Modelling Nonalcoholic Steatohepatitis In Vivo-A Close Transcriptomic Similarity Supports the Guinea Pig Disease Model.

Josephine Skat-Rørdam,Pernille Tveden-Nyborg,Jens Lykkesfeldt,Markus Latta,Stefan E Seemann,David H Ipsen

doi:10.3390/biomedicines9091198

Josephine Skat-Rørdam, Pernille Tveden-Nyborg + Show 4 more

Open Access

https://doi.org/10.3390/biomedicines9091198

Copy DOI

Journal: Biomedicines	Publication Date: Sep 10, 2021
Citations: 3	License type: CC BY 4.0

Affiliation: University of Copenhagen, Novo Nordisk (Denmark)

Abstract

The successful development of effective treatments against nonalcoholic steatohepatitis (NASH) is significantly set back by the limited availability of predictive preclinical models, thereby delaying and reducing patient recovery. Uniquely, the guinea pig NASH model develops hepatic histopathology and fibrosis resembling that of human patients, supported by similarities in selected cellular pathways. The high-throughput sequencing of guinea pig livers with fibrotic NASH (n = 6) and matched controls (n = 6) showed a clear separation of the transcriptomic profile between NASH and control animals. A comparison to NASH patients with mild disease (GSE126848) revealed a 45.2% overlap in differentially expressed genes, while pathway analysis showed a 34% match between the top 50 enriched pathways in patients with advanced NASH (GSE49541) and guinea pigs. Gene set enrichment analysis highlighted the similarity to human patients (GSE49541), also when compared to three murine models (GSE52748, GSE38141, GSE67680), and leading edge genes THRSP, CCL20 and CD44 were highly expressed in both guinea pigs and NASH patients. Nine candidate genes were identified as highly correlated with hepatic fibrosis (correlation coefficient > 0.8), and showed a similar expression pattern in NASH patients. Of these, two candidate genes (VWF and SERPINB9) encode secreted factors, warranting further investigations as potential biomarkers of human NASH progression. This study demonstrates key similarities in guinea pig and human NASH, supporting increased predictability when translating research findings to human patients.

Highlights

Hepatic fibrosis is the primary prognostic marker of mortality in nonalcoholic steatohepatitis (NASH) globally affecting millions [1]
We propose a selection of candidate genes present in both human and guinea pig fibrotic NASH that may prove valuable in future drug discovery and biomarker development
This paper shows the first comparison of human and guinea pig NASH transcriptomes, and reveals the high translational potential of this model compared to the included murine models

Summary

Introduction

Hepatic fibrosis is the primary prognostic marker of mortality in nonalcoholic steatohepatitis (NASH) globally affecting millions [1]. Many mouse and rat models do not develop NASH with advanced fibrosis (bridging fibrosis and cirrhosis) without the use of hepatotoxins and micronutrient-deficient diets, compromising construct and, by extension, predictive validity [2,3,4]. Guinea pigs and humans share an LDL-dominant lipoprotein profile in contrast to the HDL-dominant profile of rats and mice, and targeted analysis of genes related to hepatic lipid metabolism, inflammation, and fibrogenesis further supports a high degree of similarity between the guinea pig disease model and human NASH [8,9]. Using RNA sequencing and human orthologue mapping to improve annotation, this paper investigates the translational validity of the guinea pig NASH transcriptome and directly compares transcriptome remodelling to profiles from two patient datasets representing either mild or advanced disease, and three frequently used mouse models. We propose a selection of candidate genes present in both human and guinea pig fibrotic NASH that may prove valuable in future drug discovery and biomarker development

Methods

Discussion

Conclusion