Norrin attenuates protease-mediated death of transformed retinal ganglion cells

Materials

Dulbecco’s modified Eagle’s medium (DMEM), Dulbecco’s phosphate buffered saline (DPBS), penicillin, and streptomycin were obtained from Invitrogen Corporation (Carlsbad, CA). Staurosporine was obtained from Alexis Biochemicals (San Diego, CA). Human glu-plasminogen (product #410) and human fibrinogen (product #431) were obtained from American Diagnostica (Stamford, CT). Recombinant Dkk1 was obtained from R&D systems (Minneapolis, MN) and 3-[4,5-dimethylthiazol-2-yl]-2,5-diphenyltetrazolium bromide (MTT) was obtained from Sigma Chemical Company (St. Louis, MO).

Cell culture

Transformed RGC-5 cells were cultured in DMEM containing 1 g/l glucose, 10% fetal bovine serum (FBS), 100 U/ml penicillin, and 100 μg/ml streptomycin. RGC-5 cells (from passage 10–20) were treated with 2.0 μM staurosporine to induce their differentiation as described previously [10,11]. Briefly, cells were cultured overnight in DMEM containing FBS. The next morning, cells were washed three times with phosphate buffered saline (PBS; 3.2 mM, Na₂HPO₄, 0.5 mM KH₂PO₄, 1.3 mM KCl, 135 mM NaCl, pH 7.4) and incubated in serum-free medium supplemented with 2.0 μM staurosporine. Where indicated, cells were also treated with SS+norrin, SS+Dkk1, and SS+H-89. Cells morphology was observed by using an inverted, phase contrast, and bright-field microscope, and digitized images were obtained by using a Nikon D100 digital camera (Nikon Corporation, Tokyo, Japan).

Cell viability

Cells plated at 4×10³ cells/ml in 96 well tissue culture plates were left untreated or treated for 48 h with 2.0 μM SS along with indicated concentrations of norrin, Dkk1, and H-89 in serum-free DMEM. At 48 h after the treatment, cell viability was determined by incubating cells with 1.2 mM MTT for 2 h at 37 °C. At the end of 2 h, the formazan product produced by viable cells was dissolved in 0.01 M HCl, and the optical density was read at 570 nm by using an automated microplate reader (Bio-Tek Instruments, Inc., Winooski, VT). Statistical significance was analyzed by nonparametric Newman-Keuls analog procedure (GB Stat Software, Dynamic Microsystems, Silver Spring, MD).

Activation of Wnt pathway

Accumulation of beta-catenin in the nucleus, an indicator for activation of Wnt pathway, was measured by western blot analysis [12]. Briefly, RGC-5 cells (2×10⁶) were plated in 100 mm tissue culture plates and incubated overnight at 37 °C. Afterwards, cells were washed three times with PBS and cells were either left untreated or treated for 24 h with 25 ng/ml norrin, SS (2.0 μM), or SS (2.0 μM) and norrin. At the end of treatment, cells were washed three times with PBS, scraped into 1 ml PBS, and transferred into individual Eppendorf tubes. Cells were centrifuged for 1 min at 16.1x g and the supernatants were discarded. Cells were then resuspended in 0.4 ml lysis buffer, composed of 10 mM HEPES, pH 7.9, 10 mM KCl, 0.1 mM EDTA, 0.1 mM EGTA, and 1 mM dithiothreitol, along with protease inhibitors (complete mini protease inhibitor cocktail, Roche Diagnostics). Cells were allowed to swell on ice for 20 min followed by the addition of 12.5 μl 10% Nonidet P-40. The tubes containing cells were vortexed vigorously for 10 s, and the homogenates were centrifuged at 16.1x g for 30 s. Supernatants containing cytoplasmic proteins were collected into fresh Eppendorf tubes. The nuclear pellets were resuspended in 25 μl ice-cold nuclear extraction buffer, composed of 20 mM HEPES, pH 7.9, 400 mM NaCl, 1 mM EDTA, 1 mM EGTA, 1 mM dithiothreitol, and protease inhibitors, then incubated on ice for 30 min with intermittent vortexing. Samples were centrifuged for 5 min at 4 °C (16.1x g), and the supernatants containing nuclear proteins were transferred into fresh Eppendorf tubes. Protein content was measured by using BCA protein assay reagent (Pierce, Rockford, IL) and aliquots containing equal amount of nuclear and cytoplasmic proteins (20 mg) were separated on 10% SDS-polyacrylamide gels. After separation, the proteins were transferred to polyvinylidene fluoride (PVDF) membranes and incubated for 1 h in 10% nonfat dry milk prepared in Tris-buffered saline containing 0.1% Tween-20 (TBS-T). PVDF membranes were incubated overnight with polyclonal antibodies against beta-catenin (Cell Signaling Technology, Danvers, MA). Membranes were washed with TBS-T and incubated for 1 h at room temperature with appropriate secondary antibodies conjugated to horseradish peroxidase (HRP). Finally, the proteins were detected by using an ECL chemiluminescence kit (Pierce, Rockford, IL). Equal loading of nuclear and cytoplasmic proteins was determined by re-probing the membranes with antibodies against TATA-binding protein (Abcam, Boston, MA) and actin, respectively. Relative level of beta-catenin was determined by using image analysis software (Scion Corporation, Frederick, MD), and the results were represented as mean arbitrary units. Statistical significance was analyzed by nonparametric Newman-Keuls analog procedure (GB Stat Software, Dynamic Microsystems).

Phosphorylation of LRP-1

RGC-5 cells (2×10⁶ cells) plated in 100 mm tissue culture plates were either left untreated or were treated 24 h with 2.0 μM SS and indicated concentrations of norrin and H-89. At the end of 24 h, cells were washed twice with DPBS. Cells were scraped into 2 ml DPBS, transferred into pre-cooled 15 ml centrifuge tubes, and centrifuged at 13.4x g for 10 min at 4 °C. The supernatant was discarded and 500 μl of RadioImmuno Precipitation Assay (RIPA) buffer (1% nonidet-P40, 20 mM Tris-HCl, 150 mM NaCl, 1 mM Na₃VO₄, pH 7.4) buffer was added to each tube, and the tubes were incubated at 4 °C for 30 min on rotary shaker. Thirty min after incubation, cells were centrifuged and supernatants were collected into 1.5 ml Eppendorf tubes. Protein content was estimated by using Bio-Rad protein assay. Aliquots containing 500 μg total proteins from each treatment were incubated overnight 4 °C with 2 μl polyclonal antibodies against lipoprotein-related receptor-1 (LRP-1; Orbigen, San Diego, CA). Next, 100 μl Protein G-Sepharose (Invitrogen Corporation) was added to each sample, and samples were incubated at 4 °C for 4 h. Tubes containing Protein G-Sepharose were centrifuged for 2 min at 13.4x g and the supernatant was discarded. SDS-loading buffer (25 μl) was added to the pellet, and the tubes containing the pellet were boiled for 5 min. Tubes were centrifuged again to remove sepharose beads, and the supernatant was loaded onto 10% SDS–PAGE gels. After separation, the proteins were transferred to PVDF membranes and incubated for 1 h in 10% nonfat dry milk prepared in TBS-T. PVDF membranes were then probed with polyclonal antibodies against phosphoserine (Cell Signaling), phosphotyrosine (clone 4G10; Millipore, Billerica, MA), and 1:1,000 dilution LRP-1 (Orbigen). After overnight incubation with primary antibodies, membranes were washed twice with TBS-T and incubated for 1 h at room temperature with appropriate secondary antibodies conjugated to HRP. Finally, the proteins on the membranes were detected by using an enhanced chemiluminescence kit (Pierce). Relative level of proteins was determined by using Scion image analysis software, and the results from 3 independent experiments were represented as mean arbitrary units. Statistical significance was analyzed by non-parametric Newman-Keuls analog procedure (GB Stat Software, Dynamic Microsystems).

Zymography

Proteolytic activity of tPA and uPA in RGC-5 cells was determined by plasminogen and fibrinogen zymography assays as previously described [10,11]. Briefly, aliquots containing 20 μl conditioned medium were mixed with 4X SDS gel-loading buffer and loaded onto 10% SDS polyacrylamide gels containing 5.5 mg/ml fibrinogen and 50 μg/ml plasminogen. After electrophoresis, the gels were washed three times with 2.5% Triton-X 100 (15 min each time) to remove SDS from the gels, and the gels were incubated overnight in 0.1 M glycine-buffer (pH 8.0) at 37 °C. After overnight incubation, the gels were stained with 0.2% Coomassie brilliant blue-R250 for 5 min and then destained with a solution containing 50% methanol and 10% acetic acid in deionized water. Proteolytic activity of tPA and uPA, evident as clear bands in the zymograms, was documented using the Kodak EDAS 290 imaging system (Eastman Kodak Company, Rochester, NY). Relative level of protease activity was determined by using image analysis software (Scion Corporation), and the results from 3 independent experiments were represented as mean arbitrary units. Statistical significance was analyzed by non-parametric Newman-Keuls analog procedure (GB Stat Software, Dynamic Microsystems).

Effect of norrin on expression of tPA and uPA, and on death of RGC-5 cells

Throughout this study, RGC-5 cells were treated with 2.0 μM staurosporine to investigate the effects of norrin. This particular concentration was chosen because RGC-5 cells synthesize elevated levels of tPA and uPA at this concentration, and elevated levels of tPA and uPA, in turn, cause death of RGC-5 cells [10,11]. Cells were left untreated or treated for 48 h with varying concentrations of norrin alone, SS (2.0 μM) alone, or SS (2.0 μM) and varying concentrations of norrin. At the end of 48 h, conditioned medium was collected and proteolytic activity of tPA and uPA was determined by plasminogen/fibrinogen zymography assays. Zymography assays indicated that RGC-5 cells secreted low levels of uPA constitutively (Figure 1A). RGC-5 cells treated with increasing doses of norrin also secreted low levels of uPA similar to the levels secreted by untreated cells (Figure 1A,B) and the levels were not significantly different from untreated cells. In contrast, compared to untreated RGC-5 cells, cells treated with SS alone or SS and norrin secreted elevated levels of both uPA and tPA (Figure 1C), and the levels were significantly different from untreated cells (Figure 1C,D). Comparison of proteolytic activity (Figure 1B,D) with cell survival (Figure 1E,F) indicates that the survival of RGC-5 cells was significantly decreased under SS-treated conditions, but not under combined SS- and norrin-treated conditions despite the fact that RGC-5 cells synthesized similar levels of tPA and uPA under both treatment conditions.

To investigate the time course effect of norrin on cell survival, we treated RGC-5 cells for 24 and 48 h with 2.0 μM SS with or without 25 ng/ml norrin. From this experiment forward, we chose 25 ng/ml norrin to treat the cells because norrin seems to elicit optimum effect on the expression of tPA and uPA, and on cell survival at this concentration (see Figure 1F). At the end of the treatment period, levels of tPA and uPA, and cell survival were investigated as described in the previous section. Zymography assays indicated that, regardless of the presence of norrin, RGC-5 cells synthesized similar levels of uPA at 24 h after treatment (Figure 2A,B) and both tPA and uPA at 48 h after treatment (Figure 2D,E). Cell viability assays indicated that low levels of uPA had no effect on cell survival at 24 h after treatment (Figure 2C). In contrast, increased levels of both tPA and uPA reduced the survival of RGC-5 cells at 48 h after treatment (Figure 2F). In addition, despite elevated levels of tPA and uPA, a significant number of RGC-5 cells survived in the presence of norrin (Figure 2F).

Effect of norrin on Wnt-pathway activation

To investigate whether Wnt pathway plays a role in norrin-mediated protective effect, we employed western blot analysis to determine the translocation of beta-catenin into the nucleus. Western blot analysis performed on nuclear extracts prepared from RGC-5 cells indicated that a low level of beta-catenin was constitutively present in nuclear proteins extracted from cells left untreated or treated with norrin (Figure 3A). In contrast, in the presence of norrin, a significant increase in beta-catenin levels was observed in nuclear proteins extracted from cells treated with norrin or both SS and norrin (Figure 3A,B) indicating that norrin, indeed, activates Wnt-beta-catenin pathway. Comparison of nuclear translocation of beta-catenin with cell viability (Figure 3C) indicated that increased survival of RGC-5 cells under combined norrin- and SS-treated conditions associated with nuclear translocation of beta-catenin.

To determine whether inhibition of Wnt-beta-catenin pathway reduces norrin's protective effect, RGC-5 cells were left untreated or treated with norrin, Dkk1 (a potent inhibitor for Wnt-beta-catenin pathway; 0.25-4.0 μg/ml) [13] or Dkk1 and norrin. In addition, cells were treated with either SS only, SS and norrin, SS and Dkk1, or combined SS, norrin, and Dkk1. Results presented in Figure 3D indicate that in the absence of SS, Dkk1 had no effect on survival of RGC-5 cells regardless of norrin’s presence. In contrast, in the presence of SS, norrin enhanced survival of RGC-5 cells (Figure 3E), consistent with results presented in Figure 1F. When cell survival under combined SS- and norrin-treated conditions was compared with combined SS-, norrin-, and Dkk1-treatment, although cell survival was reduced to a large extent with Dkk1, cell survival was not reduced to a level observed under SS alone-treated conditions (Figure 3E). These results suggest that beta-catenin activation alone is insufficient to account for norrin-mediated protective effect on RGC-5 cells.

Effect of norrin on phosphorylation of LRP-1

To investigate the mechanisms (other than Wnt/beta-catenin pathway) by which norrin attenuated tPA and uPA-mediated death of RGC-5 cells, we have focused our attention on LRP-1 because tPA and uPA induces death of RGC-5 cells by interacting with this cell surface receptor. First, we investigated whether norrin downregulates the expression of LRP-1. Our western blot studies indicated that RGC-5 cells left untreated or treated with either SS only or combined SS and norrin expressed similar levels of LRP-1 protein (data not shown). These results indicate that downregulation of LRP-1 is not responsible for norrin-mediated survival of RGC-5 cells.

Since recent studies indicated that phosphorylation status of the cytoplasmic domain of LRP-1 plays a crucial role in signal transduction [14-16], we have hypothesized that phosphorylation status of LRP-1 may underlie norrin-mediated survival of RGC-5 cells. To investigate this hypothesis, RGC-5 cells were left untreated or treated with norrin alone, SS alone, or SS and norrin. At the end of the treatment, proteins were extracted and immunoprecipitated by using antibodies against LRP-1. Immunoprecipitated protein complexes were then subjected to western blot analysis by using antibodies against pSer and pTyr residues. Western blot analysis indicated constitutive phosphorylation of LRP-1 at both Ser and Tyr-residues in cells left untreated or treated with norrin (Figure 4A,B). When cells were treated with SS, LRP-1 phosphorylation was reduced at Ser-residues, but not at Tyr-residues (Figure 4A,B). In contrast, Ser-phosphorylation of LRP-1, reduced by SS, was increased largely when cells were treated with SS and norrin (Figure 4A,B). These results indicate that norrin attenuates the effect of SS, in part, by regulating the phosphorylation of LRP-1. Comparison of phosphorylation of LRP-1 with tPA and uPA levels (Figure 4C,D) indicated that the phosphorylation status of LRP-1 does not affect the levels of tPA and uPA. However, comparison of phosphorylation of LRP-1 (Figure 4A) with MTT assays (Figure 4E) indicated that phosphorylation status of LRP-1 affected the survival of RGC-5 cells. Results presented in Figure 4 show that despite elevated levels of tPA and uPA (Figure 4C,D), a significant number of RGC-5 cells survived (Figure 4E) as long as LRP-1 was phosphorylated at both Ser and Tyr-residues (Figure 4A).

To identify the kinases responsible for LRP-1 phosphorylation, we have concentrated our efforts on protein kinase A (PKA) and protein kinase C (PKC) because these kinases have been shown to regulate the phosphorylation of LRP-1. Since SS (inhibitor for PKCs) reduced the phosphorylation of LRP-1 at Ser-residues in RGC-5 cells (see Figure 4A), we have ruled out the role of PKCs in LRP-1 phosphorylation under SS and norrin-treated conditions. To investigate whether PKA plays a role in norrin-mediated phosphorylation of LRP-1, RGC-5 cells were left untreated or treated with 5 μM H-89 (a PKA inhibitor) [16], norrin alone, or combined norrin and 5 μM H-89. In addition, RGC-5 cells were treated with SS only, SS and H-89, SS and norrin, or combined SS, norrin, and H-89. After the treatment, phosphorylation of LRP-1 was assessed by western blot analysis as described in the previous section. Western blot analysis indicated that LRP-1 was phosphorylated at Ser- and Tyr-residues when treated with norrin alone or norrin plus SS (Figure 5A,B), consistent with the results presented in Figure 4A. When LRP-1 phosphorylation was compared between norrin only or norrin and H-89-treated cells, Ser-phosphorylation of LRP-1 was reduced by about 15%–20%. In the presence of SS and norrin, H-89 led to an additional reduction in LRP-1 phosphorylation, suggesting that PKA plays a role in norrin-mediated phosphorylation of LRP-1.

Finally, to investigate the role of PKA on survival of RGC-5 cells, cells were left untreated or treated with 0.625–10 μM H-89 only, norrin only, or combined norrin and 0.625–10 μM H-89, with or without SS. At the end of the treatment cell viability was determined by MTT assays (Figure 5C,D). Results presented in Figure 5C indicate that 2.5−10 μM H-89 alone, in the presence or absence of norrin, significantly reduced the viability of RGC-5 cells, indicating that PKA, in part, plays a role in survival of RGC-5 cells under normal conditions. In addition, when survival of RGC-5 cells was compared between SS- and norrin-treated conditions, and combined SS-, norrin-, and H-89-treated conditions (Figure 5D), cell survival was further reduced under 2.5–10 μM H-89-treated conditions. These results indicate that PKA, in part, plays a role in norrin-mediated survival of RGC-5 cells.