Hepcidin Deficiency Disrupts Iron Homeostasis and Induces Ferroptosis in Zebrafish Liver

Abstract

Hepcidin is a key regulator of systemic iron homeostasis, which is essential for maintaining iron balance and cellular health. To investigate its role in zebrafish, we empolyed a hepcidin knockout model. Morphological and histological analyses revealed pale livers and significant iron accumulation in hep^−/− zebrafish, particularly in liver, skin, and egg tissues. RNA sequencing identified 1,424 differentially expressed genes (DEGs) between wild-type (WT) and hep^−/− zebrafish, with significant enrichment in pathways related to ferroptosis, fatty acid degradation, and heme binding. Western blot analysis showed reduced levels of key iron-related proteins, including GPX4, Fth1, and ferroportin (FPN), indicating impaired iron transport and increased oxidative stress. Gene Ontology (GO) and KEGG analyses highlighted disruptions in iron metabolism and lipid oxidation, linking iron overload to ferroptosis in the absence of hepcidin. These findings demonstrate that hepcidin deficiency leads to profound dysregulation of iron homeostasis, driving lipid peroxidation and ferroptosis in the zebrafish liver. Our study provides mechanistic insights into the molecular consequences of hepcidin loss, advancing our understanding of iron-related oxidative damage and its physiological impacts.

1. Introduction

Iron is a vital element for vertebrates, playing indispensable roles in numerous biological processes, including oxygen transport via hemoglobin and myoglobin, electron transport, oxidative phosphorylation, the tricarboxylic acid cycle, and the synthesis of iron-containing proteins [1]. However, iron imbalance, whether deficiency or overload, can result in severe physiological disorders, including anemia, oxidative stress, and even death [2]. Achieving systemic iron homeostasis is critical and relies on tightly regulated pathways involving iron uptake, storage, utilization, and export. Among these mechanisms, the iron-regulatory hormone hepcidin plays a central role and is evolutionarily conserved across vertebrates, from fish to mammals [3]. Despite extensive studies on hepcidin in mammals, the systemic impact of its absence, particularly in the liver, remains poorly understood in fish.

Hepcidin was first identified as a liver-derived antimicrobial peptide [4,5] before being recognized as a key regulator of systemic iron metabolism [6,7,8]. Structurally, hepcidin comprises three domains, namely a signal peptide, a pro-peptide, and a bioactive mature peptide, stabilized by four intramolecular disulfide bonds formed by eight conserved cysteine residues [9,10]. Hepcidin achieves iron homeostasis by binding to ferroportin (FPN), the only known cellular iron exporter, promoting its internalization and degradation, thereby limiting iron efflux into circulation [11,12,13]. In hepcidin-deficient states, uncontrolled ferroportin activity leads to systemic iron overload, as observed in hepcidin-null mice, which exhibit excessive iron deposition in multiple tissues [8,14]. While these studies provide a foundation, the downstream consequences of hepcidin loss, especially in liver-specific contexts, demand further exploration.

Iron metabolism is orchestrated by an interconnected network of proteins, including transferrin receptors (TFR), divalent metal transporter 1 (DMT1), ferritin, and ferroportin [1,15]. While TFR and DMT1 mediate cellular iron uptake, ferritin stores excess intracellular iron, and ferroportin exports iron to the extracellular environment [16]. Hepcidin fine-tunes this system by regulating ferroportin activity to prevent excessive iron release [13,17]. Disruption of this equilibrium in hepcidin-deficient models often results in pathological iron accumulation. However, the liver’s dual roles as an iron reservoir and the primary site of hepcidin production make it a particularly vulnerable organ under conditions of hepcidin loss.

Iron overload is intricately linked to lipid peroxidation [18] and ferroptosis [19], a specialized form of iron-dependent cell death driven by the accumulation of lipid peroxides [20]. Glutathione peroxidase 4 (GPX4), a key regulator of ferroptosis, protects cells from oxidative damage by efficiently neutralizing lipid peroxides [21,22]. In conditions of iron overload, reduced GPX4 activity exacerbates oxidative stress, suggesting a direct link between iron dysregulation and liver pathophysiology. Despite this connection, the role of hepcidin in mediating ferroptosis in the liver remains largely unexplored.

Despite the well-established role of hepcidin in regulating systemic iron homeostasis through ferroportin degradation, critical gaps remain in understanding its tissue-specific functions, particularly in aquatic vertebrates. Previous studies in mammals have primarily focused on systemic iron overload phenotypes in hepcidin^−/− models [8,14], yet unresolved issues highlight the necessity of this study: Evolutionary conservation of hepcidin’s function, while mammalian hepcidin is known to mediate the disorder of iron homeostasis [8,18], its protective role in teleost fish—where iron metabolism exhibits unique adaptations to aquatic environments—remains poorly characterized. Zebrafish offer a valuable model for investigating genetic and physiological processes due to their high genetic similarity to humans—approximately 70% of human genes have identifiable zebrafish orthologs—and their compatibility with gene-editing technologies [23]. Zebrafish provide unique opportunities to study the systemic and tissue-specific effects of hepcidin loss, particularly in the liver, where the interplay between iron homeostasis and oxidative stress can be examined in detail.

In this study, we utilized a stable hepcidin knockout zebrafish model as previously described [24], to investigate the effects of hepcidin deficiency. By integrating transcriptomic profiling, iron quantification, and histological analyses, we demonstrated that hepcidin loss results in significant liver iron overload, lipid peroxidation, and ferroptosis. Our findings reveal the critical role of hepcidin in maintaining liver iron homeostasis and preventing oxidative damage, providing novel insights into its function in vertebrate iron regulation and its potential implications for liver pathophysiology.

2. Materials and Methods

2.1. Zebrafish Husbandry

All zebrafish procedures were approved by the Shanghai Ocean University Animal Care and Use Committee (approval date: 2 March 2024; approval code: SHOU-DW-2024-030). The wild-type zebrafish (AB strain) were purchased from the Shanghai Institute of Biochemistry and Cell Biology, Chinese Academy of Sciences (SIBCB). The light cycle was maintained at 10 h of darkness and 14 h of light.

2.2. Construction of Zebrafish Mutant by CRISPR/cas9 Genome Editing

As previously described [24], CRISPR gRNAs were designed using the online tool ZiFiT. The double sgRNA target sequences for hepcidin (NCBI accession number: NC_007127.7) are shown in

Table 1. The primers used in this study.

Primer	Sequence (5′-3′)	Location in GRCz11	Primer Binding Location	Product Size (bp)
Hep-F1	GATTAAAGGGACTACGCCGAATG	chr16:50,289,518-50,289,540	5′ UTR	986
Hep-R1	TAGGCCAAAATCATATCCCTGT	chr16:50,290,678-50,290,699	Intron2
Hep-F2	CTTCCAGATCACAGCCGTTCCCTTCATACAG	chr16:50,290,064-50,290,094	Exon1	170
Hep-R2	GTGAATGTGTATTGTCTATATGTCCCCATAG	chr16:50,290,203-50,290,233	Intron1
Hep-F3	AAATATCAGAGCCGAGCAGAAG	chr16:50,289,922-50,289,943	Exon1	306
Hep-R3	GTGTATTGTCTATATGTCCCCATAGG	chr16:50,290,202-50,290,227	Intron1
sgRNA1	GATCACTAATACGACTCACTAT GGAAGAGCTAAAGCGTCACAAGTTTTAGAGCTAGAAATAGC	chr16:50,289,873-50,289,892	5’ UTR	123
sgRNA-scaffold	AAAAGCACCGACTCGGTGCCACTTTTTCAAGTTGATAACGGACTAGCCTTATTTTAACTTGCTATTTCTAGCTCTAAAAC
sgRNA2	GATCACTAATACGACTCACTAT GGTACAGGATGAGCATCATGAGTTTTAGAGCTAGAAATAGC	chr16:50,290,543-50,290,562	Exon2	123

. The sgRNA were annealed with a sgRNA-scaffold containing the T7 promoter sequence and transcribed in vitro (Invitrogen, Waltham, MA, USA, AM1314). Cas9 protein was purchased from Genscript. The final concentrations of Cas9 protein and the gRNA mix were 800 ng/µL and 100 ng/µL, respectively. In total, 1 nL of the mixture was injected into each 1-cell stage embryo.

Injected embryos were raised to adulthood as F0 chimeric founders. Tail-clip genotyping confirmed indels at the target sites. Positive F0 founders were crossed with wild-type zebrafish, and the resulting F1 embryos with indels were raised as stable mutant lines. F1 adult mutants of different sexes were crossed to produce F2 embryos. Approximately 25% of F2 individuals were homozygous mutants (hep^−/−), consistent with Mendelian inheritance.

2.3. Genotyping

To genotype hepcidin knockout zebrafish, a small piece of tail tissue was clipped and digested in 50 µL of 50 mM NaOH at 95 °C for 20 min. For 48 hpf zebrafish embryos, whole embryos were digested in 10 µL of 50 mM NaOH at 95 °C for 20 min [25]. The lysates were neutralized with a 1/10 volume of 1 M Tris-HCl (pH 8.0). One microliter of lysate was used for PCR genotyping. The information of the primers is shown in Table 1.

2.4. RNA Isolation and Sequencing

Adult zebrafish tissues were dissected as previously described. Total RNA was extracted from the liver using Trizol reagent (Invitrogen, MA, USA, AM9738) following the manufacturer’s protocol. RNA quality was assessed via 1% agarose gel electrophoresis, and concentrations were measured using a spectrophotometer (Thermo Fisher, Waltham, MA, USA, Nanodrop 2000). Libraries for mRNA sequencing were prepared using the TruSeq RNA Sample Prep Kit (Illumina, San Diego, CA, USA, RS-122-2001) and sequenced on an Illumina HiSeq 2500 system (Illumina, San Diego, CA, USA).

2.5. Liver Bulk RNA Sequencing Data Processing

Samples were sequenced on the Illumina HiSeq 2500 platform (Illumina, San Diego, CA, USA). Raw sequencing data underwent quality control using FastQC (version 0.12.1), and low-quality bases were trimmed and filtered with Fastp (version 0.22.0) using the default parameters. Reads were mapped to the zebrafish (Danio rerio) genome (GRCz11, Ensembl release 112) using HISAT2 (version 2.2.1) with the default settings. Read counts were generated with featureCounts (version 2.0.3), using the parameters -p and -t exon to assign reads to exonic regions and aggregate counts by gene.

The variance stabilizing transformation (VST) expression matrix was used as input for principal component analysis (PCA). PCA was performed using the PCA function from the FactoMineR (version 2.8) package in R. Visualizations of sample clustering were generated with the fviz_pca_ind function from factoextra (version 1.0.7), plotting principal components with 95% confidence ellipses for each group.

Pearson correlation analysis was performed on the expression matrix using the cor function in R (method = “pearson”) and visualized with the pheatmap R package (version 1.0.12). DEGs were identified using DESeq2 (version 1.38.3), applying a p-value cutoff of <0.05. Genes with log₂ fold changes > 0.5 were considered upregulated, while those with log₂ fold changes < −0.5 were considered downregulated. Volcano plots were generated using ggplot2 (version 3.4.4).

2.6. Enrichment Analysis

Gene Ontology (GO) enrichment and Kyoto Encyclopedia of Genes and Genomes (KEGG) pathway analyses were conducted using the clusterProfiler R package (version 4.6.2) on the basis of hypergeometric distribution. For GO enrichment, the enrichGO function was used with the parameters OrgDb = org.Dr.eg.db, ont = “ALL”, and pAdjustMethod = “fdr”. For KEGG pathway analysis, zebrafish gene symbols were converted to Entrez IDs using the toTable function (parameter: org.Dr.egSYMBOL). Pathway enrichment analysis was performed with the enrichKEGG function (parameters: gene = ENTREZID, organism = “dre”, keyType = “kegg”, and pAdjustMethod = “fdr”). Terms or pathways with p-values < 0.05 were considered to be significantly enriched.

2.7. Visualization of Differential Gene Expression

To visualize gene expression patterns within enriched pathways, size factors for each sample were estimated using the estimateSizeFactors function in DESeq2 to normalize sequencing depth differences. Dispersion parameters for each gene were calculated using estimateDispersions, and the vst function was applied for variance stabilizing transformation (VST) to reduce technical noise.

The transformed expression matrix was extracted using the assay function, standardized using the scale function, and visualized with the ComplexHeatmap package (version 2.14.0).

2.8. Statistical Analysis

All statistical analyses were performed in GraphPad Prism 8 (GraphPad, San Diego, CA, USA). Data are presented as the mean ± SEM. Differences between groups were assessed using unpaired, two-tailed Student’s t-tests. Statistical significance was defined as p < 0.05, while p < 0.01 indicated high significance.

2.9. Western Blot Assay

Livers from 18-month-old zebrafish were collected in a lysis buffer containing 1/100 volume PMSF (Beyotime, SHA, CHN, ST506) and a 1/10 volume cocktail (Sigma, St. Louis, MO, USA, P8340), homogenized on ice for 5 min, and centrifuged at 12,000 rpm for 15 min at 4 °C. The supernatant was collected, and 4× SDS-PAGE loading buffer (Takara, Kyoto, Japan) was added before heating at 95–100 °C for 10 min.

Proteins were separated on a 10–12% SDS-PAGE gel, transferred to 0.45 µm polyvinylidene difluoride membranes (Millipore, St. Louis, MO, USA, IPVH00010), and blocked overnight in TBST containing 5% non-fat milk at 4 °C. Membranes were incubated with the following primary antibodies for 2 h at room temperature: rabbit anti-SLC40A1 (1:1000), rabbit anti-GAPDH (1:1000), rabbit anti-GPX4 (1:1000), and rabbit anti-ferritin (1:1000). After three washes with TBST, membranes were incubated with HRP-conjugated goat anti-rabbit IgG (1:4000) for 1 h at room temperature. Chemiluminescence was detected using SuperSignal™ West Pico PLUS substrate (Thermo Fisher, 34580), and images were captured on an imager (Amersham Imager 600, GE, Boston, MA, USA).

2.10. Iron Measurement by ICP-MS

Tissue iron content was measured by inductively coupled plasma mass spectrometry. Zebrafish tissues (n = 15 per sample, three replicates) were collected from adult zebrafish, weighed, and digested in 70% HNO₃ in a microwave oven for 4 h. Digested samples were analyzed using an iCAP6300 ICP-MS system (Thermo Scientific, Waltham, MA, USA, iCAP6300).

2.11. Tissue Section Preparation and Perls’ Prussian Blue Staining

Fe³⁺ levels were detected using Perls’ Prussian Blue staining. Dewaxed paraffin sections were treated with fresh a Perls’ dye solution (5% potassium ferrocyanide and 10% HCl; Sangon) for 30 min at room temperature, rinsed in deionized water, and counterstained with 0.1% Nuclear Fast Red for 5 min. Sections were rinsed and imaged using a microscope (Leica, DM300, Wetzlar, Germany).

3. Results

3.1. Zebrafish Lacking Hepcidin Exhibit Iron Overload

To investigate the in vivo role of hepcidin, we used a hepcidin knockout zebrafish line previously established by our group [24] (Figure A1a). The hepcidin mutant carries a 675 bp genomic deletion, including a 200 bp deletion in the transcription region, with 178 bp in Exon 1 and the remaining 22 bp in Exon 2. This deletion eliminates the open reading frame (ORF) in the remaining exons (Figure A1c). Sequencing of the residual hepcidin mRNA in mutants, followed by ORF finder analysis, confirmed the absence of any alternative ORF resembling the hepcidin peptide.

Stable homozygous mutants were established from F2 progeny obtained by self-crossing F1 heterozygous mutants. PCR genotyping confirmed that the genotypic ratios among F2 individuals were 28% homozygous mutants (hep^−/−), 51% heterozygotes (hep^+/−), and 21% wild-type (hep^+/+), consistent with Mendel’s expected 1:2:1 ratio (Figure A1b).

Consistent with prior studies [12], hepcidin expression is nearly restricted to the liver. To evaluate the effects of hepcidin deletion on liver physiology and iron homeostasis, we compared livers from wild-type (WT) and hep^−/− adult zebrafish. Gross morphological examination revealed that hep^−/− livers appeared pale and lacked the characteristic reddish hue of WT livers (Figure 1a). Histological analysis using Prussian Blue and Nuclear Fast Red staining further corroborated these observations. In WT livers, numerous red-stained nuclei, indicative of normal blood cells’ presence, were observed (Figure 1b). Conversely, hep^−/− livers exhibited sparse red-stained nuclei and prominent blue staining, reflecting abnormal iron deposition (Figure 1b).

Inductively coupled plasma mass spectrometry (ICP-MS) was used to measure the levels of various metals in mice [26]. Iron quantification using ICP-MS revealed significantly elevated total iron levels in hep^−/− zebrafish compared with the WT controls, particularly in the liver, skin, and egg tissues (p < 0.05) (Figure 1c). Interesting, iron increased in the eggs of adult fish, suggesting that the initial iron overload is maternally derived [24]. These findings demonstrate that hepcidin is essential for maintaining iron homeostasis in zebrafish, with its absence leading to disrupted iron storage and distribution.

3.2. RNA-Seq Reveals Ferroptosis and Key Regulatory Genes in hep^−/− Zebrafish Livers

To explore the molecular consequences of hepcidin deficiency, we performed RNA-seq on liver samples from WT and hep^−/− zebrafish. Principal component analysis (PCA) revealed distinct clustering between WT and hep^−/− groups, indicating significant transcriptional differences (Figure 1d). Pearson correlation heatmaps validated the high consistency within replicates for each genotype (Figure 1e).

Differential gene expression analysis identified 1,424 differentially expressed genes (DEGs), including 596 downregulated and 828 upregulated genes in hep^−/− livers compared with the WT (Figure 1f). Notably upregulated genes included cyp2k8, rnd2, and smad9, while downregulated genes included rnasel2, cacna1g, and slc16a6. Genes directly involved in iron regulation, such as fth1a and fthl27 [27] (Figure 1f and Figure 2c,e), as well as lipid metabolism-related genes like cidec [28] (Figure 1f), exhibited significant expression changes (Figure 2f).

Gene Ontology (GO) and KEGG pathway enrichment analyses of DEGs highlighted significant alterations in pathways associated with iron metabolism, lipid metabolism, and ferroptosis (Figure 2a,b). Notably, the “iron ion binding” pathway was highly enriched (p = 7.87 × 10⁻⁹), as was “heme binding” (p = 1.76 × 10⁻⁹), emphasizing disruptions in iron homeostasis. Additionally, “ferroptosis” (p = 6.41 × 10⁻⁴) and “fatty acid degradation” (p = 4.80 × 10⁻³) were significantly enriched, indicating lipid metabolic dysregulation.

Heatmaps of gene expression in these enriched pathways (Figure 2c–f) demonstrated distinct transcriptional profiles between WT and hep^−/− livers. Genes associated with ferroptosis, including fthl27 and sat2b (Figure 2c), were significantly upregulated in hep^−/− livers, consistent with iron overload and oxidative stress. Similarly, genes in the “iron ion binding” and “heme binding” pathways, such as fth1a (Figure 2e) and hbae5 (Figure 2f), exhibited altered expression, reflecting disruptions in iron storage and transport. In the “fatty acid degradation” pathway, key genes such as acsl4a and cyp2u1 (Figure 2d) displayed significant changes, further supporting the hypothesis of lipid metabolic dysregulation in hep^−/− zebrafish.

These findings collectively reveal that hepcidin deficiency induces extensive changes in the liver iron metabolism, lipid oxidation, and ferroptosis pathways. The observed iron overload and ferroptosis underscore the essential role of hepcidin in maintaining iron homeostasis and preventing oxidative damage.

3.3. Validation of Key Differentially Expressed Genes via Western Blotting

To validate the protein-level expression of key DEGs identified via RNA-seq, we performed Western blot analysis. The results showed that the protein levels of FPN (SLC40A1), Fth1, and GPX4 were significantly reduced in hep^−/− zebrafish compared with the WT (Figure 3). These findings align with the transcriptomic data and support the conclusion that hepcidin deficiency disrupts critical pathways in iron metabolism and ferroptosis regulation.

4. Discussion

In this study, we empolyed a hepcidin-deficient zebrafish model and demonstrated that the complete knockout of the hepcidin gene leads to iron accumulation in the liver, even when zebrafish are maintained on an iron-normal diet (Figure 1b,c). This phenotype mirrors observations in hepcidin-deficient mice [8], further supporting the conserved role of hepcidin as a key regulator of systemic iron homeostasis across vertebrates. Notably, iron accumulation was also observed in the eggs (Figure 1c), suggesting maternal transfer of iron overload to subsequent generations [24]. These findings highlight the critical role of hepcidin in preventing iron dysregulation in both somatic and reproductive tissues.

Iron metabolism involves tightly regulated processes, including absorption, storage, and export, to maintain appropriate cellular and systemic iron levels and avoid toxicity [29]. The dysregulation observed in hep^−/− zebrafish results from two key factors: increased iron absorption and decreased iron efflux. Given that the WT and hep^−/− zebrafish were fed identical iron-normal diets, differences in intestinal iron absorption are unlikely to explain the iron overload phenotype. Instead, reduced expression of FPN, the sole known iron exporter, appears to be the primary cause. Consistent with findings from Jiang et al. [30], who observed reduced FPN levels in zebrafish with blocked hepcidin splicing, our study demonstrates significantly decreased FPN expression in hep^−/− zebrafish (Figure 3). This reduction limits iron release from cells into circulation, exacerbating intracellular iron retention.

In contrast, TFR, which facilitates cellular iron uptake [31], was upregulated in hep^−/− zebrafish compared with the WT controls. The simultaneous reduction in iron export and the increase in cellular iron uptake likely amplify the iron overload phenotype. Additionally, elevated levels of the iron storage protein ferritin (fthl27) in hep^−/− zebrafish. Collectively, these findings reveal that hepcidin deficiency disrupts the balance of iron transport and storage, resulting in systemic iron overload.

Beyond iron dysregulation, our transcriptomic analysis revealed profound changes in global gene expression profiles between hep^−/− and WT zebrafish. DEGs were significantly enriched in pathways related to ferroptosis, fatty acid metabolism, iron ion binding, and heme binding (Figure 2a,b). Ferroptosis, a form of iron-dependent, non-apoptotic cell death, is characterized by lipid peroxidation driven by excessive iron [20]. Key regulators of ferroptosis, such as GPX4, play a central role in mitigating lipid peroxide accumulation [23]. While gpx4 mRNA levels were not significantly altered in hep^−/− zebrafish, protein levels of GPX4 were markedly reduced (Figure 3). This discrepancy suggests that post-transcriptional or translational mechanisms may contribute to the induction of ferroptosis under conditions of iron overload. Reduced GPX4 protein levels, combined with iron accumulation, provide strong evidence of ferroptosis in hep^−/− zebrafish livers.

The heme binding pathway was also significantly disrupted in hep^−/− zebrafish (Figure 2b). Heme, a critical molecule comprising iron complexed in a porphyrin ring, plays essential roles in oxygen transport, electron transfer, and enzymatic processes [32]. Alterations in heme binding suggest that hepcidin deficiency impairs the utilization of heme for vital biological functions, potentially contributing to systemic metabolic dysfunction. Dysregulated heme metabolism may further exacerbate oxidative stress and ferroptosis, compounding the observed liver phenotype in hep^−/− zebrafish.

In conclusion, our study highlights the indispensable role of hepcidin in regulating iron homeostasis and preventing iron-mediated toxicity in zebrafish. Hepcidin deficiency disrupts iron transport and storage, leading to systemic iron overload, lipid peroxidation, and ferroptosis in the liver. These findings not only deepen our understanding of hepcidin’s conserved function across vertebrates but also provide insights into the molecular mechanisms linking iron metabolism to ferroptosis. Further studies are needed to elucidate the downstream consequences of heme dysregulation and ferroptosis for zebrafish physiology and potential compensatory mechanisms that may mitigate iron toxicity over time. The zebrafish model provides a robust platform for studying early-stage iron toxicity and ferroptosis, which are challenging to investigate in mammalian embryos due to in utero development.

5. Conclusions

This study demonstrates that hepcidin deficiency disrupts iron homeostasis in zebrafish liver, leading to iron overload, oxidative stress, and ferroptosis. Our findings highlight the conserved role of hepcidin in regulating systemic iron metabolism across vertebrates, as similar mechanisms have been observed in mammalian models.