Ferulic Acid Exerts Neuroprotective Effects via Autophagy Induction in C. elegans and Cellular Models of Parkinson’s Disease

Ferulic acid (FA, F103701), bafilomycin A1 (Baf, B101389), and levodopa (L-Dopa, D111048) were purchased from Aladdin Bio-Chem Technology Co., Ltd. (Shanghai, China). 6-Hydroxydopamine (6-OHDA, H4381), 5-fluoro-2′-deoxyuridine (FUDR, F0503), and rapamycin (Rap, V900930) were purchased from Sigma-Aldrich (St. Louis, USA). C. elegans strains used in this study were as follows: wild-type Bristol N2, transgenic BZ555 strain [Pdat-1::gfp], N5901 strain [unc-54p::alpha synuclein::YFP + unc-119(+)], BC12921 strain [rCes T12G3.1::GFP + pCeh361], and DA2123 strain [lgg-1p::GFP::lgg-1 + rol-6(su1006)]. All strains were obtained from the Caenorhabditis Genetics Center (CGC) and cultured at 20°C on nematode growth medium (NGM) agar plates carrying a lawn of Escherichia coli OP50 as a food source unless otherwise stated. Synchronized eggs (embryos) were isolated from gravid adults by bleaching solution (0.5 M NaOH and 1% NaClO) and incubated in M9 buffer at 20°C overnight to obtain synchronized L1 larvae. Then, L4 larval worms were transferred onto NGM plates containing 5 mg/L FUDR to prevent progeny from hatching.

Age-synchronized NL5901 nematodes were bred on NGM agar plates with or without FA to reach the L4 developmental stage. Then, worms were transferred onto new NGM plates containing FUDR (5 mg/L) to prevent the growth of progeny. When the animals grew to L4 + 5 or L4 + 10 days, they were independently placed in a drop of M9 buffer and allowed to recover for 90 s to avoid observing behavior caused by stress. Then, the number of body bends was counted in 20 s. At least 20 animals were used to analyze in each experiment, and all experiments were carried out in triplicate.

NL5901 nematodes were used to determine the accumulation of α-synuclein protein as reported previously [15, 16]. In brief, synchronized NL5901 L1 larvae were incubated on NGM plates with or without FA (50 μM) for 7 days and 12 days at 20°C. Then, worms were washed with M9 buffer 3 times and transferred to 2% agarose pad slides with 100 mM sodium azide. Immobilized animals were captured under a positive fluorescence microscope (Leica DM6B, Leica Microsystems GmbH, Germany) to monitor the accumulation of α-synuclein, and the fluorescence intensity was quantified by using ImageJ software (National Institutes of Health, Bethesda, MD, USA). At least 20 animals were used to analyze in each group, and all experiments were carried out in triplicate.

The assay of DA neurodegeneration was performed in 6-OHDA-induced BZ555 worms as described previously [15, 16]. In brief, synchronized L3 BZ555 worms were incubated with 50 mM 6-OHDA and 10 mM ascorbic acid in S-medium blended with OP50 in the presence or absence of L-Dopa or FA. After treatment for 1 h, the worms were washed with M9 buffer 3 times and transferred onto NMG plates containing L-Dopa or FA for 24 h. Then, 5 mg/L of FUDR was added to inhibit the production of progeny. After 48 h of treatment, BZ555 worms were washed 3 times with M9 buffer and then mounted onto a glass slide with 2% agarose pad using 100 mM sodium azide and enclosed with a coverslip. The immobilized animals were observed and photographed under a positive fluorescence microscope, and the fluorescence intensity of DA neurons was measured by using ImageJ software. At least 20 animals were used to analyze in each group, and all experiments were carried out in triplicate.

The food-sensing behavioral test was performed in 6-OHDA-induced BZ555 worms according to the method described previously [15, 16]. In brief, synchronized L3 BZ555 worms were incubated with 50 mM 6-OHDA and 10 mM ascorbic acid in S-medium blended with OP50 in the presence or absence of L-Dopa or FA for 1 h. Then, worms were washed with M9 buffer 3 times and transferred onto NMG plates containing L-Dopa or FA. After 72 h, BZ555 worms were transferred to the center of NGM plates spotted with or without E. coli OP50 lawn and allowed to recover for 90 s to avoid observing behavior caused by stress. Then, the numbers of body bending of worms were counted in 20 s intervals. The slowing rate of body bending was calculated by the following formula:

where N represents the total number of body bending in the presence or absence of bacteria.

For H₂O₂-induced N2 worms, 4 gravid adults were allowed to lay eggs for 6 h in NGM plates containing H₂O₂ (1 mM) in the presence or absence of FA (50 μM) or NAC (5 mM). Then, the adult worms were picked out, and the NGM plates were continued to culture at 20°C for 72 h. After treatment, worms were washed with M9 buffer and incubated in 1 mL M9 containing 100 μM dihydroethidium (DHE) for 1 h in the dark. After incubation, the tubes were centrifuged at a speed of 2000 rpm, and the worms were mounted onto a glass slide in M9 medium containing 100 mM sodium azide. Images were captured using a positive fluorescence microscope. For quantifying the RFP fluorescence intensity, images were analyzed and quantified by ImageJ software.

The survival assay of C. elegans under acute oxidative stress was performed according to the previous study [17]. Briefly, synchronized L1 N2 worms were maintained at 20°C and treated with FA for 48 h. After that, worms were transferred to new NGM medium containing 40 mM H₂O₂, and the survival rate of worms was counted after 0-12 h of H₂O₂ treatment.

The autophagy effect of FA was monitored by counting GFP-positive LGG-1/Atg8 puncta in the hypodermal seam cells of DA2123 strain and by detecting the GFP fluorescence of p62/SQST-1-GFP fusion protein in the BC12921 strain. In brief, the two strains were maintained at 20°C and treated with FA or Rap (a positive control) on the day after eggs hatching for 48 h. Then, the worms were mounted onto a glass slide with 2% agarose pad using 100 mM sodium azide to observe and capture representative photographs under a positive fluorescence microscope. The fluorescence intensity representing the total number of GFP::LGG-1/Atg8 puncta in the DA2123 strain or indicating p62 expression in the BC12921 strain was measured by using ImageJ software as previously described [18].

RNA interference (RNAi) of the 3 key autophagy-related genes including lgg-1, vps-34, and unc-51 was performed according to the previously described procedures [19]. In brief, the synchronized L1 worms of NL5901 or BZ555 were transferred onto the NGM plates with the E. coli HT115 bacteria expressing the double-stranded RNA of individual genes or the control bacteria HT115 in the presence or absence of FA. After treatment, the adult nematodes were used for the analysis of α-synuclein and ROS accumulation and degeneration of DA neurons as described above.

Stable GFP-RFP-LC3 U87 cells were a generous gift kindly provided by Dr. Xiaoming Zhu (Macau University of Science and Technology, Macao, China) and cultured in α-MEM supplemented with 10% fetal bovine serum (FBS) and 50 U/mL penicillin and 50 μg/mL streptomycin (Invitrogen, Scotland, UK). PC-12 cells were purchased from the American Type Culture Collection (ATCC) (Rockville, MD, USA). PC-12 cells were cultured in DMEM supplemented with 10% horse serum (HS), 5% FBS, and 50 U/mL penicillin and 50 μg/mL streptomycin. All cells were maintained in the 5% CO₂ incubator with 75% humidity at 37°C. To prevent direct interaction between FA and 6-OHDA and H₂O₂ in the culture medium, PC-12 cells were pretreated with FA for 12 h prior to the treatment of 6-OHDA or H₂O₂. At the end of FA treatment, the medium was exchanged for fresh medium containing 6-OHDA or H₂O₂. Then, the PC-12 cells were incubated with 6-OHDA or H₂O₂ for another 12 h.

The PC-12 cells were seeded onto 96-well plates at a density of 4 × 10³ cells/well and treated with the test drugs for 24 h. After treatment, 10 μL of MTT solution (5 mg/mL) was added into each well followed by incubation for 4 h. Then, the medium was removed, and 100 μL of DMSO was added into each well. After shaking at a low speed for 10 min at room temperature, the colorimetric reading of each well was obtained with a spectrophotometer (BioTek, VT Lab, USA) at 570 nm. The cytotoxicity was measured by calculating the cell viability according to the following formula: cell viability (%) = 100 × OD value_{Cells treated}/OD value_{Cells untreated}. The data were obtained from three separate experiments.

Cell apoptosis was measured with a flow cytometer using the Annexin V-FITC/PI Detection Kit (BD Biosciences, CA, USA). In brief, after treatment, cells were collected and centrifuged at a speed of 2000 rpm for 5 min. The supernatant was removed, and the cell pellet was resuspended with 500 μL of 1× Annexin V solution containing 4 μL of propidium iodide (PI) and 2 μL of FITC solution according to the manufacturer's instructions. After 15 min incubation in the dark, the cells were analyzed with a FACSVerse flow cytometer (BD Biosciences, San Jose, CA, USA). Data acquisition and analysis were performed by using BD FACSuite v1.0.6 software (BD Biosciences, San Jose, CA, USA).

PC-12 cells were seeded onto 96-well plates at a density of 4 × 10³ cells/well and treated with the test drugs for 24 h. After treatment, PC-12 cells were washed with PBS 3 times. After that, the cells were stained with 5 mg/L Hoechst 33342 and 5 mg/L PI solution for 5 min. After incubation, the liquid was sucked dry, and 100 uL of PBS was added into wells. Representative images were captured and analyzed under a fluorescence microscope (Nikon ECLIPSE 80i, Tokyo, Japan). The cell apoptosis was measured by calculating the percentage of cells with PI signal (red fluorescence) in cells with Hoechst signal (blue fluorescence).

The intracellular ROS levels were determined as previously described by flow cytometry analysis using the H₂DCFDA fluorescence probe [20]. Briefly, after treatment, PC-12 cells were trypsinized and collected for centrifugation at a speed of 1000 rpm for 5 min. Then, the cell pellet was washed twice with PBS. After that, the supernatant was removed, and the remaining cell pellet was then resuspended in PBS with 20 μM H₂DCFDA and continued to incubate for 20 min in the dark. After incubation, the cell suspension was then subjected to flow cytometry analysis for the measurement of ROS levels on a FACSVerse flow cytometer (BD Biosciences, San Jose, CA, USA). Data acquisition and analysis were performed by using the FlowJo software (BD Biosciences, San Jose, CA, USA).

For the quantification of GFP-LC3 puncta formation, stable GFP-RFP-LC3 U87 cells were employed. In brief, cells were treated with FA at the indicated concentrations or Rap (0.5 μM) for 24 h. After treatment, the cells were fixed with 4% paraformaldehyde (PFA) for 20 min at room temperature, followed by the wash with PBS twice. Then, the slides were brought out for air-drying and mounted by using FluorSave^™ mounting media (Calbiochem, San Diego, CA, USA). Representative images of cells were captured by using a Nikon ECLIPSE 80i fluorescence microscope equipped with a CCD digital camera Spot RT3^™ (Diagnostic Instruments, Inc., Melville, NY, USA). The autophagic effect was evaluated by calculating the total number of punctate GFP-LC3 fluorescence formations in the GFP-positive cells. In this study, at least 60 GFP-positive cells were counted from 3 randomly selected fields of the slides.

All data in this study were obtained from three or more independent experiments. The data expressed as means ± SD were analyzed by using the GraphPad Prism 6.0 software (San Diego, CA, USA) using one-way ANOVA followed by the pos hoc Tukey test. p < 0.05 was considered to have significant differences among the compared groups.

In this study, we investigated the effect of FA on the inhibition of α-synuclein accumulation and the improvement of motility in transgenic NL5901 nematodes. It is reported that NL5901 worms express human α-synuclein fused with yellow fluorescent protein (YFP) under the control of the unc-54 promoter in the body wall muscle cells [21], and there are three advantages in this model. First, the unc-54 promoter is powerful, and muscle cells are the largest and easiest cell type to score, which allows accurate detection of α-synuclein accumulation and its subcellular localization. Second, the expression and subsequent aggregation of α-synuclein in the muscle result in PD-like progressive decline of motility in C. elegans, demonstrating the in vivo toxicity of α-synuclein aggregates [22]. Third, this model has been successfully used to identify compounds possessing anti-PD effects and modifier genes of PD in previous reports [23, 24]. In the present study, we found that 25–200 μM of FA significantly decreased the YFP intensity in NL5901 worms (Figure 1(a)), suggesting that FA can inhibit the accumulation of α-synuclein in vivo. Meanwhile, 25–200 μM of FA also improved the motor ability of NL5901 worms, as evidenced by the increased number of body bends in 20 s (Figure 1(b)). Among these concentrations, 50 μM of FA exhibited the best effect. Thus, 50 μM was selected for FA as the optimal concentration in the subsequent experiments. In addition, we further investigated the inhibitory effect of FA on the accumulation of α-synuclein in NL5901 worms at different adulthood stages. The results showed that FA treatment significantly reduced the YFP intensity of NL5901 worms at both day 7 (day 5 of adults) and day 12 (day 10 of adults) with the reduced rates of 13.9% and 18.3%, respectively (Figures 1(c) and 1(d)). With the production and aggregation of α-synuclein, the locomotor capacity of NL5901 worms decreases with aging. It was found that FA significantly alleviated the motor dysfunction of NL5901 worms (Figure 1(e)). Taken together, our results demonstrate that FA can ameliorate the pathology of PD in C. elegans, manifesting in the reduction of α-synuclein accumulation and the improvement of locomotor capacity.

There is a growing body of evidence indicating that the degeneration or death of DA neurons is one of the main pathological features of PD [1]. C. elegans possess exactly eight DA neurons, containing two pairs of cephalic (CEP) neurons and one pair of anterior deirid (ADE) neurons in the head region and one pair of posterior deirid (PDE) neurons in the posterior lateral region [23, 25]. It has been reported that 6-OHDA can enter into these DA neurons through the DAT-1 dopamine transporter and thus trigger their degeneration in worms, while the addition of L-Dopa, which serves as the precursor of dopamine and is converted to dopamine by aromatic l-amino acid decarboxylase, leads to an increase in dopamine levels, thus effectively protecting DA neuron loss in 6-OHDA-exposed worms [26]. In this study, we evaluated whether FA protected against the degeneration of DA neurons using the transgenic BZ555 nematodes. It specifically expressed the green fluorescent protein (GFP) in DA neurons under the promoter of the dat-1 gene (Pdat-1::GFP), which encodes for the dopamine transporter, DAT-1. Due to the specific fluorescent expression in DA neurons, Pdat-1::GFP was used as the reporter construct to monitor the changes of dopaminergic neuronal viability [27]. As expected, 6-OHDA exposure led to the degeneration of DA neurons in BZ555 worms, as indicated by the decreased GFP intensity. However, the supplement of FA or L-Dopa remarkably recovered the GFP intensity to 15.0% and 27.8%, respectively (Figures 2(a) and 2(b)). These results suggest that FA and L-Dopa inhibit 6-OHDA-induced DA neurodegeneration in C. elegans.

In addition, it has been confirmed that 6-OHDA-induced BZ555 worms exhibit the food-sensing behavioral deficit [28]. Normally, C. elegans bend their bodies for transportation and foraging. When C. elegans come across the bacteria lawn or food source, they decline their bending frequency to feed themselves more effectively. Thus, the bending frequency is commonly used to estimate the food-sensing behavior of worms. As shown in Figure 2(c), the BZ555 worms without treatment exhibited a decline in bending frequency with a slowing rate of 40.5% upon contact with the bacteria food. However, 6-OHDA-induced BZ555 worms failed to decrease the bending frequency with a slowing rate of 3.6%. After treatment with FA or L-Dopa, 6-OHDA-induced BZ555 worms markedly slowed down their bending frequency, and their slowing rates were improved to 20.1% and 30.3%, respectively (Figure 3(c)). Collectively, these results indicate that FA decreases the degeneration of DA neurons in 6-OHDA-induced BZ555 C. elegans.

Previous studies have proved that FA is an effective scavenger of free radicals, suggesting that it possesses great antioxidant capacity [29]. Here, we investigated whether FA could eliminate the excessive ROS generation in 6-OHDA-induced BZ555 worms using the fluorescent dihydroethidium (DHE) probe. As shown in Figures 4(a) and 4(b), the BZ555 worms in the 6-OHDA group exhibited a significant increase in DHE intensity compared to the untreated BZ555 worms, while the treatment with FA or L-Dopa could attenuate the DHE intensity in 6-OHDA-induced BZ555 worms. Therefore, these data suggest that FA inhibits the generation of ROS and exerts antioxidative activity in 6-OHDA-induced BZ555 worms.

To further confirm the antioxidative effect of FA in C. elegans, H₂O₂ was used to induce oxidative stress in wild-type N2 nematodes, and the ROS levels were measured using the fluorescent DHE probe. The results showed that H₂O₂-treated worms exhibited an increase in the DHE fluorescence signal, which was significantly reduced by the supplement of FA or NAC, suggesting that FA can reduce the ROS levels in H₂O₂-treated C. elegans (Figures 4(c) and 4(d)). In addition, we also assessed the resistance of FA to acute oxidative stress induced by 40 mM H₂O₂. By measuring the lifespan of nematodes, we found that FA significantly increased the survival rate of C. elegans (Figure 4(e)), suggesting that FA extends the lifespan of H₂O₂-treated C. elegans via the antioxidation mechanism.

Mounting evidence demonstrates that autophagy is a crucial cellular catabolic process involved in the degradation of various cytosolic components, while the defects in autophagy are critically associated with aging and aging-related diseases such as PD. In this study, two C. elegans strains including BC12921 and DA2123 were employed to evaluate the autophagy induction of FA in vivo. In the BC12921 strain, the autophagy substrate protein SQST-1/p62 fused to GFP is expressed under the control of the promoter of the sqst-1 gene. In the DA2123 strain, a number of puncta labeled with GFP::LGG-1 accumulate in the hypodermal seam cells. Our results showed that the treatment with FA or Rap significantly reduced the GFP intensity in BC12921 worms, suggesting that FA or Rap can inhibit the expression of SQST-1/p62 protein (Figures 3(a) and 3(b)). Meanwhile, the treatment with FA or Rap remarkably increased the number of LGG-1-GFP-positive puncta in DA2123 worms (Figures 3(c) and 3(d)). Furthermore, qRT-PCR results showed that FA administration significantly increased the mRNA expression levels of 3 key autophagy-related genes including lgg-1, vps-34, and unc-51 in N2 worms. Collectively, these data strongly demonstrate that FA induces autophagy in C. elegans.

To investigate whether FA inhibits the accumulation of α-synuclein in NL5901 worms via autophagy induction, we knocked down autophagy-related genes by feeding NL5901 worms with RNAi bacteria and then examined the expression of α-synuclein. As shown in Figures 5(a) and 5(b), the treatment of FA or L-Dopa significantly reduced the accumulation of α-synuclein in NL5901 worms supplied with the control bacteria HT115. However, the effect of FA on the inhibition of α-synuclein was largely compromised when NL5901 worms were fed with the RNAi bacteria of 3 autophagy-related genes including lgg-1, vps-34, and unc-51. In addition, the mobility of NL5901 worms was also evaluated by counting the body bending frequency of worms in 20 s. The results showed that the treatment with FA or L-Dopa significantly enhanced the locomotor capacity of NL5901 worms fed with the control bacteria HT115, while the feeding of RNAi bacteria eliminated the effect of FA on the improvement of locomotion capacity (Figure 5(c)). Together, these data suggest that FA degrades α-synuclein and improves the locomotion capacity of NL5901 C. elegans via autophagy induction.

It has been reported that autophagy dysfunction is closely related to the degeneration of DA neurons in neurodegenerative diseases [8]. In this study, we further studied whether FA inhibited the degeneration of DA neurons in 6-OHDA-induced BZ555 worms via autophagy induction. The results showed that the treatment with FA or L-Dopa significantly inhibited the degeneration of DA neurons in 6-OHDA-induced BZ555 worms, which was indicated by the recovered GFP intensity (Figures 6(a) and 6(b)). However, the feeding of RNAi bacteria targeting autophagy-related genes including lgg-1, vps-34, and unc-51 successfully eliminated the effect of FA on the inhibition of DA neuron degeneration in 6-OHDA-induced BZ555 worms (Figures 6(a) and 6(b)). Meanwhile, the feeding of RNAi bacteria also attenuated the improvement effects of FA on the ROS accumulation and food-sensing behavior in 6-OHDA-induced BZ555 worms (Figures 6(c) and 6(d)). Taken together, these data suggest that FA inhibits the DA neuron degeneration and ROS production and improves the food-sensing behavior in 6-OHDA-induced BZ555 C. elegans via autophagy induction.

To determine whether FA exerts the neuroprotective effects in vitro, the PD neurotoxin 6-OHDA was used to induce oxidative damage in PC-12 cells. Firstly, the cytotoxicity of 6-OHDA and FA against PC-12 cells was examined by the MTT assay (Figures 7(a) and 7(b)). Then, the cytoprotective effect of FA in 100 μM of 6-OHDA-induced PC-12 cells was evaluated. The results showed that FA dose-dependently recovered the viability of PC-12 cells (Figure 7(c)). It is known that the neurotoxicity of 6-OHDA is closely associated with the generation of ROS and the possible direct inhibition of the mitochondrial respiratory chain complex I [30]. Therefore, we here investigated the cytoprotective effect of FA in the H₂O₂-induced oxidative damage model of PC-12 cells. The MTT assay showed that the cell viability of PC-12 was decreased by H₂O₂ (Figure 7(d)), while the treatment with FA or NAC (a ROS scavenger) significantly recovered the viability of 200 μM H₂O₂-induced PC-12 cells in a dose-dependent fashion (Figure 7(e)). Since the cell death induced by 6-OHDA or H₂O₂ was highly accompanied by the increase in intracellular ROS levels, we then investigated whether FA could decrease the production of ROS levels in 6-OHDA- or H₂O₂-treated PC-12 cells. As shown in the flow cytometric analysis in Figures 8(a) and 8(b), 6-OHDA and H₂O₂ significantly increased the ratio of cells with H₂DCFDA fluorescence representing the ROS levels in cells, which was remarkably decreased by FA treatment. Together, our results demonstrate that FA can enhance the cell viability and reduce the ROS levels in 6-OHDA- or H₂O₂-treated PC-12 cells.

The emerging evidence indicates that 6-OHDA and H₂O₂ induce neuronal apoptosis, which is an important feature of PD. Thus, we investigated whether FA could protect neurons against 6-OHDA- and H₂O₂-induced apoptosis in PC-12 cells by the PI/Hochest33342 staining assay and flow cytometry analysis. As shown in Figures 9(a) and 9(b), PC-12 cells were observed with a high percentage of cells with the PI signal in cells with the Hoechst signal after exposure to 6-OHDA and H₂O₂ for 24 h, respectively, while pretreatment with FA significantly reduced the ratio of PI/Hoechst. In addition, 6-OHDA and H₂O₂ exposure significantly increased the apoptosis of PC-12 cells with the rate of 44.86% and 45.89%, respectively (Figures 9(c) and 9(d)). However, the pretreatment of FA could decrease the apoptosis rate to 10.54% and 8.64%. Together, these results suggest that FA can markedly protect the cell death of PC-12 cells by inhibiting apoptosis.

Increasing evidence indicates that autophagy plays a prominent role in maintaining cellular homeostasis via resisting cellular damage [31]. Therefore, we here investigated whether FA induced autophagy and improved 6-OHDA- and H₂O₂-induced injury via autophagy induction in PC-12 cells. As shown in Figures 10(a) and 10(b), FA treatment dose-dependently increased the formation of GFP-LC3 puncta in stable RFP-GFP-LC3 U87 cells. Additionally, the autophagic effect was also evaluated by measuring the conversion of LC3-I to LC3-II and the levels of Beclin 1 and SQSTM1/p62 proteins in PC-12 cells using the Western blotting method. The results showed that FA treatment significantly improved the ratio of LC3-II/LC3-I and Beclin 1 proteins and reduced the expression of SQSTM1/p62 proteins in a dose-dependent manner (Figures 10(c)–10(f)). Together, these data suggest that FA can induce autophagy in cells. Then, we employed Baf, an inhibitor of the late phase of autophagy, to inhibit the effect of FA on autophagy induction and determined the viability of 6-OHDA- and H₂O₂-treated PC-12 cells. The results showed that FA treatment significantly increased the viability of PC-12 cells in the presence of 6-OHDA or H₂O₂, while Baf successfully eliminated the protective effect of FA on rescuing 6-OHDA- or H₂O₂-induced cell death in PC-12 cells (Figures 10(g) and 10(h)). In addition, we also detected the ROS levels in 6-OHDA- or H₂O₂-induced PC-12 cells with the treatment of FA and Baf. The flow cytometry results showed that FA alone supplement remarkably reduced ROS levels in 6-OHDA- or H₂O₂-treated PC-12 cells, which were significantly eliminated by the addition of Baf (Figures 10(i) and 10(j)). Thus, these results suggest that FA improves cell viability and reduces ROS levels in 6-OHDA- or H₂O₂-induced PC-12 cells via activating autophagy.

All the figures used to support the findings of this study are included within the article.