Snyder and Champness Molecular Genetics of Bacteria. Tina M. Henkin
are 64 (4 x 4 x 4) possible codons that can be made of 4 different nucleotides taken 3 at a time. Thus, without redundancy, there would be far too many codons for only 20 amino acids. As shown in Table 2.2, some amino acids are encoded by a single codon (e.g., tryptophan), while others use as many as six different codons (e.g., arginine).
Figure 2.32 trans-Translation by transfer-messenger RNA (tmRNA). (A) A ribosome translating an mRNA that lacks a termination codon will stall, because a release factor is unable to bind and release the translation complex. (B) tmRNA, which has features of both a tRNA and an mRNA, enters the A site. (C) The ribosome switches from translation of the mRNA to translation of the coding sequence in the tmRNA, which results in addition of a short polypeptide tag to the carboxy terminus of the nascent polypeptide. (D) The ribosome and mRNA are released, and the tmRNA-encoded tag targets the polypeptide for degradation by the Clp protease system.
Exceptions to the Code
One of the greatest scientific discoveries of the 20th century was that of the universal genetic code. Whether human, bacterium, or plant, for the most part, all organisms on Earth use the same three bases in nucleic acids to designate each of the amino acids. However, although the code is almost universal, there are exceptions to this general rule. In some situations, a codon can mean something else. We gave the example of initiation codons that encode different amino acids when internal to a gene than they do at the beginning of a gene, where they invariably encode methionine (see the text). Also, some organelles and primitive microorganisms use different codons for some amino acids. For example, in mammalian mitochrondria, UGA, which is normally a termination codon, instead designates tryptophan. Also, some protozoans use the termination codons UAA and UAG for glutamine. In these organisms, UGA is the only termination codon. Some yeasts of the genus Candida, the causative agent of thrush, ringworm, and vaginal yeast infections, recognize the codon CUG as serine instead of the standard leucine. In bacteria, the only known exceptions to the universal code involve the codon UGA, which encodes the amino acid glutamine in some bacteria of the genus Mycoplasma, which are responsible for some plant and animal diseases.
Some exceptions to the code occur only at specific sites in the mRNA. For example, UGA encodes the rare amino acid selenocysteine in some contexts. This amino acid exists at one or a very few positions in certain bacterial and eukaryotic proteins. It has its own unique aaRS, translation elongation factor (analogous to EF-Tu), and tRNA, to which the amino acid serine is added and then converted into selenocysteine. This tRNA then inserts the amino acid selenocysteine at certain UGA codons, but only at a very few unique positions in proteins and not every time a UGA appears in frame. How, then, does the tRNA distinguish between these sites and the numerous other UGAs, which usually signify the end of a polypeptide? The answer seems to be that the selenocysteine-specific EF-Tu has extra sequences that recognize the mRNA sequences around the selenocysteine-specific UGA codon, and only if the UGA codon is flanked by these particular sequences will this EF-Tu allow its tRNA to enter the ribosome. It is a mystery why the cell goes to so much trouble to insert selenocysteine in a specific site in only a very few proteins. In some instances where selenocysteine was replaced by cysteine, the mutated protein still functioned, albeit less efficiently. However, it may be required in the active center of some enzymes involved in anaerobic metabolism, and this amino acid has persisted throughout evolution, existing in organisms from bacteria to humans.
Another striking deviation from the code is found in the methanogenic archaea (archaea that produce methane). These bacteria insert the lysine analog pyrrolysine at the normal termination codon UAG. Unlike selenocysteine, which is chemically derived from serine already on its tRNA, pyrrolysine is synthesized and then loaded onto a dedicated tRNA by a dedicated aaRS. It therefore qualifies as the 22nd amino acid. Its aa-tRNA uses the normal EF-Tu and is inserted whenever the codon UAG appears within the mRNA.
Recently, efforts have been made to purposely reprogram the genetic code to allow targeted insertion of nonnatural amino acids into specific sites of individual target proteins with the goal of generating new classes of proteins with novel activities. This work relies on engineering systems to attach a new type of amino acid to a dedicated tRNA so that the tRNA will insert its amino acid only at a specific codon within the target protein. Approaches of this type will enable both specific labeling of proteins in the cell and development of new enzymatic activities.
Other exceptions violate the rule that the code is read three bases at a time until a termination codon is encountered. This happens with high-level frameshifting and readthrough of termination codons. In high-level frameshifting, the ribosome can back up one base or go forward one base before continuing translation. High-level frameshifting usually occurs where there are two cognate codons next to each other in the RNA, for example, in the sequence UUUUC, where both UUU and UUC are phenylalanine codons that are presumably recognized by the same tRNA through wobble. The ribosome with the tRNA bound can slip forward or backward by one nucleotide before it continues translating in the new reading frame, creating a frameshift. Sites at which high-level frameshifting occurs are designated “shifty sequences” and usually have common features. They often have a secondary structure, such as a pseudoknot, in the RNA (see Figure 2.2) just downstream of the frameshifted region, which causes the ribosome to pause. They also may have a sequence similar to an S-D sequence just upstream of the frameshifted site to which the ribosome then binds through its 16S rRNA, shifting the ribosome 1 nucleotide on the mRNA and causing the frameshift. Sometimes both the normal protein and the frameshifted protein, which has a different carboxyl end, can function in the cell. Examples are the E. coli DNA polymerase accessory proteins γ and τ, which are both products of the dnaX gene (see Table 1.1) but differ because of a frameshift that results in the formation of a truncated protein. Frameshifting can also allow the readthrough of termination codons to make “polyproteins,” as occurs in many retroviruses, such as human immunodeficiency virus (the acquired immune deficiency syndrome [AIDS] virus). Moreover, high-level frameshifting can play a regulatory role, for example, in the regulation of the RF2 gene in E. coli. The RF2 protein causes release of the ribosome at the termination codons UGA and UAA (see “Translation Termination” in the text). The gene for RF2 in E. coli is arranged so that its function in translation termination can be used to regulate its own synthesis through frameshifting (see chapter 11). How long the ribosome pauses at a UGA codon depends on the amount of RF2 in the cell. If there is a lot of RF2 in the cell, the pause is brief and the polypeptide is quickly released by RF2. If there is less RF2, the ribosome will pause longer, allowing time for a –1 frameshift. The RF2 protein is translated in the –1 frame, so this is the correct frame for translation of RF2, and more RF2 will be made if there is not enough for rapid termination.
In the most dramatic cases of frameshifting, the ribosome can hop over large sequences in the mRNA and then continue translating. This is known to occur in gene 60 of bacteriophage T4 and the trpR gene of E. coli. Somehow, the ribosome stops translating the mRNA at a certain codon and “hops” to the same codon further