Sandbox 201

Structural domains

T4 Rnl1 is a 374-amino acid polypeptide that consists of two structural domains.

The N-terminal domain is responsible for the adenylyltransferase activity of the enzyme. It includes an a helix (a1), followed by antiparallel ß-sheets (ß1–ß4). It is the central ensemble of ß strands and loops that forms the nucleotide-binding pocket.

The pocket is lined by six peptide motifs (I, Ia, III, IIIa, IV, and V). These motifs characterize the covalent nucleotidyltransferase super-family that includes DNA ligases, RNA ligases, and mRNA capping enzymes.

The N-terminal polypeptide from from 1 to 254 amino acids suffices for the ligase–adenylylation reaction of T4 Rnl1.

The C-terminal domain is an all-helical domain, specific to the Rnl1. This all-helical domain is not found in C-terminal domains of DNA ligases and RNA capping enzymes.

The C domain confers the specificity for tRNA repair. The physiological substrate for Rnl1 is a tRNA containing a single break in the anticodon loop. :Deleting the C domain abolishes the preference of the enzyme for this broken tRNA substrate, but does not abolish adenylyltransferase activity.

A conserved Arg318–Lys319 dipeptide is a possible candidate that determines the tRNA specificity.

Ligands binding sites

The N- and C-terminal domains are both able to form interactions with ATP and ATP analogues. But the ß-strands of the core region in the N-terminal domain contain most of the residues involved in binding ATP. , (motif I), (motif Ia), (motif V), and (motif V) interact with the phosphate groups of ATP and ATP analogues.

Lys99 is the site of adenylation in Rnl1, ^[1] but in this structure this residue seems to be situated at a distance incompatible with covalent interaction with the phosphate of ATP (more than 3 Å).^[2] That could suggest that the formation of a covalent bond needs some conformational changes. But we do not know if a conformational change has to occur to allow the covalent bond formation, or if the formation of this bond leads to a conformational change.

• Divalent cation binding sites

Metal ions are essential for the nucleotidyltransferase catalysis by all T4 DNA and T4 RNA ligases, which use the same two-metal ions mechanism.

The enzyme binds Mg²⁺. The true substrate in the adenylation reaction is the ATP-Mg²⁺ complex, ^[3] but nucleotidyltransferase enzymes cannot bind ATP-Mg₂ directly. They bind ATP-Mg first, then a second Mg²⁺ ion. Each oh these cations with one phosphoryl oxygen from AMPcPP, three water molecules and two residues (Gly269 and Asp272), which both belong to the C-terminal domain.

The enzyme binds Ca²⁺. are coordinated to six water molecules. They do not directly interact with the enzyme, but via water molecules interacting with Glu227, Glu159, Lys99, Glu100, and Tyr246 via hydrogen bonds. They also interact with one phosphoryl oxygen of the AMPcPP. are coordinated to four water molecules and interact with three enzyme residues (Ile211 and Asp212) via hydrogen bonds.

Calcium is very important for enzyme Rnl1 structural biology, because the enzyme crystallizes only in présence of Ca²⁺. This could be explained by the fact that interactions between Ca²⁺ and negatively charged surface of each Rnl1 allow interactions between several enzymes at crystallization interfaces.^[2]

• Anion binding sites

The enzyme binds Cl^-.

• RNA binding site

The RNA-Rnl1 complex have not be crystallized yet, because the enzyme seems to crystallize only with AMPcPP, which is incompatible with the presence of RNA in the active site. That's why we are not able to define precisely the RNA binding site. But there are some elements tending to indicate of a possible RNA biding site in the  :

- The analysis of charge distribution on the protein surface shows that the Rnl1 surface is negatively charged, apart from the C-terminal domain. The positive charges present on this domain could interact with the polyanion-like RNA backbone.

- The RNA have to be close to the ATP binding site to enable the AMP transfer from Lys99. The RNA could bind at the surface of the C-terminal domain, allowing the anticodon loop to be positionned toward the ATP binding site.

- Moreover, the C-terminal helical structure matches the tRNA structure.

The Rnl1 C-terminal domain is unique in nucleotidyltransferase family, so the architecture of this domain could allow T4 RNA ligase to bind specifically tRNA^Lys in vivo.

A chloride ion is also positionned in the active site, and could mimic the 5'-phosphate of the incoming RNA.^[4][5][2]

T4 RNA ligase is also able to bind . This molecule does not allow the nucleotidyltransferase function of the enzyme, but it seems necessary to Rnl1 cristallyzation. Each AMPcPP with ten residues (Tyr37, Arg54, Lys75, Tyr98, Lys99, Glu100, Lys119, Glu159, Lys 240 and Lys242), one Ca²⁺, one Mg²⁺ and one Cl^-. It also interacts with several other enzyme residues via hydrophobic interactions.