OUP user menu

Pharmacological stimulation of sigma-1 receptors has neurorestorative effects in experimental parkinsonism

Veronica Francardo, Francesco Bez, Tadeusz Wieloch, Hans Nissbrandt, Karsten Ruscher, M. Angela Cenci
DOI: http://dx.doi.org/10.1093/brain/awu107 1998-2014 First published online: 22 April 2014

Summary

The sigma-1 receptor, an endoplasmic reticulum-associated molecular chaperone, is attracting great interest as a potential target for neuroprotective treatments. We provide the first evidence that pharmacological modulation of this protein produces functional neurorestoration in experimental parkinsonism. Mice with intrastriatal 6-hydroxydopamine lesions were treated daily with the selective sigma-1 receptor agonist, PRE-084, for 5 weeks. At the dose of 0.3 mg/kg/day, PRE-084 produced a gradual and significant improvement of spontaneous forelimb use. The behavioural recovery was paralleled by an increased density of dopaminergic fibres in the most denervated striatal regions, by a modest recovery of dopamine levels, and by an upregulation of neurotrophic factors (BDNF and GDNF) and their downstream effector pathways (extracellular signal regulated kinases 1/2 and Akt). No treatment-induced behavioural-histological restoration occurred in sigma-1 receptor knockout mice subjected to 6-hydroxydopamine lesions and treated with PRE-084. Immunoreactivity for the sigma-1 receptor protein was evident in both astrocytes and neurons in the substantia nigra and the striatum, and its intracellular distribution was modulated by PRE-084 (the treatment resulted in a wider intracellular distribution of the protein). Our results suggest that sigma-1 receptor regulates endogenous defence and plasticity mechanisms in experimental parkinsonism. Boosting the activity of this protein may have disease-modifying effects in Parkinson’s disease.

  • mouse
  • neuroprotection
  • 6-hydroxydopamine
  • Parkinson’s disease
  • MAPK signalling

Introduction

Several studies have reported neuroprotective effects of sigma-1 receptor (Sig-1R) ligands in animal models of brain disorders, including Alzheimer’s disease (Jansen et al., 1993; Mishina et al., 2008), amyotrophic lateral sclerosis (Mancuso et al., 2012; Prause et al., 2013), and stroke (Harukuni et al., 2000; Ajmo et al., 2006; Ruscher et al., 2011, 2012). At the end of the 70s, the Sig-1R was mistakenly classified as an orphan opioid receptor (Martin et al., 1976), but more recent studies have shown that Sig-1R is a molecular chaperone at the mitochondrion-associated endoplasmic reticulum membrane where it regulates calcium signalling between the two organelles (Hayashi and Su, 2007). Under conditions of cellular stress (Hayashi and Su, 2007) or upon pharmacological stimulation (Hayashi and Su, 2003a, b), Sig-1Rs dissociate from this membrane protein complex and redistribute widely to support various cellular functions (Maurice and Su, 2009; Su et al., 2010). These include regulation of lipid transport (Hayashi and Su, 2003b), modulation of synaptic signalling (Kourrich et al., 2012, 2013), stimulation of axonal outgrowth (Kimura et al., 2013) and dendritic spine arborization (Tsai et al., 2009).

Substances that increase the activity and intracellular trafficking of Sig-1R are defined as Sig-1R agonists. Several in vitro and in vivo studies have indicated that Sig-1R agonists may counteract apoptosis and excitotoxicity by multiple mechanisms, such as upregulation of protective genes (Yang et al., 2007; Meunier and Hayashi, 2010), suppression of microglial activation (Griesmaier et al., 2012; Mancuso et al., 2012), maintenance of mitochondrial integrity with reduced production of reactive oxygen species (Meunier and Hayashi, 2010) and nitric oxide (NO) (Goyagi et al., 2001; Vagnerova et al., 2006; Yang et al., 2010). All these actions are potentially relevant to Parkinson’s disease, as mitochondrial dysfunction and nitrosative stress are critical mediators of nigrostriatal dopamine degeneration (Liberatore et al., 1999; Iravani et al., 2002; Cali et al., 2012), and neuroinflammation is widely recognized as an important contributor to neurodegeneration in parkinsonian disorders (Hirsch and Hunot, 2009; Tansey and Goldberg, 2010). To date, possible neurorestorative effects of Sig-1R agonists in animal models of Parkinson’s disease have not yet been reported.

In the present study, we used mice with 6-hydroxydopamine lesions to evaluate the expression of Sig-1R in the damaged nigrostriatal system and the effects of its pharmacological modulation. Chronic treatment with the Sig-1R agonist, 2-morpholin-4-ylethyl 1-phenylcyclohexane-1-carboxylate (PRE-084) produced a dose-dependent behavioural and histological neurorestoration. These effects were accompanied by an upregulation of neurotrophic factors and activation of cell survival pathways [extracellular signal-regulated kinases 1 and 2 (Erk1/2), Akt] in the nigrostriatal system. Parallel experiments in Sig-1R knockout mice provided a validation of Sig-1R being the molecular target of the neurorestorative treatment.

Materials and methods

Animals

The study was performed in male C57BL/6 mice (Charles River Laboratories), and Sig-1R knockout mice of both genders on a mixed C57BL/6Jx129S/SvEv background. As a Sig-1R knockout line we chose the well-characterized Sigmar1Gt(OST422756)Lex mouse (Sabino et al., 2009; Mavlyutov et al., 2010; Chevallier et al., 2011), obtained from the Mutant Mouse Resource Regional Centre (MMRRC) at the University of California, Davis. Mice were housed under a 12-h light/dark cycle with free access to food and water. Housing conditions and experimental treatments had been approved by the Malmö-Lund Ethical Committee on Animal Research.

Experimental design

A total of 309 mice were used in this study. Treatment with PRE-084 or its vehicle (saline) was administered daily for 35 days starting on the same day as the 6-hydroxydopamine lesion. One additional experiment was carried out to examine the effects of a delayed-start treatment, where PRE-084 was given daily for 35 days starting on Day 8 following 6-hydroxydopamine lesion (Supplementary material). Behavioural tests were carried out once per week to evaluate spontaneous turning behaviour and forelimb use asymmetry. At the end of the behavioural pharmacological studies animals were either perfusion-fixed for immunohistochemistry, or decapitated for either biochemical assays [measurements of dopamine, 5-hydroxytryptamine (5-HT) and their metabolites] or western immunoblotting (GDNF, BDNF, ERK1/2, Akt).

6-Hydroxydopamine lesions

Lesions were performed on mice weighing ∼25 g (corresponding to ∼8–9 weeks of age) according to previously described procedures (Francardo et al., 2011).

Briefly, mice were anaesthetized with isofluorane (Isoba® vet, Apoteksbolaget) and placed in a stereotaxic frame on a flat-skull position. Injections of 6-hydroxydopamine (3.2 µg/µl free base concentration in 0.02% ascorbate-saline) were made at the following coordinates, given in mm relative to bregma, sagittal suture and dural surface (cf. Paxinos and Franklin, 2001): (i) AP = +1.0, L = −2.1, DV = −2.9; (ii) AP = +0.3, L = −2.3; DV = −2.9. Toxin solution (1 µl) was injected at each site.

PRE-084 treatment

PRE-084 (Tocris) was dissolved in physiological saline immediately before use and injected at the volume of 0.1 ml/10 g body weight in a single subcutaneous injection per day, which was administered between 5 and 6 pm. In a first pilot experiment we tested four doses of PRE-084 (0.1, 0.3, 1.0 and 3.0 mg/kg/day) or saline treatment for 7 days (n = 6–11 per group). After treatment with 0.3 and 1 mg/kg PRE-084, tyrosine hydroxylase (TH)-positive neurons in the lesioned substantia nigra compacta were 71.8 ± 4.5 and 68.8 ± 3.4 of the contralateral side, respectively, indicating a significant neuroprotective effect (cf. 55.2 ± 4.5% in saline-treated controls, P < 0.05 for both doses). The doses of 0.1 and 3.0 had no effect. Thus, the doses of 0.3 and 1 mg/kg were selected for the chronic experiment (35 days treatment). In each experiment, treatment with the selected doses of PRE-084 or saline was administered also to sham-lesioned mice, which were pooled into one single group (‘sham’) for statistical analysis (as the drug treatment did not have any effect in these animals).

Behavioural testing

Three different behavioural tests were used to monitor the effects of chronic treatment with PRE-084. A test of spontaneous rotation was used to assess the postural and locomotor asymmetry resulting from unilateral nigrostriatal lesions. Unlike drug-induced turning, spontaneous ipsilateral rotations are a transient response elicited by novelty or stress (Cenci et al., 2002). This behaviour subsides gradually within weeks post-lesion (Francardo et al., 2011; Lindgren et al., 2012). Forelimb use asymmetry during vertical exploration (cylinder test) provides a validated measure of akinesia in hemiparkinsonian rodents (Lundblad et al., 2002). An impaired capacity to perform adjusting steps during experimenter-imposed movements (stepping test) reflects both akinetic and postural deficits relevant to Parkinson’s disease (Schallert et al., 2000; Blume et al., 2009).

These tests were carried out between 9 am and 2 pm at the end of each treatment week, and data were analysed by an experimentally blind investigator.

Spontaneous rotational activity

A videotracking system monitoring the relative position of the mouse head, body and tail (Viewer2, Biobserve BmbH) was used to measure the animals’ spontaneous rotational activity in an open field, according to previously described procedures (Francardo et al., 2011). Briefly, each mouse was placed individually in the centre of the open field (50 × 50 cm) and immediately recorded for 10 min, corresponding to the period of maximal activity (Francardo et al., 2011). Data were expressed as the total number of net full turns (360°) per session towards the side ipsilateral to the lesion.

Cylinder test

To detect forelimb use asymmetry during spontaneous vertical movements, mice were placed individually in a glass cylinder (10 cm diameter, 14 cm height) and video-recorded for 3–5 min (further details can be found in Francardo et al., 2011). The number of supporting wall contacts performed by the paw contralateral to the lesion was expressed as a percentage of all supporting wall contacts in each session.

Stepping test

Forelimb use asymmetry during experimenter-imposed horizontal movements was assessed using a modification of the method described in Blume et al. (2009). The mouse was placed at the beginning of a corridor (7-cm wide, 1-m long, flanked by 10-cm high plastic walls), gently lifted by the tail and pulled backwards with a fixed speed (1 m/4 s). Each trial was recorded using a digital videocamera (Canon Legria FS406), and videorecordings were used to count the number of adjusting steps performed with the forelimb ipsilateral and contralateral to the lesion. Trials in which mice turned their body by 90° towards the walls of the corridor (in an attempt to escape or explore) were discarded. In each session, mice were tested until three valid trials per animal were obtained. Results were expressed as the percentage of adjusting steps performed with the forelimb contralateral to the lesion (left paw) over the total number of steps (mean of three trials per session).

Tissue preparation for histology

Mice allocated to immunohistochemistry were transcardially perfused 12–20 h after the last PRE-084/saline injection. Animals were deeply anaesthetized with sodium pentobarbital (240 mg/kg, intraperitoneal; Apoteksbolaget) and perfusion-fixed with 0.9% saline, followed by buffered ice-cold, 4% paraformaldehyde (VWR) (pH 7.4). After rapid extraction, brains were post-fixed in the same fixative for 2 h and then cryoprotected in ice-cold 25% phosphate-buffered sucrose overnight. Coronal sections of 30-μm thickness were cut on a freezing microtome and stored at −20°C in a non-freezing solution (30% ethylene glycol and 30% glycerol in 0.1 M phosphate buffer).

Other animals were anaesthetized as above and immediately decapitated. Their brains were rapidly extracted and frozen on crushed dry ice. Tissue samples were dissected out for biochemical analyses or western blotting (see below).

Bright-field immunohistochemistry and quantitative analyses

Bright-field immunohistochemistry was performed according to established protocols (Francardo et al., 2011), using the following primary antibodies: rabbit anti-TH (Pel-Freez, diluted 1:1000), rabbit anti-Sigma-1 receptor (Santa Cruz biotechnology, diluted 1:800), rat anti-CD68 (AbD Serotec, diluted 1:1000), which labels activated microglia and other immune cells, such as macrophages (Perry and Teeling, 2013) Quantitative image analysis was performed by an experimentally blinded investigator as described below.

The number of TH-positive cell bodies in the substantia nigra compacta was determined by unbiased stereology according to the optical fractionator method (West, 1999). Analysis was performed using a Nikon 80i microscope with an x–y motorized stage controlled by the NewCAST software (Visiopharm). The sampling fraction was chosen so as to count at least 100 neurons per side per animal following an established protocol (Francardo et al., 2011), and the total number of TH-positive neurons in the substantia nigra pars compacta was then estimated using the optical fractionator formula: number of neurons = 1/ssf (slice sampling fraction) × 1/asf (area sampling fraction) × 1/tsf (thickness sampling fraction) × Σ (number of objects counted) (West, 1999).

The overall optical density of TH immunostaining in different striatal subregions was assessed on low magnification pictures (Fig. 1) using the freeware NIH ImageJ 1.43 (National Institute of Health, Bethesda, MD; downloadable from http://rsbweb.nih.gov/). The density (number and thickness) of TH-immunoreactive fibres in highly denervated regions was then assessed using an image segmentation software VIS (Visiopharm Integrator System; Visiopharm). To this end, the dorsolateral striatum and the substantia nigra pars reticulata were outlined at low magnification (×4 objective) in a Nikon Eclipse 80i microscope. A systematic random sampling was then applied under a ×100 objective (sample area size, 0.010 mm2) to cover a fixed percentage of the structure of interest (10% of the dorsolateral striatum, 50% of the substantia nigra reticulata). Images were captured with a digital camera (Olympus DP72), obtaining 25–30 areas per mouse per structure. On each image, immunopositivity associated with TH fibres was separated from background objects using a Bayesian algorithm-based pixel classifier (Westin et al., 2006). Results were expressed as the total number of immunopositive pixels per sample area.

Figure 1

Overviews of four rostrocaudal levels through the striatum (A) and the substantia nigra (B) used for quantitative immunohistochemical analysis. Scale bars = 1 mm and 0.5 mm in A and B, respectively. DL = dorsolateral; DM = dorsomedial.

Counts of CD68-positive cells in the striatum and the substantia nigra were performed using a Nikon 80i microscope with an x–y motorized stage controlled by the NewCAST software (Visiopharm). A mask for delineating the striatum and substantia nigra was defined at ×4 magnification. Counts of CD68-positive cells were carried out on four sections per mouse corresponding to the rostrocaudal levels depicted in Fig. 1A (level I, bregma +1.34/+1.18; II, bregma +0.74/+0.50; III, bregma +0.14; IV, bregma −0.34/−0.50) and 1B (level I, bregma −2.92; II, bregma −3.08; III, bregma −3.16; IV, bregma −3.52) (Paxinos and Franklin, 2001). Only CD68-positive cells with microglial morphology (as described in Cicchetti et al., 2002; Perry and Teeling, 2013) were counted, under a ×40 objective, in the entire area of interest. Data from the 6-hydroxydopamine-lesioned side were expressed as a percentage of the measurements from the intact side in each section.

Dual-antigen immunofluorescence

Sections through the substantia nigra and the striatum were processed for dual-antigen immunofluorescence to detect Sig-1R expression in neurons, dopaminergic neurons, or glial cells using the following primary antibodies: rabbit anti Sig-1R (1:200, Santa Cruz Biotechnology, Inc); mouse anti-neuronal nuclei (NeuN) (1:400, Millipore); mouse anti-glial fibrillary acidic protein (GFAP) (1:400, Millipore); rat anti-dopamine transporter (DAT) (1:800, Millipore). The antibodies were diluted in potassium phosphate buffered-saline (KPBS) containing 1% normal serum (from goat, horse or donkey, where appropriate) and 0.3% Triton™ X-100. Sections were subsequently incubated for 1 h at room temperature with Alexa Fluor® 488-conjugated goat anti-rabbit antibody (1:200; Molecular Probes, Invitrogen) and Cy3-conjugated horse anti-mouse/donkey anti-rat antibodies (1:200, Jackson ImmunoResearch), diluted in 0.3% Triton™ X-100-KPBS. Sections were then rinsed in KPBS, mounted, coverslipped with polyvinyl alcohol mounting medium with DABCO® (PVA/DABCO) (Sigma-Aldrich), and viewed under a confocal laser scanning fluorescence microscope (LSM510 Zeiss).

Biochemical analysis

A brain slice spanning rostrocaudal levels, +1.18 to −0.34 mm relative to bregma, was extracted using a mouse brain mould. The dorsolateral and dorsomedial striatum were dissected out using a scalpel blade, and samples were kept frozen until the analysis. Tissue samples from the substantia nigra were taken with a tissue puncher of 2-mm diameter spanning the following rostrocaudal levels relative to bregma, −2.70 to −3.80. The samples were kept frozen until further analysis.

For biochemical measurements of dopamine, 5-HT and their metabolites, the samples underwent an ultrasound homogenization (Sonifier Cell Disruptor B30; Branson Sonic Power Co.) in 0.1 M perchloric acid and 2.5 mM of Na2EDTA, and were then centrifugated. The supernatant was processed for high pressure liquid chromatography (HPLC) followed by electrochemical detection according to well-established methods (Lindgren et al., 2010).

Western blot analysis

Tissue homogenates were prepared in cell lysis buffer [20 mmol/l Tris (pH 7.5), 150 mmol/l NaCl, 1 mmol/l EDTA, 1 mmol/l EGTA, 1% Triton™ X-100, 2.5 mmol/l sodium pyrophosphate, 1 mmol/l β-glycerolphosphate, 1 mmol/l Na3VO4, 1 μg/ml leupeptin, and 1 mmol/l phenylmethylsulphonyl fluoride] (Rickhag et al., 2008). Twenty micrograms of protein per sample were separated on a 10% SDS PAGE gel. Proteins were transferred onto polyvinyldifluoride membranes, which were incubated in blocking buffer (20 mM Tris, 136 mM NaCl, pH 7,6, 0,1% Tween 20, 5% non-fat dry milk). Thereafter, membranes were incubated overnight at 4°C using one of the following primary antibodies: rabbit polyclonal anti GDNF (Santa Cruz Biotechnologies, CA, 1:1000); rabbit polyclonal anti BDNF (Santa Cruz Biotechnologies, CA, 1:1000); rat polyclonal anti CD68 (AbD Serotec, 1:1000); rabbit polyclonal antibodies against total or Thr202/Tyr204-phosphorylated ERK1/2 (p44/42-MAPK; Cell Signaling Technology Inc, 1:2000); rabbit polyclonal antibodies against total and phosphorylated-Ser473 Akt (Cell Signaling Technology Inc, 1:2000). After appropriate washing steps, membranes were incubated with horseradish peroxidase-linked secondary antibodies (Sigma-Aldrich; 1:15 000). Signals were visualized using a chemiluminescence kit (Merck Millipore) and images were acquired using a CCD camera (LAS1000 system, Fuji Films). Optical density was measured on specific immunoreactive bands using NIH ImageJ Software. Membranes were then stripped and reprobed with β-actin antibodies (Sigma-Aldrich; 1:50 000). The optical density of specific bands was then normalized to the corresponding β-actin levels.

Statistical analysis

Behavioural data recorded on multiple testing sessions over a chronic treatment period were compared using repeated measures ANOVA and post hoc Bonferroni test. All remaining analyses were performed using either unpaired t-test or one-factor analysis of variance (ANOVA) and post hoc Bonferroni test, where appropriate. The level of statistical significance was set at P < 0.05. All data are expressed as mean ± standard error of the mean (SEM). The exact P- and F-values of the ANOVAs are given in the figure legends, and post hoc pairwise comparisons are presented in the ‘Results’ section as being either significant or non-significant.

Results

Chronic treatment with PRE-084 induces motor recovery in 6-hydroxydopamine-lesioned mice

Wild-type mice sustaining 6-hydroxydopamine lesions or sham lesions were treated daily with PRE-084 (0.3 or 1 mg/kg) or saline (‘saline’ group) for 35 days, starting on the same day as the lesion (n = 11–14 per group). Behavioural tests were performed in the last two days of each treatment week.

A comparison of spontaneous turning behaviour between groups and testing sessions revealed significant overall differences (Fig. 2A). Throughout the experiment, animals treated with 0.3 mg/kg PRE-084 showed significantly lower rotational counts than saline-treated 6-hydroxydopamine-lesioned controls. Mice treated with the higher dose of PRE-084 (1 mg/kg) differed significantly from saline-treated 6-hydroxydopamine-lesioned mice only in the first 2 weeks of treatment (P < 0.05 versus saline-treated group; Fig. 2A). By the end of the treatment, animals receiving either dose of PRE-084 displayed a very low rotational activity, which was comparable to the values measured in sham-lesioned controls (Fig. 2A, P < 0.05 for 0.3 mg/kg PRE-084 versus saline on Week 5).

Figure 2

Chronic treatment with PRE-084 induces behavioural motor recovery. Spontaneous rotations (A) and forelimb use asymmetry (cylinder test, B and stepping test, C) were assessed once a week in mice treated for 5 weeks with either PRE-084 (0.3/1 mg/kg) or saline solution. Sham lesioned mice were treated with either PRE-084 or saline solution, and then merged in the same group. Results are expressed as number of spontaneous ipsilateral rotations (A) or as a percentage of supporting wall contacts (B)/adjusting steps (C) performed with the paw contralateral to the lesion (left paw). Significant overall differences were found (A) in the counts of spontaneous ipsilateral rotations, (B) in the cylinder test and (C) in the stepping test. Repeated measures ANOVA, (A) Treatment F(3,44) = 21.84, P < 0.001; Time F(4,176) = 57.84, P < 0.001; Interaction F(12,176) = 7.54, P < 0.001; (B) Treatment F(3, 44) = 38.06, P < 0.001; Time F(4,176) = 8.72, P < 0.001; Interaction F(12,176) = 3.57, P < 0.001; (C); Treatment F(3,44) = 87.42, P < 0.001; Time F(4,176) = 15.48; Interaction F(12,176) = 3.35, P < 0.001; post hoc Bonferroni, P < 0.05 # versus Sham; € versus PRE0.3. In A, B and C the number of mice used was the following: saline (n = 11), PRE0.3 (n = 11), PRE1 (n = 12), Sham (n = 14). Saline = 6-hydroxydopamine-lesioned mice treated with saline solution for 35 days; PRE0.3 or 1 = 6-hydroxydopamine-lesioned mice treated with PRE-084 0.3 or 1 mg/kg for 35 days.

In the cylinder test, the percentage use of the contralateral forepaw showed significant overall differences between groups and testing sessions (Fig. 2B). All lesioned groups exhibited a significant forelimb use asymmetry compared to the sham-lesion controls in the first 3 weeks of treatment (Fig. 2B). However, the group receiving 0.3 mg/kg PRE-084 showed a gradual improvement in contralateral forepaw usage, finally reaching levels of performance comparable with the sham-lesion controls (P < 0.05 versus saline and > 0.05 versus sham in Week 5; Fig. 2B).

Significant overall differences between groups and testing sessions were also found in the stepping test (Fig. 2C). In this test too, all 6-hydroxydopamine-lesioned mice showed a similar deficit during the first week, regardless of treatment allocation. Thereafter, animals treated with 0.3 mg/kg PRE-084 showed a reduction in forelimb stepping asymmetry, differing significantly from saline-treated 6-hydroxydopamine-lesioned controls from the second through the fifth week (Fig. 2C, P < 0.05 for 0.3 mg/kg PRE-084 versus both saline and sham in Weeks 2–5).

When treatment with PRE-084 was started 7 days post-lesion, it did not produce any significant motor improvement (Supplementary material).

Neuroprotective and neurorestorative effects of PRE-084 treatment

Stereological counts of TH-positive cells in the substantia nigra pars compacta revealed significant group differences (Fig. 3A). In saline-treated mice (Fig. 3A and B), the number of TH-positive cells on the side ipsilateral to the 6-hydroxydopamine lesion amounted to 44% of the values on the contralateral side (P < 0.05 versus sham-lesioned group). Mice treated with 0.3 mg/kg PRE-084 (Fig. 3A and C) showed a significantly larger value (∼62% of contralateral side; P < 0.05 versus both saline-injected mice and sham-lesioned group, Fig. 3A and D). Mice treated with 1.0 mg/kg PRE-084 showed a trend towards a higher number of TH-positive cells (∼53% TH-positive neurons on the lesioned side), not differing significantly from either the 0.3 mg/kg dose or saline treatment (but P < 0.05 versus sham-lesioned group, Fig. 3A).

Figure 3

Chronic treatment with PRE-084 induces neurohistological restoration. (A) Stereological counts of TH-positive cells were performed in the substantia nigra pars compacta (SNc). Values represent the mean ± SEM of seven sections per animal throughout the substantia nigra pars compacta and are expressed as percentage of lesion/intact side. Representative photomicrographs were taken under a ×10 objective from the substantia nigra of 6-hydroxydopamine lesioned mice treated with saline (B) or 0.3 mg/kg PRE-084 (C) and from a sham-lesion control (D). Image segmentation analysis of TH-positive fibres at high magnification (cf. photomicrographs F, G, I and J) was performed both in the dorsolateral striatum (E) and in the substantia nigra (H) of PRE-084 (0.3 or 1 mg/kg) and saline-treated mice. Results represent the mean ± SEM of four levels per animal throughout striatum or substantia nigra, respectively. Values are expressed as pixels/sample area analysed. Representative photomicrographs were taken under a ×100 objective from the dorsolateral (DL) striatum of mice treated with saline or the effective dose of PRE-084 (0.3 mg/kg) (F and G, respectively). Photomicrographs of the same treatment groups were taken also from the substantia nigra (I and J, respectively). Scale bar = 10 μm. One-way ANOVA, (A) [F(3,38) = 28.42, P < 0.001]; (E) dorsolateral striatum [F(2,24) = 4.07, P = 0.0300]; (H) substantia nigra [F(2,22) = 10.65, P = 0.0007]; post hoc Bonferroni, P < 0.05 *versus saline; #versus sham; °versus saline and PRE 1 mg/kg. In A, the number of mice used was the following: saline (n = 11), PRE0.3 (n = 11), PRE1 (n = 12), sham (n = 8). In B, the number of mice used was the following: saline (n = 8), PRE0.3 (n = 10), PRE1 (n = 9). In C, the number of mice used was the following: saline (n = 7), PRE0.3 (n = 8), PRE1 (n = 8). pxls = pixels; saline = 6-hydroxydopamine-lesioned mice treated with saline solution for 35 days; PRE0.3 or 1 = 6-hydroxydopamine-lesioned mice treated with PRE-084 0.3 or 1 mg/kg for 35 days.

Possible restorative effects of the treatment on the striatal dopaminergic innervation were first examined by measuring the overall optical density of TH immunostaining on low-magnification pictures. Because this 6-hydroxydopamine lesion model produces an increasing rostrocaudal and mediolateral gradient of TH fibre loss, the analysis was performed separately in the dorsolateral and dorsomedial striatum on four different rostrocaudal levels (cf. Fig. 1). The results of this analysis showed a significant increase in TH immunostaining by PRE-084 treatment only in regions most spared by the lesion (i.e. the dorsomedial striatum at rostral levels, where TH optical density was ≥60% of the contralateral side in saline-treated animals) (Supplementary material and Supplementary Table 1.1). Because TH optical density measurements on low-magnification pictures would not detect the presence of residual fibres in highly denervated regions, we applied an image segmentation method to high-magnification pictures from the dorsolateral striatum. We also examined the substantia nigra pars reticulata, where recovery of TH-positive neuritic processes can be seen following trophic factor delivery (Lindgren et al., 2012). This analysis was performed on 6-hydroxydopamine-lesioned mice treated with PRE-084 and saline-treated controls, but not on sham-lesioned mice due to an impossibility to resolve distinct fibres in this group. The density of TH-positive neuritic processes in the dorsolateral striatum and substantia nigra pars reticulata was ∼4-fold and 1.5-fold, respectively, in mice treated with PRE-084 (0.3 mg/kg) compared with saline-treated animals (P < 0.05, Fig. 3E–G and 3H–J). The higher dose of PRE-084 (1 mg/kg) did not significantly increase TH fibre density in either region (P > 0.05 versus saline, Fig. 3E and H).

Levels of dopamine, DOPAC (3,4-dihydroxyphenylacetic acid) and homovanillic acid striatal tissue levels were determined in separate groups of 6-hydroxydopamine-lesioned mice where only the most effective dose of PRE-084 was tested (0.3 mg/kg/day for 35 days). In this experiment too, treatment with 0.3 mg/kg PRE-084 significantly improved behavioural deficits in spontaneous rotations, cylinder test and stepping test (P < 0.05 for 0.3 mg/kg PRE-084 versus saline treatment in all tests on the fifth week; ANOVA and post hoc Bonferroni test; data not shown). In the dorsolateral striatum, levels of dopamine and its metabolites were depleted by at least 80% in all 6-hydroxydopamine-lesioned animals (Table 1). Mice treated with 0.3 mg/kg PRE-084 exhibited an 8-fold increase in dopamine levels compared with the values measured in saline-treated animals, although the degree of dopamine depletion remained quite pronounced compared with the sham-lesioned group (P < 0.05 for 0.3 mg/kg PRE-084 versus both saline-treated 6-hydroxydopamine-lesioned mice and sham group, Table 1). In addition, mice treated with 0.3 mg/kg PRE-084 showed a reduction by 36% in dopamine turnover rate (P < 0.05 versus both saline and sham-lesioned controls, Table 1). A reduced dopamine turnover is in-line with other studies reporting such an effect following treatment with neuroprotective agents in rodent Parkinson’s disease models (Lange et al., 1994; Muralikrishnan and Mohanakumar, 1998).

View this table:
Table 1

Striatal tissue levels of dopamine, 5-HT and their metabolites

Wild-type mice, 35 days treatmentDorsolateral striatumDorsomedial striatum
SalinePRE0.3ShamSalinePRE0.3Sham
Dopamine0.05 ± 0.02#0.40 ± 0.21*#4.59 ± 0.682.04 ± 0.292.37 ± 0.334.44 ± 0.54
DOPAC0.12 ± 0.04#0.35 ± 0.15#2.59 ± 0.322.13 ± 0.42#2.01 ± 0.394.48 ± 0.39
HVA0.42 ± 0.07#0.81 ± 0.20#2.19 ± 0.271.56 ± 0.25#1.47 ± 0.22#3.05 ± 0.29
DOPAC + HVA/DA3.08 ± 0.33#1.98 ± 0.38*#0.058 ± 0.051.73 ± 0.131.50 ± 0.131.82 ± 0.16
5-HT0.76 ± 0.081.22 ± 0.18*0.98 ± 0.061.69 ± 0.261.90 ± 0.171.57 ± 0.16
5-HIAA0.47 ± 0.050.72 ± 0.090.54 ± 0.050.84 ± 0.100.94 ± 0.0090.90 ± 0.07
5-HIAA/5-HT0.62 ± 0.030.61 ± 0.030.54 ± 0.030.55 ± 0.040.51 ± 0.020.57 ± 0.01
  • Values represent the monoamine concentrations as pmol/mg of striatal tissue, expressed as mean ± SEM. Data have been log-transformed for statistical analysis. One-way ANOVA, dorsolateral striatum, dopamine F(2,26) = 38.79, P < 0.001; DOPAC, F(2,26) = 24.57, P < 0.001; HVA, F(2,26) = 24.00, P < 0.001; DOPAC + HVA/DA, F(2,26) = 34.28, P < 0.001; 5-HT, F(2,26) = 3.55, P = 0.0433; 5-HIAA, F(2,26) = 3.092, P = 0.0624; 5-HIAA/5-HT, F(2,26) = 2.88, P = 0.0749; dorsomedial striatum, dopamine, F(2,27) = 2.49, P = 0.10; DOPAC, F(2,27) = 6.28, P = 0.0057; HVA, F(2,27) = 6.44, P = 0.0051; DOPAC + HVA/dopamine, F(2,26) = 1.20, P = 0.31; 5-HT, F(2,27) = 0.57, P = 0.56; 5-HIAA, F(2,27) = 0.57, P = 0.56; 5-HIAA/5-HT, F(2,25) = 1.19, P = 0.31. One-way ANOVA and post hoc Bonferroni; P < 0.05 *versus saline, #versus sham.

  • DOPAC = 3,4-dihydroxyphenylacetic acid; HVA = homovanilic acid; 5-HT = 5-hydroxytryptamine; 5-HIAA = 5-hydroxyindoleacetic acid; saline = 6-OHDA-lesioned mice treated with saline solution; PRE0.3 = 6-OHDA-lesion mice treated with PRE-084 0.3 mg/kg; sham = sham-lesion mice treated either with saline or PRE-084 0.3 mg/kg.

Treatment with PRE-084 also increased 5-HT and 5-HIAA tissue contents in the dorsolateral striatum achieving a statistically significant effect on 5-HT levels (P < 0.05 versus saline, Table 1).

In the region with most dopamine sparing (the dorsomedial striatum), neither dopamine levels nor dopamine turnover rate were significantly modified following PRE-084 treatment (P > 0.05 versus saline, Table 1).

Chronic treatment with PRE-084 upregulates neurotrophic factors and activates ERK1/2 and Akt

The gradual behavioural improvement and significant increase in TH fibre density induced by PRE-084 were suggestive of a neurorestorative effect promoted by trophic factors. To explore this hypothesis, additional mice were subjected to 6-hydroxydopamine lesion, followed by 35-days treatment with PRE-084 (0.3 mg/kg) or saline (n = 6 per group). Mice were prepared for western blot analysis of GDNF (Fig. 4A–D) and BDNF (Fig. 4E–H) using striatal and nigral protein extracts.

Figure 4

Treatment with an effective dose of PRE-084 increases striatal and nigral protein levels of GDNF and BDNF. Western immunoblotting analysis for GDNF (A and D) and BDNF (E and G) on striatal and nigral protein extracts from 6-hydroxydopamine-lesioned mice treated with PRE-084 (0.3 mg/kg) or saline for 35 days. Unpaired t-test, GDNF (B) striatum P = 0.023, (D) substantia nigra P = 0.0031; BDNF (F) striatum P = 0.0036, (H) substantia nigra P = 0.47. In B and D, the number of mice used was the following: saline (n = 6), PRE0.3 (n = 6).

In the striatum, animals treated with PRE-084 showed a small but significant increase in GDNF levels (∼6% above saline treatment, P < 0.05; Fig. 4A and B) and a 3-fold increase in BDNF levels (P < 0.05 versus saline, Fig. 4E and F). In the substantia nigra, only GDNF protein levels were significantly upregulated in PRE-084-treated mice (∼18% above the saline-treated group, P < 0.05; Fig. 4C and D).

In the same animals, we measured the levels of phosphorylated (active) ERK1/2 and Akt (protein kinase B), which are major intracellular mediators of trophic factor effects (Namikawa et al., 2000; Garcia-Martinez et al., 2006; Ries et al., 2006). Mice treated with PRE-084 showed an increase in pERK1/2 levels by ∼40% in both the striatum (Fig. 5A and B) and the substantia nigra (Fig. 5C and D) (P < 0.05 versus saline in both structures). The levels of pAkt were similarly upregulated in the striatum (P < 0.05 for PRE-084 versus saline, Fig. 5E and F) but not in the substantia nigra (P > 0.05, Fig. 5G and H).

Figure 5

Treatment with an effective dose of PRE-084 increases striatal protein levels of pERK1/2 and pAkt and nigral protein levels of pERK1/2. Western blots for pERK1/2 (A–D), totERK1/2 (B’ and D’), pAkt (E–H) and totAkt (F’ and H’) were done on striatal (A and E) and nigral protein extract (C and G) of saline and PRE-084 0.3 mg/kg treated mice. Unpaired t-test, pERK, (B) striatum P = 0.014, (D) substantia nigra P = 0.049; ERK1/2, (B’) striatum P = 0.48, (D’) substantia nigra P = 0.13; (F) striatum pAkt, P = 0.013; (F’) striatum totAkt P = 0.088; (H) substantia nigra, pAkt P = 0.79; (H’) totAkt P = 0.238. In B, D, B’, D’, F, F’, H and H’ the number of mice used was the following: saline (n = 6), PRE0.3 (n = 6). totERK = total ERK1/2; pERK = pERK1/2; totAkt = total Akt.

PRE-084 treatment reduces microglial activation

The intrastriatal injection of 6-hydroxydopamine is accompanied by sustained microglial activation, occurring both locally and in the substantia nigra (Cicchetti et al., 2002). At 7 days post-lesion, a massive expression of CD68-positive cells occurred in the dorsolateral part of the lesioned striatum in both saline- and PRE-084-treated mice, corresponding to a 23-fold and 13-fold increase, respectively, above the values measured on the contralateral side (but P > 0.05 for saline versus PRE-084; Fig. 6A and B). A clear microglial activation was observed also in the ipsilateral substantia nigra in both groups (∼145% and 116% of intact side in saline- and PRE-084-treated mice, respectively, Fig. 6D and E). Treatment with 0.3 mg/kg PRE-084, however, achieved a statistically significant, 20% decrease in CD68 expression in this structure (P < 0.05 versus saline, Fig. 6F).

Figure 6

Treatment with PRE-084 (0.3 mg/kg) reduces microglial activation in the substantia nigra of 6-hydroxydopamine-lesioned mice. Computer-assisted counts of CD68 positive cells from four sections per animal were performed in the striatum (C and I) and in the substantia nigra (F and L) of animals receiving 0.3 mg/kg PRE-084 or saline solution for 7 or 35 days, respectively. Values are expressed as percentage of lesion/intact side (mean ± SEM). Representative photomicrographs were taken under a ×10 objective from the striatum (A, B, G, H) and the substantia nigra (D, E, J, K) of saline and PRE-084 treated mice. Scale bar = 0.1 mm. Unpaired t-test, 7 days treatment, striatum P = 0.303; substantia nigra P = 0.0395. The number of animals used in C and F is the following: saline (7 days) (n = 8), PRE0.3 (7 days) (n = 7). Unpaired t-test 35 days treatment, striatum P = 0.0007; substantia nigra P = 0.0069. The number of animals used in I and L is the following: saline (35 days) (n = 11), PRE0.3 (35 days) (n = 10).

By 35 days post-lesion, the microglial activation in the lesioned striatum had declined, and the number of CD68-positive cells amounted to ∼140% of the values measured on contralateral side (Fig. 6G and I). By this time point, treatment with PRE-084 had significantly reduced the number of active microglial cells in the lesioned striatum by ∼40% compared to saline treatment (P < 0.05, Fig. 6H and I). In the ipsilateral substantia nigra, the number of CD68-positive microglial cells amounted to ∼129% and 105% of the contralateral side in saline-treated and PRE-084-treated mice, respectively, and the reduction in the PRE-084-treated group was statistically significant (Fig. 6J and L; Fig. 6K and L).

Treatment with PRE-084 is not effective in sigma-1 receptor knockout mice

To verify the treatment’s target specificity, the effective dose of PRE-084 was tested in Sig-1R knockout mice. Sig-1R knockout mice sustained 6-hydroxydopamine lesions or sham lesions and were treated with 0.3 mg/kg PRE-084 or saline for 35 days. Knockout mice responded to the 6-hydroxydopamine lesion with features of spontaneous rotation and impaired forelimb use quite similar to those found in wild-type mice (cf. Fig. 2, Supplementary Tables 1.1 and 1.2). Sig-1R knockout-lesioned mice treated with 0.3 mg/kg PRE-084 or saline showed similar levels of spontaneous rotational asymmetry and forelimb use asymmetry (Fig. 7A–C; P < 0.05 versus sham-lesioned controls for both treatment groups). Treatment with PRE-084 did not have any restorative effects on either striatal TH optical density (Supplementary material and Supplementary Table 1.2) or TH fibre density in highly denervated regions (Fig. 7L and O). Striatal and nigral levels of GDNF and BDNF remained unaffected (Fig. 7D–K).

Figure 7

Treatment with PRE-084 is ineffective in Sig-1R knockout mice. Sig-1R knockout mice treated for 35 days either with PRE-084 (0.3 mg/kg) or saline solution underwent weekly tests for rotational behaviour (A) and forelimb use asymmetry (cylinder test, B and stepping test, C). The results are expressed as number of spontaneous ipsilateral rotations (A) or as percentage of supporting wall contacts (B)/adjusting steps (C) performed with the paw contralateral to the lesion (left paw). Western immunoblotting analysis for GDNF (D and G) and BDNF (H and K) on striatal and nigral protein extracts from 6-hydroxydopamine-lesioned mice treated with PRE-084 (0.3 mg/kg) or saline. Image segmentation analysis of TH-positive fibres was performed both in the dorsolateral striatum (L–N) and substantia nigra (O–Q) of 6-hydroxydopamine-lesioned Sig-1R knockout mice treated with saline or PRE-084 0.3 mg/kg. Repeated measures ANOVA, (A) spontaneous rotations, treatment F(2,18) = 28.03, P < 0.001, Time F(4,72) = 51.06, P < 0.001, Interaction F(8,72) = 16.68, P < 0.001; (B) Cylinder test, treatment F(2,18) = 22.65, P < 0.001, Time F(4,72) = 0.619, P = 0.64, Interaction F(8,72) = 0.839, P = 0.59; (C) Stepping test, treatment F(2,18) = 29.50, P < 0.001, Time F(4,72) = 1.374, P = 0.251, Interaction F(8,72) = 0.313, P = 0.958; post hoc Bonferroni, P < 0.05 # versus sham. In A–C, the number of mice used was the following: saline (n = 5), PRE0.3 (n = 6), sham (n = 11). Unpaired t-test (E, G, I, K, L, O), P > 0.05 versus saline. In E, G, I, K, L and O, the number of mice used was the following: saline (n = 5), PRE0.3 (n = 6). W1–5 = Week 1–5; PRE 0.3/1 = 6-hydroxydopamine-lesioned mice treated with PRE-084 0.3/1 mg/kg; saline = 6-hydroxydopamine-lesioned mice treated with saline solution; sham = sham-lesion mice receiving either PRE 0.3 mg/kg or saline solution; ko = knockout.

Sigma-1 receptor immunohistochemistry offers clues about target localization and regulation

Dual-antigen immunofluorescence and confocal microscopy on striatal and nigral sections from wild-type mice revealed that Sig-1R was expressed in both neurons (Fig. 8A–F) and astrocytes (Fig. 8G–L), while it did not appear to colocalize with a marker of active microglia (CD68; not shown). In the striatum, neuronal expression was weak (Fig. 8A–C), whereas nearly all GFAP-positive astrocytes displayed clear Sig-1R immunoreactivity (Fig. 8I). Conversely, in the substantia nigra (both pars compacta and pars reticulata), Sig-1R immunoreactivity was more prominent in neurons than in astrocytes (cf. Fig. 8D–F and Fig. 8J and L). Dual-antigen immunofluorescence for Sig-1R and DAT confirmed the expression of this protein in nigral dopaminergic neurons (Fig. 9A–H’).

Figure 8

Sig-1R expression in neurons and astroglia. Photomicrographs of striatal (A–C, G–I) and nigral sections (D–F, J–L) dually immunostained for Sig-1R, NeuN (neuronal marker) and GFAP (astrocytic marker) were taken in a Zeiss confocal microscope under a ×40 objective. (A–C) Sig-1R and NeuN, striatum. (D–F) Sig-1R and NeuN, substantia nigra. (G–I) Sig-1R and GFAP, striatum. (J–L) Sig-1R and GFAP, substantia nigra. Scale bar = 20 μm. The animals represented here had sustained 6-hydroxydopamine lesion and saline treatment.

Figure 9

Subcellular redistribution of Sig-1R in neuronal and astroglial cells of PRE-084-treated animals. Representative photomicrographs of nigral (A–H’) or striatal (I–P’) sections dually immunostained for Sig-1R and DAT (dopamine transporter), or Sig-1R and GFAP (astrocytic marker) were taken in a Zeiss confocal microscope under a ×40 objective (A–C, E–G, I–K, M–O) or ×63 objective (D, D’, H, H’, L, L’, P, P’). The upper row in each panel shows saline-treated animals (A–D’ or I–L’), whereas the lower row shows PRE-084 treated mice (E–H’ and M–P’). Scale bar = 20 μm. The animals represented here had sustained 6-hydroxydopamine lesion, and saline or PRE-084 treatment at the dose of 0.3 mg/kg/day had been given for 35 days.

Although treatment with PRE-084 did not affect the total number of Sig-1R-immunoreactive cells (Supplementary Fig. 3.1), it appeared to alter the subcellular localization of this protein in both neurons (Fig. 9, cf. D-D’ with H-H’) and astrocytes (Fig. 9, cf. L-L’ with P-P’). As exemplified by the Sig-1R/DAT co-stained specimens (Fig. 9A–H’), PRE-084-treated mice showed widespread Sig-1R immunoreactivity along the neuronal processes (Fig. 9G and H’), whereas Sig-1R immunoreactivity was more prominent in the cell bodies compared to the cell processes in saline-treated animals (Fig. 9C and D’). In the striatum (where Sig-1R immunoreactivity was predominantly astrocytic), localization of Sig-1R in astrocytic processes was more frequently encountered following PRE-084 treatment (Fig. 9 cf. O and P’ versus K and L’). The increased levels of Sigma-1 immunoreactivity in cellular processes relative to somatic (perinuclear) regions suggests that the treatment with PRE-084 had stimulated the intracellular trafficking of this protein (discussed below).

Discussion

This is the first study addressing the expression and regulation of the Sig-1R in the nigrostriatal pathway, and the possibility to target this protein to achieve motor recovery and neurorestoration in a parkinsonian animal.

Daily treatment of unilaterally 6-hydroxydopamine-lesioned mice with the selective sigma-1 agonist PRE-084 significantly improved asymmetries in postural control and forelimb use. The recovery of contralateral forelimb use had a gradual course. In the cylinder test, no improvement was detected during the first three treatment weeks, but the final performance of mice treated with 0.3 mg/kg PRE-084 did not differ from that of sham-lesioned animals. An improvement in spontaneous rotation was induced by both doses of PRE-084 (0.3 and 1 mg/kg) already during the first week of treatment. The different pattern of treatment-induced recovery in rotational behaviour versus forelimb use may reflect differences in the neural control of these two responses. Indeed, forelimb use is dependent on a dopaminergic activity in the lateral striatum (Winkler et al., 1996; Chang et al., 1999; Tillerson et al., 2001; Lane et al., 2006), whereas a dopaminergic stimulation restricted to the substantia nigra may be sufficient to elicit rotation or locomotor activity (Robertson and Robertson, 1988; Bergquist et al., 2003). It is therefore relevant to point out that both doses of PRE-084 exerted a modest neuroprotective effect on nigral dopamine neurons at both 7 and 35 days, whereas only the 0.3 mg/kg dose produced an increase in dopamine fibre density in the lateral striatum upon chronic treatment.

The results of the delayed start experiment (Supplementary material) show that neither behavioural nor histological restoration can be achieved if treatment with PRE-084 starts 1 week post-lesion. In 6-hydroxydopamine-lesioned rodents, the first week post-lesion represents a critical window of opportunity for neurorestorative interventions that promote endogenous plasticity mechanisms (Tillerson et al., 2001). These mechanisms may involve, for example, a stimulation of axon terminal and spine growth, enabling the restoration of synaptic contacts (van Waarde et al., 2011), which would be aided by reduced levels of neuroinflammation (Di Filippo et al., 2008). A neuroplasticity-boosting action is likely to be the main mechanism through which chronic treatment with PRE-084 improved behaviour. Indeed, the substantial motor recovery in mice treated with 0.3 mg/kg PRE-084 does not seem proportional to their modest degree of dopaminergic restoration, given that both TH fibre density and striatal dopamine levels remained quite below normal levels in these mice. A recent study in a parkinsonian mouse model has shown that physical exercise restores dendritic arborizations and spine density in striatal projection neurons even in the absence of significant dopaminergic neuroprotection (Toy et al., 2014). This study indicates that treatments restoring synaptic connectivity in Parkinson’s disease models may produce functional recovery independent of an increase in dopamine levels (Toy et al., 2014).

Supporting a neuroplasticity-boosting effect of PRE-084 treatment, western immunoblotting analysis of striatal and nigral protein extracts revealed a significant upregulation of both GDNF and BDNF in animals treated with 0.3 mg/kg PRE-084 compared to saline-treated controls. This was accompanied by higher levels of phosphorylated ERK1/2 and Akt, the main survival and plasticity pathways activated by tyrosine kinase receptors (Brunet et al., 2001; Onyango et al., 2005; Lee et al., 2006; Villegas et al., 2006) (Supplementary material). In light of previous literature, our findings suggest that the increased TH fibre density induced by 0.3 mg/kg PRE-084 in the dorsolateral striatum and substantia nigra pars reticulata reflects a neurotrophin-mediated neuritogenic effect (Lindgren et al., 2012). In addition, the upregulation of BDNF may underlie the increased striatal levels of serotonin produced by treatment with PRE-084 (0.3 mg/kg), because serotonin neurons are very responsive to BDNF (for review see Mattson et al., 2004) (Supplementary material).

Previous studies have suggested that Sig-1R agonists have anti-inflammatory properties (Griesmaier et al., 2012; Mancuso et al., 2012). Counts of cells with microglial morphology and immunoreactive for CD68 (which is expressed in activated microglia and macrophages) (Perry and Teeling, 2013) indicate that treatment with 0.3 mg/kg PRE-084 had attenuated the 6-hydroxydopamine lesion-induced microglial activation in both the striatum and the substantia nigra after a 35-day treatment, (Fig. 6I and L). In mice sustaining intrastriatal 6-hydroxydopamine lesions, a pronounced microglial activation in these structures contributes to the neurodegenerative process during the rapid phase of dopamine cell death (Alvarez-Fischer et al., 2008), and then persists for weeks (Cicchetti et al., 2002). A reduction of microglial activation by PRE-084 may have contributed to some of the effects observed. In particular, the inefficacy of PRE-084 treatment in the ‘delayed start experiment’ may partly reflect a failure to dampen the lesion-induced neuroinflammatory response that is triggered by the initial toxic insult but then becomes self-sustaining (Cicchetti et al., 2002).

Expression and regulation of sigma-1 receptor in the nigrostriatal dopamine system

The Sig-1R is widely distributed in the brain (Gundlach et al., 1986; Alonso et al., 2000). However, its expression and regulation in the nigrostriatal system have not been previously addressed. Immunohistochemistry for Sig-1R was performed in both striatal and nigral sections. The protein was found to be expressed in both neurons and astrocytes, in agreement with previous studies (Hayashi and Su, 2003a; Ruscher et al., 2011). The 6-hydroxydopamine lesion was found to increase the total number of Sig-1R-immunoreactive cells in the ipsilateral striatum (though not significantly in the substantia nigra) at 7 days post-surgery (cf. Supplementary material). Although treatment with PRE-084 did not modulate the number of Sig-1R-immunoreactive cells, it affected the protein’s subcellular distribution, inducing higher Sig-1R immunoreactivity in cellular processes relative to the perinuclear region. These observations are consistent with previous studies in animals treated with Sig-1R agonists, showing relatively high levels of Sigma-1 immunoreactivity over the entire endoplasmic reticular structure and cellular processes both in neurons and glial cells (Hayashi and Su, 2003b; Ruscher et al., 2011). Accordingly, in vitro studies have revealed that Sig-1R agonists cause this protein to dissociate from the mitochondrion-associated endoplasmic reticulum membrane and redistribute widely within the cell (Su et al., 2010). The treatment-induced redistribution of Sig-1R appears to be linked to an increased intracellular trafficking of lipid rafts and biomolecules required for brain repair (Ruscher et al., 2011). It is also most likely linked to the increased intracellular processing and secretion of neurotrophins, as observed following Sigma-1 agonist application to cell lines (Fujimoto et al., 2012).

Concluding remarks

Our results are the first to show that chronic pharmacological stimulation of Sig-1Rs induces substantial behavioural improvement in a parkinsonian animal model. The overall pattern of effects produced by PRE-084 treatment, along with the modest degree of dopamine restoration, suggest that the functional improvement relied on a stimulation of general, intrinsic mechanisms of cell repair and plasticity, possibly favoured by a mild anti-inflammatory action. Significant neurorestorative effects were seen only with one of the tested doses (0.3 mg/kg PRE-084). Sig-1R agonists usually display a bell-shaped dose-response curve (Maurice and Su, 2009; Villard et al., 2011). This property has, however, not prevented the initiation of clinical trials with this class of compounds (Ishikawa and Hashimoto, 2010) (see also clinical trial NCT01832285), and new Sig-1R agonists are currently being developed for therapeutic purposes by several groups (partly reviewed in Kourrich et al., 2012).

The intrastriatal 6-hydroxydopamine lesion model accurately mimics both the heterogeneous pattern and the biphasic time course of nigrostriatal degeneration in Parkinson’s disease, although within a much compressed time frame (Supplementary material). However, the suitability of this lesion model to test neuroprotective/restorative treatments is often questioned, particularly after recent reports that the neurorestorative action of GDNF on the nigrostriatal pathway is abrogated by α-synuclein overexpression in dopamine neurons (Decressac et al., 2012), a feature that is not present in 6-hydroxydopamine-lesioned animals. Nevertheless, putaminal GDNF infusion can induce dopaminergic activity in nigrostriatal axon terminals in patients with Parkinson’s disease (reviewed and discussed in Cenci and Brooks, 2013). Moreover, several documented effects of Sig-1R agonist, such as the facilitation of intracellular trafficking and inter-organelle signalling (Hayashi and Su, 2003b; Su et al., 2010), could potentially alleviate mechanisms of α-synuclein toxicity (reviewed in Wales et al., 2013). Finally, pharmacological treatments that promote endogenous plasticity mechanisms have been shown to improve both behavioural and pathological features in transgenic α-synuclein overexpressing mice (Kohl et al., 2012; Ubhi et al., 2012). Thus, the results obtained in this study call for further investigations on the neurorestorative potential of Sig-1R agonists using also other experimental models of parkinsonian disorders.

Several compounds with agonistic activities at the Sig-1R are already approved for clinical use and seem to be well tolerated (reviewed in Ishikawa and Hashimoto, 2010). Interestingly, some symptomatic medications used in Parkinson’s disease have agonistic activity at the Sig-1R (Peeters et al., 2004; van Waarde et al., 2011). Our results raise the important question, whether such medications may have disease-modifying properties if administered early during the course of Parkinson’s disease.

Funding

This study was supported by grants to M.A.C. from the Michael J Fox Foundation for Parkinson's Research, the Swedish Parkinson Foundation, The Swedish Research Council (project no. K2012-61X-13480-13-5), and European Community's Seventh Framework Programme FP7/2008 under grant agreement no. 215618 (project acronym, Neuromodel), the Basal Ganglia Disorders Linnaeus Consortium (BAGADILICO), Greta and Johan Kocks Foundation, Åhlen Foundation.

Supplementary material

Supplementary material is available at Brain online.

Acknowledgements

The authors warmly thank Ann-Christin Lindh and Natallia Maslava for their excellent technical assistance.

Abbreviations
ERK1/2
extracellular signal-regulated kinases
PRE-084
2-morpholin-4-ylethyl 1-phenylcyclohexane-1-carboxylate
Sig-1R
sigma-1 receptor

References

View Abstract