The acceptor stem is the site of attachment of amino acids to transfer RNA (tRNA). It is formed by 7 base pairs and has 4 single-stranded nucleotides. Nucleotides 1 to 7 from the 5′ end of the tRNA base pair with nucleotides 72-66, respectively, from the 3′ end of the molecule. Whereas the 5′ end of the RNA has a monophosphate group, the 3′ end contains a 3′-hydroxyl group, which is the site of esterification to the amino acid. The four 3′-end single-stranded nucleotides include residue 73, the discriminator base and the well-conserved nucleotides C74, C75, A76, the 3′-CCA sequence. In the three-dimensional structures of tRNA, the acceptor stem is stacked on the T arm, forming the acceptor "branch" of the L-shaped RNA fold (see Fig. 1 of Transfer RNA). The acceptor branch usually forms a regular A-type double helix, with the four single-stranded nucleotides extending in a regular helical continuity. In tRNA-like domains of some viruses, the acceptor stem is formed by a single RNA chain that folds into a helix due to the presence of a pseudoknot (1).
Specific aminoacylation of tRNA by their cognate aminoacyl-tRNA synthetases is dependent on the presence of a series of identity elements. Limited in number, these elements are preferentially located in the anticodon loop and in the acceptor stem (2-4). Residue 73, next to the CCA end, is called the "discriminator" base. The hypothesis that it contributes to discrimination of tRNA by cognate aminoacyl tRNA synthetases (5) has been largely confirmed. Residue 73 contributes strongly to specific aminoacylation of 17 different Escherichia coli tRNAs (6). It is the element making the major thermodynamic contribution toward aminoacylation of E. coli tRNACys, E. coli tRNAHis, yeast tRNAHis and E. coli tRNALeu. Replacement of A73 to G73 in human tRNASer converts it to a tRNA for isoleucine (7). The crystallographic structures of two tRNA-cognate synthetase complexes suggests two possible mechanisms by which the discriminator base contributes to aminoacylation specificity. One is a direct mechanism, involving direct hydrogen interactions with the synthetase. The other is an indirect mechanism that confers a conformational change to the acceptor end of the tRNA to facilitate aminoacylation. In the complex of yeast tRNAAsp/aspartyl-tRNA synthetase, nucleotide G73 fits into the active site, forming hydrogen bond interactions with side chains of the synthetase. It does not cause a conformational change in the acceptor stem (8). Alternatively, the 2-amino group of G73 of E. coli tRNAGln hydrogen bonds with the phosphate oxygen of the previous nucleotide, folding the backbone of G73 back toward the 3′ end of the tRNA. The formation of this fold-back hairpin enables the synthetase to open the first base pair of the acceptor stem and reach the second and third base pairs for specific interactions (9). In both complexes, additional contacts exist between the enzyme and the acceptor stems of the tRNA. The aspartyl-tRNA synthetase interacts with the tRNAAsp acceptor stem via the major groove of the RNA helix, whereas glutaminyl-tRNA synthetase interacts via the minor groove.
Acceptor stem identity-element nucleotides, as well as additional structural features, are important for synthetase recognition. Alanyl-tRNA synthetase is sensitive to both the exocyclic NH2 group of G3 and local conformation of the helix due to structural characteristics of the G-U wobble pair (1013). Alanine acceptance by alanyl-tRNA is modulated by additional signals within the acceptor stem, namely, A73, G1-C72, G2-C71 and G4-C69. The 5′ end of histidine-specific tRNA has an additional nucleotide, residue number -1, which is a guanosine. This nucleotide, opposite the discriminator base, does form a base pair with the discriminator in many, but not all, histidyl-tRNA. This -1 guanosine is the major histidine identity element. Its influence is complemented by base pairs U2-A71 and G3-C70. Base pairs 1-72, 2-71, 3-70, and/or 4-69 contribute to aminoacylation in several cases. Glutamine identity requires a weak 1-72 base pair, G2-C71, G3-C70, in addition to signals elsewhere in the tRNA. Glycine identity is dependent on C2-G71, G3-C70 sequences, and serine identity involves G2-C71, among other signals. Methionine identity is based on the presence of A73, G2-C71, C3-G70, in addition to the CAU anticodon.
At the three-dimensional level, tRNA molecules fold into a two-domain L-shaped structure with the amino acid acceptor terminus and the anticodon at opposite ends. Minihelices (small RNA molecules containing only part of a tRNA) that mimic the amino acid acceptor domain, specifically, the acceptor stem stacked on top of the TY stem, have been shown to be useful tools for determining the contribution of the acceptor stem to tRNA function (14, 15). Minihelices containing the identity elements from a tRNA are efficient substrates for its cognate aminoacyl-tRNA synthetases. Thus, alanyl-tRNA synthetase, glycyl-tRNA synthetase, and histidyl-tRNA synthetase aminoacylate minihelices efficiently. Minihelices containing partial identity sets of six additional tRNAs are specific substrates for their cognate synthetases. Since identity elements are located very close to the CCA end, minihelices may be reduced in size to microhelices, consisting only of the amino acceptor stem closed by a loop and remain active. The smallest substrate for an aminoacyl-tRNA synthetase is derived from yeast tRNAAsp and consists of only 14 nucleotides. There are three base pairs closed by a tetraloop, plus the discriminator base and the CCA sequence (16). Double-stranded RNA duplexes and RNA/DNA heteroduplexes may also be aminoacylated (15).
The aminoacylation efficiency of minihelices is generally reduced significantly compared to that of the corresponding full-length tRNA. The amino acid charging is, however, very specific and depends on just a few nucleotides. These minimal RNA substrates are devoid of the anticodons that read the genetic code and provide the link between amino acid and codon. Since the relationship between the sequences and structures of the minisubstrates and the specific amino acids is maintained, there appears to be an operational RNA code for primitive aminoacylation. This operational code may be the primitive code from which the contemporary genetic code evolved (17-20)
Acceptor stem properties are important for tRNA recognition by proteins other than synthetases. Aminoacylated initiator tRNA in prokaryotes is a substrate for a methionyl-tRNA transformylase that converts methionyl-tRNAMet to formylmethionine tRNAMet. Recognition of initiator tRNA by the formylase depends on the presence of methionine. Sequence and/or structural elements in the tRNA that are important for formylation by methionyl-tRNA transformylase are clustered at the end of the acceptor stem (21). The key determinants appear to be a mismatch or a weak base pair between nucleotides 1 and 72, a G-C base pair between nucleotides 2 and 71, and a C-G or, less preferably, a G-C base-pair between nucleotides 3 and 70. Mutations at G4-C69 also affect formylation kinetics slightly. In addition to the positive elements A73, G2-C71, C3-G70, and G4-C69, the occurrence of a G-C or a C-G base pair between positions 1 and 72 acts as a major negative determinant for the formylase. Formylation is a prerequisite for interaction with initiation factor IF2, which delivers the initiator tRNA to the P site of the ribosome. Special structural features within the acceptor stem of eubacterial initiator tRNA contribute to their discrimination from elongator tRNA. They lack a Watson-Crick base pair between nucleotides 1 and 72 at the end of the acceptor stem. Eukaryotic initiator tRNA almost always has an A1-U72 pair, a feature not found in eukaryotic elongator tRNA.
A high-resolution X-ray crystallography structure of the ternary complex, formed by aminoacylated yeast tRNAPhe, elongation factor Tu from Thermus aquaticus, and an analog of GTP, revealed numerous contacts between the protein and the acceptor stem of the tRNA (22). These contacts include specific interactions with residue A76, with phosphates 74, 75, 67, 64, 3, and with riboses 2, 3, 63, 64, 65 along the acceptor stem and the T stem.
The stability of the 1-72 base pair also governs the degree of sensitivity of a peptidyl-tRNA to peptidyl-tRNA hydrolase. This enzyme converts the peptidyl-tRNA of N-acetylaminoacyl-tRNA into free tRNA plus peptides or N-acetyl amino acids. It is believed to play a role in the translational apparatus through the recycling of free tRNA from immature peptidyl-tRNA created by abortive protein synthesis (23).
The acceptor stem also contains specific information required during synthesis of the tRNA. Ribonuclease P, an enzyme that removes extra nucleotides from the 5′ end of tRNA during tRNA biosynthesis, recognizes the -74CCA76- sequence in addition to nucleotides within the TY loop. Moreover, this enzyme locates its cleavage site by "measuring" the length of the helix formed by the amino acid acceptor stem fused to the TY stem. Processing at the 3′ end of the tRNA, as well as the addition and repair of the conserved CCA sequence by ATP(CTP)-tRNA nucleotidyltransferase, requires information within the acceptor stem. This enzyme interacts with tRNA at the corner of the structure, where the T and D loops interact, and extends that interaction across the aminoacyl stem to the 3′ end (24, 25).
