EVALUATION OF ELECTROSTATIC AND HYDROPHOBIC EFFECTS ON THE INTERACTION OF MITOCHONDRIAL SIGNAL SEQUENCES WITH PHOSPHOLIPID-BILAYERS

The information that directs a nuclear-coded protein to be imported into mitochondria resides in an N-terminal extension, called a signal sequence. The primary sequences of all known ones differ. The only common feature is their ability to theoretically form an amphiphilic, positively charged, alpha-helix. We previously showed that a short stable helical segment was required for a peptide to be functional in import [Wang, Y., & Weiner, H. (1993) J. Biol. Chem. 268, 4759-4765]. Here we investigate the interaction of three altered signal sequences with phospholipid membranes containing cardiolipin to ascertain the importance of electrostatic and hydrophobic interactions with the membrane. The three already described peptides were derivatives of the signal sequence from aldehyde dehydrogenase, which is composed of three segments, two helices separated by a linker. ANCN had the C-helix replaced by the N-helix of the signal sequence of cytochrome c oxidase subunit IV, ANCC had the C-terminal helix replaced by the C-terminal random coil of cytochrome oxidase subunit IV, and linker deleted had the linker region deleted. ANCC, which functioned poorly as a signal sequence, had a very low affinity for binding to the negatively charged membranes. In contrast, both ANCN and linker deleted showed a relatively high affinity for the membranes and were capable of functioning as a good leader sequence. It appears that linker deleted possessed a stronger hydrophobic effect with membranes while ANCN had a higher electrostatic interaction. On the basic of these studies, a model was proposed to describe the interaction of mitochondrial signal sequences with negatively charged phospholipid membranes involving electrostatic interaction for initial binding and hydrophobic interaction for insertion. If the peptide does not have the ability to form a stable helix, it will not be inserted into the membrane. Though not proven, it is tempting to postulate that there is an initial interaction between the signal sequence and the mitochondrial membrane prior to binding to the protein translocation apparatus.