Pleiotrophin receptor RPTP-ζ/β expression is up-regulated by l-DOPA in striatal medium spiny neurons of parkinsonian rats

Authors


Address correspondence and reprint requests to Juan E. Ferrario, INSERM U679, Groupe Hospitalier Pitié-Salpêtrière, 47 Bd de l’Hôpital, Bâtiment Pharmacie, Paris, France. E-mail: ferrario@ccr.jussieu.fr

Abstract

l-DOPA is still the drug of choice to treat Parkinson’s disease although adverse side effects appear after several years of treatment. These are thought to be the consequence of plastic re-arrangements of the nigrostriatal connections, such as sprouting of the dopaminergic terminals or post-synaptic changes. Pleiotrophin, a trophic factor that we have shown to be up-regulated in the striatum of parkinsonian rats after long-term l-DOPA treatment may play a role in these plastic changes. To determine whether one of the three known pleiotrophin receptors [N-syndecan, receptor protein tyrosine phosphatase type zeta beta (RPTP-ζ/β) and anaplastic lymphoma kinase] might be implicated in these putative plastic effects, we quantified their expression levels by real-time RT-PCR in the striatum and mesencephalon of rats with partial lesions of the nigrostriatal pathway undergoing l-DOPA treatment. Both pleiotrophin and RPTP-ζ/β expression was up-regulated in the striatum but not in the mesencephalon of lesioned rats and RPTP-ζ/β expression was even further increased by l-DOPA. The levels of the RPTP-ζ/β protein were also increased in the striatum of l-DOPA-treated lesioned rats. Immunofluorescence labeling showed the protein to be constitutively expressed in striatal medium spiny neurons, which are innervated by both the corticostriatal glutamatergic and nigrostriatal dopaminergic systems. RPTP-ζ/β might therefore be implicated in the plastic changes triggered by l-DOPA treatment and might merit further study as a potential candidate for Parkinon’s disease therapy.

Abbreviations used
6-OHDA

6-hydroxydopamine

ALK

anaplastic lymphoma kinase

LID

l-DOPA-induced dyskinesias

MSNs

medium spiny neurons

PBS

phosphate-buffered saline

PBS-T

PBS with 0.1 M Triton X-100 (0.15%)

PD

Parkinson’s disease

PTN

pleiotrophin

qRT-PCR

quantitative RT-PCR

RPTP-ζ/β

receptor protein tyrosine phosphatase type zeta beta

SNpc

substantia nigra pars compacta

TH

tyrosine hydroxylase

Parkinson’s disease (PD) is a neurodegenerative disorder characterized by the progressive degeneration of the dopaminergic neurons of the substantia nigra pars compacta (SNpc) that results in the denervation of their striatal targets. l-DOPA is still the most effective pharmacological treatment for PD despite the appearance of adverse side effects such as l-DOPA-induced dyskinesias (LID) after several years of therapy (Hirsch 2000). Both the loss of the dopaminergic neurons and dopamine replacement therapy trigger a myriad of adaptive mechanisms in an attempt to maintain the function of the denervated striatal neurons (Hirsch 2000; Cenci 2007) such as spontaneous (Blanchard et al. 1996; Finkelstein et al. 2000) and l-DOPA-induced (Murer et al. 1998; Datla et al. 2001) recovery of dopaminergic nerve terminals together with altered gene expression, protein phosphorylation, and enhanced receptor signaling in the post-synaptic neurons (Cenci and Lundblad 2006).

We identified pleiotrophin (PTN), a novel cytokine, in an analysis of differential gene expression in the striatum of l-DOPA-treated rats with partial lesions of the nigrostriatal system, undertaken to identify the elements implicated in these adaptive process (Ferrario et al. 2004). PTN is a secreted heparin-binding protein, which is highly conserved and widely expressed during development both in CNS and PNS, where it plays a role in cell migration and replication, angiogenesis, neurite outgrowth, axon guidance, and synaptogenesis (Rauvala and Peng 1997; Deuel et al. 2002). In the adult brain, PTN is expressed in the hippocampus, cortex, and olfactory bulb (Wanaka et al. 1993; Lauri et al. 1996) as well as in striatal interneurons (Taravini et al. 2005). It is also up-regulated after brain lesions (Takeda et al. 1995; Poulsen et al. 2000), including lesions of the nigrostriatal system (Hida et al. 2003) and in the SNpc of patients with PD (Marchionini et al. 2007). In vitro studies show that PTN promotes the maturation of dopaminergic neurons in primary mesencephalic cultures (Mourlevat et al. 2005) and increases the differentiation of dopaminergic neurons from embryonic stem cells (Jung et al. 2004). In addition, it improves functional recovery of grafted dopaminergic neurons in the striatum of parkinsonian rats (Hida et al. 2007). PTN has three known membrane receptors: the heparin sulfate proteoglycan N-syndecan (Raulo et al. 1994), the receptor protein tyrosine phosphatase type zeta beta (RPTP-ζ/β) (Maeda et al. 1996), and the tyrosine kinase anaplastic lymphoma kinase (ALK) (Stoica et al. 2001).

To determine whether one of these molecules might be involved in the plastic changes induced during l-DOPA treatment in the lesioned brain, we quantified the expression of PTN and its receptors by real time RT-PCR [quantitative RT-PCR (qRT-PCR)] in the striatum and ventral mesencephalon of rats with 6-hydroxydopamine (6-OHDA)-induced partial lesions of the nigrostriatal dopaminergic pathway which were treated with either vehicle or l-DOPA for 21 days. We found that the mRNAs of PTN and RPTP-ζ/β were up-regulated in the striatum of the lesioned rats and that l-DOPA increased RPTP-ζ/β expression even further, but no changes in expression were detected in the ventral mesencephalon. Western blot and immunohistological analyses showed that the RPTP-ζ/β protein is also up-regulated in the striatum of l-DOPA-treated lesioned rats and is localized in both the dopaminergic neurons in the SNpc and in their targets, the medium spiny neurons (MSNs) in the striatum.

Materials and methods

Animals

Adult male Wistar rats (Centre d’Elevage René Janvier, Le Genest St Ile, France), weighing 250–300 g at the beginning of the experiment, were housed two or three per cage, with free access to food and water, under controlled temperature and a 12-h light/dark cycle. All animal procedures were carried out in compliance with the European Directive No.: 86/609/EEC and the guidelines of the local Institutional Animal Care and Use Committee. All efforts were made to minimize the number of animals used and their suffering.

Dopaminergic lesions and l-DOPA treatment

The nigrostriatal dopaminergic neurons were lesioned unilaterally by injections of 6-OHDA in the right striatum, as previously described (Ferrario et al. 2003; Debeir et al. 2005). Two independent pools of animals were produced for molecular (seven rats per group) and biochemical (six rats per group) analysis. Briefly, rats were anesthesized by injection of equithesin (3 mL/kg) and ketamine (0.2 mL/kg) before the stereotaxic injection of 20 μg of 6-OHDA (free base; Sigma-Aldrich, Saint-Quentin Fallavier, France) in 4 μL of vehicle (0.9% NaCl, 0.2% ascorbic acid), at a rate of 0.5 μL/min, at 0.2 mm anteroposterior, −3.4 mm lateral, and −5.4 mm ventral to bregma. The incisor bar was fixed at −3.3 mm, according to the rat brain atlas of Paxinos and Watson 1986. Control rats were sham-injected with 4 μL of vehicle. One week after surgery, animals were randomly assigned to receive l-DOPA or vehicle (NaCl, 0.9%) treatment. l-DOPA-methyl ester/benserazide (Sigma-Aldrich) was administered at doses of 40/10 mg/kg (i.p.), which is in the range of doses used in similar protocols (Cenci 2007), once a day around 6:00 pm for 3 weeks, as in a previous study (Ferrario et al. 2003), followed by 2-day washout. After a lethal injection of pentobarbital 60 mg/kg, (i.p.), the animals were perfused through the heart with 100 mL of NaCl (0.9%) to wash out blood components. The brains were then removed and the striatum and ventral mesencephalon were dissected and quickly frozen on dry ice until processed.

RNA extraction

Total RNA was extracted from striatum and ventral mesencephalon using the RNeasy Mini kit (Qiagen, Courtaboeuf, France), according to the manufacturer’s instructions. Residual DNA was digested during the process with DNase I (Qiagen). The RNA was eluted with 50 μL of water; the concentration was determined with a Nanodrop ND-1000 spectrophotometer (Nanodrop Technologies, Wilmington, DE, USA) and the quality was evaluated with an Agilent Bioanalyzer (Agilent Technologies, Santa Clara, CA, USA). The RNA used in our experiments had a RNA integrity number higher than 8 on a scale of 10. Each RNA sample was diluted once, aliquoted, and stored at −80°C until use.

Quantitative real-time RT-PCR

Quantitative RT-PCR was performed in one step using the QuantiTect SYBR Green Kit (Qiagen) by following the manufacturer’s instructions. The primers for quantifying the expression of the reference genes (ß-actin [QT00193473], hypoxanthine-guanine phosphoribosyl transferase (HPRT) [QT00199640], glyceraldehyde-3-phosphate dehydrogenase (GAPDH) [QT00199633], and β2-microglobuline [QT00176295]), genes reflecting the 6-OHDA lesions (cyclic AMP-related phosphoprotein 32 kDa; DARPP-32) [QT00428897] and tyrosine hydroxylase (TH; [QT00185024]), PTN and its receptors (PTN [QT00181447], RPTP-ζ/β [QT00182910], ALK [QT00446166], and N-Syndecan [QT00179774]) were also obtained from Qiagen (product reference is in brackets). Before the experiments, the amplification efficiency of each set of primers was tested with serial dilutions to construct a standard curve. The RNA samples were then tested in the concentration range where the slope of the standard curve was near −3.3 (we accepted −3.1 to −3.5, which represents a 100 ± 10% of efficiency). Each experiment was performed on 50 ng/5 μL of the total RNA in a final volume of 50 μL. Real time PCR amplification was performed on the ABI Prism 7700 sequence detection system (Applied Biosystems, Foster City, CA, USA). The PCR conditions were 30 min incubation at 50°C for the reverse transcription, 15 min at 95°C for activation of the polymerase, followed by 40 cycles of 15 s denaturation at 94°C, 30 s annealing at 55°C, and 30 s extension at 72°C. Specificity of the PCR product was confirmed by analysis of the dissociation curve. Each run included a negative control [No Template Control (NTC)] and a tube without reverse transcriptase (No RT control). All samples and controls for each experiment were performed in triplicate and always in the same plate.

The stability of each reference gene in the different samples was determined with geNorm software [Vandesompele et al. 2002 (Ghent, Belgium)]. β-actin, hypoxanthine-guanine phosphoribosyl transferase (HPRT), and glyceraldehyde-3-phosphate dehydrogenase (GAPDH) were the most stable in both the striatum and the mesencephalon and were used in combination to determine the relative levels of expression of the lesion- and PTN-associated genes, which was calculated with qBASE software [Hellemans et al. 2007 (Ghent, Belgium)]. The fold increase in expression was determined with respect to the average values of the sham-vehicle group, arbitrarily defined as 1.

Protein extraction and western blot

Striatal tissue was gently thawed and homogenized with a rotor-stator (Ultra-Turrax T8; IKA-Werke, Staufen, Germany) in lysis buffer containing 150 mM NaCl, 50 mM Tris–HCl, pH 8, 1% NP-40 (Sigma-Aldrich, St. Louis, MO, USA), and a cocktail of protease inhibitors (Complete, Roche, Neuilly-sur-Seine, France). Total protein concentrations of the samples were measured by the method of Bradford using the Protein Assay Kit (Bio-Rad Laboratories, Hercules, CA, USA).

To obtain a single band at the appropriate molecular weight, RPTP-ζ/β was deglycosylated by chondroitinase before electrophoresis, as described (Dobbertin et al. 2003). Briefly, each homogenate was incubated by 3 h at 37°C in equal volume of digestion solution containing, at final volumes, 0.1 U/mL of chondroitinase ABC (EC 4.2.2.4; Sigma-Aldrich), 50 mM Tris, pH 8, 60 mM sodium acetate, and 0.02% bovine serum albumin.

Forty micrograms of total proteins was electrophoresed on a 3–8% (for RPTP-ζ/β) or 12% (for TH) pre-stacked acrylamide and bis-acrylamide gel (Invitrogen, Cergy-Pontoise, France) and electrotransferred to a nitrocellulose membrane (Protran; Perkin-Elmer, Courtaboeuf, France). Each membrane was blocked for 2 h in 5% dry fat milk at 20–25°C and incubated overnight with the following antibodies at 4°C: mouse anti-RPTPb, (1 : 250; BD Transduction Laboratories, Erembodegm, Belgium); mouse anti-TH (1 : 1000, DiaSorin, Saluggia, Italy); or mouse anti-β-actin (1 : 2000, Sigma-Aldrich). Then, they were incubated with a horseradish peroxidase-conjugated anti-mouse antibody (1 : 1000, 1 h at 20–25°C; Pierce Biotechnology, Brebiers, France) and developed with standard chemiluminescent procedures (SuperSignal West Dura; Thermo Scientific, Courtaboeuf, France). Chemiluminescence was visualized with a Kodak Image Station 4000MM (Raytest, La Defense, France) and quantified with the Image J software (NIH ImageJ, Bethesda, MD, USA) (Rasband 1997-2007). Internal controls were repeated in each membrane and values were normalized to actin. Experiments were repeated twice.

Statistics

Results are presented as the mean ± SEM. Differences between biological conditions were calculated by two-way anova using PrisM 4 (GraphPad Software; GraphPad Software Inc., San Diego, CA, USA), with treatment and lesion as anova factors; post hoc comparisons were made with the Bonferroni test. The level of significance was set at < 0.05 in all cases.

Histology

For histological analysis, groups of three normal rats were also injected with 6-OHDA or saline as above. Two weeks after surgery, they were perfused with saline followed by 200 mL of 4% paraformaldehyde in 0.1 M phosphate buffer, pH 7.4. The brains were then removed, post-fixed for 2 h in the same fixative, cryoprotected for 24 h in 30% sucrose solution, frozen in isopentane (−30°C), and stored at −80°C until processing for immunodetection. The striatum and the midbrain were cut on a freezing microtome in series of 25-μm-thick coronal sections. Slices were stored in 0.1 M phosphate-buffered saline (PBS) containing 0.02% sodium azide.

Immunohistochemistry was performed on free-floating sections, which were rinsed in PBS with 0.1 M Triton X-100 (0.15%) (PBS-T), then incubated 30 min in a blocking solution containing 0.3% Triton X-100 and 5% normal goat serum in 0.1 M PBS; endogenous peroxidase reactivity was blocked in 0.3% H2O2 in PBS-T by 15 min. After a rinse in PBS-T, the sections were incubated with the primary antibody, with slow agitation, for 2 h at 20–25°C with rabbit anti-TH (1 : 500, Pel Freeze Biologicals, Rogers, AR, USA) or three overnights at 4°C with mouse anti-RPTP-ζ/β (anti-RPTPb, 1 : 20, BD Transduction Laboratories). Slides were then rinsed three times in PBS-T and incubated for 2 h at 20–25°C with the corresponding biotin-labeled antibody (1 : 250, Vector Laboratories, Paris, France), followed by an avidin–biotin peroxidase complex (1 : 125, Vector Laboratories), and developed with 0.1% 3,3′-diaminobenzidine and 0.001% H2O2 in PBS-T. The sections were mounted on gelatin-coated slides, dehydrated, and coverslipped. The anti-RPTP-ζ/β antibody has been widely used in the detection of this protein (Kawachi et al. 1999; Dobbertin et al. 2003; Hayashi et al. 2005); it recognizes only the intracellular portion of the RPTP-ζ/β to avoid non-specific labeling of other proteoglycans and specially of phosphacan, a secreted peptide which has the same molecular structure as the extracellular portion of the RPTP-ζ/β. Only the band at 250 kDa is detected by western blot after chondroitinase treatment, as previously described (Levy et al. 1993; Dobbertin et al. 2003). Furthermore, histological detection of RPTP-ζ/β in the cortex (not shown here), where it is highly expressed, gave results similar to those previously published (Kawachi et al. 1999). Photographs were taken with a Leitz Laborlux S microscope (Carl Zeiss, Le Pecq, France).

Double-fluorescent labeling was carried out sequentially and RPTP-ζ/β was always immunolabeled first. The primary antibodies used were: anti-RPTP-ζ/β and anti-TH as above, rabbit anti-S100β (1 : 5000, 1 overnight, Sigma) to identify glial cells, and goat anti-DARPP-32 (1 : 100, two overnights, Santa Cruz Biotechnology, Tebu-Bio, Le Perray en Yvelines, France) to label MSNs. All washes and incubations were carried out in PBS-T and 0.2% of gelatin. Non-specific binding was avoided by pre-incubation of the slices for 30 min in a blocking solution containing 5% normal goat serum or bovine serum albumin in the case of DARPP-32. Anti-RPTP-ζ/β immunoreactivity was detected by incubation with a biotin-labeled anti-mouse IgG antiserum for 2 h (1 : 250; Vector Laboratories), followed by rinsing, and then incubation with streptavidin-Alexa Fluor 488 (1 : 1000; Molecular Probes, Cergy-Pontoise, France) for other 2 h. Other primary antibodies were detected with a Cy3-conjugated anti-rabbit or anti-goat (1 : 500; Jackson Labs, Suffolk, UK). To label nuclei, sections were counterstained with 4,6-diamidino-2-phenylindoldihydrochloride (DAPI) (1 μg/mL). The sections were mounted on glass slides with Mowiol medium. Images were acquired with a Leica SP2 AOBS confocal (Leica, Rueil Malmaison, France) connected to a Leica DM IRE 2 microscope using the Leica Confocal Software. The contrast and brightness of colored images were adjusted to obtain similar background levels on all photographs of a given plate.

Results

Evaluation of the model and l-DOPA treatment

To determine the extent of the dopaminergic denervation in 6-OHDA treated rats, we evaluated the loss of dopaminergic terminals in the striatum by western blot of TH and immunohistochemistry. Partial denervation of the dorsal striatum was obtained (Fig. 1a and b), as previously reported (Ferrario et al. 2003; Debeir et al. 2005). We also quantified TH mRNA in the mesencephalon by qRT-PCR. TH mRNA levels decreased about 50% in 6-OHDA-lesioned animals (Fig. 1c). l-DOPA also induced a decrease in TH expression, which was statistically significant in the sham group. The down-regulation of TH mRNA by l-DOPA therapy is in accordance with observations in patients with PD (Kastner et al. 1993) and confirms that l-DOPA was correctly delivered to the brain.

Figure 1.

 Neuronal specificity of 6-hydroxydopamine (6-OHDA) lesions. (a) Quantification of striatal tyrosine hydroxylase (TH) levels by western blot show a reduction of about 50% on the lesioned side. (b) Labeling of dopaminergic fibers in the striatum of sham-lesioned and 6-OHDA-lesioned rats. (c) Quantification of nigral TH mRNA levels evaluated by quantitative RT-PCR (qRT-PCR) also shows a decrease in expression. d) The specificity of 6-OHDA for dopaminergic neurons was confirmed by the absence of changes, in the striatum, in the relative levels of DARPP-32 mRNA evaluated by qRT-PCR. Data were analyzed by a two-way anova (lesion and treatment as factors) [F(1,20) = 32, *< 0.0001, for TH western blot and F(1,23) = 22.56, †< 0.0001, for TH qRT-PCR]. Post hoc analysis shows that the TH expression is down-regulated in l-DOPA-treated rats; *< 0.05. DARPP-32, cyclic AMP-related phosphoprotein 32 kDa.

To determine whether the striatal injections of 6-OHDA lesion specifically the nigral dopaminergic neurons or striatal neurons as well, we quantified the mRNA of DARPP-32, a marker of striatal GABAergic MSNs, which are the main neural population in this structure. No changes in this marker were detected in any group (Fig. 1d), as expected from previous studies (Raisman-Vozari et al. 1990; Campbell and Bjorklund 1995). Thus striatal neurons are not affected by the 6-OHDA.

Expression analysis of PTN and its receptors

The expression of PTN and its three known receptors (N-Syndecan, RPTP-ζ/β, and ALK) was quantified by qRT-PCR in the striatum and mesencephalon of control and 6-OHDA-lesioned rats, treated with l-DOPA or vehicle for 21 days. The expression of PTN and RPTP-ζ/β were both significantly up-regulated as a consequence of the lesion (Fig. 2), and RPTP-ζ/β was further up-regulated in lesioned rats by l-DOPA treatment (Fig. 2b). The other two PTN receptors, N-Syndecan and ALK, were not significantly affected by either the lesion or the treatment. In the mesencephalon, neither the 6-OHDA lesion nor the l-DOPA treatment had significant effects on the expression of PTN or its receptors RPTP-ζ/β and N-Syndecan. ALK expression was not detected in this structure (Fig. 3).

Figure 2.

 Expression of pleiotrophin (PTN) and its receptors in the striatum of 6-hydroxydopamine (6-OHDA)-lesioned rats treated with l-DOPA or vehicle. The levels of PTN (a) and receptor protein tyrosine phosphatase type zeta beta (RPTP-ζ/β) (b) mRNA quantified by quantitative RT-PCR (qRT-PCR) is up-regulated in the striatum of 6-OHDA-lesioned rats, as determined by a two-way anova (lesion and treatment as factors) [F(1,24) = 5.75, †= 0.024, for PTN and F(1,23) = 7.14, †= 0.022, for RPTP-ζ/β]. Post hoc analysis shows that the mRNA levels of RPTP-ζ/β is up-regulated in l-DOPA-treated rats; *< 0.05. There are no significant differences in the expression of N-syndecan (c) or anaplastic lymphoma kinase (ALK) (d) mRNA.

Figure 3.

 Expression of pleiotrophin (PTN) and its receptors in the mesencephalon of 6-hydroxydopamine (6-OHDA)-lesioned rats treated with l-DOPA or vehicle. There are no significant differences in the expression of PTN (a), receptor protein tyrosine phosphatase type zeta beta (RPTP-ζ/β) (b), or N-syndecan (c) mRNA, quantified by quantitative RT-PCR.

Quantification of RPTP-ζ/β protein 

The analysis of gene expression suggests that the PTN receptor, RPTP-ζ/β, is up-regulated as consequence of both dopaminergic denervation and l-DOPA treatment. To determine if gene up-regulation is followed by an increase in protein levels, we quantified RPTP-ζ/β protein levels by western blot in another similar group of animals. We detected RPTP-ζ/β at the expected molecular weight of 250 kDa after chondroitinase digestion (Fig. 4a), as described (Levy et al. 1993; Dobbertin et al. 2003). After quantification, we found that the levels of RPTP-ζ/β protein were increased in the striatum of l-DOPA-treated lesioned rats (Fig. 4b).

Figure 4.

 Quantification by western blot of receptor protein tyrosine phosphatase type zeta beta (RPTP-ζ/β) protein levels in the striatum of 6-hydroxydopamine (6-OHDA)-lesioned rats treated with l-DOPA or vehicle. a) Representative western blots of RPTP-ζ/β after chondroitinase (chABC) digestion (top) and samples used in quantitative analysis (bottom); SV: sham-vehicle, LV: lesion-vehicle, LD: lesion-l-DOPA, SD: sham-l-DOPA. b) Quantification of RPTP-ζ/β protein shows that it is up-regulated in the striatum of 6-OHDA-lesioned rats, as determined by a two-way anova (lesion and treatment as factors) [F(1,19) = 13.24, †< 0.002]. Post hoc analysis shows that RPTP-ζ/β is increased in l-DOPA-treated rats; *< 0.05. TH, tyrosine hydroxylase.

Immunodetection of RPTP-ζ/β

To determine where this receptor is located, we immunolabeled RPTP-ζ/β alone or in combination with cell-specific markers on sections of striatum and mesencephalon of normal rats. At low magnification, weak immunolabeling of the receptor was observed (Fig. 5b). At high magnification (Fig. 5), RPTP-ζ/β immunoreactivity was widely distributed throughout the substantia nigra; labeling was stronger, however, in the pars compacta than in the pars reticulata. RPTP-ζ/β immunoreactivity could not be detected in the ventral tegmental area (Fig. 5b). Double-labeling with anti-TH showed that RPTP-ζ/β is located on cell bodies and proximal processes of most of dopaminergic neurons in the SNpc (Fig. 5c–e), but not in those of the ventral tegmental area (Fig. 5f–h), confirming a previous microarray study (Chung et al. 2005).

Figure 5.

 Cellular localization of receptor protein tyrosine phosphatase type zeta beta (RPTP-ζ/β) in the mesencephalon of normal rats. Immunohistochemistry of tyrosine hydroxylase (TH) (a) and RPTP-ζ/β (b) showing labeling in the substantia nigra pars compacta (SNpc) (arrow heads) but not in the ventral tegmental area (VTA) (large arrows). A higher magnification of RPTP-ζ/β immunoreactivity is shown in the inset. c–h) Immunofluorescent labeling of RPTP-ζ/β (green) and TH (red) in the SNpc (c,d) and VTA (f,g). Merged images (e,h) show that RPTP-ζ/β is expressed in the cell bodies and proximal processes of TH-positive dopamine neurons in the SNpc (arrow heads in c–e) but not in the VTA. All fluorescent images were taken from the same section and with the same confocal setting. Scale bar: 50 μm.

In the striatum, RPTP-ζ/β immunoreactivity was distributed homogeneously over the whole structure, on the membranes and proximal projections of neurons. RPTP-ζ/β immunoreactivity colocalized with DARPP-32, confirming its presence on MSNs (Fig. 6a–c). No co-localization was observed with the glial marker S100ß (Fig. 6d–f). We also detected some labeling on fibers which appear to be non-dopaminergic, as we did not detect colocalization with TH-positive dopaminergic fibers (Fig. 6g–i). RPTP-ζ/β immunoreactivity was also present and not visibly altered in the dopaminergic denervated striatum (data not shown).

Figure 6.

 Receptor protein tyrosine phosphatase type zeta beta (RPTP-ζ/β) is found in medium spiny neurons (MSNs) in the striatum of normal rats. Confocal images show that RPTP-ζ/β (green) colocalizes with DARPP32 (red), a marker of MSNs (arrow heads in a–c), but not with the glial marker S100β (red, d–f) or tyrosine hydroxylase (TH) (red, g–i) in dopamine fibers (arrow heads in g and h show a fiber with RPTP-ζ/β immunoreactivity that does not colocalize with TH). Nuclei are stained with 4,6-diamidino-2-phenylindoldihydrochloride (DAPI) (blue). Scale bar: 10 μm. DARPP32, cyclic AMP-related phosphoprotein 32 kDa.

Discussion

We have shown here that RPTP-ζ/β is co-upregulated with PTN following a partial lesion of the nigrostriatal dopaminergic system, and is further up-regulated in the striatum by l-DOPA treatment. In addition, we showed that RPTP-ζ/β is constitutively expressed by MSNs in the striatum, the major output neurons in that structure which integrate glutaminergic signals from the cortex and dopaminergic signals from the SNpc. Together, these data suggest that RPTP-ζ/β might be the receptor that mediates the putative plastic effects of PTN, such as the striatal synaptic rearrangements caused by l-DOPA treatment.

Receptor protein tyrosine phosphatase type zeta beta is a chondroitin sulfate proteoglycan that is exclusively expressed in the nervous system (Levy et al. 1993), and particularly in the developing brain where it has been related to neurite outgrowth, axon guidance, and synaptic assembly (Johnson and Van Vactor 2003). It is also located postsynaptically in the adult cortex and hippocampus (Hayashi et al. 2005). It has been shown to regulate the phosphorylation level of important structural molecules, such as β-catenins (Meng et al. 2000), β-adducin (Pariser et al. 2005a), and Fyn (Pariser et al. 2005b) and strongly interacts with the scaffolding, post-synaptic density protein 95 (Kawachi et al. 1999; Fujikawa et al. 2007). These observations, in addition to the specific localization of RPTP-ζ/β in the cell bodies of the nigral dopaminergic neurons and striatal MSNs, strongly suggest that it is a constitutive tyrosine phosphatase of these two major elements of the nigrostriatal pathway.

Several plastic changes take place in the DA-denervated striatum either spontaneously or induced by l-DOPA such as dopaminergic sprouting (Murer et al. 1998), and several post-synaptic rearrangements, such as altered protein phosphorylation, enhanced receptor signaling, or overactivation of GABAergic neurons (Cenci and Lundblad 2006). The role of PTN in these processes in the lesioned striatum is still unclear, but several hypotheses are suggested by its functions in the developing brain. PTN is a neuritogenic factor, but it is also a potent angiogenic factor (Deuel et al. 2002); thus, it might induce blood vessel growth in the denervated striatum. Interestingly, an increase in striatal microvasculature has been shown in l-DOPA-treated animals that developed dyskinesias (Westin et al. 2006). Although this structural rearrangement seems to be exclusively related to LID, not to l-DOPA treatment alone.

Pleiotrophin via RPTP-ζ/β might also act directly on dopaminergic neurons or their post-synaptic elements. The former seems unlikely as (i) RPTP-ζ/β was not present in dopaminergic terminals; (ii) immunoneutralization of RPTP-ζ/β had no effect on the maturation of dopaminergic neurons in culture (M. E. Gomes and J. E. Ferrario, unpublished data); (iii) the observed reduced number of striatal interneurons expressing PTN and the low amounts of the protein in the structure (Taravini et al. 2005) would not be expected for a potent growth factor. Indeed, the concomitant up-regulation of PTN and RPTP-ζ/β in the lesioned striatum and the localization of RPTP-ζ/β in MSNs described herein is compatible with a role of these molecules in post-synaptic striatal plasticity.

The striatal adaptive re-arrangements described above might be positive attempts to restore the function of the nigrostriatal system; however, too extensive and/or inappropriate synaptic plasticity in an adult brain might contribute to the development of undesirable side effects such as dyskinesias (Cenci 2007). The up-regulation of RPTP-ζ/β following l-DOPA treatment suggests that it might be involved in l-DOPA-induced plasticity, probably LID. This hypothesis is supported by observations suggesting that the PTN/RPTP-ζ/β signal controls intracellular molecules which regulate both NMDA glutamate receptors (Dunah et al. 2004) and D1 receptors (Gu et al. 2007; Zhang et al. 2007), both of which play a crucial role in LID (Cenci and Lundblad 2006). However, the functional implication of RPTP-ζ/β in the development of dyskinesias remains to be further analyzed in a suitable animal model of LID.

In conclusion, we found that RPTP-ζ/ß is present on soma and proximal processes of striatal GABAergic neurons and might play an important role in the interaction of afferent neurons with their post-synaptic targets. The up-regulation of RPTP-ζ/β in the lesioned striatum, and even more in l-DOPA-treated rats points to a role in striatal plasticity suggesting that RPTP-ζ/β might therefore be a potential target in the study of plastic re-arrangements associated with l-DOPA treatment in Parkinson’s disease.

Acknowledgments

We would like to thank Merle Ruberg, Oscar Gershanik, Elena Avale, and Irene Taravini for helpful discussion, and Héctor Ardila, José Fernandez, and Laure Ginestet for technical advice and assistance. JEF was financed by the FRM (Fondation pour la Recherche Médicale, France, ACE20040700891 and ACE20050704104), AERM by CONACYT, (Mexico, code 76101), MSO by Naturalia et Biologia (France), and MZG and CS by CAPES-COFECUB (France/Brazil).

Ancillary