566018
research-article2015
JRA0010.1177/1470320314566018Journal of the Renin-Angiotensin-Aldosterone SystemWhite et al.
Original Article
Retinal ganglion cell neuroprotection by an angiotensin II blocker in an ex vivo retinal explant model
Journal of the Renin-AngiotensinAldosterone System 1–9 © The Author(s) 2015 Reprints and permissions: sagepub.co.uk/journalsPermissions.nav DOI: 10.1177/1470320314566018 jra.sagepub.com
Andrew JR White1–4, Janosch P Heller3, Johahn Leung5, Alessia Tassoni3 and Keith R Martin3,4,6
Abstract Purpose: An ex vivo organotypic retinal explant model was developed to examine retinal survival mechanisms relevant to glaucoma mediated by the renin angiotensin system in the rodent eye. Methods: Eyes from adult Sprague Dawley rats were enucleated immediately post-mortem and used to make four retinal explants per eye. Explants were treated either with irbesartan (10 µM), vehicle or angiotensin II (2 μM) for four days. Retinal ganglion cell density was estimated by βIII tubulin immunohistochemistry. Live imaging of superoxide formation with dihydroethidium (DHE) was performed. Protein expression was determined by Western blotting, and mRNA expression was determined by RT-PCR. Results: Irbesartan (10 µM) almost doubled ganglion cell survival after four days. Angiotensin II (2 µM) reduced cell survival by 40%. Sholl analysis suggested that irbesartan improved ganglion cell dendritic arborisation compared to control and angiotensin II reduced it. Angiotensin-treated explants showed an intense DHE fluorescence not seen in irbesartan-treated explants. Analysis of protein and mRNA expression determined that the angiotensin II receptor At1R was implicated in modulation of the NADPH-dependent pathway of superoxide generation. Conclusion: Angiotensin II blockers protect retinal ganglion cells in this model and may be worth further investigation as a neuroprotective treatment in models of eye disease. Keywords Neuroprotection, retinal explant, glaucoma, angiotensin Date received: 14 August 2014; accepted 16 November 2014
Introduction An ex vivo organotypic retinal explant model has been developed by our laboratory to examine retinal survival mechanisms and as a rapid screening tool for potential novel neuroprotective agents in a retinal neurodegenerative model.1–3 Retinal explants can be kept viable for a number of days with apoptotic retinal ganglion cell (RGC) death occurring at a consistent rate over this time. RGC counts are used to assay the neuroprotective effect of candidate drugs.2 Previous work has shown that the rate of RGC apoptosis can be pharmacologically altered by pan caspase inhibitors2 and some United States Food and Drug Administration (FDA)-approved drugs.2,3 Some of these strategies may have utility in the treatment of ocular neurodegenerative diseases such as glaucoma, which is our area of interest. In addition, the preparation lends itself to investigation of the mechanisms involved in any neuroprotective activity demonstrated by candidate agents. The fact
that RGCs are preserved in situ with their normal adjacent retina architecture creates a study environment closer to in vivo conditions than cell culture systems, without the need for chronic treatment of animals as in other in vivo models 1Centre
for Vision Research, Westmead Millennium Institute, University of Sydney, Australia 2Save Sight Institute, University of Sydney, Australia 3John van Geest Centre for Brain Repair, University of Cambridge, UK 4Cambridge NIHR Biomedical Research Centre, UK 5School of Medical Sciences, Discipline of Physiology, University of Sydney, Australia 6Wellcome Trust – MRC Cambridge Stem Cell Institute, University of Cambridge, UK Corresponding author: Andrew JR White, Dept. of Ophthalmology, Westmead Hospital, B4a, Westmead NSW, 2145, Australia. Email: [email protected]
Creative Commons CC-BY-NC: This article is distributed under the terms of the Creative Commons Attribution-NonCommercial 3.0 License (http://www.creativecommons.org/licenses/by-nc/3.0/) which permits non-commercial use, reproduction and distribution of the work without further permission provided the original work is attributed as specified on the SAGE and Open Access page (http://www.uk.sagepub.com/aboutus/openaccess.htm). Downloaded from jra.sagepub.com at Lucia Campus Library on June 16, 2015
2
Journal of the Renin-Angiotensin-Aldosterone System
of glaucoma such as induced ocular hypertension or optic nerve crush. Our laboratory has previously used retinal explants to screen several licenced medications for neuroprotective potential.2 Based on initial screening findings (data not shown), we focussed on irbesartan, a commonly used angiotensin II blocker in clinical usage, as the most promising candidate agent and examined its mechanism of action further in a wholemount preparation, which had not been conducted in our previous studies. There is already some evidence that angiotensin blockers may have a neuroprotective role in in vivo glaucoma models though the mechanism had not been investigated directly.4
Methods Drug screening, cell counts and cellular morphology Home Office United Kingdom (UK) and local institutional ethical approval was granted in compliance with the UK Animals (Scientific Procedures) Act (1986) and all experiments complied with the ARVO Statement for the Use of Animals in Ophthalmic and Visual Research. Eyes from adult male Sprague Dawley rats (n = 27) were enucleated immediately post-mortem. Intact retina was dissected from the optic cup to make four retinal explants per eye. All experiments were performed in triplicate. Half of each retina was treated with the candidate drug and half with vehicle alone per eye thus half the retina tissue from the same animal was an internal control for the treated section. Allocation of the tissue to treatment or control was randomised to minimise variability. Explants were placed ganglion cell side up on millipore filters (Millipore; Millicell; Cork, Ireland) with 300 μl N2/B27 culture media per well in a humidified incubator as described previously.3 To investigate the effect of angiotensin II blockade, irbesartan (10 µM) (Sigma, UK) dissolved in dimethyl sulfoxide (DMSO) was applied to wholemount retina explants. The effect of this treatment was compared with vehicle alone and angiotensin II (2 μM) (Sigma, UK) using protocols in keeping with our previous studies.2 The dosages were derived from manufacturer advice and simple titration experiments (data not shown). At the conclusion of each experiment, explants were fixed in wholemount in 4% paraformaldehyde (PFA). The explants were cut from the plastic casing of the millipore filters and placed on glass slides, ganglion cell side facing up. RGC density and morphology were estimated by βIII tubulin (mouse anti-βIII-tubulin, clone 5G8 Promega, Southampton, UK 1:1000) immunohistochemistry using the staining protocol described previously.1–3 Fluorescent secondary antibodies (Invitrogen, UK) were used at a concentration of 1:1000 as previously described.1–3 To ensure correct section orientation, explants were counterstained
with 4',6-diamidino-2-phenylindole (DAPI) at a concentration of 1:10,000 as previously described (Invitrogen, UK). This allowed easy identification of the photoreceptor layer of the explants and examination of the general condition of the tissue preparation. It also helped to ensure correct orientation of the sections and focal plane for cell counts. Imaging was performed at 20× magnification using an epifluorescence microscope (DM6000; Leica Microsystems UK). Ganglion cells were identified based on morphology and plane of focus of the wholemount. Images were taken by a masked observer at three locations for each explant. Cell counts were performed using image analysis software (ImageJ, United States National Institutes of Health (NIH)). This software was used to trace dendritic morphology of individual ganglion cells as an overlay. The original image was then subtracted and Sholl analysis was performed on the resultant image. Between 10 and 20 dendritic tracings could be made for each image meaning a minimum of 240 dendritic tracings were analysed for each condition tested. Data points were generated at 2 μm intervals to a radius of 70 µm from the cell centre. This is typically the limit of identified dendritic arbors seen in section. Live imaging of superoxide formation. Retinal explants were prepared as above in triplicate and exposed to either: vehicle, angiotensin II (2 μM) or irbesartan (10 µM) in culture media as above. The explants were placed ganglion cell side down in glass-bottom culture wells (P35G-0-14C Matek Corp, Ashland, MA, USA) and incubated for 30 minutes at 37°C in a humidified incubator with 10 µM dihydroethidium (DHE) added to the medium (Sigma, UK) prior to rinsing with phosphatebuffered saline (PBS) and reincubation with the media plus drug or vehicle. At 30-minute intervals, explants were removed from the incubator, treated with a premixed Hoechst stain (NucBlu: Invitrogen, UK) as per the manufacturer’s instructions and imaged with a humidified inverted immunofluorescence microscope at 100× magnification (DMI 6000B Leica Microsystems, UK). Experiments were performed in triplicate and each explant was imaged only once. Successful imaging could be carried out for six hours using this technique. Images were taken with standardised intensity settings on three filter channels to ensure no bleed through of immunofluorescence between the channels. Intensity of DHE staining at each timepoint was quantified by image analysis software (ImageJ, NIH) by four masked observers who identified the points of greatest intensity in each image independently so as to serve as internal controls. This typically correlated with the Hoechst stain of cell nuclei. If no focal DHE intensity was detected in an image, observers were asked to plot five random points on the image. Pixel values of those points were quantified and compared to background intensity using a
Downloaded from jra.sagepub.com at Lucia Campus Library on June 16, 2015
3
White et al. Tabel 1. Primary antibodies used. Antigen
Host
Supplier
Reference
Blocking
Dilution
β-actin NOX2/gp91phox NOX4
Mouse Rabbit Rabbit
Abcam Abcam Abcam
ab6276 ab31092 ab16534
5% milk 5% milk 5% milk
1:15,000 1:1000 1:1000
NOX: nicotinamide adenine dinucleotide phosphate (NAPDH) oxidase 2.
custom software routine developed using MATLAB (Mathworks, Cambridge UK). Western blotting. Explants were prepared as described above. The retinal explants were kept for four hours in culture. The explants were in close contact with the medium supplemented with DMSO alone (control) or angiotensin II (2 μM) or irbesartan (10 µM) as described above. At the end of each timepoint, the retinae were carefully lifted out of the wells and transferred into tubes. The tissue was washed three times with PBS to remove any remaining medium. After centrifugation, the supernatant was discarded and the tissue stored at −80°C until RNA and protein isolation. Half of the retinal tissue was used for total protein isolation. The tissue was thawed on ice and for every ~5 mg tissue ~300 µl radioimmunoprecipitation assay (RIPA) buffer (Roche, supplemented with protein inhibitor) were added. The tissue was broken apart using an electric homogeniser (IKA-Werke, Staufen, Germany). Subsequently, the solution was incubated for up to two hours at 4°C with constant rotation. After lysis, the solution was centrifuged for 15 minutes at 4°C at 13,000 g to remove cell debris. The cell pellet was discarded and the supernatant, containing the proteins, was transferred into a fresh tube. The total protein concentration was determined using the bicinchoninic (BCA) protein assay kit (Thermo Scientific UK) following the manufacturer’s instructions. The proteins were denatured under reducing conditions in a buffer of sodium dodecyl sulphate (SDS), β-mercaptoethanol, glycerol and tris-glycine for five minutes at 95°C. A total of 20 µg of total proteins for each condition were then separated in a 4%–12% SDS Bis-tris gel (Invitrogen, UK) for one hour at 120 V. After the separation, the proteins were transferred onto a polyvinylidene difluoride (PVDF) membrane (Invitrogen, UK) for two hours at 50 V. The PVDF membrane was then blocked using tris-buffered saline supplemented with 0.5% tween-20 (TBS-T) with 5% skimmed milk solution for up to one hour at room temperature. This was followed by the incubation with the primary antibody solution (primary antibody diluted in blocking solution, see Table 1) overnight at 4°C. The next day, the blots were washed four times with TBS-T at room temperature (RT) and they were then incubated in secondary antibody solution (horseradish peroxidase (HRP)-conjugated secondary antibody (Vector) diluted 1:50,000 in TBS-T) for one hour at RT. Membranes were washed with TBS-T four
times before proteins were visualised using a chemiluminescent substrate (ECL, Amersham). To assess the size of the protein bands, Top of Form Precision Plus Protein™ Dual Color Standard (Biorad) was used. We visualised the protein bands in an Alliance chemiluminescence system (Uvitec, UK).
Reverse transcription-polymerase chain reaction (RT-PCR) From the experiments to determine protein expression, the other half of the retinal tissue was used for total RNA isolation according to the manufacturer’s instructions (RNA/ DNA/Protein Purification Kit, Norgen Biotek, Niagaraon-the Lake, Canada). In addition to the total RNA, proteins were also isolated and used for Western blot (see above). Following RNA isolation, residual genomic DNA was digested with DNase to avoid false-positive bands after PCR. Then, the messenger RNA (mRNA) was transcripted into cDNA using the Superscript R III first-strand kit (Invitrogen, UK) following the manufacturer’s instructions. The cDNA was then used for PCR. MangoTaq DNA polymerase (Bioline, UK) was used for all PCR reactions. Forward and reverse primers were designed to amplify the DNA fragment of interest. Glyceraldehyde 3-phosphate dehydrogenase (GAPDH) was used as a PCR positive control and housekeeping standard. Primers were 20 bp in length with a guanine/cytosine ratio that did not exceed 55%. Primers were designed to span exon/exon junctions in the mRNA to discriminate RNA bands from contaminating residual genomic DNA bands. Each PCR reaction contained 100 ng cDNA, 0.5 mM of each primer, 500 mM dNTPs, 1.5 mM MgCl2, 5× Taq DNA polymerase buffer, 5 U MangoTaq and ddH2O up to a final volume of 25 µl. First, the DNA template was denatured for five minutes at 94°C before amplification was started. The reaction was stopped after 35 cycles. Each cycle consisted of 30 seconds at 94°C, 30 seconds at 58°C and 30 seconds at 72°C. The final cycle was followed by a final extension of five minutes at 72°C. After the PCR, the products were separated on a 1% agarose gel (agarose in Trisborate-ethylenediaminetetraacetic acid (TBE) buffer (Invitrogen, UK) supplemented with 0.5 mg/ml ethidium bromide). Electrophoresis took place for 30 minutes at 90 V in TBE buffer. DNA bands were visualised using a longwave UVP transilluminator (Uvitec). Hyperladder II
Downloaded from jra.sagepub.com at Lucia Campus Library on June 16, 2015
4
Journal of the Renin-Angiotensin-Aldosterone System
Figure 1. Representative wholemount sections at 20× magnification treated either with: (a) control (dimethyl sulfoxide (DMSO)), (b) irbesartan (10 µM) or (c) angiotensin II (2 μM). Tissue is Day 4 post-treatment. Scale bar is 100 μM.
Figure 2. Ganglion cell counts from wholemount explant preparations. Average cell count for explants treated either with irbesartan (10 μM) or angiotensin II (2 μM) were compared to average cell counts of control explants treated with vehicle only (dimethyl sulfoxide (DMSO)). Error bars are standard error of the mean. Asterisk denotes p < 0.05.
(Bioline UK) served as a base pair standard. The brightness of the DNA bands of interest was compared to the housekeeping standard (GAPDH) to assess the concentration of the mRNAs of interest.
Results The retinal explant preparations were fixed in PFA at Day 4, which has previously been determined as the optimal time to differentiate neuroprotective effects.2 Average cell density for control-treated explants was 878 cells/mm2. Peak density was 1666 cells/mm2. Representative wholemount sections at 20× magnification treated either with irbesartan (10µM), control (DMSO) or angiotensin II (2 μM) are shown in Figure 1. Note that tissue treated with
angiotensin II was often significantly degraded compared to the other preparations consistent with the cell counts shown in Figure 2. Figure 2 shows average cell counts from three wholemount preparation experiments of three explants each treated either with irbesartan (10 µM), angiotensin II (2 µM) or vehicle (DMSO) normalised to control explant counts. There is a marked increase in ganglion cell counts in explants treated with irbesartan (10 µM) compared to control counts treated with vehicle (DMSO) after four days of culture. Conversely, there is a marked reduction in cell counts in explants treated with angiotensin II (2 µM) compared to control explants (p < 0.05, t test). The wholemount preparation allowed for limited analysis of cellular morphology of ganglion cells. Figure 3
Downloaded from jra.sagepub.com at Lucia Campus Library on June 16, 2015
5
White et al.
Figure 3. Sholl analysis of dendritic morphology compared to average profile of control sections stained in wholemount for BIII tubulin imaged at 20× magnification. Analysis was carried out on individual cell soma and dendritic morphology traced from the original photomicrographs (analysis of variance (ANOVA), p < 0.05). Error bars denote standard error of the mean.
shows a Sholl analysis of dendritic morphology compared to average profile of control sections. Analysis was carried out on individual cell soma and dendritic morphology traced from the original photomicrographs. A minimum of 200 cells were analysed from triplicate experiments for each condition. Only irbesartan (10 µM)-treated wholemounts showed preserved dendritic morphology at Day 4 (analysis of variance (ANOVA), p < 0.05).
Live imaging or reactive oxygen species (ROS) generation Figure 4 shows cells in the wholemount explant preparation stained with Hoechst stain and counterstained for intracellular reactive oxygen species (ROS) with DHE. Hoechst stain in live culture stains only active cell nuclei. Figure 4(a) shows the amount of ROS generated at four hours in a control explant. Figure 4(b) shows an explant at four hours treated with angiotensin II (2 μm) demonstrating a comparative increase in DHE staining compared to control. Figure 4(c) shows an explant at four hours treated with irbesartan (10 µM) with a comparative decrease in DHE staining intensity compared to control. Only active cell nuclei stain with Hoechst stain. Figure 4(b) shows a cell with intense DHE staining but no Hoechst stain (marked with arrow). The intensity of DHE staining, used as a surrogate measure of intracellular ROS generation, was quantified over time. The results are shown in Figure 5, which demonstrates an angiotensin II-related bimodal increase in ROS generation compared to control, which is almost completely attenuated by irbesartan (10 µM).
Modulation of the angiotensin II receptors AIIR1a and AIIR2 in treated retinal explants Modulation of angiotensin II receptor mRNA expression by irbesartan (10 µM) and angiotensin II (2 μM) was determined by RT-PCR at four hours post-dissection of the retinal explant. The results are shown in Figure 6. Treatment of retinal explants by angiotensin II completely silenced mRNA expression of AIIRIa and AIIR2. Irbesartan treatment increased mRNA expression of AIIRIa (p < 0.01, t test) and slightly decreased expression of AIIR2 mRNA, though this result was not statistically significant.
Modulation of subunits of the nicotinamide adenine dinucleotide phosphate (NADPH)dependent pathway in treated explants Modulation of expression of NAPDH oxidase 2 (NOX2), a principal source of intracellular superoxide generation at the ganglion cell layer in the retina,13 could be demonstrated in explants treated with either vehicle, angiotensin II (2 μM), or irbesartan (10 µM). The results are shown in Figure 7. Angiotensin II (2 μM) substantially increased expression of NOX2 compared to control (p < 0.01, t test). Irbesartan (10 µM) also slightly increased expression of NOX2 though this result was not significant. Intensity values are plotted on the graph at the right (Figure 7(b)). mRNA expression of NOX2 was determined by RT-PCR in explants treated by vehicle, angiotensin II or irbesartan (10 µM) at four hours. The results are shown in Figure 7(c) and (d). Angiotensin II (2 μM) completely
Downloaded from jra.sagepub.com at Lucia Campus Library on June 16, 2015
6
Journal of the Renin-Angiotensin-Aldosterone System
Figure 4. Live imaging of retinal explants at 100× magnification taken at four hours post-dissection.
Explants were treated either with vehicle (dimethyl sulfoxide (DMSO)), angiotensin II or irbesartan (10µM) at the concentrations shown. The top panels show active cell nuclei stained with Hoechst stain. Corresponding photomicrographs in the bottom panel show intracellular reactive oxygen species (ROS) stained with dihydroethidium (DHE). Note the corresponding cell (arrow) staining brightly with DHE but with a dim Hoechst stain. Scale bar shows 20 μm.
Figure 5. Dihydroethidium (DHE) staining intensity values normalised to the intensity of control explants measured at hourly intervals. Error bars are standard error of the mean.
silenced mRNA expression of NOX2 in retinal explants (p < 0.01, t test). Irbesartan (10 µM) slightly increased NOX 2 mRNA expression in keeping with protein expression (p < 0.01, t test, Figure 7(a) and (b)).
Discussion Organotypic ex vivo retinal explants examined in wholemount are a useful way to explore both the magnitude and mechanisms of neuroprotective strategies. Our study
Downloaded from jra.sagepub.com at Lucia Campus Library on June 16, 2015
7
White et al.
Figure 6. (a) Angiotensin II receptors At1Ra and AT2R messenger RNA (mRNA) expression was determined by reverse transcription-polymerase chain reaction (RT-PCR) at four hours in explants treated either with irbesartan (10 μM) or angiotensin II (2 μM) or control (dimethyl sulfoxide (DMSO)). mRNA expression of glyceraldehyde 3-phosphate dehydrogenase (GAPDH) is given as a reference. (b) Intensity value of angiotensin II receptors At1Ra and AT2R mRNA expression compared against control in explants treated with irbesartan (10 μM). Treatment of retinal explants by angiotensin II completely silenced mRNA expression of AIIRIa and AIIR2. Error bars are standard error of the mean. Asterisk denotes p < 0.05.
showed that modulation of the renin-angiotensin system in the eye, specifically modulation of angiotensin II receptors, appears to confer neuroprotection in wholemount cultures. This is in keeping with previous in vivo findings utilising candesartan4,6 and in vitro findings utilising telmisartan,7 other common angiotensin II blockers in clinical usage. The mechanism of these neuroprotective effects appears to involve modulation of production of ROS.6,7 We cannot provide a direct link between apoptosis and a rise in superoxide generation in our model but the circumstantial evidence derived from it is compelling. It has been shown previously that elevated intraocular pressure leads to the generation of ROS as a prelude to apoptosis8 in in vitro and in vivo models.9 To date, the neuroprotective effects of angiotensin II blockers have not been directly linked to modulation of free radical formation in an in vivo chronic ocular hypertensive model of
glaucoma, though the neuroprotective effect of candesartan has been validated without demonstration of mechanism.4 Candesartan modulated amelioration of intracellular free radical generation has, however, been demonstrated in an ocular ischemic/reperfusion model.6 Angiotensin II-dependent modulation of intracellular free radical generation associated with a neuroprotective effect has not been described in an axotomised, nutrient-deprivation model such as a retinal explant previously. Angiotensin II receptor distribution is widespread in the rodent retina10,11 and is known to be localised to RGCs in addition to other ocular structures.11 Receptor distribution on interneurons such as amacrine cells has suggested angiotensin II may have a role other than regulating ocular perfusion pressure, for example, in cellular signalling.10 Components of the renin-angiotensin system in the human retina have also been demonstrated.5,12,13 The results of our
Downloaded from jra.sagepub.com at Lucia Campus Library on June 16, 2015
8
Journal of the Renin-Angiotensin-Aldosterone System
Figure 7. (a) Western blot of nicotinamide adenine dinucleotide phosphate oxidase 2 (NOX2) expression in retinal explants at four hours post-dissection. Protein band was detected at 65 kDa. Beta actin with a band at 42 kDa is included as a reference. (b) Western blot intensity of NOX2 compared to control (dimethyl sulfoxide (DMSO)) in explants treated with irbesartan (10μM) or angiotensin II (2 μM). Error bars are standard deviation. Asterisk denoted p < 0.05. (c) NOX2 messenger RNA (mRNA) expression in explants assayed by reverse transcription-polymerase chain reaction (RT-PCR) at four hours. Glyceraldehyde 3-phosphate dehydrogenase (GADPH) is included as a reference. (d) Intensity of mRNA expression in explants treated with irbesartan (10 μM) or angiotensin II (2 μM) compared to control explants (DMSO). Error bars are standard error of the mean. Asterisk denoted p < 0.05.
study demonstrate a direct neuroprotective effect on RGCs in a retinal explant model by irbesartan (10 µM), an angiotensin II blocker. Moreover, modulation of this system could also lead to increased cell death by application of angiotensin II, a direct agonist. Not only were gross wholemount cell counts reduced (Figure 2) by application of angiotensin II but dendritic arborisation, a measure of cellular health, was also reduced by application of angiotensin II (Figure 3). The reverse of this observation was seen in explants treated with irbesartan (10 µM). We did not undertake studies of ganglion cell function in this experimental paradigm as it was outside the scope of our study. Angiotensin II modulates intracellular generation of free radical formation (Figures 4 and 5). This appears to occur at the angiotensin II receptor level, expression of which is modulated by drug therapy within a matter of hours (Figure 6). Generation of ROS is markedly increased by application of angiotensin II and almost completely silenced by application of irbesartan. NOX2, an NADPH oxidase found predominantly at the ganglion cell layer of the rodent retina,14 seems to play a role in this and is heavily influenced by the activity of angiotensin II (Figure 7(c) and (d)).
The mechanism behind this effect is clearly complex and multifactorial. However, our findings show modulation of AIIr1a is involved with AIIR2 having an antagonistic effect (Figure 6). At1R modulation of generation of superoxide species intracellularly is associated with cell death, demonstrated in our experiments by wholemount cell counts (Figure 2). The modulated ‘burst’ pattern (Figure 5) is similar to what has been previously described15 and implies some of the effect of superoxide generation relates to an intracellular signalling mechanism yet to be fully elucidated. The fact that NOX2 expression is maintained and even slightly increased in irbesartan-treated cells as detected by Western blot (Figure 7(a) and (b)) and RT-PCR (Figure 7(c) and (d)) suggests at least some superoxide generation is necessary for homeostasis in the eye. Angiotensin II seems to overstimulate this system, leading to a high level of NOX2 protein expression (Figure 7(a) and (b)) and ROS generation coinciding with a decrease in Hoechst staining (Fig 4, centre panel), suggestive of reduced nuclear activity, perhaps as a pre-apoptotic event. Angiotensin II stimulation of the explants also completely silences mRNA expression of NOX2 at four hours post-treatment (Figure 7(c)
Downloaded from jra.sagepub.com at Lucia Campus Library on June 16, 2015
9
White et al. and (d)), suggestive of a negative feedback mechanism in action at the same time. The exact mechanism is yet to be elucidated. In addition, angiotensin II is known to influence neuronal NMDA currents,16 leading to increased NOX 2-dependent generation of ROS.17 Modulation of this system is also implicated in observed neuroprotective effects in the ocular ischemia reperfusion model.6 Thus, our study adds to the circumstantial evidence that the renin-angiotensin system in the eye may play a role in modulating known triggers of apoptosis such as N-methylD-aspartate (NMDA) excitotoxicity and generation of intracellular ROS and so may provide a therapeutic basis for future neuroprotective strategies in ocular neurodegenerative diseases such as glaucoma. Acknowledgements For advice with the live imaging experiments, we thank Dr Richard Eva from the Centre for Brain Repair, Cambridge, and Dr Gillian Groeger from University College, Cork. We thank Li Wen for help in preparing the manuscript figures.
Conflict of interest None declared.
Funding This work was supported by the Cambridge Eye Trust, Fight for Sight UK and the Jukes Glaucoma Research Fund.
References 1. Johnson TV and Martin KR. Development and characterization of an adult retinal explant organotypic tissue culture system as an in vitro intraocular stem cell transplantation model. Invest Ophthalmol Vis Sci 2008; 49: 3503–3512. 2. Bull ND, Johnson TV, Welsapar G, et al. Use of an adult rat retinal explant model for screening of potential retinal ganglion cell neuroprotective therapies. Invest Ophthalmol Vis Sci 2011; 52: 3309–3320. 3. Bull N, Johnson T and Martin K. Organotypic explant culture of adult rat retina for in vitro investigations of neurodegeneration, neuroprotection and cell transplantation. Protoc exch 2011. DOI:10.1038/protex.2011.215. 4. Yang H, Hirooka K, Fukuda K, et al. Neuroprotective effects of angiotensin II type 1 receptor blocker in a rat model of chronic glaucoma. Invest Ophthalmol Vis Sci 2009; 50: 5800–5804. 5. White AJ, Cheruvu SC, Sarris M, et al. Expression of classical components of the renin angiotensin system in the
human eye. J Renin Angiotensin Aldosterone Syst. Epub ahead of print 6 October 2014. 6. Fujita T, Hirooka K, Nakamura T, et al. Neuroprotective effects of angiotensin II type 1 receptor (AT1-R) blocker via modulating AT1-R signaling and decreased extracellular glutamate levels. Invest Ophthalmol Vis Sci 2012; 53: 4099–4110. 7. Ozawa Y, Yuki K, Yamagishi R, et al. Renin-angiotensin system involvement in the oxidative stress-induced neurodegeneration of cultured retinal ganglion cells. Jpn J Ophthalmol 2013; 57: 126–132. 8. Ju WK, Liu Q, Kim KY, et al. Elevated hydrostatic pressure triggers mitochondrial fission and decreases cellular ATP in differentiated RGC-5 cells. Invest Ophthalmol Vis Sci 2007; 48: 2145–2151. 9. Moreno MC, Campanelli J, Sande P, et al. Retinal oxidative stress induced by high intraocular pressure. Free Radic Biol Med 2004; 37: 803–812. 10. Downie LE, Vessey K, Miller A, et al. Neuronal and glial cell expression of angiotensin II type 1 (AT1) and type 2 (AT2) receptors in the rat retina. Neuroscience 2009; 161: 195–213. 11. Wheeler-Schilling TH, Kohler K, Sautter M, et al. Angiotensin II receptor subtype gene expression and cellular localization in the retina and non-neuronal ocular tissues of the rat. Eur J Neurosci 1999; 11: 3387–3394. 12. Savaskan E, Löffler K, Meier F, et al. Immunohistochemical localisation of angiotensin-converting enzyme, angiotensin II and AT1 receptor in human ocular tissue. Ophthal Res 2004; 36: 312–320. 13. Wagner J, Jan Danser A, Derkx F, et al. Demonstration of renin mRNA, angiotensin mRNA, and angiotensin converting enzyme mRNA expression in the human eye: Evidence for an intraocular renin-angiotensin system. Br J Ophthalmol 1996; 80: 159–163. 14. Bhatt L, Groeger G, McDermott K, et al. Rod and cone photoreceptor cells produce ROS in response to stress in a live retinal explant system. Mol Vis 2010; 16: 283–293. 15. Lieven CJ, Thurber KA, Levin EJ, et al. Ordering of neuronal apoptosis signalling: A superoxide burst precedes mitochondrial cytochrome c release in a growth factor deprivation model. Apoptosis 2012; 17: 591–599. 16. Moreno MC, Sande P, Marcos HA, et al. Effect of glaucoma on the retinal glutamate/glutamine cycle activity. FASEB J 2005; 19: 1161–1162. 17. Wang G, Coleman CG, Chan J, et al. Angiotensin II slow-pressor hypertension enhances NMDA currents and NOX2-dependent superoxide production in hypothalamic paraventricular neurons. Am J Physiol Regul Integr Comp Physiol 2013; 304: R1096–R1106.
Downloaded from jra.sagepub.com at Lucia Campus Library on June 16, 2015
research-article2015
JRA0010.1177/1470320314566018Journal of the Renin-Angiotensin-Aldosterone SystemWhite et al.
Original Article
Retinal ganglion cell neuroprotection by an angiotensin II blocker in an ex vivo retinal explant model
Journal of the Renin-AngiotensinAldosterone System 1–9 © The Author(s) 2015 Reprints and permissions: sagepub.co.uk/journalsPermissions.nav DOI: 10.1177/1470320314566018 jra.sagepub.com
Andrew JR White1–4, Janosch P Heller3, Johahn Leung5, Alessia Tassoni3 and Keith R Martin3,4,6
Abstract Purpose: An ex vivo organotypic retinal explant model was developed to examine retinal survival mechanisms relevant to glaucoma mediated by the renin angiotensin system in the rodent eye. Methods: Eyes from adult Sprague Dawley rats were enucleated immediately post-mortem and used to make four retinal explants per eye. Explants were treated either with irbesartan (10 µM), vehicle or angiotensin II (2 μM) for four days. Retinal ganglion cell density was estimated by βIII tubulin immunohistochemistry. Live imaging of superoxide formation with dihydroethidium (DHE) was performed. Protein expression was determined by Western blotting, and mRNA expression was determined by RT-PCR. Results: Irbesartan (10 µM) almost doubled ganglion cell survival after four days. Angiotensin II (2 µM) reduced cell survival by 40%. Sholl analysis suggested that irbesartan improved ganglion cell dendritic arborisation compared to control and angiotensin II reduced it. Angiotensin-treated explants showed an intense DHE fluorescence not seen in irbesartan-treated explants. Analysis of protein and mRNA expression determined that the angiotensin II receptor At1R was implicated in modulation of the NADPH-dependent pathway of superoxide generation. Conclusion: Angiotensin II blockers protect retinal ganglion cells in this model and may be worth further investigation as a neuroprotective treatment in models of eye disease. Keywords Neuroprotection, retinal explant, glaucoma, angiotensin Date received: 14 August 2014; accepted 16 November 2014
Introduction An ex vivo organotypic retinal explant model has been developed by our laboratory to examine retinal survival mechanisms and as a rapid screening tool for potential novel neuroprotective agents in a retinal neurodegenerative model.1–3 Retinal explants can be kept viable for a number of days with apoptotic retinal ganglion cell (RGC) death occurring at a consistent rate over this time. RGC counts are used to assay the neuroprotective effect of candidate drugs.2 Previous work has shown that the rate of RGC apoptosis can be pharmacologically altered by pan caspase inhibitors2 and some United States Food and Drug Administration (FDA)-approved drugs.2,3 Some of these strategies may have utility in the treatment of ocular neurodegenerative diseases such as glaucoma, which is our area of interest. In addition, the preparation lends itself to investigation of the mechanisms involved in any neuroprotective activity demonstrated by candidate agents. The fact
that RGCs are preserved in situ with their normal adjacent retina architecture creates a study environment closer to in vivo conditions than cell culture systems, without the need for chronic treatment of animals as in other in vivo models 1Centre
for Vision Research, Westmead Millennium Institute, University of Sydney, Australia 2Save Sight Institute, University of Sydney, Australia 3John van Geest Centre for Brain Repair, University of Cambridge, UK 4Cambridge NIHR Biomedical Research Centre, UK 5School of Medical Sciences, Discipline of Physiology, University of Sydney, Australia 6Wellcome Trust – MRC Cambridge Stem Cell Institute, University of Cambridge, UK Corresponding author: Andrew JR White, Dept. of Ophthalmology, Westmead Hospital, B4a, Westmead NSW, 2145, Australia. Email: [email protected]
Creative Commons CC-BY-NC: This article is distributed under the terms of the Creative Commons Attribution-NonCommercial 3.0 License (http://www.creativecommons.org/licenses/by-nc/3.0/) which permits non-commercial use, reproduction and distribution of the work without further permission provided the original work is attributed as specified on the SAGE and Open Access page (http://www.uk.sagepub.com/aboutus/openaccess.htm). Downloaded from jra.sagepub.com at Lucia Campus Library on June 16, 2015
2
Journal of the Renin-Angiotensin-Aldosterone System
of glaucoma such as induced ocular hypertension or optic nerve crush. Our laboratory has previously used retinal explants to screen several licenced medications for neuroprotective potential.2 Based on initial screening findings (data not shown), we focussed on irbesartan, a commonly used angiotensin II blocker in clinical usage, as the most promising candidate agent and examined its mechanism of action further in a wholemount preparation, which had not been conducted in our previous studies. There is already some evidence that angiotensin blockers may have a neuroprotective role in in vivo glaucoma models though the mechanism had not been investigated directly.4
Methods Drug screening, cell counts and cellular morphology Home Office United Kingdom (UK) and local institutional ethical approval was granted in compliance with the UK Animals (Scientific Procedures) Act (1986) and all experiments complied with the ARVO Statement for the Use of Animals in Ophthalmic and Visual Research. Eyes from adult male Sprague Dawley rats (n = 27) were enucleated immediately post-mortem. Intact retina was dissected from the optic cup to make four retinal explants per eye. All experiments were performed in triplicate. Half of each retina was treated with the candidate drug and half with vehicle alone per eye thus half the retina tissue from the same animal was an internal control for the treated section. Allocation of the tissue to treatment or control was randomised to minimise variability. Explants were placed ganglion cell side up on millipore filters (Millipore; Millicell; Cork, Ireland) with 300 μl N2/B27 culture media per well in a humidified incubator as described previously.3 To investigate the effect of angiotensin II blockade, irbesartan (10 µM) (Sigma, UK) dissolved in dimethyl sulfoxide (DMSO) was applied to wholemount retina explants. The effect of this treatment was compared with vehicle alone and angiotensin II (2 μM) (Sigma, UK) using protocols in keeping with our previous studies.2 The dosages were derived from manufacturer advice and simple titration experiments (data not shown). At the conclusion of each experiment, explants were fixed in wholemount in 4% paraformaldehyde (PFA). The explants were cut from the plastic casing of the millipore filters and placed on glass slides, ganglion cell side facing up. RGC density and morphology were estimated by βIII tubulin (mouse anti-βIII-tubulin, clone 5G8 Promega, Southampton, UK 1:1000) immunohistochemistry using the staining protocol described previously.1–3 Fluorescent secondary antibodies (Invitrogen, UK) were used at a concentration of 1:1000 as previously described.1–3 To ensure correct section orientation, explants were counterstained
with 4',6-diamidino-2-phenylindole (DAPI) at a concentration of 1:10,000 as previously described (Invitrogen, UK). This allowed easy identification of the photoreceptor layer of the explants and examination of the general condition of the tissue preparation. It also helped to ensure correct orientation of the sections and focal plane for cell counts. Imaging was performed at 20× magnification using an epifluorescence microscope (DM6000; Leica Microsystems UK). Ganglion cells were identified based on morphology and plane of focus of the wholemount. Images were taken by a masked observer at three locations for each explant. Cell counts were performed using image analysis software (ImageJ, United States National Institutes of Health (NIH)). This software was used to trace dendritic morphology of individual ganglion cells as an overlay. The original image was then subtracted and Sholl analysis was performed on the resultant image. Between 10 and 20 dendritic tracings could be made for each image meaning a minimum of 240 dendritic tracings were analysed for each condition tested. Data points were generated at 2 μm intervals to a radius of 70 µm from the cell centre. This is typically the limit of identified dendritic arbors seen in section. Live imaging of superoxide formation. Retinal explants were prepared as above in triplicate and exposed to either: vehicle, angiotensin II (2 μM) or irbesartan (10 µM) in culture media as above. The explants were placed ganglion cell side down in glass-bottom culture wells (P35G-0-14C Matek Corp, Ashland, MA, USA) and incubated for 30 minutes at 37°C in a humidified incubator with 10 µM dihydroethidium (DHE) added to the medium (Sigma, UK) prior to rinsing with phosphatebuffered saline (PBS) and reincubation with the media plus drug or vehicle. At 30-minute intervals, explants were removed from the incubator, treated with a premixed Hoechst stain (NucBlu: Invitrogen, UK) as per the manufacturer’s instructions and imaged with a humidified inverted immunofluorescence microscope at 100× magnification (DMI 6000B Leica Microsystems, UK). Experiments were performed in triplicate and each explant was imaged only once. Successful imaging could be carried out for six hours using this technique. Images were taken with standardised intensity settings on three filter channels to ensure no bleed through of immunofluorescence between the channels. Intensity of DHE staining at each timepoint was quantified by image analysis software (ImageJ, NIH) by four masked observers who identified the points of greatest intensity in each image independently so as to serve as internal controls. This typically correlated with the Hoechst stain of cell nuclei. If no focal DHE intensity was detected in an image, observers were asked to plot five random points on the image. Pixel values of those points were quantified and compared to background intensity using a
Downloaded from jra.sagepub.com at Lucia Campus Library on June 16, 2015
3
White et al. Tabel 1. Primary antibodies used. Antigen
Host
Supplier
Reference
Blocking
Dilution
β-actin NOX2/gp91phox NOX4
Mouse Rabbit Rabbit
Abcam Abcam Abcam
ab6276 ab31092 ab16534
5% milk 5% milk 5% milk
1:15,000 1:1000 1:1000
NOX: nicotinamide adenine dinucleotide phosphate (NAPDH) oxidase 2.
custom software routine developed using MATLAB (Mathworks, Cambridge UK). Western blotting. Explants were prepared as described above. The retinal explants were kept for four hours in culture. The explants were in close contact with the medium supplemented with DMSO alone (control) or angiotensin II (2 μM) or irbesartan (10 µM) as described above. At the end of each timepoint, the retinae were carefully lifted out of the wells and transferred into tubes. The tissue was washed three times with PBS to remove any remaining medium. After centrifugation, the supernatant was discarded and the tissue stored at −80°C until RNA and protein isolation. Half of the retinal tissue was used for total protein isolation. The tissue was thawed on ice and for every ~5 mg tissue ~300 µl radioimmunoprecipitation assay (RIPA) buffer (Roche, supplemented with protein inhibitor) were added. The tissue was broken apart using an electric homogeniser (IKA-Werke, Staufen, Germany). Subsequently, the solution was incubated for up to two hours at 4°C with constant rotation. After lysis, the solution was centrifuged for 15 minutes at 4°C at 13,000 g to remove cell debris. The cell pellet was discarded and the supernatant, containing the proteins, was transferred into a fresh tube. The total protein concentration was determined using the bicinchoninic (BCA) protein assay kit (Thermo Scientific UK) following the manufacturer’s instructions. The proteins were denatured under reducing conditions in a buffer of sodium dodecyl sulphate (SDS), β-mercaptoethanol, glycerol and tris-glycine for five minutes at 95°C. A total of 20 µg of total proteins for each condition were then separated in a 4%–12% SDS Bis-tris gel (Invitrogen, UK) for one hour at 120 V. After the separation, the proteins were transferred onto a polyvinylidene difluoride (PVDF) membrane (Invitrogen, UK) for two hours at 50 V. The PVDF membrane was then blocked using tris-buffered saline supplemented with 0.5% tween-20 (TBS-T) with 5% skimmed milk solution for up to one hour at room temperature. This was followed by the incubation with the primary antibody solution (primary antibody diluted in blocking solution, see Table 1) overnight at 4°C. The next day, the blots were washed four times with TBS-T at room temperature (RT) and they were then incubated in secondary antibody solution (horseradish peroxidase (HRP)-conjugated secondary antibody (Vector) diluted 1:50,000 in TBS-T) for one hour at RT. Membranes were washed with TBS-T four
times before proteins were visualised using a chemiluminescent substrate (ECL, Amersham). To assess the size of the protein bands, Top of Form Precision Plus Protein™ Dual Color Standard (Biorad) was used. We visualised the protein bands in an Alliance chemiluminescence system (Uvitec, UK).
Reverse transcription-polymerase chain reaction (RT-PCR) From the experiments to determine protein expression, the other half of the retinal tissue was used for total RNA isolation according to the manufacturer’s instructions (RNA/ DNA/Protein Purification Kit, Norgen Biotek, Niagaraon-the Lake, Canada). In addition to the total RNA, proteins were also isolated and used for Western blot (see above). Following RNA isolation, residual genomic DNA was digested with DNase to avoid false-positive bands after PCR. Then, the messenger RNA (mRNA) was transcripted into cDNA using the Superscript R III first-strand kit (Invitrogen, UK) following the manufacturer’s instructions. The cDNA was then used for PCR. MangoTaq DNA polymerase (Bioline, UK) was used for all PCR reactions. Forward and reverse primers were designed to amplify the DNA fragment of interest. Glyceraldehyde 3-phosphate dehydrogenase (GAPDH) was used as a PCR positive control and housekeeping standard. Primers were 20 bp in length with a guanine/cytosine ratio that did not exceed 55%. Primers were designed to span exon/exon junctions in the mRNA to discriminate RNA bands from contaminating residual genomic DNA bands. Each PCR reaction contained 100 ng cDNA, 0.5 mM of each primer, 500 mM dNTPs, 1.5 mM MgCl2, 5× Taq DNA polymerase buffer, 5 U MangoTaq and ddH2O up to a final volume of 25 µl. First, the DNA template was denatured for five minutes at 94°C before amplification was started. The reaction was stopped after 35 cycles. Each cycle consisted of 30 seconds at 94°C, 30 seconds at 58°C and 30 seconds at 72°C. The final cycle was followed by a final extension of five minutes at 72°C. After the PCR, the products were separated on a 1% agarose gel (agarose in Trisborate-ethylenediaminetetraacetic acid (TBE) buffer (Invitrogen, UK) supplemented with 0.5 mg/ml ethidium bromide). Electrophoresis took place for 30 minutes at 90 V in TBE buffer. DNA bands were visualised using a longwave UVP transilluminator (Uvitec). Hyperladder II
Downloaded from jra.sagepub.com at Lucia Campus Library on June 16, 2015
4
Journal of the Renin-Angiotensin-Aldosterone System
Figure 1. Representative wholemount sections at 20× magnification treated either with: (a) control (dimethyl sulfoxide (DMSO)), (b) irbesartan (10 µM) or (c) angiotensin II (2 μM). Tissue is Day 4 post-treatment. Scale bar is 100 μM.
Figure 2. Ganglion cell counts from wholemount explant preparations. Average cell count for explants treated either with irbesartan (10 μM) or angiotensin II (2 μM) were compared to average cell counts of control explants treated with vehicle only (dimethyl sulfoxide (DMSO)). Error bars are standard error of the mean. Asterisk denotes p < 0.05.
(Bioline UK) served as a base pair standard. The brightness of the DNA bands of interest was compared to the housekeeping standard (GAPDH) to assess the concentration of the mRNAs of interest.
Results The retinal explant preparations were fixed in PFA at Day 4, which has previously been determined as the optimal time to differentiate neuroprotective effects.2 Average cell density for control-treated explants was 878 cells/mm2. Peak density was 1666 cells/mm2. Representative wholemount sections at 20× magnification treated either with irbesartan (10µM), control (DMSO) or angiotensin II (2 μM) are shown in Figure 1. Note that tissue treated with
angiotensin II was often significantly degraded compared to the other preparations consistent with the cell counts shown in Figure 2. Figure 2 shows average cell counts from three wholemount preparation experiments of three explants each treated either with irbesartan (10 µM), angiotensin II (2 µM) or vehicle (DMSO) normalised to control explant counts. There is a marked increase in ganglion cell counts in explants treated with irbesartan (10 µM) compared to control counts treated with vehicle (DMSO) after four days of culture. Conversely, there is a marked reduction in cell counts in explants treated with angiotensin II (2 µM) compared to control explants (p < 0.05, t test). The wholemount preparation allowed for limited analysis of cellular morphology of ganglion cells. Figure 3
Downloaded from jra.sagepub.com at Lucia Campus Library on June 16, 2015
5
White et al.
Figure 3. Sholl analysis of dendritic morphology compared to average profile of control sections stained in wholemount for BIII tubulin imaged at 20× magnification. Analysis was carried out on individual cell soma and dendritic morphology traced from the original photomicrographs (analysis of variance (ANOVA), p < 0.05). Error bars denote standard error of the mean.
shows a Sholl analysis of dendritic morphology compared to average profile of control sections. Analysis was carried out on individual cell soma and dendritic morphology traced from the original photomicrographs. A minimum of 200 cells were analysed from triplicate experiments for each condition. Only irbesartan (10 µM)-treated wholemounts showed preserved dendritic morphology at Day 4 (analysis of variance (ANOVA), p < 0.05).
Live imaging or reactive oxygen species (ROS) generation Figure 4 shows cells in the wholemount explant preparation stained with Hoechst stain and counterstained for intracellular reactive oxygen species (ROS) with DHE. Hoechst stain in live culture stains only active cell nuclei. Figure 4(a) shows the amount of ROS generated at four hours in a control explant. Figure 4(b) shows an explant at four hours treated with angiotensin II (2 μm) demonstrating a comparative increase in DHE staining compared to control. Figure 4(c) shows an explant at four hours treated with irbesartan (10 µM) with a comparative decrease in DHE staining intensity compared to control. Only active cell nuclei stain with Hoechst stain. Figure 4(b) shows a cell with intense DHE staining but no Hoechst stain (marked with arrow). The intensity of DHE staining, used as a surrogate measure of intracellular ROS generation, was quantified over time. The results are shown in Figure 5, which demonstrates an angiotensin II-related bimodal increase in ROS generation compared to control, which is almost completely attenuated by irbesartan (10 µM).
Modulation of the angiotensin II receptors AIIR1a and AIIR2 in treated retinal explants Modulation of angiotensin II receptor mRNA expression by irbesartan (10 µM) and angiotensin II (2 μM) was determined by RT-PCR at four hours post-dissection of the retinal explant. The results are shown in Figure 6. Treatment of retinal explants by angiotensin II completely silenced mRNA expression of AIIRIa and AIIR2. Irbesartan treatment increased mRNA expression of AIIRIa (p < 0.01, t test) and slightly decreased expression of AIIR2 mRNA, though this result was not statistically significant.
Modulation of subunits of the nicotinamide adenine dinucleotide phosphate (NADPH)dependent pathway in treated explants Modulation of expression of NAPDH oxidase 2 (NOX2), a principal source of intracellular superoxide generation at the ganglion cell layer in the retina,13 could be demonstrated in explants treated with either vehicle, angiotensin II (2 μM), or irbesartan (10 µM). The results are shown in Figure 7. Angiotensin II (2 μM) substantially increased expression of NOX2 compared to control (p < 0.01, t test). Irbesartan (10 µM) also slightly increased expression of NOX2 though this result was not significant. Intensity values are plotted on the graph at the right (Figure 7(b)). mRNA expression of NOX2 was determined by RT-PCR in explants treated by vehicle, angiotensin II or irbesartan (10 µM) at four hours. The results are shown in Figure 7(c) and (d). Angiotensin II (2 μM) completely
Downloaded from jra.sagepub.com at Lucia Campus Library on June 16, 2015
6
Journal of the Renin-Angiotensin-Aldosterone System
Figure 4. Live imaging of retinal explants at 100× magnification taken at four hours post-dissection.
Explants were treated either with vehicle (dimethyl sulfoxide (DMSO)), angiotensin II or irbesartan (10µM) at the concentrations shown. The top panels show active cell nuclei stained with Hoechst stain. Corresponding photomicrographs in the bottom panel show intracellular reactive oxygen species (ROS) stained with dihydroethidium (DHE). Note the corresponding cell (arrow) staining brightly with DHE but with a dim Hoechst stain. Scale bar shows 20 μm.
Figure 5. Dihydroethidium (DHE) staining intensity values normalised to the intensity of control explants measured at hourly intervals. Error bars are standard error of the mean.
silenced mRNA expression of NOX2 in retinal explants (p < 0.01, t test). Irbesartan (10 µM) slightly increased NOX 2 mRNA expression in keeping with protein expression (p < 0.01, t test, Figure 7(a) and (b)).
Discussion Organotypic ex vivo retinal explants examined in wholemount are a useful way to explore both the magnitude and mechanisms of neuroprotective strategies. Our study
Downloaded from jra.sagepub.com at Lucia Campus Library on June 16, 2015
7
White et al.
Figure 6. (a) Angiotensin II receptors At1Ra and AT2R messenger RNA (mRNA) expression was determined by reverse transcription-polymerase chain reaction (RT-PCR) at four hours in explants treated either with irbesartan (10 μM) or angiotensin II (2 μM) or control (dimethyl sulfoxide (DMSO)). mRNA expression of glyceraldehyde 3-phosphate dehydrogenase (GAPDH) is given as a reference. (b) Intensity value of angiotensin II receptors At1Ra and AT2R mRNA expression compared against control in explants treated with irbesartan (10 μM). Treatment of retinal explants by angiotensin II completely silenced mRNA expression of AIIRIa and AIIR2. Error bars are standard error of the mean. Asterisk denotes p < 0.05.
showed that modulation of the renin-angiotensin system in the eye, specifically modulation of angiotensin II receptors, appears to confer neuroprotection in wholemount cultures. This is in keeping with previous in vivo findings utilising candesartan4,6 and in vitro findings utilising telmisartan,7 other common angiotensin II blockers in clinical usage. The mechanism of these neuroprotective effects appears to involve modulation of production of ROS.6,7 We cannot provide a direct link between apoptosis and a rise in superoxide generation in our model but the circumstantial evidence derived from it is compelling. It has been shown previously that elevated intraocular pressure leads to the generation of ROS as a prelude to apoptosis8 in in vitro and in vivo models.9 To date, the neuroprotective effects of angiotensin II blockers have not been directly linked to modulation of free radical formation in an in vivo chronic ocular hypertensive model of
glaucoma, though the neuroprotective effect of candesartan has been validated without demonstration of mechanism.4 Candesartan modulated amelioration of intracellular free radical generation has, however, been demonstrated in an ocular ischemic/reperfusion model.6 Angiotensin II-dependent modulation of intracellular free radical generation associated with a neuroprotective effect has not been described in an axotomised, nutrient-deprivation model such as a retinal explant previously. Angiotensin II receptor distribution is widespread in the rodent retina10,11 and is known to be localised to RGCs in addition to other ocular structures.11 Receptor distribution on interneurons such as amacrine cells has suggested angiotensin II may have a role other than regulating ocular perfusion pressure, for example, in cellular signalling.10 Components of the renin-angiotensin system in the human retina have also been demonstrated.5,12,13 The results of our
Downloaded from jra.sagepub.com at Lucia Campus Library on June 16, 2015
8
Journal of the Renin-Angiotensin-Aldosterone System
Figure 7. (a) Western blot of nicotinamide adenine dinucleotide phosphate oxidase 2 (NOX2) expression in retinal explants at four hours post-dissection. Protein band was detected at 65 kDa. Beta actin with a band at 42 kDa is included as a reference. (b) Western blot intensity of NOX2 compared to control (dimethyl sulfoxide (DMSO)) in explants treated with irbesartan (10μM) or angiotensin II (2 μM). Error bars are standard deviation. Asterisk denoted p < 0.05. (c) NOX2 messenger RNA (mRNA) expression in explants assayed by reverse transcription-polymerase chain reaction (RT-PCR) at four hours. Glyceraldehyde 3-phosphate dehydrogenase (GADPH) is included as a reference. (d) Intensity of mRNA expression in explants treated with irbesartan (10 μM) or angiotensin II (2 μM) compared to control explants (DMSO). Error bars are standard error of the mean. Asterisk denoted p < 0.05.
study demonstrate a direct neuroprotective effect on RGCs in a retinal explant model by irbesartan (10 µM), an angiotensin II blocker. Moreover, modulation of this system could also lead to increased cell death by application of angiotensin II, a direct agonist. Not only were gross wholemount cell counts reduced (Figure 2) by application of angiotensin II but dendritic arborisation, a measure of cellular health, was also reduced by application of angiotensin II (Figure 3). The reverse of this observation was seen in explants treated with irbesartan (10 µM). We did not undertake studies of ganglion cell function in this experimental paradigm as it was outside the scope of our study. Angiotensin II modulates intracellular generation of free radical formation (Figures 4 and 5). This appears to occur at the angiotensin II receptor level, expression of which is modulated by drug therapy within a matter of hours (Figure 6). Generation of ROS is markedly increased by application of angiotensin II and almost completely silenced by application of irbesartan. NOX2, an NADPH oxidase found predominantly at the ganglion cell layer of the rodent retina,14 seems to play a role in this and is heavily influenced by the activity of angiotensin II (Figure 7(c) and (d)).
The mechanism behind this effect is clearly complex and multifactorial. However, our findings show modulation of AIIr1a is involved with AIIR2 having an antagonistic effect (Figure 6). At1R modulation of generation of superoxide species intracellularly is associated with cell death, demonstrated in our experiments by wholemount cell counts (Figure 2). The modulated ‘burst’ pattern (Figure 5) is similar to what has been previously described15 and implies some of the effect of superoxide generation relates to an intracellular signalling mechanism yet to be fully elucidated. The fact that NOX2 expression is maintained and even slightly increased in irbesartan-treated cells as detected by Western blot (Figure 7(a) and (b)) and RT-PCR (Figure 7(c) and (d)) suggests at least some superoxide generation is necessary for homeostasis in the eye. Angiotensin II seems to overstimulate this system, leading to a high level of NOX2 protein expression (Figure 7(a) and (b)) and ROS generation coinciding with a decrease in Hoechst staining (Fig 4, centre panel), suggestive of reduced nuclear activity, perhaps as a pre-apoptotic event. Angiotensin II stimulation of the explants also completely silences mRNA expression of NOX2 at four hours post-treatment (Figure 7(c)
Downloaded from jra.sagepub.com at Lucia Campus Library on June 16, 2015
9
White et al. and (d)), suggestive of a negative feedback mechanism in action at the same time. The exact mechanism is yet to be elucidated. In addition, angiotensin II is known to influence neuronal NMDA currents,16 leading to increased NOX 2-dependent generation of ROS.17 Modulation of this system is also implicated in observed neuroprotective effects in the ocular ischemia reperfusion model.6 Thus, our study adds to the circumstantial evidence that the renin-angiotensin system in the eye may play a role in modulating known triggers of apoptosis such as N-methylD-aspartate (NMDA) excitotoxicity and generation of intracellular ROS and so may provide a therapeutic basis for future neuroprotective strategies in ocular neurodegenerative diseases such as glaucoma. Acknowledgements For advice with the live imaging experiments, we thank Dr Richard Eva from the Centre for Brain Repair, Cambridge, and Dr Gillian Groeger from University College, Cork. We thank Li Wen for help in preparing the manuscript figures.
Conflict of interest None declared.
Funding This work was supported by the Cambridge Eye Trust, Fight for Sight UK and the Jukes Glaucoma Research Fund.
References 1. Johnson TV and Martin KR. Development and characterization of an adult retinal explant organotypic tissue culture system as an in vitro intraocular stem cell transplantation model. Invest Ophthalmol Vis Sci 2008; 49: 3503–3512. 2. Bull ND, Johnson TV, Welsapar G, et al. Use of an adult rat retinal explant model for screening of potential retinal ganglion cell neuroprotective therapies. Invest Ophthalmol Vis Sci 2011; 52: 3309–3320. 3. Bull N, Johnson T and Martin K. Organotypic explant culture of adult rat retina for in vitro investigations of neurodegeneration, neuroprotection and cell transplantation. Protoc exch 2011. DOI:10.1038/protex.2011.215. 4. Yang H, Hirooka K, Fukuda K, et al. Neuroprotective effects of angiotensin II type 1 receptor blocker in a rat model of chronic glaucoma. Invest Ophthalmol Vis Sci 2009; 50: 5800–5804. 5. White AJ, Cheruvu SC, Sarris M, et al. Expression of classical components of the renin angiotensin system in the
human eye. J Renin Angiotensin Aldosterone Syst. Epub ahead of print 6 October 2014. 6. Fujita T, Hirooka K, Nakamura T, et al. Neuroprotective effects of angiotensin II type 1 receptor (AT1-R) blocker via modulating AT1-R signaling and decreased extracellular glutamate levels. Invest Ophthalmol Vis Sci 2012; 53: 4099–4110. 7. Ozawa Y, Yuki K, Yamagishi R, et al. Renin-angiotensin system involvement in the oxidative stress-induced neurodegeneration of cultured retinal ganglion cells. Jpn J Ophthalmol 2013; 57: 126–132. 8. Ju WK, Liu Q, Kim KY, et al. Elevated hydrostatic pressure triggers mitochondrial fission and decreases cellular ATP in differentiated RGC-5 cells. Invest Ophthalmol Vis Sci 2007; 48: 2145–2151. 9. Moreno MC, Campanelli J, Sande P, et al. Retinal oxidative stress induced by high intraocular pressure. Free Radic Biol Med 2004; 37: 803–812. 10. Downie LE, Vessey K, Miller A, et al. Neuronal and glial cell expression of angiotensin II type 1 (AT1) and type 2 (AT2) receptors in the rat retina. Neuroscience 2009; 161: 195–213. 11. Wheeler-Schilling TH, Kohler K, Sautter M, et al. Angiotensin II receptor subtype gene expression and cellular localization in the retina and non-neuronal ocular tissues of the rat. Eur J Neurosci 1999; 11: 3387–3394. 12. Savaskan E, Löffler K, Meier F, et al. Immunohistochemical localisation of angiotensin-converting enzyme, angiotensin II and AT1 receptor in human ocular tissue. Ophthal Res 2004; 36: 312–320. 13. Wagner J, Jan Danser A, Derkx F, et al. Demonstration of renin mRNA, angiotensin mRNA, and angiotensin converting enzyme mRNA expression in the human eye: Evidence for an intraocular renin-angiotensin system. Br J Ophthalmol 1996; 80: 159–163. 14. Bhatt L, Groeger G, McDermott K, et al. Rod and cone photoreceptor cells produce ROS in response to stress in a live retinal explant system. Mol Vis 2010; 16: 283–293. 15. Lieven CJ, Thurber KA, Levin EJ, et al. Ordering of neuronal apoptosis signalling: A superoxide burst precedes mitochondrial cytochrome c release in a growth factor deprivation model. Apoptosis 2012; 17: 591–599. 16. Moreno MC, Sande P, Marcos HA, et al. Effect of glaucoma on the retinal glutamate/glutamine cycle activity. FASEB J 2005; 19: 1161–1162. 17. Wang G, Coleman CG, Chan J, et al. Angiotensin II slow-pressor hypertension enhances NMDA currents and NOX2-dependent superoxide production in hypothalamic paraventricular neurons. Am J Physiol Regul Integr Comp Physiol 2013; 304: R1096–R1106.
Downloaded from jra.sagepub.com at Lucia Campus Library on June 16, 2015
