Pomegranate seed oil attenuates palmitic acid‐induced hepatic injury through modulation of oxidative stress: insights from GC‐Q‐TOF‐MS‐based metabolomics

Abstract

BACKGROUND

Non‐alcoholic fatty liver disease (NAFLD) is a leading cause of chronic liver disease worldwide, with oxidative stress and lipid dysregulation as central pathogenic drivers. Pomegranate seed oil (PSO), rich in the conjugated fatty acid punicic acid, has demonstrated potential antioxidant and lipid‐modulating properties, yet its hepatoprotective mechanisms remain incompletely characterized at the metabolic level.

RESULTS

Gas chromatographic–mass spectrometric analysis of the fatty acid composition of supercritical CO₂‐extracted PSO revealed that punicic acid was the predominant fatty acid, comprising 74.47% of the total identified fatty acids, which collectively accounted for approximately 95% of the fatty acid profile. In palmitic acid (PA)‐challenged C57BL/6 mice, PA administration significantly increased serum and hepatic lipid levels, liver function markers, oxidative stress indicators, and inflammatory cytokines. PSO intervention dose‐dependently ameliorated several of these abnormalities and markedly reduced hepatic lipid accumulation. Gas chromatography–quadrupole time‐of‐flight mass spectrometry‐based serum metabolomics revealed that PSO markedly reversed PA‐induced metabolic disturbances, identifying 27 key differential metabolites predominantly associated with amino acid and lipid metabolism. Pathway enrichment analysis highlighted oxidative stress‐related pathways, including glutamate, taurine, and fatty acid metabolism. PSO significantly upregulated antioxidant metabolites (cysteine, glutamate, hypotaurine, α‐tocopherol) while downregulating oxidative markers (uric acid, xanthine). Complementary in vitro experiments in PA‐treated L02 human hepatocytes further demonstrated that PSO alleviated lipotoxicity by attenuating reactive oxygen species generation and preserving membrane integrity.

CONCLUSION

These findings demonstrate that PSO exerts hepatoprotective effects through multi‐target, multi‐pathway synergistic mechanisms, particularly via enhancing endogenous antioxidant defense systems, providing scientific evidence for PSO as a promising nutritional intervention strategy for NAFLD prevention and treatment. © 2026 Society of Chemical Industry.

ABBREVIATIONS

ALP

alkaline phosphatase

ALT

alanine aminotransferase

AST

aspartate aminotransferase

AUC

area under the glycemia curve

ELISA

enzyme‐linked immunosorbent assay

FAMEs

fatty acid methyl esters

FC

free cholesterol

FFA

free fatty acids

FI

fluorescence intensity

FO

fish oil

GP

generalized polarization

HDL‐c

high‐density lipoprotein cholesterol

4‐HNE

4‐hydroxynonenal

H₂O₂

hydrogen peroxide

IL‐1β / 6

interleukin‐lβ / 6

LDH

lactate dehydrogenase

LDL‐c

low‐density lipoprotein cholesterol

MCT

medium‐chain triglyceride

MDA

malondialdehyde

NAFLD

non‐alcoholic fatty liver disease

NO

nitric oxide

O²⁻

superoxide anion

PA

palmitic acid

PI

propidium iodide

PSO

pomegranate seed oil

RI

retention indices

TC

total cholesterol

TG

triglyceride

TNF‐α

tumor necrosis factor‐α

XO

xanthine oxidase

ROS

reactive oxygen species

—SH

sulfhydryl groups

INTRODUCTION

Liver diseases represent a significant global health burden, with non‐alcoholic fatty liver disease (NAFLD) emerging as the most common chronic liver disorder worldwide, affecting approximately 25% of the global population. ¹ NAFLD encompasses a spectrum of conditions ranging from simple steatosis to non‐alcoholic steatohepatitis (NASH), which can progress to cirrhosis and hepatocellular carcinoma (HCC). The pathogenesis of NAFLD is multifactorial, wherein lipotoxicity and oxidative stress serve as central mechanistic drivers of disease initiation and progression. ² On March 14, 2024, the US Food and Drug Administration approved resmetirom (Rezdiffra) for the treatment of adults with non‐cirrhotic MASH accompanied by moderate‐to‐advanced liver fibrosis, to be used in conjunction with diet and exercise – marking the first drug approval in this therapeutic area after more than two decades of research efforts.

Saturated fatty acids, particularly palmitic acid (PA), as a major component of Western diets, play a pivotal role in the pathogenesis of NAFLD. Lipotoxicity arises when excessive free fatty acids (FFAs), especially saturated fatty acids such as PA, accumulate in non‐adipose tissues, leading to cellular dysfunction and apoptosis. In hepatocytes, PA accumulation triggers multiple cellular stress responses, including endoplasmic reticulum stress, mitochondrial dysfunction, and increased production of reactive oxygen species (ROS), ultimately leading to oxidative damage.^{3, 4} In the context of NAFLD, oxidative stress exacerbates lipotoxicity‐induced hepatocellular injury, promotes inflammation, and accelerates disease progression. ⁵ Therefore, mitigating oxidative stress represents a promising therapeutic strategy for preventing and treating NAFLD.

Vegetable oils are complex mixtures composed primarily of fatty acids and various bioactive minor components, and are generally regarded as beneficial to human health. The biological functions of vegetable oils are largely determined by the chain length, degree of unsaturation, and double‐bond configuration of their constituent fatty acids. Notably, the conjugated structure of conjugated fatty acids reduces oxidative stability while enhancing conformational and spatial stability. Polyunsaturated fatty acids (PUFAs) play essential roles in maintaining membrane fluidity and cellular function, facilitating cholesterol esterification, lowering cholesterol and triglyceride levels, reducing blood viscosity, improving microcirculation, and supporting neuronal activity and cognitive performance. These physiological effects are mediated through mechanisms involving oxidative stress regulation, inflammatory response modulation, endoplasmic reticulum stress alleviation, and endothelial protection, contributing to cardiovascular health and the prevention or management of inflammatory, neurodegenerative, and metabolic diseases.^{6, 7} Pomegranate seed oil (PSO), extracted from the seeds of Punica granatum L., is distinguished by its exceptionally high content of punicic acid (PA; 9c,11 t,13c‐18:3), a conjugated linolenic acid isomer that accounts for approximately 60–90% of total fatty acids.^{8, 9, 10} As the principal bioactive component of PSO, punicic acid possesses unique physicochemical properties and diverse biological activities, including anti‐inflammatory, anticancer, insulin‐sensitizing, skin barrier‐enhancing, and cardiovascular protective effects. Importantly, its conjugated triene structure is considered a key structural basis for its potent antioxidant capacity.^{11, 12, 13, 14, 15, 16, 17, 18, 19, 20, 21, 22}

Metabolomics is a powerful, unbiased analytical approach for systematically characterizing global metabolic alterations and identifying disrupted biological pathways. Among available platforms, gas chromatography–quadrupole time‐of‐flight mass spectrometry (GC‐Q‐TOF‐MS) offers high sensitivity, resolution, and broad metabolite coverage, making it particularly suitable for metabolomic investigations.^{23, 24, 25, 26} Importantly, integration of metabolomics with conventional biochemical and histopathological assessments enables a more comprehensive and mechanistic understanding of disease progression and therapeutic intervention by linking metabolic reprogramming to functional and structural outcomes.

Given the pivotal role of oxidative stress in NAFLD pathogenesis and the reported antioxidant properties of PSO, we hypothesized that PSO exerts hepatoprotective effects against PA‐induced hepatic injury through modulation of oxidative stress‐related metabolic pathways. To test this hypothesis, an integrated experimental strategy combining in vivo and in vitro models with GC‐Q‐TOF‐MS‐based metabolomic profiling was employed, followed by mechanistic validation in PA‐challenged L02 hepatocytes. To our knowledge, this is the first study to systematically characterize the hepatoprotective metabolic reprogramming induced by PSO using untargeted metabolomics. Our results demonstrate that PSO alleviates PA‐induced hepatotoxicity via multi‐target regulation of oxidative stress and lipid metabolism, providing novel mechanistic insights and supporting its therapeutic potential for NAFLD intervention.

MATERIALS AND METHODS

Instruments and equipment

Supercritical CO₂ extraction was performed using a supercritical fluid extraction system (SFE‐2, Applied Separations, Allentown, PA, USA). Cell viability measurements were conducted using a microplate reader (gen5, BioTek Instruments, Winooski, VT, USA). Metabolomic analysis was performed on a GC‐Q‐TOF‐MS system (7890B‐7200, Agilent Technologies, Santa Clara, CA, USA).

Materials

Palmitic acid (PA; CAS 408‐35‐5), adonitol (CAS 488‐81‐3), n‐alkane, fish oil (CAS 8002‐50‐4), and Supelco 37 component FAME Mix were purchased from Sigma‐Aldrich (Saint Louis, MO, USA). 2′,7′‐Dichlorofluorescin diacetate (DCFH‐DA; CAS 4091‐99‐0, Aladdin), 3‐(4,5‐dimethylthiazol‐2‐yl) ‐2,5‐diphenyltetrazolium bromide (MTT; CAS 298‐93‐1, Sangon‐Biotech), BSTFA containing 1% TMCS (CAS 25561‐30‐2, Solarbio), O‐methoxyamine HCl (CAS 593‐56‐6, Macklin), propidium iodide (PI; CAS 25535‐16‐4, Solarbio), MitoSOX Red (CAS 1003197‐00‐9, TargetMol), lucigenin (CAS 2315‐97‐1, Adamas), Amplex Red (ADHP; CAS 119171‐73‐2, energy chemical), horseradish peroxidase (HRP; CAS 9003‐99‐0, Orileaf), 2,4‐dinitrophenylhydrazine (DNPH; CAS 119‐26‐6, Macklin), dihydroethidium (CAS 104821‐25‐2, Bide Pharmatech Ltd), menadione (CAS 58‐27‐5, Bide Pharmatech Ltd), Griess reagent (Beyotime), LDH kit (Beyotime), Enhanced BCA Protein Assay Kit (Beyotime), margaric acid (CAS 506‐12‐7, Adamas Life), and medium‐chain triglyceride (MCT, Adamas Life) oil were purchased from Biotechnology of China. Dulbecco's Modified Eagle Medium (DMEM), fetal bovine serum (FBS), and antibiotic‐antimycotic solution were obtained from Gibco (Thermo Fisher Scientific, Waltham, MA, USA).

Fresh pomegranates (Punica granatum L.) were purchased from local markets in Yecheng County, Kashgar, Xinjiang (37.89 N, 77.49 E) during the harvest season (September–October 2022).

Extraction and pretreatment of PSO

Fresh pomegranate fruits were manually peeled, and the seeds were thoroughly rinsed with distilled water to remove residual pulp. The cleaned seeds were dried at 50 °C for 48 h, ground, and sieved to obtain particles between 18 and 60 mesh. The resulting powder was stored at −20 °C until extraction.

PSO was extracted using supercritical carbon dioxide (SC‐CO₂) under optimized conditions (280 bar, 45 °C, 3 h) with ethanol (4%, v/v) as a co‐solvent. The extraction yielded a clear, bright‐yellow oil, which was collected and stored at −20 °C for subsequent analyses.

Fatty acid component analysis of PSO

Preparation of fatty acid methyl esters

PSO samples (0.1 mL) were dissolved in a mixture of n‐hexane (0.7 mL) and diethyl ether (0.3 mL), followed by the addition of methanol (1 mL) and 0.8 mol L⁻¹ potassium hydroxide in methanol (10 mL). After standing at room temperature for 15 min, distilled water was added to induce phase separation. The upper organic phase containing fatty acid methyl esters (FAMEs) was collected for GC‐MS analysis.

GC‐MS analysis

FAMEs were analyzed using a GC‐Q‐TOF‐MS system (7200, Agilent Technologies) equipped with an HP‐5MS capillary column (30 m × 250 μm × 0.25 μm). Helium was used as the carrier gas at a constant flow rate of 1.0 mL min⁻¹, and the injector temperature was set at 280 °C.

The oven temperature program was as follows: initial temperature of 60 °C held for 3 min, increased to 240 °C at a rate of 4 °C min⁻¹, and held for 3 min. Samples (0.5 μL) were injected in split mode (20:1). Mass spectra were acquired in electron impact (EI) mode at 70 eV over an m/z range of 50–500.

Retention indices (RI) were calculated using a homologous series of n‐alkanes under identical chromatographic conditions. Fatty acid components were identified by comparing both RI values and mass spectra with entries in the NIST 14 mass spectral library.

Animal experiments

Animals and housing

Seventy‐two male SPF‐grade C57BL/6 mice (8–10 weeks old, 20–25 g) were purchased from the Experimental Animal Center of Xinjiang Medical University. Animals were housed under controlled conditions (22 ± 2 °C, 50–60% relative humidity, 12 h light/dark cycle) with free access to standard chow and sterile water. The animal study followed the ‘Guide for the Care and Use of Laboratory Animals’ established by the Xinjiang Medical University and was approved by the Institutional Animal Care and Use Committee of Xinjiang Medical University (Approval No. IACUC‐JT‐20250626‐17).

Experimental design and grouping

After a 7‐day acclimatization period, mice were randomly assigned to six groups (n = 12 per group) using a computer‐generated randomization protocol: (i) control; (ii) model group (PA); (iii) fish oil (FO); (iv) low‐dose PSO (PSO‐L); (v) high‐dose PSO (PSO‐H); and (vi) PSO control (PSO‐C). Body weight and general health status were monitored throughout the experimental period.

Drug administration and NAFLD model induction

All mice received intragastric administration for 14 consecutive days. The control and PA groups were administered MCT oil (4.5 mL kg⁻¹) as vehicle control. The FO group was administered fish oil at 0.8 g kg⁻¹, while the PSO‐L and PSO‐H groups received PSO at doses of 0.2 and 0.8 g kg⁻¹, respectively, The PSO‐C group received PSO at 0.8 g kg⁻¹.

Beginning on day 3 of gavage (2 h post‐administration), NAFLD was induced by alternating intraperitoneal and tail vein injections of PA, as previously described. ²⁷ The control and PSO‐C groups received 600 μmol L⁻¹ bovine serum albumin (BSA) solution (20 mL kg⁻¹), while the remaining groups received PA‐BSA complexes (5 mmol L⁻¹ PA, 20 mL kg⁻¹). Injections were administered once daily for 12 consecutive days.

Sample collection

Two hours after the final gavage on day 14, mice were anesthetized with pentobarbital sodium (45 mg kg⁻¹, i.p.). Blood was collected via cardiac puncture, allowed to clot at room temperature for 30 min, and centrifuged at 3000 rpm for 15 min at 4 °C to obtain serum, which was stored at −80 °C until analysis.

Following euthanasia by cervical dislocation, liver, heart, kidney, and spleen tissues were rapidly excised, rinsed with ice‐cold PBS, blotted dry, and weighed. Organ coefficients were calculated as organ weight/body weight (mg g⁻¹). Portions of liver tissue from the left lobe were either snap‐frozen in liquid nitrogen for biochemical analyses or fixed in 4% paraformaldehyde for histological evaluation by Oil Red O staining. ²⁸

Biochemical analyses

Serum biochemical parameters, including alanine aminotransferase (ALT), aspartate aminotransferase (AST), alkaline phosphatase (ALP), total cholesterol (TC), triglycerides (TG), low‐density lipoprotein cholesterol (LDL‐c), high‐density lipoprotein cholesterol (HDL‐C), free cholesterol (FC), and free fatty acids (FFA), were measured using commercial assay kits following the manufacturers' protocols. ²⁹ Oxidative stress biomarkers in serum and liver tissues, including malondialdehyde (MDA), xanthine oxidase (XO), sulfhydryl groups (—SH), and 4‐hydroxynonenal (4‐HNE), were quantified using enzymatic or colorimetric kits. Pro‐inflammatory cytokines (TNF‐α, IL‐6, and IL‐1β) were determined by enzyme‐linked immunosorbent assay.

Serum metabolomic analysis

Serum metabolomic profiling was performed using GC‐Q‐TOF‐MS following established protocols, with minor modifications.^{30, 31} Metabolites were extracted using methanol–acetonitrile–water (2:2:1, v/v/v), with adonitol (10 μL, 0.01 mg mL⁻¹) added as an internal standard. After lyophilization, samples were derivatized with methoxypyridine hydrochloride (20 mg mL⁻¹) at 37 °C for 90 min, followed by BSTFA containing 1% TMCS at 70 °C for 60 min. GC‐MS analysis was conducted on an Agilent 7200 Q‐TOF GC/MS system equipped with a DB‐5MS + DG column (30 m × 250 μm × 0.25 μm). The oven temperature program was as follows: 80 °C (2 min), ramped to 180 °C at 10 °C min⁻¹, to 250 °C at 5 °C min⁻¹, and to 290 °C at 15 °C·min⁻¹, with a final hold of 9.33 min. Helium was used as the carrier gas at 1 mL min⁻¹. Mass spectra were acquired in EI mode (70 eV) over a mass range of m/z 50–1000. Samples (1 μL) were injected in splitless mode with six technical replicates per sample.

Raw data were processed using MS‐DIAL (version 4.8.0) and metabolites were identified against the MassBank and NIST libraries. Multivariate analyses, including principal component analysis (PCA) and orthogonal partial least squares discriminant analysis (OPLS‐DA), were performed using MetaboAnalyst. Differential metabolites were selected based on variable importance in projection (VIP) > 1 and fold change >1.2 or < 0.83 (P < 0.05). KEGG pathway enrichment analysis was conducted to identify affected metabolic pathways.

In vitro experiment protocol

Preparation of saponified PSO

Saponified PSO was prepared by mixing 1.0 g oil with 0.2 g NaOH and 2.5 mL deionized water, followed by heating at 95 °C under continuous stirring for 1 h. While the mixture remained hot, poloxamer 188 solution (30 mg mL⁻¹) was added, and the final concentration was adjusted to 4 mg mL⁻¹ to obtain a stock solution for cell‐based experiments (pH 7.2–7.4). The stock solution was diluted with DMEM medium to the desired working concentrations before use. ³²

Cell culture

Human normal hepatic L02 cells, kindly provided by Xinjiang Medical University, were routinely cultured in this laboratory. Cells were maintained in DMEM supplemented with 10% FBS and 1% penicillin–streptomycin under standard culture conditions.

Cell viability assay

L02 cells in logarithmic growth phase were harvested and seeded in 96‐well plates at a density of 2.5 × 10⁴ cells per well. Cells were allowed to adhere overnight before treatment. The experimental design included a control group (L02), PSO group, FO group and PA group, with each group consisting of three to four replicate wells. Cell viability across all groups was evaluated using the MTT assay.

Effect of PSO on PA‐induced lipotoxicity in L02 cell

L02 cells in logarithmic growth phase were seeded in 96‐well plates at a density of 2 × 10⁴ cells per well and allowed to adhere overnight. Cells were pretreated with PSO or FO for 2 h, followed by exposure to 400 μmol L⁻¹ PA for an additional 24 h. Each experimental condition was performed in three to five replicate wells.

1. Cell viability and cytoprotection were evaluated using the MTT assay and methylene blue staining. ³³

2. Culture supernatants were collected to determine nitric oxide (NO) production and lactate dehydrogenase (LDH) release.^{34, 35}

3. Intracellular ROS levels were measured using the DCFH‐DA fluorescent probe. ³⁶

4. Hydrogen peroxide (H₂O₂) generation was quantified using Amplex Red (50 μmol L⁻¹) in the presence of horseradish peroxidase (1 U mL⁻¹). ³⁷

5. Superoxide anion (O₂^•−) production was assessed using lucigenin‐enhanced (20 μmol L⁻¹) chemiluminescence. ³⁸ O₂^•− production in menedione (Mene)‐induced L02 cells was evaluated using a dihydroethidium‐based fluorescence assay after PSO or FO pretreatment. ³⁹

6. Mitochondrial ROS levels were determined by staining cells with MitoSOX Red (2.5 μmol L⁻¹). ⁴⁰

7. Lipid peroxidation was assessed by measuring MDA levels following derivatization with DNPH.^{41, 42}

8. Plasma membrane integrity was evaluated using PI staining. ⁴³

9. Membrane fluidity was assessed by Laurdan staining, and generalized polarization (GP) values were calculated as GP = (FI₄₄₀‐FI₄₉₀)/(FI₄₄₀ + FI₄₉₀). ⁴⁴

Statistical analysis

Statistical differences among groups were assessed by one‐way analysis of variance followed by Tukey's honestly significant difference (HSD) post hoc test for multiple comparisons, employing GraphPad Prism 9.5 (GraphPad Software, Boston, MA, USA). A significance threshold of P < 0.05 was adopted for determining statistical significance.

RESULTS

Chemical compositions of PSO

The yield of PSO obtained by supercritical CO₂ extraction was about 8%. The fatty acids of PSO were analyzed by GC‐MS, and the total ion current chromatogram is shown in Fig. 1. The RI of each peak was determined according to the retention time of n‐alkanes, and the composition was determined by comparison with the NIST14 library (National Institute of Standards and Technology, Gaithersburg, MD, USA). As shown in Table 1, nine fatty acids were identified in PSO, among which C18:3 was present in a high amount (relative content of 60–90%). The total content of PUFAs in PSO is about 90%, while the contents of saturated fatty acids and monounsaturated fatty acids are lower. ⁴⁵ This ingredient characteristic makes PSO have high biological activity, which may be the material basis for its protective effect. ⁴⁶

Figure 1.

Pomegranate seed oil methyl ester products (FAMEs) – HP‐5MS column total ion current plot.

Table 1.

Gas chromatographic–mass spectrometric analysis of fatty acid components in pomegranate seed oil

	Fatty acid	RT	CAS	MF	MW	RI	Percentage (%)	FAMEs fragments m/z
1	Hexadecanoic acid (C16:0)	30.051	57‐10‐3	C16H32O2	256.42	1926	0.91	74, 87, 43, 55, 143, 83, 227, 129
2	Linolic acid (C18:2)	33.229	60‐33‐3	C18H32O2	280.4	2091	1.02	67, 81, 95, 55, 82, 41, 96, 79
3	Oleic acid (C18:1)	33.336	112‐80‐1	C18H34O2	282.5	2092	1.75	55, 69, 74, 83, 97, 87, 84, 96
4	Stearate (C18:0)	33.809	57‐11‐4	C18H36O2	284.5	2128	0.71	74, 87, 43, 55, 143, 83, 71, 255
5	Punicic acid (C18:3)	35.72	544‐72‐9	C18H30O2	278.4	2239	74.47	79, 91, 93, 67, 77, 105
6	Eleostearic acid (C18:3)	35.78	4337‐71‐7	C18H30O2	278.4	2239	9.12	91, 79, 93, 77, 67, 105, 121
7	Jacaric acid (C18:3)	36.017	28 872‐28‐8	C18H30O2	278.4	2253	8.79	91, 79, 93, 77, 67, 105
8	Linolenic acid (C18:3)	36.484	506‐26‐3	C18H30O2	292.46	2279	1.24	80, 91, 79, 93, 77, 67, 107, 121, 81
9	Eicosanoic acid (C20:0)	36.83	506‐30‐9	C20H40O2	326.56	2329	0.31	55, 69, 81, 96, 250, 123, 87, 143, 326, 283, 75
	SFA						1.62
	MUFA						1.75
	PUFA						93.4
	Total						96.77

RI means retention index in NIST 14, Italicized m/z values indicate compound‐specific fragment ions retrieved from PubChem, whereas non‐italicized m/z values represent fragment ions observed in the mass spectra. RT, retention time, CAS, Chemical Abstracts Service Registry Number; MF, molecular formula; MW, molecular weight; FAMEs, fatty acid methyl esters.

The effect of PSO on the PA‐induced NAFLD in mice

General observations and organ coefficients

Throughout the experimental period, the general health status of the mice was monitored every other day through systematic evaluation of behavioral and physical parameters. Mice in the control group remained in a stable and healthy condition, exhibiting normal mental alertness, regular food and water intake, a glossy coat, and high levels of spontaneous activity. In contrast, mice in the PA model group exhibited progressive deterioration following model induction, characterized by reduced locomotor activity, loss of coat luster, and greater body weight gain compared with the control group. As illustrated in the body weight curve, the PA group displayed pronounced fluctuations during the modeling period, whereas body weight in the PSO intervention groups and the positive control group (fish oil) remained relatively stable, closely approximating the trajectory observed in the control group. (Fig. 2(A)). Treatment groups exhibited intermediate phenotypes between the control and PA groups, with the PSO‐H and PSO‐C groups showing markedly improved general conditions relative to the PA model group.

Figure 2.

Pomegranate seed oil (PSO) attenuates palmitic acid (PA)‐induced hepatic steatosis in a dose‐dependent manner. *P < 0.05, **P < 0.01 versus PA group. (A) Body weight monitoring over 21‐day treatment period. (B) Hepatosomatic index. (C) Cardiac index. (D) Kidney index. (E) Spleen index. (F) Representative Oil Red O staining showing lipid accumulation (200×, scale bar = 50 μm).

Organ coefficient analysis revealed significant hepatomegaly and splenomegaly in the PA model group compared with the control group (P < 0.05; Fig. 2(B),(E)), providing evidence for effective model induction. In contrast, heart and kidney coefficients were not significantly altered (Fig. 2(C),(D)), suggesting that PA‐induced alterations predominantly affected the liver, with potential involvement of immune‐related organs. This selectivity may be attributed to the liver's central role in lipid metabolism and its preferential uptake of PA. Notably, the liver coefficient in the PSO‐C group was comparable to that of the control group, indicating good tolerability of PSO at 0.8 g kg⁻¹. In addition, no significant abnormalities were observed in serum biomarkers of liver function, lipid metabolism, or oxidative stress in the PSO‐C group, although longer‐term toxicological evaluation is still warranted.

Oil Red O staining showed that hepatocytes in the control and treatment groups largely preserved a cord‐like hepatic architecture, although evident inter‐field heterogeneity was observed. In contrast, the PA model group exhibited extensive cytoplasmic lipid droplet accumulation, stained orange‐red to deep red, sharply contrasting with the pale‐blue nuclei. Lipid deposition was diffusely distributed and manifested as both microvesicular steatosis, characterized by numerous small droplets with relatively central nuclei, and macrovesicular steatosis, in which large droplets occupied most of the cytoplasm and displaced nuclei to the cell periphery. Extensive lipid droplet accumulation and steatosis were observed in hepatocytes from the PA group, while PSO and FO treatments significantly alleviated hepatic lipid deposition. (Fig. 2(F)).

Serum biochemical parameters in PA‐induced hepatic injury model

PA injection successfully induced lipid metabolic dysregulation, oxidative stress, and inflammation in mice. Serum AST levels were significantly elevated, whereas ALT and ALP remained unchanged, suggesting early‐stage or mild hepatic injury rather than overt hepatocellular necrosis (Fig. 3(A)–(C)). Lipid metabolism was markedly disrupted, as evidenced by significant increases in TC and FC, reflecting a hypercholesterolemic state (Fig. 3(D),(H)), providing translational relevance for investigating lipid‐driven hepatic injury. ⁴⁷

Figure 3.

Effects of pomegranate seed oil (PSO) intervention on serum indicators in a palmitic acid (PA)‐induced mouse non‐alcoholic fatty liver disease model. *P < 0.05, **P < 0.01 versus PA group. (A) Alanine aminotransferase (ALT) levels. (B) Aspartate aminotransferase (AST) levels. (C) Alkaline phosphatase (ALP) levels. (D) Total cholesterol (TC) content. (E) Triglyceride (TG) content. (F) Low‐density lipoprotein cholesterol (LDL‐c) content. (G) High‐density lipoprotein cholesterol (HDL‐c) content. (H) Free cholesterol (FC) levels. (I) Free fatty acid (FFA) levels. (J) Malondialdehyde (MDA) levels. (K) Total sulfhydryl (—SH) content. (L) 4‐Hydroxynononal (4‐HNE) activity. (M) Xanthine oxidase (XO) activity. (N) Tumor necrosis factor‐α (TNF‐α) levels. (O) Interleukin‐1β (IL‐1β) levels. (P) Interleukin‐6 (IL‐6) levels.

Oxidative stress represents a critical pathological link in NAFLD progression from simple steatosis to NASH. ⁴⁸ Lipid peroxidation‐derived aldehydes, such as MDA and 4‐HNE, are highly reactive species capable of modifying proteins, lipids, and DNA, thereby activating inflammatory signaling pathways, promoting hepatic stellate cell activation, and facilitating hepatic fibrosis. ⁴⁹ Oxidative stress was pronounced, with significant elevations in MDA, XO, and 4‐HNE, accompanied by a marked depletion of total sulfhydryl groups (Fig. 3(J)–(M)). In parallel, inflammatory responses were activated, as indicated by significant increases in TNF‐α and IL‐1β/IL‐6 levels (Fig. 3(N)–(P)). The PA injection model has been reported to exhibit advantages such as a shortened modeling period and high reproducibility, rendering it more suitable for rapid drug efficacy screening and mechanistic investigations. ⁵⁰

PSO intervention exerted broad protective effects. Both low‐ and high‐dose PSO significantly reduced AST levels and FC concentrations (Fig. 3(A),(H)), indicating improvements in hepatic function and lipid metabolism. PSO treatment also attenuated oxidative stress by significantly decreasing XO activity and restoring total sulfhydryl levels (Fig. 3(K),(M)). Moreover, PSO‐L significantly reduced TNF‐α levels, while both PSO doses markedly suppressed IL‐1β and IL‐6 levels (Fig. 3(N)–(P)). In contrast, PSO did not significantly affect ALT, FFA, or 4‐HNE levels.

Hepatic biochemical parameters in PA‐induced liver injury model

PA injection effectively disrupted hepatic lipid metabolism and provoked oxidative stress‐mediated injury in mice.^{51, 52} Liver pathology showed mild injury (modest ALT rise, no AST change; Fig. 4(A),(B)), marked lipid accumulation (elevated TC, TG, FC, FFA; Fig. 4(C)–(F)), enhanced oxidative stress (increased MDA and XO; Fig. 4(G),(I)), indicating lipid peroxidation and free radical surge; and activated inflammation (raised TNF‐α, IL‐1β/6; Fig. 4(J)–(L)). PSO intervention exerted multi‐target protective effects at the hepatic tissue level. Both PSO‐L and PSO‐H significantly reduced hepatic TC, TG, and FC contents (Fig. 4(C)–(E)), alleviating lipid accumulation, whereas hepatic FFA levels were not significantly altered (Fig. 4(F)). PSO treatment also enhanced antioxidant capacity by significantly decreasing MDA levels and XO activity while restoring total sulfhydryl content (Fig. 4(G)–(I)). In addition, PSO markedly suppressed TNF‐α and IL‐1β levels, demonstrating pronounced anti‐inflammatory effects (Fig. 4(J), (K)). Collectively, these findings indicate that PSO selectively modulates hepatic lipid metabolism and lipid peroxidation‐related pathways, exhibiting hepatoprotective efficacy comparable to that of fish oil at the tissue level. Collectively, these findings indicate that PSO confers multi‐target protective effects against PA‐induced hepatic injury with a favorable safety profile, as supported by normal biochemical parameters in the PSO‐C group.

Figure 4.

Effects of pomegranate seed oil (PSO) intervention on liver tissue indicators in a palmitic acid (PA)‐induced mouse non‐alcoholic fatty liver disease model. *P < 0.05, **P < 0.01 versus PA group. (A) Alanine aminotransferase (ALT) levels. (B) Aspartate aminotransferase (AST) levels. (C) Total cholesterol (TC) content. (D) Triglyceride (TG) content. (E) Free cholesterol (FC) levels. (F) Free fatty acid (FFA) levels. (G), Malondialdehyde (MDA) levels. (H) Total sulfhydryl (—SH) content. (I) Xanthine oxidase (XO) activity. (J) Tumor necrosis factor‐α (TNF‐α) levels. (K) Interleukin‐1β (IL‐1β) levels. (L) Interleukin‐6 (IL‐6) levels.

GC‐Q‐TOF‐MS‐based serum metabolomics profiling

The metabolic effects of PSO on PA‐induced hepatic injury were investigated using GC‐Q‐TOF‐MS‐based metabolomics (Supporting Information, Fig. S1). ⁵³ PCA of 30 samples (n = 6 per group) revealed a discernible clustering trend among experimental groups, with PC1 and PC2 explaining 17.2% and 11.8% of the total variance, respectively (cumulative variance: 28.94%; Fig. 5(A)). Although the cumulative variance explained by the first two principal components was modest, as is common in high‐dimensional metabolomics datasets, a visible separation tendency between the control and PA groups was observed, suggesting PA‐induced alterations in the overall metabolic (Supporting Information, Table S1) profile. PSO‐treated groups exhibited a distinct metabolic distribution and a tendency to shift toward the control group (Supporting Information, Tables S2 and S3), indicating partial restoration of PA‐disrupted metabolic profiles. To further delineate group differences and identify metabolites of biological relevance, supervised multivariate analysis was subsequently performed.

Figure 5.

Metabolomic analysis reveals palmitic acid (PA)‐induced metabolic alterations and pomegranate seed oil (PSO)‐mediated restoration. (A) Principal component analysis score plot displaying global metabolic profiles with quality control sample clustering. (B) Orthogonal partial least squares discriminant analysis (OPLS‐DA) score plot of control versus PA groups. (C) OPLS‐DA score plot of PA versus PSO groups. (D) Venn diagram showing the distribution of differential metabolites among control, PA, and PSO groups. (E, F) Volcano plots of differential metabolites in control versus PA (E) and PA versus PSO (F) comparisons. PSO‐C, PSO control; FO, fish oil; QC, quality control.

Multivariate statistical analysis was performed using both PLS‐DA and OPLS‐DA to evaluate group‐level metabolic separation. The control versus PA comparison yielded a robust OPLS‐DA model (R ² X = 0.171, R ² Y = 0.922, Q ² = 0.735), confirming pronounced metabolic perturbation induced by PA treatment. Similarly, the FO versus PA comparison demonstrated clear intergroup separation in both models. In contrast, neither PLS‐DA nor OPLS‐DA achieved statistically significant discrimination for the PSO versus PA comparison, suggesting that PSO intervention did not produce globally distinct metabolic reprogramming relative to the PA group at the whole‐profile level. This pattern is consistent with a partial, rather than complete, metabolic restoration. Accordingly, subsequent differential metabolite identification was prioritized based on feature‐level statistical significance rather than overall model discriminability. Using stringent criteria (FC > 1.2 or < 0.83, P < 0.05, VIP > 1.0), Venn diagram analysis identified 198 differential metabolites (control vs. PA: 69; PSO vs. PA: 68; FO vs. PA: 61), with 27 common metabolites representing core alterations (Fig. 5(D)). Volcano plot analysis showed that PSO induced 68 differential metabolites (18 upregulated, 50 downregulated; Fig. 5(E),(F)), demonstrating extensive modulation with pronounced downregulatory effects and mechanistic divergence from fish oil.

The 27 common differential metabolites comprised fatty acids (five, 18.5%), amino acids (six, 22.2%), organic acids (five, 18.5%), sugars and sugar acids (four, 14.8%), amino acid derivatives (three, 11.1%), purine metabolites (two, 7.4%), and steroid hormones and vitamins (one each, 3.7%) (Fig. 6(A),(B)). Five fatty acids – butyric acid, palmitoleic acid, stearic acid, oleic acid, and docosanoic acid – indicated that PA disrupted fatty acid homeostasis, which PSO effectively restored. ⁵⁴

Figure 6.

Pathway enrichment analysis reveals pomegranate seed oil (PSO)‐mediated reversal of palmitic acid (PA)‐induced metabolic dysregulation. (A) Chemical classification of differential metabolites identified across experimental groups. (B) Heatmap with hierarchical clustering of differential metabolites. (C) KEGG pathway enrichment analysis highlighting oxidative stress‐related pathways. (D) Bubble plot of enriched KEGG pathways based on differential metabolites.

KEGG pathway enrichment revealed that 76.9% (10/13) of enriched pathways highly correlated with oxidative stress, identifying it as the central mechanism (Fig. 6(C)). Integrated pathway analysis elucidated PSO's multifaceted mechanisms: (i) enhanced antioxidant defense (upregulated cysteine, glutamic acid, hypotaurine, α‐tocopherol); (ii) reduced oxidative damage (decreased uric acid, xanthine); (iii) restored lipid homeostasis; (iv) normalized energy metabolism (TCA cycle, glycolysis/gluconeogenesis); (v) anti‐inflammatory effects (cortisone regulation); and (vi) modulated amino acid metabolism (phenylalanine, tyrosine) supporting tissue repair (Fig. 6(D)).

Effect of PSO on PA‐induced hepatocellular injury

The liver serves as the center of lipid metabolism, with hepatocytes playing a pivotal role in lipid uptake, storage, and catabolism. Owing to their stable phenotype and metabolic characteristics, human L02 hepatocytes are widely used to model lipid‐induced oxidative injury and lipotoxicity associated with metabolic disorders, including NAFLD and fatty liver disease. ⁵⁵ Cell viability assays demonstrated that PSO and FO at concentrations up to 50 μg mL⁻¹ maintained cell viability above 80%, indicating acceptable cytocompatibility for subsequent experiments (Fig. 7(A),(B)). PA exposure induced significant cytotoxicity at concentrations exceeding 500 μmol L⁻¹; therefore, 400 μmol L⁻¹ was selected to establish the lipotoxic injury model (Fig. 7(C)).

Figure 7.

Cell activity. **P < 0.01, compared with L02. (A) pomegranate seed oil (PSO) cell activity. (B) Fish oil (FO) cell activity. (C) Palmitic acid (PA) cell activity.

PA treatment markedly reduced L02 cell viability and triggered pronounced oxidative stress, as evidenced by elevated levels of NO, total and mitochondrial ROS, H₂O₂, O₂⁻, and MDA (Fig. 8(A)–(I)). Concurrently, plasma membrane integrity was compromised, indicated by increased PI uptake, enhanced LDH release, and altered membrane fluidity (Fig. 8(J)–(L)). PSO intervention conferred dose‐dependent cytoprotection, significantly reducing LDH leakage and PI incorporation while suppressing oxidative stress markers, including NO, ROS, H₂O₂, and MDA (Fig. 8(C)–(E),(I)‐(K)). PSO at 50 μg mL⁻¹ notably restored membrane fluidity (Fig. 8(L)), suggesting that PSO mitigates PA‐induced hepatocyte injury primarily by attenuating oxidative stress and preserving membrane integrity, although its superoxide anion scavenging capacity was limited (Fig. 8(F)–(H)). FO treatment exhibited comparable protective effects to high‐dose PSO, likely attributable to their shared enrichment in unsaturated fatty acids with known antioxidant and anti‐inflammatory properties. ⁵⁶

Figure 8.

Lipotoxicity protection activity of pomegranate seed oil (PSO). *P < 0.05, ** P < 0.01, compared with palmitic acid (PA) group. (A) MTT method to evaluate the protective activity of PSO against PA lipotoxicity. (B) Methylene blue staining method to evaluate the protective activity of PSO against PA lipotoxicity. (C) Effect of PSO on NO content induced by PA. (D) Effect of PSO on reactive oxygen species (ROS) content induced by PA. (E) Effect of PSO on H₂O₂ content induced by PA. (F) Effect of PSO on mitochondrial ROS content induced by PA. (G) Effect of PSO on O²⁻ content induced by PA. (H) Effect of PSO on O²⁻ content induced by menedione. (I) Effect of PSO on MDA content induced by PA. (J) Effect of PSO on lactate dehydrogenase (LDH) content induced by PA. (K) Propidium iodide (PI) method was used to detect the effect of PSO on the cell membrane of L02 model induced by PA. (L) Effect of PSO on the membrane fluidity of L02 cell model induced by PA was detected by Laurdan method. FO, fish oil; GP, generalized polarization.

Collectively, these findings demonstrate that PSO exerts multifaceted protective effects against PA‐induced hepatocyte damage through oxidative stress suppression and membrane stabilization. Consistent with in vivo results, PSO significantly alleviated hepatic steatosis, improved liver function, reduced lipid peroxidation, enhanced antioxidant defenses, and attenuated inflammatory responses in both cellular and animal models. These results provide mechanistic evidence supporting PSO as a promising natural agent for NAFLD management and related metabolic disorders.

DISCUSSION

Vegetable oils contain fatty acids with diverse biological activities conferred by chain length, saturation, and double bond configuration. PUFAs maintain membrane fluidity, regulate cholesterol metabolism, and modulate oxidative stress, inflammation, and endoplasmic reticulum stress, providing cardiovascular and neuroprotective benefits.^{57, 58} Punicic acid, the primary component of PSO, is a unique conjugated fatty acid conferring distinct physicochemical properties and biological activities, including anti‐inflammatory, anticancer, insulin‐sensitizing, and cardiovascular protective effects.^{59, 60, 61, 62, 63}

PSO obtained via supercritical CO₂ extraction (yield ~8%) comprised predominantly C18:3 (74.47%, mainly punicic acid), with total PUFAs ~90%, providing the material basis for its biological activities. ⁶⁴ The PA injection model successfully recapitulated NAFLD pathological features: hepatomegaly, hepatic steatosis, dyslipidemia, oxidative stress (elevated MDA, 4‐HNE, XO; decreased —SH), and inflammation (elevated TNF‐α, IL‐1β, IL‐6). In this study, NAFLD was induced in C57BL/6 mice via tail vein and intraperitoneal injection of PA – a model characterized by a short induction period, high reproducibility, and clear hepatic lipid accumulation and inflammatory phenotypes, making it well suited for rapid screening of bioactive natural products. However, certain limitations warrant consideration. The pathophysiological basis of PA‐induced hepatotoxicity differs from human NAFLD, which typically develops through a chronic, multifactorial process encompassing progressive steatosis, insulin resistance, and fibrosis driven by prolonged dietary excess and genetic susceptibility. In contrast, acute PA administration rapidly elevates circulating saturated fatty acids, triggering oxidative stress and inflammation without recapitulating the chronic metabolic remodeling characteristic of human disease. Consequently, this model may overestimate the contribution of acute lipotoxicity while underestimating the roles of chronic lipid dysregulation and insulin resistance. Future studies employing chronic dietary models, such as HFD or CDAA/HFHC diet‐induced NAFLD, are warranted to validate the long‐term efficacy of PSO intervention and more comprehensively evaluate its translational potential.

PSO demonstrated potent antioxidant activity, significantly reducing hepatic MDA and XO while restoring —SH content. In L02 hepatocytes, PA (>500 μmol L⁻¹) induced oxidative stress, decreased viability, increased NO, ROS, H₂O₂, O₂ ⁻ and MDA production, impaired membrane integrity (increased LDH release and PI fluorescence), and reduced membrane fluidity.^{65, 66} High‐dose PSO (15–50 μg mL⁻¹) significantly attenuated these effects. PSO also suppressed pro‐inflammatory cytokines, potentially through NF‐κB and inflammasome pathway inhibition.^{67, 68} GC‐Q‐TOF‐MS metabolomics revealed distinct metabolic phenotype separation between PA and control groups, significantly ameliorated by PSO intervention. Twenty‐seven common differential metabolites were identified, with KEGG analysis showing 76.9% (10/13) of enriched pathways highly correlated with oxidative stress.^{69, 70} PSO upregulated antioxidant metabolites (cysteine, glutamic acid, hypotaurine, α‐tocopherol) while downregulating oxidative markers (uric acid, xanthine), demonstrating multi‐target, multi‐pathway hepatoprotective mechanisms. ⁷¹ These findings support PSO's therapeutic potential for NAFLD management, warranting molecular target elucidation.

The KEGG pathway enrichment analysis revealed that the predicted metabolic pathways do not function in isolation but rather converge into an interconnected regulatory network. At the core of this network lies a self‐reinforcing triad of oxidative stress, NF‐κB‐mediated inflammation, and mitochondrial dysfunction. Specifically, the perturbation of fatty acid metabolism, as evidenced by the altered levels of oleic acid, palmitoleic acid, and stearic acid, may impose an excessive substrate burden on mitochondrial β‐oxidation, thereby promoting electron leakage from the electron transport chain and subsequent overproduction of ROS. Concurrently, the accumulation of succinate, a key TCA cycle intermediate identified among the differential metabolites, has been reported to stabilize HIF‐1α and potentiate NF‐κB‐driven pro‐inflammatory signaling. Under conditions of sustained oxidative stress, the Nrf2/ARE antioxidant defense system may become progressively overwhelmed, as reflected by the dysregulation of cysteine–hypotaurine metabolism and altered α‐tocopherol levels observed in this study. The resultant imbalance between ROS generation and antioxidant capacity further activates the IKK/NF‐κB cascade, amplifying the release of pro‐inflammatory cytokines such as TNF‐α and IL‐1β, which in turn exacerbate mitochondrial membrane permeabilization and perpetuate lipid metabolic disorders.

To further substantiate these mechanistic inferences derived from the metabolomic profiling, subsequent in vitro experiments were conducted using a PA‐induced hepatocyte steatosis model. Given that metabolomics data indicate that excessive production of ROS and mitochondrial dysfunction are the main pathological events, various methods were used to assess the intracellular ROS levels in order to evaluate the degree of oxidative stress; the integrity of the cell membrane was determined by LDH/PI release test and Laurdan staining method to quantify cell damage. These complementary approaches enabled verification at the cellular level of whether PSO exerts its hepatoprotective effects through modulation of the ROS‐NF‐κB signaling axis and preservation of mitochondrial membrane homeostasis.

Collectively, the integrated metabolomic and cellular evidence presented herein supports the therapeutic potential of PSO in attenuating NAFLD progression through multi‐target modulation of oxidative stress, inflammatory signaling, and mitochondrial function. Nevertheless, several limitations warrant consideration. The current findings are derived from animal models and in vitro systems, and the extent to which these mechanisms translate to human NAFLD pathophysiology remains to be determined. Furthermore, the specific molecular targets through which PSO's bioactive constituents, particularly punicic acid, interact with the identified signaling nodes require elucidation through targeted approaches such as gene silencing or pharmacological inhibition studies. Future clinical investigations are warranted to validate the translational relevance of these findings.

CONCLUSIONS

This study demonstrates that PSO exerts significant hepatoprotective effects against PA‐induced NAFLD through coordinated regulation of oxidative stress, lipid metabolism, and inflammatory signaling. Integrated GC‐Q‐TOF‐MS‐based metabolomics with biochemical and cellular validation revealed 27 differential metabolites mainly involved in glutamate, fatty acid, and taurine metabolic pathways. PSO intervention markedly ameliorated hepatic steatosis, improved liver function, suppressed lipid peroxidation, enhanced antioxidant defense, and attenuated inflammatory responses in both mouse and cellular models. These findings provide mechanistic insights into the protective effects of PSO and highlight its potential as a natural therapeutic agent for NAFLD. Further studies focusing on clinical validation and molecular target identification are warranted to promote translational application.

CONFLICT OF INTEREST

The authors declare that they have no known competing financial interests or personal relationships that could influence the work reported in this study.

AUTHOR CONTRIBUTIONS

Ma Xueping: conceptualization; methodology; software; writing – original draft. Cheng Feng, Lu Xiuxiang: data curation; investigation. Abula Reyanggu, Bakri Mahinur, Wu Tao: software. Maitinuer Maiwulanjiang: supervision; funding acquisition; writing – review and editing.

Supporting information

Figure S1. Representative total ion current chromatograms of Samples. (A) Representative total ion current chromatograms of a control sample. (B) Representative total ion current chromatograms of a PA sample. (C) Representative total ion current chromatograms of a PSO sample. (D) Representative total ion current chromatograms of a FO sample. (E) Representative total ion current chromatograms of a PSO‐C sample.

Figure S2. Heat maps of differentially abundant metabolites.

Figure S3. Clustering situation of the QC samples in the PCA graph (QC: green; PA: red; control: blue).

Table S1. Differential metabolites between control and PA.

Table S2. Differential metabolites between PSO and PA.

Table S3. Differential metabolites between FO and PA.

Table S4. Differential metabolites between control and PA.

Table S5. Differential metabolites between PSO and PA.

Table S6. Differential metabolites between FO and PA.

DATA AVAILABILITY STATEMENT

The data that support the findings of this study are available from the corresponding author upon reasonable request.