Salsolinol as an RNA m6A methylation inducer mediates dopaminergic neuronal death by regulating YAP1 and autophagy.

Parkinson’s disease (PD) is the second most common progressive neurodegenerative disorder worldwide (Tolosa et al., 2021). The pathological characteristics of PD include the loss of the dopaminergic neurons in the substantia nigra pars compacta (SNpc) in the midbrain and the depletion of dopamine in the striatum (Chung et al., 2020; Tolosa et al., 2021; Dong et al., 2024; Park et al., 2024). Another key characteristic of PD is the presence of Lewy bodies in neurons that are mainly composed of abnormal aggregated α-synuclein (Poewe et al., 2017). Only approximately 5%–10% cases of PD are familial PD. Most PD cases are “sporadic” and caused by genetic and environmental factors and gene-environment interactions (Cherian and Divya, 2020). Epidemiologic and experimental studies have shown that the risk of PD is associated with exposure to environmental toxicants, such as pesticides, metals, and other pollutants (Goldman, 2014; Bellou et al., 2016). Salsolinol (1-methyl-6,7-dihydroxy-1,2,3,4-tetrahydroisoquinoline, Sal) is a tetrahydroisoquinoline (TIQ) that is synthesized from the condensation of dopamine and acetaldehyde through a Pictet-Spengler reaction or catalyzed by Sal synthase (Chen et al., 2018). Previous studies showed that the concentration of Sal is higher in the cerebrospinal fluid and urine of patients with PD compared with other samples or patient groups (Moser et al., 1996; Mravec, 2006). Moreover, its derivative, N-methyl-Sal, is structurally similar to 1-methyl-4-phenyl-1,2,3,6-tetrahydropyridine, which is an environmental toxin that causes PD (Herraiz, 2016; Voon et al., 2020). High production of Sal and N-methyl-Sal was observed in α-synuclein-overexpressing PD models (Zhang et al., 2013; Kurnik et al., 2015). Sal causes the release of cytochrome C, which leads to an increase in oxidative stress and causes oligomeric α-synuclein (Su et al., 2013). Sal is present at a higher concentration than other TIQs in the human brain, especially in areas with high dopamine synthesis and turnover, such as the ventral midbrain and striatum, respectively (DeCuypere et al., 2008). The levels of Sal and N-methyl-Sal were lower in the caudate nuclei of PD samples than that in normal human brain, and this reduction was caused by nigrostriatal dopaminergic cell death (DeCuypere et al., 2008). These findings indicated that Sal might play two contrasting roles, which might be the consequence or the cause of the loss of dopaminergic neurons (Kurnik-Łucka et al., 2018, 2020). Sal has been established as a neurotoxin that can alter the function of dopaminergic neurons and dopamine metabolism in the central nervous system (Bae et al., 2008; Voon et al., 2020; Wang et al., 2022). However, the mechanism by which Sal mediates dopaminergic neuronal death remains unclear. Many studies have shown that epigenetic modifications, including DNA methylation, miRNA expression, and histone modifications, play an important role in the progression of PD (Wüllner et al., 2016; Pavlou and Outeiro, 2017; Zhang et al., 2023). Methylation of adenosine at the nitrogen-6 position (N 6 -methyladenosine, m 6 A) is the most prevalent RNA modification in eukaryotes (Chen et al., 2021b). m 6 A regulates the maturation, stability, and translation of mRNAs (An and Duan, 2022). m 6 A RNA methylation occurs at the consensus sequence 5′-RRACU-3′ (R = A or G) and is dynamically and reversibly mediated by methyltransferases (“writers”) and demethylases (“erasers”) (Chen et al., 2021b). In mammalian cells, the m 6 A methyltransferase complex contains methyltransferase-like 3 (METTL3), methyltransferase-like 14, and Wilms tumor 1-associated protein (Oerum et al., 2021). The only two identified m 6 A demethylases are fat-mass and obesity-associated protein (FTO) and alkB homolog 5 (ALKBH5), which both belong to the AlkB protein family and are α-ketoglutarate-dependent dioxygenases (Shen et al., 2022). m 6 A-binding proteins with YTH domains (YTHDF1–3 or YTHDC1–2) act as m 6 A “readers” and specifically bind m 6 A-modified RNA to regulate its degradation or translation (Zhao et al., 2020). Dysregulation of m 6 A RNA modification has been linked to neurological diseases (Zhang et al., 2022). Studies have found that m 6 A upregulation strongly induces defects in dopaminergic neurotransmission. In the midbrain and striatum of FTO-deficient mice, several mRNAs in the dopamine signaling pathway, which are related to the transmission of dopamine, showed increased m 6 A modification levels and altered expression (Hess et al., 2013). The dysfunction of the nigro-striatal projection system in mice caused by manganese exposure was rescued by the overexpression of FTO, which showed m 6 A can reverse dopamine neurodegeneration caused by environmental factors (Qi et al., 2022). Arsenite, an environmental toxin, increases the risk of neurological disorders by altering dopamine levels. m 6 A upregulation was shown to promote the arsenite-induced neurotoxic process (Bai et al., 2018). However, reduction of m 6 A in the SNpc region caused by the deletion of methyltransferase-like 14 induced motor dysfunction, followed by downregulation of tyrosine hydroxylase (TH), an important rate-limiting enzyme in dopamine synthesis (Teng et al., 2021). We previously showed that m 6 A reduction increases oxidative stress and Ca 2+ influx, which promotes the apoptosis of dopaminergic neurons (Chen et al., 2019). FTO inhibitors showed efficacy in protecting dopaminergic neurons (Selberg et al., 2021). These studies indicated that balance of m 6 A might affect the transmission of dopamine and dopaminergic neurodegeneration. Sal has neurotoxicity and is associated with dopamine levels in the brain. m 6 A, a widely expressed modification in the brain, plays a crucial role in regulating dopamine nerve regeneration. Therefore, we validated the impact of Sal on m 6 A. In this study, we hypothesized that Sal might regulate dopaminergic neuronal death via m 6 A RNA modification. We found that Sal significantly upregulated the global level of m 6 A RNA modification in dopaminergic neurons. We then used RNA sequencing (RNA-Seq) and FTO and ALKHB5 knockdown models to identify the targets involved in the toxic effects of dopaminergic neurons. This study provides insight into the neurotoxic mechanism of Sal in regulating RNA methylation and betters our understanding of the neurotoxic effects of TIQs and the pathogenesis of PD.

PC12 and HEK-293T cell lines were obtained from the Cell Resource Center, Peking Union Medical College (PCRC). PC12 cells (CCRID: 1101RAT-PUMC000024, RRID: CVCL_0481) were cultured in RPMI-1640 supplemented with 10% horse serum and 5% fetal bovine serum. HEK-293T cells (CCRID: 1101HUM-PUMC000010, RRID: CVCL_0063) were cultured in high glucose-containing Dulbecco’s modified Eagle’s medium (Gibco, Grand Island, NY, USA) supplemented with 10% fetal bovine serum. Both culture media contained 1% penicillin-streptomycin (Solarbio, Beijing, China). The cells were cultured at 37°C with 5% CO 2 . Both cells have a STR identification report at the time of purchase. PC12 cells were treated with various concentrations (0, 20, 50, 100, 200, 500, and 800 µM) of Sal (Cat# ab120603, Abcam, Cambridge, UK) for 24 hours. Cell viability was measured by the CCK-8 assay. In experiments assessing the induction of autophagy, cells were treated with 100 nM of Bafilomycin A1 (BafA1; Cat# A7823, Applichem, Darmstadt, Germany) and Sal for 6 hours. Verteporfin (Cat# HY-B0146, MCE, Monmouth Junction, NJ, USA) was used as a YAP1 inhibitor. Thirty adult male Sprague-Dawley rats (180–200 g body weight, 8–10 weeks old) (Beijing Vital River, Beijing, China, license No. SCXK (Jing) 2021-0006) were maintained in a pathogen-free environment under a 12-hour light/dark cycle at 24 ± 2°C and relative humidity of 50%–60%, with free access to food and water. Animal experiments were approved by the Animal Management Committee of Beijing Rehabilitation Hospital and the International Association for Animal Research (approval No. 2019bkky037, September 3, 2019). This study complied with all relevant ethical regulations regarding the use of research animals. Rats were housed in three groups (shFTO, shALKBH5, and Scramble; n = 10/group) for 3 days before experimental procedures. The recombinant (r)AAV-shFto, AAV-shAlkbh5 and AAV-scramble vectors were constructed by ViGene Biosciences (Jinan, Shandong Province, China). Animals were placed on a stereotaxic frame (Stoelting, Wood Dale, IL, USA), and the vectors were administered into the right SNpc of rats (coordinates from bregma: A/P –5.3/–6.0 mm; M/L +2.0 mm; D/V –7.2 mm) using an automatic injector (Stoelting) with a vector volume of 1 µL and a flow rate of 0.1 µL/min. The control animals received intracerebral inoculation of rAAV-scramble vectors. After 40 days, the rats were anesthetized by 3% pentobarbital sodium (40 mg/kg; Sigma-Aldrich, St. Louis, MO, USA, Cat# P3761), followed by cervical dislocation. The brains were removed after heart perfusion, and the cerebral cortex, striatum, hippocampus, and midbrain of the rats were separated on ice and stored at –80°C until further analysis. Five rats per group were used for separation of the striatum and midbrain. Total RNA was extracted from PC12 cells or rat brains with the RNAsimple Total RNA Extraction Kit (TIANGEN, Beijing, China), and mRNA was isolated from total RNA using the Dynabeads mRNA DIRECT Micro Purification Kit (Invitrogen, Carlsbad, CA, USA). RNA or mRNA was applied to a nitrocellulose membrane using Bio-Dot equipment (Bio-Rad, Hercules, CA, USA) and tightly attached by the ultraviolet cross-linking assay. The membrane was incubated with rabbit anti-m 6 A polyclonal antibody (1:500, Synaptic Systems, Goettingen, Germany, Cat# 202003, RRID: AB_2279214) at 4°C overnight; methylene blue (Cat# M9140, Sigma-Aldrich) staining served as a loading control. RNA was extracted from PC12 cells treated with 200 µM Sal and control cells. RNA high-throughput sequencing was performed by Cloud-Seq Biotech (Shanghai, China). Briefly, rRNAs were removed from total RNA using the NEBNext rRNA Depletion Kit (New England Biolabs, Inc., Beverly, MA, USA) following the manufacturer’s instructions. RNA libraries were constructed using the NEBNext® Ultra TM II Directional RNA Library Prep Kit (New England Biolabs, Inc.) following the manufacturer’s instructions. The libraries were controlled for quality and quantified using the BioAnalyzer 2100 system (Agilent Technologies, Inc., Santa Clara, CA, USA). Library sequencing was performed using an Illumina HiSeq instrument (Illumina, San Diego, CA, USA) with 150 bp paired-end reads. Paired-end reads were harvested using the Illumina HiSeq 4000 sequencer and quality controlled by Q30. After the 3′ adaptors were trimmed and low-quality reads were removed by the Cutadapt software (v1.9.3; Martin, 2011), the high-quality clean reads were aligned to the reference genome (RN5) with HISAT2 software (v2.0.4; http://ccb.jhu.edu/software/hisat2/index.shtml). HTSeq software (v0.9.1; Anders et al., 2015) was used to obtain the raw count, and edgeR (Robinson et al., 2010) was used to perform normalization. After multiple-test correction using the Benjamini-Hochberg method (Benjamini and Hochberg, 1995), differentially expressed mRNAs were identified. Then, Gene Ontology (GO) and Kyoto Encyclopedia of Genes and Genomes (KEGG) pathway enrichment analyses of differentially expressed mRNAs were performed by Cloud-Seq Biotech. PC12 cells were lysed in RIPA buffer (LABLEAD, Beijing, China) supplemented with 1 mM PMSF for 30 minutes at 4°C. Frozen tissues were thawed and homogenized in RIPA buffer. Cellular or tissue lysates were centrifuged at 12,000 × g to obtain the supernatant. The proteins were separated by sodium dodecyl sulfate–polyacrylamide gel electrophoresis and transferred onto a polyvinylidene fluoride membrane activated by methanol. The membrane was blocked with 5% skimmed milk and incubated with mouse anti-FTO monoclonal antibody (1:1000, Abcam, Cat# ab92821, RRID: AB_10565042), rabbit anti-ALKBH5 monoclonal antibody (1:1000, Abcam, Cat# ab195377, RRID: AB_2827986), rabbit anti-METTL3 monoclonal antibody (1:1000, Abcam, Cat# ab195352, RRID: AB_2721254), mouse anti-YAP1 monoclonal antibody (1:8000, Proteintech, Chicago, IL, USA, Cat# 66900-1-Ig, RRID: AB_2882229), mouse anti-glyceraldehyde 3-phosphate dehydrogenase (GAPDH) monoclonal antibody (1:100,000, Proteintech, Cat# 60004-1-Ig, RRID: AB_2107436), rabbit anti-YTH domain-containing family protein 1 (YTHDF1) polyclonal antibody (1:2000, Proteintech, Cat# 17479-1-AP, RRID: AB_2217473), rabbit anti-YTH domain-containing family protein 2 (YTHDF2) polyclonal antibody (1:5000, Proteintech, Cat# 24744-1-AP, RRID: AB_2687435), mouse anti-YTH domain-containing family protein 3 (YTHDF3) monoclonal antibody (1:1000, Santa Cruz Biotechnology, Texas, CA, USA, Cat# sc-377119, RRID: AB_2687436), rabbit anti-p62 polyclonal antibody (1:1000, MBL, Tokyo, Japan, Cat# PM045, RRID: AB_1279301), rabbit anti-light chain 3 beta (LC3B) polyclonal antibody (1:1000, Cell Signaling Technology, Boston, MA, USA, Cat# 2775, RRID:AB_915950), and rabbit anti-phospho-YAP1 (Ser127) (1:1000, Cell Signaling Technology, Cat# 13008, RRID: AB_2650553) at 4°C overnight. Membranes were then incubated with horseradish peroxidase (HRP)-conjugated goat anti-rabbit secondary antibody (1:5000, ZSBIO, Beijing, China, Cat# ZB-2301, RRID:AB_2747412) or goat anti-mouse secondary antibody (1:5000, ZSBIO, Cat# ZB-2305, RRID:AB_2747415) for 1 hour at 25°C, and bands were visualized using an ECL kit (Merck Millipore, Billerica, MA, USA). The density value of protein bands were determined using the Tanon 4600 image analysis system (Tanon, Shanghai, China). Protein expression was normalized to GAPDH. Each experiment was performed at least three times. Total RNA extracted from PC12 cells was subjected to reverse transcription using the All-in-One First-Strand Synthesis MasterMix (LABLEAD). Quantitative reverse transcription polymerase chain reaction (qRT-PCR) analysis was performed using Taq SYBR® Green qPCR Premix (LABLEAD) following the manufacturer’s instructions. Relative gene expression was calculated using the 2 –ΔΔCt method (Livak and Schmittgen, 2001). GAPDH mRNA served as the internal control. All experiments were conducted at least three times. Primer sequences are listed in Additional Table 1 . The primer sequences analyzed by quantitative reverse transcription polymerase chain reaction Alkbh5: alkB homolog 5; Fto: fat mass and obesity-associated protein; Gapdh: glyceraldehyde 3-phosphate dehydrogenase; Mettl3: methyltransferase-like 3; Yap1: Yes-associated protein 1. The promoter sequences of the rat Fto and Alkbh5 genes were obtained from the NCBI database ( https://www.ncbi.nlm.nih.gov/ ) and retrieved from the rat genome library (Fto promoter: NC_051354.1 , Alkbh5 promoter: NC_051345.1 ). The primer sequences for the Fto and Alkbh5 promoters are listed in Additional Table 2 . The promoter sequences were inserted into the pGL3-Basic plasmid (P0193, Miaolingbio, Wuhan, Hubei Province, China). The constructed plasmid and a reporter plasmid containing the Renilla luciferase gene were co-transfected into PC12 cells using Lipofectamine®3000 transfection reagent (Cat# L3000015, Invitrogen) following the manufacturer’s protocol. At 24 hours after transfection, Sal was added to cells. Cells were collected, and the Dual-Luciferase Reporter Gene Assay Kit (Yeasen, Shanghai, China) was used to detect the luciferase activity. The primer sequences of the promoters of Fto and Alkbh5 Alkbh5: alkB homolog 5; Fto: fat mass and obesity-associated protein. The rat Fto and Alkbh5 coding sequences were inserted into the Pcdh-CMV-MCS-EF1-Puro plasmid (MY1480, MEIYAN Biotechnology, Shanghai, China). The plasmids were transiently transfected into PC12 cells with Lipofectamine®3000 (Invitrogen). Control shRNA (Scramble) and shRNA constructs targeting rat Fto (shFTO) and Alkbh5 (shALKBH5) were inserted into the Plvx-shRNA2-Puro plasmid. Oligonucleotides were purchased from Shanghai GenePharma Co., Ltd. (Shanghai, China). Plasmids were transfected into HEK293T cells along with Plp1, Plp2, and VSVG plasmids, and lentivirus was collected. PC12 cells were infected with lentivirus, and 3 µg/mL puromycin (Cat# P8833, Sigma-Aldrich) was used to select stably transfected cells. siRNAs targeting Yap1 and Ythdf2 and negative controls were synthesized by Shanghai GenePharma Co., Ltd. siRNAs were transfected into PC12 cells using siRNA-mate (GenePharma) following the manufacturer’s instructions. The siRNA sequences are as follows (5ʹ to 3ʹ): negative control, 5ʹ-TTC TCC GAA CGT GTC ACG T-3ʹ; rat Yap1 siRNA #1, #2 and #3, 5ʹ-GGT CAG AGA TAC TTC TT-3ʹ, 5ʹ-GGG CCT CTT CCT GAT GGA T-3ʹ and 5ʹ-GGC AAT ACG GAA TAT CAA T-3ʹ, respectively; and rat Ythdf2 siRNA, 5ʹ-GGT TCT GCA TAG ACT GCA G-3ʹ. RNA methylation immunoprecipitation (MeRIP) was performed using the m 6 A MeRIP kit (Cat# A-17–10499, A&D Technologies, Beijing, China) following the manufacturer’s instructions. RNA (100 µg) of PC12 cells was fragmented and mixed with m 6 A antibody, followed by immunoprecipitation using A/G magnetic beads. A magnetic frame was used to immobilize bead-bound compounds, and the unbound material was removed by washing. RNA was extracted and purified following the manufacturer’s instructions. The SRAMP website ( http://www.cuilab.cn/sramp/ ; Zhou et al., 2016) was used to predict m 6 A modification sites in Yap1 mRNA. qRT-PCR analysis was performed, and the primers are listed in Additional Table 3 . The primer sequences of RNA méthylation immunoprecipitation- quantitative reverse transcription polymerase chain reaction Yapl: Yes-associated protein 1. siYTHDF2-transfected PC12 cells were treated with 5 µg/mL actinomycin D (Cat# A9415, Sigma-Aldrich). After incubation for specific durations, RNA was isolated for qRT-PCR analysis. The half-life (t1/2) of the precursor and mature Yap1 mRNA was calculated, and Gapdh was used for normalization. Data are expressed as the mean ± standard deviation (SD). All measurements were statistically analyzed and significant differences were graphically represented. Data were analyzed by Student’s t -test or one-way analysis of variance followed by Tukey’s multiple comparison test using GraphPad Prism 8 for Windows (GraphPad Software, Boston, MA, USA, www.graphpad.com ). P < 0.05 indicated statistical significance. The sample size was not pre-determined, but the concentrations of proteins extracted from each brain region were quantified.

Several studies have shown that Sal has pro-apoptotic properties and promotes the pathogenesis of PD through its function as a neurotoxin (Voon et al., 2020). Some in vitro studies showed that high and low concentrations of Sal might have neurotoxic and neuroprotective effects, respectively (Możdżeń et al., 2015; Kurnik-Łucka et al., 2020). We thus initially analyzed the cytotoxicity of Sal using the PC12 cell line, which is derived from a rat adrenal medullary tumor (pheochromocytoma) and widely used as a model for studying cell toxicology, neuronal differentiation, and neurodegenerative diseases. After 24 hours of treatment, Sal showed cytotoxicity at concentrations higher than 50 µM ( Figure 1 A ). Sal has cytotoxicity and upregulates m 6 A levels in PC12 cells. (A) Cell viability of PC12 cells treated with different concentrations of Sal (0, 10, 10, 50, 100, 200, 500, and 800 µM) for 24 hours. (B) The effect of different concentrations of Sal on the expression of FTO and ALKBH5 in PC12 cells. (C) The m 6 A levels in total RNA and mRNA in PC12 cells treated with 200 µM Sal for 24 hours. Methylene blue staining was used as a loading control. (D) Quantitative reverse transcription polymerase chain reaction analyses of Fto , Alkbh5 , and Mettl3 mRNA levels in PC12 cells treated with 200 µM Sal. (E) FTO, ALKBH5, and METTL3 protein expressions in PC12 cells treated with 200 µM Sal (electrophoretic gel image is shown in Additional Figure 1 ). (F) The Fto and Alkbh5 promoters were inserted into the pGL3 reporter (left); cells transfected with luciferase reporters and controls were treated with 200 or 500 µM Sal for 24 hours and evaluated with dual-luciferase reporter assays. Data in D–F were normalized to levels in the control group. Data are expressed as the mean ± SD ( n = 3/group). * P < 0.05, ** P < 0.01, *** P < 0.001, vs . Ctrl group (Student’s t -test). ALKBH5: alkB homolog 5; Ctrl: control; FTO: fat mass and obesity-associated protein; m 6 A: N 6 -methyladenosine; METTL3: methyltransferase-like 3; Sal: salsolinol. To examine whether Sal impacted m 6 A levels, we next examined the effect of Sal on m 6 A-related proteins and m 6 A RNA methylation. We found that 200 and 500 µM Sal significantly inhibited the expression of FTO and ALKBH5 ( P < 0.001; Figure 1 B ), but no changes were at observed at concentrations under 50 µM. The effect of Sal on the expression of demethylases was consistent with its inhibitory effect on viability. Quantitative dot-blot analysis was performed to evaluate m 6 A levels. The m 6 A levels in both total RNA and purified mRNA increased considerably after treatment with 200 µM Sal compared with controls ( Figure 1 C ). These results suggested that the increased m 6 A levels in cells treated with Sal may be from the downregulated expression of demethylases. We selected 200 µM Sal for use in subsequent experiments unless stated otherwise. To confirm the effect of Sal on the expression of m 6 A-regulated proteins, we determined the mRNA and protein levels of FTO, ALKBH5, and METTL3 in cells treated with 200 µM Sal. Sal significantly inhibited the mRNA and protein levels of FTO and ALKBH5 and upregulated the expression of METTL3 ( Figure 1 D and E and Additional Figure 1 ). To explore the mechanism by which Sal impacted Fto and Alkbh5 mRNA levels, we conducted a dual-luciferase reporter assay using luciferase reporters driven by the Fto or Alkbh5 promoter ( Figure 1 F ). We found that Sal significantly decreased luciferase activity driven by the reporters, which indicated that Sal inhibited the expression of RNA demethylases by reducing gene transcription. These results confirmed the toxic effect of Sal on PC12 cells. Sal reduced the mRNA and protein expression of demethylases FTO and ALKBH5 by inhibiting their promoters, thereby upregulating the overall m 6 A level in PC12 cells. In addition, Sal also upregulated the mRNA expression of methyltransferase METTL3, which may also be involved in its neurotoxicity. These results revealed a new function of Sal in RNA methylation and provided new insights into the role of Sal in inducing the apoptosis of dopaminergic neurons. To explore the mechanism by which Sal causes neurotoxic effects, we performed RNA-seq in PC12 cells treated with Sal for 24 hours. Quality control assessment showed that the quality of the data was high ( Additional Table 4 ). The results revealed 2405 differentially expressed genes in cells treated with Sal (fold change ≥ 2.0, P -value ≤ 0.05), including 740 upregulated genes and 1665 downregulated genes ( Figure 2 A and B ). In line with our results in PC12 cells, RNA-seq data revealed a significant decrease in Fto after Sal treatment compared with controls; Alkbh5 was also downregulated although the difference was not significant ( Figure 2 C ). RNA quantification and quality assurance by NanoDrop ND-1000 Ctrl: control; Sal: salsolinol. Changes in the gene expression profiles and signaling pathways in PC12 cells treated with Sal. (A) Heatmaps of the relative expression levels of total DEGs between PC12 cells treated with Sal and controls. (B) The results identified 740 upregulated genes and 1665 downregulated genes after Sal treatment. (C) The mRNA levels of m6A-related genes, as determined by RNA sequencing. (D) KEGG analysis of downregulated DEGs in the Sal treatment group. (E) The mRNA levels of genes involved in the Hippo signaling pathway, as determined by RNA sequencing. Data in C and E were normalized by the control group. Data are expressed as the mean ± SD ( n = 3/group). * P < 0.05, *** P < 0.001, vs. Ctrl group (Student’s t -test). ALKBH5: alkB homolog 5; Ctrl: control; DEG: differentially expression gene; FTO: fat mass and obesity-associated protein; LATS1/2: large tumor suppressor homolog 1/2; m 6 A: N 6 -methyladenosine; METTL14: methyltransferase-like 14; METTL3: methyltransferase-like 3; MST1/2: STE20-like serine/threonine kinases1/2; Sal: salsolinol; TAZ: transcriptional coactivator with PDZ-binding motif; WTAP: Wilms tumor 1-associated protein; YAP1: Yes-associated protein 1. As our results showed that Sal increased total m 6 A levels, and m 6 A upregulation has been shown to decrease mRNA stability (Boo and Kim, 2020), therefore, we performed GO enrichment analysis and KEGG pathway analysis of the 1665 downregulated genes identified by RNA-Seq. In the molecular functions of the GO analysis, the function of binding, protein binding, and RNA binding changed most significantly, indicated that m 6 A RNA modification can affect its interaction with the binding protein (further discussion later). KEGG analysis showed that the downregulated differentially expressed genes were mainly enriched in proteoglycans in cancer, regulation of actin cytoskeleton, and the Hippo signaling pathway ( Figure 2 D ). The Hippo signaling pathway is a key determinant of organ size and regulates cell proliferation, differentiation, cell cycle, and apoptosis (Meng et al., 2016; Ma et al., 2019). The Hippo-YAP pathway is a highly conserved kinase cascade in mammals, and its key player is STE20-like serine/threonine kinase 1/2 (MST1/2), which phosphorylates large tumor suppressor homolog 1/2 (LATS1/2) (Ouyang et al., 2020). The interaction between MST1/2 and LATS1/2 leads to the phosphorylation of YAP1 (Kodaka and Hata, 2015). Phosphorylated YAP1 is localized in the cytoplasm, and its continued phosphorylation leads to its degradation. When the Hippo pathway is turned off, non-phosphorylated YAP1 accumulates in the nucleus, where it acts as a transcription cofactor to activate the expression of target genes (Holden and Cunningham, 2018). RNA-Seq data showed significantly decreased levels of Lats1/2 and Yap1 in Sal-treated cells ( Figure 2 E ). To confirm the RNA-Seq results, we examined YAP1 expression in PC12 cells treated with different concentrations of Sal. The mRNA and protein levels of YAP1 were significantly downregulated upon Sal treatment, and this decrease was inversely related to the concentration of Sal ( Figure 3 A and B ). The level of phospho-YAP1 (Ser127) was also downregulated in parallel with the decrease of YAP1. The effect of Sal on YAP1 expression and m 6 A modification of Yap1 mRNA. (A) YAP1 and p-YAP1 expression in PC12 cells treated with different concentrations of Sal. (B) Yap1 mRNA levels in cells treated with Sal. (C) SRAMP website prediction of the m 6 A modification sites in rat Yap1 mRNA. Red indicates very high reliability, and blue indicates high reliability. (D) The 1510 and 1520 sites in rat Yap1 mRNA underwent m 6 A modification; IgG was used as the negative control. (E) The 2156 site in rat Yap1 mRNA underwent m 6 A modification; IgG was used as the negative control. (F) The 2795 site in rat Yap1 mRNA underwent m 6 A modification; IgG was used as the negative control. (G) The m 6 A levels at the 1510 and 1520 sites in rat Yap1 mRNA increased significantly in cells treated with Sal. (H) The m 6 A levels at the 2156 site in the rat Yap1 mRNA increased significantly after Sal induction. (I) The m 6 A levels at the 2795 site in the rat Yap1 mRNA increased significantly after induction by Sal. Data are expressed as the mean ± SD ( n = 3/group). * P < 0.05, ** P < 0.01, *** P < 0.001, vs . Ctrl group (Student’s t -test). Ctrl: Control; GAPDH: glyceraldehyde 3-phosphate dehydrogenase; IP: immunoprecipitation; m 6 A: N 6 -methyladenosine; p-YAP1: phospho-YAP1; Sal: salsolinol; YAP1: Yes-associated protein 1. Our data showed that Sal increased levels of m 6 A, which regulates gene transcription, translation, and RNA degradation. We thus speculated that Sal might regulate the expression of YAP1 by influencing m 6 A RNA modification. Several studies have identified m 6 A-modified sites in mouse and human Yap1 mRNA, and the regulation of YAP1 by m 6 A RNA methylation was found to be associated with the development of various types of human cancer (Jin et al., 2020; Ke et al., 2022). Whether m 6 A-modified sites are present in rat Yap1 mRNA and whether they regulate the function of nervous system are not known. We investigated m 6 A sites in rat Yap1 mRNA using the SRAMP website and found that rat Yap1 mRNA contains eight highly reliable m 6 A sites (other sites were not detected because of their low reliability) ( Figure 3 C ). We next performed MeRIP-qPCR analysis to evaluate m 6 A-modified Yap1 mRNA. The results showed that m 6 A modification only occurred at 1510, 1520, 2156, and 2795 sites ( Figure 3 D–F ). The 1510 and 1520 sites are located in the CDS region near the stop codon, and the 2156 and 2795 sites are located at the 3′UTR. As the 1510 site is close to the 1520 site, only one pair of qPCR primers was used for these sites. MeRIP-qPCR analysis showed that Sal treatment resulted in significantly increased m 6 A modification at 1510, 1520, 2156, and 2795 sites in Yap1 mRNA ( Figure 3 G–I ). These results demonstrated that Sal upregulated m 6 A modification of Yap1 mRNA. To elucidate the mechanism by which Sal stimulates m 6 A modification of Yap1 mRNA, we established PC12 cell lines with stable knockdown of FTO or ALKBH5 (shFTO or shALKBH5) and a scramble control (Scr). Knockdown efficiency was confirmed by immunoblotting and qPCR analyses ( Additional Figure 2 ). We first performed MeRIP-qPCR analysis to determine whether the two demethylases regulated methylation of Yap1 mRNA. m 6 A levels at the 1510, 1520, and 2156 sites in Yap1 mRNA were significantly upregulated in shFTO and shALKBH5 cells ( Figure 4 A–C ), indicating both RNA demethylases affected Yap1 mRNA m 6 A level. YAP1 protein level was downregulated in ALKBH5 knockdown cells but not in FTO knockdown cells ( Figure 4 D and E ). m 6 A RNA methylation regulates YAP1 expression in vitro and in vivo . (A) m 6 A modification at the 1510 and 1520 sites of rat Yap1 mRNA after FTO or ALKBH5 was knocked down. (B) m 6 A modification at the 2156 site of the rat Yap1 mRNA after FTO or ALKBH5 was knocked down. (C) m 6 A modification at the 2795 site of the rat Yap1 mRNA after FTO or ALKBH5 was knocked down. (D) YAP1 protein expression did not change in PC12 cells after knocking down FTO. (E) YAP1 protein expression decreased in PC12 cells after knocking down ALKBH5. (F) YAP1 protein expression remained unchanged In the striatum of the rat brain after FTO was knocked down and decreased after ALKBH5 was knocked down. (G) YAP1 protein expression remained unchanged in the midbrain of the rat after FTO was knocked down and decreased after ALKBH5 was knocked down. Data were normalized by the control group. Data are expressed as the mean ± SD ( n = 3/group). * P < 0.05, ** P < 0.01, **** P < 0.0001, vs. Scr group (Student’s t -test). ALKBH5: alkB homolog 5; FTO: fat mass and obesity-associated protein; GAPDH: glyceraldehyde 3-phosphate dehydrogenase; IP: immunoprecipitation; m 6 A: N 6 -methyladenosine; Scr: scramble; shALKBH5: short hairpin RNA of alkB homolog 5; shFTO: short hairpin RNA of fat mass and obesity-associated protein; YAP1: Yes-associated protein 1. Dopaminergic neurons are mostly located in the SNpc of the midbrain and extend elaborate projections into the striatum. We established FTO-knockdown and ALKBH5-knockdown animal models by stereotactic injection of AAV-9 lentivirus into the SNpc of rats. The regions of the midbrain and the striatum were separated, and all regions showed satisfactory knockdown of FTO or ALKBH5 ( Additional Figures 3 and 4 ). YAP1 expression was decreased in the midbrain and striatum of the shALKBH5 rat brain; however, it did not change in these regions in the shFTO rat brain ( Figure 4 F and G ). These results were consistent with our in vitro results. Our findings suggest that ALKBH5 regulates the expression of YAP1 in the nervous system or dopaminergic neurons by influencing its mRNA m 6 A levels. We next investigated the m 6 A reader protein that may be involved in this regulation. YTHDF2 is the most effective m 6 A reader and decreases the stability of mRNAs by recognizing and localizing m 6 A-containing transcripts to the processing body (Wang et al., 2014). Sal increases the m 6 A levels of Yap1 mRNA, leading to decreased mRNA levels. We speculated that YTHDF2 might function in the degradation of Yap1 mRNA. We found no changes in YTHDF2 in cells treated with Sal ( Figure 5 A ). We overexpressed or knocked down YTHDF2 in PC12 cells and examined YAP1 levels. Overexpression of YTHDF2 decreased the expression of YAP1, while knockdown of YTHDF2 did not alter the expression of YAP1 ( Figure 5 B–E ). To assess whether the decrease in YAP1 expression was from YTHDF2-mediated RNA decay, we measured the mRNA lifetime of Yap1 by inhibiting transcription with actinomycin D. The half-life of Yap1 mRNA increased after YTHDF2 was knocked down ( Figure 5 F ), indicating that YTHDF2 is involved in the degradation of Yap1 mRNA. YTHDF2 induces degradation of Yap1 mRNA in dopaminergic neurons. (A) The effect of different concentrations of Sal on the expression of YTHDF1, YTHDF2, and YTHDF3. (B) Confirmation of YTHDF2 overexpression in PC12 cells. (C) YAP1 protein expression decreased after YTHDF2 was overexpressed. (D) Confirmation of YTHDF2 knockdown in PC12 cells. (E) YAP1 expression was not changed after YTHDF2 was knocked down. (F) NC and siYTHDF2 cells were treated with actinomycin D for 90 minutes, and Yap1 mRNA was analyzed after pre-determined durations. The half-life of Yap1 mRNA increased after YTHDF2 was knocked down. Data were normalized by the NC group. Data are expressed as the mean ± SD ( n = 3/group). ** P < 0.01, *** P < 0.001, **** P < 0.0001, vs . NC or ctrl group (Student’s t -test). Ctrl: Control; GAPDH: glyceraldehyde 3-phosphate dehydrogenase; NC: normal control; oe-YTHDF2: overexpression of YTHDF2; Sal: salsolinol; siYTHDF2: small interfering RNA of YTHDF2; YAP1: Yes-associated protein 1; YTHDF1/2/3: YTH domain-containing family protein 1/2/3. Based on the comparative analysis of the mRNA and protein levels of YAP1 after treatment with 200 µM Sal for 24 hours, we found that the level of Yap1 mRNA decreased by about 15%, whereas the protein level decreased by more than 50% ( Figure 3 A and C ), suggesting that Sal can impact both the stability of the Yap1 mRNA and the efficiency of protein translation. These findings indicated that the regulation of YAP1 expression by Sal might not only be from degradation of Yap1 mRNA mediated by YTHDF2. In addition, we assessed the expression of YTHDF1 and YTHDF3 and found that Sal significantly inhibited YTHDF1 and YTHDF3 protein levels ( Figure 5 A ). While YTHDF2 is associated with mRNA destabilization and degradation, binding of YTHDF1 and YTHDF3 to m 6 A-modified mRNA enhances protein translation (Wang et al., 2015; Zhang et al., 2019). Some studies have shown that YTHDF1 and YTHDF3 cancombine with m 6 A in Yap1 mRNA topromote translation, through a mechanism requiring the translation factor Eif3a/b (Jin et al., 2020; Li et al., 2022). Therefore, Sal-induced decrease in the expression of YTHDF1 and YTHDF3 might play a negative role in the translation of Yap1 . We speculated that mRNA translation mediated by YTHDF1/YTHDF3 and mRNA degradation mediated by YTHDF2 may influence the expression of YAP1 in dopaminergic neurons. Research has shown that YAP1 regulates the level of autophagy, and the YAP-autophagy signal strongly influences the development of several diseases, such as neurodegenerative disease, pancreatic cancer, endometriosis, and diabetic kidney disease (Sun et al., 2021; Pei et al., 2022; Yin et al., 2022). m 6 A RNA modification regulates the transcription, translation, and degradation of many autophagy-related genes, including Atg5 , Atg7 , and CAMKK2 genes (Wang et al., 2020; Chen et al., 2021c). We speculated that the upregulation of the m 6 A level in dopaminergic neurons by Sal might lead to changes in autophagy. Therefore, we evaluated autophagy in dopaminergic neurons after treatment with Sal. The autophagy-related protein LC3B strongly influences the formation of autophagosomes. When autophagy is activated, LC3-I is converted into LC3-II, which accelerates the formation of autophagosomes; LC3-II is considered to be a marker of autophagy induction (Kabeya et al., 2000). Sequestosome 1 (p62/SQSTM 1) acts as a “cargo protein” of autophagy and is degraded during the autophagic flux. Its accumulation indicates the inhibition of autophagy (Ploumi et al., 2021). We found that p62 expression was downregulated with increased concentrations of Sal, while LC3-I and LC3-II levels significantly increased ( Figure 6 A ). These results indicated that autophagy flux increased in cells treated with Sal. p62 protein is degraded by recruiting “waste” to the autophagosome or by the autophagy lysosomal pathway. We next used the lysosome inhibitor BafA1, which blocks autophagy, to examine the effect of Sal on LC3-II and p62. The results showed high levels of LC3-II accumulated when Sal was combined with BafA1 ( Figure 6 B ). Autophagy regulated by Sal is partly related to the inhibition of YAP1. (A) Sal treatment decreased the expression of p62 and increased LC3. (B) Autophagy-related proteins were examined in cells treated with Sal alone or in combination with BafA1 treatment. (C) In PC12 cells with YAP1 knockdown, the levels of p62 decreased and LC3 increased. Data are expressed as the mean ± SD ( n = 3/group). * P < 0.05, *** P < 0.001, vs. NC group (Student’s t -test). (D) In PC12 cells treated with Verteporfin, the levels of p62 decreased and LC3 increased. (E) A schematic model in which Sal acts as an RNA m 6 A methylation inducer, mediating dopaminergic neuronal death by regulating YAP1 and autophagy. ALKBH5: alkB homolog 5; BafA1: Bafilomycin A1; FTO: fat mass and obesity-associated protein; GAPDH: glyceraldehyde 3-phosphate dehydrogenase; LC3: light chain 3; m 6 A: N 6 -methyladenosine; METTL3: methyltransferase-like 3; NC: normal control; p62: Sequestosome 1; Sal: salsolinol; siYAP1: small interfering RNA of YAP1; YAP1: Yes-associated protein 1; YTHDF2: YTH domain-containing family protein 2. To explore whether YAP1 is involved in autophagy in PC12 cells, we downregulated YAP1 using siRNA and found that the level of p62 decreased and LC3-I accumulated ( Figure 6 C ). We also assessed autophagy after treatment with Verteporfin, an inhibitor of YAP1, and the results were the same as those obtained after YAP1 was knocked down ( Figure 6 D ). In conclusion, the effect of YAP1 on autophagy was consistent with the effect of Sal on autophagy, which indicated that autophagy regulated by Sal is related to the inhibition of YAP1, but it did not depend on YAP1 inhibition.

Sal, a type of TIQ or catechol isoquinoline, is present in food sources, such as beef, port wine, and fruit, and can also be formed endogenously from dopamine and aldehydes. Most studies on Sal have focused on the risk of PD and alcohol addiction because of its association with the abnormal metabolism of dopamine and acetaldehyde (Hipólito et al., 2012; Kurnik-Łucka et al., 2018). Sal inhibits catecholamine uptake in brain synaptosomes and inhibits the activity of monoamine oxidase, catechol-O-methyltransferase, and tyrosine hydroxylase in the striatum, which participate in dopamine synthesis and metabolism (Kurnik-Łucka et al., 2018). Sal also regulates the release of prolactin, which alters the synthesis and distribution of dopamine (Székács et al., 2007; Oláh et al., 2011). Most studies have shown that Sal exhibits pro-apoptotic properties, as it induces the apoptosis of dopaminergic neurons, while others have suggested that Sal might have neuroprotective properties at lower concentrations (Kurnik-Łucka et al., 2018, 2020). However, the neuro-modulatory role of Sal is poorly understood. m 6 A RNA methylation plays a key role in brain development, neuronal signaling, and neurological disorders (Sokpor et al., 2021; Zhang et al., 2022). A previous study showed that global m 6 A modification of mRNAs is downregulated in dopaminergic cells and the striatum of the 6-OHDA-induced PD rat model (Chen et al., 2019). Many studies have investigated the role of m 6 A modification in the pathogenesis of PD using 1-methyl-4-phenyl-1,2,3,6-tetrahydropyridine, paraquat, and manganese as PD model inducers (Chen et al., 2021a; Qi et al., 2022; Su et al., 2022; Yu et al., 2022). Some studies have examined the relationship between m 6 A RNA methylation and dopamine transmission and found that FTO-mediated hypermethylation is related to dopamine deficiency (Hess et al., 2013). Sal is present at a higher concentration than other TIQs in regions of the brain with high levels of dopamine synthesis, such as the ventral midbrain and the striatum (DeCuypere et al., 2008). In this study, we found that Sal at concentrations higher than 50 µM was cytotoxic and significantly increased the global m 6 A RNA modification levels. We speculated that m 6 A RNA methylation induced by Sal might regulate dopaminergic neurotoxicity and participate in the morbidity of PD. In this study, we found that hypomethylation of RNA was induced by the downregulation of the expression of FTO and ALKBH5 after treatment with Sal. Dual-luciferase reporter assays showed that Sal reduced the transcriptional activity of the promoters of both these demethylases. Studies have shown that FTO and ALKBH5 regulate metabolic disorders and participate in the regulation of different diseases, including cancer, neurodegenerative disease, diabetes, and obesity (Jiang et al., 2021). Therefore, FTO and ALKBH5 may be potential therapeutic targets for disease. Some small-molecule FTO inhibitors were shown to exhibit strong anti-tumor effects in various types of cancer (Huang et al., 2019). CS1 and CS2 specifically target FTO and efficiently suppress its demethylase activity; these compounds significantly inhibit the viability of human acute myelocytic leukemia cells (Su et al., 2020). Entacapone, an inhibitor of catechol-O-methyltransferase that is used to treat PD, is an FTO chemical inhibitor that upregulate the expression of FOXO1. It has been used for treating metabolic disorders, such as diabetes and obesity (Peng et al., 2019). Our results suggest Sal may be a potential inhibitor of FTO and ALKBH5; research into its potential regulatory effects on metabolic disorders is warranted. RNA-Seq analysis showed that the Hippo-YAP pathway was suppressed in PC12 cells treated with Sal. We further found that Sal strongly inhibited the expression of YAP1 in a concentration-dependent manner. YAP1 is a downstream effector of the Hippo signaling pathway, which regulates the loss of dopaminergic neurons. Studies showed that YAP1 inhibits the synthesis of the cell death–related protein PTEN by inducing the expression of miR-130a, which can prevent the loss of mesencephalic dopaminergic neurons (Zhang et al., 2017). YAP1 is also the regulatory target of m 6 A RNA methylation and strongly upregulates or downregulates the progression of cancer (Jin et al., 2020; Cui et al., 2021). We found that Sal upregulated m 6 A methylation of Yap1 mRNA and YTHDF2 mediated degradation of Yap1 mRNA, reducing YAP1 expression. Moreover, ALKBH5, but not FTO, regulated m 6 A modification of Yap1 mRNA. ALKBH5 directly removes the methyl group from m 6 A methylated adenosine, and FTO converts m 6 A via oxidation to N 6 -formyladenosine, which is eventually hydrolyzed to adenine (Chen and Wong, 2020). FTO preferentially demethylates m 6 Am (another reversible RNA modification) instead of m 6 A (Relier et al., 2021). Our in vitro and in vivo experiments showed that knocking down FTO did not affect the expression of YAP1, which might be because of the different mode of demethylation of FTO. Many studies on oncology have shown that YAP1 is involved in the regulation of autophagy signals, whereas abnormalities in autophagy also regulate YAP through autophagic degradation (Sun et al., 2021; Pei et al., 2022; Yin et al., 2022). The YAP-autophagy axis may be a new therapeutic strategy for treating cancer. However, its role in the nervous system is poorly understood. We found that the downregulation of YAP1 induced by Sal can cause dysregulation of autophagy in dopaminergic neurons. These results suggested that YAP-autophagy is an important part of Sal neurotoxicity. Additionally, the expression of several autophagy-related genes (Atg), such as Atg5 and Atg7, is regulated by m 6 A methylation (Wang et al., 2020). Autophagy impairment was shown to lead to the accumulation of α-synuclein and degeneration of dopaminergic neurons, two major features of PD (Xilouri et al., 2016; Hou et al., 2020). We speculate that hypomethylation of RNA after treatment with Sal might also influence autophagy and this should be explored in future studies. Increased autophagy levels by Sal are not solely related to the inhibition of YAP1, but also regulated by other mechanisms. Therefore, further studies in animal models are necessary to confirm the role of Sal in inducing m 6 A RNA methylation. Given that m 6 A RNA methylation has been reported to be closely linked to autophagy, we speculate that Sal may regulate autophagy through other m 6 A-related pathways beyond YAP1 inhibition. In conclusion, we elucidated the effect of Sal on m 6 A RNA methylation and found that Sal acts as an m 6 A inducer in dopaminergic neurons by inhibiting the expression of FTO and ALKBH5. We also identified the m 6 A-YAP-autophagy axis, which may play a key role in neurotoxicity mediated by Sal. We found that Sal downregulated the expression of YAP1 through ALKBH5 and the m 6 A recognition protein YTHDF2. Knocking down YAP1 dysregulated autophagy, which may play a role in its neurotoxic effects ( Figure 6 E ). Our study revealed the function of Sal in m 6 A RNA methylation. These findings might provide a new strategy for investigating the neurotoxic effects of catechol isoquinolines and may be a reference for further determining the role of RNA methylation in the pathogenesis of PD.

Additional Figure 1 : The effect of 200 μM Sal on the expression of m6A-regulated proteins. The effect of 200 μM Sal on the expression of m 6 A-regulated proteins. (A) The expression of FTO. (B) The expression of ALKBH5. (C) The expression of METTL3. ALKBH5: alkB homolog 5; Ctrl: control; FTO: fat mass and obesity-associated protein; Gapdh: glyceraldehyde 3-phosphate dehydrogenase; m6A: N6-methyladenosine; Mettl3: methyltransferase-like 3; Sal: salsolinol. Additional Figure 2 : Knockdown efficiency of FTO or ALKBH5 in PC12 cells. Knockdown efficiency of FTO or ALKBH5 in PC12 cells. (A) Knockdown efficiency of FTO in PC12 cells. (B) Knockdown efficiency of ALKBH5 in PC12 cells. (C) Fto mRNA level was decreased after knockdown of Fto in PC12 cells. (D) Alkbh5 mRNA level was decreased after knockdown of Alkbh5 in PC12 cells. Data were normalized by control group. The data are expressed as the mean ± SD ( n = 3/group). * P < 0.05, *** P < 0.001, vs . Scr group (Student’s t -test). ALKBH5: alkB homolog 5; FTO: fat mass and obesity-associated protein; GAPDH: glyceraldehyde 3-phosphate dehydrogenase; Scr: scramble shALKBH5: short hairpin RNA of alkB homolog 5; shFTO: short hairpin RNA of fat mass and obesity-associated protein. Additional Figure 3 : Expression of FTO or ALKBH5 in the striatum is decreased after FTO or ALKBH5 knockdown in the SNpc of rat brain. Expression of FTO or ALKBH5 in the striatum is decreased after FTO or ALKBH5 knockdown in the SNpc of rat brain. (A) Expression of FTO in striatum was decreased after it was knockdown in SNpc of rat brain. (B) Expression of ALKBH5 in striatum was decreased after it was knockdown in SNpc of rat brain. Data were normalized by Scr group. The data are expressed as the mean ± SD ( n = 3/group). ** P < 0.01, *** P < 0.001, vs . Scr group (Student’s t -test). ALKBH5: alkB homolog 5; FTO: fat mass and obesity-associated protein; GAPDH: glyceraldehyde 3-phosphate dehydrogenase; Scr: scramble; shALKBH5: short hairpin RNA of alkB homolog 5; shFTO: short hairpin RNA of fat mass and obesity-associated protein; SNpc: substantia nigra pars compacta Additional Figure 4 : Expression of FTO or ALKBH5 in the midbrain is decreased after FTO or ALKBH5 knockdown in the SNpc of rat brain. Expression of FTO or ALKBH5 in the midbrain is decreased after FTO or ALKBH5 knockdown in the SNpc of rat brain. (A) Expression of FTO in midbrain was decreased after it was knockdown in SNpc of rat brain. (B) Expression of ALKBH5 in midbrain was decreased after it was knockdown in SNpc of rat brain. Data were normalized by Scr group. The data are expressed as the mean ± SD ( n = 3/group). * P < 0.05, *** P < 0.001, vs . Scr group (Student’s t -test). ALKBH5: alkB homolog 5; FTO: fat mass and obesity-associated protein; GAPDH: glyceraldehyde 3-phosphate dehydrogenase; Scr: scramble; shALKBH5: short hairpin RNA of alkB homolog 5; shFTO: short hairpin RNA of fat mass and obesity-associated protein; SNpc: substantia nigra pars compacta. Additional Table 1 : The primer Sequences analyzed by quantitative reverse transcription polymerase chain reaction. Additional Table 2 : The primer sequences of the promoters of Fto and Alkbh5. Additional Table 3 : The primer sequences of RNA methylation immunoprecipitation- quantitative reverse transcription polymerase chain reaction. Additional Table 4 : RNA quantification and quality assurance by NanoDrop ND-1000.