PPARβ/δ Agonist Provides Neuroprotection by Suppression of IRE1α–Caspase-12-Mediated Endoplasmic Reticulum Stress Pathway in the Rotenone Rat Model of Parkinson’s Disease
Abstract Two recent studies demonstrated that peroxisome proliferator-activated receptor β/δ (PPARβ/δ) agonists exerted neuroprotective effects in mouse model of Parkinson’s disease (PD). However, the underlying mechanisms remain unknown. Endoplasmic reticulum (ER) stress plays a major role in rotenone-induced dopaminergic neuronal degenera- tion. In the present study, we explored whether GW501516, a selective and high-affinity PPARβ/δ agonist, could protect the dopaminergic neurons against degeneration and improve PD behavior via suppressing the ER stress in the rotenone rat model of PD. GW501516 was administered intracerebroven- tricular infusion. Catalepsy and open field tests were used to test catalepsy and locomotor activities. The levels of dopa- mine and its metabolites were determined using high- performance liquid chromatography. Western blot and immu- nohistochemistry analysis were performed to assess dopami- nergic neuronal degeneration. Quantitative real-time RT-PCR and Western blot analysis were executed to detect ER stress. TUNEL and immunohistochemistry assays were used to de- tect ER stress-mediated apoptosis. Our results showed that GW501516 ameliorated the catalepsy symptom and increased locomotor activity. Meanwhile, GW501516 partially reversed the loss of dopaminergic neurons. Moreover, GW501516 sup- pressed the activation of ER stress markers including inositol- requiring enzyme 1α (IRE1α) and caspase-12. Furthermore, GW501516 inhibited caspase-12-mediated neuronal apopto- sis. These findings suggest that GW501516 conferred neuro- protection of not only biochemical and pathological attenua- tion but also behavioral improvement in the rotenone rat mod- el of PD. More importantly, we demonstrated for the first time that suppressing IRE1α–caspase-12-mediated ER stress path- way may represent one potential mechanism underlying the neuroprotective effects of PPARβ/δ agonist in the rotenone rat model of PD.
Keywords : Parkinson’s disease . PPARβ/δ . ER stress . Neuroprotection
Introduction
Parkinson’s disease (PD) is clinically characteristic of rigidity,bradykinesia, resting tremor, and postural instability. The
pathological hallmark of PD is the progressive degeneration of dopaminergic neurons in the substantia nigra pars compacta (SNpc) [1], which leads to dopamine (DA) deficiency in the striatum. However, the etiology and pathogenesis of PD re- main largely unknown. Moreover, currently available drug therapy cannot halt the disease progression and only provides symptomatic relief. Accumulating studies have demonstrated that endoplasmic reticulum (ER) stress plays a crucial role in the development of PD [2–4].
ER stress is buffered by activation of the unfolded protein response (UPR). A majority of the UPR signaling pathways have been unraveled [5–7]. Initiation of the canonical UPR involves activation of three key signaling proteins: inositol- requiring enzyme/endonuclease 1α (IRE1α), activating tran- scription factor-6 (ATF6), and double-stranded RNA-activat- ed protein kinase (PKR)-like ER kinase (PERK). Normally, these transmembrane proteins are bound to the intra-luminal binding immunoglobulin protein/glucose-regulated protein 78 (Bip/GRP78). In the presence of stress, the large excess of unfolded proteins sequesters Bip/GRP78 from transmem- brane ER proteins, thereby inducing the UPR. Bip/GRP78 is a central regulator of the UPR and plays a crucial role in the regulation of the ER dynamic homeostasis [8, 9]. Upregula- tion of Bip/GRP78 is acknowledged as sign of ER stress.
However, prolonged ER stress and failure to adapt to ER stress result in activity of CCAAT-enhancer-binding protein (C/EBP) homologous protein (CHOP) (a proapoptotic factor) and subsequent cell apoptosis [10, 11]. Accumulating evi- dence has demonstrated that rotenone induces ER stress in vitro and in vivo [2, 12–14]. Otherwise, a molecular mech- anism linking ER stress and apoptosis has been recently de- scribed, in which the ER-resident caspase-12 is activated, leading to caspase-3 activation and cellular death. Caspase- 12 mediates the ER-specific apoptotic pathway [15, 16]. CHOP and caspase-12 are critical mediators in the course of ER stress-mediated apoptosis.
Peroxisome proliferator-activated receptors (PPARs) are members of the nuclear receptor superfamily of ligand- activated transcription factors [17] and composed of three subtypes including α, γ, and β (also known as δ). PPARβ/δ is widely expressed in neuronal nuclei in both the SNpc and the striatum [18], and it is expressed in dopaminergic neurons and oligodendrocytes [19]. Meanwhile, two recent studies in- dicated that PPARβ/δ agonists could provide neuroprotective effect in PD mouse model [20, 21], but the two results were not completely concordant. Otherwise, the underlying mech- anism remains poorly understood. In addition, there were studies showing that PPARβ/δ agonists could prevent ER stress in vitro and in vivo [22, 23]. Based on the evidences that ER stress played a major role in rotenone-induced dopa- minergic neuronal loss and PPARβ/δ agonist could suppress ER stress, therefore, the aim of our present study was to ex- plore whether GW501516 can result in the neuroprotective effect via suppressing ER stress signaling pathway in the ro- tenone rat model of PD.
Materials and Methods
Reagents and Preparation
GW501516, rotenone, dimethyl sulfoxide (DMSO), and sun- flower oil were purchased from Sigma-Aldrich (St. Louis, MO, USA). Rotenone was emulsified in sunflower oil at 2.5 mg/ml for injection. GW501516 was dissolved with 30 % DMSO/saline at concentrations of 10 mg/ml for injec- tion. Micro-osmotic pumps (2006, Alzet Corporation, Palo Alto, CA) were incubated in sterile 0.9 % saline at 37 °C overnight, according to the manufacturer’s instructions.
Animal Management
Animals
Male Sprague–Dawley rats (200–220 g) were obtained from the Experimental Animals Center of Nanjing Medical Univer- sity. Rats were housed in a standard animal room under a 12/ 12-h light/dark cycle and given free access to food and water. Body weight was recorded every week to monitor health, and animals were fed Nutrison milk powder (5 g in water, twice daily) when they lost weight. Animal care and experimental protocols were carried out according to the Guide for the Care and Use of Laboratory Animals of Nanjing Medical Univer- sity and were approved by the Biological Research Ethics Committee of Nanjing Medical University.
Experimental Groups
Experimental rats were randomly divided into four groups: control group, rotenone group, rotenone + GW501516 group, and GW501516 group. Control group was treated with sun- flower oil and 30 % DMSO/saline. Rotenone group was given rotenone and 30 % DMSO/saline. Rotenone + GW501516 group was administered rotenone and GW501516. GW501516 group was treated with sunflower oil and GW501516.
Drug Administration
Rotenone or sunflower oil was administrated intraperitoneally once a day at 1 ml/kg for 6 weeks. The administration route, dose, and duration were chosen based on previous researches [24, 25]. GW501516 does not readily cross the blood–brain barrier, so GW501516 or vehicle was administrated intracerebroventricularly. The dose for GW501516 was deter- mined according to a previous study from Iwashita and col- leagues [20]. Rats were anesthetized with 10 % chloral hy- drate (0.35 ml/100 g) and placed in a stereotactic frame (David Kopf Instrument Inc., USA). Rats were chronically implanted with brain infusion cannula (Brain Infusion Kit 2; ALZET Inc., USA) coupled to a micro-osmotic pump into the right lateral cerebral ventricle at the following coordinates: 0.8 mm posterior to the bregma, 1.5 mm lateral to the midsagittal suture, and 4.0 mm ventral to the skull. GW501516 was ad- ministered at 1 h before rotenone treatment, and the infusion volume was adjusted to 0.15 μl/h.
Neurological Behavioral Measurements (Open Field Test and Catalepsy Test)
On day 41 of treatment with rotenone or vehicle, all rats were tested in the open field box for locomotor activities and the next day following this for catalepsy test. Tests were carried out between 9 am and 2 pm, always in the same context and under standard conditions.
Open Field Test
The open field consisted of a square plastic box measuring 100×100×40 cm and painted black inside. Each animal was placed in the box and allowed to freely explore the area for 5 min. A video camera was fixed on the top of the box. The movement and behavior of the rats were recorded on the vid- eo, and the film was later observed for off-line. Two motor parameters (number of rearing and time of inactive sitting) were quantified with the use of automatic software.
Catalepsy Test
The catalepsy test consisted of a vertical grid and a horizontal bar. A gridiron 25.5 cm wide and 44 cm high with a space of 1 cm between wires was used for grid test. Each rat was hung by all four paws on the vertical grid, and a stopwatch was started. When the rats removed a paw from the wire, the stop- watch was stopped, and the time taken was recorded as de- scent latency. Maximum descent latency time was fixed at 120 s.The second part of the test was the bar test. Each rat was gently placed with its forepaws on a metal rod suspended 9 cm above the floor. The time elapsing before it climbed down from the bar was recorded. Maximum descent latency time was fixed at 120 s.
HPLC Analysis
Levels of DA and its metabolites, dihydroxyphenylacetic acid (DOPAC) and homovanillic acid (HVA), in the striatum were measured in the HPLC with electrochemical detection [26]. The striatum of the rats was rapidly dissected out on an ice- cold dish. Samples were immediately frozen in liquid nitrogen before storage at −80 °C until analysis. On the day of the assay, the tissue samples were homogenized with an ultrasonic disrupter in a 0.1 M perchloric acid. After centrifugation (15, 000g, 15 min, 4 °C), Twenty microliters of the sample was injected into the HPLC system for quantification. The mobile phase consisted of 0.1 M citrate buffer, 0.02 mM ethylenedi- aminetetraacetic acid (EDTA), 1 mM sodium octane sulfonic acid, and 10 % methanol at a flow rate of 1.0 ml/min. DA, DOPAC, and HVA were detected using an ultraviolet detector.
The chromatogram was recorded and analyzed with Breeze software version 3.2 (Waters).
Quantitative Real-Time RT-PCR
RNAwas extracted from frozen midbrain tissues using TRIzol (Invitrogen, Carlsbad, CA) according to the manufacturer’s instructions. The concentration of total RNA was quantified by spectrophotometry, and RNA was reverse-transcribed using the M-MLV Reverse Transcriptase System (TaKaRa Biotechnology, Dalian, China) and oligo (dT). Total cDNA was amplified with LightCycler FastStart DNA Master SYBR Green I (TaKaRa Biotechnology, Dalian, China). Real-time PCR was carried out using the following primers: Bip/ GRP78, 5′-GCAGTTGCTCACGTGTCTTG-3′ (sense), 5′- TCCAAGGTGAACACACACACCC-3′ (anti-sense); Cas- pase-12, 5′-GGCCGTCCAGAGCACCAGT-3′ (sense), 5′- CAGTGGCTATCCCTTTGCTTG-3′ (anti-sense); CHOP, 5′- CCAGCAGAGGTCACAAGCAC-3′ (sense), 5′-CGCACT GACCACTCTGTTTC-3′ (anti-sense); GAPDH, 5′-ACAG CAACAGGGTGGTGGAC-3′ (sense), 5′-TTTGAGGGTGCAGCGAACTT-3′ (anti-sense). Two-step qPCR was per- formed using the ABI 7500 Real-Time PCR System (Applied Biosystems, Carlsbad, CA). After 30 s of initial denaturation at 95 °C, the thermal profile included two-step cycles: step 1 for denaturation at 95 °C for 5 s and step 2 for annealing and extension at 60 °C for 34 s. Data were collected at step 2. The results were obtained using the comparative Ct method and the arithmetic formula 2−△△Ct.
Western Blot Analysis
After extraction of midbrain protein, different samples with an equal amount of protein were separated with 15 % SDS- PAGE, transferred to PVDF membranes, and blocked in 5 % nonfat milk for 2 h at room temperature. Membranes were incubated with primary antibodies specific to tyrosine hydrox- ylase (TH; 1:8000; Sigma), Bip/GRP78 and procaspase-12 (1:200; Santa Cruz Biotechnology, Santa Cruz, CA, USA), p-IRE1αser724 and IRE1α (1:1000; Novus Biologicals, MO, USA), CHOP (1:1000; Cell Signaling Technology, Danvers, MA, USA), and β-actin (1:1000; Santa Cruz Biotechnology) at 4 °C overnight. Membranes were then washed with TBS buffer containing 0.1 % Tween-20 and incubated with horse- radish peroxidase (HRP)-labeled secondary antibody (1:2000; Sigma) for 2 h at room temperature. Finally, membranes were developed using the enhanced chemiluminescence (ECL) sys- tem. Immunoreactivity was quantified using ImageJ software.
Immunohistochemistry Analysis
After behavioral tests, rats were perfused with 0.9 % saline (pH 7.4) followed by a fixative solution containing 4 % paraformaldehyde (PFA) in 0.9 % saline (pH 7.4). Afterward, the brains were removed and placed in the same PFA solution at 4 °C overnight. The brains were dissected out from 24.5 to 26.2 mm caudal to bregma (the region containing the SNc) and embedded in paraffin. Coronal sections (5 μm) were cut according to standard procedures. Afterward, the paraffin- embedded sections were dewaxed and hydrated while endog- enous peroxidase activity was blocked with 3 % H2O2 for 30 min. After antigen retrieval, the slides were treated with 0.5 % Triton-X 100 for 30 min and 5 % bovine serum albumin for 30 min. The sections were incubated with primary anti- bodies: polyclonal rabbit anti-caspase-12 (1:100; Santa Cruz Inc., USA) and monoclonal mouse anti-TH (1:2000; Sigma) overnight at 4 °C and then incubated with the HRP- conjugated secondary antibody (1:1000; Sigma-Aldrich Inc.) for 1 h. Positive staining was visualized with diaminobenzi- dine. Lastly, the sections were counterstained with Mayer’s hematoxylin (Sigma-Aldrich Inc.), dehydrated, mounted on the slides, and examined by a digital camera connected to a microscope. Immunoreactivity was performed on three ran- domly selected nonoverlapping fields in SNpc by an observer who was blinded to the experimental groups. Immunoreactiv- ity was quantified using GEDA-801D morphological image analysis system.
For double immunofluorescence assay, deparaffinized sec- tions were blocked with 5 % bovine serum albumin and then incubated overnight at 4 °C with a mixture of the polyclonal rabbit anti-caspase-12 (1:100; Santa Cruz Inc., USA) and monoclonal mouse anti-TH (1:2000; Sigma) overnight at 4 °C. The sections were rinsed and sequentially incubated for 1 h at 37 °C with anti-rabbit IgG tagged with FITC (Invitrogen, Carlsbad, CA, USA; 1:200) or anti-mouse IgG tagged with TRITC (Invitrogen; 1:200). Then, the sections were further incubated with 0.5 μg/ml 4,6-diamidino-2- phenylindole (DAPI) for 10 min and were sealed with cover- slips. The sections were examined with an Olympus fluores- cence microscope (Olympus, Tokyo, Japan).
TUNEL Assay
For TUNEL assay, a cell death detection kit was used (In Situ Cell Death Detection Kit, POD; Roche, USA). The paraffin- embedded sections received deparaffinization and rehydration treatments and then were incubated with proteinase-K for 15 min at room temperature followed by wash in PBS. The TdT enzyme and nucleotide mixture were added and incubat- ed for 60 min at 37 °C. Sections were washed with PBS again and then incubated with peroxidase–streptavidin conjugate solution for 30 min at room temperature. Subsequently, the sections were washed again with PBS and then exposed to 0.03 % diaminobenzidine in 0.01 % H2O2. Finally, the sec- tions were counterstained with Mayer’s hematoxylin (Sigma- Aldrich Inc.), dehydrated, mounted on the slides, and examined by a digital camera connected to an optical micro- scope. Neurons with deep black nuclei were identified as TUNEL-positive neurons. Neuron counting was performed on three randomly selected nonoverlapping fields per slide under high-power magnification by an observer who was blinded to the experimental groups and was expressed as num- ber per square millimeter. Data obtained in every field were added together to make a final data count for each slide and expressed as percentage of total cell number within the relevant fields.
Statistical Analysis
Data were presented as mean±standard deviation (SD), and the statistical analysis was carried out using SPSS version 19.0 for Windows (SPSS, Chicago, IL). Tests of variance homoge- neity, normality, and distribution were performed. Statistically significant differences were evaluated by one-way analysis of variance (ANOVA) followed by the Tukey post hoc test. Dif- ferences were considered significant at P value <0.05. Results GW501516 Attenuates the Hypokinesis of PD Rats Seven rats died prior to behavioral tests and were excluded from the study: three rats in the rotenone group and four rats in the rotenone + GW501516 group. In order to investigate the neuroprotective effect of GW501516 in the rotenone rat of PD, we performed the open field test and catalepsy test. GW501516 Increases Locomotor Activity of PD Rats The open field test was performed to test locomotor activities. As shown in Fig. 1a, the rotenone infusion remarkably de- creased the numbers of rearing and prolonged the time of inactive sitting when compared with the control group (n= 10, P<0.001). The GW501516 + rotenone group demonstrat- ed an obvious increase in the number of rearing (n =10, P<0.01) and a significant reduction in the time of inactive sitting (n=10, P<0.001) in comparison to the rotenone group. However, there was no difference to be observed in the num- ber of rearing and in the time of inactive sitting between the control group and GW501516 group (n=10, P>0.05).
GW501516 Ameliorates the Catalepsy Symptom of PD Rats
The catalepsy test was performed to test the catalepsy symp- tom. As shown in Fig. 1b, chronic rotenone treatment resulted in significantly prolonged descent latency as compared to the control group (n = 10, P < 0.001). The rotenone-induced catalepsy was significantly antagonized by treatment with GW501516 (n = 10, P < 0.001), but the rotenone + GW501516 group showed significantly prolonged descent la- tency compared with the control group (n=10, P<0.001). GW501516 Decreases the Depletion of DA in the Striatum of PD Rats To confirm the neuroprotective effect of GW501516 in PD rats, we measured the levels of DA and its metabolites in the HPLC with electrochemical detection. In the striatum, the amounts of DA, DOPAC, and HVA were remarkably reduced in the rotenone group as compared to the control group. GW501516 treatment partially abolished the reduction in the level of DA in the striatum (n=6, P<0.001) but had no effect on the DOPAC and HVA (n =6, P> 0.05). Besides, GW501516 alone demonstrated no effects on the DA and its metabolites (n=6, P>0.05) (Fig. 2).
GW501516 Reverses the Loss of Dopaminergic Neurons in the Midbrain of PD Rats
To evaluate the effect of GW501516 on the loss of dopami- nergic neurons in PD rats, we detected the TH protein and TH- immunoreactive (TH-ir) neurons in the midbrain using West- ern blot analysis and immunohistochemistry analysis, respec- tively. Rotenone markedly decreased the TH protein expres- sion (Fig. 3a; n=6, P<0.001) and caused TH-ir neuron loss (Fig. 3b; n=4, P<0.001) as compared to the control group, and GW501516 treatment defended against rotenone-induced toxic effect (P<0.05). No difference was observed between the GW501516 group and the control group on the neuron loss (P>0.05).
GW501516 Suppresses ER Stress in the Midbrain of PD Rats
To investigate the underlying mechanism involved in the neu- roprotection, we evaluated the effect of GW501516 on ER stress in the PD rats. Western blot and quantitative real-time RT-PCR were used to examine the ER stress marker expres- sion. As shown in Fig. 4a, rotenone induced significant in- crease in messenger RNA (mRNA) expression of ER stress markers including Bip/GRP78, CHOP, and caspase-12 (n=6, P<0.001). Meanwhile, consistent with the changes in tran- scription levels, significant increments in protein products of Bip/GRP78 and CHOP were also observed (Fig. 4b; n=6, P<0.001). It is noteworthy that rotenone caused decrease in the protein level of procaspase-12, implying that procaspase- 12 is activated by being cleaved. Collectively, these findings indicated that ER stress was involved in the rotenone rat mod- el of PD. As shown in Fig. 4a, GW501516 treatment markedly downregulated the mRNA expressions of Bip/GRP78, CHOP, and caspase-12 in the midbrain of PD rats (n=6, P<0.001). Meanwhile, the protein expressions of Bip/GRP78 and CHOP were decreased in parallel (Fig. 4b; n=6, P<0.001). It was noticeable that GW501516 treatment remarkably increased the level of procaspase-12 (Fig. 4b; n=6, P<0.001). Taken together, these results demonstrated that GW501516 sup- pressed rotenone-induced ER stress. But, GW501516 alone could not affect the mRNA and protein expression of Bip/ GRP78, CHOP, and caspase-12 (Fig. 4a and b; n =6, P>0.05). As indicated by Fig. 4c, the level of IRE1α phos- phorylated at ser724 was elevated significantly by rotenone, implying activation of IRE1α, and GW501516 reduced the IRE1α phosphorylated at ser724 (n=6, P<0.001). GW501516 Inhibits ER Stress-Mediated Dopaminergic Neuronal Apoptosis in the Midbrain of PD Rats To further investigate whether the ER stress-mediated apopto- sis was involved in the dopaminergic neuronal degeneration and the effect of GW501516 on ER-specific apoptosis, we detected ER-resident caspase-12 expression in the midbrain using immunohistochemistry assay and then measured the neuronal apoptosis in the midbrain using TUNEL assay. Last- ly, double immunofluorescence assay was used to confirm the expression of caspase-12 in dopaminergic neurons. As displayed by Fig. 5a, rotenone caused significantly increased expression of caspase-12 in the midbrain of PD rats which was attenuated by GW501516 (n=4, P<0.001). Meanwhile, rote- none induced noticeable neuronal apoptosis in the midbrain of PD rats which was partially reversed by GW501516 (Fig. 5b; n=4, P<0.001). However, GW501516 alone had no effect on caspase-12 and neuronal apoptosis in the midbrain of normal rats. Double immunofluorescence assay demonstrated that the caspase-12 immunoreactivity in dopaminergic neurons in rotenone-induced midbrain of PD rats was stronger than that in control rats, while treatment with GW501516 markedly attenuated this increase in caspase-12 immunoreactivity (Fig. 6). Discussion The major finding of this study is that a PPARβ/δ agonist GW501516 confers neuroprotection of not only behavioral improvement but also biochemical and pathological attenua- tion in the rotenone-induced rat model of PD. More impor- tantly, we demonstrate for the first time that these neuropro- tective effects may be achieved by suppression of IRE1α– caspase-12-mediated ER stress pathway. In the current experiment, we showed that GW501516 attenuated rotenone-induced depletion of DA in the stria- tum. And, this outcome is supported by the work of Iwashita et al. in mouse model of PD [20]. However, this is in contrast to Martin et al. who found that GW0742 (another PPARβ/δ agonist) infusion did not affect 1- methyl-4-phenyl-1,2,3,6-tetrahydropyridine (MPTP)-in- duced decreases in DA and its metabolites in the striatum [21]. The differences among previous studies and present experiment may arise from variations in the animal spe- cies, infusion site, infusion duration, and doses of agonist used. It should be noted that there were no significant differences in striatal DOPA and HVA levels to accompa- ny with DA between rotenone + GW501516 group and rotenone group, which is in contrast to Iwashita et al. [20], and it seems interesting and contradictory apparently. It is plausible that GW501516 represses the metabolism of DA in PD rats. Since Quinn et al. showed that a PPARγ agonist, pioglitazone, protected against MPTP-induced neurotoxicity by inhibition of monoamine oxidase B (MAO-B) [27]. The underling mechanism may be that GW501516 suppresses the activity of MAO which is a rate-limiting enzyme in the metabolism of DA in the stri- atum. In the midbrain of PD rats, the increased TH pro- tein level after GW501516 treatment suggested that the degeneration or loss of dopaminergic neurons was re- versed. These data indicated that the PPARβ/δ agonist rescued dopaminergic neurons in the SNpc. In addition, Iwashita et al. confirmed that the loss of TH-positive neu- rons in the SNpc in a mouse model of PD is well correlated with the decrease in DA and metabolite content in the striatum [28]. Our present results supported the notion that the quantification of DA content is a suitable substitute to quantifying the dopaminergic neuronal degen- eration by TH immunostaining. Nevertheless, GW501516 alone had no effect on TH and DA levels, which indicated that PPARβ/δ agonist was not necessary in normal status. Our present results demonstrated that rotenone dramatically upregulated mRNA and protein expressions of Bip/GRP78 and CHOP, indicating that chronic rotenone treatment induced per- sistent and severe ER stress. GW501516 treatment inhibited the expressions of Bip/GRP78 and CHOP, implying that GW501516 suppressed ER stress. Meanwhile, our results showed that rotenone infusion increased IRE1α phosphoryla- tion, which was supported by a previous research conducted in cellular models of PD [2]. GW501516 treatment attenuated the IRE1α phosphorylation, consistent with previous studies con- ducted in several cell lines [22, 23]. Caspase-12 is localized to the ER and activated in ER stress-induced apoptosis, but not by membrane- or mitochondrial-targeted apoptotic signals. It is found to mediate ER-specific apoptotic pathway in prion disor- ders and Alzheimer’s disease [15, 29]. In our experiments, rote- none infusion led to the decrease in the level of procaspase-12 protein and the increase in the level of caspase-12 protein in the midbrain of PD rats, suggesting that caspase-12 was activated, which was accompanied by increased neuronal apoptosis in the midbrain of PD rats. Indeed, the increase of neuronal apoptosis was well correlated with the elevated caspase-12 expression in the middle brain of PD rats. In addition, double immunofluores- cence assay demonstrated that the caspase-12 immunoreactivity in dopaminergic neurons in rotenone-induced midbrain of PD rats was stronger than that in control rats, implying that caspase- 12-mediated apoptosis was involved in the dopaminergic neuro- nal degeneration. Treatment with GW501516 markedly attenu- ated the rotenone-induced increase in the level of caspase-12 immunoreactivity and neuronal apoptosis, suggesting GW501516 anti-apoptotic effect. Taken together, these results demonstrated that caspase-12 was involved in rotenone- induced ER stress and ER stress-mediated apoptosis whereas GW501516 suppressed the rotenone-induced ER stress and sub- sequent ER-specific apoptosis. Indeed, Iwashita et al. showed that the neuroprotective effects of the selective PPARβ/δ ago- nists were closely correlated with anti-apoptotic effect in their in vitro cell death model [20]. Notably, The IRE1α activates caspase-12 and the c-Jun N- terminal kinase (JNK) pathway [30, 31], initiating apoptotic phase. Indeed, IRE1α has been linked to several proapoptotic pathways, the most well defined being the IRE1α–caspase-12 and IRE1α–JNK axes. Consistently, our present results demon- strated that the IRE1α phosphorylation was closely correlated with caspase-12 activation. Taken together, our findings indicat- ed that IRE1α–caspase-12 pathway is involved in the rotenone rat model of PD and suppressing ER stress via inhibiting IRE1α–caspase-12 pathway contributes to the neuroprotection of GW501516. Moreover, the recent groundbreaking discoveries of novel IRE1α regulatory events reveal that IRE1α signaling is persistent during ER stress and define IRE1α as a master regulator in cell fate determination under ER stress [32–34]. Additionally, ER stress interacting with inflammatory and oxidative stress contributes to degeneration of dopaminergic neurons [35]. Meanwhile, PPARβ/δ is also expressed in oli- godendrocytes and astrocytes which regulate inflammatory responses in the brain, and PPARβ/δ agonist exhibited anti- inflammation in vitro and in vivo [36–38]. Therefore, further researches are required to explore whether other mechanism and pathway are involved in the neuroprotection of PPARβ/δ agonist in PD. In conclusion, the present study clarifies that PPARβ/δ agonist provides neuroprotection in rotenone-induced rat model of PD and further elucidates for the first time that suppressing GW 501516 ER stress by inhibiting IRE1α–caspase-12 pathway is involved in the neuroprotection.