Multiparameter flow cytometric characterization of epidermal cell suspensions prepared from normal and hyperproliferative human skin using an optimized thermolysin-trypsin protocol

Reliable flow cytometric analysis of normal and diseased skin requires pure epidermal single-cell suspensions. Several methods to separate the dermis from the epidermis are available. The proteolytic enzyme thermolysin separates the epidermis from the dermis at the lamina lucida and therefore permits reliable dermoepidermal separation. In the present study an optimized cell isolation procedure using thermolysin and trypsin is described, which is particularly suitable for punch biopsies. A 16–20-h (overnight) incubation of biopsies taken from normal and hyperproliferative skin with thermolysin (0.5 mg/ml) at 4°C produced a selective separation of the dermis and epidermis. After a 30-min trypsin incubation (0.25 mg/ml) at 37°C a cell suspension was produced which was characterized by minimal cell damage (cellular debris and clumps), a high recovery of basal cells and high quality DNA histograms. Furthermore, dermal contamination was very low. The thermolysin-trypsin separation methodology followed by triple-labelling flow cytometry provided a precise quantification of the percentage of keratin 10-positive cells, vimentin-positive cells and cells in S and G₂M phases. Proliferative activity was selectively measured in the basal, the suprabasal and the non-keratinocyte compartment at various time intervals during epidermal regeneration after adhesive tape stripping. In contrast to the non-keratinocytes, the percentage of cells in S and G₂M phases in the basal keratinocytes and in the suprabasal compartment increased 44–48 h after stripping. The increased proliferation following tape stripping was paralleled by an increased invasion of vimentin-positive cells into the epidermis and preceded by a decreased number of keratin 10-positive cells. Thermolysin-trypsin separation followed by three-colour flow cytometry permits a highly selective characterization of normal and hyperproliferative epidermis.     

In vivo quantification of epidermis pigmentation and dermis papilla density with reflectance confocal microscopy: variations with age and skin phototype

Reflectance confocal microscopy (RCM) may help to quantify variations of skin pigmentation induced by different stimuli such as UV radiation or therapeutic intervention. The objective of our work was to identify RCM parameters able to quantify in vivo dermis papilla density and epidermis pigmentation potentially applicable in clinical studies. The study included 111 healthy female volunteers with phototypes I-VI. Photo-exposed and photo-protected anatomical sites were imaged. The effect of age was also assessed. Four epidermis components were specifically investigated: stratum corneum, stratum spinosum, basal epidermal layer and dermo-epidermal junction. Laser power, diameter of corneocytes and upper spinous keratinocytes, brightness of upper spinous and interpapillary spinous keratinocytes, number of dermal papillae and papillary contrast were systematically assessed. Papillary contrast measured at the dermo-epidermal junction appeared to be a reliable marker of epidermis pigmentation and showed a strong correlation with skin pigmentation assessed clinically using the Fitzpatrick's classification. Brightness of upper spinous and interpapillary spinous keratinocytes was not influenced by the skin phototype. The number of dermal papillae was significantly lower in subjects with phototypes I-II as compared with darker skin subjects. A dramatic reduction in the number of dermal papillae was noticed with age, particularly in subjects with fair skin. The method presented here provides a new in vivo investigation tool for quantification of dermis papilla density and epidermal pigmentation. Papillary contrast measured at the dermo-epidermal junction may be selected as a marker of skin pigmentation for evaluation in clinical studies.     

Comparative study of autoantigens for various bullous skin diseases by immunoblotting using different dermo-epidermal separation techniques

We investigaged the reactivity of pemphigus vulgaris (PV), pemphigus vegetans, pemphigus foliaceus (Pf), Brazilian Pf, bullous pemphigoid (BP), and epidermolysis bullosa acquisita (EBA) sera with an immunoblot analysis using human epidermal and dermal extracts as a source of antigen. To obtain epidermal and dermal extracts three different dermo-epidermal separation methods were used: namely, ethylenediaminetetraacetic acid (EDTA) separation, heat separation, and dispase separation. All the 15PV and the seven pemphigus vegetans sera demonstrated a 130-kDa PV antigen in epidermal extracts obtained by all the three methods. Furthermore, three PV sera also showed a 160-kDa Pf antigen, desmoglein. Ten of 14 Pf sera and six of 15 Brazilian Pf sera reacted with desmoglein in the same pattern in all the three epidermal extracts. Fifteen of the 22 BP sera showed reactivity with 230-kDa BP antigen in the same pattern in all the three epidermal extracts, whereas 14 BP sera delected the 180-kDa BP antigen in extracts of EDTA-and heat-separated epidermis but not in dispase-separated epidermal extract. Dermal extracts were obtained by EDTA- and heat-separated dermis, and all six EBA sera labelled a 290-kDa EBA antigen in both samples. These results suggest that heat-separated skin is as useful as EDTA-separated skin for detecting various autoantigens, but heat separation is preferable because the preparation time is shorter.     

Use of the polymerase chain reaction in quantification of interleukin 8 mRNA in minute epidermal samples.

Quantitative studies of cytokine gene expression in vivo are necessary in order to properly describe the cytokine network and to elucidate its role in skin inflammation. Ideally, one should be able to follow cytokine gene expression in epidermal, dermal, and blood compartments. However, such studies are limited by small amounts of available material. Here we report a polymerase chain reaction (PCR) cDNA amplification protocol useful for quantification of specific mRNAs in small skin samples. We found that analysis of dilution series of each sample permitted establishment of quantitative PCR amplification conditions using only picogram to nanogram amounts of total RNA. Cytokine mRNA amounts could then be measured relative to an internal standard species, co-reverse transcribed, and co-amplified with the cytokine species as a measure of cDNA input. Large numbers of samples can be screened rapidly with initial short dilution series identifying cytokine-positive samples and the correct dilution range for each, followed by closer analysis in this range. Epidermal samples obtained through curettage of a small skin area, 2-mm dermal biopsies from the scraped sites, and a few blood drops from the biopsy sites all yielded sufficient RNA for analysis by this protocol. Any mRNA of known sequence can be studied. We analyzed interleukin 8 mRNA levels in more than a hundred epidermal samples from patients and normal test persons and found a variation over several orders of magnitude that seemed to follow the degree of inflammation of the skin.     

The Caucasian and African skin types differ morphologically and functionally in their dermal component

Abstract:  In the literature, most reported differences between African and Caucasian skin properties concern pigmentation and barrier function of the stratum corneum and related photoprotective properties. However, little is known about differences in morphology and possibly related biological functions. In this study, we investigated: (i) architectural differences of Caucasian and African mammary skin biopsies using microscopy, (ii) comparative constitutive expression of cytokines, matrix metalloproteinase 1 (MMP-1) and its inhibitors in papillary dermal fibroblast (pF) and reticular dermal fibroblast (rF) cultures in order to reveal biological features. (i) Neither epidermis thickness nor superficial dermis thickness was significantly different in African versus Caucasian subjects. However, the dermal–epidermal junction (DEJ) length in African skin was about threefold that in Caucasian skin. No differences were noticed as regards elastic and collagen fibre organization. (ii) In papillary fibroblast cultures, a significantly higher level of monocyte chemotactic peptide-1 (MCP-1) protein was found in cell cultures from African donors when compared with that from Caucasians. With regard to keratinocyte growth factor (KGF), the ratio of papillary to reticular fibroblast expression was found to be twofold greater in cell cultures from African donors compared with that from Caucasian donors. The same trend was found regarding MMP-1 and tissue inhibitor metalloproteinase protein 1 (TIMP-1) protein expression. African skin displays a greater convolution of the DEJ and a higher papillary fibroblast activity. These findings reveal that differences between African and Caucasian skin do not only affect upper epidermis but also dermal functions and dermal–epidermal cellular interactions.     

Possible role of connective tissue in epidermal neoplasia

Interactions between epidermal cells have been defined within a proposed mathematical model of mammalian skin. Testing the model in a computer suggests that in highly proliferating conditions of the epidermis competition for cell space in the basal layer may be sufficient to generate considerable forces in the papillary dermis. Data shown from human and pig epidermal hyperplasia indicate that basal cells are submitted to considerable lateral forces and that these and not dermal hyperplasia are the forces responsible for the increasingly folded dermo-epidermal junction. When the model was examined in condition of persistently high mitolic rate it was found that it could remain stable only if new connective tissue synthesis was not induced by the developing papillary tension. This complex and counterproductive relationship that may occur between epidermis and dermis and its possible role in the development of neoplasia are discussed.     

Basement membrane zone as a target for human neutrophil elastase in psoriasis

Summary Human neutrophil elastase was found, by indirect immunofluorescence using rabbit anti-elastase antiserum, to be bound to basement membrane of psoriatic plaques in vivo. The enzyme was also identified inside the migrating neutrophils in the reticular dermis and dermal papillae, as well as outside the cells in micro-abscesses in psoriatic skin. In vitro incubation of normal skin with human neutrophil elastase resulted in the destruction of hemidesmosomes and separation of the epidermis from the dermis above localizations of bullous pemphigoid antigen. These findings are direct evidence that human neutrophil elastase could play a role in psoriasis in in vivo destruction of the epidermal — dermal junction.     

Thermolysin treatment: a new method for dermo-epidermal separation

The epidermis of superficial human skin samples could easily be separated from the dermis following incubation at +4 degrees C for 1 h in a solution containing 250-500 micrograms/ml thermolysin, a proteolytic enzyme hitherto mostly used for protein analysis. Light and electron microscopy revealed that the dermo-epidermal separation occurred at the basement membrane between the sites of bullous pemphigoid antigen and laminin and that the hemidesmosomes were selectively disrupted. The cohesion and morphology of the separated epidermis as well as the immunologic parameters investigated were not altered by this procedure. The clear cut dermo-epidermal separation produced by thermolysin treatment differed from the separation obtained with trypsin, which predominantly occurred between basal and suprabasal cells by disruption of desmosomes.     

Epithelium-lining macrophages in psoriasis

Summary Epithelium-lining macrophages are spindle-shaped cells which line the epidermis and hair follicles. We studied the distribution and phenotype of this hitherto neglected member of the dermal monocyte/macrophage system in 25 lesional psoriatic, and five normal skin biopsies. Epithelium-lining macrophages were inconspicuous in normal skin, whereas their number was increased in almost two thirds of psoriatic cases; in nine out of 25 lesional skin biopsies, these flattened cells formed an almost continuous single-cell row at the dermo-epidermal junction.
Immunophenotyping revealed that these cells expressed the leucocyte common antigen CD45. and the macrophage markers CD14, CD36 and CD4, but not CD11b. Epithelium-lining macrophages strongly expressed HLA-DR-antigens and CD 11a, but lacked the Langerhans cell marker CD1, and CD34. The dermal dendrocyte marker factor XIIIa was expressed in only a minority of these cells.
It is concluded that epithelium-lining macrophages represent a separate subset of dermal monocytes/macrophages with a distinct tissue localizaton and immunophenotype. Their restricted distribution and close association with the epidermis may suggest a role in the regulation of epidermal growth. Alternatively, the expression of several immune-associated molecules may indicate that epithelium-lining macrophages are involved in the antigen-dependent or-independent activation of T cell.     

Microanatomical compartments of clonal and reactive T cells in mycosis fungoides: molecular demonstration by single cell polymerase chain reaction of T cell receptor gene rearrangements

Mycosis fungoides (MF) is a cutaneous T cell lymphoma, clinically characterized by patches, plaques and tumors occurring in successive stages of the disease. In early MF, an infiltrate consisting of mainly reactive T cells is seen in the papillary dermis while tumor cells are mostly confined to the epidermis. By contrast, later stages show nodular infiltrates formed mostly of tumor cells in the dermis while the epidermis is relatively devoid of tumor cells; however, knowledge of the localization of clonal T cells has been based on histomorphologic features and immunohistochemical stainings visualizing certain V-beta subfamilies of the T cell receptor (TCR). As these techniques do not allow for an unequivocal identification of clonal tumor cells, we used micromanipulation and single cell PCR amplifying the TCR chain gene rearrangement. A total number of 387 single T cells was isolated from six skin biopsies in five patients in patch, plaque, and tumor stages. Of these, 180 T cells were picked from the epidermis and 207 from the dermal infiltrate. The rearranged TCR-gamma DNA could be sequenced from 181 of 387 T cells. In three of six patients representing all three stages, epidermal T cells with a clonal rearrangement could be amplified. In early plaque stage a higher degree of epidermal T lymphocytes was found than in initial patch, later plaque, and tumor stages with an inverse distribution found for reactive T lymphocytes. In two patients a biallelic rearrangement was demonstrated that had not been detected in prior PCR analysis from blood and skin samples. These data show that clonal (neoplastic) and non-clonal (reactive) T lymphocytes in MF preferentially infiltrate different microanatomical compartments of the skin, depending on the stage of disease. The microanatomically distinct localization of reactive and clonal T cells suggests that the absence of direct contact between tumor and host-defense lymphocytes may contribute to tumor persistence and progression in epidermis, peripheral blood, and deep dermal tumor cell nests, respectively.     

Inducible nitric oxide synthase in pityriasis lichenoides lesions

Background:  Pityriasis lichenoides (PL) is an inflammatory skin disease of unknown etiology. Nitric oxide (NO) has emerged as an important mediator of many physiological functions. The importance of NO-mediated signaling in skin diseases has been reported by several studies.
Methods:  A review of clinical records and histopathological slides of 34 patients diagnosed with PL was performed. Three different groups of skin biopsies including PL chronica (24 patients), PL et varioliformis acuta (10 patients) and 15 normal skin samples were subjected to the immunohistochemistry technique for inducible nitric oxide synthase (iNOS) detection.
Results:  Normal skin group exhibited a few number of iNOS-positive cells in the dermis and rare positive cells in the upper epidermis, unlike abundant epidermal and dermal iNOS expression observed in both PL groups.
Conclusion:  According to our results, we hypothesize that NO produced by iNOS could participate in PL pathogenesis. Abnormal and persistent responses to unknown antigens, probably a pathogen, associated with NO immunoregulatory functions could contribute to the relapsing course observed in PL. NO anti-apoptotic effect on T-cell lymphocytes could play a role on maintenance of reactive T cells, leading to a T-cell lymphoid dyscrasia.     

Detection of human immunodeficiency virus-DNA and RNA in the skin of HIV-infected patients using the polymerase chain reaction.

The presence of HIV genomic-associated nucleic acids (DNA and RNA) within biopsies of normal-appearing skin and various skin lesions obtained from a group of 33 HIV-infected patients was investigated by using the polymerase chain reaction (PCR). In order to define the localization (dermal vs. epidermal) of HIV, the PCR was carried out separately on the dermis and the epidermis in 21 of the specimens. Altogether, HIV-DNA and HIV-RNA were detected, respectively, in 89% and 47% of the specimens included in this study; both DNA and RNA were detected more frequently in the dermis (90% and 43%, respectively) than in the epidermis (62% and 5%, respectively). No correlation could be established between the presence of HIV genomic material, the nature (normal-appearing vs. diseased) of the skin specimen studied, and the clinical or biologic severity of HIV infection, as evidenced by the CDC stage classification and the number of peripheral CD4+ cells. It seems, therefore, that the HIV is very frequently present within the skin during the course of HIV infection; however, its precise cellular localization and pathologic significance await further investigation.     

FGF expression allows nevus cells to survive in three-dimensional collagen gel under conditions that induce apoptosis in normal human melanocytes.

Melanocytes, the pigment forming cells of the skin, form an almost nonproliferating cell population located to the lowermost part of the epidermis. Normally melanocytes are not found higher in the epidermis or in the dermis. Nevi consist of melanocytes with altered growth characteristics and localization. The common pigmented nevus, a benign skin lesion, develops when melanocytes proliferate in the dermo-epidermal junction or in the dermis. Here we report growth characteristics of in vitro cultured normal human melanocytes and dermal nevus-derived melanocytes. As previously reported, nevus cells have a moderate to high FGF-2 expression level. Here we demonstrate that dermal nevus cells are able to survive in three-dimensional type 1 collagen culture, while normal human melanocytes rapidly undergo apoptosis. Melanocytes also, however, survive in collagen cultures in the presence of exogenous FGF-2. The survival of nevus cells in collagen is suppressed by protamine, an inhibitor of FGF-mediated cell stimulation. The in vivo growth environment of dermal nevus cells consists largely of type I and type III collagens. The results suggest that FGF-2 expression by nevus cells allows them to adapt to grow in the dermis. FGF-2 obviously has importance as a melanocyte survival factor and probably also in the development of malignant melanoma.     

Neuropeptides and general neuronal marker in psoriasis an immunohistochemical study

Nerve fibres immunoreactive to antibodies to vasoactive intestinal polypeptide (VIP) and substance P (SP) were increased in lesional psoriatic skin when assessed semi-quantitatively. Biopsies from psoriatic plaques on the arm were studied in 13 patients and compared with biopsies from non-lesional areas (in three of the same psoriatic subjects) and from normal skin in seven non-psomtic controls. Immunohistochemical methods were used on cryocut skin sections to demonstrate the neuropeptides SP, VIP, calcitonin gene-related peptide and neuropeptide Y, and the general neuronal marker protein gene product (PGP) 9.5. The immunofluorescence was examined by semiquantitative and, for PGP 9.5, by quantitative methods. VIP reactive nerve fibres were increased at areas of eccrine sweat glands throughout the dermis, at the dermo-epidermal junction, and in the epidermis, in psoriasis lesional skin. SP reactive nerve fibres were increased at the dermo-epidermal junction, where the nerves ran parallel with and perpendicularly through the junction. PGP 9.5 reactive nerve fibres showed an increase at the dermo-epidermal junction, in the papillary dermis, and at the eccrine sweat glands in lesional psoriatic skin but not in non-lesional, or in control skin. These findings support the hypothesis that neuropeptides may be involved in the pathogenesis of psoriasis.     

UV-induced DNA damage initiates release of MMP-1 in human skin

Destruction of collagen is a hallmark of photoaging. The major enzyme responsible for collagen 1 digestion, matrix metalloproteinase-1 (MMP-1), is induced by exposure to sunlight. To study the molecular trigger for this induction, human skin was ultraviolet-B (UVB)-irradiated and treated with liposome-encapsulated DNA repair enzymes. The photolyase-mediated DNA repair of epidermal UV damage was associated with a reduction of MMP-1 mRNA and protein expression in both the epidermal and dermal compartments of the skin. The role of the epidermal cells in MMP-1 induction in the fibroblasts was examined when human epidermal keratinocytes were irradiated with UVB and their media were transferred to unirradiated human dermal fibroblasts. Transfer of media from irradiated keratinocytes to unirradiated fibroblasts enhanced MMP-1 mRNA and protein. Thus, UV damage to keratinocytes of the epidermis may participate in the destruction of collagen in the dermis by release of soluble mediators that signal fibroblasts to release MMP-1. The MMP-1 induction was reduced when the keratinocytes were treated with DNA repair enzymes T4 endonuclease V or UV endonuclease prior to transfer of the media to fibroblasts. This implies that UVB, which deposits most of its energy on the chromatin of the epidermal keratinocytes and to a lesser extent in the upper dermis, has a significant role in photoaging. DNA damage in the keratinocytes initiates one of the signals for MMP-1 release, and enhancing DNA repair can reduce MMP-1 expression in human skin cells and tissue.     

Characterization of the glucocorticoid receptor in human epidermis and dermis

The binding of [³H]triamcinolone acetonide to cytosol proteins in human epidermis and dermis has been characterized. Qualitatively the binding in both cytosols was very similar. The binding was of high affinity (dissociation constant = 1·07 × 10^?9 mol/l and 1·36×10^?9mol/l for epidermal and dermal binding proteins respectively) and low capacity (27·3 fmol/mg protein and 53·0 fmol/mg protein for epidermal and dermal binding proteins respectively). The binding protein was labile m heating at 40°C and proteolytic enzymes. Glucocorticoids showed an affinity which approximated to their therapeutic potency in both epidermis and dermis and non-glucocorticoids had little or no binding affinity. Quantitatively more than 90% of the binding protein in whole skin was present in the dermis and less than 10% in the epidermis, which approximates to a difference in tissue mass. There were 12·O× 10¹⁰ receptors in a disc of whole skin of 1 cm² surface area of which 11·0×10¹⁰ were in the dermis, 0·95 × 10¹⁰ in the epidermis with 1706 and 1214 receptor sites for epidermal and dermal cells respectively. Receptor concentration is highest in dermis when expressed per cytosol protein and in epidermis when expressed per DNA. It is concluded that the binding protein characterized in epidermis and dermis is the physiological glucocorticoid cytosol receptor. In the absence of qualitative differences between the epidermal and dermal receptors quantitative differences may help in the development of less toxic therapeutic agents.     

The localization of the target antigens and antibodies in linear IgA disease is heterogeneous, and dependent on the methods used

Fifty-nine patients with linear IgA disease, 24 with onset in childhood and 35 with adult onset, were studied. Sera from all patients were tested by indirect immunofluorescence, using as substrates intact normal skin and normal skin which had been split through the lamina lucida region of the basement membrane zone by suction and by prolonged incubation with molar NaCl. This enabled the site of the target antigen for the circulating IgA antibodies to be determined. The sites of deposition of the IgA antibodies in vivo were detected by raising a suction blister in eight patients, and splitting seven patients’ biopsies by prolonged incubation with molar NaCl. Eighteen sera were positive with intact skin, and 34 with split skin. Twenty-nine sera were positive with suction blisters as substrate; 14 bound to the epidermal aspect of the split skin, seven in a combined pattern (binding to the epidermis and dermis) and six to the dermal aspect. Thirty-one sera bound to salt-split skin, 24 to the epidermal side and seven on the dermal side. There was discordance between the two methods of skin splitting in 15 sera. Seven sera gave a combined pattern with suction but with salt-split skin, five of these bound epidermally, one was dermal, and one negative. Five sera showed epidermal binding on saltsplit skin and were negative on suction blisters, and the reverse was seen with one serum. Two sera gave variable results on suction blisters. Direct immunofluorescence studies showed dermal binding on all eight patients with suction blisters, and epidermal binding in four and dermal binding in three patients with salt splitting. These results demonstrate that the location of the target antigens and the sites of deposition of the antibodies are dependent on the methods used. They also suggest that there are at least two different antigens, an epidermal- and a dermal-associated antigen. The sera reacting in the combined pattern may represent antibodies reacting with a different epitope of the epidermal antigen, with a further epidermal antigen, or with two target antigens.