A comparative analysis of hepatic pathological phenotypes in C57BL/6J and C57BL/6N mouse strains in non-alcoholic steatohepatitis models
www.nature.com/scientificreports
OPEN
Received: 25 June 2018
Accepted: 26 November 2018
Published: xx xx xxxx
A comparative analysis of
hepatic pathological phenotypes
in C57BL/6J and C57BL/6N
mouse strains in non-alcoholic
steatohepatitis models
Eri Kawashita, Keiichi Ishihara, Madoka Nomoto, Mika Taniguchi & Satoshi Akiba
C57BL/6J (BL6J) and C57BL/6N (BL6N) inbred substrains are most widely used to understand the
pathological roles of target molecules in a variety of diseases, including non-alcoholic steatohepatitis
(NASH), based on transgenic mouse technologies. There are notable differences in the metabolic
phenotypes, including glucose tolerance, between the BL6J and BL6N substrains, but the phenotypic
differences in NASH are still unknown. We performed a comparative analysis of the two mouse
substrains to identify the pathological phenotypic differences in NASH models. In the CCl4-induced
NASH model, the BL6J mice exhibited a more severe degree of oxidative stress and fibrosis in the liver
than the BL6N mice. In contrast, in the high-fat diet-induced NASH model, more accumulation of
hepatic triglycerides but less weight gain and liver injury were noted in the BL6J mice than in the BL6N
mice. Our findings strongly suggest caution be exercised with the use of unmatched mixed genetic
background C57BL6 mice for studies related to NASH, especially when generating conditional knockout
C57BL6 mice.
Non-alcoholic fatty liver disease (NAFLD) is the most common chronic liver disease covering a spectrum of
histopathological changes in the liver ranging from simple steatosis to non-alcoholic steatohepatitis (NASH)
that may progress to hepatic fibrosis, cirrhosis, or hepatocellular carcinoma1–3. In the discovery of molecular
mechanisms and new drugs for NAFLD/NASH, a number of mouse models have been used: gene-deleted (e.g.
Alms1−/− or Mc4r−/−), high-fat and -carbohydrate diet-fed, nutrient deficient diet-fed, and CCl4 chronically
administered mouse models4. Genetic and lifestyle factors can lead to obesity, insulin resistance, and disorders
of lipid metabolism, resulting in the accumulation of free fatty acids in the liver and, as a consequence, mitochondrial dysfunction with oxidative stress, endoplasmic reticulum (ER) stress, hepatocyte cell death, and the
production of inflammatory chemokines and cytokines, such as monocyte chemotactic protein-1 (MCP-1) and
tumor necrosis factor (TNF-α). Thus, the multiple-hit hypothesis is the most widely accepted explanation of the
mechanisms underlying the progression of NASH5,6.
Genetically modifying systems, including transgenic and knockout technologies, have been utilized to understand the pathological roles of target molecules in a variety of diseases, including NAFLD/NASH7. One of the
most widely used mouse strains is the C57BL/6, with more than 20 inbred substrains derived from C57BL/6J
(BL6J) with “J” for Jackson and C57BL/6N (BL6N) with “N” for NIH. It has become clear that there are multiple
genetic differences between the BL6J and BL6N substrains. A whole-genome sequence comparison between the
substrains identified 34 coding single-nucleotide polymorphisms (SNPs), leading to amino acid substitutions in
the encoded protein, 2 coding small insertions or deletions (indels), 146 noncoding SNPs, 54 noncoding small
indels, and 43 structural variants including the nicotinamide nucleotide transhydrogenase (Nnt) mutation8,9. The
most widely known difference is the spontaneous deletion of exon 7–11 in the Nnt gene, resulting in a complete
Department of Pathological Biochemistry, Kyoto Pharmaceutical University, 5 Misasaginakauchi-cho, Yamashina-ku,
Kyoto, 607-8414, Japan. Correspondence and requests for materials should be addressed to S.A. (email: akiba@
mb.kyoto-phu.ac.jp)
Scientific REporTS |
(2019) 9:204 | DOI:10.1038/s41598-018-36862-7
1
www.nature.com/scientificreports/
Figure 1. CCl4-induced oxidative stress in BL6J and BL6N mice. BL6J and BL6N mice were intraperitoneally
administered CCl4 in corn oil at 0.31 mL/kg (2 times/week) for 6 weeks. (A) Levels of 13-HPODE-adducted
protein relative to those of GAPDH in the livers of BL6J and BL6N mice were determined by Western blotting.
The cropped blots were displayed in (A) and the full-length blots were presented in Supplementary Fig. 2. (B)
The relative intensity was measured using the NIH ImageJ software program and normalized to that of GAPDH.
The bar graphs represent the means ± SE (arbitrary units: A.U., n = 8–10/group). Significance was evaluated
using an ANOVA with the LSD post-hoc test. *P < 0.05, **P < 0.01, ns: non-significant.
absence of NNT, in the BL6J substrains, but not in the BL6N substrains10. Recently, Mekada et al. showed that
even the BL6N-derived substrains (C57BL/6NJ, C57BL/6Ntac and C57BL/6NCrSlc), have identified SNPs11.
These small genetic differences between the BL6J and BL6N substrains lead to notable differences in the metabolic phonotypes, including differences in the glucose tolerance, insulin secretion, weight regulation, energy
expenditure, and O2 consumption9. These findings suggest that caution be exercised regarding the use of mice
with mixed BL6J and BL6N genetic backgrounds in research on metabolic syndromes, such as diabetes. However,
the phenotypic differences between the BL6J and BL6N strains in NASH models remain unclear.
In the present study, we aimed to elucidate the pathological differences between these two substrains.
Results
Greater CCl4-induced hepatic oxidative stress in BL6J mice than in BL6N mice. Trichloromethyl
radical (.CCl3) metabolized from CCl4 by CYP2E1 in hepatocytes evokes lipid peroxidation and hepatotoxicity12,13. We confirmed that there was no marked difference in the mRNA expression of Cyp2e1 between the BL6J
and BL6N mice (Supplementary Fig. 1). The degree of lipid peroxidation was assessed by detecting the liver levels
of 13-hydroperoxyoctadecanoic acid (13-HPODE)-modified proteins, which reacts specifically with an anti-HEL
antibody (Fig. 1 and Supplementary Fig. 2). The intensity of a band of approximately 72 kDa was increased by
chronic CCl4 administration, and the increased level was significantly higher in the BL6J mice than in the BL6N
mice, indicating that more severe oxidative stress was induced by CCl4 administration in the BL6J mice than in
the BL6N mice. In addition, the level of the 13-HPODE-modified protein of 28 kDa was markedly higher in the
BL6J mice than in the BL6N mice in vehicle-injected groups, suggesting that the BL6J mice have greater oxidative
stress than BL6N mice even at the basal level.
Comparable CCl 4-induced hepatotoxicity and liver inflammation between the BL6 substrains. The degree of CCl4-induced hepatotoxicity in the BL6J and BL6N mice was compared by measuring
the serum levels of AST and ALT, markers of liver injury. As shown in Fig. 2A, the serum levels of AST and ALT
were significantly increased by chronic CCl4 administration, but there were no marked differences in (...truncated)