RETRACTED: Characterization of a transformed rat retinal ganglion cell line

Molecular Brain Research 86 (2001) 1–12 www.elsevier.com / locate / bres

Research report

Characterization of a transformed rat retinal ganglion cell line R.R. Krishnamoorthy a , P. Agarwal b , G. Prasanna c , K. Vopat a , W. Lambert a , H.J. Sheedlo a , I.-H. Pang c,d , D. Shade c,d , R.J. Wordinger a , T. Yorio c , A.F Clark a,c,d , N. Agarwal a , * a

Department of Pathology and Anatomy, UNT Health Science Center, 3500 Camp Bowie Blvd., Fort Worth, TX 76107, USA b Texas Academy of Math and Science, UNT, Denton, TX, USA c Department of Pharmacology, UNT Health Science Center, Fort Worth, TX, USA d Alcon Research Ltd., Fort Worth, TX, USA Accepted 29 August 2000

Abstract The purpose of the present study was to establish a rat retinal ganglion cell line by transformation of rat retinal cells. For this investigation, retinal cells were isolated from postnatal day 1 (PN1) rats and transformed with the c2 E1A virus. In order to isolate retinal ganglion cells (RGC), single cell clones were chosen at random from the transformed cells. Expression of Thy-1 (a marker for RGC), glial fibrillary acidic protein (GFAP, a positive marker for Muller cells), HPC-1 / syntaxin (a marker for amacrine cells), 8A1 (a marker for horizontal and ganglion cells) and neurotrophins was studied using reverse transcriptase-polymerase chain reaction (RT-PCR), immunoblotting and immunocytochemistry. One of the retinal cell clones, designated RGC-5, was positive for Thy-1, Brn-3C, Neuritin, NMDA receptor, GABA-B receptor, and synaptophysin expression and negative for GFAP, HPC-1, and 8A1, suggesting that it represented a putative RGC clone. The results of RT-PCR analysis were confirmed by immunocytochemistry for Thy-1 and GFAP. Upon further characterization by immunoblotting, the RGC-5 clone was positive for Thy-1, negative for GFAP, 8A1 and syntaxin. RGC 5 cells were also positive for the expression of neurotrophins and their cognate receptors. To establish the physiological relevance of RGC-5, the effects of serum / trophic factor deprivation and glutamate toxicity were analyzed to determine if these cells would undergo apoptosis. The protective effects of neurotrophins on RGC-5 after serum deprivation was also investigated. Apoptosis was studied by terminal deoxynucleotidyl transferase-mediated fluoresceinated dUTP nick end labeling (TUNEL). Serum deprivation resulted in apoptosis and supplementation with both BDNF and NT-4 in the growth media, protected the RGC-5 cells from undergoing apoptosis. On differentiation with succinyl concanavalin A (sConA), RGC-5 cells became sensitive to glutamate toxicity, which could be reversed by inclusion of ciplizone (MK801). In conclusion, a transformed rat retinal cell line, RGC-5, has certain characteristics of retinal ganglion cells based on Thy-1 and Brn-3C expression and its sensitivity to glutamate excitotoxicity and neurotrophin withdrawal. These cells may be valuable in understanding of retinal ganglion cell biology and physiology including in vitro manipulations in experimental models of glaucoma.  2001 Elsevier Science B.V. All rights reserved. Theme: Development and regeneration Topic: Neuronal death Keywords: Thy-1; Glial fibrillary acidic protein; Glaucoma; c2E1A virus; Serum deprivation; Apoptosis; Neurotrophin; Excitotoxicity

1. Introduction Glaucoma is a heterogeneous group of optic neuropathies characterized by cupping of the optic disc and optic nerve degeneration resulting in a progressive loss of the *Corresponding author. Tel.: 11-817-735-2094; fax: 11-817-7352610. E-mail address: [email protected] (N. Agarwal).

visual field, which may cause blindness. Although the mechanisms of optic nerve damage in glaucoma have not been completely determined, it appears that the optic nerve head is a major site of damage [37]. While several etiologies of glaucoma exist, it is thought that the ultimate cause of vision loss is through apoptosis of retinal ganglion cells [37]. Apoptotic cells are characterized by condensation of the cell nucleus, chromosomal DNA fragmentation, and formation of multiple membrane

0169-328X / 01 / $ – see front matter  2001 Elsevier Science B.V. All rights reserved. PII: S0169-328X( 00 )00224-2

R.R. Krishnamoorthy et al. / Molecular Brain Research 86 (2001) 1 – 12

2

blebs. The apoptotic bodies are phagocytized by surrounding healthy cells [24]. Apoptosis of RGCs was first shown in a rat optic nerve axotomy model [9,18,19,25,38,39,]. Other investigators also found apoptosis of the RGCs in the monkey model of experimental glaucoma [19,37]. Studies in human patients with primary open angle glaucoma revealed that more than 50% of the patients showed TUNEL-positive labeling in the RGC layer, compared to less than 10% TUNEL-positive cells in the control retinas [39]. Thus, in vivo studies have shown that glaucoma is associated with RGC death via apoptosis. Currently, there is no clear understanding of the mechanism(s) of apoptotic cell death of RGCs in glaucoma. The fact that there are no established RGC cultures available makes it even more difficult to manipulate these cells with a view to study apoptotic mechanisms of RGC death. Although a number of laboratories have been successful in growing RGCs in primary culture from isolated neonatal retinas, the number of viable RGCs decreases rapidly [32,40,43]. Thy-1 antigen is a glycoprotein found in highest concentrations in brain and thymus; and in the retina, where it is specifically expressed in retinal ganglion cell layer [5,8,22,33]. Furthermore, Thy-1 labeling as well as retrograde labeling of granular blue or other fluorescent label has been used as a specific marker for RGCs [8,21,29,43,53]. A number of laboratories have emphasized the need for having a RGC cell line [7,14,22,23,25,48,52]. In order to overcome the labor intensive process of growing primary cultures of RGCs, a permanently transformed RGC culture may provide a convenient system to investigate the mechanisms of apoptotic cell death, as well as to test the effects of various growth factors on the survival and regeneration of these cells [25]. The objective of the present study was to establish a permanent stable transformed rat RGC culture. To accomplish this clonal populations of cells were isolated from a virally transformed mixed retinal cell culture from PN1 rats. One prospective clone, RGC-5, was determined to be of ganglion cell origin based upon expression of the Thy-1 surface antigen, Brn-3C, Neuritin, NMDA receptor, GABA-B receptor, synaptophysin expression and neurotrophins and their receptors. The RGC-5 cells, when serum deprived, underwent apoptosis and inclusion of neurotrophins such as brain-derived neurotrophic factor (BDNF) and neurotrophin-4 (NT-4) in their growth medium protected them from undergoing apoptosis. Thus, the RGC-5 cell line appears be useful to study various aspects of RGC biology including the mechanisms of ganglion cell death in glaucoma. 2. Materials and methods

2.1. Materials The c2E1A virus was a kind gift from Dr. Gail Seigel

[45] of University of Rochester, Rochester, NY, who received this virus originally from Dr. Roger Cone’s laboratory at Cold Spring Harbor, NY. The following antibodies were purchased commercially: Thy-1 (Santa Cruz Biotechnologies, Santa Cruz, CA), GFAP (Dako, Glostrup, Denmark), and 8A1 and HPC-1 / syntaxin antibodies (Leinco Technologies, St. Louis, MO). A Tropix Chemiluminescence Kit (alkaline phosphatase) was purchased from Tropix (Milford, MA). Dulbecco’s minimum essential medium (DMEM) was from Cellgro, Mediatech (Herndon, VA). The neurotrophins were purchased from Sigma (St. Louis, MO). AMV reverse transcriptase, dNTPs, random primers, Rnasin, and Taq polymerase were all purchased from Promega, Madison, WI. Antibody to Taq Polymerase was purchased from Clonetech Laboratories, Palo Alto, CA.

2.2. Culture of mixed retinal cells Rats were maintained and treated in accordance with NIH guidelines. Rat retinal cells were isolated as reported earlier [36]. Briefly, retinas from PN1 non-pigmented Sprague–Dawley rats were dissected in Dulbecco’s modified Eagle’s media (nutrient mixture F12, DMEM / F12, Gibco, Grand Island, NY). Retinas were dissociated following incubation with a digestion solution containing 10 mg papain (34 units / ml) (Sigma), 2 mg DL-cysteine (3.3 mM; Sigma) and 2 mg bovine serum albumin (0.4 mg / ml; Sigma) in 5 ml of DMEM / F12 for 25 min at 378C. The cells were washed 3 times with 5 ml of media, which contains DMEM (Life Technologies, Gibco-BRL, Gaithersburg, MD), supplemented with 10% fetal bovine serum (Hyclone, Logan, UT), 4 mM glutamine (Gibco), 100U / ml penicillin and 100 mg / ml streptomycin (Sigma). All cultures were maintained in growth medium in a humidified atmosphere of 95% air and 5% CO 2 at 378C. After 4 h in culture, the cells were overlaid with c2 E1A viral culture supernatant and incubated for 4 h. The viral culture supernatant was discarded, fresh virus was added and incubated for an additional 4 h. Cultures of mixed retinal cells were washed free of viral supernatant by rinsing several times with fresh media. Mixed retinal cells were selected by G418 (geneticin, a neomycin analog, Life Technologies, Gibco-BRL) and infected cells expressing the viral neomycin / G418 resistance gene survived. Since the whole retina was used, a mixed population of cells was transformed including ganglion cells. The c2 E1A retrovirus used was replication incompetent and, therefore, no infective virus was present after the initial infection of retinal cells.

2.3. Reverse transcription-polymerase chain reaction ( RT-PCR) analysis Total RNA from 14 putative clones was extracted using the RNAzol B reagent (Tel-Test, Friendswood, TX) and

R.R. Krishnamoorthy et al. / Molecular Brain Research 86 (2001) 1 – 12

subjected to cDNA synthesis using AMV reverse transcriptase. PCR primers for Thy-1, Brn-3C, Neuritin, NMDA receptor, GABA-B receptor, synaptophysin, GFAP, ciliary neurotrophic factor (CNTF), brain-derived neurotrophic factor (BDNF), glial cell-derived neurotrophic factor (GDNF), neurotrophin-3 (NT-3), neurotrophin-4 (NT-4), nerve growth factor (NGF), neuritin, Trk-A, TrkB, Trk-C, and p75NTR are listed in Table 1. The primers were purchased either commercially or designed from published sequences using Oligo Software (National Biosciences, Plymouth, MN) and Primer 3 (MIT, Cambridge, MA) program downloaded from the internet [1]. All test samples were amplified simultaneously with a particular primer pair as per the annealing temperature of the individual set of primers, using a master mix containing all of the components in the PCR reaction, except the target cDNA, or in the case of the control, water. Although all of the RT-PCR products gave the expected product size, the authenticity of these products

3

was established by either Southern blot hybridization or by sequencing.

2.4. Immunoblot analysis Rat RGC-5 cells, and isolated rat retinas were homogenized in a buffer containing 0.25 M sucrose, 10 mM Hepes, 1 mM MgCl 2 , pH 6.5, with a mixture of protease inhibitors (10 mg / ml antipain, 2 mM leupeptin and 10 KIU / ml aprotinin). Total homogenate protein (20–50 mg) was separated in 12% sodium dodecyl sulfate (SDS)–polyacrylamide gels [28]. Following electrophoresis, the gels were equilibrated in transfer buffer (192 mM glycine, 20% methanol and 25 mM Tris–HCl, pH 8.3) for 30 min and electroblotted to nitrocellulose for 1 h at 110 V using a Bio-Rad electroblotting unit. Prior to immunoblotting, the nitrocellulose membranes were blocked with I-Block from Tropix Chemiluminescence Kit (for monoclonal primary antibodies) and 4% normal goat serum in PBS (for

Table 1 PCR primer sequences used for screening gene expression by RGC-5 cells and their product size Gene

Primer sequence (59–39)

Size (bp)

Thy-1

(S) TGCCTGGTGAACCAGAACCTT (A) TCACAGAGAAATGAAGTCCGTGGC (S) ACATCGAGATCGCCACCTAC (A) ATCTCCACGGTCTTCACCAC (S) TGTGATGGTGGGAATGGGTCAG (A) TTTGATGTCACGCACGATTTC (S) ACCCAAATTCTCCAGTCTGCACTCTG (A) GCGTGCCAGCAGGCTCTCATCAAAGC (S) CCAAGGGAGCAGCTTTCTATCCTGG (A) GGCAGTGTCAAGGGAATGCTGAAGT (S) GATCCAGGCGGATATCTTGA (A) AATCATCGGCTGGAATTCTG (S) CTCCTGAGTGGGACCTCTTG (A) CACTCACTGCATCGCACAC (S) GGGACTTAAGTTGAACGGCA (A) ACCCAGCTTGAGCAAACAGT (S) TGAAGGAAAGGGGACAGCTA (A) TGCTGTTAGTGTCTGTGGGG (S) CTTTTGGTAATGCTGTTTCTG (A) CGCGGCGATCTGCTGGGCTAT (S) GAGTCTGATGCGAGCCCTAC (A) GACCGTCGACCGTACTTGTT (S) AGCCAACCAGACCGTGTGTG (A) TTGCAGCTGTTCCACCTCTT (S) TGGCTTTCACAGAGCATTCA (A) GAGCAGTCAGGTCTGAACGA (S) CCTCCTCTTCTCTTTCTGCTG (A) GGCAGTGTCAAGGGAATGCTGAAGT (S) CAGAGGGAAAGGTCGCAGAGG (A) TGGAGCCAGGGTCAGATACAT (S) CCGGAACCAAAAGCTGATTA (A) GTGTAGCCCACGGTGAGAAT (S) AACGACCACTTCACTCCCAC (A) GGACAGAGGGCTTGTCAGAG (S) ACTACTCCTCGTCGGCTGAA (A) GTTCCTTGCATGTGTTCCCT

415

GFAP b-Actin Brn 3-C NGF NT-3 NT-4 Neurtin TrkA TrkB TrkC P75 CNTF BDNF GDNF GABA-B-R NMDA-R Synaptophysin

174 535 450 189 151 207 142 133 648 160 663 101 369 355 186 112 294

4

R.R. Krishnamoorthy et al. / Molecular Brain Research 86 (2001) 1 – 12

polyclonal antibodies), for 30 min. Membranes were incubated individually with monoclonal anti-8A1 (1:500), monoclonal anti HPC-1 (1:500) or polyclonal anti-GFAP (1:1000, Dako, Glostrup, Denmark) in phosphate-buffered saline (PBS) for 2 h. The membranes were then incubated with appropriate secondary antibodies conjugated to alkaline phosphatase (1:20 000) for 30 min. X-ray film was exposed to membranes for 30 min and later developed to visualize protein bands which was compared with the known molecular weight markers. For Thy-1 antigen expression, membranes containing cell lysates (50 mg protein) of rat RGC-5 cells, and retinas were incubated with monoclonal anti-Thy-1 (10 mg; Pharmingen, San Diego, CA) and rotated for 2 h at 48C. After this step, 150 ml of Pansorbin-A cells (Calbiochem, La Jolla, CA) was added to each sample and rotated for another 1.5 h at 48C. The samples were then quickly centrifuged and the supernatant was discarded. SDS-sample buffer (80 ml) was added to the pellets and the samples were placed on ice. The samples were boiled for 5 min, quickly centrifuged and proteins were separated in 12% SDS–PAGE gels and electroblotted as described above using the monoclonal anti-Thy-1 antibody (1:1000) (Santa Cruz Biotechnologies).

2.5. Immunocytochemistry RGC-5 cells were seeded onto 12-mm glass coverslips, then fixed in buffered 4% paraformaldehyde in PBS (pH 7.4) for 30 min. The fixed cells were subjected to immunolocalization studies for Thy-1 (1:500) and GFAP (1:500) using specific antibodies as described earlier using appropriate fluorescent second antibodies [2,27].

2.6. Serum deprivation of retinal ganglion cells RGC-5 cells were subjected to serum deprivation by first washing RGC cultures with Eagle’s balanced salt solution (EBSS) several times before incubating in serum-free DMEM at 5% CO 2 and 95% air for various time periods. Apoptosis of RGC-5 cells was monitored during a time course of serum deprivation for 2, 3, 4, and 6 days. Morphological changes, with respect to cell shape and cell rounding was also monitored microscopically. To determine the effect of serum deprivation on cell viability and apoptosis, RGC-5 cells were seeded in 96-well plates. One set of wells containing RGC-5 cells were maintained in serum-free media and compared with another set of cells maintained in complete media, using a formazan assay kit (Promega, Madison, WI). We also determined if the inclusion of neurotrophins (NGF, BDNF, NT-3 and NT-4 at 100 or 500 ng / ml for 2 days) in the serum-free media could protect the serum-deprived RGC-5 cells from undergoing apoptosis. The number of cells surviving under serum-deprived and neurotrophin supplemented conditions were monitored by formazan assay kit [47].

2.7. 39 end labeling of fragmented DNA by TUNEL RGC-5 cells were subjected to serum deprivation as described above. Apoptosis of RGCs was monitored during a time course of serum deprivation for 2, 3, 4 and 6 days. The TUNEL procedure as described [2,20] was employed using a commercially available fluorescent apoptosis kit (in situ cell death detection kit, BoehringerMannheim, Indianapolis, IN) as per the supplier’s instructions.

2.8. Succinyl concanavalin-A induced differentiation of RGC-5 cells Differentiation of RGC-5 cells was performed as previously described with minor modifications [46]. Briefly, the cells were cultured in the absence of serum for 24 h, after which the cells were cultured in growth media supplemented with succinyl concanavalin A (sConA; 50 mg / ml; Sigma) for 7 days. Incubation of the cells with sConA induced differentiation and rendered the cells responsive to glutamate toxicity.

2.9. Glutamate toxicity study of RGC-5 cells The response of RGC-5 cells to glutamate was tested using a modified procedure [36]. Differentiated or undifferentiated RGC-5 cells were serum deprived for 24 h, then treated with 500 mM of L-glutamate for 24 h both in presence and absence of 1 mM dizocilpine (MK-801). Cytotoxicity was evaluated using the Neutral red (Gibco) uptake viability assay, which was modified from that reported by Borenfreund and Puerner [10]. Neutral red dye was added to a final concentration of 0.033% to cell culture media after the cells were treated with the indicated compounds. Cells were then incubated an additional 2 h and gently washed with 2 volumes of a Hepes buffer (125 mM NaCl, 5 mM KCl, 1.8 mM CaCl 2 , 2 mM MgCl 2 , 0.5 mM NaH 2 PO 2, 5 mM NaHCO 3 , 10 mM D-glucose, 10 mM Hepes, pH 7.2). The cells were allowed to air dry for 20 min, then treated with ice-cold solubilization buffer (1% acetic acid / 50% ethanol; 300 ml). Twenty minutes later, 100 ml aliquots were transferred to wells of flat-bottomed 96-well plates and optical densities of samples were read at 570 nm.

3. Results

3.1. Culture of RGC-5 cells Fourteen individual clones were selected from the mixed transformed retina cell cultures. Clone RGC-5 was selected based on its positive Thy-1 expression. RGC-5 cells have been maintained in culture for over 50 passages with no apparent slowing of mitotic activity or loss of Thy-1

R.R. Krishnamoorthy et al. / Molecular Brain Research 86 (2001) 1 – 12

5

3.2. Expression of neuronal markers by RGC-5 cells

Fig. 1. Morphology of RGC-5 cells. Cells from the mixed retinal cultures were subjected to clonal selection and 14 clones were picked at random. Clone RGC-5 is shown. The culture is characterized by neuronal-like cells with extended neurites (arrows).

expression. Morphological characterization of RGC-5 cells showed that they grew as a monolayer and exhibited axonal processes (Fig. 1). Later, the cells grew to contact each other through projected extensions and often multiple cells were interconnected (Fig. 1). RGC-5 cells grew to confluence in culture and proliferated rapidly with a doubling rate of approximately 24 h (data not shown).

RT-PCR analysis was performed to study the expression of Thy-1 as well as other cell markers (Fig. 2). The results of RT-PCR analysis showed that RGC-5 cells expressed Thy-1, but were negative for GFAP expression. The RGC5 cells were found to express various neurotrophins including NGF, BDNF, NT-3, and NT-4 and other neurotrophic factors such as CNTF and GDNF. They also expressed various neurotrophin receptors such as Trk-A, Trk-B, Trk-C and p75NTR. RGC-5 cells were positive for Brn-3C (another ganglion cell marker), neuritin and synaptophysin (neuronal cell markers). Low levels of expression of both NMDA receptors, and GABA-B receptors were also observed, compared to retinas. BALB / c mouse retinas were included as a positive control for these studies (Fig. 2). It was established in separate experiments that all primers including Thy-1 gave a positive RT-PCR reaction for both mouse and rat retinas (data not shown).

3.3. Immunocytochemical staining for Thy-1 and GFAP in RGC-5 cells To confirm and extend the results of RT-PCR analyses, we performed immunocytochemistry on RGC-5 cells using

Fig. 2. RT-PCR analysis of the RGC-5 clone for expression of various retinal ganglion cell markers and neurotrophins and their receptors. RGC-5 cells express mRNA for Thy-1, NMDA receptor, GABA-B receptor, and synaptophysin, Brn-3C and neuritin but were negative for GFAP expression (A). The RGC-5 cells express various neurotrophins including NT-3, NT-4, NGF, and BDNF and neurotrophic factors, CNTF, and GDNF. They also express the neurotrophin receptors Trk-A, Trk-B, Trk-C, and p75 (B). BALB / c retina was included as a positive control for these studies.

6

R.R. Krishnamoorthy et al. / Molecular Brain Research 86 (2001) 1 – 12

Fig. 3. Immunolocalization studies for Thy-1 and GFAP in RGC-5 cells. The RGC-5 clone was positive for Thy-1, but negative for GFAP. A single typical control without primary antibody is included. These results support that the RGC-5 clone, though transformed, represents retinal ganglion-like cells.

Thy-1 and GFAP antibodies. The results indicate that the RGC-5 cells showed a diffuse expression of Thy-1 antigen (Fig. 3). RGC-5 cells were analyzed for GFAP expression, a marker for Muller cells, by immunocytochemistry and were negative for GFAP labeling.

3.4. Immunoblot analysis of RGC-5 cells for expression of Thy-1, GFAP, HPC-1 and 8 A1 proteins The expression of Thy-1 in RGC-5 cells as determined by RT-PCR and immunocytochemistry was further confirmed by immunoblot analysis. In addition, immunoblot analyses was performed for GFAP, HPC-1 (syntaxin) and 8A1 proteins. Rat retinas were used as positive controls, for this study. RGC-5 cells expressed a Thy-1-positive band (26 kDa) on immunoblots (Fig. 4). As expected, rat retinas also showed the presence of Thy-1 proteins. GFAP was barely detectable in RGC-5 cells as compared to normal rat retinas, in which it was expressed at low levels as a 55-kDa band. This is not an unexpected result, since GFAP is expressed at low levels in normal retina and its expression is increased in dystrophic retinas [16,41]. RGC5 cells did not express HPC-1 (35-kDa band), a marker for amacrine cells [6] or 8A1 (70-kDa band) a neurofilament marker of retinal horizontal and ganglion cells. In contrast, these proteins were expressed in rat retinas.

3.5. Glutamate toxicity of undifferentiated and differentiated RGC-5 cells

Fig. 4. Immunoblot analysis of RGC-5 clone for Thy-1, GFAP, HPC-1, and 8A1 proteins. Lysates from the RGC-5 cells, and rat retinas were subjected to SDS–PAGE followed by immunoblotting using specific antibodies for Thy-1, GFAP, 8A1 and HPC-1 (Syntaxin). The primary antibodies were visualized using alkaline phosphatase linked secondary antibody (chemiluminescence). RGC-5 cells express Thy-1, but not GFAP (Muller cell marker) and HPC-1 (amacrine cell marker).

Undifferentiated RGC-5 cells were not sensitive to glutamate-induced toxicity. Cell survival was not affected by incubation with as high as 1 mM glutamate for as long as 3 days of incubation. RGC-5 cells began to show signs of cytotoxicity only when treated with 2 mM or higher concentrations of glutamate. When RGC-5 cells were pretreated with sConA, they became sensitive to the cytotoxic effects of glutamate. A preliminary dose response study indicated an EC 50 of approximately 50 mM for glutamate excitotoxicity with near maximal effects at 500 mM. The treatment of sConA RGC-5 cells with 500 mM glutamate caused about 50% cell death that was almost totally abrogated by treatment with the NMDA antagonist, dizocilpine (Fig. 5). The glutamate-induced excitotoxic cell death of sConA RGC-5 cells was very similar to that seen in primary rat retinal ganglion cell cultures [36].

R.R. Krishnamoorthy et al. / Molecular Brain Research 86 (2001) 1 – 12

7

3.6. Serum deprivation causes apoptosis of RGC-5 cells

Fig. 5. Effects of glutamate on survival of RGC-5 cells with or without differentiation with sConA. RGC-5 cells were grown in 500 mM glutamate in presence or absence of 1 mM dizocilpine (NMDA receptor antagonist) for 24 h. Bars represent mean cell number and S.E.M. of each group (n56). The mean survival levels of groups without glutamate treatment correspond to 100%. The * indicates statistical difference (P,0.05, ANOVA, then Dunnett’s test) from the control group of undifferentiated cells.

RGC-5 cells were maintained in serum-free growth media for various time periods up to 6 days in order to study the effect of serum deprivation and withdrawal of neurotrophins. Changes in RGC-5 cell shape was observed as early as 2 days in serum-free conditions (Fig. 6). A large population of the cells started to round up by days 4 and 6 and the number of cells appeared to be considerably reduced. TUNEL assay was performed to further confirm that death of RGC-5 cells following serum deprivation was due to apoptosis. Control RGC-5 cells were maintained in complete media. Apoptosis of RGC-5 cells was monitored during 1–6 days of serum deprivation. No significant change was seen when these cells were maintained up to 24 h in serum-free media. However, after 2 days in serum-free media, a number of cells underwent apoptosis (Fig. 7) which continued for up to 6 days of serum deprivation. These data suggest that RGC-5 cells are dependent on trophic support for survival, which is one of the important characteristics of most neurons, including the retinal ganglion cells.

Fig. 6. Effects of serum deprivation on RGC-5 cells. RGC-5 cells were transferred to serum-free growth medium and maintained for various time periods up to 6 days. Control RGC-5 cells were maintained in complete growth medium. The cells were observed microscopically for morphological changes. Changes in cell shape and rounding up of cells were observed as early as 2 days after serum deprivation and continued for up to days 4 and 6.

8

R.R. Krishnamoorthy et al. / Molecular Brain Research 86 (2001) 1 – 12

Fig. 7. Apoptosis of RGC-5 cells following serum deprivation. Loss of RGC-5 cells on serum deprivation was due to apoptosis as shown by TUNEL assay. Apoptosis of RGC-5 cells was monitored during the time course of serum deprivation. The bright fluorescent nuclei represent the apoptotic nuclei of RGC-5 cells after 2–6 days of serum deprivation.

3.7. BDNF and NT-4 rescue RGC-5 cells from cell death by serum deprivation A cell survival assay was used to determine the number of RGC-5 cells surviving in serum-deprived conditions following neurotrophin supplementation. BDNF and NT-4 at 500 ng / ml were able to protect RGC-5 cells from undergoing apoptosis when cultured in conditions of serum deprivation for 2 days (Fig. 8). However, other neurotrophins, such as NGF and NT-3 at 500 ng / ml were not effective in protecting the RGC-5 cells under these same conditions. BDNF and NT-4 at 100 ng / ml were not effective in protecting the serum-deprived RGC-5 cells (data not shown).

4. Discussion Glaucoma is an optic neuropathy most commonly associated with elevated intraocular pressure. The disease is multi-factorial affecting several sites including the trabecular meshwork, lamina cribrosa and optic nerve head. The final common pathway in glaucoma is apoptosis of retinal ganglion cells in animal models of the disease [19,37] as well as in human primary open angle glaucoma [25]. A number of studies have employed primary RGC

cultures to better understand the molecular basis of cell death. The basic mechanisms of RGC apoptotic cell death are currently not understood. A number of mutations in the GLC-1A / myocilin (TIGR) gene [51] or a cytochrome P-450 gene (CYP 1B1) [49,50], have been shown to result in glaucoma. A permanent RGC line would facilitate studies to elucidate molecular mechanisms of RGC cell death in glaucoma. To better understand the basic mechanisms of RGC death in glaucoma, there is a need to establish a culture model that will allow a detailed molecular analysis of RGC apoptosis. In the current study, a permanently transformed rat RGC line was established from retinal cells from PN1 rats. Confirmation of RGC-5 cells to be of retinal ganglion cell origin was made by using specific cell type markers, such as Thy-1, Brn-3C (specific to retinal ganglion cells), Neuritin, NMDA receptor, GABA-B receptor, and synaptophysin. Additionally other markers including GFAP for ¨ Muller cells, syntaxin or HPC-1 for amacrine cells and 8A1 for horizontal and ganglion cells were also tested. RGC-5 cells expressed Thy-1, a cell surface glycoprotein of the immunoglobulin superfamily, which has been shown to be specifically expressed in retinal ganglion cells in retinas. Thus, Thy-1 expression has been used as a specific marker for retinal ganglion cells in retinas [5,8,33]. Thy-1 expression in rat retinas associated with the RGC begins at embryonic day 19 and increases until PN14. Increased

R.R. Krishnamoorthy et al. / Molecular Brain Research 86 (2001) 1 – 12

Fig. 8. Effects of supplementation of neurotrophins on the protection of RGC-5 cells under conditions of serum deprivation. RGC-5 cells were seeded in 96-well plates. The cells were subjected to serum deprivation (SF) for 2 days along with controls with complete medium (S). A batch of SF-treated RGC-5 cells were fortified with either BDNF, NGF, NT-3, or NT-4 at 500 ng / ml in the growth medium to assess their effect on cell viability by formazan assay. The absorbance was recorded at 490 nm and used as an index of cell viability and cell numbers were generated by use of the standard curve. ANOVA was used to statistically analyze the data and the asterisk above S, BDNF and NT-4 histograms denotes significant difference (P,0.001) as compared to SF. These results show that BDNF and NT-4 are most effective in protecting the RGC-5 cells from undergoing cell death following serum deprivation.

Thy-1 expression is paralleled by a thickening of the inner plexiform layer [44]. Recent studies demonstrated that the RGC phenotype is maintained in dissociated cell cultures and these cells continue to express Thy-1 [21,29,43,53]. The positive expression of Thy-1 by RGC-5 cells was confirmed by RT-PCR, immunocytochemistry and im¨ munoblot analysis. In order to differentiate Muller cells which are GFAP positive, from RGCs, we performed immunoblot analysis for GFAP expression. Negligible levels of GFAP were observed in RGC-5 cells. Furthermore, RGC-5 cells were negative for the expression of syntaxin / HPC-1, which is an amacrine cell marker. 8A1 is a neurofilament protein and although it has been shown to be expressed in retinal ganglion as well as horizontal cells, RGC-5 cells did not express detectable levels of 8A1 as

9

observed by immunoblot analysis (Fig. 4). This could be either due to low levels of expression of 8A1 by RGC-5 or because of their transformed nature. Furthermore, RGC-5 cells expressed Brn-3c, a member of POU-domain transcription factors as determined by RT-PCR analysis. These POU-domain transcription factors have been shown to be expressed specifically by retinal ganglion cells [55]. Glutamate causes excitotoxicity by overstimulation of glutamate receptor (N-methyl-D-aspartate (NMDA) receptor) and generating a cascade of events resulting in an overload of intracellular calcium and release of nitric oxide, which has free radical properties. The nitric oxide combines with reactive oxygen species causing oxidative damage, which may lead to apoptosis. A recent study reported elevated glutamate levels in vitreous of both humans with glaucoma and monkeys with experimental glaucoma [15]. Retinal ganglion cells were shown to be sensitive to low doses of glutamate (5–500 mM) after 3 days in culture [35,36]. Initially, RGC-5 cells were not sensitive to glutamate toxicity, but when treated with a differentiation agent, sConA, RGC-5 cells differentiated and showed sensitivity to glutamate. The insensitivity of RGC-5 cells to glutamate could possibly be due to their undifferentiated and transformed nature. Succinyl ConA has been reported to cause differentiation of transformed neuroretinal cells in vitro [46]. Similar to results observed in chick retina cells [46], treatment of the transformed RGC-5 cells with sConA appeared to restore their responsiveness to glutamate excitotoxicity. The glutamate toxicity in the sConA-treated RGC-5 cells is comparable in many ways to that observed in primary rat RGC cultures [36]. In both systems, this glutamate excitotoxicity was blocked by the NMDA antagonist, dizocilpine. Low sensitivity of RGC-5 cells to glutamate was further corroborated by RT-PCR results exhibiting that RGC-5 cells express lower levels of NMDA and GABA-B receptor mRNAs as compared to total retina (Fig. 2). Neurotrophic factors play critical roles in the development and survival of the mammalian nervous system [30]. The most commonly known and intensely investigated neurotrophic factors are the neurotrophins (NGF, BDNF, NT-3, NT-4 / 5). Neurotrophins contribute to a complex pattern of connectivity in the nervous system by their interactions with both neuronal and non-neuronal cell types [11]. One compelling hypothesis in glaucoma is that blockage of reterograde transport of neurotrophins leads to apoptosis of RGCs. In order to test this hypothesis in vitro, we cultured RGC-5 cells under the conditions of serum deprivation in an attempt to mimic the in vivo situation. Serum deprivation resulted in a time-dependent decrease in survival of RGC-5 cells, and these cells died via apoptosis as shown by TUNEL assays. Similar to these results, serum deprivation has been used to induce apoptosis of differentiated PC-12 cells via a decrease in mitochondrial membrane potential [54].

R.R. Krishnamoorthy et al. / Molecular Brain Research 86 (2001) 1 – 12

10

A number of studies have addressed the ability of various neurotrophins to promote survival and regeneration of injured retinal neurons [3]. Application of BDNF to the chick embryo resulted in a 70% increase in the number of RGCs in both embryonic day 6 and day 9 retinas [17]. Short-term survival of axotomized RGCs was increased after intravitreal injection of NT-4 / 5. Similar studies demonstrated that a single intraocular injection with glialderived neurotrophic factor (GDNF) slowed the initiation of RGC death and increased the density of surviving RGCs at 7, 10, and 14 days postaxotomy [26]. Our RT-PCR results show that RGC-5 cells expressed mRNA for NGF, BDNF, NT-3 and NT-4, and their cognate receptors, suggesting these cells are capable of producing the neurotrophins but apparently their secreted amounts are not sufficient to protect the cells from serum deprivation in vitro. Furthermore, it is possible that either autocrine signalling may not play a significant role in survival of RGCs or that the neurotrophin levels are down-regulated under serum deprivation. How the levels of various neurotrophins and their receptors in RGC-5 cells are affected by serum deprivation is currently being investigated. When the serum deprived RGC-5 cells were treated with BDNF and NT-4, they were protected from undergoing apoptosis. Thus, RGC-5 cells may prove to be a model to study apoptosis of RGCs by trophic factor deprivation and protection by BDNF and NT-4 as a method of intervention of apoptosis in glaucoma. A wide range of concentrations of neurotrophins has been used to study the neuroprotection of RGCs. The levels of various neurotrophins including BDNF, NT-3, and NT-4 used in one study were 100–300 ng / ml [4]. These concentrations of neurotrophins were effective in inducing retinal neuronal outgrowth in retinal explant cultures. In another study [12] NT-4 and BDNF at 50 ng / ml were able to induce a significant neurite outgrowth in retinal explant tissue as compared with the untreated explants. Johnson et al. [23] demonstrated a protective role of BDNF at 100 ng / ml in maintaining survival of mature RGC primary cultures. Further, in optic nerve axotomized rats, intravitreal injections of BDNF at 5 mg / ml, were protective of RGC as compared with the untreated axotomized rats [31]. In our

Table 2 Comparison of RGC-5 marker phenotype with expected phenotypes for other retinal cells Marker

RGC-5 cells

Muller cells

Amacrine cells

Horizontal cells

Thy-1 Brn-3C GFAP HPC-1 8A1

111 111 2 2 2

2 2 111 2 2

1(?) 2 2 111 2

2 2 2 2 111

studies, neuroprotection of RGC-5 cells was observed at 500 ng / ml of both BDNF and NT-4. In summary, based on expression pattern of various specific cell markers (Table 2), we have characterized a transformed rat RGC-5 cell line that displays many of the characteristic phenotypes of retinal ganglion cells. These cells expressed Thy-1, Brn-3C, various neurotrophins and their receptors and showed a dependence on trophic factors for survival. Thus, RGC-5 cells may provide a model to study various aspects of RGC cell biology, including apoptosis by trophic factor deprivation and glutamate toxicity as occurs in glaucoma.

Acknowledgements We wish to thank Anne-Marie Brun and Dr. Lawrence Oakford for their expert technical help during the course of these studies. We would also like to acknowledge the efforts of Higinio Cantres Carmonas and Vilay Thankom. Part of these studies include the Barry M. Goldwater Scholarship and Intel Science Talent Search proposal for PA. These studies were supported in part by a grant from National Glaucoma Research of American Health Assistance Foundation to NA and RJW.

References [1] N. Agarwal, K. Mehta, Possible involvement of Bcl-2 pathway in retinoid X receptor alpha-induced apoptosis of HL-60 cells, Biochem. Biophys. Res. Commun. 230 (1997) 251–253. [2] N. Agarwal, C. Jomary, S.E. Jones, K. O’Rourke, M.H. Chaitin, R.J. Wordinger, B.F. Murphy, Immunocytochemical colocalization of clusterin in apoptotic photoreceptor cells in retinal degeneration slow rds mutant mouse retinas, Biochem. Biophys. Res. Commun. 225 (1996) 84–91. [3] A.J. Aguayo, D.B. Clarke, T.N. Jelsma, P. Kittlerova, H.C. Friedman, G.M. Bray, Effects of neurotrophins on the survival and regrowth of injured retinal neurons, Ciba Found. Symp. 196 (1996) 135–144. [4] J. Atkinson, M.K. Panni, R.D. Lund, Effects of neurotrophins on embryonic retinal outgrowth, Dev. Brain Res. 112 (1999) 173–180. [5] C.J. Barnstable, U.C. Drager, Thy-1 antigen: a ganglion cell specific marker in rodent retina, Neuroscience 11 (1984) 847–855. [6] C.J. Barnstable, R. Hofstein, K. Akagawa, A marker for early amacrine cell development in rat retina, Brain Res. 352 (1996) 286–290. [7] B.A. Barres, B.E. Silverstein, D.P. Corey, Immunological morphological and electrophysiological variation among retinal ganglion cells purified by panning, Neuron 1 (1988) 791–803. [8] R. Beale, N.N. Osborne, Localization of the Thy-1 antigen to the surfaces of rat retinal ganglion cells, Neurochem. Int. 4 (1982) 587–595. [9] M. Berkalaar, D.B. Clarke, Y.C. Yang, A.J. Aquayo, Axotomy results in delayed death and apoptosis of retinal ganglion cells in adult rats, J. Neurosci. 14 (1994) 4368–4374. [10] E. Borenfreund, J. Puerner, Toxicity determined in vitro by morphological alterations and neutral red absorption, Toxicol. Lett. 24 (1985) 119–124.

R.R. Krishnamoorthy et al. / Molecular Brain Research 86 (2001) 1 – 12 [11] M. Bothwell, Functional interactions of neurotrophins and their receptors, Annu. Rev. Neurosci. 18 (1995) 223–253. [12] A. Cohen, G.M. Bray, A.J. Aguayo, Neurotrophin-4 / 5 (NT-4 / 5) increases adult rat retinal ganglion cell survival and neurite outgrowth in vitro, J. Neurobiol. 25 (1994) 953–959. [14] E.G. de Araujo, R. Linden, Survival of retinal ganglion cells in vitro: the effect of conditioned medium from retinal cell aggregates, Brazil J. Med. Biol. Res. 23 (1990) 743–746. [15] E.B. Dreyer, D. Zurakowski, R.A. Shumer, S.M. Podos, S.A. Lipton, Elevated glutamate levels in the vitreous body of humans and monkeys with glaucoma, Arch. Ophthamol. 114 (1996) 299–305. [16] A.J. Eisenfeld, A.J. Buntmilam, P.V. Sarthy, Muller cell expression of glial fibrillary acidic protein after genetic and experimental photoreceptor degeneration in the rat retina, Invest. Ophthalmol. Vis. Sci. 25 (1984) 1321–1328. [17] J.M. Frade, P. Bovolenta, J.R. Martinez-Morales, A. Arribas, J.A. Barbas, A. Rodriguez-Tebar, Control of early cell death by BDNF in the chick retina, Development 124 (1997) 3313–3320. [18] E. Garcia-Valenzuela, W. Gorczyca, Z. Darzynkiewicz, Apoptosis in adult retinal ganglion cells after axotomy, J. Neurobiology 25 (1994) 431–438. [19] E. Garcia-Valenzuela, S. Shareef, J. Walsh, S.C. Sharma, Programmed cell death of retinal ganglion cells during experimental glaucoma, Exp. Eye Res. 61 (1995) 33–44. [20] Y. Gavrieli, Y. Sherman, S.A. Ben-Sasson, Identification of programmed cell death in situ via specific labeling of nuclear DNA fragmentation, J. Cell Biol. 119 (1992) 493–501. [21] E. Guenther, S. Schmid, R. Grantyn, E. Zrenner, In vitro identification of retinal ganglion cells in culture without the need of dye labeling, J. Neurosci. Methods 51 (1994) 177–181. [22] D.N. Hu, R. Ritch, Tissue culture of adult human retinal ganglion cells, J. Glaucoma 6 (1997) 37–43. [23] J.E. Johnson, Y.A. Barde, M. Schwab, H. Thoenen, Brain derived neurotrophic factor support the survival of cultured retinal ganglion cells, J. Neurosci. 6 (1986) 3031–3038. [24] J.F. Kerr, A.H. Wyllie, A.R. Currie, Apoptosis: a basic biological phenomena with wide ranging implications in tissue kinetics, Br. J. Cancer 26 (1972) 239–257. [25] L.A. Kerrigan, D.J. Zack, H.A. Quigley, S.D. Smith, M.E. Pease, TUNEL-positive ganglion cells in human primary open-angle glaucoma, Arch. Ophthalmol. 115 (1997) 1031–1035. [26] P.D. Koeberle, A.K. Ball, Effect of GDNF on retinal ganglion cell survival after axotomy, Vision Res. 38 (1998) 1505–1515. [27] R.R. Krishnamoorthy, M.J. Crawford, M.M. Chaturvedi, S.K. Jain, B.B. Aggarwal, M. Al-Ubaidi, N. Agarwal, Photo-oxidative stress down modulates the activity of nuclear factor-kappa B via involvement of caspase-1, leading to apoptosis of photoreceptor cells, J. Biol. Chem. 274 (1999) 3734–3743. [28] U.K. Laemmli, Cleavage of structural protein during the assembly of the head of bacteriophage T4, Nature (London) 227 (1970) 680– 685. [29] D. Leifer, S.A. Lipton, C.J. Barnstable, R.H. Masland, Monoclonal antibody to Thy-1 enhances regeneration of processes by rat retinal ganglion cells in culture, Science 224 (1984) 303–306. [30] R. Levi-Montalcini, The nerve growth factor 35 years later, Science 237 (1987) 1154–1162. [31] S. Mansour-Robaey, D.B. Clarke, Y.C. Wang, G.M. Bray, A.J. Aguayo, Effects of ocular injury and administration of brain-derived neurotrophic factor on survival and regrowth of axotomized retinal ganglion cells, Proc. Natl. Acad. Sci. USA 91 (1994) 1632–1636. [32] C.A. McCaffery, T.R. Raju, M.R. Bennett, Effect of cultured astroglia on the survival of neonatal rat retinal ganglion cells in vitro, Dev. Biol. 104 (1984) 441–448. [33] R. Morris, F. Grosveld, Expression of Thy-1 in the nervous system of the rat and mouse, in: A.E. Reif, M. Schlesinger (Eds.), Cell

[35]

[36]

[37]

[38]

[39]

[40]

[41]

[43]

[44]

[45] [46]

[47]

[48]

[49]

[50]

[51]

[52] [53]

[54]

11

Surface Antigen Thy-1: Immunology, Neurology, and Therapeutic Applications, Marcel Dekker, New York, 1989, pp. 121–148. Y. Otori, J.Y. Wei, C.J. Barnstable, Neurotoxic effects of low doses of glutamate on purified rat retinal ganglion cells, Invest. Ophthalmol. Vis. Sci. 39 (1998) 972–981. I.H. Pang, E.M. Wexler, S. Nawy, L. DeSantis, M.A. Kapin, Protection by eliprodil against excitotoxicity in cultured rat retinal ganglion cells, Invest. Ophthalmol. Vis. Sci. 40 (1999) 1170–1176. H.A. Quigley, R.W. Nickells, L.A. Kerrigan, M.E. Pease, D.J. Thibault, D.J. Zack, Retinal ganglion cell death in experimental glaucoma and after axotomy occurs by apoptosis, Invest. Ophthalmol. Vis. Sci. 36 (1995) 774–786. S.A. Rabacchi, M. Ensini, L. Bonfanti, Nerve growth factor reduces apoptosis of axotomized retinal ganglion cells in the neonatal rat, Neuroscience 63 (1994) 969–973. S.A. Rabacchi, L. Bonfanti, X.H. Liu, L. Maffei, Apoptotic cell death induced by optic nerve lesion in the neonatal rat, J. Neurosci. 14 (1994) 5292–5301. M. Raff, K. Fields, H. Itiroh-Sen, R. Mirsky, R. Pruss, J. Winter, Cell-type specific markers for distinguishing and studying neurons and major classes of glial cells in culture, Brain Res. 174 (1979) 283–308. R.S. Roque, R.B. Caldwell, Muller cell changes precede vascularization of the pigment epithelium in the dystrophic rat retina, Glia 3 (1990) 464–475. P.V. Sarthy, B.M. Curtis, W.A. Catterall, Retrograde labeling, enrichment, and characterization of retinal ganglion cells from neonatal rat, J. Neurosci. 3 (1983) 2532–2544. S. Schmid, E. Guenther, K. Kohler, Changes in Thy-1 antigen immunoreactivity in the retina during pre- and postnatal development, Neurosci. Lett. 199 (1995) 91–94. G.M. Seigel, Establishment of an E1A-immortalized retinal cell culture, In Vitro Cell. Dev. Biol. Anim. 32 (1996) 66–68. G.M. Seigel, M.F.D. Notter, Lectin-induced differentiation of transformed neuroretinal cells in vitro, Exp. Cell Res. 199 (1992) 240–247. H.J. Sheedlo, J.E. Turner, Influence of a retinal pigment epithelial cell factor(s) on rat retinal progenitor cells, Dev. Brain Res. 93 (1996) 88–99. S. Shiosaka, H. Kiyama, M. Tohyama, A simple method for the separation of retinal sublayers from the entire retina with special reference to application for cell culture, J. Neurosci. Methods 10 (1984) 229–235. I. Stoilov, A.N. Akarsu, M. Sarfarazi, Identification of three different truncating mutations in cytochrome P4501B1 (CYP1B1) as the principal cause of primary congenital glaucoma (Buphthalmos) in families linked to the GLC3A locus on chromosome 2p21, Hum. Mol. Genet. 6 (1997) 641–647. I. Stoilov, A.N. Akarsu, I. Alozie, A. Child, M. Barsoum-Homsy, M.E. Turacli, M. Or, R.A. Lewis, N. Ozdemir, G. Brice, S.G. Aktan, L. Chevrette, M. Coca-Prados, M. Sarfarazi, Sequence analysis and homology modeling suggests that primary congenital glaucoma on 2p21 results from mutations disrupting either the hinge region or the conserved core structures of cytochrome P-4501B1, Am. J. Hum. Genet. 62 (1998) 573–584. E.M. Stone, J.H. Fingert, W.L.M. Alward, Identification of a gene that causes primary open angle glaucoma, Science 275 (1997) 668–670. N. Takahashi, D. Cummins, J. Caprioli, Rat retinal ganglion cells in culture, Exp. Eye Res. 53 (1991) 565–572. H. Taschenberger, R. Grantyn, Several types of Ca12 channels mediate glutamatergic synaptic response to activation of single Thy-1 immunolabeled rat retinal ganglion cells, J. Neurosci. 15 (1995) 2240–2254. J.S. Wadia, R.M.E. Chalmers-Redman, W.J.H. Ju, G.W. Carlile, J.L. Phillips, A.D. Fraser, W.G. Tatton, Mitochondrial membrane po-

12

R.R. Krishnamoorthy et al. / Molecular Brain Research 86 (2001) 1 – 12

tential and nuclear changes in apoptosis caused by serum and nerve growth factor withdrawal: time course and modification by deprenyl, J. Neurosci. 18 (1998) 932–947. [55] M. Xiang, L. Zhou, J.P. Macke, T. Yoshioka, S.H.C. Hendry, R.L.

Eddy, T.B. Shows, J. Nathans, The Brn-3 family of POU-domain factors: primary structure, binding specificity, and expression in subsets of retinal ganglion cells and somatosensory neurons, J. Neurosci. 15 (1995) 4762–4785.