Cereblon enhances hepatic hepcidin production through estrogen-related receptor gamma activation

Abstract

Hepcidin (encoded by the HAMP gene), produced primarily by hepatocytes, is the master regulator of systemic iron homeostasis. Its dysregulation contributes to various iron-related metabolic disorders. Cereblon (CRBN) has been implicated in metabolic regulation, while estrogen-related receptor gamma (ESRRG) is known to govern energy homeostasis and mitochondrial function. In this study, we demonstrate a novel CRBN-ESRRG signaling pathway that mediates endoplasmic reticulum (ER) stress-induced hepatic HAMP expression. In mice and primary hepatocytes exposed to tunicamycin-induced ER stress, gene expression and biochemical analyses revealed significant increases in the transcript levels of hepatic Crbn, Esrrg, and Hamp. Correspondingly, hepcidin protein levels were elevated, accompanied by reduced serum iron levels and increased cellular iron levels, consistent with hepcidin-mediated regulation of iron distribution. Overexpression of Crbn enhanced ESRRG expression and increased hepatic hepcidin production, while knockdown of either Crbn or Esrrg attenuated this response. Chromatin immunoprecipitation assays demonstrated enhanced recruitment of ESRRG to the Hamp promoter. Collectively, these findings identify a CRBN-ESRRG regulatory axis that drives hepatic HAMP expression under ER stress and suggest a potential therapeutic target for ER stress-associated metabolic and iron disorders.

INTRODUCTION

Iron is an essential trace element necessary for fundamental biological processes, including oxygen transport, deoxyribonucleic acid (DNA) synthesis, and cellular energy metabolism. Systemic iron homeostasis is primarily regulated by hepcidin, a peptide hormone produced by the liver and encoded by the HAMP gene (1, 2). Hepcidin regulates plasma iron concentrations by binding to ferroportin, the only identified cellular iron exporter. This binding promotes the internalization and lysosomal degradation of ferroportin. As a result, this regulatory mechanism limits intestinal iron absorption and the release of iron from macrophages and hepatocytes, thereby maintaining iron homeostasis under both physiological and pathological conditions (3-5).

Dysregulated hepcidin expression is linked to a wide range of iron-related disorders, including anemia of chronic disease, hereditary hemochromatosis, and iron-loading anemias (6, 7). Hepcidin synthesis is tightly regulated by various physiological signals, including systemic iron availability, inflammatory cytokines like interleukin-6 (IL-6), which activates the signal transducer and activator of transcription 3 (STAT3) pathway, and bone morphogenetic proteins (BMPs), particularly BMP6, that signal through suppressor of mother against decapentaplegic (SMAD) transcription factors. Emerging evidence also underscores the role of metabolic stress responses, especially the unfolded protein response (UPR) triggered by endoplasmic reticulum (ER) stress, as novel modulators of hepcidin transcription (8, 9).

Cereblon (CRBN), a substrate receptor of the cullin 4-RING E3 ubiquitin ligase (CRL4-CRBN) complex, was initially identified as the main molecular target responsible for the teratogenic effects of thalidomide (10-12). Beyond its historical relevance, CRBN has recently emerged as a key regulator of metabolic processes, such as hepatic gluconeogenesis and insulin sensitivity (13, 14). Mechanistically, CRBN regulates with various transcription factors and signaling mediators, including AMP-activated protein kinase (AMPK), Krüppel-like factor 15 (KLF15), and cAMP-responsive element-binding protein H (CREBH). This interaction modulates pathways that are central to cellular energy and nutrient homeostasis (14-16). Despite these insights, the potential role of CRBN in regulating hepatic iron levels and controlling HAMP production has yet to be explored.

Estrogen-related receptor gamma (ESRRG; NR3B3) is an orphan nuclear receptor that serves as a crucial transcriptional regulator of mitochondrial biogenesis, oxidative phosphorylation, and glucose metabolism (17-21). ESRRG is highly expressed in metabolically active tissues, particularly in the liver, where its activity increases during nutrient deprivation and cellular stress (17, 19). By regulating the transcription of genes involved in oxidative metabolism and lipid catabolism, ESRRG plays a crucial role in maintaining cellular energy homeostasis (19, 21). However, despite its well-documented metabolic functions, the involvement of ESRRG in systemic iron regulation and the transcriptional control of HAMP in ER stress remains to be elucidated.

Given the emerging roles of CRBN and ESRRG in metabolic regulation and cellular stress responses, we hypothesized that CRBN modulates hepatic hepcidin expression and production through ESRRG under ER stress conditions. In this study, we investigated the functional interplay between CRBN and ESRRG in the transcriptional regulation of the HAMP gene. Using complementary in vivo and in vitro approaches, we demonstrated that CRBN enhances ESRRG expression and promotes its recruitment to the Hamp promoter, leading to increased hepcidin production. However, this effect was negated by silencing Crbn and Esrrg in primary hepatocytes and liver of mice. Collectively, these findings reveal a previously unrecognized CRBN-ESRRG regulatory axis in hepatic iron homeostasis and suggest potential therapeutic targets for managing iron-related metabolic disorders during ER stress.

RESULTS

CRBN and ESRRG are upregulated by ER stress and enhance hepatic hepcidin expression

To investigate the role of ER stress in the regulation of hepatic hepcidin, we administered tunicamycin (TU), a potent ER stress inducer, to mice and primary hepatocytes. TU treatment significantly increased hepatic mRNA levels of Crbn, Esrrg, and Hamp compared to control mice (Fig. 1A). Correspondingly, ELISA analysis revealed that TU markedly elevated serum hepcidin concentrations (Fig. 1B), while serum iron levels were prominently reduced (Fig. 1C), consistent with enhanced hepcidin-mediated iron sequestration. Additionally, TU treatment significantly increased Crbn, Esrrg, and Hamp mRNA expression (Fig. 1D) and elevated hepcidin secretion in primary hepatocytes (Fig. 1E). Furthermore, TU-treated hepatocytes displayed increased intracellular iron accumulation (Fig. 1F), aligning with enhanced hepcidin-dependent iron retention. Collectively, these findings suggest that ER stress promotes hepatic hepcidin expression and function through the transcriptional upregulation of Crbn and Esrrg.

Fig. 1.

ER stress increases the expression of Crbn, Esrrg, and Hamp genes in mice. (A) Wild-type (WT) mice were injected intraperitoneally with tunicamycin (TU, 0.5 mg/kg body weight) for 7 days. Gene expressions were analyzed by qPCR using gene-specific primers. (B) Serum hepcidin and (C) serum iron contents in the observed mice groups. (D) Mouse primary hepatocytes were treated with tunicamycin (TU, 5 μg/ml) for 12 h. Gene expression analysis using various gene-specific primers in primary hepatocytes. (E) Cellular hepcidin and (F) iron concentrations were measured in TU-treated cells. n = (4-6) mice per group. *P < 0.05 and **P < 0.01 vs. untreated control (CON) mice (A-C) or untreated control hepatocytes (D-F).

CRBN increases ESRRG and hepcidin levels in hepatocytes and liver of mice

To evaluate the functional consequences of CRBN induction, we overexpressed Crbn using an adenoviral system. Mice infected with Ad-Crbn exhibited a striking increase in hepatic mRNA levels of Crbn, Esrrg, and Hamp, with Crbn rising over 100-fold compared to Ad-GFP controls (Fig. 2A). Immunoblotting revealed a corresponding increase in CRBN, ESRRG, and hepcidin protein levels (Fig. 2B), resulting in a significant elevation of serum hepcidin concentrations (Fig. 2C). In contrast, knockdown of Crbn by Ad-shCrbn markedly reduced the mRNA expression of both Esrrg and Hamp (Fig. 2D). Although TU administration induced Crbn, Esrrg, and Hamp expression, this induction was strikingly attenuated by Crbn silencing (Fig. 2E). Immunoblotting confirmed reduced protein expression of CRBN, ESRRG, and hepcidin in the liver lysates of Ad-shCrbn-infected mice under ER stress (Fig. 2F). ELISA measurements indicated that the TU-induced elevation in serum hepcidin was notably blunted by Crbn knockdown (Fig. 2G). Together, these findings demonstrate that CRBN is essential for ER stress-induced ESRRG expression and subsequent hepcidin production in vivo.

Fig. 2.

CRBN promotes hepatic hepcidin expression and production in mice. WT mice were injected via tail-vein with Ad-GFP and Ad-Crbn for 7 days. (A) Gene expression levels were analyzed by qPCR using various gene-specific primers. (B) Liver tissue extracts from the indicated groups were analyzed by Western blot analysis with various antibodies. (C) Serum hepcidin contents in the indicated groups. (D) WT mice were transduced with Ad-Scramble (Ad-Scram) and Ad-shCrbn for 7 days. (E) These mice were then injected intraperitoneally with TU for 7 days. Total RNAs were utilized for qPCR analysis with gene-specific primers. (F) Liver tissue extracts were analyzed by Western blot analysis with the indicated antibodies. (G) Serum hepcidin contents in the observed mice groups. n = (4-6) mice per group. *P < 0.05, **P < 0.01, and ***P < 0.001. For panels A-C: vs. Ad-GFP. For panels D: vs. Ad-Scram. For panels E-G: *P < 0.05 and **P < 0.01 vs. TU ＋ Ad-Scram for knockdown comparisons.

ESRRG is required for CRBN-mediated induction of hepcidin production in mice

To determine whether ESRRG alone is sufficient to induce hepcidin expression, we overexpressed Esrrg in mouse liver using Ad-Esrrg. Adenoviral transduction of Esrrg resulted in a robust increase in hepatic Esrrg mRNA (~80-fold) and a significant upregulation in Hamp expression compared to Ad-GFP controls (Fig. 3A). Consistently, ELISA analysis revealed a 1.8-fold increase in circulating hepcidin levels in Ad-Esrrg-exposed mice (Fig. 3B), accompanied by approximately a 23% reduction in serum iron (Fig. 3C). To assess whether ESRRG is necessary for the hepcidin-inducing effects of CRBN and ER stress, we knocked down Esrrg using Ad-shEsrrg with and without TU treatment. As anticipated, TU treatment upregulated both Esrrg and Hamp expression, which was significantly reduced by Esrrg silencing (Fig. 3D). Concordantly, TU-induced hepcidin secretion was markedly reduced in Ad-shEsrrg-infected mice (Fig. 3E), demonstrating that ESRRG is required for ER stress-induced hepcidin synthesis. Collectively, these findings indicate that ESRRG is both necessary and sufficient to drive hepatic hepcidin expression and serves as a critical mediator of CRBN-dependent regulation under ER stress conditions.

Fig. 3.

ESRRG elevates hepatic hepcidin gene expression and production in mice. WT mice were intravenously injected with Ad-GFP and Ad-Esrrg for 7 days. (A) Gene expression analysis using gene-specific qPCR primers. (B) Serum hepcidin levels were measured. (C) Serum iron levels in the indicated groups. WT mice were transduced with Ad-Scram and Ad-shEsrrg for 7 days and were then exposed with TU for 7 days. (D) Total RNAs were utilized for qPCR analysis with gene-specific primers. (E) Measurement of serum hepcidin in the indicated groups of mice. n = (4-6) mice per group. *P < 0.05 and ***P < 0.001 vs. Ad-GFP (A-C) or vs. TU-treated Ad-Scram mice (D-E).

CRBN facilitates ESRRG recruitment to the Hamp promoter

To investigate the role of CRBN in Esrrg-dependent transcriptional regulation of hepcidin, we conducted a series of luciferase reporter assays and chromatin immunoprecipitation (ChIP) analyses. In AML-12 hepatocytes, TU treatment significantly increased Hamp promoter activity. Co-transfection of Crbn and Esrrg produced a greater induction of Hamp promoter activity than either construct alone, and this effect was further amplified by TU treatment (Fig. 4A), suggesting cooperative transcriptional activation of the Hamp promoter under ER stress conditions. Deletion analysis of the Hamp promoter revealed that transcriptional activation by TU or Esrrg was localized to the −300 to −100 bp region upstream of the transcription start site, with a notable decrease in activity when only the −100 bp region was present (Fig. 4B-C). To assess whether Esrrg directly targets a specific motif in the Hamp promoter, we created a mutant construct (Hamp-Luc mt) that disrupted the Esrrg response element located at −130 bp to −123 bp. The luciferase activity of the mutant construct was significantly reduced compared to the wild-type (Hamp-Luc wt), even in the presence of TU, confirming the importance of this region for Esrrg-mediated transcriptional activation (Fig. 4D). Subsequently, we performed ChIP-qPCR using an ESRRG antibody to evaluate its binding at the Hamp promoter. TU treatment increased ESRRG recruitment specifically to the proximal promoter region (−250 to −50 bp), while binding at a distal control region (−1800 to −2000 bp) remained unchanged (Fig. 4E). Importantly, knockdown of Crbn eliminated TU-induced ESRRG binding to the proximal promoter, indicating that CRBN is essential for effective ESRRG recruitment under ER stress conditions. In summary, these findings demonstrate that CRBN enhances Esrrg-dependent transcriptional activation of the Hamp gene by facilitating ESRRG recruitment to its promoter, revealing a novel mechanism by which CRBN amplifies hepcidin expression during ER stress (Fig. 5).

Fig. 4.

ESRRG acts as a crucial partner of hepcidin gene transcription in hepatocytes. (A) AML-12 cells were co-transfected with the indicated reporter gene, Crbn, and Esrrg. After transfection for 36 h, cells were treated with TU for 12 h. (B) AML-12 cells were transfected with wild type and deletion mutant of the Hamp reporters. After transfection for 36 h, cells were treated with TU for 12 h. (C) AML-12 cells were transfected with wild type and deletion mutant of the Hamp reporters, and Esrrg for 36 h. (D) Schematics of the wild-type (wt) and mutant form (mt) of the Hamp promoter encompassing constructs from −130 to −123 bp under the indicated conditions. AML-12 cells were co-transfected with the wild-type or mutant form of the Hamp promoters (mt), and Esrrg expression vector. Cells were treated with TU under the observed conditions. (E) ChIP assay for the occupancy of ESRRG on the Hamp gene promoter. WT mice were transduced with Ad-Scramble (Ad-Scram) and Ad-shCrbn for 7 days and were then injected intraperitoneally with TU for 7 days. Purified DNAs after chromatin immunoprecipitations were utilized for qPCR using specific primers binding to the specific proximal (Pro) and nonspecific distal (Dis) regions on the Hamp gene promoter. Input represents 10% of the purified DNA from each sample. *P < 0.05 and **P < 0.01 vs. vehicle-treated control (A-D) or vs. TU ＋ Ad-Scram (E).

Fig. 5.

Proposed schematic model for ER-mediated regulation of hepatic hepcidin gene expression and its production via CRBN-ESRRG signaling pathway. ER stress significantly elevates hepatic hepcidin gene expression by upregulating the CRBN-ESRRG signaling pathway, which subsequently increases the production of hepcidin. Elevated hepcidin production results in the regulation of systemic iron homeostasis. This model summarizes the CRBN-ESRRG-HAMP signaling axis linking ER stress to iron metabolism. Solid arrows indicate experimentally supported relationships; dashed arrows indicate untested mechanisms.

DISCUSSION

In this study, we identify a functional role for CRBN in the induction of hepatic hepcidin during ER stress and provide evidence that this response occurs, at least in part, through an ESRRG-dependent mechanism. We demonstrate that TU-induced ER stress increases the expression of CRBN, ESRRG, and hepcidin in both mouse liver and primary hepatocytes, which is accompanied by elevated hepcidin production and reduced circulating iron. These findings support previous research indicating that ER stress directly modulates hepcidin transcription via UPR-associated factors (CREBH and CHOP) and metabolic regulators (AMPK and KLF15), and expand this paradigm by identifying CRBN as an additional ER stress-responsive regulator (13, 15, 22, 23). Our findings enhance these roles by identifying CRBN as an inducible regulator in ER stress signaling that affects ESRRG-dependent hepatic hepcidin regulation.

Tunicamycin is a commonly used inducer of ER stress in hepatic models; however, prolonged treatment may also trigger broader cellular stress responses (24, 25). Importantly, the consistent regulation of the CRBN-ESRRG-HAMP pathway in both tunicamycin-treated mice and primary hepatocytes highlights the involvement of ER stress signaling in this regulatory mechanism. A significant finding of this study is that CRBN is essential for the complete induction of ESRRG and hepcidin during ER stress. In vivo overexpression of Crbn significantly increased hepatic ESRRG levels and boosted hepcidin expression, while Crbn knockdown reduced the tunicamycin-induced increases in both ESRRG and hepcidin (Fig. 1-3). Although these results clearly indicate a positive correlation between CRBN levels and ESRRG abundance, they do not clarify whether CRBN directly regulates ESRRG transcription, affects ESRRG protein stability, or influences ESRRG indirectly through broader cellular stress pathways. Additionally, Crbn silencing lowers both ESRRG expression and its occupancy at the Hamp promoter, suggesting that decreased recruitment may stem from a combination of diminished ESRRG availability and impaired transcriptional activation (Fig. 4). This observation aligns with the established transcriptional role of ESRRG in hepatic gluconeogenesis and oxidative metabolism (17, 18, 20, 26). Our data confirms that CRBN acts as an upstream regulator of ESRRG during ER stress, but the exact mechanism is still unclear. CRBN may enhance Esrrg transcription, stabilize ESRRG protein, or boost its transcriptional activity through co-factor–dependent mechanisms. Future studies should clarify these possibilities by evaluating Esrrg promoter activity, ESRRG protein half-life, ubiquitination status, and CRBN-dependent protein interactions under ER stress conditions.

Additionally, our promoter and ChIP analyses provide further evidence for a CRBN-ESRRG axis in hepcidin regulation (Fig. 4). ESRRG increased Hamp promoter activity and exhibited enhanced binding to the proximal promoter region during ER stress, which aligns with previous studies that identified ESRRG as a stress-responsive metabolic transcription factor (26-29). Mutating the putative ESRRG response element (ERRE) significantly decreased reporter activity, confirming its functional importance. While CRBN did not change the binding location of ESRRG in the Hamp promoter, knockdown of Crbn greatly reduced ESRRG occupancy under ER stress conditions. This suggests that CRBN enhances ESRRG responsiveness. Consequently, these findings position CRBN as a modulator that amplifies ESRRG-dependent transcription rather than acting as a traditional DNA-binding transcription factor.

In contrast, classical pathways such as BMP-SMAD and IL-6-STAT3 respond to systemic signals, iron overload, and inflammation, while ER stress-mediated regulation reflects disturbances in intracellular metabolism (9, 30-32). The present findings identify a CRBN-ESRRG axis that regulates hepatic HAMP expression under ER stress conditions; however, the relationship of this pathway to canonical BMP-SMAD signaling was not directly examined. As a result, our data do not establish whether the CRBN–ESRRG pathway functions independently, in parallel, or cooperatively with established hepcidin-regulatory pathways such as BMP-SMAD or IL-6-STAT3. Future studies will be necessary to clarify the extent of mechanistic crosstalk between these pathways in regulating hepatic iron homeostasis. During ER stress, increased CRBN expression promotes higher ESRRG protein levels and enhances ESRRG occupancy at the Hamp promoter, which drives hepcidin transcription. This ER stress–driven hepcidin induction contributes to reduced circulating iron levels, consistent with prior reports that ER stress activates hepcidin transcription through UPR-responsive transcription factor binding at the Hamp promoter (8, 23). We have identified a previously unrecognized CRBN-ESRRG axis that may play a role in coordinating proteotoxic stress responses with the regulation of systemic iron homeostasis. Clinically, this pathway could help explain the elevation of hepcidin seen in chronic metabolic or inflammatory liver diseases, where ER stress is a common pathological feature. Modulating CRBN or ESRRG activity might thus offer a therapeutic strategy for conditions marked by inappropriate hepcidin expression, such as anemia of inflammation or hereditary iron-loading disorders.

The clinical implications of this signaling pathway are potentially significant. Dysregulation of hepcidin contributes to both iron-restricted anemias and iron-loading disorders. In anemia of chronic disease, elevated hepcidin traps iron in macrophages and hepatocytes, limiting its availability for erythropoiesis. Our findings suggest that selectively inhibiting the CRBN-ESRRG axis may reduce stress-induced overproduction of hepcidin and improve iron mobilization in this context. Conversely, in conditions of systemic iron overload, such as hereditary hemochromatosis, activating the CRBN-ESRRG pathway could enhance hepcidin expression, decrease intestinal iron absorption, and mitigate toxic iron accumulation. Additionally, this pathway may be relevant in chronic liver disease and metabolic syndromes, where ER stress and iron imbalance often coexist.

Several limitations remain for future research. First, the precise molecular mechanism by which CRBN regulates ESRRG needs clarification. It is unclear whether CRBN primarily acts at the transcriptional level, stabilizes the ESRRG protein post-translationally, or modulates its activity through co-factor interactions. Second, the extent to which the CRBN-ESRRG-HAMP axis functions across various forms of metabolic stress and disease states requires investigation. Third, the upstream UPR signaling pathway responsible for CRBN expression induced by ER stress is still unclear. Although tunicamycin is a well-established pharmacologic inducer of endoplasmic reticulum stress, the present study did not directly validate the activation of canonical UPR markers such as CHOP, ATF4, or XBP1s in the primary experimental analyses. Therefore, the observed effects are interpreted within the context of a tunicamycin-induced ER stress model rather than through a direct examination of specific UPR pathway activation. Future studies will be necessary to elucidate how defined ER stress signaling pathways integrate with the CRBN–ESRRG axis.

Beyond upstream regulation, the downstream functional consequences of iron export also require further clarification. In our model, ER stress-induced hepcidin expression was associated with reduced serum iron levels and increased intracellular iron accumulation, consistent with hepcidin-mediated iron sequestration. However, the present study did not directly assess ferroportin expression or internalization. Therefore, the involvement of the ferroportin-hepcidin axis is inferred based on established mechanisms of hepcidin function. Future studies examining ferroportin protein levels, localization, and turnover will be crucial for confirming the functional impact of the CRBN-ESRRG-HAMP axis on systemic iron export. Additionally, since CRBN is the known target of immunomodulatory drugs such as thalidomide and its analogs, it will be important to determine whether pharmacological modulation of CRBN affects systemic iron metabolism through this axis. Such findings could have translational relevance for both oncology and hematology.

While hepcidin expression is regulated by multiple iron-related pathways, including BMP-SMAD signaling and cellular iron-sensing mechanisms, the present study specifically focused on defining the CRBN-ESRRG axis under ER stress conditions. Canonical regulators, including BMP6, SMAD7, and transferrin receptor 1 (TfR1), play significant roles in systemic iron homeostasis and the regulation of hepcidin (30, 31, 33). These pathways were not thoroughly evaluated in the current study. However, our preliminary observations indicate that CRBN may influence hepatic ferritin expression, suggesting a broader role in iron metabolism beyond just regulating hepcidin (data not shown). Additionally, ER stress has been linked to iron regulatory pathways through UPR-associated transcription factors like CHOP and CREBH (8, 23, 34), implying that CRBN operates within a more complex regulatory network. Further studies are needed to explore how CRBN integrates ER stress signaling with established iron-regulatory pathways and overall iron homeostasis in both physiological and pathological contexts.

In conclusion, our study reveals a novel CRBN-ESRRG-HAMP signaling pathway that connects ER stress with hepatic hepcidin production and systemic iron regulation (Fig. 5). By defining this CRBN-ESRRG-HAMP axis, our findings enhance the regulatory network governing hepatic hepcidin expression and highlight potential therapeutic targets for anemia and ER stress-related metabolic disorders.

MATERIALS AND METHODS

Animal experiments

Male C57BL/6J mice (8 weeks old) were purchased from Samtako BioKorea (Osan, Republic of Korea) and maintained under specific pathogen-free conditions with ad libitum access to standard chow and water. For the tunicamycin (TU) stimulation experiments, wild-type (WT) mice were injected intraperitoneally with TU (Sigma-Aldrich, St. Louis, MO, USA, 0.5 mg/kg of body weight) for 7 days, as previously described (24, 25). For viral transduction of Crbn and Esrrg, adenoviruses expressing the observed genes (Ad-Crbn, Ad-Esrrg) were delivered via tail-vein injection into WT mice (a single dose of 1 × 10⁹ plaque-forming units) for 7 days. For the disruption of viral transductions, WT mice were injected with adenovirus silencing Crbn (Ad-shCrbn) and Esrrg (Ad-shEsrrg) via tail-vein injection (a single dose of 1 × 10⁹ plaque-forming units). At the end of the experiments, the mice were sacrificed at indicated time points for liver and serum collection. Animal experiments were performed by the relevant institutional guidelines for the care and use of laboratory animals according to the policy of the National Institutes of Health (NIH).

Measurement of metabolic parameters

Serum and culture supernatant hepcidin concentrations were determined using a mouse hepcidin ELISA kit (CUSABIO, Wuhan, Hubei Province, China), according to the manufacturer’s instructions. Iron levels were measured using an iron assay kit (Abcam, Cambridge, MA, USA), as mentioned previously (13, 35).

Primary hepatocytes isolation and culture

Mouse primary hepatocytes (MPH) were isolated from the livers of 8-week-old male C57BL/6 mice (Samtako) by a two-step portal vein collagenase (Collagenase IA, Sigma-Aldrich, St. Louis, MO, USA) perfusion method and purified by centrifugation. Freshly prepared hepatocytes were cultured in M199 medium (Cellgro, Herndon, VA, USA), as described previously (14). After trypan blue staining for determination of viability and attachment, cells were infected with the indicated adenoviruses and then treated with TU (5 μg/ml).

Cell culture and transient transfection assays

AML-12 cells (immortalized mouse hepatocyte) were cultured in DMEM/F-12 medium (Gibco-BRL, Grand Island, NY, USA) supplemented with 10% fetal bovine serum (FBS, Hyclone, Logan, UT, USA), insulin-transferrin-selenium (ITS, Gibco-BRL), dexamethasone (40 ng/ml, Sigma-Aldrich), and antibiotics (Gibco-BRL). The cells were maintained in a humidified atmosphere containing 5% CO₂ at 37°C. Transient transfection assays were performed with AML12 cells as reported previously (14, 36). Briefly, transient transfection was conducted using Lipofectamine 2000 (Invitrogen, Carlsbad, CA, USA) with the observed promoter together with expression vectors and/or exposed with TU. The total amount of DNA was adjusted to 0.8 μg per well by the addition of each appropriate amount of empty vector and β-galactosidase plasmids as the internal control. Luciferase activity was measured and normalized to β-galactosidase activity. All data are representative of at least three independent experiments.

Plasmids and DNA constructs

The reporter plasmid containing the mouse Hamp gene promoter and Esrrg cDNA were kindly provided by Dr. Hueng-Sik Choi (Chonnam National University, Gwangju, Republic of Korea), as previously described (20, 26). Expression vectors of Crbn were used as described previously (13, 37). The point mutation form of Hamp promoters was generated using the site-directed mutagenesis kit (Stratagene, La Jolla, CA, USA) using the specific primers: forward, 5’-TGGCTGTAAAAGACACAACC-3’, and reverse, 5’-GG-TTGTGTCTTTTACAGCCA-3’. All constructs or plasmids were confirmed by DNA sequencing.

Recombinant adenoviruses

Adenoviruses encoding full-length of Crbn (Ad-Crbn), Esrrg (Ad-Esrrg), green fluorescent protein (GFP), Ad-shCrbn and Ad-shEsrrg were conducted as previously described (13, 14, 26). Adenoviruses were intravenously injected to WT mice for 7 days, which was conducted according to the manufacturer’s instructions.

RNA isolation and quantitative RT-PCR

Total RNA was isolated from primary hepatocytes and mouse liver tissues using TRIzol reagent (Invitrogen). Complementary DNA (cDNA) was synthesized with the Maxima^Ⓡ First Strand cDNA synthesis kit (Fermentas, Vilnius, Lithuania), as previously mentioned (13, 26). Quantitative PCR was performed with Power SYBR Green Master Mix (Applied Biosystems, Warrington, UK) on a StepOne Plus Real-time PCR system (Applied Biosystems). We determined the expressions of Crbn, Esrrg, and Hamp genes using qPCR, as previously described (13, 20, 37). Relative gene expression was calculated using the ∆∆Ct method and normalized to ribosomal protein L32 mRNA.

Western blot analysis

Liver tissue lysates were prepared in a radioimmunoprecipitation assay (RIPA) buffer (Elpis-Biotech, Dajeon, Republic of Korea) containing protease inhibitors. Equal amounts of proteins were separated by SDS-PAGE and transferred to nitrocellulose membranes (Amersham Biosciences, Piscataway, NJ, USA), as described previously (13, 20, 27). The membranes were blocked with 5% non-fat milk in TBS-T and probed with primary antibodies against CRBN (Cell signaling Technology, Danvers, MA, USA), ESRRG (R&D systems, Minneapolis, MN, USA), β-actin (Santa Cruz Biotechnology, Santa Cruz, CA, USA), and hepcidin (Abcam) antibodies. Horseradish peroxidase–conjugated secondary antibodies were applied, and signals were detected using an ECL Western Blot Detection Reagent (Amersham Biosciences).

Chromatin immunoprecipitation assay (ChIP)

ChIP assays were conducted according to the manufacturer’s protocol, as described previously (13, 26, 37). Briefly, cross-linked chromatin was sonicated and immunoprecipitated with an ESRRG (R&D systems) antibody. After elution, DNA fragments were purified and analyzed by qPCR using specific primers spanning the proximal (Pro, −250/−50 bp) and distal (Dis, −2000/−1800 bp) regions of the Hamp promoter. Results are presented as fold enrichment relative to input DNA. The specific primers used for the analyses were: proximal, forward 5’-GTTTAGGGGAAAGAAGGGGA-3’ and reverse 5’-TCATTTCCAAGAAAATAGTG-3’; distal, forward 5’-TGGAAAATGGACAAAAAATA-3’ and reverse 5’-ATCAGCACCCCAGGGCCTCA-3’.

Statistical analysis

All data are expressed as mean ± standard error of the mean (SEM). Statistical significance of the differences between the two groups was determined using Student’s t-test, and multiple comparisons were analyzed using one-way analysis of variance (ANOVA). P values < 0.05 were considered statistically significant. GraphPad Prism (GraphPad Software, CA, USA) software was used for all analyses.