Safley S A, Cluff C W, Marshall N E, Ziegler H K
Department of Microbiology and Immunology, Emory University School of Medicine, Atlanta, GA 30322.
J Immunol. 1991 May 15;146(10):3604-16.
Using a murine model, we investigated the role of the bacterial exotoxin listeriolysin O (LLO) in cellular immunity to Listeria monocytogenes. A correlation between LLO production by infecting bacteria and generation of protective immunity to virulent LLO-producing bacteria was noted. Using isogeneic hemolysin (Hly+ or Hly-) strains of L. monocytogenes, we demonstrated that LLO production by infecting bacteria is required to elicit T cells reactive both to bacteria-associated Ag and to the secreted LLO molecule as measured by IL-2 production in vitro. Distinct sets of T cells specific for largely nonoverlapping pools of antigenic determinants represented by LLO and cell-associated Ag (heat-killed L. monocytogenes) are generated after infection. We have used models for prediction of T cell epitopes based on primary structure of LLO, and synthetic amphipathic LLO peptides were evaluated as Ag in vitro or as immunogenes in vivo. Infection of several strains of mice (H-2k and H-2d) with LLO-producing L. monocytogenes resulted in the generation of T cells that could respond consistently to two peptides, LLO 215-234 and LLO 354-371. Mouse strains lacking expression of I-E molecules (e.g., B10.A(4R) and C57BL/6) responded to LLO but not to the peptides tested. With C3HeB/FeJ mice, antibodies to I-Ek blocked the presentation of LLO 215-234. The importance of the N-terminal portion of LLO 215-234 was evidenced by the drastic reduction in antigenic activity of truncated peptides (e.g., LLO 221-234 and LLO 224-234). LLO 215-234, the strongest and most consistent activator of T cells from L. monocytogenes-immune mice, fit well some models for antigenic peptides in several ways. It could be predicted to form an amphipathic alpha-helix, it contained multiple "Rothbard motifs" (charged residue or glycine, two or three hydrophobic amino acids and then a glycine or polar residue), it had a net charge of +2, and it contained the correct spacing of amino acids (five to six residues between a hydrophobic and basic amino acid) that is characteristic of I-Ek-binding peptides. Immunization with 8 of 10 synthetic LLO peptides generated T cells that recognized the immunizing peptide in vitro, but such T cells were only poorly reactive with LLO. Our results indicate that LLO is an important target Ag for stimulation of CD4+ L. monocytogenes-specific T cells, and that LLO 215-234 is antigenically dominant in C3HeB/FeJ mice.
利用小鼠模型,我们研究了细菌外毒素李斯特菌溶血素O(LLO)在针对单核细胞增生李斯特菌的细胞免疫中的作用。我们注意到感染细菌产生的LLO与对产生毒性LLO的细菌产生保护性免疫之间存在相关性。利用单核细胞增生李斯特菌的同基因溶血素(Hly+或Hly-)菌株,我们证明感染细菌产生的LLO是引发T细胞所必需的,这些T细胞对细菌相关抗原以及分泌的LLO分子均有反应,这通过体外IL-2产生来衡量。感染后会产生针对LLO和细胞相关抗原(热灭活的单核细胞增生李斯特菌)所代表的大量非重叠抗原决定簇池的不同T细胞组。我们利用基于LLO一级结构的T细胞表位预测模型,并评估了合成两亲性LLO肽在体外作为抗原或在体内作为免疫原的效果。用产生LLO的单核细胞增生李斯特菌感染几种小鼠品系(H-2k和H-2d)后,产生了能够一致地对两种肽LLO 215 - 234和LLO 354 - 371作出反应的T细胞。缺乏I-E分子表达的小鼠品系(如B10.A(4R)和C57BL/6)对LLO有反应,但对所测试的肽无反应。对于C3HeB/FeJ小鼠,抗I-Ek抗体阻断了LLO 215 - 234的呈递。截短肽(如LLO 221 - 234和LLO 224 - 234)抗原活性的急剧降低证明了LLO 215 - 234 N端部分的重要性。LLO 215 - 234是来自单核细胞增生李斯特菌免疫小鼠的T细胞最强且最一致的激活剂,在几个方面很好地符合了抗原肽的一些模型。它可被预测形成两亲性α-螺旋,包含多个“罗斯巴德基序”(带电荷残基或甘氨酸、两到三个疏水氨基酸,然后是一个甘氨酸或极性残基),净电荷为+2,并且包含I-Ek结合肽特有的氨基酸正确间距(疏水和碱性氨基酸之间有五到六个残基)。用10种合成LLO肽中的8种进行免疫产生了在体外能识别免疫肽的T细胞,但这种T细胞与LLO的反应性很差。我们的结果表明,LLO是刺激CD4+单核细胞增生李斯特菌特异性T细胞的重要靶抗原,并且LLO 215 - 234在C3HeB/FeJ小鼠中具有抗原优势。