receptor transactivation refers to the ability of a primary agonist, via binding to its receptor, to activate a receptor for another ligand. This situation can occur with G protein-coupled receptors (GPCRs), leading to activation of growth factor receptors. In the November 2006 issue of AJP-Cell Physiology, an interesting example is provided by Chen and colleagues (4a), in which binding of an agonist to a GPCR in primary cells leads to release of epidermal growth factor (EGF), which in turn activates the EGF receptor (EGFR).1 In contrast to the pro-mitogenic role of EGFR transactivation that is typically reported, the authors have shown that EGFR transactivation in this situation results in negative feedback on GPCR-induced protein secretion. The study presented in this issue further elucidates the mechanism of the pathway. The resulting scenario illustrates several important points. First, growth factor receptor transactivation via GPCRs does not occur solely in tumor cells. Second, EGF itself (as opposed to other EGFR ligands) can be released via GPCR stimulation. Third, transactivation of a growth factor can have a negative effect on GPCR signaling. Thus, as we encounter more examples of receptor transactivation, it becomes apparent that there is a greater level of complexity to this phenomenon than was initially appreciated.
One of the more complex and intriguing phenomena involved in signal transduction networks involves receptor “transactivation”, the ability of one receptor to activate another receptor via signaling events. These pathways can be viewed as a receptor transactivation “cascades”, by analogy to protein phosphorylation cascades. Transactivation of various growth factor receptors, including the epidermal growth factor receptor (EGFR), and the platelet-derived growth factor (PDGF) receptor, by G protein-coupled receptors (GPCRs) has been documented in multiple cellular model systems (3, 4, 11, 13, 27). GPCR-induced transactivation of Trk neurotrophin receptors has also been reported (3, 11). Receptor transactivation can potentially occur through several different mechanisms. One mechanism is through activation of intracellular protein tyrosine kinases, such as c-Src and PKC (2, 33), which can phosphorylate the growth factor receptor and thereby promote its activation. A more commonly reported mechanism involves release of membrane-tethered growth factors from the membrane by activation of matrix metalloproteinases (MMPs) (12, 27, 28). These growth factors, such as pro-heparin-bound (HB)-EGF (20), are released from their precursors and then bind to their cognate growth factor receptors (17). Since pro-mitogenic signaling by the primary GPCR can be mediated primarily by transactivation of a growth factor receptor, the downstream signaling cascade can be significantly attenuated when generation or response of the growth factor is inhibited (6, 12). Strategies that have been used to interrupt the transactivation pathway include EGFR kinase inhibitors (10, 24), anti-EGF antibodies (8, 20), and MMP inhibitors (12, 30). However, it is not yet clear whether transactivation of growth factor receptors is universally required for the mitogenic responses mediated by GPCRs.
GPCR-induced transactivation of EGFRs has been studied in some detail. GPCR activation leads to cleavage of pro-EGF (e.g., pro-HB-EGF) ligands from the membrane (2, 16, 31), initiating the transactivation (see Fig. 1). Agonist-bound GPCRs activate numerous MMPs (6), including MMP-3 (19), MMPs 2 and 9 (18, 28), as well as members of the ADAM family of metalloproteases: ADAM10, ADAM15, and ADAM17 (7, 29, 32). The molecular mechanisms underlying GPCR-induced MMP activation are not yet clear. MMPs, which are synthesized in a proenzyme form, are activated via proteolytic cascades (15). Their activation state is tightly regulated at the transcriptional and posttranscriptional levels by multiple mechanisms. The step that is stimulated by GPCRs has not been defined. EGFR transactivation has been shown to occur primarily through shedding of HB-EGF from the membrane (20, 22, 37). Although release of other ligands for the EGFR family of receptors has been observed (5, 26), cleavage of pro-EGF itself has been only rarely documented. Two cases have been previously reported in which pro-EGF is cleaved to activate EGFRs (14, 29). The article by Chen and colleagues (4a) provides an interesting new example of this alternative pathway.
The phenomenon of EGFR transactivation has been particularly well studied in carcinoma cells, where EGFRs are commonly overexpressed (1, 7, 35). Overexpression, as well as transactivation, of EGFRs plays a prominent role in regulating tumor growth and survival. EGFR transactivation has been shown to play a potential role in cancers, such as the prostate (20, 27), ovarian (22), colon (23), and breast (3). Transactivation of the EGFR results in intracellular signaling that leads to growth, proliferation, and migration of cancer cells (4, 6). Considerable work in this regard has been done in prostate cancer cells, where ligands, including bombesin, lysophosphatidic acid, endothelin, thrombin, and carbachol can transactivate the EGFR (27). In the PC-3 human prostate cancer cell line, binding of each of these ligands to its respective GPCR leads to MMP activation (12). The activated MMPs cleave pro-HB-EGF to initiate EGFR transactivation, leading to subsequent proliferation, migration, and metastasis (12, 20). Transactivation of the EGFR appears to be responsible for lysophosphatidic acid-induced Erk1/2 activation in PC-3 cells, based on experiments using EGFR inhibitors (12). In an interesting twist, a recent article describes an EGFR transactivation cascade in breast cancer cells that involves three ligands and receptors (34). In this model, estrogen stimulates release of sphingosine-1-phosphate, which then binds to a GPCR, which activates MMPs to stimulate EGFR transactivation. Thus studies in transformed cells have focused in some detail on the pro-mitogenic roles of EGFR transactivation.
EGFR transactivation has been studied in primary nontransformed cells to a limited extent. Cholinergic agonists can transactivate the EGFR in goblet cells (9). Angiotensin II transactivates the EGFR in normal prostate stromal cells (17). EGFR transactivation has been studied by several investigators in vascular smooth muscle cells. Angiotensin II transactivates the EGFR in these cells via activation of MMP2/9 (18). This transactivation leads to subsequent downstream signaling events that include Erk1/2 activation (36), Akt activation (36), and smooth muscle contraction (18). Downstream actions of ligands, including angiotensin II, heparin, and endothelin have been blocked in smooth muscle cells using EGFR and/or MMP inhibitors (8, 30, 36). Thus transactivation of the EGFR seems to be crucial for downstream mitogenic signaling mediated by some GPCRs. Transactivation may also be important for differentiation, as a recent report shows that knockout of the 5-HT2B receptor reduces expression of ErbB-2, leading to embryonic lethality and implying that the serotonin receptor acts via ErbB-2 transactivation (29). One would expect that additional examples of this phenomenon in primary cells, as described by Chen et al. (4a), will be forthcoming. The pathways involved are clearly not unique to transformed cells.
The role of EGFR transactivation in primary lacrimal gland epithelial cells is described by Chen and co-workers (4a). Tear secretions from the lacrimal gland are regulated by sympathetic and parasympathetic nerves within the gland. α1-Adrenergic agonists stimulate secretion (21). Previous work (25) also established that the EGFR is transactivated by the α1-adrenergic receptor (α1-AR) agonist phenylephrine in lacrimal gland acinar cells. The manuscript that is the subject of the focus article explores the role of protein secretion/shedding from the lacrimal gland and its regulation by EGFR transactivation. The authors show that the α1D-AR is responsible for peroxidase secretion and Erk1/2 activation (see Ref. 4a), as well as for the previously demonstrated EGFR transactivation (25). An MMP inhibitor (GM6001) and an ADAM-17 inhibitor (TAPI-1) both inhibit phenylephrine-induced Erk1/2 activation, and enhance peroxidase secretion. These results suggest that EGFR transactivation results in an Erk-mediated decrease in protein secretion, and provide insight into the MMPs that may be involved. Subsequent studies examined the EGFR ligand responsible for the response. The authors identified a 170-kDa precursor of EGF in the lacrimal gland, and showed that phenylephrine decreases the level of the precursor. Shedding of mature EGF was demonstrated, with EGF appearing in the ducts as well as the apical and basolateral membranes of the acini of the lacrimal gland. Medium harvested from stimulated lacrimal gland pieces activates EGFR phosphorylation and Erk activation in lacrimal gland acini, further confirming the shedding of EGF. Finally, an EGF neutralizing antibody decreases phenylephrine-stimulated Erk activation. Taken together, these data establish that mature EGF is released from lacrimal gland acinar cells in response to α1D-AR stimulation, and that the released EGF exerts an inhibitory effect on phenylephrine-induced protein secretion. Interestingly, the inhibitory effect on protein secretion appears to be mediated by the Erk pathway, which is typically observed to play a positive role in cell proliferation. Thus, EGFR transactivation represents a negative feedback loop with respect to GPCR-induced secretion.
The results presented by Chen and co-workers (4a) are unique in that they describe a transactivation response mediated by EGF itself, in a setting in which EGF exerts negative feedback on an acute response to a GCPR agonist. Several interesting questions remain with respect to receptor transactivation cascades in general. Is there signaling significance to the fact that some cells release HB-EGF, whereas others release EGF itself? Is GPCR-induced release of growth factors important only in an autocrine sense, or is there a paracrine role as well? Are there other instances in which the EGFR exerts a negative effect on GPCR signaling? Are pathways for growth factor receptor transactivation similar in tumor cells compared with nontransformed cells? What is the molecular mechanism whereby GPCRs stimulate MMP activation? Is EGFR transactivation solely responsible for the pro-mitogenic effects mediated by GPCRs, or is it a melding of direct GPCR signals with indirect EGFR signals that defines the complete mitogenic response? As Chen and colleagues point out, phenylephrine-induced Erk activation is not completely blocked by inhibitors of the EGF response, suggesting that other autocrine factors or pathways are involved in GPCR-initiated mitogenesis. The findings described in this issue suggest that we need to “think outside the box”, or at least outside the pro-mitogenic paradigm, in considering the roles of growth factor receptor transactivation in GPCR-mediated responses.
This work was supported by the Department of Defense Congressionally Directed Medical Research Program Grant DAMD17-01-0730.
We thank Andrea Meier for graphic design and Patricia Murphy for informational technology support.
↵1 This editorial focus should have been printed in the same issue as the article in focus by Chen et al. We apologize for the oversight.
- Copyright © 2007 the American Physiological Society