Hohl D, Mehrel T, Lichti U, Turner M L, Roop D R, Steinert P M
Dermatology Branch, National Cancer Institute, National Institutes of Health, Bethesda, Maryland 20892.
J Biol Chem. 1991 Apr 5;266(10):6626-36.
We have isolated and characterized a full-length cDNA clone encoding human loricrin. Curiously, this protein displays major differences from the recently described mouse loricrin (Mehrel, T., Hohl, D., Nakazawa, H., Rothnagel, J.A., Longley, M.A., Bundman, D., Cheng, C.K., Lichti, U., Bisher, M.E., Steven, A. C., Steinert, P.M., Yuspa, S.H., and Roop, D.R. (1990) Cell 61, 1103-1112). Although both proteins are glycine-serine-cysteine-rich, the sequences have not been conserved. However, analysis of the sequences reveals a common motif of quasi-peptide repeats of an aliphatic or aromatic amino acid residue followed by several glycine and/or serine and cysteine residues. These sequences are interspersed and flanked by short glutamine- or glutamine/lysine-rich peptides. Thus loricrins consist of a family of cell envelope proteins of highly variable sequences that nevertheless retain common structural elements. We show that unlike all other putative protein components of the cell envelope, loricrins are highly insoluble, due at least in part to cross-linking by disulfide bonds. Furthermore, we have isolated four peptides from purified human cell envelopes that contain recognizable loricrin sequences and which are cross-linked by the N epsilon-(gamma-glutamyl)lysine isodipeptide bond. The presence of such bonds thus affords an explanation for the extraordinary insolubility of loricrin by cross-linking to the cell envelope and can also explain the low steady-state levels of monomeric loricrin in cytoskeletal extracts of epidermis. This study represents the first report of this isodipeptide cross-link in a protein component of the cornified cell envelope. We propose a model for the structure of loricrin in which (i) the unusual glycine-serine-rich sequences adopt a flexible loop conformation, indexed on the recurrent aliphatic residues; (ii) inter- or intramolecular isodipeptide and disulfide cross-links induce or stabilize folding of loricrin so as to form a more compact rosette-like structure; and (iii) the presence of the flexible glycine-rich loops necessarily will impact a flexible character to the cell envelope and entire epithelium.
我们已经分离并鉴定了一个编码人兜甲蛋白的全长cDNA克隆。奇怪的是,这种蛋白质与最近描述的小鼠兜甲蛋白存在重大差异(Mehrel, T., Hohl, D., Nakazawa, H., Rothnagel, J.A., Longley, M.A., Bundman, D., Cheng, C.K., Lichti, U., Bisher, M.E., Steven, A. C., Steinert, P.M., Yuspa, S.H., and Roop, D.R. (1990) Cell 61, 1103 - 1112)。尽管这两种蛋白质都富含甘氨酸、丝氨酸和半胱氨酸,但它们的序列并不保守。然而,对这些序列的分析揭示了一个共同的基序,即由一个脂肪族或芳香族氨基酸残基的准肽重复序列,随后是几个甘氨酸和/或丝氨酸以及半胱氨酸残基。这些序列相互穿插,并由富含谷氨酰胺或谷氨酰胺/赖氨酸的短肽侧翼包围。因此,兜甲蛋白由一系列细胞包膜蛋白组成,其序列高度可变,但仍保留共同的结构元件。我们发现,与细胞包膜的所有其他假定蛋白质成分不同,兜甲蛋白高度不溶,这至少部分归因于二硫键的交联。此外,我们从纯化的人细胞包膜中分离出四种肽,它们含有可识别的兜甲蛋白序列,并通过Nε-(γ-谷氨酰)赖氨酸异二肽键交联。这种键的存在因此解释了兜甲蛋白通过与细胞包膜交联而具有的非凡不溶性,也可以解释表皮细胞骨架提取物中单体兜甲蛋白的低稳态水平。这项研究代表了在角质化细胞包膜的蛋白质成分中首次报道这种异二肽交联。我们提出了一个兜甲蛋白结构模型,其中:(i) 异常的富含甘氨酸 - 丝氨酸的序列采用柔性环构象,以反复出现的脂肪族残基为索引;(ii) 分子间或分子内的异二肽和二硫键交联诱导或稳定兜甲蛋白的折叠,从而形成更紧凑的玫瑰花结样结构;(iii) 富含甘氨酸的柔性环的存在必然会赋予细胞包膜和整个上皮组织柔性特征。