The role of TWEAK/Fn14 signaling in the MPTP-model of Parkinson’s disease

Highlights • We investigate the role of TWEAK and Fn14 in a model of Parkinson’s disease.• Ablation of TWEAK or Fn14 had no effect on acute MPTP toxicity.• TWEAK neutralizing antibody provided neuroprotection in the sub-acute MPTP-model.• Suggestion of a possible role for TWEAK in Parkinson’s disease.


INTRODUCTION
Parkinson's disease (PD) is one of the main neurodegenerative disorders and is symptomatically characterized by debilitating movement impairments. The main underlying pathology of PD is the degeneration of dopaminergic neurons in the nigrostriatal pathway. Systemic administration of the neurotoxin 1-methyl-4-phe nyl-1,2,3,6-tetrahydropyridine (MPTP) causes a similar nigrostriatal-selective dopaminergic cell loss in the mouse brain and therefore provides a suitable animal model to study PD neuropathology. Mechanisms such as microglial activation, oxidative stress and inflammatory cytokine release can mediate dopaminergic cell loss and their roles in PD neuropathology have been studied using the MPTP model (Litteljohn et al., 2010).
More recently characterized members of the TNF superfamily include the TNF-like weak inducer of apoptosis (TWEAK) and its fibroblast growth factorinducible 14 (Fn14) receptor (Chicheportiche et al., 1997;Wiley et al., 2001). TWEAK is a soluble cytokine widely expressed in neurons, astrocytes and macrophages and it acts as a ligand for the Fn14 receptor, the expression of which is up-regulated after tissue injury, including in injured hepatic, vascular and neuronal cells, and in tumors (Desplat-Jego et al., 2005;Winkles, 2008). TWEAK-Fn14 signaling has been linked to NFjB nuclear translocation and signal transduction (Brown et al., 2003;Tran et al., 2005). Furthermore, NFjB controls transcription of the gene for cycoloxygenase-2 (COX-2) which underlies PD neuropathology as demonstrated in the MPTP mouse model and observed in human PD tissue (Teismann et al., 2003). Thus, proinflammatory cytokines such as TWEAK can activate NFjB and its transcriptional control of downstream inflammatory mediators such as COX-2 and can therefore initiate and/or further exacerbate the neuroinflammatory and degenerative processes.
Neuronal death and neuroinflammatory processes have long been implicated in the pathogenesis of PD. Substantial evidence demonstrates that TWEAK-Fn14 signaling mediates neuroinflammatory processes and apoptotic cell death in in vitro and in vivo models of cerebral edema, ischemic stroke and multiple sclerosis (reviewed by Yepes (2007)). In models of pathological conditions such as ischemic stroke, chronic injury is associated with up-regulation of TWEAK and Fn14 expression (Potrovita et al., 2004;Inta et al., 2008). The neuronal cell death observed in these in vivo ischemia models, as well as in primary culture of cortical neurons undergoing oxygen-glucose deprivation, is associated with the observed increase in TWEAK and subsequent binding to Fn14 and activation of NF-jB expression (Potrovita et al., 2004;Polavarapu et al., 2005). Additionally it was shown, that the observed increase in the expression of monocyte chemoattractant protein-1 (MCP-1) and the recruitment of neutrophils after middle cerebral artery occlusion was absent in TWEAK À/À and Fn14 À/À mice (Haile et al., 2010). Reactive microglia and astrocytes mediate inflammation processes that contribute to the neurodegenerative process in PD Teismann and Schulz, 2004) and TWEAK has been shown to act on and mediate pro-inflammatory cytokine expression in astrocytes (Saas et al., 2000).
To date, little is known about the involvement of the TWEAK Fn-14 signaling complex in PD neurodegeneration. The aim of the present study is to determine the role of TWEAK-Fn14 in PD neuropathology as modeled using the MPTP neurotoxin in mice. We observed the effect of genetic ablation of TWEAK and Fn14 and neutralizing of TWEAK on MPTP-mediated neuropathologies such as dopaminergic cell loss in the substantia nigra, as well as dopamine depletion and fiber degeneration in the striatum.
For the neutralizing antibody experiments, adult male wild-type C57BL/6 mice weighing 20-25 g (Charles River Laboratories, Ormiston, UK) were treated with either an acute (4 Â 18 mg/kg, i.p., 2 h apart) or sub-acute regimen of MPTP (5 Â 30 mg/kg, i.p., over five consecutive days) and were injected i.p. with 200 lg anti-mouse TWEAK neutralizing antibody (Biogen Idec, Inc.) or isotype control antibody (Biogen Idec) 30 min prior to MPTP treatment (Potrovita et al., 2004). In the latter study the authors showed that the antibody crosses the blood-brain barrier. For striatum monoamine HPLC measurements and substantia nigra and striatal TH immunohistochemistry, mice undergoing the acute MPTP-regimen were killed 7 days after MPTP injection, mice undergoing the sub-acute MPTP-regimen 21 days after MPTP injection.
For endogenous TWEAK protein expression experiments, adult male wild-type C57BL/6 mice weighing 20-25 g (Charles River Laboratories, Ormiston, UK) were treated with an acute regime of MPTP (4 Â 18 mg/kg, i.p., 2 h apart). Control mice received saline-vehicle. Sub-groups of mice were sacrificed at various time points following MPTP administration (0, 1, 2, 4, 7, 14 and 21 days). All procedures complied with the European Community Council Directive (2010/63/UE) and the Animals (Scientific Procedures) Act 1986 and were approved by the Home Office or the Regierungspra¨sidium Karlsruhe, Germany. Mice were housed in appropriately sized cages with access to food and water ad libitum, with a 12-hour light-dark cycle (lights on at 7 am).

Human samples
Human nigral tissue was obtained from the UK Parkinson's Disease Tissue Bank at Imperial College, London. Selected PD patients and controls were matched for age at death. Controls were (n = 5) 80.4 ± 2.66 years at death (mean ± SEM) and PD patients (n = 6) 79 ± 2.02 years with an average disease duration of 9.08 ± 1.20 years (mean ± SEM). The age at onset of disease seen was 69.83 ± 2 years. None of the PD patients had a family history of the disease. Samples were isolated in NP-40 buffer with protease inhibitors 1:20 (wt/vol) by hand homogenization. Extracts were centrifuged at 14,000 rpm (18,620Âg;Mikro 200R) for 20 min at 4°C and supernatants retained. All procedures were approved by the responsible ethics committee (North of Scotland Research Ethics Committees).

Western blot analysis
The midbrain and striatum were dissected out from the mouse brain; total protein was isolated in ice-cold NP-40 buffer (20 mM Tris, 137 mM NaCl, 10% glycerol, 1% NP-40, 2 mM EDTA). The protein concentration was assessed using bicinchoninic acid (BCA) assay kit (Pierce, Rockford, IL, USA). 20 lg of protein (in Laemmli buffer and mercaptoethanol) was loaded onto a 12% SDS-polyacrylamide gel (SDS-PAGE) and transferred to a nitrocellulose membrane. After blocking with 5% milk powder in phosphate-buffered saline (PBS) containing 0.05% Tween-20 for 1 h the blots were incubated with the following primary antibodies: TWEAK (1:500, Cell Signalling) and b-actin (1:25,000, Sigma-Aldrich, UK) at 4°C overnight, then rinsed with 0.05% PBS-Tween followed by an incubation at room temperature for 1 h with a horseradish peroxidase (HRP)-conjugated secondary antibody (anti-rabbit or anti-mouse 1:10,000, Amersham). After washing in PBS-Tween the blots were exposed to a homemade ECL solution (Luminol sodium salt in 0.1 mM Tris HCl and Para-hydroxycoumarin in DMSO) for 1-2 min and then viewed in the Alpha Imager and quantified using the Alpha imaging software.
Using the above protocol Western blot analysis was also performed on post mortem human brain samples.

Immunohistochemistry studies
Mouse brains were dissected out and fixed in 4% paraformaldehyde (overnight at 4°C) and then transferred to 30% (w/v) sucrose solution. The brains were then frozen by immersion in isopentane and stored at À80°C. Coronal sections of frozen brains were cut (30 lm) in a cryostat. Ventral midbrain and striatal sections were collected in PBS with 0.01% sodium azide. The sections were washed with Tris-buffered saline (TBS) (three times for 5 min each), incubated with 10% methanol and 3% hydrogen peroxide, washed with TBS; incubated for 1 h in 5% normal goat serum followed by an incubation for 48 h with primary TH antibody (1:1000), rinsed with TBS (three times for 5 min each) and then incubated with biotinylated anti rabbit (1:200, Vector labs) antibody. After washes in PBS (three times for 5 min each) the sections were incubated for 1 h in HRP-avidin biotin complex (Vector), followed by washes in PBS (three times for 5 min each) and incubated for 15 min in diaminobenzidine (in 0.1 M Tris-HCl, pH = 7.6). All the sections were then mounted and dehydrated, counterstained with Nissl before viewing under a brightfield microscope. TH-and Nissl-positive cell counts in the substantia nigra pars compacta (SNpc) were quantified and estimated using the optical fractionator method (Stereoinvestigator, Microbrightfield Bioscience, Magdeburg, Germany) at 100Â magnification. Striatal fiber staining was quantified by comparing the average optical density (AOD) of staining in the striatum with the surrounding cortical tissue, using Scion Image software.
HPLC analysis of striatal monoamine levels HPLC with electrochemical detection was used to measure striatal levels of dopamine, DOPAC and serotonin using a method that has been described (Sathe et al., 2012;Nuber et al., 2008). Briefly, after mice were killed, striata were dissected out and snap frozen directly on dry ice. Striata were then homogenized in 1% perchloric acid (PCA) (1:30 [i.e. tissue weight: PCA volume]) and centrifuged at 14,000g at 4°C for 20 min. Following centrifugation, 20 ll of the supernatant sample was injected onto a C18 column (Dionex, Germering, Germany). The mobile phase consisted of 90% 50 mM sodium acetate, 35 mM citric acid, 105 mg/L octane sulfonic acid, 48 mg/L sodium EDTA solution, and 10% methanol at pH 4.3. Flow rate was 1 ml/min. Peaks were detected by an ESA Coulochem II electrochemical detector (ESA), and the detector potential was set at 700 mV. Data were collected and processed using the Chromeleon computer system (Dionex).

Statistical analysis
All values expressed are mean ± the standard error of the mean (SEM). Differences between means were analyzed using Student's t-test. Differences among means were analyzed using a one-way ANOVA. When ANOVA showed significant differences, pair-wise comparisons between means were assessed using the Newman-Keuls post hoc test. Null hypothesis was rejected at 0.05 levels. All analysis was performed using SPSS for windows and Graphpad Prism softwares.

TWEAK protein expression in the acute MPTP model and human PD samples
To determine the expression pattern of TWEAK, we looked at the protein levels of TWEAK in the ventral midbrains and striata of saline and MPTP-treated mice at different time points. No change in nigral protein levels of TWEAK was seen in acute MPTP-treated mice when compared to saline at any of the various time points after MPTP treatment (p > 0.05, ANOVA with Newman-Keuls post-hoc, Fig. 1A). However, an increase was seen in striatal levels of TWEAK protein on 2, 4 and 7 days after treatment with MPTP ( Ã p = 0.03, compared to saline, ANOVA with Newman-Keuls post-hoc, Fig. 1B).
The protein expression levels for TWEAK were also quantified in post-mortem human control and PD samples. No change was seen in TWEAK expression in either substantia nigra (Fig. 1C) or striatal (Fig. 1D) tissue of PD patients when compared to extracts from control human tissue.

Genetic ablation of TWEAK and Fn14 does not alter acute MPTP toxicity
According to HPLC analysis, TWEAK and Fn14 ablation had no significant effect on levels of dopamine in the striatum compared to their wild-type counterparts (Table 1).
In the SNpc, the number of TH-and Nissl-positive dopaminergic neurons in acute MPTP-treated TWEAK À/À or Fn14 À/À mice did not show any changes when compared to wild-type MPTP-treated mice (Table 1). Striatal TH-immunoreactivity appeared to be reduced in TWEAK À/À mice in response to MPTP as compared to wild-type MPTP-treated mice, however, the effect was not significant (Table 1). Thus, genetic ablation of TWEAK À/À or Fn14 À/À does not attenuate the effect of acute MPTP on SNpc dopaminergic cell counts and striatal dopaminergic fiber density.

Neutralizing TWEAK mitigates effects of sub-acute MPTP on nigral dopaminergic neurons
Having established that TWEAK levels in the striatum are altered by MPTP treatment (Fig. 1B), we assessed whether neutralization of TWEAK had an effect on MPTP-induced cell death. Neutralizing TWEAK did not lead to a change in SNpc cell counts of TH-positive dopaminergic neurons in mice dosed with acute MPTP at 18 mg/kg ( Fig. 2 and Table 1) or at a lower MPTP dose of 16 mg/kg (data not shown) when compared to cell counts from mice that had received MPTP alone. Using the less severe sub-acute regimen significantly more dopaminergic neurons survived after MPTPinduced cell death in mice receiving the TWEAK antibody compared to controls treated with the isotype control Ig (Fig. 2 and Table 1).
In HPLC analysis we did observe higher striatal dopamine levels caused by neutralizing TWEAK antibody treatment combined with MPTP although not significant (Table 1).

DISCUSSION
Using both genetic knockout and neutralizing antibody methods, we aimed to determine whether TWEAK-Fn14 signaling is involved in MPTP-induced effects in the substantia nigra (SN) and striatum of mice. We also measured TWEAK protein expression in substantia nigra and striatal tissue from MPTP-treated mice and PD patients. Our results show that TWEAK-neutralizing antibody leads to attenuation of MPTP-induced dopaminergic cell death in the SNpc.    Table 1). Three weeks after acute MPTP treatment, numbers of TH-positive neurons are not significantly different between mice receiving isotype control Ig or TWEAK neutralizing antibody (middle panel, Table 1); whereas after sub-acute MPTP-treatment mice initially receiving also TWEAK-neutralizing antibody treatment show higher numbers of TH-positive neurons, than do mice who received isotype control Ig (lower panel, Table 1). Striatal optical density was comparable between salinetreated mice receiving TWEAK-neutralizing antibody and isotype control Ig (upper panel, Table 1). Treatment with TWEAK neutralizing antibody had no effect on MPTP-induced loss of striatal optical density in the acute (middle panel, In both human and mouse substantia nigra tissue, no significant differences in TWEAK protein levels were observed between PD patients vs. healthy controls and MPTP-treated mice vs. saline-treated mice. However, in striatal tissue of MPTP-treated mice, there was increased TWEAK protein levels detected 2 days after treatment. This increase in striatal TWEAK protein was detected also on days 4 and 7 after MPTP-treatment. In post-mortem human striatal tissue from PD patients, we observed a trend of increased TWEAK protein in comparison to healthy controls, which if it were a statistically significant effect would have mimicked the transient striatal TWEAK expression seen in the mouse MPTP model. The expression levels of TWEAK and Fn14 proteins were also measured in human cortex samples obtained from healthy controls and PD patients with PINK and IPD mutations (data not shown). There was no difference in protein expression of either TWEAK or Fn14 between controls and the hereditary mutation samples suggesting that TWEAK and Fn14 expression changes are not associated with these hereditary forms of the disease and/or are not observed in the cortex.
It is interesting to observe at first view that TWEAK antibody is able to protect nigral dopaminergic neurons partially against MPTP-toxicity, whereas TWEAK as well as Fn14 ablation fails to provide any significant benefit. Different factors need to be kept in mind, which could have contributed to the observed differences. First of all, in the antibody study, a sub-acute regimen of MPTP administration was used. This model is linked closer to apoptotic induced cell death (Tatton and Kish, 1997), than the acute model, where cell death occurs mainly via necrosis (Jackson-Lewis et al., 1995). Another possibility could be that ablation of TWEAK or Fn14 leads to compensatory mechanisms, which then result in the observed effect, or anti-TWEAK resulted in Fc-mediated killing of TWEAK-expressing neuroinflammatory cells such as microglia.
Although we did not investigate glial activation in this study, it is interesting to note that TWEAK has been shown to be expressed in glial cells and has been reported to induce apoptosis by interacting with endogenous TNF-a and TNF receptor 1 (TNFR1) (Schneider et al., 1999). TNF-a expression in microglial cells has been shown in PD patients . It has been demonstrated that TNFR1 ablation leads to neuroprotection in the MPTP-model of PD, suggesting the participation of TNF-a via TNFR1 in MPTPinduced cell death and in PD (Sriram et al., 2002;Ferger et al., 2004). As TNF-a and TNFR1 factors are mainly involved in apoptotic cell death, perhaps TWEAK as well as Fn14 ablation did not result in neuroprotection in the acute MPTP-model since cell death in the acute model mainly occurs via necrosis (Jackson-Lewis et al., 1995). Although a TWEAK antibody did provide partial neuroprotection in the sub-acute model, it was not a substantial protection, indicating, that the cell death processes in the MPTP-model are not mainly triggered via TWEAK. Although, TWEAK has been shown to interact with TNF-a, this seems not to be the main case in the MPTP-model. Also, we did not find a significant increase of TWEAK in the SNpc of PD patients.
Taking together our data suggest a minor role for TWEAK in PD. Still, when generating a neuroprotective therapy for the treatment of PD, one should keep in mind the potential role of TWEAK, and it might still prove to be a useful in a multi-target therapy, which in the end could be the way forward for the treatment of this debilitating disorder. DISCLOSURES L.B. is an employee and stockholder of Biogen Idec, Inc., which holds patents and pending patent applications in the United States and abroad on TWEAK-related molecules, including U. S. Patent No. 7129061, 7109298, 7087725, 7695934, 7566769, and 8048422.