Just as important as starting cellular signalling pathways is switching them off again. It seems that the Cbl protein has a dual function in accelerating the degradation of certain signalling molecules.
The cells that make up our bodies live in tight-knit communities and communicate with each other almost constantly. They often do so by using messenger molecules such as growth factors; when these molecules bind to receptors on the cell surface, they activate a plethora of cellular processes that set specific gene programmes in motion. The receptors must then be inactivated so that signalling can be stopped, and breakdown of the deactivation mechanisms often leads to cancer. One of the main inactivation mechanisms entails rapid clearance of the receptor from the surface — a process termed receptor-mediated endocytosis — and degradation of messenger–receptor complexes in an acidic cellular compartment, the lysosome. Receptors may spend hours to days at the cell surface, but they are removed within seconds after messenger binding. On pages 183 and 187 of this issue, Soubeyran and colleagues1 and Petrelli and co-workers2 shed light on the rapid sequence of events that take place in this critical window of time (Fig. 1).
A major source of information on receptor-mediated endocytosis is the equally rapid recycling of membranes at synapses, the specialized junctions between nerve cells. When nerves are stimulated, neurotransmitter-laden synaptic vesicles — tiny, membrane-clad sacks — fuse with the plasma membrane and discharge their contents outside the cell. As the synapse needs to stay the same size, the vesicle membrane must then be retrieved, and this is achieved by endocytosis: areas of the plasma membrane invaginate and pinch off to form intracellular vesicles. These areas can be identified because they are coated with the clathrin protein. Membrane proteins to be internalized are sorted into the invaginations, or 'pits', through adaptor proteins, which recognize specific linear amino-acid sequences in the membrane proteins.
Another important component is dynamin — a protein that hydrolyses the molecule GTP and is involved in the pinching off (scission) of the progressively invaginating pits. Although it is unclear how the GTP-hydrolysis activity regulates scission, analyses of dynamin-interacting proteins have ascribed a key function for lipid modifications. One dynamin-interacting protein, synaptojanin, removes phosphate groups from specific membrane lipids. Another, endophilin, has lysophosphatidic acid acyltransferase activity, and might be involved in altering the curvature of membranes.
Receptor-mediated endocytosis shares many similarities with vesicle recycling; for example, clathrin-coated pits, adaptor proteins and dynamin are all involved. But, at least in the case of the epidermal growth factor receptor (EGFR), endocytosis is independent of linear sequence motifs, suggesting that membrane proteins such as growth-factor receptors are sorted into invaginating pits by a different mechanism.
The first clue to the process underlying internalization of active growth-factor receptors (also known as receptor tyrosine kinases, or RTKs) came from studies of vulva development in worms. Genetic screens identified the SLI-1 protein, a worm relative of the mammalian Cbl protein, as an inhibitor of EGFR-induced differentiation of vulva precursor cells3.
Subsequent in vitro studies of Cbl revealed that it helps to sort active RTKs into invaginating pits (reviewed in ref. 4; see Fig. 1). When EGFRs or other receptors bind their respective growth factors, certain tyrosine amino acids in the receptors become modified with phosphate groups (phosphorylated). The phosphorylated tyrosines bind to the so-called SH2 domain of Cbl. The region of Cbl next to the SH2 domain — a 'ring-finger' domain — then recruits an enzymatic machine that covalently tags EGFR with a 76-amino-acid molecule called ubiquitin. Tagging with a single ubiquitin molecule seems to be sufficient to sort out those membrane proteins that are to be internalized5. Indeed, Cbl-mediated ubiquitination of EGFR accelerates its endocytosis and delivery to lysosomes for degradation.
So the sorting of RTKs to be internalized is controlled by two types of reversible modification — phosphorylation and ubiquitination. The recycling of synaptic proteins, by contrast, depends on sequence motifs in those proteins. But the two new papers1,2 suggest a key similarity between receptor-mediated endocytosis and synaptic-vesicle recycling: lipid modifications are essential for both. What's more, Cbl is involved in these modifications, as well as in protein sorting, during receptor endocytosis.
Petrelli et al.2 used one of the non-neuronal forms of endophilin to identify a new endophilin-interacting protein — CIN85, a member of a family of adaptor proteins. The interaction occurs through a proline-rich sequence in CIN85, but the flanking SH3 domains bind to several other molecules, including Cbl. In a reciprocal approach, Soubeyran et al.1 used the carboxy-terminal end of Cbl to isolate CIN85, a portion of which enabled them to unravel the interaction with endophilin. Subsequent experiments in cultured cells revealed that a complex comprising Cbl, CIN85 and endophilin is recruited to growth-factor receptors — specifically EGFR1 and c-Met2, the receptor for hepatocyte growth factor. Maximal assembly of the complex at receptors depends on phosphorylation of receptors (to recruit Cbl) and Cbl (to increase binding to CIN85).
What is the outcome of complex assembly? By using mutants of the various components and biochemical assays, the two groups1,2 compellingly show that, once assembled at an active RTK, the complex accelerates endocytosis, thereby terminating the biological activities of the two receptors they studied.
Perhaps most unexpected is the emerging dual function of Cbl in receptor desensitization: its ability to enhance receptor ubiquitination must be complemented by its ability to recruit endophilin molecules. How endophilin participates in receptor endocytosis remains unclear, although studies of neural synapses have provided hints. For instance, injection of anti-endophilin antibodies into the giant reticulospinal synapse of lampreys interferes with vesicle recycling6, and a mutant form of endophilin can inhibit the formation of synaptic-like microvesicles in an in vitro assay7. According to existing models, endophilin acts as an 'effector' of dynamin: by adding the unsaturated fatty acid arachidonate to lysophosphatidic acid (LPA, a membrane lipid), it converts this inverted-cone-shaped lipid to phosphatidic acid, a cone-shaped lipid that favours negative curvature of the plasma membrane's inner leaflet. This is thought to facilitate the invagination of membrane pits or vesicle scission.
Rigorous proof that membrane bending by a Cbl–endophilin complex helps to dispatch ubiquitin-tagged RTKs on their long intracellular journey will require analysis by electron microscopy and precise definition of endophilin's catalytic site. Moreover, endophilin's enzymatic activity in solution is very weak (it converts just one molecule of LPA per minute); if it is important in receptor endocytosis, it has to work much more quickly in cells. Alternatively, endophilin might bend the membrane by forming a complex with other endophilin molecules and directly invaginating lipid bilayers8.
On the ubiquitination side, a major unanswered question is how the clathrin coat identifies ubiquitin-tagged RTKs. A ubiquitin-interacting amino-acid motif, found in both endosomal and proteasomal proteins, could provide the link, but firm evidence is lacking. More generally, because receptor sorting occurs in other endocytic compartments, as well as in the cellular pathway for protein synthesis and sorting, the newly discovered ability1,2 of Cbl to couple sorting events to membrane bending may become a prototype mechanism in vesicular transport.
Soubeyran, P., Kowanetz, K., Szymkiewicz, I., Langdon, W. Y. & Dikic, I. Nature 416, 183–187 (2002).
Petrelli, A. et al. Nature 416, 187–190 (2002).
Yoon, C. H., Lee, J., Jongeward, G. D. & Sternberg, P. W. Science 269, 1102–1105 (1995).
Thien, C. B. & Langdon, W. Y. Nature Rev. Mol. Cell Biol. 2, 294–307 (2001).
Hicke, L. & Riezman, H. Cell 84, 277–287 (1996).
Ringstad, N. et al. Neuron 24, 143–154 (1999).
Schmidt, A. et al. Nature 401, 133–141 (1999).
Farsad, K. et al. J. Cell Biol. 155, 193–200 (2001).
About this article
Dimerization leads to changes in APP (amyloid precursor protein) trafficking mediated by LRP1 and SorLA
Cellular and Molecular Life Sciences (2018)
Annals of Biomedical Engineering (2012)
Integrative Biology (2011)
The Oncologist (2011)
The Oncologist (2010)