Bradykinin B2 Receptor Contributes to Inflammatory Responses in Human Endothelial Cells by the Transactivation of the Fibroblast Growth Factor Receptor FGFR-1

Elevated levels of bradykinin (BK) and fibroblast growth factor-2 (FGF-2) have been implicated in the pathogenesis of inflammatory and angiogenic disorders. In angiogenesis, both stimuli induce a pro-inflammatory signature in endothelial cells, activating an autocrine/paracrine amplification loop that sustains the neovascularization process. Here we investigated the contribution of the FGF-2 pathway in the BK-mediated human endothelial cell permeability and migration, and the role of the B2 receptor (B2R) of BK in this cross-talk. BK (1 µM) upregulated the FGF-2 expression and promoted the FGF-2 signaling, both in human umbilical vein endothelial cells (HUVEC) and in retinal capillary endothelial cells (HREC) by the activation of Fibroblast growth factor receptor-1 (FGFR-1) and its downstream signaling (fibroblast growth factor receptor substrate: FRSα, extracellular signal–regulated kinases1/2: ERK1/2, and signal transducer and activator of transcription 3: STAT3 phosphorylation). FGFR-1 phosphorylation triggered by BK was c-Src mediated and independent from FGF-2 upregulation. Either HUVEC and HREC exposed to BK showed increased permeability, disassembly of adherens and tight-junction, and increased cell migration. B2R blockade by the selective antagonist, fasitibant, significantly inhibited FGF-2/FGFR-1 signaling, and in turn, BK-mediated endothelial cell permeability and migration. Similarly, the FGFR-1 inhibitor, SU5402, and the knock-down of the receptor prevented the BK/B2R inflammatory response in endothelial cells. In conclusion, this work demonstrates the existence of a BK/B2R/FGFR-1/FGF-2 axis in endothelial cells that might be implicated in propagation of angiogenic/inflammatory responses. A B2R blockade, by abolishing the initial BK stimulus, strongly attenuated FGFR-1-driven cell permeability and migration.


Introduction
Inflammation and angiogenesis are closely integrated processes regulating a number of physiological and pathological settings, including wound healing, rheumatoid arthritis, diabetic retinopathy, arteriosclerosis, and cancer [1][2][3]. Fibroblast growth factor-2 (FGF-2) exerts a key role in the cross-talk between angiogenesis and inflammation by interacting with various surface molecules, including tyrosine kinase receptors Fibroblast growth factor receptor 1 to 4 (FGFR-1 to FGFR-4), heparan-sulfate proteoglycans, integrins and syndecans [4]. In endothelial cells (EC), through the FGFR-1 activation, FGF-2 promotes the inflammatory response during the angiogenic process by inducing vasoactive effects, such as vasodilation and vascular permeability [5][6][7]. Further, the FGF-2/FGFR-1 signaling pathway promotes the expression of a wide variety of inflammation-related genes in EC, which have a pivotal role in the neovascularization [8,9], and, in turn, several inflammatory mediators trigger FGF-2 release from EC [10,11]. In a previous report, we showed that prostaglandin E2 (PGE2) induces angiogenesis by an autocrine FGF-2 mobilization from the EC extracellular matrix, resulting in FGFR-1 signaling activation [12,13]. Of note, FGF-2 itself also upregulates its expression in EC, and that of other growth factors, including vascular endothelial growth factor (VEGF) [14][15][16]. Thus, the EC activation by FGF-2 appears to be a concerted action between an autocrine loop and that of other factors originating in a paracrine modality from inflammatory cells, including prostanoids, cytokines, and other chemokines [17,18].
Two G-protein-coupled receptors, BK receptor 1 and 2 (B1R and B2R) transduce BK signals to the effector molecules mentioned above [19]. Activation of these receptors elicits pro-angiogenic responses in EC, as well as important changes of their tone (vasodilatation), permeability, and the enhanced recruitment of inflammatory cells [17,20,21]. Recently, in the oxygen induced retinopathy (OIR) model, in mice, we demonstrated that the BK/B2R signaling plays a pathogenic role in retinal neovascularization, and that its effects correlate with FGF-2 upregulation in retinal vessels [22]. Here we investigated the contribution of the FGF-2/FGFR-1 pathway to BK/B2R-mediated human endothelial cell permeability and migration in two EC lines, human umbilical vein endothelial cells (HUVEC) and human retinal capillary endothelial cells (HREC). Our data demonstrates that BK transactivated the FGF-2/FGFR-1 signaling in both HUVEC and HREC cells, which was implicated in cell permeability and migration. B2R blockade, by abolishing the initial BK stimulus, strongly attenuated FGFR-1-driven inflammatory responses.
We also studied the perinuclear translocation of FGFR-1 in response to BK, a known mechanism linked to tyrosine kinase receptor activation [12].

Phosphorylation of FGFR-1 by BK Requires the Activation of c-Src
To investigate the mechanism whereby BK stimulated the phosphorylation of FGFR-1, we designed experiments in which FGF-2 was either ablated (knockout) or its signal transduction was impaired through the application of a non-permeant specific neutralizing antibody to EC. Both maneuvers failed to influence the FGFR-1 phosphorylation elicited by the BK exposure ( Figure 4A,B), leading to the conclusion that BK activates the FGFR-1 pathway independently from the extant FGF-2. (1 µM, 30 min), then stimulated with BK (1 µM) for 15 min. Results were normalized to ERK1/2 and STAT3, respectively. (C) FGFR-1 expression evaluated using western blot analysis in HUVEC transfected with two different shRNA for FGFR-1 knock-down (Sh#1 and Sh#2). EV = empty vector. (D,E) Western blot analysis for ERK1/2 and STAT3 phosphorylation in HUVEC transfected with Sh#1 and Sh#2 and stimulated with BK (1 µM) for 15 min. (F) FGF-2 expression was evaluated in HUVEC treated with STAT3 inhibitor (10 µM, 30 min) and then stimulated with BK (1 µM) for 24 h. Actin was used as a loading control. The results presented are representative of three independent experiments (n = 3) with similar results. Quantification was expressed as an arbitrary density unit (ADU). ** p < 0.01; *** p < 0.001 vs. Ctr; ### p < 0.001 vs. BK treated cells.

Phosphorylation of FGFR-1 by BK Requires the Activation of c-Src
To investigate the mechanism whereby BK stimulated the phosphorylation of FGFR-1, we designed experiments in which FGF-2 was either ablated (knockout) or its signal transduction was impaired through the application of a non-permeant specific neutralizing antibody to EC. Both maneuvers failed to influence the FGFR-1 phosphorylation elicited by the BK exposure ( Figure 4A,B), leading to the conclusion that BK activates the FGFR-1 pathway independently from the extant FGF-2.  Given the observed irrelevancy of the extracellular FGFR-1 domain in the downstream signal propagation activated by BK/B2R system, we focused on the cell cytosol, particularly on c-Src, a kinase recently shown to serve as a signaling mediator both downstream and upstream of the epidermal growth factor receptor activation [26]. BK application to EC (HUVEC and HREC) promoted a robust increase over basal in c-Src phosphorylation via B2R, an effect sensitive to fas blockade (5-fold decrease over BK treated cells; Figure 5A,B). Inhibition of FGFR-1 and FRSα phosphorylation mediated by BK through the known c-Src inhibitors PP1 and SU566, provided evidence that FGFR-1 lay upstream of c-Src ( Figure 5C-E). We conclude that BK activated FGFR-1 through a c-Src-dependent mechanism that appeared to be independent to the FGF-2-mediated receptor activation. Given the observed irrelevancy of the extracellular FGFR-1 domain in the downstream signal propagation activated by BK/B2R system, we focused on the cell cytosol, particularly on c-Src, a kinase recently shown to serve as a signaling mediator both downstream and upstream of the epidermal growth factor receptor activation [26]. BK application to EC (HUVEC and HREC) promoted a robust increase over basal in c-Src phosphorylation via B2R, an effect sensitive to fas blockade (5-fold decrease over BK treated cells; Figure 5A,B). Inhibition of FGFR-1 and FRSα phosphorylation mediated by BK through the known c-Src inhibitors PP1 and SU566, provided evidence that FGFR-1 lay upstream of c-Src ( Figure 5C-E). We conclude that BK activated FGFR-1 through a c-Src-dependent mechanism that appeared to be independent to the FGF-2-mediated receptor activation.  Given the observed irrelevancy of the extracellular FGFR-1 domain in the downstream signal propagation activated by BK/B2R system, we focused on the cell cytosol, particularly on c-Src, a kinase recently shown to serve as a signaling mediator both downstream and upstream of the epidermal growth factor receptor activation [26]. BK application to EC (HUVEC and HREC) promoted a robust increase over basal in c-Src phosphorylation via B2R, an effect sensitive to fas blockade (5-fold decrease over BK treated cells; Figure 5A,B). Inhibition of FGFR-1 and FRSα phosphorylation mediated by BK through the known c-Src inhibitors PP1 and SU566, provided evidence that FGFR-1 lay upstream of c-Src ( Figure 5C-E). We conclude that BK activated FGFR-1 through a c-Src-dependent mechanism that appeared to be independent to the FGF-2-mediated receptor activation.

BK Induces Endothelial Permeability and Migration through FGFR-1 Activation
As previously demonstrated, BK produces a significant increase of endothelial permeability, a common histopathological marker of inflammation, measured as paracellular flux of fluorescent conjugated dextran [17]. In EC, BK increased paracellular flux. Co-treatment of EC with fasitibant (1 µM) or SU5402 (1 µM) abolished the BK-induced paracellular flux increase, restoring the flux to the control level ( Figure 6A,B).
In a condition of in vitro confluence, cells regulate permeability through the expression of celltype-specific transmembrane adhesion proteins, such as vascular endothelial-cadherin (VEC), at adherens junctions, and zonula occludens-1 (ZO-1), at tight junctions. Consistent with its permeability effects, BK drastically reduced the typical pattern of fluorescence localization of either VEC ( Figures 6C,E

BK Induces Endothelial Permeability and Migration through FGFR-1 Activation
As previously demonstrated, BK produces a significant increase of endothelial permeability, a common histopathological marker of inflammation, measured as paracellular flux of fluorescent conjugated dextran [17]. In EC, BK increased paracellular flux. Co-treatment of EC with fasitibant (1 µM) or SU5402 (1 µM) abolished the BK-induced paracellular flux increase, restoring the flux to the control level ( Figure 6A,B).
In a condition of in vitro confluence, cells regulate permeability through the expression of cell-type-specific transmembrane adhesion proteins, such as vascular endothelial-cadherin (VEC), at adherens junctions, and zonula occludens-1 (ZO-1), at tight junctions. Consistent with its permeability effects, BK drastically reduced the typical pattern of fluorescence localization of either VEC ( Figures 6C,E

BK Induces Endothelial Permeability and Migration through FGFR-1 Activation
As previously demonstrated, BK produces a significant increase of endothelial permeability, a common histopathological marker of inflammation, measured as paracellular flux of fluorescent conjugated dextran [17]. In EC, BK increased paracellular flux. Co-treatment of EC with fasitibant (1 µM) or SU5402 (1 µM) abolished the BK-induced paracellular flux increase, restoring the flux to the control level ( Figure 6A,B).
In a condition of in vitro confluence, cells regulate permeability through the expression of celltype-specific transmembrane adhesion proteins, such as vascular endothelial-cadherin (VEC), at adherens junctions, and zonula occludens-1 (ZO-1), at tight junctions. Consistent with its permeability effects, BK drastically reduced the typical pattern of fluorescence localization of either VEC ( Figures 6C,E  We also evaluated the contribution of FGFR-1 activation in the pro-angiogenic effect of BK by studying endothelial cell migration. In EC, BK induced migration while SU5402 co-incubated with BK, and the knock-down of FGFR-1 reduced its effects ( Figure 7A,C and Figure 7B,D for quantification), indicating that FGFR-1 was involved in the pro-angiogenic effect of BK in EC.
We also evaluated the contribution of FGFR-1 activation in the pro-angiogenic effect of BK by studying endothelial cell migration. In EC, BK induced migration while SU5402 co-incubated with BK, and the knock-down of FGFR-1 reduced its effects ( Figure 7A,C and 7B,D for quantification), indicating that FGFR-1 was involved in the pro-angiogenic effect of BK in EC.

Discussion
In this study, we demonstrate that stimulation of FGF-2/FGFR-1 signaling in endothelial cells contributes to BK/B2R-induced permeability changes and migration. These findings suggest that FGF-2 and BK signaling might orchestrate the pathogenesis of vascular disorders through induction of inflammatory and proangiogenic changes in the vascular endothelium.
The half-life of BK is regulated locally and is propagated through stimulation of other local signaling [27]. Recently, in a model of OIR, in mice, we demonstrated that BK/B2R is involved in the pathological retinal neovascularization and that this effect correlated with upregulation of FGF-2 in

Discussion
In this study, we demonstrate that stimulation of FGF-2/FGFR-1 signaling in endothelial cells contributes to BK/B2R-induced permeability changes and migration. These findings suggest that FGF-2 and BK signaling might orchestrate the pathogenesis of vascular disorders through induction of inflammatory and proangiogenic changes in the vascular endothelium.
The half-life of BK is regulated locally and is propagated through stimulation of other local signaling [27]. Recently, in a model of OIR, in mice, we demonstrated that BK/B2R is involved in the pathological retinal neovascularization and that this effect correlated with upregulation of FGF-2 in retinal vessels [22]. A limited number of studies suggest the existence of a functional link between the two systems in several inflammatory/angiogenic disorders [28][29][30]. Although FGF-2 can trigger neovessel formation per se, the major finding of this study concerns the BK ability to transactivate FGFR-1 signaling in endothelial cells, as evidenced by its phosphorylation and translocation from the plasma membrane to cytosol, and, in turn, by the phosphorylation of second messengers downstream to FGFR-1, as FRSα, ERK1/2, and STAT3. All events occur within minutes from BK addition to EC, suggesting that the peptide functions as an initial trigger for a robust pro-angiogenic response. Further, the delayed upregulation of FGF-2 observed in EC challenged with BK indicate that the B2R signaling also functions as a trigger for an amplified pro-angiogenic/inflammatory response mediated by FGF-2 signaling. Fasitibant application to EC, a selective full antagonist of B2R, halted the FGFR-1 activation, therefore inhibiting the functional downstream sequelae of its activation. Of note, from several of our findings, in absence of BK, the antagonist appeared to affect the B2R signaling per se. Further experiments are needed to investigate this point. Interfering with the FGF-2/FGFR-1 system by a FGFR-1 blockade (inhibition of the receptor through the low selective SU5402 [31], or its knocking-down through shRNA) also inhibited BK activity on ERK1/2 and STAT3, and similarly, STAT3 inhibition suppressed BK activity on FGF-2 expression, indicating hat the FGF-2/FGFR-1 system plays a functional role in the proangiogenic/proinflammatory effects of BK. It is notable that STAT3 is a key player in inflammation. The interactions of growth factors and cytokines with their membrane-bound receptors frequently trigger STAT activation [32][33][34]. FGFR-1 influences the STAT3 pathway, and in turn, impacts on pro-inflammatory and proangiogenic responses [32][33][34]. In EC exposed to BK, STAT3 appears instrumental for FGF-2 upregulation. Further, the endothelium, activated through B2R stimulation and by the FGFR-1 cascade products, exhibits the typical inflammatory/angiogenic phenotype, as shown by enhanced permeability and motility of endothelial cells.
In light of these results, we propose a model for the angiogenic/inflammatory switch in EC based on the FGFR-1 signaling stimulation and FGF-2 upregulation by BK (see Figure 8). Previously, we reported that FGFR-1 acts as a master switch in neovascularization induced by inflammatory mediators by initiating a positive autocrine/paracrine loop of FGF-2 synthesis and FGFR-1 activation [13]. In EC, BK appears to act as a primer of this switch by directly transactivating FGFR-1 signaling and FGF-2 expression. In conclusion, the functional mechanistic association between the BK and FGF-2 signaling pathways here described provides the basis for a defined targeting of molecules involved in microvascular disease initiation and progression. The blockade of the B2R by a selective antagonist might restrict the pathological angiogenesis, reducing the acute inflammatory and angiogenic responses of the vascular endothelium and the following amplification and propagation through the FGFR-1/FGF-2 pathway. 2 signaling pathways here described provides the basis for a defined targeting of molecules involved in microvascular disease initiation and progression. The blockade of the B2R by a selective antagonist might restrict the pathological angiogenesis, reducing the acute inflammatory and angiogenic responses of the vascular endothelium and the following amplification and propagation through the FGFR-1/FGF-2 pathway.