The major findings of this work are as follows: (1) Iron, in the form of FeNTA, up-regulates HMOX1
gene expression in human Huh-7, and cell lines (9-13 and CNS3) stably expressing HCV proteins (Figure ); (2) Iron significantly increases Nrf2 protein levels in human hepatoma cells, and silencing the Nrf2
gene with Nrf2-specific siRNA abrogates the up-regulation of HMOX1 by iron (Figures and ); (3) Iron does not significantly change Bach1 protein levels in human hepatoma cells (Figure ); (4) Iron increases ROS (Figures and ) but decreases HCV gene expression (Figure ) in human hepatoma cells; and (5) These effects are blocked by the selective iron chelator DFO (Figures -). However, none of the effects is produced by Na3
NTA, establishing that they are due to iron and not to the NTA anion. These results show clearly that iron is capable of acting directly on hepatoma cells and on HCV gene expression in hepatoma cells, without the need for mediation of effects by other tissues, organs, or cell types. Thus, it appears that iron exerts manifold effects on HCV-infected hepatocytes. On the one hand, it increases ROS and oxidative stress, acting in concert with HCV proteins, especially the core protein. On the other hand, it induces HMOX1 by increasing expression and activity of Nrf2 (Figures , and ), and it decreases levels of CNS3 or NS5A mRNA and protein expression (Figure ). These latter effects are likely mediated by the recently described iron-dependent inactivation of the HCV RNA polymerase NS5B[43
HMOX1 is a heat shock protein (also known as HSP 32), which can be induced to high levels, not only by heat shock, but also by a large number of physiological or pathological stressors[30-33
]. Nrf2 is a basic leucine zipper transcriptional activator[44,45
]. It protects cells against oxidative stress through antioxidant response element (ARE)-directed induction of several phase 2 detoxifying and antioxidant enzymes, including HMOX1[35,46
mice displayed a dramatically increased mortality associated with liver failure when fed doses of ethanol that were tolerated by wild type mice, establishing a central role of Nrf2 in the natural defense against ethanol-induced liver injury[47
]. Cobalt protoporphyrin (CoPP)-mediated induction of HMOX1 involves increased Nrf2 protein stability in human hepatoma Huh-7 cells[35
]. In this study, silencing Nrf2 by Nrf2-siRNA markedly abrogated FeNTA-mediated up-regulation of HMOX1 mRNA levels. Therefore Nrf-2 plays a central role in up-regulation of HMOX1
gene expression by FeNTA (Figure ).
Bach1, a member of the basic leucine zipper family of proteins, has been recently shown to be a transcriptional repressor of HMOX1, and to play a critical role in heme-, CoPP-, SnMP- and ZnMP-dependent up-regulation of the HMOX1
]. Upon exposure to heme, heme binds to Bach1 and forms antagonizing heterodimers with proteins in the Maf-related oncogene family. These heterodimers bind to MAREs, also known as AREs, and suppress expression of genes that respond to Maf-containing heterodimers and other positive transcriptional factors. Surprisingly, ZnMP does not bind to Bach1, but it still produces profound post-transcriptional down-regulation of Bach1 protein levels by increasing proteasomal degradation and transcriptional up-regulation of HMOX1[53
]. In contrast, iron does not affect levels of Bach1 mRNA (data not shown) or protein (Figure ), suggesting that Bach1 is not involved in up-regulation of the HMOX1
gene by iron.
Expression of HMOX1 was recently reported to be decreased in human livers from patients with chronic hepatitis C[54,55
] including some with only mild fibrosis. The reasons for this are not known currently. It is known that levels of expression of the HMOX1
gene depend in part upon genetic factors (lengths of GT repeats in the promoter[56-59
] and a functional polymorphism (A/T) at position -413 of the promoter[60,61
]. Higher expression and/or induction of HMOX1 are probably beneficial to mitigate liver cell injury in HCV infection, as well as in other liver diseases. This may be a therapeutic goal, achieved by treatment with heme or CoPP or with silymarin[62
] or other herbal products or compounds that combine anti-oxidant, iron-chelating and HMOX1-inducing effects.
Recently, we showed that HCV expression in CNS3 cells increases the levels of HMOX1 mRNA and protein[41
]. This induction is likely in response to oxidative stress. More recently, we showed that micro RNA-122, which is expressed at a high level in hepatocytes, causes down-regulation of Bach1, which, as already described, tonically down-regulates the HMOX1
]. In addition, we and others have shown that expression of micro RNA-122 is required for HCV replication in human hepatoma cells[40,63,64
]. Whether iron affects levels of micro RNA-122 has not yet been assessed, to our knowledge.
Others recently reported that iron binds to NS5B, the RNA dependent RNA polymerase of HCV, and inhibits its enzymatic activity[43
]. The HCV replicon system used in that study showed changes in the gene expression of certain genes involved in iron metabolism, including down-regulation of ceruloplasmin and transferrin receptor 1 but up-regulation of ferroportin thus producing an iron-deficient phenotype[65
]. The authors speculated that the HCV genes and proteins somehow produced these changes in order to diminish the effects of iron to inhibit NS5B RNA polymerase activity and to decrease HCV protein expression.
Regardless of these results in cell culture models, the preponderance of clinical evidence[10-15,24-29
] supports the view that iron acts as a co-morbid or synergistic factor in chronic hepatitis C infection. Because both iron and HCV infection increase oxidative stress within hepatocytes, one attractive mechanistic explanation for the additive or synergistic affects of these two perturbations is that they act, at least in part, by increasing oxidative stress in the form of highly reactive oxygen species. These considerations provide additional rationale for the notion that reduction of iron and anti-oxidant therapy[62
] may be of benefit in the management of difficult to cure chronic hepatitis C[10-15,24-29,66-68
]. Iron reduction has usually been achieved with therapeutic phlebotomies. However, deferasirox (Exjade) recently has been approved in the USA and other countries as oral chelation therapy for iron overload states. Thus, studies of deferasirox for therapy of chronic hepatitis C are timely and important[69
], especially because the therapy of chronic hepatitis C currently is fraught with side effects, difficulties of adherence and rates of response that are not better than about 50%[70-72
In conclusion, iron can cause or exacerbate liver damage, including viral hepatitis. In the work reported in this paper we assessed effects of iron and iron chelators on liver cells, some of which also expressed genes and proteins of the HCV. Iron increased oxidative stress and led to up-regulation of the HMOX1 gene, a key cytoprotective gene. A mechanism for this action was to increase expression of the positive transcription factor Nrf-2. In contrast, iron did not affect expression of Bach1. Iron decreased expression of HCV genes and proteins. All the effects of iron were abrogated by DFO. The induction of HMOX1 helps to protect liver cells from the damaging effects of the HCV.