ATP-driven cation pumps: alignment of sequences.
ATP-driven cation pumps: alignment of sequences.
复制标题
ATP 驱动的阳离子泵：序列比对。
DOI：
10.1042/bst0170972
复制
发表时间：
1989
影响因子：
3.9
通讯作者：
N. Green
中科院分区：
文献类型：
--
作者：
N. Green
早期的论文Ii 31描述了ATP驱动的阳离子泵序列的保守片段在分子的不同结构域内的位置，并讨论了它们的进化和功能意义。讨论的基础是ATP酶序列的比对，这是hcrc（图I，pp. 970-97 1）。比对包括了198 Y年6月之前发表的所有P型阳离子泵的序列，这些序列与该类的任何其他成员的同一性小于70%。这将比对限制为每种类型的一个代表，总共12个[11-12]，30个已经测序的。此处显示的最接近的相似性是Na+/K+和H+/K+泵之间的相似性（64%同一性）。后者的顺序只有当它不同于Na+/K+泵的顺序时才是b 'wen。序列根据它们来自动物、植物、原生动物或细菌的来源分组。它们被注释以显示高保守性区域（第1行）、疏水性跨膜区段和共有二级结构（第6行）和可能的结构域结构（第7行）。手动进行比对，从明确身份的区域开始。这使得大部分长的疏水片段相互对应。然后使用极性图1141确定这些区段的边界，其中三个残基的窗口强调将阻止插入膜的短极性中断。然后将这些片段排成一行，对它们相对于保守的膜外片段的位置和疏水区域内极性位点的对齐给予权重。疏水性身份没有高度加权，除非它们是保守运行的一部分，其中有很少。避免了漏洞。由于排列的疏水片段通常具有相同的长度，为了便于展示，任意修剪10组片段中的每一组。间隙也保持在最低限度的细胞质区域，其位置被限制在尽可能的区域的环和弯曲的共同所有的序列。疏水片段和明确预测的链和螺旋没有中断。一致二级结构预测和保守性评分是基于六个序列的初步比对。这些包括来自慢收缩肌肉1141的Ca+泵（未示出）、来自Torpedo 1141的Na+/K+泵（未示出）和其他四个，其标题在图1中加下划线。同一组被用于预测核苷酸结合结构域的共有结构[16]。Ca？+从质膜泵包含几个大的插入，这些插入写在单独的行上，以避免重新编号的必要性。来自细菌泵的三个长插入物和来自Leishrnuniu的一个长插入物也以这种方式使用字母处理。用Garnier等的方法分析了除跨膜片段外的6个序列。17 J，a、p和线圈+匝数概率自动求和。指示了预测的链（=> 3个残基）和螺旋（==> 4个残基）。基于其他序列的后续预测与这一共识基本一致。
An earlier paper Ii3 1 describes the location of conserved segments of the sequences of ATP-driven cation pumps within different structural domains of the molecule and discusses their evolutionary and functional significance. The basis for the discussion was an alignment of the ATPase sequences which is presented hcrc (Fig. I, pp. 970-97 1). The alignment includes sequences of all P-type cation pumps published up to June l98Y, which show less than 70% identity to any other member of the class. This limits the alignment to one representative of each type, 12 in all 11-12], out of 30 which have been sequenced. The closest similarity shown here is that between the Na+/K+ and H+/K+ pumps (64% identity). The sequence of the latter is b'wen only when it differs from that of the Na+/K+ pump. The sequences are grouped according to their origin from animals, plants, protozoa or bacteria. They are annotated to show regions of high conservation (line 1), hydrophobic transmembrane segments and consensus secondary structure (line 6) and probable domain structure (line 7). The alignment was performed manually, starting with regions of clear identity. This brought most of the long hydrophobic segments into correspondence. The boundaries of these segments were then determined using a polarity plot 1141 in which the window of three residues emphasized short polar interruptions which would block insertion into the membrane. These segments were then brought into line, giving weight both to their position relative to conserved extramembranous segments and to alignment of polar sites within the hydrophobic region. Hydrophobic identities were not highly weighted unless they were part of a conserved run, of which there were very few. Gaps were avoided. Since the aligned hydrophobic segments were usually of the same length, each of the 10 sets of segments was arbitrarily trimmed for convenience of display. Gaps were also kept to a minimum in the cytoplasmic region and their location was restricted as far as possible to regions of loops and bends common to all the sequences. Hydrophobic segments and clearly predicted strands and helices were not interrupted. This did not lead to any significant conflict with alignment of identities.The consensus secondary structure prediction and the conservation scores were based on a preliminary alignment of six sequences. These included the Ca" pump from slow twitch muscle 1141 (not shown), the Na+/K+ pump from Torpedo [IS I (not shown) and four others, the titles of which are underlined in Fig. 1. The same group was used to predict a consensus structure for the nucleotide-binding domain [16]. The Ca?+ pump from plasma membrane contains several large insertions which are written on a separate line to avoid the necessity of renumbering the preliminary alignment. Three long insertions from bacterial pumps and one from Leishrnuniu are also dealt with in this way using lowercase letters. The six sequences, excluding transmembrane segments, were analysed by the method of Garnier et ul. 17 J and the a, p and coil+ turn probabilities were summed automatically. Predicted strands (---> 3 residues) and helices (===> 4 residues) are indicated. Subsequent predictions based on the other sequences were mainly consistent with this consensus.