Literature DB >> 25733819

A computational and functional study elicits the ameliorating effect of the Chinese herbal formula Huo Luo Xiao Ling Dan on experimental ischemia-induced myocardial injury in rats via inhibition of apoptosis.

Xiang-Dong Han¹, Zhi-Wei Zhou², Wei Yang¹, Hang-Cheng Ye¹, Ying-Zi Xu¹, Yun-Feng Huang¹, Tong Zhang¹, Shu-Feng Zhou³.

Abstract

Ischemic heart disease (IHD) is the leading cause of death worldwide and remains a major life-threatening factor in humans. Apoptosis has been implicated in the pathogenesis of IHD. The Chinese herbal formula Huo Luo Xiao Ling Dan (HLXLD), one of the commonly used Chinese herbal formulas, consists of Salviae miltiorrhizae, Angelica sinensis, Gummi olibanum, and Commiphora myrrha, with a wide spectrum of pharmacological activity. However, the mechanism of action and molecular targets of HLXLD in the treatment of IHD are unclear. This study aimed to computationally predict the molecular interactions between the major active components of HLXLD and key regulators of apoptosis and then examine the effect of HLXLD on coronary artery ligation-induced acute myocardial ischemia in rats. The molecular interactions between the major active components of HLXLD, including ferulic acid, ligustilide, succinic acid, vanillic acid, tanshinone IIA, tanshinone IIB, danshensu, salvianolic acid A, salvianolic acid C, protocatechuic aldehyde, and β-boswellic acid and human protein molecules including B cell lymphoma-extra large (Bcl-xl), B cell lymphoma 2 antagonist/killer 1 (Bak1), B cell lymphoma 2 (Bcl-2), procaspase 3, and caspase 9 with regard to hydrogen bond formation, charge interaction, and π-π stacking using Discovery Studio(®) program 3.1. The 12 HLXLD components were predicted by ADMET (absorption, distribution, metabolism, excretion and toxicity) Predictor to have favorable pharmacokinetic and low hepatotoxicity profiles. The acute myocardial ischemia was established by surgical ligation of the left anterior descending coronary artery. The rats were divided into a sham operative group, a model group, a positive control group treated with 0.2 mg/kg isosorbide mononitrate, and groups treated with 2.7, 5.4, or 10.8 g/kg HLXLD. The results showed that administration of HLXLD increased mean arterial pressure, left ventricular systolic pressure, heart rate, and maximal rate of rise/descent of left ventricular pressure levels. Administration of HLXLD significantly ameliorated coronary artery ligation-induced tissue damage in the left ventricle, with restored arrangement of myocardial fibers and recovered myoplasm in rats. Furthermore, HLXLD markedly increased the expression level of Bcl-2 but decreased the level of cleaved caspase 3. Taken together, administration of HLXLD attenuated acute myocardial ischemia-induced damage in cardiomyocytes and inhibited apoptotic death of cardiomyocytes, thereby exerting a cardioprotective effect in rats with IHD. These findings suggest that HLXLD may represent a promising herbal formula for the treatment of cardiovascular disease by counteracting apoptotic cell death via multiple active compounds. More studies are warranted to fully elucidate the mechanisms of action, identify the therapeutic targets, and validate the efficacy and safety of HLXLD in the treatment of IHD.

Entities: CellLine Chemical Disease Gene Mutation Species

Keywords: Huo Luo Xiao Ling Dan; apoptosis; coronary artery ligation; hemodynamics; molecular docking; myocardial ischemia

Mesh：

Substances：

Year: 2015 PMID： 25733819 PMCID： PMC4342182 DOI： 10.2147/DDDT.S76336

Source DB: PubMed Journal: Drug Des Devel Ther ISSN： 1177-8881 Impact factor: 4.162

Introduction

Ischemic heart disease (IHD, also known as coronary heart disease) is the most common type of cardiovascular disease, is the leading cause of death worldwide, and places a huge burden on the individual, family, and community.1–3 There were 7,249,000 deaths resulting from IHD worldwide in 2008, accounting for 12.7% of total global mortality.1 Worldwide, there is over 20-fold variation in IHD mortality rates between countries.4 In the USA, IHD is the leading cause of mortality, resulting in over 500,000 deaths every year.1,4 In the People’s Republic of China, IHD is the second leading cause of cardiovascular disease-related death, accounting for 22% of cardiovascular deaths in urban areas and 13% in rural areas.5 In the past few decades, although there has been great improvement in our understanding of the etiology of IHD and advances in the diagnostics and therapeutics for IHD, the disease remains a major clinical challenge with life-threatening outcomes. Therefore, there is an urgent need for new agents and novel therapeutic approaches in the treatment of IHD to reduce the mortality and improve quality of life for patients with IHD. Accumulating evidence shows that apoptosis is an important contributing factor in the development and progression of IHD.6,7 The balance between proapoptotic and antiapoptotic proteins is critical to maintain the cellular status of cardiomyocytes. Apoptosis can augment ischemic conditions. The B-cell lymphoma 2 (Bcl-2) protein family is a determining factor in the commitment of cells to apoptosis.8 Manipulation of apoptosis by regulating the Bcl-2 family has been proposed to be a promising strategy for the treatment of many diseases, including IHD and cancer.8 Furthermore, the caspase family, aspartate-specific cysteine proteases, and members of the interleukin-1β-converting enzyme family, are closely associated with apoptosis and have an important role in the induction, transduction, and amplification of intracellular apoptotic signals.9,10 Caspases can be regulated by Bcl-2 family proteins, calpain, and calcium ions.9,10 It has been reported that targeting caspase 3 may represent an attractive approach to attenuate or reverse heart failure.9,10 However, the therapeutic effect and safety of the abovementioned approaches in the treatment of cardiovascular diseases need to be confirmed. There is increasing evidence showing that natural products have substantial therapeutic effects on cardiovascular diseases, including IHD. Huo Luo Xiao Ling Dan (HLXLD) is a classic prescription that has been documented in “Records of Tradition Chinese and Western Medicine in Combination” since the Qing Dynasty.11 HLXLD consists of Radix Salvia miltiorrhiza, Angelica sinensis, Gummi olibanum, and Commiphora myrrha. It has a wide range of pharmacological activity,12–17 such as removing blood stasis, relieving cold pain in the heart and abdomen, ameliorating back pain or lumbago, sprains, and bruises, improving internal and external sores and ulcers, and decreasing lump accumulation in the abdomen. In recent years, increasing evidence suggests that HLXLD may be effective for the treatment of IHD, such as coronary atherosclerotic cardiopathy, due to the numerous pharmacologically active components, such as ferulic acid, ligustilide, succinic acid, vanillic acid, tanshinones, danshensu, salvianolic acids, protocatechuic aldehyde, and β-boswellic acid.18,19 However, mechanistic and functional studies of the therapeutic effects and targets of HLXLD in IHD are still sparse. In particular, there is a lack of investigation of the molecular interactions between the major active components of HLXLD and apoptotic regulators and the regulating effect of HLXLD on apoptosis in IHD. In the present study, we aimed to predict the molecular interactions between the major active components of HLXLD and apoptotic regulators and to examine the therapeutic effect of HLXLD in a rat model of acute myocardial ischemia induced by coronary artery ligation.

Materials and methods

Molecular docking

We first predicted the molecular interactions between the major active components of HLXLD (ferulic acid, ligustilide, succinic acid, vanillic acid, tanshinone IIA, tanshinone IIB, danshensu, salvianolic acid A, salvianolic acid B, salvianolic acid C, protocatechuic aldehyde, and β-boswellic acid) and the key regulators involved in apoptotic cell death using the Discovery Studio® program 3.1 designed by Accelrys Inc (San Diego, CA, USA). The two-dimensional structure of the 12 components was retrieved from the PubChem database (Figure 1) and the three-dimensional crystal protein structures were obtained from the Research Collaboratory for Structural Bioinformatics Protein Data Bank (PDB; http://www.rcsb.org/pdb/). There were five key regulators of apoptosis that were selected for the molecular docking experiments, including human Bcl-xl (PDB ID: 4TUH), Bak1 (PDB ID: 2JCN), Bcl-2 (PDB ID: 2O21), procaspase 3 (PDB ID: 4JRO), and caspase 9 (PDB ID: 2AR9). The molecular docking experiments were performed as previously described, with some modifications.20,21 Briefly, the protein and ligand molecules were prepared prior to docking. For protein preparation, Bcl-xl, Bak1, Bcl-2, procaspase 3, and caspase 9 were cleaned, modified, and prepared for defining and editing the binding site. During preparation for ferulic acid, ligustilide, succinic acid, vanillic acid, tanshinone IIA, tanshinone IIB, danshensu, salvianolic acid A, salvianolic acid B, salvianolic acid C, protocatechuic aldehyde, and β-boswellic acid, the duplicate structures were deleted, and ionization change, tautomer or isomer generation, Lipinski filter, and three-dimensional generator were all set true. A harmonic potential with a force constant of 300 kcal/mol was applied outside the grid boundary. Following preparation of the proteins, ligands, and grid setting, the 12 components were docked into the binding sites of Bcl-xl, Bak1, Bcl-2, procaspase 3, and caspase 9. Electrostatic energy and van der Waals forces were considered during the docking process. For each defined van der Waals force or electrostatic probe, the interactions with all protein atoms were stored at predetermined grid points. For ligand atoms located between grid points, a trilinear interpolation was used to approximate the energies.

Figure 1

Chemical structures of the major active components of Huo Luo Xiao Ling Dan, including ferulic acid, ligustilide, succinic acid, vanillic acid, tanshinone IIA, tanshinone IIB, danshensu, salvianolic acid A, salvianolic acid B, salvianolic acid C, protocatechuic aldehyde, and β-boswellic acid.

Prediction of ADMET of the major active components of HLXLD

Following the prediction of the molecular interactions, we also computationally evaluated the absorption, distribution, metabolism, excretion and toxicity (ADMET) of ferulic acid, ligustilide, succinic acid, vanillic acid, tanshinone IIA, tanshinone IIB, danshensu, salvianolic acid A, salvianolic acid B, salvianolic acid C, protocatechuic aldehyde, and β-boswellic acid using the Discovery Studio® program 3.1 designed by Accelrys Inc as previously mentioned.21 Briefly, the chemical structures of the 12 major active components of HLXLD were retrieved from the PubChem database (Figure 1). All the structures were introduced into the Discovery Studio® program 3.1 and prepared. During the process of structure preparation, duplicate chemical structures were deleted, and ionization change, tautomer or isomer generation, Lipinski filter, and three-dimensional generator were all set true. Following the structure preparation, ferulic acid, ligustilide, succinic acid, vanillic acid, tanshinone IIA, tanshinone IIB, danshensu, salvianolic acid A, salvianolic acid B, salvianolic acid C, protocatechuic aldehyde, and β-boswellic acid were subject to ADMET evaluation using the ADMET Predictor module. The solubility, absorption, permeability across the blood–brain barrier, interactions with cytochrome P450 (CYP)2D6, hepatotoxicity, and plasma protein binding were predicted. There are different levels for description of ADMET and their rules. Absorption levels ranged from 0, 1, 2, to 3, indicating good, moderate, low, or very low absorption, respectively. The predicted aqueous solubility level was classified into extremely low (0), no, very low, but possible (1), yes, low (2), yes, good (3), yes, optimal (4), no, too soluble (5), and warning, molecules with one or more unknown AlogP98 types (6). The predicted blood–brain barrier permeability levels included very high (0), high (1), medium (2), low (3), undefined (4), and warning, molecules with one or more unknown AlogP calculation (5). There are two predicted classes of CYP2D6 ligand, ie, noninhibitor (0) and inhibitor (1). The predicted hepatotoxicity was categorized into nontoxic (0) and toxic (1) effects. The predicted plasma protein binding levels included binding <90% (0), binding ≥90% (1), and binding ≥95% (2).

Chemicals and reagents

HLXLD extract (batch no.: 2010001) was prepared in the Pharmaceutical Laboratory at Shanghai University of Traditional Chinese Medicine, Shanghai, People’s Republic of China. Xinya Danxiao tablets were purchased from Shanghai New Asia Pharmaceutical Minhang Inc. (Shanghai, People’s Republic of China). Urethane was obtained from Sinopharm Chemical Reagent Inc. (Shanghai, People’s Republic of China). Penicillin sodium injections were sourced from Shanghai Asia Pioneer Pharmaceutical Inc. (Shanghai, People’s Republic of China). Normal (0.9%) saline was purchased from Shanghai Changzheng Fumin Jinshan Pharmaceutical Inc. (Shanghai, People’s Republic of China). Superoxide dismutase (SOD), creatine kinase (CK), and lactate dehydrogenase (LDH) assay kits were purchased from Nanjing Jiancheng Technology Inc. (Nanjing, People’s Republic of China). A Pierce™ bicinchoninic acid protein assay kit, skim milk, RIPA buffer, and Pierce enhanced chemiluminescence Western blotting substrate were obtained from Thermo Scientific Inc. (Hudson, NH, USA). Primary antibodies against rat caspase 3 and glyceraldehyde-3-phosphate dehydrogenase (GAPDH) were sourced from Cell Signaling Technology Inc. (Beverly, MA, USA) and the primary antibody to Bcl-2 was from Biolegend Inc. (San Diego, CA, USA). The IRDye® 680RD secondary antibody donkey anti-mouse immunoglobulin G was obtained from Li-Cor Biosciences Inc. (Lincoln, NE, USA).

Herbal preparation

A. sinensis, G. olibanum, and C. myrrha (500 g each) were mixed and dissolved in distilled water (1:10, w/v). A 9-mL aliquot of volatile oil was extracted from the mixture. Next, 240 g of powder was obtained from the aqueous solution of Chinese angelica, frankincense, and myrrh mixture. Salvia miltiorrhiza (500 g) was dissolved in alcohol (1:10, w/v) and heated under reflux for 1 hour, and 500 mL of alcoholic extraction was collected by centrifuging at 3,000× g for 30 minutes. The residues were dissolved in water (1:10, w/v). The water solution was heated under reflux for 1 hour twice, and the resulting liquids were combined with the previous extractions. The mixture was then concentrated by centrifugation to 250 mL. The extracts of the drugs were mixed into a suspension and diluted with distilled water to prepare the low-dose (2.7 g/kg), mid-dose (5.4 g/kg), and high-dose (10.8 g/kg) test.

Animals

Male Sprague-Dawley rats (200–250 g, aged 6–8 weeks) were provided by the Experimental Animal Center of Shanghai University of Traditional Chinese Medicine. The rats were housed in the accredited animal housing facility under controlled temperature (22°C±2°C) and a 12-hour dark/light cycle. All rats were allowed to acclimate for 5–7 days prior to the experiments. All experiments were conducted according to the animal protocol approved by the animal ethics committee of Shanghai University of Traditional Chinese Medicine.

Animal experiment

In order to examine the effect of HLXLD on IHD in rats, we first performed coronary artery ligation to establish an acute myocardial ischemic rat model using thoracotomy. Briefly, the animals were anesthetized by intraperitoneal injection of 25% urethane (1.25 g/kg) and then placed in a dorsal position with limbs been fixed. Thoracotomy was then performed and coronary artery ligation was achieved by inserting a needle 2 mm below the bottom of the left auricula sinistra. The chest was quickly closed and the incision was sutured. The rats were monitored by an animal electrocardiogram machine. Establishment of the model was considered successful with one of following conditions: horizontal shifts of the ST-segment downward or upward by >0.1 mV; T-wave elevation over 1/2 of the R wave in the same lead; T-wave elevation accompanied by ST-segment shifts; oblique or horizontal shifts of ST-segment <0.1 mV; and a flat low pattern or two-way inversion of the T-wave. The established animal models were randomly divided into five groups (n=8), including a model group, a positive control group (isosorbide mononitrate tablets), and groups treated with HLXLD 2.7, 5.4, or 10.8 g/kg. Additionally, a sham operation group of rats underwent thoracotomy with insertion of a line but no coronary artery ligation. After surgery, each animal was administered sodium penicillin (40,000 U/day) by intramuscular injection for 3 consecutive days to prevent postoperative infection. At 24 hours post surgery, all experimental groups were given the test drug by intragastric administration once a day for 4 weeks. The model control group and sham operation group were given the same volume of distilled water by intragastric administration. After 4 weeks, the animals were fasted for 12 hours and then anesthetized. The left and right carotid arteries were separated and a PE-50 polyethylene catheter containing 50 units/mL heparin sodium solution was inserted. The animals were connected to a physiological recorder via a pressure transducer for recording the changes in electrocardiography and pressure. Hemodynamic parameters, including heart rate, mean arterial pressure (MAP), left ventricular systolic pressure (LVSP), left ventricular end-diastolic pressure (LVEDP), and maximum rate of left ventricular pressure rise and fall (±dp/dtmax) were measured and recorded. Thereafter, blood specimens were collected from the carotid artery. Serum was separated by centrifugation at 300× g and then used for CK, LDH, and SOD assays. Additionally, tissue specimens were taken from the left ventricle and divided into two parts for Western blotting assay and morphologic examination.

Measurement of selected enzymes in rat plasma

The blood samples were collected from the carotid artery and separated by centrifugation at 300× g. The effect of HLXLD in rats with coronary artery ligation-induced acute myocardial ischemia was evaluated by examining plasma levels of SOD, CK, and LDH. Assay kits for SOD, CK, and LDH were used to measure their levels in blood according to the manufacturer’s instructions.

Histopathological examination

To examine the effect of HLXLD on the morphology of cardiomyocytes, left ventricle tissue samples were collected for routine histopathological examination as previously described.22 The tissue samples were fixed with formalin, embedded in paraffin, and sectioned. Sections 4 μm thick were stained with hematoxylin and eosin and observed under light microscopy (Nikon Eclipse 50i™; Kanagawa, Japan). Magnification was 40×.

Western blotting assay

The expression levels of rat heart caspase 3 and Bcl-2 were determined using the Western blotting assay as described previously.23 Samples of rat left ventricle tissue (5 mg each) were washed with pre-cold phosphate-buffered saline and homogenated in 200 μL of RIPA buffer containing a protease inhibitor cocktail. The supernatant was collected twice via centrifugation at 15,000× g for 15 minutes at 4°C. The protein concentration were measured by a bicinchoninic acid protein assay kit. An equal amount of protein sample (20 μg) was electrophoresed on 12% sodium dodecyl sulfate polyacrylamide gel electrophoresis (SDS-PAGE) mini-gel after thermal denaturation for 5 minutes at 95°C. The proteins were transferred onto a methanol-activated polyvinylidene difluoride membrane at 100 V for 2 hours at 4°C. Subsequently, the membranes were blocked with 5% skim milk and probed with primary antibodies to caspase 3 and Bcl-2 overnight at 4°C and then incubated with the respective secondary antibody. Visualization was performed using an infrared fluorescence scanning and imaging system (Li-Cor Biosciences Inc., Lincoln, NE, USA). The protein level was normalized to the matching densitometric value of GAPDH.

Statistical analysis

The results are expressed as the mean ± standard deviation. Comparison of means between groups was performed using one-way analysis of variance (ANOVA) followed by Tukey’s multiple comparison. A value of P<0.05 was considered to be statistically significant.

Results

HLXLD components interact with human Bcl-xl, Bak1, Bcl-2, procaspase 3, and caspase 9

Following the prediction and validation of the molecular targets, we examined the possible interactions between ferulic acid, ligustilide, succinic acid, vanillic acid, tanshinone IIA, tanshinone IIB, danshensu, salvianolic acid A, salvianolic acid B, salvianolic acid C, protocatechuic aldehyde, and β-boswellic acid with human Bcl-xl, Bak1, Bcl-2, procaspase 3, and caspase 9 using the Discovery Studio® program 3.1. Ferulic acid, ligustilide, succinic acid, vanillic acid, tanshinone IIA, tanshinone IIB, danshensu, salvianolic acid A, salvianolic acid C, protocatechuic aldehyde, and β-boswellic acid were easily docked into the active sites of human Bcl-xl, Bak1, Bcl-2, procaspase 3, or caspase 9, except for salvianolic acid B. As shown in Table 1, the CDOCKER interaction energy ranged from −3.43985 to 33.9346 kcal/mol between ferulic acid and human Bcl-xl, Bak1, Bcl-2, procaspase 3, and caspase 9 (Table 1 and Figure 2A–E). Ferulic acid interacted with human Bcl-xl via hydrogen bond formation, charge interaction, and π-π interaction with Leu108, Arg102, and Phe105 at the active site, respectively (Figure 2). There was a charge interaction and π-π interaction between ferulic acid and Bak1 with Glu33 and Phe40, respectively. Ferulic acid interacted with procaspase 3 via hydrogen bond formation and charge interaction at Lys159, and ferulic acid bound to caspase 9 via hydrogen bond formation at Arg180 and Gln285 and charge interaction at Arg180, Gln285, and Ser287 (Figure 2). In contrast, there was no hydrogen bond formation, charge interaction, or π-π interaction between ferulic acid and Bcl-2 (Figure 2).

Table 1

Molecular interactions between ferulic acid, ligustilide, succinic acid, vanillic acid, tanshinone IIA, tanshinone IIB, danshensu, salvianolic acid A, salvianolic acid B, salvianolic acid C, protocatechuic aldehyde, and β-boswellic acid and human Bcl-xl, Bak1, Bcl-2, procaspase 3, and caspase 9 using Discovery Studio® program 3.1

Binding target	CDOCKE interaction energy (kcal/mol)	H-bond number	Residues involved in H-bond formation	Charge interactions	Residues involved in charge interactions	π-π stacking	Residues involved in π-π stacking
Ferulic acid
Bcl-xl (PDB ID: 4TUH)	27.8853	1	O-Leu108	1	O-Arg102	1	Phe105
Bak1 (PDB ID: 2JCN)	23.8894	0	–	1	O-Glu33	1	Phe40
Bcl-2 (PDB ID: 2O21)	−3.43985	0	–	0	–	0	–
Procaspase 3 (PDB ID: 4JRO)	32.8532	1	O-Lys159	1	O-Lys159	0	–
Caspase 9 (PDB ID: 2AR9)	33.9346	2	O-Arg180O-Gln285	3	O-Arg180O-Gln285O-Ser287	0	–
Ligustilide
Bcl-xl (PDB ID: 4TUH)	24.0792	0	–	0	–	0	–
Bak1 (PDB ID: 2JCN)	25.9178	0	–	0	–	1	Phe40
Bcl-2 (PDB ID: 2O21)	−15.9741	1	O-Phe195	0	–	0	–
Procaspase 3 (PDB ID: 4JRO)	22.4759	1	O-Arg15	0	–	1	Arg15
Caspase 9 (PDB ID: 2AR9)	24.9135	0	–	0	–	0	–
Succinic acid
Bcl-xl (PDB ID: 4TUH)	23.8093	0	–	1	O-Arg102	0	–
Bak1 (PDB ID: 2JCN)	17.7988	0	–	1	O-Glu33	0	–
Bcl-2 (PDB ID: 2O21)	20.7002	0	–	0	–	0	–
Procaspase 3 (PDB ID: 4JRO)	40.3633	2	O-Asn37O-Gly38	2	O-Arg15O-Arg15	0	–
Caspase 9 (PDB ID: 2AR9)	34.9818	3	O-Arg180(br/O-Gln285O-Ser287	3	O-Arg178O-Arg180O-Gln285	0	–
Vanillic acid
Bcl-xl (PDB ID: 4TUH)	27.1178	0	–	0	–	0	–
Bak1 (PDB ID: 2JCN)	19.8383	1	H-Phe111	1	O-Leu118	0	–
Bcl-2 (PDB ID: 2O21)	16.0496	0	–	0	–	0	–
Procaspase 3 (PDB ID: 4JRO)	29.7815	4	H-Gly12O-Arg15O-Gly18O-Asn90	1	O-Arg15	1	O-Arg15
Caspase 9 (PDB ID: 2AR9)	33.0769	1	O-Arg180	2	O-Arg178O-Arg180	0	–
Tanshinone IIA
Bcl-xl (PDB ID: 4TUH)	10.3489	0	–	0	–	5	Arg139Arg139Phe105Phe105Phe105
Bak1 (PDB ID: 2JCN)	19.5827	0	–	0	–	3	Phe40Phe40Phe40
Procaspase 3 (PDB ID: 4JRO)	30.0502	0	–	0	–	1	Arg15
Caspase 9 (PDB ID: 2AR9)	30.1231	0	–	0	–	0	–
Tanshinone IIB
Bcl-xl (PDB ID: 4TUH)	9.27765	1	H-Ala142	0	–	5	Arg139Arg139Phe105Phe105Phe105
Bak1 (PDB ID: 2JCN)	16.1665	0	–	0	–	3	Phe40Phe40Phe40
Procaspase 3 (PDB ID: 4JRO)	33.0863	3	O-Ile17H-Gly92O-Gly92	0	–	3	Arg15
Caspase 9 (PDB ID: 2AR9)	35.7031	0	–	0	–	1	Phe351
Danshensu
Bcl-xl (PDB ID: 4TUH)	30.1811	1	O-Arg102	0	–	1	O-Arg139
Bak1 (PDB ID: 2JCN)	27.7588	1	O-Gln66	0	–	1	Phe40
Bcl-2 (PDB ID: 2O21)	13.0233	0	–	0	–	1	Ile186
Procaspase 3 (PDB ID: 4JRO)	39.3839	5	H-Ser14O-Ser14H-Arg15O-Gly18O-Asn-90	2	O-Arg15O-Asn37	1	Arg15
Caspase 9 (PDB ID: 2AR9)	34.2947	1	O-Gln285	1	O-Arg180	1	Arg180
Salvianolic acid A
Bcl-xl (PDB ID: 4TUH)	−25.7507	3	H-Glu98H-Ser145O-Ser145	0	–	3	Arg102Phe105Arg132
Procaspase 3 (PDB ID: 4JRO)	53.6318	3	H-Ser14O-Ser14O-Asn37	1	O-Arg15	2	Arg15Arg95
Caspase 9 (PDB ID: 2AR9)	−14.8647	4	O-Arg180H-Thr181H-Thr181O-Gln285	0	–	2	Arg178Arg180
Salvianolic acid C
Bcl-xl (PDB ID: 4TUH)	−37.3485	3	O-Arg102H-Glu129H-Glu129	0	–	3	Arg102Arg102Arg102
Procaspase 3 (PDB ID: 4JRO)	52.9465	7	O-Gly18O-Asn35O-Asn35O-Asn37O-Ala62O-Val64O-Asn90	2	O-Arg15O-Gly38	2	Arg15Arg15
Caspase 9 (PDB ID: 2AR9)	32.6288	5	O-Arg180O-Arg180O-Gln285O-Ser287O-Gly360	1	O-Asp186	2	Arg178Arg180
Protocatechuic aldehyde
Bcl-xl (PDB ID: 4TUH)	24.6477	2	O-Arg102H-Ser145	0	–	0	–
Bak1 (PDB ID: 2JCN)	18.8431	0	–	0	–	1	Phe40
Bcl-2 (PDB ID: 2O21)	11.8606	0	–	0	–	0	–
Procaspase 3 (PDB ID: 4JRO)	16.6186	4	H-Gly12H-Arg15O-Gly18O-Asn90	0	–	1	Arg15
Caspase 9 (PDB ID: 2AR9)	21.4411	2	O-Arg180H-Asp186	0	–	0	–
β-boswellic acid
Procaspase 3 (PDB ID: 4JRO)	31.6175	1	O-Arg15	0	–	0	–
Caspase 9 (PDB ID: 2AR9)	7.52648	1	O-Arg178	0	–	0	–

Note: The molecular docking of salvianolic acid B to all protein molecules failed.