The general procedure for synthesis of MCGP (98% or higher purity) and its structure was characterized by one-dimensional nuclear magnetic resonance (NMR) spectrometer (Supplementary Materials). MCGP was dissolved in dimethyl sulfoxide (DMSO, Biosharp, China) to prepare a 40 mM stock solution and stored at −20°C before use for cell administration. MCGP was resuspended in 5% CMC-Na solution for the intragastric administration.

Human liver cancer cell line 7,721 and HepG2, human hepatic stellate cell line LX-2, human monocytic leukemia cell line THP-1, and murine normal liver cell AML12 were purchased from Procell Life Science & Technology Co., Ltd. (Wuhan, China). Cells were cultured as monolayers in DMEM-high glucose (7,721 and HepG2)/DMEM (LX-2)/RIPM-1640 (THP-1) and D/F12 (AML12) medium (Hyclone, United States) supplemented with 10% fetal bovine serum (Procell, China) and 1% antibiotics of penicillin/streptomycin (100 units/mL) (Invitrogen, United States). Cells were grown in an atmosphere of 5% CO₂ in air at 37 °C.

For the detection of the cytotoxicity effect, 25, 50, 100, 200, 400, 800, and 1,000 μM of MCGP were applied to 7,721, LX-2, and THP-1 cells. 25, 50, 100, 200, and 400 μM of MCGP were applied to HepG2 and AML12 cells for 48h. DMSO (0.1%) was used as control. Then, cell counting kit 8 (CCK8, Biosharp, China) was added to each well for 1–4 h, and absorbance was detected at 450 nm using a microplate reader (Thermo Fisher Scientific, United States).

For the cell viability assay, cells were incubated with 50, 100, and 200 μM of MCGP for 18 h followed by 30 mM of CCl₄ for 6 h. Then, cell counting kit 8 was added to each well for 1–4 h, and absorbance was detected at 450 nm. For the cell apoptosis assay, cells were incubated with 50, 100, and 200 mM of MCGP for 18 h followed by 30 mM of CCl₄ or 25 mM of APAP for 6 h. Then, the cells were determined using an Annexin V-fluorescein isothiocyanate (FITC) apoptosis detection kit (KeyGEN, Jiangsu, China) according to the manufacturer’s protocol. For the ROS generation assay, the cells were incubated with 50, 100, and 200 mM of MCGP administration for 18 h, after which 400 μM of H₂O₂ was introduced to generate ROS for 6 h. Then, the cells were fixed with 4% cold paraformaldehyde for 15 min. Thereafter, ROS (Sigma, United States) were stained and scanned using a digital microscopy scanner Pannoramic MIDI (3DHISTECH, Hungary).

To evaluate the antiradical activity of MCGP, 25–1,000 µM of MCGP was placed in 96-well plates (Corning, NY, United States), and 130 µl of 2,20-AzinoBis-(3-ethylbenzoThiazoline-6-Sulfonic acid) (ABTS^•+) (Macklin, Shanghai, China) radicals was added. The ABTS^•+ reagent was produced by reacting the ABTS solution (10 mM) with 30% H₂O₂ in advance. After incubation for 16 h at room temperature, 10 ml of this solution was diluted in 80 ml of sodium acetate hydrochloric acid buffer (Macklin, Shanghai, China). Ascorbic acid (25–1,000 µM) (Macklin, Shanghai, China) was used as a standard antioxidant. After 30 min of incubation in the dark at room temperature, the plates were read at 650 nm using a microplate reader (Thermo Fisher Scientific, United States).

Total RNAs were extracted with TRIzol reagent (Invitrogen, Thermo Fisher Scientific, United States) from HepG2 cells treated with 40 μM of MCGP or DMSO (control) for 24 h. Three biological replicates for the control group and the sample group were sequenced by Beijing Genomics Institute (BGI) using the BGISEQ-500 platform for each group. The differentially expressed genes (DEGs) were screened with a q value (adjusted p value) ≤ 0.01 and |fold change| ≥ 2. Bioinformatics Workflow including data filtering, mapping transcript prediction, differential gene expression analysis, Kyoto Encyclopedia of Genes and Genomes (KEGG) pathway analysis, and GO function analysis were performed by the platform established at BGI.

Specific-pathogen–free (SPF) male C57BL/6 and Balb/c mice (20–22 g) were purchased from Beijing HFK Bio-Technology Co. Ltd. (Beijing, China) and housed in a SPF environment at the experimental animal center of Tongji Medical College, Huazhong University of Science and Technology (Wuhan, China). After adapting feed for 1 week, mice were subjected to the following administration.

For the CCl₄-induced acute liver injury, male C57BL/6 were randomly divided into 5 groups. 1) Control group (n = 10). 2) Model group (n = 10). 3) MCGP-M group (n = 10, 100 mg/kg). 4) MCGP-H group (n = 10, 200 mg/kg). 5) Positive group. (n = 10, 50 mg/kg of silymarin). Mice were first orally given MCGP once a day for 10 days, and then followed with a single intraperitoneal injection of 1% CCl₄ (in olive oil, 5 ml/kg) 1 h after the last MCGP administration. Mice were killed 24 h after CCl₄ injection, and plasma and liver tissue were collected.

For the acetaminophen (APAP)-induced acute liver injury, male Balb/c mice were randomly divided into 6 groups. 1) Control group (n = 10). 2) Model group (n = 10). 3) MCGP-L group (n = 10, 50 mg/kg). 4) MCGP-M group (n = 10, 100 mg/kg). 5) MCGP-H group (n = 10, 200 mg/kg). 6) Positive group (n = 10, 50 mg/kg of silymarin). Mice were first orally given MCGP once a day for 10 days, and then followed with intragastric administration of APAP (300 mg/kg). Mice were killed 6 h after APAP administration, and plasma and liver tissue were collected.

All mice were cared for and killed according to guidelines provided by the Institutional Animal Care and Use Committee of Tongji Medical College.

The livers were fixed in 10% formaldehyde and embedded in paraffin. The liver tissue lesions were stained with hematoxylin and eosin (H&E) for histopathology. Immunohistochemistry was performed using CYP2E1 (Proteintech), BCL-xL (Abcam), PCNA, and Bax antibodies (Cell Signaling Technology). Aperio Image Scope software was applied to analyze the quantification of PCNA positive cells.

Whole-blood samples were allowed to coagulate for 15 min at room temperature and then centrifuged at 4,000 g at 4°C for 10 min to separate the serum. The liver homogenate was centrifuged at 12000 g, 4°C for 10 min. Serum concentrations of alanine aminotransferase (ALT), aspartate aminotransferase (AST), liver homogenate of glutathion (GSH), superoxide dismutase (SOD), and malondialdehyde (MDA) were detected with commercial test kits according to the manufacturer’s instructions (Nanjing Jiancheng Bioengineering Institute, Nanjing, Jiangsu, China).

Total RNA was reverse-transcribed into cDNA using a transcription kit (ABP, United States). Quantitative RT-PCR (qRT-PCR) was performed using SYBR Green qPCR Mix (Vazyme Biotech Co., Ltd., China) with 0.2 μM forward and reverse primers in a final volume of 10 μl, and detected by ABI QuantStudio 5 (Thermo Fisher Scientific, United States). The resulting cDNA was amplified by incubating at 95°C for 5 min, 40 cycles of denaturation at 95°C for 10 s, annealing at 60°C for 20 s, and extension at 72°C for 30 s. Values were exhibited relative to β-actin. The corresponding primer sequences are listed in Supplementary Table S1.

Total proteins from livers were lysed in radioimmunoprecipitation assay (RIPA, Beyotime, China) buffer, and 30 μg total proteins were used for each blot. The samples were separated by SDS-PAGE and transferred onto a nitrocellulose filter (NC, Millipore, United States) membrane by electroblotting. The membranes were blocked for 1 h in blocking buffer (Beyotime, China) and then incubated overnight with 1:1,000 dilutions of anti-Nrf2, anti-NQO1, anti-HO-1, anti-Keap1, anti-BCL-xL, anti-GSDMD, anti-caspase 1, anti-PCNA, anti-NLRP3, and anti-ASC (Cell Signaling Technology, United States) and 1:2,000 dilutions of anti-Bax, anti-IL1β, and anti-IL18 (Abcam, United States). After incubation with the secondary antibodies anti-mouse IgG (H + L) (DyLight™ 800, Cell Signaling Technology, United States) at 1:50,000 dilutions, membranes were imaged using a LiCor Odyssey scanner (LI-COR, United States). The protein expressions were normalized using β-actin as the reference (Cell Signaling Technology, United States) in the same sample.

The liver tissues were fixed for at least 24 h in 4% paraformaldehyde, dehydrated in graded ethanol, and embedded in paraffin. The paraffin sections (5 μm) were incubated in a TUNEL reaction mixture from a kit according to the manufacturer’s instructions. The slides were scanned using a digital microscopy scanner Pannoramic MIDI (3DHISTECH, Hungary).

The liver homogenate was centrifuged at 12000 g, 4°C for 10 min. The liver homogenate was collected for IL1β and IL18 ELISA assay (Boster, Wuhan, China) according to the manufacturer’s instruction.

The experimental results are expressed as means ± SEM of at least triplicate measurements. The differences were evaluated by Student’s t-test with GraphPad Prism software (GraphPad Prism version 5.01 for Windows, Sand Diego, CA). The differences with *P and #p < 0.05 were considered statistically significant.

MCGP was synthesized according to the methods in Figure 1A. We first detected the cytotoxicity effect of MCGP on different cell lines. As a result, MCGP exhibited little cytotoxicity on 7,721 and LX-2 at dosages of 800 and 1,000 μM (Figures 1B,D). Interestingly, MCGP slightly promoted the cell growth at dosages of 25–400 μM at all the detected cell lines (Figures 1B–F). Accordingly, we presumed that MCGP might also promote cell growth in other conditions.

APAP and CCl₄ are two well-used hepatotoxins that can induce hepatocyte apoptosis and liver injury in vitro and in vivo. Therefore, we identified the cell viability of AML12 cells co-treated by MCGP and APAP or CCl₄. As expected, MCGP promoted the cell viability of AML12 cells treated by APAP or CCl₄ (Figures 2A,B). Moreover, MCGP also inhibited cell apoptosis of AML12 cells injured by APAP or CCl₄ (Figures 2C,D). In addition, we also found that MCGP could suppress the ROS generation induced by H₂O₂ (Figure 2E). This result prompted that MCGP might suppress oxidative stress by scavenging free radicals. However, MCGP exerted little effect on ABTS^•+ radical inhibition (Figure 2F). Considering this, we supposed that MCGP could not directly scavenge free radicals, while it could scavenge through intracellular processes.

In order to comprehensively understand the effect of MCGP on liver cells, RNA-seq was conducted to compare the gene expression between 40 μM MCGP-treated HepG2 cells and DMSO control. As a result, 100 differentially expressed genes were screened between MCGP and control, including 58 down-expressed and 42 up-expressed genes regulated by MCGP (Figure 3A). The KEGG pathway analysis of these genes indicated an enrichment in metabolic pathways and PI3K-Akt signaling pathway (Figures 3B,C). The gene ontology (GO)-function analysis showed an enrichment in components of membranes (Figures 3D,E).

Oxidative stress, hepatocyte apoptosis, and liver regeneration are common processes engaged in APAP- or CCl₄-induced hepatotoxicity (Scholten et al., 2015; Ramachandran and Jaeschke 2019). Hence, we screened the gene expression of related genes in RNA-seq results. As shown in Figure 3F, the Nrf2/HO-1/NQO1 signaling pathway components were almost upregulated by MCGP. In addition, MCGP also increased the liver regeneration–related genes PCNA, Ki67, BCL-2, and IL18, while it decreased hepatocyte nuclear factor 4 alpha (HNF4A) (Figure 3G). Taking together, we supposed that MCGP might protect the liver cell from APAP- or CCl₄-induced hepatotoxicity through the activation of the Nrf2 signaling pathway. However, the effect and mechanism of MCGP in vivo still remains unknown.

To further verify the hepatoprotective effect of MCGP in vivo, we used APAP- or CCl₄-induced liver injury mice models. As shown in Figure 4A, the liver histological evaluation in mice 6 h after APAP administration showed that pretreatment of MCGP alleviated APAP-induced intrahepatic hemorrhage and nuclear pyknosis in mice. Similarly, pretreatment of MCGP also improved liver damage in mice 24 h after CCl₄ induction (Figure 5A). MCGP also inhibited the APAP-/CCl₄-induced mice weight loss, and decreased liver weight and liver index (Figures 4B,C and Figure 5B). The liver necrotic areas were increased in CCl₄-administrated mice, while MCGP reduced the increased necrotic areas (Figure 5C). MCGP also significantly inhibited the increase of the serum ALT/AST level in APAP-/CCl₄-induced mice (Figure 4D and Figure 5D).

GSH, SOD, and MDA are critical markers indicating liver damage during APAP-CCl₄-injured liver. GSH and SOD can protect cells from oxidative damage. Previous studies have revealed significant decrease in liver antioxidase activity GSH and SOD in APAP-/CCl₄-induced acute liver injury (Xie et al., 2016). In addition, the lipid peroxidation product MDA was detected to reflect the liver oxidative damage (Ingawale et al., 2014). As a result, the activity of GSH and SOD were notably decreased in APAP-/CCl₄-treated mice, accompanied by a significant increase in the MDA level. The pretreatment of MCGP improved GSH and SOD levels and inhibited the MDA level in both APAP- or CCl₄-treated mice (Figures 4E,F and Figure 5E). Silymarin, a well-used hepatoprotective herbal drug, was used as the positive control.

A small amount of hepatotoxins (such as APAP and CCl₄) are metabolized by cytochrome P450 enzymes, mainly via CYP2E1 isoform. APAP is metabolized by CYP2E1 to NAPQI, which exhausts GSH and leads to mitochondrial damages and necrotic cell death (Lancaster et al., 2015). Similarly, CCl₄ is metabolized by CYP2E1 to highly reactive free radical metabolites that initiate membrane lipid peroxidation (Fahmy et al., 2016). The overactivation of CYP2E1 induced by APAP or CCl₄ leads to oxidative stress and resultant liver damage. In our result, MCGP significantly depressed the protein increase of liver CYP2E1 on APAP-/CCl₄-induced mice (Figures 6A,E). Glutamate–cysteine ligase catalytic subunit (Gclc) and glutamate–cysteine ligase modifier subunit (Gclm) are two subunits of glutamate–cysteine ligase, the first rate-limiting enzyme of glutathione synthesis (Jin et al., 2019). The glutathione S-transferase (GST) family includes glutathione S-transferase pi 1 (Gstp1) and glutathione S-transferase pi 2 (Gstp2), which play an important role in detoxification by catalyzing the conjugation of many hydrophobic and electrophilic compounds with reduced glutathione (Mari et al., 2009). CCl₄ administration inhibits the transcription level of Gclc, Gclm, Gstp1, and Gstp2, while MCGP rescued their decrease (Figure 6B).

Oxidative stress is closely relevant with APAP-/CCl₄-induced acute liver injury (Lu et al., 2016; Wang S. et al., 2017). Nrf2 is a transcription factor that plays an important role against oxidative stress by mediating the expression of many endogenous antioxidants such as GSTs, GCLs, and quinine oxidoreductase 1 (NQO1) (Bataille and Manautou 2012). MCGP significantly activated the gene and protein expression of Nrf2, NQO1, and HO-1 in APAP-/CCl₄-induced mice liver (Figures 6C–F). In addition, MCGP suppresses the expression of kelch-like ECH associated protein 1 (Keap1) (Figures 6D–F), thus dissociating Nrf2 from Keap1 and leading to the nuclear translocation of Nrf2. These results are also consistent with the RNA-seq result that MCGP upregulated the transcription of Nrf2 signaling molecules (Figure 3F).

Nrf2 has been confirmed with the function on promotion of liver regeneration and inhibition of hepatocyte apoptosis (Chan et al., 2021; Ghanim and Qinna 2022). Therefore, we explored the effect of MCGP on liver regeneration and apoptosis. As shown in Figure 7A, both APAP and CCl₄ increased the number of PCNA-staining hepatocytes in livers from APAP-/CCl₄-intoxicated mice. Similarly, MCGP markedly increased the number of PCNA-staining hepatocytes in livers from APAP-/CCl₄-intoxicated mice (Figures 7A,B). In addition, MCGP promoted the protein expression of PCNA in APAP-/CCl₄-intoxicated mice livers (Figure 7G). Hepatocyte nuclear factor 4 alpha (HNF4A), colony stimulating factor 1 receptor (CSF1), MYB proto-oncogene, transcription factor (MYB), and cAMP responsive element binding protein 5 (CREB5) are downregulated genes revealed by RNA-seq after MCGP administration, while research studies showed that the four genes were related to liver regeneration, anti-apoptosis, and cell proliferation (Lancaster et al., 2015; Wang Y. et al., 2017; Wu et al., 2018; Zhou et al., 2020). As expected, we found the increase of these genes in MCGP-treated livers from CCl₄-intoxicated mice, which is incompatible with the RNA-seq result (Figure 7C).

CCl₄ exposure significantly increased hepatocyte apoptosis, while MCGP remarkably decreased the TUNEL-positive hepatocytes in livers from CCl₄-intoxicated mice (Figure 7D). In addition, the apoptosis-related gene and protein expressions of Bax and Bcl-xL in the liver tissue were also detected by immunohistochemistry, qRT-PCR, and Western blotting. As a result, CCl₄ had a slight effect on the expression of BCL2-like 1 (Bcl-XL) (Figures 7E,G), and significantly increased the transcription level of BCL2 apoptosis regulator (BCL-2) (Figure 7F). MCGP notably promoted the expression of Bcl-XL and BCL-2 in APAP-/CCl₄-induced mice livers (Figures 7E–G). In addition, MCGP also decreased the gene and protein level of BCL2 associated X, apoptosis regulator (BAX) (Figures 7E–G).

Previous studies revealed that IL18 accelerated liver regeneration after partial hepatectomy or APAP-induced liver injury (Zhang et al., 2014; Shi et al., 2020). IL-1β and IL-18 are two main pro-inflammatory cytokines produced during the recruitment of caspase 1 to cleaved GSDMD into its active form GSDMD-N and insert in cell membranes (de Vasconcelos et al., 2016; Shojaie et al., 2020). Here, we found that MCGP activated the transcription of caspase 1, IL1β, and IL18 and elevated the liver content of IL1β and IL18, inducing the protein expression of GSDMD-N, cleaved caspase 1, IL1β, and IL18 (Figures 8A–C).

APAP- or CCl₄-induced liver injury is characterized by sterile inflammation, functioning as damage-associated molecular patterns (DAMPs), which can transcriptionally activate inflammatory cytokines (TNFα, IL1β, IL6, and IL10) and chemokines (MCP-1, MIP-2, and IL8) (Dambach et al., 2002; James et al., 2005). As a result, CCl₄ markedly elevated the serum level of pro-inflammatory cytokines IL6 and TNFα (Figure 8D). In contrast, MCGP decreased the level of IL6 and TNFα (Figure 8A). In addition, MCGP inhibited the elevated transcription level of inflammation factors IL6, TNFα, MCP-1, and IL12 by CCl₄, while it increased the expression of anti-inflammatory factor IL10 (Figure 8E).