Rotenone induces degeneration of photoreceptors and impairs the dopaminergic system in the rat retina

Neurobiology of Disease 44 (2011) 102–115

Contents lists available at ScienceDirect

Neurobiology of Disease j o u r n a l h o m e p a g e : w w w. e l s ev i e r. c o m / l o c a t e / y n b d i

Rotenone induces degeneration of photoreceptors and impairs the dopaminergic system in the rat retina Julián Esteve-Rudd a, Laura Fernández-Sánchez a, Pedro Lax a, Emilio De Juan a, José Martín-Nieto a, b, Nicolás Cuenca a, b,⁎ a b

Departamento de Fisiología, Genética y Microbiología, Facultad de Ciencias, Universidad de Alicante, E-03080 Alicante, Spain Instituto Multidisciplinar para el Estudio del Medio ‘Ramón Margalef’, Universidad de Alicante, E-03080 Alicante, Spain

a r t i c l e

i n f o

Article history: Received 25 February 2011 Revised 24 May 2011 Accepted 16 June 2011 Available online 25 June 2011 Keywords: Retina Parkinson disease Rotenone Retinal degeneration Pesticide Mitochondria Photoreceptor Dopaminergic neuron

a b s t r a c t Rotenone is a widely used pesticide and a potent inhibitor of mitochondrial complex I (NADH-quinone reductase) that elicits the degeneration of dopaminergic neurons and thereby the appearance of a parkinsonian syndrome. Here we have addressed the alterations induced by rotenone at the functional, morphological and molecular levels in the retina, including those involving both dopaminergic and non-dopaminergic retinal neurons. Rotenone-treated rats showed abnormalities in equilibrium, postural instability and involuntary movements. In their outer retina we observed a loss of photoreceptors, and a reduced synaptic connectivity between those remaining and their postsynaptic neurons. A dramatic loss of mitochondria was observed in the inner segments, as well as in the axon terminals of photoreceptors. In the inner retina we observed a decrease in the expression of dopaminergic cell molecular markers, including loss of tyrosine hydroxylase immunoreactivity, associated with a reduction of the dopaminergic plexus and cell bodies. An increase in immunoreactivity of AII amacrine cells for parvalbumin, a Ca2+-scavenging protein, was also detected. These abnormalities were accompanied by a decrease in the amplitude of scotopic and photopic a- and b-waves and an increase in the b-wave implicit time, as well as by a lower amplitude and greater latency in oscillatory potentials. These results indicate that rotenone induces loss of vision by promoting photoreceptor cell death and impairment of the dopaminergic retinal system. © 2011 Elsevier Inc. All rights reserved.

Introduction Parkinson disease (PD) is a neurodegenerative disorder characterized mainly by the degeneration of dopaminergic neurons of the central and peripheral nervous systems, and also affecting the visual system. Previous studies have provided a body of evidence concerning the alterations of visual functions in Parkinson disease patients and in the 1-methyl-4phenyl-1,2,3,6-tetrahydropyridine (MPTP)-treated monkey model (Bodis-Wollner, 2003, 2009; Tatton et al., 1990; Witkovsky, 2004), most of which can be reversed upon L-DOPA administration (Barbato

Abbreviations: CB, calbindin; COX, cytochrome c oxidase; ERG, electroretinogram; GCL, ganglion cell layer; INL, inner nuclear layer; IPL, inner plexiform layer; MPTP, 1methyl-4-phenyl-1,2,3,6-tetrahydropyridine; ONL, outer nuclear layer; OPL, outer plexiform layer; PD, Parkinson disease; PKCα, α isoform of protein kinase C; PV, parvalbumin; ROS, reactive oxygen species; TH, tyrosine hydroxylase. ⁎ Corresponding author at: Departamento de Fisiología, Genética y Microbiología, Universidad de Alicante, Campus Universitario San Vicente, P.O. Box 99, E-03080 Alicante, Spain. Fax: + 34 96 590 9569. E-mail addresses: [email protected] (J. Esteve-Rudd), [email protected] (L. Fernández-Sánchez), [email protected] (P. Lax), [email protected] (E. De Juan), [email protected] (J. Martín-Nieto), [email protected] (N. Cuenca). Available online on ScienceDirect (www.sciencedirect.com). 0969-9961/$ – see front matter © 2011 Elsevier Inc. All rights reserved. doi:10.1016/j.nbd.2011.06.009

et al., 1994; Bodis-Wollner and Tzelepi, 1998; Kupersmith et al., 1982; Price et al., 1992). Rotenone is a natural compound widely used as an insecticide, pesticide and piscicide (Bové et al., 2005) which elicits the appearance of neuropathological and biochemical defects similar to those detected in PD patients (Betarbet et al., 2006; Schober, 2004). Rotenone acts systemically as a potent inhibitor of mitochondrial complex I (NADHubiquinone oxidoreductase), whose activity levels are usually reduced in the substantia nigra, frontal cortex, blood platelets and skeletal muscle of parkinsonian patients (Büeler, 2009; Greenamyre et al., 2001; Mizuno et al., 1989; Parker et al., 1989; Schapira, 1998). Complex I inhibition causes bioenergetic defects and mitochondrial dysfunction, which is associated with the generation of oxidative stress (Büeler, 2009). Mutations in mitochondrial DNA genes that encode different subunits of complex I cause a severe thinning of the nerve fiber layer in Leber hereditary optic neuropathy (Huoponen, 2001). Also, PD patients and animal models show a thinning of this and/or other layers of the inner retina (Hajee et al., 2009; Inzelberg et al., 2004). In MPTP-treated animals (Chen et al., 2003; Ghilardi et al., 1988; Mariani et al., 1986; Wong et al., 1985) and in PD patients (Harnois and Di Paolo, 1990; Nguyen-Legros et al., 1993) the inadequate processing of visual information has been correlated with decreased retinal levels of dopamine and its rate-limiting synthesizing enzyme, tyrosine hydroxylase (TH), in dopaminergic processes. In this context, a significant decrease

J. Esteve-Rudd et al. / Neurobiology of Disease 44 (2011) 102–115

of dopamine levels and/or the number of retinal TH-immunoreactive neurons has been found in MPTP-treated mammals, such as mice (Mariani et al., 1986; Tatton et al., 1990), rabbits (Wong et al., 1985) and macaques (Cuenca et al., 2005), as well as in rats that have been chronically treated with rotenone (Biehlmaier et al., 2007). Alterations in retinal neurons other than dopaminergic cells have not been deeply studied in parkinsonian mammals, and in particular in the rotenone-treated rat. An important question is whether the set of visual dysfunctions associated with parkinsonism can be explained solely on the basis of dopaminergic neuronal degeneration or whether other retinal cell types are affected as well, as we have documented in MPTP-treated monkeys (Cuenca et al., 2005). Here we show that rotenone treatment induces a series of physiological, morphological and molecular changes in a variety of retinal cell types, in addition to dopaminergic amacrine cells, and including photoreceptor cell death. Materials and methods Animals Forty-nine Sprague–Dawley rats (Harlan, IN) weighing approximately 300 g and aged 2–3 months at the beginning of the experiment were divided into two groups, one of which was treated with rotenone and the other kept as control. Animals were housed in cages under controlled photoperiod (LD 12:12), temperature (23 °C) and humidity (55–60%). Dry food and water were available ad libitum. All animals were handled in accordance with the National Institutes of Health (NIH) and the European Union guidelines (Directive 86/609/EEC) for the use of laboratory animals, in order to minimize animal suffering and numbers used for experiments. Rotenone treatment, body weight assessment and mortality Rotenone was purchased from Sigma (St. Louis, MO) and dissolved in sterile natural oil Miglyol 812 (Acofarma, Barcelona, Spain) at 3 mg/ ml. Rotenone was given to the treated group subcutaneously every 48 h at 3 mg/kg for 60 days, whereas Miglyol alone was injected as vehicle to control rats (1 ml/kg). In order to adjust the amount of rotenone administered, their body weight was measured before drug administration. Rats that did not survive the treatment were not considered for further analysis. At the end of the two-month period rotenone-administered rats weighed approximately 19% less than control rats, and a mortality rate of 23% in the treated group was calculated. One week after the last administration of rotenone, video recordings of control and rotenone-treated rats were obtained, in order to evaluate motor impairments. Electroretinograms (ERG) Prior to recording, rats were adapted to darkness overnight. Animals were then anesthetized under a dim red light by i.p. injection of a ketamine (100 mg/kg) plus xylazine (4 mg/kg) solution, and maintained on a heated pad at 37 °C. Pupils were dilated by topical application of 1% tropicamide (Alcon Cusí, Barcelona). A drop of 0.2% polyacrylic acid carbomer (Viscotears; from Novartis, Barcelona) was instilled on the cornea to prevent dehydration and allow electrical contact with the recording electrode. Recordings were obtained using DTL fiber electrodes (X-Static silver-coated nylon conductive yarn) from Sauquoit Industries (Scranton, PA). A 25-gauge platinum needle inserted under the scalp between the two eyes served as the reference electrode. A gold electrode was placed in the mouth and served as ground. Anesthetized animals were placed on a Faraday cage and all experiments were performed in complete darkness. Scotopic flashinduced ERG responses were recorded from both eyes in response to light stimuli produced by a Ganzfeld stimulator. Light flashes were given for 10 ms at eleven different increasing intensities (ranging

103

from −3.2 to 2.0 log cd·m − 2). Three to ten consecutive recordings were averaged for each light presentation. The interval between light flashes was 10 s for dim flashes and up to 20 s for the highest intensity. Photopic responses were obtained after light adaptation at 10 cd·m − 2 during 20 min, and stimuli were of the same intensity and duration as for scotopic conditions. The ERG signals were amplified and band-pass filtered (1–1000 Hz, without notch filtering) using a DAM50 data acquisition board (World Precision Instruments, Aston, UK). Stimulus presentation and data acquisition (4 kHz) were performed using a PowerLab system (ADInstruments, Oxfordshire, UK). Recordings were saved on a computer and analyzed off-line. To visualize oscillatory potentials, the signal recorded was filtered between 100 and 1000 Hz. The amplitude of the a-wave was always measured from the baseline to the trough of the a-wave and the results were averaged. Likewise, the amplitude of the b-wave was measured from the trough of the a-wave to the peak of the b-wave and not to the peak of oscillations (which could exceed the b-wave apex), and averaged. For oscillatory potentials the maximum peak-totrough amplitude was considered. Quantitative RT-PCR The neural retina was dissected free from retinal pigment epithelium and other ocular tissues. Total RNA was extracted using the TRIzol reagent from Invitrogen (Carlsbad, CA, USA) and further treated with DNase I (Ambion; Austin, TX, USA) following the manufacturers' instructions. Reverse transcription to cDNA was carried out with random decamers using the RETROscript kit (Ambion). Real-time PCR was performed using specific primers for the rat Th (forward: 5′-CAGGGCTGCTGTCTTCCTAC-3′, reverse: 5′GGGCTGTCCAGTACGTCAAT-3′), Vmat2 (forward: 5′-TGCCTTCTCCAGCAGCTATG-3′, reverse: 5′-TGGAGCACAAAGAGCTGAATAG-3′) and Gapdh (forward: 5′-GGTGCTGAGTATGTCGTGGA-3′, reverse: 5′CTTCTGAGTGGCAGTGATGG-3′) cDNAs, and the qPCR FastStart SYBR Green reaction mix (Roche; Barcelona), in an ABI PRISM 7000 Sequence Detection system (Applied Biosystems; Foster City, CA, USA). PCR reactions were carried out per triplicate, and at least 3 animals per group were used. Results were analyzed by the ΔΔCt method, and are presented as the ratio between levels of the mRNA of interest to the mRNA of an internal housekeeping gene, Gapdh. Western blotting Soluble retinal proteins were extracted and subjected to immunoblotting analysis as previously described (Esteve-Rudd et al., 2010; Martínez-Navarrete et al., 2007). Proteins (50 μg/lane) were resolved by SDS-PAGE on 5–20% polyacrylamide-gradient gels and electrotransferred to Hybond-P PVDF membranes (GE Healthcare, Buckinghamshire, UK). Blots were blocked with 5% non-fat dry milk and then probed with primary antibodies to TH or β-actin (Table 1) overnight at 4 °C. Thereafter, they were incubated with horseradish peroxidase-conjugated goat anti-mouse IgG secondary antibodies (Pierce, Rockford, IL) at a 1:10,000 dilution for 1 h at room temperature. Proteins were detected by enhanced chemiluminescence using the SuperSignal West Dura substrate from Pierce followed by Hyperfilm ECL film (GE Healthcare) exposure to blots. Films were digitized on an Image Scanner system (GE Healthcare), protein band intensities were quantified using the NIH ImageJ 1.42 software, and data were normalized to β-actin levels. Retinal immunohistochemistry Cryostat vertical sections and retinal whole-mounts were obtained and processed for immunolabeling following well established procedures (Cuenca and Kolb, 1998; Cuenca et al., 1995; Martínez-Navarrete et al., 2007; Esteve-Rudd et al., 2010). For objective comparison, retinas from control and rotenone-treated rats were fully processed in parallel,

104

J. Esteve-Rudd et al. / Neurobiology of Disease 44 (2011) 102–115

Table 1 Primary antibodies used in this work. Molecular marker

Antibody

Source

Working dilution

Application

β-Actin Bassoon Brn3a Calbindin (CB) D-28K Calbindin D-28K Cytochrome c oxidase (COX), subunit IV Parvalbumin (PV)

Mouse, clone AC-15 Mouse, clone SAP7F407 Goat polyclonal Mouse, clone CB-955 Rabbit polyclonal Mouse, clone 20E8 Rabbit polyclonal

Sigma (St. Louis, MO), A-5441 Stressgen (Ann Arbor, MI), VAM-PS003 Santa Cruz Biotechnology, sc-31984 Sigma, C9848 SWant (Bellinzona, Switzerland), CB-38 Molecular Probes (Eugene, OR), A21348 SWant, PV-28

Protein kinase C (PKC), α isoform Tyrosine hydroxylase (TH)

Rabbit polyclonal Mouse, clone F-11

Santa Cruz Biotechnology, sc-10800 Santa Cruz Biotechnology (Santa Cruz, CA), sc-25269

1:10,000 1:1000 1:500 1:250 1:500 1:1000 1:500 1:250 1:100 1:1000 1:500

WB IHC sections IHC sections IHC whole mounts IHC sections IHC sections IHC sections IHC whole mounts IHC sections, whole mounts WB, IHC sections IHC whole mounts

Antigens are indicated in the 1st column. The 2nd column lists the animal origin of each antibody, and the clone designation is given for monoclonal antibodies. The 3rd column provides the commercial company and catalog reference. The 4th column gives the dilution at which each antibody was used depending on the application, listed in the last column. WB, Western blotting; IHC, immunohistochemistry.

including their mounting and staining on the same slide (Cuenca et al., 2005). All primary antibodies used in this work (summarized in Table 1) have been utilized in several previous studies and are well characterized by us and other authors regarding specific cell type immunostaining. Eyeballs were fixed in 4% paraformaldehyde, and cryoprotected sequentially in 15, 20 and 30% sucrose. After washing in 0.1 M sodium phosphate buffer pH 7.4 (PB), the cornea, lens and vitreous body were removed, and the retinas were processed either for vertical sections or whole-mounts. For cryostat sections, retinas were embedded in Tissue-Tek OCT (Sakura Finetek, Zoeterwouden, Netherlands) and frozen in liquid N2. Sixteen micrometer-thick sections were obtained at −25 °C, mounted on Superfrost Plus slides (Thermo Scientific, Madrid, Spain) and air dried. Before further use, slides were thawed and washed three times in PB, and treated with blocking solution (10% normal donkey serum, 0.5% Triton X-100 in PB) for 1 h. Sections were subjected to single or double immunostaining overnight at room temperature with antibodies against a set of molecular markers for distinct retinal neuronal types, at the dilutions indicated in Table 1 in PB plus 0.5% Triton X-100. The secondary antibodies used were donkey anti-mouse or anti-rabbit IgG conjugated to Alexa Fluor 488 or 555 (Molecular Probes, Eugene, OR) at a 1:100 dilution. The nuclear dye TO-PRO-3 iodide (Molecular Probes) was added at 1 μM simultaneously to secondary antibodies and slides were incubated for 1 h at room temperature. Images were finally obtained under a Leica (Wetzlar, Germany) TCS SP2 confocal laserscanning microscope. Final images from control and experimental subjects were processed in parallel using Adobe Photoshop 10.0. Immunoreactivity differences in vertical retinal sections were analyzed from their corresponding gray-intensity (range 0–255) profile plots obtained with the NIH ImageJ 1.42 software. For whole-mount immunohistochemistry, retinas were dissected out from the choroid and flat-mounted on a nitrocellulose filter, with the photoreceptor layer side up. The retinas were subjected to single or double immunostaining with primary antibodies at the dilutions indicated in Table 1 for 4 days at 4 °C. After washing with PB, retinas were incubated with the corresponding Alexa Fluor-conjugated secondary antibodies at a 1:100 dilution for 24 h. Retinas were then washed, mounted in Citifluor (Citifluor Ltd., London, UK) with the ganglion cell layer side up, and coverslipped for viewing under the Leica TCS SP2 confocal microscope.

Morphometric analysis Maps of whole-mount retinas were drawn at a 10× magnification, and the location of TH-positive cell bodies was represented by black spots. Cell counts were carried out on these drawings, and cell density was quantified and analyzed statistically. Measurements of the average thickness of retinal layers and counts of ganglion cells were

performed using the ImageJ software, and means obtained from 4–5 pictures of central retinal sections from each control and rotenonetreated rats were compared.

Statistical analysis A 2-tailed Student t-test was applied using the GraphPad software from Prism (La Jolla, CA), and p values b 0.05 were considered statistically significant. Data were plotted as the average ± standard error of the mean (SEM).

Results Rotenone-treated compared to control rats revealed a number of motor deficiencies, such as rotatory movements, postural disturbances and balance alterations. These symptoms were apparent to different degrees among treated rats (data not shown), likely reflecting varying degrees of brain dopaminergic cell loss.

Electroretinograms In order to evaluate the effects of rotenone on retinal function, ERG recordings from 14 control and 9 treated rats were obtained 1 week after the last dose of rotenone under scotopic and photopic conditions. As shown in Figs. 1A–F, rats treated with rotenone exhibited a decrease in the amplitude of a- and b-waves under both conditions. The maximum a-wave (89.4 ± 9.0 μV) and b-wave (272.8 ± 23.1 μV) recorded in rotenone-treated animals under scotopic conditions (Figs. 1A, C and D) were significantly lower than those obtained in control rats (a-wave: 245.1±9.9 μV; b-wave: 584.1±22.8 μV; pb 0.0001 in both cases). Also, the ERG responses recorded under photopic conditions (Figs. 1B, E and F) showed a significant decrease in the maximum amplitude of both waves in rotenone-treated rats (a-wave: 16.1±2.1 μV; b-wave: 110.1± 10.5 μV) in comparison with the control group (a-wave: 27.9±3.1 μV; b-wave: 181.8±15.4 μV; pb 0.05 and pb 0.005, respectively). These results suggested that rotenone had elicited damage to both rod- and cone-mediated visual pathways. We also detected a small difference, though significant (pb 0.05), in the b-wave latency, this being longer in rotenone-treated rats (63.9±1.9 ms) than in control rats (58.4±1.0 ms). No difference was found, however, in the a-wave latency between the two groups (control: 11.9±0.2 ms; rotenone: 12.7±0.3 ms). Oscillatory potentials in the scotopic ERG from rotenone-treated rats (amplitude: 59.9 ± 9.4 μV; latency: 32.8 ± 0.6 ms) showed a significantly lower amplitude (Fig. 1G; pb 0.0001), as well as a higher latency (Fig. 1H; pb 0.0001), compared with control animals (amplitude: 209.3±8.4 μV; latency: 27.6±0.6 ms).

J. Esteve-Rudd et al. / Neurobiology of Disease 44 (2011) 102–115

105

Fig. 1. Physiological and structural alterations in the rotenone-treated rat retina. (A) Examples of ERG traces from control (left) and rotenone-treated (right) rats recorded under scotopic conditions. (B) Idem A under photopic conditions. (C and D) Stimulus–response curves for the a-wave (C) and b-wave (D) of control (circles; n = 14) and rotenone-treated (squares; n = 9) rats obtained under scotopic conditions. (E and F) Idem C and D, respectively, in control (n= 13) and rotenone-treated (n= 7) rats under photopic conditions. (G) Examples of filtered oscillatory potential traces from scotopic ERGs recorded in control (upper graph) and rotenone-treated (lower graph) rats in response to a 1 cd·s/m2 stimulus. (H) Amplitude (upper graph) and latency (lower graph) of maximum oscillatory potentials from control (n= 11) and rotenone-treated (n= 7) rats. (I and J) Vertical sections of retinas from control (I) and rotenone-treated (J) rats, showing TO-PRO-labeled cell nuclei. (K) Quantification of the thickness of retinal layers of control (n= 4) and rotenone-treated (n= 5) rats. (L and M) Correlation between ONL thickness and scotopic maximum a-wave (L) or b-wave (M) amplitude measured in individual animals. H and K: Student's t-test (*p b 0.01; **p b 0.005; ***pb 0.0001; n.s., not significant, p N 0.05). L and M: Pearson's correlation coefficient (**pb 0.005; ***pb 0.0001). Error bars represent SEM values. ONL: outer nuclear layer, OPL: outer plexiform layer, INL: inner nuclear layer, IPL: inner plexiform layer, GCL: ganglion cell layer. Scale bars= 20 μm.

106

J. Esteve-Rudd et al. / Neurobiology of Disease 44 (2011) 102–115

Photoreceptor degeneration Since some of the changes observed in the ERG from rotenone-treated rats reflected possible deficiencies in the activity of photoreceptors, the outer retina was examined by means of confocal microscopy. Both the outer nuclear layer (ONL) and the outer plexiform layer (OPL) were noticeably thinner in rotenone-treated rats (Fig. 1J) compared to control subjects (Fig. 1I). This thinning of the outer retina was not uniform throughout the retina, but there were areas more affected by rotenone where the narrowing of the retina was more evident. Significant decreases in the thickness of both the ONL (48.5±0.4 μm in control group, 33.5±2.0 μm in treated group; pb 0.01) and the OPL (8.2±0.4 μm in control group, 4.9±0.3 μm in treated group;b 0.005) were observed in the rotenone-treated rat retina (Fig. 1K). This contrasted with observations made in the inner retina, where the thickness of the inner nuclear layer (INL; 26.3±1.0 μm in control group, 24.6±2.6 μm in treated group) and the inner plexiform layer (IPL; 49.05±5.89 μm in control group,

47.48±3.55 μm in treated group) did not exhibit any significant change. We also evaluated the possible correlation between retinal function (by means of ERG) and ONL thickness (taken as an indicator of the number of photoreceptors in the retina), which was plotted as a function of the maximum amplitudes of scotopic a- (Fig. 1L) and b-waves (Fig. 1M). Indeed, we observed a statistically-significant linear correlation between ONL thickness and the amplitudes of both the a- (slope, 0.102±0.0175; R2 =0.8506; Pearson's r=0.9223) and b-wave (slope, 0.044±0.007; R2 =0.8628; Pearson's r=0.9289) in the scotopic ERG.

Loss of synaptic connectivity in the OPL Given the significant reduction in the thickness of the OPL and in the amplitudes of scotopic and photopic b-waves in rotenone-treated rats, synaptic connectivity between photoreceptors and horizontal as well as bipolar cells was assessed.

Fig. 2. Alterations in synaptic connectivity between photoreceptors and rod bipolar cells in the retina of rotenone-treated rats. (A and B) Immunolabeling of retinal vertical sections for PKCα in control (A) and rotenone-treated (B) rats. Arrows in A point to rod bipolar cell bodies and arrowheads to their dendrites in the OPL. (C and D) Immunolabeling for Bassoon in control (C) and rotenone-treated (D) rats. (E and F) Double immunolabeling for PKCα (green) and Bassoon (red) in control (E) and rotenone-treated (F) rats. Arrowheads in F point to rod bipolar cell dendrites sprouting into the ONL. Nuclei are labeled with TO-PRO (blue) in all panels. Scale bars = 10 μm.

J. Esteve-Rudd et al. / Neurobiology of Disease 44 (2011) 102–115

We first analyzed rod bipolar cells in retinal vertical sections by immunolabeling for the α isoform of protein kinase C (PKCα), a specific marker of such cell subtype in mammals. In control rats, the cell bodies of rod bipolars are located in the distal part of the INL (Fig. 2A, arrows), and their primary dendrites branch in the OPL (Fig. 2A, arrowheads). In contrast, in rotenone-treated rats rod bipolar cells exhibited a significant reduction in the length and density of their dendrites (Fig. 2B). Vertical sections of retinas from control and rotenone-treated rats were immunolabeled with antibodies against Bassoon, a constituent protein of synaptic ribbons present in the presynaptic axon terminals of photoreceptors (Brandstätter et al., 1999). In control rats, a continuous punctate staining uniformly distributed along the OPL was observed (Fig. 2C). In contrast, in rotenone-treated rats the Bassoon-positive staining in the OPL was discontinuous and with fewer immunoreactive dots, indicating a decreased number of photoreceptor terminals (Fig. 2D). A double immunostaining for PKCα and Bassoon showed that in control rats the dendritic tips of rod bipolar cells (labeled for PKCα in green, Fig. 2E) were associated with rod spherules (labeled for Bassoon in red, Fig. 2E). In rats treated with rotenone, the reduction of rod bipolar dendritic terminals was accompanied by a diminished immunoreactivity for Bassoon, this showing a discontinuous distribution along the OPL (Fig. 2F). Additionally, some bipolar cells lacked their characteristic dendritic arborization, and isolated dendrites that sprouted into the ONL could be observed (Fig. 2F, arrowheads). Horizontal cells were visualized in retinal vertical sections by immunolabeling with antibodies against calbindin, a specific marker for this neuronal type in rodents which forms a sublayer in the outermost part of the INL. In control rats, a continuous and relatively dense plexus formed by horizontal cell processes was observed in the OPL (Fig. 3A, arrowheads), whereas in animals treated with rotenone these cells exhibited fewer processes, thus resulting in a significant thinning of the plexus in the OPL (Fig. 3B). Double staining with antibodies against Bassoon and calbindin allowed to visualize synaptic contacts between photoreceptor axon terminals and horizontal cell processes. In control rats, terminals of horizontal cell processes were associated with dots of Bassoon immunoreactivity (Fig. 3C). Accordingly, this double staining systematically labeled pairs of red and green structures, corresponding to photoreceptor synaptic ribbons and horizontal cell dendrites. The same double immunostaining of rotenone-treated rat retinas showed a clear reduction in the number of Bassoon-calbindin pairs, and those remaining were discontinuously distributed across the OPL (Fig. 3D). Synaptic connectivity between horizontal and rod bipolar cells was additionally analyzed in whole-mount retinas immunostained with antibodies against calbindin and PKCα. There was an evident reduction in the dendritic plexus formed by both cell types in the OPL, which harbored a discontinuous distribution across the retina in rotenone-treated rats (Figs. 3F and H) compared with control subjects (Figs. 3E and G). Defects in dopaminergic amacrine cells We assessed the effects of rotenone at the molecular and cellular levels on the retinal dopaminergic system of rats. The expression levels of TH, the rate-limiting enzyme of dopamine biosynthesis and a molecular marker of dopaminergic neurons, was evaluated by Western blotting in the retina of control and rotenone-treated rats (Fig. 4A). TH protein levels were approximately 70% lower (31.0± 10.9%) in the retina of rotenone-treated than in control rats (Fig. 4B), which was indicative of a dramatic loss of retinal dopaminergic neurons and/or their plexus. qRT-PCR analysis revealed a decrease in the expression of Th (Fig. 4C) and Vmat2 (Fig. 4D) genes at the mRNA level in the retina of rotenone-treated rats, their relative expression levels being 57.7 ± 8.4% (Fig. 4C) and 66.0 ± 1.0% (Fig. 4D), respectively. This led us to analyze TH immunoreactivity in vertical sections of the retina of control and rotenone-treated rats. Dopaminergic neurons exhibited a normal cell

107

morphology in control rats (Figs. 4E and G), forming a dense, THpositive dendritic plexus in the outermost stratum (S1) of the IPL, and with some dendritic processes also observable in the OPL and INL (Figs. 4E and G). In contrast, in vertical sections of rotenone-treated rat retinas TH immunofluorescence from cell bodies was decreased, and there were significantly fewer dopaminergic processes in stratum S1 of the IPL (Figs. 4F and H). The corresponding density profile plots of control (Fig. 4I) and rotenone-treated (Fig. 4J) rat retinas showed that the TH-immunoreactivity peak area in this stratum had decreased dramatically in the second group, by 50% on average compared with control subjects (Fig. 4K). We next set to quantify the number of dopaminergic amacrine neurons in control and parkinsonian rat retinas. To this end, we counted TH-positive cells in whole-mount retinas from control (Fig. 5A) and rotenone-treated subjects (Fig. 5B). A loss of dopaminergic cell bodies in the second group, by an average of 47.3% (untreated rats, 15.0± 2.1 cell bodies/mm 2; rotenone-treated rats, 7.9 ± 3.5 cell bodies/mm2), was observed (Fig. 5C). Furthermore, in TH-labeled whole-mount retinas of rotenone-treated rats, the complexity of the dopaminergic dendritic network in the IPL S1 stratum was found to be reduced in both the central (Fig. 5E) and peripheral (Fig. 5G) retinas, when compared with untreated control rats (Figs. 5D and F, respectively). Defects in AII amacrine cells Synaptic connectivity was also assessed between dopaminergic and AII amacrine cells, the main cell subtype postsynaptic to dopaminergic amacrines. With this purpose, whole-mount retinas were immunolabeled with antibodies against TH and parvalbumin (PV), a marker for AII amacrine cells in rodents. In control rats, this double immunostaining showed the meshwork of fine, varicose dendrites of dopaminergic neurons (Fig. 6A, green) encircling AII cell bodies (Fig. 6A, red) and forming ring-like structures (Fig. 6A, arrows). The same view of a rotenone-treated rat retina showed a marked loss of the structures usually formed by dopaminergic dendrites around AII cells (Fig. 6B). Pictures obtained at a higher magnification show more clearly the reduction of dopaminergic ring-like structures around AII cells in rotenone-treated rats (Fig. 6D) compared with control subjects (Fig. 6C), which conceivably should translate into a significant loss of synaptic connectivity between the two neuronal subtypes in rotenone-treated rats. We also analyzed the morphology of AII amacrine cells. As illustrated in PV-immunolabeled sections of control rat retinas, shown in Fig. 6E, AII cells bear a mitral-shaped cell body located in the innermost INL stratum, and a primary stout dendrite from which lobular appendages full of mitochondria emerge restricted to sublamina a of the IPL. In addition, bushy distal dendrites ramify down into the lowest stratum of sublamina b. There were no apparent morphological changes in AII amacrine cells of rotenone-treated rats, although an increase in PV immunoreactivity levels was observed (Fig. 6F). To confirm this observation, the density profiles from PVimmunostained control and rotenone-treated rat retinas were obtained and quantitated (Figs. 6G and H, respectively). A statistically-significant increase of PV immunoreactivity was found in rotenonetreated rats (p b 0.05), by 65% on average compared with control subjects (Fig. 6I). This increase was observed both in cell bodies located in the INL (Fig. 6F arrows, and 6H segment 1) and in dendritic processes in the IPL (Fig. 6F arrowheads, and 6H segment 2), and was attributable to overexpression of this protein in response to rotenoneinduced stress either by amacrine AII themselves or by other bipolar cells located in the INL that normally express lower PV levels. Ganglion cell loss In order to assess if rotenone treatment affected ganglion cell number, we immunostained vertical sections with antibodies against Brn3a, a specific marker for ganglion cells. Fig. 7A shows a control

108

J. Esteve-Rudd et al. / Neurobiology of Disease 44 (2011) 102–115

Fig. 3. Alterations in synaptic connectivity between photoreceptors and horizontal cells in the retina of rotenone-treated rats. (A and B) Immunolabeling of retinal vertical sections for calbindin (CB) in control (A) and rotenone-treated (B) rats. Arrowheads in A point to horizontal cell dendrites in the OPL. (C and D) Double immunolabeling for CB (green) and Bassoon (red) in control (C) and rotenone-treated (D) rats. Nuclei are labeled with TO-PRO (blue). (E–H) Immunolabeling of whole-mount retinas for CB (E and F) or CB plus PKCα (G and H) in control (E and G) and rotenone-treated (F and H) rats. Scale bars = 10 μm.

retina stained for Brn3a (green) and TO-PRO (blue). In the ganglion cell layer, double-stained neurons were observed together with cells labeled only with TO-PRO, the latter corresponding to displaced amacrine cells (Figs. 7 A and B arrows). In treated animals a decrease of ganglion cells immunostained with the Brn3a antibody were observed (Fig. 7B), but the quantification of ganglion cell number in vertical sections showed no statistical significance (39 ± 3.29 cells/mm in control group, 32.4 ± 2.18 cells/mm in treated group; Mann–Whitney test, p b 0.26) (Fig. 7C).

Mitochondrial impairments Since rotenone is known to act as a mitochondrial complex I inhibitor, we evaluated the presence of mitochondria in the retina of control and rotenone-treated rats. Vertical retinal sections were immunolabeled with antibodies against subunit IV of the cytochrome c oxidase (COX) complex of the mitochondrial respiratory chain. In the outer retina of control rats, mitochondria were mainly distributed in the inner segments of photoreceptors, as well as in their axon terminals

J. Esteve-Rudd et al. / Neurobiology of Disease 44 (2011) 102–115

109

Fig. 4. Reduced expression of dopaminergic cell molecular markers in the retina of rotenone-treated rats. (A) Western blotting analysis of tyrosine hydroxylase (TH) in 3 control and 3 rotenone-treated rats. (B) TH protein levels normalized to those of β-actin from control (Ctrl; n = 5) and rotenone-treated (Rot; n = 7) rat retinas. (C) Expression levels of Th mRNA in control (n = 3) and rotenone-treated (n= 3) rats, normalized to Gapdh mRNA levels. (D) Idem Vmat2 mRNA in control (n= 3) and rotenone-treated (n = 4) rats. (E–H) Immunostaining for TH of retinal sections from control (E and G) and rotenone-treated (F and H) rats. Nuclei are stained with TO-PRO (blue). (I and K) Profile plots of average gray intensity for each horizontal line of panels G and H, respectively. (K) Relative TH-immunoreactivity levels in control (n= 3) and rotenone-treated (n= 3) rats. B–E and L: Student's t-test (*p b 0.05; ***pb 0.0005; n.s, p N 0.05). Error bars represent the SEM. Scale bars= 20 μm.

(Figs. 8A and C). Here a single, giant mitochondrion is localized in each rod spherule that could be readily visualized by confocal microscopy (Johnson et al., 2007) (Figs. 8C and D, arrowheads). In rotenone-treated rats, a dramatic loss of mitochondria was observed in the inner

segments of photoreceptors, as well as in their axon terminals located in the OPL (Figs. 8B and D). In the inner plexiform layer no apparent differences in COX immunoreactivity were found between control and rotenone-treated animals (Figs. 8A and B).

110

J. Esteve-Rudd et al. / Neurobiology of Disease 44 (2011) 102–115

Fig. 5. Decreased dopaminergic neuron number and plexus in the retina of rotenone-treated rats. (A and B) Representative low-magnification pictures of TH-immunolabeled whole-mount retinas with focus at the level of dopaminergic cell somata in the INL of control (A) and rotenone-treated (B) rats. (C) Dopaminergic neuron density in control (n = 4) and rotenone-treated (n= 5) rats quantitated from whole-mount retinal preparations immunolabeled for TH (as those shown in A and B). (D–G) High-magnification confocal stacks showing the somata and dendritic plexus of dopaminergic neurons in the central (D, control; E, rotenone) and peripheral (F, control; G, rotenone) retina. C: Student's t-test (**pb 0.01). Error bars represent the SEM. Scale bars: A and B = 100 μm; D–G = 20 μm.

Discussion Rotenone is a widely used pesticide that elicits the death of midbrain dopaminergic neurons and hence the development of parkinsonism, thus being commonly used to generate animal models of idiopathic PD (Terzioglu and Galter, 2008; Uversky, 2004). In this study, we have characterized its effects at the functional and cellular levels in the retina

of rats. We have found that rotenone induces several functional alterations, including a decrease in the amplitude of a- and b-waves. These functional defects correlated with a loss of photoreceptors and a decrease of the synaptic connectivity in the OPL, as well as with a severe impairment of the retinal dopaminergic system. These alterations were associated with a loss of mitochondria located in the inner segments and axon terminals of photoreceptors. To our knowledge, this work provides

J. Esteve-Rudd et al. / Neurobiology of Disease 44 (2011) 102–115

111

Fig. 6. Loss of synaptic connectivity between dopaminergic and AII amacrine cells in parkinsonian rats. (A and B) Immunolabeling for TH (green) and parvalbumin (PV; red) of whole-mount retinas from control (A) and rotenone-treated (B) rats. (C and D) Magnified areas of the retina from control (C) and rotenone-injected (D) rats labeled for TH and PV. (E and F) PV-immunostaining of retinal sections from control (E) and rotenone-treated (F) rats. Nuclei are stained with TO-PRO (blue). (G and H) Profile plots of average gray intensity for each horizontal line of panels E and F, respectively (1: immunoreactivity in the INL; 2: immunoreactivity in the IPL). (I) Relative, added-up PV-immunoreactivity levels in control (n = 4) and rotenone-treated (n = 5) rats. Scale bars: A, B, E and F = 20 μm; C, D = 10 μm. I: Student's t-test (*p b 0.05). Error bars represent the SEM.

the first evidence that functional and morphological alterations occur in several retinal cell types after rotenone administration, which may induce mitochondrial impairments and oxidative stress in the retina.

In our experiments, rotenone-treated rats showed abnormalities in equilibrium, postural instability and rotatory movements, in agreement with previous studies in which parkinsonian rotenone-treated rats

112

J. Esteve-Rudd et al. / Neurobiology of Disease 44 (2011) 102–115

Fig. 7. Quantification of ganglion cell number in rotenone-treated rats. (A and B) Vertical sections of retinas showing ganglion cells doubly stained with antibodies against Brn3a (green) and with TO-PRO (blue) from control (A) and treated (B) animals. (C) Quantification of the number of ganglion cells in control (Ctrl, n= 4) and rotenone-treated (Rot, n= 6) rats, showing no statistically significant differences. Mann–Whitney test (p b 0.26). Scale bars: A and B= 100 μm.

exhibited akinesia, bradykinesia, rigidity and/or stereotypic rotations (Alam and Schmidt, 2002; Betarbet et al., 2000; Sindhu et al., 2005). We also found that at the end of the rotenone treatment the rats' weight was approximately 19% less than in controls. Kraft et al. (1987) described the effects on retinal function of zinc and vitamin A deficiency in rats. They used 21 day-old animals, which were fed for 90 days a diet lacking zinc and vitamin A. This caused them a severe double deficiency of zinc and vitamin A, and many died by the end of the treatment. ERG analyses performed on the surviving animals showed that they required a 1.5 log unit increase in the light stimulus intensity to obtain a scotopic b-wave amplitude equivalent to that obtained in control rats. Our rats were fed a standard diet during the 60 days of rotenone treatment, and when the loss of weight became evident, food was added directly into the cage in order to circumvent the possibility that rats were not reaching the food due to motor alterations. Moreover, the functional and morphological alterations described here are more severe and were obtained in a shorter time frame than those described by Kraft et al. (1987). Therefore, while we cannot rule out the possibility that a lack of vitamin A and/or zinc could be involved to a certain extent in the alterations described in the present article, we can argue that rotenone effects are the major cause of those alterations. In the present study, the outer retina of rotenone-treated rats showed a loss of photoreceptors, as well as a reduced synaptic connectivity between them and their postsynaptic neurons: horizontal and bipolar cells. These results are in accordance with the decrease we found in the amplitude of scotopic and photopic a- and b-waves, as well as the increase in the b-wave implicit time. In fact, the decreases in the ONL and OPL thickness were significantly correlated with the lower-than-normal maximum amplitudes of a- and b-waves, respectively. In accordance with our experiments, deficiencies in the b-wave are more evident than in the a-wave in PD patients (Ellis and Ikeda, 1988; Ikeda et al., 1994) and MPTP-treated animals (Ghilardi et al., 1988; Wong et al., 1985). Moreover, there is a greater latency in the b- compared to the a-wave in the ERG from parkinsonian humans

(Archibald et al., 2009; Bodis-Wollner, 2003; Gottlob et al., 1987; Langheinrich et al., 2000; Nightingale et al., 1986; Peppe et al., 1998; Stanzione et al., 1990) and monkeys (Ghilardi et al., 1988). The above mentioned changes in the functionality of the retina cannot be explained solely on the basis of decreased levels of dopamine induced by rotenone treatment. A good deal of evidence suggests that the retina is sensitive to excitotoxic damage induced by excessive activation of ionotropic NMDA and/or AMPA glutamate receptors or, more generally, mediated by abnormally-high levels of intracellular Ca 2+. Indeed, massive influx of Ca 2+ into the cytoplasm promotes the activation of proapoptotic pathways and hence neuronal death. In this regard, excitotoxic damage has been demonstrated in glaucoma (Dreyer et al., 1996; Casson, 2006) and in animal models of retinitis pigmentosa (Delyfer et al., 2005) and diabetic retinopathy (Semkova et al., 2010). Under normal conditions, the incoming Ca 2+ can be sequestered in the mitochondria or buffered in the cytosol by the action of Ca 2+binding proteins, such as calbindin or parvalbumin. Ellipsoids of photoreceptors bear a high density of mitochondria, and the presynaptic region of rod photoreceptors contains a single giant mitochondrion in each spherule, whereas cones contain an average of five medium-sized mitochondria in each pedicle (Szikra and Krizaj, 2007; Johnson et al., 2007). The presence of rotenone would cause a mitochondrial impairment through the blockage of the mitochondrial respiratory chain at complex I, which in turn would cause a significant reduction in ATP levels and an increase of reactive oxygen species (ROS). Subsequently, this would induce the activation of mitochondria-dependent apoptotic pathways (Dauer and Przedborski, 2003; Vila and Przedborski, 2004) and an increase of cytosolic Ca 2+ (Nicholls, 2008; Yadava and Nicholls, 2007). Neurodegeneration should start in synaptic terminals of neurons in the retina, where mitochondria are expected to exhibit a lower respiratory capacity (Davey et al., 1997). Therefore, rotenone-induced bioenergetic defects could be related to the deficiencies in synaptic connectivity we observed in this work in the OPL of rotenone-treated rats. The fact that calbindin exerts a neuroprotective role against Ca 2+-mediated neurodegeneration could be the reason why we did not observe a loss of horizontal cells in the retina of rats treated with rotenone, despite the strong reduction of their dendritic and axonal tips. This hypothesis would be consistent with the fact that calbindin-positive dopaminergic neurons in the substantia nigra of PD patients and MPTP-treated animals are spared from degeneration (Damier et al., 1999; German et al., 1992). In the inner retina, a decrease of the IPL thickness and number of ganglion cells was observed. This result was in keeping with the thinning of the inner retina found in Parkinson patients (Hajee et al., 2009). However, the decrease we found in rotenone-treated rats was not statistically significant, probably due to the short time of treatment or the variable degree of sensitivity of rotenone likely existing among rats. In the INL we observed a decrease in the expression of dopaminergic cell molecular markers, and an associated loss of TH immunoreactivity derived from a reduction of both the dopaminergic plexus and the number of dopaminergic neurons, all of which was consistent with previous studies made in different animal models of PD (Chen et al., 2003; Cuenca et al., 2005; Tatton et al., 1990; Wong et al., 1985). The loss of TH-positive amacrine cells we found in the present study was consistent with that reported by Biehlmaier et al. (2007) in rats that were chronically treated with rotenone. However, these authors did not describe any change in the thickness of the different retinal layers. The discrepancy between our results and theirs could be explained by the regime of rotenone administration, rather than by the dose. They administered rotenone at 2.5 mg/kg intraperitoneally daily during 60 days, whereas we administered rotenone at 3 mg/kg subcutaneously every 48 h during 60 days. Therefore, the cumulative amount of rotenone received by an

J. Esteve-Rudd et al. / Neurobiology of Disease 44 (2011) 102–115

113

Fig. 8. Mitochondrial impairments in rotenone-treated rat retinas. (A and B) Immunolabeling for cytochrome c oxidase (COX; red) of retinal sections from control (A) and rotenoneinjected (B) rats. (C and D) High magnification of the outer retina showing a decrease of COX immunoreactivity in photoreceptor outer segments and a loss of giant mitochondria (arrowheads) located in rod spherules in rotenone-treated animals (D) compared with controls (C). Nuclei are stained with TO-PRO (blue). OS: outer segments; IS: inner segments; ONL: outer nuclear layer, OPL: outer plexiform layer, INL: inner nuclear layer. IPL: inner plexiform layer, GCL: ganglion cell layer. Scale bars: A and B = 20 μm, C and D = 10 μm.

individual rat at the end of the treatment was higher in the Biehlmaier et al. (2007) regime than in ours. This also suggests that the subcutaneous way is more efficient for rotenone administration than intraperitoneal injection, at least with the purpose of inducing alterations in the retina. AII amacrine cells are the major neurons postsynaptic to dopaminergic cells, playing a crucial role in the scotopic visual pathway by allowing the transmission of rod-generated signals to the cone-mediated visual pathway. In this context, we have found a dramatic loss in both the dopaminergic plexus and the number of dopaminergic neurons in the retina of rotenone-treated rats. This, together with the degeneration of synaptic terminals of horizontal, bipolar and dopaminergic cells observed here in the IPL and OPL of rotenone-treated rats and previously by our group in MPTP-treated monkeys (Cuenca et al., 2005) may explain, at least in part, the impairments in light adaptation under scotopic conditions, as well as the loss of contrast sensitivity observed in PD patients. Dopaminergic neurons in the retina of rats treated with rotenone showed a reduced number of dendritic rings surrounding the bodies of AII cells. This loss of connectivity was not associated with apparent

morphological changes in AII amacrine cells, but instead with an increase in immunoreactivity for parvalbumin, a Ca 2+-scavenging protein. This result contrasts with previous observations in the retina of MPTP-treated macaques, where we found a decrease in calretinin (a marker of AII amacrine cells in the retina of primates) immunoreactivity levels, a loss of the lobular appendages full of mitochondria of AII cells in strata S1 and S2 of the IPL, and a decrease in their dendritic arborization in strata S3–S5 (Cuenca et al., 2005). The discrepancies between our present results and those obtained in parkinsonian monkeys could be explained, in addition to speciesattributable differences, by the fact that monkeys were treated with MPTP for 2 years, whereas the rats tested in this study were rotenonetreated for a much shorter period, of 2 months. In addition, rotenone was administered subcutaneously and is able to diffuse through cell membranes freely and systemically, whereas MPTP was administered intravenously and is selectively taken up by dopaminergic neurons (Uversky, 2004). Diverse mechanisms candidate to be involved in the selective degeneration of dopaminergic neurons induced by rotenone in the midbrain have been proposed. This selectivity could be due to the

114

J. Esteve-Rudd et al. / Neurobiology of Disease 44 (2011) 102–115

dopamine biosynthesis and degradation pathways, exclusive of these neurons. Indeed, the oxidation of dopamine and its metabolites leads to the production of ROS. Therefore, any deficiency in the cellular protective systems against oxidative stress (ubiquitin–proteasome, antioxidant proteins) could trigger their degeneration (Ogawa et al., 2005; Stokes et al., 1999). Additionally, the rotenone-promoted degeneration of dopaminergic neurons could be due to the intrinsic physiology of these cells. In this context, young dopaminergic neurons generate pacemaker potentials mediated by Na + channels, whereas in adult neurons these potentials are mediated by voltage-dependent Ca2+ channels (Chan et al., 2009; Surmeier, 2007). This Ca2+ dependence increases with age and leads to elevated and sustained Ca2+ intracellular levels, particularly within the dendrites of dopaminergic neurons (Chan et al., 2007, 2009; Surmeier, 2007). Ca2+, in turn, can stimulate the respiratory metabolism and thereby the production of ROS (Orrenius et al., 2003). This adds to the fact that mitochondria also play an important role in regulating Ca2+ intracellular levels through its sequestration, as mentioned above. Therefore, mitochondrial dysfunction induced by rotenone in amacrine dopaminergic cells can contribute to the deregulation of cytosolic Ca2+ levels, resulting in cytotoxic damage and neuronal death. With regard to AII amacrine cells, it can be envisioned that overexpression of the Ca2+-buffering protein, parvalbumin, could represent a protective mechanism for these neurons against such cytotoxic damage. This would be consistent with the increase of PV levels in dopaminergic neurons reported in the substantia nigra of PD patients (Soós et al., 2004). To summarize, our results show a correlation between functional and structural alterations in the retina of rotenone-treated rats, the latter extending to photoreceptors and their synaptic connections with second order neurons. Our findings provide as well possible mechanisms involved in the rotenone-induced retinal defects, and explain at the cellular level many of the visual dysfunctions detected by means of electrophysiological (ERG, visual evoked potentials) and psychophysical techniques in patients and animal models of PD (Archibald et al., 2009; Bodis-Wollner, 2003, 2009; Witkovsky, 2004). It would be therefore of great interest to assess on a regular basis retinal structure and function in parkinsonian patients. Acknowledgments We thank David Williams for advice and critical reading of the manuscript. This research was supported by grants from the Spanish Ministerio de Ciencia e Inovación (MICIN; BFU2009-07793/BFI), Instituto de Salud Carlos III, (RETICS RD07/0062/0012), Organización Nacional de Ciegos de España (ONCE), Fundación Lucha contra la Ceguera (FUNDALUCE) and Fundación Médica Mutua Madrileña. References Alam, M., Schmidt, W.J., 2002. Rotenone destroys dopaminergic neurons and induces parkinsonian symptoms in rats. Behav. Brain Res. 136, 317–324. Archibald, N.K., Clarke, M.P., Mosimann, U.P., Burn, D.J., 2009. The retina in Parkinson's disease. Brain 132, 1128–1145. Barbato, L., Rinalduzzi, S., Laurenti, M., Ruggieri, S., Accornero, N., 1994. Color VEPs in Parkinson's disease. Electroencephalogr. Clin. Neurophysiol. 92, 169–172. Betarbet, R., Sherer, T.B., MacKenzie, G., Garcia-Osuna, M., Panov, A.V., Greenamyre, J.T., 2000. Chronic systemic pesticide exposure reproduces features of Parkinson's disease. Nat. Neurosci. 3, 1301–1306. Betarbet, R., Canet-Aviles, R.M., Sherer, T.B., Mastroberardino, P.G., McLendon, C., Kim, J.H., et al., 2006. Intersecting pathways to neurodegeneration in Parkinson's disease: effects of the pesticide rotenone on DJ-1, α-synuclein, and the ubiquitin–proteasome system. Neurobiol. Dis. 22, 404–420. Biehlmaier, O., Alam, M., Schmidt, W.J., 2007. A rat model of Parkinsonism shows depletion of dopamine in the retina. Neurochem. Int. 50, 189–195. Bodis-Wollner, I., 2003. Neuropsychological and perceptual defects in Parkinson's disease. Parkinsonism Relat. Disord. 9, S83–S89. Bodis-Wollner, I., 2009. Retinopathy in Parkinson disease. J. Neural Transm. 116, 1493–1501. Bodis-Wollner, I., Tzelepi, A., 1998. The push–pull action of dopamine on spatial tuning of the monkey retina: the effects of dopaminergic deficiency and

selective D1 and D2 receptor ligands on the pattern electroretinogram. Vision Res. 38, 1479–1487. Bové, J., Prou, D., Perier, C., Przedborski, S., 2005. Toxin-induced models of Parkinson's disease. NeuroRx 2, 484–494. Brandstätter, J.H., Fletcher, E.L., Garner, C.C., Gundelfinger, E.D., Wässle, H., 1999. Differential expression of the presynaptic cytomatrix protein bassoon among ribbon synapses in the mammalian retina. Eur. J. Neurosci. 11, 3683–3693. Büeler, H., 2009. Impaired mitochondrial dynamics and function in the pathogenesis of Parkinson's disease. Exp. Neurol. 218, 235–246. Casson, R.J., 2006. Possible role of excitotoxicity in the pathogenesis of glaucoma. Clin. Exp. Ophthalmol. 34, 54–63. Chan, C.S., Guzman, J.N., Ilijic, E., Mercer, J.N., Rick, C., Tkatch, T., et al., 2007. ‘Rejuvenation’ protects neurons in mouse models of Parkinson's disease. Nature 447, 1081–1086. Chan, C.S., Gertler, T.S., Surmeier, D.J., 2009. Calcium homeostasis, selective vulnerability and Parkinson's disease. Trends Neurosci. 32, 249–256. Chen, S.T., Hsu, J.R., Hsu, P.C., Chuang, J.I., 2003. The retina as a novel in vivo model for studying the role of molecules of the Bcl-2 family in relation to MPTP neurotoxicity. Neurochem. Res. 28, 805–814. Cuenca, N., Kolb, H., 1998. Circuitry and role of substance P-immunoreactive neurons in the primate retina. J. Comp. Neurol. 393, 439–456. Cuenca, N., De Juan, J., Kolb, H., 1995. Substance P-immunoreactive neurons in the human retina. J. Comp. Neurol. 356, 491–504. Cuenca, N., Herrero, M.T., Angulo, A., De Juan, E., Martínez-Navarrete, G.C., López, S., et al., 2005. Morphological impairments in retinal neurons of the scotopic visual pathway in a monkey model of Parkinson's disease. J. Comp. Neurol. 493, 261–273. Damier, P., Hirsch, E.C., Agid, Y., Graybiel, A.M., 1999. The substantia nigra of the human brain: II. Patterns of loss of dopamine-containing neurons in Parkinson's disease. Brain 122, 1437–1448. Dauer, W., Przedborski, S., 2003. Parkinson's disease: mechanisms and models. Neuron 39, 889–909. Davey, G.P., Canevari, L., Clark, J.B., 1997. Threshold effects in synaptosomal and nonsynaptic mitochondria from hippocampal CA1 and paramedian neocortex brain regions. J. Neurochem. 69, 2564–2570. Delyfer, M.N., Forster, V., Neveux, N., Picaud, S., Léveillard, T., Sahel, J.A., 2005. Evidence for glutamate-mediated excitotoxic mechanisms during photoreceptor degeneration in the rd1 mouse retina. Mol. Vis. 11, 688–696. Dreyer, E.B., Zurakowski, D., Schumer, R.A., Podos, S.M., Lipton, S.A., 1996. Elevated glutamate levels in the vitreous body of humans and monkeys with glaucoma. Arch. Ophthalmol. 114, 299–305. Ellis, C.J.K., Ikeda, H., 1988. Evidence for retinal dopamine deficiency in Parkinson's disease. In: Bodis-Wollner, I., Piccolino, M. (Eds.), Dopaminergic Mechanisms in Vision. Alan R. Liss, New York, pp. 239–251. Esteve-Rudd, J., Campello, L., Herrero, M.-T., Cuenca, N., Martín-Nieto, J., 2010. Expression in the mammalian retina of parkin and UCH-L1, two components of the ubiquitin–proteasome system. Brain Res. 1352, 70–82. German, D.C., Manaye, K.F., Sonsalla, P.K., Brooks, B.A., 1992. Midbrain dopaminergic cell loss in Parkinson's disease and MPTP-induced parkinsonism: sparing of calbindin-D28k-containing cells. Ann. N Y Acad. Sci. 648, 42–62. Ghilardi, M.F., Bodis-Wollner, I., Onofrj, M.C., Marx, M.S., Glover, A.A., 1988. Spatial frequency-dependent abnormalities of the pattern electroretinogram and visual evoked potentials in a parkinsonian monkey model. Brain 111, 131–149. Gottlob, I., Schneider, E., Heider, W., Skrandies, W., 1987. Alteration of visual evoked potentials and electroretinograms in Parkinson's disease. Electroencephalogr. Clin. Neurophysiol. 66, 349–357. Greenamyre, J.T., Sherer, T.B., Betarbet, R., Panov, A.V., 2001. Complex I and Parkinson's disease. IUBMB Life 52, 135–141. Hajee, M.E., March, W.F., Lazzaro, D.R., Wolintz, A.H., Shrier, E.M., Glazman, S., et al., 2009. Inner retinal layer thinning in Parkinson disease. Arch. Ophthalmol. 127, 737–741. Harnois, C., Di Paolo, T., 1990. Decreased dopamine in the retinas of patients with Parkinson's disease. Invest. Ophthalmol. Vis. Sci. 31, 2473–2475. Huoponen, K., 2001. Leber hereditary optic neuropathy: clinical and molecular genetic findings. Neurogenetics 3, 119–125. Ikeda, H., Head, G.M., Ellis, C.J., 1994. Electrophysiological signs of retinal dopamine deficiency in recently diagnosed Parkinson's disease and a follow up study. Vision Res. 34, 2629–2638. Inzelberg, R., Ramirez, J.A., Nisipeanu, P., Ophir, A., 2004. Retinal nerve fiber layer thinning in Parkinson disease. Vision Res. 44, 2793–2797. Johnson Jr., J.E., Perkins, G.A., Giddabasappa, A., Chaney, S., Xiao, W., White, A.D., et al., 2007. Spatiotemporal regulation of ATP and Ca2+ dynamics in vertebrate rod and cone ribbon synapses. Mol. Vis. 13, 887–919. Kraft, S.P., Parker, J.A., Matuk, Y., Rao, A.V., 1987. The rat electroretinogram in combined zinc and vitamin A deficiency. Invest. Ophthalmol. Vis. Sci. 28, 975–984. Kupersmith, M.J., Shakin, E., Siegel, I.M., Lieberman, A., 1982. Visual system abnormalities in patients with Parkinson's disease. Arch. Neurol. 39, 284–286. Langheinrich, T., Tebartz van Elst, L., Lagrèze, W.A., Bach, M., Lücking, C.H., Greenlee, M.W., 2000. Visual contrast response functions in Parkinson's disease: evidence from electroretinograms, visually evoked potentials and psychophysics. Clin. Neurophysiol. 111, 66–74. Mariani, A.P., Neff, N.H., Hadjiconstantinou, M., 1986. 1-Methyl-4-phenyl-1,2,3,6tetrahydropyridine (MPTP) treatment decreases dopamine and increases lipofuscin in mouse retina. Neurosci. Lett. 72, 221–226. Martínez-Navarrete, G.C., Martín-Nieto, J., Esteve-Rudd, J., Angulo, A., Cuenca, N., 2007. αSynuclein gene expression profile in the retina of vertebrates. Mol. Vis. 13, 949–961. Mizuno, Y., Ohta, S., Tanaka, M., Takamiya, S., Suzuki, K., Sato, T., et al., 1989. Deficiencies in complex I subunits of the respiratory chain in Parkinson's disease. Biochem. Biophys. Res. Commun. 163, 1450–1455.

J. Esteve-Rudd et al. / Neurobiology of Disease 44 (2011) 102–115 Nguyen-Legros, J., Harnois, C., Di Paolo, I., Simon, A., 1993. The retinal dopamine system in Parkinson's disease. Clin. Vis. Sci. 8, 1–12. Nicholls, D.G., 2008. Oxidative stress and energy crises in neuronal dysfunction. Ann. N Y Acad. Sci. 1147, 53–60. Nightingale, S., Mitchell, K.W., Howe, J.W., 1986. Visual evoked cortical potentials and pattern electroretinograms in Parkinson's disease and control subjects. J. Neurol. Neurosurg. Psychiatry 49, 1280–1287. Ogawa, N., Asanuma, M., Miyazaki, I., Diaz-Corrales, F.J., Miyoshi, K., 2005. L-DOPA treatment from the viewpoint of neuroprotection. Possible mechanism of specific and progressive dopaminergic neuronal death in Parkinson's disease. J. Neurol. 252, IV23–IV31. Orrenius, S., Zhivotovsky, B., Nicotera, P., 2003. Regulation of cell death: the calciumapoptosis link. Nat. Rev. Mol. Cell Biol. 4, 552–565. Parker Jr., W.D., Boyson, S.J., Parks, J.K., 1989. Abnormalities of the electron transport chain in idiopathic Parkinson's disease. Ann. Neurol. 26, 719–723. Peppe, A., Stanzione, P., Pierantozzi, M., Semprini, R., Bassi, A., Santilli, A.M., et al., 1998. Does pattern electroretinogram spatial tuning alteration in Parkinson's disease depend on motor disturbances or retinal dopaminergic loss? Electroencephalogr. Clin. Neurophysiol. 106, 374–382. Price, M.J., Feldman, R.G., Adelberg, D., Kayne, H., 1992. Abnormalities in color vision and contrast sensitivity in Parkinson's disease. Neurology 42, 887–890. Schapira, A.H., 1998. Human complex I defects in neurodegenerative diseases. Biochim. Biophys. Acta 1364, 261–270. Schober, A., 2004. Classic toxin-induced animal models of Parkinson's disease: 6-OHDA and MPTP. Cell Tissue Res. 318, 215–224. Semkova, I., Huemmeke, M., Ho, M.S., Merkl, B., Abari, E., Paulsson, M., et al., 2010. Retinal localization of the glutamate receptor GluR2 and GluR2-regulating proteins in diabetic rats. Exp. Eye Res. 90, 244–253. Sindhu, K.M., Saravanan, K.S., Mohanakumar, K.P., 2005. Behavioral differences in a rotenone-induced hemiparkinsonian rat model developed following intranigral or median forebrain bundle infusion. Brain Res. 1051, 25–34.

115

Soós, J., Engelhardt, J.I., Siklós, L., Havas, L., Majtényi, K., 2004. The expression of PARP, NF-κB and parvalbumin is increased in Parkinson disease. Neuroreport 15, 1715–1718. Stanzione, P., Fattapposta, F., Tagliati, M., D'Alessio, C., Marciani, M.G., Foti, A., et al., 1990. Dopaminergic pharmacological manipulations in normal humans confirm the specificity of the visual (PERG-VEP) and cognitive (P300) electrophysiological alterations in Parkinson's disease. Electroencephalogr. Clin. Neurophysiol. Suppl. 41, 216–220. Stokes, A.H., Hastings, T.G., Vrana, K.E., 1999. Cytotoxic and genotoxic potential of dopamine. J. Neurosci. Res. 55, 659–665. Surmeier, D.J., 2007. Calcium, ageing, and neuronal vulnerability in Parkinson's disease. Lancet Neurol. 6, 933–938. Szikra, T., Krizaj, D., 2007. Intracellular organelles and calcium homeostasis in rods and cones. Vis. Neurosci. 24, 733–743. Tatton, W.G., Kwan, M.M., Verrier, M.C., Seniuk, N.A., Theriault, E., 1990. MPTP produces reversible disappearance of tyrosine hydroxylase-containing retinal amacrine cells. Brain Res. 527, 21–31. Terzioglu, M., Galter, D., 2008. Parkinson's disease: genetic versus toxin-induced rodent models. FEBS J. 275, 1384–1391. Uversky, V.N., 2004. Neurotoxicant-induced animal models of Parkinson's disease: understanding the role of rotenone, maneb and paraquat in neurodegeneration. Cell Tissue Res. 318, 225–241. Vila, M., Przedborski, S., 2004. Genetic clues to the pathogenesis of Parkinson's disease. Nat. Med. 10, S58–S62. Witkovsky, P., 2004. Dopamine and retinal function. Doc. Ophthalmol. 108, 17–40. Wong, C., Ishibashi, T., Tucker, G., Hamasaki, D., 1985. Responses of the pigmented rabbit retina to NMPTP, a chemical inducer of parkinsonism. Exp. Eye Res. 40, 509–519. Yadava, N., Nicholls, D.G., 2007. Spare respiratory capacity rather than oxidative stress regulates glutamate excitotoxicity after partial respiratory inhibition of mitochondrial complex I with rotenone. J. Neurosci. 27, 7310–7317.