2.1 Animals

Male Sprague–Dawley rats weighing 180–200 g were purchased from the animal center of Fujian Provincial Center for Disease Control and Prevention. Rats were kept in an environmental temperature- and humidity-controlled room (22–25°C, 40%–70% humidity) with 12-h light/12-h dark cycles and with food and water ad libitum. All animal experiments were approved by the Animal Care Committee of Fujian Medical University and carried out in accordance with the Guidelines for Animal Experiments at Fujian Medical University (approval number: 2019–0107).

2.2 Experimental design and the rat model of ACLF

To induce ACLF in the rats, after 7 days' stabilization, the rats were first intraperitoneally (ip) injected with 50% carbon tetrachloride (CCl₄, 1 mL/kg, every Tuesday and Friday) (CN: 61554; Hengxing) which was dissolved in peanut oil, and porcine serum (0.5 mL, every Wednesday) (S9060; Solarbio) for 6 weeks to induce chronic liver injury (CLI).^{11, 12} Then, lipopolysaccharide (LPS, 80 μg/kg) (L2880; Sigma-Aldrich) and d-galactose (d-GalN, 0.5 g/kg) (G0500; Sigma-Aldrich) were administrated ip once to further induce the acute liver injury (ALI).^{13, 14}

To explore the effects of RvD1 on the rat model with ACLF, in total, 42 rats were used in this study. The rats (n = 42) were randomly allocated into four groups: (1) normal control group (Con, n = 6): rats were only treated with normal saline; (2) ACLF rat group (ACLF, n = 12): rats were treated with normal saline 1 h before the acute hepatic injury; (3) low-dose RvD1 group (RvD1-L, n = 12): rats were ip injected with RvD1 (10012554; Cayman) 0.3 μg/kg 1 h before the induction of ALI; and (4) high-dose RvD1 group (RvD1-H, n = 12): rats were ip injected with RvD1 1 μg/kg 1 h before the induction of ALI. Specimens of groups 2–4 were collected at 8 and 24 h after LPS +d-GalN injection, respectively (n = 6). After the indicated treatment time period, rats were killed, and samples of whole blood, liver, and spleen were collected for the subsequent assessment.

2.3 Histological assessment

The liver samples were fixed in 10% formalin solution for 24 h, followed by embedding in paraffin and cutting into 10 μm sections. Necrosis of hepatic cells was assessed by hematoxylin-eosin (H&E) staining, and hepatic fibrosis was assessed by Sirius red staining. The slides were observed with a microscope (Leica DM2500). The confluent necrosis and/or bridging necrosis (CN/BN) scores were calculated according to the histological grading and staging of chronic hepatitis (Ishak's score),^{15, 16} and the criteria were shown in Table 1. The histological assessment was performed and evaluated blindly by a professional pathologist.

2.4 Biochemical analysis

Serum was collected by centrifuging the whole blood at 4000 rpm for 10 min at 4°C. The serum levels of aspartate aminotransferase (AST) and alanine aminotransferase (ALT) were measured by Backman Kurt au5800 automatic biochemical analyzer (Beckman Kurt Co., Ltd.).

2.5 Flow cytometry assay

The spleen samples were ground and filtered to obtain the single-cell suspension. Red blood cell lysis buffer (R1010; Solarbio) was added to lyse the erythrocytes. Cells were first stained with CD4 monoclonal antibody (OX35), APC, eBioscience™ (17-0040-80; Thermo Fisher Scientific) for 30 min at 4°C, then the nuclear membrane was broken by Foxp3 transcription factor staining buffer set (00-5523-00; Thermo Fisher Scientific), and finally, cells were stained with Foxp3 monoclonal antibody (FJK-16s), PE, eBioscience™ (12-5773-80; Thermo Fisher Scientific) for 30 min at room temperature to mark the Treg cells. The detection of Treg cells was performed using a BD Accuri C6 Plus flow cytometer (BD).

2.6 Real-time polymerase chain reaction (RT-PCR)

Total RNA of rat liver was extracted by RNA isolation total RNA extraction reagent (R401-01; Vazyme). The complementary DNA (cDNA) was synthesized by HIScript®Ⅲ RT SuperMix for qPCR (+gDNA wiper) (R323-01; Vazyme). RT-PCR analysis of the cDNA was performed using the ABI 7500 Real-Time PCR system (Life) with Taq Pro Universal SYBR qPCR Master Mix (Q712-02; Vazyme). The relative messenger RNA (mRNA) expression level of the interested genes was determined using the cycle threshold method ($${2}^{-\unicode{x02206}\unicode{x02206}{C}_{{\rm{t}}}}$$)¹⁷ with Actb as an internal housekeeping gene. Primer sequences are shown in Table 2.

2.7 Data analysis

All data were analyzed using GraphPad Prism version 8 software. The results are presented as mean ± standard error of the mean. For the comparison among several groups one-way analysis of variance followed by Holm–Sidak multiple comparison test was used. A p < .05 was considered statistically significant.

3.1 The rat model of ACLF

To induce the pathological changes of ACLF, the rat model of ACLF was induced by subsequent injection of different compounds as illustrated in Figure 1A,^{13, 14} and the mortality of rats is show in Table 3. Compared with the healthy rats, the liver capsules of ACLF rats were thickened and there was hard texture adhered to the surrounding tissues. Sirius red staining of liver tissue showed that in the liver of ACLF rats, there was a significant increase in the proliferation of collagenous fiber and the formation of fibrous septa, while healthy rats had no fibrosis (Figure 1B). Next, the CN/BN of hepatic cells was assessed by H&E staining. As shown in Figure 1B, the liver tissue organization of the healthy rats was normal, and their hepatic lobules were well-arranged. While H&E staining of rat liver pathology showed the disordered structure of hepatic lobule, the swelling, and apoptosis of hepatic cells, inflammatory cells infiltration, and dilated and bleeding hepatic sinus, accompanied by confluent necrosis in ACLF (8 h) rats. The CN/BN scores were significantly higher than the healthy rats (Figure 1C). Though H&E staining of the rat liver pathology in ACLF (8 h) rats also showed the disordered structure of hepatic lobule, in ACLF (24 h) rats, there was a significant decrease in the swelling and apoptosis of the hepatic cells, the inflammatory cells, and the necrosis of the hepatic cells(Figure 1C). Serum transaminases are the classical indicators of hepatocyte inflammation and necrosis. The serum levels of ALT (Figure 1D) and AST (Figure 1E) in ACLF rats were increased substantially at 8 h but showed a dramatic decline at 24 h, which was in line with the previous study.¹³ Thus, the changes in the gross specimens, liver histology, and serological indicators were observed by 8 h but not 24 h after the ALI was induced on the CLI status, which indicated that the rat of the ACLF model was successfully induced at the early time point. Therefore, an early time point (8 h) after ACLF generated in rats was further explored.

3.2 The protective effects of RvD1 on ACLF in rat models

To explore the protective effects of RvD1 in rat models of ACLF, different doses of RvD1 were ip injected 1 h before the induction of ALI in ACI rats (Figure 2A). The Sirius red staining of rat liver pathology showed that compared with the healthy rats, the proliferation of the collagenous fiber and the formation of fibrous septa were significantly increased in the liver of ACLF rats regardless of RvD1 treatment (Figure 2B). H&E staining of rat liver showed that RvD1-L rather than RvD1-H administration improved the pathologic changes identified in ACLF rats (Figure 2B). To be specific, the disordered structure of the hepatic lobule, the swelling, and apoptosis of hepatic cells, a little inflammatory cell infiltration, and the inflammatory cells and necrosis of hepatic cells significantly reduced the RvD1-L group (Figure 2B). H&E staining of rat liver pathology of the RvD1-H group showed the disordered structure of hepatic lobule, the swelling, and apoptosis of hepatic cells, inflammatory cells infiltration, dilated and bleeding hepatic sinus, accompanied by confluent necrosis (Figure 2B). As shown in Figure 2C, the RvD1-L group had lower CN/BN scores than the ACLF group. However, the RvD1-H group still has severe necrosis. In consistence with these pathological changes in the liver, compared with the ACLF rats, a low dose of RvD1 significantly decreased the serum ALT level but a high dose of RvD1 led to an even higher level of ALT (Figure 2D). These findings indicate that a low dose of RvD1 may have a protective effect on liver function in ACLF rats.

Both the anti-inflammatory effect and the increased ability of hepatocyte regeneration could contribute to the recovery of hepatic pathology in ACLF rats. To evaluate if the liver regeneration in the rat can be affected by the disease of ACLF and/or RvD1 treatment, a marker of cell proliferation (Ki-67) was assessed in the liver by immunohistochemical staining and the percentage of Ki-67 positive cells was further calculated. It was shown that in the ACLF group, the percentage of Ki-67 positive cells was induced by approximately three times compared with the Con group while the treatment of RvD1-L did not play a role in cell proliferation in the rat model of ACLF (Figure S1A,B).

3.3 The effects of RvD1 administration on Treg cells

To explore the potential underlying mechanisms of the protective effects of a low dose of RvD1 on liver function in a rat model of ACLF, Treg cells which have been linked to maintaining immune homeostasis^18-20 were further evaluated. The percentage of Treg cells in the spleen was examined using flow cytometry assays and the results showed that the percentage of Treg cells in the spleen of ACLF rats was lower than that in the healthy rats, which was restored by a low-dose treatment of RvD1 (Figure 3A,B). The gene expression of the classical downstream cytokines including interferon γ (IFNγ) secreted by T-helper type 1 (Th1) cells, interleukin (IL)-4 secreted by Th2 cells, and IL-17 secreted by Th17 cells were detected by RT-PCR in the liver tissue. The mRNA levels of Ifng and Il-4 were significantly increased in the liver of ACLF rats compared with the control rats, while these changes could not be affected by the treatment of RvD1-L (Figure 3C,D). Furthermore, the increased liver mRNA levels of Il-17a in ACLF rats were significantly attenuated by the treatment of a low dose of RvD1 (Figure 3E). Next, the investigation of the ratio of Foxp3/RORγt (the critical transcription factor of Treg²¹ and Th17,²² respectively) which represents a classic way for evaluating the balance of Treg and Th17 cells was further explored in the liver in this study. No significant difference was identified in the changes in Foxp3 and in Rorc between the ACLF group and the RvD1-L group (Figure 3F,G). However, the ratio of Foxp3/Rorc in mRNA levels was higher in the RvD1-L group than that in the ACLF group (Figure 3H). These results indicate that RvD1 treatment could increase the proportion of Treg cells in the spleen of ACLF rats and regulate the balance of Treg and Th17 in the liver.

3.4 The effects of RvD1 on immunologic cytokines in the liver microenvironment

To explore the possible mechanisms underlying the increased proportion of Treg cells in the liver of ACLF rats treated with RvD1-L, we further determined the immunologic cytokines (e.g., transforming growth factor-β [TGF-β] and IL-6) as they are closely related to the Treg cell differentiation in the liver microenvironment.^23-26 Surprisingly, compared with the Con group, the mRNA expression of Tgfb1 was not affected in the liver of ACLF rats regardless of the treatment of RvD1 (Figure 4A). However, the mRNA levels of Il-6 were significantly elevated in the liver of ACLF rats compared with the control rats, but this increase in the Il-6 gene expression restored to a control level by a low-dose treatment of RvD1 in ACLF rats (Figure 4B).