Cytochrome b₅ reductase-3 overexpression reduces renal damage associated to intravascular hemolysis

Abstract

Intravascular hemolysis can cause both acute kidney injury (AKI) and chronic kidney disease (CKD). When erythrocytes break down in the bloodstream, they release hemoglobin and heme derivatives, which accumulate in renal cells and lead to increased oxidative stress, inflammation, cell death, and mitochondrial dysfunction. The enzyme NADH-cytochrome b₅ reductase 3 (CYB5R3) plays a role in NAD⁺ metabolism and helps reduce oxidative stress and inflammation while enhancing mitochondrial function. In this study, we examined whether overexpressing CYB5R3 offers kidney protection against AKI caused by intravascular hemolysis and prevents the progression from AKI to CKD, including analysis of sex-based differences. We created an experimental model to assess acute and chronic kidney damage related to intravascular hemolysis in wild-type (WT) and CYB5R3-overexpressing transgenic (TG) mice. Additionally, we performed in vitro experiments to determine if pharmacological induction of CYB5R3 with tetrahydroindenoindole (THII) reduced heme-mediated nephrotoxicity. AKI resulting from intravascular hemolysis impaired kidney function, increased damage to tubular epithelium and podocytes, promoted oxidative stress and inflammation, caused mitochondrial dysfunction, altered autophagy-related markers, and cell death—effects that were more severe in males than females. CYB5R3 overexpression notably diminished these pathological changes and slowed CKD progression by decreasing inflammation and fibrosis, regardless of sex. In cellular systems, induction of CYB5R3 with THII lowered heme-induced reactive oxygen species (ROS) production, inflammation, and cell death, and restored NAD⁺ levels. These protective effects were eliminated by pharmacological inhibition of CYB5R3 activity. Our findings suggest that CYB5R3 could be a promising therapeutic target for preventing AKI related to intravascular hemolysis and its progression to CKD.

1. Introduction

Massive intravascular hemolysis, which involves the lysis of red blood cells within the circulation [1], is a feature of diverse pathological conditions, such as sickle cell disease, and thrombotic microangiopathies, among others [2–8]. As a consequence of erythrocytes break down, large quantities of hemoglobin (Hb) and heme-derivates are released into the bloodstream, which are subsequently accumulated in the kidney, particularly within proximal tubular epithelium cells and podocytes [9,10]. Once internalized by renal cells, hemoproteins trigger nephrotoxic effects, including oxidative stress, inflammation and cell death [1,8,11], resulting in the development of acute kidney injury (AKI), which eventually progresses towards chronic kidney disease [7].

NADH-cytochrome b₅ reductases (CYB5R, EC 1.6.2.2) constitute a family of flavoprotein oxidoreductases that catalyse the oxidation of NADH to NAD⁺ coupled with the reduction of a variety of both endogenous and exogenous acceptors. Five CYB5R isoforms (CYB5R1–5) have been identified to date; among them, CYB5R3 is the best characterized one owing to its central role in the regulation of redox equilibrium and cell metabolism [12–15]. CYB5R3 exists in two main forms: a soluble isoform found in erythrocytes, where it reduces methemoglobin to hemoglobin thereby supporting efficient oxygen transport to tissues [12, 16,17], and a membrane-bound isoform that is widely distributed across cellular membranes, including the cytosolic side of the outer mitochondrial membrane (OMM), the endoplasmic reticulum (ER), and the plasma membrane. Membrane-associated CYB5R3 facilitates electron transfer from NADH to a variety of electron acceptors, primarily hemoproteins. Among the best-characterized are the two cytochrome b₅ isoforms: CYB5A, localized at the ER, and CYB5B, at the OMM. The CYB5R3–CYB5A axis plays essential roles in several metabolic processes, including cholesterol and steroid hormone biosynthesis [18], fatty acid elongation and desaturation [19], and cytochrome P450-mediated drug hydroxylation [20]. On the other hand, the CYB5R3–CYB5B system uses NADH to regenerate cytosolic ascorbate from its semioxidized form, the ascorbate free radical [21]. Additionally, mitochondrial CYB5R3 may interact with VDAC1 to maintain intracellular NAD⁺/NADH balance and facilitate the transport of ascorbate radicals into the intermembrane space [22]. Other hemoprotein targets of CYB5R3 include soluble guanylate cyclase (sGC) in smooth muscle cells, which becomes sensitized to nitric oxide (NO); cytoglobin in endothelial cells, which modulates NO bioavailability; and myoglobin, which enhances oxygen delivery to mitochondria in muscle cells [23]. Beyond hemoproteins, CYB5R3 also reduces non-heme molecules such as coenzyme Q (CoQ). At the plasma membrane, reduced CoQ (ubiquinol) regenerates key antioxidants like α-tocopherol and extracellular ascorbate, thereby amplifying their antioxidant effects, preventing lipid peroxidation, preserving membrane integrity, and protecting against ceramide-induced apoptosis [15, 24–28]. Recent studies have highlighted a functional interaction between CYB5R3 and NOX4 at the OMM, where CYB5R3-driven reduction of CoQ regulates superoxide and hydrogen peroxide production by NOX4, which plays a critical role in controlling oxidative stress and inflammation in endothelial cells [29].

Overexpression of CYB5R3 reduced oxidative stress and inflammation, improved glucose metabolisms, protected from chemically induced cancer, and increased longevity in transgenic mice [30]. However, the mechanisms underlying these beneficial effects depend on sex: transgenic males exhibited elevated carbohydrate metabolism and increased adiposity [30], whereas transgenic females displayed enhanced fatty acid oxidation and lower fat mass [31]. Moreover, metabolic adaptations elicited in transgenic mice not only exhibit sexual dimorphism, but are also tissue-specific [31,32]. In skeletal muscle, CYB5R3 overexpression enhanced mitochondrial biogenesis and function, with more prominent results in females [33]. In the kidney, CYB5R3 overexpression resulted in changes in mitochondrial morphology, abundance and mitochondria-reticulum contact sites in distal tubular cells, mainly in females as well [34]. On the other hand, the upregulation of biochemical markers consistent with enhanced mitochondrial function in cardiomyocytes has been predominantly observed in transgenic males [35]. In liver, transgenic females and males exhibited distinct metabolic reprogramming which also involved systemic regulatory mechanisms [36]. By the contrast, genetic CYB5R3 deficiency or mutations impaired reduction of methemoglobin, resulting in cyanosis or neurological disorders [16,37,38]. Defects in CYB5R3 also reduced CoQ levels, promoting cardiac hypertrophy and sudden death in males but not females [39]. For all these reasons, CYB5R3 has emerged as a potential therapeutic target in diseases associated to redox imbalance. Tetrahydroindeno [1,2-b] indole (THII) upregulated CYB5R3 expression and activity in various tissues [40], resulting in reduced inflammation and increasing life expectancy [30].

To date, no previous studies have assessed the role of CYB5R3 in the development of hemolysis-AKI and the subsequent AKI-CKD transition, nor whether activating this enzyme could protect renal cells from hemoprotein-induced toxicity. To explore these questions, we developed an experimental model of intravascular hemolysis in both male and female mice of wild-type (WT) and CYB5R3-overexpressing transgenic (TG) genotypes. We then investigated whether the CYB5R3 activator THII could reduce heme-mediated damage in renal cells. Our findings suggest that CYB5R3 upregulation may serve as a potential therapeutic target for treating AKI caused by significant intravascular hemolysis.

2. Material and methods

2.1. Animal model

Transgenic mice that overexpress the CYB5R3 enzyme (TG) were generated as described by Martín-Montalvo et al. [30]. We performed an experimental model of AKI associated with intravascular hemolysis in 24-week-old male and female C57BL6 mice of wild-type (WT) or TG genotypes by injecting a freshly prepared phenylhydrazine solution (150 mg/kg body weight, Cat. 114715, Sigma, MO, USA, i.p,). In a first set of experiments, animals were sacrificed 72 h after hemolysis induction. In another set of experiments, we aimed to evaluate the role of the CYB5R3 enzyme in the transition from AKI to CKD. For this, we administered an initial injection of phenylhydrazine (150 mg/kg body weight), followed by weekly 75 mg/kg doses for three consecutive weeks. Animals were sacrificed 28 days after the first injection. Blood, urine and kidney tissue samples were collected in both experiments. We used blood samples for biochemical analysis (Atellica Solution, Siemens Healthineers) and heme serum determination using a commercial assay kit (Cat. MAK316, Sigma-Aldrich). Urine albumin levels were obtained by enzyme-linked immunosorbent assay (ELISA, Cat. ab108792, Abcam (Cambridge, UK)) and normalized to urine creatinine concentration (Creatinine Assay Kit; Cat.11802, Biosystems). Kidneys were dissected and fixed in 4% paraformaldehyde and embedded in paraffin for histological studies, or snap-frozen for RNA and protein expression determination. The study was performed according to Directive 2010/63/EU of the European Parliament and was approved by the bioethics committee of the University of Córdoba and authorized by the Consejería de Agricultura, Pesca y Desarrollo Rural, Junta de Andalucía, Spain (authorization code: 25/08/2021/117).

2.2. Cell cultures

Murine tubular-epithelial renal cells (MCT) were cultured in RPMI 1640 (Cat. R0883, Sigma) supplemented with 10% heat-inactivated foetal bovine serum (FBS, Cat. F7524, Sigma), glutamine (2 mmol/L, Cat. G7513, Sigma), and penicillin/streptomycin (100 U/ml; Cat. P0781, Sigma) in a 5% CO₂ atmosphere at 37°C. Cells were deprived of FBS for 16 h when their confluence reached 70–80% and pre-treated during 24 h with the CYB5R3 inducer THII (0–2000 μg/ml, obtained through chemical synthesis, kindly provided by Rafael de Cabo (NIA, NIH, Baltimore, USA)) in the presence/absence of the specific CYB5R3 enzyme inhibitor ZINC05626394 (Z, 0–200 μM, Cat. 189057–68–9, Sigma-Aldrich). After that, the cells were stimulated with heme (0.5–5 μM, Cat. H9039, Sigma-Aldrich, dissolved in 50% triethanolamine (Cat. T58300, Sigma-Aldrich) at different time points.

2.3. Hematoxylin/Eosin and Sirius Red staining

Histological analyses were conducted on formalin-fixed, paraffin-embedded renal Section (3 μm). Samples were deparaffinized and hydrated in increasing concentrations of ethanol. Renal injury was assessed using hematoxylin/eosin staining (Cat. 10034813 and Cat. 6766007, ThermoFisher). Also, we evaluated interstitial fibrosis in renal section using Picrosirius Red staining. In this case, the samples were immersed for one hour in a saturated aqueous solution containing 0.5 g of Sirius red (“Direct Red 80”, Cat. 365548, Sigma-Aldrich) in 500 ml of 1.3% picric acid (Cat. P6744–1GA, Sigma-Aldrich). The samples were washed in 0.05% acidic water (5 ml of glacial acetic acid in 1 Liter of distilled water), dehydrated, and mounted with non-aqueous mounting medium DPX (Cat. 100579, Sigma-Aldrich). Images were captured using a Leica DM2000 LED microscope coupled with a Leica MC190 HD camera (Leica Microsystems GmbH; Advanced Optical Microscopy Unit, IMIBIC). The percentage of positive staining per field was quantified in 10 randomly selected areas per tissue section at 20x objective magnification

2.4. Immunochemistry and immunofluorescence

Renal sections were deparaffinized and subjected to antigen retrieval using citrate buffer (pH 6.0, Cat. S/3280/53, ThermoFisher) to perform F4/80 immunohistochemistry. Peroxidase activity was blocked by treating the samples with 3% (v/v) H₂O₂ for 30 min. Non-specific antibody binding was prevented by blocking with 10% (v/v) goat serum (Cat. 50197Z, ThermoFisher) diluted in PBS buffer containing 4% (w/v) bovine serum albumin (BSA, Cat. 10326640, ThermoFisher). The samples were incubated for 1 h at room temperature. Sections were incubated overnight at 4°C with the primary antibody F4/80 (1:80, Cat.70076, Cell Signaling). Following incubation, sections were washed with PBS and subsequently incubated at room temperature for 1 h with a secondary goat anti-rabbit IgG (H+L) biotinylated antibody (ready to use, Cat. BP-9100, Vector Laboratories). This was followed by a 30-minute incubation at room temperature, protected from light, with an avidin-biotin peroxidase complex (Vectastain ABC kit, Cat. PK-4000, Vector Laboratories). Finally, the sections were developed using 3,3’-diaminobenzidine (DAB substrate kit, Cat. ab64238, Abcam), counterstained with hematoxylin, mounted using DPX non-aqueous mounting medium (Cat. 100579, Sigma-Aldrich) and sealed with coverslips. Images were captured using a Leica DM2000 LED microscope coupled with a Leica MC190 HD camera (Leica Microsystems GmbH; Advanced Optical Microscopy Unit, IMIBIC). The percentage of positive staining per field was quantified in 10 randomly selected areas per tissue section at 20x objective magnification.

Cell death was assessed in renal sections with the terminal deoxynucleotidyl transferase dUTP nick-end labeling (TUNEL) assay, according to the manufacturer’s protocol (Cat. S7110, Millipore). Nuclei were counterstained with 4′-6-diamidino-2-phenylindole (DAPI; Cat. D9542, Sigma-Aldrich), and slides were mounted using FluorSave mounting medium (Cat. 345789, Millipore). Images were acquired with a Leica TCS SP5 inverted confocal microscope. TUNEL-positive cells were quantified by counting the number of labeled cells per field in at least 10 randomly selected areas per tissue section at 20x magnification. The average number of labeled cells per field was reported as TUNEL-positive cells. Image analysis was performed using Image-Pro Plus software (Version 4.5.0.29, USA).

2.5. Immunofluorescence in cells

MCT cells were cultured in RPMI 1640 medium (Cat. R0883, Sigma) with 10% FBS until reaching 70–80% confluence. They were then incubated in RPMI 1640 medium containing 0% FBS for 16 h and then stimulated with THII (60 μg/ml) for 24 h. Afterward, the cells were fixed with 4% paraformaldehyde, washed with PBS 1X, and permeabilized with Triton X-100 0.25% (Cat. T8787, Sigma-Aldrich). The cells were incubated with a primary antibody against CYB5R3 (1/100, Cat. 16837583, Thermofisher) at 4°C overnight, followed by incubation with a secondary anti-rabbit Alexa Fluor 488 antibody (1/200, Cat. 10424752, Thermofisher). The nuclei were stained with 4′-6-diamidino-2-phenylindole (DAPI; Cat. D9542, Sigma-Aldrich), and the samples were mounted with FluorSave mounting medium (Cat. 345789, Millipore) and stored at 4°C. Images were captured using a Leica TCS SP5 confocal microscope and analyzed with Image-Pro-Plus software.

2.6. Isolation of mitochondria-enriched fractions

We performed subcellular fractionation to obtain mitochondrial extracts from kidney tissue previously stored at −80°C. Each kidney sample was minced and resuspended in homogenization buffer (20 mM Tris-HCl, pH 7.6, 40 mM KCl, 0.2 M sucrose, 10 mM EDTA) supplemented with 1 mM DTT, 1 mM phenylmethylsulfonyl fluoride (PMSF), and 20 μg/ml CLAP (chymostatin, leupeptin, antipain, and pepstatin A). Tissue samples were homogenized in a steel ball mill (MM400, Retsch, Germany) for 5 min at a speed of 25 strokes/second using pre-chilled adaptors. Following homogenization, the samples were centrifuged at 420 × g for 10 min at 4 °C, producing a pellet containing the nuclei and cellular debris fraction, which was then discarded. The resulting supernatants were collected and subjected to a second centrifugation at 6700 × g for 10 min to isolate a mitochondria-enriched fraction. The supernatants were resuspended in 100 μl of isolation buffer and stored at −80°C until further analysis.

2.7. Western-Blot

Extraction of protein from tissue samples or cultured cells was carried out using a lysis buffer containing 50 mM Tris-HCl, 150 mM NaCl, 2 mM EDTA, 2 mM EGTA, 0.2% Triton X-100, 0.3% NP-40, 0.2 mM PMSF, 0.1 mM dithiothreitol (DTT), and a protease and phosphatase inhibitor cocktail (Cat. PPC1010, Sigma). The lysates were centrifuged at 12,000 × g for 15 min at 4°C, and the supernatants were collected. Protein concentrations were determined using a BCA protein assay kit (Cat. 23225, ThermoFisher). Equal amounts of protein (30 μg) from each lysate were separated by SDS-PAGE and transferred onto PVDF membranes (Cat. 10600021, GE Healthcare) for Western blot analysis. Protein loading consistency was verified using Ponceau S staining. The membranes were blocked for 1 h at room temperature with 5% w/v skimmed milk powder in Tris-buffered saline (TBS) containing 0.5% v/v Tween-20 (Cat. 10491081, ThermoFisher), then incubated overnight at 4°C with the appropriate primary antibodies (Table 1). After primary antibody incubation, membranes were washed with TBS containing 0.5% Tween-20 and incubated with horseradish peroxidase-conjugated secondary antibodies (Table 1) for 1 h at room temperature. Following additional washes, blots were developed using an enhanced chemiluminescence detection system (ECL Luminata Crescendo; Cat. WBLUR0500, Millipore). Densitometric analysis of the blots was performed using the Chemi Doc XRS+ system (Bio-Rad). Images were analysed with QuantityOne software (Bio-Rad), and protein quantification was expressed in arbitrary densitometric units (AU). The intensity of each target protein band was normalized to the signal from Ponceau S-stained lanes. The normalized values of the control group were averaged to establish a reference value. Fold changes in protein expression were then calculated relative to this average reference value.

Table 1.

Antibodies used in Western-Blot.

Primary antibodies	Dilution	Reference
anti-HO-1	1:1000	ADI-OSA-150-D, Enzo Life Sciences
anti-Ferritin	1:1000	701934, ThermoFisher
anti-Nephrin	1:1000	BP5030, Origene
anti-Synaptopodin	1:1000	65194, Progen
anti-OxPhos	1:4000	45-8099, ThermoFisher
anti-Mfn1	1:1000	sc- 50330, Santa Cruz Biotechnology
anti-Mfn2	1:1000	sc-50331, Santa Cruz Biotechnology
anti-DRP1	1:500	sc-32989, Santa Cruz Biotechnology
anti-TFAM	1:1000	sc-2358, Santa Cruz Biotechnology
Anti-Nrf1	1:2000	sc-33771, Santa Cruz Biotechnology
anti-Tom20	1:1000	PA5-52843, ThermoFisher
anti-Nrf2	1:1000	12721, Cell Signaling
anti-p62	1:3000	P0067, Sigma-Aldrich
anti-LC3 AB	1:1000	4108, Sigma-Aldrich
anti-Parkin1	1:1000	2132, Cell Signaling
anti-Pink1	1:1000	sc-33796, Santa Cruz Biotechnology
anti-GADPH	1:5000	MAB374, Sigma-Aldrich
anti-α-tubulin	1:5000	T6199, Sigma-Aldrich
Secondary antibodies	Dilution	Reference
anti-rabbit IgG	1:5000	NA934, GE Healthcare
anti-mouse IgG	1:5000	NA931, GE Healthcare
anti-goat IgG	1:5000	A-5420, Sigma-Aldrich

2.8. RNA extraction and RT-PCR

Total RNA from renal tissue samples and cultured cells was extracted using the TRIzol method with TRI Reagent (Cat. MRC-TR-118–200, Sigma-Aldrich) and converted to cDNA with the High-Capacity cDNA Reverse Transcription Kit (Cat. 4368813, ThermoFisher). Target gene expression was quantified by RT-qPCR on a CFX-Connect Real-Time System (Bio-Rad) using SYBR Green probes specific for the following mouse genes: Lcn2 (Neutrophil Gelatinase-Associated Lipocalin, NGAL), Hacvr1 (KIM-1, Kidney injury molecule-1), Hmox1 (Heme oxygenase 1, HO-1), Fth1 (Ferritin), Sod1 (Superoxide dismutase 1, SOD1), Cat (Catalase), Chop (C/EBP homologous protein), ATF4 (Activating transcription factor 4, ATF4), sXBP1 (Spliced X-box binding protein 1, sXBP1), Ccl2 (C-C motif chemokine ligand 2), Il6 (Interleukin 6), Nphs1 (Nephrin), Synpo (Synaptopodin), Fn1 (Fibronectine-1) and Col1 (Collagen type-1). GAPDH (Glyceraldehyde 3-phosphate dehydrogenase) levels were used to normalize target gene expression and quantified using the 2-ΔΔCt method. Gene sequences are provided in Table 2.

Table 2.

Mouse SYBR gene sequences.

Gene	Forward	Reverse
Lcn2	5′-TCCTTGAGGCCCAGAGACTT-3′	5′-CTCAGAACTTGATCCCTGCC-3′
Hacvr1	5′-CTGGAATGGCACTGTGACATCC-3′	5′-GCAGATGCCAACATAGAAGCCC-3′
Hmox1	5′-CATGAAGAACTTTCAGAAGGG-3′	5′-TAGATATGGTACAAGGAAGCC-3′
Fth1	5′-AGACCGTGATGACTGGGAGA-3′	5′-TGAAGTCACATAAGTGGGGATCA-3′
Sod1	5′-CACTCTAAGAAACATGGTGG-3′	5′-GATCACACGATCTTCAATGG-3′
Cat	5′-CTCCATCAGGTTTCTTTCTTT-3′	5′-CAACAGGCAAGTTTTTGATG-3′
Ccl5	5′-CAGGGTCAGAATCAAGAAAC-3′	5′-AGGAGTATTTCTACACCAGC-3′
Chop	5′-CCACCACACCTGAAAGCAGAA-3′	5′-AGGTGAAAGGCAGGGACTCA-3′
ATF4	5′-GGGTTTGTCTTCCACTCCA-3′	5′-AAGCAGCAGAGTCAGGCTTTC-3′
sXBP1	5′-CTGAGTCCGAATCAGGTGCAG-3′	5′-GTCCATGGGAAGATGTTCTGG-3′
Ccl2	5′-TTGGTGACAAAAACTACAGC-3′	5′-CAAGATGATCCCAATGAGTAG-3′
Il6	5′-AAGAAATGATGGATGCTACC-3′	5′-GAGTTTCTGTATCTCTCTGAAG-3′
Nphs1	5′-AATCCTCTCCATCCTAGTTT-3′	5′-ACATATTCCTGTCCAGCTAC-3′
Synpo	5′-CATCCTCTATCCTGTAGCTC-3′	5′-TAGATTTGAGACAGAGTGTATTG-3′
Fn1	5′-CCTATAGGATTGGAGACACG-3′	5′-GTTGGTAAATAGCTGTTCGG-3′
Col1	5′-CGTATCACCAAACTCAGAAG-3′	5′-GAAGCAAAGTTTCCTCCAAG-3′
GADPH	5′-TGCACCACCAACTGCTTAGC-3′	5′-CGCATGGACTGTGGTCATGAG-3′

2.9. Respirometry analysis in kidney isolated mitochondria

To analyse the oxygen consumption rate in renal mitochondrial extracts, we measured the maximal respiratory rate of complexes II+III+IV as well as the maximal respiration rate mediated solely by complex IV using a Seahorse XFe24 flux analyser (Agilent, Santa Clara, CA, USA), following the protocol outlined by Corey Osto et al. [41]. Each assay contained 5 μg of mitochondrial extracts resuspended in MAS buffer (70 mM sucrose, 220 mM mannitol, 5 mM KH₂PO₄, 5 mM MgCl₂, 1 mM EGTA, 2 mM HEPES, pH 7.4) and was supplemented with cytochrome c (10 μg/ml, Cat. C2506, Sigma-Aldrich). The samples were loaded into a 24-well Seahorse plate in a volume of 50 μl and centrifuged at 3200 g for 20 min at 4°C to ensure mitochondrial adhesion to the bottom of the plate. Then, 450 μl of MAS buffer with cytochrome c were carefully added to each well. The maximum respiration rate through complexes II+III+IV was calculated after activation with succinate (0.65 mM) in the presence of rotenone (0.26 mM), followed by inhibition with antimycin A (0.46 μM). The maximum respiration rate for complex IV was measured by stimulation with N,N,N′,N′-tetramethyl-p-phenylenediamine (TMPD, 50 μM) plus ascorbic acid (0.1 mM), and subsequent inhibition with sodium azide (4.6 mM). At the conclusion of the experiment, results were normalized to mitochondrial protein concentration.

2.10. Oxidative stress determination

We measured intracellular ROS production, glutathione (GSH), lipid peroxidation, and mitochondrial ROS using the molecular probes 2′,7′-dichlorodihydrofluorescein diacetate (H2DCFDA) (Cat. C6827; Invitrogen), CMFDA Cell Tracker (Cat. C2925, Invitrogen), Bodipy (Cat. D3861, Invitrogen), and MitoSOX (Cat. M36007, ThermoFisher), respectively. MCT cells were pretreated with THII (60 μg/ml) and/or Z (25 μM) for 16 h before being stimulated with heme (0.5 and 5 μM) for 2–16 h. After incubation, the cells were washed with PBS and centrifuged at 1200 rpm for 5 min at 4°C. The resulting pellet was resuspended in PBS, and the samples analysed using a BD FACSCanto II cytometer (BD Biosciences, San Jose, CA, USA) and FlowJo software (BD Biosciences). Additionally, malondialdehyde (MDA) levels in kidney extracts were quantified using the Thiobarbituric Acid Reactive Substances (TBARS) assay kit (Cat. CA995, Canvax), according to the supplier recommendation.

2.11. Cytotoxicity and cell viability assay

Citotoxicity of THII or Z was determined in MCT cells cultured in 96-well plates. When they reached 70–80% confluence, the cells were deprived of FBS for 16 h and then treated with THII (0–2000 μg/ml) or Z (0–200 μM). After 24 h, the medium was removed, and the cells were incubated with 1 mg/ml MTT (3-[4,5-dimethylthiazol-2-yl]-2,5-diphenyltetrazolium bromide), Cat. A2231, PanReac) reagent for 5–10 min at 37°C. Once the wells displayed a purple colour, the reagent was removed, and the cells allowed to dry at room temperature. The cells were then resuspended in 100 μl of DMSO, and the absorbance was measured at 570 nm.

Cell viability was determined in renal cells stimulated with heme for 24 h in presence/absence of THII (60 μg/ml) and Z (25 μM) using a commercial kit (WST-1, Cat. 5015944001, Roche).

2.12. NAD⁺/NADH determination

MCT cells were cultured in 6-well plates. When the cells reached 70–80% confluence, they were incubated without FBS for 16 h. Simultaneously, the cells were pretreated with THII (60 μg/ml) and/or Z (25 μM). Subsequently, the cells were stimulated with heme (5 μM) for 6 h. NAD⁺ and NADH levels were measured in total cell extracts using the NAD⁺/NADH-Glo^™ bioluminescent assay (Cat. G9071, Promega, Madison, WI, USA). The resulting values were normalized to protein concentrations determined using the BCA protein assay kit (Cat. 23225, ThermoFisher).

2.13. Statistical analysis

The results obtained in this study were analysed using the statistical program GraphPad Prism 8 (San Diego, CA, USA). Differences between experimental groups were evaluated using the Shapiro-Wilk normality test. Data with a parametric distribution were subjected to one-way or two-way ANOVA, followed by Tukey or Sidak post hoc analysis to identify significant differences between groups. For non-parametric distribution data, the Kruskal-Wallis test was applied, followed by Dunn's multiple comparison test. All data are presented as the mean ± SEM. Data were considered significantly different when the p-value was less than 0.05.

3. Results

3.1. CYB5R3 overexpression reduces renal injury associated to intravascular hemolysis

To evaluate the role of CYB5R3 in renal damage associated with intravascular hemolysis, we performed an experimental model in male and female mice of both WT and TG genotypes. Intravascular hemolysis was induced through intraperitoneal administration of phenylhydrazine, as previously reported [8]. The existence of hemolysis was confirmed by decreased red blood cells and hematocrit as well as increased heme serum levels in WT and TG mice treated with phenylhydrazine, independently of sex or genotype (Supplemental Figure 1). Hemolysis induction resulted in acute loss of renal function, as characterized by increased serum creatinine and blood urea nitrogen (BUN) levels. However, this effect was higher in WT than in TG males (Fig. 1A and B). Similarly, WT females with hemolysis showed an increase in BUN levels, although this effect was less pronounced in TG females (Fig. 1E). In contrast, no significant changes in serum creatinine levels were observed in females of either genotype (Fig. 1F). Consistent with these findings, the tubular injury markers NGAL (Lcan2) and KIM-1 (Hacvr1) were upregulated under hemolysis induction (Fig. 1C–D, G–H). Histological analysis confirmed renal injury, evidenced by the presence of pyknotic nuclei, loss of the brush border, and tubular dilation (Fig. 1I), along with increased cell death as determined by TUNEL staining (Fig. 1J–L). All these deleterious effects associated with intravascular hemolysis were significantly attenuated in TG mice compared to WT, with no observable sex-related differences. We next analyzed renal CYB5R3 protein levels in response to hemolysis in both WT and TG mice. Induction of hemolysis reduced renal CYB5R3 levels in both males and females independently of CYB5R3 genotype. However, TG mice displayed significantly higher levels of CYB5R3 compared to WT counterparts both at basal state and under hemolytic conditions, independently of sex (Fig. 1M and N).

Fig. 1.

CYB5R3 overexpression reduces renal injury associated to intravascular hemolysis. Hemolysis was induced by administering phenylhydrazine (150 mg/ body weight) to 12-week-old C57BL6 male and female mice of wild-type (WT) and transgenic CYB5R3-overexpressing (TG) genotypes (n = 4–9 per group). Mice were sacrificed 72 h after hemolysis induction. (A, E) Blood uremic nitrogen (BUN) and (B, F) serum creatinine levels in males and females. (C, G) Gene expression of tubular damage markers NGAL (Lcn2) and (D, H) KIM1 (Hacvr1) in males and females. (I) Representative images of Hematoxylin/Eosin staining of kidney sections at 20X magnification in both males and females. Scale bar 50 μm. Arrows indicate signs of kidney damage: loss of tubular epithelium, pyknotic nuclei and cellular debris in the tubular lumen. (J) Representative image of dead cells in kidney sections at 20X magnification by TUNEL immunofluorescence technique (green). Nuclei stained with DAPI (blue). Scale bar 100 μm. (K, L) Semi-quantification of TUNEL-positive cells in males and females. (M, N) Semi-quantification and representative image of CYB5R3 enzyme expression in mitochondria isolated from kidney in males and females. Protein expression and representative image of Nephrin (Nphs1) and Synaptopodin (Synpo) in males (O, P) and females (R, S) obtained by Western-Blot. Gene expression of Nphs1 and Synpo in males (Q) and females (T) obtained by real-time PCR. Results are expressed as mean ± SEM. ^a p < 0,05 compared to WT saline. ^b p < 0,05 compared to TG saline. ^c p < 0,05 compared to TG vs WT under hemolytic conditions. The main effect of hemolysis (regardless of genotype) is represented as “Hemolysis”, while the main effect of genotype (regardless of hemolysis) is represented as “WT-TG”. The interaction “hemolysis x genotype” is also indicated when relevant. * p < 0,05 in the analysis of overall effect (genotype, hemolysis or interaction).

Similar to tubular epithelial cells, podocytes can internalize hemoglobin released during intravascular haemolysis, a process that compromises the viability of these cells and disrupts glomerular filtration barrier [9]. In this line, we observed a decreased expression of key structural proteins of podocytes, such as Synaptopodin (Synpo) and Nephrin (Nphs1), under hemolytic conditions (Fig. 1O–T). Importantly, CYB5R3 overexpression prevented the hemolysis-induced loss of Synaptopodin and Nephrin in both males and females. Consistent with these findings, hemolytic WT showed an increased urinary albumin/creatinine ratio, indicative of glomerular injury, whereas this response was reduced in TG mice, with no sex-dependent differences (Supplemental Figure 1D and H).

3.2. CYB5R3 overexpression decreases oxidative stress, inflammation and autophagy caused by massive intravascular hemolysis

CYB5R3 plays an essential role in maintaining cellular redox balance [14]. During intravascular hemolysis, the accumulation of hemoglobin and heme leads to increased production of ROS [42]. Therefore, we examined whether CYB5R3 overexpression could influence oxidative stress linked to intravascular hemolysis. Our results showed a significant rise in renal MDA levels following hemolysis induction, with this increase being notably more significant in WT mice compared to TG mice, in both males and females (Fig. 2A and E).

Fig. 2.

CYB5R3 overexpression decreases oxidative stress, inflammation and autophagy caused by massive intravascular hemolysis. Hemolysis was induced by administering phenylhydrazine (150 mg/ body weight) to 12-week-old C57BL6 male and female mice of wild-type (WT) and transgenic CYB5R3-overexpressing (TG) genotypes (n = 4–9 per group). Mice were sacrificed 72 h after hemolysis induction. (A, E) Malondialdehyde (MDA) levels in kidney tissue in males and females. Semi-quantification and representative image of heme oxygenase 1 (HO-1), Ferritin (Fth1) and nuclear transcription factor erythroid 2 (Nrf2) levels in males (B, C) and females (F, G). Semi-quantification and representative image of Nox4 in male (D) and female (H). Gene expression of HO-1 (Hmox1), Ferritin (Fth1), catalase (Cat) and superoxide dismutase (Sod1) in males (I, J) and females (L, M). Chop, ATF4 and sXBP1 gene expression in males (K) and females (N). (O, R) Gene expression of proinflammatory cytokines Il-6 and Ccl2 in males and females by real-time PCR. Semi-quantification and representative image of autophagy-related proteins, p62/SQSTM1 and LC3-II/LC3-I, in males (P, Q) and females (S, T). Results are expressed as mean ± SEM. ^a p < 0,05 compared to WT saline. ^b p < 0,05 compared to TG saline. ^c p < 0,05 compared to TG vs WT under hemolytic conditions. The main effect of hemolysis (regardless of genotype) is represented as “Hemolysis”, while the main effect of genotype (regardless of hemolysis) is represented as “WT-TG”. The interaction “hemolysis x genotype” is also indicated when relevant. * p < 0,05 in the analysis of overall effect (genotype, hemolysis or interaction).

Hemolysis also induced an increase in the transcription factor Nrf2 and its downstream targets HO-1 and Fth1, with a similar response observed in WT mice of both sexes (Fig. 2B, C, F, and G). However, the increase in HO-1 and Fth1 protein levels under hemolytic conditions was reduced in males overexpressing CYB5R3 compared to WT males under the same conditions (Fig. 2B and C). A similar pattern was seen in females, although no significant differences between genotypes were found (Fig. 2F and G). Although Hmox1 and Fth1 gene expression increased with hemolysis, this upregulation was diminished in TG males (Fig. 2I). A similar trend was observed in females, although significant genotype differences were only found for Hmox1 (Fig. 2L). We also examined the expression of Nox4, a key enzyme generating ROS in the kidney in our model. Results showed that hemolysis decreased Nox4 protein levels in both males and females of WT genotype. Interestingly, Nox4 expression remained steady in hemolytic mice overexpressing CYB5R3, regardless of sex (Fig. 2D and H). Additionally, hemolysis caused a reduction in Sod1 and Cat expression in WT males but not in females (Fig. 2J and M); however, expression of these antioxidant enzymes remained high after hemolysis induction in TG females (Fig. 2M). Intravascular hemolysis increased the expression of genes involved in the unfolded protein response (UPR), such as Chop and sXBP1, but this increase was less pronounced in TG mice of both sexes (Fig. 2K and N). For ATF4, expression changes were seen only in WT males after hemolysis (Fig. 2K and N).

Our results showed increased expression of pro-inflammatory cytokines Ccl2 and Il-6 under hemolytic conditions, regardless of sex (Fig. 2O and R). However, TG females, but not TG males, displayed lower levels of these pro-inflammatory markers after hemolysis. Altered autophagy flux, as shown by an increased LC3-II/LC3-I ratio and higher expression of p62/SQSTM1, was also observed under hemolytic conditions, with more conspicuous effects in males than in females (Fig. 2P and Q, S and T). Additionally, TG males had lower levels of both autophagy markers following hemolysis, while in TG females, changes were only seen in p62/SQSTM1 (Fig. 2P and Q, S and T).

3.3. CYB5R3 overexpression ameliorates changes in mitochondrial dynamics, respiratory capacity and mitophagy caused by massive intravascular hemolysis

Mitochondria play a fundamental role in AKI [42], and previous studies have demonstrated the relevance of CYB5R3 in regulating mitochondrial dynamics and function in muscle and liver [31]. Based on this, we analyzed the effect of CYB5R3 overexpression in our experimental model. Protein levels of selected markers of mitochondrial complexes were altered in hemolytic conditions, showing a reduction in complex I, III, and IV markers in both males and females (Fig. 3A–D). Mice overexpressing CYB5R3 displayed higher levels of complex I after hemolysis compared to WT mice under the same conditions, but no other genotype-dependent differences were observed for the remaining mitochondrial complexes, regardless of sex. To further assess mitochondrial function, we analyzed the respiratory capacity of isolated kidney mitochondria. Maximal oxygen consumption linked to complexes II+III+IV was higher in mitochondria isolated from TG males compared to their corresponding WT controls (Fig. 3E–G), both under basal conditions and following hemolysis. In contrast, females showed a different pattern: respiratory capacity associated with complexes II+III+IV increased in TG mice compared to WT under hemolytic conditions (Figs. 3H and 3J), whereas oxygen consumption solely linked to complex IV decreased in WT animals after hemolysis but remained unchanged in TG mice (Figs. 3I to 3J). These findings suggest that mitochondrial respiratory efficiency following hemolytic conditions may be influenced by CYB5R3 overexpression and that this effect depends on sex.

Fig. 3.

CYB5R3 overexpression improves mitochondrial dynamics, respiratory capacity and mitophagy caused by massive intravascular hemolysis. Hemolysis was induced by administering phenylhydrazine (150 mg/ body weight) to 12-week-old C57BL6 male and female mice of wild-type (WT) and transgenic CYB5R3-overexpressing (TG) genotypes (n = 4–9 per group). Mice were sacrificed 72 h after hemolysis induction. Quantification and representative image of the expression of mitochondrial complexes (OXPHOS) in males (A, B) and females (C, D) obtained by Western-Blot. Respiratory capacity associated with complexes II+III+IV of mitochondria isolated from kidney tissue in males (E) and in females (H). Respiratory capacity associated with Complex IV of mitochondria isolated from kidney in males (F) and in females (I). (G, J) Representative graph of respirometry assay in mitochondria isolated from kidney tissue in both sexes. Semi-quantification and representative Western-Blot image of fusion proteins (Mfn1, Mfn2), fission (DRP1), biogenesis-associated transcription factors (TFAM, Nrf1) and mitochondrial mass (TOM20) in males (K-M) and in females (N-P). (Q-R) Expression of mitophagy-associated proteins, PARKIN1 and PINK1, in males and in females (S-T). Results are expressed as mean ± SEM. ^a p < 0,05 compared to WT saline. ^b p < 0,05 compared to TG saline. ^c p < 0,05 compared to TG vs WT under hemolytic conditions. The main effect of hemolysis (regardless of genotype) is represented as “Hemolysis”, while the main effect of genotype (regardless of hemolysis) is represented as “WT-TG”. The interaction “hemolysis x genotype” is also indicated when relevant. * p < 0,05 in the analysis of overall effect (genotype, hemolysis or interaction).

Finally, we found that hemolysis increased the expression of proteins involved in mitochondrial fusion (Mfn1, Mfn2), fission (DRP1), mitochondrial biogenesis (TFAM and Nrf1), and mitophagy (PARKIN1 and PINK1) in both males and females (Fig. 3K to P). CYB5R3 overexpression reduced these changes. In contrast, levels of the outer mitochondrial membrane protein Tom20 decreased under hemolytic conditions, although this reduction was less significant in TG mice, regardless of sex (Fig. 3K to P).

3.4. CYB5R3 overexpression reduces inflammation and fibrosis in a chronic model of intravascular hemolysis

Sustained accumulation of hemoproteins in the kidney may promote long-term chronic renal damage [43,44]. Therefore, we examined the impact of CYB5R3 in a model of recurrent hemolysis. Our results showed an upregulation of the renal injury markers NGAL (Lcn2) and KIM1 (Hacvr1) after chronic exposure to hemolysis-derived products, although this effect was less pronounced in TG mice, regardless of sex (Fig. 4A and B, D and E). A similar trend was observed when we analyzed inflammation and fibrotic responses in the kidney. Thus, prolonged hemolysis led to increased expression of pro-inflammatory cytokines and macrophage infiltration in renal tissue, yet these effects were significantly attenuated in CYB5R3 overexpressing animals, in both males and females (Fig. 4C, F, G to I). Regarding fibrosis, we found that chronic hemolysis increased fibrotic mediators such as Collagen I and III and fibronectin, as shown by RT-PCR and Picrosirius Red staining. These fibrotic changes were notably reduced in TG mice, regardless of sex (Fig. 4J to R).

Fig. 4.

CYB5R3 overexpression decreases inflammation and fibrosis in a chronic model of intravascular hemolysis. Experimental model of intravascular hemolysis induced by weekly administration of phenylhydrazine (75–150 mg/kg mouse weight) in males and females of WT and TG genotypes (24 weeks old, n = 6–12 per group). Mice were sacrificed 30 days after induction of hemolysis. Gene expression of kidney damage markers NGAL (Lcn2) and KIM1 (Hacvr1) obtained by real-time PCR in males (A, B) and females (D, E). Gene expression of proinflammatory cytokines Il-6, Ccl2 and Ccl5 in males (C) and females (F). Representative image (G) and semi-quantification of F4/80 macrophage content in males (H) and females (I). Scale bar 50 μm. Semi-quantification and representative Western-Blot image of Fibronectin (Fn1) and Collagen I (Col1) in males (J, K) and females (M, N). Gene expression of Fn1 and Col1 in males (L) and in females (O) by real-time PCR. Representative image (P) and semi-quantification of Sirius Red staining in kidney sections of males (Q) and females (R). Results are expressed as mean ± SEM. ^a p < 0,05 compared to WT saline. ^b p < 0,05 compared to TG saline. ^c p < 0,05 compared to TG vs WT under hemolytic conditions. The main effect of hemolysis (regardless of genotype) is represented as “Hemolysis”, while the main effect of genotype (regardless of hemolysis) is represented as “WT-TG”. The interaction “hemolysis x genotype” is also indicated when relevant. * p < 0,05 in the analysis of overall effect (genotype, hemolysis or interaction).

3.5. Pharmacological induction of CYB5R3 reduces heme-induced oxidative stress in cultured tubular cells

Once we demonstrated the protective effect of CYB5R3 overexpression under hemolytic conditions, we aimed to assess whether pharmacological targeting of this enzyme could be effective against heme-induced renal injury. To do this, we stimulated renal cells with heme in the presence or absence of THII, a CYB5R3 activator. We also used the specific CYB5R3 inhibitor ZINC05626394 (Z) to verify whether CYB5R3 activation was responsible for the effects mediated by THII. To determine the optimal non-toxic concentrations for our in vitro study, we first performed toxicity assays on renal cells treated with a range of THII (0–2000 μg/ml) and Z (0–200 μM) concentrations (Supplemental Material Fig. 2A and B). Based on these results, we selected 60 μg/ml THII and 25 μM Z as the highest non-toxic concentrations for all subsequent experiments. We also found that the chosen THII dose increased CYB5R3 protein levels in cultured renal cells (Fig. 5A and B). In MCT cells, THII significantly reduced heme-mediated total (Fig. 5C) and mitochondrial ROS production (Fig. 5D), abated GSH loss (Fig. 5E) as well as lipid peroxidation (Fig. 5F). Consistent with these results, THII also prevented depletion of intracellular NAD⁺ and NADH levels caused by heme exposure (Fig. 5G to I). Similarly, THII reduced heme-induced loss of cell viability (Fig. 5J) and decreased gene expression of pro-inflammatory cytokines such as Il-6, Ccl2, and TNF-α (Fig. 5K to M). Finally, selective inhibition of CYB5R3 with ZINC05626394 abolished the protective effects of THII against heme-induced oxidative stress, inflammation, and cell death (Fig. 5C to M).

Fig. 5.

Pharmacological induction of CYB5R3 induction reduces heme-induced oxidative stress in tubular cells. Murine tubulo-epithelial cells (MCT) were stimulated with tetrahydroindenoindole (THII) and ZINC05626394 (Z) for 24 h. (A) Representative image of CYB5R3 expression in cells treated with THII (60 μg/ml). Scale bar 100 μm. Nuclei were labeled with DAPI (blue). (B) Semi-quantification and representative image of CYB5R3 expression after THII treatment (0–100 μg/ml). We subsequently performed experiments on MCT cells that were pretreated with THII (60 μg/ml) for 16 h and subsequently stimulated with heme group (5 μM) in the presence/absence of Z (25 μM). (C) ROS production (H2DCFDA) (D) representative image of mitochondrial ROS production (MitoSOX) in cells treated with heme group for 3 h. DAPI-labeled nuclei (blue). Scale bar 100 μm. Intracellular GSH levels (E) and lipid peroxidation (Bodipy) (F) obtained after stimulation with heme group (5 μM) for 2–16 h. (G-I) Levels of NAD⁺, NADH and NAD⁺/NADH ratio after 6 h of heme group stimulation. (J) WST-1 cell viability assay in cells stimulated with heme group for 24 h. (K-L) Gene expression of proinflammatory cytokines Il-6, Ccl2 and TNF-α in MCT cells stimulated with heme group for 6 h. Results expressed as mean ± SEM. * p < 0.05 compared to control cells, ≠ p < 0.05 compared to cells treated with heme group, ɸ p < 0.05 compared to cells treated with heme group and THII.

4. Discussion

CYB5R3 plays a key role in maintaining redox homeostasis [13]. However, its role in renal damage-associated to intravascular hemolysis has not been previously analyzed. In our study, transgenic mice overexpressing CYB5R3 exhibited reduced kidney injury, improved renal function, decreased oxidative stress, and lower rates of cell death during hemolysis. These findings are consistent with previous reports showing that CYB5R3 overexpression attenuated oxidative stress and inflammation, improves metabolism, and extends lifespan [30].

Several mechanisms may account for these protective effects. First, CYB5R3-driven reduction of CoQ may decrease Nox4-mediated production of O₂^.− at the OMM [29]. Since Nox4 is a key source of ROS in AKI associated to intravascular hemolysis [42], CYB5R3 overexpression could mitigate oxidative stress by reducing Nox4-mediated ROS production. Second, CYB5R3 regulates the redox state of sGC, an enzyme involved in nitric oxide (NO)-cGMP signaling pathway [23]. By maintaining sGC in its reduced (Fe²⁺) state, CYB5R3 preserves NO responsiveness and cGMP synthesis [45], thereby preserving vascular function and limiting oxidative stress [46,47]. This mechanism is particularly relevant in hemolytic conditions, where free heme reduce NO bioavailability. In addition, CYB5R3 maintains cytoglobin in its functional Fe²⁺ state, facilitating oxygen transport to the mitochondrial electron transport chain and contributing to vascular tone regulation [48]. CYB5R3 also cooperates with CYB5B at the OMM to promote the reduction of ascorbate free radicals to ascorbate, coupled with NADH oxidation in the cytosol [14]. This activity sustains cytosolic NAD⁺/−NADH levels, which are essential for redox balance and metabolic homeostasis [49], while allows regenerated ascorbate to reach the mitochondrial intermembrane space, where it donates electrons to cytochrome c, thus enhancing mitochondrial respiration. Furthermore, in biological membranes, CYB5R3 reduces CoQ, α-tocopherol, and ascorbate, thereby decreasing lipid peroxidation, preserving plasma membrane integrity [50–52], and protecting against ceramide-induced apoptosis [15,24–27]. Therefore, although not directly addressed in our study, these CYB5R3-associated functions may reduce heme-renal damage. However, it is important to note thatthe effects of CYB5R3 activity may be cell-type dependent and may not be uniformly beneficial across tissues. For example, in endothelial cells, CYB5R3 maintains α-globin in its reduced Fe²⁺ state, increasing its NO-scavenging capacity and thereby limiting vasodilation [53,54]. Therefore, while CYB5R3 overexpression may be protective in the kidney during hemolysis, its effects may vary depending on the cellular context and should be evaluated accordingly.

Importantly, oxidative stress, inflammation and cell death are hallmarks of AKI secondary to intravascular hemolysis [55]. The accumulation of hemoglobin and heme in tubular cells and podocytes is highly toxic, increasing ROS production, triggering ER stress and inflammation, and promoting cell death [8,42,56–58]. All these detrimental effects were evident in kidneys from WT mice following hemolysis induction, but were attenuated in CYB5R3-overexpressing mice, likely due to the antioxidant and redox-regulating functions of CYB5R3, as discussed above. In this line, it is important to note that CYB5R3 overexpression attenuated the induction of Nrf2 downstream targets under hemolytic conditions, as evidenced by the reduced HO-1 and Fth1 levels. CYB5R3 is transcriptionally regulated by Nrf2 [50]; however, a direct protein-protein interaction between these molecules has not been demonstrated. Given that the Keap1–Nrf2 system is highly sensitive to oxidative stress [59], it is likely that CYB5R3-Nrf2 modulation may be secondary to redox regulation rather than direct molecular interaction.

In addition, mitochondrial integrity appears to be a central target of CYB5R3-mediated protection. Mitochondria are particularly vulnerable to heme-induced toxicity, which promotes organelle dysfunction and structural damage [60,61], as reported in kidneys from patients with hemolysis-induced AKI [62]. In agreement with our results, renal tissues exposed to heme display reduced oxygen consumption, downregulation of mitochondrial complexes I and IV, increased mitochondrial fission, and activation of mitophagy pathways [60,62]. These changes may represent adaptive responses to limit oxidative damage and maintain mitochondrial function under hemolytic stress. In this regard, the reduced activation of these pathways in mice overexpressing CYB5R3 may indicate a lower degree of mitochondrial injury and a diminished need for compensatory responses due to the protective effects provided by CYB5R3. Interestingly, previous reports have documented that the interaction of CYB5R3 with Mitoregulin, a mitochondrial microprotein, specifically stimulates Complex I, most likely by providing a favourable lipid composition of the membrane [63]. Moreover, a role for Mitoregulin in preserving Complex I function has been confirmed in mouse kidney [64]. Therefore, the involvement of a Mitoregulin-Complex I interaction in the beneficial effects of CYB5R3 overexpression on mitochondrial function during hemolysis cannot be excluded. Nonetheless, it remains possible that some of the mitochondrial changes observed in kidneys from hemolytic mice may also contribute to damage rather than protection. Indeed, previous studies in sepsis-induced AKI have reported increased TFAM expression and NLRP3-mediated inflammation and cell death [65], whereas studies in ischemia-reperfusion models have shown that inhibition of Nrf1 promoted NLRP3-dependent cell death [66].

Repeated exposure to heme derivatives in the kidney can lead to permanent structural changes and CKD development, as reported in patients with chronic hemolytic disorders [67,68]. This chronic loss of renal function is associated with inflammation and fibrosis [69,70]. Intravascular hemolysis triggers acute inflammation via the TLR4/NF-κB axis [56] and in our study, chronic exposure to hemolytic products increased cytokine expression and macrophage infiltration. Similar effects were observed in other models of heme protein overload, such as rhabdomyolysis [71]. Interestingly, CYB5R3 overexpression reduced this chronic inflammatory response. Likewise, lower inflammation has been reported in double transgenic mice overexpressing CYB5R3 and NQO1, a related NAD(P)H dehydrogenase, in a liver cancer model [30]. Since oxidative stress and persistent inflammation drive renal fibrosis and subsequent CKD [70,72], the greater fibrosis observed in WT animals compared to TG animals suggests a protective role of CYB5R3 against fibrosis under hemolytic conditions. Supporting this hypothesis, CYB5R3 inhibition in alveolar epithelial cells increased fibrosis through ERK1/2 phosphorylation and TGF-β1 activation [73].

Sex-based differences also influence AKI risk, severity and recovery [74–76]. Males are generally more susceptible to AKI and show worse renal outcomes than females [77], mainly due to hormonal and metabolic differences affecting inflammation, antioxidant capacity and mitochondrial dynamics [76–80]. To our knowledge, no previous study examined sex-differences in hemolysis-induced AKI. Our data indicate that males experience higher renal dysfunction, inflammation, and cell death. This is in line with clinical observations in men with sickle cell disease showing faster decline of kidney function [81,82]. Although both sexes benefited from CYB5R3 overexpression, we observed distinct sex-based differences in several injury markers, particularly those related to mitochondrial function. Thus, TG males exhibited higher respiratory capacity associated with complexes II+III+IV than WT animals under both basal and hemolytic conditions, whereas TG females showed this enhancement only during hemolysis. In contrast, oxygen consumption linked to complex IV was higher in WT females at baseline but decreased under hemolytic stress in both genotypes. Thus, CYB5R3 overexpression modulates renal mitochondrial respiration in response to hemolysis in a sex-dependent way. Previous data from our group indicated that metabolic adaptations elicited in transgenic mice not only exhibit sexual dimorphism, but are also tissue-specific [31,32]. In skeletal muscle, CYB5R3 overexpression enhanced mitochondrial biogenesis and function, with more prominent results in females [33]. In the kidney, CYB5R3 overexpression resulted in changes in mitochondrial morphology, abundance and mitochondria-reticulum contact sites in distal tubular cells, mainly in females [34]. However, enhanced mitochondrial function in cardiomyocytes was predominantly observed in transgenic males [35], while, in liver, females and males exhibited distinct metabolic reprogramming which also involved systemic regulatory mechanisms [36].

THII is a pharmacological activator of CYB5R3 with potent antioxidant effects [30,83–85]. Previous studies indicated that THII supplementation induced antioxidant and detoxification enzymes and inhibited lipid peroxidation [86], and improved glucose metabolism [86] and mitochondrial function [40]. In our study, THII attenuated heme-induced damage by reducing total and mitochondrial ROS, lipid peroxidation, cytokine production and cytotoxicity, while restoring glutathione and NAD⁺ levels. Co-treatment with the CYB5R3 inhibitor ZINC05626394 abolished these protective effects, supporting that THII-renoprotective actions are mediated through direct targeting of CYB5R3.

Our study has several limitations. Although hemolysis was associated with alterations in autophagy-related markers, the absence of dynamic autophagy flux measurements precludes definitive conclusions regarding autophagic activity and limits the interpretation of the beneficial effects of CYB5R3 overexpression on autophagy in this pathological setting. On the other hand, although the phenylhydrazine-induced hemolysis model is widely used and effectively reproduces key features of intravascular hemolysis, phenylhydrazine itself can induce systemic oxidative stress, potentially confounding the interpretation of the in vivo findings. To address this limitation, we conducted complementary in vitro experiments with heme, confirming CYB5R3-dependent protection, thereby dissociating heme-mediated toxicity from potential off-target effects of phenylhydrazine.

In conclusion, our findings identify CYB5R3 as a key factor in kidney resilience during intravascular hemolysis. Genetic overexpression or pharmacological activation of CYB5R3 with THII significantly reduced heme-induced oxidative stress, maintained mitochondrial function, restored autophagic flux, and decreased tubular cell death, inflammation, and fibrosis. These results identify CYB5R3 as an essential regulator of renal redox balance and mitochondrial metabolism and highlight THII as a promising therapy for both acute and chronic kidney injuries caused by severe intravascular hemolysis. However, further studies are required to validate these results in clinical settings and to assess safety and efficacy in humans.

Abbreviations:

AKI

Acute kidney injury

ATF4

Activating transcription factor 4

BUN

Blood Urea Nitrogen

BSA

bovine serum albumin

Cat

Catalase

Ccl2

C-C motif chemokine ligand 2

Ccl5

Motif chemokine ligand 5

Col1

Collagen 1

Chop

C/EBP homologous protein

CLAP

Chymostatin, leupeptin, antipain, and pepstatin A

CKD

chronic kidney disease

CoQ

Coenzyme Q

CYB5

Cytochrome b₅

CYB5R3

Cytochrome b₅ reductase-3

DAB

3,3’-Diaminobenzidine

DAPI

4′-6-Diamidino-2-phenylindole

DRP1

Dynamin-related protein 1

DTT

Dithiothreitol

ER

Endoplasmic reticulum

FBS

Fetal bovine serum

Fn1

Fibronectine-1

Fth1

Ferritin

GADPH

Glyceraldehyde phosphate dehydrogenase

GSH

Glutathione

HO-1

Heme-oxygenase 1

Il6

Interleukin 6

ITS

Insulin-transferrin-selenite

KIM1

Kidney injury molecule-1

LC3

Microtubule-associated protein light chain 3

MCT

Murine renal proximal tubular epithelial cells

MDA

Malondialdehyde

Mfn1

Mitochondrial fusion protein 1 Mitofusin 1

Mfn2

Mitochondrial fusion protein 2 Mitofusin 2

MTT

3-[4,5-Dimethylthiazol-2-yl]-2,5-diphenyltetrazolium bromide

MtROS

Mitochondrial reactive oxygen species

NAD⁺

Nicotinamide adenine dinucleotide

NGAL

Neutrophil Gelatinase-Associated Lipocalin

Nox4

NADPH oxidase 4

Nphs1

Nephrin, Nrf2, Nuclear factor erythroid 2-related factor 2

PMSF

Phenylmethylsulfonyl fluoride

ROS

Reactive oxygen species

SOD1

Superoxide dismutase 1

SXBP1

Spliced X-box binding protein 1

Synpo

Synaptopodin

TBARS

Thiobarbituric acid reactive substances

TFAM

Mitochondrial transcription factor A

THII

Tetrahydroindenoindole

TMPD

N,N,N′,N′-tetramethyl-p-phenylenediamine

TUNEL

Terminal deoxynucleotidyl transferase dUTP nick-end labeling

TOM20

Translocase of the outer membrane 20

UPR

Unfolded protein response

Z

ZINC05626394

Appendix A. Supporting information

Supplementary data associated with this article can be found in the online version at doi:10.1016/j.biopha.2026.119437.

Data availability

Data will be made available on request.