Retinal Dysfunction in Cancer-Associated Retinopathy Is Improved by Ca²⁺ Antagonist Administration and Dark Adaptation

Abstract

purpose. It was recently found that recoverin acts as an autoantigen recognized by sera of patients with cancer-associated retinopathy (CAR), and that CAR-like retinal dysfunction is produced by intravitreous administration of anti-recoverin antibody in Lewis rat eyes. To examine the pathologic molecular mechanism of CAR, and to elucidate an effective therapy for CAR, the function and morphology of CAR were compared with those of phototoxic retinal damage, another form of photoreceptor dysfunction, and the effect of nilvadipine, a Ca²⁺ antagonist, on the retinal degenerations was studied, using these models.

methods. Under different illumination conditions and/or medication with nilvadipine, the functional and morphologic properties of the retinas were evaluated after intravitreous injection of anti-recoverin antibody into Lewis rat eyes (six rats, 12 eyes in each experimental condition), using electroretinogram (ERG), rhodopsin phosphorylation, and light microscopy.

results. Anti-recoverin antibody administered into the vitreous of Lewis rat eyes induced a significant decrease and increase of ERG responses and rhodopsin phosphorylation levels, respectively, under cyclic or continuous light. Similar changes were observed in eyes of rats bred under continuous illumination that did not receive anti-recoverin antibodies. However, anti-recoverin antibody–induced retinal dysfunctions were not observed in rat eyes under dark conditions. Administration of nilvadipine, a Ca²⁺ antagonist, to the anti-recoverin antibody–treated rats and rats with phototoxic retinal dysfunction caused significant improvement of the deterioration of ERG and normalization of rhodopsin phosphorylation.

conclusions. The present data indicate that anti-recoverin antibody–induced retinal dysfunction was functionally similar to phototoxic retinal dysfunction and was markedly suppressed under dark conditions or by systemic administration of a Ca²⁺ antagonist.

Cancer-associated retinopathy (CAR), characterized by photopsia, progressive visual loss with a ring scotoma, attenuated retinal arterioles, and abnormalities of the a- and b-waves of ERG, ¹ has been recognized in patients with small-cell carcinoma of the lung and other malignant tumors. ^{2 3 4 5 6 7 8} Based on histopathologic and immunologic studies, it has been suggested that in CAR, photoreceptor loss is primarily caused by an autoimmune reaction against a photoreceptor-specific 23-kDa calcium-binding protein, recoverin. ^{9 10} Functionally, recoverin has been identified as playing a major role in light and dark adaptation by regulating rhodopsin phosphorylation and dephosphorylation in a calcium-dependent manner. ^{11 12} Aberrant expression of recoverin has been identified in cancer cells of several patients, including those with CAR, ^{13 14 15 16 17} suggesting that aberrant expression of recoverin in cancer cells may trigger an autoimmune reaction. In addition, other retinal antigens including a 65-kDa protein, ^{18 19 20} a 48-kDa protein, ⁸ enolase (46-kDa protein), ²¹ and neurofilament (58–62-kDa, 145-kDa, and 205-kDa proteins) ²² are also recognized by sera of some patients with CAR. Among these retinal autoantigens, we have identified the 65-kDa protein as heat shock cognate protein 70 (hsc70) ²⁰ and have found that CAR-like retinal dysfunction is produced by intravitreal injection of anti-recoverin antibody and that this anti-recoverin–induced retinal dysfunction is worsened by coadministration with anti-hsc70 antibody in Lewis rats. ²³ Therefore, we suggest that autoimmune reactions to recoverin and hsc70 play significant roles in the pathologic molecular mechanisms of CAR. 

In terms of the molecular mechanisms causing the retinal dysfunction through the anti-recoverin antibody, it has been reported that the anti-recoverin antibody is internalized in photoreceptor cells and induces apoptotic cell death in a retinal cell culture system, ²⁴ and that intravitreous administration of antibody against recoverin in Lewis rat eyes also causes apoptotic death of photoreceptor cells in vivo. ^{25 26} In addition, we found that anti-recoverin antibody, administered intravitreously, internalized into photoreceptors, bound recoverin, and blocked the recoverin function that inhibits rhodopsin phosphorylation in a Ca²⁺-dependent manner. ²⁶ Therefore, based on these observations, we speculated that the effects of inhibition of recoverin function by anti-recoverin antibody—that is, higher levels of rhodopsin phosphorylation and continuous opening of cGMP-gated channels resulting in accumulation of intracellular Ca²⁺ within photoreceptor cells—may represent critical steps in photoreceptor degeneration in CAR. If our speculation is correct, decrease of the light-dependent rhodopsin phosphorylation levels by dark or suppression of the increase of intracellular levels of Ca²⁺ by Ca²⁺ antagonist may have a beneficial effect on retinal dysfunction. 

In the present study, to elucidate the effect of light on retinopathy in CAR, anti-recoverin antibody–induced retinal dysfunction was evaluated under different illumination conditions and compared with phototoxic photoreceptor degeneration. Furthermore, we examined the effects of Ca²⁺ antagonist on anti-recoverin–induced retinal dysfunction. 

Materials and Methods

All experimental procedures were designed to conform to both the ARVO statement for Use of Animals in Ophthalmic and Vision Research and our own institution’s guidelines. Unless otherwise stated, all procedures were performed at 4°C or on ice, with ice cold solutions. 

Antibodies

Anti-recoverin IgG obtained by immunization of purified bovine recoverin was purified using protein G Sepharose column chromatography, as described previously. ²³ The specificity and titers were examined by Western blot analysis, using a bovine retinal soluble fraction, as described in our previous study, ²³ before the antibody penetration. Both antibodies were diluted with phosphate-buffered saline (PBS) to adjust the IgG concentration to 1 mg/ml. 

Anesthesia

In the present study, 8-week-old Lewis rats or Brown Norway rats (approximately 250 g) reared in cyclic light (1000 lux; 12 hours on-12 hours off) were used. To induce retinal damage, continuous strong illumination (2500 lux) by fluorescent light was used, according to the method of Aonuma et al. ²⁷ For induction of anesthesia, rats inhaled diethylether. Once unconscious, the animals were injected intramuscularly with a mixture of ketamine (80–125 mg/kg) and xylazine (9–12 mg/kg). Adequacy of the anesthesia was tested by tail clamping, and supplemental doses of the mixture were administered intramuscularly if needed. 

Vitreous Injection of Antibodies

Intravitreous injection of antibody was performed as described by Ohguro et al. ²³ Briefly, in rats under anesthesia, 5μ l PBS or anti-recoverin IgG (5 μg) was administered into the vitreous cavity of a rat eye. The injection was performed with a 26-gauge microneedle syringe (Hamilton, Reno, NV) through the sclera at a point 1 mm from the limbus, to avoid puncturing the lens. Animals showing apparent traumatic changes, such as cataract, after vitreous injection were excluded from the present study. After the surgery, a drop of 0.5% ofloxacin was administered to avoid infection. 

Drug Administration

5-Isopropyl 3-methyl 2-cyano-6-methyl-4-(3-nytrophenol)-1,4-dihydro-3,5- pyridine dicarboxylate (nilvadipine; Fujisawa Pharmaceutical Co. Ltd., Tokyo, Japan) was dissolved in a mixture of ethanol-polyethylene glycol 400-distilled water (2:1:7) at a concentration of 0.1 mg/ml, diluted twice with physiological saline before use, and injected intraperitoneally (0.5 ml/kg) into anesthetized rats once a day for 3 weeks. In control rats, the same solution without nilvadipine (vehicle solution) was administered the same as the antibodies. 

Electroretinography

While under anesthesia, each rat was laid on its side with its head fixed in place with surgical tape in an electrically shielded room and dark adapted for at least 1 hour. The pupils were dilated with drops of 0.5% tropicamide. ERGs were recorded with a contact electrode equipped with a suction apparatus to fit on the cornea (Kyoto Contact Lens Co., Kyoto, Japan). A grounding electrode was placed on the ear. Responses evoked by white flashes (3.5 × 10² lux, 200-msec duration) were recorded (Neuropack MES-3102; Nihon Kohden, Tokyo, Japan), as described by Ohguro et al. ²³ The a-wave amplitude was determined from the baseline to the bottom of the a-wave. The b-wave amplitude was determined from the bottom of the a-wave to the top of the b-wave. 

Light Microscopy

Anesthetized animals were transcardially perfused with 100 ml 82 mM sodium phosphate buffer (pH 7.2), containing 4% paraformaldehyde. Posterior segments (5 × 5 mm² containing the optic disc) dissected from the enucleated eyes were embedded in paraffin. Retinal sections were cut vertically through the optic disc at 4-μm thickness, mounted on subbed slides, and dried. The sections were processed with hematoxylin-eosin staining after deparaffinization with graded ethanol and xylene solutions. 

Rhodopsin Phosphorylation

Rhodopsin phosphorylation of rat eyes was studied as retinal photoreceptor functions by using isolated rod outer segments, as described by Ohguro et al., ¹² with some modifications. Briefly, after dark adaptation of enucleated eyeballs (two eyes for each condition) for 1 hour on ice, retinas were dissected and homogenized in 0.5 ml 45% sucrose in buffer A (100 mM Na-phosphate buffer [pH 7.2] containing 5 mM MgCl₂). After centrifugation at 13,000 rpm for 5 minutes, the supernatant was diluted twice with buffer A and centrifuged again at 13,000 rpm for 5 minutes. The pellet was dissolved in 200 μl of buffer A containing 0.5 mM[γ -³²P]-adenosine triphosphate (ATP; 100 cpm/pmol) and incubated at 30°C for 5 minutes under a 100-W lamp from a distance of 10 cm. The reaction was terminated by addition of buffer B (200 mM Na-phosphate buffer [pH 7.2] containing 5 mM adenosine, 100 mM KF, 200 mM KCl, and 200 mM EDTA) and centrifuged at 13,000 rpm for 5 minutes. The pellet was dissolved in 50 μl SDS-PAGE sample buffer and analyzed by SDS-PAGE using 12.5% gel. The gels were stained and destained with Coomassie blue and dried, followed by autoradiography. The band corresponding with rhodopsin was cut out and dissolved in 0.5 ml H₂O₂, and radioactivities were counted in a scintillation cocktail. 

Statistical Analysis

Retinal sections were photographed, and each retinal layer was measured at temporal and nasal points apart from 1 mm from the optic disc (two points per section), and compared between control and anti-recoverin antibody–treated animals under different conditions in five different sections from five rats in each condition. The experimental data including ERG amplitudes and rhodopsin phosphorylation (n = 6 rats, 12 eyes in each conditions) and thickness of retinal layers (n = 10) are shown as mean ± SD. Significant differences between groups were found, by using the Mann-Whitney test with a significance level of less than P < 0.05, 0.01, or 0.001. 

Results

To determine whether the anti-recoverin antibody–induced photoreceptor cell death is photodependent, ERG responses were analyzed in eyes intravitreously administered with anti-recoverin antibody under different illumination conditions, consisting of dark, continuous light, and cyclic light. Within 3 weeks after intravitreal injection, amplitudes of b-wave of ERG were not affected in eyes with PBS under dark and cyclic light (Fig. 1 , bars 1, 2) but was significantly decreased under continuous light (bar 3). In contrast, ERG responses were significantly affected in the anti-recoverin antibody–treated eyes under cyclic and continuous light (bars 5 and 6), but these changes were not observed in the dark conditions (bar 4). Histopathologically, the thickness of all retinal neuronal layers of anti-recoverin antibody–treated eyes was diminished throughout the 3-week period after intravitreal injection under cyclic light (Fig. 2 , bars and micrographs 2–4); whereas in the dark the retinal layer was not affected (bar and micrograph 5). In contrast, in phototoxic retinal degeneration, among the retinal layers only the outer segment (OS) layer was thinner (bar and micrograph 6). These observations suggest that activation of the light-dependent phototransduction pathway may be required for development of anti-recoverin antibody–induced retinal dysfunction similar to phototoxic retinal dysfunction. 

To study the effects of a Ca²⁺ antagonist on the retinal dysfunction induced by the anti-recoverin antibody, nilvadipine, which among the Ca²⁺ antagonists is known to be the most effective penetrator of the central nervous system in clinical practice, ^{28 29 30} was administered intraperitoneally every other day for 3 weeks to rats treated with anti-recoverin antibody or PBS or to rats raised under continuous or cyclic illuminations. Thereafter ERG responses and rhodopsin phosphorylation reactions were evaluated. The ERG responses and rhodopsin phosphorylation reactions of untreated rats (n = 5 rats, 10 eyes) were not changed by the intraperitoneal administration under cyclic light of nilvadipine or its vehicle solution (Fig. 3) . In addition, nilvadipine caused no significant changes in retinal morphology (data not shown). However, anti-recoverin antibody–induced reduction (Fig. 4 , bar 2) and continuous-light–dependent reduction (bar 4) in ERG responses significantly recovered after the administration of nilvadipine (bars 3, 5). These effects of nilvadipine on the light-induced reduction of ERG responses were also observed in a pigmented rat species, Brown Norway rats (bars 6, 7). 

To study the effects of light- or anti-recoverin antibody–induced retinal damage on the residual recoverin function, we examined levels of rhodopsin phosphorylation in rat eyes exposed to continuous illumination or treated with anti-recoverin antibody and found significant enhancement in the levels of rhodopsin phosphorylation in both of the affected eyes (Fig. 5 , bars 3, 5). In addition, these increases in rhodopsin phosphorylation levels were significantly suppressed by the administration of nilvadipine, in both continuously illuminated (Fig. 5 ; bar 4) and anti-recoverin antibody–treated (bar 6) rat eyes, although nilvadipine did not affect the rhodopsin phosphorylation levels of control condition (PBS, cyclic light conditions; bars 1, 2). 

Discussion

With regard to the pathologic molecular mechanisms of CAR, it has been suggested that anti-recoverin antibody plays a pivotal role in retinal photoreceptor degeneration, ^{20 23 24 25 26} and that anti-hsc70 antibody may enhance anti-recoverin antibody–induced retinal degeneration. ²³ Regarding the mechanism of the anti-recoverin antibody–induced retinal degeneration, it has been reported that anti-recoverin antibody is localizes in photoreceptor cells and blocks recoverin function and regulation of rhodopsin kinase in a Ca²⁺-dependent manner, resulting in enhancement of rhodopsin phosphorylation and induction of apoptotic cell death. ^{24 25 26} Based on these observations, we suggest that after intravitreal administration of anti-recoverin antibody, the following set of events may happen: migration of the anti-recoverin antibody into photoreceptor cells; binding of the anti-recoverin antibody with recoverin; blocking of recoverin function and inhibition of rhodopsin kinase in a calcium-dependent manner, causing enhancement of rhodopsin phosphorylation; marked suppression of light-dependent transducin activation; continued opening of cGMP-gated channels in plasma membranes, resulting in increased intracellular Ca²⁺ levels; and activation of the caspase-dependent apoptotic pathway. ²⁶  

In the present study, we found that the anti-recoverin antibody–induced retinal dysfunction was exclusively photodependent and did not occur in the dark and that changes in ERG responses and rhodopsin phosphorylation levels of anti-recoverin–induced retinal dysfunction were similar to those of phototoxic retinal dysfunction caused by continuous illumination. ^{27 31 32} Therefore, these observations suggest that uncontrolled states of the phototransduction pathway, by blockage of recoverin with anti-recoverin antibodies or by continued activation of rhodopsin under continuous illumination, may constitute a common mechanism in photoreceptor cell death. This idea is supported by two lines of evidence: First, mutations in proteins involved in the phototransduction pathway, including rhodopsin, arrestin, and cGMP phosphodiesterase, are found in patients with retinitis pigmentosa (RP) and in animal models of RP. ^{33 34} Second, absence of or abnormally high levels of rhodopsin phosphorylation are possible mechanisms of retinal degeneration in RP. ^{35 36 37}  

Regarding therapy for patients with CAR and other paraneoplastic syndromes, such as paraneoplastic cerebellar degeneration and Lambert-Eaton myasthenia syndrome, steroid administration, immunomodulation, and plasmapheresis have been clinically performed in conjunction with anti-neoplastic therapy. ^{38 39 40} For CAR, no definitive therapy has been established, although it has been reported that these treatments may be effective in some patients. ^{3 5 6 8 20 41} Recently, it has been reported that lowering of intracellular Ca²⁺ by Ca²⁺ antagonists and other drugs effectively suppresses the retinal neuronal apoptosis induced in some experimental animal models by ischemia-reperfusion ^{42 43} and intravitreal injection of N-methyl-d-aspartate (NMDA). ⁴⁴ These observations allowed us to speculate that similar effects by a Ca²⁺ antagonist could be expected in our CAR model and phototoxic model rats. 

In a prior study, we found no significant effects of steroid or cyclosporin A on anti-recoverin antibody–induced retinal dysfunction. ²³ In the current study, nilvadipine demonstrated marked effects on anti-recoverin antibody–induced retinal dysfunction. At present, the precise mechanisms of nilvadipine have not been elucidated. However, we speculate that the lowering of intracellular Ca²⁺ levels by nilvadipine may inhibit the Ca²⁺-dependent apoptotic pathway. It is of interest that Frasson et al. ⁴⁵ recently reported rod photoreceptor rescue by lowering intracellular Ca²⁺ levels in photoreceptor cells using d-cis-diltiazem, a Ca²⁺ channel blocker in a different animal model of RP, the rd mouse, in which the gene encoding cGMP phosphodiesterase is affected. However, recently, Bush et al. ⁴⁶ reported that a Ca²⁺ antagonist, diltiazem, had no effects on photoreceptor degeneration in the rhodopsin P23H rat. Therefore, it seems likely that Ca²⁺ channel blockers have protective effects on the retinal degeneration in some disease models, but these effects may be variable among different models, different species, disease stages, and different Ca²⁺ antagonists. 

Calcium antagonists, which have been widely used in treatment of systemic hypertension, inhibit the intracellular entry of the calcium ion, relax vascular smooth muscle cells, and increase regional blood flow in several organs. ^{28 29 30} It has been suggested that some of the calcium antagonists effectively retard the progression of visual field defects in some glaucoma patients, ^{47 48 49 50} especially in normal tension glaucoma (NTG), through their vasodilating effects on intraocular blood flow. ^{47 50} More recently, it has been reported that nilvadipine, another dihydropyridine calcium antagonist, had minimum effects on systemic blood pressure in subjects without hypertension, ⁵¹ increased vertebrate blood flow more effectively than nifedipine or nicardipine in dog, ⁵² and increased blood velocity and blood flow in the optic nerve head, as well as in the choroid and retina in rabbit. ⁵³ In addition, oral administration of calcium antagonists produced clinically beneficial effects on glaucomatous visual field losses in some patients with primary open-angle glaucoma (POAG) and in some with NTG. ⁵⁴ Based on these observations, we speculated that these beneficial effects of nilvadipine may be related to its vasodilating action on vessels within the central nervous system as well as its ability to decrease high levels of intracellular calcium, which in turn triggers apoptotic cell death in neurons. 

In summary, our present study suggests that nilvadipine may be effective in the treatment of retinal degeneration in animal models of CAR. Further laboratory and/or prospective clinical study is indicated to determine whether these findings are applicable to CAR in humans. 

 Supported by grants from the Japanese Ministry of Health, Naito Memorial Foundation, Ciba-Geigy Foundation for the Promotion of Science, The Mochida Memorial Foundation for Medical and Pharmaceutical Research, Uehara Memorial Foundation, and JRPS Research Foundation.

 Submitted for publication February 6, 2001; revised June 4, 2001; accepted June 27, 2001.

 Commercial relationships policy: N.

 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.

 

Figure 1.

 

Effects of illumination conditions on ERG in rat eyes treated with anti-recoverin antibody or PBS. Either PBS or 5 μg anti-recoverin antibody was injected intravitreously in Lewis rat eyes. After treatment for 3 weeks under different illumination conditions of dark, cyclic light, and continuous light, ERG measurement were performed in 12 eyes (six rats) in each condition. (A) ERG traces; (B) mean ± SD of the b-wave amplitudes. Most eyes produced similar ERG responses in all conditions (B; bar and corresponding condition numbers): (1) 12 of 12 eyes; (2) 12 of 12; (3) 10 of 12; (4) 12 of 12; (5) 11 of 12; and (6) 12 of 12.

Figure 2.

 

Changes in the thickness of retinal neuronal layers in an anti-recoverin antibody–treated eye. Hematoxylin-eosin–stained retinal sections near the posterior pole from Lewis rat eyes that had been treated with PBS or anti-recoverin antibody under different light conditions for 24 hours, 1 week, or 3 weeks. (A) Representative micrographs. GCL, ganglion cell layer; IPL, inner plexiform layer; INL, inner nuclear layer; OPL, outer plexiform layer; ONL, outer nuclear layer; OS, outer segment. Scale bar, 50 μm. (B) Each retinal layer was measured at temporal and nasal points 1 mm away from the optic disc in five different rat sections (total 10 points). The mean ± SD of the thickness (%) was plotted. The numbered bars correspond to the numbered micrographs.

Figure 3.

 

Effects of nilvadipine administration on ERG and rhodopsin phosphorylation in untreated rats. Nilvadipine or its vehicle solution was administered intraperitoneally every day for 3 weeks to untreated Lewis rats (n = 5, 10 eyes). ERGs were then measured, OS from two retinas were prepared, and light-dependent phosphorylation by[γ -³²P] ATP was examined. After the reaction, the radioactivity of the rhodopsin bands in SDS-PAGE was counted in a scintillation cocktail and plotted. Experiments were performed in triplicate. All eyes (10/10) produced similar ERG responses in all conditions.

Figure 4.

 

Effects of nilvadipine administration on ERG in anti-recoverin antibody–induced or phototoxic retinal dysfunction in rat eyes. Either PBS or 5 μg anti-recoverin antibody was injected intravitreously in Lewis rat eyes. After treatment for 3 weeks, PBS-injected rats were reared under cyclic light or continuous light, and the anti-recoverin antibody–treated rats were reared under cyclic light. Nilvadipine or vehicle solution was administered intraperitoneally every other day for 3 weeks and then ERG measurements were performed in 12 eyes (six rats) in each condition. (A) ERG traces; (B) mean ± SD of the b-wave amplitudes. Most eyes produced the same ERG responses (bar and corresponding condition numbers): (1) 12 of 12 eyes; (2) 11 of 12; (2) 10 of 12; (3) 10 of 12; (4) 11 of 12; (5) Brown-Norway; 11 of 12; (6) Brown Norway; 10 of 12.

Figure 5.

 

Effects of nilvadipine administration on rhodopsin phosphorylation in anti-recoverin antibody–induced or phototoxic retinal dysfunction in rat eyes. Rats were treated as described in Figure 4 . ROS was prepared from two retinas and light-dependent phosphorylation by[γ -³²P]-ATP was examined. After the reaction, the radioactivity of rhodopsin bands in SDS-PAGE was counted in a scintillation cocktail and plotted. Experiments were performed in triplicate.