Introduction

Parkinson’s disease (PD) is the second most common age-related neurodegenerative disorder. Progressive loss of dopaminergic (DA-) neurons in the substantia nigra (SN) pars compacta (SNc) and the formation of Lewy bodies are pathological features of PD. Although the etiology of PD remains unclear, mounting evidence indicates that multiple factors lead to this disease, including oxidative stress, mitochondrial dysfunction, proteasome dysfunction, inflammatory or immune response, disturbances of calcium homeostasis, excitotoxicity, and apoptosis [1–3]. In recent years, we have focused on studying the mechanism of α-synuclein (α-SYN), which is the main component of Lewy bodies [4, 5]. α-SYN, a characteristic marker of various types of PD, is degraded by the ubiquitin–proteasome system (UPS) and the autophagy-lysosome pathway (ALP) [5, 6]. Mounting evidence suggests that autophagic impairment induced by pathological factors, such as oxidative stress, results in excessive accumulation of misfolded proteins, which contributes to the onset and progression of PD [7, 8].

Acidosis is a common cause of tissue impairment. Recent studies have shown that significant reduction of extracellular pH can activate the special voltage-independent, proton-gated cation-selective channels called acid-sensing ion channels (ASICs), which are widely distributed in the central and peripheral sensory nervous system [9, 10]. So far, at least seven subtypes of ASICs (ASIC1a, lb, 2a, 2b, 3a, 3b and 4) encoded by four different genes have been identified [11]. Of all the ASICs, homomeric ASIC1a and heteromeric ASIC1a/2b channels have permeability to Ca2+ [12, 13]. A number of studies have demonstrated that acidosis within the brain is involved in various nervous system degenerative diseases, including acute brain injury and PD, and that ASIC1a is found to be expressed in the vulnerable DA-neurons in the SN [14, 15]. In addition, amiloride, a non-selective ASICs blocker, protects DA-neurons against 1-methyl-4-phenyl-1,2,3,6-tetrahydropyridine (MPTP)-induced degeneration in mice. Psalmotoxin-1 (PcTx1), a subtype-selective ASIC1a blocker has also been reported to weaken the affinity of dopamine (DA) for dopamine transporter (DAT) [14, 16]. All these data suggest that ASICs are related to the damage of DA-neurons. Therefore we hypothesize that the degeneration of DA-neurons may be caused by chronic and pathological activation of ASICs, and that selective inhibition of some subunits of ASICs may delay the progression of PD.

To date, no direct evidence demonstrating a relationship between ASICs and autophagy has been found. Our previous experimental data revealed that amiloride protects PC12 cells against 1-methyl-4-phenylpyridinium (MPP+) and acidic damage, and that this protective effect is closely related to the changes of autophagy-specific proteins LC3-II and LAMP2a [17, 18]. These findings led us to presume that enhancement of α-SYN autophagic degradation, via its action on ASICs, may protect DA-neurons.

Paeoniflorin (PF), a monoterpene glycoside, is the main biologically active component extracted from the roots of Paeonia lactiflora and is used in traditional Chinese medicine. Previous studies have demonstrated that PF has a variety of pharmacological effects, including anti-oxidant and anti-inflammatory effects and inhibition of intracellular Ca2+ overload [19, 20]. Recent studies also suggest that PF has neuroprotective properties and can be used in clinic to treat cerebral ischemia, epilepsy, and neurodegenerative disorders, especially PD [21]. However, its mechanism of neuroprotection remains unclear. Our previous work has shown that PF protects PC12 cells against MPP+ and acidic damage through regulating autophagy, just like amiloride does [17, 18]. In this study, the 6-hydroxydopamine (6-OHDA)-induced PD rat model was used to evaluate the protective effects of PF on DA-neurons and to further investigate PF’s molecular mechanisms. We expect to provide evidence that PF is a new neuroprotectant for PD and to determine whether ACIS1a is a potential therapeutic target in PD.

Material and Methods

Drugs and Chemicals

PF (monomer, purity >98 %) was purchased from the National Institute for the Control of Pharmaceutical and Biological Products of China (Beijing). Psalmotoxin-1 (PcTx1) was obtained from Alomone Labs (Jerusalem, ISR). 6-OHDA and amiloride were purchased from Sigma–Aldrich (St Louis, MO, USA). Apomorphine (APO) was obtained from Enzolife Science in Japan. The sources of the primary antibodies used in these experiments are as follows: mouse anti-tyrosine hydroxylase (TH) antibody (Sigma, USA), rabbit anti-LC3 (Abcam, Cambridge, UK), mouse anti-α-SYN (Abcam), rabbit anti-p62 (Enzo, Japan), and rabbit anti-ASIC1a (Alomone Labs). The small interfering RNA (siRNA) sequences targeting rat ASIC1a were as follows: siRNA-1, 5′-CACUGUGCAAUCUCAAUGATT-3′ and 5′-UCAUUGAGAUUGCACAGUGTT-3′; siRNA-2, 5′-GACUCGGAUUUCUUCGACUTT-3′ and 5′-AGUCGAAGAAAUCCGAGUCTT-3′; and siRNA-3, 5′-CCUACGAGGUCAUUAAGCATT-3′ and 5′-UGCUUAAUGACCUCGUAGGTT-3′. Negative control siRNA was purchased from GenePharma (Shanghai, China).

Animals and Experimental Procedure

All 140 male Sprague–Dawley (SD) rats (180–220 g, 2–3 month old) were purchased from the Center for Experimental Animals (Soochow University, China). Animals were housed under specific conditions: quiet environment, day-night rhythm (12 h light–dark cycles), and free access to food and water. The experimental procedures were agreed to by the Institutional Animal Care and Use Committee of Soochow University (Suzhou, China).

6-OHDA-Induced PD Model and the Behavioral Test

The rats were anesthetized with 3.6 % chloral hydrate [0.18 g/kg, intraperitoneal (i.p.)] and were fixed on a stereotaxic apparatus (David Kopf Instruments, CA, USA). According to the stereotaxic brain atlas of rat [22], 6-OHDA (10 µg 6-OHDA hydrochloride in 5 µL of saline solution, 5 µg 6-OHDA at each point), or sterile saline (2.5 µL at each point) were injected at two points in the right striatum (Bregma coordinates for two injection points—first point: AP, +1.0 mm; ML, +3.0 mm; DV, −4.5 mm; and second point: AP, −0.2 mm; ML, +3.0 mm; DV, −6.0 mm) at the rate of 0.5 µL/min by 10-µL Hamilton syringe. After the injection, the needle was retained in the striatum for 3 min. Then the needle was withdrawn at the rate of 1 mm/min. After surgery, the rats were returned to the same environment they were in prior to surgery.

Three weeks after the operation, the rats’ contralateral rotations induced by APO [0.5 mg/kg, subcutaneous (s.c.)] were measured and recorded with a video camera once a week for two consecutive weeks. The duration of each recording time was 30 min. Rats with a rotating frequency of over 7 turns/min were included in the PD model. These selected model rats were divided into the following six groups: vehicle treatment group (sterile saline, i.p.), PF 15 mg/kg, i.p. treatment group, PF 30 mg/kg, i.p. treatment group, PF 60 mg/kg, i.p. treatment group, PcTx1 0.7 µg/kg, intranasal (i.n.) treatment group, and amiloride 10 mg/kg, i.p. treatment group. At the same time, sham-operated rats received vehicle or PF (60 mg/kg) treatment. All these rats were treated with vehicle or PF or amiloride or PcTx1 once a day for 21 days. And the changes in the contralateral rotational behavior caused by APO (0.5 mg/kg, s.c.) were tested once a week for 3 weeks after the rats received treatment.

High-Performance Liquid Chromatography (HPLC) Determination of DA, Serotonin, and Their Metabolites in the Striatum

The levels of DA, 3,4-dihydroxyphenylacetic acid (DOPAC), 4-hydroxy-3-methoxyphenylacetic acid (HVA), serotonin (5-HT) and 5-hydroxyindole-3-acetic acid (5-HIAA) in the striatum were measured by HPLC. A Waters 2695 Separation Module HPLC system (Waters Company, Milford, MA, USA) equipped with a Waters 2475 Fluorescence detector (Waters Company) was used. A C18 chromatographic column (250 × 4.6 mm, 5 µm, Akzonobel, AMS, NL) was used to qualitatively and quantitatively analyze the neurotransmitters, and the column temperature was maintained at 30 °C. Data acquisition and integration were controlled by Applied Biosystems Analyst Software (Applied Biosystems/MDS SCIEX, version 1.4.2). The rats were sacrificed and the striatum was dissected from brain. Then the tissue was homogenized and sonicated in 200 µL saline on ice for 10 s, three times; then, the same volume of 200 µL perchloric acid (4 °C, 0.4 mol/L) was added. The mixture was centrifuged at 10,000 rpm for 15 min at 4 °C; then, the supernatant mixed with 200 µL perchloric acid (4 °C, 0.4 mol/L) was centrifuged at 10,000 rpm for 10 min at 4 °C again. Next, the supernatant was diluted in perchloric acid (4 °C, 0.4 mol/L); and the total volume was 1 mL. The supernatant was then filtered (0.45 µm, Millipore, Boston, MA, USA), and 10 µL of sample was injected into the HPLC system for analysis. The mobile phase included 0.1 M sodium acetate, methanol (99.9 %) (9:1, v/v) and was adjusted to pH 5.1. The flow rate of the sample and mobile phase was set at 1 mL/min. The excitation wavelength was set at 290 nm and emission wavelength was set at 330 nm. The levels of monoamine neurotransmitters were expressed as µg/g wet tissue.

Western Blot Analysis

Midbrain tissue of rats was dissected and homogenized in radioimmunoprecipitation assay tissue lysis buffer (1:10, w/v, Beyotime Institute of Biotechnolog, Shanghai, China), to which protease inhibitor cocktail was added. The lysate was placed on ice for 30 min and centrifuged at 132,000 rpm for 15 min at 4 °C. Then the supernatant was collected; and protein concentration was measured by bicichoninic acid protein assay kit (Beyotime). The samples (20 µL) were separated with 12 % sodium dodecyl sulfate–polyacrylamide gels and transferred onto polyvinylidine fluoride membranes. After blocking with 2 % bovine serum albumin (BSA) in phosphate buffered saline (PBS) with Tween-20 (3.2 mM Na2HPO4, 0.5 mM KH2PO4, 0.1 % Tween-20, pH 7.4) at 37 °C for 1 h, membranes were incubated with primary antibodies, including anti-LC3 (1:1000), mouse anti-α-SYN (1:500), anti-p62 (1:1000), and anti-ASIC1a (1:200) at 4 °C overnight. Next, the membranes were incubated with corresponding secondary antibodies (1:5000) at room temperature for another hour. The special protein bands were detected by ECL chemiluminescence kit (GE Healthcare, Buckinghamshire, UK), and the density of protein bands was analyzed by using Image J software (National Institute of Health, Bethesda, MD, USA).

Immunohistochemistry

The rats were deeply anesthetized with 3.6 % chloral hydrate and perfused with 4 % paraformaldehyde (PFA). The intact brain tissues were collected and then fixed in 4 % PFA at 4 °C for 24 h. The tissues were immersed in 15 % sucrose/PBS solution (0.1 M) and 30 % sucrose/PBS solution (0.1 M) sequentially at 4 °C until they sunk to the bottom of the tube. After being stored at −80 °C for 30 min, frozen sections containing SN were cut at a thickness of 12 µm and mounted on gelatin-coated glass slides. Briefly, fresh sections were treated in 3 % H2O2 for 15 min to get rid of endogenous peroxidase, blocked with 5 % goat serum in Triton X-100-PBS for 30 min, and then incubated with primary antibody (anti-TH, 1:1,000, Sigma–Aldrich; anti-ASIC1a, 1:50, Alomone Labs) overnight at 4 °C. Next, slides were incubated with horseradish peroxidase (HRP)-marked polymers (Gene Tech Company Limited, Shanghai, China) or corresponding Alexa Fluor secondary antibodies (Life Technologies, Carlsbad, CA, USA) for 30 min at room temperature. HRP sections were treated with diaminobenzidine solution to visualize immunostaining; then, the sections were observed and photographed with a Zeiss microscope (AXIOSCOPE A1, Zeiss Corp, Jena, Germany). After Hoechst 33342 (1 µg/mL, Sigma–Aldrich) was added to stain the nuclei, Alexa Fluor sections were observed and photographed under confocal microscope (LSM700, Zeiss Corp). The numbers of TH-positive (TH+) neurons in the SN were counted with the aid of a superimposed grid to estimate the loss of DA-neurons. Five sections from each rat in the same atlas levels were analyzed by two researchers blind to the treatment group.

Cell Culture and siRNA Transfection

PC12 cell lines, provided by the Chinese Academy of Sciences (Shanghai, China), were incubated in Dulbecco’s modified Eagle’s medium containing 10 % fetal bovine serum and 1 % penicillin/streptomycin at 37 °C in a 5 % humidified CO2 atmosphere. Cells were plated onto a 24-well plate (1 × 105 cells/well) and allowed to grow for 24 h prior to further experimentation. Then, these cells were cultured with serum-free medium without antibiotics for 4 h, followed by transfection with negative siRNA or ASIC1a siRNAs using Lipofectamine 2000 (Life Technologies). After 24 h, the cells were collected, total protein was extracted from PC12 cells, and ASIC1a-protein expression was examined by western blot to monitor the gene-silencing effect. Next, the ASIC1a siRNA transfected cells and negative siRNA transfected cells were further treated with 6-OHDA (50 µM) or vehicle for 12 h. The levels of ASIC1a, α-SYN, and autophagic markers (LC3-II, p62) were then measured by western blot.

Statistical Analysis

All data were expressed as the mean ± standard error of the mean (SEM). Results were assessed by SPSS Version 6.1 software (SPSS, Chicago, IL, USA) using one-way analysis of variance followed by a post hoc (Turkey) analysis for multiple comparisons among groups. Differences were considered significant when the P value was less than 0.05.

Results

PF and ASICs Blockers Affected the Behavioral Symptoms in 6-OHDA-Lesioned Rats

To evaluate the curative effects of PF and ASICs blockers (PcTx1, amiloride), the successful PD model rats were selected in the 4th week following unilateral injection of 6-OHDA into the right striatum and then treated with vehicle (saline) or PF (15 or 30 or 60 mg/kg) or ASICs blockers (PcTx1 0.7 µg/kg, amiloride 10 mg/kg) for 21 continuous days. No rotational behavior was observed throughout the test in sham-operated and sham-operated plus PF (60 mg/kg) treatment rats (Fig. 1). The number of contralateral rotations in vehicle-treated 6-OHDA lesioned rats increased gradually from the 4th to 7th weeks after surgery (## P < 0.01), but this phenomenon did not occur in the 6-OHDA-lesioned rats treated with PF or ASICs blockers (PcTx1, amiloride) (*P < 0.01, **P < 0.05) (Fig. 2). These data show the remarkable neurotoxicity induced by 6-OHDA. The data also suggest that PF and ASICs blockers (PcTx1, amiloride) slowed down the progress of rotational behavior induced by APO in 6-OHDA-lesioned rats in a time-dependent manner.

Fig. 1
figure 1

Behavioral test in 6-OHDA-lesioned rats. The contralateral rotations induced by APO developed in the 6-OHDA-lesioned rats in a time-dependent manner, but occurrence of contralateral rotational behavior could not be observed in the sham-operated and sham-operated plus PF (60 mg/kg) treatment groups. Mean ± SEM. n = 10 in sham group, n = 10 in sham plus PF group, n = 12 in 6-OHDA-lesioned group, ## P < 0.01 versus week 4

Full size image
Fig. 2
figure 2

Administration with PF and ASICs-blocker (PcTx1, amiloride) improved the contralateral rotational behavior in unilaterally 6-OHDA-lesioned rats. PF (15 or 30 or 60 mg/kg) or PcTx1 (0.7 µg/kg) or amiloride (Ami, 10 mg/kg) was given daily from the 4th to 6th weeks after unilateral injection of 6-OHDA. Mean ± SEM. n = 12 in each group, ## P < 0.01 versus week 4, *P < 0.05, **P < 0.01 versus vehicle control group at same time point. Ami amiloride

Full size image

PF and ASICs Blockers Prevented Progression of the DA-Neuronal Loss in the SN of 6-OHDA-Lesioned Rats

To further investigate the effects of PF and ASICs blockers (PcTx1, amiloride) on neuronal degeneration, immunohistochemistry was used to detect the number of TH+ neurons in the SNc in unilaterally 6-OHDA-lesioned rats. The number of TH+ neurons in the intact side of 6-OHDA-lesioned and in sham-operated rats had no apparent reduction, but remarkable loss of TH+ neurons in the 6-OHDA-lesioned side was observed. In the rats, treated with PF (30 mg/kg) or ASICs blockers (PcTx1 0.7 µg/kg, amiloride 10 mg/kg) once daily for 21days, the loss of TH+ neurons was prevented (Fig. 3a). This was further confirmed by counting the number of TH+ neurons in the SN (Fig. 3b). Compared with sham rats, the number of TH+ neurons in 6-OHDA-lesioned side treated with saline was reduced by 56.3 % (**P < 0.01). The number of TH+ neurons was significantly higher in the 6-OHDA-lesioned side treated with PF and ASICs blockers (PcTx1, amiloride) than that in saline-treated rats (## P < 0.01). These findings suggest the possible neuroprotective effects of PF and ASICs blockers (PcTx1, amiloride) on DA-neuronal degeneration induced by 6-OHDA.

Fig. 3
figure 3

Effects of PF and ASICs-blocker (PcTx1, amiloride) treatment on 6-OHDA-induced TH+ neuronal degeneration. a Representative pictures of immunohistochemistry showing PF (30 mg/kg) or PcTx1 (0.7 µg/kg) or Ami (10 mg/kg) treatment alleviated TH+ neuronal loss in SNc of 6-OHDA-lesioned rats. b The number of TH+ neurons in the right side of the SNc was counted at 7 weeks after surgery; then the mean value of the TH+ neuron number calculated from four rats was regarded as the data for each group. A total of five sections were counted and analyzed in each rat. Data are expressed as the mean ± SEM, n = 4, **P < 0.01 versus sham group; ## P < 0.01 versus saline-treated 6-OHDA group. Total neurons in the SNc were observed by increased cell size. Photos were taken at ×50 magnification (Scale bar 200 µm) (a)

Full size image

PF and ASICs Blockers Treatment Reduced the Levels of DA and its Metabolites in the Striatum

Next, the levels of neurotransmitters (DA and its metabolites, 5-HT and its metabolites) in the striatum were examined with HPLC. On the whole, we found no apparent difference in the levels of DA and its metabolites between the intact sides of PD rats and the striatum of sham-group rats. Compared with those in sham-operated rats, the levels of DA and DOPAC, as well as HVA in 6-OHDA-lesioned striatum treated with saline were markedly reduced by approximately 72.1, 50.3, and 52.2 %, respectively (**P < 0.01), whereas the ratio of DA metabolites to DA was raised by about 46.8 % (Fig. 4a–d). These data suggest that 6-OHDA induced significant DA-neuronal loss in the striatum and confirmed that a successful PD rat model was established. We also observed that the levels of DA and its metabolites were significantly higher and that the ratio of DA metabolites to DA was significantly lower in the injured striatum of PD rats treated with PF (15 or 30 or 60 mg/kg) or ASICs blockers (PcTx1 0.7 µg/kg, amiloride 10 mg/kg) than the ratio in saline-treated rats (# P < 0.05, ## P < 0.01) (Fig. 4a–d). However, we did not observe a marked difference in the levels of 5-HT and its metabolites between the rats in the sham group and those in the PD group (Fig. 4e, f). These results suggest that the lesion of the DA- terminals is relatively selective and demonstrate once again that PF and ASICs blockers (PcTx1, amiloride) protected DA-neurons against 6-OHDA-induced damage.

Fig. 4
figure 4

Effects of PF and ASICs-blocker (PcTx1, amiloride) treatment on DA and its metabolites reduction in the striatum. The levels of DA and its metabolite DOPAC and HVA (a–d) and the levels of 5-HT and its metabolite 5-HIAA (e, f) in the striatum of 6-OHDA-lesioned rats were determined by HPLC. Left (intact side), indicated as a hollow column; right (injured side), indicated as black column. The results are expressed as µg/g tissue. Data are expressed as the mean ± SEM, n = 8, **P < 0.01 versus the sham group; # P < 0.05, ## P < 0.01 versus the injured striatum in saline-treated 6-OHDA group

Full size image

PF and ASICs Blockers Treatment Suppressed the Up-Regulation of ASIC1a Expression in the SN

Our previous experiment revealed that PF protected PC12 cells against MPP+ or acidic damage. Thus, we designed a series of studies to investigate more deeply the relationship between the PF’s neuroprotection and ASIC1a in an invivo model. We observed that expression of ASIC1a was remarkably higher in 6-OHDA-lesioned SN treated with saline compared to expression in the SN of sham rats (**P < 0.01) (Fig. 5a), and the increased expression of ASIC1a in lesioned SN of PD rats co-localized with TH (Fig. 5b, c). These findings indicate that the activation of ASIC1a was caused by 6-OHDA toxicity which may have induced the changes of pH. PF (30 mg/kg) and ASICs blockers (PcTx1 0.7 µg/kg, amiloride 10 mg/kg) each significantly inhibited the increase of ASIC1a expression in the lesioned SN to different degrees compared with ASIC1a expression in the saline-treated group (## P < 0.01) (Fig. 5a). These data suggest that, like ASICs blockers (PcTx1, amiloride), PF inhibited the activation of ASIC1a induced by 6-OHDA.

Fig. 5
figure 5

Effects of PF and ASICs-blocker (PcTx1, amiloride) treatment on up-regulation of ASIC1a expression in the SN. a Treatment with PF (30 mg/kg), PcTx1 (0.7 µg/kg) and Ami (10 mg/kg) for 3 weeks inhibited the increase of ASIC1a expression in 6-OHDA-lesioned SN, n = 5. b, c Seven weeks after 6-OHDA injection, this toxin induced significant increase of ASIC1a expression in the lesioned SN. Data are expressed as the mean ± SEM, n = 4. **P < 0.01 versus the sham group; ## P < 0.01 versus the injured SN in saline-treated 6-OHDA group. Pictures were observed and taken with a confocal microscope. Photos were taken at ×200 magnification (Scale bar 50 µm) (b) and ×630 magnification (Scale bar 20 µm) (c)

Full size image

PF and ASICs Blockers Inhibited the α-SYN Accumulation in the SN Induced by 6-OHDA

α-SYN aggregation likely contributes to PD pathogenesis, and the ALP is a critical route for this protein degradation [4, 5, 23–25]. A previous study in our lab demonstrated that PF protects PC12 cells by increasing the autophagic degradation of α-SYN [18]. Thus, in the present study, we continued to investigate whether PF could also reduce the level of α-SYN in vivo. We found that α-SYN expression in the 6-OHDA-lesioned side treated with saline was higher than that in sham rats (**P < 0.01) (Fig. 6a). As expected, both PF (30 mg/kg) and ASICs blockers (PcTx1 0.7 µg/kg, amiloride 10 mg/kg) dramatically reduced the α-SYN level in 6-OHDA-lesioned SN to different degrees (## P < 0.01) (Fig. 6a). These data suggest that 6-OHDA neurotoxicity can cause abnormal α-SYN accumulation, as previous studies reported [26]. Numerous studies have investigated the correlation between α-SYN overexpression and cell death. A previous report showed that overexpression of human α-SYN causes human DA neuronal death [27]. Other studies also showed that knockdown of endogenous α-SYN levels could attenuate some neurotoxicity in vitro [28, 29]. Thus, we infer that PF and ASICs blockers (PcTx1, amiloride) might protect DA neurons in vivo by alleviation of the α-SYN accumulation.

Fig. 6
figure 6

Effects of PF and ASICs-blocker (PcTx1, amiloride) treatment on α-SYN, LC3-II and p62 up-regulated expression in the SN after 6-OHDA lesion. The protein expression was tested by western blotting analysis. aTreatment with PF (30 mg/kg), PcTx1 (0.7 µg/kg) and Ami (10 mg/kg) for 3 weeks attenuated 6-OHDA-induceded increase of α-SYN expression in the SN. b, c Treatment with PF, PcTx1 and Ami for 3 weeks decreased expression of LC3-II and p62. Data are expressed as the mean ± SEM, n = 4. **P < 0.01 versus the sham group; ## P < 0.01 versus the injured SN in the saline-treated 6-OHDA group

Full size image

Our previous data showed that the MPP+-induced autophagic impairment potentially leads to α-SYN aggregation in PC12 cells. Thus, we further tested the expression of LC3-II and p62, which are autophagic markers to evaluate the autophagic activity in vivo. Western blotting showed that LC3-II and p62 levels in lesioned SN of PD rats treated with saline were markedly higher than those in sham rats (**P < 0.01) (Fig. 6b, c). However, both PF (30 mg/kg) and ASICs blockers (PcTx1 0.7 µg/kg, amiloride 10 mg/kg) significantly decreased the levels of LC3-II and p62 (## P < 0.01) (Fig. 6b, c). These results imply that, despite autophagic activation, the autophagic process in DA-neurons was incomplete, and also suggest that PF and ASICs blockers (PcTx1, amiloride) enhanced α-SYN degradation by ameliorating autophagic impairment.

ASIC1a Knockdown Alleviated the α-SYN Accumulation in PC12 Cells Induced by 6-OHDA

Because our results indicated that ASICs blockers (PcTx1, amiloride) could enhance the clearance of α-SYN, we directly examined the involvement of ASIC1a in 6-OHDA-induced α-SYN accumulation by evaluating the α-SYN expression in ASIC1a deficient cells using an RNA–interfering technique. Western blotting showed that, compared with the control siRNA cells, the transfection of ASIC1a siRNA-3 led to a significant downregulation of ASIC1a-protein expression in PC12 cells (**P < 0.01) (Fig. 7a). Thus, the siRNA-3 was used in the following experiment. After knockdown of ASIC1a, PC12 cells were incubated in the presence or absence of 6-OHDA (50 µM), and then we evaluated the ASIC1a and α-SYN expression by western blotting. In the control siRNA cells, we observed an increase in ASIC1a and α-SYN expression in 6-OHDA treated cells (*P < 0.05). However, knockdown of ASIC1a inhibited the increase of ASIC1a and a-SYN levels in 6-OHDA-induced PC12 cells (# P < 0.05, ## P < 0.01) (Fig. 7b, c). These results indicate that the α-SYN clearance was enhanced when the activation of ASIC1a was suppressed.

Fig. 7
figure 7

Down-regulated expression of α-SYN, LC3-II and p62 in PC12 cells after knockdown of ASIC1a. a PC12 cells were transfected with ASIC1a siRNA-1, siRNA-2, siRNA-3, or negative siRNA as a control. ASIC1a siRNA-3 effectively inhibited the ASIC1a protein level and was used in the following study. b, c After knockdown of ASIC1a, the increased levels of ASIC1a and α-SYN were inhibited in 6-OHDA-induced PC12 cells. d Due to knockdown of ASIC1a, levels of LC3-II and p62 significantly decreased in ASIC1a siRNA PC12 cells even with 6-OHDA (50 µM) compared with levels in 6-OHDA-treated control siRNA cells. Data are expressed as the mean ± SEM, n = 3. *P < 0.05, **P < 0.01 versus the control or control siRNA; # P < 0.05, ## P < 0.01 versus the 6-OHDA-treated control siRNA

Full size image

As the ALP is critical for clearing misfolded α-SYN [30], we examined the autophagic activity by measuring the levels of autophagic markers (LC3II and p62) after ASIC1a knockdown. Western blotting also showed the upregulation of LC3-II and p62 in the 6-OHDA-induced control siRNA cells, just like the result in the rat model (*P < 0.05). However, the levels of LC3-II and p62 decreased partly in 6-OHDA treated- PC12 cells in response to ASIC1a-gene knockdown (# P < 0.05) (Fig. 7d).

Discussion

PF can cross the blood–brain barrier and has extensive pharmacological effects in the nervous system [19, 31]. Recently, the effect of PF on DA-neuronal degeneration in PD has attracted increasing attention. Accumulating evidence suggests that PF modulates the autophagic activity, which is a main pathway to clear misfolded or aggregated proteins such as α-SYN, a possible toxic mediator of pathology in PD [24, 32, 33]. Our previous work found that both PF and amiloride could protect PC12 cells against MPP+ and acidic injury by enhancing autophagic degradation of α-SYN and that PF’s protective effects may be closely associated with its action on ASICs. All these findings encouraged us to further evaluate the therapeutic value of PF and test whether its therapeutic effect is connected with its inhibitory effect on ASIC1a in an in vivo model.

6-OHDA, has been used to induce death of DA-neurons in PD models for over 30 years [34]. Because of the similarity to DA and high affinity to DAT, 6-OHDA is often mistakenly taken up by DA-neurons [35]. The neuronal damage and eventual death induced by 6-OHDA is mainly attributable to the massive oxidative stress caused by the toxin. In this study, a model based on the unilateral striatal injection of 6-OHDA was chosen because the process of neuronal degeneration caused by 6-OHDA is similar to the clinical progression of PD [36]. As the loss of TH+ neurons in the SNc after intrastriatal 6-OHDA infusion reaches its maximum value at week 4, the onset of experimental intervention at this time point was designed to test the neuroprotective effect of PF [37].

In recent years, numerous studies have investigated ASICs. ASICs, a type of H+-gated cation channel, are characterized by acid sensitivity and can be activated by changes in pH [9, 10]. Most nervous system diseases such as cerebral ischemia, seizures and PD, are accompanied by brain tissue acidosis; given the sensitivity of ASICs to pH, the activation of ASICs may play an important role in the pathogenesis of these diseases [38–41]. Amiloride has been used as a diuretic since 1967 for the inhibition of the epithelial sodium channels (ENaCs) in the renal tubules [14]. Amiloride also blocks ASICs, members of the ENaC/Degenerin family. PcTx1 is a toxin that is extracted as a minor component (<1 % of protein content) from the venom of the South American tarantula, Psalmopoeus cambridgei [42]. Amiloride and PcTx1 have neuroprotective effects in rodent models of PD, Huntington’s disease, multiple sclerosis, and cerebral ischemia [14, 43–45]. Thus, we chose amiloride and PcTx1 as positive controls in this study. We found that PF and these ASICs blockers, amiloride and PcTx1, markedly ameliorated APO-induced rotational behavior in unilaterally 6-OHDA-lesioned rats. Meanwhile, the immunohistochemistry and HPLC data further showed that PF and the ASICs blockers (amiloride, PcTX1) alleviated the loss of DA-neurons in the SNc and decreased the levels of DA and its metabolites in the striatum of PD model rats. Though the above-mentioned data show that PF and the ASICs blockers only got the effect of partial restoration due to irreversible changes occurring before treatment initiation, we believe this limitation is consistent with the clinical condition being investigated. These results reveal that PF and the ASICs blockers could indeed protect DA-neurons in the SN against 6-OHDA-induced damage. These results, together with those from previous reports, demonstrate that PF may be a potential therapeutic drug for PD [31, 46]. These findings further suggest a possible relationship between the protective effects of PF and its action on ASICs.

Until now, no direct evidence had confirmed a decrease of pH in DA-neurons that were subjected to 6-OHDA toxin. However, the accumulated toxin in the cytoplasm results in acceleration of oxidation and inhibition of mitochondrial oxidative respiratory chain complexes I and IV, so that glycolysis becomes active and lactic acid accumulates, which may ultimately lead the pathological activation of ASICs [47]. Of course, 6-OHDA can also cause cell toxicity via other pathways [26]. In our study, western blotting showed the enhancement of ASIC1a levels in the lesioned side of model rats, and immunofluorescence also showed significantly increased expression of ASIC1a, which was co-localized with TH, in the SN with 6-OHDA-induced damage. Similar results were found in 6-OHDA-induced PC12 cells. These findings indicate pathological activation of ASIC1a induced by 6-OHDA and further suggest that ASIC1a may be related to the pathogenesis of PD, consistent with previous reports and our previous results [14, 17, 18].

Previous reports did not provide direct evidence of the relationship between the protective effects of PF and its inhibitory effects on ASICs. But our previous research revealed that PF could protect PC12 cells against acidic damage. Additionally, whole-cell patch recording configuration revealed that PF could block the acid-induced current in PC12 cells in a dose-dependent manner just as amiloride or PcTx1 did [17, 18]. In the present study, western blotting results further showed that PF reduced the ASIC1a levels significantly, just like ASICs blockers (PcTx1, amiloride). These results clearly indicate the action of PF on ASIC1s. Thus, we infer that PF may protect DA-neurons by inhibiting the activation of ASICs, especially ASIC1a.

α-SYN is one of the pathological hallmarks of PD [5]. Aggregation of abnormal proteins such as α-SYN, is toxic and may result in damage or even death of neurons [48]. Therefore, inhibition of the abnormal accumulation of α-SYN is becoming a promising therapeutic strategy for PD. In PD, intraneuronal α-SYN is maintained at normal levels via the UPS and ALP. ALP plays an especially important role in the onset and progression of PD [23–25]. To a certain extent, autophagy can protect cells against a variety of damage, but excessive activation of autophagy or autophagic dysfunction will result in the pathogenesis of some neurodegenerative diseases such as PD [49]. The level of LC-3 II reflects autophagic activity and is closely related to the quantity of autophagosomes [50]. p62/SQSTM1, one of the substrates and a ubiquitin-conjugated protein, is degraded through autophagy when bound to the LC3 and ubiquitinated protein [51, 52]. Although 6-OHDA does not produce α-SYN oligomers in the nigrostriatal pathway, several previous studies showed that 6-OHDA-induced neurotoxicity not only triggered an oxidative burst and dysregulation autophagy, but also caused α-SYN accumulation in vitro and in vivo [26, 53, 54]. We also found that the levels of α-SYN, LC-3 II, and p62 were higher in rates with 6-OHDA-induced injury than those in rats from the sham group with no exception. These data suggest that autophagic activation may be a neuroprotective mechanism to degrade aberrant protein elicited by 6-OHDA toxicity, but the disruption of autophagic flux, which is reflected by increased p62 levels will lead to the accumulation of α-SYN.

Thusfar, we know that ASICs are closely linked to the UPS [43], a protein degradation pathway, but there is no adequate evidence to verify the direct connection between ASICs and ALP activity. Our earlier studies have shown that acidosis induces α-SYN accumulation, enhancement of α-SYN autophagic degradation via blocking ASICs may contribute to protecting PC12 against MPP+ or acidic damage. An interesting discovery in this study is that PF partly reversed the effect of 6-OHDA toxin on α-SYN accumulation and disruption of autophagic flux, and ASICs blockers (PcTx1, amiloride) exhibited similar effects. Thus, these data suggest that the activation of ASIC1a may be involved in the dysfunction of the α-SYN autophagic degradation pathways.

To prove a direct relationship between ASIC1a and 6-OHDA-induced abnormal α-SYN accumulation, we first used specific ASIC1a siRNA to evaluate the expression of α-SYN, LC-3 II, and p62 in ASIC1a -suppressed PC12 cells. In agreement with our results in vivo, we found increased levels of α-SYN, LC-3 II, and p62 in 6-OHDA-induced control siRNA cells. Knockdown of ASIC1a resulted in the enhancement of α-SYN clearance and partial restoration of autophagic injury in siRNA transfected cells treated with 6-OHDA. These findings imply that activation of ASIC1a is probably linked to the impairment of α-SYN clearance, and ASIC1a may be a vital upstream regulator in the autophagic degradation process.

ASIC1a is permeable to Ca2+, so the activation of these channels can induce calcium influx in neurons [55]. A recent study found that an intracellular Ca2+ increase activates a variety of kinases and proteases associated with an autophagic signal [56]. Hcyer-Hansen et al. also demonstrated that the activation of autophagy was mediated by Ca2+/calmodulin-dependent kinase kinase-beta and adenosine monophosphate-activated protein kinase [57]. In addition, Ca2+ could regulate autophagy by the Ca2+- calpain -Gsa non-mTOR pathway [58]. Both our previous data and the present study provide evidence that PF can suppress the activation of ASICs. Thus, we believe PF’s neuroprotective effects result from its inhibition of ASIC1a activity, which in turn ameliorates autophagic dysfunction induced by calcium influx.

In conclusion, our present data clearly demonstrate that systemic administration of PF or ASICs blockers improved behavioral symptoms, delayed DA neuronal loss and attenuated the reduction in levels of DA and its metabolites in PD model. PF may be a neuroprotectant by its effect on ASIC1a; and, thus, ASIC1a may be a potential target for treatment of PD because of the close relationship between ASIC1a and autophagic degradation.