Int J App Pharm, Vol 16, Issue 2, 2024, 237-246Original Article

EXPLORATION OF THE ACTIVE COMPOUNDS OF MORINGA OLEIFERA LAM AS HIV-1 REVERSE TRANSCRIPTASE INHIBITOR: A NETWORK PHARMACOLOGY AND MOLECULAR DOCKING APPROACH

MELANDA FITRIANA1, ABDUL MUN’IM2*, FIRDAYANI3, WIRAWAN ADIKUSUMA4

1Graduate Program, Faculty of Pharmacy, Universitas Indonesia, Gedung Fakultas Farmasi Kampus UI-16424, Indonesia. 2Department of Pharmacognosy-Phytochemistry, Faculty of Pharmacy, Universitas Indonesia, Gedung Fakultas Farmasi Kampus UI-16424, Indonesia. 3,4Research Center for Vaccine and Drugs, National Research and Innovation Agency (BRIN), South Tangerang-15314, Indonesia. 4Department of Pharmacy, University of Muhammadiyah Mataram, Mataram 83127, Indonesia
*Corresponding author: Abdul Mun’im; *Email: munim@farmasi.ui.ac.id

Received: 13 Nov 2023, Revised and Accepted: 01 Jan 2024

ABSTRACT

Objective: This study aims to predict the active compound of Moringa oleifera for the treatment of Human Immunodeficiency Virus (HIV), specifically targeting the HIV-1 reverse transcriptase (HIV-1 RT) enzyme using network pharmacology and molecular docking approach.

Methods: The active ingredients of M. oleifera, were screened from the Knapsack database. Subsequently, HIV-1 RT and its related target compounds were retrieved from the Genecard database. The analysis of common targets involved protein-protein interactions (PPI) analysis using string databases and constructing interaction IDs using Cytoscape software. Gene Ontology (GO) functional and the Kyoto Encyclopedia of Genes and Genomes (KEGG) pathway analyses were performed. Molecular docking studies were conducted using AutoDock Vina software to validate the results of the network pharmacological analysis.

Results: A total of 63 active ingredients and 8601 targets related to HIV-1 RT were identified. The network analysis, encompassing GO and KEGG enrichment, revealed strong associations of common targets with key signaling pathways such as Tumor Necrosis Factor (TNF), Toll-Like Receptor (TLR), and apoptosis. Additionally, 11 compounds of M. oleifera, including apigenin, benzyl isothiocyanate, benzylamine, caffeic acid, ferulic acid, epicatechin, kaempferol, gallic acid, luteolin, syringic acid and vanillin were identified as potential vital compounds. Molecular docking analysis highlighted apigenin and kaempferol as the most promising compounds, exhibiting the lowest binding affinity to the HIV-1 RT enzyme. These compounds correlated with caspase-3(CASP3), caspase-9 (CASP9), and BCL2 Apoptosis Regulator (BAX) protein, stimulating cell apoptosis through multiple pathways.

Conclusion: The study highlighted that apigenin and kaempferol are potential compound of M. oleifera in HIV-1 treatment through inhibition activity at HIV-1 RT Enzyme.

Keywords: HIV-1 reverse transcriptase, Moringa oleifera, Network pharmacology, Molecular docking

© 2024 The Authors. Published by Innovare Academic Sciences Pvt Ltd. This is an open access article under the CC BY license (https://creativecommons.org/licenses/by/4.0/)
DOI: https://dx.doi.org/10.22159/ijap.2024v16i2.49855 Journal homepage: https://innovareacademics.in/journals/index.php/ijap

INTRODUCTION

Human Immunodeficiency Virus (HIV) remains a global health issue and causes the deaths of 630,000 people by the end of 2022 [1]. The objectives of HIV treatment are to impede the advancement of HIV infection to Acquired Immune Deficiency Syndrome (AIDS) and extend the lifespan of individuals with AIDS while maintaining a high standard of living. Antiretroviral medications (ARV) can impede viral replication by suppressing the activity of enzymes implicated in viral replication, specifically reverse transcriptase (RT), protease (PR), and integrase (IN). One of the most efficient interventions occurs at the stage of the reverse transcription process [2]. HIV-1 RT is crucial in facilitating the creation of double-stranded DNA from single-stranded HIV RNA [3]. A downside of ARV therapy is the presence of numerous side effects, leading to reduced patient adherence and stimulating the emergence of resistance [4]. Resistance to antiretroviral drugs are the essential development of research on compounds as anti-HIV agents, including those from a natural sources [5].

oleifera is traditionally used in HIV treatment in South Africa [6]. This plant is native and grows widely in northern India and spread to the continents of Africa and Asia [6]. The plant is widely used for treatment due to its high nutritional content and has various activities such as an antioxidant [7], antimicrobial, anti-inflammatory, antihypertensive, immunomodulator, and as hypoglycemic agent [6, 8] and hepatoprotector [9]. Based on the analysis carried out by Kerdsomboon [10], the leaf extract contains flavonoids and phenolics including apigenin, protocatechuic acid, gallic acid, chlorogenic acid, vanillic acid, caffeic acid, catechin, rutin, para-coumaric acid, ferulic acid, quercetin, rosmarinic acid, cinnamic acid, and kaempferol. The major compounds are quercetin, ellagic acid, chlorogenic acid, niazimicin, gallic acid, rutin, ferulic acid, and moringin [11]. Furthermore, this leaf also contains glycoside compounds such as apigenin glucoside, kaempferol 3-O-glucoside, kaempferol diglucoside, and kaempferol-di-acetyl-rhamnoside [12].

A previous study revealed that water and 50% methanol extracts of M. oleifera have inhibitory activity on the HIV-1 RT with IC50 values of 9 µg/ml and 79 µg/ml, respectively [3]. However, the active compound that has HIV-1 RT inhibition activity remains unclear. The present study aims to determine the active compound in the M. oleifera leaves that is responsible for the inhibitory activity of the HIV-1 RT. This active compound could be predicted through network pharmacology and molecular docking. By constructing an interactive network connecting drug, component, target, pathway, and disease and using GO and KEGG pathway enrichment analysis along with molecular docking technology, the potential molecular mechanism of M. oleifera as an HIV-1 RT inhibitor was explored. Our research findings guide future basic experimental research.

MATERIALS AND METHODS

Data mining of active compounds in M. oleifera

Bioactive compounds from M. oleifera were retreived from the Knapsack database (http://www.knapsackfamily.com/KNApSAcK/, NAIST Comparative Genomics Laboratory, Japan), accessed on July 23, 2023. The Knapsack family database integrates information about plant species and their secondary metabolite content, supporting various plant-related research area, including metabolomics and plant taxonomy studies [13]. The physicochemical and absorption, distribution, metabolism, excretion and toxicity (ADMET) properties of the compounds were obtained from PKCSM Pharmacokinetics (https://biosig.lab.uq.edu.au/pkcsm/, American Chemical Society) [14] and Tox Prediction (https://tox-new.charite.de/protox_II/ index.php) [15], accessed on July 23, 2023 using the canonical name. These pharmacokinetics (ADMET) studies and drug-likeness were predicted by adopting Lipinski’s rule of five parameters: molecular weight ≤ 500 Da, Log p ≤ 5, Hydrogen bond acceptor (HBA) ≤ 10, Hydrogen bond donor (HBD) ≤ 5 [16, 17]. A Log P value greater than 5 indicates that the compound has poor permeation and absorption. Therefore, a molecule will have good bioavailability if it meets the criteria of Lipinski's Rule of 5 [17]. Human intestine absorption (HIA) should be ≥ 30% and not act as Cytochrome P450 2D6 (CYP2D6) inhibitors with class toxicity>3.

Target of HIV disease

The target proteins corresponding to the active compounds of M. oleifera were sourced from GeneCard (www. genecards. org) [18], accessed on July 23, 2023, using the compound name as a keyword. The top 30 targets listed were collected based on their codes and target names. The target of HIV disease was collected from hsa05170 HIV 1 infection on KEGG (https://www.genome.jp/entry/hsa05170, Keisha Laboratories). Specifically, HIV-1 RT, was extracted from GeneCard (www.genecards.org), accessed on July 23, 2023, using the keyword "HIV-1 Reverse Transcriptase". The entire set of targets listed was taken based on data collection of codes and target names.

Active compounds-target interaction

The relationship between target compounds and disease was visualized using a Venn diagram (https://bioinfogp.cnb.csic.es/tools/venny/) by inputting data related to the target protein for HIV disease and the target protein of the active compounds from M. oleifera.

Analysis of protein-protein interaction (PPI)

PPI analysis was conducted using the Search Tool for the Retrieval of Interacting Genes (STRING, https://string-db.org/) by inputting the target protein set obtained from Venn diagram intersection. Homo Sapiens species restrictions and confidence score>0.4 were applied to ensure the reliability and confidence of the results [19, 20]. Cytoscape 3.9.1 was utilized to create networks and quantify the data obtained from STRING by calculating closeness, degree, and between centrality as degree values. Degree, a vital parameter within the topological structure, is employed to evaluate the importance of each node in the PPI network. It represents the number of connections to a node and is commonly used to illustrate the topological significance of a protein within the PPI network. Furthermore, hub target genes are identified as nodes with a 'Degree' exceeding double the median degree of the network [20]. Subsequently, the top 10 targets with the highest values were selected as the main targets [21].

Functional enrichment analysis

Enrichment analysis, covering GO functions (molecular function–MF, biological process–BP, and cellular component–CC) and KEGG pathways, was conducted using ShinyGO 0.77 (http://bioinformatics.sdstate.edu/go/, South Dakota State University) [22]. The target protein of HIV-1 RT and the active compound of M. oleifera served as input, with the species "Human", False Discovery Rate (FDR) cutoff at 0.05, and display of 30 pathways [23]. ShinyGO is an intuitive, visual application specifically designed for enrichment analysis. This tool offers more comprehensive gene sets associated with TF and miRNA target genes for human, mouse, and Arabidopsis. By using ShinyGO, it becomes possible to gain a summary of gene associations through GO enrichment and explore KEGG pathways [22]. The GO functions were visualized using a histogram bar plot and the resulting pathways were correlated with HIV disease.

Molecular docking

The active compounds of M. oleifera were drawn using the Marvin application (Chemaxon https://www.chemaxon.com, trial version) with the canonical smile and were saved in PDB format. The protein target was obtained through the RCSB PDB website (https://www.rcsb.org/) using "HIV-1 Reverse Transcriptase" as a keyword. The X-Ray crystal 3MEE was considered due to the lower resolution (<2.5 A) [24] and its complex with TMC278 (Rilpivirine), which is a second-generation HIV-1 RT. When creating the macromolecule 3MEE, the mutated amino acids from the first-generation were considered [25]. Proteins were downloaded in PDB format, visualized, and corrected for missing residues using the Chimera 1.17.1 application (UCSF Resource for Biocomputing, University of California, San Francisco) [26]. The process of preparing protein molecules as both receptors and ligands involved the utilizing Discovery Studio 2021 (BIOVIA, Dassault Systèmes, Biovia Discovery Studio, License, San Diego: Dassault Systèmes, 2023) [27]. This is included the removal of water molecules and heteroatoms.

Molecular docking is conducted to validate the relationship between compounds and the target. Molecular docking was performed using the PyRx software (version 0.8) [28] to investigate a range of ligand orientations, conformations, and binding affinities at the active sites of HIV-1 and ligands. To ensure compatibility with AutoDock Vina docking, all ligands were converted into the PDBQT format using Open Babel software[29]. Subsequently, a grid box was established to encompass the entire protein for a blind docking strategy. Validation of native ligands and receptors was carried out using the AutoDock Vina application, utilizing grid box settings derived from optimization. The grid box settings include a center at X: 10.9796, Y: 12.5195, Z: 16.2002, and dimensions at X: 12.4367, Y: 11.7488, Z: 14.0701. Gridbox validation was conducted using the Yasara application [30] with a Root-Mean-Square Deviation (RSMD) acceptable value of<2 Å (Angstroms). Following this, the results of molecular docking were visualized, and the nonbonding interactions between protein-ligand complexes and docking poses were assessed using BIOVIA Discovery Studio Client 2021 [27]. The conformation with the most negative binding affinity (in kcal/mol) and RSMD value closest to 0 was selected as the lead compound [31].

RESULTS

Active compounds of Moringa oleifera

A total of 63 M. oleifera active compounds were obtained from the Knapsack database, accessed on July 23, 2023. Screening for physicochemical properties was carried out using Lipinski’s rules of five, which include criteria of molecular weight ≤ 500 Da, Log p ≤ 5, HBA ≤ 10, (HBD ≤ 5 [16, 17] and 35 active compounds (table 1) met these criteria and were used in further analysis.

Fig. 1: Venn diagram of M. oleifera active compounds, HIV-1 RT Enzyme and HIV-1 Infection

Active compounds-target interaction

The outcomes derived from the Genecard search, accessed on July 23, 2023, revealed 309 targets associated with the active components found in M. oleifera, along with 8,601 targets related to HIV-1 RT. Furthermore, in the KEGG pathway 'HIV-1 infection,' 82 target proteins were identified. Subsequently, all the targets will be analyzed using a Venn diagram to illustrate the overlapping proteins. The Venn diagram (fig. 1) shows that there were 10 intersections between the target active compounds of M. oleifera, HIV-1 RT, and HIV-1 Infection. It was shown that 77% of M. oleifera target compounds were associated with HIV-1 RT. Additionally, 10 proteins named Caspase 3 (CASP3), BCL2 Apoptosis Regulator (BCL2), Caspase 8 (CASP8), CD4 Molecule (CD4), C-X-C Motif Chemokine Receptor 4 (CXCR4), Toll Like Receptor 2 (TLR2), Toll Like Receptor 4 (TLR4), Caspase 9 (CASP9), C-C Motif Chemokine Receptor 5 (CCR5),BCL2 Associated X, Apoptosis Regulator (BAX) were identified as potential targets correlated with M. oleifera active compounds and HIV-1 disease. These targets will be analyzed for protein-protein interactions using the STRING database and visualized using Cytoscape 3.9.1.

Table 1: Information on the 35 active compounds of M. oleifera that meet Lipinski’s rules of five and ADMET parameters

No Compound MW LogP HBA HBD HIA CYP2D6 inhibitor Toxicity class
1 Epicatechin 290,271 1.5461 6 5 65,425 No 6
2 alpha-Phellandrene 136,238 3.1648 0 0 96,548 No 6
3 Niazimin A 383,397 0.7098 8 3 44,845 No 5
4 Niazicin A 385,438 0.4846 8 3 45,796 No 5
5 Niazimicin A 357,428 0.3039 7 4 50,995 No 5
6 Niazinin B 343,401 0.3251 7 4 51,997 No 5
7 Niazicinin A 369,370 0.3197 8 3 56,015 No 5
8 Niazidin 369,370 0.3197 8 3 56,015 No 5
9 Niazinin A 343,401 0.3251 7 4 56,775 No 5
10 Niaziminin B 399,465 1,286 8 3 58,560 No 5
11 Niazirine 279,292 -0.04102 6 3 59,546 No 5
12 Niazirinine 321,329 0.52978 7 2 60,000 No 5
13 Niaziminin A 399,465 1,286 8 3 68,647 No 5
14 Caffeic acid 180,159 1,1956 3 3 69,407 No 5
15 Isorhamnetin 316,265 2.2910 7 4 69,585 No 5
16 Rhamnetin 316,265 2.2910 7 4 72,598 No 5
17 Kaempferol 286,239 2.2824 6 4 74,290 No 5
18 Luteolin 286,239 2.2824 6 4 75,170 No 5
19 Gentisic acid 154,121 0.796 3 3 80,078 No 5
20 Apigenin 270,240 2.5768 5 3 90,555 No 5
21 p-Coumaric acid 164,160 1.4900 2 2 93,494 No 5
22 o-Coumaric acid 164.16 1.49 2 2 93.4941 No 5
23 Genistein 270.24 2.5768 5 3 93,831 No 5
24 Daidzein 254,241 2.8712 4 3 94,772 No 5
25 Gallic acid 170,120 0.5016 4 4 43,374 No 4
26 Moringyne 312,318 -0.73986 7 4 54,312 No 4
27 Syringic acid 198,174 1.1076 4 2 74,963 No 4
28 Vanillin 152,149 1.2131 3 1 78,206 No 4
29 Ellagic acid 302,194 1.3128 8 4 83,817 No 4
30 Benzylamine 107,156 1.1453 1 1 87,626 No 4
31 Salicylic acid 138,122 1.0904 2 2 89,961 No 4
32 Eugenol 164,204 2.1293 2 1 91,651 No 4
33 Sinapic acid 224,212 1.5072 4 2 91,779 No 4
34 Ferulic acid 194,186 1.4986 3 2 93,490 No 4
35 Benzyl isothiocyanate 149,218 2.2894 2 0 94,988 No 4

MW: Molecular Weight, Log p: Log of octanol/water partition coefficient, HBA: Hydrogen Bond Acceptor, HBD: Hydrogen Bond Donor, HIA: Human intestinal absorption

PPI network

The establishment of the PPI network involved entering 10 protein targets into the STRING database. In this network, there were 10 nodes and 32 edges that correlated with M. oleifera and HIV disease. Finally, an analysis of these 10 proteins using Cytoscape software was performed to determine their total degree, closeness, and betweenness centrality values, illustrating the proteins with the greatest impact within this network. These targets are essential in HIV therapy, underscoring their significance in potential treatment strategies the information on the total degree value of this network is shown in table 2 and fig. 2.

Table 2: Information of total degree value on PPI Network

No Protein code Protein name Betweenness Closeness Degree Total
1 CASP3 Caspase 3 0,15 1,00 9 10,15
2 BCL2 BCL2 Apoptosis Regulator 0,15 1,00 9 10,15
3 CASP8 Caspase 8 0,11 1,00 8 9,11
4 CD4 CD4 Molecule 0,06 1,00 8 9,06
5 CXCR4 C-X-C Motif Chemokine Receptor 4 0,06 1,00 8 9,06
6 TLR2 Toll Like Receptor 2 0,00 1,00 7 8,00
7 TLR4 Toll Like Receptor 4 0,00 0,00 8 8,00
8 CASP9 Caspase 9 0,07 0,80 7 7,87
9 CCR5 C-C Motif Chemokine Receptor 5 0,00 1,00 6 7,00
10 BAX BCL2 Associated X, Apoptosis Regulator 0,00 0,64 4 4,64

Fig. 2: PPI network of the targets in M. oleifera, HIV-1 RT and HIV-1 Infection the darker color and node size indicated to the impact of the network

Results of gene ontology and KEGG pathway analysis

Intersection target proteins obtained from the Venny platform were analyzed using ShinyGo 0.77 by inputting 10 target proteins, specifying the species as “human”, and setting FDR cut off 0.05. GO enrichment was performed, including GO BP, GO MF, and GO CC. The GO enrichment results were in 1000 BP pathway, 56 CC pathway, and 107 MF pathway. The top 30 of each GO pathway are shown in fig. 3. The KEGG enrichment analysis indicated that apoptosis was the primary pathway associated with those 10 targets, as shown by fold enrichment (fig. 4a). This demonstrates the extent to which proteins/genes from a specific pathway are present in the experimental data compared to random data. However, when evaluated through FDR enrichment (fig. 4b), the HIV-1 infection exhibited the most minimal FDR enrichment value (FDR<0.05). This suggests that this pathway significantly contributes to the conducted networking analysis.

(a)

(b)

(c)

Fig. 3: Top 30 gene ontology pathways, a. GO Biological process (BP), b. GO Cellular component (CC), c. GO molecular function (MF)

(a)

(b)

Fig. 4: KEGG pathway enrichment, the analyzed results based on fold enrichment (a) and FDR enrichment (b)

Moreover, these 10 shared proteins were linked to 11 bioactive compounds found in M. oleifera (fig. 5), specifically apigenin, benzyl isothiocyanate, benzylamine, caffeic acid, epicatechin, ferulic acid, gallic acid, kaempferol, luteolin, syringic acid, and vanillin. Thereafter, these compounds were subsequently employed as ligands for the molecular docking analysis.

Fig. 5: M. oleifera compounds–HIV-1 disease protein interaction

Molecular docking

Molecular docking, using Autodock Vina, was applied to predict ligands and HIV-1 RT binding affinity. Docking simulations were conducted using Autodock Vina, resulting in 9 ligand conformations bound to the receptor and ranked based on their binding energy [32]. The HIV-1 RT X-Ray crystal structures (3MEE: 2,4 A) in complex with TMC278 (Rilpivirine) a second generation of Non-Nucleoside Reverse Transcriptase Inhibitors (NNRTI) [33–35] were downloaded from the Protein Data Bank (https://www.rcsb.org/ accessed on 23 July 2023). Molecular docking was validated by ensuring that the grid_box encompassed the receptor and ligand binding site. Fig. 6 demonstrated that the native ligand before and after a docking overlapped and had the same amino acid arrangement (fig. 7 and table 3). Confirmation results through the Yasara application showed that the generated RMSD was 0.5159 Å. Such validation of a grid box is expected to improve the results of molecular docking with a flexible ligand/protein binding site and distinguish the compounds that can bind to a protein target from false positives [25].

Fig. 6: Visualization of native ligands before (blue molecules) and after docking (yellow molecules)

(a) (b)

Fig. 7: Visualization of the interaction of receptor amino acids and native ligands before docking (a) and after docking (b)

Table 3: Amino acid interaction of native ligand and macromolecule (3MEE)

No Native ligand Amino acid
1 TMC278 before docking GLY A: 190, VAL A: 179, PRO A: 225, PHE A: 227, LEU A: 100, LEU A: 234, TYR A: 181, LYS A: 101, PRO A: 236, TYR A: 318, HIS A 235, PRO A: 95, LYS A: 103, VAL A: 106, LEU A: 228, TRP A: 229, TYR A: 188, LYS A: 102
2

TMC278

after docking

 LYS A: 102, LYS A: 103, TYR A: 318, HIS A: 235, PRO A: 225, VAL A: 106, PRO A: 236, LEU A: 234, PHE A: 227, GLY A: 190, LEU A: 100, LEU A: 228, LYS A: 101, PRO A: 95, VAL A: 179, TYR A: 181, TRP A: 229, TYR A: 188

Table 4: Molecular docking result

No Compound name Binding affinity (kcal/mol)
1 Native Ligand -12.1
2 Nevirapine -10.2
3 Apigenin -9.5
4 Kaempferol -9.5
5 Luteolin -9.3
6 Epicatechin -9.0
7 Ferulic acid -9.0
8 Caffeic acid -7.0
9 Benzyl isothiocyanate -6.7
10 Syringic acid -6.5
11 Gallic acid -6.2
12 Vanillin -5.9
13 Benzylamine -5.8

In general, a lower energy level in the conformation of the receptor and ligand increases the likelihood of interaction, and an affinity energy of ≤ −5 kcal/mol is considered indicative of a strong binding affinity [21]. Table 4 indicates that our ligands exhibited binding affinities with HIV-1 RT in the range of -5.8 to -9.5 kcal/mol. These results demonstrated that all selected ligands have strong binding affinity with HIV-1 RT as target molecules. Among them, apigenin and kaempferol showed the highest binding affinity to HIV-1 RT at-9.5 kcal/mol. The binding occurs in the subunit p66, which plays a catalytic role in the polymerase domain [36] and forms hydrogen bonding with Lys 101 residues of HIV-1 RT. This binding affinity can be stabilized by van der Waals, pi-sigma and pi-alkyl amino acid interaction (fig. 8). These results show that the docked ligands have a stable complex with protein and have antiviral activity.

(A) (B)

Fig. 8: Protein and ligands interaction (A) apigenin (B) kaempferol

DISCUSSION

M. oleifera was used for HIV treatment in South Africa [6] and widely used for its high nutritional content and various medicinal properties. It had antioxidant, antimicrobial, anti-inflammatory, antihypertensive activities, immunomodulatory, antiglycemic activity and antiviral activity [6, 8, 37]. In Indonesia, M. oleifera leaf is traditionally used to treat skin disorders like ringworm and rashes, beriberi, and gout [38]. The potential for its utilization could have been further developed, including as an antiviral agent. Research regarding the activity of M. oleifera leaf extracts against HIV has been conducted on the HIV-1RT enzyme, showing that both water and 50% methanol extracts exhibited inhibitory activity against the HIV-1 RT enzyme with IC50 values of 9 µg/ml and 79 µg/ml, respectively [3]. However, the active compounds responsible for inhibiting the HIV-1 RT enzyme in M. oleifera leaf extracts have not been identified. Therefore, pharmacological network and molecular docking methods can be utilized to determine the active compounds. Nowadays, network pharmacology is widely used in the drug development and utilization process. Through this method, essential drug characteristics can be optimized and, speeding up and simplifying the drug discovery process [39].

First, screening for physicochemical properties was carried out using Lipinski's rule of five, an approach based on the principle of drug-likeness that relates to the bioavailability of oral drugs, including molecular weight, log P, and the number of hydrogen donors and acceptors [21, 40]. Molecular weight shows the size, density, and mass/volume of active compound, which modulates ligand binding [41]. Low molecular weight, typically 300-500 Da or less can facilitate efficient binding interactions [42, 43]. Small molecules possess high permeability and affect membrane transport. This membrane transport is also affected by Log-P, which determines whether compounds are lipophilic or lipophobic [43]. The H-bond donor and H-bond acceptor present in the structure of active compound also play a vital role in membrane transport, protein-compound interaction, distribution, and solubility [41]. Based on our study there are 35 active compounds were obtained from M. oleifera that meet the rules (table 1) and have the potential for future analysis.

In addition, other crucial properties in drug discovery are related to absorption, distribution, metabolism, excretion, and toxicity. The causes of some failures in the discovery of new chemical entities (NCEs), accounting for 50%-60%, are poor ADMET properties and the occurrence of drug side effects. The bioavailability of a drug is determined by various factors, such as its ability to dissolve within the digestive system, its permeability across intestinal membranes, and its metabolic interactions in the liver. In the case of the majority of compounds, roughly 64% of their bioavailability is predominantly shaped by the process of intestinal absorption. Consequently, forecasting intestinal absorption serves as the initial stage in anticipating the bioavailability of these compounds [44]. The assessment of HIA involves a comparison between the amount reaching the vein and the oral dose, resulting in categorization as follows: HIA+(HIA ≥ 70%) denotes strong absorption, HIA±(30%<HIA<70%) indicates moderate absorption and HIA–(HIA ≤ 30%) signifies limited absorption [45]. In this study, an HIA threshold of>30% is used to obtain compounds with moderate and strong absorption, aiming for a favorable level of bioavailability. CYP2D6 is an enzyme that plays a key role in the metabolism of various drugs and chemical compounds within the human body. This enzyme is essential for drug metabolism and can influence on the effectiveness and safety of pharmaceuticals. Consequently, inhibiting CYP2D6 may raise the potential for toxicity. Hence, in this research, compounds that do not hinder the CYP2D6 enzyme were selected [46]. According to the Globally Harmonized System of Classification and Labelling of Chemicals (GHS), there are six acute toxicity classes. The choice of these toxicity classes from classes 4-6 is based on their classification as non-toxic, with LD50 values in class 4 ranging from 300 mg/kg to 2000 mg/kg, class 5 ranging from 2000 mg/kg to 5000 mg/kg, and class 6 exceeding>5000 mg/kg, all falling within the non-toxic category [15].

The interaction between active compound targets and diseases is constructed through protein-protein analysis, thus providing an overview of potential protein targets in HIV treatment. The HIV-1 life cycle is an intricate sequence of events that includes various stages, starting from the viral entry into the host cell and culminating in the assembly of new viral particles. At the core of this cycle is the activity of various viral proteins, notably reverse transcriptase, integrase, and protease enzymes. The process initiates with the attachment of the viral envelope protein, glycoprotein 120 (gp120), to the CD4 receptor on host cells. This is followed by the attachment to chemokine receptors (such as CCR5 or CXCR4) and subsequent merging of the viral and host cell membranes. The viral capsid containing the genome and enzymes then enters the cell, and undergoes transcription, integration, and translation, leading to the assembly of new viral particles [36]. In our study, we focused on the critical stage of transcription, particularly the action of the HIV-1 RT enzyme. This enzyme converts the single-stranded viral RNA into double-stranded DNA, making it a prime target for antiretroviral therapy.

Our network pharmacology and molecular docking analyses revealed two potential inhibitors of HIV-1 RT from M. oleifera: apigenin and kaempferol. Both compounds exhibited the lowest binding affinity compared to others (table 4). This binding affinity is associated with the amino interaction of both compounds in the hydrophobic pocket of the allosteric site, which contains the amino acids Leu A: 234, Leu A: 100, Lys A: 101, Lys A: 103, Val A: 106, Val A: 179, Phe A: 227, Trp A: 229, Tyr A: 188, and Tyr A: 181. These amino acids are essential in the HIV-1 RT binding site and these interactions are similar to the NNRTIs group, including nevirapine, efavirenz, etravirine, and rilpivirine [25]. These results align with the study conducted by Muruganantham [47], that showed apigenin from Hybanthus enneaspermus had the best binding affinity at-7.21 kcal/mol and interacted with the amino acid Lys: 102 through a hydrogen bond [47]. These results suggest their promising potential as anti-HIV agents.

Apigenin has antiviral activity, such as inhibiting several viruses, including enterovirus 71 (EV71), herpes simplex viruses HSV-1 and HSV-2, hepatitis C virus, and influenza virus. The inhibitory mechanism involves inhibiting viral replication by inhibiting intracellular ROS (reactive oxygen species) and disrupting cytokine regulation, resulting in apoptosis. Apigenin disrupts hepatitis C virus replication by interfering with the function of the viral internal ribosome entry sites (IRES) and preventing the activation of the N-terminal c-Jun (JNK) kinase, which is essential for viral replication triggered by EV71 [48]. In the HIV, apigenin contained in Punica granatum leaf extract exhibits HIV-1 RT-RNase inhibitor activity with an IC50 value of 3.7±0.5 µM [49]. Apigenin compound was also tested on the HIV-1 RT enzyme and showing a high percentage of inhibition, at>100% and an IC50 value of 8.5 µg/ml [50].

Kaempferol exhibits a wide range of pharmacological effects, including anti-inflammatory properties via blocking the NF-KB pathway and suppressing the release of pro-inflammatory cytokines, including IL-8 and TNF-alpha in cell line-based in vitro experiments. Apart from that, the anti-inflammatory mechanism of kaempferol also targets the Mitogen-Activated Protein Kinases (MAPK) activation pathway, thereby inhibiting the formation of inflammatory cytokines [51]. In addition, kaempferol isolated from Securigera securidaca has acted as an HIV-1 RT inhibitor with an IC50 value of 50 µg/ml [52]. Kaempferol compound was also tested on HIV-1 RT enzyme and showed a good percentage of inhibition at 88.08 % and an IC50 value of 41.4 µg/ml [50].

Apigenin and kaempferol, identified from M. oleifera, demonstrated inhibitory potential against HIV-1 RT. Apigenin exhibited anti-HIV activity by inhibiting the activation of BAX, CASP9, and CASP3, critical players in the apoptotic pathway [53, 54]. Additionally, both compounds displayed inhibitory effects on HIV-1 RT, a crucial enzyme in viral replication, by binding to the enzyme's active site. These findings suggest that M. oleifera possesses a dual mechanism against HIV, targeting both host cells and the virus itself [36].

Based on the results of GO and KEGG enrichment analysis (fig. 3 and 4), it appears that the pathway includes a fusion of virus membrane, apoptosis signaling pathway, virus receptor activity, protein domain specific binding, apoptosis pathway, and HIV-1 infection pathway. Apoptosis is a form of cell death that involves a series of activation processes, including intrinsic and extrinsic pathways The extrinsic pathway begins with activating apoptotic receptors on the cell surface, which belong to the TNF receptor family. The main pathways in the apoptosis process that have an important role in HIV infection include the Fas/FaL pathway, TNF-related apoptosis-inducing ligand (TRAIL), and TNF pathway. Meanwhile, the BCL2 gene family tightly regulates the intrinsic pathway, which consists of proteins that control the release of special factors and caspases from mitochondria. The main effect of this pathway is to release cytochrome-c from mitochondria, which forms a complex with Apoptotic Protease Activating Factor 1 (APAF1). This complex triggers the activation of CASP9, which then activates the executive caspase responsible for triggering the apoptosis process. HIV infection is characterized by viral proteins that have proapoptotic and antiapoptotic properties, including the proteins gp120, Tat, Vpr, Vpu, and Nef [55].

In infectious pathways such as cancer and viral infection, apigenin increases the sensitivity of TNF-alpha, which can cause apoptosis by inhibiting NFKB. The induction of the apoptosis process occurs through increased production of Reactive Oxygen Species (ROS) and calcium (Ca2+), elevating BAX levels and inhibiting BCL2 levels, changing the BAX/BCL2 ratio. This eventually leads to a decrease in mitochondrial membrane permeability, followed by the release of cytochrome C, AIF, and Endo G from mitochondria, and induces activation of CASP9 and CASP3 [54]. The apoptotic pathway through mitochondria is mostly caused by non-receptor disorders, including the presence of toxins, hyperthermia, hypoxia, oxidants, and viral infections, which can affect the permeability of the mitochondrial membrane, causing the pores of the mitochondria to open and release pro-apoptosis factors. Pro-apoptosis includes cytochrome-c, the Second mitochondria-derived activator of caspase (Smac)/direct IAP binding protein with low pI (DIABLO), and the serine protease HtrA2/Omi. This protein activates caspases, including CASP9, which ultimately causes cell apoptosis. Apart from activated caspases, there is activation of the BCL2 protein group, which plays a pro-apoptotic role, such as BAX, and BCL10 [55]. This is followed by the activity of M. oleifera in signaling the apoptotic pathway, where the M. oleifera fraction inhibits the proliferation of the SCC15 cell line and causes apoptosis by involving the activation of CASP3 and BAX [56].

Our study sheds light on potential HIV-1 RT inhibitors derived from M. oleifera. However, it is crucial to recognize the study's limitations. The study's in silico approach heavily relies on computational predictions and molecular docking, necessitating validation through robust experimental investigations. Moreover, the study focused on a specific subset of compounds, yet M. oleifera encompasses a diverse array of compounds that may harbor further promising candidates for combating HIV. To mitigate these limitations, future research should encompass experimental validation of the identified compounds' inhibitory effects of HIV-1 RT, employing both in vitro and in vivo studies. Additionally, a deeper understanding of potential synergistic effects among compounds within the M. oleifera extract is vital for a thorough comprehension its anti-HIV properties.

CONCLUSION

In summary, our study utilized a combination of network pharmacology and molecular docking analyses to shed light on the relationship between M. oleifera and the HIV-1 RT enzyme. The finding identified apigenin and kaempferol as potential marker compounds within M. oleifera with inhibitory effects on HIV-RT inhibitors.

These compounds display dual mechanisms against HIV. Firstly, M. oleifera activates the apoptosis signaling pathway and involves in HIV-1 infection signaling pathway on host cells. Secondly, it exhibits inhibitory activity on the HIV-1 RT enzyme, thwarting viral replication by hindering double-stranded DNA formation. These findings offer a valuable foundation for future research in developing natural HIV treatments, which should be further validated through in vitro and in vivo studies.

ACKNOWLEDGMENT

This research was supported by The National Agency of Drug and Food Control (NADFC) and The Ministry of Research and Technology/BRIN Republic of Indonesia.

FUNDING

The National Agency of Drug and Food Control (NADFC).

AUTHORS CONTRIBUTIONS

Conceptualization: MF, Formal analysis: MF, AM, F, WA, Methodology: MF, Writing–review and editing: MF, AM, F, WA

CONFLICT OF INTERESTS

There is no conflict of interest in the study.

REFERENCES

  1. WHO. UNAIDS data 2023. Geneva: joint United Nations programme on HIV/AIDS; 2023.

  2. Sharma A, Rangari V. HIV-1 reverse transcriptase and protease assay of methanolic extracts of Adansonia digitata L. Int J Pharm Pharm Sci. 2016 Sep 1;8(9):124-7. doi: 10.22159/ijpps.2016v8i9.12485.

  3. Ndhlala A, Cele K, Mulaudzi R, du Plooy C, Venter S, Mashela P. The potential of Moringa oleifera lam. to manage HIV-1 infections and its positive pharmaco-synergy with antiretroviral therapies. Planta Med. Georg Thieme verlag KG. 2016 Dec 14;81(S 01):S1-S381.

  4. Patel PH, Zulfiqar H. Reverse transcriptase inhibitors. Frontiers in HIV Research. Stat. 2023 Jun 25:44-61. PMID 31855348.

  5. Ndhlala AR, Van Staden J, Mashela PW, Du Plooy CP, Abdelgadir HA. Pharmaco-synergistic potential of Moringa oleifera lam. with antiretroviral therapies in managing HIV-1 infections. S Afr J Bot. 2017 Mar;109:358-9.

  6. Aprioku JS, Robinson O, Obianime AW, Tamuno I. Moringa supplementation improves immunological indices and hematological abnormalities in seropositive patients receiving HAARTs. Afr Health Sci. 2022;22(2):1-11. doi: 10.4314/ahs.v22i2.2, PMID 36407401.

  7. Hamed MM, Abdalla AM, Ghareeb MA, Saleh SA. Chemical constituents, in vitro antioxidant activity, oral acute toxicity and LD50 determination of Moringa oleifera leaves. Int J Pharm Pharm Sci. 2017 May 1;9(5). doi: 10.22159/ijpps.2017v9i5.17834.

  8. Chhikara N, Kaur A, Mann S, Garg MK, Sofi SA, Panghal A. Bioactive compounds, associated health benefits and safety considerations of Moringa oleifera L.: an updated review. Nutr Food Sci. 2021;51(2):255-77. doi: 10.1108/NFS-03-2020-0087.

  9. Salama RA, Mohamed M, Shakweer M, Mohamed MM, Elwahab MBA, Shakweer MM. Assessment effect of Aloe vera, Azadirachta indica and Moringa oleifera aqueous extracts on carbon tetrachloride-induced hepatotoxicity in rats. Int J Pharm Pharm Sci. 2016;8(4):83-9.

  10. Kerdsomboon K, Chumsawat W, Auesukaree C. Effects of Moringa oleifera leaf extracts and its bioactive compound gallic acid on reducing toxicities of heavy metals and metalloid in Saccharomyces cerevisiae. Chemosphere. 2021 May 1;270:128659. doi: 10.1016/j.chemosphere.2020.128659, PMID 33757277.

  11. Ntlhamu MI, Ndhlala AR, Masoko P. Exploring the anti-HIV-1 reverse transcriptase, anti-inflammatory, anticancer activities and cytotoxicity of two fermented commercial herbal concoctions sold in limpopo province of South Africa. BMC Complement Med Ther. 2021 Dec 1;21(1):151. doi: 10.1186/s12906-021-03321-2, PMID 34039320.

  12. Hamany Djande CY, Piater LA, Steenkamp PA, Madala NE, Dubery IA. Differential extraction of phytochemicals from the multipurpose tree, Moringa oleifera, using green extraction solvents. S Afr J Bot. 2018 Mar 1;115:81-9. doi: 10.1016/j.sajb.2018.01.009.

  13. Afendi FM, Okada T, Yamazaki M, Hirai Morita A, Nakamura Y, Nakamura K. KNApSAcK family databases: integrated metabolite-plant species databases for multifaceted plant research. Plant Cell Physiol. 2012 Feb;53(2):e1. doi: 10.1093/pcp/pcr165, PMID 22123792.

  14. Pires DE, Blundell TL, Ascher DB. pkCSM: predicting small-molecule pharmacokinetic and toxicity properties using graph-based signatures. J Med Chem. 2015 May 14;58(9):4066-72. doi: 10.1021/acs.jmedchem.5b00104, PMID 25860834.

  15. Banerjee P, Eckert AO, Schrey AK, Preissner R. ProTox-II: a webserver for the prediction of toxicity of chemicals. Nucleic Acids Res. 2018 Jul 2;46(W1):W257-63. doi: 10.1093/nar/gky318, PMID 29718510.

  16. Walters WP. Going further than Lipinski’s rule in drug design. Expert Opin Drug Discov. 2012 Feb 13;7(2):99-107. doi: 10.1517/17460441.2012.648612, PMID 22468912.

  17. Ya’u Ibrahim Z, Uzairu A, Shallangwa G, Abechi S. Molecular docking studies, drug-likeness and in silico ADMET prediction of some novel β-Amino alcohol grafted 1,4,5-trisubstituted 1,2,3-triazoles derivatives as elevators of p53 protein levels. Sci Afr. 2020 Nov;10:e00570.

  18. Stelzer G, Rosen N, Plaschkes I, Zimmerman S, Twik M, Fishilevich S, Nudel R, Lieder I, Mazor Y, Kaplan S, Dahary D, Warshawsky D, Guan‐Golan Y, Kohn A, Rappaport N, Safran M, Lancet D. The gene cards suite: from gene data mining to disease genome sequence analyses. Curr Protoc Bioinformatics. 2016 Jun 20;54(1).

  19. Liu F, Li Y, Yang Y, Li M, Du Y, Zhang Y. Study on mechanism of matrine in treatment of COVID-19 combined with liver injury by network pharmacology and molecular docking technology. Drug Deliv. 2021 Jan 1;28(1):325-42. doi: 10.1080/10717544.2021.1879313, PMID 33517789.

  20. Zhuang Z, Wen J, Zhang L, Zhang M, Zhong X, Chen H. Can network pharmacology identify the anti-virus and anti-inflammatory activities of shuanghuanglian oral liquid used in Chinese medicine for respiratory tract infection? Eur J Integr Med. 2020 Aug;37:101139. doi: 10.1016/j.eujim.2020.101139, PMID 32501408.

  21. Huang X, Rehman HM, Szollosi AG, Zhou S. Network pharmacology-based approach combined with bioinformatic analytics to elucidate the potential of curcumol against hepatocellular carcinoma. Genes (Basel). 2022 Apr 7;13(4):653. doi: 10.3390/genes13040653, PMID 35456457.

  22. Ge SX, Jung D, Yao R. ShinyGO: a graphical gene-set enrichment tool for animals and plants. Bioinformatics. 2020 Apr 15;36(8):2628-9. doi: 10.1093/bioinformatics/btz931, PMID 31882993.

  23. Li X, Wen Z, Si M, Jia Y, Liu H, Zheng Y. Exploration of hanshi zufei prescription for treatment of COVID-19 based on network pharmacology. Chin Herb Med. 2022 Apr;14(2):294-302. doi: 10.1016/j.chmed.2021.06.006, PMID 35382000.

  24. Alom MdM, Bonna RP, Islam A, Alom MdW, Rahman MdE, Faruqe MO. Unveiling neuroprotective potential of spice plant-derived compounds against Alzheimer’s disease: insights from computational studies. Int J Alzheimers Dis. 2023 Sep 15;2023:8877757. doi: 10.1155/2023/8877757, PMID 37744007.

  25. Tarasova O, Poroikov V, Veselovsky A. Molecular docking studies of HIV-1 resistance to reverse transcriptase inhibitors: mini-review. Molecules. 2018 May 21;23(5):1233. doi: 10.3390/molecules23051233, PMID 29883406.

  26. Pettersen EF, Goddard TD, Huang CC, Couch GS, Greenblatt DM, Meng EC. UCSF Chimera-a visualization system for exploratory research and analysis. J Comput Chem. 2004 Oct;25(13):1605-12. doi: 10.1002/jcc.20084, PMID 15264254.

  27. DVD, S Visualizer. Accelrys software Inc. Discov Studio Visualizer Visualizer; 2005. p. 2.

  28. Dallakyan S, Olson AJ. Small-molecule library screening by docking with PyRx; 2015. p. 243-50.

  29. O’Boyle NM, Banck M, James CA, Morley C, Vandermeersch T, Hutchison GR. Open babel: an open chemical toolbox. J Cheminform. 2011 Dec 7;3(1):33. doi: 10.1186/1758-2946-3-33, PMID 21982300.

  30. Krieger E, Vriend G, Spronk C. YASARA-yet another scientific artificial reality application. YASARA Org. 2013;993:51-78.

  31. Nurhasanah N, Fadilah F, Bahtiar A. Prediction of active compounds of muntingia calabura as potential treatment for chronic obstructive pulmonary diseases by network pharmacology integrated with molecular docking. Int J App Pharm. 2023 Jan 7;15(1):274-9. doi: 10.22159/ijap.2023v15i1.46281.

  32. Iheagwam FN, Ogunlana OO, Ogunlana OE, Isewon I, Oyelade J. Potential anti-cancer flavonoids isolated from Caesalpinia bonduc young twigs and leaves: molecular docking and in silico studies. Bioinform Biol Insights. 2019 Jan 7;13:1177932218821371. doi: 10.1177/1177932218821371, PMID 30670919.

  33. Lansdon EB, Brendza KM, Hung M, Wang R, Mukund S, Jin D. Crystal structures of HIV-1 reverse transcriptase with etravirine (TMC125) and rilpivirine (TMC278): implications for drug design. J Med Chem. 2010 May 27;53(10):4295-9. doi: 10.1021/jm1002233, PMID 20438081.

  34. Crisan L, Bora A. Small molecules of natural origin as potential anti-HIV agents: a computational approach. Life (Basel). 2021 Jul 20;11(7):722. doi: 10.3390/life11070722, PMID 34357094.

  35. Azijn H, Tirry I, Vingerhoets J, de Béthune MP, Kraus G, Boven K. TMC278, a next-generation nonnucleoside reverse transcriptase inhibitor (NNRTI), active against wild-type and NNRTI-resistant HIV-1. Antimicrob Agents Chemother. 2010 Feb;54(2):718-27. doi: 10.1128/AAC.00986-09, PMID 19933797.

  36. Spach D. Antiretroviral medications and initial therapy-core concepts; 2023.

  37. Silprasit K, Seetaha S, Pongsanarakul P, Hannongbua S, Choowongkomon K. Anti-HIV-1 reverse transcriptase activities of hexane extracts from some Asian medicinal plants. J Med Plants Res. 2011;5(19):4899-906.

  38. Mardisiswoyo S, Mangunsudarso HR, Nenek Moyang C Puyang W. Jakarta: balai pustaka. Indonesian; 1985.

  39. Guan M, Guo L, Ma H, Wu H, Fan X. Network pharmacology and molecular docking suggest the mechanism for the biological activity of rosmarinic acid. Evid Based Complement Alternat Med. 2021 Apr 11;2021:5190808. doi: 10.1155/2021/5190808, PMID 33936238.

  40. Noe MC, Peakman MC. Drug discovery technologies: current and future trends. Comprehensive medicinal chemistry III. Elsevier; 2017. p. 1-32.

  41. Tijjani H, Olatunde A, Adegunloye AP, Ishola AA. In silico insight into the interaction of 4-aminoquinolines with selected SARS-CoV-2 structural and nonstructural proteins. Coronavirus drug discovery. Elsevier; 2022. p. 313-33.

  42. Bauman JD, Patel D, Dharia C, Fromer MW, Ahmed S, Frenkel Y. Detecting allosteric sites of HIV-1 reverse transcriptase by X-ray crystallographic fragment screening. J Med Chem. 2013 Apr 11;56(7):2738-46. doi: 10.1021/jm301271j, PMID 23342998.

  43. Neidle S. Design principles for quadruplex-binding small molecules. Therapeutic applications of quadruplex nucleic acids. Elsevier; 2012. p. 151-74.

  44. Hou T, Wang J, Li Y. ADME evaluation in drug discovery. The prediction of human intestinal absorption by a support vector machine. J Chem Inf Model. 2007 Nov 1;47(6):2408-15. doi: 10.1021/ci7002076, PMID 17929911.

  45. Radchenko EV, Dyabina AS, Palyulin VA, Zefirov NS. Prediction of human intestinal absorption of drug compounds. Russ Chem Bull. 2016 Feb 8;65(2):576-80. doi: 10.1007/s11172-016-1340-0.

  46. Zhai J, Man VH, Ji B, Cai L, Wang J. Comparison and summary of in silico prediction tools for CYP450-mediated drug metabolism. Drug Discov Today. 2023 Oct;28(10):103728. doi: 10.1016/j.drudis.2023.103728, PMID 37517604.

  47. Muruganantham S. In silico molecular docking analysis of anti HIV-1 RT from Indian medicinal plant Hybanthus enneaspermus. J Nat Sci Math. 2021.

  48. Wang M, Firrman J, Liu LS, Yam K. A review on flavonoid apigenin: dietary intake, ADME, antimicrobial effects, and interactions with human gut microbiota. BioMed Res Int. 2019;2019:7010467. doi: 10.1155/2019/7010467, PMID 31737673.

  49. Sanna C, Marengo A, Acquadro S, Caredda A, Lai R, Corona A. In vitro anti-HIV-1 reverse transcriptase and integrase properties of Punica granatum L. leaves, bark, and peel extracts and their main compounds. Plants (Basel). 2021 Oct 1;10(10). doi: 10.3390/plants10102124, PMID 34685933.

  50. Ko Y. Flavonoids as potential inhibitors of retroviral enzymes. J Korean Soc Appl Biol Chem. 2009;52(4):321-6. doi: 10.3839/jksabc.2009.057.

  51. Yang L, Gao Y, Bajpai VK, El-Kammar HA, Simal Gandara J, Cao H. Advance toward isolation, extraction, metabolism and health benefits of kaempferol, a major dietary flavonoid with future perspectives. Crit Rev Food Sci Nutr. 2023;63(16):2773-89. doi: 10.1080/10408398.2021.1980762, PMID 34554029.

  52. Behbahani M, Sayedipour S, Pourazar A, Shanehsazzadeh M. In vitro anti-HIV-1 activities of kaempferol and kaempferol-7-O-glucoside isolated from Securigera securidaca. Res Pharm Sci. 2014;9(6):463-9. PMID 26339261.

  53. Brentnall M, Rodriguez Menocal L, De Guevara RL, Cepero E, Boise LH. Caspase-9, caspase-3 and caspase-7 have distinct roles during intrinsic apoptosis. BMC Cell Biol. 2013 Jul 9;14(1):32. doi: 10.1186/1471-2121-14-32, PMID 23834359.

  54. Chung. Apigenin induces caspase-dependent apoptosis in human lung cancer A549 cells through Bax-and Bcl-2-triggered mitochondrial pathway. Int J Oncol. 2010;36(6). doi: 10.3892/ijo_00000634.

  55. Jabea Ekabe C, Asaba Clinton N, Agyei EK, Kehbila J. Role of apoptosis in HIV pathogenesis. Adv Virol. 2022;2022:8148119. doi: 10.1155/2022/8148119, PMID 35462964.

  56. Luetragoon T, Pankla Sranujit RP, Noysang C, Thongsri Y, Potup P, Suphrom N. Anti-cancer effect of 3-hydroxy-β-ionone identified from Moringa oleifera lam. leaf on human squamous cell carcinoma 15 cell line. Molecules. 2020 Aug 1;25(16). doi: 10.3390/molecules25163563, PMID 32764438.