Molecular Vision 2003; 9:549-558 <http://www.molvis.org/molvis/v9/a67/> Received 1 July 2003 | Accepted 6 October 2003 | Published 8 October 2003

Download Reprint

Normal vascular development in mice deficient in endothelial NO synthase: Possible role of neuronal NO synthase

Mohamed Al-Shabrawey,^1,2 Azza B. El-Remessy,^1,3,4 Xiaolin Gu,^1,5 Steven S. Brooks,^1,3 Mohamed S. Hamed,² Paul Huang,⁶ Ruth B. Caldwell^1,3,5
 
 

¹Vascular Biology Center and the Departments of ³Ophthalmology, ⁴Pharmacology, and ⁵Cellular Biology and Anatomy, Medical College of Georgia, Augusta, GA; ²Mansoura Faculty of Medicine, Mansoura, Egypt; ⁶Department of Medicine, Cardiovascular Research Center, Harvard Medical School, Boston, MA

Correspondence to: Ruth B. Caldwell, Ph.D., Vascular Biology Center, Medical College of Georgia, Augusta, GA, 30912-2500; Phone: (706) 721-6145; FAX: (706) 721-9799; email: rcaldwel@mail.mcg.edu

Abstract

Purpose: Nitric oxide formation by nitric oxide synthase (NOS) has been implicated in vascular injury and retinal neovascularization during oxygen-induced retinopathy. However, the role of NOS in normal retinal vascular development and growth has not been studied. The purpose of these experiments was to characterize the expression of NOS in relation to vascular development and to determine the effect of deleting endothelial NOS (eNOS) on this process.

Methods: Retinal vascular development was analyzed in 150 eNOS+/+ and eNOS-/- mice ranging from 1 day to 6 months old by using a combination of morphometric and biochemical approaches. The pattern of vascular development was analyzed in retinal tissue sections and whole-mount preparations labeled with fluorescein-conjugated Griffonia simplicifolia lectin. Analysis of vascular density and arterial diameter were performed with the lectin-labeled whole-mounts using computer-assisted morphometry. NO production was quantified by measuring retinal levels of nitrate/nitrite accumulation using the Greiss reaction. Western blotting techniques with isoform-specific NOS antibodies were used to evaluate differences in levels of NOS protein expression. Retinal distribution of nNOS was characterized using nNOS immunocytochemistry and NADPH diaphorase histochemistry.

Results: These analyses showed that the rate and pattern of retinal vascular development in eNOS-/- mice were comparable with those in wild-type control mice. Measurement of vascular density showed no significant differences between the two strains. The amount of NO production in the eNOS-/- retina was also equivalent to that in the eNOS+/+ retina. Analysis of nNOS expression within the eNOS+/+ and eNOS-/- mice showed similar levels of total nNOS protein in the two strains. Inducible NOS was not detected in either strain. Studies of nNOS distribution showed intense labeling of the deep capillary plexus in the eNOS-/- retina. This was not seen in the wild-type retinas. The number of neuronal cells showing NADPH-diaphorase activity was also significantly increased in the eNOS-/- mice.

Conclusions: Development of the retinal vasculature occurs normally without eNOS. The observations of similar levels of NO production, perivascular redistribution of nNOS and increased numbers of NADPH-diaphorase reactive neurons in the eNOS-/- retinas suggest that increases in vascular-associated nNOS activity compensate for the eNOS deficiency in the developing mutant retina.

Introduction

Nitric oxide (NO) is a ubiquitous second messenger with many physiological functions, ranging from dilation of blood vessels to neuronal development and synaptic activity [1,2]. NO is synthesized by the conversion of L-arginine to citrulline by a group of isoenzymes known as nitric oxide synthases (NOS). There are three known isoforms of NOS, two of which are constitutively produced and activated by increased intracellular calcium. One is found predominantly in vascular endothelial cells (eNOS) [3], another in neurons (nNOS) [4]. A third isoform (iNOS) is not constitutively expressed, but can be induced in macrophages, smooth muscle cells, glia and many other cell types upon cytokine stimulation or under conditions of inflammation. The iNOS isoform is calcium/calmodulin independent and is active upon formation [5].

The roles of NO produced by constitutive NOS as neurotransmitter/neuromodulator and vasodilator are well established [1,2,6-8]. In endothelial cells, NO has been shown to signal the actions of vascular endothelial growth factor (VEGF) in increasing cell proliferation, migration, and permeability [9-14], suggesting a role in vascular growth and remodeling. This relationship, and the well-known role that VEGF has in stimulating angiogenesis in the developing retina [15-17], led us to hypothesize that eNOS-derived NO may also be a critical factor regulating normal retinal vascular development. Furthermore, eNOS has recently been implicated in sensitizing capillaries in the developing retina to oxygen-induced retinopathy, thereby potentiating vascular injury and the potential for retinopathy of prematurity (ROP) [18,19].

Previous studies of the developing retina have suggested a role for nNOS in neuronal development and maturation [20,21]. A close association of nNOS-positive amacrine cells with retinal vessels has been described in the adult rat and human retinas [22]. However, the role of NOS in retinal vascular development has not been determined. We were therefore interested in better understanding the role that NOS might play in retinal vascular development.

To look at this issue experimentally, we performed immunohistochemical and biochemical studies to analyze retinal vascular development in relation to NOS activity. Similar analyses were also performed in a genetically modified mouse with a knock-out deletion of the eNOS gene. The effects of the eNOS deletion on vascular development and NOS activity were evaluated by morphometric and biochemical methods.

Methods

Animals

These experiments were done using C57Bl/6 (eNOS+/+) mice and eNOS-deficient mutant mice (eNOS-/-). The eNOS-/- mice, which have been described previously [23], have been backcrossed to the C57BL/6 strain for ten generations. Animal care guidelines comparable to those published by the US Public Health Service (Public Health Service Policy on Humane Care and Use of Laboratory Animals) were followed. A total of 150 mice ranging in age from postnatal day 1 (P1) to 6 months were used for these analyses. P1 is defined as the first day after birth.

Blood pressure determination

Blood pressure was determined in conscious mice aged 3 to 6 months using a computerized tail-cuff system (Visitech Systems, Inc., Apex, NC), which has been described previously [24]. Four restraining units are mounted on a surface maintained at 37 °C. Mice tails were passed through a cuff and immobilized between light source above and a photoresistor below. Blood flow in the tails produces oscillating wave forms that are digitally sampled 200 times per second per channel. The system first determines full waveform amplitude (A) and then seeks waveforms having amplitudes less than 20% of A. When five consecutive waveforms are less than 20% of A, this will accurately represent blood pressure [25]. Twenty-four measurements were used for determination of the BP of each mouse.

Vascular morphology

Retinal vascular distribution was analyzed using frozen sections (12 μm) and flat mounts labeled with biotinylated Griffonia (Bandeiraea) simplicifolia isolectin B4 (Vector Laboratories, Burlingame, CA) and fluorescein-conjugated Avidin D (Vector Laboratories) as described previously [26]. For quantitation of vascular density, confocal microscope images were collected systematically from each retinal quadrant. Beginning 0.2 mm from the optic disc three non-overlapping photographs (0.5 mm² in area) were taken in each quadrant for each retina. The central retina was defined as the 4 areas surrounding the optic disk. The peripheral retina was defined as the remaining areas (i.e. the area beginning 0.7 mm from the optic disk and extending to the ora seratta). The number of photographs available for this peripheral region ranged from 4 to 8 per retina due to occasional tissue damage during dissection and processing of the flat mounts. Density of the microvasculature within each image was determined by using computer-assisted morphometric software (Image-1/Metamorph Imaging System, Universal Imaging Corporation) to calculate microvessel length per unit area. The major arterioles and venules were excluded from this analysis. Arterial size in the eNOS-/- and eNOS+/+ mice was determined by measuring the diameter of the central retinal artery branches immediately before and after the first branch. These quantitative analyses were based on samples from 5 mice in each age and strain group.

Retinal angiography

Vascular patency was evaluated using retinal angiography performed as described previously [27]. Briefly, mice were deeply anesthetized (with 0.5 ml/kg of a mixture of 1.5 ml xylazine HCl, 1.5 ml ketamine HCl, 0.5 ml acepoject injected IM) and injected through the left ventricle with 1 ml of PBS containing 50 mg of FITC-dextran (2x10⁶ kDa, Sigma). The mice were sacrificed, and then the eyeballs were removed and fixed in 4% paraformaldehyde in PBS. Retinas were dissected, flat mounted, and analyzed by fluorescence microscopy.

Measurement of NO

NO production was determined by measuring the levels of nitrite and nitrate, the oxidized products of NO, in the supernatant of PBS-homogenate retinas by modified Greiss reagent assay. Briefly, 210 ml of homogenate was incubated with nitrate reductase enzyme (10 mU) and NADPH (12.5 mM) for 30 min at 37 °C. Then the total nitrite in each sample was determined by addition of 200 mU of L-glutamate dehydrogenase, 100 mM ammonium chloride and freshly prepared 4 mM of α-ketogluterate. Enzymatic reduction of nitrate to nitrite was monitored by including a nitrate standard. The mixture was incubated at 37 °C for 10 min followed by addition of 250 ml of Greiss reagent and incubation for another 5 min at 37 °C. The absorbance at 543 nm was recorded and concentrations of NO₂ were calculated from a standard curve constructed using sodium nitrite and sodium nitrate as standards. Protein levels were measured by the Bradford method (Bio Rad, Hercules, CA) and nitrite/nitrate level was expressed as μmoles/mg protein.

Western blotting

Proteins (50 μg) from retinas, abdominal aortas, and cerebral capillaries prepared as described [28] were electrophoresed on a 10% polyacrylamide gel. Protein was transferred to nitrocellulose membrane and reacted with antibodies against nNOS, eNOS, or iNOS (Transduction Lab, Lexington, KY). The primary antibodies were detected by peroxidase-conjugated secondary antibodies. Visualization of the immunoreactive proteins was accomplished using ECL reagents (Amersham Life Science Inc., Arlington Heights, Illinois). Relative densities of the immuno-specific protein bands were determined by densitometry.

Immunolocalization of nNOS

Retinal distribution of nNOS was determined using immunohistochemical techniques performed according to our previous methods with minor modifications [24]. Briefly, 12 μm cryosections were incubated for 72 h at 4 °C with polyclonal anti-nNOS antibody (1:100, Transduction Lab). The primary antibody was detected with fluorescein-labeled secondary antibodies (Fisher, Pittsburgh PA). Results were analyzed by confocal microscopy. In control experiments where the primary antibody was omitted, no immunostaining was evident.

NADPH-diaphorase staining

NADPH-diaphorase (NADPH-d) staining was done on retina flat mounts and 12 μm frozen sections using previously described methods [29]. Tissues were fixed briefly in 4% paraformaldhyde in 0.1 M phosphate buffered saline (pH 7.6), washed in Tris buffer (pH 7.6), permeabilized with 5% Triton X-100 in Tris buffer (pH 7.6) and reacted for NADPH-d using 0.25 mg/ml nitroblue tetrazolium (NBT), 1 mg/ml β-NADPH (Sigma) and 0.5% Triton X-100 in 0.1 M PBS at 37 °C for 1 h. Retinas from age-matched eNOS+/+ and eNOS-/- mice were processed simultaneously under identical reaction conditions. Whole retinas or sections were mounted on gelatinized slides, dehydrated, cover-slipped and analyzed by light microscopy. In control experiments, in which β-NADPH was omitted from the reaction, staining was not observed (data not shown).

Camera lucida and digital planimetry methods were used for quantitation of the numbers of NADPH-d-positive cells. For this analysis drawings of labeled cells were made of 4 to 8 systematically selected 1 mm² square areas in each retina beginning at 200 μm from the optic disc and continuing to the ora serrata.

Statistical analysis

Each experiment was repeated at least three times. Results were expressed as mean ± standard error. Statistical significance was determined by unpaired t-tests or ANOVA. A p value less than 0.05 was considered significant.

Results

Retinal vessels develop normally in eNOS-/- mice

Analysis of lectin-labeled frozen sections showed identical patterns of vascular development in both eNOS-/- and eNOS+/+ retinas. The developing vessels were restricted to the nerve fiber layer until P7 at which time the penetrating vessels began to appear within the inner plexiform layer (Figure 1A,B). By P9 numerous penetrating vessels had reached the outer plexiform layer (Figure 1C,D). Thus, the period between P7 and P9 marks the angiogenic sprouting stage of retinal vascular development. By P14 the deep vessels were more densely packed and were located closer to the ora serrata than before in both strains (Figure 1E,F). At this time the deep vessels were seen in both inner and outer plexiform layers with communicating channels connecting them to the superficial vessels. By P30 the deep vessels had reached the ora serrata (Figure 1G,H).

Analysis of retinal flat mounts also showed comparable patterns of vascular formation in the two strains at each age examined between P1 and P30. By P7 the radial organization of the vessels was clearly apparent (Figure 2A,C). Veins and arteries could be distinguished on the basis of a continuous capillary network around the veins and a capillary-free zone around the arteries. For quantitative analysis of microvascular density, computer assisted morphometic analysis was done on skeletonized digital images from the retinal flat mount preparations (Figure 2B,D). This analysis showed no significant difference between the two strains in vessel length/100 μm² at P7, P14, or P30 (Figure 2E,F; p>0.05). Measurement of the diameter of the major arteries before and after the first branch point at P7 also showed no difference between the two strains (Table 1). However, as the animals matured to young adults (P30), the arterial diameter in the eNOS-/- retina was significantly smaller than that in the eNOS +/+ retina (p<0.05). This is consistent with the increase in blood pressure seen in adult eNOS-/- mice (145±14 mm Hg vs 107±6.1 mm Hg in the eNOS+/+ mouse). Measurement of blood pressure in younger mice was impossible due to the small size of the mice.

Retinal angiography with high molecular weight fluorescein-dextran confirmed that the vessels in the eNOS-/- retina were patent and intact throughout development and into young adulthood (Figure 3A,B,C). However, after three months occasional, small areas of focal vascular leakage were observed (Figure 3D).

Levels of NOS activity in the eNOS-/- retina are normal

The above data showing that retinal vessels in the eNOS-/- mouse develop normally suggested that increases in activity of another NOS isoform compensate for the eNOS deficiency. In order to test this possibility, studies were done using an assay for nitrate/nitrite formation. The results showed that total nitrite/nitrate levels in the eNOS-/- retinas were the same or slightly higher than those in the eNOS+/+ retinas, suggesting that a compensatory increase occurs in activity of another NOS isoform (μmoles/mg protein mean±stadard error=85.87±16.76 μmoles/mg protein for eNOS-/- retinas vs 77.25±13.86 for eNOS+/+ retinas, p>0.05, unpaired t-test, n=4).

Retinal vessels in the eNOS-/- mouse are immunoreactive for nNOS

Because nNOS has been found to compensate for eNOS deficiency in the pial arteries of eNOS-/- mice [30], additional studies were done to see if changes occur in either the amount or distribution of nNOS protein in the mutant retina. Western blotting analysis showed similar levels of nNOS protein in both mutant and wild-type mice at P9, the time of maximum angiogenic sprouting (Figure 4). Further analysis of samples isolated from retina, aorta, and brain capillaries at P30 also showed similar levels of nNOS protein. These experiments also confirmed the lack of eNOS expression in the eNOS-/- mouse. iNOS protein was not detected in either strain.

We next evaluated whether changes in the distribution of nNOS occur in the eNOS-/- retina. Immunolocalization analyses of the eNOS+/+ retinas showed strong nNOS immunoreactivity in the nerve fiber, ganglion cell, inner plexiform, inner nuclear and outer plexiform layers. However, the deep vascular plexus was not reactive. By contrast, the deep vessels in the eNOS-/- retina were strongly immunoreactive for nNOS (Figure 5).

NADPH-diaphorase reactive cells are increased in the eNOS-/- retina

In order to further evaluate the apparent alteration in nNOS distribution/activity in the mutant retina, we also studied the distribution of NADPH-diaphorase activity. NADPH-diaphorase has been shown to be a reliable marker of nNOS activity in neuronal tissue [31-33]. Analyses of retina flat-mounts from eNOS+/+ mice showed diaphorase activity beginning at P7. Weak activity was seen in large vessels and in a few neurons. By P9, numerous diaphorase-positive cells with neuron-like morphology were evident (Figure 6A). At P14 and P30, the reaction in the larger vessels was stronger and diaphorase activity was also evident within the microvasculature (Figure 6B,C). Many of the diaphorase-positive cells were closely related to the blood vessels. In the eNOS-/- retinas, NADPH-diaphorase activity was also present in the retinal vasculature and neuron-like cells (Figure 7). However, the vascular labeling appeared slightly later (beginning at P9) and was weaker than that in the eNOS+/+ retinas. In contrast with the reduction in vascular reactivity in eNOS-/- retina, the number of NADPH-d-reactive neurons was markedly increased. Quantitative analysis showed significant increases in the numbers of NADPH-d-reactive cells in eNOS-/- retina at P9 and P14 as compared with the age-matched eNOS+/+ retinas (Figure 8, p<0.05). As development continued to P30, the number of NADPH-d-reactive cells in the eNOS-/- retina declined to control levels.

Analysis of frozen sections showed qualitatively similar distributions of NADPH-d-labeling in the eNOS+/+ and eNOS-/- retinas, with labeling in the nerve fiber layer, ganglion cell layer, inner nuclear layer, and both inner and outer plexiform layers (Figure 9). However, as noted above, NADPH-d-reactive neurons were more numerous in the developing eNOS-/- retina.

Discussion

Angiogenesis in the retina is a critical process in vascular development as well as recovery from injury. It may also have pathological, sight-threatening consequences when it produces neovascularization of the vitreous (e.g., ROP, diabetes, retinal vein occlusion) or subretinal space (e.g., age-related macular degeneration, ocular histoplasmosis). Elucidating the molecular mediators of angiogenesis is therefore of great clinical importance.

Our study revealed that mice lacking a functional eNOS gene demonstrate normal development in terms of both timing and organization of the retinal vasculature. Although this finding does not exclude a role for NO, it clearly shows that eNOS is not critical in this regard.

Because VEGF is known to be essential for both vasculogenesis and angiogenesis [34,35] and because NO is critically involved in VEGF signaling [9,11,36], we conclude that NO formation in the eNOS-/- retina most likely occurs by compensatory production from another NOS isoform. Because iNOS was not detected and because NO-mediated VEGF signaling is calcium dependent [36], it is likely that nNOS substitutes for eNOS in the mutant retina. This substitution phenomenon has been previously reported to occur in the pial arteries of eNOS-/- mice, in which acetylcholine-induced vasorelaxation was mediated by nNOS-dependent activation of GMP cyclase [30]. We suggest that the compensatory activity of nNOS is important for normal vascular development in the eNOS-/- mouse retina and for maintaining adequate oxygen delivery during retinal metabolic activity. Expression of the VEGF receptors VEGFR-1 and VEGFR-2 has been demonstrated within extravascular neural cells of the developing mouse retina and in cultured Muller glial cells [37]. Thus, VEGF could induce nNOS activation and NO production in the eNOS-/- retinas via activation of VEGFR-2 in perivascular neurons and/or glia.

The concept of a compensatory mechanism for the eNOS deficiency is strongly supported by the results of our nitrate/nitrite assay showing equal levels of NOS activity in the eNOS-/- and eNOS+/+ retinas as well as by our immunolocalization analysis showing perivascular redistribution of nNOS protein within the retinas of the eNOS-/- mice. In the wild-type retinas, perivascular nNOS labeling was associated only with larger vessels of the inner retina. In contrast, eNOS-/- mice showed strong nNOS immunoreactivity associated with the deep vascular plexus as well as the superficial vessels. Our western blotting analysis showed similar amounts of nNOS protein in the eNOS-/- and eNOS+/+ retinas. However, compensatory increases in nNOS activity could occur without any increase in total amount of nNOS protein due to local changes in nNOS distribution and/or increases in nNOS activation upon the accumulation of intracellular calcium. Further investigation is required to specify the cells responsible for the vascular-associated nNOS expression in the eNOS-/- retina and also to show whether or not endothelial cells in eNOS-/- mice are able to express nNOS.

NADPH-diaphorase analyses also showed a redistribution of NOS activity within the retinas of eNOS-/- mice. The number of neuron-like cells showing NADPH-diaphorase reaction was significantly increased in the eNOS-/- retina at P9 and P14. Moreover, the processes of these NADPH-reactive neuron-like cells were often in close proximity to the blood vessels. NADPH-diaphorase histochemistry showed that NOS activity was first detectable within the blood vessels of wild-type retinas between P7 and P9. Vascular NOS activity was evident slightly later in eNOS deficient retinas, beginning at P9. This is the same time that prominent vascular-associated nNOS immunoreactivity is seen in the mutant retina. Because the P7 to P9 period is the stage of vascular development in which the penetrating vessels begin to form, these observations suggest that constitutive NOS activity is more important in formation of the deep network than in formation of the primary network. The physiological significance of this difference is unclear. Recent studies suggest that both superficial and deep vascular networks are formed by the same mechanism of angiogenic sprouting [38,39]. However, the superficial vessels differ from the deep vessels in that they form by aligning along the retinal glial network. It is possible that the delay in the appearance of vascular NADPH-diaphorase activity in the mutant retina represents the time required for the retinal cells to compensate for the absence of eNOS by redistributing and/or upregulating nNOS activity within the perivascular glial processes.

Other, NO-independent, compensatory mechanisms may also occur that allow normal development of retinal vessels in eNOS-/- mice. For example, basic fibroblast growth factor and transforming growth factor-β are both NO-independent mediators of angiogenesis, and both are present in the retina [40,41]. Further studies are needed to investigate this issue.

Our finding that mean arterial blood pressure is higher in the adult eNOS-/- mice than in the eNOS+/+ mice suggests that nNOS compensation, if it is indeed occurring, is either transient or incomplete. We were not able to measure blood pressure in the neonatal mice, but our analysis of diameters of the major retinal arterioles showed no significant difference between the eNOS-/- and wild type mice on P7, whereas by P30 the large arteries in eNOS-/- mice showed a statistically significant decrease in arterial caliber. It is likely that this change is a result of the vaso-constriction that occurs in the absence of endothelium-derived NO. The reason why this vasoconstriction effect is not seen prior to P30 is not clear, but this difference could be due to differences in regulation of vascular function between the mature and developing retinas.

It is important to note that the eNOS-/- mouse retinas did not show any evidence of vascular occlusion. Fluorescein angiography did show some areas of increased vascular permeability in the older eNOS-/- mice. This effect is probably due to altered vascular function associated with chronic hypertension [42].

In summary, eNOS and nNOS are both expressed in the developing mouse retina. Deletion of eNOS does not prevent normal vascular development, or cause significant changes in vascular density or permeability in the developing retina. The perivascular redistribution of nNOS around the smallest vessels in eNOS-/- mice may indicate a compensatory mechanism to increase production of NO within the microvasculature.

Acknowledgements

This work was completed as partial fulfillment of requirements for the PhD degree from Mansoura University, Mansoura, Egypt. The studies were supported in part by grants from the NIH (NIH-EY01766; NIH-EY04617); an unrestricted grant to the Department of Ophthalmology of the Medical College of Georgia from Research to Prevent Blindness, Inc.; and a pre-doctoral fellowship from the Egyptian Cultural Bureau Ph.D. to M. Al-Shabrawey.

References

1. Moncada S, Palmer RM, Higgs EA. Nitric oxide: physiology, pathophysiology, and pharmacology. Pharmacol Rev 1991; 43:109-42.

2. Snyder SH, Bredt DS. Biological roles of nitric oxide. Sci Am 1992; 266:68-71, 74-7.

3. Forstermann U, Gorsky LD, Pollock JS, Ishii K, Schmidt HH, Heller M, Murad F. Hormone-induced biosynthesis of endothelium-derived relaxing factor/nitric oxide-like material in N1E-115 neuroblastoma cells requires calcium and calmodulin. Mol Pharmacol 1990; 38:7-13.

4. Bredt DS, Snyder SH. Isolation of nitric oxide synthetase, a calmodulin-requiring enzyme. Proc Natl Acad Sci U S A 1990; 87:682-5.

5. Lowenstein CJ, Glatt CS, Bredt DS, Snyder SH. Cloned and expressed macrophage nitric oxide synthase contrasts with the brain enzyme. Proc Natl Acad Sci U S A 1992; 89:6711-5.

6. Furchgott RF, Zawadski JV. The obligatory role of endothelial cells in the relaxation of arterial smooth muscle by acetylcholine. Nature 1980; 288:373-6.

7. Ignarro LJ, Buga GM, Wood KS, Byrns RE, Chaudhuri G. Endothelium-derived relaxing factor produced and released from artery and vein is nitric oxide. Proc Natl Acad Sci U S A 1987; 84:9265-9.

8. Garthwaite J. Neural nitric oxide signalling. Trends Neurosci 1995; 18:51-2.

9. Leibovich SJ, Polverini PJ, Fong TW, Harlow LA, Koch AE. Production of angiogenic activity by human monocytes requires an L-arginine/nitric oxide-synthase-dependent effector mechanism. Proc Natl Acad Sci U S A 1994; 91:4190-4.

10. Papapetropoulos A, Garcia-Cardena G, Madri JA, Sessa WC. Nitric oxide production contributes to the angiogenic properties of vascular endothelial growth factor in human endothelial cells. J Clin Invest 1997; 100:3131-9.

11. Ziche M, Morbidelli L, Choudhuri R, Zhang HT, Donnini S, Granger HJ, Bicknell R. Nitric oxide synthase lies downstream from vascular endothelial growth factor-induced but not basic fibroblast growth factor-induced angiogenesis. J Clin Invest 1997; 99:2625-34.

12. Feng Y, Venema VJ, Venema RC, Tsai N, Behzadian MA, Caldwell RB. VEGF-induced permeability increase is mediated by caveolae. Invest Ophthalmol Vis Sci 1999; 40:157-67.

13. Fischer S, Clauss M, Wiesnet M, Renz D, Schaper W, Karliczek GF. Hypoxia induces permeability in brain microvessel endothelial cells via VEGF and NO. Am J Physiol 1999; 276:C812-20.

14. Morales-Ruiz M, Fulton D, Sowa G, Languino LR, Fujio Y, Walsh K, Sessa WC. Vascular endothelial growth factor-stimulated actin reorganization and migration of endothelial cells is regulated via the serine/threonine kinase Akt. Circ Res 2000; 86:892-6.

15. Alon T, Hemo I, Itin A, Pe'er J, Stone J, Keshet E. Vascular endothelial growth factor acts as a survival factor for newly formed retinal vessels and has implications for retinopathy of prematurity. Nat Med 1995; 1:1024-8.

16. Stone J, Itin A, Alon T, Pe'er J, Gnessin H, Chan-Ling T, Keshet E. Development of retinal vasculature is mediated by hypoxia-induced vascular endothelial growth factor (VEGF) expression by neuroglia. J Neurosci 1995; 15:4738-47.

17. Pierce EA, Foley ED, Smith LEH. Vascular endothelial growth factor: a critical factor in the vaso-obliterative phase of retinopathy of prematurity. Invest Ophthalmol Vis Sci 1996; 37:S128.

18. Brooks SE, Gu X, Samuel S, Marcus DM, Bartoli M, Huang PL, Caldwell RB. Reduced severity of oxygen-induced retinopathy in eNOS-deficient mice. Invest Ophthalmol Vis Sci 2001; 42:222-8.

19. Ando A, Yang A, Mori K, Yamada H, Yamada E, Takahashi K, Saikia J, Kim M, Melia M, Fishman M, Huang P, Campochiaro PA. Nitric oxide is proangiogenic in the retina and choroid. J Cell Physiol 2002; 191:116-24.

20. Kim KY, Ju WK, Oh SJ, Chun MH. The immunocytochemical localization of neuronal nitric oxide synthase in the developing rat retina. Exp Brain Res 2000; 133:419-24.

21. Tay D, Diao YC, Xiao YM, So KF. Postnatal development of nicotinamide adenine dinucleotide phosphate-diaphorase-positive neurons in the retina of the golden hamster. J Comp Neurol 2002; 446:342-8.

22. Roufail E, Stringer M, Rees S. Nitric oxide synthase immunoreactivity and NADPH diaphorase staining are co-localised in neurons closely associated with the vasculature in rat and human retina. Brain Res 1995; 684:36-46.

23. Huang PL, Huang Z, Mashimo H, Bloch KD, Moskowitz MA, Bevan JA, Fishman MC. Hypertension in mice lacking the gene for endothelial nitric oxide synthase. Nature 1995; 377:239-42.

24. Jiang B, Liou GI, Behzadian MA, Caldwell RB. Astrocytes modulate retinal vasculogenesis: effects on fibronectin expression. J Cell Sci 1994; 107:2499-508.

25. Krege JH, Hodgin JB, Hagaman JR, Smithies O. A noninvasive computerized tail-cuff system for measuring blood pressure in mice. Hypertension 1995; 25:1111-5.

26. Jiang B, Behzadian MA, Caldwell RB. Astrocytes modulate retinal vasculogenesis: effects on endothelial cell differentiation. Glia 1995; 15:1-10.

27. D'Amato R, Wesolowski E, Smith LE. Microscopic visualization of the retina by angiography with high-molecular-weight fluorescein-labeled dextrans in the mouse. Microvasc Res 1993; 46:135-42.

28. Tontsch U, Bauer HC. Isolation, characterization, and long-term cultivation of porcine and murine cerebral capillary endothelial cells. Microvasc Res 1989; 37:148-61.

29. Scherer-Singler U, Vincent SR, Kimura H, McGeer EG. Demonstration of a unique population of neurons with NADPH-diaphorase histochemistry. J Neurosci Methods 1983; 9:229-34.

30. Meng W, Ayata C, Waeber C, Huang PL, Moskowitz MA. Neuronal NOS-cGMP-dependent ACh-induced relaxation in pial arterioles of endothelial NOS knockout mice. Am J Physiol 1998; 274:H411-5.

31. Norris PJ, Hardwick JP, Emson PC. Localization of NADPH cytochrome P450 oxidoreductase in rat brain by immunohistochemistry and in situ hybridization and a comparison with the distribution of neuronal NADPH-diaphorase staining. Neuroscience 1994; 61:331-50.

32. Beesley JE. Histochemical methods for detecting nitric oxide synthase. Histochem J 1995; 27:757-69.

33. Blottner D, Grozdanovic Z, Gossrau R. Histochemistry of nitric oxide synthase in the nervous system. Histochem J 1995; 27:785-811.

34. Carmeliet P, Ferreira V, Breier G, Pollefeyt S, Kieckens L, Gertsenstein M, Fahrig M, Vandenhoeck A, Harpal K, Eberhardt C, Declercq C, Pawling J, Moons L, Collen D, Risau W, Nagy A. Abnormal blood vessel development and lethality in embryos lacking a single VEGF allele. Nature 1996; 380:435-9.

35. Ferrara N. Role of vascular endothelial growth factor in regulation of physiological angiogenesis. Am J Physiol Cell Physiol 2001; 280:C1358-66.

36. He H, Venema VJ, Gu X, Venema RC, Marrero MB, Caldwell RB. Vascular endothelial growth factor signals endothelial cell production of nitric oxide and prostacyclin through flk-1/KDR activation of c-Src. J Biol Chem 1999; 274:25130-5.

37. Robinson GS, Ju M, Shih SC, Xu X, McMahon G, Caldwell RB, Smith LE. Nonvascular role for VEGF: VEGFR-1, 2 activity is critical for neural retinal development. FASEB J 2001; 15:1215-7.

38. Fruttiger M. Development of the mouse retinal vasculature: angiogenesis versus vasculogenesis. Invest Ophthalmol Vis Sci 2002; 43:522-7.

39. Gariano RF. Cellular mechanisms in retinal vascular development. Prog Retin Eye Res 2003; 22:295-306.

40. D'Amore PA. Mechanisms of retinal and choroidal neovascularization. Invest Ophthalmol Vis Sci 1994; 35:3974-9.

41. Campochiaro PA. Retinal and choroidal neovascularization. J Cell Physiol 2000; 184:301-10.

42. Wolfensberger TJ, Tufail A. Systemic disorders associated with detachment of the neurosensory retina and retinal pigment epithelium. Curr Opin Ophthalmol 2000; 11:455-61.