Citrus depressa peel extract and its polymethoxyflavones mitigate acetaminophen‐induced chronic liver injury in mice through antioxidant activation, epigenetic regulation, and gut microbiota modulation

Abstract

BACKGROUND

Preventing chronic liver damage is essential for maintaining overall health and reducing the risk of serious liver diseases. Acetaminophen (APAP) overdose can lead to severe liver damage due to oxidative stress, and citrus fruits are rich in bioactive compounds that may have health benefits. The present study aimed to examine the hepatoprotective effects of Citrus depressa peel ethanolic extract (CD‐EE) and its major compounds, nobiletin (Nob) and tangeretin (Tan), in BALB/c mice suffering from chronic liver injury caused by APAP.

RESULTS

CD‐EE [250 and 500 mg kg⁻¹ body weight (bw)], Nob (50 mg kg⁻¹ bw) and Tan (50 mg kg⁻¹ bw), or N‐acetylcysteine (NAC, 600 mg kg⁻¹ bw, as a positive control), were orally administered to the mice daily for 9 weeks. Subsequently, starting from the second week, mice received intraperitoneal injections of APAP (400 mg kg⁻¹ bw) twice weekly. We found that APAP‐induced liver damage was reduced through the enhancement of antioxidant enzyme activity, the reduction of lipid peroxidation and the modulation of the Nrf2‐mediated antioxidant pathway by CD‐EE, Nob, Tan and NAC treatments. The levels of DNA methyltransferases and histone deacetylases were influenced by these treatments in epigenetic regulation. Additionally, they partially restored the gut microbiota composition that was altered by APAP.

CONCLUSION

These results suggest that CD‐EE, Nob, and Tan are associated with hepatoprotection against APAP‐induced chronic liver injury in mice. The modulation of antioxidant mechanisms and epigenetic regulation potentially contributes to this protection, whereas the restoration of gut microbiota composition may represent an additional pathway. © 2026 The Author(s). Journal of the Science of Food and Agriculture published by John Wiley & Sons Ltd on behalf of Society of Chemical Industry.

INTRODUCTION

The liver plays a vital role in metabolism and detoxification by neutralizing toxins, drugs, and alcohol absorbed from the small intestine. Through a two‐phase chemical reaction involving multiple enzymes, fat‐soluble toxins are converted into water‐soluble compounds for excretion in urine or bile. ¹ Acetaminophen (APAP) is a drug that is commonly used for analgesia and pain, but excessive use can cause liver damage and hepatotoxicity. Recent research has revealed new metabolic and mitochondrial mechanisms behind APAP‐induced liver injuries, highlighting the importance of cellular energy reprogramming in addition to glutathione (GSH) depletion and oxidative stress. ² The liver metabolizes APAP, resulting in intermediate metabolites in Phase I and transforming them into APAP‐glucuronide and APAP‐sulfate in Phase II. Cytochrome P450 enzymes convert a small portion of APAP into N‐acetyl‐p‐benzoquinone imine (NAPQI). ³ NAPQI binding to cysteine groups on mitochondrial proteins results in oxidative stress and hepatocyte death when GSH is depleted. ⁴ N‐acetylcysteine (NAC) supplementation increases GSH synthesis, effectively counteracting the toxicity of APAP. ⁵ Other antioxidant enzymes, such as superoxide dismutase, convert free radicals into hydrogen peroxide, which is decomposed by catalase. Glutathione peroxidase (GPx) is responsible for reducing peroxides and converting GSH to oxidized GSH, which can be reversed by glutathione reductase (GRd). ⁶ The expression of the antioxidant system is initiated by the basic family member of leucine zipper transcription factor, nuclear factor erythroid 2‐related factor 2 (Nrf2). ⁷ Gene expressions of phase II detoxification and antioxidant enzymes, such as hemeoxygenase‐1 (HO‐1), NAD(P)H:quinone oxidoreductase 1 (NQO1) and UDP‐glucuronosyl transferase 1A (UGT1A), can be triggered by Nrf2 to protect the hepatocytes from oxidative damage caused by NAPQI. ⁴

DNA methylation, histone modifications and epitranscriptomic remodeling play significant roles in regulating inflammatory responses and regeneration in chronic and drug‐induced liver diseases. ⁸ DNA methylation typically occurs on the cytosine of adjacent CpGs, and methylated CpGs work with methyl‐DNA binding proteins to stop transcription factors from attaching to DNA. ⁹ DNA methyltransferase (DNMT) 1 maintains the current state of DNA methylation, whereas DNMTs 3a and 3b are responsible for establishing the DNA methylation pattern of the genome by acting on both hemi‐methylated and unmethylated CpG sites. ¹⁰ Moreover, it is well‐known that histones are subjected to various modifications. Histone acetyltransferases acetylate histones, which can increase the loose chromatin structure consisting of histones and DNA. ¹¹ When histone deacetylases (HDACs) cause histone deacetylation, the chromatin structure will tighten to cause gene silencing. ¹²

The gut–liver axis theory examines how gut microbiota and liver diseases, such as viral hepatitis, alcoholic hepatitis, non‐alcoholic fatty liver disease, liver fibrosis and liver cancer, are connected. ¹³ APAP‐induced liver injury was also confirmed by changes in gut microbiota composition and metabolite signaling. ¹⁴ A diet supplemented with phytochemicals can alter the imbalance of gut microbiota induced by high‐fat feeding, resulting in a decrease in liver damage caused by inflammatory responses in non‐alcoholic fatty liver disease. ¹⁵ Dietary changes may impact the toxicity of APAP metabolites in acute liver injury caused by APAP. ¹⁶

Citrus depressa Hayata (CD), Japanese name Shiikuwasha, is distributed in the mountains of southeastern Japan, Okinawa and Taiwan. ¹⁷ The most important active ingredients are flavonoids, including flavanone (such as naringenin and hesperetin), flavonoid glycosides (such as naringin and hesperidin) and polymethoxyflavonoids (PMFs). ¹⁸ Nobiletin (Nob) and tangeretin (Tan) have a variety of biological functions, including neuroprotection, antioxidant, antitumor and anti‐inflammatory effects.^{19, 20} Recent evidence highlights that Nob protects against drug‐induced liver injury by mitigating oxidative stress, whereas Tan exerts similar hepatoprotection by modulating Nrf2‐related antioxidant pathways.^{21, 22} The present study aimed to assess how the CD peel ethanolic extract (CD‐EE) and its primary compounds, Nob and Tan, protect against chronic hepatic injury induced by APAP in mice. Additionally, we examined the underlying mechanisms, including the Nrf2‐mediated antioxidant pathway, epigenetic regulation, and alterations in the gut microbiota.

MATERIALS AND METHODS

Reagents and materials

CD was obtained from a fruit farm located in Pingtung, Taiwan. AB DIET and birch wood mixed litter were purchased from WF Fisher & Son (Somerville, NJ, USA) and Tapvei Estonia OÜ (Harjumaa, Estonia), respectively. Methanol, 95% ethanol, aluminum chloride and potassium acetate were purchased from Jingming Chemical (Miaoli, Taiwan). AP, Nob, Tan, Triton X‐100, sodium dodecyl sulfate (SDS), acetic acid, Folin–Ciocalteu's phenol reagent, sodium carbonate, gallic acid and quercetin were purchased from Sigma‐Aldrich (St Louis, MO, USA). Synephrine, naringin, hesperidin and naringenin were purchased from TCI (Tokyo, Japan). Neohesperidin was obtained from Tauto Biotech (Shanghai, China). Nob, Tan, 4′‐hydroxy‐5,6,7,8‐tetramethoxyflavone and 5,7,3′,4′‐tetramethoxyflavone were synthesized by Dr Guor‐Jien Wei (Institute of Food Safety and Health Risk Assessment, National Yang Ming Chiao Tung University, Taipei, Taiwan).

Preparation of CD‐EE

Fresh CD fruits were washed, peeled and dried. The dried peel was finely ground and sieved. The dry peel powder and 95% ethanol (ratio 1:40; g mL⁻¹) were mixed at room temperature for 24 h. The extract was then filtered to obtain CD‐EE. After concentrating, the extract was subjected to vacuum freeze‐drying and stored as dried CD‐EE at −20 °C. The extraction yield of CD‐EE is 2.307 g kg⁻¹ dried CD‐EE peel.

Total phenolic content

The CD‐EE was dissolved in 95% methanol to prepare a 5 mg mL⁻¹ sample solution. Standard solutions of gallic acid were prepared. Then, 50 μL of each sample or standard solution was mixed with 100 μL of Folin–Ciocalteu's phenol reagent and 400 μL of 700 mm sodium carbonate. The mixture was allowed to react at room temperature for 1 h. Afterward, 200 μL of the resulting solution was transferred to a 96‐well plate, and absorbance was measured at 765 nm. The outcomes were presented as milligrams of gallic acid equivalent (GAE) per gram of dried CD‐EE.

Total flavonoid content

The CD‐EE was reconstituted in 95% methanol to prepare a 5 mg mL⁻¹ sample solution. Standard solutions of quercetin were also prepared. Then, 50 μL of each sample or standard solution was sequentially mixed with 350 μL of methanol, 20 μL of 10% aluminum chloride and 20 μL of 1.0 m potassium acetate, followed by 500 μL of distilled water. The mixture was thoroughly vortexed and reacted at room temperature for 30 min. Next, 200 μL of the supernatant was transferred into a 96‐well plate, and the absorbance was measured at 415 nm. The results were presented as milligrams of quercetin equivalent (QE) per gram of dried CD‐EE.

HPLC analysis

CD‐EE was dissolved in methanol to prepare a 10 000 ppm solution. The mobile phases consisted of methanol (solvent A) and water containing 1% acetic acid (solvent B). The HPLC system was equipped with a pump (Pu‐2080 Plus; JASCO, Tokyo, Japan) and UV‐visible light detector (UV‐2075 Plus; JASCO), and the analytical column used was a C18 (5 μm particle size, 4.6 × 250 mm) (Agilent Technologies, Santa Clara, CA, USA) maintained at 30 °C. Each sample injection volume was 20 μL, and the flow rate was set at 0.7 mL min⁻¹. The detection wavelength was set at 280 nm. The gradient elution was as follows: methanol increased from 18% to 40% from 0 to 20 min, then increased from 40% to 50% from 20 to 40 min, further increased from 50% to 80% from 40 to 45 min, and maintained at 80% from 45 to 50 min. Standard calibration curves (12.5–100 ppm) were utilized for the quantification of synephrine, naringin, hesperidin, neohesperidin, naringenin, 4′‐hydroxy‐5,6,7,8‐tetramethoxyflavone, 5,7,3′,4′‐tetramethoxyflavone, Nob and Tan in the CD‐EE.

Animal experimental design

The animal experimental protocols employed in the present study were reviewed and approved by the Institutional Animal Care and Use Committee (IACUC) of Chung Yuan Christian University under approval number 105032. Four‐week‐old BALB/c mice were purchased from the National Laboratory Animal Center (Taipei, Taiwan) and were housed in cages in the Department of Bioscience Technology's animal facility at Chung Yuan Christian University (Taoyuan, Taiwan). The mice were maintained under a 12:12 h light/dark photocycle, at a room temperature of 22 ± 2 °C and a relative humidity of 50–70%. They were provided ad libitum access to food and water. The experiment was conducted when the mice were 6 weeks old. A total of 56 mice were randomized into seven groups, with eight mice per group, consistent with our previous experience and published APAP‐induced liver injury studies, in which six to eight animals per group provided sufficient power to detect biologically relevant changes in serum aspartate aminotransferase (AST)/alanine aminotransferase (ALT) and hepatic oxidative stress markers.^{23, 24} In the experimental groups, Group A served as the normal control group, Group B comprised the APAP‐treated negative control group and Group C (APAP+NAC) was the positive control group; Groups D to G (APAP+CD‐EE at 250 or 500 mg kg⁻¹ bw, APAP+Nob and APAP+Tan) were used to evaluate the potential hepatoprotective effects of CD‐EE, Nob and Tan under the same chronic APAP exposure conditions. The experiment lasted 9 weeks, during which each group received daily oral doses of the samples. From the second week onwards, Group A was injected intraperitoneally with 0.9% NaCl solution, whereas Groups B to G received intraperitoneal injections of 400 mg kg⁻¹ bw APAP twice a week for 8 weeks. At the end of the experiment, all mice were euthanized. During euthanasia, blood was collected from the heart, and the liver, spleen and kidneys were harvested and weighed. Liver sections were examined qualitatively under light microscopy to assess APAP‐induced hepatic lesions and the apparent improvement after CD‐EE, Nob and Tan treatment. Blood samples were examined for AST, ALT, cholesterol and triglycerides. Homogenized liver tissues were prepared for the determination of antioxidant enzyme activities, including GPx, GRd, glutathione S‐transferase (GST) and catalase, as well as lipid peroxidation by measuring the levels of thiobarbituric acid‐reactive substances (TBARS) from the reaction with malondialdehyde. The procedures for histopathology, serum biochemical analyses, and hepatic antioxidant and TBARS assays are described in the Supporting Information (Data S1).

Protein expression

The liver tissue was homogenized in RIPA lysis buffer (Cell Signaling Technology, Danvers, MA, USA), followed by centrifugation at 17 000 × g for 30 min at 4 °C. The supernatant was collected as the liver protein extract and stored at −80 °C. After sonication and centrifugation at 4 °C and 17 000 × g for 30 min, the protein concentration of the lysates was determined using a bicinchoninic acid (i.e. BCA) assay kit (Energenesis Biomedical, Taipei, Taiwan). A total of 25 μg of protein was mixed with loading dye containing SDS. The mixture was heated at 95 °C for 5 min to denature the proteins. Subsequently, SDS‐polyacrylamide gel electrophoresis was performed, followed by the transfer of proteins onto a polyvinylidene difluoride (PVDF) membrane. After the transfer, the membrane was sequentially incubated with primary and secondary antibodies, and a chemiluminescent substrate was used to visualize the proteins on the PVDF membrane, which was then photographed. The primary antibodies targeted proteins including Nrf2, HO‐1, NQO1, DNMT3a, DNMT3b and HDAC1 (ABclonal, Woburn, MA, USA); UGT1A, HDAC2 and HDAC3 (GeneTex, Irvine, CA, USA); DNMT1 (Santa Cruz, Dallas, TX, USA); and GAPDH (Proteintech, Rosemont, IL, USA). The secondary antibodies (ABclonal) were specific to either mouse or rabbit IgG. Protein bands were visualized using an enhanced chemiluminescence detection kit (Bio‐Rad, Hercules, CA, USA) and quantified by densitometry using ImageJ software (NIH, Bethesda, MD, USA). For western blot quantification, liver protein extracts from six individual mice per group (n = 6) were analyzed. The representative blot images shown in the Results correspond to two mice from each group.

RNA expression

Liver tissue RNA extraction was carried out using a Total RNA Isolation Kit from Genedirex (Taoyuan, Taiwan). Subsequently, mRNA was reverse transcribed into first‐strand cDNA templates using the iScript cDNA synthesis kit (Bio‐Rad). Real‐time quantitative PCR (RT‐qPCR) was performed on a CFX Connect™ Real‐Time PCR Detection System (Bio‐Rad) using SYBR® Green Master Mix (Bio‐Rad) to determine the expression levels of nrf2 and its downstream antioxidant and detoxifying enzyme genes. The primer sequences (5′‐ to 3′) for nrf2, ho‐1, nqo1 and ugt1a1, as well as the reference gene, are listed in Supporting information (Table S1). The primers were synthesized by Integrated DNA Technologies (Clareville, IA, USA). Three independent liver samples per group (n = 3) were used as biological replicates and each sample was measured in technical duplicate.

Gut microbiota

DNA/RNA Shield buffer (ZYMO Research, Tustin, CA, USA) was used to collect and store fresh fecal samples. The DNA was extracted from the fecal samples using the innuPREP Stool DNA Kit (Jena, Germany) in accordance with to the manufacturer's recommendations. The V3–V4 hypervariable region of bacterial 16S rRNA genes was targeted by primers used for PCR on the Illumina MiSeq NGS system (Illumina, San Diego, CA, USA). The amplified segments contain a portion of this highly variable region, which is used to identify and analyze bacterial species after sequencing. Paired‐end reads were trimmed using Trimmomatic, version 0.39 (https://trimmomatic.com) and demultiplexed by an in‐house script. Primer sequences were removed using Cutadapt (https://cutadapt.readthedocs.io). QIIME classified reads into OTUs in the GreenGenes database (http://greengenes.lbl.gov). Analysis included Venn diagrams, KRONA plots, ²⁵ alpha and beta diversity, linear discriminant analysis effect size and PICRUSt (https://picrust.github.io/picrust/) for functional predictions using Kyoto Encyclopedia of Genes and Genomes (KEGG) (https://www.genome.jp/kegg) ortholog groups. STAMP (https://github.com/donovan-h-parks/STAMP) identified significant differences (P < 0.05) in relative abundance. ²⁶ For each experimental group, fecal samples from three individual mice (n = 3) were collected and analyzed as biological replicates.

Statistical analysis

The experimental results are presented as the mean ± SD. Differences among experimental groups were analyzed using one‐way analysis of variance, followed by Duncan's new multiple range test for post‐hoc comparisons. Statistical analyses were performed using SAS software (SAS Institute, Cary, NC, USA) P < 0.05 was considered statistically significant.

RESULTS

CD‐EE contains phenolic and flavonoid compounds

A diet rich in polyphenols and flavonoids is associated with a reduced risk of chronic diseases. ²⁷ In the present study, CD‐EE was analyzed for total phenolic and flavonoid contents (Table 1). CD‐EE exhibited a total phenolic content of 30.31 ± 2.09 mg GAE g⁻¹ and a total flavonoid content of 17.91 ± 0.95 mg QE g⁻¹. HPLC analysis further revealed hesperidin (54.88 ± 0.21 mg g⁻¹), synephrine (11.00 ± 0.51 mg g⁻¹), naringin (0.85 ± 0.03 mg g⁻¹), neohesperidin (0.49 ± 0.02 mg g⁻¹) and naringenin (0.07 ± 0.03 mg g⁻¹), as well as PMFs including Nob (81.21 ± 0.10 mg g⁻¹), Tan (25.41 ± 0.08 mg g⁻¹), 4′‐hydroxy‐5,6,7,8‐tetramethoxyflavone (8.70 ± 0.04 mg g⁻¹) and 5,7,3′,4′‐tetramethoxyflavone (1.17 ± 0.02 mg g⁻¹) (Fig. 1 and Table 2). These findings indicate that CD‐EE is abundant in phenolic and flavonoid compounds that could be involved in hepatic antioxidant and detoxification responses.

Table 1.

Total phenol and total flavonoid contents in CD‐EE ^a

	CD‐EE
Total phenols (mg GAE g−1 dried extract) b	30.31 ± 2.09
Total flavonoids (mg QE g−1 dried extract) c	17.92 ± 0.95

^a Data are shown as the mean ± SD (n = 3).

^b GAE: gallic acid equivalent.

^c QE: quercetin equivalent.

Figure 1.

HPLC profiles of standards (A) and CD‐EE (B).

Table 2.

Content of major compounds in CD‐EE ^a

Peak number	Compound	Retention time (min)	Content in CD‐EE (mg g−1 dried extract)
1	Synephrine	3.4	11.00 ± 0.51
2	Naringin	10.8	0.85 ± 0.03
3	Hesperidin	11.2	54.88 ± 0.21
4	Neohesperidin	11.7	0.49 ± 0.02
5	Naringenin	17.4	0.07 ± 0.03
6	4′‐Hydroxy‐5,6,7,8‐tetramethoxyflavone	19.1	8.70 ± 0.04
7	5,7,3′,4’‐Tetramethoxyflavone	20.4	1.17 ± 0.02
8	Nobiletin	23.1	81.21 ± 0.10
9	Tangeretin	25.6	25.41 ± 0.08

^a Data are shown as the mean ± SD (n = 3).

Effect of CD‐EE, Nob and Tan on the body weights and relative organ weights of mice with APAP‐induced liver injury

Table 3 summarizes body weight and relative organ weights. There were no significant differences in body weight among groups at either the start or the end of the 9‐week feeding period. By contrast, relative liver weight was significantly higher in all APAP‐treated groups, including those receiving NAC, CD‐EE, Nob or Tan, compared to the control group (P < 0.05), reflecting APAP‐induced liver swelling (approximately a 17.6% increase). Mice fed NAC (600 mg kg⁻¹ bw), CD‐EE(L) (250 mg kg⁻¹ bw), CD‐EE(H) (500 mg kg⁻¹ bw), Nob (50 mg kg⁻¹ bw) or Tan (50 mg kg⁻¹ bw) showed similarly elevated relative liver weights compared to the control group, with slightly lower values in the NAC, CD‐EE(H), Nob and Tan groups than in the APAP group. Relative kidney and spleen weights did not differ significantly among groups.

Table 3.

Effect of CD‐EE, Nob and Tan on body weights, relative weights of organs, serum biochemical analysis, and liver oxidant stress and antioxidant enzyme activities in APAP‐treated mice ^{a , b , c}

	Control	APAP	APAP+NAC	APAP+CD‐EE(L)	APAP+CD‐EE(H)	APAP+Nob	APAP+Tan
Body weight
Initial (g)	23.3 ± 1.1	23.1 ± 1.1	23.8 ± 1.2	22.9 ± 1.6	22.9 ± 1.0	23.5 ± 1.1	23.0 ± 1.3
Final (g)	27.2 ± 1.9	26.7 ± 1.4	26.0 ± 1.5	27.1 ± 1.3	26.7 ± 1.0	27.9 ± 1.7	26.7 ± 1.6
Relative organ weight
Liver (g 100 g−1 bw)	5.10 ± 0.44 b	6.00 ± 0.53 a	5.83 ± 0.40 a	6.09 ± 0.57 a	5.98 ± 0.39 a	5.87 ± 0.53 a	5.75 ± 0.52 a
Kidney (g 100 g−1 bw)	1.95 ± 0.10	1.94 ± 0.14	2.05 ± 0.17	2.04 ± 0.12	2.04 ± 0.08	2.06 ± 0.11	2.05 ± 0.09
Speen (g 100 g−1 bw)	0.43 ± 0.07	0.47 ± 0.10	0.41 ± 0.09	0.50 ± 0.09	0.42 ± 0.05	0.45 ± 0.06	0.46 ± 0.13
Serum biochemical analysis
ALT (U L−1)	49.7 ± 20.8 b	193.3 ± 107.8 a	87.5 ± 79.5 b	65.7 ± 37.1 b	53.5 ± 22.4 b	101.0 ± 59.2 b	69.4 ± 47.2 b
AST (U L−1)	104.4 ± 41.5 b	266.6 ± 93.6 a	169.2 ± 74.3 b	149.2 ± 34.1 b	132.4 ± 32.5 b	149.0 ± 45.4 b	133.4 ± 35.9 b
Cholesterol (mg dL−1)	148.2 ± 25.1	141.3 ± 31.5	146.8 ± 33.7	138.7 ± 36.3	145.5 ± 29.4	131.1 ± 38.7	136.9 ± 31.1
Triglyceride (mg dL−1)	165.3 ± 9.0	145.8 ± 22.7	172.4 ± 30.6	147.0 ± 35.7	161.9 ± 10.1	137.8 ± 32.1	157.9 ± 27.8
Liver oxidant stress
TBARS (nmol mg−1 protein)	0.275 ± 0.067 b	0.353 ± 0.076 a	0.283 ± 0.080 ab	0.282 ± 0.047 b	0.277 ± 0.046 b	0.267 ± 0.058 b	0.314 ± 0.085 ab
Liver antioxidant enzyme activities
GPx (nmol min mg−1 protein)	303.3 ± 71.9 a	224.1 ± 48.3 b	300.8 ± 74.9 a	311.7 ± 71.3 a	287.8 ± 54.5 ab	269.2 ± 72.6 ab	276.9 ± 70.0 ab
GRd (nmol min mg−1 protein)	75.4 ± 15.3 b	72.1 ± 12.9 b	157.3 ± 142.6 a	87.2 ± 25.1 b	83.4 ± 19.5 b	73.8 ± 18.1 b	76.4 ± 18.2 b
GST (nmol min mg−1 protein)	1929.7 ± 329.6 ab	1655.2 ± 351.4 b	2180.1 ± 412.2 a	2096.8 ± 332.4 ab	2175.3 ± 491.5 a	2124.7 ± 519.5 ab	1762.4 ± 493.7 ab
Catalase (μmol min mg−1 protein)	299.1 ± 74.2 ab	233.8 ± 94.7 b	360.0 ± 109.0 a	318.3 ± 102.5 ab	350.1 ± 85.1 a	303.7 ± 81.8 ab	326.5 ± 64.0 a

^a The intraperitoneal dose of APAP was 400 mg kg⁻¹ bw. N‐acetylcysteine (NAC) was orally administered at a dosage of 600 mg kg⁻¹ bw as a positive control. The oral dosages of CD‐EE (L), CD‐EE (H), Nob and Tan were 250, 500, 50 and 50 mg kg⁻¹ bw, respectively.

^b After a 9‐week experimental period, final body weight, relative organ weights, serum biochemical analysis and liver oxidant stress and antioxidant enzyme activities were assessed in BALB/c mice.

^c Data are presented as mean ± SD (n = 8). Groups with different lowercase letters indicate significant differences (P < 0.05).

Effect of CD‐EE, Nob and Tan on serum ALT and AST activities, cholesterol levels and triglyceride levels in mice with APAP‐induced liver injury

To evaluate hepatoprotection, serum ALT and AST activities were measured after 9 weeks (Table 3). APAP treatment caused a 3.9‐fold increase in ALT and a 1.6‐fold increase in AST compared to the control group. NAC, CD‐EE(L), CD‐EE(H), Nob and Tan all significantly reduced APAP‐elevated ALT and AST activities (P < 0.05), with CD‐EE(H) showing the greatest reduction in ALT (approximately 72%), followed by CD‐EE(L) (approximately 66%) and Tan (approximately 64%), whereas NAC, CD‐EE(L), CD‐EE(H), Nob and Tan reduced AST by approximately 36–50%. These data indicate that chronic high‐dose APAP injection induced marked liver injury, whereas CD‐EE, Nob and Tan mitigated this damage. By contrast, APAP and the different treatments did not significantly affect serum cholesterol or triglyceride levels compared to either the control or APAP groups.

Effects of CD‐EE, Nob and Tan on TBARS levels and antioxidant enzymes in the livers of mice with APAP‐induced liver injury

Hepatic TBARS levels and antioxidant enzyme activities are shown in Table 3. APAP significantly increased TBARS levels to approximately 0.353 nmol mg⁻¹ protein compared to the control group (P < 0.05). CD‐EE(L), CD‐EE(H) and Nob reduced TBARS to approximately 0.282, 0.277 and 0.267 nmol mg⁻¹ protein, respectively (P < 0.05), whereas NAC and Tan also lowered TBARS but did not reach statistical significance. Regarding antioxidant enzymes, APAP significantly reduced GPx activity compared to the control, whereas NAC and CD‐EE(L) significantly restored GPx activity (P < 0.05), and other treatments tended to increase GPx activity. APAP slightly decreased GRd and GST activities; NAC significantly increased GRd and GST compared to the APAP group, and CD‐EE(H) significantly elevated GST compared to the APAP group (P < 0.05), with CD‐EE(L), Nob and Tan showing non‐significant upward trends. For catalase, NAC, CD‐EE(H) and Tan significantly increased activity relative to APAP (P < 0.05), whereas CD‐EE(L) and Nob produced modest increases. Consequently, CD‐EE, Nob and Tan partially counteracted APAP‐induced oxidative stress by lowering lipid peroxidation and enhancing antioxidant enzyme activities.

Effects of CD‐EE, Nob and Tan on liver histopathology in mice with APAP‐induced liver injury

Macroscopic and histopathological changes in the liver are shown in Fig. 2. APAP injection induced liver swelling and pale brown discoloration compared to the healthy red appearance of control livers. Histologically, APAP caused nuclear disappearance, hepatocellular necrosis, inflammatory cell infiltration and vacuolization around the central vein. By contrast, livers from NAC‐, CD‐EE(L)‐, CD‐EE(H)‐, Nob‐ or Tan‐treated mice more closely resembled controls, with attenuated necrosis and inflammatory changes, supporting the protective effects of these treatments against APAP‐induced liver injury.

Figure 2.

Macroscopic and histopathological changes in livers of mice with APAP‐induced chronic liver injury and the effects of CD‐EE, Nob and Tan. (A) Representative gross appearance of livers from each group. (B) Representative hemotoxylin and eosin‐stained liver sections (200×; scale bar = 100 μm). The control group shows normal hepatic architecture with radiating hepatic cords. The APAP group displays centrilobular hepatocellular necrosis, inflammatory cell infiltration and cytoplasmic vacuolization (arrows), indicative of APAP‐induced liver injury. In the APAP+NAC, APAP+CD‐EE(L), APAP+CD‐EE(H), APAP+Nob and APAP+Tan groups, these lesions are noticeably attenuated, with more preserved hepatic architecture and fewer necrotic foci.

Effect of CD‐EE, Nob and Tan on the Nrf2 antioxidant pathway in the livers of mice with APAP‐induced liver injury

The effects of CD‐EE, Nob and Tan on the hepatic Nrf2 pathway are shown in Fig. 3. APAP injection significantly reduced the protein expression of Nrf2 and its downstream targets HO‐1, NQO1 and UGT1A compared to the control group (P < 0.05) (Fig. 3(A), (B)). CD‐EE(H) and NAC significantly increased Nrf2 protein expression relative to APAP (P < 0.05), whereas Nob and Tan caused modest increases. CD‐EE(H) significantly increased HO‐1 protein levels, whereas all treatment groups showed significantly higher NQO1 protein levels than the APAP group (P < 0.05). UGT1A protein expression was slightly increased by CD‐EE, Nob and Tan, and markedly enhanced by NAC (P < 0.05). Consistent with these protein changes, APAP lowered hepatic ugt1a1 mRNA and slightly reduced nrf2, ho‐1 and nqo1 expression (Fig. 3(C)). CD‐EE(H) and Tan significantly increased nrf2 mRNA; CD‐EE(H) significantly upregulated ho‐1; and NAC, CD‐EE(H), Nob and Tan significantly elevated nqo1 mRNA (P < 0.05). Tan also significantly increased ugt1a1 mRNA compared to APAP (P < 0.05), whereas other treatments showed non‐significant upward trends. These findings indicate that CD‐EE, Nob, and Tan can partially restore Nrf2‐mediated antioxidant gene expression in APAP‐injured livers.

Figure 3.

Effect of CD‐EE, Nob and Tan on the protein and mRNA expression of Nrf2 and antioxidation enzymes in the liver of mice treated with APAP. The relative protein and mRNA levels of Nrf2, HO‐1, NQO1 and UGT1A in the liver of BALB/c mice were measured after a 9‐week trial period. (A, B) Representative western blot images from two mice per group are shown. Data are shown as the mean ± SD from six individual mice per group (n = 6). (C) Data are shown as the mean ± SD (n = 3). There is a significant difference between groups that share different superscript letters (P < 0.05).

Effect of CD‐EE, Nob and Tan on epigenetics‐related enzymes in the livers of mice with APAP‐induced liver injury

We next examined epigenetic regulators of Nrf2, including DNMTs and HDACs (Fig. 4). APAP significantly increased hepatic protein levels of DNMT1, DNMT3a and DNMT3b compared to controls (P < 0.05). Administration of CD‐EE(L), CD‐EE(H), Nob, Tan or NAC significantly reduced DNMT1 expression relative to APAP (P < 0.05), and CD‐EE(L), CD‐EE(H), Nob and NAC also significantly lowered DNMT3a and DNMT3b expression (P < 0.05). By contrast, APAP did not significantly alter the protein levels of HDACs 1–3. NAC significantly decreased HDAC1 expression compared to APAP (P < 0.05), whereas CD‐EE, Nob and Tan had minimal effects on HDACs. These results support a role for DNMT modulation, rather than broad HDAC suppression, in the regulation of Nrf2‐related pathways by CD‐EE, Nob, Tan and NAC in APAP‐induced liver injury.

Figure 4.

Effect of CD‐EE (A), Nob and Tan (B) on the protein expression of DNA methyltransferases (DNMTs) and histone deacetylase (HDACs) in the liver of mice treated with APAP. Relative levels of DNMTs and HDACs in the liver of BALB/c mice were measured after a 9‐week trial period. Representative western blot images from two mice per group are shown. Data are shown as the mean ± SD (n = 6, from six individual mice per group). There is a significant difference between groups that share different superscript letters (P < 0.05).

Effect of CD‐EE, Nob and Tan on gut microbiota in the mice with APAP‐induced liver injury

The results of 16S rDNA sequencing of mouse fecal microbiomes are presented in Fig. 5. PCoA indicated that the APAP group had a distinct gut microbial community compared to the control group, and that the APAP+CD‐EE(H), APAP+NAC, APAP+Nob and APAP+Tan groups exhibited shifted community compositions compared to both the control and APAP groups (Fig. 5(A)–(C)). However, CD‐EE(H), NAC, Nob and Tan did not fully restore the microbiota to the control pattern.

Figure 5.

Effects of CD‐EE, NAC, Nob and Tan on gut microbiota composition and predicted functional pathways in mice with APAP‐induced chronic liver injury. (A–C) PCoA plots based on 16S rDNA sequencing showing β‐diversity among the groups. (D–F) Heatmaps of the relative abundance of major bacterial phyla (upper) and genera (lower). Data are based on fecal samples from three individual mice per group (n = 3).

The heatmap (Fig. 5(D)) showed that APAP intervention significantly altered the microbiota composition, increasing the phyla Cyanobacteria, Bacteroidetes, TM7, Tenericutes and Verrucomicrobia, at the same time as decreasing Deferribacteres, Firmicutes, Actinobacteria and Proteobacteria. Compared with APAP, CD‐EE(H) treatment further increased Cyanobacteria, Bacteroidetes, TM7 and Deferribacteres, and clearly decreased Tenericutes, Verrucomicrobia, Firmicutes, Actinobacteria and Proteobacteria at the phylum level. At the species level, APAP increased Streptococcus, Sutterella, Akkermansia, Carnobacterium, Burkholderia, Anaeroplasma, Coprobacillus and Gemella compared to the control group, whereas these taxa were reduced in the APAP+CD‐EE(H) group. By contrast, the APAP+CD‐EE(H) group displayed higher abundances of Candidatus_Arthromitus, Clostridium, Odoribacter, Allobaculum, Pseudomonas, Parabacteroides, Turicibacter and Victoria, and lower abundances of Dehalobacterium, Dorea, Adlercreutzia and Oscillospira, compared to both the control and APAP groups.

Regarding NAC treatment (Fig. 5(E)), NAC partially reversed the APAP‐induced increases in Tenericutes, Bacteroidetes, TM7, Actinobacteria and Verrucomicrobia, whereas it tended to increase Firmicutes, Cyanobacteria, Deferribacteres and Proteobacteria at the phylum level. At the genus/species level, NAC attenuated the elevated abundances of Gemella, Akkermansia, Burkholderia, Coprobacillus, Sutterella, Anaeroplasma, Oscillospira, Turicibacter, Parabacteroides and Streptococcus induced by APAP and mitigated the decreases in Desulfovibrio, Candidatus_Arthromitus and Coprococcus. NAC also increased Clostridium, Mucispirillum, Enterobacter, Odoribacter and Victoria, which were relatively low in both the control and APAP groups.

Nob and Tan exhibited broadly similar effects in counteracting APAP‐induced dysbiosis (Fig. 5(F)). Both reduced the phyla TM7, Verrucomicrobia, Bacteroidetes and Tenericutes that were increased by APAP and partly restored Firmicutes, which had been suppressed. In addition, both Nob and Tan increased Cyanobacteria and Proteobacteria, whereas Tan further elevated Actinobacteria and Deferribacteres, which were relatively low in the control and APAP groups. At the species level, APAP increased Akkermansia, Burkholderia, Anaeroplasma and Sutterella compared to the control group, whereas both Nob and Tan reduced these taxa toward control‐like levels. Nob treatment additionally suppressed APAP‐induced increases in Coprobacillus, Parabacteroides and Streptococcus. Compared with the control and APAP groups, Nob and Tan increased Desulfovibrio, Oscillospira, Bifidobacterium and Lactococcus; both also increased Turicibacter, Dehalobacterium and Odoribacter, whereas Tan increased Adlercreutzia, Gemella, Mucispirillum, Allobaculum and Clostridium.

Predicted functional KEGG pathways reflected these compositional changes. APAP administration upregulated pathways related to apoptosis, PPAR signaling and nucleotide metabolism, at the same time as downregulating fructose and mannose metabolism, glyoxylate and dicarboxylate metabolism, and linoleic acid metabolism (P < 0.05) (Fig. 6(A)). In the APAP+CD‐EE(H) group, higher levels of inorganic ion transport and metabolism, amino sugar and nucleotide sugar metabolism, sphingolipid metabolism, ion channels, cyanoamino acid metabolism, and phenylpropanoid biosynthesis were observed, accompanied by decreased biosynthesis and biodegradation of secondary metabolites, transcription‐related proteins, cytoskeleton proteins and other transporters (P < 0.05) (Fig. 6(B)). NAC treatment reduced pyruvate metabolism, glycolysis/gluconeogenesis and limonene and pinene degradation, at the same time as increasing ion channel–related functions (P < 0.05) (Fig. 6(C)). Tan treatment increased functions classified as unknown and glyoxylate and dicarboxylate metabolism but decreased glycolysis/gluconeogenesis, peptidoglycan biosynthesis, valine, leucine and isoleucine biosynthesis, tropane/piperidine/pyridine alkaloid biosynthesis, methane metabolism, lipid biosynthesis proteins, base excision repair, aminoacyl‐tRNA biosynthesis, and GSH metabolism (P < 0.05) (Fig. 6(D)). Compared with APAP, Nob treatment significantly increased sphingolipid metabolism, function unknown, inorganic ion transport and metabolism, phenylpropanoid biosynthesis, ascorbate and aldarate metabolism, and galactose metabolism, at the same time as decreasing various types of N‐glycan biosynthesis pathways, apoptosis, glycolysis/gluconeogenesis, metabolism of cofactors and vitamins, fatty acid biosynthesis, peptidoglycan biosynthesis, lysine degradation, glutamatergic synapse, aminoacyl‐tRNA biosynthesis, sulfur relay system, methane metabolism, pyruvate metabolism, lipid biosynthesis proteins, basal transcription factors and base excision repair (P < 0.05) (Fig. 6(E)). Therefore, these microbiota and pathway changes imply that CD‐EE, NAC, Nob and Tan may support hepatoprotection not only via hepatic antioxidant and epigenetic regulation, but also through modulation of the gut‐liver axis toward a more homeostatic state in APAP‐challenged mice.

Figure 6.

Predicted functional KEGG pathways significantly altered by APAP and each treatment in mice with APAP‐induced chronic liver injury. (A–E) Predicted KEGG functional pathways significantly altered by APAP and each treatment, analyzed using PICRUSt and STAMP (P < 0.05). Data are based on fecal samples from three individual mice per group (n = 3).

DISCUSSION

Citrus fruits, belonging to the genus Citrus of the family Rutaceae, are among the most widely consumed fruits worldwide and are commonly used as fresh fruits or processed juices. Bioactive compounds from citrus peel or pulp have attracted interest for pharmaceutical and nutraceutical applications. ²⁸ Citrus fruits contain diverse health‐promoting components, including phenolic compounds, vitamins, minerals, dietary fiber, essential oils and carotenoids. ²⁹ Citrus‐derived PMFs, such as Nob and Tan, have been reported to confer hepatoprotection in chronic and drug‐induced liver injury models, partly by activating the Nrf2 pathway and modulating gut microbiota. ²³ These recent findings indicate that CD‐EE contains a physiologically relevant combination of bioactive compounds that may act via similar mechanisms, as a result of the high levels of Nob, Tan, and related PMFs. In the present study, CD‐EE contained substantial levels of total phenolics and flavonoids (Table 1), with a concentration of phenolics and methoxyflavonoids of at least 172.78 mg g⁻¹, including Nob, Tan, 4′‐demethyltangeretin, 5‐demethylnobiletin, 5,7,3′,4′‐tetramethoxyflavone and 5‐demethyltangeretin (Fig. 1 and Table 2). Among these, approximately Nob at 81.21 mg g⁻¹ and Tan at 25.41 mg g⁻¹ were identified. Because Nob has one additional methoxy group at the 3′‐position compared to Tan, we evaluated the hepatoprotective effects of CD‐EE, Nob and Tan in parallel in the APAP‐induced chronic liver injury model. Recent reviews of citrus‐derived PMFs, specifically Nob and Tan, also highlight their ability to improve liver function, attenuate oxidative stress and inflammation, and remodel gut microbiota in various liver disease models, which aligns with the protective profile observed for CD‐EE, Nob and Tan in the present study.^{21, 22, 30} The oral doses of CD‐EE (250 and 500 mg kg⁻¹ bw) and Nob/Tan (50 mg kg⁻¹ bw) were chosen based on our previous work and other in vivo studies using citrus peel extracts and PMFs in rodents, where similar dose ranges (e.g. 10–50 mg kg⁻¹ for single PMFs and 200–500 mg kg⁻¹ for citrus extracts) were well tolerated and effectively attenuated liver injury and oxidative stress parameters.^{30, 31, 32, 33}

Prolonged and excessive APAP consumption can cause overproduction of NAPQI, leading to oxidative stress, hepatocellular damage, and increased serum ALT and AST activities. ³⁴ In the present study, repeated APAP administration induced clear liver morphological damage (Fig. 2) and elevations in serum ALT and AST (Table 3). A 9‐week protocol with twice‐weekly APAP injections was used to establish a chronic liver injury model rather than an acute toxicity setting, and daily oral supplementation with CD‐EE, Nob or Tan throughout this period was intended to mimic long‐term exposure to citrus‐derived PMFs under sustained APAP‐induced stress. ²⁴ NAC, CD‐EE, Nob and Tan all reduced APAP‐elevated ALT and AST activities, indicating hepatoprotective effects. APAP‐induced liver injury is associated with depletion of GSH, enhanced lipid peroxidation, and reduced activities of antioxidant and detoxifying enzymes, including GPx, GRd, GST and catalase.^{3, 4, 6, 34} Consistent with this, the present study showed significant decreases in hepatic GPx, GST and catalase activities and increased lipid peroxidation in APAP‐treated mice, whereas CD‐EE, Nob and Tan administration was associated with higher antioxidant enzyme activities and lower lipid peroxidation (Table 3), supporting a potential contribution of antioxidant mechanisms to the observed hepatoprotective effects rather than proving a direct causal relationship.

The Nrf2/antioxidant response element (i.e. ARE) pathway regulates phase II detoxifying and antioxidant enzymes, including UGT1A, NQO1 and HO‐1, which protect hepatocytes from oxidative damage.^{35, 36, 37} Citrus peel extracts and citrus flavonoids such as naringin and hesperidin have been reported to activate Nrf2 signaling and alleviate APAP‐induced liver injury. ³¹ In the present study, APAP decreased the mRNA expression of nrf2, ho‐1, nqo1 and ugt1a (Fig. 3(C)) and reduced hepatic protein levels of Nrf2, HO‐1, NQO1 and UGT1A (Fig. 3(A), (B)). CD‐EE(H), Nob, and Tan significantly or modestly restored the expression of these genes and proteins, indicating activation or recovery of the Nrf2‐mediated antioxidant pathway. Polyphenols can enhance Nrf2 activity via epigenetic mechanisms, including inhibition of DNMTs and HDACs. ³⁸ APAP increased the protein expression of DNMT1, DNMT3a and DNMT3b (Fig. 4), whereas CD‐EE, NAC, Nob and Tan tended to reduce these proteins, whereas HDACs 1–3 levels were largely unaffected. Collectively, these findings suggest that recovery of the Nrf2‐mediated antioxidant pathway observed with CD‐EE, Nob and Tan may be related, at least in part, to changes in DNA methylation in mice with chronic APAP‐induced liver injury, although direct causal relationships remain to be established. Thus, the current data support an association between Nrf2‐related epigenetic modulation and hepatoprotection, but they do not yet provide pathway‐level functional evidence. In addition, protein oxidation adducts, an important marker of oxidative protein damage, ³⁹ were not measured in the present study and should be addressed in future work. Moreover, we did not perform immunohistochemistry or separate cytoplasmic and nuclear western blot analyses for Nrf2 and related proteins; future studies incorporating IHC and subcellular fractionation with nuclear Nrf2 analysis will be important to clarify their localization and the dynamics of Nrf2 activation in APAP‐induced chronic liver injury.^{40, 41}

APAP‐induced acute liver failure has been associated with alterations in gut microbiota composition, ⁴² but the impact of chronic APAP exposure on the gut–liver axis is less well characterized. Using 16S rDNA sequencing of fecal samples, we observed distinct changes in gut microbiota composition in mice with APAP‐induced chronic liver injury and in those treated with CD‐EE(H), NAC, Nob or Tan compared to controls (Fig. 5(A)–(C)). CD‐EE(H), NAC, Nob and Tan partially restored APAP‐induced alterations at the phylum and genus levels, although the microbiota profiles did not fully return to the control pattern (Fig. 5(D)–(F)). Predicted functional KEGG pathways were also significantly altered by APAP and the various treatments (Fig. 6(A)–(E)), suggesting potential protection against APAP‐induced liver injury through modulation of microbiota‐associated metabolic pathways. Together with previous studies showing that gut microbiota and their metabolites modulate APAP metabolism, GSH homeostasis and hepatic inflammatory and Nrf2 signaling pathways, these microbiota and KEGG changes support a functional gut–liver crosstalk whereby CD‐EE, Nob, and Tan may influence hepatic detoxification and antioxidant defenses via microbiota‐dependent mechanisms. Functionally, CD‐EE and Nob restore sphingolipid metabolism, thereby regulating APAP‐induced hepatic apoptosis and inflammation. ⁴² Additionally, microbial modulation of GSH and secondary metabolite pathways aids host Phase II detoxification, acting through the gut–liver axis to enhance NAPQI neutralization and prevent oxidative damage. ⁴³ Dietary polyphenols, particularly citrus‐derived ones, can modulate the gut–liver axis, restore microbial balance and ameliorate APAP‐induced liver injury in animal models.^{43, 44} Our findings add to this evidence by demonstrating that long‐term supplementation with CD‐EE(H), Nob and Tan influences the composition of the gut microbiota under chronic APAP exposure. These observations support a potential contribution of gut microbiota modulation to the hepatoprotective effects of CD‐EE, Nob and Tan, although further functional studies will be required to confirm causality and clarify the underlying mechanisms. In particular, future work employing pathway‐level approaches, such as genetic or pharmacologic manipulation of Nrf2 and epigenetic enzymes, and targeted modulation or transplantation of the gut microbiota, will be important for substantiating the coordinated roles of antioxidant, epigenetic and gut–liver axis pathways in APAP‐induced chronic liver injury.

CONCLUSIONS

The present study demonstrates the hepatoprotective properties of CD‐EE and its active compounds, Nob and Tan, as well as the positive control NAC, against APAP‐induced chronic liver injury in mice. CD‐EE, NAC, Nob and Tan treatments significantly prevented APAP‐induced liver damage by increasing antioxidant enzyme activity, decreasing lipid peroxidation and modulating the Nrf2‐mediated antioxidant pathway. Moreover, we propose that these treatments may act through epigenetic mechanisms, including DNA methylation and histone acetylation. Our investigation also reveals the influence of APAP‐induced liver injury on the composition of gut microbiota. The hepatic protection afforded by CD‐EE, NAC, Nob and Tan may be achieved by modulating the gut microbiota and associated metabolic pathways. Future studies may entail clinical trials, microbiota interventions and the exploration of epigenetic mechanisms to enhance our understanding of liver diseases and devise effective strategies.

CONFLICTS OF INTEREST

The authors declare that they have no conflicts of interest.

Supporting information

Table S1. Primer sequences used for RT‐qPCR¹.

Data S1. Supporting information.

DATA AVAILABILITY STATEMENT

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