NeuroToxicology 54 (2016) 161–169 Contents lists available at ScienceDirect NeuroToxicology Full length article Rotenone down-regulates HSPA8/hsc70 chaperone protein in vitro: A new possible toxic mechanism contributing to Parkinson’s disease Gessica Salaa,f,* , Daniele Mariniga,b,f , Chiara Rivaa,f , Alessandro Arosioa,f , Giovanni Stefanonia,c , Laura Brighinac,f , Matteo Formentid,e , Lilia Alberghinad,e,f , Anna Maria Colangelod,e,f , Carlo Ferraresea,c,f a Lab. of Neurobiology, School of Medicine and Surgery, University of Milano-Bicocca, Milano, Italy PhD Program in Neuroscience, University of Milano-Bicocca, Milano, Italy c Dept. of Neurology, San Gerardo Hospital, Monza, Milano, Italy d Lab. of Neuroscience R. Levi-Montalcini, Dept. of Biotechnology and Biosciences, University of Milano-Bicocca, Milano, Italy e SYSBIO Centre of Systems Biology, University of Milano-Bicocca, Milano, Italy f NeuroMI Milan Center for Neuroscience, University of Milano-Bicocca, Milano, Italy b A R T I C L E I N F O Article history: Received 4 November 2015 Received in revised form 30 March 2016 Accepted 26 April 2016 Available online 28 April 2016 Keywords: Heat shock cognate protein 70 Rotenone Parkinson’s disease Autophagy A B S T R A C T HSPA8/hsc70 (70-kDa heat shock cognate) chaperone protein exerts multiple protective roles. Beside its ability to confer to the cells a generic resistance against several metabolic stresses, it is also involved in at least two critical processes whose activity is essential in preventing Parkinson’s disease (PD) pathology. Actually, hsc70 protein acts as the main carrier of chaperone-mediated autophagy (CMA), a selective catabolic pathway for alpha-synuclein, the main pathogenic protein that accumulates in degenerating dopaminergic neurons in PD. Furthermore, hsc70 efficiently fragments alpha-synuclein fibrils in vitro and promotes depolymerization into non-toxic alpha-synuclein monomers. Considering that the mitochondrial complex I inhibitor rotenone, used to generate PD animal models, induces alpha-synuclein aggregation, this study was designed in order to verify whether rotenone exposure leads to hsc70 alteration possibly contributing to alpha-synuclein aggregation. To this aim, human SH-SY5Y neuroblastoma cells were treated with rotenone and hsc70 mRNA and protein expression were assessed; the effect of rotenone on hsc70 was compared with that exerted by hydrogen peroxide, a generic oxidative stress donor with no inhibitory activity on mitochondrial complex I. Furthermore, the effect of rotenone on hsc70 was verified in primary mouse cortical neurons. The possible contribution of macroautophagy to rotenone-induced hsc70 modulation was explored and the influence of hsc70 gene silencing on neurotoxicity was assessed. We demonstrated that rotenone, but not hydrogen peroxide, induced a significant reduction of hsc70 mRNA and protein expression. We also observed that the toxic effect of rotenone on alpha-synuclein levels was amplified when macroautophagy was inhibited, although rotenone-induced hsc70 reduction was independent from macroautophagy. Finally, we demonstrated that hsc70 gene silencing up-regulated alpha-synuclein mRNA and protein levels without affecting cell viability and without altering rotenone- and hydrogen peroxide-induced cytotoxicity. These findings demonstrate the existence of a novel mechanism of rotenone toxicity mediated by hsc70 and indicate that dysfunction of both CMA and macroautophagy can synergistically exacerbate alpha-synuclein toxicity, suggesting that hsc70 up-regulation may represent a valuable therapeutic strategy for PD. ã 2016 Elsevier Inc. All rights reserved. 1. Introduction * Corresponding author at: Lab. of Neurobiology, Milan Center for Neuroscience, School of Medicine and Surgery, University of Milano-Bicocca, via Cadore, 48, 20900 Monza, MB, Italy. E-mail address: gessica.sala@unimib.it (G. Sala). http://dx.doi.org/10.1016/j.neuro.2016.04.018 0161-813X/ ã 2016 Elsevier Inc. All rights reserved. HSPA8/hsc70 (70-kDa heat shock cognate) protein represents a constitutively expressed protein belonging to the heat shock protein 70 (hsp70) chaperone family (Liu et al., 2012). Hsc70 is mainly localized in the intracellular space, possesses a highly 162 G. Sala et al. / NeuroToxicology 54 (2016) 161–169 conserved amino acid sequence and plays a critical role in a variety of cellular mechanisms including endocytosis, protein folding and degradation (Stricher et al., 2013). Recent studies reported that hsc70 is able to confer to the cells resistance against metabolic stress, hyperthermia and oxidative challenges (Chong et al., 2013; Wang et al., 2013a). Furthermore, particularly relevant is the involvement of hsc70 protein in the autophagic pathway known as chaperone-mediated autophagy (CMA), a selective device for the degradation of aberrant proteins containing the consensus peptide sequence KFERQ, which are directly transported to the lysosomes by a translocation system constituted by specific carrier proteins including cytosolic hsc70. Hsc70 is also localized into the lysosomal lumen where it allows the translocation of the substrate protein across the lysosomal membrane (Cuervo and Wong, 2014). Dysfunction of the CMA pathway is known to be closely associated with Parkinson’s disease (PD) (Alvarez-Erviti et al., 2010; Cuervo et al., 2004; Kabuta et al., 2008; Xilouri et al., 2009); in particular, a significant reduction of hsc70 levels was evidenced in the substantia nigra pars compacta and amygdala of PD brains (Alvarez-Erviti et al., 2010) and in lymphomonocytes obtained from sporadic PD patients (Sala et al., 2014). A significant downregulation of HSPA8/hsc70 was also observed in Alzheimer’s disease post-mortem brain tissues (Silva et al., 2014), suggesting that loss of expression of this molecular chaperone should play a critical role in the neuronal death associated not only with PD but also with other neurodegenerative diseases. Since a crucial pathogenic role in PD is recognized to be played by intraneuronal accumulation and aggregation of alpha-synuclein, the demonstration that CMA represents the main catabolic system for alpha-synuclein (Cuervo et al., 2004; Mak et al., 2010) has strengthened the link between CMA dysfunction and PD pathology. Further reinforcing the connection between hsc70 and PD, hsc70 has been demonstrated in vitro to bind to both soluble alpha-synuclein, slowing down its assembly into fibrils, and fibrillar form even with higher affinity (Pemberton et al., 2011; Pemberton and Melki, 2012), thus limiting the prion-like alphasynuclein spreading known to amplify PD-associated neurodegeneration. Lastly, a very recent study demonstrated that HSPA8/hsc70 represents the main constituent of a disaggregase system that efficiently fragments alpha-synuclein fibrils in vitro into shorter fibrils and promotes their depolymerization into nontoxic alpha-synuclein monomers (Gao et al., 2015). These findings identify other protective mechanisms exerted by hsc70 against the cytotoxicity associated with alpha-synuclein aggregation and inter-neuronal propagation occurring in PD. It is well known that exposure to rotenone, an inhibitor of the mitochondrial complex I, is able to reproduce PD pathology both in animal and cellular models, as indicated by the degeneration of nigrostriatal dopaminergic neurons and the formation in nigral neurons of alpha-synuclein-positive cytoplasmic inclusions (Betarbet et al., 2000; Gao et al., 2002; Sherer et al., 2003), although with some important limitations (Höglinger et al., 2006). Considering that rotenone induces alpha-synuclein aggregation and that hsc70 has a disaggregant effect on alpha-synuclein, this study was designed in order to verify whether rotenone exposure leads to hsc70 alteration possibly contributing to alpha-synuclein aggregation. To this aim, human SH-SY5Y neuroblastoma cells were treated with rotenone and hsc70 mRNA and protein expression was assessed. The protein levels of other heat shock proteins, hsp70 and hsp90, were evaluated to examine the specificity of rotenone-induced hsc70 reduction; the effect of rotenone on hsc70 was compared with that exerted by hydrogen peroxide, a generic oxidative stress donor with no inhibitory activity on mitochondrial complex I. Furthermore, the effect of rotenone on hsc70 was confirmed in primary mouse cortical neurons. As we observed a rotenone-induced autophagosome accumulation, the possible contribution of macroautophagy to rotenone-induced modulation of hsc70 was explored. The influence of hsc70 reduction on neurotoxicity was verified through HSPA8/hsc70 gene silencing. 2. Material and methods 2.1. Cell cultures Human neuroblastoma SH-SY5Y cells were grown in Dulbecco’s modified Eagle’s medium-F12 (EuroClone) supplemented with 10% fetal bovine serum (EuroClone), 100 U/mL penicillin (EuroClone), 100 mg/ml streptomycin (EuroClone) and 2 mM L-glutamine (EuroClone), at 37
C in an atmosphere of 5% CO2 in air. SH-SY5Y cells were used at a number of passages of growth ranging from thirteen to seventeen. Cortical neurons were prepared as previously described (Cirillo et al., 2014). Animal experiments were carried out using protocols approved by the University of Milano-Bicocca Animal Care and Use Committee and by the Italian Ministry of Health (protocol number 14-2011). This study complies with the ARRIVE guidelines. Briefly, cortices were dissected from neonatal (P1-P2) C57BL/6J mice (Charles River Laboratories), washed in dissociation medium and digested by trypsin (0.15%) with deoxyribonuclease (DNAse, 1 mg/ml, Sigma-Aldrich) at 37
C for 20 min. After mechanical dissociation, cells (1 106/ml) were plated onto poly-D-lysine (1 mg/ml) coated dishes in Neurobasal medium (NB; Invitrogen) containing B27 (Invitrogen), bFGF 10 ng/ml (Invitrogen), glutamine 1 mM (Sigma-Aldrich) and antibiotics (Sigma-Aldrich). Cultures were maintained at 37
C in 5% CO2 and used after 8 days in vitro (DIV). To evaluate purity of cultures (99–99.5%), cells were plated onto 12 mm poly-D-lysine coated coverslip (5000/well) and assessed by immunocytochemistry using anti-bIII-tubulin (Cell Signaling), as previously described (Cirillo et al., 2014). Neurons were imaged under a reversed microscope Olympus CX40 (X20) equipped with an Olympus camera. 2.2. Cytotoxicity assays The effect of rotenone or hydrogen peroxide on cell viability was assessed by the MTT assay based on reduction of the yellow tetrazolium salts (MTT) to the purple formazan by mitochondrial dehydrogenases. After exposure to rotenone (from 100 to 800 nM) or hydrogen peroxide (from 50 to 200 mM) for 6 or 24 h, SH-SY5Y cells were incubated with 0.5 mg/ml MTT (Sigma-Aldrich) in standard medium for 45 min at 37
C in an atmosphere of 5% CO2 in air. The effect on cell viability of 5 mM 3-methyladenine (3-MA) for 24 h, alone or in combination with 200–400 nM rotenone, was also assessed in SH-SY5Y cells. Similarly, cortical neurons (5000 cells/well) at DIV8 were treated with rotenone (from 100 to 800 nM) for 6 or 24 h and incubated with MTT (0.5 mg/ml) for 4 h. After cell solubilization with DMSO, absorbance was quantified (wavelength 570 nm) using a multimode microplate reader (FLUOstar Omega, BMG LABTECH) and cell viability expressed as% vs. vehicle-treated cells. Since rotenone affects mitochondrial function, rotenone-induced cell death was also evaluated with the Trypan blue exclusion test in order to independently confirm results obtained at MTT assay. 2.3. Whole-cell reactive oxygen species (ROS) levels The dye 20 ,70 -dichlorofluorescein diacetate (DCF-DA, Sigma-Aldrich) was used to quantify the levels of whole-cell ROS. After medium removal, cells were exposed to 10 mM DCF-DA G. Sala et al. / NeuroToxicology 54 (2016) 161–169 163 the following conditions: 50
C for 2 min, 95
C for 10 min, 40 cycles of: 95
C for 15 s, 60
C for 30 s. The following primer pairs were used: human hsc70-F (CAGGTTTATGAAGGCGAGCGTGCC) and human hsc70-R (GGGTGCAGGAGGTATGCCTGTGA); human alpha-synuclein-F (GCAGCCACTGGCTTTGTCAA) and human alphasynuclein-R (AGGATCCACAGGCATATCTTCCA); human beta-actin-F (TGTGGCATCCACGAAACTAC) and human beta-actin-R (GGAGCAATGATCTTGATCTTCA); mouse hsc70-F (CCTCGGAAAGACCGTTACCA) and mouse hsc70-R (TTTGTTGCCTGTCGCTGAGA); mouse beta-actin-F (GTCGAGTCGCGTCCACC) and mouse beta-actin-R (GTCATCCATGGCGAACTGGT). For relative quantification of each target vs. beta-actin mRNA, the comparative CT method was used as previously described (Sala et al., 2010). 2.6. Western blotting Cell pellets were lysed in cell extraction buffer (Invitrogen) supplemented with 1 mM PMSF and protease inhibitor cocktail (Sigma-Aldrich) and protein concentrations determined by Fig. 1. (A) Cytotoxicity studies. Cell viability was assessed by MTT assay after 24 h exposure to 100–800 nM rotenone and 50–200 mM hydrogen peroxide (H2O2). Values are expressed as% vs. vehicle. N = 6, repeated measures ANOVA test, followed by Dunnett’s post-test; * p < 0.05, ** p < 0.01 vs. vehicle. (B) Whole-cell intracellular ROS production. The effect on ROS production of 24 h exposure to 200–400 nM rotenone or 50–100 mM hydrogen peroxide (H2O2) was shown and expressed as% vs. vehicle of dichlorofluorescein fluorescence units (DCF FU) normalized to protein content. N = 4, repeated measures ANOVA test, followed by Dunnett’s post-test; * p < 0.05, ** p < 0.01 vs. vehicle. in Locke’s buffer (154 mM NaCl, 5.6 mM KCl, 3.6 mM NaHCO3, 2.3 mM CaCl2, 5.6 mM glucose, 5 mM Hepes, 1.2 mM MgCl2, pH 7.4) for 45 min at 37
C in an atmosphere of 5% CO2 in air. Cells were washed in Locke’s buffer without glucose, harvested and lysed. Fluorescence units (FU) were quantified (excitation 488 nm, emission 525 nm) and related to the total protein content assessed using the method of Bradford. 2.4. RNA extraction and cDNA synthesis Total RNA was extracted using the RNeasy Mini kit (Qiagen), according to the manufacturer instructions. RNA concentration was determined spectrophotometrically at 260 nm. RNA (2 mg) was retro-transcribed into cDNA using the SuperScript1 VILOTM cDNA Synthesis Kit (Invitrogen) at the following conditions: 10 min at 25
C and 60 min at 42
C. The reaction was terminated at 85
C for 5 min and cDNAs stored at 20
C. 2.5. Real-time quantitative PCR (qPCR) cDNAs obtained from RNA (50 ng for hsc70 and 100 ng for alpha-synuclein) were amplified in triplicate in the ABI Prism 7500HTSequence Detection System (Applied Biosystems) using the Platinum1 SYBR1 Green qPCR SuperMix-UDG (Invitrogen) at Fig. 2. Effect of rotenone treatment on hsc70 mRNA and protein levels in human SH-SY5Y cells. (A) Relative quantification (RQ) of hsc70 mRNA levels after treatment with rotenone (200–400 nM) for 6 or 24 h. Hsc70 mRNA levels were calculated as ratio to beta-actin and expressed as fold change vs. vehicle (RQ = 1). (B) Hsc70 protein levels normalized to beta-actin and expressed as% vs. vehicle. (C) Representative Western blot image showing the effect of 6 or 24 h rotenone exposure on hsc70 protein expression. Immunoreactivity of beta-actin, used as internal standard, was also shown. N = 10, repeated measures ANOVA test, followed by Dunnett’s post-test; * p < 0.05, ** p < 0.01 vs. vehicle. 164 G. Sala et al. / NeuroToxicology 54 (2016) 161–169 multiple comparison test, was used to assess the significance of differences between groups. 3. Results 3.1. Effect of rotenone and hydrogen peroxide on hsc70 mRNA and protein expression in human SH-SY5Y cells Preliminary experiments were carried out to establish rotenone and hydrogen peroxide concentrations able to evoke similar cytotoxic and pro-oxidant effects in SH-SY5Y cells. While rotenone is a specific mitochondrial complex I inhibitor that secondarily results in cell damage by promoting generation of peroxides, hydrogen peroxide is a well-known oxidative stress donor. Exposure for 24 h to rotenone concentrations ranging from 100 to 800 nM or hydrogen peroxide (from 50 to 200 mM) causes a dose-dependent cell death, as shown in Fig. 1A. Whole-cell intracellular ROS production was also quantified as index of rotenone- and hydrogen peroxide-induced oxidative stress. Exposure for 24 h to 200 or 400 nM rotenone caused a significant increase (65 and 90%, respectively; p < 0.01) of ROS production with respect to vehicle-treated cells (Fig. 1B). Similarly, 24 h Fig. 3. Effect of rotenone treatment on hsp70 and hsp90 protein levels in human SH-SY5Y cells. Cells were treated with 200–400 nM rotenone for 24 h. (A) Hsp70 and hsp90 protein levels normalized to beta-actin and expressed as% vs. vehicle. (B) Representative Western blot image showing the effect of 24 h rotenone exposure on hsp70 and hsp90 protein expression. Immunoreactivity of beta-actin, used as internal standard, was also shown. N = 4, repeated measures ANOVA test, followed by Dunnett’s post-test; ** p < 0.01 vs. vehicle. Bradford’s method. After denaturation, 10 and 30 mg (for mouse and human hsc70, respectively), 30 mg (for hsp70, hsp90 and beclin-1) or 50 mg (for LC3II and alpha-synuclein) proteins were separated by electrophoresis in 8% or 4–12% SDS-PAGE and transferred to nitrocellulose. Blots were blocked for 1 h, incubated overnight at 4
C with specific primary antibodies (hsc70 Abcam, 1:3,000 dilution, hsp70 Enzo Life Sciences, 1:3,000 dilution, hsp90Cell Signaling, 1:1,000 dilution, beclin-1Cell Signaling, 1:1,000 dilution; LC3B Cell Signaling, 1:500 dilution; alphasynuclein BD Biosciences, 1:1,000 dilution) and then with the HRP-linked anti-mouse or – rabbit IgG for 1 h. Beta-actin (Sigma, 1:30,000 dilution) was used as internal standard. Signals were revealed by chemiluminescence, visualized on X-ray film and quantified by GS-710 Imaging Densitometer (Bio-Rad). 2.7. siRNA transfection Gene silencing by small interfering RNA (siRNA) was carried out by transfecting cells with a siRNA-lipid complex composed by 10 nM Silencer1 Select siRNA (Ambion1 by Life Technologies) targeted to HSPA8/hsc70 gene and Lipofectamine1 RNAiMAX Transfection Reagent (Invitrogen by Life Technologies) in 1:1 ratio. 10 nM Silencer1 Select Negative Control siRNA (Ambion1 by Life Technologies) was used as non-targeting negative control. After 48 h, transfection reagents were washed out and cells were pelleted or stimulated for 24 h with 400 nM rotenone or 100 mM hydrogen peroxide. 2.8. Statistical analysis All data are shown as mean  standard deviation (SD). Statistical analysis was performed using GraphPad Prism 4.0. Repeated measures ANOVA, followed by Dunnett’s or Tukey’s Fig. 4. Effect of 24 h hydrogen peroxide (H2O2) treatment on hsc70 mRNA and protein levels in human SH-SY5Y cells. (A) Relative quantification (RQ) of hsc70 mRNA levels calculated as ratio to beta-actin and expressed as – fold change vs. vehicle (RQ = 1). (B) Hsc70 protein levels, expressed as% vs. vehicle of the ratio between hsc70 and beta-actin optical density. (C) Representative Western blot image showing the effect of 24 h H2O2 exposure on hsc70 protein expression. The immunoreactivity of beta-actin, used as internal standard, was also shown. N = 10, repeated measures ANOVA test. G. Sala et al. / NeuroToxicology 54 (2016) 161–169 treatment with hydrogen peroxide (50 or 100 mM) results in a 30% (p < 0.05) and 80% (p < 0.01) ROS increase (Fig. 1B). Based on results derived from cytotoxicity studies, we decided to evaluate hsc70 gene and protein expression levels after treatment with rotenone or hydrogen peroxide at concentrations responsible for a mild to moderate cytotoxic effects after 24 h exposure. Therefore, SH-SY5Y cells were exposed to 200 and 400 nM rotenone or 50 and 100 mM hydrogen peroxide for 24 h. The same concentrations of toxins was also administered to the cells for 6 h when no cytotoxic effect of rotenone was evidenced and only a 20% cell death (p < 0.05) was observed after exposure to 100 mM hydrogen peroxide. After treatment with rotenone for 6 or 24 h, levels of mRNA encoding for hsc70 were quantified by realtime PCR. Relative quantification (RQ) of hsc70 normalized to betaactin in SH-SY5Y is represented in Fig. 2A. Rotenone induced a significant dose- and time-dependent decrease of mRNA encoding for hsc70, as compared to vehicle-treated cells. Interestingly, instead, western blot analyses showed that 6 h treatment with rotenone (200 or 400 nM) resulted in a progressive increase of hsc70 protein levels, while prolonged incubation of SH-SY5Y cells with the same rotenone concentrations for 24 h caused a dosedependent reduction of hsc70 protein expression (Fig. 2B and C). To 165 examine the specificity of rotenone-induced hsc70 reduction, the expression of other two molecular chaperones, hsp70 and hsp90, constitutively expressed under normal conditions to maintain protein homeostasis and induced upon environmental stress, was assessed. The exposure to 200 and 400 nM rotenone for 24 h resulted in a 40 and 60% reduction (p < 0.01) of hsp70 protein levels, while no significant change was observed for hsp90 (Fig. 3A and B). Hsc70 expression was also assessed after exposure to hydrogen peroxide, used as paradigm of oxidative stress. No significant change in hsc70 mRNA and protein levels was observed after 24 h exposure to 50 or 100 mM hydrogen peroxide (Fig. 4). No change was evidenced in hsc70 gene and protein levels after 6 h treatment with hydrogen peroxide (data not shown). 3.2. Effect of rotenone on hsc70 mRNA and protein expression in mouse cortical neurons To validate the neurotoxic effect of rotenone on hsc70 expression, we used mouse cortical neurons. As shown in Fig. 5A, exposure of cortical neurons to increasing concentrations of rotenone (100–800 nM) for 24 h caused a dose-dependent cell death, while progressively increasing their loss of neuronal Fig. 5. Survival of mouse cortical neurons exposed to rotenone for 24 h. A. Cell viability was assessed by MTT assay after 24 h exposure to 100–800 nM rotenone. Values, expressed as% vs. vehicle, are the mean  SD of two separate experiments, each including five separate samples for each condition. * p < 0.05, ** p < 0.01 vs. vehicle; ANOVA test, followed by Dunnett’s multiple comparisons test. B. Representative images of cortical neurons exposed for 24 h to rotenone at the indicated concentrations. Neurons were imaged under a reversed microscope Olympus CX40 (X20) equipped with an Olympus camera. 166 G. Sala et al. / NeuroToxicology 54 (2016) 161–169 To explore a putative role for macroautophagy in rotenoneinduced modulation of hsc70, hsc70 mRNA and protein levels were assessed in presence of a specific macroautophagy inhibitor, 3-methyladenine (3-MA, 5 mM, 1 h before rotenone treatment). We observed that 3-MA, while was able to prevent or partially counteract cell death induced by 200 or 400 nM rotenone (Fig. 8A), did not affect hsc70 mRNA and protein levels and did not modify the reduction of hsc70 mRNA (Fig. 8B) and protein levels caused by rotenone exposure (Fig. 8C and D). Furthermore, the effect of macroautophagy inhibition was also verified on alpha-synuclein expression during rotenone toxicity. Treatment with 3-MA resulted in a significant reduction (-37%, p < 0.05) of alphasynuclein mRNA levels (Fig. 8B) and increase of its protein levels (+41%, p < 0.05) (Fig. 8C and D). As previously reported (Sala et al., 2013), exposure to rotenone caused an increase of alpha-synuclein mRNA and protein levels and macroautophagy inhibition resulted in a further increase of alpha-synuclein mRNA (+42% vs. rotenonetreated cells, p < 0.05) and protein (+49% vs. rotenone-treated cells, p < 0.05) levels (Fig. 8). 3.4. Effect of HSPA8/hsc70 silencing on neurotoxicity To test whether HSPA8/hsc70 gene silencing influenced neurotoxicity, SH-SY5Y cells were subjected to 48 h transfection with a specific siRNA recognizing hsc70. Hsc70 silencing was confirmed through the assessment of hsc70 mRNA and protein expression by qPCR and Western blot, respectively. Hsc70 siRNA resulted in about 60% reduction of mRNA (Fig. 9A) and 50% decrease of protein levels (Fig. 9B and C) with respect to negative control. Hsc70 silencing resulted in a significant 2.5-fold increase of alpha-synuclein mRNA levels (p < 0.01, Fig. 9A) paralleled by an increase of its protein expression (+80%, p < 0.01, Fig. 9B and C). Furthermore, we demonstrated that hsc70 silencing did not affect cell viability and did not alter the cytotoxic effect exerted by 24 h Fig. 6. Effect of rotenone treatment on hsc70 mRNA and protein levels in mouse cortical neurons. A. Relative quantification (RQ) of hsc70 mRNA levels after treatment with rotenone (200–400 nM) for 6 or 24 h. Hsc70 mRNA content was calculated as ratio to beta-actin and expressed as – fold change vs. vehicle (RQ = 1). B. Hsc70 protein levels normalized to beta-actin and expressed as% vs. vehicle. C. Representative Western blot image showing the effect of 6 or 24 h rotenone exposure on hsc70 protein expression. Immunoreactivity of beta-actin, used as internal standard, was also shown. N = 4, repeated measures ANOVA test, followed by Dunnett’s post-test; * p < 0.05, ** p < 0.01 vs. vehicle. branched morphology (Fig. 5B). No significant cytotoxic effect was evidenced on cortical neurons after 6 h exposure to all tested rotenone concentrations (data not shown). Hsc70 mRNA levels in primary neurons challenged with 200–400 nM rotenone for 6 or 24 h showed a trend similar to that observed in neuroblastoma cells, as indicated by the significant timedependent reduction of hsc70 mRNA levels (Fig. 6A). Western blot analyses showed that 6 h treatment with rotenone did not affect hsc70 protein levels, while an increased hsc70 protein level was observed at 6 h in SH-SY5Y cells. Instead, similarly to SH-SY5Y cells, exposure to rotenone (200–400 nM) for 24 h caused a significant reduction of hsc70 protein expression (Fig. 6B and C). 3.3. Effect of macroautophagy inhibition on rotenone-induced hsc70 and alpha-synuclein modulation in SH-SY5Y cells Next, we evaluated whether rotenone induces macroautophagy in SH-SY5Y cells. To this end, expression of LC3II and beclin-1, two proteins typically used to monitor macroautophagy, were assessed in SH-SY5Y cells exposed to 100–400 nM rotenone for 24 h. Western blot analysis revealed that rotenone induced a significant increase of LC3II, while no change in beclin-1 protein expression was observed in rotenone-treated cells (Fig. 7A and B). Fig. 7. Effect of rotenone on macroautophagy. A. LC3II and beclin-1 protein levels after treatment with rotenone (100–400 nM) for 24 h. Hsc70 mRNA content was, expressed as% vs. vehicle of the ratio between target protein and beta-actin optical density. B. Representative Western blot image showing the effect of rotenone treatment for 24 h on LC3II and beclin-1 protein expression. Immunoreactivity of beta-actin, used as internal standard, was also shown. N = 6, repeated measures ANOVA test, followed by Dunnett’s post-test; * p < 0.05, ** p < 0.01 vs. vehicle. G. Sala et al. / NeuroToxicology 54 (2016) 161–169 167 Fig. 8. Effect of macroautophagy inhibition on rotenone-induced cell death and modulation of hsc70 and alpha-synuclein expression. 3-methyladenine (3-MA, 5 mM) was used to inhibit macroautophagy and administered 1 h before rotenone treatment (200 nM, 24 h). A. Cell viability was assessed by MTT assay. Values are expressed as% vs. vehicle. N = 4, repeated measures ANOVA test, followed by Tukey’s post-test; * p < 0.05, ** p < 0.01 vs. vehicle, x p < 0.01 vs. rotenone-treated cells. B. Relative quantification (RQ) of hsc70 and alpha-synuclein (asyn) mRNA levels calculated as ratio to beta-actin and expressed as fold change vs. vehicle (RQ = 1). C. Hsc70 and asyn protein levels, expressed as% vs. vehicle of the ratio between target proteins and beta-actin optical density. D. Representative Western blot image showing the effect of rotenone, 3-MA or rot/ 3-MA co-treatment on hsc70 and asyn protein expression. Immunoreactivity of beta-actin, used as internal standard, was also shown. N = 3, repeated measures ANOVA test, followed by Tukey’s post-test; * p < 0.05, ** p < 0.01 vs. vehicle; x p < 0.05 vs. rotenone-treated cells. exposure to 400 nM rotenone or 100 mM hydrogen peroxide (Fig. 9D). 4. Discussion Based on knowledge that hsc70 is a chaperone protein endowed with a critical role in several intracellular mechanisms known to be closely associated with PD, the aim of this study was to establish the existence of a possible modulation of hsc70 following exposure to PD-related stimuli. No previous study specifically explored the effect of rotenone on the constitutively expressed HSPA8/ hsc70 chaperone protein. Therefore, we assessed hsc70 mRNA and protein expression after mitochondrial inhibition by rotenone (a well-known inhibitor of the mitochondrial complex I) and oxidative stress, two major pathogenic mechanisms contributing to the loss of dopaminergic neurons in PD (Jenner and Olanow, 1998; Schapira et al., 1998). To this purpose, we used human neuroblastoma SH-SY5Y cells, a neuron-like cell line expressing dopaminergic markers and widely used to study PD pathomechanisms, and primary mouse cortical cultures, representing a neuronal population affected in late-stage PD. Maintaining proper levels of hsc70 appears fundamental for at least three processes whose dysfunction is known to lead to PD; in fact, hsc70 i) is required for the physiological activity of the CMA pathway, being hsc70 the principal carrier protein of CMA also essential to allow the entry of substrates into the lysosomal lumen (Cuervo and Wong, 2014); ii) it also cooperates to confer resistance against different metabolic cellular stresses (Chong et al., 2013; Wang et al., 2013a); iii) it prevents alpha-synuclein deposition and propagation through the binding to both soluble and fibrillar forms (Pemberton et al., 2011; Pemberton and Melki, 2012) and efficiently fragments alphasynuclein fibrils in vitro into shorter fibrils and promotes depolymerization into non-toxic alpha-synuclein monomers (Gao et al., 2015). Based on these considerations, we exposed SH-SY5Y cells to different concentrations of rotenone for 6 and 24 h and we measured the effects of this treatment on hsc70 expression. We found that rotenone determined a dose- and time-dependent reduction of hsc70 mRNA levels, together with an initial (6 h) increase of hsc70 protein levels followed by a dose-dependent reduction after 24 h rotenone exposure (Fig. 2). These findings suggest that, after 24 h rotenone exposure, the reduction of hsc70 protein expression is ascribable to the down-regulation of its synthesis. Instead, the rotenone-induced increase of hsc70 protein levels observed at 6 h can be interpreted as a consequence of an involvement of other mechanisms possibly affected by rotenone, such as a compromised protein degradation, that overcome and hide, at least for a limited time lapse, the reduced hsc70synthesis. The observed reduction of hsp70 protein levels after rotenone exposure indicates that the influence of rotenone is not specific for hsc70, but rather involves other heat shock proteins as already described in several brain areas of rotenone-treated rats (Sonia Angeline et al., 2012; Thakur and Nehru, 2014). Nevertheless, our findings identify for the first time hsc70 as adjunctive target of rotenone toxicity possibly contributing to alpha-synuclein accumulation. The rotenone-induced reduction of hsc70 mRNA levels observed in SH-SY5Y cells was also confirmed in primary cortical neurons, which showed a similar trend in hsc70 mRNA levels when challenged with rotenone (Fig. 6). While after 24 h rotenone 168 G. Sala et al. / NeuroToxicology 54 (2016) 161–169 Fig. 9. Effect of HSPA8/hsc70 silencing on neurotoxicity. Analysis of 48 h hsc70 silencing on alpha-synuclein (asyn) gene (A) and protein levels (B and C) and on cytotoxicity (D) in SH-SY5Y cells. A. Hsc70 and asyn mRNA levels were normalized on beta-actin levels and expressed as – fold change vs. negative control (NC, mRNA level = 1). B. Hsc70 and asyn protein levels, expressed as% vs. negative control (NC) of the ratio between target proteins and beta-actin optical density. C. Representative Western blot image showing hsc70 and asyn protein expression in cells transfected with siRNA-hsc70 or NC. Immunoreactivity of beta-actin, used as internal standard, was also shown. D. Cell viability was assessed by MTT assay in cells transfected with siRNA-hsc70 or NC for 48 h and then treated with 400 nM rotenone or 100 mM hydrogen peroxide (H2O2) for 24 h. Values are expressed as% vs. vehicle. N = 4, repeated measures ANOVA test, followed by Dunnett’s post-test; * p < 0.05 vs. corresponding vehicle (NC or siRNA-hsc70); ** p < 0.01 vs. NC. exposure a significant reduction of hsc70 protein levels was observed also in primary cortical neurons, differently from what found in SH-SY5Y cells, no alteration in hsc70 protein expression emerged after 6 h treatment (Fig. 6). This last result seems to indicate that in cortical neurons hsc70 protein expression after 6 h rotenone treatment does not reflect the reduced synthesis, as well as already reported for neuroblastoma cells which even showed increased hsc70 protein levels at this time point. We then compared the effect of the mitochondrial inhibitor rotenone, which also causes oxidative stress, versus the oxidative stress donor hydrogen peroxide. It was noteworthy to observe that concentrations of rotenone and hydrogen peroxide that caused similar cytotoxicity and intracellular ROS raise in SH-SY5Y cells differently affected hsc70 expression. In fact, while rotenone significantly reduced hsc70 gene expression (Fig. 2A), hydrogen peroxide did not affect hsc70 mRNA and protein levels (Fig. 4). These data suggest that mechanisms used by rotenone to reduce hsc70 expression may involve mitochondrial inhibition, but they are not necessarily directly linked to increased oxidative stress. Rather, it is conceivable that the effect of rotenone on hsc70 expression might be directly or indirectly linked to decreased mitochondrial bioenergetics, a primary outcome of mitochondrial complex I inhibition, leading to decreased ATP levels and induction of macroautophagy. Indeed, we found that treatment of SH-SY5Y cells with rotenone for 24 h caused an increase of LC3-II levels, while leaving unchanged beclin-1 expression (Fig. 7), in line with the rotenone-induced autophagosome accumulation already observed in these cells (Mader et al., 2012). Although we did not perform a complete study with specific pharmacological inhibitors of the autophagic steps, these data are indicative of a block of the autophagic flux caused by rotenone, as also previously reported in both in vitro and in vivo studies (Giordano et al., 2014; Mader et al., 2012; Wu et al., 2015). Using a specific inhibitor, we demonstrated that macroautophagy inhibition, although was able to partially prevent rotenone-induced cell death (Fig. 8A), did not influence hsc70 expression, as evidenced by the similar rotenone-induced reduction in mRNA and protein hsc70 levels observed in the presence or absence of 3-MA (Fig. 8B-D). These findings support the view that the effect of rotenone on hsc70 gene and protein expression is not ascribable to the rotenone cytotoxicity and that the protective effect of 3-MA on rotenone-induced cell death might be due to an induction of CMA, which represents a cellular protective mechanism against necrosis and massive cell death. Furthermore, we demonstrated that macroautophagy inhibition by 3-MA did not modify per se hsc70 mRNA and protein levels under basal conditions. Based on knowledge that macroautophagy participates, together with CMA, to alphasynuclein degradation (Webb et al., 2003), through macroautophagy inhibition, we also explored the specific contribution of CMA in the accumulation of alpha-synuclein caused by rotenone, as previously demonstrated in both in vivo and in vitro models (Betarbet et al., 2006; Betarbet et al., 2000; Sala et al., 2013). Interestingly, we observed that macroautophagy inhibition by 3-MA was able to cause a significant increase of alphasynuclein protein levels (Fig. 8B-C), according to the knowledge that this protein is physiologically partially degraded by this pathway. Accumulation of alpha-synuclein protein (Fig. 8B) could represent an inhibitory feedback responsible for the decrease of alpha-synuclein synthesis, as indicated by the reduction of mRNA levels in presence of 3-MA (Fig. 8A). The increase of alpha-synuclein mRNA and protein levels induced by rotenone was further enhanced by macroautophagy inhibition, which likely renders cells more susceptible to rotenone toxicity, thus confirming the need of a proper activity of both CMA and macroautophagy for maintaining the cellular homeostasis. G. Sala et al. / NeuroToxicology 54 (2016) 161–169 Finally, hsc70 gene silencing allowed us to demonstrate that hsc70 reduction is per se sufficient to cause an up-regulation of alpha-synuclein gene and protein expression that is likely to favor protein accumulation and aggregation typical of PD and other synucleinopathies. 5. Conclusions The main finding of this study is the demonstration that the constitutively expressed HSPA8/hsc70 chaperone protein represents a new intracellular target of rotenone toxicity. Our data suggest that mechanisms used by rotenone to reduce hsc70 expression may involve mitochondrial inhibition, although a role for increased oxidative stress cannot be excluded. This finding implies the existence of an adjunctive toxic mechanism of rotenone possibly contributing to PD pathogenesis and worthy of study in the attempt of identifying new therapeutic targets for the disease. Further studies exploring the molecular mechanisms responsible for rotenone-induced hsc70 reduction are needed, including the possible involvement of specific microRNA deregulation already reported to be associated with hsc70 reduction in PD (Alvarez-Erviti et al., 2013; Wang et al., 2013b), as well as a better understanding of specific mechanisms involved in hsc70-mediated modulation of alpha-synuclein. Conflict of interest This manuscript is not under consideration for publication elsewhere and is free of any conflict of interest or financial implications. Acknowledgements This work was carried out within the framework of the Ivascomar project, Cluster Tecnologico Nazionale Scienze della Vita ALISEI, Italian Ministry of Research (MIUR, to CF and AMC). This work was also supported by MIUR grants (PRIN2007 to AMC; SYSBIONET-Italian ROADMAP ESFRI Infrastructures to LA and AMC) and Associazione Levi-Montalcini (fellowships to MF). References Alvarez-Erviti, L., Rodriguez-Oroz, M.C., Cooper, J.M., Caballero, C., Ferrer, I., Obeso, J. A., Schapira, A.H., 2010. Chaperone-mediated autophagy markers in Parkinson disease brains. Arch. Neurol. 67, 1464–1472. Alvarez-Erviti, L., Seow, Y., Schapira, A.H., Rodriguez-Oroz, M.C., Obeso, J.A., Cooper, J.M., 2013. Influence of microRNA deregulation on chaperone-mediated autophagy and a-synuclein pathology in Parkinson’s disease. Cell Death Dis. 4, e545. Betarbet, R., Sherer, T.B., MacKenzie, G., Garcia-Osuna, M., Panov, A.V., Greenamyre, J.T., 2000. Chronic systemic pesticide exposure reproduces features of Parkinson’s disease. Nat. Neurosci. 3, 1301–1306. Betarbet, R., Canet-Aviles, R.M., Sherer, T.B., Mastroberardino, P.G., McLendon, C., Kim, J.H., Lund, S., Na, H.M., Taylor, G., Bence, N.F., Kopito, R., Seo, B.B., Yagi, T., Yagi, A., Klinefelter, G., Cookson, M.R., Greenamyre, J.T., 2006. Intersecting pathways to neurodegeneration in Parkinson’s disease: effects of the pesticide rotenone on DJ-1, alpha-synuclein, and the ubiquitin-proteasome system. Neurobiol. Dis. 22, 404–420. Chong, K.Y., Lai, C.C., Su, C.Y., 2013. Inducible and constitutive HSP70s confer synergistic resistance against metabolic challenges. Biochem. Biophys. Res. Commun. 430, 774–779. Cirillo, G., Colangelo, A.M., Berbenni, M., Ippolito, V.M., De Luca, C., Verdesca, F., Savarese, L., Alberghina, L., Maggio, N., Papa, M., 2014. Purinergic modulation of spinal neuroglial maladaptive plasticity following peripheral nerve injury. Mol. Neurobiol. 52 (3), 1440–1457. Cuervo, A.M., Wong, E., 2014. Chaperone-mediated autophagy: roles in disease and aging. Cell Res. 24, 92–104. 169 Cuervo, A.M., Stefanis, R., Lansbury, P.T., Sulzer, D., 2004. Impaired degradation of mutant alpha-synuclein by chaperone-mediated autophagy. Science 305, 1292– 1295. Gao, H.M., Hong, J.S., Zhang, W., Liu, B., 2002. Distinct role for microglia in rotenoneinduced degeneration of dopaminergic neurons. J. Neurosci. 22, 782–790. Gao, X., Carroni, M., Nussbaum-Krammer, C., Mogk, A., Nillegoda, N.B., Szlachcic, A., Guilbride, D.L., Saibil, H.R., Mayer, M.P., Bukau, B., 2015. Human hsp70 disaggregase reverses Parkinson’s-linked a-synuclein amyloid fibrils. Mol. Cell 59, 781–793. Giordano, S., Dodson, M., Ravi, S., Redmann, M., Ouyang, X., Darley Usmar, V.M., Zhang, J., 2014. Bioenergetic adaptation in response to autophagy regulators during rotenone exposure. J. Neurochem. 131, 625–633. Höglinger, G.U., Oertel, W.H., Hirsch, E.C., 2006. The rotenone model of parkinsonism—the five years inspection. J. Neural Transm. 269–272. Jenner, P., Olanow, C.W., 1998. Understanding cell death in Parkinson’s disease. Ann. Neurol. 44, S72–S84. Kabuta, T., Furuta, A., Aoki, S., Furuta, K., Wada, K., 2008. Aberrant interaction between Parkinson disease-associated mutant UCH-L1 and the lysosomal receptor for chaperone-mediated autophagy. J. Biol. Chem. 283, 23731–23738. Liu, T., Daniels, C.K., Cao, S., 2012. Comprehensive review on the HSC70 functions, interactions with related molecules and involvement in clinical diseases and therapeutic potential. Pharmacol. Ther. 136, 354–374. Mader, B.J., Pivtoraiko, V.N., Flippo, H.M., Klocke, B.J., Roth, K.A., Mangieri, L.R., Shacka, J.J., 2012. Rotenone inhibits autophagic flux prior to inducing cell death. ACS Chem. Neurosci. 3, 1063–1072. Mak, S.K., McCormack, A.L., Manning-Bog, A.B., Cuervo, A.M., Di Monte, D.A., 2010. Lysosomal degradation of alpha-synuclein in vivo. J. Biol. Chem. 285, 13621– 13629. Pemberton, S., Melki, R., 2012. The interaction of Hsc70 protein with fibrillar a-synuclein and its therapeutic potential in Parkinson’s disease. Commun. Integr. Biol. 5, 94–95. Pemberton, S., Madiona, K., Pieri Kabani, L.M., Bousset, L., Melki, R., 2011. Hsc70 protein interaction with soluble and fibrillar alpha-synuclein. J. Biol. Chem. 286, 34690–34699. Sala, G., Brighina, L., Saracchi, E., Fermi, S., Riva, C., Carrozza, V., Pirovano, M., Ferrarese, C., 2010. Vesicular monoamine transporter 2 mRNA levels are reduced in platelets from patients with Parkinson’s disease. J. Neural Transm. 117, 1093–1098. Sala, G., Arosio, A., Stefanoni, G., Melchionda, L., Riva, C., Marinig, D., Brighina, L., Ferrarese, C., 2013. Rotenone upregulates alpha-synuclein and myocyte enhancer factor 2D independently from lysosomal degradation inhibition. Biomed. Res. Int. 846725. Sala, G., Stefanoni, G., Arosio, A., Riva, C., Melchionda, L., Saracchi, E., Fermi, S., Brighina, L., Ferrarese, C., 2014. Reduced expression of the chaperone-mediated autophagy carrier hsc70 protein in lymphomonocytes of patients with Parkinson’s disease. Brain Res. 1546, 46–52. Schapira, A.H., Gu, M., Taanman, J.W., Tabrizi, S.J., Seaton, T., Cleeter, M., Cooper, J.M., 1998. Mitochondria in the etiology and pathogenesis of Parkinson’s disease. Ann. Neurol. 44, S89–S98. Sherer, T.B., Kim, J.H., Betarbet, R., Greenamyre, J.T., 2003. Subcutaneous rotenone exposure causes highly selective dopaminergic degeneration and alphasynuclein aggregation. Exp. Neurol. 179, 9–16. Silva, P.N., Furuya, T.K., Braga, I.L., Rasmussen, L.T., Labio, R.W., Bertolucci, P.H., Chen, E.S., Turecki, G., Mechawar, N., Payão, S.L., Mill, J., Smith, M.C., 2014. Analysis of HSPA8 and HSPA9 mRNA expression and promoter methylation in the brain and blood of Alzheimer’s disease patients. J. Alzheimers Dis. 38, 165–170. Sonia Angeline, M., Chaterjee, P., Anand, K., Ambasta, R.K., Kumar, P., 2012. Rotenone-induced parkinsonism elicits behavioral impairments and differential expression of parkin, heat shock proteins and caspases in the rat. Neuroscience 220, 291–301. Stricher, F., Macri, C., Ruff, M., Muller, S., 2013. HSPA8/HSC70 chaperone protein: structure, function, and chemical targeting. Autophagy 9, 1937–1954. Thakur, P., Nehru, B., 2014. Long-term heat shock proteins (HSPs) induction by carbenoxolone improves hallmark features of Parkinson’s disease in a rotenonebased model. Neuropharmacology 79, 190–200. Wang, B.S., Yang, Y., Yang, H., Liu, Y.Z., Hao, J.J., Zhang, Y., Shi, Z.Z., Jia, X.M., Zhan, Q. M., Wang, M.R., 2013a. PKCl counteracts oxidative stress by regulating Hsc70 in an esophageal cancer cell line. Cell Stress Chaperones 18, 359–366. Wang, Q., Hu, W., Lei, M., Wang, Y., Yan, B., Liu, J., Zhang, R., Jin, Y., 2013b. MiR-17-5p impairs trafficking of H-ERG K+ channel protein by targeting multiple er stressrelated chaperones during chronic oxidative stress. PLoS One 8, e84984. Webb, J.L., Ravikumar, B., Atkins, J., Skepper, J.N., Rubinsztein, D.C., 2003. Alphasynuclein is degraded by both autophagy and the proteasome. J. Biol. Chem. 278, 25009–25013. Wu, F., Xu, H.D., Guan, J.J., Hou, Y.S., Gu, J.H., Zhen, X.C., Qin, Z.H., 2015. Rotenone impairs autophagic flux and lysosomal functions in Parkinson's disease. Neuroscience 284, 900–911. Xilouri, M., Vogiatzi, T., Vekrellis, K., Park, D., Stefanis, L., 2009. Abberant alphasynuclein confers toxicity to neurons in part through inhibition of chaperonemediated autophagy. PLoS One 4, e5515.