Male Sprague Dawley rats (200-250 g) were obtained from Chungang Animal Laboratory (Korea). The aniamals were housed in cages in a standard barrier facility, and maintained on a 12-h light: 12-h dark cycle at 23℃. The research was conducted in accordance with the internationally accepted principles for laboratory animal use and care as found in the NIH guidelines (USA).

To collect samples, six rats were deeply anesthetized with chloral hydrate (Fluka, Switzerland) (375 mg/kg body weight, intra-peritoneal injection), and the eyes were enucleated. The anterior segments of both eyes were removed from each rat, and the retina of one eye was separated for Western blot analysis. The eyecup of the second eye was fixed by immersion in 10% formalin in 0.1 M phosphate buffered saline (PBS), pH 7.4, for 24 h before paraffin embedding.

The antibodies used in this study included mouse monoclonal anti-caveolin-1 (IgG1, clone 2297), and anti-caveolin-2 (IgG1, clone 65), all obtained from BD Biosciences (USA). Mouse monoclonal anti-beta-actin was obtained from Sigma (USA). The specificity of each monoclonal antibody to caveolin-1, or -2 has been well characterized; no cross-reactivity between antibodies has been detected [10,11,14].

The retinas and hearts from rats were separately homogenized in lysis buffer (40 mM Tris, 120 mM NaCl, 0.1% Nonidet 40, 2 mM Na₃VO₄, 1 mM PMSF, 10 µg/ml aprotinin, 10 µg/ml leupeptin) and then centrifuged to remove any insoluble material. Samples were mixed with loading buffer and electrophoresed under denaturing conditions by 12% SDS-polyacrylamide gel electrophoresis. The proteins were then electroblotted in transfer buffer to nitrocellulose membranes (Schleicher & Schuell, USA) for 2 h at 100 V. Residual binding sites on the membranes were blocked by incubation with 5% non-fat milk in Tris-buffered saline (TBS; 10 mM Tris-HCl, pH 7.4, and 150 mM NaCl) for 1 h. Membranes were then incubated for 1 h with mouse antibody against either caveolin-1, or caveolin-2. The blots were washed three times in TBS containing 0.1% Tween-20, and then probed with horseradish peroxidase-conjugated anti-mouse IgG (Vector, USA) for 1 h. The blots were developed using enhanced chemiluminescence reagents (Amersham, USA), according to the manufacturer's instructions. Heart tissue was used as a positive control for caveolin-1 and -2 [15]. After imaging, the membranes were stripped and reprobed using monoclonal anti-beta-actin antibody as the primary antibody (Sigma, USA).

Paraffin sections (5 µm) of rat eyecups were deparaffinized and pretreated with citrate buffer (0.01 M, pH 6.0) in a microwave oven for 2 min. After hydration, the sections were treated with 0.3% hydrogen peroxide in distilled water for 20 min to block endogenous peroxidase activity. After three washes in PBS, the sections were exposed to 10% normal horse serum and then incubated for overnight at 4℃ with mouse monoclonal antibodies against either caveolin-1, or caveolin-2 (all at 1 : 200 dilution). After three further washes, the sections were incubated with biotinylated second antibody, followed by formation of avidin-biotin peroxidase complexes using the Elite kit (Vector, USA). The peroxidase reaction was developed with a diaminobenzidine substrate (Vector, USA). Before mounting, the sections were counterstained with hematoxylin. In the negative control, no immunostaining was seen in sections where primary antiserum was omitted (Fig. 2B).

Western blotting revealed the presence of caveolin-1 (Fig. 1A), and caveolin-2 (Fig. 1B) in the retinas of adult rats. Caveolin-1 immunoreactivity was detected in a band with an approximate molecular weight of 22 kDa. The same band was also observed in rat heart extracts, and observed to run in parallel, confirming the specificity of the antibody for caveolin-1 (Fig. 1A). For caveolin-2, there were two bands typically detected in retinas, a major band of approximately 20 kDa, and a minor band of approximately 18 kDa (Fig. 1B).

Expression of caveolin-1 was detected in the internal limiting layer and optic nerve fiber layer (Fig. 2A). In the ganglion cell layer, caveolin-1 was weakly localized to the cell membranes and nuclei of ganglion cells (Fig. 2C). Caveolin-1 was diffusely expressed in the plexiform layer, which contained bipolar axons, ganglion cell dendrites, and amacrine cell processes (Fig. 2C). In the inner nuclear layer, caveolin-1 immunoreactivity was weakly detected in the cellular membranes and nuclei of amacrine cells, and bipolar cells (Fig. 2D). Caveolin-1 staining was diffuse in the outer plexiform layer, which contains axonal ends of photoreceptor cells, dendritic processes of the bipolar cells and processes of the horizontal cells (Fig. 2E). In the outer nuclear layer, weak caveolin-1 immunoreactivity was seen in photoreceptor cells, and in the external limiting membrane, but little was detected in inner segments of rods and cones (Fig. 2F).

The immunostaining pattern of caveolin-2 in the retina was largely the same as that of caveolin-1, but was much weaker (Fig. 3A). Caveolin-2 immunostaining was however, intensely detected in vessels (Fig. 3A). Also, caveolin-2 was diffusely stained in inner plexiform layer (Fig. 3B), inner nuclear layer (Fig. 3C), and outer plexiform layer (Fig. 3D). Caveolin-2 was immunostained in the processes of glial cells (Fig. 3A, arrowhead) and of identical Muller cells (Fig. 3C, arrow). Caveolin-2 immunoreactivity was very limited in retinal neuronal cells including the ganglion cells, amacrine cells, bipolar cells, horizontal cells, and photoreceptor cells (Fig. 3B-D).