Absence of Photoreceptor Rescue with d-cis-Diltiazem in the rd Mouse

Abstract

purpose. Because of a previous report suggesting that d-cis-diltiazem slows retinal degeneration in rd mice, this study was undertaken to examine the effect of d-cis-diltiazem on photoreceptor structure and function in this line of mice.

methods. Mice were randomly assigned to daily intraperitoneal injections of d-cis-diltiazem or saline between postnatal days 9 and 24. On postnatal day 26 or 27, retinal function was assessed by recording dark-adapted bright-flash ERGs in all animals. Retinal morphology was examined in fixed sections and in immunolabeled frozen sections. Examiners were masked to the treatment group assignment.

results. On postnatal days 26 and 27, diltiazem- and saline-treated mice had only one row of remaining photoreceptor cells throughout most of the central retina. Cone cells in the periphery had remnants of inner segments. Total cell counts and separate counts of rod and cone photoreceptor cells by immunostaining were similar in the diltiazem- versus saline-treated mice. Both groups of mice had, on average, comparable subnormal ERG amplitudes.

conclusions. d-cis-Diltiazem had no detectable effect on preservation of photoreceptor structure and function in rd mice.

The retinal degeneration (rd) mouse carries a mutation in the gene encoding the β subunit of rod cyclic guanosine monophosphate (cGMP) phosphodiesterase (PDE6B) ¹ ; this mutation is the cause of approximately 2% of cases of retinitis pigmentosa in the United States. ^{2 3} Patients with this form of retinitis pigmentosa, similar to rd mice, have severe retinal degeneration. ³ In a light-dependent manner, rod PDE6B normally helps to regulate the cGMP level within rod photoreceptors, ⁴ which in turn governs the cationic conductance through specialized cGMP-sensitive channels across photoreceptor outer segment membranes. ^{5 6 7} In the rd mouse, it is hypothesized that these channels remain open because of an abnormally high cGMP level, ⁸ which ultimately leads to the death of rod photoreceptor cells. ⁹  

d-cis-Diltiazem (Cardizem, Hoechst-Marion Roussel, Kansas City, MO) is a well-known calcium channel–blocking agent used for the treatment of cardiac disorders in humans. ¹⁰ Although the d-cis form of this drug has been reported to have only a mild effect on cGMP-gated conductance in rod outer segments, ^{5 11} Frasson et al. ¹² have reported photoreceptor rescue in the rd mouse model of retinitis pigmentosa after treatment with this agent. On postnatal day (P)25, mice injected intraperitoneally with d-cis-diltiazem were reported to have more numerous rod and cone photoreceptor cells and larger ERG a- and b-waves than mice injected with saline. ¹² Frasson et al. have speculated that their rescue effect with the d-cis form of diltiazem was most likely not by a direct action of this drug on rod outer segments but by some more proximal effect, perhaps involving blockage of L-type voltage-gated calcium channels. 

The present study was conducted in an attempt to confirm the findings of Frasson et al., because successful treatment of rd mice with d-cis-diltiazem would provide a rationale for a trial of Cardizem (Hoechst-Marion Roussel) in patients with retinitis pigmentosa with mutations in the PDE6B gene. 

Methods

d-cis-Diltiazem Injection

Five rd adult female mice (C3H/HeNCrlBR) with litters (Charles River Laboratories, Wilmington, MA) were reared under 65- to 140-lux white fluorescent cyclic (8:00 AM light to 8:00 PM dark) light. On P9, each of the five litters was separated from its birth mother, and pups were randomly assigned to one of four new foster mothers, so that each of the foster mothers had a litter (n = 5–6) containing mice from each of the five original litters (total of 22 mice). Twice daily, from P9 to P24, mice were weighed and given intraperitoneal injections of d -cis-diltiazem (Sigma, St. Louis, MO) in 0.9% saline (2.25 mg/mL) or saline alone. The source and dosage schedule of d -cis-diltiazem was the same as that used by Frasson et al. (Sahel JA, written communication, October 1999). Based on body weight, the dosage was 28 mg/kg on P9 (average weight, ∼4 g) and 53 mg/kg on P17 (average weight, ∼7 g). 

We conducted a pilot study of four normal mice to be certain that our preparation of d-cis-diltiazem injected intraperitoneally affected retinal function. We confirmed that a single injection led to reductions in ERG b-wave amplitude in all four mice within 30 minutes, as described previously by Frasson et al. (Fig. 1) . 

ERGs

Because normal mice treated with diltiazem have significantly reduced ERG b-waves soon after injection of diltiazem, we allowed 2 to 3 days for clearance of the drug before recording ERGs. On P26 or P27, after overnight dark adaptation, rd mice were anesthetized with intraperitoneal sodium pentobarbital (80 mg/kg) and had their pupils dilated with 0.2% phenylephrine and 0.02% cyclopentolate hydrochloride. Dark-adapted (rod dominant), full-field ERGs were elicited with xenon flashes of bright white light (2.75 log foot-lambert/sec) presented at 4-minute intervals. These flashes were generated by a bright flash stimulator (L. K. C., Gaithersburg, MD) positioned above a diffuser in a Ganzfeld dome. ^{13 14} Light-adapted (cone) ERGs were then elicited in response to the same bright white flashes delivered in the presence of a background illumination of 12 foot-lamberts that had been on for 10 minutes. ¹⁵ Responses were monitored with a gold wire electrode placed on the cornea topically anesthetized with 0.5% proparacaine hydrochloride. A saline cotton wick was placed in the mouth as the reference electrode, and a subdermal electrode was placed in the neck as the ground. Responses were differentially amplified at a gain of 5000 (−3 dB at 2 and 300 Hz), digitized at a sampling rate of 1302 Hz, and summed (n = 2) with custom software. a-Wave amplitudes were quantified from baseline to the peak of the cornea-negative deflection, and b-wave amplitudes were quantified from the latter to the major cornea-positive peak. 

Histology

Immediately after ERG testing, mice were killed with an overdose of intraperitoneal sodium pentobarbital (910 mg/kg). Right eyes were enucleated for light microscopy and left eyes, taken from a subgroup of saline- (n = 5) and diltiazem- (n = 4) injected mice, were enucleated for immunocytochemistry. For light microscopy, eyes were fixed in 1% formaldehyde and 2.5% glutaraldehyde in 0.1 M cacodylate buffer and embedded in Epon. One-micrometer-thick sections were cut along the vertical meridian at a level through the optic nerve head and stained with Richardson stain. For immunocytochemistry, eyes were fixed in 4% formaldehyde in phosphate-buffered saline overnight. The anterior segment and lens were removed. The eyecups were cryoprotected in 30% sucrose, frozen in optimal cutting temperature (OCT) compound, and cut along the superior to inferior meridian at a thickness of 10 μm. Sections were stained with primary antibodies for rod opsin (1D4 and 4D2) ¹⁶ or cone opsins (UV- and medium-wavelength [M] cones) ¹⁷ followed by Cy3-conjugated goat anti-rabbit secondary antibody. By light microscopy, counts of the total number of remaining photoreceptor cells (rod and cone photoreceptor cell nuclei) were made on both sides of the optic nerve head along the entire length of the section up to the ora serrata. Similarly, counts of remaining rod or cone cell nuclei were made in immunolabeled sections. 

All ERG recordings and histologic evaluations were performed by examiners masked to treatment group assignment. All animals were cared for and studied in accordance with the ARVO Statement for the Use of Animals in Ophthalmic and Vision Research. 

Statistics

Group differences for ERG amplitudes and photoreceptor cell counts were evaluated by Student’s t-test. 

Results

Histology

By light microscopy, both diltiazem- (n = 9) and saline- (n = 13) treated rd mice generally had only one remaining row of photoreceptor nuclei visible in the central retina. An occasional second row of cells could be seen in the far periphery in both groups. The total number of remaining photoreceptor nuclei along the entire length of the retinal sections was not significantly different between the diltiazem- and saline-treated mice (Table 1) . Although there were no visible inner or outer segments throughout the central retina in either group of mice, inner segment remnants were characteristic in the peripheral retina in both groups of mice (Fig. 2) . 

Immunocytochemistry

With anti-rhodopsin 1D4 and 4D2 antibody staining, numerous rod nuclei could be seen scattered throughout the retinal sections in both groups of rd mice. However, no significant difference was observed in the total number of remaining rod nuclei between the diltiazem- and saline-treated mice (Table 1) . By immunofluorescence, cone opsin (UV and M) antibody–treated sections revealed that cone cells were the predominant remaining cell type at P26 and P27 in both groups of mice. Cone opsins were especially evident in the inner segment remnants in the peripheral retina in both groups of mice (Fig. 3) . Counts of the total number of remaining cone cells over the length of the retinal sections were not significantly different between the diltiazem- and saline-treated mice (Table 1) . 

ERGs

Both d-cis-diltiazem- and saline-treated rd mice had markedly subnormal dark-adapted ERG responses on P26 and P27. Responses from both groups of mice were predominantly cornea-negative waveforms with little, if any, cornea-positive waves to quantify (Fig. 4) . There was no significant difference in mean dark-adapted ERG amplitudes between the diltiazem- and saline-treated mice (Table 2) . Light-adapted ERG responses for both groups of mice were similar in waveform to the dark–adapted responses but smaller (not shown). There was also no significant difference in mean light-adapted ERG amplitudes between the diltiazem- and saline-treated mice (Table 2) . 

Discussion

In contrast to the results reported by Frasson et al., ¹² we found that rd mice given daily intraperitoneal injections of d-cis-diltiazem from P9 to P24 did not have greater numbers of rod and cone cells or significantly larger ERGs than did rd mice injected with saline alone during the same period. Total photoreceptor cell counts, as well as separate counts of rod and cone nuclei, were not significantly different between diltiazem- and saline-treated rd mice in the present study. Moreover, many cone photoreceptor inner segment remnants were present in the peripheral retina in both groups of mice. This morphology in both diltiazem- and saline-treated mice in the present study differs from the preserved inner–outer segment remnants found only in the diltiazem-injected mice by Frasson et al. The claim of Frasson et al. that significant group differences were found in cone cell counts for diltiazem- versus saline-injected mice at P25 may be largely due to their exceptionally small within-group variances (coefficient of variation, 1.55–4.54), which are much smaller than the variances observed by us (coefficient of variation, 13.7–28.3) as well as by others in this animal model. ¹⁸ In the present study, ERGs recorded on P26 or P27 from 22 mice were essentially cornea-negative–appearing waveforms with little, if any, cornea-positive waves to quantify. Mean amplitudes were comparable for diltiazem- and saline-injected mice. 

Certain methodological differences between our study and that by Frasson et al. should be noted. First, Frasson et al. ¹² included counts from wholemount retinas that may have given a more accurate estimate of remaining rod photoreceptor cells than our counts in retinal sections alone. However, their cone cell counts were made by an indirect method of subtracting the relatively small number of remaining rod cells from the total number of cells counted in retinal sections, whereas our measure of remaining cone photoreceptor cells was made by a direct count of immunolabeled cones. This could perhaps account for their unusually low within-group variances for cone cell counts. Second, with a long-duration (i.e., 300 ms) light stimulus to elicit ERGs, they obtained unusual long latency responses in diltiazem-treated mice that can be affected by eye movements. Our test flash of approximately 10-ms duration elicited faster responses, thereby lessening the possible confounding effects of eye movements. Last, their strain of rd mice (C3H/HeHanRj; Sahel JA, written communication, December 2001), with fewer remaining photoreceptor cells at P25 compared with the present study, appears to have a slightly faster rate of disease progression than our strain (C3H/HeNCrlBR). We cannot therefore exclude the possibility that a difference in mouse strain may have contributed to differences in study outcome. 

In conclusion, we could not confirm a photoreceptor rescue effect of d-cis-diltiazem in our strain of rd mouse. Similarly, Pearce-Kelling et al. ¹⁹ noted no therapeutic effect of d-cis-diltiazem in a canine model (rcd1) of retinal degeneration with a mutation in the PDE6B gene. An absence of rescue effect of d-cis-diltiazem was also reported by Bush et al. ²⁰ in P23H rhodopsin transgenic rats. Therefore, there is no compelling rationale for testing d-cis-diltiazem (Cardizem; Hoechst-Marion Roussel) as a possible treatment for retinitis pigmentosa. 

 Supported by Grants EY10309 and EY00169 from the National Eye Institute, Bethesda, Maryland, and the Foundation Fighting Blindness, Owings Mills, Maryland.

 Submitted for publication August 17, 2001; revised January 8, 2002; accepted January 25, 2002.

 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.

 Corresponding author: Eliot L. Berson, Berman-Gund Laboratory, Massachusetts Eye and Ear Infirmary, 243 Charles Street, Boston, MA 02114.

 

Figure 1.

 

Dark-adapted, full-field ERGs from a normal (C57BL/6) mouse before and 30 minutes after an intraperitoneal injection of 200 μL of d-cis-diltiazem (2.25 mg/mL). Responses were elicited with 2.75-log foot-lamberts/sec white flashes.

Table 1.

 

Effect of d-cis-Diltiazem Versus Saline on Photoreceptor Cell Counts in rd Mice

	Diltiazem	n	Saline	n	P
Total cell counts*	391 ± 11	9	402 ± 9	13	0.45
Rod cells, †	151 ± 24	4	188 ± 13	5	0.20
Cone cells, †	232 ± 16	4	244 ± 31	5	0.75

 Data are the mean ± SEM.

*  Light microscopy.

†  Immunocytochemistry.

Figure 2.

 

Light micrographs of central and peripheral retina of diltiazem- and saline-injected rd mice. Arrowheads: inner-segment remnants in the peripheral retina. INL, inner nuclear layer; ONL, outer nuclear layer; RPE, retinal pigment epithelium. Bar, 20 μm.

Figure 3.

 

Immunofluorescence of cone opsins within inner-segment remnants in the peripheral retina of both diltiazem- and saline-injected rd mice. GCL, ganglion cell layer; INL, inner nuclear layer; ONL, outer nuclear layer; RPE, retinal pigment epithelium.

Figure 4.

 

Representative dark-adapted full-field ERGs from a normal mouse, a diltiazem-injected rd mouse, and a saline-injected rd mouse. Responses were elicited with 2.75-log foot-lamberts/sec white flashes.

Table 2.

 

Effect of d-cis-Diltiazem Versus Saline on ERG Cornea-Negative Amplitude in rd Mice

	Diltiazem	n	Saline	n	P
Dark-adapted response	35.2 ± 3.1	9	30.7 ± 2.7	13	0.28
Light-adapted response	24.7 ± 2.9	5	24.5 ± 2.3	8	0.95

 Data are mean microvolts ± SEM.