Yeast have a protein complex known as 'retromer' that retrieves the receptor Vps10 (vacuolar protein sorting-10) from endosomes and returns it the Golgi. In mammalian cells, endosome-to-trans-Golgi-network (TGN) retrieval is poorly understood, although several proteins or protein complexes are thought to be involved in the endosome-to-TGN retrieval of mannose-6-phosphate receptors (MPRs) — the mammalian counterparts of Vps10. So, might the mammalian retromer be one of the protein complexes involved?

Two papers in The Journal of Cell Biology now answer this question with a resounding “yes”. In the first paper, Matthew Seaman showed that VPS26, a mammalian retromer subunit, is localized to endosomal membranes. As this location fits with the suspected role for the mammalian retromer, he then studied endosome-to-TGN retrieval in a cell line that was derived from Vps26−/− mice and in HeLaM cells that had been treated with small interfering RNA (siRNA) against VPS26.

Seaman found that, in both cases, a loss of VPS26 expression disrupted the trafficking of the cation-independent MPR (CI-MPR). In the mouse Vps26−/− cells, there was an increased degradation of CI-MPR in lysosomes, whereas in the siRNA-treated cells, there was an increased cycling of CI-MPR between endosomes and the cell surface. This difference is probably due to cell-type-specific trafficking pathways for CI-MPR.

In the final part of his study, Seaman used reporter constructs and antibody-uptake assays to show that a CI-MPR reporter construct moves from the cell surface to the Golgi through VPS26-positive endosomes and that, in the anti-VPS26 siRNA-treated cells, this construct is retained in endosomes. Together, these data “...provide compelling evidence that [VPS26] (and therefore retromer) is required for efficient retrieval of the CI-MPR from endosomes to the Golgi.”

In the second paper, Juan Bonifacino and colleagues used a yeast two-hybrid system and deletion mutants to show that two non-overlapping regions of the CI-MPR cytosolic domain can independently interact with VPS35 (another mammalian retromer subunit). This prompted them to look at the potential role of retromer in CI-MPR trafficking, and they used immunofluorescence microscopy to show that retromer mainly localizes to endosomes, where it partially colocalizes with CI-MPR.

Using immunogold cryo-electron microscopy for increased resolution, they showed that retromer is associated with tubular–vesicular profiles that emanate from early endosomes or from intermediates in the maturation pathway from early to late endosomes. They also showed that CI-MPR can be found both in these retromer-containing tubules (which are probably en route back to the TGN) and in the intralumenal vesicles of endosomes.

To establish a functional link between retromer and CI-MPR, Bonifacino and co-workers used siRNA to deplete HeLa cells of VPS26 or VPS35. They observed that this resulted in a marked decrease in CI-MPR levels, which was due to an increase in its delivery to, and degradation in, lysosomes. These studies therefore “...indicate that retromer prevents the delivery of CI-MPR to lysosomes, probably by sequestration into endosome-derived tubules from where the receptor returns to the TGN.” And, together, these papers have highlighted a retrieval role for the mammalian retromer.