B R A I N R E S E A R C H 1 3 5 2 ( 2 0 1 0 ) 7 0 – 8 2

ava i l ab l e a t www.sc i enced i r ec t . com

www.e l sev i e r . com/ loca te /b ra i n res

Research Report

Expression in the mammalian retina of parkin and UCH-L1,two components of the ubiquitin-proteasome system

Julián Esteve-Rudda, Laura Campelloa, María-Trinidad Herrerob,Nicolás Cuencaa,c,⁎, José Martín-Nietoa,c

aDepartamento de Fisiología, Genética y Microbiología, Facultad de Ciencias, Universidad de Alicante, E-03080 Alicante, SpainbNiCE-CIBERNED, Departamento de Anatomía Humana y Psicobiología, Facultad de Medicina, Universidad de Murcia, E-30071 Murcia, SpaincInstituto 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

⁎ Corresponding author.Departamento de FisiP.O. Box 99, E-03080 Alicante, Spain. Fax: +34

E-mail address: cuenca@ua.es (N. CuencaAbbreviations: CB, calbindin; CR, calretinin

layer; OPL, outer plexiform layer; PV, parvalbtyrosine hydroxylase

0006-8993/$ – see front matter © 2010 Elsevidoi:10.1016/j.brainres.2010.07.019

A B S T R A C T

Article history:Accepted 7 July 2010Available online 15 July 2010

The ubiquitin-proteasome system (UPS) functions as a major degradation pathway formisfolded and damaged proteins with an important neuroprotective role in the CNS against avariety of cellular stresses. Parkin and ubiquitin C-terminal hydrolase L1 (UCH-L1) are tworelevant components of theUPS associatedwith a number of neurodegenerative disorders.Wehere address the expression profile of parkin and UCH-L1 in the mammalian retina, withspecial emphasisonprimates.Wedescribe for the first timethepresenceofparkin in the retinaof mammals, including humans. Parkin and UCH-L1 genes were expressed at the mRNA andprotein levels in the retina of all species examined. The immunolocalization pattern of parkinwas quite widespread, being expressed by most retinal neuronal types, includingphotoreceptors. UCH-L1 was localized to horizontal cells and specific subtypes of bipolar andamacrine cells, as well as to ganglion cells and their axons forming the nerve fiber layer. Inrodents no UCH-L1 immunoreactivity was found in cone or rod photoreceptors, whereas thisprotein was present along the whole length of cones in all othermammals. Remarkably, UCH-L1wasexpressedbydopaminergic amacrinecells ofprimates.Theampledistributionofparkinand UCH-L1 in the mammalian retina, together with the crucial role played by the UPS innormal neuronal physiology in the brain, points to a participation of these two proteins in theubiquitin-proteasomal pathway of protein degradation in most retinal cell types, where theycould exert a protective function against neuronal stress.

© 2010 Elsevier B.V. All rights reserved.

Keywords:RetinaProteasomeParkinUCH-L1Gene expressionParkinson disease

1. Introduction

Ubiquitin (Ub) is a small protein modifier with a critical role incell protection under health and disease. Ub is attached

ología, Genética yMicrobi96 590 9569.).; GCL, ganglion cell layer;umin; Ub, ubiquitin; UCH

er B.V. All rights reserved

covalently as oligo-Ub chains to protein substrates, acting as atag for protein targeting to the 26S proteasome, amulti-subunitdegradation complex (Glickman and Raveh, 2005). The Ub-proteasomesystem(UPS) thereby functionsasa cellular garbage

ología, Universidad de Alicante, Campus Universitario San Vicente

INL, inner nuclear layer; IPL, inner plexiform layer; NFL, nerve fibe-L1, Ub C-terminal hydrolase L1; UPS, Ub-proteasome system; TH

.

,

r

,

71B R A I N R E S E A R C H 1 3 5 2 ( 2 0 1 0 ) 7 0 – 8 2

removal service which clears damaged and misfolded proteinsthat form within the cell under normal and stress conditions(Hershko et al., 2000; Tan et al., 2009). Protein ubiquitination notonly regulates protein degradation, but also cell cycle progres-sion, developmental processes, metabolism, signal transduc-tion, transcriptional activity and synaptic plasticity (DiAntonioand Hicke, 2004; Shang and Taylor, 2004). Ubiquitinatingenzymes are of three types: Ub-activating (E1), Ub-conjugating(E2) andUb-ligating (E3) proteins (Hershko et al., 2000; Dev et al.,2003; Tan et al., 2009). The substrate specificity of this process iscontroled by E3 Ub-protein ligases, which react with Ub-boundE2 enzymes and carry out ubiquitination of the substrate.Conversely, Ub-protein conjugates are disassembled by deubi-quitinating proteases, among which Ub C-terminal hydrolasesconstitute amajor class (Fischer, 2003). The released substrate isthen transferred to the proteasome for degradation in thecytoplasm, and free Ub regenerated for reutilization (Fischer,2003; Gongand Leznik, 2007). The balance between the activitiesof ubiquitinating and deubiquitinating enzymes thus deter-mines the rate of substrate degradation, as well as the levels offree Ub and Ub conjugates. The UPS is highly active in neurons(DiAntonio andHicke, 2004), and its dysfunction is involved in anumber of neurodegenerative disorders, such as Parkinson,AlzheimerandHuntingtondiseases (GongandLeznik, 2007; Tanet al., 2009), and including retinitis pigmentosa and diabeticretinopathy in the visual system (Mendes et al., 2005; Adachi-Uehara et al., 2006).

Parkin, encoded by the PARK2 gene, is a 465-amino acidenzyme responsible for the ubiquitination of misfolded anddamaged proteins by acting as an E3-type Ub-protein ligase(Dev et al., 2003; Tan et al., 2009). Parkin interacts with E2enzymes by virtue of its RING finger domains and is also ableto ubiquitinate itself and promote its own degradation. Thisprotein is especially abundant in the mammalian brain,including the substantia nigra (Horowitz et al., 1999; Shimuraet al., 1999; Stichel et al., 2000; Zarate-Lagunes et al., 2001),where it locates in the cytoplasm of neurons, at synapticterminals and associated with membranes (Biskup et al.,2008). Mutations in the PARK2 gene have been found in ca. 50%of patients with autosomal recessive Parkinson and in 10–15%of sporadic cases of this disease with early onset. A greatvariety of point mutations, deletions spanning several exons,insertions, duplications and splice-site mutations have beenidentified (OMIM 602544), most of which result in loss ofparkin Ub ligase activity. Also, decreased parkin levels havebeen reported in the substantia nigra of Parkinson patientsand animal models of parkinsonism (Shimura et al., 1999;Zarate-Lagunes et al., 2001). In addition to its role in the UPS,parkin has been proposed to have an involvement in synapticfunction, mitochondrial homeostasis and nervous systemdevelopment, and is in general considered to exert aprotective role in neurons of the CNS against a wide spectrumof cellular stresses (Dev et al., 2003; Biskup et al., 2008;Henchcliffe and Beal 2008; Tan et al., 2009). These includeoxidative stress derived from mitochondrial dysfunction(Palacino et al., 2004; Periquet et al., 2005).

Ubiquitin C-terminal hydrolase L1 (UCH-L1, formerlyPGP9.5), encoded by the UCHL1 gene (locus PARK5, OMIM191342), is a 223-amino acid neuron-specific Ub thiolesterasethat catalyzes the hydrolysis of C-terminal poly-Ub chains to

produce Ub monomers. As a component of the UPS, it has arelevant involvement in Ub turnover and protein degradation,and hence in overall neuronal health in the CNS (Fischer, 2003;Gong and Leznik, 2007). This enzyme, however, exhibits a dualactivity, being able to function also as an Ub ligase upon itsdimerization (Liu et al., 2002). UCH-L1 is highly expressed inneuronal cell bodies and processes in various regions of thebrain, among them the substantia nigra (Doran et al., 1983;Saigoh et al., 1999; Choi et al., 2004; Barrachina et al., 2006).Although highly infrequent, heterozygous mutations in theUCHL1 gene leading to loss of hydrolase activity have beenrelated to parkinsonism with Lewy body formation in men(Leroy et al., 1998), and in homozygosity to the gracile axonaldystrophy (gad) phenotype inmice (Saigoh et al., 1999), whereasloss of its ligase activity appears to confer a lower susceptibilityto Parkinson development (Liu et al., 2002). Besides, UCH-L1protein levels are down-regulated in the brain of patients withidiopathic Parkinson, Alzheimer and related diseases (Choi etal., 2004; Barrachina et al., 2006). Other than in Ub homeostasis,roles for UCH-L1 have been suggested in the modulation ofneurogenesis and apoptosis, oxidative stress protection, andsynaptic plasticity and function (Harada et al., 2004; Lansbury,2006; Gong and Leznik, 2007; Biskup et al., 2008).

In themammalian retina, which is an extension of the CNS,knowledge on the UPS is scarce, and the role of Ub and relatedenzymes in retinal aging and disease essentially unknown.Apart from studies in the brain, the expression and distribu-tion of parkin in the normal retina of mammals has not beenaddressed so far, and that of UCH-L1 has not been character-ized in depth. A growing body of experimental evidence hasaccumulated concerning visual dysfunction and morpholog-ical impairments in the retina of parkinsonian patients andexperimental animals (Witkovsky, 2004; Cuenca et al., 2005a;Archibald et al., 2009; Bodis-Wollner, 2009). We thus set toanalyze the expression of parkin and UCH-L1 genes at themRNA and protein levels and to characterize their localizationpattern in the distinct retinal layers and cell types. This studyinvolved a spectrum of mammalian species ranging fromrodents to humans.

2. Results

By means of RT-PCR, we detected expression of the PARK2gene at the mRNA level in the retina of all mammals studied,including the rodent (mouse and rat), bovine and primate(monkey and human) retinas, as shown in Fig. 1A. Immuno-blotting analysis revealed the presence in the retina of the52 kDa parkin polypeptide encoded by the canonical PARK2transcript (Fig. 1B). In addition, two smaller parkin isoforms, ofca. 48 and 37 kDa, were detected in the retina of most species.These molecular sizes were in keeping with several previousreports in the mammalian brain (Horowitz et al., 1999; Stichelet al., 2000; Pawlyk et al., 2003).

Immunohistochemical analysis using polyclonal antibo-dies to parkin from Chemicon revealed that this proteindistributed throughout the cell bodies and dendrites of retinalneurons, as shown for the mouse (Fig. 2A), rat (Fig. 2C), bovine(Fig. 2E) and human (Fig. 2G) retinas. A fairly consistent parkinexpression pattern was found among these mammals. It was

Fig. 1 – Expression of the parkin gene in the retina ofmammals. A: RT-PCR analysis of PARK2 mRNA in the retinafrom the indicated species (Rat, Sprague–Dawley). Theamplified bands were verified by DNA sequencing and theirmolecular sizes are given to the right. B: Immunoblottinganalysis of parkin protein. Molecular masses of bandsobtained are given to the right, where the closed arrowheadindicates the main parkin isoform and open arrowheadsother detected parkin variants.

72 B R A I N R E S E A R C H 1 3 5 2 ( 2 0 1 0 ) 7 0 – 8 2

observed that photoreceptors were immunolabeled alongtheir entire length in all species. Parkin immunoreactivitywas also found in the outermost inner nuclear layer (INL),where horizontal cells are located, in bipolar cells and in thelayer of amacrine cells (Fig. 2A, C, E and G; arrows). As well,parkin was present in most neurons in the ganglion cell layer(GCL) of the mouse (Fig. 2A) and rat (Fig. 2C) retinas, where itslevels were especially prominent. In contrast, in the bovine(Fig. 2E) and human (Fig. 2G) retinas parkin was expressed bycertain ganglion cell subtypes. Finally, weakly-immunoreac-tive processes were also observed in the inner plexiform layer(IPL) of all mammals analyzed. Similar results were obtainedwhen using polyclonal antibodies to parkin from Santa CruzBiotechnology. Negative control stainings in which parkinantibodies were omitted failed to show any immunoreactivityin the retina of mammals above (Fig. 2B, D, F and H).

The UCHL1 mRNA was also expressed in the retina of allspecies studied, as revealed by RT-PCR (Fig. 3A). Immunoblot-ting analysis revealed that this transcript was effectivelytranslated into protein to yield the expected UCH-L1 26 kDapolypeptide (Fig. 3B).

We also used confocal microscopy to analyze the UCH-L1expression profile in the retina of different mammals. In themouse (Fig. 4A) and rat (Fig. 4B) retinas this protein was foundin the cell bodies of horizontal cells as well as in theirprocesses and terminal tips in the outer plexiform layer (OPL;arrowheads). UCH-L1 immunoreactivity was also observed inspecific subtypes of bipolar and amacrine cells located in the

INL, as well as in ganglion cell bodies and axons (Fig. 4A and B).In the squirrel retina, UCH-L1 was also present in horizontalcell bodies and processes located in the OPL (Fig. 4C), while aweaker immunostaining was found in some amacrine andbipolar cells in the INL. Ganglion cells of the squirrel alsodisplayed UCH-L1 expression, although a particularly promi-nent labeling was found in the nerve fiber layer (NFL) (Fig. 4C).Certain subtypes of horizontal, bipolar and amacrine cellswere found to contain UCH-L1 as well in the rabbit, bovine,human and monkey retinas (Figs. 4D–F and 6A, respectively),together with the somata and axons of ganglion cells. Incontrast to rodents, expression of UCH-L1 was detected incone photoreceptors of those species, including their outerand inner segments. UCH-L1 was also present in the IPL,although without a clear stratification pattern in the mouse(Fig. 4A) and rat (Fig. 4B) retinas. However, in the IPL of thesquirrel two strata were observed to exhibit a higher numberof immunopositive processes, namely S1 and S5, togetherwithaweaker staining in stratumS2/S3 (Fig. 4C, arrowheads). In therabbit (Fig. 4D), bovine (Fig. 4E), human (Figs. 4F and 5E) andmonkey (Fig. 6A) retinas the IPL strata S1, S3 and S5 werefound to be UCH-L1-immunoreactive, as illustrated(arrowheads).

We have further characterized the UCH-L1 distributionpattern in the retina of primates. Fig. 5A shows in higher detailexpression of UCH-L1 by human cones, extending from theirouter segments to pedicles. Also, we illustrate the presence ofUCH-L1 in horizontal cells of the human (Fig. 5B, arrow) andmacaque (Fig. 6B, arrows) retinas, as well as by bipolar (Figs. 5Cand 6B, arrowheads) and amacrine cells (Figs. 5B and 6B) ofthese primates. Prominent expression of UCH-L1 by neuronsin the GCL is also shown for the human (Fig. 5D) and monkey(Fig. 6B). In order to gain insight into the particular neuronalsubtypes containing UCH-L1 in the primate retina, we carriedout double immunostaining experiments with antibodies toUCH-L1 and to several molecular markers of subsets of retinalneurons, namely calbindin (CB), parvalbumin (PV), tyrosinehydroxylase (TH) and calretinin (CR). A double labeling for CBrevealed expression of UCH-L1 by H2 horizontal cells and byDB3 and DB5 diffuse cone bipolar cells (Fig. 6F, arrowheads),which are the specific neuronal subtypes known to be positivefor CB in the monkey retina INL (Hopkins and Boycott 1995;Wässle et al., 2000). All CB-positive horizontal cells (i.e. of theH2 subtype) contained UCH-L1 (Fig. 6F, arrows), although notall UCH-L1-expressing horizontal neuronswere stained for CB.Thus, we next carried out double labelings for UCH-L1 and PV,a marker for both H1 and H2 horizontal cell subtypes in theprimate retina (Röhrenbeck et al., 1989). As shown, allhorizontal cells in the human (Fig. 5C) and monkey (Fig. 6G)retinas expressed UCH-L1. Regarding cone bipolar cells,Fig. 6C–E shows in detail a UCH-L1-immunoreactive conebipolar of the DB3 subtype, whose axon terminal lies in the IPLstratum S3 in the monkey retina (Grünert et al., 1994; Hopkinsand Boycott, 1995). Among amacrine cells immunopositive forUCH-L1 some of them displayed an intense labeling, whileothers showed a weak expression (Fig. 6A and B). Given thatthe monkey IPL S1 exhibited a high immunoreactivity (Fig. 6Aand B), we next addressed the possibility that dopaminergiccells were among the amacrine subtypes expressing UCH-L1.This followed the observation that some of the amacrine cells

Fig. 2 – Parkin immunolocalization pattern in the mammalian retina. Retinal sections from (A) mouse, (C) rat, (E) bovine, and(G) human retinas immunolabeled for parkin are shown. The corresponding immunostaining controls lacking primaryantibody are presented in panels B, D, F and H. Arrows point to amacrine cells. 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 = 20 μm.

73B R A I N R E S E A R C H 1 3 5 2 ( 2 0 1 0 ) 7 0 – 8 2

intensely labeled resembled morphologically dopaminergicneurons, together with the fact that the dendritic plexusformed by this cell subtype is well known to extend along thestratum S1 (Nguyen-Legros et al, 1984; Cuenca et al., 2005a).Therefore, we carried out double immunostaining experi-ments using antibodies to UCH-L1 and to TH, the prototypicaldopaminergic-cell marker. As clearly seen in Fig. 7A, thesecells were found to exhibit UCH-L1 in their cytoplasm, as wellas in their dendritic plexus stratifying in the IPL S1 (Fig. 7C–E,arrowheads). In addition to the dopaminergic plexus, otherUCH-L1-positive dendritic processes were present in thisstratum (Fig. 7B–E, arrows). It was also noticed that somedendrites immunoreactive for both TH and UCH-L1 werelocated in the S1 plexus at the base of other amacrine cellbodies either displaying low UCH-L1 levels (Fig. 7B, arrow-heads) or that did not show any expression of this protein(Fig. 7B, double arrowheads). In order to investigate theseparticular amacrine subtypes, double immunolabeling wascarried out with antibodies to UCH-L1 and CR, a specificmarker for AII amacrine cells in the monkey (Kolb et al., 2002).

From Fig. 7F–H it was concluded that AII cells were UCH-L1-negative. Similarly, a double labeling for PV, which has beenrecently reported to be a marker for the knotty type 2 ofglycinergic amacrines in the monkey (Klump et al., 2009),revealed that these cells lacked as well UCH-L1 expression(Fig. 7I–K). Finally, most of monkey retinal cells in the GCL,bearing somata of very different sizes, were found to containUCH-L1 in both their cell bodies and dendrites, theseramifying in all strata of the IPL (Fig. 7A and B). Furthermore,and consistently to other species studied in this work, humanand monkey ganglion-cell axons constituting the NFL alsoexhibited high levels of UCH-L1, as clearly observed in sectionsobtained both longitudinally (Figs. 5E and 6H) and transver-sally (Fig. 6I) with respect to axons in this layer.

3. Discussion

The UPS is crucial for both normal CNS function and neuronalprotection under stress conditions. Its impairment thus leads

Fig. 3 – Expression of the UCH-L1 gene in the retina ofmammals. A: RT-PCR analysis of UCHL1 mRNA in the retinafrom the indicated species (Rat, Sprague–Dawley). Theamplified bands were verified by DNA sequencing and theirmolecular sizes are given to the right. B: Immunoblottinganalysis of UCH-L1 protein. The arrowhead points to the 26kDa UCH-L1 band.

74 B R A I N R E S E A R C H 1 3 5 2 ( 2 0 1 0 ) 7 0 – 8 2

to neurodegeneration associated with a number of hereditaryand idiopathic diseases, mainly attributable to the toxic build-up of aggregates of proteins that are not effectively cleared bythe UPS, and derived mitochondrial dysfunction (Henchcliffeand Beal 2008; Neutzner et al., 2008; Tan et al., 2009). Parkinand UCH-L1 are two relevant components of the Ub-protea-somal pathway that have been related not only to Parkinson,but also to Alzheimer, Huntington and Lewy body diseases(Barrachina et al., 2006; Gong and Leznik, 2007; Humbert et al.,2007; Biskup et al., 2008). In this work we have undertaken theanalysis of parkin and UCH-L1 expression in the retina ofdifferent mammals.

Parkin has been reported to be present throughout thebrain, including the midbrain, basal ganglia, cerebral cortexand cerebellum (Horowitz et al., 1999; Shimura et al., 1999;Stichel et al., 2000; Zarate-Lagunes et al., 2001). Our resultsdemonstrate that parkin exists as well in the retina, where wehave found the PARK2 gene and its encoded protein to be

Fig. 4 – UCH-L1 immunolocalization pattern in the retina of mam(D) rabbit, (E) bovine, and (F) human retinas immunolabeled for Uhorizontal cell processes in the OPL, and in panels C-F to IPL stranuclear layer; OPL, outer plexiform layer; INL, inner nuclear layer;fiber layer. Scale bars=20 μm.

expressed in all of mammalian species studied. The 52 kDaisoform we have detected on retinal immunoblots is consis-tent with the molecular mass of native parkin encoded by thePARK2 canonical transcript, predicted to be 51.6 kDa, and withprevious reports in the mammalian brain (Horowitz et al.,1999; Stichel et al., 2000; Pawlyk et al., 2003). However, we havealso found two smaller isoforms, of 48 and 37 kDa, which havebeen previously reported in the brain (Horowitz et al., 1999;Stichel et al., 2000; Pawlyk et al., 2003) and are either productsof post-translational proteolytic cleavage or, more likely,encoded by PARK2 transcripts lacking specific exon(s). Exten-sive alternative splicing of the PARK2 gene has been reportedin the human and rodent brains (Stichel et al., 2000; D'Agataand Cavallaro 2004; Humbert et al., 2007), and we haveobtained evidence that this is also the case for themammalianretina (Campello, Esteve-Rudd, Cuenca and Martín-Nieto,unpublished). By immunohistochemical methods we havefound parkin to localize in all main retinal neuronal types,including photoreceptors. Hence the distribution of thisprotein in the retina is quite widespread, as it occurs in thebrain.

UCH-L1 is a protein especially abundant in neurons andneuroendocrine cells (Doran et al., 1983; Trowern et al., 1996;Saigoh et al., 1999). In this work we have found expression ofthe UCH-L1 gene at bothmRNA and protein levels in the retinaof all mammals studied. This protein migrated with anapparent molecular mass of 26 kDa, consistent with thatreported in the mammalian brain (Doran et al., 1983; Trowernet al., 1996; Saigoh et al., 1999; Barrachina et al., 2006). Ouranalysis of its immunolocalization pattern in the retinarevealed that UCH-L1 distributed through horizontal, bipolarand amacrine cells in the INL, as well as ganglion cells. Thisprotein was also present through the entire length of conephotoreceptors in all mammals analyzed with the exceptionof rodents, these including both nocturnal (mouse and rat) anddiurnal (squirrel) vision animals, whose cones were found tolack UCH-L1. In contrast, this protein was absent from rods inall mammalian species. Using the same polyclonal antibodiesto UCH-L1, Bonfanti et al. (1992) had reported this protein to berestricted to horizontal and ganglion cells in a number ofmammals by means of avidin–biotin–peroxidase (ABC) immu-nostaining, whereas we have identified additional UCH-L1-positive neuronal types in the retina, i.e. cones, bipolar andamacrine cells. It seems likely that the observed differences instaining patterns result from variations in the labelingmethodand/or fixation and processing of the retinas, i.e. paraffin-embedded tissue in their study vs. cryosections in the presentwork. Further analysis led us to conclude that all horizontalcells contained UCH-L1 in the human and monkey retinas, inkeeping with previous results in the rabbit (Reichenbach et al.,1994), and hence this might be a feature of all mammals. Also,among bipolar cells two CB-positive subtypes, namely DB3and DB5, were identified to express UCH-L1 in the monkey

mals. Retinal sections from (A) mouse, (B) rat, (C) squirrel,CH-L1 are shown. Arrowheads in panels A and B point tota S1, S3 (S2/S3 in C) and S5. OS, outer segments; ONL, outerIPL, inner plexiform layer; GCL, ganglion cell layer; NFL, nerve

75B R A I N R E S E A R C H 1 3 5 2 ( 2 0 1 0 ) 7 0 – 8 2

Fig. 5 – UCH-L1 immunolocalization pattern in the human retina. A: Detail of human cones immunolabeled for UCH-L1. B: Ahorizontal cell (arrow) and an amacrine cell in the INL immunopositive for UCH-L1 are shown at highmagnification. C: A doubleimmunostaining for UCH-L1 and parvalbumin (PV) colabels all horizontal cell subtypes (arrows) in the human retina. A bipolarcell is indicated by an arrowhead. D, E: Detail of the innermost retina sectioned longitudinally to optic nerve fibers (E, arrow) tohighlight UCH-L1-positive ganglion cell bodies and their axons. Arrowheads point to IPL strata S1, S3 and S5 in E. Retinal layerabbreviations are as in the legend of Fig. 4. Scale bars=10 μm.

76 B R A I N R E S E A R C H 1 3 5 2 ( 2 0 1 0 ) 7 0 – 8 2

together with some subtypes of amacrines, these includingdopaminergic cells. The latter finding is interesting, asdopaminergic neurons modulate the balance of center-surround organization of foveal ganglion cells and therebyspatial frequency tuning, which is greatly impaired inparkinsonian subjects (Bodis-Wollner and Tzelepi 1998;Bodis-Wollner, 2009). Also, dopaminergic amacrines arecrucial in the modulation of the scotopic visual pathway,which is severely affected together with foveal vision inParkinson disease patients and animal models (Wink andHarris, 2000; Witkovsky, 2004; Archibald et al., 2009; Bodis-Wollner, 2009), and have been described by us to undergodegeneration in the peripheral retina of parkinsonianmonkeys (Cuenca et al., 2005a). It is also important to note

that a significant thinning of the INL has been recentlyreported in the fovea of parkinsonian humans (Hajee et al.,2009). Finally, our staining found in the IPL of all speciesstudied is consistent with the previous localization of UCH-L1in axon terminals in the brain (Saigoh et al., 1999; Choi et al.,2004) and its proposed roles in synaptic function and plasticity(Gong and Leznik, 2007; Biskup et al., 2008). In this context, wehad as well observed a derangement of synaptic connectivityin this layer mediated by AII amacrines and rod bipolar cells(Cuenca et al., 2005a). However, it must be emphasized thatUCH-L1 deficiency cannot be regarded as the sole or mainmechanism accounting for neuronal degeneration in theparkinsonian retina, since for instance AII cells have beenfound in this work to lack UCH-L1.

Fig. 6 – UCH-L1 immunolocalization pattern in the monkey retina. A: Retinal section from the macaque retina immunolabeledfor UCH-L1. Arrowheads point to the IPL strata S1, S3 and S5. B: Higher magnification of the UCH-L1-immunostained monkeyretina, showing layers from the OPL to GCL. Arrows point to horizontal cells and arrowheads to bipolar cells. C-E: A detail of aDB3 cone bipolar cell doubly immunolabeled for UCH-L1 and calbindin (CB) is shown in C and D, respectively, and the mergedimage in E. F: A double immunolabeling for UCH-L1 and CB reveals H2 horizontal cells (arrows) and certain subtypes of conebipolars (arrowheads) colocalizing both proteins whose cell bodies reside in the outermost INL sublayer. G: Doubleimmunostaining for UCH-L1 and parvalbumin (PV) colabels all horizontal cell subtypes in the monkey retina. H, I: Detail of theinnermost retina sectioned longitudinally (H) or transversally (I) to optic nerve fibers to highlight UCH-L1-positive ganglion cellbodies and their axons. Retinal layer abbreviations are as in the legend of Fig. 4. Scale bars, A, B, F–I=20 μm; C–E=5 μm.

77B R A I N R E S E A R C H 1 3 5 2 ( 2 0 1 0 ) 7 0 – 8 2

Fig. 7 – UCH-L1 immunolocalization in amacrine cells of themonkey retina. A, B: A double immunolabeling for UCH-L1 (green) andtyrosine hydroxylase (TH; red) reveals dopaminergic cells in the IPL stratum S1 (labeled in yellow), as well as TH-negative,non-dopaminergic amacrines (in green). Nuclei were stained with TO-PRO-3 (blue). In panel B synaptic contacts made byTH-positive dendrites with a weakly UCH-L1-immunoreactive cell body are pointed by arrowheads, and with a UCH-L1-negativecell body bydouble arrowheads. Arrows point to non-dopaminergic dendrites. C–E: Double immunolabeling for UCH-L1 (C) and TH(D) reveals theUCH-L1-positive dopaminergic plexus (arrowheads) in themonkey retina (yellow in themerged image in E). Arrowspoint to non-dopaminergic dendrites. F–H: Double immunolabeling for UCH-L1 and calretinin (CR) in panels F and G, respectively(merged image in H), reveals the lack of UCH-L1 expression by AII amacrines. I-K: Double immunolabeling for UCH-L1 (I) andparvalbumin (PV) (J) reveals no colocalization inmonkey amacrine cells, although a clear coexpression is detected in ganglion cells(K). Retinal layer abbreviations are as in the legend of Fig. 4. Scale bars, A, B=20 μm; C–K=10 μm.

78 B R A I N R E S E A R C H 1 3 5 2 ( 2 0 1 0 ) 7 0 – 8 2

The retina in general, and photoreceptors in particular,appear to have very active ubiquitination systems, includingE1, E2 and E3 ubiquitinating enzymes and Ub C-terminal

hydrolases (Shang and Taylor, 2004). Thus, the majority of Ubin the retina is covalently bound to proteins, and retinalextracts are capable of Ub addition to substrate proteins and of

79B R A I N R E S E A R C H 1 3 5 2 ( 2 0 1 0 ) 7 0 – 8 2

disassembly and degradation of Ub-protein conjugates in vitro(Naash et al., 1991). In bovine rod outer segments both free Uband Ub conjugated to proteins, including rhodopsin andtransducin, have been identified which are thereafter degrad-ed (Obin et al., 1996, 2002). Interestingly, human cone innersegments are rich in Ub, a subcellular location where we havepresently found parkin to be coexpressed with UCH-L1. Theseresults are in keeping with the proposed role of the UPS inmodulating the levels of proteins responsible for phototrans-duction (Obin et al., 1996), a process in which parkin and UCH-L1 could participate. Our data are in line aswellwith thenotionthat these two enzymes could be important not only inphotoreceptors, but also in ganglion cells and other neuronaltypes in the INL where they localize, including horizontal,bipolar and amacrine cells. Precisely, upon light-induced(Naash et al., 1997) or ischemic (Harada et al., 2004) retinalinjury a strong increase of Ub and Ub-protein conjugatesoccurs in the INL and GCL of rodents. The neuroprotective roleof the UPS against retinal disease is supported by the up-regulation of Ub expression found in the rat retina uponelevated intraocular pressure (Dibas et al., 2008) and in adiabetic retinopathymousemodel (Adachi-Uehara et al., 2006).

The ample distribution of parkin and UCH-L1 we havequalitatively found in the mammalian retina is in line withtheir extensive expression and high levels in CNS neurons,and supportive of a relevant role of these proteins within theUb-proteasomal pathway in many cell types of the mamma-lian retina, including cones. A role for parkin is recognized inthe regulation of normal mitochondrial respiratory function(Palacino et al., 2004; Periquet et al., 2005) and prevention ofoxidative and a variety of other stresses in neurons (Hench-chcliffe and Beal 2008; Neutzner et al., 2008; Tan et al., 2009),while UCH-L1 appears involved in a general neuroprotectiveresponse to protein aggregation occurring in neurodegenera-tive diseases (Lansbury, 2006; Gong and Leznik, 2007). There-fore, their potential neuroprotection functions in the retinaconstitute an issue deserving further investigation. Neverthe-less, not all retinal neurons were found to contain parkin and/or UCH-L1, and thus differences in the UPS components andprotein degradation mechanisms must exist among retinalcell types (e.g. cones versus rods) and subtypes. In this context,parkin constitutes the third E3 Ub ligase identified in theretina up to date, following Ubr1, recently involved in UPS-dependent rhodopsin degradation in a retinal inflammationmouse model (Ozawa et al., 2008), and TRIM2, whosedeficiency triggers neurodegeneration in mice (Balastik et al.,2008). A UCH-L1 homolog, the ubiquitous Ub C-terminalhydrolase UCH-L3, is present aswell in the retina and enrichedin the inner segments of mouse photoreceptors (Sano et al.,2006). Finally, the widespread localization of parkin and UCH-L1 in the retina, together with the importance of the UPS inthis tissue (Shang and Taylor, 2004), makes envisionable thattheir altered expression or dysfunction is associated withretinal impairments. Consequently, it should be of interest toscreen their encoding genes for mutations or alterations intheir expression levels in patients with hereditary or sporadicretinal neurodegenerations. In addition, therapies based onthe use of neuroprotective agents currently tested in the braincould be taken into consideration for the treatment of theseretinal diseases.

4. Experimental procedures

4.1. Biological material

Mammalian species studied in this work were: mouse (Musmusculus, C57BL/6J), rat (Rattus norvegicus, Long Evans unlessotherwise stated), California ground squirrel (Spermophilusbeecheyi), European wild rabbit (Oryctolagus cuniculus), cow (Bostaurus), cynomolgus monkey (Macaca fascicularis) and human(Homo sapiens). Mice and rats were held at the Universidad deAlicante animal-care facility. Monkeys were kept as described(Martínez-Navarrete et al., 2007) at the Universidad de Murcia,Spain, in accordance with guidelines given by the InternationalPrimate Society. All animal handling was carried out incompliance with rules set by the National Institutes of Healthand the European Directive 86/609/EEC. Animals were anesthe-sized and euthanized according to well-standardized, interna-tionally-approved protocols, and eyeballs were immediatelyenucleated followingsacrifice. Squirrel fixedeyeswereprovidedby Dr. Kenneth A. Linberg. Rabbit and bovine eyes were freshlyobtained from local farms and slaughterhouses, respectively.Post-mortemeyes fromover 45-year old humanswere providedby the Utah Lions Eye Bank. For RNA and protein extraction,mouse (4–6months old) and rat (3–4 monthsold) eyeswerekeptin RNAlater solution from Ambion (Austin, TX, USA) at −20 °Cbefore processing. Bovine (8–9 months old) and monkey (5–6years old) eyes were snap-frozen in liquid N2 and immediatelystored at −80 °C without fixation. For immunohistochemistry,eyes were fixed in 4% paraformaldehyde and then subjected tosucrose cryoprotection (Cuenca et al., 2005b) before storage at−80 °C.

4.2. RT-PCR

Eyes were thawed and the neural retina was dissected free fromretinal pigment epithelium and other ocular tissues. RNA wasextracted using the RNAqueous-4PCR kit (Ambion) and furthertreated with DNase I following the manufacturers’ instructions.RNA from human retina was purchased from Clontech (Moun-tain View, CA, USA). Reverse transcription to cDNA was carriedout using the RETROscript kit (Ambion), and thereafter PCRamplification was performed essentially as described (Martínez-Navarrete et al., 2007) using the PARK2 or UCHL1 primerssummarized in Table 1. PCR reactions (50 μl) contained 10 ng ofcDNA, forward and reverse primers (0.4 μMeach), the four dNTPs(0.2 mM each) and 1 U of GoTaq DNA polymerase (Promega,Madison, WI, USA). After an initial denaturation step at 95 °C for3 min, amplificationwasperformed for 35 cycles eachof 95 °C for10 s, 60 °C for 20 s and 72 °C for 30 s, and concluded with a finalelongation at 72 °C for 2 min. PCR products were run on 2%agarose gels stained with SYBR Green I (Sigma, St. Louis, MO,USA), and were verified by automated DNA sequencing in allcases. A Macaca fascicularis PARK2 cDNA covering the wholeparkin coding sequence was amplified and cloned from neuralretina mRNA using the oligonucleotides 5'-CCACCTACCCAGT-GACCATG-3' and 5'-GGGTATGCTCCCCCAGGATG-3' as forwardand reverse primers (aligning with exons 1 and 12, respectively),and its sequence was submitted to GenBank under accessionnumber FJ531869.

Table 1 – Primers used for RT-PCR in this work.

Gene Species Accession number Primer sequences Exon

PARK2 Mouse NM_016694 Fw: 5’-AACTGTGACCTGGAACAACA-3' 3Rat NM_020093 Rv: 5'-CTGGGYCAAGGTGAGCGTTG-3' 4

Bovine XM_609337 Fw: 5'-MRYTGTGACCTGGATCAGCA-3' 3Monkey FJ531869 Rv: 5'-CTGGGYCAAGGTGAGCGTTG-3' 4Human NM_004562

UCHL1 Mouse NM_011670 Fw: 5'-CCAAGTGTTTCGAGAAGAACG-3' 5Rat NM_017237 Rv: 5'-GCTAAAGCTGCAAACCAAGG-3' 9

Bovine NM_001046172 Fw: 5'-GATGTTCTGGGACTGGAGGA-3' 3Rv: 5'-CCATCCACGTTGTTAAACAGAA-3' 7

Monkey AB125199 Fw: 5'-CTGTGGCACAATCGGACTTA-3' 4Human NM_004181 Rv: 5'-CCATCCACGTTGTTAAACAGAA-3' 7

Forward (Fw) and reverse (Rv) primers were designed following our previous criteria (Martínez-Navarrete et al., 2007), based on cDNA sequencesdeposited inGenBankunder the indicated accessionnumbers.An “R” symbol indegenerate primer sequencesdenotesAorG, and “Y”denotesCorT.The exon to which each primer aligns is indicated in the last column.

80 B R A I N R E S E A R C H 1 3 5 2 ( 2 0 1 0 ) 7 0 – 8 2

4.3. Immunoblotting

Retinal proteins were extracted and subjected to immuno-blotting analysis as previously described (Martínez-Navarreteet al., 2007). Proteins (50 μg/lane) were resolved by SDS-PAGEon 5–20% polyacrylamide-gradient gels and electrotransferredto Hybond-P PVDF membranes (GE Healthcare, Buckingham-shire, UK). These were probed at 4 °C overnight with goatpolyclonal antibodies to human parkin (N-terminus) fromSanta Cruz Biotechnology (Santa Cruz, CA, USA; Catalog No.sc-13279) at a 1:500 dilution in 25 mM Tris (pH 8.0), 150 mMNaCl, 2.7 mM KCl (TBS), or with rabbit polyclonal antibodies tofull-length human PGP9.5 (Ultraclone, Isle of Wight, UK; Cat.No. RA95101) at a 1:5,000 dilution. Thereafter, they wereincubated at room temperature for 1 h with horseradishperoxidase-conjugated rabbit anti-goat (Sigma) or goat anti-rabbit (Pierce, Rockford, IL, USA) IgG at a 1:10,000 dilution.Detection was performed by enhanced chemiluminescenceusing the SuperSignal West Dura system (Pierce).

4.4. Immunohistochemistry

Cryostat vertical 16 μm retinal sections were obtained andprocessed for immunohistochemistry following well estab-lished procedures (Cuenca et al., 2005a, 2005b; Martínez-Navarrete et al., 2007, 2008). They were immunostained atroom temperature overnight with goat polyclonal antibodies tohuman parkin N-terminus from Santa Cruz Biotechnology (Cat.No. sc-13279) at a 1:50 dilution in 0.1 mM sodium phosphatebuffer (pH 7.4), 0.5% Triton X-100 (PBX), or to human parkinamino acids 83–97 fromChemicon (Temecula, CA, USA; Cat. No.AB5240) at a 1:500–1:1,000 dilution. Alternatively, slides wereprobedwith rabbit antibodies to human PGP9.5 (Ultraclone; Cat.No. RA95101) at a 1:5,000 dilution, alone or in combination withmousemonoclonal antibodies to calbindin D-28K (Sigma; cloneCB-955, Cat. No. C9848) at a 1:500 dilution, to parvalbumin(Sigma; clone PARV19, Cat. No. P3088) at a 1:300 dilution, or totyrosinehydroxylase (SantaCruzBiotechnology; clone F-11, Cat.No. sc-25269) at a 1:1,000 dilution, or else with goat polyclonalantibodies to calretinin (Swant, Bellinzona, Switzerland;Cat.No.CG1) at a 1:500 dilution. Control slides in which primaryantibodies were omitted were processed in parallel, with no

immunoreactivity found inany case. Subsequently, thesectionswere incubated at room temperature for 1 h with donkeysecondary antibodies conjugated to Alexa Fluor 488 or 546 fromMolecular Probes (Eugene, OR, USA) at a 1:100 dilution. Thenuclear dye TO-PRO-3 iodide (Molecular Probes) was added at 1μMsimultaneously to secondary antibodies. Imageswere finallyobtained under a Leica (Wetzlar, Germany) TCS SP2 confocallaser-scanning microscope.

Acknowledgments

We thank Dr. Kenneth A. Linberg (University of California atSanta Barbara, CA, USA) and the Utah Lions Eye Bank(University of Utah, Salt Lake City, UT, USA) for providingthe squirrel and human eyes, respectively, used in this study.This research was supported by grants from the SpanishInstituto de Salud Carlos III (ISCIII; PI09/1623) to J.M.-N.,Ministerio de Ciencia e Innovación (MICINN; SAF2007-62262),Fundación Séneca (FS/05662/PI/07) and Centro de InvestigaciónBiomédica en Red sobre Enfermedades Neurodegenerativas(CIBERNED) to M.-T.H.; and MICINN (BFU2009-07793/BFI),ISCIII (RETICS RD07/0062/0012), Generalitat Valenciana(ACOMP/2009/139), Organización Nacional de Ciegos de España(ONCE) and Fundación Lucha contra la Ceguera (FUNDALUCE) toN.C. J.E.-R. and L.C. are recipients of predoctoral contractsfrom the Universidad de Alicante.

R E F E R E N C E S

Adachi-Uehara, N., Kato, M., Nimura, Y., Seki, N., Ishihara, A.,Matsumoto, E., Iwase, K., Ohtsuka, S., Kodama, H., Mizota, A.,Yamamoto, S., Adachi-Usami, E., Takiguchi, M., 2006.Up-regulation of genes for oxidative phosphorylation andprotein turnover in diabetic mouse retina. Exp. Eye Res. 83,849–857.

Archibald, N.K., Clarke, M.P., Mosimann, U.P., Burn, D.J., 2009. Theretina in Parkinson's disease. Brain 132, 1128–1145.

Balastik, M., Ferraguti, F., Pires-da Silva, A., Lee, T.H.,Alvarez-Bolado, G., Lu, K.P., Gruss, P., 2008. Deficiency inubiquitin ligase TRIM2 causes accumulation of neurofilament

81B R A I N R E S E A R C H 1 3 5 2 ( 2 0 1 0 ) 7 0 – 8 2

light chain and neurodegeneration. Proc. Natl Acad. Sci. USA19, 12016–12021.

Barrachina, M., Castaño, E., Dalfó, E., Mães, T., Buesa, C., Ferrer, I.,2006. Reduced ubiquitin C-terminal hydrolase-1 expressionlevels in dementia with Lewy bodies. Neurobiol. Dis. 22,265–273.

Biskup, S., Gerlach, M., Kupsch, A., Reichmann, H., Riederer, P.,Vieregge, P., Wüllner, U., Gasser, T., 2008. Genes associatedwith Parkinson syndrome. J. Neurol. 255, 8–17.

Bodis-Wollner, I., 2009. Retinopathy in Parkinson disease. J. NeuralTransm. 116, 1493–1501.

Bodis-Wollner, I., Tzelepi, A., 1998. The push-pull action ofdopamine on spatial tuning of themonkey retina: the effects ofdopaminergic deficiency and selective D1 and D2 receptorligands on the pattern electroretinogram. Vision Res. 38,1479–1487.

Bonfanti, L., Candeo, P., Piccinini, M., Carmignoto, G., Comelli,M.C., Ghidella, S., Bruno, R., Gobetto, A., Merighi, A., 1992.Distribution of protein gene product 9.5 (PGP 9.5) inthe vertebrate retina: evidence that immunoreactivity isrestricted to mammalian horizontal and ganglion cells. J.Comp. Neurol. 322, 35–44.

Choi, J., Levey, A.I.,Weintraub, S.T., Rees, H.D., Gearing, M., Chin, L.-S.,Li, L., 2004. Oxidative modifications and down-regulation ofubiquitin carboxyl-terminal hydrolase L1 associated withidiopathic Parkinson's and Alzheimer's disease. J. Biol. Chem. 279,13256–13264.

Cuenca, N., Herrero, M.-T., Angulo, A., De Juan, E.,Martínez-Navarrete, G.C., López, S., Barcia, C., Martín-Nieto, J.,2005a. Morphological impairments in retinal neurons of thescotopic visual pathway in a monkey model of Parkinson'sdisease. J. Comp. Neurol. 493, 261–273.

Cuenca, N., Pinilla, I., Sauvé, Y., Lund, R., 2005b. Early changesin synaptic connectivity following progressivephotoreceptor degeneration in RCS rats. Eur. J. Neurosci. 22,1057–1072.

D'Agata, V., Cavallaro, S., 2004. Parkin transcript variants in ratand human brain. Neurochem. Res. 29, 1715–1724.

Dev, K.K., Van der Putten, H., Sommer, B., Rovelli, G., 2003. Part I:Parkin-associated proteins and Parkinson's disease.Neuropharmacology 45, 1–13.

DiAntonio, A., Hicke, L., 2004. Ubiquitin-dependent regulation ofthe synapse. Annu. Rev. Neurosci. 27, 223–246.

Dibas, A., Yang, M.H., He, S., Bobich, J., Yorio, T., 2008. Changes inocular aquaporin-4 (AQP4) expression following retinal injury.Mol. Vis. 14, 1770–1783.

Doran, J.F., Jackson, P., Kynoch, P.A., Thompson, R.J., 1983.Isolation of PGP 9.5, a new human neurone-specific proteindetected by high-resolution two-dimensional electrophoresis.J. Neurochem. 40, 1542–1547.

Fischer, J.A., 2003. Deubiquitinating enzymes: their roles indevelopment, differentiation, and disease. Int. Rev. Cytol. 229,43–72.

Glickman, M.H., Raveh, D., 2005. Proteasome plasticity. FEBS Lett.579, 3214–3223.

Gong, B., Leznik, E., 2007. The role of ubiquitin C-terminalhydrolase L1 in neurodegenerative disorders. Drug NewsPerspect. 20, 365–370.

Grünert, U., Martin, P.R., Wässle, H., 1994. Immunocytochemicalanalysis of bipolar cells in the macaque monkey retina. J.Comp. Neurol. 348, 607–627.

Hajee, M.E., March, W.F., Lazzaro, D.R., Wolintz, A.H., Shrier, E.M.,Glazman, S., Bodis-Wollner, I.G., 2009. Inner retinal layerthinning in Parkinson disease. Arch. Ophthalmol. 127, 737–741.

Harada, T., Harada, C., Wang, Y.L., Osaka, H., Amanai, K., Tanaka,K., Takizawa, S., Setsuie, R., Sakurai, M., Sato, Y., Noda, M.,Wada, K., 2004. Role of ubiquitin carboxy terminal hydrolase-L1in neural cell apoptosis induced by ischemic retinal injury invivo. Am. J. Pathol. 164, 59–64.

Henchcliffe, C., Beal, M.F., 2008. Mitochondrial biology andoxidative stress in Parkinson disease pathogenesis. Nat. Clin.Pract. Neurol. 4, 600–609.

Hershko, A., Ciechanover, A., Varshavsky, A., 2000. Basic MedicalResearch Award. The ubiquitin system. Nat. Med. 6,1073–1081.

Hopkins, J.M., Boycott, B.B., 1995. Synapses between cones anddiffuse bipolar cells of a primate retina. J. Neurocytol. 24,680–694.

Horowitz, J.M., Myers, J., Stachowiak, M.K., Torres, G., 1999.Identificationanddistributionof Parkin in rat brain.NeuroReport10, 3393–3397.

Humbert, J., Beyer, K., Carrato, C., Mate, J.L., Ferrer, I., Ariza, A.,2007. Parkin and synphilin-1 isoform expression changes inLewy body diseases. Neurobiol. Dis. 26, 681–687.

Klump, K.E., Zhang, A.J., Wu, S.M., Marshak, D.W., 2009.Parvalbumin-immunoreactive amacrine cells of macaqueretina. Vis. Neurosci. 13, 1–10.

Kolb, H., Zhang, L., Dekorver, L., Cuenca, N., 2002. A new look atcalretinin-immunoreactive amacrine cell types in the monkeyretina. J. Comp. Neurol. 453, 168–184.

Lansbury Jr., P.T., 2006. Improving synaptic function in a mousemodel of AD. Cell 126, 655–657.

Leroy, E., Boyer, R., Auburger, G., Leube, B., Ulm, G., Mezey, E.,Harta, G., Brownstein, M.J., Jonnalagada, S., Chernova, T.,Dehejia, A., Lavedan, C., Gasser, T., Steinbach, P.J., Wilkinson,K.D., Polymeropoulos, M.H., 1998. The ubiquitin pathway inParkinson's disease. Nature 395, 451–452.

Liu, Y., Fallon, L., Lashuel, H.A., Liu, Z., Lansbury Jr., P.T., 2002. TheUCH-L1 gene encodes two opposing enzymatic activities thataffect α-synuclein degradation and Parkinson's diseasesusceptibility. Cell 111, 209–218.

Martínez-Navarrete, G.C., Martín-Nieto, J., Esteve-Rudd, J., Angulo, A.,Cuenca, N., 2007. α-Synuclein gene expression profile in the retinaof vertebrates. Mol. Vis. 13, 949–961.

Martínez-Navarrete,G.C.,Angulo,A.,Martín-Nieto, J., Cuenca,N., 2008.Gradualmorphogenesisof retinalneurons in theperipheral retinalmargin of adult monkeys and humans. J. Comp. Neurol. 511,557–580.

Mendes, H.F., Van der Spuy, J., Chapple, J.P., Cheetham, M.E.,2005. Mechanisms of cell death in rhodopsin retinitispigmentosa: implications for therapy. Trends Mol. Med. 11,177–185.

Naash, M., Izbicka, E., Anderson, R.E., 1991. Rat retina has an activeand stable ubiquitin-protein conjugating system. J. Neurosci.Res. 30, 433–441.

Naash, M.I., Al-Ubaidi, M.R., Anderson, R.E., 1997. Light exposureinduces ubiquitin conjugation and degradation activities in therat retina. Invest. Ophthalmol. Vis. Sci. 38, 2344–2354.

Neutzner, A., Benard, G., Youle, R.J., Karbowski, M., 2008. Role ofthe ubiquitin conjugation system in the maintenance ofmitochondrial homeostasis. Ann. N.Y .Acad. Sci. 1147,242–253.

Nguyen-Legros, J., Botteri, C., Phuc, L.H., Vigny, A., Gay, M., 1984.Morphology of primate's dopaminergic amacrine cells asrevealed by TH-like immunoreactivity on retinal flat-mounts.Brain Res. 295, 145–153.

Obin, M.S., Jahngen-Hodge, J., Nowell, T., Taylor, A., 1996.Ubiquitinylation and ubiquitin-dependent proteolysis invertebrate photoreceptors (rod outer segments). Evidence forubiquitinylation of Gt and rhodopsin. J. Biol. Chem. 271,14473–14484.

Obin, M., Lee, B.Y., Meinke, G., Bohm, A., Lee, R.H., Gaudet, R.,Hopp, J.A., Arshavsky, V.Y., Willardson, B.M., Taylor, A.,2002. Ubiquitylation of the transducin βγ subunit complex.Regulation by phosducin. J. Biol. Chem. 277, 44566–44575.

Ozawa, Y., Nakao, K., Kurihara, T., Shimazaki, T., Shimmura, S.,Ishida, S., Yoshimura, A., Tsubota, K., Okano, H., 2008. Roles ofSTAT3/SOCS3 pathway in regulating the visual function and

82 B R A I N R E S E A R C H 1 3 5 2 ( 2 0 1 0 ) 7 0 – 8 2

ubiquitin-proteasome-dependent degradation of rhodopsinduring retinal inflammation. J. Biol. Chem. 283, 24561–24570.

Palacino, J.J., Sagi, D., Goldberg, M.S., Krauss, S., Motz, C.,Wacker,M.,Klose, J., Shen, J., 2004. Mitochondrial dysfunction and oxidativedamage in parkin-deficientmice. J. Biol. Chem. 279, 18614–18622.

Pawlyk, A.C., Giasson, B.I., Sampathu, D.M., Perez, F.A., Lim, K.L.,Dawson, V.L., Dawson, T.M., Palmiter, R.D., Trojanowski, J.Q.,Lee, V.M.-Y., 2003. Novel monoclonal antibodies demonstratebiochemical variation of brain parkin with age. J. Biol. Chem.278, 48120–48128.

Periquet, M., Corti, O., Jacquier, S., Brice, A., 2005. Proteomicanalysis of parkin knockout mice: alterations in energymetabolism, protein handling and synaptic function. J.Neurochem. 95, 1259–1276.

Reichenbach, A., Kasper, M., Schnitzer, J., Osborne, N.N.,Pritz-Hohmeier, S., 1994. A set of early-born neurons isdistinctly labeled by several defined antibodies in the adultrabbit retina. J. Hirnforsch. 35, 391–395.

Röhrenbeck, J., Wässle, H., Boycott, B.B., 1989. Horizontal cells inthe monkey retina: Immunocytochemical staining withantibodies against calcium binding proteins. Eur. J. Neurosci. 1,407–420.

Saigoh, K., Wang, Y.L., Suh, J.G., Yamanishi, T., Sakai, Y.,Kiyosawa, H., Harada, T., Ichihara, N., Wakana, S., Kikuchi,T., Wada, K., 1999. Intragenic deletion in the gene encodingubiquitin carboxy-terminal hydrolase in gad mice. Nat.Genet. 23, 47–51.

Sano, Y., Furuta, A., Setsuie, R., Kikuchi, H., Wang, Y.L., Sakurai, M.,Kwon, J.,Noda,M.,Wada,K., 2006. Photoreceptor cell apoptosis inthe retinal degeneration of Uchl3-deficient mice. Am. J. Pathol.169, 132–141.

Shang, F., Taylor, A., 2004. Function of the ubiquitin proteolyticpathway in the eye. Exp. Eye Res. 78, 1–14.

Shimura, H., Hattori, N., Kubo, S., Yoshikawa, M., Kitada, T.,Matsumine, H., Asakawa, S., Minoshima, S., Yamamura, Y.,Shimizu, N., Mizuno, Y., 1999. Immunohistochemical andsubcellular localization of Parkin protein: absence of proteinin autosomal recessive juvenile parkinsonism patients. Ann.Neurol. 45, 668–672.

Stichel, C.C., Augustin, M., Kühn, K., Zhu, X.R., Engels, P., Ullmer,C., Lübbert, H., 2000. Parkin expression in the adult mousebrain. Eur. J. Neurosci. 12, 4181–4194.

Tan, J.M.M., Wong, E.S.P., Lim, K.-L., 2009. Protein misfolding andaggregation in Parkinson's disease. Antiox. Redox Signal. 11,2119–2134.

Trowern, A.R., Laight, R.,MacLean, N., Mann, D.A., 1996. Detection ofneuron-specific protein gene product (PGP) 9.5 in the rat andzebrafish using anti-human PGP9.5 antibodies. Neurosci. Lett.210, 21–24.

Wässle, H., Dacey, D.M., Haun, T., Haverkamp, S., Grünert, U.,Boycott, B.B., 2000. The mosaic of horizontal cells in themacaque monkey retina: with a comment on biplexiformganglion cells. Vis. Neurosci. 17, 591–608.

Wink, B., Harris, J., 2000. A model of the Parkinsonian visual system:support for the dark adaptation hypothesis. Vision Res. 40,1937–1946.

Witkovsky, P., 2004. Dopamine and retinal function. Doc.Ophthalmol. 108, 17–40.

Zarate-Lagunes, M., Gu, W.J., Blanchard, V., Francois, C., Muriel, M.P., Mouatt-Prigent, A., Bonici, B., Parent, A., Hartmann, A.,Yelnik, J., Boehme, G.A., Pradier, L., Moussaoui, S., Faucheux, B.,Raisman-Vozari, R., Agid, Y., Brice, A., Hirsch, E.C., 2001. Parkinimmunoreactivity in the brain of human and non-humanprimates: An immunohistochemical analysis in normalconditions and in parkinsonian syndromes. J. Comp. Neurol.432, 184–196.