Iganidipine, a New Water-Soluble Ca²⁺ Antagonist: Ocular and Periocular Penetration after Instillation

Abstract

purpose. To evaluate the potential of topical iganidipine ophthalmic solution to exert Ca²⁺-antagonistic activity in the posterior parts of the eye without inducing systemic effects, ocular and periocular penetration of topically instilled iganidipine was studied in pigmented rabbits.

method. First, ¹⁴C-iganidipine solution (0.03%, 30 μL) was instilled into one eye, and vehicle into the other eye to determine the intraocular penetration of iganidipine and to measure the radioactivity of ocular tissues 0.25, 0.5, 1, 2, 4, and 12 hours after a single instillation (n = 3, respectively). Second, iganidipine (0.03%) or betaxolol (0.5%) was unilaterally instilled twice daily for 20 days to study the effects on intravitreously injected various doses of endothelin (ET)-1-induced retinal artery constriction to evaluate whether a pharmacologically active level of the drug penetrated to the posterior retina and to estimate the drug level in the posterior retina (n = 6, respectively). Third, iganidipine (0, 10, or 30 μg/kg: n = 6, 3, and 6, respectively) was intravenously injected to study the effects on intravitreously injected ET-1-induced retinal artery constriction to evaluate iganidipine levels in the posterior retina. Fourth, periocular penetration of iganidipine was studied by means of whole-head autoradiography after a single instillation of ¹⁴C-iganidipine (0.09%, 30 μL; n = 5).

results. Penetration of topically applied iganidipine to the cornea or aqueous humor was high and estimated to be at least 10 times higher than that reported for timolol or carteolol. Concentrations in the iris-ciliary body or retina-choroid were much higher than in the plasma, both in the treated and control eyes, suggesting that iganidipine binds to uveal pigments. Twice-daily 20-day instillation of iganidipine (0.03%), but not of betaxolol (0.5%), significantly suppressed constriction of the retinal arteries induced by intravitreous injection of ET-1 at a dose of 2.5 or 0.5 ng in the ipsilateral eye. Intravenous injection of iganidipine at a dose of 30 μg/kg (giving a free plasma concentration of approximately 10⁻⁸ M), but not at a dose of 10 μg/kg, significantly suppressed intravitreous ET-1-induced (0.5 ng) constriction of the retinal artery to a similar degree as twice-daily 20-day instillation of 0.03% iganidipine. After a single instillation of 0.09% iganidipine, the equivalent concentration of iganidipine in the ipsilateral retrobulbar periocular space estimated from autoradiography was approximately 3.9 × 10⁻⁸ M between 15 minutes and 1 hour after instillation, consistently higher than in the untreated contralateral eyes by approximately 3.0 × 10⁻⁸ M (P = 0.043).

conclusions. In rabbits, topically instilled iganidipine, a Ca²⁺ antagonist, in a 0.03% solution reaches the ipsilateral posterior retina or retrobulbar periocular space by local penetration at concentrations sufficient to act as a Ca²⁺ antagonist.

Ca²⁺ antagonists, which are widely used for the treatment of systemic hypertension, reduce vascular tone by inhibiting the entry of Ca²⁺ intracellularly, thus causing relaxation of the vascular smooth muscle cells. Vasodilation, induced by Ca²⁺ antagonists, has been shown to increase regional blood flow in several organs. ^{1 2 3} Some Ca²⁺ antagonists such as nifedipine, ^{4 5} nimodipine, ⁶ and brovincamine, ^{7 8} are used by ophthalmologists in some patients with normal-tension glaucoma (NTG). Glaucomatous tissues may have a significantly favorable response to the vasodilating effects of nifedipine ⁴ or brovincamine. ⁷  

Iganidipine hydrochloride, a new dihydropyridine-derivative Ca²⁺ antagonist, has more effective and continuous vasodilative action than nifedipine or nicardipine, and has superior stability under light ^{9 10 11 12} (Fig. 1) . Further, this substance is relatively water soluble ⁹ and is the only dihydropyridine-derivative Ca²⁺ antagonist presently available that is easily prepared as an ophthalmic solution. 

Several studies suggest that a topically instilled drug with Ca²⁺-antagonistic activity favorably affects circulation in the posterior part of the eye. Netland et al. ¹³ have reported that topical verapamil increases optic nerve head (ONH) blood flow in humans. Harris et al. ¹⁴ have reported that 4-week twice-daily instillation of betaxolol, a selective β-1 antagonist with weak Ca²⁺ antagonistic activity, ¹⁵ decreases the average resistance index in retrobulbar vessels, which suggests that betaxolol reaches the local retrobulbar tissue at pharmacologic concentrations after long-term instillation. Yoshida et al. ¹⁶ and Tamaki et al. ¹⁷ have also reported that 2-week or 3-week twice-daily unilateral instillation of betaxolol in normal humans increases retinal and ONH circulation, independent of the reduction in intraocular pressure, only in the ipsilateral eye. 

To affect the circulation locally in the retina, ONH, or posterior choroid, a topically instilled drug must reach a pharmacologically active concentration, either at the target tissue or retrobulbar tissue where there are arteries supplying blood to these tissues. ^{18 19} It is not known whether this is possible for topically instilled drugs. 

To evaluate the potential of topical iganidipine ophthalmic solution to exert Ca²⁺-antagonistic activity in the posterior parts of the eye without inducing systemic effects, we studied ocular and periocular penetration of topically instilled iganidipine in pigmented rabbits. In rabbits, topically instilled iganidipine in a 0.03% solution reached the ipsilateral posterior retina and posterior periocular tissue, mainly through local penetration at concentrations of at least 1.0 × 10⁻⁸ M, which is higher than the median inhibitory concentration (IC₅₀) of this drug in isolated vessels. ^{9 12}  

Materials and Methods

Experimental Animals

Dutch rabbits weighing 1.5 to 2.5 kg were used and handled in accordance with the ARVO Statement for the Use of Animals in Ophthalmic and Vision Research. The animals were entrained to a light schedule of alternating 12-hour periods of light and dark (lights on at 8 AM) for at least 3 weeks before the experiments. 

Drugs

The ophthalmic solution of 0.03% iganidipine—3-(4-allyl-1-piperazinyl)-2,2-dimethylpropyl methyl-1,4-dihydro-2,6-dimethyl-4-(3-nitrophenyl)-3,5-pyridinedicarboxylate dihydrochloride—and vehicle solution (0.1% sodium acetate, 0.005% benzalkonium chloride, and 0.9% sodium chloride in sterile purified water) were supplied by Senju Pharmaceutical Company, Ltd. (Kobe, Japan). 

¹⁴C-iganidipine was prepared by Daiichi Pure Chemical Company, Ltd. (Ibaragi, Japan) in a 0.03% or 0.09% solution. The specific activity of ¹⁴C-iganidipine was 2.77 MBq/mg, and the radiochemical purity was higher than 99%. 

Intraocular Penetration of Topically Instilled Iganidipine: Experiment 1

¹⁴C-iganidipine eye drops of 0.03% (30 μL) were instilled into the conjunctival cul-de-sac of one randomly chosen eye and vehicle into the other eye. Rabbits were killed by an overdose injection of pentobarbital (Nembutal; Abbott Laboratories, North Chicago, IL) 15 and 30 minutes and 1, 2, 4, and 12 hours after instillation (n = 3, respectively). After an incision was made along the edge of the orbit, the eyeballs were enucleated by grasping both clamped eyelids together with the nictitating membrane, conjunctiva, extraocular muscles, and optic nerve. Eyeballs with surrounding tissues were washed well with physiologic saline, and excess water or blood was thoroughly wiped away with filter paper. The aqueous humor was collected immediately by a syringe with a 27-gauge needle, and eyeballs were frozen in a dry ice-ethanol mixture. The cornea was removed along the limbus and the conjunctiva and extraocular muscles were removed. The sclera was cut anularly at approximately 2 mm from the limbus, and the iris-ciliary body (ICB) was removed. After the lens was removed, the scleral cup was opened by four meridional cuts, and after the vitreous was removed, the retina-choroid was exfoliated from the inside of the sclera. The optic nerve was cut from the sclera. Blood samples were obtained from the abdominal aorta. 

Radioactivity was measured with a liquid scintillation counter (Tri-Carb Liquid Scintillation Analyzer 2700TR; Perkin Elmer, Wellesley, MA). Aliquots of aqueous humor were directly mixed with a 1,2,4-trimethylbenzene scintillation cocktail (Pseudocumene, Hionicfluor; Perkin Elmer). Aliquots of other ocular tissues were air-dried after weighing and dissolved in a tissue solubilizer (Soluene-350; Perkin Elmer). All samples were radio assayed with a liquid scintillation counter, and data were corrected for quenching by automatic external standardization. 

In Vivo Evaluation of the Penetration of Topically Instilled Iganidipine or Betaxolol into the Posterior Retina: Experiment 2

Experiment 2 was performed to evaluate whether topically instilled iganidipine reaches the posterior retina at a pharmacologically active concentration. 

Single Instillation of Iganidipine

General anesthesia was induced by urethane (1 g/kg, intravenously) at approximately 8:30 AM. Body temperature was maintained with a heating pad, without the use of artificial ventilation. The pupils were dilated with one drop of 0.4% tropicamide in both eyes. Iganidipine (0.03% solution; 30 μL) was instilled into one randomly chosen eye and vehicle into the other eye at 9 AM, and 30 minutes later, one of two concentrations (5 × 10⁻⁸ or 1 × 10⁻⁸ M in 20 μL, n = 6, respectively) of endothelin (ET)-1 (human ET-1; Peptide Institute, Osaka, Japan) was injected through the pars plana into the central part of the vitreous body over the disc, with a 30-gauge needle attached to a 0.1-mL syringe. Color fundus photographs were taken 5 minutes before and 15, 30, and 60 minutes after the injection of ET-1 into both eyes at a 50° angle (TRC-WT3; Topcon, Tokyo, Japan). The photographs were then digitized into a computer (Photoshop; Adobe, San Jose, CA) using a flatbed scanner (resolution, 300 dpi) and enlarged to 4× magnification. Retinal artery diameters were then measured with an on-screen digital slide caliper tool. 

The diameters of the two major retinal arteries at the rim of the ONH and the diameter of the ONH itself were measured, and the average diameter of the two arteries was divided by the ONH diameter (the normalized diameter). The normalized diameter 15, 30, and 60 minutes after injection of ET-1, was expressed as a percentage of the diameter 5 minutes before the injection, and the results in both eyes compared 60 minutes after the injection. ²⁰ Inhibitory effects on intravitreous ET-1-induced constriction of the retinal arteries were used to indicate the presence of a pharmacologically active level of iganidipine. The normalized diameter was measured by an investigator blind to the treatment or time points. 

Twenty-Day Instillation

Iganidipine (30 μL; 0.03% solution) was instilled into one randomly chosen eye and vehicle into the other eye at 9 AM and 9 PM for 20 days (iganidipine group). On experimental day 21, general anesthesia was induced by 1 g/kg intravenous urethane at approximately 8:30 AM, and the same measurements were made as described earlier, except that four concentrations of ET-1 (2.5 × 10⁻⁷, 5 × 10⁻⁸, 1 × 10⁻⁸, or 2 × 10⁻⁹ M, n = 6 for all) were used. 

In a separate group of rabbits, 30 μL of 0.5% betaxolol was instilled into one randomly chosen eye and physiologic saline into the other eye at 9 AM and 9 PM daily for 20 days for comparison with iganidipine. On experimental day 21, general anesthesia was induced and the same measurements were made, except that only two concentrations of ET-1 (5 × 10⁻⁸ and 1 × 10⁻⁸ M, n = 6 for both) were used. Normalized diameter was measured as described earlier. 

Effect of Intravenous Iganidipine on Intravitreous ET-1-Induced Constriction of Retinal Arteries: Experiment 3

In experiment 3, we measured iganidipine levels in the posterior retina. After general anesthesia was induced as earlier, the pupils were dilated with 1 drop of 0.4% tropicamide in both eyes. Color fundus photographs were taken in both eyes 30 minutes later, and immediately thereafter iganidipine at a dose of 10 or 30 μg/kg or the same volume of vehicle solution (1 mL) was slowly injected intravenously over 5 minutes (n = 3, 6, and 6, respectively). Five minutes after completion of the intravenous injection of iganidipine, 20 μL ET-1 (1 × 10⁻⁸ M) was injected into the vitreous as just described, and color fundus photographs were taken 15, 30, and 60 minutes after the injection. Concurrently, the free plasma concentration of iganidipine was determined at 15, 30, and 60 minutes after the intravitreous injection of ET-1 (20, 35, and 65 minutes after completing the intravenous injection of iganidipine) using a liquid chromatography-tandem mass spectrometry method. The diameter was measured by an investigator blind to the treatment or time points. 

Periocular Penetration of Topically Instilled Iganidipine: Experiment 4

Rabbits were killed by an overdose injection of pentobarbital 15 and 30 minutes and 1 hour after a single 30-μL instillation of 0.09% ¹⁴C-iganidipine in the right eye (n = 1, 3, and 1, respectively). 

Whole-head autoradiography was performed, essentially as described by Sar and Stumpf ²¹ Immediately, the inside of the conjunctival sac was thoroughly rinsed with physiologic saline three times. The conjunctival cul-de-sac, nasal cavity, and anus were blocked with 5% carboxymethyl cellulose sodium (CMC-Na), and the whole body frozen in dry ice-acetone. Thereafter, the samples were dried at approximately −20°C until the next day with lyophilization, and the fur was clipped with electric clippers. After the head was separated from the frozen carcass, the head was embedded in 5% CMC-Na and again frozen with dry ice-acetone. The head was mounted on a microtome stage and cut in 35-μm sections with a cryomicrotome (Cryomicrocut; Leila MicroSystems GmbH, Nussloch, Germany) from the left to the right. The section through the optic nerve insertion and midline in both eyes was exposed to an imaging plate (Bass-III; Fuji Photograph Film, Tokyo, Japan) for 7 days to develop an autoradiogram that was visualized with a bioimage analyzer (Fujix Bas2000; Fuji Photograph Film). 

Quantitative analysis of ¹⁴C-iganidipine levels was performed in periocular tissues 15, 30, and 60 minutes after instillation. Luminescent intensity in the autoradiograph was translated into photostimulated luminescence (PSL) per square millimeter. Measurements were obtained from triangular areas between the extraocular muscles and eye wall, rectangular or square areas around the optic nerve insertion, in the extraocular muscles, and in the harderian gland (see Fig. 8 ). For a rough estimate of the equivalent concentration of iganidipine from viewing the autoradiographs, the ratios of the averaged equivalent concentration of iganidipine obtained in experiment 1 to the averaged PSL per square millimeter obtained from the extraocular muscles (Fig. 2 and Table 1 ) at 15, 30, and 60 minutes were calculated for both treated and untreated eyes. At 30 minutes, when three autoradiographs from three rabbits were available, the data from three rabbits were averaged. Thus, the calculated ratios were further averaged to yield a conversion factor: estimated equivalent concentration from autoradiograph = (X _psl − B _psl) × C, where C is the conversion factor, B _psl is background PSL per square millimeter, and X _psl is the mean PSL per square millimeter obtained from all periocular areas. 

Measurements of luminescent intensity were performed by an investigator blind to the treatment and time points. 

Data Analysis

All data are presented as the mean ± SE. Paired t-test or Wilcoxon signed rank test was used for paired variates and Mann-Whitney test was used for unpaired variates. P of < 0.05 was considered to be statistically significant. 

Results

Ocular Penetration of Topically Instilled Iganidipine: Experiment 1

Equivalent concentrations of iganidipine at each time point obtained from three rabbits after a single instillation are shown in Figure 2 . In the vehicle-treated eye, samples from three rabbits were pooled and measured, because radioactivity in the vehicle-treated eye was expected to be very low. Radioactivity in the cornea, aqueous humor, lens, and vitreous in the vehicle-treated eye was at or below the limits of detection. 

The equivalent concentration of iganidipine in the cornea was high (30 μM at 15 minutes), suggesting high permeability of iganidipine to corneal epithelium and the aqueous concentration peaked 1 hour after instillation. The equivalent concentration in the ICB was higher than that in the aqueous humor and that in the retina-choroid higher than that in the plasma. The equivalent concentration in the ICB or retina-choroid in the vehicle-treated eye was also higher than that in the plasma. 

In Vivo Evaluation of the Penetration of Topically Instilled Iganidipine to the Posterior Retina: Experiment 2

A preliminary study performed before commencing experiment 2 demonstrated that the coefficients of reproducibility ( difference between the two measurements /the mean of the two measurements) of the normalized diameter at 20-, 35-, and 65-minute-interval measurements in untreated rabbit eyes were 0.12 ± 0.02, 0.20 ± 0.04, and 0.18 ± 0.05, respectively (n = 6). 

Single Instillation of Iganidipine

After intravitreous injection of ET-1 at a dose of 2.5 or 0.5 ng (20 μL; 5 × 10⁻⁸ M or 1 × 10⁻⁸ M), there was no significant difference in the extent of retinal artery constriction between the iganidipine- and vehicle-treated eyes at any time point (n = 6, Figs. 3a 3b ). 

Twenty-Day Instillation of Iganidipine or Betaxolol

After intravitreous injection of ET-1 at a dose of 12.5 ng (20 μL; 2.5 × 10⁻⁷ M), there was no significant difference in the extent of retinal artery constriction between the iganidipine- and vehicle-treated eyes at any time point (n = 6, Fig 4a ). In contrast, after the intravitreous injection of ET-1 at a dose of 2.5 or 0.5 ng (20 μL; 5 × 10⁻⁸ M or 1 × 10⁻⁸ M), there was significantly less constriction in the iganidipine-treated eye than in the vehicle-treated eye at 60 minutes (n = 6, Figs. 4b 4c , respectively; 49.1% ± 16.5% vs. 5.2% ± 5.2% after a 2.5-ng ET-1 injection, P = 0.028 and 96.7% ± 5.9% vs. 49.2% ± 11.8% after a 0.5-ng ET-1 injection; P = 0.028, Wilcoxon signed rank test). There was a similar tendency at 15 and 30 minutes. After intravitreous injection of ET-1 at a dose of 0.1 ng (20 μL; 2 × 10⁻⁹ M), there was no significant constriction in either eye (n = 6, Fig. 4d ). 

After intravitreous injection of ET-1 at a dose of 2.5 ng (20 μL; 5 × 10⁻⁸ M), there was no significant difference in the extent of retinal artery constriction between the betaxolol- and vehicle-treated eyes (n = 6, Fig. 5a ). After intravitreous injection of ET-1 at a dose of 0.5 ng (20 μL; 1 × 10⁻⁸ M), there tended to be less constriction in the betaxolol-treated eyes than in vehicle-treated eyes at 15 minutes, but there was no significant difference at 60 minutes (Fig. 5b) . 

Effect of Intravenous Iganidipine on Intravitreous ET-1-Induced Constriction of Retinal Arteries: Experiment 3

Intravitreous injection of ET-1 at a dose of 0.5 ng (20 μL; 1 × 10⁻⁸ M) constricted the retinal artery to a similar extent after slow intravenous injection of iganidipine (10 μg/kg) as it did after slow intravenous injection of the vehicle solution (Fig 6 ; n = 3 and 6, respectively). However, the retinal artery was not significantly constricted after slow intravenous injection of 30 μg/kg iganidipine (n = 6). 

In comparison, there was a significant difference in the extent of constriction after slow intravenous injection of vehicle solution at 60 minutes (91.1% ± 5.0% vs. 44.0% ± 6.8%; P = 0.037, Mann-Whitney test). 

The free concentration of iganidipine in the plasma after slow intravenous injection of 30 μg/kg iganidipine ranged from 1.85 (15 minutes) to 0.69 (60 minutes) × 10⁻⁸ M (Table 2) . 

Periocular Penetration of Topical Iganidipine: Experiment 4

The whole-head slice included the optic nerve insertion and midline in both eyes, and the results are shown in Figure 7 . In an untreated rabbit, there was no significant nonspecific reaction not attributable to radioactivity in the present method. In the treated side, there was relatively high radioactivity in the harderian gland and in the extraocular muscles and also between the eyeball and extraocular muscles in as little as 15 minutes. The radioactivity levels were higher in the treated than in the untreated side, suggesting local penetration. Table 1 summarizes the PSL/mm² data in the extraocular muscle obtained from autoradiographs (Fig. 8) and the ratios of the averaged equivalent concentrations of iganidipine obtained in experiment 1 at 15, 30, and 60 minutes. The ratios indicate fair agreement, ranging from 0.71 to 1.42. The conversion factor was calculated as 3.42 × [10⁻⁸ M/(PSL/mm²)][mean of ratios × (0.09/0.03)]. 

Estimated equivalent concentrations in the area between extraocular muscles and eye wall, around the optic nerve insertion, and in the harderian gland are summarized in Table 3 . The concentration in the treated side was consistently higher than that in the untreated side (at 15 minutes, n = 1; at 30 minutes, n = 3; at 60 minutes, n = 1) and the difference between sides at 30 minutes was approximately 3.4 × 10⁻⁸ M (Table 3) . If all the results are pooled, the value was significantly higher in the treated side (P = 0.043, Wilcoxon signed rank test) in all tissues examined. 

Discussion

The dihydropyridine-derivative Ca²⁺-channel blocker iganidipine has potent vasodilating effects with a relatively long duration of action. ^{9 10 11 12} Because of its unique chemical structure (a hydrochloride), this drug is relatively water soluble, unlike other dihydropyridine-derivative Ca²⁺-channel blockers and is easily formulated as an ophthalmic solution. To evaluate the potential of topical iganidipine ophthalmic solution to exert Ca²⁺-channel blocking activity in the posterior parts of the eye without inducing systemic effects, we investigated ocular and periocular penetration of topically instilled iganidipine in the pigmented rabbit eye. 

After topical instillation, iganidipine was rapidly absorbed into the cornea. Twenty minutes after a single instillation of 1% timolol, the concentrations of timolol in the cornea and aqueous humor were reportedly 885 × 10⁻⁸ and 63.2 × 10⁻⁸ M, respectively, ²² and 60 minutes after a single instillation of 2% carteolol, the concentrations of carteolol in the cornea and aqueous were 263 × 10⁻⁸ and 17.1 × 10⁻⁸ M, respectively. ²³ In contrast, iganidipine concentrations in the cornea and aqueous humor 30 and 60 minutes after a single instillation of 0.03% iganidipine were 157 × 10⁻⁸ and 2.67 × 10⁻⁸ M (30 minutes) and 127 × 10⁻⁸ and 3.34 × 10⁻⁸ M (60 minutes), respectively, suggesting that intraocular penetration of iganidipine is several times higher than that of timolol or carteolol, given the same molar concentration of topically instilled solution. Because the molecular mass of iganidipine (599.6 daltons [D]) is approximately twice that of timolol (316.4 D) or carteolol (292.4 D), this higher penetration is probably due to the higher lipid solubility of iganidipine versus timolol or carteolol. The log partition coefficient, an index of distribution into cellular membrane, of iganidipine is 3, which is attributable to the fact that a significant portion of iganidipine is not dissociated at higher pH (>5). In contrast, those of timolol and carteolol are −0.16 and −0.81, respectively. ²⁴ The equivalent iganidipine concentration was higher in the ICB than in the aqueous humor in the treated eye. It was also higher in the retina-choroid (both eyes) and in the ICB (vehicle-treated eye) than that in the plasma. These findings suggest that iganidipine binds to uveal pigments. 

To investigate whether topically instilled iganidipine reaches the ipsilateral posterior retina at pharmacologically active concentrations, through local penetration and not through systemic circulation, we compared the intravitreous ET-1-induced constriction of retinal arteries between drug- and vehicle-treated fellow eyes after a single instillation and a twice-daily 20-day regimen. Constriction of retinal arteries just after intravitreous injection of ET-1 should be due to the effect of ET-1 on the retinal arterial smooth muscles, because the concentration of ET-1 should be higher in the vitreous-retina interface than that in the optic nerve, orbit, or plasma. 

ET-1 activates ET-1 receptors on the cellular membrane of vascular smooth muscle, which stimulates entry of Ca²⁺ ions through Ca²⁺ channels and the release of Ca²⁺ ions from the sarcoplasmic reticulum, which leads to subsequent muscular contraction by activating myosin light chain kinase. ^{25 26} If there is a pharmacologically active concentration of a Ca²⁺ channel blocker in the retina, ET-1-induced contraction of smooth muscle cells should be partly inhibited. After a single instillation of iganidipine, there was no difference in the time course of retinal artery constriction between the drug- and vehicle-treated eyes at intravitreous doses of ET-1 of 2.5 or 0.5 ng. After twice-daily 20-day instillation, however, the ET-1-induced constriction of retinal arteries was significantly less on the drug-treated side than on the vehicle-treated side. At a higher intravitreous ET-1 dose (12.5 ng), there was no significant difference in the time course of retinal artery constriction between the drug- and vehicle-treated fellow eyes, and at a low intravitreous ET-1 dose (0.1 ng), there was no significant constriction on either side. This finding indicates that topically instilled iganidipine, a Ca²⁺ antagonist, reached the posterior retina in pharmacologically active concentrations through local penetration rather than through systemic circulation at least 30 minutes after the last instillation of a twice-daily 20-day regimen. In a separate group of rabbits, we measured retinal artery constriction induced by intravitreous injection of ET-1 at a dose of 0.5 ng after slow intravenous injection of various doses of iganidipine. After intravenous injection of vehicle or 10 μg/kg iganidipine, ET-1-induced constriction of the retinal arteries was not different from that observed in the vehicle-treated eyes (Fig. 2c) , whereas after intravenous injection at a dose 30 μg/kg, constriction was significantly inhibited, just as was observed in the iganidipine-treated eyes in experiment 2. The free iganidipine concentration in the plasma during the first hour was between 1.9 × 10⁻⁸ and 0.7 × 10⁻⁸ M, which suggests that the iganidipine concentration in the ipsilateral posterior retina 30 minutes after the last instillation of the twice-daily 20-day unilateral treatment with 0.03% solution was approximately 1.0 × 10⁻⁸ M or higher and that in the vehicle-treated eye was approximately 0.3 × 10⁻⁸ M [(10 μg/30 μg) × 1.0 × 10⁻⁸ M] or lower. Thus, the iganidipine concentration in the posterior retina due to local penetration in experiment 2 was approximately 0.7 × 10⁻⁸ M [(1.0 − 0.3) × 10⁻⁸ M] or higher. The route by which topically instilled iganidipine reached the ipsilateral posterior retina remains unknown. The route from the anterior chamber through the vitreous seems unlikely, because the vitreous concentration was negligible, even after twice-daily 14-day instillation, as determined as in experiment 1 (data not shown). Diffusion from the anterior chamber through the anterior and posterior choroid also seems unlikely, because of the relatively long distance. As discussed later, the route from the posterior periocular tissues cannot be excluded, but is difficult to determine from the present results. Intravitreous 0.5-ng ET-1-induced constriction of retinal arteries in the vehicle-treated eye was significantly less after a twice-daily 20-day instillation regimen than after a single instillation (49.2% ± 11.8% vs. 9.6% ± 5.8%, P < 0.039). This is probably due to differences in the plasma concentration of iganidipine. After a single instillation, plasma iganidipine was less than 10⁻⁹ M (Fig. 2) , but was approximately 3 × 10⁻⁹ M after twice-daily 14-day instillation (Isaka M, unpublished data of Senju Pharmacological Co, 1999). 

Twice-daily 20-day instillation of betaxolol did not significantly inhibit the constriction induced by the intravitreous injection of ET-1 at a dose of 2.5 or 0.5 ng. Osborne et al. ²⁷ reported that betaxolol instilled eight times at 30-minute intervals reaches the retina at a concentration of 1.4 × 10⁻⁶ M in rabbits. The difference between iganidipine and betaxolol was probably due to differences in the potency of the drugs’ Ca²⁺-antagonistic activities. 

The in vivo method of estimation of the iganidipine concentration in the posterior retina used in the present study has two advantages over the method of isolating the retina only and determining the drug concentration. The current method avoids possible contamination during tissue dissection and can estimate the concentration of the pharmacologically active drug rather than the whole drug or metabolites in the tissue. ²⁷  

In contrast, Yu et al. ²⁸ reported that betaxolol at concentrations of 1 × 10⁻¹⁰ M or higher significantly inhibits ET-1-induced (1 × 10⁻⁹ M) constriction in isolated porcine or human retinal arteries. The results obtained for iganidipine in experiment 2 suggest that the betaxolol concentration in the ipsilateral posterior retina is on the order of at least 1 × 10⁻⁹ M after twice-daily 20-day instillation. The discrepancy between the present results for betaxolol and those of Yu et al. may be due at least partly to the difference in the species used (rabbits versus pigs or humans), experimental design (in vivo versus isolated vessels), and/or route by which betaxolol reached the vascular smooth muscle cells (from outside the vessel versus from inside the vessel). 

Sponsel et al. ²⁹ reported that the periocular tissue in patients with glaucoma undergoing long-term topical administration of timolol or betaxolol accumulates a substantial amount of a β-antagonist (5.4–11.1 μg/mg), to a level several times higher than the maximal intraocular concentration. They also suggested the possibility that a topically instilled drug penetrates to retrobulbar space just posterior to the eye through periocular connective tissue and exerts pharmacologic effects on short posterior ciliary arteries that supply the ONH and choroid. ^{17 18} To investigate whether this applies to topically instilled iganidipine, whole-head semimicro autoradiographs were made after a single instillation of ¹⁴C-iganidipine. The rapid penetration of topically instilled iganidipine to the ipsilateral retrobulbar periocular tissues was an unexpected finding. 

The PSL per square millimeter in the extraocular muscle and posterior periocular tissues in the treated side were consistently higher than those in the untreated side at all time points. There was a similar tendency in the harderian gland, but relatively higher concentrations in the untreated side (3.9 × 10⁻⁸ M) suggest that iganidipine was deposited from the plasma. The unchanged ratio of iganidipine in intraocular tissues (aqueous humor, ICB and retina-choroid) was more than 70% 15 or 60 minutes after a single instillation in rabbits (Isaka M, unpublished data of Senju Pharmacological Co, 1999). The present study suggests that topical instillation of 0.03% iganidipine locally penetrates to the posterior periocular tissue at a concentration of approximately 0.8 × 10⁻⁸ M {0.7 × [(3.9–4.3) − (0.5–0.9)] × 0.03/0.09} at 30 minutes, a concentration higher than the IC₅₀ of iganidipine determined in isolated arteries (≈10⁻¹⁰ M). ⁹ This concentration was determined after a single instillation; after repeated instillation, the level may be higher. Topical iganidipine instillation has the potential to dilate short posterior ciliary arteries by local penetration to posterior periocular tissues. 

In conclusion, the current experiments demonstrated that topically instilled iganidipine, a potent Ca²⁺ antagonist, locally penetrates to the ipsilateral posterior retina or retrobulbar periocular tissue at concentrations higher than the IC₅₀ obtained in experiments in vitro in rabbits. According to Hughes, ³⁰ the axial length of rabbit eye is 18 mm, and, according to Sarnat and Shanedling, ³¹ the orbital volume in the rabbit is approximately 6 mL ³¹ ; both parameters are significantly smaller than those of human adults. Thus, in humans, the level of iganidipine that locally penetrates the retrobulbar periocular tissues or posterior parts of the eye would be lower than that determined in the rabbits in this study. Nevertheless, the present results suggest the possibility that a Ca²⁺ antagonist can be locally delivered to the posterior parts of the eye or to the retrobulbar periocular tissues at pharmacologically active concentrations by topical instillation in humans. This may also be the case with some other topical drugs. 

 Supported in part by Grant-in-Aid (B) 01870349 for Developmental Science Research, from the Ministry of Education, Science, Sports, and Culture of Japan.

 Submitted for publication May 20, 2002; revised August 23, 2002; accepted September 20, 2002.

 Commercial relationships policy: E (HS, MW); F (all others).

 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.

 

Chemical structure of iganidipine.

Figure 2.

 

Time course of equivalent change in concentration of iganidipine after a single instillation of 0.03% solution. ( ) Plasma, (•) cornea (treated-eye), (▪) ICB (treated eye), (□) ICB (vehicle-treated eye), (★) aqueous humor (treated-eye), (▾) lens (treated-eye), (+) vitreous (treated-eye), (▴) retina-choroid (treated-eye), (▵) retina-choroid (vehicle-treated eye), (♦) extraocular muscle (treated-eye), (⋄) extraocular muscle (vehicle-treated eye). Bars indicate SE of mean results in three rabbits.

Table 1.

 

PSL in the Extraocular Muscles (Experiment 4) and Ratios of the Equivalent Concentration of Iganidipine Obtained (Experiment 1) in Them at 15, 30, and 60 Minutes

	Treated Side					Untreated Side
		PSL/mm2	Equivalent Concentration* (10−8 M)	Ratio, †	PSL/mm2	Equivalent Concentration* (10−8 M)	Ratio, †
15 minutes	4.08	4.33	1.07	1.17	0.83	0.71
30 minutes	1.85, ‡	2.34	1.26	0.80, ‡	0.83	1.04
60 minutes	1.18	1.67	1.42	0.54	0.67	1.24

Figure 8.

 

A rabbit whole-head slice showing areas (boxes and triangles) in which PSL per millimeter square was measured. A, area between eye wall and extraocular muscle; Ao, area around optic nerve insertion; Ex, extraocular muscle; H, harderian gland.

Figure 3.

 

Time course of change in normalized diameter of retinal artery after intravitreous injection of ET-1 (a: 0.5 ng; b: 0.1 ng) in iganidipine-treated (▪) or vehicle-treated (•) eyes (n = 6). Each data point represents the average percentage of normalized diameter to that at −5 minutes, with the bar denoting SE. (↓) Time point of ET-1 intravenous injection.

Figure 4.

 

Time course of change in normalized diameter of retinal artery after intravitreous injection of ET-1 (a: 5 ng; b: 2.5 ng; c: 0.5 ng; d: 0.1 ng) in iganidipine-treated (▪) or vehicle-treated (•) eyes (n = 6). Each data point represents the average percentage of normalized diameter to that at −5 minutes, with the bar denoting SE. (↓) Time point of ET-1 intravenous injection.

Figure 5.

 

Time course of change in normalized diameter of retinal artery after intravitreous injection of ET-1 (a: 2.5 ng; b: 0.5 ng) in betaxolol-treated (□) or vehicle-treated (○) eyes (n = 6). Each data point represents the average percentage of normalized diameter to that at −5 minutes, with the bar denoting SE. (↓) Time point of ET-1 intravenous injection.

Figure 6.

 

Time course of intravitreous ET-1 induced (0.5 ng) change in normalized diameter of the retinal artery after slow intravenous injection of iganidipine at a dose of 10 μg/kg (▴; n = 3) or 30 μg/kg (▪; n = 6) or of vehicle solution (•; n = 6). Each data point represents the average percentage of normalized diameter to that at −5 minutes, with a bar denoting SE. (↓) Time point of ET-1 intravenous injection.

Table 2.

 

Free Iganidipine Concentration in Plasma after Slow Intravenous Injection

Time*	Blood Level (10−8 M)
15 minutes	1.85 ± 0.06
30 minutes	1.06 ± 0.01
1 hour	0.69 ± 0.30

Figure 7.

 

A whole-head slicethrough the rabbit optic nerve heads (a) and a semimicro autoradiograph after topical instillation of ¹⁴C-iganidipine (0.09%) (b) 15, (c) 30, and (d) 60 minutes after instillation. CR, cornea; Ex, extraocular muscle; H, the harderian gland; I, iris; L, lens; O, optic nerve; R, retina-choroid; S, sclera; V, vitreous; Vs, venous sinus.

Table 3.