Comparison Guide: Carbon Tetrachloride Induction and Choline Deficient Diet Models
Compare the key features of popular commercial mouse models for accelerated NAFLD and NASH including fibrosis, such as carbon tetrachloride induction and dietary choline and methionine deficiency, and next generation choline deficient models.
Developing Preclinical NASH Fibrosis Models
NASH fibrosis is driven by several mechanistic axes, along which some fast, conventional rodent models may act. This provides a range of rapid models which can prove suitable for certain drug candidates:
- Toxin induction of inflammation, steatosis, and cell death using carbon tetrachloride
- Hepatic steatosis induced through dietary choline and methionine deficiency
- Next generation choline deficient models which enhance NASH without the severe weight loss observed in the earlier model
This post compares these conventional fibrosis models, looking at their advantages and limitations for fibrosis and NASH drug development.
Carbon Tetrachloride Induction Model vs Choline Deficient Diet Models
|Modality||Chemical toxicity||Nutrient deficiency||Nutrient deficiency||Nutrient deficiency with high fat caloric content||Nutrient deficiency with high fat caloric content|
|Mechanism||Fat accumulation from impaired lipid metabolism and Golgi function; inflammation and general toxicity causes cell death and scarring||Impaired secretion of hepatic triglycerides via VLDL||Impaired secretion of hepatic triglycerides via VLDL||Impaired secretion of hepatic triglycerides via VLDL||Impaired secretion of hepatic triglycerides via VLDL|
|Weight change||Neutral or mild loss||Severe loss||Mild loss to mild gain||Mild gain||Mild gain|
|Hyperlipidemia||No||No||No||None to mild||No|
|Serum ALT & AST||Moderate||Moderate||Mild||High||High|
|Liver fibrosis||Severe||Mice: severe
Rats: none to mild
|Fibrotic induction duration||Mice: 4-8 weeks
Rats: 8 weeks
|Mice: 4 weeks||Mice: 20 weeks
Rats: 8-12 weeks
|6-8 weeks||6-9 weeks|
|Post-treatment fibrotic reversibility||High||No||Yes||Moderate||N/A|
Carbon Tetrachloride NASH Fibrosis Models
The solvent carbon tetrachloride (CCl4) is a toxin metabolized by liver cytochrome P450 to the trichloromethyl radical (–CCl3). This reacts with numerous hepatic targets to achieve multiple NASH-like effects:
- inflammation triggered by mitochondrial reactive oxygen species
- steatosis from reduced activity of lipid metabolizing enzymes and impairment to Golgi-driven VLDL secretion
- generalized cellular damage resulting in cell death and fibrosis
These deleterious effects occur without obesity or insulin resistance, and involve a significant amount of cell death that is independent of lipid storage and metabolism. Therefore, this does not represent true NAFLD. Nevertheless, there are three key advantages of the CCl4 model.
First, it is very simple to perform, typically involving twice weekly intraperitoneal injections of CCl4 for four to six weeks, depending on the susceptibility of the mouse strain. The popular and affordable lab strains BALB/c, C57BL/6, and DBA/2 have well-documented high to intermediate susceptibility. Popular rat strains including Sprague Dawley and Wistar are also suitable.
Second, the model presents many of the histological features of human disease. This includes progressive fibrosis, proceeding from pericentral areas toward severe bridging fibrosis, then hepatocellular carcinoma. Leakage of liver transaminases (ALT, AST) provides easily quantified circulatory biomarkers for liver disease progression, while cytokines TNFα and IL-6 are markers for inflammation. Severity of the disease is dose-dependent, and fibrosis may be spontaneously reversible if a suboptimal regimen is used, potentially confounding efficacy for NASH-reversing test articles. Overall, the CCl4 model is a well characterized system for evaluating efficacy of NASH therapies starting from a predictable severity.
Finally, CCl4 hepatoxicity involves many mechanisms that are targets for drug development efforts toward the cure for NASH. These include the apoptosis-inducing NLRP3 inflammasome and the profibrotic activity of transforming growth factor β. Notably, expression of the farnesylX receptor (FXR) is suppressed by CCl4. Pharmacological rescue of FXR activity using obeticholic acid (OCA) was shown to be protective in this model; OCA is widely expected to become the first FDA approved NASH drug later this year.
Aside from lack of true translatability to human NAFLD/NASH, the significant drawback of this model is systemic toxicity affecting nonhepatic organs and tissues. Another challenge of CCl4 is potential chemical hazard to the technician. This model should be performed only by trained operators with proper protective equipment and chemical safety infrastructure.
Choline Deficient Diet NASH Fibrosis Models
Choline deficiency has been leveraged in several generations of rodent dietary NASH models. Choline, along with methionine, are key nutrients required for hepatic mitochondrial β oxidation and very low density lipoprotein (VLDL) synthesis. When fed a methionine/choline deficient (MCD) diet, triglyceride export is inhibited, rapidly inducing hepatic steatosis and inflammation.
Like CCl4, the MCD diet is popular for its easiness and rapid induction time. Robust elevations in serum ALT/AST occur within two weeks, and clinical NAFLD activity scores reach near maximal values in the same time span. Fibrosis also occurs rapidly, becoming apparent in as little as 4 weeks and severe within 10 weeks. Fibrosis is reversible after a time course not exceeding 16 weeks, therefore the model should be run for the extended timeframe for evaluating agents to ameliorate NASH. However, animal selection is more restricted than for a CCl4 model. C57Bl/6 mice are the susceptible mouse strain of choice; rats will develop robust steatosis but only minor inflammation and no fibrosis.
There are numerous translational and practical disadvantages of the MCD model. Most saliently, animals experience rapid and severe weight loss—exceeding 40% from diet onset—as well as decreased serum glucose, triglycerides, and cholesterol. This degree of weight loss presents welfare implications and poses animal handling and dosing challenges for the operator, while the metabolic phenotypes are directly opposite to the human condition. The model’s transcriptomic alterations have also been criticized as poorly representative of human NAFLD.
Second Generation Choline Deficient Diets
Second generation choline deficient diets have also proved popular for inducing NAFLD/NASH in rodents without the severe weight loss of the MCD diet. The choline deficient, L amino acid defined (CDAA) diet replaces protein content with a precise equivalence of L amino acids. While not represented in the acronym, the CDAA diet is also reduced (but not depleted) for methionine. C57Bl/6 and BALB/c mice on CDAA develop steatosis and inflammation within 3 weeks. Fibrosis onset is at 6 weeks and fibrosis becomes mild-moderate around week 21, and this is also reversible.
This diet is also suitable for inducing NAFLD/NASH in Wistar and Fisher rats, with similar time frames to disease onset and progression as in mice. Although weight loss is not a drawback of this model in either rodent, human metabolic phenotypes that precede NAFLD (insulin resistance, hyperglycemia, lipidemia, etc.) are absent in this model.
Enrichment of the caloric fat content of CDAA from 10-20% to 60% (choline-deficient, L amino acid defined, high fat diet; CDAHFD) accelerate and exacerbate NASH, and significantly diminish spontaneous reversibility of fibrosis. This modification is well suited for evaluating fibrolytic agents in NASH therapy. This model also advances to cirrhosis, portal hypertension, and liver failure within 24 weeks in the C57Bl/6 mouse so is quite versatile for liver disease research projects. Nevertheless, hallmarks of metabolic syndrome, including obesity, hyperglycemia, and hypertriglyceridemia, are absent in this iteration of the choline deficient diet.
A final variation on the CDAHFD that produces even more severe liver injury and hepatic fibrosis without severe weight loss is a new cholesterol added, choline deficient fibrosis (CCDF) diet. Free cholesterol has toxic properties that stimulate inflammatory responses, enhancing hepatocyte death and fibrosis. CCDF incorporates clinically relevant amounts of cholesterol (approximately 1% by weight). Data in both Wistar rats and C57BL/6 mice demonstrate that the CCDF diet provides a valuable platform for inducing hepatic steatosis and fibrosis in as little as 6-9 weeks, and produces one of few rat models of NASH/fibrosis on high fat diet.
Limitations of Conventional Rodent Models
One major caveat for these models, and a challenge for preclinical development of NASH drugs in general, is their translatability. An earlier blog post delved into diets which integrate the poor eating habits of Western populations, fed to animals with susceptibility or polygenic predisposition for metabolic diseases. This current generation of models are best-in-class with regards to recapitulating human disease, but require long induction times for endpoints, and present high inter-animal variability.
The classic CCl4 model of hepatotoxicity, and multiple iterations of choline deficiency-induced steatosis, provide a well characterized and broadly available suite of rodent models for NASH and liver fibrosis. When choosing carbon tetrachloride induction or choline deficient diet models, along with their advantages you must consider their numerous limitations regarding translatability for the human condition. All of the models summarized here do not present hallmarks of metabolic syndrome that would typically precede human fatty liver disease.
Within appropriate scientific contexts, however, they allow for rapid and relevant preclinical characterization of NASH therapeutics, and may be justifiable alternatives to using slower but more translatable best-in-class models involving Westernized diets and rodent strains with polygenic predispositions for liver disease.