PerspectiveCell Biology

The Ins and Outs of Protein Translocation

See allHide authors and affiliations

Science  05 Dec 1997:
Vol. 278, Issue 5344, pp. 1728-1729
DOI: 10.1126/science.278.5344.1728

Most cells manufacture proteins destined for export, which will be used by the organism either as structural elements or information-carrying signaling molecules. In preparation for their secretion, these proteins are first translocated as a long chain of amino acids through a pore, also made of protein, into the endoplasmic reticulum (ER), a complex network of membrane tubes inside the cell. Once inside the ER, these proteins fold into their mature state, assemble with other proteins, and move toward their final destination outside the cell.

But errors can happen, so eukaryotic cells carefully monitor the accuracy of protein folding and assembly in the ER. When unfolded proteins accumulateor proteins are misfolded, unassembled, or denatured as a result of various stressesthe cell sends a signal to the nucleus to activate a program to destroy these mistakes. For many years it was thought that misfolded or unassembled proteins were destroyed in the lumen of the ER. But surprising evidence has shown recently that misfolded proteins are actually exported out of the ER to the cytosol in order to be tagged with ubiquitin, a highly conserved 76-residue polypeptide that marks them for destruction by an organelle specialized for protein degradation, the proteasome. On page 1806 of this issue, Biederer et al. (1) describe a new component of this machinery, Cue1p, a protein that docks crucial parts of the ubiquitination apparatus at the ER membrane. Without Cue1p, cells cannot efficiently ubiquitinate and degrade misfolded proteins. Indeed, in the absence of Cue1p, misfolded proteins remain in the ER lumen, suggesting that their export is usually coupled to their ubiquitination.

When proteins are imported into the ER during their translation from messenger RNA, they pass through an aqueous channel, found in a protein structure called the translocon (2) (see the figure). During this cotranslational import, the cytoplasmic face of the aqueous pore is sealed by docking of the ribosome, thus maintaining the permeability barrier of the ER membrane and ensuring that the nascent polypeptide is directed into the ER lumen. In the lumen, the nascent precursor is greeted by enzymes, including chaperones, oligosaccharyltransferase, and signal peptidase, which modify, fold, and assemble these unfolded proteins into native structures. The chaperone BiP is required for import into the ER, but cleavage of the signal peptide or glycosylation is not. Apparently, only unfolded proteins are translocated through the aqueous pore, so proper folding of the imported proteins ensure their unidirectional transport.

Quality control.

Before protein import (top) into the ER, the import translocon (blue) is opened on the cytoplasmic side to receive the ribosome, which binds to the translocon and creates a seal. The lumenal gate can then be opened, and the polypeptide is translocated to the ER lumen where interaction with BiP and protein folding ensure unidirectional translocation. For export of misfolded proteins (bottom), BiP binds to a misfolded protein, and the lumenal gate opens to allow delivery of the misfolded protein to the export translocon (blue). It is not clear whether or how the exported protein is unfolded and delivered to the export translocon. The Cue1p complex ensures ubiquitination of the exported protein and its unidirectional export. For simplicity, the import and export translocons are drawn the same, but there may be components that are specific to one of the two. Both translocons are closed by a cytoplasmic gate (orange) and a lumenal gate (red). These may or may not be independent of the translocons.

Proteins that do not succeed in folding or assembling properly in the ER are shuttled back to the cytoplasm; some of the same machinery that is used for their import is also used for their export. For instance, yeast cells mutant for BiP (Kar2p) and two components of the yeast translocon, Sec61p and Sec63p, cannot export from the ER a fully translocated, mutant carboxypeptidase Y that does not fold properly (3). Proteins destined for export can be crosslinked to Sec61p, indicating that they are bound to this molecule, and in vitro experiments also implicate Sec61p in ER export (4). These findings raise several interesting issues.

Can a protein be recognized as hopelessly misfolded? How could such a misfolded protein be directed to the export channel? Although unfolded and misfolded proteins interact with the chaperone, BiP, it is not known whether this enzyme can differentiate between an unfolded protein that will eventually fold from one that will not. In addition, a misfolded protein may have too many folded domains to pass through the aqueous pore even though the pore is about 20 in diameter (5). Can BiP unfold these proteins? If so, BiP could also direct the unfolded protein to the export translocon, because BiP is known to bind to the DnaJ-like domain of Sec63p.

It is clear that some gene products are shared by the import and export translocons, but are they identical? Thus far, two components of the import transloconSec61p and Sec63phave been shown to be required for protein export. Mutants in two genes encoding membrane proteins that are required for degradation of misfolded proteins, but not ER import, have been isolated (6). Functional characterization of these genes may help to clarify this issue.

During protein import the lumenal side of the channel is closed until the nascent peptide reaches about 70 amino acid residues in length. A putative gating protein for this side of the channel has been discussed (2), but the mechanism by which the lumenal side of the translocon is sealed is still unknown. How is the permeability barrier of the ER maintained during export of misfolded proteins? The export channel must be sealed on the lumenal side of the membrane while the exported protein is emerging on the cytoplasmic side. Therefore, export of misfolded proteins through an export translocon requires coordination of the events on the two sides of the ER membrane. How can this be accomplished?

How is unidirectional export ensured? Insight into this question comes from the new work by Biederer et al. Protein ubiquitination requires at least two enzymes: E1, which activates ubiquitin and transfers it to the second required protein, E2, a ubiquitin-conjugating enzyme (Ubc). In yeast, there are 13 Ubcs. Only one of these, Ubc6p, is a transmembrane protein, although it has some overlapping specificity with the soluble Ubc, Ubc7p. These two enzymes participate in the ubiquitination of both soluble and membrane-bound substrates. The new work reports that an ER protein, Cue1p, is required to recruit the soluble Ubc7p to the membrane. A deletion of Cue1p gives the same phenotype for ER degradation as a deletion of Ubc7p, suggesting that Ubc7p must be assembled with Cue1p to be active. In a yeast strain defective for ubiquitination of misfolded proteins, the latter are stabilized and remain in the lumen of the ER. This result suggests that attachment of ubiquitin may be required to drive unidirectional export of unfolded proteins from the ER.

If the presence of a ubiquitin branch on a protein can contribute to the unidirectional export of proteins from the ER, it may also be able to block import through the translocon. For ubiquitination to occur during cotranslational import into the ER, the seal between the ribosome and the membrane must be broken to allow access of the Ubcs to the protein. Evidence suggests that the ribosome seal can be broken during the translocation of transmembrane proteins without compromising the ER permeability barrier (7). This would allow for the ubiquitination and degradation of a misfolded transmembrane protein to begin before translation has terminated. Does this occur and could this serve to degrade improperly inserted membrane proteins? Given the rapid rate of recent progress in the fields of ER translocation and degradation, answers to these enigmas should come soon.


View Abstract

Stay Connected to Science

Editor's Blog

Navigate This Article