Effects of Somatostatin Analogues on Retinal Angiogenesis in a Mouse Model of Oxygen-Induced Retinopathy: Involvement of the Somatostatin Receptor Subtype 2

Abstract

purpose. To determine whether selective activation or blockade of the somatostatin (SRIF) receptor 2 (sst₂) with two SRIF analogues, octreotide and D-Tyr⁸ cyanamid 154806 (CYN), influences retinal vascularization and levels of vascular endothelial growth factor (VEGF) and its receptors VEGFR-1 and -2 in a mouse model of oxygen-induced retinopathy (OIR).

methods. Wild-type (WT), sst₁-knockout (KO), and sst₂-KO mice were used. The OIR model was used to test the effects of octreotide and CYN administered subcutaneously. Retinopathy was assessed by a retinal scoring system using fluorescein-perfused retinal wholemounts. Retinal levels of VEGF, VEGFR-1, and -2 were evaluated with quantitative RT-PCR, Western blot, and ELISA.

results. In both WT and sst₁-KO mice, OIR-induced neovascularization was reduced by octreotide, whereas it was increased by CYN. No effects of octreotide and CYN on retinal neovascularization were observed in sst₂-KO retinas. Hypoxia upregulated the expression of VEGF and its receptors. Compared with WT retinas, the increase in VEGF, but not in VEGF receptors, was less pronounced in sst₁-KO retinas in which sst₂ is known to be overexpressed. The hypoxia-induced increase in VEGF and its receptors was affected by SRIF analogues, with ameliorative effects of octreotide and worsening effects of CYN, which were more pronounced in the presence of sst₂ overexpression.

conclusions. These data suggest that sst₂ regulates angiogenic responses to the hypoxic insult through a modulation of retinal levels of VEGF and its receptors. The present results further support the possibility of the use of sst₂-selective ligands in the treatment of retinopathy.

Neovascular ocular diseases represent a major health threat to all age groups and especially the rapidly increasing diabetic patient population. Proliferative diabetic retinopathy is the leading cause of blindness in the young-to-middle-age population in the Western world. Although the vascular insult varies across different neovascular ocular diseases, the proliferation of aberrant blood vessels is common to all. 

The naturally occurring growth hormone inhibitor, somatostatin (somatotropin release-inhibiting factor, SRIF) and its receptors (sst₁–sst₅) ¹ are found in the neuroretina of humans and other mammals, and growing evidence suggests that in the retina, SRIF acts both as a neuromodulator and an antiangiogenic factor (for a review, see Ref. ² ). In particular, there is a general agreement that SRIF combats neovascularization associated with diabetic retinopathy, ³ although the mechanism of action remains to be elucidated. 

There is evidence mostly from in vitro studies that sst₂ may mediate the angioinhibitory activity of SRIF. ⁴ However, little is known of the effects of sst₂ activation on neovascularization and its associated factors in in vivo models. Ameliorating effects of the sst₂-preferring agonist octreotide on neovascularization with concomitant reduction of the levels of vascular endothelial growth factor (VEGF) have been recently demonstrated in rat models of portal hypertension. ^{5 6} In recent years, octreotide has been used in clinical trials of diabetic retinopathy treatments, but with conflicting results. ^{7 8} In a mouse model of oxygen-induced retinopathy (OIR), there is indication that sst₂ agonists, including octreotide, inhibit retinal neovascularization. ^{9 10 11} In this model, we have recently demonstrated that the severity of angiogenic responses to hypoxia is correlated to the expression level of sst₂ in the retina. ¹² The lack of sst₂, as in sst₂-knockout (KO) retinas, is associated with the more severe effects of hypoxia on retinal neovascularization and proangiogenic factors, whereas a chronic overexpression of sst₂ (consequent to the genetic deletion of sst₁ as in sst₁-KO retinas ^{13 14} ) attenuates hypoxia’s effects on proangiogenic factors. Thus, the use of these KO mice represents a useful tool for assessing the effect of SRIF analogues on hypoxia-induced retinal neovascularization. In the present study, retinas of wild-type (WT) and sst₁- and sst₂-KO mice were rendered hypoxic and were used to investigate whether sst₂ activation or blockade influences vascular responses to the hypoxic insult and its associated factors. sst₂-KO mice were used as a control to assess the specific action of SRIF analogues on sst₂. Since the severity of angiogenic responses to hypoxia appears to be related to the expression level of sst₂, in sst₁-KO mice we evaluated whether sst₂ overexpression influences octreotide’s or CYN′s effects on angiogenic responses to the hypoxic insult. 

Materials and Methods

Octreotide was purchased from NeoMPS (Strasbourg, France). The PCR mastermix (iQ Sybr Green Supermix) was from Bio-Rad (Hercules, CA). Primers were obtained from MWG Biotech (Ebersberg, Germany). A nucleic acid gel stain (GelStar) was obtained from Cambrex (East Rutherford, NJ). Rabbit polyclonal antibodies to VEGF, VEGFR-1, and VEGFR-2 were obtained from Santa Cruz Biotechnologies (Santa Cruz, CA). Enhanced chemiluminescence reagent was from Millipore (Billerica, MA). All other chemicals were obtained from Sigma-Aldrich (St. Louis, MO). 

Animals

Experiments were performed on 265 mice of WT (C57BL/6) and sst₁- or sst₂-KO strains of both sexes at postnatal day (PD) 17 (6 g body weight). sst₁- and sst₂-KO mice were generated as previously reported. ^{15 16} Experiments were performed in accordance with the ARVO Statement for the Use of Animals in Ophthalmic and Vision Research and in compliance with the Italian law on animal care No. 116/1992 and the EEC/609/86. All efforts were made to reduce the number of animals used. 

Model of Hypoxia-Induced Retinopathy

In a typical model of hypoxia-induced retinopathy, ¹⁷ litters of mouse pups with their nursing mothers were exposed in an infant incubator to high oxygen concentration (75% ± 2%) between PD7 and PD12, before return to room air between PD12 and PD17. Oxygen was checked twice daily with an oxygen analyzer (Miniox I; Bertocchi srl Elettromedicali, Cremona, Italy). Individual litters were reared in either oxygen or room air. Within litters, differential treatment was performed including no treatment, octreotide, CYN, and sham injection. The animals were treated with injections for 5 days from PD12 to PD16. They were anesthetized by ip injection of Avertin (1.2% tribromoethanol and 2.4% amylene hydrate in distilled water; 0.02 mL/g body weight). All experiments were performed at the same time of day to exclude possible circadian influences. The data were collected from both males and females and the results combined, as there was no apparent sex difference. 

Administration of SRIF Analogues

The sst₂-preferring peptidyl agonist octreotide and the sst₂-selective peptidyl antagonist D-Tyr ⁸ cyanamid 154806 (CYN) ^{18 19} were given twice daily subcutaneously from PD12 to PD16. Octreotide was used at 0.02 mg kg⁻¹ dose⁻¹. ^{10 11} CYN was used at concentrations ranging from 0.01 to 2.0 mg kg⁻¹ dose⁻¹ in agreement with previous works using sst₂ antagonists. ^{20 21 22 23 24} Comparable concentrations of nonpeptide SRIF receptor agonists were recently used by Palii et al. ⁹ in the OIR model. In rats, the SRIF analogue BIM-23627 has been found to exert antagonistic effects on sst₂ after subcutaneous administration. ²¹ The analogues were dissolved in 33 mM acetate buffer (pH 5) with 135 mM NaCl. Sham injections were performed with that vehicle. 

Assessment of Retinal Vascularization

Fluorescein-conjugated dextran perfusion of the retinal vessels was performed as previously described. ²⁵ Briefly, in anesthetized animals a median sternotomy was performed, and the left ventricle was perfused with 2 mL of a 25-mg/mL solution of fluorescein-conjugated dextran in 0.15 M phosphate buffer. The eyes were enucleated, the retinas were dissected, and flatmounts were obtained. They were viewed by fluorescence microscopy (Eclipse E800; Nikon, Badhoevedorp, Netherlands), and images were acquired (DFC320 camera; Leica Microsystems, Wetzlar, Germany). Neovascularization was evaluated by using the retinopathy scoring system shown in Table 1according to previously published protocols. ¹¹ Three trained observers evaluated the number of clock hours with abnormal vessels for each retina. The data were averaged and are expressed in values ranging from 0 to 11. 

Isolation of Retinal RNA and cDNA Preparation

After death, the eyes were rapidly removed and retinas were dissected, immediately frozen in liquid nitrogen, and stored at −80°C. Total RNA from four samples (each containing three retinas) for each experimental condition was extracted (RNeasy Mini Kit; Qiagen, Valencia, CA), according to the manufacturer’s instructions. The purified RNA was suspended in RNase-free water and quantified spectrophotometrically (SmartSpec 3000; Bio-Rad). First-strand cDNA was generated from 1 μg of total RNA (QuantiTect Reverse Transcription Kit; Qiagen). cDNA samples were aliquoted and stored at −20°C. 

Real-Time Quantitative RT-PCR

Real-time quantitative RT-PCR (QPCR) was performed with a kit (SYBR Green on MiniOpticon Two-Color Real-time PCR detection system; Bio-Rad). QPCR primer sets were designed with Primer3 software ²⁶ (VEGFR-1 and -2) or obtained from either Primer Bank ²⁷ (VEGF) or RTPrimerDB ²⁸ (Rpl13a). Forward and reverse primers were chosen to hybridize to unique regions of the appropriate gene sequence; their sequences are listed in Table 2 . PCR reactions including 1 μL of cDNA, 300 nM of each primer, 7.5 μL of mastermix (iQ Sybr Green Supermix; Bio-Rad), and RNase-free water to complete the reaction mixture volume to 15 μL. Each target gene was run concurrently with Rpl13a, a constitutively expressed control gene. Negative control reactions were set up as just described, but without any template cDNA. All reactions were run in triplicate. The QPCR was performed with hot-start denaturation step at 95°C for 3 minutes, and then was performed for 40 cycles at 95°C for 10 seconds and 58°C for 20 seconds. The fluorescence was read during the reaction, allowing a continuous monitoring of the amount of PCR product. Primers were initially used to generate a standard curve over a large dynamic range of starting cDNA quantity which allows calculation of the amplification efficiency for each of the primer pairs (Table 2) . After amplification, first derivative melting curve analysis was used to confirm that the signal corresponded to a unique amplicon. QPCR products were analyzed by nucleic acid staining on a gel stain (GelStar; Cambrex) containing 3% agarose to verify the correct product size. As previously described, samples were compared by using the relative cycle threshold (CT method). ²⁹ The increase or decrease (x-fold) was determined relative to a control after normalizing to RPL13a. All reactions were run in triplicate. After statistical analysis, the data from the different experiments were plotted and averaged in the same graph. 

Western Blot Analysis

Western blot analysis was performed on proteins extracted from three samples for each experimental condition, as previously described. ¹⁴ Each sample contained five retinas. Briefly, retinas were homogenized in 10 mM Tris-HCl (pH 7.6) containing 5 mM EDTA, 3 mM EGTA, 250 mM sucrose, 1 mM phenylmethylsulfonyl fluoride, 1 μM pepstatin, 10 μg/mL leupeptin, and 2 μg/mL aprotinin (buffer A) and centrifuged at 22,000g for 30 minutes at 4°C. The supernatant was used for VEGF detection. The pellet was resuspended in 20 mM HEPES (pH 7.4) containing 150 mM NaCl, 5 mM EDTA, 3 mM EGTA, 4 mg/mL n-dodecyl-β-maltoside, and the proteinase inhibitors just listed and centrifuged at 22,000g for 30 minutes at 4°C. The supernatant was used to detect either VEGFR-1 or -2. Protein concentration was determined according to Bradford, ³⁰ with bovine serum albumin used as a standard. Aliquots of each sample containing equal amounts of protein were subjected to 10% sodium dodecyl sulfate–polyacrylamide gel electrophoresis according to Laemmli. ³¹ β-Actin was used as the loading control. The gels were transblotted onto polyvinylidene difluoride membrane, and the blots were blocked for 1 hour at room temperature with 3% milk. The blots were then incubated overnight at 4°C with primary rabbit polyclonal antibodies directed to VEGF (1:200 dilution), VEGFR-1 (1:100 dilution), and VEGFR-2 (1:100 dilution) or with a primary mouse monoclonal antibody directed to β-actin (1:2500 dilution). Finally, blots were incubated for 1 hour at room temperature with mouse anti-rabbit horseradish peroxidase–labeled secondary antibody (1:5000 dilution) or with rabbit anti-mouse horseradish peroxidase-labeled secondary antibody (1:25,000 dilution) and developed with the enhanced chemiluminescence reagent. The optical density (OD) of the bands was evaluated on computer (Quantity One software; Bio-Rad). The data were normalized to the level of β-actin. All experiments were run in duplicate. After statistical analysis, data from the different experiments were plotted and averaged in the same graph. 

Enzyme-Linked Immunosorbent Assay

VEGF protein concentration was determined on retinal lysate by enzyme-linked immunosorbent assay (ELISA) according to the manufacturer’s protocol (Quantikine Mouse VEGF ELISA kit; R&D Systems, Minneapolis, MN). The measurement was performed on three samples for each experimental condition. According to Nagai et al., ³² two retinas for each sample were placed into 200 μL of buffer A and sonicated for 30 seconds at 50 Hz. The lysate was centrifuged at 22,000g for 15 minutes at 4°C. Protein concentration was measured as just described. VEGF concentration was determined spectrophotometrically (Microplate Reader 680 XR; Bio-Rad) at 450 nm. In each experiment, all samples and standards were measured in duplicate. Data were collected as picograms VEGF per milligrams of total protein and, after statistics, averaged in the same graph. 

Statistical Analysis

Retinopathy scores were evaluated by the Kruskal-Wallis test for the overall group and by the Dunn’s post hoc test to determine differences between groups. Retinopathy score data are represented as the median (25th, 75th quartiles). All other data were analyzed by the Kolmogorov-Smirnov test on verification of normal distribution. Statistical significance was evaluated with ANOVA followed by the Newman-Keuls multiple-comparison test. The results are expressed as mean ± SE of data for the indicated n values (Prism; GraphPad Software, San Diego, CA). Differences with P < 0.05 were considered significant. 

Results

Retinal Neovascularization

In agreement with Dal Monte et al., ¹² we found that hypoxic retinas of WT and sst₁-KO mice did not significantly differ in their median total retinopathy score, whereas the median retinopathy score in sst₂-KO retinas was significantly higher than in WT or sst₁-KO mice (P < 0.01, Table 3 ). Vehicle injections did not affect neovascularization in any strain. WT and sst₁-KO pups treated with octreotide had a significant improvement in their retinopathy scores when compared with oxygen and oxygen+vehicle–treated animals (P < 0.05, Table 3 ). In both WT and sst₁-KO mice, the retinopathy score was significantly increased by CYN at a dose of 0.5 mg kg⁻¹ (P < 0.01 in WT and P < 0.05 in sst₁-KO, Table 3 ). In sst₂-KO mice treated with octreotide or CYN, the median retinopathy score did not differ from that measured in oxygen and oxygen+vehicle–treated mice. No differences were found between oxygen and oxygen+vehicle–treated mice when compared with mice treated with octreotide or with CYN with respect to the different subscores, except for the score for blood vessels tufts. In both WT and sst₁-KO mice, octreotide treatment caused a significant decrease in blood vessel tufts with respect to untreated or vehicle-treated animals. The opposite (i.e., a significant increase in blood vessel tufts) was obtained after CYN treatment. Representative retinal wholemounts of WT and KO mice are shown in Figure 1 . 

VEGF, VEGFR-1, and VEGFR-2

We evaluated whether selective activation or blockade of sst₂ with SRIF analogues may be related to variations in retinal levels of VEGF and its receptors, VEGFR-1 and -2, in the OIR model of WT and sst₁-KO mice. In all experiments, vehicle treatment did not affect retinal the levels of VEGF, VEGFR-1, or VEGFR-2. 

Messengers

QPCR yielded amplified products at 105, 116, and 114 bp, which correspond to VEGF, VEGFR-1, and VEGFR-2 mRNA, respectively (Fig. 2A) . In agreement with Dal Monte et al., ¹² our findings showed that hypoxic retinas of WT and sst₁-KO mice displayed significantly increased levels of VEGF, VEGFR-1, and VEGFR-2 mRNA. In particular, the relative increase in VEGF mRNA was lower in sst₁-KO than in WT mice, whereas that in VEGF receptor messengers did not differ significantly between WT and sst₁-KO mice (Figs. 2B 2C 2D) . As shown in Figure 3 , in WT, octreotide decreased significantly the hypoxia-induced increase in VEGF and VEGFR-2 messengers (∼27% and 30%, respectively, P < 0.05), whereas it did not affect VEGFR-1 mRNA. In sst₁-KO, octreotide significantly decreased the VEGF, VEGFR-1, and VEGFR-2 messengers (∼38%, 44% and 31%, respectively, P < 0.05). In both WT and sst₁-KO mice, CYN at a dose of 0.5 mg kg⁻¹ increased significantly the relative levels of VEGF and VEGFR-2 messengers (in WT: ∼62%, P < 0.01 and 47%, P < 0.05, respectively; in sst₁-KO: ∼40% and 32%, respectively, P < 0.05), whereas it did not affect VEGFR-1 mRNA. In sst₁-KO mice, the CYN-induced increase in VEGF mRNA was significantly lower than that observed in WT mice (∼15%, P < 0.05). 

Proteins

To test whether the effects of octreotide or CYN on VEGF and VEGF receptor messengers would have comparable effects on VEGF and VEGF receptor proteins, mice were treated with SRIF analogues, and protein content was analyzed by Western blot and ELISA. Densitometric analysis of the immunoblots demonstrated that normoxic retinas of WT and sst₁-KO mice had comparable levels of VEGF, VEGFR-1, and VEGFR-2 and that hypoxia significantly increased these levels, in agreement with previous results (Fig. 4) . ¹² As shown in Figure 5A , in hypoxic mice treated with vehicle, VEGF levels were significantly lower in sst₁-KO than in WT (∼15%, P < 0.05). As also shown in Figure 5A , octreotide significantly decreased VEGF protein expression in both WT and sst₁-KO mice (∼19% and 22%, respectively, P < 0.05) with stronger effects in sst₁-KO than in WT (∼17%, P < 0.05). In contrast, CYN at 0.5 mg kg⁻¹ significantly increased VEGF protein in both WT and sst₁-KO (∼20% and 18%, respectively, P < 0.05). The CYN-induced increase in VEGF was significantly lower in sst₁-KO than in WT (∼16%, P < 0.05). The relative levels of VEGF receptors did not differ significantly between WT and sst₁-KO (Figs. 5B 5C) . Octreotide did not affect VEGFR-1 protein in the WT, but reduced its level in the sst₁-KO (∼25%, P < 0.05). In contrast, VEGFR-2 protein was comparably reduced by octreotide in either WT or sst₁-KO (∼16% and 30%, respectively, P < 0.05). VEGFR-1 protein was unaffected by CYN in both WT and sst₁-KO, whereas VEGFR-2 protein was comparably increased by CYN in both strains (∼27% and 19%, respectively, P < 0.05). To confirm the data obtained with Western blot and to evaluate whether the effects of SRIF analogues in sst₁-KO mice were dependent on sst₂ overexpression, VEGF protein levels were quantitated with ELISA. In agreement with results from the Western blot analysis, normoxic VEGF was significantly increased by hypoxia in both WT and sst₁-KO retinas (Fig. 6A) . The VEGF increase was lower in sst₁-KO than in WT (∼28%, P < 0.01). Measurements of VEGF levels in normoxic retinas are in line with previous results in the rodent retina. ^{32 33 34} Our determination of VEGF levels after hypoxia gave values in the range of those previously reported in mice. ^{34 35} As shown in Figure 6B , ELISA quantitation confirmed a significant decrease in VEGF protein formation in octreotide-treated animals of both strains (∼32% and 31%, respectively, P < 0.05). After octreotide, the VEGF level in sst₁-KO retinas was lower than in WT retinas (∼23%, P < 0.05). WT mice treated with CYN at 0.5 mg kg⁻¹ showed a significant increase in VEGF levels (∼99%, P < 0.001). In sst₁-KO mice, a dose-dependent increase in VEGF was observed after treatment with increasing concentrations of CYN with no effects at 0.01 mg kg⁻¹, an increase at 0.5 mg kg⁻¹ (∼77%, P < 0.01) but significantly lower than in WT (∼36%, P < 0.01), and reaching a maximum at 2.0 mg kg⁻¹, a concentration four times higher than that used in the WT (∼150%, P < 0.001). 

Discussion

The results of this study add further evidence to the notion that sst₂ is protective against angiogenesis and indicate for the first time that antiangiogenic effects of sst₂ may be mediated by an sst₂-induced downregulation of the VEGF system in the OIR model. 

SRIF Analogues

An ameliorating effect of octreotide on retinal neovascularization has been previously demonstrated. ^{9 10 11} In addition, in the OIR model, systemic administration of non–peptidyl SRIF receptor agonists has been shown to inhibit retinal neovascularization in a dose-dependent manner, comparable to octreotide. ⁹ As previously reported, octreotide may act on SRIF receptors other than sst₂ (i.e., sst₃ and sst₅), according to the affinity data in vitro, but octreotide affinity for sst₂ is higher than that for sst₃ and sst₅. ³⁶ Although it is difficult to conjecture about the actual drug concentration that reaches the retina, we cannot exclude an involvement of sst₃ and/or sst₅ in the observed effects. On the other hand, evidence of sst₃ and sst₅ presence/localization in the retina are still elusive. ² In addition, selective agonists for sst₃ and sst₅ have been found to elicit no effects in the mouse retina. ³⁷ Finally, our previous finding that the lack of sst₂ is associated with heavier effects of hypoxia on retinal neovascularization ¹² highlights an important role of sst₂ on retinal angiogenesis. That ameliorating effects of octreotide in the OIR model are mediated by sst₂ is demonstrated by the lack of effects in sst₂-KO mice shown in the present study. 

CYN is a synthetic peptide with in vitro and in vivo properties consistent with a pure sst₂ antagonist. ^{18 19} CYN effects in vivo are scarcely documented. In rats, CYN administration affects the activity of mesenteric afferent fibers innervating the jejunum. ²⁴ In the acute pilocarpine rat seizure model, the anticonvulsant effects of angiotensin IV are influenced by intracerebroventricular administration of CYN. ³⁸ In the rat striatum, CYN administration using in vivo microdialysis blocks the SRIF-induced stimulation of dopamine. ³⁹ The results of the present work are the first demonstration of the sst₂ antagonistic properties of CYN in an OIR model. In our work, CYN concentrations affecting both retinal neovascularization and the VEGF system are in line with those used in previous work. Indeed, the intravenous administration of CYN doses ranging from 1.0 to 3.0 mg kg⁻¹ blocks the octreotide-evoked inhibition of baseline discharge of mesenteric afferent nerves of the jejunum in the rat. ²⁴ In addition, the microinjection in the rat dorsal vagal complex of 0.1 nanomoles CYN (∼0.5 μg kg⁻¹) prevents the SRIF-induced inhibition of pancreatic secretion. ²⁰ Moreover, another sst₂ antagonist, BIM-23627, dose dependently decreases growth hormone secretion in dexamethasone-treated rats, with minimum effects at 0.02 mg kg⁻¹ and maximum effects at 2.0 mg kg⁻¹. ²³ In another study, BIM-23627 at 0.5 mg kg⁻¹ has been found to improve the catabolic effects of dexamethasone on different parameters, including growth, epididymal fat accumulation, glucose homeostasis, and insulin activity. ²¹ In addition, the intracerebroventricular application of 2.5 nanomoles BIM-23627 (∼0.01 mg kg⁻¹) blocks sst_2A internalization in a rat model of middle cerebral artery occlusion. ²² In our OIR model, there were findings in favor of the specificity of CYN. Indeed, the lack of CYN′s effects on retinal neovascularization in sst₂-KO mice confirmed the specific action of CYN at sst₂. In addition, the dose dependency of CYN′s effects on the VEGF levels in sst₁-KO retinas (described in the next section) indicated that the CYN concentration required to activate overexpressed sst₂ was four times higher than in WT retinas. 

Effects of SRIF Analogues on Neovascularization

In agreement with previous results, ¹² the neovascularization response to hypoxic insult is significantly worsened by the lack of sst₂, whereas a chronic overexpression of sst₂ (as in sst₁-KO retinas) does not attenuate retinal neovascularization. In agreement with previous findings, ^{9 10 11} this study describes ameliorative effects of octreotide on retinal neovascularization in the OIR model. Recently, nonpeptide SRIF receptor agonists specifically targeting sst₂ have been found to inhibit retinal neovascularization in the OIR model. ⁹ Our additional result that blocking of sst₂ results in a worsened angiogenic response to hypoxic insult is in line with the finding that retinal neovascularization increases in sst₂-KO mice ¹² and suggests a constitutive activity of the receptor in hypoxic conditions. In line with this possibility, a constitutive activity of sst₂ in ischemic conditions has been recently demonstrated in an in vitro model of the ischemic mouse retina. ⁴⁰ In this respect, a recombinant truncated form of the rat sst₂ has been shown to display agonist-independent, constitutive activity, ⁴¹ and constitutive activation of sst₂ has been demonstrated in mouse pituitary corticotroph cells. ⁴²  

Effects of SRIF Analogues on the VEGF System

Both experimental and clinical studies have reported an important role of VEGF in pathologic retinal angiogenesis. ⁴³ However, the sites of VEGF release and those of VEGF actions in the retina are poorly understood. Of the VEGF receptors, VEGFR-2 is considered to be the receptor that mediates functional VEGF signaling in endothelial cells, whereas the function of VEGFR-1 and the possibility that VEGFR-1 affects endothelial cells remain to be fully disclosed. ⁴⁴ Our finding that hypoxia increases the levels of VEGF and its receptors is consistent with previous findings. ^{12 45 46 47 48} Our additional finding that SRIF analogues influence VEGF production is in agreement with previous results from in vitro studies. In human glioma cells, for instance, SRIF or the sst₂-selective agonists octreotide and L-054522 potently inhibit basal VEGF production with less effect on hypoxia-induced VEGF release. ⁴⁹ In human retinal pigment epithelial cells, SRIF reduces VEGF mRNA through sst₂. ⁵⁰ In pituitary tumor cells, SRIF influences VEGF secretion differently, depending on the spectrum of expressed SRIF receptor subtypes, including sst₂. ⁵¹ A few results are available in in vivo models of proliferating endothelium. A recent study, in particular, demonstrates that octreotide markedly decreases splanchnic neovascularization and VEGF expression in portal hypertensive rats. ⁶ In addition, octreotide ameliorates histomorphologic changes, including vessel dilatation and VEGF increase associated with portal hypertensive enteropathy in rats. ⁵ Our recent results in the mouse retina demonstrate that enhanced somatostatinergic function at sst₂, as in sst₁-KO mice, limits the hypoxia-induced VEGF increase whereas sst₂ loss upregulates this increase, ¹² thus suggesting a role for sst₂ in the regulation of VEGF homeostasis. In this study, we provided the first demonstration that sst₂ is coupled to a modulation of the VEGF system in an in vivo model of retinal neovascularization. Indeed, the hypoxia-induced increase in VEGF and VEGFR-2 was consistently reduced by octreotide whereas it was drastically increased by sst₂ blockade with CYN. In particular, the fact that VEGFR-2, but not VEGFR-1, was influenced by SRIF analogues is in line with the finding that VEGFR-2 is the receptor principally involved in retinal angiogenesis and suggests that it may act downstream of the sst₂-induced modulation of retinal neovascularization. In this respect, the finding that in hypoxic retinas, octreotide downregulates VEGF and VEGFR-2 is particularly intriguing in light of a possible therapeutic use of sst₂ agonists to counteract retinal neovascularization by influencing the VEGF system. The hypoxic level of VEGF after octreotide treatment was lower in sst₁-KO than in WT mice, indicating that in WT, sst₂ may be saturated by 0.02 mg kg⁻¹ octreotide. In sst₁-KO retinas, we also observed a dose dependency of CYN′s effects on VEGF levels, thus confirming a specific action of CYN at sst₂ and suggesting that overexpressed sst₂ is not saturated by CYN at doses below 2.0 mg kg⁻¹. The finding that hypoxic levels of VEGFR-1 are downregulated by octreotide in sst₁-KO but not in WT mice is in line with previous results demonstrating that the density of retinal sst₂ represents the rate-limiting factor for sst₂-mediated effects. ³⁷  

Receptor-Mediated Effects of SRIF Analogues

Despite the growing evidence that sst₂ agonists may exert angioinhibitory activity, the exact role of receptor-mediated effects in the retina is still unknown. The simplest explanation is that in our model SRIF analogues act on sst₂ expressed by endothelial cells of retinal blood vessels. However, sst₂ does not seem to be expressed by retinal endothelial cells in rodents, although there is evidence that sst₂ is localized to numerous cells in the neuroretina. ⁵² One possibility is that quiescent vascular endothelial cells do not express sst₂ and that this receptor is expressed when the endothelial cells begin to grow. In this respect, it has been reported that, in cultured human veins, proliferating angiogenic sprouts of vascular endothelium express sst₂. ⁵³ In addition, immunostaining for sst₂ can be detected in proliferating but not quiescent human umbilical vein endothelial cells. ⁵⁴ Moreover, in human eyes with late stages of age-related maculopathy and characterized by choroidal neovascularization, newly formed endothelial cells strongly express sst₂. ⁵⁵ Finally, the pattern of sst₂ expression in patients with diabetic retinopathy indicates that beneficial effects of sst₂ agonists may depend on the presence of sst₂ on newly formed vessels. ⁵⁶ Recently, a relationship between sst₂ levels and octreotide efficacy to inhibit angiogenesis has been demonstrated in rats with advanced stages of portal hypertension. ⁶ The issue of dynamic changes in sst₂ expression in mouse models of hypoxia-induced growing vascular endothelium remains to be addressed. 

Clinical Implication

Most of the therapeutic interventions against diabetic retinopathy have focused on the mechanisms and the factors leading to new vessel growth. In this respect, based on experimental evidence, there has been great interest in the development of anti-VEGF compounds for the treatment of proliferative diabetic retinopathy, such as antibodies, antisense oligonucleotides, and small interfering RNAs that target VEGF. However, anti-VEGF therapy shows a loss of efficacy as the neovasculature develops. Moreover, the therapy stabilizes the disease rather than improving vision, and the regression of neovascularization is rarely permanent. ⁵⁷ On the other hand, several limited clinical trials using the SRIF analogue octreotide in the treatment of retinopathy has been performed giving mixed results about the efficacy of the treatment (for a review, see Ref. ⁸ ). For instance, in a small-scale randomized controlled study of 23 patients with severe nonproliferative diabetic retinopathy or early proliferative diabetic retinopathy, octreotide has been found to reduce the requirement for laser photocoagulation compared with conventional treatment. However, the two large-scale, muticenter, randomized placebo-controlled clinical trials initiated by Novartis (Basel, Switzerland), have been inconclusive on the effect of octreotide on the progression of retinopathy, but have shown some benefit on visual acuity. ³ Our finding that in hypoxic retinas, octreotide downregulates VEGF and VEGFR-2 is particularly intriguing in light of a possible therapeutic use of sst₂ agonists to counteract retinal neovascularization by influencing the VEGF system and strongly supports the validity of combination drug therapy for vasoproliferative disorders in the retina. 

Conclusions

SRIF analogues may interfere with the retinopathy-producing cascade at different levels, including proangiogenic factors. We have demonstrated that SRIF analogues with high specificity for sst₂ regulate retinal angiogenesis and that their effects involve a modulation of the VEGF system. Thus, the effectiveness of SRIF analogues in diabetic retinopathy could be due to local growth factor–lowering effects, although direct antiangiogenic effects of these compounds cannot be excluded. 

 Supported by the Italian Ministry of University and Research (MUR, PRIN, Grant 2005052312).

 Submitted for publication January 15, 2009; revised March 3, 2009; accepted June 5, 2009.

 Disclosure: M. Dal Monte, None; C. Ristori, None; M. Cammalleri, None; P. Bagnoli, None

 The publication costs of this article were defrayed in part by page charge payment. This article must therefore be marked “advertisement” in accordance with 18 U.S.C. §1734 solely to indicate this fact.

 

Table 1.

 

Retinopathy Scoring System

Criteria	Points
		0	1	2	3	4
Blood vessel tufts	None	In <3 clock hours	In 3–5 clock hours	In 6–8 clock hours	In 9–12 clock hours
Central vasoconstriction	None	Mild, early zone 1 (inner 50% of zone 1)	Moderate, throughout zone 1 (outer 50% of zone 1)	Severe, extending to zone 2
Retinal hemorrhage	Absent	Present
Blood vessel tortuosity	None	<1/3 of vessels	1/3–2/3 of vessels	>2/3 of vessels

 Flatmounted retinas were examined by fluorescence microscopy, and retinopathy was quantified according to four criteria. Points received for each criterion were summed to obtain the retinopathy score. A higher score (range 0–11) indicates more severe retinopathy. The scoring system was adapted from Higgins et al. ¹¹

Table 2.

 

Primers for PCR Analysis

Gene	Primer Sequence	Product Length (bp)	QPCR: Amplification Efficiency* (%)	Gene Bank Accession No.
VEGF	Forward: GCACATAGGAGAGATGAGCTTCC	105	98.8	NM_009505
	Reverse: CTCCGCTCTGAACAAGGCT
VEGFR-1	Forward: GAGGAGGATGAGGGTGTCTATAGGT	116	95.5	NM_010228
	Reverse: GTGATCAGCTCCAGGTTTGACTT
VEGFR-2	Forward: GCCCTGC TGTGGTCTCACTAC	114	96.4	NM_010612
	Reverse: CAAAGCATTGCCCATTCGAT
Rpl13a	Forward: CACTCTGGAGGAGAAACGGAAGG	182	97.3	NM_009438
	Reverse: GCAGGCATGAGGCAAACAGTC

*  Data calculated by Opticon Monitor 3 software (Bio-Rad).

Table 3.

 

Retinopathy Subscores and Total Score

	WT							sst1 KO							sst2 KO
		Hypoxia	+ Vehicle	+ Octreotide	+ Cyanamid	Hypoxia	+ Vehicle	+ Octreotide	+ Cyanamid	Hypoxia	+ Vehicle	+ Octreotide	+ Cyanamid
Number of animals	6	6	12	8	6	6	10	8	6	6	6	6
Blood vessel tufts	2 (1, 2)	2 (1.5, 2.5)	1 (1, 2)*	3 (2, 3.5), †	2 (1, 2)	1.5 (1, 2)	1 (1, 1)*	2.5 (2, 3)*	4 (2.5, 4), ‡	3.5 (2.5, 4), §	3.5 (2.5, 4)	4 (2.5, 4)
Central vasoconstriction	1 (1, 1)	1 (1, 1)	1 (1, 1)	1 (1, 2)	1 (1, 1)	1 (1, 1.5)	1 (1, 2)	1 (1, 1.5)	3 (2.5, 3), ‡	3 (2, 3), §	3 (2.5, 3)	3 (2.5, 3)
Hemorrhage	1 (0, 1)	1 (0, 1)	1 (0, 1)	1 (0, 1)	1 (0, 1)	1 (0.5, 1)	1 (0, 1)	1 (0.5, 1)	1 (1, 1)	1 (1, 1)	1 (1, 1)	1 (1, 1)
Blood vessel tortuosity	2 (1, 2)	2 (1, 2)	2 (1, 2)	2 (1, 2)	2 (1, 2)	2 (1, 2)	2 (1, 2)	2 (1, 2)	3 (2, 3), ‡	3 (2, 3), §	3 (2.5, 3)	3 (2, 3)
Total retinopathy score	6 (5, 6)	6 (5, 6)	5 (4, 5)*	7 (6, 8.5), †	6 (5, 6)	6 (5, 6)	5 (4, 5)*	6 (5.5, 7)*	10 (9, 11), ‡	10 (8.5, 10.5), §	10 (9, 11)	10 (9, 11)

 Data are expressed as the median (25th, 75th quartiles).

*   P < 0.05 vs. respective vehicle treated.

†   P < 0.01 vs. respective vehicle treated.

‡   P < 0.01 vs. hypoxic WT.

§   P < 0.01 vs. vehicle treated WT (Kruskal-Wallis, followed by Dunn’s post hoc test).

Figure 1.

 

Flatmounted retinas perfused with fluorescein-dextran of PD17 WT (A, D, G, J), sst₁ (B, E, H, K)- and sst₂ (C, F, I, L)-KO mice exposed to room air (A–C) or to 75% ± 2% oxygen from PD7 to PD12 (D–L), untreated (D–F) or treated with SRIF analogues (octreotide in G–I and CYN in J–L). Hyperoxia followed by normoxia for 5 days produced the central loss of blood vessels and the formation of vessel tufts with more evident effects in sst₂-KO retinas. In both WT and sst₁-KO retinas, octreotide provided improvement in blood vessel tufts, whereas CYN increased pathologic retinal vascularization. SRIF analogues were without effects in sst₂-KO retinas. Scale bar, 1 mm.

Figure 2.

 

Levels of VEGF, VEGFR-1, and VEGFR-2 mRNAs in mouse retinas. (A) PCR products of VEGF (105 bp), VEGFR-1 (116 bp), and VEGFR-2 (114 bp) and the housekeeping gene Rpl13a (182 bp) in normoxic WT retinas. (B–D) VEGF, VEGFR-1, and VEGFR-2 mRNAs in normoxic (□) and hypoxic (▪) conditions. QPCR evaluation showed that VEGF, VEGFR-1, and VEGFR-2 messengers were increased by hypoxia (*P < 0.05 and **P < 0.01 versus the respective normoxic; ANOVA). The increase in VEGF mRNA (B) was lower in sst₁-KO than in WT retinas (§P < 0.05 versus hypoxic WT; ANOVA). Data were analyzed by the formula 2^−ΔΔCT using Rpl13a as internal standard. Each column represents the mean ± SE of data from four samples. Each sample refers to the mRNA extracted from three retinas.

Figure 3.

 

QPCR evaluation of VEGF (A), VEGFR-1 (B) and VEGFR-2 (C) in hypoxic WT (□) and sst₁-KO (▪) retinas after treatment with vehicle, 0.02 mg kg⁻¹ octreotide or 0.5 mg kg⁻¹ CYN. (A) Octreotide decreased VEGF levels in both WT and sst₁-KO retinas (*P < 0.05 versus the respective untreated retinas; ANOVA), whereas CYN increased VEGF mRNA (*P < 0.05 and **P < 0.01 versus the respective untreated retinas; ANOVA). The CYN-induced increase was lower in sst₁-KO than in WT retinas (§P < 0.05 versus CYN-treated WT; ANOVA). (B) Octreotide did not affect VEGFR-1 mRNA in WT retinas, whereas it decreased VEGFR-1 mRNA in sst₁-KO retinas (*P < 0.05 versus the respective untreated retinas; ANOVA). CYN did not influence VEGFR-1 mRNA neither in WT nor in sst₁-KO. (C) VEGFR-2 mRNA was decreased by octreotide, whereas it was increased by CYN in both WT and sst₁-KO retinas (*P < 0.05 versus the respective untreated retinas; ANOVA). Data were analyzed by the formula 2^−ΔΔCT using Rpl13a as internal standard. Each column represents the mean ± SE of data from four samples. Each sample refers to the mRNA extracted from three retinas. oct, octreotide.

Figure 4.

 

Levels of VEGF (A), VEGFR-1 (B), and VEGFR-2 (C) in normoxic (□) and hypoxic (▪) conditions, as evaluated by Western blot using β-actin as a loading control in WT and sst₁-KO retinas. (A–C) Densitometric analysis showed that VEGF, VEGFR-1, and VEGFR-2 were increased by hypoxia (*P < 0.05 and **P < 0.01 versus the respective normoxic; ANOVA). The VEGF increase (A) was lower in sst₁-KO than in WT retinas (§P < 0.05 versus hypoxic WT; ANOVA). Each column represents the mean ± SE of data from three samples. Each sample refers to the protein extracted from five retinas.

Figure 5.

 

Levels of VEGF (A), VEGFR-1 (B), and VEGFR-2 (C) in hypoxic WT (□) and sst₁-KO (▪) retinas after treatment with vehicle, 0.02 mg kg⁻¹ octreotide or 0.5 mg kg⁻¹ CYN, as evaluated by Western blot using β-actin as a loading control. (A) VEGF level in vehicle-treated sst₁-KO was significantly lower than in WT (§P < 0.05 versus vehicle-treated WT; ANOVA). In both WT and sst₁-KO retinas, octreotide decreased VEGF which was, in contrast, increased by CYN (*P < 0.05 versus the respective vehicle-treated retinas; ANOVA). Compared with WT, in sst₁-KO retinas, the VEGF decrease after octreotide was higher, whereas the increase in VEGF after CYN was lower (§P < 0.05 versus octreotide- or CYN-treated WT retinas; ANOVA). (B) Octreotide did not affect VEGFR-1 in WT retinas, whereas it decreased VEGFR-1 levels in sst₁-KO retinas (*P < 0.05 versus the respective untreated retinas; ANOVA). CYN did not influence VEGFR-1 in WT or in sst₁-KO retinas. (C) VEGFR-2 was decreased by octreotide, whereas it was increased by CYN in both WT and sst₁-KO retinas (*P < 0.05 versus the respective untreated retinas; ANOVA). Each column represents the mean ± SE of data from three samples. Each sample refers to the protein extracted from five retinas. oct, octreotide.

Figure 6.

 

VEGF protein levels in WT and sst₁-KO retinas, as quantified with ELISA. (A) VEGF levels in normoxic (□) and hypoxic (▪) conditions. In both WT and sst₁-KO, normoxic VEGF was increased by hypoxia (***P < 0.001 versus the respective normoxic; ANOVA). The VEGF increase was lower in sst₁-KO than in WT retinas (§§P < 0.01 versus hypoxic WT; ANOVA). (B) VEGF levels in hypoxic WT (□) and sst₁-KO (▪) retinas after treatment with vehicle or SRIF analogues. VEGF level in vehicle-treated sst₁-KO was significantly lower than in WT (§P < 0.05 versus vehicle-treated WT; ANOVA). In both WT and sst₁-KO, octreotide was used at 0.02 mg kg⁻¹. In WT, CYN was used at 0.5 mg kg⁻¹, whereas increasing concentrations were used in sst₁-KO. In both WT and sst₁-KO retinas, octreotide decreased VEGF (*P < 0.05 versus respective vehicle-treated retinas; ANOVA). In sst₁-KO retinas, the VEGF decrease after octreotide was higher than in WT (§P < 0.05 versus octreotide-treated WT; ANOVA). WT mice treated with CYN showed a significant increase in retinal levels of VEGF (***P < 0.001 versus vehicle-treated WT; ANOVA). In sst₁-KO retinas, a dose-dependent increase of VEGF was observed after treatment with increasing concentrations of CYN, with no effects at 0.01 mg kg⁻¹; an increase at 0.5 mg kg⁻¹ (**P < 0.01 versus vehicle-treated sst₁-KO; ANOVA), but a significantly lower than in WT (§§P < 0.01 vs. 0.5 mg kg⁻¹ CYN-treated WT; ANOVA); and maximum at 2.0 mg kg⁻¹ (***P < 0.001 versus vehicle-treated sst₁-KO; ANOVA). Each column represents the mean ± SE of data from three samples. Each sample refers to the protein extracted from two retinas. oct, octreotide.

The authors thank Angelo Gazzano and Gino Bertolini (University of Pisa, Italy) for assistance with the mouse colonies.