Prime time for the recycling endosome

Receptor crosstalk is the phenomenon by which one cell surface receptor communicates with another to modulate its activity. In this issue, Smith et␣al (2021) demonstrate that such crosstalk may involve endocytic trafficking, as ligands promoting FGFR2b recycling induce a specific “priming” EGFR phosphorylation to direct unliganded EGFR to the recycling endosome, slow the lysosomal degradation of ligand‐stimulated EGFR, and enhance signaling and cell proliferation.

New work implicates endocytic trafficking and cross‐priming phosphorylation as new paradigms for crosstalk between signaling receptors.

To understate the obvious, signaling networks are complex. Extracellular signaling molecules (such as hormones, growth factors, cytokines, or neurotransmitters) bind their designated cell surface receptors to alter intracellular biochemistry. The cell's response to such signals is determined not just by which pathways are activated, but also by the magnitude and duration of activation; not too little and not too much keeps the cell in homeostasis. Important regulatory mechanisms exist for controlling the extent of signaling, including tissue‐specific secretion of ligands, ligand‐specific affinities, receptor dimerization partners, the subcellular distribution of downstream effectors, and feedback loops. However, the most universal, yet incompletely understood, regulation is via endocytic trafficking of receptors.

Clinical evidence highlights the physiologic importance of proper receptor trafficking. A number of cancers are characterized by receptor mutations that alter their trafficking or by mutation of one of the proteins that guide a given receptor through the endocytic pathway. For example, a subset of lung cancer patients exhibits somatic mutations in exon 14 of the receptor tyrosine kinase (RTK) Met, which impair its lysosomal degradation and result in sustained signaling (Kong‐Beltran et␣al, 2006). Still, the underlying molecular mechanisms have not been fully elucidated. Most earlier studies on signaling and receptor endocytosis focused on the spatio‐temporal regulation of signaling, but it was never clear what was functionally more important: The rate of ligand: receptor complex degradation? Or the proximity of the activated receptor to downstream signaling molecules? Identifying how receptor trafficking impacts signaling is important for utilizing this information for therapies. This new study by Chiara Francavilla and coworkers (Smith et␣al, 2021) proposes a third option: trafficking regulates receptor crosstalk.

The authors' entry point to these question was their using of triple‐negative breast cancer cells to characterize how membrane trafficking would affect signaling by fibroblast growth factor receptor 2b (FGFR2b). Previous work had shown that FGFR2b trafficking could be modulated by the choice of stimulating ligand. Fibroblast growth factor 10 (FGF10) promotes recycling of the receptor, whereas fibroblast growth factor 7 (FGF7) directs the receptor to the lysosome for degradation (Fig 1). Furthermore, stimulation with FGF10 favors cell motility, but it had remained unclear which signaling pathways are activated. Once identified, such pathways might be selectively targeted for new anti‐cancer therapies.

Figure 1. FGFR2b signaling is dependent on its endocytic trafficking.

While FGF7 induces FGFR2b internalization and lysosomal degradation, the FGF10:FGFR2b complex cross‐primes EGFR on recycling endosomes to slow degradation, enhance ERK‐mediated signaling, and promote cell proliferation. See text for details.

Smith et␣al (2021) started out with a phosphoproteomic screen of five breast cancer cell lines treated with either FGF10 or FGF7, to differently stimulate trafficking pathways. From this unbiased approach, they used hierarchical clustering and found an enrichment of proteins in signaling pathways, adhesion, and endocytosis and transport to be common among all FGFR2b signaling events, as well as ligand‐specific differences in the phosphoproteome. Among the proteins that were phosphorylated specifically in response to recycling ligands were those that localized to the early endosome and were involved in FGFR2b and EGFR recycling [i.e., Transferrin receptor trafficking protein (TTP) and Rab‐coupling Protein (RCP)]. This validation of the approach provided the impetus to examine how long‐term exposure to recycling ligands impacted signaling. After treatment with FGF10 and FGF7 for short (8 min) and long (40 min) periods, they performed the phosphoproteomic analysis again and identified 113 phosphoproteins after sustained treatment with the recycling ligand. Of these, 22 were particularly interesting because of their known roles in signaling and trafficking. One, however, EGFR phosphorylated on threonine 693 (EGFR_T693), stood out for several reasons: first, because it indicated that there is crosstalk between FGFR2b and EGFR in breast cancer cells. Second, threonine 693 is located in the non‐catalytic domain of the receptor, so it is not directly involved in propagating signaling. Finally, threonine 693 phosphorylation is known to be required for EGFR internalization (Heisermann et␣al, 1990) and to occur upon stimulation by the EGFR recycling ligand TGFα. Taken together, this suggested a novel mode of receptor crosstalk.

Once Smith et␣al (2021) had established that the FGF10:FGFR2b complex mediates phosphorylation of EGFR_T693 via the kinase ERK, the authors studied the biological significance of this phosphorylation event. With regard to receptor trafficking, pre‐treatment with FGF10 slowed the rate of ligand‐mediated EGFR degradation. Further investigation revealed that the phosphorylation of EGFR_T693 is required for recruiting EGFR to the recycling endosome. As shown by immunofluorescence and co‐immunoprecipitation experiments, this endosome serves as a nidus of interaction for FGFR2b and EGFR.

Additional phosphoproteomic analysis highlighted important signaling differences between wild‐type and internalization‐defective EGFR variants. Once primed by FGF10:FGFR2b signaling, EGFR ligands sustained ERK phosphorylation, increased expression of ERK target genes, and enhanced cell proliferation. When EGFR endocytosis was blocked by EGFR T693A mutation, ERK‐mediated signaling decreased both in response to FGF10 and TGFα. Expression of the T693 mutant further reduced phosphorylation of the cell cycle regulator CDK1 and decreased cell proliferation. On the other hand, abolishing EGFR priming phosphorylation did not affect cell invasion (Fig 1). A particularly interesting facet of the study is that the crosstalk is bidirectional. Phosphorylation of EGFR_T693 controls FGFR2b trafficking, as phosphorylation‐defective mutants of EGFR_T693 result in more rapid recycling of and altered signaling by FGFR2b.

These findings imply a new model for receptor crosstalk (Fig 1). Previous models viewed crosstalk as one signaling pathway modulating another receptor or effector. While the FGF10:FGFR2b complex does “activate” EGFR, the distinction is that it does not qualitatively stimulate signaling cascades. Instead, it directs trafficking to control the magnitude of a response. In a sense, it allows FGFR2b to call in reinforcements to maximize the proliferative response. In the absence of EGFR internalization, the cell tempers cell cycle progression, migration, and motility. There may well be additional layers of EGFR:effector communication that mediate such cell‐biological outcomes. In addition, this provides a new way to think about receptor trafficking, which was so far largely recognized as either constitutive or ligand‐mediated. Nutrient receptors (e.g., those for transferrin and low‐density lipoprotein, LDL) enter the cell independent of whether they are loaded with their cargo. In contrast, signaling receptors such as RTKs typically have a low constitutive rate of endocytosis, which however increases with ligand binding. The findings of Smith et␣al (2021) extend this by demonstrating how one receptor can direct the internalization of another.

Many cell biologists will be interested in these findings, as this mode of crosstalk is likely not unique to FGFR2b and EGFR. Since FGFR2b primes EGFR via the kinase ERK, any receptor with that recognition motif should be a candidate. Amino acid sequence alignment of EGFRs indicates that an analogous threonine residue can be found in other ErbB family members. In fact, many receptors may be modified by recycling FGFR2bs. Conversely, one must consider that the FGF10: FGFR2b complex is not the only receptor that can prime EGFR and that EGFR priming likely extends to other ligand: receptor combinations that promote recycling. In fact, the significance of ligands that differentially direct receptor trafficking may have been underestimated. Many receptors, particularly RTKs, have multiple ligands that differentially regulate receptor trafficking. EGFR, for instance, has naturally occurring ligands that promote rapid recycling, slow recycling, or lysosomal degradation (Roepstorff et␣al, 2009). Guided by this new model of receptor crosstalk, one could envision how the regulated release of one ligand versus another could provide the necessary fine‐tuning needed to maintain tissue homeostasis.

While these findings will excite cell biologists that think about how signaling networks are coordinated, there may be some important clinical implications as well. Priming phosphorylation of EGFR_T693 may be a prognostic indicator of a metastatic cancer. Alternatively, inhibiting phosphorylation of EGFR_T693 may be a target for anti‐cancer therapy.

The EMBO Journal (2021) 40: e108758.