Featured Articles
Protein trafficking: Glycan structure sorts it out
Functional Glycomics (12 November 2009) | doi:10.1038/fg.2009.33Standfirst
A mutant cell line deficient in the transport of GPI-linked proteins initiates the discovery of an unknown phosphodiesterase as critical for GPI maturation and sorting.
© Kok Leng Yeo, C-C 2.0
Proteins acquire a variety of post-translational modifications that determine their final function or location in the cell, including glycosylation, prenylation, and the attachment of glycosylphosphatidylinositol (GPI) anchors. GPI chains, synthesized and attached to proteins in the ER, promote the association of GPI-anchored proteins (GPI-APs) with specific lipid membrane domains. However, the specific signal that triggers the exit of GPI-APs from the ER was not known. Now, Kinoshita and colleagues have developed a mutant cell line that provides insight into intermediate protein transport steps. They report, in Cell, the identification of a specific enzyme that regulates trafficking by remodeling the glycan moiety of GPI, thus defining the importance of glycan structure in the transport pathway.
Investigations of protein trafficking are complicated by the multiple steps involved. Before a GPI-AP appears on the surface, the protein and glycolipid must be synthesized and assembled, with an additional lipid remodeling step known to occur on the assembled construct. To investigate this complex process, the authors used a temperature-sensitive mutant cell line in which protein trafficking can be halted artificially by the addition of doxycycline at a non-permissive temperature. By shifting the cells to a permissive temperature, trafficking is restored. Further mutation of this cell line led to the identification of two mutants that showed a delayed transport phenotype at the permissive temperature.
Through a series of elegant experiments, the authors demonstrated that the protein responsible for this altered phenotype is a metallophosphoesterase with a previously unknown function. Mutations to this protein, now termed PGAP5, slowed trafficking of GPI-APs but not other transmembrane proteins, denoting a GPI-specific function. Similarly, the protein was localized at sites relevant to GPI-AP trafficking, such as the ER exit site and the ER-Golgi intermediate compartment, and was also colocalized with proteins known to be important for GPI-AP transport or to be modified by GPI chains.
Finally, biochemical assays and the use of point mutants served to establish the role of PGAP5. The glycan portion of GPI is a complex carbohydrate, consisting of a linear pentasaccharide with multiple modifications. One modification, the attachment of an ethanolamine phosphate to the second mannose residue in the chain (yielding 'H8'), has been observed in the ER, but is typically not present in GPI-APs on the cell surface. It has been assumed that a simpler construct without this ethanolamine ('H7') is the biologically relevant structure for protein attachment, or that the two forms are similar in function. Using in vitro assays, Kinoshita and colleagues demonstrated that PGAP5 removes the ethanolamine group from H8 when linked to a target protein, strongly suggesting that PGAP5 function serves as the immediate trigger for GPI-AP exit.
The discovery of PGAP5 as a mediator of glycan remodeling highlights how the richness of carbohydrate structures can coordinate complex cellular processes. This study should also lay important foundations to search for additional proteins involved in ER export, further expanding our understanding of protein transport.