Development of the Rabbit NASH Model Resembling Human NASH and Atherosclerosis

Non-alcoholic steatohepatitis (NASH) is a chronic liver disease which may progress into liver fibrosis and cancer. Since NASH patients have a high prevalence of atherosclerosis and ensuing cardiovascular diseases, simultaneous management of NASH and atherosclerosis is required. Currently, rodents are the most common animal models for NASH and accompanying liver fibrosis, but there are great differences in lipoprotein profiles between rodents and humans, which makes it difficult to reproduce the pathology of NASH patients with atherosclerosis. Rabbits can be a promising candidate for assessing NASH and atherosclerosis because lipoprotein metabolism is more similar to humans compared with rodents. To develop the NASH model using rabbits, we treated the Japanese White rabbit with a newly developed high-fat high-cholesterol diet (HFHCD) containing palm oil 7.5%, cholesterol 0.5%, and ferrous citrate 0.5% for 16 weeks. HFHCD-fed rabbits exhibited NASH at 8 weeks after commencing the treatment and developed advanced fibrosis by the 14th week of treatment. In addition to hypercholesterolemia, atherosclerotic lesion developed in the aorta after 8 weeks. Therefore, this rabbit NASH model might contribute to exploring the concurrent treatment options for human NASH and atherosclerosis.

The liver is a central organ for the systemic lipid metabolism, and lipoproteins play a key role in the transport of lipids between the liver and peripheral tissues. Very lowdensity lipoprotein (VLDL) is involved in the efflux of fatty acids from the liver to the 7.5%, cholesterol 0.5%, and ferrous citrate 0.5%) with free access to sterilized water for up to 16 weeks. The composition of LRC4 is listed in Supplementary Table S1. A timecourse study was also conducted under the same conditions in the control group (12 weeks, n = 3), and in the 4-week (n = 3), 8-week (n = 4), and 14-week (n = 4) groups. Routine blood sampling was performed during the trial period. After each test period, rabbits were anesthetized and sacrificed for blood and tissue sampling. The blood samples were collected from the cardiac cavity and maintained at −80 • C until the assay. Each organ was harvested, weighed, and the parts of the organs were rapidly frozen with liquid nitrogen for subsequent assays. The liver and aorta were fixed with 10% formalin neutral buffer solution (FUJIFILM Wako Pure Chemical Corporation, Osaka, Japan) for histological analysis. All animal experiments were conducted in accordance with the National Academy of Sciences. The animal study protocol [#020042 (15 October 2020)] was approved by the Shinshu University School of Medicine "Guide to the Care and Use of Experimental Animals".

Histological Analysis
Formalin-fixed liver tissues and aorta were embedded in paraffin, cut into 3-µm thick sections, and stained with hematoxylin and eosin (HE) or Azan-Mallory method.

Statistical Analysis
Data are expressed as the mean ± SEM. Statistical analysis was performed with two-tailed unpaired Student's t test. A p value of less than 0.05 was considered statistically significant.

HFHCD Induces NASH with Advanced Liver Fibrosis and Severe Atherosclerosis in Rabbits for 16 Weeks
We generated HFHCD containing palm oil 7.5%, cholesterol 0.5%, and ferrous citrate 0.5% and administered it to Japanese White rabbits for 16 weeks. Their livers were markedly enlarged and looked yellowish ( Figure 1A). Histological findings of the liver revealed macrovesicular and microvesicular steatosis with massive pericellular fibrosis. Additionally, atherosclerotic plaques emerged in the aortic wall ( Figure 1B). Serum AST, ALT, NEFA, TG, T-Chol and TBA levels were significantly increased in HFHCD-fed rabbits ( Figure 1C). Hepatic contents of T-Chol and TBA were much higher in HFHCD rabbits compared to normal rabbits, and F-Chol also tended to increase; however, other lipid species such as NEFA, TG, and PL did not increase, presumably due to the progression of severe fibrosis in the liver. These results indicate that Japanese White rabbits fed HFHCD for 16 weeks exhibited pathologies resembling human NASH with advanced fibrosis and atherosclerosis.

Time-Course Study of HFHCD-fed Rabbits
We conducted the time-course study to elucidate the progression of pathology in HFHCD-fed rabbits. The liver-to-body weight ratio was gradually increased. The spleento-body weight also increased in HFHCD-fed rabbits, suggesting the advancement of liver fibrosis ( Figure 2A). Hepatic histology revealed microvesicular steatosis around the central vein for a 4-week HFHCD treatment. In addition to microvesicular steatosis, hepatocyte ballooning and inflammatory cell infiltration were observed in the 8-week HFHCD group. Hepatocyte ballooning and infiltration of inflammatory cells exacerbated in the 14-week HFHCD group. HFHCD feeding caused mild perivenular fibrosis within 8 weeks, and extensive perivenular/pericellular fibrosis at the time of 14 weeks. Atherosclerotic plaques emerged in the aortic wall in the 8-week HFHCD group and extended to the entire circumference with the marked accumulation of foam cells in the 14-week HFHCD group ( Figure 2B). The remarkable ballooned hepatocytes appeared in 14-week HFHCD rabbit livers, and the crystal structures were also observed in the hepatocytes ( Figure 2C). s 2023, 11, 384 4 of 19 exposed to horseradish peroxidase-conjugated secondary antibody (115-035-003, Jackson ImmunoResearch Laboratories, West Grove, PA) for 1 hour at room temperature. Specific bands were detected using ECL Select Western Blotting Detection Reagent (Cytiva, Tokyo, Japan) on ChemiDoc Touch Imaging System (Bio-Rad Laboratories, Inc. Hercules, CA). Band intensities were quantified using the manufacture's software (Image Lab 6.0.1, Bio-Rad Laboratories, Inc. Hercules, CA), and normalized to GAPDH [37][38][39][40].

Statistical Analysis
Data are expressed as the mean ± SEM. Statistical analysis was performed with twotailed unpaired Student's t test. A p value of less than 0.05 was considered statistically significant.

HFHCD Induces NASH with Advanced Liver Fibrosis and Severe Atherosclerosis in Rabbits for 16 Weeks.
We generated HFHCD containing palm oil 7.5%, cholesterol 0.5%, and ferrous citrate 0.5% and administered it to Japanese White rabbits for 16 weeks. Their livers were markedly enlarged and looked yellowish ( Figure 1A). Histological findings of the liver revealed macrovesicular and microvesicular steatosis with massive pericellular fibrosis. Additionally, atherosclerotic plaques emerged in the aortic wall ( Figure 1B). Serum AST, ALT, NEFA, TG, T-Chol and TBA levels were significantly increased in HFHCD-fed rabbits ( Figure 1C). Hepatic contents of T-Chol and TBA were much higher in HFHCD rabbits compared to normal rabbits, and F-Chol also tended to increase; however, other lipid species such as NEFA, TG, and PL did not increase, presumably due to the progression of severe fibrosis in the liver. These results indicate that Japanese White rabbits fed HFHCD for 16 weeks exhibited pathologies resembling human NASH with advanced fibrosis and atherosclerosis.

Time-Course Study of HFHCD-fed Rabbits
We conducted the time-course study to elucidate the progression of pathology in HFHCD-fed rabbits. The liver-to-body weight ratio was gradually increased. The spleento-body weight also increased in HFHCD-fed rabbits, suggesting the advancement of liver fibrosis ( Figure 2A). Hepatic histology revealed microvesicular steatosis around the central vein for a 4-week HFHCD treatment. In addition to microvesicular steatosis, hepatocyte ballooning and inflammatory cell infiltration were observed in the 8-week HFHCD group. Hepatocyte ballooning and infiltration of inflammatory cells exacerbated in the 14week HFHCD group. HFHCD feeding caused mild perivenular fibrosis within 8 weeks, and extensive perivenular/pericellular fibrosis at the time of 14 weeks. Atherosclerotic plaques emerged in the aortic wall in the 8-week HFHCD group and extended to the entire circumference with the marked accumulation of foam cells in the 14-week HFHCD group ( Figure 2B). The remarkable ballooned hepatocytes appeared in 14-week HFHCD rabbit livers, and the crystal structures were also observed in the hepatocytes ( Figure 2C).
Serum AST and ALT levels in HFHCD group were 2-3-fold higher than in the control group at each time point. Serum lipid concentrations increased in the 4-week HFHCD group, especially T-Chol and TBA, in a time-dependent manner ( Figure 2D). Hepatic lipid contents, including TBA, were markedly elevated in the 4-week HFHCD treatment. Although T-Chol, F-Chol, and TBA sustained higher levels at 14 weeks, NEFA, TG, and PL slightly decreased as time passed, resembling the pathological course of human NAFLD, called burned-out NASH ( Figure 2E) [23,41]. These data indicate that HFHCD induced hepatic steatosis and hyperlipidemia within 4 weeks, emergence of hepatic fibrosis, and atherosclerotic plaque at 8 weeks, and the progression of advanced fibrosis and extensive atherosclerotic plaques at 14 weeks of HFHCD administration. Serum AST and ALT levels in HFHCD group were 2-3-fold higher than in the control group at each time point. Serum lipid concentrations increased in the 4-week HFHCD group, especially T-Chol and TBA, in a time-dependent manner ( Figure 2D). Hepatic lipid contents, including TBA, were markedly elevated in the 4-week HFHCD treatment.
Although T-Chol, F-Chol, and TBA sustained higher levels at 14 weeks, NEFA, TG, and PL slightly decreased as time passed, resembling the pathological course of human NAFLD, called burned-out NASH ( Figure 2E) [23,41]. These data indicate that HFHCD induced hepatic steatosis and hyperlipidemia within 4 weeks, emergence of hepatic fibrosis, and atherosclerotic plaque at 8 weeks, and the progression of advanced fibrosis and extensive atherosclerotic plaques at 14 weeks of HFHCD administration.

HFHCD Alters Gene Expression Associated with Cholesterol Metabolism
A range of genes related to the lipid metabolism has changed in the liver of NAFLD/ NASH patients [42,43]. To understand the mechanism of cholesterol accumulation in the liver during HFHCD feeding, we investigated the expression of several genes involved in the cholesterol metabolism. When excess cholesterol is present in the liver, sterol regulatory element-binding protein-2 (SREBP-2) is inactivated and its target gene-encoding HMG-CoA reductase (HMGCR), a rate-limiting enzyme of cholesterol synthesis, and LDLR are down-regulated. In the 4-week HFHCD treatment, HMGCR mRNA levels were decreased significantly, but returned to a similar level to the control group at 8 and 14 weeks. Other cholesterol synthesis-related gene encoding farnesyl-diphosphate farnesyltransferase 1 (FDFT1) was up-regulated only in the 4-week HFHCD group and opposed to squalene monooxygenase (SQLE). No significant changes were observed at the mRNA levels in 8 and 14 weeks of HFHCD group ( Figure 3A). The gene expression associated with cholesterol uptake, the LDLR mRNA levels, were significantly decreased, whereas CD36 and VLDL receptor (VLDLR) mRNA levels were increased by HFHCD feeding ( Figure 3B). In genes related to hepatic cholesterol excretion, ATP-binding cassette subfamily A member 1 (ABCA1) mRNA levels were considerably increased by HFHCD in a time-dependent manner, while ATP-binding cassette subfamily G member 8 (ABCG8) mRNAs were increased slightly ( Figure 3C). Other genes encoding apoB (APOB), microsomal triglyceride transfer protein (MTTP), and ATP-binding cassette subfamily G member 5 (ABCG5) were unchanged between HFHCD and control group. Although LDLR mRNA levels were decreased, a significant up-regulation of CD36/VLDLR mRNA expression is one of the causes of excessive cholesterol accumulation in the liver through the enhancement of the uptake of cholesterol from the circulation.  mRNAs were increased slightly ( Figure 3C). Other genes encoding apoB (APOB) somal triglyceride transfer protein (MTTP), and ATP-binding cassette subfamily ber 5 (ABCG5) were unchanged between HFHCD and control group. Althoug mRNA levels were decreased, a significant up-regulation of CD36/VLDLR mRNA sion is one of the causes of excessive cholesterol accumulation in the liver thro enhancement of the uptake of cholesterol from the circulation.

HFHCD Disrupts BA Homeostasis in the Liver.
A series of genes involved in BA synthesis and excretion also contributes to cholesterol metabolism. Cytochrome P450 7A1 (CYP7A1) is a rate-limiting enzyme of BA synthesis, while bile salt export pump (BSEP, encoded by ABCB11) is a major pump excreting BA from hepatocytes, and both genes are negatively and positively regulated by farnesoid X receptor (FXR). The gene expression of CYP7A1 and ABCB11 was not altered in the livers of HFHCD ( Figure 3E). Similar to these data, the expression of other genes regulated by FXR, such as cytochrome P450 family 8 subfamily B member 1 (CYP8B1), cytochrome P450 family 27 subfamily A member 1 (CYP27A1), and multidrug resistance-associated protein 2 (MRP2), was largely unchanged. The mRNA levels of solute carrier family 10 member 1 (NTCP) were down-regulated presumably due to enhanced hepatic inflammation and increased expression of tumor necrosis factor α (TNFα) and interleukin 1β (IL1β) [44]. Since multiple drug resistance 1 (MDR1) acts as a cytoprotective function through

HFHCD Disrupts BA Homeostasis in the Liver
A series of genes involved in BA synthesis and excretion also contributes to cholesterol metabolism. Cytochrome P450 7A1 (CYP7A1) is a rate-limiting enzyme of BA synthesis, while bile salt export pump (BSEP, encoded by ABCB11) is a major pump excreting BA from hepatocytes, and both genes are negatively and positively regulated by farnesoid X receptor (FXR). The gene expression of CYP7A1 and ABCB11 was not altered in the livers of HFHCD ( Figure 3E). Similar to these data, the expression of other genes regulated by FXR, such as cytochrome P450 family 8 subfamily B member 1 (CYP8B1), cytochrome P450 family 27 subfamily A member 1 (CYP27A1), and multidrug resistance-associated protein 2 (MRP2), was largely unchanged. The mRNA levels of solute carrier family 10 member 1 (NTCP) were down-regulated presumably due to enhanced hepatic inflammation and increased expression of tumor necrosis factor α (TNFα) and interleukin 1β (IL1-β) [44]. Since multiple drug resistance 1 (MDR1) acts as a cytoprotective function through the excretion of toxic compounds, the induction of MDR1 seemed to be a compensation to protect liver function. These data indicated that the regulation of FXR-based BA homeostasis was collapsed, and the accumulation of excess BA in the liver and the rise of serum BA levels occurred.
Cholesterol ester is converted to F-Chol via neutral cholesterol ester hydrolase 1 (NCEH1), and the inverse reaction is performed by acetyl-CoA acetyltransferase 2 (ACAT2). Elevated levels of F-Chol in the HFHCD livers may involve increased NCEH1 expression with unchanged ACAT2 expression ( Figure 3D).
Overall, excessive accumulation of hepatic T-Chol and F-Chol is considered to be the result of an increased level of cholesterol uptake and dysregulation of BA homeostasis, the unsuitable expression of these cholesterol metabolic genes.

HFHCD Does Not Affect the Expression of Genes Related to Fatty Acid Metabolism
Acetyl-CoA carboxylase encoded by ACACA gene is a rate-limiting enzyme of longchain fatty acid synthesis. The mRNA levels of ACACA were upregulated in the livers of the HFHCD group. In contrast, the expression of diacylglycerol acyltransferase 2 (DGAT2) involved in the final stage of TG synthesis was decreased ( Figure 4A). Concerning fatty acid uptake, fatty acid-binding protein 1 (FABP1) mRNA tended to decrease, while hepatic triglyceride lipase (HTGL) increased in HFHCD group. Long-chain fatty acid CoA ligase 1 (ASCL1) mRNA was elevated only in the 4-week HFHCD group ( Figure 4B). No substantial changes in the expression of genes related to fatty acid oxidation were observed in the HFHCD livers ( Figure 4C). acid uptake, fatty acid-binding protein 1 (FABP1) mRNA tended to decrease, while hepatic triglyceride lipase (HTGL) increased in HFHCD group. Long-chain fatty acid CoA ligase 1 (ASCL1) mRNA was elevated only in the 4-week HFHCD group ( Figure 4B). No substantial changes in the expression of genes related to fatty acid oxidation were observed in the HFHCD livers ( Figure 4C).

HFHCD Up-Regulates Gene Expression Related to Inflammation and Fibrosis
Since HFHCD caused NASH with fibrosis, we investigated the expression of the genes associated with inflammation and ensuing fibrogenesis. The mRNA levels of macrophage marker CD68 and the inflammatory cytokines, such as TNFα, C-C motif chemo-

HFHCD Up-Regulates Gene Expression Related to Inflammation and Fibrosis
Since HFHCD caused NASH with fibrosis, we investigated the expression of the genes associated with inflammation and ensuing fibrogenesis. The mRNA levels of macrophage marker CD68 and the inflammatory cytokines, such as TNFα, C-C motif chemokine ligand 2 (CCL2), IL1-β, and interleukin 18 (IL18), were significantly increased in the HFHCD livers ( Figure 5A). In addition to transforming growth factor β1 (TGFB1), an inducer of fibrosis, the gene expression of the fibrogenic factors, such as actin alpha 2 (ACTA2), collagen type I alpha 1 chain (COL1A1), and collagen type III alpha 1 chain (COL3A1), was elevated in the HFHCD livers ( Figure 5D). The mRNA levels of toll-like receptors (TLRs) and their adaptor (TLR2, TLR4, and CD14) and inflammasome components, such as NLR family pyrin domain containing 3 (NLRP3) and caspase 1 (CASP1), were also induced ( Figure 5B,C). F-Chol causes ER stress and the resulting oxidative stress, stimulating TLRs and inducing inflammation. Cholesterol crystals activate the NLRP3 inflammasome and contribute to the inflammatory signaling. ER stress-related genes, such as PRKR-like endoplasmic reticulum kinase (PERK) and DDIT3 and ER stress-induced lipoapoptosis, such as Bcl-2-like protein 11 (BIM) and p53 upregulated modulator of apoptosis (PUMA), were increased ( Figure 6A). Furthermore, oxidative stress-producing genes (cytochrome b-245 F-Chol causes ER stress and the resulting oxidative stress, stimulating TLRs and inducing inflammation. Cholesterol crystals activate the NLRP3 inflammasome and contribute to the inflammatory signaling. ER stress-related genes, such as PRKR-like endoplasmic reticulum kinase (PERK) and DDIT3 and ER stress-induced lipoapoptosis, such as Bcl-2-like protein 11 (BIM) and p53 upregulated modulator of apoptosis (PUMA), were increased ( Figure 6A). Furthermore, oxidative stress-producing genes (cytochrome b-245 beta chain, CYBB; and neutrophil cytosolic factor 1, NCF1) were increased in the HFHCD group ( Figure 6B), but oxidative stress-eliminating genes, NAD(P)H quinone dehydrogenase 1 (NQO1), heme oxygenase 1 (HMOX1), and superoxide dismutase 1 (SOD1) were upregulated as well, and the amount of the 4-hydroxy-nonenal (4-HNE), a marker of the lipid peroxide, was not increased ( Figure 6C). These data demonstrate that cholesterol accumulation in the liver and ensuing ER stress, inflammasome activation and enhanced TLR signaling triggers hepatic inflammation and fibrosis progression in this rabbit NASH model.

Discussion
We developed the original HFHCD containing palm oil 7.5%, cholesterol 0.5%, and ferrous citrate 0.5% and administered it to Japanese White rabbits, which have close sim-

Discussion
We developed the original HFHCD containing palm oil 7.5%, cholesterol 0.5%, and ferrous citrate 0.5% and administered it to Japanese White rabbits, which have close similarities to lipoprotein and BA metabolism in humans. We found the progression of NASH with advanced fibrosis and atherosclerosis at 14-weeks of HFHCD treatment. Investigation of hepatic lipid profiles and gene expression revealed the significance of hepatic cholesterol/BA accumulation and ER stress to the pathogenesis of NASH in this model. Our model may be useful to evaluate the pathogenesis of NAFLD/atherosclerosis and seek concurrent treatment for both diseases.
Several animal models have been used to evaluate the pathogenesis of NAFLD/NASH. Mice and rats are the most common due to their easy handling and genetic manipulation, and short life cycles. However, there are some great differences between these animals and humans, and differences in lipid metabolism seem to be one of the most crucial factors as regards considering translatability of subclinical studies to clinical trials. Indeed, mice and rats have high circulating HDL and low LDL, which is opposite to humans. Rodents have both apoB-48 and -100 in VLDL particles, and humans only have apoB-100. VLDL particles including apoB-48 have a more rapid fractional catabolic rate than VLDL particles that only have apoB-100. Additionally, BA composition, a key determinant for cholesterol homeostasis, is quite different between rodents and humans as well. Cholic acid (CA) and chenodeoxycholic acid (CDCA) are synthesized as primary BA in humans, while mice and rats have CYP2C70, which converts CDCA to muricholic acid (MCA); therefore, approximately half of the BA pool of mice is MCAs. Although CDCA has the highest affinity for FXR, MCA is antagonistic to FXR. Therefore, the BA profile may affect the high synthetic rate of BA in mice and the resulting rapid cholesterol metabolism. The RCT system is also different because rodents have no cholesterol ester transfer protein (CETP) activity in plasma, which contributes to the high HDL and low LDL levels. It is known that statin's effect is not shown on mice, possibly due to the high synthetic capacity of cholesterol [28]. According to the abovementioned reasons, choosing animals whose lipid metabolism resembles that of humans as closely as possible is desired. Rabbits have more similarities in lipoprotein metabolism with that of humans compared with rodents. Plasma LDL levels are high and HDL levels are low due to CETP activity [28], and the BA composition is predominantly hydrophobic and FXR agonistic [30].
Our HFHCD-fed rabbits represented typical pathological features of NASH, such as hepatic steatosis, pericellular fibrosis, and hepatocyte ballooning, and clinical features, including hypertransaminasemia, cholesterol-dominant hyperlipidemia, and accompanying atherosclerosis. It is reported that macrophages in atherogenic lesions in humans and rabbits have VLDLR but do not have it in mice [45,46]. Although our model did not show obesity and hyperglycemia, key contributors to NAFLD/NASH, it was documented that non-obese NAFLD men were prone to coronary artery calcification [47]. Therefore, this rabbit model seems to be a suitable animal to evaluate human NAFLD/NASH, especially with hypercholesterolemia and atherosclerosis, and assess the efficacy of therapeutic interventions on NASH/atherosclerosis. Cholesterol, especially F-Chol, plays a critical role in the development of NAFLD and progression to NASH [48][49][50]. The imbalance of F-Chol content in the ER membrane causes ER stress, and chronic ER stress results in inflammation and cell death. When an excessive amount of F-Chol accumulates in the cells, a cholesterol crystal is formed. Crystalized cholesterol is detected in the hepatocytes of NASH patients. Crystalized cholesterol activates NLRP3 inflammasome, inducing hepatocyte death [49,51]. Subsequently, Kupffer cells are activated and release the proinflammatory cytokines, and hepatic stellate cells are activated as well, leading to hepatic inflammation and fibrosis.
In this HFHCD-fed rabbit model, the hepatic expression levels of some genes involved in cholesterol metabolism were markedly changed. The mRNA levels of CD36 and VLDLR were increased via HFHCD. These genes are normally expressing at very low levels in the hepatocytes, but CD36 expression increases in NASH patients and VLDLR increases in some pathological conditions, such as obesity, diabetes, and NAFLD [52]. Changes in the expression of these genes could contribute to the elevated absorption of cholesterol into the liver.
Greater accumulation of F-Chol in the liver prompted us to speculate about the alteration of the cholesterol-esterizing pathway in this HFHCD-fed rabbit model. F-Chol is generated from cholesterol ester via hydrolysis with NCEH1, while ACAT2 plays the reverse function of esterifying F-Chol. The balance of intracellular free and esterified cholesterol is preserved by both genes. In HFHCD livers, NCEH1 mRNA significantly increased, but ACAT2 was unchanged. Since hepatic NCEH1 is reportedly up-regulated in NAFLD patients [53], this model is considered to be more appropriate for investigating the contribution of F-Chol to the pathogenesis of NASH.
Excess amount of F-Chol is excreted in the bile by ABCG5 and ABCG8; however, in the HFHCD-fed livers, the gene expression of both genes was not significantly changed. ABCA1 is involved in the production of HDL and thus contributes to the extracellular excretion of cholesterol. ABCA1 is one of the target genes of the liver X receptor (LXR) the ligand of which is oxysterol, oxidized derivatives from cholesterol. Excessive levels of F-Chol are present in the liver of HFHCD-fed rabbits, and significant amounts of oxysterol are expected. Activated LXR induces ABCA1 gene expression and promotes cholesterol efflux from the hepatocyte. This effect is regarded as the protective effect from the F-Chol toxicity.
The strong points of this model are similarities of disease phenotypes of dyslipidemic NAFLD/NASH patients with accompanying atherosclerosis, and relatively speedy progression of liver fibrosis and atherogenic plaques through singular HFHCD for 8-14 weeks. Kim et al. documented that the treatment of New Zealand white rabbits with a high-cholesterol diet (1%) for 3 months led to the rabbits exhibiting mild fatty changes in the liver with sporadic fibrosis [54]. Such a difference might stem from differences in diet composition and/or the genetic background of rabbits. Considering the speedy progression of NASH fibrosis/atherosclerosis and similarities in systemic cholesterol/lipoprotein metabolism and VLDLR expression in macrophages of atherogenic plaques to humans, utilizing this rabbit model would save time to assess NASH fibrosis/atherosclerosis-retarding effects of candidate agents, such as cholesterol-lowering statins and ezetimibe.
Recently, the association between hepatic zonation and NAFLD/NASH development has been gathering attention [55]. Differences in hepatic zone-dependent enzyme expression and nutrient metabolism between humans and rodents might be associated with the fact that mouse models of NAFLD/NASH are unable to replicate the human disease [24,56]. To date, rabbit liver zonation has not been fully investigated. If hepatic zonation is more similar to humans compared with that of rodents, this finding would corroborate the relevance of using rabbit models in NAFLD/NASH research.

Conclusions
We demonstrated a new rabbit NASH model using Japanese White rabbits and a newly developed HFHCD. Although further studies are needed to validate the usefulness of this model, it is expected that this rabbit NASH model may contribute to the development of favorable treatments for NASH, liver fibrosis, and atherosclerosis.