Investigating the Effect of Ciliary Body Photodynamic Therapy in a Glaucoma Mouse Model

Abstract

purpose. To investigate the morphologic and functional effects of verteporfin ciliary body photodynamic therapy (PDT) in a murine glaucoma model and normal mouse eyes.

methods. A glaucomatous mouse strain, DBA/2J and a normal control mouse strain (C57BL/6) were used in the study. Verteporfin was injected intravenously at doses of 1.0 (DBA/2J) or 2.0 or 4.0 (C57BL/6) mg/kg. Transscleral irradiation of the ciliary body was performed with light at a wavelength of 689 nm delivered through an optical fiber, with irradiance of 1800 mW/cm² and fluence of 100 J/cm². Laser irradiation was applied for 360° of the corneoscleral limbus in C57BL/6 normal mice and for 180° in DBA/2J mice. Retreatment was performed in C57BL/6 normal mice that had been treated with 2.0 mg/kg of verteporfin at post-PDT day 7. One eye of each animal was treated, and the fellow eye served as the control. The morphologic effect of PDT on the ocular structures was assessed by light and electron microscopy. The IOP was measured using an applanation tonometer with a fiber-optic pressure sensor. Surviving retinal ganglion cells (RGCs) in DBA/2J mice eyes were retrogradely labeled with a neurotracer dye at 12 weeks after PDT.

results. In all groups, almost all ciliary body blood vessels in the treated area were thrombosed 1 day after PDT. In DBA/2J mice, ciliary epithelium and stroma were severely damaged 1 day after PDT. The mean IOP in treated eyes was significantly reduced compared with that in the control eyes in all groups. The reduction of mean IOP in DBA/2J mouse eyes persisted for 7 weeks, although the mean IOP in normal mouse eyes treated with 2 or 4.0 mg/kg verteporfin returned to the level of the fellow control eyes by 7 and 17 days after treatment, respectively. The mean number of RGCs in the DBA/2J treated eyes was significantly higher than in control eyes.

conclusions. Ciliary body PDT resulted in morphologic changes in the ciliary body, significant reduction of IOP, and prevention of ganglion cell loss in a mouse glaucoma model. These results suggest that ciliary body PDT is a more selective cyclodestructive technique with potential clinical application in the treatment of glaucoma.

Glaucoma is the most common cause of blindness and the second leading cause of irreversible blindness among African Americans in the United States. ¹ It is also the leading cause of blindness among U.S. Hispanics. ^{1 2} Lowering intraocular pressure (IOP) using medical or surgical therapy is the mainstay of therapy, designed to control and limit this common disease. However, treating glaucoma remains a difficult task, since most medications have side effects, lose their efficacy, and require patients’ life-long compliance. Surgical methods have a higher risk of complication. Ciliary body destruction by cryotherapy or laser irradiation represents a useful alternative for the management of glaucoma that is resistant to other modes of therapy. ^{3 4 5} However, the current cyclodestructive techniques have a high rate of side effects including loss of vision, hypotony, macular edema, or phthisis bulbi. ^{6 7 8}  

Photodynamic ciliary body destruction has been reported to be a better cyclodestructive technique, with potential clinical application in the treatment of glaucoma. ^{9 10 11 12} Previous studies of transscleral ciliary body photodynamic therapy (PDT) in normal rabbit eyes using the photosensitizer chloraluminum sulfonated phthalocyanine (CASPc) have shown a decrease in IOP for 2 weeks after treatment. ^{12 13} In this study, we used liposomal benzoporphyrin derivative (verteporfin) as a photosensitizer, because it is a clinically approved and well-tolerated PDT drug and is also a suitable drug for ciliary body PDT. Verteporfin is the first light-activated drug approved by regulatory authorities for the treatment of choroidal neovascularization secondary to age-related macular degeneration. ¹⁴ It is a synthetic chlorin-like porphyrin that is activated by low-intensity, nonthermal, laser light at a wavelength of ≈690 nm. ^{15 16} It is prepared as a liposome-based formulation, associates with serum lipoproteins including low-density lipoproteins (LDL), and is thought to be taken up preferentially by cells expressing high levels of LDL receptors, including neovascular endothelial cells and tumor cells. ^{17 18 19} Zheng et al. ²⁰ have investigated the distribution of LDL receptors in rats, specifically using immunohistochemistry to localize gp330 and the LDL receptor-related protein (LRP/α2MR). They reported that ciliary and retinal epithelial cells were the only cells that expressed both LRP/α2MR and gp330, suggesting that ciliary body epithelium would be a target for verteporfin. 

To our knowledge, no studies of ciliary body PDT have been performed in a glaucoma model, and treatment effects in normal eyes have been short in duration. It is important to test PDT in a model with elevated IOP, to determine whether PDT cyclodestruction may be a useful therapy in glaucoma. The strain of DBA/2 mice was established in the 1930s and has a predisposition to development of various tumors, calcific heart lesions, gastritis, duodenal polyps, calcareous pericarditis, and degenerative processes of the myocardium, skeletal muscle, subcutaneous adipose tissue, and blood vessels. ²¹ Abnormalities in the anterior segment of the eye in this strain were first reported in 1986. ²² Shortly thereafter, it was reported that DBA/2NNia mice exhibit peripheral anterior synechiae, iris atrophy, pigment dispersion, depletion of retinal ganglion cells (RGCs), and degeneration of the optic nerve. ²³ In 1998, similar changes were reported in DBA/2J mice, another DBA substrain. ²⁴ Mice of this strain have spontaneous development of essential iris atrophy, pigment dispersion, and glaucomatous changes. The IOP was elevated in most mice by the age of 9 months, followed by ganglion cell loss, optic nerve atrophy, and optic nerve cupping. The mean IOP of eyes of 6-month-old females was significantly higher than that of the eyes of males of similar age, and this sex difference in IOP was still evident at the age of 9 months. As glaucoma developed, the ganglion cell loss and mild cupping of the optic nerve were present in some animals by 11 months and in most of the mice by the age of 22 months. 

In this study, we used a liposomal benzoporphyrin derivative (verteporfin) as a photosensitizer and used female DBA/2J mice aged 8 months as a glaucoma mouse model to investigate the effect of ciliary body PDT on IOP and glaucomatous morphologic changes. We also used a normal mouse strain (C57BL/6) to evaluate the cyclodestructive effect of ciliary body PDT based on morphologic and functional alterations. 

Materials and Methods

Animal Model

All experiments were performed in accordance with the ARVO Statement for the Use of Animals in Ophthalmic and Vision Research using a protocol approved by the Animal Care Committee of the Massachusetts Eye and Ear Infirmary. C57BL/6 female mice aged 3 months (n = 20) were used as a normal control, and DBA/2J female mice aged 8 months (n = 20) were used as a glaucoma mouse model. The number and type of animals for each phase of the study is shown in Table 1 . One eye of each animal was treated, and the fellow eye served as the control. Animals were housed in covered cages, fed with a standard rodent diet ad libitum, and kept in a constant 12-hour light-dark cycle. For general anesthesia, a mixture of ketamine (100 mg/kg) and xylazine (9 mg/kg; both from Phoenix Scientific, Inc., St. Joseph, MO) was administered intraperitoneally. 

Photodynamic Therapy

The photosensitizer, verteporfin (Novartis Ophthalmics, Inc., Duluth, GA) was used and prepared according to the manufacturer’s instructions. Anesthetized mice received the photosensitizer drug by tail vain injection with a 30-gauge needle 5 minutes before laser irradiation. A verteporfin dose of 2.0 or 4.0 mg/kg was used in C57BL/6 normal mice, and 1.0 mg/kg was used in DBA/2J mice (total volume 0.25–0.4 mL on the basis of weight). The laser irradiation to the ciliary body was applied transsclerally using a 600-μm optical fiber at a wavelength of light of 689 nm, with a diode laser (Lumenis, Santa Clara, CA), with irradiance 1800 mW/cm² and fluence of 100 J/cm² (measured at the sclera). Transscleral irradiation of the ciliary body was performed with applications for 360° of the corneoscleral limbus in C57BL/6 normal mice and for 180° in DBA/2J mice. Sixteen and eight spots, respectively, were applied to overlap each other (Table 2) . Each treatment spot took 55 seconds and therefore it took 16 or 8 minutes to complete the laser irradiation. However, in the eyes prepared for lectin labeling and terminal deoxynucleotide transferase-mediated nick end labeling (TUNEL), PDT was performed in C57BL/6 mice to cover 180° of the ciliary body, using the 4.0-mg/kg verteporfin dose to compare the degree of staining between the irradiated and nonirradiated areas. 

Retreatment was performed in C57BL/6 normal mice, which had been treated with 2.0 mg/kg of verteporfin after the IOP measurement at post-PDT day 7, with the same light parameters. 

In this study we used a lower dose of verteporfin and smaller area of irradiation in the DBA/2J mice, because preliminary experiments using 2.0 mg/kg in DBA/2J mice and irradiation for 360° of the corneoscleral limbus resulted in severe damage, with corneal edema, hyphema, and corneal neovascularization in 9 of 10 eyes. After the reduction of the extent of treatment by half, these complications were still seen in two eyes of three. After we reduced both drug dose and extent of treatment by half, these complications did not develop. 

IOP Measurement

The IOP was measured with a previously reported applanation tonometer (Ahamed E, et al. IOVS 2003;44:ARVO E-Abstract 3336) ²⁵ (n = 12 in DBA/2J mice, n = 5 in C57BL/6 mice in each dose). The applanation tonometer employs a fiber-optic pressure sensor that is designed around a Fabry-Pérot interferometer (FPI; FTI-10; FISO Technologies, Inc., Saint-Jean Baptiste City, Quebec, Canada). This sensor is designed to measure the pressure on the surface of the fiber’s tip (550 μm), and it provides high-fidelity performance and in situ pressure measurement through the same minimally invasive procedure that is used to apply a medical device to obtain such measurements as intravenous blood pressure. FPI consists of two mirrors facing each other—the space separating the mirrors being called the cavity length. Light reflected in the FPI is wavelength modulated in exact accordance with the cavity length. FPI gauges convert pressure into cavity length variations. The instrument averages the data after 10 repeated pressure measurements via the tip attached to the mouse corneal surface. To calibrate the FOP-MIV fiber-optic pressure sensor, we used a pressure monitor (CyQ 103/301 BPM01; CyberSense Inc.). We connected the device to the mouse’s anterior chamber to monitor pressure and sequentially changed the IOP and measured the corneal surface pressure using the FOP-MIV fiber optic pressure sensor. We used the data to develop a calibration curve and found a linear relationship (R ² = 0.9895). IOP was measured in the morning, taking account of circadian rhythm with the mice under general anesthesia. Standardization of the plane of anesthesia was obtained as previously described, ²⁶ and anesthesia was administered to one animal at a time. After the animal lost consciousness and failed to respond to touch, IOP was measured as soon as possible (typically, within 2–3 minutes), with the placing the animal on a platform. The pressure sensor applanated a central area of mouse cornea with visualization by a dissecting microscope, the IOP was measured by the device 10 times automatically. The IOP was measured every 2 days during the first week, then twice a week until 4 weeks after PDT and then once a week until 8 weeks after PDT. 

Histology

Eyes for light and transmission electron microscopic examination were enucleated at 6 hours (DBA/2J only), 1 and 7 days after treatment and at the end of IOP follow-up (8 weeks after treatment; n = 2 in each group at each time point). The eyes were bisected behind the limbus and placed in modified Karnovsky’s fixative at 4°C overnight and then transferred to 0.1 M cacodylate buffer at pH 7.4. Ciliary body specimens in normal mouse eyes in a 4.0-mg/kg dose group were observed before postfixation by stereomicroscopy at 1 day after PDT. Tissues were postfixed in aqueous 2% osmium tetroxide, stained en bloc with uranyl acetate, dehydrated in graded ethanols, and embedded in Epon. One-micrometer-thick sections were stained with 0.5% toluidine blue in borate buffer for light microscopy and examined with a photomicroscope (Leica Microsystems, Wetzlar, Germany). Thin sections were stained with aqueous uranyl acetate and Sato’s lead stain, and examined with a transmission electron microscope (CM 10; Phillips, Eindhoven, The Netherlands). 

Lectin Labeling of Vascular Endothelial Cells and TUNEL Staining

Lectin labeling of vascular endothelial cells and TUNEL staining was performed 1 day after PDT in C57BL/6 normal mouse eyes (n = 2). In DBA/2J mice, only TUNEL staining was performed (n = 2). PDT was performed to cover 180° of the ciliary body using 4.0 (C57BL/6) and 1.0 (DBA/2J) mg/kg verteporfin to compare the degree of staining between the irradiated and nonirradiated areas. Non–surgically treated fellow eyes (n = 2) were used as the control. Anesthetized animals were perfused with 8 mL phosphate-buffered saline (PBS) administered via a catheter in the left ventricle. After PBS perfusion, fluorescein-isothiocyanate (FITC)–coupled concanavalin A lectin (20 μg/mL in PBS [pH 7.4] 5 mg/kg BW; Vector Laboratories, Burlingame, CA) was perfused to stain vascular endothelial cells. Paraformaldehyde (4%) in 0.1 PBS (pH 7.4) was perfused after PBS, to remove residual unbound lectin. Enucleated eyes were fixed in 4% paraformaldehyde in 0.1 PBS overnight and were immersed in graded sucrose in 0.1 PBS. The samples were embedded in optimal cutting temperature (OCT) compound (Tissue-Tek, Tokyo, Japan) and sectioned vertically at 10 μm. TUNEL staining was performed according to the protocol of the manufacturer (Fluorescein In Situ Apoptosis Detection Kit; Chemicon International, Temecula, CA) to detect retinal cell death induced by PDT. Sections were counterstained with 1 μg/mL DAPI (Sigma-Aldrich, St. Louis, MO). 

RGC Count in DBA/2J Mice

Four weeks after the end of IOP follow-up, RGCs were counted in the DBA/2J mouse eyes (11 months old), as previously reported, ^{27 28} with a slight modification (n = 4). In mice under anesthesia, the skin over the cranium was incised and the scalp exposed. Holes approximately 2 mm in diameter were drilled in the skull 4 mm posterior to the bregma and 1 mm lateral to the midline on both sides of the midline raphe. A neurotracer dye (4% solution in saline; Fluorogold; Fluorochrome, Englewood, CO) was directly applied (1 μL, at a rate of 0.5 μL/min) at a depth of 2 mm from the brain’s surface through a syringe (Hamilton, Reno, NV). Skull openings were then sealed with antibiotic ointment. The overlying skin was sutured and antibiotic ointment applied externally. 

Seven days after the application of the fluorescent tracer, the eyes were enucleated and the retinas were dissected, fixed in 4% paraformaldehyde (PFA) and flatmounted on glass slides. Cell counting was performed as previously described ²⁹ under a fluorescence microscope (Leica Microsystems) using a UV filter set. RGC densities were determined by counting the tracer-labeled RGCs in 12 distinct areas of 9.0 × 10⁻² mm² each (three areas per retinal quadrant at one sixth, one half, and five sixths of the retinal radius). The density of fluorochrome-labeled RGCs was defined as the average number of cells in the 12 fields. Cell counting was performed in a masked fashion. 

Statistical Analysis

All data are presented as the mean ± SD. A paired group of two were compared by paired t-test. To compare three groups, data were compared by ANOVA, with post hoc comparisons tested using the Bonferroni procedure. P < 0.05 was considered to be statistically significant. 

Results

Histology

Stereomicroscopic Findings.

Ciliary processes appeared pale and swollen compared with the control eyes in C57BL/6 normal mice 1 day after PDT with the 4.0-mg/kg dose (Fig. 1) . 

Light and Electron Microscopic Findings.

All ciliary body blood vessels in C57BL/6 normal mouse eyes were open in control eyes (Fig. 2A) . Electron microscopy showed patent capillaries surrounded by pigmented and nonpigmented epithelium (Fig. 2B) . The basal plasma membrane of the pigmented epithelium had marked infolding, indicating that the cells were actively involved in ion transport (Fig. 2C) . One day after PDT, the ciliary processes in normal mouse eyes treated with 2.0 mg/kg verteporfin were enlarged due to edema, compared with the control (Figs. 2A 2D) . Most ciliary body blood vessels appeared thrombosed (Fig 2D) . Iris changes or anterior chamber reactions were not seen in C57BL/6 mice eyes. The same findings were observed in normal mouse eyes treated with 4.0 mg/kg verteporfin. The significant morphologic changes in the ciliary body included vascular injury and thrombosis, with vacuolization of endothelial cells and extravasation of leukocyte and erythrocytes. The basal infoldings were abnormally separated due to edema (Fig. 2E) . The basal processes of the pigmented epithelium around the vessels were distended. However, cell nuclei and mitochondria in the epithelial layers appeared unchanged (Fig. 2F) . Seven days after PDT, ciliary body blood vessels were recanalized in C57BL/6 normal mouse eyes treated with 2.0 or 4.0 mg/kg verteporfin (Fig. 2G) . The size of the ciliary processes appeared normal compared with the control (Fig. 2G) . All capillaries were normally organized. Layering and organization of the ciliary processes were normal (Fig. 2H) . Basal processes of the pigment epithelium were again abundant, though they appeared more compact. Cytoplasmic and nuclear morphology were normal at 7 days after PDT (Fig. 2I) . There was no significant morphologic difference between the 2.0- and 4.0-mg/kg dye doses in C57BL/6 mice at 1 and 7 days after PDT. 

In the treated DBA/2J mice, most ciliary body blood vessels were thrombosed, and large spaces containing exudative material were seen in the epithelial layer at 6 hours after PDT (Fig. 3A) . Electron microscopy showed that ciliary body vessels were thrombosed with erythrocytes and platelets. Endothelia of blood vessels were damaged and erythrocytes and fluid were leaking from the vessels. The basal infoldings were abnormally separated by edema. Most epithelial cells were edematous, and some epithelial cells were disrupted. Organelle dissolution and rupture of the plasma membrane were seen (Fig 3B) . One day after PDT, most ciliary body blood vessels were still thrombosed, and the ciliary epithelium and stroma were severely damaged and edematous. Large intercellular spaces were occupied by exudates. Electron microscopy showed that basal infoldings were stretched and swollen with exudate. Epithelial edema and abnormal intercellular fluid were noted (Figs. 3C 3D) . At 7 days after PDT, the edema was decreasing in the epithelial layers. Some of the ciliary body blood vessels were still thrombosed, and some ciliary epithelia still showed edematous swelling. Electron microscopic findings showed that the basal infoldings were still separated due to edema. Some ciliary blood vessels were recanalized and endothelia were normally organized. Possible regenerative epithelia were seen, and vacuolations were still visible in the epithelial layer (Fig. 3F) . At 8 weeks after PDT, ciliary body blood vessels were recanalized, and edema had disappeared. Some ciliary processes were small and flattened. However, it was difficult to tell the difference between the PDT-treated and the untreated ciliary body, because even some of the untreated ciliary processes showed atrophic changes at this age. We could not detect any abnormality, including retinal detachment, in the peripheral retina in the PDT-treated area. 

Lectin Labeling of Vascular Endothelial Cells and TUNEL Staining

A section through the untreated and treated areas of the ciliary body in C57BL/6 normal mice 1 day after PDT with 4.0 mg/kg verteporfin are shown in Figures 4A and 4B , respectively. The vessels in the ciliary body and sclera on the untreated side showed abundant staining by lectin (green fluorescence), and no TUNEL-positive cells (red fluorescence) were observed (Fig. 4A) . In contrast, there were some TUNEL-positive cells in the ciliary body and retina on the PDT-treated side. The vessels in the ciliary body on the treated side showed very few cells stained by lectin (green fluorescence) because of thrombosis (Fig. 4B) . In DBA/2J mouse eyes, only TUNEL staining was performed 1 day after PDT. Many TUNEL-positive cells were seen in the ciliary processes in both the treated and untreated regions of the treated eye, whereas no TUNEL-positive cells were noted in the nontreated eye (Figs. 4C 4D 4E) . During the laser irradiation, we found that the eyes of the DBA/2J mice were readily transilluminated compared with the C57BL/6, probably due to loss of pigment caused by pigment dispersion. In other words, in DBA/2J mice, the retina-choroid on the side opposite the laser received light irradiation, probably due to these pigment abnormalities. A few TUNEL-positive cells were seen in the peripheral retina in treated region, but there are no TUNEL-positive cells in the retina and choroid in the untreated region. TUNEL-positive cells in the untreated side of the treated eye were located only within the ciliary processes, suggesting that laser light reached the opposite untreated region. 

IOP Measurement

In normal mouse eyes treated with 2.0 mg/kg verteporfin, the mean baseline IOP was 13.2 ± 2.4 mm Hg in the left eye and 13.6 ± 2.1 mm Hg in the right eye (Fig. 5A) . There was no significant difference between the baseline measurements in control and treated eyes. The mean IOP in treated eyes was significantly reduced at 1 and 3 days after treatment compared with the control eyes (by 35.1%, P < 0.05, and by 41.3%, P < 0.01, respectively). The mean IOP in the 2.0-mg/kg verteporfin group returned to the level of the fellow control eyes by 7 days after treatment. However, retreatment at 7 days after the first PDT significantly reduced the IOP again. The mean IOP in treated eyes was 8.7 ± 2.6 and 9.5 ± 2.1 mm Hg at 10 and 14 days after the first PDT, respectively. The mean IOP returned to the level in the fellow control eyes by 17 days after PDT. The mean IOP in treated eyes was significantly lower than the level in fellow control eyes at 28 and 42 days after PDT. 

In normal mouse eyes treated with 4.0 mg/kg verteporfin, the mean baseline IOP was 14.8 ± 1.4 mm Hg in the left eye and 14.2 ± 0.9 mm Hg in the right eye (Fig. 5B) . There was no significant difference between the baseline measurements in control and treated eyes. The mean IOP of treated eyes was significantly reduced (9.2 ± 2.4 mm Hg, P < 0.05) at 1 day after treatment and reached a minimum (7.2 ± 1.9 mm Hg, P < 0.01) at 5 days after treatment compared with the control eyes, by 42.1% and 47.3%, respectively. Significant reduction of mean IOP lasted for 14 days and returned to the level of the fellow control eyes by 17 days after treatment. 

In DBA/2J glaucomatous mice, the mean baseline IOP was 18.8 ± 1.9 mm Hg in the left eye and 18.8 ± 4.3 mm Hg in the right eye at 8 months of age (Fig. 5C) . The mean IOP 1 day after treatment was 18.2 ± 6.5 and 15.9 ± 5.7 in fellow control eyes and treated eyes, respectively, and there was no significant difference between the two groups 1 day after PDT. The mean IOP of treated eyes was significantly reduced by 40.8% compared with fellow control eyes from 3 days after PDT (9.5 ± 3.4 mm Hg, P < 0.05). Significant reduction of mean IOP lasted for 7 weeks and returned to the level of the fellow control eyes by 8 weeks after treatment. In control fellow eyes, the mean IOP at 4 (9 months old) and 8 (10 months old) weeks after PDT were 15.6 ± 3.7 and 12.6 ± 4.7 mm Hg, respectively (Fig. 5C 5D) . Although there was no significant difference between the mean IOP at 8 and 9 months of age, the mean IOP at 10 months of age was significantly lower than that at 8 months of age (P < 0.01; Fig. 5D ). 

RGC Count in DBA/2J Mice

RGCs stained by the gold neurotracer in the eyes of DBA/2J mice at 11 months of age are shown in Figure 6 . The number of ganglion cells in DBA/2J control and treated eyes were 2701 ± 798 and 3905 ± 627/mm², respectively. The loss of RGCs over the 3-month course of follow-up was significantly reduced in mice receiving PDT compared with the DBA/2J control (P < 0.05; Fig. 7 ). 

Discussion

The notable morphologic changes after PDT of the ciliary body were vascular injury, thrombosis, and severe edema. Lectin staining confirmed the lack of perfusion of ciliary body blood vessels after PDT. Once verteporfin is activated by light in the presence of oxygen, highly reactive, short-lived singlet oxygen, and reactive oxygen radicals are generated, causing damage to the targeted cells. Verteporfin PDT has been shown to injure endothelial cells, resulting in vessel occlusion. Damaged endothelium is known to release von Willebrand factor, ³⁰ eicosanoids such as thromboxane and prostacyclin through the cyclo-oxygenase pathway, and leukotrienes through the lipoxygenase pathway. ³¹ PDT also causes changes in the cytoskeleton and cell shape, ³² which can lead to the interruption of the tight junctions between the endothelial cells, and causes exposure of parts of subendothelial connective tissue. Von Willebrand factor and other clotting factors such as thromboxane are released at the exposed site, which leads to platelet activation. The activated plates stick to the gaps between the damaged endothelial cells, and further aggregation of platelets starts to occlude the vessel. ³¹ In addition, eicosanoids such as prostacyclin induce increased vascular permeability and leakage. 

In this study ciliary body PDT significantly reduced the IOP in both normal and glaucomatous eyes. In normal mouse eyes, we were able to demonstrate a dose-dependent effect of the drug, with a higher dose resulting in a longer period of IOP reduction in normal C57BL/6 mice. In contrast, both drug doses resulted in similar histologic findings at 1 and 7 days after PDT. We have previously reported that lowering the dye dose appeared to reduce damage to surrounding tissues after PDT in experimental choroidal neovascularization. ¹⁸ Further studies may demonstrate a gradation of histologic effects. To induce a more prolonged IOP reduction, we applied a second treatment in normal mouse eyes using 2.0 mg/kg verteporfin as soon as the effect of the first treatment vanished. The retreatment significantly reduced the IOP, which also lasted 1 week. Although the mean IOP returned to the level of the fellow control eyes within 1 week, there was a suggestion that the hypotensive effect was more intense and prolonged in some of the eyes after the second treatment. Of interest, a significant IOP reduction was observed at 28 and 42 days after PDT. This may be secondary to some variability in the response to treatment. Electron microscopy after PDT in normal mouse eyes showed that basal processes of the pigment epithelium around the vessels were distended, but cell nuclei and mitochondria in both layers appeared unchanged. The findings suggest that transscleral ciliary body PDT may damage the ciliary body more mildly than other cyclodestructive methods, including cryotherapy and laser irradiation, and has the potential for better control of IOP. 

We found that treatment of the glaucomatous eyes in the DBA/2J mice resulted in a more prolonged IOP reduction, lasting 7 weeks, compared with the effect in normal mouse eyes that lasted only 7 to 17 days. The strategy of ciliary body PDT is to injure the ciliary body epithelium, which produces aqueous humor. In DBA/2J mice, we found not only vascular thrombosis but also ciliary body epithelial injury in histologic examination. In other words, we could get effective IOP reduction with less laser power and a smaller area of treatment. One explanation is that pigment may play a protective role against PDT and the more pigmented C57BL/6 mice are more resistant to treatment. Tsilimbaris et al. ^{12 13} reported that the IOP reduction in pigmented rabbits by ciliary body PDT with CASPc lasted a relatively shorter period than that in albino rabbits. Because DBA/2J mice have pigment dispersion, ²⁴ there may be a decrease in pigment in the ciliary body. Experimentally, we found that the eyes of the DBA/2J mice were readily transilluminated during treatment compared with those in the C57BL/6 mice, which have a more pigmented eye. In light and electron microscopic findings, ciliary body blood vessel occlusions were visible, even in the untreated side of treated eyes (data not shown). TUNEL staining also suggested that the untreated side had been irradiated and damaged to some extent. Although pigmentation augments-laser induced cyclodestruction, ³³ in PDT, it can act as a protective barrier for the target tissue, preventing its exposure to light during photosensitizer circulation. ¹³ There are some reports of the relationship between intensity or extent of cyclophotocoagulation and IOP reduction in the rabbit eye. ^{34 35 36} van der Zypen et al. ³⁵ reported that the eye with normal perfusion pressure requires extensive vascular irradiation damage, and more than 50% of the total secretory surface must be destroyed before effective and long-term reduction in IOP is attained. In our preliminary experiments, we irradiated 180° of the ciliary body in normal mouse eyes, using 2.0 mg/kg verteporfin, but we did not get a significant IOP reduction. In contrast PDT with 1.0 mg/kg verteporfin and treatment of 180° resulted in a significant reduction of IOP in the DBA/2J mice. By the age of 11 to 15 months, the ciliary processes in DBA/2J mice are usually fewer, shorter, and narrower than those in young DBA/2J mice. ²⁴ The ciliary body epithelium of DBA/2J mice at age 8 to 10 months may have already started to decrease in size and number and therefore may not be able to compensate for the injury induced by PDT. 

In this study, we used a lower dose of verteporfin and smaller area of irradiation in the DBA/2J mice. Preliminary experiments using 2.0 mg/mL of verteporfin in DBA/2J mice and irradiation for 360° of the corneoscleral limbus resulted in severe damage including corneal edema, hyphema, and corneal neovascularization. As mentioned, one reason for this increased sensitivity to PDT may be the loss of pigmentation that augments the destructive effect of PDT. Another reason may be the breakdown of the blood–ocular barrier in DBA/2J mice. Mo et al. ³⁷ reported that DBA/2J mice experienced breakdown of the blood–ocular barrier from 4 months of age and an inflammatory response was sustained for several months. They also reported that fluorescein leakage was observed in all eyes by 7 months of age, entering the anterior chamber from behind the iris and from the iris stroma. These findings suggest that DBA/2J mice may be susceptible to inflammation and thus to more pronounced damage from PDT. 

DBA/2J mouse eyes exhibit a progressive form of secondary angle-closure glaucoma that leads to ganglion cell loss in a time-dependent fashion. ²⁴ Using retrogradely labeled RGCs, Schuettauf et al. ³⁸ reported that significant age-dependent loss of RGCs is observed in 6- and 9-month-old DBA/2J mice and time-dependent loss of RGCs can be inhibited by conventional anti-glaucomatous therapy with β-blockers at 9 months of age. In contrast, using cell counts of histologic sections, John et al. ²⁴ reported that the number of surviving RGCs is normal by 8 to 9 months of age and in 5 of 16 eyes decreased by 33% to 40% by 11 to 15 months of age. In our study the age-dependent loss of RGCs between 8 and 11 months of age was significantly reduced in mice receiving PDT compared with the DBA/2J control. In this experiment we used only female DBA/2J mice, which are known to have higher IOPs than those of the male at 6 to 9 months of age. This selection may have increased the likelihood of finding a significant difference in treated and untreated eyes. 

We detected a few, TUNEL-positive cells in the peripheral retina of the direct laser-irradiated area. Because the size of the laser probe was slightly large for the mouse eye, irradiation of some peripheral retina was unavoidable. However, histologic analysis after 2 months of follow-up showed no abnormality in the peripheral retina. These data suggest that the retinal PDT injury was not severe. Further development of the probe for treatment of the ciliary body, will reduce retinal damage to a minimum. In this study we used a 600-μm optical fiber and irradiated 8 or 16 spots during 8 to 16 minutes. In future we will examine ciliary body PDT in a bigger eye or in humans. As the length of human ciliary process is approximately 2000 μm, we will have to use a larger fiber in human treatment. Thus, the number of laser spots in human may be almost the same as in the mouse. Because pigment can act as a protective barrier for the target tissue in PDT, we have to give careful consideration to that point in future human experiments. Further considerations on not only laser parameters but also probe size are needed for clinical application. Additional studies comparing PDT and transscleral cyclophotocoagulation with diode or Nd:YAG laser would be of interest. 

The most significant advantage of ciliary body PDT compared with conventional laser cyclodestruction techniques is the less-destructive nature of the lesions induced by PDT. Moreover, we were able to achieve significant and more prolonged IOP reduction as well as prevention of ganglion cell loss in glaucomatous DBA/2J mouse eyes in this study. Although further investigations are needed to refine treatment parameters further and the need for and timing of retreatment, our results suggest that ciliary body PDT may lead to a more selective cyclodestructive technique with potential clinical application in the treatment of glaucoma. 

 Supported by the Massachusetts Lions Eye Research Fund and National Eye Institute Grant EY014104.

 Submitted for publication July 23, 2005; revised December 28, 2005; accepted March 31, 2006.

 Disclosure: A. Matsubara, None; T. Nakazawa, None; D. Husain, None; E. Iliaki, None; E. Connolly, None; N.A. Michaud, None; E.S. Gragoudas, None; J.W. Miller, None

 The publication costs of this article were defrayed in part by page charge payment. This article must therefore be marked “advertisement” in accordance with 18 U.S.C. §1734 solely to indicate this fact.

 

Table 1.

 

The Number and Type of Animals for Each Phase of the Study

Strain	Histology					Lectin and TUNEL (1 day)	IOP Measurement (with 8 Weeks of Follow-up)	Total Animals
		(6 h)	(1 d)	(7 d)
C57BL/6 (2 mg/kg)	0	2	2	0	5 (2 mice for histology on completion of follow-up)	9
C57BL/6 (4 mg/kg)	0	2*	2	2	5 (2 mice for histology on completion of follow-up)	11
DBA/2J	2	2	2	2, †	12 (2 mice for histology and 4 mice for RGC count on completion of follow-up)	20

*  Examined with stereomicroscopy before postfixation.

†  Underwent only TUNEL staining.

Table 2.

 

Treatment Parameters for Each Group

Mice	Verteporfin Dose (mg/kg)	Treated Area (deg)	Spots (n)	Retreatment
C57BL/6	2	360	16	+
C57BL/6	4	360*	16, †	—
DBA/2J	1	180	8	—

*  180° in the eyes prepared for lectin labeling and TUNEL staining.

†  Eight spots in the eyes prepared for lectin labeling and TUNEL staining.

Figure 1.

 

Left: control eye; right: treated eye, showing pale, swollen ciliary processes (arrow) after PDT.

Figure 2.

 

(A–C) Control eye in normal mice. (A) All ciliary body vessels are open (toluidine blue). (B) Patent capillary ( ) surrounded by two layers of epithelium. (C) Basal infoldings of the pigmented epithelium have a normal appearance ( ). (D–F) A treated eye 1 day after PDT in a normal mouse using 2.0 mg/kg verteporfin. (D) Ciliary body vessels were thrombosed (arrows). The ciliary processes were enlarged by edema (toluidine blue). (E) All capillaries were damaged ( ). Endothelial cells were damaged and leukocytes and erythrocytes were leaking from the vessel (arrow). The basal infoldings were abnormally separated due to edema (arrowheads). (F) Erythrocytes and fluid were noted outside of the vessel. Basal processes of the pigment epithelium around the vessels were distended. Cell nuclei and mitochondria in both layers appeared unchanged. (G–I) Treated eye 7 days after PDT in a normal mouse treated with 2.0 mg/kg verteporfin. (G) Ciliary body vessels were recanalized. The size of ciliary processes appeared normal (toluidine blue). (H) All capillaries were normally organized (arrows). Layering and organization were normal. (I) Basal processes of the pigmented epithelium were again abundant, although they appeared more compact ( ). Cytoplasmic and nuclear morphology were normal. PE, pigmented epithelium; NPE, nonpigmented epithelium; L, leukocyte; Er, erythrocyte; V, vessel. Scale bars: (B) 5 μm; (C, F, I) 2 μm; (E, H) 10 μm. Magnification: (A, D, G) ×50.

Figure 3.

 

Representative histologic images of ciliary bodies in the treated DBA/2J mice. (A) Light micrograph of ciliary process 6 hours after PDT (toluidine blue). Ciliary body blood vessels were thrombosed (arrows), and large spaces containing exudative material were seen in the epithelial layer (arrowheads). (B) Electron micrograph of a ciliary process 6 hours after PDT. Ciliary body vessels were thrombosed with erythrocytes and platelets (arrows). Endothelial cells of blood vessels were damaged and erythrocytes and fluid were leaking from the vessels. The basal infoldings were abnormally separated due to edema ( ). Some epithelial cells were disrupted (arrowhead). (C) Light micrograph of a ciliary process 24 hours after PDT (toluidine blue). Most ciliary body blood vessels were thrombosed (arrows). Ciliary processes were swollen because of ciliary epithelial and stromal edema. (D) Electron micrograph of ciliary process 24 hours after PDT. The endothelia of the blood vessels was severely damaged or missing (arrow). Large intercellular spaces were occupied by exudates (arrowhead). Basal infoldings were extremely enlarged ( ). (E) Light micrograph of a ciliary process 7 days after PDT (toluidine blue). The edema had decreased in the epithelial layers. Some of ciliary body blood vessels were still thrombosed (arrows). (F) Electron micrograph of a ciliary process 7 days after PDT. The basal infoldings were still separated due to edema (asterisk). Regenerative epithelia were seen (arrows), and vacuolations were still present in the epithelial layer (arrowhead). Scale bar: (B, F) 10 μm; (D) 5 μm. Magnification: (A, C, E) ×40.

Figure 4.

 

Images of lectin and TUNEL staining of a section through the untreated side (A) and treated side (B) of ciliary body in C57BL/6 normal mice. (A) The vessels in the ciliary body and sclera on the untreated side were well stained by lectin (green fluorescence) and no apoptotic cells (red fluorescence) were visible. DAPI counterstaining (blue fluorescence). (B) The vessels in the ciliary body on the treated side showed that very few were stained by lectin. There were some apoptotic cells in the ciliary body, retina, and sclera (arrow). Images of TUNEL staining of a section through the untreated side (C) and treated side (D) of the ciliary body in DBA/2J mice. (C) In treated DBA/2J eyes, TUNEL-positive cells were visible, even in the untreated side. (D) Many TUNEL-positive cells were present in the ciliary process. (E) TUNEL staining of the ciliary body in the nontreated DBA/2J control mouse eye showed no TUNEL-positive cells. C, cornea; CB, ciliary body; R, retina; V, vitreous. Magnification, ×20.

Figure 5.

 

(A) Mean IOP changes after PDT treatment of normal mice with 2.0 mg/kg verteporfin. Both first and second treatments resulted in a significant but temporary reduction of IOP in the treated eye. (B) Mean IOP changes after PDT treatment of normal mice with 4.0 mg/kg verteporfin. Significant reduction of mean IOP lasted for 14 days and returned to the level of the fellow control eyes by 17 days after treatment. (C) Mean IOP changes after PDT treatment of DBA/2J mice with 1.0 mg/kg verteporfin. (D) Mean IOP in the control eye in DBA/2J mice as a function of age. Data are the mean ± SD. *P < 0.01 compared with the control. †P < 0.05 compared with the control. §P < 0.01 compared with the control at 8 months of age.

Figure 6.

 

Retrogradely labeled RGCs from DBA/2J mice at 11 months of age. (A) Control; (B) treated with PDT at 8 months of age. Scale bar, 50 μm.

Figure 7.

 

RGC survival at 11 months in the DBA/2J mice comparing control eyes to eyes treated with PDT 3 months earlier. Data are the mean ± SD. *P < 0.05 compared with the control.