Effective and sustained delivery of hydrophobic retinoids to photoreceptors

Abstract

Purpose: Delivery of hydrophobic compounds to the retina/RPE has been challenging. The purpose of this study was to develop an effective method for the sustained delivery of retinoids to rod and cone photoreceptors of young mice lacking a normal supply of 11-cis retinal.

Methods: Solubilized basement membrane matrix (Matrigel; BD Biosciences, San Jose, CA) loaded with 9-cis retinal was administered subcutaneously into Rpe65(-/-) mouse pups for assessment of delivery to rods and cones and to Rpe65(-/-)Rho(-/-) mouse pups for assessment of delivery to cones. Intraperitoneal injections of 9-cis retinal were used for comparison. Cone density and opsin localization were evaluated with immunohistochemistry. Cone opsin protein levels were assayed with immunoblots, and cone function was analyzed by electroretinography (ERG) recordings. Retinoid content was determined by high-performance liquid chromatography analysis of retinal extracts. Pigment levels were quantified in homogenized retinas by absorption spectroscopy before and after light exposure.

Results: Single administration of Matrigel loaded with 9-cis retinal to Rpe65(-/-) mice increased cone densities in all analyzed regions of the retina compared with mice treated using intraperitoneal delivery. Cone opsin levels increased to near wild-type levels. Similar treatment in Rpe65(-/-)Rho(-/-) mice increased b-wave ERG amplitudes significantly, indicating the maintenance of cone function. Matrigel was shown to continuously release 9-cis retinal for periods up to 1 week.

Conclusions: As a method for sustained drug delivery, subcutaneous administration using Matrigel proved more efficacious than intraperitoneal injection for in vivo delivery of retinoids to cone photoreceptors. These experiments are the first to show a sustained delivery of retinoids in mice and suggest a strategy for potential clinical therapeutic development.

Figure 1.

Total cone density increases in Rpe65^−/− animals with delivery of 9-cis retinal by Matrigel. Data are obtained from P30 dark-reared mice that began treatment at P10. (A) Schematic representation of retinal flat mount, with six regions analyzed for cone density spanning the dorsal ventral axis. FD, far dorsal; MD, mid-dorsal; CD, central dorsal; ONH, optic nerve head; CV, central ventral; MV, mid ventral; FV, far ventral. (B) Quantification of PNA-positive cells by region from Wt (n = 7), Rpe65^−/− mice treated with 9-cis retinal (0.25 mg/animal) using Matrigel (n = 7) or IP injection (0.05 mg/animal ×5) (n = 6) and control Rpe65^−/− mice (n = 6). Matrigel administration of 9-cis retinal significantly increased cone density at all analyzed regions of the retina compared with IP injection; *P < 0.01. Two-way ANOVA was used to test for differences in cone counts by treatment group and retinal region with post hoc pairwise t-tests used to compare individual groups. Differences in maximum trough to peak, between left and right eyes, were averaged for each treatment group and analyzed using ANOVA with Tukey's post hoc tests to compare individual groups. A two-sided P-value of 0.05 was considered significant. Bars indicate ±SD. (C) Images of flat-mounted retina labeled with FITC-conjugated peanut agglutinin lectin.

Figure 2.

M/L-opsin protein increases with Matrigel delivery of 9-cis retinal in Rpe65^−/− mice. Data obtained from Rpe65^−/− P30 animals. (A) Cross-section images of the central dorsal retina are shown for Wt (top) and Rpe65^−/− mice treated with 9-cis retinal using Matrigel (bottom). For control mice, 9-cis retinal was omitted from the Matrigel preparation (Matrigel-only; middle). Paraffin-embedded retinas were sectioned and stained with hematoxylin-eosin (left column); cryoprotected retinas were sectioned and stained with both M/L-opsin-specific primary antibody (red) and DAPI (blue) for nuclei (right column). OS, outer segment; IS, inner segment; ONL, outer nuclear layer; OPL, outer plexiform layer. (B) Western blot images for M/L-opsin (top) and β-actin (bottom) are shown for Wt and Rpe65^−/− Matrigel, IP, and control mice. (C) Densitometry band analyses are displayed as percentages of Wt. Matrigel alone and 9-cis retinal (RAL) intraperitoneally injected Rpe65^−/− mice displayed significantly decreased M/L-opsin protein levels compared with Wt; *P < 0.02. Black bar: Wt, n = 3. White bar: Matrigel alone, n = 3. Red bar: 9-cis retinal IP, n = 3. Blue bar: Matrigel plus 9-cis retinal, n = 3. Data are presented as mean ± SD and were analyzed by two-sample t-test, accepting a significance value of P < 0.05.

Figure 3.

Maintenance of cone function in Rpe65^−/−Rho^−/− mice with Matrigel delivery of 9-cis retinal. Rpe65^−/−Rho^−/− mice were treated with 9-cis retinal using Matrigel (0.25 mg/mouse at P10) or by IP injection (0. 5 mg/animal, beginning at P10, ×5) and were analyzed at P30 with single white flash (intensity range, 1.56–24.8 cd · s/mm²) ERG recordings under dark-adapted conditions. For control mice, 9-cis retinal was omitted from the Matrigel preparation (Matrigel-only). (A) Representative ERG traces from Matrigel/9-cis retinal (left), 9-cis retinal IP (middle), and control (right) animals. Dashed line: onset of flash. (B) Quantification of b-wave amplitudes from Matrigel/9-cis retinal (black, n = 5), 9-cis retinal IP (gray, n = 5) and control (dashed, n = 5) animals. Mice administered 9-cis retinal using Matrigel showed significantly increased cone responses from 2.48 to 24.8 cd · s/mm² flash intensities compared with those that did not receive 9-cis retinal therapy; *P < 0.05. Data are presented as mean ± SD and were analyzed by two-sample t-test, accepting a significance value of P < 0.05.

Figure 4.

Matrigel maintains a sustained delivery of 9-cis retinal to photoreceptors. P45 Rpe65^−/− animals were administered 9-cis retinal (0.25 mg/animal) with a single subcutaneous Matrigel or IP injection. For control animals, 9-cis retinal was omitted from the Matrigel preparation (Matrigel-only). Animals were reared under cyclic light until 2 days after treatment and then were dark reared for either 3 or 7 days, and the retinas were analyzed for extracted retinoids or isorhodopsin levels (4 eyes/sample). (A) Representative HPLC profile of retinoid extract from Rpe65^−/− mice dark reared for 3 days and administered 9-cis retinal using Matrigel. Peaks were identified by comparison with known standards. Detection was at 360 nm. (B) syn-9-cis Retinal oxime levels after 3 and 7 days of dark rearing in Rpe65^−/− mice administered 9-cis retinal with a single injection of Matrigel (black, n = 6), IP administration (gray, n = 6), or control (white, n = 6). Administration of 9-cis retinal with Matrigel significantly increased syn-9-cis retinal oxime levels compared with IP administration; *P < 0.002. Mice administered 9-cis retinal with Matrigel showed significantly increased syn-9-cis retinal oxime levels between 3 and 7 days of dark rearing; †P < 0.05. (C) Isorhodopsin levels from whole retinas of mice administered 9-cis retinal using Matrigel injection after 3 (n = 3) and 7 (n = 3) days of dark rearing. Analysis showed a significant increase in isorhodopsin levels between both time points, indicating sustained delivery of 9-cis retinal; øP < 0.01. Data are presented as mean ± SD and analyzed by two-sample t-test, accepting a significance value of P < 0.05.