2.1 Animal model
Animal experiments were conducted in compliance with the National Research Council's Guide for the Care and Use of Laboratory Animals. And the Animal Care and Use Committee of Fuwai Hospital, Peking Union Medical College of Medicine approved our experiments to ensure humanitarian care for animals. We purchased forty-eight 3-week-old male C57BL/6N mice from Beijing HFK Bioscience Co., Ltd. and gave a twelve-hour light and dark cycle starting from 06:00 (am) to 18:00 (pm). Mice were randomly assigned after 2 weeks of adaptive feeding and the following three groups were used for experiments: (1) Control group (n = 16): Mice were given normal water and diet and received intraperitoneal administration of solvent (5% DMSO + 40% Peg300 + 5% Tween 80 + 50% ddH2O) 3 ml/kg from week 13 to 17, once a week. (2) HFpEF group (n = 16): A double-hit model has been designed, in which metabolic and mechanical stress work together resulting in HFpEF [19]. Briefly, C57BL/6N mice had unrestricted access to a HFD (D12492, Research Diet) starting from 5 weeks old. Meanwhile, L-NAME (N5751, Sigma) was supplied in drinking water (0.5 g/L) for HFpEF groups, after adjusting the pH to 7.4. The above solvent was administrated in the same manner from week 13 to 17, once a week. (3) HFpEF + Levo group (n = 16): According to the previous study [20], HFpEF mice were received 3 mg/kg levosimendan (S2446, Selleck) (Dissolve 1 mg of levosimendan in 50 µL of DMSO, subsequently dilute to 1 mg/mL with the above solvent) intraperitoneally from week 13 to 17, once a week.
On completion of treatment, all the animals were anesthetized and euthanized by cervical dislocation. The wet to dry weight ratio of the lung tissue weight (LW wet/dry) and the heart weight to tibia length ratio (HW/TL) were assessed.
2.2 Echocardiography
Transthoracic echocardiography was undertaken by employing the VisualSonics Vevo 2100 imaging system (FUJIFILM, Toronto, Ontario, Canada) at week 17. Systolic function was measured from the short-axis M-mode at mid-ventricular level. By pulse wave and tissue Doppler imaging, a four-chamber view of the apex was obtained to measure diastolic function at the mitral valve level. Cardiac function parameters were collected, including LVEF, fractional shortening (FS), the ratio between early mitral inflow velocity (E wave) and atrial systolic mitral inflow velocity (A wave) (E/A), the ratio between early mitral inflow velocity (E wave) and mitral annular early diastolic velocity (E' wave) (E/E'). The above parameters were measured three times and averaged.
2.3 Non-invasive blood pressure measurement
Systolic and diastolic blood pressure were monitored by a smart non-invasive sphygmomanometer (BP-2010A, Softron) by the tail-cuff method. The mice were placed on 37°C platforms and recorded under steady-state conditions. Recording blood pressure at least five measurements, maximum and minimum values were removed and the average was taken.
2.4 Treadmill exercise performance test
Mice ran on treadmill (FT-200, Techman, Chengdu, China) with a 10% slope. After sufficient acclimatization, mice began to climb uphill at a pace of 4 m/min for 5 minutes, then up to 9 m/min for 2 minutes. The pace was then increased by 2 m/min per 2 minutes till mice no longer continued to run within 10 s on the grid under electrical stimulation. Record running time and calculate running distance.
2.5 Histological analysis
Half of the hearts was transected and stained by hematoxylin and eosin (H&E) to compare the wall thickness among each group.
2.6 Western blot
The same amount of tissue proteins was isolated by NuPAGE Bis-Tris 4%-12% Gels (NP0326BOX, ThermoFisher) electrophoresis then transferred to nitrocellulose filter or polyvinylidene fluoride membranes. Membranes were incubated with primary antibodies including rabbit anti-system xc− (xCT) (1:1000, ab175186, Abcam), anti-glutathione peroxidase 4 (GPX4) (1:1000, ab125066, Abcam), anti-NADPH oxidase 4 (NOX4) (1:1000, 14347-1-AP, Proteintech), anti-ferroptosis suppressor protein 1 (FSP-1) (1:2000, 20886-1-AP, Proteintech), anti-β-catenin (1:1000, 17565-1-AP, Proteintech), anti-Cx43 (1:1000, ab235585, Abcam), anti-integrin alpha-5 (1:1500, 10569-1-AP, Proteintech), anti-occludin (1:1500, 27260-1-AP, Proteintech), anti-cytochrome C (1:1500, 10993-1-AP, Proteintech), and anti-GAPDH antibody (1:5000, ab181602, Abcam), at 4 ℃ overnight, then incubated with the targeted secondary antibodies for one hour at ambient temperature. Visualization was then performed with a ECL Plus Detection Reagent (170–5060, BIORAD) and the density of the strips was analyzed using Image-J software.
2.7 Immunofluorescence staining
Cardiac tissue paraffin sections were boiled in EDTA (PH = 9.0) to retrieve antigens, blocked and permeabilized in goat serum containing Triton X-100 for ninety minutes. The used primary antibodies were mouse anti-4-hydroxynonenal (4-HNE) (1:25, ab48506, Abcam), anti-cTnT (1:500, ab8295, Abcam), rabbit anti-CD31 (1:200, ab222783, Abcam), anti-β-catenin (1:100, 17565-1-AP, Proteintech), anti-Cx43 (1:100, ab235585, Abcam), anti-integrin alpha-5 (1:100, 10569-1-AP, Proteintech), anti-occludin (1:100, 27260-1-AP, Proteintech), and anti-mitofilin (1:200, 10179-1-AP, Proteintech. After being incubated the primary antibodies at 4 ℃ overnight, the slides were incubated with secondary antibody including goat anti-mouse/rabbit IgG (H&L) (Alexa Fluor® 488/594) (1:500, ab150113/ab150077/ ab150116, Abcam). The cardiac sections were directly incubated with Alexa Fluor® 594 mouse anti-alpha smooth muscle actin (α-SMA) (1:200, ab202368, Abcam). Nuclei were counterstained using 4′,6-diamidino-2-phenylindole (DAPI) (ZLI-9557, ZSGB-bio). Images were visualized under confocal microscope (SP8, Leica, USA). Five fields from each section have been chosen randomly to carry out statistical evaluation using Image-J. Microvascular density was quantified via the number of CD31 + circles per field. The relative content of 4-HNE was analyzed by fluorescence intensity. The endothelial barrier structure of arterioles was shown by double staining of β-catenin/Cx43/integrin alpha-5/occludin and α-SMA. The myocardial mitochondrial structure was determined by double staining of mitofilin and cTnT.
The cross-sectional area of CMs was determined by wheat germ agglutinin (WGA) staining. After deparaffinization, slides would be incubated with Alexa Fluor™ 594 conjugate of anti-WGA antibody (1:500, W11262, Invitrogen) for 30 min at room temptation. The average cross-sectional area of CMs was calculated using Image-J.
2.8 Detection of reduced glutathione (GSH)/oxidized glutathione (GSSG)
GSH/GSSG ratio in mouse heart tissue was determined by GSSG/GSH quantitative kit (G263, Dojindo) referring to the manufacturer's instructions.
2.9 Malondialdehyde (MDA) content assay
The heart tissue samples were homogenized and the supernatant was collected. The concentration of protein was calculated with the BCA Protein Assay Reagent (P0012S, Beyotime). MDA as an index of lipid peroxidation was measured using MDA Assay Kit (S0131S, Beyotime) according to the protocol provided by the manufacturer. MDA content in heart samples can be calculated by dividing the measured MDA content by the protein concentration.
2.10 Iron assay
To measure the Fe2+ and total iron content, the cardiac tissues were weighed, homogenized and centrifuged (16,000g) for 10 min. The supernatant for quantitative determination of Fe2+ and total iron content was collected, then measured using Iron Assay Kits (I291, Dojindo) according to manufacturer’s instruction.
2.11 Evaluating the permeability of microvascular endothelium
Endothelial permeability has been evaluated by injecting Evans Blue dye (EB, E2129, Sigma) [21]. Briefly, at 17 week, four mice were randomly selected from each group. Mice were given 1% EB/0.5 mL physiological saline solution per 100 g body weight through the tail vein. One day later, the heart was collected after PBS perfusion, then frozen sections were made and coated with DAPI. The fluorescence of Texas Red was measured under the fluorescence microscope.
2.12 Flow cytometry of heart single cell suspension
Single-cell suspensions from adult mouse hearts were prepared as previously described [22]. Briefly, hearts were cut into pieces about 1 mm in diameter and placed in RPMI1640 medium containing collagenase type II (1.5 mg/mL) (LS004176, Worthington), elastase (0.25 mg/mL) (LS006363, Worthington) and Deoxyribonuclease I (0.5 mg/mL) (LS006342, Worthington), and enzymatically digested at 37 ° C for 1 h. Digested cell mixture was filtered through 70-µm cell strainers and washed. The suspensions were incubated with antibodies or probes used as follows: rabbit anti-cTnT antibody (ab209813, Abcam), 2,7-dichlorofluorescein diacetate (DCFH-DA) probe (S0033M, Beyotime), dihydroethidium (DHE) probe (S0063, Beyotime). ROS in CMs were detected by DCFH-DA and cTnT double staining. Superoxide anion was detected using DHE fluorescent probe. Flow cytometry was performed with the Accuri C6™ Flow Cytometer (BD Biosciences, CA, USA) to measure fluorescence intensity.
2.13 Statistic analysis
GraphPad Prism version 8 was available for analysis of statistical data. One-way analysis of variance (ANOVA) and post hoc Tukey test were used for the continuous variables between three groups. Chi-square test with fisher’s exact test was used for the categorical variables. All the data were expressed as means ± standard error (SEM). Statistical significance was set at *p < 0.05, **p < 0.01, ***p < 0.001, ****p < 0.0001.