In vitro skin diseased models

Tissue engineering concerns both the generation of healthy tissues and diseased tissues in order to better understand various pathological conditions. The establishment of human cell-based in vitro engineered disease model systems could represent a paradigm shift from inadequate conventional monolayer cell cultures, or moderately successful animal models, towards more physiologically tissue-relevant, patient-specific approaches.

This concept of in vitro diseased skin models is illustrated by the models for recessive epidermolysis bullosa simplex (REBS) and cutaneous squamous cell carcinoma (SCC) that were developed in our laboratory. REBS is characterised by severe intra-epidermal blister formation, resulting from fragility of the basal keratinocytes that lack keratin tonofilaments due to a homozygotic null mutation in the keratin 14 gene. In vivo EBS models, including murine and canine models, are limited and do not represent the human microenvironment. Earlier in vitro studies have indicated an essential role for fibroblasts in the REBS phenotype; however, the limited availability of human REBS skin samples hampered further in vitro research on dermal-epidermal interactions in REBS. To overcome this, an explant approach was used in which REBS skin biopsies were placed on a dermal equivalent in which REBS-associated fibroblasts were seeded.

A similar approach was used to construct an in vitro model for human SCC, which is a malignant tumour of epidermal keratinocytes characterised by invasive growth into the dermis. After basal cell carcinoma, SCC is the most common malignancy in white populations with epidemic incidence rates. Traditional in vivo SCC models rely on the use of chemical, genetic or mechanical induction or propagation of carcinogenesis in mice. Current in vitro approaches are limited to the use of cell lines, which often lack a true representation of primary skin cancer. The development of a representative in vitro skin carcinoma model based on primary SCC biopsies allows for a better understanding of fundamental carcinogenesis mechanisms and may serve as a validated pre-screening platform for candidate drugs, thereby eradicating the need for animals for these purposes.

Figure 1: Mimicking skin diseases in-vitro: (A) Clinical manifestation of Recessive Epidermolysis Bullosa Simplex (Blister disease); (B) REBS biopsy cultured onto the skin model; (C) Cross section of a skin model in which REBS has been mimicked; (D) Clinical manifestation of a squamous cell carcinoma; (E) SCC cultured onto the skin model; Cross section of a skin model mimicking SCC.

Projects

• ZONMW: “Development of validated organotypic in vitro models leading to improved therapy of skin cancer”. Project Leaders, Frank de Gruijl, Kees Tensen en A El Ghalbzouri.
• STW: "Generation of an atopic dermatitis reconstructed skin model as a tool to screen for therapeutics”.In collaboration with Prof.dr.J.Bouwstra, Faculty of Science, Leiden/Amsterdam Center for Drug Research, Drug Delivery Technology.

Skin models as an alternative to animal testing

The overall mission of the EU is to ensure that people’s health and safety at work are properly safeguarded. However, the information on which regulatory decisions regarding human health are based comes in part from studies conducted in experimental animals. The 7th Amendement of the Cosmetics Directive is transposed into national law in September 2004. It is a new Directive that pertains to testing of finished cosmetic products and ingredients. The most important rules for the near future are:

• As of 11 March 2009; a ban on animal testing of finished cosmetic ingredients within the EU.
• From March 2009; a ban for all human health effects with the exception of repeated-dose toxicity, reproductive toxicity and toxicokinetics. For these specific health effects the marketing ban will apply step by step as soon as alternative methods are validated and adopted in EU legislation with due regard to the OECD validation process, but with a maximum cut-off date of 10 years after entry into force of the Directive, i.e., 11 March 2013, irrespective of the availability of alternative non-animal tests.

All together this clearly shows that there is a need for validated in vitro methods for toxicity testing of chemicals, e.g. skin sensitizers. Many years ago, researchers have started the search for alternative methods. The key steps in the skin sensibilization process, (allergen detection, uptake and processing, cytokine signaling, migration/maturation of antigen-presenting cells (DC’s), activation and proliferation of T cells), have been used as read-out parameters in these tests. Although some tests look very promising, they are still far from being perfect.

Figure 2: HSEs are being used to test compounds for skin irritation, skin corrosion or skin sensibilisation (A). The epidermis detaches from the dermal part when 2% of SLS is topically applied onto the HSE (B). The HSE contains a competent barrier, shown is a Thin Layer Chromatography profile of the lipid composition of HSEs and native skin (C).

Although the skin models reproduce to a large extent the barrier properties of normal human skin, still some small differences are present. These differences might affect the predictability of a tested compound. In a collaborative project conducted at the LACDR we aim to further improve the barrier properties of the skin models by modifying lipids composition in the stratum corneum. In previous studies we have shown that LEMs are reliable and producible epidermal models that are suitable for the screening of potential skin irritants. The test substances can be applied topically and their irritant potential can be evaluated using various endpoints, such as the induction of tissue damage or the release of various pro-inflammatory mediators, changes in protein and mRNA expression profiles. Similar approach can be followed for testing other compounds such as corrosive compounds. Studies with HSEs can therefore contribute to our knowledge on the basic biochemical mechanisms underlying irritant reactions, and can be used to understand the structural features of molecules, which may be responsible for eliciting an irritant reaction. In addition, generation of epidermal equivalents populated with both keratinocytes and melanocytes makes it possible to study the regulation of melanogenesis, melanocyte-keratinocyte interactions, and how these processes are affected by UV irradiation. Such a model can also be used for testing the phototoxic or photo-protective potentials of various compounds and sunscreens. In another collaborative project with the RIVM we anticipate to identify biomarkers that can be used to discriminate skin sensitizers from skin irritants in order to develop simple and reliable test systems. For this purpose we will use different cell types including cell lines, primary keratinocytes, dendritic cells and/or epidermal skin models. In addition, signal transduction pathways that are known to play an important role in skin sensitization, such as the KEAP1/Nrf2 pathway will be studied in depth to better understand the skin sensitization process.  

Projects

• LACDR: "Improvement of the barrier properties in reconstructed human skin models”. In collaboration with Prof.dr.J.Bouwstra.
• Nederlands Toxicologisch Centrum (NTC): “Validation of molecular markers of skin sensitization by gene silencing in human keratinocytes and 3D reconstructed skin models”. In collaboration with Prof.Dr H van Loveren (RIVM).

Mimicking skin conditions in-vitro (e.g. wounding, aging)

Since human skin models are easy to modulate, different skin conditions such as scar formation, skin aging and wound healing can be mimicked and studied. For this purpose, cells originating from skin tissues representing a certain physiological condition or disease can be isolated, cultured and incorporated into the human skin model. For example, by isolating cells from skin tissue obtained from young and aged individuals we can mimic several processes that occur during skin aging (e.g. thinning of the epidermis, decrease in collagen deposition). Most of these processes are extensively studied in mouse models or in conventional mono cell layers.

By using human skin models we can mimic a micro-environment that is highly similar to that of the in-vivo human tissue, in contrast to conventional monolayer cultures and mouse models. Cells from aged individuals often are characterized by improper protein expression such as collagenase up-regulation and down-regulation of collagen production. Cells that are "aged" in vitro display similar characteristics (e.g. low collagen I deposition, cell senescence, apoptosis, delayed wound healing). In addition, by exposing the skin models to UV, we can also study processes involved in photo aging by looking for example at DNA-damage. Because skin models are easy to handle, different wounds can be introduced. By eliminating a small fragment of the epidermis, a superficial wound is created. It’s also possible to introduce full thickness wounds in order to mimic burn wounds. This approach allows us to investigate processes that occur in for example burn patients. Burn patients are a prime target for infections with their open wounds and long hospital stays. It has recently been reported that multisystem organ failure (for which an infection was responsible in 45.9% of the cases) is the most frequent cause of death of burn patients.

Furthermore, a surgical skin graft is the usual way to restore burned skin, but with bacterial colonization/infections likely such a procedure is often delayed. Moreover, the infectious agents may not respond optimally to the antibiotic treatment as antibiotic resistant strains are often encountered and/or bacteria may form biofilms. In a collaborative project with the department of infection disease we will investigate the molecular basis of the interactions between bacteria (biofilms) and human cells in (thermally-injured) human skin models co-cultured with bacteria. This will give us the opportunity to engineer an in vitro model that can be used to determine the effects of experimental medication, antibiotics and/or antimicrobial peptides.

Projects

• Cosmetic and Pharmaceutical Industry: “Identification of biomarkers for skin aging using reconstructed human skin models”.
• Nederlandse Brandwonden Stichting (NBS):"New strategies for the prevention and treatment of burn wound infections in skin models”. In collaboration with Dr. P.Nibbering, Infection Disease, LUMC. 

Figure 3: HSE have been wounded with a deep wound (liquid nitrogen) or superficial wound (scalpel). A new epidermis is formed that is active in migration  and proliferation as demonstrated with a hyperprolifertaion  associated marker Keratin 17 in (upper picture).

Different types of Human skin models

Within our department several project are conduct where different types of human skin equivalents are used. Depending on the research questions the following skin models can be engineered: Leiden Epidermal model (LEM), the Full-Thickness model (FTM) and the Fibroblasts-Derived Matrix model FDM). 1) LEM consists of keratinocytes seeded on a non-cellular matrix (e.g. inert filter membrane or de-epidermized dermis. These epidermal models are suitable for e.g. corrosivity, irritation, or penetration tests. 2) FTM consists of keratinocytes, melanocytes and a fibroblast-populated three-dimensional collagen matrix. This model closely resembles native human skin. This full-thickness skin model can be used for tests, predictive screening and research on for example wound healing that requires the complexity of human skin, i.e. where the interaction between epidermal and dermal cells is crucial. 3) FDM is similar to the FTM model, but the dermal compartment consists of human fibroblast-derived extracellular matrix. This model can be used as a tool to evaluate the effect of e.g. ingredients on dermal processes. All three types can be used for predictive screening or contract research that requires the complexity of human skin. Variations of these models can be generated by incorporation of other cell types (e.g. endothelial cells). Models can also be grown at different humidity and oxygen concentrations.

Figure 4:  Different types of skin models.

Principal investigator

Dr. A. El Ghalbzouri 

Human skin equivalents (HSE) are three-dimensional systems that are engineered by seeding fibroblasts into a three-dimensional dermal matrix. After specific culture conditions a HSE is formed that recapitulates most of the in vivo characteristics and in which cellular processes may be normalized compared to the conventional monolayer cultures. HSE are therefore an attractive tool to study cell-cell, cell-matrix, dermal-epidermal interactions and other processes that are involved in epidermal morphogenesis. In addition the HSE are also an excellent tool to mimic diseased skin disorders in vitro (e.g. Epidermolysis Bullosa Simplex and Squamous Cell Carcinoma) in order to test therapeutics.

Figure 5: Shown is a (left) and a cross section of the cultured skin (right).