Powering the powerhouse: Heme oxygenase-1 regulates mitochondrial function in non-alcoholic fatty liver disease (NAFLD)

In this issue, Li et al.¹ advanced our understanding of heme oxygenase-1 (HO-1) in non-alcoholic fatty liver disease (NAFLD). They developed hepatocyte-specific HO-1 knockout (KO) mice (HO-1^HEPKO) and used LO-2 human hepatocytes with HO-1 knocked down or overexpressed.

For the past decade, there has been a conundrum regarding the role of HO-1 in liver diseases. Most studies show a protective action of HO-1 in NAFLD.^2–4 However, in 2014, Jais et al. created hepatocyte-specific HO-1 knockout mice they named Lhoko and showed the opposite, that is, HO-1 seemed to exert deleterious effects and promote inflammation in NAFLD.⁵ This work demonstrated increased HO-1 mRNA expression in diet-induced fatty liver, similar to the present Li et al. study,¹ which found that HO-1 mRNA levels were also significantly higher with high-fat feeding-induced NAFLD. However, the two studies have entirely different findings regarding the role of HO-1 in NAFLD.

The dichotomy in the findings could arise from how the mice were generated in each study. Li et al. utilized CRISPR technology to generate the HO-1^HEPKO mice,¹ and Jais et al. used classical homologous recombination to insert LoxP sites flanking exon 2 of the HO-1 gene locus.⁵ Jais et al. crossed these animals with Alb-Cre transgenic mice generating the Lhoko mice,⁵ which caused a frameshift in exon 3 and an early stop codon that resulted in a truncated peptide consisting of nine amino acids. One must wonder whether the truncated nine amino acid peptide originating from the HO-1 gene in the Lhoko mice might have some protective action causing a differential finding compared to the HO-1^HEPKO mice described by Li et al.¹ The vast majority of the scientific literature has demonstrated the protective effects of HO-1 as a protector against oxidative stress and as an inhibitor of lipid accumulation with favorable effects on health and NAFLD (Figure 1).

FIGURE 1.

Heme Oxygenase-1 (HO-1) and its production of bilirubin in normal liver, NAFLD, and NAFLD with reduced lipids (improved NAFLD). The activity of HO-1 to produce bilirubin has been shown to be lower in NAFLD and increased with reducing liver fat content. Li et al. showed that reducing HO-1 activity increases Tomm20 that induces mitochondrial dysfunction.

In the study by Li et al.¹ they determined the effects of HO-1 on mitochondria with and without treatments with palmitic acid to mimic the environment for determining the mechanism by which HO-1 interacts with mitochondrial function in fat-loaded hepatocytes. HO-1 has been previously shown to be located in the mitochondria of the liver and controls mitochondrial heme and its metabolism to bilirubin.⁶ To identify the function of HO-1 on mitochondria, Li et al. used mitochondrial microarray analysis in their HO-1^HEPKO and littermate mice,¹ in which they identified that HO-1 inhibits the gene translocase of outer mitochondrial membrane 20 (Tomm20). They show that the loss of hepatic HO-1 in mice and human hepatocytes significantly increased Tomm20 mRNA by ~eightfold in KO compared to ~fourfold in WT. This was correlated with higher mitochondrial dysfunction with lower levels of HO-1, which was indicated by increased mitochondrial fragmentation and mitochondrial fission. To support that Tomm20 induces mitochondrial dysfunction, they used shRNA to suppress levels and demonstrate that this improves mitochondria function. The effect of HO-1 on mitochondrial function was further evidenced by showing that when the HO-1 levels are low, it impairs mitochondrial membrane potential, decreases ATP production, enhances mitochondrial reactive oxygen species (ROS) production, and causes mitochondrial DNA damage both in vivo and in vitro. The induction of HO-1 can be done via dietary supplements, which are protective against NAFLD.⁷

The profound effect that the induction of HO-1 has in attenuating ROS production is likely due to increased levels of bilirubin generation.⁸ Bilirubin is one of the most potent endogenous antioxidants in the body,⁹ and mice lacking the biliverdin reductase A (BVRA) enzyme, which is responsible for the reduction of biliverdin to bilirubin, exhibit high levels of ROS production.^10,11 Bilirubin can also impact mitochondrial function by acting as a hormone through its interaction with the nuclear receptor, PPARα.^12–15 Bilirubin binding to PPARα causes an exchange of specific coregulators, recruiting coactivators, and dissociating corepressors, which drive the expression of genes involved in the regulation of mitochondrial function, such as uncoupling protein-1 (UCP1) and β3 adrenergic receptor (Adrb3) in adipose tissue.¹³ Several studies have recently demonstrated the important role of bilirubin in protecting the liver against NAFLD. One study showed that treatment of obese mice with NAFLD with bilirubin nanoparticles reduced lipid accumulation and improved liver function as measured by the aspartate transaminase (AST) liver dysfunction biomarker.¹⁶ The bilirubin nanoparticles raised the hepatic β-oxidation pathway through stimulation of PPARα transcriptional activity, which induced fat utilization and burning, resulting in significantly higher plasma levels of the ketone β-hydroxybutyrate.¹⁶ Further supporting the positive actions of the heme oxygenase pathway on NAFLD, mice that are deficient in hepatocyte BVRA exhibited reduced hepatic PPARα activity, increased activity of glycogen synthase kinase 3β (GSK3β), and developed significantly more lipid accumulation in the liver during high-fat feeding.^11,17 Adipose-specific BVRA KO mice also showed significantly reduced mitochondria number and larger adipocyte size.¹⁸

Another HO metabolite, carbon monoxide (CO),⁹ has also been demonstrated to have positive effects on mitochondria, which might also be beneficial for NAFLD. Previous studies have demonstrated the anti-obesity effects of carbon monoxide-releasing molecules (CORMs).^19–21 Chronic CO administration has profound effects on the remodeling of mitochondria through upregulation of UCP1, peroxisome proliferator-activated receptor gamma coactivator 1-alpha (PGC-1α), and nuclear respiratory factor 1 (NRF1).¹⁹ CO has been reported to elevate mitochondrial oxygen consumption, reduce ATP-linked respiration, and increase glycolysis.²¹ These changes in mitochondrial function are mirrored by enhanced oxygen consumption and heat production in mice chronically treated with CO donors.^19,20 The loss of hepatic HO activity could impact mitochondrial function and promote NAFLD progression through decreases in local CO production. However, these are yet to be tested.

In summary, Li et al.¹ provide a better understanding of how HO-1 improves NAFLD, which likely occurs via enhancing mitochondrial function, possibly by its production of bilirubin and/or CO. This work highlights the important protective role of the hepatic heme oxygenase system (HO-biliverdin-bilirubin-CO). It suggests that HO-1 and induction of this pathway significantly improve NAFLD opening a future potential therapeutic application.