Recent Advances in Archaeal Translation Initiation

Abstract

Translation initiation (TI) allows accurate selection of the initiation codon on a messenger RNA (mRNA) and defines the reading frame. In all domains of life, translation initiation generally occurs within a macromolecular complex made up of the small ribosomal subunit, the mRNA, a specialized methionylated initiator tRNA, and translation initiation factors (IFs). Once the start codon is selected at the P site of the ribosome and the large subunit is associated, the IFs are released and a ribosome competent for elongation is formed. However, even if the general principles are the same in the three domains of life, the molecular mechanisms are different in bacteria, eukaryotes, and archaea and may also vary depending on the mRNA. Because TI mechanisms have evolved lately, their studies bring important information about the evolutionary relationships between extant organisms. In this context, recent structural data on ribosomal complexes and genome-wide studies are particularly valuable. This review focuses on archaeal translation initiation highlighting its relationships with either the eukaryotic or the bacterial world. Eukaryotic features of the archaeal small ribosomal subunit are presented. Ribosome evolution and TI mechanisms diversity in archaeal branches are discussed. Next, the use of leaderless mRNAs and that of leadered mRNAs having Shine-Dalgarno sequences is analyzed. Finally, the current knowledge on TI mechanisms of SD-leadered and leaderless mRNAs is detailed.

Keywords: Shine-Dalgarno; evolution; leaderless; mRNA; ribosomal proteins.

Figure 1

Schematic views of the translation initiation (TI) steps in the three domains of life. The figure illustrates the main steps in bacteria (left), in archaea (middle), and in eukarya (right). Bacterial 30S subunit recruits the messenger RNA (mRNA), often due to the base pairing between a Shine-Dalgarno sequence (SD) with an ASD sequence at the 3'-end of 16S rRNA. Three initiation factors, IF1, IF2, and IF3 favor the recruitment of the initiator tRNA and its pairing with the start codon. The formyl-methionyl moiety of the initiator tRNA is important for recognition by IF2. After start codon recognition, IF3 is released and the large ribosomal subunit is recruited with the help of IF2 (see Mechulam et al., 2011; Rodnina, 2018 for reviews). Archaea and eukarya share a common set of factors comprising e/aIF1A, e/aIF1, e/aIF2, and e/aIF5B (see also Figure 2). e/aIF2 heterotrimer is represented with a three-color code (α subunit in cyan, β subunit in red, and γ subunit in green). In canonical eukaryotic translation, a pre-initiation complex, containing the small ribosomal subunit, the methionylated initiator tRNA, and initiation factors, forms at the 5'-capped end of the mRNA. The complex then scans the mRNA until a start codon in a suitable environment is found. Base-pairing of the tRNA anticodon with the AUG start codon triggers eIF1 release followed by the release of Pi resulting from GTP hydrolysis by eIF2 (Algire et al., 2005). In turn, eIF2, eIF3, and eIF5 are released; eIF5B-GTP is recruited and favors joining with the large ribosomal subunit (see Hinnebusch, 2017 for a review). Archaea often use an SD sequence for mRNA recruitment. The 30S subunit is then definitely positioned with the start codon in the P site thanks to base-pairing with the tRNA anticodon. Overall, the four initiation factors aIF1, aIF1A, aIF2, and aIF5B play similar roles as their eukaryotic counterparts (see text and Schmitt et al., 2019 for a mechanism-oriented review). In the three cases, the translation competent IC is formed after the release of e/aIF1A (or IF1 in bacteria) and e/aIF5B (or IF2 in bacteria). In eukarya, the complex formed by eIF4E + eIF4G + eIF4A is known as eIF4F. eIF3, composed of 6 (yeast) to 13 (mammals) subunits is represented as a yellow oval. The figure is adapted from Schmitt et al. (2019).

Figure 2

Translation initiation factors in the three domains of life. The structures of the archaeal translation initiation factors and of their orthologues in eukaryotes and bacteria (when present) are shown. e/aIF2 is colored as in Figure 1. The unknown structure of the N-domain specific of eukaryotic eIF2β is shown as an oval. The structure of aIF2 is from PDB 3V11 (Schmitt et al., 2012), those of aIF1 and aIF1A are from Coureux et al. (2016). The structures of eIF2, eIF1, and eIF1A are from PDB 6FYX (Llacer et al., 2018). IF1 is from PDB 3I4O (Hatzopoulos and Mueller-Dieckmann, 2010). Bacterial IF3 is a two-domain protein. The correspondence between IF3 and e/aIF1 is based on a structural and functional resemblance of the IF3 C-terminal domain with e/aIF1. Despite this resemblance, the topologies of the two α–β folds are different. This suggests that they do not derive from a common ancestor. aIF5B is from PDB 1G7T (Roll-Mecak et al., 2000), eIF5B is from PDB 4N3N (Kuhle and Ficner, 2014), and IF2 is from PDB 5LMV (Hussain et al., 2016). The color code for e/aIF5B/IF2 is as follows: G-domain and domain II in green, domain III in light orange, linker in yellow, and domain IV in red. The specific archaeal helix in domain IV is shown in blue. ^*The catalytic γ and ε subunits of eIF2B are missing in archaea. The function of the eIF2B α, β, δ homologues in archaea is not clear and may be unrelated to translation initiation (Dev et al., 2009; Gogoi et al., 2016). ^**The aIF4A orthologue is present in many archaea. However, deletion of the corresponding gene in Haloferax volcanii showed only a small phenotype (Gäbel et al., 2013).

Figure 3

The small ribosomal subunit in the three domains of life. (A) 30S from Thermus thermophilus (PDB 5LMV; Hussain et al., 2016). (B) 30S from Pyrococcus abyssi (PDB 6SWC; Coureux et al., 2020). (C) 40S from Kluyveromyces lactis (PDB 6FYX; Llacer et al., 2018). Ribosomal proteins are colored as follows; universal in green, bacterial in cyan, eukaryotic and archaeal in dark blue, and eukaryotic only (as compared to P. abyssi) in orange. The P site tRNA is in yellow spheres and the mRNA in light blue spheres. rRNA is in gray. Regions of the ribosomal RNA of the P. abyssi small subunit that are not in the common core as defined in Bernier et al. (2018) are shown with red spheres (middle view, Table 2).

Figure 4

Analysis of translation start regions. (A) Analysis of the translation start regions in Sulfolobus solfataricus, P. abyssi, and the Asgard Candidatus Prometeoarchaeum syntrophicum. DNA sequences (60 nt around the first base of the start codon) were extracted from the genomic sequences (She et al., 2001; Cohen et al., 2003; Imachi et al., 2020). Annotations as corrected by Wurtzel et al. (2010) have been used for S. solfataricus. Sequence logos were created using Weblogo (Crooks et al., 2004). (B) Detailed analysis of the translation start regions in S. solfataricus. See also Tolstrup et al. (2000) for an earlier analysis. For each indicated category of transcript (number of ORFs in parentheses), the percentage of AUG, GUG, and UUG start codons are indicated. The position of potential 16S rRNA binding sites (Shine-Dalgarno sequences) in the upper two logos is shown by a blue line. Note that in fully leaderless genes (0 nt), the occurrence of T at the −1 position and the avoidance of UUG as start codon are likely linked to signals for RNA polymerase.

Figure 5

Comparison of the mRNA exit channels in the three domains of life. Surface representations of the mRNA exit channels of representative structures in the three domains of life. The mRNA is shown in blue and the 3'extremity of the rRNA is shown in orange. R-proteins are labeled using the Ban et al. (2014) nomenclature. The figure illustrates the similarity of the archaeal and eukaryotic mRNA exit channels vs. the bacterial channel. TACK and Asgard Archaea have three additional proteins in their SSU as compared to thermococcales (eS25, eS26, and eS30). The structures are from PDB 4VY4 (Yusupova et al., 2006), PDB 6SWC (Coureux et al., 2020), and PDB 6FYX (Llacer et al., 2018). The figure is adapted from Coureux et al. (2020).

Figure 6

Steps of translation initiation in P. abyssi. Surface representation of successive P. abyssi translation initiation complexes. aIF2, aIF1, and aIF1A are shown with the same color code as in Figure 2. mRNA is in dark blue, initiator tRNA is in yellow, and the h44 helix is in black. The figure shows start codon selection in the full archaeal TI complex where IC0-P_REMOTE and IC1-P_IN are in equilibrium until a start codon is found in the P site (Coureux et al., 2016). Codon:anticodon pairing stabilizes the IC1 state and triggers aIF1 release. In IC2A, the initiator tRNA fully accommodates. Release of aIF1 would cause both Pi release from aIF2γ and h44 adjustments leading to irreversible aIF2 release. The figure is adapted from Coureux et al. (2020).

Figure 7

Interaction of the initiator tRNA at the P site with the universal proteins uS9, uS13, and uS19. (A) Overall view of the accommodated tRNA as observed in the IC2 complex from P. abyssi (Coureux et al., 2016). The color code is the same as in Figure 6. (B) Close up showing the interaction of the strictly conserved terminal arginine of uS9 with the codon:anticodon duplex. (C) Close up showing the interaction of the uS13 and uS19 tails with the major groove of the anticodon stem-loop of the initiator tRNA.

Figure 8

Sequence alignments of the uS13 and uS19 C-terminal tails. uS13 and uS19 sequences were extracted and aligned using Pipealign (Plewniak et al., 2003). After visual inspection, several families regarding the C-terminal tail specificities in Archaea were identified. For uS13, ca. 250 archaeal sequences were used to which we added manually 100 sequences from halobacteria. For uS19, ca. 600 archaeal sequences were used. Typical representatives of each family are shown. The Homo sapiens sequence is used as a eukaryotic reference for comparison.

Figure 9

Domain specificities of uS13 and uS19 cores. The three panels show the B1a-B1b/c bridge. (A) Archaeal case. The view is a composite using the SSU from PDB 6SWC (Coureux et al., 2020) and the LSU from PDB 4V6U (Armache et al., 2013). (B) Structure of human ribosome from PDB 6Y0G (Bhaskar et al., 2020). (C) Structure of Escherichia coli ribosome from PDB 5AFI (Fischer et al., 2015). The views show that archaeal and eukaryotic uS13 and uS19 have specific extensions (in red and pink, respectively) that contribute to the intersubunit bridge. Bacteria have instead a specific ribosomal protein bL31.