BACKGROUND

Keloids are thick fibrous scars that are refractory to treatment and unique to humans. The lack of keloid animal models has hampered development of effective therapies. The authors' goal was to develop an animal model of keloids using grafted engineered skin substitutes composed of keloid-derived cells. To demonstrate the model's utility, differences between deep and superficial keloid fibroblasts were investigated.

METHODS

Engineered skin substitutes were prepared using six combinations of cells: 1, normal <em>keratinocytes</em> and normal fibroblasts; <em>2</em>, normal <em>keratinocytes</em> and deep keloid fibroblasts; 3, normal <em>keratinocytes</em> and superficial keloid fibroblasts; 4, keloid <em>keratinocytes</em> and normal fibroblasts; 5, keloid <em>keratinocytes</em> and deep keloid fibroblasts; and 6, keloid <em>keratinocytes</em> and superficial keloid fibroblasts. Engineered skin substitutes stably grafted to athymic mice were evaluated for wound area, thickness, and gene expression.

RESULTS

Deep keloid fibroblasts displayed elevated expression of type 1 collagen alpha 1 (COL1A1), transforming growth factor β-1, periostin, plasminogen activator inhibitor <em>2</em>, and inhibin beta A compared with superficial keloid fibroblasts and normal fibroblasts. After grafting, engineered skin substitutes in group 5 were significantly thicker than controls and had increased COL1A1 expression. Engineered skin substitutes in group 6 showed significantly increased area. Histologic analysis revealed abnormal collagen organization in engineered skin substitutes containing deep keloid fibroblasts or superficial keloid fibroblasts.

CONCLUSIONS

Aspects of the phenotypes of engineered skin substitutes prepared with keloid cells are analogous to thickening and spreading of human keloid scars. Therefore, use of keloid engineered skin substitutes is a valuable new tool for the study of keloid scarring.

BACKGROUND

Neuropilin 1 (NRP1) is expressed on several cell types including neurons and endothelial cells, where it functions as an important regulator in development and during angiogenesis. As a cell surface receptor, NRP1 is able to bind to members of the VEGF family of growth factors and to secreted class 3 semaphorins. Neuropilin 1 is also highly expressed in keratinocytes, but the function of NRP1 in epidermal physiology and pathology is still unclear.

RESULTS

To elucidate the role of NRP1 in skin in vivo we generated an epidermis-specific neuropilin 1 knock out mouse model by using the Cre-LoxP-System. Mice were viable and fertile and did not display any obvious skin or hair defects. After challenge with UVB irradiation, we found that deletion of epidermal NRP1 leads to increased rates of apoptosis both in vitro and in vivo. NRP1-deficient primary keratinocytes cultured in vitro showed significantly higher rates of apoptosis 24 hours after UVB. Likewise, there is a significant increase of active caspase 3 positive cells in the epidermis of Keratin 14-Cre-NRP1 (-/-) mice 24 hours after UVB irradiation. By Western Blot analysis we could show that NRP1 influences the cytosolic levels of Bcl-2, a pro-survival member of the Bcl-2 family. After UVB irradiation the amounts of Bcl-2 decrease in both protein extracts from murine epidermis and in NRP1-deficient keratinocytes in vitro, whereas wild type cells retain their Bcl-2 levels. Likewise, levels of phospho-Erk and Rac1 were lower in NRP1-knock out keratinocytes, whereas levels of pro-apoptotic p53 were higher.

CONCLUSIONS

NRP1 expression in keratinocytes is dispensable for normal skin development. Upon UVB challenge, NRP1 contributes to the prevention of keratinocyte apoptosis. This pro-survival function of NRP1 is accompanied by the maintenance of high levels of the antiapoptotic regulator Bcl-2 and by lower levels of pro-apoptotic p53.

Bone morphogenetic proteins (BMPs), which were originally identified by their novel ability to induce de novo cartilage and bone formation in vivo, are multifunctional proteins structurally related to transforming <em>growth</em> facto-beta s, activins, and inhibins. As a first step to elucidate the precise physiological function as well as the action mechanism of BMPs, we have examined the distribution of the specific cellular binding proteins for BMP-<em>2</em> on a wide variety of cell types. A single class of high affinity-specific binding sites for BMP-<em>2</em> were identified not only on osteoblastic cells but also on major types of non-hematopoietic cells in a rather ubiquitous fashion (1,<em>2</em>00-60,000 receptors/cell, Kd = 35-<em>2</em>30 pM); these cells included fibroblasts, <em>keratinocytes</em>, astrocytes, kidney epithelial cells, and tumor cells of bone, muscle, lung, liver, kidney, stomach, colon, prostate, and neuronal tissue. Other <em>growth</em> <em>factors</em> including transforming <em>growth</em> <em>factor</em>-beta 1, activin A, and inhibin A did not compete for the binding of 1<em>2</em>5I-labeled BMP-<em>2</em> to the cells. Affinity cross-linking of radiolabeled BMP showed five components with apparent molecular masses of 170, 105, 90, 80, and 70 kDa common to all three fibroblast cell lines analyzed. On the other hand, no specific binding sites for BMP-<em>2</em> were identified on vascular endothelial cells or on hematopoietic cells including RPMI 1788 and RPMI 8<em>2</em><em>2</em>6 (B-lymphocyte lineage), MOLT-3 and MOLT-4 (T-lymphocyte lineage), HL-60 (myeloid lineage), and K-56<em>2</em> (erythroid lineage). These results suggest that major types of cells other than hematopoietic cells and vascular endothelial cells may be potential targets for BMP-<em>2</em> action.

Two <em>growth</em> inhibitors were identified in culture medium conditioned by a human <em>keratinocyte</em> cell line, HaCat. TGF-beta was detected in media conditioned by <em>growing</em> or confluent HaCat cells, as well as in media conditioned at physiological (1 mM) or low (0.03 mM) Ca<em>2</em>+ concentrations. However, a considerable part of transforming <em>growth</em> <em>factor</em> beta (TGF-beta) in media conditioned at a physiological Ca<em>2</em>+ concentration was in active form, whereas most TGF-beta in media conditioned at a low Ca<em>2</em>+ concentration was latent. The other <em>growth</em>-inhibitory activity, which was detected only in media conditioned by confluent cells at a physiological Ca<em>2</em>+ concentration, was purified to homogeneity by a four-step procedure. The N-terminal amino acid sequence of the 33-kDa protein was identical with that of insulin-like <em>growth</em> <em>factor</em> binding protein-6 (IGFBP-6). Purified IGFBP-6 inhibited the <em>growth</em> of HaCat and Balb/MK <em>keratinocyte</em> cell lines, as well as Mv1Lu cells. The <em>growth</em> activity was also demonstrated by human recombinant IGFBP-6. In summary, HaCat cells secrete at least two possible autocrine <em>growth</em> inhibitors: TGF-beta which is secreted constitutively, but activated in a Ca(<em>2</em>+)-dependent manner, and IGFBP-6 which is secreted in a cell density- and Ca(<em>2</em>+)-dependent manner.

Matriptase, a membrane-associated serine protease, plays an essential role in epidermal barrier function through activation of the glycosylphosphatidylinositol (GPI)-anchored serine protease prostasin. The matriptase-prostasin proteolytic cascade is tightly regulated by hepatocyte <em>growth</em> <em>factor</em> activator inhibitor (HAI)-1 such that matriptase autoactivation and prostasin activation occur simultaneously and are followed immediately by the inhibition of both enzymes by HAI-1. However, the mechanisms whereby matriptase acts on extracellular substrates remain elusive. Here we report that some active matriptase can escape HAI-1 inhibition by being rapidly shed from the cell surface. In the pericellular environment, shed active matriptase is able to activate hepatocyte <em>growth</em> <em>factor</em> (HGF), accelerate plasminogen activation, and shed syndecan 1. The amount of active matriptase shed is inversely correlated with the amount of antithrombin (AT) bound to the surface of the <em>keratinocytes</em>. Binding of AT to the surface of <em>keratinocytes</em> is dependent on a functional heparin binding site, Lys-1<em>2</em>5, and that the N-glycosylation site Asn-135 be unglycosylated. This suggests that β-AT, and not α-AT, is responsible for regulation of pericellular matriptase activity in <em>keratinocytes</em>. <em>Keratinocytes</em> appear to rely on AT to regulate the level of pericellular active matriptase much more than breast and prostate epithelial cells in which AT regulation of matriptase activity occurs at much lower levels than <em>keratinocytes</em>. These results suggest that <em>keratinocytes</em> employ two distinct serine protease inhibitors to control the activation and processing of two different sets of matriptase substrates leading to different biological events: 1) HAI-1 for prostasin activation/inhibition, and <em>2</em>) AT for the pericellular proteolysis involved in HGF activation, accelerating plasminogen activation, and shedding of syndecans.

In various immunological disorders the pathomechanisms of tissue damage are causally associated with specific patterns of locally produced cytokines. To study the molecular and cellular mechanisms involved in the manifestation of psoriasis vulgaris we have assessed the cytokine mRNA profile expressed in lesional psoriatic skin and in T cell clones (TCC) that were established from skin lesions of patients with psoriasis. As demonstrated by use of the polymerase chain reaction (PCR), psoriasis lesions consistently exhibit transcription of a complex pattern of cytokines. It includes mediators selectively produced by T lymphocytes [interferon (IFN)-gamma, tumor necrosis <em>factor</em> (TNF)-beta, interleukin (IL)-<em>2</em>, IL-3 and IL-5] as well as cytokines secreted by various cell types [transforming <em>growth</em> <em>factor</em> (TGF)-alpha/-beta, TNF-alpha, IL-6/-8 and granulocyte-macrophage-colony stimulating <em>factor</em>], while IL-4 is missing. With the exception of TGF-alpha, this cytokine profile was also observed in lesional psoriatic T cell clones yielding supernatants mitogenic for <em>keratinocytes</em> in vitro (MTCC), but not in T cell clones yielding supernatants that inhibited <em>keratinocyte</em> proliferation (STCC). The congruent cytokine expression of psoriatic skin lesions and MTCC emphasizes that inflammation in psoriasis is driven by a sofar unrecognized regulatory T cell subset that may serve to control epidermal regeneration and convey immunosurveillance over epithelial surfaces. It is characterized by the combined expression of IFN-gamma, TGF-beta, IL-<em>2</em> and IL-5 in the absence of IL-4 and by its selective capacity to enhance <em>keratinocyte</em> proliferation. This newly defined combination of regulatory properties of a distinct T cell population cannot be reconciled with an immune response of the T helper cells (TH)0, TH1 or TH<em>2</em> type.

Transforming <em>growth</em> <em>factor</em>-beta s (TGF-beta s) are potent regulators of cell <em>growth</em> and differentiation. Expression of the closely related TGF-beta subtypes in vivo is differentially regulated both temporally and spatially. Members of the steroid hormone superfamily may play an important role in this gene- and tissue-specific regulation. We have shown that anti-estrogens induce the production of TGF-beta 1 in mammary carcinoma cells and fetal fibroblasts, whereas retinoic acid specifically induces TGF-beta <em>2</em> in primary epidermal <em>keratinocytes</em>. The induction of TGF-beta <em>2</em> by retinoids is accompanied by an increase in TGF-beta <em>2</em> mRNAs, but little change in transcription rates, suggesting an effect of retinoids on message stability or processing. In contrast, TGF-beta 1 mRNA levels are unchanged by anti-estrogen treatment, suggesting these compounds may regulate the translatability of the TGF-beta 1 message or some post-translational processing event. We have identified a stable stem-loop structure in the 5' untranslated region (UTR) of the TGF-beta 1 mRNA that inhibits translation of a heterologous reporter gene, and we are investigating the possibility that anti-estrogens may regulate the activity of this element, and hence the translatability of the TGF-beta 1 message. A significant fraction (<em>2</em>5-90%) of the TGF-beta induced by retinoids and anti-estrogens is in the biologically active rather than the latent form. We have shown that active TGF-beta has a much shorter in vivo half-life than latent TGF-beta, suggesting that the TGF-beta induced by retinoids and steroids may act locally at the site of production. Since many tumor cells retain sensitivity to the <em>growth</em> inhibitory effects of active TGF-beta, the use of members of the steroid hormone superfamily for inducing this potent <em>growth</em> inhibitor locally at the tumor site may have therapeutic potential.

Feeder cells of irradiated mouse fibroblasts are commonly used for, and are generally necessary for, the in vitro maintenance and <em>growth</em> of many fastidious cell types, particularly embryonic stem cells or induced pluripotent stem cells. Quantitative and semiquantitative immunoassays of conditioned media were performed to identify some of the soluble cytokines, chemokines, protein hormones, and cell matrix/adhesion molecules that are elaborated from two commonly used feeder cells, STO and CF-1. Among those quantitatively assayed, the most abundant cytokine proteins expressed by the feeder cells were activin A, hepatocyte <em>growth</em> <em>factor</em> (HGF), insulin-like <em>growth</em> <em>factor</em> 1, insulin-like <em>growth</em> <em>factor</em> <em>2</em>, insulin-like <em>growth</em> <em>factor</em> binding protein (IGFBP)-6, macrophage colony-stimulating <em>factor</em> (a.k.a. CSF-1), and pigment epithelium-derived <em>factor</em> (a.k.a. serine protease inhibitor, clade F, member 1). CF-1 cells expressed ten times more activin A than STO cells and also produced larger amounts of interleukin-6 and IGFBP-<em>2</em>, IGFBP-3, IGFBP-4, and IGFBP-5. Conversely, STO cell produced almost ten times more HGF and five times more stem cell <em>factor</em> (a.k.a. c-kit ligand) than CF-1 cells. Assayed semiquantitatively, relatively large amounts of chemokines were produced by both feeder cells including fractalkine (CX3CL1), interferon-inducible protein 10 (a.k.a. CXCL10 and cytokine-responsive gene-<em>2</em>, CRG-<em>2</em>), monocyte chemotactic protein (MCP)-1 (a.k.a. CCL<em>2</em> and junctional epithelium chemokine (JE), MCP-5/CCL1<em>2</em>), <em>keratinocyte</em>-derived chemokine (a.k.a. CXCL1 and <em>growth</em>-related oncogene alpha, GROα), nephroblastoma overexpressed gene (CCN3, IGFBP-9), stromal cell-derived <em>factor</em> 1 (CXCL1<em>2</em>), and serpin E1 (PAI-1). In contrast to one another, STO produced more CXCL16 than CF-1 cells, and CF-1 cell produced more MCP-5 (CCL1<em>2</em>), macrophage inflammatory protein (MIP)-1α (CCL3), MIP-1β (CCL4), pentraxin-3 (TSG-14), and platelet <em>factor</em>-4 (CXCL4) than STO cells. Soluble adhesion molecule, sICAM (ICAM-1, CD54), was expressed by CF-1 cells, but not STO cells, and similarly, the cell matrix-associated molecules endocan (endothelial cell-specific molecule 1), endostatin (collagen XVIII), and matrix metalloproteinase 3 were expressed more by CF-1 cells. Tissue inhibitor of metalloproteinases 1 was robustly expressed by both feeder cells. Other proteins primarily detected from CF-1 cells included retinol-binding protein 4 and FGF<em>2</em>1, while STO cells secreted more interferon gamma. Both feeder cells produced no or low amounts of LIF, tumor necrosis <em>factor</em> alpha, vascular endothelial <em>growth</em> <em>factor</em> (VEGF), VEGF-B, prolactin, various interleukins, fibroblast <em>growth</em> <em>factor</em> (FGF)-1, FGF-<em>2</em>, FGF-7, EGF, HB-EGF, and amphiregulin. The results may explain some of the cell <em>growth</em> and maintenance responses by various types of cells co-cultured on STO or CF-1 feeder cells.

By stimulating collagen synthesis and myofibroblasts differentiation, transforming <em>growth</em> <em>factor</em>-beta (TGF- beta) plays a pivotal role in tissue repair and fibrosis. The early <em>growth</em> response-1 (Egr-1) transcription <em>factor</em> mediates profibrotic TGF-beta responses, and its expression is elevated in biopsies from patients with scleroderma. NGF1-A-binding protein <em>2</em> (Nab<em>2</em>) is a conserved transcriptional co<em>factor</em> that directly binds to Egr-1 and positively or negatively modulates Egr-1 target gene transcription. Despite the recognized importance of Nab<em>2</em> in governing the intensity of Egr-1-dependent responses, the regulation and function of Nab<em>2</em> in the context of fibrotic TGF-beta signaling is unknown. Here we show that TGF-beta caused a time-dependent stimulation of Nab<em>2</em> protein and mRNA in normal fibroblasts. Ectopic expression of Nab<em>2</em> in these cells blocked Egr-1-dependent transcriptional responses, and abrogated TGF-beta-induced stimulation of collagen synthesis and myofibroblasts differentiation. These inhibitory effects of Nab<em>2</em> involved recruitment of the NuRD chromatin remodeling complex to the COL1A<em>2</em> promoter and were accompanied by reduced histone H4 acetylation. Mice with targeted deletion of Nab<em>2</em> displayed increased collagen accumulation in the dermis, and genetic or siRNA-mediated loss of Nab<em>2</em> in fibroblasts was associated with constitutively elevated collagen synthesis and accentuation of Egr-1-dependent TGF-beta responses in vitro. Expression of Nab<em>2</em> was markedly up-regulated in skin biopsies from patients with scleroderma, and was localized primarily to epidermal <em>keratinocytes</em>. In contrast, little Nab<em>2</em> could be detected in dermal fibroblasts. These results identify Nab<em>2</em> as a novel endogenous negative regulator of Egr-1-dependent TGF-beta signaling responsible for setting the intensity of fibrotic responses. Defective Nab<em>2</em> expression or function in dermal fibroblasts might play a role in persistent fibrotic responses in scleroderma.

OBJECTIVE

SSc is associated with an increased prevalence of atherosclerosis (ATS). This study assessed the prevalence of subclinical ATS as measured by carotid US and explored serum proteins to identify potential biomarkers of SSc-ATS.

METHODS

Forty-six SSc female patients and 46 age- and ethnicity-matched controls underwent carotid US to assess the presence of plaque and carotid intima media thickness (CIMT). Abstracted data included demographics, ATS risk factors and serum measurements [cholesterol, proinflammatory high-density lipoprotein (piHDL), CRP, lipoproteins]. Serum cytokines/proteins analyses included circulating type I IFN activity by quantifying IFN-inducible genes, soluble junctional adhesion molecule A (sJAM-A) and 100 serum proteins by using a microplate-based multiplex platform. Proteins significant at P < 0.05 on bivariate analyses for the presence of plaque were used to develop a composite measure.

RESULTS

Patients with SSc had more plaque (45.6% vs 19.5%, P = 0.01) but similar CIMT compared with controls. Multiplex analysis detected significant associations between serum proteins of inflammation, vasculopathy and fibrosis with ATS in SSc, including IL-2, IL-6, CRP, keratinocyte growth factor, intercellular adhesion molecule 1, endoglin, plasminogen activator inhibitor 1 and insulin-like growth factor binding protein 3 associated with carotid plaque. Myeloid progenitor inhibitory factor 1, serum amyloid A, thrombomodulin, N-terminal pro-brain natriuretic peptide (BNP), and Clara cell secretory protein 16 kD correlated with CIMT. The median composite score for the plaque group was 6 and for the no plaque group it was 2 (P < 0.0001).

CONCLUSIONS

Patients with SSc have a higher prevalence of carotid plaque than matched controls, and patients with SSc-plaque vs patients without plaque have elevated serum proteins implicated in both vasculopathy and fibrosis. Further studies are needed to evaluate the role of these proteins in SSc compared with healthy controls.

Based on evidence that arsenic modulates proinflammatory events that are involved in skin carcinogenecity, we hypothesized that in normal human epidermal <em>keratinocytes</em> (NHEK) arsenic increases expression of the procarcinogenic enzyme cyclooxygenase-<em>2</em> (COX-<em>2</em>) and that this occurs via specific mitogen and stress signaling pathways. To test this hypothesis, NHEK were exposed to sodium arsenite, and COX-<em>2</em> expression, prostaglandin E<em>2</em> (PGE(<em>2</em>)) secretion, mitogen-activated protein kinase (MAPK) phosphorylation, and DNA synthesis were quantified. Inhibitors of p4<em>2</em>/44 and p38 MAPKs were used to evaluate the contribution of mitogen and stress signaling to the modulation of COX-<em>2</em>. Our results demonstrate that arsenite (0.005-5 microM) elevates COX-<em>2</em> expression, PGE(<em>2</em>) secretion (<em>2</em>.5-5 microM), and DNA synthesis (1-5 microM). Arsenite stimulated p4<em>2</em>/44 but not p38 MAPK phosphorylation (<em>2</em>.5 microM), responses different than those produced by epidermal <em>growth</em> <em>factor</em>. Inhibition of mitogen-activated protein kinase kinase (MAPKK) and p38 MAPK using PD98059 (<em>2</em>0 microM) and SB<em>2</em>0<em>2</em>190 (5 microM), respectively, attenuated the elevation of COX-<em>2</em> protein induced by arsenite, whereas physiological concentrations of three COX-<em>2</em> inhibitors (e.g., NS-398, piroxicam, and aspirin) reduced arsenite-stimulated DNA synthesis. These data indicate that arsenite elevates COX-<em>2</em> in NHEK at the transcriptional and translational levels as well as increases PGE(<em>2</em>) secretion. Compounds that inhibit COX-<em>2</em> expression and activity may be useful in the scientific study, prevention, and treatment of arsenic skin carcinogenesis and deserve further investigation.

Low oxygen tension has recently been shown to stimulate cell <em>growth</em> and clonal expansion, as well as synthesis and transcription of certain <em>growth</em> <em>factors</em> and extracellular matrix components. These results have been obtained by exposing cell cultures to a hypoxic environment. Using an oxygen probe, we have now studied how experimental conditions affect the oxygen tension detectable at the cell surface. Dissolved oxygen tension was directly related to the height of the medium above the cell surface (r = 0.8793, P = 0.0<em>2</em>1), but was constant when no cells were present in the flask (r = -0. 973<em>2</em>, P = 0.001). In both human dermal fibroblasts and NIH/3T3 cultures, oxygen tension decreased linearly as cell density increased (r = -0.835, P < 0.0001; r = -0.916, P < 0.0001, respectively). When human dermal fibroblasts were exposed to <em>2</em>% O(<em>2</em>), maximum hypoxic levels (0 mmHg) were achieved within approximately 15 min, and the recovery time was within a similar time frame. The addition of rotenone, an inhibitor of cellular respiration, blocked this decrease in oxygen tension at the cell surface, suggesting that cellular consumption of oxygen is responsible for the decline. Finally, we examined the cell-surface oxygen tension in control and acutely wounded human skin equivalents (HSE), consisting of a <em>keratinocyte</em> layer over a type I collagen matrix containing fibroblasts. We found that oxygen tension dropped significantly (P < 0.0001) in acutely wounded areas of HSE as compared to unwounded areas of HSE and that this drop was prevented by the addition of mitomycin C. These results indicate that cell-surface oxygen tension is indirectly related to cell density, and that the amount of detectable oxygen at the cell surface is a function of cell density, the oxygen tension in the incubator, and increased cellular activity, as occurs after injury.

Reversible phosphorylation of CTP:phosphocholine cytidylyltransferase, the rate-limiting enzyme of phosphatidylcholine biosynthesis, is thought to play a role in regulating its activity. In the present study, the hypothesis that proline-directed kinases play a major role in phosphorylating cytidylyltransferase is substantiated using a c-Ha-ras-transfected clone of the human <em>keratinocyte</em> cell line HaCaT. Cellular extracts from epidermal <em>growth</em> <em>factor</em>-stimulated HaCaT cells and from ras-transfected HaCaT cells phosphorylated cytidylyltransferase much stronger as compared with extracts from quiescent HaCaT cells. The tryptic phosphopeptide pattern of cytidylyltransferase phosphorylated by cell-free extracts from ras-transfected HaCaT cells was similar compared with the patterns of cytidylyltransferase phosphorylated by p44mpkmitogen-activated protein kinase and p34cdc<em>2</em> kinase in vitro, whereas in the case of casein kinase II the pattern was different. Furthermore, in c-Ha-ras-transfected HaCaT cells the in vivo phosphorylation state of cytidylyltransferase was <em>2</em>-fold higher as compared with untransfected HaCaT cells. This higher phosphorylation of cytidylyltransferase in the ras-transfected clone was reduced to a level below the phosphorylation of cytidylyltransferase in untransfected cells, using olomoucine, a specific inhibitor of proline-directed kinases. The reduced phosphorylation of cytidylyltransferase in olomoucine-treated cells correlated with an enhanced stimulation of enzyme activity by oleic acid.

An early cellular event in the development of psoriatic lesions is infiltration of target tissue by macrophages and activated T lymphocytes. Lesional psoriatic skin contains activated memory T lymphocytes with production of mRNA for lymphokines such as interleukin-<em>2</em>, interferon-gamma, and tumor necrosis <em>factor</em>-alpha that is elevated relative to normal or uninvolved psoriatic skin. That the T-cell activation and cellular lymphokine production have a crucial role in the maintenance of epidermal hyperplasia in the psoriatic lesion is indicated by the beneficial effect of immunosuppressive agents in the treatment of psoriasis (cyclosporin A, FK506, anti-CD3, anti-CD4). A link between immune activation and psoriasis is also indicated by immunogenetic associations in this disease. Also, psoriatic <em>keratinocytes</em> appear to have been modulated by T-cell lymphokines in vivo, because they abnormally express molecules uniquely induced on <em>keratinocytes</em> by the T-cell product interferon-gamma. Indeed, T cells producing interferon-gamma have been cloned from psoriatic lesions, and they are able to induce <em>keratinocyte</em> class II major histocompatibility complex and intercellular adhesion molecule expression. These lesion-derived T-cell clones can induce <em>growth</em> of <em>keratinocytes</em>, and specifically lesional psoriatic T cells produce <em>factors</em> that induce increased <em>keratinocyte</em> colony formation, as well as increased cell cycle entry of the normally quiescent stem cell population. Interferon-gamma, although a <em>growth</em> inhibitor on its own, acts cooperatively with other T-cell-produced <em>growth</em> <em>factors</em> to cause <em>keratinocyte</em> <em>growth</em> induction. Furthermore, relative to normal stem cells, <em>keratinocyte</em> stem cells (beta 1 integrin+ K1/K10-) in psoriatic uninvolved epidermis are significantly hyperresponsive to the <em>growth</em>-stimulatory lymphokine milieu created by lesional T lymphocytes. Whether such abnormalities in responsiveness are associated with new genetic linkages reported in families of psoriasis patients is unknown. As the epidermis of lesional psoriatic skin can be demonstrated to produce elevated levels of <em>factors</em> that can further potentiate T-cell activation, a self-sustaining cycle can be constructed of T-cell recruitment, intralesional activation, release of <em>factors</em> that preferentially stimulate psoriatic epidermal stem cells to proliferate, and further epidermal potentiation of the T-cell-mediated lesions.

The oral use of chewing tobacco has greatly increased in recent years, and this usage is associated with cancers of the mouth, lip, nasal cavities, esophagus and gut. Oral cancer accounts for 3% of all cancers in U.S.A. and is the seventh most common cancer. Previous studies in our laboratory have demonstrated the protective abilities of a novel IH636 grape seed proanthocyanidin extract (GSPE) against reactive oxygen species both in vitro and in vivo models, and provided significantly better protection as compared to vitamins C, E and beta-carotene. In the recent past, we have demonstrated smokeless tobacco (STE)-induced oxidative stress, apoptotic cell death in a primary culture of normal human oral <em>keratinocytes</em> (NHOK), and have compared the protective abilities of vitamins C and E, singly and in combination, and GSPE in this pathobiology [Free Rad. Biol. Med., <em>2</em>6, 99<em>2</em>-1000 (1999)]. In the present study, we have assessed the protective role of vitamins C and E, and GSPE against STE-induced modulation of intracellular oxidized states in NHOK cells as demonstrated by laser scanning confocal microscopy. Approximately 11%, <em>2</em>6%, <em>2</em>8% and 50% protection were observed following incubation with vitamin C, vitamin E, a combination of vitamins C plus E, and GSPE, respectively. DNA fragmentation was assessed as an index of oxidative DNA damage and similar results were observed. Furthermore, the cellular viability and functional roles of Bcl-<em>2</em>, p53 and c-myc genes were assessed in STE-induced oxidative stress in NHOK cells. NHOK cells were treated with STE (0-<em>2</em>00 micrograms/ml) for <em>2</em>4 h and changes in the expression of Bcl-<em>2</em>, p53 and c-myc genes were measured by reverse transcriptase-polymerase chain reaction (RT-PCR), and the protective effect of GSPE was assessed. Approximately a <em>2</em>.0-fold increase in p53 gene expression was observed following incubation of the oral <em>keratinocytes</em> with 100 micrograms/ml of STE, beyond which the expression of p53 decreased, confirming increased apoptotic cell death with a higher concentration of STE as reported earlier. GSPE significantly modulated STE-induced changes in p53. The expression of antiapoptotic Bcl-<em>2</em> gene decreased with STE treatment and the expression of Bcl-<em>2</em> gene increased significantly following preincubation with GSPE. No significant change in the expression of transcription <em>factor</em> c-myc gene responsible for cell cycle <em>growth</em> was observed following incubation with STE and/or GSPE. Thus, c-myc may not be involved in STE-induced cytotoxicity towards NHOK cells. These results suggest that antioxidant protection of STE-induced cellular injury is associated with alterations in Bcl-<em>2</em> and p53 expression.

<em>Keratinocyte</em> <em>growth</em> <em>factor</em> receptor (KGFR), also known as FGFR<em>2</em> IIIb, is a splice variant of FGFR-<em>2</em>. KGFR is expressed in many types of epithelial cell and is activated with four known ligands [FGF-1, FGF-3, FGF-7 (also known as KGF) and FGF-10] that are predominantly synthesized by mesenchymal cells. KGFR is highly expressed in the late-proliferative phase of a normal endometrium and in endometrial adenocarcinoma. In the present study, we attempted to determine the expression and localization of KGFR in human cervical cancer cell lines and cervical cancer tissues. The KGFR protein was detected in CaSki and HeLa cells, but not in ME-180 cells of cervical cancer cell lines. In non-cancer cervical tissues, KGFR immunoreactivity was weakly expressed in the surface of squamous epithelial cells and vascular smooth muscle cells. Immunohistochemically, the KGFR protein was detected in squamous cell carcinoma in 36 of 4<em>2</em> (86%) cervical cancer patients. In cervical cancer tissues, KGFR was detected in 17 of 18 (94%) of patients with the keratinizing type and 19 of <em>2</em>4 (79%) of patients with the non-keratinizing type of cervical cancer. Furthermore, KGFR was prominently localized in proliferating reserve cells and squamous metaplastic reserve cells adjacent to cancer cells. In contrast, KGFR was not detected in cervical ductal cells in cancer or non-cancer cervical tissues. These findings may indicate that KGFR mediates the <em>growth</em> and differentiation of reserve cells and squamous cell carcinoma in the cervix.

Alveolar epithelial cell (AEC) migration mediated by matrix metalloproteinases (MMPs) is required for lung development and repair after injury such as hyperoxia. Of specific interest in lung remodeling are the gelatinases, which are upregulated in AEC after hyperoxia. We correlated migration and gelatinase production in AEC cultured from fetal, adult, and hyperoxic rats. Fetal AEC (19-<em>2</em>0 days) had higher MMP-<em>2</em> and MMP-9 gelatinase expression than adult AEC, with fivefold higher MMP-9 activity, and were migratory through gelatin, responding to epidermal <em>growth</em> <em>factor</em>, <em>keratinocyte</em> <em>growth</em> <em>factor</em>, and fibroblast <em>growth</em> <em>factor</em>-10. MMP-<em>2</em> and MMP-9 expression and migratory activity could be detected from the time of plating. In contrast, adult AEC migrated and expressed MMP-<em>2</em> and MMP-9 proteins only after 48 h of culture. AEC from hyperoxic rats were significantly more migratory through gelatin than control adult AEC, with significantly higher MMP-9 activity. Inhibition of MMPs with doxycycline reduced the migration of AEC from hyperoxic rats to the level of control adult AEC. Fibronectin-cultured "hyperoxic" AEC acquired a temporary capacity for migration similar to the A549 lung cancer cell line, which is both highly migratory and invasive and is derived from the AEC type <em>2</em> lineage. These data suggest that MMP activity is associated with a migratory phenotype in fetal, hyperoxic, and transformed AEC in vitro, and we speculate that MMPs may play a key mechanistic role in AEC migration in vivo during lung development and repair.

Apoptosis with premature termination of hair follicle <em>growth</em> induces several types of hair loss and is one of the crucial <em>factors</em> of hair loss. Astragaloside IV, which is a major component of Astragalus membranaceus, is a cycloartane triterpene saponin. Although an anti-apoptotic effect of Astragaloside IV has been reported, its effects against hair loss have not been investigated. To explore the underlying mechanisms of Astragaloside IV on apoptotic signaling in hair follicle, the dorsal skin of depilated C57BL/6 mice was topically treated with 1 and 100 μM Astragaloside IV for 14 days. In Astragaloside IV-treated group, TUNEL-positive cells were reduced. We found that Astragaloside IV blocked the procaspase-8, resulting in the inhibition of caspase-3 and procaspase-9 activities. The changes were accompanied with down-regulation of Bax and p53, and up-regulation of Bcl-<em>2</em> and Bcl-xL by Astragaloside IV treatment. In addition, activation of NF-κB and phosphorylation of IκB-α were inhibited, along with decreases in three MAPKs: ERK, SAPK/JNK and p38 by Astragaloside IV. The expressions of KGF, p<em>2</em>1, TNF-α and IL-1β, which are <em>keratinocyte</em> terminal differentiation markers associated with catagen, were modulated by treatment with Astragaloside IV. These results demonstrated that Astragaloside IV is concerned with blocking the Fas/Fas L-mediated apoptotic pathway, which would be an alternative therapy for hair loss.

Induction of PAI-<em>2</em> by <em>2</em>,3,7,8-tetrachlorodibenzo-p-dioxin (TCDD) has been studied in human primary hepatocytes, hepatoma HepG<em>2</em> cells and monocytic U937 cells, extending recent findings in human <em>keratinocytes</em>. PAI-<em>2</em> represents a serpine-type protease inhibitor with wide-ranging implications in fibrinolysis, extracellular matrix proteolysis, <em>growth</em> <em>factor</em> activation and carcinogenesis. PAI-<em>2</em> was induced by >10(-9) M TCDD in hepatocytes and HepG<em>2</em> cells and by >10(-10) M TCDD in U937 cells. In the latter cell line, PAI-<em>2</em> induction by TCDD and by 1<em>2</em>-O-tetradecanoyl phorbol-13-acetate (TPA) has been compared. TCDD appeared to be less efficient than TPA as an inducer of PAI-<em>2</em>. In contrast to induction by TPA, PAI-<em>2</em> induction by TCDD was found to be biphasic, with an early peak of mRNA at 1-3 h and a late peak at 1<em>2</em>-<em>2</em>4 h. A biphasic response was also seen at the protein level although production of PAI-<em>2</em> protein lagged behind the corresponding mRNA. PAI-<em>2</em> is known to contain AP-1 sites, i.e. Jun/Fos protein-binding sites, in its promotor region. Hence, PAI-<em>2</em> induction by TCDD has originally been conceived to be due to an indirect response, secondary to the induction of Jun/Fos proteins. Therefore, expression of jun/fos genes and their AP-1 activity were studied at the early phase of PAI-<em>2</em> induction by TCDD. TCDD did not increase mRNA of c-fos, c-jun, junB or junD (in contrast to TPA which markedly increased the expression of c-fos and junB), nor did TCDD increase AP-1 activity. In conclusion, the findings suggest that PAI-<em>2</em> induction by TCDD is not restricted to human <em>keratinocytes</em> but includes liver cells and monocytic U937 cells. The induction mechanism is complex but the early phase does not appear to involve Jun/Fos proteins.

During terminal differentiation, <em>keratinocytes</em> lose the ability to divide. One indicator of responsiveness to certain <em>growth</em> <em>factors</em> is a transient rise in the intracellular concentration of free calcium ions ([Ca<em>2</em>+]i). The aim of our experiments was to discover whether or not terminally differentiating <em>keratinocytes</em> have lost the ability to exhibit an increase in [Ca<em>2</em>+]i in response to <em>factors</em> that stimulate [3H]thymidine incorporation and increase [Ca<em>2</em>+]i in undifferentiated <em>keratinocytes</em>. [Ca<em>2</em>+]i was measured with the calcium indicator dye FURA-<em>2</em> and by a ratio imaging method. Expression of involucrin, a precursor of the <em>keratinocyte</em> cornified envelope, was used as a marker of terminal differentiation. Measurements were made on stratified colonies of cells grown in standard medium (containing 1.8 mM calcium ions) and on cell monolayers in low calcium medium (0.1 mM). Treatment of serum-starved monolayers with substance P, bombesin or complete <em>growth</em> medium containing 10% fetal calf serum resulted in increased [3H]thymidine incorporation. A switch from low calcium to standard medium also stimulated [3H]thymidine incorporation whether or not the cells had been serum-starved. In each experiment some cells showed an increase in [Ca<em>2</em>+]i while others did not. However, the heterogeneity in the [Ca<em>2</em>+]i response did not reflect the terminal differentiation status of individual cells: both involucrin-positive and -negative cells were found in the responding and nonresponding populations. Involucrin-positive and -negative areas of stratified cultures also underwent a transient increase in [Ca<em>2</em>+]i in response to serum-containing medium. Our data therefore indicate that both proliferating (involucrin-negative) and post-mitotic, terminally differentiating (involucrin-positive) <em>keratinocytes</em> can respond to mitogenic stimuli by an increase in [Ca<em>2</em>+]i.(ABSTRACT TRUNCATED AT <em>2</em>50 WORDS)

Primary cultures of mouse <em>keratinocytes</em> maintain a basal cell phenotype in 0.05 mM Ca<em>2</em>+ medium, while culture in 1.4 mM Ca<em>2</em>+ results in terminal differentiation and inhibition of DNA synthesis. Induction of differentiation by Ca<em>2</em>+ results in a 10- to <em>2</em>0-fold increase in the expression of transforming <em>growth</em> <em>factor</em>-beta <em>2</em> (TGF-beta <em>2</em>) mRNA and peptide, but a decrease in the expression of TGF-beta 1. In contrast, binding and cross-linking analyses show that the number of available surface 80 kilodalton (kDa) and 65 kDa TGF-beta receptor types decrease during differentiation. However, a mild acid wash significantly increases the number of available receptor sites on the differentiated <em>keratinocytes</em>, indicating that the TGF-beta receptors are unavailable for binding due to masking by endogenous ligand. A significant level of TGF-beta <em>2</em> secretion and receptor binding occur before the decrease in DNA synthesis, suggesting that the inhibition of DNA synthesis associated with differentiation of <em>keratinocytes</em> is mediated through the production and autocrine action of TGF-beta <em>2</em>.

We previously reported data regarding the mechanism of neoplastic transformation in JB6 Cl41 mouse skin epidermal cells. However, experimental in vitro models for studying neoplastic transformation of human cells could provide further insight into the mechanisms of human cancer development. In this study, we have established a neoplastic transformation model with HaCaT cells, a human <em>keratinocyte</em> cell line, and showed the usefulness of this cell line for studying the mechanisms of neoplastic transformation. Epidermal <em>growth</em> <em>factor</em> (EGF) treatment induced a dose-dependent anchorage-independent cell transformation in HaCaT cells. Furthermore, PD98059, a mitogen-activated protein (MAP) kinase/ERK kinase (MEK) inhibitor, or SP6001<em>2</em>5, c-Jun N-terminal kinase (JNK) inhibitor, decreased cell <em>growth</em>, EGF-induced DNA synthesis and transformation. Unlike observations in the JB6 mouse epidermal cell model, SB<em>2</em>03580, a stress-activated protein kinase-<em>2</em>/p38 alpha and beta (p38) inhibitor, increased EGF-induced transformation in HaCaT cells. These results suggest that extracellular-signal regulated kinase (ERK), JNK, or p38 are implicated in EGF-induced neoplastic transformation of human cells.

OBJECTIVE

Total parenteral nutrition (TPN) induces epithelial cell (EC) apoptosis. <em>Keratinocyte</em> <em>Growth</em> <em>Factor</em> (KGF) increases EC-<em>growth</em>; however, little is known of its effect on apoptosis. This study aims to determine if mRNA expression of Bcl-<em>2</em> proteins (major mediators of epithelial cell apoptosis) is altered with TPN, and if KGF-administration influences Bcl-<em>2</em> family expression.

METHODS

C57BL/6J mice (n = 6 per group) received oral feeding (control), TPN (TPN), or TPN plus intravenous KGF daily (TPN + KGF). After 7 days, intestine was harvested and EC isolated. Apoptosis was identified using flow cytometry. EC mRNA expression of Bcl-<em>2</em> family members was measured by reverse transcriptase polymerase chain reaction; Bcl-<em>2</em> protein level was measured by immunoblot analysis.

RESULTS

EC apoptotic rates were: control, 14.4% +/- 5.1%; TPN, <em>2</em>9.4% +/- 11.3%; KGF, 17.<em>2</em>% +/- 5.6%. Pro-apoptotic Bcl-<em>2</em> proteins changed minimally with TPN or KGF; however, the antiapoptotic protein Bcl-<em>2</em> changed significantly: control, 0.78 +/- 0.<em>2</em>4; TPN, 0.10 +/- 0.13; KGF, 0.76 +/- 0.36. EC Bcl-<em>2</em> protein levels were: control, 0.16 +/- 0.13; TPN 0.18 +/- 0.16; and TPN + KGF 0.47 +/- 0.19.

CONCLUSIONS

TPN-induced apoptosis decreased Bcl-<em>2</em> mRNA expression. KGF decreased EC apoptosis and increased Bcl-<em>2</em> expression. Modalities to increase endogenous KGF, or KGF-administration may have benefit in patients on TPN.

Mice lacking the alpha6 integrin chain die at birth with severe skin blistering. To further study the function of alpha6 integrin in skin, we generated conditionally immortalized cell lines from the epidermis of wild-type and alpha6 deficient mouse embryos. Mutant cells presented a decreased adhesion on laminin 5, the major component of the basement membrane in the skin, and on laminins 10/11 and <em>2</em>. A DNA array analysis revealed alterations in the expression of extracellular matrix (ECM) components including laminin 5, cytoskeletal elements, but also membrane receptors like the hemidesmosomal components integrin beta4 and collagen XVII, or <em>growth</em> <em>factors</em> and signaling molecules of the TGFbeta, EGF, and Wnt pathways. Finally, an increase of several epidermal differentiation markers was observed in cells and tissue at the protein level. Further examination of the mutant tissue revealed alterations in the filaggrin signal. These differences may be linked to an upregulation of the TGFbeta and the Jun/Fos pathways in mutant <em>keratinocytes</em>. These results are in favor of a role for integrin alpha6beta4 in the maintenance of basal <em>keratinocyte</em> properties and epidermal homeostasis.