Activated Protein C Rescues the Retina from Ischemia-Induced Cell Death

Abstract

Purpose.: Ischemia causes severe and persistent visual loss in many eye diseases, including central retinal vein occlusion (CRVO) and diabetic retinopathy. Activated protein C (APC) has been demonstrated to reduce the cell death associated with ischemia in the brain and kidney. This study was performed to examine the ability of APC to rescue hypoxia-induced retinal cell death in vitro and in vivo.

Methods.: Retinal pigment epithelium (RPE) and photoreceptor cells were placed in either a normoxic or a hypoxic chamber. Immediately before they were subjected to ischemia, the cultures were treated with APC (3–240 μg/mL). Incubation was followed by an MTT assay to determine the number of viable cells. The activity of caspase-3, -8, and -9 in RPE cells was also analyzed. Various concentrations of APC were intravitreally injected in a rat CRVO model, followed by TUNEL staining to detect the in vivo effects of APC.

Results.: Lower concentrations of APC (0.3–30 μg/mL) showed a cell-protective effect against hypoxia in vitro, whereas higher concentrations (≥120 μg/mL) demonstrated cytotoxicity in both RPE and photoreceptor cells. Caspase-3, -8, and -9 were activated when the cells were exposed to hypoxia, but this activation was significantly inhibited by APC. Experimental CRVO-induced retinal cell apoptosis was reduced dramatically by intravitreal injection of APC.

Conclusions.: APC can reduce ischemia-induced cytotoxicity both in vitro and in vivo via blocking the activation of caspase-3, -8, and -9. APC may be a promising candidate for protecting the retina from ischemia.

Retinal ischemia is a major cause of visual impairment and blindness. At the cellular level, ischemic retinal injury consists of a self-reinforcing destructive cascade involving neuronal depolarization, calcium influx, and oxidative stress initiated by energy failure and increased glutamatergic stimulation. The ischemic cascade, once activated, results in cell death, which can occur by the classic and rapid necrotic process or by the slower process of apoptosis. Several animal models and analytical techniques have been used to study retinal ischemia, and an increasing number of treatments have been shown to interrupt the ischemic cascade and attenuate the detrimental effects of retinal ischemia. ¹ However, success in the laboratory has not translated to the clinic. Furthermore, neuroprotection-based treatment strategies for retinal ischemia have so far been disappointing. 

To date, there is compelling evidence that activated protein C (APC) could be an ideal neuroprotectant candidate for ischemic stroke therapy. ² APC, a plasma serine protease with systemic anticoagulant, anti-inflammatory, and antiapoptotic activities, and direct neuronal protective activity, ^{3 –8} has been shown to attenuate microvascular injury during systemic hypoxia and to reduce tissue hypoxia in traumatized skeletal muscle during endotoxemia. ^9,10 Meanwhile, APC is also believed to protect neurons and neurovascular cells from cell death after transient brain ischemia and embolic stroke in rodents. ^4,6,7  

Therefore, we hypothesize that APC may be a promising therapeutic approach in the treatment of retinal ischemic diseases. Thus, we performed this study to characterize the biological effects of APC in both normoxic and hypoxic conditions, and further investigated whether APC had the ability to block hypoxia-induced cytotoxicity in vitro and in vivo. 

Methods

Reagents

Human plasma-derived APC was kindly provided by the Chemo- Sero- Therapeutic Research Institute (Kumamoto, Japan). Caspase detection kits (SR-VAD-FMK for caspase-3, FAM-LETD-FMK for caspase-8, and FAM-LEHD-FMK for caspase-9) were purchased from Peninsula Laboratories, Inc. (San Carlos, CA). Caspase inhibitors (Z-DEVD-FMK for caspase-3, Z-IETD-FMK for caspase-8, Z-LEHD-FMK for caspase-9) were purchased from BioVision (Mountain View, CA). An in situ terminal deoxynucleotidyl transferase-mediated digoxigenin-dUTP nick-end labeling (TUNEL) kit was obtained from Roche Applied Science (Penzberg, Germany). 

ARPE-19 and 661w Cell Culture

661w cells, a cell line cloned from retinal tumors of a transgenic mouse demonstrating cellular and biochemical characteristics of cone photoreceptors, ¹¹ were kindly provided by Muayyad R. Al-Ubaidi (University of Oklahoma Health Sciences Center, Oklahoma City, OK). ARPE-19, cultured human retinal pigment epithelial cells, and 661w cells were maintained in Dulbecco's modified Eagle's medium (DMEM) containing 3 mM l-glutamine, 10% fetal bovine serum, 100 U/mL penicillin G, and 100 μg/mL streptomycin sulfate at 37°C in an environment containing 95% O₂ and 5% CO₂. Cell suspensions (4 × 10⁵ cells/mL) were seeded onto microplates (tissue culture grade, 96-well, flat bottomed) in a final volume of 100 μL culture medium per well. The cells were incubated in 96-well plates overnight to reach 70% to 80% confluence before being exposed to the treatment. 

APC Biological Effects under Normoxic Conditions

Dose-dependent APC biological effects were examined by incubating ARPE-19 and 661w cells with serial concentrations of APC (3, 6, 12, 30, 60, 120, and 240 μg/mL) for 24 hours. To detect the effects over time, the cells were cultured with APC (3, 30, and 120 μg/mL) for 0.5 (12 hours), 1, 3, 7 days. 

At the indicated time points, the cellular activity of mitochondrial dehydrogenase was measured based on the cleavage of the yellow tetrazolium salt 3-(4,5-dimethylthiazol-2-yl)-2,5-diphenyl tetrazolium bromide (MTT; Roche Applied Science, Mannheim, Germany). 

Hypoxia-Induced Cytotoxicity in RPE and Photoreceptor Cells and Inhibition by APC

Cell suspensions (4 × 10⁵ of ARPE-19 and 1 × 10⁵ of 661w cells/mL) were seeded onto 96-well microplates in a final volume of 100 μL culture medium per well. Hypoxia-related cytotoxicity over time was investigated by placing the cells in an anaerobic chamber (<1% oxygen) for 4, 8, and 12 hours. At the indicated time points, an MTT assay was performed. 

Caspase-3, -8 and -9 Activities

Caspase activities in ARPE-19 cells cultured for 8 hours under normoxic conditions were used as a control. Caspase activity in hypoxic conditions was detected by placing ARPE-19 cells into an anaerobic chamber for 8 hours. Then, 3.3 μL of 30× caspase detection reagents (SR-VAD-FMK for caspase-3, FAM-LETD-FMK for caspase-8, FAM-LEHD-FMK for caspase-9) were added per 100 μL of sample volume. After 1 hour of incubation at 37°C, a fluorescent microscope and a fluorescent plate reader were used to visualize and analyze the activities of caspase-3, -8 and -9. The ratio of optical density (OD) of experimental groups to that of control was used to represent the caspase activity. 

To evaluate the effect of caspase inhibition, caspase inhibitors (Z-DEVD-FMK for caspase-3, Z-IETD-FMK for caspase-8, and Z-LEHD-FMK for caspase-9) were added to the culture medium for 8 hours under hypoxic conditions. Then, an MTT assay was performed to determine cell viability. 

APC Effects on Hypoxia-Induced Caspase Activation

ARPE-19 cells were placed into an anaerobic chamber with APC (3, 6, 12, and 30 μg/mL) for 8 hours. Then caspase activities were evaluated as just described. 

In Vivo CRVO Model

We used pigmented rats weighing from 150 to 200 g. All animals were treated in accordance with the ARVO Statement for the Use of Animals in Ophthalmic and Vision Research. The CRVO model was established as previously described. ^12,13 Briefly, immediately after an injection of rose bengal (2 mg/kg; Nacalai Tesque, Kyoto, Japan) into the tail vein, a laser was applied to the major retinal veins adjacent to the optic nerve head to induce a thrombus. The laser settings were as follows: wavelength, 514 nm; spot size, 50 μm; power, 100 mW; duration, 0.5 second; and number of applications, one to three shots for each vein. Retinal vein occlusion was confirmed by fluorescein angiography, and eyes with hemi-CRVO or BRVO were excluded from the study. 

One hour after CRVO induction, 2 μL of balanced salt solution (BSS, Alcon, Kyoto, Japan) or APC (0.03 μg/μL or 0.3 μg/μL) were intravitreally injected into the eyes (n = 6 for each group). The final concentrations of APC (∼3 and 30 μg/mL), were calculated based on a previously reported method. ¹⁴  

Twenty-four hours after intravitreal injection, the rats were euthanatized with an overdose of ketamine. Immediately after euthanatization, the eyes were enucleated and fixed with 4% paraformaldehyde in phosphate-buffered saline at 4°C overnight. Then, 8-μm cryosections across the optic disc and the injection site were stained using the TUNEL method. Ten sections from each eye were stained, and the number of all TUNEL-positive cells in each section was counted manually under a microscope, except for those cells within a disc diameter from the optic disc, which were directly damaged by laser irradiation. 

Statistical Analysis

All experiments were repeated at least four times, and the results are expressed as the means ± standard deviations (SDs). One-way analysis of variance followed by least-significant difference (LSD) t-test was used to evaluate the significance of any differences, and P < 0.05 was considered statistically significant. 

Results

Biological Effects of APC on Normoxic ARPE-19 and 661w Cells

Lower concentrations of APC (3–60 μg/mL) showed a cell protective effect, whereas higher concentrations of APC (≥120 μg/mL) developed cytotoxicity in both ARPE-19 and 661w cells (Fig. 1A; Table 1A). APC at lower doses (3 and 30μg/mL) showed no cytotoxicity on normoxic ARPE-19 and 661w cells over time, however, APC at higher doses (≥120 μg/mL) showed significant cytotoxicity in the first 12 hours, and increased dramatically over time (Fig. 1B, Table 1B). 

Figure 1.

 

Biological effects of APC on normoxic and hypoxic ARPE-19 and 661w cells. (A) Two contrary effects of APC on normoxic cells according to dose variation at 24 hours of exposure. (B) Biological effects of APC on normoxic cells over time. (C) Time course of hypoxia-induced cell death. *P < 0.01 versus control.

Table 1.

 

Effects of APC on ARPE-19 and 66lw Cells in Normoxia and Hypoxia

A. Two Contrary Effects on Normoxic Cells According to Dose Variation
Cells	APC (μg/mL)
	0	3	6	12	30	60	120	240
ARPE-19	100	109	114	116	104	94.3	7.9	4.6
661w	100	102	109	102	110	109	33.7	35.7

B. Biological Effects on Normoxic Cells Over Time
Cells/APC (μg/mL)	Time Course (days)
	0	0.5	1	3	7
ARPE-19
0	100	100	100	100	97.8
3	100	114	115	109	106
30	100	107	108	104	102
120	100	35.6	7.9	6.4	4.0
661w
0	100	150	188	138	92.6
3	100	144	179	140	91.3
30	100	131	178	127	96.9
120	100	118	34.6	19.1	15.1

C. Time Course of Hypoxia-Induced Cell Death
Cells/APC (μg/mL)	Time Course (h)
	0	4	8	12
ARPE-19
Normoxia
0	100	101	99.1	97.5
0	100	82.8	49.8	46.9
0.3	100	102	106	105
Hypoxia
3	100	101	97.9	99.1
30	100	103	97.4	104
661w
Normoxia
0	100	120	148	165
0	100	91.4	78.3	62.7
0.3	100	94.9	93.7	83.5
Hypoxia
3	100	96.3	97.9	90.0
30	100	92.3	103	95.7

 Data are the percentage of viable cells.

Hypoxia-induced Cytotoxicity and Cytoprotection of APC on Hypoxic ARPE-19 and 661w Cells

In normoxic conditions, no cell death was observed in 24 hours. When ARPE-19 and 661w cells were cultured in an anaerobic chamber, cell death developed in a time-dependent manner (Fig. 1C, Table 1C). The most significant cell death induced by hypoxia was observed after 8 hours of treatment. APC ranging from 0.3 to 30 μg/mL significantly protected ARPE-19 and 661w cells from hypoxia-induced cell death, and APC-mediated cytoprotection had no obvious difference due to the concentrations of the applied doses (Fig. 1C, Table 1C). 

Involvement of Caspase-3, -8, and -9 in Hypoxia-Induced Cytotoxicity

After ARPE-19 cells were incubated in an anaerobic chamber for 8 hours, the activation of caspase-3, -8, and -9 was observed (Fig. 2A–G; Table 2A). When inhibitors of caspase-3, -8, and -9 were added to the culture medium, hypoxia-induced cell death was significantly reduced (Fig. 2H, Table 2B). There was no observed difference in the reduction of cell death between the inhibitors of caspase-8 and -9; however, caspase-3 inhibition showed a less protective effect (P < 0.05). 

Figure 2.

 

Involvement of caspase-3, -8, and -9 in hypoxia-induced cytotoxicity for ARPE-19. (A–G) Increased activity of caspase-3, -8, and -9 at 8 hours of hypoxia. (H) Inhibition of activated caspase-3, -8 and -9 attenuated hypoxia-induced cytotoxicity. Data are the mean ± SD of results in four independent experiments. I-cas, caspase inhibitor.*P < 0.01. Scale bar, 100 μm.

Table 2.

 

Activation and Inhibition of Caspase-3, -8, and -9

A. Increasing Activity 8 Hours of Hypoxias
	Normoxia			Hypoxia
	Caspase-3	Caspase-8	Caspase-9	Caspase-3	Caspase-8	Caspase-9
Caspase activity	1.00	1.00	1.00	3.62	4.80	4.30
SD	0	0	0	0.60	0.30	0.35

B. Effect of Inhibition of Caspase on Hypoxia-Induced Cytotoxicity
	Normoxia Control	Hypoxia
		Control	1-cas 3	1-cas 8	1-cas 9	1-cas 8, 9
	100	39.8	69.4	82.9	78.8	87.9
SD	0	3.2	2.0	6.4	4.3	3.0

 Data are the percentage of viable cells. Icas, caspase inhibitor.

APC-Mediated Inhibition of Caspase Activity In Vitro

APC dosages ranging from 3 to 30 μg/mL totally abolished hypoxia-induced activation of caspase-3, -8, and-9 (Fig. 3, Table 3). There were no observed differences in caspase inhibition effects due to the various APC concentrations. These results strengthen the evidence that APC may protect cells from hypoxia-induced death by inhibiting caspase pathways. 

Figure 3.

 

Cytoprotection of hypoxia-exposed ARPE-19 cells after 8 hours of treatment by APC inhibition of activated caspase-3, -8, and -9. (A) Results from the fluorescence plate reader. (B–S) Results from fluorescence microscopy. Values are the mean ± SD of results in four independent experiments. *P < 0.01, versus control. Scale bar, 100 μm.

Table 3.

 

Cytoprotection of APC by Inhibiting Activation of Caspase -3, -8, and -9

	APC (μg/mL)
	Normoxia	Hypoxia
	0	0	3	6	12	30
Caspase-3	1.0	3.6	1.2	1.0	1.0	1.0
Caspase-8	1.0	4.8	1.1	1.2	1.2	1.2
Caspase-9	1.0	4.3	1.0	1.0	1.1	1.0

 Data are the ratio of caspase activity (OD) versus the control.

Hypoxia-Induced Apoptosis and APC-Mediated Cytoprotection in a CRVO Model

Few TUNEL-positive cells were detected in normal rat retinas, whereas a significant increase in the number of TUNEL-positive cells (Fig. 4, P < 0.01, Table 4) was observed in the experimental CRVO eyes injected with balanced salt solution. Apoptotic cells were mainly located in the outer nuclear layer but seldom in the other layers. Ischemia-associated apoptosis was reduced dramatically by intravitreal injection of APC at lower (3 μg/mL) and higher (30 μg/mL) doses (P < 0.01). There were no significant differences (P > 0.05) between the two applied concentrations of APC. No extra morphologic damage was caused by APC injection. 

Figure 4.

 

TUNEL staining in vivo. (A) TUNEL-positive cells were counted in retinas after various treatments for 24 hours after induction of experimental CRVO. (B–E) TUNEL staining results obtained by fluorescence microscopy. *P < 0.01. Scale bar, 100 μm.

Table 4.

 

TUNEL-Stained Cells In Vivo

Group	Normal	CRVO
		Saline	APC (3 μg/mL)	APC (30 μg/mL)
Apoptotic cells, n	0	52.5	5.33	6.30
SD	0	3.60	1.05	1.72

Discussion

Occlusion of retinal vessels leading to retinal ischemia is a feature shared by several disease processes including central retinal vein occlusion (CRVO), branch retinal vein occlusion (BRVO), diabetic retinopathy (DR), and retinopathy of prematurity (ROP), diseases that are collectively referred to as ischemic retinopathies. ¹⁵ An interruption in the supply of blood to the retina leads to tissue ischemia, which causes rapid failure of energy production and subsequent cell death through necrosis or apoptosis. The optimal therapy should have the benefit of restoring the normal retinal environment as well as preventing irreversible retinal damage. 

APC, which is neuroprotective during transient ischemia ^3,16 and promotes activation of antiapoptotic mechanisms in brain cells by acting directly on endothelium and neurons, ^3,5,6,8 may be a promising candidate to treat ischemic retinopathy. To our knowledge, there has been no investigation of the biological effects of APC on retina and retinal cells, to date. In our study, APC was found to have two contrary effects on cultured normoxic ARPE-19 and 661w cells, according to the applied dosage. APC at a dose of lower than 60 μg/mL had protective effects, whereas APC at a dose of higher than 120 μg/mL induced significant cytotoxicity. 

In our in vitro experiment, viability of ARPE-19 and 661w cells in the presence of hypoxia was related to the duration of exposure. APC dosages ranging from 0.3 to 30 μg/mL were demonstrated to protect against hypoxia-induced cell death. This result agrees with those in previous studies of other tissues that have shown that APC prevents glucose-induced apoptosis in endothelial cells and podocytes in diabetic nephropathy, and that APC blocks p53-mediated apoptosis in ischemic human brain endothelium. ^3,17 APC possibly reduces hypoxia-induced cell death by inhibiting the activation of caspase-3, -8, and -9. Our speculation is supported by previously reported data that APC blocks caspase-3, -8, and -9 activation in hypoxic brain endothelial cells. ⁷  

The caspases, a family of cysteine proteases and the central regulators of apoptosis, have been demonstrated to play an important role in ischemia-induced cytotoxicity in vivo and in vitro. ^{7,18 –21} Initiator caspases, including caspase-8 and -9, are closely coupled to proapoptotic signals. Once activated, these caspases cleave and activate downstream effector caspases, including caspase-3, -6, and -7, which in turn execute apoptosis by cleaving cellular proteins containing specific Asp residues. In this study, caspase-3, -8, and -9 were significantly activated by hypoxia. Yang et al. ²² reported that caspase-8 level in ARPE-19 cells was low, whereas we found elevation of caspase-8 in those cells. This discrepancy may be caused by the differences in induction; TNF-α was used in their study and hypoxia in our study. Inhibition of caspase-3, -8, or -9 produced a good protective effect on hypoxia-treated ARPE-19 cells, with no significant difference of protective effects between inhibition of caspase-8, -9, or both -8 and -9. Inhibition of caspase-8 or -9 retained a higher rate of viable cells than inhibition of caspase-3 did, suggesting that other candidates besides caspase-3, such as caspase-6 and -7, downstream of caspase-8 and -9, may be activated and contribute to hypoxia-mediated cell death. 

Next, we studied whether our findings in vitro translate in vivo by using a rat model of CRVO. Experimental CRVO-induced ischemia alone was shown to produce significant cell death, whereas APC, which was applied at two dosages (3 and 30 μg/mL) and demonstrated no cytotoxicity in vitro, had the ability to block apoptosis induced by CRVO-derived ischemia. Dead cells in CRVO retina were far more prevalent in the ONL than in other retinal layers, which may be due to the higher sensitivity of photoreceptor cells to retinal ischemia than other retinal cells. ^2,23 Our results are in good agreement with previous data showing that APC is neuroprotective during ischemia in vivo. ^2,3,5,16 Several clinical studies argue that some patients with CRVO exhibit APC resistance, but large-sample epidemiologic studies have proven that APC resistance has no major role in the pathogenesis of CRVO and other venous thrombosis diseases. ^{24 –26} Thus, we believe that this study will translate well to patients with CRVO in vivo. 

In summary, our data show that APC has a protective effect on ischemia-induced cytotoxicity both in vivo and in vitro. Therefore, APC may hold great promise in the treatment of ischemic retinal diseases. 

Footnotes

 Supported by Grant-in-Aid for Scientific Research 15591853 from the Ministry of Education, Science and Culture of Japan.

Footnotes

 Disclosure: Z.-J. Du, None; T. Yamamoto, None; T. Ueda, None; M. Suzuki, None; Y. Tano, None; M. Kamei, None