Multi-Target Cardioprotection from Berberis kaschgarica Extract in Zebrafish via AMPK Pathway Activation

TL;DR

A plant extract from Berberis kaschgarica protects zebrafish hearts by reducing stress and inflammation, possibly through activating a key energy pathway.

Contribution

This study reveals the cardioprotective mechanisms of Berberis kaschgarica extract in heart failure via AMPK pathway activation.

Findings

BKRE contains 14 bioactive compounds with antioxidant properties and drug-like qualities.

BKRE improves heart function in zebrafish by reducing oxidative stress and inflammation.

BKRE activates the AMPK-PPARα-PGC-1α pathway, offering multi-target cardioprotection.

Abstract

Background: Heart failure (HF) has a complex pathogenesis involving oxidative stress, inflammation, and energy metabolism disorders, and requires multi-target agents. Berberis kaschgarica Rupr. (BKR) is used in Uyghur folk medicine to improve cardiovascular health, but its cardioprotective mechanisms against HF remain unclear. Methods: UPLC-MS/MS was used to identify BKRE components; DPPH/ABTS assays evaluated antioxidant activity. The MTC of BKRE was determined in zebrafish, and its effects on ISO-induced HF zebrafish were assessed via cardiac function, apoptosis, oxidative stress, and inflammation indicators. Network pharmacology, molecular docking, transcriptomics, and qRT-PCR clarified targets and pathways. Results: BKRE contained 14 bioactive flavonoids/alkaloids with favorable drug-likeness, showing concentration-dependent DPPH and ABTS scavenging. In HF zebrafish, BKRE (5/10/20 μg/mL) dose-dependently improved cardiac function, inhibited apoptosis, reduced ROS and TNF-α/IL-6, restored GSH/T-SOD, activated the AMPK-PPARα-PGC-1α pathway by binding ALOX5/NQO1, etc. Conclusions: BKRE exerts multi-mechanistic cardioprotective effects, validating BKR’s ethnopharmacological value and highlighting it as a promising HF agent/functional food.

Figures

Full text

1. Introduction

Heart failure (HF) is a serious pathological condition with diverse presentation. It is a major cause of death and is the most common clinical syndrome resulting from various cardiovascular diseases. One of its most notable characteristics is the significant economic burden that it causes to societies, alongside the increased morbidity and mortality rates. Thus, HF is considered a major global public health issue of our time. According to the epidemiological data, the prevalence of chronic heart failure (CHF) is more than 64 million people worldwide. After a great deal of progress in terms of diagnostics, therapeutics, and clinical care, the rates of readmission and mortality associated with HF have remained at a very high level, and therefore, there has been no noticeable decline in these rates over the last few decades [1,2]. The first piece of evidence presented by the China Hypertension Survey shows how serious HF is, as the main cause of a heavy disease burden in the country. The burden of HF in China is evaluated to be 1.3% approximately among people aged 35 years and over. This indicates that there are around 8.9 million patients across the country. The result is that China is becoming the country with the highest HF burden in the world [3,4]. Finding new and effective therapeutic targets is therefore a critical issue for the mitigation of the disease and the improvement in clinical outcomes for patients. The pathological profile of CHF is very complex, as it is characterized by increased activation of the sympathetic nervous system, secretion of dysregulated humoral factors, overexpression of pro-inflammatory cytokines, impaired immune homeostasis, and interstitial cardiac remodeling [5,6]. During cardiac ischemia and hypoxia, the sympathetic nervous system (SNS) is rapidly activated, and then it interacts with the adrenergic receptors that are present on cardiac tissues. This acute compensatory response provides a temporary cardioprotective effect, which is the main reason for maintaining cardiac function during times of acute stress. But as cardiac function reaches the limit of its reserve, prolonged SNS activation still continues. However, it no longer protects, and instead causes further damage to cardiac function [7].

Isoproterenol (ISO) is basically a primary agent that is largely employed in the study of cardiovascular diseases. As a poison of the heart, it can bring about a wide range of cardiac disorders such as cardiac hypertrophy, arrhythmias, and myocardial infarction. In a word, the long-term use of a low dose of ISO causes chronic progressive cardiac dysfunction in animal models, and a short exposure to a high dose of ISO can be used to mimic acute cardiac injury in vivo. Eventually, the model of cardiotoxicity induced by ISO became so widely used in the experimental platform that it is now a reference for the investigation of acute and chronic cardiac damage, as evidenced by ample reports from both local and international research communities [8,9,10]. High-dose ISO intraperitoneal injection increases myocardial oxygen consumption greatly, excessively stresses the heart, and eventually causes myocardial injury accompanied by histopathological changes similar to those of myocardial infarction. For instance, ISO administration in animal models causes a similar range of cardiac pathological changes that strongly correlate with oxidative stress, inflammatory responses, and apoptosis signaling pathways. High-dose ISO treatment results in the dysregulated upregulation of myocardial injury biomarkers and cardiac dysfunction, and thus, the pathological phenotypes of acute myocardial infarction in humans are replicated, as studies have consistently evidenced. Because it has these particular features, ISO is now labeled an experimental model of the gold standard for evaluating the effectiveness of preventive interventions and protective strategies against myocardial ischemia, infarction, and heart failure [11,12]. Over-accumulated in vivo, ISO perturbates the generation of endogenous radicals that are able to trigger a redox cascade. Free radicals are highly reactive species that contain at least one unpaired electron and form the bioactive derivatives of reactive oxygen and nitrogen species (RONS). Among them are peroxynitrite anions, superoxide anion radicals, and hydroxyl radicals. The operon for ROS overproduction exacerbates the impairment of myocardial functions and increases the etiopathogenesis of cardiovascular diseases [13]. Tight ROS generation, on the other hand, is instrumental for numerous regulatory signaling cascades. Nevertheless, the overproduction and accumulation of radicals bring about an imbalance in cardiac homeostasis. Excess ROS derive their damaging potential from DNA damage, protein and lipid peroxidation, and cellular dysfunction, in which the genesis of pathological cardiac injury has been implicated. The overproduction of reactive oxygen species (ROS) not only mediates oxidative stress but also contributes to cardiomyocyte apoptotic cell death [14,15].

Berberis kaschgarica Rupr. (BKR) is a perennial deciduous shrub of the Berberis family. Berberidaceae is mainly found scattered around forest edges, mountain slopes, valley terraces, and areas that are covered with shrubs. Berberis kaschgarica is a species from the Kashgar area of Xinjiang, China, and is recognizable by its bright red berries. Wild BKR grows in large, wide-ranging areas, but there are only a few places in which it has been grown artificially. Its fruits have been used as ethnomedicine for a very long time and are now considered a source of great ethnomedicinal and nutraceutical potential. Mili is the local name for the dried berries, which are frequently made into a herbal tea and taken as a cure as well as for food in the local area. Based on our previous research [16], we have observed preliminary evidence of BKR’s antihypertensive and cardiovascular protective potential; however, its direct cardioprotective effects against HF and the underlying molecular mechanisms remain unclear; this constitutes a critical research gap to be addressed. Therefore, the core objectives of the present study are as follows: (1) To evaluate the in vivo cardioprotective efficacy of BKR in an ISO-induced HF zebrafish model. (2) To identify the key bioactive components of BKR via UPLC-MS/MS. (3) To systematically elucidate its multi-target mechanisms through integrated approaches including network pharmacology, molecular docking, transcriptomics, and qRT-PCR. The current project aims to clarify the protective role of BKR in the context of heart failure using an ISO-induced heart failure model in zebrafish. The choice of this model was decided after weighing up several factors because it offers unique advantages for cardiovascular research. One of the reasons is that zebrafish are genetically very close to humans; this increases the applicability of our findings to humans. Moreover, their clear body allows the functioning of the heart to be seen and recorded under a microscope throughout the experiment. Additionally, this model is very compatible with the requirements of high-throughput drug screening, in which the testing of several compounds can be completed at a fast rate. Furthermore, the cardiac damage brought about by ISO in zebrafish mirrors the essential pathological features of human heart failure, such as ventricular dysfunction, excessive production of reactive oxygen species, and programmed cell death of heart muscle cells. These similarities make the zebrafish model a powerful tool in the investigation of the therapeutic potential of BKR [10,17].

2. Materials and Methods

2.1. Animals and Experimental Design

2.1.1. Animals Breeding

Three-month-old AB wild-type adult zebrafish were purchased from the Institute of Hydrobiology, Chinese Academy of Sciences (Wuhan, China). Male and female fish were paired 1:1 in transparent aquariums with a baffle, which was removed the next morning for natural spawning. Larvae were collected 5 days post-hatching at 28 °C, and those with normal development were selected via a stereomicroscope for subsequent experiments.

To avoid observer bias, all morphological observations and quantitative measurements followed a unified single-blinding protocol. Experimental groups were labeled with random alphanumeric codes instead of explicit names during sample preparation, imaging, and analysis. Assessments were performed by an independent researcher unaware of group assignments, and codes were decoded only after full data collection and analysis to ensure result reliability. All animal procedures were approved by the Laboratory Animal Ethics Committee of Xinjiang Medical University (Ethical Approval No.: IACUC-20221004-192 approved on 4 October 2022), complying with animal research ethical guidelines.

2.1.2. Maximum Tolerated Concentration (MTC) Assay

This pre-experiment aimed to determine the safe dose range of BKRE. Zebrafish embryos (6 hpf) were randomly divided into 6 groups (n = 3 biological replicates per group, 10 embryos/well): negative control group (maintained in standard E3 medium) and 5 BKRE concentration groups (1, 10, 50, 200, 1000 μg/mL). The well was defined as the independent experimental unit. Embryos were continuously exposed to the respective BKRE concentrations for 72 h (from 6 to 78 hpf), with daily medium renewal. At 78 hpf, lethality rate, malformation rate (characterized by body axis curvature and edema), and heart rate were recorded to confirm the MTC (20 μg/mL). Mortality was monitored daily during the exposure period, and mortality rates were calculated as (number of dead larvae/total larvae per group) × 100, presented as mean ± standard error of mean (SEM). Cardiac rate was measured after anesthetizing larvae with 0.01% tricaine methane sulfonate (MS-222) to minimize movement without affecting cardiac function. Cardiac beats were counted manually for 20 s per larva, with 3 replicates per individual, under a stereomicroscope (SZ61, Olympus, Japan), and the average value was calculated for each larva.

2.1.3. ISO-Induced HF Model and Therapeutic Assay

Zebrafish larvae (24 hpf) were randomly assigned to 5 groups (n = 3 biological replicates per group, 10 larvae/well): control group (E3 medium), ISO model group (0.02 mg/mL ISO), and 3 BKRE treatment groups (0.02 mg/mL ISO + 5/10/20 μg/mL BKRE). The well was defined as the independent experimental unit. All groups were continuously exposed for 72 h (24–96 hpf), with daily medium renewal. At 96 hpf, cardiac function, apoptosis, ROS, and molecular markers were detected.

2.2. Main Reagents and Instruments

Main reagents included: BKRE, self-prepared with voucher specimen No. XJMUYXY20201020; ISO from Sigma-Aldrich, St. Louis, MO, USA; DPPH assay kit and ABTS assay kit from Biosharp, Hefei, Anhui, China; GSH assay kit (Catalog No. A061-1), CAT assay kit (Catalog No. A007-1), NO detection kit (Catalog No. AN121-1), and BCA protein assay kit from Nanjing Jiancheng Bioengineering Institute, Nanjing, Jiangsu, China; T-SOD assay kit (Catalog No. E-BC-K019-M), TNF-α ELISA kit (Catalog No. E-EL-M0049), and IL-6 ELISA kit (Catalog No. E-EL-M0044) from Elabscience, Wuhan, Hubei, China; Annexin V-FITC Apoptosis Detection Kit (Catalog No. A13201) and MF20 antibody from Thermo Fisher Scientific, Waltham, MA, USA; DCFH-DA fluorescent probe (Catalog No. D6883) and MS-222 (tricaine methane sulfonate) from Sigma-Aldrich, St. Louis, MO, USA; TRIzol reagent from BioLit, SRL, New Delhi, India; PrimeScript™ RT reagent kit from Takara Bio Inc., Kusatsu, Shiga, Japan; and KAPA SYBR FAST one-step kit from Kapa Biosystems, Wilmington, MA, USA. Main instruments included: UHPLC system (SHIMADZU LC30) coupled with TripleTOF 6600 mass spectrometer from AB Sciex, Framingham, MA, USA; ACQUITY UPLC® HSS T3 column (2.1 mm × 100 mm, 1.8 μm) from Waters Corporation, Milford, MA, USA; stereomicroscope (SZ61) from Olympus, Tokyo, Japan; microplate reader from Thermo Fisher Scientific, Waltham, MA, USA; spectrophotometer from Shimadzu, Kyoto, Japan; Illumina NovaSeq 6000 sequencer from Illumina, San Diego, CA, USA; Light Cycler 96 qRT-PCR system from Roche Diagnostics GmbH, Mannheim, Germany; GraphPad Prism 10.0.0 software from GraphPad Software, San Diego, CA, USA; and ImageJ v1.8.0 software from National Institutes of Health, USA.

2.3. Preparation of Extracts from the Fruits of Berberis kaschgarica Rupr.

The fruits of B. kaschgarica were collected in Akto County, Kashgar Prefecture, Xinjiang Uygur Autonomous Region, China, in October 2023. The plant material was botanically identified by Prof. Palida Abulizi (Department of Natural Medicinal Chemistry and Pharmacognosy, School of Pharmacy, Xinjiang Medical University). A voucher specimen (No. XJMUYXY20201020) has been stored in the Ethnical Herbs Specimen Museum of Traditional Chinese Medicines at the same university. The collected fruits were air-dried at ambient temperature, pulverized into a fine powder, and sieved through a 20-mesh sieve. The powdered material was then extracted using the method described in Reference [16], and the resulting extract was stored in a cool, dark environment until further use.

2.4. LC-MS/MS Method for Analyzing Bioactive Components in BKR Fruit Extract

Precisely 100 mg of fruit extract from Berberis kaschgarica Rupr. was weighed out, 1 mL of ultrapure water was introduced to the sample, and vortex mixing was performed until complete dissolution of the extract was achieved, after which 3 mL of ethanol was supplemented; subsequent revortexing was conducted to ensure homogeneous mixing prior to subjecting the mixture to ultrasonic extraction at ambient temperature for 10 min, followed by static incubation at 4 °C for 12 h, centrifugation at 4000× g and 4 °C for 10 min to harvest the resulting supernatant, and nitrogen blow-down to evaporate ethanol from the supernatant to dryness, yielding a purified concentrated solution to which ultrapure water was added to bring the final volume to 1 mL prior to mass spectrometry (MS) detection, with the resulting solution passed through a 0.22 μm membrane filter and preserved for subsequent experiments; for MS analysis, the injection volume was set at 10 μL with each sample analyzed in three technical replicates, all samples were maintained in a 4 °C autosampler throughout the entire detection process, ultra-high-performance liquid chromatography (UHPLC)-based separation was conducted using a SHIMADZU LC30 system coupled to an ACQUITY UPLC^®^ HSS T3 column (2.1 mm × 100 mm, 1.8 μm; Waters Corporation, Milford, MA, USA) with the operational parameters configured as follows: column temperature 40 °C, flow rate 0.3 mL/min, mobile phase consisting of Phase A (0.1% formic acid aqueous solution) and Phase B (0.1% formic acid in acetonitrile), and a gradient elution protocol of 0–1 min isocratic at 0% Phase B, 1–2 min linear gradient from 0% to 30% Phase B, 2–12 min linear gradient from 30% to 50% Phase B, 12–21 min linear gradient from 50% to 100% Phase B, 21–26 min isocratic at 100% Phase B, 26–26.1 min linear gradient from 100% to 0% Phase B, and 26.1–30 min isocratic at 0% Phase B; MS detection was implemented on an AB Sciex TripleTOF 6600 mass spectrometer, with samples ionized using an electrospray ionization (ESI) source post-UHPLC separation and subjected to analysis in both positive ion mode (ESI^+^) and negative ion mode (ESI^−^), and MS data acquisition parameters configured as follows: total acquisition duration 30 min, time-of-flight MS (Tof MS) scanning range 90–1500 m/z, product ion scanning range 50–1500 m/z, information-dependent acquisition (IDA) trigger threshold where the top 18 monovalent charged ions exhibiting an intensity exceeding 100 cps in the Tof MS scan were selected for fragmentation, and dynamic background subtraction functionality activated.

2.5. DPPH Radical Scavenging Activity of the Extract

Free radical scavenging capacity of different extracts was assessed via the 2,2-diphenyl-1-picrylhydrazyl (DPPH) assay. For this assay, 1 mL of ethanol-based extract solutions and standard solutions (concentration gradient: 0.25–5 mg·mL^−1^) were mixed with 2.4 mL of DPPH ethanol solution in test tubes. Each mixture was subjected to dark incubation at 25 °C for 30 min, followed by absorbance determination at 517 nm with a spectrophotometer. Extracts without DPPH addition were set as the blank control, and vitamin C (VC) was utilized as the positive control. DPPH scavenging activity was calculated by the formula: DPPH scavenging activity = [(A_n_ − A_v_)/A_n_] × 100, where A_n_ denotes the absorbance of the control group and A_v_ represents that of the test samples at 517 nm.

2.6. ABTS Radical Scavenging Activity of the Extract

Free radical scavenging capacity of different extracts was assessed via the 2,2′-azino-bis(3-ethylbenzothiazoline-6-sulfonate) (ABTS) assay, with ultraviolet–visible (UV–Vis) spectrophotometric detection at 734 nm. For this assay, 1 mL of ethanol-based extract solutions or standard solutions (concentration gradient: 0.1–1.6 mg·mL^−1^) was mixed with 3 mg·mL^−1^ ABTS solution at a 1:1 (v/v) volume ratio in test tubes. Each mixture was incubated at 25 °C for 15 min, followed by absorbance determination at 734 nm with a microplate reader. Extracts without ABTS addition were set as the blank control, and vitamin C (VC) was utilized as the positive control. ABTS radical scavenging activity percentage was calculated by the formula: ABTS scavenging activity = [(A_n_ − A_v_)/A_n_] × 100, where A_n_ denotes the absorbance of the control group and A_v_ represents that of the test samples at 734 nm.

2.7. Screening for BKRE and HF Targets

Standardized molecular structures of the identified components from Berberis kaschgarica Rupr. (BKR) were obtained as SMILES (Simplified Molecular-Input Line-Entry System) identifiers, sourced from the PubChem database (https://pubchem.ncbi.nlm.nih.gov/; accessed on 5 May 2025). Specific targets of these BKR-derived components were identified using comprehensive data resources from the SIB Swiss Institute of Bioinformatics. After that, the researcher searched widely for HF-related targets by looking up three reliable databases: GeneCards, OMIM (Online Mendelian Inheritance in Man), and DrugBank. Rigorous search criteria were formulated, with inclusion restricted to targets achieving a score of ≥20 to ensure the relevance and credibility of the identified targets.

2.8. Network Construction

A compound–protein target interaction network was created through Cytoscape software (version 3.9.1). We specifically chose this version because of its capability to easily integrate active compounds and their corresponding targets related to the test agent. After this stage, protein–protein interaction (PPI) information was obtained from the STRING database. We also used the same Cytoscape version (3.9.1) to build a PPI network from the retrieved datasets. The resulting protein–protein interaction network illustrated the intricate interactions between the proteins that interacted most comprehensively. Such visualization made it very easy to analyze and interpret the data deeply; thus, it was a great help in understanding the biological mechanisms in more depth.

2.9. Differentially Expressed Genes Screening

Relevant to heart failure, clinical datasets had been retrieved from the Gene Expression Omnibus (GEO) database, with a particular emphasis on the GSE76701 dataset created through the GPL570 platform. The dataset acquired was rich in details of clinical samples. One of the major goals of the present study was to uncover and understand the new aspects of cardiac remodeling that accompany HF progression. For this purpose, the expression of genes in the tissues of the ischemic failing (F) and non-failing (NF) hearts was compared. Following this, the gene expression data were compared with the proteomic changes occurring in the F and NF hearts to get a more comprehensive picture of the molecular changes leading to HF. We conducted differential expression analysis using R Studio 4.3 to identify genes with significant expression changes. Statistical significance was determined using the following criteria: an absolute log_2_ fold change (|log_2_FC|) > 0.5 and a p-value.

2.10. Functional Enrichment Analysis

Gene Ontology (GO) analysis consists of three major categories: biological processes (BPs), molecular functions (MFs), and cellular components (CCs). These three interrelated categories operate together to define the functional features of the target genes [18]. In order to understand the hub genes that were identified here more, we used the Kyoto Encyclopedia of Genes and Genomes (KEGG) database for the enrichment analysis—a method that helped to locate and figure out the enrichment profiles of various biological signaling pathways [19]. After that, we went on to perform GO and KEGG enrichment analysis using the R programming language (version 4.3.0) and Bioconductor ClusterProfiler package (version 4.4.4). By combining the initial analysis results in R and using the ClusterProfiler package’s analytical capabilities, we carried out enrichment analysis at the level of gene cluster-specific, thus allowing more advanced and detailed investigation. The main purpose of this methodological framework was to closely define the functional characteristics and the ways of interaction of the hub genes that were pinpointed; thus, the final goal was to figure out their biological roles at the systemic level [20].

2.11. Molecular Docking Assay for Component-Target Binding Validation

We employed an integrated approach that merged network pharmacology and bioinformatics to carry out a comprehensive screening strategy, which in turn led to the identification of five crucial hub genes: ALOX5, NQO1, ALOX5AP, PDE5A, and PLA2G2A. Exploring structural and functional features of the target genes through their three-dimensional (3D) conformations became an important step. Hence, we got the 3D structures of these hub genes in PDB format from the Protein Data Bank (PDB) [21]. After that, we used AutoDock 4.2.6 to preprocess the target proteins (e.g., Titan). As part of the protein preprocessing operation that was going on, they took out the water molecules and put hydrogen atoms in—these changes were done to make the docking simulations that would come later more reliable. The protein structures that had been optimized were later recorded in PDBQT format [22] for subsequent analyses. In the HPLC/MS-based detection assay, 14 bioactive components were listed for the next work along with the detailed description of their structural properties. We used the Traditional Chinese Medicine Systems Pharmacology Database (TCMSP) and PubChem database for structural retrieval to get the small-molecule structures of these components. For example, TCMSP helped us get the 3D structural files of the components derived from the herbs, and PubChem provided additional structural data (e.g., SDF, PDB formats) for cross-validation. For ligand preparation, we refined the retrieved small-molecule structures using AutoDock 4.2.6—redundant water molecules were removed, hydrogen atoms incorporated, and Gasteiger charges calculated. Such operations were necessary to ensure the accuracy of the molecular conformations. The ligand files that were optimized were recorded in PDBQT format for the next molecular docking simulations. Before docking, the structural files were loaded into AutoDock 4.2.6, which is a very important step for starting the docking operation. In the course of this, we determined the lowest binding energy between receptors and small-molecule ligands and also their binding affinities. After the binding energies had been figured out, the PDBQT files were changed to PDB format by OpenBabel 2.4.1 [23]. The change was necessary to ensure that the research would be continued using other software tools without any compatibility issues. Finally, we employed PyMOL 2.5, a Python-based 3.11 graphical interface, to depict and scrutinize the molecular docking maps obtained. The depiction allowed us to understand the biological implications of the complex interactions of small-molecule parts with the hub genes. Moreover, it enabled the inference of possible binding conformations and regulatory pathways.

2.12. Evaluate the Biocompatibility of BKRE in the Zebrafish Embryos

Experimental grouping refers to Section 2.1.2 (Maximum Tolerated Concentration Assay), with zebrafish embryos (6 hpf) randomly placed in 6-well plates (5 mL culture medium per well, 10 embryos per well) and three biological replicates per group (180 embryos total, each as an individual experimental unit); the negative control group was maintained in standard E3 medium while treatment groups were continuously exposed to respective BKRE concentrations throughout the experiment, with daily medium renewal and timely removal/recording of dead embryos to ensure data reliability; at 72 hpf, endpoint evaluations were performed to determine the maximum tolerable concentration (MTC): lethality rate was calculated per well as the primary metric for MTC determination, cardiac rate was measured by randomly selecting 10 larvae per group, anesthetizing them with 0.01% MS-222, and manually counting cardiac beats for 20 s (3 replicates per larva) under a stereomicroscope (SZ61, Olympus, Japan), and tissue structural integrity was assessed under the unified single-blinding protocol (consistent with Section 2.1.1) to record structural abnormalities (e.g., body axis curvature, tissue disruption) with integrity graded as “intact,” “partially disrupted,” or “severely disrupted,” representative phenotypes were photographed, and the structural disruption rate was calculated, with core assessment parameters including zebrafish lethality rate, cardiac rate, and tissue structural disruption rate at 72 hpf.

2.13. Development of a Zebrafish Model of ISO-Induced Heart Failure

At 24 hpf, zebrafish embryos were stringently screened in order to select those that were at a homogeneous developmental stage. Subsequently, the selected embryos were transferred into 6-well plates at a density of 10 embryos per well. This screening and seeding step was instrumental in maintaining the homogeneity of the experimental groups. The control group was always maintained in standard E3 medium throughout the study. In contrast, the model group was exposed to ISO at a concentration of 0.02 mg/mL continuously up to 72 hpf—a concentration of ISO that was responsible for cardiac dysfunction in zebrafish, as had been verified earlier. The culture medium was changed daily during the study in order to maintain stable exposure conditions. We used two main criteria to establish the successful generation of the model: firstly, we found the cardiac rate to be significantly lower in the model group as compared to the control group; secondly, prominent cardiac morphological anomalies were apparent in the model group, which further corroborated the induction of cardiac impairment.

2.14. Effect of BKRE on ISO-Induced Cardiomyocyte Apoptosis

A total of 120 zebrafish larvae at 24 hpf were randomized into experimental groups (6-well plates, 5 mL medium/well, 10 larvae/well, 3 biological replicates). The model and BKRE-treated groups were exposed to 0.02 mg/mL ISO, with the latter additionally receiving varying concentrations of BKRE. Medium was refreshed daily to ensure consistent drug exposure. At 72 hpf, apoptotic cells were detected using acridine orange (AO) staining—a fluorescent dye that specifically labels apoptotic cells—following the manufacturer’s protocol for the apoptosis assay kit. Post-staining, larvae were anesthetized with 0.01% MS-222 (tricaine methaneulfonate) to facilitate imaging. Lateral-view images were captured from 10 randomly selected larvae per group, and the number of apoptotic cells in the cardiac and cerebral regions was quantified using ImageJ software.

2.15. ROS Detection in ISO-Induced HF Zebrafish

Zebrafish larvae at 24 hpf were randomly distributed into 6-well plates (5 mL medium/well, 10 larvae/well, 3 biological replicates), with 120 larvae used in total. The model group and BKRE intervention groups were exposed to 0.02 mg/mL ISO, and the intervention groups were co-treated with gradient concentrations of BKRE. Daily medium renewal was maintained throughout the experiment. At 72 hpf, larvae were incubated with the ROS-sensitive fluorescent probe DCFH-DA (2′,7′-dichlorodihydrofluorescein diacetate) for 1 h in the dark—DCFH-DA is oxidized to fluorescent DCF in the presence of intracellular ROS, enabling quantitative assessment. After incubation, larvae were anesthetized with 0.01% MS-222, and lateral-view fluorescence images were acquired from 10 randomly selected larvae per group for subsequent ROS level analysis.

2.16. NO Quantification

Zebrafish larvae were first euthanized with an overdose of 0.1% tricaine methane sulfonate (MS-222) for 5 min until no cardiac activity was observed, in compliance with the AVMA Guidelines for the Euthanasia of Animals. Subsequent experiments involving whole-body protein extraction (Section 2.17, Section 2.18, Section 2.19 and Section 2.20) adopted the same euthanasia method. Whole-body protein extracts were prepared by homogenizing euthanized larvae in ice-cold phosphate-buffered saline (PBS, pH 7.4) using a glass homogenizer, followed by centrifugation at 12,000× g for 15 min at 4 °C to collect supernatants. Nitric Oxide (NO) levels were assayed using a commercial NO detection kit. After standardizing protein concentrations via a bicinchoninic acid (BCA) assay, 50 μL aliquots of the 1 mg/mL extracts were processed with kit-specific reagents, including incubation at 37 °C and subsequent centrifugation. Absorbance was measured at 540 nm, and NO concentrations were derived from a standard curve, with final values normalized against total protein content.

2.17. Reduced GSH Determination

The quantification of glutathione (GSH) levels in whole-body protein extracts of zebrafish was conducted using a microplate-based GSH assay kit. Initially, protein concentrations in the extracts were standardized via the bicinchoninic acid (BCA) method. Subsequently, 0.1 mL aliquots of the extracts, which had been adjusted to a concentration of 1 mg/mL, were mixed with the reagents provided in the kit and incubated at 37 °C. After the incubation period, we spun down the samples in a centrifuge and then recorded the absorbance at 405 nm. Using a previously validated standard curve, we determined the GSH amounts in the samples. In the end, to allow for consistent comparison and easier understanding of the results, we adjusted the GSH levels according to the total protein content of each sample.

2.18. CAT Activity Measurement

Catalase (CAT) activity in zebrafish whole-body protein extracts was determined using a microplate-based CAT assay kit. Protein concentrations were standardized to 1 mg/mL via the BCA method, and 0.1 mL aliquots of the normalized extracts were mixed with kit-supplied reagents, incubated at 37 °C for 20 min, centrifuged, and absorbance measured at 405 nm. CAT activity was calculated using the kit’s validated standard curve and normalized to total protein content for consistent comparison.

2.19. T-SOD Activity Measurement

We measured total superoxide dismutase (T-SOD) activity using the hydroxylamine method with an Elabscience T-SOD kit. Firstly, we adjusted the protein concentration of the extract to 1 mg/mL by the BCA assay. Then, 0.1 mL aliquots of the protein-normalized extracts were mixed vigorously with the reagents supplied by the kit. The mixtures were incubated, and after this incubation step, the absorbance was measured at the wavelength range of 540–560 nm. Based on the formula provided in the kit manual, we calculated the T-SOD activity. In order to make the results more reliable and comparable, we adjusted the calculated T-SOD activity values according to the total protein content of each sample.

2.20. Determination of TNF-α and IL-6

The levels of Tumor Necrosis Factor-α (TNF-α) and Interleukin-6 (IL-6) were measured in protein extracts from the whole bodies of zebrafish using enzyme-linked immunosorbent assays (ELISA). Briefly, for the detection of TNF-α, we employed the Elabscience TNF-α ELISA Kit and for IL-6 quantification, the respective IL-6-specific ELISA Kit from the same manufacturer was used. This method allows for the accurate and specific determination of these two pro-inflammatory cytokines in the zebrafish samples. After BCA-based normalization of protein concentrations, we processed the samples by strictly following the manufacturers’ standard operating procedures (SOPs): we added extract supernatants to pre-coated microplate wells, followed by the stepwise addition of biotinylated detection antibodies, streptavidin–horseradish peroxidase (streptavidin–HRP), and substrate solutions. Subsequently, we measured the absorbance at 450 nm with a reference wavelength calibration at 540 nm using a microplate reader. Concentrations of TNF-α and IL-6 were calculated based on standard curves generated with kit-supplied standard solutions, and the final values were normalized to the total protein content of the extracts.

2.21. Transcriptomics of Zebrafish

Thirty randomly selected zebrafish per group were euthanized as described in Section 2.16. Total RNA was extracted using TRIzol reagent; purity was verified via NanoDrop 2000c and integrity via 1% agarose gel electrophoresis. RNA samples with A260/A280 = 1.8–2.0 and RIN ≥ 7.0 were used for library construction with NEBNext^®^ Ultra™ RNA Library Prep Kit, followed by sequencing on Illumina NovaSeq 6000 (Illumina, USA) to generate 150 bp paired-end reads. Raw data were filtered with Trimmomatic, and clean reads were mapped to zebrafish reference genome (GRCz11) using Hisat2. Differential expression analysis was done with DESeq2 (|log2(fold change)| ≥ 1, adjusted p < 0.05), and functional annotation via GO/KEGG enrichment.

2.22. Gene Expression

Total RNA was extracted from the homogenates of thirty randomly selected individuals per group using TRIzol reagent. Complementary DNA (cDNA) was synthesized from 1 μg total RNA with PrimeScript™ RT reagent kit (Takara Bio Inc., Japan) to eliminate genomic DNA. Based on transcriptome data, five metabolic genes (PPAR-α, AMPK-α, HK2, PFKL, PPARGC-1α, mTOR) were selected for qRT-PCR validation using a KAPA SYBR FAST one-step kit on a Light Cycler 96 system. Cycling conditions: 95 °C for 3 min, followed by 40 cycles of 95 °C for 10 s and 60 °C for 30 s. β-actin served as the internal control, and relative expression was calculated via the 2^−ΔΔCt^ method (Livak and Schmittgen). All reactions were run in triplicate with non-template controls.

2.23. Statistical Analysis

All experimental data were analyzed using GraphPad Prism 10.0.0 software. Prior to parametric statistical analysis, the Shapiro–Wilk test was performed to verify the normality of all datasets, and the Levene’s test was used to assess variance homogeneity, confirming that all quantitative data followed a normal distribution (p > 0.05) and met the variance homogeneity assumption (p > 0.05). For in vivo experiments (including MTC assay and ISO-induced HF model studies), the well (containing 10 embryos/larvae) was defined as the independent experimental unit, with each group containing 3 biological replicates (n = 3 per group); individual embryos/larvae within a well were considered as subsamples. For in vitro assays (DPPH/ABTS radical scavenging), the experimental unit was a single reaction tube with 3 technical replicates per concentration. Quantitative data are presented as mean ± standard error of mean (SEM) for both in vivo and in vitro experiments, calculated based on biological replicates. Pairwise comparisons between two groups (e.g., control vs. ISO-induced model group, BKRE vs. vitamin C at the same concentration) were conducted using the independent samples t-test. Multiple-group comparisons (e.g., control group, ISO model group, and BKRE treatment groups; different BKRE concentration groups in MTC assay) were analyzed by one-way analysis of variance (ANOVA) followed by Tukey’s HSD post hoc t-test to perform pairwise comparisons among groups. A two-tailed p-value < 0.05 was considered statistically significant.

3. Results

3.1. LC-MS/MS Analysis of Extracts from the Fruits of Berberis kaschgarica Rupr.

For chemical profiling of BKRE, UPLC-MS/MS analysis identified 14 bioavailable, drug-like compounds, primarily categorized into flavonoids and alkaloids (Table 1). The flavonoid component encompassed Kaempferol, Tamarxetin, Luteolin, Tangeretin, and Baicalein, whereas the alkaloid fraction included Diosmetin, Nobiletin, Dihydroberberine, Isocorydine (+), Oripavine, (R)-N-Methylcoclaurine, Tetrandrine, Dihydropalmatine, and Orydaline. All identified compounds satisfied the predefined criteria for bioavailability (OB) and drug-likeness, with key analytical parameters (retention time, m/z, adduct type, PPM, and detection limit) in line with standard UPLC-MS/MS characterization criteria. The total ion chromatogram (TIC) and qualitative analysis results of the identified components are shown in Figure 1.

3.2. Antioxidant Activity of BKRE

For evaluation of the in vitro antioxidant activity of BKRA, DPPH and ABTS radical scavenging assays were performed (Figure 2). In the DPPH radical scavenging assay (Figure 2A), a concentration-dependent elevation in scavenging activity was observed for BKRA: as its concentration increased from 0 to 6 mg/mL, the DPPH radical scavenging rate increased progressively to ~55%. Consistently, the ABTS assay (Figure 2B) revealed a concentration-dependent enhancement in BKRA’s antioxidant capacity: with concentrations ranging from 0 to 2.0 mg/mL, its relative ABTS radical scavenging activity rose to ~65%. Overall, the results obtained in the current investigation clearly show that BKRA yields a measurable antioxidant effect that varies with the concentration, i.e., it is a concentration-dependent effect. An important point to note is that the effect is not very strong, rather it lies within the moderate range, thus indicating that the antioxidant efficacy of BKRA is regulated by its concentrations in the experimental system.

3.3. Network Pharmacology and PPI Network Analysis

Utilizing 14 bioactive pharmaceutical compounds and their 421 associated targets, we assembled a combined compound–target network through Cytoscape version 3.9.1 (Figure 3A). This elaborate network consists of one parent drug entity, 14 unique compounds derived therefrom, and 421 corresponding targets. Compounds in the network were individually classified based on their degree values—higher degree values signify that the compound is involved in more varied biological processes and has a greater functional significance. In addition to that, by means of Venn diagram analysis, we identified 389 common genes between drug targets and disease-associated genes (Figure 3B). After that, PPI data were fetched from the STRING database and merged into Cytoscape 3.9.1 to produce a PPI network diagram (Figure 3C). Core proteins such as AKT1, SRC, EGFR, and ESR1 in this network are indicated with more intense color—this representation is in line with the higher weights coming from their DEGREE values.

3.4. Identification of Differentially Expressed Genes and Subsequent Functional Enrichment Analysis

Data pertaining to the HF model were systematically sourced from the GEO database and subsequently subjected to in-depth analysis via R Studio (available at https://cran.rstudio.com/, accessed on 5 May 2025) to identify differentially expressed genes (DEGs), with a total of 186 DEGs being identified, comprising 83 upregulated and 103 downregulated genes; to visually represent the distribution characteristics of these DEGs, a volcano plot (Figure 4D) and a heatmap (Figure 3E) were generated, both clearly showcasing gene expression alterations; moreover, ALOX5, NQO1, ALOX5AP, PDE5A, and PLA2G2A were selected as key target genes (Figure 4A), which have been extensively involved in various drug and disease studies and exhibited prominent upregulated or downregulated trends in our dataset; to elucidate potential signaling pathways, Gene Ontology (GO) and Kyoto Encyclopedia of Genes and Genomes (KEGG) enrichment analyses were performed using the ClusterProfiler package in R Bioconductor (Figure 4), where GO analysis indicated that overlapping genes were enriched in biological processes (BP) such as response to chemical stimuli and detoxification (Figure 4C), cellular components (CCs) including the extracellular region and collagen-containing extracellular matrix (Figure 4D), and molecular functions (MFs) like ion binding and cargo receptor activity (Figure 4E), all functionally associated with AMPK-mediated metabolic homeostasis and signal transduction, while KEGG analysis (Figure 4B) highlighted significant enrichment in metabolism-related pathways, with the AMPK signaling pathway (marked in Figure 4B) showing a high target count, high enrichment score, and low p-value, suggesting it may act as a core regulatory axis in the biological processes mediated by the target genes.

3.5. Molecular Docking

We conducted molecular docking analysis with the core targets of BKR as follows: ALOX5 (PDB ID: 3V99), NQO1 (PDB ID: 1D4A), ALOX5AP (PDB ID: 2Q7M), PDE5A (PDB ID: 1UHO), and PLA2G2A (PDB ID: 1OJA) were used as receptors, and all these PDB IDs were verified to be human-derived proteins. Meanwhile, the active ingredients of BKR (e.g., Kaempferol, Tamaricetin, and other analogs) were used as ligands in this setup. In order to measure the binding affinity of BKR components with these targets, binding energies were analyzed as follows: When the conformations of both the ligand and the receptor are stable, a more negative (lower) energy value indicates a higher probability of stable binding. According to the docking criteria, a binding energy of ≤−4.25 Kcal/mol is indicative of initial binding capacity; ≤−5.00 Kcal/mol shows binding affinity; and ≤−7.00 Kcal/mol signifies strong and stable binding interactions. The binding energy patterns of each component–-target pair have been put together in the heatmap (Figure 5A). The color gradient from orange to dark blue here represents the binding energy intensity, with dark blue being the more negative (stronger) binding interactions. We depicted the representative molecular binding modes such as the interaction between ALOX5 and Tamaricetin, and that of NQO1 and Isocorydine in Figure 5B. The figure shows the 3D binding interactions (e.g., hydrogen bonds and hydrophobic interactions) between BKR components (ligands) and target proteins (receptors) in their respective binding sites. It is worth mentioning that some of the BKR components had a strong binding affinity (energy ≤ −7.00 Kcal/mol) for the targets as evidenced by the dark blue areas in Figure 5A. Additionally, ALOX5–ligand complexes were extremely stable binding interactions and for this reason, the details are shown in Figure 5B. The exact binding energy values for all the pairs of components and targets are given in Supplementary Table S1.

3.6. Experiment on the Maximum Tolerated Concentration (MTC) of the Extract

To evaluate the toxicity of BKRE, toxicity tests were conducted on zebrafish larvae, focusing on tissue structural integrity, survival rate, and cardiac function parameters after exposure to a concentration gradient of 1–1000 μg/mL. As shown in Figure 6A, larvae in the control group exhibited intact tissue structure, while concentration-dependent structural changes were observed in the treated groups: exposure to 1–10 μg/mL caused no obvious structural abnormalities; 50 μg/mL led to distinct body axis curvature; and 200 μg/mL resulted in severe tissue structural disruption. Data in Figure 6B show the survival rate (calculated as the mean survival percentage per well) was over 90% at concentrations ≤20 μg/mL, then declined sharply at higher concentrations-~60% survival at 30 μg/mL, ~30% at 50 μg/mL, and nearly 0% at ≥500 μg/mL (p < 0.01 vs. control group). The structural disruption rate (Figure 6C) followed a similar trend: minimal (≤10%) at ≤10 μg/mL, increasing to ~60% at 50 μg/mL, and reaching 100% at ≥200 μg/mL (p < 0.01 vs. control group), with body axis curvature as the main structural abnormality. Regarding cardiac function (Figure 6D), heart rate (beats per 20 s) slightly increased at low BKRE concentrations (1–20 μg/mL), then gradually decreased with increasing exposure. At 200 μg/mL, the heart rate dropped to approximately 40 beats per 20 s, a ~30% reduction compared to the control group (p < 0.05). Based on these toxicological data, 20 μg/mL was determined as the maximum tolerable concentration (MTC, highest safe therapeutic dose), with 5 μg/mL and 10 μg/mL set as low- and medium-dose regimens, respectively.

3.7. Effect of BKRE on ISO-Induced Cardiac Function Changes

We have examined the effects of BKRE on cardiac injury caused by ISO in zebrafish larvae, in detail, in order to substantiate the cardioprotective potential of BKRE. Our pilot study has helped us to determine the ISO modeling concentration of 30 μg/mL, and Figure S1 describes the procedures in detail. The choice of this concentration is in line with the method used by [10] in their study of cardiac dysfunction induction in both embryonic and adult zebrafish by chronic isoproterenol stimulation [10]. We have used fluorescent imaging to assess the cardiac morphology of zebrafish larvae in which the cardiomyocytes were labeled by the MF20 antibody (green fluorescence). The cardiac structures of the control group animals were visually intact, well-organized, and clearly defined in Figure 7A. In contrast, the architecture of the heart in the ISO-induced model group was severely disrupted; this disorganization was essentially due to the lack of any regular pattern in the cardiac tissue. This is indicative of the fact that the extreme disorder of the tissue structure is a major feature of the model group. The repair of the heart tissue could be attributed to BKRE administration at doses of 5, 10, and 20 μg/mL, revealing a dose–response effect. Cardiac performance of the zebrafish larvae was assessed by us through a set of experimental measurements. Heart rate as a core parameter of cardiac functionality was significantly lowered in the model group that was subjected to ISO as compared to the control group (p < 0.01). A dose-dependent manner of the heart rate elevation by BKRE treatment was seen as follows: at each of the tested doses (5, 10, and 20 μg/mL), the heart rate was significantly higher than that in the model group (p < 0.01), as shown in Figure 7B. Cardiac output, which is another essential parameter that depicts the total volume of blood pumped by the heart per unit time, has also changed dramatically. The model group exhibited a major decrease in cardiac output when compared to the control group (p < 0.01), thus confirming the development of cardiac dysfunction due to ISO. BKRE administration relieved the cardiac output reduction in a dose-dependent manner so that the cardiac output of the three concentration groups was significantly higher as compared to the model group (p < 0.01), as shown in Figure 7C. We also measured the atrial–ventricular (AV) distance to understand the spatial relationship between the atrium and ventricle and to be able to infer from it cardiac morphological changes. ISO exposure resulted in a substantial enlargement of the AV distance in the model group as compared to the control group (p < 0.01). It is worth noting that BKRE treatment dose-dependently prevented the effect of ISO, leading to the transition of AV distance back toward the control level as at each concentration tested, the AV distance was significantly shorter than that of the model group (p < 0.01), as shown in Figure 7D. The volume of cardiac edema—one of the most common signs of cardiac dysfunction and fluid retention—was also quantified in order to gauge the extent of the cardiac insult. The model group demonstrated a significant elevation of cardiac edema volume in comparison to the control group (p < 0.01), which is indicative of cardiac dysfunction. The BKRE treatment was able to lower the cardiac edema volume in a dose-dependent manner. The reduction in cardiac edema volume was significant in all three BKRE concentration groups compared with the model group (p < 0.01), as shown in Figure 7E. In summary, these extensive data unequivocally demonstrate that BKRE has a pleiotropic protective effect against ISO-induced cardiac damage in zebrafish larvae. In addition to the mitigation of functional deficits (e.g., decreased heart rate and cardiac output), the drug also brought structural changes back to normal, such as cardiac tissue reorganization and shortened AV distance. These findings, therefore, indicate the therapeutic potential of BKRE for cardiac protection.

3.8. Regulatory Effect of BKRE on ISO-Induced Cardiomyocyte Apoptosis

To investigate the anti-apoptotic potential of BKRE during ISO-induced cardiac injury, we used the Annexin V-FITC Apoptosis Detection Kit to identify apoptotic cells in zebrafish larvae. Figure 8A demonstrates that the fluorescent imaging data showed almost no green fluorescence in the control group—an indicator of apoptotic cells. In sharp contrast, the ISO-induced model group showed bright green fluorescence, reflecting a large amount of cardiomyocyte apoptosis. Importantly, BKRE treatment with 5, 10, and 20 μg/mL displayed a dose-dependent reduction in green fluorescence intensity, which suggests a reduction in cardiomyocyte apoptosis. The corresponding quantitative analysis presented in Figure 8B confirmed this beyond doubt: the relative apoptotic fluorescence intensity in the model group was significantly higher than that in the control group (p < 0.01). BKRE treatment was associated with a dose-dependent decrease in this fluorescence intensity. For example, 5 μg/mL BKRE brought about a slight decrease compared to the model group (p < 0.05), while 10–20 μg/mL BKRE had a significantly stronger inhibitory effect on the intensity (p < 0.01). Therefore, these data provide direct evidence that BKRE is able to hamper cardiac apoptosis elicited by ISO in zebrafish larvae, and this anti-apoptotic property is likely to underpin its protective effects against ISO-induced cardiac dysfunction.

3.9. Detection of the Effect of BKRE on ISO-Induced Cardiac ROS Levels

We employed DCFH-DA, a well-known fluorescent probe, for the detection of ROS to assess the antioxidant potential of BKRE to rescue the cardiac injury of zebrafish larvae induced by ISO. As the images in Figure 9A show, the control group exhibited weak green fluorescence, which is indicative of basal ROS levels. On the other hand, the model group that was induced by ISO showed a significant increase in ROS as evidenced by the strong fluorescence. BKRE at 5, 10, and 20 μg/mL lightened the fluorescence intensity in a dose-dependent manner, and the effect of the 10–20 μg/mL groups was more apparent. The quantitative results presented in Figure 9B are consistent with the imaging data and show that the ROS levels in the model group are dramatically higher than in the control group (p < 0.01). The treatment with BKRE at 10 μg/mL and 20 μg/mL decreased the ROS fluorescence intensity to a great extent (vs. model group, p < 0.05), whereas the 5 μg/mL group did not show a statistically significant decrease (ns, not significant). Altogether, these data suggest that BKRE has antioxidant activity in a manner that inhibits the overproduction of ROS induced by ISO in zebrafish larvae, which could be the main mechanism through which it exerts cardiac protective effects against ISO-induced dysfunction.

3.10. Determination of Oxidative Stress-Related Indices in ISO-Induced Myocardial Injury Following BKRE Intervention

In order to deeply examine the regulation of BKRE on the myocardial injury of zebrafish larvae induced by ISO, we measured a series of key oxidative stress-related biomarkers and inflammatory cytokines. These included GSH, NO, T-SOD, CAT, and the cytokines IL-6 and TNF-α. The results in Figure 10 demonstrated that in comparison with the control group, the ISO-induced model group had significantly lowered GSH content (Figure 10A), T-SOD activity (Figure 10C), and CAT activity (Figure 10F), while the NO levels (Figure 10B) and IL-6 (Figure 10D) and TNF-α (Figure 10E) were increased (vs. Ctrl, p < 0.05, p < 0.01). The administration of BKRE at 10–20 μg/mL led to the restoration of GSH, T-SOD, and CAT levels in a dose-dependent manner, reduction in NO content, and decrease in IL-6 and TNF-α levels (vs. model, p < 0.05, p < 0.01). It is worth mentioning that the 5 μg/mL BKRE treatment group did not demonstrate any statistically significant differences (ns) in some indicators (including CAT activity) compared with the model group. In summary, these data provide evidence that BKRE is able to alleviate ISO-induced myocardial injury in zebrafish larvae through a dual mechanism of oxidative stress regulation and suppression of inflammatory response. The former is achieved through an increase in antioxidant capacity (restoring GSH, T-SOD, and CAT activities) and a decrease in the production of oxidative products.

3.11. Zebrafish Transcriptomic Analysis and Validation of Related Genes

We carried out transcriptomic profiling and subsequently validated core genes in order to understand the molecular mechanisms by which BKRE alleviates ISO-induced myocardial injury. Figure 11A (volcano plot) illustrates that the exposure to ISO in the zebrafish larvae resulted in the differential expression of 1527 genes that were downregulated and 749 genes that were upregulated. Figure 11B (heatmap) also reveals that the gene expression cluster patterns were very different between the ISO and ISO + BKRE groups, which indicates that BKRE had regulatory effects on the transcriptomic landscape. The pathway enrichment analysis (Figure 11C) pinpointed the AMP-activated protein kinase (AMPK) signaling pathway and glycolysis/gluconeogenesis as the core enriched pathways (indicated by the red boxes)—pathways consistent with AMPK’s central role in cardiac energy metabolism and stress responses, and closely intertwined with mechanistic target of rapamycin (mTOR) signaling in metabolic regulation. The follow-up validation of the key genes (Figure 11D–I) indicated that ISO treatment caused the transcriptional levels of AMPK-α, hexokinase 2 (HK2), phosphofructokinase, liver type (PFKL), peroxisome proliferator-activated receptor alpha (PPARα), peroxisome proliferator-activated receptor gamma coactivator 1-alpha (PPARGC1A), and mTOR to decrease significantly compared with the control group (p < 0.05). Notably, mTOR—an essential regulator of energy homeostasis and cell growth closely linked to ISO-induced cardiac metabolic dysfunction—exhibited the same downregulation trend as other metabolic genes. On the contrary, BKRE administration at 10–20 μg/mL was able to reverse the changes in the expression of these genes in a dose-dependent manner (vs. model, p < 0.05, p < 0.01). In sum, these results provide evidence that BKRE is capable of alleviating ISO-induced myocardial injury in zebrafish through the regulation of the AMPK-mTOR signaling pathway and genes associated with glycolysis-mediated energy metabolism, thus restoring cardiac energy homeostasis and improving myocardial function by modulating this key metabolic regulatory axis.

4. Discussion

HF is a significant issue for global health. The condition is complicated by a range of pathophysiological mechanisms, including oxidative stress, cardiomyocyte apoptosis, inflammation, and alterations in energy metabolism. These intertwined mechanisms highlight the importance of targeting multiple pathways simultaneously for therapeutic interventions [2,24]. The present work is pioneering in that it thoroughly investigates the protective mechanisms of BKRE on the heart in a zebrafish model of ISO-induced HF through various approaches including chemical profiling [25], in vitro and in vivo experiments, network pharmacology, and transcriptomic analysis to identify the rationale. The findings corroborate ample experimental evidence confirming the ethnopharmacological utilization of BKR to alleviate cardiovascular complications.

Chemical profiling through UPLC-MS/MS of BKRE led to the characterization of 12 biologically active compounds in the extract. Flavonoids (e.g., Kaempferol, Luteolin, Baicalein) and alkaloids (e.g., Dihydroberberine, Tetrandrine, Isocorydine) are the major constituents of the compounds identified. In addition, their bioavailability and drug-like properties were also tested and were found to be favorable for all of the detected bioactive compounds. The chemical composition here serves as the molecular basis of BKRE’s cardioprotective effects; on the one hand, flavonoids are widely accepted as the primary components that can relieve myocardial injury by ROS scavenging and anti-inflammatory properties of the compounds involved [11,16], while on the other, alkaloids (e.g., Dihydroberberine) play the most important roles in regulating cardiac energy metabolism and prohibiting apoptotic signaling pathways [26]. It is interesting to note that Kaempferol and Dihydroberberine are the major components of BKRE, and their roles in alleviating ISO-induced myocardial injury have been well elucidated [27,28]; therefore, these components may work as the pivotal agents that mediate the cardioprotective activity of BKRE.

Abnormally high oxidative stress is a major driver of cardiac problems caused by ISO. Excessive ROS leads to lipid peroxidation, DNA damage, and cell death of the heart tissue, as has been supported by the literature [29,30]. Our in vitro DPPH and ABTS experiments corroborate that BKRE has a free radical-neutralizing effect that is dependent on the concentration, and the DPPH and ABTS scavenging rates reached about 55% and 65%, accordingly. These in vitro results were validated in vivo as well: BKRE administration notably inhibited cardiac ROS production in the ISO-stressed zebrafish, and at the same time, it replenished the GSH level and boosted the T-SOD activity. This ROS scavenging effect is in line with the direct alleviation of oxidative injury induced by ISO and is harmonious with the role of ROS exacerbating HF progression and previous citations, suggesting that natural flavonoids give cardiac protective effects through modulating oxidative stress. From the point of view of the mechanism, the flavonoid materials in BKRE are capable of releasing the metal ions and capturing the radicals, whereas the alkaloid Dihydroberberine, in particular, lowers ROS generation by regulating mitochondrial redox homeostasis [7]. Hence, it forms a mutually reinforcing antioxidant network. Apoptosis of cardiomyocytes and abnormal inflammation are the major pathological events that characterize the progression of ISO-induced HF [15]. The AO staining and imaging results revealed that BKRE administration resulted in a significant reduction in the apoptotic cells in the cardiac region of the damaged zebrafish, and this was accompanied by a decrease in the expression of pro-inflammatory cytokines (TNF-α and IL-6). This implies that BKRE actively inhibits not only apoptosis but also inflammation processes, which are the main mechanistic pathways leading to cardiac protection. Research has already shown that TNF-α and IL-6 promote apoptosis in cardiomyocytes and cardiac remodeling in HF [20]; at the same time, flavonoids like Luteolin and Kaempferol interfere with apoptotic signaling pathways (e.g., caspase cascade) and suppress NF-κB-mediated inflammatory reactions [16]. Our current discoveries are a further step in understanding those earlier observations since we demonstrate that BKRE turns on these synergistic pathways in order to reduce ISO-induced cardiac structural damage such as decreases in atrial–ventricular distance, reductions in pericardial edema, and improvements to functional deficits including increased heart rate and normalized cardiac output.

BKRE’s multi-target mode of action was elucidated to a greater extent by network pharmacology and molecular docking analyses, with ALOX5, NQO1, ALOX5AP, PDE5A, and PLA2G2A being identified as the core therapeutic targets. These genes are significantly involved in the development of HF: ALOX5 and ALOX5AP regulate the biosynthesis of inflammatory mediators, NQO1 modulates oxidative stress responses, PDE5A impairs cardiac contractility [5], and PLA2G2A promotes lipid peroxidation [21]. Molecular docking experiments (e.g., Tamaricetin–ALOX5, Isocorydine–NQO1) substantiated the direct molecular interactions accounting for BKRE’s capacity to simultaneously target multiple pathological pathways. This multi-target property differentiates BKRE from single-target synthetic drugs, thus allowing effective intervention in complex HF pathophysiological processes. Transcriptomic analysis integrated with qRT-PCR validation indicated a pivotal regulatory role of the AMPK signaling pathway. ISO-induced suppression of AMPK-α, HK2, PFKL, PPARα, PPARGC1A, and mTOR—central genes for glycolysis and cardiac energy metabolism—was significantly reversed by BKRE application, highlighting BKRE’s ability to restore the interconnected AMPK-PPARα-mTOR regulatory network that governs cardiac metabolic homeostasis. As the upstream master regulator of energy metabolism, AMPK activation directly phosphorylates and activates PPARα, a nuclear receptor that forms a functional complex with PPARGC1A to transcriptionally upregulate glycolytic genes (HK2, PFKL) and mitochondrial biogenesis-related targets. This synergy is critical for reversing ISO-induced metabolic dysfunction; reduced AMPK activity not only impairs PPARα-PPARGC1A-mediated glycolytic flux but also disrupts the AMPK-mTOR balance, in which AMPK normally inhibits excessive mTORC1 activity under energy stress to conserve resources, while mTOR activation supports metabolic recovery once energy equilibrium has been restored [31]. Notably, ISO-induced suppression of both AMPK and mTOR reflects a dysregulated metabolic state, as uncoupled AMPK-mTOR signaling exacerbates cardiomyocyte energy depletion and contractile dysfunction, a hallmark of HF progression. These findings are in line with the latest reports that AMPK agonists relieve ISO-induced HF by regulating metabolic remodeling [32,33,34,35,36]. This regulatory pattern not only corroborates the well-recognized mechanistic links between AMPK-PPARα-mTOR axis dysfunction and HF pathogenesis in both clinical, epidemiological, and preclinical ISO-induced injury studies, but also extends the current understanding of natural product-mediated metabolic remodeling in cardiac protection, while validating the translational relevance of the zebrafish model for investigating this conserved signaling axis in HF. The comprehensive cardioprotective mechanism of BKRE is visualized in the mechanism diagram (Figure 12). Concerning single-component validation, priority compounds (Kaempferol, Dihydroberberine, Tamaricetin) with strong binding affinity to core targets identified by network pharmacology and molecular docking analyses will be the focus of subsequent isolation and in vitro mechanistic validation experiments. Furthermore, the combination of in vitro (antioxidant assays), in vivo (zebrafish experiments), and in silico (network pharmacology, molecular docking) approaches help to reduce the risk of biases and increases the rigor of our findings. As a whole, the research demonstrates that BKRE offers cardioprotection against ISO-induced HF in zebrafish through a complex mechanism: (1) elimination of ROS and increase in antioxidant capacity through the flavonoid and alkaloid constituents of the extract; (2) reduction in cardiomyocyte apoptosis and suppression of inflammatory responses by regulation of the TNF-α/IL-6 signaling pathways; (3) interaction with the main genes (ALOX5, NQO1, PDE5A) encoding the molecular mechanisms of HF; and (4) recovery of cardiac energy equilibrium through AMPK signaling pathway activation. The findings, therefore, not only validate the ethnopharmacological importance of BKR but also provide a scientific rationale for promoting BKRE as a novel drug candidate in HF therapy.

5. Conclusions

This study clarifies the multi-target cardioprotective mechanisms of BKRE in ISO-induced HF zebrafish, with three core findings: (1) Its efficacy stems from 14 bioactive flavonoids and alkaloids (e.g., Kaempferol, Dihydroberberine) with favorable drug-likeness and bioavailability. (2) It acts synergistically on oxidative stress, cardiomyocyte apoptosis, inflammation, and energy metabolism, restoring antioxidant capacity (GSH, T-SOD, CAT), suppressing TNF-α/IL-6, inhibiting apoptosis, and improving cardiac function (heart rate, cardiac output, AV distance). (3) The AMPKα-PPARα-PGC-1α-mTOR axis serves as the central regulatory pathway, rescuing cardiac energy metabolism by upregulating glycolysis-related genes (HK2, PFKL). Beyond validating the ethnopharmacological value of B. kaschgarica in Uyghur folk medicine, this work offers broader implications for HF therapy. BKRE’s multi-component, multi-pathway mode aligns with HF’s complex pathophysiology, providing a safer and more comprehensive alternative to single-target synthetic drugs. Its favorable biocompatibility (MTC = 20 μg/mL) supports its development as a functional food ingredient or adjuvant therapy for HF, addressing the unmet need for low-toxicity, long-term interventions. Additionally, this study establishes an integrated “in vitro–in vivo–in silico” paradigm for investigating ethnic medicinal plants, serving as a scientific template for unlocking underutilized natural products.

Future research will focus on advancing BKRE’s translational potential, in the following directions: (1) isolating priority bioactive components to validate their individual and synergistic effects; (2) extending studies to mammalian HF models to confirm translational relevance and pharmacokinetic profiles; (3) elucidating component–target binding modes via structural biology; (4) evaluating efficacy in combination with conventional HF drugs; and (5) optimizing extraction and formulation to enhance stability and bioavailability. These efforts will further position BKRE as a promising natural product-based agent for HF management.