A Peptide Derived from Type 1 Thrombospondin
Repeat–Containing Protein WISP-1 Inhibits Corneal and
Marisol del Valle Cano,1,2Emmanouil D. Karagiannis,2,3Mohamed Soliman,1
Belal Bakir,1Wenjuan Zhuang,1Aleksander S. Popel,3and Peter L. Gehlbach1
PURPOSE. Ocular neovascularization is the primary cause of
blindness in a wide range of prevalent ocular diseases includ-
ing proliferative diabetic retinopathy, exudative age-related
macular degeneration, and retinopathy of prematurity, among
others. Antiangiogenic therapies are starting to give promising
results in these diseases. In the present study the antiangio-
genic potential of an 18-mer peptide derived from type 1
thrombospondin repeat-containing protein WISP-1 (wisposta-
tin-1) was analyzed in vitro with human retinal endothelial cell
proliferation and migration assays. The peptide was also tested
in vivo in the corneal micropocket and the laser-induced cho-
roidal neovascularization (CNV) mouse models.
METHODS. Human retinal endothelial cells were treated with
the WISP-1 peptide and in vitro migration and proliferation
assays were performed. Also evaluated was the antiangiogenic
effect of this peptide in vivo using the corneal micropocket
assay and the laser-induced CNV model.
RESULTS. Wispostatin-1 derived peptide demonstrated antimi-
gratory and antiproliferative activity in vitro. Wispostatin-1
completely abolished bFGF-induced neovascularization in the
corneal micropocket assay. The peptide also demonstrated
significant inhibition of laser-induced CNV.
CONCLUSIONS. An inhibitory effect of Wispostatin-1 on ocular
neovascularization was found in vitro and in vivo. The iden-
tification of novel and potent endogenous peptide inhibitors
provides insight into the pathogenesis of corneal and cho-
roidal neovascularization. The results demonstrate potential
for therapeutic application in prevalent ocular disease. (In-
vest Ophthalmol Vis Sci. 2009;50:3840–3845) DOI:10.1167/
proliferative diabetic retinopathy, exudative age-related macu-
lar degeneration, retinopathy of prematurity, ischemic central
and branch retinal vein occlusions, infectious keratitis, trauma,
and various inflammatory ocular diseases.1
In the normal and developmentally mature ocular vascular
system, angiogenic stimulating factors, such as vascular endo-
thelial growth factor (VEGF) and angiogenic inhibitors exist in
a homeostatic balance.2In a variety of pathologic conditions,
such as hypoxia, ischemia, inflammation, infection, and
trauma, the balance between angiogenic stimulators and inhib-
itors is disrupted, leading to formation of pathologic vessels.
Abnormal growth of new vessels in the eye can limit light
transmission and affect physiological function.3
During the past two decades, the mechanisms of endoge-
nous suppression of angiogenesis have been elucidated by
studying several antiangiogenic proteins and derivative pep-
tides including pigment endothelium–derived factor (PEDF),
kallistatin, angiostatin, and thrombospondin-1 (TSP-1). PEDF
has been identified as a potent antiangiogenic factor.4,5It is a
potent inhibitor of endothelial cell (EC) proliferation and mi-
gration.6Kallistatin levels are significantly reduced in the vit-
reous from patients with proliferative diabetic retinopathy and
in the retina of diabetic rats, suggesting that it is implicated in
diabetic retinopathy.7,8Kallistatin inhibits retinal neovascular-
ization and reduces vascular leakage in the retina.9In patients
with diabetic retinopathy significant elevation of vitreous an-
giostatin and decreased VEGF concentrations are observed in
eyes previously treated with laser photocoagulation.10The
TSP-1 protein has been identified as an important ocular angio-
genesis inhibitor.11,12The protein is primarily expressed in the
limbal area where it is speculated that it creates a natural
barrier to vascular invasion into the corneal stroma.13
We have identified numerous peptides derived from pro-
teins containing TSP-1 repeats that share similarities with the
known angiogenesis inhibitors Mal-II and -III.14–16In the
present study, one of these peptides, wispostatin-1 (WISP-1),
composed of 18 amino acids, derived from TSP-1 repeat-con-
taining protein WISP-1 (human sequence: SPWSPCSTSCGL-
GVSTRI, and its orthologous mouse sequence: SPWSPCST-
TCGLGISTRI) inhibited proliferation and migration in human
retinal EC assays (human sequence). WISP-1 also significantly
inhibited neovascularization in the corneal micropocket assay
(mouse sequence) and the laser-induced choroidal neovascu-
larization (CNV) model (both human and mouse sequences).
WISP-1 is 52% identical with the known antiangiogenic pep-
tides (Mal-II and -III) derived from the TSP-1 domains of the
TSP-1 protein (Fig. 1A). It is a small peptide with molecular
mass of 1837 Da. It is composed of mostly hydrophilic amino
acids, apart from a small stretch of five amino acids near the
C-terminus that are hydrophobic (Fig. 1B). WISP-1 is slightly
positively charged at neutral pH (charge ?1.9) and has an
isoelectric point of 9.2, which means that under neutral pH
environments the peptide retains its positive charge.
cular neovascularization is the primary cause of blindness
in a wide range of prevalent ocular diseases, including
From the Departments of1Ophthalmology, Wilmer Eye Institute,
3Biomedical Engineering, Johns Hopkins University, School of
Medicine, Baltimore, Maryland.
2Contributed equally to the work and therefore should be consid-
ered equivalent authors.
Supported by Fight for Sight (MdC), the Sheila West Research
Grant Award (MdC), the JG Foundation (PLG); an unrestricted grant
from Research to Prevent Blindness (Wilmer Eye Institute); a Research
to Prevent Blindness Career Development award (PLG); the Jack and
Gail Baylin Philanthropic Fund; the Johns Hopkins University Fund for
Medical Discovery (PLG); a gift from Kenneth and Brenda Richardson
(PLG), and a gift form Mr. and Mrs. George Laniado.
Submitted for publication July 28, 2008; revised December 7,
2008, and January 10, 2009; accepted June 15, 2009.
Disclosure: M. del V. Cano, None; E.D. Karagiannis, P; M.
Soliman, None; B. Bakir, None; W. Zhuang, None; A.S. Popel, P;
P.L. Gehlbach, None
The publication costs of this article were defrayed in part by page
charge payment. This article must therefore be marked “advertise-
ment” in accordance with 18 U.S.C. §1734 solely to indicate this fact.
Corresponding author: Peter L. Gehlbach, Retina Division, Wilmer
Eye Institute, Johns Hopkins University, School of Medicine, 1550
Orleans Street, Baltimore, MD 21231; firstname.lastname@example.org.
Investigative Ophthalmology & Visual Science, August 2009, Vol. 50, No. 8
Copyright © Association for Research in Vision and Ophthalmology
MATERIALS AND METHODS
Using a bioinformatics algorithm, we searched within the human
proteome for antiangiogenic peptides and identified a set of peptides
17 to 20 amino acids in length that are derived from the TSP-1 repeats
of human proteins and share similarities with the known angiogenesis
inhibitors Mal-II and -III.14–16We named the predicted peptides ac-
cording to the names of the proteins of origin. We have demonstrated
that these peptides inhibit the proliferation and migration of human
umbilical vein ECs (HUVECs) in vitro. In the present study, we exper-
imentally tested the ocular antiangiogenic potency of WISP-1 derived
from the protein WISP-1 (Wnt-1 induced secreted protein 1) in vitro
using human retinal EC (HREC) proliferation and migration assays and
an in vivo corneal micropocket and a CNV mouse model. The prop-
erties of the peptide were calculated with the ProtScale tool.17
Peptide Synthesis and Handling
The peptides were produced by a commercial provider (Abgent, San
Diego, CA) using a solid-phase synthesis technique. HPLC and mass
spectrometry analyses of each peptide were performed. In each case,
the synthetic procedure yielded ?95% pure peptide. Scrambled pep-
tides were also produced to be used as a negative control.
HRECs from a single donor were obtained from The Applied Cell
Biology Research Institute (Kirkland, WA). The cells were propagated
in EGM-2 medium, consisting of a basal cell medium with 2% FBS,
growth factors (hbFGF and VEGF) and antibiotics (gentamicin/ampho-
tericin B). All the cells used were from passages 3 to 6.
In Vitro Cell Viability Assay
We assessed the effects of WISP-1 on the proliferation of ECs by
measuring the metabolic activity of the live cells using a colorimetric
cell proliferation reagent (WST-1; Roche, Indianapolis, IN). Approxi-
mately 2000 cells/well were seeded in a 96-well microplate without
any extracellular matrix substrate and exposed for 3 days to different
peptide concentrations: 0.01, 0.1, 1, and 10 ?g/mL and 20, 30, and 40
?g/mL. The molecular mass of the peptides is approximately 2 kDa;
thus, the aforementioned concentrations are equivalent to 5 nM, 50
nM, 500 nM, 5 ?M, 10 ?M, 15 ?M, and 20 ?M. Each of the concen-
trations was tested simultaneously in quadruplicate, and each of the
experiments was repeated three times. As a positive control, we
applied 100 ng/mL (0.22 ?M) TNP-470. As a negative control (normal
viability) the cells were cultured without any agent in full medium
containing growth factors and serum.
In Vitro Cell Imaging
Two thousand cells/well were seeded in a 96-well microplate without
any extracellular matrix substrate and were allowed to attach over-
night, as previously described, and exposed for 3 days to different
peptide concentrations: 10, 20, 30, and 40 ?g/mL. The medium with
the peptide was removed and a DiI cell-labeling solution (Vybrant;
Invitrogen, Carlsbad, CA) was added according to the manufacturer’s
protocol and allowed to stand for 15 minutes. The cells were then
washed three times with medium and imaged at 549 nm absorbance
and 565 nm emission.
In Vitro Cell Migration Assay
A modified Boyden chamber migration assay (BD Biosciences, San Jose,
CA) was used to examine EC migration in the presence of an activator and
the peptide solution. In our case, we used 20 ng/mL VEGF (Invitrogen)
and 1, 10, and 30 ?g/mL of the tested peptide solution. A serum- and
growth factor–free medium was used as a negative control, and 20 ng/mL
VEGF was used as a positive control. The chambers were then incubated
for 20 hours at 37°C. The cells that had migrated into the lower chamber
were stained with calcein (Invitrogen–Molecular Probes, Eugene, OR) for
90 minutes before termination of the experiment and counted.
All animal studies were conducted in accordance with an animal
protocol reviewed and approved by the Johns Hopkins University
Animal Care and Use Committee and in accordance with the ARVO
Statement for the Use of Animals in Ophthalmic and Vision Research.
Normal C57BL/6 mice were used in these studies. Six- to 8-week-old
male C57BL/6 mice were obtained from Harlan (Indianapolis, IN) and
peptide, WISP-1. (A) The peptide is 52% identical with the known anti-
angiogenic peptides derived from the TSP-1 protein (Mal-II and -III). (B)
WISP-1 is mostly hydrophilic apart from a short stretch of five amino acids
near the C terminus that are hydrophobic. Blue: the hydrophilic residues;
red: the hydrophobic residues. (C) The probability that those amino acids
are buried in an aqueous environment is low (?12%).
The properties of the newly identified ocular antiangiogenic
IOVS, August 2009, Vol. 50, No. 8
WISP-1 Peptide Inhibits Ocular Neovascularization3841
were housed on a 12-hour light–dark cycle, with food and water
provided ad libitum. All mice were fed with a standard caloric diet for
Preparation of Pellets
Hydron pellets (polyhydroxyethyl-methacrylate [polyHEMA], Inter-
feron Sciences, New Brunswick, NJ) were made so that each one
contained approximately 40 ng of bFGF (Peprotech Inc., Rocky Hill,
NJ), and 30 ?g of sucrose aluminum sulfate (Sucralfate; Sigma-Aldrich,
St. Louis, MO). Briefly, in this modified technique, we prepared 6 ?L of
a suspension containing 4.1 ?L of 12% Hydron in ethanol, 1250 ng of
bFGF (1 ?L of a 1250 ng/?L solution), and 900 ?g of sucralfate (0.9 ?L
of a 1 mg/?L solution). Subsequently, 2 ?L of this suspension was
taken with the use of a pipetter (0.1 to 2.5 ?L; Eppendorf North
America, Westbury, NY) and deposited onto an autoclaved 15 ?
15-mm piece of nylon mesh (3–300/50, approximate pore size 0.4 ?
0.4 mm; Tetko, Lancaster, NY). Ten holes of this mesh were filled with
each 2 ?L of our suspension, which resulted in a pellet volume of
approximately 0.2 ?L. Under the aid of a surgical microscope (Carl
Zeiss Meditec, Inc.), the fibers of the mesh were pulled apart, and
uniformly sized pellets were selected for implantation. In the case of a
peptide application the Hydron pellets were made in a similar manner
as described for bFGF. In this case, we prepared 8 ?L of a suspension
containing 3.5 ?L of 12% Hydron in ethanol, 1625 ng of bFGF (1.3 ?L
of a 1250 ng/?L solution), 200 ?g of peptide (2 ?L of a 100-?g/?L
solution), and 1200 ?g of sucralfate (1.2 ?L of a 1-mg/?L solution). The
suspension was deposited onto the nylon mesh as just described. Each
pellet of approximately 0.2 ?L contained approximately 40 ng of bFGF,
100 ?g/mL of peptide, and 30 ?g of sucralfate.
Corneal Micropocket and Pellet Implantation
Mice received general anesthesia with an intraperitoneal injection of
ketamine/xylazine (45 and 4.5 mg/kg, respectively), and topical anes-
thesia (1 drop of proparacaine 0.5%) before each intervention. Surger-
ies were performed by the same surgeon (WZ) with the use of an
ophthalmic operating microscope (Carl Zeiss Meditec, Inc.). Using a
modification of the technique described by Kenyon et al.,18a 1-mm
stromal linear keratectomy was performed with a disposable 15° sur-
gical blade (OR Specialties, Inc., Baltimore, MD). The incision was
made parallel to the insertion of the lateral rectus muscle, approxi-
mately 1.5 mm away from the temporal limbus. An intrastromal micro-
pocket (1 ? 0.5 mm) was dissected toward the limbus with a lamellar
dissecting blade originally designed for deep anterior lamellar kerato-
plasty (Dutch Ophthalmics, Kingston, NH). A single pellet was im-
planted and advanced toward the temporal corneal limbus, within 0.7
to 1.0 mm. A drop of levofloxacin 0.5% (Quixin, Santen Inc., Napa, CA)
was instilled in each eye immediately after the surgical procedure. In
addition, topical levofloxacin eye drops were also administered three
times daily, for 6 days. For analgesia, SC buprenorphine (0.1 mg/kg)
was administered to each animal every 12 hours for 4 days.
Design of the Corneal Micropocket Experiment
Before the surgery, the animals were randomly distributed to receive
one of three pellet types: bFGF, bFGF with peptide, or empty. The eyes
were photographed at day 6 after pellet implantation with a digital camera
(Power Shot S2 1S; Canon, Tokyo, Japan) with the aid of a dissecting
microscope (Steamii 2000 C; Carl Zeiss Meditec) and then the animals
were killed. The neovascularization emerging from the limbal vessel was
(NV): the maximal vessel length (VL) extending from the limbal vascula-
ture toward the pellet, and the contiguous circumferential zone of NV
(clock hours [CN] of NV, where 1 clock hour equals 30° of arc). The area
(millimeters) of NV was then calculated, as described by Kenyon et al.18
The areas of NV are expressed in percentage, with the growth factor
control (bFGF) representing 100% of neovascularization, and the empty
pellet control representing 0% of neovascularization. One week after
pellet implantation, the animals were killed, and the neovascularization
emerging from the limbal vessel was evaluated. Two parameters were
used to assess NV: the maximal VL extending from the limbal vasculature
toward the pellet, and the contiguous circumferential zone of NV (CN of
NV, where 1 clock hour equals 30° of arc). The area (in square millime-
ters) of NV response was then calculated using the formula of half an
ellipse: 0.5 ? ? ? R1 ? R2, where the conversion factor for R2 (mm) ?
CN ? 0.4 (mm) and R1 (mm) ? VL (mm), and Area (mm2) ? ? ? VL
(mm) ? CN ? 0.2 (mm).19The area of NV is expressed as a percentage,
with the positive control (bFGF) representing 100% of neovascularization,
and the negative control (empty pellet) representing 0% of neovascular-
Adult C57BL/6 mice were used for these experiments. The mice were
anesthetized and a drop of tropicamide was applied for pupillary
dilatation. Bruch’s membrane was ruptured with laser photocoagula-
tion at three locations in each eye. Briefly, laser photocoagulation
(400-nm wavelength, 50-?m spot size, 0.05-second duration, and
120-mW intensity) was delivered via the slit lamp delivery system with
a handheld cover slide used as a contact lens. Burns were performed in
the 9, 12, and 3 o’clock positions two to three disc diameters from the
optic nerve. Production of a vaporization bubble at the time of laser-
ing, which indicates rupture of the Bruch’s membrane, is an important
factor in obtaining CNV, and therefore only burns in which a bubble
was produced were included in the study.
After laser rupture of Bruch’s membrane, the mice were injected
intravitreously with 100 ?g/mL of the peptide studied. Intravitreous
injections were performed with a pump microinjection apparatus
(Harvard Apparatus, Holliston, MA) and pulled glass micropipettes.
Each micropipette was calibrated to deliver 1 ?L of vehicle containing
the peptide on depression of a foot switch. The mice were anesthe-
tized and the pupils dilated. Under a dissecting microscope, the sharp-
ened tip of the micropipette was passed through the sclera, just behind
the limbus into the vitreous cavity, and the foot switch was depressed.
The injections were repeated 7 days after Bruch’s membrane rupture
to replenish the peptide concentration in the vitreous cavity.
Measurement of the Sizes of Laser-Induced
Two weeks after laser treatment, the sizes of CNV lesions were mea-
sured in choroidal flatmounts.20Mice used for the flatmount technique
were anesthetized and perfused with 1 mL phosphate-buffered saline
(PBS) containing 50 mg/mL fluorescein-labeled dextran (500,000 aver-
age molecular weight; Sigma-Aldrich), as previously described.21The
eyes were removed and fixed in 10% phosphate-buffered formalin. The
cornea and lens were removed, and the entire retina was carefully
dissected from the eye cup. Radial cuts (four to seven; average, five)
were made from the edge to the equator, and the eye cup was
flatmounted in aqueous medium (Aqua Poly/Mount; Polysciences, Inc.,
Warrington, PA) with the sclera facing down. The flatmounts were
examined by fluorescence microscopy (Axiovert 200M; Carl Zeiss
Meditec, Thornwood, NY), and images were digitized with a CCD
camera (Axiocam MRc5). Image analysis software (AxioVision Soft-
ware; Carl Zeiss Meditec) was used to measure the total area of
hyperfluorescence associated with each burn, corresponding to the
total fibrovascular scar. The areas within each eye were averaged to
obtain one experimental value, and mean values were calculated for
each treatment group and compared by Student’s paired t-test.
For the in vitro experiments, Student’s t-test was used. For NV, de-
scriptive statistics were expressed as the mean ? SE. Nonparametric
tests were used for the analysis of NV. A paired Student’s t-test was
used to conduct a statistical analysis for CNV.
3842Cano et al.
IOVS, August 2009, Vol. 50, No. 8
Peptide Effects: Proliferation and Migration
Two key characteristics of the angiogenic process are the prolif-
eration of ECs near a maternal vessel, where the novel sprouting
bud emerges, and their coordinated migration along a chemotac-
tic gradient. To evaluate the antiproliferative and antimigratory
potency of WISP-1, we used in vitro proliferation and migration
assays with HRECs and measured the peptide potency compared
to positive and negative control experiments. We tested the
ability of the peptide to inhibit the proliferation of HRECs in an in
vitro assay (Fig. 2A). As a negative control, we added only the full
medium to the cells; as a positive control, we added the full
medium with 100 ng/mL of TNP-470 (fumagillin). TNP-470 is a
microtubule-stabilizing agent that has the ability to induce EC
apoptosis. The optical signal from the proliferation assay was
scaled so that 0% represented the signal from the negative control
was added. We expressed this scaled result as peptide activity
relative to the activity of TNP-470. The tested peptide reached a
maximum efficacy of approximately 20% proliferation inhibition.
To observe whether the peptide induces any alterations to the
morphology of the ECs we imaged the cells after applying increas-
ing concentrations of WISP-1 (10, 20, 30, and 40 ?g/mL). The
cells were stained with Dil-LDL, and images were taken after 3
days of peptide application. From the images, it was apparent that
at increasing peptide concentrations. Also the round morphology
of the cells at higher peptide concentrations suggests the de-
creased viability of the cells in the presence of WISP-1 (Fig. 2B).
right: the positive and negative controls used to scale the results from the proliferation and migration experiments. (B) The morphology of the ECs
was altered after increasing concentrations of WISP-1 were applied for 3 days. After 3 days of peptide application, the cells were stained with
Dil-Ac-LDL and then imaged with phase microscopy. At increasing peptide concentrations, the cells were less dense and acquired a round
morphology typical of apoptosing cells. Scale bar, 100 ?m.
In vitro screening of WISP-1 using human retinal ECs. (A) The activity of the peptide in the proliferation and migration assays. Insets,
(asterisk) at postoperative day 6. bFGF (positive control) showed abundant neovascularization (arrow), and the empty pellet (negative control)
showed none. Right: the mean neovascularization area ? SE on day 6 after pellet implantation. WISP-1 reduced neovascularization by 96.6%
compared with the untreated control (n ? 10).
Inhibition of NV by WISP-1 (mouse sequence). Left: a representative mouse from each group, showing NV (arrows) and pellet
IOVS, August 2009, Vol. 50, No. 8
WISP-1 Peptide Inhibits Ocular Neovascularization3843
To assess the ability of the peptides to inhibit the migration
of HRECs, we used a modified Boyden chamber assay to mea-
sure the suppression of VEGF-stimulated migration of HRECs
through a porous membrane covered by laminin (Fig. 2A). As
a positive control (causing decreased motility), we incubated
the cells in serum- and VEGF-free cell medium. As a negative
control, we incubated the cells with VEGF alone, without
peptide. WISP-1 suppressed the migration of retinal ECs
through the laminin coated porous membrane by 40%.
In Vivo Assays of Neovascularization
WISP-1 demonstrated significant bioactivity in the corneal mi-
cropocket assay. Treatment with the mouse peptide sequence
(Fig. 3) suppressed experimentally induced NV to levels com-
parable to that induced by empty pellets. WISP-1 also signifi-
cantly inhibited laser-induced CNV (Figs. 4, 5). In control mice,
laser injury resulted in well-circumscribed hyperfluorescent
areas at the sites of Bruch’s membrane injury indicating the
formation of CNV. These lesions predominantly represent out-
growths of the choroidal vasculature at sites of laser injury. In
contrast, laser injury in mice treated with intravitreous WISP-1
(mouse sequence) showed a 43% inhibitory effect on the area
of CNV (P ? 0.05). Mice treated with a WISP-1 human se-
quence exhibited an approximately 28% reduction in the area
of CNV (P ? 0.059). Treatment with a scrambled amino acid
sequence peptide resulted in a slight increase in the area of
neovascularization (the human sequence was used; data not
Several prevalent ocular diseases have been identified that are
characterized by pathologic angiogenesis, including age-related
macular degeneration, diabetic retinopathy, and retinopathy of
prematurity. Discovery of angiogenesis inhibitors contributes
to the development of therapeutic treatments of these dis-
eases. Moreover, the identification of endogenous inhibitors
also contributes to fundamental understanding of the angio-
genic balance in the ocular tissue.
The TSP-1 protein has been identified as a potent angio-
genic inhibitor that plays an important role in maintaining
vascular homeostasis in the eye.22TSP-1 is synthesized and
secreted by cultured RPE cells and is upregulated by vitamin
A.23Immunohistochemistry-stained sections of human eyes
indicate that in the macula region TSP-1 is found in Bruch’s
membrane, the choriocapillaris, and the larger choroidal ves-
sels.24The aqueous humor and vitreous from normal human
eyes also contain elevated levels of the TSP-1 protein.25Of
interest, the TSP-1 levels in the aqueous humor and vitreous are
downregulated in some diabetic animal models. This associa-
tion suggests that ocular vascular abnormalities in the diabetic
animals results from decreased levels of the thrombospondin
protein.22TSP-1 is also implicated in the pathogenesis of CNV
in the setting of macular degeneration.23There is evidence
suggesting a relationship between spatial localization of TSP-1
and disease in eyes with macular degeneration.24Notably,
expression of antiangiogenic TSP-1 is significantly decreased in
Bruch’s membrane and choriocapillaris in AMD eyes and TSP-1
levels correlate inversely with severity of disease.24
In the present study, we tested the activity of an endoge-
nous peptide derived from a protein containing a throm-
bospondin repeat, in two in vivo ocular angiogenesis models,
the corneal micropocket model and the laser-induced CNV
model. The peptide was previously identified by using a bioin-
formatics computational approach, and the predictions were
validated in human umbilical vein EC proliferation and migra-
tion assays.15WISP-1 strongly inhibited bFGF-induced neovas-
cularization in the corneal micropocket model. Qualitatively
lar neovascularization. C57BL/6 mice underwent laser-induced rupture
of Bruch’s membrane at three locations in each eye. They were then
injected with 1 ?L of vehicle containing the WISP-1 at a concentration
of 100 ?g/mL, and the injection was repeated on day 7. Fourteen days
after laser treatment, the mice were perfused with fluorescein-labeled
dextran, and the area of CNV was measured by image analysis of
choroidal wholemounts (A–D). Fourteen days after rupture of Bruch’s
membrane, there were large areas of CNV at rupture sites in the
untreated control eyes (PBS injected) (A, C, arrows). Mice that re-
ceived an intravitreous injection of WISP-1 mouse and human se-
quences (B, D, respectively) showed less CNV (arrows) than the
baseline level at 14 days after laser treatment (A, C). Scale bar, 100 ?m.
Intravitreous injection of WISP-1 caused inhibition of ocu-
area of CNV was measured on cho-
roidal flatmounts by computerized
image analysis of PBS-injected con-
trols (n ? 30) and of eyes given in-
travitreous injection of WISP-1 (n ?
30); both mouse and human se-
quences were used. (A) WISP-1 in-
hibited CNV by 43% in the mouse
sequence (*P ? 0.02); (B) in the hu-
man sequence, it inhibited neovascu-
larization by 28% (P ? 0.059); the
difference is nearly statistically signif-
WISP-1 inhibits CNV. The
3844Cano et al.
IOVS, August 2009, Vol. 50, No. 8
similar results were reported when histidine-rich glycoprotein
(HRGP) was used, which masked the antiangiogenic epitope of
Furthermore, addition of WISP-1 to bFGF-loaded pellets
completely suppressed experimentally induced NV. Topically
administered bevacizumab (Avastin; Genentech, San Francisco,
CA), reduces NV by 40%.27The effect of addition of bevacizumab
to bFGF pellets in this model is not known. WISP-1 was also
potent in suppressing angiogenesis in the laser-induced CNV
model. In this murine model, the mouse WISP-1 peptide se-
quence reduced the mean area of CNV by 43%. By way of
comparison, use of a mouse monoclonal antibody against VEGF
in this model results in reduction of neovascularization by
69%.28Intravitreous administration of the human instead of
mouse WISP-1 peptide sequence in the mouse model of CNV
resulted in a smaller (28%) reduction in CNV area that did not
reach statistical significance in this study (P ? 0.059). This
interspecies differential effect potentially indicates an incom-
plete cross-reactivity between human and mouse forms of the
ligand in vivo. This result also demonstrates the critical impor-
tance of peptide sequence to bioactivity.
Identification of novel endogenous antiangiogenic peptides
that can play a role in both physiological and pathologic con-
ditions, such as the peptide described herein, have the poten-
tial to increase our understanding of neovascularization in
health and disease. They can also serve as a basis for novel,
potent, and potentially synergistic therapeutics for various oc-
ular neovascularization-related diseases.
1. Dorrell M, Uusitalo-Jarvinen H, Aguilar E, Friedlander M. Ocular
neovascularization: basic mechanisms and therapeutic advances.
Surv Ophthalmol. 2007;52(suppl 1):S3–S19.
2. Campochiaro PA. Molecular targets for retinal vascular diseases.
J Cell Physiol. 2007;210:575–581.
3. Zhang SX, Ma JX. Ocular neovascularization: Implication of endog-
enous angiogenic inhibitors and potential therapy. Prog Retin Eye
4. Araki T, Taniwaki T, Becerra SP, Chader GJ, Schwartz JP. Pigment
epithelium-derived factor (PEDF) differentially protects immature
but not mature cerebellar granule cells against apoptotic cell
death. J Neurosci Res. 1998;53:7–15.
5. Tombran-Tink J, Johnson LV. Neuronal differentiation of retino-
blastoma cells induced by medium conditioned by human RPE
cells. Invest Ophthalmol Vis Sci. 1989;30:1700–1707.
6. Dawson DW, Volpert OV, Gillis P, et al. Pigment epithelium-
derived factor: a potent inhibitor of angiogenesis. Science. 1999;
7. Hatcher HC, Ma JX, Chao J, Chao L, Ottlecz A. Kallikrein-binding
protein levels are reduced in the retinas of streptozotocin-induced
diabetic rats. Invest Ophthalmol Vis Sci. 1997;38:658–664.
8. Ma JX, King LP, Yang Z, Crouch RK, Chao L, Chao J. Kallistatin in
human ocular tissues: reduced levels in vitreous fluids from pa-
tients with diabetic retinopathy. Curr Eye Res. 1996;15:1117–
9. Gao G, Shao C, Zhang SX, Dudley A, Fant J, Ma JX. Kallikrein-
binding protein inhibits retinal neovascularization and decreases
vascular leakage. Diabetologia. 2003;46:689–698.
10. Spranger J, Hammes HP, Preissner KT, Schatz H, Pfeiffer AF.
Release of the angiogenesis inhibitor angiostatin in patients with
proliferative diabetic retinopathy: association with retinal photo-
coagulation. Diabetologia. 2000;43:1404–1407.
11. Good DJ, Polverini PJ, Rastinejad F, et al. A tumor suppressor-
dependent inhibitor of angiogenesis is immunologically and func-
tionally indistinguishable from a fragment of thrombospondin.
Proc Natl Acad Sci U S A. 1990;87:6624–6628.
12. Volpert OV, Stellmach V, Bouck N. The modulation of throm-
bospondin and other naturally occurring inhibitors of angiogenesis
during tumor progression. Breast Cancer Res Treat. 1995;36:119–
13. Chan CK, Pham LN, Chinn C, et al. Mouse strain-dependent het-
erogeneity of resting limbal vasculature. Invest Ophthalmol Vis
14. Karagiannis ED, Popel AS. Peptides derived from type I throm-
bospondin repeat-containing proteins of the CCN family inhibit
proliferation and migration of endothelial cells. Int J Biochem Cell
15. Karagiannis ED, Popel AS. Anti-angiogenic peptides identified in
thrombospondin type I domains. Biochem Biophys Res Commun.
16. Karagiannis ED, Popel AS. A systematic methodology for pro-
teome-wide identification of peptides inhibiting the proliferation
and migration of endothelial cells. Proc Natl Acad Sci U S A.
17. Wilkins MR, Gasteiger E, Bairoch A, et al. Protein identification and
analysis tools in the ExPASy server. Methods Mol Biol. 1999;112:
18. Kenyon BM, Voest EE, Chen CC, Flynn E, Folkman J, D’Amato RJ.
A model of angiogenesis in the mouse cornea. Invest Ophthalmol
Vis Sci. 1996;37:1625–1632.
19. Kenyon BM, Browne F, D’Amato RJ. Effects of thalidomide and
related metabolites in a mouse corneal model of neovasculariza-
tion. Exp Eye Res. 1997;64:971–978.
20. Edelman JL, Castro MR. Quantitative image analysis of laser-in-
duced choroidal neovascularization in rat. Exp Eye Res. 2000;71:
21. Tobe T, Ortega S, Luna JD, et al. Targeted disruption of the FGF2
gene does not prevent choroidal neovascularization in a murine
model. Am J Pathol. 1998;153:1641–1646.
22. Sheibani N, Sorenson CM, Cornelius LA, Frazier WA. Throm-
bospondin-1, a natural inhibitor of angiogenesis, is present in
vitreous and aqueous humor and is modulated by hyperglycemia.
Biochem Biophys Res Commun. 2000;267:257–261.
23. Uno K, Kuroki M, Hayashi H, Uchida H, Kuroki M, Oshima K.
Impairment of thrombospondin-1 expression during epithelial
wound healing in corneas of vitamin A-deficient mice. Histol
24. Nguyen NQ, Tabruyn SP, Lins L, et al. Prolactin/growth hormone-
derived antiangiogenic peptides highlight a potential role of tilted
peptides in angiogenesis. Proc Natl Acad Sci U S A. 2006;103:
25. Hiscott P, Paraoan L, Choudhary A, Ordonez JL, Al-Khaier A,
Armstrong DJ. Thrombospondin 1, thrombospondin 2 and the eye.
Prog Retin Eye Res. 2006;25:1–18.
26. Simantov R, Febbraio M, Crombie R, Asch AS, Nachman RL, Silver-
stein RL. Histidine-rich glycoprotein inhibits the antiangiogenic
effect of thrombospondin-1. J Clin Invest. 2001;107:45–52.
27. Manzano RP, Peyman GA, Khan P, et al. Inhibition of experimental
corneal neovascularisation by bevacizumab (Avastin). Br J Oph-
28. Campa C, Kasman I, Ye W, Lee WP, Fuh G, Ferrara N. Effects of an
anti-VEGF-A monoclonal antibody on laser-induced choroidal neo-
vascularization in mice: optimizing methods to quantify vascular
changes. Invest Ophthalmol Vis Sci. 2008;49:1178–1183.
IOVS, August 2009, Vol. 50, No. 8
WISP-1 Peptide Inhibits Ocular Neovascularization3845