| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
Regular Articles |
From the Departments of Ophthalmology and Neuroscience,*
The Johns Hopkins University School of Medicine, Baltimore; and Genetic
Therapy,
A Novartis Company,
Gaithersburg, Maryland
 |
Abstract |
|---|
 Top Abstract IntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
2a blocks tumor
angiogenesis and causes regression of hemangiomas, but has no
effect on choroidal neovascularization (CNV). Therefore,
inhibitors of tumor angiogenesis do not necessarily inhibit ocular
neovascularization. In this study, we used an intravenous
injection of adenoviral vectors containing a sig-mEndo
transgene consisting of murine immunoglobulin 
-chain leader sequence
coupled to sequence coding for murine endostatin to investigate the
effect of high serum levels of endostatin on CNV in mice. Mice injected
with a construct in which sig-mEndo expression was
driven by the Rous sarcoma virus promoter had moderately high serum
levels of endostatin and significantly smaller CNV lesions at sites of
laser-induced rupture of Bruch’s membrane than mice injected with null
vector. Mice injected with a construct in which
sig-mEndo was driven by the simian cytomegalovirus
promoter had 
10-fold higher endostatin serum levels and had nearly
complete prevention of CNV. There was a strong inverse correlation
between endostatin serum level and area of CNV. This study provides
proof of principle that gene therapy to increase levels of endostatin
can prevent the development of CNV and may provide a new treatment for
the leading cause of severe loss of vision in patients with age-related
macular degeneration.
|
Introduction |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
The potential for tissue-specific features makes it hazardous to
predict how a protein or drug will affect neovascularization in one
disease process based on its effects in another. Transforming growth
factor-ß and thrombospondin inhibit neovascularization in some
settings and stimulate it in others.2-5
The retina and
choroid are highly specialized tissues with vasculatures that have
unique features making it particularly difficult to predict their
response to purported stimulators and inhibitors. Tissue inhibitor of
metalloproteinase-1 has been touted as an inhibitor of
neovascularization,6
but it stimulates vascular
endothelial growth factor-induced neovascularization in the
retina.7
Interferon 2a causes dramatic involution of
hemangiomas8
and inhibits iris neovascularization in a
model of ischemic retinopathy,9
which led to the
prediction that it would inhibit choroidal neovascularization (CNV).
However, a multicenter, randomized, placebo-controlled trial
demonstrated that patients with CNV who received interferon 2a do
not have any involution of CNV and end up with worse vision than those
treated with placebo.10
Therefore, testing in relevant
animal models is necessary to predict the effect of proteins or drugs
on ocular neovascularization.
Endostatin is a cleavage product of collagen XVIII that inhibits tumor
angiogenesis resulting in inhibition of tumor growth.11
Gene transfer provides a strategy to achieve sustained release of
endostatin and can circumvent difficulties arising from handling the
protein. Adenoviral vectors transduce the liver efficiently after
systemic administration, and consequently they can produce high serum
levels of encoded secreted proteins. In this study we performed
intravascular injections of adenoviral vectors containing a transgene
consisting of murine immunoglobulin (Ig) chain signal sequence
coupled to sequence coding for murine endostatin. This allowed us to
determine the effect of high serum levels of endostatin in a model of
CNV.
|
Materials and Methods |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
The adenoviral vectors used in the experiments described
here have the following genes deleted: E1A, E1B, E2, and E3. The
expression cassettes have been placed in the position of the E1A gene.
Construction of Av3mEndo, in which the Rous sarcoma virus promoter
drives expression of murine endostatin fused to the IgG signal
peptide, has been described.12
Av3CsmEndo is identical to
Av3mEndo except that the simian cytomegalovirus promoter was
substituted for the Rous sarcoma virus promoter. Av3Null and Av3CsNull
are identical to Av3mEndo and Av3CsmEndo, respectively, but lack the
cDNA inserts.
The Av3mEndo, Av3Null, Av3CsmEndo, and Av3CsNull vectors were reamplified in S8 cells with 0.3 µmol/L dexamethasone in Richter’s conditioned medium containing 5% fetal bovine serum until cytopathic effect was observed. The adenoviral vector titer (particles per ml) and biological titer (plaque-forming units per ml) were determined as previously described.13 The correct genome structures of each of the purified vectors were confirmed by restriction digests and Southern blot analysis. The Av3mEndo and Av3CsmEndo were confirmed to be negative for replication-competent adenovirus by the Gene Therapy Core Technologies Molecular Core Laboratory at Genetic Therapy, Inc.
Gene Transfer to Mice and Induction of CNV
Viral vectors were injected into the tail vein of adult C57BL/6 mice. Mice were injected with 2 x 1011 particles of either Av3mEndo (n = 18) or Av3Null (n = 17) or with 6 x 1010 particles of either Av3CsmEndo or Av3CsNull. Four days after viral vector injection, the mice were anesthetized with ketamine hydrochloride (100 mg/kg body weight), pupils were dilated with 1% tropicamide, and krypton laser photocoagulation was used to rupture Bruch’s membrane at three locations in each eye of each mouse as previously described.14 Briefly, krypton laser photocoagulation (100 µm spot size, 0.1 seconds duration, 120 mW) was delivered using the slit-lamp delivery system of a Coherent model 920 photocoagulator and a hand-held cover slide as a contact lens. Burns were performed in the 9, 12, and 3 o’clock positions 2 to 3 disk diameters from the optic nerve. Production of a vaporization bubble at the time of laser, which indicates rupture of Bruch’s membrane, is an important factor in obtaining CNV,14 so only burns in which a bubble was produced were included in the study. A bubble was not produced for one burn in mice injected with Av3mEndo and three burns in mice injected with Av3Null. The cornea of one eye of a mouse that had been injected with Av3mEndo had a corneal scar that prevented visualization for laser and that eye could not be used.
Measurement of the Size of Laser-Induced CNV Lesions
Two weeks after laser treatment, the size of CNV lesions was evaluated by one of two different techniques, measurement of the integrated area of CNV on serial sections as previously reported15 or measurement of the area of CNV in choroidal flat mounts.16 For mice injected with Av3mEndo, 10 mice were evaluated by the flat mount technique and 8 by serial sections, and for mice injected with Av3Null, 10 mice were evaluated by the flat mount technique and 7 by serial sections.
Mice used for the flat-mount technique were anesthetized and perfused with 1 ml of phosphate-buffered saline containing 50 mg/ml of fluorescein-labeled dextran (2 x 106 average molecular weight; Sigma, St. Louis, MO) as previously described.17 The eyes were removed and fixed for 1 hour in 10% phosphate-buffered formalin. The cornea and lens were removed and the entire retina was carefully dissected from the eyecup. Radial cuts (4 to 7, average 5) were made from the edge of the eyecup to the equator and the eyecup was flat-mounted in Aquamount with the sclera facing down and the choroid facing up. Flat mounts were examined by fluorescence microscopy and images were digitized using a 3-charge-coupled device color video camera and a frame grabber. Image-Pro Plus was used to measure the total area of hyperfluorescence associated with each burn, corresponding to the total fibrovascular scar. For mice injected with Av3mEndo, a total of 19 eyes were evaluated (one eye had a pre-existent corneal scar that precluded laser treatment) and there was one burn that had not been associated with a bubble, so that 56 lesions were measured. For mice injected with Av3Null, a total of 20 eyes were evaluated and because there were three burns that had not been associated with a bubble, 57 lesions were measured. The areas within each eye were averaged, and after log transformation, regression analysis with generalized estimating equations was performed using SAS software (SAS Institute Inc., Cary, NC) to compare average areas in eyes of mice receiving Av3mEndo versus Av3Null. This analysis adjusts for correlation between right and left eyes of each mouse.
Mice used to measure the integrated area of CNV on serial sections were sacrificed 2 weeks after laser treatment and eyes were rapidly removed and frozen in optimum-cutting temperature embedding compound (OCT; Miles Diagnostics, Elkhart, IN). Frozen serial sections (10 µm) were cut through the entire extent of each burn and histochemically stained with biotinylated Griffonia simplicifolia lectin B4 (GSA; Vector Laboratories, Burlingame, CA) that selectively binds to vascular cells. Slides were incubated in methanol/H2O2 for 10 minutes at 4°C, washed with 0.05 mol/L Tris-buffered saline, pH 7.6 (TBS), and incubated for 30 minutes in 10% normal porcine serum. Slides were incubated 2 hours at room temperature with biotinylated GSA and after rinsing with 0.05 mol/L TBS, they were incubated with avidin coupled to alkaline phosphatase (Vector Laboratories) for 45 minutes at room temperature. After being washed for 10 minutes with 0.05 mol/L TBS, slides were incubated with Histomark Red (Kirkegaard and Perry) to give a red reaction product that is distinguishable from melanin. Some slides were counterstained with Contrast Blue (Kirkegaard and Perry).
To perform quantitative assessments, GSA-stained sections were examined with an Axioskop microscope and images were digitized using a 3-charge-coupled device color video camera and a frame grabber. Image-Pro Plus software was used to delineate and measure the area of GSA-stained blood vessels in the subretinal space. For each lesion, area measurements were made for all sections on which some of the lesion appeared and added together to give the integrated area measurement. Measurements were analyzed as described above for area measurements on flat mounts.
Comparison of Mice Injected with Av3mEndo to Those Injected with Av3CsmEndo
Mice were injected in the tail vein with 2 x 1011 particles of Av3mEndo (n = 10) or Av3Null (n = 9), or they were injected with 6 x 1010 particles of Av3CsmEndo (n = 11) or Av3CsNull (n = 11). A no injection control group (n = 11) was also included. Four days after injection, Bruch’s membrane was ruptured with laser in three places in each eye of each mouse as described above. Seven days after injection, blood was drawn from the tail vein of 47 of the 52 mice (despite multiple attempts, serum could not be obtained from five mice) and serum was stored at -80°C for enzyme-linked immunosorbent assays (ELISAs). Eighteen days after injection and 14 days after laser, the area of CNV was assessed on choroidal flat mounts as described above.
ELISAs for Endostatin
Endostatin serum levels were determined with a murine endostatin ELISA kit (Accucyte murine endostatin; CytImmune Sciences, College Park, MD) according to the manufacturer’s instructions.
Measurement of ß-Galactosidase Activity
Livers and eyes were snap-frozen after removal from mice. On the day of the assay, livers or eyes were homogenized in lysis buffer [40:1 v/v 1x reporter lysis buffer (Promega, Madison, WI) and protease inhibitor cocktail (Sigma, St Louis, MO)]. Protein content was determined by Bradford assay (BioRad, Hercules, CA). ß-galactosidase activity was determined using the Galacto-Light system (Tropix, Bedford, MA).
|
Results |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
The construction and characterization of Av3mEndo has been described.12 The supernatant from Av3mEndo-transformed S8 cells contains a 20-kd protein, the expected size of endostatin, that potently inhibits vascular endothelial growth factor-165-induced migration of human umbilical vein endothelial cells. An ELISA assay demonstrated that 106 Av3mEndo-transduced Hep3B cells secrete 1 to 2 µg of murine endostatin per 24 hours.
Intravascular Injection of Av3mEndo Inhibits Development of CNV
In initial experiments, the amount of CNV at sites of
laser-induced rupture of Bruch’s membrane was compared in mice
injected with Av3mEndo and mice injected with Av3Null. The amount of
CNV was assessed by two different techniques; measurement of the area
of CNV perfused by fluorescein-labeled dextran on choroidal flat mounts
and measurement of the area of CNV on serial sections through the
entire lesion, a technique that we have used in previous
studies.15,18
Computerized image analysis performed by
investigators masked with respect to treatment group showed that the
mean area of perfused CNV lesions on flat mounts was significantly less
in mice injected with Av3mEndo compared to Av3Null-injected
controls (Table 1)
. The integrated area
of CNV obtained by adding together the area of CNV on each serial
section, which assesses size in three dimensions, confirmed that there
was significantly less CNV at sites of Bruch’s membrane rupture in
mice injected with Av3mEndo compared to Av3Null injected-mice (Table 1)
. Differences are larger between the treated and untreated groups
using the first technique, possibly because with the second technique,
we measured the entire hyperfluorescent area and in treated eyes,
vessels did not always occupy the entire area of hyperfluorescence
(Figure 1E)
. Also, the second technique
ignores any difference in the thickness of CNV. Therefore, the second
measurement technique underestimates treatment effect, but because both
techniques provide similar information, only choroidal flat mounts were
used in subsequent experiments. Also, in subsequent experiments, some
mice that did not receive any vector injections were treated with laser
to evaluate for a possible null vector effect.
|
|
shows the typical appearance of laser-induced CNV in a
choroidal flat mount of a no-injection control mouse. The CNV had a
similar appearance in mice injected with Av3Null (Figure 1B)
, but
appeared smaller in mice injected with Av3mEndo (Figure 1C)
. Serial
sections stained with GSA show the CNV as red-stained tissue in the
subretinal space. The section through the largest part of a typical CNV
lesion in a mouse injected with Av3Null is shown in Figures 1F
(low
power) and 1H (high power). The distance between the arrows
represents the approximate diameter of the lesion and the arrowheads
along the superior border help to point out the thickness. The section
through the largest part of a typical CNV lesion in a mouse injected
with Av3mEndo is shown in Figure 1G
(low power) and 1I (high power).
Note that the approximate diameter and thickness are less than that in
the typical lesion of a null vector-injected mouse (1F). Inverse Correlation between Endostatin Serum Levels and Area of CNV
Characterization of a second vector construct,
Av3CsmEndo, demonstrated that intravascular injection of
6 x 1010
particles resulted in 10-fold
higher maximal endostatin levels compared to levels in mice injected
with the maximum tolerated dose (2 x 1011
particles) of Av3mEndo (Figure 2A)
. The
palindromic series I repeats in the simian cytomegalovirus promoter
behave as strong basal enhancers and cyclic AMP response elements
boosting expression.19
The kinetics of expression differ
for the two vector constructs, but serum endostatin attains maximal
levels 4 to 7 days after intravenous injection and remains greater than
background (30 to 150 ng/ml, results not shown) for at least 1 month.
|
appeared smaller than those in mice
injected with Av3CsNull (1D). Image analysis confirmed that the area of
CNV lesions was significantly less in mice injected with either
Av3CsmEndo or Av3mEndo compared to controls (Table 2)
. A plot of the mean area of CNV
lesions versus endostatin serum level in each mouse showed a
strong inverse correlation with r = -0.66 (Figure 2B)
.
|
To determine whether systemic administration of adenoviral vectors
results in significant transduction of the eye, a group of mice was
injected with Av3nBg. This vector expresses ß-galactosidase from an
Rous sarcoma virus promoter. After 5 days, the mice were sacrificed and
ß-galactosidase activity was measured in homogenates of the eye and
liver using a chemiluminescence assay. In the livers of mice that
received vector, levels of ß-galactosidase activity were 1000-fold
higher than uninjected controls, whereas in the eye, the levels of this
enzyme activity were similar between vector-injected and control
animals (Figure 3)
. The absence of
detectable ß-galactosidase activity in the eye after administration
of an adenovirus expressing this enzyme suggests that the
anti-angiogenic effect after intravascular injection of endostatin
vectors is because of systemically produced rather than locally
produced endostatin.
|
|
Discussion |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
Intravascular injections of large doses of recombinant endostatin inhibit tumor growth in mice and even cause tumor regression.11 Endostatin-treated tumors are much less vascular, suggesting that inhibition of tumor angiogenesis is the basis for the tumor regression. Gene transfer provides a means to achieve sustained release of endostatin and avoid storage, potential loss of activity, and multiple injections of recombinant protein. Two studies have explored nonviral approaches for endostatin gene transfer in tumor models.22,23 Intramuscular injection of a synthetic polymer containing an expression plasmid encoding a secretable form of mouse endostatin resulted in peak serum levels reported as 8 ng/ml, which is less than the range of endogenous endostatin levels in our experiments. This suggests that the increase in endostatin levels were quite modest, but they resulted in decreased growth of subcutaneous renal cell carcinoma by 40%, and a sixfold decrease in lung metastases.22 Similar results were obtained in nude mice implanted with MDA-MB-435 tumors in which liposomes complexed to plasmids encoding angiostatin or endostatin were injected.23
Compared to nonviral approaches, endostatin gene transfer with recombinant adenoviral vectors results in higher serum levels and more dramatic effects. In one study,24 peak endostatin serum levels were 1740 ng/ml and growth of subcutaneous breast cancer was slowed resulting in a 60% reduction in tumor size. The effect on pre-established Lewis lung carcinoma tumors was investigated by injecting vector containing the murine endostatin gene or control vector 7 days after subcutaneous tumor implantation. There was a 78% reduction in tumor size and complete absence of metastases in the treated group compared to controls, but no tumor regression. In another study,25 peak serum levels of endostatin were 1770 ng/ml resulting in a 40% inhibition of MC38 adenocarcinoma growth. Using the same Av3mEndo vector reported on in the present study, Chen and colleagues12 found that compared to Av3Null-injected mice, which all died from liver metastasis 2 months after implantation of poorly differentiated human stage IV colon carcinoma, Av3mEndo-injected mice survived significantly longer, with 25% still alive at 4 months. Although these results are good, they are not comparable to the dramatic tumor regression achieved with injection of 10 to 20 mg/kg/day of recombinant endostatin.11
Our data show a strong correlation between endostatin serum levels and prevention of CNV after rupture of Bruch’s membrane. To our knowledge, this is the first demonstration of endostatin-induced inhibition of intraocular neovascularization. This is an important finding, because as noted above, there are agents that inhibit tumor angiogenesis and have no effect on CNV. Furthermore, CNV is a major public health problem; it is responsible for the vast majority of severe visual loss in patients with age-related macular degeneration, the most common cause of visual morbidity in patients older than the age of 60 in developed countries.26 The magnitude of this already huge problem is continuing to increase as our population ages. Our current treatments for CNV are inadequate, because they are all directed at ablation of the abnormal vessels and do nothing to address the stimuli for blood vessel growth. As a result, they are all plagued by recurrences. Treatments that antagonize or neutralize the stimuli for blood vessel growth are urgently needed.
Our data suggest that endostatin could potentially provide a means to counter the strong stimuli for neovascularization that are present in the eyes of patients with age-related macular degeneration and related diseases. Because sustained high levels of endostatin are needed for the effect, and efficacy increases with increasing concentration, gene transfer offers advantages as a possible therapeutic approach. In this study, we used systemic gene therapy to provide proof of concept. Because endostatin is an endogenous protein and is not known to have toxic effects, it is conceivable that systemic gene therapy could be used in nonmalignant neovascular diseases such as CNV. However, it is also conceivable that high serum concentrations of endostatin could have deleterious effects, and recently patients with systemic sclerosis have been noted to have elevated serum levels of endostatin.27 Highest levels of endostatin were found in patients with cutaneous scars or ulcers and it was postulated that endostatin could contribute to ischemic complications. Local delivery of vectors to the eye provides an appealing alternative to systemic delivery that deserves investigation.
Adenoviral vectors provide rapid onset of fairly high level expression and therefore are ideal for proof of concept, but do not provide long-term expression, which is disadvantageous for treatment of chronic diseases such as CNV in age-related macular degeneration. Helper-dependent adenoviral vectors provide long-term expression in baboons while maintaining high serum levels of secreted protein,28 or alternatively adeno-associated virus or lentivirus vectors could be considered. Although many details must be worked out, our data suggest that endostatin gene therapy may provide a new approach in patients with age-related macular degeneration who are at high risk for development of CNV.
The implications of our study go beyond using endostatin for prevention of CNV. Experiments are underway to determine whether increasing levels of endostatin after development of CNV results in regression. We have previously demonstrated that vascular endothelial growth factor kinase inhibitors prevent the development of CNV;15,18 it will be important to determine whether combining endostatin gene therapy with vascular endothelial growth factor kinase inhibitors provides synergy with regard to prevention and/or regression of CNV. Also, there are several other proteins and/or protein cleavage products, in addition to endostatin, that are purported endogenous inhibitors of angiogenesis including angiostatin,29 antithrombin III,30 platelet factor-4,31 and pigment epithelium-derived factor32 just to name a few. Additional studies are needed to assess the relative efficacy and safety of these and other proteins alone and in combination, because the present study demonstrates that the general approach of using purported protein inhibitors may have merit for treatment of ocular neovascularization and justifies the enormous amount of work and resources that such studies will require.
|
Footnotes |
|---|
Supported by a grant from the Michael Panitch Fund; a grant from Mrs. Harry J. Duffey; Public Health Service grants EY05951 and EY12609; a core grant P30EY1765 from the National Eye Institute; a training grant from the Juvenile Diabetes Foundation (to P. G.); a Lew R. Wasserman Merit Award (to P. A. C.); an unrestricted grant from Research to Prevent Blindness, Inc.; a grant from CIBA Vision, Inc., a Novartis Company; the Rebecca P. Moon, Charles M. Moon, Jr., and Dr. P. Thomas Manchester Research Fund; and a grant from Dr. and Mrs. William Lake.
P. A. C. is the George S. and Dolores Dore Eccles Professor of Ophthalmology and Neuroscience. Peter A. Campochiaro is a paid consultant for CIBA Vision, Inc. This arrangement is being monitored by the Conflict of Interest Committee of the Johns Hopkins University.
Accepted for publication March 23, 2001.
|
References |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
This article has been cited by other articles:
 |
|
 |
|
  M. A. Economou, J. Wu, D. Vasilcanu, L. Rosengren, C. All-Ericsson, I. van der Ploeg, E. Menu, L. Girnita, M. Axelson, O. Larsson, et al. Inhibition of VEGF Secretion and Experimental Choroidal Neovascularization by Picropodophyllin (PPP), an Inhibitor of the Insulin-like Growth Factor-1 Receptor Invest. Ophthalmol. Vis. Sci., June 1, 2008; 49(6): 2620 - 2626. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 I. A. Bhutto, K. Uno, C. Merges, L. Zhang, D. S. McLeod, and G. A. Lutty Reduction of Endogenous Angiogenesis Inhibitors in Bruch's Membrane of the Submacular Region in Eyes With Age-Related Macular Degeneration Arch Ophthalmol, May 1, 2008; 126(5): 670 - 678. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
K. Kiuchi, M. Matsuoka, J. C. Wu, R. Lima e Silva, M. Kengatharan, M. Verghese, S. Ueno, K. Yokoi, N. H. Khu, J. P. Cooke, et al. Mecamylamine Suppresses Basal and Nicotine-Stimulated Choroidal Neovascularization Invest. Ophthalmol. Vis. Sci., April 1, 2008; 49(4): 1705 - 1711. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
M. Ju, C. Mailhos, J. Bradley, T. Dowie, M. Ganley, G. Cook, P. Calias, N. Lange, A. P. Adamis, D. T. Shima, et al. Simultaneous but Not Prior Inhibition of VEGF165 Enhances the Efficacy of Photodynamic Therapy in Multiple Models of Ocular Neovascularization Invest. Ophthalmol. Vis. Sci., February 1, 2008; 49(2): 662 - 670. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
O. Tatar, K. Shinoda, E. Kaiserling, G. Pertile, C. Eckardt, A. Mohr, E. Yoeruek, P. Szurman, K. U. Bartz-Schmidt, and S. Grisanti Early Effects of Triamcinolone on Vascular Endothelial Growth Factor and Endostatin in Human Choroidal Neovascularization Arch Ophthalmol, February 1, 2008; 126(2): 193 - 199. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 A. G. Marneros, H. She, H. Zambarakji, H. Hashizume, E. J. Connolly, I. Kim, E. S. Gragoudas, J. W. Miller, and B. R. Olsen Endogenous endostatin inhibits choroidal neovascularization FASEB J, December 1, 2007; 21(14): 3809 - 3818. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 O. Tatar, A. Adam, K. Shinoda, T. Eckert, G. B Scharioth, M. Klein, E. Yoeruek, K. U. Bartz-Schmidt, and S. Grisanti Matrix metalloproteinases in human choroidal neovascular membranes excised following verteporfin photodynamic therapy Br. J. Ophthalmol., September 1, 2007; 91(9): 1183 - 1189. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
K. Yokoi, S. Kachi, H. S. Zhang, P. D. Gregory, S. K. Spratt, R. J. Samulski, and P. A. Campochiaro Ocular Gene Transfer with Self-Complementary AAV Vectors Invest. Ophthalmol. Vis. Sci., July 1, 2007; 48(7): 3324 - 3328. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
O. Tatar, K. Shinoda, A. Adam, T. Eckert, C. Eckardt, K. Lucke, C. Deuter, K. U. Bartz-Schmidt, and S. Grisanti Effect of verteporfin photodynamic therapy on endostatin and angiogenesis in human choroidal neovascular membranes Br. J. Ophthalmol., February 1, 2007; 91(2): 166 - 173. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 N. Jo, C. Mailhos, M. Ju, E. Cheung, J. Bradley, K. Nishijima, G. S. Robinson, A. P. Adamis, and D. T. Shima Inhibition of Platelet-Derived Growth Factor B Signaling Enhances the Efficacy of Anti-Vascular Endothelial Growth Factor Therapy in Multiple Models of Ocular Neovascularization Am. J. Pathol., June 1, 2006; 168(6): 2036 - 2053. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
A. G. Marneros and B. R. Olsen Physiological role of collagen XVIII and endostatin FASEB J, May 1, 2005; 19(7): 716 - 728. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
H. Akiyama, T. Tanaka, H. Itakura, H. Kanai, T. Maeno, H. Doi, M. Yamazaki, K. Takahashi, Y. Kimura, S. Kishi, et al. Inhibition of Ocular Angiogenesis by an Adenovirus Carrying the Human von Hippel-Lindau Tumor-Suppressor Gene In Vivo Invest. Ophthalmol. Vis. Sci., May 1, 2004; 45(5): 1289 - 1296. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
I. A. Bhutto, S. Y. Kim, D. S. McLeod, C. Merges, N. Fukai, B. R. Olsen, and G. A. Lutty Localization of Collagen XVIII and the Endostatin Portion of Collagen XVIII in Aged Human Control Eyes and Eyes with Age-Related Macular Degeneration Invest. Ophthalmol. Vis. Sci., May 1, 2004; 45(5): 1544 - 1552. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 M. Tomita, H. Yamada, Y. Adachi, Y. Cui, E. Yamada, A. Higuchi, K. Minamino, Y. Suzuki, M. Matsumura, and S. Ikehara Choroidal Neovascularization Is Provided by Bone Marrow Cells Stem Cells, January 1, 2004; 22(1): 21 - 26. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
J.-M. Rakic, V. Lambert, L. Devy, A. Luttun, P. Carmeliet, C. Claes, L. Nguyen, J.-M. Foidart, A. Noel, and C. Munaut Placental Growth Factor, a Member of the VEGF Family, Contributes to the Development of Choroidal Neovascularization Invest. Ophthalmol. Vis. Sci., July 1, 2003; 44(7): 3186 - 3193. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
H. Funatsu, H. Yamashita, H. Noma, H. Mochizuki, T. Mimura, T. Ikeda, and S. Hori Outcome of Vitreous Surgery and the Balance between Vascular Endothelial Growth Factor and Endostatin Invest. Ophthalmol. Vis. Sci., March 1, 2003; 44(3): 1042 - 1047. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 N. Ortega and Z. Werb New functional roles for non-collagenous domains of basement membrane collagens J. Cell Sci., November 15, 2002; 115(22): 4201 - 4214. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 A. Ando, A. Yang, H. Nambu, and P. A. Campochiaro Blockade of Nitric-Oxide Synthase Reduces Choroidal Neovascularization Mol. Pharmacol., September 1, 2002; 62(3): 539 - 544. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
K. Mori, P. Gehlbach, A. Ando, D. McVey, L. Wei, and P. A. Campochiaro Regression of Ocular Neovascularization in Response to Increased Expression of Pigment Epithelium-Derived Factor Invest. Ophthalmol. Vis. Sci., July 1, 2002; 43(7): 2428 - 2434. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 B. J. Raisler, K. I. Berns, M. B. Grant, D. Beliaev, and W. W. Hauswirth Adeno-associated virus type-2 expression of pigmented epithelium-derived factor or Kringles 1-3 of angiostatin reduce retinal neovascularization PNAS, June 25, 2002; 99(13): 8909 - 8914. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
K. Mori, P. Gehlbach, S. Yamamoto, E. Duh, D. J. Zack, Q. Li, K. I. Berns, B. J. Raisler, W. W. Hauswirth, and P. A. Campochiaro AAV-Mediated Gene Transfer of Pigment Epithelium-Derived Factor Inhibits Choroidal Neovascularization Invest. Ophthalmol. Vis. Sci., June 1, 2002; 43(6): 1994 - 2000. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
K. Mori, P. Gehlbach, A. Ando, K. Wahlin, V. Gunther, D. McVey, L. Wei, and P. A. Campochiaro Intraocular Adenoviral Vector-Mediated Gene Transfer in Proliferative Retinopathies Invest. Ophthalmol. Vis. Sci., May 1, 2002; 43(5): 1610 - 1615. [Abstract] [Full Text] [PDF]
|
|
| |||||||||||||||||||||||||||||||||||||
