Protective Effect of Bradykinin against Glutamate Neurotoxicity in Cultured Rat Retinal Neurons

Abstract

purpose. To identify the localization and expression of bradykinin (BK)-B2 receptors in rat retina and examine the effects of BK on glutamate-induced neurotoxicity using cultured rat retinal neurons.

methods. An immunohistochemical study using a specific antibody against BK-B2 receptor was performed with rat retina. Primary cultures were obtained from the retina of fetal rats (gestation day 17–19). Expression of BK-B2 receptor mRNA was determined by reverse transcription–polymerase chain reaction (RT-PCR) using total RNA obtained from cultured retinal neurons. Cultured cells were exposed to glutamate (1 mM) for 10 minutes and followed by incubation in glutamate-free medium for 1 hour. The effects of BK were assessed by simultaneous application of BK with glutamate. The neurotoxic effects on retinal cultures were quantitatively assessed by the trypan blue exclusion method.

results. Immunohistochemical study demonstrated that BK-B2 receptors were expressed in the ganglion cell, inner nuclear layers, and outer nuclear layers. Furthermore, BK-B2 receptor mRNA expression was observed in cultured retinal neurons. Cell viability was markedly reduced by 10-minute exposure to 1 mM glutamate followed by a 1-hour incubation in glutamate-free medium. Simultaneous application of BK at concentrations of 0.001 to 1 μM with glutamate demonstrated dose-dependent protection against glutamate neurotoxicity. The protective action of BK (1 μM) was inhibited by simultaneous application of BK-B2 receptor antagonist, Hoe140 (1 μM). Furthermore, 1 μM BK had protective effects on neurotoxicity induced by 1 μM ionomycin, a calcium ionophore, and sodium nitroprusside (SNP, 500 μM), a nitric oxide (NO)–generating agent. However, BK did not inhibit neurotoxicity induced by 3-morpholinosydnonimine (SIN-1, 10 μM), an NO and oxygen radical donor.

conclusions. These results suggest that BK-B2 receptors were distributed in rat retinas and cultured retinal neurons and that BK had a protective action against glutamate neurotoxicity through BK-B2 receptors in cultured retinal neurons. It is suggested that BK-induced protection against glutamate neurotoxicity took place downstream to NO generation and upstream to oxygen radical generation.

Ischemic preconditioning is a phenomenon in which exposure of the tissues to a brief period of ischemia causes it to adapt itself quickly and become resistant to a subsequent prolonged ischemic insult. The phenomenon of ischemic tolerance was first documented in the myocardium. ¹ Thereafter, a similar phenomenon was reported in the brain. ² Recent study has indicated that this phenomenon also induces retinal tolerance to ischemia in vivo. ^{3 4} The mechanism underlying this phenomenon remains unclear, but it is known to be triggered by several endogenous mediators. Bradykinin (BK) has been proposed to play a central role in this phenomenon among several endogenous mediators. ⁵  

BK is a nonapeptide with a wide range of actions. It is known to be involved in inflammation, edema, pain, and contraction or relaxation of smooth muscles. The actions of BK are mediated through at least two subtypes of receptor, B1 and B2. ⁶ Most biologic actions are mediated by B2 receptors. It is reported that BK-B2 receptors are abundantly distributed in vascular tissues and smooth muscles cells but also in human brain ⁷ and retinal ⁸ neurons. Despite the abundance, the functional role of BK in the retinal neurons is not known. 

In our previous study, we showed the involvement of glutamate and nitric oxide (NO) neurotoxicity in ischemia–reperfusion-induced retinal injury in vivo. ⁹ Glutamate, one of the excitatory neurotransmitters in the retina, ^{10 11 12} has a toxic action ^{13 14 15} when it is present in excess under pathologic conditions, such as retinal hypoxia ¹⁶ and ischemia. ^{17 18 19}  

Therefore, the present study was undertaken to elucidate the effects of BK on glutamate-induced neurotoxicity mediated through N-methyl-d-aspartate (NMDA) receptors by using cultured rat retinal neurons. First, we identified the localization and expression of BK-B2 receptors in rat retina, because the distribution of BK-B2 receptors in rat retina was not previously known. Our study suggests that BK-B2 receptor stimulation provides protection for retinal neurons against glutamate neurotoxicity. The protective action is suggested to occur downstream to NO synthesis and is presumed to involve some process concerning oxygen radical formation in glutamate neurotoxicity. 

Materials and Methods

All animals were treated in accordance with the ARVO Statement for the Use of Animals in Ophthalmic and Vision Research. 

Immunohistochemistry

To determine the presence of BK-B2 receptors and identify the distribution of BK-B2 receptors in rat retina, immunohistochemical study using a specific antibody against BK-B2 receptor ²⁰ was performed with rat retina. The adult Wistar rats were anesthetized by inhalation of diethyl ether. The eyes were enucleated and the anterior segments removed to make eyecups. The eyecups were then embedded in OCT compound (Miles, Elkhart, IN) and 20-μm frozen sections were cut in a cryostat. The sections were air dried for 30 minutes at 4°C and then soaked for 30 minutes in 95% ethanol at 4°C and for 1 minute in acetone. The specimens were washed for 10 minutes with 0.1 M phosphate buffer (PB) and then incubated with 20% skim milk (Dainihon-Seiyaku, Osaka, Japan) in 0.1 M PB containing 0.005% saponin (Merck, Darmstadt, Germany) for 10 minutes to block nonspecific antibody binding. They were then incubated with primary antibody diluted in 5% skim milk in 0.1 M PB containing 0.005% saponin for 24 hours at 4°C. Antibody and concentration used in this study were rat monoclonal anti-BK-B2 receptor (1:1000; Peptide Institute, Osaka, Japan). On the following day, the specimens were washed with 0.1 M PB three times for 10 minutes and incubated with secondary antibody for 24 hours at 4°C: fluorescein isothiocyanate (FITC)-conjugated sheep anti-rabbit immunoglobulin (Amersham, Buckinghamshire, UK) diluted 1:100 in 0.1 M PB containing TO-PRO3 (1:100,0000; Molecular Probes, Eugene, OR). Sections were then washed with 0.1 M PB three times for 10 minutes and mounted with glycero-phosphate-buffered saline (PBS; 1:1). Negative control sections without primary antibody were processed under the same conditions. Sections were observed with a laser-scanning confocal microscope (model 1024; Bio-Rad, Hercules, CA). 

Cell Culture

Primary cultures were obtained from fetal Wistar rat retinas (17–19 days’ gestation). The procedures have been described previously. ^{15 21 22 23 24} In brief, retinal tissues were mechanically dissociated, and single-cell suspensions were plated on plastic coverslips (1.0 × 10⁶ cells/ml). Ten coverslips were placed in a 60-mm dish (Falcon Labware, Oxnard, CA). Approximately 15 to 20 dishes were obtained and used for a single experiment. Retinal cultures were incubated with Eagle’s minimal essential medium (Eagle’s salts; Nissui, Tokyo, Japan) containing 2 mM glutamine, 11 mM glucose (total), 24 mM sodium bicarbonate, and 10 mM HEPES with 10% heat-inactivated fetal calf serum added during the first week and supplemented with 10% horse serum for the remaining 10 to 11 days. Ten micromolar cytosine arabinoside (ara-C) was added to the culture on the sixth day to eliminate proliferating cells. We used only those cultures maintained for 9 to 10 days in vitro and used only isolated cells in this study. Clusters of cells were excluded from the results, because cells located in the clusters could not be used for histologic experiments. ²¹ A previous immunocytochemical study revealed that these isolated cells mainly consist of amacrine cells. ²¹  

RT-PCR and Agarose Gel Analysis

Expression of BK-B2 receptor mRNA was examined by reverse transcription–polymerase chain reaction (RT-PCR) using total RNA obtained from cultured retinal neurons. Total RNA was extracted from 5 × 10⁵ cultured retinal neurons by using a kit (QuickPrep Total RNA Extraction Kit; Pharmacia Biotech, Uppsala, Sweden) and a guanidine thiocyanate (GTC) method. All the total RNA obtained from preparation were resuspended in 10 to 20 μl RNase-free water, and the sample was used for the after RT-PCR. Synthesis of the first-strand cDNA was performed with a kit (TaKaRa RNA LA PCR Kit[ AMV]) Ver.1.1 (Takara, Siga, Japan) with the reaction mixture in a final volume of 20 μl containing 1 μl of the total RNA sample and 20 pM BK-B2 receptor downstream primer. All the first-strand cDNA obtained from the reverse transcriptase reaction were used for the subsequent PCR reaction. PCR was performed for 25 cycles (94°C for 30 seconds, 60°C for 30 seconds, 72°C for 90 seconds) in a thermal cycler. The first PCR reaction mixture in a final volume of 100 μl contained all the reverse transcription product, 20 pM upstream primer and TaKaRa LA Tag (Takara). Nested PCR reaction mixture in a final volume of 100 μl contained 5 μl of first PCR product, 100 pM each of both the primers and TaKaRa EX Tag (Takara). The PCR primers were designed using sequences in the coding regions of BK-B2 receptor (Accession Number: M59967) ²⁵ and β-actin (Accession Number: J00691) ²⁶ gene. BK-B2 receptor primers used were 5′-AAATGCACTGTTCTTGGAAGCGACC-3′ (first PCR, upstream, nucleotides 89–113), 5′-TGGCTTGTGTTCACTGCTTGTTCCC-3′ (first PCR, downstream, nucleotides 1490–1466), 5′-TCTGCCCGAAGACACAGGCTGTCGT-3′ (nested PCR, upstream, nucleotides 309–333) and 5′-TTCAGCAGCATGTTG-GTGAACACCT-3′ (nested PCR, downstream, nucleotides 966–990). β-Actin primers used were 5′-ACGATATGGAGAAGATTTGGCACCA-3′ (upstream) and 5′-ATAGTGATGACCTGACCGTCAGGCA-3′ (downstream). After PCR amplification, 12 μl of the products were loaded on 0.9% electrophoresed agarose gel and visualized by ethidium bromide staining. 

Drug Application

In our previous study using cultured rat retinal neurons, we demonstrated that cell viability was markedly reduced by exposure to glutamate (1 mM) for 10 minutes followed by postincubation in glutamate-free medium for more than 1 hour, ^{15 21 22} and we showed that there was no significant difference between the values of reduction in cell viability between 1-hour and 24-hour incubations. ²¹ Therefore, in this study, cultures were exposed to drugs as follows. Glutamate neurotoxicity was assessed by 10-minute exposure to 1 mM glutamate followed by 1-hour incubation in glutamate-free medium. Ionomycin, sodium nitroprusside (SNP) and 3-morpholinosydnonimine (SIN-1) were tested in a manner similar to glutamate. According to our previous study on the dose–response relationship in the neurotoxic effects of NO-generating agents, the concentrations at 500 μM consistently reduced cell viability to 30% to 40%. ¹⁵ Therefore, we used this concentration to examine NO-induced neurotoxicity. Effects of BK and D-Arg-[Hyp³, Thi⁵, D-Tic⁷, Oic⁸]-bradykinin (Hoe140) were assessed by simultaneous application of the drugs with glutamate. To investigate the effects of simultaneous drug application, drugs were added to the incubation medium during glutamate exposure and removed from culture medium during the postincubation period. 

The following drugs were used: monosodium l-glutamate (Nacalai Tesque, Kyoto, Japan), bradykinin (Peptide Institute), Hoe140 (Peptide Institute), ionomycin (Biomol Research, Plymouth Meeting, PA), SNP (Wako, Osaka, Japan), SIN-1 (Dojindo, Kumamoto, Japan). 

Measurement of Neurotoxicity

The neurotoxic effects of glutamate and the protective effects of drugs on the retinal cultures were quantitatively assessed by the trypan blue exclusion method, as described previously. ^{15 21 22 23 24 27} At each session of the experiment, we randomly chose five coverslips from different dishes, which constituted the number of samples (n = 5) for measurement of neurotoxicity. All experiments were performed in Eagle’s solution at 37°C. After the completion of drug treatment, cell cultures were stained with 1.5% trypan blue solution at room temperature for 10 minutes and were then fixed with isotonic formalin (pH 7.0, 2–4°C). The fixed cultures were rinsed with physiological saline and examined under Hoffman modulation microscopy at ×400 (Hoffman Modulation Optics, Greenvale, NY). More than 200 cells on each of five coverslips were randomly counted to determine the viability of cell culture. The cell counts were made by a blind observer. Viability of culture was calculated as the percentage of the ratio of the number of unstained cells (viable cells) to the total number of cells counted (viable cells plus nonviable cells). In each experiment, five coverslips were used to obtain mean values ± SEM of cell viability. The significance of data was determined by Dunnett’s two-tailed test. 

Results

Localization of BK-B2 Receptor

Immunohistochemical study using a specific antibody against BK-B2 receptor was performed in rat retina. Most cells in the ganglion cell layer (GCL), inner nuclear layer (INL), and outer nuclear layer (ONL) were stained with the BK-B2 receptor antibody (Fig. 1A ). In contrast, sections incubated without primary antibody showed no staining under identical incubation and development conditions (Fig. 1B) . This demonstrated that BK-B2 receptors were present and distributed at high levels in the GCL, INL, and ONL in rat retina. 

Expression of BK-B2 Receptor mRNA

Expression of BK-B2 receptor mRNA was examined by RT-PCR using total RNA obtained from cultured retinal neurons. The RT-PCR products were visualized on electrophoresed agarose gel stained with ethidium bromide (Fig. 2) . The expected size of the first RT-PCR products for BK-B2 receptor mRNA (1400 bp) was not identified, and then the nested PCR were performed. The expected size of RT-PCR products for BK-B2 receptor mRNA (680 bp) was identified in cultured retinal neuron (Fig. 2 , lane 1).β -Actin gene expression was determined by RT-PCR, which served as a control to assure the quality and quantity of the total RNA used (Fig. 2 , lane 3). Expression of BK-B2 receptor mRNA was detected in cultured retinal neurons. 

Effects of BK on Glutamate-Induced Neurotoxicity

Figure 3 demonstrates an example of the effect of BK on glutamate-induced neurotoxicity. Most cells in nontreated culture (control) were not stained by trypan blue (Fig. 3A) , which is normally excluded by living cells. However, numerous cells were stained by trypan blue, and cell viability was markedly reduced by 10-minute exposure to 1 mM glutamate followed by 1-hour incubation in glutamate-free medium (Fig. 3B) . Simultaneous application of BK (1 μM) with glutamate reduced the number of cells stained by trypan blue, and cell death was markedly reduced (Fig. 3C) . Furthermore, cells treated with both BK (1 μM) and Hoe140 (1 μM), a BK-B2 receptor antagonist, showed an increased number of stained cells, and cell viability was markedly reduced (Fig. 3D) . 

Figure 4 summarizes the dose–response effect of BK on glutamate-induced neurotoxicity. Cell viability was markedly reduced by 10-minute exposure to 1 mM glutamate followed by 1-hour incubation in glutamate-free medium. Simultaneous application of BK at concentrations of 0.001 to 1 μM with glutamate demonstrated dose-dependent protections against glutamate neurotoxicity. A significant difference (P < 0.01, by Dunnett’s two-tailed test) was noted between cell viability of cultures treated with BK at concentrations of 0.1 to 1 μM and that of glutamate-treated cultures. The maximal protection was observed in the culture treated with BK at a concentration of 1 μM. 

Effect of Hoe140 on BK-Induced Protection against Glutamate Neurotoxicity

To investigate whether the BK-induced neuroprotection is mediated by a specific BK receptor, the effect of a selective BK-B2 receptor antagonist was examined. Figure 5 shows the effect of Hoe140, a selective BK-B2 receptor antagonist, on BK-induced action against glutamate neurotoxicity. As shown in Figure 1C , Hoe140 and BK were added to the incubation medium during glutamate exposure and removed from culture medium followed by 1-hour incubation. The protective action of BK (1 μM) was inhibited by simultaneous application of Hoe140 (1 μM), whereas Hoe140 alone did not affect cell viability. 

Effects of BK on Ionomycin-, SNP- and SIN-1–Induced Neurotoxicity

Figure 6 summarizes the effect of BK on ionomycin-, SNP- and SIN-1–induced neurotoxicity. Cell viability was markedly reduced by 10-minute exposure to ionomycin (1 μM), a calcium ionophore; SNP (500 μM), an NO-generating agent; or SIN-1 (10 μM), an NO- and oxygen radical–generating agent, followed by 1-hour incubation in ionomycin-free, SNP-free, or SIN-1-free medium. Simultaneous application of BK (1 μM) with ionomycin or SNP demonstrated protective effects on neurotoxicity induced by ionomycin and SNP. By contrast, BK did not inhibit neurotoxicity induced by SIN-1. 

Discussion

Recently, Ma et al. ⁸ demonstrated that BK-B2 receptors are abundantly distributed in human retinal neuronal cells including the GCL, INL, and ONL. To our knowledge, distribution of BK-B2 receptors in rat retina has not yet been reported. Therefore, in the present study, we identified the localization and expression of BK-B2 receptors in rat retina. The immunohistochemical study using a specific antibody against BK-B2 receptor identified cellular localization of BK-B2 receptors. They were expressed at high levels in the GCL, INL, and ONL with a cellular localization similar to that demonstrated in human retina by in situ hybridization with antisense riboprobe of BK-B2 receptor. Furthermore, RT-PCR and agarose gel analysis have detected expression of BK-B2 receptor mRNA in cultured retinal neuron. This is the first report that the localization and expression of BK-B2 receptors are identified in rat retina. 

In this study, we demonstrate that the neurotoxic effect of glutamate was greatly reduced by simultaneous application of BK, and the protective effects of BK on glutamate-induced neurotoxicity were blocked by a BK-B2 receptor antagonist, Hoe140. These results suggest that protective effects of BK against glutamate neurotoxicity are mediated by BK-B2 receptors in cultured rat retinal neurons. 

We have demonstrated in the cultured retinal neurons that an influx of Ca²⁺ induced by stimulation of NMDA receptor, ^{15 21 22 28 29 30 31} a subtype of glutamate receptors, activates nitric oxide synthase (NOS) and that an excess amount of NO produced by activation of NOS interacting with oxygen radicals ^{15 21 22 23} mediates glutamate neurotoxicity. NO alone had no toxic effects on cultured retinal neurons. The present study demonstrated that BK had protective effects on neurotoxicity induced by ionomycin, a calcium ionophore, and SNP, an NO-generating agent, but did not inhibit neurotoxicity induced by SIN-1, an NO and oxygen radical–generating agent. ³² It is thus suggested that BK-induced protection against glutamate neurotoxicity took place downstream to NO generation and upstream to oxygen radical generation, although none of these molecules was analyzed by direct measurement. 

Intracellular Ca²⁺ overload is well known to trigger glutamate-induced neuronal death. In our cultured retinal neurons, Ca²⁺ is essential for glutamate neurotoxicity. ³³ In this context, it is interesting to note that BK is known to cause an increase in cytosolic free Ca²⁺ concentration in various cells and that actual stimulation of BK-B2 receptors even inhibited Ca²⁺-induced neuronal death in our cultured retinal neurons. Recently, Stout et al. ³⁴ demonstrated that very high levels of cytoplasmic Ca²⁺ are not necessarily toxic to cultured rat forebrain neurons and that potential-driven uptake of Ca²⁺ into mitochondria is required to trigger glutamate-induced neuronal death. Thus, it is tempting to speculate that the protective action seen in response to BK-B2 receptor stimulation against glutamate may involve some process of inhibiting mitochondrial Ca²⁺ uptake and mitochondrial membrane depolarization, resulting in reduced oxygen radical formation. However, the downstream consequences of glutamate toxicity and the role of mitochondria are still very controversial. Glutamate can produce oxygen radicals by various pathways, including membrane arachidonic acid release and metabolism, xanthine–xanthine oxidase activation, and even NOS activation. The mechanism of glutamate neurotoxicity that we hypothesized is one possibility. Further studies are necessary to determine the mechanism of the protective effect induced by BK against glutamate neurotoxicity. 

In conclusion, we have demonstrated that BK-B2 receptors were distributed in rat retinas in situ and in vitro, and that BK had a protective effect on neurotoxicity induced by glutamate through BK-B2 receptors in cultured retinal neurons. It is suggested that BK-induced protection against glutamate neurotoxicity took place downstream to NO generation and upstream to oxygen radical generation. 

 Submitted for publication October 7, 1999; revised January 19, 2000; accepted January 26, 2000.

 Commercial relationships policy: N.

 

Figure 1.

 

Immunohistochemical localization of BK-B2 receptors in rat retina. (A) Most cells in the GCL, INL, and ONL were stained with a specific antibody against BK-B2 receptor. (B) In contrast, sections incubated without primary antibody showed no staining under identical incubation and development conditions.

Figure 2.

 

Agarose gel analysis of RT-PCR products of total RNA obtained from cultured retinal neuron. After PCR amplification, the products were visualized on electrophoresed agarose gel stained with ethidium bromide. The expected size of RT-PCR products for BK-B2 receptor mRNA (680 bp) was identified in cultured retinal neuron (lane 1). β-Actin gene expression (510 bp) was determined by RT-PCR, which serves as a control to assure the quality and quantity of the total RNA used (lane 3). Lanes 2 and 4: control reactions performed in the absence of reverse transcriptase. A band observed in lane 2 is genomic DNA contamination. Lane M: 50- to 2500-bp DNA marker.

Figure 3.

 

Photomicrographs showing the effect of bradykinin (BK) on glutamate-induced neurotoxicity in cultured rat retinal neurons. All cultures were photographed after trypan blue staining followed by formalin fixation using modulation microscopy. Cells stained with trypan blue dye were regarded as nonviable. (A) Nontreated cells (control). Cells showed almost no stain. (B) Cells treated with glutamate (1 mM) for 10 minutes, followed by a 1-hour incubation with glutamate-free medium. Marked cell death occurred. (C) Cells treated with simultaneous application of BK (1μ M) with glutamate (1 mM) for 10 minutes, followed by a 1-hour incubation with glutamate- and BK-free medium. Cell death was markedly reduced. (D) Cells treated with simultaneous application of both BK (1 μM) and Hoe140 (1 μM), BK-B2 receptor antagonists, with glutamate (1 mM) for 10 minutes, followed by a 1-hour incubation with glutamate-, BK-, and Hoe140-free medium. The number of stained cells increased, and cell viability was markedly reduced. Bar, 50 μm.

Figure 4.

 

The dose–response effects of BK on glutamate-induced neurotoxicity. Cell viability was markedly reduced by 10-minute exposure to 1 mM glutamate followed by 1-hour incubation in glutamate-free medium. Simultaneous application of BK at concentrations of 0.001 to 1 μM with glutamate demonstrated dose-dependent protection against glutamate neurotoxicity. The maximal protection was observed in the culture treated with BK at a concentration of 1 μM. (*P < 0.05, **P < 0.01, compared with the glutamate-only group). Error bars in this and the subsequent figure represent SEMs (n = 5).

Figure 5.

 

The effect of Hoe140, a selective BK-B2 receptor antagonist, on BK-induced protection against glutamate neurotoxicity. Hoe140 and BK were added to the incubation medium during glutamate exposure and were removed from culture medium followed by 1-hour incubation. The protective action of BK (1 μM) was inhibited by simultaneous application of Hoe140 (1 μM), whereas Hoe140 alone did not affect cell viability. (**P < 0.01, compared with glutamate-only group).

Figure 6.

 

The effect of BK on ionomycin-, SNP- and SIN-1–induced neurotoxicity. (A) Cell viability was markedly reduced by 10-minute exposure to ionomycin (1 μM), a calcium ionophore, followed by a 1-hour incubation in ionomycin-free medium. Simultaneous application of BK (1 μM) with ionomycin demonstrated protective effects on neurotoxicity induced by ionomycin. (**P < 0.01, compared with the ionomycin-only group). (B) Cell viability was markedly reduced by 10-minute exposure to SNP (500 μM), an NO-generating agent, followed by 1-hour incubation in SNP-free medium. Simultaneous application of BK (1 μM) with SNP demonstrated protective effects on neurotoxicity induced by SNP. (**P < 0.01, compared with the SNP-only group). (C) Cell viability was markedly reduced by 10-minute exposure to SIN-1 (10 μM), an NO and oxygen radicals donor, followed by 1-hour incubation in SIN-1-free medium. Simultaneous application of BK (1 μM) with SIN-1 did not inhibit neurotoxicity induced by SIN-1.