The "+70 pause": hypothesis of a translational control of membrane protein assembly.

Sequences of 66 genes encoding bacterial or yeast membrane proteins have been examined for the respective positioning of putative transmembrane domains and translational pauses. The latter were operationally defined as clusters of at least 17 non-preferred codons along the mRNA. The putative transmembrane domains were defined as stretches of at least 17 hydrophobic amino acids in the encoded protein. For yeast non-mitochondrial membrane proteins, it was observed that clusters of non-preferred codons occur more frequently about 56 to 75 codons after a hydrophobic stretch in the encoded protein. About 40 amino acid residues are required to span the large ribosomal subunit. Such clusters were thus predicted to cause a severe slow-down in peptide elongation, just when the hydrophobic stretch fully protrudes from the ribosome. This transient slow-down of the ribosome pace has consequently been named the "+70 pause". This pause was not observed for mitochondrial or bacterial membrane proteins, which are thought to insert post-translationally in their respective membranes. Because insertion of yeast proteins in the endoplasmic reticulum membrane is generally cotranslational instead, it is possible that the "+70 pause" reflects the coupling of translation, targeting, insertion and folding in this case. The pause may, for instance, give time for productive interaction of the newly synthesized hydrophobic domain with the proper targeting/insertion machineries. Thus, it would favor entrance of the stalled protein domain into the proper pathway.