742

INTRODUCTION

Age-related macular degeneration (AMD), associated with choroidal neovascularization (CNV), is a major cause of severe vision loss in the elderly population.1,2

The newly formed choroidal vessels enter the subretinal space and eventually leak fluid beneath the retina, ultimately causing subretinal scarring, retinal pigment epithelium (RPE) atrophy, and permanent loss of cen-tral vision.3–5 The precise pathogenesis of this disease is not fully understood, and therefore, no single highly effective treatment of CNV has yet been developed.

A few treatment modalities, such as laser photocoagulation,6 photodynamic therapy,7 and macular translocation8 have been used to treat CNV

Current Eye Research, 35(8), 742–750, 2010Copyright © 2010 Informa Healthcare USA, Inc.ISSN: 0271-3683 print/ 1460-2202 onlineDOI: 10.3109/02713683.2010.486519

ORIGINAL ARTICLE

Intraocular Properties of a Repository Urokinase Receptor Antagonist Å36 Peptide in

Rabbits

Iryna A. Falkenstein1, Lingyun Cheng1, Terence R. Jones2, William R. Freeman1, Bruce Babson3, Igor Kozak1, Ajay M. Tammewar1, and Erin C. Barron1

1Joan and Irwin Jacobs Retina Center, Department of Ophthalmology, Shiley Eye Center, University of California, La Jolla, California, USA

2Ångstrom Pharmaceuticals, Inc., San Diego, California, USA3MicroConstants, San Diego, California, USA

ABSTRACT

Purpose: To evaluate the intraocular properties of Å36, a peptide that directly antagonizes the cell surface urokinase receptor and so prevents pericellular urokinase plasminogen activator activity.Methods: A total of 41 rabbits were used. The toxicity study tested three doses of Å36: 1 mg/ eye, 0.3 mg/eye, and 0.1 mg/eye. At 2 and 12 weeks, eyes were evaluated by ERG and histology. Phar-macokinetics were studied in rabbit eyes with the dose of 1 mg/eye in two different formulations: a micronized preparation and a non-micronized formulation. Eyes were enucleated at months 1, 2, 3, 4, and 5. Vitreous, retina, and choroid were collected separately for active Å36 analysis.Results: We did not find ocular toxicity with low and medium doses. At the highest dose, there was a transient toxicity at 2 weeks but was not notable at 3 months. The target choroid concentra-tion of Å36 was chosen as ≥100 nM. The micronized formulation at months 1, 2, and 3 combined, showed variable levels in the choroid giving 5/10 (50%) of the therapeutic level; the non-micronized formulation at months 4 and 5 combined, gave 6/7 (86%) of the therapeutic level, although this difference was not statistically significant.Conclusion: Å36 appears to be long lasting; the non-micronized formulation of Å36 gave concentrations above therapeutic level in the choroid at months 4 and 5. Optimization of the for-mulation of Å36, particularly the particle size, may result in a promising new compound for exu-dative age-related macular degeneration treatment.

KEYWORDS: Anti-angiogenesis; Ocular drug delivery; Ocular toxicity; Rat CNV model; Urokinase receptor antagonist

Received 09 July 2009; accepted 14 April 2010

Correspondence: Lingyun Cheng, M.D., Joan and Irwin Jacobs Retina Center, Department of Ophthalmology, Shiley Eye Center, University of California, San Diego (UCSD), 9415 Campus Point Drive, La Jolla, CA 92037. E-mail: cheng@eyecenter.ucsd.edu

09 July 2009

14 April 2010

© 2010 Informa Healthcare USA, Inc.

2010

Current Eye Research

0271-36831460-2202

10.3109/02713683.2010.486519

35

742750

8

Cur

r E

ye R

es D

ownl

oade

d fr

om in

form

ahea

lthca

re.c

om b

y U

nive

rsity

of

Nor

th T

exas

on

11/1

0/14

For

pers

onal

use

onl

y.

Intraocular Properties of Å36 743

© 2010 Informa Healthcare USA, Inc.

lesions of different locations. Unfortunately, even with successful closure of CNV without evidence of its recurrence, the final outcome of these treatments is simply stabilization of vision and a delay in visual acuity deterioration.9–11

Intravitreal antiangiogenic (anti-vascular endothe-lial growth factor (VEGF)) medications have been introduced in the past few years, such as pegaptanib sodium (Macugen),12 ranibizumab (Lucentis),13 and bevacizumab (Avastin),14,15 and have shown lesion stabilization and moderate clinical improvement of vision. Although showing benefits, the most effec-tive of these medications are disadvantageous in that they require monthly injections. Also, the clini-cal use of these medications showed that they do not seem to prevent the progression of retinal and subretinal fibrosis and RPE atrophy, which prevent many eyes from recovering good vision, even after a long period of treatment.16 The binding of urokinase plasminogen activator (uPA) to its receptor uPAR triggers twin cascades of events: first, destruction of the extracellular matrix (ECM) and second, intracel-lular signaling to program gene expression leading to cell migration, cell invasion, cancer metastasis, and angiogenesis.17–22 In ocular disease, high levels of uPA were observed in neovascular tissue in patients with proliferative diabetic retinopathy.23 VEGF-induced sustained pericellular permeability in cultured retinal microvascular endothelial cells is mediated by uPA/uPAR.24 Overexpression of the uPA/uPAR system has been shown during retinal neovascularization,25 in surgically excised CNV,26 and in laser-induced CNV.26 uPA/uPAR has been found to play a key role in the migration of human retinal pigment epithelial cells through ECM-like layers in vitro.27 Enhanced levels of uPAR have been found in surgically excised CNV and proliferative vitreoretinopathy membranes.27 Oxy-gen-induced retinal neovascularization is reduced in homozygous uPAR−/- mice.25 Homozygous uPA−/- mice also demonstrate decreased CNV development in a laser-induced CNV model.26

The above studies led to the testing of the linear octapeptide Å6, derived from the non-receptor-binding region of uPA and acting on the uPA/uPAR axis by an as yet incompletely defined mechanism, in efficacy studies of ocular disease in animals—it was active. Å6 inhibits hypoxia-induced retinal neovas-cularization in the mouse,22 and laser-induced CNV in mouse,28 rats,29 and monkey.30 More recently, Å6 was shown to inhibit uPA/uPAR-mediated migration and invasion of retinal endothelial cells31 and diabetic microvascular permeability in vivo.32 Specific block-ade of uPAR to prevent the twin cascades referred to above has been demonstrated in the field of cancer33–35 but not in ophthalmology. Preliminary studies on the

specific role of uPAR in ocular disease have given discordant results: Rakic et al. observed robust CNV at the site of laser-induced trauma in both homozy-gous uPAR−/- and wild-type mice—a result discount-ing a role for uPAR in CNV.26 Against this, McGuire et al. observed reduced retinal neovascularization in homozygous uPAR−/- mice.25 Furthermore, Le Gat et al. showed that an adenovirus-delivered, amino-terminal fragment of uPA, that contains the EGF-like growth factor domain that blockades uPAR, inhib-ited hypoxia-induced retinal neovascularization in mice by 78%.36 The amino-terminal fragment spans residues 1–135 of uPA and does not contain the Å6 sequence 136–143. However, it does contain the krin-gle domain in addition to the growth factor domain and the kringle may confer biological activity addi-tional to and separate from the blockade of uPAR. Alternatively, these discordant results may owe to differences in the underlying molecular pathologies that are induced by hypoxia and by laser insult in these models. We decided to assess the specific block-ade of uPAR in ocular neovascularization as a means to a new drug therapy.

Å36 is a rationally designed, 13-amino acid cyclic peptide of molecular formula C81H107N21O20 and molecular weight of 1695 Daltons (Figure 1, Ång-strom Pharmaceuticals, San Diego, California, USA). Å36 competitively inhibits uPAR with Kd = 4 nM in a surface plasmon resonance experiment and with Kd = 30 nM in a cell-based experiment.37 Inhibition of uPAR is expected to inhibit both extracellular proteol-ysis and signal transduction. Thus, Å36 inhibited the migration of human umbilical vein endothelial cells in a dose-related manner and is potently active in the 40 nM range. It inhibited the invasiveness of highly metastatic human prostatic carcinoma PC3MLN4 cells at concentrations of 100 and 500 nM.27 Because of the variability of results from these different assays, we chose 100 nM as our goal therapeutic concentration.

Å36 is poorly soluble in water; this arises from the dominance of hydrophobic residues (valine, tyrosine, phenylalanine, isoleucine, histidine, tryptophan, and the pendant phenyl group) that are readily visible in its structure. Its solubility is in stark contrast to that of Å6, which is highly water soluble, and which property probably explains the low choroidal concentrations

O

OO

Val-Ser-Asn-Lys-Tyr-Phe-Ser-Asn-Ile-His-Trp

NHNH

NH

HN

O

FIGURE 1 Chemical structure of Å36.

Cur

r E

ye R

es D

ownl

oade

d fr

om in

form

ahea

lthca

re.c

om b

y U

nive

rsity

of

Nor

th T

exas

on

11/1

0/14

For

pers

onal

use

onl

y.

744 I. A. Falkenstein et al.

Current Eye Research

observed following an intraocular injection of Å6.38 It is reasonable to assume that the hydrophobic nature of Å36 will suit it to treat CNV by local intravitreal injection. Lipophilic Å36 should penetrate freely through the retina into the target tissue—the chor-oid. In contrast to a systemic administration for vit-reoretinal disease, local drug administration bypasses the blood-ocular barriers, allowing higher intraocular levels for a longer duration, as well as avoiding pos-sible side effects associated with systemic therapy. We, therefore, undertook to study Å36. The aim of this study was to evaluate the toxicity and the pharma-cokinetics of an aqueous Å36 suspension, delivered via single intravitreal injection, in rabbit eyes. The results from this study were seen as necessary to establish the dose and dosing schedule for an efficacy study to come later.

MATERIALS AND METHODS

Formulation of Å36

The non-micronized formulation had a larger vari-ety of particles and a broader distribution of sizes from 15 to 150 µm in diameter, while the micron-ized formulation was more homogeneous and had a distribution of 15 to 40 µm in diameter under a calibrated histology microscope along with an image analysis program. Micronization was achieved by a manual crushing of particles until all were less than 40 micrometers in size. Å36 was sterilized with 70% pure ethanol and dried aseptically under vacuum. The particle sizes were not measured after ethanol sterilization. The sterilized Å36 was suspended in 0.9% sterile saline to form the formulation for the intravitreal injection.

Toxicity Study

The experiment was carried out using 24 New Zealand Red (NZR) rabbits, weighing between 1.8 and 3.0 kg. All procedures were performed with strict adherence to the Association for Research in Vision and Oph-thalmology (ARVO) guidelines for animal care and experimentation and the guidelines of the University of San Diego, Office of Veterinary Affairs.

Intravitreal Injection Technique

Twenty-four study rabbits were examined with indi-rect ophthalmoscopy prior to the injections, in order to exclude initial fundus abnormalities. Three concen-

trations of Å36 were used to evaluate toxicity: high (20 mg/ ml), medium (6 mg/ml), and low (2 mg/ ml). Solution was injected intravitreally in a volume of 0.05 ml (50 µl). The high dose eyes received 1 mg, medium dose eyes 0.3 mg, and low dose eyes 0.1 mg of the drug. The dose was determined based on that Å36 inhibited the migration of human umbilical vein endothelial cells in a dose-related manner and was potently active in the 40 nM range. The molecular weight of Å36 is 1695 Daltons and the potently active concentration will be 67.8 ng/ml. Rabbits have about 1.5 ml of vitreous and that translates to a 101.7 ng per rabbit eye. Since Å36 is minimally water soluble, the lowest dose was chosen to be 0.1 mg, which is 1000 times the dose needed to show active inhibition of human umbilical vein endothelial cell migration. From the lowest dose of 0.1 mg per eye, half-log escalation was used to yield 0.3 mg and 1 mg per eye.

The twenty-four rabbits were divided into three study groups of eight. In all groups, the right eye was injected using the aforementioned dosage of Å36, while the left eye was used as a control and injected with 0.05 ml of 0.9% saline. The intravitreal injection was as described previously by us.39 The injection site was located approximately 2 mm behind the limbus, in the temporal region of the globe, through the conjunc-tiva and sclera. Post injection, each eye was topically treated with 0.3% Tobramycin ointment, once daily for 5 days, to prevent possible inflammation.

Indirect ophthalmoscopy, slit lamp examination, and intraocular pressure (IOP) measurements were performed on day 3, weeks 1, 2, 3, 5, 7, 8, 10, and 12 (weekly up to 3 months post-injection). Intraocu-lar pressure was measured before dilation using a TonoPen (Medtronic Solan, Jacksonville, Florida, USA); averages of three consecutive measures (with ≤5% error) were recorded for each eye and aver-aged. Slit lamp anterior segment examinations, with vitreous clarity, as well as indirect ophthalmos-copy were evaluated using the methods previously described by our group.40,41 Representative fundus photos from each group were taken 2 and 12 weeks post-injection.

At week 2 and 12 after injection, eyes were evalu-ated by full-field electroretinography (ERG) and 12 rabbits were then sacrificed at each time point; eyes were enucleated and analyzed by light and electron microscopy.

ERG Technique

Eyes were fully dilated with 2.5% phenylephrine hydrochloride and 1% tropicamide and dark-adapted for 30 min. A dim red light was then used to complete

Cur

r E

ye R

es D

ownl

oade

d fr

om in

form

ahea

lthca

re.c

om b

y U

nive

rsity

of

Nor

th T

exas

on

11/1

0/14

For

pers

onal

use

onl

y.

Intraocular Properties of Å36 745

© 2010 Informa Healthcare USA, Inc.

the procedure as described previously.40 ERG analysis was based on both the amplitudes and the implicit time measurements of the a- and b-waves.

Histopathology

There were two time points for rabbit sacrifice: 2 weeks and 12 weeks (12 rabbits at each time point); rabbits were randomly chosen from each of three dosage groups, four animals from each group at each ending time point (2 and 12 weeks). The rabbits were deeply anesthetized via an intramuscular injection of ketamine (21 mg/kg) and xylazine (5.25 mg/kg) and then euthanized immediately after using an intracar-dial injection of 390 mg sodium pentobarbital. After sacrifice, eyes were enucleated and immersed in a 2% paraformaldehyde and 2.5% glutaraldehyde fixative and left to sit at 4°C overnight. The following day, each globe was bisected vertically through the optic nerve. One half of the globe was post-fixed in 10% formalin overnight, and later blocked in paraffin and processed for the light microscopic examination. Sec-tions of 5 microns were cut along the vertical meridian through the optic nerve using a microtome. The sec-tions were stained with hematoxylin and eosin. The remaining half was reserved for electron microscopy in fixative containing 2% paraformaldehyde and 2.5% glutaraldehyde.

Drug Level in Ocular Tissues

Seventeen New Zealand Red (NZR) rabbits, weigh-ing between 2.3 and 3.7 kg, were used in the pharma-cokinetic study. All procedures adhered to the ARVO Statement for the Use of Animals in Ophthalmic and Vision Research and the guidelines set forth by IACUC and the University of San Diego, Office of Veterinary Affairs.

The right eyes of all rabbits were intravitreally injected with 0.05 ml (50 µl) of the predetermined high-est non-toxic dose of Å36 suspension—20 mg/ml (1 mg per eye) through the pars plana, using a 5/8 inch 25G needle. All fellow eyes (left) were injected with 0.05 ml of 0.9% saline.

We used two formulations of the Å36 suspension: non-micronized (heterogeneous, large particles, with an average size of 80 microns), and micronized (a more homogeneous suspension with smaller par-ticles <40 microns). The animals injected with the micronized formulation were followed for up to 3 months, as it was hypothesized that this formulation would most likely release active Å36 for this period of time. Animals injected with the non-micronized

formulation were followed longer, up to 5 months, with the rationale that this formulation would allow slower and longer release of the active drug. At each monthly sacrificing time point, prior to anesthesia, rabbits’ eyes were fully dilated, using a topical application of 1% tropicamide and 2.5% phenyleph-rine hydrochloride. They were anesthetized and a vitreous tap, in a volume of 0.05 ml, was performed in an area away from the visible drug depot. Then animals were euthanized and the enucleated globes were kept on ice until we were able to perform the frozen dissection described previously.39 The globe was briefly submerged in -75°C 2-methylbutane for 50 sec and the globe was cut into two halves with the optic nerve parallel to the cutting surface and to one side of the globe. The cut made starting at the top of the globe and then through the cornea and optic nerve until the globe was cut into two halves, the blade held perpendicular to the cutting surface the entire time. The anterior part of each half was cut off at 2 mm behind the limbus. Anterior chamber frozen blocks were collected first then the vitreous blocks. Then the retina along with retinal pigment epithelium were easily scraped off the Bruch’s mem-brane choroidal complex after the retina changed from white to transparent. Finally, the Bruch’s mem-brane choroidal complex was forcefully scraped off the sclera. In this way there is little chance for contamination among various eye tissues. Retinal pigment epithelium can be sampled together with neuroretina or the Bruch’s membrane choroidal complex. In this study, retinal pigment epithelium was sampled together with neuroretina. Dissected tissues from the same eyes were stored separately in pre-weighed and pre-labeled vials. The vials were kept at –70°C until analysis. Assessment of the drug’s concentration in the target tissues was done using HPLC/mass spectroscopy. To prepare the samples for analysis, choroidal and retinal tissue samples and vitreous fluid samples were internally standardized with an analogue peptide, A17, and then subjected to sonic disruption in an aqueous mixture containing acetonitrile and perchloric acid. After centrifugation, a portion of the supernatant was further purified using solid phase extraction on a Varian ENV SPE cartridge (styrene-divinylbenzene type resin). After elution from the resin the extracts were chromatographed using gradient elution in a normal phase mode on a YMC PVA-Sil column (150 × 2 mm, 5 μm). Detection and quantification was performed by a tandem mass spectrometer (Quattro Ultima, Micromass, Waters, Milford, Mas-sachusetts, USA) in the electrospray positive mode using mass transitions from m/z 847.70 to 109.9 and m/z 855.0 to109.9 for A36 and the internal standard,

Cur

r E

ye R

es D

ownl

oade

d fr

om in

form

ahea

lthca

re.c

om b

y U

nive

rsity

of

Nor

th T

exas

on

11/1

0/14

For

pers

onal

use

onl

y.

746 I. A. Falkenstein et al.

Current Eye Research

respectively. The calibration range of the assay was 26.7 to 26,700 ng per gram of tissue or per milliliter of fluid.

Statistical Analysis

ERG parameters were compared between the drug injected eyes and the control eyes using Dunnett’s test. For IOP analysis, to take advantage of repeated mea-surements we used generalized estimating equations to compare the least squared means among the groups at two separate periods, within 2 weeks and between 3 and 12 weeks, using SAS (9.1, SAS Institute Inc. Cary, North Carolina, USA).

RESULTS

Toxicity Study

Two rabbits (8%) had a mild anterior chamber (AC) reaction and vitreous cells at week 1 in the fellow eyes, which were injected with 0.9% saline. None of the rabbit’s eyes injected with Å36 suspension showed AC or vitreal inflammation. The inflamed eyes were treated with a topical 0.3% Tobramycin ointment. AC

inflammation disappeared after 1 week of treatment. Lenses were clear in all eyes, and neither traumatic nor pharmacologic cataracts were observed (Figure 2). Eyes injected with the drug were compared clinically (indirect ophthalmoscopy and biomicroscopy) with their fellow eyes and did not show any significant difference.

IOP in all study eyes was not affected (Table 1) throughout the experiment. The electroretinography results did not differ between eyes injected with a high dose of Å36 and those injected with 0.9% saline at both the 2- and 12-week time points (Table 2, Figure 3). All eyes demonstrated normal waveforms.

The dissolution rate of each dose (low, medium, and high), in the vitreous, was estimated by indirect ophthalmoscopy at multiple time points during the first 3 months. The visible drug-depot in the vitre-ous was compared to the optic disc diameter and estimated in disc diameter units (DD). The initial size was assigned a value of 100% relative to which lesser values were recorded as the depot decreased over time. The average results for each dose over time are shown in Figure 4. Later, in the pharmacokinetic study, the amount of drug was assessed in target tis-sues precisely (Table 3).

Light microscopy histopathology was performed at 2 and 12 weeks. At 2 weeks, one of the four high

A B

C D

FIGURE 2 Fundus photos of rabbit eyes injected with micronized and non-micronized Å36 formulations. (A) This photograph represents a vitreous non-micronized high dose drug-depot of Å36 2 weeks after the injection and (B) shows the same eye 3 months after the injection. The white drug-depot is only slightly reduced in size and density, showing a very slow dissolution. (C) Fundus photo of a micronized high dose drug-depot of Å36, 2 weeks after the injection. (D) The same eye 3 months after the injection, showing the micron-ized formulation dissolves in the vitreous much faster than the non-micronized formulation, leaving a comparably smaller drug-depot.

TABLE 1 Intraocular pressure measurements

GroupShort term

(3 d, 1 wk, 2 wks)Long term

(3, 4, 5, 7, 8, 10, 12 wks)Control (0.9% saline)

17 ± 4 (n = 69) 16 ± 4 (n = 84)

Low dose (0.1 mg/eye)

16 ± 5 (n = 24) 17 ± 3 (n = 28)

Medium dose (0.3 mg/eye)

17 ± 4 (n = 24) 17 ± 3 (n = 28)

High dose (1 mg/eye)

16 ± 5 (n = 24) 17 ± 3 (n = 28)

p value 0.33 (groups); 0.40 (time points)

0.14 (groups); 0.22 (time points)

TABLE 2 ERG parameters

Group

2 weeks 12 weeksb wave implicit

b wave amplitude

b wave implicit

b wave amplitude

Control (0.9% saline)

43 ± 9 (n = 12)

90 ± 36 (n = 12)

58 ± 11 (n = 12)

86 ± 13 (n = 12)

Low dose (0.1 mg/eye)

44 ± 6 (n = 4)

81 ± 19 (n = 4)

59 ± 4 (n = 4)

80 ± 21 (n = 4)

Medium dose (0.3 mg/eye)

45 ± 4 (n = 4)

70 ± 20 (n = 4)

56 ± 10 (n = 4)

75 ± 13 (n = 4)

High dose (1 mg/eye)

44 ± 8 (n = 4)

85 ± 35 (n = 4)

50 ± 7 (n = 4)

78 ± 16 (n = 4)

p value >0.05 >0.05 >0.05 >0.05

Cur

r E

ye R

es D

ownl

oade

d fr

om in

form

ahea

lthca

re.c

om b

y U

nive

rsity

of

Nor

th T

exas

on

11/1

0/14

For

pers

onal

use

onl

y.

Intraocular Properties of Å36 747

© 2010 Informa Healthcare USA, Inc.

dose-injected eyes showed excessive vacuolization in inner and outer segments of photoreceptors (Figure 5A). The other three eyes showed a normal morphology as seen in the control eyes (Figure 5B). The structural changes in the vacuolized eye on elec-tron microscopy (Figure 6B—asterisks, 6D—arrow) were more striking. By 3 months, however, histology and electron microscopy did not show any abnormal-ities in all eyes even with the high doses (Figure 5C, Figure 6G, Figure 6H.)

Drug Level in Ocular Tissues

The analysis of tissues studied (vitreous tap, whole vit-reous, retina, and choroid) showed variable Å36 tissue concentrations (Table 3). Using the micronized formula-tion, the median concentration of Å36 in the choroid (the target tissue) exceeded the targeted therapeutic level of 100 nM in month 1 and month 2. As expected, the level dropped below the target in month 3 and no further samples were obtained. Overall, 50% of the tissues had choroid concentration above the therapeutic level. Using the non-micronized formulation, Å36 choroid concentrations were observed in month 4 and month 5. The concentrations were above the therapeutic concentrations in 86% of the samples analyzed. These results were consistent with the expectation that non-micronized formulation will release the drug slowly. No samples were analyzed prior to month 4.

DISCUSSION

Cell migration is a critical step in the angiogenesis cascade that involves proteolysis of the basement membrane and the extracellular matrix surrounding existing blood vessels. The urokinase plasminogen activator (uPA/uPAR) system has, for some decades, been seen involved in cellular invasion and angio-genesis, leading to tumor spread. More recently, expression of urokinase and its receptor (uPAR) has been noted in both retinal and choroidal neovascu-larization. Significant inhibition of CNV has been observed when cell surface-associated activity of uPA/uPAR is prevented with a specific inhibitor of this proteinase system. Inhibition of the uPA/uPAR system also causes reduction in the activity of VEGF, matrix metalloproteinases (MMPs), and fibroblast growth factor (FGF) in target tissues. Overproduc-tion of each of these factors contributes to abnormal angiogenesis, proliferation and inflammation during CNV development and simultaneous inhibition of all of these factors might be a more potent approach for CNV treatment than the inhibition of VEGF alone. As the current options for CNV treatment are not optimal, targeting the urokinase uPA/uPAR system appears to be an attractive method for an alternative, non-VEGF-based pharmacological therapy for CNV in AMD.

Å36 was developed as a specific inhibitor of uPAR. It is far less soluble than the previously studied Å6,38 and hence, has a more sustained presence in tissue, properties which make it more suitable as an intraocu-lar CNV treatment.

In this study, Å36 was shown to be non-toxic for ocular tissues in the highest therapeutic concentra-tion. Even though the signs of an early retinal tox-

0

0

20

40

60

80

100−100−40

−20

200

Time (mS)

ERG at 12 weeks with high dose Å36

ER

G a

mpl

itude

(m

Vol

t)

300

R99OD Å36 1mgR99OS with BSS

400 500

FIGURE 3 ERG waveforms from the eye (OD) that received the high dose of Å36 and from the eye (OS) that received 0.9% saline. Both waveforms and implicit times are normal.

0.01 2 3 4 5 6 7

Weeks

dose

Per

cent

age

of th

e in

itial

dru

g de

pot s

ize

8 9 10 11 12

0.10.20.30.40.50.60.70.80.91.0

H L M

FIGURE 4 The graph depicts the dissolution of Å36 non-micronized formulation in vitreous over a period of 3 months. Visible drug depot was estimated by indirect oph-thalmoscopy and measured in disc diameter units (DD). The low dose was reduced by more than half by the end of 3 months, and the high dose depot lost approximately 1/3 of its initial volume at 3 months.

Cur

r E

ye R

es D

ownl

oade

d fr

om in

form

ahea

lthca

re.c

om b

y U

nive

rsity

of

Nor

th T

exas

on

11/1

0/14

For

pers

onal

use

onl

y.

748 I. A. Falkenstein et al.

Current Eye Research

icity (2 weeks post-injection) were noted in one of 12 rabbits, the long-term (3 months post-injection) histopathologic results did not show any structural abnormalities. The mechanism of this toxicity in this eye is unclear.

Å36 appears to be long lasting; as our study shows, the non-micronized formulation dissolves slowly in the vitreous (Figure 4) with a drug-depot still visible at 5 months after a single injection. The non-micron-ized (larger particle size) formulation gave concentra-tions of Å36 in the choroid well above therapeutic levels at month 4, with a tendency to increase further by month 5 (Table 3). The assay was able to detect intact drug in the choroid, which implies that Å36 is biologically functional in the tissues. Long-lasting release and biologically functional in ocular tissues and no evidence of toxicity of drug accumulation are favorable features for a prospective new anti-CNV treatment.

We acknowledge the limitation of this study that the measured Å36 concentration in ocular tissues was highly variable and some pharmacokinetic param-eters cannot be reliably derived. One of the reasons is that Å36 is minimally water soluble and is a nominal concentration which often leads to a large variation from one injection to the other. In addition, at the time we performed the study there was not enough data available about the uniformity and stability of the Å36 suspension formulations. The suspension was kept at 4°C for 24 to 48 hr before dosing, which might have caused drug particles to aggregate and be less uniform. This might have also contributed to the large variation of the drug levels observed. To

improve the consistency of drug levels, the particle size of the drug-depot should be further optimized for more controlled vitreal release. Nonetheless, the current study demonstrated that optimization of the formulation of Å36, particularly the particle size, could result in a promising new compound for exudative AMD treatment.

TABLE 3 A36 formulation and drug levels in the ocular tissues

FormulationTime-point (months)

Vitreous tap (nM)

Wholevitreous (nM)

Retina (nM)

Choroid (nM)*

Micronized 1 139.2 14.3 113.9 209Micronized 1 750.7 16.3 23 11Micronized 1 0 38.6 12.7 6837.2Micronized 2 626.3 77.6 60 158.2Micronized 2 0 45.6 130 294.3Micronized 2 0 130.3 0 15.2Micronized 3 0 52.4 1332.4 2033.1Micronized 3 0 214.4 0 0Micronized 3 12 655 17.3 50.5Micronized 3 0 4.5 11 94Non-micronized 4 60 343.3 128 821Non-micronized 4 25 1363.8 234.5 121.7Non-micronized 4 57.5 1942.2 125.3 132.9Non-micronized 5 26.7 1903.2 5900 5864.3Non-micronized 5 28 2257.3 47.3 1735.7Non-micronized 5 242.3 1408.6 70.4 31.5Non-micronized 5 25.1 1554.7 7566.2 1941.7Our chosen therapeutic level of Å36 in target tissue is ≥100 nM

A B

C D

FIGURE 5 Light microscopy results at 2 and 12 weeks. (A) This figure shows retina structures 2 weeks after a high dose of Å36 injection. Compared to (B) that received 0.9% saline, (A) demonstrated excessive vacuoles in the inner and outer segments of the photoreceptors (*). Figure (C) shows a cross section of a retina 3 months after a high dose Å36 injection, which did not reveal a significant difference from (D), which was at 3 months but received 0.9% saline injection.

Cur

r E

ye R

es D

ownl

oade

d fr

om in

form

ahea

lthca

re.c

om b

y U

nive

rsity

of

Nor

th T

exas

on

11/1

0/14

For

pers

onal

use

onl

y.

Intraocular Properties of Å36 749

© 2010 Informa Healthcare USA, Inc.

ACKNOWLEDGMENTS

Declaration of interest: The authors report no conflicts of interest. The authors alone are responsible for the content and writing of the paper.

REFERENCES

[1] The Eye Diseases Prevalence Research Group. Prevalence of age-related macular degeneration in the United States. Arch Ophthalmol. 2004;122:564–572.

[2] Age-Related Eye Disease Study Research Group. Poten-tial public health impact of age-related eye disease study results: AREDS report No. 11. Arch Ophthalmol. 2003;121:1621–1624.

[3] Grossniklaus HE, Green WR. Choroidal neovascularization. Am J Opthtalmol. 2004;137:496–503.

[4] Gass JD. Biomicroscopic and histopathologic consideration regarding the feasibility of surgical excision of subfoveal neo-vascular membranes. Am J Ophthalmol. 1994;118:285–289.

[5] Submacular Surgery Trials Research Group. Histopathology and ultrastructural features of surgically excised subfoveal choroidal neovascular lesions. Report No. 7. Arch Ophthal-mol. 2005;123:914–921.

[6] Macular Photocoagulation Study Group. Occult choroidal neovascularization. Influence on visual outcome in patients with age-related macular degeneration. Arch Ophthalmol. 1996;114:400–412.

[7] Sharma S, Bakal J, Oliver-Fernandez A, et al. Photodynamic therapy with verteporfin for subfoveal choroidal neovascu-larization in age-related macular degeneration: Results of an effectiveness study. Arch Ophthalmol. 2004;122:853–856.

[8] Sadda SR, Pieramici DJ, Marsh MJ, et al. Submacular sur-gery trials pilot study investigators. Changes in lesion size after submacular surgery for subfoveal choroidal neovas-cularization in the submacular surgery trials pilot study. Retina. 2004;24:888–899.

[9] American Academy of Ophthalmology. Macular transloca-tion. Ophthalmology. 2000;107:1015–1018.

[10] Armbrecht AM, Aspinal PA, Dhillon B. A prospective study of visual function and quality of life following PDT in pat-ents with wet age related macular degeneration. Br J Oph-thalmol. 2004;88:1270–1273.

[11] TAP Study Group. Photodynamic therapy of subfoveal chor-oidal neovascularization in age-related macular degeneration with verteporfin: One-year results of 2 randomized clinical trials—Tap report. Arch Ophthalmol. 1999;117:1329–1345.

[12] Gragoudas ES, Adamis AP, Cunningham ET Jr, et al. VEGF inhibition study in ocular neovascularization clinical trial group. Pegaptanib for neovascular age-related macular degeneration. N Engl J Med. 2004;351:2805–2816.

[13] Rosenfeld PJ, Brown DM, Heier JS, et al. MARINA Study Group. Ranibizumab for neovascular age-related macular degeneration. N Engl J Med. 2006;355:1419–1431.

[14] Spaide RF, Laud K, Fine HF, et al. Intrravitreal bevacizumab treatment of choroidal neovascularization secondary to age-related macular degeneration. Retina. 2006;26:383–390.

[15] Avery RL, Pieramici DJ, Rabena MD, et al. Intravitreal bev-acizumab (Avastin) for neovascular age-related macular degeneration. Ophthalmology. 2006;113:363–372.

[16] Freeman WR, Falkenstein I. Avastin and new treatments for AMD: Where are we? Retina. 2006;26:853–858.

[17] Dano K, Andreasen PA, Grondahl-Hansen J, et al. Plasmi-nogen activators, tissue degradation, and cancer. Adv Cancer Res. 1985;44:139–266.

[18] Andreasen PA, Kjoller L, Christensen L, et al. The urokinase type plasminogen activator system in cancer metastasis: A review. Int J Cancer. 1997;72:1–22.

[19] Dear AE, Medcalf RL. The urokinase-type-plasminogen activator receptor (CD87) is a pleiotropic molecule. Eur J Biochem. 1998;252:185–193.

[20] Sidenius N, Blasi F. The urokinase plasminogen activator system in cancer: Recent advances and implications for prog-nosis and therapy. Cancer Metastasis Rev. 2003;22:205–222.

[21] Schmitt M, Wilhelm OG, Reuning U, et al. The urokinase plasminogen activator system as a novel target for tumour therapy. Fibrinolysis Proteolysis. 2000;14:114–132.

[22] Blasi F, Carmeliet P. uPAR: A versatile signalling orchestra-tor. Nat Rev Mol Cell Biol. 2002;3:932–943.

[23] Das A, McGuire PG, Eriqat C, et al. Human diabetic neo-vascular membranes contain high levels of urokinase and

A B

C D

E F

G H

FIGURE 6 Electron microscopy results at 2 weeks (A to D) and 3 months (E to H). (A) and (C) are from the eye that had 0.9% saline; B and D are from the eye shown in Figure 5A and received high dose Å36. It is clear that (B) showed swelling of the outer segments of the photoreceptors (asterisks) and (D) showed a damaged ganglion cell with disrupted nuclear mem-branes, enlarged perinuclear space, and the nipple of nuclear membranes (arrow). (E) and (F) are from the control eye; (G) and (H) are from the eye injected with a high dose of Å36, showing normal retinal structures.

Cur

r E

ye R

es D

ownl

oade

d fr

om in

form

ahea

lthca

re.c

om b

y U

nive

rsity

of

Nor

th T

exas

on

11/1

0/14

For

pers

onal

use

onl

y.

750 I. A. Falkenstein et al.

Current Eye Research

metalloproteinase enzymes. Invest Ophthalmol Vis Sci. 1999;40:809–813.

[24] Behzadian MA, Windsor LJ, Ghaly N, et al. VEGF-induced paracellular permeability in cultured endothelial cells involves urokinase and its receptor. FASEB J. 2003;17:752–754.

[25] McGuire PG, Jones TR, Talarico N, et al. The urokinase/uroki-nase receptor system in retinal neovascularization: Inhibition by Å6 suggests a new therapeutic target. Invest Ophthalmol Vis Sci. 2003;44:2736–2742.

[26] Rakic JM, Lambert V, Munaut C, et al. Mice without uPA, tPA,or plasminogen genes are resistant to experimental choroidal neovascularization. Invest Ophthalmol Vis Sci. 2003;44:1732–1739.

[27] Elner SG, Elner VM, Kindzelskii AL, et al. Human RPE cell lysis of extracellular matrix: Functional urokinase plasmi-nogen activator receptor (uPAR), collagenase and elastase. Exp Eye Res. 2003;76: 585–595.

[28] Das A, Boyd N, Jones TR, et al. Inhibition of choroidal neo-vascularization by a peptide inhibitor of the urokinase plas-minogen activator and receptor system in a mouse model. Arch Ophthalmol. 2004;122:1844–1849.

[29] Koh HJ, Bessho K, Cheng L, et al. Inhibition of choroidal neovascularization in rats by the urokinase-derived peptide Å6. Invest Ophthalmol Vis Sci. 2004;45:635–640.

[30] Koh HJ, Freeman WR, Azen SP, et al. Effect of a novel octapeptide urokinase fragment, Å6, on experimen-tal choroidal neovascularization in the monkey. Retina. 2006;26:202–209.

[31] Colombo ES, Menicucci G, McGuire PG, et al. Hepatocyte growth factor/scatter factor promotes retinal angiogenesis through increased urokinase expression. Invest Ophthalmol Vis Sci. 2007;48:1793–1800.

[32] Navaratna D, Menicucci G, Maestas J, et al. A peptide inhibi-tor of the urokinase/urokinase receptor system inhibits

alteration of the blood-retinal barrier in diabetes. FASEB J. 2008;22:3310–3317.

[33] Kobayashi H, Gotoh J, Fujie M, et al. Inhibition of metastasis of Lewis lung carcinoma by a synthetic peptide within growth factor-like domain of urokinase in the experimental and spon-taneous metastasis model. Int J Cancer. 1994;57:727–733.

[34] Min HY, Doyle LV, Vitt CR, et al. Urokinase receptor antag-onists inhibit angiogenesis and primary tumor growth in syngeneic mice. Cancer Res. 1996;56:2428–2433.

[35] Tressler RJ, Pitot PA, Stratton JR, et al. Urokinase receptor antagonists: Discovery and application to in vivo models of tumor growth. APMIS. 1999;107:168–173.

[36] Le Gat L, Gogat K, Bouquet C, et al. In vivo adenovirus-me-diated delivery of a uPA/uPAR antagonist reduces retinal neovascularization in a mouse model of retinopathy. Gene Ther. 2003;10:2098–2103.

[37] Jones TR, Haney, DN, Varga J, et al. Cyclic peptide ligands that target urokinase plasminogen activator receptor. US Patent Application 20040204348, October 14, 2004. Retrieved August 8, 2009 from www.uspto.gov.

[38] Koh HJ, Cheng L, Bessho K, et al. Intraocular properties of urokinase-derived antiangiogenic Å6 peptide in rabbits. J Ocul Pharmacol Ther. 2004;20:439–449.

[39] Cheng L, Hostetler KY, Lee J, et al. Characterization of a novel intraocular drug-delivery system using crystalline lipid antivial prodrugs of ganciclovir and cyclic cidofovir. Invest Ophthalmol Vis Sci. 2004;45:4138–4144.

[40] Cheng L, Hostetler KY, Chaidhawangul S, et al. Treatment or prevention of herpes simplex virus retinitis with intravitre-ally injectable crystalline 1-O-hexadecylpropanediol-3-phos-pho-ganciclovir. Invest Ophthalmol Vis Sci. 2002;43:515–521.

[41] Lu S, Cheng L, Hostetler KY, et al. Intraocular properties of hexadecyloxypropyl-cyclic-cidofovir in Guinea pigs. J Ocul Pharmacol Ther. 2005;21:205–209.

Cur

r E

ye R

es D

ownl

oade

d fr

om in

form

ahea

lthca

re.c

om b

y U

nive

rsity

of

Nor

th T

exas

on

11/1

0/14

For

pers

onal

use

onl

y.