Crosstalk between TGF-β and EGF receptors via direct phosphorylation

Signal transduction pathways execute the physiological communication between cells. Signaling can control developmental processes, homeostatic adaptation, and maintenance of tissue integrity. Consequently, misregulation of signal transduction characterizes human diseases, including cancer (1). Among many signaling pathways, the epidermal growth factor (EGF) family provides positive signals for cell proliferation and migration. The transforming growth factor β (TGF-β) family provides signals maintaining homeostasis by limiting cell proliferation, inhibiting cell differentiation but also promoting cell migration (2, 3).

Linked to tumorigenesis is the genetic alteration of key signaling components, of which the EGF family receptor HER2/ERBB2 is a classic case. Mutant, hyperactive HER2 tyrosine kinase characterizes various epithelial tumors, e.g., breast, lung, and liver cancer (2). Despite our deep understanding of HER2 and its utility as a target of pharmacological treatments of cancer patients, key elements of its function remain unsolved, especially those elements that may explain why breast cancer patients with hyperactive HER2 signaling develop resistance to anti-HER2 treatment modalities (2).

It is accordingly relevant to understand basic mechanisms of signal transduction crosstalk that contribute to disease progression and resistance to clinical treatment. A classic case of cooperative signaling in cancer progression is the EGF/HER2 and TGF-β crosstalk (4) that has been associated with breast cancer survival, migratory behavior, and metastatic potential, as demonstrated in studies with double (HER2/Neu and TGF-β/TGF-β receptor) transgenic mouse models (5, 6). These mouse studies were among the first that confirmed the ability of TGF-β to suppress the pre-malignant stage of the disease and promote invasive and metastatic development, once malignancy has been established.

Mechanisms by which the HER2-TGF-β crosstalk can be manifested in breast cancer cells have been studied and cover diverse cell biological processes. A simplified model could propose that HER2, in response of EGF ligands or due to its oncogenic mutations, enhances one signaling pathway, e.g., mitogen-activated protein kinase (MAPK) activity. TGF-β, via its serine/threonine kinase receptors, activates a distinct signal, most commonly through SMAD proteins. Crosstalk should entail an interaction of one signaling protein with, or phosphorylation from, the other. Such an elegant mechanism has been demonstrated earlier by the group of Ye-Guang Chen. HER2 activates the Akt protein kinase that phosphorylates SMAD3, thus directing TGF-β signaling toward the transcriptional regulation of genes that promote breast cancer cell migration (7). Following the same mechanistic logic, strong signals from EGF receptors guide SMAD proteins to induce expression of the AP-1 transcription factors, which then transcriptionally induce many genes involved in breast cancer cell migration (8). Alternatively, cooperating HER2 and TGF-β signaling induce expression of cytokines that are secreted and then promote the migratory response to the same cells in an auto-/paracrine manner (4). Via a distinct pathway, HER2 signaling facilitates mRNA translation of a specific isoform of the C/EBPβ transcription factor, which inhibits physiological tumor suppressive gene responses to TGF-β signaling (9).

Turning to the signaling receptors themselves, uncharacterized TGF-β signaling mechanisms cause clustering of HER2 in lamellipodia and facilitate actin cytoskeleton-supported breast cancer cell migration (10). Alternatively, by activating the Src kinase, TGF-β induces HER2 and integrin receptor co-clusters to facilitate migration (11). These mechanisms involve Rac and Cdc42 family small GTPase activation. An intermediate activator of these enzymes can be the Cdc42 GTPase activating protein CdGAP, whose expression is transcriptionally induced by TGF-β/SMAD signaling (12). Yet none of these mechanisms provide insight into direct signaling input generated by either the TGF-β or the HER2 receptors.

This important question is addressed by work in this issue from the group of Ye-Guang Chen (13). The novelty of the work that revisits the well-established crosstalk between HER2 and TGF-β receptors is twofold: (i) Although many reports have previously claimed that TGF-β receptors can activate Akt and MAPK signaling (reviewed in 3), Shi and colleagues now demonstrate that such activation requires signaling by the HER2 receptor. (ii) The beauty of this mechanism is that it is direct. It entails phosphorylation by the TGF-β type I receptor kinase of Ser779 of the HER2 receptor. Ser779 resides in the short αC-β4 loop of the HER2 kinase domain that auto-inhibits the kinase. Upon its phosphorylation, the αC-β4 loop may either change conformation or facilitate receptor dimerization in order to activate HER2 tyrosine kinase activity. The latter model deserves future structural and molecular modeling analysis.

Elegantly shown, this mechanism regulates physiological development and maturation of the mouse mammary gland since a Cre-mediated knock-in transgenic animal that expresses the HER2 Ser779Ala mutant presents defective mammary alveolar tree morphogenesis. Furthermore, the mechanism is relevant to breast cancer: (i) Reconstitution of the mutant HER2 Ser779Ala mutant in human breast cancer cells that lack endogenous HER2 expression inhibits mesenchymal trans-differentiation, migration of the cancer cells induced by TGF-β, and metastatic dissemination to lungs. (ii) A novel antibody that recognizes HER2 phosphorylated on Ser779 revealed abundant phospho-HER2 levels in tumor specimen of women with breast cancer after quantitative immunoblot analysis. Interestingly, but a bit difficult to comprehend based on the current model, breast cancer patients with mutations on the very same Ser779 are discussed by the authors. Whether such mutations activate HER2 or are selected in patients in order to divert TGF-β signaling during tumorigenesis remains to be tested.

The new evidence establishes HER2 as a receptor acting “downstream” from the TGF-β receptors. Since phosphorylation of a protein by a kinase depends on their interaction, the authors demonstrate this elegantly using in vitro assays with recombinant proteins, as a complement to the plethora of cell-based assays. The phosphorylation mechanism was uncovered after mass spectrometric analysis of HER2 phospho-peptides extracted from cells, in which the signaling of TGF-β was activated. The signaling mechanism was confirmed in independent breast cancer cell models, normal mammary epithelial cells, and even human or mouse skin, liver, lung, and cervical cell models. In addition, the phenotypic analysis of mice expressing HER2 Ser779Ala in their mammary glands is complemented with ex vivo primary culture of mammary organoids, in which expression of HER2 Ser779Ala impaired branching morphogenesis.

Shi et al. (13) provide a fresh mechanism that directly links the TGF-β receptor to HER2 and thus provides a robust signaling link to the previously reported pathway crosstalk. Does this mechanism also explain direct participation of TGF-β receptors in the above examples of HER2-integrin clustering and small GTPase signaling at the leading edge of migratory cancer cells? Does HER2 also phosphorylate the TGF-β receptors on specific tyrosines, thus mediating additional control of the signaling crosstalk? From a TGF-β signaling perspective, the paper provides strong evidence for a new mechanism by which the type I receptor can signal. By incorporating HER2 as a novel substrate of the TGF-β receptor, the textbook mechanism of SMAD phosphorylation by the same type I receptor kinase is enriched significantly. From a HER2 perspective, the Ser779 phosphorylation provides new structural confirmation of the key regulatory function of the αC-β4 loop, and the new antibody that recognizes this site provides a valuable tool for cell biological but also diagnostic studies in tissues from patients with many forms of cancer.