Using rare codon pairs to attenuate viruses.

A friend of mine who works at the CDC turned me on to a new technique they're working on to construct safe attenuated viruses to use as vaccines: Genome wide changes in codon pair bias.

A quick review of codons: each of the twenty amino acids is encoded by at least one 3 letter codon in RNA: for example, GCC is translated by ribosomes into an Alanine. Because there are 4 nucleotides availible, there are 4^3, or 64 possible 3 letter codons. Since only 21 (20 AA + a stop codon) are actually required, many amino acids are encoded by multiple sets of codons (eg. UCU, UCA, UCG, and UCC all encode Serine) The human genome, however, preferentially uses certian codons (GCC is four times as prevelent as GCG, though both encode Ala), and pairs of codons (GCAGAG and GCCGAA encode the same two amino acids, but the former is 7 times as prevelent in the human genome). As it turns out, our cellular machinery is tuned to translate common codon pairs more effecitvely than rare codon pairs.

A group of researchers has developed a way to build long DNA sequences, and with those sequences, generate poliovirus de novo, without any template. Using this technique, they were able to develop equivalent versions of poliovirus with identical protiens, but encoded with a varying degree of rare codons and rare codon pairs.

Because our cellular machinery translates rare codons poorly, a very large number of rare codons renders the virus totally ineffective at infection - creating no immune response. By including some rare codons, but some standard codons, the researchers were able to create a virus that is effective enough to trigger an immune response (and subsequent immunity - remember, the virus has identical protiens), but weak enough to prevent a true viral infection - an effectively attenuated virus.

As you may remember, some current live, attenuated vaccines have the possibility of reverting to virulence: for example, the oral polio vaccine, which has only 5 attenuating mutations, can, on rare occasions, revert to a virulent form - which is why it is no longer available on the US market. A rare codon virus, on the other hand, has hundreds of mutations, each contributing, in a small way, towards attenuating the virus; this leaves no chance of the virus reverting to virulence. This technique also allows for the systematic weakening of a virus - potentially allowing vaccine developers to precisely adjust the strength of an attenuated virus to maximize immunogenic effectiveness.

Reference:
J. Robert Coleman, Dimitris Papamichail, Steven Skiena, Bruce Futcher, Eckard Wimmer, and Steffen Mueller
Science 27 June 2008 320: 1784-1787 [DOI: 10.1126/science.1155761] (in Reports)