This article is mentioned in:

Introduction

One of the largest tyrosine kinase families, erythropoietin-producing hepatocyte receptors (Ephs) and their corresponding ligands, ephrins, are expressed in neighboring cells with reciprocal pattern (1). The interaction between receptors and ligands plays an important role in different kinds of diseases (2–8). Interactions between Eph receptors and NMDA receptors and/or metabotropic glutamate receptors have been reported to cause excitotoxic neuronal death in central nervous system (CNS) (8–13). Eph receptors have also shown to be important in both experimental glaucomatous animals and spontaneous glaucomatous DBA/2J mice (14,15). Our previous study has shown that interaction between ephrinB/EphB signaling and AMPA receptor subunits GluA2 contributes to retinal ganglion cells (RGCs) apoptosis in a rat chronic ocular hypertension (COH) model (16). Since Eph receptor signaling casts a wide net on cellular behavior, there are probably other pathways involved in RGCs injury induced by ephrinB/EphB signaling in COH model. The study on the optic nerve head (ONH) of the DBA/2J mice showed that EphB2-Fc increased intracellular Ca2+ concentrations in single RGCs axons (14). Although different kinds of Ca2+ permeable receptors which mediated Ca2+ overload have been indicated in the RGCs apoptosis in retina (17–19), elevated levels of Ca2+ concentrations through voltage-gated Ca2+ channels have been associated with diminished vision in healthy aging (20). Also, according to study on glaucoma prevalence, calcium oral intake without calcium channel antagonist can increase the risk for glaucoma (21,22). Based on these evidences, the aim of the present study was to investigate the role of calcium channels in a rat COH model and whether these channels can be regulated by ephrinB/EphB signaling.

Materials and methods

Animals and rat COH model

Male Sprague Dawley rats, that were 3 to 4 weeks of age and weighted 100–250 g, were obtained from Shanghai Laboratory Animal Center Laboratory Animal Co., Ltd. All the animals were housed in an environment with temperature of 22±1°C, relative humidity of 50±1% and a light/dark cycle of 12/12 h. Both food and water were provided ad libitum. All experimental procedures described were in accordance with the National Institutes of Health guidelines for the Care and Use of Laboratory Animals and the Guidelines of the Yangtze University on the Ethical Use of Animals. Care and use of animals were also approved by the ethics committee of First Affiliated Hospital of Yangtze University (Jingzhou, China).

COH rat model was established and validated following the previously described approach (16,23,24). Briefly, rats were anesthetized with a mixture of ketamine (25 mg/kg, im) and xylazine (10 mg/kg, im); eyes were locally anesthetized with topical 0.4% oxybuprocaine hydrochloride drops (Benoxil; Santen Pharmaceutical, Co., Ltd., Ishikawa, Japan). Three episcleral veins in the left eye were carefully separated and cauterized under an OPMI VISU 140 microscope (Carl Zeiss, Oberkochen, Germany). A sham operation, following a similar procedure (except for not occluding the vines), was conventionally performed on the eyes of other rats. After surgery, eyes were flushed with antibiotic eye drops and covered with antibiotic ointment. Intraocular pressure (IOP) was measured using a handheld digital tonometer (Tonopen XL; Mentor O&O, Norwell, IL, USA) under general and local anesthesia as described above. The average value of five consecutive acceptable measurements with a deviation <5% was recorded. All measurements were performed in the morning to avoid possible circadian differences. IOPs in both eyes were measured before surgery (baseline), immediately after surgery (day 0), the first day after surgery (G1 day), the third day after surgery (G3 days) and weekly thereafter.

Western blotting

Western blot analysis was conducted as previously described (16,23,24), with some modifications. Previous studies have demonstrated that Cav3.1, Cav3.2 and Cav3.3 subunits of T-type Ca2+ channels and the Cav1.2 subunit of L-type Ca2+ channels are mainly expressed in rat retinal Müller cells and RGCs layer (25). Since both ganglion cell layer and Müller cells are vulnerable to IOP, we examined Ca2+ channels subunits change in COH retinas using Western blot. Retinas were homogenized in RIPA lysis buffer (cat. no. 89900; Pierce; Thermo Fisher Scientific, Inc., Waltham, MA, USA) supplemented with protease and phosphatase inhibitor cocktail (cat. no. 88661&88662; Roche Applied Science, Mannheim, Germany). The concentration of total proteins was measured using a standard bicinchoninic acid assay kit (cat. no. 23227, Pierce; Thermo Fisher Scientific, Inc., Waltham, MA, USA). The extracted whole protein samples (50 µg) were resolved by 8% SDS-PAGE gel and electroblotted onto PVDF membranes (cat. no. ISEQ0001, Immobilon-P; EMD Millipore, Billerica, MA, USA) using Mini-PROTEAN 3 Electrophoresis System and Mini Trans-Blot Electrophoretic Transfer System (Bio-Rad Laboratories, Inc., Hercules, CA, USA). After being blocked in 5% nonfat milk at room temperature for 1.5 h, the membranes were incubated at 4°C overnight with the following primary antibodies: monoclonal mouse anti-β-actin (cat. no. A2228, 1:3,000 dilution; Sigma-Aldrich; Merck KGaA, Darmstadt, Germany), polyclonal rat anti-Cav3.1 (cat. no. ACC-021), anti-Cav3.2 (cat. no. ACC-025), anti-Cav3.3 (cat. no. ACC-009), anti-Cav1.2 (cat. no. CC-003) (1:200 dilution for all the three; Alomone Labs, Jerusalem, Israel). After being washed in Tris-buffered saline-Tween-20 (TBST) for three times (5–10 min per time), the blots were incubated with horseradish-peroxidase-(HRP-) conjugated donkey anti-mouse (Cat no. 715-035-151) or donkey anti-rabbit (cat. no. 715-035-151) secondary antibody (Jackson ImmunoResearch Laboratories, Inc., West Grove, PA, USA) for 2 h at room temperature. Blots were then washed in TBST for another three rounds, and incubated with chemofluorescent reagent (cat. no. 34080; Pierce; Thermo Fisher Scientific, Inc.) followed by exposure to X-ray film in a dark room. Experiments were performed in triplicate. The protein bands were quantitatively analyzed with NIH Image Analysis software (Image J, version 1.38x; National Institutes of Health, Bethesda, MD, USA). Protein levels were normalized to the corresponding β-actin levels.

Immunofluorescence analysis

As previously described (16,23), retinal sections were examined by immunofluorescence technique. Rats were anesthetized with ethyl carbamate (1.25 g/kg, ip; Sinopharm Chemical Reagent Co. Ltd., Shanghai, China) and transcardially perfused with 4% paraformaldehyde (PFA; in 0.1 M PB, pH 7.4). Eyes were post-fixed in 4% PFA for 2–4 h and then dehydrated with graded sucrose solutions at 4°C (4 h in 20% and overnight in 30% solutions). Retinas were vertically sectioned (14 µm; Leica Microsystems GmbH, Wetzlar, Germany), and sections were mounted on chrome-alum-gelatin-coated slides (Thermo Fisher Scientific, Inc.). Sections were then blocked for 2 h in 6% normal donkey serum, 1% bovine serum, and 0.2% Triton X-100, and dissolved in PBS at room temperature. Consequently, sections were incubated with the following primary antibodies at 4°C for 24 h: polyclonal rat anti-Cav3.1, anti-Cav3.2, (1:200 dilution; Alomone Labs, Jerusalem, Israel), Goat anti-CTB (cat. no. 703; List Biological Laboratories, Campbell, CA, USA), mouse anti-glial fibrillary acidic protein (GFAP, cat. no. G3893; Sigma-Aldrich; Merck KGaA). Binding sites of primary antibodies were visualized by incubating sections with: 488-conjugated donkey anti-rabbit secondary antibodies (cat. no. 711-546-152, 1:400; Jackson ImmunoResearch Laboratories, Inc.) for single Cav3.2 vision, and cy3-conjugated donkey anti-rabbit (cat. no. 711-165-152), cyanine conjugated donkey anti-goat (cat. no. 705-175-147), 488-conjugated donkey anti-mouse (cat. no. 715-545-150) were used to stain the co-localization of Cav3.2, CTB and GFAP. All the secondary antibodies were incubated for 2 h at room temperature. Finally, sections were visualized and photographed with a Leica SP2 confocal laser-scanning microscope. To avoid reconstruction stacking artifacts, double labeling was evaluated by sequential scanning on single-layer optical sections at 1.0 µm intervals.

CTB injection for retrograde labeling of RGCs

Retrograde labeling of RGCs was previously described in detail (23). Briefly, after anesthetizing rats with 40 mg/ml sodium pentobarbital (0.1 ml/100 g), 1% Cholera Toxin B subunit (CTB, cat. no. 104; List Biological Laboratories) was injected into the superior colliculus (6.0 mm posterior and 2.0 mm lateral to the bregma, and 4–5 mm deep from the cortical surface). After a survival period of 5–7 days, RGCs were clearly labeled for immunofluorescence analysis.

Clustering of EphB2-Fc

The EphB2-Fc (cat. no. 467-B2-200, 2 µg/µl; R&D Systems, Inc., Minneapolis, MN, USA) was clustered by a 40 min incubation at 37°C in buffer containing an AffiniPure goat anti-human IgG, Fc-specific (1:2.5 w/w) (Jackson ImmunoResearch, Milan, Italy) (26).

Intravitreal injection

Intravitreal injection was performed according to previously described method (16,24). When the pupil of the anesthetized eye was dilated with tropicamide drops, clustered EphB2-Fc (2 µl,), or Mibefradil (no. M5441, 3 mM; Sigma-Aldrich; Merck KGaA) dispersed in 2 µl of 0.9% saline, was injected into the vitreous space through a postlimbus spot using stereoscopic microscope (Carl Zeiss). A 30-gauge needle was then inserted 2 mm behind the temporal limbus and directed toward the optic nerve. Eyes of vehicle or negative control (Ctr) group were injected with saline or clustered human Ig-Fc (2 µg; R&D Systems, Inc.), respectively.

Cell apoptosis

To investigate cell apoptosis, terminal deoxynucleotidyl transferase-mediated biotinylated UTP nick end labeling (TUNEL) assay (16,23) was performed on whole flat-mounted retinas using the DeadEnd Fluorometric TUNEL System G3250 kit (Promega Corporation, Madison, WI, USA), following the manufacturer's instructions. TUNEL signals were visualized with a laser confocal scanning microscope through a 20× objective (FluoView 1000; Olympus Corporation, Tokyo, Japan). The retinas were mounted with the ganglion cell layer (GCL); a serial deep scanning was performed in the GCL according to the DAPI staining results. All TUNEL-positive signals that merged well with DAPI in each retina in GCL, were consequently counted (comparing with INL and ONL, the cell body in GCL was the largest).

Statistical analysis

The data derived from the TUNEL assay and the IOP monitoring were presented as mean ± SEM. For Western blot experiments, the expression level of a protein was first normalized to the corresponding β-actin levels. The relative expression levels were averaged for all samples. The mean values of the data obtained during different postoperational time periods were normalized according to the mean value of the control group. The data were presented as the mean ± SEM. A one-way ANOVA with LSD post hoc test (for protein analysis and RGCs apoptosis), or t test (paired data for IOP analysis) were used and P<0.05 was considered to indicate a statistically significant difference.

Results

Expression change of Ca2+ channels in rat COH model

The rat COH model, with increased IOP levels, was successfully established and validated according to previously described method (16). Briefly, significantly higher IOP was observed in COH rats compared to sham and unoperated group (COH: 25.0±0.7 to 28.2±1.2 mmHg, n=10–74; sham: 19.0±0.6 to 19.7±0.9 mmHg, n=10–74; unoperated eyes: 19.8±1.0 mmHg, n=9, all P<0.001).

As the summary data showed, the average intensity of Cav3.1 increased from G1 day until G3 weeks while that of Cav3.2 increased later from G1 week until G4 weeks (Fig. 1B and C). The protein level of Cav3.1 on G1 day was increased to 123.9±3.6% of control (n=4, P=0.006 and further increased to 208.0±4.3, 241.3±9.5, 224.4±10.6 and 257.7±13.5% of control at G3 days, G1 week, G2 weeks and G3 weeks respectively (all n=4, P<0.001) and then unexpectedly returned to the control level at G4 weeks (113.1±6.3% of control, n=4, P=0.109) (Fig. 1B). As for Cav3.2, the protein level didn't change much in the initial period of time (G1 day and G3 days) and started increasing in G1 week (164.7±3.4% of control, n=4, P=0.002) and remained at stable higher level thereafter, which were 175.7±3.2% of control at G2 weeks (n=4, P=0.001), 228.5±15.6% of control at G3 weeks (n=4, P<0.001), and 210.6±15.9% of control at G4 weeks (n=4, P<0.001). In contrast, the average intensity of Cav3.3 subunit of T-type Ca2+ channels and Cav1.2 subunit of L-type Ca2+ channels didn't change during the whole postoperational period.

Figure 1. Differences in the protein levels of Cav3.1, Cav3.2, Cav3.3 and Cav1.2 in the retinas of sham and COH rats. (A) Cav3.1, Cav3.2, Cav3.3 and Cav1.2 expression in sham-operated and COH retinal extracts at different time points (G1 day, G3 days, G1 week, G2 weeks, G3 weeks and G4 weeks). Bar charts summarizing the average densitometric quantification of immunoreactive bands of (B) Cav3.1 and (C) Cav3.2 expression during different time points. All data were normalized to control and are presented as the mean ± standard error of the mean. **P<0.01 vs. Ctr. Ctr, sham-operated control; G1 day, 1 day following surgery; G3 days, 3 days following surgery; G1-4 weeks, 1–4 weeks following surgery; Cav, calcium channel; COH, chronic ocular hypertension.

Immunofluorescence analysis further demonstrated that Cav3.1 and Cav3.2 subunit of T-type Ca2+ channels were expressed in different locations. Cav3.1 was mainly expressed in the outer plexiform layer (OPL) and cell membranes in inter nuclear layer (INL) (Fig. 2); during G1 day and G3 days, most of the Cav3.1 signal was located in the OPL (Fig. 2Ab and c), while later the signal was also found in the INL (from G1 week to G3 weeks, Fig. 2Ad-f). Cav3.2 was mainly located in GCL layer; the signal increased from G1 week through G4 weeks (Fig. 3Ad-g and Cd-g). Furthermore, all those data were consistent with western blot results. Moreover, since GCL layer was most vulnerable to IOP, we then verified whether RGCs and Müller cells were both positive for Cav3.2 expressions (Fig. 4A and E), by co-localization with CTB (Fig. 4B) and GFAP (Fig. 4D), respectively.

Figure 2. Cav3.1 expression in COH rats examined by immunofluorescence. (A) Cav3.1 expression profiles in rat retinal vertical slices taken from sham-operated retinas (Ctr; image a), and those obtained from COH rats at different time points (images b-g). (B) DAPI staining in retinas of control rats (image a), and those obtained from COH rats at different time points (images b-g). (C) Merged images of Cav3.1 and DAPI in retinas of control rats (a), and those obtained from COH rats at different time points (images a-g). Scale bars=20 µm. GCL, ganglion cells layer; IPL, inner plexiform layer; INL, inner nuclear layer; OPL, outer plexiform layer; ONL, outer nuclear layer; Ctr, sham-operated control; G1 day, 1 day following surgery; G3 days, 3 days following surgery; G1-4 weeks, 1–4 weeks following surgery; Cav, calcium channel; COH, chronic ocular hypertension.

Figure 3. Cav3.2 expression in COH rats examined by immunofluorescence. (A) Cav3.2 expression profiles in rat retinal vertical slices taken from sham-operated retinas (Ctr; image a), and those obtained from COH rats at different time points (images b-g). (B) DAPI signal in retinas of control (image a), and those obtained from COH rats at different time points (images b-g). (C) Merged images of Cav3.2 and DAPI in retinas of control (image a), and those obtained from COH rats at different time points (images b-g). Scale bars=20 µm. GCL, ganglion cells layer; IPL, inner plexiform layer; INL, inner nuclear layer; OPL, outer plexiform layer; ONL, outer nuclear layer; Ctr, sham-operated control; G1 day, 1 day following surgery; G3 days, 3 days following surgery; G1-4 weeks, 1–4 weeks following surgery; Cav, calcium channel; COH, chronic ocular hypertension.

Figure 4. Cav3.2 expression location in COH retinas (G2 weeks) by immunofluorescence analysis. (A) Immunofluorescence labeling for Cav3.2 in G2 week rat retinal vertical slices following an injection of CTB to the superior colliculus bilaterally. (B) Immunofluorescence labeling for CTB (positive for retinal ganglion cells) in the retinal vertical slices. (C) Immunofluorescence labeling for DAPI in the retinal vertical slices. (D) Immunofluorescence labeling for GFAP (positive for Müller cells) in the retinal vertical slices. (E) Merged images of Cav3.2, CTB, DAPI and GFAP in the retinal vertical slices. Scale bars=20 µm. Cav, calcium channel; COH, chronic ocular hypertension; G2 weeks, 2 weeks following surgery; GCL, ganglion cells layer; IPL, inner plexiform layer; INL, inner nuclear layer; CTB, Cholera Toxin B subunit; GFAP, glial fibrillary acidic protein.

Mibefradil reduced RGCs apoptosis in EphB2-Fc injected rat retina

In our previous study, we have shown that intravitreal injection of EphB2-Fc induces apoptosis in RGCs. Since Cav3.2 was increased in the RGCs and Müller cells in COH rats, in the present study we used T type Ca2+ channel blocker Mibefradil to further investigate whether calcium channels were involved in the process of EphB2-Fc induced apoptosis in RGCs. As Fig. 5 showed, average TUNEL signals that matched well with DAPI in GCL by EphB2-Fc injection were 497.4±25.3, significantly higher than control Fc treatment (17.2±2.9, n=5, P<0.001). Pretreatment of Mibefradil (2 µl, 3 mM), injected intravitreally 1 day before EphB2-Fc, reduced the TUNEL signals to 302.4±16.3 (n=5, P<0.001, compared with EphB2-Fc group) while saline pretreatment had no effect (505.2±26.1, n=5, P=0.64, compared with EphB2-Fc group).

Figure 5. Pretreatment with Mibefradil reduced EphB2-Fc induced apoptosis in retinal ganglion cells. (A) TUNEL signaling, (B) DAPI staining and (C) merged images in the retinas of the control, EphB2-Fc and Mebefradil with EphB2-Fc groups. Scale bars=50 µm. Ctr, sham-operated control; EphB2-Fc, erythropoietin-producing hepatocyte receptor B2-fragment crystallizable region.

Discussion

We have previously reported that elevated reverse EphB/ephrinB signaling contributes to RGCs apoptosis by interacting with GluR2 subunit of AMPA receptors in a COH rat model (16). By using the same ocular hypertension rat model, we found that Cav3.1 and Cav3.2 protein expression increased in retinas, but at different locations and at different timepoints. It is necessary to emphasize that Cav3.2 increased both in RGCs and Müller cells in COH retinas, which was similar to the expression pattern of ephrinB2, with the exception that ephrinB2 starts to increase from second week (16) while increment of Cav3.2 started in G1 week. Consequently, RGCs apoptosis caused by intravitreal injection of clustered EphB2-Fc could be partly relieved by Mibefradil, a T type Ca2+ channel blocker. Numerous studies have shown that disturbance of calcium homeostasis which occurs through several pathways contributes to neuronal damage in many neurodegenerative diseases (27–32). Moreover, the deregulation of Ca2+ channel activities, which is one of the pathways causing Ca2+ homeostasis disturbance, has very important role in neurodegenerative disease like ALS (33) although related reports are relatively rare and controversial (34). In glaucoma, like in many other neurodegenerative diseases, Ca2+ channel blockers have shown to be beneficial (reviewed by Chihiro Mayama in reference) (35). The most often tested Ca2+ channel blockers in experimental models are L type Ca2+ channel blockers, nifedipine and nimodipine. Their benefits include lowered IOP, improved blood flow to RGCs and neuron protection due to less Ca2+ influx. In the present study, we didn't test L type Ca2+ channel blockers' effect on the EphB2-Fc induced RGCs apoptosis, since no changes in L type Ca2+ channel expression were observed in COH retinas. Nonetheless, we didn't totally rule out the possibility that L type Ca2+ channel could play a role, for in our other separated preliminary experiments (data not shown here), clustered EphB2-Fc increased mixed Ca2+ currents in isolated Müller cells. If this was the same as with RGCs or whether it was related with Mibefradil's protection on RGCs needs to be further examined. Our results suggest that elevated EphB/ephrinB signaling regulates RGCs apoptosis under COH through more than one pathway, while calcium channels, i.e., T type calcium channel, might be a target for EphB/ephrinB signaling. Another important point that needs to be addressed is the way EphB/ephrinB signaling regulates calcium channels in rat retinas. It is well known that phosphorylation of calcium channels is important for the function. Previous studies have shown that L type calcium channel could be regulated by tyrosine kinase src (36–38). According to our previous findings, EphB/ephrinB signaling in retina could interact with src (16). Src kinase acts as a downstream of Ca2+/calmodulin-dependent protein kinase II (CaMK II) in intracellular free [Ca2+]i glomerular mesangial cells (39), while in striatum basal level of autophosphorylation of CaMK II it requires T type, but not L or N type calcium channels (40). Although, using high resolution MS approach several threonine and serine residues with no tyrosine phosphorylation have been detected in normal rat brain tissue (41), the situation in retina, especially under higher IOP remains an open question.

Acknowledgements

This study was supported by the Natural Science Foundation of Hubei Province (no. 2014CFB213), the Health and Birth Control Department of Hubei Province (no. 81000380/H1204) and the Jingzhou Science and Technology funding (2014AC47B). We thank the grants that supported our study and MedSci (www.MedSci.com) for its linguistic assistance during the preparation of this manuscript.

References