Current Status of ACLF Animal Model Research

Acute-on-chronic liver failure (ACLF) refers to a clinical syndrome characterized by acute decompensation of liver function in a short period on the basis of chronic liver disease. The pathogenesis of ACLF is complex, the condition is perilous, progresses rapidly, and is accompanied by multiple organ failure, with a high short-term mortality rate and no specific effective treatment measures currently available. Therefore, establishing a suitable, stable, and highly clinically simulated ACLF animal model is of great value and significance for elucidating the pathogenesis of ACLF, helping clinicians develop more effective early diagnosis and treatment strategies, thereby improving the success rate of clinical ACLF treatment. However, due to the complex etiology and numerous complications of ACLF, there have been significant challenges in the development of its animal models. This article reviews the currently commonly used methods for establishing ACLF animal models.

1Overview of ACLF

Currently, there is no unified standard for the definition of ACLF domestically and internationally. In China, ACLF is defined as a syndrome characterized by acute jaundice deepening (total bilirubin ≥ 10 times the upper limit of normal or daily increase ≥ 17.1 μmol/L) and coagulopathy (prothrombin activity ≤ 40% or international normalized ratio ≥ 1.5) caused by various triggers on the basis of chronic liver disease, which may be accompanied by complications including hepatic encephalopathy, ascites, electrolyte disturbances, infections, hepatorenal syndrome, and hepatopulmonary syndrome, as well as extrahepatic organ failure. According to different underlying chronic liver diseases, ACLF can be divided into three types: Type A: ACLF occurring on the basis of chronic non-cirrhotic liver disease; Type B: ACLF occurring on the basis of compensated cirrhosis, usually within 4 weeks; Type C: ACLF occurring on the basis of decompensated cirrhosis. The pathogenesis of ACLF is not yet fully elucidated and may be related to susceptible constitution, triggering factors, inflammatory response, organ failure, etc. Common risk factors primarily include bacterial infection, alcohol, variceal bleeding, hepatotoxic drugs, obesity, and dyslipidemia. Currently, there is a lack of specific effective treatment options for ACLF, which mainly relies on drug therapy, artificial liver support therapy, and liver transplantation.

2Methods for Establishing ACLF Animal Models

Currently, the approach to establishing ACLF animal models involves inducing chronic liver injury in experimental animals through various methods, and then implementing acute attacks on the basis of chronic injury, resulting in acute liver failure.

Different methods are used in the chronic liver injury induction phase, mainly including CCl₄ injection, thioacetamide injection, serum albumin injection, and bile duct ligation. In the acute attack phase, different acute attack methods can be used, including the sole use of CCl₄ or D-galactosamine (D-GaL N) combined with LPS and D-GaL N, etc. Hassan et al. provided a comprehensive summary of the methods currently used to establish ACLF animal models. This article mainly summarizes, compares, and reflects on four commonly used methods: CCl₄ method, thioacetamide method, serum albumin method, and bile duct ligation method.

Rodents, especially mice, are the preferred animals for ACLF animal models due to their small size, short lifespan, short gestation period, ease of breeding in captivity, and significant genetic similarity to humans, along with the convenience of genetic manipulation. Therefore, mice have become the most commonly used experimental animals for simulating ACLF. Additionally, larger animal models can be chosen from pigs, rabbits, and non-human primates. However, non-human primates such as monkeys, apes, and baboons are very similar to humans, but their cost and the strictness of modeling methods make it difficult to conduct large-scale studies, thus few researchers establish ACLF animal models using them.

2.1 CCl₄ Combined with LPS/D-GaL N Method

CCl₄ is a compound that can cause hepatocyte necrosis and is commonly used as a modeling agent to induce various liver injuries. CCl₄ generates free radicals CCl3· and CCl3OO· through the activation of cytochrome P450 in liver cells, which attack the polyunsaturated fatty acids on membrane structures, promoting lipid peroxidation of biological membranes such as cell membranes and mitochondrial membranes, leading to the production of lipid peroxides that further damage various biological membranes, reducing their stability and integrity, increasing permeability, causing the outflow of various enzymes within cells and other cellular damage, ultimately resulting in hepatocyte apoptosis and necrosis. Lipopolysaccharide (LPS), also known as endotoxin, is a component on the surface of Gram-negative bacterial cell walls that stimulates immune cells to release inflammatory factors, leading to hepatocyte apoptosis and necrosis. D-GaL N can cause hepatocyte damage by inhibiting mRNA synthesis and partially inhibiting the transfer of glycoproteins, while activating macrophages and neutrophils, promoting the occurrence of inflammatory responses, leading to hepatocyte apoptosis. The toxic dose of D-GaL N primarily causes hepatocyte necrosis by depleting ATP and inhibiting the synthesis of macromolecules. Under the synergistic effect of LPS and D-GaL N, a large number of hepatocytes in experimental animals die within a short period, severely impairing liver physiological function.

The ACLF animal model induced by CCl₄ combined with LPS/D-GaL N can simulate liver injury caused by human chemical toxic substances and is currently a commonly used method. This method often uses intraperitoneal injection as the administration route, which is easy to operate and allows for precise control of the injection dose, while the absorption of the drug is faster than that of subcutaneous injection, avoiding local skin inflammation, and slower than intravenous injection, preventing acute drug poisoning. However, researchers have certain differences in the selection of experimental animals, induction methods, and induction doses.

Currently, the ACLF animal model established by CCl₄ combined with LPS/D-GaL N can be divided into rat models and mouse models. Most studies have confirmed that during the chronic liver injury induction phase, whether using rats or mice, about 12 weeks of modeling time can lead to significant chronic liver injury in either species. During the 12-week modeling process, to maintain the liver function in a damaged state and prevent complete recovery, while avoiding frequent dosing that could lead to death due to overdose, the recommended administration frequency is twice a week. Depending on the type of rodent used for modeling, the injection dose of CCl₄ varies. Zhang Huiyun et al. injected 20% CCl₄ oil solution at a dose of 5 mL/kg into 3-4 week old male BALB/C mice; Gao An et al. used 3-4 week old SPF male C57BL/6 mice as experimental animals, injecting 20% CCl₄ (diluted with olive oil) at a dose of 5 mL/kg, twice a week for a total of 12 weeks. The steps for inducing chronic liver injury in rats are relatively more complicated. Research has shown that if male SD rats (body weight 160-170 g) are selected, although the same method of intraperitoneal injection of 50% CCl₄ plant oil solution is used continuously for 12 weeks with an interval of every 3 days, the dosage needs to be adjusted accordingly compared to mice, where the dosage for the first month is 1.5 mL/kg and for the second and third months is 2.0 mL/kg. Additionally, to prevent rats from developing resistance to CCl₄, the dosage needs to be increased after 4 weeks, usually to about 0.3 times the initial dosage, while the dosing for inducing liver cirrhosis in mice remains relatively stable.

In the acute attack phase, different acute attack methods can be used, divided into the sole use of CCl₄ or D-GaL N, or LPS combined with D-GaL N. Zhang Haiyan et al. simulated the above three acute attack methods to compare whether different methods could successfully induce acute liver failure on the basis of chronic liver injury. After successfully inducing chronic liver injury in rats, the experimental rats were divided into groups A, B, and C, with group A receiving 2 g/kg of D-GaL N; group B receiving 100 μg/kg of LPS combined with 0.5 g/kg of D-GaL N; and group C receiving 5 mL/kg of 50% CCl₄ plant oil solution, all administered via intraperitoneal injection. The results showed that all three experimental groups exhibited large or sublarge areas of hepatocyte necrosis, consistent with the manifestations of acute liver failure, with group B showing a greater degree of hepatocyte necrosis than groups A and C, with group C being the least affected. This study indicates that all three methods can successfully induce acute liver failure on the basis of chronic liver injury. Among them, the sole injection of D-GaL N resulted in the longest survival time, suggesting that this model has a longer therapeutic window, making it more suitable for drug screening studies, while the occurrence of acute liver failure induced by sole CCl₄ injection, although rapid, did not show as severe hepatocyte necrosis as that induced by LPS combined with D-GaL N. The acute liver failure induced by LPS combined with D-GaL N occurred quickly, with the most severe degree of hepatocyte necrosis, thus, compared to the other two treatment methods, LPS combined with D-GaL N may be more suitable for experimental research on the pathophysiological mechanisms of ACLF. However, different studies may use different doses of LPS and D-GaL N to induce the occurrence of acute liver failure. Zhang Huiyun et al. established an ACLF mouse model by intraperitoneally injecting 1 g/kg of D-GaL N and 10 μg/kg of LPS 3 days after the last administration at the end of 12 weeks. Gao An et al. established an ACLF mouse model by giving a single intraperitoneal injection (LPS 0.5 mg/kg, D-GaL N 400 mg/kg) 10 days after the last administration at the end of 12 weeks. It can be seen that when using the acute attack method of LPS combined with D-GaL N, researchers have differences in the selection of doses for LPS and D-GaL N, considering that LPS has a strong toxic effect, the dose required for acute attack is usually much lower than that of D-GaL N, but the optimal doses for the combined use of these two drugs during acute attacks still need further exploration.

In summary, multiple studies have confirmed that the CCl₄ combined with LPS/D-GaL N method for establishing ACLF animal models is simple, stable, and suitable for various clinical studies. However, some studies have indicated that after the inducing factors are stopped, the degree of liver fibrosis and inflammation gradually decreases in the CCl₄ model. Notably, some scholars believe that bacterial infections are extremely common among ACLF patients, and acute bacterial infection is one of the most common triggering events for ACLF in Asia, often closely associated with systemic inflammation, poor clinical outcomes, and high mortality rates. Merely using CCl₄ or D-GaL N injections does not adequately simulate the role of infection in the occurrence of liver failure and damage to extrahepatic organs. Therefore, the CCl₄ combined with LPS/D-GaL N method still has limitations and deficiencies in simulating the progression of ACLF. In recent years, some studies have established more complete ACLF mouse models that simulate the disease course and clinical characteristics of ACLF, which are divided into three stages: inducing chronic liver injury, acute liver injury, and bacterial infection. Specifically, male C57BL/6J mice are first injected intraperitoneally with CCl₄ (0.2 mL/kg) twice a week for 8 weeks to induce chronic liver injury. In the acute attack phase, a double dose of CCl₄ (0.4 mL/kg) is injected, followed by intraperitoneal injection of Klebsiella pneumoniae strains or performing cecal ligation and puncture. This method successfully constructs a severe liver injury ACLF mouse model by inducing bacterial infection, better simulating the core disease course of ACLF and the potential damage to extrahepatic organs during the disease course. This model not only has an appropriate survival period for intervention studies but also can be well standardized, is easy to obtain materials, quick to implement, and easy to promote, providing a reliable platform for related mechanism studies and new therapeutic target screening for ACLF.

2.2 Thioacetamide (TAA) Combined with LPS Method

TAA is a hepatotoxic drug that, after being ingested by the human body, can be metabolized into TAA-sulfoxide by cytochrome P450 mixed-function oxidase in liver cells, which is further metabolized into intermediate metabolites and other polar molecules and combines with macromolecules in the liver, causing changes in liver cell function and subsequently necrosis. A single intraperitoneal injection of TAA can lead to acute hepatitis, while repeated intraperitoneal injections can lead to hepatocyte necrosis, regeneration nodule formation, capillary bile duct proliferation, portal hypertension, and ultimately liver cirrhosis. The mechanism of chronic liver injury induced by TAA is similar to the changes in human liver cirrhosis caused by various etiologies in terms of histological and biochemical metabolic changes, and the liver fibrosis induced is more stable and persistent.

de Mesquita et al. administered 200 mg/kg of TAA dissolved in 0.9% saline to male Wistar rats via intraperitoneal injection about 1 hour before the TAA injection, twice a week for 12 weeks, to establish a chronic liver disease animal model. Tripathi et al. injected male rats (150-200 g) with saline-dissolved TAA at a dose of 250 mg/kg, twice a week for 10 weeks, to establish a chronic liver disease rat model without ascites, followed by intraperitoneal or intravenous injection of LPS (1 mg/kg) in compensated rats to establish an ACLF rat model.

The TAA-induced animal model can also simulate liver injury caused by human chemical toxic substances, and compared to the CCl₄ model, its induced liver injury is more stable and persistent, making it less reversible. However, due to the strong toxicity of TAA and its significant harm to humans, it is not commonly used for establishing ACLF animal models.

2.3 Serum Albumin Method

Bhunchet et al. proposed that repeated exposure to allogenic serum leads to the production of large amounts of immune complexes in the body, promoting the production of numerous cytokines by endothelial cells and Kupffer cells, ultimately leading to liver fibrosis. Currently, bovine serum albumin (BSA), human serum albumin (HSA), and porcine serum (PS) are mainly used to induce immune-mediated liver injury models, and subsequently combined with D-GaL N and LPS to establish ACLF animal models. This method usually selects rats as experimental animals, and the differences in establishing ACLF models using serum albumin mainly lie in the type of albumin, dosage, administration route, and timing of administration.

Commonly used serum albumins include BSA, HSA, and PS. Liu Xuhua et al. were the first to use HSA combined with D-GaL N/LPS to establish a rat ACLF model in China. The BSA and HSA methods for inducing chronic liver injury are usually divided into two stages: the immune phase, where serum albumin is diluted with saline and emulsified with an equal volume of complete Freund’s adjuvant, with subcutaneous injections of 0.5 mL (containing 4 mg of albumin) administered four times (on days 1, 15, 25, and 35). The second phase is the immune attack phase, where serum albumin is injected intravenously. Liu Ziqian et al. subcutaneously injected male SPF rats with 0.5 mL of emulsified solution containing 4 mg of BSA at multiple sites on days 1, 15, 25, and 35 (four times in total); then intravenously injected BSA solution into the tail vein twice a week for a total of 6 weeks (12 times), starting with a BSA content of 2 mg per injection and gradually increasing to 4 mg per injection. Liu Xuhua et al. administered 4 subcutaneous injections of 0.5 mL of HSA emulsified solution to female Wistar rats, followed by intravenous injection of 2.5 mg of HSA, gradually increasing by 0.5 mg until reaching 4.5 mg, maintaining this dosage with two injections per week for a total of 6 weeks. Wang et al. administered multiple subcutaneous injections of 0.5 mL of emulsified solution containing 4 mg of HSA to male Wistar rats. After sensitization with HSA, the rats were intravenously injected with 4 mg of HSA twice a week for 6 weeks, leading to chronic liver injury. The main difference between the BSA and HSA methods lies in the dosage of albumin administered intravenously after sensitization; the HSA method can either start with 2.5 mg and gradually increase by 0.5 mg until maintaining at 4.5 mg or directly use a maintenance dose of 4 mg, while the BSA method typically starts at 2 mg and gradually increases to 4 mg. The method of gradually increasing the dosage may reduce the mortality rate of rats and improve the success rate of modeling, but further exploration is needed to determine which method is superior for better simulating human ACLF. Compared to the relatively complicated steps of inducing chronic liver injury with BSA or HSA, Li et al. administered 0.5 mL of PS via intraperitoneal injection to male Wistar rats twice a week for 11 weeks to establish a liver fibrosis model. The PS method for inducing chronic liver injury is simpler, easier to operate, repeatable, and safer than subcutaneous or intravenous injections, leading to a higher modeling success rate.

In the acute attack phase, the BSA, HSA, and PS methods all employ a combined use of D-GaL N and LPS, but there are differences in the doses and administration routes of D-GaL N and LPS due to the different types of albumin used to induce chronic liver injury. Among them, both the BSA and HSA methods can use the same acute attack method, where at the end of the sixth week, a combined intraperitoneal injection of 100 μg/kg of LPS and 400 mg/kg of D-GaL N is administered to induce acute liver failure based on chronic liver injury. For rats induced with PS, they will receive an intravenous injection of 50 μg/kg of LPS after 11 weeks, followed by an intraperitoneal injection of 600 mg/kg of D-GaL N 30 minutes later to induce acute liver failure.

In summary, all three methods (BSA, HSA, and PS) can successfully establish ACLF rat models. Currently, the most common modeling approach using immune induction to establish ACLF animal models is the HSA method, but the mortality rate of the models established by this method is relatively high, at 23%. Compared to the ACLF rat models established by the BSA and HSA methods, the PS model has a lower mortality rate, is easier to operate, and is more similar to human ACLF.

2.4 Bile Duct Ligation (BDL) Method

BDL induces liver fibrosis by ligating the bile duct, causing bile reflux and resulting in liver cell damage. After BDL induction, the rats become more sensitive to endotoxins, usually using D-GaL N or LPS to induce acute liver injury.

Cong Wei successfully established an ACLF rat model using this method. First, male SD rats (body weight 250-280 g) were fasted for 12 hours before surgery, and the common bile duct was ligated. After 4 weeks, a cholestatic liver cirrhosis model can be established. Then, the rats that survived after 4 weeks of bile duct ligation (having established a cholestatic liver cirrhosis model) were fasted for 12 hours, and D-GaL N was administered at a dose of 1.4 g/kg (concentration 0.2-0.4 g/mL), inducing acute liver injury based on cirrhosis after 24 hours of intraperitoneal injection.

Tripathi et al. established an ACLF animal model using BDL by first ligating the common bile duct of male rats (200-225 g) to induce secondary biliary liver cirrhosis (for a total of 28 days). In the first 4 hours before hemodynamic studies, LPS at a dose of 1 mg/kg was administered intraperitoneally or intravenously to rats in the decompensated stage of cirrhosis to establish the ACLF rat model.

The BDL model can be used to simulate liver injury caused by human bile duct obstruction, but the BDL model is complex to operate and relatively unstable.

3Challenges and Prospects in Establishing ACLF Animal Models

In summary, the CCl₄ combined with LPS/D-GaL N method, TAA combined with LPS method, serum albumin method, and bile duct ligation method are currently common methods for establishing ACLF animal models. The CCl₄ model and TAA model are both based on chemical toxic injury, which is simple and convenient to operate, but there are certain requirements for dosage selection; excessive dosage can easily lead to animal death, while too small a dosage does not meet diagnostic requirements. The serum albumin method mainly simulates immune-mediated liver injury, and the model is stable and repeatable. The BDL method simulates liver injury caused by human bile duct obstruction, which is complex to operate and causes significant damage to animals, making it relatively unstable. Therefore, there are various methods for establishing ACLF animal models, and researchers can choose a suitable method based on specific research purposes. However, the etiology and pathophysiological mechanisms of ACLF are complex and diverse, and there is heterogeneity between experimental animals and humans, while clinical situations are even more complicated, with multiple triggering factors and complications occurring simultaneously or sequentially, leading to various diseases and organ failures, making it impossible to accurately simulate the occurrence and development process of human ACLF.

ACLF occurs on the basis of chronic liver disease, usually including viral liver disease, alcoholic liver disease, drug-induced liver disease, autoimmune liver disease, etc., with a variety of causes, and different mechanisms of liver injury caused by different causes. The reactivation of HBV is the main cause of ACLF in Asia, with a high prevalence rate, and HBV-related acute-on-chronic liver failure accounts for over 70% of ACLF cases. However, the currently established ACLF animal models, including those simulating chemical toxic injury, immune-mediated liver injury, and cholestatic liver injury, cannot simulate viral liver injury. Therefore, establishing animal models of acute-on-chronic liver failure related to HBV is a significant challenge and difficulty at present. Moreover, the selection of experimental animals is also crucial for establishing suitable ACLF animal models; depending on different research purposes, it is essential to choose the most appropriate animal model. Considering the advantages of rodents, rats or mice are usually selected as experimental animals.

An ideal ACLF animal model should closely simulate the pathogenesis of liver injury in humans. Although this process can be achieved by using different drugs, toxins, or other invasive operations, adjusting dosages or other methods can attempt to simulate the mortality and survival rates of human ACLF as closely as possible, there are still challenges due to the heterogeneity between animals and humans and the uncontrollability of external factors, making it currently impossible to completely replicate human ACLF. Therefore, researchers need to further explore the selection of animals that more closely align with human pathogenesis, better toxic substances, and appropriate drug dosages.