Skip to main page content
U.S. flag

An official website of the United States government

Dot gov

The .gov means it’s official.
Federal government websites often end in .gov or .mil. Before sharing sensitive information, make sure you’re on a federal government site.

Https

The site is secure.
The https:// ensures that you are connecting to the official website and that any information you provide is encrypted and transmitted securely.

Access keys NCBI Homepage MyNCBI Homepage Main Content Main Navigation
. 2021 Jun 21;49(11):6529-6548.
doi: 10.1093/nar/gkab436.

The structure of the mouse ADAT2/ADAT3 complex reveals the molecular basis for mammalian tRNA wobble adenosine-to-inosine deamination

Affiliations

The structure of the mouse ADAT2/ADAT3 complex reveals the molecular basis for mammalian tRNA wobble adenosine-to-inosine deamination

Elizabeth Ramos-Morales et al. Nucleic Acids Res. .

Abstract

Post-transcriptional modification of tRNA wobble adenosine into inosine is crucial for decoding multiple mRNA codons by a single tRNA. The eukaryotic wobble adenosine-to-inosine modification is catalysed by the ADAT (ADAT2/ADAT3) complex that modifies up to eight tRNAs, requiring a full tRNA for activity. Yet, ADAT catalytic mechanism and its implication in neurodevelopmental disorders remain poorly understood. Here, we have characterized mouse ADAT and provide the molecular basis for tRNAs deamination by ADAT2 as well as ADAT3 inactivation by loss of catalytic and tRNA-binding determinants. We show that tRNA binding and deamination can vary depending on the cognate tRNA but absolutely rely on the eukaryote-specific ADAT3 N-terminal domain. This domain can rotate with respect to the ADAT catalytic domain to present and position the tRNA anticodon-stem-loop correctly in ADAT2 active site. A founder mutation in the ADAT3 N-terminal domain, which causes intellectual disability, does not affect tRNA binding despite the structural changes it induces but most likely hinders optimal presentation of the tRNA anticodon-stem-loop to ADAT2.

PubMed Disclaimer

Figures

Graphical abstract
Graphical abstract
ADAT catalyses tRNAs wobble adenosine-to-inosine modification through ADAT3 recognition of tRNAs 3D structure and anticodon-loop presentation to ADAT2 catalytic subunit. ADAT3 V128M mutation hampers presentation, leading to neurodevelopmental disorders.
Figure 1.
Figure 1.
Sequence alignment of mouse ADAT2 and ADAT3 subunits and Escherichia coli TadA. The alignment of the sequences of mouse (M.musculus) ADAT2 (pink) and ADAT3 (N-terminal domain, cyan; C-terminal domain, blue) subunits and of E. coli TadA (black) shows an overall conservation of their deaminase domain (lower part of alignment) which extends to their secondary structure elements. ADAT3 has diverged from ADAT2 and TadA, including a specific N-terminal domain (cyan) and a shorter C-terminal α-helix (boxed). Residues not observed in density are italicized and the ADAT3-specific loop removed for crystallization is shown in grey and boxed. Numbering as well as secondary structure elements for mouse ADAT2 and ADAT3 are shown above and below the alignment, respectively. Yellow stars: zinc binding residues. The fourth zinc ligand in ADAT3 is represented by a red-framed yellow star. Blue star: TadA (and potentially ADAT2) glutamate involved in proton shuttling. Purple diamonds: TadA (and potentially ADAT2) residues involved in wobble adenine binding. Orange diamonds: ADAT3 residues participating in ADAT3 V128 (red circled yellow diamond) hydrophobic core. The sequence alignments were produced with Espript (48).
Figure 2.
Figure 2.
Structure of the mouse ADAT2/ADAT3 complex. (A) Ribbon representation of the 2.1 Å resolution structure of the mouse ADAT complex. ADAT2 is shown in pink, while ADAT3 N- and C-terminal domains are shown in cyan and blue, respectively, using the same colour code as in Figure 1. The zinc ions of both subunits are shown as orange spheres. ADAT2 active site and ADAT3 very N-terminus, which participates to the central β-sheet of ADAT3 C-terminal domain, are indicated. (B) Superposition of ADAT3 N-terminal domains from both structures. The different positioning of ADAT3 very N-terminus, which is linked to ADAT3 C-terminal domain, reveals that ADAT3 N- and C-terminal domains can rotate by at least 90° with respect to each other. (C andD) Close-ups of the ADAT3-N and ADAT3-C interfaces in both ADAT structures. The residues that participate in the loose interactions between the two domains are shown as sticks and labelled. Interactions between ADAT3 very N-terminus (residues 25–33) and ADAT3-C are shown, demonstrating the tight and stable interface between these two regions.
Figure 3.
Figure 3.
Molecular basis of ADAT3 inactivity. (A) Close-up view of the ADAT2 (left panel) and ADAT3 (right panel) zinc-binding pockets and their superpositions (middle panel). Zinc ions are shown as orange spheres. While ADAT2 (left panel) shows a canonical zinc coordination, including a catalytic water (red sphere), ADAT3 (right panel) has a fourth protein ligand (D348, the before last ADAT3 residue) that replaces the catalytic water. Capping of the zinc-binding pocket by the last ADAT3 residue, P349, as well as replacement of the catalytic glutamate by a valine (V225) in ADAT3 further provide the molecular basis of ADAT3 inactivation. These different interactions within the zinc-pocket of ADAT3 are allowed by the shortening of ADAT3 α8 helix compared to the longer α7 helix of ADAT2 (middle panel). (B) Wall-eye stereo view of ADAT3 zinc binding site. The 2Fo-Fc electron density is contoured at 1σ.
Figure 4.
Figure 4.
ADAT2 but not ADAT3 can accommodate an anticodon stem-loop in its active site. (A) Structure of an anticodon-stem-loop (ASL) bound to S. aureus TadA (PDB code: 2b3j) (17). The ASL is shown as an orange ribbon and the catalytic zinc as a light orange sphere. The non-hydrolysable adenine analog, nebularine, at the wobble position is shown as sticks and labeled (Neb34). The αC-helix of TadA is required for ASL binding. (B) Model based on our structures and the structure shown in (A) of an ASL binding to mouse ADAT in the ADAT3 zinc-binding pocket. The shorter α8 helix and the position of the ADAT3 C-terminus are incompatible with the ASL binding. (C) Same as in (B) but with the ASL binding in the ADAT2 active site. The ASL and the wobble adenosine could be recognized, provided minor conformational rearrangements of the ADAT complex. ADAT2 long α7 helix could participate in ASL binding. (D) Close-up from (A) of the interaction of wobble Nebularine within TadA. The zinc and catalytic water are shown as light orange and red spheres, respectively. (E) Close-up from (B) of the putative interaction of wobble Nebularine within ADAT3 zinc-binding pocket. Binding is incompatible with the position of ADAT3 C-terminus, notably aspartate 348 and C-terminal proline 349. (F) Close-up from (C) of the putative interaction of wobble Nebularine within ADAT2 active site. TadA and ADAT2 share the same recognition determinants of the wobble base.
Figure 5.
Figure 5.
ADAT3-N harbours a ferredoxin-like domain (FLD) found in other tRNA-modifying enzymes. (A) Superposition of mammalian ADAT3-N, archeal Trm5-D1 (purple) and ThiI-FLD (ferredoxin-like domain; orange). All three domains adopt a FLD fold but show specific structural features. Specifically, ADAT3-N contains an additional structural subdomain (aquamarine) tightly bound to its FLD (cyan). This additional subdomain is mostly responsible for the loose interactions made between ADAT3-N and ADAT catalytic domain. (B) Cognate tRNA bound at the surface of Trm5. The electrostatic potential is displayed (blue, positively charged; red, negatively charged) at the surface of Trm5, showing that the tRNA interacts with the positive (blue) electrostatic patches in Trm5-FLD. (C) Same as in (b) for the ThiI/tRNA complex. (D andE) Electrostatic potential at the surface of the ADAT complex for both structures obtained, showing positive electrostatic patches that could interact with an incoming tRNA. For figures in (B–E), the FLDs are displayed as ribbons in the same orientation, after their superposition, showing that each FLD should interact with different surfaces with tRNAs. (F) Structure of archeal Trm5 bound to a cognate tRNA (PDB code: 2zzm). The electrostatic potential is represented at the surface of the Trm5 protein and the tRNA is shown as orange ribbon. The active site of Trm5 is not seen, being in the back of the enzyme. (G) Structure of the prokaryotic TadA/ASL complex (PDB code: 2b3j) with the same features as in (F). The orientation of the ASL is identical as in (F). The ASL makes a limited number of contacts with TadA, only nucleosides 32 to 38 interacting with the protein. (H) Model of an ADAT/tRNA complex based on the structures shown in (F and G), keeping the position of the ASL as in (F) and (G), and rotating slightly the position of the ADAT3-N domain from its observed position in the high resolution ADAT structure. The ASL might undergo conformational changes upon ADAT binding, as observed in the TadA/ASL complex (not included in the current model). The model shows that the ADAT3-N domain could participate to tRNA binding by interacting, notably through its FLD, with parts of the ASL stem, the D-arm and possibly, but to a lesser extent, with the variable arm of the incoming tRNAs. Basic residues changed to glutamates in ADAT3-acidic1 and ADAT3-acidic2 mutants are coloured white.
Figure 6.
Figure 6.
Deamination activity of ADAT, ADAT2, ADAT3 and TadA and various ADAT mutants. (A andB) The percentages of deamination activity of ADAT(ADAT2/ADAT3), ADAT2, ADAT3 and TadA and of various ADAT mutants on four different tRNAs (tRNAVal(AAC), tRNAArg(ACG) and tRNAAla(AGC), and the pseudo-cognate mutant tRNAGly(ACC)) are shown as circles decreasing in size for a decreasing deamination activity. The threshold for activity has been set at 5%. The full data for all complexes is provided in Supplementary Table S3. Different tRNA:enzyme ratios (1:3, 1:0.3, 1:0.03) were used for the measurements. In (A) only the 1:3 ratio is shown for all complexes, while in (B) the three ratios are shown for a few complexes that did not appear to lose activity at high enzyme concentration but turn out to be less active when the ratio is decreasing. Only ADAT and TadA were shown to have a robust deamination activity. All enzyme mutants showed perturbation of the deamination activity, albeit to a lesser extent in the case of tRNAAla(AGC) that appears to bind slightly differently to ADAT, but still requires ADAT3-N. TadA only deaminates tRNAArg(ACG), and ADAT can partially deaminate a pseudo-cognate mutant tRNAGly(ACC), where the wobble cytosine has been replaced by an adenosine.
Figure 7.
Figure 7.
Position of V128 within ADAT3-N and effect of its mutation in leucine (V128L). (A) Ribbon structure of ADAT3-N with the residues forming its central V128 (purple) hydrophobic core shown as spheres. The residues are coloured according to the subdomain they belong to: FLD (cyan) and additional structural subdomain (aquamarine). (B) Same as (A) for the V128L mutant. The same colour code is used with slightly darker colours. (C) Superposition of the ADAT3-N WT and V128L domains shown as Cα ribbons. Slight changes are observed in the main chain position (movements of 0.3–0.9 Å) of residues from the V128 hydrophobic core and of neighbouring residues upon the V128L mutation. These movements propagate notably within the ADAT3-N specific structural subdomain but much less in the FLD. A V128M mutation is expected to exacerbate these changes.
Figure 8.
Figure 8.
In vivo effects of mutant ADAT complexes. (A) Scheme representing the in utero electroporation procedure used to follow the migration of GFP+ neurons expressing WT and mutant ADAT2 and/or ADAT3. (B) Western blot of extract from N2A cells transfected with an empty vector or the indicated ADAT2 (upper panel) or HA-tagged ADAT3 (lower panel) constructs showing similar expression of WT and catalytically inactive ADAT2 (E73A) proteins and of WT, mutant and truncated ADAT3 proteins. α-Tubulin was used as a loading control. (C) Coronal sections of E18.5 cortices, 4 days after IUE with NeuroD-IRES-GFP and the indicated ADAT2 and ADAT3 constructs, showing impaired distribution of GFP-positive electroporated cells (green) in all conditions tested. Nuclei are stained with DAPI (blue); scale bar: 100 μm. (D andE) Histograms (means ± S.E.M) showing the distribution of GFP-positive neurons in different regions (Up CP, Upper cortical plate; Lo CP, Lower cortical plate; IZ, intermediate zone; SVZ, subventricular zone) for all conditions as indicated. Significance was calculated by two-way ANOVA (Bonferroni’s multiple comparisons test). Number of embryos analysed: empty vector and A2-WT/A3-WT, n = 8; A2-WT/A3-V128L and A2-WT/A3-V128I, n = 6; A2-WT/A3-V128M, n = 5; A2-WT/A3-Δloop, A2-E73A/A3-WT and A2-E73A/A3-ΔC, n = 8; A2-WT/A3-ΔN, n = 9; ns, non-significant; *P < 0.05; **P < 0.01; ***P < 0.001; ****P < 0.0001. A2 and A3 stand for ADAT2 and ADAT3, respectively.
Figure 9.
Figure 9.
Proposed TadA and ADAT cognate tRNAs deamination mechanisms. The modes of tRNA binding and recognition for prokaryotic TadA (left) and ADAT (right) enzymes are indicated. While TadA can specifically and directly bind and recognize within its active site its cognate tRNAArg(ACG) from the pool of cellular RNAs using the specific anticodon-stem-loop conformation and the anticodon loop base composition of tRNAArg(ACG), ADAT appears to have evolved a two-step mechanism. ADAT first binds to tRNAs irrespective of their anticodon composition through its N-terminal domain, most likely upon recognizing the tRNAs specific three-dimensional structure. Upon rotation of ADAT3 N-terminal domain, the anticodon of the tRNAs is presented to the ADAT2 active site that recognizes its cognate tRNAs in a process requiring residues from both ADAT2 and ADAT3. Both TadA and ADAT2 deaminate the cognate tRNAs wobble adenosine into inosine using the same conserved mechanism (centre), requiring a catalytic water bound to their zinc ion as well as a proton shuttling glutamate residue.

Similar articles

Cited by

References

    1. Boccaletto P., Machnicka M.A., Purta E., Piatkowski P., Baginski B., Wirecki T.K., de Crecy-Lagard V., Ross R., Limbach P.A., Kotter A.et al. .. MODOMICS: a database of RNA modification pathways. 2017 update. Nucleic Acids Res. 2018; 46:D303–D307. - PMC - PubMed
    1. El Yacoubi B., Bailly M., de Crecy-Lagard V.. Biosynthesis and function of posttranscriptional modifications of transfer RNAs. Annu. Rev. Genet. 2012; 46:69–95. - PubMed
    1. Tuorto F., Lyko F.. Genome recoding by tRNA modifications. Open Biol. 2016; 6:160287. - PMC - PubMed
    1. Kirchner S., Ignatova Z.. Emerging roles of tRNA in adaptive translation, signalling dynamics and disease. Nat. Rev. Genet. 2015; 16:98–112. - PubMed
    1. Schimmel P. The emerging complexity of the tRNA world: mammalian tRNAs beyond protein synthesis. Nat. Rev. Mol. Cell Biol. 2018; 19:45–58. - PubMed

Publication types