Production of IL-3 by non-transformed primary neonatal murine keratinocytes: evidence for constitutive IL-3 gene expression in neonatal epidermis

Immunity at the surface: homeostatic mechanisms of the skin immune system

The coordinated function of multiple epidermal and dermal cell populations allows the skin immune system to respond rapidly and effectively to a wide variety of insults occurring at the interface of the organism and its environment. Keratinocytes are the first line of defense in the skin immune system, and keratinocyte-derived cytokines are pivotal in mobilizing leukocytes from blood and signaling other cutaneous cells. Cytokine-mediated cellular communication also enables dermal fibroblasts and endothelial cells lining the cutaneous vasculature to participate in immune and inflammatory responses. Skin is an important site for antigen presentation, and both epidermal Langerhans cells and dermal dendritic cells play pivotal roles in T cell-mediated immune responses to antigens encountered in skin. Proinflammatory signaling pathways are necessarily balanced by a variety of regulatory pathways that help maintain the homeostatic functioning of the skin immune system.

Comparative study of the role of professional versus semiprofessional or nonprofessional antigen presenting cells in the rejection of vascularized organ allografts

The immune systems of transplant recipients are progressively challenged with exposure to the multiple lineages of donor cells that comprise the vascularized organ allograft. Each lineage of such donor tissue constitutively expresses or can be induced to express varying densities of MHC antigens ranging from no expression of MHC to MHC class I only to both MHC class I and class II. In addition, the cell surface expression of a diverse assortment of costimulatory and cell adhesion molecules also varies in density in a tissue specific fashion within the allograft. The MHC class I/II molecules displayed on the donor cells contain within their clefts a constellation of processed protein antigens in the form of peptides derived from intracellular and to some extent extracellular sources. Therefore, the potential for each cell lineage to induce alloactivation and serve as a target for allospecific immune responses is dependent on the diversity and density of peptide-bearing MHC molecules, costimulatory molecules, and cell adhesion molecules. In addition, the T cell receptor repertoire of the recipient also contributes to the magnitude of the allogeneic response. Consequently, the variety of clinical outcomes following organ transplantation even with the institution of potent immunosuppressive (drug) therapies is not surprising, as it appears reasonable for such therapies to influence the allogeneic response against distinct lineages differentially. Our failure to prevent chronic human allograft rejection may therefore be due to our limited appreciation of the full spectrum of alloactivating experiences encountered by host T cells as they interact with donor cells of diverse tissue lineages. Investigations by our laboratory of the immunopathogenesis of chronic cardiac allograft rejection have revealed an intrinsic inability of human cardiac myocytes to process and present antigens, not only for primary but also for secondary alloimmune responses. One obvious explanation for this phenomenon is the fact that cardiac myocytes do not constitutively express MHC class II molecules and express only low levels of class I molecules. However, this immunological unresponsiveness is maintained even after the induction of MHC class II and upregulation of MHC class I on these cells by interferon-gamma (IFN-gamma). Similar results have also been reported for cells of different tissue lineages (e.g. chondrocytes, keratinocytes, neural cells). Until now, cells have been defined as professional or nonprofessional for the purposes of defining their potential for antigen presentation to T cells. Professional antigen presenting cells have been identified as cells that are of haematopoietic origin, that constitutively express MHC class I and class II molecules as well as potent costimulatory molecules, and that are able to induce both primary and secondary immune responses, whereas nonprofessional antigen presenting cells are not bone marrow derived, do not constitutively express MHC class II, but may in some cases initiate primary and secondary immune responses after induction of MHC class II antigen by proinflammatory cytokines (e.g. IFN-gamma). The findings of our laboratory and others suggest that cells of certain lineages be considered in the separate class of 'nonantigen presenting cells'. Indeed, nonprofessional antigen presenting cells can be reclassified into three categories: semiprofessional-, nonprofessional-, or nonantigen presenting cells that are able to present antigen to and activate naive T cells, activated T cells, or no T Cells, respectively. The aim of this review is to identify and (re)examine the antigen presentation characteristics of cells of different tissue lineages in terms of their ability to activate different subsets of T cells. This approach is taken in an attempt to synthesize these concepts into a unified picture of T cell activation in the context of antigen processing and presentation by different cell types.

Failure to detect interleukin (IL)-3 mRNA or protein in human keratinocytes: antibodies to granulocyte macrophage-colony-stimulating factor or IL-6 (but not IL-3) neutralize "IL-3" bioactivity

Interleukin (IL)-3-like bioactivity has been found in culture supernatants from human and murine keratinocytes. However, there is controversy as to the presence of IL-3 mRNA in human keratinocytes. Using highly sensitive reverse transcription-polymerase chain reaction and enzyme-linked immunosorbent assay techniques, we examined human keratinocytes from four different donors (neonatal foreskins) and were unable to detect IL-3 mRNA or IL-3 protein. Despite successful amplification of DNA from an IL-3 cDNA, no product could be obtained by amplification of keratinocyte RNA treated with reverse transcription-polymerase chain reaction. Analysis of concentrated (up to 50-fold) supernatants failed to detect IL-3 protein by enzyme-linked immunosorbent assay. Because ultraviolet radiation up-regulates many cytokines, we irradiated human keratinocytes with 300 J/m2 ultraviolet B and collected supernatants 24 h post-irradiation. Supernatants concentrated 50-fold were also negative for IL-3 protein by enzyme-linked immunosorbent assay. When assayed on the IL-3-responsive M-07e cell line, unirradiated supernatants stimulated M-07e proliferation 22-fold over background levels. Irradiated supernatants stimulated M-07e proliferation 128-fold. Neither the unirradiated nor the irradiated supernatant activity could be neutralized with antibody to human IL-3. However, incubation of irradiated supernatants with antibody to granulocyte macrophage-colony-stimulating factor (GM-CSF) reduced the M-07e proliferation by 90%. Antibodies against GM-CSF and IL-6 completely abrogated proliferation. Reverse transcription-polymerase chain reaction confirmed a concomitant elevation of IL-6 (2.6- to 5.6-fold) and of GM-CSF mRNA (2.7- to 4.3-fold) at 6 and 24 h after ultraviolet B irradiation in keratinocytes, but no IL-3 amplification products could be detected. IL-3 mRNA was also not detected in adult keratinocytes. Even after stimulation by IL-1 alpha, tumor necrosis factor-alpha, or phobol myristate acetate, IL-3 mRNA was not detected in either neonatal or adult human keratinocytes. We have been unable to detect IL-3 mRNA or IL-3 protein in human keratinocytes. The IL-3-like activity in human keratinocytes is mainly due to GM-CSF, with a small contribution from IL-6.