CaMKIIαB Mediates a Survival Response in Retinal Ganglion Cells Subjected to a Glutamate Stimulus

Abstract

purpose. During N-methyl-d-aspartate–induced cell death in the neural retina, levels of the nuclear isoform of CaMKIIα, CaMKIIαB, previously reported to be detected only in the midbrain and diencephalon, become elevated. The purpose of this study was to investigate whether CaMKIIαB is present specifically in retinal ganglion cells (RGCs) and to determine whether it can be implicated in the cell death or cell survival of signal transduction pathways.

methods. Pan-purified RGCs were obtained from the retinas of postnatal day (P)6 to P8 Sprague–Dawley rats. The expression level of CaMKIIαB was investigated in RGCs with the aid of RT-PCR and immunostaining under normal and glutamate-stressed conditions. siRNA targeted to CaMKIIαB was used to knock down the level of endogenous mRNA in RGCs, and cell viability was tested. The putative role of CaMKIIαB in the downstream expression of survival genes such as BDNF was evaluated in CaMKIIαB knocked-down RGCs with the aid of RT-PCR, real-time PCR, and immunofluorescence microscopy.

results. Basal levels of CaMKIIαB were expressed in RGCs. Expression levels became increased in response to glutamate treatment and were translocated to the nuclei after a glutamate stimulus. In pan-purified RGCs with knocked down levels of CaMKIIαB, a glutamate stimulus led to an increase in cell death. When CaMKIIαB was knocked down in RGCs, a corresponding decrease occurred in the level of BDNF expression.

conclusions. These data indicate that the presence of basal levels of CaMKIIαB in RGCs may afford them some ongoing protection from a stressful environment. In response to the glutamate stimulus, the expression of survival genes such as BDNF may be enhanced through elevation of this particular isoform of the CaMKIIα gene.

Amultifunctional Ser/Thr protein kinase, CaMKII comprises a family of isoforms derived from four closely related genes (α, β, γ, δ). CaMKII is ubiquitously expressed in most cell types, but the predominant isoforms in the brain and the eye are α and β. The proteins play important roles in response to an increase in intracellular calcium ¹ by controlling a variety of cellular functions, ^{2 3} including the regulation of carbohydrate metabolism, membrane current (Ca²⁺, Cl⁻, and K⁺ channels, and ligand-gated channels), neurotransmitter synthesis and release, transcription (C/EBPβ and CRE-binding protein), cytoskeletal organization (τ and microtubule-associated protein 2), long-term potentiation, and neuronal memory. Recently, CaMKII has been shown to be involved in signaling pathways related to cell death and cell survival. The mechanisms by which CaMKII regulates cell death and survival are largely unknown, but it seems that both antiapoptotic ^{4 5 6 7 8 9} and proapoptotic ^{10 11 12 13 14 15 16 17} properties have been attributed to CaMKII in a variety of cells. 

The death of retinal ganglion cells (RGCs) is a leading cause of blindness in patients with retinal diseases such as glaucoma and retinal ischemia. Extensive reports in the literature have shown that glutamate release and N-methyl-d-aspartate (NMDA) receptor–mediated excitotoxicity are likely important contributors to RGC death in these conditions, ^{18 19 20 21 22 23 24 25 26} particularly with regard to secondary RGC degeneration, ^{27 28} though some studies fail to find the link between elevated glutamate levels and glaucoma. ^{29 30 31 32 33} A clinical trial with memantine, an NMDA receptor antagonist, is in progress. ^{34 35} In view of such trials, it seems important to explain the signal transduction pathways that could be involved. CaMKII is an important enzyme downstream of such receptors, and it responds to increases in intracellular Ca²⁺ resulting from hyperstimulation of the NMDA receptor. This calcium-sensitive enzyme plays an important role in controlling multiple cellular functions. We focused our studies on the role of the α isoform of CaMKII in cell death/survival responses in the retina, especially in the RGCs, for several reasons. First, CaMKIIα, one of the dominant isoforms in the retina, is located in the cells of the inner nuclear and ganglion cell layers, including RGCs. ^{36 37} Second, CaMKIIα activity and expression levels are altered in vivo in the retina in response to NMDA stimulation. ^{38 39} Third, neuroprotection within these cell layers of the retina is afforded in vivo by treatment with the autocamtide-2–related inhibitory peptide (AIP), a specific inhibitor for CaMKII. ¹⁰ In addition, inhibition of CaMKIIα with AIP provides a retinal ganglion cell line with neuroprotection against glutamate treatment in vitro. ⁴⁰  

The presence of discrete subcellular pools, together with the specific pattern of localization of CaMKIIα close to its many substrates, has important regulatory consequences. ^{38 41} Although cytoplasmically localized CaMKIIα has been implicated in regulating cell death by directly or indirectly activating caspase-3 ^{38 40} and by inhibiting the release of BDNF, ⁴² the function of the nuclear localized CaMKIIα, CaMKIIαB, remains largely unknown in RGCs or any other cells. CaMKIIαB results from alternative splicing of the α gene and contains an 11-amino acid insert, the nuclear localizing signal (NLS), in the variable region of its regulatory domain. The function of the NLS is to target the CaMKIIα protein from the cytoplasm to the nucleus. ^{13 43} The nuclear localization of CaMKIIαB is most likely indicative of a functional role for CaMKIIα in regulating gene expression, especially during Ca²⁺-mediated transcriptional regulation of genes. CaMKIIαB is reportedly most abundant in the midbrain and diencephalon regions of the brain. ⁴⁴ We have previously shown that an early and transient increase in CaMKIIαB mRNA expression occurs in the retina after intravitreal injection of NMDA. ³⁸ The transient increase of the CaMKIIαB transcript correlates with an increase in the amount of CaMKIIα protein in nuclear extracts several hours later. ⁴⁰ It is possible that elevated levels of CaMKIIαB aid in the regulated expression of other genes, which, in turn, regulate the cell death and survival responses. 

In the present study, we have investigated whether basal levels of CaMKIIαB are present specifically in RGCs and whether CaMKIIαB can be implicated in cell death and survival responses to glutamate stimulation. Our findings have revealed that CaMKIIαB is expressed not only at basal levels in RGCs but that its expression level and subcellular distribution are altered after glutamate treatment. In purified RGCs, knockdown of CaMKIIαB with the aid of RNA interference, followed by glutamate treatment, led to an increase in cell death. In addition, our study showed that corresponding to the knockdown of CaMKIIαB, BDNF expression was decreased, suggesting that CaMKIIαB may aid in regulating the expression of survival genes, such as BDNF, and thus may play a role in the cell survival response. 

Materials and Methods

RGC Culture

All animals were handled in accordance with the regulations of the Institutional Animal Care and Use Committee, and all the procedures adhered to the ARVO Statement for the Use of Animals in Ophthalmic and Vision Research. RGCs from postnatal Sprague–Dawley (SD) rat retinas were purified as previously described by Barres et al. ^{45 46} Briefly, eyes were enucleated from SD rats (postnatal days 6–8) and rinsed with Dulbecco phosphate-buffered saline (Invitrogen, Carlsbad, CA). Retinas were dissected under a microscope and were dissociated with the aid of reagents (Papain Dissociation System kit; Worthington Biochemicals, Lakewood, NJ), at 37°C for 40 minutes, to create a single-cell suspension. RGCs were isolated from this suspension using sequential immunopanning. ⁴⁵ Purified RGCs were seeded on poly-d-lysine/laminin–coated 12-mm glass coverslips or 24-well plates at a density of 2 × 10⁴ RGCs per coverslip or well. Cells were maintained in B27-supplemented medium (Neurobasal; Invitrogen) containing bovine serum albumin (100 μg/mL), progesterone (60 ng/mL), insulin (5 μg/mL), pyruvate (1 mM), glutamine (1 mM), putrescine (16 μg/mL), sodium selenite (40 ng/mL), transferrin (100 μg/mL), triiodo-thyronine (30 ng/mL), brain-derived neurotrophic factor (BDNF; 50 ng/mL), ciliary neurotrophic factor (CNTF; 20 ng/mL), bFGF (10 ng/mL), forskolin (5 μM), inosine (100 μM), and antibiotics (Sigma-Aldrich, St. Louis, MO). RGCs were identified by the expression of cell markers, including Thy-1, and by their characteristic cell morphology. Cultures were maintained at 37°C in a humidified environment of 10% CO₂ and 90% air. ^{46 47} Cells in cultures for 1 to 2 weeks were used for all the experiments. ⁴⁶  

RGC-5 cells (a kind gift from Neeraj Agarwal, University of North Texas Health Science Center, Fort Worth, TX) were maintained in Dulbecco modified Eagle medium (DMEM)–low glucose containing 10% fetal bovine serum, 100 U/mL penicillin, and 100 μg/mL streptomycin in a humidified atmosphere of 90% air and 10% CO₂ at 37°C. ⁴⁸ The cells were trypsinized and subcultured using a 1:20 split after they had reached confluence. 

Cell Treatment

After 1 to 2 weeks in culture, purified RGCs were exposed to glutamate (10–500 μM) for various periods of time. For RGC-5 cells that were induced to overexpress CaMKIIαB, the cells were plated in slide chambers (Nalge Nunc International, Naperville, IL) and were grown for 8 hours. Then the cells were treated with 5 mM glutamate in serum-free DMEM for 24 hours. In RNA interference experiments, purified RGCs were plated at a density of 1 × 10⁴ in PDL/laminin–coated eight-well slide chambers and were transfected with specific siRNA for 6 hours. Twenty-four hours later, the cells were treated with or without 100 μM glutamate prepared in culture medium for 24 hours. 

RT-PCR and Real-Time RT-PCR

Total RNA was extracted from purified RGCs using an RNA isolation kit (PicoPure; Arcturus, Sunnyvale, CA) according to manufacturer’s directions. Total RNA was also extracted from SD rat retinas and RGC-5 cells. The yield and purity of RNA were estimated by optical density at 260/280 nm. After DNAse treatment, cDNA was synthesized from RNA with the use of reverse transcription reagents kit (TaqMan; Applied Biosystems, Foster, CA) with random hexamers as primers, in accordance with the manufacturer’s specifications. Polymerase chain reactions were performed in a PCR system (Gene-Amp PCR System 2400; PerkinElmer, Norwalk, CT) with a PCR mix (MasterMix; Eppendorf, Hamburg, Germany). The following primers were used: for CaMKIIαB, ⁴⁴ the forward primer corresponded to nucleotides 946–965 (5′-CCATCCTCACCACTATGCTG-3′) and the reverse primer to nucleotides 1211–1230 (5′-ATCGATGAAAGTCCAGGCCC-3′); for β-actin, ³⁸ the forward primer corresponded to nucleotides 48–74 (5′-AGCCAGGTCCAGACGCAGGATGGCATG-3′) and the reverse primer to nucleotides 558–584 (5′-GATGATATCGCCGCGCTCGTCGTCGAC-3′). The PCR products for CaMKIIα and β-actin were loaded together in the same gel lane, and the expected PCR products were 284 bp for CaMKIIα, 317 bp (α+33) for CaMKIIαB, and 536 bp for β-actin. The authenticity of all PCR products was established by sequencing (data not shown). For densitometric analysis, the density of individual PCR bands was calculated with a computerized image analysis system (Alpha Innotech, San Leandro, CA) as integrated density values, normalized to β-actin, and compared with the level of CaMKIIαB in purified RGCs, whose expression level was taken as 1. 

To determine the expression levels of CaMKIIαB and BDNF, real-time PCR was performed. Total RNA was extracted from purified RGCs or RGC-5 cells treated with or without glutamate, as described. After DNAse treatment, cDNA was synthesized from 50 ng RNA using reverse transcription reagents (TaqMan; Applied Biosystems, Foster City, CA), followed by real-time PCR. Primers were designed with primer express software (Applied Biosystems, Foster, CA). For CaMKIIαB, the forward primer was 5′-AGAAAGTCCAGTTCCAGCG-3′, and the reverse was 5′-TGATAATTTCCTGTTTGCGC-3′. For BDNF, ⁴² the forward primer was 5′-GGCCCAACGAAGAAAACCAT-3′, and the reverse primer was 5′-GCACTTGACTGCTGAGCATC A-3′. PCR was performed with 2 μL cDNA and a reagent kit (SYBR Green PCR Core 7000 Sequence Detection System; ABI Prism). SYBR green data were analyzed with sequence detection software (7000 Sequence Detection System; ABI Prism). Relative expression levels of the target genes were analyzed according to the 2^-ΔΔCt method ⁴⁹ by normalization with GAPDH gene expression and were presented as the percentage change compared with controls. Experiments were performed in triplicate for each gene and were repeated three times using independent biological replicates. 

Construction of CaMKIIαB Expression Vectors and Overexpression of CaMKIIαB in RGC-5

First-strand cDNA was synthesized from 5 μg rat brain RNA using reverse transcriptase (Superscript II; Gibco BRL, Gaithersburg, MD) with random primers according to the manufacturer’s specifications. The following primers for CaMKIIαB were designed and used in PCR amplification according to Schulman et al. ⁴⁴ : sense, 5′-GGTGGATCCAGGATGGCTACCATCACCTGC-3′; antisense, 5′-CAGGATATCACATTCCATGGACAAAG-3′. The PCR mix contained 1/20 volume from the reverse-transcribed reaction for CaMKIIαB, and all amplifications were performed on a PCR system (GeneAmp 9600; Perkin-Elmer, Foster City, CA). Five microliters of the PCR mixture was loaded on a 1% agarose gel, and the product was visualized with ethidium bromide staining. The expected PCR product was approximately 1.5 kb. The PCR product was cut with BamH1 and EcoRV, inserted into appropriately digested pIRES-hyg3 vector (Clontech, Palo Alto, CA), and subcloned (pIRES-hyg3/aB). The clones contained the 1.5-kb insert were verified by restriction digest and sequencing_. RGC-5 cells were plated at a density of 80% confluence in 24-well plates and were transfected with 1 μg pIRES-hyg3/aB or pIRES-hyg3 (Lipofectamine 2000; Invitrogen) for 48 hours. Transfected RGC cells were selected with 400 μg/mL aminoglycoside antibiotic (Hygromycin B; A.G. Scientific, San Diego, CA) in culture medium for 1 week, and the percentage of transfected cells was calculated using immunofluorescence microscopy. 

RNA Interference

Target sequences for CaMKIIαB small interfering (si)RNA were designed with the Web-based tool to locate siRNA target sites (Target Finder; Ambion, Austin TX). CaMKIIαB siRNAs were tested in initial transfection experiments, and the most effective knock down was obtained by transfecting 100 nM CaMKIIαB-1011 siRNA, named after the nucleotide start site in the CaMKIIα sequence (GenBank accession no. NM_012920). To knock down CaMKIIαB in purified RGCs or RGC-5 that overexpressed CaMKIIαB, cells were plated in an eight-well slide chamber or in six-well plates at a density of 1 × 10⁴ or 1 × 10⁵cells/well and were transfected with 100 nM (or 50–200 nM in some experiments) CaMKIIαB-1011 siRNA (Ambion) for 6 hours with a reagent (Lipofectamine 2000; Invitrogen) in accordance with the manufacturer’s instructions. Nonspecific siRNA served as control. In addition, a mock transfection control (without siRNA) was included. Transfection efficiency was monitored (BlOCK-iT fluorescent Oligo; Invitrogen). The knockdown of CaMKIIαB was tested 24 to 72 hours later by RT-PCR/real-time PCR, immunofluorescence staining, and immunoblotting, and cell viability was assessed (Calcerin/EthD-1 staining; Molecular Probes, Eugene, OR). 

Western Blotting

Cytoplasmic and nuclear extracts were obtained (NE-PER Nuclear and Cytoplasmic Extraction Reagent kit; Pierce Biotechnology, Rockford, IL) according to the manufacturer’s protocol. The concentration of all protein samples was determined by Coomassie Plus Protein Assay (Pierce Biotechnology). Equal amounts of protein samples were separated on 8% SDS-PAGE gels and transferred to polyvinylidene difluoride membranes (Millipore, Bedford, MA). Membranes were blocked overnight at 4°C in 0.1% Tween-20 Tris-buffered saline solution containing 5% nonfat dry milk and then were incubated with anti-CaMKIIα (Sigma, St. Louis, MO). Antibody binding was detected with horseradish peroxidase–conjugated anti–mouse (Chemicon International Inc., Temecula, CA) secondary antibodies and ECL Western blotting detection reagents (Amersham Life Sciences, Buckinghamshire, UK). 

Immunocytochemistry

Purified RGCs or RGC-5 cells were plated on poly-l-lysine/laminin–coated coverslips or chamber slides (Nalge Nunc International, Naperville, IL). Immunostaining for CaMKIIα or BDNF, or double-immunofluorescence labeling of CaMKIIα and BDNF, was performed. Cells were fixed for 10 minutes in 4% paraformaldehyde, washed three times in PBS, and permeabilized with 0.1% Triton-X-100 in PBS for 5 minutes After blocking, cells were incubated with primary antibodies overnight at 4°C. Primary antibodies used were anti–CaMKIIα monoclonal antibody (BD Transduction Laboratories, Lexington, KY) and anti–BDNF polyclonal antibody (Santa Cruz Biotechnology, Santa Cruz, CA). CaMKIIα was visualized with Cy3-conjugated goat anti–mouse secondary antibody (Chemicon International); BDNF was visualized with Alexa 488-conjugated goat anti–rabbit secondary antibody (Molecular Probes, Eugene, OR). The slides were mounted with antifade mounting medium with or without DAPI (Vector Laboratories, Burlingame, CA) and were viewed with the aid of a fluorescence microscope. Images were recorded with equal exposure conditions for each specific antibody. 

Cell Viability Assay

Cell viability was determined with the aid of cell death kit (Live/Dead Viability/Cytotoxicity kit; Molecular Probes). Live cells are distinguished by the presence of ubiquitous intracellular esterase activity, determined by the enzymatic conversion of the virtually nonfluorescent cell-permeant calcein AM to the intensely fluorescent calcein (green), whereas the dead cells produce a bright red fluorescence resulting from the entering of EthD-1 through a damaged membrane. ⁵⁰ Briefly, cells were incubated with 2 μM calcein AM and 4 μM EthD-1 for 30 minutes at room temperature (RT) and then were mounted with PBS and examined with the aid of a fluorescence microscope. Six random fields of cells were counted for viability in each of three wells per condition. Survival rates were presented as the percentage of the total number of the cells in parallel control cultures. 

Statistical Analysis

All quantitative data were expressed as mean ± SD. At least three independent repeats with triplicate determinates were performed for each quantitative assay. The Student’s t-test was used for two-group comparisons. ANOVA was used for multiple comparisons, followed by Newman-Keuls paired comparison. A significance cutoff of P < 0.05 was used. 

Results

Expression of CaMKIIαB, the Nuclear Isoform of CaMKIIα, in Purified RGCs

The presence of the CaMKIIαB isoform in the retina has been reported using total RNA isolated from the whole retina. ³⁸ However, the specific cellular expression/localization of CaMKIIαB in the retina has not been identified. To determine whether CaMKIIαB is expressed specifically in RGCs, total RNA was extracted from RGCs freshly purified from 10 retinas (P6-P8), and CaMKIIαB mRNA was assessed with the aid of RT-PCR. For purposes of comparison with the whole retina, total RNA was extracted from retinas of four rats at the same age (P6-P8). Two independent RT-PCR procedures were performed using the same amount of RNA. Expression levels for CaMKIIα or CaMKIIαB were represented as the averages of the two experiments after normalization to β-actin. 

Like the expression pattern for the whole retina, pan-purified RGCs expressed the CaMKIIα and CaMKIIαB transcripts (Fig. 1A) . Although the ratio of basal levels of the CaMKIIαB transcript to the CaMKIIα transcript was a little lower in RGCs than in retinas, total amounts of the CaMKIIα and CaMKIIαB transcripts were similar in RGCs and retinas (Figs. 1A 1B) . This expression pattern did not change in RGCs after culture for 1 to 2 weeks (data not shown). CaMKIIαB expression in RGC-5 cells was also included because RGC-5 cells were used for some experiments. As reported, ⁴⁰ RGC-5 cells expressed lower levels of CaMKIIαB and CaMKIIα compared with the whole retina and, as seen here, with the pan-purified RGCs (Figs. 1A 1B) . This is the first demonstration of a relatively high level of basal expression of this transcript in RGCs. 

Effect of Glutamate on Pan-Purified RGC Survival

Application of increasing concentrations of glutamate caused a dose-dependent decrease in RGC survival after 24 hours (Fig. 2) . Figure 2Ashowed the morphologic changes and cell death in cultures treated with glutamate. Treatment of RGCs with 100 and 500 μM glutamate reduced the cell survival rate to 73.8% ± 8.4% and 39.2% ± 7.7%, respectively (Fig. 2B) . The concentration of 100 μM glutamate was selected for all the other experiments based on its modest neurotoxic effect on RGCs. 

Changes in CaMKIIαB Expression and Intracellular Distribution in RGCs in Response to Glutamate Treatment

We have previously shown that CaMKIIαB expression transiently increased in the retina in the in vivo condition after intravitreous injection of NMDA. ³⁸ The increase could be detected as early as 30 minutes after the stimulus. To determine whether a glutamate response occurred specifically in RGCs, pan-purified RGCs were treated with or without (control) 100 μM glutamate for 2 and 24 hours. CaMKIIαB expression was assayed by real-time PCR. It was observed that CaMKIIαB transiently increased by 64% ±15.9% at 2 hours and then declined to the control level by 24 hours (Fig. 3A)in response to this glutamate treatment. After this transient increase in the CaMKIIαB transcript, a change in CaMKIIα protein distribution in RGCs was detected 4 hours later. As shown by immunostaining (Fig. 3B) , CaMKIIα was located primarily throughout the cytoplasm in cells before glutamate treatment, though there was some basal level of staining in the nuclei. In cells treated with glutamate (4 hours), though the cytoplasmic staining for CaMKIIα was still observed, nuclear staining was significantly elevated in most of the cells, indicating a nuclear distribution for CaMKIIα protein in RGCs in response to glutamate treatment. These data suggest that elevation of the alternatively spliced transcript CaMKIIαB is part of the normal RGC response to glutamate treatment. 

Overexpression of CaMKIIαB in RGC-5 Cells Enhanced Cell Survival against Glutamate Treatment

Because the pan purification and culture of RGC cells is labor intensive, initial investigations to determine the experimental parameters needed for knocking down CaMKIIαB were performed with the aid the RGC-5 cell line. Cells were first transfected with pIRES-hyg3/αB or control vector pIRES-hyg3. Overexpression of CaMKIIαB in RGC-5 was determined by RT-PCR and immunostaining (Fig. 4) . As shown in Figure 4A , CaMKIIαB was significantly overexpressed, and the overexpressed CaMKIIαB protein was largely located in the nucleus of the cells (Fig. 4B) . No significant change in cell morphology or cell viability occurred after CaMKIIαB overexpression (data not shown). However, the data demonstrated that CaMKIIαB overexpression in the RGC-5 cells enhanced the cell survival response to glutamate treatment, as assayed by the cell death kit (Live/Dead Viability/Cytotoxicity kit; Molecular Probes). Treatment with 5 mM glutamate for 24 hours reduced cell survival to approximately 60% in mock-transfected cells. By contrast, more than 95% of the cells that overexpressed CaMKIIαB survived glutamate treatment (ANOVA; P < 0.01) (Figs. 4C 4D) . These data strongly suggest a role for CaMKIIαB in supporting cell survival. 

Specific Knockdown of CaMKIIαB

To specifically knock down the CaMKIIαB, we designed several siRNAs that span the CaMKIIα sequence and the 33-bp insertion sequence. ⁵¹ Among three available pairs of siRNAs, the most effective knockdown was obtained from CaMKIIαB siRNA-1011. RGC-5 cells that overexpressed CaMKIIαB were transfected with siRNA-1011 (50, 100, and 200 nM) or nonspecific siRNA for 6 hours. The specific knockdown of CaMKIIαB was evaluated at 24 to 48 hours with the aid of real-time PCR, immunostaining, and Western blotting. As shown by fluorescence microscopy, though nuclear CaMKIIαB expression was significantly and specifically inhibited by siRNA-1011 in a concentration-dependent manner but not by nonspecific siRNA, cytoplasmic staining for CaMKIIα remained relatively unchanged (Fig. 5A) . Compared with controls (mock or nonspecific siRNA), CaMKIIαB was knocked down by 60% to 70% with siRNA-1011 at a concentration of 100 nM, as assayed by real-time PCR (Fig. 5B ; ANOVA; P < 0.01). This was further confirmed by Western blotting, which showed a significant decrease in the amount of CaMKIIα in nuclear extracts, whereas the amount of cytoplasmic CaMKIIα remained unchanged, indicating a specific knockdown of CaMKIIαB, not the cytoplasmic CaMKIIα (Fig. 5C) . 

Knockdown of CaMKIIαB Decreased Cell Survival in Purified RGCs in Response to Glutamate Treatment

To determine the role of CaMKIIαB in cell survival responses, specifically of RGCs, we subsequently used pan-purified primary cultured RGCs and used the siRNA approach described above to knock down endogenous CaMKIIαB. Cells were transfected with 100 nM CaMKIIαB-specific siRNA-1011 or nonspecific siRNA for 6 hours, and cell viability was assayed at 24 to 48 hours Knockdown of CaMKIIαB was determined by RT-PCR. Cell viability was assessed with the aid of the live/dead cell death kit for CaMKIIαB knockdown alone and for CaMKIIαB knockdown followed by glutamate treatment. As shown in Figure 6A , CaMKIIαB expression was significantly reduced by siRNA-1011, whereas CaMKIIα remained almost unchanged in purified RGCs. Knockdown of CaMKIIαB alone caused a trend for cell survival to decrease compared with the controls (nonspecific siRNA), but this change was not statistically significant (Fig. 6B) . However, knockdown of the CaMKIIαB followed by a stimulus with 100 μM glutamate significantly enhanced RGC death even in the presence of BDNF supplement in the culture medium. Less than 55% of RGCs survived the treatment compared with controls (Fig. 6C) . These data further support results from the previous overexpression/knockdown experiments and indicate that CaMKIIαB is indeed important for RGC survival after glutamate stimulation. 

BDNF Expression Was Regulated by CaMKIIαB

To determine the mechanisms underlying how CaMKIIαB might be involved in such a survival response to the glutamate stressor, we tested the possibility that CaMKIIαB may aid in regulating the expression of survival genes. Pan-purified RGCs were used, and CaMKIIαB was knocked down with the aid of siRNA. Expression levels of CaMKIIαB and of the cell survival/growth factor BDNF were determined with real-time PCR and double immunostaining. Pan-purified RGCs expressed endogenous levels of BDNF in the in vitro condition. However, the expression level of BDNF was significantly decreased when CaMKIIαB was knocked down, as shown by real-time PCR (Fig. 7A) . Thus, when CaMKIIαB expression was inhibited by 60% with the aid of RNA interference, a corresponding and significant decrease in BDNF expression occurred in the RGCs (Student’s t-test; P < 0.05). Double labeling with antibodies to CaMKIIα and BDNF in purified RGCs further revealed that knockdown of CaMKIIαB resulted in decreased BDNF staining intensity (Fig. 7B) . The level of BDNF immunostaining can reflect the level of endogenous BDNF, BDNF taken up from the culture medium, or both. However, the RT-PCR data helped to distinguish between these possibilities. Decreased BDNF immunostaining matched the changes in BDNF mRNA levels, indicating that endogenous BDNF was the major contributor to the BDNF immunolabeling detected. The data support the possibility that CaMKIIαB might be a regulator of BDNF expression. 

Discussion

CaMKIIα is one of the dominant isoforms of the CaMKII family expressed in brain and retina. The nuclear localized isoform, CaMKIIαB, is a splice variant of the α gene. Although CaMKIIα is widely expressed in the brain, CaMKIIαB is reportedly restricted to the midbrain and diencephalon, where it has approximately equal amounts of CaMKIIα and CaMKIIαB. Our laboratory has previously reported the presence of CaMKIIαB in the retina, but the cell-specific expression had not been identified in these earlier studies. ³⁸ Detection of cell-specific expression/localization of CaMKIIαB in vivo is difficult because of the high homology between CaMKIIα and CaMKIIαB transcripts. The only difference between CaMKIIαB and CaMKIIα is the 33-bp insertion that encodes NLS, which targets CaMKIIαB to the nucleus. In this study, we have demonstrated for the first time that CaMKIIαB is expressed at relatively high levels in the RGCs. In addition, we found that the expression pattern of CaMKIIα and CaMKIIαB in pan-purified RGCs is similar to that of the whole retina, though the ratio of CaMKIIαB transcript to CaMKIIα transcript appears a little lower in the RGCs than in the whole retina. Whether this minor difference is caused by the procedures of cell purification is unknown. It could also be related to the presence of CaMKIIα in the amacrine cells. Given the lack of a significant change in the expression pattern for CaMKIIαB in RGCs in culture, we investigated the role of CaMKIIαB in the cell death/survival responses of the RGCs. 

Although it has been widely reported that RGCs are highly vulnerable to glutamate and NMDA excitotoxicity in vitro ^{52 53 54 55} and in intact retinas ^{56 57} ex vivo or in vivo, ^{58 59 60 61} some investigators have suggested that RGCs are resistant to glutamate receptor agonist treatment. Luo et al. ⁶² report that adult RGCs are resistant to glutamate treatment in mixed retinal cultures. Ullian et al., ⁴⁷ using highly purified RGCs and serum-free medium, show the invulnerability of RGCs to glutamate or NMDA treatment. In this study, with the aid of the cell death kit, we have shown that panned RGCs are susceptible to glutamate treatment, though a relatively high concentration of glutamate is used here to induce the same amount of cell death seen in some other studies. ^{53 54} These variabilities may stem from the criteria used to define the surviving cells, the age of the cells, or perhaps the culture conditions. 

We have previously shown the change in CaMKIIαB mRNA expression in the retina after intravitreal injection of the glutamate receptor agonist NMDA. ³⁸ In this study, we have demonstrated this is also the case in isolated RGCs. Glutamate treatment induced a transient increase in the CaMKIIαB transcript at an early stage in purified RGCs. Corresponding to this change, the CaMKIIα protein increases in the nucleus several hours later, indicating that glutamate stimulation induces alternative splicing of the α gene whose product is targeted to the nucleus. These findings strongly suggest that CaMKIIαB is part of a normal signal transduction response to glutamate treatment. 

As an initial experiment to explore the role of CaMKIIαB in cell death/survival responses, RGC-5 cells were used first. RGC-5 cells expressed lower levels of endogenous CaMKIIα and CaMKIIαB than primary cultured RGCs. We used gene transfection and overexpressed CaMKIIαB in these cells. In case of overexpression, CaMKIIαB enhanced cell survival against glutamate treatment, suggesting that CaMKIIαB might be involved in a cell survival signaling pathway. However, though these cells are useful for screening purposes, caution should be used when interpreting the data because RGC-5 is an E1A virus-transformed cell line and may not be completely representative of the RGCs. In addition, it is unknown whether the overexpressed CaMKIIαB functions in exactly the same way as the endogenous transcript. To overcome some of these problems, we used primary cultured RGCs in combination with RNA interference methods to specifically knock down the endogenous CaMKIIαB. 

The methodological parameters were determined, and the effect of specific siRNAs for CaMKIIαB was first evaluated in overexpressing RGC-5 cells. This allowed us to show that the specific knockdown of CaMKIIαB in purified RGCs is feasible. Subsequent experiments demonstrate that knockdown of endogenous CaMKIIαB decreases the survival rate of glutamate-stressed RGCs. To our knowledge, this study is the first to show the involvement of CaMKIIαB in a cell survival signaling pathway. 

The precise mechanisms underlying the role of CaMKIIαB in cell death/survival responses remain unclear, but it seems likely that it involves the regulation of gene expression. ⁶³ It has been shown that CaMKIIα plays a role in Ca²⁺-mediated transcriptional regulation of genes through the phosphorylation of transcription factors such CREB, ^{64 65} ATF, ^{66 67} CCAAT/enhancer-binding protein (C/EBP), ^{68 69} and serum response factor. ⁷⁰ Recently, another transcription factor, NeuroD, has been found to be phosphorylated by CaMKIIα in granule cells. ⁷¹ Ultimately, CaMKIIα and its downstream signaling cascade are involved in regulating a wide variety of cellular events, including proliferation, differentiation, and even apoptosis. In such cases, an increase in intracellular Ca²⁺ is critical in the CaMKIIαB signaling pathway. Although we have not measured the levels of intracellular Ca²⁺ in RGCs in response to glutamate stimulation here, other investigators have. For example, Otori et al. ⁵⁴ show that glutamate can activate Ca²⁺-influx through AMPA-KA receptors in early postnatal RGCs maintained in the same condition used in our study. 

To determine whether CaMKIIαB is indeed involved in regulating the expression of genes critical for RGC survival, we looked at the expression of BDNF. In the retina, BDNF has been proposed to play critical roles not only in the development and differentiation ^{72 73} but also in the survival of retinal neuronal cells of the mature animal. ^{58 74 75 76} There are two source methods of BDNF for RGCs: retrograde transport and local synthesis. ⁷⁷ Although retrogradely transported BDNF has been recognized as an important trophic factor for RGC survival, ^{78 79 80} locally synthesized BDNF has been implicated in RGC protection. ^{81 82 83 84} Local levels of BDNF mRNA and protein in the retina have been shown to be modulated by injury to the optic nerve, ⁸¹ by ocular hypertension, ⁸² and by injection of NMDA into the eye. ⁸⁵ Transgenic expression of the BDNF gene prolongs the survival of RGCs in experimental glaucoma models, supporting the potential role of locally synthesized BDNF in RGC protection. ^{86 87} In vitro, where the retrogradely transported BDNF is not as much a factor, supplements of trophic factors including BDNF appear to be mandatory for RGCs to survive. Neutralizing BDNF secreted from cells or blocking its cognate receptor, TrkB, using specific antibodies enhances RGC death in vitro.⁴² Taken together, these studies suggest the presence of an important paracrine/autocrine mechanism for BDNF support of RGCs, especially under stress. In the present study, our data have revealed that when CaMKIIαB is knocked down, there is a corresponding decrease in the level of BDNF protein in RGCs. Considering that CaMKIIαB knockdown enhances RGC death, our data may indicate an involvement of CaMKIIαB in regulating BDNF expression and thus cell survival responses. This may be especially true for in vivo conditions in which the microenvironment of cells is intact and, therefore, the locally synthesized BDNF is significant for maintaining cell survival. ⁸⁸  

It should be noted that, in our experiments, after the knockdown of CaMKIIαB and the reduction of endogenous BDNF, the supplement of exogenous BDNF in the culture medium did not protect the cells from dying. It seems likely that BDNF is not the only survival gene regulated by CaMKIIαB. It is assumed that other genes are involved. For example, it has been reported that the inhibition of CaMKII suppresses Bcl-2 expression and accelerates neuronal damage after exposure to glutamate. ⁸⁹ Bcl-2 is a well-known antiapoptotic gene. We have also shown an increase in Bcl-2 expression in RGC-5 cells that overexpress CaMKIIαB (data not shown), suggesting that Bcl-2 might be another survival-promoting gene regulated by CaMKIIαB. Indeed, target genes whose expression is regulated by CaMKIIαB are the subject of further investigation in this laboratory. 

In summary, the present study has demonstrated the presence of the nuclear isoform of CaMKIIα, namely CaMKIIαB, specifically in RGCs. The study has also shown that CaMKIIαB is involved in the cell survival signaling pathways in response to glutamate treatment. This probably occurs as the result of CaMKIIαB-mediated phosphorylation of transcription factors and the regulation of gene expression for growth factors such as BDNF or of other antiapoptotic genes. Elucidating the complete signaling pathways involved in cell death and survival in the cells most affected in diseases and disorders of the nervous system is an important task and may yield new pharmaceutical targets. Additional studies to clarify how CaMKIIαB regulates gene expression and which transcription factor(s) and genes are involved in the cell death and survival pathways in RGCs are ongoing in this laboratory. 

 Supported in part by National Institutes of Health/National Center for Research Resources Grant 2 P20 RR016481.

 Submitted for publication November 15, 2006; revised February 13, 2007; accepted June 11, 2007

 Disclosure: W. Fan, None; X. Li, None; N.G.F. Cooper, None

 The publication costs of this article were defrayed in part by page charge payment. This article must therefore be marked “advertisement” in accordance with 18 U.S.C. §1734 solely to indicate this fact.

 

Figure 1.

 

Expression of CaMKIIαB in RGCs. Total RNA was extracted from RGCs pan-purified from 10 retinas (P6-P8). For comparison with retina or RGC-5 cells, total RNAs were also extracted from four retinas of rats at the same age (P6-P8) and RGC-5 cells. CaMKIIαB mRNA was assessed with the aid of RT-PCR. (A) CaMKIIαB (317 bp) and CaMKIIα (284 bp) was detected in purified RGCs. (B) For densitometric analysis, the density of individual PCR bands was calculated with a computerized image analysis system as the integrated density value, normalized to that of β-actin, and compared with CaMKIIαB of purified RGCs, whose expression level was taken as 1. The expression level for CaMKIIα or CaMKIIαB was presented as the average of the two independent experiments using the same amount of respective RNA. The number over each bar represents the relative percentage amount of CaMKIIαB or CaMKIIα.

Figure 2.

 

Effect of glutamate on RGC survival. (A) Images of RGCs incubated with or without (control) glutamate for 24 hours showed the changes in cell morphology and cell death characterized by the loss of neuritis, cell debris, and nuclear condensation. (B) Dose-dependent effect of glutamate on RGC survival assayed by counting calcein AM–positive cells. Data are presented as mean ± SD (n = 8; ANOVA; *P < 0.01).

Figure 3.

 

Upregulation of CaMKIIαB expression and redistribution of CaMKIIα in purified RGCs in response to glutamate treatment. (A) RGCs were treated with or without (control) 100 μM glutamate for 2 and 24 hours. Expression of CaMKIIαB was assayed by real-time PCR. Data are presented as mean ± SD (n = 9; Student’s t-test; *P < 0.01). (B) Immunostaining of CaMKIIα in RGCs treated with or without glutamate (4 hours) showed a change in intracellular distribution pattern, with more nuclear localization of CaMKIIα in glutamate-treated cells.

Figure 4.

 

Overexpression of CaMKIIαB in RGC-5 cells enhanced cell survival against glutamate treatment. CaMKIIαB expression vector was constructed, and RGC-5 cells were transfected with pIRES-hyg3/αB or control pIRES-hyg3 vector (mock-transfected) for 24 hours (A) RT-PCR and (B) immunostaining for CaMKIIα were performed to show the overexpression and nuclear localization of CaMKIIαB in RGC-5 cells. (C) RGC-5 cells that overexpressed CaMKIIαB or the mock-transfected cells were plated in a slide chamber and grown for 8 hours. Then the cells were treated with 5 mM glutamate in serum-free DMEM for 24 hours. Cell viability was determined with the aid of a cell death kit. Live cells were stained green (calcein), and dead cells produced a bright red fluorescence (EthD-1). (D) Treatment of 5 mM glutamate for 24 hours reduced the cell survival rate to 60% in mock-transfected cells. By contrast, more than 95% of cells that overexpressed CaMKIIαB survived the glutamate treatment. Data are presented as mean ± SD (n = 9; ANOVA; *P < 0.01).

Figure 5.

 

Specific knockdown of CaMKIIαB by RNA interference in CaMKIIαB-transfected RGC-5 cells. Those that overexpressed CaMKIIαB were transfected with siRNA-1011 (50, 100, and 200 nM), with nonspecific siRNA, or without siRNA (mock) for 6 hours. The specific knockdown of CaMKIIαB was tested at 24 to 48 hours (A) Immunofluorescence antibody labeling showed the knockdown of CaMKIIαB in RGC-5 cells with 50 nM, 100 nM, and 200 nM specific siRNA. Nonspecific siRNA served as control (upper left). (B) Real-time PCR showed that CaMKIIαB was knocked down by 60% to 70% using specific siRNA. Data are presented as mean ± SD (n = 9; Student’s t-test; *P < 0.01). (C) Immunoblot assay for CaMKIIα in the nuclear and cytoplasmic extract of CaMKIIαB-transfected cells showed a reduction of nuclear CaMKIIα after transfection with specific siRNA relative to the nonspecific siRNA and mock transfection with Lipofectamine reagent. β-Actin served as a loading control.

Figure 6.

 

Knockdown of CaMKIIαB decreased cell survival in pan-purified RGCs in response to glutamate treatment. Purified RGCs were used, and RNA interference was used to knock down the endogenous CaMKIIαB. The cells were transfected with 100 nM specific siRNA-1011 or nonspecific siRNA for 6 hours and were assayed at 24 to 48 hours. In some experiments, glutamate treatment was performed after CaMKIIαB knockdown. Cell viability was assessed before and after glutamate treatment using a cell death kit. (A) Significant knockdown of CaMKIIαB, not CaMKIIα, in purified RGCs, as shown by RT-PCR. (B) After CaMKIIαB knockdown, there was a trend for the cell survival rate to decrease compared with control (nonspecific siRNA), but it was not statistically significant. (C) Knockdown of CaMKIIαB followed by glutamate treatment significantly enhanced cell death in purified RGCs. Data are presented as mean ± SD (n = 9; Student’s t-test; *P < 0.05).

Figure 7.

 

BDNF expression was regulated by CaMKIIαB in RGCs. Purified RGCs were transfected with 100 nM specific siRNA-1011 or nonspecific siRNA for 6 hours and were assayed at 24 to 48 hours. (A) Real-time PCR showed that CaMKIIαB was knocked down by 60%. Correspondingly, there was a significant decrease in BDNF expression in RGCs. Data are presented as mean ± SD (n = 9; Student’s t-test; *P < 0.05). (B) Double labeling of CaMKIIα and BDNF in purified RGCs revealed that the knockdown of CaMKIIαB resulted in decreased BDNF immunostaining intensity.