Collagen is an important extracellular matrix component that directs many fundamental cellular processes including differentiation, proliferation and motility. The signalling networks driving these processes are propagated by collagen receptors such as the β1 integrins and the DDRs (discoidin domain receptors). To gain an insight into the molecular mechanisms of collagen receptor signalling, we have performed a quantitative analysis of the phosphorylation networks downstream of collagen activation of integrins and DDR2. Temporal analysis over seven time points identified 424 phosphorylated proteins. Distinct DDR2 tyrosine phosphorylation sites displayed unique temporal activation profiles in agreement with in vitro kinase data. Multiple clustering analysis of the phosphoproteomic data revealed several DDR2 candidate downstream signalling nodes, including SHP-2 (Src homology 2 domain-containing protein tyrosine phosphatase 2), NCK1 (non-catalytic region of tyrosine kinase adaptor protein 1), LYN, SHIP-2 [SH2 (Src homology 2)-domain-containing inositol phosphatase 2], PIK3C2A (phosphatidylinositol-4-phosphate 3-kinase, catalytic subunit type 2α) and PLCL2 (phospholipase C-like 2). Biochemical validation showed that SHP-2 tyrosine phosphorylation is dependent on DDR2 kinase activity. Targeted proteomic profiling of a panel of lung SCC (squamous cell carcinoma) DDR2 mutants demonstrated that SHP-2 is tyrosine-phosphorylated by the L63V and G505S mutants. In contrast, the I638F kinase domain mutant exhibited diminished DDR2 and SHP-2 tyrosine phosphorylation levels which have an inverse relationship with clonogenic potential. Taken together, the results of the present study indicate that SHP-2 is a key signalling node downstream of the DDR2 receptor which may have therapeutic implications in a subset of DDR2 mutations recently uncovered in genome-wide lung SCC sequencing screens.
The present study characterizes integrin and DDR2 signalling networks activated by collagen. Using clustering approaches, DDR2-specific signalling components such as SHP-2 were identified. We further demonstrate that SHP-2 is phosphorylated by a subset of DDR2 lung cancer mutants.
cell signalling; collagen; discoidin domain receptor; lung cancer; mass spectrometry; phosphoproteomics; CDK1, cyclin-dependent kinase 1; DDR, discoidin domain receptor; DMEM, Dulbecco’s modified Eagle’s medium; DYRK1A, dual-specificity tyrosine-phosphorylation-regulated kinase 1A; EGFR, epidermal growth factor receptor; ERK, extracellular-signal-regulated kinase; EV, empty vector; GO, Gene Ontology; HEK, human embryonic kidney; HRP, horseradish peroxidase; IL, interleukin; IMAC, immobilized metal-ion-affinity chromatography; KD, kinase domain; mAb, monoclonal antibody; MCAM, multiple clustering analysis methodology; NCK1, non-catalytic region of tyrosine kinase adaptor protein 1; PIK3C2A, phosphatidylinositol-4-phosphate 3-kinase, catalytic subunit type 2α; PLCL2, phospholipase C-like 2; RFB, radiometric filter binding; RTK, receptor tyrosine kinase; SCC, squamous cell carcinoma; SHIP-2, SH2 (Src homology 2)-domain-containing inositol phosphatase 2; SHP-2, Src homology 2 domain-containing protein tyrosine phosphatase 2; SRM, selective reaction monitoring; TDA, template-directed assembly; TEAB, triethylammonium bicarbonate; TFA, trifluoroacetic acid
Almost all human cancers display dysregulated expression and/or function of one or more receptor tyrosine kinases (RTKs). The strong causative association between altered RTK function and cancer progression has translated into novel therapeutic strategies that target these cell surface receptors in the treatment of cancer. Yet, the full spectrum of RTKs that may alter the oncogenic process is not completely understood. Accumulating evidence suggests that a unique set of RTKs known as the Discoidin Domain Receptors (DDRs) play a role in cancer progression by regulating the interactions of tumor cells with their surrounding collagen matrix. The DDRs are the only RTKs that specifically bind to, and are activated by collagen. Hence, the DDRs are part of the signaling networks that translate information from the extracellular matrix thereby acting as key regulators of cell-matrix interactions. Under physiological conditions, DDRs control cell and tissue homeostasis by acting as collagen sensors, transducing signals that regulate cell polarity, tissue morphogenesis, and cell differentiation. In cancer, DDRs are hijacked by tumor cells to disrupt normal cell-matrix communication and initiate pro-migratory and pro-invasive programs. Importantly, several cancer types exhibit DDR mutations, which are thought to alter receptor function and contribute to cancer progression. Other evidence suggests that the actions of DDRs in cancer are complex, either promoting or suppressing tumor cell behavior in a DDR type/isoform specific and context dependent manner. Thus, there is still a considerable gap in our knowledge of DDR actions in cancer tissues. This review summarizes the current knowledge on DDR expression and function in cancer and discusses the potential implications of DDRs in cancer biology. It is hoped that this effort will encourage more research into these poorly understood but unique RTKs, which have the potential of becoming novel therapeutics targets in cancer.
discoidin domain receptor; tyrosine kinase; collagen; extracellular matrix; signaling; cell migration; metastasis
The discoidin domain receptors, DDR1 and DDR2, are receptor tyrosine kinases that bind to and are activated by collagens. Similar to collagen-binding β1 integrins, the DDRs bind to specific motifs within the collagen triple helix. However, these two types of collagen receptors recognize distinct collagen sequences. While GVMGFO (O is hydroxyproline) functions as a major DDR binding motif in fibrillar collagens, integrins bind to sequences containing Gxx’GEx”. The DDRs are thought to regulate cell adhesion, but their roles have hitherto only been studied indirectly. In this study we used synthetic triple-helical collagen-derived peptides that incorporate either the DDR-selective GVMGFO motif or integrin-selective motifs, such as GxOGER and GLOGEN, in order to selectively target either type of receptor and resolve their contributions to cell adhesion. Our data using HEK293 cells show that while cell adhesion to collagen I was completely inhibited by anti-integrin blocking antibodies, the DDRs could mediate cell attachment to the GVMGFO motif in an integrin-independent manner. Cell binding to GVMGFO was independent of DDR receptor signalling and occurred with limited cell spreading, indicating that the DDRs do not mediate firm adhesion. However, blocking the interaction of DDR-expressing cells with collagen I via the GVMGFO site diminished cell adhesion, suggesting that the DDRs positively modulate integrin-mediated cell adhesion. Indeed, overexpression of the DDRs or activation of the DDRs by the GVMGFO ligand promoted α1β1 and α2β1 integrin-mediated cell adhesion to medium- and low-affinity integrin ligands without regulating the cell surface expression levels of α1β1 or α2β1. Our data thus demonstrate an adhesion-promoting role of the DDRs, whereby overexpression and/or activation of the DDRs leads to enhanced integrin-mediated cell adhesion as a result of higher integrin activation state.
The discoidin domain receptors, DDR1 and DDR2, are constitutively dimeric receptor tyrosine kinases that are activated by triple-helical collagen. Aberrant DDR signaling contributes to several human pathologies, including many cancers. We have generated monoclonal antibodies (mAbs) that inhibit DDR1 signaling without interfering with collagen binding. The crystal structure of the monomeric DDR1 extracellular region bound to the Fab fragment of mAb 3E3 reveals that the collagen-binding discoidin (DS) domain is tightly associated with the following DS-like domain, which contains the epitopes of all mAbs. A conserved surface patch in the DS domain outside the collagen-binding site is shown to be required for signaling. Thus, the active conformation of the DDR1 dimer involves collagen-induced contacts between the DS domains, in addition to the previously identified association of transmembrane helices. The mAbs likely inhibit signaling by sterically blocking the extracellular association of DDR1 subunits.
► Monoclonal antibodies inhibit DDR1 signaling without blocking collagen binding ► The DDR1 extracellular region consists of a DS and a DS-like domain ► The collagen-binding DS domain contains a patch that is essential for signaling ► The mAbs bind to the DS-like domain, preventing formation of the active DDR dimer
Collective cell migration occurs in a range of contexts: cancer cells frequently invade in cohorts while retaining cell-cell junctions. Here we show that collective cancer cell invasion depends on reducing actomyosin contractility at sites of cell-cell contact. When actomyosin is not down-regulated at cell-cell contacts migrating cells lose cohesion. We provide a novel molecular mechanism for this down-regulation. Depletion of Discoidin Domain Receptor 1 (DDR1) blocks collective cancer cell invasion in a range of 2D, 3D and ‘organotypic’ models. DDR1 co-ordinates the Par3/6 cell polarity complex through its C-terminus binding PDZ domains in Par3 and Par6. The DDR1/Par3/6 complex controls the localisation of RhoE to cell-cell contacts where it antagonizes ROCK-driven actomyosin contractility. Depletion of DDR1, Par3, Par6 or RhoE leads to increased actomyosin at cell-cell contacts, a loss of cell-cell cohesion and defective collective cell invasion.
The discoidin domain receptors, DDR1 and DDR2 are cell surface receptor tyrosine kinases that are activated by triple-helical collagen. While normal DDR signalling regulates fundamental cellular processes, aberrant DDR signalling is associated with several human diseases. We previously identified GVMGFO (O is hydroxyproline) as a major DDR2 binding site in collagens I–III, and located two additional DDR2 binding sites in collagen II. Here we extend these studies to the homologous DDR1 and the identification of DDR binding sites on collagen III. Using sets of overlapping triple-helical peptides, the Collagen II and Collagen III Toolkits, we located several DDR2 binding sites on both collagens. The interaction of DDR1 with Toolkit peptides was more restricted, with DDR1 mainly binding to peptides containing the GVMGFO motif. Triple-helical peptides containing the GVMGFO motif induced DDR1 transmembrane signalling, and DDR1 binding and receptor activation occurred with the same amino acid requirements as previously defined for DDR2. While both DDRs exhibit the same specificity for binding the GVMGFO motif, which is present only in fibrillar collagens, the two receptors display distinct preferences for certain non-fibrillar collagens, with the basement membrane collagen IV being exclusively recognised by DDR1. Based on our recent crystal structure of a DDR2-collagen complex, we designed mutations to identify the molecular determinants for DDR1 binding to collagen IV. By replacing five amino acids in DDR2 with the corresponding DDR1 residues we were able to create a DDR2 construct that could function as a collagen IV receptor.
DDR, discoidin domain receptor; DS, discoidin homology; HEK, human embryonic kidney; RTK, receptor tyrosine kinase; VWF, von Willebrand factor; Collagen receptor; Discoidin domain receptor; Receptor tyrosine kinase; Cell–extracellular matrix interaction; Collagen binding specificity
Spondylo-meta-epiphyseal dysplasia (SMED) with short limbs and abnormal calcifications (SMED-SL) is a rare, autosomal recessive human growth disorder, characterized by disproportionate short stature, short limbs, short broad fingers, abnormal metaphyses and epiphyses, platyspondyly and premature calcifications. Recently, three missense mutations and one splice-site mutation in the DDR2 gene were identified as causative genetic defects for SMED-SL, but the underlying cellular and biochemical mechanisms were not explored. Here we report a novel DDR2 missense mutation, c.337G>A (p.E113K), that causes SMED-SL in two siblings in the United Arab Emirates. Another DDR2 missense mutation, c.2254C>T (p.R752C), matching one of the previously reported SMED-SL mutations, was found in a second affected family. DDR2 is a plasma membrane receptor tyrosine kinase that functions as a collagen receptor. We expressed DDR2 constructs with the identified point mutations in human cell lines and evaluated their localization and functional properties. We found that all SMED-SL missense mutants were defective in collagen-induced receptor activation and that the three previously reported mutants (p.T713I, p.I726R and p.R752C) were retained in the endoplasmic reticulum. The novel mutant (p.E113K), in contrast, trafficked normally, like wild-type DDR2, but failed to bind collagen. This finding is in agreement with our recent structural data identifying Glu113 as an important amino acid in the DDR2 ligand-binding site. Our data thus demonstrate that SMED-SL can result from at least two different loss-of-function mechanisms: namely defects in DDR2 targeting to the plasma membrane or the loss of its ligand-binding activity.
The discoidin domain receptors, DDR1 and DDR2, are widely expressed receptor tyrosine kinases that are activated by triple-helical collagen. They control important aspects of cell behavior and are dysregulated in several human diseases. The major DDR2-binding site in collagens I–III is a GVMGFO motif (O is hydroxyproline) that also binds the matricellular protein SPARC. We have determined the crystal structure of the discoidin domain of human DDR2 bound to a triple-helical collagen peptide. The GVMGFO motifs of two collagen chains are recognized by an amphiphilic pocket delimited by a functionally critical tryptophan residue and a buried salt bridge. Collagen binding results in structural changes of DDR2 surface loops that may be linked to the process of receptor activation. A comparison of the GVMGFO-binding sites of DDR2 and SPARC reveals a striking case of convergent evolution in collagen recognition.
The adhesion receptors known as integrins perform key functions for hematopoietic cells. The platelet integrin αIIbβ3 is critical in hemostasis, and the β1 and β2 integrins on leukocytes have many roles in cell-mediated immunity. Mutations in the β2 subunit lead to integrin nonexpression and to an immune deficiency, leukocyte adhesion deficiency-1. Mutations in either the α or β subunit of αIIbβ3 usually lead to integrin nonexpression and a bleeding tendency termed Glanzmann thrombasthenia. Here we describe a unique patient with clinical features of both Glanzmann thrombasthenia and leukocyte adhesion deficiency-1. The patient has normal expression of β1, β2, and β3 integrins, but all are dysfunctional. The key findings are that “inside-out” signaling pathways leading to integrin activation are defective and that this is associated with abnormal integrin clustering. The integrins themselves are intact and capable of function following extracellular stimulation. T cell motility is normal, as are the expression levels and electrophoretic characteristics of all cytoskeletal and signaling proteins tested, except PKC-α, which has enhanced expression in the patient’s cells. To our knowledge, this is the first description of a dysfunction affecting three classes of integrins. We propose that it is caused by a lesion in an intracellular factor or signaling pathway essential for integrin activation in hematopoietic cells and results in lack of regulation of clustering, an essential component of integrin-mediated adhesion.
A subset of integrin α subunits contain an I domain, which is important for ligand binding. We have deleted the I domain from the β2 integrin lymphocyte function-asssociated antigen-1 (LFA-1) and expressed the resulting non–I domain-containing integrin (ΔI-LFA-1) in an LFA-1-deficient T cell line. ΔI-LFA-1 showed no recognition of LFA-1 ligands, confirming the essential role of the I domain in ligand binding. Except for I domain monoclonal antibodies (mAbs), ΔI-LFA-1 was recognized by a panel of anti-LFA-1 mAbs similarly to wild-type LFA-1. However, ΔI-LFA-1 had enhanced expression of seven mAb epitopes that are associated with β2 integrin activation, suggesting that it exhibited an “active” conformation. In keeping with this characteristic, ΔI-LFA-1 induced constitutive activation of α4β1 and α5β1, suggesting intracellular signaling to these integrins. This “cross-talk” was not due to an effect on β1 integrin affinity. However, the enhanced activity was susceptible to inhibition by cytochalasin D, indicating a role for the cytoskeleton, and also correlated with clustering of β1 integrins. Thus, removal of the I domain from LFA-1 created an integrin with the hallmarks of a constitutively active receptor mediating signals into the cell. These findings suggest a key role for the I domain in controlling integrin activity.