Phytochemical library screening reveals betulinic acid as a novel Skp2-SCF E3 ligase inhibitor in non-small cell lung cancer

Abstract

Skp2 is overexpressed in multiple cancers and plays a critical role in tumor development through ubiquitin/proteasome-dependent degradation of its substrate proteins. Drugs targeting Skp2 have exhibited promising anticancer activity. Here, we identified a plant-derived Skp2 inhibitor, betulinic acid (BA), via high-throughput structure-based virtual screening of a phytochemical library. BA significantly inhibited the proliferation and migration of non-small cell lung cancer (NSCLC) through targeting Skp2-SCF E3 ligase both in vitro and in vivo. Mechanistically, BA binding to Skp2, especially forming H-bonds with residue Lys145, decreases its stability by disrupting Skp1-Skp2 interactions, thereby inhibiting the Skp2-SCF E3 ligase and promoting the accumulation of its substrates; that is, E-cadherin and p27. In both subcutaneous and orthotopic xenografts, BA significantly inhibited the proliferation and metastasis of NSCLC through targeting Skp2-SCF E3 ligase and upregulating p27 and E-cadherin protein levels. Taken together, BA can be considered a valuable therapeutic candidate to inhibit metastasis of NSCLC.

Keywords: E-cadherin; NSCLC; Skp2; betulinic acid; metastasis.

FIGURE 1

Online database analyses show that the higher expression of Skp2 is negatively associated with patients’ prognosis. A, The expression of Skp2 in the Hou Lung, Landi Lung, Selamat Lung, and Su Lung of Oncomine datasets is shown. B, The Kaplan‐Meier survival curve of non–small cell lung cancer (NSCLC) patients with high or low expression levels of Skp2 is determined by the kmplot database. C, The association between the expression levels of Skp2 and the prognosis of patients is determined by the prognoscan database. **P < .01, and ***P < .001 vs normal group

FIGURE 2

Identification of betulinic acid (BA) as a Skp2 inhibitor. A, Plant‐derived compounds were docked to the binding pocket of Skp2‐Skp1. Skp2 and compounds are shown as cartoon and line, respectively. B, Co‐immunoprecipitation (co‐IP) experiments were performed to examine the effects of four selected compounds and SZL‐P1‐41 on Skp2‐Skp1 binding in 293T cells transfected with Skp2 and Flag‐Skp1. C, Endogenous Skp2‐Skp1 interactions with or without BA in H1299 cells were examined by co‐IP. D, 293T cells were transfected with Flag‐Skp1 to examine the Skp2‐Skp1 interaction with or without BA treatment. E, Ubiquitination assay of exogenous p27 in 293T cells transfected with Myc‐p27, HA‐Ub, and Skp2 upon BA or SZL‐P1‐41 treatment. F, Ubiquitination assay of endogenous p27 in A549 cells upon BA or SZL‐P1‐41 treatment. The asterisk indicates the bands of IgG light chain. G, Ubiquitination assay of exogenous E‐cadherin in 293T cells transfected with Flag‐E‐cadherin, HA‐Ub, and Skp2 upon BA treatment. H, Ubiquitination assay of endogenous E‐cadherin in A549 cells upon BA treatment. I, J, A cellular thermal shift assay (CETSA) of Skp2 was conducted by western blotting (I); quantification of the cellular thermal curve shift of Skp2 protein (J)

FIGURE 3

Betulinic acid (BA) directly interacts with Skp2 at Lys145 residue. A, Molecular docking revealed the potential binding residue of BA in Skp2; BA is shown as sticks and the residues composed of the pocket are labeled and shown as lines. B, BA (shown as sticks) was demonstrated to disrupt the interaction of Skp2 (shown as electrostatic potential surface) with Skp1 (shown as cartoon) by docking. C, 293T cells were transfected with Flag‐Skp1 and HA‐Skp2 WT or HA‐Skp2 K145A to perform Skp2‐Skp1 binding assay with or without BA treatment for 6 h. D, 293T cells were transfected with HA‐Skp2 WT or HA‐Skp2 K145A to perform a cellular thermal shift assay (CETSA) with or without BA (20 μmol L⁻) treatment. E, Quantification of the cellular thermal curve shift of Skp2 protein

FIGURE 4

Betulinic acid (BA) decreases Skp2 and increases p27 and E‐cadherin in non–small cell lung cancer (NSCLC) cells. A, B H1299 cells were treated with cycloheximide (CHX) in the presence or absence of BA for indicated time; protein levels of Skp2 were detected by western blotting (A); Skp2 protein level was quantified by densitometry analysis (B). C, D The protein levels of Skp2 and p27 in A549 and H1299 cells were examined by western blotting upon BA or SZL‐P1‐41 treatment for 24 h (C); protein levels of Skp2 and p27 were quantified by densitometry analysis (D). E, F The protein levels of the indicated proteins in H1299, A549, and Lewis lung carcinoma cell line (LLC) cells were examined by western blotting upon BA treatment for 24 h (E); protein levels of Skp2 and E‐cadherin were quantified by densitometry analysis (F). G, H The protein levels of Skp2, p27, and E‐cadherin were detected by western blotting after BA (20 μmol L⁻) treatment for indicated time points (G); quantification of Skp2, p27, and E‐cadherin by densitometry analysis (H). All data are expressed mean ± SEM. ^# P < .05 vs BA treatment groups. *P < .05, **P < .01, and ***P < .001 vs control

FIGURE 5

Betulinic acid (BA) inhibits cell migration and stemness of non–small cell lung cancer (NSCLC). A, B Wound healing assay of NSCLC cells treated with BA: cells were imaged and visualized by light microscopy (A); relative migration rate was quantified by ImageJ software (B). C, D Transwell assay of NSCLC cells treated with DMSO or BA (C); cells were counted from three independent experiments (D). E, A549 and H1299 cells cultured in non‐adherent conditions were treated with DMSO or BA for 7 days. Spheroids were photographed by light microscopy. F, A549 and H1299 cells cultured in three‐dimensional Matrigel were administrated with DMSO or BA for 7 days. Images were obtained via optical microscope. The values are expressed as the means ± SEM; *P < .05, **P < .01, and ***P < .001 vs control

FIGURE 6

Overexpression of Skp2 reverses the antimigration effect of betulinic acid (BA). A, Protein levels of Skp2 and E‐cadherin in non–small cell lung cancer (NSCLC) cells stably overexpressing Skp2 were examined by western blotting. B, H1299 cells with Skp2 overexpression or EV were treated with BA for 24 h; the expression of Skp2 and E‐cadherin were examined by western blotting. C, D Wound healing assays of A549 and H1299 cells stably overexpressing Skp2 in the presence or absence of BA (C); relative migration rate was quantified by ImageJ software (D). E, F Transwell assay of A549 and H1299 cells that stably overexpress Skp2 (E); cells were counted from three independent experiments (F). EV, empty vector. The values are expressed as mean ± SEM. *P < .05, **P < .01, and ***P < .001 vs controls. ^# P < .05, ^## P < .01, and ^### P < .001 vs EV group

FIGURE 7

Betulinic acid (BA) suppresses tumor growth and metastasis in vivo. A‐E, Images of subcutaneous xenografted tumors of Lewis lung carcinoma cell line (LLC) cells were dissected and photographed A. Tumor volumes were measured and calculated B. Representative pictures of lung metastases (upper panel) and the H&E staining of metastatic nodules (lower panel) in the lung tissue of each group are shown (C). Number of metastasis nodules was counted (D). Protein levels of Skp2, p27, and E‐cadherin were detected by western blotting and quantified by densitometry analysis from four mice from each group (E). F‐I, LLC cells were injected into C57BL/6 mice via tail vein to establish lung metastasis model. Representative pictures of lung metastases (upper panel) and H&E staining of lung tissue sections (lower panel) are shown (F). The number of metastasis nodules in the lung tissue was counted (G). Overall survival analysis of mice receiving indicated treatments (H). Immunohistochemistry (IHC) analysis of E‐cadherin, Skp2, and Ki67 expression in the lung tissues (I). J‐L, A549‐Luciferase cells were intravenously injected into BALB/C nude mice. Bioluminescence images were obtained (J), and the relative luminescence intensity in the mice was calculated (K). Representative pictures of lung metastases (upper panel) and H&E staining of lung tissue sections (lower panel) of the nude mice (L). Black arrows: metastasis nodules in the lung tissue. The values are expressed as the mean ± SEM. *P < .05, **P < .01, and ***P < .001 vs vehicle control

FIGURE 8

Schematic identification of betulinic acid as a novel Skp2‐SCF E3 ligase inhibitor to inhibit proliferation and metastasis of non–small cell lung cancer (NSCLC)