In vivo studies of factors influencing mast cell numbers in rat skin

In order to examine the biological relevance of known in vitro stimuli for mast cell growth, the following substances were injected at two day intervals into the skin of Wistar rats: Ascaris Ag, epidermal growth factor, basic fibroblast growth factor, fibronectin, phytohemagglutinin-stimulated spleen cell supernatants, L-cell fibroblast supernatants and horse serum alone or in combination with L-cell supernatants. In some experiments, rats were also injected with fresh or cultured peritoneal cells. Single injections of the different factors had no significant effect on mast cell numbers. After multiple injections (4-10 x), deep dermal and submuscular mast cell numbers increased most markedly in ascaris sensitized animals at sites of ascaris antigen injections and in normal animals in response to a combination of horse serum and L-cell supernatants. Less pronounced increases occurred with all other test substances except for epidermal growth factor which was inactive. Mast cell numbers were also increased at sites of injections of immature, cultured mast cells and less so after injections of mast cell precursors and mature cells. Taken together, these data show that in vivo growth and differentiation of cutaneous mast cells can be influenced by several fibroblast- and lymphocyte-derived growth factors.     

Induction of manganese superoxide dismutase in human dermal fibroblasts: a UV-B-mediated paracrine mechanism with the release of epidermal interleukin 1 alpha,interleukin 1 beta,and tumor necrosis factor alpha

BACKGROUND: Reactive oxygen species generated in the skin by UV irradiation promote photoaging and photocarcinogenesis. The manganese (Mn) superoxide dismutase (SOD) is a primary antioxidant enzyme that crucially contributes to the homeostasis of oxygen radicals within the mitochondria, and thus critically participates in the control of senescence and tumor generation. OBJECTIVE: To determine whether repetitive UV-B exposure, as practiced for light hardening during phototherapy for various photodermatoses, can enhance the adaptive antioxidant response by up-regulating MnSOD activity in either the epidermal or the dermal skin compartment. DESIGN: In vitro experiments to determine MnSOD activity levels in cultured human dermal fibroblasts and epidermal cells (HaCaT cells and primary keratinocytes) at different times after direct UV-B exposure or after incubation of human dermal fibroblasts with supernatants from UV-B-irradiated epidermal cells. SETTING: Photobiological research laboratory in a university dermatology department. INTERVENTION: Irradiation of cultured human dermal fibroblasts and epidermal cells with UV-B. MAIN OUTCOME MEASURES: Manganese SOD messenger RNA and activity levels in cultured irradiated or mock-treated skin cells. RESULTS: No increase in MnSOD activity could be detected in fibroblasts or epidermal cells until 24 hours after UV-B irradiation. However, fibroblasts incubated with supernatants from UV-B-irradiated epidermal cells showed a marked increase in specific MnSOD messenger RNA and activity. Removal of interleukin 1alpha, interleukin 1beta, and tumor necrosis factor alpha from the supernatants led to a significant reduction of MnSOD mRNA in fibroblasts. CONCLUSION: Irradiation of the epidermal cells with UV-B induced a release of soluble factors that amplified MnSOD activity in fibroblasts via a paracrine mechanism.     

Effect of activated human mast cells and mast cell-derived mediators on proliferation,type I collagen production and glycosaminoglycans synthesis by human dermal fibroblasts

Although an increased number of mast cells in fibrotic tissues such as scleroderma, keloid or healing wound has been highlighted, it is still unclear whether or not mast cells are fibrogenic. The aim of the present study is to determine whether functionally active human mast cells can provide human dermal fibroblasts directly with fibrogenic properties. In order to examine the effects of IgE-mediated mast cell activation on fibroblast proliferation and synthesis of type I collagen, we utilized an in vitro defined system in which cultured human mast cells were co-cultured with human dermal fibroblasts. We also employed a three-dimensional fibroblast culture system using supplementation of L-ascorbic acid as an assay system to investigate the effects of mast cell-derived mediators on synthesis of glycosaminoglycans by human fibroblast. Fibroblast proliferation was actively stimulated with IgE-activated mast cells. However, this stimulatory effect was canceled in co-cultures with a higher number of IgE-activated mast cells. In the presence of a higher number of activated mast cells, the fibroblast cell layer was destroyed, in contrast to an intact cell layer in the presence of same number of the mast cells without activation. Type I collagen synthesis was unchanged in fibroblasts co-cultured with mast cells. The total amount of main disaccharide units, particularly DELTADi-HA, was increased when fibroblasts were exposed to histamine. Thus, we conclude that other factors, in addition to mast cells, are important in the development of human tissue fibrosis or sclerosis.     

Characteristics of an epidermal cell thymocyte-activating factor (ETAF) produced by human epidermal cells and a human squamous cell carcinoma cell line

In this study we show that both cultured normal human epidermal cells (EC) and a human squamous cell carcinoma (SCC) cell line produce a thymocyte-activating factor (ETAF). EC-ETAF and SCC-ETAF both have a Mr of 15,000 and were eluted from chromatofocusing at the same isoelectric points of 7.2, 5.8, and 5.0. Both activities were maintained at alkaline pH and were destroyed at temperatures above 60 degrees C. In addition to stimulating thymocyte proliferation, human ETAF exhibited a variety of other pertinent biologic activities. Although EC-ETAF or SCC-ETAF by themselves exhibited no T-cell growth factor activity, both ETAF preparations enhanced Interleukin 2 production by cultured human peripheral blood lymphocytes when stimulated with polyclonal T-cells stimulants (Concanavalin A and phorbol myristate acetate). Human ETAF also was chemotactic for rabbit polymorphonuclear leukocytes and was directly mitogenic for cultured human dermal fibroblasts. Injection of human ETAF into C3H/HeJ mice, resulted in inducing serum amyloid A (SAA) production by murine hepatocytes. The thymocyte growth-enhancing activity, the fibroblast-stimulating activity, and the SAA-inducing capacity of ETAF all coeluted off AcA54 gel. These biologic as well as biochemical properties of human keratinocyte-derived ETAF are identical with those of human macrophage-derived Interleukin 1. The ability of keratinocytes to release an immunomodulating factor with such diverse consequences may play an important role in normal wound healing and in diseases involving epithelial tissues.     

Repeated exposure of human fibroblasts to UVR induces secretion of stem cell factor and senescence

Background Some of chronic hyperpigmentary diseases, such as melasma, induced by multiple factors including chronic sunlight exposure, can recur even after chemical epidermal removal. Dermal factors may be involved in the pathogenesis of melasma. Changes in dermal fibroblasts resulting from chronic sun exposure might cause melanocytes to synthesize melanin in the epidermis. Objective This study aimed at determining the effects of repetitive ultraviolet (UV) radiation on cultured fibroblasts and the secretion of melanogenic factors. Methods Cultured human fibroblasts were exposed to ultraviolet A (UVA) or ultraviolet B (UVB) for five consecutive days. After each irradiation, the supernatant medium was isolated from each dish and measured for levels of stem cell factor (SCF) and hepatocyte growth factor using an ELISA kit assay. To assess the effect of the keratinocyte-derived factors on fibroblast-secretion of SCF and hepatocyte growth factor, we added supernatants of the UV-irradiated keratinocytes to the non-irradiated fibroblasts. Finally, the irradiated fibroblasts were stained with senescence associated-β-galactosidase to assess their senescent change. Results Fibroblasts irradiated with UVA or UVB for five consecutive days, secreted SCF at levels that increased with repeated UVA or UVB exposure. Conditioned culture medium from UV-irradiated keratinocytes also induced SCF release from fibroblasts, depending on the number of UV exposures. UVA- or UVB-irradiated fibroblasts stained positive for senescence associated-β-galactosidase, and the staining intensity increased with repeated exposure. Conclusion These results suggest that fibroblast senescence and increased SCF secretion after repeated UV irradiation may be related to the pathogenesis of recurring hyperpigmentation disorders induced by chronic sun exposure.     

Modulation of keratinocyte growth factor (KGF) mRNA expression in human dermal fibroblasts grown in monolayer or within a collagen matrix

Abstract In this study, we analysed the modulation of keratinocyte growth factor (KGF) mRNA expression in human dermal fibroblasts cultured either in monolayer or within a collagen matrix (dermal equivalent). In monolayer cultures, KGF expression by quiescent fibroblasts was stimulated by different growth substances such as serum, epidermal growth factor and basic fibroblast growth factor. Moreover, we demonstrated that the induction of this gene was mediated by at least 2 different signalling pathways involving protein kinase C (PKC) and cAMP. In dermal equivalents, we observed that the collagen matrix negatively modulated KGF mRNA expression. Indeed, among the growth substances used, only the serum slightly stimulated KGF expression. Nevertheless, as in monolayers, this induction involved at least PKC and cAMP signalling pathways. As the collagen matrix can modulate fibroblast growth, we also studied KGF expression in growing fibroblasts from either monolayer cultures or dermal equivalents. We then showed that this collagen matrix negatively influenced KGF expression independently of the proliferative state of fibroblasts. All these results underline the fact that KGF mRNA expression by human dermal fibroblasts is induced by different substances; however this expression can be modulated by fibroblast-matrix interactions.     

Interleukin 6 indirectly induces keratinocyte migration

IL-6-deficient transgenic mice (IL-6 KO) display significantly delayed cutaneous wound healing. To further elucidate the role of IL-6 in skin wound healing, epidermal keratinocyte and dermal fibroblast cells were isolated from neonatal IL-6 KO mice and treated with rmIL-6. It was found that rmIL-6 alone did not significantly modulate the proliferation or migration of cultured IL-6 KO keratinocytes. rmIL-6, however, significantly induced the migration of IL-6 KO keratinocytes (up to 5-fold) when co-cultured with dermal fibroblasts. Culture supernatants from IL-6-treated fibroblasts were also found to induce the migration of keratinocytes to a similar degree. Genomics analysis of treated fibroblasts indicated that rmIL-6 does not induce any known soluble keratinocyte migratory factors. rmIL-6 treatment of fibroblast, however, induced a rapid and sustained phosphorylation of STAT3 protein. These data indicate that IL-6 could influence wound healing by inducing keratinocyte migration through the production of a soluble fibroblast-derived factor, and its activity may be associated with STAT3 activation.     

Emerging role of dermal compartment in skin pigmentation: comprehensive review

The variations in human skin colour mainly occur due to differences in the distribution of melanin pigment throughout the body, synthesized by epidermal melanocytes which are further taken up by keratinocytes present in epidermis. Recently, it has been discovered that besides these cells, dermis derived fibroblast factors also play a prominent role in regulating skin pigmentation. There exists a signal crosstalk between epidermal melanocytes, keratinocytes and dermal fibroblasts and any impairment in these signalling pathways may give rise to pigmentary disorders. Vitiligo is a hypopigmentary disorder and alteration in the expression level of several fibroblast-specific factors has been reported in the lesional skin of vitiligo patients. In such patients, there is decrease in the expression levels of factors such as basic fibroblast growth factor, stem cell factor (SCF) and keratinocyte growth factor (KGF) along with a steep increase in the expression levels of Dickkopf 1. Patients affected with hyperpigmentary disorder like melasma exhibit a marked increase in SCF and KGF expression levels leading to increase in melanin production and those affected with solar lentigo experience upregulation in the expression levels of SCF, KGF and HGF (hepatocyte growth factor). Hence, we conclude that new therapeutic strategies can be adopted to cure these pigmentary disorders by targeting factors involved in crosstalk signalling between epidermal melanocytes, keratinocytes and dermal fibroblasts.     

Human keratinocytes constitutively express interleukin-18 and secrete biologically active interleukin-18 after treatment with pro-inflammatory mediators and dinitrochlorobenzene

Interleukin-18 is a potent inducer of interferon-gamma by activated T cells, macrophages, and monocytes and is synthesized as an inactive precursor. Pro-interleukin-18 must be cleaved by interleukin-1-beta-converting enzyme for secretion of the biologically active form. We report that among selected non-bone marrow derived skin cells, interleukin-18 mRNA is constitutively expressed by human keratinocytes and not by dermal microvascular endothelial cells, dermal fibroblasts, or melanocytes. Interleukin-18 mRNA and intracellular protein levels are neither changed in human keratinocytes nor induced in human dermal microvascular endothelial cells, dermal fibroblasts, or melanocytes by exposure to pro-inflammatory stimuli. Exposure of human keratinocytes to phorbol 12-myrisate 13-acetate, lipopolysaccharides or the contact sensitizer DNCB results in the secretion of immunoprecipitable interleukin-18 protein. Human keratinocyte-secreted interleukin-18 is biologically active, in that conditioned media from phorbol 12-myrisate 13-acetate, lipopolysaccharide and DNCB-treated human keratinocytes induce interferon-gamma expression by peripheral blood mononuclear cells. This bioactivity is neutralized by anti-interleukin-18, but not anti-interleukin-12 antibodies. By immunohistochemistry, interleukin-18 protein is detected in basal keratinocytes of normal human skin, but its expression is markedly upregulated in suprabasal keratinocytes in psoriasis. These findings indicate that human keratinocytes are a source of biologically functional interleukin-18 and thus are capable of playing an initiating part in the local interferon-gamma-dependent inflammatory processes through expression, activation, and secretion of interleukin-18.     

Quantitative analysis of the synthesis and secretion of type VII collagen in cultured human dermal fibroblasts with a sensitive sandwich enzyme-linked immunoassay

Type VII collagen is the major component of anchoring fibrils in the epidermal basement membrane. Its expression has been analyzed by immunostaining or Northern blotting, but rarely at the protein level. In this study, we have quantitatively examined the effects of ascorbic acid and various cytokines/growth factors on the protein synthesis and secretion of type VII collagen by human dermal fibroblasts in culture, using a developed, highly sensitive sandwich enzyme-linked immunoassay with two kinds of specific monoclonal antibodies against the non-collagenous domain-1. Ascorbic acid and its derivative induced a twofold increase in type VII collagen synthesis, and markedly increased the secretion of type VII collagen into the medium when compared with the control culture. This effect was not influenced by the presence of transforming growth factor-beta1 (TGF-beta1). The synthesis of type VII collagen was elevated by TGF-beta1, platelet-derived growth factor, tumor necrosis factor-alpha, and interleukin-1beta, but not by TGF-alpha. Thus, our data indicate that the synthesis and secretion of type VII collagen in human dermal fibroblasts are regulated by ascorbate and the enhancement of type VII collagen gene expression by cytokines/growth factors is accompanied with elevated production of type VII collagen at the protein level.     

The differentiation and function of myofibroblasts is regulated by mast cell mediators.

Myofibroblasts are fibroblasts that express certain features of smooth muscle differentiation. Increased numbers of myofibroblasts and mast cells are frequently found together in a wide variety of settings, such as normal wound repair and scleroderma skin, which suggests that mediators produced by the mast cells could play a role in the regulation of myofibroblast differentiation and function. We used a human mast cell line, HMC-1, to determine if mast cells can induce normal human dermal fibroblasts to differentiate into functional myofibroblasts in vitro. We monitored the differentiation process by assaying two properties of the myofibroblast phenotype: expression of alpha-smooth muscle actin and functional capacity to contract a collagen matrix. In both a simple coculture system and in a skin-equivalent culture system, HMC-1 cells induced alpha-smooth muscle actin expression by fibroblasts. HMC-1 cells also stimulated fibroblast contraction of collagen gels, and the relative amount of contraction was dependent upon the number of HMC-1 cells present. To characterize the individual contributions made by specific mast cell products, we examined the effects of histamine, tumor necrosis factor alpha, and tryptase. Histamine induced a clear increase in alpha-smooth muscle actin expression, but it did not appear to stimulate fibroblast contraction. Tumor necrosis factor alpha had no effect in either assay. Purified human tryptase induced alpha-smooth muscle actin expression, and blocking the proteolytic activity of tryptase with specific inhibitors reduced that response. Tryptase inhibitors also eliminated the ability of HMC-1 cells to stimulate fibroblast contraction, suggesting that tryptase secreted by the HMC-1 cells may be one of the active mast cell mediators.     

Immunohistochemical analysis of keratinocyte growth factor and fibroblast growth factor 10 expression in psoriasis

The pathogenic mechanism underlying the hyperproliferation of keratinocytes in psoriasis is still not completely clarified. The production of cytokines released by activated T lymphocytes infiltrating the upper dermis probably has a crucial role. Even dermal fibroblasts can participate in the process through the secretion of growth factors, and some studies have reported an increased expression of the insulin-like growth factor 1. Few studies, however, have focused on the possible involvement of the keratinocyte growth factor (KGF/FGF-7) and the fibroblast growth factor 10 (FGF-10/KGF-2), which are secreted by fibroblasts and stimulate keratinocyte proliferation acting through a receptor specifically expressed by epithelial cells. The aim of this study was to investigate the expression of KGF and FGF-10 on the skin of patients with psoriasis by immunohistochemical analysis and to evaluate the correlation with the lymphocyte infiltrate and the epidermal proliferation. Immunostaining for KGF and FGF-10 showed that both the growth factors are upregulated in the upper dermis of psoriatic skin, and that the expression is correlated with the presence of T-cell infiltrate and with keratinocyte proliferation. Our data suggest that in psoriatic lesions activated lymphocytes can stimulate fibroblasts to produce KGF and FGF-10, which in turn contribute to sustain the hyperproliferative status of the keratinocytes.     

Glu-Leu-Arg-negative CXC chemokine interferon gamma inducible protein-9 as a mediator of epidermal-dermal communication during wound repair

Normal wound healing is a complex, highly regulated dynamic process that requires co-ordinate responses of both epidermal and dermal compartments. To accomplish the healing process several growth factors, chemokines, and matrix elements signal both cell proliferation and migration during the inflammatory and reparative phases and limit these responses during the remodeling phase. We have found that the Glu-Leu-Arg-negative CXC chemokines interferon gamma inducible protein 10, monokine induced by interferon gamma, and platelet factor 4, limit fibroblast responsiveness to growth factors, but the functioning of these factors in wound healing remains uncertain. We hypothesized that the keratinocyte-derived member of this Glu-Leu-Arg-negative CXC family, interferon gamma inducible protein 9 (IP-9) CXCL11 (also known as I-TAC, beta-R1, and H-174) signals to the dermal compartment to synchronize the re-epithelialization process. Interferon gamma inducible protein 9 was produced after mechanical wounding of a keratinocyte monolayer, suggesting for the first time that this could be a wound response factor. Interferon gamma inducible protein 9 limited epidermal growth factor (EGF)-induced fibroblast motility (57+/-7%) by the same protein kinase A (KA)-mediated inhibition of calpain activation and cell de-adhesion as described for interferon gamma inducible protein 10. Surprisingly, interferon gamma inducible protein 9 enhanced growth factor-induced motility in undifferentiated keratinocytes (137+/-19%) as determined in a two-dimensional in vitro wound healing assay, and interferon gamma inducible protein 9 alone promoted motility in undifferentiated keratinocytes (49+/-10% of epidermal growth factor-induced motility). A stimulated keratinocyte/target cell coculture system revealed that interferon gamma inducible protein 9 acts as a soluble keratinocyte-derived paracrine factor for both fibroblasts and keratinocytes. Further, we found that in both fibroblasts and undifferentiated keratinocytes, interferon gamma inducible protein 9 exerted its action through modulation of a cytosolic protease, calpain. Interestingly, interferon gamma inducible protein 9 increased calpain activity in undifferentiated keratinocytes, whereas the same chemokine inhibited the calpain activity in fibroblasts. This provides for a model whereby redifferentiated basal keratinocytes could limit fibroblast repopulation of the dermis underlying healed wounds while simultaneously promoting re-epithelialization of the remaining provisional wound.     

Transforming growth factor-beta1 and ultraviolet A1 radiation increase production of vascular endothelial growth factor but not endothelin-1 in human dermal fibroblasts

BACKGROUND: Normal and dysregulated wound healing involves fibroblast activation and angiogenesis, in which polypeptide factors such as transforming growth factor (TGF)-beta, vascular endothelial growth factor (VEGF) and endothelin-1 (ET-1) play an important part. Ultraviolet (UV) A1 (365 nm) has recently received attention as a possible treatment for some dermal fibrotic disorders. OBJECTIVES: The aim of this study was to evaluate the effects of TGF-beta1 and UVA1 radiation, as well as that of cobalt chloride, reported to mimic hypoxia both in vivo and in vitro, on the expression of VEGF and ET-1 by cultured human dermal fibroblasts. METHODS: Levels of VEGF and ET-1 were measured by enzyme-linked immunosorbent assay and expression of neutral endopeptidase (NEP, CD10), known to degrade ET-1, was quantified by flow cytometric analysis after cell trypsinization. RESULTS: Our results showed that the cells released minor amounts of VEGF and ET-1. Both TGF-beta1 and UVA1 strongly increased VEGF secretion in a dose- and time-dependent manner, without significantly affecting ET-1 release. Irradiation of TGF-beta1-stimulated fibroblasts resulted in a synergistic effect on increasing levels of VEGF but not ET-1 after 48 h. Cobalt chloride stimulated the secretion of VEGF by fibroblasts; the effects of TGF-beta1 and cobalt were additive. However, no significant effect of cobalt chloride on ET-1 secretion was observed, suggesting that ET-1 production in fibroblasts is not oxygen-sensitive. The expression of NEP was not modified by TGF-beta1 or UVA1 radiation. Addition of a neutralizing anti-CD10 antibody to fibroblast cultures downregulated CD10 expression at the cell surface without changing ET-1 levels in cell supernatants after 24 or 48 h. This suggests that membrane-bound NEP has minimal or no activity against secreted ET-1. CONCLUSIONS: Taken together, these results underline the major role played by TGF-beta1 in increasing VEGF secretion by fibroblasts. This, as well as the documented effect of UVA1 on increasing VEGF production, may have implications for wound healing in vivo.     

Inhibition of cytokine secretion from human leukemic mast cells and basophils by H1- and H2-receptor antagonists

H1-type antihistamines have recently been reported to inhibit cytokine secretion from human and murine mast cells and basophils. In order to confirm and expand these studies, we have compared several H1-blockers and the H2-blocker ranitidine for their effect on TNF-alpha, IL-3, 6, 8 and GM-CSF release from human leukemic mast (HMC-1) and basophilic (KU812) cells, compared to dexamethasone. Cells were stimulated for 24 h with phorbol myristate acetate (25 ng/ml) and calcium ionophore A 23187 (2.5x10(-7) M) alone or with the drugs added at 10(-4) to 10(-15) M, and production of cytokines was measured by ELISA. All antihistamines caused a dose-dependent inhibition of TNF-alpha release from HMC-1 cells, with maximal effects at 10(-12) M for azelastine, 10(-9) M for loratadine and cetirizine, and 10(-8) M for ranitidine. The inhibitory potency of H1-blockers on cytokines from HMC-1 cells was TNF-alpha >IL-8> or =IL-6> or =IL-3, with no significant effects on GM-CSF. In KU812 cells which failed to secrete TNF-alpha and GM-CSF, the sequence was IL-6 >IL-8 after preincubation. Dexamethasone inhibited all cytokines, but ranitidine only TNF-alpha and IL-3. Antihistamines had no effect on calcium flux in resting or stimulated cells. At the mRNA level, inhibition was only seen with KU812 cells and IL-8 in the presence of azelastine at 10-(10) M. These data show thus distinct inhibitory patterns for different antihistamines during cytokine production from human mast cells and basophils which may contribute to the anti-inflammatory effects of these drugs during treatment of allergic diseases.     

Establishment of basal cell carcinoma in culture: evidence for a basal cell carcinoma-derived factor(s) which stimulates fibroblasts to proliferate and release collagenase

The connective tissue adjacent to basal cell carcinomas (BCC) is frequently abnormal and contains increased numbers of fibroblasts and increased extractable collagenase. To determine whether BCC could produce these alterations by releasing mediators that regulated fibroblast function, we established BCC in culture and tested the ability of their culture supernatants to alter fibroblast proliferation and production of collagenase. Using tissue culture plates coated with type IV collagen and containing x-irradiated 3T3 feeder cells, we established epithelial colonies from 47% of the BCC cultured. The BCC-derived colonies differed from normal epidermal cell colonies in their morphology, growth rate, and keratin production. Culture supernatants from 4 out of 5 confluent BCC-derived colonies contained factors that stimulated fibroblasts to proliferate and release collagenase. These findings show that BCC-derived epidermal cell colonies release mediators which alter fibroblast functions and suggest that some of the connective tissue changes associated with BCC in vivo are the result of BCC-fibroblast interactions.     

Keratinocytes suppress transforming growth factor-beta1 expression by fibroblasts in cultured skin substitutes

Transforming growth factor (TGF)-beta1 is a multipotent growth factor with an important role in tissue homeostasis. This growth factor regulates cell proliferation, adhesion, migration and differentiation, as well as extracellular matrix deposition. The temporal secretion and activation of latent TGF-beta1 is thus of major importance to physiological and pathological processes and in wound healing and tumour formation. Cultured skin substitutes, as used to treat extensive acute or chronic skin wounds, offer an attractive model to investigate cellular interactions in cytokine and growth factor expression and response in vitro. In the present investigation, expression of TGF-beta1 was analysed in keratinocyte, fibroblast and melanocyte monolayer cultures, as well as in the dermal vs. epidermal components of reconstituted human skin. Immunohistology, enzyme-linked immunosorbent assay (ELISA) and Northern blotting were used to demonstrate expression at the RNA and protein level. In the monolayer cultures, levels of TGF-beta1 synthesized by melanocytes were observed to be considerably elevated when compared with keratinocytes. Most TGF-beta1, however, was secreted by fibroblasts. The relative contribution of the epidermal and dermal components of the skin substitutes to overall TGF-beta1 levels was determined by comparing results obtained for either component in the presence and absence of fibroblasts and keratinocytes. From results obtained by ELISA it was apparent that TGF-beta1 levels generated predominantly by fibroblasts within the skin substitutes were greatly reduced over time in the presence of keratinocytes. Suppression of fibroblast TGF-beta1 expression in the presence of keratinocytes was also demonstrable at the RNA level by Northern blotting. Results obtained by immunohistochemistry suggest that most, if not all, of the growth factor was present in the latent form. It is therefore most likely that the observed effect results from a factor secreted by keratinocytes, which is capable of suppressing TGF-beta1 synthesis by fibroblasts. These results suggest that expression of TGF-beta1 by fibroblasts is downregulated by paracrine actions of keratinocytes in healing skin.