Cell biology

Taxi service for lipids

With its role in intracellular protein transport already known, the FAPP2 protein has now also been implicated in lipid transfer and synthesis. What is more, these two FAPP2-mediated events seem to be linked.

The cellular pathways mediating the secretion of proteins such as hormones and antibodies were first described by George Palade and colleagues1 in the 1960s. Many gaps have been filled since the heyday of their work, but the basic principles remain as originally described. Proteins that travel along the secretory pathway begin their life in the endoplasmic reticulum. They then move to the Golgi complex, where they are processed further and sorted for transport to their final destinations. The vehicles for protein transport are membrane vesicles, which bud off from one compartment and fuse with the next. By contrast, synthesis and transport of membrane lipids has received less attention. This is, in part, due to a widespread misperception that lipids are simply the glue that holds proteins in place, and that their own transport within cells has little more than a passive role in secretion. On page 62 of this issue, D'Angelo et al.2 challenge this view, showing that there is an unexpected link between lipid biosynthesis and protein secretion.

Lipids come in many shapes and sizes. The three most common types of lipid in mammalian cellular membranes are glycerolipids (of which phosphatidylinositol and its derivatives are the most famous), sterols and sphingolipids. Interest in sphingolipids has been renewed during the past two decades after the discovery of their function in molecular signalling pathways and in the formation of membrane microdomains known as lipid rafts. Complex sphingolipids are formed by the addition of various head groups to a backbone ceramide molecule.

Glycosphingolipids are one such class of complex sphingolipid and are essential components of cell membranes, participating in various cellular processes including communication, differentiation, proliferation and adhesion. A bewildering number of these lipids can be formed by the sequential addition of sugar residues to sphingolipids. One of the simplest glycosphingolipids is glucosylceramide, which is formed by the addition of glucose to ceramide. Glucosylceramide is a precursor for the more complex glycosphingolipids.

Lipids, like membrane proteins, are synthesized mainly in the endoplasmic reticulum. Glycosphingolipids are an exception to this; although synthesis of the ceramide backbone occurs in the endoplasmic reticulum, complex glycosphingolipids are formed in the Golgi complex3. Glucosylceramide is even more unusual as, in contrast to more complex glycosphingolipids, which are all synthesized on the luminal (inside) surface of the Golgi complex, it is synthesized on the cytoplasmic side4,5.

Sphingolipid aficionados had spent much time and energy attempting to rationalize why glucosylceramide is synthesized on the opposite side of the membrane from the more complex glycosphingolipids. D'Angelo et al.2 might finally have found the answer. They show that glycosphingolipid synthesis depends on a cytoplasmic protein known as FAPP2, which was previously shown6,7 to be associated with protein transport from the distal compartments of the Golgi complex — the compartments that are farthest away from the endoplasmic reticulum — to the cell membrane (Fig. 1a).

Figure 1: Function of FAPP2 in glycosphingolipid biosynthesis.

a, Secreted and membrane proteins, as well as most lipids, are synthesized in the endoplasmic reticulum, and are then transferred to the Golgi complex. While passing through the compartments of the Golgi complex — early and distal — they undergo further modifications and are sorted for transport to their final destinations. b, Glucosylceramide (GlcCer), which is synthesized from ceramide in the Golgi complex, has to be carried to more distal Golgi compartments for processing into complex glycosphingolipids. D'Angelo et al.2 show that the FAPP2 protein mediates transport of GlcCer between Golgi compartments. It picks up GlcCer from the membrane of one Golgi compartment and, on reaching its destination, binds to the membrane by interacting with phosphatidylinositol 4-phosphate (PtdIns4P). Consequently, GlcCer is released and is translocated into the lumen of this compartment. Protein transport from distal Golgi compartments to the cell surface is spatially and temporally linked to ongoing GlcCer synthesis and transport, but the molecular details are not known.

Previous work had shown that FAPP2 contains a domain that is similar to the glycosphingolipid-transfer protein8. So D'Angelo and colleagues asked whether FAPP2 could also transfer glucosylceramide: the answer is yes, in a highly specific manner. The group's findings indicate that FAPP2 binds to the surface of the Golgi complex facing the cytoplasm, picks up a molecule of newly synthesized glucosylceramide and delivers it to another, probably distal, Golgi compartment. Here, FAPP2 binds to the membrane by interacting with phosphatidylinositol 4-phosphate, a membrane lipid involved in the regulation of Golgi function9 (Fig. 1b).

On their own, these findings might seem esoteric. However, the bigger picture emerged when D'Angelo and co-workers addressed the question of whether there is a link between the function of FAPP2 in glycosphingolipid synthesis and its previously described6,7 role in regulating cellular trafficking. They found that, as expected, reintroducing normal FAPP2 into FAPP2-depleted cells restores both glucosylceramide transfer and protein transport. However, when a version of FAPP2 that could not mediate glucosylceramide transfer was reintroduced into FAPP2-depleted cells, protein transport could not occur.

These results clearly show that glucosylceramide synthesis in early Golgi compartments, as well as its transport by FAPP2 to distal Golgi compartments, is required for protein transport out of the distal compartments. The molecular details of how these pathways are connected have not yet been worked out, but presumably glucosylceramide, or a more complex glycosphingolipid, is required for an aspect of vesicle budding or formation at the distal Golgi compartment. Whether this is connected to the movement of glucosylceramide from the outside to the inside of the Golgi compartments, where it is metabolized further to complex glycosphingolipids, is also not known.

There might be further twists to this story, with the path of lipid transfer being perhaps less straightforward than the authors suggest. A recent study10 shows that FAPP2 can also mediate the backward transport of glucosylceramide from the Golgi complex to the endoplasmic reticulum, where it is processed to complex glycosphingolipids. But, although many details remain unresolved, together the two studies2,10 imply an unexpected function for FAPP2 in glycosphingolipid synthesis, and, as a consequence, in protein trafficking.

An exciting concept emerging from these studies is that events in the lumen of the Golgi compartments can be controlled from the cytoplasm; that is, FAPP2 performs its lipid-transfer activity in the cytoplasm, eventually binding to the Golgi membrane by interacting with a phosphoinositide lipid. This is reminiscent of other intracellular transport and signalling pathways. For instance, ligand binding to extracellular receptors initiates signalling cascades. These also often involve phosphoinositides, which regulate intracellular transport pathways. In his wish list1, Palade stated, “there is a necessity to obtain comprehensive data on the chemistry and function of the different membranes of the secretory pathway and their interactions”. The discovery2,10 of a set of extraordinarily complex feedback and feedforward loops to regulate glycosphingolipid synthesis, and their link to protein transport and secretion, is an exciting step towards achieving this goal.


  1. 1

    Palade, G. Science 189, 347–358 (1975).

  2. 2

    D'Angelo, G. et al. Nature 449, 62–67 (2007).

  3. 3

    Futerman, A. H. & Riezman, H. Trends Cell Biol. 15, 312–318 (2005).

  4. 4

    Coste, H., Martel, M. B. & Got, R. Biochim. Biophys. Acta 858, 6–12 (1986).

  5. 5

    Futerman, A. H. & Pagano, R. E. Biochem. J. 280, 295–302 (1991).

  6. 6

    Godi, A. et al. Nature Cell Biol. 6, 393–404 (2004).

  7. 7

    Vieira, O. V. et al. Proc. Natl Acad. Sci. USA 103, 18556–18561 (2006).

  8. 8

    Malinina, L., Malakhova, M. L., Teplov, A., Brown, R. E. & Patel, D. J. Nature 430, 1048–1053 (2004).

  9. 9

    Choudhury, R. R., Hyvola, N. & Lowe, M. Biochem. Soc. Symp. 72, 31–38 (2005).

  10. 10

    Halter, D. et al. J. Cell Biol. (in the press).

Download references

Author information

Rights and permissions

Reprints and Permissions

About this article

Cite this article

Futerman, A. Taxi service for lipids. Nature 449, 35–37 (2007) doi:10.1038/449035a

Download citation

Further reading


By submitting a comment you agree to abide by our Terms and Community Guidelines. If you find something abusive or that does not comply with our terms or guidelines please flag it as inappropriate.