Animals
Male MS-NASH mice (formally FATZO) [18] were developed by Crown Bioscience as a new generation of mouse model presenting high translatable phenotypes in human diseases such as obesity, metabolic disorder, diabetes and NAFLD/NASH [19, 20, 22]. The animals for this study were bred and then housed individually in IVC cages (Taicang, China) or open ventilated cages (Indianapolis, IN) with room temperature maintained at 22-26 °C, a 12-hour light cycle (06:00–18:00), and distilled water ad libitum. The animals were fed control diet (CD, Purina 5008 chow, LabDiet, St. Louis, MO) for 8 weeks after birth, then stratified into different experimental groups based on body weight, serum ALT and AST. C57Bl/6J mice (The Jackson Laboratory, Ellsworth, Maine) were used as control strain. All mice were maintained and treated in accordance with the guidelines of Association for Assessment and Accreditation of Laboratory Animal Care (AAALAC). Experimental protocols were approved by the Institutional Animal Care and Use Committee (IACUC).
Effects of CCl4 in MS-NASH mice fed Western diet supplemented with fructose (WDF)
The first study aimed to, 1) confirm the characterization of MS-NASH mice fed WDF (40% kCal fat, 43% kCal carbohydrate, 17% kCal protein, D12079B, Research Diets, New Brunswick, NJ) to induce liver phenotypes; and 2) examine the dose effect of CCl4 to shorten the induction time and to enhance liver fibrosis.
The original CCl4 solution was further diluted in olive oil (Sigma Aldrich) at final concentrations of 0.02 and 0.005 mL/mL, with a dosing volume of 10 and 1.5 mL/kg, injected intraperitoneally (IP) twice weekly (BIW) at a final dose of 0.2 and 0.08 mL/kg for high and low dose group, respectively.
High dose CCl4 (0.2 mL/kg, BIW) for 3 weeks: After 8 weeks on CD, MS-NASH mice were divided into following 3 groups: 1) CD (n=8): continued on CD for 11 weeks; 2) WDF (n=8); and 3) WDF + CCl4 (n=6): switched to WDF for the rest of 11 weeks to induce liver phenotypes; after 8 weeks on WDF, vehicle or CCl4 was injected IP, BIW for 3 weeks.
Low dose CCl4 (0.08 mL/kg, BIW) for 8 weeks After 8 weeks on CD, MS-NASH mice were switched to WDF for 16 weeks to induce liver phenotypes. At 8 weeks after WDF, the animals were divided into 2 groups: 1) WDF (n = 4); and 2) WDF + CCl4 (n=11), with IP injection of vehicle or low dose CCl4, BIW, respectively for the final 8 weeks.
Effects of Obeticholic acid (OCA) in mice fed WDF plus CCl4
After 8 weeks on CD, MS-NASH or C57BI/6 mice were fed WDF for 16 weeks to induce liver phenotypes. After 8 weeks on WDF, the animals were injected IP with low dose CCl4 (0.08 mL/kg, BIW) and divided into vehicle (n = 11) and OCA (n = 10) groups for an additional 8 weeks during which, vehicle (1% methylcellulose, Sigma Aldrich) or OCA (Toronto Research Chemicals, New York, ON, Canada, 30 mg/kg) was administrated orally once daily. C57Bl/6 mice were compared with the same protocol in vehicle (n=9) or OCA (n=9) groups.
Sample collection, processing and measurements
In all studies, body weights were recorded every 4 weeks. At the end of studies, all mice were euthanized by CO2 inhalation and with cervical dislocation approximately 24 hours after the last CCl4 administration.
Blood samples Blood samples were collected ~72 hours after CCl4 dosing during the course of the experiment from the tail or ~24 hours after the last dose of CCl4 at the end of the experiment via cardiac exsanguination, from which, serum was prepared for measuring AST and ALT by a clinical analyzer (Beckman-Coulter AU480; Brea, CA). A separate experiment was performed to measure the acute time course of ALT and AST at 24, 48 and 72 hours in response to a single dose of CCl4 at 0.2 mL/kg in MS-NASH mice on CD.
Liver contents The right lobe of the liver (~ 200 mg/animal) was collected and snap frozen in liquid nitrogen, placed in Lysing Matrix D Tubes with distilled water at 20% concentration (MP Biomedicals, Santa Anna, CA), and homogenized in a Fastprep-FP120 cell disrupter (Thermo Fisher Savant) in cold condition for 30 seconds. The liver contents of triglyceride and cholesterol were analyzed by a clinical analyzer (Beckman-Coulter AU480) within 30 min of sample preparation. Hydroxyproline was measured on the BioTek Synergy 2 Multi-Mode Microplate Reader utilizing a colorimetric assay kit (BioVision, Catalog #: K555-100) after hydrolyzed for 3 hours at 120oC at 1:1 ratio of sample homogenate to 12N Hydrochloric acid (RICCA Chemical Co., Arlington, TX).
Liver histology The left lobe of the liver was fixed in 10% neutral buffered formalin for 24 hours followed by bath in alcohol then xylene for paraffin embedding, cut into 5-mm sections and stained with Hematoxylin and Eosin (H&E) and Picro Sirius Red (PSR). A whole slide digital imaging system (Aperio Scan Scope CS system, 360 Park Center Drive, Vista, CA) was used to scan the slides at 20x in 1.5 to 2.25 min.
Liver Histopathology analysis
Semi-quantitative scoring by a pathologist Digital images were evaluated by a research pathologist blinded to different study groups with the standard NASH criteria for semi-quantitative scoring for hepatosteatosis, lobular inflammation, and hepatocyte balloon degeneration, respectively from H&E staining and then summated as a standard NAFLD Activity Score (NAS), commonly used in preclinical animal models and in patients [28, 29]. Fibrosis score was assessed systemically with pattern recognition from PSR staining. Three representative areas per liver were examined and the scores of each parameter from individual animal were averaged.
Computerized quantitative analysis Computer software with automatic intelligence (AI) machine learning algorithm for histology analysis from Halo (Indica Labs, Albuquerque, NM) or ImageDx (Reveal Biosciences, San Diego, CA) were used to analyze digitally scanned images of H&E and PSR staining for quantitative analysis of steatosis, ballooning, inflammation or fibrosis in the same set slides evaluated by the pathologist. The analysis process included automated tissue identification, followed by segmentation of regions of interest for quantification of the following metrics: 1) Steatosis percentage: the area of total lipid accumulation subcategorized micro- or macro-vesicular within the entire section area; 2) Ballooning hepatocyte density: the density of ballooning hepatocytes within the entire section area; 3) Inflammatory cell density: the total number of inflammatory cells within the entire section area. All 3 parameters above were analyzed in the H&E stained section. 4) Fibrosis percentage: the total fibrosis area within the entire section area in the PSR stained section.
Statistical analysis
All values are reported as mean ± standard error of mean (SEM), unless noted otherwise. Data were compared in MS-NASH mice on CD or WDF with or without CCl4; and effects of OCA were compared to vehicle with one-Way ANOVA for multiple groups or Holm-Sidak t-test for 2 groups. Survival curves of MS-NASH and C57Bl/6 mice treated with CCl4 were compared using Log-rank test for trend. Parametric correlation tests were conducted between pathologist scores and ImageDx quantitative analysis using Pearson correlation coefficient r. Statistical differences were denoted as two-sided p < 0.05 or p < 0.005. Prism software (GraphPad, version 8.3) was used for the statistical analysis and graphing.