Loss of Vacuole Membrane Protein 1 protects against acetaminophen-induced liver injury by modulating inflammation and hepatocyte proliferation

Abstract

Acetaminophen (APAP) overdose is a leading cause of severe liver injury and acute liver failure (ALF). Increased hepatic macrophages promote inflammation and clear necrotic cell debris, thereby promoting a regenerative state for liver repair during APAP-induced liver injury. Vacuole membrane protein 1 (VMP1) is an ER transmembrane protein essential for autophagy. Deletion of VMP1 in the liver impairs autophagy. The role of VMP1 in APAP-induced liver injury is unknown. In this study, the human APAP ALF RNA sequencing database was analyzed. Liver-specific Vmp1 knockout (Vmp1^LKO) and matched wild-type (Vmp1^LWT) mice were administered 500 mg/kg APAP. Liver injury, necrosis, and other biochemical analyses were assessed. Increased mRNA levels of VMP1 and autophagy-related genes were found in human APAP ALF livers. Serum ALT and hepatic necrosis were increased in WT mice following APAP treatment. LC3B-II and p62 were increased, but VMP1 protein decreased in WT mouse livers following APAP treatment. Liver injury was markedly decreased in Vmp1^LKO mice compared with their WT littermates. Basal hepatic CYP2E1 and glutathione (GSH) levels were similar between genotypes, but increased lipid-laden macrophages (foam cells) were present in Vmp1^LKO livers. Interestingly, APAP protein adduct formation decreased in Vmp1^LKO mouse livers, but mitochondrial damage was similar in both phenotypes. Peroxynitrite formation was observed in APAP-treated Vmp1^LWT hepatocytes but was limited in foam cells of Vmp1^LKO mouse livers. Furthermore, GSH replenishment and liver regeneration were enhanced in Vmp1^LKO mouse livers. In conclusion, loss of hepatic VMP1 protects against APAP-induced liver injury by replenishing GSH, modulating inflammation, and enhancing liver regeneration.

Acetaminophen (APAP) is an analgesic and antipyretic drug that is commonly used worldwide. Although APAP is a safe drug at therapeutic levels, an overdose can cause severe liver injury and even acute liver failure (ALF) in animals and humans (Larson 2007). It is well established that the formation of a reactive metabolite, presumably N-acetyl-p-benzoquinone imine (NAPQI), which causes the depletion of cellular glutathione (GSH) and formation of APAP protein adduct (APAP-AD), is critical for the initiation of toxicity (Larson 2007; McGill and Jaeschke 2013). It was recognized that APAP overdose causes mitochondrial dysfunction, which is critical for triggering hepatocyte necrosis and subsequent liver injury. APAP overdose activates JNK, and activated JNK then translocates to the mitochondria, resulting in a signaling cascade that compromises mitochondrial electron transport, amplifies mitochondrial oxidant stress, and leads to hepatocyte necrosis (Xie et al. 2015; Win et al. 2024). The effects on mitochondria are thought to be initiated by mitochondrial APAP-AD formation, which is vital for APAP-induced mitochondrial damage and for amplifying cell death signaling events that contribute to APAP-induced necrosis and liver injury (Jaeschke et al. 2012). Although APAP-AD is critically involved in subsequent mitochondrial dysfunction, previous studies showed that the highly reactive peroxynitrite radical, a potent oxidant that is formed from nitric oxide and superoxide, can nitrate proteins, especially on tyrosine residues, and form nitrotyrosine protein adducts to induce mitochondrial dysfunction and hepatocyte necrosis during APAP overdose (Ramachandran et al. 2011; Radi 2018). GSH administration or treatment with a mitochondria-targeted SOD mimic, beyond the metabolism phase, effectively protected against cell death without preventing the mitochondrial oxidant stress but still eliminating nitrotyrosine (Knight et al. 2002; Du et al. 2019).

Cells can respond and adapt to various forms of stress, which may limit the extent of cell death. One adaptation and protective mechanism to control cellular stress, maintain protein homeostasis, and maintain mitochondrial quality is autophagy. Autophagy is a highly conserved intracellular degradation pathway, which is activated in response to adverse environmental conditions, such as the deprivation of nutrients or growth factors, as a survival mechanism (Mizushima et al. 2008). Autophagy protects cells by selectively removing toxic protein aggregates and damaged mitochondria (Ding and Yin 2008; Czaja et al. 2013; Williams et al. 2015; Ma et al. 2024a, 2024b). We previously demonstrated that activation of autophagy protects against APAP-induced hepatotoxicity by removing APAP-AD and damaged mitochondria in primary cultured hepatocytes and mouse livers (Ni et al. 2012a, 2016; Hinz et al. 2024).

Vacuole membrane protein 1 (VMP1), an ER-resident transmembrane protein, is important for the elongation and closure of autophagosomes (Morita et al. 2018). Deletion of VMP1 leads to the accumulation of unclosed autophagosomal membranes, resulting in impaired autophagy. It has also been shown that VMP1 is involved in many diseases, including liver and pancreatic diseases (Jiang et al. 2022; Wang et al. 2022; Ding et al. 2024; Chen et al. 2026b). VMP1 acts as a phospholipid scramblase; loss of hepatic VMP1 disrupts hepatic phospholipid homeostasis, decreases ER-mitochondrial contact, and impairs hepatic very-low-density lipoprotein (VLDL) secretion, resulting in severe steatosis, hepatocyte death, inflammation, and fibrosis, phenocopying metabolic dysfunction-associated steatotic liver disease (MASLD) (Jiang et al. 2022; Chen et al. 2026a, 2026b). However, the role of VMP1 in APAP-induced liver injury has not been studied.

Materials and methods

Antibodies

Antibodies used in this study were SQSTM1/p62 (#H00008878-M01) from Abnova, β-actin (#A5441) from Sigma, endonuclease G (#4969), proliferating cell nuclear antigen (PCNA, #13110), ATG6/Beclin-1(#3495), phosphorylated c-Jun N-terminal kinase (p-JNK, #4668), JNK (#9252), glyceraldehyde 3-phosphate dehydrogenase (GAPDH, # 2118), VMP1 (#12929), phosphor-S6 (#4858), S6 (#2217), phosphor-4EBP1 (#9451), 4EBP1 (#9452), phosphor-YAP S127 (#4911), YAP (#14074), Sox9 (#82630) from Cell Signaling Technologies, apoptosis-inducing factor (AIF, #ab32516), cytochrome P450 2E1 (CYP2E1, #ab28146), 3-nitrotyrosine (#ab61392) from Abcam, Bcl-2-associated X protein (Bax, #556467), cytochrome c (#556433) from BD Pharmingen, voltage-dependent anion channel (VDAC, #529534) from Calbiochem. Glutamate-cysteine ligase catalytic subunit (GCLC), and glutamate-cysteine ligase modifier subunit (GCLM) were kindly provided by Dr Terry Kavanagh (University of Washington, Seattle, Washington); the rabbit anti-APAP-adducts antibody was kindly provided by Dr Lance Pohl (National Heart, Lung and Blood Institute) (Ryan et al. 2012). The rabbit polyclonal anti-LC3 antibody was described previously (Ma et al. 2023). The secondary antibodies used in this study were horseradish peroxidase (HRP)-conjugated goat anti-mouse or rabbit antibodies (#115-035-062 and # 111-035-045, respectively, Jackson ImmunoResearch) for immunoblots and ImmPRESS HRP Anti-Rabbit IgG (Peroxidase) Polymer Detection Kit (#MP-7401-50, Vector Laboratories) for immunohistochemistry analysis.

Animal experiments

C57BL/6J WT mice were purchased from Jackson Laboratory. Vmp1^flox mice (C57BL/6N) were purchased from The European Mouse Mutant Archive as described previously (Jiang et al. 2022) and were crossed with Albumin-Cre mice (Alb-Cre, C57BL/6J) (Jackson Laboratory) to generate hepatic Vmp1 knockout (Vmp1^LKO). All animals received humane care. All procedures were approved by the Institutional Animal Care and Use Committee of the University of Kansas Medical Center. APAP solution (30 mg/ml) was prepared by adding APAP (Sigma Aldrich #A7085) to 0.9% sodium chloride solution (Hyclone #Z1377), and the mixture was heated without boiling until the powder was completely dissolved. All the mice were without fasting. Two to three-month-old male C57BL/6J WT mice were administered either saline or APAP (500 mg/kg) by peritoneal injection and blood and liver tissues were collected at 2, 6, 24, and 48 h after APAP treatment. Two to three-month-old male Vmp1^LKO and matched WT mice were administered either saline or APAP (500 mg/kg) by peritoneal injection, and blood and liver tissues were collected at 2, 6, and 24 h after APAP treatment. Three to six mice were used in each group. Liver injury was assessed by the determination of serum alanine aminotransferase (ALT) activities and hematoxylin and eosin (H&E) staining of liver sections as we described previously (Chao et al. 2024).

Immunoblot and densitometric analyses

Immunoblot and densitometric analyses were conducted as described previously (Chao et al. 2023). Briefly, total liver lysates were extracted using RIPA buffer (1% NP40, 0.5% sodium deoxycholate, 0.1% sodium dodecyl [lauryl] sulfate) supplemented with proteinase inhibitor cocktail (#B14001, Selleck Chemicals). Thirty micrograms of liver lysates from each sample were separated by SDS-polyacrylamide gel electrophoresis and transferred to polyvinylidene fluoride membranes. The membranes were stained with primary antibodies followed by secondary HRP-conjugated antibodies. The membranes were further developed with SuperSignal West Pico plus chemiluminescent substrate (#34580, ThermoFisher Scientific). The densitometric analysis was performed using ImageJ. The protein levels of immunoblots were calculated as fold change (FC) $\pm$standard error (SE).

Histology and immunohistochemistry

Immunohistochemistry was performed as described previously (Qian et al. 2021). Briefly, paraffin-embedded liver sections were stained with H&E, and immunohistochemistry staining for PCNA and nitrotyrosine was employed with corresponding primary antibodies followed by secondary antibodies. The tissues were developed with NOVA red and counterstained with hematoxylin. Terminal deoxynucleotidyl transferase dUTP nick end labeling (TUNEL) staining was performed in paraffin-embedded liver sections using the In Situ Cell Death Detection TUNEL Kit (#11684817910) from Sigma-Aldrich.

Necrotic areas were defined by the absence of hepatocytes surrounding the central veins. The entire section was first examined under a microscope for an overview, followed by 6 to 8 images per sample, randomly taken, using 10× magnification. The necrotic areas are quantified using ImageJ to measure the area of necrosis, which is then divided by the total area of the picture to give the percentage of necrotic areas.

Hepatic GSH measurement

GSH was measured using a commercial GSH assay Kit (#100 T, MedChem Express) following the manufacturer’s instructions.

Subcellular fraction

Subcellular fractions were extracted from mouse livers as described previously (Wang et al. 2019). Livers were homogenized in mitochondria isolation buffer (70 mM sucrose, 200 mM mannitol, 5 mM HEPES, 0.5 mM EGTA and 0.1% (w/v) fatty acid-free BSA; pH 7.2). Homogenates were centrifuged at 1,000 × g for 10 min at 4°C, and the supernatants were centrifuged once at 10,000 × g for 10 min at 4°C. The mitochondrial pellets were suspended in the isolation buffer, and total protein was determined using a BCA protein assay.

Quantitative real-time polymerase chain reaction

RNA was extracted from mouse liver tissue using TRIzol reagent (#15596026, ThermoFisher Scientific), and complementary DNA (cDNA) was reverse-transcribed utilizing RevertAid H minus reverse transcriptase (#EP0451, Thermo Fisher Scientific) (Chen et al. 2026a). Quantitative real-time PCR was conducted on a Bio-Rad CFX384 detection system utilizing SYBR Green mix (#B21202, Bimake). Expression of Gclc, Gclm, and Nqo was quantified using qRT-PCR analysis, and Actb was used as an internal control. The fold change of mRNA was expressed as 2^−ΔΔCt. The forward and reverse primer sequences used in this study were as follows: Actb, 5′-TGTTACCAACTGGGACGACA-3′ and 5′-GGGGTGTTGAAGGTCTCAAA-3′; Gclc, 5′-AACACAGACCCAACCCAGAG-3′ and 5′-CCGCATCTTCTGGAAATGTT-3′; Gclm, 5′- TGTGTGATGCCACCAGATTT-3′ and 5′-GATGATTCCCCTGCTCTTCA-3′; and Nqo1, 5′- CAGATCCTGGAAGGATGGAA-3′ and 5′-TCTGGTTGTCAGCTGGAATG-3′.

RNA sequencing analysis

The mouse liver RNA-seq data are accessible from GEO (GSE338635). Total RNA was extracted from mouse livers using TRIzol reagent (ThermoFisher Scientific). A detailed RNA sequencing analysis was performed as described previously (Chen et al. 2024). Briefly, the raw data for RNA sequencing is stored in the fastq.gz file. The data quality was assessed by the FASTQC software (version 0.11.8). The bases with a per-base sequence quality lower than 30 were removed later in the QC process. Adaptors were also removed using Trimmomatic (version 0.39) software, if any. The adaptor-removed files were mapped to the mouse genome (grcm38) with HISAT2 software (version 2.0.1). The total counts for each read were accessed by HISAT2 using Mus_musculus.GRCm38.100.gtf file. The count files were then imported into RStudio (version 2023.3.0.386, PBC, Boston, Massachusetts, United States) for downstream analysis using the DESeq package (version 3.19). The count numbers for each gene was log10 transformed and divided by the size factor for normalization using the package before performing principal component analysis (PCA). Quantification of gene expression was generated with HTSeq-counts v0.6.0. Significant differentially expressed genes (DEGs) were identified with the R package DESeq2. Multiple comparisons were taken into account by converting P values to Benjamini − Hochberg’s false discovery rate (FDR).

Two human RNAseq datasets (GSE255777 and GSE223559) for APAP-induced ALF were analyzed. GSE223559 dataset includes 3 human healthy and 2 APAP-ALF explants. Three healthy and 3 APAP-ALF livers (2 of the APAP-ALF samples were from the same patient with 2 distinct geographical tissue blocks) were utilized for spatial transcriptome analysis by 10× Genomics Visium. The raw UMI counts were normalized using SCTransform/LogNormalization within the Seurat environment to eliminate technical artifacts related to local sequencing depth while preserving biological heterogeneity. GSE255777 includes 2 healthy explants and 1 clinical APAP-ALF explant. Data were aligned using Space Ranger and subjected to standard target-sum scaling library normalization (LogNormalize, scale factor = 10,000) to yield normalized expression matrices. The corresponding heatmaps were generated by utilizing pre-processed and log-transformed datasets that were deposited by the respective authors. GSE25577 dataset has a limited sample size, and no statistical analysis was performed. GSE223559 dataset underwent a strict sample-level pseudobulk analysis design that avoids false-positive pseudo-replication, with Log2Fold Change (Log2FC), followed by a two-tailed, unpaired Welch’s t-test on the sample-level averages to calculate raw P-values and convert P-values to Benjamini − Hochberg’s false discovery rate (FDR).

Statistical analysis

Experimental data were subjected to One-way ANOVA analysis with Tukey’s post hoc test for multi-group comparison. The value of P ≤ 0.05 was considered statistically significant among groups. All experimental data were reported as mean ± standard error of the mean (SEM).

Results

Alteration of VMP1 and autophagy in APAP-induced liver injury in humans and mice

The mRNA levels of VMP1 and autophagy-related genes, SQSTM1/p62 and LC3B, were examined in 2 human APAP ALF RNA-seq databases (GSE255777 and GSE223559). The mRNA expression levels of VMP1, SQSTM1/p62, and MAP1LC3B (Microtubule-associated protein 1 light chain 3 beta, alias LC3B) were increased in human APAP ALF liver samples (Fig. 1A). WT mice were treated with APAP (500 mg/kg) for up to 48 h. APAP induced central lobular necrosis in mouse livers, with necrosis peaking at 6 h and resolving by 24 h, as analyzed by H&E staining. The mouse livers were recovered at 48 h post-APAP treatment (Fig. 1B). Consistent with necrotic areas, serum ALT activities peaked at 6 h, decreased at 24 h and returned to the basal level at 48 h post-APAP treatment (Fig. 1C). Similar to human ALF livers, mRNA expression of Vmp1 increased in mouse livers with APAP at 24 and 48 h although the expression of Sqstm1/p62 and Lc3b was not significantly altered (Fig. 1D). After APAP treatment, VMP1 protein levels in WT mouse livers decreased at 6 h, began to recover at 24 h, and fully recovered at 48 h (Fig. 1E). Protein levels of SQSTM1/p62 and LC3BII were increased following APAP overdose. APAP also increased the levels of phosphorylated JNK protein at 6 h and gradually decreased them at 24- and 48-h post-APAP (Fig. 1E). These data indicate that APAP overdose induces dynamic alterations in VMP1 expression in both humans and mice, which are associated with APAP-induced liver injury.

Fig. 1.

Alteration of VMP1 and autophagy in APAP-induced liver injury in humans and mice. A) Heatmap of autophagic genes in 2 publicized RNA-seq databases of human APAP ALF samples. B) Two to three-month-old C57BL/6J wild-type mice were injected peritoneally with APAP 500 mg/kg for the indicated time. H&E staining of liver tissues from mice. Necrotic areas were circled and labeled with solid stars. Scale bars, 100 µm. C) Serum ALT activities were measured in mice injected with APAP. D) Heatmap of autophagic genes in mouse livers. Data represent mean ± SEM and were analyzed by one-way ANOVA with a post hoc Tukey test. E) Total liver lysates were extracted from APAP-injected mice and subjected to immunoblot analyses for the indicated antibodies. The expression of proteins was quantified as fold change (FC) $\pm$ standard error (SE). *APAP vs saline; ^APAP 24 h vs APAP 48 h; ^#APAP 6 h vs APAP 24 h.

Hepatic deletion of VMP1 protected against APAP-induced liver injury in mice

APAP overdose slightly increased the liver-to-body weight ratio in WT mice (Fig. 2A). Vmp1^LKO mice had a slightly higher liver-to-body weight ratio than WT mice at baseline, and APAP further increased it (Fig. 2A). Serum ALT activities were significantly elevated in WT mice at 6 h and reduced at 24 h in APAP-treated WT mice. However, serum ALT activities were only moderately increased in Vmp1^LKO mice following APAP treatment (Fig. 2B). H&E staining showed markedly increased areas of necrosis in WT mouse livers treated with APAP, whereas areas of necrosis were significantly decreased in Vmp1^LKO mouse livers compared with WT mice (Fig. 2C and D). Interestingly, increased numbers of foam cells, which are lipid-laden cells primarily derived from macrophages, were readily observed in Vmp1^LKO mouse livers but not in WT mice, regardless of APAP treatment (Fig. 2C). To confirm that lipid-laden cell clusters are indeed foam cells, 2 macrophage markers, F4/80 and CD68, were stained in Vmp1^LWT and Vmp1^LKO mouse livers. The foam cell clusters showed both F4/80 and CD68-positive staining (Fig. 2E). Furthermore, the number of TUNEL-positive cells was markedly increased in WT mouse livers post-APAP. However, the pattern of TUNEL-positive staining was different in Vmp1^LKO mouse livers, in which most foam cells, but not hepatocytes, were TUNEL-positive even in the saline control group (Fig. 2F). These data suggest that mice with hepatic Vmp1 deletion are resistant to APAP-induced liver injury.

Fig. 2.

Hepatic deletion of VMP1protected against APAP-induced liver injury in mice. Two to three-month-old Vmp1^LWT and Vmp1^LKO mice were injected peritoneally with 500 mg/kg for the indicated time. A) Liver/body weight ratio in mice. B) Serum ALT activity was measured in mice. C) Representative images of H&E staining of liver tissues from Vmp1^LWT and Vmp1^LKO mice. Necrotic areas were circled. * indicated foam cells. Scale bars: 100 µm. D) Necrotic areas were quantified. 6 to 8 images were randomly taken per sample using 10× magnification. E) Representative images of F4/80 and CD68 IHC staining. * indicated foam cells. Scale bars, 100 µm. F) Representative images of TUNEL staining of liver tissues from Vmp1^LWT and Vmp1^LKO mice. TUNEL-positive areas were quantified. Data represent mean ± SEM and were analyzed by one-way ANOVA with a post hoc Tukey test (A, B, and D). PV, portal vein; CV, central vein.

Hepatic deletion of VMP1 decreases APAP adduct levels, with faster replenishment of GSH in mouse livers

To examine whether hepatic deletion of VMP1 may alter APAP bioactivation, hepatic levels of GSH were measured in mouse livers. Depletion of GSH was observed in both Vmp1^LWT and Vmp1^LKO mouse livers at 2 h of APAP treatment, and replenishment began at 6 h and was complete by 24 h (Fig. 3A). No GSH differences were observed in the saline control groups of WT and KO mice. Interestingly, hepatic GSH concentration was significantly higher in Vmp1^LKO mouse livers than in WT mouse livers at 6 h post-APAP (Fig. 3A). The protein levels of CYP2E1 were not different in WT and Vmp1^LKO mouse livers (Fig. 3C). The levels of APAP-AD were higher in WT mouse livers compared to Vmp1^LKO littermates at 2 h post APAP treatment (Fig. 3B). The levels of APAP-AD slightly decreased at 6 h, but markedly decreased in Vmp1^LKO mouse livers at 24 h post APAP treatment (Fig. 3C and E). However, the levels of phosphorylated JNK were lower in Vmp1^LKO mouse livers at 6 h than in WT mice following APAP treatment (Fig. 3C). Although hepatic APAP-AD levels were similar in WT and Vmp1^LKO mice at 6 h, serum APAP-AD levels were lower in Vmp1^LKO mice than in WT mice (Fig. 3D), which is likely due to reduced necrosis observed in Vmp1^LKO mice. Together, these data indicate that hepatic Vmp1 deletion does not affect APAP adduct formation but decreases APAP adduct levels in the mouse livers.

Fig. 3.

Hepatic deletion of VMP1reduced APAP adduct formation and replenished GSH faster. Two to three-month-old Vmp1^LWT and Vmp1^LKO mice were injected peritoneally with 500 mg/kg for the indicated time. A) Hepatic GSH concentrations were measured. Data represent mean ± SEM and were analyzed by one-way ANOVA with a post hoc Tukey test. B) Total liver lysates were extracted from Vmp1^LWT and Vmp1^LKO mice at 2-hour post APAP overdose and subjected to immunoblot analyses for the indicated proteins. C) Total liver lysates were extracted from Vmp1LWT and Vmp1LKO mice at -hour post APAP overdose and subjected to immunoblot analyses for the indicated proteins. D) Serum APAP adducts were analyzed by immunoblot in mice. E) Hepatic APAP adducts were analyzed by immunoblot in mice. The expression of proteins was quantified as fold change (FC) $\pm$ standard error (SE). *All groups vs WT saline; ^KO APAP vs WT APAP; ^#KO APAP vs KO saline.

Hepatic deletion of VMP1 decreases APAP-induced nitrotyrosine protein adduct formation in hepatocytes

Mitochondrial protein adduct formation is central to APAP pathophysiology and subsequent mitochondrial dysfunction (Jaeschke and Ramachandran 2024). Cytosolic and mitochondrial fractions were isolated from APAP-treated mouse livers. Western blot analysis revealed similar levels of mitochondrial APAP-AD between WT and Vmp1^LKO mice after APAP treatment. However, the levels of mitochondrial phosphorylated JNK were much lower in Vmp1^LKO mice than in WT mice after APAP treatment (Fig. 4A). APAP treatment increased the release of mitochondrial AIF, endo G, and cytochrome c into the cytosol to a similar extent in WT and KO mice, although mitochondrial Bax translocation was slightly higher in KO mice than in WT mice following APAP treatment (Fig. 4A). Protein levels of p62 and LC3B were slightly increased in APAP-treated WT mouse livers, p62 but not LC3B were further increased in APAP-treated Vmp1^LKO mice (Fig. 4B). The protein level of Beclin-1, a key autophagy protein, decreased in WT mouse livers 6 h post APAP-treatment. The protein level of Beclin-1 was already lower in Vmp1^LKO mouse livers at the basal level, which further decreased at 6 h post APAP treatment (Fig. 4B).

Fig. 4.

Hepatic deletion of VMP1 decreases APAP-induced nitrotyrosine protein adduct formation in hepatocytes. Two to three-month-old Vmp1^LWT and Vmp1^LKO mice were injected peritoneally with APAP for 6 h. A) Cytosolic and mitochondrial fractions were extracted from mice and subjected to immunoblot analysis with the indicated antibodies. * KO cytosolic fraction vs WT cytosolic fraction; ^KO mitochondrial fraction vs WT mitochondrial fraction. B) Total liver lysates were extracted from mice 6 h post APAP treatment and subjected to immunoblot analysis with the indicated antibodies. The expression of proteins was quantified as fold change (FC) $\pm$ standard error (SE). *All groups vs WT saline; ^KO APAP vs WT APAP; ^#KO APAP vs KO saline. C) Hepatic mRNA levels of Nrf2 target genes were analyzed by qPCR and normalized to Actb mRNA. D) Representative images of nitrotyrosine staining in Vmp1^LWT and Vmp1^LKO mice injected peritoneally with 500 mg/kg. The positive areas were quantified. Scale bars: 100 µm. Abbreviations: CV, central vein; LE, long exposure; PV, portal vein; SE, short exposure.

GCLC and GCLM, two Nrf2 target genes, are critical for GSH synthesis (Win et al. 2023). Although GCLC and GCLM proteins were reduced in WT mouse livers following APAP overdose, their levels were much higher in APAP-treated Vmp1^LKO mice compared with WT mice following APAP treatment (Fig. 4B). mRNA expression levels of Nrf2 target genes, Gclc, Gclm, and Nqo1, were similar in WT and Vmp1^LKO mice in saline and APAP 2-h treatment groups, mRNA levels of Gclc and Nqo1 were significantly higher in Vmp1^LKO mice 6 h post-APAP (Fig. 4C). The levels of nitrotyrosine protein adduct were increased in APAP-treated WT mouse livers, but nitrotyrosine staining was only found in foam cells in Vmp1^LKO mouse livers (Fig. 4D). No Nitrotyrosine-positive cells were observed in both WT and KO mouse livers when the livers were stained with second antibody only, followed by substrate development, ruling out the possibility of false staining (Fig. 4D). These data suggest that hepatic deletion of VMP1 accelerates GSH resynthesis and reduces nitrotyrosine protein adduct formation in hepatocytes after APAP overdose, despite similar mitochondrial protein adduct formation and mitochondrial damage compared with WT mice.

Hepatic deletion of VMP1 increases hepatocyte proliferation

Liver regeneration is a critical step in recovery from APAP-induced liver injury. Fewer PCNA-positive hepatocytes were observed in Vmp1^LWT saline control mouse livers. However, the number of PCNA-positive hepatocytes was significantly increased in mouse livers 24 h post-APAP (Fig. 5A and B). Vmp1^LKO mice already had an increased number of PCNA-positive hepatocytes in untreated mouse livers, and the number of PCNA-positive hepatocytes was further increased 24 h post-APAP (Fig. 5A and B). Consistent with the PCNA IHC staining results, the PCNA protein level was higher in APAP-treated WT mouse livers at 24 h but not at 6 h. The PCNA protein level increased slightly in Vmp1^LKO mouse livers without APAP treatment and increased further in APAP-treated Vmp1^LKO mice, reaching almost 4-fold higher than in APAP-treated WT mice at 6 and 24 h post-APAP (Fig. 5C). The protein levels of p62 and LC3B-II increased in Vmp1^LKO mice, whereas p62 decreased and LC3B-II increased in WT mice. The protein levels of Beclin-1decreased in both WT and Vmp1^LKO mice at 24 h APAP treatment (Fig. 5C).

Fig. 5.

Hepatic deletion of VMP1 increases hepatocyte proliferation. A). Representative images of PCNA staining in Vmp1^LWT and Vmp1^LKO mice injected peritoneally with APAP for 24 h. Scale bars: 100 µm. B) PCNA-positive hepatocytes were quantified. C) Immunoblot analysis of PCNA in mouse livers. D) Immunoblot analysis of autophagy and mTOR pathway in mouse livers. E) Immunoblot analysis of SOX9 and YAP in mouse livers. The expression of proteins was quantified as fold change (FC) $\pm$ standard error (SE). *All groups vs WT saline; ^KO APAP vs WT APAP; ^#KO APAP vs KO saline.

mTOR is critical for cell proliferation by increasing anabolic protein synthesis. Two downstream mTOR substrate proteins, S6 and 4EBP1, were examined in mouse livers following APAP treatment. The levels of phosphorylated S6, but not phosphorylated 4EBP1, were increased 24 h post-APAP treatment in WT mouse livers. The levels of both phosphorylated S6 and 4EBP1 were higher in Vmp1^LKO mouse livers without APAP treatment, but APAP overdose didn’t further increase phosphorylated S6 and 4EBP1 proteins (Fig. 5D). We next determined the levels of SRY-box transcription factor 9 (SOX9) and yes-associated protein (YAP), two liver progenitor cell markers, which have recently been shown to be critical in APAP or concanavalin A (ConA) induced liver injury (Poudel et al. 2021; Feng et al. 2023). In APAP-treated WT mouse livers, SOX9 levels increased, whereas phosphorylated YAP S127 (a cytosolic inactive form) decreased. Interestingly, SOX9 and phosphorylated YAP S127 levels were much lower in Vmp1^LKO mouse livers, regardless of APAP treatment (Fig. 5E). These data suggest that loss of hepatic VMP1 promotes liver regeneration and repair after APAP-induced liver injury in mice, likely independent of YAP and SOX9.

Discussion

APAP in the United States and Japan, or paracetamol in Europe and the rest of the world, is a common, widely available analgesic and antipyretic drug that has been used for decades and is safe and effective at therapeutic doses (Dart and Bailey 2007; Ohashi and Kohno 2020). However, an overdose of APAP can lead to severe liver injury, which could progress to ALF and death (Fisher and Curry 2019). APAP overdose is the number one cause of ALF in the United States and the UK (Bernal and Wendon 2013). APAP metabolism mainly occurs in the liver, and the majority of APAP (approximately 90%) is conjugated to sulfide or glucuronate and secreted in the urine. The remaining 5% to 10% is further metabolized by cytochrome P450 enzymes (mainly CYP2E1) to generate NAPQI. NAPQI further conjugates with cellular GSH without obvious harmful effects on hepatocytes at therapeutic doses. Upon APAP overdose, cellular GSH is depleted, and NAPQI then forms covalent links with the sulfhydryl group on cysteine residues of many proteins, forming APAP-AD. NAPQI-adducted proteins may alter the functions of these proteins and contribute to APAP-induced liver injury (McGill and Jaeschke 2013). Among them, mitochondrial protein APAP-AD formation and mitochondrial damage are thought to play crucial roles in APAP-induced liver injury (McGill et al. 2012; Xie et al. 2015).

Although acetaminophen overdose has been well documented to induce JNK activation in mouse livers and humans, and increased mitochondrial JNK translocation is associated with decreased mitochondrial bioenergetics and APAP-induced liver injury (Hanawa et al. 2008), the role of JNK in APAP-induced liver injury remains controversial. Notably, the liver contains two JNK isoforms, JNK1 and JNK2, which may have redundant yet distinct roles in hepatic pathophysiology. It has been reported that acetaminophen-induced liver injury is prevented only in Jnk2 KO mice but not in Jnk1 KO mice (Gunawan et al. 2006). Jnk1, Jnk2 double-knockout mice showed more severe liver injury than WT mice after APAP overdose (Cubero et al. 2016). Although total and mitochondrial phosphorylated JNK levels were lower in APAP-treated Vmp1 KO mice than in WT mice, there was no difference in the release of mitochondrial intermembrane space proteins between Vmp1 KO and WT mice after APAP treatment. These observations suggest that JNK activation may not be the critical trigger for APAP-induced mitochondrial damage or that JNK activation is a secondary effect in response to liver injury and stress.

In response to the insult of APAP overdose and its metabolites, cells have developed adaptive responses to protect hepatocytes against these detrimental effects caused by APAP overdose (Ni et al. 2012b; Hinz et al. 2024). We have previously demonstrated that pharmacological induction of autophagy protects against APAP-induced liver injury, likely via removal of APAP-induced APAP-AD and damaged mitochondria (Ni et al. 2012a, 2016; Hinz et al. 2024). Paradoxically, genetic deletion of an essential autophagy-related gene, Atg5, in the mouse livers also protected against APAP-induced liver injury. This protection is attributed to p62-mediated non-canonical Nrf2 activation and rapid replenishment of hepatic GSH in liver-specific Atg5 KO mice. Similar to liver-specific Atg5 KO mice, deletion of VMP1 in mouse livers also led to rapid replenishment of hepatic GSH, likely due to increased Nrf2 activation after APAP treatment. However, this increase was much milder in liver-specific Vmp1 KO mice than in liver-specific Atg5 KO mice.

In addition to Nrf2 activation, several other mechanisms may also contribute to protection against APAP-induced liver injury in liver-specific Vmp1 KO mice. Repair of APAP-induced liver damage requires the removal of APAP-induced necrotic cell debris by inflammatory cells and the regeneration of lost hepatocytes and non-parenchymal cells, which are vital for liver recovery and restoration of its function. Although the mechanisms remain unclear, loss of hepatic VMP1 leads to massive accumulation of foam cell clusters in the liver, which may enhance the clearance of necrotic cell debris induced by APAP. Increased nitrotyrosine adducts in the foam cells in liver-specific Vmp1 KO mice may support this notion. Moreover, it has been reported that Kupffer cells/macrophages surrounding the area of necrosis can produce IL-10, which promotes the proliferation of both hepatocytes and sinusoidal endothelial cells for liver repair (Kato et al. 2011; Nguyen et al. 2022). Furthermore, activated hepatic stellate cells can also support liver regeneration by producing HGF (Chang et al. 2017). A previous study also showed hepatocyte-specific deletion of Yap had a significantly earlier and higher pro-regenerative inflammatory response following APAP overdose in mice (Poudel et al. 2021). In ConA-induced immune-mediated ALF, it was demonstrated that monocyte-derived macrophages were rapidly recruited to encapsulate necrotic cells and promote the formation of SOX9⁺ hepatocytes. These SOX9⁺ hepatocytes are derived from mature hepatocytes located near the necrotic lesions and act as a protective barrier against further injury (Feng et al. 2023; Nguyen et al. 2025). Because YAP and SOX9 are significantly reduced in Vmp1^LKO mice, increased proliferation in APAP-treated Vmp1^LKO mice is likely independent of YAP and SOX9. Therefore, increased foam cells in liver-specific Vmp1 KO mice may also promote hepatocyte proliferation and liver repair by increasing the production of pro-proliferative cytokines or by activating stellate cells following APAP treatment. Future studies are needed to further investigate whether and how foam cells would promote liver regeneration following APAP overdose. Notably, developmental Vmp1 KO mice have been used in the study, and these mice have increased basal levels of inflammation and proliferation, which may contribute to the protective effects against APAP-induced liver injury. Future studies are needed to investigate whether VMP1 actively contributes to APAP hepatotoxicity by using Vmp1 knockdown mice or inducible KO mice.

In summary, we demonstrated that loss of hepatic VMP1 protects against APAP-induced liver injury by promoting hepatic GSH replenishment, liver regeneration, modulating inflammation and reducing APAP-AD formation (Fig. 6).

Fig. 6.

Summary of the role of VMP1 in APAP overdose. In wild-type mice, APAP overdose depletes GSH and forms APAP adducts, especially mitochondrial APAP adducts in hepatocytes. Mitochondrial APAP adducts activate JNK, which further causes mitochondrial oxidative or nitrosative stress and leads to hepatocyte necrosis and liver injury. Hepatic deletion of VMP1 activates the Nrf2 pathway and promotes GSH synthesis, which ameliorates mitochondrial stress and, in turn, alleviates hepatocyte necrosis and liver injury. Furthermore, Hepatic deletion of VMP1 in mice increased inflammation and promoted liver regeneration.

Author’s contributions

K.N. performed experiments, data analysis and interpretation, and preparation of the manuscript; M.N. performed experiments, data analysis and interpretation; H.-M.N. supervised the project, participated in research design, performed experiments, data analysis and interpretation, and wrote the manuscript. All authors read and approved the final manuscript.

Funding

The research was supported in part by National Institute of Diabetes and Digestive and Kidney Diseases DK129234 (to H.-M.N.), National Institute of General Medical Sciences of Centers of Biomedical Research Excellence (COBRE) grant P20GM144269 (to S.W.).

Conflicts of interest. None declared.