The role of CaMKII in BDNF-mediated neuroprotection of retinal ganglion cells (RGC-5)

BR A IN RE S EA RCH 1 0 67 ( 20 0 6 ) 4 8 –57

a v a i l a b l e a t w w w. s c i e n c e d i r e c t . c o m

w w w. e l s e v i e r. c o m / l o c a t e / b r a i n r e s

Research Report

The role of CaMKII in BDNF-mediated neuroprotection of retinal ganglion cells (RGC-5) Wei Fan a , Neeraj Agarwal c , Nigel G.F. Cooper a,b,⁎ a

Department of Anatomical Sciences and Neurobiology, 500 S. Preston St., Louisville, KY 40292, USA Department of Ophthalmology and Visual Sciences, University of Louisville School of Medicine, Louisville, KY 40202, USA c Department of Cell Biology and Genetics, University of North Texas Health Science Center, Fort Worth, TX 76107, USA b

A R T I C LE I N FO

AB S T R A C T

Article history:

The purpose of the study is to determine if expression or secretion of brain-derived

Accepted 9 October 2005

neurotrophic factor (BDNF) in retinal ganglion cells (RGC-5) is mediated by NFκB or Ca2+/

Available online 6 December 2005

calmodulin-dependent protein kinase II (CaMKII). RGC-5 cells were exposed to 1 mM glutamate for various periods of time, in the presence or absence of prospective regulatory molecules. BDNF mRNA and protein expression were assessed with the aid of real-time PCR

Theme: Development and regeneration

and immunoblots, respectively, and BDNF secretion was determined by ELISA. The NFκB

Topic:

inhibitor (TLCK and PTD-p65), or a specific CaMKII inhibitor (m-AIP), was used to study

Neuronal death

association of NFκB or CaMKII with BDNF expression/secretion in RGC-5 cells. Glutamate stimulated a transient increase in BDNF mRNA and protein in RGC-5 cells, and also

Keywords:

stimulated an early release of BDNF into the culture media. Neutralizing the BDNF or

BDNF

blocking the TrkB receptor enhanced the glutamate-induced cytotoxicity. NFκB nuclear

CaMKII

translocation was revealed in response to glutamate treatment. Application of TLCK or PTD-

NFκB

p65 inhibited the glutamate-induced BDNF expression and secretion. Inhibition of CaMKII

Cytotoxicity

by m-AIP did not affect expression but significantly enhanced the release of BDNF from

AIP

glutamate challenged cells. Our data suggest that glutamate treatment may stimulate

Neuroprotection

expression of BDNF in RGC-5 cells through NFκB activation. A novel mechanism for neuroprotection is proposed for the CaMKII inhibitor, AIP, which appears to protect RGC-5 cells from cytotoxicity by enhancing the release of BDNF from glutamate challenged cells. © 2005 Elsevier B.V. All rights reserved.

1.

Introduction

Brain-derived neurotrophic factor (BDNF) is a member of the protein family of neurotrophins showing widespread expression in the developing and adult mammalian brain. It is believed that BDNF plays an important role in neuronal survival, differentiation, and synaptic plasticity (Lessmann et al., 2003). In addition, BDNF is also important for protection of neurons in excitotoxic, hypoxic, and hypoglycemic insult,

suggesting an essential survival promoting role of this neurotrophin after injury (Lessmann et al., 2003; Murer et al., 2001). The biological functions of BDNF are mediated through the binding of this trophic factor to its high affinity receptor, TrkB (tyrosine kinase receptor), and a low affinity neurotrophin receptor P75NTR, respectively (Kaplan and Miller, 2000; Roux and Barker, 2002). In the retina, BDNF has not only been proposed to play a critical role in the development and differentiation of the

⁎ Corresponding author. Fax: +1 502 852 1475. E-mail address: [email protected] (N.G.F. Cooper). 0006-8993/$ – see front matter © 2005 Elsevier B.V. All rights reserved. doi:10.1016/j.brainres.2005.10.030

BR A IN RE S E A RCH 1 0 67 ( 20 0 6 ) 4 8 –5 7

retina, but also in protection of retinal neuronal cells. For example, retinal ganglion cell (RGC) death is a hallmark of optic neuropathy and retinal ischemia. Administration of exogenous BDNF protects RGCs from optic nerve axotomy (Mey and Thanos, 1993; Peinado-Ramon et al., 1996), retinal ischemia (Unoki and LaVail, 1994), or N-methyl-D-aspartate (NMDA)-induced neuronal death in vivo (Kido et al., 2000). In the in vitro paradigm, supplements of BDNF in the culture media are known to enhance RGC survival (Johnson et al., 1986; Thanos et al., 1989) and has been shown to rescue RGC-5 cells from cell death following serum deprivation (Krishnamoorthy et al., 2001). In addition, BDNF-transfected iris pigment epithelial cells rescues NMDA-induced retinal neuron death both in vivo and in vitro (Kano et al., 2002). BDNF and its receptor TrkB are expressed by RGCs (Garcia et al., 2003; Vecino et al., 1998, 2000). The endogenous levels of BDNF mRNA and protein in the retina have been shown to be modulated by injury to the optic nerve (Gao et al., 1997), or by induced ocular hypertension (Rudzinski et al., 2004), or by injection of NMDA into the eye (Vecino et al., 1999), suggesting an involvement of endogenous BDNF during injury. In addition, transgenic expression of BDNF gene prolongs the survival of RGCs in rat experimental glaucoma models, further supporting the potential role of BDNF in RGC protection (Martin et al., 2003; Mo et al., 2002). Given the central role that BDNF plays in the developing and adult nervous system, as well as its neuroprotective effects, it is critical to understand if BDNF is regulated and how it functions so that it may be used in more effective treatments for neurodegenerative disorders. The purpose of this study was to investigate the mode of BDNF expression/ secretion and the possible signaling pathway involved in RGC cells exposed to glutamate. Glutamate is the major excitatory neurotransmitter in the CNS and in the eye and it has also been implicated in neuronal death in many pathological conditions, including glaucoma and retinal ischemia. It is well documented that exposure to glutamate or its analogues in the in vivo and in vitro (Dreyer et al., 1994; Hahn et al., 1988; Lucas and Newhouse, 1957; Otori et al., 1998; Sisk and Kuwabara, 1985; Sucher et al., 1991, 1997; Vorwerk et al., 1996) models can kill retinal neurons, especially ganglion cells. Excessive glutamate has been shown to be toxic to neurons in culture mainly through two distinct mechanisms: excitotoxicity, which is mediated by the glutamate receptors (Sucher et al., 1997); and oxidative toxicity, which is mediated by the disturbance of the intracellular redox system (Maher and Hanneken, 2005). In both cases, influx of Ca2+ via activation of ionotropic glutamate receptor or other Ca2+ channels is a major pathway for the cells to activate the cell death cascade (Maher and Hanneken, 2005; Sucher et al., 1997). Both mechanisms may be involved in glutamate-induced toxicity in the retina. Whereas this study could be done using primary RGC cultures, there is an inherent problem in getting sufficient cells from primary cultures, all in the same condition at the same time for high-throughput molecular and pharmacological studies. Thus, in this study, we used an immortalized rat RGC cell line, RGC-5 cells. RGC-5 cells have been described to retain many of the characteristics of primary RGCs (Krishna-

49

moorthy et al., 2001). They are also known to be sensitive to glutamate-induced excitotoxicity (Krishnamoorthy et al., 2001; Martin et al., 2004) or oxidative toxicity (Aoun et al., 1992; Maher and Hanneken, 2005). We used this cell line to characterize the response of the cells to glutamate and, in particular, to examine the linkage between the glutamateinduced cytotoxicity and the neurotrophic factor survival response.

2.

Results

While BDNF has been shown to provide neuroprotection to retinal neurons, the mechanisms involved are not clear, and its regulated expression requires further analysis. In this study, the glutamate-induced expression of this growth factor with time was evaluated in RGC-5 cells with the aid of real-time PCR. In addition, a previous study (Fan et al., 2005) has shown that the CaMKII inhibitor, m-AIP, can protect RGC-5 cells from glutamate-induced cell death. Therefore, the effect of m-AIP on BDNF expression was also investigated. As shown in Fig. 1, glutamate treatment caused a bimodal change in BDNF expression. First, a transient increase in transcription of BDNF was revealed at 30 min and became more robust at 2 h following the addition of glutamate to the culture medium. The expression was increased by 60% above the level seen in the unstimulated controls at 2 h. Then, by 6 h, the expression level had fallen back to the control level. At 12 h and 24 h, the level of expression was lower than that of the unstimulated controls, and was about 40% lower at 24 h. Whereas glutamate induced a transient increase in BDNF expression at early time points, the inclusion of m-AIP, a CaMKII inhibitor, had no effect on the glutamate-induced change in BDNF expression seen at 2 h (data not shown). To determine if the glutamate-induced increase in transcription of BDNF resulted in a protein translation, RGC-5 cells

Fig. 1 – Real-time PCR analysis of BDNF expression in RGC-5 cells treated with glutamate (1 mM). Glutamate treatment caused a bimodal change in BDNF expression. Data are presented as % change relative to time course matched controls (without glutamate). Results presented are the mean ± SEM of triplicate determinations in three independent experiments. Student's t test was used for statistical analysis. *P b 0.05, significantly different from respective time point matched control.

50

BR A IN RE S EA RCH 1 0 67 ( 20 0 6 ) 4 8 –57

were lysed and Western blots were used to assay BDNF protein (the 14 kDa mature form). There was an initial small but detectable increase in the amount of BDNF protein at 6 h, which peaked at 12 h. This increased level of translation fell significantly by 24 h after the initial addition of glutamate (Fig. 2). Although the precise timing of the twin peaks of transcription and translation is not known due to the discontinuous assay method, there was clearly a delay of several hours between them. The results shown in Fig. 2 also revealed that when the cells were treated with glutamate in the presence of m-AIP, the amount of intracellular BDNF was significantly reduced compared to the untreated or glutamate treated cells at 2 and 6 h, suggesting that the BDNF store is mobilized. The m-AIP-induced reduction in amount of intracellular BDNF became less obvious at 12 h compared to the unstimulated controls. At 24 h, there was no significant difference in amount of BDNF among unstimulated controls and glutamate-stimulated cells in the absence or presence of m-AIP. The secretion of BDNF is thought to provide a natural protection of RGCs in response to stress. To test if glutamate stimulated a release of BDNF from RGC-5 cells, conditioned media were collected from the same cell cultures as that for Western blots described above, and assayed for BDNF with the aid of the ELISA method. The effect of m-AIP on glutamate stimulated BDNF secretion

Fig. 2 – Analysis of intracellular BDNF (mature proteins, 14 kDa) in RGC-5 cells. Upper panel: Immunoblots of BDNF in RGC-5 cells treated with 1 mM glutamate in the absence or presence of m-AIP (10 μM). Lower panel: The digitized data expressed as fold change in amounts of BDNF. Glutamate treatment caused an increase in the level of BDNF protein in the cells at 6–12 h. Addition of m-AIP in glutamate treated cells led to decrease in the amount of intracellular BDNF from 2–12 h. All data were normalized to β-actin and the values for controls were taken as 1. Values are expressed as the means ± SEM from three independent experiments. C, control (without glutamate); G, glutamate treatment; G/A, glutamate treatment in the presence of m-AIP. One-way ANOVA followed by Newman–Keuls paired comparison was used for statistical analysis. *P b 0.05.

Fig. 3 – ELISA analysis of secretion of BDNF by RGC-5 cells treated with glutamate (1 mM) in the absence or presence of m-AIP (10 μM). Glutamate treatment initiated a small increase in BDNF release at 2 h. From 6–12 h, there was no significant difference in the amounts of BDNF being released when compared to the nontreated controls. Application of m-AIP in glutamate treated RGC-5 cells dramatically enhanced the release of BDNF from 2–12 h. AIP alone also increased BDNF release by the cells from 2–12 h. Data were presented as means ± SEM of triplicate determinations in three independent experiments. One-way ANOVA followed by Newman–Keuls paired comparison was used for statistical analysis. *P b 0.05. was also investigated. As shown in Fig. 3, glutamate stimulated a small but significant increase in the release of BDNF as seen at 2 h. From 6–12 h, when the intracellular levels of BDNF, as detected by Western blots, were evidently on the rise (see Fig. 2), there was no significant difference in the amounts of BDNF being released when compared to the controls. Thus, the glutamate stimulated release is a short-lived effect. In contrast to these results, the addition of m-AIP dramatically increased the levels of BDNF being released by the glutamate-treated RGC-5 cells. There were six- and two-fold increases at 2 and 12 h, respectively. This effect was prolonged but it was not evident at 24 h, a time point when both glutamatestimulated BDNF transcription (Fig. 1) and protein translation (Fig. 2) had become minimal. In order to better understand the role of m-AIP on BDNF release, the effect of AIP alone was also tested. It was noted that m-AIP alone could increase BDNF release from RGC-5 cells at 2–12 h (Fig. 3). Taken together, these data suggest that CaMKII was involved in the BDNF release. Also, it appears that m-AIP blocked CaMKII-mediated inhibition in BDNF release, whether CaMKII was at basal or stimulated levels of autophosphorylation. The presence of TrkB, the high affinity receptor for BDNF, was evident in the RGC-5 cells as shown by immunostaining (Fig. 4A). The immunoblots (Fig. 4B) revealed that TrkB was present as two products, with masses of 145 and 95 kDa, which correspond to the two known protein products encoded by the TrkB gene. The 95 kDa product lacks the intracellular portion but is otherwise identical to the 145 kDa product. To determine if the endogenous BDNF plays a role in neuroprotection in the RGC-5 cells, the effects of neutralizing antibodies for BDNF or TrkB were tested with respect to glutamate-induced cell death. Cell death was

BR A IN RE S E A RCH 1 0 67 ( 20 0 6 ) 4 8 –5 7

51

Fig. 4 – Functional assays of BDNF and its receptor, TrkB, and their roles in the protection of RGC-5 cells from cytotoxicity. (A) TrkB was expressed in RGC-5 cells as detected by immunostaining. (B) Immunoblot of TrkB in RGC-5 cells, showing two bands at 145 and 95 kDa, respectively, corresponding to the two known protein products encoded by TrkB gene. RGC-5 cells were treated with glutamate (1 mM) in the absence or presence of either (C) BDNF neutralizing antibody or (D) TrkB blocking antibody for 24 h and the cell viability was assayed by measuring LDH leakage from the nonviable cells. BDNF or TrkB blocking antibody alone caused an increase in the levels of the cell death, suggesting that loss of BDNF signaling was sufficient to cause some death in these cells. Glutamate plus anti-BDNF or anti-TrkB acted synergistically and induced cytotoxicity in a dose-dependent manner. Data were presented as means ± SEM of triplicate determinations in three independent experiments. One-way ANOVA followed by Newman–Keuls paired comparison was used for statistical analysis. *P b 0.05, significantly different from nontreated controls; **P b 0.05, significantly different from glutamate or blocking antibody alone.

assayed by measuring the LDH leakage from the nonviable cells. The data revealed that glutamate or blocking antibody treatment alone caused a modest increase in the level of cell death above that seen in untreated controls at 24 h. Application of BDNF neutralizing antibody (Fig. 4C) or TrkB blocking antibody (Fig. 4D) plus glutamate acted synergistically and caused an increase in the levels of glutamateinduced cytotoxicity in a dose-dependent manner. NFκB has also been shown to be involved in glutamateinduced excitoxicity in neurons (Grilli and Memo, 1999; Guerrini et al., 1997; Uberti et al., 2000). To determine if NFκB is regulated by glutamate treatment of RGC-5 cells, this transcription factor was evaluated by immunostaining the cells with a monoclonal antibody that selectively reacts with the transcriptionally active, nuclear form of NFκB p65. It was found that the NFκB was located in the cytoplasm in the untreated control cells, but in glutamate stimulated cells, it was revealed in the nuclei (Fig. 5A). Cell counting analysis showed that NFκB translocated to the nucleus in glutamate treated cultures in about 50% of the RGCs, as compared to about 5% in the nontreated cultures (Fig. 5B). The findings that NFκB might be involved in glutamateinduced responses in RGC-5 cells led us to further investi-

gate if NFκB played a role in a cell survival response. Since expression of BDNF increased in response to glutamate treatment, experiments were designed to test if BDNF expression was associated with the NFκB activation. TLCK, a general inhibitor for NFκB, blocks the phosphorylation and degradation of the inhibitory protein, IκB (Epinat and Gilmore, 1999). The phosphorylation of this protein is normally required for activation and translocation of NFκB. TLCK was used to inhibit the activation of NFκB in the RGC-5 cells. Since TLCK also inhibits other proteases, a specific inhibitor for the p65 subunit of NFκB, PTD-p65 (Takada et al., 2004) was also used. Cells were pretreated with TLCK or PTD-p65 prior to glutamate treatment and the expression level and secretion of BDNF were assayed by real-time PCR and ELISA, respectively. TLCK or PTD-p65 inhibited glutamate-induced BDNF expression in a dose-dependent manner as measured at 2 h postglutamate exposure (Figs. 6A and 7A). The ELISA assay showed a significant decrease in glutamate-stimulated BDNF secretion in response to 50 μM TLCK pretreatment both at 2 and 12 h postglutamate exposure (Fig. 6B). Pretreatment of the cells with 200 μM PTD-p65 also reduced BDNF secretion at 12 h (Fig. 7B). In summary, glutamate-stimulated activation and translocation of NFκB appear to be involved in the expression and

52

BR A IN RE S EA RCH 1 0 67 ( 20 0 6 ) 4 8 –57

Fig. 5 – Nuclear translocation of NFκB in RGC-5 cells in response to glutamate treatment. RGC-5 cells were treated with 1 mM glutamate for 30 min to 1 h. NFκB was evaluated in cells with a monoclonal antibody that selectively reacts with the transcriptionally active, nuclear form of NFκB p65. Glutamate treatment induced a NFκB nuclear translocation in RGC-5 cells. (A) Control cells (without glutamate); (B) glutamate treated cells; (C) quantitative analysis of NFκB translocation by cell counting. Data were presented as means ± SEM of duplicate determinations in three independent experiments. Student's t test was used for statistical analysis. *P b 0.05.

ultimately the secretory release of BDNF, a cell survival response.

3.

Discussion

There are three major findings resulting from this study. First, glutamate stimulates a small increase in release of BDNF from RGC-5 cells. This appears to be a neuroprotective response because blocking antibodies against BDNF or its trk-B receptor lead to an elevated level of glutamate-stimulated cell death. In spite of this release of BDNF into the media, some enhanced level of glutamate-stimulated cell death occurs over and above that in the unstimulated control cells. The early release of BDNF is followed by a transient increase in BDNF expression which peaks at about 2 h. This is followed by an increase in translation and the level of cellular BDNF protein peaks at about 12 h. Second, m-AIP, a specific inhibitor of Ca2+/calmodulindependent protein kinase-II (CaMKII), leads to a large increase in the level of BDNF released into the culture media. This correlates with an inhibition of glutamate-stimulated cell death in these RGC-5 cells (Fan et al., 2005). Together, these two pieces of information may explain in part the neuroprotective effect of CaMKII inhibition on cells in the inner nuclear and ganglion cell layers of the retina (Laabich and Cooper, 2000). The findings also support the case that vulnerability to glutamate-mediated cell death in RGC-5 cells and in the retinal cell layers is directly related to extracellular levels of BDNF. Third, NFκB activation is required for the glutamateinduced BDNF expression. To our knowledge, this is the first report to demonstrate that NFκB plays an important role in mediating BDNF expression and CaMKII is involved in regulating BDNF secretion in RGC-5 cells. A novel mechanism for m-AIP in neuroprotection is proposed by this study.

BDNF gene expression in adult brain varies according to physiological activity and also subsequent to injury. BDNF, like all other neurotrophins, is generated as pre–pro-BDNF, which is further processed in the endoplasmic reticulum, trans-Golgi network, and secretory vesicles, until they are eventually secreted as mature homodimer proteins into the extracellular space (Lessmann et al., 2003). The effects of BDNF stem from its binding to the membrane bound receptor, TrkB, which activates PI3-K/Akt and/or mitogen-activated protein kinase signaling pathways, and thereby mediates numerous cellular functions, including inhibition of apoptosis (Chaum, 2003). BDNF and its receptor TrkB are expressed in RGCs in vivo, and there is evidence to show that BDNF provides autocrine and paracrine trophic support to the RGCs in the neural retina (Chaum, 2003). Our study has revealed that RGC5 cells not only express and secrete BDNF, but that they also have the BDNF receptor protein, TrkB, thus providing a valuable in vitro model for studying the modulation of BDNF expression and secretion, as well as signaling pathways. Whereas excitotoxic stress has been implicated as an important cause leading to neuronal death, glutamate, a potential stressor, and its analogues have also been shown to be able to induce expression of BDNF in CNS neurons (Aliaga et al., 1998; Bessho et al., 1993; Canals et al., 1998; Favaron et al., 1993; Gwag and Springer, 1993; Lauterborn et al., 2000; Lindefors et al., 1992; Murray et al., 1996; Timmusk and Metsis, 1994). In the retina, there are few studies relating excitotoxic stress with BDNF expression. One study shows that NMDA injection into the eye induced an increase in both BDNF mRNA and protein in the retinal ganglion cells, suggesting an involvement of endogenous BDNF in RGC defensive reactions (Vecino et al., 1999). In the present study, glutamate treatment induces an early and transient increase in both BDNF mRNA and protein in RGC-5 cells. Also, glutamate stimulation initiates an early and small increase in BDNF release into culture medium. This seems to be consistent with previous studies of the retina and is also supportive of the use of RGC-5

BR A IN RE S E A RCH 1 0 67 ( 20 0 6 ) 4 8 –5 7

53

shown here, and cells in the inner nuclear and ganglion cell layers of the retina do begin to die within 24 h after exposure to certain concentrations of glutamate or its analogue, NMDA (Laabich and Cooper, 2000). It is to be noted that although the level of BDNF protein within the RGC-5 cells starts to increase at 6 h and peaks at 12 h after exposure to glutamate, there is no corresponding increase in its release at these later time points, and this could be a critical point with regard to the eventual cell death. AIP, a specific inhibitor for CaMKII, has been shown to be a neuroprotectant for RGCs treated with NMDA or glutamate both in vivo (Laabich and Cooper, 2000) and in vitro (Fan et al., 2005). The mechanism for the neuroprotective role of AIP is not completely clear and may be mediated through multiple mechanisms. In this study, it is demonstrated that m-AIP can enhance glutamate-stimulated BDNF release in RGC-5 cells. This effect was seen at 30 min after exposure to glutamate (data not shown) and a peak is seen at 2 h, when glutamateinduced BDNF production is hardly underway. m-AIP promotes the release of BDNF for a prolonged period, perhaps for even longer than the glutamate stimulated period of BDNF

Fig. 6 – Effect of TLCK on the expression and secretion of BDNF in glutamate-treated RGC-5 cells. (A) Real-time PCR analysis of BDNF expression. TLCK inhibited glutamateinduced BDNF expression in a dose-dependent manner as measured at 2 h postglutamate exposure. Data are presented as % change relative to the control (without glutamate). *P b 0.05, significantly different from the control; **P b 0.05, significantly different from glutamate alone treated cells. (B) TLCK also inhibited secretion of BDNF by RGC-5 cells at 2 and 12 h postglutamate exposure, *P b 0.05 (Data were presented as means ± SEM of triplicate determinations in three independent experiments. One-way ANOVA followed by Newman–Keuls paired comparison was used for statistical analysis). cells as a model for evaluation of mechanisms involved in cell death and cell survival. The use of cell viability assays with BDNF neutralizing antibody or TrkB blocking antibody has demonstrated that the endogenous BDNF and its high affinity receptor, TrkB, are functional, and that they clearly can protect the cells from glutamate-induced cytotoxicity. The initial glutamate-induced release of BDNF may represent a natural autoprotective mechanism and may be triggered by transmitter-dependent depolarization (see for example Lessmann et al., 2003). In fact, at this early stage, although BDNF mRNA is on the rise, the protein translation is not yet underway. Therefore, the early release of BDNF is most likely derived from preexisting pools within these cells. The released BDNF may exert some protection as a normal part of transmitter release for glutamate challenged cells. However, the protection by this small BDNF release seems limited and perhaps insufficient to protect all cells because RGC-5 cells

Fig. 7 – Effect of PTD-p65 on the expression and secretion of BDNF in glutamate-treated RGC-5 cells. (A) Real-time PCR analysis of BDNF expression. PTD-p65 inhibited glutamateinduced BDNF expression in a dose-dependent manner as measured at 2 h postglutamate exposure. Data are presented as % change relative to the control (without glutamate). *P b 0.05, significantly different from the control; **P b 0.05, significantly different from glutamate alone or p65 control peptide treated cells. (B) PTD-p65 also inhibited secretion of BDNF by RGC-5 cells at 12 h postglutamate exposure, *P b 0.05, significantly different from glutamate-treated or p65-control treated cells. (Data were presented as means ± SEM of triplicate determinations in three independent experiments. One-way ANOVA followed by Newman–Keuls paired comparison was used for statistical analysis).

54

BR A IN RE S EA RCH 1 0 67 ( 20 0 6 ) 4 8 –57

synthesis. These results suggest that AIP promotes the release of the preexisting pool of BDNF, and possibly, of newly synthesized BDNF. Obviously, this relatively fast acting and long lasting role for m-AIP in promoting BDNF release is of importance in neuroprotection, and could be the reason why cell survival in vivo is evidently enhanced in the presence of AIP (Laabich and Cooper, 2000). The mechanisms by which BNDF release is regulated are not yet known, but an increase in the activity of cytoplasmic CaMKII induced by glutamate treatment may be involved (Sucher et al., 1997). CaMKIIα, a major isoform of CaMKII mainly expressed in neurons, is associated with synaptic vesicles (Benfenati et al., 1992) and has been shown to serve as a negative activity-dependent regulator of neurotransmitter release at hippocampal synapses (Hinds et al., 2003). This is possibly the case for BDNF release in RGC-5 cells because RGC5 cell expresses CaMKIIα (Fan et al., 2005), and as shown here, inhibition of CaMKII by m-AIP enhances glutamate-stimulated cells to release more BDNF. Our data have also demonstrated that m-AIP alone can increase BDNF release by the cells that are at control conditions (without glutamate), implicating that CaMKII is also one of the molecules that inhibit basal levels of BDNF release. It should be noted that in vivo, the synthesis and release of BDNF, and TrkB receptor expression and activation, may be more tightly regulated and influenced by neighboring cells and factors that are not present in the in vitro models. Further studies will be needed to show that the findings in the in vitro model apply to RGCs in vivo. The signaling pathway that leads to glutamate-induced BDNF expression is not yet clear. Glutamate induces gene transcription in numerous physiological and pathological conditions and several transcription factors may be involved. Among those glutamate-responsive transcription factors, NFκB has been implicated in neuronal survival and death (Scholzke et al., 2003). Glutamate has been shown to activate NFκB in neurons (Guerrini et al., 1995; Kaltschmidt et al., 1995) and more recently, NFκB p65 has been revealed to be activated in NMDA-induced retinal neurotoxicity (Kitaoka et al., 2004). Both pro- and anti-apoptotic properties have been attributed to NFκB in neurons (de Erausquin et al., 2003; Mattson et al., 2000; Pizzi et al., 2002). The association of NFκB and BDNF expression has been recently implicated in some studies (Jiang et al., 2003; Marini et al., 2004), showing that NFκB is required in NMDA-induced BDNF expression in cerebellar granular cells. The data presented here support this notion and demonstrate an early involvement of the NFκB P65 in response to glutamate treatment. Furthermore, NFκB inhibitor reduces both the expression and secretion of BDNF induced by glutamate, indicating that glutamate increases BDNF expression through an NFκB-dependent pathway in RGC-5 cells. This is supportive of a role for NFκB as an anti-apoptotic transcriptional factor in this case. It should be noted that BDNF gene expression may also be regulated by other transcription factors, such as the cAMP response element binding protein (Jiang et al., 2003) or CaMKIIαB (Takeuchi et al., 2000). CaMKIIαB is a splice variant for CaMKIIα that has a nuclear localization signal (Schulman, 2004). This variant is particularly interesting to us, because we have previously shown that an increase in CaMKIIαB is associated with an increase in cell death in an in vivo rat

model, in which NMDA is injected into the eye (Laabich et al., 2000). In RGC-5 cells, it is observed that glutamate can also stimulate increased expression of CaMKIIαB (Fan et al., 2005). Furthermore, CaMKIIαB is able to translocate into the nucleus upon glutamate stimulation (Fan et al., 2005). Previous studies have indicated an association between high levels of CaMKIIαB and BDNF expressions (Takeuchi et al., 2000). However, our results indicate that m-AIP had no effect on BDNF expression. It is possible that that AIP was not able to act at the nuclear level in the conditions used here. Alternatively, CaMKIIαB may not be involved in BDNF expression in RGC-5 cells. This is being investigated.

4.

Experimental procedures

4.1.

Cell culture

The RGC-5 cells have been shown to express RGC-specific markers thy-1 and Brn-3C. They do not express Muller cell marker GFAP, amacrine cell marker HPC-1, or horizontal cell marker 8A1, suggesting they represent an RGC phenotype (Krishnamoorthy et al., 2001). Cultures of the cells were maintained in Dulbecco's modified Eagle's medium (DMEM)-low glucose containing 10% fetal bovine serum, 100 U/ml penicillin, and 100 μg/ml streptomycine in a humidified atmosphere of 95% air and 5% CO2 at 37 °C. The cells were trypsinized and subcultured using a 1:20 split after they had reached confluence. 4.2.

Cell treatments

To determine the expression levels of BDNF in response to glutamate treatment, early passages (within passage 10) of RGC5 cells were plated at a density of 200 cells/mm2 in 100 mm-culture dishes (Corning Incorporated, NY) and grown for 8 h. Then the cells were treated with or without (controls, time course-matched) 1 mM L-glutamate (Sigma, St. Louis, MO) prepared in serum-free DMEM for various periods of time. For functional assays of the endogenous BDNF on RGC-5 survival, anti-BDNF neutralizing antibodies (Chemicon International Inc., Temecula, CA) or antiTrkB blocking antibodies (BD Pharmingen) were used. Cells were treated with glutamate as described above in the presence of antiBDNF (2 and 20 μg/ml) or anti-TrkB (1:50 and 1:500 dilution) for 24 h and cell viability was assessed (see below). The cells without glutamate, or treated with glutamate or blocking antibodies alone served as controls. NFκB inhibitors, N-α-tosyl-L-phenylalanine chloromethyl ketone (TLCK) (Sigma, St. Louis, MO) (Epinat and Gilmore, 1999) and PTD-p65 (Imgenex, San Diego, CA), or the specific CaMKII inhibitor, the myristoylated autocamtide-2-related inhibitory peptide (m-AIP) (Calbiochem, La Jolla, CA) (Ishida et al., 1995, 1998), were used to test the effects on BDNF expression and/or secretion. The RGC-5 cells were treated with TLCK (10, 50, and 100μM), PTD-p65 (50, 100, and 200 μM), or m-AIP (10μm) for 1 h or 30 min, respectively, prior to glutamate exposure as described above. The cells treated with or without glutamate, or glutamate plus p65 control peptide served as controls. 4.3.

Fluorescence microscopy

RGC-5 cells were plated on poly-L-lysine coated chamber slides (Nalge Nunc International, Naperville, IL). The cells were exposed to 1 mM glutamate treatment for 30 min to 1 h. The untreated cells served as controls. The cells were fixed in ice-cold methanol: acetone (1:1) and air dried. Then the cells were incubated with anti-NFκB p65 monoclonal antibody (BD Transduction Laboratories), followed by incubation with a Cy3-conjugated goat antimouse antibody (Chemicon International Inc., Temecula, CA). The

BR A IN RE S E A RCH 1 0 67 ( 20 0 6 ) 4 8 –5 7

55

TrkB receptor was also detected in RGC-5 cells using the anti-TrkB monoclonal antibody (BD Transduction Laboratories) as described above. The slides were mounted with mounting medium with or without DAPI (Vector Laboratories, Burlingame, CA) and viewed with the aid of a fluorescence microscope. For quantitative analysis of NFκB translocation, cells showing nuclear localization of NFκB were counted in at least six random fields of each chamber (∼2000 cells per chamber). Data are presented as the percentage of the total number of cells in the six fields. The experiments were performed in duplicate wells and repeated three times.

BDNF (Santa Cruz Biotechnology, Inc.), or anti-TrkB (BD Transduction Laboratories). The antibody binding was detected with horseradish peroxidase-conjugated anti-rabbit or -mouse (Chemicon International Inc. Temecula, CA) secondary antibodies and ECL Western blotting detection reagents (Amersham Life Science, Buckinghamshire, England). For quantitative assays, the density of the immunolabeled bands from three independent experiments was calculated with a computerized image analysis system (Alpha Innotech, CA) as the integrated density value, normalized to that of β-actin and compared with the controls, whose expression level was taken as 1 for each time point.

4.4.

4.7.

Real-time PCR

Total RNA was extracted from the RGC-5 cells treated with or without glutamate for various amounts of time, and cDNAs were synthesized from 200 ng total RNA using Taqman reverse transcription reagents (Applied Biosystems, Foster, CA). Realtime PCR analysis was then employed. Since the rat BDNF gene presents a complex structure with four short 5′ noncoding exons (exons I–IV) containing separate promoters and one 3′ exon (exon V) encoding the mature BDNF protein (Marmigere et al., 2003), primers for BDNF exon V were designed with primer-express software (Applied Biosystems): Forward: 5′GGCCCAACGAAGAAAACCAT-3′; Reverse: 5′-GCACTTGACTGCTGAGCATCA-3′. The PCR was carried out by using 2 μl cDNAs with the aid of SYBR Green PCR Core reagent kit in an ABI Prism 7000 Sequence Detection System. Reactions containing no cDNA or no primers served as negative controls. The SYBR green data were analyzed by the ABI Prism 7000 Sequence Detection System software. The relative expression levels of BDNF genes were analyzed using the 2− ΔΔCt method (Livak and Schmittgen, 2001) by normalizing with GAPDH gene expression and presented as the percent changes compared with time course matched controls (without glutamate). 4.5.

Cell death was assayed by measuring the lactate dehydrogenase (LDH) leaked from nonviable cells into the culture media with the Cytotox-One Homogenous Membrane Integrity Assay kit (Promega, Madison, WI), as described in the manufacturer's protocol. Briefly, 5 × 103 cells were plated in 96-well black-walled tissue culture plates and treated with 1 mM glutamate in the presence or absence of blocking antibodies for 24 h. No-cell-controls were included to determine the background fluorescence that might be present in the culture medium. Maximum LDH release controls, which were the same as each experimental treatment, were also included to determine the maximum LDH release in a given treatment group. These maximum releases were determined by the addition of 2 μl Lysis Solution prior to performing the assays. Plates were equilibrated to 22 °C after incubation for the desired time and Cytotox-ONE™ reagent added to each well. After a 10 min incubation, the Stop Solution was added and the fluorescence was recorded in a fluorescence plate reader (SPECTRAFluor Plus, Tecan). The excitation wavelength was 560 nm and the emission wavelength was 590 nm. Percent cytotoxicity was calculated in the following manner: % Cytotoxicity = 100 × (Experimental − Culture Medium Background) / (Maximum LDH Release − Culture Medium Background).

ELISA assay 4.8.

Conditioned media collected from RGC-5 cells were concentrated ten-fold and used for ELISA assays using BDNF Emax® ImmunoAssay system (Promega, Madison, WI) according to product specifications. Briefly, 96-well plates were coated with anti-BDNF monoclonal antibody in 0.025 M sodium bicarbonate/ sodium carbonate, pH 9.7, overnight at 4 °C. The plates were washed twice in TBST (20 mM Tris–HCl, 150 mM NaCl, and 0.05% Tween 20, pH 7.6) and incubated in blocking buffer for 1 h at room temperature (RT). Then, 100 μl conditioned media were added to each well and incubated 2 h at RT. Wells were then washed 5× with TBST and incubated with chicken antiBDNF IgY in blocking buffer for 2 h, followed by extensive washing and incubating with anti-IgY HRP conjugate for 1 h at RT. Reactions were developed using TMB solution for 10 min at RT. Reactions were stopped by the addition of 1 N hydrochloric acid. Colorimetric values were determined by measuring the absorbance at 450 nm. The level of secreted BDNF was expressed as pg/106 cells. 4.6.

Cell viability assay

Western blots

After the media were collected for ELISA assay as described above, the RGC-5 cells were lysed in Cell Lysis/Extraction buffer (Sigma, St. Louis, MO) containing a protein inhibitor cocktail (Sigma, St. Louis, MO) on ice for 15 min. After centrifugation at 4 °C for 15 min at 14,000 × g, the supernatants were collected and stored at −80 °C until use. Equal amounts of protein samples were separated on SDS-PAGE gels and transferred to polyvinylidene difluoride membranes (Millipore, Bedford, MA). The membranes were blocked overnight at 4 °C in 0.1% Tween-20 Tris-buffered saline solution containing 5% nonfat dry milk and incubated with anti-

Statistical analysis

All quantitative data were expressed as means + SEM. Experiments presented here were repeated three to four times with triplicate samples. The Student's t test was used for two group comparisons. One-way ANOVA was used for multiple comparisons followed by Newman–Keuls paired comparison. A P b 0.05 significance cut off was used.

Acknowledgment Supported in part by NIH:NCRR 2 P20 RR016481.

REFERENCES

Aliaga, E., Rage, F., Bustos, G., Tapia-Arancibia, L., 1998. BDNF gene transcripts in mesencephalic neurons and its differential regulation by NMDA. NeuroReport 9, 1959–1962. Aoun, P., Simpkins, J.W., Agarwal, N., 2003. Role of PPAR-{gamma} ligands. Neuroprotection Against Glutamate-Induced Cytotoxicity in Retinal Ganglion CellsInvest. Ophthalmol. Visual Sci., 44, pp. 2999–3004. Benfenati, F., Valtorta, F., Rubenstein, J.L., Gorelick, F.S., Greengard, P., Czernik, A.J., 1992. Synaptic vesicle-associated Ca2+/ calmodulin-dependent protein kinase II is a binding protein for synapsin I. Nature 359, 417–420. Bessho, Y., Nakanishi, S., Nawa, H., 1993. Glutamate receptor

56

BR A IN RE S EA RCH 1 0 67 ( 20 0 6 ) 4 8 –57

agonists enhance the expression of BDNF mRNA in cultured cerebellar granule cells. Brain Res. Mol. Brain Res. 18, 201–208. Canals, J.M., Marco, S., Checa, N., Michels, A., Perez-Navarro, E., Arenas, E., Alberch, J., 1998. Differential regulation of the expression of nerve growth factor, brain-derived neurotrophic factor, and neurotrophin-3 after excitotoxicity in a rat model of Huntington's disease. Neurobiol. Dis. 5, 357–364. Chaum, E., 2003. Retinal neuroprotection by growth factors: a mechanistic perspective. J. Cell. Biochem. 88, 57–75. de Erausquin, G.A., Hyrc, K., Dorsey, D.A., Mamah, D., Dokucu, M., Masco, D.H., Walton, T., Dikranian, K., Soriano, M., Verdugo, J.M.G., Goldberg, M.P., Dugan, L.L., 2003. Nuclear translocation of nuclear transcription factor-kappa B by alphaamino-3-hydroxy-5-methyl-4-isoxazolepropionic acid receptors leads to transcription of p53 and cell death in dopaminergic neurons. Mol. Pharmacol. 63, 784–790. Dreyer, E.B., Pan, Z.H., Storm, S., Lipton, S.A., 1994. Greater sensitivity of larger retinal ganglion cells to NMDA-mediated cell death. NeuroReport 5, 629–631. Epinat, J.C., Gilmore, T.D., 1999. Diverse agents act at multiple levels to inhibit the Rel/NF-kappaB signal transduction pathway. Oncogene 18, 6896–6909. Fan, W., Agarwal, N., Kumar, M.D., Cooper, N.G., 2005. Retinal ganglion cell death and neuroprotection: involvement of the CaMKIIalpha gene. Brain Res. Mol. Brain Res. 139, 306–316. Favaron, M., Manev, R.M., Rimland, J.M., Candeo, P., Beccaro, M., Manev, H., 1993. NMDA-stimulated expression of BDNF mRNA in cultured cerebellar granule neurones. NeuroReport 4, 1171–1174. Gao, H., Qiao, X., Hefti, F., Hollyfield, J.G., Knusel, B., 1997. Elevated mRNA expression of brain-derived neurotrophic factor in retinal ganglion cell layer after optic nerve injury. Invest. Ophthalmol. Visual Sci. 38, 1840–1847. Garcia, M., Forster, V., Hicks, D., Vecino, E., 2003. In vivo expression of neurotrophins and neurotrophin receptors is conserved in adult porcine retina in vitro. Invest. Ophthalmol. Visual Sci. 44, 4532–4541. Grilli, M., Memo, M., 1999. Possible role of NF-kappaB and p53 in the glutamate-induced pro-apoptotic neuronal pathway. Cell Death Differ. 6, 22–27. Guerrini, L., Blasi, F., Denis-Donini, S., 1995. Synaptic activation of NF-kappa B by glutamate in cerebellar granule neurons in vitro. Proc. Natl. Acad. Sci. U. S. A. 92, 9077–9081. Guerrini, L., Molteni, A., Wirth, T., Kistler, B., Blasi, F., 1997. Glutamate-dependent activation of NF-kappaB during mouse cerebellum development. J. Neurosci. 17, 6057–6063. Gwag, B.J., Springer, J.E., 1993. Activation of NMDA receptors increases brain-derived neurotrophic factor (BDNF) mRNA expression in the hippocampal formation. NeuroReport 5, 125–128. Hahn, J.S., Aizenman, E., Lipton, S.A., 1988. Central mammalian neurons normally resistant to glutamate toxicity are made sensitive by elevated extracellular Ca2+: toxicity is blocked by the N-methyl-D-aspartate antagonist MK-801. Proc. Natl. Acad. Sci. U. S. A. 85, 6556–6560. Hinds, H.L., Goussakov, I., Nakazawa, K., Tonegawa, S., Bolshakov, V.Y., 2003. Essential function of alpha-calcium/calmodulindependent protein kinase II in neurotransmitter release at a glutamatergic central synapse. Proc. Natl. Acad. Sci. U. S. A. 100, 4275–4280. Ishida, A., Kameshita, I., Okuno, S., Kitani, T., Fujisawa, H., 1995. A novel highly specific and potent inhibitor of calmodulindependent protein kinase II. Biochem. Biophys. Res. Commun. 212, 806–812. Ishida, A., Shigeri, Y., Tatsu, Y., Uegaki, K., Kameshita, I., Okuno, S., Kitani, T., Yumoto, N., Fujisawa, H., 1998. Critical amino acid residues of AIP, a highly specific inhibitory peptide of

calmodulin-dependent protein kinase II. FEBS Lett. 427, 115–118. Jiang, X., Zhu, D., Okagaki, P., Lipsky, R., Wu, X., Banaudha, K., Mearow, K., Strauss, K.I., Marini, A.M., 2003. N-methyl-Daspartate and TrkB receptor activation in cerebellar granule cells: an in vitro model of preconditioning to stimulate intrinsic survival pathways in neurons. Ann. N. Y. Acad. Sci. 993, 134–145. Johnson, J.E., Barde, Y.A., Schwab, M., Thoenen, H., 1986. Brainderived neurotrophic factor supports the survival of cultured rat retinal ganglion cells. J. Neurosci. 6, 3031–3038. Kaltschmidt, C., Kaltschmidt, B., Baeuerle, P.A., 1995. Stimulation of ionotropic glutamate receptors activates transcription factor NF-kappa B in primary neurons72. Proc. Natl. Acad. Sci. U. S. A. 92, 9618–9622. Kano, T., Abe, T., Tomita, H., Sakata, T., S.i. Ishiguro, M., 2002. Protective effect against ischemia and light damage of iris pigment epithelial cells transfected with the BDNF gene. Invest. Ophthalmol. Visual Sci. 43, 3744–3753. Kaplan, D.R., Miller, F.D., 2000. Neurotrophin signal transduction in the nervous system. Curr. Opin. Neurobiol. 10, 381–391. Kido, N., Tanihara, H., Honjo, M., Inatani, M., Tatsuno, T., Nakayama, C., Honda, Y., 2000. Neuroprotective effects of brain-derived neurotrophic factor in eyes with NMDA-induced neuronal death. Brain Res. 884, 59–67. Kitaoka, Y., Kumai, T., Kitaoka, Y., Lam, T.T., Munemasa, Y., Isenoumi, K., Motoki, M., Kuribayashi, K., Kogo, J., Kobayashi, S., Ueno, S., 2004. Nuclear factor-kappa B p65 in NMDA-induced retinal neurotoxicity. Mol. Brain Res. 131, 8–16. Krishnamoorthy, R.R., Agarwal, P., Prasanna, G., Vopat, K., Lambert, W., Sheedlo, H.J., Pang, I.H., Shade, D., Wordinger, R.J., Yorio, T., Clark, A.F., Agarwal, N., 2001. Characterization of a transformed rat retinal ganglion cell line. Brain Res. Mol. Brain Res. 86, 1–12. Laabich, A., Cooper, N.G., 2000. Neuroprotective effect of AIP on N-methyl-D-aspartate-induced cell death in retinal neurons. Brain Res. Mol. Brain Res. 85, 32–40. Laabich, A., Li, G., Cooper, N.G., 2000. Calcium/calmodulindependent protein kinase II containing a nuclear localizing signal is altered in retinal neurons exposed to N-methyl-Daspartate 3. Brain Res. Mol. Brain Res. 76, 253–265. Lauterborn, J.C., Lynch, G., Vanderklish, P., Arai, A., Gall, C.M., 2000. Positive modulation of AMPA receptors increases neurotrophin expression by hippocampal and cortical neurons. J. Neurosci. 20, 8–21. Lessmann, V., Gottmann, K., Malcangio, M., 2003. Neurotrophin secretion: current facts and future prospects. Prog. Neurobiol. 69, 341–374. Lindefors, N., Ballarin, M., Ernfors, P., Falkenberg, T., Persson, H., 1992. Stimulation of glutamate receptors increases expression of brain-derived neurotrophic factor mRNA in rat hippocampus. Ann. N. Y. Acad. Sci. 648, 296–299. Livak, K.J., Schmittgen, T.D., 2001. Analysis of relative gene expression data using real-time quantitative PCR and the 2-[Delta][Delta]CT method. Methods 25, 402–408. Lucas, D.R., Newhouse, J.P., 1957. The toxic effect of sodium L-glutamate on the inner layers of the retina. AMA Arch. Ophthalmol. 58, 193–201. Maher, P., Hanneken, A., 2005. The molecular basis of oxidative stress-induced cell death in an immortalized retinal ganglion cell line. Invest. Ophthalmol. Visual Sci. 46, 749–757. Marini, A.M., Jiang, X., Wu, X., Tian, F., Zhu, D., Okagaki, P., Lipsky, R.H., 2004. Role of brain-derived neurotrophic factor and NF-kappaB in neuronal plasticity and survival: from genes to phenotype. Restor. Neurol. Neurosci. 22, 121–130. Marmigere, F., Rage, F., Tapia-Arancibia, L., 2003. GABA-glutamate interaction in the control of BDNF expression in hypothalamic neurons. Neurochem. Int. 42, 353–358. Martin, K.R., Quigley, H.A., Zack, D.J., Levkovitch-Verbin, H.,

BR A IN RE S E A RCH 1 0 67 ( 20 0 6 ) 4 8 –5 7

Kielczewski, J., Valenta, D., Baumrind, L., Pease, M.E., Klein, R.L., Hauswirth, W.W., 2003. Gene therapy with brainderived neurotrophic factor as a protection: retinal ganglion cells in a rat glaucoma model. Invest. Ophthalmol. Visual Sci. 44, 4357–4365. Martin, P.M., Ola, M.S., Agarwal, N., Ganapathy, V., Smith, S.B., 2004. The sigma receptor ligand (+)-pentazocine prevents apoptotic retinal ganglion cell death induced in vitro by homocysteine and glutamate. Mol. Brain Res. 123, 66–75. Mattson, M.P., Culmsee, C., Yu, Z., Camandola, S., 2000. Roles of nuclear factor kappaB in neuronal survival and plasticity. J. Neurochem. 74, 443–456. Mey, J., Thanos, S., 1993. Intravitreal injections of neurotrophic factors support the survival of axotomized retinal ganglion cells in adult rats in vivo. Brain Res. 602, 304–317. Mo, X., Yokoyama, A., Oshitari, T., Negishi, H., Dezawa, M., Mizota, A., Adachi-Usami, E., 2002. Rescue of axotomized retinal ganglion cells by BDNF gene electroporation in adult rats. Invest. Ophthalmol. Visual Sci. 43, 2401–2405. Murer, M.G., Yan, Q., Raisman-Vozari, R., 2001. Brain-derived neurotrophic factor in the control human brain, and in Alzheimer's disease and Parkinson's disease. Prog. Neurobiol. 63, 71–124. Murray, K.D., Wood, P.L., Rosasco, C., Isackson, P.J., 1996. A metabotropic glutamate receptor agonist regulates neurotrophin messenger RNA in rat forebrain. Neuroscience 70, 617–630. Otori, Y., Wei, J.Y., Barnstable, C.J., 1998. Neurotoxic effects of low doses of glutamate on purified rat retinal ganglion cells. Invest. Ophthalmol. Visual Sci. 39, 972–981. Peinado-Ramon, P., Salvador, M., Villegas-Perez, M.P., Vidal-Sanz, M., 1996. Effects of axotomy and intraocular administration of NT-4, NT-3, and brain-derived neurotrophic factor on the survival of adult rat retinal ganglion cells: a quantitative in vivo study. Invest. Ophthalmol. Visual Sci. 37, 489–500. Pizzi, M., Goffi, F., Boroni, F., Benarese, M., Perkins, S.E., Liou, H.C., Spano, P., 2002. Opposing roles for NF-kappa B/Rel factors p65 and c-Rel in the modulation of neuron survival elicited by glutamate and interleukin-1beta. J. Biol. Chem. 277, 20717–20723. Roux, P.P., Barker, P.A., 2002. Neurotrophin signaling through the p75 neurotrophin receptor. Prog. Neurobiol. 67, 203–233. Rudzinski, M., Wong, T.P., Saragovi, H.U., 2004. Changes in retinal expression of neurotrophins and neurotrophin receptors induced by ocular hypertension. J. Neurobiol. 58, 341–354. Scholzke, M.N., Potrovita, I., Subramaniam, S., Prinz, S., Schwaninger, M., 2003. Glutamate activates NF-kappaB through calpain in neurons. Eur. J. Neurosci. 18, 3305–3310. Schulman, H., 2004. Activity-dependent regulation of calcium/

57

calmodulin-dependent protein kinase II localization. J. Neurosci. 24, 8399–8403. Sisk, D.R., Kuwabara, T., 1985. Histologic changes in the inner retina of albino rats following intravitreal injection of monosodium L-glutamate. Graefe Arch. Clin. Exp. Ophthalmol. 223, 250–258. Sucher, N.J., Aizenman, E., Lipton, S.A., 1991. N-methyl-Daspartate antagonists prevent kainate neurotoxicity in rat retinal ganglion cells in vitro. J. Neurosci. 11, 966–971. Sucher, N.J., Lipton, S.A., Dreyer, E.B., 1997. Molecular basis of glutamate toxicity in retinal ganglion cells. Vision Res. 37, 3483–3493. Takada, Y., Singh, S., Aggarwal, B.B., 2004. Identification of a p65 Peptide that selectively inhibits NF-{kappa} B activation induced by various inflammatory stimuli and its role in down-regulation of NF-{kappa}B-mediated gene expression and up-regulation of apoptosis. J. Biol. Chem. 279, 15096–15104. Takeuchi, Y., Yamamoto, H., Miyakawa, T., Miyamoto, E., 2000. Increase of brain-derived neurotrophic factor gene expression in NG108-15 cells by the nuclear isoforms of Ca2+/ calmodulin-dependent protein kinase II30. J. Neurochem. 74, 1913–1922. Thanos, S., Bahr, M., Barde, Y.A., Vanselow, J., 1989. Survival and axonal elongation of adult rat retinal ganglion cells. In vitro effects of lesioned sciatic nerve and brain derived neurotrophic factor. Eur. J. Neurosci. 1, 19–26. Timmusk, T., Metsis, M., 1994. Regulation of BDNF promoters in the rat hippocampus. Neurochem. Int. 25, 11–15. Uberti, D., Grilli, M., Memo, M., 2000. Contribution of NF-kappaB and p53 in the glutamate-induced apoptosis. Int. J. Dev. Neurosci. 18, 447–454. Unoki, K., LaVail, M.M., 1994. Protection of the rat retina from ischemic injury by brain-derived neurotrophic factor, ciliary neurotrophic factor, and basic fibroblast growth factor. Invest. Ophthalmol. Visual Sci. 35, 907–915. Vecino, E., Caminos, E., Ugarte, M., Martin-Zanca, D., Osborne, N.N., 1998. Immunohistochemical distribution of neurotrophins and their receptors in the rat retina and the effects of ischemia and reperfusion*. Gen. Pharmacol. 30, 305–314. Vecino, E., Ugarte, M., Nash, M.S., Osborne, N.N., 1999. NMDA induces BDNF expression in the albino rat retina in vivo. NeuroReport 10, 1103–1106. Vecino, E., Garcia-Grespo, D., Garcia, M., Martinez-Millan, L., Sharma, S.C., Carrascal, E., 2002. Rat retinal ganglion cells co-express brain derived neurotrophic factor (BDNF) and its receptor TrkB. Vision Res. 42, 151–157. Vorwerk, C.K., Lipton, S.A., Zurakowski, D., Hyman, B.T., Sabel, B.A., Dreyer, E.B., 1996. Chronic low-dose glutamate is toxic to retinal ganglion cells. Toxicity blocked by memantine. Invest. Ophthalmol. Visual Sci. 37, 1618–1624.