Cell biology

On the endocytosis rollercoaster

Endocytosis is a process by which molecules gain access to a cell. An unusual mode of endocytosis has now been shown to regulate cell signalling, and to be highjacked by bacterial toxins. See Article p.460 & Letter p.493

The plasma membrane that surrounds cells forms a barrier that gates access to the cell, permitting entry to nutrients and extracellular messenger molecules, but locking out hazardous compounds and deadly viruses1. Clathrin protein coats some regions of this membrane, and controls cellular entry of beneficial molecules through a process called clathrin-mediated endocytosis1. Cells are also thought to use clathrin-independent modes of endocytosis2, but these have proved difficult to pinpoint. In this issue, Boucrot et al.3 (page 460) describe a fast, clathrin-independent pathway for the import of activated receptor proteins. In a separate study, Renard et al.4 (page 493) show that this pathway is hijacked by toxins from bacteria of the genera Shigella or Vibrio (which causes cholera), to allow them to gain entry to the cell.

During clathrin-mediated endocytosis (CME), nutrients, hormones, or other ligands that bind to receptor proteins on clathrin-coated membrane regions, are shuttled into cells in small vesicles, which form from invaginations in the membrane. CME not only controls uptake of receptor-bound molecules, but also serves a general function in regulating the turnover of many membrane-bound proteins. In addition to CME, the cell uses several atypical, clathrin-independent routes of endocytosis. Molecules thought to be imported through these pathways include bacterial toxins such as Shiga toxin B (ref. 5), from Shigella, and certain membrane-bound receptor proteins2,6 that, when on the cell surface, mediate transmission of signals from extracellular signalling factors to the cell nucleus.

Boucrot and colleagues set out to investigate the molecular nature of clathrin-independent endocytosis by studying a membrane-deforming protein, endophilin. Previous genetic analysis of mice lacking all three isoforms7 of endophilin shows that one function of this protein is to promote clathrin-coat shedding from vesicles during the late stages of CME. However, there is evidence that endophilin can also interact with membrane-bound receptors — either directly, as with the β1-adrenergic receptor8, or indirectly through an intermediate adaptor protein, as with certain growth-factor receptors9,10.

The authors confirmed that endophilin can directly or indirectly associate with many activated receptors. For example, they found that activation of the β1-adrenergic receptor triggers the recruitment of endophilin to clathrin-free membrane sites at the cell's leading edge (the edge at the front of the cell during migration). This leads to the rapid internalization of the receptor into small vesicles and tubules. The researchers named this entry mechanism fast endophilin-mediated endocytosis (FEME), owing to its apparent speed.

FEME could be abrogated by depleting endophilin, but not by depleting or inhibiting clathrin and its partner proteins, indicating that this mode of entry is not clathrin-dependent (Fig. 1). Furthermore, the vesicles formed during FEME seem to be distinct from those formed during other clathrin-independent modes of endocytosis, such as macropinocytosis. In the absence of endophilin, receptors accumulate on the cell surface and continue to signal, instead of being internalized — this finding may be important, because increased growth-factor signalling through such receptors is a hallmark of cancer.

Figure 1: Mechanisms of endocytosis.

a, Boucrot et al.3 describe a method of cell entry dubbed fast endophilin-mediated endocytosis (FEME), which is triggered by activation of membrane-bound receptor proteins, or by bacterial toxins such as Shiga toxin B. The pathway begins with the conversion of the phospholipid molecule PI(4,5)P2 to PI(3,4)P2, which leads to recruitment of the protein endophilin to the cell membrane. The membrane invaginates into tubules, which are subsequently snipped to form vesicles. Renard et al.4 show that this process requires the coordinated activity of endophilin, dynamin and actin proteins. b, By contrast, in clathrin-mediated endocytosis (CME), receptors (including the activated receptors that are involved in FEME) and other cargo are captured into clathrin-protein-coated membrane invaginations, which then undergo dynamin-mediated scission. Clathrin is subsequently removed from the vesicle surface. Like FEME, CME requires PI(3,4)P2, but this is produced through a different mechanism10.

Impressively, Boucrot and colleagues determined the mechanism by which receptor activation causes endocytosis. First, activation triggers the addition of a phosphate group to phosphatidylinositol 4,5-bisphosphate (PI(4,5)P2) — a membrane-bound phospholipid molecule that activates many cell-signalling pathways — to form another phospholipid, PI(3,4,5)P3. This transformation is transient, because PI(3,4,5)P3 is rapidly converted to PI(3,4)P2, which binds to a protein called lamellipodin that is distributed at the cell's leading edge. Lamellipodin binds endophilin, and thus endophilin accumulates at the leading edge. Accumulated endophilin induces the formation of vesicles that then become internalized, transporting their cargo to the cell's centre.

Renard et al.4 convincingly demonstrate that Shiga and cholera toxins hijack the FEME pathway. They show that the A2 isoform of endophilin coats tubule-shaped membrane invaginations that arise when Shiga toxin B binds to the cell membrane. A similar endophilin-based coat may also form in receptor-activated FEME, although this was not investigated. The authors further show that, during FEME, both endophilin and dynamin (another protein involved in endocytosis) cooperate with actin, a structural protein that can polymerize into branched filaments, to snip off the tubule from the membrane, internalizing Shiga toxin B. Such severing probably occurs through a mechanism based on friction, which may be induced by interactions between the endophilin coat and the underlying lipids as the tubule extends and actin polymerizes.

For the first time, the molecular machinery that controls a specific clathrin-independent mode of endocytosis has been defined. FEME might be required in many cell types, including neurons, which use both CME and clathrin-independent endocytosis to recycle the vesicles responsible for the release of neurotransmitter molecules at neuronal junctions (synapses)11. A notable feature of FEME is that it must be triggered to occur, either by receptor-activated signalling cascades or by toxin-induced membrane deformation. By contrast, CME is believed to be largely constitutive.

FEME is unlikely to be the only pathway of clathrin-independent endocytosis, as Boucrot et al. note. They show that, in the absence of either CME or FEME, growth-factor receptors such as epidermal growth factor receptor can still be internalized when activated, through macropinocytosis. It will also be interesting to determine the relative contributions of FEME and CME to general membrane recycling under physiological conditions, because previous studies have yielded conflicting results regarding the relative contributions of CME12 and clathrin-independent pathways2 to this process.

The current findings open up avenues for further mechanistic studies. It is unclear how endophilin is recruited to either the CME or the FEME pathway, for instance in synapses, where both processes occur. Lamellipodin might control recruitment for FEME, but the protein's equivalent in CME is unknown. Furthermore, FEME and CME share several components, including dynamin and PI(3,4)P2 (although this lipid is synthesized by distinct routes in FEME and CME13). This not only presents possibilities for simultaneous pharmacological inhibition of both pathways, but also indicates that cells can reuse functional protein modules, for example the proteins involved in membrane scission, and can direct them to more than one pathway. Such direction might depend on environmental conditions, for example cell type or physiological stimulus. Future studies will be required to fully understand the rollercoaster ride of endophilin-mediated endocytosis.


  1. 1

    Doherty, G. J. & McMahon, H. T. Annu. Rev. Biochem. 78, 857–902 (2009).

    CAS  Article  Google Scholar 

  2. 2

    Howes, M. T., Mayor, S. & Parton, R. G. Curr. Opin. Cell Biol. 22, 519–527 (2010).

    CAS  Article  Google Scholar 

  3. 3

    Boucrot, E. et al. Nature 517, 460–465 (2015).

    ADS  CAS  Article  Google Scholar 

  4. 4

    Renard, H.-F. et al. Nature 517, 493–496 (2015).

    ADS  CAS  Article  Google Scholar 

  5. 5

    Römer, W. et al. Cell 140, 540–553 (2010).

    Article  Google Scholar 

  6. 6

    Lamaze, C. et al. Mol. Cell 7, 661–671 (2001).

    CAS  Article  Google Scholar 

  7. 7

    Milosevic, I. et al. Neuron 72, 587–601 (2011).

    CAS  Article  Google Scholar 

  8. 8

    Tang, Y. et al. Proc. Natl Acad. Sci. USA 96, 12559–12564 (1999).

    ADS  CAS  Article  Google Scholar 

  9. 9

    Soubeyran, P., Kowanetz, K., Szymkiewicz, I., Langdon, W. Y. & Dikic, I. Nature 416, 183–187 (2002).

    ADS  CAS  Article  Google Scholar 

  10. 10

    Petrelli, A. et al. Nature 416, 187–190 (2002).

    ADS  CAS  Article  Google Scholar 

  11. 11

    Kononenko, N. L. et al. Neuron 82, 981–988 (2014).

    CAS  Article  Google Scholar 

  12. 12

    Bitsikas, V., Correa, I. R. & Nichols, B. J. eLife 3, e03970 (2014); http://dx.doi.org/10.7554/eLife.03970

    Article  Google Scholar 

  13. 13

    Posor, Y. et al. Nature 499, 233–237 (2013).

    ADS  CAS  Article  Google Scholar 

Download references

Author information



Corresponding author

Correspondence to Volker Haucke.

Rights and permissions

Reprints and Permissions

About this article

Verify currency and authenticity via CrossMark

Cite this article

Haucke, V. On the endocytosis rollercoaster. Nature 517, 446–447 (2015). https://doi.org/10.1038/nature14081

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.


Nature Briefing

Sign up for the Nature Briefing newsletter — what matters in science, free to your inbox daily.

Get the most important science stories of the day, free in your inbox. Sign up for Nature Briefing