Vascular endothelial growth factor-A165b ... - Clinical Science

Clinical Science: this is an Accepted Manuscript, not the final Version of Record. You are encouraged to use the Version of Record that, when published, will replace this version. The most up-to-date version is available at http://dx.doi.org/10.1042/CS20170102. Please cite using the DOI 10.1042/CS20170102

VEGF-A165b inhibits diabetic retinal barrier function 1

Vascular endothelial growth factor-A165b ameliorates outer-retinal barrier and

2

vascular dysfunction in the diabetic retina

3 4

Ved N1,2, Hulse RP1, Bestall SM1,3, Donaldson LF3, Bainbridge JW 2, Bates DO1

5 6

1

7

University of Nottingham, Nottingham NG7 2UH,2 Department of Genetics, UCL

8

Institute of Ophthalmology, 11-43 Bath Street, London EC1V9EL, and 3School of Life

9

Sciences, University of Nottingham, Nottingham NG7 2UH,

Cancer Biology, Division of Cancer and Stem Cells, School of Medicine,

10 11 Clinical Significance

13

1. Diabetes results in leakage of fluid into the retina through the blood retinal barrier

14

(BRB). This permeability increase is modulated by VEGF-A. VEGF-A165b is a splice

15

form of VEGF-A that is anti-angiogenic. We therefore tested to see whether it could

16

also be anti-permeability.

17

2. VEGF-A165b inhibited permeability of the outer BRB in vitro and retinal

18

permeability in vivo

19

3. This suggests that increasing VEGF-A165b expression could prevent increased

20

permeability in diabetic retinopathy.

ACCEPTED MANUSCRIPT

12

21

1 Use of open access articles is permitted based on the terms of the specific Creative Commons Licence under which the article is published. Archiving of non-open access articles is permitted in accordance with the Archiving Policy of Portland Press ( http://www.portlandpresspublishing.com/content/open-access-policy#Archiving).


Abstract (250 words)

23

Diabetic retinopathy (DR) is one of the leading causes of blindness in the

24

developed world. Characteristic features of DR are retinal neurodegeneration,

25

pathological angiogenesis and breakdown of both the inner and outer retinal barriers

26

of the retinal vasculature and RPE-choroid respectively. Vascular endothelial growth

27

factor-A (VEGF), a key regulator of angiogenesis and permeability, is the target of

28

most pharmacological interventions of DR. VEGF-A can be alternatively spliced at

29

exon 8 to form 2 families of isoforms, pro- and anti-angiogenic. VEGF-A165a is the

30

most abundant pro-angiogenic isoform, is pro-inflammatory and a potent inducer of

31

permeability. VEGF-A165b is anti-angiogenic, anti-inflammatory, cytoprotective and

32

neuroprotective. In the diabetic eye, pro-angiogenic VEGF-A isoforms are up-

33

regulated such that they overpower VEGF-A165b. We hypothesized that this

34

imbalance may contribute to increased breakdown of the retinal barriers and by

35

redressing this imbalance, the pathological angiogenesis, fluid extravasation and

36

retinal neurodegeneration could be ameliorated. VEGF-A165b prevented VEGF-A165a

37

and hyperglycaemia-induced tight junction breakdown and subsequent increase in

38

solute flux in RPE cells. In streptozotocin-induced diabetes, there was an increase in

39

Evans’ blue extravasation after both 1 and 8 weeks of diabetes, which was reduced

40

upon intravitreal and systemic delivery of rhVEGF-A165b. 8-weeks diabetic rats also

41

showed an increase in retinal vessel density, which was prevented by VEGF-A165b.

42

These results show rhVEGF-A165b reduce DR-associated BRB dysfunction,

43

angiogenesis and neurodegeneration and may be a suitable therapeutic in treating

44

DR.

45

2

VEGF-A165b inhibits diabetic retinal barrier function 46 47 48

Introduction Diabetic retinopathy (DR) affects both type 1 and type 2 diabetic patients[1]

49

and is the leading cause of blindness in the working population of the western world.

50

Proliferative diabetic retinopathy (PDR), like many other neovascular retinopathies,

51

occurs as a result of an ischaemic retina. Persistent hyperglycaemia causes

52

structural, functional and haemodynamic changes in the vasculature[2] that can

53

occur early on in diabetes before noticeable signs of retinopathy, eventually resulting

54

in focal ischaemia[3]. As a consequence of hypoxia, angiogenesis is triggered by

55

increased vascular endothelial growth factor (VEGF-A) expression[4]. Newly formed

56

vessels rarely recruit pericytes[5], lack vessel integrity and are prone to rupture, as

57

evidenced by microaneurysms and haemorrhage[6]. Diabetic macular edema (DME)

58

occurs from blood-retina-barrier (BRB) breakdown causing accumulation of fluid and

59

plasma proteins from leaky blood vessels[7]. Both DME and PDR worsen with

60

persistent hyperglycaemia and VEGF-A upregulation[8].

61

Treatments for DR can involve surgical and/or pharmacological intervention.

62

Laser photocoagulation for both PDR and DME aims to increase the amount of

63

oxygen that is available to the retina[9]. Since the discovery of VEGF-A in the

64

pathogenesis of DR[4], the use of anti-VEGF-A agents such as ranibizumab

65

(Lucentis, Genentech, San Francisco, USA), bevacizumab (Avastin, Genentech, San

66

Francisco, USA) and aflibercept (Eyelea, Regeneron, New York, USA) have been

67

widely investigated in their treatment of DR[10]. These anti-VEGF-A treatment

68

strategies are mainly targeted at the inner blood retinal barrier (BRB). Anti-VEGF-A

69

therapy is effective in treating DR in approximately 50% of patients[11]. Conversely,

70

anti-VEGF-A therapy in treating wet AMD is much more effective with up to 87% of 3


patients responding positively[12]. Neutralising or inhibiting an endogenous cytokine

72

such as VEGF-A could prove to be deleterious. Within the eye itself, suppression of

73

VEGF-A may lead to widespread geographic atrophy[13].

74

The blood-retina barrier is a selectively permeable region formed by tight-

75

junctions (TJs) on retinal endothelial cells (RECs)[14] and retinal pigmented

76

epithelial cells (RPE)[15], which form the inner and outer BRB respectively. The BRB

77

protects the neural retina from contents of the retinal and choroidal circulations, has

78

a role in homeostatic maintenance of the neural retina and controls fluid and

79

molecular flux between cells.

80

Unlike the retinal vasculature, the choroidal vasculature is fenestrated to allow

81

transport of nutrients from the systemic circulation to the RPE, which in turn

82

transports them to the outer retina[16, 17]. This transcellular movement is also used

83

to transport outer retinal waste products back to the choroidal circulation[16]. Due to

84

its close association with the choroidal circulation, tight junction integrity of the RPE

85

is critical for the maintenance of the outer BRB. Moreover, the fenestrae of the

86

choroidal plexus may allow macromolecular leakage from the choroid that may

87

otherwise be deleterious in the retina. As well as providing a barrier function and a

88

transporter of nutrients and waste, the RPE layer also protects against oxidative

89

stress by expressing superoxide dismutase, lipofuscin and melanin[18].

90

VEGF-A pre-mRNA can be alternatively spliced at exon 8[19] to generate two

91

families of isoforms. Selection of either a proximal splice site or a distal splice site

92

results in two different family of isoforms respectively[19]. The VEGF-Axxxb (x=amino

93

acid number) family has a different terminal six amino acid sequence (SLTRKD) to

94

the pro-angiogenic family, VEGF-Axxxa (CDKPRR)[20]. VEGF-A165b is the most

95

characterised of the VEGF-Axxxb family, and has properties that differ from VEGF4


A165a, its pro-angiogenic sister isoform. VEGF-A 165b binds to VEGFR2, however

97

unlike VEGF-A165a, it does not fully stablise neuropilin 1 binding, and only weakly

98

phosphorylates VEGFR2[21]. VEGF-A165a and VEGF-A165b competitively bind to

99

VEGFR2, so VEGF-A165b prevents pro-angiogenic activity in vivo.

100

VEGF-Axxxb isoforms are found in the adult human vitreous indicating that

101

both families of isoforms are required for adequate ocular function[22]. However, in

102

diabetic patients, this balance shifts in favour of VEGF-Axxxa, with the percentage of

103

VEGF-A in the vitreous that is anti-angiogenic decreasing (12.5±3%) compared to

104

that of normal eyes (65.3±7.2%)[22]. This indicates that progression of DR may be

105

influenced by a switch in VEGF-A splicing. Intravitreal administration of VEGF-A165b

106

reduces neovascularization in oxygen-induced retinopathy (OIR)[23] and in choroidal

107

neovascularisation[24] and has been shown to increase RPE cytoprotection[25].

108

VEGF-A165b is also neuroprotective and anti-apoptotic[26].

109

It therefore seems possible that VEGF-A165b could ameliorate the vascular

110

aspect of DR, and as it appears to be cytoprotective to retinal pigmented epithelial

111

cells, could act on the outer BRB, which would be an advantageous addition to “anti-

112

VEGF-A” therapies. We therefore tested the hypothesis that the cytoprotective, anti–

113

angiogenic and anti-permeability properties of VEGF-A165b enables it to prevent BRB

114

breakdown in vitro and in vivo in an animal model of diabetic retinopathy

115 116

Methods

117

Animal Experiments.

118

All animal experiments were undertaken under a UK Home Office project license,

119

and approved by the local ethics review board.

5

VEGF-A165b inhibits diabetic retinal barrier function 120 121

Streptozotocin–induced diabetes – 8 weeks A total of 20 female Sprague-Dawley rats (200-300g, Charles River, UK) were

122

weighed and fasted overnight, prior to diabetes induction. Rats were given a single

123

intraperitoneal (i.p) injection of streptozotocin (STZ, 50mg/kg, Sigma Aldrich, MO,

124

USA). A total of 10 control rats were injected with 300ȝl saline i.p. A third of an

125

insulin capsule (LinShin, ON, Canada) was implanted subcutaneously to maintain

126

body weight over the following 8 weeks in the diabetic rats. On day 4 post–induction,

127

blood glucose was tested from the tail vein. Rats with blood glucose of 15mmol/l and

128

above were deemed diabetic. STZ–injected rats that did not become hyperglycaemic

129

on day 4 were re-fasted overnight and then re-injected with STZ the following

130

morning. For the chronic experiment, diabetic rats were treated with 20ng/g

131

recombinant human (rh)VEGF-A165b (i.p, R&D Systems, MN, USA) or saline

132

biweekly for the experimental period. Control groups remained untreated for the

133

duration of the experiment.

134

Streptozotocin–induced diabetes – 1 week

135

A total of 10 rats were used for the 1 week Induction diabetes. STZ treatment was as

136

described above on 5 rats but without insulin treatment. On day 6 post-induction, rats

137

with blood glucose •15mmol/l received an intravitreal injection of vehicle (5µl PBS) in

138

one eye and 50ng VEGF-A165b in the contralateral eye as described below. Five

139

control animals had vehicle in one eye and no injection in the contralateral eye.

140

Intravitreal injections

141

Rats were anaesthetised with a single 10mg/ml intra-peritoneal (i.p) injection

142

of Dormitor (medetomine hydrochloride, Pfizer, UK) and Ketaset (ketamine

143

hydrochloride, Zoetis, NJ, USA). A 1.5cm 34-gauge hypodermic needle (Hamilton,

144

NV, USA) attached to a 5µl syringe (World Precision Instruments, FL, USA) was

6


inserted into the posterior chamber of the eye at a 45∨ angle. 5µl of either sterile PBS

146

or 50ng rhVEGF-A165b was injected into the vitreous. The following day animals were

147

perfused with Evans’ blue dye as described below.

148

Evans’ blue dye perfusion

149

The Evans’ blue dye perfusion technique was used as described [27] but with minor

150

alterations. Evans’ blue dye (Sigma Aldrich, MO, USA) was prepared by dissolving in

151

normal saline for a final concentration of 45mg/ml. Rats were anaesthetised (ip) with

152

10mg/ml Vetalar (ketamine hydrochloride, Boehringer Ingelheim, Germany) and

153

Domitor (medetomine hydrochloride, Pfizer, UK) with additional anaesthesia

154

provided as needed (iv). The left jugular vein and femoral artery were cannulated

155

with 33 gauge tubes and Evans’ blue (45mg/ml) injected via the left jugular vein over

156

10 seconds. Two minutes post-infusion, 200µl blood was removed from the femoral

157

artery, to establish the initial vascular Evans’ blue concentration. Fifteen minutes

158

post-infusion, 100µl blood was removed, and this volume was withdrawn every 15

159

minutes for two hours. Eyes were kept moist using Viscotears (Novartis,

160

Switzerland). Two hours post-infusion, the chest cavity was opened and 200µl blood

161

was withdrawn from the left ventricle, to determine the final plasma Evans blue

162

concentration. Rats were then perfused with 50ml saline through the left ventricle at

163

a physiological pressure of 120 mmHg, and exsanguinated. Both eyes were

164

enucleated and bisected at the equator. The retinas were dissected and weighed

165

(wet weight) and then dried at 70∨C overnight, and weighed again (dry weight). 120µl

166

formamide was added to each sample and incubated at 70∨C overnight. Following

167

dye extraction, the samples were centrifuged at 12000 rpm at 4∨C for 45 minutes.

168

The supernatant was used to determine the Evans blue concentration. The blood

7


samples were centrifuged at 4∨C at 12000 rpm for 45 minutes and diluted 1/100 in

170

formamide (Sigma Aldrich, MO, USA). The absorbance of all samples was then

171

measured at 620nm, and the concentration of Evans’ blue dye in formamide was

172

calculated using a standard curve of Evans’ blue in formamide. Formulae for solute

173

flux and permeability – surface area product (PA) used are:

174

a) Evans blue (EB) absorbance in wet weight (µg/g): ߤ݃ ‫ ܿ݊݋ܿܤܧ‬ቀ݈݉ ቁ ݂ܺ‫)݈݉( ݈݋ݒ݁݀݅݉݅݉ݎ݋‬ ܴ݁‫)݃( ݐ݄݃݅݁ݓݐ݁ݓܽ݊݅ݐ‬

175 176

EB absorbance in dry weight (µg/g): same as (a), substituting wet weight for dry

177

weight

178 179 180

181 182 183

184

b) EB accumulation in tissue fluid (µg/ml): ߤ݃ ‫ ܿ݊݋ܿܤܧ‬ቀ݈݉ ቁ ‫)݈݉(݈݋ݒ݁݀݅݉ܽ݉ݎ݋݂ݔ‬ ܴ݁‫ ) ݃( ݐ݄݃݅݁ݓݐ݁ݓܽ݊݅ݐ‬െ ݀‫)݃( ݐ݄݃݅݁ݓݕݎ‬

c) EB wet weight solute flux (µg/min/g):

ߤ݃ ‫) ݃ ( ܾ݁ܿ݊ܽݎ݋ݏܾܽݐ݄݃݅݁ݓݐ݁ݓܤܧ‬ ܶ݅݉݁ (min)

d) EB dry weight permeability-surface area product (ml/min/g): ߤ݃ ݉݅݊ ‫( ݔݑ݈݂݁ݐݑ݈݋ݏݐ݄݃݅݁ݓݕݎ݀ܤܧ‬ ) ݃ ߤ݃ ܶ݅݉݁ܽ‫ ܿ݊݋ܿܤܧܽ݉ݏ݈ܽܲ݀݁݃ܽݎ݁ݒ‬ቀ ቁ ‫݁݉݅ݐ݈ܽݐ݋ݐݔ‬ሺ݄‫)ݎ‬ ݈݉

185

The control permeability and solute flux were different between the 1 week and 8

186

week group. To compare the effect of 8 week and 1 week the values were

187

normalised to their control cohort by subtracting the relevant control cohort values

188

from the STZ treated (or STZ+VEGF-A165b treated) values..

8


Isolation of human primary RPE

190

All RPE isolation was performed as described [28]. Upon reaching 80% confluence,

191

cells were split and either transferred to a T75 flask (Greiner Bio One, Austria) or

192

used for experimental purposes. RPE were used until passage 8 and cultured in

193

DMEM:F12 + GlutaMAX (Invitrogen, CA, USA) supplemented with 10% FBS

194

(Invitrogen, CA, USA) and 1% penicillin / streptomycin (Invitrogen, CA, USA).

195

Trans–epithelial electrode resistance

196

Trans-epithelial electrode resistance (TEER) was measured using electric

197

cell-substrate impedance sensing (ECIS, Applied Biophysics, NY, USA) using a

198

1600R ECIS model. Cells were plated on 8 well plates (8W10E+ plates, Applied

199

Biophysics). Wells were pre-treated with 10mM L-cysteine (Sigma Aldrich, MO, USA)

200

for 30 minutes at 37∨ to neutralise the electrodes. Wells were then washed with

201

distilled water and stabilised in complete medium. Sixty thousand cells were plated

202

per well and were deemed confluent when the resistance was greater than 1000ȍ.

203

Measurements were collected at multiple frequencies (125Hz, 250Hz, 500Hz,

204

1000Hz, 2000Hz, 4000Hz, 8000Hz and 16000Hz) for one hour, to establish baseline

205

data. Once baseline was established, growth factor treatment was prepared in 50µl

206

medium. 50µl of medium was carefully aspirated from each well of the plate and

207

replaced with 50µl of growth factor treatment, and was allowed to equilibrate for 30

208

minutes before recommencing TEER evaluation. Plates were then assayed for

209

changes in TEER for at least 15 hours. Data was analysed at 500Hz to give an

210

indication of tight-junction integrity and paracellular flux.

211 212

Immunoblotting

9


Excised retinae or sub-confluent cells were lysed in RIPA lysis buffer supplemented

214

with protease inhibitor cocktail (Sigma Aldrich, MO, USA). A total of 50 ȝg of total

215

protein was resuspended in sample buffer, heated at 95°C for 5 minutes, and

216

subjected to SDS-PAGE under reducing conditions. Subsequently, proteins were

217

electrotransferred for 2 hours at 4°C to polyvinylidene fluoride membranes (PVDF,

218

Bio-Rad, CA, USA). The membranes then were exposed to primary antibodies

219

(rabbit anti-ZO1 [Invitrogen], mouse anti-occludin [Invitrogen], rabbit anti-VEGFR2

220

Ty115 [Cell Signaling Technology] all 1:500 5% BSA and mouse anti-Χ tubulin

221

[Santa Cruz] 1:1000 5% BSA. PVDF was washed in Tris-buffered saline/0.3%

222

Tween 20 (Sigma Aldrich, MO, USA), (TBS-T) and incubated with a secondary

223

peroxidase-conjugated antibody at a 1:10,000 dilution (Thermo Fisher Scientific, MA,

224

USA). Signals were detected by enhanced chemiluminescence (ECL) substrate (GE

225

Healthcare, UK). Western blots were also detected using a fluorescent-labelled

226

secondary antibody (donkey anti-mouse IR dye 1:7000, LI-COR Biosciences, NE,

227

USA), incubated at room temperature for 1 hour, followed by detection using LI-COR

228

Odyssey imaging system.

229 230 231

Immunofluorescence and Quantification To stain for tight junctions, RPE cells grown on 13 mm glass coverslips

232

(Thermo Fisher Scientific, MA, USA) were treated with either VEGF-A 165a, VEGF-

233

A165b, anti-VEGFxxxb, (clone 56/8) or glucose (Sigma Aldrich, MO, USA). Cells were

234

fixed at room temperature with 4% paraformaldehyde (Sigma Aldrich, MO, USA) 24

235

hours post treating, permeabilised in 0.05% Triton-X (Sigma Aldrich, MO, USA) in

236

PBS and then blocked in 5% normal goat serum (Sigma Aldrich, MO, USA) in PBS,

237

all for 30 minutes at room temperature. All cells were incubated in primary antibody

10


in PBS, at 4∨C overnight (mouse anti-occludin and rabbit anti-ZO1 [both Invitrogen at

239

4µg/ml]). Cells were then incubated with Alexa-Fluor 488 conjugated secondary

240

antibody (1:500 Invitrogen, CA, USA) in PBS, for 45 minutes at room temperature,

241

followed by incubation with Hoescht (1:5000, Sigma Aldrich, MO, USA), for ten

242

minutes at room temperature. Coverslips were mounted on glass slides with

243

Vectashield (both Thermo Fisher Scientific, MA, USA).

244

To stain retinal flatmounts, eyes were enucleated from culled rats and fixed in

245

4% PFA for 1 hour. Eyes were then hemisected and retinae were excised and

246

transferred to blocking solution (5% NGS, 3% Triton-X100 and 1%BSA) for 2 hours.

247

Retinae were incubated at 4∨C overnight in mouse anti-NeuN (Merck Millipore, MA,

248

USA, 2µg/ml) and Isolectin B4 (biotin conjugated from Bandeiraea simplicifolia,

249

Sigma Aldrich, MO, USA, 5µg/ml) prepared in blocking solution followed by 3 PBS

250

washes and then incubation with Alexa-fluor 488 (1:500, diluted in blocking solution)

251

at room temperature for 2 hours. After the last wash, samples were carefully flat-

252

mounted onto a microscope slide (VWR International, PA, USA) and mounted with

253

Vectashield containing DAPI (Thermo Fisher Scientific, MA, USA).

254

Tight junction staining in cells were imaged using an epifluorescence microscope

255

(Nikon Eclipse E400) ad analysed using a custom made macro on Fiji [29]. Flat

256

mounted retinae were imaged using confocal microscopy (Leica SPE). Tortuosity

257

and vascular density was determined using Imaris software (Bitplane, UK) and

258

confirmed using Fiji.

259

2.12. Statistical Analyses

260

Unless otherwise stated, all data are shown as mean ± SEM. All data, graphs and

261

statistical analyses were calculated with Microsoft Excel (Microsoft Office Software),

11


GraphPad Prism v6 (GraphPad Software Inc. CA, USA), Fiji and Imaris parametric

263

and non-parametric statistical tests were chosen upon the results of the D’Agostino-

264

Pearson normality test in Graphpad Prism. Curve fitting was carried out using Prism

265

All results were considered statistically significant at p