Psoriasis is an inflammatory skin disorder, suggested to be triggered or exacerbated by a number of genetic, environmental, or immunological factors. The disease is characterized by hyperproliferation of keratinocytes; abnormal epidermal differentiation; and infiltration of inflammatory cells (Stern, 1997). These include Th1 cells that have been activated by dermal dendrocytes and neutrophils in a dense focal accumulation in the upper epidermis (Gillitzer et al, 1996). Pro-inflammatory cytokines and chemokines, released from multiple cell types, interact in a complex pattern to influence the pathological events leading to the development of the psoriatic plaque (Prens et al, 1990;Bata-Csorgo et al, 1995;Gillitzer et al, 1996;Mizutani et al, 1997;Fukuoka et al, 1998).
Research into psoriasis, and the subsequent development of therapeutic strategies, has been hindered by the absence of relevant models of the disease. The severe combined immunodeficiency (SCID) mouse has been used as an in vivo model, in which psoriatic plaques transplanted onto the animal have maintained the clinical, histological, and immunological characteristics of the disease for sufficient time to allow the efficacy of potential therapeutic compounds to be tested (Raychaudhuri et al, 2001). Animal models, however, can prove unsatisfactory, as allogeneic differences can cause artifactual immune responses, furnished by the xenograph, which may differ from that of psoriasis. Various attempts to model aspects of the disease in vitro have been reported, using both monolayer and stratified keratinocyte cultures (Saiag et al, 1985;Chapman et al, 1990;Mils et al, 1994;Van Ruissen et al, 1996a,b;Pol et al, 2002). Some previous studies have demonstrated the importance of crosstalk between the keratinocyte and the fibroblast in modeling the disease in vitro (Krueger and Jorgensen, 1990;Konstantinova et al, 1996;Krueger and Jorgensen, 1997). An organotypic culture system that accurately models the diseased tissue architecture, with the added potential of the introduction of multiple cell types, however, has not been previously characterized. A defined, serum-free environment is particularly advantageous, by allowing the precise control of experimental conditions.
The studies described here were carried out to compare proliferation and gene expression in skin models derived from normal cells with those derived from involved and uninvolved psoriatic cells, with the aim of developing a robust in vitro model of psoriasis that could be used as a valuable tool in basic research and drug discovery. Genes investigated were relevant to keratinocyte proliferation (Ki67), hyperproliferation (cytokeratin 16), and differentiation (cytokeratin 10, involucrin, and filaggrin) or have previously been implicated in the pathophysiology of psoriasis, i.e., primary cytokines (TNF-
, IFN-
, IL-6) and the chemokine receptors CXCR1 and CXCR2, and their ligands IL-8 and GRO-.
Results
Patient demographics
All of the 12 patients who were biopsied had chronic plaque psoriasis. Their mean age was 50 y (range 34–75 y), nine were female, mean duration of psoriasis was 19 y (3–40 y) and mean psoriasis area and severity index(PASI) at biopsy was 9.7 (2.6–22.2).
Reconstruction of normal and psoriatic skin in vitro
Skin models derived from normal, involved psoriatic, and uninvolved psoriatic cells models were grown from keratinocyte and fibroblast cultures with <0.1% contaminating CD3 or CD45 positive cells as determined by immunocytochemistry. All displayed excellent epidermal morphology using the defined, serum-free culture system (Figure 1). A compact basal layer gave rise to larger spinous layer of differentiating cells. Several flattened cell layers formed the granular layer, containing keratohyalin granules, beneath the stratum corneum. The stratum corneum was in most cases anuclear; however, occasionally models of all types displayed some parakeratosis. All of the major morphological layers were evident by day 7; continuation of the culture period to day 21 resulted in a more highly differentiated epidermis, with a progressively thicker cornified layer. Dermal fibroblasts in the collagen gel aligned horizontally with the basement membrane. No obvious differences in epidermal thickness were observed between normal and psoriatic models. Ki67 staining present in the basal and lower suprabasal layers of the models, however, showed a significant increase in proliferation in the psoriatic models compared to the normal controls over 7, 14, and 21 d p<0.02, n=4 (Figure 2). No difference in proliferation was observed between the involved and uninvolved models. Figure 2 also shows that the proliferation rates were similar between 7 and 14 d but were significantly lower after 21 d in both normal and psoriatic models.
Figure 1.
Reconstruction of human skin in vitro in a fully defined, serum-free environment. Keratinocytes were seeded onto collagen gels containing dermal fibroblasts and grown to a confluent monolayer. Subsequent exposure to the air–liquid interface for 21 d resulted in the formation of a fully differentiated epidermis. Figure shows typical morphology of normal, psoriatic involved and psoriatic uninvolved models after 7, 14, and 21 d of air exposure.
Full figure and legend (231K)Figure 2.
Proliferation of keratinocytes in normal and psoriatic models of human skin. Figure shows the mean number of Ki67 positive cells in normal (open columns), involved (black filled columns) and uninvolved (gray filled columns) psoriatic models cultured for 7, 14, and 21 d. Models derived from involved or uninvolved psoriatic skin showed higher rates of proliferation compared to normal models p<0.02 n=4. Proliferation rates were similar after 7 and 14 d culture but were significantly reduced after 21 d for all cases tested p<0.002. Error bars represent+SD.
Full figure and legend (9K)Patterns of gene expression (Figure 3) in psoriatic uninvolved cultures were the same as those for involved cultures for all genes investigated (uninvolved data not shown). Cytokeratin 16 was not detected in normal skin biopsies (in vivo) but was strongly expressed in suprabasal psoriatic keratinocytes in skin models from normal, psoriatic–involved, and psoriatic-uninvolved sources. Also indicative of the hyperproliferative phenotype was the precocious expression of involucrin, which was present in all suprabasal keratinocytes in all in vitro skin models and in psoriasis in vivo, although localized to the upper spinous and granular layers in normal skin. In contrast, the expression patterns of cytokeratin 10 and filaggrin were not altered in psoriasis in vivo and again this was reflected in vitro: cytokeratin 10 was consistently expressed in all suprabasal keratinocyte layers, whereas filaggrin was localized to the granular layer.
Figure 3.
Markers of keratinocyte proliferation and differentiation in normal and psoriatic models of human skin. Figure shows expression of Ki67, cytokeratin 16, cytokeratin 10, involucrin and filaggrin in organotypic models of normal and psoriatic (involved) human skin after 21 d of air exposure. Models derived from uninvolved psoriatic skin showed identical patterns of gene expression to those derived from involved skin (data not shown). In vivo normal and psoriatic (involved) skin is shown for comparison.
Full figure and legend (163K)Characterization of the psoriatic phenotype in vitro
Primary cytokines
No expression of TNF-, IFN- or IL-6 was observed in normal skin, either in vivo or in vitro (Figure 4). TNF- and IFN- were localized to the granular layer keratinocytes in psoriatic lesions in vivo, particularly around microabcesses, with weak expression also in the spinous layer. Some staining in mononuclear cells was observed around the dermal papillae. The psoriatic skin models showed strong staining for both TNF- and IFN- throughout the keratinocyte layers, with intense staining in the granular layer. IL-6, in contrast, showed general staining throughout the basal and suprabasal keratinocyte layers of both normal and psoriatic skin, in vivo and in vitro. Occasional keratinocytes displayed intense cytoplasmic staining, particularly in the in vitro models. The lowest level was seen in normal skin in vivo. Fibroblasts showed strong staining, particularly in psoriatic in vitro skin.
Figure 4.
Expression of primary cytokines in normal and psoriatic human skin models. Figure shows expression of TNF-, IFN-, and IL-6 in models of normal and involved psoriatic human skin after 21 d of air exposure. Models derived from uninvolved psoriatic skin showed identical patterns of gene expression to those derived from involved skin. In vivo normal and involved psoriatic skin are also shown for comparison.
Chemokines and chemokine receptors
Expression of the interleukin-8 receptors, CXCR1 and CXCR2, are shown in Figure 5. There was absence of CXCR1 in keratinocytes both in vivo and in vitro. This receptor, however, was clearly expressed in infiltrating inflammatory cells in vivo. Figure 5 shows some striking differences in CXCR2 expression between normal and psoriatic skin, both in vivo and in vitro. CXCR2 was not detectable in normal in vivo skin and in the normal in vitro models. In contrast, in psoriatic skin, CXCR2 was expressed in the suprabasal keratinocytes, with a concentrated band of expression in the terminally differentiating granular layer. This pattern of distribution was precisely reproduced in the in vitro model of psoriasis, regardless of whether the cells were derived from the involved or the uninvolved body site. (This was evident by day 7 at the air–liquid interface—data not shown). The expression profile of CXCR2 was confirmed at the mRNA level by RT-PCR of mRNA extracted from skin models derived from three psoriatic patients and three controls. Figure 6 shows a representative result of the RT-PCR analysis of CXCR2 expression compared to the housekeeping gene GAPDH using mRNA extracted from in vitro models prepared from one case of psoriasis, one control skin, and in vivo tissues from the same cases.
Figure 5.
Expression of CXCR1, CXCR2, and their major ligands in normal and psoriatic human skin models. The figure shows expression of CXCR1, CXCR2, IL-8, and GROa in normal and involved psoriatic human skin and in vitro models after 21 d of air exposure. Models derived from uninvolved psoriatic skin showed identical patterns of gene expression to those derived from involved skin (not shown). In vivo normal and involved psoriatic skin are also shown for comparison.
Full figure and legend (145K)Figure 6.
Representative RT-PCR analysis of in vivo tissues and in vitro models showing CXCR2 and GAPDH mRNA expression. Gel electrophoresis of RT-PCR reactions using mRNA extracted from the following samples: in vivo skin: normal (1 and 2), psoriatic involved (3 and 4); in vitro skin models: normal (5 and 6), uninvolved (7 and 8), involved (9 and 10) after 14 d culture. Odd tracks show PCR reactions without a reverse transcriptase step (-) even tracks show complete RT-PCR reactions (+). PCR cycles were limited for CXCR2 to observe different levels of expression.
Full figure and legend (112K)Immunohistochemical analysis of IL-8 and GRO-, the major ligands of CXCR2, is shown in Figure 5. In agreement with in vivo observations, IL-8 expression was minimal in the normal in vitro models. The low levels of IL-8 expression were in line with expectations, given the hyperproliferative in vitro environment. In contrast, both involved and uninvolved psoriatic models expressed high concentrations of IL-8 throughout the viable keratinocyte layers. This pattern of distribution agreed with that observed in psoriasis in vivo. In all cases, the stratum corneum was negative for IL-8.
GRO- expression in normal epidermis in vivo was negligible, and this was reflected in the normal in vitro models. Like IL-8, GRO- was highly expressed in psoriasis in vivo and in involved and uninvolved models in vitro (Figure 5). In contrast to IL-8, however, a change in the pattern of distribution of GRO- was noted during the culture period. At days 7 and 14, the chemokine was strongly expressed in all of the viable keratinocyte layers (data not shown). By day 21, expression became concentrated into a strong band in the granular keratinocytes, with much weaker staining in the other layers. This pattern of expression was also observed in some in vivo psoriatic skin samples. In all cases, the stratum corneum remained negative, both in vivo and in vitro.
Expression of IL-8 and GRO-in vitro was further investigated by quantitation of the chemokines in conditioned culture media using multiplex ELISA technology (Figure 7). Inter-donor variation was observed in both normal and psoriatic models. Examination of the data from individual donors, however, revealed that those producing more IL-8 also produced more GRO- (individual data not shown). All donors showed a clear, statistically significant reduction in IL-8 production over the 21-d culture period, with the most marked reductions occurring before Day 7. This phenomenon was also observed in terms of GRO- levels produced by the psoriatic models. In contrast, GRO- levels remained fairly constant during the differentiation phase in the normal models, with no significant difference between time points. At all times, conditioned medium from skin models derived from psoriatic donors contained higher levels of both IL-8 and GRO- in involved samples relative to the uninvolved. Surprisingly, in the case of GRO- it was also observed that the normal models produced considerably (up to 5–10-fold) higher quantitative levels of GRO- than any of the psoriatic models.
Figure 7.
ELISA detection of chemokine levels in conditioned culture media from skin models using luminex beads. Each panel shows the mean level of each chemokine detected in the culture medium for media sampled at 0, 2, 7, 14, and 21 days. Panels show the levels of IL-8 (A), IL-6 (B) and GRO- (C) for normal skin models (open columns) (n=4) and psoriatic skin models (n=3) Black filled columns show results of models derived from involved skin and gray filled columns from uninvolved skin. Error bars represent mean
SEM.
Discussion
This study represents a comprehensive characterization of psoriatic human skin reconstructed in vitro, and demonstrates the potential of this model as a valuable tool in drug discovery. The reconstruction of skin from multiple psoriatic donors demonstrated a robust culture system, which allowed many genes of relevance to the disease to be expressed in a fully defined, serum-free environment. The optimized culture conditions allowed important differences between normal and psoriatic cells to be retained in vitro through at least two subcultures.
Skin undergoing reconstruction in the in vitro environment is, by nature, hyperproliferative, and therefore, displays some phenotypic features in common with in vivo psoriasis (Boelsma et al, 2000;Gibbs et al, 2000), as clearly demonstrated in Figure 3. These results suggest that certain characteristics of psoriasis, relating to the hyperproliferative keratinocyte and its subsequent abnormal differentiation, may be open to study using in vitro models. It should be noted that none of these cultures, regardless of tissue source, show the psoriasiform morphology and regenerative maturation features of lesional skin (Figure 1). The involved and uninvolved psoriatic models, however, showed higher rates of proliferation compared to the normal models (Figure 2). This observation is consistent with the idea that proliferation rates are modified in vitro, but still retain in vivo differences between the cell populations. Proliferation, therefore, depends upon the inclusion of growth-promoting agents such as human epidermal growth factor (hEGF) in the medium and the intrinsic production of these agents such as IL-8 and Gro-a by the keratinocytes and fibroblasts in the model. The minimal media used in these studies have only low levels of recombinant hEGF (0.1 ng per mL) and much higher levels (1–2 ng per mL) are required for increased suprabasal keratinocyte proliferation (Maas-Szabowski et al, 2003). Previous research has shown that the epithelial–mesenchymal signaling is altered in psoriasis generating increased hyperplasia of the epidermis even in uninvolved skin (Ting et al, 2000;Chen et al, 2001). This increased sensitivity to proliferation has been shown to be mediated by dermal fibroblasts from psoriatic donors which can change the proliferation rate of keratinocytes from normal skin (Saiag et al, 1985;Krueger and Jorgensen, 1997).
The release of primary cytokines plays a key role in the initiation of psoriasis. TNF- is produced by both dermal dendrocytes and epidermal keratinocytes (Nickoloff et al, 1991) and induces the release of cytokines such as IL-8 (Barker et al, 1990;Giustizieri et al, 2001), which subsequently act as important keratinocyte mitogens in the hyperproliferative response. IFN-, produced by T cells and keratinocytes, also stimulates keratinocyte growth, both directly (Bata-Csorgo et al, 1995) and through stimulation of IL-1
production (Wei et al, 1999). Downregulation of epidermal TNF- and IFN- levels has been shown to correlate with clinical improvement of psoriasis (Reich et al, 2001). There is also some evidence that primary cytokines such as TNF- and IFN- are expressed by keratinocytes in psoriasis (Livden et al, 1989;Kristensen et al, 1993), and this was supported by the results of our studies. The principal source of these pro-inflammatory cytokines, however, is known to be activated T cells in the lesion. Keratinocyte-derived TNF- and IFN- may exaggerate the psoriatic phenotype by extending the T cell-mediated signal (Wei et al, 1999).
IL-6 has also previously been implicated in the pathogenesis of psoriasis (Prens et al, 1990;Ameglio et al, 1997), with potentially important roles in both inflammation and keratinocyte mitogenesis (Grossman et al, 1989). Studies byOyama et al (1998) suggested that IL-6 might potentiate keratinocyte growth partly through the induction of the EGF receptor. Our studies demonstrated that the hyperproliferative in vitro environment stimulated IL-6 expression in both normal and psoriatic keratinocytes during the first 7 d of growth as also shown byGoretsky et al (1996) andFayyazi et al (1999). It is possible that the fibroblasts present in these models also contribute to the total levels of this cytokine. Future studies should investigate the expression and role of IL-1 and IL-1 in the psoriatic models, since recent studies have shown that this is a dominant pro-inflammatory cytokine (Steude et al, 2002).
The CXCR2/IL-8/GRO- pathway is considered to play an important role in wound healing (Devalaraja et al, 2000) and the pathogenesis of psoriasis (Jiang et al, 2001;Reich et al, 2001). IL-8 and GRO- are CXC chemokines that are barely detectable in normal epidermis (Gillitzer et al, 1996), but are upregulated in psoriasis (Anttila et al, 1992). The results of our studies confirmed these findings both in vivo and in vitro although GRO- was also expressed in the normal in vitro models. In addition, IL-8 has been shown to be a potent stimulator of epidermal cell proliferation (Tuschil et al, 1992) as well as angiogenesis and keratinocyte migration (Koch et al, 1992), both important aspects of psoriatic pathology.Steude et al (2002) have recently shown that exogenous IL-8 (100 ng per mL) added to the culture medium stimulate normal keratinocyte growth in organotypic cultures of human skin. Our studies have shown that endogenous secreted levels of IL-8 of 70 ng per mL are highest at the onset of stratification in organotypic cultures, but rapidly decline as a differentiated epidermis forms to 1–2 ng per mL after 21 d. GRO- levels, however, remained high in the media of normal models throughout the culture period suggesting that the turnover of this ligand was much lower in the normal model.
The receptor CXCR2 is overexpressed in the differentiated keratinocytes of the psoriatic plaque (Beljaards et al, 1997;Kulke et al, 1998) and in response to cutaneous injury (Nanney et al, 1995;Devalaraja et al, 2000) but is barely detectable in normal epidermis.Steude et al (2002) also demonstrated that CXCR2 is the receptor responsible for mediating the keratinocyte growth response to IL-8 and GRO-in vitro. Although we were able to detect high levels of IL-8 and GRO-, and CXCR2 in the psoriatic models, this was not accompanied by the development of a psoriasiform morphology due to extended hyperproliferation of keratinocytes. The IL-8 levels that were released into the culture medium after 7 d in our studies were significantly lower than those used bySteude et al (2002) to promote further keratinocyte growth. Therefore, the production and release of these ligands may require other T cell-mediated signals to extend keratinocyte growth in the psoriatic model. The immunohistochemistry data have shown that IL-8 and GRO- were both expressed and retained in the psoriatic models in agreement with in vivo staining, but in the case of the normal models, the majority was secreted into the medium. The source of these ligands is likely to be the keratinocytes but they could also come from the fibroblasts as they have been shown to produce higher levels of Il-8 (Konstantinova et al, 1996) and IL-6 (Grossman et al, 1989) in psoriasis. Thus release of these chemokines and the overexpression of the CXCR2 receptor may be key regulatory factors influencing the growth of keratinocytes in psoriasis. Future work will investigate the importance of CXCR2 signaling in psoriatic models by the use of blocking antibodies and gene transfer experiments in psoriatic and normal models.
The current studies have shown that psoriatic skin reconstructed in vitro partially displays the psoriatic phenotype. This includes an abnormal differentiated phenotype typical of a hyperproliferative environment; expression of pro-inflammatory cytokines; and the presence of the chemokine receptor CXCR2 and its ligands IL-8 and GRO-. The model supports the hypothesis that the keratinocytes and fibroblasts of individuals susceptible to psoriasis possess inherent differences from normal individuals, which contribute to the disease phenotype. The model developed in these studies could provide a starting point to investigate the mechanisms of abnormal keratinocyte growth and gene expression in psoriasis, as well as acting as a valuable tool in drug discovery. Future studies on blocking receptor signaling and incorporating other cell types, such as T cells, could provide important clues regarding the interaction between cell types in psoriasis.
Materials and Methods
Patient details
Patients with active chronic plaque psoriasis (n=12) were recruited to the study from the dermatology outpatient's clinic at Leicester Royal Infirmary (age, sex, duration of disease, severity PASI). All patients gave informed written consent. During 1 mo prior to biopsy, they had received no therapy for the psoriasis. Six millimeter punch biopsies were taken under local anesthetic (2% lignocaine with no added adrenaline) from chronic plaques, which had been present for 1–4 wk. Biopsies were also taken from uninvolved skin approximately 50 mm from the active lesion. Immediately after excision, biopsies were transported to the laboratory in Leibovitz L-15 medium and processed for skin reconstruction. Alternatively, excision biopsies were mounted in optimal cutting temperature of freezing medium (OCT) on cork, frozen in isopentane precooled in liquid nitrogen, and stored in liquid nitrogen vapor until further processing for immunohistochemistry. Normal skin from non-psoriatic patients (n=12) was obtained, following informed patient consent, from reduction mammoplasty. The subjects were Caucasian females aged between 23 and 62. This study was conducted in agreement with the Declaration of Helsinki and performed under the guidelines of the local medical ethics committee for University Hospitals of Leicester NHS Trust, England, with written consent of the patients.
Isolation and culture of epidermal keratinocytes and dermal fibroblasts from human skin
Following overnight incubation at 4°C, in 10 U per mL protease in Leibovitz L-15 medium (both from Sigma-Aldrich, Poole, UK), the epidermis was peeled away from the dermis. The epidermis was incubated in 0.05% trypsin/0.02% (wt/vol) EDTA (Life Technologies, Paisley, UK) for 20 min at 37°C. Trypsin activity was neutralized with 10% fetal calf serum (FCS) in phosphate-buffered saline (PBS) (Life Technologies), the cells were pelleted by centrifugation and resuspended in Clonetics Keratinocyte Growth Medium (KGM) (Biowhittaker, UK) at 0.06 mM calcium. Keratinocytes were seeded into culture in KGM on a coating of fibronectin and collagen I (FNC, from Biological Research Faculty and Facility, Ijamsville, Maryland). Following a 4-h attachment period at 37°C/5% (vol/vol) CO2, the medium was replaced to remove any non-viable or unattached cells. The dermis was incubated in 0.25% (wt/vol) collagenase A (Roche, Mannheim, Germany) in Dulbecco's Modified Eagles Medium (Life Technologies) with 10% (vol/vol) FCS for 24 h at 37°C/5% (vol/vol) CO2. Fibroblasts were pelleted by centrifugation, resuspended and seeded into culture in DMEM with 10% (vol/vol) FCS. All cultures were incubated at 37°C/5% (vol/vol) CO2, and re-fed with fresh medium every 2–3 d. The purity of the cell populations was confirmed by immunocytochemistry using primary antibodies to CD3, CD68, and CD45 to detect contaminating T cells, macrophages or leucocytes, and anti-cytokeratin and anti-fibroblast antigen 1 to confirm the presence of keratinocytes and fibroblasts. Keratinocytes were used to generate organotypic cultures at passage numbers 1 and 2, and fibroblasts cultures were used up to passage 3.
Generation of organotypic cultures
The culture system was a modified version of that described byStark et al (1999). All experiments generated at least triplicate models from each patient source. Fibroblasts were harvested from monolayer culture using 0.05% trypsin/0.02% (wt/vol) EDTA. The cells were pelleted by centrifugation and resuspended in FCS at a concentration of 2 
106 per mL. Collagen gels were prepared by mixing eight volumes of rat tail collagen I (Pharmingen, San Diego, California) with one volume of 10 Hanks Balanced Salt Solution (HBSS) (Life Technologies), neutralizing the pH with 0.2 M sodium hydroxide, and adding one volume of the resuspended fibroblasts (passage 1) in FCS. This gave a final fibroblast concentration of 2 105 per mL. The gel mixture was added to polycarbonate membrane filter inserts (Becton Dickinson) (2.5 mL per insert) in six-well culture plates. After an initial incubation period of 1 h at 37°C/5% (vol/vol) CO2 to allow the gels to set, 2 mL DMEM was added to the surface of the gels and a further 2 mL to the culture dish. The gels were incubated for 3 d to allow the onset of deposition of extracellular matrix proteins. After this time, keratinocyte medium was used throughout the culture period. The composition of the medium was modified to give a fully defined system in the absence of bovine pituitary extract. Clonetics KGM was purchased as a calcium-free "bullet kit" and supplemented with the following: EGF, insulin, hydrocortisone, transferrin, and epinephrine were added from the bullet kit to give final concentrations as defined by the manufacturer. Gentamicin (5 
g per mL) (Life Technologies), L-ascorbic acid (50 g per mL) (Sigma-Aldrich), bovine serum albumin (0.1% w/v) (Sigma-Aldrich), and calcium chloride (0.06 or 1.8 mM as specified) (BDH, Poole, UK) were also added. The modified medium was denoted as Supplemented Keratinocyte Growth Medium (SKGM). Keratinocytes were harvested from monolayer culture using a 20 min pre-incubation in 0.05% (wt/vol) EDTA in PBS to loosen the intercellular junctions, followed by a 1 min incubation in 0.05% trypsin/0.02% EDTA (wt/vol). This method minimized exposure to trypsin, which can be particularly damaging to keratinocytes. The keratinocytes were seeded onto the surface of the gels, using 1 106 cells per gel, in 350 L SKGM at 0.06 mM calcium. Keratinocytes were always derived from the same donor as the fibroblasts; both normal and psoriatic cells seeded onto gels at passage 2. After an initial attachment period of 30 min, the medium was made up to a total of 2 mL on the surface of the gel and 2 mL in the culture dish. When a confluent monolayer had formed (usually after 24 h), the calcium concentration was raised to 1.8 mM to stimulate the onset of differentiation. After a further 48 h, the cultures were exposed to the air–liquid interface by adjusting the level of the medium: 2 mL was applied to the culture dish only, in order to feed the skin models from the basal side. The models were maintained at the air–liquid interface for up to 21 days, and were re-fed with 2 mL fresh medium every 2–3 d. Conditioned medium was collected for analysis of IL-8, GRO-, and IL-6 levels by ELISA. Skin models were harvested at specified time points, fixed in buffered formalin and processed in paraffin wax for immunohistochemistry.
ELISA
A capture sandwich immunoassay was carried out using LabMAP Capture Microspheres (Luminex Corporation, Austin, Texas). Capture antibody beads were incubated with samples, prior to addition of LabMAP biotin-conjugated detection antibody for cytokine and the subsequent addition of streptavidin. Briefly, LabMAP Cytokine Capture Microspheres (GRO-, IL-6 and IL-8, Luminex, Austin, Texas) were diluted 1:25 in PBSBN (PBS pH 7.4, 1% BSA-fraction V and 0.05% sodium azide) and added together in a multiplex in a 96-well plate (1.2 m pore filter plate, Millipore; 50 L total volume per well). Standard curves were prepared using human cytokines (GRO-, IL-6 and IL-8, Pharmingen, San Diego, California) suitably diluted in PBSBN. Aliquots of the human cytokine standard curves (50 L total volume) were added to the capture antibody beads in columns 1 and 12 of the plate. Biological samples were added to the remaining columns of the plate (50 L total volume), mixed using a plate shaker and incubated in the dark at room temperature for 1 h. After washing twice with PBSBN (100 L per wash), LabMAP biotin-conjugated cytokine detection antibodies (IL-6, IL-8 and GRO- Luminex) were diluted 1:25 and added to the corresponding wells in the 96-well plate (50 L total volume per well). In addition, PBSBN (50 L) was added to each well. The plate was shaken and then incubated in the dark at room temperature for a further 1 h. After washing twice with PBSBN (100 L per wash), streptavidin/PE (10 g per mL, 50 L; Molecular Probes, Leiden, Holland) and PBSBN (50 L) was then added to the 96-well plate and incubated in the dark at room temperature for 30 min. After two further washes with 100 L PBSBN and one wash with 100 L PBS, 100 L PBS was added to each well before measurement. Samples (50 L) were assayed for cytokine levels using the Luminex100 96-well plate reader, counting 200 events per bead.
Immunohistochemistry
In vivo skin and in vitro skin models were fixed in 10% (v/v) buffered formalin prior to processing in paraffin wax for immunohistochemistry. Paraffin sections were cut onto aminopropyltriethoxysilane-coated slides and dried overnight at 37°C. Immunohistochemical staining to characterize the skin models was carried out using antibodies listed in Table I at the stated dilutions using the streptavidin–biotin complex (ABC) method. Appropriate antigen retrieval steps were applied to optimize staining. Briefly, the sections were treated for antigen retrieval, if appropriate, after deparaffinization either for 2.5 min at 700 W 2 in citrate buffer pH 6.0 in a suitable microwave oven or digested in proteinase K (5 g per mL) for 1 h at 37°C. The sections were equilibrated in Tris-buffered saline pH 7.6 (TBS) prior to blocking in 20% normal rabbit serum (NRS) in TBS for 10 min, and then incubated with the primary antibody diluted in 20% NRS for 18 h at 4°C in a humidity chamber. After washing in TBS, sections were incubated in biotinylated rabbit anti-mouse F(ab)2 (Dakopatts, Glostrup, Denmark) diluted 1:400 in 20% NRS for 30 min, followed by 30 min incubation in pre-alkaline phosphatase conjugated ABC (Dako). Following the final washes in TBS, sections were detected with Fast Red TR salt (Sigma F-2768) and Naphthol AS-BI phosphate (Sigma N-2250) in a Veronal acetate buffer (VAB). The sections were then counterstained in hematoxylin and mounted using an aqueous mounting medium. For each case, a negative control with omission of the primary antibody was included. Two observers (A.K.G., J.H.P.) scored the immunostaining. For proliferation studies, Ki67 positive stained cells were counted for each model from 20 fields using an objective with magnification of 40. Four cases of psoriasis with involved and uninvolved models and four normal models, each in duplicate, were analyzed for proliferation.
RT-PCR
mRNA was extracted from tissue sections of normal and psoriatic skin models using oligo d(T)-labeled paramagnetic beads according to the manufacturers instructions (Dynal, UK). The sections were immediately dropped into 100 L of lysis/binding buffer (100 mM Tris-HCl, pH 8.0; 500 mM LiCl; 10 mM EDTA, pH 8.0; 1% wt/vol sodium dodecyl sulfate; 5 mM dithiothreitol). mRNA was extracted and processed using oligo-dT-linked Dynabeads (Dynal, Bromborough, UK), as described previously byBicknell et al (1996). Reverse transcription reactions were carried out at 42°C for 1 h using Expand-RT (Boehringer Mannheim) in accordance with the manufacturer's instructions. In a total volume of 25 L, reactions contained 5 L Expand RT buffer, 10 mM dithiothreitol (DTT), 1 mM each of dNTPs (all from Boehringer Mannheim) and 30 U RNasin (Promega) in diethylpyro-carbonate(DEPC)-treated water. To the positive RT reactions, 50U RT was added, and to negative RT reactions, an equal volume of DEPC-treated water was added. A series of primers for CXCR1 (NM_000634 5'-TCC TCA TCT TCC TGC TTT GC-3' 862:881, 5-'GTA GGA GGT AAC ACG ATG ACG-3' 1136:1116), CXCR2 (NM_001557 5'-TCC TCA TCT TCC TGC TCT GC-3' 1178:1197, 5'-AGT GTG CCC TGA AGA AGA GC-3' 1476:1457), and glyceraldehyde-3-phosphate dehydrogenase (GAPDH, NM_002046 5'-AGA ACA TCA TCC CTG CCT C-3' 686:704, 5'-GCC AAA TTC GTT GTC ATA CC-3' 1032:1013) were used and PCR was performed using a Perkin-Elmer Thermocycler (Perkin-Elmer, Foster City, California). Typical reactions were carried out with 1 L of cDNA template in a total volume of 50 L containing PCR buffer (45 mM Tris pH 8.8, 11 mM (NH4)2SO4, 4.5mM MgCl2, 200 mM dNTP, 110 g per mL bovine serum albumin (BSA), 6.7 mM -mercaptoethanol and 4.4 mM EDTA, pH 8.8) and 10 pmol each of forward and reverse primers. DNA was denatured at 94°C for 5 min and held at the primer annealing temperature during the addition of 1 IU of Taq polymerase (Promega, Madison, Wisconsin), followed by an extension step at 72°C for 30 s. For CXCR1 and CXCR2 primers the following cycle profile was used: 95°C for 1 min, 58°C for 45 s, 72°C for 1 min for 35 cycles, with a final extension step at 72°C for 7 min. For GAPDH, amplification was performed for 25 cycles with a profile of 95°C for 1 min, 60°C for 30 s, 72°C for 30 s, with final extension of 72°C for 7 min. The PCR products were loaded on a 2% agarose gel containing 15 g per mL ethidium bromide and gels were run at 150 V for 2 h. The resulting bands were visualized on a UV transilluminator and images captured using a gel documentation system. The results were examined for models derived from three cases of psoriasis and three controls.
Statistical analysis
Data were analyzed using Student's t test and expressed as meansSD. Statistical significance was assumed for p<0.05.
References
| 1. | Ameglio F, Bonifati C & Fazio M et al. Interleukin-11 production is increased in organ cultures of lesional skin of patients with active plaque-type psoriasis as compared with nonlesional and normal skin. Similarity to interleukin-1 beta, interleukin-6 and interleukin-8. Arch Dermatol Res (1997) 289: 399–403 10.1007/s004030050211. | Article | PubMed | ISI | ChemPort | |
| 2. | Anttila HS, Reitamo S, Erkko P, Ceska M, Moser B & Baggiolini M. Interleukin-8 immunoreactivity in the skin of healthy subjects and patients with palmoplantar pustulosis and psoriasis. J Invest Dermatol (1992) 98: 96–101 10.1111/1523-1747.ep12495817. | Article | PubMed | ISI | ChemPort | |
| 3. | Barker JN, Sarma V, Mitra RS, Dixit VM & Nickoloff BJ. Marked synergism between tumor necrosis factor-alpha and interferon-gamma in regulation of keratinocyte-derived adhesion molecules and chemotactic factors. J Clin Invest (1990) 85: 605–608. | PubMed | ISI | ChemPort | |
| 4. | Bata-Csorgo Z, Hammerberg C, Voorhees JJ & Cooper KD. Intralesional T-lymphocyte activation as a mediator of psoriatic epidermal hyperplasia. J Invest Dermatol (1995) 105: 89S–94S. | Article | PubMed | ChemPort | |
| 5. | Beljaards RC, Van Beek P, Nieboer C, Stoof TJ & Boorsma DM. The expression of interleukin-8 receptor in untreated and treated psoriasis. Arch Dermatol Res (1997) 289: 440–443. | Article | PubMed | ISI | ChemPort | |
| 6. | Bicknell GR, Shaw JA, Pringle JH & Furness PN. Amplification of specific mRNA from a single human renal glomerulus, with an approach to the separation of epithelial cell mRNA. J Pathol (1996) 180: 188–193. | Article | PubMed | ISI | ChemPort | |
| 7. | Boelsma E, Gibbs S, Faller C & Ponec M. Characterization and comparison of reconstructed skin models: Morphological and immunohistochemical evaluation. Acta Dermatol Venereol (2000) 80: 82–88. | ChemPort | |
| 8. | Chapman ML, Dimitrijevich SD, Hevelone JC, Goetz D, Cohen J, Wise GE & Gracy RW. Inhibition of psoriatic cell proliferation in in vitro skin models by amiprilose hydrochloride. In Vitro. Cell Dev Biol (1990) 26: 991–996. | ChemPort | |
| 9. | Chen G, McCormick TS, Hammerberg C, Ryder-Diggs S, Stevens SR & Cooper KD. Basal keratinocytes from uninvolved psoriatic skin exhibit accelerated spreading and focal adhesion kinase responsiveness to fibronectin. J Invest Dermatol (2001) 117: 1538–1545. | Article | PubMed | ISI | ChemPort | |
| 10. | Devalaraja RM, Nanney LB, Du J, Qian Q, Yu Y, Devalaraja MN & Richmond A. Delayed wound healing in CXCR2 knockout mice. J Invest Dermatol (2000) 115: 234–244 10.1046/j.1523-1747.2000.00034.x. | Article | PubMed | ISI | ChemPort | |
| 11. | Fayyazi A, Schweyer S & Soruri A et al. T lymphocytes and altered keratinocytes express interferon-gamma and interleukin 6 in lichen planus. Arch Dermatol Res (1999) 291: 485–490 10.1007/s004030050442. | Article | PubMed | ISI | ChemPort | |
| 12. | Fukuoka M, Ogino Y, Sato H, Ohta T, Komoriya K, Nishioka K & Katayama I. RANTES expression in psoriatic skin, and regulation of RANTES and IL-8 production in cultured epidermal keratinocytes by active vitamin D3 (tacalcitol). Br J Dermatol (1998) 138: 63–70 10.1046/j.1365-2133.1998.02027.x. | Article | PubMed | ISI | ChemPort | |
| 13. | Gibbs S, Silva Pinto AN, Murli S, Huber M, Hohl D & Ponec M. Epidermal growth factor and keratinocyte growth factor differentially regulate epidermal migration, growth, and differentiation. Wound Repair Regen (2000) 8: 192–203 10.1046/j.1524-475x.2000.00192.x. | Article | PubMed | ISI | ChemPort | |
| 14. | Gillitzer R, Ritter U, Spandau U, Goebeler M & Brocker EB. Differential expression of GRO-alpha and IL-8 mRNA in psoriasis: A model for neutrophil migration and accumulation in vivo. J Invest Dermatol (1996) 107: 778–782 10.1111/1523-1747.ep12371803. | Article | PubMed | ISI | ChemPort | |
| 15. | Giustizieri ML, Mascia F & Frezzolini A et al. Keratinocytes from patients with atopic dermatitis and psoriasis show a distinct chemokine production profile in response to T cell-derived cytokines. J Allergy Clin Immunol (2001) 107: 871–877 10.1067/mai.2001.114707. | Article | PubMed | ISI | ChemPort | |
| 16. | Goretsky MJ, Harriger MD, Supp AP, Greenhalgh DG & Boyce ST. Expression of interleukin-1alpha, interleukin-6, and basic fibroblast growth factor by cultured skin substitutes before and after grafting to full-thickness wounds in athymic mice. J Trauma (1996) 40: 894–899 discussion 899–900. | PubMed | ISI | ChemPort | |
| 17. | Grossman RM, Krueger J & Yourish D et al. Interleukin 6 is expressed in high levels in psoriatic skin and stimulates proliferation of cultured human keratinocytes. Proc Natl Acad Sci USA (1989) 86: 6367–6371. | PubMed | ChemPort | |
| 18. | Jiang WY, Chattedee AD, Raychaudhuri SP, Raychaudhuri SK & Farber EM. Mast cell density and IL-8 expression in nonlesional and lesional psoriatic skin. Int J Dermatol (2001) 40: 699–703 10.1046/j.1365-4362.2001.01262.x. | Article | PubMed | ISI | ChemPort | |
| 19. | Koch AE, Polverini PJ & Kunkel SL et al. Interleukin-8 as a macrophage-derived mediator of angiogenesis. Science (1992) 258: 1798–1801. | PubMed | ISI | ChemPort | |
| 20. | Konstantinova NV, Duong DM, Remenyik E, Hazarika P, Chuang A & Duvic M. Interleukin-8 is induced in skin equivalents and is highest in those derived from psoriatic fibroblasts. J Invest Dermatol (1996) 107: 615–621 10.1111/1523-1747.ep12584215. | Article | PubMed | ISI | ChemPort | |
| 21. | Kristensen M, Chu CQ, Eedy DJ, Feldmann M, Brennan FM & Breathnach SM. Localization of tumour necrosis factor-alpha (TNF-alpha) and its receptors in normal and psoriatic skin: Epidermal cells express the 55-kD but not the 75-kD TNF receptor. Clin Exp Immunol (1993) 94: 354–362. | PubMed | ISI | ChemPort | |
| 22. | Krueger GG & Jorgensen CM. Experimental models for psoriasis. J Invest Dermatol (1990) 95: 56S–58S 10.1111/1523-1747.ep12505791. | Article | PubMed | ChemPort | |
| 23. | Krueger GG & Jorgensen CM. Defined system to assess the in vitro induction of a psoriasis phenotype on normal keratinocytes by fibroblasts from psoriatic subjects. J Cutaneous Med Surg (1997) 2: 20. |
| 24. | Kulke R, Bornscheuer E, Schluter C, Bartels J, Rowert J, Sticherling M & Christophers E. The CXC receptor 2 is overexpressed in psoriatic epidermis. J Invest Dermatol (1998) 110: 90–94 10.1046/j.1523-1747.1998.00074.x. | Article | PubMed | ISI | ChemPort | |
| 25. | Livden JK, Nilsen R, Bjerke JR & Matre R. In situ localization of interferons in psoriatic lesions. Arch Dermatol Res (1989) 281: 392–397. | Article | PubMed | ISI | ChemPort | |
| 26. | Maas-Szabowski N, Starker A & Fusenig NE. Epidermal tissue regeneration and stromal interaction in HaCaT cells isinitiated by TGF-alpha. J Cell Sci (2003) 116: 2937–2948 10.1242/jcs.00474. | Article | PubMed | ISI | ChemPort | |
| 27. | Mils V, Basset-Seguin N, Moles JP, Tesniere A, Leigh I & Guilhou JJ. Comparative analysis of normal and psoriatic skin both in vivo and in vitro. Differentiation (1994) 58: 77–86 10.1046/j.1432-0436.1994.5810077.x. | Article | PubMed | ISI | ChemPort | |
| 28. | Mizutani H, Ohmoto Y, Mizutani T, Murata M & Shimizu M. Role of increased production of monocytes TNF-alpha, IL-1beta and IL-6 in psoriasis: Relation to focal infection, disease activity and responses to treatments. J Dermatol Sci (1997) 14: 145–153 10.1016/S0923-1811(96)00562-2. | Article | PubMed | ISI | ChemPort | |
| 29. | Nanney LB, Mueller SG, Bueno R, Peiper SC & Richmond A. Distributions of melanoma growth stimulatory activity of growth- regulated gene and the interleukin-8 receptor B in human wound repair. Am J Pathol (1995) 147: 1248–1260. | PubMed | ISI | ChemPort | |
| 30. | Nickoloff BJ, Karabin GD & Barker JN et al. Cellular localization of interleukin-8 and its inducer, tumor necrosis factor-alpha in psoriasis. Am J Pathol (1991) 138: 129–140. | PubMed | ISI | ChemPort | |
| 31. | Oyama N, Sekimata M, Nihei Y, Iwatsuki K, Homma Y & Kaneko F. Different growth properties in response to epidermal growth factor and interleukin-6 of primary keratinocytes derived from normal and psoriatic lesional skin. J Dermatol Sci (1998) 16: 120–128 10.1016/S0923-1811(97)00040-6. | Article | PubMed | ISI | ChemPort | |
| 32. | Pol A, Bergers M, van Ruissen F, Pfundt R & Schalkwijk J. A simple technique for high-throughput screening of drugs that modulate normal and psoriasis-like differentiation in cultured human keratinocytes. Skin Pharmacol Appl Skin Physiol (2002) 15: 252–261 10.1159/000066010. | Article | PubMed | ISI | |
| 33. | Prens EP, Benne K, van Damme J, Bakkus M, Brakel K, Benner R & van Joost T. Interleukin-1 and interleukin-6 in psoriasis. J Invest Dermatol (1990) 95: 121S–124S 10.1111/1523-1747.ep12874991. | Article | PubMed | ChemPort | |
| 34. | Raychaudhuri SP, Dutt S, Raychaudhuri SK, Sanyal M & Farber EM. Severe combined immunodeficiency mouse-human skin chimeras: A unique animal model for the study of psoriasis and cutaneous inflammation. Br J Dermatol (2001) 144: 931–939 10.1046/j.1365-2133.2001.04178.x. | Article | PubMed | ISI | ChemPort | |
| 35. | Reich K, Garbe C & Blaschke V et al. Response of psoriasis to interleukin-10 is associated with suppression of cutaneous type 1 inflammation, downregulation of the epidermal interleukin-8/CXCR2 pathway and normalization of keratinocyte maturation. J Invest Dermatol (2001) 116: 319–329 10.1046/j.1523-1747.2001.01248.x. | Article | PubMed | ISI | ChemPort | |
| 36. | Saiag P, Coulomb B, Lebreton C, Bell E & Dubertret L. Psoriatic fibroblasts induce hyperproliferation of normal keratinocytes in a skin equivalent model in vitro. Science (1985) 230: 669–672. | PubMed | ISI | ChemPort | |
| 37. | Stark HJ, Baur M, Breitkreutz D, Mirancea N & Fusenig NE. Organotypic keratinocyte cocultures in defined medium with regular epidermal morphogenesis and differentiation. J Invest Dermatol (1999) 112: 681–691 10.1046/j.1523-1747.1999.00573.x. | Article | PubMed | ISI | ChemPort | |
| 38. | Stern RS. Psoriasis. Lancet (1997) 350: 349–353 10.1016/S0140-6736(97)05257-4. | Article | PubMed | ISI | ChemPort | |
| 39. | Steude J, Kulke R & Christophers E. Interleukin-1-stimulated secretion of interleukin-8 and growth-related oncogene-alpha demonstrates greatly enhanced keratinocyte growth in human raft cultured epidermis. J Invest Dermatol (2002) 119: 1254–1260 10.1046/j.1523-1747.2002.19616.x. | Article | PubMed | ISI | ChemPort | |
| 40. | Ting KM, Rothaupt D & McCormick TS et al. Overexpression of the oncofetal Fn variant containing the EDA splice-in segment in the dermal-epidermal junction of psoriatic uninvolved skin. J Invest Dermatol (2000) 114: 706–711 10.1046/j.1523-1747.2000.00871.x. | Article | PubMed | ISI | ChemPort | |
| 41. | Tuschil A, Lam C, Haslberger A & Lindley I. Interleukin-8 stimulates calcium transients and promotes epidermal cell proliferation. J Invest Dermatol (1992) 99: 294–298 10.1111/1523-1747.ep12616634. | Article | PubMed | ISI | ChemPort | |
| 42. | Van Ruissen F, de Jongh GJ, van Erp PE, Boezeman JB & Schalkwijk J. Cell kinetic characterization of cultured human keratinocytes from normal and psoriatic individuals. J Cell Physiol (1996a) 168: 684–694 10.1002/(SICI)1097-4652(199609)168:3<684::AID-JCP21>3.3.CO;2-4. | Article | ChemPort | |
| 43. | Van Ruissen F, de Jongh GJ, Zeeuwen PL, Van Erp PE, Madsen P & Schalkwijk J. Induction of normal and psoriatic phenotypes in submerged keratinocyte cultures. J Cell Physiol (1996b) 168: 442–452 10.1002/(SICI)1097-4652(199608)168:2<442::AID-JCP23>3.3.CO;2-G. | Article | ChemPort | |
| 44. | Wei L, Debets R, Hegmans JJ, Benner R & Prens EP. IL-1 beta and IFN-gamma induce the regenerative epidermal phenotype of psoriasis in the transwell skin organ culture system. IFN-gamma up-regulates the expression of keratin 17 and keratinocyte transglutaminase via endogenous IL-1 production. J Pathol (1999) 187: 358–364 10.1002/(SICI)1096-9896(199902)187:3<358::AID-PATH253>3.3.CO;2-#. | Article | PubMed | ISI | ChemPort | |
Acknowledgments
This work was funded by AstraZeneca R&D Charnwood, Discovery Bioscience, Loughborough.
