1

A HYPOXIA RESPONSIVE GLIAL CELL-SPECIFIC GENE THERAPY VECTOR FOR 1

TARGETING RETINAL NEOVASCULARIZATION 2

Manas R Biswal,1 Howard M Prentice,2,3 C Kathleen Dorey,4 and Janet C Blanks2 3

1Integrative Biology Ph.D. Program, 2Center for Complex Systems and Brain Sciences, 4

Charles E. Schmidt College of Science, 3Charles E. Schmidt College of Medicine, 5

Florida Atlantic University, 4 Virginia Tech Carilion School of Medicine. 6

Supported by NIH Grant EYO16119 (JCB), American Heart Association grant 7

0815022E (MRB), Seed grant from Neuroscience Research Priority Grant funded by 8

Florida Atlantic University (JCB and HMP) 9

Word Count (Abstract: 248 words, Manuscript: words 3267 without figure legends) 10

Corresponding author: Janet C. Blanks, Center for Complex Systems and Brain 11

Sciences, Florida Atlantic University, 777 Glades Road, Boca Raton, FL 33431. Email: 12

blanks@fau.edu 13

Preliminary data was presented in abstract form at Society for Neuroscience, San 14

Diego, CA 2011. 15

Short title: Hypoxia-regulated gene therapy in retina 16

Key Words: Hypoxia, GFAP, Müller cell, HIF-1 responsive element, OIR model, gene 17

therapy 18

19

IOVS Papers in Press. Published on November 6, 2014 as Manuscript iovs.14-13932

Copyright 2014 by The Association for Research in Vision and Ophthalmology, Inc.

2

ABSTRACT: 20

Purpose: Müller cells, the major glial cell in the retina, play a significant role in retinal 21

neovascularization in response to tissue hypoxia. We previously designed and tested a 22

vector using a hypoxia-responsive domain and a glial fibrillary acidic protein (GFAP) 23

promoter to drive green fluorescent protein (GFP) expression in Müller cells in the 24

murine model of oxygen-induced retinopathy (OIR). This study compares the efficacy of 25

regulated and unregulated Müller cell delivery of endostatin in preventing 26

neovascularization in the OIR model. 27

Methods: Endostatin cDNA was cloned into plasmids with hypoxia regulated GFAP or 28

unregulated GFAP promoters, and packaged into self complementary Adeno 29

Associated virus serotype 2 vectors (scAAV2). Before placement in hyperoxia on 30

Postnatal day (P) 7, mice were given intravitreal injections of regulated or unregulated 31

scAAV2, capsid, or PBS. On P17, five days after return to room air, neovascular and 32

avascular areas, as well as expression of the transgene and vascular endothelial growth 33

factor (VEGF), were compared in OIR animals treated with a vector, capsid, or PBS. 34

Results: The hypoxia-regulated, glial-specific vector expressing endostatin reduced 35

neovascularization by 93% and reduced the central vaso-obliteration area by 90%, 36

matching the results with the unregulated GFAP-Endo vector. Retinas treated with the 37

regulated endostatin vector expressed substantial amounts of endostatin protein, and 38

significantly reduced VEGF protein. Endostatin production from the regulated vector 39

was undetectable in retinas with undamaged vasculature. 40

3

Conclusion: These findings suggest that the hypoxia-regulated, glial cell-specific vector 41

expressing endostatin may be useful for treatment of neovascularization in proliferative 42

diabetic retinopathy. 43

44

45

46

47

48

49

50

51

52

53

54

55

56

57

4

INTRODUCTION 58

Neovascularization in age-related macular degeneration and diabetic retinopathy 59

remains a major cause of vision loss and legal blindness in the world today. 60

Reduction, occlusion or loss of retinal or choroidal vessels are recognized as risk 61

factors for retinal, vitreal or choroidal neovascularization in diabetic retinopathy, 62

retinopathy of prematurity, and age-related macular degeneration. The consequent 63

retinal hypoxia increases levels of HIF-1, the master regulator of angiogenesis, which 64

regulates expression of many target genes, including at least thirty-one proangiogenic 65

factors and nine anti-angiogenic factors.1 66

The major growth factor associated with both physiological and pathological 67

angiogenesis is Vascular Endothelial Growth Factor (VEGF), expressed by several 68

ocular cell types, including Müller cells, RPE cells, pericytes, vascular endothelial cells, 69

and ganglion cells.2–5 VEGF mediates multiple events in an angiogenic program 70

characterized by increased vascular permeability (causing macular edema), 71

recruitment, proliferation, migration, adhesion, and organization of endothelial cells to 72

form tubular new vessels. The critical role of VEGF in many forms of ocular 73

neovascularization made it an attractive target for development of VEGF-targeting 74

therapies.6–8 The introduction of successful anti-VEGF treatments has dramatically 75

reduced vision loss in diabetic macular edema9,10 and neovascular forms of age-related 76

macular degeneration.11,12 However, better therapies are needed: the requirements for 77

multiple injections and frequent office visits place a burden on patients, 78

neovascularization may reappear when the treatment is stopped, and the injections are 79

associated with a low risk for elevated intraocular pressure, uveitis, vascular occlusion, 80

5

vitreous hemorrhage or retinal detachment.13–15 Moreover, there is wide variation in 81

patient responses to treatment; robust gain in vision is observed in about 30% of 82

patients, and 10% of patients do not respond to anti-VEGF treatments.8,16–18 A recent 83

large study (N=835 patients) found no link between SNPs in VEGF receptors, and 84

patient responses19 suggesting that continuous suppression of the VEGF signaling may 85

be more efficacious than monthly injections, and/or that other pathways are contributing 86

to the angiogenic responses to ocular hypoxia. 87

Gene therapy offers the advantage of local and sustained delivery of anti-88

angiogenic molecules, and has proven to efficiently suppress neovascularization in 89

animal models.20–24 Restricting expression of anti-VEGF molecules to periods when the 90

disease process is active should reduce the potential for complications associated with 91

prolonged reduction in VEGF, an important survival factor for photoreceptors, 92

endothelial cells, ganglion cells25, RPE and the ciliary process. VEGF inhibition has 93

been associated with thinning of the outer nuclear layer (ONL) in patients26; vessel 94

occlusion in animals and patients27,28 and in animals, with decreased ERGs29, reduced 95

thickness of the inner and outer nuclear layers30 and damage to RPE31 . (Despite these 96

observations, the overall safety record of intravitreal anti-VEGF drugs has been good; 97

the changes reported may be too subtle for routine clinical observation.) 98

Since hypoxia is a common feature of pathological NV leading to vision loss, 99

physiologically regulated gene delivery by low oxygen levels offers great potential for 100

more controlled gene therapy for ischemic ocular diseases. In order to target genes to 101

hypoxic cells, several hypoxia-sensitive gene switches have been developed, all of 102

which are based on hypoxia responsive elements (HRE).32–38 103

6

The HREs are targeted by hypoxia inducible factor—a heterodimer of HIF-1α and HIF-104

1β. In normoxia, HIF-1α is hydroxylated and rapidly degraded by the proteasome. In 105

hypoxic conditions, HIF1α is stabilized and accumulates in the cytoplasm to form HIF1 106

dimers that bind the HREs in target genes. Incorporating HREs in our gene therapy 107

make it HIF-1 regulated, so the therapy will be activated only in retinal regions 108

experiencing hypoxia, or other pathological conditions activating HIF-1 regulated 109

angiogenesis. 110

We chose to target Müller cells, the major retinal glial cell, since it traverses the 111

retina, and would therefore experience hypoxic stress following capillary loss in 112

diabetes39,40 and expresses VEGF in oxygen-induced retinopathy.3 It is known that 113

Müller cell derived VEGF is a significant contributor to NV41 in the retina.42 Gene 114

therapy targeting Müller cells is an important addition to existing gene therapies that 115

target RPE cells; together they provide a means to selectively deliver antiangiogenic 116

therapy to either inner retina or outer retina/choroid. 117

To suppress ocular neovascularization, we selected endostatin for its profound 118

effects on angiogenesis affecting more than VEGF-dependent pathways. A cleavage 119

product of Collagen XVIII it inhibits endothelial cell proliferation, migration and survival 120

43, and it increases expression of anti-angiogenic factors (e.g. thrombospondin), and 121

inhibits levels of pro-angiogenic factors (e.g. HIF1a, ephrins, Ids)44. Endostatin blocks 122

VEGF-induced microvascular permeability by rapidly stabilizing occludin and tight 123

junctions in vessels.45 Following internalization into endothelial cells, either through 124

endocytosis or through clathrin-coated pits46, endostatin localizes to the nucleus47 and 125

7

reduces expression of VEGF and increases the expression of PEDF.48 The C-terminal 126

peptide of endostatin competes with VEGF for binding to VEGFR3.49 127

We previously reported that a promoter containing a hypoxia-responsive domain 128

together with a GFAP-cell specific promoter maintained cell-specificity and was hypoxia-129

inducible both in vitro and in vivo.38 The current study tested the hypothesis that our 130

hypoxia-regulated, glial cell specific AAV vector expressing endostatin would be as 131

effective in reducing neovascularization in the OIR model as gene therapy with 132

constitutively expressed endostatin.20,50 133

METHODS: 134

Construction of Plasmids and generation of scAAV2. 135

Previously tested promoter cassettes38 having the GFAP promoter or the REG- 136

GFAP (hypoxia regulated GFAP) promoter were cloned into the scAAV plasmid. The 137

human GFAP promoter (GfaABC1D domain of GFAP promoter, 681bp size), a gift from 138

Dr. Michael Brenner, University of Alabama, is a truncated promoter with subfragments 139

which in combination were previously shown to drive astrocyte specific gene expression 140

in transgenic mice.51 To make this promoter hypoxia responsive, we incorporated two domains 141

immediately upstream of the 5’ end of GFAP promoter: firstly, positioned most 5 prime, we 142

inserted a silencing region (122bp) (HRSE; provided by Keith Webster, University of Miami 143

Miller School of Medicine) and secondly, between the HRSE and the GFAP promoter we 144

inserted six copies of the previously described hypoxia response element (6XHRE; total size 145

108bp) of the phosphoglycerate kinase gene35

.To make endostatin encoding vectors, the 146

endostatin open reading frame of 612bp size (NCBI id NM_030582.3) without the stop 147

8

codon was amplified from a commercially available plasmid (Catalog # puno1-hendo18, 148

Invivogen, San Diego, CA) by PCR using following primers (forward primer; 5’-149

GACaaccggtATGTACAGGATGCAACTC-3’ and reverse primer; 5- 150

GTATacgcgtCTACTTGGAGGCAGTCATG-3’ ) and cloned into self-complementary 151

Adeno-Associated Virus (scAAV) plasmid. To detect exogeneous endostatin, a FLAG 152

epitope sequence with appropriate restriction sites was synthesized (IDT DNA 153

Technology, Coralville, IA) and cloned into the 3’ end of endostatin (Fig. 1). The new 154

constructs were named as GFAP-Endo (scAAV-GFAP-Endostatin-FLAG) and REG-155

Endo (scAAV-HRSE-6XHRE-GFAP-Endostatin-FLAG). ScAAV serotype-2 viruses 156

were produced at the Gene Therapy Vector Core, University of North Carolina (Chapel 157

Hill, NC) and titers were determined by standard dot blot analysis. 158

Intravitreal Injection and Oxygen Induced retinopathy (OIR) model 159

All procedures involving mice were performed in accordance with the Association 160

for Research in Vision and Ophthalmology (ARVO) statement for the use of animals in 161

ophthalmic research and approved by the Institutional Animal Care and Use Committee 162

at Florida Atlantic University. The murine OIR model was used to compare the efficacy 163

of the unregulated GFAP-Endo and the hypoxia-regulated REG-Endo. Briefly, forty 164

pups age P7 (C57 Bl/6J mouse strain) were anesthetized with ketamine/xylazine and 165

given intravitreal injections of 1µl (2x109 viral particles) of either GFAP-Endo or REG-166

Endo (Fig. 1) in one eye and phosphate buffered saline or the empty AAV2 capsid 167

vector in the contralateral eye. P7 pups with their mothers were then either maintained 168

in normal room air, (room air controls 21% oxygen) or exposed to 75% O2 for 5 days in 169

a chamber, (OIR animals). At P12 pups were moved to room air conditions for 5 days. 170

9

Exposure of neonatal mice to high oxygen (P7-P12) causes central vaso-obliteration by 171

P12; the reduced retinal vessels are insufficient in room air and the consequent 172

physiologic hypoxia induces peripheral neovascularization over the period P12-P17. All 173

the neonatal animals used for this study developed neovascularization. Following 174

ketamine/xylazine euthanasia at P17, eyes were rapidly collected for preparation of 175

flatmounts or for analysis of endostatin and protein expression. 176

Flat Mounts 177

Following removal of the cornea and lens, the eye cups were fixed with 4% 178

paraformaldehyde for an hour and washed in PBS. The retina was isolated from the eye 179

cup and stained with Isolectin B4-594 (Alexa Fluor 594 – I21413, Life Technologies, 180

Grand Island , NY, USA) at room temperature overnight. The retina was divided into 181

quadrants, which were flattened on a slide with ganglion cells down, cover-slipped with 182

Floromount G (Qiagen, Valencia, CA, USA ) and imaged using a Nikon inverted 183

fluorescence microscope. Flat mount techniques were successfully practiced in room air 184

control animals. 185

Analysis of vaso-obliteration and neovascularization 186

The percentage of vaso-obliteration was calculated by comparing the central 187

avascular area to total retinal area using Photoshop software.52,53 To quantify 188

peripheral neovascularization, the outer 2/3rd of each retinal quadrant was selected. 189

Using lasso tool the peripheral vascular area was selected and neovascular tufts were 190

quantified using magic tool as described by Connor et al (2009).52 The ratio of 191

peripheral neovascular area to the total vascular area was measured and expressed as 192

10

percentage. For each retina, neovascular areas in each quadrants were averaged to 193

represent that retina. 194

Western Blotting & ELISA 195

The retina was collected in radioimmunoprecipitation assay (RIPA) buffer with 196

protease inhibitors, homogenized on ice, and centrifuged. For western blots, 197

supernatant proteins (20 microg/well) were electrophoresed on a 12% Tris Glycine gel, 198

transferred to nitrocellulose, blocked for an hour with 5% dried milk in Tris-buffered 199

saline and incubated overnight with rabbit anti-FLAG primary antibody (Sigma, St. 200

Louis, MO, USA). Β-actin antibody was used as loading control for each sample. 201

Endostatin bands were detected by enhanced chemiluminescence using horse-radish 202

peroxidase conjugated secondary antibodies, and following the manufacturer’s 203

instructions (Pierce, Thermo Fisher Scientific, Rockford, IL, USA). 204

Endostatin and VEGF concentrations in retinal lysates were measured using the 205

RayBio® human endostatin and the RayBio® Mouse VEGF ELISA Kits (Ray Biotech 206

Inc.,Norcross, GA, USA) according to the recommended protocol. The concentration of 207

endostatin expressed by either GFAP-Endo or REG-Endo vector was normalized to that 208

in the PBS treated eyes. VEGF was normalized to the amount of protein in each 209

sample. Each experiment was performed in duplicate (n = 2), and each data point was 210

performed in triplicate. 211

11

Statistical Analysis 212

Data are expressed as mean +/- SEM. Significance was determined using 213

analysis of variance with Newman-Keuls Multiple comparison test). Differences 214

between conditions were regarded as significant if p<0.05. 215

RESULTS: 216

Reduction in peripheral neovascular and central avascular areas: 217

Robust neovascularization was observed in both capsid and PBS treated eyes 218

(Fig. 2) from oxygen exposed mice. In comparison, the angiostatic effects of both 219

endostatin expressing vectors were clearly visible in flat mounted retinas (Compare A 220

and B in Figs. 2 and 3, respectively). Eyes treated with either GFAP-Endo or REG-Endo 221

had neovascular areas that were 90% lower (P< 0.001 for both) than those in PBS or 222

capsid injected eyes (Fig. 4). Neovascular areas in PBS and Capsid-injected eyes were 223

not significantly different. 224

Comparison of the vector-treated eyes (Figure 2 and 3; A and C) with their 225

contralateral control eyes treated with capsid or PBS (Figure 2 and 3; B and D) reveals 226

the dramatic reduction in central avascular area. The mean avascular areas in the 227

PBS and capsid treated eyes were almost 10-fold larger than in eyes treated with 228

GFAP-Endo or REG-Endo (P< 0.001; Fig. 5). 229

12

Vector-Driven Endostatin Expression 230

Western blots clearly illustrate that exogenous endostatin levels achieved in the OIR 231

retinas were greater in the GFAP-Endo treated eyes than in the REG-Endo eyes. As 232

expected, the flagged endostatin was barely detectable in REG-Endo treated eyes of 233

animals maintained in room air (Fig. 6). ELISA of the flagged human endostatin 234

revealed 13-fold greater expression of endostatin by REG-Endo in OIR retinas than in 235

room air retinas ( 2.7 ± 0.8 ng/mg versus 0.2 ± 0.3ng/mg, respectively; P<0.01) (Fig. 7). 236

In room air mice, endostatin levels were 10-fold higher in GFAP-Endo treated retinas 237

than in REG-Endo treated retinas (2.3 ± 0.4 ng/mg versus 0.2 ± 0.3ng/mg, respectively; 238

P< 0.01 (Fig. 7). The 13-fold difference in endostatin expression by REG-Endo in room 239

air and in OIR retinas is comparable to the 12-fold difference observed in our in vitro 240

experiments (Fig. 4 in our previous manuscript38). Endostatin expression in GFAP-241

Endo injected eyes was not significantly different in OIR animals by comparison to room 242

air animals. 243

Reduction of VEGF levels in OIR retina 244

VEGF protein levels in P17 control retinas (PBS injected eyes) exposed to high 245

oxygen were significantly higher than those in room air animals (365 ± 20 pg/mg versus 246

85 ± 23 pg/mg; P<0.001) (Fig. 8). Intravitreal injection of either GFAP-Endo (144± 28 247

pg/mg) or REG- Endo (153 ± 38 pg/mg) significantly reduced VEGF levels (P< 0.001; 248

Fig. 8) compared to PBS injected eyes. No significant difference was observed between 249

VEGF levels in eyes treated with GFAP-Endo or REG- Endo vectors. 250

13

DISCUSSION 251

Therapeutic Efficacy 252

These data represent the first demonstration of the efficacy of hypoxia-regulated, 253

cell-specific anti-angiogenic gene therapy for the retina. In the OIR model, our new 254

REG-Endo vector reduced neovascular area by 93% and the central avascular area by 255

90%. These results are comparable to those achieved with the constitutively expressed 256

GFAP-Endo, and with reports that the number of pre-retinal endothelial cell nuclei was 257

reduced by 86% by AAV-mediated endostatin50 and by 75% by an adenoviral 258

endostatin vector.54 259

The Müller cell specificity of our promoter has important implications because this 260

major glial cell traverses the width of the retina and may experience a distinct pattern of 261

hypoxia. While our glial cell specific and hypoxia regulated vectors will express 262

endostatin selectively in Müller cells, questions remain regarding the effects of the 263

vectors in respect to the location of hypoxia within the retina that may trigger gene 264

expression. For example a specific defined band of hypoxia in a region of the retina may 265

result in transgene expression from Müller cells that traverse this region but one 266

constraint may be the magnitude of cell surface for individual Müller cells that 267

experience oxygen deprivation. For a particular Müller cell, once the transgene is 268

induced it is likely that the endostatin protein product would be secreted from all 269

locations on the Müller cell membrane and therefore at a wide range of depths within 270

the retina. 271

14

We have demonstrated regulated expression from our REG-Endo vector both in 272

terms of increased protein product (as shown in ELISA at P17 by comparison to 273

expression in room air) and in terms of functional outcome (inhibition of 274

neovascularization). From the literature and our own data, it is clear that extensive 275

inhibition of neovascularization is obtained with AAV-CMV-Endostatin50as well as from 276

our own GFAP-Endo and REG-Endo vectors. Importantly the mechanisms of 277

endostatin action are poorly understood and include inhibition of VEGF induced 278

permeability, actions in the nucleus, decreased VEGF signaling and/or production and 279

increased PEDF expression. Hence the production of endostatin ubiquitously under the 280

CMV promoter may result in differences in endostatin action which occur in a very 281

different and progressively changing multi-cellular microenvironment from that obtained 282

with a Müller restricted vector or from REG-Endo. A novel feature of our vectors is 283

secretion of endostatin specifically from Müller cells or from hypoxic Müller cells eliciting 284

successful inhibition of neovascularization. Such inhibition of neovascularization by 285

endostatin has not been elicited previously from either Müller cell-specific or combined 286

Müller cell- specific/hypoxia regulated vectors. While the exact details of endostatin 287

action are only partially understood, it is likely that eliciting the more selective patterns 288

of endostatin distribution and release will result in distinct patterns of vascular inhibition 289

compared to the more diffuse ubiquitous patterns of multicellular secretion obtained with 290

AAV-CMV-endostatin. Furthermore the REG-Endo vector will result in activated therapy 291

that is disease specific and corresponds to emergence of a hypoxic microenvironment 292

instead of continuous and unvarying rates of production initiating soon after vector 293

delivery. (Note: The activation of our promoter is likely to differ in astrocytes that reside 294

15

in the inner layers of the retina and therefore may experience different levels of hypoxia 295

compared to cells. As a result, the responses in these two cell types could be different 296

both in timing and in levels of activation.) 297

It has been suggested Müller that sustained delivery of a transgene may lead to 298

tissue toxicity over time23 ; as discussed earlier, reduced VEGF has been linked to 299

retinal damage in animals. If the transgene is regulated in response to the pathologic 300

state of surrounding tissue, it can be effective during the period of the disease and 301

turned off once the disease progression is stabilized. If the patient has a recurrence of 302

the pathology, the expression will turn on again. Although our vector successfully 303

inhibited neovascularization within a short period of time, the effectiveness in ensuring 304

more long-term transgene expression when necessary requires further investigation. 305

Two major advantages of this strategy are that the anti-angiogenic therapy is 306

delivered as vessels start to form, when they are most amenable to treatment, and that 307

there is a high therapeutic index because therapy is delivered only when and where 308

needed; the remaining healthy retina would be exposed to a much lower dose. and only 309

for the duration needed, as expression is suppressed in normoxic tissue. 310

Retinal Endostatin Levels 311

Our regulated vector produced approximately 2-3 ng of endostatin/mg of retinal 312

homogenate (Fig. 7), which corresponds closely to the 1-2 ng of endostatin reported 313

previously with the constitutive VMD2-RPE specific promoter.21 While these levels are 314

less than 10-14 ng of endostatin delivered by a CMV promoter, it is likely that high 315

16

absolute levels of anti-angiogenic protein are not necessary to achieve a substantial 316

therapeutic effect. The key difference between REG-Endo and CMV-endostatin is that 317

for REG-Endo, levels of endostatin produced will be more than 10-fold lower in 318

normoxic retina than in OIR animals. 319

GFAP activity and endostatin delivery. 320

Endostatin levels were higher in GFAP-Endo treated eyes from OIR mice than in 321

those from similarly treated room air mice (Fig.7). Müller cells become reactive and 322

induce GFAP in response to most types of retinal injury. Reactive gliosis is a hallmark 323

of the retinal response in the OIR model.55 As a result, the basal level of the unregulated 324

GFAP-Endo will increase as GFAP expression increases, and will remain active during 325

retinal inflammation and damage. In contrast, REG-Endo will be constitutively silent, 326

activated in hypoxia and shut-off by declining HIF-1 levels, at least by P26, when the 327

OIR retina has “resorbed” the newly formed vessels.2 Once activated by hypoxia, the 328

increase in GFAP activity that is associated with the reactive phenotype in Müller cells 329

will further boost the production of endostatin. 330

Change in VEGF levels 331

Like others, we found that OIR retinas had significantly higher VEGF than room 332

air retinas. The amount of VEGF produced in the OIR retinas was comparable with that 333

in other studies.56 VEGF levels at P17 were approximately 60% lower in the endostatin 334

eyes. The significant reduction in central avascular area would certainly contribute to a 335

decline in VEGF. It is interesting that the reduction in VEGF in these experiments is 336

17

comparable to that observed in eyes injected with 500 times higher (1 µg) dose of 337

endostatin on P12 in the OIR model.57 This might indicate that the REG-Endo produces 338

endostatin at close to saturating levels. 339

Hypoxia-regulated gene therapy 340

Hypoxia-regulated gene therapy was introduced by Prentice et al (1997) as a 341

means to deliver cardioprotective genes to myocardium experiencing transient or 342

intermittent ischemia.36 The field has grown slowly since that time with reports on 343

hypoxia-regulated gene therapy targeting hypoxic regions of tumors, followed by reports 344

on cardiac ischemia and/or ischemia reperfusion injury to heart or neural tissue. 345

Bainbridge et al first reported hypoxia-regulated expression of a reporter gene in the 346

retina; it was neither cell-specific nor suppressed in normoxic retina32. Our group 347

previously reported a retinal pigment epithelium (RPE)-specific AAV vector that was 348

robustly responsive to hypoxia and silenced in normoxia35 as a potential platform for 349

treatment of choroidal neovascularization. To develop a platform targeting intraretinal 350

neovascularization, a new hypoxia-responsive, glial-specific AAV vector was 351

constructed to target Müller cells, and its specificity and responsiveness were verified in 352

vitro, and in the OIR model of retinal neovascularization.38 353

The highly sensitive response of the REG-Endo vector to hypoxia results from its 354

design. We utilized hypoxia response elements which are consensus DNA binding sites 355

for the HIF-1 transcription factor, which is kept at low levels in normoxia by the oxygen-356

dependent degradation of the HIF-1alpha subunit. Hypoxia regulated vectors have 357

been of value in different model systems for expressing a range of factors including pro-358

18

survival components, angiogenic factors and growth factors.58–61 By incorporating 359

multimers of the HRE, we increased the response of the promoter to hypoxic conditions; 360

the neuronal silencing element further ensures very low basal expression in normoxia 361

prior to hypoxic induction. REG-ENDO was made cell specific by virtue of the 681 bp 362

hybrid regulatory domain from the GFAP gene (See Methods section). 363

HRE-regulated switch in normoxia 364

A key question regarding timing of transgene expression will be whether 365

endostatin expression ceases once the pathology has subsided. Cell specific and 366

hypoxia-regulated vectors have been shown to switch off in other systems including 367

heart 36,62and cancer.63,64 Given the tight correspondence between tissue hypoxia and 368

the induction, maintenance and termination of HIF signaling, it is anticipated that the 369

activity of REG-Endo will closely correlate with retinal hypoxia. 370

Implications for the future. 371

This new strategy for anti-angiogenic gene therapy opens an avenue of 372

prophylactic gene therapy for patients at high risk for neovascularization. Theoretically, 373

high risk patients could be treated with a gene therapy that would only be activated 374

when pathological conditions begin to stimulate neovascularization. Therapy would 375

initiate before the patient experienced any symptoms or loss of vision. 376

377

378

19

SUMMARY 379

The major findings of the present study were the use of a hypoxia-regulated, 380

retinal glial cell specific promoter driving the expression of the human endostatin gene 381

to dramatically reduce peripheral neovascularization and central in the OIR mouse 382

model of neovascularization. Our results demonstrated the effect of exogenous 383

endostatin on VEGF expression levels in retina, and confirmed that the regulated vector 384

produced sufficient endostatin to inhibit hypoxia-induced neovascularization in the 385

murine OIR model. These results suggest that the localized production of endostatin in 386

ischemic retina reduces pathological angiogenesis and promotes physiological 387

angiogenesis. 388

ACKNOWLEDGMENTS 389

The authors would like to thank Sanghamitra Das for providing technical assistance on 390

this project. The authors would like to acknowledge Dr. Michael Brenner (University of 391

Alabama) for contributing the plasmid containing the GFAP promoter. Dr. George 392

(Tyler) Smith offered helpful suggestions in the design of the experiments. 393

394

20

REFERENCES 395

1. Krock BL, Skuli N, Simon MC. Hypoxia-induced angiogenesis: good and evil. 396

Genes Cancer 2011;2(12):1117-1133. doi:10.1177/1947601911423654. 397

2. Pierce EA, Avery RL, Foley ED, Aiello LP, Smith LE. Vascular endothelial growth 398

factor/vascular permeability factor expression in a mouse model of retinal 399

neovascularization. Proc. Natl. Acad. Sci. 1995;92(3):905-909. 400

3. Dorey C, Aouididi S, Reynaud X, Dvorak HF, Brown LF. COrrelation of vascular 401

permeability factor/vascular endothelial growth factor with extraretinal 402

neovascularization in the rat. Arch. Ophthalmol. 1996;114(10):1210-1217. 403

doi:10.1001/archopht.1996.01100140410008. 404

4. Aiello LP. Vascular endothelial growth factor and the eye: biochemical mechanisms 405

of action and implications for novel therapies. Ophthalmic Res. 1997;29(5):354-362. 406

5. Mousa SA, Lorelli W, Campochiaro PA. Role of hypoxia and extracellular matrix-407

integrin binding in the modulation of angiogenic growth factors secretion by retinal 408

pigmented epithelial cells. J. Cell. Biochem. 1999;74(1):135-143. 409

6. Adamis AP, Shima DT, Tolentino MJ, et al. Inhibition of vascular endothelial growth 410

factor prevents retinal ischemia-associated iris neovascularization in a nonhuman 411

primate. Arch. Ophthalmol. 1996;114(1):66-71. 412

7. Eyetech Study Group. Anti-vascular endothelial growth factor therapy for subfoveal 413

choroidal neovascularization secondary to age-related macular degeneration: 414

phase II study results. Ophthalmology 2003;110(5):979-986. doi:10.1016/S0161-415

6420(03)00085-X. 416

21

8. Rosenfeld PJ, Brown DM, Heier JS, et al. Ranibizumab for neovascular age-related 417

macular degeneration. N. Engl. J. Med. 2006;355(14):1419-1431. 418

doi:10.1056/NEJMoa054481. 419

9. Stewart MW. Anti-VEGF Therapy for Diabetic Macular Edema. Curr. Diab. Rep. 420

2014;14(8):510. doi:10.1007/s11892-014-0510-4. 421

10. Stefanini FR, Badaró E, Falabella P, Koss M, Farah ME, Maia M. Anti-VEGF for the 422

management of diabetic macular edema. J. Immunol. Res. 2014;2014:632307. 423

doi:10.1155/2014/632307. 424

11. Finger RP, Wu Z, Luu CD, et al. Reticular pseudodrusen: a risk factor for 425

geographic atrophy in fellow eyes of individuals with unilateral choroidal 426

neovascularization. Ophthalmology 2014;121(6):1252-1256. 427

doi:10.1016/j.ophtha.2013.12.034. 428

12. Sloan FA, Hanrahan BW. The effects of technological advances on outcomes for 429

elderly persons with exudative age-related macular degeneration. JAMA 430

Ophthalmol. 2014;132(4):456-463. doi:10.1001/jamaophthalmol.2013.7647. 431

13. Nuzzi R, Tridico F. Local and Systemic Complications after Intravitreal 432

Administration of Anti-Vascular Endothelial Growth Factor Agents in the Treatment 433

of Different Ocular Diseases: A Five-Year Retrospective Study. Semin. Ophthalmol. 434

2013. doi:10.3109/08820538.2013.835833. 435

14. Abedi F, Wickremasinghe S, Richardson AJ, Islam AFM, Guymer RH, Baird PN. 436

Genetic influences on the outcome of anti-vascular endothelial growth factor 437

treatment in neovascular age-related macular degeneration. Ophthalmology 438

2013;120(8):1641-1648. doi:10.1016/j.ophtha.2013.01.014. 439

22

15. Ventrice P, Leporini C, Aloe JF, et al. Anti-vascular endothelial growth factor drugs 440

safety and efficacy in ophthalmic diseases. J. Pharmacol. Pharmacother. 441

2013;4(Suppl 1):S38-42. doi:10.4103/0976-500X.120947. 442

16. Broadhead GK, Hong T, Chang AA. Treating the untreatable patient: current 443

options for the management of treatment-resistant neovascular age-related macular 444

degeneration. Acta Ophthalmol. (Copenh.) 2014. doi:10.1111/aos.12463. 445

17. Brown DM, Kaiser PK, Michels M, et al. Ranibizumab versus verteporfin for 446

neovascular age-related macular degeneration. N. Engl. J. Med. 447

2006;355(14):1432-1444. doi:10.1056/NEJMoa062655. 448

18. Kovach JL, Schwartz SG, Flynn HW, Scott IU. Anti-VEGF Treatment Strategies for 449

Wet AMD. J. Ophthalmol. 2012;2012:786870. doi:10.1155/2012/786870. 450

19. Hagstrom SA, Ying G, Pauer GJT, et al. VEGFA and VEGFR2 gene polymorphisms 451

and response to anti-vascular endothelial growth factor therapy: comparison of age-452

related macular degeneration treatments trials (CATT). JAMA Ophthalmol. 453

2014;132(5):521-527. doi:10.1001/jamaophthalmol.2014.109. 454

20. Mori K, Ando A, Gehlbach P, et al. Inhibition of Choroidal Neovascularization by 455

Intravenous Injection of Adenoviral Vectors Expressing Secretable Endostatin. Am. 456

J. Pathol. 2001;159(1):313-320. doi:10.1016/S0002-9440(10)61697-5. 457

21. Kachi S, Binley K, Yokoi K, et al. Equine infectious anemia viral vector-mediated 458

codelivery of endostatin and angiostatin driven by retinal pigmented epithelium-459

specific VMD2 promoter inhibits choroidal neovascularization. Hum. Gene Ther. 460

2009;20(1):31-39. doi:10.1089/hum.2008.046. 461

23

22. Hou X, Kumar A, Lee C, et al. PDGF-CC blockade inhibits pathological 462

angiogenesis by acting on multiple cellular and molecular targets. Proc. Natl. Acad. 463

Sci. U. S. A. 2010;107(27):12216-12221. doi:10.1073/pnas.1004143107. 464

23. Askou AL, Pournaras J-AC, Pihlmann M, et al. Reduction of choroidal 465

neovascularization in mice by adeno-associated virus-delivered anti-vascular 466

endothelial growth factor short hairpin RNA. J. Gene Med. 2012;14(11):632-641. 467

doi:10.1002/jgm.2678. 468

24. Boye SE, Boye SL, Lewin AS, Hauswirth WW. A comprehensive review of retinal 469

gene therapy. Mol. Ther. J. Am. Soc. Gene Ther. 2013;21(3):509-519. 470

doi:10.1038/mt.2012.280. 471

25. Foxton RH, Finkelstein A, Vijay S, et al. VEGF-A is necessary and sufficient for 472

retinal neuroprotection in models of experimental glaucoma. Am. J. Pathol. 473

2013;182(4):1379-1390. doi:10.1016/j.ajpath.2012.12.032. 474

26. Ozdemir O, Erol MK. Morphologic changes and visual outcomes in resolved central 475

serous chorioretinopathy treated with ranibizumab. Cutan. Ocul. Toxicol. 476

2014;33(2):122-126. doi:10.3109/15569527.2013.812104. 477

27. Chuang L-H, Wang N-K, Chen Y-P, Wu W-C, Lai C-C. Mature vessel occlusion 478

after anti-VEGF treatment in a retinal arteriovenous malformation. BMC 479

Ophthalmol. 2013;13:60. doi:10.1186/1471-2415-13-60. 480

28. Ford KM, Saint-Geniez M, Walshe T, Zahr A, D’Amore PA. Expression and role of 481

VEGF in the adult retinal pigment epithelium. Invest. Ophthalmol. Vis. Sci. 482

2011;52(13):9478-9487. doi:10.1167/iovs.11-8353. 483

24

29. Tokunaga CC, Mitton KP, Dailey W, et al. Effects of anti-VEGF treatment on the 484

recovery of the developing retina following oxygen-induced retinopathy. Invest. 485

Ophthalmol. Vis. Sci. 2014;55(3):1884-1892. doi:10.1167/iovs.13-13397. 486

30. Saint-Geniez M, Maharaj ASR, Walshe TE, et al. Endogenous VEGF is required for 487

visual function: evidence for a survival role on müller cells and photoreceptors. PloS 488

One 2008;3(11):e3554. doi:10.1371/journal.pone.0003554. 489

31. Ford KM, D’Amore PA. Molecular regulation of vascular endothelial growth factor 490

expression in the retinal pigment epithelium. Mol. Vis. 2012;18:519-527. 491

32. Bainbridge JWB, Mistry A, Binley K, et al. Hypoxia-regulated transgene expression 492

in experimental retinal and choroidal neovascularization. Gene Ther. 493

2003;10(12):1049-1054. doi:10.1038/sj.gt.3301945. 494

33. Binley K, Iqball S, Kingsman A, Kingsman S, Naylor S. An adenoviral vector 495

regulated by hypoxia for the treatment of ischaemic disease and cancer. Gene 496

Ther. 1999;6(10):1721-1727. doi:10.1038/sj.gt.3301001. 497

34. Binley K, Askham Z, Martin L, et al. Hypoxia-mediated tumour targeting. Gene 498

Ther. 2003;10(7):540-549. doi:10.1038/sj.gt.3301944. 499

35. Dougherty CJ, Smith GW, Dorey CK, Prentice HM, Webster KA, Blanks JC. Robust 500

hypoxia-selective regulation of a retinal pigment epithelium-specific adeno-501

associated virus vector. Mol. Vis. 2008;14:471-480. 502

36. Prentice H, Bishopric NH, Hicks MN, et al. Regulated expression of a foreign gene 503

targeted to the ischaemic myocardium. Cardiovasc. Res. 1997;35(3):567-574. 504

25

37. Tang YL, Tang Y, Zhang YC, et al. A hypoxia-inducible vigilant vector system for 505

activating therapeutic genes in ischemia. Gene Ther. 2005;12(15):1163-1170. 506

doi:10.1038/sj.gt.3302513. 507

38. Prentice HM, Biswal MR, Dorey CK, Blanks JC. Hypoxia-Regulated Retinal Glial 508

Cell–Specific Promoter for Potential Gene Therapy in Disease. Invest. Ophthalmol. 509

Vis. Sci. 2011;52(12):8562-8570. doi:10.1167/iovs.10-6835. 510

39. Mowat FM, Luhmann UFO, Smith AJ, et al. HIF-1alpha and HIF-2alpha are 511

differentially activated in distinct cell populations in retinal ischaemia. PloS One 512

2010;5(6):e11103. doi:10.1371/journal.pone.0011103. 513

40. Lin M, Chen Y, Jin J, et al. Ischaemia-induced retinal neovascularisation and 514

diabetic retinopathy in mice with conditional knockout of hypoxia-inducible factor-1 515

in retinal Müller cells. Diabetologia 2011;54(6):1554-1566. doi:10.1007/s00125-011-516

2081-0. 517

41. Bai Y, Ma J, Guo J, et al. Müller cell-derived VEGF is a significant contributor to 518

retinal neovascularization. J. Pathol. 2009;219(4):446-454. doi:10.1002/path.2611. 519

42. Xin X, Rodrigues M, Umapathi M, et al. Hypoxic retinal Muller cells promote 520

vascular permeability by HIF-1-dependent up-regulation of angiopoietin-like 4. Proc. 521

Natl. Acad. Sci. U. S. A. 2013;110(36):E3425-3434. doi:10.1073/pnas.1217091110. 522

43. O’Reilly MS, Boehm T, Shing Y, et al. Endostatin: an endogenous inhibitor of 523

angiogenesis and tumor growth. Cell 1997;88(2):277-285. 524

44. Benezra R, Rafii S. Endostatin’s endpoints-Deciphering the endostatin 525

antiangiogenic pathway. Cancer Cell 2004;5(3):205-206. 526

26

45. Campbell M, Collery R, McEvoy A, Gardiner TA, Stitt AW, Brankin B. Involvement 527

of MAPKs in endostatin-mediated regulation of blood-retinal barrier function. Curr. 528

Eye Res. 2006;31(12):1033-1045. doi:10.1080/02713680601013025. 529

46. Chen Y, Wang S, Lu X, Zhang H, Fu Y, Luo Y. Cholesterol sequestration by 530

nystatin enhances the uptake and activity of endostatin in endothelium via 531

regulating distinct endocytic pathways. Blood 2011;117(23):6392-6403. 532

doi:10.1182/blood-2010-12-322867. 533

47. Song N, Ding Y, Zhuo W, et al. The nuclear translocation of endostatin is mediated 534

by its receptor nucleolin in endothelial cells. Angiogenesis 2012;15(4):697-711. 535

doi:10.1007/s10456-012-9284-y. 536

48. Tan H, Mu G, Zhu W, Liu J, Wang F. Down-regulation of vascular endothelial 537

growth factor and up-regulation of pigment epithelium derived factor make low 538

molecular weight heparin-endostatin and polyethylene glycol-endostatin potential 539

candidates for anti-angiogenesis drug. Biol. Pharm. Bull. 2011;34(4):545-550. 540

49. Han K-Y, Azar DT, Sabri A, et al. Characterization of the interaction between 541

endostatin short peptide and VEGF receptor 3. Protein Pept. Lett. 2012;19(9):969-542

974. 543

50. Auricchio A, Behling KC, Maguire AM, et al. Inhibition of retinal neovascularization 544

by intraocular viral-mediated delivery of anti-angiogenic agents. Mol. Ther. J. Am. 545

Soc. Gene Ther. 2002;6(4):490-494. 546

51. Lee Y, Messing A, Su M, Brenner M. GFAP promoter elements required for region-547

specific and astrocyte-specific expression. Glia 2008;56(5):481-493. 548

doi:10.1002/glia.20622. 549

27

52. Connor KM, Krah NM, Dennison RJ, et al. Quantification of oxygen-induced 550

retinopathy in the mouse: a model of vessel loss, vessel regrowth and pathological 551

angiogenesis. Nat. Protoc. 2009;4(11):1565-1573. doi:10.1038/nprot.2009.187. 552

53. Banin E, Dorrell MI, Aguilar E, et al. T2-TrpRS inhibits preretinal neovascularization 553

and enhances physiological vascular regrowth in OIR as assessed by a new 554

method of quantification. Invest. Ophthalmol. Vis. Sci. 2006;47(5):2125-2134. 555

doi:10.1167/iovs.05-1096. 556

54. Le Gat L, Gogat K, Bouquet C, et al. In vivo adenovirus-mediated delivery of a 557

uPA/uPAR antagonist reduces retinal neovascularization in a mouse model of 558

retinopathy. Gene Ther. 2003;10(25):2098-2103. doi:10.1038/sj.gt.3302122. 559

55. DeNiro M, Al-Mohanna FH, Al-Mohanna FA. Inhibition of reactive gliosis prevents 560

neovascular growth in the mouse model of oxygen-induced retinopathy. PloS One 561

2011;6(7):e22244. doi:10.1371/journal.pone.0022244. 562

56. Yu W, Bai Y, Han N, et al. Inhibition of pathological retinal neovascularization by 563

semaphorin 3A. Mol. Vis. 2013;19:1397-1405. 564

57. Zhang M, Yang Y, Yan M, Zhang J. Downregulation of vascular endothelial growth 565

factor and integrinbeta3 by endostatin in a mouse model of retinal 566

neovascularization. Exp. Eye Res. 2006;82(1):74-80. 567

doi:10.1016/j.exer.2005.05.005. 568

58. Pachori AS, Melo LG, Hart ML, et al. Hypoxia-regulated therapeutic gene as a 569

preemptive treatment strategy against ischemia/reperfusion tissue injury. Proc. Natl. 570

Acad. Sci. U. S. A. 2004;101(33):12282-12287. doi:10.1073/pnas.0404616101. 571

28

59. Ruan H, Su H, Hu L, Lamborn KR, Kan YW, Deen DF. A hypoxia-regulated adeno-572

associated virus vector for cancer-specific gene therapy. Neoplasia N. Y. N 573

2001;3(3):255-263. doi:10.1038/sj/neo/7900157. 574

60. Jiang B, Dong H, Zhang Z, Wang W, Zhang Y, Xu X. Hypoxic response elements 575

control expression of human vascular endothelial growth factor(165) genes 576

transferred to ischemia myocardium in vivo and in vitro. J. Gene Med. 577

2007;9(9):788-796. doi:10.1002/jgm.1070. 578

61. Hu H-W, Li X-K, Zheng R-Y, Xiao J, Zeng J-Q, Hou ST. bFGF expression mediated 579

by a hypoxia-regulated adenoviral vector protects PC12 cell death induced by 580

serum deprivation. Biochem. Biophys. Res. Commun. 2009;390(1):115-120. 581

doi:10.1016/j.bbrc.2009.09.077. 582

62. Webster K. Molecular switches for regulating therapeutic genes. Gene Ther. 583

1999;6(6):951-953. doi:10.1038/sj.gt.3300959. 584

63. Harvey TJ, Hennig IM, Shnyder SD, et al. Adenovirus-mediated hypoxia-targeted 585

gene therapy using HSV thymidine kinase and bacterial nitroreductase prodrug-586

activating genes in vitro and in vivo. Cancer Gene Ther. 2011;18(11):773-784. 587

doi:10.1038/cgt.2011.43. 588

64. Luo Y, Zhu D. Combinatorial control of transgene expression by hypoxia-responsive 589

promoter and microrna regulation for neural stem cell-based cancer therapy. 590

BioMed Res. Int. 2014;2014:751397. doi:10.1155/2014/751397. 591

29

Figure legends: 592

Figure 1: Schematic diagram of vectors used in this study. 593

The promoter elements and endostatin cDNA are incorporated into self-complementary 594

Adeno Associated Vectors (ScAAV2). (A) GFAP-Endo vector; this vector has human 595

glial fibrillary acidic protein (GFAP) promoter (681bp) and human endostatin (Endo) 596

cDNA open reading frame (612bp) fused with FLAG epitope . (B) REG-Endo vector; 597

this vector includes hypoxia regulated sequence (HRSE-6XHRE, 230bp) along with 598

GFAP promoter and endostatin cDNA fused with FLAG epitope. SV40 poly A (SV40pA) 599

sequence and inverted tandem repeat sequences (ITR) are present in scAAV2 vectors. 600

(dITR denoted the mutated ITR in 5’ end). Construction of above vectors is described in 601

methods section. 602

Figure 2: Representative micrographs of flat mounted P17 retinas treated with 603

GFAP-Endo. 604

Light micrographs of retinas from mice exposed to high oxygen were isolated, retinal 605

whole mounts made, and blood vessels identified with lectin staining. In response to 606

high oxygen followed by room air, the central retina loses much of the normal 607

vasculature whereas neovascularization occurs in the peripheral retina indicated by the 608

formation of neovascular tufts. Retina from GFAP-Endo treated eye (A) shows 609

reduction of both the central avascular area (solid white square) and peripheral 610

neovascular area (square with dashed white lines) compared to contralateral capsid 611

injected eye (B). Magnified view of the peripheral retina shows reduced neovascular 612

30

tufts in GFAP- Endo treated eye (arrows, C) compared to capsid injected eye (arrows, 613

D). Magnified view of central avascular area shows reduction of vaso-obliteration and 614

increased vascularity in GFAP- Endo treated eye (arrows, E) compared to capsid 615

injected eyes (arrows, F). Note, the “hot spot” in the retina (asterisk, A) is an artifact 616

due to accidental puncture during the injection. (Scale bar; A and B = 1000um; C, D, E, 617

and F = 100um) 618

Figure 3: Representative micrographs of flat mounted P17 retinas from eyes 619

treated with REG-Endo. 620

Light micrograph of retinal whole mount from REG-Endo treated eye (A) shows 621

reduction of both central avascular area (white square) and peripheral neovascular area 622

(square with dashed white lines) compared to contralateral PBS injected eye (B). 623

Magnified view of peripheral neovascular retina shows reduced neovascular area in 624

REG- Endo treated eye (arrows, c) compared to PBS injected eye (arrows, D). 625

Enlarged view of central avascular area shows amelioration of vaso-obliteration and 626

normal vascular patterning in REG- Endo treated eye (arrows, E) compared to 627

contralateral PBS injected eye (F). The bright strands in (e) and (f) are remnants of 628

hyloid vessels. (Scale bar; a and b = 1000um; c, d, e, and f = 100um) 629

Figure 4: Reduction of peripheral neovascularization. 630

A) Mean neovascular areas (Mean ± S. E) in eyes treated with GFAP or REG-631

endostatin vector, PBS, or Capsid . Eyes treated with either GFAP-Endo or REG-Endo 632

had neovascular areas (2.2± 0.5% and 1.6± 0.6%, respectively) that were 90% lower 633

31

than eyes injected with either PBS or capsid injected eyes (*P<0.001). Neovascular 634

areas in PBS (n=11) and Capsid-injected eyes (n=10) were not significantly different 635

(24.8 ± 3.2 %, and 22.8 ± 4.1%, respectively) (P < 0.056 ). B) Paired comparison of 636

neovascular area in eyes from mice injected with endostatin vector in one eye and 637

Capsid or PBS in the contralateral eye. Treatments and mouse number are indicated 638

under each pair of bars. 639

Figure 5: Reduction of central avascular area. 640

A) Central avascular area (Mean ± S. E) in retinas treated with GFAP or REG-641

Endostatin vector, PBS, or capsid. Significant reduction of the central avascular area 642

was observed in GFAP-Endo (2.34 ± 1.4 %, n =9) and REG-Endo (2.31 ± 1.6 %, n = 9) 643

injected eyes compared to either PBS (21.1 ± 1.9 %, n =11) or Capsid (21.87 ± 2.5 %, n 644

= 10) injected eyes (*P<0.001). There was no significant difference in the central 645

avascular area in eyes injected with either PBS or Capsid. (* P < 0.02 ).B) Paired 646

comparison of avascular areas in eyes injected with endostatin vector in one eye and 647

either capsid or PBS in the contralateral eye. (n = Number of individual mice). 648

Figure 6: FLAG tagged endostatin in mouse retina at P17: 649

Lane 1 shows that eyes treated with REG-Endo vector resulted in barely detectable 650

expression of endostatin in room air. Lane 2 and Lane 3 demonstrate expression of 651

endostatin (22 kDa) by REG-Endo vectors was greater in OIR retinas but still less than 652

the expression from the GFAP-Endo vector. 653

32

Figure 7: Expression of endostatin by regulated and unregulated vectors in room 654

air or OIR retinas. 655

ELISA analysis of retinal endostatin expression in mice that were injected with GFAP-656

Endo or REG-Endo, then placed in oxygen for the OIR model or raised in room air. In 657

the room air groups, retinal endostatin concentration in the REG-Endo mice was only 658

10% of that in the GFAP-Endo mice. Endostatin concentration in the retinas of the 659

REG-Endo mice in the OIR model was 12-fold greater than in REG-Endo mice in room 660

air. Retinal endostatin expression in the OIR model was statistically equivalent in the 661

REG-Endo and GFAP-Endo mice. Groups with different letters differ significantly 662

(P<0.01 for all comparisons; N=6 in all groups). 663

Figure 8: VEGF levels measured by ELISA at P17. 664

VEGF levels were higher in retinas from PBS treated OIR mice than in room air 665

controls. Treatment with either GFAP-Endo or REG-Endo reduced VEGF levels in the 666

OIR retinas. ( * P < 0.001) 667