicodynamic effects on the liver and skin. Therefore, by facilitating longer and/or repeated applications, the Chip2 can support safety assessments by linking relevant gene modulation with systemically available parent or metabolite(s). The rate of metabolism was in accordance with the short half-life observed in in vivo in humans, thus supporting the relevance of the findings. In conclusion, the skin-liver Chip2 provides route-specific information on metabolic fate and toxicodynamics that may be relevant to safety assessment.

Development of a microphysiological skin-liver-thyroid Chip3 model and its application to evaluate the effects on thyroid hormones of topically applied cosmetic ingredients under consumer-relevant conditions

All cosmetic ingredients registered in Europe must be evaluated for their safety using non-animal methods. Microphysiological systems (MPS) offer a more complex higher tier model to evaluate chemicals. Having established a skin and liver HUMIMIC Chip2 model demonstrating how dosing scenarios impact the kinetics of chemicals, we investigated whether thyroid follicles could be incorporated to evaluate the potential of topically applied chemicals to cause endocrine disruption. This combination of models in the HUMIMIC Chip3 is new; therefore, we describe here how it was optimized using two chemicals known to inhibit thyroid production, daidzein and genistein. The MPS was comprised of Phenion^® Full Thickness skin, liver spheroids and thyroid follicles co-cultured in the TissUse HUMIMIC Chip3. Endocrine disruption effects were determined according to changes in thyroid hormones, thyroxine (T₄) and 3,3',5-triiodothyronine (T₃). A main part of the Chip3 model optimization was the replacement of freshly isolated thyroid follicles with thyrocyte-derived follicles. These were used in static incubations to demonstrate the inhibition of T₄ and T₃ production by genistein and daidzein over 4 days. Daidzein exhibited a lower inhibitory activity than genistein and both inhibitory activities were decreased after a 24 h preincubation with liver spheroids, indicating metabolism was via detoxification pathways. The skin-liver-thyroid Chip3 model was used to determine a consumer-relevant exposure to daidzein present in a body lotion based on thyroid effects. A "safe dose" of 0.235 μg/cm² i.e., 0.047% applied in 0.5 mg/cm² of body lotion was the highest concentration of daidzein which does not result in changes in T₃ and T₄ levels. This concentration correlated well with the value considered safe by regulators. In conclusion, the Chip3 model enabled the incorporation of the relevant exposure route (dermal), metabolism in the skin and liver, and the bioactivity endpoint (assessment of hormonal balance i.e., thyroid effects) into a single model. These conditions are closer to those in vivo than 2D cell/tissue assays lacking metabolic function. Importantly, it also allowed the assessment of repeated doses of chemical and a direct comparison of systemic and tissue concentrations with toxicodynamic effects over time, which is more realistic and relevant for safety assessment.

Predicting the Skin Sensitization Potential of Small Molecules with Machine Learning Models Trained on Biologically Meaningful Descriptors

In recent years, a number of machine learning models for the prediction of the skin sensitization potential of small organic molecules have been reported and become available. These models generally perform well within their applicability domains but, as a result of the use of molecular fingerprints and other non-intuitive descriptors, the interpretability of the existing models is limited. The aim of this work is to develop a strategy to replace the non-intuitive features by predicted outcomes of bioassays. We show that such replacement is indeed possible and that as few as ten interpretable, predicted bioactivities are sufficient to reach competitive performance. On a holdout data set of 257 compounds, the best model (“Skin Doctor CP:Bio”) obtained an efficiency of 0.82 and an MCC of 0.52 (at the significance level of 0.20). Skin Doctor CP:Bio is available free of charge for academic research. The modeling strategies explored in this work are easily transferable and could be adopted for the development of more interpretable machine learning models for the prediction of the bioactivity and toxicity of small organic compounds.

Demonstration of the first-pass metabolism in the skin of the hair dye, 4-amino-2-hydroxytoluene, using the Chip2 skin-liver microphysiological model

We used TissUse's HUMIMIC Chip2 microfluidic model, incorporating reconstructed skin models and liver spheroids, to investigate the impact of consumer-relevant application scenarios on the metabolic fate of the hair dye, 4-amino-2-hydroxytoluene (AHT). After a single topical or systemic application of AHT to Chip2 models, medium was analysed for parent and metabolites over 5 days. The metabolic profile of a high dose (resulting in a circuit concentration of 100 μM based on 100% bioavailability) of AHT was the same after systemic and topical application to 96-well EpiDerm™ models. Additional experiments indicated that metabolic capacity of EpiDerm™ models were saturated at this dose. At 2.5 μM, concentrations of AHT and several of its metabolites differed between application routes. Topical application resulted in a higher C_max and a 327% higher area under the curve (AUC) of N-acetyl-AHT, indicating a first-pass effect in the EpiDerm™ models. In accordance with in vivo observations, there was a concomitant decrease in the C_max and AUC of AHT-O-sulphate after topical, compared with systemic application. A similar alteration in metabolite ratios was observed using a 24-well full-thickness skin model, EpiDermFT™, indicating that a first-pass effect was also possible to detect in a more complex model. In addition, washing the EpiDermFT™ after 30 min, thus reflecting consumer use, decreased the systemic exposure to AHT and its metabolites. In conclusion, the skin-liver Chip2 model can be used to (a) recapitulate the first-pass effect of the skin and alterations in the metabolite profile of AHT observed in vivo and (b) provide consumer-relevant data regarding leave-on/rinse-off products.

Skin Doctor CP: Conformal Prediction of the Skin Sensitization Potential of Small Organic Molecules

Skin sensitization potential or potency is an important end point in the safety assessment of new chemicals and new chemical mixtures. Formerly, animal experiments such as the local lymph node assay (LLNA) were the main form of assessment. Today, however, the focus lies on the development of nonanimal testing approaches (i.e., in vitro and in chemico assays) and computational models. In this work, we investigate, based on publicly available LLNA data, the ability of aggregated, Mondrian conformal prediction classifiers to differentiate between non- sensitizing and sensitizing compounds as well as between two levels of skin sensitization potential (weak to moderate sensitizers, and strong to extreme sensitizers). The advantage of the conformal prediction framework over other modeling approaches is that it assigns compounds to activity classes only if a defined minimum level of confidence is reached for the individual predictions. This eliminates the need for applicability domain criteria that often are arbitrary in their nature and less flexible. Our new binary classifier, named Skin Doctor CP, differentiates nonsensitizers from sensitizers with a higher reliability-to-efficiency ratio than the corresponding nonconformal prediction workflow that we presented earlier. When tested on a set of 257 compounds at the significance levels of 0.10 and 0.30, the model reached an efficiency of 0.49 and 0.92, and an accuracy of 0.83 and 0.75, respectively. In addition, we developed a ternary classification workflow to differentiate nonsensitizers, weak to moderate sensitizers, and strong to extreme sensitizers. Although this model achieved satisfactory overall performance (accuracies of 0.90 and 0.73, and efficiencies of 0.42 and 0.90, at significance levels 0.10 and 0.30, respectively), it did not obtain satisfying class-wise results (at a significance level of 0.30, the validities obtained for nonsensitizers, weak to moderate sensitizers, and strong to extreme sensitizers were 0.70, 0.58, and 0.63, respectively). We argue that the model is, in consequence, unable to reliably identify strong to extreme sensitizers and suggest that other ternary models derived from the currently accessible LLNA data might suffer from the same problem. Skin Doctor CP is available via a public web service at https://nerdd.zbh.uni-hamburg.de/skinDoctorII/.

Characterization of application scenario-dependent pharmacokinetics and pharmacodynamic properties of permethrin and hyperforin in a dynamic skin and liver multi-organ-chip model

Microphysiological systems (MPS) aim to mimic the dynamic microenvironment and the interaction between tissues. While MPS exist for investigating pharmaceuticals, the applicability of MPS for cosmetics ingredients is yet to be evaluated. The HUMIMIC Chip2 ("Chip2″), is the first multi-organ chip technology to incorporate skin models, allowing for the topical route to be tested. Therefore, we have used this model to analyze the impact of different exposure scenarios on the pharmacokinetics and pharmacodynamics of two topically exposed chemicals, hyperforin and permethrin. The Chip2 incorporated reconstructed human epidermis models (EpiDerm™) and HepaRG-stellate spheroids. Initial experiments using static incubations of single organoids helped determine the optimal dose. In the Chip2 studies, parent and metabolites were analyzed in the circuit over 5 days after application of single and repeated topical or systemic doses. The gene expression of relevant xenobiotic metabolizing enzymes in liver spheroids was measured to reflect toxicodynamics effects of the compounds in liver. The results show that 1) metabolic capacities of EpiDerm™ and liver spheroids were maintained over five days; 2) EpiDerm™ model barrier function remained intact; 3) repeated application of compounds resulted in higher concentrations of parent chemicals and most metabolites compared to single application; 4) compound-specific gene induction e.g. induction of CYP3A4 by hyperforin depended on the application route and frequency; 5) different routes of application influenced the systemic concentrations of both parents and metabolites in the chip over the course of the experiment; 6) there was excellent intra- and inter-lab reproducibility. For permethrin, a process similar to the excretion in a human in vivo study could be simulated which was remarkably comparable to the in vivo situation. These results support the use of the Chip2 model to provide information on parent and metabolite disposition that may be relevant to risk assessment of topically applied cosmetics ingredients.

Skin Doctor: Machine Learning Models for Skin Sensitization Prediction that Provide Estimates and Indicators of Prediction Reliability

The ability to predict the skin sensitization potential of small organic molecules is of high importance to the development and safe application of cosmetics, drugs and pesticides. One of the most widely accepted methods for predicting this hazard is the local lymph node assay (LLNA). The goal of this work was to develop in silico models for the prediction of the skin sensitization potential of small molecules that go beyond the state of the art, with larger LLNA data sets and, most importantly, a robust and intuitive definition of the applicability domain, paired with additional indicators of the reliability of predictions. We explored a large variety of molecular descriptors and fingerprints in combination with random forest and support vector machine classifiers. The most suitable models were tested on holdout data, on which they yielded competitive performance (Matthews correlation coefficients up to 0.52; accuracies up to 0.76; areas under the receiver operating characteristic curves up to 0.83). The most favorable models are available via a public web service that, in addition to predictions, provides assessments of the applicability domain and indicators of the reliability of the individual predictions.

Non-animal methods to predict skin sensitization (II): an assessment of defined approaches *

Skin sensitization is a toxicity endpoint of widespread concern, for which the mechanistic understanding and concurrent necessity for non-animal testing approaches have evolved to a critical juncture, with many available options for predicting sensitization without using animals. Cosmetics Europe and the National Toxicology Program Interagency Center for the Evaluation of Alternative Toxicological Methods collaborated to analyze the performance of multiple non-animal data integration approaches for the skin sensitization safety assessment of cosmetics ingredients. The Cosmetics Europe Skin Tolerance Task Force (STTF) collected and generated data on 128 substances in multiple in vitro and in chemico skin sensitization assays selected based on a systematic assessment by the STTF. These assays, together with certain in silico predictions, are key components of various non-animal testing strategies that have been submitted to the Organization for Economic Cooperation and Development as case studies for skin sensitization. Curated murine local lymph node assay (LLNA) and human skin sensitization data were used to evaluate the performance of six defined approaches, comprising eight non-animal testing strategies, for both hazard and potency characterization. Defined approaches examined included consensus methods, artificial neural networks, support vector machine models, Bayesian networks, and decision trees, most of which were reproduced using open source software tools. Multiple non-animal testing strategies incorporating in vitro, in chemico, and in silico inputs demonstrated equivalent or superior performance to the LLNA when compared to both animal and human data for skin sensitization.

Licochalcone A activates Nrf2 in vitro and contributes to licorice extract-induced lowered cutaneous oxidative stress in vivo

The retrochalcone licochalcone A (LicA) has previously been shown to possess antimicrobial and anti-inflammatory properties. In this study, we focused on pathways responsible for the antioxidative properties of LicA. In vitro, LicA protected from oxidative stress mediated by reactive oxygen species (ROS) by activating the expression of cytoprotective phase II enzymes. LicA induced nuclear translocation of NF-E2-related factor 2 (Nrf2) in primary human fibroblasts and elevated the expression of the cytoprotective and anti-inflammatory enzymes heme oxygenase 1 and glutamate-cysteine ligase modifier subunit. LicA-treated cells displayed a higher ratio of reduced to oxidized glutathione and decreased concentrations of ROS in UVA-irradiated human dermal fibroblasts, as well as in activated neutrophils. In vivo, ultraweak photon emission analysis of skin treated with LicA-rich licorice extract revealed a significantly lowered UVA-induced luminescence, indicative for a decrease in oxidative processes. We conclude from these data that topical application of licorice extract is a promising approach to induce Nrf2-dependent cytoprotection in human skin.

Genomic allergen rapid detection in-house validation--a proof of concept

Chemical sensitization is an adverse immunologic response to chemical substances, inducing hypersensitivity in exposed individuals. Identifying chemical sensitizers is of great importance for chemical, pharmaceutical, and cosmetic industries, in order to prevent the use of sensitizers in consumer products. Historically, chemical sensitizers have been assessed mainly by in vivo methods, however, recently enforced European legislations urge and promote the development of animal-free test methods able to predict chemical sensitizers. Recently, we presented a predictive biomarker signature in the myeloid cell line MUTZ-3, for assessment of skin sensitizers. The identified genomic biomarkers were found to be involved in immunologically relevant pathways, induced by recognition of foreign substances and regulating dendritic cell maturation and cytoprotective mechanisms. We have developed the usage of this biomarker signature into a novel in vitro assay for assessment of chemical sensitizers, called Genomic Allergen Rapid Detection (GARD). The assay is based on chemical stimulation of MUTZ-3 cultures, using the compounds to be assayed as stimulatory agents. The readout of the assay is a transcriptional quantification of the genomic predictors, collectively termed the GARD Prediction Signature (GPS), using a complete genome expression array. Compounds are predicted as either sensitizers or nonsensitizers by a Support Vector Machine model. In this report, we provide a proof of concept for the functionality of the GARD assay by describing the classification of 26 blinded and 11 nonblinded chemicals as sensitizers or nonsensitizers. Based on these classifications, the accuracy, sensitivity, and specificity of the assay were estimated to 89, 89, and 88%, respectively.

Role of fibroblasts in the pathogenesis of atopic dermatitis

BACKGROUND

Atopic dermatitis (AD) is a common dermatosis that highly impairs a patient's quality of life. The recent discovery that epidermal barrier defects caused by an aberrant differentiation process of keratinocytes are comparably important to the well-characterized changes in immune response patterns attributed a crucial role to the keratinocytes. Fibroblasts are able to alter proliferation and differentiation of keratinocytes, but their role in AD is not yet fully understood.

OBJECTIVE

We sought to determine the role of fibroblasts in skin proliferation and differentiation in patients with AD.

METHODS

We used human 3-dimensional organotypic skin cultures consisting of atopic fibroblasts and healthy keratinocytes, as well as healthy fibroblasts and atopic keratinocytes, and compared them with their controls. The expression of differentiation markers in these organotypic cultures were analyzed by using immunohistology and quantitative RT-PCR. Furthermore, the fundamental role of fibroblast-secreted leukemia inhibitory factor was assessed by using small interfering RNA-mediated knockdown cultures.

RESULTS

We observed that atopic fibroblasts influence the proliferation of keratinocytes and the terminal differentiation process, resulting in an in vivo-like morphology of AD. Subsequently, healthy fibroblasts were able to restore the structural deficits of the epidermis consisting of atopic keratinocytes. Partially, these effects were due to a reduced expression of the differentiation-associated cytokine leukemia inhibitory factor by atopic fibroblasts.

CONCLUSION

These data demonstrate that fibroblasts and the modulation of fibroblast-derived factors might be new therapeutic targets for the alleviation of AD.

SWAP-70 associates transiently with macropinosomes

Cells accomplish the non-selective uptake of extracellular fluids, antigens and pathogens by the endocytic process of macropinocytosis. The protein SWAP-70 is a widely expressed, pleckstrin-homology (PH) domain-containing protein that marks a transitional subset of actin filaments in motile cells. Here we report that the protein SWAP-70 associates transiently with macropinosomes in dendritic cells and NIH/3T3 fibroblasts. The association of SWAP-70 with macropinosomes is preceded by the accumulation of Rac-GTP and followed by that of Rab5. Three regions of SWAP-70, the N-terminal region, the PH domain and the C-terminal region, contribute in a combinatorial manner to the transient association with newly formed macropinosomes in the cell periphery and occasionally with aged macropinosomes on their passage to the cell center. These data identify SWAP-70 as a transient component of early macropinosomes.

Functional analysis of the methylmalonyl-CoA epimerase from Caenorhabditis elegans

Methylmalonyl-CoA epimerase (MCE) is an enzyme involved in the propionyl-CoA metabolism that is responsible for the degradation of branched amino acids and odd-chain fatty acids. This pathway typically functions in the reversible conversion of propionyl-CoA to succinyl-CoA. The Caenorhabditis elegans genome contains a single gene encoding MCE (mce-1) corresponding to a 15 kDa protein. This was expressed in Escherichia coli and the enzymatic activity was determined. Analysis of the protein expression pattern at both the tissue and subcellular level by microinjection of green fluorescent protein constructs revealed expression in the pharynx, hypodermis and, most prominently in body wall muscles. The subcellular pattern agrees with predictions of mitochondrial localization. The sequence similarity to an MCE of known structure was high enough to permit a three-dimensional model to be built, suggesting conservation of ligand and metal binding sites. Comparison with corresponding sequences from a variety of organisms shows more than 1/6 of the sequence is completely conserved. Mutants allelic to mce-1 showed no obvious phenotypic alterations, demonstrating that the enzyme is not essential for normal worm development under laboratory conditions. However, survival of the knockout mutants was altered when exposed to stress conditions, with mutants surprisingly showing an increased resistance to oxidative stress.