SOLUBLE GUANYLYL CYCLASE ACTIVATION PROMOTES ANGIOGENESIS

Soluble guanylyl cyclase (sGC) is a cGMP-generating enzyme, carrying a heme prosthetic group that functions as a nitric oxide (NO) sensor. sGC is present in most cells types, including the vascular endothelium, where its biological functions remain largely unexplored. Herein, we have investigated the role of sGC in angiogenesis and angiogenesis-related properties of endothelial cells (EC). Initially, we determined that sGC was present and enzymatically active in the chicken chorioallantoic membrane (CAM) during the days of maximal angiogenesis. In the CAM, inhibition of endogenous sGC inhibited neovascularization, while activation promoted neovessel formation. Using zebrafish as a model for vascular development, we did not detect any effect on vasculogenesis upon sGC blockade, but we did observe an abnormal angiogenic response involving the cranial and intersegmental vessels, as well as the posterior cardinal vein. In vitro, pharmacological activation of sGC, or adenovirus-mediated sGC gene transfer promoted EC proliferation and migration, while sGC inhibition blocked tube-like network formation. In addition, sGC inhibition blocked the migratory response to vascular EC growth factor. Cells infected with sGC-expressing adenoviruses exhibited increased ERK1/2 and p38 MAPK activation that was sensitive to sGC inhibition by ODQ, suggesting that these MAPKs are downstream effectors of sGC in EC. A functional role for p38 in cGMP-stimulated migration was demonstrated using SB203580; pharmacological inhibition of p38 attenuated BAY 41-2272- and sGC overexpression-induced EC mobilization. We conclude that sGC activation promotes the expression of angiogenesis-related properties by EC and that sGC might represent a novel target to modulate neo-vessel formation.


Introduction
Angiogenesis, the formation of new blood vessels from pre-existing structures, is a highly orchestrated process that requires degradation of the extracellular matrix, proliferation and migration of endothelial cells, followed by organization of the EC into stable patent structures supported by mural cells (Folkman and Shing, 1992;Conway et al., 2001). In the adult, angiogenesis is tightly regulated with vessel growth being limited to a few tissues; deregulated angiogenesis has been proposed to contribute to several disease processes including tumor growth, psoriasis, arthritis and diabetic retinopathy (Carmeliet, 2003). During angiogenesis, EC integrate signals form various soluble and matrix-bound molecules to form new vessels (Bischoff, 1997;Conway et al., 2001). Among the endogenous mediators proposed to play an important role in neovascularization, is the labile diatom molecule nitric oxide (NO) (Morbidelli et al., 2003). Exogenously applied NO donors have been shown to stimulate EC growth and migration in vitro (Ziche et al., 1994;Isenberg et al., 2005); moreover, endogenous NO mediates many of the effects of the prototype angiogenic factor vascular endothelial growth factor (VEGF) (Papapetropoulos et al., 1997;Ziche et al., 1997). In vivo, ischemia-induced angiogenesis was attenuated in eNOS knockout animals and VEGF-stimulated vessel formation in the cornea could be blocked by administration of the NOS inhibitor L-NAME (Ziche et al., 1997;Murohara et al., 1998). However, in spite of the wealth of information of the angiogenic effects of NO, the downstream pathways mediating its effects remain poorly characterized.
Soluble or NO-sensitive guanylyl cyclase is a ubiquitously expressed enzyme that acts as a "receptor" for NO (Hobbs, 1997). The low basal activity of sGC increases several hundred-fold upon activation with NO converting GTP to cGMP (Lucas et al., 2000). This article has not been copyedited and formatted. The final version may differ from this version. JPET Fast Forward. Published on August 29, 2006as DOI: 10.1124 at ASPET Journals on April 27, 2022 jpet.aspetjournals.org Downloaded from JPET #108878 5 The most common form of the obligate sGC heterodimer is α 1/β1 that is present in large amounts in smooth muscle, nerve cells and platelets (Hobbs, 1997). In the vascular system, the biological role of sGC has been mostly studied in the context of smooth muscle tone and platelet aggregation (Lucas et al., 2000;Friebe and Koesling, 2003). We have previously shown that sGC subunit mRNA and catalytic activity are also present in endothelial cells from different vascular beds (Papapetropoulos et al., 1996); however, the physiological role of sGC in vascular endothelium remains for the most part unexplored. With respect to angiogenesis, some evidence for the involvement of sGC exists as NO-stimulated EC proliferation is cGMP-dependent and a cell-permeable analogue of cGMP promotes EC migration (Parenti et al., 1998;Kawasaki et al., 2003). The aim of the present study was to characterize the contribution of sGC in neovessel formation; to this end, we studied the effects of pharmacological activators and inhibitors of sGC in two in vivo models; moreover, we sought to determine whether alterations in sGC levels or activity affect the migratory and proliferative potential of EC and to elucidate the pathway(s) involved.
This article has not been copyedited and formatted. The final version may differ from this version.  (Traver et al., 2003). Embryos were incubated with 5 or 10 µM ODQ in embryo water. Analyses shown here were carried out with 5µM ODQ. Control embryos were incubated in 0,05% DMSO in embryo water. Embryos were fixed in 4%PFA overnight and then imaged.

Luciferase activity
African green monkey COSm6 (2 x 10 5 cells) were plated in 6-well plates and grown overnight. Cells were then transfected using the jetPEI transfection reagent (3 µg DNA and 6 µl of jetPEI per well). After 24 h, cells were collected and approximately 1x10 4 cells added inside each O-ring at 6, 9 and 12 days of embryogenesis. Following a 48hr incubation, CAM tissues were removed and assayed for activity using the luciferase reporter assay system according to the manufacturer's instructions.

Western blotting
This article has not been copyedited and formatted. The final version may differ from this version. They were then incubated with the NOS inhibitor L-NAME for 10min; tissues were then stimulated with sodium nitroprusside (100 µ M) in the presence of the phosphodiesterase inhibitor IBMX (1 mM) for 20 min. Media were then aspirated and tissues were homogenised in 0.1 N HCl to extract cGMP. After 30min HCl extracts were collected and cGMP was analysed using a commercially available enzyme immunoassay kit following the manufacturer's instructions.
Construction of adenoviral plasmids and production of adenoviruses. The adenoviral plasmids used in our study were constructed using standard methodology. Briefly, rat α 1 and β 1 cDNAs were subcloned into the pShuttle-CMV vector and recombined with pAdeasy-1 in BJ5183 cells. Recombinants were identified via restriction analysis and transfected into HEK cells (2 x 10 6 ) using the jetPEI reagent. Replication incompetent adenoviruses were then propagated in HEK cells and titered using the This article has not been copyedited and formatted. The final version may differ from this version. In brief, phase-contrast photomicrographs of endothelial cell cultures were recorded and imported into the freeware image analysis program Scion Image (Release Beta 4.0.2, Scion Corporation). Images were converted to a binary format and the binary threshold was adjusted to obtain the best contrast of tubules. The images were then skeletonised and total tubule length was measured in pixels. The area occupied by aggregates of cells was considered as noise. The total length calculated was then expressed as a precent of control. Cell proliferation

HUVEC cells were seeded in 24-well plates at 6x10 3 cells/cm 2 and incubated in M199
supplemented with FBS and ECGS for 24 hours. Cells were incubated with the indicated concentration of BAY 41-2272 (0.1 and 1 µ M) and allowed to proliferate for 48 hr. After this time they were trypsinised and cell number determined using a hemocytometer.

Cell migration
Cells were serum-starved overnight. After trypsinization 1x10 5 cells were added to transwells (8 µ M pore size) in 600 µl of serum-free medium containing 0.25% BSA.

Data Analysis
Data are expressed as means + SEM of the indicated number of observations.
Statistical comparisons between groups were performed using ANOVA followed by a post-hoc test, or Student's t-test, as appropriate. Differences were considered significant when p<0.05. This article has not been copyedited and formatted. The final version may differ from this version.

Pharmacological manipulation of sGC activity results in altered angiogenic responses in vivo.
To determine if sGC subunits are expressed in tissues exhibiting an active angiogenic response, we performed western blot analysis of extracts from CAM tissue. In these experiments, both subunits of the most common sGC isoform (α1/β1) were detected (Fig.1A); α 1 and β 1 levels were developmentally regulated, showing peak expression during days 9-12.
Using a heterologous system we demonstrated that endogenous sGC protein levels correlate with α 1 promoter activity ( We next examined the impact of sGC inhibition in angiogenesis in the CAM. Incubation with the selective sGC inhibitor ODQ (Garthwaite et al., 1995)caused a This article has not been copyedited and formatted. The final version may differ from this version. Inhibition of sGC using a lower ODQ concentration (5µM) added at 10hpf allowed normal formation and differentiation of EC; however, in these embryos we observed that pericardial edema developed, thinner intersegmental vessels existed and the posterior cardinal vein was disorganized (Fig. 2 B, D, E compare with A, C). Overexpression of sGC promotes EC proliferation, migration and tube-like structure formation. We next examined EC properties associated with new blood vessel formation in cells transduced with adenoviruses to overexpress the sGC subunits.
After infection, HUVEC expressed significantly higher α 1 and β 1 protein levels and sGC activity as compared to uninfected or GFP-infected cells (data not shown). sGC overexpressing cells exhibited higher proliferation rates than control cells (Fig.4A).
Similar results were obtained in migration assays where α 1/β1 overexpressing cells showed a 2-fold greater basal migration rate as compared to cells infected with a This article has not been copyedited and formatted. The final version may differ from this version. 14 GFP-expressing adenovirus and exhibited an augmented migratory response to VEGF (Fig.4B). Increased sGC expression also correlated with an increase in the ability of EC to form tube-like networks on Matrigel ® ; the increase in network formation (40%) was of similar magnitude to that seen with naïve EC stimulated with BAY 41-2272 ( Fig.4C).
Mechanisms of sGC-triggered angiogenic responses. We next sought to determine the pathways involved in sGC-regulated responses that are relevant to angiogenesis. As MAPK members have been implicated in EC proliferation and migration, we tested whether sGC-overexpressing cells exhibit increased levels of ERK1/2 and p38 activation. Indeed, cells infected with viruses containing the sGC transgenes displayed a significantly higher pERK1/2/total ERK1/2 ratio indicating that in these cells ERK1/2 is activated (Fig.5A&B). Similar results were obtained for p38 (Fig.5A&B).

Discussion
Although present in substantial amounts in EC, our knowledge on the role of sGC in EC biology is limited. In the present study, we set out to determine the role of sGC in EC properties associated with angiogenesis and to evaluate the contribution of sGC in blood vessel growth. Initial experiments showed that α 1 promoter activity, α 1/β1 immunoreactivity and sGC activity in the CAM were expressed in a manner that coincided with maximal angiogenic activity in this tissue (days 9-12) (Maragoudakis et al., 1988). 16 signals through NO/cGMP, is critical for angioblast formation and differentiation into arterial endothelium in zebrafish (Nasevicius and Ekker, 2000). Similarly, EC are formed during development from angioblasts/endothelial precursors in response to VEGF-induced flk-1 activation in mice (Shalaby et al., 1995;Carmeliet et al., 1996) .
We In order to evaluate the direct effects of sGC activation on EC properties that are important for neovascularization, we used HUVEC, exposed them to BAY 41-2272 and determined their proliferation rate. BAY 41-2272 is a new NO-independent heme-dependent activator of sGC (Stasch et al., 2001). We chose to use this agent to enhance sGC activity and increase intracellular cGMP levels over NO-generating cGMP elevating agents since we wanted to eliminate any cGMP-independent actions of NO donors, like S-nitrosylation, tyrosine nitration, interaction with heme-and non-This article has not been copyedited and formatted. The final version may differ from this version. inhibitory effects on proliferation (Isenberg et al., 2005).
Another property of EC that is important for angiogenesis is the ability of these cells to organize into patent capillary structures. Using Matrigel ® to drive network-like formation in vitro we observed that EC incubated with ODQ engaged less in network formation. In agreement to these results, we have previously shown that inhibition of endogenous NO production attenuates VEGF-and transforming growth factor β 1induced capillary-like structure formation in three dimensional collagen gels (Papapetropoulos et al., 1997). Finally, we tested whether activation of sGC or incubation with a cell-permeable analogue of cGMP affects the ability of EC to migrate. Both BAY 41-2272 and 8Br-cGMP promoted a 4-fold increase in EC migration in the absence of a growth factor, suggesting that sGC activation per se is sufficient to promote EC mobilization. Taken together, the data presented so far indicate that sGC plays an important role in all of the EC properties examined that are linked to angiogenesis. This conclusion is further strengthened by the observation that overexpression of sGC in EC using recombinant adenoviruses, increased the proliferation rate, migratory ability and organization of EC into network-like structures mimicking the responses obtained using pharmacological activators of the enzyme. 18 VEGF is among the best characterized angiogenic factors (Ferrara et al., 2003;Zachary, 2003). Exposure of EC to VEGF increases NO production; the NO released then acts in an autocrine manner to promote angiogenesis (Papapetropoulos et al., 1997;Ziche et al., 1997;Ferrara et al., 2003). However, the contribution of cGMP to the angiogenic properties of VEGF hasn' t been fully explored. In one study, inhibition of sGC by ODQ blocked the increase in VEGF-stimulated EC proliferation (Parenti et al., 1998). To determine if cGMP formation also mediates the action of VEGF with respect to migration, we treated EC with ODQ and determined the migratory response to VEGF. ODQ-treated cells exhibited a blunted migratory response, suggesting that sGC activation is required for this classic growth factor to transmit some of its angiogenic signals. Moreover, our observation that cells infected with sGC adenoviruses exhibit an increased migratory response to VEGF lends further credence to the hypothesis that sGC plays an important role in mediating the biological responses to VEGF.
Mitogen activated protein kinase cascades have been linked to many biological responses associated with angiogenesis (Seger and Krebs, 1995). Members of this family of kinases become fully active after phosphorylation on both threonine and tyrosine residues(Johnson and Lapadat, 2002). EC proliferation in response to VEGF and other growth factors depends on ERK1/2 activation, while p38 has been shown to mediate the migration of EC (Zachary, 2003). Previous reports have shown that NO donors are capable of activating ERK1/2 (Parenti et al., 1998;Oliveira et al., 2003).
NO released from pharmacological concentrations of nitrovasodilators causes Snitrosylation of Cys118 of p21Ras, leading to the recruitment and activation of This article has not been copyedited and formatted. The final version may differ from this version.  (Lander et al., 1997). On the other hand, cell-permeable analogues of cGMP have been proposed to activate Ras using a different, yet unidentified pathway (Oliveira et al., 2003). To investigate whether the EC proliferation observed in HUVEC overexpressing sGC correlates with an increase in ERK activation, we determined the phosphorylated/total ratio for this kinase. Indeed, higher sGC levels resulted in increased ERK1/2 phosphorylation in an ODQ-sensitive manner. ODQ also inhibited basal ERK1/2 phosphorylation, indicating that tonic production of cGMP contributes to basal activity of ERK1/2 in EC. Our data are in line with the observations that the NO-stimulated EC proliferation can be blocked by the MEK inhibitor PD98059 (Parenti et al., 1998).
Although the role of ERK1/2 in proliferation is well established, the biological significance of p38 in EC has only recently started to be investigated (Parenti et al., 1998;Zachary, 2003;McMullen et al., 2005). Activation of the p38 MAPK by angiogenic growth factors has been shown to occur in vitro and proposed to play a role in their ability to stimulate migration (McMullen et al., 2004). In addition, overexpression of MEK6, an upstream activator of p38, promotes EC migration (McMullen et al., 2005). To find out whether activation of p38 occurs in cells overexpressing sGC, we used a phospho-specific Ab for this kinase to determime the phospharylation status of p38; indeed, higher ratios pp38/p38 ratios were observed in cells infected with sGC adenoviruses. The finding that the sGC inhibitor ODQ blocks the increase in p38 phosphorylation offers additional support to the hypothesis that increased cGMP levels promote p38 activation. To provide a functional link between p38 activation and sGC-driven migration, we studied the effect of SB203580 on the migratory behavior of cells exposed to BAY 41-2272 or This article has not been copyedited and formatted. The final version may differ from this version.       This article has not been copyedited and formatted. The final version may differ from this version.