Nonsense Suppression (Molecular Biology)

Nonsense mutations create stop codons within the coding regions of genes. Nonsense suppression occurs when mutations in the translational apparatus allow the decoding of nonsense codons as sense codons. Most nonsense suppressor mutations alter tRNA anticodons, making them cognate the stop codon; however, a handful of suppressor mutations have also altered release factors or other translational components. Suppression requires that the amino acid inserted is compatible with protein function. Furthermore, because the nonsense mutation is decoded as "stop" at least some of the time, suppression allows only a fraction of the normal level of polypeptide synthesis. Despite these limitations nonsense suppressors have been remarkably useful research tools. Other forms of mutational suppression are described in Genetic Suppression.

There are three termination codons in the universal genetic code, UAG, UAA, and UGA. Historically, UAG nonsense mutations are called "amber," named after the mother of a pioneering worker who first isolated such mutants in phage T4 (Harris Bernstein — Bernstein means "Amber" in German. Interestingly, the molecular nature of amber mutations was not known at the time, see Ref. 1). After their discoveries, the other nonsense codons were also named according to a "color code," with UAA as "ochre" and UGA as "opal." tRNA suppressors have been isolated that can suppress each color, and because of U:G wobble decoding ochre suppressors will also suppress amber mutations.


Nonsense mutations are a type of conditional mutation, because the mutant phenotype is dependent on the absence of a suitable suppressor. This property has been used extensively in genetic studies of phage, bacteria, yeast, and the multicellular eukaryote Caenorhabditis elegans (2). For example, a large number of phage mutants have been isolated based on the ability to form plaques only on hosts that express nonsense suppressors. Such mutants are referred to as "suppressor-sensitive" and are commonly abbreviated as "sus" in the literature. sus mutations can serve as markers for genetic mapping and other routine genetic manipulations. In addition, studies of the molecular phenotypes during the nonproductive condition have helped illuminate the functions of the genes that carry nonsense mutations.

Nonsense suppression has also been extremely useful for studies of translational mechanisms (3). Suppressor tRNA genes are easy to manipulate, and suppression efficiencies are readily quantified; together, these properties make suppression the assay of choice for probing translational mechanisms. Suppressor tRNAs must compete with the peptide release factors that normally decode nonsense codons. Suppressors that are good competitors allow for efficient suppression of the nonsense mutation. Thus, the level of suppression can be used as a measure of the translational efficiency of tRNAs. Surprisingly, perhaps, tRNA suppressors can vary dramatically in their translational efficiency. Yarus observed that suppression efficiency was correlated with the tRNA nucleotides near the anticodon and proposed that decoding efficiency was a property of an "extended anticodon" in which the nucleotides near the anticodon contribute to its ability to translate codons (4). This hypothesis was confirmed by making measuring the effects mutations that saturate the anticodon arm of an amber suppressor. It was found that the native sequence was optimal for suppression, and that anticodon arm mutations reduce suppression efficiency by various degrees (5). Nonsense suppressors have also been used to show that nucleoside modifications near the anticodon increase translational efficiency. Mechanisms are not always clear, but a common theme is that bulky modification 3′ to the anticodon increases translational efficiency, as if increased base stacking stabilizes anticodon:codon complexes (6, 7).

Studies of suppressors have also shown that nucleotides outside of the anticodon region can affect translational efficiency (8-11). Mutational analyses show that the central or hinge region of the tRNA is important for translation, but probably not by directly affecting anticodon:codon pairing. Instead, this region may participate in a conformational change in the tRNA that must occur during ribosomal acceptance of the aminoacyl-tRNA. Mutations that interfere with this change affect the likelihood that the tRNA will be selected, regardless of whether the anticodon is perfectly matched to the codon (12). Certain mutations in this region allow, for example, reading with a first-position U:G pair or a third-position C:A pair. It is thought the hinge mutations facilitate conformation changes leading to ribosomal acceptance despite the mismatched base pairs.

McClain and co-workers have used nonsense suppressor tRNAs to demonstrate changes in the aminoacylation specificities of several tRNAs and thus map the determinants for tRNA amino acid "identity" (13). They engineered an amber termination site near the 5′ end of the gene for the easily isolable dihydrofolate reductase enzyme. They then mutagenized tRNA suppressors and determined which amino acids the variants insert by sequencing the amino termini of isolated enzymes. They found that a small number of nucleotides are mostly responsible for defining tRNA identity. The use of suppressor tRNAs was critical for these studies because there is ambiguity about which tRNA actually decodes the amber site. Thus, a change in the amino acid inserted can be unambiguously attributed to the changes made in the amber suppressor tRNA.

Nonsense suppressors are also used to show that codon translation is affected by neighboring nucleotides (codon context). A large number of studies show that nonsense codons are more readily suppressed if the 3′ neighbor nucleotide is a purine. This context effect has at least two molecular sources: the anticodon:codon complex is stabilized by base stacking with the 3′ purine (14, 15), and the release factors are highly dependent on the neighbor for termination with 3′ U being optimal (16, 17).

Nonsense suppressors will also be used for protein engineering. In one approach, Abelson and collaborators have constructed an extensive set of tRNA suppressors that can be used to insert a wide range of amino acids at amber mutations (18). These tRNAs may allow systematic tests of the effects of various amino acids on protein structure and function. Proteins engineered to contain amino acids with fluorescent or especially reactive side chains would be of great use in studies of protein structure and function and for biotechnological applications. Recently, it has been demonstrated that certain nonsense suppressors can be aminoacylated with novel amino acids, and that modified tRNAs can direct the incorporation into proteins that have specific nonsense mutations within their coding sequences (19). Currently, only a few novel amino acids may be used in this way. But, hopefully, a large variety of specific labels will become available.

Next post:

Previous post: