PerspectiveCell Biology

New Roles for Lysosomal Trafficking in Morphogen Gradient Sensing

See allHide authors and affiliations

Science Signaling  03 May 2011:
Vol. 4, Issue 171, pp. pe24
DOI: 10.1126/scisignal.2002053


The way in which cells recognize their position in a gradient of morphogen controls differentiation during embryogenesis. New findings indicate that the rate at which internalized morphogen receptors are trafficked to lysosomes is key to the accurate and precise sensing of morphogen gradients and the appropriate initiation of differentiation programs during development.

Concentration gradients of morphogenic growth factors (morphogens) establish themselves at particular times during development, and these control the way in which embryos are patterned. Morphogen gradients command key developmental processes, such as differentiation and cell migration, but the way in which target cells interpret these gradients has not been fully elucidated. A recent study shows that the appropriate intracellular trafficking of morphogen receptors is necessary for target cells to interpret their position in a morphogen gradient and initiate their differentiation program accordingly. Nowak et al. (1) found that the ubiquitin ligase Cbl is required for the interpretation of fibroblast growth factor 8 (Fgf8) morphogen gradients during zebrafish development. Previously, Jullien and Gurdon (2) showed that another ubiquitin ligase complex, consisting of Smurf2 and Smad7, enables activin gradients to coordinate spatial and temporal aspects of cell differentiation in Xenopus larvae. A common strand to both these studies is that it is the rate at which internalized morphogen receptors are transported from early endosomes to lysosomes in the target cell that dictates the extent of downstream signaling and, in turn, their appropriate differentiation.

Morphogens can control embryo patterning by promoting cell migration and chemotaxis (3), or they can trigger position-dependent cell fate decisions (4). During both these processes, cells need to distinguish between small differences in growth factor concentrations. During morphogen-induced patterning, it appears that cells are calibrated to recognize a particular concentration of growth factor and to respond to it in a certain way, while responding quite differently to a concentration of the same molecule that varies by less than threefold (5). Moreover, because morphogen gradients are often transiently imposed, it is sometimes necessary for target cells to “memorize” the highest concentration of morphogen to which they have been exposed so that they can respond to it by differentiating several hours after the gradient itself has collapsed (6). Thus, the way in which a cell’s morphogen receptors and intracellular signaling pathways are calibrated to recognize particular concentrations of morphogen with accuracy and precision is likely to be key to establishing its position in the morphogen gradient. However, the fact that most growth factor receptors exhibit very slow ligand-receptor off-rates, and that they are normally internalized into intracellular compartments after occupation, indicate that the mechanism by which receptors accurately report morphogen concentrations to the cell’s signaling apparatus may not be straightforward.

Postendocytic receptor trafficking influences the intensity, type, and duration of downstream signaling (7), indicating the likelihood that it plays an important role in controlling embryonic growth and development. The first port of call for internalized receptors is the early (or sorting) endosome. From here, receptors are triaged according to their type and occupation status and can then follow different intracellular routes (Fig. 1). Unoccupied receptors, including those that are internalized without ligand and those that have been divested of ligand after endocytosis, are routed from early to recycling endosomes and then return to the cell surface. On the other hand, ligand-occupied receptors do not return to the plasma membrane and, by recruiting various adaptor proteins, promote active signalosome assembly at endosomal membranes (7). At the same time, these actively signaling receptors bind to E3 ubiquitin ligases that catalyze the conjugation of ubiquitin to certain residues in their cytoplasmic domains (8). Ubiquitin acts as a signal to recruit the ESCRT complex, which sorts receptors into vesicles within the lumen of endosomes, and the resulting complex membranous structures are termed multivesicular bodies (MVBs) (Fig. 1). MVBs then fuse with lysosomes, leading to degradation of the receptor cargo and termination of signaling. Thus, a picture is now emerging in which receptors perform much of their intracellular signaling en route from early endosomes to MVBs, and the speed of this transport process will dictate the duration of their effective signaling (Fig. 1).

Fig. 1

Endosomal trafficking of activated receptors. Ligand-occupied receptors are internalized into early endosomes in a dynamin- and Rab5-dependent manner. Internalized receptors can then take different routes. They can recycle back to the plasma membrane through a Rab11-mediated pathway or, after Cbl-mediated ubiquitylation, can be sequestered by the ESCRT complex into the lumen of MVBs and targeted to lysosomes for degradation. It is thought that high-strength signaling to downstream pathways occurs after internalization but before sequestration in the lumen of MVBs. Ub, ubiquitin; P, phospho group.


A study from the Brand laboratory has provided evidence that the rate at which morphogen receptors are transported between early endosomes and MVBs is key to how target cells recognize morphogen gradients. The study was focused on the recognition of Fgf8 gradients during zebra­fish gastrulation (1). Fgf8 is expressed and secreted by cells positioned at the embryonic margin during gastrulation. Fgf8 then diffuses away from this source, forming a concentration gradient across the neighboring tissue to induce graded expression of target genes that promote differentiation.

Fgf8 binds to the receptor tyrosine kinase FgfR1 (fibroblast growth factor receptor 1) on the surface of target cells, and this triggers endocytosis of the ligand-receptor complex, which then subsequently progresses through early endosomes toward MVBs and lysosomes for degradation. The conjugation of ubiquitin to FgfR1 cytoplasmic domains, which is key to this progression, is catalyzed by the E3 ubiquitin ligase Cbl (9). Cbl activity can be specifically opposed by expressing a dominant-negative mutant (Cbl-YF) that can still bind to FgfR1 but lacks ubiquitin ligase activity because of a tyrosine to phenylalanine substitution in the E3 ligase domain. In zebrafish, microinjection of mRNA encoding Cbl-YF or knocking down endogenous Cbl protein with morpholino oligonucleotides delayed transport of FgfR1 from endosomes to MVBs and lysosomes, and correspondingly increased the number of internalized receptors that were associated with signaling adaptors (1).

To investigate how FgfR1 trafficking altered the target cell’s ability to sense a morphogen gradient, Novak et al. (1) implanted a source of exogenous Fgf8 (Fgf8-loaded beads) into embryos and monitored the amount of endosomal Fgf8 in cells at various distances from the beads. The amount of endosomal Fgf8 was high near the bead and decreased with increasing distance from the source. However, after expression of Cbl-YF, the differential between the amount of endosomal Fgf8 in cells at various distances from the bead was markedly reduced. This indicated that the rate at which internalized receptors are transported to lysosomes dictates the relationship between the morphogen concentration and the signaling response of the cell. Normal rates of endosome to lysosome transport yield a close relationship between Fgf8 dose and the cellular response (for example, high Fgf8 concentrations give a high response and low Fgf8 concentrations yield a low response), whereas inhibition of lysosomal targeting flattens this relationship such that the extent of FgfR1 signaling differs less over a broader range of distances from the source (Fig. 2A). This interpretation was borne out when the authors looked at the effect of Cbl-YF on the relationship between Fgf8 expression and the activation of its target genes, spry4 (which encodes sprouty4) and pea3, in zebrafish embryos. Normally, spry4 and pea3 are found in a relatively restricted region close to the source of Fgf8 in the embryonic margin, consistent with a close relationship between morphogen concentration and the cellular response. However, expression of Cbl-YF (although it did not alter the nature of the Fgf8 gradient itself) broadened the regions of spry4 and pea3 expression, such that they were now expressed in cells considerably farther from the source.

Fig. 2

(A) The degradative pathway dictates the signaling response to an Fgf8 concentration gradient. Normal rates of endosome to lysosome transport yield a close relationship between Fgf8 dose and the cellular response. Inhibition of lysosomal targeting flattens the relationship between Fgf8 concentration and downstream signaling such that the extent of FgfR1 signaling differs less over a broad range of distances from the source. (B) The time spent by activin receptors in endosomes en route to MVBs and lysosomes dictates signaling duration. Activin binds to TGF-β receptors at the cell surface. The receptor-ligand complex is internalized and trafficked through early endosomes to MVBs and lysosomes. The activin–TGF-β complex activates Smad2 en route from the plasma membrane to MVBs, yielding a signal duration of ~4 hours (red line). Expression of the ubiquitin ligase complex Smad2 and Smurf7 accelerates TGF-β receptor degradation and curtails the duration of receptor signaling and the ability of activin to evoke target gene expression. Blockade of TGF-β receptor endocytosis by expressing dominant-negative dynamin (dynamin DN) opposes the ability of TGF-β receptors to signal to Smad2.


A previous study from the Gurdon laboratory also showed that trafficking of receptors to MVBs and lysosomes influences interpretation of morphogen gradients (2). In Xenopus larvae, the morphogen activin is produced at an early stage of development. However, increases in the abundance of one of activin’s target genes, Xbra, occurs much later and only reaches a peak long after the activin gradient has dissipated (6). Thus, to recognize its position within an activin gradient, a cell needs to correctly interpret the morphogen concentration and “remember” this for several hours after collapse of the gradient. Like FgfRs, activin receptors [which belong to the transforming growth factor–β (TGF-β) receptor superfamily] are internalized after occupation and transported through early endosomes and onward to MVBs and lysosomes for degradation (10). As for other receptor types, key to this progression is conjugation of ubiquitin to receptor cytoplasmic domains, which for activin and TGF-β receptors is catalyzed by an E3 ubiquitin ligase complex consisting of Smurf2 and Smad7 (10). Jullien and Gurdon (2) found that microinjection of mRNAs encoding Smurf2 and Smad7 to increase receptor ubiquitinylation accelerated the progress of activin receptors to the lysosome. Conversely, addition of the proteasome inhibitor MG132 to oppose sorting of ubiquitinylated receptors into the lumen of MVBs delayed degradation of internalized activin-receptor complexes. TGF-β receptors are serine threonine kinases that signal by recruiting and phosphorylating effector proteins such as Smad2, which are then transported to the nucleus to activate target genes (10). Similar to receptor tyrosine kinases, activated TGF-β receptors carry out much of their signaling after they have been internalized, by recruiting Smads and other accessory proteins to endosomal membranes (10). Consistent with this, Jullien and Gurdon (2) found that the time spent by activin receptors in endosomes en route to MVBs and lysosomes dictates their effective signaling strength and duration. For instance, they found that blockade of internalization opposed activin receptor signaling completely. Moreover, they showed that expression of Smad2 and Smurf7 to accelerate receptor degradation curtailed the duration of receptor signaling and the ability of activin to evoke target gene expression by several hours (Fig. 2B). Thus, to effectively “memorize” an activin gradient, receptors must remain signaling in endosomes for just the right amount of time before they are sequestered into the lumen of MVBs and targeted to lysosomes for degradation.

Taken together, these studies indicate that the time spent by receptors in signaling endosomes en route to lysosomes determines the strength and duration of downstream signaling. However, the intensity and duration of signaling is unlikely to be the only determinant in fate decisions. The characteristics of the endosomal membrane in question, in particular its lipid composition and curvature, are likely to influence the type of signaling generated in endosomes. Further characterization of the membrane composition and signaling microenvironment at FgfR1- and TGF-β receptor-positive endosomes will help to determine how and why signaling generated here is most effective to drive differentiation in zebra­fish and Xenopus models. Fluorescence resonance energy transfer–based approaches to measure association of receptors with their signaling adaptors in living cells (11) will probably help resolve details of how FgfR1 and TGF-β receptors signal from endosomes in zebrafish and Xenopus embryos. Moreover, compartmentalization of hepatocyte growth factor signaling in endosomes can transduce weak signals to the nucleus (12), leading to the suggestion that the juxtanuclear positioning of signaling endosomes may directly contribute to nuclear import of downstream signaling elements. A more detailed characterization of the positioning of morphogen signaling endosomes within target cells and the type of signalosomes assembled upon them may help to determine how endocytosis influences the encoding of receptor signaling during morphogenesis.

The conclusions of the Nowak (1) and Jullien (2) studies were reached by imposing artificial manipulations on receptor trafficking. However, various signaling pathways can impinge on receptor ubiquitinylation and lysosomal trafficking, indicating the possibility of tangential inputs that might naturally tune morphogen signaling. Indeed, in epithelial cells, Src induces FgfR1 stabilization by opposing Cbl-mediated ubiquitinylation (13), and Nowak et al. (1) raised the possibility that Src might influence Cbl to generate more complex embryo patterning. Furthermore, both FgfR1 and FgfR4 are abundant during Xenopus mesoderm formation (14), and evidence from mammalian cell culture systems suggests that these Fgf receptors are trafficked differently (15). Whereas ubiquitinylation of FgfR1 targets it to lysosomes, FgfR4 seems to be recycled irrespective of its ubiquitinylation status (16). Thus, it is likely that the different intracellular trafficking of these two Fgf receptors will be important in the sensing of gradients during Fgf-dependent stages of embryogenesis such as mesoderm formation, left/right axis specification, and neural induction.

The Nowak (1) and Jullien (2) studies have both shed light on how cells are calibrated to recognize particular concentrations of morphogen and initiate their response accordingly, but it is not clear how these mechanisms intersect with those that control cell migration toward a chemoattractant source. During chemotaxis, it is not so much the recognition of absolute morphogen concentration that is important, but more a target cell’s ability to sense small differences in growth factor concentration along their length and to translate this into directed cell movement. Receptor endocytosis and Cbl-mediated trafficking are required for directional migration of Drosophila border cells during oogenesis and are thought to operate by spatially restricting and reinforcing signaling at the cell front (17). Future work will reveal further details of how E3 ubiquitin ligases control receptor trafficking to calibrate a cell’s response to absolute morphogen concentrations to dictate differentiation, and to assist in the recognition of relative morphogen concentration across a cell’s length to guide cell migration, as the way in which these processes are integrated is likely to be key to the generation of complex structures during embryogenesis.

References and Notes

  1. Funding. E.R. and J.C.N. are funded by Cancer Research, UK.
View Abstract

Navigate This Article