## Bioinformatics analysis
The expression profile datasets of patients with MI were searched from the publicly available Gene Expression Omnibus (GEO) database (http://www.ncbi.nlm.nih.gov/geo). In the GEO dataset (GSE141512), a transcriptome profiling in peripheral blood mononuclear cells of 6 MI patients and 6 healthy individuals was performed and the expression of EPHB2 was calculated.
CTD and String databases were used to screen out the signaling pathways through which EPHB2 affected MI. The pathways associated with myocardial infarction were screened from the CTD database. The pathways affected by EPHB2 are retrieved from the String database. The intersection of the two was taken to obtain the signaling pathway that mediated EPHB2's effect on MI, which was prepared for further verification.
## Mice feeding and model establishment
C57BL/6J mice were purchased from Charles River Co., Ltd, China. Mice were allowed unlimited access to food and water throughout the study. The circadian was twelve hours in the light (7:00 am~ 7:00 pm) and the rest of twelve hours in the dark. The ambient temperature was 22±2°C. Relative humidity was 55±5%.
To knockdown EPHB2 specifically, AAV9-shEPHB2 was injected into mice via the jugular vein. The detailed procedure is as follows. Mice were anaesthetized with 1% isoflurane in oxygen, while viral solution (3×1011 vector genomes (vg)/mouse) was slowly injected via the jugular vein. MI model was established 4 weeks after administration.
To establish MI mouse model, left anterior descending (LAD) coronary occlusion was performed. As described above , The mice were anaesthetized by intraperitoneal injection of 1% phenobarbital sodium (30 mg/kg). After thoracotomy at the fourth intercostal space, the left thoracic cavity was exposed. The LAD coronary artery was ligated with a 6-0 suture at the lower edge of the left atrial appendage. The success of establishing the MI model was confirmed if there was an immediate color change on the heart surface after LAD ligation. Sham animals underwent the same surgical procedures, except the LAD was not occluded. Follow-up tests were conducted 4 weeks after the model was established.
Total RNA was extracted using Trizol reagent. then reverse transcribed into DNA using a RT-PCR Kit (Thermo #K1622), according to the manufacturer's instructions. Real-time PCR was performed with SYBR Green PCR kit (Thermo F-415XL) on a Real-Time PCR System (ABI-7500, USA) with GAPDH gene being used as internal control . The 2-ΔΔCt method was used to analyze the data . The primers for qPCR were purchased from Sangon Biotech Company (Shanghai, China) .primers were designed through Primer-BLAST tool . Each sample was tested in triplicate.
The sequences of the primers for EPHB2:
Forward primer, TGGTCGTCATTGCCATCGTA,
Reverse primer, TTCATGCCTGGGGTCATGTG.
The sequences of the primers for CTGF:
Forward primer, AGAACTGTGTACGGAGCGTG,
Reverse primer, GTGCACCATCTTTGGCAGTG.
The sequences of the primers for Collagen I:
Forward primer, CGATGGATTCCCGTTCGAGT,
Reverse primer, CGATCTCGTTGGATCCCTGG.
The sequences of the primers for Collagen III:
Forward primer, TGACTGTCCCACGTAAGCAC,
Reverse primer, GAGGGCCATAGCTGAACTGA.
The sequences of the primers for GADPH:
Forward primer, GGGTCCCAGCTTAGGTTCAT,
Reverse primer, CCCAATACGGCCAAATCCGT.
## Western blot assay
Western blot assay was used to detect protein expression levels. Proteins were extracted using RIPA total protein lysate (cat. no. P0013C, Beyotime, Inc, China) following the manufacturer's instructions. Proteins from each sample were separated by electrophoresis on 10% SDS-PAGE gels and transferred onto PVDF membranes (cat. no. FFP39, Beyotime, Inc, China). After incubation with blocking solution (cat. no. P0252, Beyotime, Inc, China), the PVDF membranes were separately incubated with anti EPHB2 antibody, anti Bax antibody, anti cleaved-caspase 3 antibody, anti Bcl-2 antibody, anti p-ERK antibody, anti ERK antibody, anti p-JNK antibody, anti JNK antibody, anti p-p38 antibody, anti p38 antibody and anti GAPDH antibody (cat.no. ab252935, ab32503, ab32042, ab182858, ab201015, ab184699, ab124956, ab179461, ab195049, ab31828 and ab8245, Abcam, Inc, USA,1:2000) at 4 °C overnight. After the blots were washed, they were incubated with HRP-conjugated Goat Anti-Rabbit IgG H&L secondary antibody(cat.no. ab6721, Abcam, Inc, USA,1:2000) for 1.5 h and detected by ECL like Western reagent (cat. no. P0018S, Beyotime, Inc, China) with a chemiluminescent imaging system (ChemiDoc MP, Bio-Rad, Inc, USA).
EPHB2 expression was detected followed standard immunohistochemical protocol as reported elsewhere . Briefly, paraffin-embedded slides (5 µM) were baked at 60°C, then dewaxed by xylene and rehydrated by fractionated ethanol. Antigen was extracted, and incubated with 3% H2O2 for 10 min. Then the slides were sealed with bovine serum for 1 h, and then anti EPHB2 antibody (cat. no. ab252935, Abcam, USA; 1:100 dilutions in 1% bovine serum albumin) were added and incubated at room temperature for 3 hours. The slides were covered with the secondary antibody and placed in a humid chamber for 1 hour, and then DAB was added to reveal the staining intensity. Sections were photographed using an Olympus IX81 microscope (Olympus, Inc, Japan).
## Echocardiographic examination
Cardiac physiological analysis of mice was performed as previously described [21,22]. In brief, mice were anesthetized with i.p. injection of 1% phenobarbital sodium (30 mg/kg). A sector scanner (Sonos 1500; Hewlett-Packard) equipped with a 12-MHz transducer was used to record two-dimensionally guided M-mode tracings to assess left ventricle wall thickness, left ventricle dimensions, and fractional shortening. Electrocardiogram recordings were acquired on anesthetized mice with a multichannel amplifier and were converted to digital signals for analysis (PowerLab system; ADInstruments).
## Hemodynamic measurements
Cardiac hemodynamic measurements were performed as described. Mice were anesthetized with i.p. injection of 1% phenobarbital sodium (30 mg/kg), and the right common carotid artery was isolated and cannulated with a 1.4-F micromanometer (Millar Instruments). Maximal values of the instantaneous first derivative of LV pressure (+dp/dt max, as a measure of cardiac contractility) and minimum values of the instantaneous first derivative of LV pressure (–dp/dt min, as a measure of cardiac relaxation) were recorded.
## The target pathway of Embelin was predicted by network pharmacology analysis
To predict the signaling pathways that embelin acts on, three networked pharmacological databases were used for analysis. Embelin was retrieved from CTD Database (Comparative Toxicgenomics Database, http://ctdbase.org/), STITCH Database (http://stitch.embl.de/) and BATMAN-TCM Database (Bioinformatics Analysis Tool for Molecular mechanism of Traditional Chinese Medicine, http://bionet.ncpsb.org.cn/batman-tcm/) to obtain the corresponding signaling pathways. Subsequently, "Kidney Cancer" was retrieved from the CTD Database and the corresponding signal pathway was obtained. Winn diagram was used to analyze the results of four retrievals and the intersection was selected as the most likely target pathway of embelin.
## Enzyme-linked immunosorbent assay (ELISA)
Blood samples were collected by extracting the eyeball. The whole blood was kept for 30 min and centrifuged at 2,000 × g for 20 min at room temperature, and the serum components were stored as aliquots at −80°C until further use. Lactate Dehydrogenase (LDH), Creatine Kinas(CK), TNF-α, IL-6, and IL-1β levels were measured using ELISA kits (cat.no. E-EL-M0419c, E-EL-M0358c, E-MSEL-M0002, E-EL-M0044c, and E-MSEL-M0003, Elabscience. Inc, China) according to manufacturer's instructions. Optical density values were measured with a microplate reader (Multiskan FC, Thermo scientific) in each well.
## Histological stain
HE staining was performed with a Hematoxylin-Eosin/HE Staining Kit (cat. no. G1120, Solarbio, Inc, China) according to the manufacturer’s instructions. In brief, after dewaxing, slices were incubated with haematoxylin for 2 min and thereafter stained with eosin.
TUNEL staining was performed using a TUNEL Apoptosis Assay Kit (cat.no. C1088, Beyotime, Inc, China). Myocardial sections of mice were obtained and blocked for 1 h at room temperature in TBS containing 0.5% Triton X-100, 0.1% Na-Citrate, and 5% normal goat serum. Sections were washed in PBS twice for 5 min and incubated with the TUNEL labeling solution for 1 h at 37°C. Sections were then washed with PBS for 10 min at room temperature and Images were captured.
Masson Trichrome staining was used to detect fibrosis with a Masson's Trichrome Stain Kit(cat.no.G1340, Solarbio, Inc, China). Briefly, alcohol (100% alcohol, 95% alcohol 70% alcohol) was used to de-paraffinize and rehydrate cardiac tissue. To improve staining quality, Bouin's solution was applied for 15 minutes at 56°C and rinse under tap water for 5-10. Next, immerged in Weigert's iron hematoxylin solution 10 minutes then wash with PBS buffer 5 minutes. Afterward, stained with Biebrich scarlet-acid fuchsin solution for 5 minutes, then washed in distilled water. Then sections underwent series of differentiation, dehydration in 75% and 90% alcoho and rinsing in tap water. and cleared in xylene and mounted.
Images were captured by an Olympus IX81 microscope (Olympus, Inc, Japan) in bright-field mode.
## Statistical analysis
The GraphPad Prism software (version 8.0) was used to perform Statistical analysis and mapping. One-way analysis of variance followed by a Bonferroni post-hoc test was used to analyze differences between groups. Data were considered significantly different when p < 0.05. Image analysis was conducted using ImageJ software (1.53c).