A stop codon is a three-nucleotide sequence in messenger RNA (mRNA) that normally signals the termination of translation, the process by which cells synthesize proteins. In the standard genetic code, the three stop codons are UAA, UAG, and UGA, written in the 5′-to-3′ direction. Unlike codons specifying an amino acid, they ordinarily recruit protein release factors rather than amino acid–carrying transfer RNAs. Their interpretation can change in particular organisms or sequence contexts.
Sequences and terminology
The standard genetic code contains 64 possible codons: 61 specify amino acids, and three serve as termination signals.
| mRNA codon | Corresponding coding-strand DNA sequence | Traditional name |
|---|---|---|
| UAA | TAA | Ochre |
| UAG | TAG | Amber |
| UGA | TGA | Opal, also called umber |
The DNA sequences in the table refer to the coding, or non-template, strand written 5′ to 3′. They match the mRNA sequence except that DNA contains thymine (T) where RNA contains uracil (U).
The term nonsense codon arose from early genetic studies distinguishing triplets that did not specify an amino acid from “sense” codons. It does not mean that a normal termination signal is biologically meaningless.
A triplet acts as a stop codon only when it is interpreted in the relevant reading frame. The same sequence can occur across codon boundaries or outside a translated region without terminating protein synthesis.
Mechanism of translation termination
During translation, the ribosome reads mRNA codons sequentially. Amino acid–carrying transfer RNAs (tRNAs) normally enter its A site, where codon recognition enables addition of another amino acid to the growing polypeptide.
When a stop codon reaches the A site, a translation release factor recognizes it. The factor promotes hydrolysis of the ester bond connecting the completed polypeptide to the tRNA in the P site. This releases the protein without adding an amino acid corresponding to the stop codon. Additional factors then help disassemble and recycle the translation machinery.
Stop-codon recognition differs among major groups of organisms:
- In bacteria, release factor RF1 recognizes UAA and UAG, while RF2 recognizes UAA and UGA. RF3 assists termination-factor turnover in many bacteria.
- In eukaryotes, eRF1 recognizes all three standard stop codons, with assistance from the GTPase eRF3.
- In archaea, aRF1 generally recognizes all three standard stop codons.
The stop codon therefore marks the end of the translated coding sequence, not necessarily the end of the mRNA. Eukaryotic mRNAs typically contain a 3′ untranslated region downstream of it. Termination of translation is also distinct from termination of transcription, which ends RNA synthesis.
Sequence context and termination efficiency
Termination is not determined by the three nucleotides alone. Nearby bases, the identity of the final amino acids in the nascent protein, RNA structure, and the availability of release factors and tRNAs can influence how efficiently translation stops.
Occasionally, a near-cognate tRNA is accepted instead of a release factor, allowing translation to continue past the stop codon. This is called stop-codon readthrough. Low-frequency readthrough can occur as a translation error, whereas programmed readthrough is a regulated feature of some genes and viral genomes. It produces an extended protein containing additional amino acids encoded downstream of the usual termination site.
Thus, a stop codon is not an absolute physical barrier. Its effect depends on competition among decoding components and on the surrounding molecular context.
Exceptions and alternative genetic codes
The designation of UAA, UAG, and UGA as stops applies to the standard genetic code, not universally to every biological translation system.
Reassignment in alternative codes
In some organisms and organelles, one or more standard stop codons have been reassigned to amino acids. For example, UGA specifies tryptophan in many mitochondrial genetic codes. In the nuclear codes of some ciliates, UAA and UAG specify glutamine, while UGA remains a termination signal.
Such differences matter when interpreting a genome: applying the wrong translation table can incorrectly split genes or predict truncated proteins.
Selenocysteine and pyrrolysine
UGA can direct incorporation of selenocysteine through specialized translation machinery. This generally requires a dedicated tRNA, accessory proteins, and an mRNA signal known as a selenocysteine insertion sequence, or SECIS element. Its position and recognition mechanisms differ between biological groups.
UAG can specify pyrrolysine in certain archaea and bacteria through a dedicated tRNA and aminoacyl-tRNA synthetase system.
These cases involve specialized decoding rather than ordinary recognition of the triplet as a termination signal. They also differ from accidental readthrough.
Mutations and RNA surveillance
A mutation that converts an amino acid–coding triplet into a stop codon is called a nonsense mutation. It creates a premature termination codon and may produce a shortened protein. Insertions or deletions that shift the reading frame can also generate premature stops.
The outcome depends on the position of the stop and the function of the affected sequence. A truncated protein may lose essential regions, retain partial activity, or have altered properties. In eukaryotes, many transcripts containing premature termination codons are targeted by nonsense-mediated mRNA decay (NMD), an RNA-surveillance pathway.
NMD does not simply destroy every mRNA with an early stop. Recognition depends on transcript architecture and the context of termination. In mammals, a commonly used predictor is a stop situated more than approximately 50–55 nucleotides upstream of the last exon–exon junction, but this is a guideline with exceptions rather than a universal rule.
Conversely, a mutation that removes the normal stop codon can cause translation to continue into a downstream region. If no suitable termination codon remains, the ribosome may stall and engage rescue and quality-control pathways.
Discovery and research applications
Stop codons were characterized during the deciphering of the genetic code in the 1960s. Studies of bacteriophage mutants and suppressor mutations helped establish that particular triplets could terminate protein synthesis. The traditional names amber, ochre, and opal originated in this period and remain common in molecular genetics.
In biotechnology, stop codons are used to define protein endpoints in expression constructs. A stop codon is usually retained when a protein should end at its native boundary, but omitted when translation must continue into a downstream fusion partner or purification tag.
Engineered suppressor tRNAs can also decode selected stop codons, especially UAG, to incorporate nonstandard amino acids. These systems enable site-specific chemical modification of proteins. Their performance depends on competition with normal termination, decoding specificity, and unintended readthrough elsewhere in the cell.