Parry David A D, Smith Thomasin A, Rogers Michael A, Schweizer Jürgen
Institute of Fundamental Sciences, Massey University, Private Bag 11-222, Palmerston North, New Zealand.
J Struct Biol. 2006 Aug;155(2):361-9. doi: 10.1016/j.jsb.2006.03.018. Epub 2006 Apr 27.
In this paper, we undertake a sequence analysis of the human keratin-associated proteins (KAP). This analysis has revealed two fundamental pentapeptide quasi-repeats (A and B) of the form C-C-X-P-X and C-C-X-S/T-S/T, respectively. The A repeats are also commonly found in two subforms A1 and A2, -C-C-Q-P-X and C-C-R-P-X, respectively-similar to those found in sheep wool 30-40 years previously. Some high-sulphur and ultra-high sulphur proteins contain predominantly A repeats or B repeats but not regular combinations of them, whereas others are characterised by a contiguous pair of pentapeptide repeats that largely (though imperfectly) alternate to generate decapeptide motifs of the form AB, A1B or A2B. The A and B repeats sometimes occur in complex runs and can generate both 19- and 20-residue repeats of the form BABB' or BA1AA, respectively, where the prime indicates a motif truncated by one residue. Likewise, a 42-residue repeat with BA1BXAAAB (40 residues) separated by a di-serine (two residues) has been observed in an ultra-high sulphur protein from cuticle. To understand the possible conformations adopted by the A and B motifs, a search was initiated of the PDB structural database for a number of overlapping pentapeptide repeats. The total number of matches was 658 and these were found in 451 different proteins. From representative and unique structures the means and standard deviations were calculated for the Phi(i) and Psi(i) angles for the C-C-X-P-X and the C-C-X-S/T-S/T motifs. Molecular modelling has been employed to represent the "average" structure found from crystallographic and nmr data determined for each motif in other proteins. The conformation of consecutive A repeats with proline residues in the cis state is akin to a string of disulphide bond-stabilised pentapeptide knots between which there is relative freedom of rotation about the single bonds that link them. For B pentapeptides, however, the likelihood that a similar disulphide bond is formed appears much lower. This may give additional conformational flexibility to the chain and hence allow the A pentapeptides greater opportunity to interact appropriately with the IF via disulphide bonds, ionic interactions and/or hydrogen bonding.
在本文中，我们对人类角蛋白相关蛋白（KAP）进行了序列分析。该分析揭示了两种基本的五肽准重复序列（A和B），其形式分别为C-C-X-P-X和C-C-X-S/T-S/T。A重复序列也常见于两种亚型A1和A2中，分别为-C-C-Q-P-X和C-C-R-P-X，类似于30-40年前在羊毛中发现的序列。一些高硫和超高硫蛋白主要包含A重复序列或B重复序列，但并非它们的常规组合，而其他蛋白的特征是一对相邻的五肽重复序列，它们在很大程度上（尽管并不完美）交替出现，形成AB、A1B或A2B形式的十肽基序。A和B重复序列有时会以复杂的形式出现，分别产生BABB'或BA1AA形式的19和20个残基的重复序列，其中带撇号表示一个基序被截短了一个残基。同样，在来自角质层的一种超高硫蛋白中，观察到了一个42个残基的重复序列，其形式为BA1BXAAAB（40个残基），中间被一个双丝氨酸（两个残基）隔开。为了了解A和B基序可能采取的构象，我们在蛋白质数据银行（PDB）结构数据库中搜索了多个重叠的五肽重复序列。匹配总数为658个，它们存在于451种不同的蛋白质中。从具有代表性和独特性的结构中，计算了C-C-X-P-X和C-C-X-S/T-S/T基序的Phi(i)和Psi(i)角的平均值和标准差。分子建模已被用于呈现从其他蛋白质中为每个基序测定的晶体学和核磁共振数据中找到的“平均”结构。脯氨酸残基处于顺式状态的连续A重复序列的构象类似于一串由二硫键稳定的五肽结，在这些结之间，连接它们的单键有相对的旋转自由度。然而，对于B五肽来说，形成类似二硫键的可能性似乎要低得多。这可能会给链带来额外的构象灵活性，从而使A五肽有更多机会通过二硫键、离子相互作用和/或氢键与中间丝（IF）进行适当的相互作用。
