FGF9, a Potent Mitogen, Is a New Ligand for Integrin αvβ3, and the FGF9 Mutant Defective in Integrin Binding Acts as an Antagonist

FGF9 is a potent mitogen and survival factor, but FGF9 protein levels are generally low and restricted to a few adult organs. Aberrant expression of FGF9 usually results in cancer. However, the mechanism of FGF9 action has not been fully established. Previous studies showed that FGF1 and FGF2 directly bind to integrin αvβ3, and this interaction is critical for signaling functions (FGF-integrin crosstalk). FGF1 and FGF2 mutants defective in integrin binding were defective in signaling, whereas the mutants still bound to FGFR suppressed angiogenesis and tumor growth, indicating that they act as antagonists. We hypothesize that FGF9 requires direct integrin binding for signaling. Here, we show that docking simulation of the interaction between FGF9 and αvβ3 predicted that FGF9 binds to the classical ligand-binding site of αvβ3. We show that FGF9 bound to integrin αvβ3 and generated FGF9 mutants in the predicted integrin-binding interface. An FGF9 mutant (R108E) was defective in integrin binding, activating FRS2α and ERK1/2, inducing DNA synthesis, cancer cell migration, and invasion in vitro. R108E suppressed DNA synthesis and activation of FRS2α and ERK1/2 induced by WT FGF9 (dominant-negative effect). These findings indicate that FGF9 requires direct integrin binding for signaling and that R108E has potential as an antagonist to FGF9 signaling.

FGF9, a potent mitogen, is a new ligand for integrin αvβ3, and the FGF9 mutant defective in integrin binding acts as an antagonist

FGF9 is a potent mitogen and survival factor, but FGF9 protein level is generally low and restricted to a few adult organs. Aberrant expression of FGF9 usually results in cancer. However, the mechanism of FGF9 action has not been fully established. Previous studies showed that FGF1 and FGF2 directly bind to integrin αvβ3 and this interaction is critical for signaling functions (FGF-integrin crosstalk). FGF1 and FGF2 mutants defective in integrin binding were defective in signaling, whereas the mutants still bound to FGFR, and suppressed angiogenesis and tumor growth, indicating that they act as antagonists. We hypothesize that FGF9 requires direct integrin binding for signaling. Here we show that docking simulation of interaction between FGF9 and αvβ3 predicted that FGF9 binds to the classical ligand-binding site of αvβ3. We showed that FGF9 actually bound to integrin αvβ3, and generated an FGF9 mutants in the predicted integrin-binding interface. An FGF9 mutant (R108E) was defective in integrin binding, activating FRS2α and ERK1/2, inducing DNA synthesis, cancer cell migration, and invasion in vitro. R108E suppressed DNA synthesis induced by WT FGF9 and suppressed DNA synthesis and activation of FRS2α and ERK1/2 induced by WT FGF9 (dominant-negative effect). These findings indicate that FGF9 requires direct integrin binding for signaling and that R108E has potential as an antagonist to FGF9 signaling.

The heparin-binding domain of VEGF165 directly binds to integrin αvβ3 and plays a critical role in signaling

VEGF-A is a key cytokine in tumor angiogenesis and a major therapeutic target for cancer. VEGF165 is the predominant isoform and is the most potent angiogenesis stimulant. VEGFR2/KDR domains 2 and 3 (D2D3) bind to the N-terminal domain (NTD, residues 1-110) of VEGF165. Since removal of the heparin-binding domain (HBD, residues 111-165) markedly reduced the mitogenic activity of VEGF165, it has been proposed that the HBD plays a critical role in the mitogenicity of VEGF165. Integrin αvβ3 has been shown to bind to VEGF165, but the role of integrin αvβ3 in VEGF165 signaling are unclear. Here we describe that αvβ3 specifically bound to the isolated HBD, but not to the NTD. We identified several critical amino acid residues in HBD for integrin binding (Arg-123, Arg-124, Lys-125, Lys-140, Arg-145, and Arg-149) by docking simulation and mutagenesis, and generated full-length VEGF165 that is defective in integrin binding by including mutations in the HBD. The full-length VEGF165 mutant defective in integrin binding (R123A/R124A/K125A/K140A/R145A/R149A) was defective in ERK1/2 phosphorylation, integrin β3 phosphorylation, and KDR phosphorylation, although the mutation did not affect KDR binding to VEGF165. We propose a model in which VEGF165 induces KDR (through NTD)-VEGF165 (through HBD)-integrin αvβ3 ternary complex formation on the cell surface and this process is critically involved in potent mitogenicity of VEGF165.

Virtual Screening of Protein Data Bank via Docking Simulation Identified the Role of Integrins in Growth Factor Signaling, the Allosteric Activation of Integrins, and P-Selectin as a New Integrin Ligand

Integrins were originally identified as receptors for extracellular matrix (ECM) and cell-surface molecules (e.g., VCAM-1 and ICAM-1). Later, we discovered that many soluble growth factors/cytokines bind to integrins and play a critical role in growth factor/cytokine signaling (growth factor-integrin crosstalk). We performed a virtual screening of protein data bank (PDB) using docking simulations with the integrin headpiece as a target. We showed that several growth factors (e.g., FGF1 and IGF1) induce a integrin-growth factor-cognate receptor ternary complex on the surface. Growth factor/cytokine mutants defective in integrin binding were defective in signaling functions and act as antagonists of growth factor signaling. Unexpectedly, several growth factor/cytokines activated integrins by binding to the allosteric site (site 2) in the integrin headpiece, which is distinct from the classical ligand (RGD)-binding site (site 1). Since 25-hydroxycholesterol, a major inflammatory mediator, binds to site 2, activates integrins, and induces inflammatory signaling (e.g., IL-6 and TNFα secretion), it has been proposed that site 2 is involved in inflammatory signaling. We showed that several inflammatory factors (CX3CL1, CXCL12, CCL5, sPLA2-IIA, and P-selectin) bind to site 2 and activate integrins. We propose that site 2 is involved in the pro-inflammatory action of these proteins and a potential therapeutic target. It has been well-established that platelet integrin αIIbβ3 is activated by signals from the inside of platelets induced by platelet agonists (inside-out signaling). In addition to the canonical inside-out signaling, we showed that αIIbβ3 can be allosterically activated by inflammatory cytokines/chemokines that are stored in platelet granules (e.g., CCL5, CXCL12) in the absence of inside-out signaling (e.g., soluble integrins in cell-free conditions). Thus, the allosteric activation may be involved in αIIbβ3 activation, platelet aggregation, and thrombosis. Inhibitory chemokine PF4 (CXCL4) binds to site 2 but did not activate integrins, Unexpectedly, we found that PF4/anti-PF4 complex was able to activate integrins, indicating that the anti-PF4 antibody changed the phenotype of PF4 from inhibitory to inflammatory. Since autoantibodies to PF4 are detected in vaccine-induced thrombocytopenic thrombosis (VIPP) and autoimmune diseases (e.g., SLE, and rheumatoid arthritis), we propose that this phenomenon is related to the pathogenesis of these diseases. P-selectin is known to bind exclusively to glycans (e.g., sLex) and involved in cell-cell interaction by binding to PSGL-1 (CD62P glycoprotein ligand-1). Unexpectedly, through docking simulation, we discovered that the P-selectin C-type lectin domain functions as an integrin ligand. It is interesting that no one has studied whether P-selectin binds to integrins in the last few decades. The integrin-binding site and glycan-binding site were close but distinct. Also, P-selectin lectin domain bound to site 2 and allosterically activated integrins.

CD40L Activates Platelet Integrin αIIbβ3 by Binding to the Allosteric Site (Site 2) in a KGD-Independent Manner and HIGM1 Mutations Are Clustered in the Integrin-Binding Sites of CD40L

CD40L is expressed in activated T cells, and it plays a major role in immune response and is a major therapeutic target for inflammation. High IgM syndrome type 1 (HIGM1) is a congenital functional defect in CD40L/CD40 signaling due to defective CD40L. CD40L is also stored in platelet granules and transported to the surface upon platelet activation. Platelet integrin αIIbβ3 is known to bind to fibrinogen and activation of αIIbβ3 is a key event that triggers platelet aggregation. Also, the KGD motif is critical for αIIbβ3 binding and the interaction stabilizes thrombus. Previous studies showed that CD40L binds to and activates integrins αvβ3 and α5β1 and that HIGM1 mutations are clustered in the integrin-binding sites. However, the specifics of CD40L binding to αIIbβ3 were unclear. Here, we show that CD40L binds to αIIbβ3 in a KGD-independent manner using CD40L that lacks the KGD motif. Two HIGM1 mutants, S128E/E129G and L155P, reduced the binding of CD40L to the classical ligand-binding site (site 1) of αIIbβ3, indicating that αIIbβ3 binds to the outer surface of CD40L trimer. Also, CD40L bound to the allosteric site (site 2) of αIIbβ3 and allosterically activated αIIbβ3 without inside-out signaling. Two HIMG1 mutants, K143T and G144E, on the surface of trimeric CD40L suppressed CD40L-induced αIIbβ3 activation. These findings suggest that CD40L binds to αIIbβ3 in a manner different from that of αvβ3 and α5β1 and induces αIIbβ3 activation. HIGM1 mutations are clustered in αIIbβ3 binding sites in CD40L and are predicted to suppress thrombus formation and immune responses through αIIbβ3.

The C-type lectin domain of CD62P (P-selectin) functions as an integrin ligand

Recognition of integrins by CD62P has not been reported and this motivated a docking simulation using integrin αvβ3 as a target. We predicted that the C-type lectin domain of CD62P functions as a potential integrin ligand and observed that it specifically bound to soluble β3 and β1 integrins. Known inhibitors of the interaction between CD62P-PSGL-1 did not suppress the binding, whereas the disintegrin domain of ADAM-15, a known integrin ligand, suppressed recognition by the lectin domain. Furthermore, an R16E/K17E mutation in the predicted integrin-binding interface located outside of the glycan-binding site within the lectin domain, strongly inhibited CD62P binding to integrins. In contrast, the E88D mutation that strongly disrupts glycan binding only slightly affected CD62P-integrin recognition, indicating that the glycan and integrin-binding sites are distinct. Notably, the lectin domain allosterically activated integrins by binding to the allosteric site 2. We conclude that CD62P-integrin binding may function to promote a diverse set of cell-cell adhesive interactions given that β3 and β1 integrins are more widely expressed than PSGL-1 that is limited to leukocytes.

Pro-Inflammatory Chemokines CCL5, CXCL12, and CX3CL1 Bind to and Activate Platelet Integrin αIIbβ3 in an Allosteric Manner

Activation of platelet integrin αIIbβ3, a key event for hemostasis and thrombus formation, is known to be mediated exclusively by inside-out signaling. We showed that inflammatory chemokines CX3CL1 and CXCL12 in previous studies, and CCL5 in this study, bound to the allosteric binding site (site 2) of vascular integrin αvβ3, in addition to the classical ligand binding site (site 1), and allosterically activated integrins independent of inside-out signaling. Since αIIbβ3 is exposed to inflammatory chemokines at increased concentrations during inflammation (e.g., cytokine/chemokine storm) and platelet activation, we hypothesized that these chemokines bind to and activate αIIbβ3 in an allosteric activation mechanism. We found that these chemokines bound to αIIbβ3. Notably, they activated soluble αIIbβ3 in 1 mM Ca²⁺ by binding to site 2. They activated cell-surface αIIbβ3 on CHO cells, which lack machinery for inside-out signaling or chemokine receptors, quickly (<1 min) and at low concentrations (1–10 ng/mL) compared to activation of soluble αIIbβ3, probably because chemokines bind to cell surface proteoglycans. Furthermore, activation of αIIbβ3 by the chemokines was several times more potent than 1 mM Mn²⁺. We propose that CCL5 and CXCL12 (stored in platelet granules) may allosterically activate αIIbβ3 upon platelet activation and trigger platelet aggregation. Transmembrane CX3CL1 on activated endothelial cells may mediate platelet–endothelial interaction by binding to and activating αIIbβ3. Additionally, these chemokines in circulation over-produced during inflammation may trigger αIIbβ3 activation, which is a possible missing link between inflammation and thrombosis.

Integrin Binding to the Trimeric Interface of CD40L Plays a Critical Role in CD40/CD40L Signaling

CD40L plays a major role in immune response and is a major therapeutic target for inflammation. Integrin α5β1 and CD40 simultaneously bind to CD40L. It is unclear if α5β1 and CD40 work together in CD40/CD40L signaling or how α5β1 binds to CD40L. In this article, we describe that the integrin-binding site of human CD40L is predicted to be located in the trimeric interface by docking simulation. Mutations in the predicted integrin-binding site markedly reduced the binding of α5β1 to CD40L. Several CD40L mutants defective in integrin binding were defective in NF-κB activation and B cell activation and suppressed CD40L signaling induced by wild-type CD40L; however, they still bound to CD40. These findings suggest that integrin α5β1 binds to monomeric CD40L through the binding site in the trimeric interface of CD40L, and this plays a critical role in CD40/CD40L signaling. Integrin αvβ3, a widely distributed vascular integrin, bound to CD40L in a KGD-independent manner, suggesting that αvβ3 is a new CD40L receptor. Several missense mutations in CD40L that induce immunodeficiency with hyper-IgM syndrome type 1 (HIGM1) are clustered in the integrin-binding site of the trimeric interface. These HIGM1 CD40L mutants were defective in binding to α5β1 and αvβ3 (but not to CD40), suggesting that the defect in integrin binding may be a causal factor of HIGM1. These findings suggest that α5β1 and αvβ3 bind to the overlapping binding site in the trimeric interface of monomeric CD40L and generate integrin-CD40L-CD40 ternary complex. CD40L mutants defective in integrins have potential as antagonists of CD40/CD40L signaling.

Stromal cell-derived factor-1 (CXCL12) activates integrins by direct binding to an allosteric ligand-binding site (site 2) of integrins without CXCR4

Leukocyte arrest on the endothelial cell surface during leukocyte extravasation is induced by rapid integrin activation by chemokines. We recently reported that fractalkine induces integrin activation without its receptor CX3CR1 through binding to the allosteric site (site 2) of integrins. Peptides from site 2 bound to fractalkine and suppressed integrin activation by fractalkine. We hypothesized that this is not limited to membrane-bound fractalkine. We studied if stromal cell-derived factor 1 (SDF1), another chemokine that plays a critical role in leukocyte arrest, activates integrins through binding to site 2. We describe here that (1) SDF1 activated soluble integrin αvβ3 in cell-free conditions, suggesting that SDF1 can activate αvβ3 without CXCR4; (2) site 2 peptide bound to SDF1, suggesting that SDF1 binds to site 2; (3) SDF1 activated integrins αvβ3, α4β1, and α5β1 on CHO cells (CXCR4-negative) and site 2 peptide suppressed the activation; (4) A CXCR4 antagonist AMD3100 or pertussis toxin did not affect the site 2-mediated integrin activation by SDF1; (5) Cell-surface integrins were fully activated in 1 min (much faster than activation of soluble αvβ3) and the activation lasted at least for 1 h. We propose that the binding of SDF1 to cell-surface proteoglycan facilitates the allosteric activation process; (6) Mutations in the predicted site 2-binding site in SDF1 suppressed integrin activation. These results suggest that SDF1 (e.g., presented on proteoglycans) can rapidly activate integrins in an allosteric manner by binding to site 2 in the absence of CXCR4 or other GPCR. The allosteric integrin activation by SDF1 is a novel target for drug discovery.

Direct binding to integrins and loss of disulfide linkage in interleukin-1β (IL-1β) are involved in the agonistic action of IL-1β

There is a strong link between integrins and interleukin-1β (IL-1β), but the specifics of the role of integrins in IL-1β signaling are unclear. We describe that IL-1β specifically bound to integrins αvβ3 and α5β1. The E128K mutation in the IL1R-binding site enhanced integrin binding. We studied whether direct integrin binding is involved in IL-1β signaling. We compared sequences of IL-1β and IL-1 receptor antagonist (IL1RN), which is an IL-1β homologue but has no agonistic activity. Several surface-exposed Lys residues are present in IL-1β, but not in IL1RN. A disulfide linkage is present in IL1RN, but is not in IL-1β because of natural C117F mutation. Substitution of the Lys residues to Glu markedly reduced integrin binding of E128K IL-1β, suggesting that the Lys residues mediate integrin binding. The Lys mutations reduced, but did not completely abrogate, agonistic action of IL-1β. We studied whether the disulfide linkage plays a role in agonistic action of IL-1β. Reintroduction of the disulfide linkage by the F117C mutation did not affect agonistic activity of WT IL-1β, but effectively reduced the remaining agonistic activity of the Lys mutants. Also, deletion of the disulfide linkage in IL1RN by the C116F mutation did not make it agonistic. We propose that the direct binding to IL-1β to integrins is primarily important for agonistic IL-1β signaling, and that the disulfide linkage indirectly affects signaling by blocking conformational changes induced by weak integrin binding to the Lys mutants. The integrin-IL-1β interaction is a potential target for drug discovery.

Crosstalk between insulin-like growth factor (IGF) receptor and integrins through direct integrin binding to IGF1

It has been generally accepted that integrin cell adhesion receptors are involved in growth factor signaling (integrin-growth factor crosstalk), since antagonists to integrins often suppress growth factor signaling. Partly because integrins have been originally identified as cell adhesion receptors to extracellular matrix (ECM) proteins, current models of the crosstalk between IGF1 and integrins propose that ECM ligands (e.g., vitronectin) bind to integrins and IGF1 binds to IGF receptor type 1 (IGF1R), and two separate signals merge inside the cells. Our research proves otherwise. We discovered that IGF1 interacts directly with integrins, and induces integrin-IGF-IGF1R complex formation on the cell surface. IGF1 signaling can be detected in the absence of ECM (anchorage-independent conditions). Integrin antagonists block both ECM-integrin interaction and IGF-integrin interaction, and do not distinguish the two. This is one possible reason why integrin-IGF1 interaction has not been detected. With these new discoveries, we believe that the direct IGF-integrin interaction should be incorporated into models of IGF1 signaling. The integrin-binding defective mutant of IGF1 is defective in inducing IGF signaling, although the mutant still binds to IGF1R. Notably, the IGF1 mutant is dominant-negative and suppresses cell proliferation induced by wt IGF1, and suppresses tumorigenesis in vivo, and thus the IGF1 mutant has potential as a therapeutic.

Proinflammatory secreted phospholipase A2 type IIA (sPLA-IIA) induces integrin activation through direct binding to a newly identified binding site (site 2) in integrins αvβ3, α4β1, and α5β1

Integrins are activated by signaling from inside the cell (inside-out signaling) through global conformational changes of integrins. We recently discovered that fractalkine activates integrins in the absence of CX3CR1 through the direct binding of fractalkine to a ligand-binding site in the integrin headpiece (site 2) that is distinct from the classical RGD-binding site (site 1). We propose that fractalkine binding to the newly identified site 2 induces activation of site 1 though conformational changes (in an allosteric mechanism). We reasoned that site 2-mediated activation of integrins is not limited to fractalkine. Human secreted phospholipase A2 type IIA (sPLA2-IIA), a proinflammatory protein, binds to integrins αvβ3 and α4β1 (site 1), and this interaction initiates a signaling pathway that leads to cell proliferation and inflammation. Human sPLA2-IIA does not bind to M-type receptor very well. Here we describe that sPLA2-IIA directly activated purified soluble integrin αvβ3 and transmembrane αvβ3 on the cell surface. This activation did not require catalytic activity or M-type receptor. Docking simulation predicted that sPLA2-IIA binds to site 2 in the closed-headpiece of αvβ3. A peptide from site 2 of integrin β1 specifically bound to sPLA2-IIA and suppressed sPLA2-IIA-induced integrin activation. This suggests that sPLA2-IIA activates αvβ3 through binding to site 2. sPLA2-IIA also activated integrins α4β1 and α5β1 in a site 2-mediated manner. We recently identified small compounds that bind to sPLA2-IIA and suppress integrin-sPLA2-IIA interaction (e.g. compound 21 (Cmpd21)). Cmpd21 effectively suppressed sPLA2-IIA-induced integrin activation. These results define a novel mechanism of proinflammatory action of sPLA2-IIA through integrin activation.

The chemokine fractalkine can activate integrins without CX3CR1 through direct binding to a ligand-binding site distinct from the classical RGD-binding site

The chemokine domain of fractalkine (FKN-CD) binds to the classical RGD-binding site of αvβ3 and that the resulting ternary complex formation (integrin-FKN-CX3CR1) is critical for CX3CR1 signaling and FKN-induced integrin activation. However, only certain cell types express CX3CR1. Here we studied if FKN-CD can activate integrins in the absence of CX3CR1. We describe that WT FKN-CD activated recombinant soluble αvβ3 in cell-free conditions, but the integrin-binding defective mutant of FKN-CD (K36E/R37E) did not. This suggests that FKN-CD can activate αvβ3 in the absence of CX3CR1 through the direct binding of FKN-CD to αvβ3. WT FKN-CD activated αvβ3 on CX3CR1-negative cells (K562 and CHO) but K36E/R37E did not, suggesting that FKN-CD can activate integrin at the cellular levels in a manner similar to that in cell-free conditions. We hypothesized that FKN-CD enhances ligand binding to the classical RGD-binding site (site 1) through binding to a second binding site (site 2) that is distinct from site 1 in αvβ3. To identify the possible second FKN-CD binding site we performed docking simulation of αvβ3-FKN-CD interaction using αvβ3 with a closed inactive conformation as a target. The simulation predicted a potential FKN-CD-binding site in inactive αvβ3 (site 2), which is located at a crevice between αv and β3 on the opposite side of site 1 in the αvβ3 headpiece. We studied if FKN-CD really binds to site 2 using a peptide that is predicted to interact with FKN-CD in site 2. Notably the peptide specifically bound to FKN-CD and effectively suppressed integrin activation by FKN-CD. This suggests that FKN-CD actually binds to site 2, and this leads to integrin activation. We obtained very similar results in α4β1 and α5β1. The FKN binding to site 2 and resulting integrin activation may be a novel mechanism of integrin activation and of FKN signaling.

The binding of monomeric C-reactive protein (mCRP) to Integrins αvβ3 and α4β1 is related to its pro-inflammatory action

The prototypic acute phase reactant C-reactive protein (CRP) is not only a marker but also a potential contributor to inflammatory diseases. CRP exists as the circulating native, pentameric CRP (pCRP) and the monomeric isoform (mCRP), formed as a result of a dissociation process of pCRP. mCRP is highly pro-inflammatory, but pCRP is not. The mechanism of pro-inflammatory action of mCRP is unclear. We studied the role of integrins in pro-inflammatory action of mCRP. Docking simulation of interaction between mCRP and integrin αvβ3 predicted that mCRP binds to αvβ3 well. We found that mCRP actually bound to integrins αvβ3 and α4β1 well. Antagonists to αvβ3 or α4β1 effectively suppressed the interaction, suggesting that the interaction is specific. Using an integrin β1 mutant (β1-3-1) that has a small fragment from the ligand binding site of β3, we showed that mCRP bound to the classical RGD-binding site in αvβ3. We studied the role of integrins in CRP signaling in monocytic U937 cells. Integrins αvβ3 and α4β1 specifically mediated binding of mCRP to U937 cells. mCRP induced AKT phosphorylation, but not ERK1/2 phosphorylation, in U937 cells. Notably, mCRP induced robust chemotaxis in U937 cells, and antagonists to integrins αvβ3 and α4β1 and an inhibitor to phosphatidylinositide 3-kinase, but not an MEK inhibitor, effectively suppressed mCRP-induced chemotaxis in U937 cells. These results suggest that the integrin and AKT/phosphatidylinositide 3-kinase pathways play a role in pro-inflammatory action of mCRP in U937 cells. In contrast, pCRP is predicted to have a limited access to αvβ3 due to steric hindrance in the simulation. Consistent with the prediction, pCRP was much less effective in integrin binding, chemotaxis, or AKT phosphorylation. These findings suggest that the ability of CRP isoforms to bind to the integrins is related to their pro-inflammatory action.

An integrin binding-defective mutant of insulin-like growth factor-1 (R36E/R37E IGF1) acts as a dominant-negative antagonist of the IGF1 receptor (IGF1R) and suppresses tumorigenesis but still binds to IGF1R

Insulin-like growth factor-1 (IGF1) is a major therapeutic target for cancer. We recently reported that IGF1 directly binds to integrins (αvβ3 and α6β4) and induces ternary complex formation (integrin-IGF1-IGF1 receptor (IGF1R)) and that the integrin binding-defective mutant of IGF1 (R36E/R37E) is defective in signaling and ternary complex formation. These findings predict that R36E/R37E competes with WT IGF1 for binding to IGF1R and inhibits IGF signaling. Here, we described that excess R36E/R37E suppressed cell viability increased by WT IGF1 in vitro in non-transformed cells. We studied the effect of R36E/R37E on viability and tumorigenesis in cancer cell lines. We did not detect an effect of WT IGF1 or R36E/R37E in cancer cells under anchorage-dependent conditions. However, under anchorage-independent conditions, WT IGF1 enhanced cell viability and induced signals, whereas R36E/R37E did not. Notably, excess R36E/R37E suppressed cell viability and signaling induced by WT IGF1 under anchorage-independent conditions. Using cancer cells stably expressing WT IGF1 or R36E/R37E, we determined that R36E/R37E suppressed tumorigenesis in vivo, whereas WT IGF1 markedly enhanced it. R36E/R37E suppressed the binding of WT IGF1 to the cell surface and the subsequent ternary complex formation induced by WT IGF1. R36E/R37E suppressed activation of IGF1R by insulin. WT IGF1, but not R36E/R37E, induced ternary complex formation with the IGF1R/insulin receptor hybrid. These findings suggest that 1) IGF1 induces signals under anchorage-independent conditions and that 2) R36E/R37E acts as a dominant-negative inhibitor of IGF1R (IGF1 decoy). Our results are consistent with a model in which ternary complex formation is critical for IGF signaling.

A dominant-negative FGF1 mutant (the R50E mutant) suppresses tumorigenesis and angiogenesis

Fibroblast growth factor-1 (FGF1) and FGF2 play a critical role in angiogenesis, a formation of new blood vessels from existing blood vessels. Integrins are critically involved in FGF signaling through crosstalk. We previously reported that FGF1 directly binds to integrin αvβ3 and induces FGF receptor-1 (FGFR1)-FGF1-integrin αvβ3 ternary complex. We previously generated an integrin binding defective FGF1 mutant (Arg-50 to Glu, R50E). R50E is defective in inducing ternary complex formation, cell proliferation, and cell migration, and suppresses FGF signaling induced by WT FGF1 (a dominant-negative effect) in vitro. These findings suggest that FGFR and αvβ3 crosstalk through direct integrin binding to FGF, and that R50E acts as an antagonist to FGFR. We studied if R50E suppresses tumorigenesis and angiogenesis. Here we describe that R50E suppressed tumor growth in vivo while WT FGF1 enhanced it using cancer cells that stably express WT FGF1 or R50E. Since R50E did not affect proliferation of cancer cells in vitro, we hypothesized that R50E suppressed tumorigenesis indirectly through suppressing angiogenesis. We thus studied the effect of R50E on angiogenesis in several angiogenesis models. We found that excess R50E suppressed FGF1-induced migration and tube formation of endothelial cells, FGF1-induced angiogenesis in matrigel plug assays, and the outgrowth of cells in aorta ring assays. Excess R50E suppressed FGF1-induced angiogenesis in chick embryo chorioallantoic membrane (CAM) assays. Interestingly, excess R50E suppressed FGF2-induced angiogenesis in CAM assays as well, suggesting that R50E may uniquely suppress signaling from other members of the FGF family. Taken together, our results suggest that R50E suppresses angiogenesis induced by FGF1 or FGF2, and thereby indirectly suppresses tumorigenesis, in addition to its possible direct effect on tumor cell proliferation in vivo. We propose that R50E has potential as an anti-cancer and anti-angiogenesis therapeutic agent (“FGF1 decoy”).

Insulin-like growth factor (IGF) signaling requires αvβ3-IGF1-IGF type 1 receptor (IGF1R) ternary complex formation in anchorage independence, and the complex formation does not require IGF1R and Src activation

Integrin αvβ3 plays a role in insulin-like growth factor 1 (IGF1) signaling (integrin-IGF1 receptor (IGF1R) cross-talk) in non-transformed cells in anchorage-dependent conditions. We reported previously that IGF1 directly binds to αvβ3 and induces αvβ3-IGF1-IGF1R ternary complex formation in these conditions. The integrin-binding defective IGF1 mutant (R36E/R37E) is defective in inducing ternary complex formation and IGF signaling, whereas it still binds to IGF1R. We studied if IGF1 can induce signaling in anchorage-independent conditions in transformed Chinese hamster ovary cells that express αvβ3 (β3-CHO) cells. Here we describe that IGF1 signals were more clearly detectable in anchorage-independent conditions (polyHEMA-coated plates) than in anchorage-dependent conditions. This suggests that IGF signaling is masked by signals from cell-matrix interaction in anchorage-dependent conditions. IGF signaling required αvβ3 expression, and R36E/R37E was defective in inducing signals in polyHEMA-coated plates. These results suggest that αvβ3-IGF1 interaction, not αvβ3-extracellular matrix interaction, is essential for IGF signaling. Inhibitors of IGF1R, Src, AKT, and ERK1/2 did not suppress αvβ3-IGF-IGF1R ternary complex formation, suggesting that activation of these kinases are not required for ternary complex formation. Also, mutations of the β3 cytoplasmic tail (Y747F and Y759F) that block β3 tyrosine phosphorylation did not affect IGF1R phosphorylation or AKT activation. We propose a model in which IGF1 binding to IGF1R induces recruitment of integrin αvβ3 to the IGF-IGF1R complex and then β3 and IGF1R are phosphorylated. It is likely that αvβ3 should be together with the IGF1-IGF1R complex for triggering IGF signaling.

Integrins αvβ3 and α4β1 act as coreceptors for fractalkine, and the integrin-binding defective mutant of fractalkine is an antagonist of CX3CR1

The membrane-bound chemokine fractalkine (FKN, CX3CL1) on endothelial cells plays a role in leukocyte trafficking. The chemokine domain (FKN-CD) is sufficient for inducing FKN signaling (e.g., integrin activation), and FKN-CD binds to its receptor CX3CR1 on leukocytes. Whereas previous studies suggest that FKN-CD does not directly bind to integrins, our docking simulation studies predicted that FKN-CD directly interacts with integrin α(v)β(3). Consistent with this prediction, we demonstrated that FKN-CD directly bound to α(v)β(3) and α(4)β(1) at a very high affinity (K(D) of 3.0 × 10(-10) M to α(v)β(3) in 1 mM Mn(2+)). Also, membrane-bound FKN bound to integrins α(v)β(3) and α(4)β(1), suggesting that the FKN-CD/integrin interaction is biologically relevant. The binding site for FKN-CD in α(v)β(3) was similar to those for other known α(v)β(3) ligands. Wild-type FKN-CD induced coprecipitation of integrins and CX3CR1 in U937 cells, suggesting that FKN-CD induces ternary complex formation (CX3CR1, FKN-CD, and integrin). Based on the docking model, we generated an integrin-binding defective FKN-CD mutant (the K36E/R37E mutant). K36E/R37E was defective in ternary complex formation and integrin activation, whereas K36E/R37E still bound to CX3CR1. These results suggest that FKN-CD binding to CX3CR1 is not sufficient for FKN signaling, and that FKN-CD binding to integrins as coreceptors and the resulting ternary complex formation are required for FKN signaling. Notably, excess K36E/R37E suppressed integrin activation induced by wild-type FKN-CD and effectively suppressed leukocyte infiltration in thioglycollate-induced peritonitis. These findings suggest that K36E/R37E acts as a dominant-negative CX3CR1 antagonist and that FKN-CD/integrin interaction is a novel therapeutic target in inflammatory diseases.

Identification of inhibitors against interaction between pro-inflammatory sPLA2-IIA protein and integrin αvβ3

Increased concentrations of secreted phospholipase A2 type IIA (sPLA2-IIA), have been found in the synovial fluid of patients with rheumatoid arthritis. It has been shown that sPLA2-IIA specifically binds to integrin αvβ3, and initiates a signaling pathway that leads to cell proliferation and inflammation. Therefore, the interaction between integrin and sPLA2-IIA could be a potential therapeutic target for the treatment of proliferation or inflammation-related diseases. Two one-bead-one-compound peptide libraries were constructed and screened, and seven target hits were identified. Herein we report the identification, synthesis, and biological testing of two pyrazolylthiazole-tethered peptide hits and their analogs. Biological assays showed that these compounds were able to suppress the sPLA2-IIA-integrin interaction and sPLA2-IIA-induced migration of monocytic cells and that the blockade of the sPLA2-IIA-integrin binding was specific to sPLA2-IIA and not to the integrin.

Cross-talk between integrin α6β4 and insulin-like growth factor-1 receptor (IGF1R) through direct α6β4 binding to IGF1 and subsequent α6β4-IGF1-IGF1R ternary complex formation in anchorage-independent conditions

Integrin αvβ3 plays a role in insulin-like growth factor-1 (IGF1) signaling (integrin-IGF1 receptor (IGF1R) cross-talk). The specifics of the cross-talk are, however, unclear. In a current model, "ligand occupancy" of αvβ3 (i.e. the binding of extracellular matrix proteins) enhances signaling induced by IGF1 binding to IGF1R. We recently reported that IGF1 directly binds to αvβ3 and induces αvβ3-IGF1-IGF1R ternary complex formation. Consistently, the integrin binding-defective IGF1 mutant (R36E/R37E) is defective in inducing ternary complex formation and IGF signaling, but it still binds to IGF1R. Like αvβ3, integrin α6β4 is overexpressed in many cancers and is implicated in cancer progression. Here, we discovered that α6β4 directly bound to IGF1, but not to R36E/R37E. Grafting the β4 sequence WPNSDP (residues 167-172), which corresponds to the specificity loop of β3, to integrin β1 markedly enhanced IGF1 binding to β1, suggesting that the WPNSDP sequence is involved in IGF1 recognition. WT IGF1 induced α6β4-IGF1-IGF1R ternary complex formation, whereas R36E/R37E did not. When cells were attached to matrix, exogenous IGF1 or α6β4 expression had little or no effect on intracellular signaling. When cell-matrix adhesion was reduced (in poly(2-hydroxyethyl methacrylate-coated plates), IGF1 induced intracellular signaling and enhanced cell survival in an α6β4-dependent manner. Also IGF1 enhanced colony formation in soft agar in an α6β4-dependent manner. These results suggest that IGF binding to α6β4 plays a major role in IGF signaling in anchorage-independent conditions, which mimic the in vivo environment, and is a novel therapeutic target.

Direct binding of the EGF-like domain of neuregulin-1 to integrins ({alpha}v{beta}3 and {alpha}6{beta}4) is involved in neuregulin-1/ErbB signaling

Integrin-growth factor receptor cross-talk plays a role in growth factor signaling, but the specifics are unclear. In a current model, integrins and growth factor receptors independently bind to their ligands (extracellular matrix and growth factors, respectively). We discovered that neuregulin-1 (NRG1), either as an isolated EGF-like domain or as a native multi-domain form, binds to integrins αvβ3 (with a K(D) of 1.36 × 10(-7) m) and α6β4. Docking simulation predicted that three Lys residues at positions 180, 184, and 186 of the EGF-like domain are involved in integrin binding. Mutating these residues to Glu individually or in combination markedly suppressed integrin binding and ErbB3 phosphorylation. Mutating all three Lys residues to Glu (the 3KE mutation) did not affect the ability of NRG1 to bind to ErbB3 but markedly reduced the ability of NRG1 to induce ErbB3 phosphorylation and AKT and Erk1/2 activation in MCF-7 and T47D human breast cancer cells. This suggests that direct integrin binding to NRG1 is critical for NRG1/ErbB signaling. Notably, stimulation of cells with WT NRG1 induced co-precipitation of ErbB3 with α6β4 and with αvβ3 to a much lower extent. This suggests that WT NRG1 induces integrin-NRG1-ErbB3 ternary complex formation. In contrast, the 3KE mutant was much less effective in inducing ternary complex formation than WT NRG1, suggesting that this process depends on the ability of NRG1 to bind to integrins. These results suggest that direct NRG1-integrin interaction mediates integrin-ErbB cross-talk and that α6β4 plays a major role in NRG-ErbB signaling in these cancer cells.

A novel fibroblast growth factor-1 (FGF1) mutant that acts as an FGF antagonist

BACKGROUND

Crosstalk between integrins and FGF receptors has been implicated in FGF signaling, but the specifics of the crosstalk are unclear. We recently discovered that 1) FGF1 directly binds to integrin alphavbeta3, 2) the integrin-binding site and FGF receptor (FGFR) binding site are distinct, and 3) the integrin-binding-defective FGF1 mutant (R50E) is defective in inducing FGF signaling although R50E still binds to FGFR and heparin and induces transient ERK1/2 activation.

PRINCIPAL FINDINGS

We tested if excess R50E affect DNA synthesis and cell survival induced by WT FGF1 in BaF3 mouse pro-B cells expressing human FGFR1. R50E suppressed DNA synthesis and cell proliferation induced by WT FGF1. We tested if WT FGF1 and R50E generate integrin-FGF1-FGFR ternary complex. WT FGF1 induced ternary complex formation (integrin-FGF-FGFR1) and recruitment of SHP-2 to the complex in NIH 3T3 cells and human umbilical endothelial cells, but R50E was defective in these functions. It has been reported that sustained ERK1/2 activation is integrin-dependent and crucial to cell cycle entry upon FGF stimulation. We thus determined the time-course of ERK1/2 activation induced by WT FGF1 and R50E. We found that WT FGF1 induced sustained activation of ERK1/2, but R50E was defective in this function.

CONCLUSIONS/SIGNIFICANCE

Our results suggest that 1) R50E is a dominant-negative mutant, 2) Ternary complex formation is involved in FGF signaling, 3) The defect of R50E to bind to integrin may be directly related to the antagonistic action of R50E. Taken together, these results suggest that R50E has potential as a therapeutic in cancer.

Abstract 272: Direct binding of neuregulin-1 to integrins is involved in neuregulin-1/ErbB signaling

Crosstalk between growth factor receptors and integrins play a role in growth factor signaling. In current models of NRG/ErbB signaling, integrins and ErbB receptors transduce signals independently upon binding to their ligands (e.g., extracellular matrix and NRGs, respectively) and their signals are merged inside the cells. We discovered that the EGF-like domain of NRG1 binds directly to integrins αvβ3 and α6β4. Docking simulation predicted that three Lys residues at the amino terminus of NRG1 are involved in integrin binding. We mutated these residues to Glu (designated the 3KE mutant). Indeed the 3KE mutation markedly reduced the NRG1 binding to integrins, while the 3KE mutation did not affect the NRG1 binding to ErbB3. This is consistent with the prediction that the Lys residues critical for integrin binding are not involved in ErbB binding. We studied the effect of the 3KE mutation on NRG1/ErbB signaling in MCF-7 and T47D human breast cancer cells. The 3KE mutation markedly reduced the levels of ErbB3 phosphorylation, AKT and Erk1/2 activation, suggesting that direct integrin binding to NRG1 is critical for NRG1/ErbB signaling. Furthermore, stimulation of cells with WT NRG1 induced co-precipitation of ErbB3 with integrin α6β4. This suggests that WT NRG1 induces integrin-NRG1-ErbB3 ternary complex formation. WT NRG1 induced co-precipitation of ErbB3 with integrin αvβ3 to a less extent than with integrin α6β4. This probably reflects low levels of integrin αvβ3 expression compared to integrin α6β4 in these cells. In contrast, the 3KE mutation markedly reduced the levels of ternary complex formation, suggesting that this process depends on the ability of NRG1 to bind to integrins. Notably, the 3KE mutant acted as an antagonist to ErbB/NRG1 signaling in vitro and effectively suppressed proliferation of breast cancer cells in vivo. These results suggest that 1) NRG1-integrin interaction mediates crosstalk of integrins with ErbB receptors, 2) integrin α6β4 plays a major role in ErbB/NRG signaling in the cancer cells tested, and 3) the 3KE mutant has potential as a therapeutic in cancer.

Citation Format: {Authors}. {Abstract title} [abstract]. In: Proceedings of the 101st Annual Meeting of the American Association for Cancer Research; 2010 Apr 17-21; Washington, DC. Philadelphia (PA): AACR; Cancer Res 2010;70(8 Suppl):Abstract nr 272.

The direct binding of insulin-like growth factor-1 (IGF-1) to integrin alphavbeta3 is involved in IGF-1 signaling

It has been proposed that ligand occupancy of integrin alphavbeta3 with extracellular matrix ligands (e.g. vitronectin) plays a critical role in insulin-like growth factor-1 (IGF-1) signaling. We found that expression of alphavbeta3 enhanced IGF-1-induced proliferation of Chinese hamster ovary cells in serum-free conditions (in the absence of vitronectin). We hypothesized that the direct integrin binding to IGF-1 may play a role in IGF-1 signaling. We demonstrated that alphavbeta3 specifically and directly bound to IGF-1 in cell adhesion, enzyme-linked immunosorbent assay-type binding, and surface plasmon resonance studies. We localized the amino acid residues of IGF-1 that are critical for integrin binding by docking simulation and mutagenesis. We found that mutating two Arg residues at positions 36 and 37 in the C-domain of IGF-1 to Glu (the R36E/R37E mutation) effectively reduced integrin binding. Interestingly, although the mutant still bound to IGF1R, it was defective in inducing IGF1R phosphorylation, AKT and ERK1/2 activation, and cell proliferation. Furthermore wild type IGF-1 mediated co-precipitation of alphavbeta3 and IGF1R, whereas the R36E/R37E mutant did not, suggesting that IGF-1 mediates the interaction between alphavbeta3 and IGF1R. These results suggest that the direct binding to IGF-1 to integrin alphavbeta3 plays a role in IGF-1 signaling through ternary complex formation (alphavbeta3-IGF-IGF1R), and integrin-IGF-1 interaction is a novel target for drug discovery.

Pro-inflammatory secretory phospholipase A2 type IIA binds to integrins alphavbeta3 and alpha4beta1 and induces proliferation of monocytic cells in an integrin-dependent manner

Secretory phospholipase A2 group IIA (sPLA2-IIA) plays an important role in the pathogenesis of inflammatory diseases. Catalytic activity of this enzyme that generates arachidonic acid is a major target for development of anti-inflammatory agents. Independent of its catalytic activity, sPLA2-IIA induces pro-inflammatory signals in a receptor-mediated mechanism (e.g. through the M-type receptor). However, the M-type receptor is species-specific: sPLA2-IIA binds to the M-type receptor in rodents and rabbits, but not in human. Thus sPLA2-IIA receptors in human have not been established. Here we demonstrated that sPLA2-IIA bound to integrin alphavbeta3 at a high affinity (K(D)=2 x 10(-7) M). We identified amino acid residues in sPLA2-IIA (Arg-74 and Arg-100) that are critical for integrin binding using docking simulation and mutagenesis. The integrin-binding site did not include the catalytic center or the M-type receptor-binding site. sPLA2-IIA also bound to alpha4beta1. We showed that sPLA2-IIA competed with VCAM-1 for binding to alpha4beta1, and bound to a site close to those for VCAM-1 and CS-1 in the alpha4 subunit. Wild type and the catalytically inactive H47Q mutant of sPLA2-IIA induced cell proliferation and ERK1/2 activation in monocytic cells, but the integrin binding-defective R74E/R100E mutant did not. This indicates that integrin binding is required, but catalytic activity is not required, for sPLA2-IIA-induced proliferative signaling. These results suggest that integrins alphavbeta3 and alpha4beta1 may serve as receptors for sPLA2-IIA and mediate pro-inflammatory action of sPLA2-IIA, and that integrin-sPLA2-IIA interaction is a novel therapeutic target.

Direct binding of integrin alphavbeta3 to FGF1 plays a role in FGF1 signaling

Integrins play a role in fibroblast growth factor (FGF) signaling through cross-talk with FGF receptors (FGFRs), but the mechanism underlying the cross-talk is unknown. We discovered that FGF1 directly bound to soluble and cell-surface integrin alphavbeta3 (K(D) about 1 microm). Antagonists to alphavbeta3 (monoclonal antibody 7E3 and cyclic RGDfV) blocked this interaction. alphavbeta3 was the predominant, if not the only, integrin that bound to FGF1, because FGF1 bound only weakly to several beta1 integrins tested. We presented evidence that the CYDMKTTC sequence (the specificity loop) within the ligand-binding site of beta3 plays a role in FGF1 binding. We found that the integrin-binding site of FGF1 overlaps with the heparin-binding site but is distinct from the FGFR-binding site using docking simulation and mutagenesis. We identified an FGF1 mutant (R50E) that was defective in integrin binding but still bound to heparin and FGFR. R50E was defective in inducing DNA synthesis, cell proliferation, cell migration, and chemotaxis, suggesting that the direct integrin binding to FGF1 is critical for FGF signaling. Nevertheless, R50E induced phosphorylation of FGFR1 and FRS2alpha and activation of AKT and ERK1/2. These results suggest that the defect in R50E in FGF signaling is not in the initial activation of FGF signaling pathway components, but in the later steps in FGF signaling. We propose that R50E is a useful tool to identify the role of integrins in FGF signaling.

The COOH-Terminal Globular Domain of Fibrinogen γ Chain Suppresses Angiogenesis and Tumor Growth

Fibrinogen is a major plasma protein (350 kDa) that induces proliferative signals by serving as a scaffold to support the binding of growth factors and to promote the cellular responses of adhesion, proliferation, and migration during wound healing, angiogenesis, and tumor growth. Fibrin(ogen) degradation products generated during fibrinolysis are implicated in tissue injury. The fibrinogen gamma chain has a COOH-terminal globular domain (gamma C, residues 151-411 of the gamma chain, 30 kDa) to which several integrin cell adhesion receptors (e.g., platelet alpha(IIb)beta(3), endothelial alpha(v)beta(3), and leukocyte alpha(M)beta(2)) bind. Integrins play a critical role in signal transduction from fibrin(ogen). We found that gamma C and its truncation mutant (designated gamma C399tr), with a deletion of the COOH-terminal 12 residues, induced apoptosis of endothelial cells and blocked tube formation of endothelial cells. DLD-1 human colon cancer cells that secrete gamma C or gamma C399tr grew at similar levels in vitro but grew much slower in vivo than mock-transfected cells. The recombinant purified gamma C399tr fragment markedly suppressed tumor growth, development of intratumoral vasculature, and tumor metastasis in vivo in the highly metastatic Met-1 breast cancer model. The determinant responsible for binding to endothelial cells is cryptic in native fibrinogen but is exposed in gamma C and gamma C399tr. These results suggest that fibrinogen has a novel cryptic determinant, which can exert apoptosis-inducing activity on endothelial cells when exposed, and polypeptides containing this determinant have therapeutic potential.