Therapeutic Potential of Deferiprone–Resveratrol Hybrid (DFP-RVT) Against Hepatic Iron Overload in β-Thalassemia Mice: A Proteomic Analysis

Abstract

Iron overload is a major pathological feature of β-thalassemia and a key driver of hepatic injury through oxidative stress and mitochondrial dysfunction. This study investigated the molecular effects of iron overload on liver mitochondria and evaluated the therapeutic potential of a deferiprone–resveratrol hybrid (DFP-RVT) in a β-thalassemia mouse model. Proteomic analysis was performed on liver tissues from baseline control, iron-overloaded, and DFP-RVT-treated mice to identify differentially expressed proteins and affected pathways. Iron overload resulted in marked downregulation of mitochondrial proteins, particularly components of oxidative phosphorylation and iron–sulfur cluster-associated pathways, including frataxin. In contrast, DFP-RVT treatment restored the expression of multiple mitochondrial proteins involved in respiratory chain function and energy metabolism. Comparative proteomic profiling revealed opposing regulation patterns between iron-overloaded and DFP-RVT-treated groups, indicating recovery of mitochondrial integrity following iron chelation therapy. These findings suggest that iron-induced hepatic injury in β-thalassemia is closely linked to mitochondrial protein dysregulation and that DFP-RVT may mitigate this process by restoring mitochondrial protein expression and iron homeostasis. This study provides mechanistic insight into iron-mediated mitochondrial dysfunction and supports the therapeutic potential of DFP-RVT for iron overload-associated liver injury.

1. Introduction

Thalassemia diseases are a heterogeneous group of anemias caused by genetic defects in globin genes. The major forms are alpha (α)- and beta (β)-thalassemia, which are classified according to the affected globin chain [1]. β-thalassemia is a genetic defect of one or two beta-globin genes located on chromosome 11, leading to either a partial reduction (β⁺-thalassemia) or complete absence (β⁰-thalassemia) of β-globin chains’ synthesis [2,3,4,5]. Clinically, β-thalassemia presents as a transfusion-dependent thalassemia (TDT), also known as β-thalassemia major (TM), or as a milder form, β-thalassemia intermedia (TI), which is typically non-transfusion-dependent (NTDT) [6].

Iron overload is a major complication in both TDT and NTDT. In TDT patients, iron overload primarily results from repeated blood transfusion to alleviate chronic anemia [4]. In NTDT patients, iron overload occurs mainly due to increased gastrointestinal iron absorption driven by ineffective erythropoiesis [7]. The resulting ineffective erythropoiesis produces elevated serum erythropoietin (EPO) levels, accompanied by suppression of hepcidin, the major regulatory of systemic iron homeostasis. This dysregulation enhances intestinal iron absorption and ultimately results in progressive iron overload [7,8]. The excess iron deposits in multiple organs, particularly the liver and heart, leading to significant morbidity. Interestingly, in NTDT patients, iron overload predominantly affects hepatocytes rather than cardiomyocytes [9,10].

The liver plays a central role in iron metabolism through the synthesis of plasma proteins such as transferrin, ceruloplasmin, haptoglobin, hemopexin, and hepcidin. Hepatocytes, the prominent cell type in the liver, serve as a major site of iron storage and possess a high capacity to synthesize ferritin, the major iron storage protein [11]. Chronic liver iron overload can progress to liver fibrosis, liver cirrhosis and hepatocellular carcinoma [12].

The deferiprone–resveratrol hybrid (DFP-RVT) was developed by Dr. Yongmin Ma and colleagues by integrating the potent iron-chelating properties of deferiprone (DFP) with the antioxidant scaffold of resveratrol. In this hybrid compound, one of the benzene rings of resveratrol, an antioxidant and investigational agent for neurodegenerative diseases, is replaced by the metal-chelating moiety of deferiprone. This novel synthetic molecule, named DFP-RVT (2-(3,5-dihydroxystyryl)-5-hydroxy-1-methylpyridin-4(1H)-one), has a molecular weight of 340 and the chemical formula C₁₄H₁₄BrNO₄. The design, synthesis, and biological evaluation of a series of deferiprone–resveratrol hybrids were first reported by Xu P et al. [13]. The synthesis of DFP-RVT aimed to combine iron-chelating and antioxidant properties within a single molecule. Previous studies have demonstrated that DFP-RVT is more lipophilic and more effective than DFP alone in reducing the labile iron pool (LIP) in red blood cells infected with Plasmodium falciparum [14]. Furthermore, DFP-RVT exhibited stronger inhibitory effects on lipid peroxidation, enhanced antioxidant capacity, and superior hepatoprotective effects. Compared with DFP, DFP-RVT also more effectively improved anemia and reduced liver damage in mice infected with Plasmodium berghei [15,16].

Recently, a study from Li J et al. (2024) reported that DFP-RVT exerts multifaceted hepatoprotective effects in iron-overloaded β-thalassemia mice, including reductions in hepatic iron content, improvements in liver enzyme levels, preservation of hepatic architecture, attenuation of oxidative stress, restoration of antioxidant defenses, and suppression of inflammatory responses [17]. These findings highlight the potential of DFP-RVT as a therapeutic agent for managing liver damage associated with iron overload. However, the molecular mechanisms underlying these protective effects, as well as the broader biological responses to DFP-RVT treatment, remain incompletely understood.

To further elucidate the mechanisms of action of DFP-RVT in iron-overloaded β-thalassemia, this study employed a liver proteomic approach to comprehensively characterize protein expression changes and biological pathways modulated by DFP-RVT treatment. Unlike previous investigations that primarily focused on biochemical and histopathological outcomes, the present study provides a systems-level analysis of molecular responses associated with DFP-RVT therapy. By integrating proteomic profiling with pathway enrichment analysis, this work aims to identify key regulatory networks involved in hepatoprotection, with particular emphasis on mitochondrial function, oxidative stress, and iron homeostasis. To our knowledge, this is the first study to apply discovery-based liver proteomics to investigate the mechanistic effects of the DFP-RVT in an iron-overloaded β-thalassemia model. These findings provide novel insight into mitochondrial-targeted mechanisms of iron detoxification and broaden the current understanding of therapeutic strategies for iron overload-associated liver injury.

2. Materials and Methods

2.1. Animal Care

Male heterozygous β-globin knockout (BKO, Hbb^th3/+) C57BL/6 mice, weighing approximately 28 g and aged 2–3 months, were used in this experiment. All animal procedures were conducted in accordance with the guidelines and regulations. All animal protocols were approved by the Animal Ethical Committee of the Faculty of Medicine, Chiang Mai University, Thailand (ethics approval number 20/2567). Additionally, this study was designed and reported following the ARRIVE guidelines to ensure ethical standards in animal research. The mice were acclimated to a controlled environment with a temperature of 20–22 °C, relative humidity of 50 ± 10%, and a 12 h light/dark cycle. Throughout the acclimatization period, they had unrestricted access to food and water. A one-week housing period was provided prior to the initiation of the experiment.

2.2. Iron Loading and Intervention

β-thalassemia heterozygous knockout (BKO; Hbb^th3/+) mice (n = 12) were intraperitoneally (i.p.) injected with iron dextran at a dose of 10 mg/day for 20 consecutive days to induce hepatic iron overload. In parallel, age-matched BKO mice (n = 6) received intraperitoneal injections of 0.85% normal saline solution (NSS) to serve as non-iron-loaded controls. All animals were then maintained without treatment for a 30-day equilibration period to allow stabilization of the iron burden.

Following this period, the iron-loaded mice were randomly assigned into two groups (n = 6 per group) and orally administered (p.o.) either deionized water (DI) or a synthesized deferiprone–resveratrol hybrid compound (DFP-RVT; 122 mg/kg/day), generously provided by Yongmin Ma, School of Pharmaceutical and Chemical Engineering, Taizhou University, China. The non-iron-loaded control group also received oral DI for the same duration. Treatment continued daily for 60 days.

At the end of the treatment period, mice were euthanized via cervical dislocation without prior anesthesia. Livers were collected, weighed, and processed for histological and biochemical analysis. Samples were snap-frozen in liquid nitrogen and stored at −80 °C for proteomic evaluation.

The experimental design and group allocation are depicted in Figure 1, and key biochemical characteristics, including the liver iron content (LIC), serum iron levels, total iron-binding capacity (TIBC), and liver enzymes (AST, ALT, and ALP), are summarized in

Table 1. Biochemical parameters reflecting iron overload and liver injury in β-thalassemia mice.

	LIC (mg/Dry Weight)	Serum Iron (µg/dL)	TIBC (µg/dL)	AST (U/L)	ALT (U/L)	ALP (U/L)
BKO (n = 6)	1.10 ± 0.09	158.78 ± 31.29	342.74 ± 54.43	242.57 ± 49.85	42.43 ± 8.07	60.00 ± 4.79
IO (n = 6)	4.21 ± 0.14	1055.45 ± 163.23	1067.45 ± 168.24	528.60 ± 104.02	168.80 ± 19.86	89.80 ± 3.54
DR-IO (n = 6)	2.10 ± 0.18	674.61 ± 178.28	832.33 ± 88.24	309.50 ± 21.98	123.50 ± 11.42	81.33 ± 5.49

Abbreviation: BKO, β-thalassemia control mice (Hbb^th3/+, non-iron-loaded); IO, iron-overloaded β-thalassemia mice; DR-IO, iron-overloaded β-thalassemia mice treated with deferiprone–resveratrol hybrid (DFP-RVT). Data are presented as mean ± SEM. These data were previously reported by Li J et al. (2024) [17] and are included here for reference and continuity to support downstream proteomic analysis.

. These biochemical findings confirmed the successful establishment of hepatic iron overload and its partial mitigation by DFP-RVT treatment.

Notably, the liver tissues used for proteomic analysis in this study were obtained from our previous work by Li J et al. (2024) [17]. The associated biochemical data from that study are referenced here to confirm successful induction of hepatic iron overload and its partial mitigation by DFP-RVT treatment, thereby validating the suitability of these samples for subsequent proteomic investigation.

2.3. Sample Preparation for Proteomic Analysis

The mice livers were dissected into small pieces and incubated with the radioimmunoprecipitation assay (RIPA) lysis buffer (Thermo Scientific, Waltham, MA, USA) containing 1× protease inhibitor cocktail on ice for 30 min. The liver tissues were ground using tissue grinders and sonicated on ice for 15 s. The protein extracted was collected by centrifugation at 15,000× g at 4 °C for 10 min. Protein extracts were precipitated by pre-chilled acetone at −20 °C for 16 h, followed by centrifugation at 15,000× g for 10 min. The protein pellet was resuspended in 0.1% RapidGest SF (Waters, Milford, MA, USA) in 10 mM ammonium bicarbonate. The protein concentration was determined using the Bradford Reagent assay kit (Thermo Scientific, Waltham, MA, USA) with bovine serum albumin (Thermo Scientific, Waltham, MA, USA) as the standard.

2.4. Protein Digestion

A total of 30 µg of protein was subjected to reduction in the sulfhydryl bonds in 10 mM dithiothreitol in 10 mM ammonium bicarbonate at 65 °C for 20 min and alkylation of the sulfhydryl groups with iodoacetamide at room temperature for 30 min. Alkylated proteins were cleaned using the Zeba Spin Desalting Columns, 7K MWCO, 0.5 mL (Thermo Scientific, Waltham, MA, USA). The flow-through was collected, followed by enzymatic digestion by trypsin protease, MS Grade (Thermo Scientific, Waltham, MA, USA), at a ratio of 1:50 (enzyme:protein), and incubated at 37 °C for 16 h. The peptide mixture was dried using a centrifugal concentrator (TOMY, Tokyo, Japan), and the dried peptides were reconstituted in 0.1% formic acid and transferred to a total recovery LCMS vial (Waters, Milford, MA, USA). Quality control for sample preparation, digestion, and clean-up was done using 0.2 mg of bovine serum albumin (n = 2) for evaluating digestion efficiency.

2.5. Tandem Mass Spectrometry Analysis

Tryptic peptides were analyzed using an Orbitrap HF hybrid mass spectrometer coupled with an UltiMate 3000 LC system. Any residue salts in peptide samples were removed using a reverse-phase C18 PepMap 100 trapping column (Thermo Scientific, Vilnius, Lithuania) and the separation was performed by a C18 PepMap 100 capillary column (Thermo Scientific, Vilnius, Lithuania). The solution of 0.1% formic acid was used to reconstitute dried peptides. The protonated peptides were introduced into the nanoLC system in the amount of 1.2 µg. The mobile phases were 0.1% formic acid in water (Phase A) and 95% acetonitrile with 0.1% formic acid (Phase B). Mass spectra were acquired in data-dependent acquisition mode, with full scans covering a mass range of 400–1600 m/z. The 15 most abundant peptide ions with charge states between 2 and 5 were subjected to fragmentation. Dynamic exclusion was conditioned for 18 s. Full scan mass spectra were recorded from m/z 400 to 1600, with an AGC target of 3 × 10⁶ ions and a resolution of 120,000. MS/MS scans were triggered once the AGC target reached 10⁵ ions, with dynamic exclusion applied for a 15 s window. Ion selection occurred within 12 s of the dynamic exclusion window. Raw MS/MS spectra were analyzed using the Proteome Discoverer software version 2.4 (Thermo Scientific, Waltham, MA, USA), which included the SEQUEST, Percolator, and Minora algorithms. Peptide identification was searched against the Mus musculus UniProtKB database. The search parameters specified a maximum of two missed tryptic cleavages, a precursor mass tolerance of 10 ppm, and a fragment mass tolerance of 0.1 Da. Carbamidomethylation of cysteine was designated as a fixed modification, with methionine oxidation designated as a variable modification. A false discovery rate (FDR) of 0.01 was used at both the peptide and protein levels. Relative protein quantification was normalized using the total peptide amount.

2.6. Bioinformatics Analysis

Following proteomic analysis, the statistically significant differential expression proteins were subjected to further bioinformatics analysis for functional annotation and classification. These analyses aimed to elucidate the biological roles and mechanisms of the identified proteins. The bioinformatics tools employed included Blast2GO v6.0.5 [18], which was used to categorize proteins into Gene Ontology (GO) terms, including biological process, molecular function, and cellular component. Additionally, the STRING database was utilized to explore and visualize protein–protein interaction networks and to identify interaction clusters relevant to the biological context of this study [19].

3. Results

3.1. Proteomic Analysis

3.1.1. Iron-Overloaded Beta-Thalassemia Mice (IO) Proteomic Analysis

The liver protein expression profiles of iron-overloaded β-thalassemia mice (IO) were compared with those of non-iron-overloaded β-thalassemia control mice (BKO). Differential protein expression analysis identified 199 significantly upregulated proteins and 135 significantly downregulated in the IO group relative to BKO (IO/BKO) (Figure 2).

Gene Ontology (GO) classification at level 3 was performed to categorize the differentially expressed proteins according to the biological process (BP), molecular function (MF) and cell component (CC). GO analysis revealed that the majority of both up- and downregulated proteins were associated with the metabolic process (BP), protein binding (MF), and intracellular anatomical structure (CC) (Figure 3A–C).

Functional annotation of 199 upregulated proteins indicated the enrichment of proteins involved in organ dysfunction, particularly those related to pancreatic secretion, inflammation, liver fibrosis and cell death (

Table 2. Highlighted differentially expressed liver proteins between iron-overloaded β-thalassemia mice (IO) and non-iron-overloaded β-thalassemia controls (BKO).

Accession	Description	Gene Symbol	Amino Acid	MW [kDa]	Unique Peptides	Fold Change	Adjust p-Value
Organ dysfunction/Pancreatic secretion
Q9R0T7	Trypsin-4	Try4	245	26.3	1	100	<0.001
Q80W15	Insulin-like growth factor-binding protein-like 1	Igfbpl1	270	28.6	1	100	<0.001
P05208	Chymotrypsin-like elastase family member 2A	Cela2a	271	28.9	5	20.54	<0.001
Q9CR35	Chymotrypsinogen B	Ctrb1	263	27.8	8	17.09	0.001
Q6P8U6	Pancreatic triacylglycerol lipase	Pnlip	465	51.4	8	10.29	0.015
P00688	Pancreatic alpha-amylase 2a5	Amy2a5	508	57.3	20	6.52	0.0225
Liver fibrosis
Q6IFX2	Keratin, type I cytoskeletal 42	Krt42	452	50.1	1	100	<0.001
Q8C6K9	Collagen alpha-6(VI) chain	Col6a6	2265	246.2	2	7.28	0.042
Immune response/Inflammation
P00683	Ribonuclease pancreatic	Rnase1	149	16.8	1	100	<0.001
Q08481	Platelet endothelial cell adhesion molecule	Pecam1	727	81.2	1	100	<0.001
O88536	Formyl peptide receptor 2	Fpr2	351	39.4	1	100	<0.001
P06342	H-2 class II histocompatibility antigen, A-Q beta chain	H2-Ab1	265	29.9	1	100	<0.001
Q9WTU6	Mitogen-activated protein kinase 9	Mapk9	423	48.2	1	100	<0.001
Q2VLH6	Scavenger receptor cysteine-rich type 1 protein M130	Cd163	1121	120.8	1	12.09	0.005
P14106	Complement C1q subcomponent subunit B	C1qb	253	26.7	1	9.81	0.018
Cell death/Apoptosis
Q921F2	TAR DNA-binding protein 43	Tardbp	414	44.5	1	100	<0.001
Q07813	Apoptosis regulator BAX	Bax	192	21.4	1	15.22	0.001
O35598	Disintegrin and metalloproteinase domain-containing protein 10	Adam10	749	83.9	1	14.24	0.001
A2A935	Histone-lysine N-methyltransferase PRDM16	Prdm16	1275	140.8	1	10.13	0.025
Q63844	Mitogen-activated protein kinase 3	Mapk3	380	43	2	9.15	0.014
Mitochondrial component: Oxidative phosphorylation/ATP biosynthesis
O35943	Frataxin, mitochondrial	Fxn	207	22.9	1	0.01	<0.001
Q9CQH3	NADH dehydrogenase [ubiquinone] 1 beta subcomplex subunit 5, mitochondrial	Ndufb5	189	21.7	1	0.01	<0.001
Q9Z1P6	NADH dehydrogenase [ubiquinone] 1 alpha subcomplex subunit 7	Ndufa7	113	12.6	2	0.01	<0.001
Q9D6J5	NADH dehydrogenase [ubiquinone] 1 beta subcomplex subunit 8, mitochondrial	Ndufb8	185	21.9	2	0.01	<0.001
P43024	Cytochrome c oxidase subunit 6A1, mitochondrial	Cox6a1	111	12.3	1	0.01	<0.001
Mitochondrial component: Transit peptide/Transport
Q9D880	Mitochondrial import inner membrane translocase subunit TIM50	Timm50	353	39.8	1	0.01	<0.001
Q9WVM8	Kynurenine/alpha-aminoadipate aminotransferase, mitochondrial	Aadat	425	47.6	1	0.01	<0.001
Q9D023	Mitochondrial pyruvate carrier 2	Mpc2	127	14.3	1	0.01	<0.001
Q924X2	Carnitine O-palmitoyltransferase 1, muscle isoform	Cpt1b	772	88.2	1	0.01	<0.001
Coagulation factor
O88947	Coagulation factor X	F10	481	54	1	0.01	<0.001
Protein translation
Q9D6S7	Ribosome-recycling factor, mitochondrial	Mrrf	262	29	1	0.01	<0.001

Fold change values < 1.00 indicate downregulation, whereas values > 1.00 indicate upregulation. Statistical significance was assessed using Benjamini–Hochberg-adjusted p-values, with an adjusted p-value < 0.05 considered statistically significant.

). In contrast, among the 135 downregulated protein, 35 proteins (approximately 25%) were identified as mitochondrial components, including proteins essential for oxidative phosphorylation, ATP synthesis, and mitochondrial protein transport. In addition, blood coagulation factor X was significantly downregulated in IO mice compared with the BKO control (Table 2).

These proteomic alterations suggest that hepatic iron overload in β-thalassemia is associated with enhanced inflammatory and injury-related responses, accompanied by the suppression of mitochondrial energy metabolism and coagulation-related pathways.

3.1.2. DFP-RVT Treatment in IO Mice (DR-IO) Proteomic Analysis

To evaluate the molecular effect of DFP-RVT treatment on iron-overloaded β-thalassemia mice, hepatic proteomic profiles of DFP-RVT treated mice (DR-IO) were compared with untreated iron-overloaded mice (IO). Differential expression analysis identified 223 significantly upregulated proteins and 173 significantly downregulated proteins in the DR-IO group relative to the IO group (DR-IO/IO) (Figure 4).

GO classification at level 3 was performed to categorize the differentially expressed proteins according to BP, MF and CC. GO analysis revealed that both up- and downregulated proteins were predominantly associated with the metabolic process (BP), protein binding (MF), and intracellular anatomical structure (CC) (Figure 5A–C).

A total of 223 upregulated proteins were identified, comprising proteins that are identified as mitochondrial components (43 proteins) and translational pathway proteins (15 proteins), including large ribosomal subunit protein, eukaryotic initiation factor 4A-II and ribosome-recycling factor (

Table 3. Highlighted differentially expressed liver proteins in DFP-RVT-treated iron-overloaded β-thalassemia mice compared with untreated iron-overloaded mice (DR-IO vs. IO).

Accession	Description	Gene Symbol	Amino Acid	MW [kDa]	Unique Peptides	Fold Change	Adjust p-Value
Mitochondrial component: Oxidative phosphorylation/ATP biosynthesis
O35943	Frataxin, mitochondrial	Fxn	207	22.9	1	100	<0.001
Q9D6J5	NADH dehydrogenase [ubiquinone] 1 beta subcomplex subunit 8, mitochondrial	Ndufb8	185	21.9	2	100	<0.001
Q9CQH3	NADH dehydrogenase [ubiquinone] 1 beta subcomplex subunit 5, mitochondrial	Ndufb5	189	21.7	1	100	<0.001
P43024	Cytochrome c oxidase subunit 6A1, mitochondrial	Cox6a1	111	12.3	1	100	<0.001
Q9QX60	Deoxyguanosine kinase, mitochondrial	Dguok	277	32.3	1	5.83	0.002
Mitochondrial component: Transit peptide/Transport
Q9D880	Mitochondrial import inner membrane translocase subunit TIM50	Timm50	353	39.8	1	100	<0.001
Q9WVM8	Kynurenine/alpha-aminoadipate aminotransferase, mitochondrial	Aadat	425	47.6	1	100	<0.001
Q9D023	Mitochondrial pyruvate carrier 2	Mpc2	127	14.3	1	100	<0.001
Q924X2	Carnitine O-palmitoyltransferase 1, muscle isoform	Cpt1b	772	88.2	1	100	<0.001
Protein translation
Q9D6S7	Ribosome-recycling factor, mitochondrial	Mrrf	262	29	1	100	<0.001
Q9CZM2	Large ribosomal subunit protein eL15	Rpl15	204	24.1	2	5.36	0.014
P10630	Eukaryotic initiation factor 4A-II	Eif4a2	407	46.4	1	4.59	0.039
P62245	Small ribosomal subunit protein uS8	Rps15a	130	14.8	3	4.39	<0.001
P62908	Small ribosomal subunit protein (40S) uS3	Rps3	243	26.7	11	2.26	0.012
Coagulation factor
O88947	Coagulation factor X	F10	481	54	1	100	<0.001
Organ dysfunction/Pancreatic secretion
Q9R0T7	Trypsin-4	Try4	245	26.3	1	0.01	<0.001
Q80W15	Insulin-like growth factor-binding protein-like 1	Igfbpl1	270	28.6	1	0.01	<0.001
P07146	Anionic trypsin-2	Prss2	246	26.2	2	0.03	<0.001
Q9CR35	Chymotrypsinogen B	Ctrb1	263	27.8	8	0.15	<0.001
Q9CQC2	Colipase	Clps	113	12.4	1	0.15	0.003
Q6P8U6	Pancreatic triacylglycerol lipase	Pnlip	465	51.4	8	0.19	0.001
P00688	Pancreatic alpha-amylase 2a5	Amy2a5	508	57.3	20	0.21	<0.001
Q9QUK9	Trypsin-5	Try5	246	26.3	1	0.35	0.003
Inflammation
P00683	Ribonuclease pancreatic	Rnase1	149	16.8	1	0.01	<0.001
P06342	H-2 class II histocompatibility antigen, A-Q beta chain	H2-Ab1	265	29.9	1	0.01	<0.001
O88536	Formyl peptide receptor 2	Fpr2	351	39.4	1	0.01	<0.001
Q2VLH6	Scavenger receptor cysteine-rich type 1 protein M130	Cd163	1121	120.8	1	0.01	<0.001
Cell death/Apoptosis
O35598	Disintegrin and metalloproteinase domain-containing protein 10	Adam10	749	83.9	1	0.01	<0.001
Q921F2	TAR DNA-binding protein 43	Tardbp	414	44.5	1	0.01	<0.001
Liver fibrosis
Q6IFX2	Keratin, type I cytoskeletal 42	Krt42	452	50.1	1	0.01	<0.001

Fold change values < 1.00 indicate downregulation, whereas values > 1.00 indicate upregulation. Statistical significance was assessed using Benjamini–Hochberg-adjusted p-values, with an adjusted p-value < 0.05 considered statistically significant.

). Interestingly, several mitochondrial proteins involved in oxidative phosphorylation and ATP synthesis, which were downregulated in iron-overloaded mice compared with controls (IO/BKO), were restored following DFP-RVT treatment.

In contrast, a total of 173 downregulated proteins in the DR-IO group were associated with pancreatic secretion, inflammation, cell death, and liver fibrosis, indicating attenuation of iron-induced pathological responses after DFP-RVT administration (Table 3).

3.2. Up- and Downregulation Protein Characterization

To demonstrate the specificity of proteins that are expressed similarly or differently when comparing different experimental groups, Venn diagrams were created to visualize shared and unique proteins among the comparisons of IO/BKO and DR-IO/IO (Figure 6). Among the upregulated proteins, 22 were found to be commonly elevated in both IO/BKO and DR-IO/IO comparisons (Figure 6A), suggesting persistent activation despite treatment. In contrast, seven proteins were downregulated across both comparisons (Figure 6B), indicating shared suppression under iron overload and after chelation therapy.

To identify proteins restored by DFP-RVT treatment, we further analyzed the intersection between proteins upregulated in IO/BKO and those downregulated in DR-IO/IO (Figure 7A). The results revealed that there are 45 upregulated proteins in IO/BKO that are downregulated in DR-IO/IO (Figure 7A). These include proteins involved in pancreatic secretion and cell wall disruption, such as Trypsin-4 (Q9R0T7), Chymotrypsinogen B (Q9CR35), Pancreatic triacylglycerol lipase (Q6P8U6), Pancreatic alpha-amylase 2α5 (P00688) and Insulin-like growth factor-binding protein-like 1 (Q80W15). Additionally, proteins associated with immune response and inflammation were significantly downregulated following the treatment, including Ribonuclease pancreatic (P00683), Formyl peptide receptor 2 (O88536), H-2 class II histocompatibility antigen, A-Q beta chain (P06342) and Scavenger receptor cysteine-rich type 1 protein M130 (Q2VLH6). Proteins involved in cell death and fibrotic process, such as TAR DNA-binding protein 43 (Q921F2) and Disintegrin and metalloproteinase domain-containing protein 10 (O35598), were also reversed (Figure 8A,

Table 4. Differentially expressed proteins shared between IO/BKO and DR-IO/IO comparisons.

Accession	Gene Symbol	Description	Fold Change IO/BKO	Fold Change DR-IO/IO
Organ dysfunction/Pancreatic secretion
Q9R0T7	Try4	Trypsin-4	100	0.01
Q80W15	Igfbpl1	Insulin-like growth factor-binding protein-like 1	100	0.01
Q9CR35	Ctrb1	Chymotrypsinogen B	17.09	0.15
Q6P8U6	Pnlip	Pancreatic triacylglycerol lipase	10.29	0.19
P00688	Amy2a5	Pancreatic alpha-amylase 2α5	6.52	0.21
Liver fibrosis
Q6IFX2	Krt42	Keratin, type I cytoskeletal 42	100	0.01
Immune response/Inflammation
P00683	Rnase1	Ribonuclease pancreatic	100	0.01
O88536	Fpr2	Formyl peptide receptor 2	100	0.01
P06342	H2-Ab1	H-2 class II histocompatibility antigen, A-Q beta chain	100	0.01
Q2VLH6	Cd163	Scavenger receptor cysteine-rich type 1 protein M130	12.08	0.01
Cell death/Apoptosis
Q921F2	Tardbp	TAR DNA-binding protein 43	100	0.01
O35598	Adam10	Disintegrin and metalloproteinase domain-containing protein 10	14.24	0.01
Mitochondrial component: Oxidative phosphorylation/ATP biosynthesis
O35943	Fxn	Frataxin, mitochondrial	0.01	100
Q9CQH3	Ndufb5	NADH dehydrogenase 1 beta subcomplex subunit 5, mitochondrial	0.01	100
Q9D6J5	Ndufb8	NADH dehydrogenase 1 beta subcomplex subunit 8, mitochondrial	0.01	100
P43024	Cox6a1	Cytochrome c oxidase subunit 6A1, mitochondrial	0.01	100
Q9Z1P6	Ndufa7	NADH dehydrogenase 1 alpha subcomplex subunit 7	0.01	2.034
Mitochondrial component: Transit peptide/Transport
Q9D880	Timm50	Mitochondrial import inner membrane translocase subunit TIM50	0.01	100
Q9WVM8	Aadat	Kynurenine/alpha-aminoadipate aminotransferase, mitochondrial	0.01	100
Q9D023	Mpc2	Mitochondrial pyruvate carrier 2	0.01	100
Q924X2	Cpt1b	Carnitine O-palmitoyltransferase 1	0.01	100
Coagulation factor
O88947	F10	Coagulation factor X	0.01	100

Fold change values < 1.0 indicate downregulation, whereas values > 1.0 indicate upregulation. All reported fold changes are statistically significant (p < 0.05).

). These findings suggested that DFP-RVT treatment reduces the inflammation, preserves the hepatic architecture, improves cell membrane damage, and prevents the development of fibrosis in hepatic iron-overloaded thalassemia mice.

Furthermore, analysis of proteins downregulated in IO/BKO and subsequently upregulated in DR-IO/IO group revealed 92 restored proteins (Figure 7B). Out of these, 22 proteins (24%) were mitochondrial components. Notably, key enzymes involved in oxidative phosphorylation and ATP synthesis were restored by DFP-RVT, including Frataxin (O35943), NADH dehydrogenase 1 beta subcomplex subunit 5 (Q9CQH3), NADH dehydrogenase 1 alpha subcomplex subunit 7 (Q9Z1P6), NADH dehydrogenase 1 beta subcomplex subunit 8 (Q9D6J5), and Cytochrome c oxidase subunit 6A1 (P43024). Additionally, mitochondria transporter proteins, such as Mitochondrial import inner membrane translocase subunit TIM50 (Q9D880), Kynurenine/alpha-aminoadipate aminotransferase (Q9WVM8), Mitochondrial pyruvate carrier 2 (Q9D023) and Carnitine O-palmitoyltransferase (Q924X2), were also significantly upregulated. Notably, the coagulation factor X (O88947) previously downregulated in IO was restored in DR-IO (Figure 8B, Table 4).

Taken together, these findings indicate that DFP-RVT improved organelle damage, especially mitochondria that are essential for energy production, heme synthesis, and iron–sulfur cluster assembly. Protecting mitochondrial function is critical for maintaining cellular metabolism and iron homeostasis, especially under chronic iron overload conditions.

3.3. DFP–RVT Restores Mitochondrial Function via Regulation of Iron–Sulfur Cluster Biogenesis in Thalassemic Iron Overload

DFP-RVT, a compound with both iron-chelating and antioxidant properties, has demonstrated efficacy in reducing the iron burden and protecting hepatocytes from iron-induced toxicity. In the present study, proteomics analysis of liver tissue revealed that iron overload (IO) mice showed predominant downregulation of mitochondrial proteins compared with baseline knockout (BKO) mice. In contrast, DFP-RVT-treated IO mice (DR-IO) showed a marked upregulation of several mitochondria proteins relative to untreated IO mice.

These altered mitochondrial proteins included the protein functioning in the oxidative phosphorylation system, particularly subunits of complex I (NADH dehydrogenase 1 subcomplex subunit 5, 7, 8) and complex IV (Cytochrome c oxidase subunit 6A1). These complexes are essential for mitochondrial electron transport and ATP synthesis and are closely associated with iron-containing molecules. The downregulation of these proteins in IO mice suggests impaired mitochondrial bioenergetics under iron-overloaded conditions, while their restoration following DFP-RVT treatment indicates a recovery of mitochondrial function. In addition, frataxin, a pivotal mitochondrial protein involved in cellular and mitochondrial iron metabolism, was significantly downregulated in IO (Table 3) and restored following DFP-RVT treatment. Frataxin plays a crucial role in the iron sulfur cluster (ISC) and heme synthesis. Reduced frataxin expression is known to disrupt mitochondrial iron homeostasis and compromise respiratory chain function [20].

Taken together, these findings suggest that DFP-RVT treatment alleviates iron-induced mitochondrial dysfunction by restoring the expression of proteins involved in oxidative phosphorylation and iron–sulfur cluster metabolism. The recovery of frataxin and mitochondrial respiratory components supports a protective role of DFP-RVT in preserving mitochondrial integrity and improving hepatic energy metabolism under iron-overloaded thalassemic conditions.

4. Discussion

In β-thalassemia disease, excessive iron accumulation results in an expanded labile iron acts as a catalyst for the Fenton reaction, leading to the generation of reactive oxygen species (ROS). This iron-driven oxidative stress causes widespread cellular injury, including lipid peroxidation, protein oxidation, and DNA damage. Accumulation of ROS promotes caspase activation and apoptotic signaling, while also inducing structural and functional damage to organelles, particularly mitochondria and lysosomes. Lipid peroxidation further compromises membrane integrity, ultimately converging on hepatocellular death and progressive tissue injury [7,12,21,22,23,24]. In parallel, lipid peroxidation products activate transforming growth factor-beta 1 (TGF-β1), stimulating collagen deposition and fibrogenesis, which contributes to chronic liver dysfunction in iron-overloaded conditions [25,26]. Accordingly, iron chelation remains a cornerstone therapeutic strategy to reduce labile iron availability and attenuate ROS-mediated toxicity [27,28,29,30].

Previous studies indicate that DFP-RVT exerts multiple beneficial biological effects, primarily attributed to its iron-chelating and antioxidant properties [15,16]. Its mechanisms of action are centered around mitigating iron overload-induced oxidative stress and restoring the antioxidant defense system [15,16,31]. DFP-RVT effectively chelates redox-active iron (Fe(II) and Fe(III)), thereby reducing intracellular iron levels, particularly non-heme iron and the labile iron pool (LIP). In addition, DFP-RVT exhibits strong radical-scavenging activity against ABTS+ and DPPH radicals, with greater antioxidant capacity than deferiprone alone [31]. The hybrid design confers a synergistic effect by combining metal chelation with direct radical scavenging, enhancing the antioxidant performance of DFP, although it does not fully match the potency of native resveratrol. Collectively, these findings indicate that molecular conjugation improves the functional antioxidant capacity of DFP, supporting the therapeutic potential of DFP-RVT as a dual-acting agent targeting both the iron burden and oxidative stress [31].

Previous research by Li J et al. (2024) demonstrated the antioxidant and iron-chelating efficacy of DFP and DFP-RVT in models of iron-overloaded β-thalassemia mice [17]. However, the molecular mechanisms underlying the hepatoprotective effects of DFP-RVT remained incompletely defined. In this study, we employed a proteomic approach to characterize alterations in liver protein expression associated with iron overload and their modulation following DFP-RVT treatment, with a particular focus on pathways related to mitochondrial function, inflammation, and cellular injury.

Proteomic alterations associated with hepatic iron overload have previously been reported. Petrák J et al. (2007) demonstrated that iron accumulation in the murine liver profoundly disrupts metabolic pathways, including the urea cycle, fatty acid oxidation, and methylation processes, reflecting both compensatory responses and pathological remodeling of hepatic metabolism [32]. More recently, Rana N.K. et al. (2025) reported distinct liver proteomic signatures between Berkeley sickle cell disease (Berk-SS) and β-thalassemia (Hbb^th3/+) mice, highlighting disease-specific inflammatory and metabolic adaptations [33]. These studies underscore the utility of proteomics in uncovering systemic responses to hemoglobinopathies and iron overload.

In this study, proteomic analysis of liver tissue revealed that iron overload (IO) in β-thalassemia mice led to a marked downregulation of mitochondrial proteins, accompanied by increased expression of proteins involved in pancreatic secretion, inflammation, liver fibrosis and cell death. These findings are consistent with previous studies indicating that mitochondria dysfunction in thalassemia arises from heightened energy demands of red blood cells, oxidative stress from iron overload, and the effects of abnormal hemoglobin chain synthesis [34,35,36,37]. In contrast, DFP-RVT treatment resulted in the restoration of mitochondrial protein expression, along with an increased abundance of proteins involved in translation and mitochondrial maintenance, suggesting a recovery of mitochondrial integrity and metabolic capacity.

Mitochondria are central regulators of cellular energy production, redox balance, apoptosis, and iron metabolism, including heme synthesis and iron–sulfur (Fe-S) cluster assembly [38,39]. Among the mitochondrial proteins altered in this study, frataxin emerged as a key regulatory node. Frataxin is essential for mitochondrial iron homeostasis and plays a central role in the biogenesis of iron-sulfur (Fe-S) clusters and heme, both critical for mitochondrial enzyme function [20]. By regulating mitochondrial iron levels, frataxin helps prevent iron accumulation and subsequent oxidative damage. Its involvement in Fe-S cluster assembly also indirectly supports the function of respiratory complexes, limiting the generation of reactive oxygen species (ROS) and preserving mitochondrial health [40,41]. Loss of frataxin disrupts mitochondrial iron distribution, leading to impaired ATP production and elevated ROS generation. Such deficits are known to underline the pathophysiology of Friedreich’s ataxia (FRDA) [42,43], an autosomal recessive neurodegenerative disorder characterized by reduced frataxin expression and widespread mitochondrial impairment [44].

FRDA primarily affects the nervous system, but frataxin deficiency also impacts the heart, skeletal muscle, and liver, highlighting its broader role in mitochondrial health [45]. Our findings suggest that frataxin may be a key determinant of mitochondrial integrity and function in the context of hepatic iron overload. The observed frataxin loss in IO mice, likely driven by iron toxicity, contributes to hepatocellular damage. Importantly, DFP-RVT treatment restored frataxin expression, which may account for the recovery of mitochondrial protein levels and overall mitochondrial function. A previous study in HbE-β-thalassemia patients demonstrated an inverse correlation between malondialdehyde (MDA) levels and FXN mRNA expression, suggesting that higher frataxin levels are associated with reduced oxidative stress. This indicates that elevated frataxin may provide a protective effect by mitigating iron-induced oxidative damage. Furthermore, their findings also showed that frataxin is essential not only for heme biosynthesis through the facilitation of mitochondrial iron incorporation but also for iron detoxification by delivering iron to mitochondrial ferritin, thereby preventing free iron-mediated mitochondrial toxicity. Together, these functions highlight frataxin’s critical role in maintaining mitochondrial integrity and defending against oxidative stress, especially in conditions of iron overload [46].

Nevertheless, in this study, the proteins involved in iron homeostasis proteins (e.g., hepcidin, ferritin, ceruloplasmin, haptoglobin or hemopexin) and classical antioxidant enzymes (e.g., GPx, SOD, GST, thioredoxin) were not significantly altered at the proteomic level in IO/BKO and DR-IO/IO comparisons. This finding suggests that iron-induced oxidative injury in this model is driven primarily by mitochondrial dysfunction and iron misdistribution rather than by altered expression of canonical antioxidant systems. This interpretation is supported by prior work demonstrating significant changes in circulating hepcidin levels without corresponding hepatic protein abundance changes [17], underscoring the importance of post-transcriptional regulation and systemic iron signaling pathways in iron-overloaded pathophysiology.

A limitation of this study is that treatment with deferiprone (DFP) was not included, which limits the ability to directly compare the effects of the hybrid iron chelator DFP-RVT with those of the standard therapy. Including a DFP treatment group in future research would provide valuable insights into the distinct mechanisms of the standard chelator, DFP, and the novel hybrid iron chelator, DFP-RVT. Although both compounds demonstrate iron-chelating and antioxidant activities, their underlying mechanisms may differ, making such comparisons important for optimizing therapeutic strategies.

In summary, this study provides proteomic evidence that DFP-RVT mitigates iron-induced hepatic injury primarily through the restoration of mitochondrial function, with frataxin emerging as a central mediator linking iron detoxification, mitochondrial bioenergetics, and oxidative stress resistance. These findings advance the mechanistic understanding of DFP-RVT action and support its potential as a mitochondria-targeted therapeutic strategy for iron-overload-associated liver disease in β-thalassemia.

5. Conclusions

This study demonstrates that hepatic iron overload in β-thalassemia is associated with mitochondrial dysfunction, metabolic disturbance, and activation of pathways related to inflammation, fibrosis, and cellular injury. Using a proteomic approach, we show that treatment with DFP-RVT partially ameliorates iron-induced hepatic damage, primarily through restoration of mitochondrial-related protein expression and improvement of mitochondrial integrity. Notably, DFP-RVT restored the expression of key mitochondrial proteins, including frataxin, suggesting a potential role in reversing iron-induced mitochondrial impairment. These findings provide mechanistic insight into how DFP-RVT may confer hepatoprotection in iron overload conditions and supports its potential as a mitochondria-targeted therapeutic strategy for iron overload-associated liver disease in β-thalassemia. Further studies incorporating direct comparison with standard chelation therapy, functional validation of mitochondrial pathways, and translational investigation will be necessary to clarify the therapeutic relevance and broader applicability of DFP-RVT.