Synthesis and biological evaluation, network pharmacology, and molecular docking of 2-benzylidene-1-indanone derivatives as antitumor agents

Abstract

In this study, 45 2-benzylidene-1-indolone compounds were synthesized and their antitumor activities were evaluated. Experimental results demonstrated that these compounds exhibited varying degrees of inhibition against cancer cell proliferation. Notably, compound 20 showed significant activity against HGC-27 gastric cancer cells, with a half-maximal inhibitory concentration (IC₅₀) as low as 2.57 µM. Finally, the antitumor mechanism of compound 20 was investigated through network pharmacology and molecular docking, and 12 key genes and 21 related tumor signaling pathways were obtained. These findings suggest that compound 20, as a potential lead compound, may provide a new direction for the development of antitumor drugs after structural optimization.

Supplementary Information

The online version contains supplementary material available at 10.1038/s41598-025-34583-2.

Introduction

2-Benzylidene-1-indanone is a compound formed through the condensation of methylene at the C2 position of 1-indone with benzaldehyde and removal of one molecule of water. In this compound, the benzene ring involved in the formation of cyclopentanone is named the ring A, the benzene ring connecting the double bond is named the ring B, and the cycloamyl group is named the ring C (Fig. 1). Structurally, this compound can be considered an extended form of chalcone, wherein the carbonyl group and its adjacent phenyl ring along with the olefinic carbon at the 2-position are expanded into a cyclopentanone structure. As such, compared to chalcone, 2-benzylidene-1-indanone exhibits higher conformational stability, whereas chalcone, due to the presence of its α,β-unsaturated carbonyl side chain, displays greater structural flexibility. Despite being derivatives of each other, both compounds share a common π-π conjugated system involving their carbonyl groups and exocyclic double bonds, which endows them with electrophilic α,β-unsaturated carbonyl characteristics. This feature significantly influences their physicochemical properties. For instance, these compounds can undergo Michael addition reactions with thiols within cells¹.

Fig. 1.

Structure of the 2-benzylidene-1-indanone.

The indanone structure is prevalent in numerous bioactive natural products. These natural products, characterized by this scaffold, are not only diverse in types but also exhibit a wide range of physiological activities². Nowadays, drug molecules containing an indanone structure or similar frameworks have been proven to possess various biological activities, such as anti-Alzheimer’s disease³, antiviral⁴, anticancer⁵, and antihypertensive⁶ properties. Specific examples include Donepezil⁷, which is used to improve symptoms of Alzheimer’s disease; Nicosesquiterpene B⁸, which shows strong resistance against tobacco mosaic virus; Dehydropterosin B⁹, which effectively inhibits the growth of human small cell lung cancer cells (NCI-H446) and pancreatic cancer cells (PANC-1); and Indanocine¹⁰, which serves as a diuretic for hypertension management (Fig. 2).

Fig. 2.

Four indenone compounds with different biological activities.

Compounds derived from benzylidene-1-indanone have been validated as effective aromatase inhibitors in cancer therapy. For instance, compound A could effectively inhibit tubulin polymerization, with an IC₅₀ value of 7.56 µM, it can inducing G2/M phase arrest and apoptosis in human lung cancer A-549 cells¹¹. Meanwhile, compound B exhibited IC₅₀ values lower than 11.70 µM for both A-549 and PNAC-1 cells, outperforming the commonly used chemotherapeutic drug cisplatin, which had IC₅₀ values of 22.68 and 39.76 µM against A-549 and PNAC-1 cells respectively (Fig. 3). The anticancer mechanism of this compound might involve increasing intracellular Ca²⁺ levels and inhibiting the activities of cathepsin L and cathepsin D, thereby inducing apoptosis in cancer cells¹².

Fig. 3.

Two 2-benzylidene-1-indanone compounds with good antitumor activity.

Network pharmacology is an innovative approach that integrates biological network analysis to understand drug interactions and therapeutic mechanisms¹³. It leverages multiple databases to identify potential drug targets and pathways of action for various disease¹⁴. Moreover, molecular docking is an effective way to verify whether a drug and a potential target have a good affinity¹⁵. Therefore, network pharmacology and molecular docking can be used for the development of new drugs.

In short, indenone derivatives can be used as medicinal molecules, but also as raw materials for drug synthesis. They offer more possibilities for the development of new drugs. In addition, the use of network pharmacology and molecular docking to explore the potential targets and mechanisms of action of indenone derivatives is an effective method.

Results and discussion

Synthesis

In the system of sodium hydroxide, water, and ethanol, 2-benzylidene-1-indanone compounds 1–45 were synthesized from substitued 1-indanones and substitued benzaldehyde at room temperature (Fig. 4)¹⁶. The 1-indanones included 4-hydroxy-1-indanone, 5-hydroxy-1-indanone, 6-hydroxy-1-indanone, 5-fluoro-1-indanone, and 5-methyl-1-indanone. Furthermore, the benzaldehydes included p-hydroxybenzaldehyde, 2,3,4-trimethoxybenzaldehyde, 3,4,5-trimethoxybenzaldehyde, 2-(trifluoromethyl)benzaldehyde, 3-(trifluoromethyl)benzaldehyde, 4-(trifluoromethyl) benzaldehyde, 2-bromobenzaldehyde, 3-bromobenzaldehyde, 4-bromobenzaldehyde, 2,4-difluorobenzaldehyde, and 3-(trifluoromethoxy)benzaldehyde. These five 1-indanones and 11 benzaldehydes are readily available.

Fig. 4.

The synthetic route of compounds 1–45.

In summary, substituents for 1-indoles included electron-donating hydroxyl and methyl groups and electron-withdrawing fluorine atoms. In addition, substituents on these benzaldehydes included electron-donating hydroxyl and methoxy groups, as well as electron-withdrawing fluorine and bromine atoms, trifluoromethyl and trifluoromethoxy groups.

Biological activity

Antitumor activity

To comprehensively evaluate the antitumor potential of the synthesized compounds, five human cancer cell lines were selected for in vitro cytotoxic assays. These include hepatocellular carcinoma cell lines (SMMC-7721 and HUH-7), a lung adenocarcinoma cell line (A-549), a glioblastoma cell line (U-87), and a gastric carcinoma cell line (HGC-27). These cell lines were chosen based on their clinical relevance, distinct tissue origins, and widespread use as representative models in antitumor drug screening. In particular, the HGC-27 cell line demonstrated the highest sensitivity to several compounds, providing a basis for further mechanistic studies.Tables 1 and 2 present the results of the antitumor activity assays. As a result, most compounds exhibited a certain level of inhibition against these five types of cells.

Table 1.

Inhibitory effects of compounds 1–45 on five tumor cells.

Comps	Percentage inhibition (%)
	SMMC−7721	A−549	U−87	HGC−27	HUH−7
1	9.64 ± 11.28	24.99 ± 0.58	−21.85 ± 14.94	73.02 ± 3.74	50.33 ± 2.53
2	−23.09 ± 4.48	21.89 ± 0.79	6.66 ± 2.85	51.12 ± 9.14	−8.89 ± 6.63
3	−29.12 ± 2.32	10.05 ± 1.12	10.71 ± 1.08	−10.87 ± 4.87	6.67 ± 16.51
4	−15.06 ± 8.60	17.18 ± 0.77	8.50 ± 6.40	−1.53 ± 11.34	−16.05 ± 13.79
5	30.72 ± 7.02	16.92 ± 4.51	−0.77 ± 1.16	41.28 ± 0.75	18.35 ± 21.93
6	−49.20 ± 15.55	41.31 ± 1.81	22.37 ± 2.54	55.46 ± 6.15	35.34 ± 6.04
7	74.58 ± 6.04	72.94 ± 0.68	30.56 ± 0.76	80.53 ± 2.44	65.60 ± 4.94
8	56.23 ± 5.50	15.13 ± 4.32	−17.59 ± 2.81	47.29 ± 1.97	−2.30 ± 18.49
9	−33.95 ± 14.47	6.99 ± 1.31	0.29 ± 2.16	72.55 ± 7.07	57.78 ± 9.82
10	−28.75 ± 13.08	8.93 ± 2.61	39.64 ± 2.37	29.61 ± 3.85	45.79 ± 10.36
11	−49.75 ± 16.43	26.50 ± 3.67	20.31 ± 3.96	67.48 ± 3.60	45.66 ± 11.19
12	5.79 ± 13.48	12.95 ± 1.58	2.86 ± 9.63	35.43 ± 3.30	33.68 ± 6.39
13	−39.49 ± 21.40	7.57 ± 8.29	−11.10 ± 4.95	73.52 ± 9.46	38.61 ± 20.32
14	−14.12 ± 7.91	18.92 ± 2.96	−2.67 ± 0.93	31.47 ± 8.67	32.27 ± 7.93
15	69.07 ± 9.79	25.92 ± 1.45	−23.32 ± 3.58	24.09 ± 6.05	46.67 ± 16.18
16	−19.03 ± 12.63	23.55 ± 1.61	−11.55 ± 3.21	30.07 ± 5.64	41.19 ± 8.26
17	47.92 ± 6.77	30.27 ± 3.24	−16.92 ± 4.11	25.36 ± 9.29	37.86 ± 9.51
18	5.75 ± 4.35	39.67 ± 1.82	18.75 ± 3.11	84.74 ± 1.65	67.39 ± 3.68
19	62.04 ± 1.11	30.81 ± 0.62	−4.07 ± 5.77	86.27 ± 4.68	60.97 ± 8.57
20	−41.64 ± 14.35	57.22 ± 1.92	N.T.	68.41 ± 2.83	74.31 ± 0.42
21	13.61 ± 15.62	37.02 ± 1.98	61.94 ± 4.16	81.59 ± 1.03	50.97 ± 13.85
22	7.64 ± 15.85	29.05 ± 2.00	35.41 ± 1.82	40.20 ± 9.89	54.06 ± 2.05
23	−53.90 ± 8.40	31.70 ± 0.49	2.365.39	77.181.89	41.259.64
24	70.92 ± 3.53	42.86 ± 0.60	16.47 ± 1.60	87.81 ± 1.34	69.95 ± 5.35
25	17.37 ± 3.88	26.43 ± 6.32	54.23 ± 2.31	89.22 ± 2.40	50.88 ± 8.99
26	5.16 ± 8.76	20.29 ± 4.06	49.56 ± 8.53	76.00 ± 7.82	38.52 ± 10.07
27	13.21 ± 10.24	9.95 ± 3.28	50.54 ± 3.36	70.56 ± 2.16	37.27 ± 4.54
28	19.92 ± 19.97	−8.33 ± 4.42	57.96 ± 1.91	80.22 ± 0.59	69.54 ± 1.77
29	60.66 ± 13.06	47.67 ± 6.31	45.45 ± 3.55	93.67 ± 2.03	72.87 ± 3.90
30	−26.28 ± 0.69	8.83 ± 0.33	−6.89 ± 4.08	0.30 ± 15.71	19.68 ± 10.27
31	−32.29 ± 2.10	14.52 ± 5.98	−2.61 ± 10.35	63.46 ± 4.84	23.91 ± 10.66
32	−42.53 ± 2.82	9.93 ± 5.59	−1.12 ± 1.67	43.93 ± 4.31	−0.57 ± 17.13
33	−19.72 ± 4.52	43.43 ± 0.85	35.97 ± 2.45	41.50 ± 3.41	42.63 ± 7.23
34	−62.41 ± 4.78	20.92 ± 4.89	10.41 ± 1.94	42.22 ± 4.40	23.80 ± 7.08
35	−49.25 ± 13.82	13.09 ± 1.71	13.89 ± 1.88	35.30 ± 9.31	49.86 ± 2.81
36	16.85 ± 20.96	10.40 ± 0.63	N.T.	14.87 ± 4.81	−5.35 ± 6.98
37	2.50 ± 14.68	7.40 ± 0.37	N.T.	2.71 ± 6.79	11.71 ± 8.40
38	−43.06 ± 4.82	7.17 ± 2.56	−17.63 ± 1.67	43.00 ± 1.76	21.87 ± 15.03
39	6.30 ± 2.49	67.50 ± 0.58	64.27 ± 2.00	67.67 ± 3.83	68.40 ± 1.71
40	−35.03 ± 2.83	−21.78 ± 7.94	N.T.	10.58 ± 17.65	14.35 ± 8.81
41	−26.85 ± 5.15	−5.37 ± 1.71	30.82 ± 1.16	15.54 ± 9.94	31.02 ± 5.19
42	−35.03 ± 2.92	3.01 ± 6.85	26.76 ± 6.64	−4.96 ± 12.00	10.84 ± 6.91
43	11.77 ± 7.52	−15.95 ± 4.67	20.50 ± 2.56	2.58 ± 6.50	36.62 ± 5.24
44	−0.23 ± 0.58	−10.69 ± 3.09	32.10 ± 2.86	3.46 ± 3.59	28.85 ± 7.32
45	−29.30 ± 3.71	N.T.	35.89 ± 4.68	31.54 ± 13.16	30.71 ± 5.88

Note: The drug screening concentration was 40 µM and N.T. indicates not measured.

Table 2.

The IC₅₀ values of some compounds against HGC-27 and BEAS-2B cells.

Compounds	HGC-27 cells	BEAS-2B cells
1	20.43	–
7	15.87	–
9	25.95	–
11	25.17	–
13	29.05	–
18	55.85	–
20	2.57	> 40
21	22.34	–
23	32.12	–
24	44.09	–
25	13.79	–
26	55.51	–
27	48.94	–
28	21.54	–
29	50.18	–
31	29.62	–
32	15.18	–
40	19.03	–
Cisplatin	22.25	–
Taxol	0.10	< 0.016
Doxorubicin	–	0.87

Note: Drug screening concentrations were 40, 8, 1.6, 0.32, 0.064 µM, respectively.

Table 1 shows the inhibitory activity of compounds 1–45 against these five types of cells. By analyzing the specific structure of these compounds, several rules were found. Firstly, introducing a bromine atom on ring B (as seen in compounds 2, 3, 4, 5, 14, 16, 17, 36, 37, 41, and 42) did not significantly enhance the inhibitory effects of these compounds on tumor cells. Secondly, compounds with three methoxy groups introduced into ring B (such as compounds 3, 10, 12, 15, 22, 23, 33, 34, and 39) exhibited enhanced antitumor activity to varying degrees. Notably, compounds 10, 12, 22, and 34 demonstrated significantly higher activity compared to compounds 3, 15, 23, and 33. This difference may be attributed to the spatial distribution of methoxys on the ring B: the former has methoxys at the C3’, C4’, and C5’ positions, while the latter has methoxys at the C2’, C3’, and C4’ positions. This specific spatial arrangement facilitates more favorable interactions between the compounds and tumor cells, thereby enhancing their antitumor efficacy. Thirdly, compounds with a trifluoromethyl group introduced on ring B (such as compounds 7, 8, 9, 18, 19, 28, 29, and 30) exhibited significantly enhanced antitumor activity. Further analysis reveals that the inhibition activity is strongest when there is a hydroxyl substitution at C6 position on ring A; it is slightly less potent with a hydroxyl substitution at C5 position on ring A, followed by a hydroxyl substitution at C4 position on ring A. The weakest inhibitory effect is observed with a fluorine substitution at C5 position on ring A. Additionally, when the substituents on ring A are identical and occupy the same positions, different substitution positions of the same substituent on ring B also significantly influence inhibitory activity. Specifically, the order of activity is: C3’ position substitution > C4’ position substitution > C2’ position substitution.

Considering the notable inhibitory effects observed against HGC-27 cells among the synthesized compounds, those exhibiting an inhibition rate exceeding 60% were chosen for further evaluation of their IC₅₀ values, with the exception of compound 39.Meanwhile, two anticancer drugs, cisplatin and paclitaxel (Taxol), were co-assayed as positive controls (Table 2). As a result, compounds 1, 7, 20, 25, 28, 32, and 40 exhibited the obvious inhibitory effects on HGC-27 cells, with IC₅₀ values ranging from 2.57 to 19.03 µM. However, the IC₅₀ value of cisplatin was 22.25 µM. Therefore, these compounds exhibited stronger inhibitory activity against HGC-27 cells than cisplatin. In particular, compound 20 demonstrated an low IC₅₀ value of 2.57 µM against HGC-27 cells, highlighting its remarkable antitumor potential and providing crucial reference data for further research into this class of compounds. To evaluate the toxicity of compound 20 towards normal cells, its IC₅₀ value against the human bronchial epithelial-like BEAS-2B cells was determined, with paclitaxel and doxorubicin serving as positive control drugs. The results indicated that the IC₅₀ value of compound 20 for BEAS-2B cells was greater than 40 µM, while the IC₅₀ values of paclitaxel and doxorubicin for BEAS-2B cells were less than 0.016 and 0.87 µM, respectively. This suggests that the toxicity of compound 20 towards BEAS-2B cells is significantly lower than that of paclitaxel and doxorubicin. This finding implies that indenone compounds have much lower toxicity to normal cells than paclitaxel and doxorubicin, indicating a higher potential for safety in drug development.

Based on these experimental findings, the structure-activity relationship (SAR) of 2-benzylidene-1-indanone compounds is summarized (Fig. 5).

Fig. 5.

Summary of the SAR of the prepared compounds.

In short, structural modifications of the compounds significantly impact their antitumor activity, particularly the position and type of substituents. Future research can further explore these structural characteristics to develop more promising antitumor drugs.

Network pharmacology

Selection of compounds in network pharmacology

Among these compounds, compound 20 exhibited the best antitumor activity, with an IC₅₀ value of 2.57 µM against HGC-27 cells, so it was chosen for network pharmacology research.

SwissADME prediction of compound 20

The properties of compound 20 were predicted using the SwissADME database, encompassing its lipophilicity, water solubility, pharmacokinetics, and druglikeness. The consensus log Po/w value of compound 20 was 3.67. Its solubility was classified as moderate, while its gastrointestinal absorption was rated as “high”. Notably, compound 20 demonstrated the ability to penetrate the blood-brain barrier (“Yes”), suggesting its potential efficacy in treating brain-related diseases. In terms of druglikeness, compound 20 received five “Yes” ratings, indicating favorable pharmacokinetic properties and bioavailability. The SwissADME predicted data for compound 20 is shown in supplementary figure S162.

Target prediction of compound 20

SwissTargetprediction and TargetNet database were utilized to predict the target gene of compound 20. After eliminating duplicate entries, a total of 128 unique target genes were identified.

Identification of tumor-related genes

The three words “carcinoma”, “cancer” and “tumor” were searched separately in the GeneCards database, and 39,279 tumor-related genes were obtained after removing duplicates.

Oncogenes prediction by compound 20

The genes predicted by compound 20 were intersected with tumor-related genes using the Venny 2.1 web tool. This intersection yielded a final set of 120 oncogenes associated with compound 20 (Fig. 6).

Fig. 6.

Intersection diagram of target genes and oncogenes.

Construction of protein-protein interaction network

In constructing the protein-protein interaction (PPI) network, a total of 120 target genes were uploaded to the STRING website. The resulting PPI network, as illustrated in Fig. 7, comprises 120 nodes and 609 edges, revealing intricate interactions among the proteins.

Fig. 7.

PPI network of the potential targets of compound 20.

Calculate the topological parameters of proteins

To analyze the data in Cytoscape 3.9.1 software, imported the table from the STRING analysis. Created a CSV file that included various parameters such as betweenness, closeness centrality, clustering coefficient, and degree values. Visualized the network based on the degree values (as shown in Fig. 8). Genes with a degree value of 5 or higher were selected as target genes, resulting in a total of 89 target genes.

Fig. 8.

PPI network of potential target of compound 20.

The target main genes were obtained by molecular Docking simulation

Compound 20 was docked with 89 target proteins using BIOVIA Discovery Studio 2023 software. Proteins with a docking score of ≥ 90 were identified as main target proteins, and their corresponding genes were classified as main target genes. Consequently, 66 genes were determined to be main target genes.

Identification of important target genes

The table obtained from the STRING analysis tool (in TSV format) was imported into Cytoscape 3.9.1 software. Genes with a degree value less than 5 and a LibDock docking score below 90 were removed. The remaining 66 genes were further analyzed using the CytoHubba plugin. Three algorithms—maximum cluster centrality (MCC), neighborhood component centrality (MNC), and degree centrality—were employed to identify hub genes. Based on their scores, the top 10 hub genes were selected. The results from these three algorithms were then intersected, yielding four key hub genes: ESR1, MMP9, RELA, and MMP2. These genes were identified as potential hub genes for compound 20 (Fig. 9). Additionally, the 66 target genes were analyzed using the MCODE plugin to identify the highest-scoring module, which comprised 16 nodes and 50 edges, scoring 6.667 points. This indicates that these 16 nodes represent important target genes, including three of the aforementioned potential hub genes.

Fig. 9.

The networks of the top 10 key genes of compound 20. (MCC, MNC, Degree, Venny charts, respectively).

Enrichment analysis

The 66 target genes were uploaded to the Metscape platform for comprehensive analysis. This analysis encompassed biological processes (BP), cellular components (CC), molecular functions (MF), and KEGG pathway enrichment. In the bar chart representation, the length of each bar inversely correlates with the P-value; longer bars signify a smaller P-value, thereby indicating a more statistically significant difference between the experimental and control groups.

For the Gene Ontology Biological Process (GO BP) analysis, the top ten most representative biological processes identified were: cellular response to nitrogen compounds, metabolic processes of organic hydroxy compounds, phosphorylation, regulation of proteolysis, modulation of chemical synaptic transmission, responses to light stimuli, xenobiotic stimuli, synaptic signaling, metal ion stimuli, and cellular responses to organic cyclic compounds (as illustrated in Fig. 10).

Fig. 10.

The analysis of GO BP.

For the Gene Ontology Cellular Component (GO CC) analysis, the top ten representative components identified were: dendrite, membrane raft, presynapse, receptor complex, serine/threonine protein kinase complex, neuromuscular junction, postsynapse, mitochondrial outer membrane, Golgi-associated vesicle, and extracellular matrix (as illustrated in Fig. 11).

Fig. 11.

The analysis of GO CC.

For the Gene Ontology Molecular Function (GO MF) analysis, the top ten representative molecular functions identified were: carbonate dehydratase activity, oxidoreductase activity, protein kinase activity, endopeptidase activity, steroid hydroxylase activity, protein kinase binding, protein tyrosine kinase activity, primary amine oxidase activity, protein serine/threonine kinase binding, and heme binding (as illustrated in Fig. 12).

Fig. 12.

The analysis of GO MF.

In the bubble map for KEGG pathway analysis, the intensity of the red color corresponds to a smaller P-value, signifying a greater statistical significance in the differences between the experimental and control pathways. Additionally, the larger the bubble size, the more genes are involved in the experimental pathway.

In this study, the first 21 cancer-related pathways were selected in order of P value from smallest to largest (as illustrated in Fig. 13). They were pathways in cancer, metabolic pathways, bladder cancer, proteoglycans in cancer, prostate cancer, NF-kappa B signaling pathway, neuroactive ligand-receptor interaction, pancreatic cancer, cAMP signaling pathway, PD-L1 expression and PD-1 checkpoint, pathway in cancer, ras signaling pathway, IL-17 signaling pathway, calcium signaling pathway, TNF signaling pathway, MAPK signaling pathway, Rap1 signaling pathway, chemical carcinogenesis-receptor activation, human T-cell leukemia virus 1 infection, human cytomegalovirus infection, acute myeloid leukemia, non-small cell lung cancer. For convenience, these 21 paths are abbreviated to PW1-21 in turn.

Fig. 13.

The analysis of KEGG pathway.

Compound 20 showed strong inhibitory activity against HGC-27 cells. Therefore, the potential targets and mechanisms of compound 20 in the treatment of gastric cancer were obtained (Fig. 14).

Fig. 14.

Potential targets and mechanisms of compound 20 against gastric cancer cells. (The pink codes represent targeted genes involved in this pathway).

Identification of the potential mechanisms of action

The analysis encompassed KEGG pathways such as pathways in cancer, pancreatic cancer, and non-small cell lung cancer, among other tumor-related pathways, all of which were correlated with bioactivity experimental results. Upon aggregating the genes from the foremost 21 pathways, a total of 57 genes were identified. These were then intersected with 66 significant genes that had Lib_dock docking scores of ≥ 90, leading to the identification of 12 key genes: ALK, CA12, MMP1, ROCK1, CASP9, BCL2A1, CNR2, FLT1, TBK1, MMP14, CYP1B1, and LCK. A mapping was established to illustrate the relationships between these 21 pathways and their associated genes (as shown in Fig. 15).

Fig. 15.

The first 21 pathways and their 57 genes.

Molecular docking simulation of compound 20 with some representative target proteins

These results indicate that compound 20 exhibits significant binding affinity towards these five proteins (as detailed in Table 3).The binding strength of compound 20 to the target proteins CYP1B1 (PDB ID: 6IQ5), MMP14 (PDB ID: 3C7X), FLT1 (PDB ID: 1RV6), ROCK1 (PDB ID: 8GDS), and CA12 (PDB ID: 7PP9) was evaluated using the Lib_dock docking score, with scores of 108.8, 105.6, 104.0, 103.6, and 100.1, respectively. A docking score of Lib_dock ≥ 90 indicates a good affinity between the compound and the target protein. Compound 20 had a docking score greater than 100 for all five proteins, indicating strong binding characteristics with these target proteins. Further CDOCKER docking analysis was conducted, where a higher absolute docking value indicates stronger affinity between the compound and the target protein. The key gene MMP14 had the lowest CDOCKER docking score of −12.62, indicating the best docking effect. Compound 20 forms hydrogen bond interactions with the amino acids GLY-331, MET-468, and PHE-467 in the MMP14 gene, and a hydrophobic interaction with the amino acid MET-422. The molecular docking results of compound 20 with CYP1B1, MMP14, FLT1, ROCK1, and CA12 are illustrated in Figs. 16, 17, 18 and 19, and 20, respectively.

Table 3.

Result of the molecular Docking simulation.

Protein	Binding affinity		Interacting amino acids
	Lib_dock	CDOCKER	Hydrophobic interaction	Hydrogen bond
CYP1B1	108.77	−4.22	ALA−330, CYS−470, ILE−471, ILE−399	ARG−117, ALA−133, THR−398
MMP14	105.61	−12.62	MET−422	GLY−331, MET−468, PHE−467
FLT1	104.04	−7.15	LYS−200, CYS−34, PRO−173, LEU−174	-
ROCK1	103.58	−2.77	PHE−233, GLY−234, LEU−123, ARG−141, MET−144, LEU−155, MET−169	MET−144, ASP−232
CA12	100.13	−6.50	LEU−198, THR−200, HIS−94, VAL−121, VAL−143	GLN−92

Fig. 16.

Binding mode of compound 20 and CYP1B1. (A) 2D binding mode diagram (B) 3D binding mode diagram.

Fig. 17.

Binding mode of compound 20 and MMP14. (A) 2D binding mode diagram (B) 3D binding mode diagram.

Fig. 18.

Binding mode of compound 20 and FLT1. (A) 2D binding mode diagram (B) 3D binding mode diagram.

Fig. 19.

Binding mode of compound 20 and ROCK1. (A) 2D binding mode diagram (B) 3D binding mode diagram.

Fig. 20.

Binding mode of compound 20 and CA12. (A) 2D binding mode diagram (B) 3D binding mode diagram.

Conclusions

A total of 45 compounds were synthesized via the Claisen-Schmidt reaction. Through in vitro bioactivity evaluations, compounds 1, 7, 20, 25, 28, 32, and 40 emerged as some promising antitumor agents. Notably, compound 20 demonstrated significant antitumor activity. To further understand its potential, the antitumor mechanism of compound 20 was explored using network pharmacology and molecular docking analyses. These findings provide valuable insights and serve as a reference for the development of 2-benzylidene-1-indanone derivatives as potential anticancer agents.

Experimental

Chemistry

The chemical reagents used in this study were purchased from Shanghai Aladdin Biochemical Technology Co., Ltd. (Shanghai, China) and were of the highest commercially available purity. All reagents were used as received without modification. The melting points were determined using a micro-melting point apparatus and the data were not corrected (Shanghai Precision Scientific Instrument Co., Ltd., Shanghai, China). NMR spectra of ¹H and ¹³C were determined in acetone-d₆ or DMSO-d₆ at 500 MHz for ¹H and 125 MHz for ¹³C (Bruker, Fallanden, Switzerland). Infrared spectra were recorded with KBr pellets (Bruker, Bremen, Germany). High-resolution mass (HRMS) data were recorded via a negative or positive ion electron impact mass spectrometry using a time-of-flight analyzer (Bruker Daltonics Inc., Boston, USA).

Synthesis

Synthesis of compounds 1–45

Substituted 1-indanone (1.00 mmol) and substituted benzaldehyde (1.00 mmol) were added to a reaction test tube. Next, absolute ethanol (5 mL) and a 5% NaOH solution (0.5 mL) were introduced into the test tube. The reaction mixture was then placed in an organic synthesis apparatus and reacted at room temperature for 1 h. The progress of the reaction was monitored using thin-layer chromatography, continuing until the reaction reached equilibrium. Upon completion of the reaction, it was quenched by carefully adding an appropriate amount of water. The mixture was subsequently extracted with ethyl acetate (3 × 10 mL) and washed with saturated saline solution (3 × 10 mL). The organic layer was collected in a conical flask, and anhydrous sodium sulfate was added to dry the solution. After allowing the mixture to stand for a period of time, the solvent was removed under reduced pressure using a rotary evaporator. Finally, the crude product was purified via medium-pressure preparative chromatography, employing a gradient of ethyl acetate and petroleum ether as eluents. The volume ratio of ethyl acetate to petroleum ether gradually increased from 0 to 1. In this case, the raw materials and the products were separated. The purified product was concentrated under reduced pressure using a rotary evaporator to yield the final target compound. The purity of each target compound was greater than or equal to 98%.

Characterization of compounds

Due to space limitations, spectral data for all final products have been placed in supplementary figures S1–S161.

Biological activity

Cytotoxicity tests

The inhibitory activities of target compounds against various cancer cell lines and a normal lung epithelial cell line, including liver cancer cells (SMMC-7721, HUH-7), lung cancer cells (A-549), glioma cells (U-87), gastric cancer cells (HGC-27), and lung epithelial cells (BEAS-2B), were evaluated using the MTS assay¹⁷. These cell lines were obtained from ATCC (Manassas, VA, USA). A single-cell suspension was prepared in DMEM or RMPI1640 medium supplemented with 10% fetal bovine serum, and 100 µL of the suspension was seeded into 96-well plates at a density of 3,000–15,000 cells per well. Cells were allowed to adhere for 12–24 h before treatment. Initial screening of compounds was performed at a concentration of 40 µM, followed by re-screening at concentrations of 0.064, 0.32, 1.6, 8, and 40 µM. After 48 h of incubation at 37 °C, the culture medium was discarded, and 20 µL of MTS solution, along with 100 µL of culture medium, was added to each well. The wells were then incubated for an additional 2–4 h to allow for complete reaction. The optical absorbance at 492 nm was measured using a multifunctional microplate reader, and the results were recorded. Three blank wells containing 20 µL of MTS solution and 100 µL of culture medium were included for background correction. The experiments were repeated to ensure the accuracy of the IC₅₀ values for each concentration. The IC₅₀ values of the tested compounds were calculated using the Reed and Muench method. Two positive controls, cisplatin and Taxol, were included in each experiment, and their IC₅₀ values were also determined using the Reed and Muench method¹⁸.

Network pharmacology

SwissADME prediction of compounds

The compound in sdf format was uploaded to the SwissADME database (http://www.swissadme.ch/) to obtain its Smiles number, and the “run” button was subsequently clicked to generate the data^17,19–21.

Target prediction of compounds

The compounds in sdf format were imported into the Swiss Target Prediction database (http://swisstargetprediction.ch/)^22,23 and the TargetNet database (http://targetnet.scbdd.com/)²⁴, respectively. “Homo sapiens” was selected as the target species, and, following the provided instructions, the corresponding targets of the compounds were obtained.

Tumor-related genes

The GeneCards database (https://www.genecards.org/) was queried using tumor-related terms to identify relevant tumor-associated genes.

Oncogenes predicted by compounds

The genes predicted for the compounds and the cancer-related genes were intersected using the Venny 2.1 website to identify the cancer-related target genes of the compounds (https://bioinfogp.cnb.csic.es/tools/venny/index.html).

Construction of PPI network

The target genes were input into the STRING database (https://string-db.org/) using the “Multiple Proteins” module. “Homo sapiens” was selected as the target organism, and “evidence” was chosen to define the meaning of the network edges. Active interaction sources, including experimental data and co-expression, were selected as parameters, with a medium confidence threshold set at 0.400. The PPI network was then generated following the analysis.

Calculation of the topological parameters of proteins

The table in tsv format generated by STRING analysis was imported into Cytoscape 3.9.1 software. The tools “Network Analyzer” and “Network Analysis” were sequentially applied, followed by the selection of “Appropriate network settings”. A csv table was then constructed, containing parameters such as betweenness centrality, closeness centrality, clustering coefficient, and degree. Proteins with a degree value ≥ 5 were considered highly connected, indicating their significant role in maintaining the connectivity of complex biological networks. Accordingly, target genes with a degree value ≥ 5 were selected for further analysis.

The main target was obtained by molecular Docking

The protein names of potential target genes were retrieved from UniProt (https://www.uniprot.org/), and verified human protein structures were obtained from the PDB database (https://www.rcsb.org/). After the proteins were dehydrated, modified, and hydrogenated, they were individually docked with compound 20 using BIOVIA Discovery Studio 2023 software (URL link: https://www.3ds.com/products-services/biovia/products/molecular-modeling-simulation/biovia-discovery-studio/).

Identification of important target genes

A table (in tsv format) listing target genes with good docking scores to compound 20 was imported into the Cytoscape 3.9.1 software, followed by analyses using the CytoHubba²⁵ and MCODE²⁶ plug-ins to obtain the important target genes.

Enrichment analysis

The genes were entered into the Metscape website (https://metascape.org/) for analysis of their GO BP, GO CC, and GO MF. Subsequently, the genes were submitted to the DAVID database (https://david.ncifcrf.gov/) to identify their associated KEGG pathways.

The pathway map was downloaded from the KEGG Mapper-Color website (https://www.genome.jp/kegg/tool/map_pathway2.html). Based on the analysis results, a cluster bar chart for GO functional enrichment and a bubble diagram for KEGG enrichment were generated.

Molecular docking simulation

The protein names of potential target genes were retrieved from the UniProt website (https://www.uniprot.org/), and experimentally validated human protein structures were obtained from the PDB website (https://www.rcsb.org/). After dehydration, modification, and hydrogenation, the proteins were individually docked with compound 20 using BIOVIA Discovery Studio 2023 software.

Supplementary Information

Below is the link to the electronic supplementary material.

Author contributions

X.L. synthesized 36 compounds and wrote the main manuscript, R.Z. evaluated the antitumor activity, T.Z. synthesized nine compounds, M.X. prepared some figures, and B.Z. provided the financial support, organized experiments and revised the article. All authors reviewed the manuscript.

Funding

The authors gratefully acknowledge the financial support from the Putian City Science and Technology Plan Project (Grant No. 2021S2001-9) and the Project Approved by Fujian Shanhe Pharmaceutical Co., LTD (Grant No. 2022AHX211(L)).

Data availability

All data both in the manuscript and supplementary information files are available upon request by contact with the corresponding author.

Declarations

Competing interests

The authors declare no competing interests.