The unidirectional translocation of thousands of distinct proteins across specific intracellular membranes is mediated by "signal" sequences. On average, a signal sequence consists of a stretch of ~15 amino acid residues that is either a transient or permanent part of the protein to be translocated. The signal sequence functions essentially as a ligand. Each signal sequence is membrane specific and is decoded by a complex machinery that is restricted in its location to one particular cellular membrane.

Two distinct mechanisms of translocation have so far been discovered. In one mechanism translocation proceeds through protein conducting channels. The diameter of the aqueous center of these protein conducting channels is limited (~2 nm) so that passage of a protein can proceed only in its unfolded configuration. A number of polypeptide binding proteins assist in keeping the protein to be translocated in an unfolded configuration. Protein conducting channels have recently been detected electrosphysiologically in the endoplasmic reticulum and the prokaryotic plasma membrane. These channels were found to be gated open by the signal sequence. The channel closes after translocation of the chain is completed. In addition to opening and closing across the membrane, the channel must also be able to open and close in a second dimension, namely to the lipid bilayer. This is necessary to permit integration of proteins into membranes. A protein to be integrated into the membrane uses a signal sequence to open the channel. Translocation proceeds until a "stop transfer" sequence of the translocating polypeptide chain interacts with the channel to open it laterally to the lipid bilayer. As a result, the segment of the chain that is located in the channel would be displaced into the bilayer with the channel simultaneously closing in both dimensions. Similar protein conducting channels are likely to exist in the outer as well as the inner membrane of chloroplasts and mitochondria, in the thylakoid membrane of chloroplasts, and in the peroxisomal membrane. The great challenge ahead is to isolate and to characterize these protein conducting channels.

The mechanism of protein translocation across the nuclear pore complex (NPC) is distinct from that of translocation across the above-described protein conducting channels. NPCs are huge organelles (estimated molecular mass: 125 million daltons) that are suspended in 100-nm wide circular openings in the nuclear envelope. An NPC can open to 25 nm in diameter. For passage across, proteins do not need to be kept in an unfolded configuration. Unlike protein conducting channels, NPCs are unable to integrate proteins into the lipid bilayer. Furthermore, transport across the NPC is bidirectional. Also, transport is not limited to proteins but includes ribonucleoproteins (RNPs). An in vitro system for signal sequence-mediated protein uptake into the nucleus has been used to isolate and to characterize cytosolic factors that are required for import. Similar in vitro RNP export systems are being developed to study export of RNPs. NPCs have been purified in quantity from yeast. An estimated 100 or so proteins make up the NPC. The challenge ahead here is to understand the structure and function of these NPC proteins and of NPC as a whole.