Uncovering ancient transcription systems with a novel evolutionary indicator

Abstract

TBP and TFIIB are evolutionarily conserved transcription initiation factors in archaea and eukaryotes. Information about their ancestral genes would be expected to provide insight into the origin of the RNA polymerase II-type transcription apparatus. In obtaining such information, the nucleotide sequences of current genes of both archaea and eukaryotes should be included in the analysis. However, the present methods of evolutionary analysis require that a subset of the genes should be excluded as an outer group. To overcome this limitation, we propose an innovative concept for evolutionary analysis that does not require an outer group. This approach utilizes the similarity in intramolecular direct repeats present in TBP and TFIIB as an evolutionary measure revealing the degree of similarity between the present offspring genes and their ancestors. Information on the properties of the ancestors and the order of emergence of TBP and TFIIB was also revealed. These findings imply that, for evolutionarily early transcription systems billions of years ago, interaction of RNA polymerase II with transcription initiation factors and the regulation of its enzymatic activity was required prior to the accurate positioning of the enzyme. Our approach provides a new way to discuss mechanistic and system evolution in a quantitative manner.

Figure 1. Schematic representation of the present evolutionary distance (d) and the distance between the first and second repeats (d_DR).

(A) Schematic representation of evolutionary distances. The evolutionary distances between two genes in the gene family are shown as d₁–d₆. (B) Calculated evolutionary distances d₁–d₆ are utilized to prepare an unrooted phylogenetic tree. (C) Schematic drawing of an unrooted phylogenetic tree of archaeal and eukaryotic genes. The position of the MRCA for both archaeal and eukaryotic genes cannot be determined on the unrooted phylogenetic tree. When archaeal genes are considered as an outer group (distal relative genes), the MRCA for eukaryotic genes can be placed (left panel). When eukaryotic genes are considered as an outer group, the MRCA for archaeal genes can be placed (right panel). (D) Relationship of gene duplication, accumulated mutations, and d_DR. The EA-gene (middle panel) is generated by a gene duplication of a prototype gene (upper panel). The d_DR value of the EA-gene (t = 0) is zero due to two identical nucleotide sequences in the direct repeat. The d_DR value of the present offspring gene (lower panel) can be utilized as an indicator of the evolutionary distance between the EA-gene and the present offspring gene. (E) Relationship between the phylogenetic tree and d_DR defined in this study. d is the path length “between two distinct genes” via their MRCA (e.g., the red line in the right panel). On the other hand, d_DR is the path length “between the first and second repeats in one gene” via the hypothetical EA-gene (e.g., the red line in the left panel). Therefore, d reflects the evolutionary distance between the present gene and one of the ancestral genes, but d_DR could be a reasonable indicator of the evolutionary distance between a present gene and its EA-gene. (F) Schematic drawing of the relationship between the phylogenetic tree and d_DR. The usual phylogenetic tree is prepared based on the evolutionary distances (d) of two genes in the gene family. On the other hand, d_DR is a reasonable indicator of the evolutionary distances between the EA-gene and each of the present offspring genes.

Figure 2. Direct repeats present within TBP and TFIIB are derived from their EA-genes generated by single gene duplication.

(A,D) Phylogenetic trees drawn with the nucleotide sequences of the conserved core region of TBP (A) and TFIIB (D) from 34 species. Abbreviations of species names are given in the footnote of Table 1. (B,E) Phylogenetic trees drawn with the nucleotide sequences of the first and second repeats of TBP (B) and TFIIB (E) from 34 species. Red and cyan indicate, respectively, the first and second repeats of the TBP (B) and TFIIB (E) genes. (C,F) The d_DR values of TBP (C) and TFIIB (F) are shown on their phylogenetic trees using red-blue coloring by the d_DR values.

Figure 3. Correlation between d_DR and the amino-acid compositions of TBP.

(A) Correlations between d_DR and the number of the specific amino-acid residues for TBP (Asp (D), Glu (E), Arg (R), Phe (F), and Ser (S)). The best fitting lines are shown in red. The correlation coefficient (r) and p-value (p) are shown in each graph. (B) Sphere models of TBP molecules from M. jannaschii (Mj), S. acidocaldarius (Sa), P. woesei (Pw), S. cerevisiae (Sc), A. thaliana (At), and H. sapiens (Hs). The upper and lower panels show front and back views, respectively. Asp and Glu residues are shown in red, Arg residues are shown in blue, Phe residues are shown in cyan, and Ser residues are shown in yellow. The d_DR value of each molecule is also shown. Black curved lines indicate the DNA binding surface of TBP. Dotted curved lines indicate the TFIIB binding surface of TBP.

Figure 4. Correlation between d_DR and the amino-acid compositions of TFIIB.

(A) Correlations between d_DR and the number of the specific amino-acid residues for TFIIB (Arg (R) and Gln (Q)). The best fitting lines are shown in red. The correlation coefficient (r) and p-value (p) are shown in each graph. (B) Sphere models of TFIIB molecules from P. woesei (Pw) and H. sapiens (Hs). The upper and lower panels show front and back views, respectively. Arg residues are shown in blue, and Gln residues are shown in green. The d_DR value of each molecule is also shown. Black curved lines indicate the DNA binding surface of TFIIB. Dotted curved lines indicate the TBP binding surfaces of TFIIB.

Figure 5. Evolutionary relationship between TBP and TFIIB.

(A,B) Evolutionary correlation between the TBP and TFIIB genes analyzed by d_DR (A) and d_Mj (B). The best fitting lines are shown in red. The correlation coefficient (r) and p-value (p) are shown. (C,D) Correlation between d_DR and d_Mj for TBP (C) and TFIIB (D). The correlation coefficient (r) and p-value (p) for each fitting are shown. Several plots near the x-axis deviate from the best fitting lines. These are plots for close relatives of M. jannaschii. Since the starting point of d_Mj-calculation is the M. jannaschii gene, d_Mj decreases quickly in the close relatives of M. jannaschii. However, d_DR does not decrease in the close relatives of M. jannaschii, because the starting point of d_DR-calculation is the EA-gene. This is the reason for the observed deviations from the best fitting line in the d_Mj plots.

Figure 6. Schematic representation of the d_DR analysis and its application.

(A) Comparison between the present method (d) and our novel analysis (d_DR). (B) Schematic representation of the evolutionary development of TBP. The d_DR values of TBP are indicated on the phylogenetic trees using a color continuum from red (low d_DR) to blue (high d_DR). (C) Our analysis implies that the TBP gene was generated after the emergence of the TFIIB gene. This is the first time that the emerging order of TBP and TFIIB genes has been reported in the study of molecular evolution, and our results should thus provide novel insights into the evolutionary development of the transcription apparatus and other systems. The order of emergence of other general transcription factors, such as TAFs, TFIIA, TFIIE, TFIIF, TFIIH, remains unknown.