A transfer RNA (tRNA) connects an mRNA codon to an amino acid. It does this by carrying two kinds of information in one molecule:

• an anticodon, a three-nucleotide RNA sequence that can base-pair with an mRNA codon;
• an amino-acid attachment site at the tRNA's $3'$ end.

A tRNA folds into a compact three-dimensional structure that places these two functional regions apart from one another.

Charging a tRNA

Before a tRNA can participate in protein synthesis, it must be covalently linked to the appropriate amino acid. This reaction is called aminoacylation or tRNA charging.

An aminoacyl-tRNA synthetase recognizes an amino acid and the tRNA or tRNAs that should carry it. ATP-dependent chemistry activates the amino acid and forms a high-free-energy aminoacyl linkage between the amino acid and the tRNA.

The product is an aminoacyl-tRNA, often called a charged tRNA.

For example, a tRNA whose anticodon recognizes a phenylalanine codon must be charged with phenylalanine if decoding is to be accurate.

The synthetase establishes much of the code's accuracy

The ribosome checks whether a tRNA anticodon pairs appropriately with the mRNA codon. It does not independently verify that the amino acid attached to that tRNA is chemically the correct one.

Aminoacyl-tRNA synthetases therefore perform a crucial recognition step: they connect the nucleotide identity of a tRNA to the correct amino-acid identity.

Anticodons pair antiparallel to codons

If an mRNA codon is

5'-A U G-3'

a complementary anticodon can pair antiparallel as

3'-U A C-5'.

Some tRNAs can recognize more than one related codon because pairing at one codon position can tolerate limited variation, a phenomenon called wobble. This helps explain why fewer distinct tRNA species are needed than there are sense codons.

A charged tRNA is therefore an adaptor: its anticodon reads nucleotide information, while its attached amino acid supplies the corresponding building block for a growing polypeptide.