Wiener J R, Joklik W K
Department of Microbiology and Immunology, Duke University Medical Center, Durham, North Carolina 27710.
Virology. 1987 Dec;161(2):332-9. doi: 10.1016/0042-6822(87)90125-5.
The sequences of the S3 genome segments of reovirus serotype 1 and 2 prototype strains are presented and these sequences are compared with the sequence of the serotype 3 S3 genome segment. The S3 genome segment encodes the nonstructural protein sigma NS which possesses affinity for ssRNA and appears to function in reovirus morphogenesis. The three S3 genome segments are closely related: all are 1198 nucleotides long and possess a single long open reading frame 366 codons long. They exhibit a serotype 1:3 relatedness pattern: there are only 13% mismatches between the S3 genome segments of serotypes 1 and 3, but 27 and 26% mismatches, respectively, between those of serotype 1 and 2 and serotype 3 and 2. The amino acid mismatches for the three sigma NS proteins are much lower (2.7, 13.9, and 13.7%, respectively), because the majority of nucleotide mismatches are in third base codon positions. The three sigma NS proteins possess a conserved secondary structure that is rich in alpha-helix content; in fact, the predicted alpha-helix content of these nonstructural proteins (about 50%) is much higher than that of the three other sigma size class proteins (about 20%), all of which are structural proteins. We also sequenced the S3 genome segment of a ts mutant of serotype 3 generated by treatment with nitrous acid and found a single nucleotide change that specifies an amino acid change that introduces a five-residue-long beta-sheet prone configuration into a long (80 amino acids) highly conserved alpha-helix in the C-terminal half of sigma NS. This change could account for this mutant's ts character. Finally, the three sigma NS proteins have diverged in only about 10% of positions, whereas the three sigma 1 proteins have diverged in about 70%. The rapid evolutionary divergence of the latter may be a result of several factors, including (i) the fact that sigma 1, but not sigma NS, is under immunologic selective pressure; (ii) the fact that the functions of sigma 1 (antigenicity and cell attachment) probably reside in two rather small domains that are not restricted spatially with respect to each other; and (iii) the fact that the functions of sigma NS, namely RNA binding and protein binding (during morphogenesis), require a highly specific overall protein configuration that may permit little variation.
本文展示了呼肠孤病毒1型和2型原型株S3基因组片段的序列，并将这些序列与3型S3基因组片段的序列进行了比较。S3基因组片段编码非结构蛋白σNS，该蛋白对单链RNA具有亲和力，似乎在呼肠孤病毒形态发生中发挥作用。这三个S3基因组片段密切相关：均为1198个核苷酸长，且都拥有一个366个密码子长的单一长开放阅读框。它们呈现出1型与3型的相关性模式：1型和3型的S3基因组片段之间只有13%的错配，但1型与2型以及3型与2型之间的错配分别为27%和26%。三种σNS蛋白的氨基酸错配要低得多（分别为2.7%、13.9%和13.7%），因为大多数核苷酸错配位于密码子的第三位。这三种σNS蛋白具有富含α螺旋的保守二级结构；事实上，这些非结构蛋白预测的α螺旋含量（约50%）远高于其他三种σ大小类别的蛋白（约20%），这三种都是结构蛋白。我们还对用亚硝酸处理产生的3型ts突变体的S3基因组片段进行了测序，发现了一个单核苷酸变化，该变化指定了一个氨基酸变化，在σNS C端一半的一个长（80个氨基酸）高度保守的α螺旋中引入了一个五残基长的β折叠倾向构型。这种变化可能解释了该突变体的ts特性。最后，三种σNS蛋白仅在约10%的位置发生了分歧，而三种σ1蛋白在约70%的位置发生了分歧。后者的快速进化分歧可能是多种因素导致的，包括：（i）σ1而非σNS受到免疫选择压力这一事实；（ii）σ1的功能（抗原性和细胞附着）可能存在于两个相对较小的结构域中，这两个结构域在空间上彼此不受限制；（iii）σNS的功能，即RNA结合和蛋白结合（在形态发生过程中），需要高度特异的整体蛋白构型，这可能几乎不允许有变化。
