Klenk H P, Zillig W
Max-Planck-Institut für Biochemie, Martinsried, Germany.
J Mol Evol. 1994 Apr;38(4):420-32. doi: 10.1007/BF00163158.
The branching topology of the archaeal (archaebacterial) domain was inferred from sequence comparisons of the largest subunit (B) of DNA-dependent RNA polymerases (RNAP). Both the nucleic acid sequences of the genes coding for RNAP subunit B and the amino acid sequences of the derived gene products were used for phylogenetic reconstructions. Individual analysis of the three nucleotide positions of codons revealed significant inequalities with respect to guanosine and cytosine (GC) content and evolutionary rates. Only the nucleotides at the second codon positions were found to be unbiased by varied GC contents and sufficiently conserved for reliable phylogenetic reconstructions. A decision matrix was used for the combination of the results of distance matrix, maximum parsimony, and maximum likelihood methods. For this purpose the original results (sums of squares, steps, and logarithms of likelihoods) were transformed into comparable effective values and analyzed with methods known from the theory of statistical decisions. Phylogenetic invariants and statistical analysis with resampling techniques (bootstrap and jackknife) confirmed the preferred branching topology, which is significantly different from the topology known from phylogenetic trees based on 16S rRNA sequences. The preferred topology reconstructed by this analysis shows a common stem for the Methanococcales and Methanobacteriales and a separation of the thermophilic sulfur archaea from the methanogens and halophiles. The latter coincides with a unique phylogenetic location of a characteristic splitting event replacing the largest RNAP subunit of thermophilic sulfur archaea by two fragments in methanogens and halophiles. This topology is in good agreement with physiological and structural differences between the various archaea and demonstrates RNAP to be a suitable phylogenetic marker molecule.
古菌（古细菌）域的分支拓扑结构是通过对依赖DNA的RNA聚合酶（RNAP）最大亚基（B）的序列比较推断出来的。编码RNAP亚基B的基因的核酸序列以及衍生基因产物的氨基酸序列都被用于系统发育重建。对密码子的三个核苷酸位置进行单独分析，发现鸟嘌呤和胞嘧啶（GC）含量以及进化速率存在显著差异。只有第二个密码子位置的核苷酸不受不同GC含量的影响，并且足够保守，可用于可靠的系统发育重建。使用决策矩阵来组合距离矩阵、最大简约法和最大似然法的结果。为此，将原始结果（平方和、步数以及似然对数）转换为可比的有效值，并使用统计决策理论中已知的方法进行分析。系统发育不变量和重采样技术（自展法和刀切法）的统计分析证实了首选的分支拓扑结构，该结构与基于16S rRNA序列的系统发育树已知的拓扑结构有显著差异。通过该分析重建的首选拓扑结构显示甲烷球菌目和甲烷杆菌目有一个共同的主干，嗜热硫古菌与产甲烷菌和嗜盐菌分离。后者与一个独特的系统发育位置相吻合，即一个特征性的分裂事件，嗜热硫古菌的最大RNAP亚基在产甲烷菌和嗜盐菌中被两个片段取代。这种拓扑结构与各种古菌之间的生理和结构差异非常吻合，并证明RNAP是一种合适的系统发育标记分子。
