2.1 Materials and instruments

Unless otherwise indicated, all reagents and anhydrous solvents were purchased from commercial sources and used without purification. ¹H NMR and ¹³C NMR spectra were recorded on a Bruker AV-500/400 spectrometer (Germany). Internal tetramethylsilane was used as chemical shift standard. HRMS was recorded on a Thermo Scientific Accela-Exactive High Resolution Accurate Mass Spectrometer (USA). Low and high resolution mass spectra (LRMS and HRMS) were given with electron impact mode. Melting points were measured on an RY-1 melting point apparatus (China). Column chromatography was performed using silica gel (300–400 mesh, Qingdao Haiyang Chemical Co., Ltd., China).

2.2 General procedure I for esterification

DCC (1.5 eq) and DMAP (0.3 eq) were added to a solution of I-1 (0.55 mmol, 1 eq) and amino acid protected by N-Boc/Cbz (1.5 eq) in anhydrous CH₂Cl₂ (5 mL). After stirring at room temperature for 12 h, the reaction solution was filtrated, and the filter residue was washed by CH₂Cl₂ (5 mL × 3). The filtrate was concentrated in vacuum, the residue was dispersed in EtOAc (25 mL), washed with saturated saline (10 mL × 3), dried over anhydrous Na₂SO₄, filtered, concentrated, and purified by silica gel column chromatography.

2.3 General procedure II for de-boc reaction

Ethyl acetate solution of saturated hydrogen chloride (3 mL) was added to a solution of Boc-protected compound (0.66 mmol, 1 eq) in EtOAc (5 mL). The reaction solution was stirred at rt. for 2 h, and concentrated by rotary evaporation under reduced pressure. The residue was rotated twice with EtOAc, and then stirred in EtOAc (10 mL) for 1 h, filtered, and washed by EtOAc to get white solid.

2.4 General procedure III for debenzylation

A mixture of benzyl protected derivative (1 eq.), ammonium formate (10 eq.) and 10 % Pd/C (10 % weight) in MeOH/THF (4:1, V/V, 5 mL) was stirred at reflux for 12 h. The reaction mixture was filtered through Celite, and the insoluble substance was washed with MeOH (10 mL × 3). The filtrate was concentrated in vacuum to give an off-white solid, which was diluted in EtOAc (3 mL), and added ethyl acetate solution of saturated hydrogen chloride (2 mL). After stirring at room temperature for 8 h, the mixture was concentrated by rotary evaporation under reduced pressure. The residue was rotated three times with EtOAc, and then stirred in EtOAc (5 mL) at room temperature for 1 h, filtered, and washed by EtOAc to get white solid.

2.5 General procedure IV for esterification

To a solution of 3-benzyloxy-16α-bromo-Estra-1,3,5 (10)-Triene-17-one (II-1, 0.100 g, 0.228 mmol, 1 eq) in DMF (5.0 mL), K₂CO₃ (0.035 g, 0.25 mmol, 1.1 eq), KI (0.011 g, 0.068 mmol, 0.3 eq) and N-Boc protected amino acid (0.274 mmol, 1.2 eq) were added. After stirring at 60 °C for 12 h, the mixture was diluted with 2 N HCl (50 mL) and extracted with EtOAc (3 × 35 mL). The combined organic layers were washed successively with saturated aqueous NaHCO₃, H₂O, and saturated saline, dried over MgSO₄, filtered, and concentrated. The residue was purified by silica gel column chromatography.

2.6 Synthesis of I-2 ∼ I-19 and II-2 ∼ II-11

2.7 MTT assay

All cells were cultivated in Dulbecco's Modified Eagle Medium with 10 % Fetal Bovine Serum at 37 °C and 5 % CO₂. The cells were seeded in 96-well plates (5 × 10³ cells per well) and cultivated for 24 h. The compounds were treated with different concentrations (20, 10, 5, 2.5, 1.25 μM) for 48 h. Each well was added 20 μL MTT (5 %) solution and cultivated for 4 h, then removed the supernatant and added DMSO. Absorbance at 570 nm was recorded. The IC₅₀ value was calculated by SPSS software, and at least 3 parallel tests were performed.

2.8 Human apoptosis antibody assay

Apoptosis-related protein expression was detected and analyzed using the Human Apoptosis Array Kit (ARY009, R&D Systems). Membranes containing immobilized apoptosis-related antibodies were sealed with bovine serum albumin for 1 h. Membranes were then incubated overnight at 2–8 °C with a cocktail of proteins extracted from cells and detection antibodies. The membranes were incubated with horseradish peroxidase labeled streptavidin for 30 min and images were visualized with chemiluminescent reagents. The average background signal was subtracted from each spot to obtain the average signal (pixel density) of paired replicate spots representing each apoptosis-related protein.

2.9 Molecular docking

Structures of proteins mTOR (PDB: 4DRI), KIT (PDB: 6GQM), ESR1 (PDB: 1SJ0), PDK2 (PDB: 4MPC), and RXRA (PDB: 3OAP) were downloaded from the RCSB protein database ( https://www.rcsb.org/). The structures of the protein molecules were opened with PyMOL, water molecules were removed, small molecule ligands and other molecules bound were removed, polar hydrogens were added, molecular docking experiments were performed using AutoDock software, the conformation with the best binding energy was selected, and the pictures were plotted using PyMOL.

3.1 Synthesis

3.1.1

3.1.1 Synthesis of 17β-estradiol (17-OH)-amino acid derivatives

As shown in Scheme 1, I-1 (1 eq) was condensed with different amino acids (1.5 eq) protected by N-Boc/Cbz in the presence of DCC (1.5 eq) and DMAP (0.3 eq) to generate 17β-estradiol (17-OH)-amino acid conjugate compounds I-2 ∼ I-7 (yield: 70 % to 92 %). Followed by deprotection in saturated HCl/EtOAc solution to afford compounds I-8 ∼ I-12 (yield: 88 % to 98 %). Finally, debenzylation under HCOONH₄ (10 eq) and 10 % Pd/C (10 %, weight %) were carried out to get target compounds I-13 ∼ I-19 (yield: 40 % to 99 %).

Close

Scheme 1

Synthesis of 17β-estradiol (17-OH)-amino acid derivatives. Reagents and conditions: (a) DCC/DMAP, CH₂Cl₂, rt. (b) Saturated HCl/EtOAc solution, EtOAc, rt. (c) Ammonium formate, 10 % Pd/C, MeOH/THF, reflux.

Export to PPT

3.1.2

3.1.2 Synthesis of 17β-estradiol (16-C)-amino acid derivatives

As shown in Scheme 2, II-1 was substituted by different N-Boc protected amino acids under KI and K₂CO₃ to give compounds II-2 ∼ II-6 (yield: 29 % to 60 %). Followed by deprotection in saturated HCl/EtOAc solution to afford compounds II-7 ∼ II-11 (yield: 65 % to 99 %).

Close

Scheme 2

Synthesis of 17β-estradiol (16-C)-amino acid derivatives. Reagents and conditions: (a) K₂CO₃, KI, DMF, 60 °C. (b) Saturated HCl/EtOAc solution, EtOAc, rt.

Export to PPT

3.2 Results of in vitro antitumor activity

The IC₅₀ values of the tested compounds against tumor cells MDA-MB-231, MCF-7, SKOV-3, T24, BEL-7402 and normal cells HL-7702 were determined by MTT assay, using 5-fluorouracil (5-FU) as a positive control (Table 1).

The results showed that compounds I-8, I-9, I-10, I-11, and I-12 derived at the 17-OH position of 17β-estradiol showed good inhibitory effects (4 ∼ 20 μM) against the tested tumor cells, but also had corresponding toxicity against HL-7702 normal cells. While compounds II-2 to II-11 derived at position 16-C of 17β-estradiol showed relatively no inhibitory activity. This difference could be viewed over the structure, indicating that 17β-OH which oxidized to a carbonyl group maybe lead the inactivation. The hydrochloride form might have little effect on the antitumor activity (I-8 vs. I-9, I-10, I-11 and I-12). The 3-OH of compounds I-13 ∼ 17 was bare and the 3-OH of compounds I-8 ∼ 12 was protected by benzyl. The cytotoxicity of compounds I-13 ∼ 17 was greatly reduced compared to compounds I-8 ∼ 12. This indicated that the inhibitory activity of the compounds could be increased when the amino group was bare and the 3-OH was protected. Among them, compound I-12 (Fig. 2) (3-benzyloxy-Estra-1,3,5 (10)-triene-17β-ol pyrrolidine-2-carboxylate hydrochloride) showed the best cytotoxicity against MDA-MB-231 cells with an IC₅₀ of 4.89 µM. I-12 showed better cytotoxicity against MDA-MB-231 cells than estradiol derivatives 17β-{6-[(Pyridin-4-ylmethyl)amino]-hexyloxy}-estra-1,3,5(10)-triene-3-ol (IC₅₀ = 9.9 ± 0.4 µM) and 16S-Bromo-3-benzyloxy-estra-1,3,5(10)-triene-17-one (IC₅₀ = 5.3 ± 0.2 µM) (Yin et al., 2023). Compound I-12 was selected for further mechanistic studies in antitumor activity on MDA-MB-231 cells.

Close

Fig. 2

Structure of compound I-12.

Export to PPT

3.3 Effect on apoptosis of MDA-MB-231 cell line

Compound I-12 was selected for its expression on apoptosis-related proteins using the Proteome Analysis Tool (Human Apoptosis Array Kit). Fig. 3 showed the images at different exposure times (1–10 min). 6 min was the best exposure time, and the expression of apoptosis-related proteins was calculated based on its gray value (Fig. 4). The grayscale value is the color depth of the point in the black and white image, see Appendix A. for details of the grayscale value.

Close

Fig. 3

Expression of apoptosis-related proteins at different exposure time (1, 2, 4, 6, 8 and 10 min) after treatment of MDA-MB-231 cells with I-12 for 48 h.

Export to PPT

Close

Fig. 4

Expression of apoptosis-related proteins (Bad, Bax, Bcl-2, Bcl-x, Pro-Caspase-3, Cleaved Caspase-3, Catalase, clAP-1, clAP-2, Claspin, Clusterin, Cytochrome c, TRAIL R1/DR4, TRAIL R2/DR5, FADD, Fas/TNFRSF6/CD95, HIF-1α, HO-1/HMOX1/HSP32, HO-2/HMOX2, HSP27, HSP60, HSP70, HTRA2/0mi, Livin, PON2, p21/CIP1/CDKN1A, p27/Kip1, Phospho-p53(S15), Phospho-p53(S46), Phospho-p53(S392), Phospho-Rad17(S635), SMAC/Diablo, Survivin, TNF RI/TNFRSF1A, XIAP) after 48 h of I-12 treatment.

Export to PPT

As shown in Figs. 3 and 4, compound I-12 could obviously cause differential expression of apoptosis related proteins on tested MDA-MB-231 cells, which indicating that it could induce apoptosis. Compared with the 0 µM group, it was clearly exhibited that I-12 (5 µM) could induce the upregulation of apoptosis-related proteins Bad, Bax, Cytochrome c, TRAIL R2/DR5, Fas/TNFRSF6/CD95, HIF-1α, PON2, p21/CIP1/CDKN1A and Phospho-p53 (S46); while induce the downregulation of cIAP, HO-2/HMOX2 and HSP27. This results suggested that I-12 may induce apoptosis in MDA-MB-231 cells through a mitochondria-mediated pathway.

3.4 Network pharmacological analysis and molecular docking

(see Appendix for details) A total of 106 targets for I-12 was found in Stitch, Swiss Target Prediction, SEA, PharmMapper, and TargetNet databases (PharmMapper: zscore > 0.5, TargetNet: prob ≥ 0.5). See Appendix A for details of the 106 targets. These 106 targets were entered into DAVID for disease enrichment analysis, and breast cancer with a high score (enrichment score: 8.91) was selected as the study disease. Breast cancer genes were searched in GeneCard, DisGeNET and GEO databases, and 163 breast cancer disease genes were obtained by screening and summarizing (GeneCard: Relevance score > 70; DisGeNET: estrogen negative: Score gda ≥ 0.4, estrogen positive: Score_gda ≥ 0.1; GEO: logFC > 0.2, high expression, logFC < -0.2, low expression). Following, these 163 disease genes were used to map in KEGG pathway mapping to select pathways with higher gene enrichment: Pathways in cancer, PI3K-Akt signaling pathway, Breast cancer pathway, EGFR tyrosine kinase inhibitor resistance, MAPK signaling pathway, Metabolic pathways, JAK-STAT signaling pathway and mTOR signaling pathway. Compound targets were then intersected with breast cancer disease genes using Venn diagrams to obtain four intersecting targets: mTOR, KIT, ESR1, PDK2.

The compound targets were entered into DAVID for enrichment. It was found that I-12 may affect biological processes related to nucleoplasm, nucleus, zinc ion binding, transcription initiation from RNA polymerase II promoter, etc. The summarized targets were enriched and analyzed in Metascape. It indicated that the targets were mainly enriched in nuclear receptors, signaling by nuclear receptors, cellular response to nitrogen compounds, and steroid metabolic process. Combining the enrichment results from DAVID and Metascape, we found that compound I-12 may have an effect on nuclear receptor signaling and its related biological processes, in which the target RXRA plays an important role. At last, the pathway enrichment was carried out on KEGG pathway mapping with selected targets, mainly looking for the relevant pathways of mTOR, KIT, PDK2, ESR1 and RXRA, which were selected as the intersection targets, and selecting the pathways with relatively large number of target enrichment and interest. Targets were entered into String to see the interactions between the targets. Finally, Cytoscape software was used to constructed disease-pathway-gene network and compound-pathway-target network.

Jiameng Qu et al. investigated the mechanism of action of dandelion against triple-negative breast cancer using a network pharmacology approach and found that dandelion may act by inhibiting the cell cycle and affecting cellular metabolism (Qu et al., 2022). Yonghuan Yan et al. revealed the potential bioactivities, active components and therapeutic targets of T. aurantialba using a network pharmacology approach, and then investigated the potential binding sites of the active compounds to the key targets by molecular docking (Yan et al., 2022). Zhenyuan Yu et al. investigated the potential mechanism of action of tangeretin in oral squamous cell carcinoma by network pharmacology, and pathway enrichment analysis showed that PI3K-AKT was the most important pathway (Yu et al., 2022).

The binding mode of compound I-12 to mTOR, KIT, ESR1, PDK2, RXRA was investigated by molecular docking. The mTOR (PDB: 4DRI), KIT (PDB: 6GQM), ESR1 (PDB: 1SJ0), PDK2 (PDB: 4MPC) and RXRA (PDB: 3OAP) were downloaded from the RCSB Protein Data Bank. Docking was carried out by AutoDock software and the images were drawn by PyMOL. The molecular docking results showed that the free energy of binding of compound I-12 to mTOR was −10.91 kcal/moL, to KIT was −7.96 kcal/mol, to ESR1 was −9.56 kcal/mol, to PDK2 was −8.28 kcal/mol and to RXRA was −8.07 kcal/mol. Compound I-12 had strong binding ability to mTOR, KIT, ESR1, PDK2, RXRA. The simulated docking results (Fig. 5) showed that compound I-12 was bound to protein residues through hydrogen bonding and hydrophobic interactions, respectively. Compound I-12 interacted with mTOR residues SER-2035 and ILE-87, KIT residues SER-639 and GLU-635, ESR1 residues SER-527 and ASP-351, PDK2 residue ASP-203, and RXRA residue GLU-394 through hydrogen bonding. This might explain the higher antitumor activity of compound I-12.

Close

Fig. 5

Binding poses of compound I-12 with mTOR (A), KIT (C), ESR1 (E), PDK2 (G) and RXRA (I). Docking sites of compound I-12 with mTOR (B), KIT (D), ESR1 (F), PDK2 (H), and RXRA (J), with blue solid lines representing hydrogen bonding and gray dashed lines representing hydrophobic interactions. (For interpretation of the references to color in this figure legend, the reader is referred to the web version of this article.)

Export to PPT

In total, we found through network pharmacological analysis that compound I-12 may act on nuclear receptors and affect biological processes related to nuclear receptor signals. The potential targets were mTOR, KIT, ESR1, PDK2, RXRA. The antitumor mechanism (Fig. 6) of I-12 may be related to the pathways involved by the potential targets, such as PI3K/Akt/mTOR pathway, MAPK pathway, estrogen receptor signal pathway, etc.

Close

Fig. 6

Possible mechanism of compound I-12 against breast cancer.

Export to PPT