Dehydrocostus lactone inhibits the proliferation and metastasis of hepatocellular carcinoma cells via modulating p53-p21-CDK2 signaling pathway

2.1 Culturing of malignant and normal human liver cell lines

HepG2 and L-O2 (a noncancerous human liver cell line) were acquired from American Type Culture Collection (VA, USA). SK-HEP-1 cells were obtained from the Cell Culture Center of the Institute of Basic Medical Sciences of the Chinese Academy of Medical Sciences (Beijing, China). We utilized Roswell Park Memorial Institute (RPMI)-1640, Dulbecco's modified Eagle medium (DMEM) and minimum essential medium (MEM) plus 10% fetal bovine serum (FBS) and 1% penicillin–streptomycin (P-S) solution for maintaining L-O2, HepG2 and SK-HEP-1 cell lines, respectively, in a 37 °C incubator filled with 5% CO₂.

2.2 Reagents and antibodies

Dehydrocostus lactone (DL) (purity, ≥ 98%; molecular weight, 230.30; formula, C₁₅H₁₈O₂) was obtained from Chendu Must Bio-Technology Co. Ltd. (see Fig. 1A for its structure). The DMEM, MEM, and P-S solution were purchased from Gibco (CA, USA), and 0.25% trypsin and FBS were from Sigma (MO, USA). Matrigel and The Transwell chambers and matrigel were procured from Corning (NY, USA). The E.Z.N.A.® Total RNA Kit was obtained from Omega Bio-Tek (GA, USA), and the kits for EdU cell proliferation assay, Annexin V-FITC staining, and cell cycle analysis were from Biyuntian Biotechnology (Shanghai, China). Z-VAD-FMK (HY-16658B) and CVT-313 (HY-15339) were purchased from MedChem Express (NJ, USA). Antibodies specific for β-actin (4970 T), Bax (41162S), Bcl-2 (15071 T), PARP (9532 T), γ-H2AX (9718 T), E-cadherin (3195 T), N-cadherin (13116 T), TCF8/ZEB1 (3396 T), β-catenin (8480 T), p53 (2527 T), p21 (2947 T), p27 (3686 T) and CDK2 (18048 T), and horseradish peroxidase-conjugated goat anti-rabbit (7074P2) and anti-mouse (7076P2) IgG were obtained from Cell Signaling Technology (MA, USA). The enhanced chemiluminescence kit for developing target bands in Western blotting was procured from GE Healthcare.

Close

Fig. 1

DL repressed HCC cell proliferation. (A) The structure of DL. (B) Viability of the two HCC cell lines incubated with DL for 24 and 48 h. (C) Representative fluorescence images of HCC cells incubated for 24 h with 5, 10, and 15 μM DL and stained by BeyoClick™EdU-488 (20 × water lens, NA 1.0). The right panel indicates the EdU-positive proportion. (D) Representative images of colonies formed by the HCC cells treated with 5 μM DL for 12 days. (E) Viability of the HCC cells and normal hepatocytes following incubation for 48 h with DL. *P < 0.05, **P < 0.01, ***P < 0.001.

Export to PPT

2.3 Cell viability assay

Briefly, 3000 cells were filled into 100 μL cell culture medium in each well of 96-well plates and allowed to attach for 24 h. After treatment with DL (at concentrations of 0, 5, 10, 15, 20, and 25 μM) for varying durations (24 h and 48 h), the cells were subjected to a 2 h incubation in the Cell Counting Kit-8 (CCK-8) reagent. Afterwards, we utilized a microplate reader to record the 450 nm optical absorbance for each well.

2.4 Colony formation assay

Briefly, cells (3000 cells per well) were inoculated into 6well plates. The cell culture medium was refreshed 24 h later and added with 5 μM DL. After culturing the cells for 12 days, they were rinsed with phosphate-buffered saline (PBS), fixed inside 4% paraformaldehyde for 15 min, and subjected to crystal violet staining for 0.5 h. Colonies larger than 0.5 mm were counted.

2.5 Cell proliferation analysis using EdU

The cells were plated and cultured for 24 h in 96-well plates, followed by incubation with DL (0, 5, 10, and 15 μM) for 48 h. After aspirating 50 μL medium from each well, the same volume of EdU-containing medium (20 μg/mL) was supplemented to incubate the cells for two hours under 37 °C. The EdU-labeled proliferating cells were detected with the Beyotime BeyoClick™ EdU Cell Proliferation Kit (Shanghai, China) following the provider’s instructions, and photographed and analyzed under a high-content screening (HCS) analysis system (Operetta CLS, PerkinElmer, USA).

2.6 Hoechst 33,258 staining

Eighty thousand cells were inoculated into each well of 6well plates, incubated for 24 h, and then subjected to DL treatments (0, 5, 10, and 15 μM). After incubating the cells with DL for 48 h, the cells were fixed for 10 min inside 4% paraformaldehyde and stained for 0.5 h by Hoechst 33,258 (10 μg/mL) at ambient temperature without being exposed to light. An Olympus IX73 fluorescence microscope (Japan) was utilized for observing and capturing the images.

2.7 Apoptosis and cell cycle assays

Cells were treated with DL as described above. Then the cells were analyzed for apoptosis with a Beyotime Annexin V-FITC kit (Shanghai, China) as per the protocol of the kit. For cell cycle assay, the cells were allowed to grow for 12 h in a medium without serum, incubated for 48 h with 10 μM DL, and analyzed by a kit for apoptosis and cell cycle assays from Beyotime (Shanghai, China) following the manual provided by the manufacturer.

2.8 Immunohistochemistry (IHC)

Cells maintained in 96-well plates were subjected to a 24 h treatment with 10 μM DL. After that, the cells were subjected to fixation with formaldehyde (4%) and then permeabilization using Triton X-100 (0.1%) for 10 and 20 min, respectively. The cells were then blocked for 1 h in 5% bovine albumin, incubated using an anti-γ-H2AX (Ser 139) antibody overnight at 4 °C, rinsed by PBS, and incubated for one hour using a DyLight 594-conjugated secondary antibody, followed by DAPI for 10 min. Images were acquired with High Content Screening (Operetta CLS, PerkinElmer, USA) using a 63 × water objective.

2.9 Wound healing assay

The cells were digested and inoculated into 12-well plates, and when their fusion grew to 90%, the cells were starved with serum-free medium for 12 h. After starvation for 12 h, scratch wounds were created on the HCC cell monolayers using 10 μL pipette tips. After the removal of dislodged cells, a serum-free medium containing 3 or 6 µM DL was added. The wound region was determined under a microscope at 0, 12, and 24 h after the addition of DL to appraise the HCC cells’ migration ability.

2.10 Matrigel transwell invasion assay

Serum-free medium was utilized for starvation treatment of cells for 12 h, then digested into a single cell suspension and adjusted the density of the cells to 1 × 106/mL. We added to the upper chamber of Matrigel-coated transwell inserts with 200 μL drug supplemented serum-free medium containing 2 × 105 cells, and lower chambers added 750 μL medium with 10% FBS, respectively. The cells were maintained in the transwell set for 24 h and then those moved across the Matrigel were sequentially fixed for 5 min and 15 min by paraformaldehyde and methanol, respectively. Afterwards, these cells were subjected to crystal violet staining for 10 min, rinsed with PBS, and counted.

2.11 Quantitative real-time PCR (qRT-PCR)

We utilized the E.Z.N.A. ®Total RNA Kit and the PrimeScript RT Reagent Kit for extraction of total RNA from treated HCC cells and reverse transcription, respectively. All the experiments were performed as per the instructions of the provider Quantitative PCR was implemented with TransStart Tip Green qPCR Kit using 4 µL cDNA as the template. The sequences of qRT-PCR primers are:

• CDK1 forward: 5′- AAACTACAGGTCAAGTGGTAGCC-3′,

• CDK1 reverse: 5′- TCCTGCATAAGCACATCCTGA-3′;

• CDK2 forward: 5′-CCAGGAGTTACTTCTATGCCTGA-3′,

• CDK2 reverse: 5′-TTCATCCAGGGGAGGTACAAC-3′;

• p53 forward: 5′- GAGGTTGGCTCTGACTGTACC-3′,

• p53 reverse: 5′- TCCGTCCCAGTAGATTACCAC-3′;

• CDKN1A forward: 5′-AAGTCAGTTCCTTGTGGAGCC-3′,

• CDKN1A reverse: 5′-GGTTCTGACGGACATCCCCA-3′;

• EAF2 forward: 5′-GGCAGAAGCTAGTCTAATGGACC-3′,

• EAF2 reverse: 5′-TGTACTGTGTCATGGTAGGATGT-3′.

2.12 Western blotting

Cells with different DL treatments (5, 10, and 15 μM for 24 h) were homogenized with RIPA buffer, and the total cell lysates were heated for 10 min at 99 °C for denaturation on an electric heating block. The samples were subjected to SDS-PAGE for separation of proteins with different molecular weights and then the proteins were blotted onto PVDF membranes. Following incubation for 1.5 h in 5% skimmed milk in TBST solution at ambient temperature, the membranes were incubated overnight with the primary antibody (1:1000) at 4 °C. The blots were rinsed by TBST the following day and then immersed for 4 h in diluted (2000 × ) secondary antibody solutions at 4 °C. After rinsing by TBST 3 times for 10 min each, the bands were developed using xx and visualized with Odyssey FC (LI-COR, Germany).

2.13 In vivo subcutaneous xenograft

Four- to five-week-old male BALB/c nude mice were procured from Beijing Vital River Laboratory Animal Technology Co. Ltd., and subcutaneously inoculated with 3 × 10⁶/200 μL HepG2 cells into the right posterior region. Once the tumors grew to a volume of 100 mm³, we randomly allocated the mice into the PBS, low-dose DL (10 mg/kg), high-dose DL (20 mg/kg), and 5-fluorouracil (5-FU) groups (n = 6 each). DL and PBS were injected intraperitoneally every day, while 5-FU was administered on alternate days. The body weight of the mice was appraised every day. Besides, the length and width of the subcutaneous tumor was measured by using electronic vernier calipers, and the volume of tumor was determined with the equation: tumor volume= (length × width × width)/2. Following the treatment regimens, the mice were subjected to euthanization, and the tumor tissues and main organs were collected and fixed in 4% paraformaldehyde. H&E staining and immunohistochemical analysis were performed as per standard protocols. All experiments on mice strictly followed the Guide for the Care and Use of Laboratory Animals.

2.14 Statistical analysis

Analyses of statistical data were implemented with GraphPad Prism 8.0. The means ± standard deviations (SD) of data from two groups were compared with unpaired two-tailed t test; while the data from ≥ 3 groups were subjected to one-way analysis of variance. P < 0.05 was deemed to signify statistical significance.

3.1 DL repressed HCC cell proliferation

Given that the unlimited proliferation of cells represents a hallmark of cancer progression, a majority of chemodrugs are designed to inhibit the proliferative capacity of tumor cells (Cardano et al., 2020). We treated SK-HEP-1 and HepG2 cells using DL at different concentrations (0–25 µM) for 24 and 48 h. As shown in Fig. 1B, DL significantly suppressed the proliferative capacity of the HCC cells dose- and time-dependently. Furthermore, the EdU incorporation assay revealed that DL markedly decreased the proportion of the EdU-labeled (green fluorescence) proliferating cells (Fig. 1C). Consistent with these results, DL treatment also significantly reduced the colony-forming ability of the HCC cells (Fig. 1D). In contrast, DL had little inhibitory effect on the normal hepatocyte L-O2 cell line (Fig. 1E), indicating that DL displayed selective cytotoxicity against the HCC cells.

3.2 DL promoted HCC cell apoptosis

Apoptosis was analyzed via Hoechst staining (Busto et al., 2015). The two HCC cell lines treated with DL displayed stronger fluorescence (Fig. 2A), which was indicative of higher apoptosis rates. Flow cytometry analysis showed that exposure to 15 μM DL significantly enhanced the percentages of HepG2 and SK-HEP-1 cells that underwent apoptosis to 53.88 ± 9.58% and 55.6 ± 5.57% respectively (Fig. 2C). In addition, Z-VAD-fmk, an inhibitor against caspases, reduced the sensitivity of the HCC cells to DL (Fig. 2B), indicating that human HCC cell apoptosis invoked by DL is triggered by the caspase signaling cascade. Consistent with this, DL treatment increased the protein level of Bax, and downregulated that of Bcl-2 in the HepG2 and SK-HEP-1 cells (Fig. 2D). Furthermore, DL also activated cleaved PARP, which plays a crucial role in apoptosis (Soldani and Scovassi, 2002). Taken together, DL attenuated HCC cell growth via triggering apoptosis.

Close

Fig. 2

DL invoked human HCC cell apoptosis. (A) Representative images of liver cancer cells incubated for 48 h with 0, 5, 10 and 15 μM DL and stained by Hoechst 33258. The apoptotic cells displayed blue fluorescence (20 × water lens, NA 1.0). (B) Viability of the two HCC cell lines incubated for 4 h with or without Z-VAD-fmk (5 μM) before treatment by various concentrations of DL for 24 h. (C) Flow cytometry plots showing percentage of apoptotic cells after DL treatment at various concentrations (0, 5, 10 and 15 μM) for 48 h. (D) Immunoblots showing Bcl-2 and Bax protein levels in cells treated for 24 h with DL (at concentrations of 0, 5, 10 and 15 μM). *P < 0.05, **P < 0.01, ***P < 0.001.

Export to PPT

3.3 DL caused DNA damage and G1 cell cycle arrest in human HCC cells

Based on the pro-apoptotic effects of DL, we hypothesized that DL may induce DNA damage, which ultimately causes cell cycle blockage and apoptotic cell death. As expected, cells treated with DL showed nuclear accumulation of γ-H2AX foci (Fig. 3A), which is indicative of DNA damage. In addition, DL significantly increased p-H2AX expression dose-dependently. (Fig. 3B). It was also exhibited that DL enhanced the proportion of G1-phase HCC cells, which suggests that it blocks cell cycle transition at the G1 stage (Fig. 3C).

Close

Fig. 3

DL induced damage of DNA and blockage of cell cycle at the G1 stage in human HCC cells. (A) Representative fluorescence images showing γ-H2AX foci in cells incubated with DL (10 μM) for 24 h (40 × water lens, NA 1.0). (B) Immunoblotting showing γ-H2AX levels in HCC cells subjected to 24-h-long DL treatments (at concentrations of 0, 5, 10 and 20 μM). (C) Flow cytometry plots showing the distribution of human HCC cells at different stages of cell cycle following treatment by 10 μM DL for 48 h. **P < 0.01, ***P < 0.001.

Export to PPT

3.4 DL attenuated the metastatic capacity of the HCC cells

As shown in the wound healing assay images in Fig. 4A, DL treatment reduced the coverage of the scratched region in the two HCC cell lines, suggesting that DL significantly reduced their migration ability. Furthermore, the number of HCC cells that invaded through the Matrigel-coated membrane in Transwell assay also decreased when treated with DL (Fig. 4B), suggesting DL can also inhibit the invasiveness of HCC cells. The metastasis of malignant tumor cells is driven by the embryonic epithelial-to-mesenchymal (EMT) program (Cho et al., 2019). DL upregulated E-cadherin, the epithelial markers and decreased the expression of mesenchymal markers, including N-cadherin, β-catenin and TCF8/ZEB1 in the HCC cells (Fig. 4C), revealing that DL suppressed the EMT of human liver cancer cells.

Close

Fig. 4

DL suppressed the migration ability and invasiveness of HCC cells. (A) Representative pictures obtained from the wound healing assay showing the migration ability of the two HCC cell lines at 0, 12 and 24 h after treatment with 0, 3 and 6 μM DL (20 × magnification). (B) Representative images of transwell assay showing the invasion of cells through Matrigel 24 h after treatment with 0, 3 and 6 μM DL (20 × magnification). (C) Immunoblots showing levels of EMT markers in the two HCC cell lines treated as indicated. *, ** and *** indicate P values<0.05, 0.01 and 0.001, respectively.

Export to PPT

3.5 The regulatory role of Dehydrocostus lactone on the p53-p21-CDK2 signaling pathway

To gain further insights into the molecular mechanisms underlying the effects of DL, we analyzed the transcriptomes of the untreated and DL-treated HCC cells through RNA sequencing. As depicted in Fig. 5A, differential genes are enriched in cell cycle and p53 signaling pathway by data analysis. DNA damage always leads to p53-p21 pathway activation and blocks cell cycle progression at the G1 phase to allow the cells to repair the DNA lesions. The activation of p21 by p53 inhibits CDK2(He et al., 2005). In addition, the findings of qRT-PCR and Western blotting displayed that DL increased the mRNA and protein abundances of p53, p21 and reduced those of CDK2 in HepG2 and SK-HEP-1 cells (Fig. 5B, C). Taken together, the modulation of the p53-p21-CDK2 signaling cascade contributed in part to the anti-cancer effect of DL in human hepatoma cells. In addition, DL is also able to downregulate the mRNA abundance of CDK1 and upregulate that of EAF2 and the protein level of p27 (Fig. 5B, C).

Close

Fig. 5

DL can regulate the p53-p21-CDK2 signaling pathway to exert an anti-hepatic carcinoma effect. (A) Cells were subjected to a 24-h DL treatment at 10 μM and subsequently transcriptome sequencing. Transcriptome data analysis to identify enriched pathways. (B) Results of qRT-PCR showing p53, p21 and CDK2 mRNA levels in the HCC cells treated with DL (0 and 10 μM) for 24 h. (C) Immunoblots showing levels of p53, p21 and CDK2 proteins in HCC cells treated for 24 h with DL (0, 5, 10 and 20 μM). *P < 0.05, **P < 0.01, ***P < 0.001.

Export to PPT

3.6 Dehydrocostus lactone exerted tumoricidal efficacy in vivo.

To verify the results of in vitro experiments, we generated HepG2 cells-derived xenografts in BALB/c mice and injected different doses of DL into the mice for 22 days. As displayed by Fig. 6B, the average volume of tumors in mice injected with high-dose DL (20 mg/kg) was markedly lower compared to that of the PBS control mice. However, the body weight of the mice was not affected by DL treatment (Fig. 6A), which was indicative of the overall safety of DL. In addition, tumor tissues derived from the DL-treated mice exhibited a marked decrease in the in-situ expression of Ki67 and MMP9 compared to that of the control mice (Fig. 6D), indicating that DL can inhibit HCC cell proliferation and metastasis in vivo. Moreover, relative to the control, DL treatment also elevated the percentages of TUNEL-positive apoptotic cells in the tumor tissues (Fig. 6D). Finally, histopathological examination of the liver, heart, lung, kidney, and spleen tissues did not reveal any signs of toxicity due to DL treatment relative to the control group (Fig. 6C). Taken together, DL can effectively inhibit HCC growth in vivo with no detectable damage to major organs such as the heart, liver, spleen, lung and kidney.

Close

Fig. 6

DL inhibited the growth of HepG2 xenograft in mice. Each mouse was injected with 3 × 10⁶ cells/0.2 mL HepG2 cells. Once tumor tissues grew to a volume of approximately 50 mm³, the animals were treated with 10 and 20 mg/kg DL, PBS or 5-FU (20 mg/kg). The body weight (A) and tumor volume (B) of the mice were determined every two days. (C) Representative pictures of H&E-stained liver, heart, lung, kidney, and spleen tissues (20 × magnification). (D) Representative images of tumor tissues showing TUNEL-positive apoptotic cells and in-situ expression of Ki67 and MMP9 (20 × magnification).

Export to PPT