Genetic Deficiency of Indoleamine 2,3-dioxygenase Aggravates Vascular but Not Liver Disease in a Nonalcoholic Steatohepatitis and Atherosclerosis Comorbidity Model

Abstract

Nonalcoholic steatohepatitis (NASH) is a chronic liver disease that increases cardiovascular disease risk. Indoleamine 2,3-dioxygenase-1 (IDO1)-mediated tryptophan (Trp) metabolism has been proposed to play an immunomodulatory role in several diseases. The potential of IDO1 to be a link between NASH and cardiovascular disease has never been investigated. Using Apoe^-/-and Apoe^-/-Ido1^-/- mice that were fed a high-fat, high-cholesterol diet (HFCD) to simultaneously induce NASH and atherosclerosis, we found that Ido1 deficiency significantly accelerated atherosclerosis after 7 weeks. Surprisingly, Apoe^-/-Ido1^-/- mice did not present a more aggressive NASH phenotype, including hepatic lipid deposition, release of liver enzymes, and histopathological parameters. As expected, a lower L-kynurenine/Trp (Kyn/Trp) ratio was found in the plasma and arteries of Apoe^-/-Ido1^-/- mice compared to controls. However, no difference in the hepatic Kyn/Trp ratio was found between the groups. Hepatic transcript analyses revealed that HFCD induced a temporal increase in tryptophan 2,3-dioxygenase (Tdo2) mRNA, indicating an alternative manner to maintain Trp degradation during NASH development in both Apoe^-/- and Apoe^-/-Ido1^-/mice^-. Using HepG2 hepatoma cell and THP1 macrophage cultures, we found that iron, TDO2, and Trp degradation may act as important mediators of cross-communication between hepatocytes and macrophages regulating liver inflammation. In conclusion, we show that Ido1 deficiency aggravates atherosclerosis, but not liver disease, in a newly established NASH and atherosclerosis comorbidity model. Our data indicate that the overexpression of TDO2 is an important mechanism that helps in balancing the kynurenine pathway and inflammation in the liver, but not in the artery wall, which likely determined disease outcome in these two target tissues.

Keywords: IDO; NASH; atherosclerosis; immunometabolism; inflammation.

Figure 1

Characterization of the NASH and atherosclerosis dual model. Apoe^−/− mice were fed chow or a high-fat-cholesterol diet (HFCD) for 7 weeks, and representative pictures of atherosclerotic disease and NASH are shown; complete descriptive data of chow- and HFCD-fed mice are shown in Table 1. Top panels show en face lipid staining with Sudan-IV; middle panels show H&E staining showing hepatocyte ballooning; and the bottom panels show collagen deposition detected by picrosirius red staining in (A) chow- and (B) HFCD-fed Apoe^−/− mice. Bar = 50 µm.

Figure 2

Effects of IDO1 genetic ablation on NASH and atherosclerosis development. Apoe^−/− or Apoe^−/−Ido1^−/− mice were fed a high-fat cholesterol diet (HFCD) for 7 weeks; pooled data from two independent experiments are shown. (A) Quantification of en face Sudan IV-stained aortic arches (n = 14–15). (B) Representative pictures of the aortic arches. (C) Representative pictures of the atherosclerotic burden stained by ORO in aortic root sections of Apoe^−/− or Apoe^−/− Ido1^−/− mice (n = 2/group). (D) Representative picture of Mac-2⁺ macrophage infiltration in the aortic roots of Apoe^−/− or Apoe^−/− Ido1^−/− mice (n = 2/group). Panel (E) shows total cholesterol and (F) triglyceride levels in plasma (n = 15–16). (G,H) Total levels of cholesterol and triglycerides in Apoe^−/− and Apoe^−/−Ido1^−/− mice livers (n = 15–16). (I) Representative pictures of H&E-stained liver sections. (J) Liver/body ratio (n = 15–16), (K,L) plasma ALT and AST levels (n = 15–16). (M) Relative hepatic collagen (Col1a1) mRNA expression (n = 15–16) and (N) hepatic hydroxyproline levels (n = 15–16). (O) Representative pictures of picrosirius red-stained liver sections. ** p < 0.01; differences were detected using the Mann–Whitney U test; dotted lines refer to baseline levels of Apoe^−/− mice fed a chow diet for 7 weeks (Table 1); orange and blue colours are used to identify female and male mice, respectively. Bar = 50 µm.

Figure 3

Effects of IDO1 genetic ablation on liver inflammation. Apoe^−/− or Apoe^−/−Ido1^−/− mice were fed a high-fat cholesterol diet (HFCD) for 7 weeks; pooled data from two independent experiments are shown. (A) Relative Cd68 mRNA expression (n = 15–16); (B) Relative Clec4f mRNA expression (n = 15); (C) Quantification of immunofluorescently stained Mac-2⁺ macrophages (n = 15); and (D) representative Mac-2⁺ fluorescence staining of liver from Apoe^−/− or Apoe^−/−Ido1^−/−; Bar = 50 µm. (E) Relative hepatic mRNA expression of M1-like macrophage markers (Cd80, Il12, and Cxcl10 mRNA, and CCL2 and TNF protein) (n = 15–16) and (F) M2-like macrophage markers (Chil3, Arg1, and Cd206 mRNA, and IL-10 protein) (n = 15–16). Orange and blue colours are used to identify female and male mice, respectively; no differences between groups were detected using a Mann–Whitney U test.

Figure 4

Systemic and local tryptophan degradation rates. Apoe^−/− or Apoe^−/− Ido1^−/− mice were fed a high-fat cholesterol diet (HFCD) for 7 weeks; pooled data from two independent experiments are shown. The L-kynrenine to Trp ratio (Kyn/Trp) in (A) plasma (n = 15–16), (B) artery homogenate (n = 14), and (C) liver homogenate (n = 15–16), was estimated using specific ELISA kits, as described in the methods. (D) shows the correlation between % lesion and Kyn/Trp ratio in the aorta. (E) shows the correlation between relative Cd68 mRNA and the Kyn/Trp ratio in the liver. *** p < 0.001; Differences were detected using the Mann–Whitney U test. Correlations were determined using simple linear regression.

Figure 5

Fatty acids and iron regulate TDO2-dependent Trp degradation and consequences for THP-1 macrophage activation. (A,B) TNF, CCL2, and IL-1β protein levels and relative Tdo2 mRNA expression in livers from Apoe^−/− and Apoe^−/−Ido1^−/− mice fed a high-fat cholesterol diet (HFCD) for 3.5 and 7 weeks. (A,B) Pooled data from two independent experiments are shown; n = 14–15/group. © Relative expression of TDO2 mRNA in HepG2 cells treated with palmitic acid (PA, 500 μM) or iron (FeSO_4, 100 μM) (n = 5). (D) Kyn/ Trp ratio in the supernatants of HepG2 cells treated with PA (500 μM), FeSO_4, (100 μM), or FeSO₄ + TDO2-inhibitor LM10 (0.62 μM) (n = 7). (E) IL-1β release from THP-1 macrophages pre-treated with conditioned media from HepG2 cells incubated with PA (500 μM), FeSO₄ (100 μM), or FeSO₄ + TDO2-inhibitor LM10 (0.62 μM); in addition, cells were stimulated with LPS (10 ng/mL, 4 h) and ATP (5 mM, 30 min) for activation of the inflammasome and IL-1β release (n = 8). ^& p < 0.06; ^# p < 0.08; * p < 0.05; ** p < 0.01; *** p < 0.001; **** p < 0.0001. (A) Differences were detected using the Mann–Whitney U test. (B,C) Differences were detected using a one-way ANOVA and Dunn’s post hoc test.