Characterization and Pharmacological Validation of a Preclinical Model of NASH in Göttingen Minipigs

Abstract

Background: Nonalcoholic fatty liver disease (NAFLD) is the leading cause of chronic liver disease, which is associated with features of metabolic syndrome. NAFLD may progress in a subset of patients into nonalcoholic steatohepatitis (NASH) with liver injury resulting ultimately in cirrhosis and potentially hepatocellular carcinoma. Today, there is no approved treatment for NASH due to, at least in part, the lack of preclinical models recapitulating features of human disease. Here, we report the development of a dietary model of NASH in the Göttingen minipig.

Methods: First, we performed a longitudinal characterization of diet-induced NASH and fibrosis using biochemical, histological, and transcriptional analyses. We then evaluated the pharmacological response to Obeticholic acid (OCA) treatment for 8 weeks at 2.5mg/kg/d, a dose matching its active clinical exposure.

Results: Serial histological examinations revealed a rapid installation of NASH driven by massive steatosis and inflammation, including evidence of ballooning. Furthermore, we found the progressive development of both perisinusoidal and portal fibrosis reaching fibrotic septa after 6 months of diet. Histological changes were mechanistically supported by well-defined gene signatures identified by RNA Seq analysis. While treatment with OCA was well tolerated throughout the study, it did not improve liver dysfunction nor NASH progression. By contrast, OCA treatment resulted in a significant reduction in diet-induced fibrosis in this model.

Conclusions: These results, taken together, indicate that the diet-induced NASH in the Göttingen minipig recapitulates most of the features of human NASH and may be a model with improved translational value to prioritize drug candidates toward clinical development.

Keywords: CDAHFD, choline-deficient amino acid-defined high fat diet; FDR, false discovery rate; FFC, fatfructose cholesterol diet; NAFLD, nonalcoholic fatty liver disease; NAS, NAFLD activity score; NASH; NASH, nonalcoholic steatohepatitis; PNPLA3, patatin-like phospholipase domain-containing 3; minipig; translational value.

Figure 1

CDAHFD triggers a significant increase in both liver and total body weight. Body weight follow-up (Panel A), liver morphology (Panel B), total liver weight (Panel C), and liver to body weight ratio (Panel D) at the end of the study (chow n = 6 & CDAHFD = 7 per group). Data shown are mean ± SEM. (∗: P < 0.05; ∗∗∗: P < 0.001 CDAAHFD vs. Chow).

Figure 2

Marked hypercholesterolemia and insulin resistance in response to CDAHFD. Kinetic analysis of plasma triglycerides (Panel A), cholesterol (Panel B), glucose (Panel C), and insulin levels (Panel D) in Chow (n = 6; white bars), and CDAHFD (n = 7, blue bars) fed minipigs. Data shown are mean ± SEM. (∗∗∗: P < 0.001 CDAHFD vs. Chow).

Figure 3

Rapid and significant liver dysfunction in response to CDAHFD. Kinetic analysis of plasma liver enzymes (Panel A&B), total and direct bilirubin levels (Panel C&D) in Chow (n = 6; white bars), and CDAHFD (n = 7, blue bars) fed minipigs. Data shown are mean ± SEM. (∗∗: P < 0.01; ∗∗∗: P < 0.001 CDAHFD vs. Chow).

Figure 4

Pronounced steatosis and limited inflammation in response to CDAHFD. Representative images from HE staining (Panel A), histological analyses (Steatosis Panel B; Inflammation Panel C; Ballooning Panel D), and NAS score determination (Panel E) in Chow (n = 6; white bars), and CDAHFD (n = 7, blue bars) fed minipigs. Data shown are mean ± SEM. (∗∗∗: P < 0.001 CDAHFD vs. Chow).

Figure 5

Time-dependent increase in liver fibrosis in response to CDAHFD. Representative pictures of Sirius red stained liver samples (Panel A). Fibrosis score (Panel B) and distribution of animals according to score (Panel C) in Chow (n = 6; white bars), and CDAHFD (n = 7, blue bars) fed minipigs. Data shown are mean ± SEM. (∗∗∗: P < 0.001 CDAHFD vs. Chow).

Figure 6

RNA Seq analysis of liver samples at the end of the study. Heatmap (Panel A), volcano plot (Panel B), and GSEA (Panel C), of differentially expressed genes from CDAHFD compared to Chow (Fold change cutoff >1.5; FDR-adjusted P-value cutoff: 0.05).

Figure 7

OCA treatment had no significant impact on body weight gain nor liver enlargement. Body weight follow-up (Panel A), liver morphology (Panel B), total liver weight (Panel C), and liver to body weight ratio (Panel D) at the end of the study in chow (white bar n = 8); CDAHFD (blue bar n = 8) and CDAHFD + OCA (grey bar n = 6). Data shown are mean ± SEM. (∗∗∗: P < 0.001 CDAHFD vs. Chow; NS: nonsignificant).

Figure 8

No impact of OCA treatment on insulin resistance, dyslipidemia and liver dysfunction in CDAHFD fed minipigs. Plasma glucose (A), insulin (B), fructosamine (C), TG (D), cholesterol (E), ALT (F), AST (G), ALP (H), and γGT (I) levels were measured at the end of the study. Data shown are means ± SEM (chow white bar n = 8; CDAHFD blue bar n = 8, CDAHFD + OCA grey bar n = 6). ∗: P < 0.05, ∗∗: P < 0.01, ∗∗∗: P < 0.001 CDAHFD vs. Chow.

Figure 9

Significant impact of OCA on fibrosis but not NASH in minipigs fed with CDAHFD. Liver TG (Panel A) and cholesterol levels (Panel B) were determined at the end of the study. Representative pictures of HE and/or Sirius red stained liver samples at the end of the study (Panel C). NAS (Panel D) and Fibrosis (Panel E) scores. Data shown are mean ± SEM. (chow white bar n = 8; CDAHFD blue bar n = 8, CDAHFD + OCA grey bar n = 6). ∗∗∗: P < 0.001 CDAHFD vs. Chow; #: P < 0.05 CDAHFD + OCA vs. CDAHFD.