A halogenated pregnenolone analogue suppresses HepG2 proliferation and induces apoptosis via PPARγ regulation

Abstract

The broad biological activity of synthetic halogenated steroid derivatives has spurred considerable recent research. In this study, we synthesized and evaluated cytotoxicity against HepG2 cells for twenty-eight halogen-containing pregnenolone derivatives. Compound 20 exhibited the highest cytotoxicity, with an IC₅₀ value of 1.490 ± 0.041 μM. It effectively inhibited cell proliferation, clone formation, and migration, as well as induced DNA damage in hepatocellular carcinoma cells. Flow cytometry revealed that compound 20 induced G0/G1 cell cycle arrest. Concurrently, experimental evidence from Hoechst 33258 staining, flow cytometry, and western blotting convincingly demonstrated that compound 20 robustly promotes apoptosis in HepG2 cells, marked by decreased expression of pro-survival proteins (caspase 3/8/9, Bcl-2) and increased expression of pro-apoptotic Bax. Network pharmacology identified PPARγ as a key potential target. Western blots using a PPARγ agonist (pioglitazone) and antagonist (T0070907) validated its crucial role in mediating compound 20-induced apoptosis in HepG2 cells. Complementary molecular modeling, including docking and molecular dynamics simulations, underscored the high binding affinity and stabilizing interaction between compound 20 and PPARγ. Furthermore, a CETSA assay provided additional support, confirming that compound 20 enhances PPARγ's thermal stability. Collectively, these findings strongly suggest that compound 20 inhibits HepG2 proliferation and promotes HepG2 apoptosis via PPARγ modulation, positioning it as a promising lead compound for hepatocellular carcinoma treatment.

Halogenated pregnenolone derivative 20 exerts potent anti-HepG2 effects by inhibiting proliferation and inducing apoptosis through PPARγ modulation.

Introduction

Cancer persists as a major cause of death and a formidable challenge to enhancing global life expectancy. The worldwide burden of cancer incidence and mortality continues to rise rapidly, driven by population aging and growth, alongside evolving prevalence and distribution dynamics of crucial risk factors. Among these, liver cancer, despite being the sixth most prevalent primary cancer, was the third leading cause of cancer-related mortality in 2022, with a mortality rate of 7.8%.¹ Hepatocellular carcinoma constitutes the predominant form of primary liver cancer, representing 80–90% of cases, while cholangiocarcinoma accounts for an additional 10–15%. The predominant drivers of primary liver cancer are chronic inflammatory conditions, frequently triggered by hepatitis B (HBV) and hepatitis C (HCV) viral infections, alcoholic steatohepatitis, and nonalcoholic steatohepatitis. Although progress in vaccination and antiviral treatments has diminished the incidence of HBV- and HCV-associated liver cancer, a concurrent rise in lifestyle-related metabolic disorders is fueling ongoing chronic inflammation and hepatic tumorigenesis, a trend particularly pronounced in Western societies.² This public health crisis is exacerbated by the exceptionally low survival rates for liver cancer in China, reported at just 14.4% between 2019 and 2021.³

Transarterial chemoembolization and oral small-molecule agents such as sorafenib represent the best supportive care and remain primary therapeutic options.⁴ However, most clinical drugs for liver carcinoma are associated with high toxicity, poor response rates, adverse side effects, and a tendency to induce drug resistance. For instance, first-line sorafenib can lead to elevated serum lipase and amylase, diarrhea, hemorrhage, neuropathy, lymphopenia, nausea, and dyspnea. Adriamycin may induce myocardial degeneration and pulmonary toxicity via free radical generation,⁵ and 5-fluorouracil (5-FU) frequently encounters chemoresistance. Neither current ablation techniques nor chemotherapy significantly improve outcomes in this devastating disease. Thus, the development of more effective and safer anti-liver cancer agents is urgently needed and represents a major focus in contemporary drug research.⁶

Steroids, characterized by a fused four-ring cycloalkane structure (Fig. 1) with diverse substituents, exhibit a broad spectrum of biological activities. Steroidal derivatives are among the most clinically versatile and therapeutically diverse classes of compounds on the global market. They have long been employed in cancer treatment and continue to be integral to modern medicine.^7–9 Considerable research has explored the relationship between steroids and cancer, particularly sex steroids such as 17β-estradiol, dihydrotestosterone, and progesterone, which modulate cell proliferation, differentiation, and metabolism.¹⁰ Pregnenolone (Δ⁵-pregnene-3β-ol-20-one), a key endogenous steroid first chemically synthesized by Butenandt in 1934 and later isolated from hog testes by Ruzicka in 1943,^11,12 has served as a template for synthesizing derivatives with cytotoxic, anti-inflammatory, anti-ulcer, antimicrobial, anti-osteoporotic, and anxiolytic properties.^13–16 Notably, pregnenolone-based compounds have demonstrated cytotoxicity against various cancer cell lines, including medulloblastoma (Daoy), lung carcinoma (H460), gastric cancer (MGC-803), prostate cancer (PC-3), esophageal cancer (EC109), bladder cancer (T24), hepatocellular carcinoma (SMMC-7721/HepG2), and breast cancer (MCF-7). In particular, the introduction of alkynylaryl and heterocyclic moieties (e.g., pyrazoline, pyridine, thiophene, thiazole) at the C-21 position of pregnenolone has shown promising cytotoxic activity against HepG2 cells.^12,17–20 Therefore, the use of pregnenolone as a starting material for structural modifications confers substantial advantages in the design and discovery of anticancer steroid-based therapeutics.

Fig. 1. Synthetic strategy of halogenated steroid analogues.

Halogen-containing compounds exhibit a wide array of biological effects, encompassing antifungal, antibacterial, and anticancer activities.^21–23 Theoretical and experimental studies have shown that halogen bonding strength increases with the size and polarizability of the halogen atom (I > Br > Cl).^24,25 This interaction plays a crucial role in orienting ligands within biological targets, enhancing protein-ligand stability and improving binding affinity and selectivity. Halogens (Cl, Br, I) act as electron acceptors with strong directionality, especially in biomolecular contexts, distinguishing halogen bonding from other intermolecular forces.²⁶ Its significance in governing ligand specificity and binding underpins the high affinity observed in many inhibitors.²⁷ Thus, halogen bonding serves as a powerful tool, comparable to hydrogen bonding, for optimizing binding affinity and selectivity.

In the ongoing search for new chemotherapeutic agents, cytotoxic drugs remain central to cancer treatment due to their high target specificity.^28,29 Structural modifications of the steroid core, such as derivatization, heteroatom incorporation, and hybridization with alkylating agents, have successfully enhanced the activity of previously inactive or moderately active compounds. The addition of a steroid moiety often results in improved efficacy compared to the alkylating agent alone. Many steroidal derivatives have demonstrated potent antineoplastic activity with high tissue selectivity, underscoring the promise of steroid-based agents as safer and more target-specific anticancer therapeutics. In this study, through the modification of pregnenolone using halogen-containing intermediates (Fig. 1), we synthesized a library of 28 halogenated pregnenolone derivatives. A systematic evaluation of their anticancer activity against HepG2 cells was performed. Notably, the preliminary mechanism of action of the highly potent compound 20 was explored, which provided crucial insights into its antitumor properties.

Results and discussion

Chemistry

Starting with pregnenolone (PREG), the 3-hydroxyl group was esterified with carboxylic acid derivatives in the presence of EDCI and DMAP to obtain compounds 1–11 (Scheme 1). Separately, the reaction of pregnenolone with succinic anhydride in the presence of triethylamine yielded succinate ester intermediate 12via nucleophilic substitution. Subsequent acylation of intermediate 12 with primary or secondary amines furnished pregnenolone derivatives 13–16. The 3-hydroxyl group was then protected as the tert-butyldimethylsilyl (TBS) ether by reaction with TBSOTf and Et₃N at −78 °C. Subsequent electrophilic iodination of this intermediate at C₂₁ with N-iodosuccinimide (NIS) generated the C₂₁-iodinated derivative 17 (87% overall yield) (Scheme 2). Compound 17 then underwent nucleophilic fluorination using tetrabutylammonium difluorotriphenylsilicate (TBAT), followed by deprotection with HF·pyridine, to afford derivatives 19 and 20. Furthermore, the reaction of compound 17 with amide intermediates under NaH conditions provided derivatives 22 and 23. In a parallel synthesis, pregnenolone was converted to the C₂₁-bromomethyl-substituted derivative 21 by reaction with copper bromide in pyridine/methanol.

Scheme 1. Synthesis of pregnenolone derivatives 1–11 and 13–16. Reagents and conditions: (a): EDCI, DMAP, DCM, rt; (b) Et₃N, DCM, rt; (c) HOBt·H₂O, EDCI, pyridine, DMF, rt.

Scheme 2. Synthesis of pregnenolone derivatives 19–24 and 28–32. Reagents and conditions: (a) (i): TBSOTf, Et₃N, DCM, −78 °C; (ii): NIS, DCM, rt; (b): TBAT, MeCN, reflux; (c) HF·pyridine, DCM, 0 °C; (d) CuBr₂, pyridine, methanol, reflux; (e) p-fluorobenzamide/p-iodobenzamide, NaH, THF, 0 °C; (f) p-fluorobenzoic acid, K₂CO₃, DMF, 40 °C; (g) Ph₃PCH₃Br, t-BuOK, THF; (h) imidazole, DCM, TBSCl; (i) 9-BBN, NaOH, ethanol, H₂O₂; (j) EDCI, DMAP, DCM, rt; (K) 30: CCl₄, imidazole, PPh₃, DCM; 31: LiBr, Li₂CO₃, DMF; 32: I₂, imidazole, PPh₃, DCM.

The Wittig reaction at the C₂₀ position of a pregnenolone precursor gave intermediate 25. Protection of the 3-hydroxyl group of 25 with TBSCl and imidazole in DCM yielded the TBS ether 26. Hydroboration–oxidation of 26 using 9-BBN at 0 °C followed by oxidation with H₂O₂/NaOH yielded the (20S)-hydroxy derivative 27.³⁰ Esterification of intermediate 27 provided pregnenolone derivatives 28 and 29. Concurrently, intermediate 27 was subjected to the Appel reaction: CCl₄/PPh₃/imidazole for chloro products or a similar halogenation system for iodo-substituted products, producing 30 and 32. In parallel, bromination of intermediate 27 with LiBr/Li₂CO₃ yielded the bromo derivative 31. The C₂₁-hydroxyl group of 27 was tosylated to intermediate 33, and subsequent nucleophilic substitution with NaN₃ furnished azide 34. The azide intermediate then underwent a Cu(i)-catalyzed click reaction with terminal alkynes, followed by TBS deprotection, to construct C₂₁-substituted 1,2,3-triazole derivatives 35 and 36 (Scheme 3). All pregnenolone derivatives were identified by HR MS, ¹H NMR and ¹³C NMR (as shown in S1 and S2).

Scheme 3. Synthesis of pregnenolone derivatives 35 and 36. Reagents and conditions: (a): TsCl, pyridine, Et₃N; (b): NaN₃, DMF; (c) alkyne-R, CuSO₄·5H₂O, sodium ascorbate; (d): HF·pyridine, DCM, 0 °C.

Cell viability and SAR

To evaluate the cell viability of the 28 synthesized pregnenolone derivatives against HepG2 cells, the MTT assay was performed, with 5-fluorouracil (5-FU) as a positive control. A preliminary screening, using a concentration of 10 μM, was conducted against HepG2 cells (Fig. 2A). IC₅₀ values for active compounds (IC₅₀ < 50 μM) were then determined and are summarized in Table 1. Halogenated benzoate esters of the pregnenolone C-3 hydroxyl group (with clog P values ranging from 5.23 to 8.18) demonstrated a significant impact on HepG2 cell viability, with strong electron-withdrawing substituents on the phenyl ring inhibiting HepG2 growth. Specifically, fluoro-substituted benzoates of pregnenolone exhibited superior activity compared to their chloro and iodo counterparts (3vs.6–8). The position of the fluorine substitution on the benzoic acid ring and the number of fluorine atoms also influenced the activity (e.g., compounds 1–3). When incubated with HepG2 cells for 24 and 48 h, the ortho-fluorinated benzoate ester of pregnenolone was more potent than the para-substituted analog (1vs.3). Among benzoate ester compounds substituted with strongly electronegative groups (F and CF₃), increasing the clog P value appropriately led to a significant enhancement in compound activity (e.g., compound 4, clog P = 7.79 vs. compound 1: clog P = 7.05). Compound 4 displayed IC₅₀ values ranging from 8.969 to 16.315 μM at different time points.

Fig. 2. Effects of pregnenolone derivatives on cell viability in different cells. (A) Cell viability of HepG2 treated with 10 μM for 24 h, 48 h, and 72 h; (B–E) IC₅₀ of compound 20 on HepG2 (B), Huh7 (C), SK-Hep-1 (D), and THLE-2 (E) cells for 24 h and 48 h.

Cell viability of pregnenolone derivatives in HepG2.

No.	Structure	clog  P	IC50 (μM)
			24 h	48 h	72 h
RO–C3
1		7.05	>50	>50	20.237 ± 0.344
2		7.19	>50	20.230 ± 0.110	29.490 ± 0.629
3		6.61	43.497 ± 0.991	14.800 ± 0.156	32.537 ± 0.513
4		7.79	16.315 ± 0.532	14.213 ± 0.031	8.969 ± 0.163
5		7.49	>50	>50	>50
6		7.61	>50	>50	>50
7		7.76	>50	>50	>50
8		8.02	>50	>50	>50
9		5.53	>50	>50	>50
10		5.23	32.883 ± 1.017	25.827 ± 0.228	20.670 ± 0.335
11		5.33	>50	>50	>50
13		6.61	>50	22.600 ± 0.286	19.360 ± 0.600
14		6.77	40.317 ± 2.720	17.483 ± 0.152	6.265 ± 0.279
15		6.37	>50	43.157 ± 0.279	35.291 ± 0.290
16		6.11	>50	>50	>50

R–C21
19	–F	3.7	>50	>50	>50
20	–I	4.67	2.095 ± 0.079	1.490 ± 0.041	7.302 ± 0.155
21	–Br	4.28	3.449 ± 0.011	15.240 ± 0.560	>50
22		5.18	>50	>50	>50
23		6.07	>50	>50	>50
24		5.83	>50	>50	>50

R–C17
28		8.18	>50	>50	>50
29		4.83	>50	>50	>50
30		6.72	>50	10.320 ± 0.145	14.913 ± 0.461
31		6.86	44.060 ± 0.609	>50	>50
32		7.25	>50	>50	>50
35		5.90	>50	10.310 ± 0.157	>50
36		7.53	>50	21.293 ± 0.140	19.147 ± 0.180
PREG			37.310 ± 2.130	>50	>50
5-FU			47.453 ± 0.499	>50	47.280 ± 0.423

Lowering the clog P value, such as with compounds 9–11 (clog P ≈ 5), revealed better activity for the pregnenolone bromoacetate ester (compound 10: IC₅₀ values between 20.670 and 32.883 at different time points), although its activity was reduced compared to compound 4. Modifications to the succinate linker structure were made to alter liposolubility (achieving a clog P value of approximately 6, as seen in compounds 13–16). These modifications resulted in compounds 13 and 14 significantly enhancing cell viability after 48 and 72 h of incubation with HepG2 cells. In particular, compound 14 exhibited an IC₅₀ of 6.265 ± 0.279 μM at 72 h. Further modifications at the C₂₁ position of pregnenolone to introduce electronegative groups (e.g., F, Br, or I) and adjust clog P values (≈4) revealed that compound 20 significantly decreased HepG2 viability at 24, 48, and 72 h, with an IC₅₀ of 1.490 ± 0.041 μM at 48 h, outperforming the pregnenolone and the 5-FU positive control. Finally, derivatization of the pregnenolone carbonyl group yielded halogenated compounds 28–32, 35 and 36 (with clog P values of 4.83–8.18). Compounds 30 and 35 demonstrated significant reduction in the viability of HepG2 cells after 48 hours of incubation, with IC₅₀ values of 10.320 ± 0.145 μM and 10.310 ± 0.157 μM, respectively. As shown above, in vitro cytotoxicity assays demonstrated that compound 20 displayed its most potent antitumor bioactivity following 48 h of treatment (Fig. 2B and Table 1). Finally, we further assessed the viability of 20 on Huh7 and SK-Hep-1 cell lines. As shown in Fig. 2C and D, compound 20 also exhibited significant cytotoxic activity against these cell lines. In parallel, the cytotoxicity of compound 20 towards normal liver cells (THLE-2) was recorded as 5.388 ± 0.254 μM after 48 h of treatment (Fig. 2E), with a selectivity index of 3.62.

Proliferation and migration effects of compound 20 on HepG2

HepG2, SK-Hep-1, and Huh7 cells were treated with different concentrations of compound 20 (0.5, 1, and 2 μM) for 48 h, and their proliferative capacity was assessed using the EdU incorporation assay.³¹ The results demonstrated that compound 20 significantly reduced cell proliferation across all treatment groups compared to the control. In HepG2 cells, EdU-positive cell percentages decreased from 58.24% (control) to 30.40% (0.5 μM), 7.24% (1.0 μM), and 4.85% (2 μM) (Fig. 3). Huh7 cells showed a similar reduction, with a percentage decrease ranging from 17.33% to 34.49% (Fig. 4). SK-Hep-1 cells also exhibited a dose-dependent inhibition, dropping from 7.02% to 39.05% (Fig. 5). These findings collectively indicate that compound 20 dose-dependently inhibits hepatocellular carcinoma cell proliferation. The colony formation assay can be used to evaluate the proliferative capacity of tumor cells and their ability to form colonies from single cell, thereby reflecting their tumorigenic potential.³² HepG2 cells were further exposed to varying concentrations (0.05, 0.1, 0.2, 0.4, 0.8 μM) of compound 20, with the drug-containing medium replaced every two days. After 14 days, colony formation was examined. As illustrated in Fig. 6A, the clonogenic ability decreased with rising concentration of 20, and was almost completely abolished at 0.8 μM.

Fig. 3. Effect of compound 20 on HepG2 cell proliferation at 48 h (10×). ****P < 0.0001 vs. control.

Fig. 4. Effect of compound 20 on Huh7 cell proliferation at 48 h (10×). ***P < 0.001, ****P < 0.0001 vs. control.

Fig. 5. Effect of compound 20 on SK-Hep-1 cell proliferation at 48 h (10×). **P < 0.01, ****P < 0.0001 vs. control.

Fig. 6. Proliferation and migration effects of compound 20 on HepG2. (A) Treatment of HepG2 cells with 20 (0.05, 0.1, 0.2, 0.4, 0.8 μM) for 14 days followed by staining with 0.1% crystal violet; (B) cell migration activity of compound 20 on HepG2. ****P < 0.0001 vs. control.

The migration of tumor cells is critical for tumor progression and represents a key characteristic of metastatic behavior. Therefore, inhibiting cell migration is essential for effective cancer therapy.³³ A wound healing assay was conducted to evaluate the effect of 20 on the migratory capacity of HepG2 cells. As depicted in Fig. 6B, compound 20 demonstrated a significant dose- and time-dependent inhibition of HepG2 cells wound healing capacity. In the control group, the wound gradually closed with a migration rate up to 46.31 ± 2.06% for 72 h, indicating substantial migratory ability of the cancer cells. In contrast, after 24 h of treatment, the migration rates in the experiment groups were 16.41 ± 2.13%, 7.44 ± 1.15%, and 5.78 ± 3.15% at 0.5, 1, and 2 μM, respectively (control: 17.40 ± 1.09%). After 48 h, the HepG2 cell migration rate in the control increased to 30.32 ± 1.44%, while the migration rates were 17.31 ± 2.46%, 11.17 ± 0.34%, and 6.73 ± 1.21% on treatment with 20 at 0.5, 1, and 2 μM, respectively. After 72 h of treatment, compared with the control group, the migration rates were 12.65 ± 1.58%, 15.24 ± 1.80%, and 34.16 ± 2.84% at given gradient concentrations.

DNA damage and cell cycle arrest by compound 20

DNA damage is a common event in cancer therapy.³⁴ To investigate whether compound 20 induces DNA damage in hepatocellular carcinoma cells (HepG2, Huh7, and SK-Hep-1), immunofluorescence staining for the DNA-damage marker γ-H2AX was performed. The results showed that treatment with compound 20 (0.5, 1, and 2 μM) for 48 h increased both the number of γ-H2AX foci and their fluorescence intensity in all three cell lines (Fig. 7–9). Among them, HepG2 cells exhibited the strongest γ-H2AX foci formation and highest fluorescence intensity compared with Huh7 and SK-Hep-1 cells. Abnormal cell cycle regulation is an important mechanism of tumorigenesis.³⁵ In this study, the cell cycle distribution of HepG2 cells was analyzed by flow cytometry. As presented in Fig. 10, with the increasing concentration of compound 20 (0.5, 1, and 2 μM) for 48 h, the results demonstrated that the proportion of HepG2 cells in the G0/G1 phase was dramatically increased from 52.78% in the control to 70.30% in the treatment group with 2 μM, associated with a decrease in the fraction of cells in the S and G2/M phases. These data demonstrate that compound 20 induces cell cycle arrest at the G0/G1 phase in HepG2 cells.

Fig. 7. Effects of 20 on γ-H2AX protein expression in HepG2 cells (100×).

Fig. 8. Effects of 20 on γ-H2AX protein expression in Huh7 cells (100×).

Fig. 9. Effects of 20 on γ-H2AX protein expression in SK-Hep-1 cells (100×).

Fig. 10. Cell cycle analysis. (A) Flow cytometry results depicting the cell cycle phases in HepG2 cells, labeled as G0/G1, S, and M; (B) flow cytometry histograms of the cell cycle in HepG2 cells. ***P < 0.001, ****P < 0.0001 vs. control.

Apoptosis effect on HepG2 induced by compound 20

The targeted induction of apoptosis in malignant cells is considered a front-running anti-cancer strategy, designed to eliminate tumors while sparing healthy cells from collateral damage.³⁶ Cell cycle analysis results revealed that compound 20 prevented cell proliferation by arresting cells in the G0/G1 phase. To determine if the antihepatoma activity of 20 was associated with apoptosis, HepG2 cells were examined using Hoechst 33258 staining. As illustrated in Fig. 11A, consistent with increasing concentrations, the incidence of cells showing intense blue fluorescence was significantly higher than that in the control group, signifying a substantial increase in apoptotic cells. Furthermore, apoptosis was quantitatively evaluated using Annexin V-FITC and propidium iodide (PI) double staining; the percentages of the total apoptotic and necrotic cells increased from 4.90% (Control) to 15.44% (0.5 μM), 70.28% (1.0 μM), and 97.42% (2 μM) (Fig. 11B). The apoptosis-related proteins were subsequently quantified by western blot analysis. The expression levels of caspase 3/8/9 exhibited a dose-dependent reduction relative to the control (Fig. 11C). Meanwhile, treatment with compound 20 led to a reduction in Bcl-2 expression concomitant with an elevation in the Bax/Bcl-2 ratio. Collectively, these results demonstrate that compound 20 suppresses the proliferation of HepG2 cells by promoting cell apoptosis.

Fig. 11. Apoptosis effect on HepG2 treated with compound 20. (A) The images of the apoptosis of HepG2 cells demonstrated by Hoechst 33258 staining; (B) Annexin V-FITC/PI staining assay for apoptosis in HepG2 cells; (C) analysis of apoptosis-related proteins (caspase-3, -8, -9, Bcl-2, Bax) by western blot in HepG2 cells. *P < 0.05, **P < 0.01, ***P < 0.001, ****P < 0.0001 vs. control. ns: no significance.

PPARG as a potential target gene for compound 20

Network pharmacology analysis was applied to elucidate potential anticancer targets of compound 20. This analysis identified 101 targets associated with compound 20 and 1069 targets relevant to hepatocellular carcinoma. Intersection analysis revealed 69 putative anti-hepatocellular carcinoma targets of 20 (Fig. 12A). These potential therapeutic targets were imported into the STRING database to obtain protein–protein interaction information (Fig. 12B), followed by network reconstruction and analysis using Cytoscape 3.10.3. Among them, 15 targets exhibited an interaction degree ≥ 28, including key genes such as TNF, PPARG, HMGCR, and ESR1 (Fig. 12C). To elucidate the biological functions of compound 20-related targets, DAVID database analysis of the 69 predicted targets using Gene Ontology (GO) and Kyoto Encyclopedia of Genes and Genomes (KEGG) identified 194 biological processes (BPs), 29 cellular components (CCs), and 70 molecular functions (MFs). The top ten significantly enriched terms from each category were visualized (Fig. 12D). The dominant BP terms were steroid metabolic process, inflammatory response regulation, and protein transport regulation; CC terms included membrane rafts, microdomains, and caveolae; and MF terms featured steroid binding, nuclear receptor activity, and ligand-modulated transcription factor activity. KEGG analysis revealed 52 enriched pathways, with the top 15 signaling pathways shown in Fig. 12E. These included metabolic pathways, pathways in cancer, and the PPAR signaling pathway. These findings suggest that modulation of metabolic pathways and the PPAR signaling pathway may represent key mechanisms underlying the anti-hepatocellular carcinoma effects of compound 20.

Fig. 12. Network pharmacology analysis of compound 20. (A) Venn diagram of potential therapeutic targets of compound 20 against hepatocellular carcinoma; (B) PPI network of overlapping targets between compound 20 and liver cancer; (C) core target interaction network analysis; (D) GO enrichment analysis; (E) KEGG enrichment analysis.

Promoting HepG2 apoptosis by compound 20via regulation of PPARγ

PPARγ, encoded by the PPARG gene, is a ligand-activated nuclear receptor critically involved in cell differentiation, immune function, and biological metabolism. Recent studies have highlighted its close association with tumor biology, demonstrating that PPARγ affects the proliferation and apoptosis of several malignant cell types, including liposarcoma, prostate cancer, breast adenocarcinoma, colorectal cancer, and lung cancer.^37–40 To investigate whether PPARγ mediates compound 20-induced apoptosis in HepG2 cells, we first analyzed the expression and phosphorylation levels of PPARγ after treatment with compound 20. The results showed that both PPARγ expression and its phosphorylation levels were effectively suppressed (Fig. 13A). Additionally, immunofluorescence assays demonstrated a marked decrease in nuclear PPARγ localization in HepG2 cells treated with 20 (2 μM) compared to the control group (Fig. 13B). The reduced green fluorescence intensity in the nuclei of treated cells strongly suggests that compound 20 interferes with PPARγ expression.

Fig. 13. Effects of compound 20 on HepG2 apoptosis via PPARγ regulation. (A) Western blot analysis of PPARγ expression in HepG2 cells treated with compound 20; (B) PPARγ levels in compound 20-treated HepG2 cells detected by immunocytochemical staining (60×); (C and D) apoptosis-related protein expression in HepG2 cells treated with pioglitazone, T0070907, and compound 20, individually or in combination. *P < 0.05, **P < 0.01, ***P < 0.001, ****P < 0.0001 vs. control. ^#P < 0.05, ^##P < 0.01, ^###P < 0.001, ^####P < 0.0001, compound 20vs. compound 20 with pioglitazone or T0070907.

To further assess the role of PPARγ, HepG2 cells were treated with pioglitazone (a PPARγ agonist, 50 μM), T0070907 (a PPARγ inhibitor, 50 μM), and compound 20 (2 μM), administered individually or in combination. Western blot analysis demonstrated that pioglitazone upregulated the expression of both PPARγ and its phosphorylated form (Fig. 13C), while increasing caspase 3/8/9 and reducing Bax and the Bax/Bcl-2 ratio (Fig. 13D). Conversely, when HepG2 cells were co-treated with compound 20 and pioglitazone, these effects were reversed. Exposure to the inhibitor T0070907 alone markedly downregulated both PPARγ and p-PPARγ levels in HepG2 cells. Consistent with T0070907, compound 20 also downregulated caspase-3/8/9 and Bcl-2, while upregulating Bax and the Bax/Bcl-2 ratio. Notably, the combination of compound 20 and T0070907 resulted in a significantly stronger pro-apoptotic effect and a more pronounced downregulation of PPARγ and its phosphorylation compared to either treatment alone. These findings collectively validate that compound 20-induced apoptosis in HepG2 cells is mechanistically linked to the suppression of PPARγ.

Molecular modeling analysis of PPARγ and compound 20

To further clarify the relationship between compound 20 and PPARγ, molecular docking was performed by using the AutoDock Vina 1.2.5. The protein structure (PDB ID: 9F7X) was obtained from the Protein Data Bank (https://www.rcsb.org/). Molecular docking simulations indicated that compound 20 binds to PPARγ via hydrogen bonds with Gln-271 and Ser-342 (Fig. 14A and B). Additionally, the iodine atom of compound 20 forms a halogen bond interaction with Ser-342. The calculated binding free energy of −8.455 kcal mol⁻¹ indicates a high binding affinity between compound 20 and PPARγ.

Fig. 14. (A and B) 3D and 2D visualizations of compound 20 interacting with PPARγ from molecular docking; (C) analysis of interactions and probabilities during protein–small molecule complex dynamics; (D) RMSD curves of the complex; (E) RMSF curve of the complex; (F) residue decomposition analysis of the binding free energy for the protein–small molecule complex; (G) hydrogen bond profiles of the protein–small molecule complex, detailing the probability of amino acid hydrogen bond formation with the small molecule; (H) thermal stabilization of compound 20 and PPARγ across a temperature gradient.

To comprehensively evaluate the binding stability and interaction characteristics of the protein–small molecule complexes, 100 ns all-atom molecular dynamics simulations were performed. The specific interaction patterns observed during the 100 ns MD simulation are shown in Fig. 14C. Key dynamic parameters, RMSD, RMSF, and hydrogen bond count were monitored to systematically assess the structural stability, flexibility, and interaction strength of each complex with the receptor. As shown in Fig. 14D, the protein RMSD rapidly reached 2.1 Å, and the small molecule RMSD reached 1.8 Å within the first 5 ns. Both then stabilized, indicating rapid attainment of a conformationally stable state for the protein–small molecule complexes. The RMSF of most amino acids remained below 0.5 Å (Fig. 14E), suggesting overall protein domain stability. Notably, amino acids involved in direct binding interactions exhibited increased fluctuations, potentially highlighting their functional importance. Fig. 14F depicts the transient intermolecular hydrogen bonds during the 100 ns simulation, with counts consistently maintained between 1 and 6. Major contributors to hydrogen bonding were Arg-280, Lys-263, and Leu-270. Water bridges primarily involved Arg-280, Arg-288, and Ser-342, while hydrophobic interactions were driven by Ile-341, Leu-330, and Ile-281 (Fig. 14G). These results underscore the potential of these complexes for further mechanistic investigations and drug development.

To validate the predicted binding affinity between compound 20 and PPARγ from molecular modeling, a cellular thermal shift assay (CETSA) was employed. Fig. 14H shows that compound 20 significantly confers thermal stability to PPARγ. This stabilization was observed in compound 20-treated cells compared to DMSO-treated controls across temperatures from 42 °C to 62 °C. Notably, at 62 °C, PPARγ degradation was drastically reduced in compound 20-treated cells (27.43 ± 8.70%) versus the untreated group (97.04 ± 1.27%). This enhancement of PPARγ thermal stability suggests that compound 20 likely promotes HepG2 cells apoptosis by modulating PPARγ.

Conclusions

In summary, twenty-eight halogenated pregnenolone derivatives were designed, synthesized, and evaluated for their cytotoxicity against HepG2 cells. Among these, compound 20 exhibited the most potent activity (IC₅₀ = 1.490 ± 0.041 μM), with favorable selectivity over normal hepatic cells (THLE-2, IC₅₀ = 5.388 ± 0.254 μM). The initial mechanism of compound 20 against HepG2 cells involves reducing cell proliferation and promoting HepG2 apoptosis through PPARγ modulation.

Experimental

Instruments and materials

ESI-HRMS was conducted using a Waters Xevo Q-TOF mass spectrometer system (Waters Corporation, USA). Melting points were measured using Yanaco WRX-4 micro melting point apparatus (Shanghai YiCe Apparatus & Equipment Co., Ltd., CN). A Thermo Scientific Ultimate 3000 type HPLC (Thermo Fisher Scientific, USA) was employed to confirm the purity of all synthesized compounds. NMR spectra were recorded on a Bruker AVANCE NEO 600 spectrometer (Bruker Switzerland AG, Switzerland), with tetramethylsilane as the internal standard, for both 1H and 13C analyses. Reaction progress was visualized by TLC using silica gel GF₂₅₄ plates (Qingdaohaiyang, China) under UV light (254 nm). Commercially available reagents and chemicals (Aladdin and Macklin) were used throughout the study. Solvents were of HPLC or analytical grade.

Synthesis of compounds 1–12

The corresponding halogenated acid (for compounds 1–11, 0.5 mmol, 5.0 eq.) was introduced into a solution of pregnenolone (0.1 mmol, 1.0 eq.), EDCI (0.4 mmol, 4.0 eq.), and DMAP (0.3 mmol, 3.0 eq.) in CH₂Cl₂ (1 mL). For compound 12: pregnenolone (0.63 mmol, 1.0 eq.) was dissolved in anhydrous CH₂Cl₂ (2 mL), then Et₃N (0.7 mL) and succinic anhydride (0.95 mmol, 1.5 eq.) were added at rt. The formed mixture was stirred at rt for 8–48 h. The reaction mixture was quenched with saturated NH₄Cl, extracted with CH₂Cl₂, washed with brine, and purified via silica gel column chromatography (CC) (eluted by petroleum ether/EtOAc, with ratios ranging from 40 : 1 to 1 : 1, PE/EtOAc = 40 : 1 → 1 : 1) to yield pure compounds 1–12.

Synthesis of compounds 13–16

EDCI (0.26 mmol, 2.2 eq.), HOBt·H₂O (0.17 mmol, 1.5 eq.), and pyridine (0.35 mmol, 3.0 eq.) were dispensed into a solution of compound 12 (0.12 mmol, 1.0 eq.) in DMF at rt under a N₂ atmosphere. After 30 min, a solution of RNH (0.14 mmol, 1.2 eq.) in DMF was added dropwise. The mixture was subsequently stirred for 15 hours at rt. Upon completion of the reaction, saturated NH₄Cl was added to quench it, then extracted with EtOAc. The organic layers were rinsed with brine and dried over Na₂SO₄. The purification of the crude product was achieved by silica gel CC (PE/EtOAc = 3 : 1), yielding compounds 13–16.

Synthesis of compounds 19–24

2 mL of anhydrous CH₂Cl₂, pregnenolone (0.16 mmol, 1.0 eq.), and Et₃N (0.81 mmol, 5.1 eq.) were introduced into a dry 10 mL round-bottom flask under an argon atmosphere. Upon cooling the reaction mixture to −78 °C, TBSOTf (0.47 mmol, 3.0 eq.) was added dropwise over 2 min. The mixture was maintained under stirring at this temperature for 1 h, and subsequently allowed to warm to rt. After purification, the resulting intermediate (1.0 eq.) was dissolved in 2 mL of CH₂Cl₂, and NIS (0.2 mmol, 1.2 eq.) was added in one portion. The reaction system was agitated at rt for 2 h, then washed with a 10 wt% Na₂S₂O₃ aqueous solution (10 mL), extracted twice with CH₂Cl₂ (5 mL), and purified by silica gel CC (PE/EtOAc = 150 : 1), yielding compound 17. For the next step, compound 17 (0.09 mmol, 1.0 eq.) and TBAT (0.13 mmol, 1.5 eq.) were added in anhydrous MeCN (2 mL), stirred and refluxed for 19 h. Following cooling to rt, the mixture was subjected to silica gel CC (PE/EtOAc = 150 : 1) to afford compound 18. Subsequently, the deprotection of protecting groups at the OH-C₃ was performed using Olah's reagent,⁴¹ leading to compounds 19 and 20.

In a separate reaction, halobenzamide (0.06 mmol, 1.0 eq.) and compound 17 (0.06 mmol, 1.0 eq.) were combined in THF (2 mL). The mixture was then treated with NaH (0.05 mmol, 0.8 eq.) and stirred at 0 °C for 1 h. Following extraction, deprotection, and purification by silica gel CC (PE/EtOAc = 10 : 1 → 2 : 1), compounds 22 and 23 were obtained. Additionally, compound 17 (0.06 mmol, 1.0 eq.), p-fluorobenzoic acid (0.04 mmol, 0.7 eq.), and anhydrous K₂CO₃ (0.02 mmol, 0.3 eq.) were added in DMF (2 mL), and the mixture was stirred at 40 °C for 4 h, then extracted, deprotected and isolated by silica gel CC (PE/EtOAc = 10 : 1 → 2 : 1) to yield compound 24. For the conversion of pregnenolone to compound 21, pregnenolone (0.16 mmol, 1.0 eq.) was dissolved in 3 mL of MeOH. After adding pyridine (1.5 eq., 0.24 mmol), the solution was refluxed for 10 min. Then, CuBr₂ (3.0 eq., 0.48 mmol) was added, and the reaction mixture was monitored by TLC before purification via silica gel CC (PE/EtOAc = 1 : 1) to yield compound 21.

Synthesis of compounds 28–32

To a mixture of Ph₃PCH₃Br (3.2 mmol, 2.0 eq.) in THF (2 mL) at rt under nitrogen, t-BuOK (3.2 mmol, 2.0 eq.) was introduced. The reaction mixture was then stirred at 50 °C for 30 min, followed by the addition of pregnenolone (1.6 mmol, 1.0 eq.). The reaction was continued by stirring for 1 h at 50 °C. Finally, the mixture was extracted and isolated by silica gel CC (PE/EtOAc = 10 : 1) to give compound 25. Compound 25 (1.1 mmol, 1.0 eq.), along with imidazole (2.2 mmol, 2.0 eq.) and TBSCl (1.3 mmol, 1.2 eq.), was dissolved in anhydrous CH₂Cl₂ (2.5 mL). The resulting solution was stirred at 0 °C under a N₂ atmosphere to protect the OH-C₃ group. Following this, the mixture was stirred at rt for 5 h to yield 26. Intermediate 27 was afforded by following the reported method.³⁰ Compounds 28 and 29 were prepared according to the esterification procedure.

To a solution of intermediate 27 (1.0 eq.) in CCl₄ (0.5 mL), PPh₃ (0.09 mmol, 2.0 eq.) was added. The reaction mixture was then heated to 63 °C for 8 h. The crude product was purified by silica gel CC (PE/EtOAc = 100 : 1) followed by deprotection to afford compound 30. LiBr (0.1 mmol, 2.2 eq.) and Li₂CO₃ (0.03 mmol, 0.6 eq.) were added to a solution of 27 (0.05 mmol, 1.0 eq.) in DMF (1 mL) under a N₂ atmosphere. The reaction mixture was stirred for 2 h at 82 °C, then purified by CC (PE/EtOAc = 60 : 1) followed by deprotection to afford compound 31. For the synthesis of compound 32 from 27, imidazole (1.02 mmol, 9.3 eq.) and I₂ (0.62 mmol, 5.6 eq.) were added to a solution of PPh₃ (0.62 mmol, 5.6 eq.) in DCM (1 mL) at 0 °C and stirred for 0.5 h. Next, compound 27 (0.11 mmol, 1.0 eq.) was added. The reaction mixture was continuously stirred for 1 h at rt. Subsequently, the crude product was purified by silica gel CC (PE/EtOAc = 50 : 1) followed by deprotection to afford compound 32.

Synthesis of compounds 35 and 36

Intermediate 27 (0.11 mmol, 1.0 eq.) was dissolved in CH₃Cl (1 mL) and pyridine (2 mL) at rt, followed by the addition of TsCl (0.33 mmol, 3.0 eq.). After stirring for 2 h, purification by silica gel CC (PE : EtOAc = 50 : 1) gave intermediate 33. Subsequently, 33 (0.08 mmol, 1.0 eq.), NaN₃ (0.25 mmol, 3.0 eq.), and DMF (1 mL) were stirred at rt for 6 h to give compound 34. The reaction was then continued by adding alkyne-RX (0.315 mmol, 1.5 eq.) in a MeOH/DCM/H₂O (v/v/v = 1 : 1 : 1) mixture (1.5 mL). Finally copper(ii) sulfate pentahydrate (0.021 mmol, 0.1 eq.) and sodium ascorbate (0.042 mmol, 0.2 eq.) were added, then stirred at rt for 12 h. The reaction mixture was purified by silica gel CC (PE/EtOAc = 1 : 1) followed by deprotection to afford compounds 35 and 36.

Cell culture

HepG2 and Huh7 cell lines were purchased from ATCC (Manassas, VA). SK-Hep-1 cells were procured from Wuhan Pricella Biotechnology Co., Ltd (Wuhan, China). The human normal hepatocyte THLE-2 was purchased from Zhejiang Noble Bio (Zhejiang, China). Cells were maintained in incubators at 37 °C with 5% CO₂, using DMEM medium that contained 10% fetal bovine serum (NOBLEEBIO, China) and 1% penicillin/streptomycin.

Cell viability

The effects of pregnenolone derivatives on hepatocellular carcinoma cell viability were assessed via the MTT assay.⁴² A cell density of 3 × 10⁴ cells per well was used for culturing in 96-well plates. After the initial 24 hour incubation period, the culture medium was adjusted with pregnenolone derivatives at concentrations from 2.5–80 μM (1% final concentration). Cells were then incubated with these compounds for 24–72 h. For the final step, a 1.5 hour incubation at 37 °C was conducted with the addition of 20 μL of MTT solution (5 mg mL⁻¹) per vial. Absorbance was subsequently recorded at 490 nm using a Synergy H1 microplate spectrophotometer (BioTek, USA).

Colony formation assay

HepG2 cells in 6-well plates were treated with compound 20 at concentrations ranging from 0.05 to 0.8 μM, with the culture medium replenished every 48 h. After a 14-day incubation, the culture medium was carefully aspirated. The cells were then rinsed in cold PBS, after which they were treated with 4% paraformaldehyde for 15 min to achieve fixation. Finally, a 15 min incubation with 0.1% crystal violet solution was performed for staining. Excess stain was subsequently rinsed off, and the plates were air-dried. Colony formation was then visually assessed and quantitatively analyzed using ImageJ software.

Wound healing assay

HepG2 cells were seeded into 6-well plates containing DMEM supplemented with 10% FBS. Once the cells reached full confluence, a linear scratch was introduced in each well with a pipette tip (200 μL type). The wells were then carefully rinsed twice with PBS to eliminate any loose cells, after which the cells were maintained in DMEM with 1% FBS. Treatment was applied by adding compound 20 at concentrations of 0.5, 1, and 2 μM. At designated time points (0, 24, 48, and 72 h), cell migration was monitored by capturing images with an inverted microscope (Nikon, Japan), and the extent of the wound closure was quantified by measuring the area of the scratch.

Cell cycle analysis

Following established protocols,⁴³ HepG2 cells (5 × 10⁴) were plated in 6-well plates and incubated for 24 h; cells received treatment with compound 20 at concentrations of 0.5, 1, and 2 μM. This was followed by a further 48 hour incubation period. All cells were then collected after two washes with PBS. To fix the cells, they were incubated overnight at 4 °C in 70% ethanol. Post-fixation, cells were washed with PBS to the ethanol and then resuspended in 500 μL of a fresh PI/RNase A (Beyotime,China) solution (prepared by combining RNase A and propidium iodide in a 1 : 9 volume ratio). The cell suspensions were incubated in the dark at rt for 30 min. Finally, the distribution of cell cycle phases was analyzed using a CytoFLEX flow cytometer (Beckman, USA).

Flow cytometric analysis of apoptosis

After 48 h of treatment, cells were collected and washed with cold PBS. 500 μL of binding buffer was used to resuspend the cell pellet. Following this, the cell suspension was exposed to 5 μL of Annexin V-FITC/APC (Keygel, China) and gently mixed. A subsequent addition of 5 μL of PI was made, and the prepared samples were incubated in dim light at rt for 15 min. Analysis of the stained cells was performed on a CytoFLEX flow cytometer (Beckman).

Hoechst 33258 staining

A seeding density of 4 × 10⁴ cells per well was used for the 6-well plates. Subsequently, the cells were treated with compound 20 at concentrations of 0.5, 1, or 2 μM, followed by incubation for 48 h. Following the initial incubation, cells were incubated with Hoechst 33258 (Solarbio, China) to a final concentration of 10 μg mL⁻¹. This staining was performed at 37 °C in the dark for 15 min. After rinsing twice with PBS, images of the cells were collected using a Nikon fluorescence microscope (Tokyo, Japan).

5-Ethynyl-2′-deoxyuridine (EdU) labelling

Cell proliferation was determined using an EdU Apollo DNA in vitro kit (Abbikine, China). HepG2, Huh7, and SK-Hep-1 cells (3 × 10⁴ cells per well) were cultured in 96-well plates for 24 h before treatment with compound 20 (0.5–2 μM) for 48 h. Cells were then exposed to 10 μM EdU and incubated for 3 h then fixed with 4% paraformaldehyde for 15 min. After washing with 1× BSA wash buffer, cells were permeabilized with 0.5% Triton X-100 in PBS for 15 min. Following incubation with 1× Click-iT staining reaction solution for 30 min in the dark, nuclei were counterstained with 1× Hoechst 33342 for 10 min. Fluorescence images were recorded with a Nikon fluorescence microscope imaging system, and analyzed with ImageJ software.

Immunofluorescence assay

Cells were seeded on coverslips in 24-well plates and treated with compound 20 after 24 h. After 48 h, the coverslips were washed thrice with PBS, fixed with 4% paraformaldehyde for 30 min, and washed three more times with PBS. Permeabilization was performed in PBS with 0.5% Triton X-100 (30 min) followed by PBS washes. After blocking with PBST containing 5% BSA at 37 °C (2 h), coverslips were incubated overnight at 4 °C with primary antibodies against γ-H2AX (Ser 139) (HUABIO, China) and PPARγ (1 : 200, 20 μL) (Proteintech, USA) in a light-protected humidified chamber. Following PBST washes, coverslips were incubated with fluorescent secondary antibody (1 : 100, 20 μL) (Proteintech, USA) in the dark (60 min), followed by PBST washes. Nuclei were counterstained with 20 μL DAPI (10 μg mL⁻¹) (Solarbio, China) in the dark for 10 min, washed with PBST, and affixed to glass slides using an antifade medium.⁴⁴ The data was collected through the operation of a laser scanning confocal microscope (Nikon, Tokyo, Japan).

CETSA assay

The CETSA assay was performed as previously described.⁴⁵ HepG2 cells were treated with DMSO or 2 μM compound 20 for 3 h, then collected, resuspended in PBS with protease inhibitors, and aliquoted into PCR tubes (80 μL each). The samples were subjected to a 3 minute heat treatment across 34–66 °C, then cooled at room temperature for 3 min. Cell lysis was achieved by three freeze–thaw cycles performed in liquid nitrogen. The resulting lysates were then clarified by centrifugation at 20 000 × g for 40 min at 4 °C. Proteins were analyzed by western blot following SDS-PAGE.

Western blot analysis

Cellular contents were liberated using RIPA buffer (Solarbio, China), fortified with 1 mM each of protease and phosphatase inhibitors. Protein concentrations were determined via a BCA assay (Solarbio, China). Subsequently, proteins were resolved on SDS-PAGE gels (10–12%) and transferred onto PVDF membranes. After a 4 °C incubation with primary antibodies against caspase 3 (Proteintech, USA), caspase 8 (HUABIO, China), caspase 9 (Proteintech, USA), Bcl-2 (Bioss, China), Bax (Zenbio, China), PPARγ (Proteintech, USA), p-PPARγ (Ser112) (Affinity, USA) and GAPDH (Proteintech, USA) overnight, the membranes underwent a trio of rinses with TBST. Next, a 2 hour incubation at room temperature with secondary antibodies ensued. The protein bands were visualized utilizing an Odyssey® CLx system (Bio-Rad, USA).

Network pharmacology analysis

In this research, a systems pharmacology methodology was adopted to explore the prospective anti-liver cancer targets associated with compound 20. Molecular structures of active constituents were formatted in SDF and uploaded to the SwissTargetPrediction platform (https://www.swisstargetprediction.ch).⁴⁴ Concurrently, targets linked to liver cancer were retrieved from the GeneCards database (https://www.genecards.org/) and UniProt (https://www.uniprot.org/). The intersecting set of targets was identified to pinpoint potential therapeutic candidates. PPI networks were constructed using the STRING database (https://string-db.org) and subsequently visualized with Cytoscape version 3.10.2. Further functional insights were gained through enrichment analyses, including Gene Ontology (GO) and KEGG pathway analysis, performed on key targets with the help of the clusterProfiler package (version 4.2.1; Bioconductor-clusterProfiler).

Molecular docking and molecular dynamic analysis

To evaluate the binding of compound 20 to PPARγ, the crystal structure (PDB ID: 9F7X) was retrieved from the Protein Data Bank (https://www.rcsb.org/). Compound 20 and PPARγ were then imported into AutoDock Vina v1.2.5, with a 25 × 25 × 25 Å grid box centered on the co-crystallized ligand. Water molecules, hydrogens, and charges were reset. Finally, PyMOL was used for visualization.

Subsequently, a 100 ns all-atom molecular dynamics (MD) simulation of the protein–small molecule complex was performed using Desmond/Maestro 2022.1 with the OPLS4 force field and SPC solvent model. The system was solvated ensuring at least 1.0 Å of distance between protein atoms and the water boundary. Sodium ions were added to neutralize the system. Initial equilibration involved: 100 ps of Brownian dynamics at 10 K with heavy atoms constrained, 12 ps of constrained NPT simulation, and 24 ps of unrestrained NPT simulation. The production MD run was conducted at 300 K and 1 bar under NPT conditions, with a 2 fs time step; trajectory frames were saved every 10 ps.

Statistical analysis

All experiments were conducted in triplicate, with results presented as mean ± SD. Statistical significance was elucidated using GraphPad Prism 10.0. Group comparisons were made using Student's t-test for pairwise comparisons and one-way ANOVA for analyses involving more than two groups. A P-value less than 0.05 was considered significant.

Author contributions

S.-Y. X. and Q.-X. L. contributed equally. S.-Y. X. and Q.-X. L.: investigation, methodology, writing original draft, validation, methodology, and writing – review and editing; Y. L., W.-H. W., and Q. L.: investigation and formal analysis; S.-G. L.: methodology, data curation, and supervision; G.-B. X. and J.-L. A.: project administration, validation, formal analysis, methodology, and writing – review and editing. All authors have given approval to the final version of the manuscript.

Conflicts of interest

There are no conflicts to declare.

Data availability

The data supporting this article have been included as part of the supplementary information (SI).

Supplementary information, including structure identification, physicochemical data of compounds 1–16, 19–24, 28–32, 35, and 36, and their ¹H NMR and ¹³C NMR spectra, is available. See DOI: https://doi.org/10.1039/d5md01142f.