Kozak's scanning hypothesis for initiation of translation in eukaryotes

Kozak's scanning hypothesis explaining synthesis of mRNA in eukaryotes
Fig. 34.6. Kozak's scanning hypothesis explaining synthesis of mRNA in eukaryotes (for details see text; note the role of false start and stop signals preceding the coding region).
In 1983, Marilyn Kozak proposed a hypothesis for initiation of translation by eukaryotic ribosome. According to this hypothesis, a eukaryotic ribosome (40S smaller subunit with its associated met-tRNA) moves down the mRNA from 5' end, until it encounters the first AUG, so that 60S subunit joins and the translation begins. The 80S ribosome, after reaching termination, releases protein and dissociates in two subunits.

It has been shown that the coding region of mRNA may be preceded by upto four start and stop signals. The first AUG codon in 90% cases occurs in the form of consensus sequence PuNNAUGG (Pu = purine; N = any nucleotide), situated between 50 and 100 residues from the 5' end of the message. In 5% cases one or more AUG codons occur upstream to the AUG codon which makes starting codon of the coding region.
These extra AUG codons are also read as envisaged in Kozak's hypothesis. Same results are obtained if an AUG codon is deliberately inserted upstream of coding region. Such AUG codons make false starts, so that if the reading frame is the same, an extended protein is obtained (Fig. 34.6). If there is a termination signal between false and real starts, a small peptide is released, but 40S subunit of ribosome does not leave mRNA. Reading of the real message starts, if real AUG is encountered within a reasonable distance. Surprisingly, it was also found that with the increase of distance between the false AUG and the real AUG, the efficiency of reinitiation increased. This was explained by assuming that after the dissociation of 60S subunit, the 40S subunit would lack met-tRNAfmet and that increasing the distance between the two start codons would provide more time for the 40S submit to acquire met-tRNAfmet to become competent to reinitiate.
Kozak's scanning hypothesis explaining synthesis of mRNA in eukaryotes
Fig. 34.6. Kozak's scanning hypothesis explaining synthesis of mRNA in eukaryotes (for details see text; note the role of false start and stop signals preceding the coding region).
It has been shown that an AUG codon can be missed if it poorly matches Kozak's consensus sequence (PuNNAUGG). It has been shown in P and C proteins of Sendi virus, that in a 2000 bp long mRNA, two AUGs give rise to different proteins formed by alternative starts, though in the same reading frame.
A third AUG in different reading frame gives rise to an entirely different protein. In adenovirus also a similar situation occurs and different AUG codons are used in living cells and cell free systems. It is, however, not known, how the choice of different AUGs is exercised and regulated. They definitely represent economical use of viral genome just like the overlapping genes discussed elsewhere in the book. (Consult Organization of Genetic Material 3.  Split Genes, Overlapping Genes and Pseudogenes for overlapping genes).