Loss of cardiomyocyte CYB5R3 impairs redox equilibrium and causes sudden cardiac death

Sudden cardiac death (SCD) in patients with heart failure (HF) is allied with an imbalance in reduction and oxidation (redox) signaling in cardiomyocytes; however, the basic pathways and mechanisms governing redox homeostasis in cardiomyocytes are not fully understood. Here, we show that cytochrome b5 reductase 3 (CYB5R3), an enzyme known to regulate redox signaling in erythrocytes and vascular cells, is essential for cardiomyocyte function. Using a conditional cardiomyocyte-specific CYB5R3-knockout mouse, we discovered that deletion of CYB5R3 in male, but not female, adult cardiomyocytes causes cardiac hypertrophy, bradycardia, and SCD. The increase in SCD in CYB5R3-KO mice is associated with calcium mishandling, ventricular fibrillation, and cardiomyocyte hypertrophy. Molecular studies reveal that CYB5R3-KO hearts display decreased adenosine triphosphate (ATP), increased oxidative stress, suppressed coenzyme Q levels, and hemoprotein dysregulation. Finally, from a translational perspective, we reveal that the high-frequency missense genetic variant rs1800457, which translates into a CYB5R3 T117S partial loss-of-function protein, associates with decreased event-free survival (~20%) in Black persons with HF with reduced ejection fraction (HFrEF). Together, these studies reveal a crucial role for CYB5R3 in cardiomyocyte redox biology and identify a genetic biomarker for persons of African ancestry that may potentially increase the risk of death from HFrEF.

a 28-guage needle were placed around the aortic arch. The skin was closed using a continuous suture pattern. Mice were allowed to recover and characterized at 7, 14 or 27 days post-surgery via echocardiography as previously described. (1)

Lipid analysis
Solvents used for extractions and mass spectrometric analyses were from Burdick and Jackson (Muskegon, MI). Heart samples (≈ 15-100 mg) were homogenized in a FastPrep-24™ 5G blender in 0.5 ml phosphate buffer 50 mM pH 7.4 in the presence of 1 mg butylated hydroxytoluene (BHT).
An aliquot of the homogenate (200 μl) was diluted with 300 μl of the same buffer and lipids extracted using the Bligh and Dyer method (2). The organic extracts were dried under N2 flow, resuspended in 200 μl methanol, and further diluted 100-fold in the same solvent.

Hematological analysis
Blood (5% v/v EDTA) was collected via left heart puncture in isoflurane-anesthetized mice and subjected to complete blood count analysis using a HemaVet 950 (Drew Scientific Inc, Miami Lakes FL).

Telemetry
Mice were anesthetized with isoflurane, and telemetry units (HDX-11, Data Sciences International [DSI]) were surgically implanted with leads placed on either side of the chest. Mice were allowed to recover for 2 weeks post-surgery before telemetry units were turned on. ECG recordings and heart rate were then measured continuously to assess baseline physiological BP. ECG recordings and heart rate measurements were recorded by Ponemah Software (DSI) and exported into Microsoft Excel (version 16.16.6) for analysis.

Open chest hemodynamics
Mice were anesthetized using inhaled isoflurane and ventilated via tracheostomy with positive pressure ventilation. Maintenance of anesthesia during the procedure is achieved using 3% isoflurane gas and between 21% and 100% oxygen delivered via inhalation tube. The mouse was intubated and placed in a supine position for longitudinal dissection of the sternum to expose the heart and great blood vessels. The left ventricle was cannulated with a 1.5F micro pressurevolume catheter (Transonic FTH-1211B-0018). The right ventricle was also catheterized, using real-time pressure tracings to identify the location for measurement of RV systolic pressure. Both catheters were secured in place with ligation sutures for recording 10 minutes of LVSP and RVSP, heart rate, stroke volume, and cardiac output. A miniature ultrasound flow meter/probe placed at the level of the pulmonary artery and ascending aorta provided Doppler ultrasound spectrographic data for computation and calculation by the Pulsed Doppler Ultrasound Digital Signal Processor Workstation (Indus Instruments).

Optical mapping of calcium transience in perfused hearts
Mice were euthanized with euthasol, the heart dissected and immediately cannulated (>60 s) at the aorta, perfused with physiological Tyrode's solution (37 o C, pH 7.4; 112 mM NaCl, 25 mM NaHCO3, 5 mM KCl, 1.8 mM CaCl2, 1.2 mM MgSO4, 1 mM KH2PO4, 50 mM D-glucose; bubbled with 95% O2 and 5% CO2) and mounted on a custom-designed chamber for optical mapping in a Langendorff apparatus. The fluorescent calcium-indicator (40 µl of 0.5 mg/ml Rhod-2/AM in DMSO) was added as a bolus injection in the bubble trap proximal to the cannula, which served as a compliance chamber. Light from a 100-W tungsten lamp was collimated, passed through a 530 ± 30 nm interference filter, reflected by a 500 nm dichroic mirror and focused on the heart with a Nikon camera lens. To measure cytosolic free Ca 2+ , fluorescent light from the heart was collected by the same lens, split by a 620 nm dichroic mirror to focus heart images at wavelengths of 570-595 nm on a CMOS camera (100 x 100 pixel, UltimaOne, Scimedia, Ltd. Tokyo Japan). Each camera scanned at 1,000 frames/s for 5 ms scan time, using an optical magnification to yield a spatial resolution of 150x150 µm 2 per pixel. Arrhythmia vulnerability was tested by varying the rate of stimulation from the basic sinus rhythm rate (typically 200 ms cycle length (CL)) to the fastest rhythm rate tolerated by the heart (typically 75 ms CL). Programmed stimulation was also used to test arrhythmia vulnerability by applying 10 impulses on the RV at S1-S1 intervals of 200 ms CL, followed by a premature impulse S2, then S1-S2 decreases by 10ms steps while S1-S2>100 ms, then decreasing S1-S2 by 2ms steps until failure of the heart to capture (refractory period) or S2 triggered arrhythmia (3)(4)(5).

Transmission electron microscopy
Heart tissue was fixed for 10 min in glutaraldehyde (2.5% in 0.1 M PBS, pH 7.4) and then cut into 1 mm 3 blocks for additional fixing (50 min). Tissue was washed (3 x 15 min) in PBS (0.1 M, pH 7.4) and then soaked in osmium tetroxide (1%) containing 1% potassium ferricyanide for 1 hr.
Wash cycles were repeated as before. Tissue was then dehydrated by a series of 15-min washes in increasingly concentrated ethanol in PBS: 30% twice, 70% once, 90% once, and 100% thrice.

Heart histology
Fixed heart tissue was also stained with a Masson's Trichrome Kit or with hematoxylin & eosin staining, as previously described. (6)

Mitochondrial-to-genomic DNA ratio quantification
Total DNA from whole heart tissue was prepared as described in (7). mtDNA/gDNA ratio compared ΔΔCT of mouse mt-nd1/TBP ratios. Reactions of DNA, TaqMan Universal Master Mix (Applied Biosystems, Waltham, MA), and 5 μM primer (10 μl multiplex reactions in triplicate) were performed with standard curve to ensure linearity of assay and test dilution effects on rations in StepOnePlus thermocycler as described in (8).

MRI sample preparation and acquisition
At the end point, following euthanasia of the mouse, the blood in the heart was flushed out by retrograde perfusion of phosphate-buffered saline (PBS) through the abdominal aorta for 2 minutes, followed by retrograde perfusion of 4% paraformaldehyde for 5 minutes for tissue fixation. The heart was then dissected out of the chest and fixed by immersion in 4% paraformaldehyde at 4 o C for 2 days, followed by rehydration in PBS at 4 o C for 2 days before transfer to 10% neutral buffered formalin. For MRI, the fixed mouse heart was removed from formalin and lightly blotted dry with paper towel to remove excess water on the heart surface.

Immunohistochemistry of oxidative stress and tunnel staining in mouse heart sections
Hearts were deparaffinized using two 5-min incubations in xylene, followed by serial incubations ImageJ was used for quantifying Hypoxyprobe-positive cells relative to total cells.

Generation of soluble CYB5R3 wildtype and CYB5R3 T117S recombinant protein
The D23 WT and D23 CYB5R3 T117S proteins were expressed as C-terminal fusions in the pET-Smt3 vector in E. coli BL21 (DE3) cells (Genlantis) (12). Donor cultures grown in LB at 37ºC overnight were diluted 1:50 into LB and grown at 37ºC until reaching an OD600 of 0.5-0.7.
Expression cultures were induced with 1 mM IPTG and protein was expressed for 4-6 hours.
Cells were harvested by centrifugation and lysed by sonication in 50 mM Tri-HCl, 150 mM NaCl, 10% v/v glycerol, pH 7.4 on ice. Soluble protein was separated by centrifugation (30,000 x g, 4ºC, 30 min) and the supernatant was applied to a NiNTA Fast Flow column (GE Lifesciences, USA) equilibrated with lysis buffer. The column was washed in lysis buffer supplemented with 60 mM imidazole and the protein was eluted in lysis buffer supplemented with 400 mM imidazole.
Fractions containing the CYB5R3 protein were combined and concentrated using centrifugal concentrators (Amicon Ultra 10K, Millipore, USA). The His-tagged Smt3-specific protease, Ulp1, was added to cleave the fusion and the mixture was applied to a Sephacryl S-300 HR size exclusion column (GE Lifesciences, USA) equilibrated in lysis buffer.
Fractions containing monodisperse protein were collected and subsequently reapplied to the NiNTA column to capture the His-tagged Smt3 and Ulp1 proteins. Flow-through containing the CYB5R3 protein was collected, and a final buffer exchange was performed using a Superdex S75 size exclusion column (GE Lifesciences, USA). Protein was concentrated, flash-frozen in liquid N2 and stored at -80ºC prior to use.

CYB5R3 activity assay
Recombinant D23 CYB5R3 wildtype or D23 CYB5R3 T117S ± CYB5B were each diluted to 0. HEK 293FT cells (Thermo Fisher) that were stably knockdown with CYB5R3 (as previously described (13)) were grown to confluency and Lipofectamine 3000 (ThermoFisher Maximum intensity projections were created and ImageJ was used to perform quantification in separate channels.