Search

Phylogenetically conserved toll-like receptors (TLR) recognize "pathogen-associated molecular patterns". Upon binding of ligands, TLR initiate innate immune response in immune and most likely epithelial cells. The TLR4 isoform is considered as a lipopolysaccharide (LPS) receptor. As shown here, a rat-tongue-derived epithelial cell line RTE2 expressed TLR4 mRNA and functional protein. LPS-treated RTE2 cells responded with the transient expression of inducible nitric oxide synthase (iNOS), an effector protein of TLR4 involved in the innate immune defense of monocytes. iNOS induction occurred along a nuclear factor-kappaB (NF-kappab)-dependent pathway and correlated with the increased production of NO. Moreover, LPS and lipid A were potent inhibitors of proliferation of RTE2 cells, of mouse keratinocytes, and mouse epidermis in vivo. The inhibition depended on lipid A structure, i.e., it was related to the endotoxin activity of LPS and at least in vitro was in part mediated by NF-kappaB. C57Bl/10 ScCr mice, lacking a functional TLR4, did not respond with growth inhibition, strongly suggesting a TLR4-mediated effect. RTE2 proliferation was also inhibited by transforming growth factor beta (TGFbeta) and tumor necrosis factor alpha (TNFalpha), whereas interferon gamma (IFNgamma) was a weak inhibitor. But the growth-inhibitory effect of LPS on RTE2 cells was not mediated by TNFalpha, TGFbeta, or NO. It is concluded that besides induction of innate immune responses, LPS specifically induces growth arrest in epithelial tongue cells and keratinocytes in vitro and in mouse epidermis in a TLR4-dependent but cytokine- and NO-independent manner.

Cyclooxygenase (COX)-1 and -2 catalyze the key reaction in prostaglandin biosynthesis. Whereas COX-1 is found in most tissues, COX-2, with a few exceptions, is not expressed in normal tissues but becomes transiently induced in the course of inflammatory reactions. In many neoplastic epithelia, COX-2 is constitutively overexpressed. Here we show that COX isozymes are spatiotemporally expressed during morphogenesis of dorsal skin epithelium of NMRI mice. COX-1 and COX-2 mRNA and protein were detected in embryonic and postnatal epidermal tissue by RT-PCR, northern blot, and immunoblot analysis indicating that both isoforms may contribute to prostaglandin production. Being barely detectable in interfollicular epidermis and resting hair follicles of adult mice, COX-2 protein appeared in embryonic skin first in epidermal precursor cells and later on in the basal cells and the peridermal layer of the stratified epidermis. In the course of pelage hair follicle morphogenesis, COX-2 remained expressed in the basal interfollicular compartment and, in addition, became apparent in elongated hair germs and hair pegs and later on in the outer root sheath cells of the distal and proximal hair follicles as well as in basal sebaceous gland cells. During the subsequent synchronous phases of hair cycling, COX-2 expression declined in catagen, was barely detectable in telogen, and was reinduced in the basal outer root sheath and basal sebaceous gland cells of anagen hair follicles. COX-1 immunosignals were detected predominantly in the interfollicular spinous and granular layers of the developing, neonatal, and adult epidermis but not in follicular epithelial cells of developing or cycling hair follicles. Dendritic cells in the interfollicular epidermis and distal hair follicles were also COX-1-positive. Transgenic overexpression of COX-2 under the control of a keratin 5 promoter in basal cells of the interfollicular and follicular epidermis induced a precocious entry into the first catagen stage of postnatal hair follicle cycling and a subsequent disturbance of hair follicle phasing. Furthermore, transgenic mice developed an alopecia. Inhibition of transgenic COX-2 activity by feeding the specific COX-2 inhibitor valdecoxib suppressed the development of alopecia, indicating that COX-2-mediated prostaglandin synthesis is involved in hair follicle biology.

Increased cell volume, accumulation of lipid droplets in the cytoplasm, and nuclear degeneration are phenomena indicating terminal differentiation of human sebocytes followed by holocrine secretion and cell death. The molecular pathways of natural and induced sebocyte elimination are still unknown, however. In this study, SZ95 sebocytes were found to exhibit DNA fragmentation after a 6 h culture followed by increased lactate dehydrogenase release after 24 h, indicating cell damage. With the help of morphologic studies and using Oil Red detection of cellular lipids, cell enlargement, accumulation of lipid droplets in the cytoplasm, and nuclear fragmentation could be observed under treatment with arachidonic acid. Staurosporine, a potent inhibitor of phospholipid Ca2+-dependent protein kinase, increased externalized phosphatidylserine levels on SZ95 sebocytes, detected by annexin V/propidium iodide flow cytometry, as early as after 1 h, whereas dose-dependent reduction of bcl-2 mRNA and protein expression, enhanced DNA fragmentation, and increased caspase 3 levels, detected by caspase 3 inhibitor/propidium iodide flow cytometry, were found after 6 h of treatment. SZ95 sebocyte death was detected as early as after 6 h of SZ95 sebocyte treatment with high staurosporine concentrations (10(-6)-10(-5) M). 5Alpha-dihydrotestosterone (10(-8)-10(-5) M) did not affect externalized phosphatidylserine levels and DNA fragmentation in SZ95 sebocytes but slightly decreased lactate dehydrogenase cell release. Neither acitretin nor 13-cis retinoic acid (10(-8)-10(-5) M) affected externalized phosphatidylserine levels, DNA fragmentation, and lactate dehydrogenase cell release, despite the increased caspase 3 levels under treatment with 13-cis retinoic acid. The combined staurosporine and 13-cis retinoic acid treatment enhanced DNA fragmentation in SZ95 sebocytes to the same magnitude as in cells only treated with staurosporine. In conclusion, SZ95 sebocytes in vitro undergo apoptosis, which can be enhanced by the terminal differentiation inductor arachidonic acid or by staurosporine and leads to cell death. 5Alpha-dihydrotestosterone inhibits SZ95 sebocyte death without involving apoptotic pathways, and retinoids did not affect the programmed death of human sebocytes. The latter result fits well with the currently reported inability of normal skin cells to undergo apoptosis after treatment with retinoids, in contrast to their malignant counterparts.

In addition to their proinflammatory activities, prostaglandins recently have been shown to be beneficial in the resolution of tissue injury and inflammation. Thus, inhibition of cyclooxygenase-2, the predominant prostaglandin endoperoxide synthase under these conditions, may not only result in attenuating the inflammatory response but also in delaying tissue regeneration and repair. To this end, we investigated cyclooxygenase isozyme expression and the effects of cyclooxygenase inhibitors on wound healing upon full-thickness incisions in mouse skin. Immunohistochemical analysis revealed prominent expression of cyclooxygenase isozymes in keratinocytes of the hyperplastic epithelium, with cyclooxygenase-1 immunosignals predominating in the suprabasal compartment and cyclooxygenase-2 immunosignals spread throughout the whole epidermis. Moreover, dendritic cells, resembling Langerhans cells, as well as endothelial cells and macrophages in the vicinity of or within the granulation tissue were found to express both isozymes. Inhibition of prostaglandin E2 synthesis by oral administration of the cyclooxygenase-1-selective inhibitor SC-560 or the cyclooxygenase-2-selective inhibitor valdecoxib did not retard wound healing in mouse skin macroscopically. Except for a slight transient retardation of epithelialization early after wounding wound-induced neoangiogenesis, collagen deposition, and the restoration of tensile strength were not delayed by these agents. Likewise, the nonselective inhibitor indomethacin had no effect on the tensile strength of incisional skin wounds.