Network pharmacology and experimental analysis reveal Ethyl caffeate delays osimertinib resistance in lung cancer by suppression of MET

Abstract

Background

Fei Yanning Formula (FYN) is extensively applied in clinical lung cancer treatment. However, the specific active constituents and targets of its therapeutic effects remain unclear.

Objective

The study aims to explore the active constituents and mechanism of FYN in delaying osimertinib resistance by network pharmacology analysis and experimental verification.

Methods

We collected the chemical constituents of the FYN based on the TCMSP database and relevant literature sources. Osimertinib resistance-related targets were acquired from the GeneCards database. We systematically construct the PPI network and KEGG analysis to explore hub targets and key pathways. The main active components of FYN were identified by molecular docking. Subsequently, we conducted in vitro experiments to verify its effect on osimertinib-resistant cells in lung cancer.

Results

The PPI network and KEGG pathways analysis revealed six key targets linked to PI3K-AKT signaling pathways (ERBB2, EGFR, MET, HSP90AA1, MCL1, and IGF1R). RT-qPCR and immunohistochemical analyses demonstrated that FYN could suppress the expression of ERBB2, MET and HSP90AA1. Molecular docking indicated that Ethyl caffeate, the primary component in FYN, had a stronger binding ability with MET. Experiments illustrated that Ethyl caffeate inhibited the migration and proliferation of osimertinib-resistant cells, promoted apoptosis, and suppressed the expression level of MET.

Conclusion

FYN might delay osimertinib resistance by downregulating the expression of MET, which can be attributed to its active ingredient, Ethyl caffeate.

Supplementary Information

The online version contains supplementary material available at 10.1186/s12935-025-03690-5.

Background

Lung cancer is one of the most prevalent malignant tumors globally [1]. Non-small cell lung cancer (NSCLC) stands as the most common type of lung cancer, frequently managed through a combination of surgical intervention, chemotherapy, radiotherapy, targeted therapy, and immunotherapy [2]. There are as many as two dozen targeted drugs having received approval from the Food and Drug Administration, and targeted therapy has notably enhanced patient prognosis [3]. Osimertinib, the third-generation epidermal growth factor receptor tyrosine kinase inhibitor (EGFR-TKI), is extensively employed in the first-line setting. Nonetheless, resistance to osimertinib inevitably emerges [4–5]. Notably, the use of osimertinib may lead to adverse reactions, including the development of rash and diarrhea [6–7]. Hence, there is an imperative need to identify novel strategies aimed at delaying resistance to osimertinib.

Many traditional Chinese medicine (TCM), whether monomer or compound, can overcome multidrug resistance [8]. Feiyanning Formula (FYN) is an anti-cancer formula based on decades of clinical trials [9] and has been demonstrated to effectively inhibit the invasion and metastasis of advanced lung cancer [10]. Previous research findings that FYN could delay the osimertinib resistance by modulating the Wnt/β-catenin pathway [11]. However, the key active components and main targets of FYN are still unclear. Ethyl caffeate, a naturally occurring phenolic substance extracted from numerous plant sources, possesses both anti-inflammatory and anti-neoplastic properties [12–13]. Research has indicated that Ethyl caffeate is capable of suppressing the migration and invasion of ovarian cancer cells [14]. Despite advancements in elucidating the impact of Ethyl caffeate on inflammation and oncogenesis, its specific role and the underlying molecular pathways in the genesis and progression of lung cancer, as well as its influence on drug resistance in lung cancer, are yet to be fully understood.

Utilizing the multi-target effects and systemic regulatory properties of TCM, we employed network pharmacology to explore the specific mechanism by which active ingredients of FYN delays the osimertinib resistance. In this work, we performed the initial prediction of the correlation between active ingredients of FYN and the binding targets associated with osimertinib resistance, along with their corresponding pathways. Furthermore, through in vitro and in vivo experiments, we substantiated the mechanism underlying the delayed osimertinib resistance by FYN and its primary active components.

Methods

Identification of potential targets of FYN and osimertinib resistance

The chemical components of FYN were identified by the TCMSP database (https://old.tcmsp-e.com/tcmsp.php) with criteria set for oral bioavailability (OB ≥ 30%) and drug likeness (DL ≥ 0.18). Relevant literature was searched for supplement missing information. The PubChem database was then utilized to retrieve the corresponding SMILE numbers for these chemical components, and Swiss Target Prediction was employed to acquire the targets. A search term “lung cancer Osimertinib resistance” was employed to retrieve key targets of lung cancer resistance to osimertinib from the GeneCards database. The Venn diagram between the targets of FYN components and disease was generated to pinpoint common targets.

PPI network analysis

We input the shared targets into the STRING database (https://cn.string-db.org/) [15]. and then establish a protein-protein interaction (PPI) network via Cytoscape 3.7.2. The identification and screening of essential targets were carried out through the analysis of network topology parameters.

KEGG enrichment analysis

To elucidate how FYN affect on signaling pathways for delaying osimertinib resistance, the shared targets were entered into Metascape (https://metascape.org/) [16] database. The GO and KEGG enrichment analyses were visualized by the the “clusterProfiler” R package.

Molecular docking analysis

We selected the key targets and the active ingredients to perform molecular docking. Subsequently, we retrieved the target proteins along with their corresponding 3D structure files from the Protein Data Bank (PDB) databases (www.rcsb.org) [17]. Discovery Studio 2016 (3dsbiovia.com/) was employed for docking and calculating binding energy, while Pymol was utilized for visualization and drawing purposes.

Cells culture

PC9 cells and HCC827 cells, initially procured from the Cell Bank of the Chinese Academy of Sciences (Shanghai, China), were cultivated in RPMI-1640 medium (BasalMedia Technologies Co., Ltd., China, L210KJ) enriched with 10% fetal bovine serum (FBS) (Lonsera, China, S711-001 S) and 1% penicillin-streptomycin (Biosharp, China, BL505A) at 37˚C in a humidified incubator with 5% CO₂ (Haier, China). The FBS used was not heat inactivated. Subsequently, osimertinib-resistant PC9 cells (PC9OR) and osimertinib-resistant HCC827 cells (HCC827OR) were developed by gradually increasing the osimertinib concentration from 5 to 3,000 nM for 5 months, as previously described [12]. PC9OR and HCC827OR cells were treated with osimertinib at a concentration of 3,000 nM to sustain the drug resistance.

CCK-8 assay

PC9OR and HCC827OR cells were seed into 96-well plate with 3,000 cells per well and exposed to various concentrations of FYN (0, 62.5, 125, 250, 500, 1000 µg/ml) and Ethyl caffeate (0,16.25, 32.5, 75, 150, 300 µM) for 48 h. After introducing 10 µL CCK-8 reagent (Beyotime Inc., China, C0039) into each well for 2 h, we determined the absorbance at 450 nm using a microplate reader and definitively computed the IC₅₀ value.

Real-time qPCR assay

Total RNA was collected from PC9OR and HCC827OR cells by the RNA Purification Kit (EZB, USA, B0004DP). Subsequently, the reverse transcription kit (Takara, Japan, RR037A) was employed. The mRNA levels were evaluated utilizing the 2^–ΔΔCt calculation method, and then normalized by GAPDH mRNA expression. Primers were purchased from Sangon Biotech and primer sets are listed in Table 1.

Table 1.

The primer sequences of MET, ERBB2,HSP90AA1

Primer name	Primer sequences(5’to3’)
MET Forward	CTGAAGCCGTTTTATGCAC
MET Reverse	CAACACCAGCAATCAATCC
ERBB2 Forward	GCTCCACACTGCCAACC
ERBB2 Reverse	TCCTCCACGCACTCCTG
HSP90AA1 Forward	CGGATGCCTAAGTAGACCA
HSP90AA1 Reverse	CAAGCCCTGTGGAGAGAT

The xenograft tumor model establishment

Five-week-old female nude mice (Shanghai SLAC Animal Laboratory) were selected. Subsequently, HCC827OR cells (1 × 10⁷) were injected subcutaneously under the left armpit. Upon reaching an average tumor volume of approximately 100 mm³ in all mice, they were randomized into control group and FYN group. The FYN group received a daily oral dose of 131.86 mg FYN powder per mouse for 14 days. In contrast, the control group was given an equivalent volumes of water.

Immunohistochemistry and TUNEL assay

Tumor specimens were immersed in a 4% paraformaldehyde solution (Beyotime Inc., China, P0099). Subsequently, the specimens were embedded in paraffin, sectioned, and subjected to immunohistochemical analysis. The observations were made using light microscopy. Tissue sections embedded in paraffin were subjected to staining using the TUNEL Detection Kit (Roche, China, 11684817910) following the prescribed protocol from the manufacturer.

Invasion assay

4 × 10⁴ PC9OR and HCC827OR cells were cultivated in the upper chamber of 24-well plates using serum-free 1640 medium, while 500 µl complete medium was filled with in the lower chamber. Ethyl caffeate was administered for a duration of 24 h. Subsequently, the chambers were fixed with PFA and stained with 0.1% crystal violet. After washing with PBS, the chambers were allowed to dry and photograph.

EDU assay

Approximately 3,000 cells were cultivated in 96-well plates. Ethyl caffeate was administered for 48 h. After that, the cells were labeled with EDU detection kit (Beyotime Inc., China, C0071S) for 2 h, and fixed, stained and observed using fluorescence microscopy.

Flow cytometry analysis

PC9OR and HCC827OR cells were cultivated in Ethyl caffeate for 48 h at the IC₅₀ concentration before being harvested. Apoptosis detection was carried out utilizing the Annexin V-FITC/PI apoptosis detection kit (BD, USA, 556420) in accordance with the instructions. The Cell cycle and apoptosis analysis kit (MULTI SCIENCES, China, AP101) was employed following the manufacturer’s protocol for cell cycle analysis,. After the addition of DNA staining solution and permeabilization solution, the treated cells were detected by the flow cytometer (BD FACS Lyric, USA). FlowJo VX was used for processing and interpreting the results.

Western blot assay

PC9OR and HCC827OR cells were exposed to Ethyl caffeate for 48 h and then lysed in ice-cold RIPA lysis buffer (Beyotime Inc., China, P0013B) supplemented with Protease and phosphatase inhibitor cocktail (Beyotime Inc., China, P1045). The protein concentrations were determined through the BCA Protein Assay Kit (Epizyme Biotech, China, ZJ101). The proteins were separated via SDS-PAGE and subsequently transferred onto PVDF membranes. After blocking with 5% milk for 2 h, the membranes were incubated overnight at 4 °C with primary antibodies: MET (1:1000, cat. no. 8198, Cell Signaling Technology, USA), PI3K (1:1000, 60225-1-Ig, Proteintech Group, China), p-PI3K (1:1000, AF3242, Affinity Biosciences, China), AKT (1:1000, 10176-2-AP, Proteintech Group, China), p-AKT (1:1000, 66444-1-Ig, Proteintech Group, China), and GAPDH (1:1000, cat. no. 5174, Cell Signaling Technology, USA). The horseradish Peroxidase secondary antibodies were applied and detected using ECL solution (New Cell Molecular Biotech, China, P10200).

Statistical analysis

All experiments were replicated 3 times. Differences between two groups were analyzed by Student’s t-test and visualized via GraphPad Prism 8.0. P < 0.05 was statistically significant.

Results

Active targets of FYN delaying osimertinib resistance

FYN is composed of Chinese herbs, which include Astragalus membranaceous (Huang Qi), Polygonatum kingianum (Huang Jing), Cornus officinalis (Shan Zhu Yu), Paris polyphylla (Chong Lou), Atractylodes macrocephala (Bai Zhu), Polistes olivaceous (Feng Fang), Salvia chinensis (Shi Jian Chuan), Bufo gargarizans Cantor (Gan Chan Pi), Ganoderma lucidum(Ling Zhi), Cremastra appendiculata (Shan Ci Gu) and Epimedium brevicomu Maxim (Yin Yang Huo). All the chemical components of 11 herbs were selected based on DL ≥ 0.18 and OD ≥ 30%. After reviewing the literature, the unpredicted active compounds in TCMSP were supplemented. The corresponding chemical structures were identified using their SMILES codes in the PubChem database, and their potential targets were identified through Swiss Target Prediction. Ultimately, a total of 146 active ingredients were identified, including 8 active ingredients in Bai Zhu, 6 active ingredients in Feng Fang, 10 active ingredients in Gan Chan Pi, 13 in Huang Jing, 13 in Huang Qi, 42 in Ling Zhi, 7 in Shan Ci Gu, 7 in Shi Jian Chuan, 16 in Yin Yang Huo, 11 in Chong Lou, 13 in Shan Zhu Yu (Supplementary Table S1). Concurrently, we retrieved 904 targets of FYN and 131 targets associated with osimertinib resistance. The Venn analysis diagrams illustrated 22 overlapping targets between FYN and osimertinib resistance, indicating potential treatment targets (Fig. 1A).

Fig. 1.

Potential therapeutic targets of FYN delaying osimertinib resistance and PPI network construction. (A) Venn diagram of FYN and osimertinib resistance intersection targets. (B) PPI network of the common targets

PPI network construction

The 22 overlapping targets were entered into the STRING database and then reconstructed using Cytoscape 3.7.2 software to generate the PPI network (Fig. 1B). Notably, the network exhibited highly connected nodes, and 11 proteins (EGFR, STAT3, ERBB2, HSP90AA1, MET, MCL1, ABCB1, BRAF, ABCG2, IGF1R, and PTPN11) were identified based on the degree greater than or equal to the median (10) (Table 2). Consequently, these targets were validated as essential targets of FYN for delaying osimertinib resistance.

Table 2.

The potential hub targets ranked top 11

S. No.	Target	Degree	Betweenness centrality	Closeness centrality
1	EGFR	17	0.1835552	0.86956522
2	STAT3	16	0.15487549	0.83333333
3	ERBB2	15	0.05170635	0.8
4	HSP90AA1	15	0.10808383	0.8
5	MET	12	0.02909051	0.71428571
6	MCL1	12	0.05803547	0.71428571
7	ABCB1	11	0.02241341	0.68965517
8	BRAF	11	0.01463033	0.68965517
9	ABCG2	11	0.02241341	0.68965517
10	IGF1R	10	0.00419173	0.66666667
11	PTPN11	10	0.01042398	0.66666667

GO and KEGG analysis

To evaluate the potential impact of FYN on osimertinib resistance-related biological processes and signaling pathways, GO and KEGG analysis were conducted using the 22 overlapping targets. The GO process encompassed three categories: molecular function (e.g., transmembrane receptor protein tyrosine kinase activity), biological process (e.g., positive regulation of phosphorylation), and cell component (e.g., receptor complex, basal plasma membrane and basal part of cell) (Fig. 2A and C). Furthermore, the top 20 KEGG pathway were visualized in Fig. 2D, illustrating the mechanism through which FYN delays osimertinib resistance by influencing multiple signaling pathways. Furthermore, cnetplots were utilized to illustrate specific GO and KEGG terms (Fig. 2E and H). Notably, based on gene numbers and pathway ratios, the PI3K-AKT pathway emerged as the most promising against osimertinib resistance, involving six key targets from the PPI network (ERBB2, EGFR, MET, HSP90AA1, MCL1, and IGF1R). A compound-target-pathway network was constructed using Cytoscape, including 128 nodes and 364 edges, based on KEGG analysis (Fig. 3A). This network visualization revealed that 96 active ingredients of FYN targeted 22 genes and modulated 20 pathways, potentially contributing to the delay of osimertinib resistance. 50 active ingredients were excluded as they did not correlate with these 22 targets (Supplementary Table S2). The findings demonstrate that a single active compound can interact with multiple targets, and conversely, a single target can be influenced by various active compounds. This highlights the multi-component and multi-target interactions of FYN. Subsequently, the compound-target-PI3K-AKT pathway relationship diagram was drew to illustrate the mechanism by which FYN delays osimertinib resistance (Fig. 3B).

Fig. 2.

Functional enrichment analysis of 22 common targets. (A, B) Biological process analysis and specific genes related to the biological process terms. (C, D) Cellular component analysis and specific genes related to the cellular component terms. (E, F) Molecular function analysis and specific genes related to the molecular function terms. (G, H) KEGG analysis and specific genes related to these pathways

Fig. 3.

Network pharmacology predicted the possible mechanism in FYN delaying osimertinib resistance. (A) Compound-target-pathway network of FYN. (B) The sankey diagram of compound-target-PI3K AKT pathway

RT-qPCR experiments

583 µg/ml FYN were used to treat HCC827OR cells for 48 h, 408 µg/ml FYN were used to treat PC9OR cells for 48 h (Supplementary Figure S1). To validate whether FYN delays osimertinib resistance by modulating 6 targets (ERB2, EGFR, MET, HSP90AA1, MCL1, and IGF1R), RT-qPCR was utilized to validate whether FYN delays osimertinib resistance by modulating 6 targets (ERB2, EGFR, MET, HSP90AA1, MCL1, and IGF1R). FYN decreased the levels of MET, ERBB2, and HSP90AA1 (P < 0.05) (Fig. 4A). Notably, FYN decreased the expression levels of MCL1, EGFR, and IGF1R, but there was no significant difference (P > 0.05) (Supplementary Figure S2). Therefore, the targets (MCL1, EGFR, and IGF1T) were excluded.

Fig. 4.

Effects of FYN against HCC827OR xenografts. (A) The mRNA levels of MET, ERBB2, and HSP90AA1 of PC9OR and HCC827OR cells treated with FYN for 48 h. (B) The TUNEL assay of xenograft tumors in each group. (C) The immunohistochemistry assay of expression of Ki-67, MET, ERBB2 and HSP90AA1 in xenograft tumors. ** P < 0.01, ***P < 0.001

TUNEL and immunohistochemistry analysis

TUNEL staining consistently indicated FYN reduced cancer apoptosis in vivo (Fig. 4B). The Ki-67 immunostaining shown that FYN markedly decreased cell proliferation. Similarly, the expression levels of MET, ERBB2 and HSP90AA1 were down-regulated with FYN treatment (Fig. 4C).

Molecular docking analysis

We performed molecular docking, including Ethyl caffeate with MET, ERBB2, and HSP90AA1. The CDOCKER energies were − 6.3, − 5.8, and − 5.7 kcal/mol, respectively (Fig. 5). The findings indicated that Ethyl caffeate had a stronger binding ability with MET, which may be a potential mechanism through which Ethyl caffeate delays lung cancer osimertinib resistance.

Fig. 5.

The docking model of Ethyl caffeate with MET, ERBB2, and HSP90AA1

Ethyl caffeate may inhibit cell proliferation in osimertinib-resistant lung cancer cells

To validate the data acquired from network pharmacology, additional experiments were conducted in vitro. Initially, the growth-inhibitory effects of Ethyl caffeate were evaluated at various concentrations (0-300 µM) after 48 h. The CCK-8 assay demonstrated that the proliferation of PC9OR and HCC827OR cells was markedly suppressed in a dose- and time-dependent manner at 48 h (Fig. 6A). The effect of 78 µM Ethyl caffeate for PC9OR cells or 82 µM Ethyl caffeate for HCC827OR was close to IC₅₀, and these two concentrations were used in the subsequent experiments. Cell migration was observed to be inhibited following 24 h of Ethyl caffeate treatment in both PC9OR and HCC827OR cells (Fig. 6B). PC9OR and HCC827OR cells were treated with IC₅₀ concentrations of Ethyl caffeate for 48 h. The size and shape of the cells were changed, and the cells became increasingly sparse compared to the control group (Fig. 6C). Notably, the EDU experiment revealed that FYN-treated PC9OR and HCC827OR cells proliferation decreased (Fig. 6D).

Fig. 6.

Effects of Ethyl caffeate on the proliferation and invasion of PC9OR and HCC827OR cells. (A) The CCK8 assay of PC9OR and HCC827OR cells treated with Ethyl caffeate for 48 h. (B) The invasion assay of PC9OR and HCC827OR cells treated with or without Ethyl caffeate for 14 days. (C) Morphological changes of PC9OR and HCC827OR cells treated with or without Ethyl caffeate for 24 h (magnification, ×40). (D) The EDU assay PC9OR and HCC827OR cells treated with Ethyl caffeate for 48 h. ** P < 0.01, ***P < 0.001

Ethyl caffeate induces apoptosis in osimertinib-resistant lung cancer cells

To demonstrate whether decreased viability was related to apoptosis induction, Annexin V-FITC/PI staining was performed. The FACS analysis exhibited that the Ethyl caffeate groups increased the apoptosis rate from 4.33 to 8.6% in PC9OR cells and 4.61–7.05% in HCC827OR cells, respectively (Fig. 7A). Moreover, The number of S phase cells in the Ethyl caffeate-treated group increased observably, indicating arrest at the S phase stage (Fig. 7B).

Fig. 7.

Effects of Ethyl caffeate on the apoptosis of PC9OR and HCC827OR cells. (A) The flow cytometry assay of PC9OR and HCC827OR cells treated with Ethyl caffeate for 48 h. (B) The cell cycle analysis of PC9OR and HCC827OR cells treated with Ethyl caffeate for 48 h. (C and D) Effect of Ethyl caffeate on the mRNA and protein expression levels of MET in PC9OR and HCC827OR cells by RT-qPCR and WB experiments. * P < 0.05, ** P < 0.01

Ethyl caffeate may delay osimertinib-resistant by down-regulating the expression of MET

The molecular docking showed that the compound Ethyl caffeate in FYN has a strong binding ability with MET. After PC9OR and HCC827OR cells were intervened with IC₅₀ Ethyl caffeate for 48 h, we performed RT-qPCR and WB experiments. We found that Ethyl caffeate could inhibit the mRNA and protein expression of MET in PC9OR and HCC827OR cells (Fig. 7C and D). Concurrently, Ethyl caffeate could reduce the protein expression of p-PI3K/PI3K and (Supplementary Figure S3). These findings demonstrated that Ethyl caffeate might regulate PI3K/AKT pathway to delay osimertinib resistance.

Discussion

Currently, there is a growing global interest in the use of TCM for the treatment and prevention of cancer [18]. FYN, as a clinical empirical formula, has demonstrated a significant impact on improving symptoms in patients with NSCLC during clinical applications [19]. Moreover, it has exhibited a certain degree of efficacy in delaying drug resistance. Despite these positive clinical outcomes, the specific mechanism of action of FYN remains unclear, particularly regarding its active compounds and key targets. Consequently, we have undertaken, for the first time, a comprehensive approach by integrating network pharmacological analysis with biological methods to delve into the potential mechanisms by which the FYN prescription may delay the onset of drug resistance.

In this work, we identified 904 potential treatment targets from the public database. Simultaneously, we gathered 131 osimertinib-resistant targets from GeneCards databases. 22 shared targets were identified by overlapping between FYN targets and osimertinib-resistant targets. We utilized PPI network analysis to obtain key genes, including EGFR, HSP90AA1, ERBB2, STAT3. The KEGG analysis were carried out to reveal that the mechanism by which FYN delays osimertinib resistance was mediated via multiple signaling pathways, including PI3K/AKT and MAPK signaling pathway. Abnormal activation of PI3K/AKT axis often induces the initiation and progression of various tumors [20]. Moreover, the activation of the PI3K/AKT/NF-κB pathway leads to gemcitabine resistance [21]. The MAPK signaling pathway serves as the principal regulatory module of many cellular processes, including cell proliferation, differentiation, and responses to stress [22]. Further, based on the KEGG pathways, the six genes (ERBB2, EGFR, MET, HSP90AA1, MCL1 and IGF1R) of the top 10 key genes identified through PPI analysis were enriched in PI3K/AKT signaling pathway. Consequently, it is plausible that the modulation of the PI3K/AKT signaling pathway represents a potential mechanism by which FYN delays osimertinib resistance.

We carried out RT-qPCR assay and found that FYN decreased the expression of ERBB2, MET and HSP90AA1 of PC9OR and HCC827OR cells. Immunohistochemistry analysis revealed that the expression levels of MET, ERBB2 and HSP90AA1 were down-regulated following FYN treatment. According to the analysis of network pharmacology, we identified that Ethyl caffeate was active components in FYN which could delay NSCLC osimertinib resistant. Ethyl caffeate, a natural phenolic compound, possesses anti-inflammatory properties [23]. The molecular docking demonstrated that Ethyl caffeate exhibited a particularly strong binding ability with MET. MET, a tyrosine kinase expressed in epithelial cells, mediates biological processes such as proliferation, migration, and invasion [24–25]. High expression of MET in NSCLC is correlated with poor prognosis and shorter survival [26–28]. Abnormal MET signaling, encompassing MET gene amplification, mutation, rearrangement, and over-expression, has the potential to instigate the development of cancer [29]. Amplification of MET can induce resistance to targeted drugs, thereby constraining the clinical effectiveness of NSCLC patents [30]. Anti-MET-targeted therapies has been considered as a viable method for treating NSCLC [31]. We performed PCR and western blotting experiments and found that Ethyl caffeate could reduce MET levels. Concurrently, Ethyl caffeate could reduce the protein expression of p-PI3K/PI3K and p-AKT/AKT. These findings demonstrated that Ethyl caffeate might regulate PI3K/AKT pathway to delay osimertinib resistance.

In summary, FYN and its active component Ethyl caffeate, could be viable alternative options for patients with NSCLC. Given the complexity of the Chinese herbal formula, our future research endeavors will delve deeper into investigating whether FYN holds promise in delaying the onset of NSCLC resistance. We aim to explore various aspects, such as the tumor immune microenvironment, tumor inflammation, and gut microbiota, to comprehensively understand FYN’s potential mechanisms of action.

Conclusion

We analyzed primary bioactive compounds, targeted proteins, and associated pathways through which FYN delays lung cancer osimertinib resistance using network pharmacology. Notably, we found that Ethyl caffeate in FYN had a strong binding ability with MET by molecular docking. Experimental evidence demonstrated that the active compound Ethyl caffeate effectively inhibited the proliferation and invasion of PC9OR and HCC827OR cells by downregulating the expression of MET.

Electronic supplementary material

Below is the link to the electronic supplementary material.

Author contributions

SS wrote the manuscript draft. YH conduct the experiments. HZ conducted the data analyses. The manuscript was finally revised by KX and YG. All authors approved the final version.

Funding

This work was supported by grants from National Natural Science Foundation of China (NO.82074339, NO.82474588, NO.82074231, NO.82374273) and the Natural Science Foundation of Shanghai, China (No.20ZR1459400).

Data availability

The original data are presented as supplementary information.

Declarations

Ethical approval

The animal study was reviewed and approved by the Ethics Committee of Yueyang Hospital of Integrated Traditional Chinese and Western Medicine.

Consent for publication

Not applicable.

Competing interests

The authors declare no competing interests.