Suppression of Sproutys Has a Therapeutic Effect for a Mouse Model of Ischemia by Enhancing Angiogenesis

Sprouty proteins (Sproutys) inhibit receptor tyrosine kinase signaling and control various aspects of branching morphogenesis. In this study, we examined the physiological function of Sproutys in angiogenesis, using gene targeting and short-hairpin RNA (shRNA) knockdown strategies. Sprouty2 and Sprouty4 double knockout (KO) (DKO) mice were embryonic-lethal around E12.5 due to cardiovascular defects. The number of peripheral blood vessels, but not that of lymphatic vessels, was increased in Sprouty4 KO mice compared with wild-type (WT) mice. Sprouty4 KO mice were more resistant to hind limb ischemia and soft tissue ischemia than WT mice were, because Sprouty4 deficiency causes accelerated neovascularization. Moreover, suppression of Sprouty2 and Sprouty4 expression in vivo by shRNA targeting accelerated angiogenesis and has a therapeutic effect in a mouse model of hind limb ischemia. These data suggest that Sproutys are physiologically important negative regulators of angiogenesis in vivo and novel therapeutic targets for treating peripheral ischemic diseases.


Introduction
Growth factor-induced signaling by receptor tyrosine kinases (RTKs) plays several essential roles in development and pathogenesis; accordingly, it is tightly controlled by a number of regulatory proteins [1][2][3]. When a ligand binds to an RTK and recruits a Grb2-Sos to the inner surface of a membrane, the Sos protein binds to Ras, causing GDP/GTP exchange and thus activating Ras. Activated Ras recruits Raf to the plasma membrane and activates the Raf/MEK/extracellular signalregulated kinase (ERK) pathway. Some growth factors, such as vascular endothelial growth factor (VEGF)-A, also activate the Raf/MEK/ERK pathway through the RTK/phospholipase C (PLC)-c/protein kinase C (PKC) pathway, which is a Rasindependent pathway [4].
Sprouty (Spry) has been genetically identified as an antagonist of fibroblast growth factor (FGF) receptor in tracheal development in Drosophila, and is a proven negative regulator of the Ras/Raf/ ERK pathway [5,6]. Four mammalian genes with sequence similarity to Drosophila Sprouty (Sprouty1-4) have been identified [1,2]. In addition, we have identified three Sprouty-related proteins known as Spred1-3 (Spreds), in which the C-terminal cysteine-rich domain found in Sprouty proteins (Sproutys) is conserved [7,8]. Since loss-of-function mutations of the SPRED1 gene have been found in human neuro-cardio-facial-cutaneous (NCFC) syndromes [9], and since these syndromes are caused by dysregulation of the Ras-ERK pathway, we conclude that SPRED1 is a negative regulator of RTK-mediated Ras/ERK activation.
In the development of the cardiovascular system of Drosophila, as in the tracheal system, the formation of new blood vessels from preexisting ones (angiogenesis) involves the sprouting of endothelial cells out of an epithelial layer and the branching of tubular structures [10]. In the adult, angiogenesis only takes place during the female reproductive cycle, during wound healing, and in pathological situations, including tumor growth, diabetic retinopathy, arthritis, atherosclerosis, and psoriasis [10,11]. Angiogenesis is tightly regulated by a balance between inducing and inhibitory signals [12]. Growth factors, such as VEGF-A, basic FGF (bFGF), and angiopoietins, positively regulate angiogenesis by binding to their cognate RTKs and thus inducing endothelial cell proliferation, migration, differentiation, and survival [12,13]. In addition, sphingosine-1-phosphate (S1P), which activates GPCRs, has also been implicated in angiogenesis [14]. In contrast, proteins that negatively regulate angiogenesis by specifically blocking RTK signaling are less well characterized.
Previous studies have demonstrated that overexpression of Sproutys inhibits VEGF-A-and bFGF-induced endothelial cell proliferation and differentiation in vitro as well as branching and sprouting of small vessels in vivo [15,16]. Moreover, Sprouty4 suppresses VEGF-A/VEGF receptor (VEGFR)-2 signaling in vitro [17][18][19]. We also know that Spreds, in contrast, inhibit VEGF-C signaling, which is important in lymphangiogenesis, and that Spred1/Spred2 double knockout (KO) (DKO) mice show abnormal lymphatic development [18]. Yet the physiological role of Sproutys in angiogenesis and lymphangiogenesis remains to be elucidated.
In this study we investigated the physiological function of Sproutys in angiogenesis by performing knockout and knockdown analyses of Sproutys. We showed that Sproutys are negative regulators for angiogenesis rather than lymphangiogenesis in vivo. Moreover, Sprouty4 KO mice were more resistant to hind limb ischemia and soft tissue ischemia than wild-type (WT) mice were, and in vivo shRNA targeting Sprouty2 and Sprouty4 accelerated angiogenesis in a mouse model of hind limb ischemia. These data suggest that Sprouty2 and Sprouty4 are important negative regulators of angiogenesis in vivo that could be new therapeutic targets for ischemic diseases.

Increased developmental angiogenesis in Sproutydeficient mice
Overexpression studies suggest that Sprouty2 and Sprouty4 possess similar negative effects on RTK-mediated ERK activation [20]. To define the overlapping functions of Sprouty2 and Sprouty4, we generated Sprouty2/Sprouty4 DKO mice. Sprouty2/ Sprouty4 DKO mice were embryonic-lethal by embryonic day 12.5 and showed very severe defects in craniofacial and limb morphogenesis [21]. They also showed very severe subcutaneous hemorrhage, edema ( Fig. 1A-D), and multiple hepatic hemangiomas (Fig. 1E,F), which suggested that they had cardiovascular defects as well. We next investigated the expression pattern of Sprouty2 and Sprouty4 in endothelial cells during embryonic development, and found that Sprouty2 and Sprouty4 were more highly expressed in blood endothelial cells (BECs) than in lymphatic endothelial cells (LECs) (Fig. 1G).
This discovery led us to examine vascularization in adult Sprouty4 single KO mice in detail, although Sprouty4 single KO mice showed no obvious vascular phenotype [21]. Sprouty4 single KO mice exhibited more vascular networks of blood vessels in the ear than WT mice did ( Fig. 2A,B). Similarly, more vascular networks of blood vessels in the ear were observed in Sprouty2 single KO mice than in WT mice (data not shown). The numbers of blood vessels in the skin were also increased in Sprouty4 KO mice (Fig. 2C,D). Lymphatic vessel networks, on the other hand, were present at the same frequency in these Sprouty4 KO mice as in WT mice ( Fig. 2A-D). Retinal vasculature is a good model system for the study of general blood vessel development [22]. Vascular development in the early embryo is difficult to observe, but the murine retinal vascular system develops after birth and is therefore easier to examine. We compared flat-mounted retinas from WT and Sprouty4 KO mice at postnatal day (PD) 3 after injecting FITC-dextran (Fig. 2E). As the image clearly shows, retinal angiogenesis was enhanced in Sprouty4 KO mice compared to WT mice.
These data suggest that, in contrast to Spred1 and Spred2, which are important negative regulators of developmental lymphangiogenesis rather than developmental angiogenesis, as previously reported [18], Sprouty2 and Sprouty4 are important negative regulators of developmental angiogenesis rather than developmental lymphangiogenesis.
Sprouty4 KO mice are more resistant to ischemia Next, we sought to investigate the effect of Sprouty4 deficiency in the ischemia-induced angiogenesis model, an adult neovascularization assay which is useful for quantifying neovascularization in Sprouty4 KO mice. We used mouse models of hind limb ischemia [23] and soft tissue ischemia [24]. We used Sprouty4 KO mice, since Sprouty4 KO mice can survive much longer than Sprouty2 KO mice [21,25].
The former model revealed that Sprouty4 KO mice were more resistant to hind limb ischemia than WT mice were (Fig. 3A). Sprouty4 KO mice showed a significantly elevated recovery of limb perfusion after induction of hind limb ischemia as compared with WT mice, and the ischemic/non-ischemic leg perfusion ratio was much more favorable in Sprouty4 KO mice than in WT mice (P,0.001) (Fig. 3B,C). Additionally, angiogenesis in the ischemic hind limb was significantly increased in Sprouty4 KO mice compared with WT mice (Fig. 3D).
The latter model was induced by creating lateral skin incisions on the dorsal surfaces of mice. The overlying skin was undermined, and a silicone sheet was inserted into each mouse to separate the skin from the underlying tissue bed. As a result, the most central portion of skin underwent the most severe ischemic insult, which, in WT mice, ultimately led to the absence of flow and necrosis in the central portion of the skin (Fig. 4A). In Sprouty4 KO mice, however, angiogenesis in the dorsal skin was significantly increased compared to that in WT mice (Fig. 4B). As a result, Sprouty4 KO mice were more resistant to soft tissue ischemia than WT mice were, and gross evidence of necrosis in the dorsal skin was more evident in WT mice than in Sprouty4 KO mice (100% and 16.7% in WT mice and Sprouty4 KO mice, respectively, n = 6) (Fig. 4A).
These data show that Sprouty4 KO mice exhibit enhanced neovascularization in ischemia-induced models.

Increased ischemia-induced neovascularization by in vivo shRNA targeting Sproutys
The increased vessel density in the skin and muscles of untreated Sprouty4 KO mice (Fig. 2C,D, Fig. 3D) provides them with elevated blood-vessel area in these regions, which is partially responsible for their increased resistance to ischemia.
To investigate in vivo the efficiency of down-regulating Sproutys as therapy for peripheral ischemic diseases, we administered ischemia treatment to the hind limbs of C57BL/6J mice, then injected shRNA targeting Sprouty2 and Sprouty4. We suppressed both Sprouty2 and Sprouty4 simultaneously, because both of the mRNAs increased during hind limb ischemia (Fig. 5A), because the phenotype of Sprouty2/ Sprouty4 DKO mice demonstrated a redundant role of Sprouty2 and Sprouty4 in angiogenesis (Fig. 1), and because we have not found any functional differences between Sprouty2 and Sprouty4 in vitro [20]. The shRNA plasmid targeting Sprouty2 and Sprouty4 efficiently suppressed the expression levels of endogenous Sprouty2 and Sprouty4, respectively, in both realtime PCR (Fig. 5B,C) and Western blot (Fig. 5D,E) analysis. The shRNA plasmids targeting both Sprouty2 and Sprouty4 enhanced VEGF-A-induced ERK and Akt activation in vitro in mouse embryonic fibroblasts (MEFs), which stably expressed VEGFR-2 ( Fig. 5F). We used MEFs because it has been very difficult to introduce shRNA into primary murine endothelial cells in vitro. Although we confirmed that shRNA against Sprouty2 or Sprouty4 alone enhanced VEGF-A-induced ERK and Akt activation (data not shown), we observed much stronger effect by the combination of Sprouty2 and Sprouty4 shRNAs. Thus we decide to use combination of both Sprouty2 and Sprouty4 shRNAs for further experiments.
First, we showed that injection of shRNA plasmids targeting Sprouty2 and Sprouty4 significantly enhanced corneal neovascularization induced by VEGF-A, compared to control shRNA plasmids in a corneal micropocket assay ( Fig. 6A-C). These data indicate that Sprouty2 and Sprouty4 shRNA plasmids can efficiently suppress the expression levels of Sprouty2 and Sprouty4 and block the effect of endogenous Sprouty2 and Sprouty4 in vitro and in vivo.
Next, we investigated whether Sprouty2 and Sprouty4 shRNA plasmids were effective in a mouse model of hind limb ischemia. Upon injection to the ischemic adductor muscle, Sprouty2 and Sprouty4 shRNA plasmids reduced Sprouty2 and Sprouty4 expression in vivo (Fig. 5A). Sprouty2 and Sprouty4 shRNA plasmids induced a significantly elevated recovery of limb perfusion after induction of hind limb ischemia as compared with control shRNA plasmids, and markedly improved the ischemic/non-ischemic leg perfusion ratio (P,0.05) (Fig. 7A,B). shRNA plasmids targeting Sprouty2 and Sprouty4 also increased capillary density compared with control shRNA plasmids (Fig. 7C). Our data clearly demonstrate that Sprouty2 and Sprouty4 negatively regulate angiogenesis in vivo and would make good therapeutic targets for peripheral ischemic diseases.

Discussion
In this study, we investigated the physiological function of Sproutys in angiogenesis by performing a knockout and knockdown analysis of Sproutys. In contrast to Spred1 and Spred2, which regulate developmental lymphangiogenesis, Sprouty2 and Spro-uty4 are important negative regulators of developmental angiogenesis in vivo. We found that the amounts of blood vessels in Sprouty4 KO mice are increased in all tissues we investigated. So we think that all peripheral blood vessels are increased in Sprouty4 KO mice. Sprouty4 deficiency enhanced ischemia-induced angiogenesis in mouse models of hind limb ischemia and soft tissue ischemia. Moreover, the suppression of Sprouty2 and Sprouty4 expression in vivo by shRNA targeting had a therapeutic effect in our model of hind limb ischemia, indicating that Sproutys should be novel therapeutic targets for treating peripheral ischemic diseases. The roles of Sprouty and Spred proteins during gastrulation in Xenopus tropicalis have been compared elsewhere [26]. Spred proteins preferentially inhibit the Ras/ERK cascade that directs mesoderm formation, whereas Sprouty proteins block the Ca 2+ and PKCd signals required for morphogenetic movements during gastrulation. Thus, the expression of Sprouty and Spred genes at specific times during gastrulation might redirect FGF signals toward mesoderm formation or morphogenesis, respectively [26].
In mammalian development, Sproutys are expressed mainly in blood endothelial cells, while Spreds are expressed mainly in lymphatic endothelial cells (Fig. 1G and Ref. [18]). In overexpression experiments, while Sproutys can inhibit VEGF-A signaling but not VEGF-C signaling, Spreds can suppress both types (Taniguchi K., unpublished data and Ref. [18]). Indeed, microRNA-126 deletion suppresses VEGF-A-induced ERK activation in endothelial cells and angiogenesis through the increase of Spred1 [27][28][29]. However, the effect of Sprouty deletion is more specific to VEGF-A signaling than to VEGF-C signaling, while the effect of Spred deletion is more specific to VEGF-C signaling than to VEGF-A signaling (Taniguchi K., unpublished data and Ref. [18]). In fact, Sprouty2 and Sprouty4 single-deficient mice showed defects of blood vessels rather than lymphatic vessels, while Spred1/ Spred2 DKO mice showed abnormal lymphatic vessel development and nearly normal blood vessel development ( Fig. 2 and Ref. [18]). In addition to this difference in expression, these results suggest that Sproutys and Spreds might have different functions in endothelial cells. VEGF-A/VEGFR-2 signaling is Ras-independent and PLC-c/PKC-dependent, while VEGF-C/VEGFR-3 signaling is Ras-dependent and PKC-independent (Taniguchi K., unpublished data and Ref. [4]). Therefore, drawing an analogy from the different functions of Sproutys and Spreds in Xenopus tropicalis, we propose that Sproutys inhibit PLC-c/PKC-dependent VEGF-A signaling and angiogenesis, while Spreds inhibit Rasdependent VEGF-C signaling and lymphangiogenesis.
Although, in ischemia-induced angiogenesis, VEGF-A is thought to be the primary angiogenesis-stimulating factor [30], angiogenesis is the more complex process, as it is triggered not only by VEGF-A but also by bFGF, S1P, angiopoietins, and others [12][13][14]. In fact, it is reported that bFGF gene therapy is effective to treat critically ischemic limb [31]. It is already known that Sproutys can inhibit various RTK signals [1,2]. We have also shown that loss of Sprouty expression results in hyperactivation of VEGF-A and bFGF signaling as well as S1P and LPA signaling (Fig. 5F, Taniguchi K., unpublished data and Ref. [19] and [21]). It is reported that in vivo shRNA targeting SHP-1 also accelerated angiogenesis in a rat model of hind limb ischemia [32]. While SHP-1 inhibits only RTK signals, Sproutys suppress both RTK and GPCR signals. Thus the suppression of Sproutys could be beneficial.
Inhibition of negative feed-back loops leading to profound and long term activation of signals often lead to a dysregulation of neovascularisation since the overshooting response is inducing immature vessels. However, excessive sprouting in response to inhibition of Sproutys results in the formation of mature vessels. Angiogenesis is a complex process that includes the recruitment and proliferation of various cells, such as endothelial cells, mural cells [smooth muscle cells (SMC) and pericytes], endothelial progenitor cells (EPCs) and others. It is reported that Sprouty-family genes are expressed in both endothelial cells and smooth muscle cells [33], and we have confirmed that Sprouty/Spred family genes are also expressed in bone marrow (Taniguchi K., unpublished data). It is possible that Sproutys function not only in endothelial cells, but also in mural cells, EPCs or myeloid cells, and that enhanced angiogenesis of mature vessels in Sprouty4 KO mice and the results of our experiments with in vivo shRNA targeting Sproutys are partially dependent on the enhanced function or the increased number of mural cells, EPCs or myeloid cells. Moreover, it is possible that Sproutys are also associated with angiopoietins signals, which are important for the maturation of blood vessels. Further study is necessary to investigate these possibilities.
Sprouty4 KO mice were more resistant to ischemia than WT mice were in mouse models of ischemia (Fig. 3, Fig. 4), and neovascularization induced by a tumor transplantation model was also accelerated by Sprouty4 deficiency (Taniguchi K., unpublished data). Moreover, in vivo shRNA targeting Sprouty2 and Sprouty4 accelerated angiogenesis in a mouse model of hind limb ischemia (Fig. 7). In this study, muscle tissue injected with the Sproutys shRNA vectors exhibited a significant decrease in Sproutys transcripts (Fig. 5A). This knockdown efficiency may be due to the fact, in skeletal muscle, the efficiency of intramuscular gene transfer has been shown to be augmented from five-to seven-fold when the injected muscle is ischemic [34]. The present study is the first to uncover these significant implications for gene therapy using the Sprouty2 and Sprouty4 shRNA vectors for the treatment of peripheral ischemic diseases. The fact that Sproutys exhibit such broad suppression activity, inhibiting a wide variety of angiogenic factors and cells, indicates that the suppression of Sproutys must enhance neovascularization.
In conclusion, Sproutys are physiologically important regulators of angiogenesis in vivo and may be useful as new therapeutic targets for peripheral ischemic diseases.

Mice
Sprouty2 KO mice and Sprouty4 KO mice have been described previously [21,25]. Sprouty2 KO mice and Sprouty4 KO mice were generated as 129/C57BL/6J mixed background, and then backcrossed into C57BL/6J at least five times. Gender-matched, WT littermates were used as controls. All experiments using these mice were approved by and performed according to the guidelines of the Animal Ethics Committee of Kyushu University, Fukuoka, Japan.

RT-PCR and real-time PCR analysis
The cells or tissues were lysed in RNAiso (TAKARA BIO, Shiga, Japan) for RNA preparation. Total RNA was isolated through fluorescence activated cell sorting (FACS), which sorted about 5.0610 4 BECs and LECs at embryonic day 14.5, as previously reported [18]. Good separation of BECs and LECs was confirmed by BEC markers (Nrp1, CD44) and LEC markers (LYVE-1, Prox1). Total RNA was reverse transcribed using the High Capacity cDNA Reverse Transcription Kit (Applied Biosystems, Foster City, CA, USA), and the product was used for further analysis. PCR products were separated on 2.0% agarose gel stained with ethidium bromide. The expression level of GAPDH was evaluated as an internal control. The primer sequences for RT-PCR were as follows: Sprouty2-F, 59-TTTT-AATCCACCGATTGCTTGG-39; Sprouty2-R, 59-GCTGCACT-CGGATTATTCCATC-39; Sprouty4-F, 59-CAGCTCCTCAAA-GACCCCTAGAAGC-39; Sprouty4-R, 59-GTGCTGCTACTGC-TGCTTACAGAGC-39; GAPDH-F, 59-ACCACAGTCCATGC-CATCAC-39 and GAPDH-R 59-TCCACCACCCTGTTGC-TGTA-39. The primer sequences for BEC and LEC markers were described elsewhere [36]. Real-time PCR was performed on cDNA samples using an ABI 7000 Sequence Detection System (Applied Biosystems) with the SYBR Green system (Applied Biosystems). The relative quantitation value is expressed as 2 2Ct , where Ct is the difference between the mean Ct value of triplicates of the sample and of the endogenous GAPDH control. The primer sequences for real-time PCR were as follows:

Retinal angiography
Flat-mounted retinas were evaluated using fluorescein-dextran angiography as described elsewhere [22]. The mice were deeply anesthetized and a 0.03 ml/g body weight 50 mg/mL solution of 2610 6 molecular weight FITC-dextran (Sigma, St Louis, MO, USA) was perfused through the left ventricle. The eyes were enucleated and fixed in 4% paraformaldehyde for at least 3 h. The corneas and lenses were then removed, and the peripheral retinas were dissected and flat-mounted on microscope slides for examination under a fluorescence microscope.

Vessel quantitative analysis
The vascular area in the ear, skin or muscle was quantified as a PECAM-1/CD31-positive area from ten 610 micrographs, using Image J software (http://rsb.info.nih.gov), as described elsewhere [37]. LYVE-1-positive vessels in the skin were quantified in a similar manner.

RNAi-mediated knockdown
The mammalian expression vector pSUPER.retro.puro (Oligoengine, Seattle, WA, USA) was used for expression of shRNA targeting murine Sprouty2. The sequence of the Sprouty2 shRNA is 59-GCCGGGTTGTCGTTGTAAA-39 and corresponds to nucleotides 1150-1168 of mSprouty2. Murine Sprouty4 shRNA (29-mer) and control plasmids were purchased from Origene (Rockville, MD, USA) and used according to the manufacturer's protocols. The specificity of Sprouty2 and Sprouty4 knockdown was confirmed by real-time PCR or immunoblotting of whole-cell lysates of MEFs with anti-Sprouty2 and anti-Sprouty4 antibodies, respectively. The relative intensities of Sprouty2 or Sprouty4 band were normalized by STAT5 expression using Image J software, as previously described [9,35].

In vivo models of ischemia
A hind limb ischemia model was performed as previously described [23,38]. Male WT and Sprouty4 KO mice (8-10 weeks old) and male C57BL/6J mice (8 weeks old) were used for a hind limb ischemia model. The proximal portion of the right femoral artery including the superficial and the deep branch and the distal portion of the saphenous artery were occluded with an electrical coagulator. After 2 weeks, we determined the ischemic (right)/nonischemic (left) limb blood flow ratio by using a laser Doppler blood flow imager (Laser Doppler Perfusion Imager System, moorLDI-Mark 2; Moor Instruments, Wilmington, DE, USA). Before initiating scanning, mice were placed on a heating pad at 37uC to minimize variations in their body temperatures. Calculated perfusion is expressed as the ratio of ischemic to nonischemic hind limb perfusion. At 2 wk after femoral resection, adductor muscles from the ischemic and control limbs were embedded in OCT compound. Eight-micron sections were stained with anti-PECAM-1/CD31 antibody (BD Pharmingen) and anti-rat Ig secondary antibody. For in vivo shRNA targeting Sprouty2 and Sprouty4, a hind limb ischemia model was performed. Immediately after ischemia was induced, either a total of 40 mg Sproutys shRNA vectors (20 mg Sprouty2 shRNA vector and 20 mg Sprouty4 shRNA vector) or a quantity of control shRNA vectors was injected into five different sites in the adductor muscle of each anesthetized mouse [32]. A model of soft tissue ischemia similar to one described elsewhere [24,39] was developed. The model consisted of lateral skin incisions (2.5 cm in length and 1.25 cm apart) created on the dorsal surface of mice, penetrating the skin, dermis, and underlying adipose tissue. The overlying skin was undermined, and a 0.13mm-thick silicone sheet was inserted to separate the skin from the underlying tissue bed. The skin was then reapproximated with 6-0 nylon sutures.

Corneal micropocket assay
The mouse corneal micropocket assay and quantification of neovascularization were performed as described elsewhere [40], using male BALB/c mice (6-10 weeks old). For local delivery, shRNA plasmids (total 10 mg/10 ml per eye) were diluted in phosphate-buffered saline (PBS) and delivered subconjunctivally. The subconjunctival injections were given after hydron pellet implantation.

Statistical analysis
Data are expressed as mean6SD or mean6SEM. Statistical significance was tested with an unpaired two-tailed Student's t-test or analysis of variance (ANOVA). The differences were considered to be significant if P,0.05.