Cho S H, Strickland I, Tomkinson A, Fehringer A P, Gelfand E W, Leung D Y
Divisions of Allergy-Immunology and Cell Biology, Department of Pediatrics, National Jewish Medical and Research Center, Denver, Colorado, USA.
J Invest Dermatol. 2001 May;116(5):658-63. doi: 10.1046/j.0022-202x.2001.01331.x.
Staphylococcus aureus is found on over 90% of atopic dermatitis skin lesions and is thought to contribute to skin inflammation via the production of potent exotoxins. In contrast, less than 5% of normal subjects harbor S. aureus. This suggests that an atopic immune response itself may play a role in preferential binding of S. aureus to the skin. To examine this issue more directly, we analyzed the S. aureus binding characteristics of skin in mice undergoing different T helper type 1 cell versus T helper type 2 cell inflammatory responses using a novel in vitro bacterial binding assay. BALB/C female mice were first sensitized to ovalbumin with alum or ovalbumin with complete Freund's adjuvant to induce T helper type 2 or T helper type 1 responses, respectively. Mice were then challenged intradermally with either saline (control) or ovalbumin. Forty-eight hours later, skin specimens were obtained from the challenge sites, and the number of S. aureus binding to each skin section was quantitated. Bacterial binding was found to be significantly greater at skin sites of BALB/C mice that had been ovalbumin/alum sensitized compared with ovalbumin/complete Freund's adjuvant sensitized (p < or = 0.01). When compared to the ovalbumin sensitized/challenged skin of wild type BALB/C mice or interferon-gamma gene knockout mice, interleukin-4, but not interferon-gamma, gene knockout mice had significantly less S. aureus binding at their ovalbumin sensitized/challenged skin sites. Mutant S. aureus strains that lacked either fibronectin- or fibrinogen-binding protein expression showed significantly reduced S. aureus binding compared with the parent wild type strain (p < 0.005). Moreover, preincubation of the wild type bacteria with fibronectin or fibrinogen, but not collagen, resulted in significantly less skin binding of S. aureus (p < 0.01). Incubation of skin with interleukin-4, and less so with interferon-gamma, led to more binding of wild type S. aureus but not of an S. aureus mutant deficient in fibronectin binding protein expression. After interleukin-4 incubation, but not interferon-gamma, epidermal immunoreactivity for fibronectin was observed in murine skin explants. These results show that a T helper type 2 inflammatory environment can promote skin binding by S. aureus and that this binding is mediated by fibronectin and fibrinogen.
超过90%的特应性皮炎皮肤损伤部位可检测到金黄色葡萄球菌，人们认为它通过产生强效外毒素导致皮肤炎症。相比之下，正常受试者中携带金黄色葡萄球菌的比例不到5%。这表明特应性免疫反应本身可能在金黄色葡萄球菌与皮肤的优先结合中发挥作用。为了更直接地研究这个问题，我们使用一种新型的体外细菌结合试验，分析了经历不同1型辅助性T细胞与2型辅助性T细胞炎症反应的小鼠皮肤的金黄色葡萄球菌结合特性。先用明矾致敏BALB/C雌性小鼠的卵清蛋白或用完全弗氏佐剂致敏卵清蛋白，分别诱导2型辅助性T细胞或1型辅助性T细胞反应。然后给小鼠皮内注射生理盐水（对照）或卵清蛋白。48小时后，从激发部位获取皮肤标本，并对结合到每个皮肤切片上的金黄色葡萄球菌数量进行定量。结果发现，与用卵清蛋白/完全弗氏佐剂致敏的BALB/C小鼠皮肤部位相比，用卵清蛋白/明矾致敏的BALB/C小鼠皮肤部位的细菌结合明显更多（p≤0.01）。与野生型BALB/C小鼠或干扰素-γ基因敲除小鼠的卵清蛋白致敏/激发皮肤相比，白细胞介素-4基因敲除小鼠（而非干扰素-γ基因敲除小鼠）在其卵清蛋白致敏/激发皮肤部位的金黄色葡萄球菌结合明显更少。缺乏纤连蛋白或纤维蛋白原结合蛋白表达的金黄色葡萄球菌突变株与亲本野生型菌株相比，金黄色葡萄球菌结合显著减少（p<0.005）。此外，用纤连蛋白或纤维蛋白原（而非胶原蛋白）对野生型细菌进行预孵育，导致金黄色葡萄球菌与皮肤的结合显著减少（p<0.01）。用白细胞介素-4孵育皮肤，用干扰素-γ孵育的效果稍差，导致野生型金黄色葡萄球菌的结合增加，但缺乏纤连蛋白结合蛋白表达的金黄色葡萄球菌突变株则不然。白细胞介素-4孵育后（而非干扰素-γ孵育后），在小鼠皮肤外植体中观察到纤连蛋白的表皮免疫反应性。这些结果表明，2型辅助性T细胞炎症环境可促进金黄色葡萄球菌与皮肤的结合，且这种结合由纤连蛋白和纤维蛋白原介导。
