B R A I N R E S E A R C H 1 3 9 6 ( 2 0 1 1 ) 7 7 – 8 7

ava i l ab l e a t www.sc i enced i r ec t . com

www.e l sev i e r . com/ loca te /b ra i n res

Research Report

ASICs mediate the modulatory effect by paeoniflorin onalpha-synuclein autophagic degradation

Xue Suna,b, Yin-Bi Caoa, Li-Fang Hub, Ya-Ping Yanga, Jun Lib,Feng Wangb, Chun-Feng Liua,b,⁎aDepartment of Neurology, Second Affiliated Hospital of Soochow University, Suzhou 215004, ChinabLaboratory of Aging and Nervous Diseases, Institute of Neuroscience, Soochow University, Suzhou 215004, China

A R T I C L E I N F O

⁎ Corresponding author at: Department of N215004, China. Fax: +86 512 68284303.

E-mail address: liucf@suda.edu.cn (C.-F. L

0006-8993/$ – see front matter © 2011 Elsevidoi:10.1016/j.brainres.2011.04.011

A B S T R A C T

Article history:Accepted 5 April 2011Available online 13 April 2011

Acid-sensing ion channels (ASICs) are ligand-gated cation channels that respond to acidicstimuli. They are expressed throughout the mammalian nervous system. Complex subunitcombinations and lack of specific blockers of native receptors result in the difficulty ofresolving the functions of ASICs. In this study, we showed that rat pheochromocytoma cells(PC12 cells) functionally express ASICs with the activity of endogenous proton-gatedconductance. PF is the principal active ingredient extracted from the root of Paeoniae alba, aChinese herb commonly used to treat neurodegenerative disorders, especially PD. It wasfound that PF significantly up regulated the expression of LC3-II, which is specificallyassociated with autophagic vacuole membranes. PF also reduced the MPP+ and acidosis-induced accumulation of α-synuclein, the major component of Lewy bodies. Moreover, PFwas highly efficacious in modulating ASICs activity and protein expression. In addition, thedata showed that PF was able to protect PC12 cells against MPP+ and acidosis-inducedcytotoxicity. In summary, these findings demonstrate for the first time that PF couldenhance the autophagic degradation of α-synuclein by regulating the expression andactivity of ASICs and thus produces protective effects against cytotoxicity. It also offers theexperimental evidence for the potential role of ASICs in the pathogenesis of PD.

© 2011 Elsevier B.V. All rights reserved.

Keywords:ASICPaeoniflorinMPP+

α-SynucleinAutophagyPC12 cell

1. Introduction

Parkinson's disease (PD) is a neurodegenerative disorder,characterized by a selective and progressive loss of dopami-nergic neurons in the substantia nigra compacta (SNc) (Caiet al., 2009; Dauer and Przedborski, 2003; Wakabayashi et al.,2007). Although its etiology remains unclear, oxidative stressand protein misfolding and accumulation have been impli-cated in the pathogenic process of PD. Increased oxidativeradicals result in the covalent modification of α-synuclein,

eurology, Second Affiliate

iu).

er B.V. All rights reserved

leading to the conformational changes and over accumulationof this protein, a critical step in the development of PD. Post-mortem analyses of PD brains also reveal a role of oxidativestress and mitochondrial dysfunction in the dopaminergicneuron degeneration.

A significant drop of tissue pH or extracellular acidosis is acommon feature of acute neurological conditions includingischemic stroke, brain trauma, neurodegenerative diseasesand epileptic seizures. Fast drops in extracellular pH (pHe)mayactivate proton-gated/acid-sensing ion channels (ASICs) in

d Hospital of Soochow University, Sanxiang Road 1055, Suzhou

.

78 B R A I N R E S E A R C H 1 3 9 6 ( 2 0 1 1 ) 7 7 – 8 7

peripheral sensory neurons (Waldmann et al., 1997) andvarious brain neurons. Six different ASIC subunits have beencloned to date. They are encoded by four genes (ASIC1–ASIC4)(Jasti et al., 2007; Krishtal, 2003; Wang et al., 2008). Thesechannels belong to the amiloride-sensitive Na+-channel/degenerin family (Wang et al., 2008). Recent studies havedemonstrated that activation of these channels by protonsplays an important role in a variety of physiological andpathological processes such as nociception, mechanosensa-tion, synaptic plasticity, and acidosis-mediated neuronalinjury as well (Xiong et al., 2008; Ziemann et al., 2008).

Amiloride is a well-accepted blocker of epithelial sodiumchannel (ENaC), thus inhibiting sodiumre-absorption in the latedistal convoluted tubules in the kidneys. It also shows effect ontheheart by blockingNa+/H+ exchanger (sodium-hydrogen anti-porter 1 or NHE-1). Additionally, amiloride blocks ASICs,members of the ENaC/DEG family. The diuretic amiloride hasrecently been proven to be neuroprotective in models of PD, aproperty attributable to its inhibition on central ASICs. Ariaset al. (2008) tested the effect of amiloride on the 1-methyl-4-phenyl-1,2,3,6-tetrahydropyridine (MPTP)-treated mouse, amodel of PD also manifesting mild lactic acidosis. Amiloridewas found to protect SNc neurons against MPTP-induceddegeneration. 1-Methyl-4-phenylpyridinium (MPP+), the meta-bolicproductofMPTP,hasalsobeenshownto inducesymptomssimilar to those of PD in experimental animals and humans. Inaddition, several in vitro studies showed that MPP+ induces α-synucleinexpressionandaggregation (Caietal., 2009;Qianetal.,2008). Given that some of the neuronal loss is a consequence ofchronic and pathological activation of ASICs (Arias et al., 2008),

Fig. 1 – Effect of Ami or PF onα-synuclein (α-syn) and LC3 expressMPP+ (1 mM) alone for 24 h or combined with Ami or PF co-treatmby Western blotting analysis. β-actin was used as a loading condramatically reduced theα-synuclein level induced byMPP+. (B)Tmarkedly decreased when the cells were exposed to MPP+ 1.0 mThis effect was remarkably reversed by the co-treatment with 10**P<0.01 versus control; #P<0.05, ##P<0.01, ###P<0.001 versus MP

this led us to propose that the protective effects by amilorideagainst dopaminergic neuron loss may be related to itsinhibition on α-synuclein aggregation by acting on ASICs.

PF is theprincipal active ingredient extracted fromthe root ofPaeoniae alba, a traditional Chinese medicine natural product.Several studies suggest that PF has neuroprotective effectsagainst neuronal damage. Now it is widely used in clinicalischemia and stroke therapy (Dong and Xu, 2002). It is also usedto treat neurodegenerative disorders, especially PD in clinic(Huanget al., 2009).However, itsmechanism isyet to bedefined.Our previous study provides the experimental evidence that PFprotects PC12 cells fromMPP+ and acidic damage via autophagicpathway (Cao et al., 2010). Here we designed series ofexperiments to further investigate its molecular mechanismsfor its protective effects and to test whether this is related to itsaction on ASICs, just like amiloride.

2. Results

2.1. PF alleviates the α-synuclein aggregation induced byMPP+

In the previous study, we reported that PF protected PC12 cellsvia autophagic pathway. In this study,we continued to examinewhether this action by PF may also affect α-synuclein aggrega-tion because autophagic pathway is one of the major routes forprotein degradation. It was found that treatment with MPP+

(1 mM) for 24 h induced a significant increase of α-synucleinlevel in PC12 cells, as consistentwith previous reports (Cai et al.,

ion after MPP+ (1 mM) treatment. PC12 cells were treatedwithent for 24 h. The LC3 and α-syn levels were determined

trol. (A) Both amiloride (Ami, 100 μM) and PF (50 μM)he expression of LC3-II and the ratio of LC3-II over LC3-I wereM for 24 h, as compared with the vehicle-treated group.0 μM amiloride or 50 μM PF. Mean±SEM, n=3. *P<0.05,P+ group.

79B R A I N R E S E A R C H 1 3 9 6 ( 2 0 1 1 ) 7 7 – 8 7

2009). As can be seen from the representative gel result, bothamiloride (Ami, 100 μM)andPF (50 μM)dramatically reduced theα-synuclein level induced by MPP+. This is confirmed by thegroup data analysis (Fig. 1A). Of interest, PF showed a morepotent effect than amiloride did.

The previous study showed that the MPP+-induced α-synu-clein aggregation mainly resulted from its inhibitory effect onautophagy. For this reason,we then tested the expression of bothLC3-I (the cytosolic precursor of LC3-II) and LC3-II (a specificmarker of phagophores and autophagosomes). It was observedthat the expression of LC3-II and more important, the ratio ofLC3-II over LC3-I (a well-accepted autophagic marker) weremarkedly decreased when the cells were exposed to MPP+

1.0 mM for 24 h, as compared with the vehicle-treated group.This effect was remarkably reversed by the co-treatment witheither amiloride (Ami, 100 μM) or 50 μMPF. These results clearlysuggest that amiloride and PF may attenuate the abnormalaccumulation of α-synuclein by enhancing its degradation byautophagic pathway.

2.2. PF attenuates the acidosis-induced upregulation ofα-synuclein

Intriguingly, it was found that acidic treatment with mediumat pH 6.0 for 24 h also resulted in a dramatic up-regulation ofα-synuclein in PC12 cells (Fig. 2A). However, the ratio of LC3-IIto LC3-I was obviously reduced by acidic treatment, implying theacidosis inhibited the autophagic degradation of α-synuclein,

Fig. 2 – Effect of Ami or PF on α-synuclein (α-syn) and LC3 exprewith pH 6.0mediumalone for 24 h andAmi or PF for 24 h. The proRepresentative gels were shown in Fig. 2. The group data were owas used as a loading control. Acidic treatment with medium pHα-synuclein in PC12 cells (Fig. 2A). Co-treatment with Ami or PF wactivating the autophagic pathway, indicated here by the increas*P<0.05, **P<0.01, ***P<0.001 versus control; #P<0.05, ##P<0.01, #

resulting in its over accumulation in cells. Co-treatment withAmi or PF was found to abolish the enhancement of α-synucleinlevel by activating the autophagic pathway, indicated here by theincrease of the ratio of LC3-II to LC3-I (Fig. 2B).

2.3. The expression of ASICs in cultured PC12 cell

Next, we continued to examine the molecular mechanisms bywhich PF produced the beneficial effects against MPP+ andacidosis-induced cell damage and α-synuclein aggregation.Several studies have shown that chronic activation of ASICsmay lead to neuronal loss. Moreover, extracellular acidosiscan activate ASICs. In addition, amiloride is a commonly usedblocker of sodium channels including ASICs. Therefore, theseclues impel us to test whether ASICs are the potentialmolecular targets of PF. To confirm this, the expression ofASICs was determined with different methods. First, RT-PCRwas performed to verify the presence of the ASICs mRNA ofvarious isoforms. As shown in Fig. 3A, the targeted bandscorresponding to ASIC1a, ASIC1b, ASIC2a, ASIC2b, ASIC3, andASIC4 were detected in cultured PC12 cells. Furthermore, thedouble staining by confocal microscopy study with thecommercially available antibodies against ASICs showedthat the ASICs proteins (ASIC1, ASIC2, ASIC3, and ASIC4)were localized in the plasma membrane and perinuclearregion in PC12 cells (Fig. 3B). These data consistentlydemonstrated the existence of various isoforms of ASICs inPC12 cells.

ssion after pH 6.0 medium treatment. PC12 cells were treatedtein expressionwas determined byWestern blotting analysis.btained by analyzing the density of specific bands. β-actin6.0 for 24 h resulted in a dramatic up-regulation ofas found to abolish the enhancement ofα-synuclein level bye of the ratio of LC3-II to LC3-I (Fig. 2B). Mean±SEM, n=4.##P<0.001 versus acidic buffer-treated group.

Fig. 3 – Detection of ASICs transcript in PC12 cells and immunolocalization of ASICs expression in PC12 cells.(A) Nonquantitative RT-PCR analysis of mRNA isolated from PC12 cells revealed the presence of mRNA for ASIC1a, 1b, 2a, 2b, 3,4 subunits. (B) PC12 cells were permeabilized and immunostained with various anti-ASICs: ASIC1, ASIC2, ASIC3, and ASIC4antibodies, followed by Cy3-labeled secondary antibodies for imaging, and nuclei were counterstained with DAPI. The proteinexpression of various ASIC subunits was assessed by immunofluorescence study combined with confocal microscopy. Wefound ASICs proteins are localized in the plasma membrane and perinuclear region in PC12 cells.

80 B R A I N R E S E A R C H 1 3 9 6 ( 2 0 1 1 ) 7 7 – 8 7

2.4. Acid activates inward current in PC12 cells

The expression and function of ASICs in PC12 cells wereverified by electrophysiological recording. It was observed that

a rapid drop of pHe from 7.4 to 6.0 evoked a transient, rapidlyinactivating inward current (Fig. 4A), in line with the previousreport. The amplitude of ASIC current in PC12 cells decreasedslightly following the formation of whole cell configuration,

Fig. 4 – Electrophysiological properties of acid-sensing ion channels (ASICs) in PC12 cells. A: Currents were recorded inwhole cell configuration at −60 mV. Representative traces showed the pH-dependent activation of ASIC currents in thedifferentiated PC12 cells. The amplitude of inward current increases with smaller drops of pH with maximal responsesrecorded at pH 5.0–6.0. B: Amiloride blocked the ASIC currents in PC12 cells. Representative traces showed the dose-dependentblockade of ASIC currents activated by pH drop from 7.4 to 6.0. Holding potential is −60 mV. C: Proton-activated currents inPC12 cells are blocked by a specific ASIC1a blocker PcTX1. The representative current recording shows the dose-dependenteffect of PcTX1 on ASIC currents. Currents were activated by pH drop to 6.0 from 7.4 at the membrane potential of −60 mV.D: ASIC currents in PC12 cells were also blocked by paeoniflorin. Representative recordings showed the dose-dependentblockade of ASIC currents by paeoniflorin. Currents were activated by pH drop from 7.4 to 6.0 atmembrane potential of −60 mV.

81B R A I N R E S E A R C H 1 3 9 6 ( 2 0 1 1 ) 7 7 – 8 7

similar to the ASIC currents reported by other investigators.However, in most cells the currents stabilized after 5–10 minand remained stable for more than 30 min. For this reason,most experiments were performed 10 min after the formationof whole cell configuration. The current completely decayedwithin 3–4 s with a single exponential time course. Theamplitude of peak current increased in a sigmoidal fashionas pHe decreased. Inmost PC12 cells, the threshold pHe to elicitthe inward current was about 7.4 and the maximum responseappeared at 5.0–6.0.

2.5. PF blocks ASIC currents in PC12 cells

Amiloride and psalmotoxin-1 (PcTX1, a specific inhibitor ofASIC1a) were used here as positive controls to test the actionof PF on proton-gated currents in PC12 cells. Inward currentswere activated by a pHe drop from 7.4 to 6.0 at a holdingpotential of −60 mV. After recording of two to three consec-utive traces that had similar amplitude, cumulative concen-trations of amiloride were added to both pH 7.4 and pH 6.0solutions. Enough time was allowed for the effect of each

82 B R A I N R E S E A R C H 1 3 9 6 ( 2 0 1 1 ) 7 7 – 8 7

concentration of amiloride to be stabilized. When the finalconcentration of amiloride stabilized, this drug was washedout. Similar to the proton-gated current in sensory and centralneurons, these proton-gated currents in PC12 cells werereversibly inhibited by amiloride in a dose-dependent manner(Fig. 4B). In addition, proton-activated current was also foundto be dose-dependently blocked by PcTX1, a specific inhibitorof ASIC1a, as shown in Fig. 4C. Likewise, PF was found to blockthe ASICs current in PC12 cells in a dose-dependent manner,as shown in Fig. 4D. Moreover, the blockade of ASICs currentby PFwas reversible upon thewashout of this drug. These dataclearly show that the ASICs currents in PC12 cells can be

Fig. 5 –Western blotting analysis of acid-sensing ion channels (ASPC12 cells were exposed to pH 6.0 medium alone for 24 h in theCompared to the group treatedwith acidicmediumat pH 6.0 alonecells. Similarly, treatment with 50 μM PF also remarkably alleviastress. Mean±SEM, n=3. *P<0.05, **P<0.01, ***P<0.001 versus congroup.

regulated by PF, implying its neuroprotective effects onneuronal damage by inhibiting ASICs.

2.6. PF attenuates the up-regulation of ASIC expression inacidic medium at pH 6.0

Series of studies were further performed to test the effect of PFon the expression of ASICs, in addition to the ASICs current. Itwas shown that treatmentwith acidicmediumat pH6.0 for 24 hsignificantly induced the up-regulation of ASIC of the fouridentified isoforms in PC12 cells, as compared with the controlgroup. Compared to the group treatedwith acidicmediumat pH

ICs) ASIC1 (A), ASIC2 (B), ASIC3 (C), andASIC4 (D) in PC12 cells.presence and absence of co-treatment with amiloride or PF., amiloride (100 μM) reduced the ASICs up-regulation in PC12ted the increase of ASICs expression induced by acidictrol; #P<0.05, ##P<0.01, ###P<0.001 versus acidic buffer-treated

83B R A I N R E S E A R C H 1 3 9 6 ( 2 0 1 1 ) 7 7 – 8 7

6.0 alone, amiloride (100 μM) reduced ASIC up-regulation inPC12 cells (Fig. 5). Similarly, treatment with 50 μM PF alsoremarkably alleviated the increase of ASIC expression inducedby acidic stress (Fig. 5). These data indicate that PF inhibit boththe expression of ASICs and its current induced by acidic stress.

2.7. PF protects against MPP+ and acidosis-induced celldeath in differentiated PC12 cells

Furthermore, the potential effects of PF and amiloride onMPP+

and acidosis-induced cytotoxicity in PC12 were also assessedby measuring the cell viability with MTT assay. Exposure ofPC12 cells to MPP+ (1.0 mM, 24 h) induced a significantdecrease of cell viability. Co-treatment with amiloride(100 μM) was found to rescue the cells from MPP+-inducedinjury (Fig. 6A). Co-treatment with PF produced similarprotective effects (Fig. 6B). Moreover, amiloride and PF werealso able to protect the differentiated PC12 cells againstacidosis-induced cell death (Fig. 6C and D).

Fig. 6 – Protective effect of PF on cytotoxicity in PC12 cells inducedsignificantly decreased after exposure to MPP+ (1 mM) for 24 h; hsignificantly protected cells against the cell death. (B) PF (50 μM)cytotoxicity. (C) The cell viability was also reduced after 24 h expoblocked this effect. (D) Exposure of PC12 cells to acidicmediumof pbe blocked by co-treatment of 50 μM PF. Mean±SEM, n=4. *P<0.###P<0.001 versus MPP+ or acidic buffer-treated group.

3. Discussion

Our previous paper reported that MPP+ cytotoxicity is associ-ated with the over accumulation of the oligomeric species ofα-synuclein (Cai et al., 2009; Cao et al., 2010; Qian et al., 2008).The rapid removal of misfolded or aggregated proteins is a keyproblem in neurons, as many neurodegenerative disordershave been linked to the aberrant accumulation of such aggre-gates. Autophagy plays an important role in the clearance ofaggregate-prone mutant proteins (Levine and Kroemer, 2008).Our present findings clearly demonstrate that PF can attenuateMPP+-induced cytotoxicity in differentiated PC12 cells and theα-synuclein accumulation by enhancing its autophagic degrada-tion pathway. Furthermore, the beneficial effects were found tobe attributable to its inhibitory effect on ASICs.

The presence of ASICs in the brain, which lacks nociceptors,indicates these channels may have important functions in thesetting of acidosis beyond nociception (Jasti et al., 2007; Krishtal,

by MPP+/pH 6.0 medium treatment. (A) The cell viability wasowever, co-treatment with amiloride (Ami, 100 μM)produced similar protective effects against MPP+-inducedsure to acidic medium of pH 6.0, however, amiloride (100 μM)H 6.0 induced amarked decrease of cell viability, which could05, **P<0.01, ***P<0.001 versus control; #P<0.05, ##P<0.01,

84 B R A I N R E S E A R C H 1 3 9 6 ( 2 0 1 1 ) 7 7 – 8 7

2003). As tissueacidosis is a common featureof cerebral ischemiaand epilepsy, activation of ASICs in brain neurons may act as amediator of pathological stimuli (Pidoplichko and Dani, 2006;Pignataro et al., 2007; Xiong et al., 2007). In the present study, bothMPP+ and acidic buffer were shown to upregulate α-synucleinexpression and alter the LC3 level. It may be argued that MPP+

treatment is causing acidification of the extracellular space,making this the common mechanistic pathway for both treat-ments. But no significant decrease of pH was detected in theculture subjected to MPP+ exposure for a period up to 24 h(unpublished data). Thus, MPP+ may affect the expression ofthese proteins via the mechanisms other than acidification ofextracellular space. In fact, Cai et al. (2009) reported that MPP+

may alter the autophagy activity and thus induce the overaccumulation of α-synuclein by impairing the dynein activity. Itwas also reported that othermechanisms such asmitochondrialcomplex I inhibition and increased production of reactive oxygenspecies and nitric oxide may be involved in MPP+-induced α-synuclein aggregation and cytotoxicity (Huang et al., 2010; Kim etal., 2007). This suggested that MPP+ can cause cell toxicity viamultiple pathways. Nevertheless, PF was found to affordsignificant protection against acidosis or MPP+-induced injury inPC12 cells. The findings further demonstrated that PF ameliorat-ed acidosis-induced ASICs activation, attenuated MPP+-inducedautophagy inhibition and prevented α-synuclein accumulation.All these data strongly suggest that PF may become a potentialneuroprotectant by modulating ASICs activity.

Previous studies reported that ASIC protein expression islimited to neuronal and some sensory epithelial tissue(Krishtal, 2003). However, recent studies indicated ASICmolecules were also expressed in other cells such as PC12cells. The researchers identified the presence of ASIC1asubunits by RT-PCR and electrophysiology (Chu et al., 2002).Inline with this, our results also found the existence of ASIC1ain PC12 cells using different methods. The acidosis-inducedcurrent was not only blocked by amiloride but inhibited byPcTX1, a specific inhibitor of ASIC1a. In addition, our resultsidentified the expression of other subunits of ASICs includingASIC1b, ASIC2a, ASIC2b, ASIC3 and ASIC4. These data impliesthat the proton gated currents in PC12 cells are not onlymediated by homomeric ASIC1a. These findings may alsohave important implications for the ASICs-related studies.PC12 cells may be an ideal cell line for the study ofphysiological and potential pathological roles of ASICs.Specifically, the combination of various approaches clearlydemonstrates the localization of ASICs to the plasma mem-brane and perinuclear area of specific cell populations. Theplasma membrane and perinuclear staining patterns suggestthat ASICs tend to be retained in a plasma membrane andperinuclear intracellular pool. A similar distribution profilehas also been observed for ENaC proteins in PC12 cells. ASICsubunits can associate in a variety of combinations to formhomo- and hetero meric channels, each with differentkinetics, external pH sensitivity, and tissue distribution(Krishtal, 2003; Wemmie et al., 2006; Wu et al., 2004). However,due to complex subunit composition, possible interaction ofprotons with other ion channels/receptors, and the lack ofspecific blockers for various ASICs, it is difficult to assess thephysiological and pathological roles of specific ASICs in nativeneurons.

Xu and his colleague reported that Radix paenoniae rubrapotently and specifically blocks the sodium current in acutelydissociated rat hippocampal CA1neurons (Dong andXu, 2002). Inthis study,wedemonstrated thatPF, asimilar compoundtoRadixpaenoniae rubra, is effective in modulating ASICs activity.Treatment with pH 6.0 medium significantly induced theactivation of ASICs and inhibited autophagy in PC12 cells.However, co-treatment with PF (50 μM) down regulated theASICs but up-regulated the autophagy. ASICs are H+-gated cationchannels, which belong to the degenerin/epithelial sodiumchannel super family (Kellenberger et al., 2002; Wang et al.,2008). They are activated quickly as the extracellular pH falls,contributing to neuronal cell death associated with extracellularacidosis, a pathogenic progress associated with different disor-ders such as seizures, cerebral ischemia, PD andHD. Amiloride, anon-specific inhibitor of ASICs and PcTX1, a specific inhibitor ofASIC1a, have been shown to beneuroprotective in rodentmodelsof PD, HD, multiple sclerosis, and traumatic brain injury (Ariaset al., 2008; Joch et al., 2007; Pidoplichko and Dani, 2006; Wonget al., 2008; Xiong et al., 2006; Xiong et al., 2007; Ziemann et al.,2008). It is therefore reasonable to believe that the blockade ofASICs may produce neuroprotective effects.

Lewy bodies and Lewy neuritis are major pathologicalhallmarks of PD (Wakabayashi et al., 2007). Importantly, α-synuclein is the major component of these proteinaceousinclusions. Autophagy is considered as the main responsiblemechanism for the removal of misfolded proteins such as α-synuclein (Cherra et al., 2010). Early studies showed that thedamaged proteins by acidosis may form toxic aggregates andultimately induce neurodegeneration in PD (Cao et al., 2010).Anti-acidosis is thereby expected to decrease α-synucleinoligomer content and prevent protein carbonyl ate formation.Furthermore, acidosis-induced ASICs activation is likely in-volved in the pathogenic progress of α-synuclein accumulation.

In summary, our data demonstrate for the first time thatASIC1, ASIC2, ASIC3 and ASIC4 are expressed in cultured PC12cells. Moreover, our data also show that ASIC protein expressionand its current can be regulated by PF as well as amiloride. Inaddition, the findings show that PFmay alleviate the α-synucleinaggregation and cytotoxicity induced by both MPP+ and acidicconditions. Therefore, the present study clearly presents that PFmay become a promising neuroprotectant by regulating ASICsactivity.

4. Experimental procedures

4.1. PC12 cell culture and treatments

A differentiated rat pheochromocytoma (PC12) cell line waspurchased from Shanghai Institute of Cell Biology, ChineseAcademy of Sciences (Shanghai, China). PC12 cells werecultured in 25 cm2 culture flasks at 37 °C under an atmosphereof 5% CO2/95% air in complete Dulbecco's Modified Eagle'sMedium (Gibco, USA), supplemented with 10% heat-inactivatedfetal bovine serum (Hyclone) and 1% penicillin/streptomycin.Mediumwas replaced every 2–3 days.

For the experiments, the cells were typsinized, harvestedand re-seeded in different culture plates in accordancewith thesubsequent experiments. After 24 h, the cultures were treated

85B R A I N R E S E A R C H 1 3 9 6 ( 2 0 1 1 ) 7 7 – 8 7

with either MPP+ (1.0 mM) or acidicmedium at pH 6.0 for 24 h inthe presence or absence of co-treatment of amiloride hydro-chloride hydrate (Ami, Sigma-Aldrich, USA) and paeoniflorin(PF, monomer, purity>98%, National Institute for the Control ofPharmaceutical and Biological Products of China, Beijing). Cellstreated with the corresponding vehicle served as the control.At the end of treatment, the cells cultured in 96-well plateswere used for MTT assay and those in 6-well plates were usedfor RT-PCR andWestern blot analyses while the cells in 24-wellplates were used for immunofluorescence microscopy andelectrophysiology.

4.2. MTT assay for cell viability

Cell viability was determined using 3-[4,5-dimethylthiazol-2-]-2,5-diphenyltetrazolium bromide (MTT, Sigma-Aldrich) assay.In brief, the culture supernatant was removed at the end oftreatment. Fresh culture medium (180 μl) was then added,followed by the addition of MTT (20 μl, 5 mg/ml) and incuba-tion for 2 h at 37 °C. After that, the supernatant was removedand DMSO (150 μl) was added to each well to dissolve theformazan and the absorbance was measured at 570 nm witha microplate reader (Model 680, BioRad, USA). PC12 cellsreceiving vehicle treatment served as control group. The cellviability was expressed as the percentage of the value againstthe control group.

4.3. Electrophysiology

The electrophysiological recordings were performed with theconventional whole-cell patch recording configuration undervoltage-clamp conditions. Patchpipetteswere pulled fromglasscapillaries with an outer diameter of 1.5 mm on a two-stagepuller (PP83, Narishige, Tokyo, Japan). The tips of the electrodeswere normally heat polished on a Narishige microforge(Scientific Instruments Laboratory, Tokyo, and model MF-83)to a final diameter of 1–2 μm. The patch electrodes hadresistance between 3 and 5 MΩ. When filled with intracellularsolution, resistance was compensated by 50%–95% in voltage-clamp experiments. Whole cell currents were recorded usingAxopatch 1-D amplifiers (Axon Instruments, Foster City, CA) inthe voltage-clamp mode. Data were filtered at 2 kHz anddigitized on-line using Digidata 1320A DAC units (Axon In-struments). The on-line acquisition was done using pClampsoftware pClamp 12.0 (Axon Instruments). The membranepotential was held at 60mV throughout the experiments. Allexperiments were carried out at room temperature (22–25 °C).

PC12 cells were washed three times and bathed in extracellular solution containing (in mM) 150 NaCl, 5.0 KCl, 10HEPES, 10 glucose, 2 CaCl2, and 2.0MgCl2, pH 7.4 usingNaOH orHCl; 320–335 mOsm (Na-rich solution). A maximal responsewas achieved with pH at 6.0 in most PC12 cells. Patchelectrodes contained 150 mM KCl, 2.0 mM MgCl2, 1.0 mMCaCl2, 10 mM HEPES, 10 mM EGTA, and 4 Mg ATP, pH 7.2,using KOH (Na-rich solution), 300 mOsm. Drugs used in thepresent experiments were purchased from Sigma except forpaeoniflorin. PcTX1 from tarantula Psalmopoeus cambridgeiwaspurchased from Alomone Labs (RTP-100).

Drugs were applied using a rapid application techniquetermed the “Y-tube” method throughout the experiment (Wu

et al., 2004). This system allows a complete exchange ofexternal solution surrounding a cell within 20 ms. Aftertransient pH drops, PcTx1, Ami or PF solution was preappliedextracellularly for 10 s before and during new pH changes.

4.4. Western blotting analyses for ASICs, LC3 andα-synuclein proteins

At the end of treatment, cells were collected, washed threetimes with ice-cold PBS, and lysed by adding 50 μl lysisbuffer (Beyotime Institute of Biotechnology, China), followedby a 10-min incubation on ice. The samples were sonicated for16 s and then centrifuged at 13,400g×15 min at 4 °C. After that,protein concentration of the supernatants was determinedusing the BCA ProteinAssay Kit (Beyotime). Samples containing30 μg proteins were mixed with sodium dodecyl sulfate (SDS)-Laemmli sample buffer (Beyotime) and boiled for 5 min todenature. Samples were then fractionated on an SDS–PAGE gel(12% for LC3 and 10% for ASICs) and transferred onto PVDFmembranes (Millipore, Bedford, MA). Themembrane blots wereblocked for 2 hwith 5%non-fatmilk in TBST (50 mmol/l Tris, pH7.6, 150 mmol/l NaCl, 0.1% Tween 20) and then incubated withanti-α-syn or polyclonal rabbit anti-LC3B antibody (1:1000,Abcam, MA) or monoclonal mouse anti-beta-actin antibody(1:1000, Beyotime) at 4 °C overnight. The sources of differentprimary antibodies against ASICs subunits are listed as follows:ASIC1 (Santa Cruz: E-15: sc-13903); ASIC2 (Alomone labs: ASC-012); ASIC3 (Alomone labs: ASC-018); ASIC4 (Santa Cruz: N-20:sc-22324). Moreover, different dilutions of these primaryantibodieswere used inWestern blotting analysis. For example,the antibodies against ASIC1, ASIC2 and ASIC4 were used in1:1000 whilst anti-ASIC3 was used in a dilution of 1:1500. Afterwashing 3×10min with TBST, immunolabeling was detectedusing the TMB stabilized substrate for HRP (Promega). Mem-branes probed with antibodies and actin were used as loadingcontrols. The images were captured and densitometric analysiswas performed using Image J software (NIH, Bethesda, MD,USA).

4.5. Reverse transcription PCR

PC12 cells were transferred to a 15-ml Falcon tube andcollected by centrifugation. Cells were washed twice andresuspended in 100 μl PBS. Total RNA was collected with theRNA SafeKit (Bio 101, Vista, CA), Poly(A+) mRNA was thenisolated with the mRNA Kit Oligo [dT]30 (Bio 101, Vista, CA).The concentration of mRNA was determined by spectropho-tometry, cDNAwas prepared from0.2 μg PC12mRNAusing theThermoscript RT-PCR Kit with oligo dT primers. Reaction wasperformed in 50 μl volume containing 1 time buffer, 1.5 mMMgCl2, 0.2 mM dNTP, 0.4 μM sense and antisense primers(Table 1), 2 μl cDNA, and 2 U Platinum Taq DNA Polymerase(Life Technologies). PCR was carried out for 35 cycles ingeneral, using the following programs: ASIC1a (30 s at 94 °C;2 min at 70 °C); ASIC2a (30 s at 94 °C; 2 min at 67 °C; 1 min at68 °C); ASIC3 (30 s at 94 °C; 1 min at 60 °C; 1 min at 72 °C). ThecDNA from rat brain was used as positive control for ASIC1aamplification. For ASIC2a and ASIC3, transcripts cloned intoPCDNA3 plasmids were used as the positive control. Reactionswere analyzed on a 2% agarose gel.

Table 1 – Primers used in PCR and length of products.

Sense primer Antisense primer Length (bp)

ASIC1a GCCCACATCTTCTCCTAT CTTGGTGACGTGGTGATA 212ASIC1b CGAAGCAGGCATCAAAGT CGAAGAAATCCGAGTCCAT 179ASIC2a GCAACCTGACACGCTACAA GGCAGCAACTTCATACGC 191ASIC2b ACCTGACACGCTACAACAA GCAACTTCATACGCCTTC 184ASIC3 ATACAACCGCAGCGAGTC TCTTCCTGGAGCAGAGTGTT 246ASIC4 TCGCTACCCAGAGCCTGACA CTCTGAGGATCTGCATTGAAGGTG 168β-actin TCAGGTCACTATCGGCAAT AAAGAAAGGGTGTAAAACGCA 432

86 B R A I N R E S E A R C H 1 3 9 6 ( 2 0 1 1 ) 7 7 – 8 7

4.6. Immunofluorescence microscopy

For immunofluorescence microscopy, cells seeded on coverslipswere cultured in 24-well dishes. After treatment with acidicmedium at pH 6.0 for 24 h, cells were fixed in 4% paraformalde-hyde (Sigma) for 15 min, washed with PBS, and then permeabi-lizedwith 0.1%Triton-X-100(Sigma) for 15 min. Theprimary anti-ASICs (ASIC1,ASIC2,ASIC3,ASIC4) antibody (1:250dilution, SantaCruz; Alomone Labs) was added into the cells and incubated atroom temperature for 2 h and Cy3-conjugated anti-goat anti-bodies (1:500; Beyotime Institute of Biotechnology) were usedfor 1 h. Finally, after the cellswere incubatedwith4′,6-diamidino-2-phenylindole: DAPI (3 μg/ml, Sigma) for 10min, the slides weremounted with Citiflour (Citiflour Ltd.). The cells were thenvisualized under a Leica TSC ST2 confocal microscope.

4.7. Statistics

All data are expressed as t-test. Two-group comparisons wereperformed using Student's t-test. Multiple-group comparisonswere performed using one-way analysis of variance andFisher's least significant difference (equal variances assumed) orDunnett's T3-test (equal variances not assumed). A probabilityless than 0.05 was considered statistically significant.

Acknowledgments

We are grateful to Dr. Jin Tao of Soochow University for hisadvice, generosity and extraordinary support with facilitiesand reagents. We also thank professor Tian-le Xu in Labora-tory of Synaptic Physiology, Institute of Neuroscience ofChinese Academy of Sciences. We also thank Xing-shun Xufor critical reading of the paper. This work is supported bygrants from the Jiangsu Province Foundation of China.

R E F E R E N C E S

Arias, R.L., Sung,M.L., Vasylyev, D., Zhang,M.Y., Albinson, K., Kubek,K., Kagan,N., Beyer,C., Lin,Q.,Dwyer, J.M., Zaleska,M.M., Bowlby,M.R., Dunlop, J., Monaghan, M., 2008. Amiloride isneuroprotective in an MPTP model of Parkinson's disease.Neurobiol. Dis. 31, 334–341.

Cai, Z.L., Shi, J.J., Yang, Y.P., Cao, B.Y., Wang, F., Huang, J.Z., Yang,F., Zhang, P., Liu, C.F., 2009. MPP+ impairs autophagic clearanceof alpha-synuclein by impairing the activity of dynein.Neuroreport 20, 569–573.

Cao, B.Y., Yang, Y.P., Luo, W.F., Mao, C.J., Han, R., Sun, X., Cheng, J.,Liu, C.F., 2010. Paeoniflorin, a potent natural compound,protects PC12 cells from MPP+ and acidic damage viaautophagic pathway. J. Ethnopharmacol. 131, 122–129.

Cherra III, S.J., Kulich, S.M., Uechi, G., Balasubramani, M.,Mountzouris, J., Day, B.W., Chu, C.T., 2010. Regulation of theautophagy protein LC3 by phosphorylation. J. Cell Biol. 190,533–539.

Chu, X.P., Miesch, J., Johnson, M., Root, L., Zhu, X.M., Chen, D.,Simon, R.P., Xiong, Z.G., 2002. Proton-gated channels in PC12cells. J. Neurophysiol. 87, 2555–2561.

Dauer, W., Przedborski, S., 2003. Parkinson's disease: mechanismsand models. Neuron 39, 889–909.

Dong, X.P., Xu, T.L., 2002. Radix paeoniae rubra suppression ofsodium current in acutely dissociated rat hippocampal CA1neurons. Brain Res. 940, 1–9.

Huang, J.Z., Chen, Y.Z., Su, M., Zheng, H.F., Yang, Y.P., Chen, J., Liu,C.F., 2010. DL-3-n-butylphthalide prevents oxidative damageand reduces mitochondrial dysfunction in an MPP+-inducedcellular model of Parkinson's disease. Neurosci. Lett. 475,89–94.

Huang, K.S., Lin, J.G., Lee, H.C., Tsai, F.J., Bau, D.T., Huang, C.Y.,Yao, C.H., Chen, Y.S., 2009. Paeoniae alba Radix promotesperipheral nerve regeneration. Evid. Based ComplementAlternat. Med.

Jasti, J., Furukawa, H., Gonzales, E.B., Gouaux, E., 2007. Structure ofacid-sensing ion channel 1 at 1.9 A resolution and low pH.Nature 449, 316–323.

Joch, M., Ase, A.R., Chen, C.X., MacDonald, P.A., Kontogiannea, M.,Corera, A.T., Brice, A., Seguela, P., Fon, E.A., 2007.Parkin-mediated monoubiquitination of the PDZ protein PICK1regulates the activity of acid-sensing ion channels. Mol. Biol.Cell 18, 3105–3118.

Kellenberger, S., Gautschi, I., Schild, L., 2002. An external sitecontrols closing of the epithelial Na+ channel ENaC. J. Physiol.543, 413–424.

Kim, Y.J., Ko, H.H., Han, E.S., Lee, C.S., 2007. Lamotrigine inhibitionof rotenone- or 1-methyl-4-phenylpyridinium-inducedmitochondrial damage and cell death. Brain Res. Bull. 71,633–640.

Krishtal, O., 2003. The ASICs: signaling molecules? Modulators?Trends Neurosci. 26, 477–483.

Levine, B., Kroemer, G., 2008. Autophagy in the pathogenesis ofdisease. Cell 132, 27–42.

Pidoplichko, V.I., Dani, J.A., 2006. Acid-sensitive ionic channels inmidbrain dopamine neurons are sensitive to ammonium,which may contribute to hyperammonemia damage. Proc.Natl. Acad. Sci. USA 103, 11376–11380.

Pignataro, G., Simon, R.P., Xiong, Z.G., 2007. Prolonged activation ofASIC1a and the time window for neuroprotection in cerebralischaemia. Brain 130, 151–158.

Qian, J.J., Cheng, Y.B., Yang, Y.P., Mao, C.J., Qin, Z.H., Li, K., Liu, C.F.,2008. Differential effects of overexpression of wild-type andmutant human alpha-synuclein on MPP+-inducedneurotoxicity in PC12 cells. Neurosci. Lett. 435, 142–146.

87B R A I N R E S E A R C H 1 3 9 6 ( 2 0 1 1 ) 7 7 – 8 7

Wakabayashi, K., Tanji, K., Mori, F., Takahashi, H., 2007. The Lewybody in Parkinson's disease: molecules implicated in theformation and degradation of alpha-synuclein aggregates.Neuropathology 27, 494–506.

Waldmann, R., Champigny, G., Bassilana, F., Heurteaux, C.,Lazdunski, M., 1997. A proton-gated cation channel involved inacid-sensing. Nature 386, 173–177.

Wang, Y., Apicella Jr., A., Lee, S.K., Ezcurra, M., Slone, R.D., Goldmit,M., Schafer, W.R., Shaham, S., Driscoll, M., Bianchi, L., 2008. Aglial DEG/ENaC channel functions with neuronal channelDEG-1 tomediate specific sensory functions in C. elegans. EMBOJ. 27, 2388–2399.

Wemmie, J.A., Price, M.P., Welsh, M.J., 2006. Acid-sensing ionchannels: advances, questions and therapeutic opportunities.Trends Neurosci. 29, 578–586.

Wong, H.K., Bauer, P.O., Kurosawa, M., Goswami, A., Washizu, C.,Machida, Y., Tosaki, A., Yamada, M., Knopfel, T., Nakamura, T.,Nukina, N., 2008. Blocking acid-sensing ion channel 1 alleviatesHuntington's disease pathology via an ubiquitin–proteasome

system-dependent mechanism. Hum. Mol. Genet. 17,3223–3235.

Wu, L.J., Duan, B., Mei, Y.D., Gao, J., Chen, J.G., Zhuo, M., Xu, L., Wu,M., Xu, T.L., 2004. Characterization of acid-sensing ionchannels in dorsal horn neurons of rat spinal cord. J. Biol.Chem. 279, 43716–43724.

Xiong, Z.G., Chu, X.P., Simon, R.P., 2006. Ca2+-permeableacid-sensing ion channels and ischemic brain injury. J. Membr.Biol. 209, 59–68.

Xiong, Z.G., Chu, X.P., Simon, R.P., 2007. Acid sensing ionchannels—novel therapeutic targets for ischemic brain injury.Front. Biosci. 12, 1376–1386.

Xiong, Z.G., Pignataro, G., Li, M., Chang, S.Y., Simon, R.P., 2008.Acid-sensing ion channels (ASICs) aspharmacological targets forneurodegenerative diseases. Curr. Opin. Pharmacol. 8, 25–32.

Ziemann, A.E., Schnizler, M.K., Albert, G.W., Severson, M.A.,Howard III, M.A., Welsh, M.J., Wemmie, J.A., 2008. Seizuretermination by acidosis depends on ASIC1a. Nat. Neurosci. 11,816–822.