Immunohistochemical localization of galectin-3 in the pig retina during postnatal development

Animals and tissue samples

The eyes of six-month-old young adult pigs (n=5) were obtained immediately after butchering in a local slaughterhouse. Two-day-old pigs were obtained from a local farm (n=3). These were killed by CO₂ inhalation. The anterior segment of the eyeball was removed, and the retinas were carefully dissected on ice. All the experiments were performed in accordance with the Guide for the Care and Use of Laboratory Animals in Jeju National University, Jeju, Korea.

For western blot analysis, retinal tissues were detached from the eyecups and kept in a deep freezer. For histology, the anterior segments of the eyes were dissected. The eyecups without anterior segments were fixed by immersion in 10% neutral-buffered formalin in phosphate-buffered saline (PBS; 137mM NaCl, 2.7mM KCl, 4.3mM Na₂PO₄, 1.4mM KH₂PO₄, pH 7.4) for 24 h. Chemicals for PBS were obtained from JUNSEI (Tokyo, Japan). Fixed tissues were dehydrated through a graded series of ethanol and embedded in tissue embedding medium (Paraplast®, Tyco healthcare group Lp, MA). The paraffin-embedded retinal tissues were cut 5μm on a rotary microtome (Leica, Nussloch, Germany).

Antibodies

Approximately 1 mg/ml rat anti-galectin-3 monoclonal antibody was purified by affinity chromatography from the supernatants of hybridoma cells (clone TIB-166^™, M3/38.1.2.8. HL.2; American Type Culture Collection, Manassas, VA). Mouse monoclonal anti-glial fibrillary acidic protein (GFAP; G-3893; Sigma-Aldrich, St. Louis, MO) was used to detect astrocytes while anti-glutamine synthetase (MAB302; Chemicon International, Temecula, CA) was employed to detect Müller cells [20]. A mouse monoclonal anti-β-actin antibody (A5441; Sigma-Aldrich) was used to detect β-actin labeling on western blots.

Western blotting

Western blot analysis of the pig retina was performed as reported previously, with minor modifications [21,22]. In brief, retinal tissues were thawed in lysis buffer that consisted of 20 mM HEPES (pH 7.2; Sigma-Aldhrich, St. Louis. MO), 1% Triton X-100 (Amresco, Solon, OH), 1% deoxycholate (Sigma-Aldhrich), 0.1% sodium dodecyl sulfate (SDS; Bio-rad, Hercules, CA), 150 mM NaCl (Junsei, Tokyo, Japan), 10 μg/ml leupeptin (Sigma-Aldhrich), 10 μg/ml aprotinin (Sigma-Aldhrich), and 1 mM phenylmethylsulfonyl fluoride (Sigma-Aldhrich). Tissues were given 10 strokes in a Dounce homogenizer before they were sonicated for 10 s. The homogenates were centrifuged at 4 °C and 19,340× g for 20 min, and the lysate supernatant was obtained. Next, 40 µg of the lysate were dissolved in the SDS sample buffer and boiled. The homogenized tissue samples were electrophoresed under denaturing conditions in 10% SDS-polyacrylamide gels. Then, the proteins were electrotransferred in transfer buffer to a nitrocellulose transfer membrane (Schleicher and Schuell; Keene, Germany) for 2 h at 100 V. The residual binding sites on the membrane were blocked by incubation for 1 h with 5% nonfat milk in Tris-buffered saline (TBS), which was composed of 10 mM Tris-HCl, pH 7.4, and 150 mM NaCl. The membranes were then incubated for 2 h either with 1:5,000 rat monoclonal anti-galectin-3 or 1:20,000 mouse monoclonal anti-glutamine synthetase (GS; Chemicon International) for 2 h. The blots were washed three times in TBS containing 0.1% Tween-20 and probed with horseradish peroxidase-conjugated anti-rat IgG (1:2,000 dilution; Santa Cruz Biotechnology, Santa Cruz, CA) for 1 h. The membrane blots were developed using an enhanced chemiluminescence reagent kit (ECL; Amersham Pharmacia Biotech, Sweden), according to the manufacturer’s instructions.

After imaging, the membranes were stripped and reprobed using monoclonal anti-β-actin antibody as the primary antibody (Sigma-Aldrich). The optical density (OD/mm²) of each band was measured with a scanning laser densitometer (GS-700; Bio-Rad, Hercules, CA) and was reported as the mean±SEM. The ratios of the density of the galectin-3 band to that of the β-actin band were compared using Molecular Analyst software (Bio-Rad). The results were analyzed statistically using one-way ANOVA (ANOVA) followed by the post hoc Student–Newman–Keuls t-test for multiple comparisons. In all cases, a p<0.05 was considered significant.

Immunohistochemistry

Paraffin sections (5 μm) of retinas were deparaffinized in xylene, rehydrated in a series of decreasingly diluted solutions of ethanol and pretreated with 0.01 M citrate buffer (pH 6.0) in a microwave for 3 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 goat serum (ABC Elite Kit; Vector Laboratories, Burlingame, CA) and then incubated for 1 h at room temperature (RT) with primary antibodies, including 1:500 monoclonal rat anti-galectin-3. After three washes, the sections were incubated with the secondary antibody, 1:100 biotinylated rabbit anti-rat IgG (Vector), and then avidin-biotin peroxidase complexes were formed using an Elite kit (Vector). The peroxidase reaction was developed with a diaminobenzidine substrate kit (Vector). Before sections were mounted, they were counterstained with hematoxylin. As a negative control, the primary antiserum was omitted for a few test sections in each experiment, and no specific labeling of cell bodies or fibers was detected in these sections.

For double immunofluorescence, the sections were incubated in the following order: with 10% normal goat serum for 1 h at RT, with either mouse monoclonal anti-GFAP (Sigma-Aldrich) or mouse monoclonal anti-glutamine synthetase (Chemicon) for 1 h, and then with fluorescein isothiocyanate (FITC)-labeled goat anti-mouse IgG (Sigma-Aldrich) for 1 h at RT. Then, the sections were washed and incubated with rat anti-galectin-3 overnight at 4 °C, after which the sections were washed and incubated with tetramethylrhodamine isothiocyanate (TRITC)-labeled goat anti-rat IgG (Sigma-Aldrich). To reduce or eliminate lipofuscin autofluorescence, we washed the sections in PBS. Three times for 1 h at RT, dipped them briefly in distilled H₂O, treated them with 10 mM CuSO₄ in ammonium acetate buffer (50 mM CH₃COONH₄, pH 5.0) for 20 min, then again dipped them in distilled H₂O before returning them to PBS. The double immunofluorescence-stained specimens were examined under an FV500 laser confocal microscope (400×; Olympus, Tokyo, Japan).

Western blot analysis of galectin-3 in pig retinas

Western blot analysis was used to detect galectin-3 (molecular weight, approximately 29–35 kDa) in the homogenates of retinas from two-day-old pigs and six-month-old pigs (Figure 1). Compared with the intensity of the galectin-3 immunoreactive band in the retinas of the two-day-old pigs (relative optical density: 1.0±0.13), that in the retinas of the six-month-old pigs was significantly increased (relative optical density: 5.91±1.98; p<0.05; Figure 1A,B). This finding suggests that galectin-3 is expressed in the retina of two-day-old pigs, and its expression increases during postnatal development to the level seen in six-month-old pigs. A western blot analysis of glutamine synthetase was performed in the same samples to observe changes during postnatal development. The results indicated that the level of glutamine synthetase in the six-month-old pig retinas was significantly greater than that in the two-day-old pig retinas (Figure 1C).

Differential localization of galectin-3 immunoreactiviy in pig retinas

In the retinas of two-day-old piglets, galectin-3 immunostaining was observed in glial processes in the innermost nerve fiber layer and the ganglion cell layer (Figure 2A). It was also detected in cell bodies in the inner nuclear layer and in a few processes in the outer limiting membrane, while there was little staining of ganglion cells in the ganglion cell layer and neuronal cells in the inner nuclear layer and outer nuclear layer. The pattern of galectin-3 immunostaining was similar between the retinas of six-month-old pigs and two-day-old pigs, with some exceptions. In the ganglion cell layer of the six-month-old pig retinas, the galectin-3 immunoreactivity was more intense and covered a broader area (Figure 2B). In the inner nuclear layer, we observed a striking increase in the intensity of galectin-3 immunoreactivity in glial cell bodies and their processes, and across all retinal layers, including the ganglion cell layer, inner plexiform layer, outer plexiform layer, and outer nuclear layer (Figure 2B), whereas galectin-3 immunoreactivity was very weak at these sites in two-day-old pig retinas.

A double-labeling experiment was performed to determine the cellular phenotype of galectin-3 expression in the pig retina. The results for six-month-old pig retinas are shown because the galectin-3 immunoreactivity was similar in both age groups. Galectin-3 was colocalized with glutamine synthetase, a marker for Müller cells [20], in the retinas of six-month-old pigs (Figure 3A-C). Most of the glutamine synthetase-positive elements overlapped with galectin-3, suggesting that Müller cells contain galectin-3 (Figure 3C, merged image). In addition, galectin-3 was also colocalized with GFAP (Figure 3D-F), suggesting that Müller cells also contain GFAP. We cannot exclude the possibility that astrocytes from the optic nerve contain galectin-3 in the innermost nerve fiber layer.