Ultrastructural study of hydrocortisone action in cultured human epidermis

Application of stereology to dermatological research

Stereology is a set of mathematical and statistical tools to estimate three-dimensional (3-D) characteristics of objects from regular two-dimensional (2-D) sections. In medicine and biology, it can be used to estimate features such as cell volume, cell membrane surface area, total length of blood vessels per volume tissue and total number of cells. The unbiased quantification of these 3-D features allows for a better understanding of morphology in vivo compared with 2-D methods. This review provides an introduction to the field of stereology with specific emphasis on the application of stereology to dermatological research by supplying a short insight into the theoretical basis behind the technique and presenting previous dermatological studies in which stereology was an integral part. Both the theory supporting stereology and a practical approach in a dermatological setting are reviewed with the aim to provide the reader with the capability to better assess papers employing stereological estimators and to design stereological studies independently.

Alternative model to human skin organ culture: A preliminary study with Leibovitz L15 medium

Organ culture has been used to maintain the three-dimensional structure of the skin and the interaction between melanocytes and keratinocytes, which is essential for melanin production. In the current study we aimed to evaluate the general morphology, viability, and distribution of human melanocytes in a system that uses Leibovitz L15 medium at room temperature. By comparison with human skin explants maintained in Dulbecco's minimum Eagle's medium at 37 degrees C, we found that the skin was better preserved with Leibovitz L15 after 7 days in culture. The addition of 10% fetal bovine serum to this medium did not promote any change. Dividing cells labeled with Ki-67 were visualized at the basal and suprabasal epidermal layers. Retinoic acid was tested at 1 microg/mL and we recorded a reduction of the corneal layer after 48 hours and a complete detachment of the epidermis after 7 days, probably due to a toxic effect in the medium. Melanin and melanocytes were detected by ammoniacal silver and Dopa stainings under light microscopy. We observed that cells were viable throughout the culture period and melanin was distributed in melanocytes and keratinocytes. In conclusion, we suggest that the use of Leibovitz L15 medium at room temperature can be a viable alternative to the normal organ culture of human skin, which is an important system to study the activity and reaction of melanocytes to dermatological products and cosmetics.

CD44 substituted with heparan sulfate and endo-beta-galactosidase-sensitive oligosaccharides: a major proteoglycan in adult human epidermis

CD44 is a group of cell surface glycoproteins that is generated from a single gene by mRNA splice variation. Its functions in matrix adhesion and tumor invasion are strongly influenced by glycosylation. We studied the glycosylated tissue forms of CD44 from extracts of normal adult human epidermis by using western blotting and immunoprecipitation from short-term skin organ cultures. An antibody for CD44 (Hermes 3) precipitated 7-17% of all 35SO4-labeled proteoglycans (PGs) synthesized in epidermis. Immunoprecipitates digested with heparitinase lost 40-68% of incorporated 35SO4 and 24-40% of [3H]glucosamine, indicating that heparan sulfate was the predominant glycosaminoglycan in epidermal CD44. Chondroitinase ABC released 10-25% and 6-12% of 35SO4 and [3H]glucosamine, respectively. Less than 5% of both isotopes were susceptible to keratanase. Five to 33% of 35SO4 and 26-37% of [3H]glucosamine, however, was released by endo-beta-galactosidase, implying marked substitution by oligosaccharides with N-acetyllactosamine repeats. Heparitinase pretreatment retarded, whereas endo-beta-galactosidase enhanced the mobility of the > or = 180-kDa polydisperse CD44 on agarose gel electrophoresis. On SDS-polyacrylamide gel electrophoresis, however, western blotting and fluorographs of 35SO4-labeled immunoprecipitates showed the main CD44 isoform at > or = 250 kDa and a shift to 180-200 kDa after heparitinase treatment. Keratanase, keratanase II, and chondroitinase ABC had minor effects. A less abundant form of CD44, with a core of 100 kDa, partly substituted with chondroitinase ABC- and endo-beta-galactosidase-sensitive chains, was also present. Therefore, the large heparan sulfate-substituted CD44 forms a significant part of all proteoglycans in normal human epidermis. Both the large and the 100-kDa variant of epidermal CD44 contain endo-beta-galactosidase-sensitive oligosaccharides not previously noted in other cells or tissues.

Hydrocortisone regulation of hyaluronan metabolism in human skin organ culture

We studied the influence of hydrocortisone (HC) on hyaluronan (HA) metabolism in explants of human skin, a model retaining normal three-dimensional architecture of dermal connective tissue and dynamic growth and stratification of epidermal keratinocytes. The synthesis of hyaluronan and proteoglycans (PGs), and DNA, were determined with 3H-glucosamine and 3H-thymidine labelings, respectively. The total content and histological distribution of hyaluronan was studied utilizing a biotinylated aggrecan-link protein complex. A low concentration of HC (10(-9) M) stimulated the incorporation of 3H-glucosamine into hyaluronan in epidermis by 23% and reduced the disappearance rate of hyaluronan by 25% in chase experiments, resulting in a 74% increase in total hyaluronan (per epidermal dry weight) after a 5-day culture in 10(-9) M HC. On the other hand, a high concentration of HC (10(-5) M) reduced both synthesis (-42%) and degradation (-46%) of epidermal hyaluronan during 24 h labeling and chase periods. The cumulative effect of a 5-day treatment was a 24% decrease of total epidermal hyaluronan. The high dose (10(-5) M) also reduced keratinocyte DNA synthesis and epidermal thickness. In dermis, only the high (10(-5) M) concentration of HC was effective, inhibiting the incorporation of 3H-glucosamine into hyaluronan by 28%. No significant influences on total hyaluronan content or the disappearance rate of hyaluronan in dermal tissue was found. All HC concentrations lacked significant effects on newly synthesized PGs in epidermal and dermal tissues, but reduced the labeled PGs diffusing into culture medium. A low physiological concentration of HC thus maintains active synthesis and high concentration of hyaluronan in epidermal tissue, while high pharmacological doses of HC slows hyaluronan turnover and reduces its content in epidermis, an effect correlated with enhanced terminal differentiation, reduced proliferation rate and reduced number of vital keratinocyte layers.

Organotypic culture of human skin to study melanocyte migration

An ex vivo model system was developed to investigate melanocyte migration. Within this model system, melanocytes migrate among other epidermal cells in the epibolic outgrowth of skin explants. This process is initiated by loss of contact inhibition of epidermal cells at the rim of the explants and by locally produced chemotactic factors. Punch biopsies provided explants of reproducible diameter. Optimal culture conditions include medium consisting of Dulbecco's Minimal Essential Medium containing 10% inactivated normal human serum and placement of explants epidermal side up at the air-liquid interphase. Within 7 days, epidermal cells completely surround the explant. Approximately 3 days after the onset of keratinocyte migration, melanocytes distribute themselves within the newly formed epidermis. Throughout the 7-day culture period, melanocytes and keratinocytes show maintenance of subcellular morphology, and the dermo-epidermal junction remains intact. Melanocyte migration was quantified using immunoperoxidase staining in combination with light microscopy and computer-aided image analysis. Preliminary results using the model system to compare migration in control and nonlesional vitiligo skin indicate that no inherent migration defect is responsible for impaired repigmentation of vitiligo lesions. The organotypic culture model system allows for investigations on melanocytes within their environment of autologous epidermal and dermal components, closely resembling in vivo circumstances in human skin.

Modulation of epidermal terminal differentiation in patients after long-term topical corticosteroids

The expression of the various markers for terminal epidermal differentiation in atrophic skin of patients after long-term topical corticosteroids (TCS) was studied by electron microscopy, immunofluorescence using antibody to profilaggrin/filaggrin (PF/FG), immunoperoxidase staining using antibody to involucrin, and oil red O stain for neural lipids of the stratum corneum. Thirty-nine patients were subdivided into two groups: (A) 19 patients suffering from rebound phenomenon after stopping TCS and (B) 20 patients without rebound phenomenon. Biopsy specimens were taken before ending the use of TCS in both groups. In group A, both the morphological markers (including the different epidermal strata, keratohyalin granules, lamellar granules, and cornified cell envelopes) and the molecular markers (including involucrin, PF/FG, and neutral lipids) of terminal epidermal differentiation were significantly suppressed. On the other hand, the differentiational markers in the atrophic skin of patients without rebound phenomena were only slightly altered. These results suggest that potent TCS not only has antiproliferative actions but also inhibits the differentiation of epidermis, resulting in structural defects in the epidermis, especially the stratum corneum.

Two-compartment model for rabbit skin organ culture

A new method was developed for rabbit skin organ culture. In a two-compartment model, skin discs were cultured on a Millicell-HA insert unit with a microporous membrane which allows transport of culture medium via the dermis into the epidermis, whereas the epidermal side remains free of direct contact with culture medium. In this relatively simple two-compartment organ culture model, rabbit skin could be cultured for 7 d in RPMI 1640 medium supplemented with fetal bovine serum, or for 2 d in RPMI 1640 medium supplemented with cofactors. The histomorphology and ultrastructure of 7-d cultured rabbit skin discs was essentially similar to that of freshly isolated rabbit skin. Keratinocytes in the stratum basale continued to divide during organ culture. The terminal differentiation of the epidermis continued in vitro as was found by the presence of keratohyalin granules, the intact stratum corneum, and keratin expression. Furthermore, glucose consumption continued until culture Day 7, but thereafter it declined rapidly. Concomitantly, degenerative changes were found. At the end of the 7-d culture period the distance between single dermal collagen fibrils had increased as compared to noncultured skin. This model of skin organ cultures can be used to study biological processes, dermal toxicity, and penetration and metabolism of xenobiotics in intact skin. Furthermore, within certain limits, processes responsible for repair and regeneration of damaged skin can also be studied in this model because the rabbit skin can be cultured for 7 d.

A fully differentiating epidermal model with extended viability: development and partial characterization

A highly differentiated porcine skin organ culture model has been developed for future investigations of membrane-coating granules (MCG) and their role in epidermal differentiation. In contrast to many previous systems, cultures do not undergo necrosis of the upper epidermis or display dermo-epidermal separation, but survive for at least 3 weeks, at which time mitotic cells are still evident. Although rete projections are gradually smoothed out and the viable epidermis thins at a rate of approximately 0.35 cells per day, the stratum corneum gains approximately 1.5 corneocytes per day. Furthermore, at 3 weeks all the major differentiation markers are expressed, including keratohyalin granules, MCG, and an orthokeratotic stratum corneum. The system is inexpensive, simple to establish, and does not require elevated oxygen levels. The main requirements are 1) the use of Dulbecco's minimal essential medium supplemented with 2) hydrocortisone (100 micrograms/ml), 3) growth at an air/liquid interface, and 4) attached connective tissue. The further addition of vitamin C (300 micrograms/ml) and/or bovine serum albumin (2 mg/ml) offered no obvious advantage. Degeneration of organ cultures in standard cell culture media was discovered to be caused by fetal bovine serum (FBS). FBS-induced degeneration was not prevented by adding any of the supplements tested, or the inclusion of 3T3 fibroblasts, even when culturing at an air/liquid interface. Complete submersion rapidly killed specimens, presumably through oxygen starvation. The ability to maintain a fully keratinizing system for several weeks, in a totally chemically defined medium, will prove valuable for research not only into the role(s) of MCG in epidermal biology but also studies of desquamation and epidermal differentiation.

Morphometric analysis of human epidermis treated with testosterone and dehydroepiandrosterone in organ culture

The influence of testosterone and dehydroepiandrosterone (DHA) on human epidermal differentiation in skin organ culture were studied with quantitative light and electron microscopic morphometric methods. The hormones had equal effects on the general tissue architecture at the concentrations of 10(-7) mol/l: they had a negligible effect on the thickness of the vital and cornified layers of epidermis, and on the type of keratinization, but caused a significant increase in the number of granular cells. The electron microscopic stereologic analyses showed that testosterone did not alter the size of either vital or cornified epidermal cells. Neither did it influence the quantities of major keratinocyte organelles (keratin filaments, desmosomes, ribosomes, mitochondria) in the different epidermal strata. However, the amounts of keratohyalin granules and keratinosomes were significantly higher in testosterone-treated granular cells than in controls. Qualitative electron-microscopic analysis of DHA-treated skin showed a similar change in the amount of keratohyalin granules and keratinosomes. The present findings suggest that although androgenic hormones exert little or no direct anabolic effect on epidermis, they may modulate keratinocyte maturation.

Ultrastructural localization of keratinocyte surface associated heparan sulphate proteoglycans in human epidermis

Fixation and staining procedures were developed for the electron microscopic demonstration of glycosaminoglycans (GAGs) in human epidermis. En bloc staining with cuprolinic blue (CB), ruthenium red (RR) and tannic acid (TA) in the primary fixative were applied for the localization of the GAGs. Removal of the epidermal basal lamina and underlying dermis was a prerequisite for stain penetration. In CB-fixed specimens 50 nm long, rod-like granules were found attached to keratinocyte cell surfaces, while the RR- and TA-fixed specimens contained round granules (luminal diameter 10 and 30 nm, respectively). The stainability of the CB-positive granules in the presence of 0.3 mol/l MgCl2 indicated that they contained sulphated GAGs. Prefixation digestions of epidermal sheets with chondroitinase ABC. Streptomyces hyaluronidase, and heparitinase showed that the RR-positive granules also contained sulphated GAGs, mostly heparan sulphate. The granules visualized with TA on keratinocytes were susceptible to heparitinase treatment, but the abundance of TA-staining suggested that TA also stained structures other than heparan sulphate. The EM data was in accordance with the 35SO4 labelling experiments showing that heparan sulphate was the major sulphated GAG synthesized in epidermis, whereas chondroitin/dermatan sulphates comprised about one fifth of the total activity incorporated. The distributions of the CB-, RR- and TA-positive granules on cell surfaces were similar. The morphology of the proteoglycan granules was probably determined by the extent of the GAG-chain collapse following binding to each of the dyes.

Influence of retinoic acid on the ultrastructure and hyaluronic acid synthesis of adult human epidermis in whole skin organ culture

Normal human skin was maintained in organ culture under chemically defined conditions. All-trans retinoic acid was added to the culture medium at the final concentration of 5 mumol/l. After 5 days in culture samples were either harvested for electron microscopy or labeled with 3H-glucosamine for 24 h. After labeling, epidermis was separated from dermis and both tissue compartments were analyzed for the content of 3H-labeled glycosaminoglycans (GAGs) using CPC-precipitation and thin layer chromatography after enzymatic degradation into specific disaccharides. Retinoic acid caused a marked change in the epidermal tissue architecture. The epidermal cells were flattened and contained fewer desmosomes and tonofilaments than control explants. Retinoic acid induced accumulation of fine granular material in the intercellular spaces in the upper, and less dense, flocculent material in the lower epidermis. The analysis of 3H-glycosaminoglycans showed that in the epidermis retinoic acid elevated the amount of labeled hyaluronate by 70%, whereas sulfated GAGs were not significantly increased. In dermis the incorporation of 3H-glucosamine into neither hyaluronate nor sulfated GAGs was stimulated by the retinoic acid. It is concluded that retinoic acid significantly modifies the differentiation of normal adult human epidermis by decreasing cytoskeleton components and by inducing the synthesis of new intercellular material, at least a part of which is hyaluronic acid. As a consequence, the cohesion between the epidermal cells was apparently weakened.

Effects of retinoic acid on adult human epidermis in whole skin organ culture

Adult human skin was cultured in whole-skin organ culture under chemically defined conditions. Retinoic acid was added to the culture at final concentrations of 5 X 10(-7) and 5 X 10(-6) M. Both concentrations elicited cell death in the upper epidermal layers and prevented the terminal differentiation of the cells to mature corneocytes. The inhibition of terminal differentiation was not permanent, as the corneocytes produced later during the culture showed no signs of inhibition. The upper vital cells in epidermis cultured with retinoic acid were very flattened and contained reduced amounts of cytoskeleton components. Fine, granular material not present in normal epidermis was observed in both the intercellular spaces and the intracytoplasmic vesicles of retinoid-treated epidermis. The present results indicate that the response of normal human skin to retinoic-acid treatment involves the same kind of modulation of the epidermal structure previously described in embryonic avian and diseased human skin.