|Home | About | Journals | Submit | Contact Us | Français|
The negative feedback system is an important physiological regulatory mechanism controlling angiogenesis. Soluble vascular endothelial growth factor (VEGF) receptor-1 (sFlt-1), acts as a potent endogenous soluble inhibitor of VEGF- and placenta growth factor (PlGF)-mediated biological function and can also form dominant-negative complexes with competent full-length VEGF receptors.
Systemic overexpression of VEGF-A in mice resulted in significantly elevated circulating sFlt-1. In addition, stimulation of human umbilical vein endothelial cells (HUVEC) with VEGF-A, induced a five-fold increase in sFlt-1 mRNA, a time-dependent significant increase in the release of sFlt-1 into the culture medium and activation of the flt-1 gene promoter. This response was dependent on VEGF receptor-2 (VEGFR-2) and phosphoinositide-3'-kinase signalling. siRNA-mediated knockdown of sFlt-1 in HUVEC stimulated the activation of endothelial nitric oxide synthase, increased basal and VEGF-induced cell migration and enhanced endothelial tube formation on growth factor reduced Matrigel. In contrast, adenoviral overexpression of sFlt-1 suppressed phosphorylation of VEGFR-2 at tyrosine 951 and ERK-1/-2 MAPK and reduced HUVEC proliferation. Preeclampsia is associated with elevated placental and systemic sFlt-1. Phosphorylation of VEGFR-2 tyrosine 951 was greatly reduced in placenta from preeclamptic patients compared to gestationally-matched normal placenta.
These results show that endothelial sFlt-1 expression is regulated by VEGF and acts as an autocrine regulator of endothelial cell function.
Vascular endothelial growth factor-A (VEGF-A) is a multifunctional cytokine induced by hypoxic stress . It plays a pivotal role in many aspects of embryonic cardiovascular development, including formation of blood vessels, cardiac morphogenesis, and development of the nervous system [2-6]. Loss and gain of function studies in mice indicate that VEGF-A levels have to be maintained within a narrow range to ensure proper cardiovascular development and embryo survival [7-9]. It has been shown that the effects of VEGF-A can be deleterious if uncontrolled. Over-expression of VEGF in experimental animals increases the leakiness of blood vessels, which may lead to severe edema, loss of limb and death [10,11]. Excess VEGF-A expression in skeletal muscle results in the induction of vascular tumors (hemangiomas) [12-14], whereas loss of VEGF-A activity due to increased production of its natural antagonist, sFlt-1 (soluble VEGF receptor-1/sVEGFR-1), as in preeclampsia, reduces angiogenesis . Thus, homeostasis requires mechanisms to regulate the functional activity of VEGF-A.
Soluble Flt-1 is generated by alternative splicing of the fms-like tyrosine kinase (flt-1) gene , and binds to all isoforms of VEGF-A and placenta growth factor (PlGF) with high affinity [16,17]. It acts as a potent soluble inhibitor of both VEGF-A and PlGF-mediated biological activities  and can also form dominant-negative complexes with competent full-length VEGF receptors . In pregnancies complicated with preeclampsia, sFlt-1 levels are elevated [15,19-21]. Maternal serum levels of sFlt-1 are elevated five weeks prior to the onset of preeclampsia , supporting the premise that sFlt-1 is a key factor responsible for the clinical manifestation of this disorder . The demonstration that sFlt-1 is fundamental to the clinical onset of preeclampsia  highlights the importance of understanding the intracellular mechanism underlying its regulation and release in endothelial cells. Recently it was shown that autocrine VEGF signaling is required for vascular homeostasis . Here we demonstrate that endothelial sFlt-1 expression is regulated by VEGF and sFlt-1 is an autocrine regulator of endothelial cell function.
Recombinant growth factors were purchased from RELIATech (Brauschweig, Germany). Rabbit polyclonal antibodies against phospho- endothelial nitric oxide synthase (eNOS) at serine-1177 (p-eNOSSer1177), phospho-ERK-1/-2 MAPK and phospho-VEGF receptor-2 (VEGFR-2) tyrosine-951 antibodies were purchased from Calbiochem (Nottingham, UK). Small inhibitory RNAs (siRNA) and oligonucleotide primers were purchased from Eurogentec (Southampton, UK). Luciferase reporter assay and cDNA synthesis kits were from Promega (Southampton, UK). All other cell culture reagents and chemicals were obtained from Sigma Aldrich (Poole, UK).
Human placental tissue was obtained from normal pregnancies and gestationally-matched pregnancies complicated by preeclampsia. Preeclampsia was defined as blood pressure > 140/90 mm Hg on at least two consecutive measurements and proteinuria of at least 300 mg per 24 hours. Informed consent was obtained from the patients and the study had the approval of the South Birmingham Ethical Committee (Birmingham, UK).
Primary human umbilical vein endothelial cells (HUVEC) were isolated and cultured as described . Cells were used at passage two or three for experiments and serum-starved in endothelial cell serum-free medium (Gibco-BRL, UK) supplemented with 0.2% bovine serum albumin for 24 hours prior to stimulation.
Sample preparation and real-time PCR was performed as described previously . Briefly, mRNA was prepared using TRIzol and DNase-1 digestion/purification on RNAeasy columns (Qiagen), and reverse transcribed with the cDNA Synthesis Kit (Promega). Triplicate cDNA samples and standards were amplified in SensiMix containing SYBR green (Quantace) with primers specific for sFlt-1 . The mean threshold cycle (CT) was normalized to β-actin and expressed relative to control.
Two siRNA sequences to the unique 3' sequence of sFlt-1 (sFlt-1 A sense: 5'-TAACAGUUGUCUCAUAUCAtt-3' and antisense: 5'-UGAUAUGAGACAACUGUUAtt-3'; sFlt-1 B sense: 5'-UCUCGGAUCUCCAAAUUUAtt-3' and antisense 5'-UAAAUUUGGAGAUCCGAGAtt-3') were designed using the Dharmacon siDESIGN tool . HUVEC (~ 1 × 106 cells) were electroporated with ~ 3 μg of sFlt-1, or a universal control siRNA (Dharmacon) using the HUVEC kit II and Amaxa nucleofector (Amaxa GmbH, Cologne, Germany) as described .
A chimeric VEGF/epidermal growth factor (EGF) receptor comprising the intracellular and transmembrane domains of VEGFR-2 fused to the extracellular domain of the human EGF receptor . EGF does not bind to VEGF receptors, therefore, it does not activate the endogenous VEGF receptors. EGDR and its tyrosine-to-phenylalanine mutants (EGDR-Y951F) were generated and cloned into the pMMP retroviral vector, and retrovirus-containing cell supernatant was harvested and used immediately to infect HUVEC . Following 16 hours of incubation, the medium was replaced with fresh growth medium and the HUVEC were used 48 hours after infection.
Total NO in conditioned media was assayed as nitrite, the stable breakdown product of NO, using a Sievers NO chemiluminescence analyzer (Analytix, Sunderland, UK) as described previously .
The formation of capillary-like structures was examined on growth factor-reduced Matrigel in 24-well plates as described previously . Tube formation was quantified by measuring the total tube length in five random x200 power fields per well using a Nikon phase-contrast inverted microscope with Image ProPlus image analysis software (Media Cybernetics, Silver Spring, USA). Mean total tube length was calculated from three independent experiments performed in duplicate.
A 1.3 Kb fragment of the human flt-1 gene corresponding to -1214 to +155 bp relative to the first exon in the pGL2 luciferase vector (Promega) was used to determine flt-1 promoter activity . Briefly, porcine aortic endothelial cells (PAEC) were transfected with the flt-1 promoter-reporter construct using Exgen 500 (Fermentas, UK) and the cell lysates assayed as described previously .
Cells lysates were immunoblotted as described previously . Membranes were probed with rabbit polyclonal antibodies against phospho-eNOS-Ser1177, anti-ERK-1/-2 or anti-VEGFR-2 phosphotyrosine-951 at 4°C overnight. Proteins were visualised using the ECL detection kit (Amersham-Pharmacia, UK).
Soluble Flt-1 (sFlt-1) levels in culture supernatants were measured as previously described .
Formalin-fixed, paraffin-embedded tissues were used for immunohistochemistry as previously described .
All data are expressed as mean ± SEM. Statistical comparisons were performed using one-way ANOVA followed by the Student-Newman-Keuls test as appropriate. Statistical significance was set at a value of p < 0.05.
To evaluate the capacity of VEGF-A to regulate the secretion of its negative regulator, sFlt-1, HUVEC were incubated with VEGF-A and the conditioned media assayed for sFlt-1 by ELISA. VEGF-A stimulated a concentration and time dependent increase in the release of sFlt-1 from HUVEC that reached a maximum at 20 ng/ml and 48 hours, respectively (Figure (Figure1a1a and and1b).1b). Consistent with these findings, qPCR revealed a greater than five-fold increase in sFlt-1 mRNA after 22 hours of VEGF-A stimulation (Figure (Figure1c).1c). In addition, VEGF-A induced flt-1 gene promoter activity in porcine aortic endothelial cells transfected with a flt-1 promoter luciferase construct (Figure (Figure1d).1d). Incubation of cells with cycloheximide abrogated the VEGF-A induced response (Figure (Figure1e),1e), which coupled with the fact that there is a negligible release with VEGF-A after two hours of stimulation, indicates that sFlt-1 secretion is due to de novo protein synthesis and not release from intracellular vesicles. Adenoviral-mediated overexpression of VEGF-A in mice caused an eight-fold increase in circulating sFlt-1 levels (Figure (Figure1f),1f), demonstrating, in vivo, that an increase in VEGF-A results in a concomitant rise in circulating sFlt-1 levels, presumably to compensate for elevated VEGF bioactivity.
To identify the VEGF receptors involved in the release of sFlt-1, HUVEC were stimulated with either VEGF-A (binds VEGFR-1 and VEGFR-2), or PlGF-1 (binds VEGFR-1) or VEGF-E (binds VEGFR-2). PlGF-1 showed no effect on sFlt-1 release, whereas VEGF-E stimulated similar levels of sFlt-1 release to those induced by VEGF-A (Figure (Figure2a),2a), suggesting that release of sFlt-1 is mediated by VEGFR-2. Preincubation of endothelial cells with SU1498, a VEGFR-2 selective inhibitor, blocked the VEGF-A induced sFlt-1 release (Figure (Figure2b),2b), confirming the importance of this receptor for the response.
To investigate the role of the PI3K pathway in VEGF-A-induced sFlt-1 release, PI3K activity was inhibited through overexpression of PTEN (Phosphatase and Tensin homolog deleted on chromosome Ten), which dephosphorylates phosphatidylinositol 3,4,5-triphosphate and has been shown to reduce VEGF-mediated signaling and cellular function [28,35]. HUVEC were infected overnight with an adenovirus encoding PTEN (PTEN(wt)) or a control adenovirus (CMV) and stimulated with VEGF-A for 24 hours. Inhibition of PI3K activity by PTEN overexpression led to a significant decrease in sFlt-1 release (Figure (Figure2c).2c). Furthermore, pre-incubation of HUVEC with LY294002, a pharmacological PI3K inhibitor, also attenuated the VEGF mediated release of sFlt-1 (Figure (Figure2d)2d) and of flt-1 gene promoter activity (Figure (Figure2e2e).
Adenoviral-mediated overexpression of sFlt-1 in HUVEC inhibited endothelial cell proliferation (Figure (Figure3a)3a) and MAP kinase ERK-1/-2 phosphorylation (Figure (Figure3a3a insert). Subsequently, to test whether knockdown of sFlt-1 would promote endothelial cell proliferation, HUVEC were transfected with two synthetic siRNA sequences targeted to the unique carboxyl-terminus region of sFlt-1. sFlt-1 siRNA transfection resulted in a substantial reduction in the release of sFlt-1 from HUVEC after 24 hours (Figure (Figure3b).3b). Endothelial cell proliferation was significantly increased (Figure (Figure3c)3c) and interestingly, sFlt-1 knockdown also led to a concomitant increase in VEGFR-2 phosphorylation at tyrosine 951 (Y951) (Figure (Figure3d).3d). In addition, sFlt-1 siRNA increased both basal and VEGF-A-mediated endothelial cell migration (Figure (Figure4a)4a) and tube formation on Matrigel (Figure (Figure4b4b and and4c).4c). VEGF stimulates eNOS activity and NO release [23,36] to mediate angiogenesis [33,37], thus we predicted that loss of sFlt-1 would increase eNOS phosphorylation in HUVEC. Phosphorylation of eNOS (ser1177) was significantly increased in cells lacking sFlt-1 (Figure (Figure4d).4d). These data provide direct evidence that sFlt-1 is itself a negative regulator of endothelial function. It is likely that sFlt-1 sequesters VEGF and PlGF to maintain a physiological steady state until angiogenesis is required, at which point this system must be overridden.
Activation of VEGFR-2 leads to an increase in eNOS expression and activation, which is essential for neovascularisation . A recent study showed that mutation of VEGFR-2 Y951 to phenylalanine caused a significant reduction in VEGF-induced angiogenesis . As preeclampsia is associated with elevated placental  and circulating  sFlt-1 and placental sFlt-1 inhibits angiogenesis , we speculated that elevated free sFlt-1 would lead to a reduction of VEGFR-2 phosphorylation in preeclamptic placenta. Using the EGFR-chimeric receptor system we show that mutation of Y951 to phenylalanine resulted in over 50% reduction in NO release (Figure (Figure5a)5a) and overexpression of sFlt-1 in endothelial cells abrogated phosphorylation of VEGFR-2 Y951 (Figure (Figure5b).5b). To assess whether VEGFR-2 phosphorylation was reduced in preeclamptic placenta, that express elevated sFlt-1, we undertook immunohistochemical analysis for phospho-VEGFR-2 Y951. Overall, phosphorylation of VEGFR-2 Y951 was greatly reduced in the preeclamptic placenta compared to gestationally-matched, normal placenta (Figure (Figure5c).5c). Together, these findings indicate that increased levels of sFlt-1 have a negative effect on VEGFR-2 tyrosine phosphorylation, which in turn would lead to concomitant inhibition of downstream function and signaling and compromise maternal vascular homeostasis and placental angiogenesis.
Endothelial cell sFlt-1 expression is regulated by VEGF and sFlt-1 is an autocrine regulator of endothelial cell function.
EGF: epidermal growth factor; eNOS: endothelial nitric oxide synthase; ERK-1/-2: extracellular signal regulated kinase -1/2; HUVEC: human umbilical vein endothelial cells; MAPK: mitogen activated kinase; NO: nitric oxide; PI3K: phosphoinositide-3'-kinase; PlGF: placenta growth factor; PTEN: Phosphatase and Tensin homolog deleted on chromosome Ten; sFlt-1: soluble vascular endothelial growth factor receptor-1; VEGF-A: vascular endothelial growth factor-A; VEGFR-1: vascular endothelial growth factor receptor-1; VEGFR-2: vascular endothelial growth factor receptor-2.
The authors declare that they have no competing interests.
SA, MJC, PWH, BA, SS and TF performed the experiments and analysed the data SA PWH and MJC contributed to the writing of the manuscript. AA designed experiments and wrote the manuscript. All authors read and approved the final manuscript.
This work was supported by grants from the Medical Research Council (G0600270, G0601295 and G0700288) and British Heart Foundation (RG/09/001/25940 and PG/06/114).