Julia Tigges

Xenobiotic metabolism capacities of human skin in comparison with a 3D epidermis model and keratinocyte-based cell culture as in vitro alternatives for chemical testing: activating enzymes (Phase I)

Abstract:  The 7th Amendment to the EU Cosmetics Directive prohibits the use of animals in cosmetic testing for certain endpoints, such as genotoxicity. Therefore, skin in vitro models have to replace chemical testing in vivo. However, the metabolic competence neither of human skin nor of alternative in vitro models has so far been fully characterized, although skin is the first‐pass organ for accidentally or purposely (cosmetics and pharmaceuticals) applied chemicals. Thus, there is an urgent need to understand the xenobiotic‐metabolizing capacities of human skin and to compare these activities to models developed to replace animal testing. We have measured the activity of the phase II enzymes glutathione S‐transferase, UDP‐glucuronosyltransferase and N‐acetyltransferase in ex vivo human skin, the 3D epidermal model EpiDerm 200 (EPI‐200), immortalized keratinocyte‐based cell lines (HaCaT and NCTC 2544) and primary normal human epidermal keratinocytes. We show that all three phase II enzymes are present and highly active in skin as compared to phase I. Human skin, therefore, represents a more detoxifying than activating organ. This work systematically compares the activities of three important phase II enzymes in four different in vitro models directly to human skin. We conclude from our studies that 3D epidermal models, like the EPI‐200 employed here, are superior over monolayer cultures in mimicking human skin xenobiotic metabolism and thus better suited for dermatotoxicity testing.

The hallmarks of fibroblast ageing

Ageing is influenced by the intrinsic disposition delineating what is maximally possible and extrinsic factors determining how that frame is individually exploited. Intrinsic and extrinsic ageing processes act on the dermis, a post-mitotic skin compartment mainly consisting of extracellular matrix and fibroblasts. Dermal fibroblasts are long-lived cells constantly undergoing damage accumulation and (mal-)adaptation, thus constituting a powerful indicator system for human ageing. Here, we use the systematic of ubiquitous hallmarks of ageing (Lopez-Otin et al., 2013, Cell 153) to categorise the available knowledge regarding dermal fibroblast ageing. We discriminate processes inducible in culture from phenomena apparent in skin biopsies or primary cells from old donors, coming to the following conclusions: (i) Fibroblasts aged in culture exhibit most of the established, ubiquitous hallmarks of ageing. (ii) Not all of these hallmarks have been detected or investigated in fibroblasts aged in situ (in the skin). (iii) Dermal fibroblasts aged in vitro and in vivo exhibit additional features currently not considered ubiquitous hallmarks of ageing. (iv) The ageing process of dermal fibroblasts in their physiological tissue environment has only been partially elucidated, although these cells have been a preferred model of cell ageing in vitro for decades.

Estradiol protects dermal hyaluronan/versican matrix during photoaging by release of epidermal growth factor from keratinocytes.

Background: Skin aging involves UVB-induced degeneration of the dermal extracellular matrix. Results: Estrogen induces epidermal growth factor expression in keratinocytes thereby stimulating hyaluronan synthase 3 and versican expression in dermal fibroblasts of UVB-irradiated skin. Conclusion: Paracrine release of epidermal growth factor in response to estrogen maintains hyaluronan and versican-rich extracellular matrix. Significance: Estrogen prevents specific aging responses in the hyaluronan matrix of photoaged skin. Hyaluronan (HA) and versican are key components of the dermis and are responsive to ultraviolet (UV)B-induced remodeling. The aim of this study was to explore the molecular mechanisms mediating the effects of estrogen (E2) on HA-rich extracellular matrix during photoaging. Hairless skh-1 mice were irradiated with UVB (three times, 1 minimal erythema dose (80 mJ/cm2), weekly) for 10 weeks, and endogenous sex hormone production was abrogated by ovariectomy. Subcutaneous substitution of E2 by means of controlled-release pellets caused a strong increase in the dermal HA content in both irradiated and nonirradiated skin. The increase in dermal HA correlated with induction of HA synthase HAS3 by E2. Expression of splice variant 2 of the HA-binding proteoglycan versican was also increased by E2. In search of candidate mediators of these effects, it was found that E2 strongly induced the expression of epidermal growth factor (EGF) in UVB-irradiated epidermis in vivo and in keratinocytes in vitro. EGF in turn up-regulated the expression of HAS3 and versican V2 in dermal fibroblasts. HAS3 knockdown by shRNA caused inhibition of fibroblast proliferation. Furthermore, HAS3 and versican V2 induction by E2 correlated positively with proliferation in vivo. In addition, the accumulation of inflammatory macrophages, expression of inducible cyclooxygenase 2, as well as proinflammatory monocyte chemotactic protein 1 were decreased in response to E2 in the dermis. Collectively, these data suggest that E2 treatment increases the amount of dermal HA and versican V2 via paracrine release of EGF, which may be implicated in the pro-proliferative and anti-inflammatory effects of E2 during photoaging.

Evidence for a novel anti-apoptotic pathway in human keratinocytes involving the aryl hydrocarbon receptor, E2F1, and checkpoint kinase 1

Exposure of keratinocytes (KC) to ultraviolet (UV) radiation results in the initiation of apoptosis, a protective mechanism that eliminates cells harboring irreparable DNA damage. Hence, a manipulation of UV-induced apoptosis may significantly influence photocarcinogenesis. We have discovered that the aryl hydrocarbon receptor (AHR), a key regulator of drug metabolism and an UVB-sensitive transcription factor, serves an anti-apoptotic function in UVB-irradiated human KC. Chemical and shRNA-mediated inhibition of AHR signaling sensitized KC to UVB-induced apoptosis by decreasing the expression of E2F1 and its target gene checkpoint kinase 1 (CHK1). The decreased expression of these cell-cycle regulators was due to an enhanced expression of p27KIP1 and an associated decrease in phosphorylation of both cyclin-dependent kinase 2 and its substrate molecule retinoblastoma protein. The subsequent inhibition of E2F1 autoregulation and downstream CHK1 expression resulted in an enhanced susceptibility of damaged cells to undergo apoptosis. Accordingly, ectopic overexpression of either E2F1 or CHK1 in AHR-knockdown KC attenuated the observed sensitization to UVB-induced apoptosis. Using an AHR-knockout SKH-1 hairless mouse model, we next demonstrated the physiological relevance of the anti-apoptotic function of AHR. In contrast to their AHR-proficient littermates, the constitutive expression of E2F1 and CHK1 was significantly reduced in the skin of AHR-knockout mice. Accordingly, a single exposure of the animals to UVB resulted in an enhanced cleavage of caspase-3 in the skin of AHR-knockout mice. These results identify for the first time the AHR-E2F1-CHK1 axis as a novel anti-apoptotic pathway in KC, which may represent a suitable target for chemoprevention of non-melanoma skin cancer.

The new aryl hydrocarbon receptor antagonist E/Z-2-benzylindene-5,6-dimethoxy-3,3-dimethylindan-1-one protects against UVB-induced signal transduction.

TO THE EDITOR The aryl hydrocarbon receptor (AhR) is a ligand-activated transcription factor that mediates the toxicity of 2,3,7, 8-tetrachlorodibenzo-p-dioxin (TCDD), polycyclic aromatic hydrocarbons, and related environmental contaminants (Abel and Haarmann-Stemmann, 2010). The unligated AhR is trapped in a cytosolic multiprotein complex, which rapidly dissociates upon ligand binding. Subsequently, the AhR shuttles into the nucleus, dimerizes with ARNT, and binds to xenobiotic-responsive elements (XREs) in the promoter of target genes, e.g., encoding cytochrome P450 (CYP) 1 monooxygenases, to enforce transcription (Abel and Haarmann-Stemmann, 2010). Furthermore, AhR-triggered activation of c-src tyrosine kinase stimulates EGFR and downstream mitogen-activated protein kinase signaling, resulting in the induction of XRE-independent genes, such as cyclooxygenase-2 (COX-2; Abel and Haarmann-Stemmann, 2010). We have previously shown that the AhR in keratinocytes is not only activated by anthropogenic chemicals but also by UVB irradiation, which leads to the intracellular formation of the tryptophan photoproduct and high-affinity AhR ligand 6-formylindolo[3,2-b]carbazole (FICZ; Rannug et al., 1995; Fritsche et al., 2007). Indeed, UVB exposure enhances AhR/XRE binding (Supplementary Figure 1 online) and accompanied CYP1A1/1B1 expression (Katiyar et al., 2000), as well as XRE-independent COX-2 expression (Fritsche et al., 2007). Because (i) overexpression of a constitutively active AhR causes inflammatory skin lesions (Tauchi et al., 2005), (ii) an increase in CYP activity leads to reactive oxygen species formation (Puntarulo and Cederbaum, 1998), (iii) CYP1 enzymes are critical for chemical-induced skin carcinogenesis (Shimizu et al., 2000), and (iv) COX-2 is involved in UV-induced inflammation and carcinogenesis (Elmets et al., 2010), it was postulated that a transient inhibition of AhR may protect human skin against the detrimental effects of UVB irradiation (Agostinis et al., 2007; Haarmann-Stemmann et al., 2012). Moreover, we have shown that the expression of matrix metalloproteinase-1 (MMP-1), which is critically involved in extrinsic skin aging, is upregulated in an AhR-dependent manner in tobacco smoke extract–exposed keratinocytes (Ono et al., 2013). Therefore, we decided to develop an AHR antagonist that is suitable for topical UV-protection. We screened a library of compounds that possess the structural prerequisites to interact with AhR and identified E/Z-2-benzylidene-5,6-dimethoxy-3,3-dimethylindan-1-one (BDDI; Figure 1a) as the most promising candidate. Figure 1 Characterization of antagonistic capacities of E/Z-2-benzylidene-5,6-dimethoxy-3,3-dimethylindan-1-one (BDDI) in normal human epidermal keratinocytes (NHEKs) In concentrations from 0.33 to 33 µM, BDDI did not negatively affect cell viability or cause cytotoxicity in normal human epidermal keratinocytes (NHEKs; Figure 1b; for description of methods see Supplementary Material online). It is noteworthy that exposure to higher concentrations of BDDI enhanced cell viability (Figure 1b) without increasing the proliferation rate (data not shown). Exposure of NHEKs to 0.33 to 33 µM BDDI or 10 µM of the specific AhR inhibitor 3′-methoxy-4′-nitroflavone (MNF; Lu et al., 1995) resulted in a concentration-dependent decrease of basal CYP1A1 expression (Figure 1c). AhR activation by 10nM FICZ or 250 nM benzo(a)pyrene [B(a)P] resulted in 12- to 14-fold induction of CYP1A1 transcription after 4 h, whereas irradiation with 100 Jm−2 UVB led to a fourfold increase in CYP1A1 expression after 8 hours (Figure 1c). Pretreatment with 10µM MNF or 33 µM BDDI significantly attenuated CYP1A1 induction. Interestingly, a dose of 3.3 µM BDDI was sufficient to repress UVB- and FICZ-stimulated, but not B(a)P-induced, CYP1A1 expression, which was probably due to the different half-lives of the inducing agents (Figure 1c). To confirm the inhibitory effect of BDDI on CYP1A1, we treated NHEKs for 24 h with 250 nM B(a)P alone or in combination with BDDI and measured CYP1A-mediated 7-O-ethoxyresorufin-deethylase (EROD) activities. A 1 hour pretreatment with 0.33 to 33 µM BDDI resulted in a dose-dependent decline of B(a)P-induced EROD activity (Figure 1d), thereby confirming the AhR antagonistic properties of BDDI. Importantly, BDDI only transiently represses AhR function. Whereas a BDDI pretreatment for 1 hour attenuated UVB-mediated CYP1A1 induction in NHEKs, a pretreatment for 24 hours was not effective (Figure 1e). BDDI treatment of NHEKs directly after irradiation also decreased UVB-mediated CYP1A1 induction, providing evidence that BDDI does not act as a UVB-filter (Figure 1f). To elucidate the mode of action of BDDI, we performed an electrophoretic mobility shift assay (EMSA) that is well established to detect a direct binding of AhR/ARNT to a XRE consensus oligonucleotide (Denison et al., 1988; Vogel et al., 2004). Upon exposure of human HaCaT keratinocytes to 10 nM TCDD or 100 nM FICZ, we observed a strong binding of the AhR/ARNT complex to its DNA target motif (Figure 2a). Co-exposure of HaCaT cells to 3.3 µM BDDI or 10 µM MNF clearly blocked TCDD-and FICZ-triggered AhR/XRE binding (Figure 2a), providing evidence that BDDI acts as a true competitive AhR antagonist. Figure 2 BDDI disturbs XRE binding of aryl hydrocarbon receptor (AhR)/ARNT and represses UVB-induced gene expression in a human in vivo study To investigate whether BDDI is suitable for UV-protection of human skin, we treated defined skin areas of 10 healthy volunteers once daily on 4 consecutive days with a formulation containing 0.5% BDDI or a placebo formulation (Figure 2b). On day 4, 2 hours after the application of the substances, volunteers were irradiated with 1.5 MED (minimal erythema dose) UVB, and 24 hours later skin biopsies were taken. Quantitative expression analyses revealed a significantly increased expression of CYP1A1, COX-2, and MMP-1 in UVB-irradiated compared with sham-irradiated skin. Topical application of BDDI, but not the placebo formulation, significantly reduced the UVB-induced expression of all these genes, indicating that BDDI penetrates human skin and blocks AhR-dependent signaling. This experiment also revealed that the AhR is activated upon UVB irradiation in human skin in vivo. Importantly, the erythema response of the volunteers was not significantly affected during the study. As CYP1A1, COX-2, and MMP-1 are critically involved in cutaneous inflammatory diseases, skin cancer, and skin aging, we propose that the topical application of this chemical inhibitor presents a promising strategy to protect human skin against UVB-induced damage. In contrast to MNF (structural safety alert), BDDI has the clinical advantage of being suitable for dermal applications in humans. Our in vitro data further indicate that BDDI may protect against the adverse effects of polycyclic aromatic hydrocarbons, which are frequently found on airborne particulate matter (Vierkotter et al., 2010). Finally, BDDI may serve as a tool to study the involvement of AhR signaling in human skin (patho)physiology.

MicroRNA-15b regulates mitochondrial ROS production and the senescence-associated secretory phenotype through sirtuin 4/SIRT4

Mammalian sirtuins are involved in the control of metabolism and life-span regulation. Here, we link the mitochondrial sirtuin SIRT4 with cellular senescence, skin aging, and mitochondrial dysfunction. SIRT4 expression significantly increased in human dermal fibroblasts undergoing replicative or stress-induced senescence triggered by UVB or gamma-irradiation. In-vivo, SIRT4 mRNA levels were upregulated in photoaged vs. non-photoaged human skin. Interestingly, in all models of cellular senescence and in photoaged skin, upregulation of SIRT4 expression was associated with decreased levels of miR-15b. The latter was causally linked to increased SIRT4 expression because miR-15b targets a functional binding site in the SIRT4 gene and transfection of oligonucleotides mimicking miR-15b function prevented SIRT4 upregulation in senescent cells. Importantly, increased SIRT4 negatively impacted on mitochondrial functions and contributed to the development of a senescent phenotype. Accordingly, we observed that inhibition of miR-15b, in a SIRT4-dependent manner, increased generation of mitochondrial reactive oxygen species, decreased mitochondrial membrane potential, and modulated mRNA levels of nuclear encoded mitochondrial genes and components of the senescence-associated secretory phenotype (SASP). Thus, miR-15b is a negative regulator of stress-induced SIRT4 expression thereby counteracting senescence associated mitochondrial dysfunction and regulating the SASP and possibly organ aging, such as photoaging of human skin.

miR-23a-3p causes cellular senescence by targeting hyaluronan synthase 2: possible implication for skin aging.

Even though aging and cellular senescence appear to be linked, the biological mechanisms interconnecting these two processes remain to be unravelled. Therefore, microRNA (miRNA/miR) profiles were analyzed ex vivo by means of gene array in fibroblasts isolated from young and old human donors. Expression of several miRNAs was positively correlated with donor age. Among them, miR-23a-3p was shown to target hyaluronan synthase 2 (HAS2). HA is a polysaccharide of the extracellular matrix that critically regulates the phenotype of fibroblasts. Indeed, both aged and senescent fibroblasts showed increased miR-23a-3p expression and secreted significantly lower amounts of HA compared with young and non-senescent fibroblasts. Ectopic overexpression of miR-23a-3p in non-senescent fibroblasts led to decreased HAS2-mediated HA synthesis, upregulation of senescence-associated markers, and decreased proliferation. In addition, siRNA-mediated downregulation of HAS2 and pharmacological inhibition of HA synthesis by 4-methylumbelliferone mimicked the effects of miR-23a-3p. In vivo, miR-23a-3p was upregulated and HAS2 was downregulated in the skin of old mice compared with young mice. Inhibition of HA synthesis by 4-methylumbelliferone in mice reduced dermal hydration and viscoelasticity, thereby mimicking an aged skin phenotype. Taken together, these findings appear to link miR-23a-3p and the HA microenvironment as effector mechanisms in both dermal aging and senescence.

Inadequate mito-biogenesis in primary dermal fibroblasts from old humans is associated with impairment of PGC1A-independent stimulation.

Extrinsic skin ageing converges on the dermis, a post-mitotic tissue compartment consisting of extracellular matrix and long-lived fibroblasts prone to damage accumulation and maladaptation. Aged human fibroblasts exhibit mitochondrial and nuclear dysfunctions, which may be a cause or consequence of ageing. We report on a systematic study of human dermal fibroblasts retrieved from female donors aged 20-67 years and analysed ex vivo at low population doubling precluding replicative senescence. According to gene set enrichment analysis of genome wide array data, the most prominent age-associated change of the transcriptome was decreased expression of mitochondrial genes. Consistent with that, mitochondrial content and cell proliferation declined with donor age. This was associated with upregulation of AMP-dependent protein kinase (AMPK), increased mRNA levels of PPARγ-coactivator 1α (PGC1A) and decreased levels of NAD(+)-dependent deacetylase sirtuin 1. In the old cells the PGC1A-mediated mito-biogenetic response to direct AMPK-stimulation by AICAR was undiminished, while the PGC1A-independent mito-biogenetic response to starvation was attenuated and accompanied by increased ROS-production. In summary, these observations suggest an age-associated decline in PGC1A-independent mito-biogenesis, which is insufficiently compensated by upregulation of the AMPK/PGC1A-axis leading under baseline conditions to decreased mitochondrial content and reductive overload of residual respiratory capacity.

Effects of the genotoxic compounds, benzo[a]pyrene and cyclophosphamide on phase 1 and 2 activities in EpiDerm™ models

The micronucleus assay in the 3D human reconstructed EpiDerm™ skin model (RSMN) is a promising new assay for evaluating genotoxicity of dermally applied chemicals. To complement the testing of metabolically activated chemicals, such as cyclophosphamide (CPA) and benzo[a]pyrene (B[a]P), we measured phase 1 (ethoxyresorufin O-deethylation (EROD) and testosterone metabolism) and 2 activities (UGTs and GSTs) in non-treated and genotoxin treated EpiDerm™ models in a study design which mimics the RSMN assay. The assay involved a three-dose dosing regimen over 72 h to take into account effects e.g. enzyme induction, which requires longer than the standard 2 dose 48-h assay. These studies demonstrated the presence of basal phase 1 and 2 activities of EpiDerm™ models. With the exception of GST, all of the activities measured did not reproducibly change over time. It was possible to measure enzyme induction using this assay design. EROD activity was significantly induced by B[a]P but not by CPA. CPA and B[a]P had little or no reproducible effects on GST and UGT activities. In conclusion, a number of metabolic enzyme activities were present in the EpiDerm™ skin model and at least the CYP1 family was inducible.

A transcriptome comparison of time-matched developing human, mouse and rat neural progenitor cells reveals human uniqueness

&NA; It is widely accepted that human brain development has unique features that cannot be represented by rodents. Obvious reasons are the evolutionary distance and divergent physiology. This might lead to false predictions when rodents are used for safety or pharmacological efficacy studies. For a better translation of animal‐based research to the human situation, human in vitro systems might be useful. In this study, we characterize developing neural progenitor cells from prenatal human and time‐matched rat and mouse brains by analyzing the changes in their transcriptome profile during neural differentiation. Moreover, we identify hub molecules that regulate neurodevelopmental processes like migration and differentiation. Consequences of modulation of three of those hubs on these processes were studied in a species‐specific context. We found that although the gene expression profiles of the three species largely differ qualitatively and quantitatively, they cluster in similar GO terms like cell migration, gliogenesis, neurogenesis or development of multicellular organism. Pharmacological modulation of the identified hub molecules triggered species‐specific cellular responses. This study underlines the importance of understanding species differences on the molecular level and advocates the use of human based in vitro models for pharmacological and toxicological research. HighlightsDifferentiating human and rodent NPC show species‐specific gene expression changes.Few regulated genes are shared between human and rodent NPC.Major neurodevelopmental processes and GO Term clusters are conserved across species.Pharmacological modulation of hub molecules trigger species‐specific responses.