- Email: [email protected]
The Keratinocyte as a Target for Staphylococcal Bacterial Toxins
The Keratinocyte as a Target for Staphylococcal Bacterial Toxins Jeffrey B. Travers, David A. Norris,* and Donald Y. M. Leung²
Departments of Dermatology, Pediatrics, Pharmacology and Toxicology, Indiana University School of Medicine, Indianapolis, Indiana, U.S.A.; *Department of Dermatology, University of Colorado School of Medicine, and Department of Veterans Affairs Hospital, Denver, Colorado, U.S.A.; ²Departments of Pediatrics, The National Jewish Medical and Research Center, and The University of Colorado School of Medicine, Denver, Colorado, U.S.A.
the production of powerful pro-in¯ammatory lipidand protein-derived cytokines in keratinocytes. Characterization of interactions between these proteins and the keratinocyte can provide a better understanding of how bacterial infection modulates in¯ammatory skin diseases, as well as provide the basis for improved therapies involving antibacterial agents. Key words: a-toxin/keratinocytes/platelet-activating factor/protein A/Staphylococcus aureus/superantigens. Journal of Investigative Dermatology Symposium Proceedings 6:225±230, 2001
Skin infections with Staphylococcus aureus are not only an important cause of morbidity and even mortality, but are thought to serve as initiation and/or persistance factors for numerous in¯ammatory skin diseases, including psoriasis and atopic dermatitis. One mechanism by which S. aureus can modulate the immune system is through the production of proteins such as superantigenic toxins, Protein A, as well through the cytolytic a-toxin. This review serves to discuss the biology of these three types of proteins, with emphasis on their ability to stimulate
S
eosinophil, make the keratinocyte a potentially important player in the cutaneous in¯ammatory response.
kin infections with Staphylococcus aureus are an important cause of morbidity and even mortality. Infections by these bacteria not only cause disease directly, but are thought to be important triggers in worsening in¯ammatory skin diseases ranging from atopic dermatitis to psoriasis (reviewed by Leung et al, 1998). One mechanism by which S. aureus can accomplish this is through the production of soluble proteins with potent immunomodulatory effects. The objective of this review is to focus on one aspect of cutaneous S. aureus bacterial infection; how interactions between keratinocytes and bacterial proteins such as superantigens, a-toxin, and Protein A can potentially affect the in¯ammatory process.
Keratinocyte-derived cytokines Keratinocytes have the capacity to synthesize and release a wide variety of immunomodulatory molecules (a partial list is found in Table I). Keratinocyte-derived pro-in¯ammatory and immunomodulatory molecules can be divided into those that are lipid versus those that are protein based. Lipid-derived in¯ammatory mediators produced by keratinocytes include derivatives of phospholipids [plateletactivating factor (1-O-alkyl-2-acetyl glycerophosphocholine; PAF) and lyso-phosphatidic acid (LPA)], arachidonic acid metabolites (eicosanoids), and others including ceramides. These mediators tend to be synthesized by keratinocytes in response to membrane damage. Of signi®cance, essentially all of these lipid-derived mediators are produced in response to an increased calcium mobilization signal, which can be seen following membrane damage. The major eicosanoid synthesized by human keratinocytes is PGE2, with lesser amounts of PGD2 and PGF2-a (Pentland and Needleman, 1986). Injection of nanogram amounts of PGE2 causes erythema in human skin (Camp and Greaves, 1987). Larger amounts induce swelling through vasodilatory effects on blood vessels, with only minimal leukocyte chemoattractive properties. PGD2 also has erythrogenic effects, as well as granulocyte chemoattractant properties (Woodward et al, 1990). Though keratinocytes synthesize hydroxyeicosateranoic acids (HETE), they do not synthesize signi®cant levels of leukotrienes (Greaves and Camp, 1988). Produced by the subsequent actions of phospholipase A2 and PAF acetyltransferase, PAF is synthesized in response to the same stimuli that induce eicosanoid production (Pinckard et al, 1994). Intradermal injection of PAF results in a painful wheal and ¯are reaction within minutes. Human keratinocytes both synthesize
THE KERATINOCYTE AS AN ACTIVE PARTICIPANT IN CUTANEOUS INFLAMMATION Keratinocytes are by far the most abundant cell type in the epidermis, and our concept of the keratinocyte's role in in¯ammatory processes has changed greatly from a passive target to an active participant. The keratinocyte has the ability to produce and release numerous lipid and protein cytokines (reviewed by Greaves and Camp, 1988; Kupper, 1990). Keratinocytes also can express adhesion molecules that allow enhanced leukocyte interactions. These properties, as well as the relative long life of a keratinocyte compared with other in¯ammatory cells such as the neutrophil or Manuscript received February 13, 2001; accepted for publication February 19, 2001. Reprint requests to: Dr. Jeffrey B. Travers, The H.B Wells Center for Pediatric Research, Riley Hospital for Children Room 2659, 635 Barnhill Drive, Indianapolis, Indiana 46202. Email: [email protected] Abbreviations: CLA, cutaneous lymphocyte-associated antigen; MHC II, major histocompatibility class II protein; PAF, platelet-activating factor; PG, prostaglandin; SAg, superantigen; TCR, T cell receptor. 1087-0024/01/$15.00
´ Copyright # 2001 by The Society for Investigative Dermatology, Inc. 225
226
TRAVERS ET AL
JID SYMPOSIUM PROCEEDINGS
Table I. Immunomodulatory molecules produced by keratinocytes Lipid-derived
Protein-derived
Prostaglandins E2, D2, F2-a 12-Hydroxyeicosatetraenoic acid Platelet-activating factor Lyso-phosphatidic acid Ceramides
Interleukins 1, 6, 8, 12, 15, 18 Tumor necrosis factor-a Transforming growth factor-a Interferon-b Granulocyte and Granulocyte macrophage colony stimulatory factors
PAF as well as express functional PAF receptors (PAF-R) (Michel et al, 1990; Travers et al, 1995). Activation of the epidermal PAF-R results in the biosynthesis of numerous mediators, including eicosanoids, TNF-a, IL-6, IL-8, and PAF itself (Pei et al, 1998). Amongst the protein-derived cytokines produced by keratinocytes, TNF-a and IL-1 have been termed primary cytokines due to their ability to induce in¯ammatory reactions in part through the ability to induce a host of other lipid and protein mediators (Kupper, 1990). TNF-a and IL-1 both exert effects on receptors expressed on keratinocytes that are linked to the nuclear factor kB pathway, with resultant production of cascades of other protein and lipid cytokines. IL-6 is a pleiotropic cytokine with effects on both B and T cells (Paquet and Pierard, 1996). IL-7 has the ability to stimulate the growth of both progenitors and mature B and T cells (Heu¯er et al, 1993). IL-8 is a potent neutrophil chemoattractant also produced by keratinocytes. IL-12 is a potent TH1 cytokine that promotes activation of macrophages that can be synthesized by keratinocytes in response to varied stimuli including phorbol esters and IFN-g (Aragane et al, 1994). In addition, bacterial superantigens appear to induce production of skin-homing (cutaneous lymphocyte-associated antigen-positive; CLA) T cells through an IL-12-dependent process (Leung et al, 1995). Though resting keratinocytes do not synthesize the T cell activating cytokine IL-15, its production can be induced by IFN-g, a process inhibited by dexamethasone (Han et al, 1999). IL-18, a novel cytokine that strongly induces IFN-g production in T cells, has also been shown to be synthesized by human and murine keratinocytes. Hyperosmolar stress induced by mannitol or NaCl is a potent stimulus for IL-18 production in both epithelial and endothelial cells, suggesting that this cytokine could be important in damage-mediated in¯ammatory responses (Takeuchi et al, 1999). Thus, a keratinocyte that sustains membrane damage has the potential to synthesize and release numerous lipid and protein cytokines that could modulate the in¯ammatory response. Keratinocyte adhesion molecules In addition to the ability of keratinocytes to modulate in¯ammation through soluble mediators, these cells can also express membrane-associated molecules that can serve to bind immune cells as well as to stimulate signal transduction pathways leading to further production of mediators. Through its ability to bind the B-2 integrin leukocyte function-associated antigen found on most leukocytes, the adhesion molecule intracellular adhesion molecule1 (ICAM-1; CD54) is an important mediator of cellular immune reactions (reviewed by Krutmann and Grewe, 1995). Though not found on resting epidermal cells, ICAM-1 surface expression can be induced by cytokines including TNF-a (Krutmann et al, 1990) and PAF (Pei et al, 2000), as well as by physical methods including tapestripping (Nickoloff and Naidu, 1994). Similar to ICAM-1, resting keratinocytes do not express the costimulatory molecule B7-1 (CD80), yet epidermal expression can be seen during allergic and contact dermatitis (Wakem et al, 2000). The signi®cance of epidermal B7-1 expression is not clear, though this signaling molecule usually associated with antigen-presenting cells prevents
the induction of T cell anergy and induces IL-2 production during antigen presentation. Of note, transgenic mice overexpressing B7-1 on keratinocytes exhibit exaggerated and persistent contact hypersensitivity reactions to a variety of haptens (Gaspari et al, 1998). In addition to adhesion and costimulatory molecules, keratinocytes can also express MHC-II molecules (i.e., HLA-DR) in response to various pro-in¯ammatory and immune stimuli (Basham et al, 1983). MHC-II molecules can signal through intracellular calcium, as well as through kinases including protein kinases A and C (PKA and PKC) (Wakita et al, 1995; Etienne et al, 1999). Keratinocyte MHC-II molecules can also trigger the production of cytokines such as TNF-a (Tokura et al, 1994). BACTERIAL SUPERANTIGENS AND THE KERATINOCYTE The biology of superantigens The term ``superantigen'' was ®rst coined by White et al in 1989 and refers to a group of microbial and viral proteins that differ in several important respects from conventional peptide antigens (White et al, 1989). First, unlike conventional protein antigens, which are taken up and processed by antigen-presenting cells, superantigens exert their effects as globular intact proteins. Similar to peptide antigens, superantigens are presented by class II MHC molecules; however, they do not interact with the MHC peptide b antigen binding groove, but instead bind to conserved amino acid residues that are on the outer walls of the peptide antigen-binding groove. Thus, whereas recognition of conventional peptide antigens by the T cell receptor is restricted by MHC alleles, recognition of superantigens is not generally MHC restricted. Second, superantigens primarily recognize and bind to the variable region of the T cell receptor b chain (VB). This is in contrast to nominal peptide antigens, which require recognition by all ®ve variable elements (i.e., VB, DB, JB, Va, Ja) of the T cell receptor. Therefore, the responding frequency of a superantigen is several orders of magnitude greater than a conventional peptide antigen. The unique ability of a superantigen to bind directly to (and signal through) MHC class II molecules, and cross-link with a large percentage of T cells expressing relevant T cell receptor VB chains, provides an explanation for the potent immune stimulation seen with these molecules (Tiedemann and Fraser, 1996). The prototypical disease due to bacterial superantigens is toxic shock syndrome caused by staphylococcal exotoxins or streptococcal enterotoxins (Deresiewicz, 1997). Superantigens and in¯ammatory skin diseases As outlined above, bacterial superantigens possess several properties that can contribute to disease pathogenesis. Superantigens can bind to and activate MHC class II molecules, whether constitutively expressed on professional antigen presenting cells such as monocytes, Langerhans, or dendritic cells, or on nonprofessional antigenpresenting cells such as the keratinocyte following cytokine stimulation. Signi®cant pathologic effects can be seen due to either local or massive systemic release of protein cytokines as well as other pro-in¯ammatory mediators. In addition, proin¯ammatory effects would be further ampli®ed by subsequent activation of T cells. Of signi®cance for the ability of these proteins to induce cutaneous in¯ammation, recent evidence suggests that superantigenic stimulation of T cells promotes skin homing by upregulation of the skin homing receptor CLA, a modi®ed selectin that promotes localization of T cells to the skin. Leung et al reported that stimulation of peripheral blood mononuclear cells with bacterial superantigens resulted in a signi®cant increase of CLA+ T cells, but not T cells expressing other homing molecules, in comparison with treatment with other agents including phytohaemogluttinin or anti-CD3 (Leung et al, 1995). Bacterial toxins were also found to induce IL-12 production in this model system. Of note, induction of toxin-induced CLA expression was blocked by anti-IL-12, and the addition of IL-12 to phytohae-
VOL. 6, NO. 3 DECEMBER 2001
mogluttinin-stimulated T cells induced CLA, but not expression of the mucosal homing receptor a e b 7-integrin (Leung et al, 1995b). These data suggest that bacterial toxins induce the expansion of skin-homing CLA+ T cells in an IL-12-dependent manner, and thus may contribute to the development of skin rashes in superantigen-mediated diseases. Psoriasis is a chronic in¯ammatory skin disorder affecting up to 2% of the general population. The characteristic lesion of psoriasis is a well-demarcated erythematous papule or plaque containing hyperproliferating keratinocytes, as well as in®ltrating neutrophils, monocytes, and T lymphocytes (Nickoloff, 1991). Though considered an autoimmune disease, considerable evidence suggests an important role for bacteria in the pathogenesis of psoriasis. Colonization and/or infection with Staphylococcus and Streptococcus have been reported to exacerbate psoriasis (Henderson and Highet, 1988; Leung et al, 1993). Superantigenic toxins produced by these bacteria have been implicated in the initiation and/or propagation of psoriasis (reviewed by Leung et al, 1998). Indeed, a group of psoriatics whose skin disease was worsening were found to be secondarily infected with superantigen-expressing S. aureus, and skin biopsies revealed lesional T cell receptor VB skewing consistent with the pattern of the isolated superantigen (Leung et al, 1993). Injection of superantigen-stimulated immunocytes into normal skin from subjects with psoriasis xenografted onto an immunode®cient mouse has also been shown to induce clinical, histologic, and immunologic changes consistent with psoriasis (Boehncke et al, 1996; Wrone-Smith and Nickoloff, 1996). The most convincing clinical and experimental association between bacterial superantigens and psoriasis, however, is in patients with acute eruptive (guttate) psoriasis (Lewis et al, 1993; Leung et al, 1995). Of note, Leung et al demonstrated up to 50% T cell receptor VB2+ T cells in perilesional skin of subjects with acute eruptive psoriasis induced by a Streptococcal pyrogenic exotoxin C (SPEC) expressing Streptococcus (Leung et al, 1995). Altogether, these studies have implicated bacterial superantigens in both the initiation as well as the worsening of psoriasis. The keratinocyte as a target for superantigens Due to the lack of MHC II molecules, the resting keratinocyte is not thought to be a direct target for superantigens (Ezepchuk et al, 1996); however, treatment of resting keratinocytes with cytokines such as g-interferon has been shown to upregulate MHC II molecules (Basham et al, 1983). Superantigens have been shown to directly bind to and stimulate g-interferon-treated primary cultures of human and murine keratinocytes as well as epidermal carcinoma cell lines (Nickoloff et al, 1993; Strange et al, 1994; Tokura et al, 1994). Early biochemical events seen as a consequence of MHC II and superantigen interactions in keratinocytes include a transient intracellular calcium mobilization (Wakita et al, 1995). Superantigens have also been shown to induce the production of TNF-a in keratinocytes in vitro (Tokura et al, 1994; Ezepchuk et al, 1996). In addition to their well-characterized effects on MHC II molecules, superantigens have also been reported to interact with MHC class I molecules on squamous cell carcinoma cell lines (Haffner et al, 1996). Topical application of superantigens to atopic skin also induces acute eczematous changes and T cell recptor VB skewing (Skov et al, 2000) Hyperreactivity of psoriatic keratinocytes to topical superantigens Signi®cant evidence based on ®nding peripheral blood or skin T cells with TCR VB skewing and the isolation of a superantigen with a relevant TCR VB pro®le has linked superantigens to either initiatory events in acute guttate psoriasis (Lewis et al, 1993; Leung et al, 1995), or to worsening of psoriasis skin disease (Leung et al, 1993). Recent studies by Travers et al have examined the effects of topical application of nanogram amounts of the staphylococcal superantigens ZEB and TSST-1, and the streptococcal superantigens SPEA and SPEC to tape-abraded skin of psoriatic patients for 48 h (Travers et al, 1999). In¯ammatory responses to these puri®ed bacterial proteins were seen in both normal subjects as well as in psoriatics; however,
STAPHYLOCOCCAL TOXINS AND KERATINOCYTES
227
patients with psoriasis exhibited signi®cantly increased in¯ammatory skin responses in comparison with controls, or subjects with atopic dermatitis or lichen planus. Reactivity of psoriatic subjects to topical superantigens correlated with the amount of skin disease, and was greatest in subjects who exhibited the isomorphic (Koebner) phenomenon. Consistent with the clinical hyperactive response, in situ hybridization studies of skin biopsies obtained 6 and 24 h following patch testing with superantigens demonstrated increased epidermal TNF-a mRNA in psoriatics than controls. Though signi®cant numbers of mononuclear cells were seen in superantigen-induced skin reactions, surprisingly, skin biopsies from the lesions did not reveal increased numbers of T cells expressing the predicted T cell receptor VB type for the superantigens. This is in contrast to the T cell receptor VB skewing found after application of larger doses of superantigens on the intact skin of subjects with atopic dermatitis (Skov et al, 2000). In addition, immunohistochemical studies revealed increased epidermal MHC-II expression in reactions in psoriatics compared with control skin. Given that superantigens have the ability to induce TNF-a production with MHC-II expressing keratinocytes in vitro (Tokura et al, 1994; Ezepchuk et al, 1996), one explanation for the increased clinical reactivity and increased epidermal TNF-a production found in psoriatic subjects, would be that the superantigens were stimulating cytokine production from MHC-II positive keratinocytes, with T cell accumulation being in response to these keratinocyte-derived cytokines. This would also explain the lack of T cell receptor VB skewing as well. Consistent with this notion that epidermal MHC-II±superantigen interactions were responsible for the enhanced skin reactions in psoriatic subjects, a mutant TSST-1 toxin that does not bind to MHC-II (TSST1G31S/S32P) did not induce skin in¯ammation, even at 10-fold higher concentrations than native superantigens (Travers et al, 1999). The lack of skin reactions to the mutant TSST-1 also suggest that the exotoxin-induced skin responses were not due to a delayed type hypersensitivity reaction to a conventional antigen. The ability of superantigens to stimulate the production of potent pro-in¯ammatory and cytotoxic cytokines such as TNF-a in MHC-II positive keratinocytes provides an alternative mechanism by which colonization or actual infection with S. aureus can induce skin in¯ammation. This pathway could be relevant in psoriasis, as psoriatic lesions are often colonized with Staphylococcus and Streptococcus (Leung et al, 1993; Ezepchuk et al, 1996), and scratching of the skin would be expected to mimic tape-stripping, by both compromising the barrier function as well as activating the epidermis (Nickoloff and Naidu, 1994). This alternative mechanism may explain why some psoriatic patients report improvement of their disease activity following treatment with oral antibiotics (Rosenberg et al, 1986). It may also account for several reports that fail to demonstrate selective expansion of T cell receptor VB in the skin of patients with plaque psoriasis (SchmittEgenolf et al, 1991; Boehncke et al, 1995). Thus, epicutaneous application of superantigens on psoriatic skin may result in an acute in¯ammatory, keratinocyte-mediated response rather than superantigen-mediated T cell activation. STAPHYLOCOCCAL a-TOXIN, PROTEIN A, AND THE KERATINOCYTE In addition to toxins that can have profound immunomodulatory effects through their abilities to act as superantigens, S. aureus can produce other proteins that can act as virulence factors and affect host responses. The production of transmembrane pores by staphylococcal a-toxin can induce a potent cytotoxic response that also results in cytokine production. Though not associated with a direct cytotoxic response, staphylococcal Protein A can also induce cytokine production in keratinocytes. In contrast to superantigens, both a-toxin and Protein A can interact with resting (i.e., MHC II-negative) keratinocytes.
228
TRAVERS ET AL
The biology of a-toxin Staphylococcal a-toxin is released by bacteria as a 293 residue single chain polypeptide. Upon interaction with a target cell, a-toxin forms a heptamer, which results in the formation of a transmembrane pore (Song et al, 1996). Recent studies using biotinylated a-toxin in susceptible versus resistant cells suggest that the signaling and cytotoxic effects are related to the ability of the a-toxin heptamer to form a transmembrane channel (Valeva et al, 1997). Formation of transmembrane pores results in membrane damage as well as an in¯ux of extracellular calcium, a potent damage-related stimulus for many cells. Similar to calcium ionophores such as A23187, small amounts of a-toxin result in signaling and potential repairable damage, whereas larger amounts result in cell death. In endothelial cells as well as in PC-12 cells, a-toxin activates PLA2 via a calcium-dependent process (Fink et al, 1989). In endothelial cells, a-toxin stimulates arachidonic acid release, as well as PAF biosynthesis (Suttorp et al, 1992; Grimminger et al, 1997). Staphylococcal a-toxin has also been reported to enhance IL-6 and G-CSF production in IL-1b-treated endothelial cell cultures (Soderquist et al, 1998). Exposure of isolated perfused rat hearts to a-toxin was shown to induce coronary vasoconstriction and loss of myocardial contractility (Sibelius et al, 2000). Signi®cant levels of thromboxane A2 were found in the perfusate following a-toxin treatment. Of note, indomethacin, acetylsalicylic acid, and a speci®c thromboxane receptor antagonist inhibited a-toxin effects, demonstrating involvement of endogenous thromboxane A2 in this model system. Thus, a-toxin can induce organ damage not only through direct cytolytic effects, but also through the ability to trigger cellular production of pro-in¯ammatory and cytotoxic soluble mediators. The keratinocyte as a target for a-toxin Staphylococcus aureus is not only a common agent of cutaneous infections, but can also be isolated from many patients with in¯ammatory skin diseases such as atopic dermatitis and psoriasis. A recent study by Ezepchuk and colleagues demonstrated that nine of 19 strains of S. aureus obtained from clinical lesions of atopic dermatitis and psoriasis contained atoxin (Ezepchuk et al, 1996). Ezepchuk et al also found that treatment of primary cultures of human keratinocytes or the keratinocyte-derived cell line HaCaT with a-toxin resulted in signi®cant TNF-a production. Staphyloccoccal a-toxin also induced cytotoxic effects in keratinocytes that were more compatible with necrosis than apoptosis (Walev et al, 1993; Ezepchuk et al, 1996). Injection of rabbit corneas with a-toxin has been shown to result in signi®cant tissue edema with death of corneal epithelial cells by both necrosis and apoptosis (Moreau et al, 1997). Though a-toxin is thought to have considerable deleterious effects, it has been suggested that a-toxin-induced keratinocyte damage can protect against superantigen-mediated effects by inhibiting the ability of the keratinocyte to activate T cells (Tokura et al, 1997). Allapatt et al recently examined the ability of a-toxin to stimulate lipid mediator production in human carcinoma cell lines.1 Using select ion monitoring gas chromatography mass spectrometry, signi®cant production of PAF as well as arachidonic acid release was found in a-toxin-treated HaCaT cells. The amount of PAF produced in response to 10 mg per ml a-toxin was about 25% of that seen in response to a potent ionophore such as A23187 (Travers et al, 1995). Because activation of the epidermal PAF-R can induce arachidonic acid release (Pei et al, 1998), the potential involvement of endogenous PAF in a-toxin-induced arachidonic acid release was tested using a model system created by retroviralmediated transduction of the PAF-R-negative epithelial cell line KB with the human PAF-R. As shown in Fig 1, a-toxin stimulated arachidonic acid release in both control (PAF-Rnegative) KBM as well as PAF-R-positive KBP cells; however, the 1 Allapatt C, Leung DYM, Johnson C, Clay K, Cosgrove J, Travers JB: Acute keratinocyte damage results in the biosynthesis of the lipid mediator platelet-activating factor. J Invest Dermatol 112:543, 1999 (abstr)
JID SYMPOSIUM PROCEEDINGS
Figure 1. Staphylococcal a-toxin stimulates increased arachidonic acid release in PAF-R expressing versus control KB cells. KB cells transduced with the PAF-R (KBP) or control retrovirus (KBM) were stimulated with 10 mg per ml puri®ed a-toxin or vehicle for 30 min, and arachidonic acid released into the supernatant was quantitated by gas chromatography mass spectrometry using deuterated arachidonic acid as internal standard as previously described (Pei et al, 1998). The data shown are mean 6 standard deviation of arachidonic acid released from duplicate samples from a representative experiment of four conducted.
Figure 2. Interactions between Staphylococcal proteins and the keratinocyte. The keratinocyte has the potential to be a target for S. aureus proteins. Enterotoxins that act as superantigens (SAg) can signal through MHC-II molecules found on activated keratinocytes, resulting in a calcium mobilization response as well as the production of TNF-a. Protein A does not appear to induce a calcium mobilization response, but is a potent inducer of keratinocyte TNF-a production through an unknown mechanism. Through its ability to form a heptameric transmembrane pore, a-toxin can trigger a calcium mobilization response as well as TNF-a production.
amounts of arachidonic acid released by a-toxin-treated KBP cells were signi®cantly greater than with KBM cells (Fig 1). Similarly, pretreatment of KBP cells with 10 mM of the PAF-R antagonist WEB 2086 partially inhibited a-toxin-induced arachidonic acid release in KBP but not KBM cells (data not shown). These data are compatible with the notion that a-toxin can generate both arachidonic acid release and PAF biosynthesis through calciummediated activation of PLA2, and that endogenous PAF generated can feed forward to induce further mediator production. Altogether, through its direct effects on keratinocyte membranes, a-toxin can induce the synthesis of both lipid and protein cytokines, as well as cause direct cytotoxic effects.
VOL. 6, NO. 3 DECEMBER 2001
The biology of Protein A The unique protein derived from the cell wall structure of S. aureus termed Protein A was discovered over 40 y ago. A characteristic property of this protein has been its ability to bind the Fc structure of IgG, making it a valuable laboratory reagent for the isolation and puri®cation of antibody molecules. Over time other biologic properties of this molecule have been elucidated. Protein A appears to have an unrivaled diversity of biologic activities ranging from immunomodulatory and antitumor to antifungal and antiparasitic effects (reviewed by Silverman, 1998). Protein A has mitogenic effects through its ability to bind to the VH sequence of the B cell receptor, and stimulates T cells by cross-linking MHC II molecules (Berk et al, 1986; Feijo et al, 1997). Protein A can also stimulate production of cytokines including g-interferon and TNF-a in immunocytes as well as hepatocyte growth factor in ®broblasts (Baroni et al, 1998; Sinha et al, 1999). The keratinocyte as a target for Protein A Though Protein A is consistently produced by essentially all strains of S. aureus, little is known about the effects of this protein on keratinocytes. Ezepchuk and colleagues have reported that Protein A is a potent stimulus for TNF-a production in both primary cultures of human keratinocytes and HaCaT cells (Ezepchuk et al, 1996). The mechanism by which Protein A exerts its effects on epithelial cells is not known; however, the ®nding that it has potent cytokine-producing effects on resting keratinocytes suggests that it is not acting via cross-linking MHC II molecules. In contrast to a-toxin, Protein A treatment is not associated with cytotoxic effects in epithelial cells (Ezepchuk et al, 1996). In addition, Protein A does not stimulate a calcium mobilization response, arachidonic acid release, or PAF biosynthesis in resting HaCaT cells (Travers, unpublished data). Given that Protein A can interact with MHC-II molecules in lymphocytes (Berk et al, 1986; Tiedemann and Fraser, 1996), however, it is possible that this protein could have similar effects on intracellular calcium levels as superantigens exert in MHC II-expressing keratinocytes. THERAPEUTIC IMPLICATIONS OF TOXIN± KERATINOCYTE INTERACTIONS The skin is a primary site of S. aureus infection and colonization. Indeed, toxins and other bacterial protein products of S. aureus are hypothesized to act as triggers or persistence factors in several in¯ammatory skin diseases. Thus, an understanding of the interactions of these proteins with target cells has important therapeutic implications. Though more information is available for other cell types, especially lymphocytes, signi®cant evidence is accumulating suggesting that the keratinocyte may be an important target for S. aureus-derived proteins such as superantigens, a-toxin, and Protein A. As shown in Fig 2, keratinocyte interactions with these proteins could result in the production of potent pro-in¯ammatory mediators such as TNF-a and PAF, as well as direct cytotoxic effects, compromising the barrier function of the skin. The capability of S. aureus proteins to stimulate cutaneous in¯ammation can explain why the judicial use of antibiotics can have therapeutic potential in the treatment strategies in some patients with in¯ammatory diseases. The authors wish to acknowledge the technical assistance of Dr. Keith Clay, Christopher Johnson, and Yong Pei. These studies were supported by grants the Indiana University Showalter Memorial Research Fund, and the National Institutes of Health grants K081993, HL62996, AR41256, HL37260, and 5M01 RR0051.
REFERENCES Aragane Y, Riemann H, Bhardwaj RS, et al: Interleukin-12 is expressed and released by human keratinocytes and epidermoid carcinoma cell lines. J Immunol 153:5366±5372, 1994 Baroni A, Perfetto B, Ruocco E, Rossano F: Lipotechoic acid and protein-A from
STAPHYLOCOCCAL TOXINS AND KERATINOCYTES
229
Staphylococcus aureus stimulate release of hepatocyte growth factor by human dermal ®broblasts. Arch Dermatol Res 290:211±214, 1998 Basham TY, Nickoloff BJ, Merigan TC, Morhenn VB: Recombinant gamma interferon induced HLA-DR expression on human keratinocytes. J Invest Dermatol 83:88±90, 1983 Berk GI, Lederman MM, Liebman ML, Ellner JJ: Staphylococcal protein A primed leukocytes enhance the autologous mixed lymphocyte reaction. Am J Med Sci 291:255±263, 1986 Boehncke WH, Dressel D, Zollner TM, et al: T-cell-receptor repertoire in chronic plaque psoriasis is restricted and lacks enrichment of superantigen-associated V beta regions. J Invest Dermatol 104:725±728, 1995 Boehncke WH, Dressel D, Zollner TM, Kaufmann R: Pulling the trigger on psoriasis. Nature 379:777, 1996 Camp RDR, Greaves MW: In¯ammatory mediators in the skin. Br Med Bull 43:401± 414, 1987 Deresiewicz RL: Staphylococcal toxic shock syndrome. In: Leung DYM, Huber B, Schlievert P, eds. Superantigens: Structure, Biology, and Relevance to Human Disease. New York: Marcel Dekker, 435±480, 1997 Etienne S, Bourdoulous S, Strosberg AD, Courad PO: MHC class II engagement in brain endothelial cells induces protein kinase A-dependent IL-6 secretion and phosphorylation of cyclic AMP response element-binding protein. J Immunol 163:3636±3641, 1999 Ezepchuk YV, Leung DYM, Middleton MH, et al: Staphylococcal toxins and protein A induce cytotoxicity and release of tumor necrosis factor from human keratinocytes. J Invest Dermatol 107:603±609, 1996 Feijo GC, Sabbaga J, Carneiro CR, Brigido MM: Variable region structure and staphylococcal protein A binding speci®city of a mouse monoclonal IgM antilaminin-receptor antibody. Immunology 91:479±485, 1997 Fink D, Contreras ML, Leikes PI, Lazarovici P: Staphylococcus aureus alpha-toxin activates phospholipases and induces a Ca2+ in¯ux in PC-12 cells. Cell Signal 1:387±393, 1989 Gaspari AA, Burns RP Jr, Kondo S, et al: Characterization of the altered reactivity of transgenic mice whose keratinocytes overexpress B7±1. Clin Immunol Immunopath 86:259±270, 1998 Greaves MW, Camp RDR: Prostaglandins leukotrienes, phospholipase, plateletactivating factor and cytokines: an integrated approach to in¯ammation of human skin. Arch Dermatol Res 280s:33±41, 1988 Grimminger F, Rose F, Sibelius U, et al: Human endothelial cell activation and mediator release in response to the bacterial exotoxins escherichia coli hemolysis and staphylococcal alpha-toxin. J Immunol 159:1909±1916, 1997 Haffner AC, Zepter K, Elmets CA: Major histocompatibility complex class I molecule serves as a ligand for presentation of superantigen staphylococcal enterotoxin B to T cells. Proc Natl Acad Sci (USA) 93:3037±3042, 1996 Han GW, Iwatsuki K, Inoue M, et al: Interleukin-15 is not a constituitive cytokine in the epidermis, but is inducible in culture or in¯ammatory conditions. Acta Dermat Ven 79:37±40, 1999 Henderson CA, Highet AS: Acute guttate psoriasis associated with Lance®eld Group C and Group G cutaneous streptococcal infections. Br J Dermatol 118:559±561, 1988 Heu¯er C, Topar G, Grasseger A, et al: Interleukin 7 is produced by murine and human keratinocytes. J Exp Med 178:1109±1114, 1993 Krutmann J, Grewe M: Involvement of cytokines, DNA damage, and reactive oxygen intermediates in ultraviolet radiation-induced modulation of intracellular adhesion molecule-1 expression. J Invest Dermatol 105:67s±70s, 1995 Krutmann J, Kock A, Schauer E, et al: Tumor necrosis factor and ultraviolet radiation are potent regulators of human keratinocyte ICAM-1 expression. J Invest Dermatol 95:127±131, 1990 Kupper TS: Immune and in¯ammatory processes in cutaneosu tissues: mechanisms and speculations. J Clin Invest 86:1783±1789, 1990 Leung DYM, Walsh P, Giorno R, Norris DA: A potential role for superantigens in the pathogenesis of psoriasis. J Invest Dermatol 100:225±228, 1993 Leung DYM, Travers JB, Giorno R, et al: Evidence for a streptococcal superantigendriven process in acute guttate psoriasis. J Clin Invest 96:2106±2112, 1995a Leung DYM, Gately M, Trumble A, et al: Bacterial superantigens induce T-cell expression of the skin-homing receptor, the cutaneous lymphocyte-associated antigen, via stimulation of interleukin 12 production. J Exp Med 181:747±753, 1995b Leung DYM, Hauk P, Strickland I, Travers JB, Norris DA: The role of superantigens in human diseases: therapeutic implications for the treatment of skin diseases. Br J Dermatol 139:17±29, 1998 Lewis HM, Baker BS, Bokth S, et al: Restricted T-cell receptor V beta gene usage in the skin of patients with guttate and chronic plaque psoriasis. Br J Dermatol 129:514±520, 1993 Michel L, Denizot Y, Thomas Y, et al: Production of PAF-acether by human epidermal cells. J Invest Dermatol 95:576±581, 1990 Moreau JM, Sloop GD, Engel LS, Hill JM, O'Callaghan RJ: Histopathological studies of staphylococcal alpha toxin: effects on rabbit corneas. Curr Eye Res 16:1221±1228, 1997 Nickoloff BJ: The cytokine network in psoriasis. Arch Dermatol 127:871±884, 1991 Nickoloff BJ, Naidu Y: Perturbation of epidermal barrier function correlates with initiation of cytokine cascade in human skin. J Am Acad Dermatol 30:535±546, 1994 Nickoloff BJ, Mitra RS, Green J et al: Accessory cell function of keratinocytes for superantigens. Dependence on lymphocyte function-associated antigen-1/ intracellular adhesion molecule±1 interaction. J Immunol 150:2148±2159, 1993
230
TRAVERS ET AL
Paquet P, Pierard GE: Interleukin-6 and the skin. Int Arch Allergy Imunol 109:308± 324, 1996 Pei Y, Barber LA, Murphy RC, et al: Activation of the epidermal platelet-activating factor receptor results in cytokine and cyclooxygenase-2 biosynthesis. J Immunol 161:1954±1961, 1998 Pei Y, Dy LC, Natarajan S, Travers JB: Activation of the epidermal plateletactivating factor receptor results in ICAM-1 expression. In Vitro Cell Dev Biol ± Animal 36:116±118, 2000 Pentland AP, Needleman P: Modulation of keratinocyte proliferation in vitro by endogenous prostaglandin synthesis. J Clin Invest 77:246±251, 1986 Pinckard RN, Woodard DS, Showell HJ, et al: Structural and (patho) physiological activity of platelet-activating factor. Clin Rev Allergy 13:329±339, 1994 Rosenberg EW, Noah PW, Zanolli MD, Skinner RB Jr, Bond MJ, Crutcher N: Use of rifampin with penicillin and erythromycin in the treatment of psoriasis. Preliminary report. J Am Acad Dermatol 14:761±764, 1986 Schmitt-Egenolf M, Boehncke WH, Christophers E, Stander M, Sterry W, Type I, Type II: psoriasis show a similar usage of T-cell receptor variable regions. J Invest Dermatol 97:1053±1056, 1991 Sibelius U, Grandel U, Buerke M, et al: Staphylococcal [alpha]-toxin provokes coronary vasoconstriction and loss in mycocardial contractility in perfused rat hearts: role of thromboxane generation. Circulation 101:78±87, 2000 Silverman GJB: cell superantigens: possible roles in immunode®ciency and autoimmunity. Sem Immunol 10:43±55, 1998 Sinha P, Ghosh AK, Das TSaG, Ray PK: Protein A of staphylococcus aureus evokes a TH1 type response in mice. Immunol Lett 67:157±165, 1999 Skov L, Olsen JV, Giorno R, et al: Application of staphylococcal enterotoxin B on normal and atopic skin induces upregulation of T cells via a superantigenmediated mechanism. J Allergy Clin Immunol 105:820±826, 2000 Soderquist B, Kallman J, Holmberg H, Vikefors T, Kihlstrohm E: Secretion of IL-6, IL-8, and G-CSF by human endothelial cells in vitro in response to staphylococcus aureus and staphylococcal exotoxins. APMIS 106:1157±1164, 1998 Song L, Hobaugh MR, Shustak C, et al: Structure of staphylococcal alpha-hemolysin, a heptameric transmembrane pore. Science 274:1859±1866, 1996 Strange P, Skov L, Baadsgaard O: Interferon-gamma-treated keratinocytes activate T cells in the presence of superantigens: involvement of major histocompatibility complex class II molecules. J Invest Dermatol 102:150±154, 1994 Suttorp N, Buerke M, Tannert-Otto S: Stimulation of PAF synthesis in pulmonary artery endothelial cells by staphylococcus aureus alpha-toxin. Thromb Res 67:243±252, 1992 Takeuchi M, Okura T, Mori T, et al: Intracellular production of interleukin-18 in
JID SYMPOSIUM PROCEEDINGS
human epithelial cell lines is enhanced by hyperosmotic stress in vitro. Cell Tiss Res 297:467±473, 1999 Tiedemann RE, Fraser JD: Cross-linking of MHC class II molecules by staphylococcal enterotoxin A is essential for antigen-presenting cell and T cell activation. J Immunol 157:3958±3966, 1996 Tokura Y, Yagi J, O'Malley M, et al: Superantigenic staphylococcal exotoxins induced T cell proliferation in presence of Langerhans cells or class II-bearing keratinocytes and stimulate keratinocytes to produce T-cell-activating cytokines. J Invest Dermatol 102:31±38, 1994 Tokura Y, Furakawa F, Wakita H, Yagi H, Ushijima T, Takigawa M: T-cell proliferation to superantigen-releasing Staphylococcus aureus by MHC Class II-bearing keratinocytes underprotection from bacterial cytolysin. J Invest Dermatol 108:488±494, 1997 Travers JB, Huff JC, Rola-Pleszcynski M, Gelfand EW, Morelli JG, Murphy RC: Identi®cation of functional paltelet-activating factor receptors on human keratinocytes. J Invest Dermatol 105:816±823, 1995 Travers JB, Harrison KA, Johnson CA, Clay KL, Morelli JG, Murphy RC: Plateletactivating factor biosynthesis induced by various stimuli in human HaCaT keratinocytes. J Invest Dermatol 107:88±94, 1996 Travers JB, Hamid QA, Norris DA, et al: Epidermal HLA-DR and the enhancement of cutaneous reactivity to superantigenic toxins in psoriasis. J Clin Invest 104:1181±1189, 1999 Valeva A, Walev I, Pinkernell M, et al: Transmembrane barrel of staphylococcal apha toxin forms in sensitive, but not resistant cells. Proc Nat Acad Sci (USA) 94:11607±11611, 1997 Wakem P, Burns RP Jr, Ramirez F, et al: Allergens and irritants transcriptionally upregulate CD80 gene expression in human keratinocytes. J Invest Dermatol 114:1085±1092, 2000 Wakita H, Tokura Y, Furukawa F, Takigawa M: Staphylococcal enterotoxin B upregulates expression of ICAM-1 molecules on IFN-gamma-treated keratinocytes and keratinocyte cell lines. J Invest Dermatol 105:536±542, 1995 Walev I, Martin E, Jonas D, et al: Staphylococcal alpha-toxin kills human keratinocytes by permeabilizing the plasma membrane for monovalent ions. Infect Immun 61:4972±4979, 1993 White J, Herman A, Pullen AM, et al: The VB-speci®c superantigen staphylococcal enterotoxin B. stimulation of mature T cells and clonal deletion in neonatal mice. Cell 56:27±35, 1989 Woodward DF, Hawley SB, Williams LS, et al: Studies on the ocular pathology of PGD2. Invest Opthalm Vis Sci 31:138±146, 1990 Wrone-Smith T, Nickoloff BJ: Dermal injection of immunocytes induces psoriasis. J Clin Invest 98:1878±1887, 1996
Departments of Dermatology, Pediatrics, Pharmacology and Toxicology, Indiana University School of Medicine, Indianapolis, Indiana, U.S.A.; *Department of Dermatology, University of Colorado School of Medicine, and Department of Veterans Affairs Hospital, Denver, Colorado, U.S.A.; ²Departments of Pediatrics, The National Jewish Medical and Research Center, and The University of Colorado School of Medicine, Denver, Colorado, U.S.A.
the production of powerful pro-in¯ammatory lipidand protein-derived cytokines in keratinocytes. Characterization of interactions between these proteins and the keratinocyte can provide a better understanding of how bacterial infection modulates in¯ammatory skin diseases, as well as provide the basis for improved therapies involving antibacterial agents. Key words: a-toxin/keratinocytes/platelet-activating factor/protein A/Staphylococcus aureus/superantigens. Journal of Investigative Dermatology Symposium Proceedings 6:225±230, 2001
Skin infections with Staphylococcus aureus are not only an important cause of morbidity and even mortality, but are thought to serve as initiation and/or persistance factors for numerous in¯ammatory skin diseases, including psoriasis and atopic dermatitis. One mechanism by which S. aureus can modulate the immune system is through the production of proteins such as superantigenic toxins, Protein A, as well through the cytolytic a-toxin. This review serves to discuss the biology of these three types of proteins, with emphasis on their ability to stimulate
S
eosinophil, make the keratinocyte a potentially important player in the cutaneous in¯ammatory response.
kin infections with Staphylococcus aureus are an important cause of morbidity and even mortality. Infections by these bacteria not only cause disease directly, but are thought to be important triggers in worsening in¯ammatory skin diseases ranging from atopic dermatitis to psoriasis (reviewed by Leung et al, 1998). One mechanism by which S. aureus can accomplish this is through the production of soluble proteins with potent immunomodulatory effects. The objective of this review is to focus on one aspect of cutaneous S. aureus bacterial infection; how interactions between keratinocytes and bacterial proteins such as superantigens, a-toxin, and Protein A can potentially affect the in¯ammatory process.
Keratinocyte-derived cytokines Keratinocytes have the capacity to synthesize and release a wide variety of immunomodulatory molecules (a partial list is found in Table I). Keratinocyte-derived pro-in¯ammatory and immunomodulatory molecules can be divided into those that are lipid versus those that are protein based. Lipid-derived in¯ammatory mediators produced by keratinocytes include derivatives of phospholipids [plateletactivating factor (1-O-alkyl-2-acetyl glycerophosphocholine; PAF) and lyso-phosphatidic acid (LPA)], arachidonic acid metabolites (eicosanoids), and others including ceramides. These mediators tend to be synthesized by keratinocytes in response to membrane damage. Of signi®cance, essentially all of these lipid-derived mediators are produced in response to an increased calcium mobilization signal, which can be seen following membrane damage. The major eicosanoid synthesized by human keratinocytes is PGE2, with lesser amounts of PGD2 and PGF2-a (Pentland and Needleman, 1986). Injection of nanogram amounts of PGE2 causes erythema in human skin (Camp and Greaves, 1987). Larger amounts induce swelling through vasodilatory effects on blood vessels, with only minimal leukocyte chemoattractive properties. PGD2 also has erythrogenic effects, as well as granulocyte chemoattractant properties (Woodward et al, 1990). Though keratinocytes synthesize hydroxyeicosateranoic acids (HETE), they do not synthesize signi®cant levels of leukotrienes (Greaves and Camp, 1988). Produced by the subsequent actions of phospholipase A2 and PAF acetyltransferase, PAF is synthesized in response to the same stimuli that induce eicosanoid production (Pinckard et al, 1994). Intradermal injection of PAF results in a painful wheal and ¯are reaction within minutes. Human keratinocytes both synthesize
THE KERATINOCYTE AS AN ACTIVE PARTICIPANT IN CUTANEOUS INFLAMMATION Keratinocytes are by far the most abundant cell type in the epidermis, and our concept of the keratinocyte's role in in¯ammatory processes has changed greatly from a passive target to an active participant. The keratinocyte has the ability to produce and release numerous lipid and protein cytokines (reviewed by Greaves and Camp, 1988; Kupper, 1990). Keratinocytes also can express adhesion molecules that allow enhanced leukocyte interactions. These properties, as well as the relative long life of a keratinocyte compared with other in¯ammatory cells such as the neutrophil or Manuscript received February 13, 2001; accepted for publication February 19, 2001. Reprint requests to: Dr. Jeffrey B. Travers, The H.B Wells Center for Pediatric Research, Riley Hospital for Children Room 2659, 635 Barnhill Drive, Indianapolis, Indiana 46202. Email: [email protected] Abbreviations: CLA, cutaneous lymphocyte-associated antigen; MHC II, major histocompatibility class II protein; PAF, platelet-activating factor; PG, prostaglandin; SAg, superantigen; TCR, T cell receptor. 1087-0024/01/$15.00
´ Copyright # 2001 by The Society for Investigative Dermatology, Inc. 225
226
TRAVERS ET AL
JID SYMPOSIUM PROCEEDINGS
Table I. Immunomodulatory molecules produced by keratinocytes Lipid-derived
Protein-derived
Prostaglandins E2, D2, F2-a 12-Hydroxyeicosatetraenoic acid Platelet-activating factor Lyso-phosphatidic acid Ceramides
Interleukins 1, 6, 8, 12, 15, 18 Tumor necrosis factor-a Transforming growth factor-a Interferon-b Granulocyte and Granulocyte macrophage colony stimulatory factors
PAF as well as express functional PAF receptors (PAF-R) (Michel et al, 1990; Travers et al, 1995). Activation of the epidermal PAF-R results in the biosynthesis of numerous mediators, including eicosanoids, TNF-a, IL-6, IL-8, and PAF itself (Pei et al, 1998). Amongst the protein-derived cytokines produced by keratinocytes, TNF-a and IL-1 have been termed primary cytokines due to their ability to induce in¯ammatory reactions in part through the ability to induce a host of other lipid and protein mediators (Kupper, 1990). TNF-a and IL-1 both exert effects on receptors expressed on keratinocytes that are linked to the nuclear factor kB pathway, with resultant production of cascades of other protein and lipid cytokines. IL-6 is a pleiotropic cytokine with effects on both B and T cells (Paquet and Pierard, 1996). IL-7 has the ability to stimulate the growth of both progenitors and mature B and T cells (Heu¯er et al, 1993). IL-8 is a potent neutrophil chemoattractant also produced by keratinocytes. IL-12 is a potent TH1 cytokine that promotes activation of macrophages that can be synthesized by keratinocytes in response to varied stimuli including phorbol esters and IFN-g (Aragane et al, 1994). In addition, bacterial superantigens appear to induce production of skin-homing (cutaneous lymphocyte-associated antigen-positive; CLA) T cells through an IL-12-dependent process (Leung et al, 1995). Though resting keratinocytes do not synthesize the T cell activating cytokine IL-15, its production can be induced by IFN-g, a process inhibited by dexamethasone (Han et al, 1999). IL-18, a novel cytokine that strongly induces IFN-g production in T cells, has also been shown to be synthesized by human and murine keratinocytes. Hyperosmolar stress induced by mannitol or NaCl is a potent stimulus for IL-18 production in both epithelial and endothelial cells, suggesting that this cytokine could be important in damage-mediated in¯ammatory responses (Takeuchi et al, 1999). Thus, a keratinocyte that sustains membrane damage has the potential to synthesize and release numerous lipid and protein cytokines that could modulate the in¯ammatory response. Keratinocyte adhesion molecules In addition to the ability of keratinocytes to modulate in¯ammation through soluble mediators, these cells can also express membrane-associated molecules that can serve to bind immune cells as well as to stimulate signal transduction pathways leading to further production of mediators. Through its ability to bind the B-2 integrin leukocyte function-associated antigen found on most leukocytes, the adhesion molecule intracellular adhesion molecule1 (ICAM-1; CD54) is an important mediator of cellular immune reactions (reviewed by Krutmann and Grewe, 1995). Though not found on resting epidermal cells, ICAM-1 surface expression can be induced by cytokines including TNF-a (Krutmann et al, 1990) and PAF (Pei et al, 2000), as well as by physical methods including tapestripping (Nickoloff and Naidu, 1994). Similar to ICAM-1, resting keratinocytes do not express the costimulatory molecule B7-1 (CD80), yet epidermal expression can be seen during allergic and contact dermatitis (Wakem et al, 2000). The signi®cance of epidermal B7-1 expression is not clear, though this signaling molecule usually associated with antigen-presenting cells prevents
the induction of T cell anergy and induces IL-2 production during antigen presentation. Of note, transgenic mice overexpressing B7-1 on keratinocytes exhibit exaggerated and persistent contact hypersensitivity reactions to a variety of haptens (Gaspari et al, 1998). In addition to adhesion and costimulatory molecules, keratinocytes can also express MHC-II molecules (i.e., HLA-DR) in response to various pro-in¯ammatory and immune stimuli (Basham et al, 1983). MHC-II molecules can signal through intracellular calcium, as well as through kinases including protein kinases A and C (PKA and PKC) (Wakita et al, 1995; Etienne et al, 1999). Keratinocyte MHC-II molecules can also trigger the production of cytokines such as TNF-a (Tokura et al, 1994). BACTERIAL SUPERANTIGENS AND THE KERATINOCYTE The biology of superantigens The term ``superantigen'' was ®rst coined by White et al in 1989 and refers to a group of microbial and viral proteins that differ in several important respects from conventional peptide antigens (White et al, 1989). First, unlike conventional protein antigens, which are taken up and processed by antigen-presenting cells, superantigens exert their effects as globular intact proteins. Similar to peptide antigens, superantigens are presented by class II MHC molecules; however, they do not interact with the MHC peptide b antigen binding groove, but instead bind to conserved amino acid residues that are on the outer walls of the peptide antigen-binding groove. Thus, whereas recognition of conventional peptide antigens by the T cell receptor is restricted by MHC alleles, recognition of superantigens is not generally MHC restricted. Second, superantigens primarily recognize and bind to the variable region of the T cell receptor b chain (VB). This is in contrast to nominal peptide antigens, which require recognition by all ®ve variable elements (i.e., VB, DB, JB, Va, Ja) of the T cell receptor. Therefore, the responding frequency of a superantigen is several orders of magnitude greater than a conventional peptide antigen. The unique ability of a superantigen to bind directly to (and signal through) MHC class II molecules, and cross-link with a large percentage of T cells expressing relevant T cell receptor VB chains, provides an explanation for the potent immune stimulation seen with these molecules (Tiedemann and Fraser, 1996). The prototypical disease due to bacterial superantigens is toxic shock syndrome caused by staphylococcal exotoxins or streptococcal enterotoxins (Deresiewicz, 1997). Superantigens and in¯ammatory skin diseases As outlined above, bacterial superantigens possess several properties that can contribute to disease pathogenesis. Superantigens can bind to and activate MHC class II molecules, whether constitutively expressed on professional antigen presenting cells such as monocytes, Langerhans, or dendritic cells, or on nonprofessional antigenpresenting cells such as the keratinocyte following cytokine stimulation. Signi®cant pathologic effects can be seen due to either local or massive systemic release of protein cytokines as well as other pro-in¯ammatory mediators. In addition, proin¯ammatory effects would be further ampli®ed by subsequent activation of T cells. Of signi®cance for the ability of these proteins to induce cutaneous in¯ammation, recent evidence suggests that superantigenic stimulation of T cells promotes skin homing by upregulation of the skin homing receptor CLA, a modi®ed selectin that promotes localization of T cells to the skin. Leung et al reported that stimulation of peripheral blood mononuclear cells with bacterial superantigens resulted in a signi®cant increase of CLA+ T cells, but not T cells expressing other homing molecules, in comparison with treatment with other agents including phytohaemogluttinin or anti-CD3 (Leung et al, 1995). Bacterial toxins were also found to induce IL-12 production in this model system. Of note, induction of toxin-induced CLA expression was blocked by anti-IL-12, and the addition of IL-12 to phytohae-
VOL. 6, NO. 3 DECEMBER 2001
mogluttinin-stimulated T cells induced CLA, but not expression of the mucosal homing receptor a e b 7-integrin (Leung et al, 1995b). These data suggest that bacterial toxins induce the expansion of skin-homing CLA+ T cells in an IL-12-dependent manner, and thus may contribute to the development of skin rashes in superantigen-mediated diseases. Psoriasis is a chronic in¯ammatory skin disorder affecting up to 2% of the general population. The characteristic lesion of psoriasis is a well-demarcated erythematous papule or plaque containing hyperproliferating keratinocytes, as well as in®ltrating neutrophils, monocytes, and T lymphocytes (Nickoloff, 1991). Though considered an autoimmune disease, considerable evidence suggests an important role for bacteria in the pathogenesis of psoriasis. Colonization and/or infection with Staphylococcus and Streptococcus have been reported to exacerbate psoriasis (Henderson and Highet, 1988; Leung et al, 1993). Superantigenic toxins produced by these bacteria have been implicated in the initiation and/or propagation of psoriasis (reviewed by Leung et al, 1998). Indeed, a group of psoriatics whose skin disease was worsening were found to be secondarily infected with superantigen-expressing S. aureus, and skin biopsies revealed lesional T cell receptor VB skewing consistent with the pattern of the isolated superantigen (Leung et al, 1993). Injection of superantigen-stimulated immunocytes into normal skin from subjects with psoriasis xenografted onto an immunode®cient mouse has also been shown to induce clinical, histologic, and immunologic changes consistent with psoriasis (Boehncke et al, 1996; Wrone-Smith and Nickoloff, 1996). The most convincing clinical and experimental association between bacterial superantigens and psoriasis, however, is in patients with acute eruptive (guttate) psoriasis (Lewis et al, 1993; Leung et al, 1995). Of note, Leung et al demonstrated up to 50% T cell receptor VB2+ T cells in perilesional skin of subjects with acute eruptive psoriasis induced by a Streptococcal pyrogenic exotoxin C (SPEC) expressing Streptococcus (Leung et al, 1995). Altogether, these studies have implicated bacterial superantigens in both the initiation as well as the worsening of psoriasis. The keratinocyte as a target for superantigens Due to the lack of MHC II molecules, the resting keratinocyte is not thought to be a direct target for superantigens (Ezepchuk et al, 1996); however, treatment of resting keratinocytes with cytokines such as g-interferon has been shown to upregulate MHC II molecules (Basham et al, 1983). Superantigens have been shown to directly bind to and stimulate g-interferon-treated primary cultures of human and murine keratinocytes as well as epidermal carcinoma cell lines (Nickoloff et al, 1993; Strange et al, 1994; Tokura et al, 1994). Early biochemical events seen as a consequence of MHC II and superantigen interactions in keratinocytes include a transient intracellular calcium mobilization (Wakita et al, 1995). Superantigens have also been shown to induce the production of TNF-a in keratinocytes in vitro (Tokura et al, 1994; Ezepchuk et al, 1996). In addition to their well-characterized effects on MHC II molecules, superantigens have also been reported to interact with MHC class I molecules on squamous cell carcinoma cell lines (Haffner et al, 1996). Topical application of superantigens to atopic skin also induces acute eczematous changes and T cell recptor VB skewing (Skov et al, 2000) Hyperreactivity of psoriatic keratinocytes to topical superantigens Signi®cant evidence based on ®nding peripheral blood or skin T cells with TCR VB skewing and the isolation of a superantigen with a relevant TCR VB pro®le has linked superantigens to either initiatory events in acute guttate psoriasis (Lewis et al, 1993; Leung et al, 1995), or to worsening of psoriasis skin disease (Leung et al, 1993). Recent studies by Travers et al have examined the effects of topical application of nanogram amounts of the staphylococcal superantigens ZEB and TSST-1, and the streptococcal superantigens SPEA and SPEC to tape-abraded skin of psoriatic patients for 48 h (Travers et al, 1999). In¯ammatory responses to these puri®ed bacterial proteins were seen in both normal subjects as well as in psoriatics; however,
STAPHYLOCOCCAL TOXINS AND KERATINOCYTES
227
patients with psoriasis exhibited signi®cantly increased in¯ammatory skin responses in comparison with controls, or subjects with atopic dermatitis or lichen planus. Reactivity of psoriatic subjects to topical superantigens correlated with the amount of skin disease, and was greatest in subjects who exhibited the isomorphic (Koebner) phenomenon. Consistent with the clinical hyperactive response, in situ hybridization studies of skin biopsies obtained 6 and 24 h following patch testing with superantigens demonstrated increased epidermal TNF-a mRNA in psoriatics than controls. Though signi®cant numbers of mononuclear cells were seen in superantigen-induced skin reactions, surprisingly, skin biopsies from the lesions did not reveal increased numbers of T cells expressing the predicted T cell receptor VB type for the superantigens. This is in contrast to the T cell receptor VB skewing found after application of larger doses of superantigens on the intact skin of subjects with atopic dermatitis (Skov et al, 2000). In addition, immunohistochemical studies revealed increased epidermal MHC-II expression in reactions in psoriatics compared with control skin. Given that superantigens have the ability to induce TNF-a production with MHC-II expressing keratinocytes in vitro (Tokura et al, 1994; Ezepchuk et al, 1996), one explanation for the increased clinical reactivity and increased epidermal TNF-a production found in psoriatic subjects, would be that the superantigens were stimulating cytokine production from MHC-II positive keratinocytes, with T cell accumulation being in response to these keratinocyte-derived cytokines. This would also explain the lack of T cell receptor VB skewing as well. Consistent with this notion that epidermal MHC-II±superantigen interactions were responsible for the enhanced skin reactions in psoriatic subjects, a mutant TSST-1 toxin that does not bind to MHC-II (TSST1G31S/S32P) did not induce skin in¯ammation, even at 10-fold higher concentrations than native superantigens (Travers et al, 1999). The lack of skin reactions to the mutant TSST-1 also suggest that the exotoxin-induced skin responses were not due to a delayed type hypersensitivity reaction to a conventional antigen. The ability of superantigens to stimulate the production of potent pro-in¯ammatory and cytotoxic cytokines such as TNF-a in MHC-II positive keratinocytes provides an alternative mechanism by which colonization or actual infection with S. aureus can induce skin in¯ammation. This pathway could be relevant in psoriasis, as psoriatic lesions are often colonized with Staphylococcus and Streptococcus (Leung et al, 1993; Ezepchuk et al, 1996), and scratching of the skin would be expected to mimic tape-stripping, by both compromising the barrier function as well as activating the epidermis (Nickoloff and Naidu, 1994). This alternative mechanism may explain why some psoriatic patients report improvement of their disease activity following treatment with oral antibiotics (Rosenberg et al, 1986). It may also account for several reports that fail to demonstrate selective expansion of T cell receptor VB in the skin of patients with plaque psoriasis (SchmittEgenolf et al, 1991; Boehncke et al, 1995). Thus, epicutaneous application of superantigens on psoriatic skin may result in an acute in¯ammatory, keratinocyte-mediated response rather than superantigen-mediated T cell activation. STAPHYLOCOCCAL a-TOXIN, PROTEIN A, AND THE KERATINOCYTE In addition to toxins that can have profound immunomodulatory effects through their abilities to act as superantigens, S. aureus can produce other proteins that can act as virulence factors and affect host responses. The production of transmembrane pores by staphylococcal a-toxin can induce a potent cytotoxic response that also results in cytokine production. Though not associated with a direct cytotoxic response, staphylococcal Protein A can also induce cytokine production in keratinocytes. In contrast to superantigens, both a-toxin and Protein A can interact with resting (i.e., MHC II-negative) keratinocytes.
228
TRAVERS ET AL
The biology of a-toxin Staphylococcal a-toxin is released by bacteria as a 293 residue single chain polypeptide. Upon interaction with a target cell, a-toxin forms a heptamer, which results in the formation of a transmembrane pore (Song et al, 1996). Recent studies using biotinylated a-toxin in susceptible versus resistant cells suggest that the signaling and cytotoxic effects are related to the ability of the a-toxin heptamer to form a transmembrane channel (Valeva et al, 1997). Formation of transmembrane pores results in membrane damage as well as an in¯ux of extracellular calcium, a potent damage-related stimulus for many cells. Similar to calcium ionophores such as A23187, small amounts of a-toxin result in signaling and potential repairable damage, whereas larger amounts result in cell death. In endothelial cells as well as in PC-12 cells, a-toxin activates PLA2 via a calcium-dependent process (Fink et al, 1989). In endothelial cells, a-toxin stimulates arachidonic acid release, as well as PAF biosynthesis (Suttorp et al, 1992; Grimminger et al, 1997). Staphylococcal a-toxin has also been reported to enhance IL-6 and G-CSF production in IL-1b-treated endothelial cell cultures (Soderquist et al, 1998). Exposure of isolated perfused rat hearts to a-toxin was shown to induce coronary vasoconstriction and loss of myocardial contractility (Sibelius et al, 2000). Signi®cant levels of thromboxane A2 were found in the perfusate following a-toxin treatment. Of note, indomethacin, acetylsalicylic acid, and a speci®c thromboxane receptor antagonist inhibited a-toxin effects, demonstrating involvement of endogenous thromboxane A2 in this model system. Thus, a-toxin can induce organ damage not only through direct cytolytic effects, but also through the ability to trigger cellular production of pro-in¯ammatory and cytotoxic soluble mediators. The keratinocyte as a target for a-toxin Staphylococcus aureus is not only a common agent of cutaneous infections, but can also be isolated from many patients with in¯ammatory skin diseases such as atopic dermatitis and psoriasis. A recent study by Ezepchuk and colleagues demonstrated that nine of 19 strains of S. aureus obtained from clinical lesions of atopic dermatitis and psoriasis contained atoxin (Ezepchuk et al, 1996). Ezepchuk et al also found that treatment of primary cultures of human keratinocytes or the keratinocyte-derived cell line HaCaT with a-toxin resulted in signi®cant TNF-a production. Staphyloccoccal a-toxin also induced cytotoxic effects in keratinocytes that were more compatible with necrosis than apoptosis (Walev et al, 1993; Ezepchuk et al, 1996). Injection of rabbit corneas with a-toxin has been shown to result in signi®cant tissue edema with death of corneal epithelial cells by both necrosis and apoptosis (Moreau et al, 1997). Though a-toxin is thought to have considerable deleterious effects, it has been suggested that a-toxin-induced keratinocyte damage can protect against superantigen-mediated effects by inhibiting the ability of the keratinocyte to activate T cells (Tokura et al, 1997). Allapatt et al recently examined the ability of a-toxin to stimulate lipid mediator production in human carcinoma cell lines.1 Using select ion monitoring gas chromatography mass spectrometry, signi®cant production of PAF as well as arachidonic acid release was found in a-toxin-treated HaCaT cells. The amount of PAF produced in response to 10 mg per ml a-toxin was about 25% of that seen in response to a potent ionophore such as A23187 (Travers et al, 1995). Because activation of the epidermal PAF-R can induce arachidonic acid release (Pei et al, 1998), the potential involvement of endogenous PAF in a-toxin-induced arachidonic acid release was tested using a model system created by retroviralmediated transduction of the PAF-R-negative epithelial cell line KB with the human PAF-R. As shown in Fig 1, a-toxin stimulated arachidonic acid release in both control (PAF-Rnegative) KBM as well as PAF-R-positive KBP cells; however, the 1 Allapatt C, Leung DYM, Johnson C, Clay K, Cosgrove J, Travers JB: Acute keratinocyte damage results in the biosynthesis of the lipid mediator platelet-activating factor. J Invest Dermatol 112:543, 1999 (abstr)
JID SYMPOSIUM PROCEEDINGS
Figure 1. Staphylococcal a-toxin stimulates increased arachidonic acid release in PAF-R expressing versus control KB cells. KB cells transduced with the PAF-R (KBP) or control retrovirus (KBM) were stimulated with 10 mg per ml puri®ed a-toxin or vehicle for 30 min, and arachidonic acid released into the supernatant was quantitated by gas chromatography mass spectrometry using deuterated arachidonic acid as internal standard as previously described (Pei et al, 1998). The data shown are mean 6 standard deviation of arachidonic acid released from duplicate samples from a representative experiment of four conducted.
Figure 2. Interactions between Staphylococcal proteins and the keratinocyte. The keratinocyte has the potential to be a target for S. aureus proteins. Enterotoxins that act as superantigens (SAg) can signal through MHC-II molecules found on activated keratinocytes, resulting in a calcium mobilization response as well as the production of TNF-a. Protein A does not appear to induce a calcium mobilization response, but is a potent inducer of keratinocyte TNF-a production through an unknown mechanism. Through its ability to form a heptameric transmembrane pore, a-toxin can trigger a calcium mobilization response as well as TNF-a production.
amounts of arachidonic acid released by a-toxin-treated KBP cells were signi®cantly greater than with KBM cells (Fig 1). Similarly, pretreatment of KBP cells with 10 mM of the PAF-R antagonist WEB 2086 partially inhibited a-toxin-induced arachidonic acid release in KBP but not KBM cells (data not shown). These data are compatible with the notion that a-toxin can generate both arachidonic acid release and PAF biosynthesis through calciummediated activation of PLA2, and that endogenous PAF generated can feed forward to induce further mediator production. Altogether, through its direct effects on keratinocyte membranes, a-toxin can induce the synthesis of both lipid and protein cytokines, as well as cause direct cytotoxic effects.
VOL. 6, NO. 3 DECEMBER 2001
The biology of Protein A The unique protein derived from the cell wall structure of S. aureus termed Protein A was discovered over 40 y ago. A characteristic property of this protein has been its ability to bind the Fc structure of IgG, making it a valuable laboratory reagent for the isolation and puri®cation of antibody molecules. Over time other biologic properties of this molecule have been elucidated. Protein A appears to have an unrivaled diversity of biologic activities ranging from immunomodulatory and antitumor to antifungal and antiparasitic effects (reviewed by Silverman, 1998). Protein A has mitogenic effects through its ability to bind to the VH sequence of the B cell receptor, and stimulates T cells by cross-linking MHC II molecules (Berk et al, 1986; Feijo et al, 1997). Protein A can also stimulate production of cytokines including g-interferon and TNF-a in immunocytes as well as hepatocyte growth factor in ®broblasts (Baroni et al, 1998; Sinha et al, 1999). The keratinocyte as a target for Protein A Though Protein A is consistently produced by essentially all strains of S. aureus, little is known about the effects of this protein on keratinocytes. Ezepchuk and colleagues have reported that Protein A is a potent stimulus for TNF-a production in both primary cultures of human keratinocytes and HaCaT cells (Ezepchuk et al, 1996). The mechanism by which Protein A exerts its effects on epithelial cells is not known; however, the ®nding that it has potent cytokine-producing effects on resting keratinocytes suggests that it is not acting via cross-linking MHC II molecules. In contrast to a-toxin, Protein A treatment is not associated with cytotoxic effects in epithelial cells (Ezepchuk et al, 1996). In addition, Protein A does not stimulate a calcium mobilization response, arachidonic acid release, or PAF biosynthesis in resting HaCaT cells (Travers, unpublished data). Given that Protein A can interact with MHC-II molecules in lymphocytes (Berk et al, 1986; Tiedemann and Fraser, 1996), however, it is possible that this protein could have similar effects on intracellular calcium levels as superantigens exert in MHC II-expressing keratinocytes. THERAPEUTIC IMPLICATIONS OF TOXIN± KERATINOCYTE INTERACTIONS The skin is a primary site of S. aureus infection and colonization. Indeed, toxins and other bacterial protein products of S. aureus are hypothesized to act as triggers or persistence factors in several in¯ammatory skin diseases. Thus, an understanding of the interactions of these proteins with target cells has important therapeutic implications. Though more information is available for other cell types, especially lymphocytes, signi®cant evidence is accumulating suggesting that the keratinocyte may be an important target for S. aureus-derived proteins such as superantigens, a-toxin, and Protein A. As shown in Fig 2, keratinocyte interactions with these proteins could result in the production of potent pro-in¯ammatory mediators such as TNF-a and PAF, as well as direct cytotoxic effects, compromising the barrier function of the skin. The capability of S. aureus proteins to stimulate cutaneous in¯ammation can explain why the judicial use of antibiotics can have therapeutic potential in the treatment strategies in some patients with in¯ammatory diseases. The authors wish to acknowledge the technical assistance of Dr. Keith Clay, Christopher Johnson, and Yong Pei. These studies were supported by grants the Indiana University Showalter Memorial Research Fund, and the National Institutes of Health grants K081993, HL62996, AR41256, HL37260, and 5M01 RR0051.
REFERENCES Aragane Y, Riemann H, Bhardwaj RS, et al: Interleukin-12 is expressed and released by human keratinocytes and epidermoid carcinoma cell lines. J Immunol 153:5366±5372, 1994 Baroni A, Perfetto B, Ruocco E, Rossano F: Lipotechoic acid and protein-A from
STAPHYLOCOCCAL TOXINS AND KERATINOCYTES
229
Staphylococcus aureus stimulate release of hepatocyte growth factor by human dermal ®broblasts. Arch Dermatol Res 290:211±214, 1998 Basham TY, Nickoloff BJ, Merigan TC, Morhenn VB: Recombinant gamma interferon induced HLA-DR expression on human keratinocytes. J Invest Dermatol 83:88±90, 1983 Berk GI, Lederman MM, Liebman ML, Ellner JJ: Staphylococcal protein A primed leukocytes enhance the autologous mixed lymphocyte reaction. Am J Med Sci 291:255±263, 1986 Boehncke WH, Dressel D, Zollner TM, et al: T-cell-receptor repertoire in chronic plaque psoriasis is restricted and lacks enrichment of superantigen-associated V beta regions. J Invest Dermatol 104:725±728, 1995 Boehncke WH, Dressel D, Zollner TM, Kaufmann R: Pulling the trigger on psoriasis. Nature 379:777, 1996 Camp RDR, Greaves MW: In¯ammatory mediators in the skin. Br Med Bull 43:401± 414, 1987 Deresiewicz RL: Staphylococcal toxic shock syndrome. In: Leung DYM, Huber B, Schlievert P, eds. Superantigens: Structure, Biology, and Relevance to Human Disease. New York: Marcel Dekker, 435±480, 1997 Etienne S, Bourdoulous S, Strosberg AD, Courad PO: MHC class II engagement in brain endothelial cells induces protein kinase A-dependent IL-6 secretion and phosphorylation of cyclic AMP response element-binding protein. J Immunol 163:3636±3641, 1999 Ezepchuk YV, Leung DYM, Middleton MH, et al: Staphylococcal toxins and protein A induce cytotoxicity and release of tumor necrosis factor from human keratinocytes. J Invest Dermatol 107:603±609, 1996 Feijo GC, Sabbaga J, Carneiro CR, Brigido MM: Variable region structure and staphylococcal protein A binding speci®city of a mouse monoclonal IgM antilaminin-receptor antibody. Immunology 91:479±485, 1997 Fink D, Contreras ML, Leikes PI, Lazarovici P: Staphylococcus aureus alpha-toxin activates phospholipases and induces a Ca2+ in¯ux in PC-12 cells. Cell Signal 1:387±393, 1989 Gaspari AA, Burns RP Jr, Kondo S, et al: Characterization of the altered reactivity of transgenic mice whose keratinocytes overexpress B7±1. Clin Immunol Immunopath 86:259±270, 1998 Greaves MW, Camp RDR: Prostaglandins leukotrienes, phospholipase, plateletactivating factor and cytokines: an integrated approach to in¯ammation of human skin. Arch Dermatol Res 280s:33±41, 1988 Grimminger F, Rose F, Sibelius U, et al: Human endothelial cell activation and mediator release in response to the bacterial exotoxins escherichia coli hemolysis and staphylococcal alpha-toxin. J Immunol 159:1909±1916, 1997 Haffner AC, Zepter K, Elmets CA: Major histocompatibility complex class I molecule serves as a ligand for presentation of superantigen staphylococcal enterotoxin B to T cells. Proc Natl Acad Sci (USA) 93:3037±3042, 1996 Han GW, Iwatsuki K, Inoue M, et al: Interleukin-15 is not a constituitive cytokine in the epidermis, but is inducible in culture or in¯ammatory conditions. Acta Dermat Ven 79:37±40, 1999 Henderson CA, Highet AS: Acute guttate psoriasis associated with Lance®eld Group C and Group G cutaneous streptococcal infections. Br J Dermatol 118:559±561, 1988 Heu¯er C, Topar G, Grasseger A, et al: Interleukin 7 is produced by murine and human keratinocytes. J Exp Med 178:1109±1114, 1993 Krutmann J, Grewe M: Involvement of cytokines, DNA damage, and reactive oxygen intermediates in ultraviolet radiation-induced modulation of intracellular adhesion molecule-1 expression. J Invest Dermatol 105:67s±70s, 1995 Krutmann J, Kock A, Schauer E, et al: Tumor necrosis factor and ultraviolet radiation are potent regulators of human keratinocyte ICAM-1 expression. J Invest Dermatol 95:127±131, 1990 Kupper TS: Immune and in¯ammatory processes in cutaneosu tissues: mechanisms and speculations. J Clin Invest 86:1783±1789, 1990 Leung DYM, Walsh P, Giorno R, Norris DA: A potential role for superantigens in the pathogenesis of psoriasis. J Invest Dermatol 100:225±228, 1993 Leung DYM, Travers JB, Giorno R, et al: Evidence for a streptococcal superantigendriven process in acute guttate psoriasis. J Clin Invest 96:2106±2112, 1995a Leung DYM, Gately M, Trumble A, et al: Bacterial superantigens induce T-cell expression of the skin-homing receptor, the cutaneous lymphocyte-associated antigen, via stimulation of interleukin 12 production. J Exp Med 181:747±753, 1995b Leung DYM, Hauk P, Strickland I, Travers JB, Norris DA: The role of superantigens in human diseases: therapeutic implications for the treatment of skin diseases. Br J Dermatol 139:17±29, 1998 Lewis HM, Baker BS, Bokth S, et al: Restricted T-cell receptor V beta gene usage in the skin of patients with guttate and chronic plaque psoriasis. Br J Dermatol 129:514±520, 1993 Michel L, Denizot Y, Thomas Y, et al: Production of PAF-acether by human epidermal cells. J Invest Dermatol 95:576±581, 1990 Moreau JM, Sloop GD, Engel LS, Hill JM, O'Callaghan RJ: Histopathological studies of staphylococcal alpha toxin: effects on rabbit corneas. Curr Eye Res 16:1221±1228, 1997 Nickoloff BJ: The cytokine network in psoriasis. Arch Dermatol 127:871±884, 1991 Nickoloff BJ, Naidu Y: Perturbation of epidermal barrier function correlates with initiation of cytokine cascade in human skin. J Am Acad Dermatol 30:535±546, 1994 Nickoloff BJ, Mitra RS, Green J et al: Accessory cell function of keratinocytes for superantigens. Dependence on lymphocyte function-associated antigen-1/ intracellular adhesion molecule±1 interaction. J Immunol 150:2148±2159, 1993
230
TRAVERS ET AL
Paquet P, Pierard GE: Interleukin-6 and the skin. Int Arch Allergy Imunol 109:308± 324, 1996 Pei Y, Barber LA, Murphy RC, et al: Activation of the epidermal platelet-activating factor receptor results in cytokine and cyclooxygenase-2 biosynthesis. J Immunol 161:1954±1961, 1998 Pei Y, Dy LC, Natarajan S, Travers JB: Activation of the epidermal plateletactivating factor receptor results in ICAM-1 expression. In Vitro Cell Dev Biol ± Animal 36:116±118, 2000 Pentland AP, Needleman P: Modulation of keratinocyte proliferation in vitro by endogenous prostaglandin synthesis. J Clin Invest 77:246±251, 1986 Pinckard RN, Woodard DS, Showell HJ, et al: Structural and (patho) physiological activity of platelet-activating factor. Clin Rev Allergy 13:329±339, 1994 Rosenberg EW, Noah PW, Zanolli MD, Skinner RB Jr, Bond MJ, Crutcher N: Use of rifampin with penicillin and erythromycin in the treatment of psoriasis. Preliminary report. J Am Acad Dermatol 14:761±764, 1986 Schmitt-Egenolf M, Boehncke WH, Christophers E, Stander M, Sterry W, Type I, Type II: psoriasis show a similar usage of T-cell receptor variable regions. J Invest Dermatol 97:1053±1056, 1991 Sibelius U, Grandel U, Buerke M, et al: Staphylococcal [alpha]-toxin provokes coronary vasoconstriction and loss in mycocardial contractility in perfused rat hearts: role of thromboxane generation. Circulation 101:78±87, 2000 Silverman GJB: cell superantigens: possible roles in immunode®ciency and autoimmunity. Sem Immunol 10:43±55, 1998 Sinha P, Ghosh AK, Das TSaG, Ray PK: Protein A of staphylococcus aureus evokes a TH1 type response in mice. Immunol Lett 67:157±165, 1999 Skov L, Olsen JV, Giorno R, et al: Application of staphylococcal enterotoxin B on normal and atopic skin induces upregulation of T cells via a superantigenmediated mechanism. J Allergy Clin Immunol 105:820±826, 2000 Soderquist B, Kallman J, Holmberg H, Vikefors T, Kihlstrohm E: Secretion of IL-6, IL-8, and G-CSF by human endothelial cells in vitro in response to staphylococcus aureus and staphylococcal exotoxins. APMIS 106:1157±1164, 1998 Song L, Hobaugh MR, Shustak C, et al: Structure of staphylococcal alpha-hemolysin, a heptameric transmembrane pore. Science 274:1859±1866, 1996 Strange P, Skov L, Baadsgaard O: Interferon-gamma-treated keratinocytes activate T cells in the presence of superantigens: involvement of major histocompatibility complex class II molecules. J Invest Dermatol 102:150±154, 1994 Suttorp N, Buerke M, Tannert-Otto S: Stimulation of PAF synthesis in pulmonary artery endothelial cells by staphylococcus aureus alpha-toxin. Thromb Res 67:243±252, 1992 Takeuchi M, Okura T, Mori T, et al: Intracellular production of interleukin-18 in
JID SYMPOSIUM PROCEEDINGS
human epithelial cell lines is enhanced by hyperosmotic stress in vitro. Cell Tiss Res 297:467±473, 1999 Tiedemann RE, Fraser JD: Cross-linking of MHC class II molecules by staphylococcal enterotoxin A is essential for antigen-presenting cell and T cell activation. J Immunol 157:3958±3966, 1996 Tokura Y, Yagi J, O'Malley M, et al: Superantigenic staphylococcal exotoxins induced T cell proliferation in presence of Langerhans cells or class II-bearing keratinocytes and stimulate keratinocytes to produce T-cell-activating cytokines. J Invest Dermatol 102:31±38, 1994 Tokura Y, Furakawa F, Wakita H, Yagi H, Ushijima T, Takigawa M: T-cell proliferation to superantigen-releasing Staphylococcus aureus by MHC Class II-bearing keratinocytes underprotection from bacterial cytolysin. J Invest Dermatol 108:488±494, 1997 Travers JB, Huff JC, Rola-Pleszcynski M, Gelfand EW, Morelli JG, Murphy RC: Identi®cation of functional paltelet-activating factor receptors on human keratinocytes. J Invest Dermatol 105:816±823, 1995 Travers JB, Harrison KA, Johnson CA, Clay KL, Morelli JG, Murphy RC: Plateletactivating factor biosynthesis induced by various stimuli in human HaCaT keratinocytes. J Invest Dermatol 107:88±94, 1996 Travers JB, Hamid QA, Norris DA, et al: Epidermal HLA-DR and the enhancement of cutaneous reactivity to superantigenic toxins in psoriasis. J Clin Invest 104:1181±1189, 1999 Valeva A, Walev I, Pinkernell M, et al: Transmembrane barrel of staphylococcal apha toxin forms in sensitive, but not resistant cells. Proc Nat Acad Sci (USA) 94:11607±11611, 1997 Wakem P, Burns RP Jr, Ramirez F, et al: Allergens and irritants transcriptionally upregulate CD80 gene expression in human keratinocytes. J Invest Dermatol 114:1085±1092, 2000 Wakita H, Tokura Y, Furukawa F, Takigawa M: Staphylococcal enterotoxin B upregulates expression of ICAM-1 molecules on IFN-gamma-treated keratinocytes and keratinocyte cell lines. J Invest Dermatol 105:536±542, 1995 Walev I, Martin E, Jonas D, et al: Staphylococcal alpha-toxin kills human keratinocytes by permeabilizing the plasma membrane for monovalent ions. Infect Immun 61:4972±4979, 1993 White J, Herman A, Pullen AM, et al: The VB-speci®c superantigen staphylococcal enterotoxin B. stimulation of mature T cells and clonal deletion in neonatal mice. Cell 56:27±35, 1989 Woodward DF, Hawley SB, Williams LS, et al: Studies on the ocular pathology of PGD2. Invest Opthalm Vis Sci 31:138±146, 1990 Wrone-Smith T, Nickoloff BJ: Dermal injection of immunocytes induces psoriasis. J Clin Invest 98:1878±1887, 1996