Peonidin-3-O-Glucoside from Purple Corncob Ameliorates Nonalcoholic Fatty Liver Disease by Regulating Mitochondrial and Lysosome Functions to Reduce Oxidative Stress and Inflammation

A frequent chronic liver condition across the world is nonalcoholic fatty liver disease (NAFLD). Oxidative stress caused by lipid accumulation is generally considered to be the main cause of NAFLD. Anthocyanins can effectively inhibit the production of reactive oxygen species and improve oxidative stress. In this work, six major anthocyanins were separated from purple corncob by semi-preparative liquid chromatography. The effects of the 6 kinds of anthocyanins against NAFLD were investigated using a free fatty acid (FFA)-induced cell model. The results showed that peonidin 3-O-glucoside (P3G) can significantly reduce lipid accumulation in the NAFLD cell model. The treatment with P3G also inhibited oxidative stress via inhibiting the excessive production of reactive oxygen species and superoxide anion, increasing glutathione levels, and enhancing the activities of SOD, GPX, and CAT. Further studies unveiled that treatment with P3G not only alleviated inflammation but also improved the depletion of mitochondrial content and damage of the mitochondrial electron transfer chain developed concomitantly in the cell model. P3G upregulated transcription factor EB (TFEB)-mediated lysosomal function and activated the peroxisome proliferator-activated receptor alpha (PPARα)-mediated peroxisomal lipid oxidation by interacting with PPARα possibly. Overall, this study added to our understanding of the protective effects of purple corn anthocyanins against NAFLD and offered suggestions for developing functional foods containing these anthocyanins.


Introduction
Nonalcoholic fatty liver disease (NAFLD) is the most common chronic liver disease in the world, and its global prevalence has doubled in the past 20 years. In humans, NAFLD is characterized by excessive accumulation of lipids in the liver, which is a common complication of obesity and is related to insulin resistance and metabolic syndrome [1]. NAFLD represents a series of progressive stages of liver disease, from simple steatosis to nonalcoholic steatohepatitis that manifests as liver inflammation, necrosis, and fibrosis [2,3].
Abnormal accumulation of saturated fats such as triglycerides in the liver has been reported to be associated with the pathogenesis of NAFLD [3]. This abnormal accumulation can lead to mitochondrial dysfunction and overproduction of ROS [4]. Excessive mitochondrial ROS exacerbates mitochondrial dysfunction and can also activate NF-κB and promote tumor necrosis factor α (TNF-α), interleukin (IL)-1β, and IL-6 production and induce liver inflammation and the development of nonalcoholic fatty liver [3][4][5][6][7].
Corn originates in the American continent and is considered one of the main food sources in the world. Corn contains many bioactive compounds that provide ideal health

Establishment of NAFLD Cell Model
The normal human liver cell line L02 was purchased from the Type Culture Collection of the Chinese Academy of Sciences (Shanghai, China) and was cultivated in a humidified incubator at a constant temperature of 37 • C with 5% CO 2 . Ten percent fetal bovine serum, 100 units/mL penicillin, and 100 units/mL streptomycin were added to RPMI 1640 medium (Gibco). We referred to the method of Yang Xu et al. of treating L02 cells with 0.5 mM FFA to establish a NAFLD model [27]. Palmitic acid (PA) was dissolved in 0.1 M NaOH at 70 • C to obtain a 100 mM PA solution. Mix 100 mM PA solution with 100 mM oleic acid (OA) in a volume of OA/PA = 2:1. The mixture was then mixed in equal amounts with 10% BSA to obtain a 50 mM FFA stock solution. When building the NAFLD cell model, dilute the FFA stock solution to 0.5 mM with cell culture medium. The NAFLD cell model was obtained after cells were cultured for 24 h in culture medium containing 0.5 mM FFA.

The Effect of Purple Corn Anthocyanin on Lipid Accumulation in Hepatocytes
As previously mentioned [28], the Nile Red staining procedure was used. L02 cells were seeded in a 6-well plate at a density of 1 × 10 5 cells per well and cultivated for 24 h before being exposed to 0.5 mM FFA and treated with various concentrations of purple corncob anthocyanins (10,100,200, and 300 µM). The cells were then rinsed twice with PBS, treated with Nile Red (0.1 M) for 30 min at 37 • C, and observed using an IX71 inverted fluorescence microscope (Olympus, Tokyo, Japan). Only cells treated with FFA served as the FFA-treated group; the nontreated group served as the control group.
Oil Red O staining was carried out using a previously published procedure with slight modifications [29]. After the L02 cells had been washed twice with PBS and fixed for 30 min at room temperature with 4% paraformaldehyde, the supernatant was discarded. The cells were then stained with Cole's hematoxylin solution (Solarbio Life Science CO., Beijing, China) for 5 min after being dyed with 0.6% (w/v) Oil Red O for 30 min at room temperature.
L02 cells were seeded and treated in a 6-well plate as described before. After 24 h of exposure, the cells were washed twice with 2 mL of PBS. The cells were then scraped and resuspended in PBS before being centrifuged to remove the PBS. Isopropyl alcohol was used to extract the triglycerides from the precipitated cells. The extract was centrifuged at 12,000 rpm for 10 min at 4 • C to collect the supernatant. The triglyceride content of the supernatant was determined using an assay kit (BC0625) purchased from Beijing Solbio Technology Co., Ltd. in China. The absorbance at 420 nm was measured with a Synergy HTX microplate reader (Agilent, Santa Clara, CA, USA).

Hepatocyte Viability
L02 cells were seeded into a 96-well plate (5 × 10 3 cells per well) and cultivated for 24 h. They were then exposed to 0.5 mM FFA for 24 h, followed by 24 h of exposure to a variety of concentrations of purple corncob anthocyanins (10,100,200, and 300 µM). A Infinite M200 micro-plate meter (TECAN, Männedorf, Switzerland) was used to evaluate the cell viability rate after adding 0.5 mg/mL of MTT into each well. The cell viability rate using a microplate meter (Tecan Infinite M200, Männedorf, Switzerland) at 490 nm.

Hepatocyte Damage Detection
We tested alanine aminotransferase (ALT) and aspartate aminotransferase (AST) to investigate the damaging effect of FFA on hepatocytes and the effect of purple corncob anthocyanins [30]. L02 cells were seeded in a 6-well plate (1 × 10 5 . L02 cells were treated with FFA and different concentrations of purple corncob anthocyanins. After that, the cells were collected and washed 2 times with 0.1 mol/L phosphate buffer. The cells were then centrifuged at 1000 rpm for 10 min and the cell precipitate was collected. To the cell precipitate, 0.1 mol/L phosphate buffer was added (500 µL), and the cells were broken by ultrasound under ice water bath conditions (ultrasound power was 450 W, working 5S, pause 30S, 10 cycles). The prepared homogenate was applied to a BCA protein concentration assay kit (P0010S, Beyotime Biotechnology, Shanghai, China) to determine the protein concentration. ALT and AST activities were measured according to the aspartate and alanine aminotransferase assay kit instructions. The results are reported in units/g of protein. −• , and GSH levels were measured using a ROS kit (DCF was used as the fluorescence probe), dihydroethidium (DHE), and naphtha-lene-2,3-dicar-acetaldehyde (NDA) fluorescent dyes [31][32][33]. The cells were washed twice with PBS after the treatment, and the supernatant was discarded. After that, three groups of L02 cells were stained with ROS kit (20 µM DCF probe), DHE (10 µg/mL), and NDA (50 µM), and kept at 37 • C for 30 min. After that, the stained cells were analyzed using the fluorescence microscope and FACS Canto II flow cytometry (FCM) (BD, Franklin Lakes, NJ, USA). In FCM analysis, the DCF fluorescence was excited by the laser beam (488 nm) and recorded at the FITC channel. The DHE fluorescence was determined at an excitation wavelength of 535 nm and emission of 610 nm by flow cytometry (FCM). The NDA fluorescence was recorded at the FITC channel.

SOD, GPx and CAT Activity Assay
L02 cells were collected with a scraper and washed once with PBS. Cell samples were lysed with cell lysis buffer and centrifuged at 12,000× g for 10 min at 4 • C. The supernatant was used to determine the enzyme activities of SOD, GPx, and CAT according to the (S0109 total superoxide dismutase assay kit with NBT, S0058 total glutathione peroxidase assay kit, S0051 catalase assay kit; three kits were purchased from Beyotime Biotechnology, Shanghai, China) kit instructions.

The Effect of P3G on Hepatocyte Mitochondria
According to the method described by Chen et al., the mitochondrial membrane potential and mitochondrial mass were detected using the Mitochondrial Membrane Potential Detection Kit and MitoTracker Green [34,35]. L02 cells were seeded in a 6-well plate and treated as previously described. Cells were washed twice with 2 mL of PBS after 24 h of exposure. The specific steps were carried out according to the kit instructions, and the concentration of MitoTracker Green in the working solution was 100 nM. The mitochondrial membrane potential was evaluated by fluorescence microscopic imaging and FCM. The cell was stained with JC-1 for 20 min at 37 • C after 24 h of P3G treatment. In FCM analysis, fluorescence was read at 488 nm excitation and 530 nm emission for green and at 540 nm excitation and 590 nm emission for red. The ratio of aggregated JC-1 (red fluorescence) and monomeric JC-1 (green fluorescence) represented the mitochondrial membrane potential. The ATP level in L02 cells was also analyzed. L02 cells were seeded in a 6-well plate and treated as previously described. Following PBS aspiration, the lysate was added at a rate of 200 µL per well, and cells were lysed by repeated blowing with a pipette for 2 min. The lysed cell suspension was centrifuged for 5 min at 4 • C at 12,000 g. The assay of the ATP level in supernatant was carried out as directed by the instruction (S0026, Beyotime Biotechnology, Shanghai, China).

The Effect of P3G on Lysosome Function
As described previously, L02 cells were treated with FFA and different concentrations of purple corncob anthocyanins. In order to detect P3G's effect on lysosome function, the cells were stained with Lyso-Tracker Green's 50 nM for 30 min at 37 • C in the dark. After staining, the cells were visualized by an inverted fluorescence microscope. The results were uttered as the mean fluorescence intensity [36].

Western Blot
The cells were washed two times with PBS, gathered by a cell scraper, and lysed on ice utilizing RIPA cell lysis buffer for 30 min. The cell lysate was centrifuged at 13,000 rpm for 15 min at 4 • C. The supernatant was collected, and the protein concentration was determined. Then, equal amounts of 30 µg protein from different samples were transferred to polyvinylidene fluoride (PVDF) membranes by 10% SDS-PAGE. The membrane blocking was accomplished by using 5% skimmed milk at room temperature for 1 h. The following antibodies were used: TFEB Rabbit Polyclonal Antibody (CAT No. AF8130), Lamin B1 Rabbit Monoclonal Antibody (CAT No. AF1408), LAMP1 Rabbit Polyclonal Antibody (CAT No. AF7353), Caspase-1 Rabbit Monoclonal Antibody (CAT No. AF1681), NF-κB p65 Rabbit Polyclonal antibody (CAT No. AN365) GAPDH Rabbit Monoclonal Antibody (CAT No. AF1186) (File S1) [29].

The Effect of P3G on Hepatocyte Inflammation
L02 cells were planted in 6-well plates at a density of 2 × 10 5 cells per well and incubated for 24 h. Following 0.5 mM FFA treatment, L02 cells were treated for 24 h with P3G at varied concentrations (100, 200, and 300 µM). The cell-free supernatant was obtained following P3G stimulation for 24 h by centrifuging for 5 min at 500× g. The ELISA method was employed to detect IL-1, IL-6, and TNF-in the supernatant according to the kit instructions (Beyotime Biotechnology, Shanghai, China). Briefly, the supernatants without dilution and cytokine standards were applied to 96-well plates covered with monoclonal antibody (coupled to HRP) and incubated at 37 • C for 30 min. The plates were then cleaned and incubated with the substrate solution for 20 min at 37 • C. The stop solution was then added, and absorbance at 450 nm was measured with a Synergy HTX microplate reader (Agilent, Santa Clara, CA, USA). The results were plotted against the standard curve's linear component.

Molecular Docking
The in vitro studies of P3G against PPARα were confirmed by computational docking studies using AutoDock Vina software [37]. From the PDB database, the PPARα's 3D-structure program database (PDB) format file (PDB ID: 2ZNN) was obtained. Before docking simulations, the protein's energy was minimized using the UCSF Chimera program [38]. The PyMOL software suite was used to remove the hetero atoms and water molecules [39]. Dehydrogenation, calculation of Gasteiger charges, and fusion of nonpolar hydrogen atoms were all done using the Autodock tools [40]. Finally, the PDBQT format, the only one supported by AutoDock Vina, was used to save the protein and ligand. To ensure that the active site is completely covered and that ligands can move freely, a grid box with the dimensions 29 × 29 × 29 Å was established and centered at the coordinates X = 2.91, Y = 4.87, and Z = 4.66 [41]. The number of maximum binding modes was set to 50. The docking results were visualized and analyzed using PyMOL.

Molecular Dynamics Study
To gather more detailed binding information, GROMACS 2020.4 was utilized for molecular dynamics (MD) simulation. Using the acpype program, the GAFF force field was used to pre-treat P3G [41]. The amber14ffSB force field was added to charge αglucosidase [42]. ACPYPE software was used to create the ligand topology [43]. Complexes were soaked in a cubic water box that was filled with water molecules; the distance between the complexes and the border of the box was 1 nm. The complexes' surface charges were reduced by adding 4 Na + . Energy minimization was conducted. The minimized system was subjected to MD that utilized an NPT ensemble after being equilibrated by NVT at 1 bar and 300 K for 100 ps. The MD simulation was then set to run for 50 ns [44]. Root mean square deviation (RMSD), solvent accessible surface area (SASA), and hydrogen bond analyses were performed on the track and structure using VMD.

qRT-PCR Analysis
The expression levels of PPAR, CPT-1A, ACOX1, and actin were determined using quantitative real-time PCR testing. L02 cells were collected with a scraper from the well of the 6-well plate and washed once with PBS. Cells were lysed with 0.5 mL Trizol (Beyotime Biotechnology, Shanghai, China) and reverse transcribed to cDNA using the Prime Script RT reagent Kit (TaKaRa, Shiga, Japan) according to the manufacturer's procedure. The primers and amplification conditions were the same as described [45]. Quantitative real-time PCR was undertaken on a CFX96 real-time PCR system (Bio-Rad Laboratories, Inc., Hercules, CA, USA), and the temperature conditions were 95 • C for 5 min, followed by 95 • C for 15 s and

Statistical Analysis
Every experiment was run at least three times, and the results were expressed as the mean ± standard deviation (SD). The fluorescence intensity and Oil Red O colored area of the photo-micrograph were analyzed using Image Pro Plus 6.0 (Media Cybernetics, Inc., Rockville, MD, USA), and the findings indicated the average fluorescence intensity and the percentage of Oil Red O colored area. One-way analysis of variance (ANOVA) was used to determine the significance between two groups, and the findings were expressed as a comparison to the results obtained for Duncan's multiple range test or Student's t-test using GraphPad Prism version 8.0.0 (GraphPad Software, San Diego, CA, USA). A p-value less than 0.05 was regarded as significant.

P3G Ameliorates Hepatocyte Injury induced by FFA
We found that FFA caused a significant increase in AST and ALT in L02 cells from 2.08 ± 0.05 units/g and 0.38 ± 0.14 units/g in the control group to 15.19 ± 0.20 and 8.13 ± 0.24 units/g, respectively. However, the P3G processing offsets the increase in AST and ALT levels ( Figure 3E

P3G Ameliorates Hepatocyte Injury Induced by FFA
We found that FFA caused a significant increase in AST and ALT in L02 cells from 2.08 ± 0.05 units/g and 0.38 ± 0.14 units/g in the control group to 15.19 ± 0.20 and 8.13 ± 0.24 units/g, respectively. However, the P3G processing offsets the increase in AST and ALT levels ( Figure 3E Eliminating ROS is considered to be an effective way to alleviate NAFLD [46,47]. As shown in Figure 4A,D, 0.5 mM FFA induced a significant overproduction of ROS in L02 cells. However, P3G generated the best results at a concentration of 300 µM, and the fluorescence intensity was decreased to 31.25 ± 11.78% (compared with the FFA-treated group, p > 0.05) as a result of its considerable inhibitory action on the rise of ROS induced by FFA. Figure 4G shows that after FFA treatment, the percentage of DCF-positive cells increased to 82.0% compared to the control group (24.6%). After different concentrations of P3G intervention, the percentage of DCF-positive cells decreased significantly. The proportion of new cells that were DCF-positive decreased to 30.8% after 300 µM P3G treatment ( Figure 4G). The flow cytometry results were consistent with the fluorescence imaging statistical results.
Similar results were obtained from the observation of O 2 −• . As shown in Figure 4B,E, compared to the control group, the O 2 −• level significantly increased after FFA (317.75 ± 6.47%). Flow cytometry analysis revealed that the percentage of DHE-positive cells increased to 19.5% after FFA treatment compared to 0.28% in the control group ( Figure 4H). After treatment with different concentrations of P3G, the percentage of DHEpositive cells was significantly reduced in a dose-dependent manner. Treatment with 300 µM P3G reduced the percentage of DCF-positive neoplastic cells to 3.62%. O 2 −• production was dose-dependently decreased by P3G administration.
GSH is a significant endogenous nonenzymatic antioxidant that can scavenge ROS. The changes in GSH content in L02 cells were detected by NDA staining. As shown in Figure 4C,F, the GSH concentration in the FFA treatment group dropped to 84.67 ± 3.60% compared to the control group. P3G significantly increased the GSH concentration in cells ( Figure 4C

P3G Strengthened the Mitochondrial Function
Cell aging and apoptosis may result from the aberrant energy metabolism and mitochondrial structure caused by oxidative stress [48]. In this work, the mitochondrial membrane potential and mitochondrial mass of L02 cells were measured using JC-1 and MitoTracker Green, respectively ( Figure 5). Figure 5A,D demonstrates that 0.5 mM FFA decreases the JC-1 fluorescence fusion result from 100% in the control group to 68.36 ± 6.63% in the experimental group. In a dosedependent manner, P3G restored the decline in mitochondrial membrane potential. The mitochondrial membrane potential rose with an increase in P3G concentration, reaching values of 71.06 ± 4.27% (10 µM), 77.08 ± 6.71% (100 µM), 91.91 ± 4.85% (200 µM), and 96.12 ± 7.66% (300 µM). The results of the flow cytometry analysis showed that the FFA group had a significantly lower red/green ratio than the control group, while P3G reversed this trend ( Figure 5A Figure 5B,E). The findings demonstrate that P3G can promote mitochondrial dysfunction caused by FFA. According to Figure S8, the ATP level in the FFA-treated group was significantly lower than that in the control group (p < 0.01). When compared to the FFAtreated group, intervention with different doses of P3G significantly reversed ATP levels in FFA-exposed L02 cells (p < 0.05).   and CAT.

P3G Strengthened the Mitochondrial Function
Cell aging and apoptosis may result from the aberrant energy metabolism and mitochondrial structure caused by oxidative stress [48]. In this work, the mitochondrial membrane potential and mitochondrial mass of L02 cells were measured using JC-1 and Mito-Tracker Green, respectively ( Figure 5).

P3G Enhances the Lysosomal Function
In rodents with NAFLD, the lysosomal function was inhibited [49]. The fluorescence intensity of LysoTracker in the FFA-treated group reduced to 87.99 ± 1.03% compared to the control, as illustrated in Figure 5C Lysosome-associated membrane protein-1 (LAMP1) is one of the main protein components on the lysosomal membrane [50,51]. The LAMP1 level in the FFA-treated L02 cells reduced (0.67 ± 0.09) in comparison to the control group, according to the Western blot results in Figure 5G,H, but P3G considerably increased the LAMP1 level. Following treatment at different concentrations of P3G, the levels of LAMP1 increased to 0.81 ± 0.06 (10 µM), 1.15 ± 0.18 (100 µM), and 1.44 ± 0.26 (200 µM).
The majority of lysosomal hydrolases contain cathepsin D, which is a member of the cathepsin protease family [52]. Studies have found that decreased liver cathepsin D levels in patients with NFALD will affect lysosomal acidification, resulting in the accumulation of lipids. The findings of the verification of the cathepsin D expression level are displayed in Figure 5G,J. The results show that the expression level of cathepsin D increased to 1.08 ± 0.20 after treatment with 200 M P3G, almost matching the expression level in the control cells. These results suggest that P3G intervention can boost lysosomal function, up-regulate LAMP1 and cathepsin D expression, and promote lysosomal biosynthesis.
TFEB is the basic transcription factor helix-loop-helix leucine zip from the NITF family [53]. TFEB is the main transcriptional regulator of autophagy and lysosomal biogenesis. Overexpression of TFEB has been reported to produce improved autophagy, which in turn helps to break down accumulated lipids in liver cells to treat various genetic and dietary models of obesity and alcoholic liver disease [54][55][56]. In this work, we examined whether P3G-induced lysosomal function in L02 cells involves TFEB. The expression of the cytoplasmic TFEB protein decreased following FFA treatment (0.63 ± 0.04), as shown in Figure 5G,I. The expression of TFEB in the cytoplasm further decreased after treatment with 300 µM P3G (0.19 ± 0.01), whereas the expression in the nucleus dramatically increased (1.18 ± 0.05). P3G treatment did not significantly increase the expression of total TFEB in cells but significantly promoted the translocation of TFEB to the nucleus.

P3G Reduces the Level of Inflammation in NAFLD
The rate of inflammation in NAFLD increases with the build-up of ROS [57][58][59]. Caspase-1 is an important target of inflammasomes, which in turn increases cytokines such as interleukin-1β and tumor necrosis factor α (TNFα) and promotes the development of inflammation in NAFLD. In rodent models of NAFLD, the development of hepatic inflammation has also been associated with increased NF-κB activity [60]. In the NAFLD model, increased stimulation of ROS promotes NF-κB expression and the development of inflammation [61,62]. The inflammatory cytokines IL-1, IL-6, and TNF-α were tested by ELISA, and the effects of FFA and P3G on the expression of caspase-1 and NF-B were examined by Western blotting. The result is shown in Figure 6. FFA treatment significantly increased the expression of caspase-1, reaching 1.80 ± 0.084 compared to the control group. After 300 µM P3G treatment, the level of caspase-1 decreased to 0.85 ± 0.082 in comparison with the control group. P3G significantly reduced the expression levels of activated NF-κB to 1.24 ± 0.27 (100 µM), 0.83 ± 0.15 (200 µM), and 0.80 ± 0.06 (300 µM) (Figure 6). At the same time, according to the results of Figure 6, P3G significantly reduced the production of IL-6, TNF-α, and IL-1β and reduced the inflammation caused by FFA.
As a transcription factor, TFEB promotes the expression of PPARα [63]. We found that P3G significantly promoted TFEB expression, which in turn ameliorated lysosomal inhibition during NAFLD. Based on this finding, we further explored P3G's effects on the expression of PPARα and the peroxisomal genes CPT1A and ACOX1. Figure 7 demonstrates that the expression of PPARα, ACOX1, and CPT1A, were downregulated in L02 cells treated with FFAs. P3G upregulates the expression of PPARα protein and downstream target proteins ACOX1 and CPT1A in a dose-dependent way (Figure 7), which in turn promotes peroxisomal degradation of aberrantly accumulated fatty acids via β-oxidation. Yaoyao Jia et al. found that cyanidin-3-O-glucoside, pelargonidin-3-O-glucoside, and delphinidin-3-O-glucoside not only induce PPAR gene expression but also exert hypolipidaemic by activating PPARα activity [64]. Based on these results, we postulated that there might exist potential intermolecular interactions between PPARα and P3G; a molecular docking analysis was therefore carried out. Figure 7C depicts the interactions between P3G and PPARα amino acid residues. Hydrophobic interactions and hydrogen bonds are the main forces between P3G and PPARα. P3G forms hydrogen bonds with bond lengths of 3.2 and 3.9 Å with residues SER 280, ALA 333 of PPARα, respectively. P3G binds to THR 279, LEU 321, VAL 324, VAL 332, and ALA 333 of PPARα, respectively, to form σ-π bonds with bond lengths of 3.9, 3.6, 3.7, 4.0, and 3.6 Å. The MD analysis showed that the number of hydrogen bonds finally stabilized at 2 during the simulation, which was consistent with the docking results ( Figure 7D). flammation has also been associated with increased NF-κB activity [60]. In the NAFLD model, increased stimulation of ROS promotes NF-κB expression and the development o inflammation [61,62]. The inflammatory cytokines IL-1, IL-6, and TNF-α were tested b ELISA, and the effects of FFA and P3G on the expression of caspase-1 and NF-B wer examined by Western blotting. The result is shown in Figure 6. FFA treatment signifi cantly increased the expression of caspase-1, reaching 1.80 ± 0.084 compared to the contro group. After 300 μM P3G treatment, the level of caspase-1 decreased to 0.85 ± 0.082 i comparison with the control group. P3G significantly reduced the expression levels o activated NF-κB to 1.24 ± 0.27 (100 μM), 0.83 ± 0.15 (200 μM), and 0.80 ± 0.06 (300 μM (Figure 6). At the same time, according to the results of Figure 6, P3G significantly reduce the production of IL-6, TNF-α, and IL-1β and reduced the inflammation caused by FFA.  (B) quantitative data of caspase-1; (C) quantitative data of NF-κB p65; (D) quantitative data of TFEB (Nuclents); (the Western blots test was repeated 3 times and the image presented was typical of these 3 independent tests. The control group was considered 1.0; # p < 0.05, ## p < 0.01 compared with control; * p < 0.05, ** p < 0.01 vs. FFA-treated group). The amount of IL-1β (E), IL-6 (F), and TNF-α (G) were determined according to the ELISA product instructions; (E) quantitative data of IL-1β; (F) quantitative data of IL-6; (G) quantitative data of TNF-α. (Control group was considered as 1.0; # p < 0.05, ## p < 0.01 compared with control; * p < 0.05, ** p < 0.01 vs. FFA-treated group.). tween P3G and PPARα amino acid residues. Hydrophobic interactions and hydrogen bonds are the main forces between P3G and PPARα. P3G forms hydrogen bonds with bond lengths of 3.2 and 3.9 Å with residues SER 280, ALA 333 of PPARα, respectively. P3G binds to THR 279, LEU 321, VAL 324, VAL 332, and ALA 333 of PPARα, respectively, to form σ-π bonds with bond lengths of 3.9, 3.6, 3.7, 4.0, and 3.6 Å. The MD analysis showed that the number of hydrogen bonds finally stabilized at 2 during the simulation, which was consistent with the docking results ( Figure 7D).

Molecular Dynamics
Molecular dynamics (MD) simulations supply elaborate information between proteins and ligands at different time scales under simulated physiological conditions, which helps us further understand the binding process [65]. As shown in Figure S9, the potential energy, temperature, pressure, and density of PPARα and PPARα/P3G complexes reached a steady state in a short time during the MD simulation. This result indicates that the topology files of protein and ligand and the parameters implied in the MD simulation process are reasonable.
To determine when the steady state will arrive and to verify the stability of the simulation, the root mean square deviation (RMSD) is used [66]. Figure 8A demonstrates that after 50 ns of simulation, free PPAR's ultimate RMSD value is 0.2 nm. In contrast, the RMSD value of the PPARα/P3G complex showed a relatively stable state at about 0.18 nm at the end of the simulation time. It may be due to the formation of hydrogen and hydrophobic bonds between P3G and PPARα, which leads to a more stable conformation of the complex. The root mean square fluctuations (RMSF) values were also calculated for all residues on all time scales ( Figure 8B). The remaining sites were comparatively constant compared to free PPARα, with the exception of the significant fluctuations in residues 300-400 in the complex system, indicating that P3G triggered flexible alterations in PPARα residues 300-400. Interestingly, we found that the radius of gyration of the complex was smaller than that of the free PPARα ( Figure 8C), suggesting that the binding of P3G pushed the protein to become more compact. The binding of P3G also reduced the solvent-accessible surface area (SASA) of the complex ( Figure 8D). The radial distribution function (RDF) is usually used to reveal the distribution density of ligands at a certain distance (r) from the central atom of the protein [67]. Figure 8E shows the RDF of P3G with a typical peak at 1 nm, representing the average distance between the central atom of the protein and P3G, after which a sharp decrease in g(r) is noted when r(nm) is large. The structural alterations in the protein and ligand before and after P3G's binding to PPAR were also looked at, along with changes in the Gibbs energy landscape profiles. The distinction between PPARα and PPARα/P3G complexes before and after MD simulations is shown in Figure 9A,B. Figure 9A,B indicate that, when P3G binds to PPARα, the structure protein changes significantly in the P3G's binding area. After the simulation, Figure 9C depicts a sizable modification in the spatial structure of P3G. The change of PPARα with P3G after MD simulation may be due to the stronger interaction between P3G and PPARα. The change of PPARα and P3G after MD simulation may be due to the stronger interaction between P3G and PPARα. The free energy shape of PPARα and PPARα/P3G complexes is depicted in Figure 9D. (both 2D and 3D landscapes). Figure 9D shows that the free energy distribution of PPARα/P3G is more concentrated than that of free PPARα, demonstrating that P3G promotes the conformation of PPARα to become more compact and stable.  The secondary structures of the PPARα and PPARα/P3G complexes as well as Ramachandran plots were also examined. Figure S10 and video S1 indicates how the distribution of PPARα's secondary structures underwent a significant change after the binding of P3G, particularly the secondary structure formed by the amino acid residues at positions 330-340. After P3G was included, the secondary structure of the amino acid residues at positions 330 to 340 altered from S BEND to 3-Helix. The Ramachandran plot of ALA333 changed from 60°, 60° (phi, psi angle) to −60°, 120° (phi, psi angle) after combining with P3G ( Figure S10), and this shift promotes the formation of 3-Helix. 3-Helix may be more advantageous than S-BEND for maintaining the structure of the protein [68]. Due to the secondary structure transformation, the PPARα /P3G complex has a more compact conformation compared to the free PPARα (the movie provided in the Supplementary Material describes this process).

Discussion
The liver is essential for lipid metabolism [69]. When lipids accumulate, fats are primarily stored as triglycerides in hepatocytes, resulting in steatosis, the main histological feature of nonalcoholic fatty liver disease [70]. Anthocyanins are polar compounds found in foods as glycosides and acylated glycosides that have a variety of health-promoting properties. Many studies have previously reported that may have significant potential in the prevention and treatment of NAFLD [71]. Anthocyanins from Bilberry, Cherry, and Grape-Skin have been shown to improve lipid metabolism and reduce triglyceride levels in the NAFLD model [72][73][74]. We isolated six anthocyanin components from purple corn cob and tested their ability to interfere with NAFLD in this study. We discovered that P3G had the greatest ability to improve NAFLD. The development of NAFLD and oxidative stress are tightly connected [75]. When NAFLD occurs, lipid accumulation leads to The secondary structures of the PPARα and PPARα/P3G complexes as well as Ramachandran plots were also examined. Figure S10 and Video S1 indicates how the distribution of PPARα's secondary structures underwent a significant change after the binding of P3G, particularly the secondary structure formed by the amino acid residues at positions 330-340. After P3G was included, the secondary structure of the amino acid residues at positions 330 to 340 altered from S BEND to 3-Helix. The Ramachandran plot of ALA333 changed from 60 • , 60 • (phi, psi angle) to −60 • , 120 • (phi, psi angle) after combining with P3G ( Figure S10), and this shift promotes the formation of 3-Helix. 3-Helix may be more advantageous than S-BEND for maintaining the structure of the protein [68]. Due to the secondary structure transformation, the PPARα /P3G complex has a more compact conformation compared to the free PPARα (the movie provided in the Supplementary Material describes this process).

Discussion
The liver is essential for lipid metabolism [69]. When lipids accumulate, fats are primarily stored as triglycerides in hepatocytes, resulting in steatosis, the main histological feature of nonalcoholic fatty liver disease [70]. Anthocyanins are polar compounds found in foods as glycosides and acylated glycosides that have a variety of health-promoting properties. Many studies have previously reported that may have significant potential in the prevention and treatment of NAFLD [71]. Anthocyanins from Bilberry, Cherry, and Grape-Skin have been shown to improve lipid metabolism and reduce triglyceride levels in the NAFLD model [72][73][74]. We isolated six anthocyanin components from purple corn cob and tested their ability to interfere with NAFLD in this study. We discovered that P3G had the greatest ability to improve NAFLD. The development of NAFLD and oxidative stress are tightly connected [75]. When NAFLD occurs, lipid accumulation leads to mitochondrial DNA and protein abnormalities, resulting in abnormal electron transfer to form superoxide or hydrogen peroxide [76]. The activities of GSH, CAT, and other antioxidant enzymes in NAFLD patients are low, so ROS cannot be cleared in hepatocytes in time [77]. Therefore, we investigated the effect of P3G on free radicals in LO2 cells. The findings demonstrated that P3G is an effective free radical scavenger. This finding is consistent with the fact that P3G exposure increases the activity of antioxidant enzymes such as SOD, GPx, and CAT in cells. Anthocyanins from purple sweet potato and mulberry have been shown to reduce ROS production, restore glutathione (GSH) and antioxidant enzyme activity, and protect experimental mice from free radical-mediated endoplasmic reticulum stress [78,79].
Mitochondria are involved in the regulation of intracellular calcium levels, cellular homeostasis, and ROS [80]. Mitochondrial dysfunction and an impaired mitochondrial respiratory chain have been reported in NAFLD patients [81]. Oxidative stress causes mitochondrial dysfunction and an impaired mitochondrial respiratory chain, which further reduces cellular catabolism of lipids and energy supply [82]. This could hasten the progression of liver inflammation and NAFLD. Reducing mitochondrial dysfunction and restoring the mitochondrial respiratory chain may be an effective way to treat NAFLD. This study also confirmed that P3G, in a dose-dependent manner, reduces mitochondrial membrane damage, improves mitochondrial information, and increases ATP supply in cells.
Lysosomes promote the degradation of polysaccharides, proteins, and lipids/free fatty acids (FFA) [83]. Lysosomes can alleviate the lipid accumulation of hepatocytes by selectively degrading triglycerides (TGs) and cholesterol [84]. Lysosomal inhibition occurs during NAFLD, and genes involved in peroxisome biogenesis and lipid metabolism are significantly downregulated following lysosomal inhibition [85]. Expression of PPARα is significantly downregulated during NAFLD [86]. Peroxisome genes such as CPT1A and ACOX1 are transcriptional targets of PPARα, and lysosomal repression, reduction in peroxisome, and fatty acid metabolism-related enzyme activities are associated with PPARα [85]. This study discovered that P3G intervention dramatically increased the expression of the lysosomal markers LAMP1 and cathepsin D. In the meantime, P3G might encourage TFEB's nuclear translocation and promote the expression of PPARα. This outcome is consistent with Yang Xu et al.'s findings [27]. The peroxisome's CPT1 and ACOX1 expression was further encouraged by increased PPARα expression. The results of molecular docking and molecular dynamics assays showed that the major interactions between P3G and PPAR are hydrogen bonds and hydrophobic contacts. The formation of hydrogen bonds and hydrophobic forces between P3G and PPAR facilitates the modification of PPARα's secondary structure. The modification of the secondary structure may aid in the stabilization of the P3G/ PPARα complex and may increase the activity of PPARα.
The variation in the OH part, the quantity and type of OH part methylation, the degree of OH part methylation, and the binding position of these linkages all contribute to the structural differences of anthocyanins [87]. In comparison to the other five acyl-containing anthocyanins, the acyl-free P3G showed better effectiveness in improving NAFLD, according to this study. In comparison to the other five acyl-containing anthocyanins, the acyl-free P3G showed better effectiveness in reducing NAFLD, according to this study. Acylation may improve anthocyanin stability in aqueous solution [88], but it may also impair anthocyanin biological activity. This could be related to anthocyanin catabolism in cells. The investigation of the impact of acyl groups on the biological activity of anthocyanins will be the primary focus of the following phase of research.

Conclusions
In this study, an established hepatocyte model was used to evaluate the effect of purple corncob anthocyanin on NAFLD, in which P3G showed the potential to improve FFAinduced NAFLD. Further studies found that P3G seems to reduce FFA-induced oxidative stress by reducing mitochondrial damage, including the increase in ROS and superoxide anion levels. P3G may enhance the function of lysosomes by regulating the translocation of TFEB, so as to promote the degradation of lipid peroxide in hepatocytes and reduce the possibility of ROS production. P3G can also reduce caspase-1 and NF-κB expression to alleviate the level of hepatocyte inflammation caused by FFA. P3G interacts with PPARA primarily through hydrogen bonds and hydrophobic forces, according to the results of molecular docking and MD simulation analysis. The interaction with P3G increases the stability and compactness of the PPARA structure. P3G may upregulate the expression of genes involved in the oxidative catabolism of fatty acids by activating PPARA in addition to upregulating the expression of PPARA through TFEB, thereby reducing the aberrant buildup of fatty acids. Our research enriches the understanding of the biological function of purple corncob anthocyanins and provides ideas for the development of functional foods to improve NAFLD.

Conflicts of Interest:
The authors declare that the research was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest.