Investigating the Mechanisms of Prunella vulgaris against Acute Lymphoblastic Leukemia through Network Pharmacology and Experimental Validation

Abstract

Objective: This study investigates the underlying mechanisms of Prunella vulgaris on apoptosis in acute lymphoblastic leukemia (ALL) cells using an integrated approach that combines network pharmacology, molecular docking, and cellular experiments. Methods: Active ingredients of P. vulgaris and their potential targets were obtained from public databases, along with target genes associated with ALL. Intersecting targets were identified and analyzed through protein–protein interaction (PPI) network analysis to screen for hub genes. Gene Ontology (GO) and Kyoto Encyclopedia of Genes and Genomes (KEGG) pathway enrichment analyses are conducted on the common targets. A network linking P. vulgaris, its active components, targets, and KEGG pathways is constructed to identify key bioactive components. Molecular docking is performed to assess the binding affinities between the active compounds and core targets. Finally, in vitro experiments validate the predictions derived from network pharmacology. Results: PPI analysis identified five hub genes of P. vulgaris against ALL: CASP3, TNF, IL6, TP53, and AKT1. GO and KEGG analyses reveal that the mechanisms primarily involve the regulation of apoptotic signaling pathways, Bcl-2 family protein complexes, the PI3K/AKT and p53 pathway. Network analysis indicates that quercetin acts as a central active component of P. vulgaris against ALL. Molecular docking revealed potent binding affinities of quercetin toward all five core targets. Cellular experiments further confirm that quercetin inhibits the proliferation of ALL cells and induces mitochondrial pathway-mediated apoptosis, potentially through the modulation of the PI3K/AKT and p53 pathways. Conclusion: This study reveals that quercetin, a key active ingredient of P. vulgaris, mediates its anti-ALL activity by modulating the PI3K/AKT/p53 axis while inducing mitochondrial apoptosis. These findings provide reliable experimental evidence for utilizing P. vulgaris in the treatment of ALL.

1. Introduction

Acute lymphoblastic leukemia (ALL) is a malignant tumor of the hematopoietic system characterized by the aberrant proliferation and accumulation of B-cell or T-cell lymphoid progenitors. Children are the most severely affected group, with this disease accounting for over 70% of childhood leukemia cases. Currently, chemotherapy remains the mainstay of treatment for ALL. Although continued optimization of chemotherapeutic regimens has significantly improved the complete remission rate to 90% in pediatric patients, the overall survival rates for adolescents and adults remain suboptimal, ranging from 50% to 70%. Moreover, the five-year survival rate postrelapse decreases sharply to between 30% and 60%. Consequently, developing innovative therapeutic strategies is imperative to enhance clinical outcomes for ALL patients.

Traditional Chinese Medicine (TCM) provides unique advantages for oncology treatment, owing to the multitarget, low-toxicity, and holistic regulatory characteristics. , Prunella vulgaris, a commonly used herb in TCM for resolving masses and dissipating nodules, exhibits multiple pharmacological activities including antitumor, anti-inflammatory, and hypoglycemic effects. − Studies indicated that extracts of P. vulgaris inhibit proliferation and induce apoptosis in multiple tumor models, including lung cancer, liver cancer, thyroid cancer, and T-cell lymphoma. − A case report found that Sijunzi Decoction combined with P. vulgaris-based formulas could sustain long-term survival and alleviate suffering in patients with chemotherapy-intolerant diffuse large B-cell lymphoma. These studies indicate the potential therapeutic value of P. vulgaris in leukemia. However, its effects on ALL originating from B-lineage or T-lineage lymphoid progenitor cells and the underlying mechanisms remain unclear.

Network pharmacology is an emerging discipline that converges with systems biology and multiomics data. It enables the efficient identification of key active constituents and core mechanisms within complex TCM systems, thereby providing a targeted direction for subsequent precise experimental validation and thus avoiding indiscriminate screening. For instance, machine learning algorithms combined with network topological features have been successfully applied to identify key gene modules in complex diseases such as Tetralogy of Fallot. Owing to its systematic, high-throughput, and efficient nature, it has become a pivotal tool for deciphering the holistic mechanisms of Chinese herbs.

To investigate the anti-ALL mechanisms of P. vulgaris, we combined multiple methods. Identify its primary active components and potential targets using network pharmacology. Molecular docking then validated the interactions between these components and the core targets. Furthermore, cellular experiments were conducted to confirm the pharmacological effects and underlying mechanisms. This integrated strategy provides a crucial theoretical and experimental basis for translating P. vulgaris into clinical practice for ALL treatment.

2. Results

2.1. Identification of Common Targets between P. vulgaris and ALL

A systematic screening process identified 11 bioactive compounds from P. vulgaris. After integration and removal of duplicates, 158 potential drug targets were obtained. A total of 3720 and 280 ALL-associated targets were screened from the databases of GeneCards and OMIM. After merging and removing duplicates, 3842 disease-related targets were retained for subsequent analysis. The intersection between P. vulgaris targets and ALL targets yielded 110 common genes (Figure A).

1.

Identification of shared targets and the core network of P. vulgaris against ALL. (A) Venn diagram illustrating the overlapping targets between P. vulgaris (158 predicted targets) and ALL (3842 disease-associated targets from GeneCards and OMIM databases), yielding 110 common potential therapeutic targets. (B) PPI network of the 110 common targets. Nodes represent proteins, and edges represent functional associations. The size and color intensity of nodes are proportional to their degree centrality. (C) The top five core targets were screened from the PPI network based on median degree centrality.

2.2. Construction of the PPI Network

A PPI network was built with the 110 common targets using the STRING database (Figure B). The resulting network comprised 102 nodes and 326 edges, with nodes representing targets and edges indicating functional associations between proteins. The importance of each node, represented by its degree centrality, was visually reflected through node size and color intensity. Using median degree as a cutoff, five genesCASP3, TNF, IL6, TP53, and AKT1were screened as core targets (Table and Figure C).

1. Five Core Genes Associated with P. vulgaris in the Treatment of ALL.

name	betweenness	closeness	degree	eigenvector	LAC	network
CASP3	24.225	0.615	9	0.238	4.444	5.912
TNF	105.895	0.686	14	0.285	5.000	10.653
IL6	34.074	0.585	10	0.187	4.800	7.578
TP53	92.561	0.667	15	0.342	5.333	12.344
AKT1	54.575	0.686	13	0.317	5.385	10.029

2.3. GO and KEGG Analyses of Intersection Targets

Gene Ontology (GO) and Kyoto Encyclopedia of Genes and Genomes (KEGG) analyses of the 110 intersection targets were conducted using R (v4.4.1). A total of 2332 significantly enriched GO terms were identified, which included 2128 biological process (BP), 50 cellular component (CC), and 154 molecular function (MF) terms. The BP terms primarily involved the regulation of apoptotic signaling pathways, responses to bacterial-derived molecules, and cellular responses to oxygen levels. The CC terms included membrane microdomains, protein kinase complexes, and Bcl-2 family protein complexes. The MF terms predominantly involved protein kinase regulator activity and cytokine receptor binding. The top 10 terms are presented in Figure A.

2.

Enrichment analysis of the intersection targets between P. vulgaris and ALL. (A) The top 10 significantly enriched GO terms in the categories of BP, CC, and MF. The bar length positively correlated with the number of genes, and the color represents the −log10­(P-value). (B) Bubble plot of the top 30 enriched KEGG pathways. The size of the bubble corresponds to the count of genes, and the color represents the −log10­(P-value). (C,D) KEGG pathway maps highlighting the positions of the common targets within the (C) PI3K-AKT signaling pathway (hsa04151) and (D) p53 signaling pathway (hsa04115). The common targets are marked in red.

KEGG pathway analysis revealed 174 significantly enriched pathways (Figure B), including PI3K/AKT signaling pathway, Cytomegalovirus infection, Apoptosis, and Chronic myeloid leukemia. Particular attention was given to pathways related to apoptosis, hsa04151 (PI3K/AKT), and hsa04115 (p53) (Figure C,D).

2.4. Construction of the P. vulgaris–Active Compounds–Targets–Pathways Network

A multilevel network illustrating the interactions among P. vulgaris, its active compounds, targets, and pathways was constructed using Cytoscape (v3.10.0) (Figure A). The network contained 143 nodes and 691 edges. Topological analysis revealed that quercetin (degree = 95), luteolin (degree = 44), and kaempferol (degree = 38) exhibited the highest connectivity. While the most connected targets were AKT1, RELA, MAPK1, CASP3, TP53, BAX, CASP9, and TNF. These results imply that the PI3K/AKT/p53 axis may represent a key mechanism underlying the anti-ALL effects of P. vulgaris, with quercetin identified as a critical bioactive compound.

3.

Integrated network and molecular docking validation of P. vulgaris against ALL. (A) The “Herb–Compounds–Targets–Pathways” network was constructed using Cytoscape (v3.10.0). The network integrates P. vulgaris (Orange triangle), its active compounds (blue circles), the common targets (pink polygon), and the significantly enriched KEGG pathways (green circles). Edges represent interactions between different nodes. (B–F) Representative three-dimensional binding conformations from molecular docking simulations between quercetin (red stick model) and the active sites of the five core target proteins (yellow highlight): (B) AKT1, (C) TP53, (D) TNF, (E) IL6, and (F) CASP3. Yellow dashed lines indicate hydrogen bonds.

2.5. Molecular Docking Validation

Molecular docking results indicated that quercetin, luteolin, kaempferol, and β-sitosterol exhibited strong binding affinities to the core target proteins, including CASP3, TNF, IL6, TP53, and AKT1 (Table ). All computed binding affinities were lower than −5 kcal/mol, indicating robust ligand–receptor interactions. Representative binding conformations of quercetin with each target are illustrated in Figure . Based on these findings, quercetin was selected for subsequent studies to elucidate its role in modulating ALL cell proliferation and apoptotic pathways.

2. Binding Affinities between Bioactive Compounds and Core Target Proteins.

bioactive components	core target proteins	PDB ID	binding affinities (kcal/mol)
quercetin	AKT1	1H10	–6.1
luteolin	AKT1	1H10	–6.3
kaempferol	AKT1	1H10	–6.1
quercetin	TP53	5O1A	–7.8
luteolin	TP53	5O1A	–7.9
quercetin	CASP3	1GFW	–7.1
luteolin	CASP3	1GFW	–7.8
kaempferol	CASP3	1GFW	–7.3
beta-sitosterol	CASP3	1GFW	–7.4
quercetin	TNF	1A8M	–8.7
luteolin	TNF	1A8M	–8.6
kaempferol	TNF	1A8M	–8.6
quercetin	IL6	1ALU	–6.9
luteolin	IL6	1ALU	–7.1

2.6. Quercetin Inhibits the Proliferation Activity of CEM Cells

CCK-8 assay was conducted after exposure to various concentrations for 24, 48, and 72 h. A significant decrease in CEM cell viability was observed with increasing quercetin concentrations and longer incubation periods (Figure A,B). The IC₅₀ values were calculated to be 117.4 μM at 24 h, 25 μM at 48 h, and 15 μM at 72 h. Morphological examination revealed that quercetin-treated cells displayed characteristic apoptotic morphology, including cell shrinkage and irregular shape. In contrast, control cells maintained a round, translucent, and intact morphology (Figure C). Due to the comparable inhibitory effects observed at 48 and 72 h, subsequent experiments were carried out using 30, 40, and 50 μM quercetin with a 48 h treatment period.

4.

Effects of quercetin on proliferation and apoptosis in CEM cells. (A) Cell viability of CEM cells exposed to a range of concentrations of quercetin for 24, 48, and 72 h. (B) Inhibition rate of CEM cells under the same treatment conditions. (C) Morphological changes in CEM cells treated with quercetin for 48 h; black arrows indicate cell shrinkage, red arrows indicate irregular cells. (D) Representative flow cytometry plots of apoptosis detection by PE Annexin V staining after 48 h of quercetin treatment. PE represents early apoptosis, while 7-AAD represents late apoptosis. (E) Quantitative analysis of the apoptotic cell population shown in (D). (F) Caspase-9 activity in CEM cells treated with quercetin for 48 h. (G) Quantitative analysis of mitochondrial membrane potential (ΔΨm) loss, as assessed by JC-1 staining and flow cytometry after 48 h of quercetin treatment. (H) Representative flow cytometry plots of ΔΨm corresponding to (G). FITC fluorescence indicates JC-1 monomers, while PE indicates JC-1 aggregates. Data are presented as mean ± standard deviation (SD) of three independent experiments; *P < 0.05, **P < 0.01, ***P < 0.001, ****P < 0.0001 versus the control (0 μM quercetin) group (Dunnett’s test following one-way ANOVA).

2.7. Quercetin Induces Apoptosis in CEM Cells via the Mitochondrial Pathway

Quercetin treatment triggered a dose-dependent elevation in the proportion of apoptotic cells, as determined by flow cytometry (Figure D,E). Western blot analysis indicated that quercetin treatment markedly upregulated the expression of cleaved Caspase-3, cleaved Caspase-9, and Bax, while downregulating Bcl-2 compared with the control group (Figure A,B). To further validate the engagement of the mitochondrial apoptotic pathway, Caspase-9 activity and mitochondrial membrane potential (ΔΨm) were determined. Significantly increased caspase-9 activity was observed in Figure F, and a pronounced loss of mitochondrial membrane potential (Figure G,H) after quercetin treatment, confirming that quercetin triggers apoptosis in CEM cells primarily through the mitochondrial pathway.

5.

Quercetin regulates the PI3K/AKT/p53 pathway in CEM cells. (A) Representative Western blot images of Bcl-2, Bax, cleaved Caspase-3, and cleaved Caspase-9 expression in CEM cells exposed to the various concentrations of quercetin for 48 h. (B) Densitometric quantification of the blots shown in (A). (C) Representative Western blot images showing the expression of AKT, p-AKT, and p53 proteins in CEM cells after quercetin treatment for 48 h. (D) Densitometric quantification of the blots shown in (C). (E) Apoptosis rate analyzed by flow cytometry following treatment with quercetin (40 μM) and/or AKT activator SC79 (4 μM) for 48 h. (F) Quantitative analysis of the apoptosis data in (E). (G) Cell viability of CEM cells was assessed after exposure to the indicated agents for 48 h. (H) Representative Western blot images of AKT and p-AKT expression under the indicated treatment conditions for 48 h. (I) Densitometric quantification of the blots shown in (H). Data are expressed as mean ± SD of three independent experiments. *P < 0.05, **P < 0.01, ***P < 0.001, ****P < 0.0001 versus the control (0 μM quercetin) group (Dunnett’s test following one-way ANOVA); #P < 0.05, ###P < 0.001, ####P < 0.0001 for comparisons among all groups (Tukey’s test following one-way ANOVA).

2.8. Quercetin Inhibits CEM Cell Proliferation by Regulating the PI3K/AKT/p53 Pathway

Western blot analysis indicated that, relative to the control group, although total AKT protein expression remained constant in quercetin-treated CEM cells, a marked downregulation was observed in both phosphorylated AKT (p-AKT) protein expression and the p-AKT/AKT ratio. At the same time, a marked upregulation in p53 protein expression was observed (Figure C,D). These results indicate that quercetin may inhibit the PI3K/AKT axis by suppressing AKT phosphorylation without altering total AKT levels, while simultaneously activating the p53 pathway, ultimately leading to the inhibition of cell proliferationconsistent with KEGG pathway predictions.

Additionally, the SC79 rescue experiment results indicate that compared to quercetin treatment alone, cotreatment with SC79 significantly restored the p-AKT/AKT ratio (Figure H,I), alleviated the suppression of cell proliferation (Figure G), and partially reversed quercetin-induced apoptosis (Figure E,F). These results further support the conclusion that quercetin exerts its antileukemic effects primarily by suppressing the PI3K/AKT axis.

2.9. Quercetin Induces Apoptosis in NALM-6 Cells

To investigate whether quercetin exerts similar pro-apoptotic effects in other ALL subtypes, we employed the B-ALL cell line NALM-6. Following 48 h of treatment, Cell viability and apoptosis rate were determined using the CCK-8 assay and flow cytometry, and a rescue experiment was performed using the AKT activator SC79. Figure A illustrates the concentration-dependent reduction in NALM-6 cell viability induced by quercetin, with a calculated IC₅₀ of 20 μM. Based on this result, concentrations of 10, 20, and 30 μM were selected for subsequent apoptosis analysis. Flow cytometry revealed that quercetin treatment increased the apoptotic rate in a dose-dependent manner (Figure B). Notably, cotreatment with SC79 partially reversed quercetin-induced apoptosis in NALM-6 cells (Figure E), suggesting that the PI3K/AKT axis is also involved in quercetin-mediated pro-apoptotic effects in this B-ALL model.

6.

Quercetin inhibits proliferation and induces apoptosis in NALM-6 cells. (A) Cell viability of NALM-6 cells exposed to a range of concentrations of quercetin for 48 h. (B) Quantitative analysis of the apoptotic cell population shown in (C). (C) Representative flow cytometry plots of apoptosis detection of quercetin treatment after 48 h. (D) Apoptosis detection following treatment with quercetin (30 μM) and/or AKT activator SC79 (4 μM) for 48 h. (E) Quantitative analysis of the apoptosis data in (D). Data are expressed as mean ± SD of three independent experiments. **P < 0.01, ***P < 0.001, ****P < 0.0001 versus the control (0 μM quercetin) group (Dunnett’s test following one-way ANOVA); ##P < 0.01, ####P < 0.0001 for comparisons among all groups (Tukey’s test following one-way ANOVA).

3. Discussion

ALL represents one of the most common childhood malignancies. Although advances in chemotherapy and immunotherapy have improved clinical outcomes, , significant challenges remain, including drug resistance, neurotoxicity, immune evasion, and high treatment costs. , As a complementary and alternative modality in oncology, TCM has garnered growing recognition in recent years. Although previous studies have demonstrated the antitumor effects of P. vulgaris in solid tumors and lymphomas, , this study represents the first integrated application of prediction and verification to systematically elucidate the anti-ALL mechanisms across both T-ALL (CEM) and B-ALL (NALM-6) cellular models.

Through network pharmacology analysis, 11 active components of P. vulgaris were identified, along with 110 overlapping targets significantly enriched in apoptosis-related pathways. PPI network analysis further revealed five core genesCASP3, TNF, IL6, TP53, and AKT1, a multitarget identification strategy consistent with recent advances in integrative omics approaches. Among these, AKT1 exhibited the highest degree of centrality, suggesting a central role in the anti-ALL network of P. vulgaris. Based on GO and KEGG enrichment analyses, particular focus was placed on the PI3K/AKT/p53 signaling pathways. The P. vulgaris–Active Compounds–Targets–Pathways network indicated that quercetin possesses a degree value of 95, substantially higher than other components, suggesting its pivotal role in combating ALL. Molecular docking revealed that quercetin exhibits strong affinity (binding energy ≤ −6.1 kcal/mol) toward all key targets, indicating stable interactions within their active sites and providing a structural basis for its multitarget regulatory mechanism. Previous studies have reported that quercetin induces mitochondrial superoxide production, activates the ERK pathway and Caspase family, and triggers mitochondrial apoptosis in HL-60 cells. Another study demonstrated that Quercetin inhibits the PI3K/AKT axis by regulating the expression levels of miR-224-3p and PTEN, ultimately accelerating apoptosis in acute myeloid leukemia (AML) cells. Thus, while quercetin’s role in inducing mitochondrial apoptosis, autophagy-mediated cell death, and inhibiting the PI3K/AKT axis is well established in AML, its specific mechanisms in ALL remain unclear. Therefore, based on its highest connectivity in the network, strong binding affinity in molecular docking, and documented multifaceted bioactivity in other hematological malignancies, quercetin was selected as the primary candidate compound for subsequent experimental validation.

Quercetin, a major flavonoid constituent of P. vulgaris, has been reported to significantly inhibit CEM cell viability and induce apoptosis time- and concentration-dependently. The endogenous apoptosis, also known as the mitochondrial pathway apoptosis, triggers the oligomerization of Bax/Bak upon activation. This promotes mitochondrial outer membrane permeabilization (MOMP) and the subsequent release of cytochrome c into the cytosol. There, cytochrome c associates with activated Caspase-9 to form the apoptosome, a complex that ultimately executes apoptosis through the cleavage of Caspase-3. Western blot results showed that quercetin treatment upregulates the cleaved Caspase-3, cleaved Caspase-9, and Bax, and downregulates Bcl-2 levels in CEM cells. Concurrently, a loss in mitochondrial membrane potential and an increase in Caspase-9 activity were observed, indicating that the pro-apoptotic effects are mediated through the mitochondrial pathway. Previous studies have found that quercetin, an active component of P. vulgaris, accelerates apoptosis in breast cancer cells by increasing the protein level of cleaved Caspase-3. Another study also reported that P. vulgaris accelerates apoptosis in thyroid carcinoma cells by elevating the mRNA levels of Bcl-2, Bax, and Caspase-3. These findings are highly consistent with our observations in ALL cells.

To further elucidate the mechanism of quercetin, the PI3K/AKT and p53 pathways identified through network pharmacology were experimentally validated. As a widely conserved signaling network in eukaryotic cells, the PI3K/AKT pathway plays a crucial role in promoting survival and modulating proliferation and the cell cycle. Studies have indicated that hyperactivation of this pathway highly correlates with the development and poor prognosis of ALL. , Furthermore, this pathway modulates mitochondrial apoptosis by regulating Bcl-2 and Caspase family proteins, , which is consistent with our earlier findings. As a well-established tumor suppressor, p53 facilitates Bax activation and mitochondrial membrane permeabilization. Western blot analysis of pathway-related proteins revealed that quercetin inhibits AKT phosphorylation and enhances p53 protein expression in CEM cells. Additionally, in CEM and NALM-6 cells, a rescue experiment using SC79, a specific AKT agonist, was conducted to further validate the critical role of the PI3K/AKT axis. , The results showed that SC79 partially reversed quercetin-induced proliferation suppression and apoptosis, and significantly upregulated p-AKT expression. These results comprehensively demonstrate that quercetin mediates its anti-ALL activity by regulating the PI3K/AKT/p53 axis. These findings are consistent with previously reported mechanisms of action for P. vulgaris in treating breast cancer and liver cancer. ,

It is noteworthy that quercetin’s dual effects of inhibiting AKT phosphorylation and elevating p53 protein levels suggest its mechanism may involve cross-regulation among multiple pathways. Previous studies indicate that AKT can directly phosphorylate and activate MDM2, a key E3 ubiquitin ligase responsible for p53 degradation. The inhibition of AKT activation by quercetin may reduce MDM2 activity, thereby preventing proteasomal degradation of p53 and leading to its accumulationa finding consistent with our experimental results. Accumulated p53 protein can further enhance pro-apoptotic effects by transcriptionally activating downstream target genes such as Bax, consistent with our observed increase in Bax expression and subsequent decrease in the Bcl-2/Bax ratio. Furthermore, activated p53 may enhance quercetin’s antileukemic efficacy by transcriptionally upregulating PTEN (a key upstream negative regulator of the PI3K/AKT pathway), thereby further inhibiting AKT signaling. In summary, quercetin induces synergistic pro-apoptotic effects in ALL by simultaneously targeting the oncogenic PI3K/AKT pathway and the tumor-suppressor p53 pathway, thereby disrupting the vicious cycle of cancer cell proliferation. This dual-targeting strategy reveals the complex mechanism underlying the multitargeted properties of TCM compounds. Combining TCM with Western pharmaceuticals may overcome resistance associated with single-target chemotherapy drugs and provide complementary therapeutic advantages.

Although this study established a complete chain of evidence for the anti-ALL activity of P. vulgaris and its key component quercetin, several limitations should be acknowledged. First, regarding experimental models, our validation was primarily conducted in the T-ALL cell line CEM, with consistent antiproliferative and pro-apoptotic effects further confirmed in the B-ALL cell line NALM-6. However, the study remains confined to in vitro systems. The absence of in vivo data precludes evaluation of key pharmacological parameters such as bioavailability, systemic toxicity, or the influence of the tumor microenvironment on the efficacy of quercetin or P. vulgaris extracts. Future studies should further protein-level validation in additional models and employ patient-derived xenograft (PDX) models or immunocompetent mouse models of ALL to assess the pharmacokinetics and systemic toxicity of quercetin in a more physiologically relevant setting. Such models would also allow investigation of their potential interactions with the host immune system and the leukemic bone marrow microenvironment. Second, in terms of active component validation, this study focused on quercetin as the core bioactive constituent identified through network pharmacology. Although this targeted approach is efficient, we did not perform direct component isolation or comparative activity assays, which limits definitive confirmation that quercetin is the principal bioactive constituent of P. vulgaris against ALL. Moreover, the potential synergistic or additive effects of other bioactive components in P. vulgaris, such as luteolin and kaempferol (which also showed high network connectivity and favorable docking scores), remain unexplored. The holistic therapeutic effect of the whole herb may arise from multicomponent synergy. Future work should include comprehensive component analysis and combination studies to determine whether quercetin acts as the primary effector or as part of a broader synergistic network. Third, on the mechanistic level, our conclusions rely largely on functional assays (viability, apoptosis) and pathway modulation inferred from protein expression changes. Although molecular docking predicted strong binding of quercetin to key targets (e.g., AKT1, TP53) and functional experiments implicated the PI3K/AKT/p53 axis, direct experimental evidence for physical interactions between quercetin and these targets, such as via cellular thermal shift assays or surface plasmon resonance, is still lacking. Further studies are needed to confirm these direct interactions and to elucidate the precise molecular events, such as the potential cross-talk between the p53 and PI3K/AKT pathways.

4. Conclusion

This study utilized a combined network pharmacology and experimental validation approach to demonstrate that quercetin is a key active component in P. vulgaris contributing to its anti-ALL activity. Quercetin induced mitochondrial-mediated apoptosis of ALL cells by inhibiting the PI3K/AKT pathway and activating the p53 signaling pathway. These findings offer a solid rationale for the development of novel therapeutic strategies targeting ALL.

5. Materials and Methods

5.1. Network Pharmacology

5.1.1. Screening of Active Ingredients and Targets of P. vulgaris

A systematic retrieval of P. vulgaris constituents was conducted using the TCMSP database (https://www.91tcmsp.com) with the keyword “xiakucao”. Selection of active compounds was performed according to the previous standards. Corresponding target proteins were identified, and their gene nomenclature was unified using the UniProt database.

5.1.2. Acquisition of ALL-Related Targets

ALL-associated target genes were collected from the Genecards (https://www.genecards.org) and OMIM (https://www.omim.org) databases. The collected targets were integrated and deduplicated using R software (v4.4.1) to establish a unified set of ALL-associated targets.

5.1.3. Construction and Analysis of PPI Network

Using the jvenn online tool (https://www.bioinformatics.com.cn/jvenn) to generate a Venn diagram to discern shared targets between P. vulgaris and ALL. These shared targets were then analyzed on the STRING platform (https://string-db.org), opting to hide disconnected nodes. After exporting the PPI data in TSV format. We employed Cytoscape (v3.10.0) for network visualization and discerned hub genes using the CytoNCA plugin.

5.1.4. GO and KEGG Enrichment Analysis

Gene IDs of the common targets were obtained. GO and KEGG enrichment analyses were performed by R (v4.4.1). The significance thresholds were set at P < 0.05. The top 10 significantly enriched terms from each GO category were retained for further analysis. For KEGG, the top 30 pathways were retained and visualized.

5.1.5. Construction of the “P. vulgaris–Compounds–Targets–Pathways” Network

An integrated multilevel network was built by incorporating the bioactive compounds of P. vulgaris, common targets, and significantly enriched KEGG pathways. We employed Cytoscape (v3.10.0) for network visualization. The topological parameter “degree” was computed to assess node importance, with a higher degree value indicating greater biological relevance. The final network visualization was exported for further interpretation.

5.1.6. Molecular Docking Validation

The two-dimensional (2D) structures of the bioactive compounds (ligands) were retrieved from the PubChem database (https://pubchem.ncbi.nlm.nih.gov). Energy minimization of each ligand structure was performed using the MM2 force field in ChemBio3D Ultra 14.0, and the optimized structures were output in.mol2 format. The three-dimensional crystal structures of the five core targetsAKT1, TP53, CASP3, TNF, and IL6were obtained from the RCSB Protein Data Bank (https://www.rcsb.org). Before docking, receptor structures were preprocessed using PyMOL (v2.4.0) to remove crystallographic water molecules and any original ligands by executing the commands remove solvent and remove organic. Both receptors and ligands were then saved in.pdbqt format using AutoDockTools 1.5.7 for docking preparation.

The binding site was defined by centering a grid box on the coordinates of the native ligand’s binding pocket. The grid spacing was set to 1.0 Å, and the box dimensions were set to 40 × 40 × 40 points. These parameters, including grid box center coordinates, were saved in a grid parameter file (.gpf). Docking simulations were run in AutoDock Vina (v1.2.x) with an exhaustiveness value of 8. For each ligand–target complex, the pose with the most favorable binding affinity (lowest binding energy in kcal/mol) was selected as the best-docked conformation. The resulting complexes were visualized and analyzed for key interactions (e.g., hydrogen bonds) using PyMOL (v2.4.0).

5.2. In Vitro Experiments

5.2.1. Reagents and Antibodies

Quercetin was acquired from Chengdu Herbpurify Co., Ltd. (Chengdu, China). T-ALL cell line CEM and B-ALL cell line NALM-6 were purchased from Wuhan Procell Life Science and Technology Co., Ltd. (Wuhan, China). The CCK-8 assay kit was obtained from Beijing Livning Biotech Co., Ltd. (Beijing, China). The PE Annexin V apoptosis kit was acquired from BD Bioscience. (Shanghai, China). RIPA lysis buffer, protease and phosphatase inhibitor mixture, 5× SDS-PAGE sample loading buffer, BCA protein quantification kit, and Caspase-9 activity assay kit were acquired from Biotime Biotechnology Research Institute. (Shanghai, China). The mitochondrial membrane potential detection kit was provided by Beijing Solabio Life Sciences. (Beijing, China).

The primary antibodies for Bax (Rabbit monoclonal, Cat#5023, 1:2000 dilution), Bcl-2 (Rabbit monoclonal, Cat#4223, 1:2000 dilution), p53 (Rabbit monoclonal, Cat#2527, 1:2000 dilution), cleaved Caspase-3 (Rabbit monoclonal, Cat#9664, 1:1000 dilution), and p-AKT (Rabbit monoclonal, Cat#4060, 1:1000 dilution) were provided by Cell Signaling Technology. (Danvers, MA). The Caspase-9 antibody (Rabbit polyclonal, Cat#WL03421, 1:1000 dilution) was obtained from Shenyang Wanlei Biotechnology Co., Ltd. (Shenyang, China); the AKT antibody (Rabbit polyclonal, Cat#10176-2-AP, 1:5000 dilution) was sourced from Wuhan Sanying Biotechnology Co., Ltd. (Wuhan, China); GAPDH (Rabbit monoclonal, Cat#AF1186, 1:3000 dilution) was purchased from Biotime Biotechnology Research Institute. (Shanghai, China); β-actin (Mouse monoclonal, Cat# K200058M, 1:3000 dilution) was provided by Beijing Solabio Life Sciences. (Beijing, China).

5.2.2. Cell Culture

CEM cells and NALM-6 cells were centrifuged at 1000 rpm for 3 min, resuspended in specialized culture medium, and cultured at 37 °C with 5% CO₂.

5.2.3. Cell Viability Assay

Cells (CEM or NALM-6) were seeded into 96-well plates at a density of 5 × 10⁵ cells/mL and treated with quercetin at final concentrations of 0, 5, 10, 20, 40, and 80 μM. Triplicate plates were prepared, and viability was measured after 24, 48, and 72 h of incubation. At the indicated time points, after adding CCK-8 reagent to all wells, the plates were incubated for 3 h, and then the absorbance at 450 nm was recorded. Cell viability was calculated according to the following formula

$$\mathrm{v}\mathrm{i}\mathrm{a}\mathrm{b}\mathrm{i}\mathrm{l}\mathrm{i}\mathrm{t}\mathrm{y}(\%)=\left(\frac{{\mathrm{O}\mathrm{D}}_{\mathrm{t}\mathrm{r}\mathrm{e}\mathrm{a}\mathrm{t}}-{\mathrm{O}\mathrm{D}}_{\mathrm{b}\mathrm{l}\mathrm{a}\mathrm{n}\mathrm{k}}}{{\mathrm{O}\mathrm{D}}_{\mathrm{c}\mathrm{o}\mathrm{n}\mathrm{t}\mathrm{r}\mathrm{o}\mathrm{l}}-{\mathrm{O}\mathrm{D}}_{\mathrm{b}\mathrm{l}\mathrm{a}\mathrm{n}\mathrm{k}}}\right)\times 100\%$$

GraphPad Prism 10.3.0 was employed to determine the half-maximal inhibitory concentration (IC₅₀). Based on the IC₅₀ value obtained at 48 h, concentrations of 30, 40, and 50 μM were selected for subsequent experiments. Triplicate wells were set up for each condition within a single experiment. This entire assay was independently repeated three times (n = 3) to ensure reproducibility.

5.2.4. Apoptosis Assay

The experimental groups included a negative control group and quercetin-treated groups at concentrations of 30, 40, and 50 μM in CEM cells and 10, 20, and 30 μM in NALM-6 cells. Cells were plated in 12-well plates at 5 × 10⁵ cells/mL and exposed to respective concentrations of quercetin. After 48 h, the cells were collected, and each tube received the following: 100 μL of 1× binding buffer, along with 5 μL each of 7-AAD and Annexin V-PE. The samples were gently vortexed and incubated for 20 min in the dark. Subsequently, an additional 400 μL of 1× binding buffer was added to all tubes before analysis on a flow cytometer. Apoptotic cells were quantified using FlowJo v10.8.1. The assay was independently repeated three times (n = 3).

5.2.5. Mitochondrial Membrane Potential Assay

Cells were treated according to the procedure outlined in Section . After 48 h, cells were collected by centrifugation, adjusting the total cell count to 1 × 10⁵ to 6 × 10⁵. And then 0.5 mL of JC-1 working solution was added, the tube was inverted to mix before incubation for 20 min at 37 °C. Following centrifugation, the pellet was washed twice with 1× JC-1 staining buffer. Finally, the cells were resuspended and assessed for changes in mitochondrial membrane potential using a flow cytometer. The assay was independently repeated three times (n = 3).

5.2.6. Caspase-9 Activity Assay

Cells were treated according to the procedure outlined in Section . After 48 h, cells were harvested and resuspended in lysis buffer, then subjected to ice-cold lysis for 15 min. The resulting lysate was then spun at 16,000 rpm for 10 min at 4 °C to isolate the supernatant. A reaction system was prepared according to the instructions, and absorbance was measured at 405 nm. Caspase-9 activity units were calculated based on a standard curve, and enzyme activity was expressed as units per μg of protein. The assay was independently repeated three times (n = 3).

5.2.7. Western Blot Analysis

Cells were treated according to the procedure outlined in Section . Referring to the previous study, proteins were quantified using a BCA kit. Proteins were separated by SDS-PAGE with 25 μg loaded per lane and transferred to a PVDF membrane via wet transfer methods. After the membrane was blocked, incubated with primary and secondary antibodies, and washed with PBST, blots were developed using an ECL substrate. Band intensities were determined with ImageJ software, with three independent replicates performed for each experiment.

5.3. Statistical Analysis

GraphPad Prism 10.3.0 was employed for all statistical analyses. Results are presented as the mean ± SD from three independent experiments. Differences between multiple groups and a single control were determined by one-way analysis of variance (ANOVA) with Dunnett’s post hoc test. Comparisons across all groups were assessed by one-way ANOVA with Tukey’s post hoc test. P < 0.05 is defined as statistical significance.

The Supporting Information is available free of charge at https://pubs.acs.org/doi/10.1021/acsomega.5c09488.

• Raw data of Western blot (PDF)

• Bioactive components of P. vulgaris, targets of P. vulgaris, deduplicated ALL Targets, common targets between ALL and P. vulgaris, and detailed Information of P. vulgaris–Compounds–Targets–Pathways” Network (XLSX)

The authors declare no competing financial interest.