6-OHDA (#H116), apomorphine (#A4393), 3-methyladenine (3-MA, #M9281), chloroquine (CQ, #CC6628) and mouse anti-TH (#1299) monoclonal antibody were purchased from Sigma-Aldrich (St. Louis, MO, USA). [D-Lys3]-GHRP-6 (#1922/5) was purchased from R&D (Minneapolis, MN, USA). Antibodies against LC3B (#2775S) and cleaved caspase 3 (#9661) were obtained from Cell Signaling Technology (Beverly, MA, USA). The antibody against TFEB (A303-673A) was obtained from Bethyl Laboratories (Montgomery, TX, USA). Antibodies against p62 (#18420-1-AP), Bax (#50599-2-Ig) and Bcl-2 (#12789-1-AP) were obtained from Proteintech Group, Inc. (Chicago, IL, USA). Anti-GAPDH (#Ab103-01) antibody was purchased from Vazyme Biotech Co. (Nanjing, China). The HRP-conjugated anti-mouse (#ZB-2305) and anti-rabbit (#ZB-2301) antibodies were obtained from ZSGB-BIO (Beijing, China). DMEM/F12 culture medium, fetal bovine serum, penicillin/streptomycin solution and trypsin EDTA were obtained from Biological Industries (Beit Haemek, Israel).

Male Sprague-Dawley rats weighing 200–220 g were purchased from Beijing HFK Bioscience Cooperation, China. All animal care and procedures were approved by Shengjing Hospital Medical Ethics Committee (2019PS054K(X1)). They were housed and bred in accordance with NIH Guidelines for the care and use of laboratory animals. The rats were housed in a12 h light/dark cycle schedule and with ad libitum access to food and water. For 6-OHDA-induced rat model, the animals were randomly divided into the following groups: 1) control group: rats received stereotaxic injection of normal saline at right striatum as described before (He et al., 2018); 2) ghrelin solely-treated group: rats received ghrelin injection (i.c.v., 100 ng); 3) 6-OHDA solely-treated group: rats received stereotaxic injection of 6-OHDA at right striatum as described before (He et al., 2018); 4) ghrelin+6-OHDA treatment group: rats received different doses of ghrelin (100, 200 and 400 ng/rat, i.c.v.) 30 min before intrastriatal injection of 6-OHDA, and to observe the long-term effect of ghrelin on 6-OHDA-induced dopaminergic neuron loss, rats received different doses of ghrelin for consecutive 14 days; 5) ghrelin + 6-OHDA+ [D-Lys3]-GHRP-6 group: rats received 100 ng ghrelin (i.c.v.) and intrastriatal injection of 10 μg [D-Lys3]-GHRP-6 30 min before intrastriatal injection of 6-OHDA as described before (Wei et al., 2015); 6) ghrelin + 6-OHDA + 3-MA group: rats received 100 ng ghrelin (i.c.v.) and intrastriatal injection of 200 nmol 3-MA 30 min before intrastriatal injection of 6-OHDA as described before (He et al., 2017); 7) ghrelin + 6-OHDA + CQ group: rats received 100 ng ghrelin (i.c.v.) and 10 mg/kg CQ (intraperitoneal injection, i.p.) 30 min before intrastriatal injection of 6-OHDA as described before (He et al., 2018). To study the effect of ghrelin on autophagy-lysosome pathway, the rats were sacrificed 1 day after corresponding treatments, as autophagy-related markers showed the most notable changes within 1 day in the 6-OHDA-induced rat model based on our previous findings (He et al., 2018). And to observe the long-term effect of ghrelin on rat behavioral changes and the survival dopaminergic neurons, the rats were sacrificed 5 weeks after corresponding treatments.

Apomorphine-induced rotation and cylinder test were assessed at the fifth week after 6-OHDA lesion before sacrifice in a double-blinded manner. For the apomorphine-induced rotation test, the rats received apomorphine 0.5 mg/kg (i.p.) and acclimated for 10 min. Data of asymmetry rotation were consecutively recorded for 30 min. For the cylinder test, the rats were placed in a glass cylinder with a height of 30 cm and diameter of 20 cm and the forelimb use was recorded for 5 min. Data of contralateral paw use was calculated as: (left paw use + bilateral paw use × 0.5)/(left paw use + right paw use + bilateral paw use) × 100%.

Human neuroblastoma cells (SH-SY5Y) were purchased from the Cell Bank of the Institute of Biochemistry and Cell Biology (Shanghai, China). The cells were cultured in DMEM/F12 medium (Biological Industries) supplemented with 10% fetal bovine serum (Biological Industries) and 100 units/ml of penicillin/streptomycin. The cells were incubated at 37°C in a humidified atmosphere with 5% CO₂.

Cell viability was assessed with cell counting kit-8 (CCK-8, Dojindo Molecular Technologies, Japan) according toe the manufacture’s instruction. SH-SY5Y cells were seeded at a concentration of 10 × 10³ per well in a 96-well plate. After different treatments for 24 h, 10 μL of CCK-8 was added to each well and incubated at 37°C for 1 h. The absorbance was measured by a microplate reader at 450 nm.

Adenovirus Ad-mCherry-GFP-LC3B was purchased from Beyotime (Beijing, China) to assess the state of autophagic flux in vitro. SH-SY5Y cells were seeded at a concentration of 10 × 10⁴ per well in a 6-well plate and infected with the adenovirus at a multiplicity of infection (MOI) of 10. The medium was changed 24 h after infection followed by corresponding treatments.

For Atg7 and TFEB knockdown, siRNAs were synthesized by GenePharma (Shanghai, China) and applied. SH-SY5Y cells were seeded in 6-well plates at a density of 20 × 10⁴ per well, and 6 μL of jetPRIME reagent (Polyplus Transfection, France) was applied for siRNA transfection as instructed by the manufacture. The silencing efficiency was assessed with RT-PCR 24 h after transfection.

LysoTracker red (Beyotime) was applied to label lysosomes according to the manufacture’s instruction. In summary, at the end of corresponding treatment, SH-SY5Y cells seeded in 6-well plates were incubated with LysoTracker red diluted at a concentration of 1:15,000 (in DMEM/F12 medium) at 37°C for 30 min and replaced with fresh DMEM/F12 medium. The images were captured under the fluorescence microscope and further analyzed with ImageJ software as described before (Bonet-Ponce et al., 2016).

Western blot was carried out with brain or whole cell extracts obtained by lysing with ripa buffer containing protease and phosphatase inhibitors, and protein concentration was determined with a BCA kit (Beyotime). Equal amount of protein samples were subjected to electrophoresis and then transferred to PVDF membranes (Millpoire, U.S.A.). The membranes were blocked in 5% non-fat milk in TBS with 0.1% Tween 20 (TBST) for 1 h at room temperature followed by incuabation of primary antibodies overnight. After washing the membranes with TBST for three times, they were further incubated with HRP-conjugated secondary antibodies for 2 h at room temperature and visualized with enhanced chemiluminescence (ELC) detection kit (Millipore). The results were quantified with ImageJ software and normalized with the loading control GAPDH.

Total RNA was isolated from SH-SY5Y cells with RNAiso Plus (Takara) according to the manufacture’s instrunction and RNA quality and concentration were determined with spectrophotometer. Rt-PCR was performed with One-Step SYBR PrimeScript RT-PCR kit (Takara) in 7,500 Fast RT-PCR System (Applied Biosystems). The expression of GAPDH mRNA was used as an internal control. And the results were analyzed with–ΔΔCt method normalized with GAPDH.

Sacrificed rats were perfused with normal saline followed by 4% paraformaldehyde. Then the whole brain was fixed in 4% paraformaldehyde overnight followed by dehydration with 30% sucrose for 2 days. Coronal brain sections containing midbrain were prepared at 15 μm with a freezing microtome (Leica, Germany). And SH-SY5Y cells were cultured on coverslips and after corresponding treatments, they were fixed with 4% paraformaldehyde in PBS for 15 min at room temperature. The fixed sections or cells were permeabilized with 0.1% Triton X-100 for 10 min at room temperature and blocked with 5% goat serum for 1 h at room temperature, followed by incubation of primary antibodies at 4°C overnight. After three times washing with PBS, they were incubated with Alexa Fluor 488 conjugated goat anti-mouse and Alexa Fluor 594 conjugated goat anti-rabbit secondary antibodies for 2 h at room temperature, and nuclei were stained with DAPI for 5 min at room temperature. All the images were acquired with a Nikon 300 microscope.

Cell apoptosis was detected with Annexin V-FITC/PI apoptosis assay kit (Bimake). After harvesting the SH-SY5Y cells with different treatments, they were resuspended in the staining buffer and stained with Annexin V-FITC and PI to assess cell death with flow cytometry (BD Bioscience). The apoptosis rate was calculated as “[Annexin V (+)PI(−) cells + Annexin V (+)PI(+) cells]/total cell × 100%”.

The data were presented as mean ± SD. And statistical analysis was carried out with Student’s t-test or one-way analysis of variance (ANOVA) followed by Bonferroni’s multiple group comparison using GraphPad Prism 6.0 Software (GraphPad Software Inc., U.S.A.) and SPSS 15.0 software (SPSS Science, U.S.A.). Statistical significance was accepted with p < 0.05.

To assess whether ghrelin could ameliorate 6-OHDA-induced motor dysfunction, apormorphine-induced rotation and cylinder test were assessed. As shown in Figure 1A, 6-OHDA induced a marked increase in the number of contralateral rotations, while different doses of ghrelin treatment (100 ng, 200 ng, 400 ng/day) could all effectively reduce apormorphine-induced contralateral rotations. Moreover, no statistical difference was observed in terms of the number of apormorphine-induced rotation among the groups with different doses of ghrelin treatment. And for the cylinder test, 6-OHDA could lead to a notable drop in the percentage of contralateral paw use, which could be partially relieved by different doses of ghrelin treatment (p < 0.01).

To study the effect of ghrelin on the survival of dopaminergic neurons, expression of tyrosine hydroxylase (TH) was assessed with western blot and immunofluorescence staining. As TH is the rate-limiting enzyme in catecholamine biosynthesis and decreased TH level is closely related to reduced dopamine synthesis, it is conventionally regarded as an important marker of dopaminergic neuron survival (Haavik and Toska, 1998). As shown in Figure 1B 6-OHDA induced a remarkable decrease in TH level in the rat substantia nigra 5 weeks after intrastriatal injection, which could be reversed by different doses of ghrelin treatment. However, unexpectedly, a dose-dependent decrease in TH expression was observed with increasing the dose of ghrelin treatment. Immunofluorescence staining of TH at the right substantia nigra (SN) also revealed a similar trend that 6-OHDA induced a dramatic loss of TH-positive neurons, which could be effectively reversed by 100 and 200 ng treatment of ghrelin (Figure 1C). Based on these findings, 100 ng ghrelin was applied for further experiments to study its regulation on autophagy and apoptosis in vivo. In SH-SY5Y cells, a similar trend could be observed, with 1 nM ghrelin treatment showing the best neuroprotective effect against 6-OHDA-induced neurotoxicity and 800 nM ghrelin treatment showing even deteriorating cell viability compared with 6-OHDA treatment alone (Figure 1D), and 1 nM ghrelin was applied for further experiments in vitro.

To further study whether the neuroprotective effect of ghrelin is dependent on its receptor, its receptor antagonist, [D-Lys3]-GHRP-6, was applied both in vivo and in vitro. As shown in Figure D, [D-Lys3]-GHRP-6 could abolish the neuroprotective effect of 1 nM ghrelin as revealed by a decline of cell viability. Moreover, by assessing the apoptosis-related markers 1 day after 6-OHDA treatment in vivo, [D-Lys3]-GHRP-6 could partially reverse the decline in bax/bcl-2 ratio as well as cleaved caspase three expression by ghrelin treatment (Figure 1E), suggesting that the neuroprotective effect of ghrelin is at least partially dependent on its receptor.

To characterize the effect of ghrelin on autophagy in 6-OHDA-induced rat model, the rats were sacrificed 1 day after 6-OHDA treatment as our previous study suggested the most notable change at this time point (He et al., 2018). The expression of autophagy-related protein LC3 and dopaminergic neuron marker TH was assessed with double immunofluorescence staining (Figure 2A). An increased LC3 signal could be observed in the right SN of the rats treated with 6-OHDA or 6-OHDA + ghrelin compared with the control group, which colocalized with TH staining, suggesting an activation of autophagy in dopaminergic neurons. However, despite a light increase in the number of TH and LC3 double positive cells in the 6-OHDA + ghrelin, there was no statistical difference compared to that of the 6-OHDA group. As for the participation of the microtubule-associated protein LC3 in autophagosome biogenesis, LC3-II is modified by addition of phosphatidylethanolamine to LC3-I and serves as a maker for complete autophagosome formation (Klionsky et al., 2016), LC3-II, instead of LC3, is a better readout of the autophagy activation process. To further characterize the effect of ghrelin on autophagy activation, several autophagy markers including Atg7, LC3-II and p62 were assessed with western blot (Figure 2B). Atg7 is a noncanonical E1 enzyme that is involved in autophagosome formation (Hong et al., 2011), and p62 is one of the substrates in the autophagy process that can serve as a readout of autophagic degradation (Benito-Cuesta et al., 2017). As shown in Figure 2B, both 6-OHDA and ghrelin could induce a notable increase in Atg7 and LC3-II expression compared with the control group. And compared with the 6-OHDA treatment group, ghrelin treatment could further upregulate the expression of Atg7 and LC3-II. In contrast, compared with the control group, a marked decrease in p62 level could be observed in either single ghrelin or 6-OHDA group, and the expression of p62 was significantly lower in the ghrelin+6-OHDA group compared with the 6-OHDA group. Taken together, these results indicated that ghrelin could further induce autophagy activation in 6-OHDA-induced PD rat model.

As our previous study has revealed autophagic flux dysfunction in 6-OHDA-induced PD model, which contributed to the neuron death (He et al., 2018), and the in vivo results above could not fully reveal the actual state of autophagic flux upon ghrelin treatment, we further applied adenovirus Ad-mCherry-GFP-LC3B to assess the effect of ghrelin on autophagic flux in the 6-OHDA-induced SH-SY5Y cell model. We tinfected SH-SY5Y cells with Ad-mCherry-GFP-LC3B, further treated with ghrelin and 6-OHDA for 24 h and assessed the autophagic flux under the fluorescent microscope. Upon infection of Ad-mCherry-GFP-LC3B, autophagosomes are labeled with yellow signals due to both mCherry and GFP fluorescence; however, upon autophagosome and lysosome fusion, the autolysosomes will present red puncta as GFP signal is more rapidly quenched by the low PH of lysosomes. As shown in Figure 3, cells in the control group showed a weak and diffused signal in both green and red, while single ghrelin treatment could lead to a marked increase in red puncta, indicating an increased autophagosome and lysosome fusion. In contrast, 6-OHDA treatment alone yielded more yellow puncta, suggesting inadequate autophagosome and lysosome fusion. This could be partially reversed by ghrelin treatment, suggesting that ghrelin could promote more autophagosome and lysosome fusion and partially alleviate the autophagic flux dysfunction induced by 6-OHDA.

To elucidate whether the neuroprotective effect of ghrelin was due to autophagic flux regulation, two inhibitors that block different autophagy stages were applied. 3-methyladenine (3-MA) could inhibit class III phosphoinositide 3-kinase-dependent formation of autophagosomes, while chloroquine could effectively inhibit the fusion between autophagosomes and lysosomes. The doses and application of 3-MA and CQ have been validated in our previous study (He et al., 2018). As shown in Figure 4A, 6-OHDA induced a remarkable increase in the number of TH and cleaved caspase three double positive cells, which could be partially reversed by ghrelin treatment. Moreover, inhibiting autophagy induction with 3-MA showed a slight decrease in the number of TH and cleaved caspase three double positive cells compared with 6-OHDA + ghrelin group, but with no statistical difference. While chloroquine could induce a remarkable increase in the number of TH and cleaved caspase three double positive cells compared with 6-OHDA + ghrelin group Western blot of apoptosis-related markers also revealed a parallel trend. Ghrelin could notably reverse 6-OHDA-induced increase in bax/bcl-2 ratio as well as cleaved caspase three level, and 3-MA treatment did not further cause any remarkable change in bax/bcl-2 ratio and cleaved caspase three level compared with 6-OHDA + ghrelin group at 1 day post treatment. On the other hand, chloroquine induced a marked increase in bax/bcl-2 ratio and cleaved caspase three level compared with 6-OHDA + ghrelin group, suggesting that chloroquine treatment could abolish the neuroprotective effect of ghrelin (Figure 4B).

As autophagosomes requires constant fusion with lysosomes for further digestion and recycling, the state of lysosomes is critical to a healthy autophagic flux. Our previous study has revealed that 6-OHDA could impair the autophagic flux by downregulating TFEB (He et al., 2018), which is a critical transcription regulator for lysosome biosynthesis and function. We further assessed the effect of ghrelin on TFEB expression with immunofluorescence staining. As shown in Figure 5A, ghrelin could upregulate TFEB expression, while 6-OHDA induced a dramatic decline in TFEB fluorescent intensity, which could be partially reversed by ghrelin co-treatment. Moreover, ghrelin could not only induce a general increase in TFEB positive signal, an increase of TFEB positive signal in the nucleus compartment could also be observed, indicating an activation of the TFEB transcriptional activity (Figure 5A). To further validate the downstream effect of TFEB activation, lysosomes were visualized with LysoTracker staining in SH-SY5Y cells. As shown in Figure 5B, compared with control group, single ghrelin treatment could lead to a notable increase in LysoTracker red fluorescence intensity, while 6-OHDA treatment led to a drmatic decline (p < 0.01), which could be partially reversed by ghrelin treatment, further indicating that ghrelin could ameliorate 6-OHDA-induced autophagic flux dysfunction by preserving TFEB level and lysosome expression.

To further validate the association between autophagic flux regulation and the neuroprotective action of ghrelin, we applied ATG7 siRNA and TFEB siRNA independently to inhibit autophagy activation and lysosome biogenesis and function. The silencing efficiency was validated with rt-PCR (Figure 6A,B). After transfecting SH-SY5Y cells with NC or siRNAs, they were further treated with 6-OHDA with or without ghrelin for 24 h. Cell apoptosis was assessed with Annexin V-PI/FITC kit through a flow cytometer. As shown in Figure 6C, 6-OHDA induced a notable increase in the number of apoptotic cells compared with the NC group, which could be partly reversed by ghrelin treatment. Moreover, ATG7 siRNA could further reduce the number of apoptotic cells compared with 6-OHDA + ghrelin treatment group (p < 0.05). In contrast, TFEB siRNA treatment showed an increase in the number of apoptotic cells, which was even higher than that in the 6-OHDA treatment group (p < 0.01). Collectively, these findings suggested that instead of via modulating autophagy activation, ghrelin exerts the neuroprotective effect against 6-OHDA more via preserving TFEB expression and thus relieving the autophagic flux dysfunction.